Source Tracking ofEscherichia coli Using the l6S-23S Intergenic Spacer Region
ofthe rrnB Ribosomal Operon
by
Tom D'Elia
Submitted in Partial Fulfillment ofthe Requirements
For the Degree of
Masters ofScience
In the
Biological Sciences
Program
YOUNGSTOWN STATE UNIVERSITY
August, 2004
,
Source Tracking ofEscherichia coli using the 16S-23S Intergenic Spacer Region
ofthe rrnb Ribosomal Operon
Tom D'Elia
I hereby release this thesis to the public. I understand that this thesis will be made
available from the OhioLINK ETD Center and the Maag Library Circulation Desk for
public access. I also authorize the University or other individuals to make copies of this 
thesis for scholarly research.
Signature:
Tom D'Elia ('
Approvals:
Dr. Carl G. Johnston, Thesis Advisor
.--'-~ ... ~~ .. /?.?.-?-.7~ ~~' ,/~;,~~
Date
Abstract
119 Escherichia coli isolates from nine different animal sources were subjected to
denaturing gradient gel electrophoresis (DGGE) to determine sequence variations within
the 16S-23S intergenic spacer region (ISR) ofthe rrnB ribosomal operon. The ISR was
analyzed to determine ifE. coli isolates from various animal sources could be
differentiated from each other. In E. coli, the 16S-23S ISR has been demonstrated to
consist ofnon-essential sequences that are subject to frequent insertion or deletion events
that may allow for differences between different isolates. DNA isolated from the E. coli
animal sources was PCR amplified to isolate the rrnB operons. To prevent PCR
amplification ofall seven E. coli ribosomal operons by PCR amplification by using
universal primers, sequence specific primers were utilized for the rrnB operon. An
additional primer set was then used on these amplimers to prepare samples ofthe 16S 
23S ISR for DGGE. DGGE results show the presence of40 unique ISR sequences from
all ofthe samples. The highest rate ofunique banding patterns, 60%, was observed for
humans. The genetic profiles established by the PCR-DGGE method revealed a high
genetic diversity for the E. coli isolates tested. There was also very little correlation
between the ISR profiles created by the DGGE bands between the host sources. These
findings suggest that the 16S-23S ISR may contain some host specificity, and that the
high diversity ofthe E. coli isolates may allow for the assessment ofenvironmental
samples to distinguish similarities.
III
Table ofContents
Abstract 111
Table ofContents IV
List ofFigures V
List ofTables VI
List ofAppendices VB
Acknowledgments V111
Chapter 1: Introduction 1
I. Microbial Source Tracking 1
II. E. coli: Ribosomal DNA and 16S-23S Intergenic 4
Spacer Region
III. Denaturing Gradient Gel Electrophoresis 12
Chapter 2: Method Development and E. coli source isolate PCR-DGGE 14
Specific Aims 14
Materials and Methods 17
Chapter 3: Results 25
ISR Amplification 25
Controls 28
DGGE Analysis ofthe rrnB ISR of 119 E. coli Isolates 35
Long term Cultures 60
Chapter 4: Discussion 65
References 73
IV
List ofFigures
Figure Page
Figure 1: Ribosomal operon location in the E. coli genome. 6
Figure 2: Gene organization ofthe rrnB operon ofE. coli. 8
Figure 3a and b: Primer locations for PCR reactions. 15
Figure 4: PCR amplification resulting in the isolation ofthe 26
rrnB operon from chicken isolates.
Figure 5: PCR amplification resulting in the isolation ofthe 29
rrnB 16S-23S ISR from chicken isolates.
Figure 6: Resolving ofdouble bands that were present after 31
PCR amplification ofthe rrnB 16S-23S ISR.
Figure 7: DGGE gel ofE. coli standard that was used to 33
compare gel migrations between gels.
Figure 8: DGGE gel of cloned Canada goose isolated number 11. 36
Figure 9: Restriction digest ofisolated plasmids from positive clones. 49
Figure 10: DNA sequence alignment oftwo isolates with 51
the same DGGE RD value.
Figure 11: DNA sequence alignment oftwo isolates with 53
different DGGE RD value.
Figure 12: DNA sequence alignment oftwo isolates with 55
different DGGE RD value.
Figure 13: DNA sequence alignment ofall isolates that were sequenced. 57
v
List ofTables
Tables:
Table 1: Frequency and distribution ofgel migrations of
the E. coli isolates from the various sources.
Table 2: Distribution and frequency ofDGGE results compared
among the isolates.
Table 3: Distribution and frequency ofDGGE results compared
between human and nonhuman isolates.
Table 4: Correlation of DGGE Bands from 8 Animal Sources.
Table 5: Results ofthe long term culture studies.
Table 6: Results ofduplicate DGGE results from independently
prepared samples compared.
38
40
43
46
61
63
VI
List ofAppendices
Appendix A- Host Isolates Label and Individual RD Values
Vll
Acknowledgments
United States Geological Survey
I would like to thank Dr. Don Stoeckel ofthe United States Geological Survey for
allowing me to use Eshcerichia coli isolates from his collection. Also, he was very
helpful through the course ofmy research in offering important advice.
Youngstown State University
This research was funded by the Youngstown State University PACER grant.
I would like to thank Dr. Carl Johnston, my research advisor, for all ofhis help and
support. He has been very helpful in all parts ofmy research, including offering lab
advice, and help with writing and presentations. I am very fortunate to have conducted
my research in his lab, and feel that I will be better prepared for future work because of
Dr. Johnston.
I would like to thank Dr. Cooper for his technical support, and use ofhis lab equipment.
As a member ofmy committee, Dr. Cooper was very important in many aspects ofmy
research, including denaturing gradient gel electrophoresis, PCR, cloning and DNA
sequencing. Additionally, he is a very good professor, and I learned a lot from his help
and from his classes.
I would also like to thank another committee member, Dr. Asch. He was very helpful in
giving advice, and the knowledge I gained from his classes was vital to my research
success. Additionally, I would like to thank Dr. Asch for the use ofhis lab equipment.
I would like to extend a special thank you to Ms. Diana Arnett for all ofthe help with
DNA sequencing.
For helping with data analysis, I would like to thank Dr. Diggins.
A special thank you goes to all the members ofmy lab.
Additional thanks goes to Jessica Gordon for all ofher support, and Donald Brown and
James Shevchuck.
A final thanks goes to my parents, family and friends.
Vlll
We shouldpreserve every scrap ofbiodiversity as
priceless while we learn to use it and come
to understand what it means to humanity.
-Edward O. Wilson
IX
Chapter 1: Introduction
I. Microbial Source Tracking
The concept that the origin offecal pollution can be traced using microbiological,
genotypic, phenotypic and chemical methods has been given the general term of
microbial source tracking (MST) (Scott et al 2002). This new technology has come about
in order to track potentially pathogenic microbes to a particular source to prevent further
contamination. Known sources offecal contamination include combined sewage
overflows (CSOs), septic systems, agricultural runoffand wildlife (Dombek et al 2000).
In general, there are two main types classification for sources offecal contamination:
point and nonpoint sources. Point sources considered the major contributor to fecal
pollution include raw sewage, storm water, CSOs and effluents from waste water
pollution. Nonpoint sources are more dispersed and include wildlife, agricultural runoff
and pleasure boats (Seurinck et aI., 2003). The importance ofmicrobial source tracking
is to determine ifthe source is human, livestock or wildlife, since the microorganisms of
human origin are regarded as having greater potential to cause disease in humans and
contain human specific enteric pathogens (Scott et al 2002, Guan et al 2002).
Furthermore, bacteria from humans found in the environment may indicate the presence
ofSalmonella spp., Shigella spp., hepatitis A virus and Norwalk group viruses which are
known human pathogens that do not colonize nonhuman species (Parveen et al 1999).
All MST technologies rely on information received by investigating a particular
indicator organism. Indicator organisms are used to determine the presence offecal
1
pollution in water and are essential for MST. For many years, fecal coliforms have been
widely used as indicators ofhuman enteric pathogens, since they are naturally occurring
in the gastrointestinal tracts ofhumans and warm blooded animals (Parveen et aI1999).
Additionally, pathogens present in the environment are often in low numbers and more
difficult to culture relative to indicator organism (McLellan et aI., 2003). In particular,
Escherichia coli has been extensively used as an indicator organism offecal pollution.
The reason E. coli has become a predominant indicator organism is due to the availability
ofthe complete genome sequence, and that the organism is easily cultured in the lab.
Furthermore, E. coli is not normally pathogenic to humans and is present in much higher
concentrations than the other environmental pathogens it predicts, and thus reveals the
presence ofhuman enteric pathogens (Scott et aI2002).
Some strains ofE. coli are primary pathogens with an enhanced potential to cause
disease, and have been linked to worldwide outbreaks ofsevere disease (Kuhnert et al
2000). Therefore, it is important to monitor the input ofE. coli into waterways, which is
a widespread problem in the U.S. and is correlated with increase risk ofseveral diseases
(Dombek et aI., 2000). When fecal coliforms, including E. coli, are found in high levels
they impair the water quality in lakes and rivers, and bring a threat to those that use the
water as evident by water borne outbreaks ofE. coli 0157:h7 (Guan et aI2002). The
effects ofan E. coli infection in humans can be very serious and life threatening.
Eshcherichia. coli infections are often enteric, and cause severe nausea and diarrhea,
while extraintesintal infections are possible that are related to urinary tract infections,
sepsis and meningitis (Kuhnert et aI., 2000). The fact that E. coli is known to exist in the
natural flora ofthe intestine, and is also a potentially severe pathogen, makes this
2
bacterium a very good indicator organism and a microbe that itselfshould be closely
monitored itselfin the environment.
Most ofthe various MST methods work by comparing environmental samples to
a data bank ofE. coli isolates from known sources. Therefore, MST is based on the
assumption that specific genetic markers or strains ofbacteria are associated with specific
animal sources (Hartel et aI., 2003). The presence ofa predominant strain ofE. coli in a
particular host is most likely due to genetic drift. There are several variables that
influence the potential success ofMST by impacting strain selection and enrichment.
These factors include the microenvironment ofthe particular host, intestine temperature,
pH and diet (Carson et aI2001). Recently, the diet ofconfined deer, compared to wild
deer, was shown to significantly affect ribotypes ofE. coli isolates (Hartel et aI., 2003).
Other assumptions associated with MST are that particular E. coli clones are more likely
to be isolated from one particular host species than another (host specificity) and that the
clonal composition ofthe species isolated from soil or water represent the clonal
composition ofthe species in the host population responsible for the fecal inputs to the
environment (Gordon, 2001).
Based on previous MST investigations, there is a substantial amount of
information about the genetic diversity, clonality and spatial and temporal distribution of
E. coli strains in different hosts from different environments currently available (Farleiter
et al 2000). This information suggests that it is possible to assign a host source to an
environmental sample ofE. coli. Rep-PCR utilizes PCR amplification ofDNA between
repetitive extragenic elements to obtain strain specific fingerprints. When analyzing 154
E. coli isolates, Rep-PCR using BOX primers has been able to correctly classify 94.7%
3
human, 100% chicken and 100% cow isolates (Dombek et aI., 2000). However, Rep 
PCR using ERIC primers was only capable ofcorrectly classifying 28.6,0 and 76%
human, bovine and pig isolates from a total of62 E. coli isolates (Leung et aI., 2004).
Other M8T methods, such as amplified fragment length polymorphisms (AFLP) and
multiple antibiotic resistance have given classification rates up to 97% when analyzing E.
coli isolates from humans, wildlife and livestock (Guan et aI., 2002). Additionally, when
E. coli isolates were grouped as from either human or non-human sources, ribotyping was
able to accurately identify 245 of247 nonhuman and 38 of40 human E. coli sources
(Carson et aI., 2001) and 67% human and 100% non-human from a total of238 E. coli
isolates (Dombek et aI., 2000). Denaturing gradient gel electrophoresis using the 168 
238 intergenic spacer region ofE. coli was also capable ofcorrectly classifying 72%
human and 94% animal isolates (8eurinck et aI., 2003). While these different M8T
methods prove promising, the influence ofthe various variables on E. coli genetic
diversity still needs more investigation, as do the issues ofthe reproducibility and cost
effectiveness ofthe various methods.
II. E. coli: Ribosomal DNA and 168-238 Intergenic 8pacer Regions
The ribosomal genes in bacteria are part ofa multigene family consisting ofa
various number ofribosomal (rrn) operons depending on the particular species (Cilia et
aI., 1996). Escherichia. coli has 7 rrn operons which consist ofthe 168 rRNA gene, an
intergenic spacer region, the 238 rRNA genes, another intergenic spacer region, and the
58 rRNA gene (figure 1) (Anton et aI., 1998, Fukushima et aI., 2002, Garcia-Martinez et
4
aI., 1999). The ISR are short regions that contain tRNA genes, target sequences for
RNase III and other recognition signals for processing the transcript (Garcia-Martinez et
aI., 1996a), including a well-known consensus antiterminator (Anton et aI., 1998). The 7
rrn operons in E. coli consist oftwo main types of 16S-23S ISR based on the number and
specific tRNA genes present. Escherichia coli has 4 ISR type 1 (ISR1) regions that have
only one tRNA gene encoding tRNAg1u-2 and are located in rrnB, rrnC, rrnE, and rrnG
(figure 2) (Brosius et aI., 1981). There are also three ISR2 regions that have two tRNA
'I 1genes, tRNA
1 e and tRNA a a , and are located in rrnA, rrnD, and rrnH(Anton et aI., 1998,
Garcia-Martinez et aI., 1996a).
After understanding the distribution and arrangement ofthe ISR in E. coli it
becomes easier to understand why this region has the potential to be used as a method of
differentiating bacteria at the sub-species level. The rRNA operon sequences have
become useful to differentiate strains because they are relatively easy to sequence (Cilia
et aI., 1996). Resent research has also utilized the characterization ofthe 16S-23S ISR
for the comparison ofclosely related organisms when 16S rRNA has been inadequate for
discriminating (Nagpal et aI., 1998). The sequences ofthe rRNA genes undergo much
slower divergences than their flanking non-genic sequences (Lia, 2000). This results in
more variability and less homogeneity between ISR sequences compared to the ribosomal
genes. The main reason the ISR has the potential to discriminate between closely related
organisms is because ofthe presence ofnucleotide sequences that apparently neither
transcribe for genes or playa vital role in the secondary structure ofthe ribosomal RNA
operon transcript. This is an important aspect to consider, since the secondary structure is
essential for the cleavage and release ofthe rRNA molecules from the primary transcript
5
Figure 1: Ribosomal operon location in the E. coli genome.
Genomic distribution ofthe rRNA genes. The size (kb) ofthe entire E. coli genome is
indicated in parenthesis. The rRNA genes are depicted as arrows, and the distances (kb)
between each gene are indicated (adapted form Cilia et aI., 1996)
6
E. coli (4639 kb)
rrnH
rrnD
7
Figure 2: Gene organization ofthe rrnB operon ofE. coli.
Horizontal arrows at the promoters specify direction oftranscription. The RNA genes are
indicated by filled bars and the open reading frames are indicated by open bars. The
numbers under the bars indicate the length (bp) ofthe genes and spacers (adapted from
Brosius et al., 1981).
8
ORF T 4416S rRNA tRN AGlu5'
R67 1542
171 193
23S rRNA
2904 593
9
by RNase III at specific recognition sites (Garcia-Martinez et aI., 1999). The functional
units within the spacer region do not sum up to more than 50% ofits whole size, and the
rest ofthe region consists ofnon-essential sequences submitted to frequent
insertion/deletion events as noticed in E. coli (Garcia-Martinez et aI., 1999). An
additional aspect ofthe ISR that makes it particularly useful for differentiating among
closely related species is that the spacer region varies in sequence and in length among
species (Fox et aI., 1998, Garcia-Martinez et aI., 1999, Scheinert et aI., 1996).
Currently, the presence ofISR sequence variations between E. coli strains and
within an individual genome has been observed when looking at E. coli K-12 and other
members ofthe E. coli reference collection (ECOR) (Anton et aI., 1998, Garcia-Martinez
et aI., 1996a). Three main variations were found among the E. coli strains, and include
dispersed nucleotide substitutions at certain locations, grouped variable sites ofdifferent
composition but preserved secondary structure and block substitutions involving putative
insertions or deletions that change the secondary structure (Anton et aI., 1998).
Specifically, E. coli K-12 was shown to have an ISR sequence ofeither 106 bp or 20 bp
upstream ofthe tRNAg1u-2 (ISR1), a block of 14 bases grouped in a stem loop secondary
structure in ISR1 only, and a 17 bp block substitution by a different 8 base sequence
(Anton et aI., 1998). Other differences observed were single base substitutions
differentiating every individual operon and the switching ofan ISR1 for an ISR2 and
consequently an ISR2 for an ISR1 in the genome ofECOR 40 (Anton et aI., 1998).
Further analysis also revealed that all sites prone to nucleotide substitutions seem not to
be involved in secondary structure, and the presence ofa stem-loop with 21 variable
positions in ISR1 with no homologous region in ISR2 (Anton et aI., 1998). These
10
variations, among different E. coli strains, are valuable when using ISR sequence analysis
for strain differentiation.
Ribosomal operons, including the rrn ofE. coli, also exhibit intercistronic
heterogeneity. This is a differing ofoperon sequences within the genome ofan
individual strain. There has been reportedly almost as much variation among different
operons ofthe same strain as among different strains. Rarely are identical operons found
even in the same genome when looking at ECOR samples (Garcia-Martinez et aI.,
1996a). Having as much variation among different operons in an E. coli genome as in
other strains could limit the usefulness ofISR sequencing as a differentiating device
(Nagpal et aI., 1998). This is a very important aspect to consider when using the ISR
sequence to compare E. coli strains.
Intercistronic heterogeneity in E. coli ribosomal operons can be overcome as a
hindrance to strain typing and can be exploited as a way to differentiate between E. coli
samples. PCR amplification ofthe 16S-23S ISR with general ribosomal operon primers
produces a broad mixture ofamplicons from each ofthe 7 rrn operons in E. coli. Direct
sequencing ofthese PCR products produces a mean sequence in which mutations in the
most variable domains become hidden. Cloning a single operon actually results in a
sequence that differs from that ofother operons, and ofthe mean sequence, by several
point mutations. For this reason a mean sequence should be avoided to identify strains at
the species level or below (Cilia et aI., 1996). Besides using cloning to overcome the
intercistronic heterogeneity, primers have been created that allow amplification ofeach
individual rrn operon in E. coli (Anton et aI., 1998). These primers are positioned on
genes or open reading frames upstream ofthe 5' end ofthe 16S rRNA genes and use a
11
universal primer for the 23S rRNA gene. Utilization ofthese primers allows for the
individual amplification ofeach operon for sequencing, avoiding production ofa mean
sequence.
The ribosomal operons in E. coli are complex in compositions having the
sequences for important genes and cleavage sites, along with a complex and important
secondary structure. Even though the ribosomal operons, and genes it encodes, are so
essential to life, there are still some areas within the operon that are prone to higher levels
ofvariation without effecting the transcription ofthe genes. Additionally, these
variations can be as high, ifnot higher, between the operons ofa single genome as
between various strains. The fact that such variations exists, along with the capability to
PCR amplify each individual operon, allows for the comparison ofindividual operons of
one E. coli isolate to that ofanother, and shows promise as a useful tool for
discriminating among samples.
III. Denaturing Gradient Gel Electrophoresis (DGGE)
DGGE is a gel separating method that can be used to distinguish two DNA
molecules that differ by as little as a single base substitution (Sheffield et aI., 1989). The
ability ofDGGE to separate PCR amplified DNA differing by a single base substitution
has contributed to the wide use ofthe method in various fields ofmicrobiology. In fact,
DGGE has become routinely used in microbiology labs around the world as a molecular
tool to compare the diversity ofmicrobial communities and to monitor population
12
dynamics (Muyzer, 1999). In parallel DGGE a gradient ofdenaturants, urea and
formamide, are established in a polyacrylamide gel. DNA samples are loaded onto the
gel and an electric current is applied in the same direction as the denaturants. As DNA
fragments migrate through a denaturing gel they remain double stranded until they reach
a concentration ofdenaturants equivalent to a melting temperature (Tm) that causes the
fragments lower melting temperature domains to melt resulting in the reduced mobility of
the fragment as the DNA denatures (Sheffield et aI., 1989). Sequence variations within
the melting domains ofvarious DNA samples will alter their melting temperatures,
resulting in fragments that stop migrating at different positions in denaturing gels
(Muyzer et aI., 1993). Therefore, it is possible to separate PCR amplified DNA samples
oftarget genes from various sources to determine sequence variations.
DGGE DNA separation is even more enhanced with the addition ofa GC-c1amp
by using specially designed primers. GC-c1amps are a series of40 guanine and cytosine
nuc1eotides that are added to one ofthe PCR primers, and they allow for separation of
single base changes in the highest melting domains (Sheffield et aI., 1989). DNA
fragments up to 1000bp can be separated by DGGE. However, smaller fragments are
easier to separate due to their increased mobility and the presence offewer melting
domains. Specifically, single base changes ofPCR amplicons up to 500 bp joined to a 40
bp GC-c1amp can be separated by DGGE (Farnleitner et aI., 2000, Sheffield et aI., 1989).
13
Chapter 2: Method Development and E. coli source isolates PCR-DGGE
Specific Aims:
The goal ofthis study was to evaluate a microbial source tracking method for
differentiating between strains ofE. coli from known sources based on DGGE analysis of
the 16S-23S ISR. Specifically, the rrnB operon was subjected to PCR amplification
using sequence specific primers. Two new primers, one ofwhich had a GC-Clamp, were
then used to prepare PCR amplified 16S-23S ISR for DGGE analysis (figure 3). These
techniques were performed on a total of 119 E. coli from 8 different sources, which were
generously donated by the United States Geological Survey (USGS). The resulting gels
were compared utilizing an E. coli PCR amplified 16S-23S ISR and commercially
available standards, which were ran on each DGGE gel. In order to verify DGGE results,
and to determine the amount ofsequence variation in the ISR, DNA sequencing ofcloned
ISR was performed. Control experiments were performed to determine the
reproducibility ofthis PCR-DGGE method, and included repeated independent PCR
amplification ofE. coli isolates followed by DGGE, cloning ofan amplified 16S-23S ISR
for PCR efficiency analysis. Additionally, 3 ofthe isolates were carried through a series
ofgenerations over the course ofthe experiment to evaluate any changes in the ISR over
time.
14
Figure 3a and b: Primer locations for peR reactions.
3a: Primer sets PrrnB and 23S1R are isolate the rrnB operon specifically in E. coli by
binding to an open reading frame upstream ofthe 16S rRNA gene and a conserved region
on the 5' end ofthe 23S rRNA gene. *Distance from the 3' end ofthe primer to the 5'
end ofthe l6S rRNA gene. **Distance from the 5' end ofthe primer to the 5' end ofthe
23S rRNA gene.
3b: Primer sets GC-G1 and L2 isolate the ISR, distance form these primers to the ISR are
listed. Open reading frames are designated as open bars, and rRNA genes are indicated
by filled bars.
15
3a:
PrrnB
--+ 955bp* tRNAG1u
23SJR
+-- 92bp**
3b:
ORF 16S rRNA ISR
L2
20b~
23S rRNA
ORF 16S rRNA
tRNAG1u
ISR 23S rRNA
16
Materials and Methods
E. coli isolates and collection:
E. coli isolates from nine animal sources (Canada goose, chicken, cow [beefand
dairy], deer, dog, horse, human, swine) were subcultured from an E. coli isolate
collection from various farms in Berkely County West Virginia. The samples consisted
ofone isolate per individual animal source and 15 individual isolates from the 8 animal
sources for a total of 120 samples. Isolates were grown overnight in 3 ml ofLuria 
Bertani broth at 37?C and then frozen in a 20% glycerol solution. The cells were stored
at -80? C. Each isolate was labeled by using a host source abbreviation followed by the
isolate number (01-15) (appendix A).
DNA Isolation:
E. coli isolates were cultured overnight in Luria-Bertani broth at 37?C. The
genomic DNA was then isolated using the GenElutetm Genomic DNA Kit (Sigma 
Aldrich, Inc.), according to its protocol. Isolated genomic DNA was subjected to PCR
followed by quantitation using agarose gel electrophoresis. The isolated DNA used in the
PCR was diluted by a factor of 1:1, 1:10, 1:100 and 1:1000, to optimize future reactions.
The optimal concentration was determined and used for subsequent reactions. The
isolated DNA was stored at -20?C.
17
PCR Reactions:
PCR I: Reagents for the Polymerase Chain Reaction.
Reagent Amount per Reaction
(1-11
lOx Gold PCR Buffer (Applied Biosystems) 5
25 mM MgClz (Applied Biosystems) 5.5
50 pmol/Ill Primer 23S1R 1
50 pmol/Ill Primer PrrnB 1
10mM dNTPs (Roche) 1
Taq polymerase 5U/IlI (Applied Biosystems) 0.4
H2O 26.1
PCR amplification was carried out in 50 III volumes using 40 III ofthe PCR I mix
and 10 III a 1:10 dilution ofisolated genomic DNA. Primer 23S1R (5' GGG TTT CCCC
A TT CGG AAA TC 3') hybridizes 96bp from the 3' ofthe 23S rRNA gene (Garcia
Martinez et. aI1996b). Primer PrrnB (5' AAC ACT GCC AGT ACC GTT TC 3') binds
to an open reading frame 955 bp from the 5' end ofthe 16S rRNA gene (Anton et aI,
1998). All PCR amplifications were carried out in a MJ Research PTC-200 thermal
cycler. An initial 1 min at 95?C was followed by 35 cycles of: 30s at 94?C, 30s at
56.8?C, 2 min at 72?C. The final cycle was followed by an additional 5 min at 72?C,
18
with a holding temperature of4?C. PCR products were visualized by agarose gel
electrophoresis.
PCR II: Reagents for the Polymerase Chain Reaction.
Reagent Amount per Reaction
(,..1)
lOx Gold PCR Buffer (Applied Biosystems) 5
25 mM MgClz (Applied Biosystems) 5
6 nM Primer GC-G1 2
400 nM Primer L2 2
10mM dNTPs (Roche) 0.2
Taq polymerase 5Ufl..tl (Applied Biosystems) 0.4
H2O 26.8
PCR amplifications were carried out using 40 III ofthe PCR II mix and 1: 100
dilutions ofPCR I products. Primer GC-G1 (5' CGC CCG CCG CGC CCC GCG CCG
T CCC GCC GCC CCC CGC CCC CGA AGT CGT AAC AAG G 3') binds within the
16S rRNA gene, approximately 40 bp upstream ofthe ISR. The G-C rich region of
primer GC-G1 is a GC clamp, and is essential to prevent complete denaturization during
DGGE analysis (Buchan et aI., 2001). Primer L2 (5' CAA GGC ATC CAC CGT 3') is
the reverse primer, and binds to a region ofthe 23S rRNA gene approximately 20 bp
19
downstream ofthe ISR (Jensen et aI., 1993). All PCR amplifications were carried out in
a MJ Research PTC-200 thermal cycler. An initial 3 min at 94?C was followed by 25
cycles of: I min at 94?C, I min at 55?C, 2 min at noc. The final cycle was followed by
an additional 7 min at noc, with a holding temperature of4?C. PCR products were
visualized by agarose gel electrophoresis.
Agarose Gel Electrophoresis:
PCR products and restriction digests were run on I% high resolution agarose
(Sigma, MO) using IX TAE buffer (40 rnM Tris, 20 rnM Acetic Acid, ImM EDTA, pH
8.3). All agarose gels were run at 84 V for 45 min. Gels were stained in 1% ethidium
bromide solutions for 20 min, followed by a 5 min rinse in water. The resulting DNA
bands were visualized using a UV light, and the size and approximate quantity ofDNA
present were determined by comparison to a molecular weight marker (Biomarker EXT
Plus, Invitrogen, CA).
Denaturing Gradient Gel Electrophoresis (DGGE)
Parallel denaturing gradient gel electrophoresis was performed on the amplified
ISR samples (PCR II products) using the DCode Universal Mutation Detection System
(BioRad Inc., CA) to detect nucleic acid differences among the samples. Samples were
loaded onto 8% (wtlvol) polyacrylamide gels. The gels were prepared using 30% and
60% denaturant acrylamide solutions (30% stock: 40% acrylamide/Bis stock solution
(BioRad Inc., CA), 12% deionized formamide, 2.1 M urea, IX TAE buffer; 60% stock:
40% acrylamide/Bis stock solution, 24% deionized formamide, 4.2 M urea, IX TAE
20
buffer). The amount ofsamples loaded was determined from the agarose gel
electrophoresis, and typically ranged from 3-5 Jll (approximately 20 ng DNA).
Electrophoretic charge was applied in the same direction as denaturants. All gels were
run in IX TAE buffer for 3h 45min at 60?C and 150 V. The amount oftime to run each
gel was determined by a time course experiment, in which samples were added to the gel
each hour for three hours and the gel was ran for a total of6 hours and 45min. After
electrophoresis, each gel was stained in 1% ethidium bromide solution for 7 min,
followed by a 10 min de-staining in 1 X TAE. Images were captured using Eagle Eye II
image capturing (Stratagene, CA) and saved as digital files.
To allow for comparisons between the denaturing gels, relative distances (RD)
were calculated by dividing the distance the individual samples migrated by the distance
ofthe E. coli standard on the same gel.
Four independent PCR amplifications were carried out on one E. coli isolate,
Canada goose isolate 11 (Cgll) and then confirmed to have the same DGGE banding
pattern. The E. coli PCR product was then subsequently used as a relative standard to
compare banding patterns between gels. DCode wild type and mutant controls (BioRad,
CA) were also run on each gel to show the capability ofthe gels to differentiate between
single nucleotide differences. Additional isolates were also separately PCR amplified
and subjected to DGGE to confirm repeatability.
Statistical Analysis ofDGGE Results:
The diversity ofthe RD values for each E. coli source was calculated using the
Shannon-Weaver index. This index has been shown to be a useful measure ofdiversity
21
for microbial communities, and is the most commonly used method for calculating
diversity (Mills et aI., 1980). DGGE band diversity for each ofthe sources was
k
n logn - Ijilogji
calculated as follows: H'= ;;1n
Where n = number ofsamples, k = the number ofcategories, or different possible DGGE
bands,fi =number or frequency ofDGGE bans present for a given category. The higher
the value ofH', the higher the diversity.
The maximum possible diversity for each E. coli source was calculated using
H'max =log k. The magnitude ofH' is affected by both the distribution and the number of
categories. Therefore, by calculating the evenness (J'), the observed diversity is
presented as a proportion ofthe maximum possible diversity. Evenness was calculated
. h ?": 11? . J' H'usmg t e 10 owmg equatIOn: = ---
H'max
SPSS 11.5 software (SPSS, IL.) was used to for bivariate correlation analysis of
the DGGE band distributions. The Pearson correlation coefficient was calculated, and
the significance (2-tailed) ofcorrelation was given for each sources compared.
Cloning:
Selected E. coli isolates were subjected to cloning and DNA sequencing ofthe
ISR based on DGGE banding patterns. Within 24 hr ofamplification, ISR PCR products
were ligated into pCR 4-TOPO vector (TOPO TA Cloning Kit for Sequencing,
Invitrogen, CA) following the manufacturer's instructions. The ligation reaction was set
up as follows: 2.5 ~l PCR products, 1.0 ~l salt solution, 1.5 ~l H20, 1.0 ~l vector. The
ligation reaction was then transformed into One Shot? TOP10 Competent cells following
22
the manufacturer's instructions. Following transformation, 50 and 100 III ofthe
transformed cells were spread onto pre-warmed (37?C) LB-ampicillin (100 Ilg/1l1
ampicillin) and incubated at 37?C for 24 h. The vector contains the lethal E. coli ccdB
gene fused to the C-terminus ofthe LacZa fragment. When the PCR product ligates to
the vector the LacZa -ccdB gene fusion is disrupted permitting the growth ofpositive
recombinants. Cells that contain non-recombinant vector are killed upon plating on
plates containing ampicillin. After each cloning procedure, 5 colonies were selected from
each plated isolate to analyses positive insertion by restriction digest with EcoRl.
Restriction Digests:
Following cloning, colonies were isolated and grown overnight in LB-ampicllin
broth at 37?C for 24 h. Plasmid DNA was then isolated using the Cyclo-Prep Plasmid
DNA isolation kit (Amresco, OH) and eluted in 65 III H20. Plasmid DNA was then
digested with EcoRl as follows: 9.6 III H20, 3 III Buffer, and 0.21l1 enzyme were
combined with 17 III ofplasmid DNA and incubated for 3h at 37?C. Digests were then
visualized by loading 5 III ofdigest onto a 1% agarose gel and electrophoresed for 45 min
at 85V. Positive inserts were noted as having an inserted 500bp band. A molecular size
marker (Biomarker EXT Plus, Invitrogen, CA) was used to determine the size ofthe
insert and to allow quantification ofthe plasmid DNA for DNA sequencing.
DNA Sequencing:
Isolated plasmids that showed successful insertion ofthe ISR PCR product were
used for DNA sequence analysis. The DNA sequencing reaction was prepared using the
23
CEQ 2000 Dye Terminator Cycle Sequencing Quick Start kit (Beckman Coulter Inc.,
CA) according to the manufacturer's instructions. PCR dye labeling was amplified
following the manufacturer's program kit (Beckman Coulter Inc., CA) using the M13F
and M13R primers for the pCR 4-TOPO plasmid (Invitrogen, CA). The DNA
sequencing reaction was ethanol precipitated following the manufacturer's instructions
(Beckman Coulter Inc., CA) and resuspended in 40 III sample loading solution and stored
at -20?C. Samples were sequenced on a Beckman Coulter CEQ 2000XL Dye Terminator
Cycle DNA Sequencer.
Long Term Cultures:
Three isolates (cow, swine and human) were chosen and ran through a series of
cultures. Samples were grown on LB plates overnight at 37?C and then transferred to LB
broth tubes for another overnight culture at 37?C. This was repeated over the course of
the experiment. Genomic DNA was isolated from the cultures before the time course
experiment began, and after 11 culture transfers. The ISR was then PCR amplified
following the same protocols as all other isolates and subjected to DGGE analyses to see
what affects multiple cultures has on the genetic stability ofthe ISR over multiple
generations ofE. coli growth.
24
Chapter 3: Results
ISR Amplification:
The ISR of 119 E. coli isolates from 8 different sources were subjected to PCR
amplification to look for DNA sequence variations among the isolates by DGGE
analysis. To avoid producing a composite sequence ofall 7 E. coli ribosomal operons,
primers specific to the rrnB (Anton et aI., 1998) were used for the first PCR (PCR I)
amplification. The size ofresulting PCR products was then determined by agarose gel
electrophoresis and compared to a size standard. Amplification ofall isolates with the
rrnB specific primers yielded a single product ofapproximately 3000 bp (figure 4),
which is consistent with other results using the same primer set (Anton et aI., 1998).
After successful amplification ofthe rrnB operon, the ISR was specifically
targeted using primers adapted from Jensen et aI., 1995. This second PCR (PCR II)
resulted in the amplification ofthe rmB ISR, which could then be subsequently used for
DGGE analysis. PCR II primers incorporated a GC-Clamp into products, which allows
for DGGE analysis. Depending on PCR I results, dilutions of 1: 100 or 1: 1000 ofPCR I
were used for the PCR II reactions. Resulting PCR products were ran on agarose gels to
determine fragment size. Previous studies using the rrnB ISR specific primers on
isolated E. coli genomic DNA have resulted in producing amplicons of480 and 540 bp
corresponding to ISR I and ISR II respectively (Buchan et. aI., 2001) or approximately
530 bp for rrnB, rrnG, rrnD the hybrid operon rrnX (Jensen et aI., 1993). In this study,
ISR PCR amplification ofrrnB operons revealed products ofapproximately 500 bp or
25
Figure 4: PCR amplification resulting in the isolation ofthe rrnB operon from
chicken isolates.
Legend: Chicken isolate numbers are listed
Lane I-BioMarker Ext PIUS 50-2500bp ladder
Lane 2-01
Lane 3-03
Lane 4-04
Lane 5-06
Lane 6-07
Lane 7-08
Lane 8-09
Lane 9-10
Lane 10-11
Lane 11-12
Lane 12-13
Lane 13-14
Lane 14-15
Lane 15-blank
26
1 2 3 4 5 6 7 8 9 10 1112 13 14
a ? ? ? ? ,
2500 bp
27
580 bp (figure 5). Each E. coli isolate that had the ISR amplified produced a single
dominate band ofeither ofthese two listed sizes. However, on certain occasions, the
PCR II products would result in two discrete double bands. These double bands were
resolved into single products by increasing the amount ofPCR I product used for the ISR
amplification (figure 6). Comparing the bands offigures 5 and 6 shows that the double
bands were able to be resolved into single bands. The size ofthe double bands present
seemed to be proportionally different from each other in all samples in which they
occurred. The cause ofthe double bands in unknown, and would require additional
investigation by DNA sequence analysis. Additionally, a faint band was often observed
around 1000 bp. The occurrence ofa faint 1000 bp band has been noted elsewhere when
using the same primer set (Buchan et aI., 200I).
Controls:
In addition to the commercially available DGGE controls, an E. coli control was
created to use for between gel comparisons. Four 100 III PCR reaction were performed
to amplify the ISR ofone E. coli isolate, Cg11. The resulting products were run on a
DGGE gel to confirm that they all would show the same gel migration patterns (figure 7).
After confirmation ofsimilar migrations, the E. coli standard, along with the commercial
controls (BioRad Inc., CA) was run on each DGGE gel to allow comparison between
gels.
In order to determine ifsequence variations apparent in different DGGE bands
were due solely on naturally occurring nucleotide differences and not from Taq
28
Figure 5: PCR amplification resulting in the isolation ofthe rrnB 16S-23S ISR from
chicken isolates.
Legend: Chicken isolate numbers are listed
Lane I-BioMarker Ext PIUS 50-2500bp ladder
Lane 2-01
Lane 3-03
Lane 4-04
Lane 5-05
Lane 6-06
Lane 7-07
Lane 8-08
Lane 9-09
Lane 10-10
Lane 11-11
Lane 12-12
Lane 13-13
Lane 14-14
Lane 15-15
29
500bp .....
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
30
Figure 6: Resolving ofdouble bands that were present after PCR amplification of
the rrnB 16S-23S ISR.
The gel pictured shows double bands that were present for chicken isolates 01-15.
Legend: Chicken isolate numbers are listed
Lane I-BioMarker Ext PLUS 50-2500bp ladder
Lane 2-01
Lane 3-03
Lane 4-04
Lane 5-05
Lane 6-06
Lane 7-07
Lane 8-08
Lane 9-09
Lane 10-10
Lane 11-11
Lane 12-12
Lane 13-13
Lane 14-14
Lane 15-15
31
500bP .....
1 2 3 4 5 6 7 8 9 10 1112 13 14 15
32
Figure 7: DGGE gel ofE. coli standard that was used to compare gel migrations
between gels.
Legend:
Lane I-Blank
Lane 2-Wildtype commercial control
Lane 3-Mutant commercial control
Lane 4- E. coli control
Lane 5- E. coli control
Lane 6- E. coli control
Lane 7- E. coli control
Lane 8-13-Blank
33
1 2 3 4 5 6 7 8 9 10 11 12 13
34
polymerase inefficiencies, E. coli isolate Cgll PCR amplified rrnB ISR was cloned and
subjected to ten independent PCR reactions. Resulting products were ran on DGGE gels
to confirm consistency in PCR amplification between samples. The resulting DGGE gels
showed that each independent PCR ofthe same clone gave the same gel migration
distance (figure 8).
DGGE Analysis ofthe rrnB ISR of119 E. coli Isolates:
DGGE analysis ofthe rrnB ISR from 119 E. coli isolates revealed the presence ofa total
of40 different possible gel migrations ofthe single band ISR product based on relative
distance (RD) values. The 40 possible DGGE bands that were present were labeled A 
00 and the frequency ofoccurrence ofeach band was recorded for each animal source
(table 1). The distribution ofRD values for each animal source overlapped quite a bit
towards the middle range. For several ofthe sources, there were unique bands present,
and when there were similar bands present, they were often present in different
frequencies. Consequently, each animal source has different frequency ofbands and
distributions, which may allow for differentiation between sources. Table 2 shows the
variations in DGGE patterns between the human and animal sources. Ofthe 15 Canada
goose E. coli isolates, a total ofnine different DGGE bands were present. Each Canada
goose sample produces a single band representing the amplified rrnB ISR. Therefore, the
ISR bands present from the 15 Canada goose samples were distributed between 9 band
possibilities on the denaturing gels. The number ofthe 15 total Canada goose samples
that were unique compared to all other sources was 7. As a result, 47 % ofall Canada
goose E. coli ISR sequences were unique.
35
Figure 8: DGGE gel ofindependently cloned Canada goose isolated number 11.
Legend:
Lane 1- E. coli control
Lane 2- Cg 11 clone 1
Lane 3- Cg 11 clone 2
Lane 4- Cg 11 clone 3
Lane 5- Cg 11 clone 4
Lane 6- Cg 11 clone 5
Lane 7- Cg 11 clone 6
Lane 8- Cg 11 clone 7
Lane 9- Cg 11 clone 8
Lane 10- Cg 11 clone 9
Lane 11- Cg 11 clone 10
Lane 12- E. coli control
36
1 2 3 4 5 6 7 8 9 10 11 12
37
Table 1: Frequency and distribution ofgel migrations ofthe E. coli isolates from the
various sources.
The table shows how the DGGE bands establish a genetic profile for each animal source.
There are unique bands present for many ofthe sources. There are overlaps in the middle
range, but the frequencies are often different between the sources. All nonhuman sources
were pooled together for comparison with human source isolates.
38
Sources
Canadian Swine Cattle Human Deer Dog Horse Chicken Nonhuman
Goose
A 3 0 0 0 0 0 0 0 3
B 1 0 0 0 0 0 0 0 1
C 0 0 0 2 0 0 0 0 0
0 0 0 0 0 0 1 0 0 1
E 0 0 0 1 0 0 0 0 0
F 0 0 1 0 0 0 0 0 1
G 1 0 0 0 0 0 0 0 1
H 3 0 1 0 0 0 0 0 4
I 0 0 0 0 0 0 0 1 1
J 1 0 0 0 0 0 0 0 1
K 0 0 0 0 1 0 0 0 1
L 0 0 0 0 1 0 0 0 1
M 0 1 0 0 0 0 1 0 2
N 0 0 0 2 0 0 0 0 0
0 0 0 0 0 0 0 0 1 1
P 0 0 2 0 2 3 1 1 9
Q 0 0 1 0 0 0 0 0 1
R 0 0 0 1 0 0 2 0 2
S 0 3 0 0 0 0 1 2 6
T 0 1 0 1 1 0 0 0 2
U 0 0 2 0 3 2 0 1 8
V 0 0 1 0 2 0 0 3 6
W 0 0 1 0 0 0 2 0 3
X 0 0 0 0 0 0 1 0 1
Y 0 2 0 1 1 4 0 1 8
Z 3 0 1 0 0 0 1 3 8
AA 1 4 1 0 0 0 1 0 7
BB 0 0 0 2 0 0 0 0 0
CC 0 0 0 0 0 1 1 0 2
DO 1 1 1 0 1 0 3 1 8
EE 0 2 1 1 1 0 0 0 4
FF 0 0 1 0 1 0 0 1 3
GG 0 0 0 0 0 1 1 0 2
HH 0 1 0 1 0 1 0 0 2
II 0 0 0 0 0 1 0 0 1
JJ 0 0 1 1 0 0 0 0 1
KK 0 1 0 0 0 0 0 0 1
LL 0 0 0 1 0 0 0 0 1
MM 1 0 0 0 0 0 0 0 1
NN 0 0 0 0 0 1 0 0 1
00 0 0 0 1 0 0 0 0 0
39
Table 2: Distribution and frequency ofDGGE results compared among the isolates.
The different DGGE profiles established for each source are compared. The number and
percentage ofunique bands present are recorded, and the diversity and evenness are given
in the table.
40
Source
Canadian
Goose Swine Cattle Human Deer Dog Horse Chicken Total
~umber of Isolates 15 15 15 15 14 15 15 15 119
~umber of Different
)GGE Bands Present 9 8 13 12 10 9 11 10 40
~umber of
Unique RDs 7 1 2 9 2 2 1 2 26
'ercentage of Unique
RD Values per
Number of Isolates 47 7 13 60 14 13 7 13 21
,hannon Weaver
Diversity H' 0.89 0.84 1.14 1.06 0.96 0.88 1 0.95 1.46
~' Max= 0.95 0.95 1.11 1.08 1 0.95 1.04 1 1.6
:venness J' 0.93 0.88 1 0.98 0.96 0.92 0.96 0.95 0.91
41
For all ofthe animal sources, the number ofpossible DGGE bands ranged from 8
to 13. The number ofunique DGGE bands ranged from 1 for swine and horse to 9 for
human E. coli isolates. Each source produce unique RD values, with the highest
proportion ofunique DGGE bands, 60%, found for human isolates. Cattle, dog and
chicken isolates each had 13% unique DGGE bands for their respective E. coli isolates.
The Shannon-Weaver Diversities (H') ofall samples were moderately high, ranging from
H '=0.84-1.14. All evenness values (J') were also high, ranging from J '=0.92-1. These
results show that there is high diversity among the possible DGGE bands present for each
animal source, and that the bands identified were evenly distributed among the possible
RD values for each human or animal source. Therefore, it is likely that any additional E.
coli isolates tested for a given source will remain unique to that particular source.
Effective MST methods should be capable ofdistinguishing human sources ofE.
coli from nonhuman sources, as the presence ofmicroorganisms from human origin are
regarded as having a greater potential to cause disease in humans (Guan et aI., 2002).
Therefore, all ofthe nonhuman sources DGGE results were combined for comparison
with the human results (table 3). The samples sizes are quite different (human n=15,
nonhuman=104). However, the results give an insight to the amount ofvariation between
the two groups. Human source isolates had 60% unique RD values, while nonhumans
had 80% unique DGGE bands. This is substantially higher than any other individual
nonhuman source. Combining the nonhuman sources also considerably increased the
diversity (H'=1.39) compared to any individual nonhuman source, and the evenness
remained relatively high (J'=0.91).
42
Table 3: Distribution and frequency ofDGGE results compared between human
and nonhuman isolates.
The different DGGE profiles established for human and nonhuman sources are compared.
The number and percentage ofunique bands present are recorded, and the diversity and
evenness are given in the table.
43
Source
Non-
Human Human
Number of Isolates 15 105
Number of Different
DGGE Bands Present 12 34
Number of
Unique RDs 9 83
% Unique RD Values
per Number of
Isolates 60 80
Shannon Weaver
Diversity H' 1.1 1.39
H'Max= 1.08 1.53
Evenness J' 1 0.91
44
Pearson correlation coefficients were calculated for the RD values ofall the
isolates tested. Table 4 shows the results ofthe correlation test between the DGGE bands
from the 8 animal sources. The correlation calculations take into consideration both the
frequency and distribution ofall RD values for each isolate source, establishing a genetic
profile or fingerprint for each source. These profiles are then compared to each other for
any correlation between the RD values. The only sources that showed any correlation
between isolate profiles were cattle/deer, cattle/chicken, deer/dog, and deer/chicken. All
other source profiles showed no correlation. Therefore, based on the isolates tested, the
majority ofthe sources could be differentiated based on their rrnB 16S-23S ISR DGGE
results. These results suggest that even though the diversity ofRD values is high among
all the sources tested, the bands that were present for each source were still unique, and
that further testing would likely reveal unique RD values for each source. It is important
to note that the sample sizes for each source was low (15 isolates/source). Nonetheless,
the results reveal the potential ofDGGE analysis ofthe rrnB 16S-23S ISR for
differentiating between E. coli isolates.
When the human and nonhuman sources were compared using Person correlation
coefficients, the two sources were nearly negatively correlated (-0.290, P=0.065 2-tailed).
Regardless, the two sources were still not correlated by RD values, and the unique
profiles established for DGGE RD values were capable ofdistinguishing each source.
Further analysis with a larger set ofhuman isolates would reveal more information about
exactly how the two sources genetic profiles are related. The data is also useful for
establishing field testing studies. It may not be necessary to determine specific sources
45
Table 4: Correlation of DGGE Bands from 8 Animal Sources.
The table shows the correlation between the DGGE bands from the 8 animal sources.
Only CattlelDeer, Cattle/Chicken, Deer/Dog, and Deer/Chicken showed any correlation
in the DGGE bands produced from amplified ribosomal B ISR. All other comparisons
have no correlation between DGGE bands. Number values listed are Pearson correlation
values.
*Correlation is significant at the 0.05 level (two-tailed).
**Correlation is significant at the 0.01 level (two-tailed).
46
Source Goose Swine Cattle Human Deer Do Horse Chicken
Goose
Swine
Cattle
Human
Deer 0.613**
Dog 0.457**
Horse
Chicken 0.365* 0.418**
47
for all nonhuman sources ifhuman and nonhuman sources can be confidently
distinguished.
To confirm the ability ofdenaturing gels to separate PCR amplified 16S-23S ISR,
specific isolates that gave the same, and slightly different, RD values for DGGE analysis
were cloned and sequenced. Positive clones were analyzed by restriction digest ofthe
isolated plasmid with EcoR1 endonuclease. EcoR1 restriction sites flank the insert on the
plasmid, so resulting fragments ofapproximately 550 bp were observed by agarose gel
electrophoresis (figure 9). The Isolates Cg11 (E. coli standard for DGGE gels) and Cd1
both gave RD values of 1.0. Sequence analysis ofthe 16S-23S ISR from these isolates
confirmed that the sequence were in fact identical (figure 10), and that the size ofthe
PCR fragment was 480 bp, as expected (Buchan et al.,2001, Jensen et al.,1993). Isolates
Sw15 had an RD value of 1.11, and isolate Dol had an RD values of 1.05, and
subsequent DNA sequence analysis confirmed the sequences did actually differ in 7
locations (figure 11). Additionally, an RD of0.96 was recorded for isolate Cg4, and
isolate Sw5 had an RD of0.93. The difference in RD values was confirmed by the
presence ofnucleotide differences determined from DNA sequence analysis (fig 12).
The two sequences differed in 6 locations. When the DNA sequences from all ofthe
isolates were aligned there were individual nucleotide differences between samples with
different RD values (figure 13). DGGE analysis ofsample Sw4 E. coli isolate gave a RD
value of0.81. This isolate did not align with any ofthe other isolates. There was an 80
nt block insert that was not present in any ofthe other isolates that were analyzed by
DNA sequencing (fig ofall). The results ofthe DNA sequence analysis confirm that
48
Figure 9: Restriction digest ofisolated plasmids from positive clones.
Legend:
Lane 1- Blank
Lane 2- BioMarker Ext PLUS 50-2500bp ladder
Lane 3- Canada goose clone plasmid showing insert
Lane 4- Blank
Lane 5- Canada goose clone plasmid showing insert
Lane 6-15- Blank
49
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
50
Figure 10: DNA sequence alignment oftwo isolates with the same DGGE RD value.
The following alignment shows the DNA sequence ofCd1 and Cgl1. Both samples had
DGGE RD values of 1.00. The sequences are 100% homologous.
51
10 20 30 40 50 60 70 80 90 100
?? .. 1.. ??1 .... 1.. ??1 .... 1?? .. 1.. ??1 .. ??1? .. ?1 .... 1.... 1?? .. 1.... 1.... 1????1 .. ??1????1 .... 1.. ??1????1
CgllR CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
CgllF CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
CdlR CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
CdlF CGCCCGGCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
110 120 130 140 150 170 180 190 200
.. ??1????1????1 .... 1.... 1.... 1.. ??1 .. ??1? .. ?1? .. ?1? .. ?1 .. ??1????1 .... 1????1? .. ?1 .. ??1 .... 1.... 1.. ??1
CgllR GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
CgllF GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
CdlR GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
CdlF GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
210 220 230 240 250 260 270 280 290 300
?? .. 1.... 1.. ??1????1????1????1.. ??1 .. ??1?? .. 1.... 1.... 1.. ??1????1? .. ?1 .... 1????1????1?? .. 1.... 1.... 1
CgllR GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
CgllF GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
CdlR GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
CdlF GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
310 320 330 340 350 360 380 390 400
.... 1.. ??1????1 .... 1.... 1.... 1.... 1????1? .. ?1 .. ??1 .... 1.... 1????1 .... 1.... 1.... 1????1????1 .... 1.... 1
CgllR TCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAAA
CgllF TCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAAA
CdlR TCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAAA
CdlF TCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAAA
410 420 430 440 450 460 470 480
.... 1.. ??1 .. ??1????1 .... 1.... 1.... 1.... 1????1? .. ?1?? .. 1.... 1.. ??1 .... 1? .. ?1????1??
CgllR TTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTC
CgllF TTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTC
CdlR TTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTG
CdlF TTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTG
52
Figure 11: DNA sequence alignment oftwo isolates with different DGGE RD values.
The following alignment shows the DNA sequence ofSwl5 and Dol. Sample Swl5 had
an DGGE RD value of 1.11, Dol RD value of 1.05. Sequence difference occur at
positions 177,220,280,283,284,288 and 379.
53
10 20 30 40 50 60 70 80 90 100
????1????1????1 .... 1.... 1.... 1.... 1.... 1? .. ?1 .. ??1 .... 1.... 1.... 1.... 1? .. ?1 .... 1.... 1.... 1.... 1.... 1
Swl5R CGCCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAA
Swl5F CGCCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAA
DolF CGCCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAA
DolR CGCCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAA
110 120 130 140 150 160 170 ~ 180 190 200
?? .. 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1?? .. 1.... 1.... 1.... 1.... 1.... 1.. ??1 .... 1.. ??1? .. ?1? .. ?1
Swl5R AGAAGCGTACTTTGCAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCC
Swl5F AGAAGCGTACTTTGCAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCC
DolF AGAAGCGTACTTTGCAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGACACGTCCCCTTCGTCTAGAGGCCC
DolR AGAAGCGTACTTTGCAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGACACGTCCCCTTCGTCTAGAGGCCC
.... I... ~t~ .I? .. ~f~ ... I... ~~~ ... I... ~f~ ... I... ~f~ ... I... ~f~ ... I... ~~~ ... I... ~~~ *1 ~~~ ... I... ~~O
Swl5R AGGACACCGCCCTTTCACGACGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCGCCGACCTCAATATCTCAAAA
Swl5F AGGACACCGCCCTTTCACGACGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCGCCGACCTCAATATCTCAAAA
DolF AGGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAA
DolR AGGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAt tt
310 320 330 350 360 370 380 400
.... 1????1 .... 1? .. ?1 .... 1.... 1.... 1? .. ?1 .... 1.... 1.... 1.... 1.... 1.... 1.. ??1 .... 1.... 1.... 1.... 1.... 1
Swl5R CTCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAA
Swl5F CTCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAA
DolF CTCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAGAGTTGTTCGTGAGTCTCTCAA
DolR CTCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAGAGTTGTTCGTGAGTCTCTCAAt
410 420 430 440 450 460 470 480
.... 1.... 1.. ??1 .... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1? .. ?1 .. ??1 .... 1.. ??1
Swl5R ATTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCC
Swl5F ATTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCC
DolF ATTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCC
DolR ATTTTCGCAACACGATGATGAATCGCAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCC
54
Figure 12: DNA sequence alignment oftwo isolates with different DGGE RD values.
The following alignment shows the DNA sequence ofCg4 and Sw5. Sample Sw5 had a
DGGE RD value of0.93, Cg4 RD value of 1.05. Sequence difference occurs at positions
64,378,417 and 425.
55
SlI'5R
sw5F
Cq4R
Cq4F
SlI'5R
sw5F
Cq4R
Cq4F
SlI'5R
sw5F
Cq4R
Cq4F
SlI'5R
sw5F
Cq4R
Cq4F
SlI'5R
sw5F
Cq4R
Cq4F
10 20 30 40 50 60 ~ 70 80 90 100
????1????1????1????1????1????1????1????1????1????1????1.... 1.... 1.... 1.... 1? .. ?1 .... 1.... 1? .. ?1 .... 1
CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTAGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTGGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
CGCCCGCCGCGCCCCGCGCCGTCCCGCCGCCCCCCGCCCCCGAAGTCGTAACAAGGTAACCGTGGGGGAACCTGCGGTTGGATCACCTCCTTACCTTAAA
110 120 130 140 150 160 170 180 190 200
.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1? .. ?1 .... 1.... 1.... 1.... 1.... 1.... 1.... 1
GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
GAAGCGTTCTTTGAAGTGCTCACACAGATTGTCTGATAGAAAGTGAAAAGCAAGGCGTCTTGCGAAGCAGACTGATACGTCCCCTTCGTCTAGAGGCCCA
210 220 230 240 250 260 270 280 290
.... 1? .. ?1 .... 1.... 1.... 1? .. ?1? .. ?1? .. ?1? .. ?1? .. ?1? .. ?1? .. ?1 .... 1.... 1? .. ?1 .... 1.... 1.... 1? .. ?1 .... 1
GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
GGACACCGCCCTTTCACGGCGGTAACAGGGGTTCGAATCCCCTAGGGGACGCCACTTGCTGGTTTGTGAGTGAAAGTCACCTGCCTTAATATCTCAAAAC
310 320 330 340 350 360 370 380 390
.... 1.... 1.... 1.... 1.... 1? .. ?1 .... 1.... 1.... 1? .. ?1 .... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1.... 1
TCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAGAGTTGTTCGTGAGTCTCTCAAA
TCATCTTCGGGTGATGTTTGAGATATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAGAGTTGTTCGTGAGTCTCTCAAA
TCATCTTCGGGTGATGTTTGAGGTATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAACGAAAGTTGTTCGTGAGTCTCTCAAA
TCATCTTCGGGTGATGTTTGAGGTATTTGCTCTTTAAAAATCTGGATCAAGCTGAAAATTGAAACACTGAACAAC~GTTGTTCGTGAGTCTCTCAAA
.... I... :~~ ... I. ~ :~~ ... I~ ...4~~ ... I... :f~ ... I... :f~ ... I... :f~ ... I....4~~ ... I... :~~.
TTTTCGCAACACGATGATGAATCGAAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTC
TTTTCGCAACACGATGATGAATCGAAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACCGTGGATGCCTTC
TTTTCGCAACACGATGGTGAATCGTAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTG
TTTTCGCAACACGATGGTGAATCGTAAGAAACATCTTCGGGTTGTGAGGTTAAGCGACTAAGCGTACACGGTGGATGCCTTG
56
Figure 13: DNA sequence alignment ofall isolates that were sequenced.
The following alignment shows the DNA sequence ofall 7 isolates sequenced. Sw4
differs by all other isolates by 90bp.
57
Cg11F
Cd1F
Sw15F
Do1F
sw5F
Cg4F
Sw4F
???? I ?.?? I ???? I ???? I .... I ???? I
5 15 25
CGCCCGCCGC GCCCCGCGCC GTCCCGCCGC
CGCCCGGCGC GCCCCGCGCC GTCCCGCCGC
CGCCCGCCGC GCCCCGCGCC GTCCCGCCGC
CGCCCGCCGC GCCCCGCGCC GTCCCGCCGC
CGCCCGCCGC GCCCCGCGCC GTCCCGCCGC
CGCCCGCCGC GCCCCGCGCC GTCCCGCCGC
CGCCCGCCGC GCCCCGCGCC GTCCCGCCGC
.... I ???? I ..?? I ?..? I ??.? I ???? I
35 45 55
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
CCCCCGCCCC CGAAGTCGTA ACAAGGTAAC
Cg11F
Cd1F
Sw15F
Do1F
sw5F
Cg4F
Sw4F
???? I ?..? I ???? I ???? I
65 75
CGTAGGGGAA CCTGCGGTTG
CGTAGGGGAA CCTGCGGTTG
CGTAGGGGAA CCTGCGGTTG
CGTAGGGGAA CCTGCGGTTG
CGTAGGGGAA CCTGCGGTTG
CGTGGGGGAA CCTGCGGTTG
CGTAGGGGAA CCTGCGGTTG
.... I ..?? I
85
GATCACCTCC
GATCACCTCC
GATCACCTCC
GATCACCTCC
GATCACCTCC
GATCACCTCC
GATCACCTCC
?... I ???. I
95
TTACCTTAAA
TTACCTTAAA
TTACCTTAAA
TTACCTTAAA
TTACCTTAAA
TTACCTTAAA
TTACCCTAAA
???? I ???? I
105
GAAGCGTTCT
GAAGCGTTCT
GAAGCGTACT
GAAGCGTACT
GAAGCGTTCT
GAAGCGTTCT
GAAGCGTACT
???? I ???? I
115
TTGAAGTGCT
TTGAAGTGCT
TTGCAGTGCT
TTGCAGTGCT
TTGAAGTGCT
TTGAAGTGCT
TTGTAGTGCT
Cg11F
Cd1F
Sw15F
Do1F
sw5F
Cg4F
Sw4F
???? I ???? I ???? I ???? I ???? I ???? I
125 135 145
CACACAGATT GTCTGATAGA AAGTGAAAAG
CACACAGATT GTCTGATAGA AAGTGAAAAG
CACACAGATT GTCTGATAGA AAGTGAAAAG
CACACAGATT GTCTGATAGA AAGTGAAAAG
CACACAGATT GTCTGATAGA AAGTGAAAAG
CACACAGATT GTCTGATAGA AAGTGAAAAG
CACACAGATT GTCTGATAGA AAGTGAAAAG
???? I ???? I ???? I ???? I
155 165
CAAGGCGTCT TGCGAAGCAG
CAAGGCGTCT TGCGAAGCAG
CAAGGCGTCT TGCGAAGCAG
CAAGGCGTCT TGCGAAGCAG
CAAGGCGTCT TGCGAAGCAG
CAAGGCGTCT TGCGAAGCAG
CAAGGCGTTT ACGCGTTGGG
???? I ??.. I
175
ACTGATACGT
ACTGATACGT
ACTGATACGT
ACTGACACGT
ACTGATACGT
ACTGATACGT
AGTGAGGCTG
Cg11F
Cd1F
Sw15F
Do1F
sw5F
Cg4F
Sw4F
???? I .?.. I ..?? I ???? I ???. I ...? I ??.? I ..?. I ???? I ..?? I .... I ???? I
185 195 205 215 225 235
CCCCTTCGTC TAGAGGCCCA GGACACCGCC CTTTCACGGC GGTAACAGGG GTTCGAATCC
CCCCTTCGTC TAGAGGCCCA GGACACCGCC CTTTCACGGC GGTAACAGGG GTTCGAATCC
CCCCTTCGTC TAGAGGCCCA GGACACCGCC CTTTCACGAC GGTAACAGGG GTTCGAATCC
CCCCTTCGTC TAGAGGCCCA GGACACCGCC CTTTCACGGC GGTAACAGGG GTTCGAATCC
CCCCTTCGTC TAGAGGCCCA GGACACCGCC CTTTCACGGC GGTAACAGGG GTTCGAATCC
CCCCTTCGTC TAGAGGCCCA GGACACCGCC CTTTCACGGC GGTAACAGGG GTTCGAATCC
AAGAGAATAA GGCCGTTCGC TTTCTATTAA TGAAAGCTCA CCCTACACGA AAATATCACG
Cg11F
Cd1F
Sw15F
Do1F
sw5F
Cg4F
Sw4F
???? I ???? I ???? I ???? I
245 255
CCTAGGGGAC GCCACTTGCT
CCTAGGGGAC GCCACTTGCT
CCTAGGGGAC GCCACTTGCT
CCTAGGGGAC GCCACTTGCT
CCTAGGGGAC GCCACTTGCT
CCTAGGGGAC GCCACTTGCT
CAACGCGTGA TAAGCAATTT
???? I ???? I ???? I ???? I ???? I ???? I ???? I ???? I
265 275 285 295
GGTTTGTGAG TGAAAGTCAC CTGCCTTAAT ATCTCAAAAC
GGTTTGTGAG TGAAAGTCAC CTGCCTTAAT ATCTCAAAAC
GGTTTGTGAG TGAAAGTCGC CGACCTCAAT ATCTCAAAAC
GGTTTGTGAG TGAAAGTCAC CTGCCTTAAT ATCTCAAAAC
GGTTTGTGAG TGAAAGTCAC CTGCCTTAAT ATCTCAAAAC
GGTTTGTGAG TGAAAGTCAC CTGCCTTAAT ATCTCAAAAC
TCGTGTCCCC TTCGTCTAGA GGCCCAGGAC ACCGCCCTTT
58
CgIIF
CdlF
Sw15F
DolF
sw5F
Cg4F
Sw4F
0000 I 0 ??? I ..? 0 I 0000 I 000. I ..? 01
305 315 325
TCATCTTCGG GTGATGTTTG AGATATTTGC
TCATCTTCGG GTGATGTTTG AGATATTTGC
TCATCTTCGG GTGATGTTTG AGATATTTGC
TCATCTTCGG GTGATGTTTG AGATATTTGC
TCATCTTCGG GTGATGTTTG AGATATTTGC
TCATCTTCGG GTGATGTTTG AGGTATTTGC
CACGGCGGTA ACAGGGGTTC GAATCCCCTA
00 ?? I .. 0. I .000 I ..? 01 0 ??? I 0000 I
335 345 355
TCTTTAAAAA TCTGGATCAA GCTGAAAATT
TCTTTAAAAA TCTGGATCAA GCTGAAAATT
TCTTTAAAAA TCTGGATCAA GCTGAAAATT
TCTTTAAAAA TCTGGATCAA GCTGAAAATT
TCTTTAAAAA TCTGGATCAA GCTGAAAATT
TCTTTAAAAA TCTGGATCAA GCTGAAAATT
GGGGACGCCA CTTGCTGGTT TGTGAGTGAA
CgIIF
CdlF
Sw15F
DolF
sw5F
Cg4F
Sw4F
CgIIF
CdlF
Sw15F
DolF
sw5F
Cg4F
Sw4F
CgIIF
CdlF
Sw15F
DolF
sw5F
Cg4F
Sw4F
CgIIF
CdlF
Sw15F
DolF
sw5F
Cg4F
Sw4F
.000 I 0000 I 0000 I 0000 I .... I .000 I 0 ??? I 0000 I .000 I .000 I 00 ?? I 0000 I
365 375 385 395 405 415
GAAACACTGA ACAACGAAAG TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC ACGATGATGA
GAAACACTGA ACAACGAAAG TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC ACGATGATGA
GAAACACTGA ACAACGAAAG TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC ACGATGATGA
GAAACACTGA ACAACGAGAG TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC ACGATGATGA
GAAACACTGA ACAACGAGAG TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC ACGATGATGA
GAAACACTGA ACAACGAAAG TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC ACGATGGTGA
AGTCACCTGC CTTAATATCT CAAAACTCAT CTTCGGGTGA TGTTTGAGAT ATTTGCTCTT
o ??? I .0 ?? I .... I .... I 0000 I 00 ?? I .000 I 0.00 I 00 ?? I 000. I 0000 I .0 ?? I
425 435 445 455 465 475
ATCGCAAGAA ACATCTTCGG GTTGTGAGGT TAAGCGACTA AGCGTACACG GTGGATGCCT
ATCGCAAGAA ACATCTTCGG GTTGTGAGGT TAAGCGACTA AGCGTACACG GTGGATGCCT
ATCGCAAGAA ACATCTTCGG GTTGTGAGGT TAAGCGACTA AGCGTACACG GTGGATGCCo
ATCGCAAGAA ACATCTTCGG GTTGTGAGGT TAAGCGACTA AGCGTACACG GTGGATGCCo
ATCGAAAGAA ACATCTTCGG GTTGTGAGGT TAAGCGACTA AGCGTACACC GTGGATGCCT
ATCGTAAGAA ACATCTTCGG GTTGTGAGGT TAAGCGACTA AGCGTACACG GTGGATGCCT
TAAAAATCTG GATCAAGCTG AAAATTGAAA CACTGAACAA CGAAAGTTGT TCGTGAGTCT
0000 I 0 ?? 01 .000 I 0000 I 0.0. I ... 01 0000 I ?... I 00 ?? I .00. I .000 I o ?? 01
485 495 505 515 525 535
TCo 0 0 ?? 0 ??? 0000.0000 000 ????? 00 0000 ??? 0.0 000 ?? 0000 ?? 0000 ??? 00
TGo 0 0 0 ??? 0 0000000000 000 ?? 00.00 0000 ??? 0 ?? 000 ?? 0000 ?? 000 ???? 00
TCo
TGo
CTCAAATTTT CGCAACACGA TGATGAATCG TAAGAAACAT CTTCGGGTTG TGAGGTTAAG
o ??? I .0 ?? I ???? I ???? I 0000 I 00 ??
545 555 565
CGACTAAAGC GTACACGGTG GATGCCTTG
59
DGGE analysis is sensitive to single nucleotide alterations in PCR products, and is a
sufficient method to compare PCR products.
Long Term Cultures:
Three individual E. coli isolates, HuOl, CdOl and SwOl were subjected to
multiple culture transfers to determine ifany natural changes in the rmB l6S-23S ISR
DNA sequence occur over time. This is important aspect to examine for future studies
using environmental samples. Each ofthe three isolates DNA was isolated at day 0, and
used for PCR-DGGE analysis. The isolates were then transferred between cultures of
LB-broth and LB-plates eleven times. The DNA was then isolated from the E. coli
isolates and examined for any differences from the day 0 isolates by PCR-DGGE. The
isolates were then transferred between the two medias 11 more times. Once again the
DNA was isolated form the isolates and subjected to PCR-DGGE analysis. Table 5
shows the resulting DGGE gel ofthe long term cultures compared to the day 0 cultures.
Duplicate runs ofseveral isolates were performed to measure the amount of
reproducibility in the method. Each isolate was independently subjected to PCR
amplification ofthe ISR for DGGE analysis. The results show a high degree of
reproducibility. Almost all RD values are within 0.02 ofeach other (table 6).
60
Table 5: Results ofthe long term culture studies.
The three long term cultures, human (HuOI), swine (SwOl) and dairy cow (CdOl) are
listed showing there day 0 RD values and there RD values after 11 culture changes from
LB broth to LB plates.
61
Source Isolate
HuOt
CdOt
SwOt
Da 0
0.68
1
0.99
RD Value
5 Culture Chan es
0.76
0.87
1.04
62
Table 6: Results ofduplicate DGGE results from independently prepared samples
compared.
The calculated RD values are given for each oftwo trials for the isolates that were tested
for reproducibility.
63
RD Value
Source
Isolate Trial 1 Trial 2
Sw01 1.01 0.99
Sw03 0.96 0.94
Sw06 1.01 1.01
Sw07 0.89 0.87
Sw08 0.96 0.97
Sw09 0.88 0.88
Sw11 0.88 0.88
Sw12 0.94 0.95
Sw13 1.00 1.01
Cd01 1.00 1.00
De10 0.9 0.9
Ho4 0.87 0.87
Ho5 0.88 0.88
Ho12 0.88 0.87
64
Chapter 4: Discussion
The results ofthe PCR-DGGE analysis ofthe rrnB 16S-23S ISR ofE. coli
isolates from eight different animal sources reveals that this method can differentiate
between isolates from different sources. Each source E. coli isolate tested produced a
single band on a denaturing gel corresponding to the DNA sequence ofthe ISR. The
fifteen isolates from each source produced composite genetic profiles oftheir source
based on the unique distribution and abundance ofeach isolates RD value. There was no
correlation between the majority ofthe sources DGGE genetic profiles. Each source also
produced high diversity and evenness ofRD values for their isolates tested. This high
diversity and evenness, and the fact that there was no correlation between most ofthe
sources, implies that the profiles produced were actually unique to the individual sources,
and that there is a high probability that any additional isolates tested would remain unique
to a particular source. Additionally, when all nonhuman sources were grouped together,
there was no correlation with the DGGE profiles produced compared to those ofhumans.
The diversity ofthe nonhuman sources actually increased compared to any individual
nonhuman source. These results support using this method to differentiate between
human and nonhuman E. coli isolates.
The DGGE results also revealed a high amount ofgenetic diversity for the E. coli
isolates from each ofthe 8 different sources. Many MST methods have reported finding
high diversity in E. coli isolates tested (Buchan et. aI., 2001, Seumick et aI., 2003,
McLellan et aI., 2003, Jarvis et aI.,2000) using various techniques. Many factors
influence the genetic variations found in E. coli populations. Host specificity, in
65
particular, is an important factor influencing E. coli populations (McLellan et aI., 2003).
Buchan et ai. found high E. coli genetic diversity when performing DGGE analysis ofthe
16S-23S ISR using primers that amplified all 7 ISR. Ofthe 132 E. coli isolates
examined, 84 unique DGGE banding patterns were identified. Similarly, when
performing the same ISR-DGGE analysis with primers for all 7 rrn operons, Seurnick et.
ai. identified 87 ISR fingerprints out ofa total of267 isolates examined. The results of
our study show that each source had a different H' value, and they were all on the high
end. Out ofthe 119 E. coli isolates evaluated by the PCR-DGGE ofthe single rrnB ISR,
only 40 DGGE bands were observed. Additionally, when our samples were grouped as
human or nohuman sources, the nonhuman sources had higher diversity than the human
sources. The nonhuman sources also had a higher diversity than any individual
nonhuman source. The higher diversity observed in nonhuman sources has been
associated with the wide host range ofall possible nonhuman sources (Parveen et aI.,
1999), and strain adaptation to various wild hosts from different regions has been shown
to be an important factor in E. coli population structure (Souza et ai. 1999).
Host specificity refers to the presence ofdominant clonal groups ofE. coli found
in a particular host. The Pearson correlation coefficient analysis performed on the eight
E. coli source ISR DGGE bands showed almost no correlation between the different
sources. Many factors are attributed to the occurrence ofhost specificity for E. coli.
When examining the ISR in particular, there are stretches ofnonfunctional DNA that are
present, and these regions should exhibit a considerable degree ofvariation due to genetic
drift (Garcia-Martinez et. alI996a). Therefore, it is likely that unique ISR sequences
may dominate specific animal sources based on genetic drift. The diet ofvarious host
66
sources also attributes to the host specificity. Recent studies have shown that diet has
affected the E. coli populations ofvarious hosts (Jarvis et aI., 2000, Hartel et aI., 2003).
Furthennore, the types ofsugars that are used by E. coli have been shown to be
associated with the taxonomic group ofthe host from which the isolates were obtained
(Souza et aI., 1999). There are also other various factors between sources that can lead to
difference in E. coli isolates. The temperature and pH ofthe host's microenvironment are
two important aspects that affect the E. coli strains present (Carson et aI., 2001). The
differences in diets and microenvironments may therefore contribute to host specificity
and the genetic drift observed in our E. coli samples. Host specificity has only been
reported to account for some ofthe observed diversity ofE. coli populations, while the
the extent to which the host influences the gentic composition ofE. coli is still unknown
(McLellan et aI., 2003). Nonetheless, E. coli isolates have still been reported to be
correctly classified to host sources, and candidate specific genetic fingerprints have been
identified using a variety ofMST methods.
The main goal ofthis investigation was to evaluate the ability ofa MST method
to differentiate between E. coli isolates fonn various sources for future work oftesting
environmental samples based on the results ofthe DGGE analysis. The high amount of
diversity among the sources tested is a positive result, along with the high evenness ofthe
DGGE band distribution for the sources tested. The low amount ofcorrelation between
sources also adds to the potential ofthis method for evaluating environmental samples.
However, the ultimate success ofthis method is still limited by the inadequate amount of
infonnation available on the fate ofE. coli in the environment. Specifically, the stability
ofthe genetic marker, the ISR, needs to be further investigated. It has been estimated
67
that a typical E. coli bacterium spends halfofits life outside ofthe host in the external
environment, and that fate ofthe clones in the external environment is poorly understood
(Gordon,2001). Furthermore, there appears to be a substantial amount ofchange in the
community composition during the transition from the host to the environment. The
conditions in the external environment that differ from the host, and may affect the clonal
composition ofE. coli, include differences in temperature, pH, nutrients, oxygen
concentrations and solar irradiation (Buchan et aI., 2001). However, isolates from the
same source have still been reported to give the same profiles (Buchan et aI., 2001).
Therefore, the high observed diversity, along with the low correlations between RD
values in this investigation, supports using this MST method to analyze environmental
samples by comparing collected isolates from the environment with isolates from the
presumed source.
Currently, most MST methods require the comparison ofenvironmental samples
to a developed host library to determine the source ofthe contamination. Using a host
library has potential drawbacks, which include the possibility that the isolates in the
library may overestimate the frequency ofa particular strain in the overall population
(McLellan et aI., 2003). This disadvantage, along with the potential success ofour MST
method, suggests that comparing different environmental isolates will be adequate to
determine the source. However, isolates collected from a stream have been shown to
have higher diversity than individual source isolates (Buchan et aI., 2001). This may be
expected due to the increase in potential sources ofE. coli in the environment, and the
possibility ofgenetic changes occurring from the transition from host to the environment.
Possible geographic and temporal genetic variations that may occur in isolates from the
68
same host source from different locations makes comparisons to a host library difficult.
This problem could be overcome by simply comparing environmental sources from their
presumed source in the same geographical area. In fact, it has been stated that given the
high amount ofE. coli strain diversity, isolate characterization may be most feasible
within a limited geographical area such as a watershed (McLellan et aI., 2003). The
ability ofour MST method to show no correlation between E. coli isolates from various
sources with high diversities suggests that the PCR-DGGE method may be able to
distinguish environmental samples, even ifthere diversities are higher than source
isolates. Additionally, the repeatability ofour method, as shown by the ability to produce
the same DGGE bands when individual isolates were independently analyzed by DGGE,
demonstrates the reproducibility ofthis technique.
69
Summary:
The results ofthis investigation show promise of 16S-23S ISR DGGE as an
effective method for microbial source tracking. DGGE analysis ofeach E. coli source
isolate resulted in the production ofa genetic profile based on the DNA sequence oftheir
ISR. This finding supports using the ISR as a genetic target for source tracking, and that
the sequence variability found in the ISR is sufficient for differentiating isolates from the
same bacterial species. Additionally, these results show that E. coli demonstrates enough
genetic variation to be considered a good candidate for an indicator organism. For each
ofthe sources, the diversity (H') and evenness (J') ofbanding profiles was relatively
high, as would be expected from the reported high genetic diversity ofE. coli. This
makes it difficult to determine were in a specific host group a new isolate would be
placed, iffurther tests were performed. However, the high diversity is important when
considering that almost all ofthe profiles were shown to have no correlation. It can
therefore be concluded that each source group produced unique genetic profiles
compared to the other sources, excluding the four sets that were shown to be correlated.
There are many conditions that have to be meet in order for a MST method to be
successful. The existence ofhost specificity ofthe indicator organism is vital to MST
methods. This research shows that there appears to be host specificity for most ofthe
sources. Each animal source had unique DGGE bands in their profile. However, the
results presented in this study are from a moderately small sample set. Therefore, the
results show that further investigation ofsource isolates by the PCR-DGGE method is
merited.
70
The results ofthis investigation also shows that it may be possible to overcome
some ofthe problems associated with other MST methods that utilize source reference
libraries to assign unknown isolates to a particular source. Reference libraries require
vast amounts ofisolates from each source to be tested, due to the possible genetic
variation in isolates from different geographica11ocations, and with different diets. Also,
reference libraries may over represent the frequency ofa particular strain. The apparent
ability ofour PCR-DGGE method to differentiate isolates from 8 different animal sources
shows that the technique can distinguish between two E. coli populations. This will
strengthen any field results that show similarities, or dissimilarities between DGGE
profiles, regardless ofthe source. Therefore, the use ofcomparing samples to a reference
library is not essential, and any results from analyzing a reference library only support the
ability to distinguish possible sources. This supports that using our PCR-DGGE method
should be sufficient for concluding iftwo samples are indeed from the same location.
The stability ofthe indicator organism and the genetic target in the environment
are very important for all MST methods. The conditions an E. coli isolate face from the
transition from host to environment differ greatly, and may affect the abilities ofa MST
method. The 16S-23S ISR examined in this study is part ofan operon found on the E.
coli chromosome that is crucial to cell survival. The amount ofvariation in this region
that may come about by genetic recombination in the environment is still not completely
understood. However, the ISR as a genetic marker may be a more stable genetic marker
than others previously studied. Other genetic targets, such as antibiotic resistance, are
present on p1asmids and often exchanged in the environment (Guan et aI., 2002, Scott et
aI., 2002).
71
Appendix A Isolate Name Chart and Individual RD Values
Canadian Goose (Cg01-15) Swine (Sw01-15) Human (Hu01-Hu15)
Relative Distance Relative Distance Relative Distance
Cg01 0.55 Sw01 0.99 Hu01 0.68
Cg02 0.55 Sw02 0.93 Hu02 0.98
Cg03 0.61 Sw03 0.94 Hu03 0.82
Cg04 0.96 Sw04 0.84 Hu04 0.68
Cg05 0.73 Sw05 0.93 Hu05 0.98
Cg06 0.55 Sw06 1.01 Hu06 1.2
Cg07 0.95 Sw07 0.87 Hu07 0.89
Cg08 0.95 Sw08 0.97 Hu08 1.09
Cg09 0.75 Sw09 0.88 Hu09 0.94
Cg10 1.1 Sw10 0.86 Hu10 0.87
Cg11 1 Sw11 0.88 Hu11 1.01
Cg12 0.77 Sw12 0.95 Hu12 1.05
Cg13 0.75 Sw13 1.01 Hu13 0.82
Cg14 0.75 Sw14 0.88 Hu14 0.7
Cg15 0.95 Sw15 1.11 Hu15 1.07
Cattle Dairy (Cd01-07) Beef (Cb01-
08) Deer (De01-15) Dog (Do01-D015)
Relative Distance Relative Distance Relative Distance
Cd01 1 Oe01 0.91 0001 1.05
Cd02 0.89 Oe02 1.07 0002 0.85
Cd03 0.95 Oe03 0.85 0003 0.85
Cd04 0.82 Oe04 0.9 0004 0.94
Cd05 0.9 Oe05 0.78 0005 0.85
Cd06 0.91 Oe06 1.01 0006 0.9
Cd07 1.07 Oe07 0.89 0007 0.99
Cb01 0.72 Oe08 0.8 0008 0.94
Cb02 0.75 Oe09 1 0009 1.06
Cb03 0.85 Oe10 0.9 0010 0.9
Cb04 0.85 Oe11 0.94 0011 0.94
Cb05 0.86 Oe12 NA 0012 1.15
Cb06 0.87 Oe13 0.91 0013 1.03
Cb07 0.96 Oe14 0.9 0014 0.94
Cb08 1.02 Oe15 0.85 0015 0.69
Horse (Ho01-H015) Chicken (Ch01-15)
Relative Distance Relative Distance
Ho01 0.95 Ch01 0.76
Ho02 0.81 Ch02 0.94
Ho03 1 Ch03 0.85
Ho04 0.87 Ch04 0.91
Ho05 0.88 Ch05 0.88
Ho06 0.85 Ch06 0.88
Ho07 0.92 Ch07 1
Ho08 1.03 Ch08 0.83
Ho09 0.92 Ch09 0.9
H010 1 Ch10 0.95
H011 1 Ch11 0.95
H012 0.87 Ch12 0.95
H013 0.93 Ch13 1.02
H014 0.96 Ch14 0.91
H015 0.99 Ch15 0.91
72
References:
1.) Anton, A. I., A. J Martinez-Murcia, F. Rodriguez-Valera. 1998. Sequence
diversity in the 16S-23S intergenic spacer region (ISR) ofthe rRNA operons in
representatives ofthe Escherichia coli ECOR collection. Journal ofMolecular
Evolution. 47:62-72.
2.) Brosius, J., T. J. Dull, D. D Sleeter and H. F. Noller. 1981. Gene organization and
primary structure ofa ribosomal RNA operon from Escherichia coli. Journal of
Molecular Biology. 148:107-127.
3.) Buchan, A., M. Alber, and R. E. Hodsom. 2001. Strain-specific differentiation of
environmental Escherichia coli isolates via denaturing gradient gel electrophoresis
(DGGE) analysis ofthe 16S-23S intergenic spacer region. FEMS Microbial Ecology.
35 :313-321.
4.) Carson, C. A., B. L. Shear, M. R. Ellersieck and A. Asfaw. 2001. Identification of
fecal Escherichia coli from humans and animals by ribotyping. Applied and
Environmental Microbiology. 67:1503-1507.
5.) Cilia, V., B. Lafay and R. Christen. 1996. Sequence heterogeneities among 16S
ribosomal RNA sequences, and their effects on phylogenetic analyses at the species
level. Molecular Biology Evolution. 13:451-461.
6.) Dombek, P. E., L. K. Johnson, S. T. Zimmerley and M. J. Sadowsky. 2000. Use of
repetitive DNA sequence and the PCR to Differentiate Escherichia coli isolates from
human and animal sources. Applied and Environmental Microbiology. 66:2572-2577.
7.) Farleitner, A. H., N. Kreuzinger, G. G. Kavka, S. Grillenberger, J. Rath and R.
L. Mach. 2000. Simultaneous detection and differentiation ofEscherichia coli
populations from environmental freshwaters by means ofsequence variations in a
fragment ofthe B-D-Glucuronidase gene. Applied and Environmental Microbiology.
66: 1340-1346.
8.) Fox, K. F., M. Nagpal, P. Steinberg and K. Heroux. 1998. Identification of
Brucella by ribosomal-spacer-region PCR and differentiation ofBrucella canis from
other Brucella spp. Pathogenic for humans by carbohydrate profiles. Journal of
Clinical Microbiology. 36:3217-3222.
9.) Fukushima, M., K. Kakinuma and R. Kawaguchi. 2002. Phylogenetic analysis of
Salmonella, Shigella, and Escherichia coli strains on the basis ofthe gyrB gene
sequence. Journal ofClinical Microbiology. 40:2779-2785.
73
10.) Garcia-Martinez, J., A. Matinez-Murcia, A. I. Anton and F. Rodriguez-Valera.
1996a. Comparison ofthe small16S to 23S intergenic spacer region (ISR) ofthe
rRNA operons ofsome Escherichia coli strains ofthe ECOR collection and E. coli
K-12. Journal ofBacteriology. 178:6374-6377.
11.) Garcia-Martinez, J., A. Matinez-Murcia, A., F. Rodriguez-Valera and I. Anton.
1996b. Molecular evidence supporting the existence oftwo major groups in
uropathogenic Escherichia coli. FEMS Immunology and Medical Microbiology.
14:231-244.
12.) Garcia-Martinez, J., S. G. Acinas, A. I. Anton and F. Rodriguez-Valera. 1999.
Use ofthe 16S-23S ribosomal genes spacer region in studies ofprokaryotic
diversity. Journal ofMicrobiological Methods. 36: 55-64.
13.) Gordon, David M. 2001. Geographical structure and host specificity in bacteria and
the implications for tracing the source ofcoliform contamination. Microbiology.
147:1079-1085.
14.) Guan, S., R. Xu, S. Chen, J. Odumeru and C. Gyles. 2002. Development ofa
procedure for discriminating among Escherichia coli isolates form animal and human
sources. Applied and Environmental Microbiology. 68:2690-2698.
15.) Hartel, G. P., J. D. Summer and W. I. Segars. 2003. Deer diet affects ribotypes
diversity ofEscherichia coli for bacterial source tracking. Water Research. 37:3263 
3268.
16.) Jarvis, G.N., M.G. Kizoulis, F. Diez-Gonzalez, and J. B. Russel. 2000. The
genetic diversity ofpredominant Escherichia coli strains isolated from cattle fed
various amounts ofhay and grain. FEMS Microbiology Ecology. 32:225-233.
17.) Jensen, M. A, J. A. Webster, and N. Straus. 1993. Rapid identification ofbacteria
on the basis ofpolymerase chain reaction-amplified ribosomal DNA spacer
polymorphisms. Applied and Environmental Microbiology. 59:945-952.
18.) Kuhnert, P., P. Boerlin and J. Frey. 2000. Target genes for virulence assessment
ofEscherichia coli isolates form water, food and the environment. FEMS
Microbiology Reviews. 24: 107-117.
19.) Leung, K. T., R. Mackereth, Y. C. Tien and E. Topp. 2004. A comparison of
AFLP and ERIC-PCR analyses for discriminating Escherichia coli from cattle, pig
and human sources. FEMS Microbiology Ecology. 47:111-119.
21.) Lia, D. 2000. Gene conversion drives within genic sequences: concerted evolution of
ribosomal RNA genes in Bacteria and Archaea. Journal ofMolecular Evolution.
51:305-317.
74
22.) McLellan S. L., A. D. Daniels and A. K. Salmore. Genetic characterization of
Escherichia coli populations from host sources offecal pollution by using DNA
fingerprinting. Applied and Environmental Microbiology. 69:2587-2594.
23.) Mills A. L., and R. A. Wassel. 1980. Aspects ofdiversity measurement for
microbial communities. Applied and Environmental Microbiology. 40:578-586.
24.) Muyzer, Gerard. 1999. DGGE/TGGE a method for identifying genes from natural
ecosystems. Current Opinion in Microbiology. 2:317-322.
23.) Muyzer, G., E. C. De Wall and A. G. Uitterlinden. 1993. Profiling ofcomplex
microbial populations by denaturing gradient gel electrophoresis analysis of
polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and
Environmental Microbiology. 59:695-700.
25.) Nagpal, M. L., K. F. Fox and A. Fox. 1998. Utility of 16S-23S rRNA spacer region
methodology: how similar are interspace regions within a genome and between
strains for closely related organisms? Journal ofMicrobiological Methods. 33:211 
219.
26.) Parveen, S., K. M. Portier, K. Robinson, L. Edmiston and M. L. Tamplin. 1999.
Discriminant analysis ofribotype profiles ofEscherichia coli for differentiating
human and nonhuman sources offecal pollution. Applied and Environmental
Microbiology. 65:3142-3147.
27.) Scheinert, P., R. Krausse, U. Ullmann, R. Soller and G. Krupp. 1996. Molecular
differentiation ofbacteria by PCR amplification ofthe 16S-23S rRNA spacer.
Journal ofMicrobiological Methods. 26: 103-117.
28.) Scott, T. M., J. B. Rose, T. M. Jenkins, S. R. Farrah and J. Lukasik. 2002.
Microbial source tracking: Current methodology and future directions. Applied and
Environmental Microbiology. 68:5796-5803.
28.) Seurinck, S., W. Verstraete, and S. D. Siciliano. 2003. Use ofthe 16S-23S rRNA
intergenic spacer region PCR and repetitive extragenic palindromic PCR analyses of
Escherichia coli isolates to identify Nonpoint fecal sources. Applied and
Environmental Microbiology. 69:4942-4950.
30.) Sheffield, V. C., D. R. Cox, L. S. Lerman and R. M. Myers. 1989. Attatchment of
a 40-base pair G+C-rich sequence (GC-clamp) to genomic DNA fragments by the
polymerase chain reaction results in improved detection ofsingle-bases changes.
Proceedings ofthe National Academy ofScience. 86:232-236.
31.) Souza, V., M. Rocha, A. Valera, and L.E. Eguiarte. 1999. Genetic structure of
natural populations ofEscherichia coli in wild hosts on different continents. Applied
and Environmental Microbiology. 65:3373-3385.
75