COMPARATIVE ANALYSIS REVEALS A HYPER-VIRULENT ESCHERICHIA COLI O157:H7 STRAIN ISOLATED FROM A SUPER-SHEDDER

By

LIN TENG

A THESIS PRESENTED TO THE GRADUATE SCHOOL

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2016

© 2016 Lin Teng

To my dear parents, Lihang Wang and Yongyu Teng, and my dear advisor, Dr. KwangCheol Casey Jeong, who is leading me into the world of science

4

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. KwangCheol Casey Jeong, for his generous

support, continuous patience and strict training during my entire M.S. program. His

training and direction help me culture my scientific thinking and evoke my interests to

explore the unknown. Thanks to Dr. Soohyoun Ahn for her priceless encouragement

and directions in my thesis. I appreciate Dr. Nicolas Dilorenzo’s kind help for the

arrangement of sample collections at North Florida Research and Education Center.

Thanks to Dr. Dongjin Park for working together with me on this project and giving me

lots of detailed directions.

Special thanks to Raies Mir, Zhengxin Ma, Dr. Alejandro Garrido-Maestu, Dr.

Sarah Markland, Minyoung Kang, Choonghee Lee Amber Ginn and Shinyoung Lee.

Raies has been taking care of me since the first day I joined the lab. With his help and

directions, I was able to quickly adjust to a new life at the University of Florida. Thanks

to Zhengxin for all the detailed directions she gave me for the experiments, course work

and daily life throughout my M.S. career. I appreciate the kind and priceless help and

encouragements from Alejandro in the toughest period during my M.S. program. Sarah

not only helped me with my experiments but also taught me to be optimistic to

overcome the problems I met in the experiments. Minyoung always supported me with

the kind help in the experiments whenever I needed. Choonghee taught me how to

conduct Western blot and maintained the cell lines for my experiments. Amber taught

me phylogenetic analysis and never got tired of my endless questions. I would like to

acknowledge Shinyoung for her efforts to optimize the condition of biofilm assay. In

addition, I appreciate the help from Cheng Ye and Guan Yang. Even though they are

not my lab members, they always cheered me up and motivated me to move forward.

5

I would like to express my deepest and greatest gratitude to my parents, Lihang

Wang and Yongyu Teng for their unconditional support throughout the years. I would

also like to thank my girlfriend, Lanwen Zhou, for her support and encouragement

during the years.

6

TABLE OF CONTENTS page

ACKNOWLEDGMENTS .................................................................................................. 4

LIST OF TABLES ............................................................................................................ 8

LIST OF FIGURES .......................................................................................................... 9

LIST OF ABBREVIATIONS ........................................................................................... 10

ABSTRACT ................................................................................................................... 11

CHAPTER

1 INTRODUCTION .................................................................................................... 13

2 LITERATURE REVIEW .......................................................................................... 15

Shiga Toxin-Producing Escherichia coli O157:H7 .................................................. 15

Virulence Factors and Virulence Related Genes .................................................... 15 Toxins ............................................................................................................... 16

Adhesion .......................................................................................................... 16 Secretion Systems ........................................................................................... 17

Effectors ........................................................................................................... 17

Reservoir and Super-Shedder ................................................................................ 18

Role of Super-Shedder ........................................................................................... 19 Factors Contributing to Supper-Shedding ............................................................... 20

Microbial Factors in Super-Shedding ............................................................... 21

Animal Factors Contributing to Super-Shedding .............................................. 25 Environmental Factors Contributing to Super-Shedding .................................. 26

Conclusion .............................................................................................................. 26

3 COMPARATIVE ANALYSIS REVEALS A HYPER-VIRULENT Escherichia coli O157:H7 STRAIN ISOLATED FROM A SUPER-SHEDDER .................................. 28

Background ............................................................................................................. 28

Materials and Methods............................................................................................ 32 Bacterial Strains ............................................................................................... 32 Genomic DNA Extraction .................................................................................. 32 Stx2 Subtyping ................................................................................................. 32

Phage Induction ............................................................................................... 33 Western Blot Analysis of Stx2 Expression ........................................................ 33 Adherence to HEp-2 Cells ................................................................................ 34 Static Biofilm Formation Assay ......................................................................... 35 Genome Sequencing and Assembly ................................................................ 36 Annotation and Comparative Genomics ........................................................... 36

7

In silico Analysis of LSPA-6 Profile and Phylogenetic Clade ............................ 37 In silico Analysis of IS629 Profile ...................................................................... 38

Function-Based Gene Comparison .................................................................. 38

Phylogenetic Analysis ...................................................................................... 38 Results .................................................................................................................... 39

Shiga Toxin Expression of KCJ1266 ................................................................ 39 Adherence to Epithelial Cell Line ...................................................................... 39 Biofilm Formation ............................................................................................. 40

Lineage Determination Based on LSPA-6 ........................................................ 40 Clade of KCJ1266 ............................................................................................ 40 IS629 Profile Distribution .................................................................................. 41

Overview of KCJ1266 Genome ........................................................................ 41

Comparative Genomic Analyses. ..................................................................... 42 Functional Characterization of the Genome ..................................................... 43 Single Nucleotide Polymorphisms .................................................................... 44

Virulence and Colonization Factors .................................................................. 44

Phage Regions in KCJ1266 ............................................................................. 46 Phylogenetic Analysis ...................................................................................... 47

Discussion .............................................................................................................. 47

Conclusion .............................................................................................................. 51 Figures .................................................................................................................... 56

LIST OF REFERENCES ............................................................................................... 67

BIOGRAPHICAL SKETCH ............................................................................................ 76

8

LIST OF TABLES

Table page 3-1 Genome statistics of KCJ1266 and reference O157 strains .............................. 52

3-2 Function-based gene comparison ..................................................................... 53

3-3 Genome encoded in the 19 phage regions of KCJ1266 .................................... 54

9

LIST OF FIGURES

Figure page 3-1 Shiga-toxin II detection.. ..................................................................................... 56

3-2 Adhesion of E. coli O157:H7 isolates to Hep-2 cells ........................................... 57

3-3 Biofilm assay of KCJ1266. EDL933 was used as a positive control and ΔfimH

mutant of EDL933 was used as a negative control............................................. 58

3-4 IS629 profile of KCJ1266, SS17, EC4115 and EDL933.. ................................... 59

3-5 Genome map of KCJ1266 .................................................................................. 60

3-6 Mauve alignments of KCJ1266 with the three reference genome. ..................... 61

3-7 Comparison of O157 plasmids of KCJ1266 and other 3 reference strains ......... 62

3-8 Subsystem category distribution ......................................................................... 63

3-9 Phage comparison of KCJ1266 with reference strains ....................................... 64

3-10 Virulence related genes in KCJ1266 and reference strains. ............................... 65

3-11 Strains are clustered into three groups which belong to Lineage I (green), Lineage II (red) and Lineage I/II (blue). .............................................................. 66

10

LIST OF ABBREVIATIONS

CFU Colony-forming unit

GI Gastrointestinal

HUS Hemolytic-uremic syndrome

HV Hype-virulence

IDP IS629 distribution profile

LEE Locus for enterocyte effacement

LSPA Lineage specific polymorphism analysis

nsSNP Nonsynonymous single-nucleotide polymorphism

PCR Polymerase chain reaction

PFGE Pulsed-field gel electrophoresis

RAJ Recto-anal junction

SB Synteny block

SNP Single-nucleotide polymorphism

SS Super-shedder

sSNP Synonymous single nucleotide polymorphism

STEC Shiga toxin-producing Escherichia coli

TBST 20 Tris-Buffered Saline and Tween 20

tir translocated intimin receptor

TTSS Type three secretion system

WGS Whole genome sequencing

11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science

COMPARATIVE ANALYSIS REVEALS A HYPER-VIRULENT ESCHERICHIA COLI O157:H7 STRAIN ISOLATED FROM A SUPER-SHEDDER

By

Lin Teng

May 2016

Chair: KwangCheol Casey Jeong Major: Animal Sciences

Shiga toxin-producing Escherichia coli (STEC) O157:H7 is a foodborne pathogen

that threatens public health on a global scale. STEC O157:H7 predominantly colonizes

the terminal recto-anal junction (RAJ) of cattle, which is the major asymptomatic

reservoir of this pathogen. Cattle shedding STEC O157:H7 ≥ 104 CFU/g of feces are

known as super-shedders and are responsible for within-farm and between-farm

transmission of STEC O157:H7. The purpose of this study was to perform genetic

characterization of KCJ1266, a strain isolated from a super-shedder steer from a farm in

North Florida. PacBio sequencing was employed for whole genome sequencing (WGS)

to characterize the genomic features of KCJ1266. A comparative genome analysis of

KCJ1266 with reference genomes including SS17 (strain isolated from a super-

shedder), EC4115 (strain related to spinach outbreak) and EDL933 (strain related to

hamburger outbreak) was conducted. WGS of KCJ1266 revealed that it has a genome

of 5,478,683 bp encoding 5,545 open reading frames and a plasmid, pO157, of 95,910

bp. In silico analysis revealed that KCJ1266 belongs to E. coli lineage I/II and clade 8,

12

which are related to disease-causing isolates. In addition, Mauve alignment showed that

KCJ1266 shares a similar genomic architecture with SS17 and EC4115. Comparative

analyses also revealed that KCJ1266 has the same virulence and similar functional

genes as SS17 and EC4115. Phylogenetic analysis showed that KCJ1266, SS17 and

EC4115 clustered in the same group. Taken together, these results reveal that super-

shedding STEC strain KCJ1266 is a hyper-virulent strain similar to that of SS17 and

EC4115. This is one of the first studies which has utilized PacBio WGS to characterize

a hyper-virulent STEC O157:H7 strain isolated from super-shedding cattle. KCJ1266

can potentially be used as a reference strain for future studies regarding the

phenomenon of super-shedding.

13

CHAPTER 1 INTRODUCTION

Shiga toxin-producing Escherichia coli (STEC) are foodborne pathogens that

threaten public health on a global scale [1]. It causes a variety of symptoms including

bloody diarrhea and life-threatening hemolytic uremic syndrome (HUS) in humans

through the expression of Shiga-like toxin [2-4]. STEC infections cause annual

economic loss of more than $1 billion USD in the US [5]. The human infection caused

by STEC is primarily attributed to the consumption of food products contaminated by

STEC.

Cattle are the primary reservoirs and asymptomatic carriers of STEC. The

terminal recto-anal junction (RAJ), as well as gastrointestinal (GI) tract, are the principle

localization sites of STEC [5]. Among the cattle colonized by STEC, a subset of cattle

shed E. coli O157:H7 at a level of > 104 colony forming unit (CFU) per gram feces and

are defined as “super-shedders” [6, 7]. Previous studies showed that super-shedders

were responsible for most of the within-farm and between-farm transmission events of

STEC [1, 8, 9]. The existence of super-shedders may lead to an increase in the

contamination of animal food products, such as beef and milk, and thus threaten public

health.

Since the phenomenon of super-shedding was first observed, researchers have

been trying to determine the bacterial, host, and environmental factors that contribute to

super-shedding [10]. However, the mechanisms of super-shedding are still unknown. In

order to determine the potential bacterial factors that contribute to super-shedding, we

hypothesized that the super-shedding phenomenon is caused by a group of specific

genomic traits of bacterial strains isolated from super-shedding cattle. The objective of

14

this study was to identify the genes which contribute to the specific phenomenon of

super-shedding using comparative genome analysis.

15

CHAPTER 2 LITERATURE REVIEW

Shiga Toxin-Producing Escherichia coli O157:H7

Shiga toxin-producing Escherichia coli O157:H7 (STEC) are foodborne

pathogens that globally threaten public health [11]. In 1982, E. coli O157:H7 was first

identified as a foodborne pathogen following an outbreak of hemorrhagic colitis in the

United States [1]. In 1993, a multistate outbreak of E. coli O157:H7 led to the

recognition of E. coli O157:H7 as a crucial pathogen in the U.S [12]. Reports of E. coli

O157:H7 outbreaks from 2003 to 2012 increased when compared to the preceding two

decades. E. coli outbreaks are historically associated with not only beef products, but

also a variety of food sources, including unpasteurized apple juice, spinach and salami

[13, 14]. E. coli O157:H7 quickly became one of the most recognized foodborne

pathogens due to the severity of the disease it caused in humans and the significant

media attention seen during outbreaks. STEC causes a variety of symptoms ranging

from bloody diarrhea to the development of life-threatening hemolytic-uremic syndrome

(HUS). Previous research has shown that STEC infections cause more than $1 billion

USD in economic loss per year [5].

Virulence Factors and Virulence Related Genes

Bacterial virulence and pathogenesis factors play a significant role in infectious

disease etiology. With the development of whole genome sequencing techniques, an

increasing number of virulence factors can be identified by whole genome comparison

[15]. A recent review concluded that about 394 coding sequences were identified as

putative or known virulence factors [15]. Without a good animal model for STEC

infection, it is difficult to determine the contribution of putative virulence genes to its

16

pathogenesis [2], therefore most of the putative virulence genes will not be discussed in

this review.

Toxins

Shiga toxins (Stxs) are encoded by Shiga-toxin converting prophages in the

chromosome of STEC. The diversity of Stxs are caused by the high heterogeneity of

Shiga-toxin converting prophages [16]. There are two major types of Shiga toxins, Stx1

and Stx2, which are encoded by two main sets of genes, stx1AB and stx2AB,

respectively. Both of Stx1 and Stx2 consist of an A subunit of 32-kDa and five identical

B subunits of 7.7kDa [17]. The main receptor of Shiga toxin is globotriaosylceramide

(Gb3) which can be found on human cells, including endothelial cells, epithelial cells

and microvascular endothelial cells [18]. Although Stx1 and Stx2 have similar structure,

their toxic dosage to different cell types vary. An in vitro study showed that Stx2 toxin

was reported to be 1000 times more cytotoxic in human renal microvascular endothelial

cells when compared with Stx1 toxin [19]. In addition, the Stx2 subtypes encoded by

stx2a, stx2b, stx2c, stx2d, stx2e and stx2f differ by a few amino acids, resulting in

diverse toxicity of Stx2 in humans [20]. Among these Stx2 subtypes, Stx2a, Stx2c and

Stx2d are often related to severe human diseases such as hemorrhagic colitis (HC) and

HUS [21-24]. A recent study showed that STEC O157:H7 harboring the stx2a gene can

express Shiga toxin at higher levels than those harboring the stx2c gene [25].

Adhesion

The ability of STEC to adhere to human epithelial cells is believed to be a critical

step for infection because adherence is the initial step in STEC infection [15]. Fimbriae

and curli are two major apparatuses which contribute to adherence ability of STEC.

Type I fimbriae are encoded by the fim operon, including fimB, fimE, fimA, fimI, fimC,

17

fimF, fimG and fimH. Curli play a significant role in surface attachment and biofilm

formation. The formation of curli is controlled by the cgs operon [26]. Long polar filament

encoded by the lpfABCC’DE operon are also involved in adherence [27].

Secretion Systems

E. coli O157 harbors a highly conserved plasmid known as pO157. There are 13

ORFs (open reading frames) named etpC to eptO which encode a type two secretion

system in pO157 [2]. It has been found that this secretion system can secrete StcE (zinc

metalloprotease protein), thus it contributes to the virulence of E. coli O157:H7 [15].

Another secretion system in STEC O157:H7 is the type three secretion system (TTSS),

a needle complex apparatus, by which bacterial effectors are injected into host cells[28].

This apparatus, made up of more than 20 proteins, is necessary for STEC to colonize to

cattle. The structure of the TTSS is encoded by genes, including eprK, eprJ, eprH,

epaQ, epaP, epaO, elvJ, elvI, elvA, elvE, elvG and elvF [15]. When STEC start

colonization, Stx prophages are activated by SOS induction followed by cell lysis of a

subgroup of STEC [28]. The Stx produced by Stx prophages bind to its receptor, Gb3,

on epithelia cells, resulting in redistribution of nucleolin and other factors to the cell

surface [28]. The changed surface structure promotes bacterial attachment by binding

to bacterial surface protein intimin (Eae) [28].

Effectors

Virulence effectors refer to the proteins that are injected to host cells using TTSS

[15]. STEC O157:H7 are able to colonize the intestinal mucosa and cause attaching and

effacing (A/E) lesions. The regions encoding proteins related to A/E lesion were

designated the locus of enterocyte effacement (LEE) [29]. The espH, espZ and espG

are effector genes encoding in LEE. In addition to effectors encoded by LEE, there are

18

numerous non-LEE-encoding effectors. These effectors are encoded by genes including

espY4, nleB1, espW, nleG8-2, espM2, espR4, nleA, nleH1-2, espM1, nleG2-2, nleG6-1,

nleG5-1, espK, espX2, espY1, nleG8-1, nleD, and nleH1-1 [15].

Reservoir and Super-Shedder

Cattle are the primary asymptomatic reservoir of STEC [30]. However, the

shedding amount of E. coli O157:H7 varies greatly from animal to animal [8, 31, 32].

The number of E. coli O157:H7 shed in feces is also affected by the climate, season

and breed of cattle. Previous reports showed that the shedding of E. coli O157:H7 was

observed as a transient event and the typical duration of E. coli O157:H7 shedding lasts

about 1 month [33]. The term “super-shedder”, was first used by Matthews et al. in

2006, which is defined as a subset of cattle that can shed ≥ 104 colony-forming unit

(CFU) /g feces of E. coli O157:H7 [6]. However, the term “super-shedding” are often

used to describe the heterogeneous role that the cattle play in the transmission of E.

coli O157:H7. Based on a review written by Chase-Topping et al. (2008) the super-

shedder is defined as the individual that can shed an extremely high number of

infectious organisms, by which the super-shedder has an increased probability of

infecting other individuals [1].

However, a part of the definition remains unclear, specifically what defines the

duration of the super-shedding state. In a previous study, the duration of super-

shedding lasted up to a month [33]. Since the duration of super-shedding may affect the

identification of a super-shedder, a proper sampling method for isolation of super

shedding bacteria from animals is extremely important. To get an accurate account of

shedding information of E. coli O157:H7, Munns et al. (2015) modified the sampling

method and had each animal sampled twice a day during a 6-day period [10]. The

19

results of that study revealed that super-shedding is a short-term phenomenon and that

shedding amount may not be consistent even between two samplings within the same

day [10]. In this case, it is not easy to identify the low-shedder and super-shedder, since

their status may change within one day.

Role of Super-Shedder

Super-shedders are of significant importance because they are considered to be

responsible for the within-farm and between-farm transmission of E. coli O157:H7, as

well as transmission of E. coli O157:H7 into the environment [1]. It has been observed

that E. coli O157:H7 are heterogeneously shed by cattle within the same farm [6]. In a

previous study, more than 96% of total E. coli O157:H7 were shed by about 9% of the

total 761 cattle tested [34]. In a mathematical modelling study, Matthews et al. (2006) [6]

showed that 20% of the most infectious individuals of cattle were responsible for about

80% of the infections that occurred within the same farm. Stephens et al. (2009) [9]

reported that super-shedders were responsible for 47% of within-farm shedding of E.

coli O157:H7, while this ratio increased to more than 90% in another study [1]. A study

involving Scottish farms showed evidence of between-farm transmission of E. coli

O157:H7, where 15 of 105 pulsed-field gel electrophoresis (PFGE) patterns were

detected in more than one farm [35]. Jeon et al. (2013) traced the presence of E. coli

O157:H7 within a herd of cattle after they were transported to a commercial feedlot,

finding that the PFGE pattern of E. coli O157:H7 isolated in the feedlot was identical to

that of commercial farm isolates [8]. This result indicated that E. coli O157:H7 could be

transferred between farms.

Since super shedder E. coli O157:H7 strains may express high level of Shiga

toxin, they, with their high shedding load, are a potential threat to public health. A study

20

performed by Chase-Topping et al. (2007) showed that the most common phage type in

Scotland was PT 21/28 (41%) [36]. This was found to be correlated with another study

in which 58% of E. coli O157:H7 human infections were caused by PT 21/28 E. coli

O157:H7 [37]. There is more evidence indicating the importance for the control and

surveillance of super-shedding animals, and the types of bacteria that are being shed in

high concentrations. Cote et al. (2015) found that an E. coli O157:H7 strain isolated

from a super-shedding animal in the U.S. Midwest possessed a hyper-virulent

characteristic having two copies of the stx2a gene, allowing the bacteria to produce high

level of Shiga toxin 2 without induction [26].

Factors Contributing to Supper-Shedding

Cattle live in various environments, interact with each other and are exposed to

different environmental pressures. There may be three major factors, including animal

factors, pathogen factors and environmental factors that cause the super-shedding

phenomenon [1, 10]. The animal factors include the status of the animal’s immune

system, susceptibility to intestinal colonization, animal to animal transmission, stress

response and animal breed [8, 10]. The environmental factors include season, animal

housing hygiene, and diet. The bacterial factors include the ability of E. coli O157:H7 to

survive in the gastrointestinal tract, competiveness of super-shedder E. coli O157:H7

vs. commensal E. coli, potential of E. coli O157:H7 to proliferate in rectal contents and

the ability to form biofilms. To date, the most studied among these factors are bacterial

factors, while experiments regarding host and environmental factors are not well

studied.

21

Microbial Factors in Super-Shedding

Previous studies reported that the E. coli O157:H7 isolated from super-shedders

in Scotland shared a primary phage type, PT21/28, suggesting that PT21/28 phenotype

may be a marker of super-shedding strains [1, 36]. However, another study reported

that no PT21/28 were identified, while the most common phage type affecting cattle was

PT4 [38]. Another study [39] examined the phage type in cattle in Canada, finding that

PT14a and PT8 were the most popular among all five of the identified phage types.

Although certain primary phage types are common in certain farms or regions, there is

no one specific phage that is commonly found on a global scale. These results showed

that phage type may not be a good marker for super-shedding strains.

PFGE is believed to be the “gold standard” for epidemiological studies involving

outbreaks [40]. In some of the previous studies, PFGE profiles showed that the strains

isolated from super-shedders and non-super-shedders have similar PFGE patterns.

Jeon et al. (2013) reported that the isolates of super-shedders, including medium-

shedders and low-shedders in North Florida, have the same PFGE patterns, indicating

that these strains are derived from the same source and have the same subtype [8].

Another study performed by Munns et al. (2015) showed a similar result that isolates of

E. coli O157:H7 from super-shedders and low-shedders contained three similar PFGE

patterns [10]. Among all these isolates, 97.6% (123/126) of them had the same PFGE

pattern. These results may reveal that strains with specific characteristics (PFGE

patterns) may have better adaptation abilities than other strains to survive and

proliferate within the host. In contrast, some other studies also showed that the super-

shedders and low-shedders shared different PFGE patterns. According to a study by

Stanford et al. (2012) PFGE patterns of E. coli O157:H7 isolates from super-shedders

22

and low-shedders of two commercial farms were varied [41]. Another study reported by

Dodd et al. (2013) confirmed the theory that super-shedders do not necessarily share

the same PFGE subtype. Arthur et al. (2013) identified 102 stains isolated from cattle of

super-shedders, finding 52 distinct PFGE subtypes [38]. However, there were no

specific PFGE subtypes found to be responsible for super-shedding. All these results

suggest that there is not a specific PFGE pattern of E. coli O157:H7 that is related to

super-shedding.

Octamer-based genome scanning methods could discriminate E. coli O157:H7

according to the size of gene markers. By using octamer-based genome scanning and

microarray-based comparative genomic hybridization, E. coli O157:H7 can be grouped

into three distinct lineages [42, 43]. These three lineages, lineage I, lineage II and

lineage I/II, are usually isolated from different sources. In most of the cases, lineage I

strains are composed of human clinical isolates. Lineage II isolates usually originate

from bovine sources. The lineage I/II isolates are related to strains that can cause

human disease, including foodborne outbreaks [44]. Within these three lineages,

lineage I and I/II are more likely to cause human disease. Studies showed that a

potential hyper-virulent strain isolated from a super-shedding animal belonged to linage

I/II according to a whole genome analysis [26]. An isolate from a super-shedding animal

in the study conducted by Jeon et al.(2013) [8] was analyzed by whole genome

sequencing, and the result showed it also belonged to lineage I/II. According to Munns

et al. (2015) 99.2% and 0.8% of all isolates from super-shedders and low-shedders,

respectively, were from lineage I and lineage I/II [45]. However, a broader study showed

that 30, 35, and 37 out of 102 strains isolated from super-shedders were identified as

23

lineage I, II and I/II isolates, respectively [38]. These results indicate that super-

shedders isolates do not belong to specific lineages.

The translocated intimin receptor (tir) single-nucleotide polymorphism (SNP) is

another typing method used to differentiate E. coli O157:H7 [46]. The translocated

intimin receptor, an effector of the type three secretion system, was previously used as

a marker of virulence because of its significant role in adhering to intestinal epithelial

cells. In this typing method, a nucleotide change (either A or T) at position 255 in the tir

gene was used to predict the virulence of an E. coli O157:H7 strain, with hyper-virulent

isolates having an increased chance to possess the T allele at this position [46]. Among

all the super-shedder strains used in a study by Arthur et al.(2013) 71% harbored the T

allele, while 29% of them harbor the A allele [38]. Munns et al.(2015) reported that in 10

strains isolated from cattle of super-shedder and low-shedder isolates, 9 strains (90%)

displayed the T allele at the 255 position of tir gene [45].

The Shiga toxin-encoding bacteriophage insertion (SBI) assay is typically used to

discriminate the E. coli O157:H7 shed from super- and low-shedders. Among a variety

of Shiga toxin-encoding genes, stx1, stx2a and stx2c are the most frequently studied in

STEC isolated from cattle. Since Shiga toxin-encoding genes may insert within or

adjacent to several distinct chromosomal loci including yehV, wrbA, argW and sbcB. Six

individual polymerase chain reactions (PCR) were previously employed to distinguish

those insertion sites [47]. It was found in a study by Arthur et al. that super-shedding

isolates have a higher chance of harboring either stx2a or stx2c than carrying both

genes [38]. Within 12 groups of super-shedding isolates, 55.9% of isolates were

identified as SBI genotype 1, 2 and 3 [38].

24

Sequencing techniques, including whole genome sequencing, are becoming

increasingly popular to analyze the characteristics of bacteria. Historically, whole

genome sequences of the E. coli O157:H7 strain isolated from the Sakai outbreak in

Japan, was compared with that of the benign laboratory strain K-12. The result of this

whole genome comparison has shown that the Sakai strain has 20 specific tRNAs and

131 potential virulence factors [48]. The whole genome sequence of another E. coli

O157:H7 outbreak strain, EDL933, was also compared with that of K-12 strain, with

1387 specific genes identified in EDL933 [49]. Since whole genome sequencing of

isolates could provide researchers with a variety of pertinent genetic information,

researchers have begun to use this powerful tool in order to determine the differences

between super-shedding and low-shedding strains at the genomic level. The whole

genome sequences of 4 low-shedding isolates and 6 super-shedding isolates were

sequenced in a study by Munns et al.(2015) [45] The results showed that, based on the

analysis of single nucleotide polymorphism, there were no significant genetic

differences between super- and low-shedding strains. Another study including the whole

genome sequencing analysis of a super-shedding strain was conducted by Cote et al.

(2015) [26]. Since phage regions of E. coli O157:H7 contain a variety of genes inserted

by phages, it is possible that the genes which cause the super-shedding phenomenon

are located in phage regions. Cote et al. (2015) compared the phage regions between

EDL933, Sakai and EC4115 [26], but no specific genes were found in these phage

regions. Researchers concluded that no specific gene in phage region was responsible

for super-shedding characteristics. It is possible that super-shedding strains harbor

specific adherence factors which give them ability to be shed in large quantities from the

25

intestinal tract. To examine this hypothesis, Cote et al. (2015) identified all the

adherence factors in a super-shedding strain, SS17 [26]. Through comparison of

adherence factors with those of reference strains, researchers found that no specific

adherence factors could be associated with super-shedding. SNPs are also a potential

factor that could be related to super-shedding bacteria. Recently, a new clustering

method, IS629 profile, was proposed. It discriminates E. coli O157:H7 according to the

presence of insertion sequences in chromosome. This new method may give

researchers more clues for discovering the nature of super-shedding.

Animal Factors Contributing to Super-Shedding

A previous study showed that the prevalence of E. coli O157:H7 strains isolated

from super-, medium- and low-shedders could be affected by genetic and physiological

factors of animals [8]. Jeon et al. (2013) reported that Brahman cattle are more resistant

to E. coli O157:H7 [8]. Some cattle probably have resistance to E. coli O157:H7,

because they cannot be infected by inoculation with different doses of E. coli O157:H7,

as was shown in previous studies [8, 50].

The microbiome of the GI tract of cattle were examined, since microbial

community could potentially reflect host health. High variation in bacterial community

composition was observed among feces of individual animals in recent study [51, 52].

By sequencing 16S rRNA genes of bacteria in feces from 11 super-shedders and 11

non-shedders, Xu et al. (2014) found variation of bacterial composition between two

sets of animals, with super-shedders harboring a richer microbiome than non-shedders

[53]. Some studies showed that changes in the microbiome may lead to a boost in

immune response and exclusion of enteric pathogens [54]. As a result, super-shedding

strains could be able to accumulate to higher numbers.

26

Environmental Factors Contributing to Super-Shedding

As previously mentioned, the environmental factors include season, diet, and

animal housing hygiene. Shedding of STEC, especially E. coli O157:H7, varies

significantly in different seasons, with a peak in summer [55-57]. Diet composition is

another factor that significantly affects super-shedding. Callaway et al. (2011) found a

decrease of E. coli O157:H7 population in ruminants after the animals were fed orange

peel product [58]. In another study, Fox et al. (2007) observed that fecal shedding of E.

coli O157:H7 increased after feeding distillers grains to cattle and claimed that the

change of volatile fatty acids (VFA) concentration was responsible for this increase [59].

When Munns et al. (2015) made a change of feed for cattle, the super-shedders stop

shedding E. coli O157:H7 at a high level [45].

Conclusion

Among all the factors that might cause super-shedding, bacterial factors are

considered to be one of the major ones causing the super-shedding phenomenon in

beef cattle. Although bacterial factors were among the most studied, the exact bacterial

factors causing this phenomenon have not been clearly defined. We have also been

unable to identify the common characteristics between super-shedding isolates. There

are some possible reasons that could explain this. First, the super-shedding

phenomenon is a transient phenomenon, which means that the status of super-

shedders and low-shedders could change in a short period of time. As a result, it may

not be easy to distinguish between super-shedders and low-shedders. In this case, the

strains isolated from “super-shedder” and “low-shedder” may not be good

representatives regarding this phenotype. Second, whole genome sequencing is

thought to be a powerful technology to study bacterial factors attributed to super-

27

shedding. However, by using Illumina sequencing or other sequencing methods, many

gaps will be generated, which means some genetic information present in gaps are

missing. Without accurate sequencing data, the gene traits that cause super-shedding

will be more difficult. A more accurate sequencing technology, such as PacBio

sequencing, should be employed in order to study these bacterial factors. Third,

researchers tend to seek out a certain gene or SNP which is responsible for this

phenomenon. However, it is likely that this phenomenon is modulated by a cluster of

genes, which make this work more difficult.

28

CHAPTER 3 COMPARATIVE ANALYSIS REVEALS A HYPER-VIRULENT Escherichia coli O157:H7

STRAIN ISOLATED FROM A SUPER-SHEDDER

Background

Shiga toxin-producing Escherichia coli (STEC) are foodborne pathogens which

threaten public health on a global scale [10]. It causes a variety of symptoms including

bloody diarrhea and life-threatening hemolytic-uremic syndrome (HUS) [2-4]. STEC

infections cause economic losses of more than $1 billion USD per year, including direct

and indirect costs [5]. The primary reason that people get infected is through the

consumption of food products contaminated by STEC. Since no effective treatment

method is available for patients infected by STEC, contamination of cattle by STEC

must be controlled at the pre-harvest level. Cattle are primary reservoirs of STEC so the

prevention of colonization of STEC in cattle has become an important approach for

preventing STEC infections in humans [30]

The terminal recto-anal junction (RAJ) as well as gastrointestinal (GI) tract, are

the principle localization site by STEC in cattle [60]. Among the cattle colonized by

STEC, a subset of cattle have been found to shed > 104 CFU of E. coli O157 /g feces

and are defined as “super-shedders” [6]. The existence of super-shedders raises the

chances of contamination of animal food products such as animal hides, beef and milk,

and thus threatens public health. Jeon et al.(2013) [8] demonstrated that cattle were

able to transmit super-shedding strains between farms. Previous reports showed that

super-shedding animals within feedlot pens may be responsible for more than 90% of

shedding of E. coli O157:H7 [1, 9]. In another study, Matthew et al. (2006) [6] suggested

that 80% of bacterial transmission was caused by 20% of infectious individuals and

reduction in shedding by the cattle will reduce human infections.

29

Three potential factors causing the phenomenon of super-shedding are bacterial,

animal host, and environmental factors. To understand characteristics of super-

shedding E. coli O157 strains, phenotypic and genotypic difference among super-

shedder strains have been noted in previous studies, including phage type [1, 45],

lineage-specific polymorphisms, tir polymorphisms [61], and the clade of E. coli O157

[62] and the difference in the presence of stx2a and stx2c [10]. Previous studies showed

that most of the STEC O157 isolated from super-shedders in Scotland had a phage

type of 21/28 (PT21/28) [1, 36]. The super-shedder isolates with phage types of PT4

and PT14a were predominantly identified in two studies conducted in the United State

and Canada [10, 38]. Based on a lineage-specific polymorphism assay E. coli O157

isolates can be classified into three lineages: lineage I, lineage II and lineage I/II [42,

63]. A study showed that super-shedder isolates are distributed among all of the three

lineages [38]. Another molecular marker, tir polymorphism, used to differentiate E. coli

O157:H7 strains was based on a nonsynonymous base change (A or T) at position 255

in the tir gene [62]. Bono et al. (2007) [46] found that the tir 255 T allele was more

frequently found in human isolates, suggesting the isolates with a T allele at the 255

position in tir may be more virulent to humans. Other studies showed that super-

shedder isolates harbored both T and A alleles at the 255 position of tir and do not have

an obvious preference [10, 38]. Clade analysis can discriminate E. coli O157:H7

isolates into 9 different clades based on SNPs in 96 loci [62]. Among all the 9 clades,

isolates in clade 8 are believed to be hyper-virulent because they have higher chances

to cause severe patient symptoms. Although the phenotypic and genetic methods have

been conducted, the specific bacterial factors causing super-shedding is still unknown.

30

Recently, a whole genome analysis revealed that a super-shedder strain SS17

isolated from the Recto-anal junction (RAJ) of cattle in U.S. Midwestern states has a

distinctive genome compared with foodborne outbreak strains, Sakai and EDL933, and

spinach associated outbreak strains, TW14359 and EC4115 [26].This study revealed

that SS17 has a unique plasmid pSS17, a large number of non-synonymous SNPs

(nsSNP) and other polymorphisms compared with the reference strains. In addition, the

strong adherence aggregative phenotype associated with SS17 was shown to be

unassociated with the locus for enterocyte effacement (LEE) [26].

Previously super-shedder strains were isolated from cattle shedding O157 at a

level of more than 105 CFU /g feces which is ten times the minimum load defining a

super-shedder [1, 8]. These super-shedder strains shared the same PFGE profile and

genotype of eae+, stx1-, stx2+ and hlyA+ [8]. Further studies showed that all these

super-shedder strains can express Shiga-toxin. With this information, we hypothesize

that the phenomenon of super-shedding is caused by a group of specific genomic traits

of bacterial strains isolated from super-shedding cattle. The representative strain,

KCJ1266, isolated from super-shedding cattle was sequenced using PacBio whole

genome sequencing method, which could generate longer reads and circular bacterial

whole genome sequence than Illumina sequencing, followed by whole genome

comparative analysis.

This study shows that KCJ1266 can adhere to human HEp-2 cell as good as

human pathogen EDL933, has similar biofilm formation ability of EDL933, belongs to E.

coli lineage I/ II and clade 8, and shares a similar genomic architecture and virulence

factors with SS17 and EC4115. Taken together, these results reveal that super-

31

shedding STEC strain KCJ1266 is a hyper-virulent strain similar to that of SS17 and

EC4115. This is one of the first studies which utilizes PacBio WGS to characterize a

hyper-virulent STEC O157 strain isolated from super-shedding cattle. KCJ1266 can

potentially be used as a reference strain for future studies regarding the phenomenon of

super-shedding.

32

Materials and Methods

Bacterial Strains

KCJ1266 is a strain isolated from a super-shedding cattle located in North Florida

in 2012. EDL933 (ATCC48935) is a laboratory strain and was linked to hamburger-

borne outbreak in 1982 in the U.S. The laboratory strains, Δeae (KCJ696) and Δtir

(KCJ698) and ΔfimH (KCJ867), are deletion mutant strains of EDL933.

Genomic DNA Extraction

Genomic DNA of KCJ1266 was isolated by the use of QIAGEN DNA mini Kit

(cat. no. 51304). Briefly, 1 mL overnight culture of KCJ1266 was centrifuged for 10 min

at 5000 x g (7500 rmp). Following the manufacturer’s protocol, the bacterial pellet was

resuspended in buffer ATL and treated with 20 μL proteinase K at 56°C for 30 mins. The

released DNAs were eluted with 100 μL of ddH2O.

Stx2 Subtyping

Polymerase chain reaction (PCR) was performed to determine the stx2 subtypes

(i.e. stx2a and stx2c gene) carried by KCJ1266. Two E. coli O157:H7 strains, EDL933

carrying stx2a and stx1 and FRIK2455 carrying stx2c, were used as controls. Primers

used for detecting stx2a (stx2a-F: 5’-CTTTTCGACCCAACAAAGTTATGT-3’ and stx2a-

R: 5’-CACAGTCCCCAGTATCGCT-3’) and stx2c (stx2c-F: 5’-

TACTGTGCCTGTTACTGGGC-3’ and stx2c-R: 5’-ACAGTGCCCAGTATCGCC-3’) were

designed according to previous study [44]. Each 25 μL PCR reaction mix contains 2.5

mL of 10X buffer, 0.5 mL of dNTP, 0.2 mL of Taq polymerase, and primers. The PCR

reaction condition was set as 94°C for 5 min; 94°C for 30 sec, 54°C for 30 sec, 72°C for

60 sec for 30 cycles and a final extension time of 10 min at 72°C. Amplified PCR

33

product was analyzed in 1% agarose gel electrophoresis and visualized by ethidium

bromide staining.

Phage Induction

Cell lysis cause by phage induction was measured by optical density (OD).

Overnight culture of bacterial strains was inoculated into LB media, Mitomycin C was

added to the cell culture at the final concentration of 0.5 μL/ mL when the OD600 of the

culture reached between 0.6 and 0.7. EDL933 was used as a positive control, while

DH5α were used as a negative control for phage induction. For phage particle

preparation, Mitomycin C was added to 25 mL cell culture when OD600 reached 0.7 to

make a final concentration of 1 μL/ mL. After 18 h incubation, unlysed cells and debris

were removed by centrifugation at 4,000 rpm for 20 min at 4°C, and the resultant

supernatant was filtered with 0.22 μm-pore-size membrane. Precipitation of phage

particles was performed using 0.25 volume of 20% polyethylene glycol 8000 and 10%

NaCl. The polyethylene glycol/NaCl solution containing phage particles was incubated

at 4°C overnight, followed by centrifugation at 12,000 rpm for 1 h. The pellet was

resuspended and stored in 1 ml SM buffer (0.58 g NaCl, 0.2 g MgSO4·7H2O, 0.01 g

Gelatin per 100 ml 1M Tris-Cl pH7.5). SDS-PAGE was used to check the phage protein

profile after an aliquot of the phage suspension was mixed with SDS-loading buffer and

boiled for 5 min.

Western Blot Analysis of Stx2 Expression

The method used for identifying Stx2 expression was described in the previous

paper [64]. Briefly, the exponential phase culture (OD600= 0.7) of a O157 strain was

treated with Mitomycin C to the final concentration of 1 μg/ mL in order to induce the

lytic cycle of phages in the O157 strain and concomitant stx gene expression. Next, the

34

O157 cell culture was incubated for 18 hrs for complete cell lysis. The cell debris was

removed by centrifugation (4,000 rpm for 20 min at 4°C) and the cell-free supernatant

was collected by using 0.22 μm-pore-size membrane filter. The total proteins in the

supernatant were precipitated by adding 1 × volume of 100% trichloroacetic acid (Fisher

Scientific BP555-1) to 4 × volume of the cell-free culture. The mixture was incubated for

30 min on ice and centrifuged at 14,000 rpm, 4°C for 30 min. The protein pellet was

finally washed using cold acetone. The total proteins were separated by 12% sodium

dodecyl sulfate polyacrylamide gel and transferred to a polyvinylidene difluoride

membrane (0.45 μm pores, Immobilon-P, Millipore) for Western blot. The membrane

was blocked using 5% skim milk in TBST (10 mM Tris-HCL [pH 7.4], 150 mM NaCl, and

0.05% Tween-20) at room temperature for 1 h. Later, the membrane was incubated

overnight at 4°C with Monoclonal Verotoxin II-a subunit antibody (Meridian, Life Science

Inc.), and washed using TBST at room temperature for 1 h. The membrane was then

incubated with HRP conjugated secondary antibody (GE Healthcare) diluted 1:10000 in

TBST, washed with TBST, and incubated with a chemiluminescent substrate (ECL Plus,

GE Healthcare). Finally, the membrane was exposed to Kodak BioMax film (Carestream

Kodak X-Omat LS film, F1274 Sigma).

Adherence to HEp-2 Cells

The ability of KCJ1266 to adhere to human epithelial type-2 (HEp-2) cells was

evaluated. EDL933 was used as a positive control and eae and tir mutant of EDL933

was used as a negative control. HEp-2 cells were maintained in Dulbecco’s Modified

Eagle Medium (DMEM) (Coring, 10-017-CV) composed of 10% (vol/vol) heat-

inactivated fetal bovine serum at 37°C and 5% CO2. Approximately 105 of HEp-2 cells

were seeded into a 24-well polystyrene plate as measured using a hemocytometer and

35

allowed to grow until 90% confluent (approximately 24 h). Bacterial cultures were grown

overnight at 37°C in LB broth and washed with sterile phosphate buffered saline (PBS)

three times. The final cell pellet was resuspended in PBS to a final concentration of

O.D.600=1 (5x108 CFU). An aliquot (20µl) of the resuspended culture was inoculated

into 480µL of DMEM. Then 500 µl DMEM containing 107 bacteria was added to each

well (MOI 100). After a three-hour incubation at 37°C with 5% CO2, the bacterial

suspension was aspirated and 500 µl of new DMEM was added to each well, and this

process was repeated once more. The infected cell monolayer was washed using sterile

PBS three times to remove any unattached bacteria. To detach the HEp-2 cells from the

plate, 1mL of 0.1% Triton X 100 in PBS buffer was added to each well and allowed to

incubate for 5 min. The media was removed from each well and serially diluted in LB

broth and plated on TSA plates to enumerate the number of attached bacterial cells.

The adherence assay of each strain was conducted in triplicate. Statistical significance

between the level of attachments of isolates to the negative and positive controls were

accessed using a student’s t-test (α = 0.05).

Static Biofilm Formation Assay

The method used in the biofilm formation assay was the same as previously

described one [26]. Briefly, EDL933 strain and ΔfimH mutant of EDL933 were used as a

positive and a negative control, respectively. Strains were inoculated in LB media at

37°C with overnight shaking. The overnight culture was diluted (1:100) into M9 medium

supplemented with glucose (0.8% wt/vol) and minerals (1.16 mM MgSO4, 2 µM FeCl3, 8

µMCaCl2 and 16 µM MnCl2) and incubated at 37 °C with shaking for 48 h. Then the

overnight cultures were diluted (1:100) into the same medium and 200 µL of diluted cell

36

cultures were pipetted into a 96-well plate (Corning, NY, USA) in 6 replicates. The cell

cultures were incubated at 30°C for 48 h, removed the supernatant and washed 3 times

using PBS. The plate was dried at 37°C for 20 min until there were no water in the well,

followed by staining the biofilm using 0.1% crystal violet (wt/vol) for 2 min. Then the

crystal violet was removed and wells were washed using PBS for 3 times to remove the

remaining crystal violet. The plate was dried at 37°C for 20 min to remove all the liquid

in the wells, followed by adding 200 µl of 80% (vol/vol) ethanol and 20% (vol/vol)

acetone to release the stain. The released stain was measured by microplate reader

using the absorbance at λ 595 nm.

Genome Sequencing and Assembly

To acquire the whole genome sequence of KCJ1266, whole genome PacBio

sequencing was performed in this study. Eighty-five thousand reads with a mean

subread length of 7660 bases were obtained with 115.7 fold coverage. To obtain

contigs of genome and plasmid, de novo assembly was conducted by Falcon

(https://github.com/PacificBiosciences/FALCON). RS HGAP Assembly.3 protocol was

used with control filtering set as ‘KeepControlReads’.1.xml. Option details were set as

shown below: Minimum subread length was 500. Minimum polymerase read quality was

set as 0.80. Minimum polymerase read length was 100. At this analysis, the minimum

seed read length was adjusted to 2000. Target coverage, overlapper error rate,

overlapper min length and overlapper K-Mer were set as 25, 0.06, 40 and 14. Contigs

were connected to form a circular DNA by overlapping each end of contigs.

Annotation and Comparative Genomics

The genome of KCJ1266 was automatically annotated by RAST (Rapid

Annotation using Subsystem Technology). A FASTA file of KCJ1266 was unloaded to

37

RAST (http://rast.nmpdr.org/). Mauve 2.3.1 (default setting) was employed to align the

genome of KCJ1266 and reference strains. The SNPs between KCJ1266 and reference

strains was identified using galaxy in HiPergator of the University of Florida.

To identify the position of phage related genes, the PHAge Search Tool (PHAST)

(http://phast.wishartlab.com/ ) was used. After assembly was done and the contig of

genome of KCJ1266 was available, FASTA file of the genome was submitted to the

PHAST web server. The genome of was circularly presented by CGview [8]. Genes

encoding known and putative virulence factors in the genome of KCJ1266 were found

using VFDB [65-68] in Patric [69-71]. Multilocus sequence typing (MLST) of KCJ1266

and reference strains was conducted using Center for Genomic of Epidemiology (CGE).

Reference genomes used for comparative analysis were E. coli O157 strain

EDL933 [NC_008957] and EC4115 [NC_011353], and super-shedding strain SS17

[NZ_CP008805].

In silico Analysis of LSPA-6 Profile and Phylogenetic Clade

Six pairs of primers described in Yang, et al. were used to perform BLAST

against whole genome sequence of KCJ1266 using SnapGen View. LSPA-6 profile of

KCJ1266 was decided based on the length of a region targeted by each pair of the

primers. An Isolate that has a LSPA-6 profile of 111111 was defined as lineage I. A

strain possessing 222222 or 222223 was defined as lineage II. Other genotypes were

defined as lineage I/II [44].

Two methods were used to identify the clade of KCJ1266. An In silico clade 8

rhsA assay, the presence of a SNP (3468C) in the rhsA gene, was used to determine

whether the isolate belongs to the Clade 8 [72]. Besides, another In silico analysis

based on 32 SNPs was also conducted [62].

38

In silico Analysis of IS629 Profile

A total of thirty-two pairs of primers were used to detect IS629-distribution profile

[44]. The primers were aligned against KCJ1266 using BLAST in NCBI. EC4115,

EDL933 and SS17 were used as control. Positive was assigned if the in silico PCR

amplicon has the same size as expected, while negative if the in silico PCR amplicon

has either no IS629 or different size of amplicon [44].

Function-Based Gene Comparison

Whole genome sequence of KCJ1266, SS17, EC4115 and EDL933 were

submitted to RAST (http://rast.nmpdr.org/). Then subsystem information of each strain

was available for function-based analysis. The RAST supply the function of function-

based comparison of two strains and output the unique genes that only exist in one of

the strain.

Phylogenetic Analysis

The whole genome sequences of 28 E. coli O157 reference strains for building

phylogenetic tree were downloaded from Patric

(https://www.patricbrc.org/portal/portal/patric/Home). Mauve was used to align all the 29

sequences, including sequences of KCJ1266 and the reference strains, using

HiPerGator, a super computer located in University of Florida. The output file was

converted to a fasta alignment using an in house perl script, followed by extracting the

variable sites within contig position of each strain using MEGA6. The MEGA6 output

fasta file containing variable sites was submitted to IQtree webserver to construct a

maximum likelihood phylogeny performing 1000 bootstrap iterations for branch support

with the Kimura-2 nucleotide substitution model. Figtree (v.1.4.2) was used to edit the

phylogenetic tree.

39

Results

Shiga Toxin Expression of KCJ1266

Super-shedder isolate of O157 strain, KCJ1266, was collected from cattle in

North Florida Research and Education Center (NFREC) [8]. Fecal samples from rectal

anal junction were collected using sterile cotton swab as previously described [8]. Since

all the strains isolated from super-shedder, medium-shedder and low-shedder have stx2

gene and same PFGE pattern, KCJ1266 was chosen as a representative strain for

further analysis [8]. PCR, Mitomycin C treatment, SDS-PAGE and Western blot were

conducted to subtype stx2 gene and identify expression of Shiga toxin (Figure 3-1). The

result shows that KCJ1266 contains both stx2a and stx2c genes and express Stx2 at a

similar level of EDL933.

Adherence to Epithelial Cell Line

The adherence ability of STEC to epithelial cells plays a significant role in

bacterial colonization and persistence in cattle. Since SS17 has been shown to have

high levels of adherence ability to Hep-2 cells [26], we hypothesized that the high

adherence ability may be a conserved genetic property in supper-shedder strains. To

test the hypothesis, we compared the adherent ability of KCJ1266 to Hep-2 cells with

that of control strains including EDL933, Δeae, Δtir and DH5α strains. The data showed

that KCJ1266 has similar adherence ability to Hep-2 cells compared with EDL933

(Figure 3-2), while the KCJ1266 shows significant difference (P < 0.05) in adherence to

Hep-2 cells compared with Δeae, Δtir and DH5α strains. This result reveals that

KCJ1266 have a high adherence ability to the epithelial cells.

40

Biofilm Formation

A better bacterial biofilm formation ability is related to a better survival ability in

the environment. In the biofilm assay, the biofilm formation of EDL933 shows significant

higher than its fimH gene mutant (Figure 3-3.). At the same time, the KCJ1266 showed

significant higher ability of biofilm formation on polystyrene surface. This result suggests

that KCJ1266 may have a strong ability to survive in the environment, contaminate farm

products and reinoculate other cattle.

Lineage Determination Based on LSPA-6

Previous studies show that O157 could be classified into three groups, lineage I,

lineage II and lineage I/II, which of each is characterized by distinct genome structure,

host origin, and virulence potential [73, 74]. Lineage I and I/II strains are usually isolated

from clinical and bovine source, while Lineage II isolates usually originate from bovine

source [44, 73]. In particular, Lineage I/II include Spinach outbreak strain that displayed

high virulence and Stx expression [75]. Previously, the super-shedder strain SS17 was

shown to belong to lineage I/II [26]. The lineage of KCJ1266 determined by LSPA-6 [43]

showed a profile of (211111) which is the same as that of SS17 (Figure 3-4.). This result

reveals that the two independent super shedder strains, SS17 and KCJ1266 belong to

lineage I/II, implying their potential epidemiological significance.

Clade of KCJ1266

Manning et al. identified the presence of a hyper-virulence clade (clade 8) among

E. coli O157 strains, based on an SNP analysis of 32 loci [62]. Clade 8 strains were

characterized by high expression of Stx and high cytotoxicity [62]. In silico analysis

shows that KCJ1266 belongs to clade 8 (Figure 3-4). Previously, SS17 was also shown

to belongs to clade 8, indicating that KCJ1266 and SS17 are closely related. The

41

findings that KCJ1266 and SS17 belong to the same hyper-virulence clade and share

super-shedding property, together with their distinct geological origins imply that this

type of O157 strain characterized by super-shedder (SS) and hyper-virulence (HV) may

be more prevalent in the natural habitat than was previously considered.

IS629 Profile Distribution

To further investigate the genetic relatedness between KCJ1266, SS17, and the

prototypic strain EC4115, we performed IS629 distribution profile (IDP) analysis, a

method that can be used for not only lineage typing but also genetic differentiation

within a given lineage [44].

IDP of KCJ1266 was determined by In silico analysis, and compared with IDP of

SS17 EC4115 (lineage I/II control) and EDL933 (lineage I control). As expected, IDP of

KCJ1266, SS17 and EC4115 displayed typical IDP of lineage I/II and were distinctive

from lineage I type IDP represented by EDL933 (Figure 3-4). Interestingly, IDPs among

the 3 lineage I/II strains were not identical but varied, reflecting their genetic divergence

at the sub-strain level. Five IDP loci are polymorphic between KCJ1266 and SS17, all of

which locate in prophage regions (Figure 2), indicating prophage polymorphisms are a

primary attribute for sub-strain diversification in E. coli O157.

Overview of KCJ1266 Genome

In order to further detail the genetic traits of KCJ1266 its whole genome

sequence was determined using PacBio sequencing technology. The KCJ1266 has a

chromosome of 5478683 bp which encodes for 5545 open reading frames, 106 tRNA

and 22 rRNA (Fig.3-5). The KCJ1266 genome contains a plasmid, pO157, of 95910 bp.

This plasmid contains 117 genes, including genes translate the type V secretion serine

protease (espP), hemolysin (hlyABCD), cytotoxin (toxB), catalase peroxidase (katP) and

42

the type III secreted proteins (etp operon). The overall genomic features of KCJ1266

are summarized in comparison to reference strains; SS17, EC4115, and EDL933 in

Table 1. The size of KCJ1266 genome is smaller than other two lineage I/II strains,

SS17 and EC4115. It contains a 95.91 Kbp plasmid which is homologous to the pO157

plasmid of SS17 and pO157 plasmid of EC4115, but lacks the additional plasmid

commonly possessed by SS17 and EC4115. These data indicate that KCJ1266

contains reduced genomic features compared to its close relatives.

Comparative Genomic Analyses.

LSPA-6 typing, clade analysis and IDP revealed that all three lineage I/II strains

investigated in this study, KCJ1266, SS17, and EC4115, commonly belong to hyper-

virulence clade but have been diverged genetically. Further, genetic makeup of

KCJ1266 and SS17 seem to be divergent each other despite they shared super-

shedding phenotype. To understand the genetic diversity of the lineage I/II strains in the

context of HV and SS properties, we conducted a series of comparative genomic

analysis of KCJ1266, SS17, and EC4115.

The homology of KCJ1266 and reference strains was showed by the use of

progressive Mauve, displaying 7 different synteny blocks (Figure 3-6). Aligned against

genome of SS17, EC4115 and EDL933, the genome of KCJ1266 has a higher

homology to SS17 and EC4115 than to EDL933. The non-homology region between

KCJ1266 and reference strains were identified to be primarily phage regions.

In details, the lengths of the 7 synteny blocks (SB) varied from ~45.4 kb to

~2,276 kb (Fig.3-6). SB1 is a ~1,249 kb sequence containing phage region 1 to 4. SB2,

a sequence of ~429 kb, SB3 with a length of ~ 45.4 kb contains part of phage region 8

(Table 3-3). SB4 of ~338.5 kb is inverted in EDL933. This block contains sections of

43

phage region 8 and 9. SB5 is a ~1,077.8 kb sequence that contains phage regions 10 to

14 and a section of phage region 9. SB6 with a ~62 kb length contains part of phage

region 15. In the whole genome of EDL933, this block was moved, inverted and inserted

between homologs of SB1 and SB2 of KCJ1266. SB7 contains phage regions of 16 to

19 and part of phage region 15 where a non-homology area was found. According to

the analysis, most of the non-homology areas between KCJ1266 and SS17 exist in

phage regions, suggesting that KCJ1266 and SS17 share highly conserved core

genome region. However, when the genome of KCJ1266 was compared to EC4115 and

EDL933, non-homology regions were frequently found in the core genome regions.

The plasmid of KCJ1266 was also compared with plasmids of reference strains.

There are 99% query coverage and 99% identity of pO157 plasmids between KCJ1266

and SS17 (Figure 3-7).

Functional Characterization of the Genome

KCJ1266 have 5545 coding sequences (CDS). Fifty-five percent (3082) of the

total CDS, including 2914 non-hypothetical and 168 hypothetical CDS, were categorized

into 27 main subsystem categories, including a total of 599 subsystems categories (Fig.

5). The other 44% (2463) of the total CDS were not included in these 599 subsystem

categories (Figure. 3-8).

Pair-wise comparisons of the functional genomes between KCJ1266 and the 3

reference strains were conducted based on the subsystem categories in order to find

unique gene entities in each genome (Table 3-2). In comparison between KCJ1266 and

SS17, KCJ1266 had three unique genes encoding nitrogen regulation protein NR (I),

DNA adenine methyltransferase (phage-associated) and KefF (glutathione-regulated

potassium-efflux system ancillary protein), while SS17 have one unique gene encoding

44

type III secretion protein SsaH. When KCJ1266 and EC4115 were compared with each

other, KCJ1266 had three unique genes encoding Nitrogen regulation protein NR (I),

DNA adenine methyltransferase and LSU ribosomal protein L32p, while EC4115 has

one unique gene encoding SsaH protein. Seventeen unique genes present in genome

of KCJ1266, when KCJ1266 were compared with EDL933. Beside the five genes shown

in Table 2, 12 phage encoding genes encoding phage tails and proteins were among

the KCJ1266 genes absent in the genome of EDL933.

All together, a majority of genes assigned to the subsystem categories in the

genome of KCJ1266 are also present in SS17 and EC4115, except a few variable

genes listed in Table 3-2. Genetic or physiological impacts of these minor differences, if

any, are not known.

Single Nucleotide Polymorphisms

By comparing coding sequences of KCJ1266 and the reference strains, SNPs of

KCJ1266 were identified. KCJ1266 has 109 and 196 SNPs compared with SS17 and

EDL933, respectively.

Of the 109 SNPs between SS17 and KCJ1266, 22 are non-synonymous single

nucleotide polymorphisms (nsSNP). ORFs containing the nsSNPs encode putative

functions including tetratricopeptide repeat protein, Alpha-ketoglutarate permease, and

PTS system glucitol/sorbitol-specific transporter subunit IIA.

Virulence and Colonization Factors

Since KCJ1266 belongs to the hyper-virulence clade 8 and displays supper-

shedder property we analyzed genes known to be involved in virulence and colonization

of this bacterium. A total of 252 genes related to virulence and colonization were

identified in KCJ1266. Of these, 232 genes are encoded by chromosome and 20 are

45

encoded by pO157 plasmid. According to their putative protein function, these genes

are divided into five groups, adhesion, effectors and toxins, secretion systems and other

virulence-associated genes.

In chromosome of KCJ1266, there are 85 genes in adhesion group, including fim

operon, encoding fimbrial proteins and csg operon encoding curli related proteins. Other

genes, such as fliC, efa-1 and ompA, which are not in operons are also found.

The effectors contain LEE-encoded effectors and non-LEE-encoded effectors.

The LEE-encoded effectors are EspA, EspB, EspF, EspG, EspH, EspZ, Tir, and

mitochondrial associated type III secreted effector protein encoded by map. Non-LEE-

encoded protein contains effectors encoded by nleA, nleC and nleD. The stx2 that

encodes A and B subunits of Shiga-Like-toxin 2 are found in chromosome of KCJ1266.

KCJ1266 contains both stx2a and stx2c genes, which is the same as reference strains,

SS17 and EC4115.

The secretion system group is consisted of genes that encode type three

secretion system (TTSS). Among these genes, TTSS components are encoded by

sepL, escD and escQ. The needle structure of TTSS is encoded by escC.

The other virulence-associated genes contain iron uptake genes, such as chuA,

chuS, chuT, chuU, chuW, chuX, chuY and iroB. These genes may play critical roles in

iron acquisition that will affect the outcome of the infection significantly in bacteria-host

interactions

Virulence factors of reference strains were also analyzed. The result shows that

KCJ1266, EC4115 and SS17 have almost the same virulence factors (Figure 3-9). All of

46

KCJ1266, SS17, EC4115 share similar composition of virulence genes, suggesting that

KCJ1266 may have the same virulence as EDL933, EC4115 and SS17.

Phage Regions in KCJ1266

A typical E. coli O157:H7 genome contains multiple lambdoid prophages which

are known to be a major contributor for the genomic diversity among strains.

Consistently, the IDP profile and MAUVE analyses in this study revealed that prophage-

related polymorphisms are a main contributor for genetic dynamics detected in

KCJ1266 and its relative strains. In addition, stx genes and a multitude of putative

virulence genes are carried by prophages. The integration sites for stx2a and stx2c in

the phage regions are shown to be argW and sbcB, respectively, in most of lineage I/II

and are conserved in KCJ1266. A total number of 19 phage regions were found across

the genome of KCJ1266 and major genes encoded in each phage region are

summarized in Table 3-3 and Figure 3-10. The completeness of phage regions are

describe as intact, incomplete and questionable based on the scoring of each phage

region, which is decided by the number of phage related coding sequences and the

homology of suspected regions to the sequences of phage regions in the database [76].

Most of these genes in phage regions of KCJ1266 were also present in phage regions

of SS17.

Previous study showed that Stx2a-encoding phage could be classified into four

major and two minor subtypes [25]. In silico analysis showed that KCJ1266 belongs to

Stx2a phage subtype of ϕStx2a_γ that can produce high level of Shiga toxin, suggesting

the hyper-virulence property of KCJ1266.

47

Phylogenetic Analysis

To understand the evolutionary position of KCJ1266, the phylogenetic analysis

was performed using whole genome sequences of 28 E. coli O157 strains, including

strains of human source, bovine source and outbreaks (i.e., Spinach outbreak and Taco

bell outbreak). Figure 3-11 shows that these strains were clustered into three major

groups which are congruent to the lineage I, lineage II, and lineage I/II. As expected,

KCJ1266, SS17, EC4115, other spinach outbreak strains, and Taco bell outbreak

strains formed a discrete lineage I/II group while EDL933 and Sakai strains belonged to

the lineage I group, and selected stains of bovine source form a lineage II group.

Lineage I/II group contained two distinct clusters; cluster 1 (10 spinach outbreak strains)

and cluster 2 (2 Taco bell outbreak strains). KCJ1266, SS17, another super shedder

SS52, EC4115 and two other spinach outbreak strains (EC4024 and EC4192) were

shown to be outgroups without forming a distinct cluster (Figure 3-11). KCJ1266 is

closest to EC4192. Eppinger et al. previously classified Spinach outbreak strains based

on SNP genotypes [73]. In this classification, all three EC4115, EC4024, and EC4192

strains were genotyped as Group B while stains in the inner Spinach outbreak cluster

were genotyped as either Group A or Bovine genotype II. All lineage I/II strains included

in this study have clade 8 genotype. All together, these data support that KCJ1266 are

closely related to the phylogenetically distinctive outgroups of lineage I/II, yet

possessing high virulence potentials.

Discussion

To discover the genetic traits that contribute to super-shedding, whole genome

sequencing and a variety of gene markers have been used in previous studies. The

Illumina sequencing method has been identified as one of the most popular methods

48

used for whole genome sequencing, however, gaps are generated within whole genome

sequence. At the same time, a large amount of Illumina sequencing data is not able to

be closed. In the current study PacBio sequencing was employed in order to generate a

gap free whole genome sequence to better characterize KCJ1266.

Whole genome sequencing data of KCJ1266 was compared with other reference

strains at a genomic level (Table.3-1), proving KCJ1266 a super-shedder-hyper-

virulence strain by analyses of phage regions, virulence factors, LSPA-6, IS629 profile,

function-based gene, whole genome, SNPs and phylogeny.

In this study we identified several critical features of E. coli O157:H7 KCJ1266

isolated from a super-shedding cow. KCJ1266, SS17 and EDL933 strains belong to

clade 8 and have the same LSPA-6, revealing they share the same phenomenon of

hyper-virulence. Second, IS629 typing shows the additional genetic variability among

these three closely related strains. This relationship and diversity are further detailed by

genomic characterization of KCJ1266 and comparative analysis with SS17 and

EC4115. In addition, virulence factors of KCJ1266 are highly similar with other

reference strains, suggesting that KCJ1266 is a hyper-virulent strain.

It is commonly believed that the phenomenon of super-shedding is caused by

three major factors which are bacterial, host and environmental factors [1, 8]. Among all

these factors, bacterial factors are studied the most based on phenotypic and genetic

characteristics. Manning et al. (2015) showed that clade 8 E. coli O157:H7 strains were

more likely to cause hemolytic uremic syndrome in humans after testing more than 500

E. coli O157:H7 strains [62]. Another study showed that E. coli O157:H7 strains of clade

8 isolated from cattle may partially be responsible for the high amount of HUS cases in

49

Argentina [77]. In this study, KCJ1266 was identified as a clade 8 strain belonging to

lineage I/II, suggesting that KCJ1266 may have hyper-virulence characteristics. This

result is consistent with the comparison of virulence related genes among KCJ1266 and

reference strains. Except one phage related gene, all the virulence related genes in

hyper-virulence reference strain (EDL933) were found to be present in KCJ1266. Ogura

et al. tested STEC O157:H7 isolates harboring different stx2 genes, finding that strains

harboring stx2a or both stx2a and stx2c could express higher levels of Stx2 than those

strains harboring only stx2c [25]. This finding is also consistent with the hyper-virulence

properties of KCJ1266. LSPA profile of KCJ1266 (211111) is the same as SS17 and

EC4115, which is consistent with phylogenetic analysis, suggesting that they are

phylogenetically related to each other. Although KCJ1266, SS17 and EC4115 are

phylogenetically related to each other, diversity in their IS629 profiles were observed.

The difference among inserted sequence based profiles may be representative of the

diversity of these three strains.

Its super-shedding and virulence gene characteristics are major factors that

identify KCJ1266 as a hyper-virulent strain. Since KCJ1266 was isolated from cattle

shedding STEC O157:H7 at a level of > 104 CFU/g feces, there is a greater chance for

this isolate to contaminate the environment and farm-related products. In addition, the

enhanced ability of KCJ1266 to produce biofilms indicates that KCJ1266 has enhanced

environmental fitness. The ability to produce biofilms means this strain may survive in

animal food products after food processing. In addition, the enhanced biofilm formation

ability of KCJ1266 suggests that it may have high affinity in cattle, which means

KCJ1266 may have enhanced within-farm and between-farm transmission.

50

The whole genome sequence of a low-shedder E. coli O157:H7 would have been

an additional reference to help identify the genetic traits that contribute to super-

shedding. However, there is no such whole genome sequence available. The reference

strains, SS17, EC4115 and EDL933, were chosen because they have similar genome

architecture and good quality. This may explain why no such specific genes responsible

for super-shedding were identified. In addition, the whole genome analysis at the

genomic level may not reflect the differences between super-shedder strains and low-

shedder strains. The existence of a gene in two different isolates does not necessarily

mean the same level of transcription and expression of this gene. Even in the same

isolate, the transcription and expression level of a gene may be affected or regulated by

other stimulus. It should be noted that a comparison of transcriptome and proteomics of

super-shedder isolates and low-shedder isolates may help understand the essence of

the phenomenon of super-shedding.

Although this research focuses on the bacterial factors related to super-shedding,

animal host factors are also related to super-shedding. A previous study showed that

the prevalence of E. coli O157:H7 strains isolated from super-, medium- and low-

shedders could be affected by genetic and physiological factors of animals [8]. Jeon et

al. (2013) reported that Brahman cattle are more resistant to E. coli O157:H7 [8]. Some

cattle probably be resistant to E. coli O157:H7, because they cannot be infected by

inoculation with different doses of E. coli O157:H7 [50, 78]. In addition, the innate and

adaptive immune response of the host are believed to be the critical host factors [10].

Although cattle have no receptor for Shiga toxin [79], its innate immune response can

be induced by colonization of E. coli in the GI tract [10]. In addition, dietary animal

51

husbandry practices play important roles in super-shedding. Diet composition, feed

processing and bactericidal properties of ruminal and volatile fatty acids could impact E.

coli shedding and surviving.

Conclusion

This study characterized a hyper-virulent STEC O157:H7 strain, KCJ1266, isolated from

a super-shedder. Although the genes related to super-shedding phenotype are not

identified in this study, the identification of the accurate, gap-free, and circular whole

genome sequence of KCJ1266 is a cornerstone for the optimization of outbreak source

tracking. In addition, KCJ1266 will serve as an ideal reference for the future study of the

super-shedding phenomenon.

There are several possible reasons explaining the result that none of the genes related

to super-shedding were identified in this experiment. The experimental setting was not

perfect for identifying super-shedding phenotype. Instead of comparing stains used in

this experiment, it will be better to compare the whole genomes of isolates from super-

shedders and low-shedders. In addition, the expression of proteins is not only related to

presence of genes, but also regulated by transcription factors. In this case, the

transcriptome and proteome of strains isolated from super- and low-shedders should be

compared. Furthermore, animal factors and environmental factors are also critical.

There is a probability that animal factors are modulating super-shedding phenotype.

52

Table 3-1. Genome statistics of KCJ1266 and reference O157 strains

KCJ1266 SS17 EC4115 EDL933

Length of Sequence (Mb) 5.48 5.52 5.57 5.53

G+C ratio (%) 50.5 50.5 50.5 50.4

Coding DNA Sequence (CDSs)

5274 5430 5608 5587

No. rRNA 22 22 22 22

No. tRNA 106 107 109 100

No. Plasmid 1 2 2 1

53

Table 3-2. Function-based gene comparison

Protein KCJ1266 SS17 EC4115 EDL933 EDL933 Function of protein

NR (I) + - - - - Nitrogen regulation

Enzyme + - - + + DNA adenine methyltransferase

KefF + - + + + Glutathione-regulated potassium-efflux system ancillary protein

SsaH - + + + + Type III secretion protein

L32p + + - - - LSU ribosomal protein

RTX + + + - - T1SS secreted agglutinin

YopD + + + - - Type III secretion host injection and negative regulator protein

tRNA dihydrouridine synthase A

+ + + - - RNA metabolism

54

Table 3-3. Genome encoded in the 19 phage regions of KCJ1266

Phage Regions

Size (Kbp)

Completeness Synteny block

Encoded genes Possible Phage name

SS17 EC4115 EDL933

1 26.5 Intact SB1 Phage DNA invertase; Phage integrase; and phage assembly protein

Entero SfI NC 027339

- + +

2 38.5 Intact SB1 Lysozyme and Attachment invasion locus protein precursor

Entero cdtI NC 009514

+ - -

3 38 Incomplete SB1 Phage assembly proteins and 3 tRNAs

Shigel POCJ13 NC 025434

- - +

4 29.4 Intact SB1 Attachment invasion locus protein precursor and phage lysis

Entero BP 4795 NC 004813

+ + +

5 50.5 Intact SB2 Integrase and phage assembly proteins

Entero BP 4795 NC 004813

+ + +

6 46.5 Intact SB2 Integrase, phage lysin, attachment invasion locus protein precursor

Entero lambda NC 001416

+ + +

7 8 Questionable SB2 Mobile element proteins Stx2 converting 1717 NC 011357

+ + +

8 57 Intact SB2 and SB3

Attachment invasion locus protein precursor, 2 tRNAs, phage lysis, acfC, and integrase

Entero HK629 NC 019711

- - -

9 94.8 Intact SB4 and SB5

Attachment invasion locus protein precursor, phage lysis 5 tRNAs and integrase

Entero YYZ 2008 NC 011356

+ + +

10 24.8 Incomplete SB5 Integrase and 3 tRNAs Entero P88 NC 026014

- + +

55

Table 3-3. Continued

Phage Regions

Size (Kbp)

Completeness Synteny block

Encoded genes Possible Phage name

SS17 EC4115 EDL933

11 54 Intact SB5 Integrase, phage lysin, inv and 5 tRNAs

Entero BP 4795 NC 004813

+ + +

12 61.2 Intact SB5

Attachment invasion locus protein precursor, integrase, yeeV, yeeX, stx2A, stx2Bc and 3 tRNAs

Stx2 converting 1717 NC 011357

+ + +

13 24 Intact SB5 Phage assembly proteins and Phage lysin

Entero P2 NC 001895

+ - -

14 73.5 Intact SB5 Integrase and attachment invasion locus protein precursor

Entero BP 4795 NC 004813

+ + +

15 77.2 Questionable

SB5, SB6 and SB7

Integrase, attachment invasion locus protein precursor, Phage lysin, stx2A, stx2B and 4 tRNAs

Stx2 converting I NC 003525

+ + +

16 23.9 Questionable SB7 Phage lysin and integrase Stx2 converting 1717 NC 011357

+ + +

17 13.1 Questionable SB7 Mobile element proteins Stx2 converting 1717 NC 011357

+ + +

18 20.5 Questionable SB7 Integrase, yeeU, yeeV, espA, espB, espD, espF, sseE, yscL, escD, eae, and tir chaperone

Entero Sf6 NC 005344

+ - +

19 21.7 Questionable SB7 Integrase and phage eae Entero mEp460 NC 019716

+ + +

56

Figures

Figure 3-1. Shiga-toxin II detection. A. Phage induction was measured. MMC was add to cell culture of EDL933,

KCJ1266 and DH5α at the 3rd, 3rd and 4th hour, respectively (Red arrows). B. PCR of stx2a and stx2c C. SDS-PAGE analysis of phage proteins. D. Western blot.

57

Figure 3-2. Adhesion of E. coli O157:H7 isolates to Hep-2 cells, MOI: 100. The Assay was performed in triplicate.

58

Figure 3-3. Biofilm assay of KCJ1266. EDL933 was used as a positive control and ΔfimH mutant of EDL933 was used as

a negative control

59

Figure 3-4. IS629 profile of KCJ1266, SS17, EC4115 and EDL933. A shaded gray box indicates the presence of the in

silico amplicon with correct size and locus, while an unshaded white box indicates a negative in silico result. The three reference strains (EDL933, EC4115 and SS17) are compared.

Str

ain

stx

1

stx

2a

(wrb

A)

stx

2a

(ye

hV

)

stx

2a

(arg

W)

stx

2c

Cla

de

LS

PA

-6

pro

file

Lin

ea

ge

LI-

US

-8

LI-

DS

-17

LI-

US

-9

LI-

DS

-11

LI-

DS

-18

LI-

DS

-1

LI-

DS

-7

LI-

US

-16

LI-

DS

-10

LI-

DS

-19

LI-

DS

-5

LI-

DS

-15

LI-

DS

-6

LI-

US

-13

LI-

DS

-14

LI/

II-D

S-1

5

LI/

II-D

S-4

LI/

II-D

S-7

LI/

II-D

S-1

0

LI/

II-D

S-1

2

LI/

II-D

S-1

3

LI/

II-D

S-1

6

LI/

II-D

S-1

8

LI/

II-D

S-3

LII

-DS

-11

LII

-DS

-3

LII

-DS

-2

LII

-DS

-8

LII

-US

-1

LII

-DS

-10

LII

-DS

-6

LII

-DS

-9

EDL933 3 111111 I

EC4115 8 211111 I/II

SS17 8 211111 I/II

KCJ1266 8 211111 I/II

* * * * * * * * * * * * * * * * * * *

60

Figure 3-5. Genome map of KCJ1266. Marked characteristics are shown from outside to the center. KCJ1266: CDS on

forward strand, tRNA and rRNA in forward strand, tRNA and rRNA in reward strand, CDS on reverse strand, GC content and GC skew; pO157: CDS on forward strand, tRNA and rRNA in forward strand, tRNA and rRNA in reward strand, CDS on reverse strand, GC content and GC skew.

61

Figure 3-6. Mauve alignments of KCJ1266 with the three reference genome. Alignment of KCJ1266 with the three

reference strains (SS17, EC4115 and EDL933) reveal 7 blocks of homology each of varying from ~45.4 kb to ~ 2,276 kb. Each block is a different color with lines connecting corresponding homologous blocks, with the white regions indicating non-homology. (* indicates the block that only exists in SS17 and EDL933)

62

Figure 3-7. Comparison of O157 plasmids of KCJ1266 and other 3 reference strains

63

Figure 3-8. Subsystem category distribution. The subsystem contains 56% of total genes. The other 46% of genes don’t

belong to any of the subsystem.

64

Figure 3-9. Phage comparison of KCJ1266 with reference strains. From inner ring: phage regions of EC4115, phage regions of SS17, phage regions of KCJ1266. The same color represent homologous regions.

65

Figure 3-10. Virulence related genes in KCJ1266 and reference strains. A. Known genes detected in KCJ1266 B. Putative

genes, hypothetical genes, mobile genes and phage related genes detected in KCJ1266

66

Figure 3-11. Strains are clustered into three groups which belong to Lineage I (green), Lineage II (red) and Lineage I/II

(blue). Within Lineage I/II, there are three subgroups which are cluster1, cluster2 and outgroup. KCJ1266, EC4115 and SS17 belong to outgroup of Lineage I/II.

67

LIST OF REFERENCES

1. Chase-Topping M, Gally D, Low C, Matthews L, Woolhouse M. Super-shedding and the link between human infection and livestock carriage of Escherichia coli O157. Nature reviews Microbiology. 2008;6(12):904-12. doi: 10.1038/nrmicro2029.

2. Gyles CL. Shiga toxin-producing Escherichia coli: an overview. J Anim Sci. 2007;85(13 Suppl):E45-62. doi: 10.2527/jas.2006-508. PubMed PMID: 17085726.

3. Nataro JP, Kaper JB. Diarrheagenic Escherichia coli. Clin Microbiol Rev. 1998;11(1):142-201. PubMed PMID: 9457432; PubMed Central PMCID: PMCPMC121379.

4. Paton JC, Paton AW. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin Microbiol Rev. 1998;11(3):450-79. PubMed PMID: 9665978; PubMed Central PMCID: PMCPMC88891.

5. RL S. Health-related costs from foodborne illness in the United States. 2010. Available at: http://tinyurlcom/l5g2x7n, accessed July 31, 2014. 2010.

6. Matthews L, Low JC, Gally DL, Pearce MC, Mellor DJ, Heesterbeek JA, et al. Heterogeneous shedding of Escherichia coli O157 in cattle and its implications for control. Proceedings of the National Academy of Sciences of the United States of America. 2006;103(3):547-52. doi: 10.1073/pnas.0503776103.

7. Matthews L, McKendrick IJ, Ternent H, Gunn GJ, Synge B, Woolhouse MEJ. Super-shedding cattle and the transmission dynamics of Escherichia coli O157. Epidemiology and infection. 2006;134(1):131-42. doi: 10.1017/S0950268805004590.

8. Jeon SJ, Elzo M, DiLorenzo N, Lamb GC, Jeong KC. Evaluation of animal genetic and physiological factors that affect the prevalence of Escherichia coli O157 in cattle. PloS one. 2013. doi: 10.1371/journal.pone.0055728.

9. Stephens TP, McAllister TA, Stanford K. Perineal swabs reveal effect of super shedders on the transmission of Escherichia coli O157:H7 in commercial feedlots. Journal of Animal Science. 2009;87(12):4151-60. doi: 10.2527/jas.2009-1967.

10. Munns KD, Selinger LB, Stanford K, Guan L, Callaway TR, McAllister TA. Perspectives on super-shedding of Escherichia coli O157:H7 by cattle. Foodborne pathogens and disease. 2015;12(2):89-103. doi: 10.1089/fpd.2014.1829.

68

11. Wells JG, Shipman LD, Greene KD. Isolation of Escherichia coli serotype O157: H7 and other Shiga-like-toxin-producing E. coli from dairy cattle. Journal of Clinical …. 1991.

12. Bell BP, Goldoft M, Griffin PM, Davis MA, Gordon DC, Tarr PI, et al. A multistate outbreak of Escherichia coli O157:H7-associated bloody diarrhea and hemolytic uremic syndrome from hamburgers. The Washington experience. JAMA. 1994;272(17):1349-53. Epub 1994/11/02. PubMed PMID: 7933395.

13. Caprioli A, Morabito S, Brugere H, Oswald E. Enterohaemorrhagic Escherichia coli: emerging issues on virulence and modes of transmission. Veterinary research. 2005;36(3):289-311. Epub 2005/04/23. doi: 10.1051/vetres:2005002. PubMed PMID: 15845227.

14. Centers for disease control. Ongoing Multistate Outbreak of Escherichia coli serotype O157:H7 Infections Associated with Consumption of Fresh Spinach --- United States, September 2006. MMWR. 2006;55:1045-6.

15. Reiland H, Omolo, M. , Johnson, T. and Baumler, D. A Survey of Escherichia coli O157:H7 Virulence Factors: The First 25 Years and 13 Genomes. Advances in Microbiology. 2014;4:390-423. doi: 10.4236/aim.2014.47046.

16. Ng TB, Wong JH, Wang H. Recent progress in research on ribosome inactivating proteins. Current protein & peptide science. 2010;11(1):37-53. Epub 2010/03/06. PubMed PMID: 20201806.

17. Fraser ME, Chernaia MM, Kozlov YV, James MNG. Crystal-Structure of the Holotoxin from Shigella-Dysenteriae at 2.5-Angstrom Resolution. Nature Structural Biology. 1994;1(1):59-64. doi: Doi 10.1038/Nsb0194-59. PubMed PMID: ISI:A1994NU32000013.

18. Lingwood CA. Verotoxin/globotriaosyl ceramide recognition: Angiopathy, angiogenesis and antineoplasia. Bioscience Reports. 1999;19(5):345-54. doi: Doi 10.1023/A:1020299819637. PubMed PMID: ISI:000086086700002.

19. Louise CB, Obrig TG. Specific Interaction of Escherichia-Coli O157/H7-Derived Shiga-Like Toxin-Ii with Human Renal Endothelial-Cells. Journal of Infectious Diseases. 1995;172(5):1397-401. PubMed PMID: ISI:A1995TA55200037.

20. Fuller CA, Pellino CA, Flagler MJ, Strasser JE, Weiss AA. Shiga Toxin Subtypes Display Dramatic Differences in Potency. Infection and Immunity. 2011;79(3):1329-37. doi: 10.1128/Iai.01182-10. PubMed PMID: ISI:000287700200036.

21. Boerlin P, McEwen SA, Boerlin-Petzold F, Wilson JB, Johnson RP, Gyles CL. Associations between virulence factors of Shiga toxin-producing Escherichia coli and disease in humans. Journal of clinical microbiology. 1999;37(3):497-503. PubMed PMID: ISI:000078579500004.

69

22. Kawano K, Okada M, Haga T, Maeda K, Goto Y. Relationship between pathogenicity for humans and stx genotype in Shiga toxin-producing Escherichia coli serotype O157. European Journal of Clinical Microbiology & Infectious Diseases. 2008;27(3):227-32. doi: 10.1007/s10096-007-0420-3. PubMed PMID: ISI:000253994000009.

23. Orth D, Grif K, Khan AB, Naim A, Dierich MP, Wurzner R. The Shiga toxin genotype rather than the amount of Shiga toxin or the cytotoxicity of Shiga toxin in vitro correlates with the appearance of the hemolytic uremic syndrome. Diagnostic Microbiology and Infectious Disease. 2007;59(3):235-42. doi: 10.1016/j.diagmicrobio.2007.04.013. PubMed PMID: ISI:000251122500001.

24. Ostroff SM, Tarr PI, Neill MA, Lewis JH, Hargrettbean N, Kobayashi JM. Toxin Genotypes and Plasmid Profiles as Determinants of Systemic Sequelae in Escherichia-Coli O157-H7 Infections. Journal of Infectious Diseases. 1989;160(6):994-8. PubMed PMID: ISI:A1989CC57500011.

25. Ogura Y, Mondal SI, Islam MR, Mako T, Arisawa K, Katsura K, et al. The Shiga toxin 2 production level in enterohemorrhagic Escherichia coli O157:H7 is correlated with the subtypes of toxin-encoding phage. Sci Rep. 2015;5:16663. doi: 10.1038/srep16663. PubMed PMID: 26567959; PubMed Central PMCID: PMCPMC4645166.

26. Cote R, Katani R, Moreau MR, Kudva IT, Arthur TM. Comparative analysis of super-shedder strains of Escherichia coli O157: H7 reveals distinctive genomic features and a strongly aggregative adherent …. PloS one. 2015. doi: 10.1371/journal.pone.0116743.

27. Giron JA, Torres AG, Freer E, Kaper JB. The flagella of enteropathogenic Escherichia coli mediate adherence to epithelial cells. Molecular microbiology. 2002;44(2):361-79. Epub 2002/04/26. PubMed PMID: 11972776.

28. Tree JJ, Wolfson EB, Wang D, Roe AJ, Gally DL. Controlling injection: regulation of type III secretion in enterohaemorrhagic Escherichia coli. Trends Microbiol. 2009;17(8):361-70. doi: 10.1016/j.tim.2009.06.001. PubMed PMID: 19660954.

29. Lim JY, Yoon J, Hovde CJ. A brief overview of Escherichia coli O157:H7 and its plasmid O157. Journal of microbiology and biotechnology. 2010;20(1):5-14. Epub 2010/02/06. PubMed PMID: 20134227; PubMed Central PMCID: PMC3645889.

30. Lim JY, Li J, Sheng H, Besser TE, Potter K, Hovde CJ. Escherichia coli O157:H7 colonization at the rectoanal junction of long-duration culture-positive cattle. Applied and environmental microbiology. 2007;73(4):1380-2. Epub 2006/12/26. doi: 10.1128/AEM.02242-06. PubMed PMID: 17189448; PubMed Central PMCID: PMC1828644.

31. LeJeune JT, Besser TE, Rice DH, Berg JL, Stilborn RP, Hancock DD. Longitudinal study of fecal shedding of Escherichia coli O157:H7 in feedlot cattle:

70

predominance and persistence of specific clonal types despite massive cattle population turnover. Applied and environmental microbiology. 2004;70(1):377-84. doi: 10.1128/AEM.70.1.377-384.2004.

32. Stanford K, Croy D, Bach SJ, Wallins GL, Zahiroddini H, McAllister TA. Ecology of Escherichia coli O157:H7 in commercial dairies in southern Alberta. Journal of dairy science. 2005;88(12):4441-51.

33. Besser TE, Hancock DD, Pritchett LC, McRae EM, Rice DH, Tarr PI. Duration of detection of fecal excretion of Escherichia coli O157:H7 in cattle. The Journal of infectious diseases. 1997;175(3):726-9.

34. Omisakin F, MacRae M, Ogden ID, Strachan NJ. Concentration and prevalence of Escherichia coli O157 in cattle feces at slaughter. Applied and environmental microbiology. 2003;69(5):2444-7. Epub 2003/05/07. PubMed PMID: 12732509; PubMed Central PMCID: PMC154535.

35. Sargeant JM, Amezcua MR, Rajic A, Waddell L. Pre-harvest interventions to reduce the shedding of E. coli O157 in the faeces of weaned domestic ruminants: a systematic review. Zoonoses and public health. 2007;54(6-7):260-77. Epub 2007/09/07. doi: 10.1111/j.1863-2378.2007.01059.x. PubMed PMID: 17803515.

36. Chase-Topping ME, McKendrick IJ, Pearce MC, MacDonald P, Matthews L, Halliday J, et al. Risk factors for the presence of high-level shedders of Escherichia coli O157 on Scottish farms. Journal of clinical microbiology. 2007;45(5):1594-603. doi: 10.1128/JCM.01690-06.

37. Locking M, Browning L, Smith-Palmer A. Gastro-intestinal and foodborne infections. HPS Weekly Report. 2010.

38. Arthur TM, Ahmed R, Chase-Topping M, Kalchayanand N, Schmidt JW, Bono JL. Characterization of Escherichia coli O157: H7 strains isolated from supershedding cattle. Applied and environmental microbiology. 2013;79(14):4294-303. doi: 10.1128/AEM.00846-13.

39. Jokinen C, Edge TA, Ho S, Koning W, Laing C, Mauro W, et al. Molecular subtypes of Campylobacter spp., Salmonella enterica, and Escherichia coli O157:H7 isolated from faecal and surface water samples in the Oldman River watershed, Alberta, Canada. Water research. 2011;45(3):1247-57. Epub 2010/10/26. doi: 10.1016/j.watres.2010.10.001. PubMed PMID: 20971491.

40. Goering RV. Pulsed field gel electrophoresis: a review of application and interpretation in the molecular epidemiology of infectious disease. Infection, genetics and evolution : journal of molecular epidemiology and evolutionary genetics in infectious diseases. 2010;10(7):866-75. Epub 2010/08/10. doi: 10.1016/j.meegid.2010.07.023. PubMed PMID: 20692376.

71

41. Stanford K, Agopsowicz CA. Genetic diversity and antimicrobial resistance among isolates of Escherichia coli O157: H7 from feces and hides of super-shedders and low-shedding pen- …. BMC veterinary …. 2012. doi: 10.1186/1746-6148-8-178.

42. Kim J, Nietfeldt J, Benson AK. Octamer-based genome scanning distinguishes a unique subpopulation of Escherichia coli O157:H7 strains in cattle. Proceedings of the National Academy of Sciences of the United States of America. 1999;96(23):13288-93. Epub 1999/11/11. PubMed PMID: 10557313; PubMed Central PMCID: PMC23940.

43. Zhang Y, Laing C, Steele M, Ziebell K, Johnson R, Benson AK, et al. Genome evolution in major Escherichia coli O157:H7 lineages. BMC Genomics. 2007;8:121. doi: 10.1186/1471-2164-8-121. PubMed PMID: 17506902; PubMed Central PMCID: PMCPMC1890555.

44. Stanton E, Park D, Dopfer D, Ivanek R, Kaspar CW. Phylogenetic characterization of Escherichia coli O157 : H7 based on IS629 distribution and Shiga toxin genotype. Microbiology. 2014;160(Pt 3):502-13. Epub 2014/01/16. doi: 10.1099/mic.0.073437-0. PubMed PMID: 24425770.

45. Munns KD. Characterization of Escherichia coli O157: H7 from super-shedding feedlot cattle. Characterization of Escherichia coli O157: H7 from super-shedding feedlot cattle. 2015.

46. Bono JL, Keen JE, Clawson ML, Durso LM. Association of Escherichia coli O157: H7 tir polymorphisms with human infection. BMC infectious …. 2007.

47. Besser TE, Shaikh N, Holt NJ, Tarr PI, Konkel ME, Malik-Kale P, et al. Greater Diversity of Shiga Toxin-Encoding Bacteriophage Insertion Sites among Escherichia coli O157:H7 Isolates from Cattle than in Those from Humans. Applied and Environmental Microbiology. 2006;73(3). doi: 10.1128/AEM.01035-06.

48. Hayashi T, Makino K, Ohnishi M, Kurokawa K, Ishii K, Yokoyama K, et al. Complete genome sequence of enterohemorrhagic Escherichia coli O157 : H7 and genomic comparison with a laboratory strain K-12. DNA Res. 2001;8(1):11-22. doi: DOI 10.1093/dnares/8.1.11. PubMed PMID: WOS:000167291700002.

49. Perna NT, Plunkett G, 3rd, Burland V, Mau B, Glasner JD, Rose DJ, et al. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature. 2001;409(6819):529-33. doi: 10.1038/35054089. PubMed PMID: 11206551.

50. Shere JA, Kaspar CW, Bartlett KJ. Shedding of Escherichia coli O157: H7 in dairy cattle housed in a confined environment following waterborne inoculation. Applied and …. 2002. doi: 10.1128/AEM.68.4.1947-1954.2002.

72

51. Durso LM, Harhay GP, Smith TPL, Bono JL, DeSantis TZ, Harhay DM, et al. Animal-to-Animal Variation in Fecal Microbial Diversity among Beef Cattle. Applied and environmental microbiology. 2010;76(14):4858-62. doi: 10.1128/Aem.00207-10. PubMed PMID: ISI:000279611500033.

52. Shanks OC, Kelty CA, Archibeque S, Jenkins M, Newton RJ, McLellan SL, et al. Community Structures of Fecal Bacteria in Cattle from Different Animal Feeding Operations. Applied and environmental microbiology. 2011;77(9):2992-3001. doi: 10.1128/Aem.02988-10. PubMed PMID: ISI:000289773100022.

53. Xu Y, Dugat-Bony E, Zaheer R, Selinger L, Barbieri R, Munns K, et al. Escherichia coli O157:H7 super-shedder and non-shedder feedlot steers harbour distinct fecal bacterial communities. PloS one. 2014;9(5):e98115. Epub 2014/05/27. doi: 10.1371/journal.pone.0098115. PubMed PMID: 24858731; PubMed Central PMCID: PMC4032279.

54. Guarner F, Malagelada JR. Role of bacteria in experimental colitis. Best Practice & Research in Clinical Gastroenterology. 2003;17(5):793-804. doi: 10.1016/S1521-6918(03)00068-4. PubMed PMID: ISI:000185868900008.

55. Barkocy-Gallagher GA, Arthur TM, Rivera-Betancourt M, Nou XW, Shackelford SD, Wheeler TL, et al. Seasonal prevalence of Shiga toxin-producing Escherichia coli, including O157 : H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. Journal of Food Protection. 2003;66(11):1978-86. PubMed PMID: ISI:000186488900003.

56. Laegreid WW, Elder RO, Keen JE. Prevalence of Escherichia coli O157:H7 in range beef calves at weaning. Epidemiology and infection. 1999;123(2):291-8. Epub 1999/12/01. PubMed PMID: 10579450; PubMed Central PMCID: PMC2810762.

57. Heuvelink AE, Bleumink B, van den Biggelaar FL, Te Giffel MC, Beumer RR, de Boer E. Occurrence and survival of verocytotoxin-producing Escherichia coli O157 in raw cow's milk in The Netherlands. Journal of Food Protection. 1998;61(12):1597-601. Epub 1999/01/05. PubMed PMID: 9874335.

58. Callaway TR, Carroll JA, Arthington JD, Edrington TS, Rossman ML, Carr MA, et al. Escherichia coli O157:H7 Populations in Ruminants Can Be Reduced by Orange Peel Product Feeding. Journal of Food Protection. 2011;74(11):1917-21. doi: 10.4315/0362-028X.JFP-11-234. PubMed PMID: ISI:000296756800017.

59. Fox JT, Depenbusch BE, Drouillard JS, Nagaraja TG. Dry-rolled or steam-flaked grain-based diets and fecal shedding of Escherichia coli O157 in feedlot cattle. J Anim Sci. 2007;85(5):1207-12. doi: 10.2527/jas.2006-079. PubMed PMID: 17224458.

73

60. Naylor SW, Low JC, Besser TE, Mahajan A. Lymphoid follicle-dense mucosa at the terminal rectum is the principal site of colonization of enterohemorrhagic Escherichia coli O157: H7 in the bovine host. Infection and …. 2003.

61. Bono JL, Keen JE, Clawson ML, Durso LM, Heaton MP, Laegreid WW. Association of Escherichia coli O157:H7 tir polymorphisms with human infection. BMC infectious diseases. 2007;7:98. Epub 2007/08/28. doi: 10.1186/1471-2334-7-98. PubMed PMID: 17718910; PubMed Central PMCID: PMC2063500.

62. Manning SD, Motiwala AS, Springman AC, Qi W, Lacher DW, Ouellette LM, et al. Variation in virulence among clades of Escherichia coli O157:H7 associated with disease outbreaks. Proc Natl Acad Sci U S A. 2008;105(12):4868-73. doi: 10.1073/pnas.0710834105. PubMed PMID: 18332430; PubMed Central PMCID: PMCPMC2290780.

63. Kim J, Nietfeldt J, Ju J, Wise J, Fegan N, Desmarchelier P, et al. Ancestral divergence, genome diversification, and phylogeographic variation in subpopulations of sorbitol-negative, beta-glucuronidase-negative enterohemorrhagic Escherichia coli O157. Journal of bacteriology. 2001;183(23):6885-97. Epub 2001/11/08. doi: 10.1128/JB.183.23.6885-6897.2001. PubMed PMID: 11698378; PubMed Central PMCID: PMC95530.

64. Jeon SJ, Oh M, Yeo WS, Galvao KN, Jeong KC. Underlying mechanism of antimicrobial activity of chitosan microparticles and implications for the treatment of infectious diseases. PLoS One. 2014;9(3):e92723. doi: 10.1371/journal.pone.0092723. PubMed PMID: 24658463; PubMed Central PMCID: PMCPMC3962451.

65. Chen L, Xiong Z, Sun L, Yang J, Jin Q. VFDB 2012 update: toward the genetic diversity and molecular evolution of bacterial virulence factors. Nucleic acids research. 2012;40(Database issue):D641-5. Epub 2011/11/10. doi: 10.1093/nar/gkr989. PubMed PMID: 22067448; PubMed Central PMCID: PMC3245122.

66. Chen L, Yang J, Yu J, Yao Z, Sun L, Shen Y, et al. VFDB: a reference database for bacterial virulence factors. Nucleic acids research. 2005;33(Database issue):D325-8. Epub 2004/12/21. doi: 10.1093/nar/gki008. PubMed PMID: 15608208; PubMed Central PMCID: PMC539962.

67. Chen L, Zheng D, Liu B, Yang J, Jin Q. VFDB 2016: hierarchical and refined dataset for big data analysis-10 years on. Nucleic acids research. 2016;44(D1):D694-7. Epub 2015/11/19. doi: 10.1093/nar/gkv1239. PubMed PMID: 26578559; PubMed Central PMCID: PMC4702877.

68. Yang J, Chen L, Sun L, Yu J, Jin Q. VFDB 2008 release: an enhanced web-based resource for comparative pathogenomics. Nucleic acids research. 2008;36(Database issue):D539-42. Epub 2007/11/07. doi: 10.1093/nar/gkm951. PubMed PMID: 17984080; PubMed Central PMCID: PMC2238871.

74

69. Gillespie JJ, Wattam AR, Cammer SA, Gabbard JL, Shukla MP, Dalay O, et al. PATRIC: the comprehensive bacterial bioinformatics resource with a focus on human pathogenic species. Infection and immunity. 2011;79(11):4286-98. Epub 2011/09/08. doi: 10.1128/IAI.00207-11. PubMed PMID: 21896772; PubMed Central PMCID: PMC3257917.

70. Snyder EE, Kampanya N, Lu J, Nordberg EK, Karur HR, Shukla M, et al. PATRIC: the VBI PathoSystems Resource Integration Center. Nucleic acids research. 2007;35(Database issue):D401-6. Epub 2006/12/05. doi: 10.1093/nar/gkl858. PubMed PMID: 17142235; PubMed Central PMCID: PMC1669763.

71. Wattam AR, Abraham D, Dalay O, Disz TL, Driscoll T, Gabbard JL, et al. PATRIC, the bacterial bioinformatics database and analysis resource. Nucleic acids research. 2014;42(Database issue):D581-91. Epub 2013/11/15. doi: 10.1093/nar/gkt1099. PubMed PMID: 24225323; PubMed Central PMCID: PMC3965095.

72. Franz E, Semenov AV. Modelling the contamination of lettuce with Escherichia

coli O157: H7 from manure‐amended soil and the effect of intervention strategies. Journal of applied …. 2008. doi: 10.1111/j.1365-2672.2008.03915.x.

73. Eppinger M, Mammel MK, Leclerc JE, Ravel J, Cebula TA. Genomic anatomy of Escherichia coli O157:H7 outbreaks. Proc Natl Acad Sci U S A. 2011;108(50):20142-7. doi: 10.1073/pnas.1107176108. PubMed PMID: 22135463; PubMed Central PMCID: PMCPMC3250189.

74. Laing CR, Buchanan C, Taboada EN, Zhang Y, Karmali MA, Thomas JE, et al. In silico genomic analyses reveal three distinct lineages of Escherichia coli O157:H7, one of which is associated with hyper-virulence. BMC Genomics. 2009;10:287. doi: 10.1186/1471-2164-10-287. PubMed PMID: 19563677; PubMed Central PMCID: PMCPMC2719669.

75. Kulasekara BR, Jacobs M, Zhou Y, Wu Z, Sims E, Saenphimmachak C, et al. Analysis of the genome of the Escherichia coli O157:H7 2006 spinach-associated outbreak isolate indicates candidate genes that may enhance virulence. Infect Immun. 2009;77(9):3713-21. doi: 10.1128/IAI.00198-09. PubMed PMID: 19564389; PubMed Central PMCID: PMCPMC2738036.

76. Zhou Y, Liang Y, Lynch KH, Dennis JJ, Wishart DS. PHAST: a fast phage search tool. Nucleic acids research. 2011;39(Web Server issue):W347-52. Epub 2011/06/16. doi: 10.1093/nar/gkr485. PubMed PMID: 21672955; PubMed Central PMCID: PMC3125810.

77. Amigo N, Mercado E, Bentancor A, Singh P, Vilte D, Gerhardt E, et al. Clade 8 and Clade 6 Strains of Escherichia coli O157: H7 from Cattle in Argentina have

75

Hypervirulent-Like Phenotypes. PloS one. 2015;10(6). doi: 10.1371/journal.pone.0127710.

78. Jeong KC, Kang MY, Kang J, Baumler DJ. Reduction of Escherichia coli O157: H7 shedding in cattle by addition of chitosan microparticles to feed. Applied and …. 2011.

79. Pruimboom-Brees IM, Morgan TW, Ackermann MR, Nystrom ED, Samuel JE, Cornick NA, et al. Cattle lack vascular receptors for Escherichia coli O157:H7 Shiga toxins. Proc Natl Acad Sci U S A. 2000;97(19):10325-9. doi: 10.1073/pnas.190329997. PubMed PMID: 10973498; PubMed Central PMCID: PMCPMC27023.

76

BIOGRAPHICAL SKETCH

Lin Teng was born in Fuzhou, Fujian, China in 1990. He enrolled at the

Huazhong Agricultural University in Wuhan, China, one of the most foremost

Agricultural Universities in China in 2008. His B.S. project focused on constructing

deletion mutant strains of extraintestinal pathogenic Escherichia coli and evaluating the

change of toxicity of mutant strains. Lin earned his Bachelor of Agronomy degree at

Huazhong Agricultural University in 2012. After graduation from the university, Lin

served in Fujian Fengze Group Cooperation, a feed and meat producer, for one year as

a research assistant. In 2014, he was accepted by the Department of Animal Science of

University of Florida as a M.S. student and joined Dr. KwangCheol Casey Jeong’s lab.

During his M.S. program, he got scientific training and participated in some other project,

such as the projects of Chitosan nanoparticles and antimicrobial resistance. His long-

term plan is to work in the industry in China. His immediate plan is to learn more

bioinformatics techniques in the current lab and finish his other ongoing projects.