Molecular characteristics of Kidney Cancer - KOPS - Universität ...
Molecular characteristics of Kidney Cancer - KOPS - Universität ...
Molecular characteristics of Kidney Cancer - KOPS - Universität ...
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
MOLECULAR CHARACTERISTICS OF<br />
KIDNEY CANCER<br />
DISSERTATION<br />
Zur Erlangung des akademischen Grades des<br />
Doktors der Naturwissenschaften an der <strong>Universität</strong> Konstanz<br />
Fachbereich Biologie<br />
Vorgelegt von<br />
KERSTIN STEMMER<br />
Konstanz, im Oktober 2008<br />
Tag der mündlichen Prüfung:<br />
15. Dezember 2008<br />
Referenten:<br />
Pr<strong>of</strong>. Dr. Daniel R. Dietrich<br />
Pr<strong>of</strong>. Dr. Christ<strong>of</strong> R. Hauck<br />
PD Dr. Sonja von Aulock<br />
I
“Success consists <strong>of</strong> going from failure to failure without loss <strong>of</strong> enthusiasm”<br />
Winston Churchill (1874-1965)<br />
III
Table <strong>of</strong> Contents<br />
Table <strong>of</strong> Contents........................................................................................................................................ IV<br />
List <strong>of</strong> Figures.............................................................................................................................................. XI<br />
List <strong>of</strong> Tables............................................................................................................................................. XIII<br />
Abbreviations ............................................................................................................................................XIV<br />
Publications and Presentations ..................................................................................................................XV<br />
Zusammenfassung..................................................................................................................................XVIII<br />
Summary...................................................................................................................................................XXI<br />
Chapter 1: Introduction........................................................................................................................... 1<br />
1.1 <strong>Kidney</strong> <strong>Cancer</strong>........................................................................................................................... 1<br />
1.1.1 Risk Factors for <strong>Kidney</strong> <strong>Cancer</strong>............................................................................................ 1<br />
1.1.2 Genetic and Epigenetic Causes <strong>of</strong> <strong>Cancer</strong>........................................................................... 2<br />
1.1.3 Models <strong>of</strong> Carcinogenesis .................................................................................................... 3<br />
1.1.4 Renal Tumor Pathology........................................................................................................ 4<br />
1.2 Renal Carcinogens.................................................................................................................. 10<br />
1.2.1 Genotoxic Carcinogens ...................................................................................................... 10<br />
1.2.2 Non-Genotoxic Carcinogens .............................................................................................. 11<br />
1.3 Renal Carcinogens Relevant for this Study............................................................................. 13<br />
1.3.1 Aristolochic acid.................................................................................................................. 13<br />
1.3.2 Ochratoxin A....................................................................................................................... 15<br />
1.3.3 Cycasin / Methylazoymethanol........................................................................................... 16<br />
1.4 Methods to Identify Renal Carcinogens................................................................................... 18<br />
1.4.1 Standard carcinogenicity testing......................................................................................... 19<br />
1.4.2 The lifetime rodent bioassay............................................................................................... 19<br />
1.4.3 The Eker Rat Model <strong>of</strong> Renal Carcinogenesis.................................................................... 21<br />
1.5 Toxicogenomics ...................................................................................................................... 22<br />
1.5.1 Principles <strong>of</strong> the Microarray Technology............................................................................. 22<br />
1.5.2 Microarrays- a Novel Tool to Study Carcinogenesis........................................................... 24<br />
1.5.3 Microarrays- Current Limitations and Future Directions ..................................................... 27<br />
Chapter 2: Work Hypotheses & Experimental Design.......................................................................... 28<br />
IV
2.1 Part I........................................................................................................................................ 28<br />
2.1.1 Experimental setup............................................................................................................. 29<br />
2.2 Part II....................................................................................................................................... 30<br />
2.2.1 Experimental setup:............................................................................................................ 30<br />
2.3 Part III...................................................................................................................................... 31<br />
2.3.1 Experimental setup............................................................................................................. 31<br />
2.4 Schematic Overview on the Experimental Design................................................................... 32<br />
Chapter 3: Manuscript I........................................................................................................................ 33<br />
3.1 Abstract ................................................................................................................................... 33<br />
3.2 Introduction ............................................................................................................................. 34<br />
3.3 Material and Methods.............................................................................................................. 36<br />
3.3.1 Compounds ........................................................................................................................ 36<br />
3.3.2 Animals............................................................................................................................... 36<br />
3.3.3 Animal treatment and sample collection ............................................................................. 36<br />
3.3.4 Histopathology.................................................................................................................... 37<br />
3.3.5 Immunohistochemistry........................................................................................................ 37<br />
3.3.6 RNA isolation and expression pr<strong>of</strong>iling ............................................................................... 37<br />
3.3.7 Microarray data processing and statistical analysis............................................................ 38<br />
3.3.8 Functional analysis <strong>of</strong> microarray data ............................................................................... 38<br />
3.3.9 Statistical analysis .............................................................................................................. 39<br />
3.4 Results .................................................................................................................................... 39<br />
3.4.1 Cell proliferation data.......................................................................................................... 39<br />
3.4.2 Preneoplastic and neoplastic pathology ............................................................................. 39<br />
3.4.3 Non- neoplastic pathology .................................................................................................. 40<br />
3.4.4 Gene expression pr<strong>of</strong>iles.................................................................................................... 42<br />
3.4.5 Functional analysis <strong>of</strong> significantly deregulated genes....................................................... 43<br />
3.4.6 Genes deregulated by aristolochic acid.............................................................................. 43<br />
3.4.7 Genes deregulated by ochratoxin A ................................................................................... 48<br />
3.5 Discussion............................................................................................................................... 57<br />
3.6 Supplement ............................................................................................................................. 62<br />
V
3.6.1 Supplementary Information ................................................................................................ 62<br />
3.6.2 Supplementary Tables........................................................................................................ 63<br />
Chapter 4: Manuscript II....................................................................................................................... 73<br />
4.1 Abstract ................................................................................................................................... 73<br />
4.2 Introduction ............................................................................................................................. 74<br />
4.3 Material and Methods.............................................................................................................. 76<br />
4.3.1 Compound.......................................................................................................................... 76<br />
4.3.2 Animals............................................................................................................................... 76<br />
4.3.3 Short-term experiments ...................................................................................................... 77<br />
4.3.4 Subchronic and chronic experiments.................................................................................. 77<br />
4.3.5 Sample collection ............................................................................................................... 77<br />
4.3.6 Histopathology.................................................................................................................... 77<br />
4.3.7 Cell proliferation analysis.................................................................................................... 78<br />
4.3.8 RNA isolation microarray hybridization and gene expression pr<strong>of</strong>iling ............................... 78<br />
4.3.9 Quantification <strong>of</strong> O6MG ...................................................................................................... 79<br />
4.3.10 Statistical analysis ......................................................................................................... 79<br />
4.4 Results .................................................................................................................................... 80<br />
4.4.1 Gene expression pr<strong>of</strong>iling................................................................................................... 80<br />
4.4.2 Non-neoplastic renal pathology .......................................................................................... 83<br />
4.4.3 (Pre)-neoplastic renal pathology......................................................................................... 84<br />
4.4.4 Cell proliferation.................................................................................................................. 87<br />
4.4.5 Detection <strong>of</strong> O6MG in MAMAc treated Eker-rats................................................................ 88<br />
4.5 Discussion............................................................................................................................... 89<br />
4.6 Supplement ............................................................................................................................. 92<br />
4.6.1 Supplementary experimental procedures: Real time PCR ................................................. 92<br />
4.6.2 Supplementary Figures ...................................................................................................... 92<br />
4.6.3 Supplementary Tables........................................................................................................ 93<br />
Chapter 5: Manuscript III...................................................................................................................... 95<br />
5.1 Abstract ................................................................................................................................... 95<br />
5.2 Introduction ............................................................................................................................. 96<br />
5.3 Material and Methods.............................................................................................................. 98<br />
VI
5.3.1 Animal treatment and tissue preparation ............................................................................ 98<br />
5.3.2 Sectioning and staining ...................................................................................................... 98<br />
5.3.3 Laser-assisted microdissection........................................................................................... 99<br />
5.3.4 RNA isolation and quality control........................................................................................ 99<br />
5.3.5 RNA amplification ............................................................................................................. 100<br />
5.3.6 Affymetrix chip hybridization and expression pr<strong>of</strong>iling ...................................................... 101<br />
5.4 Results .................................................................................................................................. 102<br />
5.4.1 Tissue fixation and staining .............................................................................................. 102<br />
5.4.2 RNA quality and yield from pooled microdissected samples ............................................ 105<br />
5.4.3 RNA amplification and microarray hybridization ............................................................... 106<br />
5.4.4 Performance <strong>of</strong> the established protocol .......................................................................... 110<br />
5.5 Discussion............................................................................................................................. 111<br />
5.6 Appendix ............................................................................................................................... 112<br />
Chapter 6: Manuscript IV ................................................................................................................... 114<br />
6.1 Abstract ................................................................................................................................. 114<br />
6.2 Introduction ........................................................................................................................... 115<br />
6.3 Material and Methods............................................................................................................ 116<br />
6.3.1 Compounds ...................................................................................................................... 116<br />
6.3.2 Animals............................................................................................................................. 116<br />
6.3.3 Sample collection ............................................................................................................. 117<br />
6.3.4 Histopathology.................................................................................................................. 117<br />
6.3.5 Immunohistochemistry...................................................................................................... 117<br />
6.3.6 Laser microdissection and RNA isolation ......................................................................... 118<br />
6.3.7 Microarray data processing and analysis ......................................................................... 118<br />
6.3.8 Statistical analysis ............................................................................................................ 119<br />
6.4 Results .................................................................................................................................. 119<br />
6.4.1 Non-neoplastic pathology ................................................................................................. 119<br />
6.4.2 Cell proliferation................................................................................................................ 120<br />
6.4.3 (Pre-) neoplastic pathology............................................................................................... 121<br />
6.4.4 Gene expression pr<strong>of</strong>iling................................................................................................. 123<br />
VII
6.4.5 Confirmation <strong>of</strong> TORC1 and TORC2 pathway activation in preneoplastic lesions ........... 127<br />
6.5 Discussion............................................................................................................................. 130<br />
6.5.1 Compound-induced tubular damage and regenerative cell proliferation is critical for the<br />
formation but not for the progression <strong>of</strong> preneoplastic lesions......................................................... 130<br />
6.5.2 Time dependent tissue adaptation to carcinogen treatment ............................................. 131<br />
6.5.3 Numbers <strong>of</strong> earliest preneoplastic lesions formed are decisive for later occurrence <strong>of</strong><br />
veritable tumors ............................................................................................................................... 132<br />
6.5.4 The balance <strong>of</strong> TORC1 and TORC2 pathways determines preneoplastic lesion progression<br />
133<br />
6.5.5 Outlook ............................................................................................................................. 134<br />
6.6 Supplemental Material........................................................................................................... 135<br />
6.6.1 Supplementary experimental procedures: Real time PCR ............................................... 135<br />
6.6.2 Supplementary Figures .................................................................................................... 136<br />
Supplemental Tables ....................................................................................................................... 140<br />
Chapter 7: Manuscript V .................................................................................................................... 145<br />
7.1 Abstract ................................................................................................................................. 145<br />
7.2 Introduction ........................................................................................................................... 146<br />
7.3 Material and Methods............................................................................................................ 148<br />
7.3.1 Animals............................................................................................................................. 148<br />
7.3.2 Blood parameters ............................................................................................................. 148<br />
7.3.3 Renal Pathology ............................................................................................................... 149<br />
7.3.4 Cell Proliferation analysis ................................................................................................. 149<br />
7.3.5 Immunohistochemistry...................................................................................................... 149<br />
7.3.6 Statistics ........................................................................................................................... 150<br />
7.4 Results .................................................................................................................................. 150<br />
7.4.1 Wistar rats show widely differing propensity towards body weight gain in response to<br />
prolonged high-fat diet exposure...................................................................................................... 150<br />
7.4.2 Increased levels <strong>of</strong> adipokines and insulin in DIOsens and DIOres rats........................... 151<br />
7.4.3 High-fat diet exposure causes preneoplasia in DIOsens and DIOres rats........................ 152<br />
7.4.4 High-fat diet-induced changes in kidney pathology are independent from α2u-globulin<br />
accumulation.................................................................................................................................... 155<br />
7.4.5 High fat diet increases cell proliferation in proximal tubule <strong>of</strong> DIOsens and DIOres rats.. 155<br />
VIII
7.4.6 Involvement <strong>of</strong> the mTOR pathway .................................................................................. 156<br />
7.5 Discussion............................................................................................................................. 157<br />
7.5.1 Similar renal preneoplasias in DIOsens and DIOres rats suggest a minor role <strong>of</strong> high body<br />
adiposity or its co-morbidities........................................................................................................... 158<br />
7.5.2 High-fat diet exposure does not induce α2u-globulin accumulation................................... 159<br />
7.5.3 mTor-S6K signaling: The link between obesity and kidney cancer................................... 160<br />
7.5.4 Conclusion........................................................................................................................ 161<br />
Chapter 8: Additional Data ................................................................................................................. 162<br />
8.1 Gender Disparity in Early Ochratoxin A and Aristolochic Acid Mediated Renal Toxicity in Shortterm<br />
Assays ......................................................................................................................................... 162<br />
8.2 Ochratoxin A ......................................................................................................................... 162<br />
8.2.1 Cell proliferation (PCNA labeling index) ........................................................................... 162<br />
8.2.2 Renal pathology................................................................................................................ 163<br />
8.2.3 Conclusion........................................................................................................................ 164<br />
8.3 Aristolochic acid .................................................................................................................... 165<br />
8.3.1 Cell proliferation (PCNA labeling index) ........................................................................... 165<br />
8.3.2 Renal pathology................................................................................................................ 165<br />
8.3.3 Conclusion........................................................................................................................ 166<br />
Chapter 9: Overall Discussion and Future Perspectives .................................................................... 167<br />
9.1 Short-term In Vivo Tests to Identify Renal Carcinogens........................................................ 167<br />
9.1.1 Prerequisite I: Short term treatment would result in detectable compound-specific gene<br />
expression changes ......................................................................................................................... 168<br />
9.1.2 Prerequisite II: Genotoxic and non-genotoxic renal carcinogens can be distinguished by<br />
early gene expression pr<strong>of</strong>iles ......................................................................................................... 171<br />
9.1.3 Prerequisite III: Manifestation <strong>of</strong> early compound-induced expression changes in kidney<br />
tumors. 172<br />
9.1.4 Summary Part I................................................................................................................. 174<br />
9.2 Cellular Signaling Cascades in Preneoplastic Lesion Progression ....................................... 175<br />
9.2.1 Pivotal role <strong>of</strong> carcinogen-induced tubular damage, regeneration and adaptation processes<br />
for the development <strong>of</strong> kidney cancer .............................................................................................. 176<br />
9.2.2 Predictive potential <strong>of</strong> preneoplastic lesions as tumor precursors .................................... 180<br />
9.2.3 Summary Part II................................................................................................................ 183<br />
IX
9.3 Obesity as biasing factor for a reliable human risk assessment from rodent bioassays........ 184<br />
9.4 Synopsis................................................................................................................................ 186<br />
Chapter 10: References ....................................................................................................................... 188<br />
Erklärung.................................................................................................................................................. 199<br />
Danksagung ............................................................................................................................................. 200<br />
X
List <strong>of</strong> Figures<br />
Figure 1.1: Overview <strong>of</strong> the morphological structures <strong>of</strong> the renal nephron.. ................................ 5<br />
Figure 1.2: Progression <strong>of</strong> renal lesions........................................................................................ 8<br />
Figure 1.3: Chemical structures <strong>of</strong> AAI and AAII......................................................................... 13<br />
Figure 1.4: Chemical structure <strong>of</strong> OTA........................................................................................ 15<br />
Figure 1.5: Chemical structures <strong>of</strong> cycasin, MAM and methyldiazonium ion............................... 17<br />
Figure 1.6: Composition <strong>of</strong> an Affymetrix chip............................................................................. 23<br />
Figure 2.1: Overview on the experimental design ....................................................................... 32<br />
Figure 3.1: Comparison <strong>of</strong> PCNA S-Phase labelling indices: short term assay ......................... 40<br />
Figure 3.2: PCNA staining patterns <strong>of</strong> control- and OTA-treated Eker rats: short term assay. ... 41<br />
Figure 3.3: AA or OTA induced renal gene expression pr<strong>of</strong>iles in short term assays ................. 42<br />
Figure 3.4: Postulated mechanistic pathways <strong>of</strong> AA and OTA induced toxicity........................... 59<br />
Figure 4.1: Metabolic pathways <strong>of</strong> cycasin and MAM activation ................................................. 75<br />
Figure 4.2: MAMAc induced renal gene expression pr<strong>of</strong>iles in short term assays ...................... 80<br />
Figure 4.3: Representative preneoplastic lesions in MAMAc treated rats ................................... 85<br />
Figure 4.4: Mean number <strong>of</strong> preneoplastic lesions in MAMAc treated Eker rats......................... 86<br />
Figure 4.5: PCNA S-Phase mean labeling indices in MAMAc treated rats.................................. 87<br />
Figure 4.6: Number <strong>of</strong> O6MG adducts detected via LC-MS/MS ................................................. 88<br />
Figure 5.1: Histology <strong>of</strong> H&E stained renal cryo-sections. ........................................................ 103<br />
Figure 5.2: RNA quality in renal cyrosections ........................................................................... 105<br />
Figure 5.3: Comparison <strong>of</strong> total RNA yields after one- or two cycle amplification .................... 107<br />
Figure 5.4: Expression pr<strong>of</strong>iles <strong>of</strong> differentially expressed genes. ........................................... 108<br />
Figure 5.5.: Principal components analysis: One cycle and two-cyce amplification ................... 109<br />
Figure 5.6. Boxplot: Analyis <strong>of</strong> Affymetrix genechip data .......................................................... 110<br />
Figure 6.1. Preneoplastic lesions in AA- and OTA-treated rats: chronic exposure.................... 122<br />
Figure 6.2: Heatmap comparing gene expression changes <strong>of</strong> microdissected lesions ............. 123<br />
Figure 6.3. Postulated mechanisms involved in chemically induced in renal cacinogenesis..... 127<br />
Figure 6.4: mTOR activation in preneoplastic lesions............................................... ............. ...129<br />
Figure 7.1: Adipokine, cytokine, and insulin levels ................................................................... 152<br />
Figure 7.2: Images <strong>of</strong> renal histopathology in DIOsens and DIOres rats .................................. 154<br />
Figure 7.3: PCNA staining in DIOsens and DIOres rats............................................................ 156<br />
Figure 7.4: pS6RP staining in DIOsens and DIOres rats .......................................................... 157<br />
Figure 8.1: PCNA labelling indices in OTA treated Eker and willd type rats.............................. 163<br />
Figure 8.2: Total numbers and incidences <strong>of</strong> bATs in OTA treated rats.................................... 164<br />
Figure 8.3: PCNA labelling indices in AA treated Eker and willd type rats. .............................. 165<br />
Figure 8.4: Total numbers and incidences <strong>of</strong> bATs in AA treated rats ...................................... 166<br />
XI
Figure 9.1: Hypothetical model <strong>of</strong> chemically-induced renal carcinogenesis ........................... 177<br />
Figure 9.2: Gene expression pr<strong>of</strong>iles from short-term treated rats and HT .............................. 179<br />
Figure 9.3: Overview <strong>of</strong> the mTOR pathway ............................................................................ 183<br />
Supplementary Figures:<br />
Supplementary figure 4.1: Real time PCR MGMT...................................................................................... 92<br />
Supplementary figure 6.1: Overview over the experimental design.......................................................... 136<br />
Supplementary figure 6.2: H&E stained paraffin sections: Renal pathology............................................. 137<br />
Supplementary figure 6.3; Immunmohistochemistry pS6RP, pAKT, Foxo1 ........................................... 138<br />
Supplementary figure 6.4: Real time PCR: OGG1 and CAT .................................................................... 139<br />
XII
List <strong>of</strong> Tables<br />
Table 1.1: Classification <strong>of</strong> kidney tumors.................................................................................... 6<br />
Table 1.2: Images <strong>of</strong> frequent cellular abnormalities in rat kidney tumors ................................... 9<br />
Table 3.1: Gene categories <strong>of</strong> genes differentially deregulated by AA ...................................... 45<br />
Table 3.2: Gene categories <strong>of</strong> genes differentially deregulated by OTA.................................... 50<br />
Table 4.1: Genes <strong>of</strong> genes differentially deregulated by MAMAc .............................................. 81<br />
Table 4.2: Non-neoplastic renal pathology in MAMAc treated Eker r......................................... 84<br />
Table 5.1: Influence <strong>of</strong> tissue preservation on RNA quality...................................................... 102<br />
Table 5.2: Influence <strong>of</strong> long-term storage on RNA quality and yield ........................................ 103<br />
Table 5.3: RNA quality and yields from pooled microdissected samples................................. 106<br />
Table 6.1: Comparison <strong>of</strong> BrdU S-Phase labeling indices: Chronic treatment ......................... 120<br />
Table 6.2: Numbers and categories <strong>of</strong> deregulated genes in preneoplastic lesions ................ 125<br />
Table 7.1: Overview <strong>of</strong> antibodies and modifications to the manufacture’s protocols .............. 149<br />
Table 7.2: Life-history traits <strong>of</strong> chow-fed rats, DIOres and DIOsens rats ................................. 151<br />
Table 7.3: Histopathology in chow-fed, DIOres, and DIOsens rats.......................................... 153<br />
Table 9.1: Pathways involved in AA-, OTA- or MAMAc-induced renal carcinogenicity ............ 170<br />
Table 9.2: Genes and pathways linked to the six hallmarks <strong>of</strong> cancer......................................181<br />
Supplementary Tables:<br />
Supplementary table 3.1: Numbers and categories <strong>of</strong> genes deregulated by AA or OTA.......................... 63<br />
Supplementary table 3.2: Incidence and numbers <strong>of</strong> AA and OTA induced renal lesions.......................... 65<br />
Supplementary table 3.3: Non-neoplastic pathology <strong>of</strong> AA or OTA treated Eker rats. ............................... 66<br />
Supplementary table 3.4: Non-neoplastic pathology <strong>of</strong> AA or OTA treated wild type rats. ........................ 70<br />
Supplementary table 4.1: Non-neoplastic pathology <strong>of</strong> MAMAc treated rats ............................................. 93<br />
Supplementary table 4.2: Incidence and numbers <strong>of</strong> MAMAc induced renal lesions. ................................ 94<br />
Supplementary table 6.1: Non-neoplastic renal pathology: 6months treatment ....................................... 140<br />
Supplementary table 6.2: (Pre-)neoplastic pathology: 6 months treatment ............................................. 141<br />
Supplementary table 6.3: Genes differentially deregulated in preneoplastic lesions ............................... 142<br />
XIII
Abbreviations<br />
α2u alpha 2 urine-globulin<br />
AA aristolochic acid<br />
AAN aristolochic acid nephropathy<br />
bAT basophilic atypical tubule<br />
bAH basophilic atypical hyperplasia<br />
BEN Balkan endemic nephropathy<br />
bw body weight<br />
CPN chronic progressive nephropathy<br />
DIO diet induced obesity<br />
ECM extracellular matrix<br />
EK Eker<br />
EST expressed sequence tag<br />
HT healthy tubule<br />
IARC International Agency for Research on <strong>Cancer</strong><br />
ICH International Conference on Harmonization<br />
i.p. intraperitoneal<br />
LC/MS Liquid Chromatography/ Mass spectrometry<br />
LOH loss <strong>of</strong> heterozygosity<br />
MAM mehylazoxymethanol<br />
MAMAc mehylazoxymethanol acetate<br />
MTD maximum tolerated dose<br />
mTOR Mammalian target <strong>of</strong> rapamycin<br />
NNM N-nitrosodimethylamine<br />
NOAEL no observable adverse effect level<br />
NTP National Toxicology Program<br />
OTA ochratoxin A<br />
ROS reactive oxygen species<br />
TSC tuberous sclerosis complex<br />
UT urothelial tumors<br />
WHO World Health Organization<br />
WT wild type<br />
XIV
Journal Articles<br />
Publications and Presentations<br />
• Kerstin Stemmer, Heidrun Ellinger-Ziegelbauer, Hans-Juergen Ahr and Daniel R.<br />
Dietrich: Carcinogen Specific Gene Expression Pr<strong>of</strong>iles in Short-Term Treated Eker<br />
and Wild Type Rats Indicative for Pathways Involved in Renal Tumorigenesis.<br />
<strong>Cancer</strong> Research (2007) May 1;67(9):4052-68<br />
• Kerstin Stemmer, Heidrun Ellinger-Ziegelbauer, Kerstin Lotz, Hans-J. Ahr and<br />
Daniel R. Dietrich: Establishment <strong>of</strong> a Protocol for the Analysis <strong>of</strong> Laser<br />
Microdissected Rat <strong>Kidney</strong> Samples on Affymetrix Genechips. Toxicology and<br />
Applied Pharmacology, 217 (2006) 134-142.<br />
• Kerstin Stemmer, Heidrun Ellinger-Ziegelbauer, Hans-Juergen Ahr and Daniel R.<br />
Dietrich: Earliest preneoplastic renal lesions provide an in-depth understanding <strong>of</strong><br />
renal carcinogenesis (American Journal <strong>of</strong> Pathology, submitted)<br />
• Kerstin Stemmer, Heidrun Ellinger-Ziegelbauer, Katja Strauch, Leonard Collins,<br />
James A. Swenberg, Hans-Juergen Ahr and Daniel R. Dietrich: Renal Carcinogenic<br />
Response <strong>of</strong> Eker Rats to Low Doses <strong>of</strong> Methylazoxymethanol Acetate (Mutation<br />
Research, submitted)<br />
• Kerstin Stemmer, Diego Perez-Tilve, Daniel R. Dietrich, Matthias H. Tschöp and<br />
Paul T. Pfluger: High-fat diet exposure is associated with renal carcinogenesis in rats<br />
(manuscript to be submitted)<br />
Poster Presentations<br />
• K. Stemmer, P.T. Pfluger, D. Perez-Tilve, M.H. Tschöp and D.R. Dietrich: Long-term<br />
exposure to high-fat diet leads to renal pathology in rats: SOT 47th Annual Meeting,<br />
2008, Seattle, USA<br />
• D.R. Dietrich and K. Stemmer: Gender disparity in early ochratoxin A mediated<br />
toxicity: SOT 47th Annual Meeting, 2008, Seattle, USA<br />
XV
• K. Stemmer and D.R. Dietrich: Characterisation <strong>of</strong> Renal Preneoplastic Lesions in<br />
Carcinogen Treated and Untreated Eker Rats: International Congress <strong>of</strong> Toxicology,<br />
Toxicology Discovery Serving Society, July 2007 Montréal, Canada<br />
• K. Stemmer, H. Ellinger-Ziegelbauer, T. Lampertsdoerfer, K. Lotz, H.J. Ahr and D.R.<br />
Dietrich: Expression Pr<strong>of</strong>iles Induced By Renal Carcinogens in Eker Rats Compared<br />
to Wildtype Rats. SOT 45th Annual Meeting, 2006, San Diego, USA<br />
• K. Stemmer, J. Haehnlein, H. Ellinger-Ziegelbauer, H.J. Ahr and D.R. Dietrich:<br />
Identification <strong>of</strong> Potential Markers for an In Vitro Test System <strong>of</strong> Renal<br />
Carcinogenesis. 5th World Congress on Alternatives & Animal Use in the Life<br />
Sciences, 2005, Berlin, Germany<br />
• K. Stemmer, H. Ellinger-Ziegelbauer, T. Lampertsdoerfer, M. Thiel, H.J. Ahr and<br />
D.R. Dietrich: Ochratoxin A-Induced Gene Expression Deregulations in the <strong>Kidney</strong> <strong>of</strong><br />
Eker Rats Analyzed on Affymetrix Chips. SOT 44th Annual Meeting, 2005, New<br />
Orleans, USA<br />
• K. Stemmer, H. Ellinger-Ziegelbauer, T. Lampertsdoerfer, M. Thiel, H.J. Ahr and<br />
D.R. Dietrich: Expression Pr<strong>of</strong>iles Induced by Renal Carcinogens in Rat <strong>Kidney</strong><br />
Analysed on Affymetrix Chips. Joint Speciality Symposium on Renal Toxicology and<br />
Toxicologic Pathology, 2004, Lindau, Germany<br />
• J. Haehnlein, K. Stemmer, H. Ellinger-Ziegelbauer, S. Michel-Kaulmann, H.J. Ahr<br />
and D.R. Dietrich: Cytotoxic Effects <strong>of</strong> Aristolochic Acid (AA) on NRK-52E and<br />
Primary Rat <strong>Kidney</strong> Cells and Associated Changes in the Expression Pattern <strong>of</strong> p21<br />
and MGMT. Joint Speciality Symposium on Renal Toxicology and Toxicologic<br />
Pathology, 2004, Lindau, Germany<br />
XVI
Oral presentations<br />
• Früherkennung Nieren-kanzerogener Substanzen. Projektforum Biotechnologie<br />
German Federal Ministry <strong>of</strong> Research and Education (BMBF), 10.10.2007,<br />
BIOTECHNIKA Hannover, Germany.<br />
• The Use <strong>of</strong> Laser Microdissection and Pressure Catapulting to Study Gene<br />
Expression in Renal Preneoplastic Lesions. P.A.L.M MicroBeam Workshop:<br />
16.11.2006, P.A.L.M Microlaser Technologies AG, Bernried, Germany.<br />
• Gene Expression Patterns <strong>of</strong> Different Stages <strong>of</strong> Renal Pre-Neoplasia: Biolago<br />
Symposium: Science Meets Economy: 13.10.2006, University <strong>of</strong> Konstanz,<br />
Germany.<br />
Funding and Awards<br />
• Three months research visit at the University <strong>of</strong> Tel Aviv in Israel. Funded by the<br />
Young Scientist Exchange Program <strong>of</strong> the German Federal Ministry <strong>of</strong> Research and<br />
Education (BMBF) and Israel’s Ministry <strong>of</strong> Science, Culture and Sports (MOST),<br />
2008.<br />
• Three months PhD-scholarship <strong>of</strong> the Office <strong>of</strong> Equal Opportunity and Women‘s<br />
Affairs, University <strong>of</strong> Konstanz, 2008.<br />
• Congress Grant by the German Research Foundation (DFG), 2008<br />
• Research Grant: <strong>Molecular</strong> Analysis <strong>of</strong> the Nephrotoxic Action <strong>of</strong> the Mycotoxin<br />
Ochratoxin A. Environment and Living Foundation, University <strong>of</strong> Konstanz, 2006<br />
• Three months PhD-scholarship <strong>of</strong> the Senate Committee for Research (AFF) <strong>of</strong> the<br />
University <strong>of</strong> Konstanz, 2006.<br />
• Travel Award <strong>of</strong> the Society <strong>of</strong> Toxicology SOT, 2006<br />
• Finalist for the CVSS (Comparative & Veterinary Speciality Section) Student Poster<br />
Award, SOT 45th Annual Meeting in San Diego, USA, 2006<br />
XVII
Zusammenfassung<br />
Hintergrund: Die Mehrheit aller Krebserkrankungen, darunter auch Nierenkrebs, werden<br />
durch Fremdst<strong>of</strong>fe verursacht. Zur Untersuchung des möglichen kanzerogenen Potentials<br />
von Fremdst<strong>of</strong>fen sind zweijährige Kanzerogenesestudien in Nagetieren vom Gesetzgeber<br />
vorgeschrieben. In diesen Tests werden nach zweijähriger Exposition mit der jeweiligen<br />
Testsubstanz die Tumorinzidenz und weitere pathologische Effekte untersucht und<br />
anschliessend eine Risikoabschätzung für den Menschen erstellt. Derartige Studien sind<br />
jedoch äußerst zeitaufwendig und teuer. Hinzu kommt ein hoher Tierverbrauch sowie ein<br />
hoher Bedarf an Testsubstanzen für die zweijährige Dosierung. Erschwerend kommt hinzu,<br />
dass die Ergebnisse dieser Langzeit-Studien <strong>of</strong>t durch unspezifische Hochdosiseffekte<br />
oder durch eine altersabhängige Pathologie verfälscht werden. Zudem können Nagetierspezifische<br />
Effekte auftreten, die für eine Abschätzung der potentiellen Risiken für den<br />
Menschen nur wenig relevant sind, Umgekehrt vernachlässigen diese Studien jedoch<br />
andere Krebsrisik<strong>of</strong>aktoren, etwa eine hochkalorische Diät oder Fettleibigeit, die die<br />
kanzerogene Wirkung eines Fremdst<strong>of</strong>fes verstärken können. Die Entwicklung neuer,<br />
hochsensitiver sowie robuster Kurzzeit-In-Vivo-Testsysteme könnte bereits nach kurzen<br />
Expositionsdauern Aussagen über das kanzerogene Potential sowie den<br />
Wirkungsmechanismus einer Testsubstanz liefern, und zu einer verbesserte<br />
Risikoabschätzung für den Menschen beitragen.<br />
Ziele: Im Rahmen dieser Doktorarbeit sollte untersucht werden, ob Nieren-kanzerogene<br />
Substanzen in Kurzeit-In-Vivo-Studien durch sensitive Microarray-Transkriptomanalysen<br />
identifiziert werden können. Des weiteren sollten molekulare Mechanismen bei der<br />
Entwicklung von Fremdst<strong>of</strong>f-induzierten Nierentumoren charakterisiert werden. Schließlich<br />
sollten die Einflüsse einer fetthaltigen Diät sowie erhöhter Fettleibigkeit auf frühe Stadien<br />
von Nierenkrebs in der Ratte bestimmt werden.<br />
Experimenteller Aufbau: I) Wildtyp Ratten sowie Eker Ratten mit einer Mutation im TSC2<br />
Tumorsuppressor Gen wurden täglich mit den genotoxischen Nierenkanzerogenen<br />
Aristolochia Säure (AA), Methylazoxymethanol Azetat (MAMAc), oder dem nicht<br />
genotoxischen Nierenkanzerogen Ochratoxin A (OTA) gavagiert. Nach 1, 3, 7 und 14<br />
Tagen wurden von Nierenkortex-Homogenaten aller behandelten und Kontroll-Tiere<br />
Genexpressionspr<strong>of</strong>ile mittels Microarrays erstellt. Paraffinschnitte wurden zur<br />
Untersuchung von Histopathologie und Zellproliferation verwendet. II) Eker Ratten wurden<br />
XVIII
mit AA, MAMAc und OTA über 3 und 6 Monate gavagiert und histopathologische<br />
Veränderungen und Zellproliferationraten bestimmt. Parallel hierzu wurde ein Protokoll<br />
entwickelt, welches erstmals reproduzierbare und verlässliche Genexpressionsanaysen<br />
von mikrodissektierten Nierentubuli ermöglichte: Nach diesem Protokoll wurden<br />
Transkriptionsanalysen aus frühen Vortumorstadien (Prä-neoplasien) sowie aus gesunden<br />
Nierentubuli AA- und OTA- und Vehikel-behandelter Eker Ratten erstellt. Die zentralen<br />
Ergebnisse dieser Genexpressionsanalysen wurden mittels immunhistochemischen<br />
Färbungen verifiziert. III) Adverse Effekte auf Nierenpathologie und zelluläre Signalwege<br />
wurden in zwei Subpopulationen von Wistar-Ratten mit hoher Sensitivitaet bzw. Resistenz<br />
gegenüber nahrungsinduzierter Adipositas nach 11 Monaten Exposition mit fetthaltiger<br />
(45%) Diät, sowie in Wistar-Ratten nach Exposition mit fettarmer (6%) Diät untersucht.<br />
Ergebnisse & Diskussion: Histopatholgische Untersuchungen sowie v.a. die Auswertung<br />
der Genexpressionspr<strong>of</strong>ile aus Nierenkortex-Homogenaten zeigten, dass früheste<br />
substanzspezifische Effekte Nieren-kanzerogener Substanzen bereits nach 1- bis 14tägiger<br />
Exposition mit AA, MAMAc bzw. OTA erkennbar sind. Des weiteren erlaubten<br />
besagte Genexpressionspr<strong>of</strong>ile auch die Unterscheidung von genotoxischen und nichtgenotoxischen<br />
Substanzen. Diese Veränderungen der Genexpression nach 1-14 Tagen<br />
Exposition mit AA, MAMA oder OTA waren zudem prediktiv für die substanzspezifische<br />
Inzidenz und Zahl von prä-neoplastischen Läsionen in Eker Ratten nach 3- bzw. 6monatiger<br />
Exposition.<br />
Genexpressionspr<strong>of</strong>ile aus mikrodissektierten präneoplastischen Läsionen von Eker Ratten<br />
nach 6-monatiger Exposition mit AA, OTA oder dem entsprechenden Vehikel zeigten im<br />
Vergleich zu mikrodissektierten gesunden Tubuli von Vehikel behandelten Ratten eine<br />
stark erhöhte Deregulation zahlreicher Gene. Interessanterweise ähnelten sich die<br />
Expressionveränderungen in den verschiedenen Stadien der Läsionen und nach<br />
Behandlung mit den verschiedenen Kanzerogenen stark. Des weiteren konnten nur<br />
geringfügige Genexpressionveränderungen in gesunden Tubuli AA- und OTA-behandelter<br />
Ratten gemessen werden, was auf deren Insensitivität gegenüber kanzerogenen<br />
Substanzen hindeutetete. Diese Daten weisen darauf hin, dass die Anzahl der<br />
präneoplastischen Läsionen, die während einer begrenzten kritischen Phase der<br />
Substanzbehandlung entstehen, ausschlaggebend für die Inzidenz und Anzahl von<br />
Tumoren am Ende der zweijährigen Kanzerogenese Studie sein könnte. Die klonale<br />
Expansion dieser Läsionen ist demgegenüber eher ein substanzunabhängiger Prozess, der<br />
möglicherweise durch eine gestörte Feedback-Inhibitierung des mTOR Signalwegs<br />
verstärkt wird.<br />
XIX
Im letzten Teil der Dissertation konnte gezeigt werden, dass eine stark fetthaltige Diät eine<br />
deutlich schädliche Wirkung auf Rattennieren ausübt. Hierbei stellte sich jedoch heraus,<br />
dass der Grad der Adipositas nur eine untergeordnete Rolle einnimmt im Vergleich zu den<br />
direkten adversen Effekten der ueber die Nahrung aufgenommenen Lipide.<br />
Immunohistochemische Färbungen der Nieren wiesen zudem auf eine bedeutende Rolle<br />
des mTOR Signalweges bei der Lipid-induzierten preneoplastischen Pathologie der Niere.<br />
Fazit: Kurzzeit-In-Vivo-Studien in Kombination mit Microarrays erlauben die<br />
Charakterisierung von frühesten molekularen Vorgängen bei der Entstehung von<br />
Fremdst<strong>of</strong>f-induzierten Nierentumoren. Des weiteren erfüllen sie alle wichtigen<br />
Vorraussetzungen für eine verlässliche Identifizierung von Nieren-kanzerogenene<br />
Substanzen, nämlich eine hohe Sensitivität, Spezifität und Prediktivität. Die Ergebnisse<br />
dieser Arbeit zeigen, dass unter Verwendung neuer und sensitiver Detektionstechniken<br />
z,B. Microarrays, zweijährige Kanzerogenesestudien deutlich verkürzt werden könnten. Auf<br />
Basis der hier gezeigten Erkenntnisse erscheint es deshalb durchaus möglich, neue<br />
verbesserte Kanzerogenese-In-Vivo-Tests zu entwickeln, die zudem zu einer Reduktion<br />
der Tierzahl führen und weiterhin molekulare Grundlagen schaffen, die zukünftig zur<br />
Schaffung tierversuchsfreier Methoden dienen könnten.<br />
XX
Summary<br />
Background: The vast majority <strong>of</strong> all human cancers, including kidney cancer, are caused<br />
by xenobiotica. Two-year rodent bioassays play the central role in evaluating the<br />
carcinogenicity <strong>of</strong> chemicals and are mainly based on a post-mortem evaluation <strong>of</strong> tumor<br />
incidences and other pathological changes after two years <strong>of</strong> exposure. However, such<br />
lifetime bioassays are time-consuming and costly, require large numbers <strong>of</strong> animals and a<br />
high amount <strong>of</strong> compound for continuous dosing over the entire period <strong>of</strong> the assay. In<br />
addition, widespread age-related pathologies, unspecific high-dose effects, or rodentspecific<br />
carcinogenic effects with little or no relevance to humans can confound the<br />
outcome <strong>of</strong> these studies. Furthermore, they do not account for possible additive effects <strong>of</strong><br />
other cancer risk factors, such as diet-induced obesity. Therefore, interest in developing<br />
short, sensitive but robust, and mechanistically based in vivo studies has grown, allowing a<br />
more reliable human risk assessment.<br />
Aims: The overall aim <strong>of</strong> this thesis was to elucidate whether short-term in vivo assays in<br />
combination with microarray technology can be used to identify renal carcinogens. In<br />
addition, a more detailed understanding <strong>of</strong> potential additive effects <strong>of</strong> high fat diet<br />
exposure, and molecular mechanisms underlying chemically induced renal carcinogenesis,<br />
should be obtained.<br />
Experimental Setup: I) TSC2-mutant Eker and wild type rats were gavaged daily with the<br />
genotoxic renal carcinogens aristolochic acid (AA) and methylazoxymethanol acetate<br />
(MAMAc), or with the non-genotoxic ochratoxin A (OTA), respectively. Subsequent 1, 3, 7<br />
and 14 days <strong>of</strong> exposure, gene-expression pr<strong>of</strong>iles from kidney cortex homogenates,<br />
histopathology, and cell proliferation were assessed. II) Eker rats were gavaged with AA,<br />
MAMAc and OTA for 3 or 6 months. Subsequently, renal histopathology and cell<br />
proliferation were evaluated. In addition, a novel protocol for reproducible and reliable gene<br />
expression analysis from laser-microdissected pre-neoplastic renal lesions was<br />
established, and gene expression pr<strong>of</strong>iles <strong>of</strong> different stages <strong>of</strong> preneoplastic lesions and<br />
healthy tubules were examined from 6 months AA-, OTA- and vehicle-treated male Eker<br />
rats. Key findings from transcriptome analyses were verified by immunohistochemistry. III)<br />
The respective impacts <strong>of</strong> dietary lipids and high body adiposity on renal pathology and<br />
pathways involved in renal cancer were delineated in diet-induced obesity-sensitive and<br />
XXI
diet-induced obesity-resistant subpopulations <strong>of</strong> male Wistar rats vs. chow-fed controls,<br />
subsequent to 11 months <strong>of</strong> high fat diet or standard chow exposure, respectively.<br />
Results & Discussion: Renal histopathology and gene expression pr<strong>of</strong>iling <strong>of</strong> kidney<br />
homogenates <strong>of</strong> short-term AA-, MAMAc-, and OTA-treated rats revealed that microarray<br />
technology is a sensitive tool to study earliest events <strong>of</strong> renal carcinogenesis. In addition,<br />
gene expression pr<strong>of</strong>iles allowed dissecting genotoxic from non-genotoxic modes <strong>of</strong> action.<br />
Most importantly, early gene expression changes after short-term exposure were predictive<br />
for the incidence and number <strong>of</strong> preneoplastic lesions after 3 or 6 months <strong>of</strong> exposure.<br />
Gene expression pr<strong>of</strong>iles from microdissected preneoplastic lesions <strong>of</strong> 6-month AA-, OTAor<br />
control rats revealed marked gene deregulations, when compared to healthy tissue <strong>of</strong><br />
control animals. Notably, gene expression changes were similar in different lesion<br />
progression stages and after treatment with genotoxic or non-genotoxic carcinogens.<br />
Furthermore, gene expression pr<strong>of</strong>iles <strong>of</strong> microdissected healthy tubules <strong>of</strong> AA- and OTAtreated<br />
rats were only marginally changed, suggesting a low compound-sensitivity. These<br />
findings suggested that the number <strong>of</strong> preneoplastic lesions initially formed during a critical<br />
period <strong>of</strong> exposure may primarily drive the incidence and number <strong>of</strong> solid tumors. In<br />
contrast, clonal expansion <strong>of</strong> these lesions may be compound-independent and instead be<br />
driven by a disturbed feedback inhibition <strong>of</strong> the mTOR pathway.<br />
Last, high fat diet exposure <strong>of</strong> male Wistar rats revealed a clear adverse effect <strong>of</strong> dietary<br />
lipids, and a minor role <strong>of</strong> the degree <strong>of</strong> body adiposity, on renal morphology. In addition,<br />
immunohistochemical analyses suggested a major role <strong>of</strong> the mTOR pathway in dietary<br />
lipid-induced pre-neoplastic pathology.<br />
Conclusion: Microarray-based short-term in vivo assays not only help to delineate<br />
molecular events important for renal cancer, but also fulfill the most important prerequisites<br />
for reliable carcinogenicity testing, i.e. high sensitivity, high specificity, and high predictivity.<br />
Thus, the duration <strong>of</strong> the standard two-year rodent bioassay can be reduced to a shorter<br />
period <strong>of</strong> time by using novel and sensitive methodology such as microarrays. Ultimately,<br />
this thesis provides encouraging first results that may help to reduce, refine and one day<br />
even replace animal experiments.<br />
XXII
1.1 <strong>Kidney</strong> <strong>Cancer</strong><br />
Chapter 1: Introduction<br />
<strong>Kidney</strong> cancer accounts for 2% <strong>of</strong> all malignancies worldwide, with approximately 210.000<br />
additional cases per year. Because <strong>of</strong> its high mortality and increasing incidence in most<br />
countries, it is considered to be one <strong>of</strong> the most important urologic cancers. The world’s<br />
highest incidence <strong>of</strong> kidney cancer is observed in Eastern European countries (Czech<br />
Republic, Estonia and Hungary). Germany has the 6th highest incidence with 13.2 cases<br />
and 6.3 mortalities per 100.000 in men and 5.9 cases and 2.9 mortalities per 100.000 in<br />
women (IARC 2002c). <strong>Kidney</strong> tumors usually do not cause signs or symptoms and <strong>of</strong>ten<br />
remain unrecognized in their early and treatable stages. Advanced stages and metastases<br />
are therefore more common with kidney cancer than with many other cancer types.<br />
Secondary tumors <strong>of</strong> the kidney are typically found in the lung, s<strong>of</strong>t tissue, bone, liver,<br />
cutaneous sites and the central nervous system (Corgna et al. 2007).<br />
1.1.1 Risk Factors for <strong>Kidney</strong> <strong>Cancer</strong><br />
The majority <strong>of</strong> human kidney tumors occur sporadically and only 1-4% <strong>of</strong> all kidney cancer<br />
cases are caused by an inherited predisposition (Pavlovich and Schmidt 2004). Sporadic<br />
tumors <strong>of</strong> the kidneys occur more frequently in men than in women (McLaughlin and<br />
Lipworth 2000). The underlying causes for the sex-specific differences are largely unknown,<br />
but may be due to differences in metabolism or hormonal balance. In both sexes kidney<br />
cancer increases with age and more than 75% <strong>of</strong> all cases occur over the age <strong>of</strong> 60.<br />
Besides sex and age, epidemiologic studies point to several other risk factors for renal<br />
cancer, including smoking, prolonged dialysis, hypertension, or the use <strong>of</strong> diuretics and<br />
other anti-hypertensive medication (McLaughlin and Lipworth 2000). In addition, 1/4 <strong>of</strong> all<br />
human kidney cancers were attributed to excess body weight (Mellemgaard et al. 1995).<br />
However, the underlying causal mechanisms that link the respective risk factors to kidney<br />
cancer still remain elusive.<br />
1
Chapter 1: Introduction<br />
Due to their predominant role in excretion, the kidneys are highly vulnerable to a variety <strong>of</strong><br />
circulating toxicants. Approximately one third <strong>of</strong> the plasma water reaching the kidney is<br />
filtered by the glomeruli and 98-99% <strong>of</strong> that filtered plasma water is reabsorbed from the<br />
tubular lumen (Casarett and Doull 1986). Consequently, any drug or chemical that is<br />
present in non-toxic concentrations in the blood can reach the renal epithelial cells in much<br />
higher concentrations after being concentrated with the urine and after being reabsorbed<br />
into the cells.<br />
1.1.2 Genetic and Epigenetic Causes <strong>of</strong> <strong>Cancer</strong><br />
I: Tumor suppressor genes and oncogenes<br />
<strong>Cancer</strong> is usually caused by mutations that are typically found in two classes <strong>of</strong> genes:<br />
<strong>Cancer</strong>-promoting oncogenes and cancer-preventing tumor suppressor genes. Gain <strong>of</strong><br />
function mutations in oncogenes, which are <strong>of</strong>ten involved in signal transduction and<br />
execution <strong>of</strong> mitogenic signals, result in a hyperactive growth and proliferation <strong>of</strong> cells<br />
(Todd and Wong 1999). Loss <strong>of</strong> function mutations in tumor suppressor genes generally<br />
result in dysregulation <strong>of</strong> the cell cycle and DNA-replication, the inhibition <strong>of</strong> programmed<br />
cell death (apoptosis) (Malumbres and Barbacid 2001), or disturbed interaction <strong>of</strong> tumor<br />
cells with protective cells <strong>of</strong> the immune system (immunoresistance) (Parsa et al. 2007).<br />
In 1975, Alfred Knudson and colleagues proposed the two-hit hypothesis, stating that both<br />
copies (alleles) <strong>of</strong> a tumor suppressor gene must be lost or mutated for cancer to occur<br />
(Knudson 1975). Although widely accepted, certain exceptions to the 'two hit' rule were<br />
found. For instance, tumor suppressor genes like p27Kip1 might exhibit haploinsufficiency<br />
after a mutation <strong>of</strong> a single allele, when the remaining functional allele does not translate<br />
into enough protein to preserve a wild-type phenotype (Fero et al. 1998). For oncogenes,<br />
gain <strong>of</strong> function mutations in one allele are usually enough to trigger cancer development<br />
(Todd and Wong 1999).<br />
Besides mutations in tumor suppressor genes and oncogenes, tumorigenesis can be<br />
supported by any event that accelerates the spontaneous rate <strong>of</strong> genetic alterations in the<br />
cell. For instance, hereditary or spontaneous damage in the cellular DNA damage repair<br />
system may favor the accumulation <strong>of</strong> additional mutations, and genetic or genomic<br />
instability <strong>of</strong> cells (Schmutte and Fishel 1999; Jefford and Irminger-Finger 2006).<br />
2
II: Epigenetic alterations<br />
Chapter 1: Introduction<br />
Most recently, several studies pointed towards a critical role <strong>of</strong> reversible epigenetic<br />
alterations in carcinogenesis (Baylin and Herman 2000, Sparmann, 2006, Vucic, 2008).<br />
Epigenetic regulation is generally organized at the level <strong>of</strong> DNA (post-replicative<br />
methylation), RNA (RNA interference), and protein (histone modifications and polycomb<br />
group protein complexes) (Momparler 2003).<br />
Abnormal patterns <strong>of</strong> DNA methylation in cancer cells have been recognized for more than<br />
20 years and global hypomethylation and region-specific hypermethylation are frequent<br />
events in tumor cells (Baylin and Herman 2000). The biological significance <strong>of</strong> DNA<br />
hypomethylation in cancer is less understood. Hypermethylation mainly occurs at CpG<br />
islands within the promoter region <strong>of</strong> a gene, resulting in its suppressed transcription. The<br />
current model suggests that promoter hypermethylation and transcriptional silencing <strong>of</strong><br />
tumor suppressor or DNA repair genes and mutations within these genes occur at least<br />
with a comparable frequency.<br />
Besides DNA methylation, alteration in histone acetylation plays another important role in<br />
the regulation <strong>of</strong> gene expression. Inactive chromatin mediated by histone deacetylation<br />
was demonstrated as one critical component in the silencing <strong>of</strong> tumor suppressor genes.<br />
Different histone deacetylase inhibitors are currently in clinical trials for the treatment <strong>of</strong><br />
cancer (Martinez-Iglesias et al. 2008).<br />
Although still much has to be learned, epigenetic regulation is already assumed to be<br />
equally important for the onset <strong>of</strong> cancer as genetic alterations. The interplay between<br />
genetic and epigenetic changes during tumor progression is thus becoming a major focus<br />
<strong>of</strong> cancer research.<br />
1.1.3 Models <strong>of</strong> Carcinogenesis<br />
I: Multistage model <strong>of</strong> carcinogenesis<br />
The process <strong>of</strong> carcinogenesis is widely accepted as a multistage process, with a sequence<br />
<strong>of</strong> multiple genetic events preceding the onset <strong>of</strong> cancer (Armitage and Doll 1954; Barrett<br />
and Wiseman 1987). The process is based on genetic alterations that have to be fixed<br />
during DNA replication, leading to growth advantage <strong>of</strong> the initiated cell (tumor initiation).<br />
Clonal expansion <strong>of</strong> these cells may lead to the development <strong>of</strong> pre-neoplastic foci with<br />
<strong>characteristics</strong> that favor genetic instability (tumor promotion). Further accumulation <strong>of</strong><br />
3
Chapter 1: Introduction<br />
additional mutations in the fast-growing cells can result in the development <strong>of</strong> a malignant<br />
phenotype that enables the cells to invade the surrounding tissue (tumor progression).<br />
Such a genetic multistage model has been described by Fearon and Vogelstein for<br />
colorectal cancer, where a linear sequence <strong>of</strong> specific genetic events can be correlated with<br />
evolving pathophysiological stages <strong>of</strong> colon carcinogenesis (Fearon and Vogelstein 1990).<br />
Ten years later the linear model <strong>of</strong> multistage carcinogenesis was elaborated by Hanahan<br />
and Weinberg (Hanahan and Weinberg 2000), who described six essential changes in cell<br />
physiology that define the development <strong>of</strong> malignant tumors: (1) self-sufficiency in growth<br />
signals, (2) insensitivity to growth inhibitory signals, (3) evasion <strong>of</strong> apoptosis, (4) limitless<br />
replicative potential, (5) sustained angiogenesis and (6) tissue invasion and metastasis. In<br />
contrast to the Fearon-Vogelstein model, the Hanahan–Weinberg model does not depend<br />
on the sequential accumulation <strong>of</strong> specific mutations for a distinct tumor type but instead<br />
depends on a set <strong>of</strong> functional changes that can result from many different genetic or<br />
epigenetic alterations, independent <strong>of</strong> cell-type, species and aetiology.<br />
II: <strong>Cancer</strong> stem cell model<br />
During recent years, the stem cell model <strong>of</strong> carcinogenesis gained major attention. The<br />
American Association for <strong>Cancer</strong> Research Stem Cell Workshop defined a cancer stem cell<br />
as a cell within a tumor that possesses the capacity to self-renew. Thus, heterogeneous<br />
lineages <strong>of</strong> cancer cells that comprise the tumor may originate from one cancer stem cell.<br />
In this model cancer cells may arise from tissue stem cells that have acquired mutations<br />
that render them cancerous. <strong>Cancer</strong> cells may also arise from differentiated progenitor cells<br />
that "re-acquired" stem cell like properties due to mutations (Clarke et al. 2006). <strong>Cancer</strong><br />
stem cells were originally identified in leukemia (Bonnet and Dick 1997) and later in solid<br />
brain (Singh et al. 2003), breast (Al-Hajj et al. 2003), and colon tumors (O'Brien et al.<br />
2007). However, to date cancer stem cells have not been detected in kidney tumors.<br />
1.1.4 Renal Tumor Pathology<br />
Renal toxicology and pathology are closely related to the kidney’s complex functional<br />
anatomy, with the cortex forming the outer part enclosing the renal medulla and papilla<br />
(Figure 1.1). The functional unit <strong>of</strong> the kidney is the nephron, consisting <strong>of</strong> the glomerulus,<br />
proximal tubule, loop <strong>of</strong> Henle, distal tubule and collecting duct. The proximal tubule is<br />
furthermore dissected into three segments: The tubulus contortus proximalis, consisting <strong>of</strong><br />
the post-glomerular portion (P1), and the more distal portion (P2) <strong>of</strong> the proximal tubule,<br />
4
Chapter 1: Introduction<br />
containing an active endocytosis/lysosomal apparatus, thus representing the site <strong>of</strong> injury<br />
related to lysosomal overload, as well as to protein-bound toxic moieties (Crist<strong>of</strong>ori et al.<br />
2007). The following straight portion <strong>of</strong> the P3 segment (tubulus rectus proximalis)<br />
represents the most susceptible site <strong>of</strong> injury via metabolic activation <strong>of</strong> xenobiotica and via<br />
their transporter-associated accumulation (Crist<strong>of</strong>ori et al. 2007).<br />
Figure 1.1: Overview <strong>of</strong> the morphological structures <strong>of</strong> the renal nephron. Source:<br />
http://de.wikipedia.org/wiki/Nephron.<br />
I: <strong>Kidney</strong> Tumors in Humans<br />
The majority <strong>of</strong> kidney tumors in humans (80-85%) are renal cell carcinoma (RCC),<br />
originating from the renal parenchyma. The remaining 15-20% are mainly transitional cell<br />
carcinoma <strong>of</strong> the renal pelvis, while other neoplasms, e.g. angiomyolipoma or<br />
nephroblastoma (Wilms' Tumor), are rare (Motzer et al. 1996).<br />
RCC are not a single entity but comprise a heterogeneous group <strong>of</strong> tumors that range from<br />
entirely benign to aggressively malignant. They were sub-classified according to their<br />
histological <strong>characteristics</strong> into clear cell RCC (70%), papillary RCC (15%), chromophobe<br />
RCC (10%), oncocytoma (5%) and collecting duct carcinoma (
Chapter 1: Introduction<br />
to correlate with distinct histological sub-types in sporadic RCC, which allowed a more indepth<br />
tumor classification and improved diagnostics (for further information see table 1.1)<br />
(Kovacs et al. 1997; Pavlovich and Schmidt 2004).<br />
Table 1.1: Classification <strong>of</strong> kidney tumors (adjusted from (Bjornsson et al. 1996; Dulaimi et al. 2004;<br />
Pavlovich and Schmidt 2004; Henske 2005))<br />
Histological<br />
type<br />
Presumed<br />
origin<br />
Clear cell RCC Proximal renal<br />
tubule<br />
Papillary RCC<br />
(basophilic<br />
and<br />
eosinophilic)<br />
Chromophobe<br />
RCC<br />
Proximal renal<br />
tubule<br />
Collecting duct<br />
(intercalated<br />
cells)<br />
Oncocytoma Collecting duct<br />
(intercalated<br />
cells)<br />
Collecting<br />
carcinoma<br />
Angiomyolipoma<br />
Frequent cytogenetic<br />
abnormalities in<br />
sporadic tumors *<br />
-3p, +5q, -Y, -8p, -9p, -<br />
14q, t(3;5)(p;q);<br />
+7, +17, -Y, +12, +16,<br />
+20<br />
-1, -2, -6, -10, -13, -17,<br />
-21<br />
-1, -Y, t(5;11)(q35;q13),<br />
t(9;11)(p23;q13)<br />
Collecting duct -1q32, -6p, -8p, -21q,<br />
highly aneuploid<br />
Mesenchymal<br />
cells (immature<br />
smooth muscle<br />
cells, fat cells)<br />
No specific<br />
chromosomal changes<br />
have been identified<br />
Frequent promoter<br />
hypermethylations in<br />
sporadic tumors<br />
VHL, RASSF1A,<br />
p16 INK4a , p14 ARF, APC,<br />
MGMT, GSTP1,<br />
RARß2, E-cadherin,<br />
TIMP3<br />
RASSF1A, p16 INK4a ,<br />
p14 ARF, APC, MGMT,<br />
GSTP1, RARß2, Ecadherin,<br />
TIMP3<br />
RASSF1A, p14 ARF,<br />
RARß2, TIMP3<br />
RASSF1A, p16 INK4a ,<br />
p14 ARF, APC, MGMT,<br />
GSTP1, TIMP3<br />
RASSF1A, p14 ARF,<br />
APC, MGMT, RARß2,<br />
TIMP3<br />
No epigenetic changes<br />
have been identified<br />
Hereditary genetic<br />
syndromes and<br />
implicated genes **<br />
Von Hippel-Lindau disease<br />
(VHL); Birt Hogg Dubé<br />
syndrome (BDH); Tuberous<br />
Sclerosis (TSC1, TSC2)<br />
Hereditary papillary renal<br />
carcinoma (c-MET);<br />
Hereditary leiomyomatosis<br />
renal cell cancer (FH);<br />
Hyperparathyroidism-jaw<br />
tumor (HRPT2); Tuberous<br />
Sclerosis (TSC1, TSC2)<br />
Birt Hogg Dubé syndrome<br />
(BDH); Tuberous Sclerosis<br />
(TSC1, TSC2)<br />
Birt Hogg Dubé syndrome<br />
(BDH); Tuberous Sclerosis<br />
(TSC1, TSC2)<br />
Hereditary leiomyomatosis<br />
renal cell cancer (FH)<br />
Tuberous Sclerosis (TSC1,<br />
TSC2)<br />
* Nomenclature cytogenetics: Loss <strong>of</strong> a chromosome (-); gain <strong>of</strong> a chromosome (+); translocation (t); short<br />
arm <strong>of</strong> a chromosome (p); long arm <strong>of</strong> a chromosome (q). ** Von-Hippel-Lindau (VHL); Birt Hogg Dubé<br />
(BDH), Fumarat hydratase (FH); Hyperparathroidism 2 (HRPT2); Tuberous Sclerosis 1 (TSC1), Tuberous<br />
Sclerosis 2 (TSC2)<br />
However, cytogenetic studies are mainly based on the analysis <strong>of</strong> gross tumors that may<br />
have gained numerous mutations after developing genetic instability. Thus, in most cases<br />
the initial genetic or epigenetic cause for the initiation <strong>of</strong> spontaneous kidney tumors<br />
remains elusive. However, studies <strong>of</strong> families with hereditary genetic syndromes have<br />
identified several predisposing genes involved in renal carcinogenesis. Due to the similar<br />
histology <strong>of</strong> hereditary and sporadic RCC, these genes were also suggested to be involved<br />
in the development <strong>of</strong> sporadic tumors (Table 1.1, right column).<br />
6
II: <strong>Kidney</strong> Tumors in Rats<br />
Chapter 1: Introduction<br />
Rats are the most common laboratory animal to study renal carcinogenesis. Although the<br />
incidence <strong>of</strong> spontaneous RCC in laboratory strains is very low (e.g. 0.08% in Sprague-<br />
Dawley and 0.28% in F344 rats) (Chandra et al. 1993), similar tumor types as those<br />
observed in humans have been experimentally induced in rodents by distinct chemicals,<br />
hormones, viruses and radiation (Hamilton 1975; Lock and Hard 2004). Studies in rodents<br />
not only allowed the analysis <strong>of</strong> the carcinogenic potential <strong>of</strong> numerous test compounds<br />
(see chapter 1.2.1) but also the identification <strong>of</strong> the earliest carcinogen-induced foci <strong>of</strong> cells<br />
(pre-neoplastic lesions) with alterations very similar to those found in end-stage tumors.<br />
From these findings it was concluded that kidney tumor development can occur in<br />
successive stages, from single atypical tubules to hyperplasia to adenoma and/or<br />
carcinoma (Figure 1.2) (Dietrich and Swenberg 1991). However, knowledge <strong>of</strong> the initial<br />
causes and the process <strong>of</strong> progression <strong>of</strong> chemically induced tumors is still limited. Current<br />
findings from rodent bioassays assume that not all carcinogen-induced pre-neoplastic<br />
lesions will progress to a neoplastic lesion and it is <strong>of</strong>ten not possible to distinguish lesions<br />
with a neoplastic potential from lesions that are self-limiting or even reversible, and thus<br />
may not have any significance for the testing <strong>of</strong> carcinogenicity (Alden and Kanerva 1982;<br />
Eustis 1989). To date, the potential contribution <strong>of</strong> pre-neoplastic lesions to carcinogeninduced<br />
tumor development is mainly assessed by histopathologic evaluation <strong>of</strong> similarities<br />
between compound-induced pre-neoplastic and neoplastic lesions, by the absence or<br />
presence <strong>of</strong> a morphological continuum between these lesions, and by the occurrence <strong>of</strong><br />
treatment-related cytotoxicity and regenerative cell proliferation (Eustis 1989).<br />
Common cytological alterations found in carcinogenic lesions in rats resemble those <strong>of</strong><br />
sporadic lesions in humans and are summarized in table 1.2. Both the chemical properties<br />
<strong>of</strong> the inducing compounds, as well as their metabolism and location <strong>of</strong> uptake within the<br />
different tubular segments, most likely determine the type and cellular origin <strong>of</strong> the lesions.<br />
For instance, clear cell/ acidophilic tumors in N-nitrosomorpholine (NNM)- treated rats were<br />
demonstrated to arise from the distal tubules and collecting ducts, which are also the main<br />
sites <strong>of</strong> NNM-induced acute cell damage (Bannasch et al. 1989; Nogueira et al. 1989).<br />
Similarly, the most common carcinogen-induced basophilic RCC as well as chromophobe<br />
RCC originate from the proximal tubules, which represent the most susceptible sites <strong>of</strong><br />
injury via metabolic activation <strong>of</strong> xenobiotica or their transporter-associated accumulation<br />
(e.g. ochratoxin A, fumonisine B1) (Boorman et al. 1992; Howard et al. 2001). Oncocytic<br />
cells were found to arise from the intercalated cells <strong>of</strong> the renal collecting duct <strong>of</strong> potassium<br />
7
Chapter 1: Introduction<br />
bicarbonate-, NNM-, streptomycin- or lead acetate-treated rats (Nogueira 1987; Mayer et al.<br />
1989; Lina and Kuijpers 2004).<br />
Preneoplastic and neoplastic lesions <strong>of</strong> mixed cellular phenotypes can occur spontaneously<br />
or subsequent to carcinogen treatment. However, it is not clear whether different cellular<br />
alterations occur by random hits in different tumor suppressor genes or oncogenes or by a<br />
transition process from one to another lesion type.<br />
Figure 1.2: Progression <strong>of</strong> renal lesions. Adjusted from Dietrich and Swenberg (Dietrich and<br />
Swenberg 1991). Microscopic images represent typical renal basophilic lesions <strong>of</strong><br />
Eker rats after exposure with renal carcinogens, and were obtained during the course<br />
<strong>of</strong> this thesis. For further information, see text.<br />
8
Chapter 1: Introduction<br />
Table 1.2: Representative images <strong>of</strong> frequent cellular abnormalities and biochemical features <strong>of</strong><br />
carcinogen-induced kidney tumors in rats. Adjusted from (Dietrich and Swenberg<br />
1991). Microscopic images kindly provided by Daniel R. Dietrich<br />
Histological type Morphological and cytological <strong>characteristics</strong> Representative image<br />
Clear cells Large translucent cytoplasm, irregular cell shape<br />
with small chromatin-dense nuclei. Cytoplasmatic<br />
accumulations <strong>of</strong> glycogen and lipids (Mayer et al.<br />
1988, Steinberg et al. 1992).<br />
Acidophilic<br />
(eosinophilic) cells<br />
Often associated with clear cells and are slightly<br />
larger than normal cells with a strongly eosinophilic<br />
cytoplasm<br />
Basophilic cells Basophilic cytoplasm. Increased abundance <strong>of</strong><br />
ribosomes, slightly enlarged nuclei with prominent<br />
nucleoli<br />
Chromophobic cells Enlarged cells with slightly translucent cytoplasm.<br />
Weak eosinophilic or basophilic staining. Extensive<br />
storage <strong>of</strong> mucopolysaccharides likely enclosed in<br />
cytoplasmatic vacuoles<br />
Oncocytic cells Large polygonal shape with acidophilic/<br />
eosinophilic and granular cytoplasm. Abundance<br />
<strong>of</strong> atypical mitochondria and a large content <strong>of</strong><br />
cytochrome c oxidase. Over-expression <strong>of</strong> GLUT1<br />
and hexokinase I.<br />
9
1.2 Renal Carcinogens<br />
Chapter 1: Introduction<br />
Based on the general assumption that multigenetic events are required for carcinogenesis,<br />
there are basically two possibilities by which a chemical carcinogen can increase the<br />
possibility <strong>of</strong> tumorigenesis within a tissue: 1) by increasing the rate <strong>of</strong> DNA or<br />
chromosomal damage, resulting in more errors per replication, or 2) by increasing the<br />
number <strong>of</strong> DNA replications and therefore the likelihood <strong>of</strong> spontaneous errors (Cohen and<br />
Arnold 2008). Initiated and clonally expanding cells might have different <strong>characteristics</strong> than<br />
their cell <strong>of</strong> origin and therefore might be differentially influenced by chemical carcinogens.<br />
For instance, it is not entirely clear whether progressing stages <strong>of</strong> transformed, neoplastic<br />
kidney cells are still capable <strong>of</strong> taking up or metabolizing carcinogens or rather<br />
dedifferentiate in favor <strong>of</strong> proliferation.<br />
Chemical carcinogens have been categorized according to two general mechanisms <strong>of</strong><br />
action into genotoxic and non-genotoxic carcinogens:<br />
1.2.1 Genotoxic Carcinogens<br />
Genotoxicity is induced by compounds that are able to directly interact with the DNA<br />
(mutagens), compounds that break chromosomes (clastogens) and compounds that<br />
damage the spindle apparatus (aneugens). Genotoxic carcinogens are either electrophilic<br />
molecules per se or are absorbed as stable pro-carcinogens that have to be bio-activated<br />
to their active electrophilic form, which can then covalently interact with any nucleophilic<br />
site in the cell, such as the DNA resulting in DNA adducts. Groups <strong>of</strong> genotoxins can<br />
therefore <strong>of</strong>ten be recognized by their structural features (e.g. alkenes, aromatic amines,<br />
nitrosamines or polycyclic aromatic hydrocarbons).<br />
According to their relatively simple mode <strong>of</strong> action, which involves a single chemical<br />
interaction with one specific target molecule, DNA adduct formation was expected as a<br />
linear function <strong>of</strong> the administered dose <strong>of</strong> the DNA-reactive compound. In fact most<br />
carcinogenicity studies that have used low doses <strong>of</strong> genotoxic compounds have been<br />
interpreted as showing no threshold (Lijinsky et al. 1988; Peto et al. 1991). It was therefore<br />
concluded that even the smallest doses <strong>of</strong> DNA-reactive compounds increase susceptibility<br />
to cancer and that a zero-risk or safe-dose may not be defined. However, it is now<br />
recognized that many factors e.g. efficiency <strong>of</strong> absorption, distribution, metabolism, DNA<br />
damage repair and mutation fixation, can affect the linear shape <strong>of</strong> the dose-response<br />
10
Chapter 1: Introduction<br />
curve, leading to a sublinear and supralinear curve progression (Williams et al. 1995;<br />
Swenberg et al. 2008). Due to these findings it is now frequently discussed, whether a<br />
threshold or no-observable-adverse-effect-level (NOAEL) may exist at least for some<br />
genotoxic compounds (Williams et al. 1995; Oesch et al. 2000; Hengstler et al. 2003).<br />
1.2.2 Non-Genotoxic Carcinogens<br />
Non-genotoxic carcinogens are a heterogeneous group <strong>of</strong> carcinogens that are not directly<br />
DNA-reactive. However, they are capable <strong>of</strong> inducing epigenetic effects on cells that 1)<br />
either indirectly result in DNA modification and subsequently pro-carcinogenic mutations, or<br />
2) promote the development <strong>of</strong> initiated cells into neoplasms. Numerous tumor promoting<br />
mechanisms have been identified for a wide range <strong>of</strong> non-genotoxic compounds. Many <strong>of</strong><br />
these are cell type-, tissue-, sex- or species-dependent and usually require sustained and<br />
high (above threshold) exposures to induce a carcinogenic effect. For kidney<br />
carcinogenesis non-genotoxic carcinogens can be categorized into three basic groups:<br />
I: Cytotoxic Carcinogens<br />
A number <strong>of</strong> non-genotoxic carcinogens were demonstrated to induce tumors at high<br />
cytotoxic doses. It is assumed that exposure to cytotoxic carcinogens results in tissue<br />
damage, and the subsequent regenerative cell proliferation may either contribute to the<br />
fixation <strong>of</strong> spontaneous mutations, or to the promotion <strong>of</strong> naturally occurring tumors in<br />
tissues with a high spontaneous tumor rate (Dietrich and Swenberg 1990; Cohen and<br />
Arnold 2008). One <strong>of</strong> the best studied species-, sex- and tissue-specific mechanisms is the<br />
chemically induced α2u-globulin nephropathy, which is associated with the onset <strong>of</strong> renal<br />
neoplasms in male rats (Swenberg 1993). α2u-globulin is a plasma protein that binds<br />
hydrophobic small molecules for subsequent absorption and lysosomal degradation by the<br />
P2 segment <strong>of</strong> the renal proximal tubule. Binding <strong>of</strong> xenobiotica, e.g. unleaded petrol,<br />
decalin or d-limonene, may result in a structural alteration <strong>of</strong> α2u-globulin and its protection<br />
from lysosomal degradation. Subsequent accumulation <strong>of</strong> lysosomal protein droplets in the<br />
kidney frequently results in tubular necrosis, compensatory cell proliferation (regeneration),<br />
and an increased chance for the fixation <strong>of</strong> spontaneous genetic damage and therefore the<br />
onset <strong>of</strong> renal tumorigenesis (Dietrich and Swenberg 1990; Swenberg and Lehman-<br />
McKeeman 1999; Dill et al. 2003).<br />
11
II: Non-Cytotoxic Carcinogens<br />
Chapter 1: Introduction<br />
Non-cytotoxic carcinogens exert their tumor promoting potential without causing prior cell<br />
loss. Instead, they are able to influence a variety <strong>of</strong> cellular signaling pathways that are<br />
<strong>of</strong>ten controlled by tumor suppressor genes and proto-oncogenes. Non-cytotoxic<br />
carcinogens may cause a cell to commence cell division by triggering mitosis or inhibiting<br />
apoptosis. Such ‘mitogens’ <strong>of</strong>ten mimic the activation <strong>of</strong> extracellular growth factors by<br />
either directly activating the corresponding receptor, or by modulating downstream<br />
signaling events through interaction with kinases, phosphatases, scaffold proteins or<br />
chaperons, ultimately inducing mitosis (Perona 2006; Lawrence et al. 2008). Other noncytotoxic<br />
carcinogens can inhibit intercellular communication via gap junctions and affect<br />
cellular growth regulation and differentiation (Chipman et al. 2003; Leithe et al. 2006).<br />
III: Indirect genotoxic carcinogens<br />
The induction <strong>of</strong> oxidative stress has been proposed as one possible mechanism <strong>of</strong> action<br />
for non-genotoxic carcinogens, either by increasing the production <strong>of</strong> reactive oxygen<br />
species (ROS), or by inhibiting the antioxidant capability <strong>of</strong> the target cell. ROS can not only<br />
modify cellular lipids and proteins but also directly oxidize and damage the DNA. The best<br />
studied oxidative DNA lesion is 8-hydroxydeoxyguanosine (OH8dG) which occurs from<br />
oxidation <strong>of</strong> guanine at the C8 position within the DNA, or from oxidation <strong>of</strong> dGTP and its<br />
later incorporation into the DNA. Both can result in site-specific mutagenesis (Shibutani et<br />
al. 1991; Hussain and Harris 1998). Recent evidence shows that ROS and oxidative DNA<br />
damage can modulate DNA methylation patterns (Turk et al. 1995) which may result in<br />
epigenetic silencing <strong>of</strong> tumor suppressor genes or activation <strong>of</strong> oncogenes. Besides their<br />
role in tumor initiation, ROS were further shown to be involved in tumor promotion and<br />
progression. High doses <strong>of</strong> cellular ROS or ROS-promoting agents are clearly cytotoxic and<br />
can result in regenerative cell proliferation and fixation <strong>of</strong> mutations. Low doses <strong>of</strong> ROS<br />
may contribute to abnormal gene expression, cell signaling and alteration <strong>of</strong> second<br />
messenger systems. For instance, ROS can activate both mitogen-activated protein (MAP)<br />
kinase/AP-1 and NFkappa-B pathways, which are critically involved in cell proliferation and<br />
apoptosis (Klaunig and Kamendulis 2004).<br />
12
1.3 Renal Carcinogens Relevant for this Study<br />
Chapter 1: Introduction<br />
For this thesis, three naturally occurring renal carcinogens were used and the following<br />
chapter will give an overview <strong>of</strong> the chemical and biological properties <strong>of</strong> the respective<br />
compounds.<br />
1.3.1 Aristolochic acid<br />
Aristolochic acid (AA) represents a mixture <strong>of</strong> aristolochic acid I (AAI) and aristolochic acid<br />
II (AAII) (Figure 1.3) and is primarily found in the Aristolochia and Asarum species <strong>of</strong> the<br />
Aristolochiaceae family <strong>of</strong> plants. It can be found in a number <strong>of</strong> botanical products such as<br />
traditional medicines, dietary supplements and weight loss remedies.<br />
Figure 1.3: Chemical structures <strong>of</strong> aristolochic acid I and aristolochic acid II, from<br />
(Balachandran et al. 2005)<br />
I: Acute and Chronic Toxicity<br />
The kidneys are the primary target organ for acute and chronic AA toxicity in both<br />
laboratory animals (rats, mice and rabbits) and humans. In rats, the oral LD50 (the quantity<br />
<strong>of</strong> a compound per kg per body weight (bw) that kills 50% <strong>of</strong> the study population) was<br />
reported to be 203.4 mg/kg bw in males and 183.9 mg/kg bw in females (Mengs 1987).<br />
Rats treated with a single or continuous high oral doses 1-10mg/kg BW) <strong>of</strong> AA developed<br />
renal failure with tubular necrosis <strong>of</strong> the corticomedullary junction (Mengs 1987; Mengs and<br />
Stotzem 1993), and had elevated plasma creatinine and urea levels (Mengs and Stotzem<br />
1993; Liu et al. 2003; Cui et al. 2005). In addition, in some studies interstitial fibrosis was<br />
observed (Debelle et al. 2002; Sun et al. 2006), although this finding could not be replicated<br />
13
Chapter 1: Introduction<br />
by others (Cosyns et al. 1998; Cui et al. 2005). In humans, consumption <strong>of</strong> AA-containing<br />
botanical products has been associated with the onset <strong>of</strong> aristolochic acid nephropathy,<br />
characterized by mild tubular proteinuria, extensive interstitial fibrosis, tubular atrophy,<br />
global sclerosis <strong>of</strong> glomeruli, and rapid progression to renal failure (Nortier and<br />
Vanherweghem 2002; Cosyns 2003).<br />
II: Carcinogenicity<br />
AA is mutagenic and carcinogenic in mice, rats, rabbits and humans (Arlt et al. 2002).<br />
Recently, AA-containing products were classified as carcinogenic to humans (Group 1) by<br />
the International Agency for Research on <strong>Cancer</strong> (IARC) (IARC 2002b). In rats, application<br />
<strong>of</strong> high oral doses (50mg/kg bw) for three days resulted in increased incidences <strong>of</strong><br />
preneoplastic and neoplastic lesions <strong>of</strong> the kidney after a latency <strong>of</strong> 6 months (Cui et al.<br />
2005). In another study, continuous gavage with 1-10mg/kg bw for 3 to 6 months resulted<br />
in a dose-dependent increase <strong>of</strong> a variety <strong>of</strong> tumors predominantly observed in the<br />
forestomach, the kidneys (epithelial adenomas and adenocarcinomas <strong>of</strong> the renal cortex)<br />
and tumors <strong>of</strong> the renal pelvis or urinary bladder (Mengs et al. 1982; Cosyns et al. 1999)<br />
A number <strong>of</strong> single case reports as well as studies from a large cohort <strong>of</strong> patients <strong>of</strong> a<br />
Belgian slimming regimen, who received AA-contaminated products by accident, have<br />
reported a high prevalence <strong>of</strong> urothelial cancer in patients with aristolochic acid<br />
nephropathy (AAN) (Cosyns et al. 1999; Nortier et al. 2000). Most recently, AA was also<br />
associated with the etiology <strong>of</strong> the Balkan Endemic Nephropathy (BEN), a form <strong>of</strong> interstitial<br />
fibrosis associated with urothelial tumors which is most prevalent in the Balkan countries<br />
Bosnia, Serbia Croatia, Romania and Bulgaria (Arlt et al. 2007; Grollman et al. 2007).<br />
However, until now it has not been unequivocally established that the inhabitants <strong>of</strong> BEN<br />
areas are indeed exposed to higher concentration <strong>of</strong> AA than inhabitants <strong>of</strong> other regions,<br />
and the routes <strong>of</strong> exposure remain largely unknown.<br />
The genotoxic action <strong>of</strong> AA is based on its bioactivation to an intermediate aristolactamnitrium<br />
ion (Krumbiegel et al. 1987; Chan et al. 2006; Stiborova et al. 2008), which is able<br />
to form bulky adducts at purine bases <strong>of</strong> the DNA (Schmeiser et al. 1988), including<br />
dAdenin-AAI adducts which were identified as the main effectors for the mutagenic and<br />
carcinogenic action <strong>of</strong> AA (Pfau et al. 1990; Pfau et al. 1991). The mutagenic potential for<br />
AA was demonstrated in a number <strong>of</strong> in vitro systems (Schmeiser et al. 1984; Arlt et al.<br />
2002; Zhang et al. 2004). Increased mutation frequencies were also seen in the kidneys,<br />
bladder and forestomach <strong>of</strong> carcinogenicity models such as the transgenic Muta TM Mouse<br />
(Kohara et al. 2002) and Big Blue rats (Mei et al. 2006). In another study, analysis <strong>of</strong><br />
14
Chapter 1: Introduction<br />
mutational spectra from tumors <strong>of</strong> AA-treated rats revealed activating AT to TA transversion<br />
mutations in codon 61 <strong>of</strong> the H-ras oncogene in forestomach, lung and ear duct tumors, but<br />
not in kidney tumors (Schmeiser et al. 1990), suggesting an H-ras independent pathway <strong>of</strong><br />
renal tumor induction. In humans, AT to TA transversion mutations were observed in the<br />
p53 tumor suppressor gene in urothelial tumors derived from AAN and BEN patients (Lord<br />
et al. 2004; Arlt et al. 2007).<br />
1.3.2 Ochratoxin A<br />
Ochratoxin A (OTA) is a naturally occurring mycotoxin, commonly produced by a variety <strong>of</strong><br />
Penicillium and Aspergillus species (Figure 1.4). It occurs as a notorious contaminant <strong>of</strong><br />
grains and cereals and is also commonly found in food and beverages, e.g. in dried fruits,<br />
chocolate, juices, wine or beer (Kuiper-Goodman 1991; Clark and Snedeker 2006).<br />
Figure 1.4: Chemical structure <strong>of</strong> ochratoxin A. From: www.wikipedia.org/wiki/Ochratoxin<br />
I: Acute and Chronic Toxicity<br />
Acute toxicity <strong>of</strong> OTA is highly species-specific with an oral LD50 ranging from 0.2-<br />
1.0mg/kg bw for species such as sheep, pigs, rabbits and dogs, to 20mg/kg bw in rats, and<br />
to 46-58mg/kg bw in mice. In rats, chronic exposure with OTA targets primarily the kidney.<br />
OTA accumulates to high amounts in the proximal tubule where it provokes morphological<br />
changes predominantly in the S3 segment <strong>of</strong> the proximal convoluted tubule, including<br />
cytoplasmatic alterations, karyomegaly, and tubular degeneration as well as the<br />
development <strong>of</strong> cortical fibrosis. Functional alterations in response to OTA treatment are<br />
reflected by a decreased creatinine, inulin and p-aminohippurate clearance, polyuria,<br />
glucosuria, and a decreased osmolarity <strong>of</strong> the urine (NTP 1989).<br />
15
II: Carcinogenicity<br />
Chapter 1: Introduction<br />
OTA is a potent renal carcinogen in rodents (NTP 1989; Boorman et al. 1992; Rasonyi et<br />
al. 1999; Mantle et al. 2005). However, due to inadequate evidence for carcinogenicity in<br />
humans, it was classified by IARC as possibly carcinogenic to humans (group 2B). Rats<br />
exposed for two years with 210µg/kg bw/day developed a massive increase <strong>of</strong> benign and<br />
malignant tumors arising from tubular renal epithelial cells (Boorman et al. 1992). The 60%<br />
tumor incidence in male rats was 10-fold higher than observed in female rats, suggesting a<br />
sex-specific mode <strong>of</strong> action. To date, the mechanism <strong>of</strong> OTA-induced carcinogenicity is still<br />
not well defined and controversial results regarding a genotoxic (Faucet et al. 2004) or a<br />
non-genotoxic (Mally et al. 2005) mode <strong>of</strong> action have been published. Since OTA has not<br />
been clearly demonstrated to form reactive intermediates capable <strong>of</strong> covalently interacting<br />
with the DNA (Kamp et al. 2005; Mally et al. 2005; Delatour et al. 2008), a non-genotoxic<br />
mechanism <strong>of</strong> action is suggested by the majority <strong>of</strong> studies. Recent data indicate that renal<br />
toxicity as well as the observed DNA damage are most likely attributable to oxidative stress<br />
(Kamp et al. 2005; Marin-Kuan et al. 2006; Domijan et al. 2007). Notably, MAPK-ERK<br />
pathway activation as well as histone deacetylase-induced gene silencing are suggested to<br />
play an important role in OTA-induced tumorigenesis (Marin-Kuan et al. 2007).<br />
Similar to aristolochic acid, OTA is assumed to be involved in the onset <strong>of</strong> BEN, based on<br />
studies that revealed a higher frequency <strong>of</strong> BEN in inhabitants <strong>of</strong> areas endemic for OTAcontaining<br />
foods (Fuchs and Peraica 2005; Clark and Snedeker 2006). Moreover, OTA was<br />
more frequently detected in blood samples <strong>of</strong> patients with BEN than in healthy individuals<br />
(Fuchs and Peraica 2005). Based on data from one human volunteer who ingested 395ng<br />
<strong>of</strong> radioactivelly labelled [3H]-OTA, the half-life <strong>of</strong> 35,5 days was about eight-fold longer the<br />
OTA half-life determined for rats (Studer-Rohr et al. 2000). Therefore, although a causal<br />
relationship between OTA and the onset <strong>of</strong> BEN and urothelial tumors still remains elusive,<br />
the long half-life in humans might contribute to the accumulation <strong>of</strong> OTA in various target<br />
organs, and thus to the onset <strong>of</strong> BEN or OTA-mediated cancers.<br />
1.3.3 Cycasin / Methylazoymethanol<br />
Methylazoxymethanol (MAM) is the biologically active metabolite <strong>of</strong> cycasin (Figure 1.5, A),<br />
which is a toxic glycoside <strong>of</strong> the cycad plants Cycas circinalis and Cycas revoluta. Nuts,<br />
roots and leaves <strong>of</strong> these plants have been sources for food and medicine for the<br />
indigenous people <strong>of</strong> the Western pacific (e.g. Guam).<br />
16
A: B: C:<br />
Chapter 1: Introduction<br />
Figure 1.5: Chemical structures <strong>of</strong> (A) cycasin, (B) MAM and (C) methyldiazonium ion. From:<br />
(Sohn et al. 2001)<br />
I: Acute and Chronic toxicity<br />
The acute oral LD50 <strong>of</strong> cycasin is highly species-dependent, ranging from less than 20<br />
mg/kg bw in the guinea pig to 562 mg/kg bw in the rat (Hirono 1972). In rats, a 6.5 fold<br />
lower oral LD50 was detected when the active metabolite MAM was directly gavaged<br />
(Hirono 1972). Acute toxic effects <strong>of</strong> MAM and cycasin mainly occurred in the liver,<br />
including glycogen, RNA and phospholipid depletion, cellular necrosis, and haemorrhage<br />
(Williams and Laqueur 1965). Moreover, MAM was described as a developmental<br />
neurotoxin in rats when administered during fetal or neonatal periods <strong>of</strong> CNS development<br />
(Singh 1980). In humans, consumption <strong>of</strong> cycasin-containing products resulted in liver<br />
damage and was associated with the etiology <strong>of</strong> the neurological disorder Amyotrophic<br />
Lateral Sclerosis-Parkinson Dementia Complex (ALS-PDC) in people <strong>of</strong> the Western<br />
Pacific (Ince and Codd 2005; Borenstein et al. 2007).<br />
II: Carcinogenicity<br />
Cycasin was found to be a potent genotoxic carcinogen in rodents, and the duration <strong>of</strong><br />
exposure was shown to highly influence the site <strong>of</strong> tumor development (Hirono 1972). In<br />
cycasin treated rats, the target sites <strong>of</strong> tumor development w highly dependent on the dose<br />
regimen. In rats, one single high oral doses <strong>of</strong> cycasin (100 to 500 mg/kg bw) (Hirono et al.<br />
1968; Laqueur 1968; Fukunishi et al. 1971; Uchida and Hirono 1981) or repeated oral<br />
doses (25mg/kg bw for 12 weeks) (Fukunishi et al. 1971) predominantly resulted in tumors<br />
<strong>of</strong> the kidneys, followed by neoplasms <strong>of</strong> the liver and colon. In contrast, life long exposure<br />
to lower doses <strong>of</strong> cycasin (4mg/kg bw via drinking water) primarily induced mammary and<br />
testicular cancer, while tumors <strong>of</strong> the liver, kidney and intestine only occurred sporadically<br />
(Fukunishi et al. 1971). Cycasin was carcinogenic only after passage through the<br />
gastrointestinal tract (Laqueur 1964), where it was deglucosylated by ß-glucosidases <strong>of</strong> the<br />
17
Chapter 1: Introduction<br />
gut micr<strong>of</strong>lora to release the principal metabolite MAM (Spatz et al. 1967; Matsushima et al.<br />
1979; Choi et al. 1996). Under physiological conditions, MAM spontaneously hydrolyzes to<br />
water, formaldehyde, and the “ultimate carcinogen” methyldiazonium ion (Figure 1.5, C), an<br />
active DNA methylating agent, responsible for N7-methylguanine- and O6-methylguanine-<br />
DNA-alkylations, as observed in vitro (Matsumoto and Higa 1966) and in vivo (Zedeck and<br />
Brown 1977; Sohn et al. 2001) after MAM exposure. However, to date a reliable<br />
determination <strong>of</strong> the respective adducts in kidneys <strong>of</strong> MAM-treated rats remains difficult.<br />
Intragastric or intraperitoneal administration <strong>of</strong> MAM induced similar tumors as observed<br />
with cycasin, although with a 15% higher incidence (Laqueur 1968). Due to the chemical<br />
instability <strong>of</strong> MAM most studies were performed using synthetic MAM acetate (MAMAc).<br />
MAMAc is highly sensitive to hydrolysis by various blood and organ esterases, and<br />
therefore is deacetylated to MAM almost immediately following administration (Zedeck and<br />
Brown 1977).<br />
Up to now, no case reports <strong>of</strong> cancer in humans exposed to cycasin or MAM have been<br />
reported. Since further epidemiological studies did not show increased cancer mortality 2 to<br />
7 years after heavy intake <strong>of</strong> cycad plants or plant products, cycasin and MAM were<br />
classified by the IARC as possibly carcinogenic to humans (group 2B carcinogen).<br />
1.4 Methods to Identify Renal Carcinogens<br />
The vast majority <strong>of</strong> all kidney cancers are expected to be caused by exogenic factors, like<br />
exposure to industrial or environmental chemicals, pesticides or mycotoxins (McLaughlin<br />
and Lipworth 2000). Identification <strong>of</strong> potential carcinogens and assessing their risk for<br />
humans is therefore the first step in disease prevention. For decades, methods for the<br />
identification <strong>of</strong> renal and other carcinogens have been extensively developed, including<br />
the rodent bioassays (see below), animal models for cancer induction at distinct organ sites<br />
(e.g. the Eker rat model <strong>of</strong> renal cancer, for details see chapter 1.4.3), in vitro methods to<br />
study the effects <strong>of</strong> carcinogens in cells and organ slice cultures, and mechanistic assays to<br />
study various molecular events such as metabolic activation and detoxification <strong>of</strong><br />
carcinogens, DNA damage and repair and mutagenicity.<br />
18
1.4.1 Standard carcinogenicity testing<br />
Chapter 1: Introduction<br />
For almost 40 years, carcinogenicity testing has been used in an attempt to predict the<br />
potential <strong>of</strong> chemicals and pharmaceuticals to cause cancer in humans. According to<br />
different guidelines in the United States, Japan and Europe, carcinogenicity testing is<br />
required when human exposure to the respective compound is expected in a continuing or<br />
intermittent manner for at least three months (according to the U.S. Food and Drug<br />
Administration (FDA)) or six months (according to Japan’s Guidelines for Toxicity Studies <strong>of</strong><br />
Drugs Manual, or to European rules governing medicinal products, respectively). At any<br />
rate, carcinogenicity testing is required when there is concern about the potential <strong>of</strong> a<br />
chemical known to have a carcinogenic effect (based e.g. on structure activity<br />
relationships).<br />
The standard carcinogenicity testing procedure consists <strong>of</strong> a battery <strong>of</strong> genotoxicity tests<br />
involving: 1) a bacterial reverse mutation test (Ames test), 2) an in vitro genotoxicity test<br />
(e.g. in vitro mammalian chromosome aberration test, in vitro mammalian cell gene<br />
mutation test, or unscheduled DNA synthesis test in mammalian cells), and 3) an in vivo<br />
genotoxicity test (e.g. mammalian erythrocyte micronucleus test, mammalian bone marrow<br />
chromosome aberration test or unscheduled DNA synthesis test with mammalian liver cells<br />
in vivo).<br />
To be designated as a genotoxic carcinogen, compounds must have been proven positive<br />
in the Ames test combined with at least one in vitro and one in vivo genotoxicity test. Only<br />
compounds with negative or substantially negative results in these tests are further<br />
evaluated in the rodent lifetime bioassay for their carcinogenic potential.<br />
1.4.2 The lifetime rodent bioassay<br />
To date almost 1500 chemicals have been tested by a standard protocol originally<br />
developed by the National Toxicology Program (NTP). The protocol involves exposing<br />
distinct strains <strong>of</strong> rats (Sprague Dawley or Wistar rats) and mice (B6C3F1) <strong>of</strong> both sexes to<br />
the respective compound for a period <strong>of</strong> 104 weeks. Typically 50 animals per sex and dose<br />
group and 4 doses, i.e. 0 (control), the maximum tolerated dose (MTD), MTD/2 and MTD/4<br />
are used The MTD <strong>of</strong> a test compound is generally determined from a preceding 90-day<br />
study using the route and method <strong>of</strong> administration that will be used in the lifetime<br />
bioassay. During the two year bioassay, animals <strong>of</strong> all groups are dosed for the indicated<br />
19
Chapter 1: Introduction<br />
time and subsequently sacrificed for a post-mortem evaluation <strong>of</strong> tumor incidences and<br />
pathological changes in 24 target organs (OECD 1981; NTP 1992).<br />
Utilizing the rodent bioassay for identifying potential cancer hazards for humans is generally<br />
based on two assumptions: a) that high dose effects can be extrapolated to lower doses<br />
and b) that an interspecies extrapolation between animals and humans is possible (Ward<br />
2007). The latter assumption may be reasonable at least for genotoxic compounds that<br />
tend to be carcinogenic across species, in both sexes and at all dose levels (e.g. aflatoxin<br />
B1 or vinyl chloride). In contrast, non-genotoxic carcinogens <strong>of</strong>ten do not induce adverse<br />
effects if dosed below a certain threshold (see 1.2.2) and may act via species-specific<br />
mechanisms that have no relevance for humans. Since current standard test procedures<br />
are based on high-dose regimens to warrant a potential detection <strong>of</strong> carcinogenic<br />
properties <strong>of</strong> a compound, an overestimation <strong>of</strong> the carcinogenic potential <strong>of</strong> a testsubstance<br />
is possible.<br />
The long duration <strong>of</strong> the rodent bioassay together with the high number <strong>of</strong> animals leads to<br />
enormous costs. Especially for drug development, low specificity <strong>of</strong> the standard test<br />
battery is a major drawback, and promising drug candidates may be taken out at late<br />
stages <strong>of</strong> the developmental process when tested “false positive”. In case <strong>of</strong> xenobiotica,<br />
positive carcinogenicity testing determines the intensity <strong>of</strong> regulatory management and<br />
restriction <strong>of</strong> marketing.<br />
To improve in vivo carcinogenicity testing without compromising human safety, current<br />
efforts try to augment the depth <strong>of</strong> studies by introducing more mechanistic data. Moreover,<br />
by adopting the directive on the protection <strong>of</strong> experimental animals in 1986, the protection<br />
and welfare <strong>of</strong> animals has become a European value. The European Union and its<br />
Member States are obliged to ensure that the number <strong>of</strong> experimental animals, their<br />
possible pain, suffering, distress or lasting harm as a consequence <strong>of</strong> procedures being<br />
conducted upon them, shall be kept to an absolute minimum. The directives thus can be<br />
summarized as the “3R- principle”, to reduce, replace and refine animal experiments.<br />
Recently, international governmental organizations and industry partners agreed upon the<br />
new International Conference on Harmonization (ICH) guideline “Testing for<br />
Carcinogenicity <strong>of</strong> Pharmaceuticals” (ICH 1997). This new guidance allows the use <strong>of</strong> only<br />
one rodent species (preferably the rat) and the replacement <strong>of</strong> the second species by newly<br />
developed models like the newborn mouse assay or one <strong>of</strong> various transgenic mouse<br />
models (e.g. the p53+/- mouse). The alternative usage <strong>of</strong> genetically engineered models<br />
might also provide additional mechanistic information not readily available from the<br />
20
Chapter 1: Introduction<br />
conventional long-term assay. In addition, the ‘International Agency for Research on<br />
<strong>Cancer</strong>’ working group has recommended the use <strong>of</strong> preneoplastic lesions as early<br />
indicators for carcinogenic activity (Williams 1999).<br />
Other promising model systems under development include non-mammalian species, in<br />
vitro cell systems, computer-based predictive toxicology models, or microarray-based<br />
toxicogenomics (described in chapter 1.6.). However, to allow reliable dose- and speciesextrapolation<br />
through the use <strong>of</strong> these systems requires a much more in-depth<br />
understanding <strong>of</strong> mechanistic processes involved in the onset and progression <strong>of</strong><br />
chemically induced carcinogenesis<br />
1.4.3 The Eker Rat Model <strong>of</strong> Renal Carcinogenesis<br />
Eker rats are a rodent model <strong>of</strong> hereditary kidney cancer. Their hereditary tumors resemble<br />
chemically induced tumors in other rat strains, as well as human renal adenomas and<br />
carcinomas. Thus, Eker rats were applied to study the etiology and mechanisms underlying<br />
kidney cancer. Their genetic predisposition is caused by a heterozygous insertion <strong>of</strong> a<br />
retroviral element into the tuberous sclerosis 2 (TSC2) tumor suppressor gene (Yeung et al.<br />
1994; Hino et al. 1995). Under physiological conditions, TSC2 encodes a protein called<br />
tuberin that binds to the TSC1 gene product hamartin to form a complex which acts as a<br />
GTPase-activating protein (GAP), thereby preventing Rheb-GTP-dependent stimulation <strong>of</strong><br />
the PI3K-Akt-Tsc1/2-Rheb-mTor pathway. Consequently, TSC2 and TSC1 are key<br />
mediators <strong>of</strong> growth factors or various nutrients to regulate cell growth, proliferation,<br />
migration and differentiation (Mak and Yeung 2004; Jozwiak 2006; Wullschleger et al.<br />
2006; Hanna et al. 2008).<br />
The inheritance <strong>of</strong> the TSC2 mutation follows Mendelian genetics. While a homozygous<br />
loss <strong>of</strong> the TSC2 gene causes embryonic lethality, the heterozygous <strong>of</strong>fspring is<br />
predisposed to the spontaneous development <strong>of</strong> multiple and <strong>of</strong>ten bilateral RCC with a<br />
100% incidence by one year <strong>of</strong> age (Everitt et al. 1995). Loss <strong>of</strong> heterozygosity (LOH) was<br />
only seen in 60% <strong>of</strong> the renal tumors in Eker rats, suggesting that a second genetic hit<br />
within the TSC2 gene is not the exclusive rate-limiting step for the development <strong>of</strong> renal cell<br />
carcinomas in this strain (Yeung, 1995). Rather, alternative pathogenic mechanisms such<br />
as haploinsufficiency seem possible.<br />
Additional studies have demonstrated that Eker rats are highly susceptible towards<br />
genotoxic and non-genotoxic renal carcinogens, which makes them a useful model for the<br />
21
Chapter 1: Introduction<br />
early detection <strong>of</strong> renal carcinogens. Eker rats treated with a single dose <strong>of</strong> the DNA<br />
alkylating dimethylnitrosamine (DMN) had a 70-fold higher average number <strong>of</strong> tumors per<br />
animal, compared to DMN-treated wild type rats (Walker, 1992). Furthermore,<br />
transplacental administration <strong>of</strong> N-ethyl-nitrosamine (ENU) resulted in the early occurrence<br />
<strong>of</strong> gross RCC after 8 weeks <strong>of</strong> age (Hino et al. 1993). No increased incidence but an<br />
advanced progression <strong>of</strong> preneoplastic lesion was observed in Eker rats that were<br />
subchronically treated with the tumor promoter sodium barbital (Wolf et al. 1998). In<br />
contrast, genotoxic non-renal carcinogens had no consistent effect on total numbers or<br />
incidence <strong>of</strong> preneoplastic and neoplastic lesions in the Eker rat kidney (Morton et al.<br />
2002). From these studies, it was concluded that Eker rats represent a useful tool to identify<br />
renal carcinogens and to study the underlying mechanisms.<br />
1.5 Toxicogenomics<br />
Microarrays have been used as a powerful and widespread tool for the simultaneous<br />
measurement <strong>of</strong> gene expression from thousands <strong>of</strong> genes within a target tissue. Soon<br />
after the first microarrays were applied to analyze molecular events that underlie chemically<br />
induced toxicity (toxicogenomics), toxicologists became fascinated by the multitude <strong>of</strong><br />
possibilities for safety assessment: Gene expression data may be used to identify<br />
mechanistic pathways involved in the dose-, tissue-, inter- and intra-specific heterogeneity<br />
<strong>of</strong> compound-induced toxicity. In addition, toxicogenomics can be used to identify and<br />
characterize novel biomarkers which might serve as early predictors for adverse effects <strong>of</strong><br />
test compounds in short term in vivo or in vitro experiments (Luhe et al. 2005).<br />
1.5.1 Principles <strong>of</strong> the Microarray Technology<br />
To date various types <strong>of</strong> microarrays are commercially available, such as cDNA arrays or<br />
oligonucleotide arrays. Currently Affymetrix GeneChips® are the most popular commercial<br />
arrays, and were also used for this thesis.<br />
Affymetrix GeneChips® consist <strong>of</strong> up to 1.3 million 25-mer oligonucleotide probes attached<br />
to a quartz surface, each with a gene-specific DNA sequence able to bind complementary<br />
target RNA <strong>of</strong> a test sample. Probe-target hybridizations can be quantified to determine the<br />
22
Chapter 1: Introduction<br />
relative abundance <strong>of</strong> an expressed gene in a test sample. The main feature <strong>of</strong> Affymetrix<br />
chips is that each gene is represented by an ensemble <strong>of</strong> probes (probe set) mapping to<br />
eleven different regions <strong>of</strong> the gene (Figure 1.6, A). Each <strong>of</strong> the eleven oligonucleotides is<br />
furthermore represented by a set <strong>of</strong> perfect match probes (PM) and identical mismatched<br />
probes (MM), with the exception <strong>of</strong> a single base difference in a central position. The<br />
mismatch probes serve as controls to determine the specificity <strong>of</strong> target-probe<br />
hybridizations and therefore to receive a higher reliability <strong>of</strong> gene expression data. As an<br />
example, Affymerix Rat Genome RAE_230_2 GeneChips® used for this thesis cover the<br />
entire genome <strong>of</strong> the rat by containing 30.199 probe sets encoding known annotated genes<br />
and unknown open reading frames (expressed sequence tags (EST)).<br />
Figure 1.6: (A) Composition <strong>of</strong> an Affymetrix chip. MM: mismatch oligonucleotide; PM: perfect<br />
match oligonucleotide. (B) Overview on Affymetrix chip hybridization. (C) Staining <strong>of</strong><br />
the DNA-cRNA hybrides using streptavidin-phycoerythrin. Modified from Affymetrix<br />
Expression Analysis: Technical manual for eukaryotic sample and array processing.<br />
23
Chapter 1: Introduction<br />
For gene expression analysis using Affymetrix chips, RNA is isolated from different test<br />
samples for subsequent cDNA preparation using a T7-Oligo(dT) promoter primer (Figure<br />
1.6, B). Following RNase H-mediated second-strand cDNA synthesis, the double-stranded<br />
cDNA serves as a template for in vitro transcription (IVT). The IVT reaction is carried out in<br />
the presence <strong>of</strong> T7 RNA polymerase and a mix <strong>of</strong> biotinylated nucleotides for<br />
complementary RNA (cRNA) amplification and biotin labeling. The biotinylated cRNA<br />
targets are then cleaned up, fragmented, and hybridized onto an Affymetrix GeneChip<br />
expression array (Figure 1.6, B). Biotinylated probe-target hybridizations are visualized by<br />
staining the arrays with a streptavidin/phycoerythrin solution and the addition <strong>of</strong> a<br />
biotinylated anti-streptavidin antibody, followed by a second streptavidin/phycoerythrin<br />
staining step (Figure 1.6, C). Hybridization and staining <strong>of</strong> Affymetrix arrays are performed<br />
with specialized robots, which increase both reproducibility and throughput.<br />
Fluorescence signals can be scanned and raw data image files (DAT) can be converted<br />
into CEL-files using Affymetrix Microarray Suite (MAS) 5.0 algorithm. In CEL files, the scan<br />
data from 36 pixels per PM/ MM probe set are averaged, background, gradients and<br />
distortions are corrected. The fluorescent intensities <strong>of</strong> the eleven PM/MM probe pairs per<br />
gene are then condensed to one intensity value per probe set, associated with a statistical<br />
detection <strong>of</strong> a p-value calculated from the intensity differences <strong>of</strong> the PM and corresponding<br />
MM oligonucleotides. This p-value indicates how reliably a transcript is detected and<br />
transcripts below a distinct quality setting p-value can be excluded from further statistical<br />
analysis. Different statistical tests can then be used to elucidate significant gene<br />
deregulations between two or more test samples.<br />
1.5.2 Microarrays- a Novel Tool to Study Carcinogenesis<br />
I: Diagnostic approach<br />
Microarrays have been used to study renal carcinogenesis and a number <strong>of</strong> publications<br />
have focused on the molecular classification <strong>of</strong> the different subtypes <strong>of</strong> kidney cancers in<br />
humans (Takahashi et al. 2006; Furge et al. 2007; Young et al. 2008). These studies not<br />
only revealed specific gene expression pr<strong>of</strong>iles for different histopathological subtypes, but<br />
also <strong>of</strong>fered a more detailed insight into the underlying pathways. In addition, a number <strong>of</strong><br />
studies have suggested the use <strong>of</strong> microarrays as promising tool for refining the diagnosis<br />
and staging <strong>of</strong> RCC, as well as for highlighting potential therapeutic targets (Tan et al.<br />
2004).<br />
24
II: Mechanistic Approach<br />
Chapter 1: Introduction<br />
Thus far, only a relatively small number <strong>of</strong> studies employing toxicogenomics to study<br />
mechanisms underlying renal carcinogenesis in rats were performed. One study was<br />
designed to identify differential gene expression in kidneys <strong>of</strong> Tsc2 heterozygous mutant<br />
Eker rats compared to wild-type rats, and thereby to elucidate whether those differences<br />
may be involved in tumorigenesis (Sen et al. 2004). The authors found only 3.2% <strong>of</strong> 4395<br />
evaluated genes significantly deregulated, which mainly belonged to the functional<br />
categories <strong>of</strong> cell cycle regulation, cell proliferation, cell adhesion and endocytosis. Notably,<br />
many <strong>of</strong> these genes appear to be directly or indirectly regulated by the PI3K/Akt pathway,<br />
thereby pointing towards mechanistic pathways that may be associated with the enhanced<br />
susceptibility <strong>of</strong> the Tsc2 mutant Eker rats to develop kidney tumors. Prior to this thesis,<br />
gene expression changes in Eker and wild-type rats were not addressed after carcinogen<br />
treatment. It was therefore unknown whether and how genotoxic and non-genotoxic<br />
carcinogens would influence the TSC2 pathway.<br />
Other mechanistic studies used microarrays for the identification <strong>of</strong> deregulated genes and<br />
pathways subsequent to treatment with known genotoxic and non-genotoxic nephrocarcinogens.<br />
For example, Marin-Kuan and colleagues applied microarray technology for<br />
the analysis <strong>of</strong> gene expression changes in kidney samples from F344 rats that were orally<br />
dosed with 300 µg/kg bw OTA during a two year carcinogenicity study. The authors<br />
demonstrated a prominent down regulation <strong>of</strong> genes involved in antioxidant defense, which<br />
was assumed to result in an increase <strong>of</strong> reactive oxygen species, leading to oxidative DNA<br />
damage and tumor initiation (Marin-Kuan et al. 2006). In another study, Chen et al. (Chen<br />
et al. 2006) demonstrated that gene expression pr<strong>of</strong>iling can reveal differential effects <strong>of</strong> AA<br />
exposure on the kidneys and liver <strong>of</strong> rats, i.e. significant alterations <strong>of</strong> genes previously<br />
associated with biotransformation and carcinogenesis, apoptosis and immune response<br />
mainly in kidney, but not in liver, Therefore, in the respective study gene expression<br />
pr<strong>of</strong>iling served to explain underlying mechanisms for the tissue-specific toxicity and<br />
carcinogenicity <strong>of</strong> AA.<br />
These and a number <strong>of</strong> other studies have demonstrated that microarrays facilitate the<br />
interpretation <strong>of</strong> a toxic or carcinogenic mechanism <strong>of</strong> action. The use <strong>of</strong> mechanistic data<br />
in preclinical trials would not only allow to assess differences between acute and chronic<br />
toxicity, but also to extrapolate toxic effects between different species. Moreover,<br />
microarrays were previously used to study gene expression changes in F344 rats<br />
subsequent to treatment with already known non-effective doses and carcinogenic doses <strong>of</strong><br />
potassium bromate. The dramatic increase in global gene expression changes in response<br />
25
Chapter 1: Introduction<br />
to the carcinogenic dose and compared to non-carcinogenic doses suggests that<br />
microarrays are indeed useful tools to determine thresholds and safety levels for test<br />
compounds (Delker et al. 2006).<br />
III: Predictive approach<br />
Histopathology and clinical chemistry subsequent to the two year rodent bioassay are still<br />
benchmarks for evaluating the toxic and carcinogenic potential <strong>of</strong> test compounds.<br />
However, since the kidneys can function relatively normal with low grade damage, altered<br />
blood or urine parameters used for clinical chemistry (e.g. blood urea, nitrogen or serum<br />
creatinine) are not readily observed until a significant portion <strong>of</strong> the kidney is damaged<br />
(Duarte and Preuss 1993). Similarly, histopathologic examinations <strong>of</strong> organ sections require<br />
that toxic lesions (e.g. pre-neoplastic and neoplastic lesions) have become manifested for<br />
microscopic detection. Early or compound-related alterations at a molecular level (e.g. DNA<br />
damage response) thus remain unobserved with these techniques. Microarrays therefore<br />
appear promising for a sensitive prediction <strong>of</strong> toxicity and carcinogenicity.<br />
In general, predictive toxicogenomics require databases with expression pr<strong>of</strong>iles derived<br />
from standardized in vivo or in vitro studies subsequent to treatment with well characterized<br />
pharmaceuticals and chemicals, including time-courses, doses, and clinical pathology<br />
information. These “training pr<strong>of</strong>iles” can then be used to extract sets <strong>of</strong> marker genes to<br />
classify unknown test compounds. Studies employing microarrays for predictions have<br />
been mainly reported for hepatotoxins to extract marker genes representing distinct classes<br />
<strong>of</strong> hepatotoxicity, e.g. liver necrosis or steatosis (Hamadeh et al. 2002; Steiner et al. 2004;<br />
Raghavan et al. 2005). These studies have demonstrated that toxicogenomics can be used<br />
to determine toxicologically relevant changes at very early time points (Steiner et al. 2004)<br />
and at very low doses (Hamadeh et al. 2002). To date, only a few in vivo and in vitro<br />
studies have utilized microarrays to identify early biomarkers that predict chronic endpoints<br />
such as cancer, and that may refine the lifetime rodent bioassay (van Delft et al. 2004;<br />
Tsujimura et al. 2006; Fielden et al. 2007). Most important for this thesis, Ellinger-<br />
Ziegelbauer and colleagues demonstrated that hepatic gene expression pr<strong>of</strong>iles <strong>of</strong> rats<br />
treated for up to 14 days with known genotoxic and non-genotoxic carcinogens allow to<br />
distinguish both groups <strong>of</strong> carcinogens (Ellinger-Ziegelbauer et al. 2005). However, none <strong>of</strong><br />
the available studies dealt with renal cancer.<br />
26
1.5.3 Microarrays- Current Limitations and Future Directions<br />
Chapter 1: Introduction<br />
After several years <strong>of</strong> research, toxicogenomics came into the focus <strong>of</strong> regulatory agencies<br />
and may influence the future decision making on drug safety. Currently, the U.S. Food and<br />
Drug Administration is developing a framework to review gene expression data from<br />
classical toxicological studies (Leighton et al. 2006). However, more research is needed to<br />
adequately assess the potential <strong>of</strong> microarrays for carcinogenicity testing and for the<br />
analysis <strong>of</strong> mechanisms underlying carcinogenesis.<br />
Most <strong>of</strong> the recent studies that analyzed gene expression changes to elucidate mechanism<br />
involved in cancer used whole tumor- or biopsy samples as a source for RNA isolation (Ellis<br />
et al. 2002; Li et al. 2002; Goley et al. 2004; Marin-Kuan et al. 2006; Jones and Pantuck<br />
2008). These samples <strong>of</strong>ten contain different amounts <strong>of</strong> non-cancerous cells, e.g. stromal<br />
or endothelial cells or infiltrating inflammatory cells, thereby complicating an accurate<br />
delineation <strong>of</strong> gene expression changes in non-tumor versus tumor samples. Nevertheless,<br />
the <strong>characteristics</strong> <strong>of</strong> a gross tumor may be influenced by the interaction with these<br />
different, non-cancerous cell types that may contribute to a microenvironment that is<br />
essential for cancer growth, invasion and metastatic progression. Thus it may be relevant<br />
for some, however not for all purposes, to take these cell types under investigation.<br />
In addition, gene expression pr<strong>of</strong>iling seems challenging when analyzing toxic effects in<br />
complex target organs. For instance, the kidney is composed <strong>of</strong> various tubular segments<br />
and cell types, which all have highly specific physiological functions and different<br />
susceptibility towards xenobiotica. Distinct segment-specific nephrotoxic effects may<br />
therefore not be readily detectable by gene expression pr<strong>of</strong>iling when “diluted” by other,<br />
unaffected nephron parts. Separation <strong>of</strong> the kidney into its basal segments (cortex, medulla<br />
and papilla) would serve to highly improve the outcome <strong>of</strong> gene expression experiments.<br />
A major improvement was the development <strong>of</strong> laser capture microdissection (Emmert-Buck<br />
et al. 1996; Schutze et al. 1998), which allows the isolation <strong>of</strong> single cells from a tissue<br />
section, and to collect pure populations <strong>of</strong> morphologically similar cells for subsequent<br />
molecular analysis. Gene expression pr<strong>of</strong>iling from isolated, early preneoplastic lesions<br />
could for instance allow a direct identification <strong>of</strong> gene expression changes underlying<br />
earliest morphologic pathologies. However, previous to this thesis, gene expression<br />
pr<strong>of</strong>iling had not been used for the analysis <strong>of</strong> microdissected pre-neoplastic lesions.<br />
27
Chapter 2: Work Hypotheses & Experimental Design<br />
Lifetime rodent bioassays play a central role in evaluating the carcinogenicity <strong>of</strong> chemicals,<br />
yet are critically viewed especially with regard to their long duration, high dose regimens,<br />
high animal numbers and financial costs. By detecting end-stage adenomas and<br />
carcinomas current protocols enable the identification <strong>of</strong> carcinogens which can act on any<br />
stage <strong>of</strong> tumorigenesis (tumor initiation, promotion and progression). However, they do not<br />
allow drawing detailed conclusions on the underlying molecular mechanisms and their<br />
relevance for human risk assessment. Accordingly, interest in improving carcinogenicity<br />
testing by developing mechanistic and predictive short-term in vivo studies has grown. One<br />
promising approach might be the use <strong>of</strong> microarrays to analyze compound-induced gene<br />
expression changes in different target organs after short-term exposure. However, at the<br />
beginning <strong>of</strong> this thesis:<br />
1. No studies had been performed that elucidated the applicability <strong>of</strong> short-term in vivo<br />
assays in predicting the carcinogenic potential <strong>of</strong> genotoxic and non-genotoxic<br />
renal carcinogens.<br />
2. The identity <strong>of</strong> deregulated genes and pathways involved in the early onset and<br />
progression <strong>of</strong> carcinogen-induced and hereditary renal preneoplastic lesions was<br />
largely unknown.<br />
3. One principal biasing factor for a reliable human risk assessment from rodent<br />
bioassays, i.e. the impact <strong>of</strong> high-fat diet and adiposity on pathways involved in<br />
renal cancer, was never addressed in detail.<br />
2.1 Part I<br />
The first aim <strong>of</strong> this thesis was to investigate the hypothesis that short-term (14 days) in<br />
vivo assays, using microarrays as the detection system, can be used as a tool to evaluate<br />
the renal carcinogenic potential <strong>of</strong> test compounds. Specifically, the most important<br />
prerequisites for short-term in vivo assays to allow for such a sensitive and specific<br />
detection <strong>of</strong> renal carcinogens should be examined, as specified in the below subhypotheses:<br />
28
Chapter 2: Work Hypotheses & Experimental Design<br />
1. Short-term treatment with known renal carcinogens will result in detectable<br />
compound-specific gene expression changes in kidney cortex homogenates <strong>of</strong> rats<br />
(Sensitivity).<br />
2. Gene expression pr<strong>of</strong>iles will serve to distinguish a genotoxic from a non-genotoxic<br />
mode <strong>of</strong> action <strong>of</strong> respective carcinogens (Specificity).<br />
3. Exposure to carcinogens with early genotoxic or non-genotoxic expression changes<br />
in short-tem assays will also result in a carcinogen-specific onset <strong>of</strong> kidney tumors<br />
after subchronic or chronic exposure (Predictivity).<br />
2.1.1 Experimental setup<br />
Short-term exposure: male and female TSC2-mutant Eker rats or wild-type rats were<br />
gavaged daily for 1, 3, 7 and 14 days with the genotoxic carcinogens AA or MAMAc, the<br />
non-genotoxic carcinogen OTA, or vehicle. Subsequent to exposure, kidneys were taken<br />
for the analysis and comparison <strong>of</strong> the following parameters:<br />
• Gene expression pr<strong>of</strong>iles derived from the kidney cortex <strong>of</strong> male Eker and wild-type<br />
rats to detect compound-induced expression changes in both a sensitive and a<br />
conventional rat strain.<br />
• Non-neoplastic and neoplastic renal histopathology in all dose groups to elucidate<br />
specific adverse effects induced by the respective compounds in both rat strains.<br />
• Cell proliferation via immunohistochemical PCNA staining to detect early mitogenic<br />
effects in all dose groups.<br />
• O6-methylguanine DNA adducts levels in 1, 7 and 14 days MAMAc-treated male and<br />
female Eker rats, respectively, indicative for the mutagenic potential <strong>of</strong> MAMAc.<br />
Subchronic and chronic exposure: Male and female Eker rats were gavaged 5 days per<br />
week for 3 or 6 months, respectively, with AA, MAMAc, OTA or vehicle. Subsequently,<br />
kidneys were harvested and analyzed for:<br />
• Sex- and compound-specific differences in non-neoplastic and neoplastic<br />
histopathology (type, incidence and total number <strong>of</strong> lesions).<br />
• Sex- and compound-specific differences in cell proliferation, analyzed by<br />
immunohistochemical BrdU staining.<br />
29
2.2 Part II<br />
Chapter 2: Work Hypotheses & Experimental Design<br />
The second hypothesis <strong>of</strong> this thesis was, that gene expression pr<strong>of</strong>iling from isolated<br />
preneoplastic lesions allows a direct identification <strong>of</strong> gene expression deregulations<br />
underlying earliest morphologic pathologies <strong>of</strong> renal cancer. Specifically the following subhypothesis<br />
should be examined:<br />
1. Different stages <strong>of</strong> preneoplastic lesions can be distinguished by their gene<br />
expression pr<strong>of</strong>iles, therefore allowing the study <strong>of</strong> mechanisms underlying<br />
preneoplastic lesion progression<br />
2. Renal carcinogens differentially affect gene expression preneoplastic lesions,<br />
based on their compound specific mode <strong>of</strong> action (genotoxic vs. non-genotoxic)<br />
3. The comparison <strong>of</strong> gene expression pr<strong>of</strong>iles from short-term treated rats with<br />
microdissected healthy tissue and preneoplastic lesions will serve to identify novel<br />
mechanistic and predictive markers involved in chemical induced renal<br />
carcinogenesis, applicable in short-term assays.<br />
2.2.1 Experimental setup:<br />
Protocol establishment: Before addressing the specific hypotheses, a protocol had to be<br />
established that allows the reliable and reproducible analysis <strong>of</strong> laser microdissected renal<br />
preneoplastic lesions by using Affymetrix Genechips. Specifically, the following conditions<br />
were optimized:<br />
• Effects <strong>of</strong> different histopathological staining techniques in combination with different<br />
sample processing techniques on RNA quantity and quality.<br />
• RNA amplification <strong>of</strong> lowest amounts <strong>of</strong> RNA from dissected lesions to allow reliable<br />
gene expression pr<strong>of</strong>iling.<br />
Microdissection and gene expression analysis: In the next step, the optimized protocol was<br />
adapted to perform large scale gene expression analyses from laser microdissected<br />
lesions:<br />
• Laser microdissection <strong>of</strong> different stages <strong>of</strong> preneoplastic lesions and healthy tubules<br />
from cryosectioned kidneys <strong>of</strong> male Eker rats treated with AA, OTA or vehicle.<br />
30
Chapter 2: Work Hypotheses & Experimental Design<br />
• RNA isolation and microarray analysis <strong>of</strong> the microdissected tissue and comparison<br />
<strong>of</strong> gene expression pr<strong>of</strong>iles between different stages <strong>of</strong> preneoplastic lesions and<br />
healthy tissue.<br />
• Verification <strong>of</strong> deregulated genes, involved in the initiation and progression <strong>of</strong><br />
preneoplastic lesions at the protein level via immunohistochemistry.<br />
2.3 Part III<br />
In the third part <strong>of</strong> this thesis it was hypothesized that chronic high-fat diet causes early<br />
stages <strong>of</strong> renal cancer in rats via mechanisms similar to those observed following exposure<br />
to renal carcinogens. The following sub-hypotheses were addressed:<br />
1. Early stages <strong>of</strong> renal cancer are a direct effect <strong>of</strong> dietary lipids, and largely<br />
independent from the degree <strong>of</strong> body adiposity.<br />
2. Exposure to chronic high fat diet results in the activation the mTOR pathway, a<br />
central component <strong>of</strong> the nutrient-hormonal signaling network and known mediator<br />
<strong>of</strong> renal carcinogenesis.<br />
2.3.1 Experimental setup<br />
From a large pool <strong>of</strong> male Wistar rats fed a customized high-fat diet for 11 months, 2<br />
subgroups <strong>of</strong> diet-induced obesity sensitive and diet-induced obesity resistant rats were<br />
selected, and compared to each other and with low-fat (standard) diet controls for the<br />
following parameters:<br />
• Life history traits (body weight, local fat depots) and plasma parameters (blood<br />
glucose, triglycerides, insulin, leptin, IL-1ß, MCP-1 and PAI-I)<br />
• Possible accumulation <strong>of</strong> renal levels <strong>of</strong> fatty acid binding protein α2u-globulin<br />
• Renal histopathology<br />
• Renal cell proliferation<br />
• Immunodetection <strong>of</strong> hyperphosphorylated S6-ribosomal protein indicative for the<br />
activation <strong>of</strong> the mTOR pathway, a central component <strong>of</strong> the nutrient-hormonal<br />
signaling network and known mediator <strong>of</strong> renal carcinogenesis<br />
31
Chapter 2: Work Hypotheses & Experimental Design<br />
2.4 Schematic Overview on the Experimental Design<br />
Figure 2.1: Overview on the experimental design<br />
32
Chapter 3: Manuscript I<br />
Carcinogen Specific Gene Expression Pr<strong>of</strong>iles in Short-Term<br />
Treated Eker and Wild Type Rats Indicative for Pathways<br />
Involved in Renal Tumorigenesis<br />
Kerstin Stemmer 1 , Heidrun Ellinger-Ziegelbauer 2 , Hans-J. Ahr 2 and Daniel R. Dietrich 1*<br />
1 Human and Environmental Toxicology, University <strong>of</strong> Konstanz, Konstanz, Germany<br />
2 <strong>Molecular</strong> and Special Toxicology, Bayer Healthcare AG, Wuppertal, Germany<br />
* Corresponding author: Daniel R. Dietrich: daniel.dietrich@uni-konstanz.de<br />
Published in <strong>Cancer</strong> Research (2007) May 1;67(9):4052-68<br />
3.1 Abstract<br />
Eker rat heterozygous for a dominant germline mutation in the tuberous sclerosis 2 (Tsc2)<br />
tumor suppressor gene, were used as a model to study renal carcinogenesis. Eker and<br />
corresponding wild type rats were exposed to genotoxic aristolochic acid (AA) or nongenotoxic<br />
ochratoxin A (OTA) in order to elucidate early carcinogen specific gene<br />
expression changes and to test whether Eker rats are more sensitive for carcinogen<br />
induced changes in gene expression. Male Eker and wild type rats were gavaged daily with<br />
AA (10mg/kg BW) or OTA (210µg/kg BW). After 1, 3, 7 and 14 days <strong>of</strong> exposure, renal<br />
histopathology, tubular cell proliferation and Affymetrix gene expression pr<strong>of</strong>iles from renal<br />
cortex/outer medulla were analyzed. AA-treated Eker and wild type rats were qualitatively<br />
comparable in all parameters assessed, suggesting a Tsc2-independent mechanism <strong>of</strong><br />
action. OTA treatment resulted in slightly increased cortical pathology and significantly<br />
elevated cell proliferation in both strains, although Eker rats were more sensitive.<br />
Deregulated genes involved in PI3K-AKT-Tsc2-mTor signalling, amongst other important<br />
genes prominent in tumorigenesis, in conjunction with the enhanced cell proliferation and<br />
presence <strong>of</strong> preneoplastic lesions suggested involvement <strong>of</strong> Tsc2 in OTA- mediated toxicity<br />
33
Chapter 3: Manuscript I<br />
and carcinogenicity, especially as deregulation <strong>of</strong> genes involved in this pathway was more<br />
prominent in the Tsc2 mutant Eker rat.<br />
3.2 Introduction<br />
Eker rats, heterozygous for a loss-<strong>of</strong>-function mutation in the tuberous sclerosis 2 (Tsc2)<br />
tumor suppressor gene, appear ideal models to study the aetiology <strong>of</strong> renal carcinogenesis<br />
(Yeung et al. 1994; Hino et al. 1995). Heredity <strong>of</strong> the Tsc2 mutation follows Mendelian<br />
genetics (Yeung et al. 1994; Kobayashi et al. 1995) and heterozygous progeny are<br />
predisposed to spontaneous development <strong>of</strong> multiple bilateral renal neoplasms originating<br />
from the proximal tubular epithelium with complete penetrance by one year <strong>of</strong> age (Everitt<br />
et al. 1995). Approximately 60% <strong>of</strong> the spontaneous renal tumors in Eker rats also exhibit a<br />
functional inactivation <strong>of</strong> the second Tsc2 allele, suggesting, in accordance with Knudson’s<br />
two-hit hypothesis that a second somatic mutation might be the rate-limiting step for the<br />
development <strong>of</strong> renal cell carcinomas (RCC) in Eker rats (Yeung et al. 1995). Eker rats<br />
have been employed to elucidate the mechanism <strong>of</strong> renal carcinogens, primarily using<br />
histopathological and statistical analyses <strong>of</strong> the number, multiplicity and progression <strong>of</strong><br />
renal lesions (Walker et al. 1992; Wolf et al. 2000). Accordingly, treatment <strong>of</strong> Eker rats with<br />
dimethylnitrosamine resulted in a 70-fold increase in the induction <strong>of</strong> renal adenomas and<br />
carcinomas, when compared to wild type rats (Walker et al. 1992). No increased lesion<br />
incidence, albeit an advanced lesion progression was observed in Eker rats, subchronically<br />
treated with the tumor promoter sodium barbital (Wolf et al. 2000). Although the latter data<br />
highlight that Eker rats are sensitive to genotoxic and non-genotoxic compounds, the<br />
involvement <strong>of</strong> Tsc2 protein (tuberin) in renal carcinogenesis remains to be established.<br />
Several studies suggest that functional Tsc2 promotes the GTP-hydrolysis <strong>of</strong> the Ras<br />
homologe Rheb, thereby acting as a negative regulator <strong>of</strong> the PI3K-Akt-Tsc1/2-Rheb-mTor<br />
pathway. Consequently, Tsc2 is suspected to play a central role in mediating growth factor,<br />
nutrient and energy sensing to regulate cell growth, proliferation, migration and<br />
differentiation (Mak and Yeung 2004).<br />
The objective <strong>of</strong> this study was to elucidate whether short-term exposure <strong>of</strong> Eker and wild<br />
type rats to a non-genotoxic and a genotoxic renal carcinogen would result in compound<br />
specific changes in renal non-neoplastic and preneoplastic pathology and cell proliferation<br />
rates. Subsequently the hypothesis was investigated, whether compound specific changes<br />
34
Chapter 3: Manuscript I<br />
in histopathology and cell proliferation can be associated with respective changes in gene<br />
expression and whether Eker and wild type rats respond differently. This should allow<br />
identification <strong>of</strong> deregulated genes involved in known and novel pathways possibly<br />
mediating carcinogen-induced renal tumorigenesis. Accordingly, Eker and wild type rats<br />
were treated with daily doses <strong>of</strong> the genotoxic and the non-genotoxic renal carcinogen<br />
aristolochic acid (AA) and ochratoxin A (OTA), respectively, for which renal tumor induction<br />
in long-term in vivo studies was previously demonstrated (Mengs et al. 1982; Boorman et<br />
al. 1992).<br />
Indeed, intragastric administration <strong>of</strong> 10mg AA /kg BW/day (representing a mixture <strong>of</strong><br />
structurally related nitrophenanthrene carboxylic acids (mostly AAI and AAII)) to rats over<br />
three months was demonstrated to induce tumors in the forestomach, kidney and the<br />
urinary bladder (Mengs et al. 1982). DNA reactivity <strong>of</strong> AA was confirmed in that the most<br />
frequent and persistent dAdenin- AAI adduct could lead to mutation and activation <strong>of</strong> the Hras<br />
oncogene in the forestomach but not in kidneys <strong>of</strong> rats (Schmeiser et al. 1991; Cheng<br />
et al. 2006) or to p53-mutations in urothelial tumors <strong>of</strong> humans (Lord et al. 2004). Despite<br />
the lack <strong>of</strong> H-ras mutations, higher levels <strong>of</strong> AA adducts were found in renal tissues than in<br />
the forestomach <strong>of</strong> orally treated Wistar rats (5mg AA/kg BW/day) after only one week <strong>of</strong><br />
exposure (Dong et al. 2006), suggesting an H-ras independent pathway <strong>of</strong> renal tumor<br />
induction. The genotoxic properties <strong>of</strong> AA are explained by the metabolic activation <strong>of</strong> AA<br />
by several phase I enzymes to a DNA-reactive aristolactam-nitriumion (Pfau et al. 1990;<br />
Arlt et al. 2002; Stiborova et al. 2003). Similarly, the mycotoxin OTA increased the<br />
incidence <strong>of</strong> renal adenoma and carcinoma in rats, when exposed for up to two years to<br />
dietary OTA (Mantle et al. 2005) or 210µg OTA /kg BW/day via gavage (Boorman et al.<br />
1992). However, as OTA has not been convincingly demonstrated to covalently interact<br />
with DNA, a non-genotoxic mechanism <strong>of</strong> action is assumed (Kamp et al. 2005; Mally et al.<br />
2005).<br />
The comparison <strong>of</strong> cell proliferation, pathology and expression pr<strong>of</strong>iles <strong>of</strong> AA- and OTAtreated<br />
Eker and wild type rats should allow for a more in-depth understanding <strong>of</strong> the<br />
involvement <strong>of</strong> the Tsc2-mTor pathway as well as <strong>of</strong> other early gene expression changes<br />
in the etiology <strong>of</strong> carcinogen induced renal tumors.<br />
35
3.3 Material and Methods<br />
3.3.1 Compounds<br />
Chapter 3: Manuscript I<br />
Ochratoxin A (>98% purity, benzene free) was kindly provided by Dr. M.E. Stack, US FDA,<br />
Washington DC. Aristolochic acid sodium salt mixture (AA I: 41% and AA II: 56%) was<br />
purchased from Sigma Aldrich, Germany.<br />
3.3.2 Animals<br />
Six to ten week old, genotyped heterozygous Tsc2 mutant Eker rats (Tsc2+/-, Long Evans)<br />
were purchased from the MD Anderson <strong>Cancer</strong> Center, Smithville, Texas, USA and<br />
maintained at the University <strong>of</strong> Konstanz animal research facility under standard conditions<br />
with food and water ad libitum. Male rats were randomly assigned to dose groups (three<br />
animals per compound (or vehicle) and time-point) and allowed to acclimatize to laboratory<br />
conditions for 4 weeks. Two weeks prior to exposure, rats were handled daily to reduce<br />
non-compound related stress during exposure.<br />
Heterozygous Eker rats were bred and wild type (Tsc2+/+) genotypes <strong>of</strong> the progeny were<br />
determined via PCR (Rennebeck et al. 1998). Two weeks prior to exposure, 8-9 week old<br />
genotyped male wild type rats, were randomly allocated to dose groups and accustomed to<br />
daily handling (see above).<br />
3.3.3 Animal treatment and sample collection<br />
Eker and wild type rats were gavaged daily with OTA (210µg/ kg BW) or AA (10mg/ kg<br />
BW), dissolved in 0,1M sodium bicarbonate. Time-matched vehicle controls were gavaged<br />
with 0,1M sodium bicarbonate. Following 1, 3, 7, and 14 days <strong>of</strong> treatment, Narcoren®<br />
(pentobarbital) anesthetized rats were sacrificed by exsanguination subsequent to<br />
retrograde perfusion with PBS. Left kidneys were collected, cross-sectioned into 5mm<br />
slices and stored in RNAlater (Qiagen, Germany) or in PBS buffered histology fixative<br />
buffer, containing 2% paraformaldehyde and 1% glutaraldehyde for subsequent paraffin<br />
embedding and sectioning.<br />
36
3.3.4 Histopathology<br />
Chapter 3: Manuscript I<br />
For histopathological examinations, H&E stained sections were randomized and<br />
pathological analysis was carried out by light microscopy at 40-400-fold magnification. Nonneoplastic<br />
changes were classified as none (0), mild (1), moderate (2), strong (3), and<br />
severe (4), including intermediate classes (e.g. 0.5, 1.5 etc.), while total numbers <strong>of</strong><br />
preneoplastic and neoplastic lesions were counted.<br />
3.3.5 Immunohistochemistry<br />
Cell proliferation was evaluated by immunohistochemical staining for proliferating cell<br />
nuclear antigen (PCNA) using monoclonal primary anti-PCNA antibody (PC-10; DAKO,<br />
Germany) in paraffin-embedded kidney sections.<br />
Sections were deparaffinized, rehydrated in a decreasing alcohol series and washed with<br />
PBS. For antigen retrieval, slides were placed in 0,1M sodium citrate buffer (pH 6.0),<br />
microwaved to boiling point 3-times and cooled to room temperature for 20min. Sections<br />
were denatured with 4 N HCl (20min at 37°C), washed with PBS (2x 5min) and non-specific<br />
protein binding blocked by preincubation with casein solution (Power BlockTM, BioGenex,<br />
USA) for 20 minutes. Sections were incubated with PC-10 primary antibody (diluted 1:50 in<br />
Power BlockTM) at 4°C for 16 hours. Antigen- antiserum complexes were visualised using<br />
the Super SensitiveTM alkaline phosphatase labelled, biotin streptavidine amplified<br />
detection system and Fast Red as chromogen according to the manufacturer’s instructions<br />
(BioGenex, USA).<br />
Cell proliferation was quantified on PCNA stained sections, randomized across all<br />
treatment and control groups. Twenty microscopic fields (10x ocular, 40x objective) were<br />
randomly chosen in the outer cortex and inner cortex/outer medulla. All tubule cell nuclei<br />
were counted, concurrently differentiating between negative and positive PCNA staining.<br />
Nuclear labeling indices for PCNA (LI%) (PCNA positive nuclei/total number <strong>of</strong> nuclei<br />
counted) were determined based on a minimum <strong>of</strong> at least 2.000 nuclei evaluated.<br />
3.3.6 RNA isolation and expression pr<strong>of</strong>iling<br />
RNA isolation from RNAlater fixed kidneys was performed as described previously<br />
(Ellinger-Ziegelbauer et al. 2005). Starting with 5µg <strong>of</strong> total RNA with a 28S/18S rRNA peak<br />
37
Chapter 3: Manuscript I<br />
ratio >1.7, biotin-labeled cRNA was prepared and subsequently hybridized on Affymetrix<br />
Rat Genome RAE230A arrays according to the manufacturer’s instructions (Affymetrix,<br />
USA; GeneChip® Expression Analysis 701194 Rev.1). This specific array contains 15.866<br />
probe sets, corresponding to approx. 5399 annotated rat genes and 10467 expressedsequence<br />
tags (ESTs).<br />
3.3.7 Microarray data processing and statistical analysis<br />
Microarray quality control was performed as described previously (Ellinger-Ziegelbauer et<br />
al. 2005) and gene expression data were submitted to the GEO repository<br />
(http://www.ncbi.nlm.nih.gov/projects/geo; accession number: GSE5923). Expressionist<br />
Analyst s<strong>of</strong>tware (Genedata AG, Switzerland) was used for statistical analysis. Significantly<br />
deregulated genes per compound were selected based on the factors treatment and time<br />
as both, single and interaction effects, in a 2-way ANOVA with a p-value cut-<strong>of</strong>f <strong>of</strong> 0.005,<br />
combined with a 1.7-fold deregulation threshold for at least one time point. Significantly<br />
deregulated genes were divided into gene groups with distinct expression pr<strong>of</strong>iles over the<br />
time-course using self-organizing map (SOM) analysis. SOM analysis also allowed<br />
deselection <strong>of</strong> genes showing inconsistent expression between the controls at different time<br />
points. Using the adjusted datasets, gene expression ratios <strong>of</strong> individual genes were<br />
calculated by dividing the respective expression values <strong>of</strong> single treated replicate samples<br />
by the mean expression value <strong>of</strong> all corresponding time-matched control samples.<br />
Heatmaps were used to graphically display the relative expression data, after onedimensional<br />
clustering <strong>of</strong> the genes (for the validation <strong>of</strong> microarray data, see<br />
supplementary information).<br />
3.3.8 Functional analysis <strong>of</strong> microarray data<br />
For functional analysis each significantly deregulated gene was characterized according to<br />
the biochemical role <strong>of</strong> its encoded protein, whenever sufficient information from<br />
databases, e.g. NetAffx from Affymetrix (update from August 2006), Swissprot, Proteome<br />
and Pubmed was available. The consequence <strong>of</strong> the direction <strong>of</strong> deregulation was<br />
interpreted specifically with regard to possible downstream pathophysiological effects. This<br />
allowed distribution <strong>of</strong> the deregulated genes into toxicological categories (Supplementary<br />
table 3.1) and facilitated the comparison <strong>of</strong> specific pathophysiological pathways, involved<br />
38
Chapter 3: Manuscript I<br />
in the response <strong>of</strong> Eker and wildy type rats to AA and OTA treatment. In addition, the<br />
pathophysiological pathways were compared with major pathways suspected to be<br />
involved in AA- and OTA- induced carcinogenesis (Tables 3.1 and 3.2) and<br />
histopathological changes observed.<br />
3.3.9 Statistical analysis<br />
Statistical analysis <strong>of</strong> histopathological and cell proliferation changes were carried out using<br />
GraphPad Prism 4® (USA) S<strong>of</strong>tware. Statistical significant differences in nuclear labeling<br />
indices (LI %) or total number <strong>of</strong> lesions in treated and control animals were analyzed by an<br />
unpaired (two tailed) t-test. A statistical significant effect <strong>of</strong> the treatment- time response<br />
was tested with a 2-Way ANOVA.<br />
3.4 Results<br />
3.4.1 Cell proliferation data<br />
AA treatment resulted in no overt change in cell proliferation rate (Figure 3.1, A and B) in<br />
either strain, despite that occasional AA groups appeared to have a significantly lower<br />
proliferation then the corresponding controls. Conversely, OTA treatment increased the<br />
proliferation rate on day 14 (Figure 3.1, C and D, Figure 3.2, A and B) 3.8- and 3.4-fold<br />
above control in Eker and wild type rats, respectively. Moreover, Eker rats appeared to be<br />
more sensitive to OTA as an increased cell proliferation rate was already observed at day 7<br />
<strong>of</strong> treatment.<br />
3.4.2 Preneoplastic and neoplastic pathology<br />
While no preneoplastic or neoplastic lesions (Dietrich and Swenberg 1991) were observed<br />
in any <strong>of</strong> the wild type rat controls, the Eker rat control groups presented with atypical<br />
tubules (AT, Figure 3.2, C), atypical hyperplasia (AH, Figure 3.2, D) and an occasional<br />
adenoma (Supplementary table 3.2). Neither AA nor OTA treatment induced an increase in<br />
preneoplastic or neoplastic lesions in wild type rats. Similarly, AA treatment <strong>of</strong> Eker rats<br />
39
Chapter 3: Manuscript I<br />
resulted in no significant increase in AT or AH. However, two carcinomas were observed in<br />
the day 14 AA-treated Eker group, and an adenoma in the day one group. In contrast, OTA<br />
treatment <strong>of</strong> Eker rats resulted in a significant increase <strong>of</strong> AT on day 14 (Figure 3.1, E).<br />
However, no significant increase in AH or neoplastic lesions were observed.<br />
Figure 3.1: Comparison <strong>of</strong> PCNA S-Phase labelling indices (LI %) in the renal cortex <strong>of</strong> AA or<br />
OTA treated Eker and wild type rats (A-D), respectively, at various treatment timepoints.<br />
E: Mean number <strong>of</strong> atypical tubules (preneoplastic lesion) per animal in<br />
control and OTA treated Eker rats. Data represent the mean + SEM <strong>of</strong> n=3 animals per<br />
group. Significant differences between treated and time-matched control groups: *p <<br />
0.05; **p < 0.01 and ***p
Chapter 3: Manuscript I<br />
Figure 3.2: Representative PCNA staining patterns <strong>of</strong> control- (A) and OTA-treated (B) Eker rats.<br />
C: Atypical tubule; D: Atypical hyperplasia; E: karyomegally (arrows) and F: apoptotic<br />
nuclei (arrow) observed in OTA treated rats.<br />
41
3.4.4 Gene expression pr<strong>of</strong>iles<br />
Chapter 3: Manuscript I<br />
Oral treatment <strong>of</strong> Eker and wild type rats with AA and OTA, respectively, led to a significant<br />
deregulation <strong>of</strong> gene expression already after one day <strong>of</strong> exposure. Compared to the<br />
respective time-matched controls, the number <strong>of</strong> significantly deregulated genes increased<br />
with the duration <strong>of</strong> exposure in both strains. At all time-points compound-treated Eker rats<br />
consistently showed a higher number <strong>of</strong> significantly deregulated genes compared to their<br />
wild type counterparts (Figure 3.3). Overall, AA treatment led to the significant deregulation<br />
<strong>of</strong> 111 non-redundant genes in Eker rats compared to 81 genes in wild type rats. However,<br />
the visualization <strong>of</strong> the expression pr<strong>of</strong>iles <strong>of</strong> the union <strong>of</strong> the Eker- and wild type- selected<br />
genes revealed a qualitatively comparable pr<strong>of</strong>ile in AA-treated Eker and wild type rats. In<br />
contrast, treatment with OTA resulted in 375 significantly deregulated, non-redundant<br />
genes in Eker rats compared to 141 genes in wild type rats, with a strikingly different<br />
expression pr<strong>of</strong>ile <strong>of</strong> at least half <strong>of</strong> the genes (Figure 3.3).<br />
Figure 3.3: Heat map <strong>of</strong> compound-specific unions <strong>of</strong> genes found significantly deregulated (red:<br />
upregulated; green: downregulated) from the corresponding control in AA or OTA<br />
induced renal gene expression pr<strong>of</strong>iles <strong>of</strong> Eker or wild type rats after 1, 3, 7 or 14 days<br />
<strong>of</strong> treatment (n=3 per time-point). Color scale (left side): gene expression ratio.<br />
42
3.4.5 Functional analysis <strong>of</strong> significantly deregulated genes<br />
Chapter 3: Manuscript I<br />
Many <strong>of</strong> the non-redundant genes, deregulated by AA (Table 3.1) or OTA (Table 3.2)<br />
treatment in Eker and wild type rats, respectively, could be associated with major<br />
pathophysiological processes involved in renal toxicity and regeneration (Supplementary<br />
table 3.1).<br />
3.4.6 Genes deregulated by aristolochic acid<br />
Metabolism and bioactivation: Treatment <strong>of</strong> Eker and wild type rats with AA led to a<br />
prominent up-regulation <strong>of</strong> genes encoding phase I or phase II biotransformation enzymes<br />
or drug transporters. Most <strong>of</strong> the genes were either constantly upregulated over the whole<br />
exposure time-frame or increasingly upregulated with prolonged duration <strong>of</strong> exposure in<br />
both Eker and wild type rats. In addition to the genes significantly deregulated in both<br />
strains, some genes met the significance criteria only in one or the other strain. Yet the<br />
direction <strong>of</strong> deregulation was mostly comparable (e.g. GSTM2 or CES3, apparently specific<br />
for Eker or wild type rats, respectively).<br />
DNA damage response (incl. oxidative stress): P53 is inducible by DNA damage and<br />
oxidative stress. The upregulation <strong>of</strong> several p53 pathway genes, as observed in Eker and<br />
wild type rats, was therefore summarized as DNA damage response (incl. oxidative stress).<br />
Most genes involved in this category showed a time-dependent increase with highest<br />
deregulation values after 14 days <strong>of</strong> treatment and met the significance criteria in both<br />
strains.<br />
Inhibited cell survival and proliferation: In conjunction with the DNA damage response<br />
described above, downregulation <strong>of</strong> anti-apoptotic genes and genes involved in DNA<br />
replication, cell cycle progression, as well as the upregulation <strong>of</strong> pro-apoptotic genes was<br />
observed. These genes demonstrated a comparable time-dependent increase in<br />
expression as the genes representing the DNA damage response (see above), with highest<br />
deregulation values after 14 days <strong>of</strong> treatment. In contrast to the DNA damage response<br />
genes, many <strong>of</strong> these genes showed an apparently strain-specific deregulation, yet again<br />
with a qualitatively similar expression pattern for most <strong>of</strong> them in both strains. An exception<br />
to the latter observation, were genes directly involved in the G2/M transition <strong>of</strong> the cell<br />
cycle, which were specifically and consistently downregulated in wild type rats.<br />
43
Chapter 3: Manuscript I<br />
Enhanced cell survivial/cell proliferation: Only three genes were assigned to this<br />
category. The Tsc1 (tuberous sclerosis 1) tumor suppressor gene, which is known to be<br />
associated with Tsc2 (Mak and Yeung 2004), was significantly downregulated in Eker but<br />
not wild type rat. Enhanced cell survival and proliferation induced by AA was suggested by<br />
the observed upregulation <strong>of</strong> a positive regulator <strong>of</strong> cell proliferation (KEG1) and the<br />
downregulation <strong>of</strong> a proapoptotic gene (WWOX).<br />
44
Chapter 3: Manuscript I<br />
Table 3.1: Categories <strong>of</strong> genes differentially deregulated by AA in Eker- and wild type rats. For the major toxicological categories the associated genes<br />
are listed together with their Genebank accession number. The main biochemical functions or pathways in which these genes are involved are<br />
indicated in column four. For Eker- and wild type rats, the fold deregulation ratios <strong>of</strong> genes that were significantly deregulated over all timepoints<br />
according to N-way ANOVA are shown in the last eight columns. Genes meeting the significance criteria are indicated in bold.<br />
Toxicological<br />
category<br />
Metabolism and<br />
Biotransformation<br />
Accession<br />
number<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
AI233740 AKR1B8 Aldo-keto reductase 1B8 Biotransformation, Phase I 2.2 3.4 2.5 23.3 1.8 2.0 2.7 2.7<br />
NM_012844 EPHX1 Epoxide hydrolase 1 (microsomal) Biotransformation, Phase I 1.7 1.7 1.5 1.9 1.2 1.3 1.4 1.8<br />
AW142784 CYP4A10 Cytochrome P450 4a10 Biotransformation, Phase I 2.0 1.6 1.6 2.1 1.6 1.8 1.9 1.8<br />
J02679 NQO1 NADPH-Quinone Oxidoreductase Biotransformation, Phase I 2.4 1.9 1.6 3.1 2.0 2.1 1.5 2.0<br />
U27518 UGT2B8 UDP-glucuronosyltransferase 2B8, micros. Biotransformation, Phase II 6.1 4.5 3.3 4.7 3.1 2.5 1.7 1.7<br />
NM_031980 UGT2B12 UDP-glucuronosyltransferase 2B12, micros. Biotransformation, Phase II 2.0 2.1 1.9 2.2 2.2 2.0 2.0 1.9<br />
M31109 UGT2B3 UDP-glucuronosyltransferase 2B3, micros. Biotransformation, Phase II 4.1 4.6 4.5 6.7 2.7 2.0 4.7 3.6<br />
M28241 GSTM1 Glutathione-S-transferase M1 Biotransformation, Phase II 4.2 5.5 5.4 15.8 3.9 3.1 3.6 7.1<br />
NM_031154 GSTM3 Glutathione-S-transferase M3 Biotransformation, Phase II 1.1 2.0 1.6 2.0 1.3 1.3 1.4 2.1<br />
NM_053293 GSTT1 Glutathione-S-transferase T1 Biotransformation, Phase II 1.3 1.6 2.0 2.0 1.3 1.3 1.6 1.8<br />
AA945082 GSTA2 Glutathione-S-transferase A2 Biotransformation, Phase II 4.1 4.5 2.9 3.8 3.9 4.3 4.2 5.0<br />
X02904 GSTP1 Glutathione-S-transferase P1 Biotransformation, Phase II 4.8 5.7 3.9 4.8 4.9 4.9 5.3 6.0<br />
NM_013215 AKR7A3 Aldo-keto reductase family 7, member A3 Biotransformation, Phase I 1.8 1.7 1.5 2.0 1.5 1.4 1.5 1.7<br />
NM_017084 GNMT Glycine-N-methyltransferase Biotransformation, Phase I 1.4 1.6 1.8 1.6 1.2 1.2 1.0 1.9<br />
NM_133586 CES2 Carboxylesterase 2 (intestine, liver) Biotransformation, Phase I -1.1 1.1 1.1 2.1 -1.4 1.1 1.5 2.9<br />
AW142784 CYP51 Cytochrome P450 51 Biotransformation, Phase I 1.1 -1.4 -1.7 -1.1 1.0 1.0 -1.6 -1.1<br />
BI285792 GSTM7-7 Similar to glutathione transferase GSTM7-7 Biotransformation, Phase II 1.5 1.7 1.5 2.1 1.3 1.5 1.3 1.5<br />
AI169331 GSTM2 Glutathione-S-transferase M2 Biotransformation, Phase II 1.1 1.2 1.7 2.1 1.0 1.2 1.3 1.9<br />
AF461738 UGT1A UDP-glucuronosyltransferase 1A, micros. Biotransformation, Phase II 1.9 1.7 1.5 1.8 1.5 1.6 1.6 1.5<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
45
DNA damage<br />
response (incl.<br />
oxidative stress<br />
response)<br />
Inhibited cell<br />
survival and<br />
proliferation<br />
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
NM_133547 SULT1C2 Sulfotransferase K1 (SULTK1) Biotransformation, Phase II -1.3 -1.2 -1.3 -2.0 1.0 -1.3 -1.1 -1.3<br />
AY082609 MDR1B Mutlidrug resistance-1B (ABCB1b) Drug transport 1.1 1.2 1.2 5.5 -1,1 1.1 1.6 2.5<br />
NM_012833 MRP2 Multidrug resistance-associated protein 2 Drug transport 1.5 1.5 1.2 4.6 1.1 1.3 1.1 1.6<br />
L46791 CES3 Carboxylesterase 3 Biotransformation, Phase I 1.1 1.5 1.5 1.7 1.4 1.6 1.6 2.3<br />
NM_134349 MGST1 Microsomal glutathione S-transferase 1 Biotransformation, Phase II 1.5 1.4 1.3 1.5 1.6 1.4 1.5 1.8<br />
AF072816 MRP3 Multidrug resistance-associated protein 3 Drug transport 4.1 -1.7 1.3 2.2 2.3 1.3 1.9 7.8<br />
NM_012861 MGMT O6-methylguanine-DNA methyltranferase DNA repair -1.1 1.2 1.7 3.1 1.1 1.2 1.9 2.7<br />
AW520812 PHLDA3 Pleckstrin homology-like domain family A3 DNA damage response 1.1 1.1 1.3 3.2 1.4 1.4 2.4 3.2<br />
Q64315 CDKN1A Cyclin-dependent kinase inhibitor 1A (p21) Cell cycle checkpoint 1.5 1.1 6.9 14.2 1.1 2.7 2.8 9.8<br />
NM_031821 SNK Serum-inducible kinase (PLK2) Cell cycle checkpoint 1.2 1.3 2.2 6.6 1.4 1.2 2.0 4.3<br />
NM_012923 CCNG1 Cyclin G1 Cell cycle checkpoint 1.1 1.5 2.3 3.8 1.1 1.6 2.1 4.1<br />
BI296301 MDM2 Ubiquitin E3 ligase Mdm2 (predicted) Cell cycle checkpoint 1.0 1.7 1.4 4.9 -1.2 1.0 1.9 2.4<br />
NM_022547 FTHFD 10-formyltetrahydr<strong>of</strong>olate dehydrogenase Cell cycle 2.0 1.9 1.5 1.8 1.7 1.7 1.6 1.8<br />
AI411345 PRODH Similar to Proline dehydrogenase (oxidase) 1 Proapoptotic 1.4 2.5 1.7 2.0 1.2 1.7 1.8 1.9<br />
NM_057153 OXR1 Oxidation resistance 1 DNA damage response -1.1 1.2 1.6 1.5 1.2 -1.1 1.2 1.7<br />
AA801395 TNFAIP8 Tumor necrosis factor α-induced protein 8 Antiapoptotic 1.1 -1.3 -1.4 -2.5 1.0 1.1 1.2 -1.4<br />
M14050 HSPA5 Heat shock 70kD protein 5 (GRP78) Antiapoptotic 1.5 -1.7 -1.6 -1.4 1.0 1.2 1.1 -1.3<br />
AF106659 USP2 Ubiquitin-specific protease 2 Proapoptotic -1.1 1.9 1.6 3.5 1.0 2.0 1.5 1.4<br />
BF288101 SNN Stannin Proapoptotic -1.2 1.1 1.1 1.9 -1.1 1.1 1.0 1.8<br />
AI714002 Ki-67 (predicted) DNA replication -2.4 -2.2 -1.9 -1.2 1.5 -1.4 1.0 -1.4<br />
AI072892 FRZB Frizzled-related protein Signalling cascades -1.1 1.2 1.8 1.3 1.6 1.2 1.0 1.3<br />
M15481 IGF1 Insulin-like growth factor 1 Regulation <strong>of</strong> proliferation -1.1 -1.2 -1.7 -2.4 1.0 1.2 -1.1 -2.0<br />
NM_022266 CTGF Connective tissue growth factor Cell adhesion / migration 1.2 -1.1 -1.1 -1.9 1.4 1.1 1.2 -1.9<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
46
Enhanced cell<br />
survival/<br />
proliferation<br />
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
BI275994 TGM2 Tissue-type transglutaminase Cell adhesion / migration -1.7 -1.5 -1.5 -2.9 -1.9 -2.0 -1.9 -1.9<br />
AI102530 NAB2 Ngfi-A binding protein 2 (predicted) Transcriptional corepressor -1.1 1.2 1.8 2.0 1.1 1.0 1.0 1.7<br />
BF417638 CDCA3 Cell division cycle associated 3 (TOME1) Cell cycle (G2/M) 1.1 -1.1 -1.3 1.3 -1.2 -1.7 -1.6 -1.5<br />
X64589 CCNB1 Cyclin B1, G2/M-specific Cell cycle (G2/M) 1.0 -1.3 -1.3 1.0 -1.4 -2.2 -1.3 -1.6<br />
AI171185 HMMR Hyaluronan mediated motility receptor Cell cycle (G2/M) 1.2 -1.4 -1.4 1.2 -1.6 -2.0 -1.4 -1.7<br />
NM_019296 CDC2 Cell division cycle 2 protein kinase (CDK1 Cyclin<br />
dependent kinase 1)<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
Cell cycle (G2/M) -1.2 -1.4 -2.5 1.2 -1.5 -2.4 -2.0 -1.6<br />
AA944180 CKS2 Cyclin-dependent kinases regulatory subunit 2 Cell cycle (G2/M) 1.0 1.1 -1.7 2.5 -1.6 -2.3 -2.6 -1.3<br />
BF396602 SFRP2 secreted frizzled-related protein 2 Antiapoptotic -1.1 -1.2 -1.6 -1.4 -1.1 -1.1 -1.4 -1.7<br />
NM_012593 KLK7 Kallikrein 7 Antiapoptotic 1,2 -1.3 -1.8 -2.2 -1.4 -1.1 -1.2 -1.8<br />
NM_021854 TSC1 Tuberous sclerosis 1 Tumor suppressor gene -1.9 -2.1 -1.3 -2.3 -1.1 -1.8 1.0 -1.5<br />
BE098555 WWOX WW-domain oxidoreductase (predicted) Proapoptotic -1.4 -1.5 -1.5 -1.7 -1.5 -1.5 -1.9 -1.9<br />
NM_134330 KEG1 kidney expressed gene 1 Regulation <strong>of</strong> proliferation 1.2 1.0 1.8 1.3 1.7 1.2 1.8 2.2<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
47
3.4.7 Genes deregulated by ochratoxin A<br />
Chapter 3: Manuscript I<br />
Biotransformation: OTA treatment downregulated the expression <strong>of</strong> several phase I and<br />
phase II enzymes and drug transporter genes in both strains. While this effect appeared<br />
constant over time for some genes, most deregulated genes demonstrated an enhanced<br />
downregulation with increasing exposure time. Besides the downregulation <strong>of</strong> numerous<br />
genes coding for components <strong>of</strong> the biotransformation machinery, OTA treatment resulted<br />
in an increased upregulation <strong>of</strong> the phase I gene CYP4A12 in both strains and a consistent<br />
upregulation <strong>of</strong> phase I ALDH6A1 in Eker rats.<br />
DNA damage response (incl. oxidative stress): OTA treatment led to significant<br />
upregulation <strong>of</strong> p53 pathway genes in both strains. However, the upregulation <strong>of</strong> these<br />
genes differed between the two strains. Eker rather than wild type rats demonstrated an<br />
upregulation <strong>of</strong> genes known to be involved in oxidative stress responses as well as<br />
downregulation <strong>of</strong> genes that code for products with extra-cellular antioxidant activities. The<br />
latter genes and the upregulated KEAP1, which suppresses the transactivation <strong>of</strong><br />
antioxidant responsive elements, were categorized as “enhanced oxidative stress”.<br />
Cellular stress: A general stress response, as indicated primarily by the upregulation <strong>of</strong><br />
several components <strong>of</strong> the stress-inducible MAP-kinase pathway, was predominantly<br />
detectable in Eker rats, since the latter genes were not significantly deregulated in wild type<br />
rats.<br />
Inhibited cell survival and proliferation: Predominantly OTA-treated Eker rats presented<br />
with an upregulated expression <strong>of</strong> tumor suppressor-, negative cell proliferation controland<br />
pro-apoptotic response genes. However, in both strains an inhibited cell survival<br />
response could be inferred from the upregulation <strong>of</strong> a pro-apoptotic gene and the<br />
downregulation <strong>of</strong> an anti-apoptotic and a DNA replication gene.<br />
Enhanced cell survival and proliferation: Numerous genes coding for regulators <strong>of</strong> cell<br />
survival signalling pathways, e.g. the IGF-PI3K-PKB pathway, anti-apoptosis, mitosis,<br />
growth and proliferation (incl. proto-oncogenes) were primarily upregulated in Eker rats,<br />
while a tumor suppressor gene and a gene coding for a signal cascade inhibitor (OVCA2)<br />
was downregulated. In comparison, the latter genes were not or only marginally<br />
deregulated in the wild type rats treated with OTA.<br />
48
Chapter 3: Manuscript I<br />
Cell cycle progression and mitosis: Upregulation <strong>of</strong> genes, more directly involved in the<br />
cell cycle progression, was categorized separately as they appeared as an entity distinct<br />
from the category “enhanced cell survival and proliferation”. Remarkably, gene deregulation<br />
that would further cell cycle progression was exclusively observed in Eker rats.<br />
Cell structure remodeling: Similar to the effects observed for genes involved in enhanced<br />
cell survival and proliferation (see above), genes coding for components and regulators <strong>of</strong><br />
the cytoskeleton, altered cell cell-adhesion or communication, and components <strong>of</strong> the Rac<br />
and Rho signalling were almost exclusively upregulated in Eker rats.<br />
Epithelial mesenchymal transition (EMT)/ Fibrosis: This category includes regulators <strong>of</strong><br />
cell proliferation, growth factor activation, cell adhesion and extracellular matrix (ECM)<br />
known to be associated in the process <strong>of</strong> EMT and/ or fibrosis, and which could be<br />
associated with the progression <strong>of</strong> renal and urothelial tumors. Such increased expression<br />
<strong>of</strong> components <strong>of</strong> the TGFß pathway and <strong>of</strong> a hepatocyte growth factor activator-inhibitor<br />
gene in conjunction with a downregulated expression <strong>of</strong> extracellular matrix protease genes<br />
were predominantly deregulated in Eker rats.<br />
49
Chapter 3: Manuscript I<br />
Table 3.2: Categories <strong>of</strong> genes differentially deregulated by OTA in Eker- and wild type rats. For the major pathophysiological categories the associated<br />
genes are listed together with their Genebank accession number. The main biochemical functions or pathways in which these genes are<br />
involved are indicated in column four. For Eker- and wild type rats, the fold deregulation ratios <strong>of</strong> genes that were significantly deregulated<br />
over all time points according to N-way ANOVA are shown in the last eight columns. Genes meeting the significance criteria are indicated in<br />
bold.<br />
Toxicological<br />
category<br />
Metabolism and<br />
Biotransformation<br />
DNA damage<br />
response<br />
Accession<br />
number<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
NM_031565 CES1B Carboxylesterase RL1 Biotransformation phase I 1.0 -1.7 -2.7 -1.9 -1.1 -1.5 -2.2 -2.2<br />
L46791 CES3 Carboxylesterase 3 Biotransformation phase I 1.0 -1.4 -2.7 -3.4 -1.2 -1.7 -2.7 -1.9<br />
NM_020538 AADAC Arylacetamide deacetylase Biotransformation phase I -1.4 -1.3 -1.9 -2.5 1.0 -1.2 -1.5 -2.0<br />
AW142784 CYP4A12 Cytochrome P450 4a12 Biotransformation phase I 1.0 1.6 1.5 1.9 1.7 1.8 2.0 1.9<br />
NM_133558 CML1 Camello-like 1 Biotransformation phase II -1.1 -1.3 -2.6 -2.9 -1.2 -1.1 -2.0 -2.1<br />
NM_022635 NAT8 N-acetyltransferase 8 Biotransformation phase II -1.2 -2.1 -3.3 -2.9 -1.3 -1.5 -2.4 -2.9<br />
AI072042 GGT6 Gamma-glutamyl transpeptidase type VI Biotransformation phase II -1.4 -2.0 -2.9 -2.7 -1.3 -1.3 -1.6 -1.8<br />
NM_134349 MGST1 Microsomal glutathione S-transferase 1 Biotransformation phase II -1.3 -1.4 -1.8 -1.8 -1.1 -1.3 -1.7 -1.7<br />
NM_134369 CYP2T1 Cytochrome P450 2T1 Biotransformation phase I 1.0 -1.1 -2.0 -1.9 1.1 -2.0 -1.3 -1.3<br />
BF283000 WBSCR21 Williams-Beuren syndr. chromosome region 21 Biotransformation phase I 1.1 -1.4 -1.7 -1.9 -1.1 -1.1 -1.3 -1.4<br />
AI407458 ALDH6A1 Aldehyde dehydrogenase 6A1 Biotransformation phase I 1.8 2.0 1.6 2.1 1.1 1.1 -1.1 -1.1<br />
NM_022270 OCTN1 Organic cation carnitine transporter 1 Drug transport -1.1 -1.1 -2.1 -1.8 1.0 -1.1 -1.5 -1.2<br />
NM_017224 OAT1 Organic anion transporter 1 (SLC22A6) Drug transport -1.1 1.1 -1.9 -1.7 1.1 -1.2 -1.3 -1.6<br />
NM_019303 CYP2F4 Cytochrome P450 2F4 Biotransformation phase I 1.3 1.0 -1.4 -1.8 1.2 -1.8 -1.8 -1.1<br />
NM_017084 GNMT Glycine-N-methyltransferase Biotransformation phase II 1.0 -1.4 -1.4 -1.6 -1.4 -1.5 -1.9 -1.5<br />
U76379 OCT1 Organic cation transporter 1 (SLC22A1) Drug transport 1.1 -1.3 -1.4 -1.7 -1.1 -1.4 -1.7 -1.7<br />
BM388545 SUPT16H Suppressor <strong>of</strong> Ty 16 homolog (predicted) DNA damage repair 1.0 1.8 1.4 1.8 1.6 1.8 2.1 2.4<br />
NM_053677 CHEK2 Checkpoint kinase 2 Cell cycle checkpoint 1.0 1.8 2.8 2.2 1.0 -1.1 1.1 -1.1<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
50
Oxidative<br />
stress<br />
response<br />
Enhanced<br />
oxidative stress<br />
Cellular stress<br />
Reduced cell<br />
survival/<br />
proliferation<br />
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
BF548539 MDM2 Ubiquitin E3 ligase Mdm2 (predicted) Cell cycle checkpoint 5.3 4.4 1.8 2.4 -1.4 -1.4 -1.4 1.0<br />
AI178158 RBBP6 retinoblastoma binding protein 6 (predicted) Cell cycle checkpoint 1.0 1.1 1.0 1.0 1.2 1.5 2.0 1.5<br />
NM_019192 SEPP1 Selenoprotein P Oxidative stress response 1.1 3.5 3.4 3.9 4.1 4.2 4.9 4.2<br />
NM_019235 GGTLA1 Gamma-glutamyltransferase-like activity 1 Oxidative stress response 1.4 2.6 1.8 4.2 1.3 -1.4 2.1 1.0<br />
NM_031614 TXNRD1 Thioredoxin reductase 1, cytoplasmic Oxidative stress response 1.7 1.7 1.4 1.7 -1.1 1.0 -1.3 -1.1<br />
AI231438 CN1 Carnosine dipeptidase 1 Oxidative stress response -1.2 -1.2 -2.9 -2.5 -1.2 -1.5 -1.8 -1.3<br />
BM386741 HSP40-3 Heat shock protein hsp40-3 (predicted) Protein folding (cytosol) -1.3 -1.8 -1.7 -1.4 1.0 1.0 -1.5 -1.1<br />
BF414210 KEAP1 Kelch-like ECH-associated protein 1 Regulation <strong>of</strong> transcription 1.7 1.6 2.0 2.4 1.1 1.0 -1.1 1.0<br />
NM_053307 MSRA Methionine sulfoxide reductase A Oxidative stress response 1.4 1.4 1.0 -1.3 1.0 -1.6 -2.0 -1.8<br />
BM387750 DUSP11 Dual specificity phosphatase 11 MAPK pathway -1.1 -1.7 -1.7 -1.5 -1.5 -1.5 -1.4 -1.8<br />
BE110108 DUSP1 Dual specificity protein phosphatase 1 MAPK pathway 1.9 1.6 2.3 2.3 -1.1 -1.3 -1.3 1.0<br />
NM_031032 GMFB Glia maturation factor beta MAPK pathway 2.4 4.0 5.9 5.6 1.5 1.3 2.7 1.5<br />
AAH61870 JNK2 c-Jun N-terminal kinase 2 MAPK pathway 1.8 1.3 1.4 1.3 1.2 -1.4 -1.1 -1.3<br />
AA851481 BRE Brain and reproductive organ-expressed protein MAPK pathway 1.8 1.7 1.5 1.7 1.1 -1.1 -1.2 -1.2<br />
L48060 PRLR Prolactin receptor MAPK pathway 1.5 2.1 1.3 1.7 1.1 1.2 1.2 1.2<br />
NM_023090 HIF2 alpha Hypoxia-inducible factor 2 alpha HIF pathway 2.5 1.8 2.0 1.7 1.2 1.1 -1.2 1.0<br />
BF403837 NFE2L1 Nuclear factor (erythroid-derived) 2 like 1 Regulation <strong>of</strong> gene expression 1.4 1.3 1.3 1.7 1.0 -1.3 -1.1 -1.1<br />
AI598399 RBM3 RNA-binding motif protein 3 RNA metabolism 1.0 1.0 -1.3 1.0 1.4 1.1 1.2 1.8<br />
BF281976 POLD4 Polymerase (DNA-directed), delta 4 DNA replication 2.0 -2.2 -2.0 -2.1 -1.1 -1.4 -1.6 -2.0<br />
NM_057138 CFLAR CASP8 and FADD-like apoptosis regulator Antiapoptotic -1.3 -1.9 -2.9 -4.3 1.2 -3.6 -2.9 -3.4<br />
AW253957 ENDOG Endonuclease G Proapoptotic 1.1 1.9 2.3 1.9 1.3 1.4 1.8 2.5<br />
AA944698 BAT3 HLA-B associated transcript 3 Proapoptotic 6.5 4.4 3.9 5.0 1.2 1.3 1.1 -1.1<br />
NM_019208 MEN1 Multiple endocrine neoplasia 1 Tumor suppressor gene 1.9 1.2 1.6 1.9 -1.1 -1.2 1.1 -1.1<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
51
Increased cell<br />
survival/<br />
proliferation<br />
Accession<br />
number<br />
AI113091 TSSC4 Tumor-suppressing subchromosomal transferable<br />
fragment 4 (predicted)<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
Tumor suppressor gene 1.8 1.4 1.2 2.2 -1.1 1.0 -1.1 1.0<br />
AA996685 PKIA cAMP-dependent protein kinase inhibitor alpha Signalling cascades 3.0 3.1 2.5 2.4 -1.3 1.2 1.3 2.6<br />
BI290885 FSTL Follistatin-like 1 Signalling cascades 1.9 1.2 1.7 1.9 1.1 -1.1 1.1 1.3<br />
BG663107 AKAP12 A kinase anchor protein 12 (gravin) Signalling cascades 1.7 1.4 2.0 1.5 -1.1 -1.2 -1.1 -1.2<br />
M86708 ID1 Inhibitor <strong>of</strong> DNA binding 1 Regulation <strong>of</strong> transcription 1.1 -1.3 1.4 1.0 -1.3 -1.1 -1.3 -1.9<br />
NM_031546 RGN Regucalcin (SMP-30) Antiapoptotic -1.4 1.9 -3.7 -4.5 1.2 -1.7 -1.9 -2.6<br />
BE108969 IGFBP-4 Insulin-like growth factor binding protein 4 (IGF)-PI3K-AKT pathway -1.3 -1.4 -3.0 -3.8 1.0 -1.3 -3.0 -2.8<br />
AI145815 MKRN-1 Makorin-1 Protein metabolism -1.1 -1.6 -1.9 -1.9 -1.3 -1.7 -1.7 -1.6<br />
BM986536 HIST1H4I Histone 1 H4i Nucleosome assembly 1.4 3.1 2.7 3.0 1.3 2.2 2.5 2.3<br />
NM_022265 PDC4 Programmed cell death 4 Antiapoptotic 1.5 2.1 1.7 1.9 1.0 -1.1 -1.3 1.1<br />
BE112895 PEA15 Phosphoprotein enriched in astrocytes 15 Antiapoptotic -1.5 1.3 1.9 1.9 1.2 1.1 1.4 1.1<br />
NM_021846 MCL1 Myeloid cell leukemia 1 Antiapoptotic 1.6 1.4 1.5 1.8 1.1 -1.3 -1.1 -1.4<br />
NM_022943 MERTK C-Mer proto-oncogene tyrosine kinase Protooncogene 3.3 4.9 4.3 2.9 -1.3 1.0 1.2 1.1<br />
NM_012807 SMOH Smoothened Protooncogene 1.8 1.3 1.6 1.5 -1.1 1.0 -1.3 1.0<br />
NM_022264 V-KIT Hardy-Zuckerman 4 feline sarcoma viral oncogene<br />
homolog (c-Kit)<br />
NM_012843 TMP Tumor-associated membrane protein (EMP1 Epithelial<br />
membrane protein 1)<br />
Protooncogene 2.9 3.7 2.4 2.5 -1.2 -1.3 -1.5 1.2<br />
Protooncogene 2.8 1.7 2.2 4.3 1.0 -1.3 -1.2 -1.3<br />
AA943541 OVCA2 Candidate tumor suppressor OVCA2 (predicted) Tumor suppressor gene -1.3 -1.4 -1.6 -1.8 -1.3 -1.1 -1.4 -1.6<br />
NM_053481 PIK3CB Phosphatidylinositol-4,5-bisphosphate 3-kinase<br />
catalytic beta subunit<br />
(IGF)-PI3K-AKT pathway 2.0 2.9 -1.3 2.0 1.3 1.0 1.0 -1.1<br />
AI102030 AKT1S1 AKT1 substrate 1 (proline-rich) (IGF)-PI3K-AKT pathway 2.4 4.0 2.0 2.1 1.0 1.0 -1.1 -1.1<br />
AI105076 AKT2 Thymoma viral proto-oncogene 2 (IGF)-PI3K-AKT pathway 1.5 1.5 1.2 1.9 1.0 1.1 -1.1 -1.2<br />
AW434982 SBF1 Similar to SET binding factor 1 (IGF)-PI3K-AKT pathway 2.4 3.5 2.2 3.6 1.2 1.0 -1.1 1.0<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
52
Cell cycle<br />
progression<br />
and mitosis<br />
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
NM_012593 KLK7 Glandular kallikrein 7 Protein metabolism 1.8 1.3 1.9 1.6 -1.1 -1.2 -1.2 -1.1<br />
BI300565 ADAM10 A-disintegrin and metalloprotease domain 10 Protein metabolism 1.6 1.8 1.3 1.4 1.1 -1.3 -1.3 -1.3<br />
NM_053665 AKAP1 A kinase (PRKA) anchor protein 1 Signalling cascades 5.9 3.8 4.1 7.8 1.0 1.2 -1.1 1.0<br />
NM_017094 GHR Growth hormone receptor Signalling cascades 1.7 2.8 1.2 2.0 1.2 -1.1 1.0 1.1<br />
NM_012850 GHRHR Growth hormone releasing hormone receptor Signalling cascades 1.2 1.3 1.8 1.7 1.6 1.4 1.1 -1.1<br />
BI294916 KLF2 Kruppel-like factor 2 Regulation <strong>of</strong> gene expression 1.6 1.4 1.9 1.9 1.1 -1.1 -1.2 1.2<br />
NM_131904 MGEA5 Meningioma expressed antigen 5 Protein metabolism 2.1 1.7 1.8 1.4 -1.1 -1.1 -1.5 -1.7<br />
AF080594 VEGF Vascular endothelial growth factor Angiogenesis 2.3 2.0 1.6 2.6 1.1 -1.3 1.0 1.0<br />
AW524517 VEZF1 Vascular endothelial zinc finger 1 Angiogenesis 1.9 1.6 1.2 1.4 1.1 1.0 -1.1 1.1<br />
NM_133569 ANGPTL2 Angiopoietin-like 2 Angiogenesis 1.3 1.9 1.2 1.9 1.2 -1.1 -1.1 1.1<br />
NM_017089 EPHB1 Ephrin B1 Angiogenesis 1.9 1.6 1.6 1.9 1.1 1.0 -1.3 1.0<br />
AB035507 MCAM Melanoma cell adhesion molecule Angiogenesis 2.1 1.5 1.6 1.3 1.0 -1.3 1.1 -1.1<br />
NM_022615 TOP1 DNA topoisomerase 1 DNA replication 1.5 1.8 1.6 2.2 1.0 -1.1 1.0 1.1<br />
BM385181 SMARCA4 SWI-SNF related matrix associated actin<br />
dependent regulator <strong>of</strong> chromatin subfamily A4<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
Chromatin remodelling 1.3 1.1 1.7 1.7 -1.1 1.4 1.0 1.1<br />
U75920 MAPRE1 Microtubule-associated protein RP/EB family 1 Mitotic spindle formation 2.4 2.6 2.0 2.8 1.1 1.0 -1.1 1.0<br />
AJ306292 AJUBA Ajuba protein Mitotic spindle formation 1.4 1.3 1.3 1.8 1.0 -1.2 -1.1 -1.2<br />
BE118382 NEK9 NIMA related kinase 9 Mitotic spindle formation 2.1 1.9 1.8 2.4 1.0 1.0 1.0 -1.1<br />
U77583 CSNK1A1 Casein kinase 1, alpha 1 Mitotic spindle formation 1.6 1.5 1.6 1.8 1.1 1.0 -1.1 1.0<br />
NM_057148 SEPT2 Septin 2 Mitotic spindle formation 2.5 2.2 2.1 2.5 -1.1 -1.1 1.0 -1.1<br />
NM_080907 PPP4R1 Protein phosphatase 4, regulatory subunit 1 Mitotic spindle formation 2.1 1.7 1.3 2.0 1.0 1.0 -1.2 -1.1<br />
NM_013194 NMMHC-A Nonmuscle myosin heavy chain A Mitotic spindle formation 1.6 1.6 1.7 2.0 1.3 -1.1 1.1 1.2<br />
Cell structure AA955773 PCDH1 Protocadherin 1 Cell adhesion molecule -1.1 -1.7 -1.6 -1.9 -1.3 -1.7 -2.2 -1.8<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
53
emodelling<br />
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
BE097805 PCDH4 Protocadherin 4 Cell adhesion molecule 2.7 1.2 3.6 2.3 1.5 1.3 1.6 1.3<br />
AA955091 ITGA6 Integrin alpha 6 Cell adhesion molecule 2.0 1.3 1.7 1.5 1.0 1.0 -1.3 -1.4<br />
NM_031699 CLDN1 Claudin 1 Cell adhesion molecule 1.6 1.2 2.0 1.8 1.0 -1.2 -1.3 -1.2<br />
NM_031329 OCLN Occludin Cell adhesion molecule 2.5 1.9 1.3 2.2 1.4 1.0 -1.1 1.0<br />
NM_013217 AF-6 Afadin Regulation <strong>of</strong> cell adhesion/<br />
migration<br />
BF284125 IQGAP1 IQ motif containing GTPase activating protein 1 Regulation <strong>of</strong> cell adhesion/<br />
migration<br />
AB020726 PODXL Podocalyxin Regulation <strong>of</strong> cell adhesion/<br />
migration<br />
NM_020085 RPTPK Receptor-like protein tyrosine phosphatase kappa<br />
extracellular region<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
1.6 1.7 1.7 1.9 1.0 1.0 -1.2 -1.2<br />
1.3 1.3 1.8 1.8 1.0 -1.1 -1.3 -1.2<br />
1.6 1.3 1.5 1.8 1.2 1.0 1.1 -1.1<br />
Signalling cascades 5.6 5.7 3.2 5.4 1.0 -1.4 -1.8 -1.4<br />
NM_057115 PTPN12 Protein tyrosine phosphatase, non-receptor type12 Signalling cascades 1.9 1.5 2.7 3.6 -1.1 -1.3 -1.2 -1.2<br />
NM_031034 GNA12 Guanine nucleotide-binding protein alpha 12 Signalling cascades 2.2 2.2 1.6 2.4 1.1 1.2 1.1 -1.1<br />
BE115857 PARVA Parvin alpha Cytoskeleton organization 2.4 2.1 1.6 2.1 1.0 1.0 1.1 -1.1<br />
NM_024401 AVIL Advillin (Pervin) Cytoskeleton organization 1.2 1.3 1.9 2.6 1.0 1.1 1.5 1.8<br />
AF054618 CTTN Cortactin Cytoskeleton organization 1.8 1.5 1.9 2.0 1.1 1.0 1.1 1.1<br />
NM_032613 LASP1 LIM and SH3 protein 1 Cytoskeleton organization 3.1 1.8 1.5 2.0 1.1 1.0 -1.2 1.1<br />
NM_023982 ARHGEF11 Rho guanine nucleotide exchange factor 11 Cytoskeleton organization 1.7 2.5 2.8 3.1 1.7 2.1 2.1 1.4<br />
AI170442 DSTN Destrin Cytoskeleton organization 1.5 1.5 1.7 2.0 1.0 1.1 1.0 -1.1<br />
NM_030873 PFN2 Pr<strong>of</strong>ilin 2 Cytoskeleton organization 1.6 2.5 2.1 3.4 1.3 -1.1 -1.2 1.2<br />
AA875047 CCTZ Chaperonin containing T-complex 1 zeta Cytoskeleton organization 9.0 2.0 6.7 5.0 -2.2 2.4 -1.5 -1.1<br />
V01217 ACTB Actin beta Cytoskeleton constituent 2.2 2.1 2.1 3.1 1.0 -1.1 -1.1 1.0<br />
NM_016990 ADD1 Alpha adducin Cytoskeleton constituent 2.0 1.3 1.5 2.1 1.2 1.2 -1.1 1.1<br />
X70706 PLS3 Plastin 3 T-is<strong>of</strong>orm (T-plastin) Cytoskeleton constituent 2.0 2.4 1.8 2.9 1.0 1.1 1.0 1.2<br />
54
EMT / Fibrosis<br />
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
NM_030863 MSN Moesin Cytoskeleton constituent 1.7 1.8 1.6 1.9 1.3 -1.2 1.0 1.1<br />
BF392456 SPNB2 Spectrin beta 2 Cytoskeleton constituent 1.9 1.9 1.5 2.4 1.1 1.1 1.0 1.0<br />
BI284344 KRT2-7 Keratin complex 2, basic, gene 7 (predicted) Cytoskeleton constituent -1.1 -1.4 -1.8 -2.0 1.0 -1.3 -1.4 -1.6<br />
BI295970 TPM3 Tropomyosin 3 gamma (Tropomyosin alpha 3) Cell morphology / motility 1.7 1.7 1.6 2.1 1.0 1.0 -1.1 -1.1<br />
NM_012678 TPM4 Tropomyosin 4 Cell morphology / motility 2.2 1.8 2.3 2.0 1.0 -1.1 -1.4 1.1<br />
NM_053986 MYO1B Myosin IB (Brush border myosin I) Cell morphology / motility 1.7 1.3 1.5 1.8 -1.1 1.1 -1.1 -1.1<br />
NM_023092 MYO1C Myosin 1C (Unconventional myosin Myr2 I heavy<br />
chain)<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
Cell morphology / motility 1.5 1.4 1.7 1.7 1.1 1.1 1.0 1.1<br />
BG380723 CAP1 Adenylyl cyclase-associated protein 1 Cell morphology / motility 2.0 1.5 1.9 2.3 -1.1 1.2 1.0 1.0<br />
AA997129 LAMC1 Laminin gamma 1 Extracellular matrix component 2.2 1.5 1.6 1.6 1.1 -1.2 1.0 1.3<br />
AW435213 NANS N-acetylneuraminic acid synthase (sialic acid<br />
synthase)<br />
Cell adhesion / migration 1.4 1.3 1.5 1.7 1.0 1.1 -1.3 -1.2<br />
U61261 LAMA3 Laminin alpha 3 (Laminin-5 alpha 3) Extracellular matrix component 1.2 1.1 1.2 -1.3 -1.2 -1.7 -1.6 -2.0<br />
BF284673 FGD1 Faciogenital dysplasia homolog Cell adhesion/ migration 1.1 1.3 1.2 1.5 1.5 1.2 1.9 2.2<br />
NM_013085 PLAU Urokinase-type plasminogen activator Cell adhesion/ migration -1.3 -1.3 -2.1 -1.9 -1.3 -2.0 -1.7 -2.1<br />
NM_013143 MEP1A Meprin A alpha-subunit Protein degradation -1.2 1.0 -2.5 -2.9 -1.2 -1.6 -1.6 -2.1<br />
L09653 TGFBR2 TGF-beta receptor type II TGF-beta family pathway 3.9 5.5 5.9 3.4 1.0 -1.4 -1.1 1.0<br />
NM_013130 SMAD1 (MADH1) Mothers against decapentaplegic homolog<br />
1<br />
NM_013095 SMAD3 (MADH3) Mothers against decapentaplegic homolog<br />
3<br />
NM_019275 SMAD4 (MADH4) Mothers against decapentaplegic homolog<br />
4)<br />
TGF-beta family pathway 2.1 1.5 1.6 1.4 1.0 1.1 1.0 1.7<br />
TGF-beta family pathway 2.1 3.5 1.6 2.8 1.2 1.1 -1.1 1.1<br />
TGF-beta family pathway 1.7 1.7 1.3 1.6 -1.2 -1.2 -1.3 -1.6<br />
NM_053357 CTNNB1 Beta-catenin Regulation <strong>of</strong> proliferation 1.7 1.6 1.3 1.9 1.0 -1.2 -1.1 1.0<br />
NM_031317 PDGFC Platelet derived growth factor C Regulation <strong>of</strong> proliferation 2.1 1.7 1.4 2.4 1.0 -1.4 -1.1 -1.1<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
55
Accession<br />
number<br />
Chapter 3: Manuscript I<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point)<br />
BE095528 HAI-1 Hepatocyte growth factor activator inhibitor 1 Regulation <strong>of</strong> proliferation 3.9 9.2 5.4 3.0 1.0 1.0 -1.1 1.0<br />
AF172255 NPHS1 Nephrosis 1 homolog (nephrin) Cell adhesion / migration 1.7 2.3 2.0 1.8 -1.1 1.0 1.0 1.1<br />
NM_012886 TIMP3 Tissue inhibitor <strong>of</strong> metalloproteinase 3 Cell adhesion / migration 1.8 1.4 1.4 1.8 1.0 -1.1 -1.4 -1.7<br />
BI278545 DPT Dermatopontin Cell adhesion / migration 1.6 1.5 1.7 1.8 1.1 1.0 1.2 1.2<br />
NM_012924 CD44 antigen (ECMR-III Extracellular matrix receptor-III) Cell adhesion / migration 2.3 3.3 1.2 1.6 -1.2 -1.1 1.1 1.2<br />
NM_024358 NOTCH2 Notch homolg 2 Differentiation / organogenesis 2.4 2.2 3.2 7.4 1.0 1.0 -1.1 -1.1<br />
AA997458 Similar to CSRP2 Cysteine and glycine-rich protein 2 Differentiation / organogenesis 1.0 -1.5 -1.8 -1.6 -1.3 -1.5 -1.6 -1.6<br />
NM_012901 AMBP Alpha-1-microglobulin/bikunin precursor Extracellular transport 1.1 2.0 2.6 1.9 1.0 1.3 1.8 1.9<br />
NM_017309 CNB Calcineurin B regulatory subunit is<strong>of</strong>orm 1 Signalling cascades 4.5 7.6 1.8 5.8 -1.1 -1.4 -1.3 -1.3<br />
Ek<br />
d1<br />
Ek<br />
d3<br />
Ek<br />
d7<br />
Ek<br />
d14<br />
wt<br />
d1<br />
wt<br />
d3<br />
wt<br />
d7<br />
wt<br />
d14<br />
56
3.5 Discussion<br />
Chapter 3: Manuscript I<br />
As expected for the short-term duration and dose-regimen employed (Mengs et al. 1982;<br />
Boorman et al. 1992), neither AA nor OTA induced pronounced non-neoplastic renal<br />
pathology in either strain <strong>of</strong> rats. While AA treated rats presented with a slightly higher<br />
inflammatory response than the corresponding controls, OTA treated rats responded with<br />
the typical pathological changes, e.g. karyomegaly, apoptosis, cell shedding, regenerative<br />
proliferation and an inflammatory response in the renal cortex, as reported earlier (Boorman<br />
et al. 1992; Rasonyi et al. 1999). However, the AA and OTA induced non-neoplastic<br />
pathology was clearly distinct from one another, and this was also reflected in the cell<br />
proliferation assessment (Figure 3.1). Lack <strong>of</strong> overt cell necrosis and subsequent cell<br />
shedding and regeneration in the tubuli <strong>of</strong> AA treated rats coincided with absence <strong>of</strong><br />
increased cell proliferation, demonstrating that AA had neither a cytotoxic nor a mitogenic<br />
effect on the kidneys <strong>of</strong> either strain <strong>of</strong> rats. In contrast, OTA- induced tubular toxicity and<br />
possibly mitogenic activity was reflected in the overt cell regeneration observed with the<br />
histopathological as well as immunohistological assessments. Moreover, the observed<br />
effects increased with the duration <strong>of</strong> treatment and were comparable in extent <strong>of</strong> effect in<br />
both strains. Based on above observations, the age <strong>of</strong> animals used and the duration and<br />
type <strong>of</strong> exposure, the expectation was confirmed that no increased prevalence (no. <strong>of</strong><br />
animals affected) or number (no. <strong>of</strong> lesions per animal) <strong>of</strong> preneoplastic and neoplastic<br />
lesions were to be encountered in the wild type control and AA treated rats. Moreover, AA<br />
treatment <strong>of</strong> the arguably more susceptible Eker rats did not increase the prevalence or<br />
number <strong>of</strong> lesions, suggesting that the Tsc2 mutation was not critical for AA induced<br />
effects. The two carcinomas observed on day 14 in AA exposed Eker rats cannot be<br />
conclusively associated with AA treatment, as adenomas were also observed in the<br />
corresponding Eker rat control and AA treatment groups already on day 1 (Supplemetary<br />
table 3.2). As the number <strong>of</strong> animals employed for this experiment limited statistical<br />
evaluation, a progression <strong>of</strong> already present adenomas to carcinomas due to AA treatment<br />
cannot be excluded.<br />
Similar to AA, OTA treatment was not associated with an increased prevalence or number<br />
<strong>of</strong> lesions in wild type rats, despite the clearly enhanced cell proliferative response in the<br />
renal cortex. The latter observation corroborates numerous earlier findings (Boorman et al.<br />
1992; Rasonyi et al. 1999) suggesting that non-genotoxic compound- induced renal<br />
carcinogenesis can only be observed after a prolonged compound exposure period. In<br />
contrast to the situation in wild type, Eker rats presented with a significantly increased<br />
57
Chapter 3: Manuscript I<br />
prevalence and number <strong>of</strong> AT on day 14 <strong>of</strong> OTA treatment (Figure 3.1, E, Supplementary<br />
table 3.2). The latter finding suggests that the increased cell proliferative stimulus (Figure<br />
3.1, C) provided for an enhanced manifestation <strong>of</strong> the predisposition for renal neoplasia<br />
mediated by the Tsc2 mutation and thus a direct or indirect interaction <strong>of</strong> OTA with Tsc2<br />
(tuberin) pathway.<br />
At the outset <strong>of</strong> this experiment it was assumed that each compound would induce a<br />
distinct gene expression pr<strong>of</strong>ile, which is reflected by the short-term pathology but also<br />
displays characteristic genes representative for pathways involved in the compoundspecific<br />
type <strong>of</strong> renal carcinogenesis. Consequently, the genotoxic AA was expected to<br />
induce a gene expression pr<strong>of</strong>ile, most likely involving genes <strong>of</strong> cell cycle arrest and DNA<br />
damage repair but not cell proliferation and most likely would not involve Tsc2. Indeed, the<br />
expression pr<strong>of</strong>iles <strong>of</strong> AA- treated Eker and wild type rats (Figure 3.3) were not distinctly<br />
different and were comparable to expression pr<strong>of</strong>iles obtained with AA in kidneys <strong>of</strong> Big<br />
Blue transgenic F344 rats (Chen et al. 2006). Several phase I genes were deregulated by<br />
AA treatment from the first day <strong>of</strong> exposure. The gene product <strong>of</strong> one <strong>of</strong> the upregulated<br />
phase I genes, NQOI, was previously demonstrated to be capable <strong>of</strong> reducing the nitro<br />
group <strong>of</strong> AAI leading to metabolic activation. NQOI up-regulation could therefore be at least<br />
partly responsible for DNA adduct formation (Stiborova et al. 2003). Although DNA adduct<br />
formation could not be measured with the study design employed, previously published<br />
reports support the assumption that AA DNA-adducts are formed (Dong et al. 2006).<br />
Indeed, upregulation <strong>of</strong> several p53 pathway genes, including p53 target genes carrying a<br />
p53 consensus sequence in the promotor region, was most prominent on day 7 and 14 <strong>of</strong><br />
exposure, indicating a DNA damage response upon bioactivation <strong>of</strong> AA. This interpretation<br />
is further supported by the fact that several <strong>of</strong> the p53 target genes, e.g. MDM2, p21 or<br />
CCNG1, have also been shown to be upregulated in rat liver after short-term exposure to<br />
different known genotoxic compounds (Ellinger-Ziegelbauer et al. 2005). DNA damage and<br />
activation <strong>of</strong> p53 pathway genes are expected to result in cell cycle arrest (Figure 3.4, A),<br />
followed by damage repair or programmed cell death (Sionov and Haupt 1999). AAtreatment<br />
led to the deregulation <strong>of</strong> several genes involved in apoptosis, with a comparable<br />
time pr<strong>of</strong>ile to the p53- pathway and target genes. Although cell cycle components are<br />
predominantly regulated on the protein level, downregulation <strong>of</strong> genes crucial for the G2/M<br />
transition, e.g. CDC2, Cyclin-B, TOME1 and CKS2 and downregulation <strong>of</strong> genes required<br />
for mitotic spindle formation like TUBA1 or HMMR, suggest a G2 arrest. Deregulation <strong>of</strong> the<br />
latter genes in wild type rats only, may be explained by 12-fold lower PI3K mRNA<br />
expression compared to Eker rats (Sen et al. 2004). Increasing evidence suggest that<br />
constitutive activation <strong>of</strong> the PI3K pathway could lead to defects in DNA damage checkpoint<br />
control (Liang and Slingerland 2003). Consequently, AA-induced G2/M arrest would<br />
58
Chapter 3: Manuscript I<br />
therefore not be readily detectable in Eker rats, although they responded with upregulation<br />
<strong>of</strong> proapoptotic genes, as well as downregulation <strong>of</strong> Ki-67, an observation also supported<br />
by the cell proliferation and pathological analysis (Figure 3.1, A and B).<br />
Figure 3.4: Postulated mechanistic pathways <strong>of</strong> AA (A) and OTA (B) induced toxicity in Eker and<br />
wild type rats and potential influence <strong>of</strong> Tsc2 in the manifestation <strong>of</strong> short- and longterm<br />
effects discussed. Black: pathways and processes suggested from the literature;<br />
Blue: pathways and processes implicated by gene expression analysis; Red:<br />
Pathways and processes implicated by histopathology and corroborated by gene<br />
expression analysis.<br />
In contrast to AA, the gene expression pr<strong>of</strong>iles <strong>of</strong> OTA-treated Eker and wild type rats were<br />
distinctly different (Figure 3.3). However, in both strains up-regulation <strong>of</strong> CYP4A12 and<br />
down-regulation <strong>of</strong> other phase I and II genes could provide for increased generation <strong>of</strong><br />
reactive oxygen species (ROS) and enhanced oxidative DNA damage (Kamp et al. 2005;<br />
Mally et al. 2005), that could lead to cellular damage and regenerative cell proliferation.<br />
Indeed, enhanced proximal tubular damage and cell regeneration/proliferation (Figure 3.1,<br />
C and D) was observed in both rat strains, corroborating the above interpretation. OTA<br />
induced oxidative stress and ensuing DNA damage in combination with enhanced cell<br />
59
Chapter 3: Manuscript I<br />
proliferation could increase the likelihood <strong>of</strong> neoplastic transformation (Dietrich and<br />
Swenberg 1991). Presence <strong>of</strong> oxidative DNA damage, as also suggested by earlier findings<br />
(Luhe et al. 2003; Cavin et al. 2006; Marin-Kuan et al. 2006), is supported by the<br />
upregulation <strong>of</strong> the p53 pathway genes SUPT16H in both strains, MDM2 and CHEK2 in<br />
Eker rats, and RBBP6 in wild type rats, as well as by the time-dependent downregulation <strong>of</strong><br />
HSP 40-3, CN1, MSRA and MGST1, responsible for the protection <strong>of</strong> cells against oxidative<br />
stress. Indeed, recent findings demonstrated that overexpression <strong>of</strong> glia maturation growth<br />
factor beta (GMFB) resulted in reduced antioxidant enzyme activities, subsequent<br />
accumulation <strong>of</strong> H2O2, and finally enhanced oxidative injury <strong>of</strong> renal proximal tubular cells<br />
(Kaimori et al. 2003). The 5.9-fold upregulation <strong>of</strong> GMFB in Eker rats, shown here, and the<br />
known reduced 8-oxoguanine-DNA glycosylase expression in Eker rats (Habib et al. 2003),<br />
may suggest that Eker rats are more susceptible to oxidative stress than wild type rats. This<br />
interpretation is supported by the observation that Eker rats responded with increased cell<br />
proliferation already on day 7 (Figure 3.1, C) and increased formation <strong>of</strong> AT (Figure 3.1, E).<br />
However, OTA induced regenerative proliferation may not have been the sole contributor to<br />
the propagation <strong>of</strong> preneoplastic lesions. Indeed, OTA has previously been assumed to<br />
have mitogenic properties (Horvath et al. 2002; Gekle et al. 2005), a hypothesis supported<br />
by downregulation <strong>of</strong> IGFBP-4, a negative regulator <strong>of</strong> the (IGF)-PI3K-AKT mitogenic<br />
pathway, in Eker and wild type rats. Despite this, only Eker rats demonstrated a significant<br />
increase <strong>of</strong> preneoplastic lesions. The latter can be explained with the renal gene<br />
expression pr<strong>of</strong>ile <strong>of</strong> OTA-treated Eker rats which contains deregulated genes characteristic<br />
<strong>of</strong> the most pertinent hallmarks involved in cancer progression (Hanahan and Weinberg<br />
2000), namely: 1) the evasion <strong>of</strong> programmed cell death via upregulation <strong>of</strong> antiapoptotic<br />
genes e.g. PDC 4, PEA15 or MCL1; 2) insensitivity to growth inhibitory signals via<br />
upregulation <strong>of</strong> protooncogenes e.g. MERTK, SMOH, V-KIT, TMP; 3) self sufficiency in<br />
growth signals e.g. via an activated (IGF)-PI3K-AKT pathway (PIK3CB, AKT1S1, AKT2 or<br />
SBF1); 4) limitless replication potential e.g. via an upregulation <strong>of</strong> genes involved in DNA<br />
replication (TOP1), chromatin remodelling (SMARCA4), or mitotic spindle formation<br />
(MAPRE1, AJUBA, NEK9); 5) sustained angiogenesis via upregulation <strong>of</strong> VEGF, VEZF1,<br />
ANGPTL2; and 6) tissue invasion and metastasis via a broad set <strong>of</strong> deregulated genes<br />
involved in cell structure remodelling (e.g. Rho guanine nucleotide exchange factor<br />
ARHGEF11 or IQGAP1, an effector for CDC42 and RAC1) and EMT (TGFßR-II, SMADs,<br />
CD44, TIMP3).<br />
As demonstrated, OTA treatment <strong>of</strong> Eker rats led to the expression <strong>of</strong> genes that could be<br />
intricately involved in, or are the result <strong>of</strong>, activated mTor signalling (Thomas 2006). Thus<br />
OTA treatment and the Tsc2-mutation may have acted in concert or separately on the<br />
(IGF)-PI3K-AKT pathway thus resulting in a co-joint activation <strong>of</strong> mTor signalling. Additional<br />
60
Chapter 3: Manuscript I<br />
stimulation <strong>of</strong> the mTor pathway could be possible by OTA-mediated ERK activation<br />
(Horvath et al. 2002). Tsc2 was demonstrated to be a direct substrate <strong>of</strong> ERK (Ma et al.<br />
2005a). As Eker rats exhibit only one functional allele <strong>of</strong> the Tsc2 gene, OTA mediated<br />
ERK-dependent inactivation <strong>of</strong> Tsc2 could lead to more drastic and therefore earlier<br />
detection <strong>of</strong> proliferative effects and neoplastic transformation in Eker rats (Figure 3.4, B).<br />
In summary, gene expression pr<strong>of</strong>ile comparisons with histopathologic findings from AAand<br />
OTA- treated Eker and wild type rats discussed here, highlight that gene expression<br />
analysis subsequent to short-term in vivo assays may have the potential to identify<br />
deregulated genes involved in compound- and strain- specific pathology. Moreover,<br />
deregulation <strong>of</strong> genes, for which a similar direction <strong>of</strong> deregulation has been reported for<br />
various types <strong>of</strong> cancers, suggests that pathways linked to tumorigenesis may be<br />
deregulated already after short-term carcinogen exposure. Whether these changes in gene<br />
expression are transient or can be causally linked to a compound-specific tumorigenicity<br />
cannot be determined without gene expression pr<strong>of</strong>ile analysis <strong>of</strong> the respective<br />
preneoplastic and neoplastic lesions in rats chronically treated with AA or OTA. This<br />
analysis, as currently carried out in this laboratory using laser-capture microdissection<br />
(Stemmer et al. 2006) will provide further information as to the relevance <strong>of</strong> the pathways<br />
identified in short-term experiments for the understanding <strong>of</strong> the mechanisms underlying<br />
AA- and OTA- induced renal carcinogenesis.<br />
Acknowledgements<br />
We would like to thank the Federal Ministry <strong>of</strong> Education and Research for funding the<br />
project (BMBF: 0313024), Tanja Lampertsdörfer and Gudrun von Scheven for skillful<br />
assistance during the whole animal experiment, Evelyn O’Brien and Alexandra Heussner<br />
for help with the animal sacrifice, and Margot Thiel and Kerstin Lotz for microarray<br />
hybridization.<br />
61
3.6 Supplement<br />
3.6.1 Supplementary Information<br />
Chapter 3: Manuscript I<br />
The significance <strong>of</strong> microarray data for biological interpretations and the necessity <strong>of</strong><br />
confirming them by another method were extensively discussed by Rockett et al. (Rockett<br />
and Hellmann 2004). The authors suggested that microarrays as well as quantitative realtime<br />
polymerase chain reaction (QRT-PCR) are at best semi-quantitative, while quantitative<br />
detection <strong>of</strong> gene expression cannot be achieved with any <strong>of</strong> the currently available<br />
methodologies. Based on 284 individual comparisons, a regression coefficient <strong>of</strong> 0.94<br />
between the natural log ratios for Affymetrix RAE230A arrays and QRT-PCR (Taqman®<br />
technology) data was previously obtained in this laboratory (data not shown). This supports<br />
the assumption that both technologies deliver at least semi-quantitatively comparable<br />
expression ratios. Therefore, it was assumed that the expression pr<strong>of</strong>iles generated with<br />
the Affymetrix platform, as presented here, can be used for semi-quantitative comparisons<br />
<strong>of</strong> gene deregulation and subsequent data interpretation.<br />
62
3.6.2 Supplementary Tables<br />
Supplementary table 3.1: Numbers and categories <strong>of</strong> genes deregulated by AA or OTA<br />
# genes<br />
Ek<br />
specific<br />
# genes<br />
Ek and wt<br />
specific<br />
# genes<br />
wt<br />
specific<br />
# genes<br />
Ek<br />
specific<br />
# genes<br />
Ek and wt<br />
specific<br />
# genes<br />
wt<br />
specific<br />
Compound AA AA AA OTA OTA OTA<br />
# all significantly<br />
deregulated genes<br />
Pathophysiological<br />
categories<br />
Metabolism and<br />
biotransformation<br />
77 34 47 314 61 80<br />
Description<br />
Chapter 3: Manuscript I<br />
10 12 3 5 6 3 Deregulation <strong>of</strong> genes encoding biotransformation enzymes and drug transporters<br />
DNA damage response 1 7 1 2 1 1 Upregulation <strong>of</strong> genes encoding DNA damage repair enzymes and p53 target genes<br />
Oxidative stress response - - - 2 1 - Upregulation <strong>of</strong> ARE A target genes and other genes known to be involved in oxidative stress<br />
responses<br />
Enhanced oxidative stress - - 1 2 3 1 Downregulation <strong>of</strong> genes mediating the cellular antioxidant defense or upregulation <strong>of</strong> genes<br />
encoding Nrf2 B inhibitors<br />
Response to unfolded and<br />
damaged proteins<br />
1 - - 6 - 2 Upregulation <strong>of</strong> genes encoding members <strong>of</strong> the heat shock- or glucose-regulated protein<br />
family and others involved in the response to unfolded or damaged proteins<br />
Cellular stress 1 - - 6 2 1 Upregulation <strong>of</strong> immediate early genes, target genes <strong>of</strong> the stress kinase-, NFkB C -, or HIF D<br />
pathways or other stress related pathways. Downregulation <strong>of</strong> genes encoding inhibitors <strong>of</strong><br />
these pathways<br />
Inhibited cell survival and<br />
proliferation<br />
Enhanced cell survival/<br />
proliferation<br />
Cell cycle progression and<br />
mitosis<br />
8 1 7 6 3 2 Downregulation <strong>of</strong> antiapoptotic genes, genes involved in cell cycle progression or genes<br />
encoding components <strong>of</strong> the mitotic spindle apparatus. Upregulation <strong>of</strong> proapoptotic genes<br />
or tumour suppressor genes<br />
1 - 3 23 2 - Upregulation <strong>of</strong> genes encoding growth and survival factors, (IGF)- PI3K-AKT E pathway<br />
components, antiapoptotic genes, genes <strong>of</strong> the VEGF F pathway and functioning in growth<br />
and survival <strong>of</strong> endothelial cells. Downregulation <strong>of</strong> tumour suppressor genes.<br />
- - - 10 1 - Upregulation <strong>of</strong> genes involved in DNA replication, mitotic spindle/ cytokineses or<br />
nucleosome formation.<br />
63
Cell structure remodelling<br />
Chapter 3: Manuscript I<br />
2 2 2 32 1 2 Up- and downregulation <strong>of</strong> genes coding for components or regulators <strong>of</strong> the cytoskeleton,<br />
cell adhesion molecules and components <strong>of</strong> the ECM. Upregulation <strong>of</strong> gene transcripts<br />
involved in glycoprotein synthesis.<br />
EMT/ Fibrosis - - - 12 2 - Upregulation <strong>of</strong> genes involved in the TGF-ßG pathway, ECMH accumulation or deregulation<br />
<strong>of</strong> other genes known to be involved in the process <strong>of</strong> epithelial mesenchymal transition<br />
(EMT) and /or fibrosis<br />
Inflammation 8 2 2 13 - 2 Upregulation <strong>of</strong> genes encoding components <strong>of</strong> the TNF I / cytokine pathway, <strong>of</strong> the antigene<br />
presentation machinery, or inflammatory factors. Downregulation <strong>of</strong> inhibitors <strong>of</strong> blood<br />
coagulation.<br />
Altered ion homeostasis 6 1 2 24 - 1 Deregulation <strong>of</strong> genes encoding proteins involved in ion storage or transport<br />
Energy mobilisation 1 2 1 17 4 3 Upregulation <strong>of</strong> genes primarily involved in glycolysis, fatty acid oxidation or the pentose<br />
phosphate pathway.<br />
Increased transcription and<br />
RNA processing<br />
6 - 2 15 1 6 Upregulation <strong>of</strong> genes encoding transcription factors, coactivators, or enzymes involved in<br />
the metabolism <strong>of</strong> nucleotides and nucleosides and RNA splicing<br />
Increased protein synthesis 1 - 2 4 1 1 Upregulation <strong>of</strong> genes involved in translation or protein folding in the endoplasmatic<br />
reticulum<br />
Increased protein<br />
trafficking<br />
- - 2 28 - - Upregulation <strong>of</strong> genes involved in protein sorting, vesicular transport or endo-/exocytosis<br />
Dedifferentiation 10 2 7 9 9 4 Downregulation <strong>of</strong> genes mediating the physiological homeostasis <strong>of</strong> the cell and/or organ.<br />
Neuronal differentiation 2 2 - 14 1 4 Upregulation <strong>of</strong> genes involved in neuronal signal transmission or neuronal differentiation<br />
Deregulation <strong>of</strong> the<br />
circadian rhythm<br />
Unknown toxicological<br />
category<br />
4 1 - - - - Deregulation <strong>of</strong> genes involved in the regulation <strong>of</strong> the circadian rythm<br />
9 1 2 20 7 12 Genes with known biochemical function, but whose deregulation could not be given a<br />
specific meaning<br />
unknown ESTs 6 1 10 60 16 35 Genes which are not annotated<br />
A B C D E<br />
: ARE: Arylhydrocarbon responsive elements; : NFkB: Nuclear transcription factor B; : Nrf2: NF-E2-related factor-2; : HIF: Hypoxia inducible factor; : (IGF)- PI3K-<br />
AKT: Insulin growth factor- Phosphoinositol-triphosphate kinase-AKT; F : VEGF: Vascular endothelial growth factor; G : TGF-ß: Transforming growth factor ß; H : ECM:<br />
Extracellular matrix; I : TNF: Tumor necrosis factor<br />
64
Chapter 3: Manuscript I<br />
Supplementary table 3.2: Incidence and mean numbers <strong>of</strong> preneoplastic and neoplastic lesions. Mean + SD <strong>of</strong> atypical tubule (AT), aypical hyperplasia (AH)<br />
and neoplastic (adenoma, carcinoma) lesions <strong>of</strong> AA or OTA treated Eker and wild type rats and their respective controls (n=3). Incidences <strong>of</strong><br />
respective lesions in treated or control animals are given in parentheses. Significant differences (unpaired t-test) between treated and timematched<br />
control groups are indicated (*p
Chapter 3: Manuscript I<br />
Supplementary table 3.3: Non-neoplastic pathology <strong>of</strong> AA or OTA treated Eker rats. Mean + SD <strong>of</strong> non-neoplastic changes. ranked according to the following<br />
severity classes: none (0). mild (1). moderate (2). strong (3). and severe (4). including intermediate classes (e.g. 0.5. 1.5) <strong>of</strong> AA or OTA treated<br />
Eker rats and their respective controls (n=3). Incidences <strong>of</strong> respective lesions in treated or control animals are given in parentheses.<br />
Necrosis<br />
Apoptosis<br />
Fibrosis<br />
Basement<br />
membrane<br />
(thickening)<br />
Mesangia<br />
(thickening)<br />
Bowmans Space<br />
with erythrocytes)<br />
Calcium cast<br />
Glomerulo-<br />
Sclerosis<br />
Glomerulo-<br />
Nephritis<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
Control (AA) AA Control (OTA) OTA<br />
d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.3 ±<br />
0.34<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.5 ±<br />
0.71<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.5 ±<br />
0.41<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
Glomeruli<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.67 ±<br />
0.47<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.67 ±<br />
0.24<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(2/3)<br />
0.83 ±<br />
0.47<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.83 ±<br />
0.24<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.5 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.5 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0,17 ±<br />
0,24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.67 ±<br />
0.47<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
66
Pigment Deposits<br />
Necrosis<br />
Apoptosis<br />
Karyomegally<br />
Vacuolization<br />
Hyaline droplets<br />
Cell shedding<br />
Proteinaceous<br />
casts<br />
Tubular dilatation<br />
Calcium casts<br />
Tubular<br />
regeneration<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
Control (AA) AA Control (OTA) OTA<br />
Chapter 3: Manuscript I<br />
d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14<br />
0.33 ±<br />
0.47<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
0.33 ±<br />
0.47<br />
(2/3)<br />
0.67 ±<br />
0.24<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.0 ±<br />
0.41<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.5 ±<br />
0.41<br />
(2/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
Cortex<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.83 ±<br />
0.24<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.67 ±<br />
0.62<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.0 ±<br />
0,41<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.83 ±<br />
0.47<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.83 ±<br />
0.62<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.83 ±<br />
0.62<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.67 ±<br />
0.47<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
1.5 ±<br />
0.41<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.67 ±<br />
0.47<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.83 ±<br />
0.24<br />
(3/3)<br />
1.0 ±<br />
0.41<br />
(3/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.33 ±<br />
0.62<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.17 ±<br />
0.24<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.50 ±<br />
0.41<br />
(3/3)<br />
67
CPN<br />
Inflammation<br />
Hypertrophy<br />
Pigment Deposits<br />
Necrosis<br />
Apoptosis<br />
Karyomegally<br />
Vacuolization<br />
Hyaline droplets<br />
Cell shedding<br />
Proteinaceous<br />
casts<br />
Tubular dilatation<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.83 ±<br />
0.24<br />
(3/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
1.5 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
Medulla/ Papilla<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.50 ±<br />
0.0<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
Chapter 3: Manuscript I<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.50 ±<br />
0.0<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.5 ±<br />
0.41<br />
(2/3)<br />
0.67 ±<br />
0.24<br />
(3/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
0.50 ±<br />
0.0<br />
(3/3)<br />
1.0 ±<br />
0.0<br />
(3/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
68
Calcium casts<br />
Tubular<br />
regeneration<br />
CPN<br />
Inflammation<br />
Hypertrophy<br />
Chapter 3: Manuscript I<br />
(0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3) (0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.24<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.50 ±<br />
0.41<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.34<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.33 ±<br />
0.47<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
1.33 ±<br />
0.94<br />
(2/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.17 ±<br />
0.24<br />
(1/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
0.0 ±<br />
0.0<br />
(0/3)<br />
69
Chapter 3: Manuscript I<br />
Supplementary table 3.4: Non-neoplastic pathology <strong>of</strong> AA or OTA treated wild type rats. Mean + SD <strong>of</strong> non-neoplastic changes, ranked according to the<br />
following severity classes: none (0), mild (1), moderate (2), strong (3), and severe (4), including intermediate classes (e.g. 0.5, 1.5) <strong>of</strong> AA or OTA<br />
treated wild type rats and their respective control (n=3). Incidences <strong>of</strong> respective lesions in treated or control animals are given in parentheses.<br />
Pigment Deposits<br />
Necrosis<br />
Apoptosis<br />
Fibrosis<br />
Basement membrane<br />
(thickening)<br />
Mesangia<br />
Bowmans Space with<br />
erythrocytes)<br />
Calcium cast<br />
Glomerulo- Sclerosis<br />
Glomerulo- Nephritis<br />
Control (AA, OTA) AA OTA<br />
d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.71 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.83 ±<br />
0.62 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.71 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.5 ± 0.71<br />
(1/3)<br />
0.67 ±<br />
0.62 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
Glomeruli<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.71 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
1.0 ± 0.71<br />
(2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
1.0 ± 0.41<br />
(3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.67 ±<br />
0.24 (3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
70
Pigment Deposits<br />
Necrosis<br />
Apoptosis<br />
Karyomegally<br />
Hyaline droplets<br />
Cell shedding<br />
Proteinaceous<br />
casts<br />
Tubular dilatation<br />
Calcium casts<br />
Tubular<br />
regeneration<br />
CPN<br />
Inflammation<br />
Hypertrophy<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
1.0 ± 0.0<br />
(3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.83 ±<br />
0.62 (2/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.67 ±<br />
0.24 (3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.33 ±<br />
0.23 (2/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
Cortex<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ± 0.0<br />
(3/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.67 ±<br />
0.24 (3/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.83 ±<br />
0.24 (3/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.83 ±<br />
0.62 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.83 ±<br />
0.24 (3/3)<br />
0.67 ±<br />
0.47 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
1.33 ±<br />
0.24 (3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
1.17 ±<br />
0.62 (3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.67 ±<br />
0.47 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.83 ±<br />
0.85 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
Chapter 3: Manuscript I<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ± 0.0<br />
(3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.67 ±<br />
0.24 (3/3)<br />
0.33 ±<br />
0.24 (2/3)<br />
0.67 ±<br />
0.47 (2/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.67 ±<br />
0.24 (3/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
1.0 ± 0.0<br />
(3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
0.83 ±<br />
0.24 (3/3)<br />
0.50 ± 0.0<br />
(3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.67 ±<br />
0.47 (2/3)<br />
0.17 ±<br />
0.24 (1/3)<br />
1.0 ± 0.0<br />
(3/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.83 ±<br />
0.85 (2/3)<br />
0.50 ±<br />
0.41 (2/3)<br />
0.83 ±<br />
0.24 (3/3)<br />
0.33 ±<br />
0.47 (1/3)<br />
71
Pigment Deposits<br />
Necrosis<br />
Apoptosis<br />
Karyomegally<br />
Vacuolization<br />
Hyaline droplets<br />
Cell shedding<br />
Proteinaceous<br />
casts<br />
Tubular dilatation<br />
Calcium casts<br />
Tubular<br />
regeneration<br />
CPN<br />
Inflammation<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ± 0.47<br />
(1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ± 0.24<br />
(1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.17 ± 0.24<br />
(1/3)<br />
0.17 ± 0.24<br />
(1/3)<br />
Medulla/ Papilla<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ± 0.47<br />
(1/3)<br />
0.33 ± 0.47<br />
(1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
Chapter 3: Manuscript I<br />
0.0 ± 0.0<br />
(0/3)<br />
0.33 ± 0.47<br />
(1/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
0.0 ± 0.0<br />
(0/3)<br />
72
Chapter 4: Manuscript II<br />
Renal Carcinogenic Response <strong>of</strong> Eker Rats to Low Doses <strong>of</strong><br />
Methylazoxymethanol Acetate<br />
Kerstin Stemmer 1 , Heidrun Ellinger-Ziegelbauer 2 , Katja Strauch 3 , Leonard Collins 3 , James A.<br />
Swenberg 3 , Hans-J. Ahr 2 and Daniel R. Dietrich 1*<br />
1<br />
Human and Environmental Toxicology, University <strong>of</strong> Konstanz, Konstanz, Germany<br />
2<br />
<strong>Molecular</strong> and Special Toxicology, Bayer Healthcare AG, Wuppertal, Germany<br />
3<br />
Environmental Sciences and Engineering, University <strong>of</strong> North Carolina at Chapel Hill, Chapel<br />
Hill, North Carolina, USA<br />
* Corresponding author: Daniel R. Dietrich: daniel.dietrich@uni-konstanz.de<br />
Submitted to Mutation Research<br />
4.1 Abstract<br />
Methylazoxymethanol-actetae (MAMAc), the stabilized aglycone <strong>of</strong> cycasin, is a potent<br />
renal carcinogen in rodents, which acts via direct alkylation <strong>of</strong> the DNA. Current knowledge<br />
on MAMAc carcinogenicity is based on the use <strong>of</strong> very high doses, which may reflect a<br />
carcinogenic mechanism different from the one active at low doses. This study compared<br />
the renal response <strong>of</strong> male and female Tsc2-mutant Eker rats exposed to low doses <strong>of</strong><br />
MAMAc (250µg/kg bw day -1 ) after 1, 3, 7 and 14 days <strong>of</strong> exposure using gene expression<br />
pr<strong>of</strong>iling, histopathology, including the assessment <strong>of</strong> preneoplastic and neoplastic lesions,<br />
cell proliferation analysis and LC-MS/MS for demonstration <strong>of</strong> O6-methylguanine (O6-meG)<br />
DNA adducts. Following 3 and 6 months <strong>of</strong> exposure renal histopathology, including<br />
preneoplastic and neoplastic lesions and cell proliferation was assessed.<br />
Gene expression pr<strong>of</strong>iles revealed only a small number <strong>of</strong> deregulated genes in short-term<br />
MAMAc treated males, which mainly revealed a reduced protein synthesis, including downregulation<br />
<strong>of</strong> tumor suppressor genes. An exposure dependent increase in O6-meG<br />
adducts was indicated following 1-14 days <strong>of</strong> exposure, whereas the number <strong>of</strong> O6-meG<br />
adducts detected were generally higher in female than in male rats. 250µg/kg bw day -1 <strong>of</strong><br />
73
Chapter 4: Manuscript II<br />
MAMAc for 3 and 6 months induced increased cell proliferation and number <strong>of</strong><br />
preneoplastic lesions only in female but not in male rats. However, surprisingly male Eker<br />
rats presented with an increased incidence and total higher number <strong>of</strong> adenomas and<br />
carcinomas compared to female rats after 6 months <strong>of</strong> exposure.<br />
In conclusion, above findings suggests that subchronic and –chronic exposure to low doses<br />
<strong>of</strong> MAMAc results in gender-specific differences in response. Indeed, in female rats the<br />
increase in early tumor stages and absence <strong>of</strong> late stages points to a tumor initiating role <strong>of</strong><br />
MAMAc, while in male rats, the increase in later tumor stages after 6 months exposure is<br />
suggestive <strong>of</strong> a tumor promoting role <strong>of</strong> MAMAc.<br />
4.2 Introduction<br />
Cycasin (methylazoymethanol-ß-D-glucoside) is a natural constituent <strong>of</strong> nuts, root and<br />
leaves <strong>of</strong> the cycad plants Cycas circinalis and Cycas revoluta that have been sources <strong>of</strong><br />
food and medicine for the indigenous people <strong>of</strong> the Western Pacific, e.g. the island <strong>of</strong><br />
Guam (Whiting 1963). Cycasin contamination mainly depend on food procession, for<br />
instance flour from washed cycad nuts from Guam only contain marginal concentrations,<br />
while up to 2.3% cycasin was detected when unwashed nuts were used (Hirono 1981). In<br />
rats (Laqueur 1964; Laqueur and Matsumoto 1966), mice (O'Gara et al. 1964), guinea pigs<br />
(Spatz 1964), rabbits (Hirono 1972) and nonhuman primates (Sieber et al. 1980), cycasin<br />
was found to be a potent carcinogen resulting primarily in epithelial and mesenchymal<br />
tumors <strong>of</strong> the kidneys, tumors <strong>of</strong> the liver and colon.<br />
The carcinogenic action <strong>of</strong> cycasin appears mainly dependent on its deglucosylation by ßglucosidases<br />
<strong>of</strong> the gut micr<strong>of</strong>lora to release its principal metabolite methylazoxymethanol<br />
(MAM) (Spatz et al. 1967; Matsushima et al. 1979; Choi et al. 1996). The release <strong>of</strong> the<br />
“ultimate carcinogen” methyldiazonium ion from MAM (Figure 4.1) is either catalyzed by<br />
different enzymes including alcohol dehydrogenase (Grab and Zedeck 1977; Notman et al.<br />
1982), choline dehydrogenase (Tan et al. 2004) and cytochrome P450-2E1 (Sohn et al.<br />
2001), however, may also occur spontaneously under normal physiological conditions<br />
(Nagasawa et al. 1972; Sohn et al. 2001). The methyldiazonium ion, is also metabolically<br />
formed from other DNA alkylating pro-carcinogens, e.g. dimethylnitrosamine (DMN) (Shank<br />
and Magee 1967; Weisburger 1978) and was demonstrated to be responsible for the<br />
formation <strong>of</strong> pro-mutagenic and pro-carcinogenic N7-methylguanine- and O6-<br />
74
Chapter 4: Manuscript II<br />
methylguanine- DNA- adducts observed in different target organs <strong>of</strong> rats, treated with DNA<br />
alkylating pro-carcinogens (Matsumoto and Higa 1966; Sohn et al. 2001).<br />
C<br />
H 3<br />
N +<br />
O -<br />
I<br />
N<br />
CH2 O<br />
O<br />
OH<br />
CH 2 OH<br />
OH OH<br />
β-glucosidase<br />
Glycoside<br />
C<br />
H 3<br />
N + N<br />
IV<br />
O -<br />
H<br />
C O<br />
enzymatic<br />
C<br />
H 3<br />
N + N<br />
DNA<br />
O -<br />
H<br />
CH2 O<br />
spontaneous<br />
II<br />
C<br />
H 3<br />
nucleophilic<br />
attack<br />
C<br />
H 2<br />
V<br />
N +<br />
III<br />
O<br />
+<br />
H3C +<br />
VI<br />
N<br />
+ OH -<br />
Figure 4.1: Metabolic pathways <strong>of</strong> cycasin and MAM activation, from (Laqueur 1968; Nagasawa et<br />
al. 1972; Sohn et al. 2001). I: cycasin; II: MAM; III: formaldehyde; IV:<br />
methylazoxyformaldehyde; V: methyldiazonium ion, VI: methylcarbonium ion.<br />
However, since earlier cycasin and MAM carcinogenicity studies used high oral doses (4-<br />
500mg kg-1 bw), either as single or as multiple applications over a short exposure period<br />
(Hirono et al. 1968; Laqueur 1968; Fukunishi et al. 1971), it remains unclear whether the<br />
exposure to acute, subchronic or chronic low doses <strong>of</strong> MAM is sufficient to induce initiate<br />
and/or promote carcinogenic lesions or other adverse effects in the rat kidney.<br />
To elucidate this question, a sensitive rat model for the detection <strong>of</strong> renal carcinogens was<br />
employed: Eker rats carrying an inherited heterozygous mutation in the tuberous sclerosis<br />
(TSC2) tumor suppressor gene (Yeung et al. 1994; Hino et al. 1995; Kobayashi et al. 1997)<br />
are predisposed to develop renal cancer early <strong>of</strong> age and were previously characterized to<br />
be highly sensitive towards renal carcinogens (McDorman and Wolf 2002; Hino 2004). For<br />
instance, when treated with a single dose <strong>of</strong> the DNA alkylating DMN (30mg/kg bw ), 8<br />
months later, Eker rats demonstrated a 70-fold higher average number <strong>of</strong> tumors per<br />
animal when compared to the corresponding wild type rats exposed under identical<br />
conditions (Walker et al. 1992 8).<br />
Based on above findings it was expected that Eker rats, continuously exposed to low doses<br />
<strong>of</strong> MAM-acetate (MAMAc, 250µg/kg bw), should present with a detectable increase <strong>of</strong> preneoplastic<br />
and neoplastic lesions already following 6 months <strong>of</strong> exposure. Gene-expression<br />
pr<strong>of</strong>iling was previously demonstrated to be a sensitive and reliable tool that allows<br />
N 2<br />
75
Chapter 4: Manuscript II<br />
detection <strong>of</strong> the activation <strong>of</strong> a specific mode <strong>of</strong> action <strong>of</strong> renal or liver carcinogens (toxicity<br />
and carcinogenicity) already after 1 to 14 days <strong>of</strong> carcinogen exposure in Eker (Stemmer et<br />
al. 2007) or Wistar rats (Ellinger-Ziegelbauer et al. 2004; Ellinger-Ziegelbauer et al. 2005),<br />
respectively. Indeed, an up-regulation <strong>of</strong> genes involved in the DNA damage response was<br />
reported in livers <strong>of</strong> DMN-treated Wistar rats after one single dose <strong>of</strong> 4mg/kg bw (Ellinger-<br />
Ziegelbauer et al. 2004). Thus, gene expression pr<strong>of</strong>iling, subsequent to 1-14 days low<br />
dose MAMAc treatment <strong>of</strong> Eker rats, should allow the detection <strong>of</strong> a possible DNA damage<br />
response as well as other specific responses associated with the toxicity and/or the<br />
carcinogenicity <strong>of</strong> MAMAc. The latter in combination with renal pathology and cell<br />
proliferation analysis would therefore allow the determination <strong>of</strong> the acute toxic effects and<br />
earliest events <strong>of</strong> MAMAc induced renal tumorigenesis, whereas renal histopathology,<br />
including the assessment <strong>of</strong> preneoplastic and neoplastic lesions, and cell proliferation<br />
analysis following 3 and 6 months <strong>of</strong> exposure would allow to determine whether<br />
subchronic and chronic low dose MAMAc treatment <strong>of</strong> Eker rats manifests in the onset <strong>of</strong><br />
preneoplastic and neoplasic renal lesions. Possible sex-specific effects as observed for<br />
other renal carcinogens (Boorman et al. 1992) were addressed via employing male and<br />
female Eker rats.<br />
4.3 Material and Methods<br />
4.3.1 Compound<br />
Highly purified methylazoxymethanol acetate (MAMAc) was purchased from the Midwest<br />
Research Institute, NCI Chemical Resource Repository (64 FR 72090, 64 FR 28205). Prior<br />
to the in vivo experiments, the stability <strong>of</strong> MAMAc in 0.1M sodium bicarbonate (NaHCO3)<br />
was confirmed by HPLC using a method published previously (Fiala et al. 1976). Stabilized<br />
MAMAc is highly sensitive to hydrolysis by various blood and organ esterases, and<br />
therefore is deacetylated to MAM almost immediately following administration (Zedeck and<br />
Brown 1977)<br />
4.3.2 Animals<br />
Six to ten week old, genotyped heterozygous Tsc2 mutant Eker rats (Tsc2+/-, Long Evans)<br />
were purchased from the MD Anderson <strong>Cancer</strong> Center, Smithville, Texas, USA and housed<br />
at the University <strong>of</strong> Konstanz animal research facility under standard conditions. Prior to<br />
76
Chapter 4: Manuscript II<br />
exposure, male and female Eker rats were randomly assigned to dose groups and allowed<br />
to acclimatize to laboratory conditions for 4 weeks.<br />
4.3.3 Short-term experiments<br />
Groups <strong>of</strong> three male and female Eker rats were gavaged daily with MAMAc (250µg/ kg<br />
BW) dissolved in sterile 0.1M NaHCO3 for 1, 3, 7, and 14 days, respectively (Table 4.1).<br />
Time-matched vehicle controls (n=3) were gavaged with corresponding volumes <strong>of</strong> 0,1M<br />
NaHCO3.<br />
4.3.4 Subchronic and chronic experiments<br />
Groups <strong>of</strong> 10 Male and female Eker rats were gavaged with MAMAc (250µg/ kg BW),<br />
dissolved in 0.1M NaHCO3 at five days a week for 3 and 6 months. Time-matched vehicle<br />
controls were gavaged with 0.1M NaHCO3. For post-mortem immunohistochemical<br />
analysis <strong>of</strong> cell proliferation, 5 <strong>of</strong> 10 rats per dose group and sex were s.c. implanted with<br />
osmotic ALZET-pumps (Model 2ML1, Charles River Laboratories, Germany) containing 5bromo-2-deoxyuridine<br />
(BrdU, 20mg/ml sterile saline, Sigma Aldrich, Germany) 5 days prior<br />
to sacrifice.<br />
4.3.5 Sample collection<br />
At the end <strong>of</strong> each treatment period, anesthetized rats were sacrificed by exsanguination<br />
subsequent to retrograde perfusion with PBS and kidneys were collected, longitudinally<br />
sectioned into 5mm slices and stored in RNAlater (Qiagen, Germany) or in PBS buffered<br />
histology fixative buffer, containing 2% paraformaldehyde and 1% glutaraldehyde for<br />
subsequent paraffin embedding and sectioning.<br />
4.3.6 Histopathology<br />
Haematoxilin and Eosin (H&E) stained renal paraffin section were randomized and<br />
histopathological analysis was carried out by light microscopy at 40-400-fold magnification.<br />
Non-neoplastic changes were classified as none (0), mild (1), moderate (2), strong (3), and<br />
severe (4), including intermediate classes (e.g. 0.5, 1.5 etc.), while preneoplastic and<br />
77
Chapter 4: Manuscript II<br />
neoplastic lesions were classified as reported previously (Dietrich and Swenberg 1991).<br />
Absolute numbers and incidences <strong>of</strong> preneoplastic and neoplastic lesions were determined.<br />
4.3.7 Cell proliferation analysis<br />
Immunostaining: Subsequent to short term experiments (1-14 days), cell proliferation was<br />
evaluated by immunohistochemical staining for proliferating cell nuclear antigen (PCNA)<br />
using monoclonal primary anti-PCNA antibody (PC-10; DAKO, Germany) in paraffinembedded<br />
kidney sections, as described previously (Stemmer et al. 2007).<br />
Five days prior to the sub-chronic (3 months) and chronic (6 months) experiments 5 <strong>of</strong> 10<br />
rats per dose group and sex were s.c. implanted with osmotic ALZET-pumps (Model 2ML1,<br />
Charles River Laboratories, Germany) containing 5-bromo-2-deoxyuridine (BrdU, 20mg/ml<br />
sterile saline, Sigma Aldrich, Germany) to allow post-mortem immunohistochemical<br />
analysis <strong>of</strong> cell proliferation. BrdU-immunostaining was performed as described previously<br />
(Stemmer et al. 2007), using a monoclonal mouse anti-BrdU primary antibody (MU247-UC,<br />
Biogenex, USA) diluted 1:100 in Power BlockTM (BioGenex, USA) in an overnight<br />
application at 4°C.<br />
Quantification <strong>of</strong> S-phase nuclei: Subsequent to short-term, sub-chronic and chronic<br />
experiments, PCNA- or BrdU- positive stained S-phase nuclei were quantified on<br />
randomized sections. 20 microscopic fields (10x ocular, 40x objective) were randomly<br />
chosen within the area <strong>of</strong> the renal outer cortex and inner cortex/outer medulla. Proximal<br />
and distal tubules and collecting ducts were counted separately per field. A minimum <strong>of</strong><br />
1.000 and 500 nuclei were counted in proximal and distal tubules and collecting ducts,<br />
respectively, distinguishing between negative and positive BrdU or PCNA-stained nuclei.<br />
Nuclear labeling indices (LI%) were calculated as positive nuclei/total number <strong>of</strong> nuclei<br />
counted.<br />
4.3.8 RNA isolation microarray hybridization and gene expression<br />
pr<strong>of</strong>iling<br />
RNA isolation from RNAlater fixed kidneys and Affymetrix Rat Genome RAE230A array<br />
hybridization was performed as described previously (Stemmer et al. 2007). Microarray<br />
quality control was performed as published by Elling-Ziegelbauer (Ellinger-Ziegelbauer et<br />
al. 2004). Statistical analysis was performed using Expressionist Analyst s<strong>of</strong>tware<br />
(Genedata AG, Switzerland). To select significantly deregulated genes due to treatment or<br />
78
Chapter 4: Manuscript II<br />
time (single effects) and to treatment and time (interaction effect), 2-way ANOVA with a pvalue<br />
cut-<strong>of</strong>f <strong>of</strong> 0.005, combined with a 1.7-fold deregulation threshold for at least one timepoint<br />
was applied. Significantly deregulated genes were subsequently divided into gene<br />
groups with distinct expression pr<strong>of</strong>iles over the time-course using self-organizing map<br />
(SOM) analysis. SOM analysis also allowed the identification <strong>of</strong> genes showing inconsistent<br />
expression between the controls at different time points. Such genes were manually<br />
discarded from the subsequent functional analysis. Using this adjusted dataset, gene<br />
expression ratios <strong>of</strong> individual genes were calculated by dividing the respective expression<br />
values <strong>of</strong> single treated replicate samples by the mean expression value <strong>of</strong> all<br />
corresponding time-matched control samples. Heat-maps were used to graphically display<br />
the relative expression data, after one-dimensional clustering <strong>of</strong> the genes.<br />
Significantly deregulated genes were characterized according to the physiological role <strong>of</strong> its<br />
encoded protein, using information from databases, e.g. NetAffx (update: August 2006),<br />
Swissprot, Proteome and Pubmed.<br />
4.3.9 Quantification <strong>of</strong> O6MG<br />
DNA was isolated and extracted from pooled kidney samples <strong>of</strong> three replicate animals<br />
using an ABI 340A nucleic acid extractor with two phenol/chlor<strong>of</strong>orm and one chlor<strong>of</strong>orm<br />
extraction, followed by precipitation with sodium acetate and isopropanol. DNA adduct<br />
formation in kidneys <strong>of</strong> MAMAc-treated rats was measured by liquid chromatography<br />
coupled with mass spectrometry. The actual analyses were conducted under a pay-for<br />
service contract in the analytical core facility <strong>of</strong> the Environmental Sciences and<br />
Engineering, University <strong>of</strong> North Carolina at Chapel Hill, USA, using undisclosed<br />
methodology.<br />
4.3.10 Statistical analysis<br />
Statistical analysis <strong>of</strong> histopathological and cell proliferation data were carried out using<br />
GraphPad Prism 4® S<strong>of</strong>tware. Significant differences in nuclear labeling indices (LI %) or<br />
total number <strong>of</strong> lesions in treated and control rats were analyzed by Bonferroni’s test for<br />
multiple comparisons. Lesion incidences were tested for significance using the 2-sided<br />
Fisher’s exact test. Ranked non-neoplastic pathology data were analyzed using the<br />
nonparametric Mann-Whitney test.<br />
79
4.4 Results<br />
4.4.1 Gene expression pr<strong>of</strong>iling<br />
Chapter 4: Manuscript II<br />
Oral treatment <strong>of</strong> male Eker rats with MAMAc for 1, 3, 7 and 14 days resulted in only a<br />
small number <strong>of</strong> significantly deregulated genes in the renal cortex. From overall 59<br />
significantly deregulated annotated genes 15 were down-regulated. Many <strong>of</strong> the up- and<br />
down-regulated genes demonstrated a transient deregulation, with highest gene expression<br />
values after one day <strong>of</strong> exposure (Figure 4.2).<br />
Figure 4.2: Heat map <strong>of</strong> compound-specific unions <strong>of</strong> genes found significantly deregulated (red:<br />
up-regulated; green: down-regulated) from the corresponding control in MAMAc<br />
induced renal gene expression pr<strong>of</strong>iles <strong>of</strong> Eker after 1, 3, 7 or 14 days <strong>of</strong> treatment<br />
(n=3 per time-point). Color scale (left side): gene expression ratio.<br />
Interpretation <strong>of</strong> significantly deregulated genes according to their physiological role and<br />
their direction <strong>of</strong> deregulation allowed categorization <strong>of</strong> genes into different functional<br />
classes. MAMAc treatment resulted in a deregulation <strong>of</strong> two prominent gene clusters<br />
pointing to a reduced transcription and protein synthesis and an up-regulation <strong>of</strong> the TGFß<br />
pathway. An overview <strong>of</strong> these categories is given in table 4.1, while specific responses are<br />
detailed below. Other clusters included genes involved in ion homeostasis, intracellular<br />
transport and regulation <strong>of</strong> the cytoskeleton. Importantly, MAMAc treatment did not result in<br />
genes encoding proteins involved in DNA damage response (Table 4.1 and below).<br />
80
Chapter 4: Manuscript II<br />
Table 4.1: Categories <strong>of</strong> genes differentially deregulated by MAMAc in Eker- and wild type rats. For the major genes cluster are listed together with their<br />
Genebank accession number and their main biochemical function. For Eker- and wild type rats, the fold deregulation ratios <strong>of</strong> genes that were<br />
significantly deregulated at least at on time point according to N-way ANOVA are indicated bold in the last four columns.<br />
Cluster Accession<br />
number<br />
Reduced<br />
transcription<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point vs. mean controls)<br />
AI235236 H2AFO H2A histone family member O Regulation <strong>of</strong> gene expression /<br />
chromatin assembly<br />
BF398015 CBX5 chromobox homolog 5 (Drosophila HP1a) Regulation <strong>of</strong> gene expression/<br />
chromatin assembly<br />
BI300772 RING1 Ring finger protein 1 Regulation <strong>of</strong> gene expression/<br />
chromatin assembly<br />
M25804 NR1D1 Nuclear receptor subfamily 1D1 (Rev-ErbAalpha)<br />
BF284190 NR1D2 Nuclear receptor subfamily 1D2 (Rev-ErbAbeta)<br />
BF420722 NFIX Nuclear factor I-X (CCAAT-binding<br />
transcription factor)<br />
Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
AI008883 FKHL18 Forkhead-like 18 (Drosophila) [predicted] Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
NM_012760 PLAGL1 Pleiomorphic adenoma gene-like 1 Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
1387874_at DBP D-site albumin promoter binding protein 1 Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
NM_133303 BHLHB3 Basic helix-loop-helix protein B3 Regulation <strong>of</strong> gene expression<br />
(Transcription factor)<br />
BG371810 BAT1A ATP-depende nt RNA helicase P4<br />
(Spliceosome RNA helicase Bat1)<br />
NM_053849 PDIA4 Protein disulfide isomerase A4 (CABP2<br />
Calcium-binding protein 2)<br />
Ek d1 Ek d3 Ek d7 Ek d14<br />
0.49 + 0.02 0.64 + 0.06 0.94 + 0.02 0.79 + 0.14<br />
0.79 + 0.20 0.72 + 0.10 0.65 + 0.04 0. 56 + 0.21<br />
0.52 + 0.17 0.83 + 0.13 0.94 + 0.07 0.69 + 0.12<br />
0.57 + 0.10 0.74 + 0.13 0.51 + 0.04 0.33 + 0.08<br />
0.66 + 0.11 0.74 + 0.02 0.45 + 0.02 0.32 + 0.07<br />
0.22 + 0.00 0.28 + 0.01 0.80 + 0.21 0.85 + 0.10<br />
0.94 + 0.07 0.50 + 0.10 1.32 + 0.25 0.73 + 0.18<br />
0.58 + 0.07 0.80 + 0.09 0.86 + 0.05 0.90 + 0.15<br />
0.53 + 0.16 0.94 + 0.09 0.81 + 0.12 1.06 + 0.03<br />
0.56 + 0.15 1.06 + 0.18 0.97 + 0.10 0.99 + 0.07<br />
RNA metabolism (Splicing) 0.88 + 0.03 0.74 + 0.03 0.48 + 0.00 1.10 + 0.07<br />
Protein metabolism (Folding) 0.67 + 0.07 0.47 + 0.05 0.61 + 0.07 0.68 + 0.05<br />
81
TGFß pathway<br />
activation<br />
DNA-damage<br />
response<br />
Accession<br />
number<br />
Chapter 4: Manuscript II<br />
Gene title Biochemical category Fold deregulation (mean <strong>of</strong> three replicate animals<br />
per time-point vs. mean controls)<br />
NM_012775 TGFBR1 TGF-beta receptor type I (ALK5) Signaling cascades (TGF-ß<br />
family pathway)<br />
L09653 TGFBR2 TGF-beta receptor type II Signaling cascades (TGF-ß<br />
family pathway)<br />
NM_013095 SMAD3 (MADH3 Mothers against decapentaplegic<br />
homolog 3)<br />
NM_020085 PTPRK Protein tyrosine phosphatase, (receptortype,<br />
kappa), extracellular region<br />
Signaling cascades (TGF-ß<br />
family pathway)<br />
Signaling cascades (TGF-ß<br />
family pathway)<br />
Ek d1 Ek d3 Ek d7 Ek d14<br />
1.97 + 0.13 1.66 + 0.19 1.32 + 0.18 1.15 + 0.19<br />
5.07 + 0.79 1.33 + 0.27 3.75 + 1.45 2.08 + 0.73<br />
2.06 + 0.21 1.44 + 0.15 1.06 + 0.11 1.17 + 0.19<br />
2.79 + 0.79 2.78 + 0.30 1.73 + 0.73 1.11 + 0.30<br />
NM_012861 MGMT O-6-methylguanine-DNA methyltransferase DNA damage repair 0.86 + 0.79 0.97 + 0.03 1.03 + 0.17 0.91 + 0.06<br />
NM_012601 MPG N-methylpurine-DNA glycosylase DNA damage repair (BER) 1.15 + 0.11 1.03 + 007 1.14 + 0.04 1.18 + 0.15<br />
AF311054<br />
NM_017141<br />
NM_021662<br />
APEX1 apurinic/apyrimidinic endonuclease 1 DNA damage repair (BER) 1.04 + 0.20 0.79 + 0.20 0.79 + 0.34 0.99 + 0.11<br />
POLB Polymerase (DNA directed). beta DNA damage repair (BER) 1.02 + 0.06 0.96 + 0.09 0.84 + 0.07 0.83 + 0.08<br />
POLD polymerase (DNA directed). delta DNA damage repair (BER) 0.92 + 0.12 1.06 + 0.34 1.11 + 0.22 1.11 + 0.35<br />
82
Chapter 4: Manuscript II<br />
DNA damage response: Due to the known genotoxic action <strong>of</strong> MAMAc, special focus was<br />
set on genes coding for proteins that could be indicative for a MAMAc induced DNA<br />
damage. However, genes involved in the expected O6meG adduct repair, including O6methylguanine-DNA<br />
methyltransferase (MGMT) and enzymes <strong>of</strong> the base excision repair<br />
pathway involving N-methylpurine-DNA glycosylase (MPG), apurinic endonuclease<br />
(APEX), DNA polymerase ß and δ (Sedgwick et al. 2007) were not up-regulated in MAMAc<br />
treated male rats (Table 4.1). Deregulating <strong>of</strong> other genes involved in DNA damage repair,<br />
cell cycle arrest or apoptosis could also not be detected (data not shown).<br />
Reduced mRNA synthesis: The most prominent deregulated gene cluster detectable after<br />
MAMAc treatment includes 11 <strong>of</strong> the overall 15 down-regulated genes point to a decreased<br />
overall mRNA synthesis, including down-regulated genes involved in chromatin remodeling<br />
(H2A, CBX5, RING1), basic transcription factors (e.g. PLAGL1, NR1D1, NR1D2, NFIX,<br />
FKHL18, DBP, BHLHB3), RNA splicing and translation (BAT1A).<br />
Signaling cascades: Up-regulated genes and gene-clusters following MAMAc treatment<br />
included 4 genes involved in TGFß signaling (TGFBR1, TGFBR2, SMAD3 and PTPRK),<br />
see table 4.1.<br />
4.4.2 Non-neoplastic renal pathology<br />
Up to 6 months treatment with MAMAc did not result in overt renal non-neoplastic<br />
pathology in male or female Eker rats when compared to the respective controls<br />
(Supplementary tables 4.1 and 4.2). Both MAMAc treated and control rats <strong>of</strong> either sex<br />
presented with minimal renal pathology i.e. cell shedding, tubular dilatation, protein casts,<br />
tubular regeneration and chronic progressive nephropathy (Table 4.2). Notably, epithelial<br />
vacuolar degeneration in the corticomedullary was observed exclusively in MAMAc treated<br />
and control female rats, while this was absent in the corresponding male rats.<br />
83
Chapter 4: Manuscript II<br />
Table 4.2: Non-neoplastic renal pathology <strong>of</strong> male and female Eker rats treated with MAMAc for<br />
three and six months respectively. Histopathological changes were ranked from none<br />
(0) to severe (4) including intermediate classes. Values are presented as median ±<br />
median absolute deviation.<br />
3 months 6 months<br />
males females males females<br />
Control MAMAc Control MAMAc Control MAMAc Control MAMAc<br />
Necrosis 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.3 0.5 + 0.3<br />
Apoptosis 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.1 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0<br />
Karyomegaly 0.0 + 0.0 0.0 + 0.0 0.0 + 0.1 0.0 + 0.1 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.1<br />
Vacuolization 0.0 + 0.0 0.0 + 0.1 0.0 + 0.2 0.0 + 0.0 0.0 + 0.0 0.0 + 0.2 2.0 + 0.4 1.5 + 0.6<br />
Cell shedding 1.0 + 0.6 0.8 + 0.3 1.0 + 0.3 0.5 + 0.3 1.0 + 0.6 0.5 + 0.3 1.0 + 0.4 1.0 + 0.3<br />
Protein casts 0.3 + 0.6 0.3 + 0.3 0.5 + 0.5 0.5 + 0.4 1.0 + 0.5 1.3 + 0.6 0.0 + 0.7 1.0 + 0.7<br />
Tubular<br />
dilatation<br />
0.5 + 0.5 0.5 + 0.2 1.0 + 0.6 0.8 + 0.4 0.5 + 1.0 1.0 + 0.3 0.5 + 0.5 0.5 + 0.3<br />
Ca 2+ Casts 0.0 + 0.4 0.0 + 0.1 0.0 + 0.2 0.0 + 0.4 0.0 + 0.0 0.0 + 0.2 0.0 + 0.0 0.0 + 0.2<br />
Regeneration 0.5 + 0.2 0.5 + 0.3 0.3 + 0.4 0.5 + 0.2 0.0 + 0.5 1.0 + 0.3 0.5 + 0.5 0.5 + 0.4<br />
CPN 0.5 + 0.2 0.5 + 0.2 0.5 + 0.2 0.5 + 0.4 0.8 + 0.5 1.0 + 0.3 1.0 + 0.3 0.8 + 0.4<br />
Inflammation 0.3 + 0.3 0.0 + 0.1 0.5 + 0.2 0.0 + 0.1 0.5 + 0.4 0.3 + 0.3 0.5 + 0.3 0.0 + 0.3<br />
Pigment<br />
deposit<br />
0.1 + 0.2 0.0 + 0.1 0.1 + 0.2 0.0 + 0.1 0.0 + 0.1 0.0 + 0.2 0.0 + 0.3 0.0 + 0.1<br />
4.4.3 (Pre)-neoplastic renal pathology<br />
Histopathological analysis demonstrated the occurrence <strong>of</strong> preneoplastic and neoplastic<br />
lesions <strong>of</strong> different progressional stages, i.e. basophilic atypical tubules (bATs), basophilic<br />
hyperplasia (bAHs), adenomas and carcinomas <strong>of</strong> a basophilic phenotype in all groups<br />
examined. Representative images <strong>of</strong> preneoplastic and neoplastic lesions are shown in<br />
Figure 4.3, A-C.<br />
One to 14 days MAMAc treatment did not result in a significant increase <strong>of</strong> preneoplastic<br />
and neoplastic lesions in either sex. Yet a clear, but non-significant increase in total bAT<br />
numbers were found in 14 days (Figure 4.4, A and B).<br />
84
Chapter 4: Manuscript II<br />
Figure 4.3: Representative preneoplastic lesions as observed in H&E stained renal sections (A)<br />
bAT; (B) bAH, (C) carcinoma, (D) BrdU staining.<br />
Although a 100% incidence <strong>of</strong> bATs was found in all treatment and control groups after<br />
three and six months <strong>of</strong> exposure, no significant increase <strong>of</strong> absolute numbers could be<br />
detected in MAMAc treated males, when compared to the respective controls (Figure 4.4, C<br />
and D). Only female Eker rats presented with a significant 1.7-fold increase <strong>of</strong> bATs after<br />
three months <strong>of</strong> MAMAc treatment (Figure 4.4, C). However, the latter effect appeared to<br />
be transient as it was not seen in the 6 month treated females (Figure 4.4, D). A higher yet<br />
non-significant total number and incidence <strong>of</strong> microscopically detected adenomas and<br />
carcinomas was present in 6 months MAMAc treated male than in female rats, when<br />
corrected for the numbers and incidences observed in the respective controls.<br />
85
Chapter 4: Manuscript II<br />
Figure 4.4: Mean number <strong>of</strong> preneoplastic lesions in male and female Eker rats after 1, 3, 7 and 14 days (A-B) or 3 and 6 month (C-D) <strong>of</strong> MAMAc exposure<br />
compared to the respective control groups.Light grey bars: bAT; white bars: bAHs, black.<br />
86
4.4.4 Cell proliferation<br />
Chapter 4: Manuscript II<br />
Site-specific cell proliferation analysis was assessed via PCNA and BrdU immunostaining<br />
(Figure 4.3, D) after acute and chronic MAMAc treatment respectively. In general nuclei<br />
with positive PCNA staining were rare (below 1% <strong>of</strong> all nuclei counted) and mainly localized<br />
in proximal tubules <strong>of</strong> the inner and outer cortex. As demonstrated in figure 4.5, A, 14 days<br />
treatment with MAMAc resulted in no significant change in cell proliferation rate in the<br />
proximal tubules in the outer or inner cortex <strong>of</strong> male Eker rats. In contrast 14 days<br />
treatment <strong>of</strong> females with MAMAc resulted in a significant 1.5 fold increase <strong>of</strong> cell<br />
proliferation in the inner cortex compared to the time-matched controls (Figure 4.5, B).<br />
Figure 4.5: (A-B) PCNA S-Phase mean labeling indices (LI %) in the outer and inner renal cortex<br />
<strong>of</strong> 14 days MAMAc treated male (A) and female (B) Eker rats, respectively (n=3 per<br />
group). (C-F) BrdU S-Phase mean labeling indices <strong>of</strong> male and female Eker rats<br />
treated with MAMAc for 3 (C-D) and 6 months (E-F), respectively (n=5 per group). Data<br />
represent the mean + SD. Significant differences between treated and time-matched<br />
control groups was analyzed using one-way ANOVA and Bonferroni's Post Test for<br />
multiple comparisons (*p < 0.05; **p < 0.01 and ***p
Chapter 4: Manuscript II<br />
BrdU-immunostaining based analysis <strong>of</strong> cell proliferation demonstrated no significant<br />
changes in cell proliferation rates in proximal tubules <strong>of</strong> three or six month treated male rats<br />
(Figure 4.5, C-F). Female rats showed a transient 1.8 fold increase <strong>of</strong> cell proliferation after<br />
3 months <strong>of</strong> MAMAc treatment in the inner cortex, whereas this was not observable after 6<br />
months <strong>of</strong> treatment (Figure 4.5, D and F). Distal tubules and collecting ducts did not show<br />
a significant increased cell proliferation compared to the control animals (data not shown).<br />
4.4.5 Detection <strong>of</strong> O6MG in MAMAc treated Eker-rats<br />
LC-MS/MS analysis <strong>of</strong> 100µg <strong>of</strong> pooled DNA, from three animals <strong>of</strong> the same treatment<br />
group i.e. 1, 7 and 14 days treatment, demonstrated a time-dependent accumulation <strong>of</strong><br />
O6MG (Figure 4.6). Concurrent O6MG adduct levels in the corresponding control rats were<br />
under the limit <strong>of</strong> detection. MAMAc treated female rats demonstrated slightly higher O6MG<br />
adduct levels than the corresponding treated males at all time-points examined.<br />
Figure 4.6: Number <strong>of</strong> O6MG adducts per 106 dG detected via LC-MS/MS in DNA <strong>of</strong> male and<br />
female Eker rats after 1, 7 and 14 days <strong>of</strong> MAMAc treatment (n=1, with DNA pooled<br />
from three replicate animals)<br />
88
4.5 Discussion<br />
Chapter 4: Manuscript II<br />
Cycasin and its active metabolite MAM(Ac) were shown to be carcinogenic in rodents when<br />
administered at very high doses (Hirono et al. 1968; Laqueur 1968; Fukunishi et al. 1971).<br />
However, the use <strong>of</strong> high doses <strong>of</strong> carcinogens may not reflect the carcinogenic potential <strong>of</strong><br />
a test compound und realistic exposure scenarios, since unspecific additional enhancing<br />
effects such as cytotoxicity and regenerative cell proliferation may occur at the high doses<br />
employed (Ames and Gold 1990; Foran 1997; Cohen and Arnold 2008). In addition, dosedependant<br />
differences in metabolism, as well as in pharmacokinetics <strong>of</strong> a compound, could<br />
also have an influence on carcinogenesis.<br />
Previous studies with rats exposed to low doses <strong>of</strong> DNA alkylating compounds<br />
demonstrated that short-term treatment with genotoxic (renal or hepatic) carcinogens<br />
resulted in a characteristic expression pr<strong>of</strong>ile, including the up-regulation <strong>of</strong> numerous<br />
genes involved in DNA damage response, cell cycle arrest and apoptosis (Ellinger-<br />
Ziegelbauer et al. 2004; van Delft et al. 2004; Ellinger-Ziegelbauer et al. 2005; Stemmer et<br />
al. 2007). Consequently and based on its genotoxic properties, it was assumed that<br />
MAMAc would result in a similar expression pr<strong>of</strong>ile as observed for other DNA alkylating<br />
compounds, even when administered at low doses. However, contrary to expectations<br />
treatment with 250µg/kg BW MAMAc did not result in an up-regulation <strong>of</strong> the expected<br />
genes, i.e. genes involved in DNA damage response, cell cycle arrest and apoptosis (Table<br />
4.1). This lack <strong>of</strong> deregulation observed in the microarray analyses was furthermore<br />
corroborated via quantitative real time PCR. For instance, MGMT, an important protein in<br />
the repair <strong>of</strong> alkylated DNA, was unchanged in male and female rats <strong>of</strong> both treatment<br />
groups, and at any exposure time (Supplementary Figure 4.1).<br />
In contrast to the latter, a time-dependent increase <strong>of</strong> MAMAc induced O6-methylguanine<br />
(O6meG) DNA-adducts in male and female Eker rats was observed (Figure 4.6). The lack<br />
<strong>of</strong> a measurable DNA damage response, as suggested from the data presented here, may<br />
be explained by: a) a time-dependent accumulation <strong>of</strong> O6meG DNA-adducts with a<br />
corresponding DNA damage response, yet changes in the DNA damage response genes<br />
were too small to be detected by gene expression analysis; b) the DNA adducts were not<br />
recognized, thereby leading to the observed lack <strong>of</strong> response at the gene expression level;<br />
and c) the DNA adducts were recognized but not sufficiently repaired with the DNA damage<br />
machinery expressed. The question, whether MAMAc-induced O6meG-adducts were<br />
sufficiently repaired and/or were not sufficiently recognized and therefore not repaired and<br />
thus have the potential to result in the onset <strong>of</strong> cancer could be answered by comparing<br />
89
Chapter 4: Manuscript II<br />
total numbers <strong>of</strong> preneoplastic lesions in 3 and 6 months MAMAc treated and control Eker<br />
rats. Indeed, 3 and 6 months exposure to renal carcinogens were previously demonstrated<br />
to be effective in the identification <strong>of</strong> genotoxic compounds in Eker rats (Walker et al. 1992;<br />
McDorman and Wolf 2002; Morton et al. 2002; Stemmer et al. 2007). Within this timeperiod<br />
genotoxic compounds may cause a second hit mutation in the remaining allele <strong>of</strong> the<br />
Tsc2 tumor suppressor gene. Therefore, a higher total number <strong>of</strong> preneoplastic renal<br />
lesions in chronically treated rats would indicate a genotoxic action <strong>of</strong> MAMAc, and an<br />
insufficient DNA damage response, when compared to vehicle control animals.<br />
However, treatment with MAMAc resulted only in a marginal increase <strong>of</strong> total numbers <strong>of</strong><br />
preneoplastic lesions (bATs and bAHs) at all time-points examined, and statistical<br />
significance was only reached in female rats after 3 months <strong>of</strong> treatment (Figure 4.4). The<br />
latter corresponding to the increased cell proliferation observed at 3 but not 6 months <strong>of</strong><br />
MAMAc treatment in female but not in male rats (Figure 4.5). The concurrent increase <strong>of</strong><br />
cell proliferation and preneoplastic lesions not only points to a critical role <strong>of</strong> cell<br />
proliferation for the formation <strong>of</strong> renal preneoplastic lesions, but also suggests a higher<br />
susceptibility <strong>of</strong> female rats towards MAMAc treatment. The latter is corroborated by the<br />
finding that an increased cell proliferation in female but not in male rats was already<br />
detectable after 14 days <strong>of</strong> MAMAc treatment. Since MAMAc treatment did not induce overt<br />
tubular damage, increased cell proliferation may rather result from a mitogenic mode <strong>of</strong><br />
action, than from a regenerative response.<br />
Above data thus suggest higher susceptibility <strong>of</strong> female rats to MAMAc-induced toxicity<br />
and/or carcinogenicity. Indeed, MAMAc treatment resulted in higher O6meG-adduct levels<br />
in females than in males (Figure 4.6). O6meG-adducts in turn may give rise to guanine to<br />
adenine transitions typically detected as a consequence <strong>of</strong> this adducts type. In<br />
combination with a higher MAMAc-induced cell proliferation, these mutations may be fixed<br />
before a sufficient repair is possible, and may explain the significantly increased number <strong>of</strong><br />
pre-neoplastic lesions in female Eker rats. In contrast in male Eker rats, prolonged<br />
treatment with MAMAc resulted in an increased total number and incidence <strong>of</strong> kidney<br />
tumors (adenomas and carcinomas), but not in a significant increase <strong>of</strong> preneoplastic<br />
lesions (Figure 4.4). Thus, in male rats low doses <strong>of</strong> MAMAc may exert their carcinogenic<br />
action rather by accelerating the progression <strong>of</strong> pre-existing (hereditary) preneoplastic<br />
lesions, than by lesion initiation. However, since MAMAc did not increase overall cell<br />
proliferation in male rats, tumor promotion cannot be explained by a mitogenic effect <strong>of</strong><br />
MAMAc. Another possible mechanism leading to MAMAc tumor promotion is suggested<br />
from the gene expression pr<strong>of</strong>iles <strong>of</strong> male Eker rats treated for 1, 3, 7 and 14 days with<br />
MAMAc. Although a relatively small number <strong>of</strong> genes appeared to be deregulated by<br />
90
Chapter 4: Manuscript II<br />
MAMAc, the most prominent gene cluster with 20% <strong>of</strong> all deregulated genes included<br />
down-regulated genes encoding basic transcription factors, and proteins involved in RNA<br />
splicing, translation and folding. This response is in agreement with previous findings <strong>of</strong> an<br />
early and transient inhibition <strong>of</strong> DNA-, RNA- and protein synthesis in the rat liver within the<br />
first hours <strong>of</strong> MAMAc treatment (Zedeck et al. 1970). More importantly, down-regulated<br />
genes included targets that are known for their tumor suppressing properties, such as the<br />
transcription factor PLAGL1, which encodes a growth suppressor protein frequently deleted<br />
in cancer (Abdollahi 2007), or BHLHB3, a candidate tumor suppressor gene in lung cancer<br />
(Falvella et al. 2008), leading to an overall reduced tumor suppressing environment in the<br />
kidney.<br />
In male rats, MAMAc furthermore resulted in an up-regulation <strong>of</strong> 4 genes involved in TGF-ß<br />
signaling, a pathway which plays an important role in the modulation <strong>of</strong> cellular functions<br />
such as immune function, cell proliferation, cell motility, glycolysis and epithelial-tomesenchymal<br />
transition (Leivonen and Kahari 2007). TGF-ß signaling is further known to<br />
posses janus-properties, i.e. it can function as a tumor suppressor at early stages <strong>of</strong> kidney<br />
cancer but it can also promote tumor progression and metastasis in later stages (Gold<br />
1999; Laping et al. 2007). Early stimulation <strong>of</strong> the TGF-ß pathway in male rats could<br />
therefore still be involved in tumor suppression, but may also result in tumor promotion at<br />
later time-points.<br />
In summary, above findings demonstrate that the combination <strong>of</strong> sensitive methods such as<br />
gene expression pr<strong>of</strong>iling and adduct analysis via LC-MS/MS with established methods e.g.<br />
cell proliferation analysis and histopathology allows to detect earliest carcinogen induced<br />
effects at low dose levels and at time-points when histopathological changes are not yet<br />
readily detectable. Above findings suggests that subchronic and –chronic exposure to low<br />
doses <strong>of</strong> MAMAc results in gender-specific differences in the carcinogenic response. In<br />
female rats the increase in early tumor stages and absence <strong>of</strong> late stages may point to a<br />
tumor initiating role <strong>of</strong> MAMAc, while in male rats, the increase in later tumor stages after 6<br />
months exposure is suggestive <strong>of</strong> a tumor promoting function.<br />
Acknowledgement:<br />
We would like to thank the Federal Ministry <strong>of</strong> Education and Research for funding the<br />
project (BMBF: 0313024), Tanja Lampertsdörfer and Gudrun von Scheven, Alexandra<br />
Heussner and Evelyn O’Brien for skilful assistance during the animal experiment, Kerstin<br />
Albrecht for microarray hybridization, Valeriy Afonin for DNA extraction for LC-MS/MS<br />
analysis and Paul T. Pfluger for critically reading the manuscript.<br />
91
4.6 Supplement<br />
4.6.1 Supplementary experimental procedures: Real time PCR<br />
Chapter 4: Manuscript II<br />
To evaluate gene expression by real time PCR, total RNA was extracted from RNAlater<br />
(Qiagen, Germany) fixed renal cortex <strong>of</strong> 14-days MAMAc treated and control male and<br />
female Eker rats (n=3) using RNEasy Mini Kit (Qiagen, Germany), according to<br />
manufacturer’s instructions. Following DNase treatment, reverse transcriptions were<br />
performed using SuperScript III (Invitrogen) and random primers (Invitrogen). Real-time<br />
PCRs <strong>of</strong> the house keeping gene hypoxanthin phosphoribosyl transferase (HPRT: forward<br />
primer: aagacagcggcaagttgaat; reverse primer: ggctgcctacaggctcatag) and <strong>of</strong> candidate<br />
genes O6-methylguanine–DNA methyltransferase (MGMT: forward primer:<br />
cagtaggaggagcgatgagg; reverse primer: gctggaagactcgaaggatg) were performed on an<br />
Eppendorf Cycler using iQ-SybrGreen Supermix (Biorad). All samples were performed in<br />
technical duplicates. Relative amounts <strong>of</strong> gene-specific mRNAs were calculated via Ct<br />
values, based on a standard curve <strong>of</strong> five points with known amounts <strong>of</strong> template cDNA.<br />
4.6.2 Supplementary Figures<br />
Supplementary figure 4.1: Gene expression analysis <strong>of</strong> MGMT in male and female Eker rats<br />
following 14-days treatment with MAMAc. Data <strong>of</strong> three replicate animals are shown<br />
as mean + SEM. One-way ANOVA with Bonferroni's post test, demonstrated no<br />
significant difference <strong>of</strong> male and female MAMAc treated Eker rats vs. the respective<br />
controls.<br />
92
4.6.3 Supplementary Tables<br />
Chapter 4: Manuscript II<br />
Supplementary table 4.1: Non-neoplastic pathology <strong>of</strong> the renal cortex <strong>of</strong> male Eker and wild type rats treated with MAMAc for 1, 3 7 and 14 days respectively<br />
(n=3 per group). Histopathological changes were ranked from none (0) to severe (4) including intermediate classes. Values are presented as<br />
median ± median absolute deviation. Incidences are given in parenthesis.<br />
Pigment<br />
Deposits<br />
Necrosis<br />
Apoptosis<br />
Karyomegally<br />
Vacuolization<br />
Cell shedding<br />
Proteinaceous<br />
casts<br />
Tubular<br />
dilatation<br />
Calcium casts<br />
Control (male Eker rats) MAMAc (maleEker rats) Control (female Eker rats) MAMAc (femaleEker rats)<br />
d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.3<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(3/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(3/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.3<br />
(2/3)<br />
0.5+0.0<br />
(3/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
93
Tubular<br />
regeneration<br />
CPN<br />
Inflammation<br />
Chapter 4: Manuscript II<br />
Control (male Eker rats) MAMAc (maleEker rats) Control (female Eker rats) MAMAc (femaleEker rats)<br />
d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.2<br />
(0/3)<br />
Supplementary table 4.2: Incidence and mean numbers <strong>of</strong> preneoplastic and neoplastic lesions <strong>of</strong> the renal cortex. Mean + SD <strong>of</strong> atypical tubule (AT), aypical<br />
hyperplasia (AH) and neoplastic (adenoma, carcinoma) lesions <strong>of</strong> male and female MAMAc treated Eker and wild type rats and their respective<br />
controls (n=3). Incidences <strong>of</strong> respective lesions in treated or control animals are given in parentheses.<br />
Group Sex AT: Mean + SD AH: Mean + SD Neoplastic lesions: Mean + SD<br />
d1 d3 d7 d14 d1 d3 d7 d14 d1 d3 d7 d14<br />
Control male 1.7 ± 0.9 1.3 ± 0.5 2.7 ± 3.8 1.3 ± 1.2 0.7 ± 0.5 0.0 ± 0.0 0.3 ± 0.6 0.3 ± 0.5 0.3 ± 0.5<br />
(3/3) (3/3) (2/3) (2/3) (2/3) (0/3) (1/3) (1/3)<br />
A<br />
0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0<br />
(1/3)<br />
(0/3) (0/3) (0/3)<br />
MAMAc male 1.7 ± 1.2 2.0 ± 1.4 1.0 ± 0.8 1.0 ± 0.8 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.3 ± 0.5<br />
(2/3) (3/3) (2/3) (2/3) (0/3) (0/3) (0/3) (0/3)<br />
A<br />
0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0<br />
(1/3)<br />
(0/3) (0/3) (0/3)<br />
Control female 3.0 ± 2.2 2.0 ± 1.4 1.7 ± 1.5 2.7 ± 0.9 0.0 ± 0.0 0.7 ± 0.5 0.0 ± 0.0 0.3 ± 0.5 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0<br />
(2/3) (3/3) (2/3) (3/3) (0/3) (2/3) (0/3) (1/3) (0/3)<br />
(0/3) (0/3) (0/3)<br />
MAMAc female 2.0 ± 0.8 2.7 ± 3.1 2.7 ± 2.5 5.0 ± 1.4 0.3 ± 0.5 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.3 ± 0.5<br />
(3/3) (2/3) (3/3) (3/3) (1/3) (0/3) (0/3) (0/3) (0/3)<br />
(0/3) (0/3)<br />
A<br />
(1/3)<br />
A C<br />
: Adenoma; : Carcinoma<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(1/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.2<br />
(0/3)<br />
0.0+0.0<br />
(0/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.2<br />
(2/3)<br />
0.0+0.0<br />
(0/3)<br />
0.0+0.2<br />
(1/3)<br />
0.5+0.0<br />
(3/3)<br />
0.0+0.0<br />
(0/3)<br />
94
Chapter 5: Manuscript III<br />
Establishment <strong>of</strong> a Protocol for the Analysis <strong>of</strong> Laser<br />
Microdissected Rat <strong>Kidney</strong> Samples on Affymetrix Genechips<br />
Kerstin Stemmer a, 1 , Heidrun Ellinger-Ziegelbauer b, 1 , Kerstin Lotz b , Hans-J. Ahr b and<br />
Daniel R. Dietrich a, *<br />
a<br />
Human and Environmental Toxicology, University <strong>of</strong> Konstanz, Konstanz, Germany<br />
b<br />
<strong>Molecular</strong> and Special Toxicology, Bayer Healthcare AG, Wuppertal, Germany<br />
1<br />
These authors contributed equally to this work.<br />
* Corresponding author: Daniel R. Dietrich: daniel.dietrich@uni-konstanz.de<br />
Published in Toxicology and Applied Pharmacology (2006) Nov 15; 217(1):134-42<br />
5.1 Abstract<br />
Laser microdissection in conjunction with microarray technology allows selective isolation<br />
and analysis <strong>of</strong> specific cell populations, e.g., preneoplastic renal lesions. To date, only<br />
limited information is available on sample preparation and preservation techniques that<br />
result in both optimal histomorphological preservation <strong>of</strong> sections and high-quality RNA for<br />
microarray analysis. Furthermore, amplification <strong>of</strong> minute amounts <strong>of</strong> RNA from<br />
microdissected renal samples allowing analysis with gene chips has only scantily been<br />
addressed to date. The objective <strong>of</strong> this study was therefore to establish a reliable and<br />
reproducible protocol for laser microdissection in conjunction with microarray technology<br />
using kidney tissue from Eker rats p.o. treated for 7 days and 6 months with 10 and 1mg<br />
aristolochic acid/kg bw, respectively. <strong>Kidney</strong> tissues were preserved in RNAlater or snap<br />
frozen. Cryosections were cut and stained with either H&E or cresyl violet for subsequent<br />
morphological and RNA quality assessment and laser microdissection. RNA quality was<br />
comparable in snap frozen and RNAlater-preserved samples, however, the<br />
histomorphological preservation <strong>of</strong> renal sections was much better following<br />
cryopreservation. Moreover, the different staining techniques in combination with sample<br />
processing time at room temperature can have an influence on RNA quality. Different RNA<br />
amplification protocols were shown to have an impact on gene expression pr<strong>of</strong>iles as<br />
95
Chapter 5: Manuscript III<br />
demonstrated with Affymetrix Rat Genome 230_2.0 arrays. Considering all the parameters<br />
analyzed in this study, a protocol for RNA isolation from laser microdissected samples with<br />
subsequent Affymetrix chip hybridization was established that was also successfully<br />
applied to preneoplastic lesions laser microdissected from aristolochic acid-treated rats.<br />
5.2 Introduction<br />
An exponentially increasing number <strong>of</strong> publications indicate that microarray technology is<br />
now an accepted and powerful tool for large-scale gene expression pr<strong>of</strong>iling. However, the<br />
outcome and interpretation <strong>of</strong> microarray studies can be strongly affected by inherent tissue<br />
heterogenicity. For example the kidney represents a tissue <strong>of</strong> high complexity with various<br />
structurally and functionally different compartments. It is therefore very difficult to causally<br />
link changes in gene expression patterns from whole kidney homogenates to cell-typespecific<br />
gene expression, the associated pathways and subsequent physiological effects. In<br />
addition, site-specific pathology-related gene expression changes, e.g., within a distinct<br />
tubular segment or a preneoplastic or neoplastic lesion can be diluted by the gene<br />
expression pr<strong>of</strong>ile <strong>of</strong> other cell types in a whole tissue homogenate and thus not be readily<br />
detectable. Laser-assisted microdissection could be a useful tool to overcome such<br />
obstacles. Two <strong>of</strong> the most common technologies are laser capture microdissection (LCM)<br />
(Emmert-Buck et al. 1996) and laser microdissection and pressure catapulting (LMPC)<br />
(Schutze et al. 1998). Both technologies allow the dissection <strong>of</strong> cell populations, single cells<br />
or even sub-cellular structures from frozen or paraffin-embedded tissues. As the<br />
microdissected structures remain morphologically intact and macromolecules are not<br />
damaged by the laser beam, proteins, DNA or RNA can be isolated for subsequent<br />
analyses. However, the combination <strong>of</strong> laser-assisted microdissection and microarray<br />
technology has to cope with three main challenges: (1) minimizing RNA degradation over<br />
several tissue (section) processing steps, (2) preserving tissue morphology for accurate<br />
histological examination <strong>of</strong> lesions selected for microdissection and (3) maintaining the<br />
proportionality <strong>of</strong> specific RNA species after several rounds <strong>of</strong> RNA amplification, required<br />
to obtain sufficient RNA for chip hybridization. It is important to conserve high-quality RNA,<br />
as degraded RNA can result in detection <strong>of</strong> an altered expression pattern (Spiess et al.<br />
2003). RNA must therefore be protected during all processing steps, e.g., tissue harvesting,<br />
storage, sectioning, staining and fixation and finally during microdissection. Most <strong>of</strong> these<br />
steps are conducted at room temperature. The use <strong>of</strong> the commercially available stabilizing<br />
reagent RNAlater (e.g., from Ambion or Qiagen) could therefore be a promising tool for<br />
96
Chapter 5: Manuscript III<br />
RNA preservation in cryo-fixed sections. As a first step <strong>of</strong> the experiments presented here,<br />
the use <strong>of</strong> RNAlater-preserved to snap frozen tissue with respect to tissue morphology and<br />
RNA quality was compared. Furthermore, the influence <strong>of</strong> long term-storage, repeated use<br />
<strong>of</strong> same cryo-samples and different staining techniques on histomorphological tissue<br />
preservation and RNA quality was determined.<br />
Besides maintaining tissue morphology and high-quality RNA, another challenge to be<br />
dealt with is the low amount <strong>of</strong> RNA obtained from microdissected material. Standard<br />
protocols for microarray analysis are based on starting amounts <strong>of</strong> 1–5μg total RNA. Yet<br />
laser-assisted microdissection yields at best nanograms <strong>of</strong> RNA. To overcome this<br />
obstacle, it is necessary to employ RNA amplification methods to generate the required<br />
microgram amounts <strong>of</strong> RNA. Linear one-cycle mRNA amplification by a T7-based in vitro<br />
transcription (IVT) protocol was first described by Van Gelder et al. (Van Gelder et al. 1990)<br />
and became the standard amplification and labeling protocol for the Affymetrix Chip<br />
technology. A second amplification round can be performed to further increase the RNA<br />
yield. However, a major concern remains as to whether the amplification via IVT maintains<br />
the proportionality <strong>of</strong> RNA species present in the original sample, since it was shown<br />
previously that gene expression ratios were not completely conserved between linearly<br />
amplified and non-amplified material (Baugh et al. 2001; Goley et al. 2004; Nygaard et al.<br />
2005). Consequently, the comparability <strong>of</strong> one- and two-cycle amplification protocols on the<br />
gene expression pr<strong>of</strong>iles <strong>of</strong> rat kidney samples was elucidated and presented here. <strong>Kidney</strong><br />
samples were taken from rats treated with vehicle control or Aristolochic acid (AA), a potent<br />
genotoxic renal toxicant and carcinogen that was previously shown to induce tumors in<br />
kidney, forestomach and urinary bladder <strong>of</strong> the rat (Mengs et al. 1982) as well as renal<br />
toxicity, progradient fibrosis and urothelial tumors in women exposed to AA in Chinese<br />
slimming teas (Vanherweghem et al. 1993; Nortier et al. 2000). Based on the results <strong>of</strong> the<br />
experiments presented here, a protocol for RNA isolation from laser microdissected<br />
samples with subsequent Affymetrix chip hybridization was established that was also<br />
successfully applied to preneoplastic lesions laser microdissected from Aristolochic acidtreated<br />
rats.<br />
97
5.3 Material and Methods<br />
5.3.1 Animal treatment and tissue preparation<br />
Chapter 5: Manuscript III<br />
Heterozygous Tsc2 mutant Eker rats were purchased at 6–8weeks <strong>of</strong> age from the<br />
University <strong>of</strong> Texas MD Anderson <strong>Cancer</strong> Center, Smithville, USA. Upon arrival, rats were<br />
allowed to acclimatize to laboratory conditions for 4 weeks. For the short-term experiment<br />
(7-day treatment) three randomly selected male Eker rats were gavaged daily with 10mg<br />
per kg BW Aristolochic acid (AA) per kg BW dissolved in 0.1M sodium bicarbonate, or with<br />
vehicle (0.1Msodium bicarbonate) alone. At the end <strong>of</strong> the experiment, rats were<br />
anesthetized and then killed via exsanguination subsequent to retrograde perfusion with<br />
PBS. <strong>Kidney</strong>s were collected, cross sectioned to 5-mm slices and either snap frozen with<br />
liquid nitrogen and stored at −80°C or fixed in RNAlater (Qiagen, Hilden, Germany)<br />
according to the manufacturer's instructions. For the latter procedure, kidney slices were<br />
incubated at room temperature for at least 1h, kept at 4°C for up to 1 week, and then<br />
removed from the RNAlater reagent. From these kidney slices, kidney cortex was isolated<br />
and cortex and medulla stored separately in cryovials at −80°C. The RNA <strong>of</strong> the kidney<br />
cortex samples fixed in RNAlater was used for the comparison <strong>of</strong> the one- and two-cycle<br />
protocols shown in Figs. 3–5. In the long-term exposure (6months) five male Eker rats per<br />
group were gavaged on 5 days per week for 6months with 1mg per kg BWAA dissolved in<br />
0.1M sodium bicarbonate and with vehicle (0.1M sodium bicarbonate) respectively.<br />
Following treatment, kidneys were collected as described for the short-term experiment<br />
(above) and then stored using two different protocols: one set <strong>of</strong> kidney cross-sections was<br />
stored in RNAlater as described above, and a second set <strong>of</strong> kidney slices was snap frozen<br />
in liquid nitrogen immediately after harvesting and subsequently stored at −80°C. The<br />
kidney samples <strong>of</strong> these animals were used for results shown in (Figs. 1, 2, and 6) and in<br />
all tables.<br />
5.3.2 Sectioning and staining<br />
Due to the high chaotrophic salt concentration <strong>of</strong> the RNAlater reagent, RNAlater preserved<br />
samples will not freeze in the cryostate and thus were thawed and then washed for 2×5min<br />
or 2×15min in ice-cold RNase-free PBS. For embedding samples were refrozen in the<br />
cryostate at −20°C. To avoid interference <strong>of</strong> frozen tissue embedding reagent OCT<br />
(Cryoblock, medite Medizintechnik GmbH, Burgdorf, Germany) with laser efficiency and<br />
RNA isolation, samples were frozen only on a small drop <strong>of</strong> OCT instead <strong>of</strong> being fully<br />
98
Chapter 5: Manuscript III<br />
embedded in the medium. Both RNAlater- and snap-frozen kidney samples were cut to 10μm-thick<br />
sections on a cryotome. Sections were mounted onto RNase-free membranecovered<br />
slides (PALM Microlaser GmbH, Bernried, Germany) and air-dried on ice for 20 s.<br />
Subsequently, sections were fixed in ice-cold 70% ethanol (RNase-free) for 3min and airdried<br />
for 5 to 10min. For H&E (hematoxylin–eosin) staining, air-dried sections were stained<br />
with RNase-free Mayer's hematoxylin (Sigma Aldrich, Taufkirchen, Germany) for 3min,<br />
rinsed with RNase-free tap water and counterstained with RNase-free eosin Y (Sigma<br />
Aldrich, Taufkirchen, Germany), for 3min. For cresyl violet staining, ethanol-fixed sections<br />
were stained with 1% (w/v) cresyl violet acetate (Sigma Aldrich, Taufkirchen, Germany) in<br />
ACS-grade 100% EtOH for 20s. Excess staining solution was removed by placing the<br />
sections on an absorbent surface. Subsequent to H&E or cresyl violet staining, sections<br />
were dehydrated by shortly dipping the sections in 70% and then 100% ethanol. For further<br />
information, see the customer protocol page <strong>of</strong> the PALM homepage at: www.palmmicrolaser.com.<br />
5.3.3 Laser-assisted microdissection<br />
For laser-assisted microdissection, a PALM laser microdissection and pressure catapulting<br />
(LMPC) system (PALM Microlaser GmbH, Bernried, Germany) was used. Areas <strong>of</strong> variable<br />
size were dissected from H&E-stained whole kidney sections (cresyl violet-stained sections<br />
were not used due to the poor histomorphological preservation <strong>of</strong> the sections, see below).<br />
The microdissected samples were collected in a non-contact and contamination-free<br />
manner, immediately lysed by addition <strong>of</strong> 300μl RLT lysis buffer (RNeasy Micro Kit, Qiagen,<br />
Hilden, Germany) containing 1% <strong>of</strong> β-mercaptoethanol (Sigma Aldrich,Taufkirchen,<br />
Germany) and stored at −80°C. For quality control, total RNA was isolated from additional,<br />
non-dissected and unstained whole sections from each kidney examined.<br />
5.3.4 RNA isolation and quality control<br />
RNA <strong>of</strong> microdissected tissue was isolated using the RNeasy Micro Kit (Qiagen, Hilden,<br />
Germany). RNA from whole kidney cortex immersed in RNAlater was isolated with Qiagen<br />
RNeasy Mini Kits. All isolations were performed according to the manufacturer's<br />
instructions. The quality <strong>of</strong> the total RNA was analyzed with the Agilent 2001 Bioanalyzer<br />
(Agilent Technologies GmbH, Germany) using RNA 6000 Nano or Pico chips. Agilent 2100<br />
expert s<strong>of</strong>tware was used to determine RNA quality, designated as RNA Integrity Number<br />
99
Chapter 5: Manuscript III<br />
(RIN). RIN is calculated from the entire electrophoretic trace by Agilent 2100 expert<br />
s<strong>of</strong>tware and thus represents the relative ratios <strong>of</strong> several RNA regions, e.g., the 5S rRNA<br />
region, the region between the 5S and 18S rRNA and the 18S and 28S rRNA bands.<br />
Based on more than 1000 RNA samples <strong>of</strong> different quality, an algorithm was generated<br />
that describes RNA integrity much more reliably then than the mere ratio <strong>of</strong> the ribosomal<br />
(18S and 28S rRNA) bands. A RIN <strong>of</strong> 7–10 is indicative <strong>of</strong> high RNA quality, whereas a RIN<br />
<strong>of</strong> 1 marks a completely degraded RNA sample (Schroeder et al. 2006). For RNA<br />
concentrations ≥5ng/μl (kidney cortex samples), the RNA quantity was determined using<br />
Ribogreen (<strong>Molecular</strong> Probes, Eugene, USA). For lower RNA concentrations (LMPC<br />
samples) the RNA quantity estimate by the Agilent 2100 expert s<strong>of</strong>tware was utilized. Data<br />
were statistically analyzed using GraphPad Prism 4 s<strong>of</strong>tware with oneway ANOVA followed<br />
by Bonferroni's post test for multiple comparisons or Student's t-test for individual<br />
comparisons.<br />
5.3.5 RNA amplification<br />
For the synthesis <strong>of</strong> biotin-labeled cRNAwith one or two rounds <strong>of</strong> amplification, the<br />
Affymetrix One-cycle and Two-cycle kits were used according to the manufacturer's<br />
instructions, the latter in conjunction with the Megascript T7 kit from Ambion (Austin, USA).<br />
5μg <strong>of</strong> total RNAwas used for the one-cycle protocol, during which the RNA is first reversetranscribed<br />
in the presence <strong>of</strong> an oligo-dT primer coupled with a primer sequence for<br />
bacterial T7-polymerase. Following the synthesis <strong>of</strong> the second cDNA strand, this cDNA<br />
template is then transcribed with T7-polymerase in the presence <strong>of</strong> biotin-labeled<br />
ribonucleotides to yield biotin-labeled antisense cRNA. The concentration, corrected for<br />
input RNA, and the RNA quality was determined based on OD 260/280 measurement and<br />
the Agilent 2100 Bioanalyzer pr<strong>of</strong>ile. Up to 50ng <strong>of</strong> total RNA was employed as starting<br />
material with the two-cycle protocol. After the first round <strong>of</strong> cDNA synthesis as described<br />
above, unlabeled antisense cRNA was synthesized with the Megascript T7 kit from Ambion<br />
(Austin, USA). The cRNA from this first round <strong>of</strong> amplification was quantified after column<br />
purification using Ribogreen. The first strand <strong>of</strong> the second round <strong>of</strong> cDNA synthesis was<br />
primed with random primers, whereas the second strand was primed with the oligo-dT-T7<br />
primer as described above. Finally, biotin-labeled cRNA was synthesized, quantified and<br />
quality controlled as described for the one-cycle kit.<br />
100
5.3.6 Affymetrix chip hybridization and expression pr<strong>of</strong>iling<br />
Chapter 5: Manuscript III<br />
15μg <strong>of</strong> the fragmented biotin-labeled cRNA samples was hybridized on Affymetrix Rat<br />
Genome RAE230_2.0 Genechips according to the manufacturer's instructions.<br />
RAE230_2.0 Genechips contain 31,000 probe sets represented by 11 pairs <strong>of</strong> 25mer<br />
oligonucleotides. Each probe pair consists <strong>of</strong> a perfect match oligo (PM) complementary to<br />
the cRNA target sequence, and a mismatch oligo (MM) with a single base change in the<br />
middle to control for background and nonspecific hybridization. Raw data image files (DAT)<br />
were converted into CEL files by averaging the scan data from the 36 pixels per oligo set.<br />
With GeneData Expressionist Refiner s<strong>of</strong>tware, dark and white spots, gradients and<br />
distortions were detected and corrected using the CEL file data. Using the MAS 5.0<br />
statistical algorithms implemented in the Refiner s<strong>of</strong>tware, the intensities <strong>of</strong> all 11 probe<br />
pairs were condensed to one intensity value per probe set associated with a statistical<br />
detection p-value calculated from the intensity differences <strong>of</strong> the PM and corresponding<br />
MM oligos. This p-value indicates how reliably a transcript is detected. After condensing,<br />
which also includes overall microarray background correction, microarrays were scaled to<br />
an average signal intensity <strong>of</strong> 100 after excluding the highest 2% and lowest 2% <strong>of</strong> the<br />
data.<br />
To select significantly deregulated genes between replicate sample groups representing<br />
kidney samples from short-term Aristolochic acid-treated and vehicle control Eker rats, a ttest<br />
(p≤0.005) combined with a 1.7-fold deregulation cut-<strong>of</strong>f implemented in Genedata<br />
Expressionist Analyst s<strong>of</strong>tware was employed. Genes with an intensity <strong>of</strong> expression
5.4 Results<br />
5.4.1 Tissue fixation and staining<br />
Chapter 5: Manuscript III<br />
I: Influence <strong>of</strong> tissue preservation on histology <strong>of</strong> the sections and on RNA<br />
quality<br />
Degradation <strong>of</strong> RNA can result in detection <strong>of</strong> altered gene expression patterns. Therefore,<br />
the question was asked whether tissue preservation with RNAlater would stabilize RNA in<br />
cryosections and whether the required PBS washing steps would negatively influence RNA<br />
stability. Two RNAlater fixed kidney pieces, each <strong>of</strong> three replicate animals was washed for<br />
2×5min or 2×15min in RNase-free, ice-cold PBS. Subsequently, cryosections were cut and<br />
RNA was isolated from a complete unstained and an H&E-stained section. As control, RNA<br />
was isolated from tissue homogenate <strong>of</strong> a third, unwashed RNAlater fixed kidney piece.<br />
Table 5.1 depicts the RNA quality <strong>of</strong> the samples, displayed as RNA integrity number<br />
(RIN).<br />
Table 5.1: Influence <strong>of</strong> tissue preservation on RNA quality. RNAlater-immersed kidney pieces<br />
from three replicate animals were washed for 2x 5min or 2x 15min in ice-cold PBS<br />
before freezing and cryo-sectioning. RNA was isolated from one unstained or one<br />
H&E-stained section each. Control RNA was isolated directly from RNAlater-fixed,<br />
homogenized kidney cortex. RNA quality is displayed as mean RNA integrity number<br />
(RIN) ± SD. One-way ANOVA and Bonferroni’s post test (p< 0.05(*) and p< 0.01 (**))<br />
suggested the absence <strong>of</strong> significant differences between the samples.<br />
Sample Staining Washing in PBS RIN (mean and SD <strong>of</strong> 3 replicates)<br />
Control N/A N/A 8,8 ± 0,5<br />
Section 1 No 2 x 5min 8,8 ± 0,6<br />
Section 2 H&E 2 x 5min 8,0 ± 0,1<br />
Section 3 No 2 x15min 8,6 ± 0,4<br />
Section 4 H&E 2 x15min 7,6 ± 0,7<br />
The RNA quality is very good after 2×5min or 2×15min <strong>of</strong> washing in ice-cold PBS, with no<br />
significant difference to the control RNA from unwashed RNAlater-fixed tissue. The<br />
comparison with the RNA quality retrieved from snap-frozen tissue (Table 5.2, see below),<br />
RNAlater preservation provided for slightly lower RNA quality.<br />
102
Chapter 5: Manuscript III<br />
Table 5.2: Influence <strong>of</strong> long-term storage on RNA quality (RIN) and yield (ng RNA). RNA was<br />
isolated from one complete section from the kidneys <strong>of</strong> 4 replicate animals. <strong>Kidney</strong><br />
samples were snap-frozen rather than RNAlater immersed. The duration the tissues<br />
were kept frozen at –80°C as <strong>of</strong> initial freeze-storing is shown in the first column.<br />
Duration frozen at –80°C RIN (mean ± SD <strong>of</strong> 4 replicates) RNA ng yield (mean ± SD <strong>of</strong> 4 replicates)<br />
5 months 9,4± 0,4 681.6 ±103.3<br />
12 months 9,2 ± 0,2 493.5 ±135.7<br />
<strong>Kidney</strong> morphology from RNAlater-fixed tissue sections was hardly discernible in unwashed<br />
kidney sections and also in sections washed for 2×5min (Figure 5.1, A). Further washing in<br />
PBS appeared to improve the preservation <strong>of</strong> renal tubular morphological <strong>characteristics</strong> <strong>of</strong><br />
the renal cortical structure (Figure 5.1, B). However, cryosections from RNAlater fixed<br />
tissue never achieved the histomorphological quality <strong>of</strong> cryosections from snap-frozen<br />
tissue even when stored up to 12months at −80°C (Figure 5.1, C). Use <strong>of</strong> snap-frozen<br />
kidneys instead <strong>of</strong> RNAlater-fixed tissue appeared to allow a more precise histological and<br />
pathological examination, especially with regard to the identification <strong>of</strong> tubular structural<br />
detail, and thus to facilitate the structural identification required for LMPC.<br />
Figure 5.1: Histology <strong>of</strong> H&E stained sections. (A) Section from RNAlater-fixed tissue after<br />
2x5min washes in PBS. (B) Section from RNAlater fixed tissue after 2x15min washes<br />
in PBS. (C) Section from snap-frozen tissue, without an RNAlater presoak.<br />
103
II: RNA stability in snap-frozen kidney samples<br />
Chapter 5: Manuscript III<br />
Since long-term storage <strong>of</strong> frozen tissue samples may have a negative impact on<br />
preservation <strong>of</strong> histological features and RNA stability, the influence <strong>of</strong> cryo-storage on<br />
histomorphological <strong>characteristics</strong> and RNA quality was investigated with snap-frozen<br />
tissue stored at −80°C for 5 or 12 months post initial cryopreservation. RNA was isolated<br />
from renal cryosections <strong>of</strong> four independent animals. No significant decrease <strong>of</strong> RNA<br />
quality (RIN) or yield (ng) could be observed in frozen tissue stored up to 12months at<br />
−80°C (Table 5.2). To elucidate whether repeated “thawing” <strong>of</strong> tissue samples from −80°C<br />
to cryostate temperature (−18°C to −20°C) has any influence on RNA stability or tissue<br />
morphology, RNAwas isolated from complete cryosections <strong>of</strong> three replicate animals, that<br />
had been snap frozen, stored at −80°C for 12months, and were then subjected to one or<br />
three cycles <strong>of</strong> “thawing” and refreezing. The mean RIN number <strong>of</strong> the respective samples<br />
only slightly decreased from 9.3±0.2 to 8.5±0.2 after the first and the third freeze–thaw<br />
cycle, respectively, yet still providing for high-quality RNA. Microscopic examination <strong>of</strong><br />
multiple thawed samples demonstrated that tissue morphology was not adversely affected,<br />
thus also allowing pathological examination and histological differentiation (see also Figure<br />
5.1, C and above).<br />
III: Influence <strong>of</strong> the duration <strong>of</strong> section staining procedure on RNA quality<br />
Current working procedures require cryosections to be microdissected at room<br />
temperature. Depending on the number <strong>of</strong> areas that have to be collected from one specific<br />
cryosection, the actual dissection could last for several hours. However, both time and<br />
temperature could be critical factors for the isolation <strong>of</strong> high-quality RNA. RNA yield (ng)<br />
and quality (RIN) may in addition be adversely influenced by different staining methods.<br />
However, for kidney pathology H&E staining is indispensable because it facilitates the<br />
distinction <strong>of</strong> morphological features as well as allows the detection <strong>of</strong> characteristic<br />
basophilic and eosinophilic preneoplastic and neoplastic lesions (Dietrich and Swenberg,<br />
1991). Therefore, the influence <strong>of</strong> H&E staining, temperature and LMPC duration on RNA<br />
yield and quality from H&E-stained cryosections was investigated. In addition, RNA stability<br />
<strong>of</strong> H&E-stained sections were compared with that <strong>of</strong> cresyl violet-stained sections, since<br />
cresyl violet is a recommended staining method for subsequent RNA applications<br />
(customer protocol page at: http://www.palm-microlaser.com). The comparison <strong>of</strong> RNA<br />
immediately retrieved from unstained control sections and H&E-stained sections<br />
demonstrated no significant influence <strong>of</strong> staining on RNA quality (Figure 5.2, 1st and 2nd<br />
bar). Additional handling <strong>of</strong> the sections at room temperature, up to 3 h, tendentially but not<br />
significantly decreased RNA quality. Additional handling time, 6h to overnight, decreased<br />
RNA quality significantly with RIN number decreasing to values below 7.0 (Figure 5.2, bar<br />
104
Chapter 5: Manuscript III<br />
2–6). In contrast cresyl violet staining resulted in a significant decrease <strong>of</strong> the RIN number<br />
directly after staining (Figure 5.2, 1st and 8th bar). However, as <strong>of</strong> then RNA quality<br />
appeared to remain stable over time (Figure 5.2, bar 8–12). Storing the differently stained<br />
sections overnight at −20 °C in general appeared to reduce the rate <strong>of</strong> RNA degradation.<br />
When comparing the RIN numbers <strong>of</strong> the H&E with the cresyl violet stained sections within<br />
the first 3 h <strong>of</strong> handling at room temperature, thus representing a reasonable processing<br />
time, H&E sections provided for a higher RNA quality than the corresponding cresyl violetstained<br />
sections. Therefore, RNA <strong>of</strong> high quality that allows additional RNA processing,<br />
e.g., chip hybridization, can be best obtained from H&E-stained cryosections within 3h post<br />
staining.<br />
Figure 5.2: <strong>Kidney</strong> tissue was frozen immediately after sacrifice and stored at –80°C for 12<br />
months. RNA was isolated from one complete kidney section <strong>of</strong> 3 independent<br />
replicate animals. Sections were stained with Hematoxylin and Eosin (H&E, black<br />
bars) or Cresyl Violet (white bars) and subsequently incubated for 0- 6 hours or over<br />
night (ON) at RT or at -20°C. The control sections (1st bar) were lysed immediately<br />
after sectioning. Significant differences, analyzed by Bonferroni’s multiple<br />
comparison test, are indicated using significance levels <strong>of</strong> p< 0.05(*) and p< 0.001<br />
(**).<br />
5.4.2 RNA quality and yield from pooled microdissected samples<br />
To investigate whether the area <strong>of</strong> pooled microdissected samples had an influence on<br />
RNA quality and yield, areas <strong>of</strong> 1mm 2 and 2mm 2 were laser microdissected from H&Estained<br />
cryosections and analyzed. These analyses as shown in table 5.3 suggested that<br />
total RNA from pooled sections were <strong>of</strong> comparably high quality (RIN >8.0). RNA yields<br />
from both 1mm2 or 2mm2 sections (Table 5.3, also see below) indicated that<br />
105
Chapter 5: Manuscript III<br />
microdissection <strong>of</strong> an approximate area <strong>of</strong> at least 2mm 2 was necessary to achieve the<br />
minimum amount <strong>of</strong> RNA (5ng to 100ng <strong>of</strong> total RNA) required for the two-cycle<br />
amplification protocol. For area comparison, a cross-sectioned glomerulus has an average<br />
area <strong>of</strong> 0.02mm 2 .<br />
Table 5.3: RNA quality and yields from pooled microdissected samples. RNA was isolated from<br />
pooled microdissected samples <strong>of</strong> an overall area <strong>of</strong> 1mm 2 or 2mm 2 .<br />
Pooled area RIN (mean ± SD <strong>of</strong> 3 replicates) Yield ng RNA (mean ± SD <strong>of</strong> 3 replicates)<br />
1mm 2 8,0 ±0,2 3,4 ±0,3<br />
2mm 2 8,3 ±0,4 8,8 ±1,7<br />
5.4.3 RNA amplification and microarray hybridization<br />
I: Influence <strong>of</strong> one-cycle and two-cycle amplification on the expression pr<strong>of</strong>ile<br />
Laser-assisted microdissection only yielded nanograms <strong>of</strong> RNA, thus necessitating global<br />
mRNA amplification for an intended subsequent microarray analysis. Therefore, microarray<br />
experiments with different starting amounts <strong>of</strong> kidney RNA were subjected to one- or twocycle<br />
amplification to investigate the influence <strong>of</strong> one-cycle or two-cycle amplification on the<br />
expression pattern. Figure 5.3 depicts that the RNA amplification factor is highly dependent<br />
on the starting amount <strong>of</strong> RNA. Using 10ng <strong>of</strong> starting RNA, two-cycle amplification led to a<br />
12,000- to 14,000-fold increase in RNA yield. With 50ng <strong>of</strong> starting RNA, the same protocol<br />
resulted in 2500- to 3000-fold RNA yield. One-cycle amplification with 5μg <strong>of</strong> starting RNA<br />
led to a 20- to 30-fold increase <strong>of</strong> the RNA yield. Irrespective <strong>of</strong> the amount <strong>of</strong> starting RNA<br />
or the amplification protocol employed, the RNA yield was comparable (Figure 5.3).<br />
106
Chapter 5: Manuscript III<br />
Figure 5.3: Comparison <strong>of</strong> total RNA yields after either one-cycle (OC) amplification with 5µg <strong>of</strong><br />
starting RNA or two-cycle (TC) amplifications with 10ng or 50ng <strong>of</strong> starting material,<br />
respectively. Data represent mean yield ±SD (n = 6 replicates).<br />
The latter observation was tested with a Bonferroni's multiple comparison test which<br />
suggested that no significant differences can be observed. This further suggests a<br />
saturation <strong>of</strong> the amplification process, an artifact that could lead to preferential<br />
amplification <strong>of</strong> certain genes and consequently in the generation <strong>of</strong> aberrant expression<br />
pr<strong>of</strong>iles. To test this hypothesis, the expression pr<strong>of</strong>iles <strong>of</strong> above samples were analyzed<br />
using Affymetrix Rat Genome RAE230_2.0 arrays. The heat maps in figure 5.4 depict the<br />
ratios <strong>of</strong> gene expression calculated by dividing the expression values in the single AAtreated<br />
replicate samples by the mean expression value <strong>of</strong> the corresponding time-matched<br />
control replicates.<br />
Most <strong>of</strong> the genes selected with both the one-cycle protocol and the two-cycle protocol<br />
starting with 10ng were similarly upregulated after one- and two-cycle amplification. Only<br />
ca. 20% <strong>of</strong> the genes were qualitatively differently expressed in the two protocols, i.e.,<br />
approximately 80% <strong>of</strong> the genes were found to be significantly and similarly deregulated<br />
following one-cycle and two-cycle amplification. Furthermore, a preliminary analysis <strong>of</strong><br />
specific genes that were deregulated following AA treatment demonstrated that the same<br />
main biological pathways were identified in the two amplification protocols thus suggesting<br />
that no preferential/aberrant gene expression is observed due to differing amplification<br />
protocols (Stemmer et al., in preparation).<br />
107
Chapter 5: Manuscript III<br />
Figure 5.4: Expression pr<strong>of</strong>iles <strong>of</strong> differentially expressed genes. Genes that were assigned to be<br />
differentially expressed after one (OC)- or Two (TC)- cycle amplification <strong>of</strong> the<br />
indicated amount <strong>of</strong> RNA from rat kidney after treatment with Aristolochic acid for 7<br />
days were subjected to one-dimensional cluster analysis using Euclidian as distance<br />
metric. The heat maps show ratios <strong>of</strong> gene expression calculated by dividing the<br />
expression values in the single treated replicate samples through the mean<br />
expression value <strong>of</strong> all corresponding time-matched control replicates. The left<br />
diagram depicts the genes commonly selected with the TC- (10ng) and the OC-<br />
protocol, the diagram in the middle those specifically selected with the TC (10ng)<br />
protocol, and the diagram on the right those specifically selected with the OC<br />
protocol. The red color illustrates up-regulated and green down-regulated genes<br />
always in comparison to the corresponding control, as indicated by the color scale.<br />
II: Reproducibility <strong>of</strong> RNA amplification and microarray hybridization<br />
Absolute expression pr<strong>of</strong>iles <strong>of</strong> all genes on a microarray may be influenced by the reagent<br />
lots used. Therefore, hybridization <strong>of</strong> cRNA prepared at different dates with different<br />
reagent lots was carried out at varying dates. This was done with RNA from the short-term<br />
AA-treated and vehicle control samples described above and whose separation in the<br />
three-dimensional space (PCA) is illustrated in figure 5.5.<br />
108
Chapter 5: Manuscript III<br />
Figure 5.5.: Principal components analysis based on the expression <strong>of</strong> all genes <strong>of</strong> the RAE230A<br />
subset (15923 probe sets) on the RAE230_2.0 chip. Black and grey dots depict kidney<br />
samples <strong>of</strong> control and <strong>of</strong> aristolochic acid-treated rats, respectively. Ellipses indicate<br />
samples processed with identical protocols. (A) Different amounts <strong>of</strong> the same RNA<br />
sample subjected to One-cycle (OC) or Two-cycle (TC) amplification. (B) 10 ng <strong>of</strong> the<br />
same RNA sample subjected to TC amplification and chip hybridization at two<br />
different dates (Feb05, Oct05). (C) 10 ng <strong>of</strong> the same RNA sample subjected to TC<br />
amplification at two different dates (Feb05, Oct05), yet hybridized on the same day.<br />
The principle components after the one- and two-cycle protocols separate from each other<br />
only in one dimension, whereas the principle components <strong>of</strong> treated and control samples<br />
separate in another dimensions namely the amplification procedure dimension and the<br />
treatment dimension. The 20% <strong>of</strong> the total deregulated genes observed that differed<br />
between the two amplification protocols may have been responsible for the shift in the<br />
dimension observed in the PCA (Figure 5.5, A). From these observations it can be<br />
concluded that genes deregulated between treated and control samples can be detected<br />
reliably, as reported above, but that microarrays from treated samples must be compared<br />
with control microarrays employing the identical amplification and hybridization protocol.<br />
Preparation <strong>of</strong> cRNA at different dates using separate reagent kits leads to a separation <strong>of</strong><br />
the overall expression pr<strong>of</strong>ile in PCA space (Figure 5.5, B). This appears mainly due to the<br />
separate cRNA synthesis and not as a result <strong>of</strong> microarray hybridization at different dates,<br />
an interpretation that is corroborated by the PCA in figure 5.5, C, for which two separate<br />
cRNA preparations were hybridized on the same day, yielding a very similar dimensional<br />
separation as observed for the PCA in figure 5.5, B. Despite the influence <strong>of</strong> amplification<br />
protocols and possibly hybridization, pr<strong>of</strong>ile separation between treated and control is<br />
always maintained. The conclusion <strong>of</strong> the above findings thus would be that in order to<br />
select differentially expressed genes due to a given treatment, samples <strong>of</strong> the control and<br />
treatment groups must be prepared using the identical protocol and within the same<br />
protocol handling run.<br />
109
5.4.4 Performance <strong>of</strong> the established protocol<br />
Chapter 5: Manuscript III<br />
To test the reliability <strong>of</strong> above protocol RNA was isolated from microdissected samples <strong>of</strong><br />
control and AA exposed male Eker rats (n=3) <strong>of</strong> the 6-month experiment for Affymetrix chip<br />
hybridization. Three different types <strong>of</strong> renal lesions, namely basophilic atypical tubules<br />
(bAT) basophilic hyperplasia (bHA) and chronic progressive nephropathy (CPN), as well as<br />
healthy tissue (HT) <strong>of</strong> the same rats were excised via LMPC. RNA (5ng) was isolated from<br />
each sample type (lesion type or HT) and analyzed on Affymetrix RAE230_2.0 chips (one<br />
chip per lesion type or HT and animal). Amplification factors were comparable for excised<br />
sample types. A box plot, displaying the absolute expression values after scaling to a mean<br />
target expression value <strong>of</strong> 100 per microarray, was used as a quality check <strong>of</strong> the<br />
hybridized chips (Figure 5.6). No outliers or significant variations in chip hybridizations were<br />
found. The data thus demonstrated that the procedure developed generated reproducible<br />
and reliable gene expression determinations.<br />
Figure 5.6. Boxplot analyis <strong>of</strong> Affymetrix genechip data obtained from microdissected renal<br />
samples. 5 ng kidney RNA isolated from various sample types <strong>of</strong> rats (n=3) treated<br />
with AA for 6 months. Sample types: healthy tissue (HT), basophilic atypical tubuli<br />
(bAT), basophilic hyperplasia (bHA), and chronic progressive nephropathy (CPN).<br />
Isolated RNA was subjected to two-cycle amplification and Affymetrix RAE230_2.0<br />
chip hybridization. After scaling to a target intensity <strong>of</strong> 100, the expression intensity<br />
distribution <strong>of</strong> each Affymetrix chip, corresponding to a given sampletype and animal,<br />
was visualized in a boxplot format. The box itself is limited by lower and upper<br />
hinges, which correspond to the 25% and 75% quantiles, respectively.<br />
110
5.5 Discussion<br />
Chapter 5: Manuscript III<br />
Gradual degradation <strong>of</strong> RNA can introduce a variable bias in the quantification and<br />
amplification <strong>of</strong> gene transcripts. This degradation, especially during a protocol<br />
encompassing various processing steps, can severely affect the results <strong>of</strong> downstream<br />
applications, e.g., gene expression pr<strong>of</strong>iling. It is therefore crucial to protect the RNA from<br />
degradation in order to obtain reliable gene expression results. RNA integrity numbers can<br />
be used as standard to define an individual minimum threshold <strong>of</strong> RNA quality. For the<br />
study presented here a RIN number threshold <strong>of</strong> 7.0 was used. This threshold was chosen<br />
based on recommendations by the manufacturer Agilent, who also provides the RINcalculating<br />
s<strong>of</strong>tware (Schroeder et al. 2006). The comparison <strong>of</strong> different preservation and<br />
staining techniques demonstrated that RNA with the required high quality as well as optimal<br />
conservation <strong>of</strong> tissue morphology can be obtained from cryosections <strong>of</strong> snap-frozen tissue<br />
samples that were stained with H&E. The latter findings, however, stand in contrast to<br />
previously published data by Ellis et al. (Ellis et al. 2002) where RNAlater-fixed and PBSwashed<br />
pancreas samples provided reasonably good tissue morphology. Thus caution is<br />
advised as to generalizations <strong>of</strong> fixation and processing protocols, while emphasizing that<br />
protocol optimization procedures, as presented here, are advisable for every animal and<br />
organ prior to embarking on expression pr<strong>of</strong>ile analyses <strong>of</strong> microdissected tissue samples.<br />
From all parameters tested, processing <strong>of</strong> cryosections at room temperature had the most<br />
adverse effect on RNA quality, emphasizing that in order to conserve high-quality RNA<br />
from stained cryosections, 3h <strong>of</strong> processing should not be exceeded. In addition to<br />
maintenance <strong>of</strong> RNA quality, standardization <strong>of</strong> RNA amplification and chip hybridization is<br />
<strong>of</strong> utmost importance. Indeed, optimization <strong>of</strong> amplification protocols also entailed starting<br />
with comparable RNA quantities in order to avoid saturation effects, as demonstrated here.<br />
Due to the tubular structure <strong>of</strong> the kidney, microdissected areas <strong>of</strong> comparable size (mm 2 )<br />
can vary in RNA yields. A minimum area <strong>of</strong> 2mm 2 should be microdissected for reliable<br />
RNA quantification and subsequent two-cycle amplification. In addition, it was<br />
demonstrated that samples should be prepared with an identical protocol during the same<br />
time frame and quality analysis by box plot analysis, to ensure reliable gene expression<br />
data. In conclusion, a protocol for laser-assisted microdissection in conjunction with<br />
Affymetrix microarray technology was established that allows investigation <strong>of</strong> rat kidney<br />
gene expression pr<strong>of</strong>iles with high reproducibility and reliability (Figure 5.6). This protocol is<br />
described in detail in Appendix A.<br />
111
Acknowledgments<br />
Chapter 5: Manuscript III<br />
We would like to thank the Federal Ministry <strong>of</strong> Education and Research (BMBF: 0313024)<br />
for funding the project, T. Lampertsdoerfer and G. von Scheven for their skillful assistance<br />
during the whole animal experiment, E. O'Brien and A.H. Heussner for help with the animal<br />
sacrifice.<br />
5.6 Appendix<br />
Recommended protocol for laser-assisted microdissection in conjunction with Affymetrix<br />
microarray technology<br />
1. Collect kidneys from sacrificed animals and section each kidney into ca. 5mm<br />
replicate slices.<br />
2. Wrap kidney slices into sterile (oven baked-RNAse free) aluminium foil and<br />
immediately snap-freeze samples in liquid nitrogen.<br />
3. Use samples directly or store at −80°C.<br />
4. For preparation <strong>of</strong> cryosections, cool cryostat to −20°C and “thaw” the respective<br />
sample from −80°C to cryostat temperature. It is possible to refreeze the samples at<br />
−80°C for later use. However 3 “freeze and thaw” cycles should not be exceeded.<br />
5. Freeze samples onto a small drop <strong>of</strong> OCT embedding media in the cryostat.<br />
6. Prepare 10 μm cryosections and mount onto special RNase-free membrane<br />
covered slides, recommended for the respective microdissection system, e.g., the<br />
LMPC system from PALM Microlaser GmbH, Bernried Germany.<br />
7. Air dry freshly cut sections for 20s in the cryostat and subsequently fix them in<br />
−20°C cold ethanol (70% in RNase-free ddH2O) for 3min.<br />
8. Air dry sections for 10min at RT.<br />
9. Stain sections with RNase-free, ice-cold Mayer's hematoxylin for 3min.<br />
10. Rinse with RNase-free tap water for 3min.<br />
11. Counterstain with RNase-free, ice-cold eosin Y for 3min.<br />
12. Remove excess staining solution on absorbent surface and dehydrate section by<br />
shortly dipping into ice-cold 70% and 100% ethanol.<br />
13. Air-dry sections at RT and directly use for laser-assisted microdissection<br />
employing, e.g., the PALM LMPC system (PALM Microlaser GmbH, Bernried<br />
Germany).<br />
112
Chapter 5: Manuscript III<br />
14. For RNA isolation from microdissected kidney samples pool a minimum area <strong>of</strong><br />
2mm 2 and collect samples into an appropriate sampling cup <strong>of</strong> the respective<br />
microdissection system.<br />
15. After completion <strong>of</strong> the sample collection, immediately lyse pooled samples via<br />
addition <strong>of</strong> 300μl lysis buffer containing 1% β-mercaptoethanol (e.g., component <strong>of</strong><br />
the RNeasy Micro Kit, Qiagen, Hilden, Germany).<br />
16. Do not exceed 3h between staining and lysis <strong>of</strong> pooled samples.<br />
17. Isolate RNA using e.g. RNeasyMicro Kit (Qiagen, Hilden, Germany) according to<br />
the manufacturer's instruction.<br />
18. Analyze total RNA quality and estimate quantity, using an Agilent Bioanalyzer. For<br />
later amplification and hybridization procedures, use only RNA samples with a RIN<br />
numbers >7.0.<br />
19. Amplify RNA and convert to biotin-labeled cRNA using e.g. the Affymetrix Twocycle<br />
Target labeling kit according to the manufacturer’s instructions. Always use<br />
same starting quantity <strong>of</strong> RNA, e.g., 5ng or 10ng RNA.<br />
20. Use 15μg biotin-labeled cRNA samples for Affymetrix chip hybridization and<br />
process the hybridizations according to the manufacturer's instruction.<br />
113
Chapter 6: Manuscript IV<br />
Earliest preneoplastic renal lesions provide an in-depth<br />
understanding <strong>of</strong> renal carcinogenesis<br />
Kerstin Stemmer 1 , Heidrun Ellinger-Ziegelbauer 2 , Hans-J. Ahr 2 and Daniel R. Dietrich 1 *<br />
1<br />
Human and Environmental Toxicology, University <strong>of</strong> Konstanz, Konstanz, Germany<br />
2<br />
<strong>Molecular</strong> and Special Toxicology, Bayer Healthcare AG, Wuppertal, Germany<br />
* Corresponding author: Daniel R. Dietrich: daniel.dietrich@uni-konstanz.de<br />
American Journal <strong>of</strong> Pathology, submitted<br />
6.1 Abstract<br />
Although the mTOR pathway appears critical both in human and rat renal tumors, it<br />
remains to be established whether both TORC1 and TORC2 pathways are involved in the<br />
genesis and/or progression <strong>of</strong> renal preneoplastic lesions to tumors. Eker rats<br />
heterozygous for a mutation in the TSC2 tumor suppressor gene, thus predisposed for the<br />
early development <strong>of</strong> renal lesions with morphologic similarities with human renal cancer,<br />
were used to analyze the specific involvement <strong>of</strong> TORC1 and TORC2 in the carcinogenand<br />
sex-dependent formation and progression <strong>of</strong> preneoplastic lesions.<br />
Our data demonstrate activation <strong>of</strong> both, TORC1 and TORC2 dependent mTOR pathways<br />
and comparable gene expression pr<strong>of</strong>iles in laser-microdissected preneoplastic lesions<br />
from genotoxic (aristolochic acid) and non-genotoxic (ochratoxin A) treated and control<br />
Eker rats. Co-localised immunostaining <strong>of</strong> phospho-ribosomal protein S6, phospho-Akt<br />
(Ser473) and cytoplasmatic FOXO1 confirmed TORC1 and TORC2 activation, thus<br />
corroborating disruption <strong>of</strong> the mTOR-AKT feedback inhibition in preneoplastic lesions.<br />
Moreover, the here presented findings strongly suggest that the compound-dependent<br />
formation <strong>of</strong> preneoplastic lesions is limited to a critical period <strong>of</strong> time, while lesion<br />
progression appears compound-independent. Therefore, preneoplastic lesion initiation<br />
seems to be influenced by multiple events, whereas clonal expansion appears primarily<br />
dependent on the deregulation <strong>of</strong> the AKT-TSC2-TORC1 and TORC2-AKT pathways.<br />
114
Chapter 6: Manuscript IV<br />
Together these findings indicate that continuous dosing with carcinogens may not have an<br />
augmenting influence exciting lesions. In addition, further in-depth understanding <strong>of</strong> the<br />
TORC1 and TORC2 balance in preneoplastic lesions may be critical for improved renal<br />
cancer treatment strategies.<br />
6.2 Introduction<br />
The mammalian target <strong>of</strong> rapamycin (mTOR) pathway is upregulated in several human<br />
cancers (Albanell et al. 2007), amongst which are renal cell carcinomas (Pantuck et al.<br />
2006; Radulovic and Bjelogrlic 2007). Under non-pathological conditions this pathway is<br />
tightly controlled by the tumor suppressor complex tuberous sclerosis 2 (TSC2, tuberin)<br />
and its obligate binding partner, TSC1 (hamartin) (Jozwiak 2006). Inactivating mutations <strong>of</strong><br />
TSC2 were shown to be accompanied by Rheb mediated activation <strong>of</strong> the raptor containing<br />
mTOR complex 1 (TORC1) and its effectors p70S6K, S6 ribosomal protein (S6RP), 4E-<br />
BP1 and eIF4G (Kenerson et al. 2002; Mak and Yeung 2004). As a consequence, TORC1<br />
dependent protein translation is activated, affecting several cell growth regulators, e.g.<br />
stabilization <strong>of</strong> the transcription factor hypoxia-inducible-growth-factor α (HIF1α)<br />
(Brugarolas et al. 2003), which in turn can drive diverse processes involved in<br />
tumorigenesis, including cell growth, angiogenesis and glycolysis (Semenza 2002).<br />
Moreover, the antiproliferative capacity <strong>of</strong> TSC2 was shown to be inhibited by the c-Myc<br />
oncogene in vitro (Rosner et al. 2003; Ravitz et al. 2007), demonstrating the broad<br />
dimension <strong>of</strong> the TSC2 tumor suppressor network. To date, it remains to be established,<br />
whether TSC2 is also involved in the regulation <strong>of</strong> rictor containing mTOR complex 2<br />
(TORC2). This complex was previously shown to function as an important regulator <strong>of</strong> the<br />
cytoskeleton through members <strong>of</strong> the Rho small GTPase family and phosphorylation <strong>of</strong><br />
Protein Kinase Cα (PKCα) (Jacinto et al. 2004; Sarbassov et al. 2004), and to<br />
phosphorylate AKT at Ser473, one <strong>of</strong> the key residues, required for full AKT activation<br />
(Sarbassov et al. 2005). AKT regulates numerous signal transduction processes involved in<br />
cell survival and proliferation, one being direct phosphorylation and inactivation <strong>of</strong> tumor<br />
suppressor FOXO1 (Manning and Cantley 2007).<br />
Although the mTOR pathway was described to be altered in human renal cancer (Pantuck<br />
et al. 2006) and functional loss <strong>of</strong> TSC2 was found in spontaneous and chemically-induced<br />
rat renal tumors (Walker et al. 1992; Kubo et al. 1994; Kenerson et al. 2002), it remains to<br />
be established whether the TORC1 and TORC2 pathways are involved in the formation<br />
115
Chapter 6: Manuscript IV<br />
and/or progression <strong>of</strong> renal preneoplastic lesions to tumors. Furthermore the influence <strong>of</strong><br />
genotoxic and non-genotoxic carcinogens on these pathways remains unknown.<br />
To characterize the specific involvement <strong>of</strong> TORC1 and TORC2 in the carcinogen- and<br />
sex-dependent formation and progression <strong>of</strong> preneoplastic lesions, Eker rats carrying a<br />
heterozygous loss-<strong>of</strong>-function mutation in the TSC2 gene, therefore predisposed for the<br />
early development <strong>of</strong> renal lesions with morphologic similarities with human renal cancer<br />
(Yeung et al. 1995; Hino 2004), appeared an ideal model. Therefore, male and female Eker<br />
rats were treated for 3 and 6 months with low doses <strong>of</strong> the genotoxic aristolochic acid (AA)<br />
(Mengs et al. 1982; Arlt et al. 2002) and the non-genotoxic carcinogen ochratoxin A (OTA),<br />
for which a 10-fold higher tumor incidence was found in male than in female F344 rats<br />
(Boorman et al. 1992; O'Brien and Dietrich 2005). Both are associated with the occurrence<br />
<strong>of</strong> human urothelial tumors and nephropathy in Balkan countries, suggesting a direct link<br />
between human and rat renal carcinogenesis (Stoev 1998; Grollman et al. 2007).<br />
Compound-induced non-neoplastic and neoplastic renal pathology, site-specific renal cell<br />
proliferation, incidence, phenotype and progression stage <strong>of</strong> the preneoplastic lesions, and<br />
the rate <strong>of</strong> preneoplastic lesion progression were determined at the 3- and 6-month timepoint<br />
in both sexes. For the first time, laser-microdissected preneoplastic lesions and<br />
healthy tubules <strong>of</strong> AA and OTA treated as well as control male Eker rats were analyzed<br />
using microarrays, allowing to investigate gene expression pr<strong>of</strong>iles specifically induced by<br />
AA and OTA and to differentiate pathways specific for the progression stage <strong>of</strong><br />
preneoplastic lesions. Activation <strong>of</strong> TORC1 and TORC2 pathways were visualized via<br />
immunohistochemical detection <strong>of</strong> phosphorylated downstream targets.<br />
6.3 Material and Methods<br />
6.3.1 Compounds<br />
Aristolochic acid sodium salt mixture (AA I: 41% and AA II: 56%) was purchased from<br />
Sigma Aldrich, Germany. Ochratoxin A (>98% purity) was kindly provided by M.E. Stack,<br />
US FDA, Washington DC.<br />
6.3.2 Animals<br />
Eker rats were purchased at 6-8 weeks <strong>of</strong> age from the University <strong>of</strong> Texas MD Anderson<br />
<strong>Cancer</strong> Center, Smithville, USA. Groups <strong>of</strong> females and males were randomly assigned to<br />
116
Chapter 6: Manuscript IV<br />
dose groups (10 rats per compound (or vehicle) and time-point) and acclimatized for 4<br />
weeks. Groups <strong>of</strong> 10 Eker rats per sex were gavaged with OTA (210µg/kg BW) or AA<br />
(1mg/kg BW), dissolved in 0.1M NaHCO3 at five days a week. Time-matched vehicle<br />
controls were gavaged with 0.1M NaHCO3. 5 <strong>of</strong> 10 rats per dose group and sex were s.c.<br />
implanted with osmotic ALZET-pumps (Model 2ML1, Charles River Laboratories, Germany)<br />
containing 5-bromo-2-deoxyuridine (BrdU, 20mg/ml sterile saline, Sigma Aldrich, Germany)<br />
5 days prior to sacrifice to allow post-mortem immunohistochemical analysis <strong>of</strong> cell<br />
proliferation. After 3 and 6 months <strong>of</strong> treatment, anesthetized rats were sacrificed by<br />
exsanguination subsequent to retrograde perfusion with PBS and kidneys were collected<br />
(Supplementary figure 6.1).<br />
6.3.3 Sample collection<br />
One half <strong>of</strong> the freshly isolated kidney was cross-sectioned to 5mm slices, slices snap<br />
frozen in liquid nitrogen and stored at -80°C for subsequent cryosectioning. The other half<br />
was fixed in PBS buffered fixative (2% paraformaldehyde and 1% glutaraldehyde) for<br />
subsequent paraffin embedding and sectioning.<br />
6.3.4 Histopathology<br />
H&E stained sections from 10 rats per group were randomized for histopathological<br />
evaluation. Non-neoplastic pathology was classified as none (0), mild (1), moderate (2),<br />
strong (3), and severe (4), while preneoplastic and neoplastic lesions were classified<br />
according to the lesion type (Dietrich and Swenberg 1991) and numbers <strong>of</strong> each lesion type<br />
counted.<br />
6.3.5 Immunohistochemistry<br />
Cell proliferation was evaluated via BrdU-immunohistochemistry. Immunostaining was<br />
carried out as described previously (Stemmer et al. 2007), yet using a monoclonal mouse<br />
anti-BrdU primary antibody (MU247-UC, Biogenex, USA) diluted 1:100 in Power BlockTM<br />
(BioGenex, USA) in an overnight application at 4°C. Cell proliferation was quantified on<br />
randomized sections. 20 microscopic fields (10x ocular, 40x objective) were randomly<br />
chosen within the area <strong>of</strong> the renal outer cortex and inner cortex/outer medulla. Proximal<br />
and distal tubules and collecting ducts were counted separately per field, distinguishing<br />
117
Chapter 6: Manuscript IV<br />
between negative and positive BrdU-stained nuclei. Nuclear labeling indices for BrdU (LI%)<br />
(BrdU positive nuclei/total number <strong>of</strong> nuclei counted) were determined. Subsequently, the<br />
LIs <strong>of</strong> proximal tubules without pathological changes were determined. For LI(%)determination,<br />
at least 1.000 and 500 nuclei were counted in proximal and distal tubules<br />
and collecting ducts, respectively.<br />
Immunohistochemistry <strong>of</strong> phospho-S6 ribosomal protein (pS6RP), phospho-Akt (pAKT) and<br />
FOXO1 in paraffin sections was carried out according to manufacturers’ instructions (Cell<br />
Signaling Technology). Primary rabbit anti-pS6RP (Ser235/236), (Cat No: 2211) was<br />
diluted 1:100, while rabbit anti-p-AKT (Ser473) (Cat No: 3787), was diluted 1:10 and rabbit<br />
anti-FOXO1 (Abcam ab39656) 1:200 in 5% normal goat serum. Antigen–antibody<br />
complexes were visualized using the Super SensitiveTM (BioGenex, USA) alkaline<br />
phosphatase-labeled, biotin–streptavidin amplified detection system and Fast Red as<br />
chromogen.<br />
6.3.6 Laser microdissection and RNA isolation<br />
For microdissection <strong>of</strong> preneoplastic lesions from H&E stained renal cryosections, a laser<br />
microdissection and pressure catapulting (LMPC) system (ZEISS, Germany) was used.<br />
Atypical tubule (bAT), basophilic atypical hyperplasia (bAH) or healthy tubules (HT) were<br />
micodissected separately (Stemmer et al. 2006) from each <strong>of</strong> three replicate rats per dose<br />
group. bAT, bAH and HT from each individual animal were pooled. RNA isolation from<br />
pooled samples and subsequent Affymetrix Rat Genome RAE_230A_2.0 chip hybridization<br />
was carried out as previously described (Stemmer et al. 2006).<br />
6.3.7 Microarray data processing and analysis<br />
Microarray quality control was performed as described previously (Ellinger-Ziegelbauer et<br />
al. 2005) and gene expression data were submitted to the NCBI-GEO repository (GSE<br />
10608). For statistical analysis Expressionist Analyst s<strong>of</strong>tware (Genedata AG, Switzerland)<br />
was used. Significantly deregulated genes per compound and lesion type vs. healthy<br />
tubules <strong>of</strong> control rats were selected by student’s T-test with a p-value cut-<strong>of</strong>f <strong>of</strong> 0.005,<br />
combined with a 2.0-fold deregulation threshold. Identical cut-<strong>of</strong>fs were applied for<br />
comparing healthy tubules <strong>of</strong> AA and OTA treated rats with healthy tubules <strong>of</strong> control rats.<br />
Heatmaps were used to graphically display the relative expression data after one-<br />
118
Chapter 6: Manuscript IV<br />
dimensional clustering <strong>of</strong> the genes. See supplemental information for details on the<br />
validation <strong>of</strong> microarray data.<br />
For functional analysis, significantly deregulated genes were characterized according to the<br />
biochemical role <strong>of</strong> its encoded protein, using information from databases, e.g. NetAffx,<br />
Swissprot, Proteome and Pubmed. Depending on the direction <strong>of</strong> deregulation, genes were<br />
further assigned to pathophysiological categories allowing the comparison <strong>of</strong> deregulated<br />
pathways in different types <strong>of</strong> lesions and healthy tubules.<br />
6.3.8 Statistical analysis<br />
Statistical analysis <strong>of</strong> histopathological and cell proliferation data were carried out using<br />
GraphPad Prism 4® (USA). Significant differences in nuclear labeling indices or total<br />
number <strong>of</strong> lesions in treated and control rats were analyzed by Bonferroni and Dunnett’s<br />
test for multiple comparisons. Lesion incidences were tested for significance using the 2sided<br />
Fisher’s exact test. Ranked non-neoplastic pathology data were analyzed using the<br />
nonparametric Mann-Whitney test.<br />
6.4 Results<br />
6.4.1 Non-neoplastic pathology<br />
Treatment <strong>of</strong> male and female Eker rats with AA or OTA resulted in increased nonneoplastic<br />
renal pathology compared to respective control rats (Supplementary table 6.1).<br />
OTA treatment significantly and time-dependently increased chronic progressive<br />
nephropathy (CPN), as characterized by a thickened basal membrane, an increased<br />
regeneration and monocytes infiltration in both sexes (Supplementary figure 6.2). In<br />
agreement with previously described OTA-induced non-neoplastic pathology (Boorman et<br />
al. 1992; Rasonyi et al. 1999), male and female Eker rats treated with OTA for 6 months<br />
presented with increased apoptosis, necrosis, karyomegally, prevalence <strong>of</strong> dilated tubule<br />
and protein casts, as well as an increased tubular regeneration and inflammatory response<br />
primarily within the inner cortex. In contrast, treatment with AA resulted in weaker pathology<br />
compared to OTA treated rats. A significantly enhanced severity <strong>of</strong> CPN and inflammation<br />
and an increased prevalence <strong>of</strong> dilated tubule and protein casts in male rats were observed<br />
only at the 6-month time-point. Moreover, AA treated females generally presented with<br />
weaker pathology than males.<br />
119
6.4.2 Cell proliferation<br />
Chapter 6: Manuscript IV<br />
Site-specific cell proliferation (Supplementary figure 6.2, C-F), i.e. specific to proximal<br />
tubule, distal tubule or collecting ducts, was assessed via BrdU immunostaining (Table<br />
6.1). Female Eker rats at the 6-month time-point showed an approximately 2-3-fold higher<br />
basal cell proliferation rate than males at all sites, whereas this was not apparent at the 3month<br />
time-point.<br />
Table 6.1: Comparison <strong>of</strong> BrdU S-Phase labeling indices (LI%) <strong>of</strong> Eker rats treated with AA or<br />
OTA for 3 and 6 months, respectively. LI%, determined for proximal tubules (PT)<br />
within randomly chosen fields <strong>of</strong> the outer or inner cortex, <strong>of</strong> healthy proximal tubule<br />
(hPT), distal tubule (DT) and collecting ducts (CD). For PT >1000 nuclei were counted,<br />
for DT and CD >500 nuclei were counted. LI% were tested for significance using oneway<br />
ANOVA and Bonferroni's Post Test for multiple comparisons.<br />
Male Eker rats Female Eker rats<br />
3 months treatment Control AA OTA Control AA OTA<br />
# animals 5 5 5 5 5 5<br />
PT outer cortex 2.4 + 0.9 2.1 + 0.7 1.9 + 0.8 1.8 + 0.4 3.3 + 2.1 2.5 + 0.8<br />
PT inner cortex 2.0 + 1.0 2.2 + 1.0 12.3 + 3.5 b,d 2.2 + 0.3 4.8 + 1.5 12.6 + 4.0 b,d<br />
Male Eker rats Female Eker rats<br />
6 months treatment Control AA OTA Control AA OTA<br />
# animals 5 5 5 5 3 5<br />
PT outer cortex 2.0 + 0.4 4.1 + 2.1 4.9 + 0.7 3.7 + 1.2 6.7 + 0.6 4.8 + 0.8<br />
PT inner cortex 2.7 + 1.3 6.6 + 2.8 a 15.6 + 2.0 b,d 4.9 + 1.8 12.2 + 0.8 b,c,e 14.8 + 3.3 b,d<br />
hPT, outer cortex 1.9 + 0.4 2.3 + 0.6 2.6 + 0.6 4.4 + 0.7 5.3 + 0.3 3.9 + 0.6<br />
hPT, inner cortex 1.9 + 0.5 1.9 + 0.4 2.8 + 0.9 4.9 + 1.1 6.3 + 1.3 4.9 + 1.9<br />
DT 1.9 + 1.1 0.9 + 0.5 1.2 + 0.4 4.5 + 2.5 5.3 + 0.8 4.2 + 1.5<br />
CD 1.5 + 1.7 0.9 + 0.5 1.9 + 2.0 4.4 + 1.9 5.4 + 0.2 5.8 + 2.0<br />
a b<br />
Significantly higher than the respective control group (p< 0.05), Significantly higher than the respective<br />
control group (p< 0.001), c Significantly higher in inner, than outer cortex, within the same treatment group<br />
(p< 0.05), d Significantly higher in inner, than outer cortex, within the same treatment group (p< 0.001), e<br />
Significantly higher in 6 months than 3 months treated rats (same sex, compound and tubular segment)<br />
(p< 0.001)<br />
OTA: 3 months OTA-treatment significantly increased cell proliferation rates 6.2-fold and<br />
5.7-fold above controls in the proximal tubules <strong>of</strong> the inner cortex <strong>of</strong> male and female rats,<br />
respectively. In contrast, a higher cell proliferation was observed following 6 months OTAtreatment<br />
in the inner cortex <strong>of</strong> male rats (5.8-fold) compared to females (3.0-fold).<br />
120
Chapter 6: Manuscript IV<br />
AA: 3 months AA-treatment did not change cell proliferation rates in the proximal tubules <strong>of</strong><br />
the inner or the outer cortex in both male and female rats. However, at the 6-month timepoint<br />
a 2.4- and 2.5-fold increase in cell proliferation in the proximal tubules <strong>of</strong> the inner<br />
cortex was observed in both males and females, respectively (Table 6.1).<br />
When restricting cell proliferation analysis to proximal tubules with a normal (nonpathologic)<br />
phenotype (Supplementary figure 6.2, A), mean cell proliferation rates were not<br />
significantly different in the outer or inner cortex <strong>of</strong> control and treated rats <strong>of</strong> both sexes<br />
(Table 6.1). No increased cell proliferation was observed in proximal tubules <strong>of</strong> the outer<br />
cortex, distal tubule or collecting ducts <strong>of</strong> either sex, irrespective <strong>of</strong> the treatment.<br />
6.4.3 (Pre-) neoplastic pathology<br />
Preneoplastic (atypical tubule (AT) and atypical hyperplasia (AH)) and neoplastic (adenoma<br />
and carcinoma) lesions observed in male and female Eker rats after 3 or 6 months <strong>of</strong><br />
treatment were primarily <strong>of</strong> the basophilic phenotype i.e. bAT (Figure 6.1, A) and bAH<br />
(Figure 6.1,B). Other lesion types, e.g. oncocytic tubules and cysts, were rare and thus not<br />
evaluated further. Despite a 100% incidence <strong>of</strong> bATs in all treatment and control groups<br />
(Figure 6.1, C and D, Supplementary table 6.2), a time- and compound-dependent increase<br />
<strong>of</strong> total numbers <strong>of</strong> bATs and bAHs was observable in treated males and females, when<br />
compared to the respective controls at both time points.<br />
OTA: 3 months OTA-treatment induced a 3.3-fold and a 2.8-fold increase in total bATs in<br />
male and female rats, respectively. A similar increase in bATs was also observed at the 6month<br />
time-point, when compared to the respective controls, suggesting continued OTAinduced<br />
bAT formation in all groups. In males, total numbers <strong>of</strong> bAH increased approx. 3fold<br />
over the time, while the total number <strong>of</strong> bAH in OTA-treated females remained at<br />
similar levels at both time-points, indicating a higher formation <strong>of</strong> preneoplastic lesions in<br />
male rats. This was confirmed by the 80% bAH incidence in male rats compared to the<br />
30% incidence in females (Figure 6.1, D).<br />
AA: A significant 3.0- and 5.6-fold increase <strong>of</strong> the bATs/rat was observed in female rats at<br />
the 3-and 6-month time-points, while male Eker rats showed a significant 3.5 fold increase<br />
<strong>of</strong> bATs at the 6-month time-point only (Figure 6.1, D). Similarly, both male and female rats<br />
presented with a higher total number <strong>of</strong> bAH following 3 and 6-months AA-treatment when<br />
compared to the corresponding controls. However, the bAH/rat ratio was significantly<br />
higher at the 6-month time-point only in female rats.<br />
121
Chapter 6: Manuscript IV<br />
Figure 6.1. Representative preneoplastic lesions as observed in H&E stained renal sections (A) bAT; (B) bAH. Mean number <strong>of</strong> preneoplastic lesions in<br />
AA- and OTA-treated male and female control Eker rats after 3 (C) and 6 months (D). Grey bars: bAT; white bars: bAH. Dunnett’s test: *p < 0.05;<br />
**p < 0.01.<br />
122
6.4.4 Gene expression pr<strong>of</strong>iling<br />
Chapter 6: Manuscript IV<br />
As preneoplastic lesions exhibited a similar basophilic phenotype in both sexes, gene<br />
expression pr<strong>of</strong>ile analyses were restricted to preneoplastic lesions <strong>of</strong> male Eker rats.<br />
Laser microdissected preneoplastic lesions (bAT and bAH) and healthy tubules <strong>of</strong> AA- and<br />
OTA-treated and control males were analyzed using microarrays in order to investigate<br />
gene expression pr<strong>of</strong>iles induced by AA and OTA as well as to differentiate pathways<br />
specific for the progression stage <strong>of</strong> the preneoplastic lesions.<br />
Preneoplastic lesions: Gene expression pr<strong>of</strong>iling <strong>of</strong> microdissected lesions resulted in<br />
570 significantly deregulated annotated and non-redundant genes when compared to<br />
healthy tubules <strong>of</strong> control rats. Although different total numbers <strong>of</strong> genes appeared to be<br />
significantly deregulated in different samples (Table 6.2), visualization <strong>of</strong> expression<br />
pr<strong>of</strong>iles <strong>of</strong> the union <strong>of</strong> selected genes revealed a qualitatively similar expression pr<strong>of</strong>ile in<br />
all preneoplastic lesions, irrespective <strong>of</strong> treatment or lesion type (Figure 6.2).<br />
Figure 6.2: Heatmap comparing gene expression changes <strong>of</strong> microdissected bATs and bAHs <strong>of</strong><br />
AA, OTA and vehicle treated Eker rats, and microdissected healthy tubule from AA<br />
and OTA treated rats. Expression pr<strong>of</strong>iles were compared to microdissected healthy<br />
tubules (HT) <strong>of</strong> control rats. Gene expression ratios are indicated by the color scale:<br />
Red: upregulated; Green: downregulated.<br />
123
Chapter 6: Manuscript IV<br />
Healthy tubules (HT): Gene expression pr<strong>of</strong>iles <strong>of</strong> microdissected morphologically<br />
“healthy” tubules (Supplementary figure 6.2, A) <strong>of</strong> AA- and OTA-treated rats were<br />
compared with HT <strong>of</strong> the respective control rats HT(C) to evaluate the effect <strong>of</strong> AA and<br />
OTA treatment on gene expression in HT after 6 months treatment. Seven and six genes<br />
were significantly deregulated in HT <strong>of</strong> AA and OTA treated males, respectively <strong>of</strong> which<br />
the most important ones are listed (Supplementary table 6.3).<br />
Functional analysis <strong>of</strong> significantly deregulated genes: Functional analysis <strong>of</strong> genes<br />
deregulated in preneoplastic lesions <strong>of</strong> treated and control rats identified a number <strong>of</strong><br />
pathophysiologically relevant endpoints. Table 6.2 specifies the criteria used to assign<br />
deregulated genes to these categories. The most prominent categories, containing the<br />
majority <strong>of</strong> deregulated genes, were similar in all lesion types. Moreover, many deregulated<br />
genes <strong>of</strong> these categories could be directly associated with activated AKT, TORC1 and/or<br />
TORC2 pathways (Supplementary table 6.3 and below), and may result from a functional<br />
loss <strong>of</strong> the TSC2 gene in preneoplastic lesions (Figure 6.3, C), i.e. genes involved in: Cell<br />
cycle progression (upregulation <strong>of</strong> CDC2, CCNB1, CCNB2, AURKB; downregulation <strong>of</strong><br />
NUP50); cell growth, survival and proliferation (upregulation <strong>of</strong> CTGF, POSTN, ITGB4,<br />
PINK, IGFBP1 and PIK3AP1; downregulation <strong>of</strong> FOXOA1); protein synthesis (upregulation<br />
<strong>of</strong> RPS15A); upregulation <strong>of</strong> HIF1α target genes involved in angiogenesis (PAI1, CXCR4)<br />
and glucose uptake and glycolysis (PFKM, PFKC, ENO2, PGAM2 and GLUT2);<br />
cytoskeleton rearrangement (upregulation <strong>of</strong> genes coding for Rac/Rho/Cdc42 signal<br />
transduction proteins e.g. RHOQ, RAC2, PAK3, IQGAP1, CDC42SE1, CDC42EP1,<br />
ELMO2, DOCK6, RAC2 and PLEKHG2; upregulation <strong>of</strong> PKCα substrates MARCKS and<br />
AKAP12)<br />
Moreover, a direct correlation between functional loss <strong>of</strong> TSC2 in preneoplastic lesions and<br />
active c-myc oncogene is indicated by upregulation <strong>of</strong> C-MYC itself and its target genes<br />
EMP1 and EMP3, which may point to additional mutations up- or down-stream <strong>of</strong> the<br />
TSC2/TSC1 complex (Figure 6.3, A and C).<br />
124
Table 6.2: Numbers and categories <strong>of</strong> significantly deregulated genes in preneoplastic lesions<br />
Endpoints % <strong>of</strong> all significantly deregulated<br />
annotated genes per group<br />
bAT bAT bAT bAH bAH bAH<br />
C AA OTA C AA OTA<br />
Description<br />
Chapter 6: Manuscript IV<br />
Oxidative stress 2.9 2.7 4.7 1.9 1.4 2.4 Up regulation <strong>of</strong> ARE target genes and other genes known to be involved in oxidative stress responses. Down<br />
regulation <strong>of</strong> genes mediating the cellular antioxidant defence<br />
DNA damage<br />
response<br />
4.4 0.9 1.6 1.1 4.1 0 Up regulation <strong>of</strong> genes encoding p53 target genes and genes coding for proteins involved in DNA damage repair<br />
Apoptosis 4.3 2.4 1.6 3.4 1.4 1.2 Up regulation <strong>of</strong> pro-apoptotic genes, down regulation <strong>of</strong> anti-apoptotic genes<br />
Tumour- and<br />
metastasis<br />
suppression<br />
Cell cycle<br />
progression<br />
Cell growth/<br />
survival/<br />
proliferation<br />
2.9 3.0 4.7 2.3 2.7 3.5 Up regulation <strong>of</strong> genes coding for tumour suppressor genes or genes. Up regulation <strong>of</strong> tissue inhibitor <strong>of</strong><br />
metalloproteinases or down regulation <strong>of</strong> genes facilitating metastasis<br />
2.9 5.1 5.5 3.1 5.5 3.5 Up regulation <strong>of</strong> genes involved in DNA replication cell cycle progression, mitotic spindle/ cytokineses or<br />
nucleosome formation.<br />
10.1 8.3 8.6 6.9 6.8 10.6 Up regulation <strong>of</strong> genes encoding growth and survival factors, (IGF)-PI3K-AKT and the mTOR/S6K pathway<br />
components, anti-apoptotic genes, oncogenes. Down regulation <strong>of</strong> tumor suppressor genes.<br />
Tumourigenesis 1.4 0.9 1.6 1.1 2.7 1.2 Deregulation <strong>of</strong> genes with unclear biochemical function, however for which a similar direction <strong>of</strong> deregulation has<br />
been reported previously in different types <strong>of</strong> cancer.<br />
Cytoskeleton<br />
rearrangement<br />
Enhanced cell<br />
adhesion<br />
7.2 7.1 7.8 10.7 6.8 7.1 Up regulation <strong>of</strong> genes coding for Rho-GTPases, other proteins involved in positive regulation <strong>of</strong> actin<br />
polymerization, actin remodelling, stress fibre formation or microtubule dynamics.<br />
1.4 1.2 0.8 0.8 4.1 2.4 Up regulation <strong>of</strong> genes coding for cell adhesion molecules<br />
Angiogenesis 2.9 4.5 5.5 1.9 4.1 4.7 Up regulation <strong>of</strong> genes <strong>of</strong> the VEGF pathway and functioning in growth and survival <strong>of</strong> endothelial cells, vascular<br />
development and remodelling.<br />
125
Endpoints % <strong>of</strong> all significantly deregulated<br />
annotated genes per group<br />
bAT bAT bAT bAH bAH bAH<br />
C AA OTA C AA OTA<br />
Description<br />
Chapter 6: Manuscript IV<br />
Metastasis 1.4 2.7 2.3 3.4 2.7 1.2 Up regulation <strong>of</strong> genes coding for serin proteinases involved in ECM degradation, down regulation <strong>of</strong> genes coding<br />
for cell adhesion molecules, or gap junction proteins.<br />
Dedifferentiation 0 6.5 3.9 7.7 0 5.9 Down regulation <strong>of</strong> genes mediating the physiological homeostasis <strong>of</strong> the cell and/ or organ. Down regulation <strong>of</strong><br />
genes involved in biotransformation<br />
Stromatogenesis 2.9 4.5 6.3 7.7 5.5 12.9 Up regulation genes encoding ECM components and components involved in stroma-cell interaction like integrins or<br />
integrin binding proteins.<br />
Immune response 10.1 15.5 15.6 19.2 15.1 15.3 Up regulation <strong>of</strong> genes encoding components <strong>of</strong> the acute phase response, inflammation, innate or adaptive<br />
immunity, complement system, TNF/ cytokine pathway, or up regulation genes coding for proteins involved in<br />
antigen presentation.<br />
Lysosomal<br />
autophagy<br />
1.4 2.7 2.3 1.9 1.4 2.4 Up regulation <strong>of</strong> genes coding for lysosomal enzymes and proteins <strong>of</strong> the lysosomal membrane<br />
Energy provision 5.8 4.5 2.3 2.7 4.1 1.2 Up regulation <strong>of</strong> genes coding for proteins supporting uptake <strong>of</strong> glucose, lipids and amino acids into the cell, as well<br />
as for proteins supporting glycogenolysis, glycolysis, gluconeogenesis, citrate cycle, and glutaminolysis.<br />
Precursor provision 5.8 8.0 8.6 8.8 5.5 5.9 Up regulation <strong>of</strong> genes coding for proteins involved in fatty acid synthesis, lipid storage, cholesterol uptake,<br />
cholesterol metabolism and phospholipid / glycolipid synthesis. Down regulation <strong>of</strong> genes involved in ß-oxidation<br />
fatty acids and amino acid degradation<br />
Protein synthesis<br />
and trafficing<br />
4.3 1.2 3.1 3.4 2.7 3.5 Up regulation <strong>of</strong> genes involved in protein sorting, vesicular transport or endo-/exocytosis. Up regulation <strong>of</strong> genes<br />
involved in translation or protein folding in the endoplasmatic reticulum<br />
Protein degradation 4.3 3.3 0 0.4 0 0 Up regulation <strong>of</strong> genes coding for components <strong>of</strong> the ubiquitin-dependent protein catabolism<br />
Neuronal<br />
differentiation<br />
2.9 3.3 3.1 2.7 5.5 2.4 Up regulation <strong>of</strong> genes involved in neuronal signal transmission or neuronal differentiation<br />
Cellular stress 4.3 2.7 1.6 3.1 8.2 3.5 Up regulation <strong>of</strong> genes involved in the stress kinase-, NFkB-, or HIF pathways pathway or other stress related<br />
pathways involved in hypoxia, acidosis or tubular damage. Down regulation <strong>of</strong> genes encoding inhibitors <strong>of</strong> these<br />
pathways and endpoints<br />
Osmotic stress 7.2 1.5 2.3 1.1 1.4 1.2 Up regulation <strong>of</strong> genes involved in water and sodium resorption, down regulation <strong>of</strong> genes cell mediating volume<br />
homeostasis. Up regulation <strong>of</strong> genes involved in hyperosmotic response.<br />
Unknown endpoint 10.1 7.7 6.3 5.0 8.2 8.2 Genes with known biochemical function, but whose deregulation could not be given a specific meaning, due to the<br />
lack <strong>of</strong> information from literature<br />
126
Chapter 6: Manuscript IV<br />
6.4.5 Confirmation <strong>of</strong> TORC1 and TORC2 pathway activation in<br />
preneoplastic lesions<br />
Above presented gene expression pr<strong>of</strong>iles point to a concomitant activation <strong>of</strong> AKT,<br />
TORC1 and/or TORC2 pathways. To prove this, immunostaining <strong>of</strong> phosphorylated<br />
ribosomal protein S6 at Ser235/236 (pS6RP) was used as an indicator <strong>of</strong> TORC1 activity<br />
(Kenerson et al. 2002). Staining <strong>of</strong> AKT phosphororylated at Ser473 (pAKT) provided<br />
confirmation <strong>of</strong> active TORC2 since active TORC2 was shown to directly phosphorylate<br />
AKT at Ser473 (Sarbassov et al. 2005).<br />
Figure 6.3. Schematic overview over the postulated processes (incl. activated mTOR, cell clone<br />
survival and expansion) involved in the formation <strong>of</strong> preneoplastic lesions. (B)<br />
Postulated time curve <strong>of</strong> tissue adaptation vs. formation <strong>of</strong> preneoplastic lesions. (C)<br />
Suggested mechanistic involvement <strong>of</strong> TORC1 and TORC2 dependent mTOR<br />
pathways involved in renal carcinogenesis. Blue: pathways and processes implicated<br />
by gene expression analysis. Red: protein expression or phosphorylation verified by<br />
immunostaining.<br />
127
Chapter 6: Manuscript IV<br />
Comparable pS6RP staining was observed within bATs and bAHs <strong>of</strong> AA, OTA and control<br />
male and female Eker rats (Figure 6.4, A and B). Sporadic preneoplastic tubules were<br />
negative for pS6RP staining (a maximum <strong>of</strong> 6 bATs in the 6 months AA treatment group).<br />
Positive pS6RP staining was also present in adenomas, carcinomas and oncocytic cysts <strong>of</strong><br />
all rats examined. pS6RP positive immunostaining <strong>of</strong> preneoplastic lesions and tumors may<br />
thus be interpreted as corroborative evidence for activation <strong>of</strong> the TORC1/S6K/S6RP<br />
pathway.<br />
pAKT-immunostaining <strong>of</strong> bAT and bAH and tumors resulted in a weak but specific<br />
cytoplasmic and sporadic nuclear staining in treated and control male rats, suggesting the<br />
involvement <strong>of</strong> TORC2/AKT pathway in renal carcinogenesis (Figure 6.4, C and D).<br />
However, as pAKT staining appeared faint, additional immunostaining <strong>of</strong> AKT target<br />
FOXO1 was employed, since FOXO1 shuttles from the nucleus to cytoplasm for<br />
degradation when phosphorylated by active AKT (Tran et al. 2003). Since, FOXO1<br />
phosphorylation was demonstrated to be strongly reduced in mice lacking TORC2 member<br />
rictor (Jacinto et al. 2006), cytoplasmatic localization <strong>of</strong> FOXO1 can be interpreted as<br />
TORC2 dependent AKT activation. Healthy tubules presented primarily with nuclear<br />
FOXO1 staining. In contrast, strong cytoplasmic and nuclear staining in bATs and bAHs<br />
was detected, as expected for deregulated FOXO1 activation and/or half-life (Figure 6.4, E<br />
and F). Thus, the FOXO1 staining pattern further supports activation <strong>of</strong> the TORC2-AKT<br />
pathway in preneoplastic renal lesions.<br />
128
Chapter 6: Manuscript IV<br />
Figure 6.4: Pro<strong>of</strong> <strong>of</strong> principle staining for concurrent TORC1 and TORC2 activation in<br />
preneoplastic lesions. Serial sectioned bAT with positive pS6RP (A), pAKT (C) and<br />
cytoplasmatic accumulation <strong>of</strong> FOXO1 (E). Representative images <strong>of</strong> bAH<br />
immunopositive for pS6RP (B), p-AKT (D), and FOXO1 (F). Additional images <strong>of</strong><br />
pS6RP, p-AKT and FOXO1 are shown in Figure S3<br />
129
6.5 Discussion<br />
Chapter 6: Manuscript IV<br />
6.5.1 Compound-induced tubular damage and regenerative cell<br />
proliferation is critical for the formation but not for the progression <strong>of</strong><br />
preneoplastic lesions<br />
Increased cell proliferation can result in carcinogenesis, most likely by the fixation <strong>of</strong><br />
mutations as well as by survival and promotion <strong>of</strong> initiated cells (Barrett and Wiseman<br />
1987; Dietrich and Swenberg 1991). By distinguishing nephron specific effects and<br />
“healthy” from pathologically altered nephron-sections, above data demonstrated that AAand<br />
OTA-induced increase <strong>of</strong> cell proliferation is restricted to pathologically altered<br />
proximal tubules as well as to preneoplastic lesions at the time points evaluated.<br />
If compound induced nephrotoxicity/genotoxicity and accompanying regenerative cell<br />
proliferation is a main contributor to renal carcinogenesis, compound-induced cell<br />
proliferation should correspond to increased numbers and/or progression rates <strong>of</strong><br />
preneoplastic lesions in a compound- and sex-specific manner. Indeed the 2-fold higher<br />
cell proliferation in OTA-treated males than in females at the 6-month time-point<br />
corresponded well with the 1.7-fold higher increase <strong>of</strong> preneoplastic lesions in males than<br />
in females (Figure 6.1, Supplementary table 6.2). Since ratios <strong>of</strong> bAT and bAH numbers<br />
remained unchanged, the 80% incidence <strong>of</strong> bAHs in males compared to the 30% in<br />
females (Figure 6.1, D) suggest an earlier onset <strong>of</strong> preneoplastic lesion development in<br />
males and thus points to a sex-dependent difference in OTA-induced formation <strong>of</strong><br />
preneoplastic lesions rather than to an OTA-mediated lesion progression. The sexdependent<br />
difference in formation <strong>of</strong> preneoplastic lesions, as reported here, may result<br />
from a higher OTA-induced oxidative stress and subsequent DNA damage in male rats<br />
(Kamp et al. 2005). This interpretation is supported by the finding that 14-days oral<br />
treatment <strong>of</strong> Eker and wild type female rats with OTA resulted in an upregulation <strong>of</strong> the 8oxoguanine<br />
DNA glycosylase (an enzyme critically involved in the repair <strong>of</strong> oxidative DNA<br />
damage) than in the corresponding males (Supplementary figure 6.4). Moreover, catalase<br />
expression was reduced in OTA-treated male Eker and wild type rats, while females<br />
showed an increased expression.<br />
The critical role <strong>of</strong> compound induced nephrotoxicity/genotoxicity and accompanying cell<br />
proliferation for the onset <strong>of</strong> preneoplastic lesion formation is corroborated by the findings in<br />
the AA-treatment groups. Indeed, the 2-fold higher cell proliferation in female than in male<br />
rats after 3 months <strong>of</strong> AA-treatment corresponded to the 2-fold higher number <strong>of</strong><br />
preneoplastic lesions in females at the same time point. Moreover, a similar sex-dependent<br />
130
Chapter 6: Manuscript IV<br />
trend for cell proliferation and preneoplastic lesions, although not as pronounced, was also<br />
observed at the 6 months time-point (Figure 6.1, Supplemetary figure 6.2, Table 6.1 and<br />
Supplemetary table 6.2). As noted for OTA-treated rats, the comparable bAT to bAH ratio in<br />
male and female AA-treated rats is suggestive <strong>of</strong> only a limited influence <strong>of</strong> AA on bAT to<br />
bAH progression. The sex-dependent difference in the formation <strong>of</strong> AA-induced lesions<br />
may result from the previously reported 2-fold higher levels <strong>of</strong> dG-AAI and dA-AAII DNA<br />
adducts in kidneys <strong>of</strong> female than male Wistar rats following treatment with oral doses <strong>of</strong><br />
0.15mg/kg body weight for 3 months (Arlt et al. 2001). Given a comparable cellular uptake<br />
and metabolism <strong>of</strong> AA in Eker and Wistar rats, the higher number <strong>of</strong> DNA adducts and<br />
regenerative cell proliferation observed in female rats may have led to a higher fixation <strong>of</strong><br />
mutations and therefore explain the observed corresponding formation <strong>of</strong> preneoplastic<br />
lesions.<br />
6.5.2 Time dependent tissue adaptation to carcinogen treatment<br />
Gene expression analysis <strong>of</strong> preneoplastic lesions demonstrated that bATs and bAHs <strong>of</strong><br />
control, AA- and OTA-treated male rats were qualitatively indistinguishable (Figure 6.2),<br />
thus supporting the above assumption that once initiated, in a compound- and sexdependent<br />
manner, clonal expansion <strong>of</strong> preneoplastic lesions is compound-independent.<br />
Moreover, while 14-days AA- and OTA- treatment <strong>of</strong> Eker and corresponding wild type rats<br />
resulted in substantial gene expression changes in the renal cortex, e.g. response to<br />
oxidative stress and DNA damage (Stemmer et al. 2007), only marginal gene expression<br />
changes were observed in healthy tubules <strong>of</strong> male rats following 6 months AA- and OTAtreatment<br />
(Figure 6.2). This suggests that part <strong>of</strong> the renal cells may have the potential to<br />
adapt to compound treatment, thus further supporting the observation that compounds are<br />
important for the formation but not the clonal expansion <strong>of</strong> preneoplastic lesions. These<br />
findings thus imply the existence <strong>of</strong> a critical, albeit compound-specific, period <strong>of</strong> exposure<br />
for genotoxic, or non-genotoxic carcinogens (Figure 6.3, B), where normal renal tissue is<br />
not yet adapted. In this critical period normal renal tissue appears incapable <strong>of</strong> sufficiently<br />
repairing compound-induced damage despite active regenerative cell proliferation, thus<br />
providing the possibility for fixation <strong>of</strong> mutations and thereby formation <strong>of</strong> preneoplastic<br />
lesions.<br />
131
Chapter 6: Manuscript IV<br />
6.5.3 Numbers <strong>of</strong> earliest preneoplastic lesions formed are decisive for<br />
later occurrence <strong>of</strong> veritable tumors<br />
The existence <strong>of</strong> a critical period <strong>of</strong> carcinogen exposure would imply, that extended<br />
compound exposure may add little to the formation <strong>of</strong> additional preneoplastic lesions<br />
(Figure 6.3, B), thus suggesting that the incidence and number <strong>of</strong> veritable tumors is<br />
primarily driven by the number <strong>of</strong> preneoplastic lesions formed during this critical period.<br />
Indeed, the incidences <strong>of</strong> preneoplastic lesions (bAH) observed in the Eker rat study (80%<br />
in males vs. 30% in females) correspond well to the tumor incidences observed at the end<br />
<strong>of</strong> the OTA 2-year bioassay (87.2% in male versus 22.9% in female F344 rats) at the same<br />
dose (Boorman et al. 1992). This could imply that the incidence and number <strong>of</strong> veritable<br />
tumors is primarily driven by the number <strong>of</strong> preneoplastic lesions initially formed during a<br />
critical period <strong>of</strong> carcinogen exposure (see above). This interpretation is supported by study<br />
<strong>of</strong> Wolf and co-workers who reported an indistinguishable number <strong>of</strong> preneoplastic lesions<br />
and tumors in Eker rats treated with 500 ppm sodium barbital (NaBB) in feed for 4.5<br />
months and sacrificed at 12 months and Eker rats treated with 500 ppm NaBB in feed for<br />
the whole 12 months (Wolf et al. 2000). The latter phenomenon is not restricted to Eker<br />
rats, as confirmed by a previous study (Dietrich and Swenberg 1991), where male NBR<br />
rats, deficient for α2u-globulin and male F344 were initiated with the genotoxic N-ethyl-N-<br />
hydroxyethylnitrosamine (EHEN) and/or promoted with α2u-globulin binding d-limonene for<br />
6 months. D-limonene exposure was demonstrated to induce concentration-dependent<br />
regenerative cell proliferation in F344 but not in the α2u-globulin deficient NBR rats. Both<br />
EHEN and d-limonene treatment resulted in the formation <strong>of</strong> preneoplastic lesions (AT and<br />
AH) in F344 rats in nearly comparable numbers but not in the formation <strong>of</strong> veritable tumors.<br />
Similarly, treatment <strong>of</strong> NBR rats with EHEN or EHEN and d-limonene led to comparable<br />
numbers <strong>of</strong> ATs and AHs but no tumors. In contrast, treatment <strong>of</strong> F344 rats with a<br />
combination <strong>of</strong> EHEN and d-limonene led to highest number <strong>of</strong> ATs and AHs as well as to<br />
the formation <strong>of</strong> veritable tumors.<br />
The latter data thus strongly support the observation that the number <strong>of</strong> preneoplastic<br />
lesions initially formed, resulting from genotoxicity, fixation <strong>of</strong> spontaneous mutations via a<br />
mitogenic stimulus or regenerative cell proliferation or a combination there<strong>of</strong> (Figure 6.3, A<br />
and B), are decisive for the later development <strong>of</strong> renal tumors in rats.<br />
132
Chapter 6: Manuscript IV<br />
6.5.4 The balance <strong>of</strong> TORC1 and TORC2 pathways determines<br />
preneoplastic lesion progression<br />
Since carcinogen treatment may mainly affect early tumor formation in a compound- and<br />
sex-specific manner, similar gene expression pr<strong>of</strong>iles in preneoplastic lesions <strong>of</strong> AA, OTA<br />
and control Eker rats may point to a central pathway involved in lesion progression.<br />
Functional analysis <strong>of</strong> deregulated genes in bATs and bAHS as well as<br />
immunohistochemical staining <strong>of</strong> pS6RP, pAKT and FOXO1 in bATs and bAHs <strong>of</strong> control,<br />
AA- and OTA-treated Eker rats clearly demonstrated that both mTOR pathways (TORC1<br />
and TORC2 dependant) were activated. The latter may result from a functional loss <strong>of</strong> the<br />
TSC2 gene in preneoplastic lesions either via loss-<strong>of</strong> heterozygosity (Yeung et al. 1995) or<br />
additional mutations up- or down-stream <strong>of</strong> the TSC2/TSC1 complex e.g. the c-myc<br />
oncogene (Figure 6.3, A and C). Eker rat preneoplastic lesion analysis demonstrated an<br />
upregulation <strong>of</strong> several HIF1α �target genes (Supplemetary table 6.3) and numerous<br />
additional genes involved in downstream responses <strong>of</strong> HIF1α e.g. angiogenesis and<br />
energy provision (Table 6.2 and S3). The latter findings correspond well to the role <strong>of</strong><br />
activated TORC1 shown to regulate several cell growth regulators, including stabilization <strong>of</strong><br />
the transcription factor HIF1α (Brugarolas et al. 2003), thereby providing for sufficient<br />
oxygen and energy supply <strong>of</strong> proliferating cells, e.g. via angiogenesis and glycolysis<br />
(Semenza 2002). Under physiological conditions, activated S6K leads to phosphorylation<br />
and degradation <strong>of</strong> insulin receptor substrate proteins (Harrington et al. 2004; Shah et al.<br />
2004) and consequently to reduced downstream PI3K-AKT-TSC1/2 activation in<br />
mammalian cells. Since intact S6K dependent negative feedback inhibition and limited<br />
oncogenic PI3K-AKT signaling was still present in growth-limited benign hamartomas<br />
(Kwiatkowski 2003; Manning et al. 2005), continued and uncontrolled activation <strong>of</strong> both<br />
S6K and AKT could be a critical trigger in the progression <strong>of</strong> benign to rapidly growing<br />
malignant lesions.<br />
The concomitant activation <strong>of</strong> TORC1 and TORC2 (Figure 6.3, C and 6.4) and therefore<br />
likely disturbance <strong>of</strong> feedback inhibition <strong>of</strong> mTOR-AKT signaling in renal preneoplastic<br />
lesions, as shown here, appears important for survival and clonal expansion <strong>of</strong><br />
preneoplastic lesions. Facilitated AKT activation due to TORC2 dependent Ser473<br />
phosphorylation (Bellacosa et al. 2005), would not only further enhance the AKT-TSC2-<br />
TORC1 pathway, but also influence other targets <strong>of</strong> the oncogenic AKT leading to reduced<br />
apoptosis and cell cycle progression (Table 6.2 and S3).<br />
Moreover, preneoplastic lesion gene expression pr<strong>of</strong>iles support the previously shown<br />
function <strong>of</strong> TORC2 as an important regulator <strong>of</strong> the cytoskeleton through members <strong>of</strong> the<br />
133
Chapter 6: Manuscript IV<br />
Rho small GTPase family and phosphorylation <strong>of</strong> Protein Kinase Cα (PKCα) (Jacinto et al.<br />
2004; Sarbassov et al. 2004), which were discussed to play a role in cell proliferation,<br />
transformation and metastasis (Jaffe and Hall 2002). TORC2 dependent AKT<br />
phosphorylation and upregulation <strong>of</strong> members <strong>of</strong> the Rho small GTPase family however<br />
also suggest that TORC2 may be important initial formation <strong>of</strong> preneoplastic lesions.<br />
Indeed, Kenerson and co-workers (Kenerson et al. 2005), demonstrated that rapamycin<br />
inhibited renal preneoplastic progression to veritable renal tumors in Eker rats but not the<br />
formation <strong>of</strong> earliest preneoplastic lesions, suggesting that the balance between rapamycin<br />
sensitive TORC1 and the insensitive TORC2 may be critical for preneoplastic lesion<br />
survival and clonal expansion. Rapamycin treatment may inhibit only the TORC1/HIF1α<br />
dependent precursor-, energy- and oxygen supply, necessary for rapid clonal expansion <strong>of</strong><br />
larger preneoplastic renal lesions to veritable tumors, while TORC2/AKT dependent cell<br />
cycle progression and apoptosis inhibition may be important for the survival and<br />
proliferation <strong>of</strong> initiated cells and earliest preneoplastic lesions.<br />
6.5.5 Outlook<br />
In view <strong>of</strong> the current interest in the use <strong>of</strong> rapamycin, and analogous inhibitors <strong>of</strong> the<br />
mTOR pathway for treatment <strong>of</strong> human renal cancer and the close analogy <strong>of</strong> rat and<br />
human renal tumors, the results presented here suggest that TORC1 inhibitors (e.g.<br />
rapamycin analogues) would reduce the rate <strong>of</strong> preneoplastic to neoplastic progression but<br />
not the origin <strong>of</strong> the disease. Concurrent treatment with rapamycin analogues and a<br />
specific TORC2 inhibitor thus holds the promise that growth and survival <strong>of</strong> both the initial<br />
preneoplastic lesions and advanced progression stages could be specifically affected.<br />
Acknowledgements<br />
We would like to thank the Federal Ministry <strong>of</strong> Education and Research for funding the<br />
project (BMBF: 0313024), T. Lampertsdörfer, G. von Scheven, A. Heussner and E. O’Brien<br />
for help with the animal experiment, M. Thiel and K. Lotz for microarray hybridization and P.<br />
Pfluger for critically reading the manuscript.<br />
134
6.6 Supplemental Material<br />
6.6.1 Supplementary experimental procedures: Real time PCR<br />
Chapter 6: Manuscript IV<br />
To evaluate gene expression by real time PCR, total RNA was extracted from frozen renal<br />
cortex <strong>of</strong> 14-days OTA treated and control male and female Eker and wild type rats (n=3)<br />
using RNEasy Mini Kit (Qiagen), according to manufacturer’s instructions. Following DNase<br />
treatment, reverse transcriptions were performed using SuperScript III (Invitrogen) and<br />
random primers (Invitrogen). Real-time PCRs <strong>of</strong> the house keeping gene 18S RNA and <strong>of</strong><br />
candidate genes 8-oxoguanine DNA glycosylase (OGG1) and catalase (CAT) were<br />
performed on Biorad iCycler using iQ SybrGreen Supermix (Biorad). All samples were<br />
performed in technical duplicates. Relative amounts <strong>of</strong> gene-specific mRNAs were<br />
calculated via Ct values, based on a standard curve <strong>of</strong> five points with known amounts <strong>of</strong><br />
template cDNA.<br />
135
6.6.2 Supplementary Figures<br />
Chapter 6: Manuscript IV<br />
Supplementary figure 6.1: Overview over the experimental design: The experiments were<br />
performed according to the German Animal Protection Law, approved by the relevant<br />
German authority, the Regierungspräsidium in Freiburg, Germany (registry number:<br />
G-03)<br />
136
Chapter 6: Manuscript IV<br />
Supplementary figure 6.2: H&E stained paraffin sections: (A) proximal tubules with a normal (nonpathologic)<br />
phenotype (healthy tubules; HT); (B) chronic progressive nephropathy;<br />
(C-F) representative BrdU staining patterns <strong>of</strong> different tubular sections: Red nuclei:<br />
BrdU positive, blue nuclei: haematoxylin counterstaining; (C) proximal tubules <strong>of</strong> the<br />
outer renal cortex (P1/P2 segments); (D) proximal tubules <strong>of</strong> the inner renal cortex (P3<br />
segments); (E) distal tubules; and (F) collecting duct.<br />
137
Chapter 6: Manuscript IV<br />
Supplementary figure 6.3; Representative immunmohistochemical staining <strong>of</strong> renal paraffin<br />
sections demonstrating positive pS6RP (A), pAKT (B) and cytoplasm accumulation <strong>of</strong><br />
Foxo1 (C) demonstrate the concurrent activation <strong>of</strong> both mTOR pathways.<br />
Representative images <strong>of</strong> bAHs positively with antibodies against pS6RP (D), pAKT<br />
(E) and Foxo1 (F).<br />
138
Chapter 6: Manuscript IV<br />
Supplementary figure 6.4: Gene expression analysis <strong>of</strong> OGG1 and CAT in male and female Eker<br />
and wild type rats following 14-days treatment with ochratoxin A (Stemmer et al.<br />
2007). Data <strong>of</strong> three replicate animals are shown as mean + SD. Newman-Keuls<br />
Multiple Comparison Test was performed to test for significance: * < 0.05, ** < 0.01.<br />
139
Supplemental Tables<br />
Chapter 6: Manuscript IV<br />
Supplementary table 6.1: Non-neoplastic renal pathology <strong>of</strong> Eker rats treated with AA or OTA for<br />
three and six months respectively. Histopathological changes were ranked from none<br />
(0) to severe (4) including intermediate classes. Values are presented as median ±<br />
median absolute deviation.<br />
Male Eker rats 3 months 6 months<br />
Control AA OTA Control AA OTA<br />
Necrosis 0.0 + 0.0 0.0 + 0.1 0.0 + 0.2 0.0 + 0.0 0.0 + 0.2 0.8 + 0.6 b<br />
Apoptosis 0.0 + 0.0 0.0 + 0.1 1.0 + 0.2 a 0.0 + 0.0 0.0 + 0.2 0.3 + 0.4 a<br />
Karyomegaly 0.0 + 0.0 0.0 + 0.2 2.0 + 0.5 a 0.0 + 0.0 0.0 + 0.1 1.8 + 0.5 c<br />
Vacuolization 0.0 + 0.0 0.0 + 1.0 0.0 + 0.4 0.0 + 0.0 0.3 + 0.7 1.5 + 0.9 b<br />
Cell shedding 1.0 + 0.6 1.0 + 0.3 1.3 + 0.5 1.0 + 0.6 1.0 + 0.6 0.5 + 0.4<br />
Protein casts 0.3 + 0.6 0.5 + 0.4 1.0 + 0.6 1.0 + 0.5 1.8 + 0.7 1.3 + 0.9<br />
Tubular dilatation 0.5 + 0.5 0.5 + 0.3 1.0 + 0.6 0.5 + 1.0 1.0 + 0.4 1.0 + 0.8<br />
Calcium Casts 0.0 + 0.4 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0 0.0 + 0.0<br />
Regeneration 0.5 + 0.2 0.5 + 0.3 1.0 + 0.3 0.0 + 0.5 1.0 + 0.6 1.0 + 0.5 a<br />
CPN 0.5 + 0.2 0.5 + 0.1 1.0 + 0.3 a 0.8 + 0.5 1.5 + 0.4 a 2.0 + 0.3 b<br />
Inflammation 0.3 + 0.3 0.5 + 0.2 0.5 + 0.3 0.5 + 0.4 1.0 + 0.6 a 1.3 + 0.6 a<br />
Deposit <strong>of</strong> pigments 0.1 + 0.2 0.0 + 0.1 0.0 + 0.3 0.0 + 0.1 0.0 + 0.2 0.3 + 0.4<br />
Female Eker rats 3 months 6 months<br />
Control AA OTA Control AA OTA<br />
Necrosis 0.0 + 0.0 0.0 + 0.1 0.0 + 0.1 0.0 + 0.3 1.0 + 0.6 1.3 + 0.6 b<br />
Apoptosis 0.0 + 0.0 0.0 + 0.1 0.5 + 0.2 c 0.0 + 0.0 0.0 + 0.2 0.0 + 0.2<br />
Karyomegaly 0.0 + 0.1 0.0 + 0.2 1.0 + 0.8 c 0.0 + 0.0 0.0 + 0.3 1.3 + 0.6 c<br />
Vacuolization 0.0 + 0.2 0.0 + 0.0 0.0 + 0.0 2.0 + 0.4 2.3 + 0.9 2.5 + 0.4 b<br />
Cell shedding 1.0 + 0.3 1.0 + 0.4 1.3 + 0.6 1.0 + 0.4 1.0 + 0.5 1.0 + 0.5<br />
Protein casts 0.5 + 0.5 0.5 + 0.3 0.8 + 0.5 0.0 + 0.7 1.8 + 0.9 1.5 + 0.6<br />
Tubular dilatation 1.0 + 0.6 0.3 + 0.7 1.0 + 0.6 0.5 + 0.5 1.0 + 1.0 1.0 + 0.5<br />
Calcium Casts 0.0 + 0.2 0.0 + 0.4 0.0 + 0.1 0.0 + 0.0 0.0 + 0.0 0.0 + 0.1<br />
Regeneration 0.3 + 0.4 0.0 + 0.3 0.0 + 0.4 0.5 + 0.5 1.0 + 0.6 1.0 + 0.5<br />
CPN 0.5 + 0.2 0.5 + 0.3 1.5 + 0.3 c 1.0 + 0.3 1.3 + 0.5 1.5 + 0.3 a<br />
Inflammation 0.5 + 0.2 0.5 + 0.2 0.5 + 0.4 0.5 + 0.3 0.5 + 0.2 1.0 + 0.2 a<br />
Deposit <strong>of</strong> pigments 0.1 + 0.2 0.0 + 0.1 0.0 + 0.2 0.0 + 0.3 0.3 + 0.4 0.0 + 0.2<br />
a b<br />
Significantly higher than the respective control group. Mann-Whitney test (p< 0.05). Significantly higher<br />
than the respective control group. Mann-Whitney test (p< 0.005). c Significantly higher than the respective<br />
control group. Mann-Whitney test (p< 0.001)<br />
140
Chapter 6: Manuscript IV<br />
Supplementary table 6.2: (Pre-) neoplastic pathology <strong>of</strong> the renal cortex <strong>of</strong> male and female Eker<br />
rats treated with AA or OTA for three and six months respectively. Incidence and<br />
numbers <strong>of</strong> bAT, bAH and microscopic (m) and gross (g) tumours (adenoma and<br />
carcinoma) are displayed. Ratios <strong>of</strong> bATs and bAHs and tumours per rat are<br />
presented as mean + standard deviation. Additional ratios <strong>of</strong> mean bAT/rat to mean<br />
bAH/rat are indicative for a bAT to bAH progression.<br />
bAT<br />
bAH<br />
Tumor<br />
bAT<br />
bAH<br />
Tumor<br />
Male Eker rats 3 months 6 months<br />
Control AA OTA Control AA OTA<br />
Incidence 10/10 10/10 10/10 10/10 10/10 10/10<br />
Outer cortex 15 27 32 20 66 42<br />
Cortex/Medulla 20 28 84 30 107 107<br />
Total bAT 35 55 116 50 173 150<br />
bAT / rat 3.5 + 2.0 5.5 + 2.6 11.6+2.0 a 5.0 + 3.3 17.3+5.7 a 15.0+2.2 a<br />
Incidence 4/10 5/10 5/10 5/10 8/10 8/10<br />
Outer cortex 1 1 3 1 16 9<br />
Cortex/Medulla 4 4 6 10 15 17<br />
Total 5 5 9 11 31 26<br />
bAH / rat 0.6 + 0.9 0.5 + 0.5 0.9 + 1.0 1.1 + 1.5 3.1 + 2.8 2.6 + 1.9<br />
Ratio 5.8 9.2 12.9 4.5 5.5 5.7<br />
Incidence (m) 0/10 3/10 1/10 1/10 3/10 2/10<br />
Total (m) 0 3 1 1 3 2<br />
Total (g + m) 1 3 1 1 3 5<br />
Tumour / rat 0.0 + 0.0 0.3 + 0.7 0.1 + 0.3 0.1 + 0.3 0.3 + 0.5 0.2 + 0.4<br />
Female Eker rats 3 months 6 months<br />
Incidence 10/10 10/10 10/10 9/9 8/8 10/10<br />
Outer cortex 16 46 24 16 67 36<br />
Cortex/Medulla 19 61 76 12 63 54<br />
Total bAT 35 107 100 28 130 90<br />
bAT / rat 3.5 + 1.7 10.7+6.2 a 10.0+3.3 a 3.1 + 1.3 16.3+4.8 a 9.0 + 3.8 a<br />
Incidence 1/10 3/10 5/10 2/9 7/8 c 3/10<br />
Outer cortex 0 4 0 2 13 2<br />
Cortex/Medulla 1 2 5 0 6 5<br />
Total 1 6 5 2 19 7<br />
bAH / rat 0.1 + 0.3 0.6 + 1.0 0.5 + 0.5 0.2 + 0.4 2.4 + 1.6 b 0.7 + 1.5<br />
Ratio 35 17.8 20.0 15.5 6.8 12.9<br />
Incidence (m) 3/10 0/10 0/10 1/10 1/10 3/10<br />
Total (m) 3 0 0 1 1 3<br />
Total (g + m) 3 0 0 1 2 4<br />
Tumour / rat 0.3 + 0.5 0.0 + 0.0 0.0 + 0.0 0.1 + 0.3 0.1 + 0.6 0.3 + 0.5<br />
a b<br />
Significantly higher than the respective control group. Dunnett`s test (p< 0.01). Significantly higher than<br />
the respective control group. Dunnett`s test (p< 0.05). c Significantly higher incidence than the respective<br />
control group. Fisher`s exact test (p< 0.05). d Mortality: One female control animal died 2 weeks before<br />
sacrifice. Two female AA treated Eker rats died 10 and 3 weeks before sacrifice.<br />
141
Chapter 6: Manuscript IV<br />
Supplementary table 6.3: Genes differentially deregulated in preneoplastic lesions compared to HT <strong>of</strong> control animals. For the major signalling pathways<br />
(activated (+) or inhibited (-)), genes are listed together with their Genebank accession number and the suspected pathophysiological<br />
consequence <strong>of</strong> deregulation (endpoint). The mean fold deregulation + SD are given for every lesion type. Significant deregulations vs. HT(C)<br />
were tested by Student’s t- test: b p< 0.005; c p< 0.001 which matched the p-value cut-<strong>of</strong>f <strong>of</strong> 0.005 and 2.0 fold deregulation cut-<strong>of</strong>f. (a)<br />
Significantly deregulated (p< 0.05), however not matching the chosen p-value cut<strong>of</strong>f
Accession<br />
number<br />
Protein synthesis NM_053982 RPS15A Ribosomal protein<br />
S15a<br />
Angiogenesis<br />
Glucose uptake<br />
and glycolysis<br />
Cytoskeleton<br />
rearrangement<br />
NM_012620 PAI1 Plasminogen activator<br />
inhibitor-1<br />
AA945737 CXCR4 CXC chemokine<br />
receptor 4<br />
NM_031715 PFKM 6-phosph<strong>of</strong>ructokinase,<br />
muscle (PFK-A)<br />
BM389769 PFKC 6-phosph<strong>of</strong>ructokinase,<br />
type C (PFKP)<br />
AF019973 ENO2 Enolase 2 (Gamma<br />
enolase)<br />
NM_017328 PGAM2 Phosphoglycerate<br />
mutase 2<br />
NM_012879 GLUT2 Glucose transporter<br />
type 2, liver (SLC2A2)<br />
NM_022590 SGLT2 Sodium/glucose<br />
cotransporter2<br />
AA849961 RHOQ Ras homolog gene<br />
family, member Q<br />
NM_019210 PAK3 p21-activated kinase 3<br />
(PAK beta)<br />
BM387072 IQGAP1 IQ motif containing<br />
GTPase activating protein 1<br />
BI275583 CDC42EP1 CDC42 effector<br />
protein<br />
BM389459 CDC42SE1 CDC42 small<br />
effector 2<br />
BM386781 ELMO2 Engulfment and cell<br />
motility protein 2<br />
AI102881 DOCK6 Dedicator <strong>of</strong> cytokinesis<br />
6 (predicted)<br />
Gene title References (Pub Med<br />
ID)<br />
PMID: 18092230<br />
PMID: 16990457<br />
Chapter 6: Manuscript IV<br />
Fold deregulation vs. HT control (mean and p-value <strong>of</strong> 3 replicate animals per<br />
lesion)<br />
1.04+<br />
0.15<br />
PMID: 18381294 1.41+<br />
0.96<br />
PMID: 18375114<br />
PMID: 15802268<br />
1.06+<br />
0.32<br />
PMID: 17661163 1.20+<br />
0.28<br />
PMID: 17661163 1.50+<br />
0.27<br />
PMID: 16227210 1.62+<br />
0.62<br />
PMID: 9688259 0.79+<br />
0.32<br />
PMID: 17203215 2.13+<br />
0.30 (a)<br />
PMID: 17581928 2.58+<br />
0.67<br />
PMID: 12234610<br />
PMID: 15467718<br />
1.03+<br />
0.26<br />
PMID: 18054038 0.65+<br />
0.35<br />
PMID: 15890984<br />
PMID: 15890984<br />
1.60+<br />
0.36<br />
NETAFFX 1.11+<br />
0.13<br />
PMID: 10816584 1.26+<br />
0.04 (a)<br />
PMID: 12879077 1.11+<br />
0.25<br />
PMID: 18291711<br />
PMID: 17196961<br />
1.67+<br />
0.03<br />
1.22+<br />
0.11<br />
4.19+<br />
3.74<br />
0.66+<br />
0.37<br />
1.13+<br />
0.67<br />
1.49+<br />
0.67<br />
1.72+<br />
0.65<br />
5.03+<br />
3.47<br />
2.43+<br />
0.49 (a)<br />
1.52+<br />
0.36<br />
1.07+<br />
0.10<br />
0.86+<br />
0.76<br />
1.11+<br />
0.50<br />
0.71+<br />
0.12<br />
1.06+<br />
0.28<br />
1.12+<br />
0.10<br />
1.48+<br />
0.18 (a)<br />
1.48+<br />
0.45<br />
5.85+<br />
2.45 (a)<br />
2.06+<br />
0.31 b<br />
2.54+<br />
0.33 b<br />
3.30+<br />
0.60 (a)<br />
1.79+<br />
0.61<br />
24.40+<br />
5.76 (a)<br />
2.99+<br />
1.39<br />
3.21+<br />
1.71<br />
1.31+<br />
0.15 (a)<br />
0.91+<br />
0.09<br />
3.89+<br />
0.80 b<br />
2.28+<br />
0.70 a<br />
2.24+<br />
0.39 b<br />
2.01+<br />
0.55 (a)<br />
1.44+<br />
0.22<br />
1.95+<br />
0.15 b<br />
4.70+<br />
2.74<br />
1.79+<br />
0.44 (a)<br />
1.60+<br />
0.16 (a)<br />
4.51+<br />
0.56 b<br />
2.22+<br />
0.30 b<br />
6.02+<br />
2.90<br />
11.25+<br />
1.34 c<br />
4.29+<br />
0.77 b<br />
2.36+<br />
0.24 c<br />
14.85+<br />
3.14 c<br />
3.24+<br />
0.7 (a)<br />
1.89+<br />
0.27 (a)<br />
2.45+<br />
0.22 c<br />
2.01+<br />
0.09 b<br />
1.55+<br />
0.25<br />
1.81+<br />
0.44 (a)<br />
13.72+<br />
1.70 c<br />
1.68+<br />
0.25<br />
0.93+<br />
0.22<br />
2.40+<br />
0.25<br />
2.83+<br />
0.88 (a)<br />
1.68+<br />
1.16<br />
13.43+<br />
3.14 c<br />
2.48+<br />
0.10<br />
1.67+<br />
0.33 (a)<br />
6.20+<br />
2.22 (a)<br />
1.75+<br />
0.20<br />
0.84+<br />
0.08<br />
1.77+<br />
0.13 (a)<br />
1.59 +<br />
0.14 (a)<br />
1.73+<br />
0.14 (a)<br />
1.85+<br />
0.48<br />
7.79+<br />
3.17 b<br />
3.11+<br />
0.71 b<br />
1.68+<br />
0.35 (a)<br />
3.09+<br />
0.71 (a)<br />
2.75+<br />
0.43 b<br />
9.54+<br />
2.61 (a)<br />
6.58+<br />
1.16 c<br />
3.51+<br />
1.94 a<br />
1.82+<br />
0.80<br />
8.28+<br />
9.93<br />
3.32+<br />
0.31 b<br />
2.04+<br />
0.27 b<br />
2.21 +<br />
0.22 c<br />
2.23+<br />
0.19 b<br />
1.68+<br />
0.09<br />
1.76+<br />
0.33 (a)<br />
5.34+<br />
3.53<br />
1.41+<br />
0.20 (a)<br />
1.65+<br />
0.32<br />
3.48+<br />
0.69 (a)<br />
2.74+<br />
0.44 b<br />
5.54+<br />
3.54<br />
6.84+<br />
0.84 c<br />
2.25+<br />
0.07 a<br />
1.72+<br />
0.35 (a)<br />
9.82+<br />
3.36 b<br />
2.76+<br />
0.10 b<br />
1.54+<br />
0.18 (a)<br />
2.11+<br />
0.39 (a)<br />
1.65 +<br />
0.28 (a)<br />
1.46+<br />
0.15<br />
2.29+<br />
0.28 b<br />
16.96+<br />
8.25 b<br />
1.97+<br />
0.47<br />
0.85+<br />
0.13<br />
2.22+<br />
0.26<br />
2.83+<br />
0.47 (a)<br />
0.30+<br />
0.12<br />
11.28+<br />
6.21 b<br />
1.97+<br />
0.26<br />
2.01+<br />
0.45 (a)<br />
4.76+<br />
3.36<br />
1.76+<br />
0.12<br />
0.94+<br />
0.16<br />
1.62+<br />
0.29<br />
1.73 +<br />
0.39 (a)<br />
2.11+<br />
0.11 b<br />
143
C-MYC<br />
Accession<br />
number<br />
AI010476 RAC2 Ras-related C3 botulinum<br />
toxin substrate 2 (p21-Rac2)<br />
BF397719 PLEKHG2 Pleckstrin homology<br />
domain containing G2<br />
NM_030862 MARCKS Myristoylated alaninerich<br />
protein kinase C substrate<br />
BG663107 AKAP12 A kinase anchor<br />
protein 12 (gravin)<br />
NM_012603 CMYC Myc proto-oncogene (cmyc)<br />
BI275741 EMP1 Epithelial membrane<br />
protein 1 (TMP Tumorassociated<br />
membrane protein)<br />
NM_030847 EMP3 Epithelial membrane<br />
protein 3<br />
Gene title References (Pub Med<br />
ID)<br />
PMID: 18230340<br />
PMID: 10843388<br />
Chapter 6: Manuscript IV<br />
Fold deregulation vs. HT control (mean and p-value <strong>of</strong> 3 replicate animals per<br />
lesion)<br />
1.34+<br />
0.39<br />
PMID: 18045877 1.52+<br />
0.21 (a)<br />
PMID: 16109477 1.13+<br />
0.13<br />
PMID: 11282019 1.09+<br />
0.40<br />
PMID: 18056446<br />
PMID: 12894220<br />
PMID: 18056446<br />
PMID: 12894220<br />
PMID: 18056446<br />
PMID: 12894220<br />
1.50+<br />
0.23<br />
1.18+<br />
0.19<br />
1.21+<br />
0.40<br />
1.26+<br />
0.36<br />
1.36+<br />
0.12<br />
0.83+<br />
0.29<br />
2.50+<br />
2.16<br />
1.07+<br />
0.36<br />
1.22+<br />
0.53<br />
0.79+<br />
0.32<br />
2.08+<br />
0.57 a<br />
2.13+<br />
1.08<br />
2.04+<br />
0.48 (a)<br />
2. 77+<br />
1.64<br />
1.87+<br />
0.22 (a)<br />
2.62+<br />
1.03 (a)<br />
2.85+<br />
1.17<br />
1.90+<br />
0.50<br />
1.51+<br />
0.20 (a)<br />
1.37+<br />
0.22<br />
6.82+<br />
0.45 b<br />
2.65+<br />
0.21 (a)<br />
2.08+<br />
0.16 b<br />
2.84+<br />
0.56 b<br />
1.73+<br />
0.32<br />
1.35+<br />
0.14<br />
1.23+<br />
0.14<br />
4.88+<br />
0.16 (a)<br />
3.09+<br />
0.40 b<br />
1.89+<br />
0.40<br />
1.62+<br />
0.10<br />
3.27+<br />
0.50 b<br />
2.25+<br />
0.28 b<br />
2.93+<br />
0.35 b<br />
7.62+<br />
6.33 (a)<br />
2.60+<br />
0.90 (a)<br />
3.38+<br />
0.81 b<br />
2.52+<br />
0.28 b<br />
1.99+<br />
0.58 a<br />
1.53+<br />
0.25 (a)<br />
1.64+<br />
0.38<br />
3.84+<br />
0.33 (a)<br />
2.21+<br />
0.32 (a)<br />
1.92+<br />
0.35 (a)<br />
2.30+<br />
0.50 (a)<br />
2.53+<br />
0.78 a<br />
1.55+<br />
0.30<br />
1.35+<br />
0.10<br />
5.65+<br />
1.63 (a)<br />
3.88+<br />
0.95 b<br />
2.52+<br />
0.50 (a)<br />
1.87+<br />
0.24<br />
144
Chapter 7: Manuscript V<br />
High-fat diet exposure is associated with renal carcinogenesis<br />
in rats<br />
Kerstin Stemmer 1* , Diego Perez-Tilve 2 , Daniel R. Dietrich 1 , Matthias H. Tschöp 2 , and Paul T.<br />
Pfluger 2<br />
1<br />
Human and Environmental Toxicology, University <strong>of</strong> Konstanz, Konstanz, Germany<br />
2<br />
Department <strong>of</strong> Psychiatry, Obesity Research Centre - Genome Research Institute, University<br />
<strong>of</strong> Cincinnati, College <strong>of</strong> Medicine, Cincinnati, Ohio, USA<br />
* Corresponding author: Kerstin Stemmer: kerstin.stemmer@uni-konstanz.de<br />
Manuscript to be submitted<br />
7.1 Abstract<br />
Animal and epidemiologic studies strongly suggest an association <strong>of</strong> renal cancer with dietinduced<br />
obesity. To date, however, it remains unclear whether the pro-carcinogenic<br />
processes in diet-induced obesity are a direct consequence <strong>of</strong> dietary lipds, or from an<br />
increase in adiposity with its co-morbidities and metabolic alterations. To elucidate the<br />
effects <strong>of</strong> high fat diet exposure and body adiposity on renal carcinogenesis, male Wistar<br />
rats were fed for 11 months with high fat diet, or chow. Life-history traits revealed two<br />
different rat subpopulations: one with a morbidly obese phenotype (Diet-Induced Obesity<br />
Sensitive, DIOsens, ca. 1.1kg bw), and one high-fat diet-resistant (DIOres) subgroup with<br />
modestly increased body weight (ca. 650g bw), and local fat depots more reminiscent <strong>of</strong><br />
chow-fed controls (ca. 470g bw).<br />
High fat-diet exposure led to a severe nephropathy, with early stages <strong>of</strong> renal<br />
carcinogenesis, such as atypical regenerative hyperplasia and preneoplasias in kidneys <strong>of</strong><br />
DIOsens and DIOres, and absence there<strong>of</strong> in control rats. In addition, increased cell<br />
proliferation was found in high-fat diet fed animals, and hyperphosphorylation <strong>of</strong> the S6<br />
ribosomal protein suggests an activated mTOR pathway in proliferating tubules in DIOsens<br />
and DIOres rats.<br />
145
Chapter 7: Manuscript V<br />
In conclusion, the similar degree <strong>of</strong> histopathological changes in DIOsens and DIOres rats<br />
indicates that the dietary fat content, rather than the degree <strong>of</strong> adiposity, is the key<br />
etiological factor for renal non-neoplastic and pre-neoplastic pathology. The here presented<br />
findings <strong>of</strong> increased mTOR activation in preneoplastic lesions <strong>of</strong> high fat diet-exposed rats<br />
independently <strong>of</strong> their degree <strong>of</strong> adiposity is consistent with a central role for this pathway<br />
in the pathogenesis <strong>of</strong> obesity induced carcinogenous lesions. With recurring cycles <strong>of</strong><br />
tissue damage, inflammation and regeneration are critical hallmarks <strong>of</strong> kidney cancer,<br />
chronic high-fat diet exposure may propagate the fixation and progression <strong>of</strong> spontaneous<br />
mutations and pre-neoplastic lesions.<br />
7.2 Introduction<br />
Obesity is defined as abnormal or excessive fat accumulation that presents a risk to health.<br />
It can arise from the interaction <strong>of</strong> both genetic factors and environmental conditions and<br />
behavior. A high availability <strong>of</strong> calorie-dense food, paired with a sedentary life style, thereby<br />
leads to a chronically positive energy balance with high caloric intake and low energy<br />
expenditure, and subsequently to the excessive storage <strong>of</strong> fat. It is assumed that<br />
differences in lifestyle, e.g. a lower spontaneous physical activity, or a lower metabolic rate<br />
(Ravussin and Danforth 1999; Ravussin 2005) are important factors influencing the<br />
relationship between high caloric intake body weight gain. Similar to human populations,<br />
inbreed rats also show varying susceptibilities to weight gain after high fat diet exposure<br />
(Levin et al. 1987), and are therefore useful tools to study the effects <strong>of</strong> high caloric or high<br />
fat exposure.<br />
In Western societies, the widespread obesity epidemia is accompanied by a concurrent<br />
increase in many accompanying diseases, such as diabetes, cardiovascular diseases and<br />
cancer (McTiernan 2005; Hotamisligil 2006; Van Gaal et al. 2006; Ogden et al. 2007). A<br />
recent survey <strong>of</strong> all available literature on adiposity and cancer (including epidemiologic,<br />
clinical, and experimental data) by the International Agency for Research on <strong>Cancer</strong><br />
(IARC), provided sufficient evidence that excess body weight increases the risk <strong>of</strong> cancers<br />
<strong>of</strong> the colon, breast (postmenopausal), endometrium, esophagus and the kidney (renal cell<br />
tumors) (IARC 2002a). Specifically, studies <strong>of</strong> populations worldwide suggested a 1.5- to<br />
2.5-fold higher relative risk <strong>of</strong> renal-cell cancer (RCC) for overweight and obese subjects<br />
compared to normal weight controls. Since kidney tumors are <strong>of</strong>ten only detected at late<br />
stages due to the absence <strong>of</strong> early symptoms, overall mortality rates are high. Worldwide,<br />
146
Chapter 7: Manuscript V<br />
kidney cancer with approximately 210.000 cases per year accounts for up to 2% <strong>of</strong> all<br />
malignancies.<br />
There is convincing pre-clinical and clinical evidence that renal cancer is being promoted or<br />
caused by high fat diet induced obesity (Mellemgaard et al. 1995; Chow et al. 2000;<br />
Bergstrom et al. 2001; Renehan et al. 2008), However, it is not clear whether renal cancer<br />
risk is increased by the high body adiposity itself with its manifold co-morbidities and<br />
disease consequences, such as subchronic inflammation, or non-physiological adipokine<br />
secretory patterns. Instead, the overload <strong>of</strong> dietary fat could directly elicit a negative effect<br />
on kidney physiology by direct stimulation <strong>of</strong> cellular signaling pathways through specific<br />
dietary fatty acids.<br />
One prime candidate for such a specific cellular effect could be the mammalian target <strong>of</strong><br />
rapamycin (mTOR) pathway, which plays a key role in both kidney cancer (Kenerson et al.<br />
2002; Albanell et al. 2007; Stemmer et al. 2007; Hanna et al. 2008) and obesity (Thomas<br />
2006). A very recent report suggested that high-fat diet exposure induced the activation <strong>of</strong><br />
the hypothalamic mTOR-S6K signaling pathway in rodents (Ono et al. 2008). Based on<br />
these findings, it was hypothesized in this study that high-fat diet exposure may induces<br />
renal mTOR-S6K activation, and thus may represent a key step in high-fat diet- induced<br />
renal tumorigenesis.<br />
The development <strong>of</strong> kidney tumors is frequently associated with chemically induced renal<br />
α2u-globulin accumulation in male rats, resulting in α2u-globulin nephropathy (Swenberg<br />
1993; Hard 1998; Doi et al. 2007). Since an early report identified α2u-globulin as kidney<br />
fatty acid binding protein (Kimura et al. 1989), it was further hypothesized that α2u-globulin<br />
might bind dietary lipids, might thereby get accumulated in rat kidneys after chronic high-fat<br />
diet exposure, and might thus result in renal pathology either directly or through an<br />
increased lipid accumulation in renal tubules.<br />
To elucidate the roles <strong>of</strong> chronic high-fat diet exposure and body adiposity on renal<br />
carcinogenesis, α2u-globulin accumulation and renal mTOR-S6K signaling, a novel rodent<br />
model <strong>of</strong> diet-induced obesity was employed. Specifically, diet-induced obesity sensitive<br />
(DIOsens) and diet-induced obesity resistant (DIOres) rats were selected from a large<br />
population <strong>of</strong> chronically high-fed diet fed Wistar rats, and life-history traits, plasma<br />
parameters, renal histopathology and cell proliferation were recorded and compared to<br />
those from a group <strong>of</strong> chow-fed animals. Thus, for the first time it was possible to dissect<br />
the direct influence <strong>of</strong> dietary fat vs. the degree <strong>of</strong> body adiposity on renal carcinogenesis.<br />
147
7.3 Material and Methods<br />
7.3.1 Animals<br />
Chapter 7: Manuscript V<br />
32 male Wistar rats (initial BW approx. 250g; Harlan, Indianapolis, IN, USA) were allowed<br />
ad libitum access to a customized high fat diet (Ref.: # D03082706, 40% butter fat, 46%<br />
carbohydrates (corn starch), 0.05% cholesterol, Research Diets, New Brunswick, NJ) for 11<br />
months. From the initial cohort, two subpopulations <strong>of</strong> 1) diet-induced obesity resistant rats<br />
(DR) and 2) diet-induced obesity sensitive rats (DIO) were studied further in comparison to<br />
a group <strong>of</strong> standard chow-fed rats (6% fat, Harlan Teklad LM-485).<br />
All rats were housed in individual cages and maintained on a 12 hr light – dark cycle in a<br />
temperature controlled environment at 23°C. Ad libitum fed animals were sacrificed by<br />
decapitation, and the kidneys rapidly excised. <strong>Kidney</strong>s were cross sectioned and sections<br />
were either frozen on dry ice and kept at -80˚C, or fixed in 10% buffered formalin for a total<br />
<strong>of</strong> 4 months before paraffin embedding. In addition, epididymal fat pads were excised and<br />
weighed. All procedures were approved by the Institutional Animal Care and Use<br />
Committees at the University <strong>of</strong> Cincinnati Office in accordance with the NIH guide for the<br />
care and use <strong>of</strong> laboratory animals.<br />
7.3.2 Blood parameters<br />
Two weeks prior to sacrifice, tail blood was collected from overnight fasted rats,<br />
immediately chilled on ice, and centrifuged at 2500g, 4°C for 10min to extract plasma.<br />
Circulating non-esterified fatty acid, triglyceride and cholesterol levels were determined<br />
using commercially available enzymatic assays (NEFA, Wako Germany; Infinity triglyceride<br />
and Infinity cholesterol, Thermo Electron). Circulating adipokine levels were measured by<br />
using the respective multiplex mouse panels from LINCOplex (St. Charles, MO, USA). All<br />
measurements were performed in duplicate according to the manufacturer’s instructions.<br />
Fed and fasting blood glucose levels were measured directly from tail blood by using a<br />
glucometer (TheraSense Freestyle).<br />
148
7.3.3 Renal Pathology<br />
Chapter 7: Manuscript V<br />
H&E stained paraffin sections were randomized for histopathological evaluation, Nonneoplastic<br />
pathology was classified as none (0), mild (1), moderate (2), strong (3), and<br />
severe (4). Total numbers <strong>of</strong> preneoplastic lesions were counted and depicted as<br />
incidences/section & animal. All pathological examinations were conducted by KS and<br />
independently verified by DRD.<br />
7.3.4 Cell Proliferation analysis<br />
Cell proliferation was evaluated by immunohistochemical staining for proliferating cell<br />
nuclear antigen (PCNA) using monoclonal primary anti-PCNA antibody, in paraffinembedded<br />
kidney sections, as described previously (Stemmer et al. 2007). For details see<br />
table 7.1. PCNA- positive stained S-phase nuclei were quantified on randomized sections.<br />
20 microscopic fields (10x ocular, 40x objective) were randomly chosen within the area <strong>of</strong><br />
the renal outer cortex and inner cortex/outer medulla. A minimum <strong>of</strong> 1000 proximal tubules<br />
were counted separately per field, distinguishing between negative and positive PCNAstained<br />
nuclei. Nuclear labeling indices (LI) were calculated as positive nuclei/total number<br />
<strong>of</strong> nuclei counted.<br />
Table 7.1: Overview <strong>of</strong> antibodies and modifications to the manufacture’s protocols<br />
Primary antibody Company Antibody retrival Dilution Incubation<br />
Mouse Anti- PCNA<br />
(PC-10)<br />
Rabbit anti-pS6RP<br />
(Ser235/236)<br />
Mouse anti-alpha2uglobulin<br />
DAKO, Cat No:<br />
M0879<br />
Cell Signaling<br />
Technologies, Cat<br />
No: 2211<br />
Kind gift from Dr. U.<br />
Deschel (Boehringer<br />
Ingelheim, Biberach,<br />
Germany)<br />
7.3.5 Immunohistochemistry<br />
10mM sodium<br />
citrate buffer (pH<br />
6.0), 10min at 98°C<br />
10mM sodium<br />
citrate buffer (pH<br />
6.0), 10min at 98°C<br />
1: 50 in 1x<br />
powerblock:<br />
Biogenex; Cat No.<br />
HK085-5K<br />
1:100 in 5% normal<br />
goat serum<br />
none 1: 200 in 1x<br />
powerblock:<br />
Biogenex; Cat No.<br />
HK085-5K<br />
Over night at<br />
4°C<br />
Over night at<br />
4°C<br />
Over night at<br />
4°C<br />
Immunohistochemistry <strong>of</strong> phospho-S6 ribosomal protein (pS6RP), PCNA, and α2u-globulin<br />
in paraffin sections was carried out according to the manufacturers’ instructions (for brief<br />
information see table 7.1). Antigen–antibody complexes were visualized using the Super<br />
149
Chapter 7: Manuscript V<br />
SensitiveTM (BioGenex, USA) alkaline phosphatase-labeled, biotin–streptavidin amplified<br />
detection system and Fast Red as chromogen.<br />
7.3.6 Statistics<br />
Significant differences in nuclear labeling indices (LI%), or in the total number <strong>of</strong><br />
preneoplastic lesions, were calculated by using Bonferroni's test for multiple comparisons.<br />
Ranked non-neoplastic pathology data were analyzed by using Kruskal-Wallis tests. All<br />
analyses were performed by using GraphPad Prism 5.0 (GraphPad S<strong>of</strong>tware, San Diego,<br />
California, USA). All results are given as Means ± SD. Results were considered statistically<br />
significant when p < 0.05 (*p < 0.05; **p < 0.01 and ***p
Chapter 7: Manuscript V<br />
Table 7.2: Life-history traits <strong>of</strong> chow-fed rats (n=8), and DIOres (n=8) and DIOsens (n=7)<br />
subpopulations exposed to high fat diet.<br />
Body weight [g]<br />
Epididymal fat pad weights<br />
[g]<br />
Fasting glucose [mg/dL]<br />
Fed glucose [mg/dL]<br />
Triglyceride<br />
Free fatty acids<br />
Chow DIOres DIOsens<br />
468.0±10.9 649.6±15.9 1031.1±34.8<br />
20,5±1.3 76.5±7.9 204.4±12.5<br />
72.63±1.7 77.1±1.9 85.1±3.1<br />
99.6±2.4 99.8±3.7 104.5±3.0<br />
192.6±15.8 243.6±72.8 244.6±46.7<br />
0.44±0.030 0.41±0.033 0.37±0.027<br />
Despite the marked changes in body weight and fat mass in DIOsens and DIOres<br />
compared to chow-fed rats, no changes were observed in total fasting triglyceride,<br />
cholesterol or free fatty acid levels between the groups (Table 7.2).<br />
7.4.2 Increased levels <strong>of</strong> adipokines and insulin in DIOsens and DIOres<br />
rats<br />
Leptin levels were increased in both DIOsens and DIOres rats, compared to chow fed<br />
controls, however, the degree <strong>of</strong> body adiposity did not influence leptin levels in DIOres vs.<br />
DIOsens rats (Figure 7.1, A). In contrast, only DIOsens rats had higher insulin levels,<br />
compared to chow-fed or DIOres rats (Figure 7.1, B). Plasminogen activator inhibitor 1<br />
(PAI-1) levels correlated to the degree <strong>of</strong> body adiposity, and were higher in DIOres and<br />
especially in DIOsens rats (Figure 7.1, E). No changes were observed for Interleukin 1β (IL-<br />
1β) or monocyte chemoattractant protein 1 (MCP-1), or in any <strong>of</strong> the groups tested (Figure<br />
7.1, C and D).<br />
151
Chapter 7: Manuscript V<br />
Figure 7.1: Adipokine, cytokine, and insulin levels in plasma from chow-fed controls (open bars,<br />
n=8), and high-fat diet fed DIOres (striped bars, n=8), and DIOsens (grey bars, n=7)<br />
rats. All rats were fed ad libitum prior to blood sampling. Significant differences<br />
between groups were evaluated by using one-way ANOVA and Bonferroni's Post Test<br />
for multiple comparisons.<br />
7.4.3 High-fat diet exposure causes preneoplasia in DIOsens and DIOres<br />
rats<br />
Chronic high fat diet exposure resulted in an increased non-neoplastic renal pathology only<br />
in DIOsens and DIOres rats, but not in rats fed a control diet (Table 7.3). Most prominent<br />
changes included a marked chronic progressive nephropathy (CPN) <strong>of</strong> the renal cortex,<br />
including tubular degeneration and regeneration, glomerulosclerosis, interstitial fibrosis and<br />
tubules containing protein casts and a widespread mononuclear cell inflammation (Figure<br />
7.2, A and B). Notably, CPN in DIOsens and DIOres rats was associated with simple<br />
tubular hyperplasia characterized by an increased number <strong>of</strong> epithelial cells that do not<br />
extend beyond the single cell layer (Figure 7.2, A and B). Importantly, some tubules within<br />
the CPN lesions had a preneoplastic phenotype: partly solid, with basophilic cytoplasm,<br />
irregular shaped nuclei and not associated with a markedly thickened basement membrane<br />
(Figure 7.2, C and D). Notably, no overt changes in any <strong>of</strong> the observed parameters were<br />
observed between DIOres and DIOsens rats.<br />
152
Chapter 7: Manuscript V<br />
Table 7.3: Histopathology in chow-fed (n=8), DIOres (n=8) and DIOsens (n=7) rats exposed to<br />
high fat diet. Histopathological changes were ranked from none (0) to severe (4)<br />
including intermediate classes. Values are presented as median ± median absolute<br />
deviation.<br />
Glomeruli<br />
Cortex/Medulla<br />
Chow DIOres DIOsens<br />
Glomerulosclerosis 0.0 + 0.0 0.8 + 0.3 1.0 + 0.3*<br />
Fibrosis 0.0 + 0.0 0.8 + 0.3 0.5 + 0.2<br />
Dilated blood vessels 0.0 + 0.0 0.5 + 0.3 0.0 + 0.2<br />
Proteincast in Bowman’s<br />
space<br />
0.0 + 0.0 0.0 + 0.2 0.0 + 0.1<br />
Necrosis 0.0 + 0.1 0.5 + 0.4 1.0 + 0.5<br />
Vacuolisation 0.0 + 0.0 0.0 + 0.5 0.0 + 0.5<br />
Apoptosis 0.0 + 0.0 0.0 + 0.0 0.0 + 0.3<br />
CPN 0.3 + 0.3 1.5 + 0.6* 2.0 + 0.5***<br />
Regeneration 0.3 + 0.3 1.0 + 0.5* 1.5 + 0.3*<br />
Regenerative Hyperplasia 0.0 + 0.0 0.8 + 0.6 1.0 + 0.2<br />
Inflammation 0.0 + 0.2 1.3 + 0.6* 2.0 + 0.6**<br />
Protein Casts 0.0 + 0.1 1.3 + 0.6** 1.5 + 0.7***<br />
Calcium Casts 0.0 + 0.1 0.0 + 0.2 0.5 + 0.2<br />
Immunhistochmenistry alpha2u-globulin 2.5 + 0.1 2.5 + 0.4 2.5 + 0.1<br />
* **<br />
Significantly higher than the respective control group. Kruskal-Wallis test (p< 0.05). Significantly higher<br />
than the respective control group. Kruskal-Wallis test (p< 0.005). *** Significantly higher than the respective<br />
control group. Kruskal-Wallis test (p< 0.001)<br />
153
Chapter 7: Manuscript V<br />
Figure 7.2: Representative images <strong>of</strong> (A, B) simple tubular hyperplasia with tubules containing a<br />
thickened basal membrane and increased number <strong>of</strong> epithelial cells that do not<br />
extend beyond the single cell layer (1). Images <strong>of</strong> atypical hyperplasia (C, D), partly<br />
solid, with basophilic cytoplasm, irregular shaped nuclei and not associated with a<br />
markedly thickened membrane (2). Protein cast (3). Representative alpha 2u stained<br />
protein droplets in DIOsens rats (E, F).<br />
154
Chapter 7: Manuscript V<br />
7.4.4 High-fat diet-induced changes in kidney pathology are<br />
independent from α2u-globulin accumulation<br />
Positively stained cytoplasmatic α2u-globulin droplets were localized within the P2 segment<br />
<strong>of</strong> the proximal convoluted tubule. When classifying α2u-globulin staining intensity as none<br />
(0), mild (1), moderate (2), strong (3), and severe (4), similar intensities and therefore<br />
amounts <strong>of</strong> α2u-globulin were found in all groups tested, independent <strong>of</strong> the diet and<br />
severity <strong>of</strong> obesity (see last row in table 7.3, and Figure 7.2, E and F). In addition, no<br />
difference in the tubular localisation was detected, compared to the control rats. Thus, α2u-<br />
globulin did not accumulate with high-fat diet exposure, and did not seem to be implicated<br />
in the observed changes in renal pathology.<br />
7.4.5 High fat diet increases cell proliferation in proximal tubule <strong>of</strong><br />
DIOsens and DIOres rats<br />
Visualization <strong>of</strong> S-Phase nuclei by immunohistochemical staining <strong>of</strong> nuclear PCNA<br />
demonstrated a marked increase <strong>of</strong> proliferating cells within the CPN lesions compared to<br />
the healthy tissue. To determine whether high fat diet targets distinct tubular segments, cell<br />
proliferation was separately analyzed in random fields in specifiable areas <strong>of</strong> the outer and<br />
inner cortex. This resulted in the detection <strong>of</strong> a similar and significant increase <strong>of</strong> cell<br />
proliferation in the p1/p2 segments and p3 segments proximal tubules <strong>of</strong> DIOsens and<br />
DIOres rats respectively, compared to chow fed control rats (Figure 7.3, A and B).<br />
155
Chapter 7: Manuscript V<br />
Figure 7.3: (A) Representative images <strong>of</strong> PCNA staining in DIOsens rats. Red: PCNA positive<br />
nuclei, blue, negative counterstaining with haematoxilin. (B) Comparison <strong>of</strong> PCNA<br />
labelling indices (LI%) <strong>of</strong> Eker rats <strong>of</strong> chow fed, DIOsens and DIOres rats exposed<br />
with high fat diet. LI%, determined for proximal tubules (PT) within randomly chosen<br />
fields <strong>of</strong> the outer or inner cortex. Within both regions at least 1000 nuclei were<br />
counted. LI% were tested for significance using one-way ANOVA and Bonferroni's<br />
Post Test for multiple comparisons.<br />
7.4.6 Involvement <strong>of</strong> the mTOR pathway<br />
To evaluate the potential activation <strong>of</strong> the mTOR-S6K pathway by high-fat diet exposure,<br />
phosphorylation <strong>of</strong> Ser235/236 residues <strong>of</strong> S6RP was measured by immunohistochemical<br />
detection. S6RP Ser235/236 specifically gets phosphorylated by activated S6K1 (p70<br />
ribosomal protein S6 kinase 1) (Flotow and Thomas 1992) and serves as marker for<br />
mTOR-S6K activation. Only single S6RP phosphorylation positive cells were found in<br />
chow-fed control rats, or in unaffected tissue <strong>of</strong> any group (Figure 7.4, A). In contrast, in<br />
both DIOsens and DIOres rats positive staining was detected in nearly all cells <strong>of</strong><br />
regenerating tubules within the CPN lesions, and especially in tubules with an atypical<br />
156
Chapter 7: Manuscript V<br />
phenotype (Figure 7.4, B-D). However, no overt changes in S6RP phoshphorylation could<br />
be observed between DIOsens and DIOres rats.<br />
Figure 7.4: Representative images <strong>of</strong> single cell pS6RP staining (1) <strong>of</strong> a chow fed control rats,<br />
proliferating pS6RP positive tubules with in DIOsens rat (B-D). pS6RP positive tubular<br />
cell undergoing mitosis (2).<br />
7.5 Discussion<br />
Renal tissue damage, inflammation and regeneration are important factors contributing to<br />
the onset and progession <strong>of</strong> RCC by propagating the fixation <strong>of</strong> spontaneous mutations and<br />
progression <strong>of</strong> pre-neoplastic lesions into solid renal tumors (Dietrich and Swenberg 1990;<br />
Cohen and Ellwein 1991; Cohen and Arnold 2008). Obesity is a known risk factor for kidney<br />
cancer, and chronic exposure to dietary lipids has previously been shown to cause<br />
inflammation, regeneration and renal pathologies in animal models for obesity (Aguila and<br />
Mandarim-De-Lacerda 2003; Armitage et al. 2005; Altunkaynak et al. 2008). However,<br />
157
Chapter 7: Manuscript V<br />
none <strong>of</strong> these previous studies could convincingly distinguish between the direct influence<br />
<strong>of</strong> dietary fat, and the impact <strong>of</strong> the high body adiposity.<br />
This study describes for the first time the effect <strong>of</strong> chronic high fat diet exposure on renal<br />
pathology in DIOsens and DIOres rats. These rats were selected from a large population <strong>of</strong><br />
high-fat diet fed inbred Wistar rats, and showed marked differences in their body weight<br />
and fat pad weight despite the highly similar genetic background. High-fat diet exposure<br />
induced marked chronic progressive nephropathy (CPN) and an increased regenerative<br />
cell proliferation <strong>of</strong> similar severity in both DIOsens and DIOres rats, compared to chow fed<br />
controls. Notably, in DIOsens and DIOres rats, CPN was not only associated with a simple<br />
tubular hyperplasia, but also with preneoplastic lesions. Although neither DIOsens nor<br />
DIOres rats presented with full adenomas or carcinomas after high fat diet exposure, it is<br />
now widely accepted that atypical hyperplasias represent a preneoplastic entity which has<br />
the potential to progress to adenomas and carcinomas (Dietrich and Swenberg 1991; Hard<br />
and Khan 2004). However, it remains unclear whether atypical tubular hyperplasias, or<br />
other renal pathologies, are the result <strong>of</strong> high adiposity with its direct physiological<br />
consequences, <strong>of</strong> co-morbidities <strong>of</strong> obesity such as hypertension or diabetes, or <strong>of</strong> direct<br />
toxic effects <strong>of</strong> dietary fat.<br />
7.5.1 Similar renal preneoplasias in DIOsens and DIOres rats suggest a<br />
minor role <strong>of</strong> high body adiposity or its co-morbidities<br />
The lack <strong>of</strong> overt differences in non-neoplastic pathology between DIOres and DIOsens<br />
rats, points to a minor role for the level <strong>of</strong> adiposity in the progression <strong>of</strong> non-neoplastic<br />
pathology. However, despite significantly higher epididymal fat pad weights, leptin levels in<br />
DIOsens rats were only slightly higher than in DIOres rats. Previously, an excess secretion<br />
<strong>of</strong> leptin in obese animals was suggested to mediate, at least in part, the effects <strong>of</strong> high<br />
body adiposity on renal carcinogenesis (Takahashi et al. 1996; Attoub et al. 2000;<br />
Horiguchi et al. 2006). Thus, leptin might in part contribute to the observed renal pathology<br />
in DIOsens and DIOres rats. In obesity, monocyte infiltration <strong>of</strong> white adipose tissue can<br />
lead to the release <strong>of</strong> cytokines, and a sub-chronic, systemic inflammation. However,<br />
similar plasma levels <strong>of</strong> MCP1 and IL-ß in all groups tested do not suggest such an<br />
increased cytokine release, or sub-chronic inflammation. PAI-1 levels were reported to be<br />
elevated in obesity and might enhance fibrinolysis in renal disease by inhibiting plasmindependent<br />
extracellular matrix turnover, thus enhancing renal pathology (Eddy and Fogo<br />
2006). Accordingly, PAI-1 levels increased with the degree <strong>of</strong> body adiposity in DIOres rats<br />
158
Chapter 7: Manuscript V<br />
and DIOsens rats, compared to chow fed controls. Adipokine levels in diet-induced obesity<br />
might therefore contribute in part to the observed effects, however, their exact role in the<br />
etiology <strong>of</strong> kidney cancer still remains elusive. For instance, A-Zip/F-1 mice, which<br />
completely lack white adipose tissue and therefore have undetectable adipokine levels,<br />
nevertheless displayed higher tumor incidences, tumor multiplicity, and decreased tumor<br />
latency than wild-type mice in two experimental tumor models (Nunez et al. 2006).<br />
Insulin resistance, diabetes type II and associated hyperglycemia itself are well-established<br />
risk factors for the onset <strong>of</strong> renal cancer (Hjartaker et al. 2008). Chronic hyperinsulinemia<br />
may thereby induce an increased synthesis and bioavailability <strong>of</strong> the insulin-like growth<br />
factor 1 (IGF1), one <strong>of</strong> the most potent activators <strong>of</strong> the oncogenic AKT signaling pathway<br />
(Renehan et al. 2006). Hyperglycemia may further induce the localized production <strong>of</strong><br />
reactive oxygen species (ROS) via multiple mechanisms such as advanced glycation or<br />
enhanced glycolysis, and may ultimately lead to diabetic nephropathy (Coughlan et al.<br />
2008; Forbes et al. 2008) and renal DNA damage (Sebekova et al. 2007), therefore<br />
increasing the risk for kidney cancer. However, despite severe adiposity especially in the<br />
DIOS rats, similar blood glucose levels (ad libitum and fasted) in DIOres, DIOsens and<br />
chow fed animals indicated that high fat diet exposure did not induce hyperglycemia, one<br />
major hallmark <strong>of</strong> type II diabetes. In addition, insulin levels were only increased in<br />
DIOsens, but not in DIOres rats, compared to chow-fed controls. Based on the similar renal<br />
pathology <strong>of</strong> DIOsens and DIOres rats, hyperglycemia and hyperinsulinemia therefore do<br />
not seem to induce the here observed changes. Thus, the here presented findings suggest<br />
that the influence <strong>of</strong> body adiposity on renal pathology and renal carcinogenesis is rather<br />
minor.<br />
7.5.2 High-fat diet exposure does not induce α2u-globulin accumulation<br />
Dietary lipids may have a direct effect on renal pathology and early markers <strong>of</strong> renal<br />
tumorigenesis. It was thus initially hypothesized that an excessive renal accumulation <strong>of</strong> the<br />
fatty acid binding protein α2u-globulin may contribute to high fat diet-induced CPN, but<br />
immunohistochemical staining <strong>of</strong> α2u-globulin could not corroborate that hypothesis.<br />
However, glomerusclerosis, and tubular damage mainly within the proximal part <strong>of</strong> the renal<br />
nephron in DIOsens and DIOres rats might point to a direct toxic effect <strong>of</strong> lipids. Such<br />
effects were previously observed after chronic high fat diet feeding in leptin-receptor<br />
deficient db/db mice, or diet-induced obese C57BL/6J mice, and were attributed in part<br />
towards a sterol regulatory element binding protein (SREBP)-mediated lipid accumulation<br />
159
Chapter 7: Manuscript V<br />
in the kidney (Jiang et al. 2005; Wang et al. 2005). Previous reports suggested a different<br />
capacity for lipid metabolism within different tubular segments, with mitochondrial betaoxidation<br />
<strong>of</strong> fatty acids being relatively small in the proximal and the distal nephron, but<br />
high in the proximal tubule. Notably, the proximal tubule has a peroxisomal beta-oxidative<br />
capacity <strong>of</strong> the same order <strong>of</strong> magnitude as in liver cells <strong>of</strong> Wistar rats (Le Hir and Dubach<br />
1982). Higher lipid turnover in the glomerulus or proximal tubule may therefore lead to the<br />
release <strong>of</strong> reactive oxygen species by mitochondria, and regional oxidative stress, one key<br />
player in the pathogenesis <strong>of</strong> renal pathology.<br />
7.5.3 mTor-S6K signaling: The link between obesity and kidney cancer<br />
<strong>Kidney</strong> cancer has been linked to an abnormal activation <strong>of</strong> the mammalian target <strong>of</strong><br />
rapamycin (mTOR) pathway (Kenerson et al. 2002; Albanell et al. 2007; Stemmer et al.<br />
2007; Hanna et al. 2008). Activation <strong>of</strong> the mTOR pathway results in activation <strong>of</strong> its<br />
downstream target S6-Kinase (S6K) and further to phosphorylation <strong>of</strong> S6 ribosomal protein<br />
(S6RP) at Ser293. Due to the lack <strong>of</strong> S6K antibodies suitable for immunohistochemistry,<br />
staining <strong>of</strong> phosphoSer293- S6RP was used as surrogate marker for an activated mTOR<br />
pathway. The here presented data clearly demonstrate that mTOR pathway activation can<br />
only be observed in DIOsens and DIOres rats. Specifically, mTOR signaling is activated<br />
specifically in proximal tubules that also demonstrate an increased proliferation and<br />
regenerative hyperplasia with atypical nuclear morphology. The functional role <strong>of</strong> such<br />
mTor/S6K pathway activation in tumor initiation and/or tumor promotion, however, has not<br />
been clearly established. Kenerson and co-workers demonstrated that rapamycin, an<br />
inhibitor <strong>of</strong> mTOR, diminished renal preneoplastic progression to veritable renal tumors in<br />
Eker rats but not the formation <strong>of</strong> earliest preneoplastic lesions (Kenerson et al. 2002). In<br />
addition, an activation <strong>of</strong> the mTOR/S6K pathway might serve as facilitator <strong>of</strong> tumor<br />
initiation by an induction <strong>of</strong> cell proliferation, thus elevating the risk for the fixation <strong>of</strong> DNA<br />
mutations, e.g. induced by increased oxidative stress.<br />
The mTOR pathway is also known to be a central regulator for the nutrient-hormonal<br />
signaling network (Cota et al. 2008; Gulati et al. 2008), and deregulation by high-fat diet<br />
exposure might therefore induce human and rodent cancer. High fat diet exposure was<br />
previously shown to induce hypothalamic S6K activation (Ono et al. 2008). In addition,<br />
Palmitoleic acid was shown to rapidly induce S6K phosphorylation in various hypothalamic<br />
and muscle cell lines in vitro, and chronic exposure <strong>of</strong> mice to high-fat diet induced the<br />
phosphorylation <strong>of</strong> skeletal muscle S6K, whereas exposure to a fat-free diet reduced<br />
160
Chapter 7: Manuscript V<br />
skeletal muscle S6K phosphorylation (Castaneda et al. 2008). Thus, it seems entirely<br />
possible that fatty acids, derived from high-fat diet, could also activate S6K in renal cells.<br />
Further studies, based on in vitro exposure <strong>of</strong> primary kidney cells or established kidney<br />
cell lines with various fatty acids, will help to clarify this open question.<br />
7.5.4 Conclusion<br />
The here presented data suggests that dietary lipids can induce renal toxicity. This effect<br />
seems largely independent <strong>of</strong> the level <strong>of</strong> adiposity, and can manifest in non-diabetic rats.<br />
The high-fat diet induced activation <strong>of</strong> the mTOR pathway by dietary lipids with the<br />
increase in cell proliferation, may promote the fixation and progression <strong>of</strong> spontaneous<br />
mutations and the development <strong>of</strong> neoplasia and solid tumors, and may causally link<br />
obesity and renal cancer.<br />
161
Chapter 8: Additional Data<br />
8.1 Gender Disparity in Early Ochratoxin A and Aristolochic<br />
Acid Mediated Renal Toxicity in Short-term Assays<br />
This chapter summarizes additional data from short-term AA- and OTA-treated female Eker<br />
and wild type rats, respectively, that were not presented in one <strong>of</strong> the previous manuscripts.<br />
The experiments and analyses in female rats were done together with those from male<br />
rats, which are shown in chapter 3. Therefore, sex-specific effects <strong>of</strong> AA or OTA treatment<br />
on cell proliferation and renal pathology in female rats can be directly compared to those<br />
derived from male rats. For details on the experimental design, see material and methods<br />
<strong>of</strong> chapter 3. The results from these gender disparity experiments were partly presented as<br />
poster at the 47 th Annual Meeting <strong>of</strong> the Society <strong>of</strong> Toxicology, 2008 in Seattle, WA, USA.<br />
8.2 Ochratoxin A<br />
8.2.1 Cell proliferation (PCNA labeling index)<br />
Treatment <strong>of</strong> male and female Eker and wild typ rats with 210µg OTA/kg bw significantly<br />
increased cell proliferation in the renal cortex after 7 and 14 days <strong>of</strong> exposure (Figure 8.1).<br />
Male rats showed a strong tendency towards a higher OTA-dependent increase in cell<br />
proliferation, compared to female rats. Furthermore, the gender difference in cell<br />
proliferation was more pronounced in wild type than in Eker rats.<br />
162
Chapter 8: Additional Data<br />
Figure 8.1: Comparison <strong>of</strong> PCNA labelling indices (LI%) in the renal cortex <strong>of</strong> OTA-treated Eker<br />
and wild type rats. Data represent the mean + SD <strong>of</strong> n=3 animals per group.<br />
Significant differences between treated and time-matched control groups: *p < 0.05;<br />
**p < 0.01 and ***p
Chapter 8: Additional Data<br />
Figure 8.2: Comparison <strong>of</strong> total numbers and incidences <strong>of</strong> basophilic atypical tubules (bATs) in<br />
male and female Eker and wild type rats treated with OTA. Significant differences<br />
between treated and time-matched control groups: *p < 0.05; **p < 0.01 and ***p<br />
8.3 Aristolochic acid<br />
8.3.1 Cell proliferation (PCNA labeling index)<br />
Chapter 8: Additional Data<br />
Treatment <strong>of</strong> male and female Eker and wild typ rats with 10mg AA/kg bw significantly<br />
increased cell proliferation in the renal cortex <strong>of</strong> female but not male rats (Figure 8.3).<br />
Female Eker rats appeared to be more sensitive to AA than corresponding wild type rats,<br />
as an increased cell proliferation rate was already observed at day 7 <strong>of</strong> exposure.<br />
Figure 8.3: Comparison <strong>of</strong> PCNA labelling indices (LI%) in the renal cortex <strong>of</strong> AA-treated Eker<br />
and wild type rats. Data represent the mean + SD <strong>of</strong> n=3 animals per group.<br />
Significant differences between treated and time-matched control groups: *p < 0.05;<br />
**p < 0.01 and ***p
Chapter 8: Additional Data<br />
significantly increase the number <strong>of</strong> bATs in comparison to vehicle controls. Higher<br />
background numbers and incidences <strong>of</strong> bATs were found in TSC2-mutant Eker rats.<br />
Figure 8.4: Comparison <strong>of</strong> total numbers and incidences <strong>of</strong> basophilic atypical tubules (bATs) in<br />
male and female Eker and wild type rats treated with AA. Mean + SD. No significant<br />
differences between treated and time-matched control groups were detected by one<br />
way ANOVA with Bonferroni post-test.<br />
8.3.3 Conclusion<br />
Although no sex-differences in total numbers and incidences <strong>of</strong> preneoplastic lesions could<br />
be detected, the increased cell proliferation in AA-treated females but not in males suggest<br />
a higher sensitivity <strong>of</strong> female rats towards AA treatment. This is corroborated by previous<br />
findings, demonstrating 2-fold higher levels <strong>of</strong> dG-AAI and dA-AAII DNA adducts in kidneys<br />
<strong>of</strong> female than male Wistar rats following treatment with oral doses <strong>of</strong> 0.15mg/kg body<br />
weight for 3 months (Arlt et al. 2001).<br />
166
Chapter 9: Overall Discussion and Future Perspectives<br />
9.1 Short-term In Vivo Tests to Identify Renal Carcinogens<br />
Exposure to renal carcinogens remains one important cause for a vast majority <strong>of</strong> all<br />
human kidney cancers and identification <strong>of</strong> potential carcinogens serves as first step in<br />
disease prevention. Classification as carcinogen is mainly based on life-time rodent<br />
bioassays. These tests assume that cancer development is a basic process with a similar<br />
pathogenesis in humans and in experimental animals (Bannasch 1986a; Bannasch 1986b).<br />
However, the scientific value <strong>of</strong> rodent bioassays has come under criticism (Monro 1993;<br />
Davies et al. 2000; Haseman et al. 2001). The criticisms are: a) that the studies are too<br />
long and should rather be terminated at earlier points <strong>of</strong> life when the animals still have<br />
essentially normal organ functions and are free from widespread age-related pathology, b)<br />
that the carcinogenic doses are too high and, due to cytotoxicity and regenerative cell<br />
proliferation or dose-dependent differences in metabolism and pharmacokinetics, do not<br />
reflect the true carcinogenic potential, and c) that some organ-specific carcinogenic effects<br />
observed in experimental animals have little or no relevance to humans (e.g. α2u-globulin<br />
nephropathy associated with renal cancer in male rats, or chemically induced peroxisomal<br />
proliferation associated with rodent liver tumors). Other criticisms are that rodent bioassays<br />
require hundreds <strong>of</strong> animals, are time-consuming, and costly, especially due to the high<br />
amounts <strong>of</strong> compounds required.<br />
Due to the current limitations <strong>of</strong> the lifetime rodent bioassay, interest in developing sensitive<br />
short-term, mechanistically based in vivo studies has grown. One promising approach is<br />
the use <strong>of</strong> microarrays for the identification and characterization <strong>of</strong> novel mechanistic<br />
biomarkers involved in the early onset <strong>of</strong> cancer. In addition, since microarrays are highly<br />
sensitive detection tools, gene expression pr<strong>of</strong>iling may allow an early detection <strong>of</strong> adverse<br />
effects <strong>of</strong> test compounds, even before those effects have manifested in histopathological<br />
changes. However, before short-term assays in combination with gene expression pr<strong>of</strong>iling<br />
can be adopted as alternatives for the lifetime rodent bioassay, further research is needed<br />
to establish and validate their sensitivity, accuracy and predictivity to detect carcinogenic<br />
compounds. Some promising studies have been performed where gene expression pr<strong>of</strong>iles<br />
from rat livers were used to identify and distinguish known genotoxic and non-genotoxic<br />
167
Chapter 9: Overall Discussion and Future Perspectives<br />
hepato-carcinogens after 1 to 14 days <strong>of</strong> treatment (Ellinger-Ziegelbauer et al. 2005;<br />
Ellinger-Ziegelbauer et al. 2008). However, at the beginning <strong>of</strong> this thesis no comparable<br />
studies on renal carcinogens had been performed.<br />
The first aim <strong>of</strong> this thesis was therefore to investigate some <strong>of</strong> the most important<br />
prerequisites needed for a sensitive, specific and predictive short-term in vivo test to<br />
identify renal carcinogens. Those are:<br />
1. Detectable and compound-specific gene expression changes in kidney<br />
homogenate <strong>of</strong> short-term treated rats (Sensitivity)<br />
2. Possible differentiation between genotoxic and non-genotoxic renal carcinogens by<br />
these gene expression pr<strong>of</strong>iles (Specificity)<br />
3. Manifestation <strong>of</strong> early genotoxic or non-genotoxic expression changes in renal<br />
preneoplastic and neoplastic lesions (Predictivity)<br />
An in-depth discussion on the respective hypotheses and obtained results can be found in<br />
the respective manuscripts. This chapter will summarize the findings <strong>of</strong> all short-term and<br />
long-term in vivo experiments, and will discuss them in context to the above named<br />
prerequisites. Thus, a more comprehensive evaluation <strong>of</strong> all findings presented in this<br />
thesis might help to clarify whether the methods and model systems used for these studies<br />
can help to establish a new standard for renal toxicity testing. Ultimately, such a new<br />
standard test could shorten the duration <strong>of</strong> the life-time rodent bioassay, and could help to<br />
achieve the goal <strong>of</strong> the three R’s, to reduce, refine and replace animal experiments.<br />
9.1.1 Prerequisite I: Short term treatment would result in detectable<br />
compound-specific gene expression changes<br />
In the first part <strong>of</strong> this thesis, gene expression pr<strong>of</strong>iles <strong>of</strong> kidney cortex homogenates <strong>of</strong><br />
male Tsc2 mutant Eker- and corresponding wild-type rats were analyzed after 1, 3, 7 and<br />
14 days <strong>of</strong> treatment with genotoxic (AA and MAMAc) and non-genotoxic (OTA) renal<br />
carcinogens and compared with compound-induced changes in renal pathology and cell<br />
proliferation (for details see chapter 3, 4 and 8). In both Eker- and wild-type rats short-term<br />
treatment with the respective carcinogens resulted in minor but compound-specific changes<br />
in histopathology and cell proliferation that were reflected by early gene expression pr<strong>of</strong>iles.<br />
168
Chapter 9: Overall Discussion and Future Perspectives<br />
1. AA and OTA resulted in an up-regulation <strong>of</strong> genes and pathways involved in<br />
inflammatory responses, which was in consistence with the histopathological<br />
finding <strong>of</strong> an infiltration <strong>of</strong> monocytes in both groups.<br />
2. OTA induced the deregulation <strong>of</strong> genes modulated by acute toxic effects (e.g.<br />
genes involved in oxidative stress response, regenerative cell proliferation, and<br />
increased extracellular matrix synthesis). These changes correlated well with the<br />
histopathological findings in OTA-treated rats demonstrating tubular damage and<br />
regeneration with signs <strong>of</strong> fibrosis.<br />
3. OTA but not AA treatment resulted in an up-regulation <strong>of</strong> numerous genes involved<br />
in cell cycle progression and cell proliferation, which is in accordance to the OTAspecific<br />
increase in cell proliferation and with signs <strong>of</strong> regeneration.<br />
4. Short-term treatment with MAMAc only resulted in a small number <strong>of</strong> deregulated<br />
genes in male Eker and wild-type rats, when compared to AA and OTA treatment,<br />
which was in accordance with the lack <strong>of</strong> effects on pathology or cell proliferation in<br />
male rats.<br />
5. Carcinogen treatment generally resulted in a higher number <strong>of</strong> deregulated genes<br />
in Tsc2 mutant Eker rats compared to wild type rats. However, when compared to<br />
the respective controls, similar pathways were affected. Solely OTA-treated Eker<br />
but not wild type rats demonstrated a number <strong>of</strong> additional genes pointing to an<br />
OTA-mediated ERK-dependent inactivation <strong>of</strong> the Tsc2 tumor suppressor gene.<br />
This finding corresponded well with the significant manifestation <strong>of</strong> preneoplastic<br />
lesions, exclusively seen in OTA-treated Eker rats after 14 days <strong>of</strong> exposure.<br />
From these findings it was concluded that gene expression pr<strong>of</strong>iling allows an in-depth<br />
mechanistic insight into earliest cellular alterations after carcinogen exposure. This notion<br />
was corroborated by the finding that deregulated genes reflect the majority <strong>of</strong> cellular<br />
pathways which are known to be involved in the carcinogenic properties <strong>of</strong> AA and OTA.<br />
Table 9.1 summarizes these pathways, and compares the here presented findings to the<br />
specific results published elsewhere (for further references see Table 9.1). In conclusion,<br />
the findings from chapter 3 and 4 demonstrate that microarrays are sufficient to identify a<br />
compound-specific mode <strong>of</strong> action in the kidney already after just 1-14 days <strong>of</strong> exposure.<br />
169
Chapter 9: Overall Discussion and Future Perspectives<br />
Table 9.1: Summary <strong>of</strong> pathways, suspected to be involved in AA-, OTA- or MAMAc-induced<br />
renal toxicity and carcinogenicity, and <strong>of</strong> genes that may be involved in the suspected<br />
pathways or downstream responses (for details see chapter 3 and 4 and GEO<br />
repository (http://www.ncbi.nlm.nih.gov/projects/geo; accession number: GSE5923).<br />
AA<br />
OTA<br />
Major pathways previously<br />
described in chronic in vivo<br />
or in vitro<br />
Phase I metabolic activation <strong>of</strong><br />
AA (Arlt et al. 2002; Kohara et<br />
al. 2002; Li et al. 2006)<br />
DNA damage (Arlt et al. 2002;<br />
Lord et al. 2004)<br />
Caspase-3 activation<br />
(Balachandran et al. 2005)<br />
Mutations in H-ras and p53<br />
(Schmeiser et al. 1991; Arlt et<br />
al. 2002)<br />
Activation <strong>of</strong> cyclin D1/cdk4<br />
(Chang et al. 2006)<br />
Transcriptional repression <strong>of</strong><br />
Nrf2 target genes (Marin-Kuan<br />
et al. 2006)<br />
Increased histone deacetylase<br />
activity and down regulation <strong>of</strong><br />
RNA synthesis (Luhe et al.<br />
2003; Marin-Kuan et al. 2006)<br />
Transcriptional repression <strong>of</strong><br />
HNF4alpha target genes<br />
(Marin-Kuan et al. 2006)<br />
Activation <strong>of</strong> the mitogenic<br />
ERK1/2 pathway (Horvath et<br />
al. 2002; Marin-Kuan et al.<br />
2007)<br />
Activation <strong>of</strong> the stress-<br />
inducible JNK and p38<br />
pathway (Gekle et al. 2000;<br />
Rached et al. 2006)<br />
Activation <strong>of</strong> the NFkappa B<br />
pathway (Rached et al. 2006)<br />
Impairment <strong>of</strong> cellular Ca 2+<br />
and cAMP homeostasis<br />
(Rahimtula and Chong 1991;<br />
Benesic et al. 2000)<br />
Suggested downstream<br />
effects causally related to<br />
deregulated pathways<br />
Deregulated genes found in this<br />
study, supportive for previous<br />
described pathways and<br />
downstream effects<br />
DNA reactivity Up: AKR1B8, AKR7A3, EPHX1,<br />
CYP4A10, CYP51, NQO1, GNMT,<br />
CES2, CES3<br />
Cell cycle arrest, DNA repair Up: MGMT, PHLDA3, CDKN1A,<br />
SNK, CCNG1, MDM2, FTHFD,<br />
PRODH, OXR1<br />
Down: CDCA3, CCNB1, HMMR,<br />
CDC2, CKS2,<br />
Apoptosis, decreased cell<br />
survival and proliferation<br />
Cell survival and proliferation Not detectable<br />
Cell cycle progression and<br />
proliferation<br />
Impaired cellular defense<br />
against oxidative stress,<br />
enhanced oxidative stress,<br />
DNA-damage<br />
Reduced transcription and<br />
protein expression, gene<br />
silencing<br />
Reduced biotransformation,<br />
decreased detoxification and<br />
excretion<br />
Active mitogenic IGF-MAPK<br />
system, cell survival, cell<br />
transformation<br />
Up: TNFAIP8, HSPA5<br />
Down: USP2, SNN, Ki-67, FRZB,<br />
IGF1, CTGF, TGM2, NAB2,<br />
SFRP2, KLK7<br />
Not detectable<br />
Up: KEAP1, SUPT16H, CHEK2,<br />
MDM2, RBBP6, SEPP1, GGTLA1,<br />
TXNRD1<br />
Down: CN1, HSP40-3, MSRA<br />
Down: H2AA, MYST1, RPP25,<br />
POLD4, NR1D2, KLF13, ID1<br />
Down: CES1B, CES3, AADAC,<br />
CYP2T1, CYP2F4, WBSCR21,<br />
ALDH6A1 , CML1, NAT8, GGT6,<br />
MGST1, GNMT, OCTN1, OAT1,<br />
OCT1<br />
Up: PI3CB, AKT2<br />
Apoptosis Up: JNK2, STAT1, CYBB<br />
Down: CFLAR<br />
Suppressed apoptosis, G1/S<br />
cell cycle transition: Cell<br />
proliferation<br />
Disturbance <strong>of</strong> hormonal Ca 2+<br />
signaling, leading to altered cell<br />
proliferation<br />
Up: GMFB, NFE2L1<br />
Up: CALB, CALR, ITPKB,<br />
PRKAR2A, PPP1R2, CNB, NCX1,<br />
GNAI3, CHP, MPP2<br />
Down: SMP-30<br />
170
MAMAc<br />
Activated TGFß signaling<br />
(Gagliano et al. 2005; Sauvant<br />
et al. 2005)<br />
Spontaneous hydrolysis to the<br />
DNA-reactive<br />
methyldiazonium ion<br />
(Matsumoto and Higa 1966;<br />
Shank and Magee 1967; Sohn<br />
et al. 2001)<br />
Inhibited RNA synthesis<br />
(Zedeck et al. 1970)<br />
Chapter 9: Overall Discussion and Future Perspectives<br />
Altered cell proliferation and<br />
growth control, epithelialmesenchymal-transition,<br />
fibrosis<br />
DNA alkylation, cell cycle arrest,<br />
DNA damage repair, Apoptosis<br />
Up: TGFBR2, SMAD1, SMAD3,<br />
SMAD4, CTNNB1, PDGFC, HAI-1,<br />
NPHS1, TIMP3, DPT, CD44,<br />
NOTCH2<br />
Not detectable<br />
Tumor suppression and others Down: PLAGL1, NR1D1, NR1D2,<br />
NFIX, FKHL18, DBP, BHLHB3,<br />
BAT1A<br />
9.1.2 Prerequisite II: Genotoxic and non-genotoxic renal carcinogens<br />
can be distinguished by early gene expression pr<strong>of</strong>iles<br />
As demonstrated in chapter 3, 1 to 14 days <strong>of</strong> treatment with AA clearly resulted in the upregulation<br />
<strong>of</strong> numerous genes involved in DNA damage response, cell cycle arrest and<br />
apoptosis, indicative for a compound-induced DNA-damaging (genotoxic) action (Branzei<br />
and Foiani 2008) In contrast, gene expression pr<strong>of</strong>iles <strong>of</strong> OTA-treated rats point to an<br />
indirect genotoxic mode <strong>of</strong> action due to increased oxidative stress, as indicated by a broad<br />
down-regulation <strong>of</strong> genes involved in anti-oxidant defense. Compared to the predominant<br />
detection <strong>of</strong> genes involved in DNA damage response in AA-treated rats, OTA treatment<br />
resulted in a marginal deregulation <strong>of</strong> these gene groups, which suggests a secondary role<br />
<strong>of</strong> oxidative DNA damage in OTA-induced kidney cancer. More prominent, OTA treatment<br />
resulted in the additional up-regulation <strong>of</strong> genes involved in mitogenic processes, such as<br />
cell cycle progression, cell survival and proliferation (for details see chapter 3). Already<br />
these findings indicate that gene expression pr<strong>of</strong>iling may allow the prediction <strong>of</strong> genotoxic<br />
or non-genotoxic/ indirect genotoxic carcinogenic potential for a compound in short-term<br />
studies.<br />
This interpretation is further supported by previous reports showing that similar groups <strong>of</strong><br />
genes (indicative either for a genotoxic or and non-genotoxic mode <strong>of</strong> action) were<br />
deregulated in rat liver after short-term exposure to different known genotoxic or nongenotoxic<br />
compounds, respectively (Ellinger-Ziegelbauer et al. 2004; Ellinger-Ziegelbauer<br />
et al. 2005). Specifically, the genotoxic liver carcinogens 2-nitr<strong>of</strong>luorene,<br />
dimethylnitrosamine; 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and aflatoxin B1 all<br />
resulted in the up-regulation <strong>of</strong> p53 target genes such as CDKN1A, MGMT, CCNG1, or<br />
MDM2. Notably, these genes were also up-regulated in kidneys <strong>of</strong> short-term AA-treated<br />
rats (see chapter 3). In contrast, gene expression pr<strong>of</strong>iles after treatment with non-<br />
171
Chapter 9: Overall Discussion and Future Perspectives<br />
genotoxic liver carcinogens methapyrilene, diethylstilbestrol, Wy-14643 and<br />
piperonylbutoxide appeared more heterogeneous, with a slight emphasis on de-regulated<br />
genes involved in DNA replication and cell cycle progression, cell survival and proliferation<br />
(Ellinger-Ziegelbauer et al. 2005; Ellinger-Ziegelbauer et al. 2008). Consistently, such a<br />
gene expression pr<strong>of</strong>ile was also found in kidneys <strong>of</strong> rats after OTA treatment (Chapter 3).<br />
In contrast to the above findings, 1-14 days <strong>of</strong> treatment with the known genotoxic, DNAalkylating<br />
compound MAMAc did not result in characteristic gene expression pr<strong>of</strong>iles<br />
previously found for other genotoxic, or non-genotoxic compounds. In a predictive<br />
approach, this data would imply a non-carcinogenic action <strong>of</strong> MAMAc, at least at the low<br />
doses used for this study. However, the additional detection <strong>of</strong> an accumulating amount <strong>of</strong><br />
pro-mutagenic O6-methylguanine adducts by LC-MS/MS in the same samples used for<br />
gene expression pr<strong>of</strong>iling corroborates the potentially genotoxic, DNA-alkylating<br />
mechanism <strong>of</strong> MAMAc in Eker rats (see chapter 4). This however would suggest that<br />
microarray analysis might not be sensitive enough to detect the consequences <strong>of</strong> low but<br />
possibly relevant amounts <strong>of</strong> DNA adducts. Another explanation for the lack <strong>of</strong> deregulation<br />
<strong>of</strong> genes involved in DNA damage repair and apoptosis could be that basal expression <strong>of</strong><br />
genes involved in DNA repair and/or apoptosis might be sufficient to deal with low amounts<br />
<strong>of</strong> DNA adducts, at least until a certain threshold dose or duration <strong>of</strong> exposure is exceeded.<br />
Together, the findings presented here suggest that gene expression pr<strong>of</strong>iling in<br />
combination with the analysis <strong>of</strong> cell proliferation and histopathology may allow the<br />
distinction <strong>of</strong> genotoxic from non-genotoxic modes <strong>of</strong> action in short-term studies. However,<br />
microarray analyses may require a minimum amount <strong>of</strong> DNA damage, resulting in a<br />
detectable up-regulation <strong>of</strong> DNA damage response genes. More research on other renal<br />
carcinogens is therefore needed to verify these findings and also to elucidate whether a<br />
carcinogen can be clearly distinguished from nephrotoxic/non-carcinogenic mechanisms in<br />
short term assays.<br />
9.1.3 Prerequisite III: Manifestation <strong>of</strong> early compound-induced<br />
expression changes in kidney tumors.<br />
Eker rats were chronically exposed with AA, OTA, or MAMAc for either 3 or 6 months.<br />
Prolonged exposure should reveal whether early gene expression changes, indicative for a<br />
genotoxic, non-genotoxic/ indirect genotoxic, or non-carcinogenic mode <strong>of</strong> action, would<br />
indeed manifest in the onset or absence <strong>of</strong> renal tumors, respectively. Eker rats were<br />
chosen as the only animal model in these chronic studies out <strong>of</strong> both ethical and financial<br />
172
Chapter 9: Overall Discussion and Future Perspectives<br />
considerations. Specifically, various studies have demonstrated a substantially higher<br />
susceptibility <strong>of</strong> Eker rats towards genotoxic and non-genotoxic carcinogens, resulting in an<br />
earlier manifestation and higher number <strong>of</strong> carcinogen-induced renal pre-neoplastic and<br />
neoplastic lesions compared to wild-type rats (Walker et al. 1992; Wolf et al. 1998; Wolf et<br />
al. 2000). Thus, the duration <strong>of</strong> the study could be significantly shortened and fewer<br />
animals and less compound could be used.<br />
As demonstrated in chapter 4 and 6, prolonged carcinogen exposure corroborated the<br />
findings <strong>of</strong> the short-term assay: Both AA and OTA treatment resulted in a significant<br />
increase in total numbers <strong>of</strong> preneoplastic lesions, and the absence there<strong>of</strong> in MAMActreated<br />
male Eker rats. Moreover, in accordance with previous in vivo studies using AA and<br />
OTA, a similar sex-specific susceptibility towards both carcinogens was found; pointing to a<br />
compound-specific carcinogenic effect <strong>of</strong> AA and OTA in Eker rats (for details see chapter<br />
6 and chapter 8).<br />
A sex-specific susceptibility towards a carcinogenic action <strong>of</strong> MAMAc has not been<br />
described before. The findings <strong>of</strong> this thesis point to a higher susceptibility <strong>of</strong> female rats,<br />
as a significant increase <strong>of</strong> MAMAc-induced preneoplastic lesions was solely observed in<br />
female rats. In male Eker rats, prolonged treatment with MAMAc did not manifest in a<br />
significant increase <strong>of</strong> preneoplastic lesions. This was in accordance with the gene<br />
expression pr<strong>of</strong>iles derived from short-term treated males that did not point to a genotoxic<br />
mode <strong>of</strong> action <strong>of</strong> MAMAc, further corroborating the predictive potential <strong>of</strong> short-term<br />
assays. The lack <strong>of</strong> a tumor-initiating action was surprising since MAMAc has been<br />
identified as a potent genotoxic renal carcinogen in male rats (Laqueur 1964; Laqueur and<br />
Matsumoto 1966; Hirono et al. 1968; Fukunishi et al. 1971). However, all previous studies<br />
on the carcinogenic action <strong>of</strong> MAMAc have used acute high doses (25 to 500 mg/kg bw),<br />
compared to the here shown study using low and chronic doses (250 µg/kg bw). It thus<br />
seems possible that tumor initiation by MAMAc might depend on a threshold dose<br />
ultimately leading to a sufficient amount <strong>of</strong> DNA adducts which then might result in<br />
transforming mutations in critical genes and tumor initiation.<br />
Importantly, although 6 month treatment with MAMAc did not result in an increased number<br />
<strong>of</strong> preneoplastic lesions in males, they showed an increased incidence and higher total<br />
number <strong>of</strong> neoplastic lesions, compared to the respective control rats (for details see<br />
chapter 4). These findings could therefore point to a differential influence <strong>of</strong> MAMAc on<br />
early and later tumor stages, and may be explained by the here detected MAMAc-induced<br />
up-regulation <strong>of</strong> the TGF-ß pathway. This pathway is known to function as a tumor<br />
suppressor at early stages <strong>of</strong> kidney cancer but it can also promote tumor progression and<br />
173
Chapter 9: Overall Discussion and Future Perspectives<br />
metastasis in later stages (Gold 1999; Laping et al. 2007). These findings may therefore<br />
indicate that short-term gene expression pr<strong>of</strong>iling could also be a useful tool to predict the<br />
carcinogenic potential <strong>of</strong> compounds acting on later stages <strong>of</strong> carcinogenesis.<br />
9.1.4 Summary Part I<br />
In summary, the initial hypothesis <strong>of</strong> this thesis, that short-term in vivo assays, using<br />
microarrays as detection system, can be used as a sensitive and specific tool to predict the<br />
renal carcinogenic potential <strong>of</strong> test compounds in the kidney, is corroborated by the here<br />
presented data. However, additional work needs to be done to evaluate gene expression<br />
pr<strong>of</strong>iles in response to different dose-regimens, to also elucidate whether one compound, in<br />
a dose-dependent manner, can simultaneously or successively adopt both non-genotoxic<br />
and genotoxic mechanisms.<br />
Furthermore it needs to be established whether short-term gene expression pr<strong>of</strong>iles would<br />
indeed allow a reliable detection <strong>of</strong> carcinogens that target genes and pathways important<br />
for later stages <strong>of</strong> renal carcinogenesis. This however would require a more detailed<br />
understanding <strong>of</strong> signaling cascades involved in tumor promotion and progression, and the<br />
impact <strong>of</strong> renal carcinogens on these processes. This question was further elucidated in the<br />
second part <strong>of</strong> this thesis and is discussed in detail the following section.<br />
174
Chapter 9: Overall Discussion and Future Perspectives<br />
9.2 Cellular Signaling Cascades in Preneoplastic Lesion<br />
Progression<br />
The two-year rodent bioassay is mainly based on the detection <strong>of</strong> end-stage adenomas and<br />
carcinomas. Shorter in vivo bioassays will however no longer be based on the detection <strong>of</strong><br />
end-stage tumors, but rather have to use specific molecular, predictive endpoints. Clearly,<br />
however, such predictive tests would require a detailed understanding <strong>of</strong> mechanisms<br />
involved in tumor initiation, promotion and progression.<br />
Currently, mechanisms involved in multistage kidney carcinogenesis are only poorly<br />
understood. Thus, the next aim <strong>of</strong> this thesis was to identify and characterize signaling<br />
pathways involved in the progression <strong>of</strong> early preneoplastic renal lesions (for details see<br />
chapter 3 and 6). This should further allow to:<br />
• Analyze the impact <strong>of</strong> genotoxic and non-genotoxic renal carcinogens on these<br />
pathways<br />
• Determine, whether these genes and pathways could be detected in short-term<br />
assays<br />
Since only prolonged treatment with AA and OTA, but not with MAMAc, resulted in a<br />
compound-specific increase <strong>of</strong> total numbers <strong>of</strong> bATs and bAHs in males, gene expression<br />
studies from microdissected lesions were restricted to male AA- and OTA-treated rats and<br />
vehicle controls. For technical reasons, later tumor stages (adenomas and carcinomas)<br />
were not further addressed, since the cryo-conserved tissue designated for laser<br />
microdissection did not comprise enough replicate tumors in each dose group for proper<br />
statistical analysis.<br />
First, a protocol was established that allowed a reproducible and reliable analysis <strong>of</strong> gene<br />
expression pr<strong>of</strong>iles from laser-microdissected preneoplastic lesions (for details see chapter<br />
5). Using this protocol, gene expression pr<strong>of</strong>iles <strong>of</strong> different stages <strong>of</strong> preneoplastic lesions<br />
(bATs and bAHs) and healthy tubules <strong>of</strong> AA-, OTA- and vehicle-treated male Eker rats<br />
were analyzed and revealed the following key findings:<br />
1. Consistent with the initial hypothesis, the here presented findings demonstrate for<br />
the first time that large-scale gene expression pr<strong>of</strong>iling from microdissected earliest<br />
stages <strong>of</strong> renal preneoplastic lesions is possible (for details see chapter 5 and 6).<br />
175
Chapter 9: Overall Discussion and Future Perspectives<br />
2. As expected at the beginning <strong>of</strong> this thesis, preneoplastic lesions showed a<br />
massive deregulation <strong>of</strong> genes compared to healthy tissue <strong>of</strong> control rats, clearly<br />
demonstrating that tumor cells are different from their cellular origin (chapter 6).<br />
3. In contrast to the initial hypothesis, different stages <strong>of</strong> microdissected preneoplastic<br />
lesions (bATs and bAHs) were similar in their transcriptome chapter 6).<br />
4. Unexpectedly, comparisons <strong>of</strong> gene expression pr<strong>of</strong>iles from microdissected<br />
preneoplastic lesions <strong>of</strong> vehicle-treated with AA- and OTA-treated rats revealed<br />
only marginal differences. Thus, carcinogen treatment does not seem to influence<br />
gene deregulation once a preneoplastic lesion has manifested.<br />
5. Chronic treatment with AA or OTA did not substantially change gene expression in<br />
microdissected healthy tubules, which was in contrast to the overt carcinogeninduced<br />
gene expression changes in short-term treated Eker rats (see chapter 3<br />
and 6).<br />
The above findings are discussed in detail in the respective manuscripts. The next chapter<br />
will summarize all findings and propose a hypothetical model <strong>of</strong> carcinogen-induced renal<br />
tumorigenesis in Eker rats and potentially also wild-type strains. This model delineates the<br />
pivotal roles <strong>of</strong> carcinogen-induced tubular damage, regeneration and adaptation<br />
processes for the development <strong>of</strong> kidney cancer (see chapter 9.2.1), and proposes a<br />
predictive potential <strong>of</strong> preneoplastic lesions for the carcinogenic outcome <strong>of</strong> the rodent<br />
bioassays (see chapter 9.2.2).<br />
9.2.1 Pivotal role <strong>of</strong> carcinogen-induced tubular damage, regeneration<br />
and adaptation processes for the development <strong>of</strong> kidney cancer<br />
Based on the differing gene expression pr<strong>of</strong>iles from kidney homogenates and healthy<br />
tubules <strong>of</strong> short-term vs. chronically treated rats, respectively, a hypothetical model <strong>of</strong><br />
carcinogen-induced renal tumorigenesis is proposed (Figure 9.1, A-E) This model assumes<br />
that proximal tubules <strong>of</strong> a similar microscopic phenotype generally differ in their sensitivity<br />
towards renal carcinogens (Figure 9.1, A). Such differences in carcinogen sensitivity could<br />
be based on variations in the cellular environment (e.g. amount <strong>of</strong> blood and carcinogen<br />
supply), cellular polymorphisms in carcinogen uptake, metabolic activation, detoxification,<br />
DNA damage repair or many other variations in intracellular carcinogen binding sites.<br />
Depending on their intrinsic sensitivity, some tubules would remain rather unaffected by<br />
carcinogen treatment, while others are sensitive towards genotoxic, non-genotoxic or<br />
176
Chapter 9: Overall Discussion and Future Perspectives<br />
cytotoxic effects. Gene expression pr<strong>of</strong>iling from short-term treated rats will therefore reveal<br />
compound-specific deregulations in sensitive tubules, as shown in chapter 3 and 4).<br />
As demonstrated for AA and OTA, prolonged carcinogen treatment can result in<br />
compound-specific cytotoxic or genotoxic effects accompanied with apoptosis or necrosis<br />
and regenerative cell proliferation (Chapter 6). Regeneration might be based on the<br />
replication <strong>of</strong> less affected or more insensitive neighboring cells (Figure 9.1, B). Thus,<br />
recurring cycles <strong>of</strong> damage and regeneration might also constitute a selection towards<br />
carcinogen-insensitive cells with a healthy phenotype (Figure 9.1, C). Gene expression<br />
pr<strong>of</strong>iles from microdissected healthy tubules <strong>of</strong> carcinogen-treated rats, when compared<br />
with healthy tubules <strong>of</strong> control rats, would thus not demonstrate overt compound-specific<br />
effects even though all healthy tubules are still systemically supplied with carcinogens<br />
(Chapter 6, Figure 6.2). However, a qualitative comparison <strong>of</strong> short-term gene expression<br />
pr<strong>of</strong>iles from kidney homogenates <strong>of</strong> AA- and OTA-treated rats with those derived from<br />
microdissected healthy tubules <strong>of</strong> AA- and OTA-treated rats, respectively, demonstrated<br />
several similarly deregulated genes (Figure 9.2). This finding may point to a low number <strong>of</strong><br />
sensitive healthy tubules remaining in AA- and OTA-treated rats.<br />
Figure 9.1: Overview <strong>of</strong> the hypothetical model <strong>of</strong> chemically-induced renal carcinogenesis in<br />
Eker rats<br />
177
Chapter 9: Overall Discussion and Future Perspectives<br />
Recurring cycles <strong>of</strong> compound-induced cell death and regeneration also allow the fixation<br />
<strong>of</strong> carcinogen-induced or spontaneous mutation and, if critical genes were affected, their<br />
clonal expansion to an atypical tubule (Figure 9.1, D). Thus, the compound-specific amount<br />
<strong>of</strong> damage and regeneration could be an important driving force for the development <strong>of</strong><br />
preneoplastic lesions that determines their total numbers. A correlation between tubular<br />
damage, increased cell proliferation and the total numbers <strong>of</strong> preneoplastic lesions was<br />
demonstrated in chapter 4 and 6 and is further supported by the findings <strong>of</strong> Cunningham<br />
and colleagues (Cunningham et al. 1993), who demonstrated that mutagenic compounds<br />
that increased renal cell proliferation also increased renal tumors, whereas mutagens<br />
having no effect on renal cell proliferation did not.<br />
The carcinogen-induced transformation <strong>of</strong> tubular cells together with the finding that<br />
atypical tubules are insensitive towards carcinogen treatment points to a carcinogen<br />
resistance <strong>of</strong> clonal expanding cells (Figure 9.1, E). A similar phenomenon is already<br />
known as multidrug resistance, where e.g. cancer cells have the ability to become resistant<br />
to chemotherapeutics. This includes protective mechanisms, e.g. an increased efflux <strong>of</strong><br />
drugs or enzymatic deactivation, but could also be the result <strong>of</strong> decreased uptake or<br />
membrane permeability, or <strong>of</strong> altered drug binding sites (Szakacs et al. 2006). Indeed, as<br />
demonstrated in chapter 6, the functional analysis <strong>of</strong> gene expression pr<strong>of</strong>iles derived from<br />
preneoplastic lesions demonstrates a broad down-regulation <strong>of</strong> transporters and metabolic<br />
enzymes previously demonstrated to be involved in AA- and OTA-mediated toxicity and<br />
carcinogenicity (Gekle and Silbernagl 1994; Stiborova et al. 2003; O'Brien and Dietrich<br />
2005), for details see GEO repository, http://www.ncbi.nlm.nih.gov/projects/geo; accession<br />
number: GSE 10608. However, it remains elusive whether the here detected downregulation<br />
<strong>of</strong> genes is a specific feature <strong>of</strong> atypical renal tubular cells, or results from an<br />
unspecific dedifferentiation <strong>of</strong> fast proliferating cells. An additional explanation for the<br />
general insensitivity <strong>of</strong> preneoplastic lesions towards AA- and OTA-treatment could be that<br />
a temporarily reduced blood supply in fast growing pre-neoplastic foci could reduce the<br />
amount <strong>of</strong> carcinogen reaching the cells, at least until angiogenesis starts.<br />
178
Chapter 9: Overall Discussion and Future Perspectives<br />
Figure 9.2: Heatmap <strong>of</strong> compound-specific unions <strong>of</strong> genes found significantly deregulated (red:<br />
up-regulated; green: down-regulated) after carcinogen treatment. Gene expression<br />
pr<strong>of</strong>iles are derived from AA- or OTA-treated Eker or wild type rats after 1 or 14 days<br />
<strong>of</strong> treatment (WT-d1 to EK/d14), and from microdissected healthy tubules <strong>of</strong> 6 month<br />
treated rats (HT). All genes were compared to their corresponding vehicle controls<br />
(n=3 per time-point). The color scale at the left depicts gene expression ratios. Note<br />
that only a qualitative comparison <strong>of</strong> genes is possible, since arrays were not<br />
processed at same time points and only 1mg/kg bw instead <strong>of</strong> 10mg/kg AA were used<br />
for chronic experiments.<br />
The here presented findings suggest that the increase <strong>of</strong> preneoplastic lesions primarily<br />
depends on the initial DNA damaging or nephrotoxic action <strong>of</strong> AA and OTA, while lesion<br />
progression is the result <strong>of</strong> compound-independent clonal expansion (Chapter 3 and 6).<br />
The extent <strong>of</strong> damage and regeneration may also determine the selection rate <strong>of</strong><br />
compound-insensitive cells. This suggests the existence <strong>of</strong> a specific time window where<br />
healthy tubules are still sensitive towards carcinogen treatment and not yet adapted. Such<br />
a critical period <strong>of</strong> carcinogen exposure would imply that extended compound exposure<br />
might add little to the formation <strong>of</strong> additional preneoplastic lesions. Instead, the number <strong>of</strong><br />
preneoplastic lesions formed during this critical period can determine the incidence and<br />
number <strong>of</strong> veritable tumors at later stages.<br />
However, not all carcinogen-induced preneoplastic lesions will progress to neoplastic<br />
lesions. Instead, some preneoplastic lesions may also be self-limiting or even reversible<br />
(Alden and Kanerva 1982; Eustis 1989; Dietrich and Swenberg 1991). The identification <strong>of</strong><br />
179
Chapter 9: Overall Discussion and Future Perspectives<br />
pathways that distinguish self-limiting or reversible preneoplastic lesions from lesions with<br />
true neoplastic potential would therefore greatly improve the predictive value <strong>of</strong><br />
preneoplastic lesions for carcinogenicity testing. Potential genes and pathways are<br />
discussed in the following section.<br />
9.2.2 Predictive potential <strong>of</strong> preneoplastic lesions as tumor precursors<br />
Tumor progression has been suggested to depend on acquiring six essential "hallmarks <strong>of</strong><br />
cancer”, that collectively dictate malignant growth (Hanahan and Weinberg 2000) (Chapter<br />
1.1.3 and Table 9.2). The basic concept <strong>of</strong> Hanahan and Weinberg is well accepted and<br />
may be shared by the majority, if not by all tumor types. However, to date it is not clear<br />
whether some or all <strong>of</strong> these acquired <strong>characteristics</strong> are detectable in preneoplastic<br />
lesions.<br />
Gene expression pr<strong>of</strong>iles derived from different progression stages <strong>of</strong> renal tumors, i.e.<br />
from early basophilic atypical tubules (bAT) were highly similar to later basophilic atypical<br />
hyperplasias (bAHs) (Chapter 6). This raised the hypothesis that early atypical tubules may<br />
already have all prerequisites to develop into neoplastic lesions. Indeed, functional analysis<br />
<strong>of</strong> gene expression pr<strong>of</strong>iles from microdissected renal bATs and bAHs identified groups <strong>of</strong><br />
deregulated genes and pathways that were indicative for all six hallmarks <strong>of</strong> cancer. Table<br />
9.2 summarizes some <strong>of</strong> the functional groups <strong>of</strong> genes and pathways that may give rise to<br />
the respective physiological changes, as proposed by Hanahan and Weinberg. Together,<br />
these findings suggest that the earliest microscopically detectable preneoplastic lesions<br />
from Eker rats may have the potential to develop into a malignant phenotype.<br />
Functional analysis <strong>of</strong> deregulated genes as well as immunohistochemical staining <strong>of</strong><br />
pS6RP, pAKT and FOXO1 in bATs and bAHs <strong>of</strong> control, AA- and OTA-treated Eker rats<br />
further demonstrated an activation <strong>of</strong> the mTOR pathway when compared to healthy tissue<br />
<strong>of</strong> control animals (for details see chapter 6). The latter may result from a functional loss <strong>of</strong><br />
the TSC2 tumor suppressor gene in preneoplastic lesions by genetic or epigenetic events,<br />
leading to a permanent activation <strong>of</strong> the TSC2-mTOR pathway.<br />
180
Chapter 9: Overall Discussion and Future Perspectives<br />
Table 9.2: Genes and pathways linked to the six hallmarks <strong>of</strong> cancer, as suggested by Hanahan<br />
and Weinberg (Hanahan and Weinberg 2000). For specific information see chapter 6<br />
and GEO repository (http://www.ncbi.nlm.nih.gov/projects/geo; accession number:<br />
GSE 10608)<br />
Hallmark <strong>of</strong> cancer Suggested functional gene groups indicative<br />
for the respective hallmarks<br />
Self-sufficiency in<br />
growth signals<br />
Insensitivity to<br />
growth inhibitory<br />
signals<br />
Evasion <strong>of</strong><br />
apoptosis<br />
Limitless replicative<br />
potential<br />
Sustained<br />
angiogenesis<br />
Tissue invasion and<br />
metastasis<br />
Up-regulation <strong>of</strong> growth factors and their<br />
receptors<br />
Significantly deregulated genes, found<br />
in preneoplastic lesions <strong>of</strong> this study,<br />
supportive for the respective hallmark<br />
Up: CTGF, HDGFRP3, IGFBP1, CNTF<br />
Up-regulation <strong>of</strong> oncogenes Up: MYC, NIBAN, ELK3, CASC4<br />
Up-regulation <strong>of</strong> members <strong>of</strong> the SOS-RAS-RAF-<br />
MAPK pathway<br />
Down-regulation <strong>of</strong> tumor suppressor genes or<br />
survival factors<br />
Down-regulation <strong>of</strong> terminal differentiation factors Down: HNF-4A<br />
Up-regulation <strong>of</strong> anti-apoptotic genes and survival<br />
factors<br />
Down-regulation <strong>of</strong> pro-apoptotic genes Down: DNASE1<br />
Up-regulation <strong>of</strong> genes involved in cell cycle<br />
progression<br />
Up: RAB2, RAB3D, RAB31, RAB6, NXN<br />
Down: PINK1, CRYL1, FOXO1A, CHN2<br />
Up: PEA-1, BNIP2, TMBIM1, ARC,<br />
SH3BGRL3, CLU, GAS6, NAP1L1, KITL,<br />
NRTN, EMP1<br />
Up:, CCNB1, CCNB2, CDC2, CDCA3,<br />
CKS2, PCTK2, AURKB, KIF15, KIF22,<br />
PLK4, UBE2C, RMBS2, POLD1, MCM2,<br />
MCM3, PAF<br />
Up-regulation <strong>of</strong> endothelial growth factors Up: PAI1 CXCR4, GP38, DDL4, DSCR1,<br />
Up-regulation <strong>of</strong> smooth muscle components Up: CRP1, PSEL,<br />
Up-regulation <strong>of</strong> proteases or cell contact<br />
inhibition<br />
Up-regulation <strong>of</strong> genes involved in epithelial-<br />
mesenchymal transition (EMT)<br />
Up: tPA, BMP1, CTSK, CAPN2, ST14,<br />
MEP1A, DPP7, KLK12, FXYD5, CD44<br />
Up: VIM, CTGF, ACVR1, TGFBI, TGFB2,<br />
Down-regulation <strong>of</strong> protease inhibitors Down: SERPINF2, HRG<br />
Down-regulation <strong>of</strong> cell adhesion molecules Down: CXN-26, PCLKC, TFF3, TNS<br />
An in-depth description <strong>of</strong> this pathway is given in the introduction <strong>of</strong> chapter 6 and is<br />
summarized in Figure 9.4. Briefly, the mammalian target <strong>of</strong> rapamycin (mTOR) is an<br />
intracellular Ser/Thr kinase that is part <strong>of</strong> two distinct protein complexes called TORC1 and<br />
TORC2. The raptor-containing mTOR complex TORC1 is inhibited by the TSC1/ TSC2<br />
complex, remains rapamycin-sensitive, and mainly activates ribosomal biogenesis,<br />
translation and nutrient import. TORC2 is rapamycin-insensitive and is required for the<br />
organization <strong>of</strong> the cytoskeleton (Mak and Yeung 2004; Jozwiak 2006; Wullschleger et al.<br />
2006). Both TORC1 and TORC2 are key regulators <strong>of</strong> numerous important cellular<br />
processes like cell growth, -survival, -proliferation, -motility and the nutrient-hormonal<br />
signaling network. However, upon loss <strong>of</strong> the tumor suppressor TSC2, e.g. by carcinogeninduced<br />
mutations, mTORC1 activity is aberrantly increased, leading to pathologies such<br />
as cancer, metabolic disorders and cardiovascular diseases (Thomas 2006).<br />
181
Chapter 9: Overall Discussion and Future Perspectives<br />
Under physiological conditions, TSC2 can be phosphorylated and inactivated by AKT in<br />
response to growth factors (e.g. insulin), thus resulting in activation <strong>of</strong> mTORC1. TORC2 in<br />
turn was demonstrated to directly phosphorylate AKT at Ser 473, resulting in the activation<br />
<strong>of</strong> several down-stream pathways, including an indirect activation <strong>of</strong> TORC1 (Wullschleger<br />
et al. 2006). Recent studies added even further complexity by suggesting the existence <strong>of</strong> a<br />
negative feedback loop from the activated TSC2-TORC1-S6K1 pathway to the upstream<br />
insulin-responsive IRS-PI3K-PDK1-AKT pathway (Figure 9.4) (Harrington et al. 2004; Um<br />
et al. 2004). It was previously demonstrated that the mTORC1-dependent negative<br />
feedback loop is still present in benign and growth-limited hamartomas (Kwiatkowski 2003;<br />
Manning et al. 2005). This was in contrast to the findings <strong>of</strong> this study, which point to a<br />
concomitant activation <strong>of</strong> TORC1 and TORC2 signaling and therefore to a disturbance <strong>of</strong><br />
this feedback loop in preneoplastic lesions from Eker rats (for details see chapter 6). In this<br />
scenario, AKT activation due to TORC2 dependent Ser473 phosphorylation (Bellacosa et<br />
al. 2005) would not only further ameliorate the AKT-TSC2-TORC1-dependent oxygen and<br />
energy supply <strong>of</strong> proliferating cells via increased angiogenesis and glycolysis, but would<br />
also reduce AKT-dependent apoptosis and cell cycle progression.<br />
Together with the functional gene expression analysis, pointing to the establishment <strong>of</strong> all<br />
six hallmarks <strong>of</strong> cancer, it might be possible that the simultaneous activation <strong>of</strong> TORC1 and<br />
TORC2 may act as a neoplastic switch in renal preneoplastic lesions. Thus, disturbance <strong>of</strong><br />
the negative feedback loop <strong>of</strong> the AKT-TSC2-mTOR pathway may be indicative <strong>of</strong> a<br />
malignant outcome. This is consistent with previous findings showing that the suppression<br />
<strong>of</strong> phosphatase and tensin homolog PTEN in TSC2 mutant cells, which efficiently blocks<br />
the phosphorylation <strong>of</strong> AKT, reactivates the PI3K-AKT pathway to generate more<br />
aggressive tumors (Ma et al. 2005b). However, further research is needed to verify a direct<br />
mechanistic correlation <strong>of</strong> these findings, for instance by using specific TORC1 or TORC2<br />
inhibitors and a combination there<strong>of</strong> in in vivo carcinogenicity studies. A specific inhibitor <strong>of</strong><br />
TORC1, rapamycin, is already used under the trade name ORISEL® (temsirolimus; Wyeth<br />
USA) to treat patients with advanced RCC. However, and in accordance with the here<br />
presented findings, new evidence suggests that mTOR might have a more central<br />
rapamycin-insensitive role in the pathology <strong>of</strong> some cancers e.g. by hyperactive TORC2<br />
signalling. In Spring 2008, a small molecule dual TORC1/TORC2 kinase inhibitor has<br />
reached clinical trials (OSI-027, OSI Pharmaceuticals, Inc), which might help to further<br />
clarify the role <strong>of</strong> this pathway in tumorigenesis.<br />
182
Chapter 9: Overall Discussion and Future Perspectives<br />
Figure 9.3: Overview <strong>of</strong> the mTOR pathway and its hypothesized function within a proximal<br />
tubule. Modified from (Wullschleger 2006, Thomas, 2006, Mak, 2004, Jacinto, 2004,<br />
Hannah, 2008).<br />
9.2.3 Summary Part II<br />
For the first time, the findings from Eker rats have demonstrated a possible adaptation <strong>of</strong><br />
(damaged and regenerating) healthy tubules and an insensitivity <strong>of</strong> preneoplastic lesions<br />
towards prolonged carcinogen treatment. Furthermore, these data indicate that the<br />
incidence and number <strong>of</strong> veritable tumors is primarily driven by the number <strong>of</strong> preneoplastic<br />
lesions initially formed during a critical period <strong>of</strong> carcinogen exposure. Clonal expansion <strong>of</strong><br />
preneoplastic lesions however seems to be a compound-independent process, and may be<br />
essentially driven by the mTOR pathway. The finding that both, TORC1 and TORC2 are<br />
activated in preneoplastic lesions <strong>of</strong> Eker rats points to a disturbed negative feedback loop<br />
<strong>of</strong> the PI3K-AKT-TSC2-mTOR pathway. Together with the establishment <strong>of</strong> the 6 hallmarks<br />
<strong>of</strong> (malignant) cancer, this finding suggests that the earliest detectable preneoplastic<br />
lesions <strong>of</strong> Eker rats have the potential to overcome growth- and proliferation-limiting<br />
barriers, and thus develop into full neoplasms. <strong>Molecular</strong> <strong>characteristics</strong> <strong>of</strong> the earliest<br />
183
Chapter 9: Overall Discussion and Future Perspectives<br />
detectable preneoplastic lesions might therefore already be indicative for their later<br />
outcome. In conclusion, the use <strong>of</strong> preneoplastic lesions as predictive endpoints in<br />
shortened rodent bioassays seems principally possible.<br />
9.3 Obesity as biasing factor for a reliable human risk<br />
assessment from rodent bioassays<br />
A large number <strong>of</strong> previous reports, as well as findings presented within this thesis, have<br />
demonstrated an important role <strong>of</strong> the activated mTOR pathway in certain types <strong>of</strong> rodent<br />
and human cancer (see 9.2.2). Next to its role in tumorigenesis, the mTOR pathway is also<br />
known to be a central regulator for the nutrient-hormonal signaling network and activation<br />
<strong>of</strong> the TORC1 complex is commonly detected in in vivo and in vitro models <strong>of</strong> obesity<br />
(Castaneda et al. 2008; Cota et al. 2008; Gulati et al. 2008. Thus, it was hypothesized that<br />
hyperactivation <strong>of</strong> the mTOR pathway might be a missing link between obesity and kidney<br />
cancer (Thomas, 2006). However, the exact impact <strong>of</strong> chronic high fat diet exposure on<br />
kidney cancer and the renal mTOR pathway remains widely elusive.<br />
In the final part <strong>of</strong> this thesis, it was therefore hypothesized that prolonged high fat diet<br />
exposure leads to the manifestation <strong>of</strong> severe renal pathology with early markers <strong>of</strong> renal<br />
carcinogenesis. Specifically, this thesis aimed to dissect the impact <strong>of</strong> high body adiposity<br />
with its known co-morbidities and disease consequences versus the direct impact <strong>of</strong> dietary<br />
lipids on cellular renal pathways involved in renal carcinogenesis, such as the mTOR-S6K<br />
pathway. Renal pathology in a diet-sensitive subpopulation ((DIOsens) characterized by a<br />
morbidly increased body weight) was examined and compared to a diet-resistant<br />
subpopulation ((DIOres) with moderately increased body weight) <strong>of</strong> male Wistar rats as well<br />
as to chow fed control rats with normal body weight. This resulted in the following key<br />
findings (for details see chapter 7).<br />
1. DIOsens and DIOres rats but not control rats demonstrated marked tubular<br />
damage, regenerative cell proliferation and chronic progressive nephropathy<br />
(CPN). CPN was not only associated with a simple tubular hyperplasia, but also<br />
with preneoplastic lesions, which might have the potential to manifest into solid<br />
tumors at later stages <strong>of</strong> life.<br />
2. In both DIOsens and DIOres rats positive staining <strong>of</strong> phosphorylated ribosomal<br />
protein S6 (P-RPS6), a downstream effecter <strong>of</strong> activated TORC1, was detected in<br />
184
Chapter 9: Overall Discussion and Future Perspectives<br />
nearly all cells <strong>of</strong> regenerating tubules within the CPN lesions, and especially in<br />
tubules with an atypical phenotype.<br />
3. A similar degree <strong>of</strong> histopathological changes in DIOsens and DIOres rats<br />
suggested a more prominent role <strong>of</strong> the dietary fat content, rather than the degree<br />
<strong>of</strong> adiposity, for the etiology <strong>of</strong> renal pathology and TORC1 activation.<br />
Together these findings demonstrate a clear adverse effect <strong>of</strong> dietary lipids on renal<br />
morphology and further implicate that dietary lipids may act via similar mechanism like<br />
nephrotoxic compounds. The findings <strong>of</strong> chapter 3, 4 and 6 implicate that renal tissue<br />
damage, inflammation and regeneration are important factors contributing to the onset and<br />
progression <strong>of</strong> RCC. High-fat diet-induced renal damage therefore may further propagate<br />
the fixation <strong>of</strong> spontaneous mutations and progression <strong>of</strong> pre-neoplastic lesions into solid<br />
renal tumors. Notably, preneoplastic lesions are frequently found within the CPN lesions<br />
and might be indicative for a cancer promoting action <strong>of</strong> dietary lipids. However, to date it<br />
cannot be determined whether such preneoplastic lesions would further develop into gross<br />
tumors. Instead, the observed lesions could be self-limiting or even reversible, since no<br />
adenomas or carcinomas could be detected in this study (see chapter 9.2). Nevertheless,<br />
the lack <strong>of</strong> gross tumors was not surprising and indeed was not expected after 11 month <strong>of</strong><br />
high-fat diet exposure, since full adenomas and carcinomas in rodent bioassays are <strong>of</strong>ten<br />
only detectable after 15 to 18 months <strong>of</strong> carcinogen treatment (Haseman et al. 2001).<br />
The findings <strong>of</strong> chapter 7 have demonstrated an increased amount <strong>of</strong> phosphorylated<br />
RPS6 in atypical and regenerating tubules <strong>of</strong> DIOsens and DIOres rats, but not in<br />
unaffected tissue or in control animals. These preliminary findings may point to a role <strong>of</strong><br />
TORC1 in dietary lipid-induced renal damage and regeneration. However, it cannot be<br />
concluded whether high fat diet directly results in an activation <strong>of</strong> TORC1 and RPS6<br />
phosphorylation in renal cells (e.g. via pathways summarized in Figure 9.4). Alternatively,<br />
TORC1 activation could simply be a secondary effect <strong>of</strong> cell regeneration, e.g. via<br />
stimulation with extracellular growth factors like IGF-1 (Moore et al. 2008). Thus, clearly<br />
more research is needed to clarify whether the nephrotoxic action <strong>of</strong> high fat diet may be<br />
involved in the onset or progression <strong>of</strong> spontaneous mutations. This question could be<br />
answered in an in vivo tumor initiation-promotion study, using e.g. aristolochic acid as<br />
kidney specific tumor initiator and high fat diet as possible promoter. In addition, further<br />
studies should be conducted in vitro in primary kidney cells or established kidney cell lines,<br />
to clarify whether direct exposure to various fatty acids might directly stimulate renal<br />
TORC1 or TORC2.<br />
185
Chapter 9: Overall Discussion and Future Perspectives<br />
Considering that obesity has reached epidemic proportions, further elucidating the impact<br />
<strong>of</strong> high fat diet on renal and other types <strong>of</strong> cancer seems not only important under a<br />
medical aspect, but also regarding carcinogenicity testing in rodents. The choice <strong>of</strong> diet can<br />
severely affect the results <strong>of</strong> the rodent bioassay. For instance ad libitum-feeding <strong>of</strong> a high<br />
caloric diet considerably enhanced weight gain, increased background tumor development<br />
and non-neoplastic lesions, and decreased the life-span <strong>of</strong> rodents, compared to those fed<br />
with a caloric restricted chow (Keenan et al. 1995; Keenan et al. 1996). Thus, the dosage<br />
and composition <strong>of</strong> a diet can significantly influence the effects <strong>of</strong> a test compound,<br />
possibly by enhancing tubular damage and regeneration or by (synergistic) a”ctivation <strong>of</strong><br />
the mTOR pathway.<br />
In conclusion, high-fat diet exposure with high body adiposity clearly represents principal<br />
biasing factors for a reliable human risk assessment <strong>of</strong> compounds in rodent bioassays. It<br />
seems entirely possible that previous carcinogenicity testing in rodents underestimated the<br />
potential impact <strong>of</strong> a carcinogen in human subpopulations exposed to a high fat/ high<br />
calorie environment. Future studies should address this important issue, and should clarify<br />
whether carcinogen exposure in obese human subpopulations indeed could be linked to an<br />
enhanced number and incidence <strong>of</strong> renal cancer.<br />
9.4 Synopsis<br />
Evaluating the carcinogenic potential <strong>of</strong> test compounds by life-time rodent bioassays is<br />
time-consuming and associated with high financial costs, high numbers <strong>of</strong> animals, and a<br />
high amount <strong>of</strong> compound is required for continuous dosing over the entire period <strong>of</strong> the<br />
assay. With respect to the kidney, the findings <strong>of</strong> this thesis suggest that it could be<br />
possible to shorten rodent bioassays and to increase the specificity <strong>of</strong> these assays by<br />
using predictive biomarkers in short-term, or intermediate in vivo assays. The key findings<br />
<strong>of</strong> this study, (a) that microarrays allow the detection and distinction <strong>of</strong> genotoxic and nongenotoxic<br />
carcinogens in short-term assays and (b) that the incidence and number <strong>of</strong><br />
veritable tumors is primarily driven by the number <strong>of</strong> preneoplastic lesions initially formed<br />
during a critical period <strong>of</strong> exposure. If the above hypothesis would be transferable to other<br />
organs, then the following possibilities seem likely:<br />
• Predictive biomarkers could serve as early and specific indicators for a genotoxic or<br />
non-genotoxic carcinogenic action <strong>of</strong> test compounds in short-term (prescreening)<br />
186
Chapter 9: Overall Discussion and Future Perspectives<br />
assays. With short term assays the use <strong>of</strong> carcinogenic compounds could be<br />
reduced.<br />
• The continuous dosage <strong>of</strong> animals could be limited to the “critical” time window<br />
where cells are not yet adapted to the carcinogen exposure. Thus, much shorter<br />
time-period <strong>of</strong> exposure with a concomitant reduction <strong>of</strong> pain and distress for the<br />
animals could be achieved.<br />
• Predictive biomarkers might identify the carcinogenic potential <strong>of</strong> preneoplastic<br />
lesions by distinguishing between a potential malignant or self-limiting phenotype.<br />
Furthermore, specific markers may help to distinguish carcinogen-induced<br />
preneoplastic lesions from spontaneous background lesions. Such mechanistic<br />
biomarkers involved in rodent and human preneoplastic lesion progression (e.g. the<br />
mTOR pathway) could therefore serve as improved predictors for the neoplastic<br />
outcome in rats and the potential risk in humans.<br />
• By further specifying the carcinogenic potential <strong>of</strong> preneoplastic lesions, these could<br />
indeed serve as reliable endpoints for carcinogenicity testing and thus shorten the<br />
assay. In addition, reduced mortality in shorter assays would also allow a reduction in<br />
the number <strong>of</strong> animals. Moreover, residual animals would not suffer from extensive<br />
tumor growth.<br />
In general, higher efficiencies <strong>of</strong> the current safety testing paradigm (shorter assays with<br />
fewer animals) would lead to the reduction <strong>of</strong> financial costs for product development. In<br />
conclusion, the here presented findings suggest that the standard rodent bioassay can<br />
indeed be improved, mainly by shortening the duration <strong>of</strong> the life-time rodent bioassay, and<br />
by using novel methodology such as gene array. Ultimately, this thesis provides<br />
encouraging first results that may one day help to establish a new rodent bioassay that<br />
ultimately reduces and refine refines and even replaces animal experiments.<br />
187
Chapter 10: References<br />
Abdollahi A (2007) LOT1 (ZAC1/PLAGL1) and its family members: mechanisms and functions. J Cell<br />
Physiol 210:16-25<br />
Aguila MB, Mandarim-De-Lacerda CA (2003) Effects <strong>of</strong> chronic high fat diets on renal function and cortical<br />
structure in rats. Exp Toxicol Pathol 55:187-195<br />
Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF (2003) Prospective identification <strong>of</strong><br />
tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 100:3983-3988<br />
Albanell J, Dalmases A, Rovira A, Rojo F (2007) mTOR signalling in human cancer. Clin Transl Oncol<br />
9:484-493<br />
Alden CL, Kanerva RL (1982) Reversibility <strong>of</strong> renal cortical lesions induced in rats by high doses <strong>of</strong><br />
nitrilotriacetate in chronic feeding studies. Food Chem Toxicol 20:935-937<br />
Altunkaynak ME, Ozbek E, Altunkaynak BZ, Can I, Unal D, Unal B (2008) The effects <strong>of</strong> high-fat diet on<br />
the renal structure and morphometric parametric <strong>of</strong> kidneys in rats. J Anat 212:845-852<br />
Ames BN, Gold LS (1990) Chemical carcinogenesis: too many rodent carcinogens. Proc Natl Acad Sci U<br />
S A 87:7772-7776<br />
Arlt VM, Pfohl-Leszkowicz A, Cosyns J, Schmeiser HH (2001) Analyses <strong>of</strong> DNA adducts formed by<br />
ochratoxin A and aristolochic acid in patients with Chinese herbs nephropathy. Mutat Res<br />
494:143-150<br />
Arlt VM, Stiborova M, Schmeiser HH (2002) Aristolochic acid as a probable human cancer hazard in<br />
herbal remedies: a review. Mutagenesis 17:265-277<br />
Arlt VM, Stiborova M, vom Brocke J, Simoes ML, Lord GM, Nortier JL, Hollstein M, Phillips DH, Schmeiser<br />
HH (2007) Aristolochic acid mutagenesis: molecular clues to the aetiology <strong>of</strong> Balkan endemic<br />
nephropathy-associated urothelial cancer. Carcinogenesis 28:2253-2261<br />
Armitage JA, Lakasing L, Taylor PD, Balachandran AA, Jensen RI, Dekou V, Ashton N, Nyengaard JR,<br />
Poston L (2005) Developmental programming <strong>of</strong> aortic and renal structure in <strong>of</strong>fspring <strong>of</strong> rats fed<br />
fat-rich diets in pregnancy. J Physiol 565:171-184<br />
Armitage P, Doll R (1954) The age distribution <strong>of</strong> cancer and a multi-stage theory <strong>of</strong> carcinogenesis. Br J<br />
<strong>Cancer</strong> 8:1-12<br />
Attoub S, Noe V, Pirola L, Bruyneel E, Chastre E, Mareel M, Wymann MP, Gespach C (2000) Leptin<br />
promotes invasiveness <strong>of</strong> kidney and colonic epithelial cells via phosphoinositide 3-kinase-, rho-,<br />
and rac-dependent signaling pathways. Faseb J 14:2329-2338<br />
Balachandran P, Wei F, Lin RC, Khan IA, Pasco DS (2005) Structure activity relationships <strong>of</strong> aristolochic<br />
acid analogues: toxicity in cultured renal epithelial cells. <strong>Kidney</strong> Int 67:1797-1805<br />
Bannasch P (1986a) Preneoplastic lesions as end points in carcinogenicity testing. I. Hepatic<br />
preneoplasia. Carcinogenesis 7:689-695<br />
Bannasch P (1986b) Preneoplastic lesions as end points in carcinogenicity testing. II. Preneoplasia in<br />
various non-hepatic tissues. Carcinogenesis 7:849-852<br />
Bannasch P, Nogueira E, Zerban H (1989) [Cytology and cytogenesis <strong>of</strong> experimental epithelial kidney<br />
tumors]. Verh Dtsch Ges Pathol 73:301-313<br />
Barrett JC, Wiseman RW (1987) Cellular and molecular mechanisms <strong>of</strong> multistep carcinogenesis:<br />
relevance to carcinogen risk assessment. Environ Health Perspect 76:65-70<br />
Baugh LR, Hill AA, Brown EL, Hunter CP (2001) Quantitative analysis <strong>of</strong> mRNA amplification by in vitro<br />
transcription. Nucleic Acids Res 29:E29<br />
Baylin SB, Herman JG (2000) DNA hypermethylation in tumorigenesis: epigenetics joins genetics. Trends<br />
Genet 16:168-174<br />
Bellacosa A, Kumar CC, Di Crist<strong>of</strong>ano A, Testa JR (2005) Activation <strong>of</strong> AKT kinases in cancer:<br />
implications for therapeutic targeting. Adv <strong>Cancer</strong> Res 94:29-86<br />
Benesic A, Mildenberger S, Gekle M (2000) Nephritogenic ochratoxin A interferes with hormonal signalling<br />
in immortalized human kidney epithelial cells. Pflugers Arch 439:278-287<br />
Bergstrom A, Hsieh CC, Lindblad P, Lu CM, Cook NR, Wolk A (2001) Obesity and renal cell cancer--a<br />
quantitative review. Br J <strong>Cancer</strong> 85:984-990<br />
Bjornsson J, Short MP, Kwiatkowski DJ, Henske EP (1996) Tuberous sclerosis-associated renal cell<br />
carcinoma. Clinical, pathological, and genetic features. Am J Pathol 149:1201-1208<br />
Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from<br />
a primitive hematopoietic cell. Nat Med 3:730-737<br />
Boorman GA, McDonald MR, Imoto S, Persing R (1992) Renal lesions induced by ochratoxin A exposure<br />
in the F344 rat. Toxicol Pathol 20:236-245<br />
188
Chapter 10: References<br />
Borenstein AR, Mortimer JA, Sch<strong>of</strong>ield E, Wu Y, Salmon DP, Gamst A, Olichney J, Thal LJ, Silbert L, Kaye<br />
J, Craig UL, Schellenberg GD, Galasko DR (2007) Cycad exposure and risk <strong>of</strong> dementia, MCI,<br />
and PDC in the Chamorro population <strong>of</strong> Guam. Neurology 68:1764-1771<br />
Branzei D, Foiani M (2008) Regulation <strong>of</strong> DNA repair throughout the cell cycle. Nat Rev Mol Cell Biol<br />
9:297-308<br />
Brugarolas JB, Vazquez F, Reddy A, Sellers WR, Kaelin WG, Jr. (2003) TSC2 regulates VEGF through<br />
mTOR-dependent and -independent pathways. <strong>Cancer</strong> Cell 4:147-158<br />
Casarett LJ, Doull J (1986) Toxic response <strong>of</strong> the kidney. Macmillan Publishing Company, New York<br />
Castaneda TR, Abplanalp W, Um S, Pfluger PT, Brown K, Grant E, Schrott E, Carnevalli L, Benoit SC,<br />
Morgan DA, Gilham D, Hui DY, Ramouhni K, Thomas G, Kozma SC, D.J. C, Tschöp MH (2008)<br />
The metabolic function <strong>of</strong> S6 kinases depends on dietary lipids. Cell Metab submitted<br />
Cavin C, Delatour T, Marin-Kuan M, Holzhauser D, Higgins L, Bezencon C, Guignard G, Junod S, Piguet<br />
D, Richoz-Payot J, Gremaux E, Hayes JD, Nestler S, Mantle P, Schilter B (2006) Reduction in<br />
Antioxidant Defences May Contribute to Ochratoxin A Toxicity and Carcinogenicity. Toxicol Sci<br />
Chan W, Cui L, Xu G, Cai Z (2006) Study <strong>of</strong> the phase I and phase II metabolism <strong>of</strong> nephrotoxin<br />
aristolochic acid by liquid chromatography/tandem mass spectrometry. Rapid Commun Mass<br />
Spectrom 20:1755-1760<br />
Chandra M, Riley MG, Johnson DE (1993) Spontaneous renal neoplasms in rats. J Appl Toxicol 13:109-<br />
116<br />
Chang HR, Lian JD, Lo CW, Chang YC, Yang MY, Wang CJ (2006) Induction <strong>of</strong> urothelial proliferation in<br />
rats by aristolochic acid through cell cycle progression via activation <strong>of</strong> cyclin D1/cdk4 and cyclin<br />
E/cdk2. Food Chem Toxicol 44:28-35<br />
Chen T, Guo L, Zhang L, Shi L, Fang H, Sun Y, Fuscoe JC, Mei N (2006) Gene Expression Pr<strong>of</strong>iles<br />
Distinguish the Carcinogenic Effects <strong>of</strong> Aristolochic Acid in Target (<strong>Kidney</strong>) and Non-target (Liver)<br />
Tissues in Rats. BMC Bioinformatics 7 Suppl 2:S20<br />
Cheng CL, Chen KJ, Shih PH, Lu LY, Hung CF, Lin WC, Yesong Gu J (2006) Chronic renal failure rats are<br />
highly sensitive to aristolochic acids, which are nephrotoxic and carcinogenic agents. <strong>Cancer</strong> Lett<br />
232:236-242<br />
Chipman JK, Mally A, Edwards GO (2003) Disruption <strong>of</strong> gap junctions in toxicity and carcinogenicity.<br />
Toxicol Sci 71:146-153<br />
Choi YJ, Kim CJ, Ji GE (1996) A partially purified beta-glucosidase from Bifidobacterium adolescentis<br />
converts cycasin to a mutagenic compound. Lett Appl Microbiol 22:145-148<br />
Chow WH, Gridley G, Fraumeni JF, Jr., Jarvholm B (2000) Obesity, hypertension, and the risk <strong>of</strong> kidney<br />
cancer in men. N Engl J Med 343:1305-1311<br />
Clark HA, Snedeker SM (2006) Ochratoxin a: its cancer risk and potential for exposure. J Toxicol Environ<br />
Health B Crit Rev 9:265-296<br />
Clarke MF, Dick JE, Dirks PB, Eaves CJ, Jamieson CH, Jones DL, Visvader J, Weissman IL, Wahl GM<br />
(2006) <strong>Cancer</strong> stem cells--perspectives on current status and future directions: AACR Workshop<br />
on cancer stem cells. <strong>Cancer</strong> Res 66:9339-9344<br />
Cohen SM, Arnold LL (2008) Cell Proliferation and Carcinogenesis. J Toxicol Pathol 21:1-7<br />
Cohen SM, Ellwein LB (1991) Genetic errors, cell proliferation, and carcinogenesis. <strong>Cancer</strong> Res 51:6493-<br />
6505<br />
Corgna E, Betti M, Gatta G, Roila F, De Mulder PH (2007) Renal cancer. Crit Rev Oncol Hematol 64:247-<br />
262<br />
Cosyns JP (2003) Aristolochic acid and 'Chinese herbs nephropathy': a review <strong>of</strong> the evidence to date.<br />
Drug Saf 26:33-48<br />
Cosyns JP, Goebbels RM, Liberton V, Schmeiser HH, Bieler CA, Bernard AM (1998) Chinese herbs<br />
nephropathy-associated slimming regimen induces tumours in the forestomach but no interstitial<br />
nephropathy in rats. Arch Toxicol 72:738-743<br />
Cosyns JP, Jadoul M, Squifflet JP, Wese FX, van Ypersele de Strihou C (1999) Urothelial lesions in<br />
Chinese-herb nephropathy. Am J <strong>Kidney</strong> Dis 33:1011-1017<br />
Cota D, Matter EK, Woods SC, Seeley RJ (2008) The role <strong>of</strong> hypothalamic mammalian target <strong>of</strong><br />
rapamycin complex 1 signaling in diet-induced obesity. J Neurosci 28:7202-7208<br />
Coughlan MT, Mibus AL, Forbes JM (2008) Oxidative stress and advanced glycation in diabetic<br />
nephropathy. Ann N Y Acad Sci 1126:190-193<br />
Crist<strong>of</strong>ori P, Zanetti E, Fregona D, Piaia A, Trevisan A (2007) Renal proximal tubule segment-specific<br />
nephrotoxicity: an overview on biomarkers and histopathology. Toxicol Pathol 35:270-275<br />
Cui M, Liu ZH, Qiu Q, Li H, Li LS (2005) Tumour induction in rats following exposure to short-term high<br />
dose aristolochic acid I. Mutagenesis 20:45-49<br />
Cunningham ML, Elwell MR, Matthews HB (1993) Site-specific cell proliferation in renal tubular cells by the<br />
renal tubular carcinogen tris(2,3-dibromopropyl)phosphate. Environ Health Perspect 101 Suppl<br />
5:253-257<br />
189
Chapter 10: References<br />
Davies TS, Lynch BS, Monro AM, Munro IC, Nestmann ER (2000) Rodent carcinogenicity tests need be<br />
no longer than 18 months: an analysis based on 210 chemicals in the IARC monographs. Food<br />
Chem Toxicol 38:219-235<br />
Debelle FD, Nortier JL, De Prez EG, Garbar CH, Vienne AR, Salmon IJ, Deschodt-Lanckman MM,<br />
Vanherweghem JL (2002) Aristolochic acids induce chronic renal failure with interstitial fibrosis in<br />
salt-depleted rats. J Am Soc Nephrol 13:431-436<br />
Delatour T, Mally A, Richoz J, Ozden S, Dekant W, Ihmels H, Otto D, Gasparutto D, Marin-Kuan M,<br />
Schilter B, Cavin C (2008) Absence <strong>of</strong> 2'-deoxyguanosine-carbon 8-bound ochratoxin A adduct in<br />
rat kidney DNA monitored by isotope dilution LC-MS/MS. Mol Nutr Food Res 52:472-482<br />
Delker D, Hatch G, Allen J, Crissman B, George M, Geter D, Kilburn S, Moore T, Nelson G, Roop B, Slade<br />
R, Swank A, Ward W, DeAngelo A (2006) <strong>Molecular</strong> biomarkers <strong>of</strong> oxidative stress associated<br />
with bromate carcinogenicity. Toxicology 221:158-165<br />
Dietrich DR, Swenberg JA (1990) Lindane induces nephropathy and renal accumulation <strong>of</strong> alpha 2uglobulin<br />
in male but not in female Fischer 344 rats or male NBR rats. Toxicol Lett 53:179-181<br />
Dietrich DR, Swenberg JA (1991) Preneoplastic lesions in rodent kidney induced spontaneously or by nongenotoxic<br />
agents: predictive nature and comparison to lesions induced by genotoxic carcinogens.<br />
Mutat Res 248:239-260<br />
Dill JA, Lee KM, Renne RA, Miller RA, Fuciarelli AF, Gideon KM, Chan PC, Burka LT, Roycr<strong>of</strong>t JH (2003)<br />
Alpha 2u-globulin nephropathy and carcinogenicity following exposure to decalin<br />
(decahydronaphthalene) in F344/N rats. Toxicol Sci 72:223-234<br />
Doi AM, Hill G, Seely J, Hailey JR, Kissling G, Bucher JR (2007) alpha 2u-globulin nephropathy and renal<br />
tumors in national toxicology program studies. Toxicol Pathol 35:533-540<br />
Domijan AM, Peraica M, Vrdoljak AL, Radic B, Zlender V, Fuchs R (2007) The involvement <strong>of</strong> oxidative<br />
stress in ochratoxin A and fumonisin B1 toxicity in rats. Mol Nutr Food Res 51:1147-1151<br />
Dong H, Suzuki N, Torres MC, Bonala RR, Johnson F, Grollman AP, Shibutani S (2006) Quantitative<br />
determination <strong>of</strong> aristolochic acid-derived DNA adducts in rats using 32Ppostlabeling/polyacrylamide<br />
gel electrophoresis analysis. Drug Metab Dispos 34:1122-1127<br />
Duarte CG, Preuss HG (1993) Assessment <strong>of</strong> renal function--glomerular and tubular. Clin Lab Med 13:33-<br />
52<br />
Dulaimi E, Ibanez de Caceres I, Uzzo RG, Al-Saleem T, Greenberg RE, Polascik TJ, Babb JS, Grizzle<br />
WE, Cairns P (2004) Promoter hypermethylation pr<strong>of</strong>ile <strong>of</strong> kidney cancer. Clin <strong>Cancer</strong> Res<br />
10:3972-3979<br />
Eddy AA, Fogo AB (2006) Plasminogen activator inhibitor-1 in chronic kidney disease: evidence and<br />
mechanisms <strong>of</strong> action. J Am Soc Nephrol 17:2999-3012<br />
Ellinger-Ziegelbauer H, Gmuender H, Bandenburg A, Ahr HJ (2008) Prediction <strong>of</strong> a carcinogenic potential<br />
<strong>of</strong> rat hepatocarcinogens using toxicogenomics analysis <strong>of</strong> short-term in vivo studies. Mutat Res<br />
637:23-39<br />
Ellinger-Ziegelbauer H, Stuart B, Wahle B, Bomann W, Ahr HJ (2004) Characteristic expression pr<strong>of</strong>iles<br />
induced by genotoxic carcinogens in rat liver. Toxicol Sci 77:19-34<br />
Ellinger-Ziegelbauer H, Stuart B, Wahle B, Bomann W, Ahr HJ (2005) Comparison <strong>of</strong> the expression<br />
pr<strong>of</strong>iles induced by genotoxic and nongenotoxic carcinogens in rat liver. Mutat Res 575:61-84<br />
Ellis M, Davis N, Coop A, Liu M, Schumaker L, Lee RY, Srikanchana R, Russell CG, Singh B, Miller WR,<br />
Stearns V, Pennanen M, Tsangaris T, Gallagher A, Liu A, Zwart A, Hayes DF, Lippman ME,<br />
Wang Y, Clarke R (2002) Development and validation <strong>of</strong> a method for using breast core needle<br />
biopsies for gene expression microarray analyses. Clin <strong>Cancer</strong> Res 8:1155-1166<br />
Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA<br />
(1996) Laser capture microdissection. Science 274:998-1001<br />
Eustis SL (1989) The sequential development <strong>of</strong> cancer: a morphological perspective. Toxicol Lett 49:267-<br />
281<br />
Everitt JI, Goldsworthy TL, Wolf DC, Walker CL (1995) Hereditary renal cell carcinoma in the Eker rat: a<br />
unique animal model for the study <strong>of</strong> cancer susceptibility. Toxicol Lett 82-83:621-625<br />
Falvella FS, Colombo F, Spinola M, Campiglio M, Pastorino U, Dragani TA (2008) BHLHB3: a candidate<br />
tumor suppressor in lung cancer. Oncogene 27:3761-3764<br />
Faucet V, Pfohl-Leszkowicz A, Dai J, Castegnaro M, Manderville RA (2004) Evidence for covalent DNA<br />
adduction by ochratoxin A following chronic exposure to rat and subacute exposure to pig. Chem<br />
Res Toxicol 17:1289-1296<br />
Fearon ER, Vogelstein B (1990) A genetic model for colorectal tumorigenesis. Cell 61:759-767<br />
Fero ML, Randel E, Gurley KE, Roberts JM, Kemp CJ (1998) The murine gene p27Kip1 is haploinsufficient<br />
for tumour suppression. Nature 396:177-180<br />
Fiala ES, Bobotas G, Kulakis C, Weisburger JH (1976) Separation <strong>of</strong> 1,2-dimethylhydrazine metabolites by<br />
high-pressure liquid chromatography. J Chromatogr 117:181-185<br />
190
Chapter 10: References<br />
Fielden MR, Brennan R, Gollub J (2007) A gene expression biomarker provides early prediction and<br />
mechanistic assessment <strong>of</strong> hepatic tumor induction by nongenotoxic chemicals. Toxicol Sci<br />
99:90-100<br />
Flotow H, Thomas G (1992) Substrate recognition determinants <strong>of</strong> the mitogen-activated 70K S6 kinase<br />
from rat liver. J Biol Chem 267:3074-3078<br />
Foran JA (1997) Principles for the selection <strong>of</strong> doses in chronic rodent bioassays. ILSI Risk Science<br />
Working Group on Dose Selection. Environ Health Perspect 105:18-20<br />
Forbes JM, Coughlan MT, Cooper ME (2008) Oxidative stress as a major culprit in kidney disease in<br />
diabetes. Diabetes 57:1446-1454<br />
Fuchs R, Peraica M (2005) Ochratoxin A in human kidney diseases. Food Addit Contam 22 Suppl 1:53-57<br />
Fukunishi R, Watanabe K, Terashi S, Kawaji K (1971) Shift <strong>of</strong> target organs in cycasin carcinogenesis.<br />
Gann 62:353-358<br />
Furge KA, Dykema K, Petillo D, Westphal M, Zhang Z, Kort EJ, Teh BT (2007) Combining differential<br />
expression, chromosomal and pathway analyses for the molecular characterization <strong>of</strong> renal cell<br />
carcinoma. Can Urol Assoc J 1:S21-27<br />
Gagliano N, Torri C, Donetti E, Grizzi F, Costa F, Bertelli AA, Migliori M, Filippi C, Bedoni M, Panichi V,<br />
Giovannini L, Gioia M (2005) Ochratoxin A-induced renal cortex fibrosis and epithelial-tomesenchymal<br />
transition: molecular mechanisms <strong>of</strong> ochratoxin A-injury and potential effects <strong>of</strong> red<br />
wine. Mol Med 11:30-38<br />
Gekle M, Sauvant C, Schwerdt G (2005) Ochratoxin A at nanomolar concentrations: a signal modulator in<br />
renal cells. Mol Nutr Food Res 49:118-130<br />
Gekle M, Schwerdt G, Freudinger R, Mildenberger S, Wilflingseder D, Pollack V, Dander M, Schramek H<br />
(2000) Ochratoxin A induces JNK activation and apoptosis in MDCK-C7 cells at nanomolar<br />
concentrations. J Pharmacol Exp Ther 293:837-844<br />
Gekle M, Silbernagl S (1994) The role <strong>of</strong> the proximal tubule in ochratoxin A nephrotoxicity in vivo:<br />
toxodynamic and toxokinetic aspects. Ren Physiol Biochem 17:40-49<br />
Gold LI (1999) The role for transforming growth factor-beta (TGF-beta) in human cancer. Crit Rev Oncog<br />
10:303-360<br />
Goley EM, Anderson SJ, Menard C, Chuang E, Lu X, T<strong>of</strong>ilon PJ, Camphausen K (2004) Microarray<br />
analysis in clinical oncology: pre-clinical optimization using needle core biopsies from xenograft<br />
tumors. BMC <strong>Cancer</strong> 4:20<br />
Grab DJ, Zedeck MS (1977) Organ-specific effects <strong>of</strong> the carcinogen methylazoxymethanol related to<br />
metabolism by nicotinamide adenine dinucleotide-dependent dehydrogenases. <strong>Cancer</strong> Res<br />
37:4182-4189<br />
Grollman AP, Shibutani S, Moriya M, Miller F, Wu L, Moll U, Suzuki N, Fernandes A, Rosenquist T,<br />
Medverec Z, Jakovina K, Brdar B, Slade N, Turesky RJ, Goodenough AK, Rieger R, Vukelic M,<br />
Jelakovic B (2007) Aristolochic acid and the etiology <strong>of</strong> endemic (Balkan) nephropathy. Proc Natl<br />
Acad Sci U S A 104:12129-12134<br />
Gulati P, Gaspers LD, Dann SG, Joaquin M, Nobukuni T, Natt F, Kozma SC, Thomas AP, Thomas G<br />
(2008) Amino acids activate mTOR complex 1 via Ca2+/CaM signaling to hVps34. Cell Metab<br />
7:456-465<br />
Habib SL, Phan MN, Patel SK, Li D, Monks TJ, Lau SS (2003) Reduced constitutive 8-oxoguanine-DNA<br />
glycosylase expression and impaired induction following oxidative DNA damage in the tuberin<br />
deficient Eker rat. Carcinogenesis 24:573-582<br />
Hamadeh HK, Bushel PR, Jayadev S, Martin K, DiSorbo O, Sieber S, Bennett L, Tennant R, Stoll R,<br />
Barrett JC, Blanchard K, Paules RS, Afshari CA (2002) Gene expression analysis reveals<br />
chemical-specific pr<strong>of</strong>iles. Toxicol Sci 67:219-231<br />
Hamilton JM (1975) Renal carcinogenesis. Adv <strong>Cancer</strong> Res 22:1-56<br />
Hanahan D, Weinberg RA (2000) The hallmarks <strong>of</strong> cancer. Cell 100:57-70<br />
Hanna SC, Heathcote SA, Kim WY (2008) mTOR pathway in renal cell carcinoma. Expert Rev Anticancer<br />
Ther 8:283-292<br />
Hard GC (1998) Mechanisms <strong>of</strong> chemically induced renal carcinogenesis in the laboratory rodent. Toxicol<br />
Pathol 26:104-112<br />
Hard GC, Khan KN (2004) A contemporary overview <strong>of</strong> chronic progressive nephropathy in the laboratory<br />
rat, and its significance for human risk assessment. Toxicol Pathol 32:171-180<br />
Harrington LS, Findlay GM, Gray A, Tolkacheva T, Wigfield S, Rebholz H, Barnett J, Leslie NR, Cheng S,<br />
Shepherd PR, Gout I, Downes CP, Lamb RF (2004) The TSC1-2 tumor suppressor controls<br />
insulin-PI3K signaling via regulation <strong>of</strong> IRS proteins. J Cell Biol 166:213-223<br />
Haseman J, Melnick R, Tomatis L, Huff J (2001) Carcinogenesis bioassays: study duration and biological<br />
relevance. Food Chem Toxicol 39:739-744<br />
Hengstler JG, Bogdanffy MS, Bolt HM, Oesch F (2003) Challenging dogma: thresholds for genotoxic<br />
carcinogens? The case <strong>of</strong> vinyl acetate. Annu Rev Pharmacol Toxicol 43:485-520<br />
191
Chapter 10: References<br />
Henske EP (2005) Tuberous sclerosis and the kidney: from mesenchyme to epithelium, and beyond.<br />
Pediatr Nephrol 20:854-857<br />
Hino O (2004) Multistep renal carcinogenesis in the Eker (Tsc 2 gene mutant) rat model. Curr Mol Med<br />
4:807-811<br />
Hino O, Kobayashi E, Hirayama Y, Kobayashi T, Kubo Y, Tsuchiya H, Kikuchi Y, Mitani H (1995)<br />
<strong>Molecular</strong> genetic basis <strong>of</strong> renal carcinogenesis in the Eker rat model <strong>of</strong> tuberous sclerosis<br />
(Tsc2). Mol Carcinog 14:23-27<br />
Hino O, Mitani H, Knudson AG (1993) Genetic predisposition to transplacentally induced renal cell<br />
carcinomas in the Eker rat. <strong>Cancer</strong> Res 53:5856-5858<br />
Hirono I (1972) Carcinogenicity and neurotoxicity <strong>of</strong> cycasin with special reference to species differences.<br />
Fed Proc 31:1493-1499<br />
Hirono I (1981) Natural carcinogenic products <strong>of</strong> plant origin. Crit Rev Toxicol 8:235-277<br />
Hirono I, Laqueur GL, Spatz M (1968) Tumor induction in Fischer and Osborne-Mendel rats by a single<br />
administration <strong>of</strong> cycasin. J Natl <strong>Cancer</strong> Inst 40:1003-1010<br />
Hjartaker A, Langseth H, Weiderpass E (2008) Obesity and diabetes epidemics: cancer repercussions.<br />
Adv Exp Med Biol 630:72-93<br />
Horiguchi A, Sumitomo M, Asakuma J, Asano T, Zheng R, Asano T, Nanus DM, Hayakawa M (2006)<br />
Increased serum leptin levels and over expression <strong>of</strong> leptin receptors are associated with the<br />
invasion and progression <strong>of</strong> renal cell carcinoma. J Urol 176:1631-1635<br />
Horvath A, Upham BL, Ganev V, Trosko JE (2002) Determination <strong>of</strong> the epigenetic effects <strong>of</strong> ochratoxin in<br />
a human kidney and a rat liver epithelial cell line. Toxicon 40:273-282<br />
Hotamisligil GS (2006) Inflammation and metabolic disorders. Nature 444:860-867<br />
Howard PC, Eppley RM, Stack ME, Warbritton A, Voss KA, Lorentzen RJ, Kovach RM, Bucci TJ (2001)<br />
Fumonisin b1 carcinogenicity in a two-year feeding study using F344 rats and B6C3F1 mice.<br />
Environ Health Perspect 109 Suppl 2:277-282<br />
Hussain SP, Harris CC (1998) <strong>Molecular</strong> epidemiology <strong>of</strong> human cancer: contribution <strong>of</strong> mutation spectra<br />
studies <strong>of</strong> tumor suppressor genes. <strong>Cancer</strong> Res 58:4023-4037<br />
IARC (2002a) International Agency for Reasearch on <strong>Cancer</strong>. IARC Handbooks <strong>of</strong> <strong>Cancer</strong> Prevention.<br />
Weight Control and Physical Activity. Lyon.<br />
IARC (2002b) International Agency for Research on <strong>Cancer</strong>. Some traditional herbal medicines, some<br />
mycotoxins, naphthalene and styrene. IARC Monogr. Eval. Carcinog. Risks Hum. 82<br />
IARC (2002c) International Agency For Research on <strong>Cancer</strong>: <strong>Cancer</strong> Incidence in Five Continents, Vol.<br />
VIII. Scientific Publication No. 155<br />
ICH (1997) International Conference on Harmonization <strong>of</strong> Technical Requirenments for the Registration on<br />
Pharmaceutical for Human Use. Harmonized Guideline. testing for Carcinogenicity <strong>of</strong><br />
Pharmaceuticals. CPMP/ICH/299/95<br />
Ince PG, Codd GA (2005) Return <strong>of</strong> the cycad hypothesis - does the amyotrophic lateral<br />
sclerosis/parkinsonism dementia complex (ALS/PDC) <strong>of</strong> Guam have new implications for global<br />
health? Neuropathol Appl Neurobiol 31:345-353<br />
Jacinto E, Facchinetti V, Liu D, Soto N, Wei S, Jung SY, Huang Q, Qin J, Su B (2006) SIN1/MIP1<br />
maintains rictor-mTOR complex integrity and regulates Akt phosphorylation and substrate<br />
specificity. Cell 127:125-137<br />
Jacinto E, Loewith R, Schmidt A, Lin S, Ruegg MA, Hall A, Hall MN (2004) Mammalian TOR complex 2<br />
controls the actin cytoskeleton and is rapamycin insensitive. Nat Cell Biol 6:1122-1128<br />
Jaffe AB, Hall A (2002) Rho GTPases in transformation and metastasis. Adv <strong>Cancer</strong> Res 84:57-80<br />
Jefford CE, Irminger-Finger I (2006) Mechanisms <strong>of</strong> chromosome instability in cancers. Crit Rev Oncol<br />
Hematol 59:1-14<br />
Jiang T, Wang Z, Proctor G, Moskowitz S, Liebman SE, Rogers T, Lucia MS, Li J, Levi M (2005) Dietinduced<br />
obesity in C57BL/6J mice causes increased renal lipid accumulation and<br />
glomerulosclerosis via a sterol regulatory element-binding protein-1c-dependent pathway. J Biol<br />
Chem 280:32317-32325<br />
Jones J, Pantuck AJ (2008) Genomics and proteomics in renal cell carcinoma: diagnosis, prognosis, and<br />
treatment selection. Curr Urol Rep 9:9-14<br />
Jozwiak J (2006) Hamartin and tuberin: working together for tumour suppression. Int J <strong>Cancer</strong> 118:1-5<br />
Kaimori JY, Takenaka M, Nakajima H, Hamano T, Horio M, Sugaya T, Ito T, Hori M, Okubo K, Imai E<br />
(2003) Induction <strong>of</strong> glia maturation factor-beta in proximal tubular cells leads to vulnerability to<br />
oxidative injury through the p38 pathway and changes in antioxidant enzyme activities. J Biol<br />
Chem 278:33519-33527<br />
Kamp HG, Eisenbrand G, Janzowski C, Kiossev J, Latendresse JR, Schlatter J, Turesky RJ (2005)<br />
Ochratoxin A induces oxidative DNA damage in liver and kidney after oral dosing to rats. Mol Nutr<br />
Food Res 49:1160-1167<br />
192
Chapter 10: References<br />
Keenan KP, Laroque P, Ballam GC, Soper KA, Dixit R, Mattson BA, Adams SP, Coleman JB (1996) The<br />
effects <strong>of</strong> diet, ad libitum overfeeding, and moderate dietary restriction on the rodent bioassay:<br />
the uncontrolled variable in safety assessment. Toxicol Pathol 24:757-768<br />
Keenan KP, Soper KA, Smith PF, Ballam GC, Clark RL (1995) Diet, overfeeding, and moderate dietary<br />
restriction in control Sprague-Dawley rats: I. Effects on spontaneous neoplasms. Toxicol Pathol<br />
23:269-286<br />
Kenerson H, Dundon TA, Yeung RS (2005) Effects <strong>of</strong> rapamycin in the Eker rat model <strong>of</strong> tuberous<br />
sclerosis complex. Pediatr Res 57:67-75<br />
Kenerson HL, Aicher LD, True LD, Yeung RS (2002) Activated mammalian target <strong>of</strong> rapamycin pathway in<br />
the pathogenesis <strong>of</strong> tuberous sclerosis complex renal tumors. <strong>Cancer</strong> Res 62:5645-5650<br />
Kimura H, Odani S, Suzuki J, Arakawa M, Ono T (1989) <strong>Kidney</strong> fatty acid-binding protein: identification as<br />
alpha 2U-globulin. FEBS Lett 246:101-104<br />
Klaunig JE, Kamendulis LM (2004) The role <strong>of</strong> oxidative stress in carcinogenesis. Annu Rev Pharmacol<br />
Toxicol 44:239-267<br />
Knudson AG, Jr. (1975) Genetics <strong>of</strong> human cancer. Genetics 79 Suppl:305-316<br />
Kobayashi T, Hirayama Y, Kobayashi E, Kubo Y, Hino O (1995) A germline insertion in the tuberous<br />
sclerosis (Tsc2) gene gives rise to the Eker rat model <strong>of</strong> dominantly inherited cancer. Nat Genet<br />
9:70-74<br />
Kobayashi T, Urakami S, Hirayama Y, Yamamoto T, Nishizawa M, Takahara T, Kubo Y, Hino O (1997)<br />
Intragenic Tsc2 somatic mutations as Knudson's second hit in spontaneous and chemically<br />
induced renal carcinomas in the Eker rat model. Jpn J <strong>Cancer</strong> Res 88:254-261<br />
Kohara A, Suzuki T, Honma M, Ohwada T, Hayashi M (2002) Mutagenicity <strong>of</strong> aristolochic acid in the<br />
lambda/lacZ transgenic mouse (MutaMouse). Mutat Res 515:63-72<br />
Kovacs G, Akhtar M, Beckwith BJ, Bugert P, Cooper CS, Delahunt B, Eble JN, Fleming S, Ljungberg B,<br />
Medeiros LJ, Moch H, Reuter VE, Ritz E, Roos G, Schmidt D, Srigley JR, Storkel S, van den Berg<br />
E, Zbar B (1997) The Heidelberg classification <strong>of</strong> renal cell tumours. J Pathol 183:131-133<br />
Krumbiegel G, Hallensleben J, Mennicke WH, Rittmann N, Roth HJ (1987) Studies on the metabolism <strong>of</strong><br />
aristolochic acids I and II. Xenobiotica 17:981-991<br />
Kubo Y, Mitani H, Hino O (1994) Allelic loss at the predisposing gene locus in spontaneous and chemically<br />
induced renal cell carcinomas in the Eker rat. <strong>Cancer</strong> Res 54:2633-2635<br />
Kuiper-Goodman T (1991) Risk assessment <strong>of</strong> ochratoxin A residues in food. IARC Sci Publ:307-320<br />
Kwiatkowski DJ (2003) Rhebbing up mTOR: new insights on TSC1 and TSC2, and the pathogenesis <strong>of</strong><br />
tuberous sclerosis. <strong>Cancer</strong> Biol Ther 2:471-476<br />
Laping NJ, Everitt JI, Frazier KS, Burgert M, Portis MJ, Cadacio C, Gold LI, Walker CL (2007) Tumorspecific<br />
efficacy <strong>of</strong> transforming growth factor-beta RI inhibition in Eker rats. Clin <strong>Cancer</strong> Res<br />
13:3087-3099<br />
Laqueur GL (1964) Carcinogenic Effects Of Cycad Meal And Cycasin, Methylazoxymethanol Glycoside, In<br />
Rats And Effects Of Cycasin In Germfree Rats. Fed Proc 23:1386-1388<br />
Laqueur GL (1968) Toxicology <strong>of</strong> cycasin. Food Cosmet Toxicol 6:577-578<br />
Laqueur GL, Matsumoto H (1966) Neoplasms in female Fischer rats following intraperitoneal injection <strong>of</strong><br />
methylazoxy-methanol. J Natl <strong>Cancer</strong> Inst 37:217-232<br />
Lawrence MC, Jivan A, Shao C, Duan L, Goad D, Zaganjor E, Osborne J, McGlynn K, Stippec S, Earnest<br />
S, Chen W, Cobb MH (2008) The roles <strong>of</strong> MAPKs in disease. Cell Res 18:436-442<br />
Le Hir M, Dubach UC (1982) Peroxisomal and mitochondrial beta-oxidation in the rat kidney: distribution <strong>of</strong><br />
fatty acyl-coenzyme A oxidase and 3-hydroxyacyl-coenzyme A dehydrogenase activities along<br />
the nephron. J Histochem Cytochem 30:441-444<br />
Leighton JK, Brown P, Ellis A, Harlow P, Harrouk W, Pine PS, Robison T, Rosario L, Thompson K (2006)<br />
Workgroup report: Review <strong>of</strong> genomics data based on experience with mock submissions--view<br />
<strong>of</strong> the CDER Pharmacology Toxicology Nonclinical Pharmacogenomics Subcommittee. Environ<br />
Health Perspect 114:573-578<br />
Leithe E, Sirnes S, Omori Y, Rivedal E (2006) Downregulation <strong>of</strong> gap junctions in cancer cells. Crit Rev<br />
Oncog 12:225-256<br />
Leivonen SK, Kahari VM (2007) Transforming growth factor-beta signaling in cancer invasion and<br />
metastasis. Int J <strong>Cancer</strong> 121:2119-2124<br />
Levin BE, Triscari J, Hogan S, Sullivan AC (1987) Resistance to diet-induced obesity: food intake,<br />
pancreatic sympathetic tone, and insulin. Am J Physiol 252:R471-478<br />
Li CM, Guo M, Borczuk A, Powell CA, Wei M, Thaker HM, Friedman R, Klein U, Tycko B (2002) Gene<br />
expression in Wilms' tumor mimics the earliest committed stage in the metanephric<br />
mesenchymal-epithelial transition. Am J Pathol 160:2181-2190<br />
Li Y, Liu Z, Guo X, Shu J, Chen Z, Li L (2006) Aristolochic acid I-induced DNA damage and cell cycle<br />
arrest in renal tubular epithelial cells in vitro. Arch Toxicol 80:524-532<br />
Liang J, Slingerland JM (2003) Multiple roles <strong>of</strong> the PI3K/PKB (Akt) pathway in cell cycle progression. Cell<br />
Cycle 2:339-345<br />
193
Chapter 10: References<br />
Lijinsky W, Kovatch RM, Riggs CW, Walters PT (1988) Dose-response study with N-nitrosomorpholine in<br />
drinking water <strong>of</strong> F-344 rats. <strong>Cancer</strong> Res 48:2089-2095<br />
Lina BA, Kuijpers MH (2004) Toxicity and carcinogenicity <strong>of</strong> acidogenic or alkalogenic diets in rats; effects<br />
<strong>of</strong> feeding NH(4)Cl, KHCO(3) or KCl. Food Chem Toxicol 42:135-153<br />
Liu MC, Maruyama S, Mizuno M, Morita Y, Hanaki S, Yuzawa Y, Matsuo S (2003) The nephrotoxicity <strong>of</strong><br />
Aristolochia manshuriensis in rats is attributable to its aristolochic acids. Clin Exp Nephrol 7:186-<br />
194<br />
Lock EA, Hard GC (2004) Chemically induced renal tubule tumors in the laboratory rat and mouse: review<br />
<strong>of</strong> the NCI/NTP database and categorization <strong>of</strong> renal carcinogens based on mechanistic<br />
information. Crit Rev Toxicol 34:211-299<br />
Lord GM, Hollstein M, Arlt VM, Roufosse C, Pusey CD, Cook T, Schmeiser HH (2004) DNA adducts and<br />
p53 mutations in a patient with aristolochic acid-associated nephropathy. Am J <strong>Kidney</strong> Dis<br />
43:e11-17<br />
Luhe A, Hildebrand H, Bach U, Dingermann T, Ahr HJ (2003) A new approach to studying ochratoxin A<br />
(OTA)-induced nephrotoxicity: expression pr<strong>of</strong>iling in vivo and in vitro employing cDNA<br />
microarrays. Toxicol Sci 73:315-328<br />
Luhe A, Suter L, Ruepp S, Singer T, Weiser T, Albertini S (2005) Toxicogenomics in the pharmaceutical<br />
industry: hollow promises or real benefit? Mutat Res 575:102-115<br />
Ma L, Chen Z, Erdjument-Bromage H, Tempst P, Pandolfi PP (2005a) Phosphorylation and functional<br />
inactivation <strong>of</strong> TSC2 by Erk implications for tuberous sclerosis and cancer pathogenesis. Cell<br />
121:179-193<br />
Ma L, Teruya-Feldstein J, Behrendt N, Chen Z, Noda T, Hino O, Cordon-Cardo C, Pandolfi PP (2005b)<br />
Genetic analysis <strong>of</strong> Pten and Tsc2 functional interactions in the mouse reveals asymmetrical<br />
haploinsufficiency in tumor suppression. Genes Dev 19:1779-1786<br />
Mak BC, Yeung RS (2004) The tuberous sclerosis complex genes in tumor development. <strong>Cancer</strong> Invest<br />
22:588-603<br />
Mally A, Pepe G, Ravoori S, Fiore M, Gupta RC, Dekant W, Mosesso P (2005) Ochratoxin a causes DNA<br />
damage and cytogenetic effects but no DNA adducts in rats. Chem Res Toxicol 18:1253-1261<br />
Malumbres M, Barbacid M (2001) To cycle or not to cycle: a critical decision in cancer. Nat Rev <strong>Cancer</strong><br />
1:222-231<br />
Manning BD, Cantley LC (2007) AKT/PKB signaling: navigating downstream. Cell 129:1261-1274<br />
Manning BD, Logsdon MN, Lipovsky AI, Abbott D, Kwiatkowski DJ, Cantley LC (2005) Feedback inhibition<br />
<strong>of</strong> Akt signaling limits the growth <strong>of</strong> tumors lacking Tsc2. Genes Dev 19:1773-1778<br />
Mantle P, Kulinskaya E, Nestler S (2005) Renal tumourigenesis in male rats in response to chronic dietary<br />
ochratoxin A. Food Addit Contam 22 Suppl 1:58-64<br />
Marin-Kuan M, Nestler S, Verguet C, Bezencon C, Piguet D, Delatour T, Mantle P, Cavin C, Schilter B<br />
(2007) MAPK-ERK activation in kidney <strong>of</strong> male rats chronically fed ochratoxin A at a dose<br />
causing a significant incidence <strong>of</strong> renal carcinoma. Toxicol Appl Pharmacol 224:174-181<br />
Marin-Kuan M, Nestler S, Verguet C, Bezencon C, Piguet D, Mansourian R, Holzwarth J, Grigorov M,<br />
Delatour T, Mantle P, Cavin C, Schilter B (2006) A toxicogenomics approach to identify new<br />
plausible epigenetic mechanisms <strong>of</strong> ochratoxin a carcinogenicity in rat. Toxicol Sci 89:120-134<br />
Martinez-Iglesias O, Ruiz-Llorente L, Sanchez-Martinez R, Garcia L, Zambrano A, Aranda A (2008)<br />
Histone deacetylase inhibitors: mechanism <strong>of</strong> action and therapeutic use in cancer. Clin Transl<br />
Oncol 10:395-398<br />
Matsumoto H, Higa HH (1966) Studies on methylazoxymethanol, the aglycone <strong>of</strong> cycasin: methylation <strong>of</strong><br />
nucleic acids in vitro. Biochem J 98:20C-22C<br />
Matsushima T, Matsumoto H, Shirai A, Sawamura M, Sugimura T (1979) Mutagenicity <strong>of</strong> the naturally<br />
occurring carcinogen cycasin and synthetic methylazoxymethanol conjugates in Salmonella<br />
typhimurium. <strong>Cancer</strong> Res 39:3780-3782<br />
Mayer D, Weber E, Kadenbach B, Bannasch P (1989) Immunocytochemical demonstration <strong>of</strong> cytochrome<br />
c oxidase as a marker for renal oncocytes and oncocytomas. Toxicol Pathol 17:46-49<br />
McDorman KS, Wolf DC (2002) Use <strong>of</strong> the spontaneous Tsc2 knockout (Eker) rat model <strong>of</strong> hereditary<br />
renal cell carcinoma for the study <strong>of</strong> renal carcinogens. Toxicol Pathol 30:675-680<br />
McLaughlin JK, Lipworth L (2000) Epidemiologic aspects <strong>of</strong> renal cell cancer. Semin Oncol 27:115-123<br />
McTiernan A (2005) Obesity and cancer: the risks, science, and potential management strategies.<br />
Oncology (Williston Park) 19:871-881; discussion 881-872, 885-876<br />
Mei N, Arlt VM, Phillips DH, Heflich RH, Chen T (2006) DNA adduct formation and mutation induction by<br />
aristolochic acid in rat kidney and liver. Mutat Res 602:83-91<br />
Mellemgaard A, Lindblad P, Schleh<strong>of</strong>er B, Bergstrom R, Mandel JS, McCredie M, McLaughlin JK, Niwa S,<br />
Odaka N, Pommer W, et al. (1995) International renal-cell cancer study. III. Role <strong>of</strong> weight,<br />
height, physical activity, and use <strong>of</strong> amphetamines. Int J <strong>Cancer</strong> 60:350-354<br />
Mengs U (1987) Acute toxicity <strong>of</strong> aristolochic acid in rodents. Arch Toxicol 59:328-331<br />
194
Chapter 10: References<br />
Mengs U, Lang W, Poch J-A (1982) The Carcinogenic Action <strong>of</strong> Aristolochic Acid in Rats. Archives <strong>of</strong><br />
Toxicology 51:107-119<br />
Mengs U, Stotzem CD (1993) Renal toxicity <strong>of</strong> aristolochic acid in rats as an example <strong>of</strong> nephrotoxicity<br />
testing in routine toxicology. Arch Toxicol 67:307-311<br />
Momparler RL (2003) <strong>Cancer</strong> epigenetics. Oncogene 22:6479-6483<br />
Monro A (1993) How useful are chronic (life-span) toxicology studies in rodents in identifying<br />
pharmaceuticals that pose a carcinogenic risk to humans? Adverse Drug React Toxicol Rev 12:5-<br />
34<br />
Moore B, Beltran L, Carbajal S, Strom S, Traag J, Hursting SD, DiGiovanni J (2008) Dietary Energy<br />
Balance Modulates Signaling through the Akt/Mammalian Target <strong>of</strong> Rapamycin Pathways in Multiple<br />
Epithelial Tissues. <strong>Cancer</strong> Prev Res June:65-76<br />
Morton LD, Youssef AF, Lloyd E, Kiorpes AL, Goldsworthy TL, Fort FL (2002) Evaluation <strong>of</strong> carcinogenic<br />
responses in the Eker rat following short-term exposure to selected nephrotoxins and<br />
carcinogens. Toxicol Pathol 30:559-564<br />
Motzer RJ, Bander NH, Nanus DM (1996) Renal-cell carcinoma. N Engl J Med 335:865-875<br />
Nagasawa HT, Shirota FN, Matsumoto H (1972) Decomposition <strong>of</strong> methylazoxymethanol, the aglycone <strong>of</strong><br />
cycasin, in D 2 O. Nature 236:234-235<br />
Nogueira E (1987) Rat renal carcinogenesis after chronic simultaneous exposure to lead acetate and Nnitrosodiethylamine.<br />
Virchows Arch B Cell Pathol Incl Mol Pathol 53:365-374<br />
Nogueira E, Klimek F, Weber E, Bannasch P (1989) Collecting duct origin <strong>of</strong> rat renal clear cell tumors.<br />
Virchows Arch B Cell Pathol Incl Mol Pathol 57:275-283<br />
Nortier JL, Martinez MC, Schmeiser HH, Arlt VM, Bieler CA, Petein M, Depierreux MF, De Pauw L,<br />
Abramowicz D, Vereerstraeten P, Vanherweghem JL (2000) Urothelial carcinoma associated with<br />
the use <strong>of</strong> a Chinese herb (Aristolochia fangchi). N Engl J Med 342:1686-1692<br />
Nortier JL, Vanherweghem JL (2002) Renal interstitial fibrosis and urothelial carcinoma associated with the<br />
use <strong>of</strong> a Chinese herb (Aristolochia fangchi). Toxicology 181-182:577-580<br />
Notman J, Tan QH, Zedeck MS (1982) Inhibition <strong>of</strong> methylazoxymethanol-induced intestinal tumors in the<br />
rat by pyrazole with paradoxical effects on skin and kidney. <strong>Cancer</strong> Res 42:1774-1780<br />
NTP (1989) Toxicology and Carcinogenesis Studies <strong>of</strong> Ochratoxin A (CAS No. 303-47-9) in F344/N Rats<br />
(Gavage Studies). Natl Toxicol Program Tech Rep Ser 358:1-142<br />
NTP US (1992) U.S. NTP. Specification for the Conduct <strong>of</strong> Studies to Evaluate the Toxic and Carcinogenic<br />
Potential <strong>of</strong> Chemical, Biological and Physical Agents in Laboratory Animals for the National Toxicology<br />
Program (NTP).<br />
Nunez NP, Oh WJ, Rozenberg J, Perella C, Anver M, Barrett JC, Perkins SN, Berrigan D, Moitra J,<br />
Varticovski L, Hursting SD, Vinson C (2006) Accelerated tumor formation in a fatless mouse with<br />
type 2 diabetes and inflammation. <strong>Cancer</strong> Res 66:5469-5476<br />
Nygaard V, Holden M, Loland A, Langaas M, Myklebost O, Hovig E (2005) Limitations <strong>of</strong> mRNA<br />
amplification from small-size cell samples. BMC Genomics 6:147<br />
O'Brien CA, Pollett A, Gallinger S, Dick JE (2007) A human colon cancer cell capable <strong>of</strong> initiating tumour<br />
growth in immunodeficient mice. Nature 445:106-110<br />
O'Brien E, Dietrich DR (2005) Ochratoxin A: the continuing enigma. Crit Rev Toxicol 35:33-60<br />
O'Gara RW, Brown JM, Whiting MG (1964) Induction Of Hepatic And Renal Tumors By Topical Application<br />
Of Aqueous Extract Of Cycad Nut To Artificial Skin Ulcers In Mice. Fed Proc 23:1382-1383<br />
OECD (1981) OECD guidelines for testing <strong>of</strong> chemicals. Paris:Organisation for Economic Co-operation<br />
and Development.<br />
Oesch F, Herrero ME, Hengstler JG, Lohmann M, Arand M (2000) Metabolic detoxification: implications for<br />
thresholds. Toxicol Pathol 28:382-387<br />
Ogden CL, Yanovski SZ, Carroll MD, Flegal KM (2007) The epidemiology <strong>of</strong> obesity. Gastroenterology<br />
132:2087-2102<br />
Ono H, Pocai A, Wang Y, Sakoda H, Asano T, Backer JM, Schwartz GJ, Rossetti L (2008) Activation <strong>of</strong><br />
hypothalamic S6 kinase mediates diet-induced hepatic insulin resistance in rats. J Clin Invest<br />
118:2959-2968<br />
Pantuck AJ, Thomas G, Belldegrun AS, Figlin RA (2006) Mammalian target <strong>of</strong> rapamycin inhibitors in renal<br />
cell carcinoma: current status and future applications. Semin Oncol 33:607-613<br />
Parsa AT, Waldron JS, Panner A, Crane CA, Parney IF, Barry JJ, Cachola KE, Murray JC, Tihan T,<br />
Jensen MC, Mischel PS, Stokoe D, Pieper RO (2007) Loss <strong>of</strong> tumor suppressor PTEN function<br />
increases B7-H1 expression and immunoresistance in glioma. Nat Med 13:84-88<br />
Pavlovich CP, Schmidt LS (2004) Searching for the hereditary causes <strong>of</strong> renal-cell carcinoma. Nat Rev<br />
<strong>Cancer</strong> 4:381-393<br />
Perona R (2006) Cell signalling: growth factors and tyrosine kinase receptors. Clin Transl Oncol 8:77-82<br />
Peto R, Gray R, Brantom P, Grasso P (1991) Dose and time relationships for tumor induction in the liver<br />
and esophagus <strong>of</strong> 4080 inbred rats by chronic ingestion <strong>of</strong> N-nitrosodiethylamine or Nnitrosodimethylamine.<br />
<strong>Cancer</strong> Res 51:6452-6469<br />
195
Chapter 10: References<br />
Pfau W, Schmeiser HH, Wiessler M (1990) Aristolochic acid binds covalently to the exocyclic amino group<br />
<strong>of</strong> purine nucleotides in DNA. Carcinogenesis 11:313-319<br />
Pfau W, Schmeiser HH, Wiessler M (1991) N6-adenyl arylation <strong>of</strong> DNA by aristolochic acid II and a<br />
synthetic model for the putative proximate carcinogen. Chem Res Toxicol 4:581-586<br />
Rached E, Pfeiffer E, Dekant W, Mally A (2006) Ochratoxin A: apoptosis and aberrant exit from mitosis<br />
due to perturbation <strong>of</strong> microtubule dynamics? Toxicol Sci 92:78-86<br />
Radulovic S, Bjelogrlic SK (2007) Sunitinib, sorafenib and mTOR inhibitors in renal cancer. J Buon 12<br />
Suppl 1:S151-162<br />
Raghavan N, Amaratunga D, Nie AY, McMillian M (2005) Class prediction in toxicogenomics. J Biopharm<br />
Stat 15:327-341<br />
Rahimtula AD, Chong X (1991) Alterations in calcium homeostasis as a possible cause <strong>of</strong> ochratoxin A<br />
nephrotoxicity. IARC Sci Publ:207-214<br />
Rasonyi T, Schlatter J, Dietrich DR (1999) The role <strong>of</strong> alpha2u-globulin in ochratoxin A induced renal<br />
toxicity and tumors in F344 rats. Toxicol Lett 104:83-92<br />
Ravitz MJ, Chen L, Lynch M, Schmidt EV (2007) c-myc Repression <strong>of</strong> TSC2 contributes to control <strong>of</strong><br />
translation initiation and Myc-induced transformation. <strong>Cancer</strong> Res 67:11209-11217<br />
Ravussin E (2005) Physiology. A NEAT way to control weight? Science 307:530-531<br />
Ravussin E, Danforth E, Jr. (1999) Beyond sloth--physical activity and weight gain. Science 283:184-185<br />
Renehan AG, Frystyk J, Flyvbjerg A (2006) Obesity and cancer risk: the role <strong>of</strong> the insulin-IGF axis.<br />
Trends Endocrinol Metab 17:328-336<br />
Renehan AG, Roberts DL, Dive C (2008) Obesity and cancer: pathophysiological and biological<br />
mechanisms. Arch Physiol Biochem 114:71-83<br />
Rennebeck G, Kleymenova EV, Anderson R, Yeung RS, Artzt K, Walker CL (1998) Loss <strong>of</strong> function <strong>of</strong> the<br />
tuberous sclerosis 2 tumor suppressor gene results in embryonic lethality characterized by<br />
disrupted neuroepithelial growth and development. Proc Natl Acad Sci U S A 95:15629-15634<br />
Rockett JC, Hellmann GM (2004) Confirming microarray data-is it really necessary? Genomics 83:541-549<br />
Rosner M, H<strong>of</strong>er K, Kubista M, Hengstschlager M (2003) Cell size regulation by the human TSC tumor<br />
suppressor proteins depends on PI3K and FKBP38. Oncogene 22:4786-4798<br />
Sarbassov DD, Ali SM, Kim DH, Guertin DA, Latek RR, Erdjument-Bromage H, Tempst P, Sabatini DM<br />
(2004) Rictor, a novel binding partner <strong>of</strong> mTOR, defines a rapamycin-insensitive and raptorindependent<br />
pathway that regulates the cytoskeleton. Curr Biol 14:1296-1302<br />
Sarbassov DD, Guertin DA, Ali SM, Sabatini DM (2005) Phosphorylation and regulation <strong>of</strong> Akt/PKB by the<br />
rictor-mTOR complex. Science 307:1098-1101<br />
Sauvant C, Holzinger H, Mildenberger S, Gekle M (2005) Exposure to nephrotoxic ochratoxin A enhances<br />
collagen secretion in human renal proximal tubular cells. Mol Nutr Food Res 49:31-37<br />
Schmeiser HH, Janssen JW, Lyons J, Scherf HR, Pfau W, Buchmann A, Bartram CR, Wiessler M (1990)<br />
Aristolochic acid activates ras genes in rat tumors at deoxyadenosine residues. <strong>Cancer</strong> Res<br />
50:5464-5469<br />
Schmeiser HH, Pool BL, Wiessler M (1984) Mutagenicity <strong>of</strong> the two main components <strong>of</strong> commercially<br />
available carcinogenic aristolochic acid in Salmonella typhimurium. <strong>Cancer</strong> Lett 23:97-101<br />
Schmeiser HH, Scherf HR, Wiessler M (1991) Activating mutations at codon 61 <strong>of</strong> the c-Ha-ras gene in<br />
thin-tissue sections <strong>of</strong> tumors induced by aristolochic acid in rats and mice. <strong>Cancer</strong> Lett 59:139-<br />
143<br />
Schmeiser HH, Schoepe KB, Wiessler M (1988) DNA adduct formation <strong>of</strong> aristolochic acid I and II in vitro<br />
and in vivo. Carcinogenesis 9:297-303<br />
Schmutte C, Fishel R (1999) Genomic instability: first step to carcinogenesis. Anticancer Res 19:4665-<br />
4696<br />
Schroeder A, Mueller O, Stocker S, Salowsky R, Leiber M, Gassmann M, Lightfoot S, Menzel W, Granzow<br />
M, Ragg T (2006) The RIN: an RNA integrity number for assigning integrity values to RNA<br />
measurements. BMC Mol Biol 7:3<br />
Schutze K, Posl H, Lahr G (1998) Laser micromanipulation systems as universal tools in cellular and<br />
molecular biology and in medicine. Cell Mol Biol (Noisy-le-grand) 44:735-746<br />
Sebekova K, Wagner Z, Schupp N, Boor P (2007) Genomic damage and malignancy in end-stage renal<br />
failure: do advanced glycation end products contribute? <strong>Kidney</strong> Blood Press Res 30:56-66<br />
Sedgwick B, Bates PA, Paik J, Jacobs SC, Lindahl T (2007) Repair <strong>of</strong> alkylated DNA: recent advances.<br />
DNA Repair (Amst) 6:429-442<br />
Semenza GL (2002) HIF-1 and tumor progression: pathophysiology and therapeutics. Trends Mol Med<br />
8:S62-67<br />
Sen B, Wolf DC, Hester SD (2004) The transcriptional pr<strong>of</strong>ile <strong>of</strong> the kidney in Tsc2 heterozygous mutant<br />
Long Evans (Eker) rats compared to wild-type. Mutat Res 549:213-224<br />
Shah OJ, Wang Z, Hunter T (2004) Inappropriate activation <strong>of</strong> the TSC/Rheb/mTOR/S6K cassette induces<br />
IRS1/2 depletion, insulin resistance, and cell survival deficiencies. Curr Biol 14:1650-1656<br />
196
Chapter 10: References<br />
Shank RC, Magee PN (1967) Similarities between the biochemical actions <strong>of</strong> cycasin and<br />
dimethylnitrosamine. Biochem J 105:521-527<br />
Shibutani S, Takeshita M, Grollman AP (1991) Insertion <strong>of</strong> specific bases during DNA synthesis past the<br />
oxidation-damaged base 8-oxodG. Nature 349:431-434<br />
Sieber SM, Correa P, Dalgard DW, McIntire KR, Adamson RH (1980) Carcinogenicity and hepatotoxicity<br />
<strong>of</strong> cycasin and its aglycone methylazoxymethanol acetate in nonhuman primates. J Natl <strong>Cancer</strong><br />
Inst 65:177-189<br />
Singh SC (1980) Deformed dendrites and reduced spine numbers on ectopic neurones in the<br />
hippocampus <strong>of</strong> rats exposed to methylazoxymethanol-acetate. A Golgi-Cox study. Acta<br />
Neuropathol 49:193-198<br />
Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J, Dirks PB (2003) Identification <strong>of</strong> a cancer<br />
stem cell in human brain tumors. <strong>Cancer</strong> Res 63:5821-5828<br />
Sionov RV, Haupt Y (1999) The cellular response to p53: the decision between life and death. Oncogene<br />
18:6145-6157<br />
Sohn OS, Fiala ES, Requeijo SP, Weisburger JH, Gonzalez FJ (2001) Differential effects <strong>of</strong> CYP2E1<br />
status on the metabolic activation <strong>of</strong> the colon carcinogens azoxymethane and<br />
methylazoxymethanol. <strong>Cancer</strong> Res 61:8435-8440<br />
Spatz M (1964) Carcinogenic Effect Of Cycad Meal In Guinea Pigs. Fed Proc 23:1384-1385<br />
Spatz M, Smith DW, McDaniel EG, Laqueur GL (1967) Role <strong>of</strong> intestinal microorganisms in determining<br />
cycasin toxicity. Proc Soc Exp Biol Med 124:691-697<br />
Spiess AN, Mueller N, Ivell R (2003) Amplified RNA degradation in T7-amplification methods results in<br />
biased microarray hybridizations. BMC Genomics 4:44<br />
Steiner G, Suter L, Boess F, Gasser R, de Vera MC, Albertini S, Ruepp S (2004) Discriminating different<br />
classes <strong>of</strong> toxicants by transcript pr<strong>of</strong>iling. Environ Health Perspect 112:1236-1248<br />
Stemmer K, Ellinger-Ziegelbauer H, Ahr HJ, Dietrich DR (2007) Carcinogen-specific gene expression<br />
pr<strong>of</strong>iles in short-term treated Eker and wild-type rats indicative <strong>of</strong> pathways involved in renal<br />
tumorigenesis. <strong>Cancer</strong> Res 67:4052-4068<br />
Stemmer K, Ellinger-Ziegelbauer H, Lotz K, Ahr HJ, Dietrich DR (2006) Establishment <strong>of</strong> a protocol for the<br />
gene expression analysis <strong>of</strong> laser microdissected rat kidney samples with affymetrix genechips.<br />
Toxicol Appl Pharmacol 217:134-142<br />
Stiborova M, Frei E, Arlt VM, Schmeiser HH (2008) Metabolic activation <strong>of</strong> carcinogenic aristolochic acid, a<br />
risk factor for Balkan endemic nephropathy. Mutat Res 658:55-67<br />
Stiborova M, Frei E, Sopko B, Sopkova K, Markova V, Lankova M, Kumstyrova T, Wiessler M, Schmeiser<br />
HH (2003) Human cytosolic enzymes involved in the metabolic activation <strong>of</strong> carcinogenic<br />
aristolochic acid: evidence for reductive activation by human NAD(P)H:quinone oxidoreductase.<br />
Carcinogenesis 24:1695-1703<br />
Stoev SD (1998) The role <strong>of</strong> ochratoxin A as a possible cause <strong>of</strong> Balkan endemic nephropathy and its risk<br />
evaluation. Vet Hum Toxicol 40:352-360<br />
Storkel S, Eble JN, Adlakha K, Amin M, Blute ML, Bostwick DG, Darson M, Delahunt B, Iczkowski K<br />
(1997) Classification <strong>of</strong> renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le<br />
<strong>Cancer</strong> (UICC) and the American Joint Committee on <strong>Cancer</strong> (AJCC). <strong>Cancer</strong> 80:987-989<br />
Studer-Rohr I, Schlatter J, Dietrich DR (2000) Kinetic parameters and intraindividual fluctuations <strong>of</strong><br />
ochratoxin A plasma levels in humans. Arch Toxicol 74:499-510<br />
Sun D, Feng J, Dai C, Sun L, Jin T, Ma J, Wang L (2006) Role <strong>of</strong> peritubular capillary loss and hypoxia in<br />
progressive tubulointerstitial fibrosis in a rat model <strong>of</strong> aristolochic acid nephropathy. Am J Nephrol<br />
26:363-371<br />
Swenberg JA (1993) Alpha 2u-globulin nephropathy: review <strong>of</strong> the cellular and molecular mechanisms<br />
involved and their implications for human risk assessment. Environ Health Perspect 101 Suppl<br />
6:39-44<br />
Swenberg JA, Fryar-Tita E, Jeong YC, Boysen G, Starr T, Walker VE, Albertini RJ (2008) Biomarkers in<br />
toxicology and risk assessment: informing critical dose-response relationships. Chem Res Toxicol<br />
21:253-265<br />
Swenberg JA, Lehman-McKeeman LD (1999) alpha 2-Urinary globulin-associated nephropathy as a<br />
mechanism <strong>of</strong> renal tubule cell carcinogenesis in male rats. IARC Sci Publ:95-118<br />
Szakacs G, Paterson JK, Ludwig JA, Booth-Genthe C, Gottesman MM (2006) Targeting multidrug<br />
resistance in cancer. Nat Rev Drug Discov 5:219-234<br />
Takahashi M, Teh BT, Kanayama HO (2006) Elucidation <strong>of</strong> the molecular signatures <strong>of</strong> renal cell<br />
carcinoma by gene expression pr<strong>of</strong>iling. J Med Invest 53:9-19<br />
Takahashi Y, Okimura Y, Mizuno I, Takahashi T, Kaji H, Uchiyama T, Abe H, Chihara K (1996) Leptin<br />
induces tyrosine phosphorylation <strong>of</strong> cellular proteins including STAT-1 in human renal<br />
adenocarcinoma cells, ACHN. Biochem Biophys Res Commun 228:859-864<br />
Tan MH, Rogers CG, Cooper JT, Ditlev JA, Maatman TJ, Yang X, Furge KA, Teh BT (2004) Gene<br />
expression pr<strong>of</strong>iling <strong>of</strong> renal cell carcinoma. Clin <strong>Cancer</strong> Res 10:6315S-6321S<br />
197
Chapter 10: References<br />
Thomas GV (2006) mTOR and cancer: reason for dancing at the crossroads? Curr Opin Genet Dev 16:78-<br />
84<br />
Todd R, Wong DT (1999) Oncogenes. Anticancer Res 19:4729-4746<br />
Tran H, Brunet A, Griffith EC, Greenberg ME (2003) The many forks in FOXO's road. Sci STKE 2003:RE5<br />
Tsujimura K, Asamoto M, Suzuki S, Hokaiwado N, Ogawa K, Shirai T (2006) Prediction <strong>of</strong> carcinogenic<br />
potential by a toxicogenomic approach using rat hepatoma cells. <strong>Cancer</strong> Sci 97:1002-1010<br />
Turk PW, Laayoun A, Smith SS, Weitzman SA (1995) DNA adduct 8-hydroxyl-2'-deoxyguanosine (8hydroxyguanine)<br />
affects function <strong>of</strong> human DNA methyltransferase. Carcinogenesis 16:1253-<br />
1255<br />
Uchida E, Hirono I (1981) Effect <strong>of</strong> phenobarbital on the development <strong>of</strong> neoplastic lesions in the liver <strong>of</strong><br />
cycasin-treated rats. J <strong>Cancer</strong> Res Clin Oncol 100:231-238<br />
Um SH, Frigerio F, Watanabe M, Picard F, Joaquin M, Sticker M, Fumagalli S, Allegrini PR, Kozma SC,<br />
Auwerx J, Thomas G (2004) Absence <strong>of</strong> S6K1 protects against age- and diet-induced obesity<br />
while enhancing insulin sensitivity. Nature 431:200-205<br />
van Delft JH, van Agen E, van Breda SG, Herwijnen MH, Staal YC, Kleinjans JC (2004) Discrimination <strong>of</strong><br />
genotoxic from non-genotoxic carcinogens by gene expression pr<strong>of</strong>iling. Carcinogenesis 25:1265-<br />
1276<br />
Van Gaal LF, Mertens IL, De Block CE (2006) Mechanisms linking obesity with cardiovascular disease.<br />
Nature 444:875-880<br />
Van Gelder RN, von Zastrow ME, Yool A, Dement WC, Barchas JD, Eberwine JH (1990) Amplified RNA<br />
synthesized from limited quantities <strong>of</strong> heterogeneous cDNA. Proc Natl Acad Sci U S A 87:1663-<br />
1667<br />
Vanherweghem JL, Depierreux M, Tielemans C, Abramowicz D, Dratwa M, Jadoul M, Richard C,<br />
Vandervelde D, Verbeelen D, Vanhaelen-Fastre R, et al. (1993) Rapidly progressive interstitial<br />
renal fibrosis in young women: association with slimming regimen including Chinese herbs.<br />
Lancet 341:387-391<br />
Walker C, Goldsworthy TL, Wolf DC, Everitt J (1992) Predisposition to renal cell carcinoma due to<br />
alteration <strong>of</strong> a cancer susceptibility gene. Science 255:1693-1695<br />
Wang Z, Jiang T, Li J, Proctor G, McManaman JL, Lucia S, Chua S, Levi M (2005) Regulation <strong>of</strong> renal lipid<br />
metabolism, lipid accumulation, and glomerulosclerosis in FVBdb/db mice with type 2 diabetes.<br />
Diabetes 54:2328-2335<br />
Ward JM (2007) The Two-Year Rodent Carcinogenesis Bioassay . Will It Survive? J Toxicol Pathol 20:13-<br />
19<br />
Weisburger EK (1978) Mechanisms <strong>of</strong> chemical carcinogenesis. Annu Rev Pharmacol Toxicol 18:395-415<br />
Whiting MG (1963) Toxicity <strong>of</strong> Cycads. Economic Bot 17:270-329<br />
Williams GM (1999) Chemically induced preneoplastic lesions in rodents as indicators <strong>of</strong> carcinogenic<br />
activity. IARC Sci Publ:185-202<br />
Williams GM, Iatropoulos MJ, Jeffrey AM (1995) Thresholds for DNA-Reactive (Genotoxic) Organic<br />
Carcinogens. J Toxicol Pathol 18:69-77<br />
Williams JN, Jr., Laqueur GL (1965) Response Of Liver Nucleic Acids And Lipids In Rats Fed Cycas<br />
Circinalis L. Endosperm Or Cycasin. Proc Soc Exp Biol Med 118:1-4<br />
Wolf DC, Goldsworthy TL, Donner EM, Harden R, Fitzpatrick B, Everitt JI (1998) Estrogen treatment<br />
enhances hereditary renal tumor development in Eker rats. Carcinogenesis 19:2043-2047<br />
Wolf DC, Goldsworthy TL, Janszen DB, Harden R, Donner EM, David CS, Everitt JI (2000) Promotion by<br />
sodium barbital induces early development but does not increase the multiplicity <strong>of</strong> hereditary<br />
renal tumors in Eker rats. Carcinogenesis 21:1553-1558<br />
Wullschleger S, Loewith R, Hall MN (2006) TOR signaling in growth and metabolism. Cell 124:471-484<br />
Yeung RS, Xiao GH, Everitt JI, Jin F, Walker CL (1995) Allelic loss at the tuberous sclerosis 2 locus in<br />
spontaneous tumors in the Eker rat. Mol Carcinog 14:28-36<br />
Yeung RS, Xiao GH, Jin F, Lee WC, Testa JR, Knudson AG (1994) Predisposition to renal carcinoma in<br />
the Eker rat is determined by germ-line mutation <strong>of</strong> the tuberous sclerosis 2 (TSC2) gene. Proc<br />
Natl Acad Sci U S A 91:11413-11416<br />
Young AN, Master VA, Paner GP, Wang MD, Amin MB (2008) Renal epithelial neoplasms: diagnostic<br />
applications <strong>of</strong> gene expression pr<strong>of</strong>iling. Adv Anat Pathol 15:28-38<br />
Zedeck MS, Brown GB (1977) Methylation <strong>of</strong> intestinal and hepatic DNA in rats treated with<br />
methylazoxymethanol acetate. <strong>Cancer</strong> 40:2580-2583<br />
Zedeck MS, Sternberg SS, Poynter RW, McGowan J (1970) Biochemical and pathological effects <strong>of</strong><br />
methylazoxymethanol acetate, a potent carcinogen. <strong>Cancer</strong> Res 30:801-812<br />
Zhang H, Cifone MA, Murli H, Erexson GL, Mecchi MS, Lawlor TE (2004) Application <strong>of</strong> simplified in vitro<br />
screening tests to detect genotoxicity <strong>of</strong> aristolochic acid. Food Chem Toxicol 42:2021-2028<br />
198
Erklärung<br />
Die vorliegende Arbeit wurde ohne unzulässige Hilfe Dritter und ohne Benutzung anderer<br />
als der angegebenen Hilfsmittel angefertigt. Die aus anderen Quellen direkt oder indirekt<br />
übernommenen Daten und Konzepte sind unter Angabe der Quelle gekennzeichnet.<br />
Weitere Personen, insbesondere Promotionsberater, waren an der inhaltlich materiellen<br />
Erstellung dieser Arbeit nicht beteiligt. Die Arbeit wurde weder im In- noch im Ausland in<br />
gleicher oder ähnlicher Form einer anderen Prüfungsbehörde vorgelegt.<br />
Eigenabgrenzung/ Kooperationen<br />
• Die Hybridisierung der Microarrays (siehe Kapitel 3, 4, 5 und 6) erfolgte in der<br />
Abteilung <strong>Molecular</strong> and Special Toxicology, Bayer Healthcare AG in Wuppertal.<br />
• Sämtliche Paraffinschnitte wurden von Mitarbeitern der Abteilung Non-Clinical Drug<br />
Safety bei Boehringer Ingelheim in Biberach hergestellt.<br />
• LC-MS/MS Analysen zur Detektion von O6MG Addukten (Kapitel 4) wurden von<br />
Mitarbeitern von Pr<strong>of</strong>. Dr. James A. Swenberg am Department <strong>of</strong> Environmental<br />
Sciences and Engineering, University <strong>of</strong> North Carolina, Chapel Hill, USA<br />
durchgeführt.<br />
• Die Exposition der Ratten mit fetthaltiger Diät, sowie die Messung der Blutparameter<br />
(Kapitel 7) wurde von Mitarbeitern von Pr<strong>of</strong>. Dr. Matthias Tschöp, im Department <strong>of</strong><br />
Psychiatry, Obesity Research Centre in Cincinnati, Ohio, USA durchgeführt.<br />
Alle weiteren Leistungen wurden, s<strong>of</strong>ern nicht explizit angemerkt, von mir selbst<br />
durchgeführt.<br />
Konstanz im Oktober 2008<br />
Kerstin Stemmer<br />
199
Danksagung<br />
Meinen herzlichsten Dank möchte ich all denen aussprechen, die mich auf dem Weg zur<br />
Entstehung dieser Doktorarbeit unterstützt haben. Mein besonderer Dank gilt hierbei:<br />
• Pr<strong>of</strong>. Dr. Daniel Dietrich für die Überlassung des Promotionsthemas, seine stete<br />
wissenschaftliche und menschliche Unterstützung, seine Förderung,<br />
Diskussionsbereitschaft und Offenheit gegenüber neuen Projektideen. Insgesamt für<br />
eine schöne und sehr lehrreiche Zeit in der AG-Dietrich, an die ich immer gerne<br />
zurückdenken werde.<br />
• Pr<strong>of</strong>. Dr. Christ<strong>of</strong> Hauck für die Übernahme des Koreferates und sein Interesse an<br />
dieser Arbeit.<br />
• Dr. Dr. Hans-Jürgen Ahr, Dr. Heidrun-Ellinger Ziegelbauer und Mitarbeitern für eine<br />
<strong>of</strong>fene und nette Projektkooperation, ihre stete Diskussionsbereitschaft auch nach<br />
Projektabschluss, das Einlernen in die komplexe Auswertung von Microarray Analysen,<br />
nicht zuletzt für die herzliche Gastfreundschaft und Unterstützung während vieler<br />
mehrwöchiger Aufenhalte bei der Bayer HealthCare AG in Wuppertal.<br />
• Pr<strong>of</strong>. Dr. Matthias Tschöp, Dr. Paul Pfluger und Dr. Diego Perez-Tilve für Ihre<br />
Unterstützung und Begeisterung, ein neues, gemeinsames Projekt auf den Weg zu<br />
bringen, ihre Gastfreundschaft, und die Überlassung eines Laborplatzes während<br />
mehrer Aufenthalte am Obesity Research Center in Cincinnati, sowie für den Einblick in<br />
die Kunst des Cabrewings. Besonders bei Paul möchte ich mich für die Korrekturen und<br />
die konstruktive Kritik an dieser Arbeit bedanken. Nicht zuletzt für unzählige<br />
wissenschaftliche Diskussionen, die so manch gute Idee und Fortschritt mit sich<br />
brachten. Erin Grant möchte ich herzlichst für die Englisch-Korrekturen danken.<br />
• Dr. Ulrich Deschl, Dr. Thomas Nolte und Mitarbeitern, für die Ermöglichung eines 2wöchigen<br />
Arbeitsaufenthaltes in der Abteilung für Non-Clinical Drug Safety bei<br />
Boehringer Ingelheim in Biberach. Ohne diesen waere mir die Planung und<br />
Durchführung der in vivo Studie für diese Doktorarbeit weitaus schwerer gefallen. Nicht<br />
zuletzt bedanke ich mich für das Anfertigen zahlreicher Paraffinschnitte.<br />
• Pr<strong>of</strong>. Dr. James A. Swenberg und Mitarbeitern vom Department <strong>of</strong> Environmental<br />
Sciences and Engineering, University <strong>of</strong> North Carolina, Chapel Hill, danke ich für die<br />
Zusammenarbeit und die Quantifizierung von DNA-Addukten mittels LC-MS/MS.<br />
• Allen ehemaligen und jetzigen Mitarbeiter/-innen der Arbeitsgruppe Dietrich für das<br />
nette und familiäre Arbeitsklima. Besonders Tanja Lampertsdörfer und Gudrun von<br />
Scheven möchte ich für die tatkräftige Unterstützung beim in vivo Versuch und bei der<br />
Verabreichung tausender Schlundsonden danken. Beiden, sowie Dan Dietrich, Alex<br />
Heussner und Evelyn O’Brien gilt mein Dank für die Mithilfe bei den Sektionen.<br />
Und vor allem meiner Familie und meinem Freund Paul für ihre stete Unterstützung in allen<br />
Lebenslagen, ihren stetigen Beistand, und ihre Ruhe während aller Höhen und Tiefen dieser<br />
Doktorarbeit.<br />
200