EVALUATION OF SUBTYPING METHODS FOR THE CHARACTERIZATION OF CAMPYLOBACTER STRAINS FROM DIFFERENT GEOGRAPHICAL AREAS Except where reference is made to the work of others, the work described in this thesis is my own or was done in collaboration with my advisory committee. This thesis does not include proprietary or classified information. Robert Scott Miller Certificate of Approval: Jong Wook Hong Assistant Professor Materials Engineering Sandra Ewald Professor Poultry Science Pathobiology Omar A. Oyarzabal, Chair Associate Professor Poultry Science George T. Flowers Dean Graduate School EVALUATION OF SUBTYPING METHODS FOR THE CHARACTERIZATION OF CAMPYLOBACTER STRAINS FROM DIFFERENT GEOGRAPHICAL AREAS Robert Scott Miller A Thesis Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Master of Science Auburn, Alabama December 19, 2008 iii EVALUATION OF SUBTYPING METHODS FOR THE CHARACTERIZATION OF CAMPYLOBACTER STRAINS FROM DIFFERENT GEOGRAPHICAL AREAS Robert Scott Miller Permission is granted to Auburn University to make copies of this thesis at its discretion, upon request of individuals or institutions and at their expense. The author reserves all publication rights. Signature of Author Date of Graduation iv VITA Robert Scott Miller graduated from the College of Science and Mathematics at Auburn University with a Bachelor of Science degree in Microbiology in August 2003. After degree completion he worked as a Research Assistant II from October 2003 until the present date in the Poultry Science Department?s Microbiology Laboratory at Auburn University. Robert married Jennifer Jones in September 1995. They have two children, Robert Sean Miller and Elizabeth Nicole Miller. v THESIS ABSTRACT EVALUATION OF SUBTYPING METHODS FOR THE CHARACTERIZATION OF CAMPYLOBACTER STRAINS FROM DIFFERENT GEOGRAPHICAL AREAS Robert Scott Miller Master?s of Science, December 19, 2008 (B.S., Auburn University, 2003) 112 Typed Pages Directed by Omar A. Oyarzabal Bacterial foodborne diseases are a common cause of lost productivity for industrially developed nations. In developing nations, these illnesses can be deadly for the young, elderly, and immunocompromised. The ability to accurately characterize the bacteria responsible for foodborne disease is a very important tool for researchers and public health providers. Reported data shows that Campylobacter is one of the most common causes of acute bacterial enteritis globally. The purpose of this study was to evaluate molecular typing techniques used to characterize Campylobacter strains. To this end, 30 C. jejuni strains from Puerto Rico, 56 C. jejuni and C. coli strains from Grenada, and 55 C. jejuni and C. coli strains from the southeastern US were evaluated with pulse-field gel vi electrophoresis (PFGE), multilocus sequence typing (MLST), and flagellin A restriction fragment length polymorphism (flaA-RFLP). FlaA-RFLP was the less expensive method and required less labor than PFGE, but the discriminatory power was low and C. jejuni and C. coli strains grouped together after the analysis. PFGE proved to be the most discriminatory, but most labor intensive, method. PFGE?s limitation as a characterization technique is the failure to restrict some Campylobacter genomes due to DNA methylation or lack of restriction site. MLST, while lacking PFGE?s discriminatory power, can characterize all strains and may provide the best tool for characterization of Campylobacter in temporal, epidemiological studies. This research suggests that a combination of two of the three techniques, depending on the purpose of the characterization, is the best approach when employing molecular typing methods to characterize C. jejuni and C. coli. vii ACKNOWLEDGMENTS My special thanks to Dr. Omar A. Oyarzabal for providing the opportunity and much needed encouragement to earn my master?s degree. I also want to thank my committee members, Dr. Sandra J. Ewald and Dr. Jong Wook Hong for their time and support. Thanks to Fernando Rebollo-Carratto for providing broiler fecal samples from Peurto Rico, Harry Hariharan for Campylobacter strains from Grenada, Dr. Lee Zhang of the Auburn University Genetics Analysis Laboratory, and Nick Hermersmann of Lucigen Corporation for providing DNA sequencing. I would be remiss if I did not thank the following students that worked alongside me in the laboratory for their assistance: Sarah Pierce, Leslie Speegle, Ami Sumners, Megan Behringer, and Anna Stanford. Your help was invaluable. There is no way to express my thanks to my wife, Jennifer, and our children, Sean and Elizabeth, for their support and love. This graduate program has been a labor of love for all of us. Thank you for your understanding. I dedicate this thesis and my degree to my parents, Robert Stanford and Kathy Diane Miller. You built the foundation that led to this accomplishment the first time you told me that I could accomplish anything I was willing to work toward. You have always made the most of every blessing you received. Thank you for passing that on to Tim and me. But perhaps most importantly thanks for loving me through everything, even when I would not listen to your advice. viii Style manual or journal used Journal of Food Protection. Computer software used Microsoft Office 2007, Bionumerics version 4.50 (Applied Maths). ix TABLE OF CONTENTS LIST OF TABLES x LIST OF FIGURES xi I. LITERATURE REVIEW 1 II. MOLECULAR TYPING OF CAMPYLOBACTER JEJUNI STRAINS ISOLATED FROM COMMERCIAL BROILERS IN PUERTO RICO 15 III. DNA CHARACTERIZATION OF CAMPYLOBACTER JEJUNI AND CAMPYLOBACTER COLI ISOLATES FROM CHICKENS IN GRENADA 44 IV. FINGERPRINT PROFILING OF CAMPYLOBACTER JEJUNI AND CAMPYLOBACTER COLI ISOLATED FROM RETAIL BROILER MEAT IN ALABAMA 62 V. EVALUATION OF FINGERPRINT PROFILING OF CAMPYLOBACTER JEJUNI AND CAMPYLOBACTER COLI FROM DIFFERENT GEOGRAPHICAL AREAS USING PULSED-FIELD GEL ELECTROPHORRESIS 81 REFERENCES 88 x LIST OF TABLES I. 1. CAMPYLOBACTER TAXA (2007) 14 II. 1. Allellic profiles and clonal complexes from C. jejuni strains isolated in this study 41 2 Index of association calculated with START and the standardized index of association calculated with LIAN showing that the tested population of C. jejuni is in linkage disequilibrium 42 3 PFGE macrorestriction profiles, ST allelic profiles of C. jejuni strains isolated from five different farms 43 III. 1. Identification of 17 Campylobacter isolates with discrepancies between biochemical tests and multiplex polymerase chain reaction (mPCR) assays 60 2. Evaluation of the discriminatory power of different typing techniques applied to Campylobacter spp. isolates from Grenada 61 IV. 1. Positive samples from plants A and B 77 2. Allellic profiles and clonal complexes from C. jejuni and C. coli strains isolated in this study 78 3. Number of profiles per technique per year 79 4. Number of profiles per technique per processing plant 80 V. 1 Band sizes and the sum of all bands for each fingerprinting profile collected with flaA-RFLP 85 xi LIST OF FIGURES I I. 1A. Lineage assignment using the UPGMA cluster analysis calculated with START version 1.0.8. 38 1B. Decomposition analysis on allelic profile data using SplitsTree 38 2. PFGE macrorestriction profiles (SmaI/KpnI) of C. jejuni isolates belonging to ST-2624 39 3. flaA-RFLP restriction profiles of all Puerto Rico strains 40 III. 1. PFGE macrorestriction and flaA-RFLP restriction of strains G185 and G187 58 2. flaA-RFLP restriction profiles of C. coli and C. jejuni isolates from Grenada 59 IV. 1. PFGE dendogram of C. jejuni and C. coli from plant A and B (2005-2007) 75 2. flaA-RFLP dendogram of C. jejuni and C. coli from plants A and B (2005-2007) 76 V. 1. PFGE macrorestriction profiles (SmaI) of C. jejuni and C. coli isolates from Puerto Rico, Grenada, and southeastern US 87 1 I. LITERATURE REVIEW Foodborne Diseases Foodborne diseases are illnesses caused by infectious or toxic substances that enter the body through ingestion of food or beverage. Over 250 foodborne diseases are known, many of which are caused by microbial pathogens, such as bacteria, viruses, fungi, protozoa, and multicellular parasites. The Centers for Disease Control and Prevention (CDC) estimates 76 million cases of foodborne disease every year in the US, leading to 325,000 hospitalizations and 5,000 deaths (Anonymous. 2005). The estimated cases of foodborne illness in the US caused by known pathogens are listed in Table 1. Worldwide estimates of foodborne diseases are difficult to determine, but the World Health Organization (WHO) estimates that 1.8 million deaths worldwide are attributed to foodborne diseases every year (Anonymous. 2007). Between 1993 and 1997, the CDC reported 2,751 outbreaks of foodborne diseases (Lynch 2006), with an increase in the number of outbreaks to 6,647 between 1998 and 2002 (Olsen 2002). Bacteria were the most common cause of foodborne outbreaks in both studies, with an incidence of 55% in the period 1998-2002 and 75% in the period 1993-1997 (Lynch 2006; Olsen 2002). It is estimated that foodborne disease outbreaks linked to five bacterial pathogens, Campylobacter, Salmonella, Escherichia coli O157:H7, E. coli non-O157:H7STEC, and Listeria, lead to approximately $6.9 billion in 2 medical costs, lost productivity, and premature death every year in the US alone (Buzby 2001). The Campylobacter Genus In 1866, Theodore Escherich described nonculturable spiral bacteria found in stool specimens and large intestinal mucosa of neonates and kittens suffering from diarrhea. The failure of the bacteria to grow on solid media after microscopic detection and typical morphology associated with gastroenteritis lead many researchers to believe that the bacteria Escherich described in 1866 was indeed Campylobacter spp. (Kist 1986). McFadyean and Stockmen, veterinary surgeons in Great Britain, reported a bacterium that resembled a vibrio in aborted ovine fetuses. The scientific name, Vibrio fetus, would not appear for another decade. Over the years, more species and subspecies of Vibrio were described, with isolation from human blood occurring in 1938, 1947, and 1958 (Butzler 2004). In 1963, the Vibrio fetus and Vibrio jejuni were transferred to the newly created Campylobacter genus (On 2001). The Campylobacter genus is currently comprised of 16 species and 6 subspecies (http://www.the-icsp.org/taxa/ Campylobacterlist.htm). Table 2 shows the current taxonomic listing for the Campylobacter genus. Epidemiology of Campylobacter Campylobacter spp. has been isolated from various wild and domesticated animals. Specifically, C. jejuni and C. coli have been shown to cause disease in cats, dogs, ferrets, mink, primates, hamsters, guinea pigs, rats, sheep, pigs, and cattle. While C. jejuni rarely 3 causes disease in birds, it appears that birds have a higher carriage rate of C. jejuni than other animals (Young and Mansfield 2005). Due to this domestic poultry are the most important C. jejuni reservoir for humans with chicken being the primary source for human infection (Allos 2001). Research to reduce Campylobacter colonization in chicken has been stymied by the lack of an obvious route of transmission to negative flocks. A study by Callicott et al. in 2006 is one of several that suggest that vertical transmission is not responsible for infection of broiler flocks. The horizontal transmission from environmental sources has been suspected for sometime but are not very well understood. Using molecular techniques to determine strain relation can be difficult due to C.jejuni's weak clonality and inter- and intraspecies horizontal genetic exchange (Suerbaum 2001, Dingle 2001). Also contributing to the difficulties of understanding Campylobacter epidemiology is the lack of a single molecular subtyping method capable of providing comprehensive characterization with a proper degree of discrimination for all C. jejuni and C. coli strains while targeting appropriate genetic variation for a given study (i.e. relationship of recent outbreak strain versus longer term evolutionary studies). It is essential to understand a few terms common to epidemiology and molecular subtyping before proceeding into a discussion of molecular subtyping methods. An isolate is a population of microbial cells in pure culture derived from a single colony on an isolation plate and identified to the species level (Riley 2004, text). A strain is an isolate or group of isolates exhibiting phenotypic and/or genotypic traits belonging to the same lineage, distinct from those of other isolates of the same species 4 (Riley 2004, text). A clone is an isolate or group of isolates descending from a common precursor strain by nonsexual reproduction exhibiting phenotypic or genotypic traits characterized by a strain-typing method to belong to the same group (Riley 2004, text). Methods for molecular subtyping of bacteria Molecular subtyping is the characterization of proteins or nucleic acids to aid in epidemiological investigations (Swaminathan 2001). Researchers use molecular subtyping to determine the relatedness of bacterial strains. Understanding strain relatedness aids in taxonomic classification of organisms into related groups, phylogenetic analysis or evolutionary development of an organism, and epidemiological studies of the distribution and determinants of disease (Riley 2004). Currently practiced molecular subtyping techniques can be organized into four different categories: hybridization and amplification, amplification and restriction, restriction and migration, and sequencing. Polymerase chain reaction (PCR) is an example of hybridization and amplification. Flagellin A restricted fragment length polymorphism (flaA-RFLP) utilizes amplification and restriction. Pulsed field gel electrophoresis (PFGE) relies on restriction and migration while multilocus sequence typing (MLST) is based on sequencing. To understand how molecular subtyping techniques are evaluated some important terms must be defined. Typeability is the proportion of strains that is unambiguously typed by the subtyping method. Reproducibility is the percent of strains classified as the same subtype upon repeated testing. Discriminatory power is the ability of a technique to generate discrete units of information from different isolates at the sub-species level 5 (Riley 2004). The discriminatory power (Di) of a method is based on its ability to differentiate between unrelated strains and is determined by the number of types assigned by the method and the relative frequency of those types (Hunter 1988). Since the early 1980s, many techniques for molecular subtyping have been proposed and used with differing levels of success. This thesis will present the results of the evaluation of three subtyping techniques, flaA-RFLP, PFGE, and MLST applied to Campylobacter jejuni and C. coli strains from different geographical areas. Flagellin A restriction fragment length polymorphism (flaA-RFLP) FlaA-RFLP is a typing system based on both sequence variation and restriction. FlaA- RFLP was one of the first subtyping techniques used on Campylobacter spp. (Nachamkin et al., 1993). FlaA-RFLP involves amplification with a polymerase chain reaction assay (PCR) and subsequent digestion of the flagellin A subunit gene with DdeI. After standard electrophoresis separation, the bands are stained and visualized with ultraviolet light. FlaA-RFLP has several advantages, such as low cost, simple PCR protocols for amplification and use of traditional electrophoresis for the capturing of the results. One of the disadvantages of this technique is the low discriminatory power, which is usually less than PFGE (Neilsen 2000). Another disadvantage relates to the inconsistencies in the mobility of the DNA fragments in the gel, a limitation of any technique based on the separation of the fragments using gel electrophoresis (Meinersmann 1997). 6 Pulsed-field gel electrophoresis (PFGE) PFGE typing relies on macrorestriction of the entire bacterial genome by infrequent cutting endonucleases. In the case of C. jejuni, the genome size is ~1.7 Mb and several enzymes can produce discernable bands by this technique. In the US, SmaI is the endonuclease most commonly used to restrict Campylobacter strains (Ribot, 2001), and is the endonuclease preferred by CDC?s Pulsenet PFGE protocol for Campylobacter jejuni (Anonymous 2008). CampyNet, the European counterpart of Pulsenet, initially proposed the use of KpnI as the primary endonuclease for PFGE digestion (Michaud 2001), but now it recommends the restriction with SmaI (Gilpin 2006). Until 1984, researchers were unable to differentiate fragments of DNA larger than 40 kilobases by standard electrophoresis methods (Schwartz and Cantor, 1984). By applying voltage gradients at various angles and for various lengths of time, Schwartz and Cantor proved that large fragments of DNA can move through an agarose gel made at higher concentrations than traditional gel electrophoresis. Two years later, the countour-clamped homogeneous electric field (CHEF) apparatus was designed by Chu et al. (1986). This machine allows for 120-degree-angle reorientation of the DNA bands between 24 electrodes organized as a hexagon as they travel through the gel. PFGE has since been used to type many pathogenic and nonpathogenic microorganisms, and is considered the ?gold standard? by epidemiologists for bacterial subtyping. Two advantages of typing with PFGE are a high level of discrimination and reproducibility among inter and intra-laboratory patterns when following the standardized protocols. Some of the disadvantages are the cost of the equipment, the time required to 7 obtain the results, the high training needed to run the technique, and the fact that some strains do not restrict with current enzymes used in standard PFGE protocols. Multilocus sequence typing (MLST) Developed by Maiden et. al (1997), MLST is a variation on the multilocus enzyme electrophoresis (MLEE) technique developed in the 1980s (Maiden 1998). MLST operates under the same philosophy as MLEE by indexing neutral genetic variation from multiple chromosome locations (alleles) while exploiting nucleotide sequencing to identify this variation (Dingle 2001). The advantage of MLST over MLEE is that it determines variation in internal fragments of allele sequences rather than determining allele variation by enzyme mobilities through electrophoresis. MLEE, like other traditional and molecular typing schemes, is not portable because the indexed variation is difficult to compare among laboratories. MLST takes advantage of the electronic portability of sequence data, and allows for exchange, over the Internet, of molecular typing data for global epidemiology between laboratories (Maiden 1998). For Campylobacter typing, MLST involves the sequencing of internal fragments of seven housekeeping genes. The sequences are compared with previously identified sequences, or alleles, at that locus and are assigned allele numbers at each of the seven loci. The seven allele numbers define the strain's allelic profile. Each allelic profile is assigned a sequence type (ST) which is used to describe the strain (Aanensen 2005). Sequence types are then grouped in clonal complexes based on their sequence similarities (Miller 2005). Housekeeping loci evolve slowly because they are under stabilizing selection for conservation of metabolic function. Therefore, by indexing nucleotide sequence 8 variations, MLST provides highly discriminated data, and also enables the population structure of an organism to be established. Consequently, these data are not sensitive to the frequent chromosome rearrangements seen in C. jejuni, which adversely affects typing techniques such as PFGE (Dingle 2002). MLST offers several advantages over other subtyping methods. Minimal experimental variation in results facilitates interlaboratory comparisons and the establishment of a standard nomenclature for DNA sequence data. Unlike gel images, sequence data provide information on the precise differences between strains (Sails 2003). MLST also requires much less time to perform than PFGE, the current gold standard for epidemiological studies, but perhaps the greatest benefit of MLST is its ability to type all strains. Because MLST employs sequence variation in housekeeping genes, it is best suited for long term epidemiology studies, since it identifies variation building slowly within a population (Dingle 2001). Therefore MLST, when used alone, has less discriminatory power than PFGE for short term epidemiological studies. According to Sails et al. (2003), MLST?s DI=0.859 is lower than PFGE?s DI=0.930. Analysis of molecular subtyping data There are two major approaches for the classification of strains, cladisitic and phenetic. The cladistic approach determines evoluntionary relationships of strains based on shared derived characteristics in extant taxa. The phenetic approach groups strains based on similarities or differences without drawing evolutionary connections. Mathematical 9 procedures can then be used to evaluate characteristics that are numerically encoded (Riley 2004b). The phenetic approached will be used in this thesis. Both phenetic and cladistic approaches order each taxon in terms of its character states which represent a range of particular characters of a specific taxon. The basic unit of the phenetic approach is the OTU or operational taxonomic unit. An OTU is defined as a strain or group of strains specified by a particular set of character states. OTUs similarities or distances can be determined by applying mathematical models (Riley 2004b). Cluster analysis or the analysis of relations between many strains, rather than pairs of strains, allows for the grouping of strains by a heirarchic structure into phenograms. In MLST, cluster analysis is achieved through nearest-neighbor clustering (also called neighbor-joining clustering). Nearest neighbor cluster analysis constructs phenograms using a similarity matrix derived from sequential pairwise comparisons of the two most similar OTUs or clusters of OTUs (Riley 2004b). Electrophoretic fingerprint patterns, like those derived from PFGE and RFLP, are most commonly clustered using unweighted pair-group method using arithmetic averages (UPGMA), although the neighbor-joining clustering can also be used. UPGMA creates phenogams by calculating the average distance between one cluster of OTUs to a comparison cluster, where each OTU is given equal weight (Riley 2004b). Throughout this thesis, UPGMA analysis will be used to evaluate PFGE and Fla-A-RFLP data, while neighbor-joining will be used for the analysis of MLST data. 10 References Aanensen, D. M., B. G. Spratt. 2005. The multilocus sequence typing network: mlst.net. Nucleic Acids Res. 33:W728-W733. Allos, B. M. 2001. Campylobacter jejuni infections: update on emerging issues and trends. Clin. Infect. Dis. 32:1201-1206. Anonymous. 2005. Foodborne illnesses. CDC Oct. 25. Coordinating Center for Infectious Diseases /Division of Bacterial and Mycotic Diseases. http://www.cdc.gov/ncidod/dbmd/diseaseinfo/foodborneinfections_g.htm#howma nycase. Anonymous. 2007. Food safety and foodborne illness?. WHO Fact sheet No. 237. March. http://www.who.int/mediacentre/factsheets/fs237/en/. Anonymous. 2008. One-day (24-26 h) standardized laboratory protocol for molecular subtyping of Campylobacter jejuni by pulsed field gel electrophoresis. CDC. http://www.cdc.gov/pulsenet/protocols/campy_protocol.pdf. Buzby, J. C. 2001. Children and Microbial Foodborne Illness. Food Review 24:32-37. Butzler, J. -P. 2004. Campylobacter, from obscurity to celebrity. Clin. Microbiol. Infec. 10:868-876. Callicott, K. A., V. Fri?riksd?ttir, J. Reiersen, R. Lowman, J. Bisaillon, E. Gunnarsson, E. Berndtson, K. L. Hiett, D. S. Needleman, and N. J. Stern. 2006. Lack of Evidence for Vertical Transmission of Campylobacter spp. in Chickens. Appl. Environ. Microb. 72:5794-5798. 11 Chu, G., D. Vollrath, and R. W. Davis. 1986. Separation of large DNA molecules by contour-clamped homogeneous electric fields. Science 234:1582-1585. Dingle, K. E., F. M. Colles, D. R. A. Wareing, R. Ure, A. J. Fox, F. E. Bolton, H. J. Bootsma, R. J. L. Willems, R. Urwin, and M. C. J. Maiden. 2001. Multilocus sequence typing system for Campylobacter jejuni. J. Clin. Microbiol. 39:14-23. Dingle, K. E., F. M. Colles, R. Ure, J. A. Wagenaar, B. Duim, F. J. Bolton, A. J. Fox, D. R. A. Wareing, and M. C.. J. Maiden. 2002. Molecular characterization of Campylobacter jejuni clones: A basis for epidemiologic investigation. Emerg. Infect. Dis. 8:949-955. Gilpin, B., A. Cornelius, B. Robson, N. Boxall, A. Ferguson, C. Nicol, and T. Henderson. 2006. Application of pulsed-field gel electrophoresis to identify potential outbreaks of campylobacteriosis in New Zealand. J. Clin. Microbiol. 44:406-412. Hunter, P. R. and M. A. Gaston. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-2466. Kist. M. 1986. Who discovered Campylobacter jejuni/coli? A review of hitherto disregarded literature. Zentralbl Bakteriol. Mikrobiol. Hyg. [A]. 261:177-186. Lynch, M., J. Painter, R. Woodruff, C. Braden. 2006. Surveillance for foodborne-disease outbreaks - United States, 1998-2002. MMWR 55(SS10):1-34. 12 Maiden, M. C. J., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang, J. Zhou, K. Zurth, D. A. Caugant, I. M. Feavers, M. Achtman, and B. G. Spratt. 1998. Multilocus sequence typing: A portable approach to the identification of clones within populations of pathogenic microorganisms. Microbiology 95:3140- 3145. Meinersmann, R. J., L. O. Helsel, P. I. Fields, and K. L.Hiett. 1997. Discrimination of Campylobacter jejuni isolates by fla gene sequencing. J. Clin. Microbiol.35:2810- 2814. Miller, W. G., S. L. W. On, G. Wang, S. Fontanoz, A. J .Lastovica, and R. E. Mandrell. 2005. Multilocus sequence typing system for Campylobacter coli, C. lari, C. upsaliensis, and C. helveticus. J. Clin. Microbiol. 43:2315-2329. Nachamkin, I., K. Bohachick, and C. M. Patton. 1993. Flagellin gene typing of Campylobacter jejuni by restriction fragment length polymorphism analysis. J. Clin. Microbiol. 31:1531-1536. Nielsen, E. M., J. Engberg, V. Fussing, L. Petersen, C. Brogren, and S. L. W. On.2000. Evaluation of phenotypic and genotypic methods for subtyping Campylobacter jejuni isolates from humans, poultry, and cattle. J. Clin. Microbiol. 38:3800-3810. Olsen S. J., L. C.. MacKinon, J. S. Goulding, N. H. Bean, L. Slutsker. 2002. Surveillance for foodborne disease outbreaks - United States, 1993-1997. MMWR 49(SS01):1- 51. On, S. L. W. 2001. Taxonomy of Campylobacter, Arcobacter, Helicobacter and related bacteria: Current status, future prospects and immediate concerns. J. Appl. Microbiol. 90:1S-15S. 13 Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. J. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. 39:1889-1894. Riley, L. W. 2004a. Pinciples and Approaches, pp. 1-28. In Molecular Epidemiology of Infectious Diseases: Principles and Practices. ASM Press, Washington D.C. Riley, L. W. 2004b. Analysis of similarity and relatedness in molecular epidemiology, pp. 91-124. In Molecular Epidemiology of Infectious Diseases: Principles and Practices. ASM Press, Washington D.C. Salazar-Lindo, E., R. B. Sack, E. Chea-Woo, B. A. Kay, Z. A. Piscoya, R. Leon-Barua, A. Yi. 1986. Early treatment with erythromycin of Campylobacter jejuni- associated dysentery in children. J. Pediatr. 109:355-360. Sails, A. D., B. Swaminathan, and P. I. Fields. 2003. Utility of multilocus sequence typing as an epidemiological tool for investigation of outbreaks of gastroenteritis caused by Campylobacter jejuni. J. Clin. Microbiol. 41:4733-739. Suerbaum, S., M. Lohrengel, A. Sonnevend, F. Ruberg, and M. Kist. 2001. Allelic diversity and recombination in Campylobacter jejuni. J. Bacteriol. 183:2553- 2559. Swaminathan, B., T. J. Barrett, S. B. Hunter, R. V. Tauxe, and the CDC PulseNet Task Force. 2001. PulseNet: The molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg. Infect. Dis. 7:382-389. Young,V. B. and L. S. Mansfield. 2005. Campylobacter Infection ? Clinical Context, pp. 1-12. In J. M. Ketley and M. E. Konkel (ed.), Campylobacter: Molecular and Cellular Biology. Horizon Bioscience, Wymondham, Norfolk, U.K. 14 Table 1 Campylobacter taxa (2007) Campylobacter jejuni Campylobacter concisus C. jejuni subsp. jejuni Campylobacter upsaliensis C. jejuni subsp. doylei Campylobacter curvus C. jejuni subsp. fetus Campylobacter rectus Campylobacter coli Campylobacter helveticus Campylobacter lari Campylobacter lanienae Campylobacter fetus Campylobacter mucosalis C. fetus subsp. fetus Campylobacter showae C. fetus subsp. venerealis Campylobacter hominis Campylobacter hyointestinalis Campylobacter gracilis C. hyointestinalis subsp. lawsonii Campylobacter sputorum C. hyointestinalis subsp. hyointestinalis C. sputorum subsp. bubulus C. sputorum subsp. sputorum 15 II. MOLECULAR TYPING OF CAMPYLOBACTER JEJUNI STRAINS ISOLATED FROM COMMERCIAL BROILERS IN PUERTO RICO Introduction Little information is available about human campylobacteriosis in Puerto Rico. After Salmonella, C. jejuni has been reported as the second most common foodborne bacterial pathogen isolated from humans from a reference laboratory in San Juan (Lopez-Ortiz and Solivan, 1999). However, there is no information on the incidence and characteristics of Campylobacter spp. in commercial broilers raised in the island. The poultry industry is the second largest agricultural industry in Puerto Rico, with broiler meat comprising more than 69% of the total meat consumed, with an average of 80 lbs of broiler meat consumed per capita for the year 2005. Yet, from the total consumption of broiler meat only 34% is produced locally while the rest is imported as frozen products from the US (Anonymous 2005a). In C. jejuni, epidemic clones responsible for human disease are characterized by unique combinations of virulence genes or their alleles (Maynard Smith et al., 1993; Musser 1996), and some clones have been associated with domestic animals raised for human consumption, primarily broiler chickens. Therefore, it is assumed that contaminated raw poultry products and the consumption of undercooked poultry are major risk factors for human campylobacteriosis (Butzler and Oosterom, 1991; Nadeau et 16 al., 2002; Tauxe 1997). However, drawing conclusions about the pathogenicity potential of C. jejuni strains from different sources has always been a challenge. The weak clonality of C. jejuni brings several limitations when accurate estimations are needed for source tracking efforts aimed at determining the attribution of infections to food sources. Besides the differences in the interpretation of the results from the same techniques applied to different C. jejuni isolates, different molecular approaches may answer specific questions but may not be complementary to each other to help identify sources and improve the prediction of the appearance of epidemic clones in foods. Few studies have addressed the complementation of methods for epidemiological studies. The use of genomotyping may represent a holistic approach to identifying potentially pathogenic clones. Yet, these methods still do not provide a complete overview on the variability in the genomic profiles among C. jejuni strains from different areas (K?renlampi et al., 2007). In addition, DNA microarrays are more expensive and more difficult to perform than molecular typing methods based on restriction profiles, such as pulsed-field gel electrophoresis (PFGE), and even sequencing methods such as multilocus sequence typing (MLST). With these challenges in mind, we decided to collect samples from commercial live broilers in Puerto Rico to isolate Campylobacter strains and analyze them with DNA-based techniques to understand their genomic relatedness and potential implication in human campylobacteriosis. Campylobacter strains were collected from fecal material taken from commercial broiler chickens and analyzed using different molecular techniques to determine possible traits that would predict their risk for humans. For this purpose, isolates were 1) identified to the species level using culture media and multiplex 17 polymerase chain reaction (mPCR) assays and 2) characterized with PFGE, MLST, and flagellin A restriction fragment length polymorphism (flaA-RFLP). Materials and Methods Sample collection Fecal samples were collected by Fernando Rebollo-Carratto, a student of the University of Turabo at Gurabo, Puerto Rico from 17 commercial broiler farms in Puerto Rico between March and May 2005 covering different areas in the center-south region of the island. Sampled farms contained chicken houses, some with double-deck houses. Five to seven samples were collected from each farm. Each sample consisted of ~2-3 g of freshly deposited fecal material collected from four different areas of the houses with sterile tongue depressors. Samples were placed in tubes containing Cary-Blair semi-solid medium supplemented with 5% lysed horse blood and shipped in refrigeration (~8-10oC) for analysis to the Department of Poultry Science, Auburn University, USA. Campylobacter isolation Fecal samples were analyzed for Campylobacter spp. by direct plating on modified Campy-Cefex (mCC, Oyarzabal et al., 2005) and on modified charcoal cefoperazone deoxycholate agar (mCCDA; Hutchinson and Bolton, 1984) within four to six days of collection. Samples were also enriched in Preston broth (Oxoid, Ltd.) for 24 h at 42oC under microaerobic conditions (10% CO2, 5% O2, and 85% N2; Airgas, Radnor, PA) generated using anaerobic jars gassed with a MACSmics Jar Gassing System (Microbiology International,Frederick, MD). Enriched samples were swabbed (~0.1 ml) 18 onto mCC and mCCDA plates. All plates were then incubated at 42oC under microaerobic conditions for 48 h and screened for typical Campylobacter colonies. Colonies were considered presumptive positive if they showed the typical morphology and motility under phase contrast microscopy (Optiphot-2, Nikon, Tokyo, Japan). All presumptive isolates were collected and individually stored at-80oC in tryptic soy broth (Difco, Detroit, MI) supplemented with 30% glycerol (vol/vol) and 5% lysed horse blood for further identification and characterization. PCR identification of Campylobacter isolates Stock cultures were transferred to mCC plates that were incubated under microaerobic conditions at 42oC for 24 h to prepare fresh bacterial DNA. DNA was extracted using PrepMan? Ultra (Applied Biosystems, Foster City, CA). DNA samples were tested with two multiplex PCR assays that were performed in 25 l aliquots as described elsewhere (Oyarzabal et al., 2005, 2007a). A PTC-100 Programmable Thermal Controller (Bio-Rad Laboratories, Hercules, CA) was used for amplification. Amplicons were detected in 1.5% UltraPure? Agarose-1000 (Invitrogen Corporation, Carlsbad, CA) and DNA bands were stained with ethidium bromide, visualized using a UV transilluminator (Gel-Doc System, Syngene, Frederick, MD) followed by recording using a computer program (GeneSnap) for molecular analysis (Syngene, Frederick, MD). All strains were also tested with a commercial hippurate test (Hardy Diagnostics, Santa Maria, CA). 19 PFGE analysis A standard PFGE protocol (Ribot et al., 2001) with the modifications suggested by the Centers for Disease Control and Preventions (http://www.cdc.gov/pulsenet/ protocols.htm) was used. A digestion of Salmonella choleraesuis ss. Choleraesuis serotype Braenderup H9812 (ATCC BAA-664) was used as the DNA size marker. This marker was included in three lanes per each PFGE gel, the two lanes at the end and one in the middle. Campylobacter DNA was digested with KpnI, SalI or SmaI, as well as a combination of SmaI/KpnI. Strains that were not digestible with SmaI were also tested with the following enzymes: ApaI, BamHI, BglII, HindIII, KspI (SacII), NotI, NruI, SacI, SpeI, and XbaI. The switch times for the SmaI/KpnI combination were initial 6.76 sec and final 35.38 sec. Restricted DNA fragments were separated with a contour-clamped homogeneous electric field (CHEF Mapper, Bio-Rad, Hercules, CA) in 1% agarose gel stained with ethidium bromide, visualized with a UV transilluminator (Gel-Doc System, Bio-Rad), and a picture recorded using GeneSnap (Syngene, Frederick, MD). PFGE profiles were used to perform a pair comparison and cluster analysis using the Dice correlation coefficient, and a dendogram was created using the unweighted pair group method using mathematical average (UPGMA) clustering algorithm of BioNumerics version 4.50 (Applied Maths, Austin, TX). The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used to analyze the different PFGE patterns (De Boer et al., 2000). MLST, sequence analysis, and phylogenetic calculations 20 MLST was performed following the guidelines available at the C. jejuni MLST website (http://pubmlst.org/ campylobacter/) and guidelines from previous publications (Dingle et al., 2001). PCR amplification was performed in 25-?l reactions using Platinum PCR Supermix (Invitrogen Corporation). Amplicons were detected on 1.5% ethidium bromide agarose gels and purified with a MiniElute PCR Purification Kit (Qiagen Inc., Valencia, CA). The concentration of amplified DNA was measured with a NanoDrop? ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE). Two pre-made PCR mixes, OmniMix? HS (Cephied, Sunnyvale, CA) and Platinum PCR Supermix (Invitrogen Corporation) were used in a second PCR amplification using the suggested sequencing primers. Products of the second amplification were sequenced at the Auburn University Genetic Analysis Laboratory. Sequences were aligned, assembled and edited using MultAlin (http://prodes.toulouse.inra.fr/multalin/multalin.html), ClustalW (http://www.ebi.ac.uk/ clustalw/index.html) or BioEdit (Hall 1999) version 7.0.5.3 (http://www.mbio.ncsu.edu/BioEdit/BioEdit.html). Phylogenetic trees calculations for specific genes based on their sequences were conducted using Neighbor-Joining analysis of MEGA (Molecular Evolutionary Genetics Analysis) version 4 (http://www.megasoftware. net). Split decomposition analysis to test for recombination was done using SplitsTree analysis (Bandelt and Dress,1992; Huson 1998) with free software (http://pubmlst.org/perl/mlstanalyse/mlstanalyse.pl?site=pubmlst&page=splits tree&referer=pubmlst.org). Linkage disequilibrium analysis was calculated with START (http://pubmlst.org/software/analysis/ start/) for the index of association and with LIAN version 3.5 (http://pubmlst.org/perl/mlstanalyse/mlstanalyse.pl?site=pubmlst) (Haubold and Hudson 2000), for the standardized index of association. 21 flaA-RFLP Flagellin A RFLP was carried out in 25 l aliquots as described elsewhere (Harrington et al 2003). A PTC-100 Programmable Thermal Controller (Bio-Rad Laboratories, Hercules, CA) was used for amplification. Amplicons were detected in 2.5% SeaKem? Gold Agarose (Lonza, Basel, Switzerland). DNA bands were stained with ethidium bromide and visualized using a UV transilluminator (Gel-Doc System, Syngene, Frederick, MD). An image was recorded (tiff format) using a computer program (GeneSnap) for molecular analysis (Syngene, Frederick, MD). Results Strain identification Thirty C. jejuni isolates were collected from 94 broiler samples. From 17 farms that were sampled, 13 farms gave Campylobacter growth, and all isolates were confirmed as C. jejuni with two multiplex PCR assays and with a hippurate test. PFGE analysis The macrorestriction profiles obtained varied among the enzymes tested. SmaI restriction resulted in nine profiles, KpnI restriction resulted in eight PFGE profiles, SalI restriction gave seven profiles, and the combination of SmaI-KpnI digestion yielded 13 different restriction profiles, with a number of DNA bands that varied from six to 14. Ten strains did not restrict with KpnI, while two strains did not restrict with SalI, SmaI, or SmaI/KpnI. Besides the suggested pulse time for KpnI (Anon. 2007a), we also tried other switch times, some described elsewhere (De Boer et al., 2000; Michaud et al., 2001), but 22 the bands in the PFGE gels were usually less defined for KpnI and SalI than for SmaI. To prove whether the lack of SmaI restriction was due to presence of endogenous proteinases, formaldehyde-treated plugs were digested with different restriction enzymes. The two strains that did not restrict with SmaI did not restrict with ApaI, BglII, HindIII, KspI (SacII), KpnI, NotI, SmaI, SpeI or XbaI either. However these strains were restricted with BamHI, NruI, SalI and SacI. It is not clear whether lack of restriction with SmaI is due to the lack of SmaI restrictions sites or the presence of methylated DNA. MLST analysis The use of both OmniMix? HS and Platinum PCR Supermix for PCR amplification gave clean amplicons for sequencing data for MLST analysis. Table 1 shows the allelic profiles and clonal complexes for the C. jejuni strains. No new allelic sequences were found for any of the seven house-keeping genes. However, the predominant allelic profile was new and assigned as ST-2624, and deposited on the MLST web site (http://pubmlst.org/campylobacter/). Among individual alleles, glnA and pgm were the genes with considerable variability in allelic profiles, with 4 alleles for pgm (10, 11, 86 and 89) and glnA (4, 17, 21 and 30). A calculation of phylogenetic trees for pgm and glnA based on their sequences and using the Neighbor-Joining feature analysis of MEGA version 4 (Tamura et al., 2007) showed a high variability within the four alleles of each gene (data not shown). The calculation of the linkage disequilibrium using the algorithms from DnaSP version 4.10.9 resulted in values of 0.77 (Kelly 1997) and 0.74 (Rozas et al., 2001) for pgm and 0.42 (Kelly 1997) and 0.53 (Rozas et al., 2001) for glnA. Three alleles were found for aspA (7, 14 and 24) and tkt (1, 3 and 59), and only two for gltA (2 and 5), 23 glyA (2 and 68) and uncA (5 and 6). Their calculated genetic diversity at individual loci (LIAN) were:aspA = 0.5398; glnA = 0.6172; gltA =0.5032; glyA = 0.5118; pgm = 0.6172; tkt = 0.4710, and uncA = 0.4258. Overall, the LIAN analysis resulted in a mean genetic diversity (H) of 0.5266 ? 0.027 (standard deviation), and showed linkage disequilibrium for this population of C. jejuni (Table 2). The phylogenetic tree of the ST profiles was calculated with the UPGMA and showed that the new profile ST-2624 was more closely related to ST-460 than the other two ST profiles. The decomposition analysis was performed with SplitsTree and resulted in a network-like shape suggesting a high degree of recombination within genes (Figure 1A & 1B). When compared to PFGE, MLST was less discriminatory for strain characterization, and strains that grouped under ST-2624 had nine different PFGE profiles (Figure 2). flaA-RFLP analysis Thirteen unique restriction profiles were obtained with DdeI enzyme with the number of DNA bands varying from 4 to 8 (Figure 3). The two largest groupings contained 5 and 6 strains respectively. Discussion We pursued this study to determine the prevalent Campylobacter species that colonizes commercial broiler flocks in Puerto Rico and to characterize these isolates using PFGE, MLST, and flaA-RFLP. From the total consumption of broiler meat, only 34% is produced locally, while the rest is imported as frozen products. Therefore, although the 24 island may be geographically separated from the continental US, the introduction of new C. jejuni strains may be possible through the importation of retail poultry products. The year average temperature in Puerto Rico is between 70 to 90?F, and the temperature recorded outside the farms was between 75 and 80?F. Only few samples were positive by direct plating, but 32% of the samples were positive after enrichment in Preston broth. This number of positive flocks is lower than the 57% reported for commercial broilers in the southeast US (Potturi-Venkata et al., 2007a), and may be accounted by the shipping of the samples in refrigeration for several days, which may have resulted in the reduction in the number of Campylobacter cells in the fecal samples. Flocks were sampled two weeks before processing, when Campylobacter is usually present at high numbers (Potturi-Venkata et al., 2007a,b). The methodology used was optimized with combinations of plates and enrichment broth for the isolation of Campylobacter spp. from fecal material collected from live broilers (Potturi-Venkata et al., 2007b). Studies that compared the macrorestriction patterns of SmaI and KpnI of human isolates of C. jejuni have concluded that KpnI is the enzyme of choice for molecular epidemiology studies of C. jejuni (K?renlampi et al., 2003; Michaud et al., 2001). However, we found that the use of SmaI appears to be better than KpnI for PFGE analysis of C. jejuni from live broilers. Although the number of restriction profiles was similar between KpnI and SmaI, only 66% of the isolates were restricted with KpnI versus more than 90% that were restricted with SmaI, or SalI. SmaI always restricts more C. jejuni strains than KpnI (Michaud et al., 2001), and it is the primary restriction enzyme for the protocols developed by the CDC, CampyNet (http://campynet.vetinst.dk/PFGE.html) and 25 studies in Denmark (Nielsen et al., 2006). SalI normally generates partial digest despite the experimental conditions (Ho and Monaco, 1995), a feature that we noticed by the presence of un-resolved fragments in the compression zone of the gels for all the strains. Therefore, this enzyme should be limited in its use for restriction profiling for fingerprinting purposes. The combination of a macrodigestion with SmaI followed by the digestion with KpnI resulted in a larger number of bands and profiles (Table 3) and may be more discriminatory to track the sources of Campylobacter isolates in poultry production. It is important to restrict the plugs first with SmaI and then with KpnI to guarantee that a larger number of strains will be restricted. The increase in the number of fragments with the SmaI/KpnI combination was more notorious in the area below 250 kbp (Figure 2), an element that made the PFGE profiles more challenging to analyze using the visual scoring system suggested by Tenover and co-workers (1995). Therefore, the use of fingerprinting analysis software, such as BioNumerics, appears to be indispensable for a reliable analysis of the results obtained with this combination. The lack of improvement by adding formaldehyde in the PFGE protocol strongly suggests that the lack of restriction patterns found for some isolates and specific enzymes is not related to the presence of DNAses in those strains (Gibson et al., 1994). The fact that all strains showed restriction with frequently cutting enzymes suggests that those strains actually lack some restriction sites or that some sites are methylated. In large eukaryotic genomes (> 108 base pairs), this lack of restriction, especially the CpG nucleotide deficiency, has been usually associated to DNA methylation, which is almost universal in these organisms (Bestor 1990). DNA methylation involves the addition of a methyl group to DNA bases, primarily to carbon #5 of the cytosine pyrimidine ring. This 26 theory has traditionally been applied to explain the lack of restriction of bacterial genome. However, recent findings question this assertion and suggest that cytosine methylation is not the primary reason for the CpG dinucleotide deficiency in bacterial genomes (Wang et al., 2004). It appears that cytosine methylation in bacteria is not generally associated with CpG dinucleotide methylation, especially in C. jejuni, which is CpG deficient (Wang et al., 2004). The methylation related problems in Campylobacter may be associated with ApC sites. PFGE macrorestriction with SmaI has been reported to be more discriminatory than MLST for outbreak investigations, although a combination of MLST plus sequencing of the flaA short variable region may provide a level of discrimination equivalent to PFGE for outbreak investigations (Sails et al., 2003).Our comparison of the PFGE, MLST and flaA-RFLP profiles of C. jejuni strains isolated in the same geographical area suggests that the PFGE profiles may change quickly within C. jejuni strains and may account for the variety of genotypes frequently found when analyzing C. jejuni with PFGE. It is apparent that a ST profile of C. jejuni, or clone, appears to dominate in a geographic area for a variable period. The prevalent ST profile in our report is a new combination of previously described alleles. The other three MLST profiles found have been identified in C. jejuni isolated from humans, animals and the environment. ST-460 has been reported by the Centers for Disease Control and Prevention, and ST-48 and ST- 353 were described in England (Dingle et al., 2001, 2002). ST-48 was the most predominant ST in a study done in New South Wales, Australia (Mickan et al., 2007). 27 Within the individual genes, glnA and pgm showed the highest diversity and the analysis of their sequence diversity showed a variability that has been described for other genes in C. jejuni (Suerbaum et al., 2001). Both the diversity within genes and the analysis of the allelic profile data using SplitsTree and the UPGMA cluster analysis resulted in graphs with alleles connected to each other by multiple pathways, with a network-like structure suggesting recombination (Suerbaum et al., 2001). Pgm has been used to analyze the recombination of C. jejuni from MLST data, and along with unc (atpA, ATP synthase alpha subunit) they appear to come to C. jejuni from other Campylobacter spp. (Fearnhead et al., 2005). In our study, the diversity of the allelic profiles for unc was minimal, with two alleles and only one base difference (C versus T) in base 864 of the atpA gene (NCBI accession number AL111168). This is different from the variety of allelic profiles, with multiple polymorphic sites, reported by French et al. (2005) for unc alleles analyzed from C. jejuni associated to cattle in the UK. It appears that the allelic variability for each of the seven housekeeping genes for MLST profiles varies according to the origin of the strains and to different geographical areas. We calculated both the index of association (IA), as described for the analysis of the degree of linkage in multilocus enzyme electrophoresis (Maynard Smith, 1999; Maynard Smith et al., 1993), and the standardized IA (sIA) that corrects for the number of loci analyzed (Haubold and Hudson, 2000). The sIA value of 0.73 and a calculated variance that was significantly higher than the expected variance (P ? 0.0127) suggested a high degree of recombination among the MLST genes within this population of C. jejuni isolates (Table 2). It is important to remember that a hypothetical sIA value of 0 represents alleles that are in linkage equilibrium due to free recombination. 28 MLST results have revealed that populations of different bacteria frequently exchange genetic material, and produce interclonal variance in virulence traits. These findings based on molecular population genetic studies have resulted in the tenet that the unit of bacterial pathogenicity is the clone, or cell line (Musser1996). However, the analyses of linkage disequilibrium on comparison of gene trees have revealed that few bacteria are indeed clonal (Feil et al., 2001; Maynard Smith et al., 1993; Spratt and Maiden, 1999; Supply et al., 2003). For MLST from C. jejuni, a high frequency of inter- and intra-species recombination, calculated to be more than 50 times the frequency of mutation, has been shown to alter the sequence profiles of the housekeeping genes (Schouls et al., 2003). This high recombination frequency plus the mutation rate may account for the high degree of variation (addition and or deletions) of restriction sites and PFGE profiles found in C. jejuni isolated from live broilers chickens (Dingle et al., 2001; Potturi-Venkata et al., 2007a,b; Schouls et al., 2003) and retail broiler products (Dickins et al., 2002; Oyarzabal et al., 2007a). Therefore, MLST profiling, although useful to determine temporal phylogenetic comparison, may be less discriminatory for spatial, short term epidemiological studies of C. jejuni strains from live broilers. The fact that we found only four ST allelic profiles, with one being completely new, versus 10 different PFGE profiles suggest that macrorestriction profiling is still a powerful system for the tracking of C. jejuni strains. FlaA-RFLP presented 13 restriction profiles. Even though flaA-RFLP has a higher number of profiles, 13, than PFGE, 12, PFGE?s discriminatory power in this study is higher than flaA-RFLP, D=0.9233 and D=0.8718 respectively. This drop in discrimination is not surprising, since flaA-RFLP targets only one gene for typing rather 29 than the entire genome. But when debating time and expense, flaA-RFLP could be substituted in the place of PFGE for C. jejuni strains. The molecular characterization of bacterial pathogens is essential to understand the short-term and long-term epidemiological implications of a given foodborne pathogen (Spratt et al., 1999). The long-term goal of an active surveillance program is to predict, with high degree of certainty, the rise and spread of a lineage with high pathogenicity potential for humans, or epidemic clone (Maynard Smith et al., 1993). For weakly clonal organisms molecular methods have yielded different data, and there appears to be a lack of agreement among methods applied to the same group of strains. For instance, different databases have been collected and maintained for C. jejuni, such as fingerprinting profiles based on PFGE or sequencing data based on MLST. MLST results in 100% typability of the isolates, but the cost of this method (Dingle et al., 2002) and the apparent low number of alleles found in a relatively small population of isolates may limit its application for epidemiological studies. Yet, the prediction of the appearance of epidemic clones in a given geographic area may still be limited even if an active surveillance is in place. The different results and the difficulty in the interpretation of these results become more a limitation in studies on pathogens with high capacity for horizontal gene transfer (Oyarzabal et al., 2007b), where even the typability of the isolates may vary according to the different methods. The small population of C. jejuni that we sampled may be a limiting factor of this study, but at the same time exemplifies some of the limitations of each of the methods employed for strain characterization. The diversity found within C. jejuni and the lack of agreement among the different methods suggests studies of larger populations of strains will result in more complex information to analyze. 30 References Anonymous. 2005a. Report on the consumption of broiler meat in Puerto Rico. Office of Agricultural Statistics, Department of Agriculture of Puerto Rico. Santurce, P.R. Bandelt, H. J. and A. W Dress. 1992. Split decomposition: a new and useful approach to phylogenetic analysis of distance data. Mol. Phylogenet. Evol. 1:242-252. Bestor, T. H. 1990. DNA methylation: Evolution of a bacterial immune function into a regulator of gene expression and genome structure in higher eukaryotes. Phil. Trans. R. Soc. Lond. B 326:179-187. Butzler, J., and J Oosterom. 1991. Campylobacter: Pathogenicity and significance in foods. Int. J. Food Microbiol. 12:1-8. Coote, J. G., D. E. S Stewart-Tull, R. J. Owen, F. J. Bolton, B. L. Siemer, D. Candlish, D. H Thompson, A. C. Wardlaw, S. L. W. On, A. Candlish, B. Billcliffe, P. J. Jordan, K. Kristiansen, and P. Borman. 2007. Comparison of virulence-associated in vitro properties of typed strains of Campylobacter jejuni from different sources. J. Med. Microbiol. 56:722-732. De Boer, P., B. Duim, A. Rigter, J. van der Plas, W. Jacobs-Reitsma, and J. A. Wagenaar. 2000. Computer-assisted analysis and epidemiological value of genotyping methods for Campylobacter jejuni and Campylobacter coli. J. Clin. Microbiol. 38:1940-1946. 31 Dickins, M. A., S. Franklin, R. Stefanova, G. E. Shutze, K. D. Eisenach, I. V. Wesley, and D. Cave. 2002. Diversity of Campylobacter isolates from retail poultry carcasses and from humans as demonstrated by pulsed-filed gel electrophoresis. J. Food Prot. 65:957-962. Dingle, K. E., F. M. Colles, R. Ure, J. A. Wagenaar, B. Duim, F. J. Bolton, A. J. Fox, D. R. A. Wareing, and M. C. J. Maiden. 2002. Molecular characterization of Campylobacter jejuni clones: a basis for epidemiologic investigation. Emerg. Infect. Dis. 8:949-955. Dingle, K. E., F. M. Colles, D. R. Wareing, R. Ure, A. J. Fox, F. E. Bolton, H. J. Bootsma, R. J. Willems, R. Urwin, and M. C. J. Maiden. 2001. Multilocus sequence typing system for Campylobacter jejuni. J. Clin. Microbiol. 39:14?23. Engberg, J., J. Neimann, E. M. Nielsen, F. M. Aarestrup, and V. Fussing. 2004. Quinolone-resistant Campylobacter infections in Denmark: Risk factors and clinical consequences. Emerg. Infect. Dis. 10:1056-1063. Fearnhead, P., N. G. C. Smith, M. Barrigas, A. Fox, and N. French. 2005. Analysis of recombination in Campylobacter jejuni from MLST population data. J. Mol. Evol. 61: 333?340. Feil, E. J., E. C. Holmes, D. E. Bessen, M. -S. Chan, N. P. Day, M. C. Enright, R. Goldstein, D. W. Hood, A. Kalia, C. E. Moore, J. Zhou, and B. G. Spratt. 2001. Recombination within natural populations of pathogenic bacteria: short-term empirical estimates and long-term phylogenetic consequences. Proc. Natl. Acad. Sci. USA 98:182?187. 32 French, N. P., M. Barrigas, P. Brown, P. Ribiero, N. J. Williams, H. Leatherbarrow, R. Birtles, E. Bolton, P. Fearnhead, and A. Fox. 2005. Spatial epidemiology and natural population structure of Campylobacter jejuni colonizing a farmland ecosystem. Environ. Microbiol. 7:1116-1126. Gibson, J. R., K. Sutherland, and R. J. Owen. 1994. Inhibition of DNAase activity in PFGE analysis of Campylobacter jejuni..Lett. Appl. Microbiol. 19:357-358. Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41:95-98. Harrington, C. S., L. Moran, A. M. Ridley, D. G. Newell, and R. H. Madden. 2003. Inter- laboratory evaluation of three flagellin PCR/RFLP methods for typing Campylobacter jejuni and C. coli: the CAMPYNET experience. J. Appl. Microbiol. 95:1321-1333. Haubold, B., and R. R. Hudson 2000. LIAN 3.0: Detecting linkage disequilibrium in multilocus data. Linkage Analysis Bioinformatics 16:847-848. Ho, M. F., and A. P. Monaco. 1995. PFGE in physical mapping, pp. 21-43. In A. P Monaco (ed.). Pulsed field gel electrophoresis, a practical approach. Oxford University Press Inc., New York, NY. Huson, D. H. 1998. SplitsTree: analyzing and visualizing evolutionary data. Bioinformatics 14, 68-73 Hutchinson, D. N., and F. J. Bolton. 1984. An improved blood-free selective medium for isolation of Campylobacter jejuni from faecal specimens. J. Clin. Pathol. 37:956- 95. 33 K?renlampi, R., H. Rautelin, and M. -L. H?nninen. 2007. Evaluation of genetic markers and molecular typing methods for prediction of sources of Campylobacter jejuni and C. coli infections. Appl. Environ. Microbiol. 73:1683-1685. Kelly, J. K. 1997. A test of neutrality based on interlocus associations. Genetics 146:1197?1206. Lopez-Ortiz, W., R. A. Solivan. 1999. Campylobacter jejuni among patients with gastroenteritis: incidence at a reference microbiology laboratory in San Juan, Puerto Rico. P. R. Health Sci. J. 18:273-276. Maynard Smith, J. 1999. The detection and measurement of recombination from sequence data. Genetics 153:1021-1027. Maynard Smith, J., N. H. Smith, M. O?Rourke, and B. G. Spratt. 1993. How clonal are bacteria? Proc. Natl. Acad. Sci. USA 90:4384-4388. Michaud, S., S. Menard, C. Gaudreau, and R. D. Arbeit. 2001. Comparison of SmaI- defined genotypes of Campylobacter jejuni examined by KpnI: a population- based study. J. Med. Microbiol. 50:1075-1081. Mickan, L., R. Doyle, M. Valcanis, K. E. Dingle, L. Unicomb, and J. Lanser. 2007. Multilocus sequence typing of Campylobacter jejuni isolates from New South Wales, Australia. J. Appl. Microbiol. 102:144-52. Musser, J. M. 1996. Molecular population genetic analysis of emerged bacterial pathogens: selected insights. Emerg. Infect. Dis. 2:1-17. 34 Nadeau, E., S. Messier, and S. Quessy. 2003. Comparison of Campylobacter isolates from poultry and humans: association between in vitro virulence properties, biotypes and pulse-field gel electrophoresis clusters. Appl. Environ. Microbiol. l69:6316-6320. Nadeau, E., S. Messier, and S. Quessy. 2002. Prevalence and comparison of genetic profiles of Campylobacter strains isolated from poultry and sporadic cases of campylobacteriosis in humans. J. Food Prot. 65:73-78. Nielsen, E. M., V. Fussing, J. Engberg, N. L. Nielsen, and J. Neimann. 2006. Most Campylobacter subtypes from sporadic infections can be found in retail poultry products and food animals. Epidemiol. Infect.134:758-767. Oyarzabal, O. A., S. Backert, M. Nagaraj, R. S. Miller, S. K. Hussain, and E. A. Oyarzabal. 2007a. Efficacy of supplemented buffered peptone water for the isolation of Campylobacter jejuni and C. coli from broiler retail products. J. Microbiol. Methods 69:129-136. Oyarzabal, O. A., K. S. Macklin, J. M. Barbaree and R. S. Miller. 2005. Evaluation of agar plates for direct enumeration of Campylobacter spp. from poultry carcass rinses. Appl. Environ. Microbiol. 71:3351-3354. Potturi-Venkata, L. -P., S. Backert, S. L. Vieira, and O. A. Oyarzabal. 2007. Evaluation of logistic processing to reduce cross-contamination of commercial broiler carcasses with Campylobacter spp. J. Food Prot. 70:2549-2554. 35 Potturi-Venkata, L. P., S. Backert, A. J. Lastovica, S. L. Vieira, R. A. Norton, R. S. Miller, S. J. Pierce, and O. A. Oyarzabal, 2007. Evaluation of different plate media for direct cultivation of Campylobacter spp. from live broilers. Poul. Sci. 86:1304-1311. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. T. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. l39:1889-1894. Rozas, J., M. Gullaud, G. Blandin, and M. Aguad?. 2001, DNA variation at the rp49 gene region of Drosophila simulans: Evolutionary inferences from an unusual haplotype structure. Genetics 158:1147?1155. Sails, A. D., B. Swaminathan, and P. I. Fields. 2003. Utility of multilocus sequence typing as an epidemiological tool for investigation of outbreaks of gastroenteritis caused by Campylobacter jejuni. J. Clin. Microbiol. l41:4733-4739. Schouls, L. M., S. Reulen, B. Duim, J.A. Wagenaar, R. J. L. Willems, K. E. Dingle, F. M. Colles, and J. D. A. Van Embden. 2003. Comparative genotyping of Campylobacter jejuni by amplified fragment length polymorphism, multilocus sequence typing, and short repeat sequencing: strain diversity, host range, and recombination. J. Clin. Microbiol. l41:15-26. Suerbaum, S., M. Lohrengel, A. Sonnevend, F. Ruberg, and M. Kist. 2001. Allelic diversity and recombination in Campylobacter jejuni. J. Bacteriol. 183:2553? 2559. Spratt, B. G., and M. C. J. Maiden. 1999. Bacterial population genetics, evolution and epidemiology. Phil. Trans. R. Soc. London. B Biol. Sci. 354:701-710. 36 Supply, P., R. M. Warren, A. -L. Ba?uls, S. Lesjean, G. D. van der Spuy, L. -A. Lewis, M. Tibayrenc, P. D. van Heldenand, and C. Locht. 2003. Linkage disequilibrium between minisatellite loci supports clonal evolution of Mycobacterium tuberculosis in a high tuberculosis incidence area. Mol. Microbiol. 47:529-538. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007.MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution, doi:10.1093/molbev/msm092. [Online.] http://www.megasoftware.net. Accessed 23 September 2007. Tauxe, R. 1997. Emerging foodborne diseases: An evolving public health challenge. Emerg. Infect. Dis. 3:425-434. Tenover, F. C., R. D. Arbeit, R. V. Goering, P. A. Mickelsen, B. E. Murray, D. H. Persing, and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: Criteria for bacterial strain typing. J. Clin. Microbiol. 33:2233-2239. Wang, Y., E. P. C. Rocha, F. C. C. Leung, and A. Danchin. 2004. Cytosine methylation is not the major factor inducing CpG dinucleotide deficiency in bacterial genomes. J. Mol. Evol. 58:692-700. 37 Figure Captions Figure 1A Lineage assignment using the UPGMA cluster analysis calculated with START version 1.0.8. ST-2624 and ST-460 were closely related and more distant from ST-48 and ST-353. B: Decomposition analysis on allelic profile data using SplitsTree (Anonymous 2007d; Hudson 1998) resulted in a network-like shape, a shape that suggests recombination. Figure 2 PFGE macrorestriction profiles (SmaI/KpnI) of C. jejuni isolates belonging to ST-2624. Pair comparison and cluster analysis performed with the Dice correlation coefficient and the unweighted pair group mathematical average clustering algorithm of BioNumerics version 4.50 (Applied Maths, Austin, TX). The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used for the determination of the different PFGE patterns. Figure 3 flaA-RFLP restriction profiles of all Puerto Rico strains 38 Figure 1 A B 39 Figure 2 40 Figure 3 41 Table 1 Allellic profiles and clonal complexes from C. jejuni strains isolated in this study. ST Allelic Profile Clonal Complex Strains % of Dataset 2624 a NA b 16 53.3 429 48 9 30.0 353 353 4 13.3 460 460 1 3.3 aNew ST profile. bNo clonal complex assigned yet. 42 Table 2 Index of association calculated with START and the standardized index of association calculated with LIAN showing that the tested population of C. jejuni is in linkage disequilibrium. Testing Null Hypotheis (H0: VD = Ve) a Summary statistics (Means) Monte Carlo (1000 Resamplings) Parametric VD 9.27 VD 0.0127 0.0136 Ve 1.71 L 1.9141 1.906 sIA b 0.73 IA c 4.35 a VD: Calculated or observed variance; Ve: Expected variance for linkage equilibrium. b sIA: Standardized index of association cIA: Index of association. significantly different (P < 0.05) by Duncan's Multiple Range Test (SAS version 9.1). 43 Table 3 PFGE macrorestriction profiles, ST allelic profiles of C. jejuni strains isolated from five different farms. a Not restricted. Isolation (Day/Place) PFGE Profiles Isolate ID Day Farm House SmaI KpnI SalI SmaI/ KpnI ST Allelic Profiles F4-H1-S1 1 4 1 S-3 NR a Sa-3 SK-4 2624 F5-H1-S1 1 5 1 S-1 NR Sa-4 SK-3 48 F10-H1-S1 2 10 1 S-2 K-2 Sa-2 SK-5 2624 F10-H1-S2 2 10 1 S-2 K-4 Sa-2 SK-5 2624 F10-H1-S3 2 10 1 S-5 K-5 Sa-2 SK-8 2624 F10-H2-S4 2 10 2 S-6 K-6 Sa-2 SK-7 2624 F10-H2-S5 2 10 2 S-6 K-6 Sa-2 SK-7 2624 F11-H1-S1 3 11 1 S-6 K-7 Sa-2 SK-5 2624 F11-H1-S3 3 11 1 S-6 K-7 Sa-2 SK-7 2624 F11-H2-S4 3 11 2 S-6 K-7 Sa-2 SK-9 2624 F11-H2-S5 3 11 2 S-6 K-7 Sa-2 SK-7 2624 F12-H1-S1 3 12 1 S-7 K-7 Sa-2 SK-7 2624 44 III. DNA CHARACTERIZATION OF CAMPYLOBACTER JEJUNI AND CAMPYLOBACTER COLI ISOLATES FROM CHICKENS IN GRENADA Introduction Grenada is an island nation in the southeastern Caribbean Sea. In 2003 Grenada produced approximately 14% of the 7.2 million kg of chicken meat consumed on the island (David 2004), importing the remaining 86%. An average of 54 lbs per capita of chicken meat is consumed annually in Grenada (Anonymous, 2008). Information concerning human campylobacteriosis in Grenada is not readily available, although statistics reveal that gastroenteritis and diarrhea are among the leading causes of infant mortality. Between 1992 and 1996, gastroenteritis increased by 60% in children less than five years of age and by 73.5% in children over five years (Anonymous, 1998). Due to the lack of information regarding Campylobacter infection along with Grenada?s high importation of chicken meat, we decided to analyze Campylobacter strains collected from broilers and layer chickens in Grenada with multiplex polymerase chain reaction assays (mPCR) for determination of their identity and with two fingerprinting techniques, restriction fragment length polymorphism of the flagellin A (flaA-RFLP) and pulsed-field gel electrophoresis (PFGE), to determine their genetic relatedness. Previously, these isolates were identified to the species level using culture media and biochemical tests (Hariharan et al., 2008). The discrepancies between the results from biochemical tests and the results from mPCR assays, and the degree of 45 discrimination of flaA-RFLP, PFGE and the combination of both techniques are presented. Materials and Methods Sample collection and biochemical identification of isolates The collection of ceca, the isolation of the Campylobacter strains and the identification using biochemical tests have been described elsewhere (Hariharan et al., 2008). Briefly, freshly grown cultures were subjected to catalase, oxidase (BBL, Becton, Dickinson and Co., Sparks, MD, USA), and hippurate tests (Remel, Lennexa, KS, USA), and latex agglutination for C. jejuni/coli/lari (JCL, Panbio, Columbia, MD, USA). All isolates were also tested for their susceptibility to nalidixic acid (30 mcg disk) and cephalothin (30 mcg disk) on Mueller-Hinton agar with 5% sheep blood. JCL positive isolates were considered to belong to C. jejuni/coli/lari group. Hippurate positive isolates were identified as C. jejuni, while nalidixic acid susceptible hippurate negative isolates as C. coli, and nalidixic acid resistant, hippurate negative isolates as C. lari. Isolates were stored at -85?C in 2% sterile, skim milk in cryovials. From these stock cultures, strains were grown on blood agar plates (24 h at 42oC under microaerobic conditions) and the growth was used to inoculate a mix of Cary-Blair and Brucella broth (1:3) supplemented with 5% lysed horse blood for a final volume of 20 ml. These tubes were shipped under refrigeration (~6-8oC) to the Microbiology Laboratory at Auburn University, USA, for further studies. At Auburn University, isolates were recovered by plating the samples on modified Campy-Cefex (mCC, Oyarzabal et al., 2005). Samples were also enriched in Bolton broth 46 (Oxoid, Ltd.) for 24 h at 42oC under microaerobic conditions(10% CO2, 5% O2, and 85% N2; Airgas, Radnor, PA) generated using anaerobic jars gassed with a MACSmics Jar Gassing System (Microbiology International, Frederick, MD). Enriched samples were swabbed (~0.1 ml) onto mCC plates. All plates were then incubated at 42oC under microaerobic conditions for 48 h and screened for typical Campylobacter colonies. Colonies were considered presumptive positive if they showed typical morphology and motility under phase contrast microscopy (Optiphot-2, Nikon, Tokyo, Japan). Presumptive isolates were identified using API Campy tests (bioMerieux, Hazelwood, MO) and DNA was extracted using PrepMan? Ultra (Applied Biosystems, Foster City, CA). Isolates were then stored at-80oC in tryptic soy broth (Difco, Detroit, MI) supplemented with 30% glycerol (vol/vol) and 5% lysed horse blood for further identification and characterization. PCR identification of Campylobacter isolates DNA samples were tested with two multiplex PCR assays that were performed in 25 l aliquots as described elsewhere (Oyarzabal et al., 2005, 2007). These mPCRs selectively detect C. coli based on amplifications of the aspA (Linton et al. 1997) and ceuE (Gonzalez et al. 1997) genes, and C. jejuni based on the amplification of the hipO gene (Persson and Olsen 2005) and an undefined gene (Winters and Slavik 1995). A PTC-100 Programmable Thermal Controller (Bio-Rad Laboratories, Hercules, CA) was used for amplification. Amplicons were detected in 1.5% UltraPure? Agarose-1000 (Invitrogen Corporation, Carlsbad, CA) and DNA bands were stained with ethidium bromide, visualized using a UV transilluminator (Gel-Doc System, Syngene, Frederick, MD) 47 followed by recording using a computer program (GeneSnap) for molecular analysis (Syngene, Frederick, MD). flaA-RFLP analysis Isolates were recovered from frozen stocks on mCC. flaA-RFLP analysis was performed in 25 l PCR aliquots as described elsewhere (Harrington et al 2003; Nachamkin et al., 1993). A PTC-100 Programmable Thermal Controller (Bio-Rad Laboratories, Hercules, CA) was used for amplification. Amplicons were detected in 2.5% SeaKem? Gold Agarose (Lonza, Basel, Switzerland) and DNA bands were stained with ethidium bromide, visualized using a UV transilluminator (Gel-Doc System, Syngene, Frederick, MD) followed by recording using a computer program (GeneSnap) for molecular analysis (Syngene, Frederick, MD). PFGE analysis A standard PFGE protocol (Ribot et al., 2001) with the modifications suggested by the Centers for Disease Control and Preventions (http://www.cdc.gov/pulsenet/protocols.htm) and described elsewhere (Oyarzabal et al., 2008) was used. The only modification to described protocols was the preparation of an OD value of 1.0 when collecting the cells. A digestion of Salmonella choleraesuis ss. Choleraesuis serotype Braenderup H9812 (ATCC BAA-664) was used as the DNA size marker. Campylobacter DNA was digested with SmaI and XmaI. Restricted DNA fragments were separated with a contour-clamped homogeneous electric field (CHEF Mapper, Bio-Rad, Hercules, CA) in 1% SeaKem? Gold Agarose (Lonza, Basel, 48 Switzerland) gel stained with ethidium bromide, visualized with a UV transilluminator (Gel-Doc System, Bio-Rad), and a picture recorded using GeneSnap (Syngene, Frederick, MD). Interpretation of results TIF images of the restriction profiles for flaA-RFLP and PFGE were incorporated into BioNumerics version 4.50 (Applied Maths, Austin, TX) and pair-comparisons and cluster analysis was made using the Dice correlation coefficient and the unweighted pair group mathematical average (UPGMA) clustering algorithm. The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used for the determination of the different restriction patterns. The Simpson?s index of diversity (Hunter and Gaston, 1988) was used to determine the discriminating power of each restriction technique. Results Strain identification From 60 cultures shipped to Auburn University, four were not able to be recovered after enrichment. The attempts to speciate the isolates using the API test resulted in several ambiguous results. Therefore, the test was not used for the determination of the species. By using two mPCR assays, 49 isolates were identified as C. coli and seven were identified as C. jejuni. Isolates identified as C. lari by biochemical tests resulted in C. coli by the mPCR assays and failed to generate amplicons in a PCR assay that uses primers specific for C. lari (Oyarzabal et al., 1997). Table 1 shows the number of isolates for 49 which discrepancies between biochemical test results and DNA results were found. Two isolates (G185 and G187) gave mixed results (C. jejuni and C. coli bands) in PCR results suggesting that they were mixed cultures of these two Campylobacter spp. These strains were re-isolated using a 0.65 ?m filter (Millipore Corp., Billerica, MA) and one isolated colony from each sample was analyzed again for PCR, flaA-RFLP and PFGE. G185 was identified as C. coli and gave a restriction profile with PFGE, while G187 was isolated as C. jejuni but did not restrict with PFGE using SmaI or XmaI (Figure 1). flaA-RFLP and PFGE results One isolate did not yield a flaA-RFLP profile using the described protocol with restriction by DdeI. The remaining 55 isolates (C. jejuni and C. coli) gave 31 flaA-RFLP profiles, with a Simpson?s index of diversity of 0.958. However, six clusters contained 47% of the strains. Of all isolates, five C. coli and one C. jejuni isolates did not show any restriction profile with the PFGE protocol using SmaI or XmaI. With PFGE, only 26 profiles were obtained and therefore this technique showed a lower discriminatory power (D = 0.9437). A combination of the flaA-RFLP and PFGE profiles results in the highest discrimination for these isolates (D = 0.9906). Some isolates that clustered together by PFGE analysis showed different flaA-RFLP patterns (Figure 2). Discussion The biochemical differentiation of C. jejuni and C. coli is based on one test, hippurate hydrolysis. C. jejuni strains carry the hipO gene which encodes the hippuricase enzyme. 50 This enzyme catalyzes the hydrolysis of hippuric acid to benzoic acid and glycine (Hani and Chan, 1995; Harvey, 1980). Although all C. jejuni test positive for the presence of this gene using molecular assays, the limitation of the biochemical test is that some C. jejuni strains do not express the enzyme for this reaction and appear to be hippurate- negative with biochemical tests (Totten et al., 1987). Similar limitations have been found for the biochemical identification of C. lari, in which some C. jejuni strains may be resistant to nalidixic acid and can be misidentified as C. lari (Oyarzabal et al., 1997). Since the incorporation of molecular techniques for routine identification of Campylobacter isolates collected from chickens, approximately 10 to 15 years ago, the report of C. lari has dropped considerable. In the US, no confirmed report of C. lari has been documented in broiler chickens in the last eight years. Even the presence of C. lari in samples collected from commercial turkeys amounts to very few strains from hundreds of Campylobacter isolates from this avian species (Smith et al., 2004). In our studies, the agreement between the two mPCR was 100%; therefore, all isolates that were found to be C. coli with one mPCR were also C. coli with the other mPCR. One of these mPCR assays uses a set of primers (COL1 and COL2) that amplifies a 894-bp fragment of the ceuE gene, a gene that encodes a lipoprotein component of enterochelin. These primers have been found to be specific for C. coli in several studies (Debruyne et al., 2008; Gonzalez et al., 1997; Nayak et al., 2005; Oyarzabal et al., 2007). The other mPCR also uses a specific set of primers (CC18F and CC519R) that reacts only with the aspartokinase gene of C. coli (Debruyne et al., 2008; Linton et al., 1997; Oyarzabal et al., 2005). Therefore, C. coli makes up the majority of the Campylobacter 51 strains from chickens in Grenada, and this Campylobacter species appears to be more prevalent than C. jejuni. Some isolates with similar PFGE profiles had different flaA-RFLP restriction patterns. Therefore, a combination of fingerprinting methods appears to be necessary for higher discrimination of isolates. But the argument for the continuous differentiation between C. jejuni and C. coli is becoming more difficult to justify under the new findings that suggest a strong convergence between these two species (Sheppard et al., 2008). It appears that the speciation of C. jejuni and C. coli should be revisited, especially considering that these two bacteria produce the same disease, are in the same niche and are transmitted by the same routes. Although in most cases C. jejuni and C. coli isolates from broiler meat tend to cluster independently when analyzed with PFGE, some isolates do become difficult to interpret because they may be confirmed as C. jejuni by molecular methods but cluster together with C. coli (Figure 2). These findings will continue to challenge our definition of species boundary between two closely related bacteria. This important discussion should be addressed before implementing any particular policy for the presence of Campylobacter spp. in food samples. In the USA, the current public health policy for other foodborne pathogens (e.g. Salmonella) is not based on species or serotypes. Only for those cases where one species is pathogenic and others are not (e.g. Listeria spp.) does speciation become necessary. Characterization of Campylobacter isolates with flaA-RFLP has several advantages, such as low cost, simplicity and a good level of standardization among laboratories (Djordjevic et al., 2007). However, flaA-RFLP usually yield less discriminatory power for C. jejuni isolates than PFGE based on SmaI (Han et al., 2007). 52 Surprisingly, we obtained more restriction profiles with flaA-RFLP than with PFGE. These results may be explained in part because most of the isolates were C. coli. To summarize, our results emphasize the need to complement the identification of isolates by biochemical tests with molecular methods if speciation is necessary. The choice of a fingerprinting profile technique sometimes relates to the techniques already in use in some laboratories. However, a thorough investigation should include at least two different techniques because in most cases those techniques will complement each other and yield more discriminatory results. 53 References Anonymous. 1998. Grenada. Pan American Health Organization. Available at: http://www.paho.org/English/HIA1998/Grenada.pdf. Anonymous. 2008. CARICOM chicken meat and egg consumption 2002, MT. Caribbean Poultry Association. Available at: http://www.caribbeanpoultry.org/docs/cpi/CARICOM_Poultry_Consumption_20 02.pdf. David, A. 2004. Livestock genetics resources report, incomplete draft. Government of Grenada, Ministry of Agriculture, Lands, Forestry, and Fisheries, Food and Agriculture Organization. Available at: ftp://ftp.fao.org/docrep/fao/010/a1250e/annexes/CountryReports/Grenada.pdf. Debruyne, L., E. Samyn, E. De Brandt, O. Vandenberg, M. Heyndrickx, and P. Vandamme. 2008. Comparative performance of different PCR assays for the identification of Campylobacter jejuni and Campylobacter coli. Res. Microbiol. 159: 88-93. Djordjevic, S. P., L. E. Unicomb, P. J. Adamson, L. Mickan, R. Rios, and the Australian Campylobacter Study Group. 2007. Clonal complexes of Campylobacter jejuni identified by multilocus sequence typing are reliably predicted by restriction fragment length polymorphism analyses of the flaA gene. J. Clin. Microbiol. 45:102-108. 54 Gonzalez, I., Grant, K.A., Richardson, P.T., Park, S.F. and Collins, M.D., 1997. Specific identification of the enteropathogens Campylobacter jejuni and Campylobacter coli by using a PCR test based on the ceuE gene encoding a putative virulence determinant. J. Clin. Microbiol. 35:759-763. Han, K., S. S. Jang, E. Choo, S. Heu, S. Ryu. 2007. Prevalence, genetic diversity, and antibiotic resistance patterns of Campylobacter jejuni from retail raw chickens in Korea. Int. J. Food Microbiol. 114:50-59. Hani, E. K. and V. L. Chan, 1995. Expression and characterization of Campylobacter jejuni benzoylglycine amidohydrolase (Hippuricase) gene in Escherichia coli. J. Bacteriol. 177:2396-2402. Hariharan H., S. Sharma, A. Chikweto, V. Matthew, and C. Deallie. 2008. Antimicrobial drug resistance as determined by the E-test in Campylobacter jejuni, C. coli, and C. lari isolates from the ceca of broiler and layer chickens in Grenada. Comp. Immunol. Microbiol. Infect. Dis. (early online publication: doi:10.1016/j.cimid.2008.01.010). Harrington, C. S., L. Moran, A. M. Ridley, D. G. Newell, and R. H. Madden. 2003. Inter- laboratory evaluation of three flagellin PCR/RFLP methods for typing Campylobacter jejuni and C. coli: the CAMPYNET experience. J. Appl. Microbiol. 95:1321-1333. Harvey, S. M., 1980. Hippurate hydrolysis by Campylobacter fetus. J. Clin. Microbiol. 11:435-437. 55 Hunter, P. R. and M. A. Gaston. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-2466. Linton, D., Lawson, A.J., Owen, R.J., Stanley, J., 1997. PCR detection, identification to species level, and fingerprinting of Campylobacter jejuni and Campylobacter coli direct from diarrheic samples. J. Clin. Microbiol. 35:2568-2572. Nachamkin, I., K. Bohachick, and C. M. Patton. 1993. Flagellin gene typing of Campylobacter jejuni by restriction fragment length polymorphism analysis. J. Clin. Microbiol. 31:1531-1536. Nayak, R., Stewart, T. M., Nawaz, M.S. 2005. PCR identification of Campylobacter coli and Campylobacter jejuni by partial sequencing of virulence genes. Mol. and Cell. Probes 19:187-193. Oyarzabal, O. A., I. V. Wesley, J. M. Barbaree, L. H. Lauerman, and D. E. Conner. 1997. Specific detection of Campylobacter lari by PCR. J. Microbiol. Methods 29:97- 102. Oyarzabal, O. A., S. Backert, M. Nagaraj, R. S. Miller, S. K. Hussain, and E. A. Oyarzabal. 2007. Efficacy of supplemented buffered peptone water for the isolation of Campylobacter jejuni and C. coli from broiler retail products. J. Microbiol. Methods 69:129-136. Oyarzabal, O. A., K. S. Macklin, J. M. Barbaree and R. S. Miller. 2005. Evaluation of agar plates for direct enumeration of Campylobacter spp. from poultry carcass rinses. Appl. Environ. Microbiol. 71:3351-3354. 56 Oyarzabal, O. A., S. Backert, L. L. Williams, A. J. Lastovica, R. S. Miller, S. J. Pierce, S. L. Vieira, and F. Rebollo-Carrato. 2008. Molecular typing of Campylobacter jejuni strains isolated from commercial broilers in Puerto Rico. J. Appl. Microbiol. 105:800-812. Persson, S, Olsen, K. E. P., 2005. Multiplex PCR for identification of Campylobacter coli and Campylobacter jejuni from pure cultures and directly on stool samples. J. Med. Microbiol. 54:1043-1047. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. T. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. l39:1889-1894. Sheppard, S. K., N. D. McCarthy, D. Falush, and M. C. J. Maiden. 2008. Convergence of Campylobacter species implications of bacterial evolution. Science 320:237-239. Smith, K., N. Reimers, J. H. Barnes, B. C. Lee, R. Siletzky, and S. Kathariou. 2004. Campylobacter colonization of sibling turkey flocks reared under different management conditions. J. Food Prot. 67:1463-1468. Totten, P. A., Patton, C. M., Tenover, F. C., Barrett, T. J., Stamm, W. E., Steigerwalt, A. G., et al. 1987. Prevalence and characterization of hippurate-negative Campylobacter jejuni in King County, Washington. J. Clin. Microbiol. 25:1747- 1752. Winters, D. K., Slavik, M. F., 1995. Evaluation of a PCR based assay for specific detection of Campylobacter jejuni in chicken washes. Mol. and Cell. Probes. 9:307-310. 57 Figure Captions Figure 1. PFGE macrorestriction (A) and flaA-RFLP restriction (B) of strains G185 (lines 3, 4, 8, 9) and G187 (lines 1, 2, 5, 6). Mix cultures (lines 1, 3, 5, 9) show more bands than pure cultures. Isolate G187 did not restrict with PFGE but showed a restriction pattern with flaA-RFLP. After obtaining a single culture, strain G185 was C. coli and G 187 was C. jejuni. Ma: DNA size marker for PFGE made from the digestion of Salmonella choleraesuis ss. Choleraesuis serotype Braenderup H9812 (ATCC BAA- 664); Mb: 100-bp marker (Promega, Madison WI). Figure 2. flaA-RFLP restriction profiles of C. coli and C. jejuni isolates from Grenada. Pair comparison and cluster analysis performed with the Dice correlation coefficient and the unweighted pair group mathematical average clustering algorithm of BioNumerics version 4.50 (Applied Maths, Austin, TX). The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used for the determination of the different PFGE patterns. 58 Figure 1 A B 1 2 Ma 3 4 5 6 Mb 8 9 59 Figure 2 60 Table 1. Identification of 17 Campylobacter isolates with discrepancies between biochemical tests and multiplex polymerase chain reaction (mPCR) assays. Number of Strains Biochemical ID mPCR ID Presence of Hippuricase Gene Hippurate Activity1 13 C. jejuni C. coli - - 3 C. lari C. coli - - 1 C. coli C. jejuni + + 1 Hippurate activity was determined using a commercial hippurate test (Hardy Diagnostics, Santa Maria, CA). 61 Table 2. Evaluation of the discriminatory power of different typing techniques applied to Campylobacter spp. isolates from Grenada. Typing Method Number of Types Strains per Type Simpson?s Index flaA-RFLP 31 2,1,2,1,1,1,1,1,1,1,1,2,5,3,1, 1,2,1,1,1,1,1,1,3,1,3,1,9,1,1, 3 0.958 PFGE 26 1,1,1,1,1,4,1,1,2,5,5,9,1,1,1, 1,1,3,1,1,3,1,1,1,1,1 0.9437 flaA-RFLP/PFGE 48 2,1,1,1,1,1,1,1,1,1,1,1,1,2,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,2,1,1,1,1,1,1,2,5,2,1,1,1, 1,1,1 0.9906 62 IV. FINGERPRINT PROFILING OF CAMPYLOBACTER JEJUNI AND CAMPYLOBACTER COLI ISOLATED FROM RETAIL BROILER MEAT IN ALABAMA Introduction The southeastern region of the US produces more broiler meat than the rest of the country combined (Anonymous 2005). According to the U. S. Department of Agriculture (USDA), 17 states account for 94% of all US broiler production (Anonymous 2008a), and nine of those states are found in the southeast (Anonymous 2005). With US broiler meat consumption increasing to 86 lbs. per person in 2006 (Anonymous 2008a) and the CDC estimating 2.4 million infections in 2006 (Anonymous 2008b), Campylobacter in broiler meat is rapidly becoming an issue of concern from the public health point of view. There is little information on the revalence of different Campylobacter spp. in broiler meat in the Southeast USA. In addition, we wanted to evaluate different DNA- based techniques for characterization of isolates to then compare the resutls with those obtained from the use of the same techniuqes in strains collected in other geographical areas (Puerto rico and Grenada). The goal of this study was to determine reaction the species of campylobacters present in broiler meat and to characterize them molecularly via pulsed-field gel electrophoresis (PFGE), multilocus sequencing typing (MLST), and flagellin A restriction fragment length polymorphism (flaA-RFLP). We purchased retail broiler meat from local supermarkets in Auburn, Alabama, during the summer months of 63 2005, 2006, and 2007. The purchased meat was processed at two commercial broiler facilities referred hitherto as Plant A and Plant B. Materials and Methods Sample collection A total of 126 broiler retail samples processed at two different broiler processing plants were purchased from different retail stores in Auburn, Alabama, between June and August over a period of three years, 2005-2007. Samples were collected once a week and consisted of breast tenders, boneless breasts and thighs (including boneless thighs). Samples were stored at ~4?C until processing within 5 h of purchase. Campylobacter isolation Twenty five grams of sample were aseptically removed from each package and placed in individual sterile Whirl-Pak? bags (Nasco, Fort Atkinson, WI). Bolton broth (100 mL) supplemented with 5% lysed horse blood (School of Veterinary Medcine, Auburn University) was added to the samples. The mix was stomached for one minute in a Laboratory Blender Stomacher 400 (Seward, London, UK). Samples with the enrichment broth were incubated under microaerobiosis conditions (10% CO2, 5% O2, and 85% N2; Airgas, Radnor, PA) generated using anaerobic jars gassed with a MACSmics Jar Gassing System (Microbiology International,Frederick, MD) at 42?C for 48 h. The enriched samples were then swabbed onto modified Campy-Cefex (mCC) plates and were incubated under microaerobiosis at 42?C for 48h. 64 Identification by PCR Presumptive identification was done by selecting colonies and observing them under phase contrast microscopy to determine cell morphology and motility. Presumptive colonies were stored at -80oC for further analysis. Stock cultures were transferred to mCC plates which were incubated under microaerobic conditions at 42oC for 24 h to prepare fresh bacterial DNA. DNA was extracted using PrepMan? Ultra (Applied Biosystems, Foster City, CA) and tested with two multiplex PCR assays that were performed in 25 l aliquots as described elsewhere (Oyarzabal et al., 2005, 2007). A PTC-100 Programmable Thermal Controller (Bio-Rad Laboratories, Hercules, CA) was used for amplification. Amplicons were detected in 1.5% UltraPure? Agarose-1000 (Invitrogen Corporation, Carlsbad, CA) and DNA bands were stained with ethidium bromide, visualized using a UV transilluminator (Gel-Doc System, Syngene, Frederick, MD), followed by recording using a computer program (GeneSnap) for molecular analysis (Syngene, Frederick, MD). PFGE analysis A standard PFGE protocol (Ribot et al., 2001) with the modifications suggested by the Centers for Disease Control and Prevention (http://www.cdc.gov/pulsenet/protocols. htm) was used. A digestion of Salmonella choleraesuis ss. Choleraesuis serotype Braenderup H9812 (ATCC BAA-664) was used as the DNA size marker. This marker was included in three lanes per each PFGE gel, the two lanes at the end and one in the middle. CampylobacterDNA was digested with SmaI. Restricted DNA fragments were separated with a contour-clamped homogeneous electric field (CHEF Mapper, Bio-Rad, Hercules, 65 CA) in 1% SeaKem? Gold Agarose (Lonza, Basel, Switzerland) gel stained with ethidium bromide, visualized with a UV transilluminator (Gel-Doc System, Bio-Rad), and a picture recorded using GeneSnap (Syngene, Frederick, MD). PFGE profiles were used to perform a pair comparison and cluster analysis using the Dice correlation coefficient, and a dendogram was created using the unweighted pair group method using mathematical average (UPGMA) clustering algorithm of BioNumerics version 4.50 (Applied Maths, Austin, TX). The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used to analyze the different PFGE patterns (De Boer et al., 2000). MLST, sequence analysis, and phylogenetic calculations MLST was performed following the guidelines available at the C. jejuni MLST website (http://pubmlst.org/ campylobacter/) and guidelines from previous publications (Dingle et al., 2001). PCR amplification was performed in 25-?l reactions using two pre-made PCR mixes, OmniMix? HS (Cephied, Sunnyvale, CA) and Platinum PCR Supermix (Invitrogen Corporation). Amplicons were detected on 1.5% ethidium bromide agarose gels, and the concentration of amplified DNA was measured with a NanoDrop? ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE). Sequencing of the amplified products using the suggested internal primers was performed at the Auburn University Genetic Analysis Laboratory and Lucigen Corporation. Sequences were aligned, assembled and edited using MultAlin (http://prodes.toulouse.inra.fr/multalin/multalin.html). 66 flaA-RFLP Flagellin A RFLP was carried out in 25 l aliquots as described elsewhere (Harrington et al 2003). A PTC-100 Programmable Thermal Controller (Bio-Rad Laboratories, Hercules, CA) was used for amplification. Amplicons were detected in 2.5% SeaKem? Gold Agarose (Lonza, Basel, Switzerland) and DNA bands were stained with ethidium bromide, visualized using a UV transilluminator (Gel-Doc System, Syngene, Frederick, MD) followed by recording using a computer program (GeneSnap) for molecular analysis (Syngene, Frederick, MD). Statistical analysis A chi square calculation was done with the percentages positive from each processing plant (A and B). Results and Discussion Strain identification Out of the 126 samples, 56 samples had presumptive Campylobacter isolates, which were identified by enrichment, selective plates, and phase contrast microscopy (Table 1). Of the 56 presumptive Campylobacter isolates, the multiplex PCR protocols identified twenty four strains as C. jejuni and thirty one as C. coli. One strain could not be stored and was lost. A higher percentage of C. coli strains (56% versus 44% C. jejuni), were isolated from retail samples compared to the ration usually found from broiler carcass rinses, in which the ratio is approximately 97% C. jejuni and the rest C. coli (Oyarzabal et al. 67 2005). Others reports have also shown a much higher percentage of C. jejuni to C. coli for processed carcasses in the US (Fitch 2005; Son et al 2007). Yet, the ratio is different when we analyze retails samples. This change in the ratio between the two Campylobacter spp. appears to be true also in some European countries. For instance, Madden et, al. (1998) found C.jejuni and C. coli percentages of 56% and 44%, respectively, in retail chicken in Northern Ireland. We still do not understand all the epidemiological events related to these different ratios found in the same products at different stages over the production chain (processed versus retail). However there appears to be a real change in the ratio because they have been shown using the same techniques for identification and speciation of isolates (Oyarzabal et al. 2005, 2007). Of the three kinds of broiler meat tested from both plant A and plant B studied (breast tenders, boneless breasts and thighs), tenders had the lowest occurrence of Campylobacter contamination, with only 27% of total tenders testing positive for Campylobacter (Table 1). Tenders are cut from the Pectoralis minor located inside the chicken breast and underneath the Pectoralis major, which makes up the breast meat. Therefore, tenders are less likely to become contaminated with Campylobacter from intestinal contents during processing. Forty three percent of boneless breasts were Campylobacter positive followed by 79% of thighs testing positive for Campylobacter (Table 1). Since both of these cuts have exterior surfaces that are exposed during evisceration, they have more opportunities to come in contact with Campylobacter contaminated material (feces) or contaminated carcasses. Boneless breasts from plant A had a higher percentage of Campylobacter positive 68 samples than boneless breasts from plant B (18% versus 9%, respectively). A chi square analysis showed that this difference was not significant, x2=3.000 (Table 1). Thighs from plant B had a higher percentage of Campylobacter positive samples than thighs from plant A (18% versus 9%, respectively). A chi square analysis showed that this difference was not significant, x2=1.667 (Table 1). PFGE analysis Thirty six macrorestriction profiles were obtained from the 55 strains that were tested using SmaI restriction and a similarity cutoff of 90%. Each profile comprised eight to 11 electrophoretic bands. All 55 strains showed electrophoretic bands when digested with SmaI. The discriminatory power of PFGE in this study was 0.9805. No more than 5 strains belonged to one unique profile. Therefore, no dominant strain of C. jejuni or C. coli was present in commercially processed retail broiler meat produced in these two plants during the sampled time. Over time the number of PFGE profiles was variable when considering total strains from both plants. The highest number of PFGE profiles (19) was found in 2007. One strain, 1084, which was identified twice by mPCRs as C. coli, showed a PFGE profile that falls into the C. jejuni grouping (Figure 1). This strain did not show any amplification with the PCR targeting the flaA gene and therefore no RFLP profile could be obtained for this strain. To corroborate that strain 1084 is indeed C. coli, we will amplify and sequence its 16S rDNA and performed ClustalW analysis for identification. We will also amplify the flaA gene of this strain to find out the actual primer set that could be used when strains do not amplify the flaA gene. 69 We have previously seen that PFGE profiling is not completely discriminatory to group C. jejuni and C. coli in different PFGE clusters. An analysis of a large database for MLST sequences has recently shown that the patterns of genetic exchange between C. jejuni and C. coli can be interpreted as ?converging?. Thus, the factors that historically prevented the gene flow between these two species are apparently disrupted and these two species are merging or ?despeciating?. The consequence of this genetic merge is that any DNA-based method for fingerprinting may not be able to cluster these species in different groupings anymore (Sheppard et al., 2008). MLST analysis Twelve strains were analyzed by MLST with the rest of the strains scheduled for completion of analysis before the publication of this data. Table 2 shows the allelic profiles and clonal complexes found for the twelve strains. Two of the ST allelic profiles are new profiles that have not been assigned a ST value yet, although the actual allele profile for each gene has been previously described in the MLST database. The two new profiles are in the process of being submitted to the MLST web site for identification. Four strains, the largest cluster, belong to ST allelic profile 353 (Table 2). There was one strain in ST allelic profiles 354, 924, 939, 1210, 1212 and 3510. According to epidemiological data, ST allelic profiles 353, 354, 1212, 1210, and 3510 have been linked to human campylobacteriosis in Europe, Australia, and North America. ST allelic profiles 924 and 939 have been found in chickens from the UK, Canada, and the US (Anonymous 2008c). 70 flaA-RFLP analysis In this study one strain (1084) failed to digest with DdeI, yielding 34 profiles based on a 90% similarity cutoff. The discriminatory power of flaA-RFLP (0.9783) was very similar to PFGE (0.9805). Unlike the powerful discrimination found for C. coli by RFLP in the strains from Grenada (Chapter III), flaA-RFLP was less discriminatory than PFGE for retails isolates of C. coli. Because flaA-RFLP is a typing process based only on the flagellin A gene, it is no surprise that C. jejuni and C. coli discrimination is not always possible considering that C. jejuni and C. coli share a common biological niche and have a high horizontal gene transfer. The fact that flaA-RFLP grouped C. jejuni and C. coli strains together (Figure 2) emphasize the fact that continuous DNA exchange between these two species makes it difficult to separate fingerprinting profiles. These two species shared the same niche (the only Campylobacter strains transmitted by commercial broiler products), use the same transmission routes and cause similar disease in humans. Therefore, a question is raised on the practicality of speciating Campylobacter isolates from broiler products. Comparisons of numbers of profiles produced over time and plant by plant were gave ambiguous results. flaA-RFLP and PFGE outperformed each other two of the three years, but both yielded 19 profiles for the 2007 strains (Table 3). When analyzing the strains originating from only one plant, fla-A RFLP produced 18 profiles to PFGE?s 14 for plant B (Table 4). 71 References Anonymous. 2005. Economic data. U.S. Poultry and Egg Association 2008. Available at: http://www.poultryegg.org/economic_data/. Anonymous. 2008a. Economic Organization of U.S. Broiler Production. In ERS Report Summary. USDA Economic Research Service June 2008. Available at: http://www.ers.usda.gov/Publications/EIB38/EIB38_ReportSummary.pdf. Anonymous. 2008b. Campylobacter. CDC May 2008. Division of Foodborne Bacterial and Mycotic Diseases. Available at: http://www.cdc.gov/nczved/dfbmd/disease_listing/campylobacter_ti.html. Anonymous. 2008c. ST query. Campylobacter jejuni and Campylobacter coli PubMLST database August 2008. Available at http://pubmlst.org/perl/mlstdbnet/mlstdbnet.pl?file=pub-cj_profiles.xml&page=st- query. Bandelt, H. J. and A. W Dress. 1992. Split decomposition: a new and useful approach to phylogenetic analysis of distance data. Mol. Phylogenet. Evol. 1:242-252. De Boer, P., B. Duim, A. Rigter, J. van der Plas, W. Jacobs-Reitsma, and J. A. Wagenaar. 2000. Computer-assisted analysis and epidemiological value of genotyping methods for Campylobacter jejuni and Campylobacter coli. J. Clin. Microbiol. 38:1940-1946. 72 Dingle, K. E., F. M. Colles, D. R. Wareing, R. Ure, A. J. Fox, F. E. Bolton, H. J. Bootsma, R. J. Willems, R. Urwin, and M. C. J. Maiden. 2001. Multilocus sequence typing system for Campylobacter jejuni. J. Clin. Microbiol. 39:14?23. Fitch, B. R., K. L. Sachen, S. R. Wilder, M. A. Burg, D. W. Lacher, W.T. Khalife, T. S. Whitman, and V. B. Young. 2005. Genetic diversity of Campylobacter sp. Isolates from retail chicken products and humans with gastroenteritis in central Michigan. J. Clin. Microbiol. 43:4221-4224. Harrington, C. S., L. Moran, A. M. Ridley, D. G. Newell, and R. H. Madden. 2003. Inter- laboratory evaluation of three flagellin PCR/RFLP methods for typing Campylobacter jejuni and C. coli: the CAMPYNET experience. J. Appl. Microbiol. 95:1321-1333. Haubold, B., and R. R. Hudson 2000. LIAN 3.0: detecting linkage disequilibrium in multilocus data. Linkage Analysis Bioinformatics 16:847-848. Madden, R. H., L. Moran, and P. Scates. 1998. Frequency of occurrence of Campylobacter spp. in red meats and poultry in Northern Ireland and their subsequent subtyping using polymerase chain reaction-restriction fragment length polymorphism and the random amplified polymorphic DNA method. J. Appl. Microbiol. 84:703-708. Oyarzabal, O. A., K. S. Macklin, J. M. Barbaree and R. S. Miller. 2005. Evaluation of agar plates for direct enumeration of Campylobacter spp. from poultry carcass rinses. Appl. Environ. Microbiol. 71:3351-3354. 73 Oyarzabal, O. A., S. Backert, M. Nagaraj, R. S. Miller, S. K. Hussain, and E. A. Oyarzabal. 2007. Efficacy of supplemented buffered peptone water for the isolation of Campylobacter jejuni and C. coli from broiler retail products. J. Microbiol. Methods 69:129-136. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. T. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. l39:1889-1894. Sheppard, S. K., N. D. McCarthy, D. Falush, and M. C. J. Maiden. 2008. Convergence of Campylobacter species implications of bacterial evolution. Science 320:237-239. Son, I., M. D. Englen, M. E. Berrang, P. J. Fedorka-Cray, and M. A. Harrison. 2007. Prevalence of Arcobacter and Campylobacter on broiler carcasses during processing. Int. J. Food Microbiol. 113:16-22. 74 Figure Captions Figure 1 PFGE dendogram of C. jejuni and C. coli from plant A and B (2005-2007). Cluster analysis was performed with the Dice correlation coefficient and the unweighted pair group mathematical average clustering algorithm of BioNumerics version 4.50. The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used for the determination of the different PFGE patterns. Figure 2 flaA-RFLP dendogram of C. jejuni and C. coli from plants A and B (2005- 2007). Cluster analysis was performed with the Dice correlation coefficient and the unweighted pair group mathematical average clustering algorithm of BioNumerics version 4.50. The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used for the determination of the different flaA-RFLP patterns. 75 Figure 1 76 Figure 2 77 Table 1 Positive samples from plants A and B No. of samples Negative Positive % Total Plant A Plant B Product Tenders 45 31 14 (27) 8 6 Boneless breast 62 35 27 (43) 18 9 Thighs (including boneless thighs) 19 4 15 (79) 5 10 Total 126 70 56 (50) 31 (24) 25 (26) 78 Table 2 Allellic profiles and clonal complexes from C. jejuni and C. coli strains isolated in this study. ST Allelic Profile Clonal Complex Strains ?a ? 1 ?a ? 1 353 353 4 354 354 1 1212 607 1 1210 353 1 924 607 1 939 353 1 3510 353 1 a New ST profile. These alleles have not been submitted to the MLST web site for identification. 79 Table 3 Number of profiles per technique per year Year Number of profiles per technique per year PFGE flaA-RFLP MLST* 2005 13 10 3 2006 7 9 4 2007 19 19 5 * Only 12 strains were analyzed with MLST. 80 Table 4 Number of profiles per technique per processing plant Plant Technique PFGE flaA-RFLP MLST* A 23 21 6 B 14 18 6 * Only 12 strains were analyzed with MLST. 81 V. SUMMARY - EVALUATION OF FINGERPRINT PROFILING OF CAMPYLOBACTER JEJUNI AND CAMPYLOBACTER COLI FROM DIFFERENT GEOGRAPHICAL AREAS USING PULSED-FIELD GEL ELECTROPHORESIS When strains from all three geographic areas were analyzed, 121 strains grouped into 58 profiles. As stated earlier, failure of PFGE enzymes to digest the genomic DNA of some strains is the largest drawback to PFGE as a molecular subtyping technique. Out of 139 strains analyzed in these studies, two C. jejuni strains from Puerto Rico, five C. jejuni strains from Grenada, and eleven C. coli strains from Grenada failed to digest with SmaI. As seen in the Puerto Rico data, the problem can be overcome to some degree by using a combination of restriction enzymes rather than a single one. Only some samples from the same geographic areas clustered at 100% identity, but clusters from the same geographic areas did not group together (Figure 1). One explanation for this is that both Puerto Rico and Grenada import a large percentage of their consumed broiler meat from the US, but both also produce a small amount of broiler meat domestically, allowing for a more diverse Campylobacter population on the two islands. More likely the diversity shown by PFGE between the three geographic areas is caused by Campylobacter?s naturally high frequency of recombination and mutation rate. The C. jejuni strains grouped into two groups split by the C. coli strains (Figure 1). The overall conclusion of this research is that one single fingerprinting method is not sufficient to identify relatedness among C. jejuni or C. coli isolates. The combination 82 of two restriction profiling methods overcomes the limitations that one method may have (low discrimination or lack of profiling with some strains). The inclusion of a sequence based method (MLST) does complement methods based on electrophoretic profiling and allow for the determination of clonality based on sequences from seven house-keeping genes. Yet, MLST is not discriminatory enough to assess the special relatedness of Campylobacter strains. MLST may be used for the temporal analysis of strains but the frequent horizontal rearrangements that these bacteria undergo makes it difficult to assess which clone is the oldest in phylogenetic studies. There is no ?quality control? procedures that have been established for the analysis of electrophoretic profiling (PFGE, RFLP, etc.) of bacterial isolates, including Campylobacter isolates. The scientific literature related to the use of these techniques for DNA relatedness emphasize the standardization of the actual eletrophoretic run (Ribot et al., 2001) but not too much attention has been paid to the limitation of DNA extraction. In the case of PFGE, the methylation of adenine in GATC sequences does limit the action of some restriction enzymes and result in some strains being refractory to PFGE analysis (Edmonds et al., 1992). In the case of RFLP, the limitation is due to changes in the sequences where the primers attached and therefore result in the lack of amplification of some strains. We are analyzing the data by converting the electrohoretic bands numbers so that each band size is represented by a number. The numbers are expressed in kilo base pairs (kbp) for PFGE bands and in base pairs (bp) for RFLP bands. Our preliminary analysis of RFLP data from the Puerto Rico is shown in Table 1. Although most strains are within the 1,700 bp range (amplicon obtained for RFLP), some strains exhibited more bands that 83 altogether make up for more than 1,700 bp (see strains F1-H1-S1, F4-H1-S1 and F10-H1- S5 in Table 1). We are trying to interpret these data to determine if a partial digestion is found for some strains. We are also analyzing the possibility that two C. jejuni strains were collected and tested together, although the PFGE analysis does point to isolated strains being originally collected (observe the PFGE profiles of strains F1-H1-S1, F4-H1- S1 and F10-H1-S5 in Figure 2, Chapter 2). We hope our efforts will result in a series of guidelines to prepare samples for successful RFLP and PFGE analysis. 84 References Edmonds, P. B. M. Hall, W. R. Edwards, and K. M. Hartline. 1992. Presence of methylated adenine in GATC sequences in chromosomal DNAs from Campylobacter species. J. Bacteriol. 174: 8156-8157. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. T. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. l39:1889-1894. 85 Table 1 Band sizes and the sum of all bands for each fingerprinting profile collected with flaA-RFLP Strain ID Bands (bp) Sum (bp) F11-H1-S3 548 348 285 229 161 93 1664 F11-H1-S1 559 346 290 228 162 99 1683 F11-H1-S4 550 355 295 233 160 97 1692 F10-H1-S4 530 339 279 224 162 1534 F1-H1-S1 977 628 542 347 285 224 158 97 3258 F4-H1-S1 970 630 544 343 278 227 160 96 3249 F10-H1-S5 1053 764 551 344 285 230 170 98 3495 F10-H1-S3 742 259 223 166 1391 . 86 Figure Captions Figure 1 PFGE macrorestriction profiles (SmaI) of C. jejuni and C. coli isolates from Puerto Rico, Grenada, and southeastern US. Pair comparison and cluster analysis performed with the Dice correlation coefficient and the unweighted pair group mathematical average clustering algorithm of BioNumerics version 4.50 (Applied Maths, Austin, TX). The optimization and position tolerance for band analysis were set at 1%, and a cutoff of 90% was used for the determination of the different PFGE patterns. 87 Figure 1 88 REFERENCES Aanensen, D. M., B. G. Spratt. 2005. The multilocus sequence typing network: mlst.net. Nucleic Acids Res. 33:W728-W733. Allos, B. M. 2001. Campylobacter jejuni infections: update on emerging issues and trends. Clin. Infect. Dis.32:1201-1206. Anonymous 1998. Grenada. Pan American Health Organization. http://www.paho.org/ English/HIA1998/Grenada.pdf. Anonymous. 2005. Economic Data. U.S. Poultry and Egg Association 2008. http://www.poultryegg.org/economic_data/. Anonymous. 2005. Report on the consumption of broiler meat in Puerto Rico. Office of Agricultural Statistics, Department of Agriculture of Puerto Rico. Santurce, P.R. Anonymous. 2005. Foodborne illnesses. CDC Oct. 25. Coordinating Center for Infectious Diseases /Division of Bacterial and Mycotic Diseases. http://www.cdc.gov/ ncidod/dbmd/diseaseinfo/foodborneinfections_g.htm#howmanycase. Anonymous. 2007. Food safety and foodborne illness?. WHO Fact sheet No. 237. March. http://www.who.int/mediacentre/factsheets/fs237/en/. Anonymous. 2008. One-day (24-26 h) standardized laboratory protocol for molecular subtyping of Campylobacter jejuni by pulsed field gel electrophoresis. http://www.cdc.gov/pulsenet/protocols/campy_protocol.pdf. 89 Anonymous. 2008. Economic Organization of U.S. Broiler Production. In ERS Report Summary. USDA Economic Research Service June 2008. http://www.ers.usda. gov/Publications/EIB38/EIB38_ReportSummary.pdf. Anonymous 2008. Campylobacter. CDC May 2008. Division of Foodborne, Bacterial and Mycotic Diseases. http://www.cdc.gov/nczved/dfbmd/disease_listing/ campylobacter_ti.html. Anonymous 2008. CARICOM Chicken Meat and Egg Consumption 2002, MT. Caribbean Poultry Association. http://www.caribbeanpoultry.org/docs/cpi/ CARICOM_Poultry_Consumption_2002.pdf. Bandelt, H.J. and A.W Dress. 1992. Split decomposition: a new and useful approach to phylogenetic analysis of distance data. Mol. Phylogenet. Evol. 1:242-252. Bestor, T.H. 1990. DNA methylation: Evolution of a bacterial immune function into a regulator of gene expression and genome structure in higher eukaryotes. Phil. Trans. R. Soc. Lond. B 326:179-187. Butzler, J., and J Oosterom. 1991. Campylobacter: Pathogenicity and significance in foods. Int. J. Food Microbiol. 12:1-8. Buzby, J.C. 2001. Children and Microbial Foodborne Illness.Food Review 24:32-37. Butzler, J.-P. 2004. Campylobacter, from obscurity to celebrity. Clin. Microbiol. Infec. 10:868-876. Callicott, K. A., V. Fri?riksd?ttir, J. Reiersen, R. Lowman, J. Bisaillon, E. Gunnarsson, E. Berndtson, K. L. Hiett, D. S. Needleman, and N. J. Stern. 2006. Lack of evidence for vertical transmission of Campylobacter spp. in chickens. Appl. Environ. Microbiol. 72:5794-5798. 90 Chu, G., D. Vollrath, and R. W. Davis. 1986. Separation of large DNA molecules by contour-clamped homogeneous electric fields. Science 234:1582-1585. Coote, J. G., D. E. S Stewart-Tull, R. J. Owen, F. J. Bolton, B. L. Siemer, D. Candlish, D. H. Thompson, A. C. Wardlaw, S. L. W. On, A. Candlish, B. Billcliffe, P. J. Jordan, K. Kristiansen, and P. Borman. 2007. Comparison of virulence-associated in vitro properties of typed strains of Campylobacter jejuni from different sources. J. Med. Microbiol. 56:722-732. David, A. 2004. Livestock Genetics Resources Report, Incomplete Draft.Government of Grenada, Ministry of Agriculture, Lands, Forestry, and Fisheries, Food and Agriculture Organization. ftp://ftp.fao.org/docrep/fao/010/a1250e/annexes/ CountryReports/Grenada.pdf. De Boer, P., B. Duim, A. Rigter, J. van der Plas, W. Jacobs-Reitsma, and J. A. Wagenaar. 2000. Computer-assisted analysis and epidemiological value of genotyping methods for Campylobacter jejuni and Campylobacter coli. J. Clin. Microbiol. 38:1940-1946. Debruyne, L., E. Samyn, E. De Brandt, O. Vandenberg, M. Heyndrickx, and P. Vandamme. 2008. Comparative performance of different PCR assays for the identification of Campylobacter jejuni and Campylobacter coli. Res. Microbiol. 159: 88-93. Dickins, M. A., S. Franklin, R. Stefanova, G. E. Shutze, K. D. Eisenach, I. V. Wesley, and D. Cave. 2002. Diversity of Campylobacter isolates from retail poultry carcasses and from humans as demonstrated by pulsed-filed gel electrophoresis. J. Food Prot. 65:957-962. 91 Dingle, K. E., F. M. Colles, D. R. Wareing, R. Ure, A. J. Fox, F. E. Bolton, H. J. Bootsma, R. J. Willems, R. Urwin, and M. C. J. Maiden. 2001. Multilocus sequence typing system for Campylobacter jejuni. J. Clin. Microbiol. 39:14?23. Dingle, K. E., F. M. Colles, R. Ure, J. A. Wagenaar, B. Duim, F. J. Bolton, A. J. Fox, D. R. A. Wareing, and M. C. J. Maiden. 2002. Molecular characterization of Campylobacter jejuni clones: a basis for epidemiologic investigation. Emerg. Infect. Dis. 8:949-955. Djordjevic, S. P., L. E. Unicomb, P. J. Adamson, L. Mickan, R. Rios, and the Australian Campylobacter Study Group. 2007. Clonal complexes of Campylobacter jejuni identified by multilocus sequence typing are reliably predicted by restriction fragment length polymorphism analyses of the flaA gene. J. Clin. Microbiol. 45:102-108. Edmonds, P. B. M. Hall, W. R. Edwards, and K. M. Hartline. 1992. Presence of methylated adenine in GATC sequences in chromosomal DNAs from Campylobacter species. J. Bacteriol. 174: 8156-8157. Engberg, J., J. Neimann, E. M. Nielsen, F. M. Aarestrup, and V. Fussing. 2004. Quinolone-resistant Campylobacter infections in Denmark: Risk factors and clinical consequences. Emerg. Infect. Dis. 10:1056-1063. Fearnhead, P., N. G. C. Smith, M. Barrigas, A. Fox, and N. French. 2005. Analysis of recombination in Campylobacter jejuni from MLST population data. J. Mol. Evol. 61: 333?340. 92 Feil, E. J., E. C. Holmes, D. E. Bessen, M. -S. Chan, N. P. Day, M. C. Enright, R. Goldstein, D. W. Hood, A. Kalia, C. E. Moore, J. Zhou, and B. G. Spratt. 2001. Recombination within natural populations of pathogenic bacteria: short-term empirical estimates and long-term phylogenetic consequences. Proc. Natl. Acad. Sci. USA 98:182?187. Fitch, B. R., K. L. Sachen, S. R. Wilder, M. A. Burg, D. W. Lacher, W.T. Khalife, T. S. Whitman, and V. B. Young. 2005. Genetic diversity of Campylobacter sp. Isolates from retail chicken products and humans with gastroenteritis in central Michigan. J. Clin. Microbiol. 43:4221-4224. French, N. P., M. Barrigas, P. Brown, P. Ribiero, N. J. Williams, H. Leatherbarrow, R. Birtles, E. Bolton, P. Fearnhead, and A. Fox. 2005. Spatial epidemiology and natural population structure of Campylobacter jejuni colonizing a farmland ecosystem. Environ. Microbiol. 7:1116-1126. Gonzalez, I., Grant, K.A., Richardson, P.T., Park, S.F. and Collins, M.D., 1997. Specific identification of the enteropathogens Campylobacter jejuni and Campylobacter coli by using a PCR test based on the ceuE gene encoding a putative virulence determinant. J. Clin. Microbiol. 35:759-763. Gibson, J. R., K. Sutherland, and R. J. Owen. 1994. Inhibition of DNAase activity in PFGE analysis of Campylobacter jejuni. Lett. Appl. Microbiol. 19:357-358. Gilpin, B., A. Cornelius, B. Robson, N. Boxall, A. Ferguson, C. Nicol, and T. Henderson. 2006. Application of pulsed-field gel electrophoresis to identify potential outbreaks of campylobacteriosis in New Zealand. J. Clin. Microbiol. 44:406-412. 93 Hall, T. A. 1999. BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41:95-98. Han, K., S. S. Jang, E. Choo, S. Heu, S. Ryu. 2007. Prevalence, genetic diversity, and antibiotic resistance patterns of Campylobacter jejuni from retail raw chickens in Korea. Int. J. Food Microbiol. 114:50-59. Hani, E. K. and V. L. Chan, 1995. Expression and characterization of Campylobacter jejuni benzoylglycine amidohydrolase (Hippuricase) gene in Escherichia coli. J. Bacteriol. 177:2396-2402. Hariharan H., S. Sharma, A. Chikweto, V. Matthew, and C. Deallie. 2008. Antimicrobial drug resistance as determined by the E-test in Campylobacter jejuni, C. coli, and C. lari isolates from the ceca of broiler and layer chickens in Grenada. Comp. Immunol. Microbiol. Infect. Dis. (early online publication: doi:10.1016/j.cimid.2008.01.010). Harrington, C. S., L. Moran, A. M. Ridley, D. G. Newell, and R. H. Madden. 2003. Inter- laboratory evaluation of three flagellin PCR/RFLP methods for typing Campylobacter jejuni and C. coli: the CAMPYNET experience. J. Appl. Microbiol. 95:1321-1333. Harvey, S. M., 1980. Hippurate hydrolysis by Campylobacter fetus. J. Clin. Microbiol. 11:435-437. Haubold, B., and R. R. Hudson 2000. LIAN 3.0: detecting linkage disequilibrium in multilocus data. Linkage Analysis.Bioinformatics 16:847-848. 94 Ho, M. F., and A. P. Monaco. 1995. PFGE in physical mapping, pp 21-43. In A. P. Monaco (ed.), Pulsed field gel electrophoresis, a practical approach. Oxford University Press Inc., New York, NY. Hunter, P. R. and M. A. Gaston. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-2466. Huson, D. H. 1998. SplitsTree: analyzing and visualizing evolutionary data. Bioinformatics 14, 68-73 Hutchinson, D. N., and F. J. Bolton. 1984. An improved blood-free selective medium for isolation of Campylobacter jejuni from faecal specimens. J. Clin. Pathol. 37:956- 95. K?renlampi, R., H. Rautelin, and M. -L. H?nninen. 2007. Evaluation of genetic markers and molecular typing methods for prediction of sources of Campylobacter jejuni and C. coli infections. Appl. Environ. Microbiol. 73:1683-1685. Kelly, J. K. 1997. A test of neutrality based on interlocus associations. Genetics 146:1197?1206. Kist. M. 1986. Who discovered Campylobacter jejuni/coli? A review of hitherto disregarded literature. Zentralbl Bakteriol. Mikrobiol. Hyg. [A]. 261:177-186. Linton, D., Lawson, A.J., Owen, R.J., Stanley, J., 1997. PCR detection, identification to species level, and fingerprinting of Campylobacter jejuni and Campylobacter coli direct from diarrheic samples. J. Clin. Microbiol. 35:2568-2572. 95 Lopez-Ortiz, W., R. A. Solivan. 1999. Campylobacter jejuni among patients with gastroenteritis: Incidence at a reference microbiology laboratory in San Juan, Puerto Rico. P. R. Health Sci. J. 18:273-276. Lynch, M., J. Painter, R. Woodruff, C. Braden. 2006. Surveillance for foodborne-Disease outbreaks - United States, 1998-2002. MMWR 55(SS10):1-34. Madden, R. H., L. Moran, and P. Scates. 1998. Frequency of occurrence of Campylobacter spp. in red meats and poultry in Northern Ireland and their subsequent subtyping using polymerase chain reaction-restriction fragment length polymorphism and the random amplified polymorphic DNA method. J. Appl. Microbiol. 84:703-708. Maiden, M. C. J., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang, J. Zhou, K. Zurth, D. A. Caugant, I. M. Feavers, M. Achtman, and B. G. Spratt. 1998. Multilocus sequence typing: A portable approach to the identification of clones within populations of pathogenic microorganisms. Microbiology 95:3140- 3145. Maynard Smith, J. 1999. The detection and measurement of recombination from sequence data. Genetics 153:1021-1027. Maynard Smith, J., N. H. Smith, M. O?Rourke, and B. G. Spratt. 1993. How clonal are bacteria? Proc. Natl. Acad. Sci. USA 90:4384-4388. Meinersmann, R.J., L. O. Helsel, P. I. Fields, and K. L.Hiett. 1997. Discrimination of Campylobacter jejuni isolates by fla gene sequencing. J. Clin. Microbiol. 35:2810-2814. 96 Michaud, S., S. Menard, C. Gaudreau, and R. D. Arbeit. 2001. Comparison of SmaI- defined genotypes of Campylobacter jejuni examined by KpnI: a population- based study. J. Med. Microbiol. 50:1075-1081. Mickan, L., R. Doyle, M. Valcanis, K.E. Dingle, L. Unicomb, and J. Lanser. 2007. Multilocus sequence typing of Campylobacter jejuni isolates from New South Wales, Australia. J. Appl. Microbiol. 102:144-52. Miller, W. G., S. L. W. On, G. Wang, S. Fontanoz, A. J .Lastovica, and R. E. Mandrell. 2005. Multilocus sequence typing system for Campylobacter coli, C. lari, C. upsaliensis, and C. helveticus. J. Clin. Microbiol. 43:2315-2329. Musser, J. M. 1996. Molecular population genetic analysis of emerged bacterial pathogens: Selected insights. Emerg. Infect. Dis. 2:1-17. Nachamkin, I., K. Bohachick, and C. M. Patton. 1993. Flagellin gene typing of Campylobacter jejuni by restriction fragment length polymorphism analysis. J. Clin. Microbiol. 31:1531-1536. Nachamkin, I. 2003. Campylobacter and Arcobacter. p. 902-14. In: Murray PR, et al. (ed.). Manual of clinical microbiology, 8th ed., ASM Press, Washington, D.C. Nadeau, E., S. Messier, and S. Quessy. 2002. Prevalence and comparison of genetic profiles of Campylobacter strains isolated from poultry and sporadic cases of campylobacteriosis in humans. J. Food Prot. 65:73-78. Nadeau, E., S. Messier, and S. Quessy. 2003. Comparison of Campylobacter isolates from poultry and humans: association between in vitro virulence properties, biotypes and pulse-field gel electrophoresis clusters. Appl. Environ. Microbiol. l69:6316-6320. 97 Nayak, R., Stewart, T. M., Nawaz, M.S. 2005. PCR identification of Campylobacter coli and Campylobacter jejuni by partial sequencing of virulence genes. Molecular and Cellular Probes 19:187-193. Nielsen, E. M., V. Fussing, J. Engberg, N. L. Nielsen, and J. Neimann. 2006. Most Campylobacter subtypes from sporadic infections can be found in retail poultry products and food animals. Epidemiol. Infect. 134:758-767. Nielsen, E. M., J. Engberg, V. Fussing, L. Petersen, C. Brogren, and S. L. W. On.2000. Evaluation of phenotypic and genotypic methods for subtyping Campylobacter jejuni isolates from humans, poultry, and cattle. J. Clin. Microbiol. 38:3800-3810. Olsen S. J., L. C.. MacKinon, J. S. Goulding, N. H. Bean, L. Slutsker. 2002. Surveillance for foodborne disease outbreaks - United States, 1993-1997. MMWR 49(SS01):1- 51. On, S. L. W. 2001. Taxonomy of Campylobacter, Arcobacter, Helicobacter and related bacteria: Current status, future prospects and immediate concerns. J. Appl. Microbiol. 90:1S-15S. Oyarzabal, O. A., I. V. Wesley, J. M. Barbaree, L. H. Lauerman, and D. E. Conner. 1997. Specific detection of Campylobacter lari by PCR. J. Microbiol. Methods 29:97- 102. Oyarzabal, O. A., K. S. Macklin, J. M. Barbaree and R. S. Miller. 2005. Evaluation of agar plates for direct enumeration of Campylobacter spp. from poultry carcass rinses. Appl. Environ. Microbiol. 71:3351-3354. 98 Oyarzabal, O. A., S. Backert, M. Nagaraj, R. S. Miller, S. K. Hussain, and E. A. Oyarzabal. 2007. Efficacy of supplemented buffered peptone water for the isolation of Campylobacter jejuni and C. coli from broiler retail products. J. Microbiol. Methods 69:129-136. Oyarzabal, O. A., S. Backert, L. L. Williams, A. J. Lastovica, R. S. Miller, S. J. Pierce, S. L. Vieira, and F. Rebollo-Carrato. 2008. Molecular typing of Campylobacter jejuni strains isolated from commercial broilers in Puerto Rico. J. Appl. Microbiol. 105:800-812. Persson, S, Olsen, K. E. P., 2005. Multiplex PCR for identification of Campylobacter coli and Campylobacter jejuni from pure cultures and directly on stool samples. J. Med. Microbiol. 54:1043-1047. Potturi-Venkata, L. -P., S. Backert, S. L. Vieira, and O. A. Oyarzabal. 2007. Evaluation of logistic processing to reduce cross-contamination of commercial broiler carcasses with Campylobacter spp. J. Food Prot. 70:2549-2554. Potturi-Venkata, L. P., S. Backert, A. J. Lastovica, S. L. Vieira, R. A. Norton, R. S. Miller, S. J. Pierce, and O. A. Oyarzabal, 2007. Evaluation of different plate media for direct cultivation of Campylobacter spp. from live broilers. Poul. Sci. 86:1304-1311. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. T. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. l39:1889-1894. 99 Riley, L. W. 2004a. Pinciples and Approaches, pp. 1-28. In Molecular Epidemiology of Infectious Diseases: Principles and Practices. ASM Press, Washington D.C. Riley, L. W. 2004b. Analysis of Similarity and Relatedness in Molecular Epidemiology, pp. 91-124. In Molecular Epidemiology of Infectious Diseases: Principles and Practices. ASM Press, Washington D.C. Rozas, J., M. Gullaud, G. Blandin, and M. Aguad?. 2001, DNA variation at the rp49 gene region of Drosophila simulans: Evolutionary inferences from an unusual haplotype structure. Genetics 158:1147?1155. Sails, A. D., B. Swaminathan, and P. I. Fields. 2003. Utility of multilocus sequence typing as an epidemiological tool for investigation of outbreaks of gastroenteritis caused by Campylobacter jejuni. J. Clin. Microbiol. l41:4733-4739. Salazar-Lindo, E., R. B. Sack, E. Chea-Woo, B. A. Kay, Z. A. Piscoya, R. Leon-Barua, A. Yi. 1986. Early treatment with erythromycin of Campylobacter jejuni- associated dysentery in children. J. Pediatr. 109:355-360. Schouls, L. M., S. Reulen, B. Duim, J. A. Wagenaar, R. J. L. Willems, K. E. Dingle, F. M. Colles, and J. D. A. Van Embden. 2003. Comparative genotyping of Campylobacter jejuni by amplified fragment length polymorphism, multilocus sequence typing, and short repeat sequencing: strain diversity, host range, and recombination. J. Clin. Microbiol. l41:15-26. Sheppard, S. K., N. D. McCarthy, D. Falush, and M. C. J. Maiden. 2008. Convergence of Campylobacter species implications of bacterial evolution. Science 320:237-239. 100 Smith, K., N. Reimers, J. H. Barnes, B. C. Lee, R. Siletzky, and S. Kathariou. 2004. Campylobacter colonization of sibling turkey flocks reared under different management conditions. J. Food Prot. 67:1463-1468. Son, I., M. D. Englen, M. E. Berrang, P. J. Fedorka-Cray, and M. A. Harrison. 2007. Prevalence of Arcobacter and Campylobacter on broiler carcasses during processing. J. Food Microbiol. 113:16-22. Suerbaum, S., M. Lohrengel, A. Sonnevend, F. Ruberg, and M. Kist. 2001. Allelic diversity and recombination in Campylobacter jejuni. J. Bacteriol. 183:2553? 2559. Spratt, B. G., and M. C. J. Maiden. 1999. Bacterial population genetics, evolution and epidemiology. Phil. Trans. R .Soc. London B Biol. Sci. 354:701-710. Supply, P., R. M. Warren, A. -L. Ba?uls, S. Lesjean, G. D. van der Spuy, L. -A. Lewis, M. Tibayrenc, P. D. van Heldenand, and C. Locht. 2003. Linkage disequilibrium between minisatellite loci supports clonal evolution of Mycobacterium tuberculosis in a high tuberculosis incidence area. Mol. Microbiol. 47:529-538. Swaminathan, B., T. J. Barrett, S. B. Hunter, R. V. Tauxe, and the CDC PulseNet Task Force. 2001. PulseNet: The molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg. Infect. Dis. 7:382-389. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007.MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution, doi:10.1093/molbev/msm092. [Online.] http://www.megasoftware.net. Accessed 23 September 2007. 101 Tauxe, R. 1997. Emerging foodborne diseases: An evolving public health challenge. Emerg. Infect. Dis. 3:425-434. Tenover, F. C., R. D. Arbeit, R. V. Goering, P. A. Mickelsen, B. E. Murray, D. H. Persing, and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol. 33:2233-2239. Totten, P. A., Patton, C. M., Tenover, F. C., Barrett, T. J., Stamm, W. E., Steigerwalt, A. G., et al. 1987. Prevalence and characterization of hippurate-negative Campylobacter jejuni in King County, Washington. J. Clin. Microbiol. 25:1747- 1752. Winters, D. K., Slavik, M. F., 1995. Evaluation of a PCR based assay for specific detection of Campylobacter jejuni in chicken washes. Mol. and Cell. Probes. 9, 307-310. Wang, Y., E. P. C. Rocha, F. C. C. Leung, and A. Danchin. 2004. Cytosine methylation is not the major factor inducing CpG dinucleotide deficiency in bacterial genomes. J. Mol. Evol. 58:692-700. Young,V. B. and L. S. Mansfield. 2005. Campylobacter Infection ? Clinical Context, pp. 1-12. In J. M. Ketley and M. E. Konkel (ed.), Campylobacter: Molecular and Cellular Biology. Horizon Bioscience, Wymondham, Norfolk, U.K.