Characterization of Salmonella enterica serovar Heidelberg from Turkey-Associated Sources

Transcription

1 APPLIED A ENVIRONMENTAL MICROBIOLOGY, Aug. 2008, p Vol. 74, No /08/$ doi: /aem Copyright 2008, American Society for Microbiology. All Rights Reserved. Characterization of Salmonella enterica serovar Heidelberg from Turkey-Associated Sources Pravin Kaldhone, 1,2 Rajesh Nayak, Aaron M. Lynne, 1 Donna E. David, 1 Patrick F. McDermott, 4 Catherine M. Logue, 5 and Steven L. Foley 1,2 * National Farm Medicine Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin 1 ; Department of Biology, University of Central Arkansas, Conway, Arkansas 2 ; Division of Microbiology, National Center for Toxicological Research, U.S. Food and Drug Administration, Jefferson, Arkansas ; Division of Animal and Food Microbiology, Center for Veterinary Medicine, U.S. Food and Drug Administration, Laurel, Maryland 4 ; and Department of Veterinary and Microbiological Sciences, North Dakota State University, Fargo, North Dakota 5 Received 18 February 2008/Accepted 19 June 2008 Salmonella enterica serovar Heidelberg strains are frequently associated with food-borne illness, with recent isolates showing higher rates of resistance to multiple antimicrobial agents. One hundred eighty S. enterica serovar Heidelberg isolates, collected from turkey-associated production and processing sources, were tested for antimicrobial susceptibility and compared by pulsed-field gel electrophoresis (PFGE) and plasmid profile analysis. The potential for the transfer of resistance between strains was studied by conjugation experiments. PFGE analysis using XbaI digestion identified eight clusters (based on 90% similarity), with the largest containing 71% of the isolates. Forty-two percent of the isolates were resistant to at least 1 of the 15 antimicrobial agents tested, and 4% of the isolates were resistant to 8 or more antimicrobial agents. Resistances to streptomycin (2%), tetracycline (0%), and kanamycin (24%) were most commonly detected. Interestingly, the XbaI PFGE profiles of selective multidrug-resistant strains (n 22) of S. enterica serovar Heidelberg from turkey-associated sources were indistinguishable from the predominant profile (JF6X ) detected in isolates associated with human infections. These isolates were further differentiated into seven distinct profiles following digestion with the BlnI enzyme, with the largest cluster comprising 15 isolates from veterinary diagnostic and turkey processing environments. Conjugation experiments indicated that resistance to multiple antimicrobial agents was transferable among strains with diverse PFGE profiles. Salmonellosis is a significant health problem. In 1999, there were an estimated 1.4 million cases of Salmonella infections in the United States, which resulted in 17,000 hospitalizations and 585 deaths (18). Since then, the overall rate of salmonellosis in the United States has decreased by approximately 9%; however, the rate of Salmonella enterica serovar Heidelberg infections increased by 25% over the same period (5). S. enterica serovar Heidelberg ranks fourth among serotypes causing human salmonellosis (6) but is often the most commonly detected serotype among turkey and chicken Salmonella isolates submitted to the U.S. Department of Agriculture s National Veterinary Services Laboratory (NVSL) (10, 11). Annually, human infections with S. enterica serovar Heidelberg lead to approximately 84,000 cases of salmonellosis and contribute to approximately 7% of the Salmonella-related deaths in the United States, the second highest percentage after S. enterica serovar Typhimurium (1, 15). The most common source of S. enterica serovar Heidelberg infections is likely the consumption of undercooked or mishandled poultry products, such as turkey, chicken, and eggs (1). Salmonella surveillance data collected in poultry processing plants as part of U.S. Department of Agriculture s Pathogen Reduction-Hazard Analysis Critical Control Point Program showed that 11.4% of broiler chickens * Corresponding author. Mailing address: Marshfield Clinic Research Foundation, 1000 North Oak Avenue, Marshfield, WI Phone: (715) Fax: (715) Published ahead of print on 27 June and 7.1% of turkeys carried Salmonella infections in 2006 (2). These findings, coupled with the fact that there has been a nearly fourfold increase in the per capita consumption of poultry products in the United States over the past half century (), indicate that the contamination of poultry products with S. enterica serovar Heidelberg is a major public health concern. Surveillance data show that antimicrobial resistance among S. enterica serovar Heidelberg isolates has been rising. Data from the National Antimicrobial Resistance Monitoring System (NARMS) indicates that the percentage of S. enterica serovar Heidelberg isolates from humans and poultry (chicken and turkey) that were resistant to cephalosporins increased overall from to 200. For example, in, none of the isolates from humans and 1.6% of poultry isolates were resistant to ceftiofur (Tio); by 200, the numbers increased to 5.2% and 7.4%, respectively (9). This increase in cephalosporin resistance is likely associated with the spread of the AmpC -lactamase, which is encoded by bla CMY (1, 26, 29). The bla CMY gene has been associated with transmissible plasmids, which is important for the spread of cephalosporin resistance (26). In addition to cephalosporin resistance, Zhao et al. (28) found a number of S. enterica serovar Heidelberg isolates from retail meat that were resistant to additional agents, including ampicillin (Amp), amoxicillin-clavulanic acid (Amc), gentamicin (Gen), kanamycin (Kan), streptomycin (), sulfisoxazole (Sul), and tetracycline (Tet). Aarestrup et al. (1) were able to transfer bla CMY -mediated cephalosporin resistance along with resistance to Amp, Amc, chloramphenicol (Chl), Kan,, Sul, and Tet among S. enterica serovar Downloaded from on January 0, 2019 by guest 508

2 VOL. 74, 2008 SEROVAR HEIDELBERG FROM TURKEY-ASSOCIATED SOURCES 509 Heidelberg strains (1). Welch et al. (25) recently described the sequence of a large transmissible Inc A/C plasmid from S. enterica serovar Newport carrying 11 resistance determinants, and there is evidence that similar plasmids are widely distributed in food isolates of various Salmonella serovars, including S. enterica serovar Heidelberg. The spread of multidrug resistance among S. enterica serovar Heidelberg isolates is a risk to the management of salmonellosis in both veterinary and human clinical practice. Therefore, an increased understanding of pathogen distribution and mechanisms of antimicrobial resistance transmission is important for development of strategies to limit salmonellosis due to multidrug-resistant strains. The main objective of this study was to characterize the distribution of antimicrobial resistance among S. enterica serovar Heidelberg isolated from turkey-associated sources. Isolates collected from multiple sources were compared using pulsedfield gel electrophoresis (PFGE) and antimicrobial susceptibility testing (AST) and subjected to plasmid analysis to compare plasmid profiles among isolates with different levels of resistance. A diverse selection of multidrug-resistant isolates was analyzed by conjugation to determine the potential for horizontal gene spread among S. enterica serovar Heidelberg isolates. MATERIALS A METHODS Bacterial strains. One hundred eighty Salmonella enterica serovar Heidelberg isolates were collected from a diverse set of turkey-associated sources from 1992 to 200 (Fig. 1). A series of isolates originated from multiple locations on a single turkey farm, including waterers, litter, barn surfaces (walls, ventilation areas, fans, feed carts, floors, and door handles), and the ceca of turkeys raised on the farm, using methods described previously (19). Isolates from veterinary diagnostic specimens were obtained from the NVSL and the North Dakota State Veterinary Laboratory. Additionally, isolates were recovered from processing plant carcass swabs and from retail ground turkey samples as described by Logue et al. (16) and Zhao et al. (28), respectively. The isolates were stored in brain heart infusion broth and 20% glycerol at 80 C. Isolates were streaked on tryptic soy agar supplemented with 5% sheep s blood (blood agar; Remel, Lenexa, KS) and incubated (5 C for 18 to 24 h) for analysis. PFGE. PFGE was performed using XbaI (Promega, Madison, WI) according to the protocol described by Ribot et al. (22). Gel electrophoresis was carried out for 18 h at 200 V and 14 C, with an initial switch time of 2.16 s and a final switch time of 6.8 s. Restriction digestion patterns were analyzed using BioNumerics software (version 4.50; Applied Maths, Kortrijk, Belgium). The images were normalized to S. enterica serovar Braenderup H9812 in-run standards, and the relatedness of the gel band patterns was calculated using Dice-based coefficients with a 1% band tolerance and 1.56% optimization. A dendrogram was generated using the unweighted-pair group method with averages to determine relatedness between bacterial isolates. Additionally, a subset of isolates with identical genotypes following XbaI analysis (subcluster A1 in Fig. 1) was subjected to PFGE analysis following restriction with BlnI (Takara, Madison, WI), using the methods described above (22). AST. Agar disk diffusion AST was carried out as described by the CLSI (formerly NCCLS) (16) for the following antimicrobial agents: amikacin (Ami), Amc, Amp, cefoxitin (Fox), ceftriaxone (Axo), Chl, ciprofloxacin (Cip), Gen, Kan, nalidixic acid (Nal), Tio,, Sul, Tet, and trimethoprim-sulfamethoxazole (Sxt). The plates were incubated (5 C for 18 to 20 h), and the diameters of the zones of inhibition were measured and interpreted according to the CLSI guidelines (21). Escherichia coli ATCC and ATCC 5218 were used as quality control strains. Plasmid analysis. Plasmid DNA was isolated using the Wizard Plus SV Minipreps DNA purification system (Promega Corp., Madison, WI), following the manufacturer s protocols. Plasmids from isolates representing at least one member of each different susceptibility and PFGE profiles were also isolated by the method of Wang and Rossman (24) to detect the presence of large plasmids that may be associated with antimicrobial resistance (26). Plasmid DNA was separated in 0.7% agarose gels prepared with Tris-borate-EDTA buffer (Fisher Biochemicals, Baltimore, MD) at 70 V for h at room temperature and stained with ethidium bromide. A supercoiled DNA ladder (2- to 10-kb size; Promega Corp.) and isolated plasmids from E. coli NCTC 50192, which contains plasmids of well-characterized sizes, ranging from 7 to 154 kb (27), were separated on gels to size the plasmids. The stained gels were visualized under UV light and analyzed to determine the plasmid sizes based on the relative degrees of migration to the corresponding size standards. Conjugation experiments. S. enterica serovar Heidelberg 16,, and 710 were used as donor strains in filter mating experiments, using the method described by Clewell et al. (8) with minor modifications. The multidrug-resistant donor strains were susceptible to Nal and resistant to Amp and displayed distinct PFGE profiles. The recipient strain, S. enterica serovar Heidelberg 819, was resistant to Nal and susceptible to Amp. Overnight cultures of donor and recipient cells were prepared, and 0.5 ml of the recipient and 0.05 ml of the donor were added to 4.5 ml of fresh tryptic soy broth (Becton Dickinson, Sparks, MD). The mixtures were collected on 0.45-mm membrane filters (Millipore Corp., Burlington, MA). The filters were placed on blood agar plates and allowed to incubate (7 C for 18 to 20 h). Cells were removed from around each of the filters, suspended in tryptic soy broth, and spread on culture plates containing Nal (2 g/ml) and Amp (2 g/ml). The culture plates were incubated (7 C for 24 h) and observed for growth. Plasmid analysis (24), AST (21), and resistance gene detection were performed on isolates that grew on plates containing both antibiotics to characterize the transconjugants. Resistance gene detection was carried out using 19 sets of primers and methods as described previously (17). The resistance genes screened are indicated in Fig. 4A. Sequence-confirmed positive-control strains and no-template controls were included with each PCR run (17). Class 1 integrons were also detected, and the resistance gene insert was identified by DNA sequencing as described previously (29). S. enterica serovar Newport and were used as positive controls for the integron PCR (29). RESULTS A DISCUSSION In the current study of 180 isolates, 41% were resistant to at least one of the antimicrobial agents tested (Table 1). These numbers are higher than those observed among S. enterica serovar Heidelberg isolates collected from humans as part of the NARMS over the period that the majority of isolates in the present study were collected ( to 200) (9). Conversely, our resistance rates were lower than those observed for NARMS isolates from turkey and ground turkey, where typically more than 50% of isolates in a given year were resistant to at least one antimicrobial agent. In the present study, resistance was most often observed alone or in combination with (1%), Tet (29%), Gen (25%), Sul (21%), and Kan (21%). Additionally, five (%) isolates were resistant to the extendedspectrum cephalosporin Tio. Thirty-one isolates (17%) were resistant to at least five antimicrobial agents. There were 26 different susceptibility profiles detected among the isolates from the turkey-associated strains of S. enterica serovar Heidelberg examined, with the most common multidrug resistance profile observed displaying resistance to Tet, Gen, Sul, Kan, Amp, and (7.2%). Interestingly, many of these isolates with the common multidrug resistance profile are closely related by PFGE to a large group of pansusceptible isolates, differing from one another by a single band. Overall, 48 different XbaI PFGE patterns were observed among the 180 isolates (Fig. 1). The patterns were assigned to eight clusters (A to H) based on 90% similarity. For analysis, a cluster was defined as containing at least three isolates. The largest was cluster A, which contained 127 isolates (71%) obtained on the farm and from processing plants (Fig. 1). The remaining isolates were grouped into clusters B (n ; 2%), C (n 9; 5%), D (n ; 2%), E (n 9; 5%), F (n 4; 2%), G (n ; 2%), and H (n 4; 2%). There were 18 unique profiles that did not fall into any of the eight clusters. Downloaded from on January 0, 2019 by guest

3 Dice (Opt:1.56%) (Tol 1.0%-1.0%) (H>0.0% S>0.0%) [0.0%-100.0%] 5040 KALDHONE ET AL. APPL. ENVIRON. MICROBIOL. PFGE-XbaI PFGE-XbaI Antibiogram (sir) Chl Tet Axo Tio Fox Amp Amc Ami Gen Kan Sul Sxt Cip Nal Isolate Source Year State Plasmids (kb) , , 6, 4,, , A1 A Processing plant A (post-chill) Processing plant A (post-chill) Processing plant A (post-chill) Processing plant A (post-chill) 4, 5 4, 4, 4, 4, 4, 4, 6, 4, 7,, 2 4 Downloaded from A , 4, , 6, on January 0, 2019 by guest , 7, ,

4 VOL. 74, 2008 SEROVAR HEIDELBERG FROM TURKEY-ASSOCIATED SOURCES 5041 A2 B C D E F G H Isolate Source Year State Plasmids (kb) MD, 2 4, 4, 4, 4, 4 4, 7, 4, 6 4, 4, 70 Processing plant A (post-chill) 5, AR MO NC AR IL MO NC 4 8, 6, 5 6, 4 8, Processing plant A (pre-scaldl) MD NC NC OH MD MD 10, 7, 4 6 6, 4, 2 4, 8,, 2 4, 8,, 2, 2 6, 6, 6, 4, 4 6, Downloaded from on January 0, 2019 by guest

5 5042 KALDHONE ET AL. APPL. ENVIRON. MICROBIOL. TABLE 1. Antimicrobial resistance profiles of isolates from turkey-associated sources Resistance profile No. of isolates a Barn b Diagn. Process. Ground Total No resistance Amc, Amp, Chl, Fox, Gen, Kan,, Sul, Tet 1 1 Amc, Amp, Chl, Fox, Kan, Nal,, Sul, Tet, Tio 1 1 Amc, Amp, Chl, Fox, Kan,, Sul, Tet, Tio 1 1 Amc, Amp, Fox, Gen, Kan,, Sul, Tet, Tio 1 1 Amc, Amp, Fox, Kan,, Tet, Tio 1 1 Amc, Amp, Gen, Kan,, Sul, Tet 2 1 Amc, Amp, Gen, Kan,, Sul, Tet, Tio 1 1 Amp, Gen, Kan,, Sul, Tet Amp, Kan,, Sul, Tet 5 5 Amp,, Tet 1 1 Gen Gen,, Sul, Tet Gen, Kan,, Sul, Tet Gen, Gen,, Sul Gen,, Tet 1 1 Gen, Tet 2 2 Kan,, Tet 7 7 Nal, Tet , Sul, Sxt, Tet 1 1, Sul, Tet 1 1 Sul 1 1 Tet Total a Diagn., veterinary diagnostic isolates from the National Veterinary Services Laboratory and North Dakota State Veterinary Laboratory; Process., turkey processing plants; Ground, ground turkey meat. b Barn swabs were collected from various locations in the turkey barn, such as walls, ventilation areas, fans, feed carts, floors, and door handles. Downloaded from When the results of the AST were compared to PFGE clusters, there were some interesting dichotomies found. For instance, within cluster A, there were two larger subclusters (A1 and A2) that differed by a single band. The subclusters were defined as isolates with more than 99% similarity to one another based on XbaI PFGE fingerprint patterns. Subcluster A1 contained 22 isolates with identical PFGE profiles, all of which originated from the upper Midwestern United States, from either a veterinary diagnostic laboratory or one of two turkey processing plants. Seventeen of these isolates were resistant to at least five antimicrobial agents, while only three were pansusceptible. Isolates in subcluster A1 shared their PFGE profiles with the most commonly detected XbaI pattern (JF6X ) reported to the Center for Disease Control and Prevention s PulseNet program (personal communication with PulseNet staff). Because of the similarity of PFGE profiles of a number of the multidrug-resistant isolates to the PulseNet JF6X pattern, further discrimination was sought to determine whether the isolates in subcluster A1 were clonal. In an attempt to further distinguish among isolates in subcluster A1, the isolates were digested with BlnI and subjected to PFGE analysis. Subcluster A was separated into seven different profiles by BlnI PFGE analysis (Fig. 2); however, the majority of isolates (15/22) remained indistinguishable. The results demonstrated the utility of using an additional enzyme to distinguish among apparently clonal S. enterica serovar Heidelberg isolates in clusters of interest, as observed in reports for other enteric pathogens (12). However, because of the additional costs associated with the multiple-enzyme PFGE approach, this study utilized the two-enzyme approach for the isolates in subcluster A1, which were of special interest due to their similarity to isolates from human infections and increased levels of antimicrobial resistance. Interestingly, the isolates in subclus- on January 0, 2019 by guest FIG. 1. XbaI PFGE, antimicrobial susceptibility, and plasmid profiles of S. enterica serovar Heidelberg isolates from turkey-related sources, including ceca, waterers, litter, barn surfaces, veterinary diagnostic areas, processing plants, and retail ground turkey. PFGE clusters A to H are defined by members sharing 90% similarity. Cluster A has two large subclusters (A1 and A2) in which the isolates share more than 99% similarity. Resistant isolates are indicated by black boxes, intermediate-susceptible isolates by dark gray boxes, and susceptible isolates by light gray boxes. The years of isolation and states of origin are indicated in the Year and State columns, with the states listed by standard postal abbreviations, except for, which represents isolates originating from processing plants in the upper Midwestern portion of the United States. The plasmid profile results presented are the sizes of plasmids (kb) detected following isolation with the Wizard Plus SV Minipreps DNA purification system. In a number of isolates, no plasmids were detected following isolation; these plasmids are designated. Representative isolates were selected for additional analysis to identify larger plasmids (Fig. ).

6 VOL. 74, 2008 SEROVAR HEIDELBERG FROM TURKEY-ASSOCIATED SOURCES 504 Dice (Opt:1.56%) (Tol 1.0%-1.0%) (H>0.0% S>0.0%) [0.0%-100.0%] PFGE_Bln1 PFGE_Bln1 Antibiogram (sir) ter A1 originated over a 9-year period, and even following the BlnI restriction analysis, two veterinary diagnostic isolates (712 and 720) remained clustered with 1 isolates from two different processing plants (Fig. 2). The results suggest that S. enterica serovar Heidelberg clones may have persisted in the turkey population in the Midwestern portion of the United States for multiple years. In subcluster A2, there were 81 isolates that originated from turkeys and sites on a West Virginia farm. The majority of the isolates (64/81) in cluster A2 did not demonstrate resistance to any of the antimicrobial agents tested (Fig. 1). This genotypic and phenotypic similarity likely indicates that there is dissemination of a particular genotype of S. enterica serovar Heidelberg throughout the production environment and among the birds present on the farm (20). The low level of resistance observed in these isolates was likely due to the fact that no antimicrobials were administered to the birds either by feed or by water during their production (19). There were, however, some isolates collected from the facility and birds that were resistant to multiple antimicrobials, with the isolate displaying the greatest level of resistance being resistant to four antimicrobial agents (, Sul, Sxt, and Tet). The large number of highly susceptible isolates from the single farm reduced the overall level of resistance detected among the turkey-associated isolates, which likely provides insight into why the overall level of resistance detected among the isolates in this study is lower than that detected in the NARMS program for turkeyassociated samples (9). The presence of subclusters A1 and A2, with very similar PFGE profiles yet significantly different susceptibility patterns, suggests differences in the carriage of resistance plasmids by isolates in these two subclusters. The initial plasmid isolation results obtained using a commercial kit did not identify any large plasmids ( 100 kb), which are often associated with Chl Tet Axo Tio Fox Amp Amc Ami Gen Kan Sul Sxt Cip Nal Isolate Source Year State Processing plant A (post-chill) Processing plant A (post-chill) Processing plant A (post-chill) Processing plant A (post-chill) FIG. 2. BlnI PFGE analysis of S. enterica serovar Heidelberg isolates from subcluster A1. The results of AST and isolate demographic information are presented to assist in isolate comparison. Resistant isolates are indicated by black boxes, intermediate-susceptible isolates by dark gray boxes, and susceptible isolates by light gray boxes. The states are listed by standard postal abbreviations, with the exception of, which represents isolates originating from processing plants in the upper Midwestern portion of the United States. multidrug resistance (Fig. 1). Hence, a more intensive plasmid isolation protocol (24) was used to screen isolates representing each susceptibility and PFGE profile to determine the prevalence of larger plasmid profiles. A number of Salmonella enterica serovar Heidelberg isolates harbored high-molecularweight plasmids ( 100 kb) (Fig. ). Seventeen isolates screened were resistant to five or more antimicrobial agents, and of these, nine (5%) contained plasmids greater than 100 kb. Among the isolates resistant to fewer than five antimicrobials, 8% (19/50) contained plasmids of at least 100 kb. These observations suggest that in some isolates either multidrug resistance is not carried on large plasmids or even with the more robust isolation methods some larger plasmids could not be isolated. Conversely, the presence of large plasmids in large, antimicrobial-susceptible strains may indicate the potential for the presence of large accessory plasmids, such as those associated with virulence, that have been reported to occur in other enteric bacteria (14). The current study did not characterize these large plasmids to confirm their identity. The next two largest clusters detected in the study were clusters C and E, with each containing nine isolates. Both clusters contain isolates from different sampled sources and geographical locations. Cluster C contained cecal isolates from the West Virginia farm, with diagnostic isolates from the NVSL originating from Arkansas and Missouri, slaughter plant isolates from the upper Midwest, and ground turkey meat isolates from Minnesota and Tennessee. Cluster E contained NVSL diagnostic isolates originating from Arkansas, Illinois, North Carolina, Missouri, and Ohio, a slaughter plant isolate from the upper Midwest, and ground turkey meat isolates from Maryland. The isolates in each cluster demonstrated variable antimicrobial susceptibility profiles, with six different AST profiles in each of the clusters. All of the isolates in cluster E were resistant to at least one antimicrobial agent, whereas five iso- Downloaded from on January 0, 2019 by guest

7 5044 KALDHONE ET AL. APPL. ENVIRON. MICROBIOL. Antibiogram (sir) Chl Tet Axo Tio Fox Amp Amc Ami Gen Kan Sul Sxt Cip Nal Isolate Plasmids (kb) *120, 4, 6, 4 *120, 4, 6, *120, 4, 6, 4 *120, 4, 10 *120, 6, 4 * *4, *12, 8,, 2 * 4, 4, *4, 4, *120, 4, 8,, 2 *>150, 4, 4, *4, 6, 4,, 2 *120, 4, 6, 4 * * 4, 4, *>150 *>150, 5 *66, 52, *6, *>150, 4, 4, *>150, 4, 4, *10, 4 *5 *>150, 5 *>150, 52, *6, *>150, 52, *6, *6, *>150, *>150, 44, 8, 5 *52, *10, *>150, 52, * * *74, *120, 8, 6, * * *52, * *6, *90, 40 *100 * * * * *100, 44, 5 * *6 * * * *110, lates in cluster C were susceptible or intermediately susceptible to each of the antimicrobial agents. Conjugation studies were done to determine the extent of resistance carried on transmissible plasmids. Plasmids carrying -lactam resistance genes were readily transferred under the selective pressure of Amp, along with genes encoding resistance to aminoglycosides ( and Kan) and Sul (Fig. 4A). Isolate, from subcluster A1, was shown to be able to conjugally transfer resistance to a recipient strain, which initially demonstrated limited resistance (Tet and Nal). Therefore, it appears likely that resistance, for at least some of the isolates in subcluster A1, is carried on a large, mobile resistance plasmid (approximately 120 kb) (Fig. 4B). Additionally, isolate 16 in cluster E was also able to transfer multipleantimicrobial resistance genes to recipient isolate 819, which is in the same PFGE cluster. One isolate (710) that did not fall into specific PFGE clusters was also able to transfer multipleantimicrobial resistance genes to a recipient via conjugation, indicating that multidrug resistance plasmids were able to be transferred among closely related as well as more-distantly related strains of S. enterica serovar Heidelberg. Similar findings indicating that conjugative plasmids can play a role in resistance transfer among S. enterica serovar Heidelberg strains have been reported (1). Zhao et al. (29) found that the bla CMY and aada genes, encoding resistance to -lactams and, respectively, were located on conjugative plasmids in S. enterica serovar Newport. Our results correspond with these findings, with all of the donors and transconjugants containing bla CMY and aada. In addition, each the strains contained bla TEM, apha1, stra, strb, sul1, sul2, teta, and tetb, which are associated with resistance to certain -lactam agents, aminoglycosides, sulfonamides, and tetracyclines (Fig. 4A). In our study, there were some small differences in the resistance phenotypes that were transferred among strains; however, we did not detect differences in the resistance genes that were transferred from each of the donor strains to the recipients. This finding likely indicates that either additional genes were not detected or differences in gene expression occurred in the different genetic backgrounds. Plasmid-sequencing studies will need to be carried out to determine the exact differences among the plasmids and detect whether the isolates share a common genetic backbone with additional resistance genes incorporated through horizontal gene transfer. All of the isolates resistant to Tio contained plasmids greater than 100 kb (Fig. ), consistent with the reported sizes of bla CMY -positive plasmids from various sources (4, 7, 25, 0). These larger plasmids likely carried the bla CMY gene, which was observed in the conjugation experiments and through sequence analysis of plasmids from other Salmonella serovars (25). The resistance genes present on the plasmids were similar Downloaded from on January 0, 2019 by guest 61 * *>150, 4, 8, *4, 4, *100, * * *140 *74, 4, 4, *100, *4 FIG.. Comparison of antimicrobial susceptibility and high-molecular-weight plasmid profiles of S. enterica serovar Heidelberg. Plasmid profile results from isolates representing at least one member of each different susceptibility and PFGE profile group are illustrated using procedures described by Wang and Rossman (24). Isolates in which no plasmids were detected are designated. Resistant isolates are indicated by black boxes, intermediate-susceptible isolates by dark gray boxes, and susceptible isolates by light gray boxes.

8 VOL. 74, 2008 SEROVAR HEIDELBERG FROM TURKEY-ASSOCIATED SOURCES 5045 A Susceptibility Profiles Antimicrobial Resistance Genes Isolate x 819 x x 819 Amc Amp Fox Tio Axo Chl Gen Kan Nal Sul Tet Sxt aada1 aada2 stra strb accc aadb aphai-iab blacmy blatem cat1 cat2 flor sul1 sul2 sul dhfr1 teta tetb tetc Integron B kb MWS x819 x x819 Downloaded from FIG. 4. Results of the conjugation experiments, including AST (A), resistance gene detection, and plasmid analysis results (B) for the donors (16,, and 710), recipient (819), and transconjugants (16 819, 819, and ) from the conjugation experiments. In panel A, resistant isolates are indicated by dark gray boxes, intermediate-susceptible isolates by light gray boxes, and susceptible isolates by white boxes. For antimicrobial resistance gene detection, a dark gray box indicates the presence of the corresponding resistance gene. The following genes were screened (the drugs for which they encode resistance are indicated in parentheses): aada1, aada2, stra, and strb (); accc and aadb (Gen); aphai-iab (Kan); bla CMY (Amp, Amc, Tio, and Axo); bla TEM (Amp); cat1, cat2, and flor (Chl); sul1, sul2, and sul (Sul); dhfr1 (Sxt); and teta, tetb, and tetc (Tet). The Integron column indicates the presence or absence of a class 1 integron in the strains. Panel B illustrates the plasmid profiles of donor, recipient, and transconjugant strains. Lane MWS contains a BAC-Tracker supercoiled DNA ladder (Epicentre, Madison, WI) size standard used to size plasmids in the conjugation experiments (kb). to those detected in sequenced resistance plasmids from other Salmonella serovars (25). It should be noted that, at the time of submission of this article, a compete sequence for S. enterica serovar Heidelberg resistance plasmid has yet to be reported. Overall, the study indicates that antimicrobial resistance is a problem among S. enterica serovar Heidelberg isolates from turkey-associated sources due to the presence of transferable multidrug resistance plasmids. The PFGE similarity between human and turkey-associated multidrug-resistant strains of S. enterica serovar Heidelberg is cause for concern. The presence of multiple-antimicrobial resistance genes on a single transmissible plasmid raises concerns, in part because an agent that is used routinely on the farm for growth promotion, disease prophylaxis, or disinfection may select for resistance to a critically important antimicrobial agent used in human medicine, which in turn would lead to increased difficulty in treating severe salmonellosis. Thus, multidrug-resistant S. enterica serovar Heidelberg is a potential emerging health concern that should be closely monitored to minimize future health impacts. In summary, our findings show that antimicrobial resistance and PFGE banding patterns in S. enterica serovar Heidelberg from turkey-associated sources can vary significantly by source and region; however, certain groups of isolates show commonality with isolates causing human infections. In addition, our data indicated that large, transferable, bla CMY -positive plasmids ( 100 kb) mediate the Tio and multidrug resistance in S. enterica serovar Heidelberg strains, which concurs with previous findings from other enteric bacterial species from the United States and other countries. In order to understand the molecular events leading to plasmid-mediated multidrug resistance in S. enterica serovar Heidelberg, further studies for determination of whole-plasmid DNA sequences are necessary. This will reveal the range of plasmid backbones extant in salmonellae and the diversity of advantageous traits underlying on January 0, 2019 by guest

9 5046 KALDHONE ET AL. APPL. ENVIRON. MICROBIOL. multidrug resistance. This genetic information will aid in the development of public health interventions designed to limit the spread of antimicrobial resistance in food animal production and processing environments. ACKNOWLEDGMENTS We appreciate the financial support for this project that was provided by the Marshfield Clinic Research Foundation (MCRF) and the National Institutes for Health-funded Arkansas Biomedical Research Infrastructure Network. We also thank Mary Stemper of the MCRF and David G. White of the FDA Center for Veterinary Medicine for their assistance with the project and Nehal Patel of the CDC PulseNet Program for providing us with information on the predominant PFGE profiles associated with human salmonellosis. The use of trade names is for identification purposes only and does not imply endorsement by the U.S. Food and Drug Administration or the U.S. Department of Health and Human Services. The views presented in this article do not necessarily reflect those of the FDA. REFERENCES 1. Aarestrup, F. M., H. Hasman, I. Olsen, and G. Sorensen International spread of bla CMY-2 -mediated cephalosporin resistance in a multiresistant Salmonella enterica serovar Heidelberg isolate stemming from the importation of a boar by Denmark from Canada. Antimicrob. Agents Chemother. 48: Reference deleted.. Buzby, J. C., and H. A. Farah Chicken consumption continues longrun rise. Amber Waves 4:5. 4. Carattoli, A., F. Tosini, W. P. Giles, M. E. Rupp, S. H. Hinrichs, F. J. Angulo, T. J. Barrett, and P. D. Fey.. Characterization of plasmids carrying CMY-2 from expanded-spectrum cephalosporin-resistant Salmonella strains isolated in the United States between 1996 and Antimicrob. Agents Chemother. 46: CDC Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food 10 states, United States, MMWR Morb. Mortal. Wkly. Rep. 55: CDC Salmonella surveillance: annual summary, Centers for Disease Control and Prevention, Atlanta,. 7. Chiu, C. H., P. Tang, C. Chu, S. Hu, Q. Bao, J. Yu, Y. Y. Chou, H. S. Wang, and Y. S. Lee The genome sequence of Salmonella enterica serovar Choleraesuis, a highly invasive and resistant zoonotic pathogen. Nucleic Acids Res. : Clewell, D. B., F. Y. An, B. A. White, and C. Gawron-Burke eptococcus faecalis sex pheromone (cam7) also produced by Staphylococcus aureus and identification of a conjugative transposon (Tn918). J. Bacteriol. 162: FDA National Antimicrobial Resistance Monitoring System Enteric Bacteria (NARMS): 200 executive report. Department of Health and Human Services, U.S. Food and Drug Administration, Washington, DC. 10. Ferris, K. E., J. M. Timm, A. M. Aalsburg, and M. Munoz.. Salmonella serotypes from animals and related sources reported during July 2001 June. Proc. Annu. Meet. U.S. Anim. Health Assoc. 106: Foley, S. L., A. M. Lynne, and R. Nayak Salmonella challenges: prevalence in swine and poultry and potential pathogenicity of such isolates. J. Anim. Sci. 86:E149 E Gupta, A., S. B. Hunter, S. A. Bidol, S. Dietrich, J. Kincaid, E. Salehi, L. Nicholson, C. A. Genese, S. Todd-Weinstein, L. Marengo, A. C. Kimura, and J. T. Brooks Escherichia coli O157 cluster evaluation. Emerg. Infect. Dis. 10: Hennessy, T. W., L. H. Cheng, H. Kassenborg, S. D. Ahuja, J. Mohle- Boetani, R. Marcus, B. Shiferaw, and F. J. Angulo Egg consumption is the principal risk factor for sporadic Salmonella serotype Heidelberg infections: a case-control study in FoodNet sites. Clin. Infect. Dis. 8(Suppl. ):S27 S Johnson, T. J., S. J. Johnson, and L. K. Nolan Complete DNA sequence of a ColBM plasmid from avian pathogenic Escherichia coli suggests that it evolved from closely related ColV virulence plasmids. J. Bacteriol. 188: Kennedy, M., R. Villar, D. J. Vugia, T. Rabatsky-Ehr, M. M. Farley, M. Pass, K. Smith, P. Smith, P. R. Cieslak, B. Imhoff, and P. M. Griffin Hospitalizations and deaths due to Salmonella infections, FoodNet, Clin. Infect. Dis. 8(Suppl. ):S142 S Logue, C. M., J. S. Sherwood, P. A. Olah, L. M. Elijah, and M. R. Dockter The incidence of antimicrobial-resistant Salmonella spp on freshly processed poultry from US Midwestern processing plants. J. Appl. Microbiol. 94: Lynne, A. M., B. S. Rhodes-Clark, K. Bliven, S. Zhao, and S. L. Foley Antimicrobial resistance genes associated with Salmonella enterica serovar Newport isolates from food animals. Antimicrob. Agents Chemother. 52: Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe Food-related illness and death in the United States. Emerg. Infect. Dis. 5: Nayak, R., P. B. Kenney, J. Keswani, and C. Ritz Isolation and characterisation of Salmonella in a turkey production facility. Br. Poult. Sci. 44: Nayak, R., and T. Stewart-King Molecular epidemiological analysis and microbial source tracking of Salmonella enterica serovars in a preharvest turkey production environment. Foodborne Pathog. Dis. 5: NCCLS.. Performance standards for antimicrobial susceptibility testing, 12th informational supplement. M100-S12, M100-S12. National Committee for Clinical Laboratory Standards, Wayne, PA. 22. Ribot, E. M., M. A. Fair, R. Gautom, D. N. Cameron, S. B. Hunter, B. Swaminathan, and T. J. Barrett Standardization of pulsed-field gel electrophoresis protocols for the subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne Pathog. Dis. : U.S. Department of Agriculture Progress report on Salmonella testing of raw meat and poultry products, Food Safety and Inspection Service, U.S. Department of Agriculture, Washington, DC. 24. Wang, Z., and T. G. Rossman Large-scale supercoiled plasmid preparation by acidic phenol extraction. BioTechniques 16: Welch, T. J., W. F. Fricke, P. F. McDermott, D. G. White, M. L. Rosso, D. A. Rasko, M. K. Mammel, M. Eppinger, M. J. Rosovitz, D. Wagner, L. Rahalison, J. E. Leclerc, J. M. Hinshaw, L. E. Lindler, T. A. Cebula, E. Carniel, and J. Ravel Multiple antimicrobial resistance in plague: an emerging public health risk. PLoS ONE 2:e Winokur, P. L., A. Brueggemann, D. L. DeSalvo, L. Hoffmann, M. D. Apley, E. K. Uhlenhopp, M. A. Pfaller, and G. V. Doern.. Animal and human multidrug-resistant, cephalosporin-resistant Salmonella isolates expressing a plasmid-mediated CMY-2 AmpC -lactamase. Antimicrob. Agents Chemother. 44: Yuan, M., H. Aucken, L. M. Hall, T. L. Pitt, and D. M. Livermore Epidemiological typing of klebsiellae with extended-spectrum beta-lactamases from European intensive care units. J. Antimicrob. Chemother. 41: Zhao, S., P. F. McDermott, S. Friedman, J. Abbott, S. Ayers, A. Glenn, E. Hall-Robinson, S. K. Hubert, H. Harbottle, R. D. Walker, T. M. Chiller, and D. G. White Antimicrobial resistance and genetic relatedness among Salmonella from retail foods of animal origin: NARMS retail meat surveillance. Foodborne Pathog. Dis. : Zhao, S., S. Qaiyumi, S. Friedman, R. Singh, S. L. Foley, D. G. White, P. F. McDermott, T. Donkar, C. Bolin, S. Munro, E. J. Baron, and R. D. Walker Characterization of Salmonella enterica serotype Newport isolated from humans and food animals. J. Clin. Microbiol. 41: Zhao, S., D. G. White, P. F. McDermott, S. Friedman, L. English, S. Ayers, J. Meng, J. J. Maurer, R. Holland, and R. D. Walker Identification and expression of cephamycinase bla CMY genes in Escherichia coli and Salmonella isolates from food animals and ground meat. Antimicrob. Agents Chemother. 45: Downloaded from on January 0, 2019 by guest