Proteome analysis of Campylobacter jejuni poultry strain 2704 survival during 45 min exposure to peracetic acid.

Peracetic acid (PAA) applied to whole poultry carcasses can reduce the number of Campylobacter, a leading cause of human gastroenteritis. However, previous modelling experiments indicated that Campylobacter survived in greater numbers when pre-treated with a thermal stress equivalent to poultry processing scalding prior to chilling with PAA than when subject to chilling with PAA only. To better understand how Campylobacter responds to PAA, proteomes of C. jejuni poultry strain 2704 were measured after exposure to PAA (60 ppm, pH 4.0) for 45 min under laboratory ambient conditions (approximately 23 °C) to establish a foundational map of survival mechanism before combining with other stresses. Analysis of 580 quantified proteins did not indicate a triggered "peroxide shock" response, nor were common heat shock responses detected. Thioredoxin, iron homeostatic, peroxiredoxins and cytochrome c peroxidases became more abundant suggesting that PAA disturbed cytoplasmic redox homeostasis resulting in antioxidant activation and increased prioritisation of iron homeostasis. The PAA treatment led to responses that included an increased priority for oxidative phosphorylation and a simultaneous decrease in central metabolism associated protein abundances. Lon protease was induced suggesting it has a role in maintaining homeostasis during non-thermal stress. Proteins in flagella and chemotaxis became more abundant though whether PAA has a chemorepellent effect requires further investigation. Overall, the proteome data suggests there was a rapid cellular response to applied PAA stress in the first 15 min with the adaptation to the stress completing between 30 and 45 min. The findings will help guide PAA implementation in commercial poultry processing in terms of processing location and length of application.

Antimicrobial resistance and genomic relationships of Salmonella enterica from Australian cattle.

The aim of this study was to evaluate phenotypic and genotypic AMR characteristics of Salmonella enterica isolates from Australian cattle collected through a structured national survey utilizing 1001 faecal samples collected from healthy cattle at slaughter. A total of 184 Salmonella isolates were subsequently derived and subjected to microbroth dilution to 16 drugs from 11 classes with interpretation of minimum inhibitory concentrations (MICs) using epidemiological cut off (ECOFF) values to distinguish between wild-type and non-wild-type populations. Most isolates were susceptible (wild type) to all antimicrobials tested, with no resistance (non-wild type) detected for colistin, nalidixic acid, meropenem, gentamicin, florfenicol or chloramphenicol. Low rates of resistance were detected for ampicillin (2.2%), cefoxitin (2.2%), ceftiofur (2.2%), ceftriaxone (2.2%), ciprofloxacin (0.5%), streptomycin (3.3%) and tetracycline (0.5%). Isolates resistant to ceftriaxone (a critically important antimicrobial, CIA) carried the extended spectrum cephalosporin gene blaCMY-2 while no known mutation in the QRDR region or qnrS genes were detected for the CIA ciprofloxacin-resistant isolate. Thirty-six serovars were detected among the 184 Salmonella isolates using whole genome sequencing, dominated by Typhimurium (n = 36), Saintpaul (n = 22) and Anatum (n = 16). Genomic analysis clustered the cattle isolates based on serovar, with the majority of serovars containing a single sequence type (ST). Further analysis of the bovine Typhimurium isolates (ST19) and comparison with publicly available data from human Typhimurium isolates from Australia, revealed the majority of cattle isolates were unrelated to human isolates. In conclusion, this study demonstrates the continued low prevalence of AMR among Salmonella within the beef, dairy and veal industries in Australia. Salmonella Typhimurium from cattle were genetically distinct from isolates sourced from human infections. Further investigations are warranted to expand on the potential clinical and public health relevance of these findings to inform risk-management of this key pathogen.

Phenotypic and Genotypic Assessment of Antimicrobial Resistance in Escherichia coli from Australian Cattle Populations at Slaughter.

ABSTRACT: Australia relies on periodic antimicrobial resistance (AMR) surveys to determine trends and changes in AMR in animal production systems. This study is a follow-up to a survey of Escherichia coli from healthy cattle at slaughter conducted in 2013, which provided baseline data on AMR prevalence across cattle groups and production practices. In this study, 591 beef cattle, 194 dairy cattle, and 216 veal calf fecal samples were collected from 25 beef and veal processing establishments in Australia, representing approximately 77% of total export volume. A total of 969 matrix-assisted laser desorption-ionization results confirmed commensal E. coli isolates from 574 beef cattle, 186 dairy cattle, and 209 veal calves were recovered, and antimicrobial susceptibility testing was carried out by microbroth dilution to 16 drugs from 10 classes interpreted against epidemiological cutoff breakpoints. Overall, a high proportion of E. coli isolates (83.8%) were wild type for all antimicrobials assessed. In addition, isolates that were non-wild type (NWT) for three or more classes of antimicrobial did not exceed 4% for any of the cattle groups. The prevalence of E. coli that were NWT for antimicrobials that are critical or of high importance to human health was very low, with 1.4% of all isolates tested determined to be NWT for fluoroquinolones, third-generation cephalosporins, or polymyxins. Genomic analysis of NWT isolates identified one beef cattle isolate (ST-10) harboring blaCMY-2 and a dairy isolate (ST-58) and two veal calf isolates (ST-58 and ST-394) that had qnrS1, which confer resistance to extended-spectrum cephalosporins and fluoroquinolones, respectively. The low levels of AMR reported in this study confirm previous Australian studies, which indicated that there is minimal evidence that specific production practices lead to widespread disproportionate development of NWT isolates.

Applicability of Enterobacteriaceae and coliforms tests as indicators for Cronobacter in milk powder factory environments.

The emergence of Cronobacter as an important potential pathogen for newborn children and its occurrence in powdered infant formulae has generated a need to develop new management practices for this food group. This includes reduction of the prevalence of Cronobacter in manufacturing environments which can be a source of Cronobacter. This study was performed to assess the suitability of qualitative and quantitative Enterobacteriaceae and coliforms indicator tests for the presence and prevalence of Cronobacter. Environmental swabs (205) from five milk powder factories were examined. The qualitative indicator tests had good sensitivity but they lacked specificity for reliable routine use. Logistic regression analyses revealed a significant relationship between the quantitative indicator tests and Cronobacter prevalence, where the Enterobacteriaceae count was a slightly stronger predictor for Cronobacter than the coliforms count. The optimum test sensitivity (81%) and specificity (66%) was obtained when the indicator count thresholds were set at ≥1 cfu/cm2. However, since 11% of samples were Cronobacter positive when counts of Enterobacteriaceae and coliforms were less than 1 cfu/cm2, specific testing for Cronobacter is advised in addition to Enterobacteriaceae testing to minimise risk of transfer of Cronobacter from the factory environment into powdered infant formulae products.

Impact of Poultry Processing Operating Parameters on Bacterial Transmission and Persistence on Chicken Carcasses and Their Shelf Life.

It is important for the poultry industry to maximize product safety and quality by understanding the connection between bacterial diversity on chicken carcasses throughout poultry processing to the end of shelf life and the impact of the local processing environment. Enumeration of total aerobic bacteria, Campylobacter and Pseudomonas, and 16S rRNA gene amplicon sequencing were used to evaluate the processing line by collecting 10 carcasses from five processing steps: prescald, postplucker, pre- and post-immersion chill, and post-air chill. The diversity throughout a 12-day shelf life was also determined by examining 30 packaged carcasses. To identify the sources of possible contamination, scald water tank, immersion chilling water tank, air samples, and wall surfaces in the air-chill room were analyzed. Despite bacterial reductions on carcasses (>5 log10 CFU/ml) throughout the process, each step altered the bacterial diversity. Campylobacter was a minor but persistent component in the bacterial community on carcasses. The combination of scalding, defeathering, and plucking distributed thermophilic spore-forming Anoxybacillus to carcasses, which remained at a high abundance on carcasses throughout subsequent processes. Pseudomonas was not isolated from carcasses after air chilling but was abundant on the wall of the air-chill room and became the predominant taxon at the end of shelf life, suggesting possible contamination through air movement. The results suggest that attention is needed at each processing step, regardless of bacterial reductions on carcasses. Changing scalding water regularly, maintaining good hygiene practices during processing, and thorough disinfection at the end of each processing day are important to minimize bacterial transmission.IMPORTANCE Culture-based and culture-independent approaches were utilized to reveal bacterial community changes on chicken carcasses at different processing steps and potential routes from the local processing environment. Current commercial processing effectively reduced bacterial loads on carcasses. Poultry processes have similar processes across facilities, but various processing arrangements and operating parameters could impact the bacterial transmission and persistence on carcasses differently. This study showed the use of a single tunnel incorporating scalding, defeathering and plucking may undesirably distribute the thermoduric bacteria, e.g., Campylobacter and Anoxybacillus, between the local environment and carcasses, whereas this does not occur when these steps are separated. The length of immersion and air chilling also impacted bacterial diversity on carcasses. Air chilling can transfer Pseudomonas from wall surfaces onto carcasses; this may subsequently influence chicken product shelf life. This study helps poultry processors understand the impact of current commercial processing and improve the chicken product quality and safety.

Changes of the bacterial community diversity on chicken carcasses through an Australian poultry processing line.

Understanding the bacterial community profile through poultry processing could help the industry to produce better poultry products. In this study, 10 chicken carcasses were randomly sampled from before and after scalding, before and after immersion chilling, and after air chilling each through a modern commercial processing line, along with the contents of 10 caeca. The sampled processing line effectively reduced the bacterial counts by > 4.6 Log10 CFU/ml for each of Total Viable Counts, Escherichia coli and Campylobacter. However, the metagenomics results suggested that Lactobacillus, Staphylococcus and unclassified Lachnospiraceae persisted at all sampling stages. Pseudomonas, Paeniglutamicibacter, Chryseobacterium and Pseudarthrobacter comprised 47.2% in the bacterial community on samples after air chilling compared to 0.3% on samples after immersion chilling, whereas TVCs were the same. Overall, the current interventions of the investigated poultry processing line were unable to eliminate persistence of certain foodborne pathogens, despite a significant reduction of the overall bacterial counts. Chilling is an important controlling point in contamination/cross-contamination, particularly extended air chilling. Lastly, the large presence of Pseudomonas on chickens after air chilling may lead to downstream spoilage related issues, which needs more investigation to explore quantitatively the effect on the shelf life of poultry products.

Antimicrobial resistance status of Enterococcus from Australian cattle populations at slaughter.

Antimicrobial agents are used in cattle production systems for the prevention and control of bacterial associated diseases. A consequence of their use is the potential development of antimicrobial resistance (AMR). Enterococcus faecium and Enterococcus faecalis that are resistant to antimicrobials are of increased concern to public health officials throughout the world as they may compromise the ability of various treatment regimens to control disease and infection in human medicine. Australia is a major exporter of beef; however it does not have an ongoing surveillance system for AMR in cattle or foods derived from these animals. This study examined 910 beef cattle, 290 dairy cattle and 300 veal calf faecal samples collected at slaughter for the presence of enterococci. Enterococcus were isolated from 805 (88.5%) beef cattle faeces, 244 (84.1%) dairy cattle faeces and 247 (82.3%) veal calf faeces with a total of 800 enterococci subsequently selected for AMR testing. The results of AMR testing identified high levels of resistance to antimicrobials that are not critically or highly important to human medicine with resistance to flavomycin (80.2%) and lincomycin (85.4-94.2%) routinely observed. Conversely, resistance to antibiotics considered critically or highly important to human medicine such as tigecycline, daptomycin, vancomycin and linezolid was not present in this study. There is minimal evidence that Australian cattle production practices are responsible for disproportionate contributions to AMR development and in general resistance to antimicrobials of critical and high importance in human medicine was low regardless of the isolate source. The low level of antimicrobial resistance in Enterococcus from Australian cattle is likely to result from comprehensive controls around the use of antimicrobials in food-production animals in Australia. Nevertheless, continued monitoring of the effects of all antimicrobial use is required to support Australia's reputation as a supplier of safe and healthy food.

National Survey of Shiga Toxin-Producing Escherichia coli Serotypes O26, O45, O103, O111, O121, O145, and O157 in Australian Beef Cattle Feces.

Escherichia coli O157 and six non-O157 Shiga toxin-producing E. coli (STEC) serotypes (O26, O45, O103, O111, O121, and O145, colloquially referred to as the "big 6") have been classified as adulterants of raw nonintact beef products in the United States. While beef cattle are a known reservoir for the prototype STEC serotype, E. coli O157, less is known about the dissemination of non-O157 STEC serotypes in Australian cattle. In the present study, 1,500 fecal samples were collected at slaughter from adult (n =628) and young (n =286) beef cattle, adult (n =128) and young (n =143) dairy cattle, and veal calves (n = 315) across 31 Australian export-registered processing establishments. Fecal samples were enriched and tested for E. coli O157 and the big 6 STEC serotypes using BAX System PCR and immunomagnetic separation methods. Pathogenic STEC (pSTEC; isolates that possess stx, eae, and an O antigen marker for O157 or a big 6 serotype) were isolated from 115 samples (7.7%), of which 100 (6.7%) contained E. coli O157 and 19 (1.3%) contained a big 6 serotype. Four of the 115 samples contained multiple pSTEC serotypes. Among samples confirmed for big 6 pSTEC, 15 (1%) contained E. coli O26 and 4 (0.3%) contained E. coli O111. pSTEC of serotypes O45, O103, O121, and O145 were not isolated from any sample, even though genes indicative of E. coli belonging to these serotypes were detected by PCR. Analysis of animal classes revealed a higher pSTEC prevalence in younger animals, including veal (12.7%), young beef (9.8%), and young dairy (7.0%), than in adult animals, including adult beef (5.1%) and adult dairy (3.9%). This study is the largest of its kind undertaken in Australia. In contrast to E. coli O157 and consistent with previous findings, this study reports a relatively low prevalence of big 6 pSTEC serotypes in Australian cattle populations.

Salmonella Typhimurium and Salmonella Sofia: Growth in and Persistence on Eggs under Production and Retail Conditions.

Salmonellosis in Australia has been linked to eggs and egg products with specific serotypes associated with outbreaks. We compared attachment to and survival on egg shells and growth in eggs of two Salmonella serotypes, an egg outbreak associated Salmonella Typhimurium and a non-egg-associated Salmonella enterica ssp. II 1,4,12,27:b:[e,n,x] (S. Sofia). Experiments were conducted at combinations of 4, 15, 22, 37 and 42 °C. No significant differences occurred between the serotypes in maximum growth rates, which were significantly greater (P < 0.001) in egg yolk (0.427 log10 CFU/mL/h) compared to whole egg (0.312 log10 CFU/mL/h) and egg white (0.029 log10 CFU/mL/h). Attachment to egg shells varied by time (1 or 20 min) and temperature (4, 22 and 42 °C), with S. Typhimurium isolates attaching at higher levels (P < 0.05) than S. Sofia after 1 min at 4 °C and S. Typhimurium ATCC 14028 attaching at higher (P < 0.05) levels at 22 °C. Survival on egg shells was not significantly different across isolates. Salmonella serotypes behaved similarly regarding growth in egg contents, attachment to egg shells and survival on eggs, indicating that other factors more likely contributed to reasons for S. Typhimurium being implicated in multiple egg-associated outbreaks.

Prevalence and Antimicrobial Resistance of Salmonella and Escherichia coli from Australian Cattle Populations at Slaughter.

Antimicrobial agents are used in cattle production systems for the prevention and control of bacteria associated with diseases. Australia is the world's third largest exporter of beef; however, this country does not have an ongoing surveillance system for antimicrobial resistance (AMR) in cattle or in foods derived from these animals. In this study, 910 beef cattle, 290 dairy cattle, and 300 veal calf fecal samples collected at slaughter were examined for the presence of Escherichia coli and Salmonella, and the phenotypic AMR of 800 E. coli and 217 Salmonella isolates was determined. E. coli was readily isolated from all types of samples (92.3% of total samples), whereas Salmonella was recovered from only 14.4% of samples and was more likely to be isolated from dairy cattle samples than from beef cattle or veal calf samples. The results of AMR testing corroborate previous Australian animal and retail food surveys, which have indicated a low level of AMR. Multidrug resistance in Salmonella isolates from beef cattle was detected infrequently; however, the resistance was to antimicrobials of low importance in human medicine. Although some differences in AMR between isolates from the different types of animals were observed, there is minimal evidence that specific production practices are responsible for disproportionate contributions to AMR development. In general, resistance to antimicrobials of critical and high importance in human medicine was low regardless of the isolate source. The low level of AMR in bacteria from Australian cattle is likely a result of strict regulation of antimicrobials in food animals in Australia and animal management systems that do not favor bacterial disease.

Mapping the carriage of flaA-restriction fragment length polymorphism Campylobacter genotypes on poultry carcasses through the processing chain and comparison to clinical isolates.

Poultry are considered a major source for campylobacteriosis in humans. A total of 1866 Campylobacter spp. isolates collected through the poultry processing chain were typed using flaA-restriction fragment length polymorphism to measure the impact of processing on the genotypes present. Temporally related human clinical isolates (n = 497) were also typed. Isolates were obtained from whole chicken carcass rinses of chickens collected before scalding, after scalding, before immersion chilling, after immersion chilling and after packaging as well as from individual caecal samples. A total of 32 genotypes comprising at least four isolates each were recognised. Simpson's Index of Diversity (D) was calculated for each sampling site within each flock, for each flock as a whole and for the clinical isolates. From caecal collection to after packaging samples the D value did not change in two flocks, decreased in one flock and increased in the fourth flock. Dominant genotypes occurred in each flock but their constitutive percentages changed through processing. There were 23 overlapping genotypes between clinical and chicken isolates. The diversity of Campylobacter is flock dependant and may alter through processing. This study confirms that poultry are a source of campylobacteriosis in the Australian population although other sources may contribute.

Quantitative effects of in-line operations on Campylobacter and Escherichia coli through two Australian broiler processing plants.

Campylobacter is an important food borne pathogen, mainly associated with poultry. A lack of through-chain quantitative Campylobacter data has been highlighted within quantitative risk assessments. The aim of this study was to quantitatively and qualitatively measure Campylobacter and Escherichia coli concentration on chicken carcasses through poultry slaughter. Chickens (n=240) were sampled from each of four flocks along the processing chain, before scald, after scald, before chill, after chill, after packaging and from individual caeca. The overall prevalence of Campylobacter after packaging was 83% with a median concentration of 0.8log10CFU/mL. The processing points of scalding and chilling had significant mean reductions of both Campylobacter (1.8 and 2.9log10CFU/carcase) and E. coli (1.3 and 2.5log10CFU/carcase). The concentration of E. coli and Campylobacter was significantly correlated throughout processing indicating that E. coli may be a useful indicator organism for reductions in Campylobacter concentration. The carriage of species varied between flocks, with two flocks dominated by Campylobacter coli and two flocks dominated by Campylobacter jejuni. Current processing practices can lead to significant reductions in the concentration of Campylobacter on carcasses. Further understanding of the variable effect of processing on Campylobacter and the survival of specific genotypes may enable more targeted interventions to reduce the concentration of this poultry associated pathogen.

Mixture effects of benzene, toluene, ethylbenzene, and xylenes (BTEX) on lung carcinoma cells via a hanging drop air exposure system.

A recently developed hanging drop air exposure system for toxicity studies of volatile chemicals was applied to evaluate the cell viability of lung carcinoma A549 cells after 1 and 24 h of exposure to benzene, toluene, ethylbenzene, and xylenes (BTEX) as individual compounds and as mixtures of four or six components. The cellular chemical concentrations causing 50% reduction of cell viability (EC50) were calculated using a mass balance model and came to 17, 12, 11, 9, 4, and 4 mmol/kg cell dry weight for benzene, toluene, ethylbenzene, m-xylene, o-xylene, and p-xylene, respectively, after 1 h of exposure. The EC50 decreased by a factor of 4 after 24 h of exposure. All mixture effects were best described by the mixture toxicity model of concentration addition, which is valid for chemicals with the same mode of action. Good agreement with the model predictions was found for benzene, toluene, ethylbenzene, and m-xylene at four different representative fixed concentration ratios after 1 h of exposure, but lower agreement with mixture prediction was obtained after 24 h of exposure. A recreated car exhaust mixture, which involved the contribution of the more toxic p-xylene and o-xylene, yielded an acceptable, but lower quality, prediction as well.

Hanging drop: an in vitro air toxic exposure model using human lung cells in 2D and 3D structures.

Using benzene as a candidate air toxicant and A549 cells as an in vitro cell model, we have developed and validated a hanging drop (HD) air exposure system that mimics an air liquid interface exposure to the lung for periods of 1h to over 20 days. Dose response curves were highly reproducible for 2D cultures but more variable for 3D cultures. By comparing the HD exposure method with other classically used air exposure systems, we found that the HD exposure method is more sensitive, more reliable and cheaper to run than medium diffusion methods and the CULTEX(®) system. The concentration causing 50% of reduction of cell viability (EC50) for benzene, toluene, p-xylene, m-xylene and o-xylene to A549 cells for 1h exposure in the HD system were similar to previous in vitro static air exposure. Not only cell viability could be assessed but also sub lethal biological endpoints such as DNA damage and interleukin expressions. An advantage of the HD exposure system is that bioavailability and cell concentrations can be derived from published physicochemical properties using a four compartment mass balance model. The modelled cellular effect concentrations EC50cell for 1h exposure were very similar for benzene, toluene and three xylenes and ranged from 5 to 15 mmol/kgdry weight, which corresponds to the intracellular concentration of narcotic chemicals in many aquatic species, confirming the high sensitivity of this exposure method.

Prevalence and concentration of Arcobacter spp. on Australian Beef Carcasses.

The International Commission on Microbiological Specifications for Foods (ICMSF) classified Arcobacter spp. as emerging pathogens in 2002. Arcobacter spp. have been isolated from numerous food products at retail and from animal carcasses and feces at slaughter. A survey was conducted to determine both the prevalence and concentration of Arcobacter spp. on pre-chill beef carcasses. Surface swab samples were collected from 130 beef carcasses at the end of processing, prior to chilling. The concentration of Arcobacter spp. was determined by a most-probable-number per square centimeter (3 by 3) method with a limit of detection of 0.12 CFU/cm(2). Of the 100 carcasses examined from export abattoirs, 20 (20.0%) were contaminated with Arcobacter spp., and 5 of these had quantifiable levels of contamination ranging from 0.12 to 0.31 CFU/cm(2). Of the 30 carcasses examined at a pet food abattoir, 25 (83.3%) were contaminated with Arcobacter spp., and 10 of these had quantifiable levels of contamination ranging from 0.12 to 0.95 CFU/cm(2). Three species of Arcobacter, A. butzleri, A. cryaerophilus, and A. skirowii, were identified by PCR. Each of the species was present in an approximately equal ratio from export abattoirs. This study demonstrates that slaughter practices at export abattoirs are sufficient to maintain both low prevalence and low levels of contamination of beef carcasses with Arcobacter spp.

The effects of transport and lairage on counts of Escherichia coli O157 in the feces and on the hides of individual cattle.

OBJECTIVES: The main objective of this study was to determine the impact of transport and lairage on the isolation rate and the number of Escherichia coli O157 on cattle. MATERIALS: Ninety animals, divided into three groups (A, B, and C) of 30 animals each, were used in this study. Individual animals were tagged, and samples were collected from the hides and feces of each at a feedlot and again after slaughter. The carcass of each animal was also sampled. Samples were also collected from the feedlot pens, the sides and floors of the transport trucks, and abattoir holding pens. The isolation rate and the number of E. coli O157 were estimated using a combination of immunomagnetic separation and the Most Probable Number technique. RESULTS: Cattle hides were more likely to be contaminated with E. coli O157 at the feedlot (31%) than at the abattoir (4%). E. coli O157 was detected in 18% and 12% of cattle feces collected at the feedlot and after slaughter, respectively. E. coli O157 was isolated from truck floors (26%), truck sides (11%), abattoir pen rails (47%), and pen floors (42%). The mean count of E. coli O157 in positive feces was log(10) 1.17 and 2.37 MPN/g at the feedlot and slaughter, respectively. A 3 log(10) increase in the number of E. coli O157 was observed between the feedlot (2.66 MPN/g) and slaughter (5.66 MPN/g) in the feces of one animal in group B. E. coli O157 was isolated from the hide and carcass of this animal. CONCLUSIONS: Transport and lairage did not lead to an increase in the number or isolation rate of E. coli O157 from cattle. APPLICATIONS: Intervention strategies for reducing E. coli O157 contamination of cattle carcasses should target mechanisms that limit the impact of animals shedding a high number throughout production and processing.

The ability of Campylobacter jejuni cells to attach to stainless steel does not change as they become nonculturable.

The ability of Campylobacter jejuni strains to attach to stainless steel as they became nonculturable during storage in distilled water at 4 degrees C for 30 days was investigated. From an initial count of approximately 7 log colony-forming units/mL all strains completely lost culturability by day 20, but the numbers of cells attaching to stainless steel remained constant at approximately 3.5 log cells/cm(2). These findings suggest that viable but nonculturable Campylobacter in a liquid matrix can still attach to stainless steel.