Campylobacter persistence in poultry processing

The Foodborne Illness Reduction Strategy 2018-2021+ released by Australia's Food Regulation Secretariat regards Campylobacter as one of the main foodborne pathogens requiring actions at all points along the food supply chain. The ultimate objective of this thesis was to help poultry processors understand the impact of current processing strategies and develop better hurdle interventions to reduce Campylobacter numbers and prevalence on chicken meat products during slaughter, which may reduce the burden of disease currently associated with this pathogen. Five aims were established for the thesis and included: 1) understanding the impact of current processing interventions on the bacterial numbers and diversity on chicken carcasses throughout commercial processing; 2) mapping changes of the bacterial community on carcasses throughout processing to the end of shelf life and to link the data to the processing environment; 3) developing an in vitro model to test Campylobacter inactivation under mimicked poultry processing conditions; 4) using the in vitro model to evaluate peracetic acid (PAA) as a potential antimicrobial agent in reducing Campylobacter; 5) understanding the potential mechanisms for C. jejuni survival during a 45 min treatment of a sub-lethal of PAA by examining timedependent global proteomic expression. Currently the success of commercial poultry processing lines is measured based on culture-based microbiological analysis, e.g. total bacterial population, count of Campylobacter, etc. This approach is unable to provide information on the entire bacterial community diversity and the importance of Campylobacter in the community in terms of its abundance and relationships with other microbes including those microbes that are not able to be isolated at various processing steps. More recently, the use of chlorine-based disinfectants has been questioned in terms of its effectiveness and potential harm to human health. Looking for alternative processing aids is in demand by the poultry industry and consumers. By completing the aforementioned aims, this thesis will utilise culture-independent techniques along with well-established culture-based methods to fill the gaps outlined, understand indepth the microbial profiles of commercial processing systems and test PAA as a potential antimicrobial aid that could be considered for the replacement of chlorine-based disinfectants. Two different types of poultry processing lines from a single abattoir in Australia, Line A and Line B, were examined. The major differences between Line A and B included: 1) Line A operates with separate steps of scalding, defeathering and plucking, whereas Line B incorporates these three steps together in a confined tunnel, and 2) the time length of immersion chilling (Line A 0.5 h and Line B 1.5 h) and air chilling (Line A 1.5 h and Line B 0.5 h) were different. Both lines produced significant declines (p <0.001) of the bacterial counts on carcasses through scalding and immersion chilling with no decrease through air chilling. However, the confined tunnel in Line B may elevate the temperature inside the tunnel and select and distribute thermoduric/thermophilic bacteria between carcasses, as the results of the culturebased and culture-independent approaches indicated the high level of presence of Campylobacter and Anoxybacillus. Psychrotolerant bacteria dominated bacterial community on carcasses after air chilling from Line A, while immersion chilling in Line B showed a greater impact on the bacterial diversity on carcasses than air chilling. Together with the bacterial diversity through chilling in Line A and B, it was noted that the longer the time that a particular chilling process occurs for (either immersion chilling or air chilling), the greater the change in the bacterial diversity. Furthermore, the culture-independent approach suggested: 1) each sampled processing step acted as a contamination source by distinctively changing the bacterial diversity on carcasses, and 2) Campylobacter was revealed to be a minor but persistent component in the bacterial community on carcasses through processing. Acknowledging the importance of scalding and chilling in reducing bacterial loads on carcasses, a poultry processing in vitro model was developed to mimic the poultry processing system. This model consisted of three laboratory-based food matrices (buffered peptone water (BPW), chicken breast meat and a meat-based broth utilising chicken breast meat to reflect the protein concentration of an immersion chilling tank) and eight processing conditions including scalding alone (54.5°C and 57°C), chilling alone, scalding + chilling, chilling with PAA (80 ppm) and scalding + chilling with PAA. PAA was added to the in vitro model to evaluate its impact on two poultry C. jejuni strains, strain 2674 and strain 2704. The meat-based broth showed a greater Campylobacter inactivation when comparing chilling with PAA to chilling without PAA, indicating the usefulness of PAA during immersion chilling even with the presence of organic matter. The condition of chilling with PAA in BPW and the chicken breast meat demonstrated the greatest Campylobacter inactivation, even greater than the combined condition of scalding + chilling with PAA in the respective food matrices. This indicates a hypothesis that prior scalding may activate a heat shock stress response that also cross-protects against oxidative stress from PAA exposure. To validate this hypothesis, it is important to understand how Campylobacter survives when subjected to a sub-lethal level of PAA over a period of exposure. Therefore, the global proteomic responses of Campylobacter poultry strain 2704 in a sub-lethal PAA (60 ppm) treatment over a 45 min time course was determined by using the sequential window acquisition of all theoretical fragment-ion spectra-mass spectrometry (SWATH-MS) protein analysis method. Amongst 591 proteins identified across 0, 15, 30 and 45 min PAA treatments, 108 proteins (~6.2% of all proteins identified from the genome) were differentially expressed based on logarithmic two-fold changes. The hypothesis was supported by the detection of two heat stress proteins, Lon protease (Lon) and 10 kDa chaperone (GroS). Other than two heat stress proteins, proteins in the energy production, amino acid-related metabolism, cell mobility and oxidative response were found to be abundant. By comparing changes in the proteome profile across time points (15, 30 and 45 min), Campylobacter was found to actively respond to PAA in the first 15 min and then reached a status quo from 30 to 45 min. In summary, this thesis contributes to further understanding the impact of current poultry processing strategies in terms of how they influence the bacterial diversity on chicken carcasses at various processing steps. Campylobacter was determined to be a minor but persistent component in the bacterial community on carcasses with high prevalence after processing. This suggest that more effort is required to reduce the level of Campylobacter in poultry processing in Australia. The environmental attributes during chilling have an important influence on the bacterial diversity on carcasses through processing. Minimizing levels of heat stress proteins of Campylobacter when implementing PAA during commercial poultry processing is recommended. Together with all the findings in this thesis, it is probably most practical to add PAA in post-chill dipping tanks at the end of processing or spray PAA onto chicken carcasses during air chilling than add PAA in immersion chilling tanks in order to achieve improved Campylobacter reduction. The findings in this thesis may prove to be of value to Australian poultry processors, to develop better processing hurdles to control Campylobacter and limit the public health burden through reducing human Campylobacter illness cases as set out by the Foodborne Illness Reduction Strategy 2018- 2021+.