Tsuji et. al. amniotic membrane-derived stem cell (2009/205260-R3 ...

Tsuji et. al. amniotic membrane-derived stem cell 

(2009/205260-R3) 

Supplemental materials 

Figure Legends for supplemental material 

Online Fig. I. Flow cytometric analysis of hAMCs 

(A) Flow cytometric analysis of hAMCs with FITC-coupled antibodies against 

human surface antigens. Vertical axis denotes cellular number and horizontal 

axis denotes fluoroscopic intensity (full scale = 4 log). Gray histogram 

corresponds to FITC positive population. (B) Summary of flow cytometric 

analysis of hAMCs. We measured the shift of peak histogram value of flow 

cytometric data. We defined each result by the following criteria (-, negative; ±, 

0.5 log shift from negative control; +, 1 log shift from negative control; ++, 2 

log shifts from negative control; +++, 3 log shifts from negative control) (n=4). 

Online Fig. II. Laser confocal microscopic view of immunocytochemistry of 

differentiated hAMCs 1. 

EGFP-labeled hAMCs (A; green) were co-cultivated with MitoTracker® Redstained 

murine cardiomyocytes (B; Mit, red). Nuclei were stained with DAPI 

(blue). Merged image of DAPI, EGFP, Mit is shown in C. There is no overlap 

of Mit and EGFP. Both Mit positive cell and EGFP positive cells express 

cardiac troponin-I (D,E, Trop-I, white). On the other hand the MitoTracker® 

Red-labeled hAMCs (G, red) were co-cultivated with EGFP-transgenic murine 

cardiomyocytes (F, green). Merged image of DAPI, EGFP, Mit is shown in H. 

There is no overlap of Mit and EGFP. Both Mit positive cells and EGFP 

positive cells express Trop-I (I,J, white). Staining pattern of Trop-I showed 

clear striation (D). Scale bar denotes 20 !m 

Online Fig. III. Laser confocal microscopic view of immunocytochemistry of 

transdifferentiated hAMCs 2. 

Suppl 1



Unmerged images of Fig. 2F are shown. Please see detail in the legend of 

Fig. 2. Striation pattern of differentiated hAMCs in the white box of panel D 

were expanded and shown in panel F-H. Clear striation pattern of !-actinin 

(red; F) is observed. !-actinin and EGFP-staining (green;G) were observed 

alternately in striated manner, suggesting !-actinin is expressed in EGFP 

positive cell. Scale bar denotes 20!m 

Online Fig. IV. Haemodynamic parameter in myocardial infarction model of 

nude rat in vivo 

Measured LV parameters by echocardiogram are averaged and shown at 2 

weeks and 4 weeks after the myocardial infarction (MI). (A) The left 

ventricular end-diastolic dimension (LVEDd) and (B) end-systolic dimension 

(LVESd) tended to improve; however, there was no statistical significance. 

There was no difference in the diameter of (C) the anterior left ventricular wall 

thickness (AW), and (D) posterior left ventricular wall thickness (PW). 

Measured (E) left ventricular systolic pressure (LVSP), (F) left ventricular enddiastolic 

pressure (LVEDP), (G) maximum positive developed left ventricular 

pressure (LV dP/dt), are averaged and shown. The MI+hAMC group showed 

slight improvement in LVSP and LV dP/dt. There is, however, no statistical 

significance. 

Online Fig V. Evidence of long term survival of EGFP-positive cardiomyocytes 

in Wistar rat heart 

(A) Number of injected cells, detected EGFP-positive/cardiac troponin-I 

positive cardiomyocytes, and % surviving cells as a function of the number of 

days after the transplantation are shown. (B-G) Representative confocal 

laser microscopic view of the immunohistochemical analysis using anticardiac 

troponin-I (Trop-I) antibody at the day after transplantation. Although 

fluorescent intensity of EGFP was decayed as a function of the time after 

Suppl 2



transplantation, there is no significant decrease in % surviving cells. Scale 

bars denote 50 µm. 

Online Fig VI Massive survival of EGFP-positive hAMCs and 

transdifferentiation into cardiomyocytes in Wistar rat heart. 

Laser confocal microscopic view of immunohistochemistry with anti-cardiac 

troponin-I antibody (Trop-I; red, C). Nuclei were stained with DAPI (blue, A). 

Many EGFP-positive (EGFP; green, B) rod-shaped cells expressed Trop-I 

and were survived in the Wistar rat heart even at 2 weeks after the 

transplantation. Images of A-C were superimposed and shown in D. Scale in 

panel A denotes 50µm. 

Online Fig VII. Laser confocal microscopic view of immunohistochemistry of 

transdifferentiated hAMCs in the Wistar rat heart. 

Unmerged images of Fig 4 C and Fig 4 D are shown in A-F and F-J, 

respectively. Please see detail in the legend of Fig. 4. 

Online Fig VIII. Laser confocal microscopic view of immunohistochemistry of 

transdifferentiated hAMCs in the EGFP-transgenic mouse heart. 

Non EGFP-labeled hAMCs were transplanted into myocardial infarction area 

of EGFP-transgenic mouse heart. Two weeks after the transplantation, laser 

confocal microscopic view of immunohistochemistry with anti-!-actinin (!- 

actinin; red C). Nuclei were stained with DAPI(blue, A). Many EGFP-negative 

(EGFPl; green, B) rod-shaped cell expressed !-actinin were survived in the 

mouse heart even at 2 weeks after the transplantation. Images of A-C were 

superimposed and shown in D. White box area in D was expanded and 

shown in E. Clear striation staining pattern of !-actinin was observed in 

EGFP-negative hAMCs derived cells. Scale in panel C denotes 50µm. 

Online Fig IX. Laser confocal microscopic view of immunohistochemistry of 

Suppl 3



regulatory T cells in the Wistar rat heart. 

Unmerged images of Fig 5 E are shown. Please see detail in the legend of 

Fig. 5. 

Online Fig X. In vitro transdifferentiation ability of hAMCs to the other organs. 

Laser confocal microscopic view of immunocytochemistry. Hepatogenic 

differentiation (A,D), pancreogenesis (B), osteogenesis (C), and neurogenesis 

(E,F) were examined (See also supplemental Material and Method 3). 

Transdifferentiated hAMCs expressed organ specific antigens, however, failed 

to show organ specific morphology in vitro (A,D,E,F). Scale bars denote 

50µm. 

Online Fig XI. Effect of hAMCs transplantation on capillary density in vivo 

Immunohistochemistry using antibody against CD34 was used to detect 

capillary density. % CD34 (+) area in Non-MI area and MI area in the MIgroup 

and MI+hAMCs group were calculated and averaged and shown in A. 

There is no difference in capillary density. Scale bars denote 50 µm. 

Online Table I. Primer sets to detect cardiomyocyte-specific and associated 

genes 

Suppl 4



Supplemental Material and Methods 

1. The isolation of the amniotic membrane-derived mesenchymal cells 

Amniotic membrane consists of fetal amniotic membrane and maternal 

deciduas. Deciduas are a part of uterine endometrium, which also 

contains a lot of cardiac precursor-like cells (Hida et al., Stem Cells, 

2008; 26(7): 1695-1704). The speed of population doubling of maternal 

endometrial cells was significantly greater than hAMCs (18.2 ± 1.7 PDs 

at 40 days in endometrial cells vs 6.68 ± 1.67 PDs at 40 days in hAMCs). 

Therefore, in our preliminary data, even if there was a small amount of 

contamination of maternal cells at the start of the culture, they quickly 

overcame the hAMCs and, finally, the majority of the cells became the 

maternal endometrial cells. Therefore, a human amniotic membrane was 

collected after delivery of a male neonate in order to check the maternal 

cell contamination in culture by the karyotype. At the 2 nd passage and 

10 th passage of hAMCs, the CEP X/Y DNA Prove Kit (Vysis) was used to 

determine the proportion of XX and XY cells in accordance with the 

manufacturer!s suggestions. 

hAMCs were isolated according to the method previously described 1 with 

slight modification. Avascular transparent amniotic membrane was 

peeled off from maternal placenta and deciduas, rinsed with phosphatebuffered 

saline (PBS), chopped into small pieces, and incubated in 2.4 

U/mL dispase II (Roche Diagnostics, 4942078) at 37ºC for 55 minutes. 

The membranes were then incubated in 0.75mg/mL type-II collagenase 

(Worthington Biochemical, NJ, USA) for 60 minutes at 37ºC. Isolated 

cells were rinsed and seeded on a 10 cm culture dish with Mesenchymal 

Stem Cell Basal Medium (MSCBM, LONZA, PT-3238). After two or three 

days, the cultured dishes became subconfluent in a humidified and 

Suppl 5



normoxic atmosphere (20% O 2 ) and 5% CO 2 at 37ºC. Passage was 

done every 3 or 4 days. 

Umbilical cord and placenta-derived mesenchymal cells were isolated by 

explant culture as described previously 2 . Umbilical cord and chorionic 

plate were chopped into small pieces, 5-8 mm 3 blocks, and put on 10 cm 

dishes with 10 ml hi-glucose Dulbecco Modified Eagle Medium (DMEM- 

H) with 10 % fetal bovine serum, 100 U/mL penicillin, 100 µg/mL 

streptomycin, and 1µg/mL of amphotericin B (Gibco). Mesenchymal cells 

came out within one or two weeks, and passage was done once a week. 

2. Calculation of cardiomyogenic transdifferentiation efficiency in vitro 

It is difficult to measure accurately the population of spontaneously 

contracted EGFP-positive hAMCs, since they usually gather together and 

generate a colony of cells with thick cell layers in the co-culture system. 

Contraction of each cell in the colonized EGFP-positive cells was usually 

unclear because it was difficult to detect the margin of each cell and to 

determine which cells were beating and which cells were not. To enable 

this, isolation of the cell is necessary. Immediately after enzymatic 

isolation of the co-cultured hAMCs, we could observe spontaneouslybeating 

EGFP-positive hAMCs. However, the number of beating cells 

was significantly reduced due to the enzymatic isolation. Therefore, we 

defined cardiac troponin-I positive cells as the cells transdifferentiated 

into cardiomyocytes. In order to stain the intracellular protein structure, 

we tried using triton-X to permeate the cellular membrane. This appeared 

to reduce the specific gravity of the cell; therefore, we could not rinse the 

antibody from the floating isolated hAMCs by centrifugation, etc. Thus, 

we could not use the FACS system in this protocol, but devised an 

original method to evaluate the cardiomyogenic induction rate. 

Suppl 6



hAMCs, one week after co-culture with murine fetal cardiomyocytes, 

were dissociated by 0.1% trypsin and 0.25 mM EDTA for 5 min at 37°C. 

These cells were then dissociated by 0.5% collagenase type-II 

(Worthington Biochemical, Lakewood, NJ) and 10 mM 2,3-butanedione 

monoxime (BDM) (Sigma) for 20-60 minutes at 37°C. After centrifugation 

(1000 rpm, 5min), the hAMCs were seeded onto poly-L-Lysine coated 

dishes. The dishes were fixed by 4% PFA and treated with 0.02% of 

triton-X, then immunocytochemically stained by the antibodies to 

troponin-I overnight at 4°C. After incubated with TRITC conjugated anti 

mouse-IgG antibody, specimens were observed by confocal laser 

microscopy (FV-1000, Olympus, Tokyo, Japan). 

3. In vitro differentiation potential to the non-cardiac organs. 

Transdifferentiation culture conditions are shown 

1 . Hepatic 

transdifferentiation was induced in DMEM-H, 10% FBS, 100nmol/L 

dexamethasone (Wako) and 100 nmol/L insulin (Sigma-Aidrich). 

Pancreatic differentiation was induced in DMEM-H, 10% FBS, 10 mmol/L 

nicotinamide (Wako). Neurogenesis was induced in DMEM-H, 10% FBS, 

30 µmol/L all trans- retinoic acid (Wako). Chondrogenic 

transdifferentiation was induced in DMEM-H medium, 10% FBS, 1 µmol/L 

insulin, 10 µg/L TGF-"1 (Wako), and 300 nmol/L fresh ascorbic acid 

(Wako). Osteogenesis was induced with DMEM-H medium, 10% FBS, 10 

µmol/L dexamethasone, 10 mol/L 1!,25-dihydroxy-vitamine D3 (Wako), 

300 nmol/L ascorbic acid, and 10 mmol/L "-glycerophosphate (Sigma- 

Aldrich). All transdifferentiation was accomplished by culturing for 2 

weeks. 

After the transdifferentiation, immunocytochemistry was performed to 

evaluate the transdifferentiation. Cells cultured on a gelatin-coated dish 

were rinsed with DMEM-H, 20% FBS, and dried. Cells were fixed with 4% 

Suppl 7



PFA (4ºC) for 20 minutes, and rinsed with phosphate-buffered saline 

(PBS) three times. Then cells were fixed in 0.02% triton-X 200 for 20 

minutes at room temperature, and rinsed with PBS. Cells were incubated 

with primary antibodies: anti-albumin (1:1000; abcam ab8940), antiglucagon 

(prediluted; abcam ab930), anti-glial fibrillary acidic protein 

(GFAP) (1:200; Santa Cruz Biotechnology sc9065), anti-nestin (1:200; 

abcam ab22035), anti-osteocalcin (1:50; Acris BP710), or anti-collagen 

type-II (1:20; Cosmobio MNS-PS042) for 1 hour at room temperature. 

Dilution buffer without primary antibody was used as the negative control. 

The samples were rinsed with PBS three times and incubated with FITCconjugated 

secondary antibodies for 30 minutes at room temperature. 

Cells were rinsed with PBS and mounted with fluorescent mounting 

medium (Dako Cytomation S3023). 

4. Calculation of number of surviving EGFP-positive cardiomyocytes in vivo 

Immediately after the hearts were excised, they were fixed by 4% 

paraformaldehyde (PFA) for 2 days at 4ºC. Then tissue was dehydrated 

by 10% sucrose containing PBS for 6 hours and then 20 % sucrose 

containing PBS for 6 hours. The dehydrated tissue was dipped with 

optimal cutting temperature (OCT) compound, and then the tissue was 

quickly frozen by liquid nitrogen. The whole tissue specimens were 

sliced by the cryostat (6 µm thickness), at intervals of 360 µm, then 

completely dried under air flow for 2 hours. They were fixed again by 4% 

PFA for 30 minutes, subsequently treated by 0.2% triton-X containing 

PBS solution for 30 minutes, and then immunohistochemistry was 

performed. Stained samples were observed by fluorescence microscope 

(IX-71, Olympus, Tokyo, JAPAN) with 10x objective lens 

(UPLFLN10XPH). EGFP-positive cardiomyocytes were defined by cells 

having clear striation staining pattern of cardiac troponin-I. The number 

Suppl 8



of the EGFP-positive cardiomyocytes was calculated in every tissue slice. 

In the present study, the size of EGFP-positive cardiomyocytes are the 

same as the host cardiomyocytes; therefore, the average size of the 

EGFP-positive cardiomyocytes was postulated as normal mammalian 

ventricular cardiomyocytes (circular cylinder 100 µm in length and 11 µm 

in radius, cell volume is calculated as 37.994 pL, 3 . Because the 

cardiomyocyte array is along a random direction, the incidence of 

appearance in a tissue section is postulated to be the same as the 

incidence of appearance, in a single slice, of a sphere of the same 

volume (radius 20.9 µm). This suggests that, if we sliced the tissue at 

every 41.8 µm, we could observe every EGFP-positive cell in the heart. 

Therefore, we assume the number of surviving EGFP-positive 

cardiomyocytes to be equal to (total count of EGFP-positive 

cardiomyocytes from every slice) x (360/41.8). The rate of surviving 

EGFP-positive cardiomyocytes was defined as the number of the 

surviving EGFP-positive cardiomyocytes divided by the number of 

injected hAMCs. 

5. Isolation of transdifferentiated cardiomyocytes and FISH experiment 

Method for isolation of cardiomyocytes was shown in our previous 

paper 4 , but with slight modification. Excised heart was perfused 

retrogradely, with nominally Ca 2+ free Tyrode!s solution (described later) 

with 0.1% bovine serum albumin (BSA, fraction-V, Gibco) for 3 minutes, 

and then switched to an enzyme solution containing 0.8 g/L of type-II 

collagenase (Worthington Biochemical, NJ, USA) and 0.2 % BSA for 25- 

30 minutes. Next, the left ventricle was minced, incubated, and gently 

agitated in enzyme solution containing 0.8 g/L of the collagenase, 3 % 

BSA, and 0.3 mmol/L CaCl 2 . Incubation with fresh enzyme solution was 

repeated 7-9 times at 10-minute intervals. The supernatant from each 

Suppl 9



digestion was filtered (120-µm mesh) and centrifuged (400 rpm for 3 

minutes). The cells were then stored in a Tyrode!s solution 

supplemented with 0.5 mmol/L CaCl 2 and 10 mmol/L of BDM at room 

temperature. Tyrode!s solution contained (mmol/L) NaCl 129.5, KCl 5, 

NaH 2 PO 4 0.9, NaHCO 3 20, MgSO 4 1.2, CaCl 2 1.8, and glucose 5.5 (pH 

adjusted to 7.4 at the 37°C) with oxygenation (0 2 95% - CO 2 5%). 

Nominally Ca 2+ -free Tyrode!s solution was prepared by simply omitting 

the CaCl 2 . 

The obtained cardiomyocytes were mounted on the inverted fluorescence 

microscope (IX71, Olympus, Tokyo, JAPAN). Rod-shaped cells with 

strong EGFP fluorescence intensity were searched in the 4x objective 

lens and confirmed by phase contrast microscope that the cells have a 

clear striation, which then defined them as EGFP-positive 

cardiomyocytes. The EGFP-positive cardiomyocytes were picked up by 

pipette and stored in the fresh Tyrode!s solution to increase the 

concentration of EGFP-positive cardiomyocytes. After every EGFPpositive 

cardiomyocyte was separated into a new dish of fresh Tyrode!s 

solution, we picked up EGFP-positive cardiomyocytes again to discard 

contaminated EGFP-negative cardiomyocytes by glass pipette (tip 

diameter was about 50~100 µm) mounted on a micromanipulator (WR-6, 

Narishige, Tokyo, Japan). We performed this step again and we finally 

obtained 100% EGFP-positive cardiomyocytes. 

EGFP-positive cardiomyocytes were mounted on the slide glass. After 

air drying, the slide glass (sample) was rinsed in 75 mmol/L KCl (Wako, 

Tokyo,Japan) at room temperature for 60 minutes. The sample was fixed 

in Carnoy!s solution (methanol 3 volumes, acetic acid 1 volume) 4 times 

for 5 minutes each at 4ºC. After drying on a slide warmer at 65ºC for 10 

minutes, the sample was rinsed in aging buffer (2xSCC, 0.1%NP-40) for 

Suppl 10



30 minutes at 37ºC, and then dehydrated in 70%, 85%, 100% ethanol for 

1 minute each at 4ºC. The sample was denatured in denaturation buffer 

(70% formamide, 2x SCC) for 5 minutes at 37ºC and dehydrated again in 

the same way as above. Human Y-FITC conjugated (Cambio, 

STARFISH, 1083-YF-02), human Alu (BIOGENEX, PR-100101), ratXbiotin 

conjugated (Cambio, STARFISH, CA1699-XB), and hybridization 

buffer (Cambio, STARFISH, HYB-1-10) for the negative control were 

applied onto the cells, covered with a coverglass and sealed with a paper 

bond. Hybridization was performed by Hybridizer (DakoCytomation, 

S245030) at 95ºC for 10 minutes, then 42 for overnight. On the 

following day, after removing the cover glass, the glass slide was rinsed 

in 50% formamide (Nakarai, Tokyo, Japan) at 2SSC, pH7.0 /HCl 4 times 

for 10min each at 45, 2SSC twice for 10min at 45, then at room 

temperature. Preblock was done in 1% block ace powder 

(Dainihonseiyaku, Tokyo, Japan)/ 2SSC for 20 min at room temperature, 

followed by blocking with Avidin/Biotin Blocking kit (Vector Laboratories, 

SP-2001) per manufacturer!s recommended protocol, biotinylated anti 

fluorescein antibody (Vector Laboratories, BA-0601) was applied for 30 

min at room temperature. After rinsing in 4SSC three times at room 

temperature, Streptavidin-Cy2 (Jackson Immuno Research, 016-220- 

084) and DAPI were applied for 30 min at room temperature. Finally, the 

glass slide was mounted with fluorescent mounting medium (Dako 

Cytomation, S3023) and inspected with a laser confocal microscope 

(FV1000, Olympus, Tokyo, Japan). 

6. Teratoma formation assay 

After hAMCs transplantation, histological analysis was performed to 

observe teratoma formation 5 of EGFP positive cells. Samples were 

examined for 89 recipient!s hearts. 21.3 ± 2.0 days after transplantation, 

Suppl 11



each heart was sliced every 50 µm or 100 µm. Hematoxylin-eosin 

staining was performed for a slice in which EGFP-positive clusters of 

cells were observed. 

7. Immunohistochemical analysis for angiogenesis in vivo 

Paraffin-embedded heart tissues were sectioned every 4µm, then 

myocardial infarction areas were selected. After deparaffinization and 

antigen retrieval, sample was pretreated with 5% skim milk in Trisbuffered 

saline (TBS) followed by anti-rat CD34 antibody (1:200, R&D 

Systems: AF4117) at 4º C for over night. Next day, each slide was 

washed with TBS and applied with biotinylated Goat immunoglobulins 

(Dako; E0466), next with strept ABC complex/HRP (Dako; K0377) and 

then with 3,3`-Diaminobenzine (DAB) substrate (Wako: K3183500). 

Finally, each sample was counterstained with Hematoxylin. Each 

sample was observed by light microscope (x100) and 5 digital images 

from the normal myocardium and the center of the myocardial infarction 

zone were collected at random. Images were randomized and analyzed 

by the blinded observer. % brown pixel area in the whole heart tissue 

was calculated by Photoshop (Adobe). The data was averaged and 

shown in Online. Fig XI. 

1. Portmann-Lanz CB, Schoeberlein A, Huber A, Sager R, Malek A, 

Holzgreve W, Surbek DV. Placental mesenchymal stem cells as 

potential autologous graft for pre- and perinatal neuroregeneration. 

Am J Obstet Gynecol. 2006;194:664-673 

2. Okamoto K, Miyoshi S, Toyoda M, Hida N, Ikegami Y, Makino H, 

Nishiyama N, Tsuji H, Cui CH, Segawa K, Uyama T, Kami D, Miyado 

K, Asada H, Matsumoto K, Saito H, Yoshimura Y, Ogawa S, Aeba R, 

Yozu R, Umezawa A. 'working' cardiomyocytes exhibiting plateau 

Suppl 12



action potentials from human placenta-derived extraembryonic 

mesodermal cells. Exp Cell Res. 2007;313:2550-2562 

3. Luo CH, Rudy Y. A dynamic model of the cardiac ventricular action 

potential. I. Simulations of ionic currents and concentration changes. 

Circ Res. 1994;74:1071-1096 

4. Fukuda Y, Miyoshi S, Tanimoto K, Oota K, Fujikura K, Iwata M, Baba 

A, Hagiwara Y, Yoshikawa T, Mitamura H, Ogawa S. Autoimmunity 

against the second extracellular loop of beta(1)-adrenergic receptors 

induces early afterdepolarization and decreases in k-channel density in 

rabbits. J Am Coll Cardiol. 2004;43:1090-1100 

5. Lensch MW, Schlaeger TM, Zon LI, Daley GQ. Teratoma formation 

assays with human embryonic stem cells: A rationale for one type of 

human-animal chimera. Cell Stem Cell. 2007;1:253-258 

Suppl 13

Tsuji et. al. amniotic membrane-derived stem cell (2009/205260-R3 ...

Create successful ePaper yourself

Delete template?

Save as template?