UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Occurrence and potential predictors of verotoxigenic Escherichia coli (VTEC), Salmonella, and Listeria… Falardeau, Justin Desmond 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2017_may_falardeau_justin.pdf [ 2.25MB ]
Metadata
JSON: 24-1.0343449.json
JSON-LD: 24-1.0343449-ld.json
RDF/XML (Pretty): 24-1.0343449-rdf.xml
RDF/JSON: 24-1.0343449-rdf.json
Turtle: 24-1.0343449-turtle.txt
N-Triples: 24-1.0343449-rdf-ntriples.txt
Original Record: 24-1.0343449-source.json
Full Text
24-1.0343449-fulltext.txt
Citation
24-1.0343449.ris

Full Text

  OCCURRENCE AND POTENTIAL PREDICTORS OF  VEROTOXIGENIC ESCHERICHIA COLI (VTEC), SALMONELLA,  AND LISTERIA MONOCYTOGENES  IN SURFACE WATER USED FOR PRODUCE IRRIGATION IN  THE LOWER MAINLAND OF BRITISH COLUMBIA, CANADA    by  JUSTIN DESMOND FALARDEAU  B.Sc., Carleton University, 2014     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE   in   THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Food Science)     THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)    April 2017  © Justin Desmond Falardeau, 2017  ii  Abstract Foodborne pathogens such as verotoxigenic Escherichia coli (VTEC), Salmonella, and Listeria monocytogenes may be present in surface waters, thus having the potential to contaminate fresh produce during crop irrigation. The objectives of this study were to determine the occurrence of VTEC, Salmonella, and L. monocytogenes in surface waters used for produce irrigation in the Lower Mainland of British Columbia, and to investigate potential predictors of their presence.  Water samples (n = 223) were collected from three and four irrigation ditches in both the Serpentine and Sumas watersheds, respectively, between February 2015 and August 2016.  VTEC colonies on water filters were detected using a verotoxin colony immunoblot developed for the detection of all VTEC serotypes, and isolates were confirmed via multiplex PCR for virulence genes vt1 and vt2.  Detection of Salmonella and L. monocytogenes was completed using Health Canada Methods MFHPB-20 and MFHPB-30, respectively.  Generic E. coli (EC) and fecal coliforms (FC) were enumerated using 3M™ Petrifilm™ Count Plates, and by membrane filtration with growth on m-FC agar with 0.01% rosalic acid; this was followed by transfer to nutrient agar containing 4-methylumbelliferyl-β-D-glucuronide.  Meteorological data were collected from Environment Canada records, and agricultural data were collected from the British Columbia Agricultural Land Use Inventories.  The most commonly occurring pathogen was L. monocytogenes (11.2%), followed by VTEC (4.93%), and Salmonella (2.68%).  Pathogen recovery was more common in the Serpentine watershed (p < 0.05), especially during the winter and fall seasons (p < 0.05).  Pathogen occurrence in these locations was correlated with FC (r = 0.431) and EC (r = 0.408), but only by iii  using the membrane filtration method.  Pathogens were still recovered when indicator concentrations were low.  Pathogen occurrence was also correlated with the proximity to upstream livestock (rs = -0.886) and the level of precipitation the day before sample collection (r = 0.203).  In conclusion, VTEC, Salmonella, and L. monocytogenes are present in surface waters used for irrigation in the Lower Mainland of British Columbia. There is potential to predict their presence, but further research is required to confirm factors affecting pathogen occurrence, which entail longer sampling times and increased sampling locations.   iv  Preface Subsection 2.2.3.1.1, Detection and isolation of Listeria monocytogenes, and subsection 2.2.3.2.1, detection and isolation of Salmonella were conducted by Silliker JR Laboratories in Burnaby, British Columbia.  The author, Justin Falardeau, submitted water samples to be tested and collected any isolates recovered from positive samples.  Subsection 2.2.3.1.2, Serotyping and pulsed-field gel electrophoresis of Listeria monocytogenes was conducted by Dr. Franco Pagotto, Ms. Karine Hébert, and Mr. Kevin Tyler at the Listeriosis Reference Service, Health Canada, Ottawa, Ontario.  The author, Justin Falardeau, submitted bacterial isolates to be tested and performed the analysis and interpretation of results.  Subsection 2.2.3.2.2, Serotyping of Salmonella was conducted by Dr. Gitanjali Arya and Dr. Cornelis Poppe at the OIE Reference Centre for Salmonellosis, National Microbiology Laboratory, Public Health Agency of Canada, Guelph, Ontario.  The author, Justin Falardeau, submitted bacterial isolates to be tested and performed the analysis and interpretation of results.  Subsection 2.2.3.3.4, Serotyping of verotoxigenic Escherichia coli was conducted by Dr. Roger Johnson and Ms. Kim Ziebell at the E. coli Reference Laboratory, Laboratory for Foodborne Zoonoses, Public Health Agency of Canada, Guelph, Ontario.  The author, Justin Falardeau, submitted bacterial isolates to be tested and performed the analysis and interpretation of results.  The rest of this research was completed solely by the author, Justin Falardeau, under the guidance of Dr. Siyun Wang. v   The work in this thesis is original and has not been previously published.  vi  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents ......................................................................................................................... vi List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xvi List of Symbols ......................................................................................................................... xviii List of Abbreviations ................................................................................................................. xix Acknowledgements .................................................................................................................... xxi Dedication ................................................................................................................................. xxiii Chapter 1 – Introduction and Literature Review ...................................................................... 1 1.1 Introduction ..................................................................................................................... 1 1.2 Literature review ............................................................................................................. 2 1.2.1 Foodborne pathogens ...................................................................................................... 2 1.2.1.1 Verotoxigenic E. coli .......................................................................................... 2 1.2.1.1.1 Characteristics and phylogeny ...................................................................... 2 1.2.1.1.2 Pathogenicity of VTEC ................................................................................. 3 1.2.1.1.3 Presence of VTEC in the agricultural environment ...................................... 5 1.2.1.1.4 Epidemiology and produce-related foodborne outbreaks of VTEC .............. 7 1.2.1.1.5 Detection and isolation of VTEC from environmental samples ................... 9 1.2.1.2 Salmonella......................................................................................................... 10 vii  1.2.1.2.1 Characteristics and phylogeny .................................................................... 10 1.2.1.2.2 Pathogenicity of Salmonella ........................................................................ 11 1.2.1.2.3 Presence of Salmonella in the agricultural environment ............................. 12 1.2.1.2.4 Epidemiology and produce-related foodborne outbreaks of Salmonella .... 13 1.2.1.2.5 Detection and isolation of Salmonella from environmental samples .......... 16 1.2.1.3 Listeria monocytogenes .................................................................................... 17 1.2.1.3.1 Characteristics and phylogeny .................................................................... 17 1.2.1.3.2 Pathogenicity of L. monocytogenes ............................................................. 18 1.2.1.3.3 Presence of L. monocytogenes in the agricultural environment .................. 19 1.2.1.3.4 Epidemiology and produce-related foodborne outbreaks of L. monocytogenes .............................................................................................................. 20 1.2.1.3.5 Detection and isolation of L. monocytogenes from environmental samples 22 1.2.2    Irrigation water as a source for foodborne pathogens in fresh produce .................... 23 1.2.3 Indicator organisms and water quality assessment ................................................... 25 1.2.4 Environmental factors associated with pathogen occurrence ................................... 28 1.3 Research purpose .......................................................................................................... 29 Chapter 2: Occurrence of Verotoxigenic Escherichia coli, Listeria monocytogenes, and Salmonella in the Irrigation Waters Used in Two Watersheds of the Lower Mainland of British Columbia ......................................................................................................................... 31 2.1 Introduction ................................................................................................................... 31 2.2 Materials and methods .................................................................................................. 32 2.2.1 Site selection ............................................................................................................. 32 2.2.2 Sample collection ...................................................................................................... 32 viii  2.2.3 Pathogen detection .................................................................................................... 33 2.2.3.1 L. monocytogenes .............................................................................................. 33 2.2.3.1.1 Detection and isolation ................................................................................ 33 2.2.3.1.2 Serotyping and pulsed-field gel electrophoresis characterization ............... 35 2.2.3.2 Salmonella......................................................................................................... 35 2.2.3.2.1 Detection and isolation ................................................................................ 35 2.2.3.2.2 Serotyping ................................................................................................... 36 2.2.3.3 Verotoxigenic Escherichia coli detection and isolation ................................... 36 2.2.3.3.1 Verotoxin immunoblot ................................................................................ 36 2.2.3.3.2 Verotoxin enzyme linked immunosorbent assay ........................................ 37 2.2.3.3.3 Polymerase chain reaction for confirmation of Escherichia coli and detection of virulence genes .......................................................................................... 38 2.2.3.3.4 Serotyping ................................................................................................... 41 2.2.4 Statistical analysis ..................................................................................................... 41 2.3 Results ........................................................................................................................... 41 2.3.1 Pathogen occurrence ................................................................................................. 41 2.3.2 Pathogen characteristics ............................................................................................ 46 2.4 Discussion ..................................................................................................................... 50 2.4.1 VTEC occurrence...................................................................................................... 51 2.4.1.1 Environmental occurrence of VTEC ................................................................ 51 2.4.1.2 Spatial and temporal distribution of VTEC ...................................................... 52 2.4.1.3 Serotype and pathotype of recovered VTEC .................................................... 54 2.4.2 Salmonella occurrence .............................................................................................. 57 ix  2.4.2.1 Environmental occurrence of Salmonella ......................................................... 57 2.4.2.2 Spatial and temporal occurrence of Salmonella ................................................ 58 2.4.2.3 Serotype analysis of recovered Salmonella ...................................................... 59 2.4.3 Listeria monocytogenes occurrence .......................................................................... 60 2.4.3.1 Environmental occurrence of Listeria monocytogenes ..................................... 60 2.4.3.2  Spatial and temporal occurrence of Listeria monocytogenes ........................... 60 2.4.3.3 Serotyping and PFGE analysis of L. monocytogenes isolates .......................... 61 2.4.4 Effect of sampling and detection methodologies ...................................................... 63 2.5 Conclusion .................................................................................................................... 65 Chapter 3: Generic Escherichia coli and Total Fecal Coliforms as Indicator Organisms for the Presence of Verotoxigenic Escherichia coli, Salmonella, and Listeria monocytogenes in Irrigation Water .......................................................................................................................... 66 3.1 Introduction ................................................................................................................... 66 3.2 Materials and methods .................................................................................................. 67 3.2.1 Enumeration of generic E. coli and total fecal coliforms ......................................... 67 3.2.1.1 Membrane filtration method ............................................................................. 68 3.2.1.2 Enumeration on 3M™ Petrifilm™ ................................................................... 68 3.2.2 Pathogen occurrence ................................................................................................. 68 3.2.3 Statistical analysis ..................................................................................................... 69 3.3 Results ........................................................................................................................... 69 3.3.1 Comparison of methods and indicator concentrations .............................................. 69 3.3.2 Predictive ability of total fecal coliforms and generic E. coli ................................... 72 3.4 Discussion ..................................................................................................................... 77 x  3.4.1 Comparison of enumeration methods ....................................................................... 78 3.4.2 Correlation and prediction of pathogen occurrence .................................................. 79 3.5 Conclusion .................................................................................................................... 85 Chapter 4 – Environmental Factors Associated with the Occurrence of Verotoxigenic Escherichia coli, Salmonella, and Listeria monocytogenes ....................................................... 86 4.1 Introduction ................................................................................................................... 86 4.2 Materials and methods .................................................................................................. 86 4.2.1 Water temperature, pH, and total dissolved solids ................................................... 86 4.2.2 Collection of weather data ........................................................................................ 87 4.2.3 Collection of geographical data ................................................................................ 87 4.2.4 Analysis of geographical data ................................................................................... 88 4.2.5 Pathogen occurrence ................................................................................................. 89 4.2.6 Statistical analysis ..................................................................................................... 89 4.3 Results ........................................................................................................................... 89 4.3.1 Water temperature, pH, and total dissolved solids ................................................... 89 4.3.2 Correlations with weather factors ............................................................................. 92 4.3.2.1 Precipitation ...................................................................................................... 92 4.3.2.2 Air temperature ................................................................................................. 94 4.3.3 Correlations with proximity and density of upstream livestock ............................... 96 4.4 Discussion ..................................................................................................................... 98 4.4.1 Geographical characteristics ..................................................................................... 98 4.4.2 Weather patterns ..................................................................................................... 103 4.4.3 Water characteristics ............................................................................................... 105 xi  4.5 Conclusion .................................................................................................................. 106 Chapter 5 – Conclusion and Future Directions ..................................................................... 107 5.1 Conclusion .................................................................................................................. 107 5.2 Future directions ......................................................................................................... 109 Bibliography .............................................................................................................................. 111 Appendix .................................................................................................................................... 130 Appendix A – Water and geographical data used for statistical analysis ............................... 130    xii  List of Tables  Table 1.1 - Recent produce-related outbreaks associated with verotoxigenic E. coli, including the number of confirmed cases in Canada, the United States, and Europe; the number of hospitalizations; number of cases of hemolytic uremic syndrome (HUS); and number of  deaths .............................................................................................................................................. 8  Table 1.2 - Recent produce-related outbreaks associated with Salmonella, including the number of confirmed cases in Canada and the United States, the number of hospitalizations, and number of deaths. ....................................................................................................................................... 15  Table 1.3 - Antigens used for serotyping of Listeria monocytogenes (McLaughlin and Rees, 2009) and their respective lineages (Orsi et al., 2011). ................................................................ 18  Table 1.4 - Recent produce-related outbreaks associated with Listeria monocytogenes, including the number of confirmed cases in Canada and the United States, the number of hospitalizations, and number of deaths. ................................................................................................................... 22  Table 2.1 – Sequences of primers used for confirmation of verotoxigenic E. coli isolates ......... 40  Table 2.2 – Occurrence of verotoxigenic E. coli (VTEC), L. monocytogenes, and Salmonella in irrigation water collected from seven sites in the Sumas and Serpentine watersheds .................. 43 xiii  Table 2.3 – Seasonal occurrence of verotoxigenic E. coli (VTEC), L. monocytogenes, Salmonella, or any of the three in water collected from irrigation ditches at three sites each in the Sumas and/or Serpentine watersheds in the Lower Mainland of British Columbia ............................................ 45  Table 2.4 – Serotype and virulence genes of verotoxigenic E. coli isolates collected from irrigation water at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia ....................................................................................................................................... 46  Table 2.5 – Serotype and PFGE patterns of L. monocytogenes isolates collected from irrigation water at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia ........................................................................................................................... 48  Table 2.6 – Serotypes of Salmonella isolates collected from irrigation water at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia ................... 49  Table 3.1 – Point-biserial correlation coefficients (rpb) of total fecal coliforms and generic Escherichia coli with the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and Listeria monocytogenes in water samples collected from three sites each in two watersheds of the Lower Mainland of British Columbia ...................................................................................................... 71  Table 3.2 – Average total fecal coliform and generic E. coli concentrations for the three sites sampled at each watershed during this experiment ....................................................................... 77 xiv  Table 4.1 – Water temperature, pH, and total dissolved solids (TDS) observed at the six primary sampling sites monitored during this study .................................................................................. 90  Table 4.2 – Point-biserial correlations (rpb) of the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes with physicochemical characteristics of the irrigation water samples collected from three sites each from the Sumas and Serpentine watersheds in the Lower Mainland of British Columbia ...................................................................................................... 91  Table 4.3 – Point-biserial correlations (rpb) of the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes in the Sumas and Serpentine watersheds with the levels of precipitation three days prior (P3), two days prior (P2), one day prior (P1), the day of (P0), and the total volume over the three days prior (P1-3) to sample collection ........................................ 93  Table 4.4 – Point-biserial correlations (rpb) of the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes in the Sumas and Serpentine watersheds with average temperatures three days prior (T3), two days prior (T2), one day prior (T1), the day of (T0), and the average temperature over the three days prior (Tavg) to sample collection ............................. 95   Table 4.5 – Spearman rank correlation coefficients (rs) for the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes with proximity and density of upstream livestock ........................................................................................................................................ 97  xv  Table A.1 – Measured data from each collected samples, including the presence of verotoxigenic Escherichia coli (V), Listeria monocytogenes (L), Salmonella (S), or any of the three (A); the temperature, pH, and total dissolved solids (TDS) of the water samples; and the concentration of fecal indicators generic E. coli (EC) and fecal coliforms (FC) measured using 3M™ Petrifilm (PF) and a membrane filtration method (MF) ............................................................................. 130  Table A.2 – Data used to measure proximity and density of upstream livestock ...................... 142     xvi  List of Figures Figure 2.1 – Location of sampling sites in the Serpentine (red border) and Sumas (green border) watersheds of the Lower Mainland of British Columbia .............................................................. 34  Figure 2.2 – Monthly recovery of verotoxigenic E. coli (VTEC), L. monocytogenes (LM), and Salmonella (SAL) between May 2015 and August 2015 from irrigation water collected at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British  Columbia ....................................................................................................................................... 44  Figure 3.1 – Correlation between the membrane filtration and petrifilm methods for enumeration of total fecal coliforms (A) and generic Escherichia coli (B).  The slope of the resulting trendlines (solid line) is compared to the expected slope of 1.0 (dashed line) ............................. 70  Figure 3.2 – Random effect – estimation method regression trees predicting the occurrence of veroxigenic E. coli, Salmonella, and L. monocytogenes relative to the concentrations of the hygiene indicator organisms, fecal coliforms and generic E. coli within two watersheds in the Lower Mainland of British Columbia ........................................................................................... 74  Figure 3.3 – Plot of the concentrations of total fecal coliforms (red) and generic E. coli (blue) observed for each sample site through the membrane filtration method, as well as samples that were positive for verotoxigenic E. coli (V), Salmonella (S), and L. monocytogenes (L).  Only samples eight through thirty-five are shown since the membrane filtration method was not used during the first seven samplings ................................................................................................... 76 xvii  Figure 4.1 – Map of the sampling area within the Sumas watershed.  Pie charts represent the sample sites, with their size being proportional to the total number of pathogens recovered and their divisions representing the individual pathogens.  Coloured polygons represent properties where livestock are present.  Upstream water flow is shown to a distance of 3 km from each sample site ................................................................................................................................... 100    Figure 4.2 – Map of the sampling area within the Serpentine watershed.  Pie charts represent the sample sites, with their size being proportional to the total number of pathogens recovered and their divisions representing the individual pathogens.  Coloured polygons represent properties where livestock are present.  Upstream water flow is shown to a distance of 3 km from each sample site ................................................................................................................................... 101     xviii  List of Symbols Aw – Water activity χ2 – Chi-square r – Pearson correlation rpb – point-biserial correlation rs – spearman’s rank correlation   xix  List of Abbreviations ALUI – Agricultural Land Use Inventory BC – British Columbia BCIG – 5-bromo-6-chloro-3-indolyl-β-D-glucuronide CFU – colony forming units DNA – deoxyribonucleic acid dNTP – deoxynucleotide tri-phosphate EPA – Environmental Protection Agency FC – fecal coliforms FDA – Food and Drug administration FSMA – Food Safety Modernization Act GAP – good agricultural practice HC – haemorrhagic colitis HGMF – hydrophobic grid membrane filter HUS – haemolytic uremic syndrome IMS – immunomagnetic separation LEE – locus of enterocyte attachment LSD – least significant difference MFB – modified Fraser broth MLST – multi-locus sequence typing MPN – most probable number mTSA-VC – modified tryptic soy agar with bile salts, vancomycin, and cefsulodin mTSB-VC - modified tryptic soy broth with bile salts, vancomycin, and cefsulodin xx  MUG – 4-Methylumbelliferyl β-D-Glucuronide NA-MUG – nutrient agar with 4-Methylumbelliferyl β-D-Glucuronide OD – optical density PAI – pathogenicity island PBS – phosphate buffered saline PBS-T – phosphate buffered saline with tween-20 PCR – polymerase chain reaction PFGE – pulsed field gel electrophoresis qPCR – quantitative polymerase chain reaction RE-EM – Random effect – estimation method RNA – ribonucleic acid TC – total coliforms TFC – total fecal coliforms TSA – tryptic soy agar TSB – tryptic soy broth US – United States USDA – United States Department of Agriculture VT – verotoxin VT-ELISA – verotoxin enzyme-linked immunosorbent assay VT-IB – verotoxin immunoblot VTEC – verotoxigenic Escherichia coli    xxi  Acknowledgements First and foremost, I would like to acknowledge Dr. Siyun Wang for her continuing supervision and mentorship.  I would like to thank her for accepting me as her student, and allowing me to pursue my Masters of Science in the Wang Lab of Molecular Food Safety.  Dr. Wang has always encouraged me to pursue any academic interests outside the scope of my project, and supported me, whenever possible, in attending workshops and conferences to improve my scientific foundations.  Secondly, I would like to thank Ms. Elsie Friesen at the British Columbia Ministry of Agriculture for conceiving and supporting this project.  Ms. Friesen was instrumental in helping me become familiar with the Lower Mainland, and in approaching growers to help with this project.  Ms. Friesen was also happy to provide me with any advice I ever needed on advancing my scientific career.  I would also like to thank my other committee members, Dr. Pascal Delaquis, Dr. David Kitts, and Dr. Sean Smukler.  They were always available when I had questions, and consistently provided thoughtful and helpful feedback to improve my experimental design.    Gratitude also goes out to Dr. Roger Johnson for advice on methodologies, and always finding time to proof-read last minute abstracts and poster presentations for meetings and conferences.  His feedback has taught me much about concise writing.  xxii  A heartfelt thank to all of my lab mates and office mates who have made these last two and a half years so much fun.  Specifically, I would like to thank Dr. Jessica Chen for encouraging me to learn out to code, Patricia Hingston for showing me what ultimate dedication to the lab looks like, and Karen Fong for being my lab buddy.  Finally, I would like to extend my sincere thanks to my family and friends who have always been supportive of my ongoing education that seems to never end.  Especially, I would like to thank my fiancé, Vicki Smar, for moving across the country with me, and always being supportive and accepting of the late nights and weekends I have put in over the course of my studies.     xxiii  Dedication I would like to dedicate this work to Victoria Smar; her love is what drives me to accomplish great things.  1  Chapter 1 – Introduction and Literature Review 1.1 Introduction Foodborne illness continues to be an issue for consumers in British Columbia, as well as the rest of Canada. In addition to  health costs, foodborne illnesses can also lead to an economic burden due to lost productivity and wages among those affected (Thomas et al., 2006). Though most foodborne pathogens, such as Listeria monocytogenes, Salmonella, and verotoxigenic Escherichia coli (VTEC), are generally associated with meat, pathogen outbreaks resulting from the consumption of fresh produce have been increasing in prevalence.  Produce accounted for more foodborne illnesses and outbreaks than any other food category in the United States between 2004 and 2013 (Fisher et al., 2015), while in Canada, there were 27 produce related outbreaks reported between 2001 and 2009 (Kozak et al., 2013).  This is of special concern, as Health Canada recommends 7-10 servings of fruits and vegetables per day for adults (Health Canada, 2016).  A likely source of contamination of fresh produce is through irrigation with contaminated water.  The use of poor quality water for irrigation has been correlated with increased incidence of foodborne infections (Steele and Odumeru, 2004), and experiments have specifically shown the effective transmission and internalization of E. coli to lettuce through spraying with contaminated water (Solomon et al., 2002).  Once attached, the bacteria are able to survive for long periods, and may not to be removed through washing (Berger et al., 2010), thus leading to considerable risks associated with any produce consumed raw, such as leafy greens and sprouts.    The objective of this project was therefore to investigate the occurrence and environmental factors that affect the presence of three foodborne pathogens in surface waters used for produce irrigation in the Lower Mainland of British Columbia: VTEC, Salmonella, and L. monocytogenes. 2   1.2 Literature review 1.2.1 Foodborne pathogens 1.2.1.1 Verotoxigenic E. coli 1.2.1.1.1 Characteristics and phylogeny Escherichia coli is a gram-negative, rod-shaped bacterium of the Enterobacteriaceae family.  It is characterized as oxidase-negative, both aerobic and facultative anaerobic, and motile or non-motile depending on the presence of peritrichous flagella.  This bacterium will produce both acid and gas on commonly fermentable carbohydrates (Sheutz and Strockbine, 2005).  Most strains are lactose fermenting, allowing them to be differentiated from Salmonella and Shigella on MacConkey agar, but exceptions do occur (Croxen et al., 2013).  One important exception is the inability of E. coli O157:H7 to ferment sorbitol (March and Ratnam, 1986).  E. coli are typically found in the small and large intestines of mammals, and as such prefer a temperature of 37°C to 42°C, but can grow under a wide range of temperatures (15°C to 48°C), and survive and persist in the natural environment under a range of adverse conditions as well (Welch, 2006).  The most commonly used serotype classification method for E. coli is based on the classification scheme developed by Kauffman, which relies primarily on the somatic “O” and flagellar “H” antigens (Nataro and Kaper, 1998). Currently there are 174 and 53 distinct E. coli O and H antigens, respectively, with non-motile strains having an H designation of NM due to lack of flagella (Croxen et al., 2013).  Other typing methods, especially for outbreak investigations, include pulsed-field gel electrophoresis (PFGE) (Swaminathan et al., 2001) and multilocus 3  sequence typing (MLST) (Jenke et al., 2011), with whole genome sequencing beginning to become more common (Wang et al., 2016).  Verotoxigenic E. coli are loosely defined as any strain of E. coli containing the gene for producing either or both variants of the Shiga-like verotoxin (VT; vt1 or vt2), which is carried on a lambdoid bacteriophage.  Only a subset of these, however, have been implicated in human illness (Croxen et al., 2013).  A further subset of VTEC, and of primary concern is Enterohemorrhagic E. coli (EHEC), which is associated with hemorrhagic colitis (HC); the most well-known and commonly occurring member is E. coli O157:H7 (Nataro and Kaper, 1998).  Other important non-O157 EHEC serotypes known to cause significant levels of human illness, termed the “Big Six”, include the serotypes O26, O45, O103, O111, O121, and O145. (Brooks et al., 2005).  Other VTEC serotypes are known to cause illness sporadically, including a large VTEC outbreak in Germany related to VTEC O104:H4 (Navarro-Garcia, 2014).  1.2.1.1.2 Pathogenicity of VTEC Pathogenicity and virulence is highly varied across VTEC isolates, and is dependent on varying combinations of virulence factors contained on mobile elements such as plasmids, pathogenicity islands (PAI), and prophages (Croxen et al., 2013).  There are many different factors leading to pathogenicity, but a select few have been shown to associate with increased severity of disease (Boerlin et al., 1999).  The most important virulence factor defining VTEC is the production of the VT, which has been associated with the development of hemolytic uremic syndrome (HUS) in infected individuals 4  (Mayer et al., 2012).  There are two types of VT, VT1 and VT2, which are encoded on prophage inserted into the bacterial chromosome (vt1 and vt2, respectively), and each possessing various sub-types.  VTEC can carry one or both types, or a combination of subtypes (Croxen et al., 2013).  The VT1 variant is almost identical to that of Shigella dysenteriae,  with VT2 being genetically and immunologically distinct (Scheutz et al., 2012).  Of note, VTEC containing vt2 are more often associated with severe human illness and HUS; though vt1-containing isolates have also been known to lead to HUS (Boerlin et al., 1999).  VT induces cell death after internalization into the cell where it removes an adenine from the 28S ribosomal RNA subunit, inhibiting protein synthesis (Gyles, 1992).  The pathogenicity of VTEC can be split across two groups based on the presence or absence of the locus of enterocyte effacement (LEE), a PAI responsible for the attachment and effacement lesions associated with Enteropathogenic E. coli (Croxen et al., 2013). The LEE contains the genes for effector proteins such as the type III secretion system, intimin, and the translocated intimin receptor, which are responsible for the creation of the attaching and effacing lesions that allow the bacterium to anchor to epithelial cells (Grant et al., 2011).  Of importance is that the presence of the intimin producing gene of LEE, eaeA, has been associated with increased virulence in VTEC (Boerlin et al., 1999).  Despite the association of LEE-positive VTEC with severe disease, not all HC and HUS causing strains contain this PAI.  Of note is a 2011 outbreak of LEE-negative E. coli O104:H4 in Germany resulting in more than 3,800 cases of gastroenteritis and more than 800 cases of HUS (Frank et al., 2011).  5  A third virulence factor in VTEC is the plasmid bound hemolysin (hlyA).  This pore-forming toxin has shown the ability to lyse sheep erythrocytes and may contribute to HUS development (Croxen et al., 2013).  The association of this virulence factor with disease is somewhat dubious, however, as hlyA has been associated strongly with the presence of eaeA which may be confounding its association with disease in humans (Boerlin et al., 1999).  1.2.1.1.3 Presence of VTEC in the agricultural environment Within the agricultural environment, VTEC may be found in various reservoirs, including animals, soil, water, and even on the produce itself.  The primary reservoir for VTEC is cattle, with a global prevalence that can be as high as 74% and 70.1% in dairy and beef cattle, respectively, but can also vary significantly across herds (Hussein and Bollinger, 2005; Hussein and Sakuma, 2005).  In Canada, VTEC O157 in cattle feces collected at abattoirs was observed to be at a level of 42.6% in Alberta (Van Donkersgoed et al., 1999), compared to 10.2% VTEC in cattle feces collected at Ontario abattoirs (Karama et al., 2008b).  Cattle positive for VTEC may typically shed the pathogen in feces at levels as high as 1 - 4 log CFU/g feces, and may do so for as long as 10 weeks (Widiasih et al., 2004).  Certain cattle on occasion may shed at levels greater than 4 CFU/g feces; these cattle are termed “super shedders” (Persad and Lejeune, 2014).  Other animals that may carry VTEC include swine (Tseng et al., 2014), as well as a variety of other domesticated and feral ruminants (Persad and Lejeune, 2014).  For example, VTEC isolates were recovered from 28.5% of fecal samples collected from swine across the United States (Fratamico et al., 2004), and the occurrence of VTEC in goats in the US was observed to be 14% (Jacob et al., 2013)  6  Surface waters are also a common place to find VTEC in the environment; though occurrence rates may vary depending on location.  The occurrence of VTEC O157 in the surface waters of southern Alberta were observed to range from 1.7% to 2.3% across two surveys (Gannon et al., 2004; Jokinen et al., 2011), and a cross Canada survey found the national occurrence to be 3% for VTEC O157 (Edge et al., 2012).  Of note, the occurrence of VTEC O157 was observed to be higher in waters of agricultural areas compared to waters collect from control regions not affected by agricultural activity (Edge et al., 2012).  In New York State, the occurrence of VTEC was determined to be 2% for surface waters (Strawn et al., 2013a), compared to a study in California which observed occurrences of 8% and 11% for O157 and non-O157 VTEC, respectively (Cooley et al., 2014).  It should be noted, however, that the California study sampled much larger volumes of water, leading to an increased likelihood of finding the pathogen when present in small concentrations.  A more recent study in British Columbia and using a recently developed method for VTEC detection found an occurrence rate of 20% (Nadya et al., 2016).  VTEC may also find its way into the soils and onto the plants of produce farms.  Soils collected from produce farms in California showed a 0.6% occurrence of the pathogen (Cooley et al., 2013) with a produce farm in New York State showing an occurrence of 2% in collected soil samples (Strawn et al., 2013a).  On harvested produce, the Molecular Data Program conducted by the USDA found yearly estimates of VTEC occurrence to be as high as 0.6%, 0.5%, and 0.18% for spinach, cilantro, and lettuce samples, respectively (Feng, 2014).  In the European Union, VTEC was found in 0.18% of produce samples tested (Feng 2014).  7  1.2.1.1.4 Epidemiology and produce-related foodborne outbreaks of VTEC The global incidence of acute illness from VTEC has been estimated at 2.8 million cases annually (Majowicz et al., 2014).  In Canada, the rate was estimated at 7 per 100,000 people in 2000/2001 (Thomas et al., 2006), but has been declining and was 1.8 per 100, 000 people in 2014 (BC Centre for Disease Control, 2017).  Here in British Columbia, the rates of VTEC infection have been consistently higher than the national average, but recently, a 10-year low at 2.4 cases per 100,000 people was observed in 2015 (BC Centre for Disease Control, 2017).  VTEC O157 has been the most common serotype observed in British Columbia, representing 35% and 47% of annual cases in 2014 and 2015, respectively (BC Centre for Disease Control, 2017).  In Canada, VTEC O157 is responsible for 6% of hospitalizations and 8% of deaths resulting from foodborne infection (Public Health Agency of Canada, 2015a).  Non-O157 serotypes also causing disease in British Columbia include O26, O121, O117, O103, and O111 (BC Centre for Disease Control, 2017).  A variety of foodborne outbreaks have been associated with produce; primarily leafy greens and sprouts.  A selection of recent outbreaks is summarized in Table 1.1.  In 2006, an outbreak of VTEC O157:H7 was traced back to spinach, that caused 192 cases, including one case in Canada.  This outbreak resulted in 97 hospitalizations, 29 cases of HUS, and 4 deaths (Kozak et al., 2013; Sharapov et al., 2016). Between 2006 and 2008, VTEC O157:H7 was implicated in multiple outbreaks caused by lettuce, leading to at least 39 confirmed cases in Canada (Kozak et al., 2013; Michigan Department of Health, 2008), followed by another lettuce related outbreak in Canada of VTEC O157:H7 in 2012, resulting in 31 cases with one confirmed case of HUS (Tataryn et al., 2014).  Other produce related outbreaks of VTEC in Canada include 235 cases resulting from Spanish onions contaminated with VTEC O157 (Kozak et al., 2013), and, perhaps the largest 8  outbreak, the 2011 outbreak of VTEC O104:H4 from sprouts, resulting in 4,068 cases with  908 cases of HUS and 50 deaths in Europe, along with 6 cases in the United States and 1 case in Canada (King et al., 2012; World Health Organization, 2011).  Table 1.1 - Recent produce-related outbreaks associated with verotoxigenic E. coli, including the number of confirmed cases in Canada, the United States, and Europe; the number of hospitalizations; number of cases of hemolytic uremic syndrome (HUS); and number of deaths. Year Food Serotype Cases Hosp. HUS Deaths Reference 2006 Spinach O157:H7 1 (Can) 191 (US) 97 29 4 (Kozak et al., 2013; Sharapov et al., 2016) 2006 Lettuce O157:H7 7 (Can) NR NR NR (Kozak et al., 2013) 2008 Lettuce O157:H7 3 (Can) 47 (US) 21 1 0 (Michigan Department of Health, 2008) 2008 Lettuce (presumed) O157:H7 29 (Can) NR NR NR (Kozak et al., 2013) 2008 Spanish Onions O157:H7 235 (Can) NR NR NR (Kozak et al., 2013) 2010 Lettuce O145 26 (US) 12 3 0 (Taylor et al., 2013) 2011 Sprouts O104:H4 1 (Can) 6 (US) 4068 (Eur)  NR 908 50 (King et al., 2012; World Health Organization, 2011) 2011 Lettuce O157:H7 58 (US) 34 3 0 (Slayton et al., 2013) 2012 Salad mix O157:H7 33 (US) 15 2 0 (Marder et al., 2014) 2012 Sprouts O26 29 (US) 7 0 0 (Centers for Disease Control, 2012) 2012 Lettuce O157:H7 31 (Can) 13 1 0 (Tataryn et al., 2014) 2014 Sprouts O121 19 (US) 7 0 0 (Centers for Disease Control, 2014) 2016 Sprouts O157 11 (US) 2 0 0 (Centers for Disease Control, 2016a) NR – not reported   9  1.2.1.1.5 Detection and isolation of VTEC from environmental samples In Canada, a Health Canada approved method only exists for detection and isolation of VTEC O157.  This method involves isolation using immunomagnetic separation (IMS) with beads coated in anti-E. coli O157 antibodies, followed by plating on selective agar media, such as Sorbitol MacConkey Agar, to capitalize on the inability of O157 isolates to ferment sorbitol (Health Canada, 2014).  A similar procedure is standard in the United States, however IMS for the “Big Six” serotypes has also been approved.  Samples are pre-screened for VT and serotype specific gene by the BAX® polymerase chain reaction (PCR) system to determine the appropriate IMS beads to use for isolation (United States Department of Agriculture, 2014).  In both Canada and the United States, confirmation of isolates by VITEK 2 or equivalent is required.  Since IMS is serotype specific, and uncommon serotypes have been known to cause unexpected outbreaks (e.g., VTEC O104:H4), other methods have been used to try and detect or isolate all VTEC serotypes.  Some studies of surface water have used PCR to detect the presence of VT producing genes (Cooley et al., 2014; Strawn et al., 2013a), though this method may not produce isolates, making it difficult to confirm their presence as a result of viable VTEC.  A variety of selective and differential media have also been developed (e.g., CHROMagar, Rainbow agar O157), but these may not detect all VTEC serotypes and tend to result in a relatively high number of false positives (Parsons et al., 2016).  Most recently, Johnson et al., (2014) developed a VT-immunoblot method (VT-IB) targeting all known variants of VT, and allowing for the identification of suspected VTEC colonies on hydrophobic grid membrane filters.  This method has also proven to be quite sensitive, more than doubling the recovery of VTEC O157 compared to the standard IMS method (Johnson et al., 2014).   10   1.2.1.2 Salmonella 1.2.1.2.1 Characteristics and phylogeny The genus Salmonella, a member of the Enterobacteriaceae family, is characterized as gram-negative, facultative anaerobic rods, usually motile due to the presence of peritrichious flagella.  They are non-lactose fermenting, allowing them to be differentiated from E. coli on MacConkey agar, but produce gas from D-glucose.  They also produce hydrogen sulfide on triple-sugar iron agar (Popoff and Le Minor, 2005).  Salmonella are mesophilic, growing across a temperature range of 5 to 46°C and having an optimal range of 35 to 37°C.  They cannot grow at Aw below 0.94 (Li et al., 2013).  The genus contains two distinct species, Salmonella enterica and Salmonella bongeri, with S. enterica being the most commonly associated with human infection (Jones et al., 2008).  S. enterica is further divided into six subspecies (McQuiston et al., 2008), with subspecies I, (subsp. enterica) being the most clinically significant, accounting for 99% of human cases (Shi et al., 2013), and contains >1,500 of the >2,500 known Salmonella serovars (Li et al., 2013).  The other subspecies are generally associated with cold-blooded animals (Brenner et al., 2000).     Identification of Salmonella serotypes has been traditionally done using the White-Kauffmann-Le Minor method, which uses antisera to determine the identity of somatic (O), flagellar (H), and capsular (Vi) antigens on the surface of the bacterium (Li et al., 2013).  Molecular approaches (e.g. PFGE and MSLT) have been attempted with varying degrees of success; though while these 11  methods do not always accurately predict serotype, they do provide better resolution between individual strains/isolates, making them better suited for outbreak investigations (Shi et al., 2013).  1.2.1.2.2 Pathogenicity of Salmonella The pathogenesis of Salmonella differs across serotypes.  Typhoid and paratyphoid fever, caused by S. Typhi and S. Paratyphi, respectively, are more common in developing nations, accounting for less than 400 cases annually in the United States (Adams et al., 2016); primarily resulting from overseas travel (Lynch et al., 2009).  On the other hand, foodborne non-typhoidal serotypes are believed to be responsible for an estimated one million illnesses annually in the United states (Scallan et al., 2011).  Non-typhoidal Salmonella serovars show differences in disease severity.  A survey of non-typhoidal Salmonella infections data collected by FoodNet between 1996 and 2006 found that the mortality rate, compared to the most common serotype in the United States, S. Typhimurium, was significantly higher in S. Dublin and significantly lower in S. Newport.  Moreover, serotypes Heidelberg, Choleraesuis, and Dublin had a much higher rate of invasive disease (13%, 57%, and 64%, respectively), compared to Typhimurium (6%).  Choleraesuis was also more likely to lead to hospitalization (Jones et al., 2008).    Disease typically occurs after the ingestion of >50,000 cells (Coburn et al., 2007), but can be as low as 1,000 cells (Public Health Agency of Canada, 2011).  Symptoms can occur anywhere from 6 to 72 hours after exposure and include abdominal pain and cramping, and diarrhea (Coburn et al., 2007). Salmonella attach to the intestinal epithelial cells and are able to induce phagocytosis through bacterial-mediated endocytosis. This encloses the bacterial cells in large vesicles which internalizes them.  This process is controlled by virulence factors injected using the type III 12  secretion system. Once inside the cellular environment, the bacterium can survive and multiply within the resulting vacuole, eventually lysing the cells and leading to inflammation and swelling of the infected area.  During this time, a reduced uptake in Na+ and an increased secretion of Cl- leads to fluid loss and diarrhea (Ohl and Miller, 2001; Ray and Bhunia, 2008).    In rare cases, invasive infections can also occur, which are more likely to lead to death compared to purely enteric infections (Jones et al., 2008).  1.2.1.2.3 Presence of Salmonella in the agricultural environment The most common animal reservoir for Salmonella is poultry, which are known to be asymptomatic excreters of the pathogen (Park et al., 2008).  Salmonella was detected from 31.2% of turkey carcasses in Quebec (Arsenault et al., 2007)  and in 5.6% of fecal samples collected from poultry in North Carolina (Alali et al., 2010).  Samples collected from various sources on poultry farms also found a 3% occurrence for Salmonella (Rodriguez et al., 2006).  Domestic swine may also be prominent carriers, as 31.5% of samples collected from pig farms in Southern Ontario showed the presence of Salmonella (Farzan et al., 2010), and 10.1% of samples collected from pig farms in the United States (Rodriguez et al., 2006).  Salmonella was found in the feces of dairy cattle at rates as high as 10% in the United States (Callaway et al., 2005), but only 0.08% of fecal samples were positive at an Alberta based abattoir (Van Donkersgoed et al., 1999).  Once present in the agricultural environment, the pathogen tends to persist for long periods through cycling through animal hosts as a result of fecal contamination of water and feces on the farm (Jacobsen and Bech, 2012).  In areas with high prevalence, wild birds may also become carriers, which can carry the pathogen long distances (Andrés et al., 2013).  13  Presence in local animal sources may also lead to contamination of surface waters in the area.  The annual mean occurrence of Salmonella in waters of agriculture areas across Canada was estimated to be 11% (Edge et al., 2012).  In southern Alberta, multiple surveys have shown occurrence in surface waters to range from 6.2% to 10.3% (Gannon et al., 2004; Johnson et al., 2003; Jokinen et al., 2011). South of the border, 9% of collected water samples in New York State were found to be positive for Salmonella (Strawn et al., 2013a), and a recovery as high as 65% was observed in California, although the California samples represented a much larger volume of water tested (Cooley et al., 2014).  Salmonella can also be routinely isolated from soil samples. On livestock farms, recovery ranging between 8.3% to 22.9% was observed (Rodriguez et al., 2006), while on produce farms, occurrence has ranged from 0.7% (Micallef et al., 2012) to 6.1% (Strawn et al., 2013b).  This presence in soil suggests a risk for the pathogen to contaminate produce grown in the area.  A survey of fresh produce in the United States found a 0.13% occurrence of Salmonella on the products sampled, with the highest occurrences being observed for cilantro (0.34%), parsley (0.29%), bagged spinach (0.29%), hot peppers (0.26%), and sprouts (0.25%).  A total of 7% of products contaminated with Salmonella had been imported from Canada (Reddy et al., 2016).  1.2.1.2.4 Epidemiology and produce-related foodborne outbreaks of Salmonella The global incidence of non-typhoidal Salmonella infection has been estimated at 93.8 million cases annually (Majowicz et al., 2010)  In Canada, the rate was 19 illnesses per 100,000 people in 2000/2001 (Thomas et al., 2006), but this has been on the rise and increased to 21.5 and 25.2 cases per 100,000 people in Canada and British Columbia, respectively (BC Centre for Disease Control, 14  2017).  Salmonella is attributed to 5% of illnesses, 24% of hospitalizations, and 16% of the deaths resulting from foodborne illness (Public Health Agency of Canada, 2016a).  The most commonly observed serovars associated with disease in Canada have consistently been Enteritidis, Heidelberg, and Typhimurium (Public Health Agency of Canada, 2015a), with Enteritidis accounting for approximately 50% of infections observed in British Columbia in 2014 and 2015 (BC Centre for Disease Control, 2017).  A variety of produce related outbreaks have been associated with Salmonella.  These have been summarized in Table 1.2.  The most commonly observed vehicles are sprouts, cucumbers, and cantaloupe.  Multiple Salmonella outbreaks from sprouts were responsible for over 360 illnesses in Canada between 1995 and 2001 (Honish and Nguyen, 2001; Sewell and Farber, 2001; Van Beneden et al., 1999).  A major outbreak of Salmonella St. Paul from peppers resulted in 1500 cases across the United States and Canada, and resulting in 2 deaths (Barton Behravesh et al., 2011).  Salmonella Brandenburg was responsible for a outbreak from cucumbers that affected 12 Canadians in 2004 (British Columbia Centre for Disease Control, 2005), and most recently in the United states, cucumbers were tied to outbreaks of Salmonella Newport and Salmonella Poona causing 275 and 907 cases respectively, which resulted in a total of 7 deaths (Angelo et al., 2015; Centers for Disease Control, 2016b).  Outbreaks associated with Salmonella contaminated cantaloupes lead to 78, 22, and 9 illnesses in Canada in 1991, 1998, and 2008, respectively (Centers for Disease Control, 2008; Deeks et al., 1998; Sewell and Farber, 2001).    15  Table 1.2 - Recent produce-related outbreaks associated with Salmonella, including the number of confirmed cases in Canada and the United States, the number of hospitalizations, and number of deaths. Year Food Serotype Cases Hosp. Deaths Reference 1991 Cantaloupe Poona 78 (Can.) >300 (US) NR NR (Centers for Disease Control, 1991; Sewell and Farber, 2001) 1995 Alfalfa sprouts Stanley 30 (Can.)  242 (US, Fin.) NR NR (Mahon et al., 1997; Sewell and Farber, 2001) 1995-1996 Alfalfa sprouts Newport 121 (Can.),  >20,000 (US, Den.) NR NR (Sewell and Farber, 2001; Van Beneden et al., 1999) 1997 Alfalfa sprouts Meleagridis 124 (Can.) NR NR (Sewell and Farber, 2001) 1998 Cantaloupe Oranienburg 22 (Can.) NR NR (Deeks et al., 1998) 2001 Mung bean sprouts Enteritidis 84 (Can.) 6 0 (Honish and Nguyen, 2001) 2004 Cucumbers Brandenburg 12 (Can.) NR NR (British Columbia Centre for Disease Control, 2005) 2005-2006 Tomatoes Multiple 459 (US) 58 0 (Centers for Disease Control, 2007) 2008 Cantaloupe Litchfield 9 (Can.) 51 (US) 16 0 (Centers for Disease Control, 2008) 2008 Peppers St. Paul 1500 (US and Can.) >286 2 (Barton Behravesh et al., 2011) 2009 Onion sprouts Cubana 20 (Can.) NR NR (Kozak et al., 2013) 2009-2011 Alfalfa sprouts Multiple 509 (US) 44 0 (Centers for Disease Control, 2016c) 2011 Papayas Agona 106 (US) 10 0 (Centers for Disease Control, 2011) 2011-2012 Cantaloupe Multiple 383 (US) 118 4 (Centers for Disease Control, 2016c) 2014 Cucumbers Newport 275 (US) 48 1 (Angelo et al., 2015) 2015 Cucumbers Poona 907 (US) 204 6 (Centers for Disease Control, 2016b) NR – not reported  16   1.2.1.2.5 Detection and isolation of Salmonella from environmental samples Both in Canada and the United States, the detection and isolation of Salmonella from environmental samples follows the same general procedure: Non-selective enrichment, selective enrichment, selective plating, and biochemical screening.  The standard Health Canada approved method (MFHPB-20; Reid, 2009) mandates non-selective enrichment in either nutrient broth or buffered peptone water, followed by selective enrichment in Tetrathionate Brilliant Green broth and Rappaport-Vassiliadis Soya Peptone broth.  This is similar to methods outlined by the USDA (MLG 4.09; U.S. Department of Agriculture, 2017) and the US FDA (Andrews et al., 2016), though the US EPA (Method 1200; U.S. Environmental Protection Agency, 2012) mandates enrichment in Tryptic Soy broth followed by selective plating directly onto modified semisolid Rappaport-Vassiliadis media to check for mobility.  Selective plating of the selective enrichments on at least two media is recommended, with Health Canada recommending Bismuth Sulfite agar and Brilliant Green Sulfa agar (Reid, 2009).  Other selective media, such as Xylose Lysine Deoxycholate  agar are approved by American agencies as well (Andrews et al., 2016; U.S. Environmental Protection Agency, 2012).  Finally, in all cases, suspected colonies are subjected to biochemical screening for carbohydrate utilization, H2S production, and gas formation from dextrose utilization on Triple Sugar Iron agar; presence and absence of lysine decarboxylase and lysine deaminase, respectively, on Lysine Iron agar; and absence of urease on Christensen’s Urea media (Andrews et al., 2016; Reid, 2009; U.S. Department of Agriculture, 2017a; U.S. Environmental Protection Agency, 2012).    17  1.2.1.3 Listeria monocytogenes 1.2.1.3.1 Characteristics and phylogeny Listeria spp. are non-spore-forming, gram-positive, aerobic/facultative anaerobic rods.  They are able to grow at temperatures ranging from < 0°C to 45°C, and a pH range between 6 and 9.  They are also motile due to the presence of peritrichious flagella at temperatures below 30°C; however, no motility is observed at 37°C (McLaughlin and Rees, 2009).  The species L. monocytogenes, in particular, is differentiated from the other species by its hemolytic ability and its production of acid from L-rhamnose and alpha-methyl-D-mannoside, but not D-xylose (Bille et al., 1992).  The species L. monocytogenes is separated into 13 distinct serotypes based on somatic (O) and flagellar (H) antigens; these are summarized in Table 1.3 (McLaughlin and Rees, 2009).  Of these serotypes, 4b has been primarily responsible for outbreaks, while 1/2a and 1/2b are responsible for most of the sporadic illnesses observed (Wiedmann, 2002); though the largest outbreak to date was caused by strains of serotypes 1/2a and 1/2b and associated with cantaloupe (McCollum et al., 2013).  Initial genotypic evaluation of the species has shown the existence of three distinct groups, or lineages: lineage I consisting of serotypes 1/2b, 3b, 4b, 4d, and 4e; lineage II consisting of serotypes 1/2a, 1/2c, 3a, and 3c; and lineage III consisting of serotypes 4a and 4c (Liu, 2006), with some atypical 4b strains also included in lineage III (Orsi et al., 2011).  Recently however, a lineage III subgroup (IIIB) has been declared to be its own distinct lineage IV (Ward et al., 2008).  Lineages I and II are considered to be the most common, with lineage I being highly clonal and highly adapted to host survival, while lineage II has shown evidence of greater horizontal gene transfer, leading to a more diverse, generalist group able to survive in diverse environments (Nightingale et al., 2005).  Moreover, human isolates are predominantly from lineage I, whereas animal isolates 18  are associated more with lineage II, although all lineages are capable of causing human disease (Haase et al., 2014)  Table 1.3 - Antigens used for serotyping of Listeria monocytogenes (McLaughlin and Rees, 2009) and their respective lineages (Orsi et al., 2011). Serovar Somatic (O) antigens Flagella (H) antigens Lineage 1/2a I, II, III A, B II 1/2b I, II, III A, B, C I 1/2c I, II, III B, D II 3a II, III, IV, (XII), (XIII) A, B II 3b II, III, IV, (XII), (XIII) A, B, C I 3c II, III, IV, (XII), (XIII) B, D I 4a III, (V), VII, IX A, B, C III, IV 4ab III, V, VI, VII, IX, X A, B, C  4b III, V, VI A, B, C I, III, IV 4c III, V, VII A, B, C III, IV 4d III, (V), VI, VIII A, B, C  4e III, V, VI, (VIII), X A, B, C  7 III, XII, XIII A, B, C    1.2.1.3.2 Pathogenicity of L. monocytogenes L. monocytogenes is unique among Listeria species in that it is pathogenic to humans. However, L. ivanovii has been shown to be an opportunistic pathogen in humans in a few rare cases (Guillet et al., 2010).  In healthy individuals, infection is rare but may lead to febrile gastroenteritis, which includes symptoms similar to mild influenza and diarrhea.  The mildness of these symptoms suggest that the number of infections may be drastically under-reported (McLauchlin et al., 2004).  In immunocompromised populations, pregnant women, and neonates, invasive systemic disease may occur, potentially leading to meningitis, septicemia, and spontaneous abortions.  Of particular concern is the mortality rate of these infections, which has been estimated to be between 20% and 40% (Public Health Agency of Canada, 2012), and survivors of severe infections will likely be left with life-long sequelae (Farber and Peterkin, 1991).  Pathogenicity of L. monocytogenes results 19  from the presence of several virulence factors present on a PAI.  The presence of these virulence factors allows the bacterium to invade epithelial cells of the intestine where it is able to translocate across cells through inducing of actin polymerization within the cytoplasm.  The cells can also survive inside macrophage, allowing them to travel through the bloodstream to various organs, across the blood brain barrier, or through the placental barrier (Ray and Bhunia, 2008).  The infectious dose is suggested to be high, ranging between 7 and 9 log CFU (Public Health Agency of Canada, 2012).  The incubation period for listeriosis can range anywhere from 3 to 70 days, with an average of 1 to 3 weeks (McCollum et al., 2013), often making it difficult to trace the food source.  Furthermore, novel food vehicles are continually being implicated in recent outbreaks.  1.2.1.3.3 Presence of L. monocytogenes in the agricultural environment Cattle are considered to the be the primary animal reservoir for L. monocytogenes, but the bacterium can also be shed by chickens, horses, and wild animals and birds (Milillo et al., 2012).  L. monocytogenes has been found in the feces of cattle farms at rates as high as 32% in New York State (Nightingale et al., 2004), 16.1% in Finland (Husu, 1990), and 9.3% in Spain (Vilar et al., 2007).  Also in Spain, the pathogen was recovered from 26.5% of feces samples collected on poultry farms (Esteban et al., 2008)  Of concern from an agricultural perspective is the ubiquity of L. monocytogenes in the environment.  The bacterium can routinely be isolated from soil, at rates ranging from 0.4% to 17% (Dowe et al., 1997; Locatelli et al., 2013a; MacGowan et al., 1994; Sauders et al., 2012; Strawn et al., 2013a), but also from surface waters. L. monocytogenes was isolated from all samples taken from seven rivers in the United Kingdom at levels ranging from 3 to >180 per litre of water 20  (Watkins and Sleath, 1981), and in 4% of surface water samples collected in Northern Greece (Arvanitidou et al., 1997).  Closer to home, L. monocytogenes was isolated from 10% of samples collected from surface waters within a Southern Ontario watershed (Lyautey et al., 2007b), and in 33% of surface water samples collected from creeks and ponds near farms in New York State (Strawn et al., 2013a).    Irrigation events have been associated with increased prevalence of L. monocytogenes in agricultural soil samples (Weller et al., 2015), suggesting transfer of the bacterium from irrigation ditches by the irrigation sprinklers.  This is of concern since L. innocua, an avirulent surrogate often used in L. monocytogenes studies (Milillo et al., 2012), has shown the bacterium to readily attach to lettuce leaves through irrigation with infected water (Oliveira et al., 2011).  A review of studies investigating the occurrence of L. monocytogenes on raw vegetables found a global average of 11.4%, including a 2.2% occurrence of L. monocytogenes on cabbage in Canada, a 2.2% occurrence on cucumbers and 36.8% occurrence on radishes in the United States, and of greater concern, a rate of occurrence of 85.7% on bean sprouts in Malaysia (Beuchat, 1996).    1.2.1.3.4 Epidemiology and produce-related foodborne outbreaks of L. monocytogenes L. monocytogenes is resistant to many of the control strategies used by food processors to minimize risks of foodborne pathogens.  Unlike other common foodborne pathogens, L. monocytogenes is capable of growing at refrigerated temperatures (Walker et al., 1990), with computer models suggesting the possibility of growth at temperatures as low as -1.5°C (Tienungoon et al., 2000).  Furthermore, the bacterium is resistant to freezing conditions in most foods (Palumbo and Williams, 1991).  L. monocytogenes is also able to survive in acidic conditions at a pH as low as 21  four (Phan-Thanh and Montagne, 1998), Aw as low as 0.90 (Nolan et al., 1992), and in salt concentrations as high as 16% (Hudson, 1992).  Global estimates of listeriosis in 2010 were 23,150 cases, resulting in 5,463 deaths (de Noordhout et al., 2014).  In Canada, listeriosis rates have remained steady between 2010 and 2013 at approximately 0.33 cases per 100,000 individuals, with similar statistics observed in British Columbia (BC Centre for Disease Control, 2017).  Despite these low incidences compared to other foodborne pathogens, L. monocytogenes is responsible for 35% of foodborne illness related deaths in Canada (Public Health Agency of Canada, 2016a).  The first confirmed outbreak of foodborne listeriosis was in Canada and was due to coleslaw produced using contaminated cabbage (serotype 4b), leading to 41 cases and 18 deaths (Schlech et al., 1983).  Since then, most outbreaks have been tied to ready-to-eat meats and cheeses (Cartwright et al., 2013), but recent outbreaks have highlighted concerns about fresh produce as a vehicle for L. monocytogenes transmission, as summarized in Table 1.4.  Since 1998 in the US, sprouts have been implicated in three individual outbreaks, leading to 27 cases of illness and two deaths (Centers for Disease Control, 2016c).   In 2010, celery was implicated in an outbreak in Texas hospitals leading to 10 cases of listeriosis that resulted in 5 deaths (Gaul et al., 2013).  In 2011, the largest outbreak of listeriosis ever recorded resulted from contaminated cantaloupes from a single farm in Colorado.  The outbreak led to 147 cases in 28 states, resulting in 33 deaths.  This outbreak was also unique in that two distinct serotypes, 1/2a and 1/2b, were involved (McCollum et al., 2013).  Continuing on the trend of novel vehicles, a major outbreak in 2014 resulting in 35 illnesses across 12 states, and 7 deaths was attributed to caramel apples (Centers for Disease 22  Control, 2015).  In addition, one case associated with this outbreak was also confirmed in Canada (Public Health Agency of Canada, 2015b).  Finally, most recently, an outbreak involving prepackaged salads led to 19 illnesses and one death in the US (Centers for Disease Control, 2016d), and 14 illnesses and three deaths in Canada (Public Health Agency of Canada, 2016b).  Table 1.4 - Recent produce-related outbreaks associated with Listeria monocytogenes, including the number of confirmed cases in Canada and the United States, the number of hospitalizations, and number of deaths. Year Food Serotype Cases Hosp. Deaths Reference 2008 Sprouts 1/2a 20 (US) 16 0 (Centers for Disease Control, 2016c) 2010 Celery 1/2a 10 (US) 10 5 (Gaul et al., 2013) 2011 Cantaloupe 1/2a, 1/2b 147 (US) 143 33 (McCollum et al., 2013) 2014 Sprouts 4b 5 (US) 5 2 (Centers for Disease Control, 2016c) 2014 Caramel Apples 4b 1 (Can) 35 (US) 34 7 (Angelo et al., 2017) 2015-2016 Packaged Salads 4b 14 (Can) 19 (US) 33 4 (Centers for Disease Control, 2016d; Public Health Agency of Canada, 2016b)   1.2.1.3.5 Detection and isolation of L. monocytogenes from environmental samples Approved methods for the detection and isolation of L. monocytogenes from environmental samples are similar across Health Canada (MFHPB-30; Pagotto et al., 2011), the USDA (MLG 8.10; U.S. Department of Agriculture, 2017b), and the US FDA (Hitchins et al., 2016).  Samples are first enriched in Listeria enrichment broth, followed by further selective enrichment in modified Fraser broth (Pagotto et al., 2011) or the addition of a Listeria selective enrichment supplement (Hitchins et al., 2016; U.S. Department of Agriculture, 2017b).  Enrichments are then plated on selective Oxford agar and a second selective agar medium, such as PALCAM or 23  modified Oxford to confirm the ability to hydrolyze esculin (Hitchins et al., 2016; Pagotto et al., 2011; U.S. Department of Agriculture, 2017b). Suspect colonies are then biochemically screened for hemolysis, motility, and carbohydrate utilization on mannitol, L-rhamnose, and D-xylose agars.  Acid production should be observed from L-rhamnose, but not the other two carbohydrates (Pagotto et al., 2011).  Confirmation through the use of VITEK 2, or equivalent is also recommended (Pagotto et al., 2011; U.S. Department of Agriculture, 2017b).  1.2.2    Irrigation water as a source for foodborne pathogens in fresh produce Contaminated irrigation water is believed to be a source of produce related foodborne illnesses, as a higher incidence of foodborne illness exists in areas that use minimally treated wastewater for irrigation purposes (Steele and Odumeru, 2004).  Furthermore, in two large outbreaks, the outbreak strain was isolated from the irrigation source water: A 2002 outbreak of S. Newport linked to tomatoes in Virginia (Greene et al., 2008), and a 2008 outbreak associated with peppers from Mexico caused by S. Saintpaul (Barton Behravesh et al., 2011).  Spray and flood irrigation methods are considered to induce the highest risk since in both these methods the water comes into direct contact with the edible parts of the plant (Olaimat and Holley, 2012)  Foodborne pathogens have shown the ability to attach to the leaves of the produce.  This has been shown for VTEC O157 on lettuce leaves (Seo and Frank, 1999) and for S. Senftenberg on various leafy greens (Berger et al., 2009).  Once present, these pathogens are able to persist for significant lengths of time.  When applied at high concentrations, VTEC O157 was still detectable on lettuce leaves for 5 weeks (Oliveira et al., 2012) and 177 days on parsley (Islam et al., 2004a).  VTEC O157 was also detectable on onions and carrots 74 and 168 days post 24  irrigation with 5 log CFU/ml of the pathogen (Islam et al., 2005).  At lower concentrations, (~1 log CFU/ml), VTEC O157 was still detectable on lettuce up to 30 days post-watering (Mootian et al., 2009); however, no viable cells of VTEC O157 were detectable on spinach after 24 hours when watered at concentrations below 126 CFU/100 ml (Ingram et al., 2011).  Salmonella have also been shown to persist, surviving 203 days on carrots grown in soil treated initially with water containing 5 log CFU / ml of the pathogen.  In this same experiment, Salmonella persisted 84 days on radishes (Islam et al., 2004b).  L. innocua, a surrogate for L. monocytogenes, was shown to survive on lettuce for up to 9 weeks after being exposed to sprinkler irrigation with water contaminated with the bacterium at 7 log CFU/ml (Oliveira et al., 2011).  A potential method for survival is the internalization of the bacterial cells into the plant tissue.  VTEC have been observed to internalize into the stomata of the plant leaf, and can be found in the intercellular space, as well as the vascular tissue (Saldaña et al., 2011).  It has also been suggested that VTEC O157 is able to internalize into plants from contaminated soil through the root system, where it can travel to edible parts of the plant (Solomon et al., 2002), but other research has suggested that this is a rare phenomenon (Erickson et al., 2010).  The presence of these pathogens on greens, even at low concentrations, is a concern. Pre-packaged salads, when subjected to temperature abuses not uncommon in retail and transport, can sustain the growth of these pathogens to harmful levels (Bovo et al., 2015; Zeng et al., 2014)   25  1.2.3 Indicator organisms and water quality assessment The testing of water for the presence of foodborne pathogens is too expensive and time consuming for routine analysis, so water quality assessment and regulation has been based on the concentrations of microbial indicators.  These are bacteria that are non-pathogenic, but are accepted to correlate with the presence of foodborne pathogens (Pachepsky et al., 2011).  Initial water quality standards were based on total coliform bacteria as an indicator.  Coliforms can be loosely defined as members of the Enterobacteriaceae that ferment lactose to form acid and gas. Although commonly associated with feces, some are capable of growth in the environment and are therefore not completely predictive of fecal contamination (Tallon et al., 2005).   More recently, total coliforms (TC) have been replaced with thermo-tolerant fecal coliforms (FC) which are a subset of coliforms able to fermenting lactose at 44.5-45.5°C, and more commonly associated with fecal contamination (Pachepsky et al., 2011).  The majority of FC are strains of E. coli, which is now considered to be the best indicator of fecal contamination since other genera of FC can also be found in the environment (Steele and Odumeru, 2004).  Other indicator organisms have been suggested, such as Bacteriodes spp., but little data currently exists to support their value as indicators of fecal pollution (Uyttendaele et al., 2015).  Indeed, a study in the Salmon River of British Columbia found FC to give stronger correlations than Bacteriodes spp. with respect to the occurrence of Salmonella, Campylobacter and VTEC O157 (Jokinen et al., 2010).   Water quality guidelines in Canada recommend fewer than 1000 TC and 100 FC per 100 ml of water for surface waters used in crop irrigation (CanAgPlus, 2015; Steele and Odumeru, 2004).  Some provinces have also instituted their own recommendations, and in British Columbia, the Ministry of the Environment recommends fewer than 200 FC per 100 ml, and fewer that 77 E. coli 26  per 100 ml in all waters used for irrigation of produce (Government of British Columbia, n.d.).  These levels are all based on single measurements.  Currently in Canada, however, no water quality standards for irrigation water have been legislated.  In the United States, recent legislation under the Food Safety Modernization Act has implemented the monitoring of trends for E. coli as a way to estimate water quality.  The law requires a geometric mean of below 126 CFU of E. coli per 100 ml across 20 consecutive samples, but no sample should exceed 410 CFU of E. coli per 100 ml.  Furthermore, no E. coli can be present in any water used for during or after harvest, or to irrigate sprouts (U.S. Food and Drug Administration, 2015).  In Canada, the recommended testing method for FC and E. coli in water samples is the Most Probable Number (MPN) method (MFHPB-19; Christensen et al., 2002).  In this method, multiple tubes of EC broth with MUG (4-methylumbelliferyl-β-D-glucuronide) inoculated with three or more decimal dilutions of the water sample are incubated at 45°C for 24 hours.  After this period, the production of gas due to fermentation of the lactose in the EC broth is indicative of the presence of fecal coliforms, while the presence of fluorescence under UV light is indicative of glucuronidase activity, suggesting the presence of E. coli.  Using the number of tubes of each dilution positive for the indicator of interest, a statistical estimate of the concentration can be determined.   Alternatively, E. coli can be enumerated by a membrane filtration method on mTEC media containing BCIG (5-bromo-6-chloro-3-indolyl-β-D-glucuronide).  A suitable volume of water sample is filtered through a 0.45 μm filter which is subsequently placed on the agar medium and 27  incubated at 45°C for 24 hours.  The resulting colonies are considered to be FC.  Strains of E. coli will hydrolyze the BCIG substrate, resulting in a red or magenta colour (Health Canada, 2012).    The use of fecal indicators as predictors for the presence of pathogenic bacteria has shown limited success in recent studies.  A recent investigation by Pachepsky et al. (2014) reviewed 81 datasets comparing concentrations of FC and E. coli in surface waters with the concentrations of one or more foodborne pathogenic organisms and only found a significant relationship in 35% of the cases reviewed.  Moreover, in a study in California, the concentration of E. coli was found to be  not associated with the occurrence of VTEC O157 or Salmonella in either sediment or in surface water (Benjamin et al., 2013).     Correlations between indicator organisms and pathogenic bacteria are more often related to the probability of finding foodborne pathogens in the long-run more so than predicting their presence at a single point in time.  For instance, a cross-Canada investigation of foodborne pathogen occurrence in surface waters found that the annual mean concentration of E. coli at a particular site was significantly correlated with the mean annual recovery of foodborne pathogens at the same site (Edge et al., 2012), and in Alberta, the mean seasonal concentration of FC in waters was correlated with the seasonal occurrences of VTEC O157 and Salmonella (Jokinen et al., 2010).  Similarly, a study in Spain found that samples positive for VTEC or Salmonella had a higher average concentration of E. coli than negative samples (Truchado et al., 2016).  Elevated levels of indicators may or may not provide an early warning for the added risk of foodborne pathogens.  For instance, a study in Belgium by Holvoet et al. (2014) found that 42% 28  of water samples with higher than 100 E. coli per 100 ml were positive for Salmonella or VTEC associated genes, compared to only 10% of the samples below this concentration of E. coli.  On the other hand, a Canadian study by Edge et al. (2012) found that 80% of water samples with fewer than 100 E. coli per 100 ml were positive for at least one foodborne pathogen.    The correlations of indicators with the likelihood of finding pathogens in surface waters may result from similar factors affecting the presence of the various indicator organisms and the pathogens themselves.  For instance, in the Salmon River of British Columbia, both FC and the pathogens VTEC O157 and Salmonella correlated significantly with precipitation.  Furthermore, concentrations of E. coli  and Salmonella were observed to be higher at sample sites in regions affected by animal agriculture when compared to non-agricultural reference sites (Edge et al., 2012).  1.2.4 Environmental factors associated with pathogen occurrence The occurrence of foodborne pathogens in water samples has been observed to correlate with various environmental factors such as landscape, weather, and season.  As mentioned above, a cross-Canada survey of surface waters found Salmonella  to occur more frequently in areas affected by animal agriculture (Edge et al., 2012).  Similarly, recovery of L. monocytogenes from surface water was correlated with proximity to the nearest upstream dairy farm in Ontario (Lyautey et al., 2007b), and proximity to land used for pasture in New York State (Chapin et al., 2014).    Pathogen occurrence has been observed to change over season, but not always consistently in different areas.  The occurrence of VTEC O157 was observed to be higher during the summer 29  months in Alberta (Gannon et al., 2004)  and Georgia (Gu et al., 2013), but was more common during the winter in California (Cooley et al., 2013).  In British Columbia, VTEC was also more common during the winter months (Nadya et al., 2016).  More consistency has been observed for L. monocytogenes, which is more prevalent during the winter months (Cooley et al., 2014; Strawn et al., 2013a), and Salmonella, which has been observed to be more common during the summer months (Haley et al., 2009; Jokinen et al., 2012).  Precipitation has also been observed to correlate with pathogen occurrence.  In Georgia, higher precipitation was shown to correlate with higher occurrence of VTEC O157 (Gu et al., 2013) and Salmonella (Haley et al., 2009).  Similarly, precipitation three days prior to sampling was significantly correlated with the occurrence of VTEC O157 in Alberta (Jokinen et al., 2012), VTEC in British Columbia (Nadya et al., 2016), and Salmonella in New York State (Jones et al., 2014).  1.3 Research purpose To date, only one study has investigated the occurrence of VTEC in the surface waters of the Lower Mainland of British Columbia, and none have investigated the occurrence of L. monocytogenes or Salmonella.  Furthermore, no studies in the area have investigated the relative occurrences these three pathogens together, or studied any geographical and environmental sources affecting occurrence.  Therefore, the primary purpose of this study was to investigate the occurrence of VTEC, Salmonella, and L. monocytogenes in surface waters used for irrigation in the Lower Mainland of British Columbia, and to assess the usefulness of various predictors of their 30  presence.  These results can be used to provide information to local produce growers to help them reduce the risk of fresh produce contamination with foodborne pathogens during irrigation.  Four hypotheses were tested during this study: 1. The three foodborne pathogens, VTEC, Salmonella, and L. monocytogenes are present in surface waters used for irrigation in the Lower Mainland of British Columbia. 2. The occurrence of VTEC, Salmonella, and L. monocytogenes in surface waters in the Lower Mainland is not uniform across space or time.  3. Fecal coliforms and/or generic E. coli are suitable indicators of the occurrence of VTEC, Salmonella, and L. monocytogenes in surface waters in the Lower Mainland. 4. The presence of VTEC, Salmonella, and L. monocytogenes in surface waters used for irrigation in the Lower Mainland is effected by geographical and environmental factors.  To test these hypotheses, three objectives were included in this study: 1. To determine the occurrence rates of VTEC, Salmonella, and L. monocytogenes in surface waters used for irrigation of fresh produce at multiple sites within two distinct water sheds in the Lower Mainland: the Sumas watershed, and the Serpentine watershed. 2. To evaluate the relationship between VTEC, Salmonella, and L. monocytogenes occurrence in surface waters with the concentrations of fecal coliforms and generic E. coli using two different methods. 3. To investigate the relationship between VTEC, Salmonella, and L. monocytogenes occurrence in surface waters with local weather patterns and the proximity to and density of upstream animal agriculture sources.  31   Chapter 2: Occurrence of Verotoxigenic Escherichia coli, Listeria monocytogenes, and Salmonella in the Irrigation Waters Used in Two Watersheds of the Lower Mainland of British Columbia  2.1 Introduction Foodborne pathogens in the environment, specifically in waters used for irrigation of fresh produce, are a potential source of contamination for a variety of produce related foodborne illness outbreaks (reviewed in Uyttendaele et al. 2015).  Studies have shown the presence of Listeria monocytogenes, Salmonella, and both O157 and non-O157 verotoxigenic Escherichia coli (VTEC) in various watersheds across Canada (Edge et al., 2012; Johnson et al., 2014; Lyautey et al., 2007b), including a recent investigation of the prevalence of VTEC in the surface waters of the Lower Mainland of British Columbia (Nadya et al., 2016).    Fresh produce is the vehicle for a significant amount of foodborne disease.  It was the food type most associated with foodborne outbreaks in the United States between 2004 and 2013 (Fisher et al., 2015), and was responsible for 27 foodborne outbreaks in Canada between 2001 and 2009 (Kozak et al., 2013).  Contaminated irrigation water has been suggested as a possible source for the transmission of foodborne pathogens to produce (Steele and Odumeru, 2004), and recent work has highlighted the potential for bacterial pathogens to attach and potentially internalize into produce during the preharvest stage (Erickson et al., 2010; Hintz et al., 2010; Mootian et al., 2009; Oliveira et al., 2011). 32   Therefore, the objective of this part of the project was to determine the occurrence of VTEC, Salmonella, and L. monocytogenes in the irrigation waters used in the Lower Mainland of British Columbia, and to investigate whether there is evidence for geographic or seasonal trends in their isolation.  2.2 Materials and methods 2.2.1 Site selection Sampling sites were located in two distinct watersheds: The Serpentine watershed in the Cloverdale region of Surrey, BC, and the Sumas watershed on the Sumas prairie near Abbotsford, BC.  Three surface water sites (i.e., ditch, creek, or stream) within each watershed were chosen which represented water adjacent to vegetable growing fields (Figure 2.1).  A fourth site (i.e., Sumas 4) was added to the Sumas watershed during the first summer of sampling.  2.2.2 Sample collection Water samples were collected from each site once per month from February 2015 to April 2015, then twice per month until August 2016, with the exception of October 2015 when only one sample was collected from each site.  Sampling began in February 2015 for Sumas 1, March 2015 for Sumas 2 and Sumas 3, April 2015 for Serpentine 2, May 2015 for Serpentine 1 and Serpentine 3, and July 2015 for Sumas 4.  Surface water samples were collected from the ditches in 532 ml Stand-Up Whirl-Pak® bags using a telescopic sampling pole designed to hold the Whirl-Pak® bags (Nasco, Fort Atkinson, WI).  The 33  sampling pole was rinsed thoroughly with 70% ethanol before and after sampling at each site.  Five sample bags were collected at each site: two to be tested for L. monocytogenes and Salmonella, and two to be tested for VTEC.  The fifth sample bag was used to measure the temperature of the sample water (Chapter 4).  2.2.3 Pathogen detection Each sample was tested for the presence of L. monocytogenes, Salmonella, and VTEC in duplicate, as follows.  If either replicate was positive for a pathogen, the site was considered positive for that pathogen on that date of sampling.  2.2.3.1 L. monocytogenes  2.2.3.1.1 Detection and isolation The presence of L. monocytogenes was detected in water samples by Silliker JR Laboratories (Burnaby, BC) using the MFHPB-30 method from Health Canada (Pagotto et al., 2011).  Briefly, 25 ml of sample was enriched in Listeria enrichment broth for 24 and 48 hours at 35°C before being subsequently used to inoculate modified Fraser Broth (MFB) and incubated for 24-26 hours at 35°C.  Positive enrichments in MFB were then streaked onto Oxford agar and PALCAM agar and incubated for up to 48 hours at 35°C.  A minimum of five typical colonies from each plating were then tested for hemolysis, motility, and carbohydrate utilization to confirm the identity as L. monocytogenes.  Positive isolates were then recovered and submitted for serotyping analysis.    34   Figure 2.1 – Location of sampling sites in the Serpentine (red border) and Sumas (green border) watersheds of the Lower Mainland of British Columbia   35  2.2.3.1.2 Serotyping and pulsed-field gel electrophoresis characterization Recovered L. monocytogenes isolates were characterized by serotyping and pulsed-field gel electrophoresis (PFGE) by Dr. Franco Pagotto, Karine Hébert, and Kevin Tyler at the Listeriosis Reference Service, Health Canada, Ottawa, Ontario.  Serotype was only predicted from ribotype due to a defective serology kit at the time of analysis.  PFGE analysis was conducted separately after digestion with the restriction endonucleases AscI and ApaI.    2.2.3.2 Salmonella  2.2.3.2.1 Detection and isolation The presence of Salmonella was detected in the water samples by Silliker JR Laboratories (Burnaby, BC) using the MFHPB-20 method from Health Canada (Reid, 2009).  Briefly, 25 ml of sample was non-selectively enriched for 18 to 24 hours at 35°C in buffered peptone water, followed by selective enrichment of these cultures in Tetrathionate Brilliant Green broth and Rappaport-Vassiliadis Soya Peptone broth for 24 ± 2 hours at 42.5°C.  Selective enrichment cultures were then streaked onto Bismuth Sulfite agar and Brilliant Green Sulfa agar and incubated for 24 ± 2 hours at 35°C.  Suspected colonies were subjected to biochemical screening for carbohydrate utilization, H2S production, and gas formation on Triple Sugar Iron agar; presence and absence of lysine decarboxylase and lysine deaminase, respectively, on Lysine Iron agar; and absence of urease on Christensen’s Urea agar.  Positive isolates were then recovered and submitted for serotyping analysis.  36  2.2.3.2.2 Serotyping Recovered Salmonella isolates were serotyped by Dr. Gitanjali Arya and Dr. Cornelis Poppe at the OIE Reference Centre for Salmonellosis, National Microbiology Laboratory, Public Health Agency of Canada, Guelph, Ontario.  2.2.3.3 Verotoxigenic Escherichia coli detection and isolation The presence of VTEC in water samples was determined using the verotoxin immunoblot (VT-IB) method previously described by Johnson et al. (2014).  A volume of 25 ml of each sample was filtered through a 0.45 μm Hydrophobic Grid Membrane Filter (HGMF; Neogen Corp., Lansing, MI).  The HGMF were then incubated on VT-IB membranes to detect presence of VT production.  Suspected colonies were confirmed using a VT Enzyme Linked Immunosorbent Assay (VT-ELISA).  The VT-ELISA confirmed isolates were then confirmed as E. coli using polymerase chain reaction (PCR) for the gadA gene, and tested to confirm the presence of either of the VT producing genes vt1 and vt2, as well as two other virulence factors, eaeA and hlyA, also using PCR.  These methods are elaborated below.  2.2.3.3.1 Verotoxin immunoblot After vacuum filtration of water samples through HGMF, the filters were over-layed on VT-capture membranes and incubated for 18 – 24 hours at 37°C on tryptic soy agar (BD, Mississauga, ON) containing 1.5 g/L bile salts No.3, 10 μg/ml vancomycin, and 10 μg/ml cefsulodin (Sigma, Oakville, ON) (mTSA-VC).    37  The VT-capture membranes consisted of 0.2 μm nitrocellulose filter membranes (Biotrace, Pall Life Sciences, Mississauga, ON) coated with rabbit anti-VT antibodies specific to all known variants of verotoxin (LFZ, Guelph, ON), and were blocked using 2% bovine serum albumin (MP Biomedicals, Solon, OH, USA) in phosphate buffered saline (PBS; Amresco, Solon, OH, USA).  After incubation, a series of needle pricks were used to mark the relative position of the HGMF on the VT-capture membrane for later re-orientation.  The VT-capture membranes were then probed using indirect detection at room temperature for the presence of VT using a set of four monoclonal mouse anti-VT antibodies (2 μg/ml; LFZ), followed by alkaline-phosphatase labelled anti-mouse IgG (0.02 μg/ml; Jackson ImmunoResearch Inc., West Grove, PA, USA) and visualized using the substrate nitroblue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate (Mandel Scientific, Guelph, ON).  Membranes were washed three times at each step using a wash buffer composed of PBS and Tween-20 (MP Biomedicals) (PBS-T).  Dark purple spots on the membrane suggested the presence of VT.  Relative location of the suspected VT-producing colony on the HGMF was measured using a transparent copy of the HGMF aligned on the VT-capture membrane using the needle pricks.  Up to eight suspected colonies were selected from the HGMF and streaked onto MacConkey agar (BD) and incubated at 37°C for 18-24 hours in preparation of confirmation of VT production using a VT-ELISA.  2.2.3.3.2 Verotoxin enzyme linked immunosorbent assay Up to six colonies of each morphology were collected from each MacConkey plate, and grown in 400 μl of Tryptic Soy agar (BD) containing 1.5 g/L bile salts No.3, 10 μg/ml vancomycin, and 10 μg/ml cefsulodin (Sigma) (mTSB-VC) for 18-27 hours at 37°C with shaking at ~150 rpm.  One 38  hundred microliters of each resulting culture where then transferred, in duplicate, to a 96-well microplate (Fisher Scientific, Ottawa, ON) pre-coated with rabbit anti-VT antibodies specific to all known variants of verotoxin (LFZ) and blocked using 2% bovine serum albumin (MP Biomedicals) in PBS, followed by incubation at room temperature for 30 minutes.  Wells were then probed for the presence of bound VT using indirect detection at room temperature for 30 minutes using 100μl each of a set of four monoclonal mouse anti-VT antibodies (2 μg/ml; LFZ), followed by horseradish peroxidase labelled anti-mouse IgG (0.2 μg/ml; Jackson ImmunoResearch Inc.).  The wells were washed five times with PBS-T between each step, and seven times afterwards.  The presence of bound VT was then visualized by incubating with the substrate tetramethylbenzidine (Sigma) at room temperature with agitation at ~150 rpm for 10 minutes, followed by addition of 100 μl of 0.2 M sulfuric acid (Sigma).  The absorbance of the wells was measured using a SpectraMax M2 Microplate Reader (MTX Lab Systems, Inc., US) at a dual wavelength of 450/620 nm and air as a blank.  Wells were considered positive if the optical density (OD) was greater than two times that of the negative control.  If the OD of both replicate wells of a particular isolate were greater than 1.5 times the negative control, but below twice that of the negative control, they were scored as suspicious and were tested a second time.  Presumptive VTEC isolates found to be positive by VT-ELISA were simultaneously streaked onto TSA to confirm them as a pure culture, and into TSB in preparation for final confirmation using PCR.  2.2.3.3.3 Polymerase chain reaction for confirmation of Escherichia coli and detection of virulence genes Overnight cultures (18-24 hours) of presumptive VTEC were confirmed as E. coli using a monoplex PCR for the gadA gene as described by Doumith et al. (2012).  They were also confirmed for both variants of the VT producing gene (i.e., vt1, vt2) and two other virulence determinants: eaeA and hlyA 39  by a multiplex PCR previously described by Paton and Paton (1998), and modified by Ms. Kim Ziebell of the Public Health Agency of Canada E. coli Reference Lab (Personal communication).  The DNA was extracted from lysates using a boiling method.  A 360 μl aliquot of overnight culture was combined with 40 μl of 10X PBS, pH 7.2 (Amresco) in a 1.5 ml microcentrifuge tube (Fisher Scientific) and incubated for ten minutes in a heating block at 96°C, followed by cooling on ice for five minutes.  The tubes were then centrifuged (Microcentrifuge 5415 R, Eppendorf, Mississauga, ON) at 13,300 rpm for 5 minutes, and the supernatant, containing the DNA, was reserved and stored at -20°C until use.  Monoplex PCR for the confirmation of the isolates as E. coli was conducted using 2 μl of DNA lysate in a 25 μl reaction mixture containing 1X buffer solution, 1X Coral Dye, 5 μl Q-solution, and 0.625 units of TopTaq DNA polymerase (Qaigen, Toronto, ON), along with 50 μM dNTP’s (Invitrogen, Thermo Fisher Scientific, Ottawa, ON) and 1 μM each of the gadA primers (Table 2.1).  The PCR reaction was carried out in a C1000 Touch™ Thermal Cycler (BioRad Laboratories, Mississauga, ON) under the following conditions: a 4 min denaturation at 94°C, followed by 30 cycles of 94°C for 30 sec, 65°C for 30 sec, and 72°C for 30 sec, and a final extension step at 72°C for 5 min.  Products were held at 4°C until visualization.   Multiplex PCR to determine the presence of virulence determinants was conducted using 2 μl of DNA lysate in a 25 μl reaction mixture containing 1X buffer solution, 1X Coral Dye, 5 μl Q-solution, and 0.625 units of TopTaq DNA polymerase (Qaigen), along with 50 μM dNTP’s (Invitrogen) and 1 μM each of the primers for eaeA and hlyA, and 0.4 μM of primers for stx1, stx2, and stx2e (Table 2.1).  The PCR reaction was carried out in a C1000 Touch™ Thermal Cycler 40  (BioRad) under the following conditions: a 3 min denaturation at 95°C, followed by 35 cycles consisting of a denaturation step at 95°C for 1 min; an annealing step at 65°C for 2 min for the first 10 cycles, decrementing 1°C per cycle to 60°C by cycle 15; and an elongation step at 72°C for 1.5 min, incrementing to 2.5 min from cycles 25 to 35; and a final extension step at 72°C for 5 min. Products were held at 4°C until visualization.  Table 2.1 – Sequences of primers used for confirmation of verotoxigenic E. coli isolates Gene Primer Sequence (5’-3’) Amplicon Size (bp) Reference gadA forward GATGAAATGGCGTTGGCGCAAG 373 (Doumith et al., 2012) reverse GGCGGAAGTCCCAGACGATATCC stx1 forward ATAAATCGCCATTCGTTGACTAC 180 (Paton and Paton, 1998) reverse AGAACGCCCACTGAGATCATC stx2 forward GGCACTGTCTGAAACTGCTCC 255 (Paton and Paton, 1998) reverse TCGCCAGTTATCTGACATTCTG eaeA forward GACCCGGCACAAGCATAAGC 384 (Paton and Paton, 1998) reverse CCACCTGCAGCAACAAGAGG hlyA forward GCATCATCAAGCGTACGTTCC 534 (Paton and Paton, 1998) reverse AATGAGCCAAGCTGGTTAAGCT stx2e forward GAACAGATGGAATTTGCAGCCA 112 (Dr. Kim Ziebell, Personal communication) reverse TAAACTTCACCTGGGCAAAGCC      41  Products of the both the monoplex and multiplex PCR reactions were visualized on 2% agarose gel stained with SYBR® Safe (Invitrogen) after electrophoresis in TAE buffer (Invitrogen) at 90V for 45 min.  Expected amplicon sizes are shown in Table 2.1.  A positive control of E. coli O157:H7 possessing vt1, vt2, eaeA, and hlyA was donated by Linda Hoang (BC Centre for Disease Control, Vancouver, BC).  Samples positive for gadA were confirmed to be E. coli, and E. coli positive for any of vt1, vt2, or vt2e were confirmed to be VTEC.    2.2.3.3.4 Serotyping Recovered VTEC isolates were serotyped by Dr. Roger Johnson and Ms. Kim Ziebell at the E. coli Reference Laboratory, Laboratory for Foodborne Zoonoses, Public Health Agency of Canada, Guelph, Ontario.  2.2.4 Statistical analysis Statistical analysis of results was conducted using R version 3.2.3 (R Core Team, 2015; http://www.R-project.org).  Uniformity of occurrence within and between watersheds was completed using either the χ2 test when appropriate, or the Fisher’s Exact test when frequencies of occurrence less than five were present in the calculations.  2.3 Results 2.3.1 Pathogen occurrence A total of 223 water samples were collected between the Sumas and Serpentine watersheds; of those, 16.6% were positive for at least one of the three pathogens tested.  The occurrence of pathogens in the water samples is outlined in Table 2.2.  The most commonly occurring pathogen 42  was L. monocytogenes which was found in 11.2% of the samples, followed by VTEC and Salmonella with 4.9% and 2.7% occurrence, respectively.    The recovery of any of the three pathogens was more common in the Serpentine watershed compared to the Sumas watershed, with 24.5% compared to 10.9% occurrence, respectively (χ2; p = 0.0119).  This difference was primarily due to the increased occurrence of L. monocytogenes in the Serpentine watershed compared to the Sumas watershed: 20.2% and 4.65%, respectively (χ2; p = 0.0006).  Neither VTEC nor Salmonella showed evidence for differences in occurrence between watersheds (χ2; p > 0.10), although all but one occurrence of Salmonella was derived from the Sumas watershed.  Site specific differences were observed for VTEC occurrence within both the Serpentine watershed (Fisher’s exact test; p = 0.046) and the Sumas watershed (Fisher’s exact test; p = 0.075), with all but one positive sample in each watershed coming from a single location. Moreover, VTEC was never recovered at 1 and 2 sites in the Serpentine and Sumas watersheds, respectively.  Similarly, Salmonella showed some evidence of site-specific occurrence in the Sumas watershed (Fisher’s exact test; p = 0.097), especially when taken into account that both sites where the bacterium was recovered were in close proximity (Figure 2.1 and Table 2.2), thereby suggesting a potential for a common source of contamination.  No significant site-specific differences were observed for L. monocytogenes in either watershed, though two thirds of positive samples from the Sumas watershed came from a single site.      43  Table 2.2 – Occurrence of verotoxigenic E. coli (VTEC), L. monocytogenes, and Salmonella in irrigation water collected from seven sites in the Sumas and Serpentine watersheds. Site # of Samples Number of positive samples (% of positive samples) VTEC L. monocytogenes Salmonella Any Pathogen Sumas 1 35 4 (11.4%)α 1 (2.86%) 3 (8.57%)α 6 (17.1%) Sumas 2 34 0 β 1 (2.94%) 0 β 1 (2.94%) Sumas 3 34 1 (2.94%)β 4 (11.8%) 0 β 5 (14.7%) Sumas 4 26 0 β 0 2 (7.69%) α 2 (7.69%) Serpentine 1 30 0 β 6 (20.0%) 1 (11.1%) 6 (20.0%) Serpentine 2 34 5 (14.7%)α 9 (26.5%) 0 12 (35.3%) Serpentine 3 30 1 (3.33%)β 4 (13.3%) 0 5 (16.7%) Total Sumas 129 5 (3.88%) 6 (4.65%)A 5 (3.88%) 14 (10.9%)A Total Serpentine 94 6 (6.38%) 19 (20.2%)B 1 (1.06%) 23 (24.5%)B Total 223 11 (4.93%) 25 (11.2%) 6 (2.69%) 37 (16.6%) A,B Pathogens at watersheds with different letters represent a significant difference in pathogen occurrence between the two watersheds (χ2; p < 0.05) α,β  Pathogens at sites with different letters represent a significant difference in pathogen occurrence between those sites within their respective watershed (Fisher’s exact test; p < 0.05)   The monthly occurrence of pathogens in the water samples is outlined in Figure 2.2.  The most positive samples were collected during February and September.  This increase in recovery is primarily the result of L. monocytogenes which showed a marked increase in occurrence during the cooler months.  No temporal trends for VTEC can be seen, though February is the only month with two positive recoveries of the pathogen.  Also, Salmonella was recovered twice in September and twice in February, representing the bulk of the occurrence save for a single recovery in August 2016.    44   Figure 2.2 – Monthly recovery of verotoxigenic E. coli (VTEC), L. monocytogenes (LM), and Salmonella (SAL) between May 2015 and August 2015 from irrigation water collected at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia.   To investigate seasonal changes in pathogen occurrence, relative occurrence was investigated for winter (December to February), spring (March to May), summer (June to August), and fall (September to November) months.  These results are summarized in Table 2.3.  The recovery of any pathogen was more common during the fall and winter months with 28.6% and 23.3% positive samples, respectively, compared to 15.6% for spring and 8.6% for summer (χ2; p = 0.033).  This was likely driven by differences in recovery of L. monocytogenes which showed 22.9% occurrence in the fall and 16.3% occurrence in the winter, compared to 7.8% and 6.2% for spring and summer, respectively (Fisher exact test; p = 0.038).  No significant difference in occurrence of VTEC or Salmonella was observed, though the lowest occurrence for both pathogens was observed to be during the summer.     45  Table 2.3 – Seasonal occurrence [number of positive samples/total number of samples collected (percent recovery)] of verotoxigenic E. coli (VTEC), L. monocytogenes, Salmonella, or any of the three in water collected from irrigation ditches at three sites each in the Sumas and/or Serpentine watersheds in the Lower Mainland of British Columbia  Winter (Dec-Feb) Spring (Mar-May) Summer (Jun-Aug) Fall (Sep-Nov) Sumas Watershed     VTEC 2/25 (8%) 2/39 (5.1%) 0/45 (0%) 1/20 (5%) L. monocytogenes 2/25 (8%) 1/39 (2.6%) 2/45 (4.4%) 1/20 (5%) Salmonella 2/25 (8%) 1/39 (2.6%) 1/45 (2.2%) 1/20 (5%) Any 5/25 (20%) 4/39 (10.3%) 3/45 (6.7%) 2/20 (10%)      Serpentine Watershed     VTEC 1/18 (5.6%) 3/25 (12%) 1/36 (2.8%) 1/15 (6.7%) L. monocytogenesα 5/18 (27.8%) 4/25 (16%) 3/36 (8.3%) 7/15 (46.7%) Salmonella 0/18 (0%) 0/25 (0%) 0/36 (0%) 1/15 (6.7%) Anyα 5/18 (27.8%) 6/25 (24%) 4/36 (11.1%) 8/15 (53.3%)      Both Watersheds     VTEC 3/43 (7%) 5/64 (7.8%) 1/81 (1.2%) 2/35 (5.7%) L. monocytogenesα 7/43 (16.3%) 5/64 (7.8%) 5/81 (6.2%) 8/35 (22.9%) Salmonella 2/43 (4.7%) 1/64 (1.6%) 1/81 (1.2%) 2/35 (5.7%) Anyα 10/43 (23.3%) 10/64 (15.6%) 7/81 (8.6%) 10/35 (28.6%) α Significant seasonal variation was observed for this pathogen(s) within respective watershed(s) (Fisher exact test; p < 0.05)   Significant seasonal differences were only observed for the Serpentine watershed.  These differences echoed those of both watersheds together, with significant differences being observed for L. monocytogenes (Fisher exact test; p = 0.016) or any of the three pathogens (Fisher exact test; p = 0.018), but not for VTEC or Salmonella.  The occurrence of any of the three pathogens in the Serpentine watershed was highest during the fall with 53.3% occurrence, compared to 27.8%, 24%, and 11.1% for winter, spring, and summer, respectively.  Similarly, the occurrence of L. monocytogenes was greater in the fall at 46.7% occurrence compared to 27.8% occurrence in the winter, 16% occurrence in the spring, and only 8.3% occurrence in the summer.    46  2.3.2 Pathogen characteristics Of the 11 positive VTEC samples, eight distinct serotypes were observed, though isolates from two positive samples were not obtained for serotyping. These results are summarized in Table 2.4.  There was no occurrence of the same serotype being observed at different sites; however, VTEC O69:H11 was observed on two consecutive sampling dates at Serpentine 2.  Also of note was the recovery of O103:H2, a member of the “Big Six” non-O157 VTEC associated with human disease.    Table 2.4 – Serotype and virulence genes of verotoxigenic E. coli isolates collected from irrigation water at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia Serotype Site  Month collected vt1 vt2 eaeA hlyA O103:H2 Sumas 1  Feb 2015 + - + + O109:H5 Sumas 1  Mar 2015 + - - - O116:H25 Serpentine 2  May 2015 - + + + Unknown 1a Serpentine 2  Sep 2015 - + + + O153:NM Sumas 1  Nov 2015 - + + - O76:H19 Sumas 1  Feb 2016 + - - + O69:H11 Serpentine 2  Serpentine 2  Feb 2016 Mar 2016 + - + + Unknown 2a Serpentine 2  Apr 2016 - + + + O34:H2 Sumas 3  May 2016 - + + - O22:H8 Serpentine 3  May 2016 + - - + aThese strains were not isolated for serotyping   In order to further characterize the recovered isolates, the presence of four genes associated with virulence were investigated: the toxin producing genes vt1 and vt2; the intimin producing gene eaeA, responsible for host-cell attachment; and hlyA, producing the pore-forming hemolysin toxin.  The results are summarized in Table 2.4.  There was an equal occurrence of both toxin genes, and no recovered isolate showed the presence of both toxin genes.  The virulence genes eaeA and hlyA were each present in 8 of the eleven VTEC isolates recovered, with six isolates showing the presence of both.   47   Three distinct serotypes of L. monocytogenes were observed over the course of this study and are summarized in Table 2.5.  No positive isolate was obtained for serotyping in two positive samples.  The serotype 1/2a was the most common (15 occurrences) followed by 4b (10 occurrences), 1/2b (2 occurrences), and 4c (1 occurrence). Within the Serpentine watershed, serotype 1/2a and 4b were equally common with 10 recoveries each, along with one recovery each of 4c and 1/2b.  This is in contrast with the Sumas watershed, where only serotype 1/2a was ever recovered except for one occurrence of 1/2b.    Also of note is that on five separate sampling dates, two different serotypes were observed at the same site.  This occurred four times for sample site Serpentine 2, and once for Serpentine 1.    48  Table 2.5 – Serotype and PFGE patterns of L. monocytogenes isolates collected from irrigation water at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia Serotype PFGE-AscI PFGE-ApaI Site Date 1/2a LMACI.0172 (03-7760) LMAAI.0217 Serpentine 2 Mar, 2015  LMACI.0906 LMAAI.0909 (PNC_9089) Sumas 3 Mar, 2015  LMACI.0084 (BOM_4) LMAAI.0531 Sumas 3 Jun, 2015  LMACI.0216 LMAAI.1338 Serpentine 1 Sep, 2015  LMACI.0041 LMAAI.0033 Serpentine 3 Sep, 2015  LMACI.0011 LMAAI.0015 Serpentine 1 Sep, 2015  LMACI.0118/221/681 LMAAI.0213 Serpentine 2 Oct, 2015  LMACI.0543 LMAAI.0524 Serpentine 3 Oct, 2015  LMACI.0195 LMAAI.0252 Sumas 2 Feb, 2016  LMACI.0155 LMAAI.0165 Sumas 3 Feb, 2016     Jul, 2016  LMACI.0122 (02-2448) LMAAI.0003 Serpentine 2 Feb, 2016  LMACI.0738 LMAAI.1076 Serpentine 2 Mar, 2016  LMACI.0044 (PNC_08-5757)/LMACI.0616 (ON_10PF0153) LMAAI.0193/0818 Serpentine 2 Jul, 2016  New New Serpentine 2 Feb, 2016  New New Serpentine 2 Feb, 2016 1/2b New New Sumas 1 Nov, 2015  New LMAAI.0548 Serpentine 2 Dec, 2015 4b LMACI.0003 (PNC_08-2076) LMAAI.0019 Serpentine 2 Mar, 2015    Serpentine 1 Jan, 2016  LMACI.0051 LMAAI.0048 Serpentine 2 Apr, 2015     Sep, 2015     Feb, 2016a     Feb, 2016a    Serpentine 1 Jan 2016  LMACI.0071 LMAAI.0022 Serpentine 3 Nov, 2015  LMACI.0822 (PNC_15-0050) LMAAI.1234 Serpentine 2 Dec, 2015  LMACI.0009 (03-7263) LMAAI.0112 Serpentine 3 Apr, 2016 4c LMACI.0051 LMAAI.0048 Serpentine 1 Sep, 2015 aThese represent two different sampling dates within the same month    49   Within the 25 water samples found positive for L. monocytogenes, there were 19 unique PFGE fingerprints observed.  Three sample sites showed recurring PFGE fingerprints on multiple sampling dates.  The same PFGE fingerprint was observed twice at Serpentine 1, four times at Serpentine 2, (the same PFGE fingerprint all four times), and twice at Sumas 3.  Furthermore, two PFGE fingerprints were observed at two different sampling sites within the Serpentine watershed: Serpentine 1 and Serpentine 2.  Of particular interest, one of these multi-site PFGE fingerprints was also a recurring fingerprint, being recovered twice at Serpentine 1, and four times at Serpentine 2.  Of the six Salmonella isolates recovered, four distinct serotypes were observed, summarized in Table 2.6.  The most common, and only recurring serotype was S. Enteritidis, which was recovered three separate times.  Interestingly, these recoveries were at two different sites; however, these two sites are in close proximity, so it is likely that this may have come from a single source.  Table 2.6 – Serotypes of Salmonella isolates collected from irrigation water at three sites each in the Sumas and Serpentine watersheds of the Lower Mainland of British Columbia Serotype Site Month Typhimurium Sumas 1 Apr 2015 Enteritidis Sumas 4 Sumas 1 Sumas 4 Sep 2015 Feb 2016 Aug 2016 Daytona Serpentine 1 Sep 2015 Heidelberg Sumas 1 Feb 2016   50  2.4 Discussion Foodborne pathogens in water used to irrigate food crops have the potential to contaminate pre-harvest produce and lead to foodborne illness in consumers (Steele and Odumeru, 2004); therefore, it is important to understand the occurrence, prevalence, and spatial/ temporal spread of these pathogens in the water used for local produce production.  The objective of this portion of the study was to investigate the occurrence of three foodborne pathogens from the irrigation waters used in the Lower Mainland of British Columbia: VTEC, Salmonella, and L. monocytogenes.    The most commonly observed pathogen was L. monocytogenes, which was expected due to its relative ubiquity in the natural environment (Vivant et al., 2013), followed by VTEC then Salmonella.  This trend differs from two previous studies which looked at the occurrence of these three specific pathogens in surface waters from agricultural areas: one from New York State (Strawn et al., 2013a), and one from California (Cooley et al., 2014).  Strawn et al. (2013a) found L. monocytogenes to be more prevalent than the other two pathogens, but a higher occurrence of Salmonella compared to VTEC was also reported.  Similarly, Cooley et al. (2014) also reported Salmonella to occur more often than VTEC, but also more often than L. monocytogenes.  The reduced recovery of VTEC in these two studies can partially be explained by differences in detection methods, as discussed below.  The observed occurrence of L. monocytogenes by Cooley et al. was not unusual, but the recovery of Salmonella (65%) was higher than in other studies.  This was likely due to a higher volume of water sampled, as discussed below.  51  2.4.1 VTEC occurrence 2.4.1.1 Environmental occurrence of VTEC The overall recovery of VTEC during the course of this experiment was 4.93%; however, recovery ranged from 3.9% to 6.4% between watersheds.  At the site level, three of seven sites showed consistently no occurrence of VTEC. Of the remaining four sites, recovery ranged from 2.9% to 14.7%.  Few other studies in Canada have investigated the occurrence of non-O157 VTEC, but one investigation of an Ontario watershed reported a recovery of 32% for all VTEC (Johnson et al., 2014). Similarly, a previous project investigating surface waters in the Lower Mainland found the occurrence for VTEC to be in the range of 20.3% (Nadya et al., 2016).  Both of these studies reported higher levels of VTEC recovery than found during this project.  Similarly, a study in California observed an 8% occurrence for VTEC O157, and 11% occurrence for non-O157 VTEC (Cooley et al., 2014), and in New York State, VTEC occurrence in surface water was found to be 2.3% (Strawn et al., 2013a).  Previous surface water studies in Canada focused on VTEC O157, indicated that the relative occurrence ranged from 1% to 3% (Edge et al., 2012; Gannon et al., 2004; Johnson et al., 2003; Jokinen et al., 2011).  Of particular note is that, unlike any of these previous studies, no VTEC O157 was recovered during this project.    Possible reasons for the range of differences in VTEC recovery may be due to differences in sampling volume or method.  For example, both Johnson et al. (2014) and Nadya et al. (2016) filtered 70-100 ml of water in duplicate samples for the VT-IB assay.  For the sake of consistency with the detection and isolation methods used for Salmonella and L. monocytogenes, as well as filter performance, a smaller volume (25 ml) of water from duplicate samples was chosen in this study.  The relative recovery in the present study was on the order of one quarter that of Nadya et 52  al., suggesting similar results were obtained considering the similar area of sampling.  Alternatively,  Cooley et al. (2014) made use of a submersed Moore swab for 24 hours, which does not allow comparison for sample volume. Instead, it is assumed that large quantities of water are filtered through the swab over the 24-hour period, implying the potential for a significantly higher recovery. However, the overall recovery, only about 19%, was lower than expected with such a high volume.  Similarly, Strawn et al. (2013) also tested a larger volume of water (250 ml) and found a lower overall recovery for VTEC.  The lower relative recovery by Cooley et al. (2014) and Strawn et al. (2013) compared to relative sample volumes may have also stemmed from the difference in detection method for VTEC.   Both groups used sample enrichment in TSB prior to detection of vt genes by real-time PCR.  Previous work by Johnson et al. (2014) showed that increased recovery of VTEC, from 7.5% to 32%, was observed by switching to smaller volumes of water with no enrichment.  It could be suggested that this is due to other microorganisms out-competing VTEC during the non-selective enrichment.  If the differences in recovery are indeed due to this updated detection procedure, then the occurrence of VTEC in agricultural waters may be much higher than previously believed.    2.4.1.2 Spatial and temporal distribution of VTEC While no significant difference was observed in the occurrence of VTEC between watersheds, there was a clear site dependence within watersheds.  In both watersheds, all but one occurrence of VTEC was from a single site.  Furthermore, in the Sumas watershed, the two sites where VTEC was recovered are in close proximity to each other and share a common source, meaning these two sites are likely not independent of each other.  This site-specific trend is consistent with previous 53  studies (Cooley et al., 2014; Johnson et al., 2014; Nadya et al., 2016) and implies that there is a geographical dependence on the likelihood of finding VTEC, such as an increase in occurrence in agricultural areas compared to control sites located away from agriculture and waste areas (Edge et al., 2012).  These differences may be attributed to proximity to a host source, such as livestock (Chapter 4) or the application of manure to neighbouring fields.   No significant difference in the occurrence of VTEC across seasons was observed, however recoveries were noticeably lower during the summer months at only 1.2% compared to 5.7%, 7%, and 7.8% for fall, winter, and spring, respectively.  The seasonal effect on VTEC occurrence has varied between different studies.  A study of VTEC O157 occurrence in Southern Alberta found that peak isolation happened in July,  and lower recoveries were observed in the spring and fall (Gannon et al., 2004).  Conversely, Cooley et al. (2014) found higher occurrence of VTEC in the winter and spring, compared to the summer and fall in California, and Nadya et al. (2016) observed higher recoveries for VTEC during the winter than in the other three seasons in the Lower Mainland of British Columbia.  No seasonal differences were observed by Johnson et al. (2014) in Southern Ontario.  Interestingly, it is accepted that the shedding of VTEC O157 by cattle, a common reservoir for this bacterium, is greater during the summer months (Hancock et al., 2001), though a more recent study found the occurrence of the VTEC O157 in beef cattle feces reported no seasonal variation (Alam and Zurek, 2006).  These different observations may imply that seasonality is not the cause of the observed differences and perhaps is confounded with a more indicative factor such as changes in agricultural practices or weather patterns during these seasons.  For instance, cooler months on the West Coast of Canada are often associated with higher rates of 54  precipitation (Chapter 4), and during the summer months, damning of the irrigation ditches results in a stopping or reducing of water flow.    2.4.1.3 Serotype and pathotype of recovered VTEC Of the VTEC recovered during this study, no recurring serotypes were observed, save for O69:H11, which was recovered twice at the same site (Serpentine 2) on successive samplings.  Interestingly, two isolates recovered from the sample site Serpentine 2 on two different occasions showed the same genotypic virulence pattern (i.e., vt2, eaeA, and hlyA) as a VTEC O116:H25 isolated from Serpentine 2 earlier in the study, but were not retained for serotyping.  This is disappointing since we may have found evidence of the recurrence of this serotype.  No other recovered serotype showed this same virulence factor pattern. The same virulence factors were shared between an O103:H2 isolate and an O69:H11 isolate, however, suggesting that the virulence genotype may be shared across serotype, and is insufficient to link untyped isolates.  No occurrence of VTEC O157:H7 was observed during this study, which was interesting since every other recent survey of water in Canada has shown the presence of VTEC O157:H7, although always at relatively low levels between 1% and 3% (Edge et al., 2012; Gannon et al., 2004; Johnson et al., 2003; Jokinen et al., 2010).  Of the serotypes that were observed herein, only one, O103:H2, is a member of the “Big 6” non-O157 serotypes associated with a high proportion of severe human illness (Croxen et al., 2013).  VTEC O103:H2 has been isolated from human cases of illness both in North America (Karama et al., 2008a), and Europe (Mariani-Kurkdjian et al., 1993; Prager et al., 2002), and was responsible for an outbreak at a nursery school in Japan (Muraoka et al., 2007).  Furthermore, serotype O103 is commonly associated with illness in British 55  Columbia (Wang et al., 2013).  Other disease associated isolates include O76:H19 which was responsible for a household outbreak of bloody diarrhea in Spain (Sanchez et al., 2014), and O22:H8 which has been associated with bloody diarrhea in Germany (Bielaszewska et al., 2006) and isolated from patients exhibiting symptoms of HUS (Constantiniu, 2002).  Members of serogroup O69 and O153 have also been isolated with patients exhibiting diarrhea in India (Gazal et al., 2014)  Of the serotypes recovered during this study, five had previously been isolated from surface waters of the Lower Mainland: O22:H8, O69:H11, O76:H19, O103.H2, and O116:H25 (Nadya et al., 2016).  Other serotypes not previously recovered in this area were O34:H2, O109:H5, and O153:NM.  Common serotypes between studies, however, did not guarantee identical virulence genotypes.    Genotype analysis of VTEC showed an even proportion of isolates possessing vt1 and vt2, with no recovered isolates possessing both toxin producing genes.  This is in contrast with the results of Nadya et al. (2016), where there was a 50% greater number of vt1 possessing isolates compared to vt2.  Nadya et al. also observed a 10% occurrence of isolates which had both variants of the toxin producing gene.  The intimin gene, eaeA, was observed in seven of the nine observed serotypes, and was more commonly associated with the vt2 variant, which is of concern since the presence of these two genes has been most associated with EHEC symptoms, (i.e., enterohemorrhagic colitis, hemolytic-uremic syndrome) in humans (Boerlin et al., 1999).  Observed serotypes with vt2 and eaeA included O116:H25, O153:NM, O34:H2, and both of the un-serotyped VTEC isolates.  On the other hand, hlyA was also present in seven of the observed 56  serotypes, but was more commonly associated with vt1 which is similar to observations by Nadya et al.  No other work has been done to investigate virulence factors of VTEC recovered from surface waters, but a recent survey of VTEC isolated from produce by the US FDA found that vt2 was over five times more common than vt1 in those isolates, with the presence of eaeA and hlyA being 9% and 61%, respectively (Feng and Reddy, 2013).    Similar VTEC serotypes recovered from a previous study in the same area did not necessarily share the same virulence factors.  The O116:H25 isolates recovered by Nadya et al. (2016) in the nearby Nicomekl watershed were positive for vt2, but not eaeA, where an O116:H25 possessed both of these virulence factors in the present study.  Furthermore, O22:H8 isolates recovered by Nadya et al. were only positive for vt2 and not the other three virulence factors, whereas the O22:H8 isolate recovered in this study was positive for vt1, and hlyA.  This discrepancy between virulence factors and serotype begs the question as to whether serotype or genotype is a better predictor of human pathogenicity, since serotype is considered to be a reasonable predictor of pathotype (Boerlin et al., 1999).  The investigation of genetic markers for molecular risk assessment of potential EHEC strains has shown promise, and that the presence of eae in combination with a variety of other virulence determinants is a strong indicator for severe human pathogenicity (Bugarel et al., 2010).  Care should be taken, however, as a major outbreak in Germany and France was the result of serotype O104:H4 which acquired the vt2 gene, but did not have any other virulence factors typically associated with VTEC or EHEC.  Instead, this strain was a member of the enteroaggregative E. coli and possessed completely different genetic mechanisms of intestinal colonization (Navarro-Garcia, 2014).  The EHEC O104:H4 outbreak also highlighted the transient nature of the Shiga-toxin producing prophage, and reminded us that new VTEC 57  serotypes are likely to develop in the future, and that targeting of virulence factors may be the best way to monitor for VTEC/EHEC occurrence in the environment compared to serotype specific methods such as immunomagnetic separation.  2.4.2 Salmonella occurrence 2.4.2.1 Environmental occurrence of Salmonella The overall recovery of Salmonella across all sites during this study was 2.69%, but ranged from 7.7% to 11.1% across the three sites that were positive for this pathogen at least once.  Four sample sites, two in each watershed, showed no occurrence of Salmonella.   These recoveries are lower than those from a previous cross-Canada study, which included the Sumas river, where the annual mean occurrence for Salmonella was 11% (Edge et al., 2012), and another study in the Salmon River watershed of British Columbia where a mean occurrence of 13% was observed (Jokinen et al., 2010), though these two studies both made use of 500 ml samples of water, much higher than the 25 ml sample volume used in this study.  Two studies in Southern Alberta observed the relative occurrence of Salmonella in surface water to be 10.3% (Gannon et al., 2004) and 6.2% (Johnson et al., 2003), respectively, but both studies observed differences in recoveries between years.  Both of these studies used 90 ml water samples for analysis, while a later study in the same watershed used 500 ml water samples and observed a relative occurrence of 8.5% for Salmonella in water samples (Jokinen et al., 2011).  Similarly, two studies in New York State found the occurrence of Salmonella in agricultural waters ranged between 9% and 11% (Strawn et al., 2013a, 2013b).  Interestingly, a study of agricultural water in California found a relative occurrence of Salmonella to be 65%; much higher than other studies (Cooley et al., 2014).  It should be noted, however, that 24-hour Moore swabs were used for sampling, meaning an indeterminately large volume of water 58  could have been “filtered” through the swab.  The authors also acknowledged that the swab will also capture sediment particles, where Salmonella are more common.  2.4.2.2 Spatial and temporal occurrence of Salmonella No significant difference in occurrence for Salmonella was observed between watersheds, though only one occurrence was observed in the Serpentine watershed compared to five occurrences in the Sumas. Site dependence within watersheds was observed, but the difference was only significant for the Sumas watershed.  It should also be noted that the two sites in the Sumas watershed where Salmonella was isolated were in close proximity and could be considered as a single site.  This suggests that site dependence is strong within the Sumas watershed.  Previous studies have also shown differences in Salmonella occurrence across various sampling sites.  Cooley et al. (2014) observed specific “hotspots” where recovery of Salmonella was as high 96%, and agricultural sites have been observed to show significantly higher levels of Salmonella compared to sites not in proximity to agricultural activity (Edge et al., 2012), which implies that proximity to a contamination source such as livestock likely plays a role in the probability of occurrence.  A significantly higher recovery of Salmonella from smaller streams, rather than larger streams in agricultural areas has also been observed (Edge et al., 2012).  Due to the limited recovery of Salmonella in this study, no seasonal effect on occurrence could be observed.  Previous studies have documented a significant increase in the occurrence of Salmonella in the spring compared to the summer (Jokinen et al., 2010) while others have seen an increase in recovery during the summer time (Cooley et al., 2014; Gannon et al., 2004).  This lack of consistent seasonal trend may suggest that apparent seasonal differences may be due to 59  confounding factors within the sampling area, such as changes in precipitation or farming practices (i.e., manure application) during that time.    2.4.2.3 Serotype analysis of recovered Salmonella Over the course of this study, four different Salmonella serotypes were recovered.  The only one recovered more than once was S. Enteritidis, which was observed at two sites in the Sumas watershed.  It should also be noted that these two sites are in close proximity and connected, suggesting the possibility that this recurrence may be from a single source.  S. Enteritidis has been implicated previously in outbreaks involving mung bean sprouts (Honish and Nguyen, 2001), and alfalfa sprouts (Centers for Disease Control, 2016c).  S. Typhimurium was also recovered, which as been implicated in outbreaks involving tomatoes (Centers for Disease Control, 2007) and a recent outbreak associated with cantaloupe (Centers for Disease Control, 2016c).  Also recovered was S. Heidelberg, which has not been implicated in any produce related outbreaks, but is commonly associated with poultry and considered to be one of the top serovars associated with human disease, and often leads to invasive infection (Foley et al., 2011).  Finally, S. Daytona has not been implicated in any foodborne outbreaks, to the best of the author’s knowledge.  S. Enteritidis, S. Typhimurium, and S. Heidelberg were also recovered from surface water in Southern Alberta; however, these serotypes were relatively rare compared to S. Rubislaw which made up 52% of the recovered isolates (Gannon et al., 2004).  A previous study in the Salmon River of British Columbia found S. Typhimurium to be the most common serotype observed (Jokinen et al., 2010).  60  2.4.3 Listeria monocytogenes occurrence 2.4.3.1 Environmental occurrence of Listeria monocytogenes The overall occurrence for L. monocytogenes in this study was 11.2%, but differed significantly between the two watersheds, ranging from 4.7% to 20.2%.  The relative occurrence was similar to that observed during a study from Southern Ontario where L. monocytogenes was recovered from 10% of surface water samples (Lyautey et al., 2007b).  It should be noted, however, that this previous study used the same detection method as used in our study, but with a greater volume of water sampled, meaning that our observations may imply a higher concentration of L. monocytogenes in the Lower Mainland water.  A higher occurrence (28%-30%) of the pathogen in agricultural waters in New York State (Strawn et al., 2013a, 2013b), and higher still in agricultural waters in California with an occurrence of 43% (Cooley et al., 2014).  Both of these American studies made use of larger volumes of water sampled, which may partly explain the disparity in observed results.    2.4.3.2  Spatial and temporal occurrence of Listeria monocytogenes There was a significant difference in L. monocytogenes occurrence between watersheds, with occurrence in the Serpentine watershed being four times greater than that of the Sumas watershed.  Differences between watersheds have also been reported in California, as well as the presence of “hotspots” where recovery for the pathogen was as high as 96% (Cooley et al., 2014).  These differences in occurrence and the presence of “hotspots” suggests geographical or environmental factors can have an effect of the presence and persistence of L. monocytogenes.  Indeed, previous work has observed L. monocytogenes to be more frequently recovered from uncultivated compared to cultivated agricultural soils (Vivant et al., 2013).   61   Seasonal dependence was also observed for L. monocytogenes occurrence, with fall and winter showing higher recovery compared to both spring and summer.  Winter and spring were found to show a higher occurrence of L. monocytogenes in California (Cooley et al., 2014), suggesting a trend that at least the cooler months show greater occurrence of L. monocytogenes compared to the heat of the summer.  Temperature may be a driving force of this phenomenon, as survival of L. monocytogenes in a water environment has been shown to double at 4°C when compared to 20°C (Budzińska et al., 2012), meaning that any contamination of the water will persist for longer periods during cooler months.  Unfortunately, many other studies did not investigate occurrence during winter due to below freezing temperatures in the study area.  2.4.3.3 Serotyping and PFGE analysis of L. monocytogenes isolates Over the course of this study, the recovered isolates of L. monocytogenes were overwhelmingly members of the serotypes 1/2a and 4b except for one isolate deemed to be serotype 4c and one isolate 1/2b.  This is of concern since the serotypes 1/2a, 1/2b, and 4b are responsible for the majority of human listeriosis cases (McLauchlin et al., 2004).  Most recently, a large multi-state outbreak related to cantaloupes was caused by strains from serotypes 1/2a and 1/2b (McCollum et al., 2013), and outbreaks related to caramel apples (Angelo et al., 2017) and packaged salads (Chen et al., 2017) were the result of strains from serotype 4b.   To the best of the author’s knowledge, no outbreaks associated with serotype 4c have been reported.  Serotype 1/2a was the most common serotype observed, representing 54% of the recovered isolates.  This increased representation of 1/2a, however, is directly tied to the Sumas watershed 62  where this serotype was the only one recovered.  In the Serpentine watershed, equal recoveries of 1/2a and 4b were observed at all three sampling sites.  Previous studies in Canada only used PCR to identify serogroups, but found that the serogroup consisting of 1/2a and 3a represented 49% and 68% of recovered isolates in watersheds in Ontario (Lyautey et al., 2007b) and Nova Scotia (Stea et al., 2015), respectively.  Similarly, in Austria, two thirds of isolates from soil and water were found to cluster into lineage II, which includes 1/2a, compared to only one third from lineage I, which includes 4b and 1/2b (Linke et al., 2014).  In contrast, surveys in California found serotype 4b isolates to make up >85% of recovered L. monocytogenes compared to 1/2a which only represented 7-8% of recovered isolates (Cooley et al., 2014; Gorski et al., 2014).  It has been suggested that isolation methods may be biased, with lineage II isolates outcompeting lineage I isolates in during enrichment (Bruhn et al., 2005); however in this study we saw isolation of both 1/2a and 4b serotypes from single samples suggesting both were enriched successfully.  Instead, these differences suggest that geography or the local environment may affect the occurrence of various serotypes.  For instance, serotype 1/2a has been observed to be more common overall within the natural environment (Schaffter et al., 2004), whereas serotype 4b has been more commonly recovered from the feces of wild animals (Orsi et al., 2011) and cattle (Schaffter et al., 2004).  Therefore, the occurrence of serotype 4b in the Serpentine watershed could be the result of an animal reservoir in the area with the observed 1/2a isolates spread across the natural environment.  Persistence of certain strains was also observed using PFGE analysis of the recovered L. monocytogenes isolates.  Multiple recoveries of isolates showing the same PFGE pattern were observed at three sites.  Isolates of serotype 1/2a showing the same PFGE pattern were recovered 63  twice at Sumas 3, five months apart, suggesting a potential upstream point source for the bacterium.  More importantly, however, isolates of serotype 4b showing a single identical PFGE pattern were recovered four times from Serpentine 2 and once from Serpentine 1 over an 11-month period, suggesting not only persistence but the potential for a mobile source of this bacterium (i.e., wildlife) since neither site is upstream of the other.  This same PFGE fingerprint was observed a second time at Serpentine 1, but interestingly was serotyped as 4c, a member of a completely different lineage (lineage III).  Lineage III is considered a related group to lineage I, and atypical isolates for 4b can be members of this group (Nightingale et al., 2005).  Therefore, it is possible that all these isolates are related, but that the five recurring 4b isolates are actually members of lineage III rather than lineage I.  Lineage III isolates are overrepresented in animal cases of listeriosis compared to human cases (Orsi et al., 2011).  One other PFGE fingerprint relating to serotype 4b was also shared between Serpentine 1 and Serpentine 2, but only occurred once at each site. Taken together, these data suggest that wild animals may be a reservoir for transporting L. monocytogenes within the Serpentine watershed, but it remains to be seen whether this poses a risk to human health.   Whole genome sequencing of these isolates in the future may provide more insight into their genetic relatedness.  2.4.4 Effect of sampling and detection methodologies At least some of the variation observed between this study and other similar studies, both in Canada and the United States, can be attributed to differences in either sampling or pathogen detection methodology.  Already discussed above is the increased recovery of VTEC without the use of pre-enrichment (Johnson et al., 2014).  It remains to be seen if this same trend will hold for other 64  pathogens, and will require the development of more sensitive assays to detect these pathogens at relatively low concentrations within the water environment.  Another major difference is in the volume of water sampled.  Including this study, sample volumes for compared studies range from 25 ml as high as 500 ml, and that does not include the use of Moore swabs which allow 24 hours of water flow to be measured.  Since these methods ultimately lead to a qualitative result, the volume of water tested will have a significant effect on the likelihood of observing the presence of a pathogen in low concentrations.  One solution to this is to convert relative recovery of pathogens into a ratio of occurrence per volume of sample.  This method was used by Edge et al. (2012) to compare the relative occurrence of pathogens in waters collected from agricultural areas to water collected from non-agricultural areas when there was a large discrepancy in the number of sampling sites within each category.  The relative occurrence per volume method, however, still does not account for the qualitative nature of pathogen detection, where two positive surface water samples may have vastly different concentrations of the pathogen of interest.  This also makes it difficult to define the level of risk.  At what sample volume does the presence of foodborne pathogens in the irrigation water become a hazard?  In the future, this question will need to be answered to not unnecessarily put too high a burden on growers as our ability to detect pathogens becomes more and more sensitive.  Therefore, consistent and sensitive pathogen detection methods should be developed and agreed upon in order to better understand the spread of pathogens world-wide, as well as to quantitatively estimate the risks associated with certain waters.  65  2.5 Conclusion The foodborne pathogens VTEC, L. monocytogenes, and Salmonella are clearly present in waters used for irrigation in the Lower Mainland of British Columbia, with L. monocytogenes being the most common, followed by VTEC and Salmonella.  Furthermore, serotypes previously shown to cause both illnesses and outbreaks were recovered, suggesting a public health risk may exist with using these waters for irrigation of fresh produce.  Recovery of pathogens was higher during the cooler months, and location specific trends were observed for their occurrence.   These observed trends suggest the potential for determining physical and environmental factors which may predict the presence of these pathogenic bacteria, allowing growers to assess water quality before irrigating their crops.   66  Chapter 3: Generic Escherichia coli and Total Fecal Coliforms as Indicator Organisms for the Presence of Verotoxigenic Escherichia coli, Salmonella, and Listeria monocytogenes in Irrigation Water  3.1 Introduction Water quality in agriculture is important for food safety, but testing for individual pathogens is both expensive and time consuming, making pathogen testing an unrealistic expectation of growers and regulators.  Instead, easily detected and enumerated indicators are often used to assess the risk of the presence of foodborne pathogens.  Since these pathogens are primarily of fecal origin, indicators are bacteria associated with the intestines of warm blooded animals such as fecal coliforms, Enterococcus, and Escherichia coli (Pachepsky et al., 2014).  Qualities that make a suitable indicator organism include easy and relatively quick detection and enumeration, a strong correlation with pathogen occurrence, and the same or slightly longer survival in the water environment than the pathogens they are used to predict (Uyttendaele et al., 2015).    In Canada, recommended water quality guidelines for good agricultural practice (GAP) suggest surface water testing three times per season, and that less than 1000 total coliforms and 100 FC per 100 ml of water should be present (CanAgPlus, 2015; Steele and Odumeru, 2004).  Some provinces in Canada have implemented their own recommendations, and the British Columbia Ministry of the Environment has set a guideline of less than 200 FC, and less than 77 E. coli in 100 ml of all water used for irrigation of produce (Government of British Columbia, n.d.).  Recent legislation in the United States under the Food Safety Modernization Act requires that all surface 67  water used for irrigation of fresh produce is based solely on levels of E. coli and requires a geometric mean of 126 CFU per 100 ml across 20 consecutive samples, with no sample exceeding 410 CFU per 100 ml.  Furthermore, any water used to irrigate spouts or used during or after harvest must contain zero E. coli, and no untreated surface water can be used for either activity (U.S. Food and Drug Administration, 2015).  Recent studies have shown that many correlations between fecal indicators and pathogen occurrence are either weak, or non-existent (Pachepsky et al., 2014), and pathogens have been observed even when indicator levels are low (Jokinen et al., 2012).  Furthermore, not all foodborne pathogens are necessarily of fecal origin (e.g., Listeria monocytogenes).  Also of interest, different methods exist for quantifying indicator organisms, but little work has been done to compare how methods may differ in their ability to predict pathogen presence.  The objective of this part of the project was therefore to investigate the suitability of two different methods of fecal coliform and E. coli enumeration to predict the occurrence of three foodborne pathogens in the irrigation waters around the Lower Mainland of British Columbia.  3.2 Materials and methods 3.2.1 Enumeration of generic E. coli and total fecal coliforms Generic E. coli and total fecal coliforms (TFC) were enumerated in each sample using two distinct methods: 1) membrane filtration followed by plating on selective differential media, and 2) 3M™ Petrifilm™ E. coli/Coliform Count Plates.  68  3.2.1.1 Membrane filtration method A 25 ml aliquot of each water sample was vacuum filtered through an 85 mm, 0.45 μm pore size GN-6 Metricel® membrane filter disc (Pall Laboratory, St. Laurent, PQ), which was subsequently transferred to m-FC agar medium (Hardy Diagnostics, Santa Maria, CA) containing 1% rosolic acid (Hach Canada Ltd., London, ON) and incubated at 44.5 ± 1°C for 18-24 hours (Myers et al., 2014).  All resulting colonies with a blue colour were counted as fecal coliforms.  The filter membranes were then transferred from the m-FC media to nutrient agar medium containing 4-methylumbelliferyl β-D-glucuronide (NA-MUG; Hardy Diagnostics) and incubated at 37°C for 4-6 hours.  Colonies producing fluorescence under long wave UV light (365 nm) were assumed to be generic E. coli (Myers et al., 2014).   3.2.1.2 Enumeration on 3M™ Petrifilm™  A 1 ml aliquot of each water sample was applied to an E. coli/Coliform Count Petrifilm™ (3M Science, London, ON) following the manufacturers directions, then subsequently incubated at 44.5 ± 1°C for 18-24 hours.  All blue and red colonies showing evidence of gas production were assumed to be fecal coliforms, and all gas-producing blue colonies E. coli.   3.2.2 Pathogen occurrence The occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes were measured as described and discussed in Chapter 2.  Only samples collected during or after June 2014 were used as that is when consistent enumeration of TFC and generic E. coli was completed using both methods.  Samples from sample site Sumas 4 were not included in the point-biserial correlations due to insufficient sampling and its close proximity to Sumas 1. 69   3.2.3 Statistical analysis All statistical analysis was conducted using R (version 3.3.2; R Core Development Team, Vienna, Austria; http://www.R-project.org).  Correlations between indicator concentrations and pathogens were calculated using the point-biserial method, which correlates a continuous variable with a dichotomous variable (Gu et al., 2013), using the ltm software package (http://www.jstatsoft.org/v17/io5/).  Regression trees were constructed using the Random Effect-Estimation Method (RE-EM) with the REEMtree package (version 0.90.3). The sample sites were set as the random variable, and the pathogen occurrence and indicator concentration were chosen as the dependent and independent variables, respectively.  This method was used since the data was longitudinal and the results from repeated sampling at the same sites were not independent (Jones et al., 2014).  Figures were produced using the ggplot2 package (http://ggplot2.org) and edited using Inkscape (version 0.91; http://inkscape.org).  3.3 Results 3.3.1 Comparison of methods and indicator concentrations Data collected using the two different enumeration methods were compared to determine if they provide similar estimates of TFC and E. coli in water samples.  The resulting slopes and correlations are shown in Figure 3.1.  A significant correlation was observed between both methods for both TFC (r = 0.60; n = 190; p < 0.001) and generic E. coli (r = 0.77; n = 176; p < 0.001); however, there was no consensus between methods on the values as the slopes deviated from the expected slope of 1.00 (slope = 0.28 and slope = 0.49 for TFC and generic E. coli, 70  respectively).  The resulting slopes of less than one indicated that higher concentrations of the indicators were observed when using the Petrifilm™ method over the membrane filtration method.     Figure 3.1 – Correlation between the membrane filtration and petrifilm methods for enumeration of total fecal coliforms (A) and generic Escherichia coli (B).  The slope of the resulting trendlines (solid line) is compared to the expected slope of 1.0 (dashed line).  The observed concentrations of both indicator bacteria were also compared to determine if the concentrations were related.  The concentrations of TFC and generic E. coli were strongly correlated when measured using the membrane filtration method (r = 0.882; n = 176; p < 0.001), but slightly less so than when measured using the Petrifilm™ method (r = 0.710; n = 221; p < 0.001).  Moreover, all individual sites which showed a significant correlation with generic E. coli by membrane filtration were also significantly correlated with TFC by membrane filtration, but not necessarily the other way around (Table 3.1).   71  Table 3.1 – Point-biserial correlation coefficients (rpb) of total fecal coliforms and generic Escherichia coli with the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and Listeria monocytogenes in water samples collected from three sites each in two watersheds of the Lower Mainland of British Columbia.  Total Fecal Coliforms Generic E. coli  Membrane Filtration 3M™ Petrifilm™ Membrane Filtration 3M™ Petrifilm™ VTEC     Sumas Watershed 0.312*** 0.084 0.309** 0.132 Sumas 1 0.573** 0.159 0.529** 0.442* Sumas 2 - - - - Sumas 3 -0.074 -0.120 -0.087 -0.063 Serp.Watershed 0.242* 0.166 0.318** 0.012 Serpentine 1 - - - - Serpentine 2 0.582*** 0.461* 0.670*** 0.425* Serpentine 3 -0.227 -0.119 -0.211 -0.077 Both Watersheds 0.273*** 0.266*** 0.320*** 0.092 Salmonella     Sumas Watershed 0.291** 0.122 0.289** 0.089 Sumas 1 0.481** 0.018 0.462* 0.175 Sumas 2 - - - - Sumas 3 - - - - Serp.Watershed 0.025 0.146 0.050 -0.027 Serpentine 1 0.089 0.225 0.174 -0.035 Serpentine 2 - - - - Serpentine 3 - - - - Both Watersheds 0.097 0.042 0.099 -0.017 L. monocytogenes     Sumas Watershed 0.169 0.049 0.144 0.118 Sumas 1 0.102 0.135 0.057 0.270 Sumas 2 0.620*** 0.041 0.616*** 0.143 Sumas 3 0.280 -0.024 0.323 0.060 Serp.Watershed 0.363*** 0.072 0.342** 0.00 Serpentine 1 0.440* 0.170 0.305 -0.059 Serpentine 2 0.213 -0.024 0.357 0.374 Serpentine 3 0.582** 0.075 0.465* 0.005 Both Watersheds 0.389*** 0.132* 0.374*** 0.061 Any pathogen     Sumas Watershed 0.297** 0.101 0.277** 0.125 Sumas 1 0.462* 0.096 0.419* 0.308 Sumas 2 0.620*** 0.041 0.616*** 0.143 Sumas 3 0.189 -0.092 0.217 0.013 Serp.Watershed 0.419*** 0.137 0.388*** -0.013 Serpentine 1 0.440* 0.170 0.305 -0.059 Serpentine 2 0.468* 0.241 0.549** 0.393* Serpentine 3 0.422* 0.011 0.323 -0.033 Both Watersheds 0.431*** 0.234*** 0.408*** 0.074 * p < 0.05; ** p < 0.01; *** p < 0.001 72  3.3.2 Predictive ability of total fecal coliforms and generic E. coli In order to determine if either method for either indicator was a reliable predictor of pathogen presence, point-biserial correlations (rpb) were calculated for each indicator/method combination with respect to each pathogen.  These correlations are summarized in Table 3.1.  The occurrence of any of the three pathogens was significantly correlated with both TFC (rpb = 0.431; n = 190; p < 0.001), and generic E. coli (rpb = 0.408; n = 176; p < 0.001) when measured using the membrane filtration method, but only with TFC (rpb = 0.234; n = 221; p < 0.001) when measured using the Petrifilm™ method; moreover, the correlation coefficient was much lower for the Petrifilm™ method.    For individual pathogens, TFC were significantly correlated with the occurrence of VTEC using both the membrane filtration method (rpb = 0.273; n = 190; p < 0.001) and by the Petrifilm™ method (rpb = 0.266; n = 221; p < 0.001).  Generic E. coli were only significantly correlated when measured using the membrane filtration method (rpb = 0.320; n = 176; p < 0.001).  The same pattern was observed for L. monocytogenes where the occurrence was significantly correlated with TFC by both the membrane filtration method (rpb = 0.389; n = 190; p < 0.001) and by Petrifilm™ (rpb = 0.132; n = 221; p = 0.049), but only significantly correlated with generic E. coli by the membrane filtration method (rpb = 0.374; n = 176; p < 0.001).  No significant correlations were observed for either indicator by either method for the occurrence of Salmonella.   Within individual watersheds, only the membrane filtration method showed any significant correlations, but these differed between the two watersheds.  Both TFC and generic E. coli were more strongly correlated with the occurrence of any of the three pathogens in the Serpentine 73  watershed (rpb = 0.419; n = 81; p < 0.001 and rpb = 0.388; n = 75; p < 0.001, respectively) compared to the Sumas watershed (rpb = 0.297; n = 109; p = 0.002 and rpb = 0.277; n = 101; p = 0.005, respectively).  For the occurrence of L. monocytogenes, significant correlations were observed in the Serpentine watershed for both TFC (rpb = 0.363; n = 81; p < 0.001) and generic E. coli (rpb = 0.342; n = 75; p = 0.003), but no significant correlations were observed in the Sumas watershed.  For the occurrence of VTEC, the correlations were more similar between the Serpentine and Sumas watersheds for both TFC (rpb = 0.242; n = 81; p = 0.028 and rpb = 0.312; n = 109; p < 0.001, respectively) and generic E. coli (rpb = 0.318; n = 75; p = 0.005 and rpb = 0.309; n = 101; p = 0.002, respectively).  Significant correlations of both TFC and generic E. coli with the occurrence of Salmonella in the Sumas watershed were observed, specifically at sample site 1 (rpb = 0.481; n = 28; p < 0.001, and rpb = 0.462; n = 26; p < 0.001, respectively), which was the only site in the Sumas watershed where Salmonella was recovered.  In order to further characterize the potential of TFC and generic E. coli by membrane filtration to predict the occurrence of these three foodborne pathogens, regression trees were produced.  The resulting splits are shown in Figure 3.2.  Across all sites tested, TFC measured by the membrane filtration method showed a single split with ≥ 7.54 CFU/ml in a sample showing an 83.3% chance of pathogen occurrence, compared to a 15.3% occurrence for samples showing < 7.54 CFU/ml TFC.  A root node only was observed for generic E. coli suggesting less predictive power across the seven sites compared to TFC.  Within the Serpentine watershed, a node split was observed for TFC with ≥ 8.23 CFU/ml showing a 100% occurrence of pathogens, and < 8.23 CFU/ml showing only a 17.5% occurrence.  Similarly, for generic E. coli in the Serpentine watershed a node split was observed with ≥ 3.59 CFU/ml showing a 75.0% chance of occurrence and < 3.59 CFU/ml 74  showing a 17.5% chance of pathogen occurrence.  No node splits were observed for either indicator in the Sumas watershed, and no nodes were observed in either or both watersheds for any individual pathogen for either indicator.     Figure 3.2 – Random effect – estimation method regression trees predicting the occurrence of veroxigenic E. coli, Salmonella, and L. monocytogenes relative to the concentrations of the fecal indicators, fecal coliforms and generic E. coli, within two watersheds in the Lower Mainland of British Columbia.  75   In order to better visualize how the occurrence of pathogens related to indicator organism trends at the six primary sampling sites, the observed levels of indicator organisms measured by membrane filtration were plotted over time along with dates where these three foodborne pathogens were recovered and are shown in Figure 3.3.  As can be seen, many sites showed a wide variation in the observed levels of both indicator organisms, but the Sumas watershed seems to have overall lower levels of these indicators when compared to the Serpentine watershed.  Sumas 2 and Sumas 3 sites, showed notably lower indicator concentrations compared to the other four sites.  These two sites also had the lowest occurrences of pathogens.  Similarly, the Serpentine watershed, which showed much higher overall concentrations of both indicator organisms had a higher occurrence of these three pathogens than Sumas 2 and Sumas 3. Also observed in Figure 3.3, the occurrence of pathogens often coincided with a spike in indicator organism concentration, resulting in the mean concentration of both fecal indicators to be, on average, higher in samples positive for the presence of a pathogen than samples where pathogens were not recovered (Wilcoxon rank sum test; p < 0.0001).  Often times, however, these pathogens also occurred when the indicators were at their lowest levels.   76    Figure 3.3 – Plot of the concentrations of total fecal coliforms (red) and generic E. coli (blue) observed for each sample site through the membrane filtration method, as well as samples that were positive for verotoxigenic E. coli (V), Salmonella (S), and L. monocytogenes (L).  Only samples eight through thirty-five are shown since the membrane filtration method was not used during the first seven samplings.   77  The mean and maximum levels of each indicator measured using the membrane filtration method were also calculated for each primary sampling site and are shown in in Table 3.2.  The concentrations of both indicators were significantly greater in the Serpentine watershed than the Sumas watershed (Wilcoxon rank sum test; p < 0.001).  Similarly, within each watershed, site-specific differences in indicators concentrations were also observed (Kruskal-Wallace test; p < 0.05).   Table 3.2 – Average total fecal coliform and generic E. coli concentrations for the three sites sampled at each watershed during this experiment. Sample Site Mean F. coliforms (CFU/100ml) Max. F. coliforms (CFU/100ml) Mean Generic E. coli (CFU/100ml) Max. Generic E. coli (CFU/100ml) Sumas 1 185 806 111 502 Sumas 2 49 246 30 154 Sumas 3 85 396 42 208 Serpentine 1 239 1108 136 630 Serpentine 2 267 970 174 700 Serpentine 3 393 926 251 720 Significantly higher mean concentrations of both indicators were observed in the Serpentine watershed than in the Sumas watershed (Wilcoxon rank sum test; p < 0.001)   3.4 Discussion The detection of foodborne pathogens in irrigation water is both time consuming and impractical due to costs.  Therefore, the quality of irrigation water in Canada, the United States, and around the world is primarily assessed using indicators such as fecal coliforms and generic E. coli (Uyttendaele et al., 2015).  Significant debate currently exists, however, on whether or not measurements based on these organisms truly predict the risk of foodborne pathogens, with a variety of studies showing both positive and negative correlations (Pachepsky et al., 2014).  The 78  purpose of this section was to investigate the ability of TFC and generic E. coli enumerations to predict risk of pathogen occurrence when measured using two different techniques.  3.4.1 Comparison of enumeration methods When comparing between the two methods, the numbers of TFC and generic E. coli observed were significantly correlated but variation was still present with r = 0.60 and r = 0.77 for TFC and generic E. coli, respectively.  This is similar to a previous study that compared E. coli concentrations from a culture method and qPCR method and found a similar correlation value (r = 0.72) (Truchado et al., 2016).  The variation between the two methods in this study may have been the result of different sampling volumes, since the 25 ml volume used for the membrane filtration, allowed for a more accurate enumeration.   Indeed, previous work comparing the Petrifilm™ and membrane filtration methods of generic E. coli enumeration has shown a stronger correlation and similarity (r > 0.9; slope = 0.9-1.0) when the volumes sampled were the same between methods (Vail et al., 2003).  The difference in volume, then, could further explain the deviation from a slope of 1.00 and intercept of 0.  At lower indicator concentrations, the small sample volume on the Petrifilm™ reduces the likelihood of observing the bacteria and errors will tend to be on the lower end since the cells in the sample are not likely to be concentrated in a random spot.  On the other hand, the larger volume of the membrane filtration method will give a more representative concentration.  This results in a shift of the trend line up the membrane filtration axis.  Since the bulk of the observations are in this area, the shift actually skews the tend line to give a reduced slope.  This is further evidenced by the fact that a greater number of observations exist on the membrane filtration 79  side of the trendline after the initial low concentration cluster (Figure 3.1).  It might also be possible that the Petrifilm™ may be more favourable to growth of injured cells and may give greater, and more accurate enumerations of TFC and generic E. coli; however, this did not lead to an improved ability to predict the occurrence of foodborne pathogens.  3.4.2 Correlation and prediction of pathogen occurrence Relative to the ability of each method to predict the occurrence of any of the three pathogens used in this study, the membrane filtration method showed a greater correlation when compared to the Petrifilm™ method.  The Petrifilm™ method was only significantly correlated on a non-site-specific basis for TFC, and at that, the correlation coefficients were reduced compared to the membrane filtration method.  Generic E. coli enumerated by Petrifilm™ significantly correlated with pathogen occurrence only on a site-specific basis.  Further, any significant correlations observed using the Petrifilm™ method were also evident with the membrane filtration method, and in all cases, the membrane filtration method produced a higher correlation coefficient.  As stated above, it is reasonable to suggest that the increased predictive power of the membrane filtration method is due to the increased volume sampled since similar results are expected when the sample volumes are consistent (Vail et al., 2003).  To the best of my knowledge, no other study has investigated the effect that sample volume has on the correlation of pathogens with microbial indicators.  For the remainder of this discussion, references to TFC and generic E. coli enumerations will be in relation to the membrane filtration method.  Total fecal coliforms showed a marginally higher correlation with total pathogen occurrence than generic E. coli, and was the only indicator able to produce a split using RE-EM tree analysis. This 80  was not necessarily the case for the occurrence of specific pathogens.  Generic E. coli was a better predictor for the occurrence of VTEC, while L. monocytogenes correlated slightly better with TFC.  Neither indicator was significantly correlated with the presence of Salmonella, but this may be the result of the overall lower occurrence of the bacterium, primarily being recovered at only one sampling site.  It makes intuitive sense that generic E. coli would be a better predictor of VTEC, a subset of the indicator species, although Gu et al. (2013) found no correlation between generic E. coli and VTEC O157 occurrence, while a weak but significant correlation was observed with TFC.   Overall, however, there was very little difference in predictive power, and a high correlation was observed between the two indicators when measured by the membrane filtration method (r = 0.88).  This correlation between the two indicators is similar to r = 0.82 (Wilkes et al., 2009) and r = 0.92 (Holvoet et al., 2014) observed in previous studies and suggests that using multiple indicators will not increase the prediction of pathogen presence.  Furthermore, TFC and generic E. coli have been found to be more robust indicators than others, such as total coliforms, Enterococcus, and Clostridium perfringens, with generic E. coli suggested to be the most predictive overall (Wilkes et al., 2009).  Correlations with individual pathogens were lower than correlations for the pathogens as a group.  No overall correlation was observed with the occurrence of Salmonella, but this is likely due the rarity of its recovery.  Indeed, a significant site-specific correlation for TFC and generic E. coli was observed for the presence of Salmonella at the site Sumas 1, where the highest recovery of the pathogen occurred.  Resutls from previous studies indicated that seasonal trends in Salmonella occurrence correlate to corresponding seasonal trends in indicator concentrations (Jokinen et al., 2010), and Wilkes et al. (2009) observed a regression tree split where 88.9% of samples with > 89 81  CFU/ml E. coli were positive for Salmonella despite only a weak correlation with the indicator (r = 0.162).  On the other hand, Benjamin et al. (2013) measured the correlation between Salmonella occurrence and sample E. coli concentration, as was done in this study, and found that there was no association between the two.  Significant correlations were observed for both indicators by membrane filtration with the occurrence of VTEC.  Site specific correlations were indeed stronger than across all sites together, and coincide mostly with Sumas 1 and Serpentine 2, which were the two sites where VTEC was most commonly recovered.  Most previous studies have only investigated the occurrence of VTEC O157 and found weak (Gu et al., 2013) or insignificant (Benjamin et al., 2013) correlations.  Interestingly, occurrences of VTEC O157 were higher (14.6% and 13.8%, respectively) than were found for all VTEC in this study.  This emphasizes that low correlation coefficients cannot necessarily be attributed to low occurrence values.  It should also be noted, however, that generic E. coli was enumerated by the most probably number method in both of these studies, and used 100 ml of sample for each enumeration.  Over the long run, however, a strong correlation (r = 0.859) was observed between the average seasonal concentration of TFC and the seasonal occurrence of VTEC O157 in the Salmon River of British Columbia (Jokinen et al., 2010).  Indicator concentration and pathogen occurrence in the Salmon River study both also correlated with total seasonal precipitation which may suggest a confounded link between the two.  Similarly, Duris et al. (2013) compared the occurrence of VTEC related genes (e.g., stx1, stx2, eaeA, etc.) in water samples that met or exceeded the recreation water quality standards in Pennsylvania, (298 CFU generic E. coli per 100 ml), and found the presence of VTEC genes was observed more often 82  in water exceeding the quality standards (χ2-test; p < 0.05). However, the genes were also found in a number of water samples that were within the recommended safe levels of the E. coli indicator.   A significant correlation between the occurrence of L. monocytogenes and both indicator organisms when measured by the membrane filtration method was observed.  This is inconsistent with previous studies that have found no relationship between this indicators and the pathogen (Arvanitidou et al., 1997; Wilkes et al., 2009).  Stronger correlations were found to exist at only certain sites, and while these site-specific correlations for the two other pathogens coincided with sites that had relatively high occurrences for the pathogen, this was not the same for L. monocytogenes.  Indeed, no correlation was observed between either indicator organism and L. monocytogenes presence for Serpentine 2, which had the highest occurrence of the bacterium over the course of this study, and the highest correlation was observed for Sumas 2, where only one occurrence of L. monocytogenes was observed, but happened to coincide with a spike in indicator concentrations (Figure 3.3).  One possibility is that these observations are driven by chance and in the long run will not continue to show a correlation.  The other possibility is that the source of L. monocytogenes occurrence differs between the sites.  L. monocytogenes is commonly associated with fecal shedding by wild life and livestock (Lyautey et al., 2007a), but can also be present and survive in environmental sources such as soil (Vivant et al., 2013).  The significant correlations observed at individual sites may point to a fecal source of contamination, and more sporadic recoveries, compared to a site with high occurrence but low correlation which suggests an environmental “point-source” reservoir which provides more consistent contamination of the water.   83  While significant correlations were observed for pathogen occurrence, recoveries of pathogens often occurred in water samples with low levels of indicators.  Indeed, just under 50% of pathogen occurrence occurred when fecal indicators were near their lowest levels (Figure 3.3).  Interestingly, though, the highest peaks for each site always correlated with the occurrence of at least one pathogen, which is likely the driver of any observed significant correlation, and leading to higher average concentrations of fecal indicators when pathogens are present.  Previous studies have shown higher mean concentrations of indicators when pathogens are present, but not necessarily in all cases of pathogen occurrence (Duris et al., 2013; Jokinen et al., 2012).  Furthermore, higher indicator concentrations have been previously observed in agricultural areas than in water not affected by agriculture, which also correlated with the presence of pathogens (Edge et al., 2012).  Similarly, higher indicator concentrations were observed in the Serpentine watershed, which correlated with overall increased pathogen recovery.  Counter-intuitively, however, the highest indicator occurrence in the Serpentine watershed was observed at site 3, which also showed the lowest pathogen recovery within the watershed.  Within the Sumas watershed, which showed a lower average level of indicators and in turn a lower recovery of pathogens, higher indicator concentrations were observed at site 1, which coincided with higher overall pathogen recovery.  These results show that high levels of indicator organisms over the long-run may signal an increased risk for pathogen presence, but low levels of indicators do not necessarily signal an absence of foodborne pathogens.  Further, it emphasizes that long term trends may be more predictive than single samplings when it comes to pathogen prediction in surface waters for agricultural use.   84  Despite low correlations between pathogen occurrence and indicator organism concentrations, RE-EM tree analysis emphasized that a there is a significantly increased risk of pathogen occurrence when indicator organisms are in higher concentrations (Figure 3.2).  These concentrations, however, are much higher than current guidelines for water quality analysis.  In British Columbia, recommended water quality assessment guidelines suggest that irrigation waters should contain TFC and generic E. coli concentrations below 100 CFU/100 ml and 77 CFU/100 ml, respectively (Government of British Columbia, n.d.).  Of interest is that only two of the six primary sites sampled during this study meet that criteria.  Furthermore, all the sites were above these criteria on multiple single samplings.  This may add a complication if mandated regulations are ever put in place with the same criteria.  Alternatively, the Health Canada regulations for recreational water mandate less than 200 E. coli per 100 ml geometric mean over at least five samplings, and no sample showing a concentration greater than 400 E. coli per 100 ml.  By these regulations, all but one sample site (Serpentine 3) fit the criteria for mean occurrence; however, four sites showed at least one sample above the maximum threshold.  While these criteria are not currently mandated for agricultural waters, the Food Safety Modernization Act (FSMA) in the United States has mandated criteria that will need to be met for any growers hoping to ship their product to our southern neighbour.  The FSMA regulations (U.S. Food and Drug Administration, 2015) require a geometric mean of E. coli of less than 126 CFU/100 ml, which was not met by any of the three Serpentine sample sites, and no single sample to have greater than 410 CFU of E. coli per 100 ml, which was, again, not met by any site in the Serpentine watershed or one site in the Sumas watershed.    85  3.5 Conclusion Total fecal coliforms and generic E. coli show some correlation with pathogen occurrence in the irrigation waters of the Lower Mainland of British Columbia, but they do not tell a complete story.  Higher concentrations of indicators at individual sites were associated with increased presence of foodborne pathogens at those particular sites, but pathogens were also recovered at sites with lower average indicator concentrations.  Moreover, indicator concentrations in individual samples provided little insight into the likelihood of pathogens being present in those samples.  Additionally, the method of indicator enumeration may have an effect on how predictive indicator organisms may be.  Using a membrane filtration method of enumeration with a higher volume of sample was more predictive of pathogen occurrence than using a smaller volume with 3M™ Petrifilm™.    Finally, many of the sites sampled for this study were found to be outside the recommended limits for water quality indicators suggested by CanadaGAP™ and mandated by FSMA.    86  Chapter 4 – Environmental Factors Associated with the Occurrence of Verotoxigenic Escherichia coli, Salmonella, and Listeria monocytogenes  4.1 Introduction Much research has demonstrated that the occurrence of foodborne pathogens in surface waters is not uniform across temporal and spatial lines, and that the presence of pathogens can be affected by season (Gu et al., 2013; Haley et al., 2009; Jokinen et al., 2012; Strawn et al., 2013a), location (Chapin et al., 2014; Lyautey et al., 2007b), and weather (Gu et al., 2013; Jokinen et al., 2012; Jones et al., 2014).  An understanding of how these factors may relate to, and therefore predict, the presence of these pathogens in the waters used for irrigation may allow for a reduction in the risk of contamination of fresh produce by applying control strategies at the right times and places.  Therefore, the objective of this section is to determine any physicochemical or environmental factors which may correlate and/or predict the presence of verotoxigenic Escherichia coli (VTEC), Salmonella, or Listeria monocytogenes in the irrigation waters of the Lower Mainland of British Columbia.  4.2 Materials and methods 4.2.1 Water temperature, pH, and total dissolved solids The temperature of the water samples was measured at the time of sampling using an alcohol thermometer.  The thermometer was inserted into the sample within one minute of collection, and allowed to equilibrate for at least 20 seconds before reading.  The pH of the samples was measured 87  within 36 hours after collection, at the laboratory in duplicate using an Accumet digital pH meter (Fisher Scientific, Ottawa, ON).   Total dissolved solids (TDS) were measured for each sample by oven drying.  Briefly, foil weigh boats were dried over night at 120°C before pre-weighing.  A 25 ml aliquot of each water sample was added to a pre-dried and pre-weighed weigh boat in duplicate after the water samples were allowed to settle for > 2 hours.  The aliquots were dried overnight (~16 hours) at 120°C and allowed to cool at room temperature in a desiccator before determining the final weight of the remaining TDS.   4.2.2 Collection of weather data Weather data was collected from the Environment Canada website for the sampling areas on each sampling date.  Data for the Serpentine watershed sampling sites were collected from the Pitt Meadows CS weather station (49°12'29.964" N; 122°41'24.076" W), and data for the Sumas watershed sampling sites were collected from the Sumas Canal weather station (49°06'48.008" N; 122°06'35.004" W) with missing data being filled in from the Mission West Abbey weather station (49°09'09.002" N; 122°16'14.001" W).  The weather data collected included total precipitation and average temperature on the date of sampling as well as on each of the three days prior to sampling.  4.2.3 Collection of geographical data Water flow direction at the sample sites was determined by visual inspection during each sampling. The upstream water sources were determined using the Drainage Mains and Drainage Open Channels datasets collected from the City of Surrey’s Open Data Site (http://data.surrey.ca), the 88  City of Abbotsford Map Viewer (maps.abbotsford.ca), and the City of Chilliwack Webmap (maps.chilliwack.com).  Livestock information was collected from the Agricultural Land Use Inventories (ALUI) for Surrey (2010), Abbotsford (2012), and Chilliwack (2012), retrieved from the Government of British Columbia (http://www2.gov.bc.ca/gov/content/industry/agriculture-seafood/agricultural-land-and-environment/strengthening-farming/planning-for-agriculture/agricultural-land-use-inventories/south-coast).    4.2.4 Analysis of geographical data Maps were produced based on the collected geographic data (section 4.2.3) using ArcMap (ArcGIS, version 10.2.2; http://www.arcgis.com).  Upstream water sources were determined up to three kilometers from each sample site.  Upstream water sources were considered to be any waterway directly connected to the sample site where the water was determined, either from retrieved geographic data or direct observation, to flow to the sample site.  If livestock was present on a property on the ALUI, that whole property was considered to be positive for the presence of that livestock type.  Livestock was considered to be cow (i.e., dairy or beef), poultry, or other (i.e., swine, sheep, goat). Livestock were considered to be upstream if the property directly bordered on to the connecting waterway.  Livestock properties separated from the waterway by a road were not considered to be bordering on the waterway.  Using the measurement tool in ArcMap, the upstream distance to the nearest livestock of each type was measured for each livestock type.  Also measured were the number of bordering livestock properties 1 km, 2 km, and 3 km upstream, and the total length of livestock associated property bordering the upstream waterways up to a distance of 1 km, 2 km, and 3 km.  Any lengths of 89  upstream waterway that passed through a livestock property were counted twice to account for bordering on two banks of the waterway.   4.2.5 Pathogen occurrence The occurrence of VTEC, Salmonella, and L. monocytogenes were measured and are shown and discussed above in Chapter 2.  Samples from sample site Sumas 4 were not included due to insufficient sampling and its close proximity to Sumas 1.  4.2.6 Statistical analysis Statistical analysis of results was conducted using R version 3.2.3 (R Core Team, 2015; http://www.R-project.org).  Fisher’s LSD tests were conducted using the agricolae software package (version 1.2-4; https://CRAN.R-projects.org/package=agricolae). Correlations between weather or water characteristics and pathogen occurrence were calculated using the point-biserial method, which correlates a continuous variable with a dichotomous variable (Gu et al., 2013), using the ltm software package (http://www.jstatsoft.org/v17/io5/).  Correlations between relative pathogen occurrence at each site (i.e., number of positive samples vs. total number of samples collected at that site) and livestock proximity and density were calculated using Spearman’s rank correlation test.  4.3 Results 4.3.1 Water temperature, pH, and total dissolved solids Water temperature, pH, and TDS were measured for each site and are summarized in Table 4.1.  No significant differences were observed in the temperature of the water between sites over the 90  course of sampling, but a significantly lower pH was observed in the Serpentine watershed compared to the Sumas watershed, specifically at Serpentine 2 (Fisher’s LSD; p < 0.05).  In addition, the mean pH at all sites was alkaline, but acidic pH was occasionally observed at Serpentine 2.  A greater amount of TDS was also observed at Serpentine 2 compared to the other five sample sites tested (Fisher’s LSD; p < 0.05).  Table 4.1 – Water temperature, pH, and total dissolved solids (TDS) observed at the six primary sampling sites monitored during this study Site Temperature (°C) pH TDS (mg/ml) Sumas 1 14.3 ± 5.4 7.64 ± 0.32a 0.254 ± 0.224a Sumas 2 15.3 ± 5.8 7.59 ± 0.28a 0.212 ± 0.097a Sumas 3 15.5 ± 6.4 7.72 ± 0.30a 0.276 ± 0.126a Serpentine 1 12.7 ± 4.4 7.35 ± 0.43b 0.205 ± 0.095a Serpentine 2 14.5 ± 5.5 7.11 ± 0.83c 0.382 ± 0.232b Serpentine 3 13.9 ± 4.8 7.56 ± 0.33ab 0.219 ± 0.283a Values in the same column with different superscripts indicated a significant difference (Fisher’s LSD; p < 0.05)  Correlations between pathogen occurrence and water characteristics were measured using the point-biserial correlation method; the resulting correlation coefficients are shown in Table 4.2.  Water temperature significantly correlated negatively with pathogen occurrence (rpb = -0.217; n = 216; p = 0.001), as did the pH of the water samples (rpb = -0.274; n = 214; p < 0.001), implying that pathogen occurrence was greater when the water was cooler and closer to a neutral pH.  Neither of these correlations are particularly strong, however.  No correlation was observed between the total dissolved solids and pathogen occurrence.    91  Table 4.2 – Point-biserial correlations (rpb) of the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes with physicochemical characteristics of the irrigation water samples collected from three sites each from the Sumas and Serpentine watersheds in the Lower Mainland of British Columbia  Temperature pH Total Dissolved Solids VTEC    Sumas Watershed -0.152 -0.031 -0.147 Sumas 1 -0.370* -0.158 -0.238 Sumas 2 - - - Sumas 3 0.041 0.085 -0.116 Serp.Watershed -0.062 -0.113 0.046 Serpentine 1 - - - Serpentine 2 -0.219 -0.062 -0.063 Serpentine 3 0.126 -0.025 -0.056 Both Watersheds -0.115 -0.104 -0.020 Salmonella    Sumas Watershed -0.093 -0.093 -0.106 Sumas 1 -0.347* -0.225 -0.153 Sumas 2 - - - Sumas 3 - - - Serp.Watershed 0.001 0.035 -0.037 Serpentine 1 0.046 0.083 -0.022 Serpentine 2 - - - Serpentine 3 - - - Both Watersheds -0.059 -0.005 -0.080 L. monocytogenes    Sumas Watershed -0.081 -0.056 -0.023 Sumas 1 -0.224 -0.114 -0.147 Sumas 2 -0.200 -0.148 -0.134 Sumas 3 -0.029 -0.108 -0.006 Serp.Watershed -0.323** -0.341*** 0.027 Serpentine 1 -0.183 0.108 -0.029 Serpentine 2 -0.510** -0.569*** 0.034 Serpentine 3 -0.209 -0.200 -0.090 Both Watersheds -0.214** -0.316*** 0.033 Any pathogen    Sumas Watershed -0.148 -0.062 -0.112 Sumas 1 -0.493** -0.184 -0.247 Sumas 2 -0.200 -0.148 -0.134 Sumas 3 -0.007 -0.057 -0.061 Serp.Watershed -0.279** -0.313** 0.062 Serpentine 1 -0.183 0.108 -0.029 Serpentine 2 -0.486** -0.471** 0.068 Serpentine 3 -0.129 -0.195 -0.110 Both Watersheds -0.217** -0.274*** 0.011 * p < 0.05; ** p < 0.01; *** p < 0.001    92  In relation to correlations with individual pathogens, both water temperature and pH only correlated significantly with the occurrence of L. monocytogenes (rpb = -0.214; n = 216; p = 0.002 and rpb = -0.316; n = 214; p < 0.001, respectively).  These correlations were primarily observed within the Serpentine watershed, specifically at sample site 2.  This is also the site where the highest number of samples positive for L. monocytogenes were recovered.  Water temperature was significantly negatively correlated with VTEC and Salmonella, but only on a site-specific basis at the Sumas 1 sample site in the Sumas watershed, with rpb = -0.370 (n = 34; p = 0.029) and rpb = -0.347 (n = 34; p = 0.041) for VTEC and Salmonella, respectively.  4.3.2 Correlations with weather factors 4.3.2.1 Precipitation Correlations between pathogen occurrence and levels of precipitation were measured using the point-biserial correlation method with the resulting correlation coefficients are shown in Table 4.3.  Pathogen occurrence was most strongly correlated with higher levels of precipitation the day before sampling (rpb = 0.203; n = 223; p = 0.002).  Less strongly correlated were precipitation levels three days before sampling (rpb = 0.175; n = 216; p = 0.010), and the total amount of precipitation over the three days prior to sample collection (rpb = 0.188; n = 223; p = 0.005).  At a site-specific level, correlations of pathogen occurrence with total precipitation the day prior to sample collection were only observed in the Sumas watershed (rpb = 0.352; n = 129; p < 0.001), with all three sample sites within that watershed showing significant correlations.  The Serpentine watershed did not show any watershed specific correlation between pathogen occurrence and precipitation one day prior to sample collection, but did show a correlation between pathogen occurrence and precipitation three days prior to sample collection (rpb = 0.257; n = 91; p = 0.013). 93  Table 4.3 – Point-biserial correlations (rpb) of the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes in the Sumas and Serpentine watersheds with the levels of precipitation three days prior (P3), two days prior (P2), one day prior (P1), the day of (P0), and the total volume over the three days prior (P1-3) to sample collection.    P3 P2 P1 P0 P1-3 VTEC      Sumas Watershed 0.041 0.016 0.266** -0.053 0.144 Sumas 1 0.168 0.108 0.613*** -0.070 0.391* Sumas 2 - - - - - Sumas 3 -0.116 -0.124 -0.098 -0.095 -0.129 Serp.Watershed 0.160 0.039 0.175 -0.046 0.156 Serpentine 1 - - - - - Serpentine 2 0.374* 0.132 0.394* -0.037 0.379* Serpentine 3 -0.100 -0.089 -0.088 -0.082 -0.126 Both Watersheds 0.093 0.024 0.205** -0.053 0.135 Salmonella      Sumas Watershed 0.096 0.145 0.072 -0.049 0.121 Sumas 1 0.119 0.286 0.137 -0.096 0.212 Sumas 2 - - - - - Sumas 3 - - - - - Serp.Watershed 0.318** 0.057 0.010 -0.045 0.176 Serpentine 1 0.550 0.095 0.015 -0.082 0.303 Serpentine 2 - - - - - Serpentine 3 - - - - - Both Watersheds 0.158* 0.114 0.075 -0.039 0.142* L. monocytogenes      Sumas Watershed 0.096 0.015 0.300*** -0.025 0.178* Sumas 1 -0.114 -0.127 0.016 -0.097 -0.075 Sumas 2 0.271 0.192 0.396* -0.022 0.348* Sumas 3 0.168 0.022 0.509** 0.005 0.306 Serp.Watershed 0.236* 0.130 0.158 0.087 0.222* Serpentine 1 0.261 -0.088 -0.036 0.387* 0.051 Serpentine 2 0.247 0.464** 0.309 -0.027 0.475** Serpentine 3 0.210 -0.066 0.187 -0.126 0.087 Both Watersheds 0.147* 0.066 0.147* 0.008 0.141* Any pathogen      Sumas Watershed 0.129 0.102 0.352*** -0.063 0.247** Sumas 1 0.110 0.221 0.452** -0.120 0.337* Sumas 2 0.271 0.192 0.396* -0.022 0.348* Sumas 3 0.097 -0.039 0.417* -0.040 0.217 Serp.Watershed 0.257* 0.106 0.116 0.048 0.204 Serpentine 1 0.261 -0.088 -0.036 0.387* 0.051 Serpentine 2 0.364* 0.427* 0.238 -0.076 0.481** Serpentine 3 0.133 -0.103 0.129 -0.155 0.018 Both Watersheds 0.175** 0.090 0.203** -0.028 0.188** * p < 0.05; ** p < 0.01; *** p < 0.001   94  With respect to individual pathogens, the amount of precipitation on the day before sample collection significantly correlated with the occurrence of VTEC (rpb = 0.205; n = 223; p = 0.002) and L. monocytogenes (rpb = 0.147; n = 223; p = 0.028).  On the other hand, Salmonella occurrence was most strongly correlated with precipitation three days prior to sample collection (rpb = 0.158; n = 216; p = 0.020).  For the precipitation level the day prior to sampling, the strongest correlations with respect to VTEC were at the Sumas 1 sampling site (rpb = 0.613; n= 35; p < 0.001), and the Serpentine 2 sampling site (rpb = 0.394; n = 34; p = 0.019), which were the two sites with the highest occurrence of the pathogen.  For L. monocytogenes, higher precipitation the day before sampling was strongly correlated for Sumas 3 (rpb = 0.509; n = 34; p = 0.002), which was the most common site of L. monocytogenes recovery in the Sumas watershed.  Alternatively, this was not observed for any sites in the Serpentine watershed where L. monocytogenes was more common.  Interestingly, the strongest correlation in the Serpentine watershed for L. monocytogenes was at site 2 and was with the total level of precipitation two days prior to sampling (rpb = 0.464; n = 34; p = 0.005) and total precipitation over the three days prior to sample collection (rpb = 0.475; n = 33; p = 0.004).  4.3.2.2 Air temperature Correlations between pathogen occurrence and mean temperature were measured using the point-biserial correlation method, with the resulting correlation coefficients shown in Table 4.4.  Pathogen occurrence significantly correlated negatively with the mean air temperature on the day of sampling (rpb = -0.155; n = 217; p = 0.022), implying increased pathogen prevalence on cooler days.  The correlation was only weak, however, and could really only be seen for Sumas 1 (rpb = -0.384; n = 35; p = 0.021) and for Serpentine 2 (rpb = -0.376; n = 32; p = 0.031), but not for any of 95  the other sampling sites.  Additionally, Sumas 1 was significantly correlated with lower temperatures on any of the three days prior to sampling, and also with the average temperature of those three days.  Table 4.4 – Point-biserial correlations (rpb) of the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes in the Sumas and Serpentine watersheds with average temperatures three days prior (T3), two days prior (T2), one day prior (T1), the day of (T0), and the average temperature over the three days prior (Tavg) to sample collection.    T3 T2 T1 T0 Tavg VTEC      Sumas Watershed -0.085 -0.115 -0.125 -0.087 -0.110 Sumas 1 -0.274 -0.270 -0.297 -0.248 -0.286 Sumas 2 - - - - - Sumas 3 0.150 0.012 0.010 0.083 0.060 Serp.Watershed 0.055 0.005 -0.022 -0.077 0.015 Serpentine 1 - - - - - Serpentine 2 -0.048 -0.092 -0.113 -0.207 -0.087 Serpentine 3 0.355 0.253 0.172 0.135 0.276 Both Watersheds -0.019 -0.059 -0.073 -0.081 -0.051 Salmonella      Sumas Watershed -0.028 -0.027 -0.043 -0.085 -0.033 Sumas 1 -0.299 -0.267 -0.224 -0.307 -0.270 Sumas 2 - - - - - Sumas 3 - - - - - Serp.Watershed 0.039 0.034 0.001 -0.001 0.026 Serpentine 1 0.064 0.055 -0.002 -0.006 0.042 Serpentine 2 - - - - - Serpentine 3 - - - - - Both Watersheds -0.007 -0.012 -0.032 -0.061 -0.017 L. monocytogenes      Sumas Watershed -0.045 -0.061 -0.082 -0.115 -0.063 Sumas 1 -0.232 -0.203 -0.217 -0.267 -0.222 Sumas 2 -0.170 -0.138 -0.122 -0.121 -0.147 Sumas 3 0.088 0.015 -0.040 -0.090 0.024 Serp.Watershed -0.127 -0.151 -0.191 -0.222* -0.162 Serpentine 1 0.098 0.033 -0.053 -0.052 0.030 Serpentine 2 -0.321 -0.297 -0.290 -0.386* -0.314 Serpentine 3 -0.128 -0.147 -0.210 -0.197 -0.165 Both Watersheds -0.092 -0.098 -0.128 -0.157* -0.108 Any pathogen      Sumas Watershed -0.043 -0.082 -0.112 -0.130 -0.079 Sumas 1 -0.381* -0.367* -0.364* -0.384* -0.379* Sumas 2 -0.170 -0.138 -0.122 -0.121 -0.147 Sumas 3 0.152 0.019 -0.032 -0.042 0.050 Serp.Watershed -0.055 -0.114 -0.169 -0.198 -0.117 Serpentine 1 0.098 0.033 -0.053 -0.052 0.030 Serpentine 2 -0.258 -0.286 -0.300 -0.376* -0.292 Serpentine 3 0.075 -0.012 -0.109 -0.103 -0.017 Both Watersheds -0.052 -0.091 -0.132 -0.155* -0.093 * p < 0.05; ** p < 0.01; *** p < 0.001   96  With respect to individual pathogens, only L. monocytogenes was correlated with air temperature, and only on the day of sample collection (rpb = -0.157; n = 217; p = 0.020), but this was only evident at the Serpentine 2 sample site (rpb = -0.386; n = 32; p = 0.027).    4.3.3 Correlations with proximity and density of upstream livestock Correlations between pathogen occurrence and nearby upstream livestock were calculated using the Spearman’s rank correlation test (rs) and the results are summarized in Table 4.5.  Pathogen occurrence correlated significantly with proximity to the nearest upstream cow or poultry farm (rs = -0.886; n = 6; p = 0.033), showing that sample sites with close proximity to livestock were more likely to be positive for the pathogens of interest.  Pathogen occurrence also correlated significantly with the number of upstream cow farms (rs = 0.828; n = 6; p = 0.042) as well as the total length of cow farms bordering upstream waterways to a distance of 1 km (rs = 0.845; n = 6; p = 0.034).  Interestingly, stronger correlations were observed when poultry farms were also included, (rs = 0.878; n = 6; p = 0.021 and rs = 0.941; n = 6; p = 0.005 for number and length, respectively), despite little to no correlation observed with poultry itself.   No correlations with total pathogen occurrence were observed at the 2 km and 3 km distances for either number or length of bordering livestock.    97  Table 4.5 – Spearman rank correlation coefficients (rs) for the occurrence of verotoxigenic E. coli (VTEC), Salmonella, and L. monocytogenes with proximity and density of upstream livestock.  VTEC Salmonella L. monocytogenes Any Pathogen Nearest LivestockA     CowB -0.377 -0.135 -0.714 -0.829 PoultryC -0.464 -0.507 0.200 -0.429 Cow or Poultry -0.696 -0.439 -0.371 -0.886* AnyD -0.493 -0.034 -0.200 -0.371 Livestock Number (1 km)E     Cow 0.105 0.122 0.828* 0.828* Poultry 0.399 0.775 -0.655 0.131 Cow or Poultry 0.396 0.693 0.293 0.878* Any 0.315 0.490 0.207 0.621 Livestock Number (2 km)F     Cow 0.626 -0.146 0.123 0.278 Poultry 0.626 0.237 -0.185 0.339 Cow or Poultry 0.721 0.017 -0.058 0.319 Any 0.672 -0.104 0.000 0.265 Livestock Number (3 km)     Cow 0.579 0.018 -0.247 0.093 Poultry 0.585 0.359 -0.334 0.273 Cow or Poultry 0.627 0.261 -0.383 0.177 Any 0.662 0.257 -0.348 0.203 Livestock Border (1 km)     Cow 0.257 0.020 0.845* 0.845* Poultry 0.399 0.775 -0.655 0.131 Cow or Poultry 0.431 0.395 0.577 0.941** Any 0.397 -0.034 0.551 0.609 Livestock Border (2 km)     Cow 0.609 -0.439 0.771 0.543 Poultry 0.308 0.359 -0.577 -0.030 Cow or Poultry 0.812* -0.372 0.543 0.486 Any 0.812* -0.372 0.543 0.486 Livestock Border (3 km)     Cow 0.377 -0.778 0.200 -0.200 Poultry 0.585 0.359 -0.334 0.273 Cow or Poultry 0.812* -0.304 0.143 0.257 Any 0.638 -0.304 -0.029 0.029 * p < 0.05; ** p < 0.01; *** p < 0.001  ANearest distance for surface water to travel BIncludes both dairy and beef cattle CIncludes chicken and turkey DIncludes bovine and poultry, as well as swine, sheep, or goats ENumber of properties bordering the surface water within a given distance upstream FTotal length of waterway bordered by properties containing livestock within a given distance     98  Looking at each pathogen individually, the occurrence of VTEC correlated significantly with the length of upstream water way that bordered on cow or poultry farms to a distance of 2 km (rs = 0.812; n = 6; p = 0.05) and 3 km (rs = 0.812; n=6; p = 0.05); however, no significant correlation was observed at the 1 km range.  Similar to the occurrence of any pathogen, the occurrence of L. monocytogenes was correlated with both the number (rs = 0.828; n=6; p = 0.042) and border length (rs = 0.845; n=6; p = 0.034) of upstream cow farms to a distance of 1 km, but unlike with the occurrence of any of the three pathogens, no significant correlation was observed when poultry was included.  No significant correlations were observed for the occurrence of Salmonella with nearby livestock, but some evidence was observed of a correlation with both the number and border length of poultry farms up to a distance of 1 km (rs = 0.775; n=6; p = 0.07 for both).  4.4 Discussion The presence of foodborne pathogens in irrigation water has been suggested to be nonuniform across environmental variables.  In this part of the project, I looked at how physicochemical, meteorological, and geospatial factors may correlate with, and potentially predict the presence of VTEC, Salmonella, and L. monocytogenes.    4.4.1 Geographical characteristics The strongest correlations were observed with proximity to and the density of upstream livestock.  The occurrence of any of the three pathogens correlated strongly with the proximity to the nearest upstream cow or poultry source, as well as the number and total length of cow and poultry sources bordering on the upstream connected water ways to a distance of 1 km.  This is consistent with a previous Canadian survey of foodborne pathogens in surface waters which found that the 99  occurrence of pathogens was higher in agricultural areas compared to control sites not affected by upstream agricultural practices (Edge et al., 2012).  Since these pathogens are primarily associated with agricultural livestock reservoirs (Nightingale et al., 2004; Persad and Lejeune, 2014; Rodriguez et al., 2006), it follows that their occurrence would be correlated with nearby proximity to these sources.  No correlations with total pathogen occurrence were observed outside of a 1 km radius, suggesting that nearby farms have the largest impact on the occurrence of these pathogens in surface waters.  It should be noted, however, that livestock associated pathogens (i.e., VTEC O157, Salmonella, and Campylobacter) can still be isolated from water sources not affected by livestock agriculture, reminding us that these are not the only sources of these bacteria (Edge et al., 2012).  Conversely, significant correlations were observed with upstream cow or poultry density and VTEC up to 2 km and 3 km, but not at 1 km. This is possibly due to the fact that Sumas 1, which accounted for a large proportion of VTEC occurrence (36%) does not have any livestock within 1 km upstream, but has a relatively large density after a distance of 1 km (Figure 4.1).  Interestingly, the significant correlation was only observed for cow and poultry together, and not with cow individually, despite the fact that cattle is considered the primary reservoir (Hussein and Bollinger, 2005; Hussein and Sakuma, 2005), with minimal occurrence of VTEC associated with domestic poultry (Persad and Lejeune, 2014).  The driving force for the inclusion of poultry in the correlation is likely Sumas 1, which accounted for 36% of recovered VTEC isolates but is surrounded primarily by poultry farms (Figure 4.1).  Serpentine 2, on the other hand, accounted for 45% of VTEC isolates and is surrounded primarily by cow farms (Figure 4.2).  Therefore, it is likely that a high level of VTEC at Sumas 1 may be coming from a small number of upstream cow farms.  It 100  should be also noted that a significantly larger amount of water flow was present at Sumas 1 compared to some other sites which could also be responsible for carrying VTEC from locations further upstream.  For instance, a previous study in California found that VTEC O157 was only carried downstream 135 meters in a low flow creek, but at times of high rains, flooding, and high water flow, transport as far as 32 km was observed (Cooley et al., 2007).   Figure 4.1 – Map of the sampling area within the Sumas watershed.  Pie charts represent the sample sites, with their size being proportional to the total number of pathogens recovered and their divisions representing the individual pathogens.  Coloured polygons represent properties where livestock are present.  Upstream water flow is shown to a distance of 3 km from each sample site.    101   Figure 4.2 – Map of the sampling area within the Serpentine watershed.  Pie charts represent the sample sites, with their size being proportional to the total number of pathogens recovered and their divisions representing the individual pathogens.  Coloured polygons represent properties where livestock are present.  Upstream water flow is shown to a distance of 3 km from each sample site.       The occurrence of L. monocytogenes correlated significantly with the density of cow farmland within 1 km upstream of sampling.  Domestic cattle are known to be a common carrier of L. monocytogenes (Lyautey et al., 2007a; Nightingale et al., 2004), and previous studies have found the occurrence of L. monocytogenes to correlate with proximity to pasture (Chapin et al., 2014) and upstream dairy farms (Lyautey et al., 2007b).  Not all sources of L. monocytogenes are related to livestock, however.  The bacterium is routinely isolated and shown to survive in soils (Vivant 102  et al., 2013), and uncultivated fields and meadows have shown a higher prevalence than agricultural land (Dowe et al., 1997).  Weaker correlation were observed between the occurrence of Salmonella and poultry farms, which reaffirms the understanding of poultry to be a primary reservoir for this pathogen (Park et al., 2008).  It should be noted, though, that three quarters of Salmonella recoveries were made at Sumas 1 which has a high concentration of upstream poultry facilities (Figure 4.1).  The other recovery of this pathogen occurred at Serpentine 1, a site with no presence of any upstream poultry, but rather was downstream of a cow farm, which is a reminder that cows can also be a reservoir for Salmonella (Rodriguez et al., 2006), as well as the potential for it to be carried by wild birds (Andrés et al., 2013).    Taken together, these results suggest that the proximity to a livestock farm, a common source of these foodborne pathogens, increases the risk of foodborne pathogens being present in the irrigation water.    103   4.4.2 Weather patterns The occurrence of any of the three pathogens was significantly correlated with precipitation on the day prior, and over the three days leading up to sample collection.  This is consistent with previous studies that have found pathogen recovery to be higher after precipitation events (Gu et al., 2013; Haley et al., 2009; Jokinen et al., 2012; Jones et al., 2014; Nadya et al., 2016).  The effect of rainfall is suggested to increase the transport of pathogens into surface waters (O’Shea and Field, 1992), and carry them longer distances downstream (Cooley et al., 2007); though it has been suggested that pathogens surviving in sediment reservoirs may also be released due to heavy rainfall (Cooley et al., 2007).  In this study, the correlation with precipitation was specifically observed within the Sumas watershed: Sumas 1 showed strong correlation between VTEC occurrence and precipitation the day before sample collection, and Sumas 2 and Sumas 3 showed a strong correlation between L. monocytogenes occurrence and precipitation before sample collection. In the Serpentine watershed, Site 2 showed significant correlation of both VTEC and L. monocytogenes with the total precipitation over the three days prior to sample collection.   These pathogen specific correlations coincide with the pathogens most commonly recovered from their respective sample sites, implying that if an upstream reservoir exists (e.g., shedding livestock), then precipitation may contribute to contamination.     Pathogen occurrence was also significantly correlated with lower average temperatures; though this observation was mostly site specific for Sumas 1 and Serpentine 2.  It is difficult to determine 104  if these observations were actually affected by temperature or if they were related to other seasonal factors (e.g., higher precipitation during the fall and winter), since, as discussed in chapter 2, seasonal variation in pathogen occurrence was also observed.  Other studies have also pointed to seasonal and temperature related differences in recovery, but as higher occurrence of L. monocytogenes in the agricultural environment has repeatedly been observed during cooler periods (Cooley et al., 2014; Strawn et al., 2013a), the summer season and higher temperatures have been previously associated with increases in VTEC O157 (Gu et al., 2013) and Salmonella (Haley et al., 2009; Jokinen et al., 2012) occurrence from surface waters.  This is of interest since temperature (Table 4.3) and seasonal differences (Chapter 2) were observed only for L. monocytogenes when reviewed on a pathogen specific level.  It is possible that trends for the other two pathogens were not observed do to lower recovery for VTEC and Salmonella of only 4.9% and 2.7%, respectively compared to 11.2% for L. monocytogenes.  The correlation observed for L. monocytogenes with temperature may also be confounded by precipitation since there was a significant correlation between the two factors (r = 0.373; p < 0.001), and precipitation was found to be highest during the fall and winter compared to the Spring and Summer (Fisher’s LSD; p < 0.05).  Another seasonally confounding variable, as previously mentioned in Chapter 2, is that during the summer growing season, dams are put in place to fill up the irrigation ditches, thereby reducing the water flow.  As water flow has previously been suggested to be an important method for downstream dissemination (Cooley et al., 2007; O’Shea and Field, 1992), this fact cannot be discounted.  Overall, no definitive conclusions can be made about the relationship between precipitation and temperature, but as a logical mechanism for pathogen occurrence exists with precipitation, I would 105  suggest that rainfall is the dominant factor in relation to weather patterns.  The potential for increased pathogen prevalence during colder temperatures cannot be discounted, but further study will be needed to tease these two factors apart in the future.  4.4.3 Water characteristics  The occurrence of any of the three pathogens was correlated with lower water temperatures at the time of sampling; however, there is no reason to believe that this observation is independent of the weather temperature correlation since we see the same trend: site specific correlations observed at Sumas 1 and Serpentine 2.  It should be noted that water temperature at Sumas 1 also correlated with VTEC and Salmonella recovery whereas these correlations were not observed for the average air temperature.  Since these two pathogens have been previously shown to occur more frequently in the  warmer months, but also after significant precipitation in previous studies, (Gu et al., 2013; Haley et al., 2009; Jokinen et al., 2012; Jones et al., 2014), it is likely that the higher precipitation observed during the cooler temperatures in this study confounds this correlation.  Finally, pH was found to significantly correlate negatively with pathogen occurrence. On a pathogen and site specific level, however, this correlation was only significant with respect to L. monocytogenes occurrence at the Serpentine 2 sample site.  The mean pH across sampling sites ranged from 7.11 to 7.72, but Serpentine 2 was singular in having a significantly lower mean pH than all of the other sample sites (Fisher’s LSD; p < 0.05).  Serpentine 2 also had the highest recovery of L. monocytogenes; likely driving this observed correlation.  Closer to neutral pH has been shown to improve the survival of L. monocytogenes in the soil environment (Locatelli et al., 2013b; McLaughlin et al., 2011), but it remains to be seen if this trend holds for surface waters as 106  well.  Furthermore, since the spread of pH across the samples did not vary largely, it could be suggested that this observed relationship is likely a coincidence or a correlation of minimal importance for pathogen prediction.  4.5 Conclusion Correlations between pathogen occurrence and environmental factors were observed, which may allow for the potential prediction of pathogen occurrence in surface waters used for irrigation.  While proximity and density of upstream livestock is a strong predictor of the likelihood of recovering VTEC, Salmonella, or L. monocytogenes, other sources such as wild animals or soil/sediment reservoirs cannot be discounted.  Given the presence of an upstream reservoir for these pathogens, recent significant precipitation and/or cooler temperatures may increase the probability of their presence in the water.  As temperature and precipitation were correlated together in this study, further investigation is required to determine if one or both of these factors is truly tied the presence of these pathogens; perhaps by only collecting samples after a minimum level of precipitation.  With these data, further understanding of the mechanisms leading to these bacteria infecting the surface waters may be developed.   107  Chapter 5 – Conclusion and Future Directions 5.1 Conclusion The occurrence and environmental factors affecting the presence of three foodborne pathogens (i.e., verotoxigenic Escherichia coli (VTEC), Salmonella, and Listeria monocytogenes) in the surface waters used for produce irrigation were investigated in the Lower Mainland of British Columbia.  All three pathogens were recovered from these waters, with L. monocytogenes being the most common with a 11.2% recovery, followed by VTEC and Salmonella, with 4.93% and 2.69% recoveries, respectively.  Of concern is that serotypes commonly associated with human illnesses in British Columbia were recovered for all three pathogens.  These results confirm the first hypothesis that VTEC, Salmonella, and L. monocytogenes are present in the surface waters used for irrigation in the Lower Mainland of British Columbia.  The second hypothesis was that the occurrence of these foodborne pathogens was not uniform across space or time.  The occurrence of L. monocytogenes was observed to be higher in the Serpentine watershed (20.2%) compared to the Sumas watershed (4.65%), while a significant difference in the recovery of VTEC at individual sites was observed in both watersheds, and significant difference in occurrence of Salmonella was observed at individual sites within the Sumas watershed.  Seasonally, L. monocytogenes was observed to occur more frequently during the fall (22.9%) and winter (16.3%) compared to the spring (7.8%) and summer (6.2%).  These findings suggest that we can accept the hypothesis that pathogen occurrence is not uniform.  Furthermore, they suggest that it may be possible to predict the occurrence of these pathogens based on local environmental factors such as weather and land use.  108  The third hypothesis was that higher numbers of total fecal coliforms (TFC) and/or generic E. coli can predict the presence of foodborne pathogens in surface waters.  Statistically significant correlations with pathogen occurrence and indicator concentrations were observed, but primarily only when a 25-ml volume of water was sampled using a filtration method.  The commonly used 1 ml method on 3M™ Petrifilm™ E. coli/Coliform count plates showed either reduced or non-existent correlations when compared to the filtration method.  Water samples that were high for either indicator organism were more likely to be positive for the presence of any of the three pathogens, but pathogens were also commonly recovered from water samples containing low levels of indicators.  The results suggest that high levels of indicators organisms indeed predict a higher risk for the presence of foodborne pathogens over the long run, but cannot accurately predict the presence or absence of foodborne pathogens at any particular point in time.  The final hypothesis was that environmental factors such as landscape and weather can predict the presence of foodborne pathogens in surface waters used for irrigation.  Proximity and density of upstream animal agriculture (i.e., cow or poultry) correlated highly with the occurrence of these three foodborne pathogens, emphasizing that domestic animals are common reservoirs of these bacteria.  Precipitation within three days prior to sample collection also correlated with the occurrence of pathogens, especially at sites with overall higher pathogen occurrence, suggesting that rain may act as a catalyst for the contamination of waterways and transport downstream.  Overall, these results suggest that being in close proximity to livestock, especially after a recent significant rainfall, may increase the risk of foodborne pathogens being present in the water.  109  In summary, foodborne pathogens associated with human illness are present in the waters used to irrigate fresh produce in British Columbia.  These bacteria are most commonly present during the fall and winter months, but can also be found during the growing and harvesting seasons.  The long-term trends of indicator organisms, as well as proximity and density of upstream sources of animal agriculture may provide an estimate of relative risk for a particular site.  Water from high risk sites should be used with caution when irrigating crops, especially after significant rainfall over the three days beforehand.  5.2 Future directions The present work revealed that pathogen occurrence is correlated to environmental factors affecting surface waters in the Lower Mainland of British Columbia, such as proximity to animal agriculture and recent precipitation, but further research is required to test whether these correlations are causative, and which may be confounded with other factors.  Hypothesis driven studies designed to investigate each factor should be conducted before any definitive conclusions can be drawn.    One primary example was that it was not possible to separate correlations between seasonal occurrence and precipitation, as rainfall and pathogen occurrence were both more common during the fall and winter months.  Precipitation provides a potential mechanism for transporting the pathogens downstream, but seasonal factors may also play a roll, such as the damming of waterways, and reducing water flow during the growing season.  Whether one or both of these are dominant factors affecting pathogen occurrence remains to be determined.  110  While this study showed the presence of foodborne illness causing organisms in the waters used for irrigation in the Lower Mainland of British Columbia, it remains to be determined at what levels these organisms pose a significant risk to human health.  To date, methods for detecting and isolating these pathogens are qualitative, often requiring a pre-enrichment step, and do not provide any information about pathogen density in water.  Furthermore, many previous studies sampled different volumes of water, making comparisons between studies difficult.  To this end, the development of novel methods that allow for the quantification of pathogens within a sample may help to better understand how concentration may vary across spatial and temporal lines, and at what concentrations these pathogens pose an actual risk to human health.  New technologies able to quantify small concentrations of DNA, such as digital droplet PCR, may hold the key to developing new quantitative methods for these pathogens.   111  Bibliography Adams, D.A., Thomas, K.R., Jajosky, R.A., Foster, L., Sharp, P., Onweh, D.H., Schley, A.W., Anderson, W.J., 2016. Summary of Notifiable Infectious Diseases and Conditions - United States, 2014. Morb. Mortal. Wkly. Rep. 63, 1–152. Alali, W.Q., Thakur, S., Berghaus, R.D., Martin, M.P., Gebreyes, W.A., 2010. Prevalence and Distribution of Salmonella in Organic and Conventional Broiler Poultry Farms. Foodborne Pathog. Dis. 7, 1363–1371. Alam, M.J., Zurek, L., 2006. Seasonal prevalence of Escherichia coli O157 : H7 in beef cattle feces. J. Food Prot. 69, 3018–3020. Andrés, S., Vico, J.P., Garrido, V., Grilló, M.J., Samper, S., Gavín, P., Herrera-León, S., Mainar-Jaime, R.C., 2013. Epidemiology of subclinical salmonellosis in wild birds from an area of high prevalence of pig salmonellosis: Phenotypic and genetic profiles of Salmonella isolates. Zoonoses Public Health 60, 355–365. Andrews, W.H., Wang, H., Jacobson, A., Hammack, T.S., 2016. Bacteriological Analytical Manual Chapter 5: Salmonella [WWW Document]. URL https://www.fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ucm070149.htm (accessed 2.24.16). Angelo, K.M., Chu, A., Anand, M., Nguyen, T.-A., Bottichio, L., Wise, M., Williams, I., Seelman, S., Bell, R., Fatica, M., Lance, S., Baldwin, D., Shannon, K., Lee, H., Trees, E., Strain, E., Gieraltowski, L., 2015. Outbreak of Salmonella Newport infections linked to cucumbers - United States, 2014. Morb. Mortal. Wkly. Rep. 64, 144–147. Angelo, K.M., Conrad, A.R., Saupe, A., Dragoo, H., West, N., Sorenson, A., BarnesS, A., Doyle, M., Beal, J., Jackson, K.A., Stroika, S., Tarr, C., Kucervova, Z., Lance, S., Gould, L.H., Wise, M., Jackson, B.R., 2017. Multistate outbreak of Listeria monocytogenes infections linked to whole apples used in commercially produced, prepackaged caramel apples: United States, 2014–2015. Epidemiol. Infect. 1–9. Arsenault, J., Letellier, A., Quessy, S., Boulianne, M., 2007. Prevalence and risk factors for Salmonella and Campylobacter spp. carcass contamination in broiler chickens slaughtered in Quebec, Canada. J. Food Prot. 70, 1820–8. Arvanitidou, M., Papa, A., Constantinidis, T.C., Danielides, V., Katsouyannopoulos, V., 1997. The occurrence of Listeria spp. and Salmonella spp. in surface waters. Microbiol. Res. 152, 395–7. Barton Behravesh, C., Mody, R.K., Jungk, J., Gaul, L., Redd, J.T., Chen, S., Cosgrove, S., Hedican, E., Sweat, D., Chávez-Hauser, L., Snow, S.L., Hanson, H., Nguyen, T.-A., Sodha, S. V, Boore, A.L., Russo, E., Mikoleit, M., Theobald, L., Gerner-Smidt, P., Hoekstra, R.M., Angulo, F.J., Swerdlow, D.L., Tauxe, R. V, Griffin, P.M., Williams, I.T., 2011. 2008 outbreak of Salmonella Saintpaul infections associated with raw produce. N. Engl. J. Med. 112  364, 918–927. BC Centre for Disease Control, 2017. Reportable Disease Dashboard [WWW Document]. URL http://www.bccdc.ca/health-info/disease-system-statistics/reportable-disease-dashboard (accessed 2.22.17). Benjamin, L., Atwill, E.R., Jay-Russell, M., Cooley, M., Carychao, D., Gorski, L., Mandrell, R.E., 2013. Occurrence of generic Escherichia coli, E. coli O157 and Salmonella spp. in water and sediment from leafy green produce farms and streams on the Central California coast. Int. J. Food Microbiol. 165, 65–76. Berger, C.N., Shaw, R.K., Brown, D.J., Mather, H., Clare, S., Dougan, G., Pallen, M.J., Frankel, G., 2009. Interaction of Salmonella enterica with basil and other salad leaves. ISME J. 3, 261–5. Berger, C.N., Sodha, S. V., Shaw, R.K., Griffin, P.M., Pink, D., Hand, P., Frankel, G., 2010. Fresh fruit and vegetables as vehicles for the transmission of human pathogens. Environ. Microbiol. 12, 2385–2397. Beuchat, L.R., 1996. Listeria monocytogenes: incidence on vegetables. Food Control 7, 223–228. Bielaszewska, M., Friedrich, A.W., Aldick, T., Schürk-Bulgrin, R., Karch, H., 2006. Shiga toxin activatable by intestinal mucus in Escherichia coli isolated from humans: predictor for a severe clinical outcome. Clin. Infect. Dis. 43, 1160–1167. Bille, J., Catimel, B., Bannerman, E., Jacquet, C., Yersin, M.N., Caniaux, I., Monget, D., Rocourt, J., 1992. API Listeria, a new and promising one-day system to identify Listeria isolates. Appl. Environ. Microbiol. 58, 1857–1860. Boerlin, P., McEwen, S.A., Boerlin-Petzold, F., Wilson, J.B., Johnson, R.P., Gyles, C.L., 1999. Associations between virulence factors of Shiga toxin-producing Escherichia coli and disease in humans. J. Clin. Microbiol. 37, 497–503. Bovo, F., De Cesare, A., Manfreda, G., Bach, S., Delaquis, P., 2015. Fate of Salmonella enterica in a Mixed Ingredient Salad Containing Lettuce, Cheddar Cheese, and Cooked Chicken Meat. J. Food Prot. 78, 491–497. Brenner, F.W., Villar, R.G., Angulo, F.J., Tauxe, R., Swaminathan, B., 2000. Salmonella nomenclature. J. Clin. Microbiol. 38, 2465–2467. British Columbia Centre for Disease Control, 2005. 2004 British Columbia annual summary of reportable disease [WWW Document]. URL http://www.bccdc.ca/resource-gallery/Documents/Statistics and Research/Statistics and Reports/Epid/Annual Reports/Epid_Stats_Research_CDAnnualReport_2004.pdf (accessed 11.14.16). Brooks, J.T., Sowers, E.G., Wells, J.G., Greene, K.D., Griffin, P.M., Hoekstra, R.M., Strockbine, N.A., 2005. Non-O157 Shiga toxin – producing Escherichia coli infections in the United 113  States , 1983–2002. J. Infect. Dis. 192, 1422–1424. Bruhn, J.B., Vogel, B.F., Gram, L., 2005. Bias in the Listeria monocytogenes enrichment procedure: lineage 2 strains outcompete lineage 1 strains in University of Vermont selective enrichments. Appl. Environ. Microbiol. 71, 961–967. Budzińska, K., Wroński, G., Szejniuk, B., 2012. Survival time of bacteria Listeria monocytogenes in water environment and sewage. Polish J. Environ. Stud. 21, 31–37. Bugarel, M., Beutin, L., Fach, P., 2010. Low-density macroarray targeting non-locus of enterocyte effacement effectors (nle genes) and major virulence factors of Shiga toxin-producing Escherichia coli (STEC): A new approach for molecular risk assessment of STEC isolates. Appl. Environ. Microbiol. 76, 203–211. Callaway, T.R., Keen, J.E., Edrington, T.S., Baumgard, L.H., Spicer, L., Fonda, E.S., Griswold, K.E., Overton, T.R., VanAmburgh, M.E., Anderson, R.C., Genovese, K.J., Poole, T.L., Harvey, R.B., Nisbet, D.J., 2005. Fecal prevalence and diversity of Salmonella species in lactating dairy cattle in four states. J. Dairy Sci. 88, 3603–3608. CanAgPlus, 2015. CanadaGAP Food Safety Manual for Fresh Fruits and Vegetables. Cartwright, E.J., Jackson, K.A., Johnson, S.D., Graves, L.M., Silk, B.J., Mahon, B.E., 2013. Listeriosis outbreaks and associated food vehicles, United States, 1998-2008. Emerg. Infect. Dis. 19, 1–9. Centers for Disease Control, 1991. Multistate Outbreak of Salmonella poona Infections - United States and Canada , 1991. Morb. Mortal. Wkly. Rep. 40, 549–552. Centers for Disease Control, 2007. Mulitistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants - United States, 2005-2006. Morb. Mortal. Wkly. Rep. 56, 909–911. Centers for Disease Control, 2008. Investigation update: outbreak of Salmonella Litchfield infections, 2008 [WWW Document]. URL http://www.cdc.gov/Salmonella/2008/cantaloupes-4-2-2008.html (accessed 11.14.16). Centers for Disease Control, 2011. Multistate Outbreak of Human Salmonella Agona Infections Linked to Whole, Fresh Imported Papayas (Final Update) [WWW Document]. URL http://www.cdc.gov/Salmonella/2011/papayas-8-29-2011.html (accessed 11.14.16). Centers for Disease Control, 2012. Multistate Outbreak of Shiga Toxin-producing Escherichia coli O26 Infections Linked to Raw Clover Sprouts at Jimmy John’s Restaurants (Final Update) [WWW Document]. URL https://www.cdc.gov/ecoli/2012/o26-02-12/index.html (accessed 2.12.17). Centers for Disease Control, 2014. Multistate Outbreak of Shiga toxin-producing Escherichia coli O121 Infections Linked to Raw Clover Sprouts (Final Update) [WWW Document]. URL https://www.cdc.gov/ecoli/2014/o121-05-14/index.html (accessed 2.12.17). 114  Centers for Disease Control, 2015. Multistate Outbreak of Listeriosis Linked to Commercially Produced, Prepackaged Caramel Apples Made from Bidart Bros. Apples (Final Update) [WWW Document]. URL https://www.cdc.gov/Listeria/outbreaks/caramel-apples-12-14/index.html (accessed 10.28.16). Centers for Disease Control, 2016a. Multistate Outbreak of Shiga toxin-producing Escherichia coli O157 Infections Linked to Alfalfa Sprouts Produced by Jack & The Green Sprouts (Final Update) [WWW Document]. URL https://www.cdc.gov/ecoli/2016/o157-02-16/index.html (accessed 2.12.17). Centers for Disease Control, 2016b. Multistate Outbreak of Salmonella Poona Infections Linked to Imported Cucumbers (Final Update) [WWW Document]. URL https://www.cdc.gov/Salmonella/poona-09-15/ (accessed 2.23.17). Centers for Disease Control, 2016c. Foodborne Outbreak Online Database (FOOD Tool) [WWW Document]. URL http://wwwn.cdc.gov/foodborneoutbreaks/ Centers for Disease Control, 2016d. Multistate Outbreak of Listeriosis Linked to Packaged Salads Produced at Springfield, Ohio Dole Processing Facility (Final Update) [WWW Document]. URL https://www.cdc.gov/Listeria/outbreaks/bagged-salads-01-16/index.html (accessed 10.28.16). Chapin, T.K., Nightingale, K.K., Worobo, R.W., Wiedmann, M., Strawn, L.K., 2014. Geographical and Meteorological Factors Associated with Isolation of Listeria Species in New York State Produce Production and Natural Environments. J. Food Prot. 77, 1919–1928. Chen, Y., Luo, Y., Pettengill, J., Timme, R., Melka, D., Doyle, M., Jackson, A., Parish, M., Hammack, T.S., Allard, M., Brown, E.W., Strain, E.A., 2017. Singleton ST382, an emerging clonal group of Listeria monocytogenes associated with three multistate outbreaks linked to contaminated stone fruit, caramel apples and leafy green salad. J. Clin. Microbiol. JCM.02140-16. Christensen, D., Crawford, C., Szabo, R., 2002. Enumeration of Coliforms, Faecal Coliforms, and of E. coli in Foods Using the MPN Method. Coburn, B., Grassl, G.A., Finlay, B.B., 2007. Salmonella, the host and disease: a brief review. Immunol. Cell Biol. 85, 112–118. Constantiniu, S., 2002. Escherichia coli Enterohemoragic – An Emerged Pathogen Of Human Infections Part II . Non-O157 Escherichia coli enterohemorrhagic. J. Prev. Med. 10, 57–73. Cooley, M., Carychao, D., Crawford-Miksza, L., Jay, M.T., Myers, C., Rose, C., Keys, C., Farrar, J., Mandrell, R.E., 2007. Incidence and tracking of Escherichia coli O157:H7 in a major produce production region in California. PLoS One 2. Cooley, M.B., Jay-Russell, M., Atwill, E.R., Carychao, D., Nguyen, K., Quiñones, B., Patel, R., Walker, S., Swimley, M., Pierre-Jerome, E., Gordus, A.G., Mandrell, R.E., 2013. 115  Development of a Robust Method for Isolation of Shiga Toxin-Positive Escherichia coli (STEC) from Fecal, Plant, Soil and Water Samples from a Leafy Greens Production Region in California. PLoS One 8. Cooley, M.B., Quiñones, B., Oryang, D., Mandrell, R.E., Gorski, L., 2014. Prevalence of shiga toxin producing Escherichia coli, Salmonella enterica, and Listeria monocytogenes at public access watershed sites in a California Central Coast agricultural region. Front. Cell. Infect. Microbiol. 4, 30. Croxen, M.A., Law, R.J., Scholz, R., Keeney, K.M., Wlodarska, M., Finlay, B.B., 2013. Recent advances in understanding enteric pathogenic Escherichia coli. Clin. Microbiol. Rev. 26, 822–880. de Noordhout, C.M., Devleesschauwer, B., Angulo, F.J., Verbeke, G., Haagsma, J., Kirk, M., Havelaar, A., Speybroeck, N., 2014. The global burden of listeriosis: A systematic review and meta-analysis. Lancet Infect. Dis. 14, 1073–1082. Deeks, S., Ellis, A., Ciebin, B., Khakhria, R., Naus, M., Hockin, J., 1998. Salmonella Oranienburg, Ontario. Canada Commun. Dis. Rep. 24, 177–179. Doumith, M., Day, M.J., Hope, R., Wain, J., Woodford, N., 2012. Improved Multiplex PCR Strategy for Rapid Assignment of the Four Major Escherichia coli Phylogenetic Groups 50, 3108–3110. Dowe, M., Jackson, E., Mori, J., Colin, B., 1997. Listeria monocytogenes Survival in Soil and Incidence in Agricultural Soils. J. Food Prot. 10, 1158–1286. Duris, J.W., Reif, A.G., Krouse, D.A., Isaacs, N.M., 2013. Factors related to occurrence and distribution of selected bacterial and protozoan pathogens in Pennsylvania streams. Water Res. 47, 300–314. Edge, T. a, El-Shaarawi,  a, Gannon, V., Jokinen, C., Kent, R., Khan, I.U.H., Koning, W., Lapen, D., Miller, J., Neumann, N., Phillips, R., Robertson, W., Schreier, H., Scott,  a, Shtepani, I., Topp, E., Wilkes, G., van Bochove, E., 2012. Investigation of an Escherichia coli environmental benchmark for waterborne pathogens in agricultural watersheds in Canada. J. Environ. Qual. 41, 21–30. Erickson, M.C., Webb, C.C., Diaz-Perez, J.C., Phatak, S.C., Silvoy, J.J., Davey, L., Payton, A.S., Liao, J., Ma, L., Doyle, M.P., 2010. Surface and internalized Escherichia coli O157:H7 on field-grown spinach and lettuce treated with spray-contaminated irrigation water. J. Food Prot. 73, 1023–1029. Esteban, J.I., Oporto, B., Aduriz, G., Juste, R.A., Hurtado, A., 2008. A survey of food-borne pathogens in free-range poultry farms. Int. J. Food Microbiol. 123, 177–182. Farber, J.M., Peterkin, P.I., 1991. Listeria monocytogenes , a Food-Borne Pathogen. Microbiol. Rev. 55, 476–511. 116  Farzan, A., Friendship, R.M., Cook, A., Pollari, F., 2010. Occurrence of Salmonella, Campylobacter, Yersinia enterocolitica, Escherichia coli O157 and Listeria monocytogenes in swine. Zoonoses Public Health 57, 388–396. Feng, P., 2014. Shiga Toxin-Producing Escherichia coli (STEC) in Fresh Produce — A Food Safety Dilemma. Microbiol. Spectr. 2, 1–11. Feng, P.C.H., Reddy, S., 2013. Prevalences of shiga toxin subtypes and selected other virulence factors among Shiga-toxigenic Escherichia coli strains isolated from fresh produce. Appl. Environ. Microbiol. 79, 6917–6923. Fisher, N., Bourne, A., Plunkett, D., 2015. Outbreak Alert! 2015 - A review of foodborne illness in the U.S. from 2004-2013. Foley, S.L., Nayak, R., Hanning, I.B., Johnson, T.J., Han, J., Ricke, S.C., 2011. Population dynamics of Salmonella enterica serotypes in commercial egg and poultry production. Appl. Environ. Microbiol. 77, 4273–4279. Frank, C., Werber, D., Cramer, J.P., Askar, M., Faber, M., an der Heiden, M., Bernard, H., Fruth, A., Prager, R., Spode, A., Wadl, M., Zoufaly, A., Jordan, S., Kemper, M.J., Follin, P., Müller, L., King, L.A., Rosner, B., Buchholz, U., Stark, K., Krause, G., 2011. Epidemic Profile of Shiga-Toxin–Producing Escherichia coli O104:H4 Outbreak in Germany. N. Engl. J. Med. 365, 1771–1780. Fratamico, P.M., Bagi, L.K., Bush, E.J., Solow, B.T., 2004. Prevalence and characterization of Shiga toxin-producing Escherichia coli in swine feces recovered in the national animal health monitoring system’s swine 2000 study. Appl. Environ. Microbiol. 70, 7173–7178. Gannon, V.P.J., Graham, T.A., Read, S., Ziebell, K., Muckle, A., Mori, J., Thomas, J., Selinger, B., Townshend, I., Byrne, J., 2004. Bacterial Pathogens in Rural Water Supplies in Southern Alberta, Canada. J. Toxicol. Enrivo. Health A. 67, 1643–1653. Gaul, L.K., Farag, N.H., Shim, T., Kingsley, M.A., Silk, B.J., Hyytia-Trees, E., 2013. Hospital-acquired listeriosis outbreak caused by contaminated diced celery-texas, 2010. Clin. Infect. Dis. 56, 20–26. Gazal, S., Taku, A., Bhat, M., Badroo, G.B.T.-I.J. of M.M., 2014. Group A rotavirus and bacterial agents associated with diarrhoea-induced hospitalisations in children below 5 years of age in Jammu. Indian J. Med. Microbiol. 32, 57. Gorski, L., Walker, S., Liang, A.S., Nguyen, K.M., Govoni, J., Carychao, D., Cooley, M.B., Mandrell, R.E., 2014. Comparison of subtypes of Listeria monocytogenes isolates from naturally contaminated watershed samples with and without a selective secondary enrichment. PLoS One 9. Government of British Columbia, n.d. B.C. Good Agricultural Practices Guide [WWW Document]. URL http://www2.gov.bc.ca/gov/content/industry/agriculture-seafood/food-safety/good-agricultural-practices (accessed 1.15.17). 117  Grant, M. a, Hedberg, C., Harris, J., Logue, C.M., Meng, J., 2011. The Significance of Non-O157 Shiga Toxin-producing Escherichia coli in food. Food Prot. Trends 33–56. Greene, S.K., Daly, E.R., Talbot, E. a, Demma, L.J., Holzbauer, S., Patel, N.J., Hill, T. a, Walderhaug, M.O., Hoekstra, R.M., Lynch, M.F., Painter, J. a, 2008. Recurrent multistate outbreak of Salmonella Newport associated with tomatoes from contaminated fields, 2005. Epidemiol. Infect. 136, 157–165. Gu, G., Luo, Z., Cevallos-Cevallos, J.M., Adams, P., Vellidis, G., Wright, A., van Bruggen, A.H.C., 2013. Occurrence and population density of Escherichia coli O157 contamination in irrigation ponds on produce farms in the Suwannee River Watershed. Can. J. Microbiol. 59, 339–46. Guillet, C., Join-Lambert, O., Le Monnier, A., Leclercq, A., Mechaï, F., Mamzer-Bruneel, M.F., Bielecka, M.K., Scortti, M., Disson, O., Berche, P., Vazquez-Boland, J., Lortholary, O., Lecuit, M., 2010. Human listeriosis caused by Listeria ivanovii. Emerg. Infect. Dis. 16, 136–138. Gyles, C.L., 1992. Escherichia coli cytotoxins and enterotoxins. Can. J. Microbiol. 38, 734–746. Haase, J.K., Didelot, X., Lecuit, M., Korkeala, H., Achtman, M., Leclercq, A., Grant, K., Wiedmann, M., Apfalter, P., 2014. The ubiquitous nature of Listeria monocytogenes clones: A large-scale Multilocus Sequence Typing study. Environ. Microbiol. 16, 405–416. Haley, B.J., Cole, D.J., Lipp, E.K., 2009. Distribution, diversity, and seasonality of waterborne Salmonellae in a rural watershed. Appl. Environ. Microbiol. 75, 1248–1255. Hancock, D., Besser, T., Lejeune, J., Davis, M., Rice, D., 2001. The control of VTEC in the animal reservoir. Int. J. Food Microbiol. 66, 71–78. Health Canada, 2012. Guidelines for Canadian Recreational Water Quality. Health Canada, 2014. Isolation of Escherichia coli O157:H7/NM from foods and environmental surface samples. Health Canada, 2016. Canada’s Food Guide [WWW Document]. URL http://www.hc-sc.gc.ca/fn-an/food-guide-aliment/index-eng.php (accessed 2.25.17). Hintz, L.D., Boyer, R.R., Ponder, M.A., Williams, R.C., Rideout, S.L., 2010. Recovery of Salmonella enterica newport introduced through irrigation water from tomato (Lycopersicum esculentum) fruit, roots, stems, and leaves. HortScience 45, 675–678. Hitchins, A.D., Jinneman, K., Chen, Y., 2016. Bacteriological Analytical Manual Chapter 10: Detection and Enumeration of Listeria monocytogenes in Foods. Holvoet, K., Sampers, I., Seynnaeve, M., Uyttendaele, M., 2014. Relationships among hygiene indicators and enteric pathogens in irrigation water, soil and lettuce and the impact of climatic conditions on contamination in the lettuce primary production. Int. J. Food 118  Microbiol. 171, 21–31. Honish, L., Nguyen, Q., 2001. Outbreak of Salmonella Enteritidis phage type 913 gastroenteritis associated with mung bean sprouts - Edmonton, 2001. Canada Commun. Dis. Rep. 27, 151–156. Hudson, J.A., 1992. Efficacy of high sodium chloride concentrations for the destruction of Listeria monocytogenes. Lett. Appl. Microbiol. 14, 178–180. Hussein, H., Bollinger, L., 2005. Prevalence of Shiga toxin-producing Escherichia coli in beef cattle. J Food Prot 68, 2224–2241. Hussein, H.S., Sakuma, T., 2005. Invited Review: Prevalence of Shiga Toxin-Producing Escherichia coli in Dairy Cattle and Their Products. J. Dairy Sci. 88, 450–465. Husu, J.R., 1990. Epidemiological Studies on the Occurrence of Listeria monocytogenes in the Feces of Dairy Cattle. J. Vet. Med. Ser. B 37, 276–282. Ingram, D.T., Patel, J., Sharma, M., 2011. Effect of repeated irrigation with water containing varying levels of total organic carbon on the persistence of Escherichia coli O157:H7 on baby spinach. J. Food Prot. 74, 709–717. Islam, M., Doyle, M.P., Phatak, S.C., Millner, P., Jiang, X., 2004a. Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. J. Food Prot. 67, 1365–70. Islam, M., Doyle, M.P., Phatak, S.C., Millner, P., Jiang, X., 2005. Survival of Escherichia coli O157:H7 in soil and on carrots and onions grown in fields treated with contaminated manure composts or irrigation water. Food Microbiol. 22, 63–70. Islam, M., Morgan, J., Doyle, M.P., Phatak, S.C., Millner, P., Jiang, X., 2004b. Persistence of Salmonella enterica serovar typhimurium on lettuce and parsley and in soils on which they were grown in fields treated with contaminated manure composts or irrigation water. Foodborne Pathog. Dis. 1, 27–35. Jacob, M.E., Foster, D.M., Rogers, A.T., Balcomb, C.C., Shi, X., Nagaraja, T.G., 2013. Evidence of Non-O157 Shiga Toxin–Producing Escherichia coli in the Feces of Meat Goats at a U.S. Slaughter Plant. J. Food Prot. 76, 1626–1629. Jacobsen, C.S., Bech, T.B., 2012. Soil survival of Salmonella and transfer to freshwater and fresh produce. Food Res. Int. 45, 557–566. Jenke, C., Lindstedt, B.A., Hamensen, D., Karch, H., Brandal, L.T., Mellmann, A., 2011. Comparison of multilocus variable-number tandem-repeat analysis with multilocus sequence typing for differentiation of hemolytic-uremic syndrome associated Escherichia coli (HUSEC) collection strains. J. Clin. Microbiol. 49, 3644–3646. 119  Johnson, J.Y.M., Thomas, J.E., Graham, T. a, Townshend, I., Byrne, J., Selinger, L.B., Gannon, V.P.J., 2003. Prevalence of Escherichia coli O157:H7 and Salmonella spp. in surface waters of southern Alberta and its relation to manure sources. Can. J. Microbiol. 49, 326–335. Johnson, R.P., Holtslander, B., Mazzocco, A., Roche, S., Thomas, J.L., Pollari, F., Pintar, K.D.M., 2014. Detection and prevalence of verotoxin-producing Escherichia coli O157 and Non-O157 serotypes in a Canadian watershed. Appl. Environ. Microbiol. 80, 2166–2175. Jokinen, C., Edge, T.A., Ho, S., Koning, W., Laing, C., Mauro, W., Medeiros, D., Miller, J., Robertson, W., Taboada, E., Thomas, J.E., Topp, E., Ziebell, K., Gannon, V.P.J., 2011. Molecular subtypes of Campylobacter spp., Salmonella enterica, and Escherichia coli O157:H7 isolated from faecal and surface water samples in the Oldman River watershed, Alberta, Canada. Water Res. 45, 1247–1257. Jokinen, C.C., Edge, T. a., Koning, W., Laing, C.R., Lapen, D.R., Miller, J., Mutschall, S., Scott, A., Taboada, E.N., Thomas, J.E., Topp, E., Wilkes, G., Gannon, V.P.J., 2012. Spatial and Temporal Drivers of Zoonotic Pathogen Contamination of an Agricultural Watershed. J. Environ. Qual. 41, 242. Jokinen, C.C., Schreier, H., Mauro, W., Taboada, E., Isaac-Renton, J.L., Topp, E., Edge, T., Thomas, J.E., Gannon, V.P.J., 2010. The occurrence and sources of Campylobacter spp., Salmonella enterica and Escherichia coli O157:H7 in the Salmon River, British Columbia, Canada. J. Water Health 8, 374–386. Jones, L.A., Worobo, R.W., Smart, C.D., 2014. Plant-pathogenic oomycetes, Escherichia coli strains, and Salmonella spp. frequently found in surface water used for irrigation of fruit and vegetable crops in New York State. Appl. Environ. Microbiol. 80, 4814–4820. Jones, T.F., Ingram, L.A., Cieslak, P.R., Vugia, D.J., Tobin-D’Angelo, M., Hurd, S., Medus, C., Cronquist, A., Angulo, F.J., 2008. Salmonellosis outcomes differ substantially by serotype. J. Infect. Dis. 198, 109–14. Karama, M., Johnson, R.P., Holtslander, R., Gyles, C.L., 2008a. Phenotypic and genotypic characterization of verotoxin-producing Escherichia coli O103:H2 isolates from cattle and humans. J. Clin. Microbiol. 46, 3569–3575. Karama, M., Johnson, R.P., Holtslander, R., McEwen, S.A., Gyles, C.L., 2008b. Prevalence and characterization of verotoxin-producing Escherichia coli (VTEC) in cattle from an Ontario abattoir. Can. J. Vet. Res. 72, 297–302. King, L.A., Nogareda, F., Weill, F.X., Mariani-Kurkdjian, P., Loukiadis, E., Gault, G., Jourdan-Dasilva, N., Bingen, E., MacÉ, M., Thevenot, D., Ong, N., Castor, C., Noël, H., Van Cauteren, D., Charron, M., Vaillant, V., Aldabe, B., Goulet, V., Delmas, G., Couturier, E., Le Strat, Y., Combe, C., Delmas, Y., Terrier, F., Vendrely, B., Rolland, P., De Valk, H., 2012. Outbreak of shiga toxin-producing Escherichia coli O104:H4 Associated with organic fenugreek sprouts, France, June 2011. Clin. Infect. Dis. 54, 1588–1594. Kozak, G.K., MacDonald, D., Landry, L., Farber, J.M., 2013. Foodborne Outbreaks in Canada 120  Linked to Produce: 2001 through 2009. J. Food Prot. 76, 173–183. Li, H., Wang, H., D’Aoust, J.-Y., Maurer, J., 2013. Salmonella Species. In: Doyle, M.P., Buchanan, R.L. (Eds.), Food Microbiology: Fundamentals and Frontiers. ASM Press, Washington, D.C., pp. 225–261. Linke, K., Rückerl, I., Brugger, K., Karpiskova, R., Walland, J., Muri-Klinger, S., Tichy, A., Wagner, M., Stessl, B., 2014. Reservoirs of Listeria species in three environmental ecosystems. Appl. Environ. Microbiol. 80, 5583–5592. Liu, D., 2006. Identification, subtyping and virulence determination of Listeria monocytogenes, an important foodborne pathogen. J. Med. Microbiol. 55, 645–659. Locatelli, A., Depret, G., Jolivet, C., Henry, S., Dequiedt, S., Piveteau, P., Hartmann, A., 2013a. Nation-wide study of the occurrence of Listeria monocytogenes in French soils using culture-based and molecular detection methods. J. Microbiol. Methods 93, 242–250. Locatelli, A., Spor, A., Jolivet, C., Piveteau, P., Hartmann, A., 2013b. Biotic and Abiotic Soil Properties Influence Survival of Listeria monocytogenes in Soil. PLoS One 8, 1–8. Lyautey, E., Hartmann, A., Pagotto, F., Tyler, K., Lapen, D.R., Wilkes, G., Piveteau, P., Rieu, A., Robertson, W.J., Medeiros, D.T., Edge, T. a, Gannon, V., Topp, E., 2007a. Characteristics and frequency of detection of fecal Listeria monocytogenes shed by livestock, wildlife, and humans. Can. J. Microbiol. 53, 1158–67. Lyautey, E., Lapen, D.R., Wilkes, G., McCleary, K., Pagotto, F., Tyler, K., Hartmann, A., Piveteau, P., Rieu, A., Robertson, W.J., Medeiros, D.T., Edge, T.A., Gannon, V., Topp, E., 2007b. Distribution and characteristics of Listeria monocytogenes isolates from surface waters of the South Nation River watershed, Ontario, Canada. Appl. Environ. Microbiol. 73, 5401–5410. Lynch, M.F., Blanton, E.M., Bulens, S., Polyak, C., Vojdani, J., Stevenson, J., Medalla, F., Barzilay, E., Joyce, K., Barrett, T., Mintz, E.D., 2009. Typhoid Fever in the United States, 1999-2006. JAMA 302, 859. MacGowan, A., Bowker, K., McLaughlin, J., Bennett, P., Reeves, D., 1994. The occurrence and seasonal changes in the isolation of Listeria spp. in shop bought food stuffs, human faeces, sewagea and soil from urban sources. Int. J. Food Microbiol. 21, 325–334. Mahon, B.E., Pönkä, A., Hall, W.N., Komatsu, K., Dietrich, S.E., Siitonen, A., Cage, G., Hayes, P.S., Lambert-Fair, M. a, Bean, N.H., Griffin, P.M., Slutsker, L., 1997. An international outbreak of Salmonella infections caused by alfalfa sprouts grown from contaminated seeds. J. Infect. Dis. 175, 876–882. Majowicz, S.E., Musto, J., Scallan, E., Angulo, F.J., Kirk, M., O’Brien, S.J., Jones, T.F., Fazil, A., Hoekstra, R.M., 2010. The Global Burden of Nontyphoidal Salmonella Gastroenteritis. Clin. Infect. Dis. 50, 882–889. 121  Majowicz, S.E., Scallan, E., Jones-Bitton, A., Sargeant, J.M., Stapleton, J., Angulo, F.J., Yeung, D.H., Kirk, M.D., 2014. Global Incidence of Human Shiga Toxin–Producing Escherichia coli Infections and Deaths: A Systematic Review and Knowledge Synthesis. Foodborne Pathog. Dis. 11, 447–455. March, S.B., Ratnam, S., 1986. Sorbitol-MacConkey medium for detection of Escherichia coli O157:H7 associated with hemorrhagic colitis. J. Clin. Microbiol. 23, 869–872. Marder, E.P., Garman, K.N., Ingram, L.A., Dunn, J.R., 2014. Multistate outbreak of Escherichia coli O157: H7 associated with bagged salad. Foodborne Pathog. Dis. 11, 593–595. Mariani-Kurkdjian, P., Denamur, E., Milon, A., Picard, B., Cave, H., Lambert- Zechovsky, N., Loirat, C., Goullet, P., Sansonetti, P.J., Elion, J., 1993. Identification of a clone of Escherichia coli O103:H2 as a potential agent of hemolytic-uremic syndrome in France. J. Clin. Microbiol. 31, 296–301. Mayer, C.L., Leibowitz, C.S., Kurosawa, S., Stearns-Kurosawa, D.J., 2012. Shiga toxins and the pathophysiology of hemolytic uremic syndrome in humans and animals. Toxins (Basel). 4, 1261–1287. McCollum, J.T., Cronquist, A.B., Silk, B.J., Jackson, K.A., O’Connor, K.A., Cosgrove, S., Gossack, J.P., Parachini, S.S., Jain, N.S., Ettestad, P., Ibraheem, M., Cantu, V., Joshi, M., DuVernoy, T., Fogg, N.W., Gorny, J.R., Mogen, K.M., Spires, C., Teitell, P., Joseph, L.A., Tarr, C.L., Imanishi, M., Neil, K.P., Tauxe, R. V, Mahon, B.E., 2013. Multistate outbreak of listeriosis associated with cantaloupe. N. Engl. J. Med. 369, 944–53. McLauchlin, J., Mitchell, R.T., Smerdon, W.J., Jewell, K., 2004. Listeria monocytogenes and listeriosis: A review of hazard characterisation for use in microbiological risk assessment of foods. Int. J. Food Microbiol. 92, 15–33. McLaughlin, H.P., Casey, P.G., Cotter, J., Gahan, C.G.M., Hill, C., 2011. Factors affecting survival of Listeria monocytogenes and Listeria innocua in soil samples. Arch. Microbiol. 193, 775–785. McLaughlin, J., Rees, C.E.D., 2009. Listeria. In: De Vos, P., Garrity, G.M., Jones, D., Krieg, N.R., Ludwig, W., Rainey, F.A., Schleifer, K.-H., Whitman, W.B. (Eds.), Bergey’s Manual of Systematic Bacteriology. Springer, New York, pp. 244–257. McQuiston, J.R., Herrera-Leon, S., Wertheim, B.C., Doyle, J., Fields, P.I., Tauxe, R. V., Logsdon, J.M., 2008. Molecular phylogeny of the Salmonellae: relationships among Salmonella species and subspecies determined from four housekeeping genes and evidence of lateral gene transfer events. J. Bacteriol. 190, 7060–7067. Micallef, S.A., Rosenberg Goldstein, R.E., George, A., Kleinfelter, L., Boyer, M.S., McLaughlin, C.R., Estrin, A., Ewing, L., Jean-Gilles Beaubrun, J., Hanes, D.E., Kothary, M.H., Tall, B.D., Razeq, J.H., Joseph, S.W., Sapkota, A.R., 2012. Occurrence and antibiotic resistance of multiple Salmonella serotypes recovered from water, sediment and soil on mid-Atlantic tomato farms. Environ. Res. 114, 31–39. 122  Michigan Department of Health, 2008. E. coli outbreak update (Oct 14) [WWW Document]. URL http://www.michigan.gov/documents/mdch/e_coli_253032_7.pdf (accessed 2.12.17). Milillo, S.R., Friedly, E.C., Saldivar, J.C., Muthaiyan, A., O’Bryan, C., Crandall, P.G., Johnson, M.G., Ricke, S.C., 2012. A review of the ecology, genomics, and stress response of Listeria innocua and Listeria monocytogenes. Crit. Rev. Food Sci. Nutr. 52, 712–25. Mootian, G., Wu, W.-H., Matthews, K.R., 2009. Transfer of Escherichia coli O157:H7 from soil, water, and manure contaminated with low numbers of the pathogen to lettuce plants. J. Food Prot. 72, 2308–2312. Muraoka, R., Okazaki, M., Fujimoto, Y., Jo, N., Yoshida, R., Kiyoyama, T., Oura, Y., Hirakawa, K., Jyukurogi, M., Kawano, K., Okada, M., Shioyama, Y., Iryoda, K., Wakamatu, H., Kawabata, N., 2007. An enterohemorrhagic Escherichia coli O103 outbreak at a nursery school in Miyazaki Prefecture, Japan. Jpn. J. Infect. Dis. 60, 410–1. Myers, D.N., Stoeckel, D.M., Bushon, R.N., Francy, D.S., Brady, A.M.G., 2014. Fecal indicator bacteria (ver. 2.1): U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A7, section 7.1 [WWW Document]. URL http://pubs.water.usgs.gov/twri9A/ (accessed 3.22.17). Nadya, S., Delaquis, P., Chen, J., Allen, K., Johnson, R.P., Ziebell, K., Laing, C., Gannon, V., Bach, S., Topp, E., 2016. Phenotypic and Genotypic Characteristics of Shiga Toxin-Producing Escherichia coli Isolated from Surface Waters and Sediments in a Canadian Urban-Agricultural Landscape. Front. Cell. Infect. Microbiol. 6, 36. Nataro, J., Kaper, J., 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11, 142. Navarro-Garcia, F., 2014. Escherichia coli O104:H4 Pathogenesis: an Enteroaggregative E. coli/Shiga Toxin-Producing E. coli Explosive Cocktail of High Virulence. Microbiol. Spectr. 2, 1–19. Nightingale, K.K., Schukken, Y.H., Nightingale, C.R., Fortes, E.D., Ho,  a J., Her, Z., Grohn, Y.T., Mcdonough, P.L., Wiedmann, M., 2004. Ecology and Transmission of Listeria monocytogenes Infecting Ruminants and in the Farm Environment Ecology and Transmission of Listeria monocytogenes Infecting Ruminants and in the Farm Environment. Appl. Environ. Microbiol. 70, 4458–4467. Nightingale, K.K., Windham, K., Wiedmann, M., 2005. Evolution and Molecular Phylogeny of Listeria monocytogenes Isolated from Human and Animal Listeriosis Cases and Foods Evolution and Molecular Phylogeny of Listeria monocytogenes Isolated from Human and Animal Listeriosis Cases and Foods. J. Bacteriol. 187, 5537–5551. Nolan, D.A., Chamblin, D.C., Troller, J.A., 1992. Minimal water activity levels for growth and survival of Listeria monocytogenes and Listeria innocua. Int. J. Food Microbiol. 16, 323–335. O’Shea, M.L., Field, R., 1992. Detection and disinfection of pathogens in storm-generated flows. 123  Can. J. Microbiol. 38, 267–276. Ohl, M., Miller, S., 2001. Salmonella: a model for bacterial pathogenesis. Annu. Rev. Med. 52, 259–74. Olaimat, A.N., Holley, R.A., 2012. Factors influencing the microbial safety of fresh produce: A review. Food Microbiol. 32, 1–19. Oliveira, M., Usall, J., Viñas, I., Solsona, C., Abadias, M., 2011. Transfer of Listeria innocua from contaminated compost and irrigation water to lettuce leaves. Food Microbiol. 28, 590–596. Oliveira, M., Viñas, I., Usall, J., Anguera, M., Abadias, M., 2012. Presence and survival of Escherichia coli O157:H7 on lettuce leaves and in soil treated with contaminated compost and irrigation water. Int. J. Food Microbiol. 156, 133–140. Orsi, R.H., Bakker, H.C. den, Wiedmann, M., 2011. Listeria monocytogenes lineages: Genomics, evolution, ecology, and phenotypic characteristics. Int. J. Med. Microbiol. 301, 79–96. Pachepsky, Y., Shelton, D., Dorner, S., Whelan, G., 2014. Can E. coli or thermotolerant coliform concentrations predict pathogen presence or prevalence in irrigation waters? Crit Rev Microbiol 7828, 1–10. Pachepsky, Y., Shelton, D.R., McLain, J.E.T., Patel, J., Mandrell, R.E., 2011. Irrigation Waters as a Source of Pathogenic Microorganisms in Produce. A Review, Advances in Agronomy. Elsevier Inc. Pagotto, F., Hébert, K., Farber, J., 2011. MFHPB-30 - Isolation of Listeria monocytogenes and other Listeria spp. from foods and environmental samples. In: Compendium of Analytical Methods, Volume 2. Health Canada. Palumbo, S.A., Williams, A.C., 1991. Resistance of Listeria monocytogenes to freezing in foods. Food Microbiol. 8, 63–68. Park, S.Y., Woodward, C.L., Kubena, L.F., Nisbet, D.J., Birkhold, S.G., Ricke, S.C., 2008. Environmental Dissemination of Foodborne Salmonella in Preharvest Poultry Production: Reservoirs, Critical Factors, and Research Strategies, Critical Reviews in Environmental Science and Technology. Parsons, B.D., Zelyas, N., Berenger, B.M., Chui, L., 2016. Detection, characterization, and typing of shiga toxin-producing Escherichia coli. Front. Microbiol. 7, 1–12. Paton, A.W., Paton, J.C., 1998. Detection and Characterization of Shiga Toxigenic Escherichia coli by Using Multiplex PCR Assays for stx 1 , stx 2 , eaeA , Enterohemorrhagic E . coli hlyA , rfb O111 , and rfb O157 Downloaded from http://jcm.asm.org/ on February 19 , 2015 by The Universi 36, 598–602. Persad, A.K., Lejeune, J.T., 2014. Animal reservoirs of Shiga toxin-producing Escherichia coli. 124  Microbiol. Spectr. 2, 1–14. Phan-Thanh, L., Montagne, A., 1998. Physiological and biochemical aspects of the acid survival of Listeria monocytogenes. J. Gen. Appl. Microbiol. 44, 183–191. Popoff, M.Y., Le Minor, L.E., 2005. Salmonella. In: Brenner, D.J., Krieg, N.R., Staley, J.T., Garrity, G.M. (Eds.), Bergey’s Manual of Systematic Bacteriology, Volume Two, Part B. Springer-Verlag New York, New York, pp. 764–799. Prager, R., Liesegang, A., Voigt, W., Rabsch, W., Fruth, A., Tschäpe, H., 2002. Clonal diversity of Shiga toxin-producing Escherichia coli O103:H2/H(-) in Germany. Infect. Genet. Evol. 1, 265–275. Public Health Agency of Canada, 2011. Salmonella enterica spp. [WWW Document]. URL http://www.phac-aspc.gc.ca/lab-bio/res/psds-ftss/Salmonella-ent-eng.php (accessed 2.26.17). Public Health Agency of Canada, 2012. Listeria monocytogenes [WWW Document]. URL http://www.phac-aspc.gc.ca/lab-bio/res/psds-ftss/Listeria-monocytogenes-eng.php (accessed 2.26.17). Public Health Agency of Canada, 2015a. Executive summary for the National Enteric Surveillance Program 2013 Annual Report [WWW Document]. URL https://www.nml-lnm.gc.ca/NESP-PNSME/surveillance-2013-eng.html (accessed 2.22.17). Public Health Agency of Canada, 2015b. Public Health Notice Update – United States Outbreak of Listeria infections related to prepackaged caramel apples that may have been distributed in Canada [WWW Document]. URL http://www.phac-aspc.gc.ca/phn-asp/2014/Listeria-eng.php (accessed 10.28.16). Public Health Agency of Canada, 2016a. Yearly food-borne illness estimates in Canada [WWW Document]. URL https://www.canada.ca/en/public-health/services/food-borne-illness-canada/yearly-food-borne-illness-estimates-canada.html (accessed 2.22.17). Public Health Agency of Canada, 2016b. Public Health Notice Update – Outbreak of Listeria infections linked to packaged salad products produced at the Dole processing facility in Springfield, Ohio [WWW Document]. URL http://www.phac-aspc.gc.ca/phn-asp/2016/Listeria-eng.php (accessed 10.28.16). Ray, B., Bhunia, A., 2008. Fundamental Food Microbiology, 4th Editio. ed. Taylor Francis, Boca Raton, FL. Reddy, S.P., Wang, H., Adams, J.K., Feng, P.C.H., 2016. Prevalence and Characteristics of Salmonella Serotypes Isolated from Fresh Produce Marketed in the United States. J. Food Prot. 79, 6–16. Reid, A., 2009. MFHPB-20 - Isolation and identification of Salmonella from food and environmental samples. In: Compendium of Analytical Methods, Volume 2. Health Canada. 125  Rodriguez, A., Pangloli, P., Richards, H.A., Mount, J.R., Draughon, F.A., 2006. Prevalence of Salmonella in diverse environmental farm samples. J. Food Prot. 69, 2576–2580. Saldaña, Z., Sánchez, E., Xicohtencatl-Cortes, J., Puente, J.L., Girón, J.A., 2011. Surface structures involved in plant stomata and leaf colonization by Shiga-toxigenic Escherichia coli O157: H7. Front. Microbiol. 2, 1–9. Sanchez, S., Cenoz, M.G., Martin, C., Beristain, X., Llorente, M.T., Herrera-Leon, S., 2014. Cluster investigation of mixed O76:H19 Shiga toxin-producing Escherichia coli and atypical enteropathogenic E. coli infection in a Spanish household. Epidemiol. Infect. 142, 1029–1033. Sauders, B.D., Overdevest, J., Fortes, E., Windham, K., Schukken, Y., Lembo, A., Wiedmann, M., 2012. Diversity of Listeria Species in Urban and Natural Environments. Appl. Environ. Microbiol. 78, 4420–4433. Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R. V., Widdowson, M.A., Roy, S.L., Jones, J.L., Griffin, P.M., 2011. Foodborne illness acquired in the United States-Major pathogens. Emerg. Infect. Dis. 17, 7–15. Schaffter, N., Zumstein, J., Parriaux, A., 2004. Factors influencing the bacteriological water quality in mountainous surface and groundwaters. Acta Hydrochim. Hydrobiol. 32, 225–234. Scheutz, F., Teel, L.D., Beutin, L., Piérard, D., Buvens, G., Karch, H., Mellmann, A., Caprioli, A., Tozzoli, R., Morabito, S., Strockbine, N.A., Melton-Celsa, A.R., Sanchez, M., Persson, S., O’Brien, A.D., 2012. Multicenter evaluation of a sequence-based protocol for subtyping Shiga toxins and standardizing Stx nomenclature. J. Clin. Microbiol. 50, 2951–2963. Schlech, W.F., Lavigne, P.M., Bortolussi, R.A., Allen, A.C., Haldane, E.V., Wort, A.J., Hightower, A.W., Johnson, S.E., King, S.H., Nicholls, E.S., Broome, C. V, 1983. Epidemic Listeriosis — Evidence for Transmission by Food. N. Engl. J. Med. 308, 203–206. Seo, K.H., Frank, J.F., 1999. Attachment of Escherichia coli O157:H7 to Lettuce Leaf Surface and Bacterial Viability in Response to Chlorine Treatment as Demonstrated by Using Confocal Scanning Laser Microscopy. J. Food Prot. 62, 3–9. Sewell, A.M., Farber, J.M., 2001. Foodborne outbreaks in Canada linked to produce. J. Food Prot. 64, 1863–1877. Sharapov, U.M., Wendel, A.M., Davis, J.P., Keene, W.E., Farrar, J., Sodha, S., Hyytia-Trees, E., Leeper, M., Gerner-Smidt, P., Griffin, P.M., Braden, C., 2016. Multistate Outbreak of Escherichia coli O157:H7 Infections Associated with Consumption of Fresh Spinach: United States, 2006. J. Food Prot. 79, 2024–2030. Sheutz, F., Strockbine, N., 2005. Escherichia. In: Brenner, D., Krieg, N., Staley, J., Garrity, G. (Eds.), Bergey’s Manual of Systematic Bacteriology, Volume Two, Part B. Springer-Verlag New York, New York, pp. 607–623. 126  Shi, C., Singh, P., Ranieri, M.L., Wiedmann, M., Moreno Switt, A.I., 2013. Molecular methods for serovar determination of Salmonella. Crit. Rev. Microbiol. 7828, 309–325. Slayton, R.B., Turabelidze, G., Bennett, S.D., Schwensohn, C.A., Yaffee, A.Q., Khan, F., Butler, C., Trees, E., Ayers, T.L., Davis, M.L., Laufer, A.S., Gladbach, S., Williams, I., Gieraltowski, L.B., 2013. Outbreak of Shiga Toxin-Producing Escherichia coli (STEC) O157:H7 Associated with Romaine Lettuce Consumption, 2011. PLoS One 8, 6–11. Solomon, E.B., Yaron, S., Matthews, K.R., 2002. Transmission of Escherichia coli O157 : H7 from Contaminated Manure and Irrigation Water to Lettuce Plant Tissue and Its Subsequent Internalization Transmission of Escherichia coli O157 : H7 from Contaminated Manure and Irrigation Water to Lettuce Plant T. Appl. Environ. Microbiol. 68, 5–9. Stea, E.C., Purdue, L.M., Jamieson, R.C., Yost, C.K., Truelstrup Hansen, L., 2015. Comparison of the Prevalences and Diversities of Listeria Species and Listeria monocytogenes in an Urban and a Rural Agricultural Watershed. Appl. Environ. Microbiol. 81, 3812–22. Steele, M., Odumeru, J., 2004. Irrigation water as source of foodborne pathogens on fruit and vegetables. J Food Prot 67, 2839-2849. Strawn, L.K., Fortes, E.D., Bihn, E. a., Nightingale, K.K., Gröhn, Y.T., Worobo, R.W., Wiedmann, M., Bergholz, P.W., 2013a. Landscape and meteorological factors affecting prevalence of three food-borne pathogens in fruit and vegetable farms. Appl. Environ. Microbiol. 79, 588–600. Strawn, L.K., Gröhn, Y.T., Warchocki, S., Worobo, R.W., Bihn, E. a., Wiedmann, M., 2013b. Risk factors associated with Salmonella and Listeria monocytogenes contamination of produce fields. Appl. Environ. Microbiol. 79, 7618–7627. Swaminathan, B., Barrett, T.J., Hunter, S.B., Tauxe, R. V., 2001. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg. Infect. Dis. 7, 382–389. Tallon, P., Magajna, B., Lofranco, C., Leung, K.T., 2005. Microbial Indicators of Faecal Contamination in Water: A Current Perspective. Water. Air. Soil Pollut. 166, 139–166. Tataryn, J., Morton, V., Cutler, J., McDonald, L., Whitfield, Y., Billard, B., Gad, R., Hexemer, A., 2014. Outbreak of E. coli O157:H7 associated with lettuc served at fast food chains in the Maritimes and Ontario, Canada, Dec 2012. Canada Commun. Dis. Rep. 40, 2–9. Taylor, E. V, Nguyen, T.A., Machesky, K.D., Koch, E., Sotir, M.J., Bohm, S.R., Folster, J.P., Bokanyi, R., Kupper, A., Bidol, S.A., Emanuel, A., Arends, K.D., Johnson, S.A., Dunn, J., Stroika, S., Patel, M.K., Williams, I., 2013. Multistate outbreak of Escherichia coli O145 infections associated with romaine lettuce consumption, 2010. J. Food Prot. 76, 939–44. Thomas, M.K., Majowicz, S.E., Sockett, P.N., Fazil, A., Pollari, F., Doré, K., Flint, J.A., Edge, V.L., 2006. Estimated numbers of community cases of illness due to Salmonella, Campylobacter and verotoxigenic Escherichia coli: Pathogen-specific community rates. 127  Can. J. Infect. Dis. Med. Microbiol. 17, 229–234. Tienungoon, S., Ratkowsky, D. a, McMeekin, T. a, Ross, T., 2000. Growth limits of Listeria monocytogenes as a function of temperature, pH, NaCl, and lactic acid. Appl. Environ. Microbiol. 66, 4979–4987. Truchado, P., Lopez-Galvez, F., Gil, M.I., Pedrero-Salcedo, F., Alarcón, J.J., Allende, A., 2016. Suitability of different Escherichia coli enumeration techniques to assess the microbial quality of different irrigation water sources. Food Microbiol. 58, 29–35. Tseng, M., Fratamico, P.M., Manning, S.D., Funk, J.A., 2014. Shiga toxin-producing Escherichia coli in swine: the public health perspective. Anim. Heal. Res. Rev. 15, 63–75. U.S. Department of Agriculture, 2017a. Isolation and Identification of Salmonella from Meat, Poultry, Pasteurized Egg, and Siluriformes (Fish) Products and Carcass and Environmental Sponges. U.S. Department of Agriculture, 2017b. Isolation and Identification of Listeria monocytogenes from Red Meat, Poultry, Ready-To-Eat Siluriformes (Fish) and Egg Products, and Environmental Samples. U.S. Environmental Protection Agency, 2012. Method 1200: Analytical Protocol for Non-Typhoidal Salmonella in Drinking Water and Surface Water. U.S. Food and Drug Administration, 2015. Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption, Federal Register. United States Department of Agriculture, 2014. Detection and isolation of non-O157 Shiga Toxin-producing Escherichia coli (STEC) from meat products and carcass and environmental sponges. Uyttendaele, M., Jaykus, L.A., Amoah, P., Chiodini, A., Cunliffe, D., Jacxsens, L., Holvoet, K., Korsten, L., Lau, M., McClure, P., Medema, G., Sampers, I., Rao Jasti, P., 2015. Microbial Hazards in Irrigation Water: Standards, Norms, and Testing to Manage Use of Water in Fresh Produce Primary Production. Compr. Rev. Food Sci. Food Saf. 14, 336–356. Vail, J.H., Morgan, R., Merino, C.R., Gonzales, F., Miller, R., Ram, J.L., 2003. Enumeration of waterborne Escherichia coli with petrifilm plates: comparison to standard methods. J. Environ. Qual. 32, 368–373. Van Beneden, C.A., Keene, W.E., Strang, R.A., Werker, D.H., King, A.S., Mahon, B., Hedberg, K., Bell, A., Kelly, M.T., Balan, V.K., Mac Kenzie, W.R., Fleming, D., 1999. Multinational outbreak of Salmonella enterica serotype Newport infections due to contaminated alfalfa sprouts. JAMA 281, 158–62. Van Donkersgoed, J., Graham, T., Gannon, V., 1999. The prevalence of verotoxins, Escherichia coli 0157:H7, and Salmonella in the feces and rumen of cattle at processing. Can. Vet. J. 40, 332–338. 128  Vilar, M.J., Yus, E., Sanjuán, M.L., Diéguez, F.J., Rodríguez-Otero, J.L., 2007. Prevalence of and risk factors for Listeria species on dairy farms. J. Dairy Sci. 90, 5083–5088. Vivant, A.-L., Garmyn, D., Piveteau, P., 2013. Listeria monocytogenes, a down-to-earth pathogen. Front. Cell. Infect. Microbiol. 3, 1–10. Walker, S.J., Archer, P., Banks, J.G., 1990. Growth of Listeria monocytogenes at refrigeration temperatures. J. Appl. Bacteriol. 68, 157–162. Wang, S., Weller, D., Falardeau, J., Strawn, L.K., Mardones, F.O., Adell, A.D., Moreno Switt, A.I., 2016. Food safety trends: from globalization of whole genome sequencing to application of new tools to prevent foodborne diseases. Trends Food Sci. Technol. 57, 188–198. Wang, X., Taylor, M., Hoang, L., Ekkert, J., Nowakowski, C., Stone, J., Tone, G., Trerise, S., Paccagnella, A., Wong, T., Galanis, E., 2013. Comparison of clinical and epidemiological features of Shiga toxin-producing Escherichia coli O157 and non-O157 infections in British Columbia, 2009 to 2011. Can. J. Infect. Dis. Med. Microbiol. 24, e102-6. Ward, T.J., Ducey, T.F., Usgaard, T., Dunn, K.A., Bielawski, J.P., 2008. Multilocus genotyping assays for single nucleotide polymorphism-based subtyping of Listeria monocytogenes isolates. Appl. Environ. Microbiol. 74, 7629–7642. Watkins, J., Sleath, K.P., 1981. Isolation and Enumeration of Listeria monocytogenes From Sewage, Sewage-Sludge and River Water. J. Appl. Bacteriol. 50, 1–9. Welch, R.A., 2006. The Genus Escherichia. Prokaryotes Prokaryotic Biol. Symbiotic Assoc. 60–71. Weller, D., Wiedmann, M., Strawn, L.K., 2015. Irrigation Is Significantly Associated with an Increased Prevalence of Listeria monocytogenes in Produce Production Environments in New York State. J. Food Prot. 78, 1132–1141. Widiasih, D.A., Ido, N., Omoe, K., Sugii, S., Shinagawa, K., 2004. Duration and magnitude of faecal shedding of Shiga toxin-producing Escherichia coli from naturally infected cattle. Epidemiol. Infect. 132, 67–75. Wiedmann, M., 2002. Molecular subtyping methods for Listeria monocytogenes. J. AOAC Int. 85, 524–531. Wilkes, G., Edge, T., Gannon, V., Jokinen, C., Lyautey, E., Medeiros, D., Neumann, N., Ruecker, N., Topp, E., Lapen, D.R., 2009. Seasonal relationships among indicator bacteria, pathogenic bacteria, Cryptosporidium oocysts, Giardia cysts, and hydrological indices for surface waters within an agricultural landscape. Water Res. 43, 2209–2223. World Health Organization, 2011. Outbreaks of E. coli O104:H4 infection: update 30 [WWW Document]. URL http://www.euro.who.int/en/health-topics/emergencies/international-health-regulations/news/news/2011/07/outbreaks-of-e.-coli-o104h4-infection-update-30 129  (accessed 2.12.17). Zeng, W., Vorst, K., Brown, W., Marks, B.P., Jeong, S., Pérez-Rodríguez, F., Ryser, E.T., 2014. Growth of Escherichia coli O157:H7 and Listeria monocytogenes in packaged fresh-cut romaine mix at fluctuating temperatures during commercial transport, retail storage, and display. J. Food Prot. 77, 197–206.    130  Appendix Appendix A – Water and geographical data used for statistical analysis  Table A.1 – Measured data from each collected samples, including the presence of verotoxigenic Escherichia coli (V), Listeria monocytogenes (L), Salmonella (S), or any of the three (A); the temperature, pH, and total dissolved solids (TDS) of the water samples; and the concentration of fecal indicators generic E. coli (EC) and fecal coliforms (FC) measured using 3M™ Petrifilm (PF) and a membrane filtration method (MF).         / = data point not measured for that particular sample       Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 16/02/15 Sumas 1  + - - +  10.4 / /  / / / / 16/03/15 Sumas 1  + - - +  8.8 7.78 0.111  5.0 / 27.5 /  Sumas 2  - - - -  10.0 7.53 0.198  0.0 / 0.5 /  Sumas 3  - + - +  12.7 7.56 0.274  0.0 / 1.5 /  Serpentine 2  - + - +  10.8 6.66 0.111  3.5 / 15.5 / 08/04/15 Sumas 1  - - + +  9.3 / 0.289  1.0 / 1.5 /  Sumas 2  - - - -  8.9 / 0.377  0.0 / 0.0 /  Sumas 3  - - - -  9.5 / 0.107  0.0 / 0.0 /  Serpentine 2  - + - +  8.5 / 0.244  2.5 / 8.0 / 131  Table A.1 - Continued                    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 08/20/15 Sumas 1  - - - -  14.9 / 0.120  0.0 / 4.0 /  Sumas 2  - - - -  20.0 / 0.173  0.0 / 0.5 /  Sumas 3  - - - -  20.5 / 0.416  0.5 / 2.5 /  Serpentine 2  - - - -  16.5 / 0.127  0.0 / 12.0 / 04/05/15 Sumas 1  - - - -  15.0 7.77 0.347  4.0 / 5.5 /  Sumas 2  - - - -  15.9 7.41 0.346  0.0 / 2.0 /  Sumas 3  - - - -  18.5 7.64 0.192  2.0 / 2.0 /  Serpentine 2  + - - +  17.0 7.39 0.663  17.5 / 27.5 / 19/05/15 Sumas 1  - - - -  19.0 7.29 0.192  0.0 / 4.0 /  Sumas 2  - - - -  21.2 7.35 0.326  0.0 / 3.0 /  Sumas 3  - - - -  22.1 7.52 0.324  1.0 / 3.5 /  Serpentine 1  - - - -  19.0 7.22 0.359  0.5 / 3.5 /  Serpentine 2  - - - -  24.5 7.34 0.252  1.5 / 6.5 /  Serpentine 3  - - - -  19.2 7.62 0.167  0.0 / 4.5 / 01/06/15 Sumas 1  - - - -  20.0 7.75 0.541  2.5 / 11.5 /  Sumas 2  - - - -  21.2 7.70 0.255  1.0 / 4.5 /  Sumas 3  - + - +  21.5 8.00 0.319  2.0 / 5.0 /  Serpentine 1  - - - -  17.1 7.23 0.401  2.0 / 18.5 /  Serpentine 2  - - - -  20.9 7.59 0.191  1.0 / 2.5 /  Serpentine 3  - - - -  16.8 7.83 1.171  1.0 / 11.5 / 132  Table A.1 - Continued                    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 22/06/15 Sumas 1  - - - -  25.7 8.89 0.166  1.5 0.3 1.5 0.3  Sumas 2  - - - -  23.1 8.02 0.303  0.0 0.2 2.0 0.4  Sumas 3  - - - -  23.5 7.94 0.375  0.0 0.2 1.5 0.4  Serpentine 1  - - - -  15.3 7.68 0.350  1.0 1.0 6.0 1.0  Serpentine 2  - - - -  18.5 7.68 0.185  0.5 0.1 1.5 0.7  Serpentine 3  - - - -  19.0 7.77 1.222  2.0 0.6 15.0 2.2 07/07/15 Sumas 1  - - - -  23.5 7.73 0.210  0.0 0.3 3.5 0.0  Sumas 2  - - - -  24.1 7.73 0.311  0.0 0.2 2.0 0.0  Sumas 3  - - - -  24.5 7.85 0.366  0.0 0.2 8.0 0.0  Serpentine 1  - - - -  15.5 7.74 0.350  2.5 2.9 18.5 6.4  Serpentine 2  - - - -  19.2 7.62 0.304  1.0 1.2 5.5 0.1  Serpentine 3  - - - -  20.8 7.70 0.214  1.0 0.5 5.0 0.0 21/07/15 Sumas 1  - - - -  21.2 8.10 1.069  3.5 / 4.0 3.8  Sumas 2  - - - -  23.0 7.93 0.168  0.0 / 0.0 0.0  Sumas 3  - - - -  23.0 8.21 0.446  0.5 / 4.5 2.4  Sumas 4  - - - -  21.1 8.11 0.370  2.0 / 2.0 3.5  Serpentine 1  - - - -  13.9 8.10 0.338  4.5 / 7.5 5.1  Serpentine 2  - - - -  19.1 8.29 0.288  7.0 / 12.0 4.2  Serpentine 3  - - - -  18.6 8.06 0.173  3.5 / 5.5 1.5 133  Table A.1 - Continued                    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 10/18/15 Sumas 1  - - - -  20.0 7.45 0.652  1.0 0.3 1.5 2.1  Sumas 2  - - - -  20.5 7.38 0.159  0.5 0.3 1.0 0.3  Sumas 3  - - - -  21.9 7.72 0.170  0.0 0.1 0.5 0.4  Sumas 4  - - - -  19.8 7.49 0.292  0.0 0.1 0.0 0.8  Serpentine 1  - + - +  15.0 7.72 0.372  0.5 0.6 0.5 2.4  Serpentine 2  - - - -  19.9 7.84 0.106  0.5 0.5 1.0 0.6  Serpentine 3  - - - -  19.0 7.77 0.174  2.0 1.8 6.5 6.1 24/08/15 Sumas 1  - - - -  19.1 7.98 0.776  0.5 0.1 0.5 0.2  Sumas 2  - - - -  20.2 7.99 0.166  0.0 0.1 0.0 0.1  Sumas 3  - - - -  20.9 8.11 0.143  0.0 0.1 0.0 0.2  Sumas 4  - - - -  19.1 8.01 0.210  0.0 0.0 0.0 0.1  Serpentine 1  - - - -  14.0 8.11 0.164  0.5 0.0 0.5 0.1  Serpentine 2  - - - -  19.0 7.95 0.142  1.0 0.2 0.5 0.4  Serpentine 3  - - - -  18.1 7.97 0.327  0.0 0.1 0.0 0.1 03/09/15 Sumas 1  - - - -  15.0 7.59 0.429  5.0 3.3 21.0 4.8  Sumas 2  - - - -  15.8 7.62 0.117  9.0 1.4 20.0 0.4  Sumas 3  - - - -  16.5 7.92 0.137  0.0 0.4 2.0 1.1  Sumas 4  - - + +  16.0 7.81 0.207  0.0 0.3 6.0 0.8  Serpentine 1  - + + +  13.8 7.54 0.194  1.5 2.7 18.0 3.6  Serpentine 2  + - - +  15.0 7.37 0.134  4.0 5.3 33.5 7.6  Serpentine 3  - + - +  14.4 7.51 0.222  0.5 1.6 6.5 3.5 134  Table A.1 - Continued                    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 23/09/15 Sumas 1  - - - -  13.1 7.75 0.132  0.0 0.3 0.0 0.8  Sumas 2  - - - -  14.3 7.64 0.121  / / / /  Sumas 3  - - - -  13.3 8.01 /  0.0 0.0 3.0 0.1  Sumas 4  - - - -  14.8 7.83 /  1.0 1.3 3.0 2.6  Serpentine 1  - + - +  9.8 7.76 /  1.0 2.5 20.5 8.4  Serpentine 2  - - - -  13.6 7.50 /  1.5 1.3 5.0 3.1  Serpentine 3  - - - -  12.1 7.69 /  0.0 1.0 1.5 3.5 13/10/15 Sumas 1  - - - -  13.9 7.75 /  2.0 0.9 2.0 3.3  Sumas 2  - - - -  15.2 7.82 /  1.0 0.1 1.0 1.1  Sumas 3  - - - -  15.3 7.59 /  0.0 0.4 0.0 0.6  Sumas 4  - - - -  14.8 7.79 /  1.0 0.3 0.5 2.6  Serpentine 1  - - - -  13.6 7.63 /  0.5 0.4 0.5 1.2  Serpentine 2  - + - +  14.1 7.43 /  7.0 3.6 6.5 9.7  Serpentine 3  - + - +  14.1 7.51 /  6.0 3.2 6.0 9.3 09/11/15 Sumas 1  - - - -  9.0 7.74 0.073  0.0 0.5 1.5 1.4  Sumas 2  - - - -  8.8 7.76 0.124  0.0 0.2 0.0 0.8  Sumas 3  - - - -  7.8 8.05 0.166  0.5 0.1 0.5 0.6  Sumas 4  - - - -  7.9 7.85 0.269  0.0 0.3 0.0 0.4  Serpentine 1  - - - -  8.2 7.28 0.167  0.0 0.0 0.0 0.4  Serpentine 2  - + - +  8.9 3.78 1.222  0.0 0.0 0.0 0.1  Serpentine 3  - - - -  9.0 7.57 0.067  2.0 1.0 2.0 3.9 135  Table A.1 - Continued    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 24/11/15 Sumas 1  + + - +  7.4 7.44 0.067  3.0 1.4 5.0 2.8  Sumas 2  - - - -  7.1 7.57 0.103  0.0 0.1 0.0 0.1  Sumas 3  - - - -  4.9 7.88 0.154  0.0 0.1 0.0 0.1  Sumas 4  - - - -  5.1 7.60 0.224  1.0 0.9 1.0 0.9  Serpentine 1  - - - -  6.2 7.01 0.148  0.0 0.1 0.0 0.2  Serpentine 2  - - - -  6.3 6.26 0.456  0.0 0.0 0.0 0.0  Serpentine 3  - + - +  6.8 7.21 0.072  4.5 6.2 5.0 9.1 01/12/15 Sumas 1  - - - -  / 7.81 0.255  0.0 0.4 0.0 0.4  Sumas 2  - - - -  / 7.68 0.092  0.0 0.0 0.0 0.0  Sumas 3  - - - -  / 7.99 0.307  0.0 0.1 0.0 0.1  Sumas 4  - - - -  / 7.83 0.174  0.0 0.3 0.0 0.4  Serpentine 1  - - - -  / 7.68 0.068  0.0 0.1 0.0 0.1  Serpentine 2  - - - -  / 7.50 0.492  0.0 0.0 0.0 0.0  Serpentine 3  - - - -  / 8.03 0.279  1.0 2.0 1.0 2.6 15/12/15 Sumas 1  - - - -  7.2 7.61 0.090  0.0 0.4 0.0 0.6  Sumas 2  - - - -  9.9 7.88 0.189  0.0 0.0 0.0 0.0  Sumas 3  - - - -  8.0 7.84 0.256  0.0 0.2 0.0 0.2  Sumas 4  - - - -  7.5 7.57 0.135  0.0 0.0 0.0 0.2  Serpentine 1  - - - -  7.5 7.33 0.113  0.0 0.1 0.0 0.3  Serpentine 2  - + - +  7.9 5.60 0.444  0.0 0.1 0.0 0.4  Serpentine 3  - - - -  7.9 7.39 0.078  0.0 1.1 0.5 2.8 136  Table A.1 - Continued    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 12/01/16 Sumas 1  - - - -  5.9 7.34 0.165  1.5 1.1 1.5 1.4  Sumas 2  - - - -  5.0 7.09 0.081  0.0 0.0 0.0 0.0  Sumas 3  - - - -  2.7 7.38 0.207  0.0 0.0 0.0 0.1  Sumas 4  - - - -  4.2 7.52 0.208  2.5 3.9 5.5 5.4  Serpentine 1  - + - +  5.0 6.65 0.124  0.0 0.4 0.0 1.3  Serpentine 2  - - - -  4.8 6.50 0.335  0.0 0.5 0.0 0.9  Serpentine 3  - - - -  5.2 7.00 0.171  5.0 5.3 6.0 7.5 26/01/16 Sumas 1  - - - -  7.8 7.62 0.117  0.5 0.3 0.5 0.4  Sumas 2  - - - -  8.3 7.86 0.179  0.0 0.0 0.0 0.0  Sumas 3  - - - -  6.8 7.96 0.104  0.0 0.0 0.0 0.0  Sumas 4  - - - -  6.9 7.70 0.195  0.0 0.1 0.0 0.2  Serpentine 1  - + - +  8.1 7.79 0.135  11.5 6.3 15.5 11.1  Serpentine 2  - - - -  7.2 7.55 0.288  1.5 0.6 29.0 3.6  Serpentine 3  - - - -  8.0 7.76 0.079  6.5 3.5 8.5 5.4 02/02/16 Sumas 1  - - + +  6.8 7.49 0.070  0.5 0.9 1.0 1.6  Sumas 2  - - - -  7.1 7.44 0.155  0.0 0.0 0.0 0.1  Sumas 3  - - - -  4.8 7.66 0.193  0.0 0.0 0.0 0.0  Sumas 4  - - - -  5.2 7.38 0.114  0.0 0.2 0.0 0.2  Serpentine 1  - - - -  6.2 6.74 0.103  2.0 3.4 4.5 4.0  Serpentine 2  - + - +  6.2 6.55 0.322  0.0 0.1 0.5 0.6  Serpentine 3  - - - -  6.9 7.46 0.081  7.0 6.6 9.0 8.1 137  Table A.1 - Continued    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 16/02/16 Sumas 1  + - + +  8.9 7.24 0.077  3.5 5.0 4.0 8.1  Sumas 2  - + - +  8.8 7.36 0.139  2.0 1.5 2.0 2.5  Sumas 3  - + - +  8.9 7.36 0.088  1.5 1.8 1.5 3.9  Sumas 4  - - - -  8.8 7.14 /0.016  4.0 3.1 5.5 4.7  Serpentine 1  - - - -  7.5 6.51 0.031  1.5 1.1 1.5 1.8  Serpentine 2  + + - +  8.1 6.82 0.167  14.0 7.0 17.0 7.7  Serpentine 3  - - - -  8.1 7.29 0.064  1.5 2.3 2.0 1.9 0/03/16 Sumas 1  - - - -  8.0 7.07 0.122  0.0 1.4 0.5 1.6  Sumas 2  - - - -  7.0 7.12 0.187  0.0 0.1 0.0 0.1  Sumas 3  - - - -  7.5 6.98 0.301  1.0 0.4 1.0 0.4  Sumas 4  - - - -  7.8 6.91 0.212  0.0 0.1 0.0 0.1  Serpentine 1  - - - -  7.9 6.81 0.215  2.0 1.9 3.5 2.5  Serpentine 2  + + - +  7.9 6.22 0.298  2.0 4.2 2.0 4.2  Serpentine 3  - - - -  8.0 6.86 0.105  1.0 3.1 1.0 3.7 30/03/16 Sumas 1  - - - -  12.1 7.35 0.122  0.5 / 0.5 0.2  Sumas 2  - - - -  12.1 6.75 0.129  0.0 / 0.0 0.9  Sumas 3  - - - -  15.8 7.03 0.178  0.0 / 0.0 0.0  Sumas 4  - - - -  16.0 7.06 0.161  0.0 / 0.0 0.4  Serpentine 1  - - - -  9.0 6.84 0.163  0.0 / 0.0 1.8  Serpentine 2  - - - -  9.8 6.59 0.360  0.5 / 1.0 3.3  Serpentine 3  - - - -  10.9 7.08 0.136  1.5 / 1.5 1.1 138  Table A.1 - Continued    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 14/04/16 Sumas 1  - - - -  10.1 7.72 0.144  0.5 1.8 0.5 2.3  Sumas 2  - - - -  10.4 7.37 0.165  0.0 0.0 0.0 0.0  Sumas 3  - - - -  10.5 7.62 0.530  1.0 1.9 1.5 4.0  Sumas 4  - - - -  11.0 7.67 0.191  0.5 1.1 0.5 1.7  Serpentine 1  - - - -  9.8 7.17 0.191  0.0 0.4 0.5 0.6  Serpentine 2  + - - +  10.4 7.17 0.480  2.5 3.8 5.5 8.3  Serpentine 3  - + - +  10.2 7.37 0.141  15.0 7.2 15.0 9.2 26/04/16 Sumas 1  - - - -  14.9 7.66 0.156  1.0 1.0 0.5 1.3  Sumas 2  - - - -  15.1 7.39 0.149  1.0 0.6 1.0 2.1  Sumas 3  - - - -  22.4 8.21 0.535  0.0 0.7 0.0 1.7  Sumas 4  - - - -  13.6 7.51 0.137  1.0 0.9 1.0 1.0  Serpentine 1  - - - -  14.9 6.95 0.216  0.0 0.1 0.0 0.3  Serpentine 2  - - - -  16.2 7.30 0.643  4.5 3.5 4.0 7.9  Serpentine 3  - - - -  12.7 7.75 0.141  1.5 1.5 1.5 1.7 02/05/16 Sumas 1  - - - -  15.9 7.37 0.114  1.0 0.8 1.5 2.2  Sumas 2  - - - -  17.1 7.31 0.125  0.5 0.4 0.5 0.7  Sumas 3  - - - -  19.5 7.25 0.156  0.5 0.4 0.5 1.6  Sumas 4  - - - -  14.7 7.07 0.118  1.5 1.0 1.5 2.0  Serpentine 1  - - - -  15.2 6.84 0.207  0.0 0.0 0.0 0.6  Serpentine 2  - - - -  18.0 7.01 0.691  0.5 1.7 0.5 1.9  Serpentine 3  - - - -  16.0 7.43 0.155  2.5 2.3 3.5 4.2 139  Table A.1 - Continued    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 17/05/16 Sumas 1  - - - -  15.9 7.83 0.143  0.5 0.3 0.5 0.6  Sumas 2  - - - -  17.7 7.92 0.191  0.5 0.3 0.5 1.0  Sumas 3  + - - +  17.0 7.87 0.195  0.0 0.2 0.0 0.4  Sumas 4  - - - -  15.3 7.72 0.137  0.0 0.2 0.0 0.3  Serpentine 1  - - - -  13.6 7.29 0.213  0.0 0.1 0.5 0.2  Serpentine 2  - - - -  15.8 7.43 0.706  0.0 0.1 0.0 0.1  Serpentine 3  - - - -  14.0 7.82 0.135  42.0 / 42.0 / 07/06/16 Sumas 1  - - - -  21.2 7.61 0.325  0.0 0.3 0.5 1.0  Sumas 2  - - - -  22.1 7.62 0.218  0.0 0.3 0.0 0.9  Sumas 3  - - - -  19.9 7.71 0.316  3.0 2.1 4.5 3.9  Sumas 4  - - - -  19.9 7.57 0.123  0.5 0.2 1.5 0.5  Serpentine 1  - - - -  16.0 7.27 0.235  1.5 0.5 2.0 1.3  Serpentine 2  - - - -  20.2 7.23 0.520  1.5 1.0 2.0 1.7  Serpentine 3  - - - -  18.9 7.59 0.135  3.5 2.2 4.0 4.0 20/06/16 Sumas 1  - - - -  14.9 7.50 0.296  1.5 3.5 1.5 2.3  Sumas 2  - - - -  17.1 7.71 0.378  2.0 0.2 0.0 0.2  Sumas 3  - - - -  15.4 7.66 0.339  12.0 0.3 0.0 0.2  Sumas 4  - - - -  17.5 7.36 0.107  1.5 1.2 0.0 0.5  Serpentine 1  - - - -  19.6 7.38 0.193  71.5 / 49.0 /  Serpentine 2  - - - -  21.0 7.42 0.438  4.0 3.6 4.0 2.7  Serpentine 3  - - - -  16.5 6.85 0.133  60.5 / 25.5 / 140  Table A.1 - Continued    Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 05/07/16 Sumas 1  - - - -  16.9 7.47 0.113  2.5 1.1 2.5 0.5  Sumas 2  - - - -  17.3 7.67 0.345  1.5 0.9 0.5 0.6  Sumas 3  - + - +  17.0 7.63 0.415  1.0 0.3 0.5 0.3  Sumas 4  - - - -  16.9 7.48 0.135  1.0 0.5 1.0 0.4  Serpentine 1  - - - -  14.4 7.75 0.180  1.5 2.5 1.5 1.7  Serpentine 2  - + - +  17.2 7.31 0.357  7.5 5.9 6.5 3.9  Serpentine 3  - - - -  16.0 7.97 0.079  1.5 1.7 1.5 1.3 19/07/16 Sumas 1  - - - -  19.2 7.57 0.170  0.5 1.0 1.5 3.0  Sumas 2  - - - -  20.9 7.79 0.430  0.0 0.2 0.5 0.2  Sumas 3  - - - -  20.5 7.75 0.471  0.0 0.4 1.5 0.5  Sumas 4  - - - -  18.1 7.60 0.144  0.5 0.9 1.0 1.4  Serpentine 1  - - - -  15.8 7.67 0.180  3.5 3.0 4.0 3.5  Serpentine 2  - - - -  17.1 7.67 0.289  1.0 0.2 1.0 0.3  Serpentine 3  - - - -  18.8 7.80 0.158  2.5 2.6 2.5 4.4 08/08/16 Sumas 1  - - - -  15.8 7.57 0.419  0.5 1.0 1.5 3.0  Sumas 2  - - - -  20.2 7.79 0.254  0.0 0.2 0.5 0.2  Sumas 3  - - - -  18.8 7.75 0.308  0.0 0.4 1.5 0.5  Sumas 4  - - - -  21.1 7.60 0.111  0.5 0.9 1.0 1.4  Serpentine 1  - - - -  21.7 7.67 0.159  3.5 3.0 4.0 3.5  Serpentine 2  - - - -  22.2 7.67 0.533  1.0 0.2 1.0 0.3  Serpentine 3  - - - -  19.5 7.80 0.113  2.5 2.6 2.5 4.4 141  Table A.1 - Continued        Pathogens  Water Characteristics  Fecal Indicators Sampling Date Site  V L S A  Temp. (°C) pH TDS (mg/ml)  EC PF (CFU/ml) EC MF (CFU/ml) FC PF (CFU/ml) FC MF (CFU/ml) 22/08/16 Sumas 1  - - - -  19.8 7.72 0.320  2.0 1.0 4.0 1.8  Sumas 2  - - - -  20.9 7.81 0.340  1.0 0.4 1.0 0.5  Sumas 3  - - - -  20.2 7.54 0.348  0.0 0.1 0.0 0.1  Sumas 4  - - + +  19.8 7.37 0.131  2.5 1.4 2.5 2.5  Serpentine 1  - + - +  15.1 7.21 0.170  0.0 0.9 0.0 0.8  Serpentine 2  - - - -  17.8 7.39 0.441  0.5 0.1 0.5 0.7  Serpentine 3  + - - +  17.1 7.52 0.136  1.0 0.5 1.0 0.7 142  Table A.2 – Data used to measure proximity and density of upstream livestock.  Serpentine Sumas  Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Nearest Livestock (km)A       CowB 0.93 0.423 1.46 1.17 1.79 1.15 PoultryC - 1.34 - 0.93 1.48 3.10 OtherD - - - - - 0.08 Number within 1 kmE       Cow 1 1 1 0 0 0 Poultry 0 0 0 1 0 0 Other 0 0 0 0 0 1 Number within 2 km       Cow 1 3 1 2 1 3 Poultry 0 2 0 2 1 0 Other 1 0 0 0 0 1 Number within 3 km       Cow 1 3 1 3 2 3 Poultry 0 2 0 4 1 0 Other 1 0 0 0 0 1 Border within 1 kmF (km)       Cow 0.214 1.153 0 0 0 0 Poultry 0 0 0 0.078 0 0 Other 0 0 0 0 0 0.318 Border within 2 km (km)       Cow 0.214 2.36 1.838 0.450 0.368 1.465 Poultry 0 0.119 0 0.719 0.152 0 Other 0 0 0 0 0 0.318 Border within 3 km (km)       Cow 0.214 2.904 1.905 1.133 1.203 3.008 Poultry 0 0.412 0 0.792 0.141 0 Other 0 0 0 0 0 0.318 ANearest distance for surface water to travel BIncludes both dairy and beef cattle CIncludes chicken and turkey DIncludes swine, sheep, or goats ENumber of properties bordering the surface water within a given distance upstream FTotal length of waterway bordered by properties containing livestock within a given distance 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0343449/manifest

Comment

Related Items