Development and deployment of a supply-chain digital tool to predict fluid-milk spoilage due to psychrotolerant sporeformers

Psychrotolerant sporeformers pose a challenge to maintaining fluid milk quality. Dynamic temperature changes along the supply chain can favor the germination and growth of these bacteria and lead to fluid milk spoilage. In this study, we aim to expand on our previous work on predicting milk spoilage due to psychrotolerant sporeformers. The key model innovations include (1) the ability to account for changing temperatures along the supply chain, and (2) a deployed user-friendly interface to allow easy access to the model. Using the frequencies and concentrations of 8 Bacillales subtypes specific to fluid milk collected in New York, the model simulated sporeformer growth in half-gallons of high-temperature, short-time (HTST) pasteurized fluid milk transported from processing facility to retail store and then to consumer. The Monte Carlo simulations predicted that 44.3% of half-gallons of milk were spoiled (defined as having a bacterial concentration >20,000 cfu/mL, a conservative estimate that represents the Pasteurized Milk Ordinance regulatory limit) after 21 d of refrigerated storage at consumer's home. Model validations showed that the model was the most accurate in predicting the mean sporeformer concentration at low temperatures (i.e., at 3°C and 4°C; compared with higher temperatures at 6°C and 10°C) within the first 21 d of consumer storage, with a root mean square error of 0.29 and 0.34 log10 cfu/mL, respectively. Global sensitivity analyses indicated that home storage temperature, facility-to-retail transportation temperature, and initial spore concentration were the 3 most influential factors for predicting milk spoilage on d 21 of shelf life. What-if scenarios indicated that microfiltration was predicted to be the most effective strategy to reduce spoilage. The implementation of this strategy (assumed to reduce initial spore concentration by 2.2 log10 cfu/mL) was predicted to reduce the percentage of spoiled milk by 17.0 percentage points on d 21 of storage and could delay the date by which 50% of half-gallons of milk were spoiled, from d 25 to 35. Overall, the model is readily deployed as a digital tool for assessing fluid milk spoilage along the supply chain and evaluating the effectiveness of intervention strategies, including those that target storage temperatures at different supply chain stages.

How Can AI Help Improve Food Safety?

With advances in artificial intelligence (AI) technologies, the development and implementation of digital food systems are becoming increasingly possible. There is tremendous interest in using different AI applications, such as machine learning models, natural language processing, and computer vision to improve food safety. Possible AI applications are broad and include, but are not limited to, ( a) food safety risk prediction and monitoring as well as food safety optimization throughout the supply chain, ( b) improved public health systems (e.g., by providing early warning of outbreaks and source attribution), and ( c) detection, identification, and characterization of foodborne pathogens. However, AI technologies in food safety lag behind in commercial development because of obstacles such as limited data sharing and limited collaborative research and development efforts. Future actions should be directed toward applying data privacy protection methods, improving data standardization, and developing a collaborative ecosystem to drive innovations in AI applications to food safety.

Expected final online publication date for the Annual Review of Food Science and Technology, Volume 14 is March 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Development of a Monte Carlo simulation model to predict pasteurized fluid milk spoilage due to post-pasteurization contamination with gram-negative bacteria

Psychrotolerant gram-negative bacteria introduced as post-pasteurization contamination (PPC) are a major cause of spoilage and reduced shelf life of high-temperature, short-time pasteurized fluid milk. To provide improved tools to (1) predict pasteurized fluid milk shelf life as influenced by PPC and (2) assess the effectiveness of different potential interventions that could reduce spoilage due to PPC, we developed a Monte Carlo simulation model that predicts fluid milk spoilage due to psychrotolerant gram-negative bacteria introduced as PPC. As a first step, 17 gram-negative bacterial isolates frequently associated with fluid milk spoilage were selected and used to generate growth data in skim milk broth at 6°C. The resulting growth parameters, frequency of isolation for the 17 different isolates, and initial concentration of bacteria in milk with PPC, were used to develop a Monte Carlo model to predict bacterial number at different days of shelf life based on storage temperature of milk. This model was then validated with data from d 7 and 10 of shelf life, collected from commercial operations. The validated model predicted that the parameters (1) maximum growth rate and (2) storage temperature had the greatest influence on the percentage of containers exceeding 20,000 cfu/mL standard plate count on d 7 and 10 (i.e., spoiling due to PPC), which indicates that accurate data on maximum growth rate and storage temperature are important for accurate predictions. In addition to allowing for prediction of fluid milk shelf life, the model allows for simulation of "what-if" scenarios, which allowed us to predict the effectiveness of different interventions to reduce overall fluid milk spoilage due to PPC through a set of proof-of-concept scenario (e.g., frequency of PPC in containers reduced from 100% to 10%; limiting distribution temperature to a maximum of 6°C). Combined with other models, such as previous models on fluid milk spoilage due to psychrotolerant spore-forming bacteria, the data and tools developed here will allow for rational, digitally enabled, fluid milk shelf life prediction and quality enhancement.

Identification, subtyping, and tracking of dairy spoilage-associated Pseudomonas by sequencing the ileS gene

Pseudomonas spp. are important spoilage bacteria that negatively affect the quality of refrigerated fluid milk and uncultured cheese by generating unwanted odors, flavors, and pigments. They are frequently found in dairy plant environments and enter dairy products predominantly as postpasteurization contaminants. Current subtyping and characterization methods for dairy-associated Pseudomonas are often labor-intensive and expensive or provide limited and possibly unreliable classification information (e.g., to the species level). Our goal was to identify a single-copy gene that could be analyzed in dairy spoilage-associated Pseudomonas for preliminary species-level identification, subtyping, and phenotype prediction. We tested 7 genes previously targeted in a Pseudomonas fluorescens multilocus sequence typing scheme for their individual suitability in this application using a set of 113 Pseudomonas spp. isolates representing the diversity of typical pasteurized milk contamination. For each of the 7 candidate genes, we determined the success rate of PCR and sequencing for these 113 isolates as well as the level of discrimination for species identification and subtyping that the sequence data provided. Using these metrics, we selected a single gene, isoleucyl tRNA synthetase (ileS), which had the most suitable traits for simple and affordable single-gene Pseudomonas characterization. This was based on the number of isolates successfully sequenced for ileS (113/113), the number of unique allelic types assigned (83, compared with 50 for 16S rDNA), nucleotide and sequence diversity measures (e.g., number of unique SNP and Simpson index), and tests for genetic recombination. The discriminatory ability of ileS sequencing was confirmed by separation of 99 additional dairy Pseudomonas spp. isolates, which were indistinguishable by 16S rDNA sequencing, into 28 different ileS allelic types. Further, we used whole-genome sequencing data to demonstrate the similarities in ileS-based phylogenetic clustering to whole-genome-based clustering for 27 closely related dairy-associated Pseudomonas spp. isolates and for 178 Pseudomonas type strains. We also found that dairy-associated Pseudomonas within an ileS cluster typically shared the same proteolytic and lipolytic activities. Use of ileS sequencing provides a promising strategy for affordable initial characterization of Pseudomonas isolates, which will help the dairy industry identify, characterize, and track Pseudomonas in their facilities and products.

A standard set of testing methods reliably enumerates spores across commercial milk powders

Contamination of dairy powders with sporeforming bacteria is a concern for dairy processors who wish to penetrate markets with stringent spore count specifications (e.g., infant powders). Despite instituted specifications, no standard methodology is used for spore testing across the dairy industry. Instead, a variety of spore enumeration methods are in use, varying primarily by heat-shock treatments, plating method, recovery medium, and incubation temperature. Importantly, testing the same product using different methodologies leads to differences in spore count outcomes, which is a major issue for those required to meet specifications. As such, we set out to identify method(s) to recommend for standardized milk powder spore testing. To this end, 10 commercial milk powders were evaluated using methods varying by (1) heat treatment (e.g., 80°C/12 min), (2) plating method (e.g., spread plating), (3) medium type (e.g., plate count milk agar), and (4) incubation time and temperature combinations (e.g., 32°C for 48 h). The resulting data set included a total of 48 methods. With this data set, we used a stepwise process to identify optimal method(s) that would explain a high proportion of variance in spore count outcomes and would be practical to implement across the dairy industry. Ultimately, spore pasteurized mesophilic spore count (80°C/12 min, incubated at 32°C for 48 h), highly heat resistant thermophilic spore count (100°C/30 min, incubated at 55°C for 48 h), and specially thermoresistant spore enumeration (106°C/30 min, incubated at 55°C for 48 h) spread plating on plate count milk agar were identified as the optimal method set for reliable enumeration of spores in milk powders. Subsequently, we assessed different powder sampling strategies as a way to reduce variation in powder spore testing outcomes using our recommended method set. Results indicated that 33-g composite sampling may reduce variation in spore testing outcomes for highly heat resistant thermophilic spore count over 11-g and 33-g discrete sampling, whereas there was no significant difference across sampling strategies for specially thermoresistant spore enumeration or spore pasteurized mesophilic spore count. Finally, an interlaboratory study using our recommended method set and a modified method set (using tryptic soy agar with 1% starch) among both university and industry laboratories showed increased variation in spore count outcomes within milk powders, which not only was due to natural variation in powders but also was hypothesized to be due to technical errors, highlighting the need for specialized training for technicians who perform spore testing on milk powders. Overall, this study addresses challenges to milk powder spore testing and recommends a method set for standardized spore testing for implementation across the dairy industry.

Nevertheless, She Resisted - Role of the Environment on Listeria monocytogenes Sensitivity to Nisin Treatment in a Laboratory Cheese Model

The growth of Listeria monocytogenes on refrigerated, ready-to-eat food products is a major health and economic concern. The natural antimicrobial nisin targets the bacterial cell wall and can be used to inhibit L. monocytogenes growth on cheese. Cell wall composition and structure, and therefore the efficacy of cell wall acting control strategies, can be severely affected by environmental and stress conditions. The goal of this study was to determine the effect of a range of pH and temperatures on the efficacy of nisin against several strains of L. monocytogenes in a lab-scale, cheese model. Cheese was made with or without the addition of nisin at different pH and then inoculated with L. monocytogenes; L. monocytogenes numbers were quantified after 1, 7, and 14 days of incubation at 6, 14, or 22°C. While our data show that nisin treatment is able to reduce L. monocytogenes numbers, at least initially, growth of this pathogen can occur even in the presence of nisin, especially when cheese is stored at higher temperatures. Several environmental factors were found to affect nisin efficacy against L. monocytogenes. For example, nisin is more effective when cheese is stored at lower temperatures. Nisin is also more effective when cheese is made at higher pH (6 and 6.5), compared to cheese made at pH 5.5, and this effect is at least partially due to the activity of cell envelope modification genes dltA and mprF. Serotype was also found to affect nisin efficacy against L. monocytogenes; serotype 4b strains showed lower susceptibility to nisin treatment compared to serotype 1/2 strains. Overall, our results highlight the importance of considering environmental conditions specific to a food matrix when developing and applying nisin-based intervention strategies against L. monocytogenes.

Bedding and bedding management practices are associated with mesophilic and thermophilic spore levels in bulk tank raw milk

Mesophilic and thermophilic spore-forming bacteria represent a challenge to the dairy industry, as these bacteria are capable of surviving adverse conditions associated with processing and sanitation and eventually spoil dairy products. The dairy farm environment, including soil, manure, silage, and bedding, has been implicated as a source for spores in raw milk. High levels of spores have previously been isolated from bedding, and different bedding materials have been associated with spore levels in bulk tank (BT) raw milk; however, the effect of different bedding types, bedding management practices, and bedding spore levels on the variance of spore levels in BT raw milk has not been investigated. To this end, farm and bedding management surveys were administered and unused bedding, used bedding, and BT raw milk samples were collected from dairy farms (1 or 2 times per farm) across the United States over 1 yr; the final data set included 182 dairy farms in 18 states. Bedding suspensions and BT raw milk were spore pasteurized (80°C for 12 min), and mesophilic and thermophilic spores were enumerated. Piecewise structural equation modeling analysis was used to determine direct and indirect pathways of association among farm and bedding practices, levels of spores in unused and used bedding, and levels of spores in BT raw milk. Separate models were constructed for mesophilic and thermophilic spore levels. The analyses showed that bedding material had a direct influence on levels of spores in unused and used bedding as well as an indirect association with spore levels in BT raw milk through used bedding spore levels. Specific bedding and farm management practices as well as cow hygiene in the housing area were associated with mesophilic and thermophilic spore levels in unused bedding, used bedding, and BT raw milk. Notably, levels of spores in used bedding were positively related to those in unused bedding, and used bedding spore levels were positively related to those in BT raw milk. The results of this study increase the understanding of the levels and ecology of mesophilic and thermophilic spores in raw milk, emphasize the possible role of bedding as a source of spores on-farm, and present opportunities for dairy producers to reduce spore levels in BT raw milk.