Can denaturing gradient gel electrophoresis (DGGE) analysis of ...

Soil Biology & Biochemistry 38 (2006) 1188–1192
www.elsevier.com/locate/soilbio
Can denaturing gradient gel electrophoresis (DGGE) analysis of amplified
16s rDNA of soil bacterial populations be used in forensic investigations
Anat Lerner a , Yaron Shor b , Asya Vinokurov b , Yaacov Okon a , Edouard Jurkevitch a, *
a Department of Plant Pathology and Microbiology and The Otto Warburg Minerva Center for Agricultural Biotechnology, Faculty of Agricultural,
Food and Environmental Quality Sciences, The Hebrew University of Jerusalem, Rehovot Campus, Israel
b The Division of Identification and Forensic Sciences, Toolmarks and Materials Laboratory, Israel Police National Headquarters, Jerusalem, Israel
Available online 8 November 2005
Abstract
In criminal investigations, information on the origin of soils may be crucial for solving cases. The biological complexity of soil may potentially
be used for sorting and differentiating between soil samples. Nucleic-acid based analyses of soil microbial populations are powerful tools,
routinely used in studies of this habitat. Application of such approaches in forensics implies that a standardized DNA extraction method has to be
applied to all samples. In this study, several DNA extraction protocols were compared. An improvement on the method proposed by Tsai and
Olson (1991) was found to be most suited to extract DNA from various soil types, including from small samples. A blind test on soils from a crime,
an alibi scene and unrelated locations was conducted to evaluate the potential of environmental PCR and denaturating gradient gel electrophoresis
for use in forensic science. In most cases, soil patterns clustered according to soil type and location.
q 2006 Elsevier Ltd. All rights reserved.
Keywords: DNA extraction; PCR; DGGE; Cluster analysis; Forensic science
1. Introduction
Criminal investigators often have to rely on tiny clues in their
search for the truth. If these clues can provide clear evidence
linking one or more individuals to a crime, they can turn to be
essential proof, or supporting evidence in convicting or
exonerating suspects. In several cases tiny amounts of soil can
play an important role in the field of physical evidence (Marumo
et al., 1995), hence the search for the origin of the soil is crucial for
the solution of the case. Soils are particularly heterogeneous and
complex habitats consisting of inorganic minerals, organic matter
and living biota (O’Donnell and Gorres, 1999), supporting a
tremendous microbial diversity that is not reflected in culturebased
approaches (Ranjard et al., 2000; Kent and Triplett, 2002).
Nucleic-acid based analyses of soils have, therefore, become
standard and powerful tools in studies of this habitat (Felske et al.,
1998; Kozdroj and van Elsas, 2000; Nannipieri et al., 2003).
Bacteria are part of the soil microflora and they have the potential
to reflect the history of a given environment (Ranjard et al., 2000).
This potential can be useful for forensic purposes. Analysis of the
living biota in soil based on nucleic acid-based methodologies
* Corresponding author.
E-mail address: jurkevi@agri.huji.ac.il (E. Jurkevitch).
0038-0717/$ - see front matter q 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.soilbio.2005.10.006
may therefore provide the criminal investigator with yet another
potent tool (Horswell et al., 2002).
However, successful application of molecular techniques
relies on effective recovery of nucleic acids from the
environment (Hurt et al., 2001). Moreover, for forensic
analyses, there is a need for a simple procedure, which can
provide sensitive detection from a wide variety of microorganisms
and a wide variety of soils (Kuske et al., 1998). This
procedure should be repeatable, be usable with small samples,
and provide a large statistical confidence in its results.
Human DNA-based forensic data is now largely used in
courts around the world and has played a major role in
numerous publicized trials (Jost, 1999). In contrast, the use of
soil bacterial DNA for forensic purposes is not a routine
procedure (Horswell et al., 2002).
The objectives of this study were: (i) to test soil DNA
extraction protocols on various soil types and their efficiency
with small sized samples; (ii) to perform a feasibility study of the
PCR-DGGE approach as a forensic tool for analyzing the
microbial diversity existing in soils collected from crime scenes.
2. Materials and methods
2.1. Crime scene
A young woman was found stabbed to death on the banks of
the Yarkon River in Tel-Aviv. No footprints, weapon or other

A. Lerner et al. / Soil Biology & Biochemistry 38 (2006) 1188–1192 1189
physical evidence was found in the crime scene. The main
suspect, arrested a couple of days later, claimed to have been
with the victim on a non-asphalted, nearby parking lot (the alibi
area). The suspect washed his clothes and shoes after the
murder and the only possible link to the crime scene was a
small soil clot (0.2 g) found inside his shoe.
2.2. Soils
Soils samples were collected from different locations
(Table 1). Soils used for the evaluation of DNA extraction
protocols were sampled 5 cm below the surface, and kept on
ice until processed. Samples originating from the crime scene
were kept in plastic bags on the shelf and in the dark for
6 months until analysis. The soil clot from the suspect’s shoe
was not made available for analysis by the court because of the
destructive nature of the analysis.
2.3. Soil wash
Soil samples in 1.8 ml of 0.05 M buffer phosphate and 0.5%
cetyltrimethylammonium bromide (CTAB) were ground using
a mortar and pestle. Samples were shaken for 3 h at 4 8C and
200 rev min K1 , and then centrifuged for 10 min at 4 8C and
700 rev min K1 . The supernatant was removed and DNA
extraction was performed.
2.4. DNA extraction
Five direct methods were used for DNA extraction from
bacterial communities: (a) Tsai and Olson (1991) (thereafter
‘T’) with slight modifications: After three freeze-thaw cycles,
proteinase K was added to the solution to a final concentration
of 50 mg ml K1 and the samples were incubated for 30 min at
37 8C; (b) Zhou et al. (1996) (‘Z’); (c) Yeates et al. (1998)
(‘Y’); (d-e) methods based on the commercial kits FastDNA
SPIN Kit for Soil (BIO 101, Qbiogene, Inc, Carlsbad, USA)
(‘F’) and UltraCleane Soil DNA kit (MO BIO Laboratories,
USA) (‘U’). The Y and F methods included a bead beating step.
Samples weighing 0.2, 1, 10, 0.6, and 0.25 g were used with
methods, T and Y, Z, F, and U, respectively. The weight of the
soil samples was the optimum recommended for each protocol.
2.5. Purification of crude DNA extracts
The QIAquick Gel Extraction kit (QIAGEN GmbH, Hilden,
Germany, and DNA Isolation kit (Biological Ind., Israel) were
used to purify DNA.
2.6. PCR amplification
One to three microlitre of each DNA preparation from
environmental sample were amplified in a PCR reaction
mixture (50 ml) using an Eppendorf Mastercycler Gradient
(Brinkmann Instruments, Inc., USA). Each PCR mixture
contained 0.8 mM of each primer, 0.3 mM of each deoxynucleotide
(dNTP), 5 ml of 10! buffer (Promega, Madison,
USA), 0.03 unit ml K1 redTaq DNA polymerase (Sigma,
Rehovot, Israel), 3.75 mM MgCl 2 ,2ml of10mgml K1 BSA
and double distilled, sterilized water to complete the mixture
volume. The primers for PCR were specific for conserved
bacterial 16S rDNA sequences (Heuer et al., 1997). PCR with
primers Gm5f (5 0 -GC-clamp-CCT ACG GGA GGC AGC
AG-3 0 ) and 907r (5 0 -CCC CGT CAA TTC CTT TGA GTT
T-3) amplified a bacterial 16S rDNA fragment from position
341–928 (Escherichia coli numbering). PCR amplification was
performed for 35 cycles as follows: after initial denaturation of
1 min at 95 8C each cycle consisted of denaturation at 95 8C for
20 s, primer annealing at 57 8C for 25 s, and primer extension
at 72 8C from 30 s. Cycling was followed by final primer
extension at 72 8C from 1 min. PCR products were visualized
by electrophoresis in 1% (w/v) agarose gels after 1 mg ml K1
EtBr staining (Sambrook et al., 1989).
2.7. Denaturing gradient gel electrophoresis
Strong PCR products of the expected size (550 bp) were
subjected to DGGE analysis. DGGE was performed with an
Ingeny phor U-2 system (Leiden, The Netherlands). Samples of
43 ml of PCR product were loaded onto 6% (w/v) polyacrylamide
gels in 1.0 strength Tris-ethylene-diamineteraacetate
Table 1
Properties of the soils used in this study
Location Texture Sampling zone Organic matter (%) pH Moisture content (%)
Kfar Menachem,
Sandy loam (Chromoxererts brown alluvial) Maize rhizosphere 0.9 7.23 5.5–6
Israel
Rehovot, Sandy (Haploxeralfs brown–red) Maize rhizosphere 0.5 8.4 27.9
Israel
Hula, Israel Peat (Lacustrine gley) Surface soil 29.3 7.26 84.5
Coconut Compost (15% polystyrene) 85 4.8–5.6 30–35
residues
medium
Crime scene Sandy clay loam (Hamric alluvial soils and gley) Surface soil N.D. N.D. N.D.
Alibi scene Sandy clay loam (Calcareous sandstone) Surface soil N.D. N.D. N.D.
Suspect’s
home
Sandy loam Surface soil N.D. N.D. N.D.
N.D.: not determined.

1190
A. Lerner et al. / Soil Biology & Biochemistry 38 (2006) 1188–1192
(TAE, pH 8.5) TAE buffer. The polyacrylamide gels were
prepared with a denaturing gradient ranging from 30 to 60%
(where 80% denaturant contained 7 M urea and 40%
formamide). The electrophoresis was run for 20 h at 85 V
at 60 8C. After the runs, gels were removed from the set up and
stained for 30 min with 2 l of 1!TAE and 100 ml of
10 mg ml K1 EtBr solution followed by washing with 1!
TAE for 15 min. The stained gels were immediately
photographed using an AlphaImagere System (Labtrade
Inc., FL, USA).
2.8. Cluster analysis
The analysis was performed as blind test in which the
operator was unaware of the origin of the samples. Cluster
analysis of profile similarity was performed using the
Discovery Series Quantity One 1-D Analysis Software Version
4.4.1, PC (Bio-Rad, Rishon Le Zion, Israel) and UPGMA.
3. Results
3.1. DNA extraction
All the protocols used in this study were based on direct
DNA extraction. Direct DNA protocols include three main
elements: chemical, physical and enzymatic lysis (Miller et al.,
1999). Each protocol, which had been tested herein, included
one or more of these elements. Of the methods tested, only
protocols T, F and Z were successful in extracting DNA from a
Rehovot and a Kfar Menachem maize rhizosphere soil.
DNA was also successfully extracted from a coconut
residues medium and from Hula soil, two highly organic
soils, using the T and F methods (Table 1). The use of the DNA
Isolation kit (Biological Ind., Israel), failed to remove humic
acids and resulted in brownish samples that could not be
amplified by PCR using primers for the 16S rRNA gene. After
a second purification step, including electrophoretic separation
of humic substances from DNA in agarose, DNA excision from
the gel and purification using QIAquick Gel Extraction kit
(QIAGEN GmbH, Hilden, Germany), PCR products were
obtained.
As a lesser amount of impurities co-extracted with DNA,
protocol T was selected as the DNA extraction method to be
used for the evaluation of soil microbial community analysis
for forensic purposes.
3.2. Microbial community analysis
Soil samples were taken at various places at and around
the murder scene, at the alibi area and near the family
parking lot at the main suspect’s home. In some forensic
cases, the amount of material collected is so low that no
replicate samples can be obtained (for example, soil
adhering to a sole, or to a piece of cloth). In order to
reflect this situation, the presented analysis was performed:
(i) without prior knowledge of the origin of the samples,
and; (ii) with only one replicate per location.
Therefore, to mimic the amount found on the suspect’s shoe
(around 0.2 g), which was not available, and to evaluate the
efficiency of the protocol T, 0.2–0.6 g of soil was extracted. DNA
was retrieved from all samples. In some instances, amplicons
were only obtained after dilution of the DNA samples.
DGGE was performed on the samples originating from the
crime scene, from the alibi scene, from the suspect’s home and
from different geographical places in Israel in a blind test.
Banding patterns were compared using cluster analysis. All
samples collected from the crime scene and surroundings
clustered together. All the samples from the alibi scene and
surroundings were clearly separated from the crime scene
samples. They clustered closer to the samples from Rehovot
and Kfar Menahem, two soils with a texture similar to that of
the alibi scene samples. Also, a Beit Dagan and a Hula soil,
both richer in organic matter appeared to be closer to the crime
scene samples. However, a sample from the suspect’s home
also clustered with crime scene samples (Fig. 1).
4. Discussion
4.1. Use of soils in forensic science
Molecular approaches have become a common tool for the
analysis of the effect of plant cover, agricultural amendments,
pollutants and environmental disorders on soil microbial
communities (Torsvik et al., 1998; Marilley and Aragno,
1999; Smit et al., 2001; Johnsen et al., 2001). Such tools may
also prove useful in criminal investigations to link a suspect to
a crime scene. The first requirement for implementing such
methods is a reliable, convenient and reproducible method of
DNA extraction from soils.
A number of methods for extracting DNA from diverse
environments such as soils are available. Thus, the first step of
this study was to compare the efficacy of DNA retrieval with
various methods on a particular soil type (Haploxeralfs brown–
red of Rehovot). The amount of DNA extracted was evaluated
by running the DNA obtained on gel agarose and comparing
the band intensity after EtBr staining. Zhou et al. (1996) found
a negative correlation between cell lysis efficiency and clay
content, in contrast to Ranjard et al. (2000) who found no such
significant correlation. Our results are in better agreement with
those of Ranjard et al. (1998) as the ‘Z’ protocol (Zhou et al.,
1996) applied to an Haploxeralfs brown–red soil with a low
clay content only yielded low amounts of DNA. Bead beating,
although recommended for sandy soils (Yeates et al., 1998),
was found to be rather inefficient for this Haploxeralfs soil.
Furthermore, only three out of the five methods evaluated
successfully extracted DNA from a loamy sand test soil
(Chromoxererts brown alluvial of Kfar Menachem). Further
comparison with other soil types clearly showed that the timeconsuming
Tsai and Olson (1991) protocol (protocol T) was
the most suitable. The combination of physical, chemical and
enzymatic attack of the cell wall in addition to initial grinding
of the sample enabled efficient recovery of DNA from
small samples (0.2 g) and from all soil types. Furthermore,
DNA could also be extracted and amplified from pieces of

A. Lerner et al. / Soil Biology & Biochemistry 38 (2006) 1188–1192 1191
(a)
0.44 0.60 0.70 0.80
1.00
Beit Dagan
(b)
cloth containing traces of soil type (crime scene) and weighing
0.2–1 g.
Another advantage of protocol T approach was the
relatively low amount of DNA shearing, which increased
when bead beating was applied. Shearing may lead to
amplification of chimeric products and to bands that do not
reflect the actual diversity in a subsequent DGGE analysis
(Wintzingerode et al., 1997; Roose-Amsaleg et al., 2001). In
addition, more humic substances were co-extracted with bead
beating (protocol F) than with protocol T.
4.2. Microbial community analysis
Hula valley
Rehovot
Crime scene
Crime scene
surrounding area
Crime scene
Suspect’s home
Kfar Menachem
Rhizosphere
Rehovot Rhizosphere
Alibi scene,10m west
Alibi scene
Alibi scene,10m east
Alibi scene,10m south
1 2 3 4 5 6 7 8 9 10 11 12 13
Fig. 1. Cluster analysis (A) and DGGE profiles (B) of bacterial communities
from different soils. Lanes: 1, crime scene; 2, Alibi scene; 3, Alibi scene, 10 m
south; 4, Alibi scene, 10 m east; 5, suspect’s home; 6, crime scene surrounding
area; 7, Alibi scene, 10 m west; 8, Hula valley; 9, Beit Dagan; 10, Rhizosphere
Rehovot; 11, Rhizosphere Kfar Menahem; 12, Rehovot, and; 13, CRIME scene.
Profiling of bacterial soil and rhizosphere communities by
denaturing gradient gels is a powerful tool for rapid analysis of
temporal and spatial variations in the soil and the rhizosphere.
Although the samples from the suspect’s home and from the
crime scene could not be separated based on cluster analysis,
DGGE reflected a large bacterial diversity, which, in most
cases clustered according to soil type and location. Plant roots
are sources for organic matter, and directly influence the
bacterial community associated with them, as reflected in
changes in PCR-DGGE patterns between bulk and rhizosphere
soil samples. In the test case described here, there was partial
plant cover on soils at the crime scene and in the garden next to
the suspect’s house, but not at the alibi scene. This might have
influenced the bacterial composition, yielding more similar
banding profiles.
The simulation presented here called for the evaluation of
non-replicated samples, a situation which can seriously limit
the sorting of samples of different origins. We suggest that by
replicating the methods used for fingerprinting and the
analytical approaches, statistical support could be achieved.
In an accompanying paper (Lerner et al., 2006), we show that
the profiling of rhizobacterial populations performed with
DGGE and different primer sets and with ribosomal intergenic
spacer analysis, and analyzed by different statistical methods
yielded similar results.
Sample collection can have an effect on subsequent analyses
and proper handling, including tool disinfection, the use of
plastic bags to slow down gas exchange and appropriate and
standardized storage conditions are recommended (Wintzingerode
et al., 1997). PCR-DGGE is powerful fingerprint tool
but it also has drawbacks: for example, bands can migrate to
the same position, there is a strong bias for dominant
populations, and multiple rrn copies from the same organism
can yield different bands (Nannipieri et al., 2003). These biases
appear in addition to those generated by differential DNA
extraction and amplification (Roose-Amsaleg et al., 2001).
4.3. Recommendations
In conclusion, the following should be considered for an
in-depth evaluation of the approach: whenever possible,
samples should be collected as soon as the crime is discovered,
as much material as possible should be retrieved from the
locations under investigation, including soils with and without
rhizosphere, from different depths and different positions at and
near the crime scene, and in unrelated areas; Sample collection
and storage should be standardized to minimize their impact on
the bacterial community, and the analysis should be performed
as rapidly as possible after sample collection (Wintzingerode
et al., 1997); Samples should be analyzed using a number of
primer sets, including phylogenetically-restricted primers
targeting specific populations. Statistical analysis should be
performed using at least two different methods such as cluster
and principal component analysis. Whenever needed, sequence
information (from bands running identically in DGGE for
example) should be sought to strengthen statistical power. We
also suggest that a molecular survey of soils from different
geographical and climatic regions should be undertaken using
the guidelines stated above to obtain a large amount of data

1192
A. Lerner et al. / Soil Biology & Biochemistry 38 (2006) 1188–1192
which could be used to calculate the statistical strength of such
analyses.
References
Felske, A., Wolterink, A., van Lis, R., Akkermans, A.D.L., 1998. Phylogeny of
the main bacterial 16S rRNA sequences in Drentse a grassland soils (The
Netherlands). Applied and Environmental Microbiology 64, 871–879.
Heuer, H., Krsek, M., Baker, P., Smalla, K., Wellington, E.M.H., 1997.
Analysis of actinomycete communities by specific amplification of genes
encoding 16S rRNA and gel-electrophoresis separation in denaturing
gradients. Applied and Environmental Microbiology 63, 3233–3241.
Horswell, J., Cordiner, J.S., Maas, E.W., Martin, T.M., Sutherland, B.W.,
Speir, T.W., Nogales, B., Osborn, A.M., 2002. Forensic comparison of soils
by bacterial community DNA profiling. Journal of Forensic Science 47,
350–353.
Hurt, R.A., Qiu, X., Wu, L., Roh, Y., Palumbo, A.V., Tiedje, J.M., Zhou, J.,
2001. Simultaneous recovery of RNA and DNA from soils and sediments.
Applied and Environmental Microbiology 67, 4495–4503.
Johnsen, K., Jacobsen, C.S., Torsvik, V., Sorensen, J., 2001. Pesticide effects
on bacterial diversity in agricultural soils—a review. Biology and Fertility
of Soils 33, 443–453.
Jost, K., 1999. DNA databases. Does expanding them threaten civil liberties
The CQ Researcher 9, 449–472.
Kent, A.D., Triplett, E.W., 2002. Microbial communities and their interactions
in soil and rhizosphere ecosystems. Annual Review in Microbiology 56,
211–236.
Kozdroj, J., van Elsas, J.D., 2000. Application of polymerase chain reactiondenaturing
gradient gel electrophoresis for comparison of direct and
indirect extraction methods of soil DNA used for microbial community
fingerprinting. Biology and Fertility of Soils 31, 372–378.
Kuske, C.R., Baton, K.L., Adorada, D.L., Stark, P.C., Hill, K.K., Jackson, P.J.,
1998. Small-scale DNA sample preparation method for field PCR detection
of microbial cells and spores in soil. Applied and Environmental
Microbiology 64, 2463–2472.
Lerner, A., Herschkovitz, Y., Baudoin, E., Nazaret, S., Moenne-Loccoz, Y.,
Okon, Y., Jurkevitch, E., 2006. Effect of Azospirillum brasilense inoculation
on rhizobacterial communities analyzed by denaturing gradient gel
electrophoresis and automated ribosomal intergenic spacer analysis. Soil
Biology & Biochemistry 38, this issue, doi:10.1016/j.soilbio.2005.10.007.
Marilley, L., Aragno, M., 1999. Phylogenetic diversity of bacterial
communities differing in degree of proximity of Lolium perenne and
Trifolium repens roots. Applied Soil Ecology 13, 127–136.
Marumo, Y., Sugita, R., Seta, S., 1995. Soil as evidence in criminal
investigation. The 11th Interpol Forensic Science Symposium, pp. 1–16.
Miller, D.N., Bryant, J.E., Madsen, E.L., Ghiorse, W.C., 1999. Evaluation and
optimization of DNA extraction and purification procedures for soil and
sediments samples. Applied and Environmental Microbiology 65, 4715–
4724.
Nannipieri, P., Ascher, J., Ceccherini, M.T., Landi, L., Pietramellara, G.,
Renella, G., 2003. Microbial diversity and soil functions. European Journal
of Soil Science 54, 655–670.
O’Donnell, A.G., Gorres, H.E., 1999. 16S rDNA methods in soil microbiology.
Current Opinion in Biotechnology 10, 225–229.
Ranjard, L., Poly, F., Nazaret, S., 2000. Monitoring complex bacterial
communities using culture-independent molecular techniques: application
to soil environment. Research in Microbiology 151, 167–177.
Roose-Amsaleg, C.L., Garnier-Sillam, E., Harry, M., 2001. Extraction and
purification of microbial DNA from soil and sediment samples. Applied
Soil Ecology 18, 47–60.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning, A
Laboratory Manual, second ed. Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, NY.
Smit, E., Leefland, P., Gommans, S., van den Broek, J., van Mil, S., Wernars,
K., 2001. Diversity and seasonal fluctuations of the dominant members of
the bacterial soil community in a wheat field as determined by cultivation
and molecular methods. Applied and Environmental Microbiology 67,
2284–2291.
Torsvik, V., Daae, F.L., Sandaa, R.A., Ovreas, L., 1998. Novel techniques for
analyzing microbial diversity in natural and perturbed environments.
Journal of Biotechnology 64, 53–62.
Tsai, Y.L., Olson, B.H., 1991. Rapid method for direct extraction of DNA from
soil and sediments. Applied and Environmental Microbiology 57, 1070–
1074.
Wintzingerode, F.V., Gobel, U.B., Stackebrandt, E., 1997. Determination of
microbial diversity in environmental samples: pitfalls of PCR-based rRNA
analysis. FEMS Microbiology Reviews 21, 213–229.
Yeates, C., Gillings, M.R., Davidson, A.D., Altavilla, N., Veal, D.A., 1998.
Methods for microbial DNA extraction from soil for PCR amplification.
Biological Procedures Online 1, 40–47.
Zhou, J., Bruns, M.A., Tiedje, J.M., 1996. DNA recovery from soils of diverse
composition. Applied and Environmental Microbiology 62, 316–322.