RISK FACTORS FOR SHIGA TOXIN-PRODUCING ESCHERICHIA COLI IN CATTLE
By
María Cristina Venegas Vargas

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Large Animal Clinical Sciences−Doctor of Philosophy
2015

ABSTRACT
RISK FACTORS FOR SHIGA TOXIN-PRODUCING ESCHERICHIA COLI IN CATTLE
By
María Cristina Venegas Vargas
Shiga toxin-producing Escherichia coli (STEC) is one of the most important food borne
pathogens of humans globally, having caused numerous outbreaks in North America and
worldwide. Severe clinical disease occurs primarily in children and immunocompromised adults
and signs range from mild diarrhea to hemorrhagic colitis to Hemolytic Uremic Syndrome,
which can result in kidney failure and mortality.
Cattle are considered the main reservoir of STEC and food or water contaminated with
cattle feces is considered to be a major source of human exposure. Common foods implicated in
STEC outbreaks include ground beef, unpasteurized milk, leafy vegetables and apple cider.
Other domesticated animals and wildlife can also shed STEC, but their importance as a source of
human exposure is considered less significant. Human infections have also been reported
following direct and indirect contact with animals at zoos, livestock exhibitions and petting
farms.
Identifying factors that influence STEC shedding and dynamics in cattle is important for
the design and implementation of strategies to prevent STEC transmission. The studies describe
in this dissertation are the results obtained from an epidemiological study performed in 11 herds
in Mid-Michigan during 2011 and 2012. The primary aims of this project were to identify risk
factors for STEC shedding and describe STEC dynamics in cattle. Of specific interest was the
potential effect on STEC shedding of pre-existing chronic disease, specifically infection with

Bovine Leukemia Virus (BLV) and Mycobacterium avium subsp. paratuberculosis (MAP), the
causative agent of Johnes disease.
We identified several variables including days in milk and number of lactations, to be
important individual factors that influence the risk of STEC shedding in dairy cattle. Intervention
strategies could be targeted towards these high risk cattle groups. We also confirmed the
importance of seasonality, more specifically warm temperatures, on STEC shedding by cattle.
No association was observed between STEC shedding and infection with BLV and MAP.
We found a significant association between the independent variable herd and rate of new
infections with STEC; also we found a significant association between herd and persistent STEC
negative status, both in dairy herds. However, we were not able to identify specific management
factors that influence the risk of STEC shedding over time. These finding highlight the complex
and multifactorial nature of STEC epidemiology in cattle.
Based on the results obtained in this dissertation, we conclude that first lactation cows
and cows in their first 30 days of lactation have the highest risk of STEC shedding. As a
consequence, these specific groups of cattle can be targeted for the design and implementation of
intervention strategies at pre-harvest with the aim of reducing STEC infection in humans.

Dedicated to my family

iv

ACKNOWLEDGMENTS

I would like to thank Dr. Grooms for his support, mentorship and positivity. I will always
appreciate your faith in me.
Also thank you to Dr. Bartlett for his mentorship and help during the sampling and analysis
during the project.
Thanks to Dr. Manning for her support, mentorship and help throughout my PhD.
I would like to thank Dr. Funk for her advice and help during my time here at MSU. You are one
of my role models.
Also thanks to Dr. Norby for his help in the statistical analysis.
Thanks to Dr. Pires for her help in the statistical analysis and her friendship.
Thanks to Dr. Manning’s lab personnel for all of their help on this project.
Thanks to my Costa Rican friends who always made me feel at home.
I would like to thank my family who always support and encourage me to pursue my goals.
Lastly, I would like to thank my friends because it would have been impossible for me to finish
this PhD without their support. You became my Michigan family.
Accomplishments have not value if you have no one to share them with.

v

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... ix
LIST OF FIGURES ....................................................................................................................... xi
KEY TO ABBREVIATIONS ....................................................................................................... xii
INTRODUCTION .......................................................................................................................... 1
REFERENCES ............................................................................................................................ 5
CHAPTER 1 ................................................................................................................................... 9
LITERATURE REVIEW: RISK FACTORS FOR SHEDDING OF SHIGA TOXINPRODUCING ESCHERICHIA COLI (STEC) IN DAIRY AND BEEF CATTLE ...................... 9
STEC ............................................................................................................................................ 9
1. Escherichia coli ................................................................................................................... 9
2. Shiga toxin-producing Escherichia coli (STEC) ............................................................... 10
Importance of STEC in Public Health ....................................................................................... 15
1. Clinical disease in humans ................................................................................................ 17
2. STEC pathology in animals ............................................................................................... 18
STEC reservoirs and transmission ............................................................................................. 19
Risk factors for STEC shedding by both dairy and beef cattle .................................................. 20
Intervention strategies at the pre-harvest level .......................................................................... 27
1. Exposure reduction strategies ............................................................................................ 28
1.1. Environmental exposure ............................................................................................ 28
1.2. Wildlife exclusion ...................................................................................................... 28
2. Exclusion strategies ........................................................................................................... 28
2.1 Probiotics .................................................................................................................... 28
2.2 Prebiotics..................................................................................................................... 29
2.3 Other diet supplements ............................................................................................... 29
3. Direct antipathogen strategies ........................................................................................... 30
3.1 Antimicrobial Compounds .......................................................................................... 30
3.2 Bacteriophage Therapy ............................................................................................... 30
3.3 Vaccination ................................................................................................................. 30
Conclusions ................................................................................................................................ 32
REFERENCES .......................................................................................................................... 34
CHAPTER 2 ................................................................................................................................. 47
RISK FACTORS FOR SHIGA TOXIN-PRODUCING ESCHERICHIA COLI (STEC)
SHEDDING IN CATTLE............................................................................................................. 47
Abstract ...................................................................................................................................... 47
Introduction ................................................................................................................................ 49
Materials and methods ............................................................................................................... 53
1. Study design and herd selection ........................................................................................ 53
1.1 Questionnaire .............................................................................................................. 54
vi

1.2 Sampling ..................................................................................................................... 54
2. Laboratory protocol for STEC detection and isolation .................................................... 55
3. Data analyses ..................................................................................................................... 56
Results ........................................................................................................................................ 58
1. Descriptive statistics .......................................................................................................... 58
2. Univariate analyses of STEC shedding in dairy herds ...................................................... 59
2.1. Individual host factors: .............................................................................................. 59
2.2. Environmental factors: ............................................................................................... 60
2.2.1 Herd characteristics ......................................................................................... 60
2.2.2 Housing characteristic ..................................................................................... 60
2.2.3 Cleaning characteristics................................................................................... 61
2.2.4 Treatment characteristics ................................................................................. 61
2.2.5 Diet ................................................................................................................... 62
2.2.6 Contact with other animals .............................................................................. 62
2.2.7 Environmental conditions................................................................................. 63
2.3. Multivariable analysis for dairy: ................................................................................ 63
3. Univariate analyses of STEC shedding in beef herds ....................................................... 64
3.1 Environmental factors: ................................................................................................ 65
3.1.1 Herd characteristics ......................................................................................... 65
3.1.2 Housing characteristics .................................................................................... 65
3.1.3 Treatment characteristics ................................................................................. 66
3.1.4 Diet ................................................................................................................... 66
3.1.5 Contact with other animals .............................................................................. 66
3.1.6 Environmental conditions................................................................................. 67
Discussion .................................................................................................................................. 68
APPENDIX ................................................................................................................................ 72
REFERENCES ........................................................................................................................ 111
CHAPTER 3 ............................................................................................................................... 118
ASSOCIATION OF BOVINE LEUKEMIA VIRUS AND MYCOBACTERIUM AVIUM SUBSP.
PARATUBERCULOSIS WITH SHEDDING OF SHIGA TOXIN-PRODUCING ESCHERICHIA
COLI ........................................................................................................................................... 118
Abstract .................................................................................................................................... 118
Introduction .............................................................................................................................. 120
Materials and methods ............................................................................................................. 122
1. Animal selection .............................................................................................................. 122
2. Fecal sample collection and analysis ............................................................................... 122
3. Blood collection and analysis .......................................................................................... 123
4. Data collection and analysis ............................................................................................ 124
Results ...................................................................................................................................... 126
1. Descriptive statistics ........................................................................................................ 126
2. Univariable models.......................................................................................................... 127
3. Multivariable models ....................................................................................................... 127
Discussion ................................................................................................................................ 129
APPENDIX .............................................................................................................................. 132
REFERENCES ........................................................................................................................ 139
vii

CHAPTER 4 ............................................................................................................................... 143
SHIGA TOXIN-PRODUCING ESCHERICHIA COLI ACQUISITION, LOSS AND
PERSISTENCE IN CATTLE ..................................................................................................... 143
Abstract .................................................................................................................................... 143
Introduction .............................................................................................................................. 145
Materials and methods ............................................................................................................. 146
1. Herd selection .................................................................................................................. 146
2. Study design .................................................................................................................... 146
2.1 Sampling ................................................................................................................... 147
3. Laboratory protocol for STEC detection and isolation ................................................... 148
4. Data analyses ................................................................................................................... 149
Results ...................................................................................................................................... 152
1. Descriptive statistics ........................................................................................................ 152
2. STEC Loss and Acquisition Rate .................................................................................... 153
3. ANY STEC LOSS and ANY STEC ACQUISITION ..................................................... 153
4. Rate of new STEC infections .......................................................................................... 154
5. Univariate analysis of herd level management risk factors ............................................. 154
Discussion ................................................................................................................................ 155
APPENDIX .............................................................................................................................. 161
REFERENCES ........................................................................................................................ 181
CONCLUSIONS AND FUTURE STUDIES ............................................................................. 186
Conclusions .............................................................................................................................. 186
Future studies ........................................................................................................................... 189

viii

LIST OF TABLES

Table 2.1. Areas explored in the questionnaire for dairy and beef herds. ................................... 73
Table 2.2. Herd identification, type of production system, total number of animals in each herd,
number of animals sampled in each herd, number of animals tested for STEC and year of
sampling. ....................................................................................................................................... 74
Table 2.3. Prevalence of genes stx1, stx2 and eaeA by total of animals, dairy animals and beef
animals. ......................................................................................................................................... 76
Table 2.4. Univariable analysis of dairy herd variables for risk of STEC shedding with herd as a
random effect ................................................................................................................................ 79
Table 2.5. Final multivariable model for dairy herds with herd as a random effect. ................... 86
Table 2.6. Univariable analysis of beef herd variables for risk of STEC shedding with herd as a
random effect ................................................................................................................................ 87
Table 2.7. Description of the abbreviations used for each variable analyzed in the beef and dairy
models. .......................................................................................................................................... 92
Table 3.1. Prevalence of bovine leukemia virus (BLV), Mycobacterium avium subsp.
paratuberculosis (MAP) and STEC by herd. * B=beef, D=dairy .............................................. 133
Table 3.2. Mean and Standard deviation (SD) for the percentage of neutrophils, lymphocytes and
L: M in cattle positive and negative to bovine leukemia virus (BLV) and Mycobacterium avium
subsp. paratuberculosis (MAP) according to STEC status. ....................................................... 134
Table 3.3. Univariable analysis to evaluate the association between risk factors and the
dependent variable for STEC shedding. ..................................................................................... 135
Table 3.4. Univariable analysis to evaluated ELISA for bovine leukemia virus (BLV) status,
Mycobacterium avium subsp. paratuberculosis (MAP) status and percentage of neutrophils,
lymphocytes, and lymphocytes monocytes ratio for their association with STEC shedding in
animals. ....................................................................................................................................... 136
Table 3.5. Final multivariable model for both beef and dairy herds to evaluate bovine leukemia
virus (BLV) and Mycobacterium avium subsp. paratuberculosis (MAP) status as determinants of
STEC shedding. .......................................................................................................................... 137
Table 3.6. Final multivariable model for dairy herds to evaluated bovine leukemia virus (BLV)
and Mycobacterium avium subsp. paratuberculosis (MAP) status as determinants of STEC
shedding. ..................................................................................................................................... 138

ix

Table 4.1. Univariate analysis of dairy herd variables associated with persistently STECnegative cattle. ............................................................................................................................ 168
Table 4.2. Univariate analysis to identify risk factors for rate of new STEC infections in dairy
cattle ............................................................................................................................................ 175

x

LIST OF FIGURES

Figure 2.1. Prevalence of STEC by herd. D= dairy, B= beef ...................................................... 75
Figure 2.2. Prevalence of stx1, stx2 and stx1/2 by herd. D= dairy, B=beef. The vertical axis has
been set up at 50% to improve visibility. B= beef and D= dairy .................................................. 77
Figure 2.3. Prevalence of enterohemorrhagic Escherichia coli and Shiga toxin-producing E. coli
strains by herd. D= dairy, B=beef. The vertical axis has been set up at 50% to improve visibility.
....................................................................................................................................................... 78
Figure 2.4. Dairy and beef questionnaires used to collect information from the herds. .............. 95
Figure 4.2. Rate STEC LOSS over time by herd. Event represents the time between one
sampling and the next one. Event 1= Phase II to Phase 3.1, Event 2= Phase 3.1 to 3.2 and Event
3= Phase 3.2 to 3.3. B= beef and D= dairy. ................................................................................ 163
Figure 4.3. Rate of STEC ACQUISITION over time by herd. Event represents the time between
one sampling and the next one. Event 1= Phase II to Phase 3.1, Event 2= Phase 3.1 to Phase 3.2
and Event 3= Phase 3.2 to 3.3. B= beef and D= dairy. ............................................................... 164
Figure 4.4. Percentage of animals that lost STEC at any time during the study by herd. B= beef
and D= dairy ............................................................................................................................... 165
Figure 4.5. Percentage of cattle that acquired STEC at any time during the study by herd. ..... 166
B= beef and D= dairy.................................................................................................................. 166
Figure 4.6. Rate of STEC NEW INFECTIONS (number of new infections in a cattle divided by
the number of times cattle was at risk or susceptible to new infection) by herd. B = beef and D =
dairy. ........................................................................................................................................... 167

xi

KEY TO ABBREVIATIONS

AA: Any STEC acquisition
AEE: Attaching and effacing E. coli
AL: Any STEC loss
BLV: Bovine Leukemia Virus
cfu: colony-forming unit
eae: gene encoding the E. coli attaching and effacing protein, intimin
EAEC: Enteroaggregative E. coli
ED: Edema disease
EHEC: Enterohemorrhagic E. coli
EIEC: Enteroinvasive E. coli
EPEC: Enteropathogenic E. coli
ETEC: Enterotoxigenic E. coli
ExPEC: Extraintestinal E. coli
DAEC: Diffusely adherent E. coli
DEC: diarrheagenic E. coli
DIM: Days in milk
Gb3: Globotriaosylceramide
Gb4: Globotetraosylceramide
HUS: Hemolytic Uremic Syndrome
IMS: Immunomagnetic separation
LEE: Locus of Enterocyte Effacement

xii

L:M: Lymphocyte to monocyte radio
MAP: Mycobacterium avium subsp. paratuberculosis
MNEC: Meningitis-associated E. coli
PP: Persistent STEC positive
RA: Rate of STEC acquisition
RL: Rate of STEC loss
RNI: Rate of new STEC infections
SRP: Siderophore receptor and porin proteins
STEC: Shiga toxin-producing Escherichia coli
Tir: Translocated intimin receptor
UPEC: Uropathogenic E. coli
VTEC: Vero toxin-producing E. coli

xiii

INTRODUCTION

Shiga toxin-producing Escherichia coli (STEC) are one of the most important foodborne
pathogens in the U.S. and other developed countries. STEC can cause hemorrhagic diarrhea and
hemolytic uremic syndrome (HUS) that can lead to kidney failure and death, particularly in
young children (Vanaja, et al 2013).
STEC, also known as Vero toxin producing E. coli, is defined by the presence of genes
encoding the Shiga toxin (Stx). Two main types of Stx can be produced by STEC: Stx1 (almost
identical to Stx from Shigella dysenteria type 1) and Stx2 (Gyles 2007; Scheutz, et al 2012).
Shiga toxins can be further broken down into subtypes based on differences in biological
properties of the toxins (Scheutz, et al 2012). Shiga toxins are bifunctional bacterial toxins,
composed by two units A and B, such as cholera toxin (O'Brien and Holmes 1987). The best
known receptor for Stx is globotriaosylceramide (Gb3), a membrane cell surface receptor
(Jacewicz, et al 1986).
STEC isolates can be further classified into those that contain the Locus of Enterocyte
Effacement (LEE) pathogenicity island and those that do not (Sahl, et al 2013). LEE encodes a
type III secretion system that injects infectors into the host cell that produce the formation of
attaching and effacing (AE) lesions (Gyles 2007; Sahl, et al 2013). Isolates positive for the LEE
island are considered enterohemorrhagic E. coli (EHEC), a subset of STEC.
STEC O157:H7 were identified for the first time in the U.S. in 1982 (Riley, et al 1983).
Based on the molecular analysis of these outbreak isolates, it was demonstrated that a lambdoid
prophage transferred into E. coli the genes required to produce Stx, resulting in a newly
emergent pathogen (O'Brien, et al 1984). Since its emergence, E. coli O157:H7 has been the
1

most commonly isolated serotype among over 100 STEC serotypes (Vanaja, et al 2013), which
has been partially due to enhanced detection protocols from human and animal feces. The burden
of illness from serotypes other than O157 (non-O157) STEC has previously been difficult to
estimate as culture systems were incapable of differentiating these strains, leading to diagnostic
limitations and inadequate surveillance (Farrokh, et al 2013; Vanaja, et al 2013). During the last
10 years, non-O157 serotypes have increased in frequency (Bettelheim 2007) and have
contributed to several large-scale outbreaks. Currently, six non-O157 STEC serogroups (O26,
O45, O103, O111, O121, and O145) and O157:H7 have been classified by the USDA-FSIS as
adulterants of all raw non-intact beef and raw intact beef intended for use in raw-intact products
in the U.S. under the Federal Meat Inspection Act (21 U.S.C. 601 (m) (1)).
Transmission of STEC is a complex process that involves different reservoirs, hosts and
environments. Ruminants, especially cattle, are the major STEC reservoir. Humans most
commonly get infected through the consumption of contaminate food, among the most common
food items have been beef products (especially ground beef) or dairy products contaminated with
feces. However there have been outbreaks with other less common sources such as unpasteurized
apple juice, spinach and salami (Chase-Topping, et al 2008; Hussein 2007; Jay, et al 2007;
Menrath, et al 2010). Produce can also get contaminated with STEC through the inclusion of
manure in soil and later uptake by plants (Callaway, et al 2013; Franz and van Bruggen 2008).
Rainfall events can wash STEC from cattle feces into drinking, recreation or irrigation water
supplies, which have led to infection in humans and other animals (Berry and Wells 2010;
Callaway, et al 2013; Franz and van Bruggen 2008; Hussein 2007; Solomon, et al 2002). Direct
contact with cattle is another route of transmission to humans (Goode, et al 2009). Other animals
species such deer and birds can also be sources of STEC (Asakura, et al 1998; Callaway, et al

2

2013; Dunn 2003; Ferens and Hovde 2011), while person-to-person transmission has also been
documented (Gyles 2007; Pennington 2010; Tarr, et al 2005).
According to data collected by FoodNet USA, there were 561 cases (1.17 per 100, 000
people) of non-O157 associated illness and 552 cases (1.15 cases per 100,000) of STEC O157
associated illness in the U.S in 2013. In this same year, there were 78 hospitalizations and 2
deaths for non-O157 and 210 hospitalizations and 2 deaths for STEC O157 (Crim, et al 2014).
Several researchers have concluded that control measures at the pre-harvest level will
have the greatest impact on the reduction of STEC infections in humans (LeJeune and Wetzel
2007; Soon, et al 2011). A solid understanding of the epidemiology of STEC is critical to
implementing control measures at the pre-harvest level. Although numerous studies have sought
to determine the prevalence of STEC in animal reservoirs and varying geographic locations,
additional studies are still needed to better understand the risk factors associated with STEC
shedding at both the herd and animal level. Similarly, more research is needed to identify which
groups of cattle have the highest risk of STEC colonization and shedding as these groups
represent the best targets for pre-harvest intervention strategies. To date, few consistent risk
factors have been identified for STEC O157 shedding in cattle across studies (Cho, et al 2013;
Menrath, et al 2010). Some studies have found similar factors while others have found
contradictory risk factors. However, most research has been focused on STEC O157 rather than
non-O157. Additional large-scale studies are therefore needed to better understand the
transmission dynamics of STEC within and across herds with varying management practices.
The overall purpose of the research conducted and presented in this dissertation was to
identify new risk factors in beef and dairy cattle that influence STEC shedding. The ultimate goal

3

was to identify management factors that could be modified to reduce STEC colonization and
shedding levels, thereby minimizing the potential risk of human infections.
In chapter one, a literature review on STEC and information available regarding studies
on STEC risk factors was presented. This information is useful in understanding what is known
and not known about STEC shedding and associated risk factors and was used to inform the
development of our subsequent studies.
In chapter two, findings on potential risk factors for STEC shedding at both the herd and
individual animal level were reported. By understanding potential risk factors, intervention
strategies can be designed to reduce pre-harvest STEC shedding, thus reducing the risk of human
infections.
In chapter three, we tested the hypothesis that cattle infected with Bovine Leukemia
Virus (BLV) and/or Mycobacterium avium subsp. paratuberculosis (MAP) were more likely to
shed STEC. BLV and MAP are chronic infectious diseases that have significant long-term
effects on the immune system (BLV) and gastrointestinal tract (MAP), thus creating a situation
where the dynamics of infection and shedding of organisms such as STEC may be altered. If
these chronic diseases have an effect on STEC shedding, then implementing management
practices to control these important diseases could indirectly influence STEC shedding as well.
In chapter four, we conducted a longitudinal study to investigate rates of STEC
acquisition, persistence and loss in cattle and to identify factors that increase or reduce STEC
acquisition or persistence. Combining this information with the findings from chapter two could
further add to and refine the development of intervention strategies at the pre-harvest level.

4

REFERENCES

5

REFERENCES

ASAKURA, H., MAKINO, S., SHIRAHATA, T., TSUKAMOTO, T., KURAZONO, H.,
IKEDA, T. & TAKESHI, K. (1998) Detection and genetical characterization of Shiga toxinproducing Escherichia coli from wild deer. Microbiol. Immunol. 42, 815-822
BERRY, E. D. & WELLS, J. E. (2010) Escherichia coli O157:H7: recent advances in research
on occurrence, transmission, and control in cattle and the production environment. Adv. Food
Nutr. Res. 60, 67-117
BETTELHEIM, K. A. (2007) The non-O157 shiga-toxigenic (verocytotoxigenic) Escherichia
coli; under-rated pathogens. Crit. Rev. Microbiol. 33, 67-87
CALLAWAY, T. R., EDRINGTON, T. S., LONERAGAN, G. H., CARR, M. & NISBET, D. J.
(2013) Shiga toxin-producing Escherichia coli (STEC) ecology in cattle and management based
options for reducing fecal shedding. Agric. Food Anal. Bacteriol. 3, 39-69
CHASE-TOPPING, M., GALLY, D., LOW, C., MATTHEWS, L. & WOOLHOUSE, M. (2008)
Super-shedding and the link between human infection and livestock carriage of Escherichia coli
O157. Nature reviews. Microbiology 6, 904-912
CHO, S., FOSSLER, C. P., DIEZ-GONZALEZ, F., WELLS, S. J., HEDBERG, C. W.,
KANEENE, J. B., RUEGG, P. L., WARNICK, L. D. & BENDER, J. B. (2013) Herd-level risk
factors associated with fecal shedding of Shiga toxin-encoding bacteria on dairy farms in
Minnesota, USA. Can. Vet. J. 54, 693-697
CRIM, S., IWAMOTO, M., HUANG, J., GRIFFIN, P., GILLISS, D., CRONQUIST, A.,
CARTTER, M., TOBIN-D'ANGELO, M., BLYTHE, D., SMITH, K., LATHROP, S., ZANSKY,
S., CIESLAK, P., DUNN, J., HOLT, K., LANCE, S., TAUXE, R. & HENAO, O. (2014)
Incidence and trends of infection with pathogens transmitted commonly through foodFoodborne diseases active surveillance network, 10 U.S. sites, 2006-2013. MMWR CDC 63
DUNN, J. (2003) The epidemiology of STEC O157:H7 in Louisiana dairy cattle, beef cattle, and
white-tailed deer In Pathobiological Sciences. Louisiana, Louisiana State University
FARROKH, C., JORDAN, K., AUVRAY, F., GLASS, K., OPPEGAARD, H., RAYNAUD, S.,
THEVENOT, D., CONDRON, R., DE REU, K., GOVARIS, A., HEGGUM, K.,
HEYNDRICKX, M., HUMMERJOHANN, J., LINDSAY, D., MISZCZYCHA, S.,
MOUSSIEGT, S., VERSTRAETE, K. & CERF, O. (2013) Review of Shiga-toxin-producing
6

Escherichia coli (STEC) and their significance in dairy production. Int. J. Food Microbiol. 162,
190-212
FERENS, W. A. & HOVDE, C. J. (2011) Escherichia coli O157:H7: animal reservoir and
sources of human infection. Foodborne Pathog Dis 8, 465-487
FRANZ, E. & VAN BRUGGEN, A. H. (2008) Ecology of E. coli O157:H7 and Salmonella
enterica in the primary vegetable production chain. Crit. Rev. Microbiol. 34, 143-161
GOODE, B., O'REILLY, C., DUNN, J., FULLERTON, K., SMITH, S., GHNEIM, G., KEEN,
J., DURSO, L., DAVIES, M. & MONTGOMERY, S. (2009) Outbreak of escherichia coli
O157: H7 infections after Petting Zoo visits, North Carolina State Fair, October-November 2004.
Arch. Pediatr. Adolesc. Med. 163, 42-48
GYLES, C. L. (2007) Shiga toxin-producing Escherichia coli: an overview. J. Anim. Sci. 85,
E45-62
HUSSEIN, H. S. (2007) Prevalence and pathogenicity of Shiga toxin-producing Escherichia coli
in beef cattle and their products. J. Anim. Sci. 85, E63-E72
JACEWICZ, M., CLAUSEN, H., NUDELMAN, E., DONOHUE-ROLFE, A. & KEUSCH, G.
T. (1986) Pathogenesis of shigella diarrhea. XI. Isolation of a shigella toxin-binding glycolipid
from rabbit jejunum and HeLa cells and its identification as globotriaosylceramide. J.E.M 163,
1391-1404
JAY, M. T., COOLEY, M., CARYCHAO, D., WISCOMB, G. W., SWEITZER, R. A.,
CRAWFORD-MIKSZA, L., FARRAR, J. A., LAU, D. K., O'CONNELL, J., MILLINGTON,
A., ASMUNDSON, R. V., ATWILL, E. R. & MANDRELL, R. E. (2007) Escherichia coli
O157:H7 in feral swine near spinach fields and cattle, central California coast. Emerg. Infect.
Dis. 13, 1908-1911
LEJEUNE, J. T. & WETZEL, A. N. (2007) Preharvest control of Escherichia coli O157 in
cattle. J. Anim. Sci. 85, E73-80
MENRATH, A., WIELER, L. H., HEIDEMANNS, K., SEMMLER, T., FRUTH, A. &
KEMPER, N. (2010) Shiga toxin producing Escherichia coli: identification of non-O157:H7super-shedding cows and related risk factors. Gut Pathogens 2

7

O'BRIEN, A. D. & HOLMES, R. K. (1987) Shiga and Shiga-like toxins. Microbiol. Rev. 51,
206-220
O'BRIEN, A. D., NEWLAND, J. W., MILLER, S. F., HOLMES, R. K., SMITH, H. W. &
FORMAL, S. B. (1984) Shiga-like toxin-converting phages from Escherichia coli strains that
cause hemorrhagic colitis or infantile diarrhea. Science (New York, N.Y.) 226, 694-696
PENNINGTON, H. (2010) Escherichia coli O157. Lancet 376, 1428-1435
RILEY, L. W., REMIS, R. S., HELGERSON, S. D., MCGEE, H. B., WELLS, J. G., DAVIS, B.
R., HEBERT, R. J., OLCOTT, E. S., JOHNSON, L. M., HARGRETT, N. T., BLAKE, P. A. &
COHEN, M. L. (1983) Hemorrhagic Colitis Associated with a Rare Escherichia coli Serotype. N.
Engl. J. Med. 308, 681-685
SAHL, J., MORRIS, C. & RASKO, D. (2013) Comparative genomics of pathogenic
Escherichia coli. In Escherichia coli: pathotypes and principles of pathogenesis. Second edn. Ed
M. DONNENBERG, Elsevier
SCHEUTZ, F., TEEL, L. D., BEUTIN, L., PIERARD, D., BUVENS, G., KARCH, H.,
MELLMANN, A., CAPRIOLI, A., TOZZOLI, R., MORABITO, S., STROCKBINE, N. A.,
MELTON-CELSA, A. R., SANCHEZ, M., PERSSON, S. & O'BRIEN, A. D. (2012)
Multicenter evaluation of a sequence-based protocol for subtyping Shiga toxins and
standardizing Stx nomenclature. J. Clin. Microbiol. 50, 2951-2963
SOLOMON, E. B., YARON, S. & MATTHEWS, K. R. (2002) Transmission of Escherichia coli
O157:H7 from Contaminated Manure and Irrigation Water to Lettuce Plant Tissue and Its
Subsequent Internalization. Appl. Environ. Microbiol. 68, 397-400
SOON, J. M., CHADD, S. A. & BAINES, R. N. (2011) Escherichia coli O157:H7 in beef cattle:
on farm contamination and pre-slaughter control methods. Anim. Health Res. Rev. 12, 197-211
TARR, P. I., GORDON, C. A. & CHANDLER, W. L. (2005) Shiga-toxin-producing
Escherichia coli and haemolytic uraemic syndrome. Lancet 365, 1073-1086
VANAJA, S., JANDHYALA, D., MALLICK, E., LEONG, J. & BALASUBRAMANIAN, S.
(2013) Enterohemorrhagic and other Shigatoxin-producing Escherichia coli. In Escherichia coli:
Pathotypes and Principles of Pathogenesis. 2nd edn. Ed M. DONNENBERG, Elsevier Inc.

8

CHAPTER 1

LITERATURE REVIEW: RISK FACTORS FOR SHEDDING OF SHIGA TOXINPRODUCING ESCHERICHIA COLI (STEC) IN DAIRY AND BEEF CATTLE

STEC
1. Escherichia coli
Escherichia coli are part of the normal intestinal microbiota and considered the major
facultative anaerobic bacterium in the intestinal tract of most mammalian species. E. coli are
Gram- negative, facultative anaerobe, rod-shaped bacteria belonging to the Enterobacteriaceae
family (Edwards and Ewing 1972; Escherich 1988; Gyles and Fairbrother 2010). Features used
for its identification include a positive indole reaction, negative tests for production of urease and
hydrogen sulfide, and failure to utilize citrate as the sole carbon source (Bettelheim 1994). E.
coli, which has its maximum concentration in the large intestine, is typically present at 107-109
organisms per gram in feces (Gyles and Fairbrother 2010). E. coli are frequently used as
indicator organisms for fecal contamination and breaches in hygiene in the areas of food safety
and public health (Farrokh, et al 2013).
E. coli are typically nonpathogenic but there is a small proportion of E. coli that has acquired
genes that enable them to cause intestinal and extraintestinal diseases in humans and animals
(Gyles 2007). Among these E. coli capable of causing disease are the pathotypes that cause
disease outside the intestines called extraintestinal E. coli (ExPEC). Examples from this group
are Uropathogenic E. coli (UPEC) and meningitis-associated E. coli (MNEC). Those E. coli that

9

cause enteric diseases are called diarrheagenic E. coli (DEC), among these pathotypes are
enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC) (traveler’s diarrhea), Vero
toxin-producing or Shiga toxin-producing E. coli (VTEC/STEC), enteroinvasive E. coli (EIEC),
enteroaggregative E. coli (EAEC), and diffusely adherent E. coli (DAEC) (Gyles 2007; Jafari, et
al 2012; Nataro and Kaper 1998). This literature review will focus on the specific pathotype
called Vero toxin-producing/Shiga toxin-producing E. coli (VTEC/STEC).

2. Shiga toxin-producing Escherichia coli (STEC)

STEC is also known as Vero toxin producing E. coli because of the cytopathic effect caused
when cultured on Vero cells (kidney epithelial cells from African green monkey) (Konowalchuk,
et al 1977; O'Brien and LaVeck 1983). STEC are characterized by their ability to produce at least
one type of Shiga toxin. The two major types of Stx that STEC can produce are Stx1 (almost
identical to Stx from Shigella dysenteria type 1, the prototype toxin for this family) and Stx2.
These two toxins have 55-60% genetic and amino acid identity homology (Jackson, et al 1987;
O'Brien, et al 1982; Stockbine, et al 1985; Strockbine, et al 1986). They are considered
genetically related but antigenically distinct, as there is no cross reaction with polyclonal
antisera. Biologically, they have similar cytotoxic, enterotoxic and lethal activities (Strockbine,
et al 1986). However, Stx1 and Stx2 cross the intestinal epithelial cell barrier by different
pathways (Hurley, et al 1999). Shiga toxins can be further broken down into subtypes based on
phenotypic differences, biological activity (serologic reactivity, receptor binding, capacity to be
activated by elastase in intestinal mucus) and hybridization properties (O'Brien, et al 1994;
Scheutz 2014). At present, 107 variants have been identified: 9 variants of Stx1a (including
10

Shiga toxin from S. dysenteriae), 4 of Stx1c, and 1 of Stx1d, and subtypes of Stx2 include 21
variants of Stx2a, 16 of Stx2b, 18 of Stx2c, 18 of Stx2d, 14 of Stx2e, 2 of Stx2f, and 4 of Stx2g
(Scheutz 2014; Scheutz, et al 2012) Some Stx2 subtypes are associated with severe symptoms in
humans, including bloody diarrhea and HUS (USDA 2012), especially the combination of eae
gene and stx2 (Persson, et al 2007). Stx2e is found almost exclusively in strains that cause
Edema Disease in pigs (Gyles and Fairbrother 2010; MacLeod and Gyles 1990; Marques, et al
1987).
In essence, Shiga toxins are holotoxins, proteins with an AB5 quaternary structure, which
means that they are composed of two subunits. The first subunit is called Shiga toxin A-subunit
(StxA) and it has enzymatic activity. The second is the five B-subunits (StxB) which bind
glycolipid cell surface receptors on the host cell surface(O'Brien and LaVeck 1983). The best
known receptor for Stx is the globotriaosylceramide (Gb3) cell surface receptor, a membrane
glycolipid of the globo-series (Jacewicz, et al 1986) located in endothelial cells, and found in
other cells (Meyers and Kaplan 2000). Once Stx and Gb3 are bound, the toxin is endocytosed and
trafficked retrograde through the Golgi apparatus to the endoplasmic reticulum. During this
process the A-subunit is proteolytically cleaved into A1 and A2 fragments. The free A1 fragment
is translocated to the cytosol, where its N-glycosidase activity cause depurination of the 28S
ribosomal RNA, resulting in cessation of protein synthesis, and leading to apoptosis (O'Brien and
Holmes 1987; Obrig, et al 1985; Vanaja, et al 2013).
Differentiation of E. coli is important for distinguishing pathogenic from nonpathogenic
types and for epidemiological investigations (Gyles and Fairbrother 2010). This differentiation
can be accomplished using phenotypic and genotypic methods. STEC can be classified by
serotyping. The serotype of an E. coli is based on the O (Ohne) antigen, which comprises the

11

polysaccharide portion of the cell wall lipopolysaccharide (LPS), and the H (Hauch) antigen,
which is found on the flagella protein (Edwards and Ewing 1972). There are more than 472
STEC/VTEC serotypes (Scheutz 2014). Serogroups are defined by O antigen only; there are
actually 174 different O antigens (Gyles and Fairbrother 2010). There are also STEC serotypes
that are nonmotile (NM) mutants of strains with an H antigen; that also can produce HUS (Gyles
2007; Gyles and Fairbrother 2010; Karch, et al 1993).
Karmali, et al. (2003) also proposed a seropathotype classification based on their reported
frequencies in human illness and their known association with outbreaks and severe outcomes.
Five seropathotype classifications have been proposed and include: seropathotype A associated
with the “highest” incidence in human disease, it consist of O157:H7 and O157: NM
(nonmotile), considered to be the most virulent. Follow by seropathotype B with a “moderate”
incidence, considered similar to seropathotype A in causing outbreaks and HUS but with a lower
frequency, it includes 13 STEC serotypes. Then seropathotype C includes serotypes infrequently
implicated in sporadic HUS but not typically with outbreaks, they are: O5:NM, O91:H21,
O104:H21, O113:H21, O121:NM and O165:H25; and seropathotype D is composed of 12
serotypes that have been implicated with sporadic cases of diarrhea but no with outbreaks of
HUS. Finally, seropathotype E included at least 14 serotypes not associated with human illness,
outbreaks or severe illness (Karmali 2003; Karmali, et al 2003; Scheutz 2014; USDA 2012). This
classification is problematic because the majority of STEC isolates are not fully serotyped nor
characterized for the presence of virulence factors. As consequence the European Food Safety
Authority Panel on Biological Hazards (BIOHAZ) concluded that this classification does not
define pathogenic VTEC nor does it provide an exhaustive list of pathogenic serotypes (Scheutz
2014).

12

STEC isolates can further be classified by the presence of putative virulence factors, for
example STEC can be classified into those that contain the Locus of Enterocyte Effacement
(LEE) pathogenicity island and those that do not (Sahl, et al 2013). The LEE encodes a type III
secretion system that injects effectors into the host cell that result in the formation of attaching
and effacing (AE) lesions (Gyles 2007; Sahl, et al 2013). Those STEC strains that are capable of
attaching to epithelial cells, effacing microvilli, and eliciting the formation of adhesion pedestals
composed of cytoskeletal proteins are called attaching and effacing E. coli (AEEC) or
enterohemorrhagic E. coli (EHEC). EHEC strains of the O157:H7 serotype are the most
important EHEC pathogens in North America, however not all O157:H7 are EHEC, some of
them lack eae/LEE and are only STEC (Kaper, et al 2004). AEEC strains that lack the
bacteriophage genes encoding Shiga toxins are classified as enteropathogenic E. coli
(EPEC)(Kaper 1996).
The first time STEC was recognized as a threat to public health was in 1982, when two
outbreaks of STEC O157:H7 (EHEC) were identified for the first time in the U.S.(Oregon and
Michigan ) (Riley, et al 1983). STEC O157:H7 was isolated from stool cultures and from
ground beef from a suspected lot of meat in Michigan (Riley, et al 1983). Since then, STEC
O157:H7 is the most commonly found of the STEC serotypes. Based on the molecular analysis
of these isolates , it was demonstrated that a lambdoid prophage, also known as bacteriophage,
transferred the genes required to produce Stx into E. coli, resulting in a newly emergent pathogen
(O'Brien, et al 1984). Phages regulate Stx production through amplification of gene copy
number, activity of phage gene promoters, and through release of Stx (Gyles 2007). Through
other horizontal gene transfer mechanism, STEC have acquired a variety of virulence factors,
such as enterotoxins and fimbriea or pili (Gyles and Fairbrother 2010; O'Brien, et al 1984; Sahl,

13

et al 2013). One of the most important virulence factors for EHEC is an outer-membrane protein
intimin, encoded by eae (E. coli attaching and effacing protein), which works as an adhesin
(Jerse, et al 1990; Moon, et al 1983). The process of attachment and interaction between
epithelial cells and eae-positive or eae-negative STEC is very different (Gyles 2007). Intimin
binds to the translocated intimin receptor (Tir), a type III-secreted effector that localizes in the
host plasma membrane after translocation into mammalian cells (Kenny, et al 1997). Intiminbinding to Tir induces a downstream signaling cascade that results in the formation of F-actin
pedestals, which promote colonization and Stx-mediated disease (Moon, et al 1983). In LEEnegative STEC, binding of STEC to the epithelium occurs in a non-intimate manner (Gyles and
Fairbrother 2010), for example, producing autoagglutinating adhesins encoded by the saa gene
(Bolton 2011; Vidal, et al 2008). LEE- negative STEC also contribute to Shiga toxin mediated
disease including HUS, for example, a study reported the cluster of three cases of HUS caused
by a STEC O113:H21 strain lacking the eae gene (Paton, et al 1999). Strain O113:H21 express a
newly identified cytotoxin, Subtilase-like toxin AB (SubAB), also an AB5 toxin (Paton, et al
2004). Other potential adherence factors, such as EibG, have been described in LEE-negative
STEC, although their significance for human disease is not as well established as for intimin (Lu,
et al 2006).
E. coli O157:H7’s lack of β-glucuronidase activity and its inability to ferment sorbitol
differentiate it from other E. coli strains (Vanaja, et al 2013). Specific biochemical characteristics
of E. coli O157:H7 allowed the development of several selective media (e.g. CHROMagar O157
and Rainbow agar) to identify and characterize this STEC strain (Bettelheim 2003). Early
screening for E. coli O157:H7 in human and animal feces was simplified by the development of
protocols specific for this STEC strain whereas the burden of illness from non-O157 STEC

14

strains was not fully recognized. The major problem in detecting non-O157 STEC is that beside
the production of Stx, they do not differ significantly in their biochemical characteristics from
typical commensal E. coli, leading to diagnostic limitations and inadequate surveillance
(Farrokh, et al 2013; Vanaja, et al 2013). There has been a steady increase in the number of
cases caused by STEC of serotypes other than O157 (Crim, et al 2014; Scallan, et al 2011). This
increase may at least partially be due to recent changes in laboratory practices by which nonO157 strains are more likely to be identified than they were in previous years (Gould, et al
2013). Currently, six non-O157 STEC serogroups (O26, O45, O103, O111, O121, and O145)
and O157:H7 have been classified as adulterants of beef in the U.S. (USDA 2011).
Consequently, rapid, accurate and reliable detection methods are necessary to test for non-O157
STEC in high risk food (Wang, et al 2013).

Importance of STEC in Public Health

STEC are one of the most relevant foodborne pathogens in U.S., Canada, and other
developed countries. STEC have been also reported in developing countries, however, the
proportion of morbidity and mortality caused by STEC in these countries is largely unknown.
STEC O157:H7 is the most common serotype isolated in U.S., whereas other non-O157 STEC
serotypes are more common in Australia, Germany and Austria (Tarr, et al 2005).
Malaysia, Thailand, Republic of Korea and China are some of the Asian countries where
STEC O157 have been reported. In 1996, one of the most largest STEC outbreaks occurred in
Japan with 9451 cases reported (Reilly 1998). E. coli O104:H4, however, caused an outbreak in
Germany that extended to other countries including the U.S. This outbreak caused more
15

problems in healthy adults than any other outbreak ever reported. This particular strain possessed
a combination of virulence genes from both EAEC and STEC (Frank, et al 2011).
The most common cause of acute renal failure in children worldwide is HUS resulted from a
gastrointestinal infection with STEC (Tarr, et al 2005). Tarr, et al. (2005) reported a 15% risk of
developing HUS in children younger than 10 years diagnosed with an E. coli O157:H7 infection.
The case fatality rate of patients with HUS is on average 2-7%, but some outbreaks targeting
elderly populations has resulted to 50% mortality (Reilly 1998). Unfortunately, the treatment for
HUS is supportive and deaths are usually associated with severe extra-renal complications
(Pennington 2010). Actually, the administration of antibiotics, antimotility agents or narcotics
during diarrheal episodes caused by STEC has been associated with an increased risk of
subsequent HUS (Tarr, et al 2005; Vanaja, et al 2013).
Studies have reported that patients with non-O157 infection were less likely to be
hospitalized than those with STEC O157 infection (Crim, et al 2014; Gould, et al 2013). In 2012,
among 496 serogrouped non-O157 STEC isolates, the most common serogroups were O26
(27%), O103 (23%), and O111 (15%) (Gillis, et al 2013). STEC O157:H7 has been reported to
cause the highest frequency of human infections, although there has been a steady increase in the
number of cases caused by STEC of serotypes other than O157 (non-O157 strains) (Crim, et al
2014; Gould, et al 2013; Scallan, et al 2011). This increase may partially be due to recent
changes in laboratory practices and diagnostics by which non-O157 strains are more likely to be
identified than they were in previous years (Gould, et al 2013).

16

1. Clinical disease in humans

In humans, STEC has a low infection dose of 10 to 100 organism, due largely to its ability to
resist highly acidic (pH 1.5-3.0) gastric environments (Menrath, et al 2010; Vanaja, et al 2013).
At least three acid resistance mechanisms have been identified in STEC O157:H7 that allow
these bacteria to survive in low pH environments. These mechanisms are a glutamate dependent
system, an acid-inducible arginine-dependent system, and oxidative systems (Audia, et al 2001;
Gyles and Fairbrother 2010).
The incubation period for STEC ranges between two to twelve days. Once established in the
intestinal tract, STEC leads to effacement of microvilli, inflammation and active chloride
secretion in the large intestine, resulting in watery diarrhea for one to three days followed by
hemorrhagic colitis in 90% of the cases. Other common signs are absence of fever and severe
abdominal pain (Bolton 2011; Pennington 2010; Tarr, et al 2005). Stx produced by STEC lead to
vascular damage and subsequent bloody diarrhea. Toxins are transported in the bloodstream to
sites rich in the Stx receptor Gb3, including the renal glomeruli, the renal proximal tubular
epithelium, and the brain (Bolton 2011; Gyles 2007). Stx2 is about 1,000 times more toxic to
human renal microvascular endothelial cells than is Stx1 (Gyles 2007).
HUS is characterized by hemolytic anemia, thrombocytopenia, and renal failure (Vanaja, et
al 2013). The release of chemokines, including IL-8 and other factors by the host, results in
platelet activation and subsequent renal thrombosis which is characteristic of HUS (Gyles 2007).
Glomerular capillaries are occluded by these thrombi resulting in ischemic damage to renal
endothelium (Vanaja, et al 2013). STEC infection in humans has also been associated with
neurological symptoms. There is evidence that Gb3 is present in neurological tissue making it
susceptible to the Stx.
17

2. STEC pathology in animals

In swine, STEC is the agent responsible for Edema Disease (ED). ED is the only animal
disease for which the role of Stx is clearly established (Gyles and Fairbrother 2010; Tseng, et al
2014). STEC with the Stx2e induces a toxemia that causes severe edema in specific sites in postweaning pigs and young finishing pigs; the most susceptible pigs to ED seems to be those with
the fastest growth (Gyles and Fairbrother 2010; Tseng, et al 2014). Pigs lacking intestinal
receptors for F18ab fimbriea are resistant to ED. The transportation of pigs and the mixing of
pigs from different sources have been mentioned as factors that predispose ED. A body of
research exists which led to the identification of the specific receptor for Stx2e called
globotetraosylceramide (Gb4), located in epithelial or vascular endothelial cells, due to the
impact that ED has had on the swine industry. However, Stx2e can also bind to Gb3. Stx2e
binding causes edema and hemorrhage and can present as sudden death without signs of illness.
Some affected pigs become inappetent, develop swelling of the eyelids and forehead and show
incoordination and respiratory distress (Gyles and Fairbrother 2010).
STEC has also been implicated in calves and lambs with diarrhea and dysentery. STEC
colonize the large intestine of calves and cause AE lesions similar to humans. Diarrhea may
result from loss of absorptive microvillus surface, activation of secretory activity in epithelial
cells, and loosening of tight junctions. Stx1 and/or Stx2 reach the blood system, presumably
causing bloody diarrhea, however no systemic signs are observed (Gyles and Fairbrother 2010).
STEC are present in the feces of healthy and diarrheal dogs. HUS occurs in about 5% of the
dogs that develop diarrhea caused by STEC. STEC has been implicated as a cause of a syndrome

18

called cutaneous and renal glomerular vasculopathy (CRGV) in racing greyhounds fed poorquality ground beef (Gyles and Fairbrother 2010).

STEC reservoirs and transmission

Transmission of STEC is a complex process that involves different reservoirs, different
hosts and different environments. To best understand transmission, we can look toward where
and how contamination or infection occurs. Ruminants, especially cattle, are the major STEC
reservoir. Humans most commonly get infected through the consumption of contaminated beef
products (especially ground beef) or unpasteurized milk and dairy products contaminated with
feces. However there have been outbreaks with other less common sources such as
unpasteurized apple juice, spinach and salami (Chase-Topping, et al 2008; Hussein 2007; Jay, et
al 2007; Menrath, et al 2010). Produce can also get contaminated with STEC through the
inclusion of manure in soil, which results in STEC uptake directly by plants, leading to human
infection (Callaway, et al 2013; Franz and van Bruggen 2008). Direct animal contact is another
source of transmission to humans. There have been STEC cases due to direct contact with
animals on farms, fairs, and petting zoos (Goode, et al 2009). Rainfall events can wash STEC
from cattle feces into drinking, recreation or irrigation water supplies, which have led to
infection in humans and other animals (Berry and Wells 2010; Callaway, et al 2013; Franz and
van Bruggen 2008; Hussein 2007; Solomon, et al 2002).
Besides cattle, other ruminants that carry STEC are deer (white-tailed deer, red deer),
sheep, and goats (Asakura, et al 1998; Callaway, et al 2013; Dunn 2003; Ferens and Hovde
2011). Other species reported to carry STEC at least transiently are pigs, rodents, and birds, such
19

as starlings, cowbirds, turkeys and egrets (Callaway, et al 2013; Cernicchiaro, et al 2012; Ferens
and Hovde 2011). EHEC O157 can be carried by amphibians, fish and invertebrates, and
mollusks, as well as insects such as flies (Berry and Wells 2010; Ferens and Hovde 2011;
Heuvelink, et al 1998). Studies have demonstrated that houseflies are not only mechanical
vectors, but E. coli O157:H7 can likely multiply in their gastrointestinal tract (Soon, et al 2011).
STEC can be transmitted from infected humans to uninfected humans through direct contact,
which is known as secondary spread. This has been most commonly reported in daycare and
elderly facilities (Gyles 2007; Pennington 2010; Tarr, et al 2005)
STEC isolates from cattle have an overall greater genetic diversity, fewer virulence
factors, but a greater tolerance for adverse conditions when compared to STEC isolates from
humans. STEC isolates associated with severe disease in humans are a minor fraction of the
strains found in cattle (Ferens and Hovde 2011). Hence, not all bovine strains are pathogenic to
humans and those that are pathogenic have particular characteristics that differentiate them and
could be used to develop specific diagnostic tests or control strategies.

Risk factors for STEC shedding by both dairy and beef cattle

The most frequent and consistently reported risk factor for STEC shedding for both beef and
dairy cattle production systems has been season of the year. The highest level of STEC shedding
occurs during warm months (Callaway, et al 2009; Callaway, et al 2013; Dunn, et al 2004;
Hancock, et al 1994; Heuvelink, et al 1998; Kondo, et al 2010; Menrath, et al 2010; Smith, et al
2013). For this reason, almost all studies include season in their study design as a cofounder in
their model and purposely programmed their sampling during the warm months (Cernicchiaro, et
20

al 2012; Cho, et al 2013; Cobbaut, et al 2009; Cobbold, et al 2004; Farrokh, et al 2013; Hussein
and Bollinger 2005; Menrath, et al 2010). Seasons not only represent a change in surface
temperature, but also season is proxy for other different things such as diet and management
practices. Changes in both the host and the environment are possible explanations for this
association. One possible reason for the seasonality of STEC shedding is the physiological
responses of cattle due to change in day length and heat stress. Another is adverse environmental
conditions, such as mud and higher temperatures that favor the growth of STEC outside the
animal (Berry and Wells 2010).
Age is also an important risk factor that has been identified for STEC shedding, although
there is not an agreement regarding which age group is at the highest risk (Cernicchiaro, et al
2009; Cho, et al 2013; Cobbaut, et al 2009; Farrokh, et al 2013; Hussein and Sakuma 2005;
Kuhnert, et al 2005). Cho, et al. (2009) reported that among all cattle, preweaned calves (calves
receiving milk or milk replacer) had a higher risk of STEC shedding than adult cows. However,
another study reported that calves aged 4 to 12 months had the highest STEC shedding rate
compared to all other age groups (Heuvelink, et al 1998). In this same study, cattle older than 3
years were more often found to be shedding STEC than cattle between 1 and 3 years of age
(Heuvelink, et al 1998). In adult dairy cows, one study reported a trend of higher shedding of
STEC in dairy cows with a parity of ≥ 4 than cows with less than 4 parities. (Cho, et al 2009). In
contrast, Menrath, et al. (2010) reported that first calf heifers were at higher risk than those cows
with ≥2 lactations (older age). Though there is no consensus regarding what age has the highest
risk for STEC, the majority of studies reported that younger animals are at higher risk.
Another risk factor for STEC shedding is contact with other infected species, such as pigs,
cats, dogs, rabbits, deer, birds, and pests, like flies and rodents (Berry and Wells 2010;

21

Cernicchiaro, et al 2009; Cho, et al 2013; Farrokh, et al 2013). Cernicchiaro, et al. (2012)
reported an increase STEC risk as the number of birds per milking cow increased. Similarly, the
use of mixed animal agriculture was also reported as a risk factor for STEC shedding in cattle
(Cernicchiaro, et al 2009).
Additional risk factors include introduction of new animals into groups/pens, animal density,
contact between adult cattle and calves (Cernicchiaro, et al 2012), size/number of farm/cattle
(Cho, et al 2013; Herbert, et al 2014), transportation and lairage, and stressful situations.
Animal density is especially important for hide contamination, which becomes a significant
risk factor for meat contamination at slaughter (Cernicchiaro, et al 2009; Herbert, et al 2014).
Cho, et al. (2013), reported that STEC shedding was more common in small dairy herds than in
large herds (≥100 cows). Bringing new cattle into the herd or recent animal movements have also
been found to increase the risk of STEC shedding at the herd level (Farrokh, et al 2013; Herbert,
et al 2014). For instance, the number of times cattle were taken to a livestock exhibition in the
previous 12 months was a risk factor for STEC O157 in cow-calf operations (Cernicchiaro, et al
2009). Transportation and lairage are also risk events for transmission or contamination among
animals. Direct or indirect transmission among animals can occur during transportation; plus the
resulting stressful situation that transportation represent to animals can add to the risk of
colonization and shedding (Callaway, et al 2013).
The presence of super-shedders is reported as a STEC risk factor in several studies for both
STEC O157 and non-O157 STEC (Berry and Wells 2010; Callaway, et al 2013; Chase-Topping,
et al 2008; Chase-Topping, et al 2007; Menrath, et al 2010). A super-shedder is “… an animal
that excretes >104 cfu per gram of feces or the simple identification of outlying counts” (Chase-

22

Topping, et al 2008). The presence of a super-shedder increases STEC transmission rate, as
statistical models have determined (Chase-Topping, et al 2008).
The cleanliness of bedding has been reported as a risk factor for STEC shedding. Herds with
clean and dry bedding, as well as with frequent change of bedding, have been associated with a
decreased risk of STEC shedding. Also, inorganic bedding, such as sand, has been shown to be
less favorable for coliform replication in general, thus decreasing risk of STEC colonization and
shedding. Frequent cleaning of the pen surface can also influence the risk of STEC transmission
and survival, as it may slow spread within a herd, although it will not completely eliminate
STEC (Callaway, et al 2013).
The exact effect of stress on STEC colonization or shedding is still unclear, but increased
stress has frequently been reported as a STEC risk factor (Berry and Wells 2010; ChaseTopping, et al 2007; Cho, et al 2013; Farrokh, et al 2013) . Besides transportation, other stressful
situations to cattle are weaning, calving, heat stress, handling, loading and unloading, changes in
climatic conditions, food and water deprivation (Callaway, et al 2013; Rostagno 2009). One
possible explanation for stress contributing to STEC shedding is that the central nervous system
and the enteric nervous system have an established communication. Thus stress can lead to the
release of hormones into the intestinal tract, that can alter the interactions between the microbiota
and the endothelial cells facilitating the infection of the intestinal tract by pathogenic microbiota
(Rostagno 2009). Studies have demonstrated that norepinephrine can influence the production of
Stx by E. coli O157:H7 and its adhesion to the cecal epithelium in cattle (Lyte, et al 1996;
Rostagno 2009).
Cattle are often subject to fasting before or during transportation to slaughter. Fasting has
been reported to increase the risk of STEC shedding at slaughter facilities while others reported

23

no effect (Callaway, et al 2009; Callaway, et al 2013; Rostagno 2009). The mechanism linked
with STEC and fasting is thought to be by decreasing short chain volatile fatty acids (VFA) and
increasing pH in the gastrointestinal tract (Callaway, et al 2013).
Feed, usually called diet and its components, are another important risk factor for STEC
shedding. There is a large body of research regarding STEC and diet, which has shown that diet
does affect E. coli O157 populations, but the magnitude and impact of diet or its components
haven’t always presented consistent results (Callaway, et al 2009). Example of diet components
that affect E. coli O157 shedding are percentage of forage and rapidly ruminally fermented
grains, among others (Callaway, et al 2009; Cernicchiaro, et al 2009). Some studies reported that
barley-based diets increase E. coli O157 shedding due to pH increased in feces. Among other
feed types linked with an increased risk for STEC shedding are corn silage (Cernicchiaro, et al
2009), distillers grain, brewers grain and wet corn gluten (Callaway, et al 2009; Callaway, et al
2013), while whole cottonseed has been linked with a decreased risk in STEC shedding
(Callaway, et al 2013). However a recent study found no association between STEC and distiller
grains (Fink, et al 2013). How feeds are processed may also have an effect on STEC colonization
and shedding. For example, steam-flaked corn has been associated with an increase risk of STEC
when compared to dry-rolled corn (Callaway, et al 2009; Callaway, et al 2013).
Studies have reported that grain-fed cattle shed more E. coli O157 than forage-fed cattle
(Callaway, et al 2013). In addition, there are conflicting results when looking at whether the
switch from a grain base to a pasture grain base diet decreased or not E. coli O157 shedding.
Some studies report a significant association towards E. coli O157 reduction when the switch
was made while others reported no association (Callaway, et al 2009; Callaway, et al 2013;
Stanford, et al 2005). One possible explanation for the differences in conclusions among studies

24

is the use of different diets. It is believed that diversity in quality and components of the forage
could be the factors responsible for the different results between studies (Callaway, et al 2009;
Callaway, et al 2013).
Ionophores (ex. monensin and lasalocid) are included in the diet to inhibit gram-positive
bacteria and promote feed efficiency. There are studies that indicated ionophores increase E. coli
O157 shedding, while another indicated a decrease, and there are even other studies that reported
zero effect (Callaway, et al 2009). For example, Cho, et al. (2013) reported that the use of
monensin for weaned calves, and the use of decoquinate (a quinolone derivative) for preweaned
calves decrease the risk of STEC shedding. The lack of consistent results about the effect of
ionophores in STEC makes necessary more research the find the right answer. Also this lack of
consistency in results exposed the importance of more consistent methodology and study design
among future studies.
In addition to the presence of other species, the type of cattle production system factors into
the risk of STEC shedding. When differences in the level of risk for STEC shedding between
dairy and beef farms have been reported, dairy farms usually present higher levels of STEC
shedding (Cobbaut, et al 2009; Cobbold, et al 2004). Similarly, the type of cattle, more
specifically, the raising of female cattle for breeding, increased the risk of STEC shedding
compared to raising of cattle for beef (Chase-Topping, et al 2007).
There are risk factors that have been reported exclusively for the dairy or beef production
system. Among the risk factors for dairy is stage of lactation. According to several studies, the
risk of STEC shedding is higher in lactating cows than in dry cows (Cho, et al 2009; Dunn, et al
2004; Fitzgerald, et al 2003; Mechie, et al 1997). Another group in dairy production at higher
risk for STEC shedding is cull cows; cull cows are those cows selected to go to slaughter. In a

25

study by Cho, et al (2009), cows that were scheduled to be culled were more likely to be
shedding STEC than those not scheduled to be culled. Menrath, et al. (2010), reported several
risk factors for detection of STEC in dairy cattle feces in Germany. They found that cows with a
somatic cell count lower than 100,000 cells/ml in milk, milk protein content higher than 3.0%
and a body condition score higher than 3.50 had significantly or tendency towards increased risk
of shedding STEC; while cows with blood urea content lower than 150 mg/L milk had a
decreased risk. These measures in milk are related to the diet, health and stress of the cow. So
they could be taken as a proxy for these other factors that have been reported to influence STEC
shedding.
Several studies reported their findings about dairy herd management practices and its
association with STEC shedding. The use of total mixed ration (TMR) for lactating dairy cows
was reported to increase STEC shedding (Cho, et al 2013). The use of manure piles for manure
storage was also reported to increase STEC shedding, and the use of three or more different
ventilation systems (ex. doors, fans, curtains) on the farm also increased STEC shedding
(Cernicchiaro, et al 2012). Garber (1999) reported a higher risk for STEC shedding in those
herds that use flushed water to remove manure compared with other methods of manure
removing. Hancock, et al. (1994), reported an increase risk for STEC presence in cattle when
owners apply slurry to pasture.
Some risk factors reported specifically for beef cattle (feedlot and/or cow-calf operations)
deal with parturition and weaning as events that increase the risk of STEC shedding in cows and
calves respectively (Gannon, et al 2002). In addition, Sargeant, et al (2003) described a positive
relationship between the water tank’s sediment and the water in those water tanks being STEC
positive as well as the cattle who drink that water in feedlots, with capacities >1000 heads, being

26

STEC positive. Also Smith, et al. (2005) reported the recovery of E. coli O157:H7 from water
tanks as a risk for STEC positive cattle. The use of corn silage supplementation in winter (silage
preparation) is another herd management practice reported to increase the risk of STEC shedding
(Cernicchiaro, et al 2009).

Intervention strategies at the pre-harvest level

Since establishing cattle as the main reservoir of STEC, control measures have been
developed and implemented during the pre and post-harvest periods to reduce the risk of beef
contamination and subsequent human infection. These measures have helped to reduce the
number of STEC cases in humans and the public health burden (Gillis, et al 2013).
Several studies have concluded that control measures at the pre-harvest level will have the
highest impact in the reduction of STEC infections (LeJeune and Wetzel 2007; Soon, et al 2011).
Callaway, et al. (2013), summarized the reasons very clearly and concisely “… 1) reducing the
amount of pathogens entering processing plants will reduce the burden on the plants and render
the in-plant interventions more effective; 2) reducing horizontal pathogen spread from infected
animals (especially in “super-shedders”) in transport and lairage; 3) will reduce the pathogenic
bacteria burden in the environment and wastewater streams; and 4) will reduce the direct risk to
those in direct contact with animals via petting zoos, open farms, rodeos and to animal workers”.
LeJeune and Wetzel (2007) grouped the pre-harvest interventions into 3 categories “1) exposure
reduction strategies; 2) exclusion strategies and 3) direct antipathogen strategies”.

27

1. Exposure reduction strategies
1.1. Environmental exposure
Avoiding muddy feedlot pens and providing dry bedding helps with the reduction of STEC
as well as preventing fecal-oral infection or re-infection (Soon, et al 2011). The treatment of
manure with carbonate and alkali has also been demonstrated to inactivate E. coli in cattle
manure (Berry and Wells 2010; LeJeune and Wetzel 2007). Some plant essential oils added to
cattle waste such as carvacrol, eugenol and thymol have been reported to reduce or eliminate E.
coli (Berry and Wells 2010; Doyle and Erickson 2012; Varel and Miller 2004). It is also
important to apply hygienic practices during transportation (Doyle and Erickson 2012).
1.2. Wildlife exclusion
Although cattle are the main reservoir for STEC, contamination of feed and water with fecal
material from wildlife could introduce into cattle new STEC strains, through the fecal-oral route,
so avoiding access of wildlife from the farm should be attempted as much as possible (Soon, et al
2011).

2. Exclusion strategies
2.1 Probiotics
Probiotics are defined as “a preparation of a product containing viable, defined
microorganisms in sufficient numbers, which alter the microflora in a compartment of the host
and that exert beneficial health effects in this host” (Schrezenmeir and de Vrese 2001). The
probiotics are also called direct fed microbials (LeJeune and Wetzel 2007). An important
probiotic that has been frequently reported capable to reduce the shedding of STEC is
Lactobacillus acidophilus. The specific strain that appears to be most efficacious and already
28

available on the market is NP51. Some point out that the selection of the right strain is very
important for successful reduction of STEC (Cull, et al 2012; LeJeune and Wetzel 2007;
Loneragan and Brashears 2005; Sargeant, et al 2007; Soon, et al 2011; Stephens, et al 2007).
Some other effective probiotics reviewed include, either individually or in combinations,
Enterococcus (Streptococcus) faecium, L. casei, L. fermentum, L.gallinarum, L. platarum,
Propionibacterium freudenreichii, and Streptococcus bovis (Berry and Wells 2010).
2.2 Prebiotics
Prebiotics are defined as “organic compounds such as fructo-oligosaccharides, inulin and
galacto-oligosaccharides that are unavailable to, or indigestible by, the host animal, but are
digestible by specific bacterial species” (LeJeune and Wetzel 2007; Soon, et al 2011). When
probiotics and prebiotics are administered together it is called Synbiotics (Doyle and Erickson
2012).
2.3 Other diet supplements
There are studies that reported an inhibitory effect of a additive product from brown seaweed
(Ascophyllum nodosum) on E. coli O157:H7 (Bach, et al 2008; Braden, et al 2004) but more
information is required to confirm this event, as some mentioned brown seaweed is not an
efficacious intervention (Loneragan and Brashears 2005).

29

3. Direct antipathogen strategies
3.1 Antimicrobial Compounds
Studies have demonstrated a reduction in STEC shedding after cattle received oral neomycin
sulfate, an aminoglyoside antibiotic. However, there are concerns regarding this practice due to
the fear of developing antibacterial resistance in humans (Berry and Wells 2010; LeJeune and
Wetzel 2007; Loneragan and Brashears 2005). Another negative side is that supplementation of
milk replacer with Neomycin may increase E. coli O157:H7 shedding in very young calves
(Berry and Wells 2010). Reports also claim STEC shedding reduction with the use of sodium
chlorate, whether via feed or water in cattle. The application of chlorate for this use is pending
U.S. Food and Drug Administration review and approval (Anderson, et al 2005; Berry and Wells
2010; LeJeune and Wetzel 2007; Loneragan and Brashears 2005; Sargeant, et al 2007).
3.2 Bacteriophage Therapy
Studies have reported the use of bacteriophage (virus of bacteria) as an effective therapy to
decrease E. coli O157:H7 in cattle and or other ruminants (Sheng, et al 2006). One advantage of
phages is their narrow target spectra, specifically in this case STEC (Soon, et al 2011).
Bacteriophages have been administered orally through water or feed and directly to the rectoanal junction (RAJ) (Berry and Wells 2010; Sheng, et al 2006). Some concerns regarding the use
of bacteriophages are the development of phage resistance and the possibility of genetic
materials being transferred to bacterial hosts (Soon, et al 2011).
3.3 Vaccination
Several studies have showed that cattle vaccination decreases shedding of E. coli O157 and
there are even results that indicate that cattle vaccination is considered the most effective
measure to reduce human exposure to E. coli O157 (Smith, et al 2013). Even with prices
30

between $2.29 and $ 9.14 USD, vaccination can be a cost effective intervention measure (Smith,
et al 2013). Loneragan, et al. (2005), discussed the potential advantages for the use of vaccine
that include: “1) cattle producers are familiar with administration of vaccines; 2) incorporation
into existing management of cattle would be fairly simple; 3) vaccines could be used in all
sectors of the industry”.
One vaccine has been developed against E. coli O157:H7 type III secreted proteins (Bioniche
Life Sciences, Inc., Belleville, Ontario, Canada) (Berry and Wells 2010). Type III secreted
proteins are critical for E. coli O157:H7 intestinal colonization in cattle. This vaccine is fully
licensed for use in Canada (Berry and Wells 2010). The vaccine license is conditional for the
U.S. market (Vande Walle, et al 2013). Another vaccine targeting siderophore receptor and porin
proteins (SRP) (Epitopix, LLC, Wilmar, MN) currently has a conditional license for use in cattle
in U.S. (Berry and Wells 2010). The SRP is a cell membrane receptor that is essential for iron
transport into cells. This vaccine produces antibodies which bind the E. coli O157:H7 SRP thus
essentially starving the cells of iron, leading to their eventual death (Cull, et al 2012). There are
studies recommending a three-dose regimen (Berry and Wells 2010; Vande Walle, et al 2013)
while others used a two-dose regimen (Cull, et al 2012). A new technical approach for E. coli
O157:H7 vaccine design is the use of bacterial ghosts (BGs), as inactivated whole-cell envelope
vaccines. BGs are empty bacterial cell envelopes, which display all surface components,
including colonization factors in a non-denatured form and are able to induce a strong mucosal
immune response (Vande Walle, et al 2013). There is still the necessity to develop more studies
to explore the efficacy of this BGs vaccine.

31

Conclusions
Historically E. coli O157 has been the “star” strain in the group of serotypes belonging to
STEC because of the frequency of reported outbreaks caused by E. coli O157. But this does not
mean that the other non-O157 STEC bacteria are a less important threat to public health. This
may be just a reflection of the lack of laboratory techniques to detect these other bacteria. As a
result, scientists have been working on the development and improvement of isolation and
detection methods for non-O157 STEC. The effort to improve non-O157 STEC detection has led
to more frequent detection of both STEC O157 and non-O157 STEC. For example, the
immunomagnetic separation (IMS) assay is a sensitive method that was design to detect E. coli
O157. This same methodology is now being adapted to non-O157 serotypes thus leading to
improved and more reliable detection.
Over the years, many different risk factors have been identified in association with STEC
shedding. Some studies have found similar factors while others have found contradictory risk
factors. It is also important to discuss the difficulty to compare results between studies, due to the
differences among laboratory methods and study design (Cobbaut, et al 2009; Sargeant, et al
2003). In careful evaluation of the literature, it is evident there is a diversity in the methodology
applied by the different researches to detect STEC. This makes it harder and sometimes
impossible to compare results between studies. For example, the association between several risk
factors, such as bedding type or house type and STEC shedding in cattle hasn’t been determined
yet, because the results among the available studies cannot be compared due to the differences
among laboratory methods. Regarding study design, the lack of uniformity in study design,
sampling strategies and animal premises analyzed (Cho, et al 2013; Cobbaut, et al 2009;
Menrath, et al 2010; Sargeant, et al 2003) makes it difficult to derive definite, well supported and
32

consistent conclusions (Sargeant, et al 2003). Therefore the different STEC research groups
should be more consistent with the methodology so results can be compared.
Risk factors that affect different production systems have been identified including
season and age. This is advantageous for the development and application of intervention
strategies aimed to control or prevent the transmission among animals and also to humans.
Although the identification of common risk factors is relevant, it will be equally beneficial and
probably easier to target risk factors specific for each production system. In this way,
intervention strategies designed for each production type could be implemented.

33

REFERENCES

34

REFERENCES

ANDERSON, R. C., HARVEY, R. B., BYRD, J. A., CALLAWAY, T. R., GENOVESE, K. J.,
EDRINGTON, T. S., JUNG, Y. S., MCREYNOLDS, J. L. & NISBET, D. J. (2005) Novel
preharvest strategies involving the use of experimental chlorate preparations and nitro-based
compounds to prevent colonization of food-producing animals by foodborne pathogens. Poult.
Sci. 84, 649-654
ASAKURA, H., MAKINO, S., SHIRAHATA, T., TSUKAMOTO, T., KURAZONO, H.,
IKEDA, T. & TAKESHI, K. (1998) Detection and genetical characterization of Shiga toxinproducing Escherichia coli from wild deer. Microbiol. Immunol. 42, 815-822
AUDIA, J. P., WEBB, C. C. & FOSTER, J. W. (2001) Breaking through the acid barrier: An
orchestrated response to proton stress by enteric bacteria. Int. J. Med. Microbiol. 291, 97-106
BACH, S. J., WANG, Y. & MCALLISTER, T. A. (2008) Effect of feeding sun-dried seaweed
(Ascophyllum nodosum) on fecal shedding of Escherichia coli O157:H7 by feedlot cattle and on
growth performance of lambs. Anim. Feed Sci. Technol. 142, 17-32
BERRY, E. D. & WELLS, J. E. (2010) Escherichia coli O157:H7: recent advances in research
on occurrence, transmission, and control in cattle and the production environment. Adv. Food
Nutr. Res. 60, 67-117
BETTELHEIM, K. A. (1994) Biochemical characteristics of Escherichia coli. In Escherichia
coli In Domestic Animals and Humans. Ed C. L. GYLES. Wallingford, UK, CAB International
BETTELHEIM, K. A. (2003) Non-O157 verotoxin-producing Escherichia coli: a problem,
paradox, and paradigm. Exp Biol Med (Maywood) 228, 333-344
BOLTON, D. J. (2011) Verocytotoxigenic (Shiga toxin-producing) Escherichia coli: virulence
factors and pathogenicity in the farm to fork paradigm. Foodborne Pathog Dis 8, 357-365
BRADEN, K., BLANTON, J. J., ALLEN, V., POND, K. & MILLAR, M. F. (2004)
Ascophyllum nodosum supplementation: a preharvest intervention for reducing Escherichia coli
O157:H7 and Salmonella spp. in feedlots steers. J. Food Prot. 67, 1824-1828

35

CALLAWAY, T. R., CARR, M. A., EDRINGTON, T. S., ANDERSON, R. C. & NISBET, D.
J. (2009) Diet, Escherichia coli O157:H7, and cattle: a review after 10 years. Curr Issues Mol
Biol 11, 67-79
CALLAWAY, T. R., EDRINGTON, T. S., LONERAGAN, G. H., CARR, M. & NISBET, D. J.
(2013) Shiga toxin-producing Escherichia coli (STEC) ecology in cattle and management based
options for reducing fecal shedding. Agric. Food Anal. Bacteriol. 3, 39-69
CERNICCHIARO, N., PEARL, D. L., GHIMIRE, S., GYLES, C. L., JOHNSON, R. P.,
LEJEUNE, J. T., ZIEBELL, K. & MCEWEN, S. A. (2009) Risk factors associated with
Escherichia coli O157:H7 in Ontario beef cow-calf operations. Prev. Vet. Med. 92, 106-115
CERNICCHIARO, N., PEARL, D. L., MCEWEN, S. A., HARPSTER, L., HOMAN, H. J.,
LINZ, G. M. & LEJEUNE, J. T. (2012) Association of wild bird density and farm management
factors with the prevalence of E. coli O157 in dairy herds in Ohio (2007-2009). Zoonoses Public
Health 59, 320-329
CHASE-TOPPING, M., GALLY, D., LOW, C., MATTHEWS, L. & WOOLHOUSE, M. (2008)
Super-shedding and the link between human infection and livestock carriage of Escherichia coli
O157. Nature reviews. Microbiology 6, 904-912
CHASE-TOPPING, M. E., MCKENDRICK, I. J., PEARCE, M. C., MACDONALD, P.,
MATTHEWS, L., HALLIDAY, J., ALLISON, L., FENLON, D., LOW, J. C., GUNN, G. &
WOOLHOUSE, M. E. J. (2007) Risk Factors for the Presence of High-Level Shedders of
Escherichia coli O157 on Scottish Farms. J. Clin. Microbiol. 45, 1594-1603
CHO, S., FOSSLER, C. P., DIEZ-GONZALEZ, F., WELLS, S. J., HEDBERG, C. W.,
KANEENE, J. B., RUEGG, P. L., WARNICK, L. D. & BENDER, J. B. (2009) Cattle-level risk
factors associated with fecal shedding of Shiga toxin-encoding bacteria on dairy farms,
Minnesota, USA. Can. J. Vet. Res.-Rev. Can. Rech. Vet. 73, 151-156
CHO, S., FOSSLER, C. P., DIEZ-GONZALEZ, F., WELLS, S. J., HEDBERG, C. W.,
KANEENE, J. B., RUEGG, P. L., WARNICK, L. D. & BENDER, J. B. (2013) Herd-level risk
factors associated with fecal shedding of Shiga toxin-encoding bacteria on dairy farms in
Minnesota, USA. Can. Vet. J. 54, 693-697
COBBAUT, K., BERKVENS, D., HOUF, K., DE DEKEN, R. & DE ZUTTER, L. (2009)
Escherichia coli O157 prevalence in different cattle farm types and identification of potential risk
factors. J. Food Prot. 72, 1848-1853

36

COBBOLD, R. N., RICE, D. H., SZYMANSKI, M., CALL, D. R. & HANCOCK, D. D. (2004)
Comparison of Shiga-Toxigenic Escherichia coli Prevalences among Dairy, Feedlot, and CowCalf Herds in Washington State. Appl. Environ. Microbiol. 70, 4375-4378
CRIM, S., IWAMOTO, M., HUANG, J., GRIFFIN, P., GILLISS, D., CRONQUIST, A.,
CARTTER, M., TOBIN-D'ANGELO, M., BLYTHE, D., SMITH, K., LATHROP, S., ZANSKY,
S., CIESLAK, P., DUNN, J., HOLT, K., LANCE, S., TAUXE, R. & HENAO, O. (2014)
Incidence and trends of infection with pathogens transmitted commonly through foodFoodborne diseases active surveillance network, 10 U.S. sites, 2006-2013. MMWR CDC 63
CULL, C. A., PADDOCK, Z. D., NAGARAJA, T. G., BELLO, N. M. & BABCOCK, A. H.
(2012) Efficacy of a vaccine and a direct-fed microbial against fecal shedding of Escherichia coli
O157:H7 in a randomized pen-level field trial of commercial feedlot cattle. Vaccine 30, 62106215
DOYLE, M. P. & ERICKSON, M. C. (2012) Opportunities for mitigating pathogen
contamination during on-farm food production. Int. J. Food Microbiol. 152, 54-74
DUNN, J. (2003) The epidemiology of STEC O157:H7 in Louisiana dairy cattle, beef cattle, and
white-tailed deer In Pathobiological Sciences. Louisiana, Louisiana State University
DUNN, J. R., KEEN, J. E. & THOMPSON, R. A. (2004) Prevalence of shiga-toxigenic
Escherichia coli O157:H7 in adult dairy cattle. J. Am. Vet. Med. Assoc. 224, 1151-1158
EDWARDS, P. & EWING, W. (1972) Identification of Enterobacteriaceae. Minneapolis, Minn,
Burgess Publishing Co.
ESCHERICH, T. (1988) The intestinal bacteria of the neonate and breast-fed infant. Review of
Infectious Diseases 10, 1220-1225
FARROKH, C., JORDAN, K., AUVRAY, F., GLASS, K., OPPEGAARD, H., RAYNAUD, S.,
THEVENOT, D., CONDRON, R., DE REU, K., GOVARIS, A., HEGGUM, K.,
HEYNDRICKX, M., HUMMERJOHANN, J., LINDSAY, D., MISZCZYCHA, S.,
MOUSSIEGT, S., VERSTRAETE, K. & CERF, O. (2013) Review of Shiga-toxin-producing
Escherichia coli (STEC) and their significance in dairy production. Int. J. Food Microbiol. 162,
190-212
FERENS, W. A. & HOVDE, C. J. (2011) Escherichia coli O157:H7: animal reservoir and
sources of human infection. Foodborne Pathog Dis 8, 465-487
37

FINK, R. C., POPOWSKI, J. M., ANDERSON, J. E., DAHLBERG, J. L., KALYANIKUTTY,
S., CRAWFORD, G. I., DICOSTANZO, A., COX, R. B. & DIEZ-GONZALEZ, F. (2013)
Impact of Management Practices and Distillers' Grains Feeding on the Prevalence of Escherichia
coli O157 in Feedlot Cattle in Minnesota. Foodborne Pathog Dis 10, 559-565
FITZGERALD, A. C., EDRINGTON, T. S., LOOPER, M. L., CALLAWAY, T. R.,
GENOVESE, K. J., BISCHOFF, K. M., MCREYNOLDS, J. L., THOMAS, J. D., ANDERSON,
R. C. & NISBET, D. J. (2003) Antimicrobial susceptibility and factors affecting the shedding of
E. coli O157:H7 and Salmonella in dairy cattle. Lett. Appl. Microbiol. 37, 392-398
FRANK, C., WERBER, D., CRAMER, J. P., ASKAR, M., FABER, M., AN DER HEIDEN, M.,
BERNARD, H., FRUTH, A., PRAGER, R., SPODE, A., WADL, M., ZOUFALY, A., JORDAN,
S., KEMPER, M. J., FOLLIN, P., MULLER, L., KING, L. A., ROSNER, B., BUCHHOLZ, U.,
STARK, K., KRAUSE, G. & TEAM, H. U. S. I. (2011) Epidemic profile of Shiga-toxinproducing Escherichia coli O104:H4 outbreak in Germany. N. Engl. J. Med. 365, 1771-1780
FRANZ, E. & VAN BRUGGEN, A. H. (2008) Ecology of E. coli O157:H7 and Salmonella
enterica in the primary vegetable production chain. Crit. Rev. Microbiol. 34, 143-161
GANNON, V. P., GRAHAM, T. A., KING, R., MICHEL, P., READ, S., ZIEBELL, K. &
JOHNSON, R. P. (2002) Escherichia coli O157:H7 infection in cows and calves in a beef cattle
herd in Alberta, Canada. Epidemiol. Infect. 129, 163-172
GARBER, L. (1999) Factors associated with fecal shedding of verotoxin-producing Escherichia
coli O157 on dairy farms. J. Food Prot 62, 307-312
GILLIS, D., CRONQUIST, A., CARTTER, M., TOBIN-D'ANGELO, M., BLYTHE, D.,
SMITH, K., ZANSKY, S., CIESLAK, P., DUNN, J., HOLT, K., LANCE, S., CRIM, S.,
HENAO, O., PATRICK, M., GRIFFIN, P. M. & TAUXE, R. (2013) Incidence and trends of
infection with pathogens transmitted commonly through food-Foodborne diseases active
surveillance network, 10 U.S. sites, 1996-2012. In MMWR, CDC. pp 283-287
GOODE, B., O'REILLY, C., DUNN, J., FULLERTON, K., SMITH, S., GHNEIM, G., KEEN,
J., DURSO, L., DAVIES, M. & MONTGOMERY, S. (2009) Outbreak of escherichia coli
O157: H7 infections after Petting Zoo visits, North Carolina State Fair, October-November 2004.
Arch. Pediatr. Adolesc. Med. 163, 42-48
GOULD, L. H., MODY, R. K., ONG, K. L., CLOGHER, P., CRONQUIST, A. B., GARMAN,
K. N., LATHROP, S., MEDUS, C., SPINA, N. L., WEBB, T. H., WHITE, P. L., WYMORE, K.,
GIERKE, R. E., MAHON, B. E. & GRIFFIN, P. M. (2013) Increased recognition of non-O157
Shiga toxin-producing Escherichia coli infections in the United States during 2000-2010:
38

epidemiologic features and comparison with E. coli O157 infections. Foodborne Pathog. Dis. 10,
453-460
GYLES, C. L. (2007) Shiga toxin-producing Escherichia coli: an overview. J. Anim. Sci. 85,
E45-62
GYLES, C. L. & FAIRBROTHER, J. M. (2010) Escherichia coli. In Pathogenesis of Bacterial
Infections in Animals. Eds C. L. GYLES, J. F. PRESCOTT, J. G. SONGER, C. O. THOEN,
Blackwell Publishing
HANCOCK, D. D., BESSER, T. E., KINSEL, M. L., TARR, P. I., RICE, D. H. & PAROS, M.
G. (1994) The Prevalence of Escherichia coli O157.H7 in Dairy and Beef Cattle in Washington
State. Epidemiol. Infect. 113, 199-207
HERBERT, L., VALI, L., HOYLE, D., INNOCENT, G., MCKENDRICK, I., PEARCE, M.,
MELLOR, D., PORPHYRE, T., LOCKING, M., ALLISON, L., HANSON, M., MATTHEWS,
L., GUNN, G., WOOLHOUSE, M. & CHASE-TOPPING, M. (2014) E. coli O157 on Scottish
cattle farms: Evidence of local spread and persistence using repeat cross-sectional data. BMC
Vet. Res. 10, 95
HEUVELINK, A. E., VAN DEN BIGGELAAR, F. L. A. M., ZWARTKRUIS-NAHUIS, J. T.
M., HERBES, R. G., HUYBEN, R., NAGELKERKE, N., MELCHERS, W. J. G., MONNENS,
L. A. H. & DE BOER, E. (1998) Occurrence of Verocytotoxin-ProducingEscherichia coli O157
on Dutch Dairy Farms. J. Clin. Microbiol. 36, 3480-3487
HURLEY, B. P., JACEWICZ, M., THORPE, C. M., LINCICOME, L. L., KING, A. J.,
KEUSCH, G. T. & ACHESON, D. W. K. (1999) Shiga Toxins 1 and 2 Translocate Differently
across Polarized Intestinal Epithelial Cells. Infect. Immun. 67, 6670-6677
HUSSEIN, H. S. (2007) Prevalence and pathogenicity of Shiga toxin-producing Escherichia coli
in beef cattle and their products. J. Anim. Sci. 85, E63-E72
HUSSEIN, H. S. & BOLLINGER, L. M. (2005) Prevalence of Shiga toxin-producing
Escherichia coli in beef cattle. J. Food Prot. 68, 2224-2241
HUSSEIN, H. S. & SAKUMA, T. (2005) Invited Review: Prevalence of Shiga Toxin-Producing
Escherichia coli in Dairy Cattle and Their Products. J. Dairy Sci. 88, 450-465

39

JACEWICZ, M., CLAUSEN, H., NUDELMAN, E., DONOHUE-ROLFE, A. & KEUSCH, G.
T. (1986) Pathogenesis of shigella diarrhea. XI. Isolation of a shigella toxin-binding glycolipid
from rabbit jejunum and HeLa cells and its identification as globotriaosylceramide. J.E.M 163,
1391-1404
JACKSON, M. P., NEILL, R. J., O'BRIEN, A. D., HOLMES, R. K. & NEWLAND, J. W.
(1987) Nucleotide sequence analysis and comparison of the structural genes for Shiga-like toxin
I and Shiga-like toxin II encoded by bacteriophages from Escherichia coli 933. FEMS Microbiol.
Lett. 44, 109-114
JAFARI, A., ASLANI, M. & BOUZARI, S. (2012) Escherichia coli: a brief review of
diarrheagenic pathotypes and their role in diarrheal diseases in Iran. Iran J. Microbiol. 4, 102-117
JAY, M. T., COOLEY, M., CARYCHAO, D., WISCOMB, G. W., SWEITZER, R. A.,
CRAWFORD-MIKSZA, L., FARRAR, J. A., LAU, D. K., O'CONNELL, J., MILLINGTON,
A., ASMUNDSON, R. V., ATWILL, E. R. & MANDRELL, R. E. (2007) Escherichia coli
O157:H7 in feral swine near spinach fields and cattle, central California coast. Emerg. Infect.
Dis. 13, 1908-1911
JERSE, A. E., YU, J., TALL, B. D. & KAPER, J. B. (1990) A genetic locus of enteropathogenic
Escherichia coli necessary for the production of attaching and effacing lesions on tissue culture
cells. Proc. Natl. Acad. Sci. U. S. A. 87, 7839-7843
KAPER, J. B. (1996) Defining EPEC. Rev Microbiol Sao Paulo 27, 130:133
KAPER, J. B., NATARO, J. P. & MOBLEY, H. L. T. (2004) Pathogenic Escherichia coli. Nat
Rev Micro 2, 123-140
KARCH, H., BÖHM, H., SCHMIDT, H., GUNZER, F., ALEKSIC, S. & HEESEMANN, J.
(1993) Clonal structure and pathogenicity of Shiga-like toxin-producing, sorbitol-fermenting
Escherichia coli O157:H. J. Clin. Microbiol. 31, 1200-1205
KARMALI, M. A. (2003) The medical significance of Shiga toxin-producing Escherichia coli
infections. An overview. Methods Mol Med 73, 1-7
KARMALI, M. A., MASCARENHAS, M., SHEN, S., ZIEBELL, K., JOHNSON, S., REIDSMITH, R., ISAAC-RENTON, J., CLARK, C., RAHN, K. & KAPER, J. B. (2003) Association
of Genomic O Island 122 of Escherichia coli EDL 933 with Verocytotoxin-Producing
Escherichia coli Seropathotypes That Are Linked to Epidemic and/or Serious Disease. J. Clin.
Microbiol. 41, 4930-4940
40

KENNY, B., DEVINNEY, R., STEIN, M., REINSCHEID, D. J., FREY, E. A. & FINLAY, B.
B. (1997) Enteropathogenic E. coli (EPEC) Transfers Its Receptor for Intimate Adherence into
Mammalian Cells. Cell 91, 511-520
KONDO, S., HOAR, B. R., VILLANUEVA, V., MANDRELL, R. E. & ATWILL, E. R. (2010)
Longitudinal prevalence and molecular typing of Escherichia coli O157:H7 by use of multiplelocus variable-number tandem-repeat analysis and pulsed-field gel electrophoresis in fecal
samples collected from a range-based herd of beef cattle in California. Am. J. Vet. Res. 71,
1339-1347
KONOWALCHUK, J., SPEIRS, J. I. & STAVRIC, S. (1977) Vero response to a cytotoxin of
Escherichia coli. Infect. Immun. 18, 775-779
KUHNERT, P., DUBOSSON, C. R., ROESCH, M., HOMFELD, E., DOHERR, M. G. &
BLUM, J. W. (2005) Prevalence and risk-factor analysis of Shiga toxigenic Escherichia coli in
faecal samples of organically and conventionally farmed dairy cattle. Vet. Microbiol. 109, 37-45
LEJEUNE, J. T. & WETZEL, A. N. (2007) Preharvest control of Escherichia coli O157 in
cattle. J. Anim. Sci. 85, E73-80
LONERAGAN, G. H. & BRASHEARS, M. M. (2005) Pre-harvest interventions to reduce
carriage of E. coli O157 by harvest-ready feedlot cattle. Meat science 71, 72-78
LU, Y., IYODA, S., SATOU, H., SATOU, H., ITOH, K., SAITOH, T. & WATANABE, H.
(2006) A New Immunoglobulin-Binding Protein, EibG, Is Responsible for the Chain-Like
Adhesion Phenotype of Locus of Enterocyte Effacement-Negative, Shiga Toxin-Producing
Escherichia coli. Infect. Immun. 74, 5747-5755
LYTE, M., ARULANANDAM, B. & FRANK, C. (1996) Production of Shiga-like toxins by
Escherichia coli O157:H7 can be influenced by the neuroendocrine hormone norepinephrine. J.
Lab. Clin. Med. 128, 392-398
MACLEOD, D. L. & GYLES, C. L. (1990) Purification and characterization of an Escherichia
coli Shiga-like toxin II variant. Infect. Immun. 58, 1232-1239
MARQUES, L. R. M., PEIRIS, J. S. M., CRYZ, S. J. & O'BRIEN, A. D. (1987) Escherichia
coli strains isolated from pigs with edema disease produce a variant of Shiga-like toxin II. FEMS
Microbiol. Lett. 44, 33-38

41

MECHIE, S. C., CHAPMAN, P. A. & SIDDONS, C. A. (1997) A Fifteen Month Study of
Escherichia coli O157:H7 in a Dairy Herd. Epidemiol. Infect. 118, 17-25
MENRATH, A., WIELER, L. H., HEIDEMANNS, K., SEMMLER, T., FRUTH, A. &
KEMPER, N. (2010) Shiga toxin producing Escherichia coli: identification of non-O157:H7super-shedding cows and related risk factors. Gut Pathogens 2
MEYERS, K. E. C. & KAPLAN, B. S. (2000) Many cell types are Shiga toxin targets. Kidney
Int. 57, 2650-2651
MOON, H. W., WHIPP, S. C., ARGENZIO, R. A., LEVINE, M. M. & GIANNELLA, R. A.
(1983) Attaching and effacing activities of rabbit and human enteropathogenic Escherichia coli
in pig and rabbit intestines. Infect. Immun. 41, 1340-1351
NATARO, J. P. & KAPER, J. B. (1998) Diarrheagenic Escherichia coli. Clin. Microbiol. Rev.
11, 142-201
O'BRIEN, A. D. & HOLMES, R. K. (1987) Shiga and Shiga-like toxins. Microbiol. Rev. 51,
206-220
O'BRIEN, A. D., KARMALI, M. A. & SCOTLAND, S. (1994) A proposal for rationalization of
the Escherichia coli cytotoxins. In Recent Advances in verocytotoxin-producing Escherichia coli
infections. Amsterdam, Netherlands, Elsevier Science, B.V. pp 147-149
O'BRIEN, A. D. & LAVECK, G. D. (1983) Purification and characterization of a Shigella
dysenteriae 1-like toxin produced by Escherichia coli. Infect. Immun. 40, 675-683
O'BRIEN, A. D., LAVECK, G. D., THOMPSON, M. R. & FORMAL, S. B. (1982) Production
of Shigella dysenteriae Type 1-Like Cytotoxin by Escherichia coli. J. Infect. Dis. 146, 763-769
O'BRIEN, A. D., NEWLAND, J. W., MILLER, S. F., HOLMES, R. K., SMITH, H. W. &
FORMAL, S. B. (1984) Shiga-like toxin-converting phages from Escherichia coli strains that
cause hemorrhagic colitis or infantile diarrhea. Science (New York, N.Y.) 226, 694-696
OBRIG, T., MORAN, T. & COLINAS, R. (1985) Ribonuclease activity associated with the 60S
ribosome-inactivating proteins rich A, phytolaccin and Shiga toxin. Biochem. Biophys. Res.
Commun 130, 879-884

42

PATON, A. W., SRIMANOTE, P., TALBOT, U. M., WANG, H. & PATON, J. C. (2004) A
New Family of Potent AB(5) Cytotoxins Produced by Shiga Toxigenic Escherichia coli. J.E.M
200, 35-46
PATON, A. W., WOODROW, M. C., DOYLE, R. M., LANSER, J. A. & PATON, J. C. (1999)
Molecular Characterization of a Shiga Toxigenic Escherichia coli O113:H21 Strain Lacking eae
Responsible for a Cluster of Cases of Hemolytic-Uremic Syndrome. J. Clin. Microbiol. 37, 33573361
PENNINGTON, H. (2010) Escherichia coli O157. Lancet 376, 1428-1435
PERSSON, S., OLSEN, K. E. P., ETHELBERG, S. & SCHEUTZ, F. (2007) Subtyping Method
for Escherichia coli Shiga Toxin (Verocytotoxin) 2 Variants and Correlations to Clinical
Manifestations. J. Clin. Microbiol. 45, 2020-2024
REILLY, A. (1998) Prevention and control of enterohaemorrhagic Escherichia coli (EHEC)
infections: memorandum from a WHO meeting. WHO Consultation on Prevention and Control
of Enterohaemorrhagic Escherichia coli (EHEC) Infections. Bull. World Health Organ. 76, 245255
RILEY, L. W., REMIS, R. S., HELGERSON, S. D., MCGEE, H. B., WELLS, J. G., DAVIS, B.
R., HEBERT, R. J., OLCOTT, E. S., JOHNSON, L. M., HARGRETT, N. T., BLAKE, P. A. &
COHEN, M. L. (1983) Hemorrhagic Colitis Associated with a Rare Escherichia coli Serotype. N.
Engl. J. Med. 308, 681-685
ROSTAGNO, M. H. (2009) Can Stress in Farm Animals Increase Food Safety Risk? Foodborne
Pathog. Dis. 6, 767-776
SAHL, J., MORRIS, C. & RASKO, D. (2013) Comparative genomics of pathogenic
Escherichia coli. In Escherichia coli: pathotypes and principles of pathogenesis. Second edn. Ed
M. DONNENBERG, Elsevier
SARGEANT, J. M., AMEZCUA, M. R., RAJIC, A. & WADDELL, L. (2007) Pre-harvest
interventions to reduce the shedding of E. coli O157 in the faeces of weaned domestic ruminants:
a systematic review. Zoonoses Public Health 54, 260-277
SARGEANT, J. M., SANDERSON, M. W., SMITH, R. A. & GRIFFIN, D. D. (2003)
Escherichia coli O157 in feedlot cattle feces and water in four major feeder-cattle states in the
USA. Prev. Vet. Med. 61, 127-135
43

SCALLAN, E., HOEKSTRA, R. M., ANGULO, F. J., TAUXE, R. V., WIDDOWSON, M. A.,
ROY, S. L., JONES, J. L. & GRIFFIN, P. M. (2011) Foodborne illness acquired in the United
States--major pathogens. Emerg. Infect. Dis. 17, 7-15
SCHEUTZ, F. (2014) Taxonomy Meets Public Health: The Case of Shiga Toxin-Producing
Escherichia coli. Microbiology Spectrum 2
SCHEUTZ, F., TEEL, L. D., BEUTIN, L., PIERARD, D., BUVENS, G., KARCH, H.,
MELLMANN, A., CAPRIOLI, A., TOZZOLI, R., MORABITO, S., STROCKBINE, N. A.,
MELTON-CELSA, A. R., SANCHEZ, M., PERSSON, S. & O'BRIEN, A. D. (2012)
Multicenter evaluation of a sequence-based protocol for subtyping Shiga toxins and
standardizing Stx nomenclature. J. Clin. Microbiol. 50, 2951-2963
SCHREZENMEIR, J. & DE VRESE, M. (2001) Probiotics, prebiotics, and synbiotics—
approaching a definition. Am. J. Clin. Nutr. 73, 361s-364s
SHENG, H., KNECHT, H. J., KUDVA, I. T. & HOVDE, C. J. (2006) Application of
Bacteriophages To Control Intestinal Escherichia coli O157:H7 Levels in Ruminants. Appl.
Environ. Microbiol. 72, 5359-5366
SMITH, D. R., MOXLEY, R. A., CLOWSER, S. L., FOLMER, J. D., HINKLEY, S.,
ERICKSON, G. E. & KLOPFENSTEIN, T. J. (2005) Use of rope devices to describe and
explain the feedlot ecology of Escherichia coli O157:H7 by time and place. Foodborne Pathog
Dis 2, 50-60
SMITH, G. G., GOEBEL, S. E., CULBERT, C. R. & GUILBAULT, L. A. (2013) Reducing the
public health risk of Escherichia coli O157 exposure by immunization of cattle. Can. J. Public
Health. 104, e9-e11
SOLOMON, E. B., YARON, S. & MATTHEWS, K. R. (2002) Transmission of Escherichia coli
O157:H7 from Contaminated Manure and Irrigation Water to Lettuce Plant Tissue and Its
Subsequent Internalization. Appl. Environ. Microbiol. 68, 397-400
SOON, J. M., CHADD, S. A. & BAINES, R. N. (2011) Escherichia coli O157:H7 in beef cattle:
on farm contamination and pre-slaughter control methods. Anim. Health Res. Rev. 12, 197-211
STANFORD, K., BACH, S., MARX, T., JONES, S., HANSEN, J., WALLINS, G.,
ZAHIRODDINI, H. & MCALLISTER, T. A. (2005) Monitoring Eschericia coli O157:H7 in
inoculated and naturally colonized feedlot cattle and their environment. J. Food Prot. 68, 26-33
44

STEPHENS, T. P., LONERAGAN, G. H., KARUNASENA, E. & BRASHEARS, M. M. (2007)
Reduction of Escherichia coli O157 and Salmonella in feces and on hides of feedlot cattle using
various doses of a direct-fed microbial. J. Food Prot. 70, 2386-2391
STOCKBINE, N., MARQUES, L., HOLMES, R. K. & O'BRIEN, A. D. (1985) Characterization
of monoclonal antibodies against Shiga-like toxin from Escherichia coli. Infect. Immun. 50, 695700
STROCKBINE, N. A., MARQUES, L. R., NEWLAND, J. W., SMITH, H. W., HOLMES, R. K.
& O'BRIEN, A. D. (1986) Two toxin-converting phages from Escherichia coli O157:H7 strain
933 encode antigenically distinct toxins with similar biologic activities. Infect. Immun. 53, 135140
TARR, P. I., GORDON, C. A. & CHANDLER, W. L. (2005) Shiga-toxin-producing
Escherichia coli and haemolytic uraemic syndrome. Lancet 365, 1073-1086
TSENG, M., FRATAMICO, P. M., MANNING, S. D. & FUNK, J. A. (2014) Shiga toxinproducing Escherichia coli in swine: the public health perspective. Anim. Health Res. Rev. 15,
63-75
USDA (2011) USDA takes new steps to fight E. coli, protect the food supply. Ed F. S. A. I.
SERVICE. United States Department of Agriculture website, United States Department of
Agriculture
USDA (2012) Risk profile for pathogenic Non-O157 shiga toxin-producing E coli (non-O157
STEC). Ed U. FOOD SAFETY AND INSPECTION SERVICE
VANAJA, S., JANDHYALA, D., MALLICK, E., LEONG, J. & BALASUBRAMANIAN, S.
(2013) Enterohemorrhagic and other Shigatoxin-producing Escherichia coli. In Escherichia coli:
Pathotypes and Principles of Pathogenesis. 2nd edn. Ed M. DONNENBERG, Elsevier Inc.
VANDE WALLE, K., VANROMPAY, D. & COX, E. (2013) Bovine innate and adaptive
immune responses against Escherichia coli O157:H7 and vaccination strategies to reduce faecal
shedding in ruminants. Vet. Immunol. Immunopathol. 152, 109-120
VAREL, V. H. & MILLER, D. L. (2004) Eugenol stimulates lactate accumulation yet inhibits
volatile fatty acid production and eliminates coliform bacteria in cattle and swine waste. J. Appl.
Microbiol. 97, 1001-1005

45

VIDAL, M., PRADO, V. & WHITLOCK, G. (2008) Subtractive hybridisation and identification
of putative adhesions in a shiga toxin-producing eae-negative Escherichia coli. Microbiology
2008, 3639-3648
WANG, F., YANG, Q., KASE, J. A., MENG, J., CLOTILDE, L. M., LIN, A. & GE, B. (2013)
Current trends in detecting non-O157 Shiga toxin-producing Escherichia coli in food. Foodborne
Pathog. Dis. 10, 665-677

46

CHAPTER 2

RISK FACTORS FOR SHIGA TOXIN-PRODUCING ESCHERICHIA COLI (STEC)
SHEDDING IN CATTLE

Abstract

Shiga toxin-producing Escherichia coli (STEC) are one of the most important foodborne
pathogens in the U.S. and other developed countries. STEC can cause hemorrhagic diarrhea, and
sometimes hemolytic uremic syndrome (HUS). STEC is defined by the presence of genes
encoding the Shiga toxin (Stx), of which Stx1 and Stx2 are the major types, but additional
subtypes have also been described. Cattle are the primary reservoir for STEC, and food or water
contaminated with cattle feces is the most common source of infections in humans.
The purpose of our study was to identify risk factors for STEC shedding in cattle. During the
summers of 2011 and 2012, a cross-sectional study was performed on 1,096 cattle in 5 dairy
herds and 6 beef herds. A fecal sample from each animal was enriched in E. coli broth (EC) and
plated on selective media. In addition, a portion of the broth was subjected to immunomagnetic
separation targeting E. coli O157, and multiplex PCR was used to detect the presence of stx1,
stx2 and the gene encoding intimin (eaeA). STEC prevalence was 21% (80/378) in beef cattle,
which was significantly higher than the 13% (95/718) in dairy cattle (OR: 1.76; 95% CI: 1.252.47). A multivariable model, with herd included as a random effect, was used to evaluate both
herd-level and cow-level risk factors for dairy cattle. Dairy cattle were more likely to shed STEC
when the average temperature was > 84°F 1-5 days before sampling (OR: 2.5; 95% CI: 1.254.91). Dairy cows were more likely to shed STEC in their first lactation (OR: 1.8; 95% CI: 1.147

2.8) and when they were < 31 days in milk (OR: 3.9; 95% CI: 2.1-7.2). Descriptive
epidemiologic studies such as this one will hopefully foster hypothesis-testing and intervention
strategies aimed at mitigating STEC shedding in cattle, thereby reducing the risk of human
infections.

48

Introduction

Shiga toxin-producing Escherichia coli (STEC) is one of the most virulent and
pathogenic foodborne pathogens in both developed and developing countries (Reilly 1998).
STEC can cause hemorrhagic diarrhea and sometimes hemolytic uremic syndrome (HUS) that
can lead to kidney failure and death, particularly in young children (Vanaja, et al 2013). STEC
belonging to serotype O157:H7 has been reported to cause the highest frequency of human
infections, although there has been a steady increase in the number of cases caused by STEC of
serotypes other than O157 (non-O157 STEC) (Crim, et al 2014; Scallan, et al 2011). This
increase may at least partially be due to recent changes in laboratory diagnostic practices by
which non-O157 strains are more likely to be identified than they were in previous years (Gould,
et al 2013). The incidence of U.S. reported non-O157 STEC cases increased from 0.12 per
100,000 population in 2000 to 0.95 per 100,000 in 2010, while the incidence of STEC O157
decreased from 2.17 per 100,000 in 2000 to 0.95 per 100,000 in 2010 (Gould, et al 2013). In year
2013, there were 561 cases (1.17 per 100, 000 people) of non-O157 and 552 (1.15 cases per
100,000) for STEC O157. In this same year there were 78 hospitalizations and 2 deaths
associated with non-O157 and 210 hospitalizations and 2 deaths associated with STEC O157
(Crim, et al 2014). These findings support the Gould et al, (2013) report that patients with nonO157 STEC infection were less likely to be hospitalized than those with O157.
STEC is defined by the presence of genes encoding Shiga toxins (Stx), which are carried
on a bacteriophage (O'Brien, et al 1984). The two major Stx types are Stx1 and Stx2, but
additional subtypes (e.g., Stx2c-2g) have also been described (Scheutz, et al 2012). The eaeA
gene, which is present on the LEE island and encodes for the intimin protein, allows STEC to
intimately adhere to the intestinal mucosa (Fagan, et al 1999; McDaniel, et al 1995). All STEC
49

strains have at least one stx subtype, though the locus of enterocyte effacement (LEE)
pathogenicity island may be variably present. STEC strains with the LEE island and stx are
referred to as enterohemorrhagic E. coli (EHEC), while STEC refers to stx-positive strains that
lack the LEE island. EHEC typically causes more severe clinical symptoms in humans relative to
STEC (Beutin, et al 2007; Reilly 1998), though the 2011 STEC O104:H4 outbreak in Germany
that contributed to over 50 deaths (Frank, et al 2011) is an exception.
Cattle are the primary reservoir for STEC, and food or water contaminated with cattle
feces is the most common source of infection for humans (Kuhnert, et al 2005).Other sources of
STEC infection include direct contact with domestic animals, such as swine, dogs and cats, and
wildlife including wild-white-tailed deer (Asakura, et al 1998; Beutin, et al 1993; Rounds, et al
2012).
STEC prevalence has been shown to vary across food animal production systems in the
U.S and other countries. For example, the prevalence of STEC O157 infections was 45%, 19%
and 8% in cow-calf operations in Ontario, feedlots in Scotland, and dairy cattle in Washington
respectively (Cernicchiaro, et al 2009; Chase-Topping, et al 2007; Hancock, et al 1994).
Additionally, worldwide the prevalence of non-O157 STEC reported in feedlots and beef cattle
on pasture ranged between 4.6% to 55.9% and 4.7% to 44.8%, respectively (Hussein 2007).
Factors associated with low or high herd prevalence estimates, however, are not fully
understood. It is therefore important to determine which production systems represent the
greatest risk of STEC infection for the efficient implementation of pre-harvest and post-harvest
intervention strategies.
Several prior studies have reported a higher prevalence of STEC shedding in pasture
cattle and dairy farms than in feedlots (Cobbold, et al 2004), while others have found differences

50

attributable to geographic location. Studies in Sweden and Korea, for instance, have reported
significant regional differences. Positive cattle samples appeared to be concentrated in the
southern and central parts of Sweden (Kistemann, et al 2004), while in Korea, the region of
Gyeonggi and Gangwon, had higher prevalence rates than other parts of the country (Kang, et al
2014). Other studies have not observed differences across region (Sargeant, et al 2003).
Although numerous studies have sought to determine the prevalence of STEC in animal
reservoirs and varying geographic locations, additional studies are still needed to better
understand the risk factors associated with STEC shedding at both the herd and animal level.
Similarly, more research is needed to identify which groups of cattle have the highest risk of
STEC colonization and shedding as these groups represent the best targets for pre-harvest
intervention strategies.
To date, few consistent risk factors have been identified for STEC O157 shedding in
cattle across studies (Cho, et al 2013; Menrath, et al 2010). This lack of consistent risk factors is
even more dramatic for non-O157 STEC, due to the scarcity of research studies (Menrath, et al
2010). For STEC O157, several risk factors including season, herd management practices
(manure removing), age, level of animal-to-animal contact, stress and diet have been suggested
to be important (Cernicchiaro, et al 2009; Cho, et al 2013; Dunn, et al 2004; Garber 1999).
However, most research has been focused on STEC O157 rather than non-O157.
To guide STEC shedding prevention strategies, additional large-scale studies are needed
to better understand the transmission dynamics of STEC within and across herds with varying
management practices. Here, we conducted a cross-sectional study of 1,096 animals from five
dairy and six beef herds during the summers of 2011 and 2012. Our goal was to identify factors
important for STEC shedding throughout Mid-Michigan. The identification of risk factors for

51

STEC shedding in cattle could aid in the improvement of intervention practices aimed at
reducing the level of STEC entering the human food supply.

52

Materials and methods

1. Study design and herd selection

A convenience sample of dairy farms and beef feedlots were contacted and selected for
inclusion in the study based on the availability of good records, proximity to East Lansing
Michigan, adequate animal handling facilities and willingness to participate in all phases of the
study. Eleven of twelve herds contacted agreed to participate. One herd chose not to participate
because of concerns regarding animal welfare. The farm owners provided written informed
consent to participate in the study, and each received a monetary incentive following study
completion. This study was approved by the Michigan State University Institutional Animal Care
and Use Committee (AN12/10-223-00) and this study was supported by the USDA NIFA Grant
#2011-67005-30004.
Phase I involved completing a questionnaire designed to collect demographic information
and data related to potential STEC risk factors. Phase II focused on sampling a representative
number of animals within each herd and culturing for STEC. Herds were visited and sampled
between May 11th and August 16th of 2011 (n=5) or between May 21st and August 27th of 2012
(n=6). Season was gauged based on the day of the equinoxes and solstices indicated on the
Gregorian calendar.

53

1.1 Questionnaire
Two questionnaires were designed; one for dairy farms and another for beef feedlots.
Both were pre-tested on the managers of representative farms. Questionnaires were administered
to the farm owners or managers during a face-to-face interview at the first visit. The same person
administered the questionnaires for all 11 farms. The questionnaires consisted of both closed and
open-ended questions addressing farm demographics, animal movements, farm management
practices, and herd health management strategies (Table 2.1 and Figure 2.4).

1.2 Sampling
The number of cattle sampled per herd was based on the type of herd and number of
cattle. In dairy herds with fewer than 175 animals, all adult cattle were sampled. In dairy herds
with greater than 175 animals, a convenience sample of 175 cattle was selected with
representation from each management group. In the beef feedlot herds, we selected cohorts of
cattle within the feedlot that were managed as one unit and then sampled all cattle within that
cohort. A summary of herd demographics and animals sampled can be found in Table 2.2.
Fresh fecal samples collected by rectal palpation using individual obstetrical sleeves,
were placed in plastic bags (Whirl pak). Samples from the first four herds (496 animals) were
transported to the laboratory on ice where they were stored at 4°C and then processed within 48
hours. The remaining seven herd’s samples were transported to the laboratory in a cooler without
ice and processed within 8 hours. This change in protocol was made because a prior study found
that ice storage decreased the likelihood of STEC recovery from feces [Mindy Brashears,
personal communication].
The date, time, latitude and longitude were recorded for each farm sampled. In addition,
54

the maximum, minimum and average temperatures from the day of sampling and the preceding
five days were recorded using data from the closest weather station (Quality Controlled Local
Climatological Data (NOAA)).

2. Laboratory protocol for STEC detection and isolation

Five grams of feces were inoculated in 2X EC broth (Oxoid Ltd.; Waltham, MA)
supplemented with novobiocin (8mg/l), rifampin (2mg/1) and potassium tellurite (1mg/1) for 24
hours at 42°C (Jason, et al 2009) followed by subculture on STEC CHROMagar™
(CHROMagar, Paris, France) and sorbitol MacConkey (SMAC) agar. A portion of the EC
culture was also processed by immunomagnetic separation using Dynabeads® (Invitrogen
Corporation, California, USA) specific for E .coli O157 followed by subculture to O157
CHROMagar (CHROMagar, Paris, France) and sorbitol MacConkey (SMAC) agar. Up to 20
presumptive STEC single colonies were selected from each plate, inoculated into Luria-Bertani
(LB) broth for growth overnight at 37°C, and confirmed by PCR using a previously described
protocol (Tarr, et al 2002) with either the Taq 2x MeanGreen Master Mix or Kappa2G Multiplex
Master Mix (Kapa Biosystems, Massachusetts). The multiplex PCR used to confirm STEC single
colonies detects the presence of stx1, stx2 and eaeA (intimin). Individual colonies with at least
one stx gene were considered to be STEC, and fecal samples from individual cattle were
considered positive if at least one STEC isolate was recovered.

55

3. Data analyses

The data was analyzed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA). The
dependent variable used in the analysis was the positive or negative STEC status of the animal.
An animal was considered positive when STEC could be recovered by culture. The analysis
performed in this study was based on the STEC results at the animal level and not at the isolate
level. There was an average of four isolates per STEC positive animal, with a range of one to 19
isolates.
The distribution of the independent variables was analyzed. Those independent variables
with non-normal distributions were transformed into binary or categorical variables based on
their average or quartiles. All the categorical variables are based on answers provided by the
farmers. The average on the temperatures variables were calculated based on the data collected
from the weather stations. For the univariate analysis the independent variables were analyzed in
groups, those groups were herd characteristics, housing, cleaning, herd treatment, diet, contact
with other animals, and environmental conditions. In Tables 2.4 and 2.6 the independent
variables analyzed were displayed. See Table 2.7 for key of the variables names.
Variables with potential confounding were identified. The univariate analysis was used to
identify variables to be included in a multivariable model, using a backward manual selection
procedure. The point of significance was P < 0.15 for inclusion into the multivariable model,
however, the point of significance for the final multivariable model was P < 0.05 (Dohoo, et al
2010). Herd was always included as a random effect in the univariate analyses. Additional
models were constructed within the groups allowing a correlation between the variables up to 0.9
(Dohoo, et al 2010). Odds ratios (ORs) and their 95% confidence intervals (95% CI) were
estimated for each variable in both the univariable and multivariable analyses. Year and season
56

variables were unique to each herd and were therefore removed from the analysis because each
herd was only sampled once.
Separate univariable models were constructed to analyze the dairy and beef data using
logistic regression and generalized linear mixed models (GLMM). The data structure was
different for beef and dairy herds and beef cattle had no cattle-specific independent variables.
Because three of the five beef herds were raised at different times at the same location, their
herd-level risk factors were mostly identical or correlated, thereby preventing construction of a
valid multivariable model.
For the dairy data, a base “full” model was created to include all variables that were
significantly associated with STEC-positivity in the univariate analyses, and then the final
multivariable model was created through a backward manual selection process. Additional
variables were evaluated with the base model, depending on their relationships to other variables,
as well as, biological plausibility. For example, variables that examined factors associated with
housing such as access to pasture or use of free stalls, were evaluated in this way.

57

Results

1. Descriptive statistics

A total of 1,108 animals were sampled during the course of this study; 724 (65%) were
dairy cattle and 384 (35%) were beef cattle. Six beef and six dairy cattle were excluded from the
analysis due to missing STEC laboratory results, leaving a total of 1,096 individual cattle in the
final analysis. Notably, STEC was detected in cattle from all 11 herds. The animal level
prevalence ranged between 6.4% and 53.7% (Figure 2.1) with an average of 16%. Among the
378 beef cattle sampled, 80 (21.2%) were positive for STEC as were 95 of the 718 (13.2%) dairy
cattle tested. STEC prevalence was significantly different between dairy and beef cattle (P <
0.0007) with beef cattle being 1.8 times more likely to be STEC positive than dairy cattle (95%
CI: 1.25- 2.47). The overall STEC prevalence was 10% for 2011 and 23% for 2012 (P < 0.0001)
and the overall STEC prevalence was 18% for spring and 15% for summer (P < 0.1888).
Among the 522 STEC isolates recovered from all 175 STEC-positives animals, stx1 and
stx2 genes were detected in 52 (29.7%) and 73 (41.7%) of animals, respectively; while 33
(18.9%) animals were positive for STEC strains with both stx1 and stx2 (stx1/2) genes present.
There were 17 (9.7%) animals that had multiple STEC strains isolates with distinct stx profiles
between them (Table 2.3). In addition, 20 (25%) beef cattle and 47 (49%) dairy cattle had stxpositive isolates without the eaeA gene, while 36 cattle (12 dairy and 24 beef) had both stx genes
as well as eaeA gene and thus, could be classified as EHEC. Differences in the stx distribution
were also observed across herds. One beef herd was positive only for stx1 (Figure 2.2), while the
other herds were all mixed with multiple stx profiles. Additionally, beef cattle had a higher
likelihood of having stx2 than did dairy cattle (OR: 2.2; 95%CI: 1.05- 4.08; p-value: 0.04). There
58

were 288 EHEC (any stx gene with eaeA gene) strains isolated, which came from a total of 108
(62%) cattle (Figure 2.3). There were 17 animals that had both an EHEC strain and also a STEC
strain. All the herds had at least one EHEC isolate, although there was a great variety in the
number of EHEC isolates among herds, some having up to 22 EHEC isolates while others had
only one or two EHEC isolates.

2. Univariate analyses of STEC shedding in dairy herds
The univariate analyses of all the variables analyzed in the dairy herds are present in
Table 2.4. In the following paragraphs we described the most relevant findings from the
univariate analysis in dairy.

2.1. Individual host factors:

Multiple host factors including number of lactations, days in milk, antibiotic treatment,
etc were evaluated to identify associations with STEC shedding. Only two variables, however,
were important in the univariate analysis. First was number of lactations. Cows in their first
lactation were at highest risk for shedding STEC (OR: 1.6; 95%CI: 1.04- 2.58, p-value: 0.04)
relative to cows with more lactations. The number of cows in their first lactation was 279 (40%);
of those 56% were sampled during 2011. Also of those 279 first lactation cows 84% was
sampled in the summer. Cows with more lactations numbered 426 (60%). Of those 59% were
sampled during 2011 and 87% were sampled in the summer.
STEC shedding was more common in the first 30 days of lactation (OR: 3.8; 95% CI:
2.07- 6.90; p-value: <0.0001) relative to cows who had been lactating more than 30 days. A total
59

of 579 (82%) cows had been lactating more than 30 days, of these 53% were sampled during
2011 and 86% were sampled in summer. Cows in the first 30 days of lactation numbered 70
(10%). Of these 70 animals 71% were sampled during 2011 and 89% were sampled in the
summer. The other 8% were dry cows. These 54 dry cows belonged to 4 of the 6 dairy herds.
These variables were further evaluated in the final model (Table 2.4).

2.2. Environmental factors:

2.2.1 Herd characteristics
Significant herd-specific variables associated with STEC shedding included the “culling
rate” (OR: 0.5; 95% CI: 0.22- 1.25; p-value: 0.15). Herds with low culling rates had a decreased
risk of STEC shedding. Also included in this category were “proportion of the herd that is
lactating” (OR: 0.4; 95% CI: 0.12- 1.35; p-value: 0.14) and “proportion of the herd that is dry”.
These last two variables were correlated, as a consequence, only the first was chosen for the
analysis. Those herds with “percentage of dry cows” between 1.7- 4.0% were more likely to shed
STEC relative to herds with either a higher or lower percentage of dry cows.

2.2.2 Housing characteristic
Cattle with access to pasture or a dry lot did not present a higher risk of shedding STEC
compared to cattle that did not have access (OR: 0.7; 95%CI: 0.29- 1.64; p-value: 0.40), although
there was a tendency for cattle with access to pasture to have an increased risk of STEC
shedding. Similarly those first lactation cows that were housed separate from cows with more
lactations had a increased tendency for STEC shedding (OR: 1.1; 95%CI: 0.44- 2.78; p-value:
60

0.84), which is in agreement with the higher rates of STEC shedding among cows that belong to
herds that housed separated transition cows from the other cows (OR: 2.0; 95%CI: 0.96- 4.11; pvalue: 0.06). Neither variable, however, was significant in the univariate analysis.

2.2.3 Cleaning characteristics
Dairy farmers that cleaned feeders every day had a trend for a lower risk of STEC
shedding when compared to those that cleaned less frequently (OR: 2.0; 95%CI: 0.96- 4.11; pvalue: 0.06).This could be due to the elimination of an environment that can favor STEC
contamination or even multiplication. Nonetheless, cow environmental cleanliness scores, which
represents a visual subjective evaluation of the farm’s cleanliness by the interviewer, were not
significantly associated with STEC shedding (OR: 1.9; 95%CI: 0.80-4.53; p-value: 0.15). The
animals and bedding cleanliness scores were high, medium and low thirds. There was a low
variability among the cleanliness scores which possibly accounts for the lack of significance.

2.2.4 Treatment characteristics
In five herds that had a history of using antimicrobials for treatment of respiratory disease
the odds of STEC shedding was significantly lower (OR: 0.3; 95%CI: 0.19- 0.52; p-value:
<0.0001) than herds that did not use antimicrobials; only one herd did not use antimicrobials.
Products reported to be used for treating respiratory disease included ceftiofur, florefenicol and
tulathromycin. In contrast, herds with a history of using antimicrobials for treatment of foot
infections (OR: 2.5; 95%CI: 0.74- 8.23; p-value: 0.14) and metritis (OR: 2.5; 95%CI; 0.74- 8.23;
p-value: 0.14) had a non-significant higher risk of STEC shedding; only one dairy herd did not
use antimicrobials. The most common product for foot infections was copper-sulfate, whereas
61

ceftiofur and oxytetracycline was used for Metritis. The prophylactic use of anthelmintics, a
measure applied by four of the six dairy herds, was significant associated with STEC shedding
(OR: 0.4; 95%CI: 0.23- 0.84; p-value: 0.01); those herds that use anthelmintics had a lower
likelihood of STEC shedding.

2.2.5 Diet
Cows fed a diet that included a “direct-fed microbial product” had less risk of STEC
shedding (OR: 0.4; 95%CI: 0.23- 0.83; p-value: 0.0111). No other diet variables such as
percentage of corn silage, distiller’s grains, and cottonseed were significant. Neither was
significantly associated the use of Rumensin on the diet with STEC shedding. We also examined
the association of STEC shedding with the use of TMR, but the association was not significant.
Neither was significant the association between level of NEL in the diet and STEC shedding. All
farms had different diets for dry and lactating cows. Some of the farms had different diets
according with the level of milk production.

2.2.6 Contact with other animals
Two herds that had continuous exposure (OR: 2.7; 95%CI: 1.09- 6.52; p-value: 0.03), and
three herds with frequent exposure to “rodents” and “raccoons” (OR: 1.3; 95%CI: 0.53- 3.00; pvalue: 0.03) were at higher risk for STEC shedding than the one herd with rare exposure to
“rodents” and “raccoons”. On the other hand, cows that did not have contact with “dogs” and
“deer” were at less risk for STEC shedding. This is in agreement with the literature as deer have
been reported to be a source of STEC, so the lack of contact with deer should reduce the risk of
STEC shedding in cattle. All the herds had frequent or constant contact with birds; as a
62

consequence, contact with birds was not significant associated with STEC shedding. Only one
dairy herd was reported to have contact with other species, more specifically horses.

2.2.7 Environmental conditions
Regarding the temperature, the “average maximum temperature 1-5 days before
sampling” was the most predictive of the correlated environmental temperature variables and
was therefore selected for analysis in the multivariable model. For example, “temperature
average” and “minimum temperature average” on the day of sampling were highly correlated.
Overall, there was higher risk of STEC shedding when the temperature was high. At an average
maximum temperature 1-5 days before the sampling less than 28.9°F, for instance, there was a
lower probability of STEC shedding relative to cattle sampled at a higher temperature (OR: 2.0;
95%CI: 0.99- 4.03; p-value: 0.05).

2.3. Multivariable analysis for dairy:

Independent variables were evaluated individually. Twenty-eight variables yielded no
significant associations with STEC-positivity at p-value cut off of ≥ 0.15 and sixteen more at pvalue ≥ 0.05 (Table 2.4). For example, “contact with cats” and “contact with raccoons” were not
significant predictors of STEC shedding in this study. Therefore, these variables were not
incorporated in the subsequent steps of the model building process.
The variables included in the final model were “average maximum temperature five days
before sampling”, “lactation status” and “days in milk (DIM)” as fixed effects with herd as a
random effect (Table 2.5). The variance of the herd random effect was 0.2012, which yielded an
63

Intra class correlation (ICC) of 0.06. In all, a total of 692 animals were included in the final
model; 26 animals had missing values for one or more of the variables examined.
Cattle in their first lactation were 1.76 times more likely to be shedding STEC than cattle
in their second or higher lactation (OR 1.8; 95% CI 1.09- 2.83; p-value: 0.0204). Also, cows in
their first 31 days of producing milk were 3.9 times more likely to be shedding STEC than cows
with 31 or more days producing milk (OR 3.9; 95% CI 2.12- 7.18). Furthermore, dry cows were
less likely to shed STEC, although the association was not significant (OR 0.7; 95% CI 0.202.41; p-value: 0.5590). Higher average temperatures (>28.9 F) in 1-5 days before sampling
increased the likelihood of STEC shedding 2.5 times compared with lower temperatures (OR:
2.5; 95%CI: 1.25- 4.91; p-value: 0.0092).

3. Univariate analyses of STEC shedding in beef herds

All the variables analyzed were displayed in Table 2.6. Eleven variables yield no
significant associations with STEC-positivity at a p-value of 0.15. Therefore, these variables
were not incorporated in the subsequent steps of the building process for the multivariable
model, that was described in the data analysis section was not possible to built. The intraclass
correlation for the herd random effect for beef herds was low (0.076) which means that there was
more variability within herds than between herds. As a consequence there was not enough
variability to use herd as a random effect.

64

3.1 Environmental factors:

3.1.1 Herd characteristics
Herds with crossbreds were less likely to shed STEC than Holstein or Angus feedlots.
Crossbreds represent 64% of the beef animals sampled, followed by 22% Holstein and 14%
Angus. The other variables feedlot animal capacity, number of cattle fed annually, weight at
arrival, weight at sale, and cattle purchased off site were significantly associated with STEC
shedding. However, there should be caution in interpreting these results as once the variables
were categorized only one herd was different than the rest. Cow environmental cleanliness
scores, which represent a visual subjective evaluation of the farm’s cleanliness was not
significant associated with STEC shedding. None of the herds was classified in the low third of
cleanliness.

3.1.2 Housing characteristics
Only one of the herds did have exposure to pasture, and this was the herd with the highest
STEC prevalence. The other variables, times the waterers were washed, times the animal’s
holding areas were clean, and the use of disinfectant in these areas were not significantly
associated with STEC shedding. Bed and animals cleanliness were a subjective evaluation of the
farm cleanliness. Only one herd was evaluated in the middle third; all others were in the cleanest
third.

65

3.1.3 Treatment characteristics
The use of “anthelmintics” was significant as those herds that used an anthelmintic had a
decreased risk of STEC shedding (OR: 0.2; 95% CI: 0.04- 0.57; p-value: 0.01). Treatment for
respiratory diseases (OR: 5.9; 95%CI: 2.79- 12.70, p-value: <0.0001) and treatment for foot
infection (OR: 5.9; 95%CI: 2.79- 12.70, p-value: <0.0001) were correlated, so only one of them
was chosen for the next model building step. Those farms that used only one type of antibiotic
for respiratory diseases had a lower risk of STEC shedding compared with farms that used
several types of antibiotics. Those herds that used only oxytetracycline for foot infections had a
higher risk of STEC shedding compared to those herds that use others antibiotics.

3.1.4 Diet
Herd managers that fed a total mixed ration (TMR) to their cattle (OR: 6.6; 95%CI: 1.7724.31; p-value: 0.01), and used ionophores (OR: 0.2; 95% CI: 0.04- 0.57; p-value: 0.01) had a
lesser risk of STEC shedding. Only one beef herd did not use either TMR or ionophores.
Because these two variables were correlated, only one was selected for the next step in the model
construction. Beef herds had very similar diets; the only herd with different diet was the herd that
raised its animal on pasture. Contrary to dairy cattle, direct feed microbials was not significant.

3.1.5 Contact with other animals
Those herds with constant contact with opossums, deer, dogs, and skunks had a higher
risk of STEC shedding. All beef herds had contact with cats. Contact with other species
presented a lower risk of STEC shedding. Voles and weasels were the other species that some of
these herds had contact with.
66

3.1.6 Environmental conditions
The herds sampled during 2011 had less risk of STEC shedding (OR: 0.19; 95%CI: 0.060.65; p-value: 0.0077). In the case of the variables that denoted temperature, all of the variables
were significant associated with STEC except the average maximum temperature at day of
sampling. Overall, the risk of STEC shedding was higher when the temperature was high.

67

Discussion

The objective of this study was to identify risk factors in dairy and beef cattle for STEC
shedding, which could ultimately lead to STEC intervention strategies.
We found that all the dairy and beef feedlot herds we sampled were positive for STEC.
These herds had at least 6% of their cattle positive to STEC shedding. In agreement with our
study, a Swiss study reported a 100% STEC farm prevalence in the dairy farms they sampled
(Kuhnert, et al 2005). Other studies had reported lower STEC prevalence; however these studies
were testing only for STEC O157 (Dunn, et al 2004; Hancock, et al 1994; Heuvelink, et al 1998;
Sargeant, et al 2003); (Cobbaut, et al 2009) or only non-O157 STEC (Renter, et al 2007), while
we were testing for all STEC. The improvement of laboratory techniques and diagnostic tools
may also be responsible for our result that all examined herds were found to contain STEC. In
addition, it could be that Michigan has a higher STEC prevalence than other parts of USA, as
this is the first study performed in Michigan. There are studies that reported differences in STEC
prevalence by region (USDA 2003) while others report no differences (Sargeant, et al 2003).
The finding of stx2 as the most frequent gene detected in our cattle sampled is in
agreement with previous studies (Kuhnert, et al 2005; Mechie, et al 1997; Polifroni, et al 2012).
Shiga toxin 2 is the more dangerous of the two Shiga toxins to humans, as a consequence finding
stx2 as the most frequent gene has important implications from the public health perspective. It is
also important to report that 68% of the animals shed EHEC isolates, which are more virulent
and likely to result in HUS (Karmali, et al 1983). Indeed, more than half of STEC isolates from
animals had the eaeA gene. It is important to identify the most common stx genes in the STEC
bacteria isolated due to the implication in case of human infection.

68

We found that beef herds had a higher risk of STEC shedding than dairy herds. This
finding is opposed to a previous reported study (Cobbold, et al 2004). A potential explanation for
this difference could be that beef cattle were younger than dairy, so dairy animals had been
already exposure to STEC, and as consequence had a good immune response already working.
Studies have reported that younger cattle have higher risk of STEC shedding than older cattle
(Cho, et al 2009; Heuvelink, et al 1998). Other possible explanation for the incongruent results
could be that our study was in the Midwest region, during summer/spring seasons whereas
Cobbold’ s study was done in the Pacific Northwest, during fall and winter so there are
environmental and management practices differences.
Dairy cattle in their first lactation were found to be at a higher risk for shedding STEC,
this is in agreement with earlier studies. Fitzgerald, et al. (2003), for instance, found that
primiparous cows shed more STEC than multiparous cows, although this difference was not
significant (p-value >0.10). Similarly, a German study found that first calf heifers were at a
higher risk of STEC shedding than older cows (Menrath, et al 2010). Also a longitudinal study in
the UK reported the prevalence of E. coli O157:H7 to be highest in cows two years of age, which
is the typical age for the first calving (Mechie, et al 1997). Other studies have found the opposite
results. Cho et al. (2009) reported that cows with parities of ≥ 4 were 1.7 times more likely to
shed STEC compared to cows with <4 parities, although this association was non-significant. A
study in Switzerland also found that as the number of lactations increased, the risk of STEC
positivity increased (Kuhnert, et al 2005). A possible explanation for the opposite results
between studies could be differences in methodology.
The reasons for our observed association between first lactations animals and STEC are
not clear but could be related to the fact that first lactation dairy cows have different energy

69

requirements and are often in more severe negative energy balance than older cows (Edrington,
et al 2004). This negative energy balance could alter the digestive microbiota composition
(Edrington, et al 2004), which could favorite STEC colonization and shedding more during the
first lactation. This negative energy balance is one of many differences between older and
younger dairy cattle.
In our study, the risk of STEC shedding was found to be highest in the first 31 days of
lactation. Other studies have reported variable findings with regard to STEC shedding and stage
of lactation. A longitudinal study in the UK found that E. coli O157:H7 shedding peaked in the
first month of lactation, followed by low levels of shedding and then a less intense increase at
seven months postpartum (Mechie, et al 1997). These investigators speculated that modifications
in diet may explain the change in STEC shedding reasoning that diet change could potentially
modify the digestive tract microbiota, favoring STEC growth. A negative energy balance could
also explain the increased risk for STEC shedding in the first 31 days of lactation. During early
lactation, there are significant physiological and metabolic changes that occur and investigators
have suggested that the high metabolic demand associated with early lactation could potentially
favor intestinal STEC colonization and shedding (Dunn 2003; Edrington, et al 2004). In contrast,
Edrington, et al. (2004) did not find any difference in STEC O157:H7 shedding between cows in
early lactation (<60 DIM) versus late lactation (>150 DIM). Similarly, Fitzgerald, et al. (2003)
reported no effect of DIM on STEC O157:H7 shedding. In complete contrast, a study performed
in Germany found that cows with more than 50 DIM have higher risk of shedding STEC with the
highest risk being in cows with more than 350 DIM (Menrath, et al 2010). There were also
differences in methodology and the way of classified the DIM among these studies.

70

Seasonal variation in STEC shedding has been widely reported, with shedding highest in
the summer months (Berry and Wells 2010; Cobbold, et al 2004; Dunn, et al 2004; Gautam, et al
2011; Kondo, et al 2010; Smith, et al 2005). Our findings support these earlier findings. Potential
reasons for these findings include increased growth and survival in the environment at higher
ambient temperature (Smith, et al 2005). Similarly, higher temperatures could lead to changes in
normal microbiota and immunological functions that may favor STEC colonization and shedding
in cattle (Smith, et al 2005).
There are two important limitations to our study with respect to the beef herds’ sample.
We found that one herd was very different than the other herds. Several independent variables
for this herd were different from the rest of the beef cattle herds. Another limitation as previously
mentioned was the correlation between the beef herds that were raised at different times at the
same location. For these reasons we were unable to build a multivariable model for the beef
herds and interpretation of the univariate analysis should be done with caution.
In summary, the implementation of control strategies in dairy herds should focus on those
groups of animals that are at higher risk for STEC shedding. Thus first lactation cows and cows
within their 31st DIM are a group with a high risk for STEC shedding. Therefore control
strategies could be specifically targeted to this group. Although these two specific groups of
cows do not usually get into the food chain, their milk can impact food safety. Also these
animals serve as potential source of contamination for other animals on the farm and the
environment. Elucidation of STEC determinants should lead to intervention strategies to control
STEC infection in cattle, and indirectly, to reduce transmission to people.

71

APPENDIX

72

Table 2.1. Areas explored in the questionnaire for dairy and beef herds.
Area

Dairy and Beef

Farm demographics

 Herd Size

Only Dairy
 Closed or open herd
 Overall culling rate

Only Beef
 Cattle breed
 Percentage cattle gender
 Average arrival and sale

weight
 Length of time on feed

Farm and animal
management

 Source and number of cattle added

during past 12 months
 Contact with other domestic or wild

species
 Cleaning process and frequency for






feedbunks, waterers and areas
where animals are housed
Percentage cattle receiving a total
mixed ration
Composition of diet
Fly control
Type of bedding
Water source







Anthelmintic used
Antimicrobial used
Morbidity and mortality
Used of feed additives
Growth promoters



Herd health
management

 Access to pasture
 Grouping of lactating,

transition and sick cows
 Number of milkings per day
 Type of lactating cow
housing

73

 Housing if animals post-

arrival

Table 2.2. Herd identification, type of production system, total number of animals in each herd, number of animals sampled in each
herd, number of animals tested for STEC and year of sampling.

Herd

Type of
herd

Total number
of animals

B1

Beef

136

D2

Dairy

B3

Number of
animals sampled
(% of total)

Number of animals
STEC tested

Year of
sampling

136 (100)

134

2011

320

154 (48)

149

2011

Beef

36

36 (100)

32

2011

D4

Dairy

3000

175 (9)

174

2011

D6

Dairy

98

94 (96)

94

2011

D7

Dairy

12000

100 (1)

100

2012

B8

Beef

54

54 (100)

54

2012

D9

Dairy

243

100 (41)

100

2012

D10

Dairy

530

101 (19)

101

2012

B11

Beef

83

83 (100)

83

2012

B12

Beef

75

75 (100)

75

2012

74

Figure 2.1. Prevalence of STEC by herd. D= dairy, B= beef

Percentage cattle positive for STEC

100

90
80
70
60
50
40
30
20
10
0
D2

D4

D6

D7

D9

D10

B1

Herds identification

75

B3

B8

B11

B12

Table 2.3. Prevalence of genes stx1, stx2 and eaeA by total of animals, dairy animals and beef
animals.

Gene prevalence

Total animals (n=175)

Dairy animals (n=95)

Beef animals (n=80)

stx1 (%)

52 (29.7)

35 (37)

17 (21)

stx2 (%)

73 (41.7)

38 (40)

35 (44)

stx1/2 (%)

33 (18.9)

11 (12)

22 (28)

stx1 and stx2

4 (2.3)

4 (4)

0

stx1 and stx1/2

3 (1.7)

2 (2)

1 (1)

stx2 and stx1/2

8 (4.6)

3 (3)

5 (6)

stx1, stx2 and stx1/2

2 (1.1)

2 (2)

0

eaeA positive (%)

108 (62)

48 (51)

60 (75)

eaeA negative (%)

67 (38)

47 (49)

20 (25)

76

Figure 2.2. Prevalence of stx1, stx2 and stx1/2 by herd. D= dairy, B=beef. The vertical axis has
been set up at 50% to improve visibility. B= beef and D= dairy

50
stx1
45

stx2

Percentage animals

40

stx1/2

35
30
25
20
15
10
5
0
D2

D4

D6

D7

D9

D10

B1

Herd Identification
stx1= Shiga toxin 1 gene
stx2= Shiga toxin 1 gene
stx1/2= Shiga toxin 1 gene and Shiga toxin 2 gene

77

B3

B8

B11

B12

Figure 2.3. Prevalence of enterohemorrhagic Escherichia coli and Shiga toxin-producing E. coli
strains by herd. D= dairy, B=beef. The vertical axis has been set up at 50% to improve visibility.

50
45
Percentage animals

40
35
30
25

EHEC

20

STEC

15
10
5
0
D2

D4

D6

D7

D9

D10

B1

Herds identification

78

B3

B8

B11 B12

Table 2.4. Univariable analysis of dairy herd variables for risk of STEC shedding with herd as a random effect

Characteristic
Year
2011
2012
Season
Spring
Summer
Temperature Aver.
> 20.6F
≤ 20.6F
Temperature Max
> 27.8F
≤ 27.8F
Temperature Min
> 15.6
≤ 15.6
Temperature Aver5 days
>19.4
≤19.4
Temperature Max5 days
>28.9
≤28.9
Temperature Min5 days
>15
≤15

No. (%) with
characteristic

No. (%) with
STEC

p-value

OR

95% CI

417 (58.1)
301 (41.9)

43 (10.3)
52 (17.3)

0.0925

0.534
ref

0.258- 1.109
ref

100 (14)
618 (86)

13 (13)
82 (13.3)

0.9685

1.024
ref

0.309- 3.395
ref

301 (41.9)
417 (58.1)

52 (17.3)
43 (10.3)

0.0925

1.871
ref

0.902- 3.883
ref

201 (28)
517 (72)

39 (19.4)
56 (10.8)

0.0631

1.990
ref

0.963- 4.111
ref

301 (41.9)
417 (58.1)

52 (17.3)
43 (10.3)

0.0925

1.871
ref

0.902- 3.883
ref

395 (55)
323 (45)

58 (14.7)
37 (11.5)

0.6062

1.265
ref

0.517- 3.099
ref

375 (52.2)
343 (47.8)

63 (16.8)
32 (9.3)

0.0519

2.002
ref

0.994- 4.031
ref

475 (66.2)
243 (33.8)

76 (16)
19 (7.8)

0.0299

2.299
ref

1.085- 4.873
ref

79

Table 2.4. (cont’d)
Longitude
< -85.241836
> -85.241837 < -84.829245
> -84.54
Lactation
1st
2nd or higher
Days in milk
0
1-30d
>= 31d
Dry
Yes
No
Breed
Jersey
Mixed
Holstein
Herd type
Closed
Open
Calves-Replacements proportion
<5.1
>48.4
From 46.9 to 48.3
Proportion herd lactating
>50
<31.7% to 49%

274 (38.2)
100 (13.9)
344 (47.9)

52 (19)
13 (13)
30 (8.7)

0.0128

2.592
1.573
ref

1.378- 4.875
0.666- 3.715
ref

279 (39.57)
426 (60.43)

49 (17.56)
44 (10.33)

0.0346

1.636
ref

1.036- 2.584
ref

54 (7.68)
70 (9.96)
579 (82.36)

5 (9.26)
21 (30)
64 (11.05)

<.0001

1.091
3.778
ref

0.397- 3.003
2.069- 6.900
ref

53 (7.42)
661 (92.58)

4 (7.55)
90 (13.62)

0.4124

0.635
ref

0.214- 1.882
ref

94 (13.1)
101 (14.1)
523 (72.8)

6 (6.4)
11 (10.9)
78 (14.9)

0.2461

0.378
0.686
ref

0.116- 1.237
0.240- 1.961
ref

149 (20.8)
569 (79.3)

13 (8.7)
82 (14.4)

0.3539

0.59
ref

0.197- 1.789
ref

94 (13.1)
100 (14)
524 (73)

6 (6.4)
13 (13)
76 (14.5)

0.3230

0.395
0.878
ref

0.117- 1.328
0.305- 2.528
ref

94 (13.1)
624 (86.9)

6 (6.4)
89 (14.3)

0.1396

0.404
ref

0.122- 1.345
ref

80

Table 2.4. (cont’d)
Proportion herd dry
1.7- 4%
>7.6%
6.2- 7.5%
Herd Size
>1000
<1000
Adding Cow/Replacements
At least 5 animals
0%
Adding Bulls
4 animals
0%
Culling Rate
Low level
High level
N° Milkings
3- 4 times
2-3 times
Loose Housing
No
Yes
Tie stanchion
No
Yes
Free stall
Yes
No
Access pasture/dry lot
No
Yes

200 (27.9)
94 (13.1)
424 (59.1)

41 (20.5)
48 (11.3)
6 (6.4)

0.0133

2.025
0.537
ref

1.109- 3.699
0.196- 1.476
ref

274 (38.2)
444 (61.8)

37 (13.5)
58 (13.1)

0.8411

1.099
ref

0.435- 2.782
ref

149 (20.8)
569 (79,3)

13 (8.7)
82 (14.4)

0.3539

0.594
ref

0.197- 1.789
ref

101 (14.1)
617 (85.9)

11 (10.9)
84 (13.6)

0.7197

0.802
ref

0.241- 2.674
ref

195 (27.2)
523 (72.8)

17 (8.7)
78 (15)

0.1452

0.525
ref

0.221- 1.250
ref

274 (38.2)
444 (61.8)

37 (13.5)
58 (13.1)

0.8411

1.099
ref

0.435- 2.782
ref

624 (86.9)
94 (13.1)

89 (14.3)
6 (6.4)

0.1396

2.473
ref

0.743- 8.227
ref

569 (79.3)
149 (20.8)

82 (14.4)
13 (8.7)

0.3539

1.684
ref

0.559- 5.073
ref

718 (100)
0

95 (13.2)
0

N/A

524 (73)
194 (27)

61 (11.6)
34 (17.5)

0.3961

0.687
ref

0.288- 1.637
ref

81

Table 2.4. (cont’d)
Lactation access pasture
No
Yes
Transition pen separate
No
Yes
Sick animals penned separated
Yes
No
1st lactations animals penned separated
Yes
No
Cow/Heifers Raised
Another farm
Off-site/ On main farm
Feeders clean Year
<365
365
Washed
No
Yes
Spray
No
Yes
Lime
No
Yes
Tx Respiratory Disease
Yes
No

524 (73)
194 (27)

61 (11.6)
34 (17.5)

0.3961

0.687
ref

0.288- 1.637
ref

201 (28)
517 (72)

39 (19.4)
56 (10.8)

0.0631

1.990
ref

0.963- 4.111
ref

368 (51.6)
350 (48.8)

43 (11.7)
52 (14.9)

0.4421

0.718
ref

0.308- 1.672
ref

274 (38.2)
444 (61.8)

37 (13.5)
58 (13.1)

0.8411

1.099
ref

0.435- 2.782
ref

94 (13.1)
624 (876.9)

6 (6.4)
89 (14.3)

0.1396

2.473
ref

0.743- 8.227
ref

201 (28)
517 (72)

39 (19.4)
56 (10.8)

0.0631

1.990
ref

0.963- 4.111
ref

469 (65.3)
249 (34.7)

54 (11.5)
41 (16.5)

0.2509

0.614
ref

0.267- 1.413
ref

569 (79.3)
149 (20.8)

82 (14.4)
13 (8.7)

0.3539

1.684
ref

0.559- 5.073
ref

524 (73)
194 (27)

76 (14.5)
19 (9.8)

0.2067

1.619
ref

0.654- 4.005
ref

618 (86.1)
100 (13.9)

67 (10.8)
28 (28)

<.0001

0.313
ref

0.189- 0.518
ref

82

Table 2.4. (cont’d)
Tx Foot Infection Disease
Yes
No
Tx Metritis
Yes
No
Feed TMR
No
Yes
% Corn silage Diet
No
Yes
% Distillers grains Diet
No
Yes
% Cottonseed Diet
No
Yes
Other SP
Horses
None
NEL
Contact with Cats
No
Yes
Contact with Deer
No
Yes

624 (86.9)
94 (13.1)

89 (14.3)
6 (6.4)

0.1396

2.473
ref

0.743- 8.227
ref

624 (86.9)
94 (13.1)

89 (14.3)
6 (6.4)

0.1396

2.473
ref

0.743- 8.227
ref

194 (27)
524 (73)

34 (17.5)
61 (11.6)

0.3961

1.456
ref

0.611- 3.469
ref

205 (30.8)
461 (69.2)

34 (16.6)
48 (10.4)

0.8115

1.117
ref

0.450- 2.769
ref

496 (74.5)
170 (25.5)

67 (13.5)
15 (8.8)

0.2053

1.738
ref

0.738- 4.092
ref

503 (75.5)
163 (24.5)

62 (12.3)
20 (12.3)

0.7234

0.860
ref

0.372- 1.987
ref

94 (13.1)
624 (86.9)

6 (6.4)
89 (14.3)

0.1396

0.404
ref

0.122- 1.345
ref

717 (99.86)

95 (13.25)

0.3511

0.419

0.067- 2.612

101 (14.07)
617 (85.93)

11 (10.89)
84 (13.61)

0.7197

0.802
ref

0.241- 2.674
ref

100 (13.93)
618 (86.07)

13 (13)
82 (13.27)

0.9685

1.024
ref

0.309- 3.395
ref

83

Table 2.4. (cont’d)
Contact with Dogs
No
Yes
Contact with Opossum
No
Yes
Contact with Raccoons
Always
Frequent
Rarely
Contact with Rodents
Always
Frequent
Rarely
Contact with Skunks
No
Yes
ACleanliness
Cleanest third
Middle third
Bed Cleanliness
Cleanest third
Middle third
Rumensin
Yes
No

374 (52.1)
344 (47.9)

65 (17.4)
30 (8.7)

0.0111

2.269
ref

1.206- 4.267
ref

101 (14.07)
617 (85.93)

11 (10.89)
84 (13.61)

0.7197

0.802
ref

0.241- 0.7197
ref

200 (27.9)
369 (51.4)
149 (20.8)

41 (20.5)
41 (11.1)
13 (8.7)

0.0311

2.671
1.257
ref

1.094- 6.520
0.528- 2.995
ref

200 (27.9)
369 (51.4)
149 (20.8)

41 (20.5)
41 (11.1)
13 (8.7)

0.0311

2.671
1.257
ref

1.094- 6.520
0.528- 2.995
ref

101 (14.07)
617 (85.93)

11 (10.89)
84 (13.61)

0.7197

0.802
ref

0.241- 2.674

523 (72.8)
195 (27.2)

78( 14.9)
17 (8.7)

0.1452

1.903
ref

0.800- 4.525
ref

523 (72.8)
195 (27.2)

78( 14.9)
17 (8.7)

0.1452

1.903
ref

0.800- 4.525

274 (38.2)
444 (61.8)

37 (13.5)
58 (13.1)

0.8411

1.099
ref

0.435- 2.782
ref

84

Table 2.4. (cont’d)
Direct Fed Microbials
Yes
No
Anthelmintic
Yes
No

344 (47.9)
374 (52.1)

30 (8.7)
65 (17.4)

0.0111

0.441
ref

0.234- 0.829
ref

518 (72.2)
200 (27.9)

54 (10.4)
41 (20.5)

0.0123

0.443
ref

0.234- 0.838
ref

85

Table 2.5. Final multivariable model for dairy herds with herd as a random effect.

Variable

Estimate

Standard
Error

p-value

0.9079
0

0.3476
ref

0.5640
0

Dry
1 – 30 days
>=31
Intercept

Individual
p-value

OR

95% CI

0.0092

2.479
ref

1.253- 4.906
ref

0.2426
ref

0.0204

1.758
ref

1.092- 2.830
Ref

-0.3731
1.3604
0

0.6338
0.3108
ref

<.0001

0.689
3.898
ref

0.197- 2.411
2.117- 7.175
Ref

-2.0173

0.2594

TempFmax5 days
> 28.9 F
<=28.9 F
Lactation
First
2 or more
DIM
0.5590
<.0001

86

Table 2.6. Univariable analysis of beef herd variables for risk of STEC shedding with herd as a random effect

Characteristic
Feedlot size capacity
<=54
>54
Feed Annual
<= 54
> 54
Proportion Beef breed
< 100%
> 100%
Breed
Holstein
Angus
Crossbreed
Proportion Steers
Mix
100% Male
Arrival average weight
< 272.2 kg
> 272.2 kg
Sale average weight
<589.7 kg
>589. 7 kg
Feedlot days
< 200
> 200

No. (%) with
characteristic

No. (%) with
STEC

p-value

OR

95% CI

54 (14.29)
324 (85.71)

29 (53.7)
51 (15.74)

0.0050

6.560
ref

1.771-24,306
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771-24,306
Ref

249 (65.9)
129 (34.1)

27 (10.8)
53 (41.1)

<.0001

0.168
ref

0.079- 0.359
Ref

83 (21.96)
54 (14.29)
241 (63.76)

13 (15.66)
29 (53.70)
38 (15.77)

0.0194

1.033
6.617
ref

0.258- 4.137
1.691- 25.887
Ref

249 (65. 9)
129 (34.1)

27 (10.8)
53 (41.1)

<.0001

5.947
ref

2.785- 12.702
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771-24,306
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771-24,306
Ref

158 (41.8)
220 (58.2)

37 (23.4)
43 (19.6)

0.7800

1.285
ref

0.220- 7.517
ref

87

Table 2.6. (cont’d)
Purchased from off-site
0%
100%
Purchased from out-state
100%
< 100%
Barn post arrival
True
False
Antibiotics at arrival
Yes
No
Waterer clean Year
Never
Yes
Sanitation areas
No
Yes
Washed
Never
Once a year
Spray Disinfectant
No
Yes
Tx Respiratory Disease
Several antibiotics
Only one type
Tx Foot Infections
Oxitetracycline
Others

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771-24,306
Ref

303 (80.2)
75 (19.8)

56 (18.5)
24 (32)

0.4546

0.462
ref

0.061- 3.518
Ref

75 (19.8)
303 (80.2)

24 (32)
56 (18.5)

0.4546

2.167
ref

0.284- 16.515
Ref

75 (19.8)
303 (80.2)

24 (32)
56 (18.5)

0.4546

2.167
ref

0.284- 16.515
Ref

75 (19.8)
303 (80.2)

24 (32)
56 (18.5)

0.4546

2.167
ref

0.284- 16.515
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.50
ref

1.771- 24.306
Ref

129 (34.1)
249 (65.9)

53 (41.1)
76 (59)

<.0001

5.947
ref

2.785- 12.702
Ref

303 (80.2)
75 (19.8)

56 (18.5)
24 (32)

0.4546

0.462
ref

0.061- 3.518
Ref

129 (34.1)
249 (65.9)

53 (41.1)
76 (59)

<.0001

5.947
ref

2.785- 12.702
Ref

129 (34.1)
249 (65.9)

53 (41.1)
76 (59)

<.0001

5.947
ref

2.785- 12.702
ref

88

Table 2.6. (cont’d)
Tx Arthritis
No
Yes
Feed TMR
0%
100%
Forage Diet
15 %
100 %
NEG
High Level
Low Level
Corn silage diet
0%
15 %
Distiller grains diet
0%
20 %
Antibiotics before sampling
No
Yes
Radius animals presence
100
400
500
Other SP
Yes
No
Pasture
Yes
No

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771- 24.306
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771- 24.306
Ref

324 (85.7)
54 (14.3)

51 (15.7)
29 (53.7)

0.0050

0.152
ref

0.041- 0.565
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771- 24.306
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771- 24.306
Ref

303 (80.2)
75 (19.8)

56 (18.5)
24 (32)

0.4546

0.462
ref

0.061- 3.518
Ref

76 (20.1)
302 (79.9)

25 (32.9)
55 (18.2)

0.2708

2.891
ref

0.436- 19.193
Ref

129 (34.13)
166 (43.92)
83 (21.96)

53 (41.09)
14 (8.43)
13 (15.66)

<.0001

3.880
0.498
ref

1.663- 9.048
0.191- 1.294
Ref

249 (65. 9)
129 (34.1)

27 (10.8)
53 (41.1)

<.0001

0.163
ref

0.079- 0.359
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771-24,306
ref

89

Table 2.6. (cont’d)
Anthelmintic
Yes
No
Rumensin
Yes
No
Direct Fed Microbials
Yes
No
Area Cleanliness
Cleanest third
Middle third
Contact with Opossum
Always
None
Contact with Deer
None
Yes
Contact with Dogs
Always
None
Contact with Skunks
Always
None
Year
2011
2012
Season
Spring
Summer

324 (85.7)
54 (14.3)

51 (15.7)
29 (53.7)

0.0050

0.152
ref

0.041- 0.565
Ref

324 (85.7)
54 (14.3)

51 (15.7)
29 (53.7)

0.0050

0.152
ref

0.041- 0.565
Ref

75 (19.8)
303 (80.2)

24 (32)
56 (18.5)

0.4546

2.167
ref

0.284- 16.515
Ref

303 (80.2)
75 (19.8)

56 (18.5)
24 (32)

0.4546

0.462
ref

0.061-3.518
Ref

129 (34.1)
249 (65.9)

53 (41.1)
76 (59)

<.0001

5.947
ref

2.785- 12.702
Ref

249 (65. 9)
129 (34.1)

27 (10.8)
53 (41.1)

<.0001

0.163
ref

0.079- 0.359
Ref

129 (34.1)
249 (65.9)

53 (41.1)
27 (10.8)

<.0001

5.947
ref

2.785- 12.702
Ref

129 (34.1)
249 (65.9)

53 (41.1)
27 (10.8)

<.0001

5.947
ref

2.785- 12.702
Ref

166 (43.9)
212 (56.1)

14 (8.4)
66 (31.1)

0.0077

0.194
ref

0.058- 0.645
Ref

188 (49.74)
190 (50.26)

40 (49.74)
40 (21.05)

0.6438

1.508
ref

0.263- 8.647
Ref

90

Table 2.6. (cont’d)
Temperature Aver.
> 20.6 C
≤ 20.6 C
Temperature Max
> 27.8 C
≤ 27.8 C
Temperature Min
> 15.6 C
≤ 15.6 C
Temperature Aver5 days
>19.4 C
≤19.4 C
Temperature Max5 days
>28.9 C
≤28.9 C
Temperature Min5 days
>15 C
≤15 C

129 (34.1)
249 (65.9)

53 (41.1)
27 (10.8)

<.0001

5.947
ref

2.785- 12.702
Ref

161 (42.59)
217 (57.41)

56 (34.78)
24 (11.06)

0.0582

3.497
ref

0.957- 12.773
Ref

212 (56.08)
166 (43.92)

66 (31.13)
14 (8.43)

0.0077

5.163
ref

1.549- 17.203
Ref

129 (34.1)
249 (65.9)

53 (41.1)
27 (10.8)

<.0001

5.947
ref

2.785- 12.702
Ref

129 (34.1)
249 (65.9)

53 (41.1)
27 (10.8)

<.0001

5.947
ref

2.785- 12.702
Ref

54 (14.3)
324 (85.7)

29 (53.7)
51 (15.7)

0.0050

6.560
ref

1.771-24,306
Ref

91

Table 2.7. Description of the abbreviations used for each variable analyzed in the beef and dairy
models.

Variable

Abbreviation

Average temperature on day of sampling
Maximum temperature on day of sampling
Minimum temperature on day of sampling
Average temperature during 5 days previous sampling
Maximum temperature during 5 days previous sampling
Minimum temperature during 5 days previous sampling
Longitude coordinates of farm
Number of lactations
Number of days the cow had been producing milk
Cow was not producing milk
Cattle's breed
Herd incorporated cattle from other farms
Proportion of the herd that were calves and replacements
cattle
Proportion of the herd that were lactating cows
Proportion of the herd that were dry cows
How many cows & heifers were added in the last year
How many bulls were added in the last year
Overall culling rate (% culled per year)
Number of milkings per day (2x/3x/Combination, 2x-3x)
Lactating cow housing included free stall
Lactating cow housing included loose housing
Did cattle have access to pasture
Lactating cow housing included tie stall/stanchion
Do lactating cows have access to pasture/dry lot

Temperature Aver
Temperature Max
Temperature Min
Temperature Aver 5 days
Temperature Max 5 days
Temperature Min 5 days
Longitude
Lactation
Days in milk
Dry
Breed
Herd type
Calves-Replacements
proportion
Proportion herd lactating
Proportion herd dry
Adding Cow/Replacements
Adding Bulls
Culling rate
N° Milkings
Free stall
Loose Housing
Pasture
Tie stanchion
Lactation access pasture
1st lactation animals penned
separated
Transition pen separate
Sick animals penned separated
Cow-Heifers Raised
Feeders clean Year
Washed

Were first lactation animals penned separately
Were transition (post-calving) animals penned separately
Were sick cows penned separately
Where were cows and heifers mostly raised
How often were feedbunks cleaned per year
How often were the processing/animal handling areas
washed/power washed per year
How often were the processing/animal handling areas
sprayed with disinfectant per year
92

Spray Disinfectant

Table 2.7. (cont’d)
How often was lime spread on the processing/animals
handling areas per year
Were treatments for respiratory disease administered
Were treatments for foot infections administered
Were treatments for arthritis/swollen joints administered
Were treatments for metritis administered
Percentage of cattle receiving TMR
Percentage of Forage in diet
Percentage of Corn silage in diet
Percentage of Distiller's grain in diet
Percentage of Cottonseed meal in diet
Was Rumensin included in ration
Were Direct fed microbials used in the ration
Were Antiparasitic agents used
How many cattle did reside within a 2-mile radius of the
farm
Did Opossum have contact with cattle environment/feed
Did cats have contact with cattle environment/feed
Did Deer have contact with cattle environment/feed
How frequent was the contact of Raccoons with cattle
environment/feed
Did Dogs have contact with cattle environment/feed
Did Skunks have contact with cattle environment/feed
How frequent was the contact of Rodents with cattle
environment/feed
Did Other species of animals have contact with cattle
environment/feed
Cattle cleanliness score in thirds
Cattle bedding cleanliness score in thirds
Did cattle receive antibiotics any route during one month
previous sampling
What was the capacity of animals that feedlot can hold
How many cattle were fed/marketed annually
Proportion of cattle that were breed beef
Proportion of cattle that were steers
In the last year, Average arrival weight of the cattle fed
Average sale weight of the cattle
Average length of time cattle were on your feedlot (days)
What percentage of incoming cattle were purchased from
off-site
93

Lime
Tx Respiratory Disease
Tx Foot Infection Disease
Tx Arthritis
Tx Metritis
Feed TMR
Forage Diet
% Corn silage Diet
% Distillers grains Diet
% Cottonseed Diet
Rumensin
Direct Fed Microbials
Anthelmintic
Radius animal presence
Contact with Opossum
Contact with Cats
Contact with Deer
Contact with Raccoons
Contact with Dogs
Contact with Skunks
Contact with Rodents
Other SP
ACleanliness
Bedcleanliness
Antibiotics before sampling
Feedlot size capacity
Feed Annual
Proportion Beef breed
Proportion Steers
Arrival average weight
Sale average weight
Feedlot days
Purchased from off-site

Table 2.7. (cont’d)
Proportion purchased out-of-state (%)
Were cattle housed in a separate barn post-arrival
Were antibiotics used in the feed or water at arrival

94

Purchased from out-state
Barn post arrival
Antibiotics at arrival

Figure 2.4. Dairy and beef questionnaires used to collect information from the herds.

Michigan State University
STEC Project
Dairy Producer Questionnaire

____________________________________________________
Herd Code:

Farm Name:

Owner/Mngr:

Appointment:

Address:

City/Town:

County:

Cell Phone:

Office Phone:

Email:

Interviewer:

Date:

Veterinarian:

Interviewee:

____________________________________________________

95

Figure 2.4. (cont’d)
Herd Information:

What is the usual herd population?
Calves & heifers (#):

Lactating Cows (#)

Dry Cows (#):

How many animals were added from off the farm in the last year?
Cows & heifers (#):

Bulls (#):

Closed herd

Overall culling rate (% culled per year):
Number of milkings per day (2x/3x/Combination, 2x-3x):
Which best describes your type of lactating cow housing (check all that apply):
Free Stall

Proportion of herd:

Loose Housing

Proportion of herd:

Access to pasture/dry lot

Proportion of herd:

Tie stall/stanchion

Proportion of herd:

Do lactating cows have access to pasture (Yes [in season] / Rarely / No):
Are first lactation animals penned separately (Yes / Sometimes / No):
Are transition (post-calving) animals penned separately (Yes / Sometimes / No):
Are sick cows penned separately (Yes / Sometimes / No):
Are cows and heifers mostly raised (on main farm / off-site / contracted from another farm):

96

Figure 2.4. (cont’d)
How often feedbunks are cleaned (times)?

Day / Week / Month:

How often waterers are cleaned (times)?

Day / Week / Month:

What methods are used to clean areas animals are housed?
Scrape

Wash/Power Wash

Spray a Disinfectant

None
How often is each type of cleaning done and where?
Scrape:
Wash/Power Wash:
Spray a disinfectant:
Spread Lime:

What are the common antibiotics/remedies used for each of the following purposes?:

Dry Cow treatment:

Clinical Mastitis:

97

Spread Lime

Figure 2.4. (cont’d)
Metritis:

Respiratory
Disease:

Foot Infections
(including
footbaths):

Are dewormers agents used?

Yes

No

If yes, what is used?

What percentage of the cows receive a total mixed ration (TMR)?
Indicate if the lactating herd receives the following as part of their diet and at what proportions:
Forage

Percentage of diet (%):

Concentrate

Percentage of diet (%):

98

Figure 2.4. (cont’d)
Total:

100%

Corn silage

Percentage of diet (%):

Distiller’s grain

Percentage of diet (%):

Cottonseed meal

Percentage of diet (%):

List the principle ingredients and approximate percentages for the most recent ration (if possible, get a copy ration for targeted
pen):

Is Rumensin included in your ration?

Yes

Are any direct fed microbials used in the ration?

No
Yes

No

If yes, what products are used?

Do you use antimicrobials in feed?

Yes

No

If yes, what antimicrobials are used and for what purpose?

99

Figure 2.4. (cont’d)
Which best describes methods of fly control (check all that apply, circle most common):

Pour-on insecticide

Premise spray

Back Rub/Duster

Eat tags

None
Feed

Predator Insects

Fly Bait

Fly sticky strips

Larvicide
Other Describe:
About how many cattle reside within a 2-mile radius of farm (Excluding your farm)?
What is the frequency the following animals have contact with cattle environment or cattle feed (None, Rarely, Frequently,
Always):
Opossums:

Cats

Deer

Raccons:

Dogs

Skunks

Rodents:

Birds

If birds, which species:
Other (describe):

100

Figure 2.4. (cont’d)
Pen/Group Number:

Description:

Average lactating cow stock density (Average cow per stall or cows per sq. ft if not free stall):
How many animals in the past month have had any of the following symptoms (new cases):
Diarrhea:

Bloat:

Respiratory:

Clinical Mastitis:

Metritis:

Displaced abomasum:

Ketosis:
List unusual health observations for this pen:

What was the annual mortality rate of the adult cows (%) ?

What is the primary housing method for cows?
Loose/group housing

Freestall dry lot

Principle bedding type:
None

Sand

Shavings

Straw

Mattress

Other (describe):

101

Pack

Hulls

Figure 2.4. (cont’d)
Which best describes the water source (never / sometimes / usually / always):
Continuos flow water tank:

Ritchie type water

Nose operated water cups/bowls

Surface Water

If surface water, describe:
Other (describe):

102

Figure 2.4. (cont’d)
Michigan State University
STEC Project
Beef Producer Questionnaire

__________________________________________________________
Herd Code:

Farm Name:

Owner/Mngr:

Appointment:

Address:

City/Town:

County:

Cell Phone:

Office Phone:

Email:

Interviewer:

Date:

Veterinarian:

Interviewee:

____________________________________________________

103

Figure 2.4. (cont’d)
Herd Information:

How many cattle pens are there?
How many cattle pens contain animals that will be here at least 90 days?
How many cattle pens can be sampled?
Feedlot Capacity:

Capacity of pen(s) to be sampled:

How many cattle are fed/marketed annually?
In the last year, what proportion of the cattle fed are:
Holstein (#):

Beef type (#):

In the last year, what proportion of the cattle fed are:
Steers (#):

Heifers (#):

In the last year, what is the average arrival weight of the cattle fed (lb)?
What is the average sale weight of the cattle (lb)?
What is the average length of time cattle are on your feedlot (days)?
What proportion of incoming cattle are purchased:
In-State (%):

Out-of-State (%):

What is the most common out-of-state location cattle came from?

104

Figure 2.4. (cont’d)
Are the cattle housed in a separate barn for 30-45 days post-arrival:

Yes

Upon arrival, are antibiotics used in the feed or water of the new cattle? :

No

Yes

No

How often are feedbunks cleaned (time(s) per day/week/month)?________ Day/Week/Month
How often are waterers cleaned (time(s) per day/week/month)? _________ Day/Week/Month
What methods are used to sanitize the processing/animal handling areas?
None

Scrape

Wash/Power Wash

Spray a Disinfectant

How often is each done?
Scrape:
Wash/Power Wash:
Spray a disinfectant:
Spread Lime:
What are the common antibiotics/remedies used for each of the following purposes?:
Respiratory Disease:

Foot Infections:

105

Spread Lime

Figure 2.4. (cont’d)
Arthritis/Swollen Joints:

Are dewormers agents used?

Yes

No

If yes, what is used?
What percentage of the cattle receive a total mixed ration (TMR)?
Indicate if the herd receives the following as part of their diet and at what proportions:
Forage

Percentage of diet (%):

Concentrate

Percentage of diet (%):
Total:

100%

Corn silage

Percentage of diet (%):

Distiller’s grain

Percentage of diet (%):

Cottonseed meal

Percentage of diet (%):

List the principle ingredients and approximate percentages for the most recent ration (if possible, get a copy ration for
targeted pen):

Is Rumensin included in your ration?

Yes

106

No

Figure 2.4. (cont’d)
Are any direct fed microbials used in the ration?

Yes

No

If yes, what products are used?

Do you use antimicrobials in feed?

Yes

No

If yes, what antimicrobials are used and for what purpose?

Do you use anticoccidials?

Yes

No

If yes, what anticoccidials are used?

Which best describes methods of fly control (check all that apply, circle most common):
None

Pour-on insecticide

Feed Larvicide
Other

Predator Insects

Premise spray
Fly Bait

Back Rub/Duster
Fly sticky strips

Describe:

About how many cattle reside within a 2-mile radius of farm (Excluding your farm)?

107

Eat tags

Figure 2.4. (cont’d)
What is the frequency the following animals have contact with cattle environment or cattle feed (None, Rarely,
Frequently, Always):
Opossums:

Cats

Deer

Raccons:

Dogs

Skunks

Rodents:

Birds

If birds, which species:
Other (describe):

____________________________________________________________

Test Group Information:

Pen Number:
Characterize the cattle in the pen (check all that apply):
Holstein:

Beef type:

Steer:

Heifer:

What was the origin of the pen?
What is the average purchase weight (lb)?
What proportion of cattle in the pen were (%):
108

Figure 2.4. (cont’d)
- Commingled by an order buyer?
- Commingled after arrival?
- All from a single source?
How many animals in the past month have had any of the following symptoms (new cases):
Diarrhea:

Bloat:

Respiratory (shipping fever):

List unusual health observations for this pen:

What was the sickness rate of this pen during the first 45 days in the feedlot (%)?
What was the mortality rate of this pen (%)?:
What antibiotics have been used in feed or water within the last two months?
Which best describe the type of housing for the test pen:
Bedded pack
Outside lot

Pasture

Slotted floor barn

Other (describe):

Principle bedding source (check all that apply):
None

Sand

Corn stocks

Straw

109

Sawdust

Figure 2.4. (cont’d)
Other (describe):
Which best describes the water source (never / sometimes / usually / always):
Continuos flow water tank:

Ritchie type water

Nose operated water cups/bowls

Surface Water

If surface water, describe:
Other (describe):

110

REFERENCES

111

REFERENCES

ASAKURA, H., MAKINO, S., SHIRAHATA, T., TSUKAMOTO, T., KURAZONO, H.,
IKEDA, T. & TAKESHI, K. (1998) Detection and genetical characterization of Shiga toxinproducing Escherichia coli from wild deer. Microbiol. Immunol. 42, 815-822
BERRY, E. D. & WELLS, J. E. (2010) Escherichia coli O157:H7: recent advances in research
on occurrence, transmission, and control in cattle and the production environment. Adv. Food
Nutr. Res. 60, 67-117
BEUTIN, L., GEIER, D., STEINRUCK, H., ZIMMERMANN, S. & SCHEUTZ, F. (1993)
Prevalence and some properties of verotoxin (Shiga-like toxin)-producing Escherichia coli in
seven different species of healthy domestic animals. J. Clin. Microbiol. 31, 2483-2488
BEUTIN, L., MIKO, A., KRAUSE, G., PRIES, K., HABY, S., STEEGE, K. & ALBRECHT, N.
(2007) Identification of Human-Pathogenic Strains of Shiga Toxin-Producing Escherichia coli
from Food by a Combination of Serotyping and Molecular Typing of Shiga Toxin Genes. Appl.
Environ. Microbiol. 73, 4769-4775
CERNICCHIARO, N., PEARL, D. L., GHIMIRE, S., GYLES, C. L., JOHNSON, R. P.,
LEJEUNE, J. T., ZIEBELL, K. & MCEWEN, S. A. (2009) Risk factors associated with
Escherichia coli O157:H7 in Ontario beef cow-calf operations. Prev. Vet. Med. 92, 106-115
CHASE-TOPPING, M. E., MCKENDRICK, I. J., PEARCE, M. C., MACDONALD, P.,
MATTHEWS, L., HALLIDAY, J., ALLISON, L., FENLON, D., LOW, J. C., GUNN, G. &
WOOLHOUSE, M. E. J. (2007) Risk Factors for the Presence of High-Level Shedders of
Escherichia coli O157 on Scottish Farms. J. Clin. Microbiol. 45, 1594-1603
CHO, S., FOSSLER, C. P., DIEZ-GONZALEZ, F., WELLS, S. J., HEDBERG, C. W.,
KANEENE, J. B., RUEGG, P. L., WARNICK, L. D. & BENDER, J. B. (2009) Cattle-level risk
factors associated with fecal shedding of Shiga toxin-encoding bacteria on dairy farms,
Minnesota, USA. Can. J. Vet. Res.-Rev. Can. Rech. Vet. 73, 151-156
CHO, S., FOSSLER, C. P., DIEZ-GONZALEZ, F., WELLS, S. J., HEDBERG, C. W.,
KANEENE, J. B., RUEGG, P. L., WARNICK, L. D. & BENDER, J. B. (2013) Herd-level risk
factors associated with fecal shedding of Shiga toxin-encoding bacteria on dairy farms in
Minnesota, USA. Can. Vet. J. 54, 693-697

112

COBBAUT, K., BERKVENS, D., HOUF, K., DE DEKEN, R. & DE ZUTTER, L. (2009)
Escherichia coli O157 prevalence in different cattle farm types and identification of potential risk
factors. J. Food Prot. 72, 1848-1853
COBBOLD, R. N., RICE, D. H., SZYMANSKI, M., CALL, D. R. & HANCOCK, D. D. (2004)
Comparison of Shiga-Toxigenic Escherichia coli Prevalences among Dairy, Feedlot, and CowCalf Herds in Washington State. Appl. Environ. Microbiol. 70, 4375-4378
CRIM, S., IWAMOTO, M., HUANG, J., GRIFFIN, P., GILLISS, D., CRONQUIST, A.,
CARTTER, M., TOBIN-D'ANGELO, M., BLYTHE, D., SMITH, K., LATHROP, S., ZANSKY,
S., CIESLAK, P., DUNN, J., HOLT, K., LANCE, S., TAUXE, R. & HENAO, O. (2014)
Incidence and trends of infection with pathogens transmitted commonly through foodFoodborne diseases active surveillance network, 10 U.S. sites, 2006-2013. MMWR CDC 63
DOHOO, I. R., MARTIN, W. & STRYHN, H. (2010) Veterinary Epidemiologic Research, VER
Inc.
DUNN, J. (2003) The epidemiology of STEC O157:H7 in Louisiana dairy cattle, beef cattle, and
white-tailed deer In Pathobiological Sciences. Louisiana, Louisiana State University
DUNN, J. R., KEEN, J. E. & THOMPSON, R. A. (2004) Prevalence of shiga-toxigenic
Escherichia coli O157:H7 in adult dairy cattle. J. Am. Vet. Med. Assoc. 224, 1151-1158
EDRINGTON, T. S., SCHULTZ, C. L., GENOVESE, K. J., CALLAWAY, T. R., LOOPER, M.
L., BISCHOFF, K. M., MCREYNOLDS, J. L., ANDERSON, R. C. & NISBET, D. J. (2004)
Examination of heat stress and stage of lactation (early versus late) on fecal shedding of E. coli
O157:H7 and Salmonella in dairy cattle. Foodborne Pathog Dis 1, 114-119
FAGAN, P. K., HORNITZKY, M. A., BETTELHEIM, K. A. & DJORDJEVIC, S. P. (1999)
Detection of Shiga-Like Toxin (stx1 and stx2), Intimin (eaeA), and Enterohemorrhagic
Escherichia coli (EHEC) Hemolysin (EHEC hlyA) Genes in Animal Feces by Multiplex PCR.
Appl. Environ. Microbiol. 65, 868-872
FITZGERALD, A. C., EDRINGTON, T. S., LOOPER, M. L., CALLAWAY, T. R.,
GENOVESE, K. J., BISCHOFF, K. M., MCREYNOLDS, J. L., THOMAS, J. D., ANDERSON,
R. C. & NISBET, D. J. (2003) Antimicrobial susceptibility and factors affecting the shedding of
E. coli O157:H7 and Salmonella in dairy cattle. Lett. Appl. Microbiol. 37, 392-398
FRANK, C., WERBER, D., CRAMER, J. P., ASKAR, M., FABER, M., AN DER HEIDEN, M.,
BERNARD, H., FRUTH, A., PRAGER, R., SPODE, A., WADL, M., ZOUFALY, A., JORDAN,
113

S., KEMPER, M. J., FOLLIN, P., MULLER, L., KING, L. A., ROSNER, B., BUCHHOLZ, U.,
STARK, K. & KRAUSE, G. (2011) Epidemic profile of Shiga-toxin-producing Escherichia coli
O104:H4 outbreak in Germany. N. Engl. J. Med. 365, 1771-1780
GARBER, L. (1999) Factors associated with fecal shedding of verotoxin-producing Escherichia
coli O157 on dairy farms. J. Food Prot 62, 307-312
GAUTAM, R., BANI-YAGHOUB, M., NEILL, W. H., DOPFER, D., KASPAR, C. &
IVANEK, R. (2011) Modeling the effect of seasonal variation in ambient temperature on the
transmission dynamics of a pathogen with a free-living stage: example of Escherichia coli
O157:H7 in a dairy herd. Prev. Vet. Med. 102, 10-21
GOULD, L. H., MODY, R. K., ONG, K. L., CLOGHER, P., CRONQUIST, A. B., GARMAN,
K. N., LATHROP, S., MEDUS, C., SPINA, N. L., WEBB, T. H., WHITE, P. L., WYMORE, K.,
GIERKE, R. E., MAHON, B. E. & GRIFFIN, P. M. (2013) Increased recognition of non-O157
Shiga toxin-producing Escherichia coli infections in the United States during 2000-2010:
epidemiologic features and comparison with E. coli O157 infections. Foodborne Pathog. Dis. 10,
453-460
HANCOCK, D. D., BESSER, T. E., KINSEL, M. L., TARR, P. I., RICE, D. H. & PAROS, M.
G. (1994) The Prevalence of Escherichia coli O157.H7 in Dairy and Beef Cattle in Washington
State. Epidemiol. Infect. 113, 199-207
HEUVELINK, A. E., VAN DEN BIGGELAAR, F. L. A. M., ZWARTKRUIS-NAHUIS, J. T.
M., HERBES, R. G., HUYBEN, R., NAGELKERKE, N., MELCHERS, W. J. G., MONNENS,
L. A. H. & DE BOER, E. (1998) Occurrence of Verocytotoxin-ProducingEscherichia coli O157
on Dutch Dairy Farms. J. Clin. Microbiol. 36, 3480-3487
HUSSEIN, H. S. (2007) Prevalence and pathogenicity of Shiga toxin-producing Escherichia coli
in beef cattle and their products. J. Anim. Sci. 85, E63-E72
JASON, V., RAJKOVIC, A., BAERT, L., DEBEVERE, J. & UYTTENDALE, M. (2009)
Comparison of enrichment conditions for rapid detection of low numbers of sublethally injured
Escherichia coli O157 in food. J. Food Prot. 72, 1862-1868
KANG, E., HWANG, S. Y., KWON, K. H., KIM, K. Y., KIM, J. H. & PARK, Y. H. (2014)
Prevalence and characteristics of Shiga toxin-producing Escherichia coli (STEC) from cattle in
Korea between 2010 and 2011. Journal of Veterinary Science 15, 369-379

114

KARMALI, M., PETRIC, M., STEELE, B. & LIM, C. (1983) SPORADIC CASES OF
HAEMOLYTIC-URAEMIC SYNDROME ASSOCIATED WITH FAECAL CYTOTOXIN
AND CYTOTOXIN-PRODUCING ESCHERICHIA COLI IN STOOLS. The Lancet 321, 619620
KISTEMANN, T., ZIMMER, S., VAGSHOLM, I. & ANDERSSON, Y. (2004) GIS-supported
investigation of human EHEC and cattle VTEC O157 infections in Sweden: Geographical
distribution, spatial variation and possible risk factors. Epidemiol. Infect. 132, 495-505
KONDO, S., HOAR, B. R., VILLANUEVA, V., MANDRELL, R. E. & ATWILL, E. R. (2010)
Longitudinal prevalence and molecular typing of Escherichia coli O157:H7 by use of multiplelocus variable-number tandem-repeat analysis and pulsed-field gel electrophoresis in fecal
samples collected from a range-based herd of beef cattle in California. Am. J. Vet. Res. 71,
1339-1347
KUHNERT, P., DUBOSSON, C. R., ROESCH, M., HOMFELD, E., DOHERR, M. G. &
BLUM, J. W. (2005) Prevalence and risk-factor analysis of Shiga toxigenic Escherichia coli in
faecal samples of organically and conventionally farmed dairy cattle. Vet. Microbiol. 109, 37-45
MCDANIEL, T. K., JARVIS, K. G., DONNENBERG, M. S. & KAPER, J. B. (1995) A genetic
locus of enterocyte effacement conserved among diverse enterobacterial pathogens. Proc. Natl.
Acad. Sci. U. S. A. 92, 1664-1668
MECHIE, S. C., CHAPMAN, P. A. & SIDDONS, C. A. (1997) A Fifteen Month Study of
Escherichia coli O157:H7 in a Dairy Herd. Epidemiol. Infect. 118, 17-25
MENRATH, A., WIELER, L. H., HEIDEMANNS, K., SEMMLER, T., FRUTH, A. &
KEMPER, N. (2010) Shiga toxin producing Escherichia coli: identification of non-O157:H7super-shedding cows and related risk factors. Gut Pathogens 2
O'BRIEN, A. D., NEWLAND, J. W., MILLER, S. F., HOLMES, R. K., SMITH, H. W. &
FORMAL, S. B. (1984) Shiga-like toxin-converting phages from Escherichia coli strains that
cause hemorrhagic colitis or infantile diarrhea. Science (New York, N.Y.) 226, 694-696
POLIFRONI, R., ETCHEVERRIA, A. I., SANZ, M. E., CEPEDA, R. E., KRUGER, A.,
LUCCHESI, P. M., FERNANDEZ, D., PARMA, A. E. & PADOLA, N. L. (2012) Molecular
characterization of Shiga toxin-producing Escherichia coli isolated from the environment of a
dairy farm. Curr. Microbiol. 65, 337-343

115

REILLY, A. (1998) Prevention and control of enterohaemorrhagic Escherichia coli (EHEC)
infections: memorandum from a WHO meeting. WHO Consultation on Prevention and Control
of Enterohaemorrhagic Escherichia coli (EHEC) Infections. Bull. World Health Organ. 76, 245255
RENTER, D. G., BOHAYCHUK, V., VAN DONKERSGOED, J. & KING, R. (2007) Presence
of non-O157 Shiga toxin-producing Escherichia coli in feces from feedlot cattle in Alberta and
absence on corresponding beef carcasses. Can. J. Vet. Res. 71, 230-235
ROUNDS, J., RIGDON, C., MUHL, L., FORSTNER, M., GT, D., KOZIOL, B. & ET AL
(2012) Non-O157 Shiga toxin-producing Escherichia coli associated with venison. Emerg.
Infect. Dis. 18
SARGEANT, J. M., SANDERSON, M. W., SMITH, R. A. & GRIFFIN, D. D. (2003)
Escherichia coli O157 in feedlot cattle feces and water in four major feeder-cattle states in the
USA. Prev. Vet. Med. 61, 127-135
SCALLAN, E., HOEKSTRA, R. M., ANGULO, F. J., TAUXE, R. V., WIDDOWSON, M. A.,
ROY, S. L., JONES, J. L. & GRIFFIN, P. M. (2011) Foodborne illness acquired in the United
States--major pathogens. Emerg. Infect. Dis. 17, 7-15
SCHEUTZ, F., TEEL, L. D., BEUTIN, L., PIERARD, D., BUVENS, G., KARCH, H.,
MELLMANN, A., CAPRIOLI, A., TOZZOLI, R., MORABITO, S., STROCKBINE, N. A.,
MELTON-CELSA, A. R., SANCHEZ, M., PERSSON, S. & O'BRIEN, A. D. (2012)
Multicenter evaluation of a sequence-based protocol for subtyping Shiga toxins and
standardizing Stx nomenclature. J. Clin. Microbiol. 50, 2951-2963
SMITH, D. R., MOXLEY, R. A., CLOWSER, S. L., FOLMER, J. D., HINKLEY, S.,
ERICKSON, G. E. & KLOPFENSTEIN, T. J. (2005) Use of rope devices to describe and
explain the feedlot ecology of Escherichia coli O157:H7 by time and place. Foodborne Pathog
Dis 2, 50-60
TARR, C. L., LARGE, T. M., MOELLER, C. L., LACHER, D. W., TARR, P. I., ACHESON, D.
W. & WHITTAM, T. S. (2002) Molecular characterization of a serotype O121:H19 clone, a
distinct Shiga toxin-producing clone of pathogenic Escherichia coli. Infect. Immun. 70, 68536859
USDA (2003) Escherichia coli O157 on U.S. dairy operations. In Info Sheet, United States
Department of Agriculture

116

VANAJA, S., JANDHYALA, D., MALLICK, E., LEONG, J. & BALASUBRAMANIAN, S.
(2013) Enterohemorrhagic and other Shigatoxin-producing Escherichia coli. In Escherichia coli:
Pathotypes and Principles of Pathogenesis. 2nd edn. Ed M. DONNENBERG, Elsevier Inc.

117

CHAPTER 3

ASSOCIATION OF BOVINE LEUKEMIA VIRUS AND MYCOBACTERIUM AVIUM
SUBSP. PARATUBERCULOSIS WITH SHEDDING OF SHIGA TOXIN-PRODUCING
ESCHERICHIA COLI

Abstract

Bovine leukemia virus (BLV) is a retrovirus that causes enzootic bovine leukosis in cattle
and Mycobacterium avium subsp. paratuberculosis (MAP) is the etiologic agent of Johnes’
disease in cattle. Both diseases are chronic in nature that leads to disruption of normal
immunological or physiological processes. Cattle are the major reservoir of Shiga toxinproducing Escherichia coli (STEC), a major cause of foodborne illness in humans. We tested the
hypothesis that cattle infected with BLV and/or MAP are more likely to shed STEC. We
conducted a cross-sectional study during the summers of 2011 and 2012 in 11 Michigan cattle
herds. A fecal sample from each animal was collected for STEC culture and multiplex PCR for
stx1, stx2, and eaeA was used to screen suspect colonies for STEC confirmation. Antibody
detection ELISA assays for BLV and MAP were used to screen serum from each animal. Blood
samples were collected from a subsample (n=497) to quantify the percentage of lymphocytes,
monocytes and neutrophils using flow cytometry. Of the animals sampled, 34.9% were BLV
positive while 2.7% were MAP positive and 16% were shedding STEC. Dairy herds had a higher
frequency of BLV and MAP than did beef herds, but beef herds had more STEC. Neither BLV
nor MAP was associated with STEC shedding. We also observed no association between
118

percentage of white blood cells and STEC status. Although controlling both BLV and MAP is
important for overall herd health and productivity, controlling BLV and MAP will not likely
have an impact on STEC shedding in cattle.

119

Introduction

Bovine leukemia virus (BLV) and Mycobacterium avium subsp. paratuberculosis (MAP)
have been associated with a suppressed immune response in cattle. Secondary health issues are
often associated with BLV and MAP because of their chronic and potentially debilitating nature
(Bartlett, et al 2014; Gonda, et al 2007).
Bovine leukemia virus (BLV) is a retrovirus that causes enzootic bovine leukosis. Most
animals infected with BLV never develop clinical signs, but 30% of BLV carriers will develop a
persistent lymphocytosis and less than 5% will develop malignant lymphosarcoma (Erskine, et al
2012). BLV affects host defense mechanisms by disrupting the homeostasis of normal
lymphocyte proliferation and programmed cell death, in both B-cells and T-cells, which can
increase susceptibility to infectious diseases (Bartlett, et al 2014). According to USDA surveys,
83% of US dairy herds have at least one infected animal (USDA 2010). The within-herd BLV
prevalence ranges from 23% to 46% in affected dairy herds (Ott, et al 2003; Sargeant, et al 1997;
Trono, et al 2001).
Mycobacterium avium subsp. paratuberculosis (MAP) causes Johnes’ disease (JD).
Initial MAP infection most likely occurs at a very early age (<6 months) yet clinical disease does
not normally occur until after 2 years of age (Blood, et al 1989). This pathogen becomes
localized in the mucosa of the small intestine and associated lymph nodes. MAP elicits T-cell
activation and clonal expansion that causes alterations in the intestine’s histology and physiology
(Manning and Collins 2001) and produces changes in the intestinal microbiota composition. An
estimated 50% of Michigan dairy herds, and 68% nationally, have MAP-infected animals
(Pillars, et al 2009).

120

Shiga toxin-producing Escherichia coli (STEC) is an important cause of foodborne
illness and hemolytic uremic syndrome (HUS) in children (Bolton 2011) which can result in
kidney failure and death in some cases. Cattle are a major STEC reservoir (Bolton 2011; Ferens
and Hovde 2011; Gyles 2007), though it does not typically cause symptomatic infections
(Kuhnert, et al 2005) except for contributing to diarrhea/dysentery in a subset of calves (Gyles
and Fairbrother 2010; Vande Walle, et al 2013). Food and water contaminated with feces is the
most common source of human exposure. For this reason, the USDA-FSIS considers STEC an
adulterant of all raw non-intact beef and raw intact beef intended for use in raw non-intact
products under the Federal Meat Inspection Act (21 U.S.C. 601(m)(1)).
Because of the chronic nature of both BLV and MAP infections and the immunosupresion
and gastrointestinal disruption, we hypothesized that both infections may have an impact on
STEC colonization and shedding in cattle. Therefore, the objective of this study was to determine
if cattle infected with MAP and/or BLV are at higher risk for shedding STEC in their feces and
thus could increase STEC contamination in the human food chain.

121

Materials and methods

1. Animal selection

A total of eleven Michigan cattle herds (6 dairy farms and 5 beef feedlots) were sampled
for STEC, BLV and MAP. The herds were chosen based on convenience and willingness to
participate in the study and were sampled during the spring-summer months of 2011 and 2012.
The number of cattle sampled was based on the type and size of herds. In the beef feedlots, all of
the animals present in each feedlot were sampled, while the quantity of the animals sampled in
the dairy herds depended on the size of the herd. All adult cows were sampled in each dairy herd
with less than 175 animals. For herds with more than 175 animals, a convenience sample of 175
animals was selected from cattle in the different management groups. This study was approved
by the Michigan State University Institutional Animal Care and Use Committee (AN12/10-22300).

2. Fecal sample collection and analysis

Fresh fecal samples were collected per rectum from 1,108 animals; a total of 724 dairy
and 384 beef cattle were sampled. For the first four herds, samples were transported to the
laboratory on ice where they were stored at 4°C and then processed within 48 hours. For the
other remaining seven herds, samples were transported to the laboratory in a cooler without ice
and processed immediately. This change in protocol was done to optimize our ability to detect
STEC in the fecal samples.
122

Samples were cultured for STEC by first enriching in Escherichia coli (EC) broth (Oxoid
Ltd.; Waltham, MA) supplemented with novobiocin (8mg/l), rifampin (2mg/l) and potassium
tellurite (1mg/l) for 25 hours at 42°C (Jason, et al 2009) and then plating on STEC
CHROMagar™ (CHROMagar, Paris, France) ans sorbitol MacConkey (SMAC) agar. The EC
broth was also used to perform immunomagnetic separation (IMS) targeting E.coli O157:H7
using Dynabeads® MAX E.coli O157 (Invitrogen Corporation, California, USA). The IMS
O157 protocol was followed by subculture to O157 CHROMagar (CHROMagar, Paris, France)
and SMAC agar. Up to 20 suspect colonies were selected from each of the three agar plates for
multiplex PCR targeting the Shiga toxin genes (stx1, stx2) and eaeA (intimin) for STEC
confirmation.
Bacterial colonies with at least one stx gene were considered to be STEC, and individual
animals were considered positive if any STEC was recovered from the fecal sample. A total of
1,096 animals (718 dairy, 378 beef) were included in the final analysis, 12 animals were
excluded because the STEC culture was missing.

3. Blood collection and analysis

At least three milliliters of blood were collected into serum separator vaccutainer tubes
from the coccygeal or jugular vein. Serum was separated by centrifugation and submitted to the
Diagnostic Center for Population and Animal Health at Michigan State University. Antibody
detection ELISA assays specific for BLV (Bovine Leukemia Virus Antibody Test Kit, VMRD,
Pullman, WA) and MAP (Paracheck, Prionics, USA Inc, Omaha, NE) were used to screen serum
from each animal. Animals were classified as either positive or negative based on the

123

manufacturer’s suggested criteria, and the optical density was recorded.
Whole blood samples (n=497) were collected in vaccutainer tubes containing Acid
Citrate Dextrose (ACD) from six herds in year two, which included 290 dairy cattle and 207 beef
cattle. These samples were used to quantify the percentage of lymphocytes, monocytes and
neutrophils, using a Becton Dickinson FACSCalibur™ flow cytometer (San Jose, CA, USA).
Cell profiles were analyzed based on size and granularity, and the level of fluorescence caused
by indirect immunofluorescence with primary monoclonal antibodies (Davis and Hamilton
1993).

4. Data collection and analysis

The daily average, maximum, and minimum ambient temperatures 1-5 days before
sampling were collected from the closest weather station (Quality Controlled Local
Climatological Data (NOAA). For dairy cattle, production data including the lactation number
(parity) and days in milk (DIM) were also recorded.
All data was analyzed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA). The
dependent variable was the STEC status (negative or positive) of each animal. The independent
variables were BLV status (binary and continuous) and MAP status (binary and continuous);
percentage of neutrophils, lymphocytes and the lymphocyte to monocyte ratio.
The distribution of the independent variables was explored as well as the presence of
confounders. The point of significance to be incorporated into the initial multivariable model was
1.5 with a correlation coefficient of 0.9 (Dohoo, et al 2010). Herd was also incorporated in the
multivariable models as a random effect as herds varied in management strategies and

124

geographic location. Both univariate and multivariate associations were examined using logistic
regression and generalized linear mixed models (GLIMMIX). The point of significance to be
incorporated in the final multivariable model was 0.05. Odds ratios (OR) and 95% confidence
intervals (95% CI) were estimated for each variable. A backward process was used to evaluate
the significance of each variable, and biologic rationale was considered prior to introducing
variables into the final model. See Table 2.7 for explanation of abbreviations used for variables.

125

Results

1. Descriptive statistics

The within-herd animal STEC prevalence ranged from 6% to 54%. The STEC prevalence
was 13% (95/715) in dairy cattle and 21% (80/378) in beef cattle. Feedlot cattle had a higher risk
of being STEC positive than did dairy cattle (OR: 1.8; 95% CI: 1.25-2.47). For BLV, the withinherd prevalence ranged from 0% to 79%. Two of the beef herds (n=129 cattle) were negative for
BLV. The individual animal prevalence of BLV among dairy and beef herds was 48% (344/717)
and 10% (38/378), respectively. Dairy cattle were 8.3 times (95% CI: 5.68- 12.22) more likely to
be positive for BLV than beef cattle (P=<0.0001). The within-herd MAP prevalence was 0% to
8%; three beef herds (n=161 animals) were negative for MAP. The individual animal prevalence
of MAP for dairy and beef herds was 3% (23/717) and 2% (7/378), respectively. Dairy cattle had
an increased frequency of MAP, however, this difference was not significant (OR: 1.76; 95% CI:
0.72- 4.91; P=0.2434). BVDV was not detected in any of the cattle tested. Data is summarized in
Table 3.1.

1.2 Immune system cells

When looking at the association between BLV status, STEC status and the percentage of
neutrophils, lymphocytes and the lymphocyte to monocyte ratio (L:M), only the ratio was
significantly associated with BLV status. As the L:M increased per unit, the likelihood of being
BLV positive also increased (OR: 1.2; 95% CI: 1.09- 1.24; p-value: < 0.0001). By contrast,
126

neither MAP nor STEC status was associated with the percentage of neutrophils, lymphocytes
and L:M (Table 3.2).

2. Univariable models

Using univariable models, there was no association between STEC and BLV when cattle
herds were analyzed together (p-value: 0.5406) or separately as dairy herds (p-value: 0.7228) or
beef herds (p-value: 0.9084). Similarly, there was no association between STEC and MAP (pvalue: 0.3126) (Table 3.3). Also, there was no association between STEC status and the
percentage of neutrophils, lymphocytes or the L:M ratio after including herd as a random effect
(Table 3.4). There was a significant association between STEC and average maximum
temperature 1-5 days before sampling and between STEC and year.

3. Multivariable models

There was no association between STEC and either BLV or MAP status. When BLV or
MAP status was examined as a continuous variable represented by the optical density of the
ELISA assays, there was still no association between STEC status and either BLV or MAP.
Additionally, production system type, maximum average temperature 1-5 days before sampling,
and county were examined in the model as potential confounders (Table 3.5). When analyzed
separately by production type, there was no association between STEC and MAP in dairy herds
either between STEC and BLV (p-value: 0.5936) (Table 3.6). The model did not converge in the

127

case of beef herds because few herds and few animals within herds were positive for both BLV
and MAP, so no conclusion could be made.

128

Discussion

In this study, we examined a population of beef and dairy cattle in Michigan to determine
whether there was an association between STEC shedding and two common chronic infections
that affect the immune system and the gastrointestinal function. Although immune suppressive
effects have been widely reported for both BLV and MAP, we did not find a higher rate of STEC
colonization among this population of BLV- or MAP-positive cattle.
A possible explanation for the lack of association could be that the immune disruption
due to BLV and MAP is generally only present in a subset of infected animals, so an effect may
have been diluted by a large number of immune competent cattle in our study, as the cattle
sampled in this study were cattle with not visual clinical signs of disease.
The lack of association between the percentage of neutrophils, lymphocytes and L:M
ratio and STEC (Table 3.4) in our study differ in part with previous studies (Hoffman, et al
2006; Vande Walle, et al 2013), which may be because we were examining natural STEC
infections. Animals infected with STEC did not have a noticeably different white blood cell
profile, further supporting the notion that STEC colonization does not stimulate major immune
responses in cattle. Because STEC is believe to be a commensal with totally asymptomatic
infection in cattle (Wells, et al 1991).
However, based on histological changes in the intestine of the colonized cattle, Vande
Walle et al. (2013) have suggested that E. coli O157:H7 represent a bovine pathogen. There is
also some evidence of innate and adaptive bovine immune response to E. coli O157:H7, such as
the production of pro-inflammatory cytokines and antibodies against secreted proteins (Vande
Walle, et al 2013). It has also been suggested that E. coli O157:H7 may cause immunosupresion

129

in cattle, preventing the onset of an antigen-specific cellular immune response. For example,
calves infected with Stx2+ O157 did not develop a lymphoproliferative response to heat-killed
Stx2+ O157 (Hoffman, et al 2006). Nevertheless, the pathogenic properties of STEC in cattle
appear to be minimal. It is therefore possible that immune suppression due to either BLV or
MAP may be largely inconsequential as a determinant of STEC colonization and shedding. More
studies in this area will help to understand the interaction between STEC and the bovine immune
system.
In summary, the lack of association between STEC and these two chronic nature diseases
could be due to the interaction between STEC, a commensal, and the immune system of cattle;
the immune system is no affected by the presence of BLV/MAP. Also could be that the cattle we
sampled was at a stage of BLV/MAP infection that does not compromise or disrupt their immune
system and the gastrointestinal function, yet. As consequence STEC have a normal interaction
with the immune system and cannot take advantage of them.
Because diet, health, and management practices vary widely among herds, herd was
included in the univariable and multivariable models as a random effect. Using herd as a random
effect allowed us to control for several known confounders and unknown confounders (Dohoo, et
al 2010). The established association between warm temperatures and STEC shedding (Cobbold,
et al 2004; Dunn, et al 2004; Gautam, et al 2011; Kondo, et al 2010; Smith, et al 2005) was the
main reason to incorporate maximum average temperature 1-5 days before sampling as a
confounder.
In the univariate and multivariable analyses, the significant difference in STEC shedding
between 2011 and 2012 may have been because that 2012 was warmer. According with NOAA,
“In 2012, the contiguous United States average annual temperature of 55.3°F was 3.2°F above

130

the 20th century average, and was the warmest year in the 1895-2012 period of record for the
nation” (NOAA National Climatic Data Center). This finding confirms the association between
STEC shedding and warm temperature/seasonality reported by several studies (Dunn, et al
2004).
Due to the high prevalence of BLV and MAP in Michigan cattle herds and their known
immune suppression effects, reducing BLV and MAP was seen as potential interventions to
control STEC shedding in the human food chain. Although controlling BLV and MAP is
important for overall herd health and productivity, based on this study, it does not appear that
controlling BLV and MAP would be an effective way to reduce STEC shedding in cattle.

131

APPENDIX

132

Table 3.1. Prevalence of bovine leukemia virus (BLV), Mycobacterium avium subsp.
paratuberculosis (MAP) and STEC by herd. * B=beef, D=dairy

Herd*

Total animals
sampled

BLV Positive
(%)

MAP Positive
(%)

STEC Positive
(%)

B1
D2
B3
D4
D6
D7
B8
D9
D10
B11
B12

134
148
32
174
94
100
54
100
101
83
75

9 (7)
43 (29)
5 (16)
56 (32)
53 (56)
54 (54)
0
58 (58)
80 (79)
24 (29)
0

2 (1)
3 (2)
0
1 (1)
4 (4)
5 (5)
0
2 (2)
8 (8)
5 (6)
0

11 (8)
13 (9)
3 (9)
24 (14)
6 (6)
13 (13)
29 (54)
28 (28)
11 (11)
13 (16)
24 (32)

Total

1,095

382 (35)

30 (3)

175 (16)

133

Table 3.2. Mean and Standard deviation (SD) for the percentage of neutrophils, lymphocytes and
L: M in cattle positive and negative to bovine leukemia virus (BLV) and Mycobacterium avium
subsp. paratuberculosis (MAP) according to STEC status.

Cell type

BLV positive

BLV negative

MAP positive

MAP negative

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

Neutrophils (%)
STEC positive
STEC negative

15.56 (14.27)
16.85 (14.81)

18.53 (12.58)
20.81 (25.16)

16.55 (4.65)
7.79 (0.28)

17.58 (13.27)
19.49 (21.61)

Lymphocytes (%)
STEC positive
STEC negative

67.87 (16.06)
64.13 (16.43)

60.74 (12.84)
62.30 (39.80)

66.01 (8.29)
68.17 (14.79)

63.01 (14.39)
62.84 (32.14)

L: M
STEC positive
STEC negative

6.8 (5.21)
5.58 (4.92)

4.73 (2.64)a
3.96 (1.97)b

7.60 (6.15)
7.85 (10.19)

5.36 (3.76)a
4.54 (2.98)b

TOTAL

208

289

477

20

134

Table 3.3. Univariable analysis to evaluate the association between risk factors and the dependent variable for STEC shedding.

Characteristic

No. (%) with
characteristic

No. (%) with STEC

p-value

OR

95% CI

MAP
Negative
Positive

30 (2.75)
1062 (97.25)

2 (6.67)
173 (16.29)

0.5012

1.672
ref

0.373- 7.504
ref

BLV
Negative
Positive

712 (65.20)
380 (34.8)

125 (17.56)
50 (13.16)

0.5406

1.139
ref

0.750- 1.730
ref

Beef
Beef
Dairy

378 (34.58)
715 (65.42)

80 (21.16)
95 (13.29)

0.2356

1.832
ref

0.673- 4.986
Ref

Temperature max 1-5 days
>28.9 C
≤ 28.9 C

502 (45.93)
591 (54.07)

116 (23.11)
59 (9.98)

0.0073

3.055
ref

1.349- 6.827
Ref

Year
2011
2012

581 (53.16)
512 (46.84)

57 (9.81)
118 (23.05)

0.0095

0.333
ref

0.145- 0.764
Ref

135

Table 3.4. Univariable analysis to evaluated ELISA for bovine leukemia virus (BLV) status, Mycobacterium avium subsp.
paratuberculosis (MAP) status and percentage of neutrophils, lymphocytes, and lymphocytes monocytes ratio for their association
with STEC shedding in animals.

Characteristic

No. (%) with characteristic

No. (%) with STEC

p-value

OR

95% CI

MAP
Negative
Positive

492 (96.09)
20 (3.91)

116 (23.58)
2 (10)

0.5012

1.672
Ref

0.373- 7.504
Ref

BLV
Negative
Positive

297 (58.01)
215 (41.99)

80 (26.94)
38 (17.67)

0.7811

0.925
Ref

0.532- 1.608
Ref

Neutrophils

0.3565

0.993

0.979- 1.008

Lymphocytes

0.8422

0.999

0.991- 1.007

L:M

0.1800

1.042

0.981- 1.106

136

Table 3.5. Final multivariable model for both beef and dairy herds to evaluate bovine leukemia virus (BLV) and Mycobacterium
avium subsp. paratuberculosis (MAP) status as determinants of STEC shedding.

Characteristic

Estimate

Standard Error

p-value

OR

95% CI

Intercept

-3.2813

0.7935

0.0033

MAP
Negative
Positive

0.7214
Ref

0.7466
Ref

0.3341

2.057
Ref

0.475- 8.902
Ref

BLV
Negative
Positive

0.08792
Ref

0.2158
Ref

0.6838

1.092
Ref

0.715- 1.667
Ref

Temperature max 1-5 days
> 28.9 C
< 28.9 C

1.1547
Ref

0.3533
Ref

0.0011

3.173
Ref

1.586- 6.346
Ref

Beef
Beef
Dairy

0.6778
Ref

0.3663
Ref

0.0645

1.970
Ref

0.960- 4.041
Ref

137

Table 3.6. Final multivariable model for dairy herds to evaluated bovine leukemia virus (BLV) and Mycobacterium avium subsp.
paratuberculosis (MAP) status as determinants of STEC shedding.

Characteristic

Estimate

Standard Error

p-value

MAP
Negative
Positive

0.1383
ref

0.7695
Ref

BLV
Negative
Positive

-0.2822
ref

Temperature max 1-5 days
> 28.9 C
< 28.9 C

OR

95% CI

0.8575

1.148
ref

0.457- 8.604
Ref

0.2653
Ref

0.2879

0.754
ref

0.695- 1.640
Ref

0.8761
ref

0.4005
Ref

0.0291

2.401
ref

0.460- 7.319
Ref

Lactation
First
2 or more

0.6285
ref

0.2545
Ref

0.0138

1.875
ref

1.137- 3.090
Ref

DIM
0
1-31
>31

-0.3612
1.4126
ref

0.6334
0.3149
Ref

<.0001

0.697
4.107
ref

0.201- 2.417
2.213- 7.621
Ref

138

Individual p-value

0.5687
<.0001

REFERENCES

139

REFERENCES

BARTLETT, P. C., SORDILLO, L. M., BYREM, T. M., NORBY, B., GROOMS, D. L.,
SWENSON, C. L., ZALUCHA, J. & ERSKINE, R. J. (2014) Options for the control of bovine
leukemia virus in dairy cattle. J. Am. Vet. Med. Assoc. 244, 914-922
BLOOD, D. C., RADOSTITS, O. M., ARUNDEL, J. H. & GAY, C. C. (1989) Veterinary
Medicine: A textbook of the diseases of cattle, sheep, pigs, goats and horses. Oxford, Great
Britain, Bailliere Tindall
BOLTON, D. J. (2011) Verocytotoxigenic (Shiga toxin-producing) Escherichia coli: virulence
factors and pathogenicity in the farm to fork paradigm. Foodborne Pathog Dis 8, 357-365
COBBOLD, R. N., RICE, D. H., SZYMANSKI, M., CALL, D. R. & HANCOCK, D. D. (2004)
Comparison of Shiga-Toxigenic Escherichia coli Prevalences among Dairy, Feedlot, and CowCalf Herds in Washington State. Appl. Environ. Microbiol. 70, 4375-4378
DAVIS, W. C. & HAMILTON, M. J. (1993) Identification of leukocyte subpopulations by flow
cytometry. In Monographs in Animal Immunology. Ed O. BAITA. Blackburg, VA, BAR-LAB,
Inc. pp b5-1--b5-12
DOHOO, I. R., MARTIN, W. & STRYHN, H. (2010) Veterinary Epidemiologic Research, VER
Inc.
DUNN, J. R., KEEN, J. E. & THOMPSON, R. A. (2004) Prevalence of shiga-toxigenic
Escherichia coli O157:H7 in adult dairy cattle. J. Am. Vet. Med. Assoc. 224, 1151-1158
ERSKINE, R. J., BARTLETT, P. C., BYREM, T. M., RENDER, C. L., FEBVAY, C. &
HOUSEMAN, J. T. (2012) Herd-level determinants of bovine leukaemia virus prevalence in
dairy farms. J. Dairy Res. 79, 445-450
FERENS, W. A. & HOVDE, C. J. (2011) Escherichia coli O157:H7: animal reservoir and
sources of human infection. Foodborne Pathog Dis 8, 465-487
GAUTAM, R., BANI-YAGHOUB, M., NEILL, W. H., DOPFER, D., KASPAR, C. &
IVANEK, R. (2011) Modeling the effect of seasonal variation in ambient temperature on the
transmission dynamics of a pathogen with a free-living stage: example of Escherichia coli
O157:H7 in a dairy herd. Prev. Vet. Med. 102, 10-21
140

GONDA, M., CHANG, Y., SHOOK, G., COLLINS, M. T. & KIRKPATRICK, B. (2007) Effect
of Mycobacterium paratuberculosis infection on production, reproduction, and health traits in US
Holsteins. Prev. Vet. Med. 80, 103-119
GYLES, C. L. (2007) Shiga toxin-producing Escherichia coli: an overview. J. Anim. Sci. 85,
E45-62
GYLES, C. L. & FAIRBROTHER, J. M. (2010) Escherichia coli. In Pathogenesis of Bacterial
Infections in Animals. Eds C. L. GYLES, J. F. PRESCOTT, J. G. SONGER, C. O. THOEN,
Blackwell Publishing
HOFFMAN, M. A., MENGE, C., CASEY, T. A., LAEGREID, W., BOSWORTH, B. T. &
DEAN-NYSTROM, E. A. (2006) Bovine immune response to shiga-toxigenic Escherichia coli
O157:H7. Clinical and vaccine immunology : CVI 13, 1322-1327
JASON, V., RAJKOVIC, A., BAERT, L., DEBEVERE, J. & UYTTENDALE, M. (2009)
Comparison of enrichment conditions for rapid detection of low numbers of sublethally injured
Escherichia coli O157 in food. J. Food Prot. 72, 1862-1868
KONDO, S., HOAR, B. R., VILLANUEVA, V., MANDRELL, R. E. & ATWILL, E. R. (2010)
Longitudinal prevalence and molecular typing of Escherichia coli O157:H7 by use of multiplelocus variable-number tandem-repeat analysis and pulsed-field gel electrophoresis in fecal
samples collected from a range-based herd of beef cattle in California. Am. J. Vet. Res. 71,
1339-1347
KUHNERT, P., DUBOSSON, C. R., ROESCH, M., HOMFELD, E., DOHERR, M. G. &
BLUM, J. W. (2005) Prevalence and risk-factor analysis of Shiga toxigenic Escherichia coli in
faecal samples of organically and conventionally farmed dairy cattle. Vet. Microbiol. 109, 37-45
MANNING, E. J. & COLLINS, M. T. (2001) Mycobacterium avium subsp. paratuberculosis:
pathogen, pathogenesis and diagnosis. Rev. Sci. Tech. 20, 133-150
NOAA NATIONAL CLIMATIC DATA CENTER, S. O. T. C. N. O. F. A. Published online
December 2012
OTT, S. L., JOHNSON, R. & WELLS, S. J. (2003) Association between bovine-leukosis virus
seroprevalence and herd-level productivity on US dairy farms. Prev. Vet. Med. 61, 249-262

141

PILLARS, R., GROOMS, D. & KANEENE, J. B. (2009) Longitudinal study of the distribution
of Mycobacterium avium subsp. paratuberculosis in the environment of dairy herds in the
Michigan Johne's disease control demonstration herd project. Can. Vet. J. 50, 1039-1046
SARGEANT, J. M., KELTON, D. F., MARTIN, S. W. & MANN, E. D. (1997) Associations
between farm management practices, productivity, and bovine leukemia virus infection in
Ontario dairy herds. Prev. Vet. Med. 31, 211-221
SMITH, D. R., MOXLEY, R. A., CLOWSER, S. L., FOLMER, J. D., HINKLEY, S.,
ERICKSON, G. E. & KLOPFENSTEIN, T. J. (2005) Use of rope devices to describe and
explain the feedlot ecology of Escherichia coli O157:H7 by time and place. Foodborne Pathog
Dis 2, 50-60
TRONO, K. G., PEREZ-FILGUEIRA, D. M., DUFFY, S., BORCA, M. V. & CARRILLO, C.
(2001) Seroprevalence of bovine leukemia virus in dairy cattle in Argentina: comparison of
sensitivity and specificity of different detection methods. Vet. Microbiol. 83, 235-248
USDA (2010) Dairy 2007, heifer calf health and management practices on U.S. dairy operations,
2007. In NAHMS. Fort Collins, CO, USDA:APHIS:VS, CEAH
VANDE WALLE, K., VANROMPAY, D. & COX, E. (2013) Bovine innate and adaptive
immune responses against Escherichia coli O157:H7 and vaccination strategies to reduce faecal
shedding in ruminants. Vet. Immunol. Immunopathol. 152, 109-120
WELLS, J. G., SHIPMAN, L. D., GREENE, K. D., SOWERS, E. G., GREEN, J. H.,
CAMERON, D. N., DOWNES, F. P., MARTIN, M. L., GRIFFIN, P. M. & OSTROFF, S. M.
(1991) Isolation of Escherichia coli serotype O157:H7 and other Shiga-like-toxin-producing E.
coli from dairy cattle. J. Clin. Microbiol. 29, 985-989

142

CHAPTER 4

SHIGA TOXIN-PRODUCING ESCHERICHIA COLI ACQUISITION, LOSS AND
PERSISTENCE IN CATTLE

Abstract

Shiga toxin-producing Escherichia coli (STEC) are one of the most common foodborne
pathogens in the USA and other developed countries. Cattle are consider the main reservoir for
STEC and food and water contaminated with cattle feces are the most common sources of human
infections with STEC. In order to develop intervention strategies for STEC in cattle at the preharvest level, understanding the factors that influence STEC shedding are necessary. However,
the dynamics of shedding are poorly understood and potential risk factors that influence
shedding over time are unclear. The objective of this study was to generate information about the
dynamics of STEC shedding in cattle over time and to identify factors associated with
acquisition, persistence and loss of the bacteria.
Fecal STEC shedding of 149 individual cattle (46% dairy and 54% beef) from 11
different herds were analyzed four consecutive times separated by an average of 19 days.
Information on potential risk factors was collected at each time point. STEC prevalence, loss rate
and acquisition rate were calculated for each visit and an analysis for risk factors was conducted.
STEC shedding was intermittent; only 5 animals shed STEC continuously throughout the study.
Twenty- seven cattle shed STEC for at least two consecutives sample times. The average STEC

143

duration of shedding was 24 days. On the other hand, 28 cattle were negative throughout the
study period. The rate of STEC loss was higher than the rate of STEC acquisition in all visits.
STEC acquisition rate and STEC loss rate did not vary between visits; however, it varied
between herds.
The percentage of animals that at any time during the study lost STEC was significantly
different between years and herds, while the percentage of animals that at any time during the
study acquired STEC was significantly different only by type of production system.
We found herd as the only factor significantly associated with being continuously STEC
negative through the study period in dairy herds. Herd was the only factor significantly
associated with the rate of STEC new infections in dairy herds as well.
In summary STEC shedding is intermittent and our study did not clearly identify any
specific factor that influence STEC shedding over time. Herd, year, and type of production
system (beef or dairy) appear to have the largest influence on STEC dynamics. Understanding
the specific factors that influence STEC dynamics will allow designing effective intervention
strategies at preharvest to prevent and reduce STEC human infections.

144

Introduction

Shiga toxin-producing Escherichia coli (STEC) is one of the most common foodborne
pathogens in the USA and other developed countries and has a significant impact on public
health. Cattle are the primary reservoir for STEC and food or water contaminated with cattle
feces is the most common source of infection for humans (Kuhnert, et al 2005).The dynamics of
STEC shedding remains poorly understood (Robinson, et al 2009). A better understanding of the
dynamics of STEC shedding by cattle, and determining management and individual factors that
influence this dynamic would help to design intervention strategies at the pre-harvest level aimed
to reducing STEC shedding and ultimately human food contamination.
Different research groups have performed longitudinal prospective studies in cattle, but
the studies have not been consistent in their findings. Among the established findings are the
seasonality and intermittence or transient shedding of STEC in cattle (Callaway, et al 2013;
Hussein 2007; Smith, et al 2013; Widiasih, et al 2004).
Our hypothesis was that there are factors that influence the dynamics of STEC shedding.
The objectives of this study were to determine the rates of STEC acquisition, persistence and loss
in these individual cattle. Additionally, we aimed to identify herd management practices and
individual cattle characteristics for STEC acquisition or persistence in dairy.

145

Materials and methods

1. Herd selection

Dairy farms and beef feedlots were contacted and selected for inclusion in the study
based on proximity to MSU, adequate animal handling facilities and willingness to participate in
all phases of the study. It was also required that the farms had completed and detailed records for
each animal, regarding management and health history, such as antibiotic administration and
diseases. From twelve herds contacted initially, eleven agreed to participate; one herd chose not
to participate because of concerns regarding animal welfare. The farm owners provided written
informed consent to participate in the study. This study was approved by the Michigan State
University Institutional Animal Care and Use Committee (AN12/10-223-00).

2. Study design

The study design was composed of three phases. Phase I involved completing a
questionnaire designed to collect demographic information and data related to potential STEC
risk factors. Phase II focused on sampling a representative number of animals within each herd
and culturing feces for STEC isolation. Phase III consisted on a longitudinal sampling of a subset
of animals, that were also sampled in Phase II. The first two phases have been described in
detail in Chapter II of this dissertation. This study was performed during years 2011 and 2012.
The herds were visited and sampled between May 11th and October 18th of 2011 (n=5) or

146

between May 29th and October 16th of 2012 (n=6). The time between Phases II and III had a
range between 10 and 33 days and mean 19 days. Phase III consisted of three visits: first, second
and third, that were in a range between 14 to 28 days (mean 19 days), 7 to 29 days (mean 18
days), and 28 to 60 days (mean 17 days) apart, respectively. The total period of the study was
between 35 to 89 days (mean 53 days).

2.1 Sampling

In Phase II, fecal samples were collected from adult cattle within the herd. The number of
cattle sampled was based on the type of herd and number of cattle. In dairy herds with fewer than
175 animals, all adult cattle were sampled. In dairy herds with greater than 175 animals, a
convenience sample of 175 cattle was selected from different management groups. In the beef
feedlots, a “herd” consisted of a pen or management group and all cattle in that group were
sampled. In Phase III, fecal samples were collected from a subsample of the animals sampled in
Phase II. An equal number of Phase II STEC positives and negatives animals were chosen from
each herd for Phase III; their STEC status was determined based on PCR performed on raw feces
at the beginning and then from culture isolates. When a positive animal was selected a negative
animal from the same pen was also selected if not from a near pen; the total number of animals
was 10 to 15 per herd. In one feedlot we were able to sample all the animals due to the easy
accessibility to the animals.
Fresh fecal samples were collected by rectal palpation using individual obstetrical sleeves
and placed in whirl-pak bags. For the first four herds (sampled in 2011), samples were
transported to the laboratory on ice where they were stored at 4°C and then processed within 48
hours. For the other seven herds (one in 2011 and six in 2012), samples were transported to the
147

lab in a cooler without ice and processed immediately. This change in protocol was made
because a prior study found that ice storage decreased the likelihood of STEC recovery from
feces (Mindy Brashears, personal communication). Some of the STEC isolates and raw sample
feces from 2011 sampling visits were lost; which made impossible to determine the STEC status
in 45 individual cattle during some specific sample points.
When each farm was sampled, the date, time, latitude and longitude were recorded. In
addition, the maximum, minimum and average temperatures from the day of each sampling and
the preceding five days during Phase II were recorded using data from the closest weather station
(Quality Controlled Local Climatological Data (NOAA)).

3. Laboratory protocol for STEC detection and isolation

Five grams of feces were inoculated in 2X EC broth (Oxoid Ltd.; Waltham, MA)
supplemented with novobiocin (8mg/l), rifampin (2mg/l) and potassium tellurite (1mg/l) for 24
hours at 42°C (Jason, et al 2009) followed by subculture on STEC CHROMagar™
(CHROMagar, Paris, France) and sorbitol MacConkey (SMAC) agar. A portion of the EC
culture was also processed by immunomagnetic separation using Dynabeads® (Invitrogen
Corporation, California, USA) specific for E .coli O157:H7 followed by subcultured to O157
CHROMagar (CHROMagar, Paris, France) and SMAC agar. Up to 20 presumptive STEC single
colonies were selected from each plate, inoculated into Luria-Bertani (LB) broth for growth
overnight at 37°C, and confirmed by PCR using a previously described protocol (Tarr, et al
2002) with either the Taq 2x MeanGreen Master Mix or Kappa2G Multiplex Master Mix (Kapa
Biosystems, Massachusetts). The multiplex PCR used to confirm STEC single colonies detects

148

the presence of stx1, stx2 and eaeA (intimin). Individual colonies with at least one stx gene were
considered to be STEC, and fecal samples from individual cattle were considered positive if at
least one STEC isolate from the sample was recovered.

4. Data analyses

The data was collected, input and analyzed using SAS 9.3 (SAS Institute Inc., Cary, NC,
USA). STEC persistence and rates of acquisition and loss were based on whether or not the
animal had a least one STEC isolate. “PERSISTENT STEC POSITIVE” (PP) shedding animal
was defined as an animal that shed STEC for two consecutive visits. A “CONTINUOUS STEC
NEGATIVE” (CN) animal was defined as an animal that was not detected as shedding STEC
throughout the four sequential samples during the entire study period. When an animal was
positive in one visit, but negative in the next one, this was called “STEC LOSS”. In contrast,
when an animal was negative in one visit, but positive in the next one, this was called “STEC
ACQUISITION”. The “Rate of STEC Acquisition” (RA) was calculated by dividing the number
of animals that acquired STEC by the number of cattle at risk of acquiring infection, i.e. the
number negative in the previous sampling. The “Rate of STEC Loss” (RL) was calculated by
dividing the number of animal that lost STEC by the number of cattle at risk of losing an
infection, i.e. the number of positives in the previously sampling. We also determined the
number of animals that had lost STEC at any point during the study; this was called ANY STEC
LOSS (AL). Also we determined the quantity of animals that had acquired STEC at any point
during the study; this was called ANY STEC ACQUISITION (AA). These AL and AA variables
have a dichotomous outcome and these were analyzed by Chi-square.

149

We also calculated the RATE of NEW STEC INFECTIONS (RNI) for all 149 animals.
This rate is the result of dividing the number of new infections by the number of times that
animal was at risk of acquiring STEC. Thus an individual animal could be at risk of a new
infection 1, 2, 3 or 4 times during the study. These variables were analyzed with general linear
models (GLM).
Univariate analyses for each variable were performed to identify herd and individual
animal factors associated with STEC shedding in dairy herds. Each variable was initially
examined in a univariate analysis to identify factors to be included in a multivariable model,
using a backward manual selection procedure. Variables with non-normal distributions and
potential confounding were identified. The point of significance was P < 0.15 for inclusion in the
first multivariable model; however, the point of significance for the final multivariable model
was P < 0.05. Odds ratios (ORs) and their 95% confidence intervals (95% CI) were estimated for
each variable in the univariable analysis.
In the first model, the dependent variable was CONTINUOUS NEGATIVE (CN) and the
animal was considered a random effect. Those cattle that were negative at all sample points were
classified as CN. In this model there were three events. The first event is the time between Phase
II and Phase III.1, the second event is the time between Phase III.1 and Phase III.2 and the third
event is the time between Phase III.2 and Phase III.3. In those animals where the result was
missing between two visits, the next consecutive result was used instead, this happened in 18
animals.
In the second model, the dependent variable was the RNI. This rate was calculated by
dividing the number of new infections by the times the animal was at risk of a new infection. In
this way, we were able to use all the animals even if we were missing some STEC results. The

150

dependent variable was binary (STEC new infection yes or no). The model had herd as a random
effect. See Table 2.7 for explanation of the abbreviations used for variables.

151

Results

1. Descriptive statistics

A total of 149 animals were sampled in this longitudinal study. During 2011 80 (54%)
cattle were sampled, whereas 69 (46%) cattle were sampled during 2012. Of the animals
sampled, 46% were dairy cattle and 54% were beef cattle. However, STEC culture results were
available for all four sampling times for only 104 cattle (36 dairy and 68 beef).
From the 104 cattle with no missing data, a total of 28 (27%) animals were STEC
negative throughout the study period. Nine of these were dairy cattle and 19 were beef cattle.
One beef herd and two dairy herds had zero CN cattle.
From the cattle with no missing data a total of 27 (26%) cattle were PP. Except for one
animal, all PP were sampled during year 2012. Of the PP cattle, 16 (59 %) were beef and 11
(41%) were dairy cattle. The PP animals were found in three dairy and four beef herds (Figure
4.1). The average duration of shedding in the PP during our study period animals was 24 days;
the range was between 14 to 43 days. Only five animals, all from the same beef herd, were STEC
positive at all sampling periods.
There were a total of 45 (30%) animals missing STEC results from at least one sample
point. These cattle were not used in the calculation of the PP or STEC negative analysis. Of the
animals missing STEC results, 33 were dairy cattle and 12 were beef cattle. As previously
mentioned, all the animals missing STEC results belonged to year 2011. Of these animals 14 had
at least one STEC positive result. Thus 60% of the 149 animals were positive at least one time
during the study. The percentage of cattle missing STEC results was calculated in each sample
point. The percentages of missing results were 17%, 7% and 17%, respectively.
152

Using all culture results for each of the four visits, the STEC point prevalence at the first
visit was 30% (45/149 cattle), at the second visit 26% (32/124), at the third visit 24% (34/139),
and at the fourth visit 24% (29/123).

2. STEC Loss and Acquisition Rate

RL and RA were calculated from the 104 animals with complete data. RL between each
of the first three visits and the subsequent one were 64%, 54% and 54%, respectively. In
contrast, the RA between each of the first three visits and the next one were 21%, 20% and 21%,
respectively. RL and RA varied by herd between each visit, as can be observed (Figure 4.2 and
Figure 4.3).

3. ANY STEC LOSS and ANY STEC ACQUISITION

AL and AA were determined using all 149 animals in the study. A total of 60 (40%)
animals lost STEC at least one time, whereas 54 (36%) animals acquired STEC at least one time.
There was a significant different between herds (P < 0.0001) (Figure 4.4) and between
years (P < 0.0001) for AL. A total of 19% of animals lost STEC in 2011, whereas 65% of
animals lost STEC in 2012. There was no significant difference in AL between beef and dairy (P
= 0.2815); AL occurred in 36% of beef animals and in 45% of dairy animals. There was no
difference between herds (P = 0.1004) (Figure 4.5) nor between years (P= 0.3063) for AA. AA
occurred in 33% of animals in 2011 and in 41% of animals in 2012. AA was significantly

153

different between beef and dairy herds (P < 0.0021); any acquisition occurred in 48% of beef
animals and in 23% of the dairy animals.

4. Rate of new STEC infections

The RNI was significant different between herds (P <0.0001) Figure 4.6. This rate was
also significant different between beef and dairy system (P =0.0036) and between years (P
<0.0001). The RNI was 27% for beef herds and 16% for dairy herds; whereas for years the rate
of STEC infection was 13% for 2011 and 33% for 2012 (P <0.0001).

5. Univariate analysis of herd level management risk factors

With the two models, CN and RNI, we did not identify any management or individualcow factor significantly associated with these outcomes. In the first model, CN, 24 variables had
a P-value less than 0.15 but when the variable HERD was included in the model, these factors
became not significant or the model did not converge (Table 4.2). In the RNI model, 10
variables had a P-value less than 0.15 but when trying to build the multivariable model, none of
these variables became significant or the value of the estimated covariance parameter estimate
was 0 (Table 4.3). As a consequence, we could not build a multivariable model with this data.

154

Discussion

In this study we determined that STEC shedding is intermittent in dairy and beef cattle
and that RL is higher than RA. Only a minority of animals shed STEC constantly and that HERD
is the most important factor that influences RNI.
Two dairy herds that provided access to pasture for their cattle and one beef herd that
kept its cattle permanently in pasture had zero negative animals thorough the study; this result is
in agreement with data that indicates cattle in pasture-based production systems may have an
increased risk for STEC shedding (Hancock, et al 1994; Jay, et al 2007; Kondo, et al 2010).
Similarly, the five animals that shed STEC throughout the study period were all from the one
herd that kept its cattle on pasture. This same herd also had the highest RNI and one of the
highest rates of RA among herds. Raising cattle in pasture is thought to increase the exposure to
sources of STEC such as water runoff and wildlife activity (Jay, et al 2007; Kondo, et al 2010;
Laegreid, et al 1999). In contrast, other studies have reported that cattle in confinement-based
production systems have higher risk of STEC shedding (Gannon, et al 2002; Ogden, et al 2004;
Synge, et al 2003) or even that there is no difference between the two production systems
(Hancock, et al 1997b). It is important to mention that only one of the 11 herds sampled was
100% pasture based, so generalization to all pasture based systems should be done with caution.
However, based on our findings and those of others (Gannon, et al 2002; Hancock, et al 1994;
Jay, et al 2007; Kondo, et al 2010; Ogden, et al 2004; Synge, et al 2003), studies designed to
specifically explore STEC shedding in pasture-based production systems should be undertaken.
The only significant variable in the two univariable models that we investigated was
HERD regardless of whether the variable HERD was included as a random effect or as fixed
effect. An explanation for this finding could be that all the management practices variables were
155

at the herd level. In line with the differences in management practices among dairy herds was our
finding that AL was significantly different between herds. Based on our statistical model we
know that herd characteristics, such as management practices, influence the dynamics of STEC
in dairy cattle, but we were not able to identify which specific practices or factors account for
this difference.
We also found a significant difference in AA and RNI between beef and dairy
productions systems. These two production systems have different management practices. For
example dairy farms divided their animals by milk production levels and gestation length, as
well as separated calf from cows. While beef farms or feedlots keep their cattle in groups by age
or weight usually until finishing, depending of their production system. Also there are
differences in diet as net energy requirements are different between beef and dairy cattle. In our
study, beef animals presented a higher risk of RA and a higher RNI. In contrast, some studies
have reported dairy cattle production systems with a higher likelihood for STEC shedding
(Cobbaut, et al 2009; Cobbold, et al 2004).
A possible explanation for the difference between production systems in AA could be
that the beef cattle were younger than the dairy cattle. Younger cattle have been reported to have
a higher risk for STEC shedding (Cho, et al 2009; Dopfer, et al 2006; Gannon, et al 2002;
Hancock, et al 1997a; Stanford, et al 2005). In addition, younger cattle usually have a less
developed immune system, anatomic and physiologic differences and are often fed different
diets, which may explain the differences observed in our study (Gannon, et al 2002).
The percentage of negative cattle was almost double in herds sampled during 2011 than
during 2012. Also the herds sampled during 2012 had higher rates of RA compared to herds
sampled in 2011. One possible explanation for this difference could be that 2012 was a warmer

156

year than 2011. The 2011 average global surface temperature was between 0.07 and 0.16 degrees
Celsius warmer than the 1981-2010 average while 2012 was between 0.14 to 0.17 degrees
Celsius above, depending on the analysis (Lindsey 2012; Osborne and Lindsey 2013). Warm
temperatures are considered one of the most important risk factors for STEC shedding (Berry
and Wells 2010; Ogden, et al 2004; Stanford, et al 2005; Widiasih, et al 2004). This is in line
with the finding that all the PP animals except for one were sampled during 2012. Also 2012 had
the highest RNI, plus AL and RL were significantly higher in 2012.
Our findings that herd, age and warm temperatures influence the dynamics of STEC
shedding in cattle, are similar to those reported by Dopfer, et al (2006). In this study, beef cattle
were followed during 2 years and a strong farm and age effect for the first detection of STEC and
EHEC was found. In addition, a significant seasonal effect for the first STEC detection was
found. Finally as age increased, EHEC and STEC were detected less frequently (Dopfer, et al
2006).
Another possible explanation for the difference we just described in STEC dynamics
between years is stress, as warmer temperatures can produce heat stress. Stress affects the
capacity of the animal’s body to produce milk or gain weight. Stress can also affect its immunity
system (Berry and Wells 2010; Rostagno 2009). With a weaker immune system, some bacteria
can reproduce or colonize easier in their host; STEC is an example of that (Dean-Nystrom, et al
2008). Also warmer temperatures influence the environment where the animal lives. Warmer
temperatures enhance the conditions for the survival and replication of bacteria, such as STEC,
outside of the animal and in the environment, thus providing greater opportunities for animal
exposure. So it could be possible that animals sampled during 2012 had more chance to have
heat stress and as a consequence were more susceptible to STEC colonization and shedding.

157

The range of duration of STEC shedding has been reported to be from < 1 week to <1
month for O157 and from < 1 week to 3 weeks for O26 (Besser, et al 1997; Khaitsa, et al 2003;
Widiasih, et al 2004). Although we did not identify the O-typing of our isolates, our findings are
consistent with these studies. A common finding among all studies is the intermittent shedding of
STEC by cattle (Besser, et al 1997; Laegreid, et al 1999; Robinson, et al 2004; Shere, et al 1998;
Widiasih, et al 2004). Some studies proposed that exposure to periods of stress could be the
cause of intermittent STEC shedding (Stanford, et al 2005). For example, Stanford et al (2005)
found that E. coli O157:H7 shedding is transitory and sporadic after weaning. In our study only
five animals were constantly STEC positive at all sampling points. Our finding that only a
minority of cattle shed STEC constantly is in agreement with other studies (Robinson, et al
2004).
One weakness of our study is that we determined the presence of STEC but we did not
determine the O-typing of the isolates. As a consequence, we could not differentiate between a
persistent infection and a re-infection; there is the possibility that an animal could be positive to
one strain at one sampling point, and positive to a different strain at the next sampling point.
From one visit to another, more than 50% of all the cattle STEC positive became negative
by the next sampling (RL). However, the RA was almost always 20%. One possible explanation
for this finding could be that as the time past by the cattle developed an immune response against
STEC, which stop STEC shedding and as a consequence decreased STEC transmission. Another
possibility could be that an animal was actually shedding STEC at very low levels below our
laboratory detection levels. The detection limits for bacteriological cultures are between102 to
106 cfu/grams depending on enrichment techniques and if immunomagnetic separation is used
(Dopfer, et al 2012). Factors such as the inoculated serotype, the inoculation level, and the initial

158

concentration of the target organism in the sample, can also influence the sensitivity of the
different techniques implemented (LeJeune, et al 2006; Verstraete, et al 2010). Our finding of
high rate of loss and low rate of acquisition is in agreement with a report that individual animals
can have short periods of increased intensity of STEC shedding (Robinson, et al 2009), although
we did not evaluate STEC quantitatively in our samples.
The limitation of finding temperature and age as risk factors is that they are not easily
modified by human interventions (Cho, et al 2009). Possible intervention strategies would be to
separate young from older animals and avoid hot temperatures in confined-production systems.
These measures may not be practical or feasible to implement, though.
A limitation from our study was the incomplete STEC results from 30% of the cattle.
This limitation potentially precluded the design of a multivariable model and also could explain
the differences found between years 2011 and 2012 as most of the missing data came from cattle
sampled during 2011. Also it could influence the lack of association between some independent
variables and the dependent variable in both statistical models. For example, some of the
management and individual factors analyzed have been reported to increase or decrease STEC
shedding; such as temperature, distiller’s grains, contact with birds and age. However, these
variables did not present an effect in the RNI per se or once other variables were included into
the model, as they became not longer significant or the statistical model did not work properly.
RL and RA were calculated using only the cattle with complete data, so there is the possibility
that these rates could be different if we had complete data for all the cattle. As a consequence, we
approached this limitation by calculating AL and AA which took into account all data for all
cattle including those with missing samples. Another limitation was the change in methodology
in the handling of the samples, after the first four herds sampled during 2011 (change from 4C to

159

ambient temperature). It could be possible that in the samples handling at 4C we were able to
recover less isolates as the reason for the change in methodology was to improve the recovery of
STEC isolates.
In addition, a bigger sample size of herds would allow more power to identify the specific
management practices influencing CN animals and the RNI or corroborate that they do not
influence them.
In summary, we found that STEC dynamics varies by herd, type of production system
temperature and exposure to pasture. Temperature appears to be an important variable that
affects STEC shedding dynamics and should be taken into consideration when implementing or
testing new control measures. STEC shedding dynamics also appear to differ by type of
production systems, although specific factors responsible for this could not be identified in this
study. Finally, we described differences in STEC dynamics between cattle in pasture-based
systems and cattle in confinement-based systems. These differences should be investigated
further because it can have relevant influence in the implementation of control strategies in
different production systems.

160

APPENDIX

161

Figure 4.1. Percentage of cattle that were culture positive for STEC on two, three or four
consecutive sampling points for a period of time between 35 to 89 days.

4 Visits
18%

3 Visits
15%
2 Visits
67%

162

Figure 4.2. Rate STEC LOSS over time by herd. Event represents the time between one
sampling and the next one. Event 1= Phase II to Phase 3.1, Event 2= Phase 3.1 to 3.2 and Event

Rate

3= Phase 3.2 to 3.3. B= beef and D= dairy.

1.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00

Event 1
Event 2
Event 3

B1

B3

D6

D7

B8

D9

D10

B11

Herd Identification

163

B12

Figure 4.3. Rate of STEC ACQUISITION over time by herd. Event represents the time between
one sampling and the next one. Event 1= Phase II to Phase 3.1, Event 2= Phase 3.1 to Phase 3.2

Rate

and Event 3= Phase 3.2 to 3.3. B= beef and D= dairy.

1.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00

Event 1
Event 2
Event 3

B1

B3

D6

D7

B8

D9

D10

Herd Identification

164

B11

B12

Figure 4.4. Percentage of animals that lost STEC at any time during the study by herd. B= beef

Percentage animals

and D= dairy

100
90
80
70
60
50
40
30
20
10
0
B1

D2

B3

D4

D6 D7
B8
D9
Herd identification

165

D10

B11

B12

Figure 4.5. Percentage of cattle that acquired STEC at any time during the study by herd.
B= beef and D= dairy.

100
90
Percentage animals

80
70
60
50
40

30
20
10
0
B1

D2

B3

D4

D6
D7
B8
D9
Herd identification

166

D10

B11

B12

Figure 4.6. Rate of STEC NEW INFECTIONS (number of new infections in a cattle divided by
the number of times cattle was at risk or susceptible to new infection) by herd. B = beef and D =
dairy.

80

Incidence rate

70
60
50
40
30
20
10
0
B1

D2

B3

D4

D6
D7
B8
D9
Herd identification

167

D10

B11

B12

Table 4.1. Univariate analysis of dairy herd variables associated with persistently STEC-negative cattle.

Characteristic
Herd
2
4
6
7
9
10
Temperature at sampling date on
Phase III
Lactating
No
Yes
Antibiotics used 2 weeks prior to
the Phase III samplings
Yes
No
Rumensin
No
Yes

N° (%) with
characteristic

N° (%) Constant
STEC Negative

p-value

OR

95% CI

11 (7.1)
22 (14.3)
19 (12.3)
36 (23.4)
36 (23.4)
30 (19.5)

9 (82)
21 (95)
16 (84)
20 (56)
22 (61)
12 (40)

0.0178

6.5
31.6
8.7
1.9
2.4
ref

0.94- 44.44
3.23- 309.78
1.63- 45.99
0.55- 6.69
0.70- 8.49
Ref

0.0252

0.9

0.86- 0.99

8 (5)
145 (95)

3 (38)
51 (35)

0.8682

0.86
ref

0.15- 5.02
Ref

3 (2)
119 (98)

2 (67)
34 (29)

0.2714

0.223
Ref

0.02- 3.32
Ref

96 (62)
58 (38)

59 (61)
41 (71)

0.2988

0.63
Ref

0.26- 1.53
Ref

168

Table 4.1 (cont’d)

Year
2011
2012
Season
Spring
Summer
Temperature Av
≤ 20.6 C
> 20.6 C
Temperature Max
≤ 27.8 C
> 27.8 C
Temperature Min
≤ 15.6 C
> 15.6 C
Temperature Av5 days
≤ 19.4 C
> 19.4 C
Temperature Max5 days
≤ 28.9 C
> 28.9 C
Temperature Min5 days
≤15C
>15C

52 (34)
102 (66)

46 (89)
54 (53)

0.0004

6.9
Ref

2.44- 19.32
ref

36 (23)
118 (77)

20 (56)
80 (68)

0.2350

0.54
Ref

0.20- 1.50
ref

52 (34)
102 (66)

46 (88)
54 (53)

0.0004

6.9
Ref

2.44- 19.32
ref

88 (57)
66 (43)

66 (75)
34 (51)

0.0109

3.1
Ref

1.31- 7.32
ref

52 (34)
102 (66)

46 (88)
54 (53)

0.0004

6.9
Ref

2.44- 19.32
ref

33 (21)
121 (79)

30 (91)
70 (58)

0.0052

6.9
Ref

1.81- 26.13
ref

66 (43)
88 (57)

45(68)
55(63)

0.5907

1.3
Ref

0.53- 3.00
ref

30 (19)
124 (81)

25 (83)
75 (60)

0.0582

3.1
ref

1- 10.2
ref

169

Table 4.1. (cont’d)
Lactation
1st
2nd or higher
Days in milk
0
1-30d
>= 31d
Dry
Yes
No
Breed
Crossbreed
Jersey
Holstein
Herd type
Closed
Open
Calves-Replacements proportion
<5.1
>48.4
From 46.9 to 48.3
Proportion herd lactating
>50
<31.7% to 49%
Proportion herd dry
1.7- 4%
>7.6%
6.2- 7.5%
Herd Size
>1000
<1000

48 (31)
106 (69)

30 (63)
70 (66)

0.7878

0.9
Ref

0.35- 2.22
ref

4 (3)
14 (11)
106 (85)

2 (50)
5 (36)
74 (70)

0.1245

0.46
0.21
Ref

0.04- 5.38
0.05- 0.98
ref

4 (3)
150 (97)

2 (50)
98 (65)

0.6449

0.6
Ref

0.05- 6.22
ref

30 (19)
19 (12)
105 (68)

12 (40)
16 (84)
72 (69)

0.0204

0.28
2.45
Ref

0.1- 0.82
0.55- 10.94
ref

11 (7)
143 (93)

9 (82)
91 (64)

0.3735

2.2
Ref

0.38- 13.10
ref

19 (12)
36 (23)
99 (64)

16 (84)
20 (56)
64 (65)

0.2036

2.9
0.6
Ref

0.63- 13.58
0.22- 1.78
ref

19 (12)
135 (88)

16 (84)
84 (62)

0.1176

3.3
Ref

0.74- 14.60
ref

72 (47)
19 (12)
63 (41)

42 (58)
16 (84)
42 (67)

0.1831

0.6
2.6
Ref

0.25- 1.58
0.53- 12.73
ref

58 (38)
96 (62)

41 (71)
59 (61)

0.2988

1.6
Ref

0.65- 3.92
ref

170

Table 4.1 (cont’d)
Adding Cow/Replacements
0%
At least 5 animals
Adding Bulls
0%
4 animals
Culling Rate
Low level
High level
N° Milkings
3- 4 times
2-3 times
Loose Housing
No
Yes
Tie stanchion
No
Yes
Access pasture/dry lot
No
Yes
Lactation access pasture
No
Yes
Transition pen separate
No
Yes
Sick animals penned separated
Yes
No

11 (7)
143 (93)

9 (82)
91 (64)

0.3735

2.2
Ref

0.38- 13.10
ref

124 (81)
30 (19)

88 (71)
12 (40)

0.0099

4.0
Ref

1.41- 11.52
ref

49 (32)
105 (68)

28 (57)
72 (69)

0.2842

0.6
Ref

0.25- 1.51
ref

58 (38)
96 (62)

41 (71)
59 (61)

0.2988

1.6
Ref

0.65- 3.92
ref

135 (88)
19 (12)

84 (62)
16 (84)

0.1176

0.3
Ref

0.07- 1.36
ref

143 (93)
11 (7)

91 (64)
9 (82)

0.3735

0.5
Ref

0.08- 2.65
ref

99 (64)
55 (36)

62 (63)
38 (69)

0.5508

0.8
Ref

0.31- 1.87
ref

99 (64)
55 (36)

62 (63)
38 (69)

0.5508

0.8
Ref

0.31- 1.87
ref

66 (75)
88 (57)

34 (51)
66 (43)

0.0109

0.3
Ref

0.14- 0.77
ref

77 (50)
77 (50)

57 (74)
43 (56)

0.0409

2.5
ref

1.04- 5.85
ref

171

Table 4.1 (cont’d)
1st lactations animals penned separated
Yes
No
Cow/Heifers Raised
Off-site/ On main farm
Another farm
Feeders clean Year
<365
365
Washed
No
Yes
Spray
No
Yes
Lime
No
Yes
Tx Respiratory Disease
Yes
No
Tx Foot Infection Disease
Yes
No
Tx Metritis
Yes
No
Feed TMR
No
Yes

58 (38)
96 (62)

41 (71)
59 (61)

0.2988

1.6
Ref

0.65- 3.92
ref

135 (88)
19 (12)

84 (62)
16 (84)

0.1176

0.3
Ref

0.07- 1.36
ref

66 (43)
88 (57)

34 (52)
66 (75)

0.0109

0.3
Ref

0.14- 0.77
ref

107 (69)
47 (31)

69 (64)
31 (66)

0.9848

1
Ref

0.40- 2.56
ref

143 (93)
11 (7.14)

91 (64)
9 (82

0.3735

0.5
Ref

0.08- 2.65
ref

99 (64)
55 (36)

64 (65)
36 (65)

0.9514

1
Ref

0.40- 2.39
ref

118 (77)
36 (23)

78 (66)
22 (61)

0.5592

1.6
Ref

0.49- 3.74
ref

135 (88)
19 (12)

84 (62)
16 (84)

0.1176

0.3
Ref

0.07- 1.36
ref

135 (88)
19 (12)

84 (62)
16 (84)

0.1176

0.3
Ref

0.07- 1.36
ref

42 (30)
141 (68)

21 (32)
66 (32)

0.5508

1.3
Ref

0.53- 3.22
ref

172

Table 4.1 (cont’d)
% Corn silage Diet
No
Yes
% Distillers grains Diet
No
Yes
% Cottonseed Diet
No
Yes
Contact with Other SP
Horses
None
Contact with Cats
No
Yes
Contact with Deer
No
Yes
Contact with Dogs
No
Yes
Contact with Opossum
No
Yes
Contact with Raccoons
Always
Frequent
Rarely

55 (36)
99 (64)

38 (69)
62 (63)

0.5508

1.3
Ref

0.53- 3.22
ref

108 (70)
46 (30)

73 (68)
27 (59)

0.4323

1.4
Ref

0.57- 3.64
ref

118 (77)
36 (23)

80 (68)
20 (56)

0.2350

1.8
Ref

0.67- 5.08
ref

19 (12)
135 (88)

16 (84)
84 (62)

0.1176

3.3
Ref

0.7- 14.6
ref

30 (19)
124 (81)

12 (40)
88 (71)

0.0099

0.2
Ref

0.09- 0.71
ref

36 (23)
118 (77)

20 (56)
80 (68)

0.2350

0.5
Ref

0.20- 1.50
ref

94 (61)
60 (39)

63 (67)
37 (62)

0.5945

1.3
Ref

0.53- 3.01
ref

30 (19)
124 (81)

12 (40)
88 (71)

0.0099

0.2
Ref

0.09- 0.71
ref

72 (47)
71 (46)
11 (7)

42 (58)
49 (69)
9 (82)

0.2954

0.3
0.6
ref

0.05- 2.16
0.10- 3.9
ref

173

Table 4.1 (cont’d)
Contact with Rodents
Always
Frequent
Rarely
Contact with Skunks
No
Yes
Area Cleanliness
Cleanest third
Middle third
Bed Cleanliness
Cleanest third
Middle third
Rumensin
Yes
No
Direct Fed Microbials
Yes
No
Anthelmintic
Yes
No
BLV
Positive
Negative

72 (47)
71 (46)
11 (7)

42 (58)
49 (69)
9 (82)

0.2954

0.3
0.6
Ref

0.05- 2.16
0.10- 3.9
ref

30 (19)
124 (81)

12 (40)
88 (71)

0.0099

0.2
ref

0.09- 0.71
ref

105 (68)
49 (32)

72( 69)
28 (57)

0.2842

1.6
ref

0.66- 4.02
ref

105 (68)
49 (32)

72( 69)
28 (57)

0.2842

1.6
ref

0.66- 4.02
ref

58 (38)
96 (62)

41 (71)
59 (61)

0.2988

1.6
ref

0.65- 3.92
ref

60 (39)
94 (61)

37 (62)
63 (67)

0.5945

0.8
ref

0.33- 1.89
ref

82 (53)
72 (47)

58 (71)
42 (58)

0.1383

1.9
ref

0.81- 4.60
ref

77 (50)
77 (50)

48 (62)
52 (68)

0.5565

0.8
ref

0.33- 1.83
ref

174

Table 4.2. Univariate analysis to identify risk factors for rate of new STEC infections in dairy cattle

Characteristic
Season
Spring
Summer
Temperature Max
≤ 27.8 C
> 27.8 C
Temperature Aver5 days
≤ 19.4 C
> 19.4 C
Temperature Max5 days
≤ 28.9 C
> 28.9 C
Temperature Min5 days
≤15 C
>15 C
Mean temperature longitudinal
≤18.9 C
>18.9 C
Range temperature longitudinal
≤10
>10
Temperature 2nd sampling
≤ 20 C
>20 C
Temperature 3rd sampling
≤18.9 C
>18.9 C

No. (%) with
characteristic

No. (%) with
STEC

p-value

OR

95% CI

12 (17)
57 (83)

9 (75)
25 (44)

0.2966

2
ref

0.55- 7.04
ref

47 (68)
22 (32)

17 (36)
17 (77)

0.1812

0.5
ref

0.17- 1.41
ref

25 (36)
44 (64)

6 (24)
28 (64)

0.0851

0.4
ref

0.14- 1.14
ref

32 (46)
37 (54)

16(50)
18(49)

0.8523

1.1
ref

0.35- 3.55
ref

20 (29)
49 (71)

7 (35)
27 (55)

0.4852

0.6
ref

0.19- 2.23
ref

10 (14)
59 (86)

2 (20)
32 (54)

0.2088

0.3
ref

0.06- 1.89
ref

47 (68)
22 (32)

23 (49)
11 (50)

0.9935

1
ref

0.29- 3.48
ref

45 (65)
24 (35)

16 (36)
18 (75)

0.1051

0.5
ref

0.19- 1.18
ref

46 (67)
23 (33)

29 (63)
5 (22)

0.0552

2.9
ref

1- 8.46

175

Table 4.2 (cont’d)
Temperature 4th sampling
≤18.3 C
>18.3 C
Lactation
1st
2nd or higher
Days in milk
0
1-30d
>= 31d
Dry
No
Yes
Breed
Crossbreed
Holstein
Jersey
Herd type
Closed
Open
Calves-Replacements proportion
<5.1
>48.4
From 46.9 to 48.3
Proportion herd lactating
>50
<31.7% to 49%
Proportion herd dry
1.7- 4%
>7.6%
6.2- 7.5%

ref

38 (55)
31 (45)

13 (34)
21 (68)

0.0961

0.5
ref

0.20- 1.14
ref

21 (30)
48 (70)

12 (57)
22 (46)

0.2612

1.6
ref

0.69- 3.91
ref

3 (5)
6 (10)
50 (85)

1 (33)
5 (83)
21 (42)

0.3901

1.6
2.2
ref

0.14- 19.13
0.66- 7.55
ref

66 (96)
3 (4)

33 (50)
1 (33)

0.7470

1.5
ref

0.13- 17.10
ref

10 (14.5)
49 (71)
10 (14.5)

8 (80)
24 (49)
2 (20)

0.3220

4.4
2.8
ref

0.63- 31.37
0.49- 15.51
ref

10 (14)
59 (86)

5 (50)
29 (49)

0.7885

1.2
ref

0.25- 6.26
ref

10 (15)
12 (17)
47 (68)

2 (20)
9 (75)
23 (49)

0.2720

0.4
1.7
ref

0.07- 2.03
0.56- 4.97
ref

10 (14)
59 (86)

2 (20)
32 (54)

0.2088

0.3
ref

0.06- 1.89
ref

24 (35)
10 (14)
35 (51)

18 (75)
2 (20)
14 (40)

0.1448

1.8
0.4
ref

0.77- 4.27
0.09- 2.32
ref

176

Table 4.2 (cont’d)
Herd Size
>1000
<1000
Adding Cow/Replacements
0%
At least 5 animals
Adding Bulls
0%
4 animals
Culling Rate
Low level
High level
N° Milkings
3- 4 times
2-3 times
Loose Housing
No
Yes
Tie stanchion
No
Yes
Access pasture/dry lot
No
Yes
Lactation access pasture
No
Yes
Transition pen separate
No
Yes

27 (39)
42 (61)

10 (37)
24 (57)

0.5098

0.7
ref

0.19- 2.28
ref

10 (14)
59 (86)

5 (50)
29 (49)

0.7885

1.2
ref

0.25- 6.26
ref

59 (86)
10 (14)

26 (44)
8 (80)

0.3651

0.5
ref

0.13- 2.15
ref

20 (29)
49 (71)

10 (50)
24 (49)

0.9028

0.9
ref

0.26- 3.24
ref

27 (39)
42 (61)

10 (37)
24 (57)

0.5098

0.7
ref

0.19- 2.28
ref

59 (86)
10 (14)

32 (54)
2 (20)

0.2088

3.04
ref

0.53- 17.46
ref

59 (86)
10 (14)

29 (49)
5 (50)

0.7885

0.8
ref

0.16- 4.05
ref

47 (68)
22 (32)

23 (49)
11 (50)

0.8022

1.2
ref

0.34- 4.00
ref

47 (68)
22 (32)

23 (49)
11 (50)

0.8022

1.2
ref

0.34- 4.00
ref

22 (32)
47 (68)

17 (77)
17 (36)

0.1812

2
ref

0.71- 5.88
ref

177

Table 4.2 (cont’d)
Sick animals penned separated
Yes
No
1st lactations animals penned separated
Yes
No
Cow/Heifers Raised
Off-site/ On main farm
Another farm
Feeders clean Year
<365
365
Washed
No
Yes
Spray
No
Yes
Lime
No
Yes
Tx Respiratory Disease
Yes
No
Tx Foot Infection Disease
Yes
No
Tx Metritis
Yes
No

37 (54)
32 (46)

12 (32)
22 (69)

0.1472

0.5
ref

0.16- 1.33
ref

27 (39)
42 (61)

10 (37)
24 (57)

0.5098

0.7
ref

0.19- 2.28
ref

59 (86)
10 (14)

32 (54)
2 (20)

0.2088

3
ref

0.53- 17.46
ref

22 (32)
47 (68)

17 (77)
17 (36)

0.1812

2
ref

0.71- 5.88
ref

47 (68)
22 (32)

20 (43)
14 (64)

0.4365

0.6
ref

0.19- 2.06
ref

59 (86)
10 (14)

29 (49)
5 (50)

0.7885

0.8
ref

0.16- 4.05
ref

47 (68)
22 (32)

23 (49)
11 (50)

0.9935

1
ref

0.29- 3.45
ref

57 (83)
12 (17)

25 (44)
9 (75)

0.4841

0.6
ref

0.15- 2.47
ref

59 (86)
10 (14)

32 (54)
2 (20)

0.2088

3
ref

0.53- 17.46
ref

59 (86)
10 (14)

32 (54)
2 (20)

0.2088

3
ref

0.53- 17.46
ref

178

Table 4.2 (cont’d)
Feed TMR
No
Yes
% Corn silage Diet
No
Yes
% Distillers grains Diet
No
Yes
% Cottonseed Diet
No
Yes
Contact with other species
Horses
None
Contact with Cats
No
Yes
Contact with Deer
No
Yes
Contact with Dogs
No
Yes
Contact with Opossum
No
Yes
Contact with Raccoons
Always
Frequent
Rarely

22 (32)
47 (68)

11 (50)
23 (49)

0.8022

0.9
ref

0.25- 2.94
ref

22 (32)
47 (68)

11 (50)
23 (49)

0.8022

0.9
ref

0.25- 2.94
ref

51 (74)
18 (26)

25 (49)
9 (50)

0.7095

1.3
ref

0.35- 4.71
ref

57 (83)
12 (17)

25 (44)
9 (75)

0.2966

0.5
ref

0.14- 1.83
ref

10 (14)
59 (86)

2 (20)
32 (54)

0.2088

0.3
ref

0.06- 1.89
ref

10 (14)
59 (86)

8 (80)
26 (44)

0.3651

1.9
ref

0.47- 7.76
ref

12 (17)
57 (83)

9 (75)
25 (44)

0.2966

2
ref

0.55- 7.04
ref

39 (57)
30 (43)

19 (49)
15 (50)

0.9355

1
ref

0.30- 3.07
ref

10 (14)
59 (86)

8 (80)
26 (44)

0.3651

1.9
ref

0.47- 7.76
ref

24 (35)
35 (51)
10 (14)

18 (75)
11 (31)
5 (50)

0.1508

1.3
0.5
ref

0.38- 4.58
0.15- 1.95
ref

179

Table 4.2 (cont’d)
Contact with Rodents
Always
Frequent
Rarely
Contact with Skunks
No
Yes
Area Cleanliness
Cleanest third
Middle third
Rumensin
Yes
No
Anthelmintic
Yes
No
BLV
Positive
Negative
Flies control
No
Yes

24 (35)
35 (51)
10 (14)

18 (75)
11 (31)
5 (50)

0.1508

1.3
0.5
ref

0.38- 4.58
0.15- 1.95
ref

10 (14)
59 (86)

8 (80)
26 (44)

0.3651

1.9
ref

0.47- 7.76
ref

49 (71)
20 (29)

24 (49)
10 (50)

0.9028

1.1
ref

0.31- 3.78
ref

27 (39)
42 (61)

10 (37)
24 (57)

0.5098

0.7
ref

0.19- 2.28
ref

45 (65)
24 (35)

16 (36)
18 (75)

0.1051

0.5
ref

0.19- 1.18
ref

35 (51)
34 (49)

18 (51)
16 (47)

0.8567

0.9
ref

0.42- 2.07
Ref

22 (32)
47 (68)

17 (77)
17 (36)

0.0768

2.3
ref

0.9- 5.7
Ref

180

REFERENCES

181

REFERENCES

BERRY, E. D. & WELLS, J. E. (2010) Escherichia coli O157:H7: recent advances in research
on occurrence, transmission, and control in cattle and the production environment. Adv. Food
Nutr. Res. 60, 67-117
BESSER, T. E., HANCOCK, D. D., PRITCHETT, L. C., MCRAE, E. M., RICE, D. H. &
TARR, P. I. (1997) Duration of Detection of Fecal Excretion of Escherichia coli O157:H7 in
Cattle. J. Infect. Dis. 175, 726-729
CALLAWAY, T. R., EDRINGTON, T. S., LONERAGAN, G. H., CARR, M. & NISBET, D. J.
(2013) Shiga toxin-producing Escherichia coli (STEC) ecology in cattle and management based
options for reducing fecal shedding. Agric. Food Anal. Bacteriol. 3, 39-69
CHO, S., FOSSLER, C. P., DIEZ-GONZALEZ, F., WELLS, S. J., HEDBERG, C. W.,
KANEENE, J. B., RUEGG, P. L., WARNICK, L. D. & BENDER, J. B. (2009) Cattle-level risk
factors associated with fecal shedding of Shiga toxin-encoding bacteria on dairy farms,
Minnesota, USA. Can. J. Vet. Res.-Rev. Can. Rech. Vet. 73, 151-156
COBBAUT, K., BERKVENS, D., HOUF, K., DE DEKEN, R. & DE ZUTTER, L. (2009)
Escherichia coli O157 prevalence in different cattle farm types and identification of potential risk
factors. J. Food Prot. 72, 1848-1853
COBBOLD, R. N., RICE, D. H., SZYMANSKI, M., CALL, D. R. & HANCOCK, D. D. (2004)
Comparison of Shiga-Toxigenic Escherichia coli Prevalences among Dairy, Feedlot, and CowCalf Herds in Washington State. Appl. Environ. Microbiol. 70, 4375-4378
DEAN-NYSTROM, E. A., STOFFREGEN, W. C., BOSWORTH, B. T., MOON, H. W. &
POHLENZ, J. F. (2008) Early Attachment Sites for Shiga-Toxigenic Escherichia coli O157:H7
in Experimentally Inoculated Weaned Calves. Appl. Environ. Microbiol. 74, 6378-6384
DOPFER, D., GEUE, L., DE BREE, J. & DE JONG, M. C. M. (2006) Dynamics of verotoxinproducing Escherichia coli isolated from German beef cattle between birth and slaughter. Prev.
Vet. Med. 73, 229-240
DOPFER, D., GEUE, L., SCHARES, S., MINTEL, B., HOFFMAN, B. & FISCHER, E. A. J.
(2012) Dynamics of shiga-toxin producing Escherichia coli (STEC) and their virulence factors in
cattle. Prev. Vet. Med. 103, 22-30
182

GANNON, V. P., GRAHAM, T. A., KING, R., MICHEL, P., READ, S., ZIEBELL, K. &
JOHNSON, R. P. (2002) Escherichia coli O157:H7 infection in cows and calves in a beef cattle
herd in Alberta, Canada. Epidemiol. Infect. 129, 163-172
HANCOCK, D. D., BESSER, T. E., KINSEL, M. L., TARR, P. I., RICE, D. H. & PAROS, M.
G. (1994) The Prevalence of Escherichia coli O157.H7 in Dairy and Beef Cattle in Washington
State. Epidemiol. Infect. 113, 199-207
HANCOCK, D. D., BESSER, T. E., RICE, D. H., HERRIOTT, D. E. & TARR, P. I. (1997a) A
longitudinal study of Escherichia coli O157 in fourteen cattle herds. Epidemiol. Infect. 118, 193195
HANCOCK, D. D., RICE, D. H., HERRIOTT, D. E., BESSER, T. E., EBEL, E. D. &
CARPENTERS, L. V. (1997b) Effects of Farm Manure-Handling Practices on Escherichia coli
O157 Prevalence in Cattle. J. Food Prot. 60, 363-366
HUSSEIN, H. S. (2007) Prevalence and pathogenicity of Shiga toxin-producing Escherichia coli
in beef cattle and their products. J. Anim. Sci. 85, E63-E72
JASON, V., RAJKOVIC, A., BAERT, L., DEBEVERE, J. & UYTTENDALE, M. (2009)
Comparison of enrichment conditions for rapid detection of low numbers of sublethally injured
Escherichia coli O157 in food. J. Food Prot. 72, 1862-1868
JAY, M. T., COOLEY, M., CARYCHAO, D., WISCOMB, G. W., SWEITZER, R. A.,
CRAWFORD-MIKSZA, L., FARRAR, J. A., LAU, D. K., O'CONNELL, J., MILLINGTON,
A., ASMUNDSON, R. V., ATWILL, E. R. & MANDRELL, R. E. (2007) Escherichia coli
O157:H7 in feral swine near spinach fields and cattle, central California coast. Emerg. Infect.
Dis. 13, 1908-1911
KHAITSA, M. L., SMITH, D. R., STONER, J. A., PARKHURST, A. M., HINKLEY, S.,
KLOPFENSTEIN, T. J. & MOXLEY, R. A. (2003) Incidence, duration, and prevalence of
Escherichia coli O157:H7 fecal shedding by feedlot cattle during the finishing period. J. Food
Prot. 66, 1972-1977
KONDO, S., HOAR, B. R., VILLANUEVA, V., MANDRELL, R. E. & ATWILL, E. R. (2010)
Longitudinal prevalence and molecular typing of Escherichia coli O157:H7 by use of multiplelocus variable-number tandem-repeat analysis and pulsed-field gel electrophoresis in fecal
samples collected from a range-based herd of beef cattle in California. Am. J. Vet. Res. 71,
1339-1347

183

KUHNERT, P., DUBOSSON, C. R., ROESCH, M., HOMFELD, E., DOHERR, M. G. &
BLUM, J. W. (2005) Prevalence and risk-factor analysis of Shiga toxigenic Escherichia coli in
faecal samples of organically and conventionally farmed dairy cattle. Vet. Microbiol. 109, 37-45
LAEGREID, W. W., ELDER, R. O. & KEEN, J. E. (1999) Prevalence of Escherichia coli
O157:H7 in Range Beef Calves at Weaning. Epidemiol. Infect. 123, 291-298
LEJEUNE, J. T., HANCOCK, D. D. & BESSER, T. E. (2006) Sensitivity of Escherichia coli
O157 detection in bovine feces assessed by broth enrichment followed by immunomagnetic
separation and direct plating methodologies. J. Clin. Microbiol. 44, 872-875
LINDSEY, R. (2012) State of the Climate: 2011 global surface temperature.
www.climate.gov/news-features/understanding-climate/state-climate-2011-global-surfacetemperature. Accessed 2-23-2015 Access, 2012
OGDEN, I. D., MACRAE, M. & STRACHAN, N. J. (2004) Is the prevalence and shedding
concentrations of E. coli O157 in beef cattle in Scotland seasonal? FEMS Microbiol. Lett. 233,
297-300
OSBORNE, S. & LINDSEY, R. (2013) 2012 State of the Climate: Earth's surface temperature.
www.climate.gov/news-features/understanding-climate/2012-state-climate-earths-surfacetemperature. Accessed 2-23-2015 Access, 2013
ROBINSON, S. E., BROWN, P. E., WRIGHT, E. J., HART, C. A. & FRENCH, N. P. (2009)
Quantifying within- and between-animal variation and uncertainty associated with counts of
Escherichia coli O157 occurring in naturally infected cattle faeces. Journal of the Royal Society,
Interface / the Royal Society 6, 169-177
ROBINSON, S. E., WRIGHT, E. J., HART, C. A., BENNETT, M. & FRENCH, N. P. (2004)
Intermittent and persistent shedding of Escherichia coli O157 in cohorts of naturally infected
calves. J. Appl. Microbiol. 97, 1045-1053
ROSTAGNO, M. H. (2009) Can Stress in Farm Animals Increase Food Safety Risk? Foodborne
Pathog. Dis. 6, 767-776
SHERE, J. A., BARTLETT, K. J. & KASPAR, C. W. (1998) Longitudinal study of Escherichia
coli O157 : H7 dissemination on four dairy farms in Wisconsin. Appl. Environ. Microbiol. 64,
1390-1399

184

SMITH, G. G., GOEBEL, S. E., CULBERT, C. R. & GUILBAULT, L. A. (2013) Reducing the
public health risk of Escherichia coli O157 exposure by immunization of cattle. Can. J. Public
Health. 104, e9-e11
STANFORD, K., BACH, S., MARX, T., JONES, S., HANSEN, J., WALLINS, G.,
ZAHIRODDINI, H. & MCALLISTER, T. A. (2005) Monitoring Eschericia coli O157:H7 in
inoculated and naturally colonized feedlot cattle and their environment. J. Food Prot. 68, 26-33
SYNGE, B. A., CHASE-TOPPING, M. E., HOPKINS, G. F., MCKENDRICK, I. J.,
THOMSON-CARTER, F., GRAY, D., RUSBRIDGE, S. M., MUNRO, F. I., FOSTER, G. &
GUNN, G. J. (2003) Factors influencing the shedding of verocytotoxin-producing Escherichia
coli O157 by beef suckler cows. Epidemiol. Infect. 130, 301-312
TARR, C. L., LARGE, T. M., MOELLER, C. L., LACHER, D. W., TARR, P. I., ACHESON, D.
W. & WHITTAM, T. S. (2002) Molecular characterization of a serotype O121:H19 clone, a
distinct Shiga toxin-producing clone of pathogenic Escherichia coli. Infect. Immun. 70, 68536859
VERSTRAETE, K., DE ZUTTER, L., MESSENS, W., HERMAN, L., HEYNDRICKX, M. &
DE REU, K. (2010) Effect of the enrichment time and immunomagnetic separation on the
detection of Shiga toxin-producing Escherichia coli O26, O103, O111, O145 and sorbitol
positive O157 from artificially inoculated cattle faeces. Vet. Microbiol. 145, 106-112
WIDIASIH, D. A., IDO, N., OMOE, K., SUGII, S. & SHINAGAWA, K. (2004) Duration and
magnitude of faecal shedding of Shiga toxin-producing Escherichia coli from naturally infected
cattle. Epidemiol. Infect. 132, 67-75

185

CONCLUSIONS AND FUTURE STUDIES

Conclusions

We performed a group of hypothesis-testing epidemiological studies with the aim of
elucidating risk factors for STEC shedding and a better understanding of STEC dynamics to
generate knowledge that could lead to the design of intervention strategies to control STEC
infection in cattle and ultimately to reduce transmission to humans.
From our study results, it is evident that the intermittent shedding pattern of STEC poses
a challenge to accurately identify STEC shedding in cattle. Additionally, the intermittent
shedding poses a challenge in understanding the dynamics of STEC shedding especially when
considering the multiple of STEC serotypes that can be present in cattle. For these reasons,
longitudinal studies will most likely provide the most accurate information when trying to further
elucidate epidemiology of STEC in cattle.
Our results support the previously reported association between STEC shedding and
seasonality, more specifically with warm temperatures. Those cattle exposed to warmer
temperatures presented higher risk for STEC shedding. Warm temperatures could not only favor
STEC shedding by cattle but as well favor the survival of STEC in the environment. This could
increase STEC infection and shedding in other reservoirs, as a consequence favoring the
transmission and infection to other cattle farms, animal species and crops.
Our results also support previously reported associations between STEC shedding and
age, more specifically that younger cattle have a higher risk of STEC shedding. This association
is also supported by our finding that cows in their first lactation are at higher risk of STEC
186

shedding compare with cows with two lactations or more, in other words, older cows. As a
consequence, younger cattle could be a target population for intervention strategies to reduce
STEC colonization and shedding.
Another population that our results support as target for intervention strategies in dairy
herds is cows in their first 30 days of milk production. Cattle early in lactation are under stress
and usually in a negative energy balance, which may increase susceptibility to STEC shedding.
Thus this specific group of cows can be also targeted for the implementation of intervention
strategies at pre-harvest aim to reduce STEC infection in humans.
All the farms that we sampled were STEC positive although the prevalence and rate of
STEC new infections varied between herds. As STEC is considered part of the normal
gastrointestinal flora in cattle, the high farm prevalence is not surprising. The fact that it is highly
likely that STEC are present in all cattle farms emphasizes the importance of designing preharvest interventions strategies to reduce colonization and shedding.
We not only isolated STEC from all farms but found differences in STEC shedding
dynamics by the type of production system. Beef cattle had a higher risk of STEC shedding than
dairy cattle. In addition, we found a diverse STEC population between and within herds, as well
as by production type. We conclude that herd, and within this term we could refer to
management practices, are associated with STEC dynamics, although we were not able to
identify which specific practices or factors in a herd are those associated.
Different stx and eaeA genes profiles were observed between herds. The gene stx2 was
the most frequent identified among the isolates we recovered. We also found that all the herds
had a least one EHEC isolate. This is an important observation from the public health
perspective, as EHEC is the most common cause of outbreaks and hospitalizations.

187

The differences between dairy and beef herds in management practices, such as diet, age
and even in genetics may influence STEC dynamics in a way that different STEC O-types/strains
can be more suitable to colonize, multiply and be shed in one production system than in the
other. These factors could also influence the type of virulence factors that STEC bacteria can
acquire or lose inside the gastrointestinal tract and influence its evolution.
Although we did not find a significant association between STEC shedding in cattle and
exposure to pasture, we did observe differences in STEC shedding, rate of new infections and
number of cattle negative to STEC in those herds that provide access to pasture to their cattle. It
could be that pasture-based systems provide the right mix of environmental factors for STEC to
survive in the environment as well as colonized and multiply in cattle’s digestive tract. These
findings deserve further investigation as pasture based systems are often advocated as having
attributes that make them more appealing to the public, as they are consider more environmental
friendly and better for cattle welfare.
We found in our study that BLV and MAP, chronic bovine diseases that have significant
long-term effects on the immune system (BLV) and the gastrointestinal tract (MAP), were not
associated with STEC shedding. Thus controlling BLV and MAP will not likely have an impact
on STEC shedding in cattle. However controlling both BLV and MAP is still important for
overall herd health and productivity.
The body of information about STEC is composed mostly by studies that focus in E. coli
O157:H7. For this reason, this dissertation offers information about all types of STEC and not
only about E. coli O157:H7. It is necessary to produce more information about non-O157 STEC
as they are more frequently identified as a cause of outbreaks thanks to the improvement in
detection and isolation techniques.

188

Future studies

The studies presented here were part of a large on-going project. There are still many
unanswered questions that potentially could be investigated with the information and E. coli
isolates gathered in this study. For example, further strain classification by O-typing or other
molecular techniques could allow us to possibly identify risk factors for specific strains. Also we
could possibly identify differences in STEC dynamics (rate loss, rate acquisition) between
different strains. Quantification (enumeration) of bacteria in the samples collected from cattle
could allow for better characterization of shedding dynamics and risk factors associated with
STEC shedding.
By more precisely characterizing the STEC strains isolated in this study, comparative
studies with isolates from human STEC cases in Michigan could be done. This comparison
would help to determine if Michigan’s cattle shed the same STEC strains that caused human
cases. In addition, this is the first study of STEC in cattle in Michigan, so it will be interesting to
determine which virulence factors are present in the STEC isolates we collected and determine if
there is any new virulence factor.
One of the questions that surfaced from results from our study was the differences in
STEC dynamics observed between pasture-based systems and more conventional confinementbased management systems. Future studies should be performed to explore if any these two
production systems favor STEC shedding.
The ultimate goal was to identify potential targets for intervention. From our study, we
identified two groups of dairy cattle that are at higher risk for STEC shedding; first lactation
heifers and cows in the first 30 days of their lactation. It would be valuable to use intervention

189

tools, such as vaccines or probiotics, in these two populations of dairy cattle as a targeted
intervention strategy. It would be also valuable to implement other management practices to
decrease stress in early lactation as this could potentially reduce STEC shedding. This is an
attractive possibility as it focuses on the highest risk populations and thus utilizes valuable
resources most effectively.

190