Survival and Prevalence of Non-O157 Shiga Toxin Producing Escherichia coli (STEC) in Beef Cattle Preharvest and in the Environment by Christine Elizabeth Palmer A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Animal Science Auburn, Alabama May 3, 2014 Keywords: Escherichia coli, non-O157 STEC, survival, pre-harvest, beef cattle Copyright 2014 by Christine Elizabeth Palmer Approved by Luxin Wang, Chair, Assistant Professor of Food Safety/Microbiology Christy Bratcher, Associate Professor of Meat Science Manpreet Singh, Associate Professor of Food Science ii Abstract Cattle are primary reservoirs of shiga toxin producing Escherichia coli (STEC) and are considered ?super-shedders? if they shed more than 104 CFU/g of E. coli O157:H7. In addition to E. coli O157:H7, six non-O157 STEC serogroups ?big six? were added to the zero-tolerance adulterant list in June 2012. Limited information is available on the prevalence of non-O157 serogroups in beef cattle, especially during the pre-harvest stages. Alabama has a large number of cow/calf operations. Most calves are birthed, weaned, and shipped from Alabama to commercial feedlots in the Midwest. Thus, the first purpose of this study was to determine the prevalence of non-O157 STEC among the beef cattle in Alabama. This was accomplished by acquiring fecal samples from cattle at 6 different Alabama Agricultural Experiment Station research centers and testing them for STEC. The second purpose was to identify shedding level changes in calves before and after weaning. For this study, 110 fecal samples were taken from calves located at E.V. Smith research center, both before and after weaning. These samples were tested for the presence of STEC. The third purpose was to isolate and serotype the STEC strains from ?super-shedder? calves. The same E. V. Smith calf fecal samples as were used for purpose 2 were also used for purpose 3. Those fecal samples exhibiting a ?super-shedder? load of STEC had their dominant STEC isolated and sent for serotyping. Based on findings, O26 proved to be the dominant ?big six? member in both the research center and the calf study. STEC O26 isolates also displayed different virulence profiles in a multiplex PCR comparison, targeting three major virulence genes. For these reasons, the iii fourth purpose was to investigate how O26?s possession of different virulence profiles affected its survival in ground beef and different environmental samples. Overnight cultures of O26 strains containing two different virulence gene profiles were inoculated at various concentrations into separate samples of ground beef, cattle feces, bedding material, and trough water. STEC levels were observed over a 10 day period for the ground beef samples and a 30 day period for the environmental materials. Results from these studies show that there were a greater number of positive STEC shedders in research centers sampled during the summer months and in herds containing calves. ?Big six? STEC were isolated from the high shedders at 3 out of 6 research centers. STEC O103 was isolated from one research center and O26 was isolated from two others. The calf study shows that calves begin shedding STEC early in their lives and that weaning does generate an impact on STEC shedding levels. Nine out of 110 calves were found to be ?super-shedders? both before and after weaning. ?Big six? member O26 was also isolated from these ?super-shedder? samples before and after weaning. Results show that virulence genes do not have an impact on STEC?s ability to survive in ground beef and in the environment (P > 0.05), while the inoculation levels and sample types do (P < 0.05). Further research should be conducted on control methods pertaining to the pre-harvest stage of cattle production. From this study?s results, the gut microflora appear to be easily influenced since the dominant STEC before weaning was not the same as after. Also, more emphasis should be placed on studying and controlling STEC O26, as it was the dominant ?big six? member in these studies. iv Acknowledgments First of all, I want to thank my family and friends for their constant love and support. I would also like to thank my major advisor Dr. Luxin Wang for taking a chance on me and for pushing me to always work my hardest over these past two years. I appreciate the guidance of my other committee members Dr. Christy Bratcher and Dr. Manpreet Singh. Their advice and review of my research was most beneficial to me. I would also like to thank Patty Tyler and my coworkers Johnathan Broady and Liang Zhao (Paul) who worked alongside me in the lab and were always willing to lend a hand or give me advice on my various projects. Lastly, I would like to extend a special thanks to the 6 research centers and E.V. Smith who provided me with as many samples as I needed to complete my research. v Table of Contents Abstract???? ???????????????????????????? ...?...ii Acknowledgements? ??????????????????????????? ..?.iv List of Tables?????... ?????????????????????????? ...ix List of Figures? ????. ???????????????????? ?????? ? x List of Abbreviations???????????????????????? ??? .....? .xi Chapter 1. Introduction????????? ????????????????? .? ? ..1 1.1 Non-O157 Escherichia coli????????????????????..? ? .1 1.1.1 Shiga toxin...?.????????????????????...... .?1 1.1.2 ?Big six???? ????????????????????? .?..3 1.1.3 Outbreaks??..????????????????????...? ....3 1.2 Virulence of STEC??????????????????????? ? ....?. 4 1.3 Non-O157 STEC O26????????????????????????? 5 1.4 Sources of STEC?????? ? ??????????????? ....????6 1.4.1 Super-shedders??????????????????? ....??? .7 1.4.2 Survival of STEC in the environment??????????? ....??.. 7 1.4.3 Survival of STEC in food???????????????? ....??9 1.5 Antibiotic Resistance of STEC??????????????????? ? ?10 1.5.1 Antibiotic usage in animals??????????????????10 1.5.2 Antibiotic resistant bacteria formation?????????????...11 vi 1.6 Pre-harvest Management and Control??????????????????.13 1.6.1 Effects of stresses on STEC shedding????????????? ...13 1.6.2 Control Methods???????????????????? ? ?14 1.7 Rapid Detection Methods???????????????.??????? ...16 1.7.1 Polymerase Chain Reaction (PCR)?????..?????????. .16 1.7.2 Pulse Field Gel Electrophoresis (PFGE)?????.??????? ...18 1.8 Summary?. ???????????????????????????? .19 Chapter 2. Materials and Methods???????????????????????? .21 2.1 Collection and processing of samples from the research centers?????.. ?. ? .21 2.2 Sample collection and processing procedures for calf testing????????? .21 2.3 DNA extraction (Simple boiling method)???????? ????????...23 2.4 DNA extraction (Prepman? boiling method)????????? .? ???? .?23 2.5 Multiplex polymerase chain reaction (mPCR)??????? ?????? ..?..23 2.6 Pulse field gel electrophoresis (PFGE)???????????????? ..?.. 25 2.7 Transformation of Escherichia coli cells??????????????? .??29 2.8 Antimicrobial tests????????????????????????? ...31 2.9 Ground Beef 10 day trials????????????????????? ..? ..32 2.10 Environmental 30 day trials??????????????????? .?? ..33 2.11 Statistical Analysis????????????????????????? .35 Chapter 3. Results and Discussion???????????????????????? .36 3.1 Research center studies..???????????????????????..36 3.1.1 Shedding and super-shedding prevalence????????????? .36 3.1.2 PCR examination and serotyping results of isolates from high-/super- shedder cattle ????????? ..............? ?? ???? ? ??? 37 vii 3.2 Calf trials????????????????????????????? ..37 3.2.1 Shedding levels and prevalence changes during weaning?????? ?38 3.2.2Super-shedders???????????????????????...38 3.2.3 PCR results of super-shedder isolates??????????????...39 3.2.4 Serotyping results??????????????????????. .39 3.3 Michigan State reference strains evaluation?????????? ..?????.. 39 3.4 Comparison of O26 isolates with reference strains???????.?... ????.40 3.4.1 PFGE comparison of O26 isolates????????????????40 3.4.2 Antimicrobial test results???????????????????..41 3.5 Survival studies???????????????????????????4 1 3.5.1 Ground beef study??????????????????? ?? ?.4 1 3.5.2 Environmental study??????????? ..?????????.. .42 Chapter 4. Conclusion?????????????????????????????45 4.1 Research center studies?? ..?????????????????????..45 4.2 Calf trials?????????????????????????????. .45 4.3 O26 isolate evaluation??????????? ?????????????. .46 4.4 Survival studies???????????????????????????46 4.5 CHROMagar? STEC????????????????????????.46 4.6 Overall conclusion?????????????????????????. ..47 References???????????????????????????????? .....49 Tables?????.. ......????????????????????? ??????? .55 Figures? ?????????????????????????????????....66 viii List of Tables Table 2.1: Volumes of reagents and primers for PCR????????????????...53 Table 2.2: PCR primer sequences???????????????????????......54 Table 2.3: Antimicrobials used???????? ?????????????????..55 Table 3.1: Research center results? .??????????????????????? 56 Table 3.2: Research center serotyping results?? ..??????????????? .?.. ...57 Table 3.3: Enumeration of STEC from calf trial??????????????????...58 Table 3.4: Shedding levels and STEC prevalence in calves at 6 and 9.5 months of age???..59 Table 3.5: Enumeration of STEC from ?super-shedders? in calf trial??????????...60 Table 3.6: Serotyping and PCR results for 9 consistently high ?super-shedder? calves???... 61 Table 3.7: Serotyping results for consistently high ?super-shedder? calves????????.62 Table 3.8: Isolates from the STEC Center at Michigan State University?????????..63 Table 3.9: Antimicrobial resistance results????????????????????....64 ix List of Figures Figure 3.1: Alabama research centers??? ..???????????????????..65 Figure 3.2: PCR results for research center samples sent for serotyping? ..??????... ? .66 Figure 3.3: Shedding level for calf trial?????????????????? ...?? ......67 Figure 3.4: PCR of the 9 ?super-shedder? isolates from the calf trial, sent for serotyping?? ...68 Figure 3.5: PCR of ?big six? clinical isolates from Michigan State University STEC center?..69 Figure 3.6: PCR of 4 farm-isolated O26 strains and 2 clinical-isolated O26 strains???... ? .70 Figure 3.7: PFGE results from 4 farm and 1 clinical-isolated O26 strain?????????.71 Figure 3.8: Dendrogram of PFGE fingerprinting profile???????????????...72 Figure 3.9: GFP transferred bacteria on supplemented LB agar ?????????? ?? ..73 Figure 3.10: Ground beef trial????????????????????????? ...74 Figure 3.11: Fecal environmental trial????????????????????? .......75 Figure 3.12: Bedding environmental trial?????????????????????.. 76 Figure 3.13: Water environmental trial?????????????????????? ..77 x List of Abbreviations CDC Center for Disease Control Ct Cycle threshold CFU Colony forming unit DNA Deoxyribonucleic acid EDTA Ethylenediaminetetraacetic acid EHEC Enterohemorrhagic Escherichia coli FSIS Food Safety and Inspection Service HC Hemorrhagic colitis HUS Hemolytic uremic syndrome GFP Green fluorescent protein LB Luria Bertani OD Optical density PCR Polymerase Chain Reaction PFGE Pulse Field Gel Electrophoresis RMAC Ramnose-MacConkey agar SDS Sodium dodecyl sulfate STEC Shiga toxin Producing Escherichia coli TBE Tris borate EDTA TE Tris EDTA xi TSA Trypticase soy agar TSB Tryptic soy broth USDA United States Department of Agriculture VTEC Verotoxigenic Escherichia coli 1 Chapter 1. Introduction 1.1 Non-O157 STEC Escherichia coli Escherichia coli is a very widely studied organism, but there is still a great deal unknown about the Enterobacteriaceae family. E. coli is a Gram negative, rod shaped, non-spore forming, facultative anaerobe, normally found in the lower gut of warm-blooded organisms. Most E. coli are part of the normal gut microflora and harmless to the human and animal population. One function of these is to inhibit other pathogenic bacteria from inhabiting the gut. There are some which have gained more notoriety because of their association with human illness such as the Verocytotoxigenic E. coli (VTEC), also known as shiga toxin- producing Escherichia coli (STEC). These E. coli have gained virulence factors via plasmids, bacteriophages, transposons, etc. which make them harmful to humans (Lim et al, 2010). The names Verotoxin and Shiga toxin have been used interchangeably and are related to cellular cytotoxin production (Mathusa et al, 2010). Both shiga toxin and shiga-like toxin come from a group of protein toxins that will bind to the surface of the cell, gain entry, and once in the cytosol, will enzymatically inhibit protein synthesis or induce apoptosis. 1.1.1 Shiga toxin Shiga toxin is made up of one catalytic A subunit and five B subunits which are capable of binding to receptors on a cell?s surface. Once the B subunits have bound, the A subunit, an N- 2 glycosidase, will enter the cell using endocytosis and venture into the cytoplasm. Here it inactivates the RNA component of the ribosome, halts protein synthesis, and causes the death of the cell (Cary et al, 2000). Shiga toxin works to attack the blood vessels in the intestinal wall, causing hemorrhaging of the mucosa and bloody feces. In more severe cases, the shiga toxins will attack the endothelial cells of the kidneys. They are thought to gain access to the kidneys via the blood stream with their entry point being the intestinal wall. The shiga toxin is attracted to specific Globo series glycolipids like Globotriacylceramide (Gb3), which contain Galactose receptors. These are found in great quantities in the kidneys, specifically the glomerulus region, well known as the filtering system of the kidneys. The invasion and death of these endothelial cells can lead to clotting and renal failure in severe cases of infection (Cary et al, 2000). The most common source of Shiga toxin is Shigella dysenteriae and a shiga-toxin producing Escherichia coli like O157:H7, O26, and other Enterohemorrhagic E. coli (EHEC). These EHEC are a subset of STEC which are known to cause bloody diarrhea, hemolytic uremic syndrome (HUS), hemorrhagic colitis (HC), fever, vomiting, and possible death in people who ingest them (Brooks et al, 2005; Mathusa et al, 2010). Symptoms of HUS are thrombocytopenia, microangiopathic hemolytic anemia, and acute renal failure, which can lead to death, especially in children (Etcheverria et al, 2013). According to Etcheverria et al. (2013), STEC usually affect children under 5, elderly, and people with immune systems which have been compromised. There are two types of shiga toxin found in EHEC: shiga toxin 1(stx1) and shiga toxin 2 (stx2). Shiga toxin 1 is associated with less severe cases of illness and more closely related to the shiga toxin in S. dysenteriae, differing by only one amino acid. Serologically, they cannot be differentiated from one another, but stx2 is less closely related to the S. dysenteriae shiga toxin 3 and is known to cause more severe symptoms in the people it infects (Boerlin et al, 1999). It shares only 56% of the DNA sequence homology with stx1. 1.1.2 ?Big six? In past decades, studies on STEC have focused on E. coli O157:H7, simply due to its earlier discovery in human illness (Monaghan et al, 2011) and its involvement in the 1982 Michigan and Oregon outbreaks (Lim et al, 2010). Unfortunately, there is very little information on non-O157:H7 shiga toxin-producing E. coli (STEC), not because they lack involvement in human illness. More than 100 serotypes have been linked to cases of human illness, however, six have recently made a big impression. They are: E. coli O26, O111, O103, O121, O45, and O145 (Brooks et al, 2005). These six serotypes are now of greater importance than O157:H7 and have been placed on the zero tolerance adulterant list as of 2012 (USDA, 2011). According to the new regulation, any raw non-intact beef trimmings, both domestic and international, found to have non-O157 STEC must be considered adulterated and removed from the food chain (Stevens, 2012). For STEC, an infectious dose is thought to consist of less than 10 cells, so it takes only a few live pathogenic bacteria cells to make a human sick (Etcheverria et al, 2013; Tilden et al, 1996). Knowing that, the USDA has attempted to identify those STEC serogroups responsible for the most human illnesses and keep them from entering the food chain through systematic tests and recalls, starting with beef trimmings. 1.1.3 Outbreaks Non-O157 STEC have been responsible for many illnesses and outbreaks all around the world. In 2006, STEC O103:H25 was involved in an outbreak which took place in Norway 4 (L'Ab?e-Lund et al, 2012). In 2011, there was a major outbreak involving STEC O104:H4, leaving 53 dead, 3,000 with acute gastroenteritis, and 600 with HUS (Hauswaldt et al, 2013). There was also a recent multistate outbreak of STEC O102 infections in the United States in March of 2013. The non-O157 STEC serogroups involved in these outbreaks cannot yet be rapidly identified and even though the number of clinical labs which test for them have increased, some still do not perform the required tests to identify these STEC. For these reasons, and according to an article on the Centers for Disease Control and Prevention website, many non-O157 STEC infections go unidentified and unrecorded (CDC, 2013). As of 2012, the CDC estimates that STEC cause 265,000 illnesses per year, 168,698 due to non-O157 STEC (CDC, 2012). Estimates of illnesses due to non-O157 STEC are quite possibly on the low side. 1.2 Virulence of STEC Some people have noticed a problem with the STEC classification system and believe virulence should play a larger role in it (Mathusa et al. 2010). Instead of focusing on the virulence profiles to determine whether or not an organism is pathogenic, STEC, in the past, has been identified as pathogenic based solely on serotype (Mathusa et al, 2010). This is an inefficient system. For example, not all E. coli O103:H2, a serogroup on the zero tolerance list, are pathogenic. The public and laboratory workers only looking to identify serogroups will classify these STEC as harmful to the public when they may not be. More focus should be placed on the virulence of each strain rather than branding the whole serogroup as pathogenic. There are several virulence factors involved in the pathogenicity of non-O157 STEC. The presence of stx1, stx2, or a combination of the genes is the most important indicator as to whether or not a bacterium is virulent (Etcheverria et al, 2013). Strains containing stx2 are 5 considered more virulent than those containing stx1 and are more often associated with HUS (Etcheverria et al, 2013). The second most important virulence factor is the eae intimin gene, which is used by the bacteria to attach to epithelial cells, inducing an attaching and effacing (A/E) lesion on the surface of the target cell (Etcheverria et al, 2013). Even though the presence of virulence genes such as eae is a good indicator of pathogenicity, its absence does not necessarily mean it is not pathogenic (Beutin et al, 2006). The presence of intimin is associated with the more severe cases, but there have been less severe cases of sickness caused by intimin- negative STEC (Beutin et al, 2004). The pathogenic non-O157 STEC were previously unrecognized as harmful in the food industry. This may be because they did not always have the virulence that they do today, or recently advanced methods of rapid detection now allow for the identification of these pathogenic bacteria that may not have necessarily been a concern in the past (Beutin et al, 2006). Research has been traditionally focused on O157:H7 (Beutin et al, 2006). Since 2000, researchers have become more aware of the existence of non-O157 STEC and testing has become more frequent, so it may only seem like these bacteria have become a greater problem these past few years (Mathusa et al, 2010). 1.3 Non-O157 STEC O26 Among the 6 STEC strains, a lot of the non-O157 STEC studies have been focused on non-O157 STEC O26 because it is notorious for having those virulence factors which can make people sick (Mathusa et al, 2010). One study in particular examines how the O26 virulence profile has changed over the years. The results from a study by Zhang et al. (2000) show that all O26 isolates collected between 1965 and 1994 expressed only stx1 while most O26 isolates 6 collected between 1995 and 1999 had either stx2 alone or a combination of stx1 and stx2. This new virulence profile was found to have a ?high pathogenic potential for humans? so more research is conducted on this STEC (Mathusa et al, 2010). Another study in France emphasizes the importance of researching the survival of non-O157, especially STEC O26, in the soil and environment (Fremaux et al, 2007). The researchers found that O26 has survivability similar to O157:H7 and its ability to survive for a very long time, up to and possibly more than 18 weeks in cow feces at 15?C (Fremaux et al, 2007). 1.4 Sources of STEC Outbreaks are most often due to the consumption of beef, dairy, or vegetable products which have come in contact with cattle feces (Etcheverria et al, 2013). Ruminant animals like cattle are reservoirs for STEC (Mathusa et al, 2010). STEC which can cause illness in humans are not normally harmful to the adult cattle that harbor them. Some serogroups have been the cause of diarrhea in calves, but nothing more serious than this has been observed (Shaw et al, 2004). Beef carcass processing has been scrutinized as a major contamination control point (Mathusa et al, 2010). During the slaughter process, pathogens from the feces, hides, or environment can contaminate the carcasses (Midgley and Desmarchelier, 2001). STEC contamination is not limited to cattle products, but there have been a higher frequency of cases involving beef or veal (Mathusa et al, 2010). Most illnesses associated with meat are due to the consumption of undercooked products (Soon et al, 2011). Even though these products are considered safe when prepared properly, such as cooking the meat to an internal temperature which would eliminate all possibility of illness, customers continue to demand a safer product. Even though there can be possible cross-contamination in the slaughter facility, 7 STEC do not originate there. Scientists can often link the STEC responsible for these illnesses back to the farm they originated from using a fingerprinting method call Pulse Field Gel Electrophoresis (Soon et al, 2011). 1.4.1 Super-shedders According to Chase-Topping et al. (2008), the definition of super-shedding cattle are those which shed 104 CFU (colony forming unit)/g STEC O157:H7 or greater in their feces. This definition has also been applied to non-O157 STEC (Menrath et al, 2010). Super-shedder cattle are thought to be the major cause of higher transmission rates and cross-contamination within cattle producing environments (Arthur et al, 2013). Therefore, focusing research on reducing the number of STEC that these individual super-shedders put out will greatly reduce the contamination of the production environment (Arthur et al, 2013; Menrath et al, 2010). A study by Baines and Erb (2013) suggests that calves infected with stx-positive strains of STEC tend to shed significantly more STEC in their feces than those infected with stx-negative strains of STEC. The presence of the stx gene could possibly have an impact on the STEC survival in the cattle gut. More research needs to be done on methods for reducing the number of non-O157 STEC in these super-shedders because it could serve as a good control point before slaughter (Menrath et al, 2010). 1.4.2 Survival of STEC in the environment Non-O157 STEC has seasonality as its numbers are highest between June and September and its peak has been observed in August (Brooks et al, 2005). Farmers and food industry workers should be aware of the greater chance of contamination during the summer months and 8 know that this is the time to take more precautionary measures. In several studies, the ability of E. coli O157:H7 and non-O157 STEC to survive long-term in feces has been observed. This being said, irrigating with a manure mixture or using the same field for another group of cattle could lead to contamination of that vegetable crop or its new occupants (Fremaux et al, 2007; Fukushima et al, 1999). In fields, fecal material has been shown to harbor E. coli for at least 150 days (Alexander et al, 2008). This also has implications for feedlots, transport trailers, and holding pens at slaughter facilities. At 15?C, cross contamination can easily happen 16-18 weeks after the cattle who shed the bacteria are gone (Fukushima et al, 1999). There have also been studies on clay and sandy soils containing non-O157 STEC and it took 40 days and 62 days for there to be a 1 log reduction of bacteria in clay and sandy soil respectively (Bolton et al, 2011). The clay had a faster reduction time because of its slightly acidic nature, which makes it a less favorable environment for STEC (Bolton et al, 2011). In another study, clay-ridden soils offered a more suitable environment for STEC because it is better able to hold bacteria and it creates a shield against microbial predators (Fremaux et al, 2007). These studies show that STEC are able to persist for an extended period of time in soils and this could be a detriment to human health. Water troughs which have been contaminated with fecal material can also be a point of cross-contamination for cattle (Midgley and Desmarchelier, 2001). Many studies have shown that E. coli has an excellent survivability in the environment for an extended amount of time and that it is also a good source of antibiotic resistance, especially on a farm (Alexander et al, 2008). These antimicrobial resistant bacteria can contaminate any new residence of the field, transferring the resistance to the microbes in the gut of the next occupants. Because E. coli is a prominent bacterium in the cattle gut and is known to readily transfer resistance to other bacteria, 9 it can serve as a reservoir for antimicrobial resistance genes within the lower intestines of cattle (Alexander et al. 2008). 1.4.3 Survival of STEC in Food Certain factors can influence the survival of STEC in foods such as temperature, pH, salt levels, and water availability (Mathusa et al, 2010). Studies have examined different food items for the presence of non-O157 STEC and have discovered them at various intensities in salads, milk, beef, pork, lamb, poultry, seafood, and cheeses (Mathusa et al, 2010). A study by Samadpour et al. (1994) searched for the stx genes in food samples and these researchers found that veal had the highest count (63%) and lamb followed with 48% of the samples being positive for stx. Some STEC are acid tolerant, surviving in low pH environments (Mand et al, 2012) and in foods like apple cider and fermented meats. A lot of preserved and processed foods have organic acids present, which some non-O157 STEC have the ability to adapt to, given the opportunity (Brudzinski and Harrison, 1998). By the mere fact that a low number of inoculants can make a person sick, they must have a resistance to the normal acid conditions of the human stomach (Cary et al, 2000). STEC strains are also very resistant to stressful conditions like high pH and salt concentrations, (Elhadidy and Muhammed, 2012) making it more difficult to find effective control measures in food industries, and improving the chances that they will come in contact with humans. As for treatments on foods for the prevention of STEC growth, Gilbreth et al. (2009) looked into how O157:H7, O145, O111, and O26 were affected by antimicrobial treatments that would be sprayed on the surfaces of meat. After treating the contaminated meat with a solution of acidified sodium chloride, octanoic acid, and peracetic acid, the researchers concluded that 10 there was no large difference in sensitivity to the treatment (Gilbreth et al, 2009). They also concluded that those treatments previously used for O157:H7 control could be just as effective on non-O157 STEC (Gilbreth et al, 2009). 1.5 Antibiotic resistance of STEC 1.5.1 Antibiotic usage in animals Antibiotics have been used since the 1950?s to increase feed efficiency, for therapy, and to prevent disease in cattle during certain times when risk of disease is high (Alexander et al, 2008). The use of antibiotics like chlortetracycline on a regular basis as a growth promoter contributes to the rise in new antibiotic resistant E. coli pathogens (Alexander et al, 2008). For this reason, the European Union in 1999 banned the use of antimicrobial growth promoters (Casewell et al, 2003). Oxytetracycline has been used to treat bacterial disease in cattle (Alexander et al, 2008). Many researchers are concerned about antibiotics used in both human and animal medicine such as tetracycline (Alexander et al, 2008). They fear that overusing these antibiotics in animals will result in bacterial resistance which could quite possibly transfer to humans and make human treatments with the same or similar antibiotics ineffective. Tetracycline is known to have a broad ranging effect on both Gram-positive and Gram- negative bacteria, and different types of it are used as growth promoters, and to treat a whole host of bacterial infections (Alexander et al, 2013). Like tetracycline, both ampicillin and streptomycin have been used as a broad spectrum antibiotic against Gram positive and negative bacteria. Streptomycin has a history of being used to treat TB in humans and is also used in Veterinary medicine for large animals. Ampicillin has been used for many years to treat bacterial 11 infections in humans. It is cause for concern when studies like Srinivasan et al. (2007) are finding high numbers of multidrug resistant STEC coming from beef and dairy cattle in the United States. Another antibiotic of concern is cephalosporin. Along with subsequent generations of this antibiotic, it has been used in human medicine as a broad spectrum treatment against both Gram positive and negative bacterial infections. There has been a noticeable increase in the number of cephalosporin resistant E. coli in the soil. Ceftiofur, a third generation cephalosporin, is used to treat foot, uterus, and respiratory infections in cattle. Due to its convenience and effectiveness, it is used a great deal by farmers. With its greater usage, scientists have been seeing an increase in enteric bacteria with ceftiofur resistance (Subbiah et al, 2012). 1.5.2 Antibiotic resistant bacteria formation Bacteria can gain resistance naturally through random genetic mutation. Resistance can also be generated by the application of a stress on a population of bacteria (de Verdier et al, 2012). In this case, only those with the resistance gene can survive, multiply, and pass their genetic code onto subsequent generations and other bacteria. Bacteria can transfer the resistance to daughter cells by vertical gene transfer or to other bacteria in the vicinity through horizontal gene transfer involving a plasmid exchange (Price et al, 2008). Antibiotics and antibiotic resistant bacteria can affect humans in a few different ways. When contaminated food is eaten, the resistant bacteria from the food may transfer resistance to other bacteria in the human gut (Sahoo et al, 2012). A study in Scotland reveals that the ?release of free stx-encoding bacteriophages into the intestinal environment of cattle by STEC O157:H7 or other EHEC may permit horizontal transfer of virulence genes to E. coli O26,? one of the six 12 zero tolerance E. coli (Chase-Topping et al, 2012). This emphasizes the possibility of transference within the intestines. Antibiotic residues in consumed food can also influence the human gut microflora by encouraging a natural selection of resistant bacteria (Sahoo et al, 2012). Gram negative bacteria such as E. coli are common inhabitants in the intestines of warm-blooded animals; these bacteria can often be carriers and conveyors of antimicrobial resistant genes, reflecting the intestine?s past exposure to antimicrobials (de Verdier et al, 2012). Another avenue of obtaining resistant bacteria is through its excretion and interaction with the environment (Sahoo et al, 2012). Resistant bacteria shed from the human or animal gut into the soil or water could interact with other E. coli in the environment (Sahoo et al, 2012). There is research by Yue et al, (2012) which says phage mediated gene transfer events can happen in the environment with the help of temperature and UV light. Feedlot sludge could also be a good environment for horizontal gene transfer, leading to the creation of new pathogenic bacteria if those genes contain the right virulence factors (Yue et al, 2012). According to S?enz et al, (2004), normal microflora ?play(s) a key role as an acceptor and donor of transmissible antimicrobial resistance mechanisms.? It is difficult to pinpoint exactly which antimicrobials most STEC are resistant to because resistance is different for every region, every animal, and every strain that is examined. It is also hard to compare different studies when they are not testing the same antimicrobials on STEC isolates. A few studies though, have noticed a high resistance to different forms of streptomycin, tetracycline, and ampicillin, among others (Mora et al, 2005 and Sasaki et al, 2012). Sasaki et al. (2012) noted, when comparing STEC O157:H7 isolates to O26 isolates in beef cattle, that the O26 strains had a significantly higher antimicrobial resistance rate. Several studies also mention an association between antimicrobial resistance and having the eae and stx virulence genes 13 (Mora et al. 2005; Sasaki et al. 2012). Mora et al. (2005) noticed a relationship between non- O157 STEC being resistant to multiple antimicrobials and also having the eae and stx1 virulence genes. Sasaki et al. (2012) had similar findings and suggests that the antimicrobial resistance has helped the O157:H7 containing both stx1 and stx2 genes to grow considerably greater in number. 1.6 Pre-Harvest Management and Control 1.6.1 Effects of stresses on STEC shedding There are various stresses which can affect the normal physiological/psychological homeostasis of an animal. For cattle, some of these stresses are weaning, transportation, fasting, diet change, and handling (Rostagno and Marcos, 2009). Such stressors will affect the gastrointestinal tract, causing an increase in pH, intestinal motility, and passage of foodborne pathogens (Rostagno and Marcos, 2009). Weaning alone provides a number of stressors because farmers most often do it abruptly, unlike the natural, gradual process of weaning. This abrupt switch to a new diet, new environment, and new social situation provides great stress to the young animal. Although it is a very stressful event, it is necessary for cattle operations to do this abrupt weaning so that they can get their breeding herd pregnant again as soon as possible (Enriqeuz et al, 2011). Some producers like to transport their calves right after weaning to save on feed costs. Enriquez et al. (2011) suggests exposing the calves to certain things before weaning like the new feed, handling, and the new environment to reduce the amount of stress on the animal. With an increase in stressors, the number of animals carrying foodborne pathogens will also increase because it suppresses their immune systems, making the GI tract more open for colonization (Salak-Johnson and McGlone, 2007). This could very well cause a shift in dominant 14 gut microflora (Edrington et al, 2011). Calves in particular are more susceptible to STEC infection and the possible development of chronic scours when stressed (Baines and Erb, 2013). While they are still young, their gut microflora is still being established and can easily be influenced. With an increase in shedding and an increase in the number of cattle open to colonization, more of the herd may be involved in cross-contamination. Also, with more fecal shedding and possible close contact with each other, more hides can become contaminated (Rostagno and Marcos, 2009). 1.6.2 Control methods Due to the importance of STEC, researchers in the past have focused on finding control measures for reducing the E. coli contamination after slaughter with little headway recently (Etcheverria et al. 2013). Therefore, focus has switched to controlling pathogenic populations in and on the cattle before slaughter. If this can be done effectively, there will be minimal carcass contamination once the cattle reach the slaughter facility. This pre-harvest intervention, combined with controls already in place at the slaughter plant, should create an effective resistance against pathogenic non-O157 STEC. Research has been done on controlling the number of Escherichia coli shed in the feces of cattle destined for slaughter by changing their diet and looking at how fasting affects the bacterial count. Cattle have shown a significantly lower concentration of E. coli in their feces when fed a high roughage diet for an extended period of time, but fasting can reverse the effects of the roughage diet, causing an increase in E. coli shedding only 24 hours after being implemented (Jordan et al, 1997). Also, although a study which fed cattle a high roughage diet 4 days before slaughter showed a significant decrease in fecal E. coli counts, it is neither practical 15 nor cost-effective for feedlot owners to switch to a diet that would reduce the rate of weight gain in their cattle (Jordan et al, 1997). Some vaccines have been developed to reduce STEC shedding in cattle. One in particular is based on the type III secreted proteins; another uses the adherence proteins intimin, Tir, and EspA (Etcheverria et al, 2013). The type III secretion system is specific to Gram negative bacteria and their ability to secrete the toxins which make people sick. Midgley and Desmarchelier. (2001) recommend some measures to be taken at the slaughter facility to reduce the spread of STEC. Providing clean feed and water to each new group of cattle, cleaning out fecal material in between groups, and holding high risk(calves) and low risk cattle(adults) in separate pens would greatly reduce cross contamination among the cattle (Midgley and Desmarchelier, 2001). Probiotics and prebiotics have also been implemented in an attempt to control STEC shedding. A study by Baines and Erb. (2013) investigated a prebiotic called Celmanax? on calves during the spring and summer months to see if it reduced STEC shedding. Celmanax? reduced the number of STEC shed and prevented the acquisition of new strains during the spring, but there was no difference seen in the summer when the calves were shedding significantly more STEC (Baines and Erb, 2013). The authors indicated that the stressful feedlot environment may have also had an influence on fecal shedding (Baines and Erb, 2013). An interesting study by Rund et al. (2013) looked at the probiotic effect of E. coli strain Nissle 1917 (EcN) on two O104:H4 isolates involved in the large Germany outbreak in 2011. In culture, it greatly reduced the adherence, growth, and stx2 production of the O104 isolates (2013). These promising results indicate that there is room for further advancement in the realm of probiotics to fight non-O157 STEC. 16 1.7 Rapid Detection Methods Non-O157 STEC do not have distinct biochemical markers such as an inability by O157:H7 to ferment sorbitol or Beta-glucuronidase activity (Fratamico et al, 2009; Fremaux et al, 2007). This is what makes studying these bacteria difficult; such identifiers would help with screening for these pathogens. There are multiple media types currently being used to isolate these STEC. One uses a base of MacConkey agar and this is supplemented with such things as sucrose, bile salts, sorbose, novobiocin, potassium tellurite, and a carbohydrate mixture (Mathusa et al, 2010). This has been used to isolate 4 specific serotypes: O26, O103, O111, and O145. They will show up as different colored colonies, which helps with identification. Another agar type being used is a chromogenic agar (Mathusa et al, 2010). This helps with isolating STEC in general, but when differing serotypes are plated on this agar, they cannot always be differentiated from one another and may appear differently based on incubation time and the number of colonies on the plate. Most colonies will appear mauve in color. A third agar being used is called Rainbow agar (Mathusa et al, 2010). STEC O26, O103, and O121 will appear pink; O45, O111, and O157 will show up as grey, light blue, or purple, and O145 may appear dark blue. An agar developed specifically for isolating E. coli O26 is called rhamnose-MacConkey agar (RMAC), based on O26?s ability to ferment rhamnose, while other STEC cannot. STEC O111 has been effectively isolated on a chromocult agar mixed with cefixime, cefsulodin, and vancomysin (Mathusa et al, 2010). Colonies appear purple on this agar. 1.7.1 Polymerase chain reaction (PCR) 17 PCR has been used to test for the presence of the shiga toxin 1(stx1) and shiga toxin 2(stx2) genes, which would indicate the presence of STEC with the potential to be pathogenic. The polymerase chain reaction amplifies a certain length of DNA millions of times in a very short amount of time. The thermo-cycler will cycle through various temperatures to create the right environment for amplification. First, the machine raises the temperature to 94-96?C, which denatures or breaks up the DNA strands into single-stranded DNA chains. Next, the temperature is lowered to 50-65?C so that the forward and reverse primers can attach or anneal to the complimentary length of base pairs on the DNA strands. After the annealing step, the temperature is raised to 72?C for 1 minute or more, allowing Taq polymerase to bind to each priming site and extend/synthesize a new DNA strand. The thermo-cycler then raises the temperature once again to 94-96?C to denature the DNA and start the cycle over again. This goes on for about 30 cycles. If the annealing temperature is too high or too low, the primers will either not bind at all or lose specificity, binding to other sites of similar sequence. Specificity is the propensity for a primer to bind to a specific target region of the DNA strand and not to other non-specific target areas (Bartlett and Stirling, 2003). This can be difficult when working with multiplex PCR, as there will be multiple primers with different melting temperatures. In preparing the DNA for PCR, the DNA must first be extracted from the cells. A simple cell suspension in water and boiling for 10 minutes will work, but it does not give the best results. Suspending the cells in the PrepMan Ultra reagent produces a purer sample of DNA. It also puts out a higher concentration of DNA so that less template needs to be used to produce the same results (Wasilenko et al, 2012) Using a TE buffer method of DNA extraction does not give as pure, or as highly concentrated a sample as PrepMan, but it does give a lower cycle threshold (Ct) value when used in real time PCR. Lower Ct values indicate strong positive reactions, so in 18 the interest of saving money, it is acceptable to use TE buffer for real time PCR preparation or the simple water boiling method for regular PCR DNA extraction. Research studies have been devoted to finding a reliable rapid detection method for identifying STEC strains in meat products and feces. PCR has become a common practice in microbiology/food safety labs, using the amplification of certain known virulence genes to test for the presence of certain pathogens. Within the same serogroup such as O26, there can be stx- positive and stx-negative versions of the bacteria (Fratamico et al. 2009). Most of the time, researchers are concerned with those containing the stx genes because they contain the means to make the toxin which makes us sick. Some researchers also use wxz and wxy, which are o- antigen gene clusters specific to the antigen and valuable for identifying individual O serogroups (Wang et al. 2010). If the goal is to identify a specific, already known serogroup within a material, the best method would be to use the wxz and wxy gene primers. This is not an effective method if the serogroup(s) is/are unknown. 1.7.2 Pulse Field Gel Electrophoresis (PFGE) In addition to PCR, another technology has been widely used to better identify the contamination sources, Pulse Field Gel Electrophoresis (PFGE). PFGE is used for separating large pieces of DNA in a gel, similar to a regular electrophoresis technique. PFGE periodically switches voltage paths in three different directions, but the DNA continues to run along one central axis. This helps pull the larger DNA fragments apart. In preparing the DNA, the protocol calls for the lysis of the organism imbedded in agarose and the usage of restriction enzymes which cut the DNA at certain restriction sites, resulting in 10 to 20 fragments. These fragments will be loaded in an agarose gel and pulled apart in the electrophoresis machine over a period of 19 approximately 19 hours. Larger DNA fragments will be located towards the top of the gel and smaller ones will be located towards the bottom. PFGE is often used to compare strains of the same serogroup. If the strains are exactly the same, the enzymes will cut the DNA strands at the same points, resulting in fragments of the same size, which will appear at the same points along the electrophoresis gel. Researchers like Miko et al. (2013) use PFGE to compare the genetic similarity of pathogenic outbreak strains to other outbreak strains or other farm strains. This way, strains that make humans sick can be traced back to the farm they originated from. After identifying the source of contamination and investigating the operation?s regular practices, more control measures may be needed to prevent any future pathogens from entering the food chain at that particular facility. If necessary, PFGE can provide the evidence needed for the recall of a contaminated product. 1.8 Summary E. coli are natural inhabitants of the gastrointestinal tracts of warm-blooded animals such as cattle, sheep, humans, etc., but there are a few specific strains in the EHEC subgroup which have been known to make people sick. Sickness can vary in severity depending on what virulence genes the strain contains and the state of the infected person?s immune system. STEC strains containing the stx2 gene have been associated with more severe cases of sickness, as these are more likely to have HUS. The first STEC to be investigated and identified as harmful was E. coli O157:H7, but recently, 6 non-O157 STEC have become a more important public health concern and have been placed on the zero tolerance list. 20 There is no well-established method for testing for these STEC, but some rapid detection methods like PCR and PFGE are currently being implemented in combination with different selective and differential medias like Rainbow, Chromogenic or a MacConkey-based agar. Many scientists are presently researching new methods for isolation and identification of the non-O157 STEC, especially the ?big six.? The classification of pathogenic STEC must also be taken into consideration when trying to identify these STEC. Identifying them as virulent based on their serotype may not be the best course of action when individual strains have been shown to vary in their virulence profiles. Another major topic of concern, related to these STEC, is the control of antimicrobial resistant bacteria. E. coli is known to readily transfer resistance to other bacteria and since it resides naturally in the gut, E. coli in general is a good reservoir for antimicrobial resistant genes. It has been suggested that the best course of action for controlling these pathogens is to use a combination of pre-harvest and post-harvest control measures. Using vaccinations, probiotics, prebiotics, proper herd management, and studying ?super-shedding? cattle are all being researched for improving pre-harvest control. Post-harvest control measures, such as acid washes used for O157:H7 control, have been fairly effective against non-O157 STEC. More research needs to be conducted on the pre-harvest management side to find the best combination of control methods. Advancement in identifying and isolating these non-O157 STEC is also a necessary endeavor. 21 Chapter 2. Materials and Methods 2.1 Collection and processing of samples from the research centers Six Alabama Agricultural Experiment Station research centers were sampled including Wiregrass, Sand Mountain, Gulf Coast, Tennessee Valley, Upper Coastal Plain, and Black Belt research centers. Approximately 30 fecal samples were taken rectally from cattle at each of these research centers and shipped back to the lab in coolers. These samples were processed immediately upon arrival. Twenty five grams of each sample was measured out into a Whirl- pak? bag (Nasco, Fort Atkinson, WI) and 100ml of 0.1% peptone water was added to this. Samples were then stomached for 2 minutes in a Smasher? lab blender (AES, Hamilton, NJ). Two 100?l samples of the slurry were then plated in duplicate on two plates of CHROMagar? STEC (CHROMagar, Paris, France). Plates were incubated overnight at 37?C and enumerated the following day. Positive STEC colonies on the agar were expected to display a mauve or pale purple color and negative colonies were expected to be colorless, blue, or inhibited. 2.2 Sample collection and processing procedures for calf testing Fecal samples were taken from 110 beef calves at both 6 months (before weaning) and 9.5 months of age (after weaning). At the E.V. Smith research center, fecal samples were taken rectally from the cattle and put in 100ml sterile specimen containers (VWR? Radnor, PA). ID numbers were written down in a notebook and on each sample cup. These numbers were located on the cattle?s ear tags. 22 Samples were placed in a cooler on ice, and taken back to the food microbiology lab in Upchurch hall for further processing. After arriving at Upchurch Hall, 25g samples of feces were removed from each cup using a clean spatula and place in a labeled, filtered Whirl-pak? bag (Nasco, Fort Atkinson, WI). 112.5 ml of 0.1% peptone water was added to the filtered Whirl-pak? bag (Nasco, Fort Atkinson, WI) and then the mixture was placed in a smasher for 2 minutes at regular speed. All fecal samples were processed as described above. Two CHROMagar? STEC (CHROMagar, Paris, France) plates, two CHROMagar? O157 (CHROMagar, Paris, France) plates, and one MacConkey plate was plated for each sample. The plates were labeled with the sample number and then 100?l of each sample was pipetted from the Whirl-pak? bags (Nasco, Fort Atkinson, WI) bag onto each plate and spread evenly. Plates were allowed to sit for 10 minutes before being inverted and placed in a 37?C incubator for 24 hours. Lastly, 111.5ml of the 2xlactose broth was added to each Whirl-pak? bag (Nasco, Fort Atkinson, WI) containing a fecal sample. These were enriched for 24 hours at 35?C. After 24 hours, the positive colonies were counted. For any samples that were completely negative for STEC or O157, samples were taken from the enriched, incubated Whirl-pak? bags (Nasco, Fort Atkinson, WI) and streaked onto the appropriate agar to determine if the samples were positive for STEC or O157:H7. These plates were put in the incubator at 37?C for 24 hours. STEC plates from super-shedder cattle were saved. Three positive, well-isolated colonies were picked from each plate and streaked on CHROMagar? O157 (CHROMagar, Paris, France) plates to make sure they were not O157:H7. Once those colonies were confirmed to be non-O157 STEC, one was transferred to trypticase soy agar (TSA) slants for further analysis. 23 2.3 DNA extraction (Simple boiling method) Purified non-O157 isolates were then tested for the presence of shiga toxin 1(stx1), shiga toxin 2 (stx2), and intimin (eaeA) genes. Overnight fresh broth cultures were obtained for each isolate and 1ml of each culture was washed and resuspended in 100?l of ultrapure water. The resuspended cultures were boiled for 10 minutes and centrifuged for 2 minutes at 12000rpm. The supernatant containing DNA was then transferred to new tubes and stored at -20?C or used right away for PCR. 2.4 DNA extraction (Prepman? boiling method) Procedures were the same as the simple boiling method except for one part. Instead of resuspending the pellet the second time with 100?l of ultrapure water, 100?l of Applied Biosystems? Prepman? Ultra (Life Technologies, Carisbad, CA) was used. The Prepman? Ultra was used for testing the ?big six? non-O157 STEC cultures obtained from the Thomas S Whittam STEC Center at Michigan State University and also the STEC O26 used in the ground beef and environmental survival tests. Any other DNA isolates were extracted using the simple boiling method. 2.5 Polymerase chain reaction (PCR) A multiplex PCR assay targeting three different virulence genes determined the virulence profiles for all six O26 isolates. Target virulence genes included the eaeA intimin gene, (stx1) and (stx2). AccuStart? II PCR Supermix (2x) (Quanta BioSciences, Gaithersburg, MD) was used for the multiplex PCR. A 1:10 dilution of the DNA samples with ultrapure water was used as a PCR template. Primers were synthesized by Integrated DNA Technologies (IDT?) 24 (Coralville, Iowa) and those primers were resuspended in a 1x TE buffer solution. The concentration of each primer stock solution was 100?M and these primers were stored at -20?C before being diluted for PCR usage. The working solution for each primer was 10?M. In each 0.1ml PCR tube, the following reagents were mixed together: Table 2.1: Volumes of Reagents and Primers for PCR Reagent/Primer Volume (?l) Concentration PCR Template 5.0 Varies with sample stx1 Forward Primer 0.25 10?M stx1 Reverse Primer 0.25 10?M stx2 Forward Primer 0.25 10?M stx2 Reverse Primer 0.25 10?M eaeA Forward Primer 0.25 10?M eaeA Reverse Primer 0.25 10?M PCR Super Mix 12.5 1x Ultrapure Water 6.0 Total Volume 25 per sample The template was replaced with TE buffer or water as a negative control. The PCR reaction was carried out in Applied Biosystems? Veriti? 96-Well Fast Thermal Cycler (Life Technologies?, Carlsbad, CA). 30 cycles of 30 sec. 94?C denaturation, 30 sec. 58?C annealing, and 1 minute 72?C extension were carried out following the initial 3 minute 94?C denaturation. Samples were then held at 4?C for 30min before being taken out of the thermocycler. After the run, PCR products were checked by running the samples on a 2% agarose gel at 120 volts for 42 minutes. Table 2.2: PCR Primer sequences PCR Primers 5?-3? Bp Tm Ta Reference stx1 Forward ATAAATCGCCATTCGTTGACTAC 180 58.4?C 52.4 (Paton and Paton, 1998) stx1 Reverse AGAACGCCCACTGAGATCATC 58.9?C stx2 Forward GGCACTGTCTGAAACTGCTCC 255 59.6?C 55.0 stx2 Reverse TCGCCAGTTATCTGACATTCTG 58.0?C eaeA Forward GACCCGGCACAAGCATAAGC 384 62.9?C 51.8 eaeA Reverse CCACCTGCAGCAACAAGAGG 61.3?C 25 2.6 Pulse field gel electrophoresis (PFGE) A PFGE was completed for the farm and clinical-isolated strains of O26 to identify any similarities. A well isolated colony was selected and streaked from test cultures onto Trypticase Soy Agar (TSA) plates for confluent growth. These cultures were incubated at 37?C for 14-18 hours. A 1% agarose mixture was made and divided into glass tubes for future use. TE buffer, cell suspension buffer, and cell lysis buffer were made according to the following procedures: -TE buffer was made using10 ml of 1 M Tris (pH 8.0) and 2 ml of 0.5 M EDTA (pH 8.0). This mixture was diluted to 1000 ml with ultrapure water. -Cell Suspension Buffer was made by mixing100 ml of 1 M Tris (pH 8.0) and 200 ml of 0.5 M EDTA (pH 8.0). This was diluted to 1000 ml using ultrapure water. -Cell Lysis Buffer was made using 25 ml of 1 M Tris (pH 8.0), 50 ml of 0.5 M EDTA (pH 8.0), and 50 ml of 10% Sarcosyl (N-lauroylsarcosin, sodium salt). This mixture was diluted to 500 ml with ultrapure water. The next day, an Eppendorf? Thermomixer? R (Fisher Scientific, Pittsburgh, PA) was set to 55?C and a regular water bath was also set to 55?C. One ml of 20% SDS was preheated to 55?C in the water bath along with 3 liters of TE buffer and 2 liters of ultrapure water. The premade agarose was reheated in the small glass tube until completely melted. Five hundred microliters of 20% pre-heated SDS per 9.5 ml agarose was added and mixed well. The final 26 concentration of SDS was 1%. The agarose tube was recapped and returned to the 55?C water bath. Fifteen milliliter centrifuge tubes were labeled with culture numbers and 5 ml of Cell Suspension Buffer was added to the tubes. A sterile loop was used to remove some of the growth from each agar plate. Cells were suspended in the cell suspension buffer by spinning the loop gently to evenly disperse them. The concentration (OD600) of cell suspensions was measured by the use of an Ultrospec 10 Cell Density Meter (GE Healthcare, Piscataway, NJ). The OD value of each sample was adjusted to 1.3-1.4 by the addition of more cells or cell suspension buffer. Along with the other samples, the Salmonella ser. Braenderup H9812 standard was prepared. It was used as a control in this experiment. Plug making: Wells of the disposable plug molds were labeled. Four hundred microliters of adjusted cell suspensions were transferred to labeled 1.5 ml centrifuge tubes. Twenty microliters of Proteinase K (20 mg/ml) was added to each tube and mixed gently by pipetting up and down 5-6 times. Four hundred microliters of the melted agarose/SDS mixture were added. The mixture was quickly but also gently mixed together, avoiding the formation of bubbles, and dispersed into wells of plug mold. Two plugs per strain were made. Plugs were allowed to solidify at 4?C for 5 min. The agarose tubes were kept in a beaker of warm water (55?C) to keep it from cooling too quickly. Lysis of cells in agarose plugs: Two milliliter centrifuge tubes were labeled with culture number and lysis reagents were measured out into the test tubes. One and a half milliliters of cell lysis buffer and 25 ?l Proteinase K (20 mg/ml) was added to each 2 ml tube. After they 27 solidified, plugs were transferred from mold to the prepared tubes (one tube per strain). Tubes were placed in a rack and incubated in the water bath at 55?C for 1 hour with constant and vigorous agitation (150-175 rpm). Washing of agarose plugs after cell lysis: The temperature of the shaking water bath was lowered to 50?C and the tubes were removed. The sterile green wash column was placed under the hood and column caps were placed wide-end-side down. A clean spatula was used to remove plugs from tubes and place them on narrow end of green lid. Caps were labeled with a marker and screwed onto the next cap. This was repeated for each sample. The green wash column was then placed inside a plastic cylinder and 750 ml of sterile water, preheated from 50?C water bath, was poured in. The lid was screwed on and the cylinder was placed securely in the gyratory water bath at 50?C for 15 min (75 rpm). The water was poured off from the cylinder and the last step was repeated once more. The water was again poured off and pre-heated (50?C) sterile TE buffer was added. The cylinder was shaken in the 50?C water bath for 15 min. TE buffer was poured off and the wash step was repeated with the pre-heated TE 3 more times. A total of 2 water washes and 4 TE washes were completed. Using a clean spatula, the plugs were transferred to 1.5 ml tubes with 1 ml TE buffer in them and stored at 4?C for Day 3. Restriction digestion of DNA in agarose plugs with XbaI: On the third day, 1.5 ml micro centrifuge tubes were labeled with culture/sample numbers. Plugs were removed one at a time from their tubes and placed on a large, clean glass slide. Excess buffer was wiped away with a Kimtech Science? Kimwipe? (Kimberly-Clark, Roswell, GA). Two pieces of 2.5mm?3mm bands were cut and transferred to the labeled 1.5 ml centrifuge tubes. A master mix was made in 28 a 2 ml tube using 87?l of water, 10?l H buffer, and 3?l of the enzyme (XbaI 10U/?l). The restriction enzyme and H buffer were kept on ice at all times. One hundred microliters of master mix was added to each tube along with the cut plugs. Tubes were closed and mixed by gentle tapping. Plugs in the enzyme mixture were incubated in the 37?C water bath for 3 hrs with constant agitation. To stop the digestion, the enzyme was removed and replaced with 200 ?l TE buffer. Plugs were dried a little, using a Kimwipe? (Kimberly-Clark, Roswell, GA), before they were added to the TE buffer. Casting the agarose gel and running PFGE: Two liters of 0.5X TBE buffer were added to the PFGE machine, the power switch was turned on, and the pump was set to a level at which the buffer would run at a smooth, consistent rate through the system. At the same time, the temperature of the cooling unit was set to 14?C. The water bath was turned on and set to 55?C. 1.2% SeaKem? Gold Agarose (SKG) (Lonza, Allendale, NJ) in 0.5X TBE was made by weighing out 1.2g of SKG into 500 ml screw-cap flask and adding 120 ml of 0.5X TBE or TE. The mixture was swirled gently to disperse the agarose and microwaved for 60 seconds. The flask was removed, swirled gently, and microwaved in 15 second intervals until the agarose was completely dissolved. After being completely dissolved, the flask was placed in a 55?C water bath. Next, the gel mold was assembled. The comb was placed on top of the mold and the gel plugs were loaded on the bottom of the comb teeth. Each plug was dried before being placed on the comb. The control (H9810) was placed on the first tooth and loaded again for every 5 samples. These were allowed to dry for about 1 minute, making sure there was no visible buffer on or surrounding the plugs. The comb was positioned in the gel mode and the plugs were confirmed to have an even alignment on the bottom of the teeth. 100 ml of warm agarose, cooled 29 to 55?C, was carefully poured into the mold. The gel was allowed to solidify for 30 min and then the comb was removed. The wells were sealed with left over warm agarose and the gel was loaded into the machine. Electrophoresis conditions were programmed into the machine and the run was started. Run time took 19 hours or less depending on the samples. Electrophoresis conditions: The Bio-Rad CHEF Mapper? XA Pulsed Field Electrophoresis System (Hercules, CA) was set up with the following conditions: Temperature 14?C, Angel 120, 30 kb-low MW, 600 kb-high MW, Run time=19.5 hour, Initial switch time= 2.16S, final switch time= 54.17 for E. coli). Gel staining and imaging: After the run completed the next day, the gel was stained with GelRed? (Biotium, Hayward, CA) for 20 minutes. After 20 minutes, the gel was de-stained in 500 ml ultrapure water for 20 min. The gel was then washed once more with fresh 500 ml ulrapure water. During this time, the buffer was drained from the PFGE machine tubing and was washed with ultrapure water several times. Lastly, the gel was placed in the GelDoc-It?2 Imager (UVP, Upland, CA) so that a photo could be taken. 2.7 Transformation of Escherichia coli cells: In preparation for the ground beef and environmental survival experiments, the farm and clinical-isolated E. coli O26 cells needed to be transformed. By doing this, the bacteria inoculated into the sample can more easily be differentiated from the other bacteria that show up on the growth media. They were transformed using a green fluorescent protein (GFP) plasmid from the bioluminescent jellyfish Aequorea victoria. 30 Making competent cells: Bacteria to be transformed were first streak plated onto regular Luria Bertani (LB) plates. These plates were grown overnight at 37?C. A Bio-Rad pGLO? Transformation Kit (Bio-Rad Laboratories, Herculues, CA) was used to help in transforming the cells. A single colony was inoculated into 25ml LB in a 250ml bottle. The bottle was shaken at 37?C for 4-6hrs. Cells were put on ice for 10min and kept cold for the rest of the procedure. The shaken and incubated samples were then centrifuged at 3800rpm for 15min at 4?C. The supernatant was discarded and the pellet was re-suspended in 10ml cold 0.1M CaCl2. The new suspension was next placed in ice for 20min. Samples were centrifuged again at 3800rpm once again for 15min. Supernatant was discarded and the pellet was resuspended in 5ml cold 0.1 CaCl2. To increase the efficacy of transformation four to six- fold, cells were stored at 4?C in CaCl2 solution for 24-48 hrs. This helped in gaining more competent cells. Transformation: Two hundred and fifty micro liters of the competent cells were placed in a small microcentrifuge tube. A sterile loop was dipped into the plasmid DNA solution, withdrawn, and mixed into the cell suspension. If this did not work, 5?l was extracted and mixed into the cell mixture. Tubes were incubated on ice for 10min and then transferred to a 42?C water bath for exactly 50 seconds. Tubes were then placed back on ice for 2 minutes. This exchange was made as rapidly as possible because the cells need to be shocked. Tubes were then placed on the bench top and 250?l of LB nutrient broth was added. Tubes were incubated for 10min at room temperature. One hundred microliters was pipetted onto LB plates supplemented with 0.1mg/ml ampicillin and 50mg/ml arabinose, spread evenly, and incubated at 37?C overnight. 31 2.8 Antimicrobial tests: The antimicrobial susceptibility of the farm and clinical-isolated O26 strains was tested against 13 different antimicrobials using the disk diffusion method. Bacteria to be tested were spread on TSA plates and incubated overnight at 37?C. Mueller Hinton Agar was prepared and antimicrobial disks were kept refrigerated at 8?C or below. Sealed packages containing the discs were removed from refrigeration one to two hours before use so that they could acclimate to room temperature before use. After opening the package, they were placed in an autoclaved container for further storage in the refrigerator or freezer. One well-isolated colony of the same morphological type was selected from each TSA plate and transferred to tubes of tryptic soy broth (TSB). These were incubated overnight at 37?C. The overnight culture was taken and diluted until an OD600 value of 1.0 was obtained. One hundred microliters of the cell suspension was added to the Mueller Hinton plates and spread evenly over the surface. The plates were divided into four parts and the appropriate drug-filled disks were placed carefully on the surface of the inoculated plates. Using sterile forceps, the disks were pressed down gently. Disks were not placed any closer than 24mm from the center of one disk to another. The disks were also kept away from the edge of the plates. The plates were then inverted and incubated at 37?C for 16-18hrs. The diameters of the zones of inhibition were examined and measured in millimeters. The organisms were reported as susceptible, intermediate, or resistant to each antimicrobial agent. Table 2.3: Antimicrobials used Class Antibody Disk concentration Penicillin Amoxicillin-clavulanic acid 20?g 32 2.9 Ground beef 10 day trials Ground Beef trials were done on STEC O26 containing one virulence gene (eaeA) and O26 containing two virulence genes (eaeA and stx1) to find out if the presence of multiple genes influenced survival. The ground beef trials were done for 10 days, samples were held at normal refrigeration temperature, and three different inoculation levels were examined: (102, 106, and 108 log CFU?s/g). Ground beef was purchased from a local store. Samples used in this experiment were 85% lean and 15% fat. Before inoculation, the ground beef was confirmed to be STEC-free by taking a 25g sample and enriching it with 225ml TSB. This was incubated at 37?C overnight. Enriched beef samples were plated on CHROMagar? STEC (CHROMagar, Paris, France) and incubated again overnight at 37?C. Positive, mauve colored colonies were further tested using PCR to determine if they contain the stx or eaeA intimin genes. LB plates supplemented with ampicillin and arabinose were made two days in advance to allow the plates time to thoroughly dry before use. Previously transformed farm and clinical-isolated strains were streaked to isolation on the supplemented LB plates. These were incubated overnight at 37?C. A 0.1% buffered peptone Penicillin Ampicillin 10 ?g Cephalosporin Ceftriaxone 30?g Cephalosporin Cephalothin 30 ?g Phenicol Chloramphenicol 30 ?g Quinolone & Fluoroquinolones Ciprofloxacin 5 ?g Cephalosporin Cefoxitin 30 ?g Cephalosporin Ceftiofur 30 ?g Aminoglycoside Gentamicin 10 ?g Quinolone & Fluoroquinolones Naladixic acid 30 ?g Sulfonamide Sulfamethoxazole ---- Sulfonamide Sulfamethoxazole/Trimethoprim 23.75/1.25 ?g Tetracycline Tetracyline 30 ?g 33 water solution, TSB, and 9ml dilution tubes were made. Measuring tools for weighing out the beef were autoclaved. For each strain, a single, well isolated colony showing fluorescence was chosen and 10ml TSB tubes were inoculated. TSB tubes were incubated overnight at 37?C. This gave approximately 109 CFU?s/ml for inoculation the next day. Bacteria cells were washed by centrifugation and resuspended in autoclaved ultrapure water. In order to reach target inoculations of 102, 106 and 108 log CFU/g, dilutions of 103, 107 and 109 log CFU/g were made by taking 1ml samples from the TSB overnight cultures and diluting them accordingly. A cocktail of bacteria containing the gene eaeA was made. The same was done with bacteria containing both eaeA and stx1 genes. Twenty five gram beef samples were then weighed out into filtered Whirl-pak? bags (Nasco, Fort Atkinson, WI). Bags were labeled according to the strain, temperature, and dilution. Each bag was inoculated with 1ml of the chosen E. coli cocktail. Every day for ten days, 6 bags containing the various inoculation levels and cocktails were removed from the 4?C refrigerator. One hundred mililiters of 0.1% buffered peptone water was added to each bag and put through the Smasher? lab blender (AES, Hamilton, NJ) for 2 minutes. Bacteria were plated by spreading 100?l of the smashed contents on a STEC plate and incubated at 37?C overnight. This plating was done in duplicated for each bag. If required, samples were diluted to reach a countable range (between 25 and 200 colonies) on the plates. If plates showed zero colonies, bags were enriched with 100ml of lactose broth, incubated overnight at 37?C, and re-plated the following day. 2.10 Environmental 30 day trials 34 Environmental trials were conducted for 30 days, and samples were held at room temperature. Only 2 different target inoculation levels were examined in these experiments (106 and 104 log CFU/g) and three different environmental sample types were tested: bovine feces, bedding, and trough water. Bedding in this case was dirt that the cattle had been milling around and laying in. The same bacteria used in the ground beef trials were streaked onto supplemented LB plates. Inoculation cocktails were prepared in the same manner as those in the beef trials, except the target inoculations here were 106 and 104 log CFU/g. About 800ml water, 1800g fecal material, and 1800g bedding were retrieved from E.V. Smith Research Center and brought back to the lab on ice. Twenty five grams of fecal or bedding samples were measured out into Whirl-pak? bags (Nasco, Fort Atkinson, WI). Bags were labeled according to the day, gene(s), and target inoculation. Each bag was inoculated with 1ml of the chosen E. coli cocktail. The farm water was separated so that there were four cups containing 100ml of the farm water. One milliliter of chosen E. coli cocktail was added and containers were labeled. On days 0, 5, 10, 15, 20, 25, and 30, 4 water cups and 8 bags of the various cocktails and inoculation levels were taken out of room temperature storage. One hundred mililiters of 0.1% peptone water was then added to each bag of fecal or bedding material and these were shaken for 2 minutes. Water cups were also shaken for 2 minutes. One hundred microliter samples were taken from these bags and cups and spread onto supplemented LB plates. If necessary, samples were diluted to reach a countable range on the agar plates. If the results from the previous sampling day gave a count below 25, then 1ml of the sample was divided and spread onto 4 LB plates to give a more accurate and lower limit of detection. All plates were incubated at 37?C and all fluorescent colonies were counted and recorded. 35 2.11 Statistical Analysis In examining the prevalence of STEC in the calf trial, before and after weaning, a Chi square test was completed along with a paired Student t-test. Statistics for the ground beef and environmental study included a one-way proc ANOVA which was completed using SAS. Inoculation levels, gene profiles, materials, and days were all examined to find out if there was a statistical difference. If P was less than 0.05, the data were considered different. 36 Chapter 3. Results and Discussion 3.1 Research Center studies Six research centers all around Alabama were sampled during the summer and fall months of 2013. Their locations are marked in (Figure 3.1). The Tennessee Valley research center, along with the Sand Mountain and Upper Coastal Plain research centers are located in north Alabama. The Gulf Coast, Black Belt, and Wiregrass research centers are located in south Alabama. Upper Coastal Plain has more of a northwestern location while Gulf Coast and Black Belt have a southwestern position. Sand Mountain and Wiregrass have more of a northeast and southeastern orientation respectively. About 30 fecal samples were taken from each research center and processed within 24 hours of being received in the lab. 3.1.1 Shedding and super-shedding prevalence Cows and calves were sampled at only two of the research centers: Wiregrass and Sand Mountain. As observed in (Table 3.1), these two research centers have the highest STEC shedding prevalence (100% and 85%). For the other four research centers, only cows were sampled; no calf samples were collected. They exhibited a relatively lower number of positive STEC shedding samples as compared to the first two research centers, but still high: (76.5%, 55.6%, 38.7%, and 68.5% positive for STEC). The cow-only herds contained fewer high and ?super-shedder? cattle as compared to the two research centers that had only calves in their herd at the time. High-shedders were defined by the author as: containing around a 3.0 log CFU/g STEC prevalence in their fecal samples. As for the prevalence of ?super-shedding? cattle, 37 Wiregrass contained only one ?super-shedder,? Sand Mountain had 11 ?super-shedders,? and Black Belt had 1 (Table 3.1). Results would be better interpreted if samples were received from each research center at the same time of the year and from the same herd type, either cows or a herd with cows and calves. The high positive STEC counts at Wiregrass and Sand Mountain may not only be due to calves being in the sampling herd, but could also be due to the samples being obtained during the warmer months of the year. 3.1.2 PCR examination and serotyping results of isolates from high-/super- shedder cattle Three isolated colonies were taken from any high- or ?super-shedder? research center fecal samples, DNA was isolated, and they were tested by PCR for three target virulence genes. Those containing at least one target gene were sent out for serotyping. Isolates from all but the Gulf Coast area were sent for serotyping (Table 3.2). The high shedders from the Gulf Coast center did not exhibit any of the target virulence genes, so serotyping was not warranted. Two STEC isolates from Wiregrass, 5 from Sand Mountain, 1 from Black Belt, 2 from Tennessee Valley, and 3 from Upper Coastal Plain were sent for serotyping. The PCR results for these isolates are shown in Figure 3.2. All of those sent out exhibited the eaeA intimin gene. Seven contained the eaeA and the stx2 genes, while three contained all 3 genes. STEC O26 was isolated from the high shedders at both Black Belt and Upper Coastal Plain while STEC O103 was isolated from the Sand Mountain research center (Table 3.2). 3.2 Calf Trials 38 Fecal samples were directly grabbed from 110 calves at E.V. Smith research center while they were still with the dam at 6 months old and just after weaning at 9.5 months of age. All fecal samples were processed, plated on CHROMagar? STEC (CHROMagar, Paris, France), and incubated. The next day they were enumerated and results were recorded. 3.2.1 Shedding level and prevalence changes during weaning All of the shedding results from the calf trial, before and after weaning, are presented in Table 3.3. Any plates showing over 250 colonies were considered too numerous to count, so the upper limit of detection for these tests was 4.14 log CFU/g. No change was seen in the shedding levels before and after weaning for 26% of the calves. This means that if the calf was a high- shedder before weaning, it was high after. If the calf was a low-shedder before weaning, then it was low after, and so on. Two calves changed from high STEC shedders to non-shedders and one calf changed from being a non-shedder to being a high shedder. Out of the rest, 34 increased in their shedding level and 45 decreased. Overall, the shedding level of 74% of the calves changed. As seen in Table 3.4, the number of calves positive for STEC went from 95.5% to 96.5%, staying about the same before and after weaning. Statistics here included a Chi square test (P > 0.05). Shedding level, on the other hand, changed from an average of 3.00?1.42 before weaning to 2.32?1.62 log CFU/g feces after. By conducting a paired Student t-test, the decrease in shedding levels between the two sampling times was discovered to be different (P < 0.05). 3.2.2 Super-shedders Colonies were picked from the consistently ?super-shedding? samples. These isolates were purified and their DNA was extracted for PCR analysis. A total of 18 isolates, 9 before and 39 9 after weaning, were tested in a multiplex PCR assay targeting the three virulence genes and then sent out for serotyping. 3.2.3 PCR results of super-shedder isolates The 9 consistently high ?super-shedders? in this study are highlighted in Table 3.5, which also outlines all of the high-shedders in the group of calves. Eighteen samples from the 9 calves, before and after weaning, were tested by multiplex PCR to see if they contained any target virulence genes: eaeA, stx1, and/or stx2. PCR results showed that 10 out of 18 were positive for at least one target gene. Also, 11 out of the 18 isolates contained eaeA and 2 out of 18 had the stx1 gene. The number of isolates discovered per serogroup and their virulence profiles are shown in Table 3.6. The PCR image of the amplified target genes for the 18 super-shedder isolates is displayed in Figure 3.4. 3.2.4 Serotyping results The serotyping results for these 18 isolates are displayed in Table 3.7, which shows the O-type of the isolates found before and after weaning. Since these dominant isolates shown in the table are the result of only a few randomly picked colonies out of 250 or more colonies on the plates, statistics could not be completed to confirm that these are the true dominant O-types for these samples. Of the ?big six,? E. coli O26 turned up most frequently, twice before and once after weaning. All three O26 isolates contained the eaeA virulence profile. One isolated before weaning contained the eaeA and stx1 virulence profile. Just from observing the potentially dominant isolates from Table 3.7, none of the calves possessed the same dominant stx-containing O-group before as they did after weaning. 3.3 Michigan State reference strain evaluation 40 Two of each ?big six? non-O157 STEC strains were obtained from the STEC Center at Michigan State University (Table 3.8) and their virulence genes were examined by multiplex PCR. As see in Figure 3.5, four of the six ?big six? did not display a different virulence profile than that which was displayed by the serogroup partner. Both O26 and O111 showed a different number of virulence genes within their individual serogroups. One O26 isolate held all three target genes while the other contained only eaeA and stx1. The same was seen with O111. One isolate had three while the other only held two of the target genes. This shows that even among the individual serogroups on the recall list, there are various pathogenic gene compositions present. 3.4 Comparison of O26 isolates Because O26 was discovered in both the E.V. Smith calf trials and the Alabama research centers, further tests were conducted to examine its survival in ground beef and the farm environment. The two clinical-isolated strains of O26 from Michigan State University were compared to the four farm-isolated O26 strains from the E.V. Smith calf testing. The multiplex PCR comparison of these six O26 can be seen in Figure 3.6. As determined by PCR, three farm O26 strains contained the eaeA intimin gene only. The fourth farm strain, along with one clinical strain (TWO7814), contained the eaeA and stx1 virulence genes. The last clinical isolate (TWO8031) contained all three. 3.4.1 PFGE comparison of O26 isolates All four farm isolates from the calf trials and one clinical isolate (TWO7814) were used in the survival tests. The PFGE of these O26 isolates showed that the farm strains were more closely related to each other than they were to the clinical strain (Figure 3.7-3.8). The farm strain 41 containing stx1 was more closely related to the clinical strain than the other three farm strains were. 3.4.2 Antimicrobial test results Antimicrobial tests were also done on the farm and clinical strains. All isolates had a similar susceptibility to the antimicrobials tested, except for one of the clinical isolates (Table 3.9). It was completely resistant to three antibiotics: Ampicillin, Sulfamethoxazole, and Tetracycline. The clinical isolate had a human source and most likely developed a resistance when the person went through an antibiotic treatment. 3.5 Survival studies The GFP transferred farm-isolated and clinical-isolated cells were added to ground beef at three different inoculation levels and to the environmental samples at two different inoculation levels. For every sampling day, previously inoculated samples were taken and plated on LB agar supplemented with ampicillin and arabinose. Examples of what the fluorescent colonies on the LB supplemented plates look like with a lot of background microflora and with very little are shown in Figure 3.9. The figure also shows what the plates looked like under normal fluorescent light, compared to UV light. As the bedding pictures show, there were oftentimes other colonies on the supplemented LB plates that looked very similar in appearance. By transforming the colonies, the inoculants could easily be distinguished from the background microflora and counted without too much trouble. 3.5.1 Ground beef study Every day for 10 days, the previously inoculated ground beef samples were taken out of the 4?C storage and plated on supplemented LB agar. Any fluorescent colonies on the plates 42 were enumerated the next day. For the 10 day ground beef tests, the counts did not vary much from the initial inoculation level (Figure 3.10). The two higher inoculations: 108 and 106 log CFU/g showed about 0.5 log reduction over time whereas the 102 log CFU/g inoculation showed about 1 log reduction over the 10 day period. From the two higher inoculations, there was no difference seen when comparing the different virulence gene profiles (P > 0.05). Even though there seems to be more variation in the lower inoculation level (102 log CFU/g), the standard deviation for that inoculation shows that the count from the two different gene profiles still did not deviate enough to be considered different from one another. There was, however, a difference in the inoculation levels (P < 0.05). 3.5.2 Environmental study The same inoculation and plating process used in the ground beef study was repeated for the environmental samples, except samples were plated every 5 days for 30 days total. Materials inoculated include: feces, bedding, and water. Cattle feed was not used in this study because preliminary tests showed that mold and other bacteria dominated the plates and made it very difficult to count the GFP transformed colonies. An accurate count would not be possible with this material unless an effective mold suppressing supplement were added to the media. The fecal test results can be observed in Figure 3.11. The graph shows a 1 log CFU/g increase in bacteria count during the first 5 days for all inoculation levels and genes profiles except the eaeA/stx1 profile at target inoculation level 104 log CFU/g. This profile also saw an increase in the first 5 days, just not as great an increase as the others. They then showed variable results over the following 25 days. Looking at the overall trend, the count did not change from the initial inoculation level (P > 0.05). 43 For the bedding test seen in Figure 3.12, the bacteria count declined steadily over the whole 30 day period. In this case, the bacterial count at the end of the 30 days was different from the initial inoculation level (P <0.05). For the farm water test seen in Figure 3.13, there was a steep decline in the bacteria count over the 30 day trial period. Similar to the bedding sample results, the bacteria count on day 30 for these water samples were different from the initial inoculation levels (P > 0.05). Most counts were 0 or close to 0 by day 30, but enriching the samples and plating showed that the STEC O26 inoculants were still present in all samples. Statistics for the ground beef and environmental study included a one-way ANOVA which was completed using SAS. Inoculation levels, gene profiles, materials, and days were all examined to find out if there was a statistical difference. If P was less than 0.05, the data were considered different. Over all the trials, there were differences seen in the types of materials that were inoculated and also in initial inoculation levels (P <0.05). The limit of detection for these trials was 0.6 log CFU/g. Looking at all of the materials, the inoculants survived best in the refrigerated ground beef samples. This is most likely due to the temperature preventing other microbes from growing and influencing the inoculants. The temperature also slows the metabolism of the O26 inoculants, so rapid growth or death was not likely to happen. The ground beef would also be the material with the least amount of background microflora, so it would see the least amount of microbial competition. The fecal samples saw the most variation as they held the nutrients to support the growth and survival of STEC, yet they also held other microflora which could offer some competition. The inoculants did not survive as well in the bedding and water samples as they did in the beef and fecal samples. There were fewer nutrients in the 44 bedding and water. For the bedding samples in particular, a greater amount of background microflora was observed on the plates. The mold and other bacteria seemed to flourish at room temperature. This may have had some influence on the results for this material. 45 Chapter 4. Conclusion 4.1 Research center studies Alabama research centers were sampled to determine the non-O157 STEC prevalence in beef cattle all around the state of Alabama. In each herd with cows only, there was only 1 super- shedder cow. The presence of only one or two super-shedder cattle per herd is supported by research (Aurthur et al, 2013). Only about 2% of the cattle displayed the ?super-shedder? status. Results indicate that the time of year and/or herd type have an impact on the number of positive STEC shedders. The presence of calves in the herd seems to have an impact on the number of high-shedders in the herd, as seen at the Sand Mountain research center, where there were 11 ?super-shedders.? 4.2 Calf trial The purpose of the E.V. Smith calf sampling, before and after weaning, was to observe the impact weaning may have on calf STEC shedding. The calves at E.V. Smith experienced a decrease in their shedding level (P < 0.05), but not in the number of calves positive for STEC (P > 0.05). Because the consistently ?super-shedder? calves did not have the same dominant STEC before weaning as they did after (Table 3.7), it suggests that their gut microflora is easily influenced while they are still young. Of course, only a few colonies were picked randomly from the plated ?super shedder? samples and multiple tests could give varying results on the dominant microflora present in the sample. There are quite possibly multiple dominant STEC in each fecal 46 sample, so it must be said that there are limitations of currently available testing protocols. As seen by the numbers in Table 3.4, weaning does have an impact on shedding level. There was a decline the amount of STEC shed by the calves on average (P < 0.05). So, it would be a good idea to look at weaning as a possible control point for reducing the amount of STEC shedding even further, as it already occurs naturally. 4.3 O26 isolate evaluation As seen in both the clinical O26 isolates obtained from the STEC center at Michigan State University and with the O26 farm isolates from the calf study, STEC can have multiple virulence profiles within the same serogroup. Therefore, the virulence profiles should be more important for determining virulence rather than looking at specific serogroups. 4.4 Survival studies Non-O157 STEC O26 has been discovered to be the dominant ?big six? member in these studies, as it showed up in both the research center study and in the E.V. Smith calf study. Because this STEC also displayed a propensity to have different virulence profiles, as discovered in the multiplex PCR analysis of the different clinical and farm isolates, it was examined further in the ground beef and environmental survival tests. Based on the results from the survival tests, it has the ability to survive for more than 30 days at different temperatures and in different samples. Having the stx1 gene, or the lack thereof, does not seem to influence its survival (P > 0.05). 4.5 CHROMagar? STEC 47 CHROMagar? STEC (CHROMagar, Paris, France) agar was the best media to fit the needs of this study. Research shows that this agar has a 98.9% specificity; only 3 out of 264 non- STEC bacteria strains could produce mauve colonies when plated (Hirovonen et al. 2012). Growth on this media is highly associated with the tellurite resistance genes (TerZABCDEF). Not all of the STEC isolated by the agar contained the stx gene, but the existence of the gene could easily be discovered by running them through a PCR assay amplifying the specific gene sequences. This agar was very effective in obtaining a total STEC count in our fecal samples and also isolating those with the stx genes. 4.6 Overall Conclusions Calves start shedding STEC early in their lives and in greater quantities than their dams. Calves are more likely to be ?super-shedders? than adult cattle. The results of this study show that weaning does influence the calf gut. This could be a good pre-harvest control point for reducing the number of STEC in the calf gut before they are shipped to a feedlot or to slaughter. More research should be done on possible methods for control, especially on the ?super- shedding? cattle. If the count in these ?super-shedding? cattle can be reduced, it will more than likely be effective for the whole herd. O26 was found to be the dominant ?big six? member in both the research center and calf studies, begging the question: does O26 have some better means of surviving in the environment? Survival studies comparing different virulence profiles revealed that the existence of the stx1 gene in STEC does not influence O26 survival in refrigerated ground beef or other environmental samples. More research should look into the stx2 gene and how it influences 48 survival, especially since it is more often associated with HUS, more severe sicknesses, and outbreaks. 49 References Alexander, T.W., L.J. Yanke, E. Topp, M.E. Olson, R.R. Read, D.W. Morck, and T.A. McAllister. 2008. Effect of subtherapeutic administration of antibiotics on the prevalence of antibiotic-resistant Escherichia coli bacteria in feedlot cattle. Appl Environ Microbiol. 74(14). 4405-4416. Alexander, T.W., X. Jin, Q. Li, S. Cook, and T.A. McAllister. 2013. Characterization of tetracycline resistant genes in Escherichia coli isolated from feedlot cattle administered therapeutic or subtherapeutic levels of tetracycline. Can J. Microbiol. 59:287-290. Arthur, Terrance M., R Ahmed, M. Chase-Topping, N. Kalchayan, J. W. Schmidt, and J. L. Bono. 2013. Characterization of Escherichia coli O157:H7 strains isolated from super- shedding cattle. AEM. 79(14): 4294-4303. Baines, Danica, and Stephanie Erb. 2013. Characterization of Shiga toxin?producing Escherichia coli infections in beef feeder calves and the effectiveness of a prebiotic in alleviating Shiga toxin - producing Escherichia coli infections. Irish Vet J. 66:17. Bartlett, John and David Stirling. PCR Protocols. 2003. Beutin, L. 2006. Emerging Enterohaemorrhagic Escherichia coli, Causes and Effects of the Rise of a Human Pathogen. J Vet Med. B53, 299-305. Beutin, L., M. G. Krause, S. Zimmermann, S. Kaulfuss, and K. Gleier. 2004. Characterization of Shiga toxin-producing Escherichia coli strains isolated from human patients in Germany over a three year period. J Clin Microbiol. 42, 1099-1108. Boerlin, Patrick, S. A. McEwen, F. Boerlin-Petzold, 1999. J. B. Wilson, R. P. Johnson and C. L. Gyles. Associations between virulence factors of Shiga toxin-producing Escherichia coli and disease in humans. J Clin Microbiol. 37(3):497. Bolton, D.J., A. Monaghan, B. Byrne, S. Fanning, T. Sweeny, and D.A. McDowell. 2011. Incidence and survival of non-O157 verocytotoxigenic Escherichia coli in soil. J Appl Microbiol. 111: 484-490. 50 Brooks, John T., E.G. Sowers, JG. Wells, K.D. Greene, P.M. Griffin, R.M. Hoekstra, and N.A. Strockbine. 2005. Non-O157:H7 Shiga Toxin-Producing Escherichia coli Infections in the United States, 1983-2002. J Infect Dis. 192:1422-9. Brudzinski, L. M.A. Harrison. 1998. Influence of incubation conditions on survival and acid tolerance response of Escherichia coli O157:H7 and non-O157:H7 isolates exposed to acetic acid. J Food Prot. 61(5):542-6. Cary, Jeffrey W., Linz, John E., Bhatnagar, Deepak. 2000. Microbial Foodborne Diseases: Mechanisms of Pathogenesis and Toxin Synthesis. p.95. Casewell, M., C. Friis, E. Marco, P. McMullen, and I. Phillips. 2003. The European ban on growth-promoting antibiotics and emerging consequences for human and animal health. J. Antimicrob Chemother. 52:159-161. Centers for Disease Control and Prevention (CDC). 2012. National Shiga toxin-producing Escherichia coli (STEC) Surveillance Overview. Atlanta, Georgia: US Department of Health and Human Services, CDC. Centers for Disease Control and Prevention(CDC). 2013. Multistate Outbreak of Shiga toxin- producing Escherichia coli O121 Infections Linked to Farm Rich Brand Frozen Food Products. CDC website accessed March 30, 2013. www.cdc.gov/e.coli/2013/O121-03- 13/index.html. Chase-Topping, Margo E., T. Rosser, L. J. Allison, et. al. 2012. Pathogenic Potential to Humans of Bovine Escherichia coli O26, Scotland. Emerg Infect Dis. Vol. 18(3). 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. Nat Rev Microbiol. 6:904?912. de Verdier, Kerstin. A. Nyman, C. Greko, and B. Bengtsson. 2012. Antimicrobial resistance and virulence factors in Escherichia coli from Swedish dairy calves. Acta Vet Scand. 54(1):2. Edrington, Tom S., B. H. Carter, R. L. Farrow, A. Islas, G. R. Hagevoort, T. H. Friend, T. R. Callaway, R. C. Anderson, and D. J. Nisbet. 2011. Influence of Weaning on Fecal Shedding of Pathogenic Bacteria in Dairy Calves. Foodborne Pathog Dis. 8(3). 51 Elhadidy, M. and M.A. Mohammed. 2012. Shiga toxin-producing Escherichia coli from raw milk cheese in Egypt: prevalence, molecular characterization and survival to stress conditions. Lett Appl Microbiol. 56: 120-127. Enriquez, D., M. J. Hotzel, and R. Ungerfeld. 2011. Minimising the stress of weaning beef calves: a review. Acta Vet Scand. 53(1): 28. Etcheverria, Analia Ines and Nora Lia Padola. 2013. Shiga toxin-producing Escherichia coli Factors involved in virulence and cattle colonization. Virulence 4:5, 366-372. Fratamico, Pina M. C. DebRoy, T. Miyamoto, and Y. Liu. 2009. PCR Detection of Enterohemorrhagic Escherichia coli O145 in Food by Targeting Gened in the E. coli O145 O-Antigen Gene Cluster and the Shiga Toxin 1 and Shiga Toxin 2 Genes. Foodborne Pathog Dis. 6(5). Fremaux, B., C. Prigent-Combaret, M. L. Delignette-Muller, B. Mallen, M. Dothal, A. Gleizal, and C. Vernozy-Rozand. Persistence of Shiga toxin-producing Escherichia coli O26 in various manure-amended soil types. J Appl Microbiol. 104(2008): 296-304. Fremaux, B., C. Prigent-Combaret, M. L. Delignette-Muller, M. Dothal, and C. Vernozy- Rozand. 2007. Persistence of Shiga toxin-producing Escherichia coli O26 in cow slurry. Lett Appl Microbiol. 45:55-61. Fukushima, Hiroshi, K. Hoshina, and M. Gomyoda. 1999. Long-term survival of shiga toxin- producing Escherichia coli O26, O111, and O157 in bovine feces. p. 5177-5181. Gilbreth, S.E., T. C. Potdburg, and P. W. Bodnaruk. 2009. Survival of non-O157 STEC on beef surface following spray treatment with different antimicrobials. 96th annual meeting of IAFP, Grapevine, TX. p.2-31. Hauswaldt, Susanne, Martin Nitschke, Friedhelm Sayk, Werner Solbach, and Johannes K.-M. Knobloch. 2013. Lessons Learned From Outbreaks of Shiga Toxin Producing Escherichia coli. Curr Infect Dis Rep. 15:4?9. Hirvonen, J.J., Siitonen, A, and Kaukoranta, S.S. (2012). Usability and performance of CHROMagar STEC medium in detection of shiga toxin-producing Escherichia coli strains. J Clin Microbiol. 50: 3586-3590. Jordan, David, and Scott A. McEwen. 1998. Effect of duration of fasting and a short-term high- roughage ration on the concentration of Escherichia coli biotype 1 in cattle feces. J Food Prot. 61(5): 531-534. 52 L'Ab?e-Lund, Trine M., Hannah J. J?rgensen, Kristin O'Sullivan, Jon Bohlin, Goro Lig?rd, Per Einar Granum, Toril Lindb?ck. 2012. The Highly Virulent 2006 Norwegian EHEC O103:H25 Outbreak Strain Is Related to the 2011 German O104:H4 Outbreak Strain. PLoS One. 7(3): e31413. Lim, Ji Youn, Jang W. Yoon, and Carolyn J. Hovde. 2010. A brief overview of Escherichia coli O157:H7 and its Plasmid O157. J Microbiol Biotechnol. 20(1): 5?14. Mand, T.D., D. Dopfer, B. Ingham, C. Ane, and C.W. Casper. 2012. Growth and survival parameter estimates and relation to RpoS levels in serotype O157:H7 and non-O157 shiga toxin-producing Escherichia coli. J Appl Microbiol. 114: 242-255. Mathusa, Emily C., Y. Chen, E. Enache, and L. Hontz. 2010. Non-O157 Shiga Toxin Producing Escherichia coli in Foods. J Food Protect, 73(9). Menrath, Andrea, L. H. Wieler, K. Heidemanns, T. Semmler, A. Fruth, and N. Kemper. 2010. Shiga toxin producing Escherichia coli: identification of non-O157:H7-Super-Shedding cows and related risk factors. Gut Pathog. 2(7). Midgley, J. and P. Desmarchelier. 2001. Pre-slaughter handling of cattle and Shiga toxin- producing Escherichia coli (STEC). Lett Appl Microbiol. 32: 307-311. Miko, A., S. Delannoy, P. Fach, N. A. Strockbine, B. A. Lindstedt, P. Mariani-Kurkdjian, J. Reetz, L. Beutin. 2013. Genotypes and virulence characteristics of Shiga toxin-producing Escherichia coli O104 strains from different origins and sources. Int J Med Microbiol. 303:8. Monaghan, Aine , Brian Byrne, Seamus Fanning, Torres Sweeny, David McDowell, and Declan J. Bolton. 2011. Serotypes and Virulence Profiles of Non-O157 Shiga Toxin-Producing Escherichia coli Isolates from Bovine Farms. Appl Environ Biology. p.8662-8668. Mora A, Blanco JE, Blanco M, Alonso MP, Dhabi G, Echeita A, Gonz?lez EA, Bern?rdez MI, Blanco J. 2005. Antimicrobial resistance of Shiga toxin (verotoxin)- producing Escherichia coli O157:H7 and non-O157 strains isolated from humans, cattle, sheep and food in Spain. Res Microbiol. 156(7):793-806. 53 Paton, A.W., Paton JC. 1998. Detection and characterization of Shiga toxigenic Escherichia coli by using multiplex PCR assays for stx1, stx2, eaeA, enterohemorrhagic E. coli hlyA, rfbO111 and rfbO157. J Clin Microbiol. 36:598?602. Price MN, Dehal PS, Arkin AP. 2008. Horizontal gene transfer and the evolution of transcriptional regulation in Escherichia coli. Genome Biol. (9)1. Rostagno, Marcos H. 2009. Can Stress in Farm Animals Increase Food Safety Risk? Foodborne Pathog Dis. 6(7). Rund SA, Rohde H, Sonnenborn U, Oelschlaeger TA. 2013. Antagonistic effects of probiotic Escherichia coli Nissle 1917 on EHEC strains of serotype O104:H4 and O157:H7. Int J Med Microbiol. 303(1):1-8. S?enz, Yolanda, L. Brinas, E. Dominguez, J. Ruiz, M. Zarazaga, J. Vila, and C. Torres. 2004. Mechanisms of resistance in multiple-antibiotic-resistant Escherichia coli strains of human, animal, and food origins. Am Society Microbiol. 48(10):3996-4001. Sahoo, K. C., A. J. Tamhankar, S. Sahoo, P.S. Sahu, S.R. Klintz, and C. S. Lundborg. 2012. Geographical variation in antibiotic-resistant Escherichia coli isolates from stool, cow- dung and drinking water. Res. Public Health. 9: 746-759. Salak-Johnson JL and McGlone JJ. 2007. Making sense of apparently conflicting data: stress and immunity in swine and cattle. J Anim Sci. 85(E Suppl.):E81-E88. Samadpour, M. J. E. Ongerth, J. Liston, N. Tran, N. Nguyen, T.S. Whittam, R.A. Wilson, and P.I. Tarr. 1994. Occurrence of Shiga-like toxin-producing Escherichia coli in retail fresh seafood, beef, lamb, pork, and poultry from grocery stores in Seattle, Washington. Appl Environ Microbiol. 60:1038-1040. Sasaki Y, Usui M, Murakami M, Haruna M, Kojima A, Asai T, Yamada Y. 2012. Antimicrobial resistance in Shiga toxin-producing Escherichia coli O157 and O26 isolates from beef cattle. J Infect Dis. 65(2):117-21. Shaw, D. J., C. Jenkins, M.C. Pearce, et al. 2004. Shedding Patterns of Verocytotoxin-Producing Escherichia coli Strains in a Cohort of Calves and Their Dams on a Scottish Beef Farm. AEM. 70(12): 7456-7465. Soon, J.M. S.A. Chadd, R.N. Baines. 2011. Escherichia coli O157:H7 in beef cattle: on farm contamination and pre-slaughter control methods. Anim Health Res Rev. 12(2): 197-211. 54 Srinivasan V, Nguyen LT, Headrick SI, Murinda SE, Oliver SP. 2007. Antimicrobial resistance patterns of Shiga toxin-producing Escherichia coli O157:H7 and O157:H7- from different origins. Microb Drug Resist. 13(1):44-51. Stevens, Shawn. 2012. FSIS Extends Deadline for Implementation of Non-O157:H7 STEC Rule. Food Safety News. Website accessed August 28, 2013. www.foodsafetycounsel.com. Subbiah, Murugan, D. H. Shah, T.E. Besser, J.L. Ullman, D.R. Call. 2012. Urine from treated cattle drives selection for cephalosporin resistant Escherichia coli in soil. PLOS one. 7(11). Tilden J. Jr, Young W., McNamara A.M., Custer, C., Boesel, B., Lambert-Fair, M.A., Majkowski, J., Vugia, D., Werner, S.B., Hollingsworth, J., and Morris, J.G. Jr. 1996. A new route of transmission for Escherichia coli: infection from dry fermented salami. Am J Public Health. 86(8):1142?5. United States Department of Agriculture. 2011. USDA takes new steps to fight E. coli, protect the food supply. Designation extends zero tolerance policy for E. coli O157:H7 to six additional E. coli serogroups. U.S.Department of Agriculture, Washington DC: http://www.usda.gov/wps/portal/usda/usdahome?contentidonly=true&contentid=2011/09 /0400.xml accessed 7 November 2013. Wang, Quan, X. Ruan, D. Wei, Z. Hu, L. Wu, T. Yu, L. Feng, L. Wang. 2010. Development of a serogroup-specific multiplex PCR assay to detect a set of Escherichia coli serogroups based on the identification of their O-antigen gene clusters. Molecular and Cellular Probes 24: 286-290. Wasilenko, Jamie L., P.M Fratamico, N. Narang, G.E. Tillman, S. Ladely, M. Simmons, and W. C. Cray Jr. 2012. Influences of primer sequences and DNA extraction method on detection of non-O157 shiga toxin-producing Escherichia coli in ground beef by real- time PCR targeting the eae, stx, and serogroup-specific genes. J Food Prot. 11: 1912- 2090. Yue, Wan-Fue, Min Du, Mei-Jun Zhu. 2012. High Temperature in Combination with UV Irradiation Enhances Horizontal Transfer of stx2 Gene from E. coli O157:H7 to Non- Pathogenic E. coli. PLoS One. 7(2). Zhang, W., M. Bielaszewska, A. Liesgang, H. Tschape, H. Schmidt, M. Bitzan, and H. Karch. 2000. Molecular characteristics and epidemiology significance of Shiga toxin-producing Escherichia coli O26 strains. J Clin Microbiol. 38:2134-2140. 55 Tables Table 3.1: Research Center Results Research Center % cattle positive for STEC Super-shedders Type of cattle Month Sampled 1 Wiregrass 100% 1 cow/calves May 2 Sand Mountain 85.70% 11 cow/calves June 3 Gulf Coast 76.50% 0 cows June 4 Black Belt 55.60% 1 cows July 5 Tennesse Valley 38.70% 0 cows October 6 Upper Coastal Plain 64.50% 0 cows November 56 Table 3.2: Research center serotyping results Research Center # Sent for Serotyping Dominant STEC 1 Wiregrass 2 O4 2 Sand Mountain 5 O103/O49 3 Gulf Coast 0 ----- 4 Black Belt 1 O26 5 Tennessee Valley 2 O177/O46 6 Upper Coastal Plain 3 O26/O10/O182 57 Table 3.3: Enumeration of STEC from calf trial Cattle Ear Tag No. 6-Month-Old log CFU/g (Before weaning) 9.5-Month-Old log CFU/g (Right after weaning) 1001 2.089848 (e)>4.14 1002 4.051599 3.790978 1003 (e)>4.14 3.248879 1004 (e)>4.14 2.129438 1005 (e)<1.74 2.518514 1007 negative 2.041393 1008 positive (e)>4.14 1009 (e)>4.14 (e)>4.14 1010 3.183608 3.786055 1011 (e)>3.83 2.915487 1012 (e)>4.14 1.890878 1013 1.985277 3.085461 1014 (e)>3.73 (e)>4.14 1015 positive 2.328408 1016 3.775566 3.202502 1017 (e)>4.14 2.390878 1018 (e)>4.14 4.071647 1019 2.779953 2.761059 1020 (e)>4.14 (e)>4.14 1021 (e)>4.14 2.430468 1022 2.73818 3.946526 1023 3.687244 3.550951 1024 positive 2.566969 1025 (e)>4.14 (e)>4.14 1026 (e)>4.14 3.435034 1027 3.373831 3.226001 1028 negative 2.279953 1029 4.166679 negative 1030 (e)>4.36 (e)>4.14 1032 positive (e)>4.14 1033 3.740145 3.258078 1034 3.228079 2.551987 1035 (e)>4.14 2.313427 1036 3.482868 (e)<1.74 1037 3.322836 3.735199 1038 3.740363 3.573434 1039 3.889636 (e)>4.14 1040 positive 3.003532 58 1041 3.314355 (e)>4.14 1042 positive (e)>4.14 1043 negative 2.518514 1044 (e)<2.04 3.051987 1045 3.394645 2.492938 1046 2.089848 3.284431 1047 3.75293 4.184228 1048 (e)>4.14 3.855715 1049 3.40054 2.041393 1050 3.542205 2.818031 1051 2.719883 2.717484 1052 (e)>4.14 (e)<2.04 1054 2.492938 4.315106 1055 3.61368 (e)>4.14 1056 3.144806 3.786372 1057 positive (e)>4.14 1058 (e)>4.14 3.470059 1059 2.478923 positive 1060 3.460062 3.30777 1062 (e)>4.14 3.142773 1063 2.298656 4.254201 1064 (e)>4.14 3.850744 1065 (e)>4.14 (e)>4.14 1067 4.256986 2.342423 1068 2.834123 3.282631 1069 (e)<2.22 4.133291 1071 4.22582 2.731498 1072 2.089848 3.336925 1073 4.118977 (e)>4.14 1074 negative 4.241672 1076 (e)>4.14 negative 1077 3.711019 3.228079 1078 2.041393 1.740363 1079 1.978923 (e)>4.14 1080 (e)>4.14 (e)<2.34 1081 (e)>4.14 2.217484 1082 3.757557 3.202502 1083 3.931479 3.071269 1084 3.23195 positive 1086 (e)4.14 1.890878 1088 3.89216 (e)>4.14 1089 4.382664 2.191908 1090 (e)4.14 2.957647 59 1091 2.739705 1.890878 1092 1.978923 positive 1094 positive 3.013634 1095 positive negative 1097 (e)4.47 (e)>4.14 1098 (e)4.14 3.257075 1099 positive 2.867999 1100 (e)4.62 2.640033 1101 negative 2.240363 1103 3.812681 (e)>4.14 1104 3.60069 3.539649 1105 (e)>4.14 2.580983 1106 2.037811 2.861882 1107 (e)>4.14 3.964762 1108 4.044529 3.146819 1109 3.451165 3.631169 1111 3.784253 3.370875 1112 (e)>4.14 3.211367 1113 3.653173 2.669029 1114 3.991098 2.328408 1115 (e)<1.74 2.129438 1116 (e)4.31 2.439333 1117 (e)>4.14 3.071742 1118 3.998035 3.010772 1119 (e)>4.14 2.901472 1120 (e)>4.14 4.018514 1121 2.390878 2.662912 1122 2.943633 4.000691 60 Table 3.4: Shedding levels and STEC prevalence in calves at 6 and 9.5 months of age STEC shedding level (log CFU/g feces) STEC prevalence (% Positive) 6-Month Old 3.00? 1.42 a 95.5% 9.5 Month Old 2.32 ? 1.62 b 96.5% 61 Table 3.5: Enumeration of STEC from ?super-shedders? in calf trial Cattle Ear Tag No. 6-Month-Old log CFU/g (Before weaning) 9.5-Month-Old log CFU/g (Right after weaning) 1002 4.05 ? 0.03 3.79 ? 0.07 1003 >4.14 3.25 ? 0.03 1004 >4.14 2.13 ? 0.12 1009 >4.14 >4.14 1012 >4.14 1.89 ? 0.21 1017 >4.14 2.39 ? 0.07 1018 >4.14 4.07 ? 0.02 1020 >4.14 >4.14 1021 >4.14 2.43 ? 0.55 1025 >4.14 >4.14 1026 >4.14 3.44 ? 0.09 1029 4.17 ? 0.02 Negative 1030 >4.36 >4.14 1035 >4.14 2.31 ? 0.38 1048 >4.14 3.86 ? 0.02 1052 >4.14 <2.04 1058 >4.14 3.47 ? 0.07 1062 >4.14 3.18 ? 0.08 1064 >4.14 3.85 ? 0.02 1065 >4.14 >4.14 1067 4.26 ? 0.11 2.34 ? 0.43 1071 4.23 ? 0.04 2.73 ? 0.12 1073 4.12 >4.14 1076 >4.14 Negative 1080 >4.14 <2.34 1081 >4.14 2.22 ? 0.01 1086 4.12 ? 0.02 1.89 ? 0.21 1089 4.38 ? 0.04 2.19 ? 0.21 1090 >4.14 2.96 ? 0.02 1097 4.47 ? 0.09 >4.14 1098 >4.14 3.26 ? 0.06 1100 4.62 ? 0.14 2.64 ? 0.08 1105 >4.14 2.58 ? 0.09 1107 >4.14 3.96 ? 0.36 1108 4.04 ? 0.01 3.15 ? 0.01 1112 >4.14 3.21 ? 0.10 1116 4.31 ? 0.01 2.44 ? 0.01 1117 >4.14 3.07 ? 0.04 1118 4.00 ? 0.09 3.01 ? 0.13 1119 >4.14 2.90 ? 0.02 1120 >4.14 4.02 ? 0.10 62 Table 3.6: Serotyping and PCR results for 9 consistently high ?super-shedder? calves Isolates # of Isolates Before Weaning # of Isolates After Weaning PCR results O26 2 1 eaeA, (eaeA, stx1) O49 2 0 eaeA O120 2 4 --- O152 1 0 --- O172 1 0 eaeA O98 1 1 eaeA, (eaeA stx1) O177 1 1 eaeA O159 0 1 --- O109 0 1 eaeA 63 Table 3.7: Serotyping results for consistently high ?super-shedder? calves Tag # Before Weaning After Weaning 1009 O26 O109 1018 O152 O159 1020 O98/O172 O26 1025 O49 O120 1030 O49 O120 1065 O120 O177 1073 O26 O120 1097 O120 O98 1120 O177 O120 64 Table 3.8: Isolates from the STEC Center at Michigan State University STEC Accession number Isolate Sources (Human) O26 TW07814 97-3250 (Female 2yr) ? USA(Idaho) O26 TW08031 MT#10 USA(Mont.) O45 TW09183 M103-19 (Male 45yr) ? USA(Mich.) O45 TW14003 MI05-14 (Male 12yr) ? USA(Mich.) O103 TW08101 MT#80 USA(Mont.) O103 TW07971 TB154A USA(Wash.) O111 TW07926 3215-99 (Female 18yr) - USA(Tex.) O111 TW014960 0201 9611 USA(Conn.) O121 TW07614 MDCH-4 (Female 51yr) ? USA(Mich.) O121 TW08039 MT#18 USA(Mont.) O145 TW09356 4865/96 Germany O145 TW09153 IH 16 Uruguay 65 Table 3.9: Antimicrobial resistance results Antimicrobials Clinical(mm) Isolated from Farm (mm) TWO8031 TWO7814 22B 41C 38H 31-1 Amox-clavulanic acid 7.00 ?0.00 6.50 ?1.22 8.75 ?0.65 8.13 ?0.85 7.88 ?0.75 8.13 ?0.75 Ampicillin 7.00 ?1.15 0.00 ?0.00 8.25 ?1.44 7.50 ?0.58 7.75 ?1.19 7.88 ?0.63 Ceftriaxone 13.88 ?0.75 14.38 ?0.48 13.88 ?0.75 13.63 ?0.25 12.50 ?0.58 14.63 ?0.48 Cephalothin 6.25 ?0.65 5.63 ?1.03 6.63 ?0.48 6.25 ?0.29 6.88 ?0.25 7.25 ?0.29 Chloramphenicol 9.38 ?0.75 9.50 ?0.41 9.88 ?0.25 9.00 ?0.58 8.50 ?0.58 9.88 ?0.25 Ciprofloxacin 14.50 ?0.41 15.00 ?1.2 13.50 ?0.71 13.63 ?0.95 14.38 ?1.03 15.88 ?1.93 Cefoxitin 9.63 ?0.75 9.75 ?0.29 9.50 ?0.41 9.88 ?0.48 10.25 ?1.44 10.38 ?0.48 Ceftiofur 11.50 ?0.41 11.75 ?0.87 11.00 ?0.00 12.00 ?0.00 11.63 ?0.75 12.00 ?0.00 Gentamicin 8.63 ?0.48 8.25 ?0.29 8.25 ?0.29 8.25 ?0.29 8.00 ?0.00 8.88 ?0.25 Naladixic acid 9.75 ?0.29 11.5 ?0.58 10.38 ?0.25 10.00 ?0.71 9.75 ?0.29 11.00 ?0.00 Sulfamethoxazole 8.13 ?0.25 0.00 ?8.13 6.88 ?0.48 8.25 ?0.29 7.25 ?0.29 6.88 ?0.48 Sulfameth/Trimeth 11.75 ?0.65 8.13 ?0.48 11.63 ?0.25 12.00 ?0.71 11.25 ?0.87 11.13 ?0.48 Tetracylcine 9.00 ?0.00 0.00 ?0.00 8.63 ?0.48 8.50 ?0.41 8.00 ?0.41 8.63 ?0.25 66 Figures Figure 3.1: Alabama research centers tested 1 2 6 5 4 3 (1) Wiregrass (2) Sand Mountain (3) Gulf Coast (4) Black Belt (5) Tennessee Valley (6) Upper Coastal Plain 67 Figure 3.2: PCR results for research center samples sent for serotyping 68 Figure 3.3: Shedding level for calf trial 69 Figure 3.4: PCR of the 9 ?super-shedder? isolates from the calf trial, sent for serotyping 1009 10 18 10 20 10 25 10 30 10 65 10 73 10 97 1120 Tag Number: eaeA stx2 stx1 70 Figure 3.5: PCR of ?big six? STEC clinical isolates from Michigan State University STEC center 71 Figure 3.6: PCR of 4 farm-isolated O26 strains and 2 clinical isolated O26 strains 72 Figure 3.7: PFGE results from 4 farm and 1 clinical O26 strain SEB 31-1 41C TWO8031 22B 38H SEB Farm Strains Clinical strain Farm Strains 73 Figure 3.8: Dendrogram of PFGE fingerprinting profile 41CEar 22B 38H 31-1 TWO8031 Farm Strains Clinical strain 74 Figure 3.9: GFP transferred bacteria on supplemented LB agar UV light Regular light With a lot of Background Microflora With little Background Microflora 75 Figure 3.10: Ground beef trial 0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 9 10 log CFU /g Days Ground Beef eaeA 8 log CFU/g eaeA 6 log CFU/g eaeA 2 log CFU/g eaeA stx1 8 log CFU/g eaeA stx1 6 log CFU/g eaeA stx1 2 log CFU/g * * * Limit of detection (1.6 log CFU/g) *Enriched and re-plated 76 Figure 3.11: Fecal environmental trial 0 1 2 3 4 5 6 7 0 5 10 15 20 25 30 log CFU /g Days Fecal eaeA 10^6 eaeA 10^4 eaeA Stx1 10^6 eaeA Stx1 10^4 Limit of detection (0.6 log CFU/g) *Enriched and re-plated 77 Figure 3.12: Bedding environmental trials 0 1 2 3 4 5 6 7 0 5 10 15 20 25 30 log CFU /g Days Bedding eaeA 6 log CFU/g eaeA 4log CFU/g eaeA stx1 6 log CFU/g eaeA stx1 4 log CFU/g * * * Limit of detection (0.6 log CFU/g) *Enriched and re-plated 78 Figure 3.13: Water environmental trials 0 1 2 3 4 5 6 7 0 5 10 15 20 25 30 log CFU/ g Days Water eaeA 10^6 eaeA 10^4 eaeA Stx1 10^6 eaeA Stx1 10^4 * * * * Limit of detection (0.6 log CFU/g) *Enriched and re-plated