Skip to main content
ARS Home » Northeast Area » Wyndmoor, Pennsylvania » Eastern Regional Research Center » Microbial and Chemical Food Safety » Research » Research Project #440376
Research Project: Advanced Methods for Predictive Modeling of Bacterial Growth and Survival in Foods

Location: Microbial and Chemical Food Safety

2024 Annual Report


Objectives
Objective 1: Utilize one-step dynamic modeling and Bayesian analysis for prediction of growth and survival of foodborne pathogens throughout the supply chain. Objective 2: Utilize logistic modeling for determination of growth and no-growth boundary of high-risk pathogens in ready-to-eat foods. Objective 3: Utilize finite element analysis for prediction of bacterial growth and survival during food processing. Objective 4: User-friendly tools for predictive modeling.


Approach
A new dynamic approach will be developed and optimized to simulate and predict the growth and survival of major foodborne pathogens in meat and poultry products exposed to complex changes in the environmental conditions during heating, cooling, and storage. The research will utilize an advanced computational framework and probabilistic Monte Carlo simulation to analyze the dynamic changes in the population of foodborne pathogens, and will develop an expert decision support system to assist the food industry and regulatory agencies in making scientifically sound food safety decisions for products of concern. This project will continue to improve and upgrade the USDA Integrated Pathogen Modeling Program (IPMP) for data analysis, and develop a new data analysis tool, IPMP Global Fit, that minimizes the global residual errors for curve-fitting of growth and survival curves.


Progress Report
For Objective 1, a study was conducted to examine the growth kinetics of mesophilic Bacillus cereus (B. cereus) in liquid egg yolk during treatment with phospholipase A2, an industrial process used to improve the emulsifying capacity and thermal stability of the finished product. The growth of mesophilic B. cereus vegetative cells in inoculated liquid egg yolk samples was observed isothermally between 9 and 50 deg. C in the presence of phospholipase A2. The growth curves were analyzed using the USDA Integrated Pathogen Modeling Program (IPMP)-Global Fit to determine the kinetic parameters and develop predictive models. The predictive mathematical models showed that the minimum, optimum, and maximum growth temperatures are 9.3, 42.7, and 48.4 deg. C for mesophilic B. cereus in liquid egg yolk, with an optimum growth rate of 1.15 log CFU/g per hour. The results of this study showed that, while it cannot grow at 50 deg. C, the mesophilic B. cereus can grow well at 20 deg. C, both of which (50 and 20 deg. C) are frequently used in the food industry to treat liquid egg yolk. This study confirmed the need to keep the temperature above 50 deg. C (such as 55 deg. C) or below 15 deg. C to prevent the growth of both thermotolerant (B. cytotoxicus) and mesophilic B. ceres during treatment of liquid egg yolk with phospholipase A2. Another study was conducted to examine the effect of curing agent (nitrite, NaNO2), curing accelerator (erythorbate), salt (flavoring agent), and sodium tripolyphosphate (STPP) on the growth of Clostridium botulinum (C. botulinum) from spores in cooked cured meats exposed to a long cooling process (5 hours from 54.4 deg. C to 26.7 deg. C and another 10 hours from 26.7 deg. C to 7.2 deg. C), as required in the USDA FSIS Appendix B, Option 1.3. The study was conducted using non-toxigenic C. botulinum mutant LNT01 and C. sporogenes (PA3679) in a fortified reinforced clostridial agar formulated with different combinations of nitrite (100-200 ppm), erythorbate (0 – 547 ppm), salt (1 – 3%), and STPP (0 – 0.5%) to develop a growth and no-growth boundary. STPP, previously found very effective in inhibiting C. perfringens, was not found to be as effective against C. botulinum mutant LNT01 and C. sporogenes. The results of this study may help the food industry to formulate meat products that may inhibit the growth of C. botulinum during food processing, storage, and distribution. To study the effect of extended sublethal exposure and incomplete lethality on growth of Staphylococcus aureus and mesophilic Bacillus cereus in beef during cooking deviation, experiments are being conducted to determine low temperature sublethal heating (50 deg. C for 30 min) for both microorganisms in irradiated ground beef. For Objective 2, a study was conducted to develop a Salmonella growth probability model as a function of the concentrations of sodium chloride (salt), sodium lactate, and sodium diacetate to identify the use of these additives to improve microbial safety of cooked meat products. Sterilized tryptic soy agar (200 µl) formulated with a combination of salt (3-8%, aw 0.98-0.93), lactate (0-2.4%), and diacetate (0-0.25%) and inoculated with a 6-strain Salmonella cocktail was placed into 96-well microplates and incubated at 37°C for 3 days under an aerobic or anaerobic (vacuum) condition. After incubation, a well showing the presence of Salmonella colonies was denoted as a growth event, otherwise a no-growth events. The growth responses of Salmonella under the aerobic or anaerobic condition were similar, indicating the packaging had no effect on the growth responses. The minimum growth-inhibiting concentrations were 3% salt with 0.8% lactate+0.2% diacetate or 1.6% lactate+0.1% diacetate, 4% salt with 2.4% lactate, 5% salt with 0.25% diacetate, 6% salt with 0.8% lactate+0.15% diacetate, 7% salt with 0.8% lactate or 0.15% diacetate, and 8% salt alone. These growth-inhibiting formulations were also shown to inhibit the growth of Salmonella in cooked meat samples. The effects of the additives on the growth responses were analyzed by logistic regression. A model was developed to describe the concentrations of these additives on the growth probability of Salmonella and showed that the additives and their interactions significantly affected the growth probability. The model can be used to estimate the growth probability and growth/no-growth boundaries of Salmonella for risk assessment and select salt, lactate, and diacetate levels for use in refrigerated and shelf-stable meat products to reduce Salmonella risk. The growth kinetics of Salmonella in cooked meat formulated with salt, lactate, and diacetate at refrigerated temperature abuse conditions is also being studied. For Objective 3, a study was conducted to investigate the growth kinetics of Clostridium perfringens (C. perfringens) in cured beef containing different levels of sodium nitrite (curing agent, 25 – 100 ppm), sodium erythorbate (250 ppm), and salt (2.5%) subjected to 15 h of long cooling time (USDA FSIS Option 1.3). The objective of this study was to determine the effect of sodium nitrite on the growth rate and lag time of C. perfringens in cured meats. The results showed that sodium nitrite added at 100 ppm could prevent the growth of C. perfringens during cooling, but it failed to inhibit the cells when added at lower levels. One-step dynamic analysis is being used to determine the growth kinetics of C. perfringens in cured beef by exposing the sample to temperature profiles simulating curing, cooking, and temperature abuse during stabilization. The results of this study may provide the food industry and regulatory agencies (such as USDA FSIS) with a predictive model to evaluate the growth of C. perfringens in cured meat in the event of process deviation during stabilization. For Objective 4, a new desktop-based dynamic predictive tool, the USDA Integrated Pathogen Modeling Program (IPMP)-Dynamic Prediction, was developed. The software was developed using Python 3 (computer language) for modeling and PYQT5 for graphical user interface. It was specifically developed for dynamic prediction of the growth and survival of foodborne pathogens in foods. The final product includes 24 validated dynamic models of Bacillus cereus, Clostridium botulinum, Clostridium perfringens, Cronobacter sakazakii, pathogenic Escherichia coli (O157:H7 and non-O157), Listeria monocytogenes, Salmonella, and Staphylococcus aureus, in various foods, such as raw and processed meat and poultry products, infant formula, cooked rice, liquid or hard-boiled eggs, and potato salad. The software is designed with an easy-to-navigate graphical user interface, allowing the users to input the temperature history of a product and then predict the growth and survival of foodborne pathogens using the dropdown menus and buttons. It is suitable for evaluating the safety of a product undergoing complex changes in temperature throughout the supply chain. IPMP-Dynamic Prediction has been compiled to a binary format and uploaded to the server for public distribution. It does not have internet security concern. In addition, the USDA Integrated Pathogens Modeling Program, originally developed in 2014 using 32-bit Python 2, is being upgraded to 64-bit Python 3 and PYQT5 since Python 2 has officially been discontinued and Python 3 has become a stable distribution. A new user graphical interface has been developed together with an improved algorithm for object-oriented programming, making it easier for future upgrades. Object-oriented programming will support easier code reuse in future upgrades. The new software, although still under development, has been expanded to include more mathematical models of application in predictive microbiology. The new software will provide additional capabilities and better graphical user interface. In addition, online applications based on Python3 is still being evaluated. The major challenge is to find a suitable vendor to deploy the products in a secure manner meeting the USDA IT security requirements. A study was conducted to examine the protective effect of animal fat during thermal inactivation of three foodborne pathogens, including Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes. The survival of these pathogens inoculated to rendered beef fat (tallow) was observed isothermally at temperatures between 55 and 80 deg. C. The experimental results showed that these pathogens could survive the heat treatment process at much higher temperatures and for longer times than those in meat and poultry. The survival curves showed that the thermal resistance of each pathogen gradually increased with the heating time at each temperature and can be described by either a 2-stage linear model or the Weibull model. The study clearly demonstrates the need to increase both heating time and temperature in order to inactivate the foodborne pathogens contaminated in the fat portion of meat and poultry products. It also suggests that it may be necessary to investigate the relative distribution of foodborne pathogens in the fat and non-fat portions of meat and poultry for development of adequate thermal processes to ensure food safety. A risk assessment evaluating the effect of potential survival of foodborne pathogens after heating in the fat-containing portion of products (such as ground beef) on public health may be needed.


Accomplishments
1. Development of the USDA Integrated Pathogen Modeling Program (IPMP)-Dynamic Prediction. IPMP-Dynamic Prediction is a new desktop software product that has been developed by ARS researchers at Wyndmoor, Pennsylvania. This software includes 24 validated dynamic models of major foodborne pathogens, such as Bacillus cereus, Clostridium perfringens, Clostridium botulinum, pathogenic Escherichia coli, Listeria monocytogenes, and Salmonella, in various foods, such as raw and processed meat and poultry products. It is designed with an easy-to-navigate graphical user interface, allowing the users to input the temperature history to predict the growth and survival of foodborne pathogens in foods undergoing complex changes in temperature throughout the supply chain. IPMP-Dynamic Prediction has been compiled to a binary format and uploaded to the server for public distribution.

2. Dynamic analysis of growth of thermo-tolerant Bacillus cereus in liquid egg yolk. Liquid egg yolk is a common household and industrial ingredient for making various food products. The commercial production of liquid egg yolk involves an enzymatic treatment process to improve its emulsifying capacity and thermal stability. Enzymatic treatment may occur at a high temperature condition for extended time, potentially allowing thermo-tolerant Bacillus cytotoxicus (B. cytotoxicus) to grow and produce an enterotoxin that affects public food safety. Under an interagency agreement with the USDA FSIS, ARS scientists at Wyndmoor, Pennsylvania, investigated the growth kinetics of B. cytotoxicus in liquid egg yolk and developed a dynamic model to predict its growth during enzymatic treatment. The results showed that B. cytotoxicus can grow prolifically between 17 and 52 degrees C. However, B. cytotoxicus gradually died off at temperature below 15 degrees C and was inactivated at 55 degrees C. The predictive model has been incorporated into the USDA IPMP-Dynamic Prediction. The results from this study may allow the food industry to design an enzymatic treatment process at temperatures to prevent the growth of B. cytotoxicus during liquid egg processing.

3. Growth kinetics of mesophilic Bacillus cereus in liquid egg yolk. Mesophilic Bacillus cereus (B. cereus) is abundant in the environment and can contaminate liquid egg yolk. This pathogen may grow during enzymatic treatment of liquid egg yolk used to improve the emulsifying capability and thermostability. ARS scientists at Wyndmoor, Pennsylvania designed a study to investigate the growth kinetics of mesophilic B. cereus liquid egg yolk and develop a mathematical model to predict its growth during enzymatic treatment. The results showed that mesophilic B. cereus can grow between 9.3 and 48.4 deg, but at a slower rate than B. cytotoxicus. The mathematical models developed in this study also can be used in the food industry to treat liquid eggs at temperatures preventing the growth of mesophilic B. cereus.

4. Computer simulation of conductive heat transfer during meat cooking. Cooking is the most effective method to kill foodborne pathogens in meat products. However, undercooking allows the survival of foodborne pathogens. It is necessary to ensure the final cooking temperature of a product, but it may not be possible to measure the product temperature in real time. ARS scientists at Wyndmoor, Pennsylvania, developed and validated a computer simulation program using the finite volume method to predict the cooking temperature in thermally conductive foods such as meat balls. Together with the thermal inactivation kinetics (D and z values), the computer simulation program can be used to design cooking processes that can effectively inactivate foodborne pathogens. In addition, the computer program also can be used to simulate the temperature distribution and history in cooked meat and poultry products during cooling to evaluate the effect of cooling on growth of C. perfringens. It can be used by the food industry and regulatory agencies to enhance the safety of cooked products.

5. Enhancing the safety of cold smoked salmon. Cold smoked salmon is a high-value seafood product enjoyed by consumers worldwide. However, commercial manufacturing of cold smoked salmon does not include a cooking process normally used to kill foodborne pathogens, such as Listeria monocytogenes. Rather, cold smoked salmon relies on various antimicrobial compounds in the liquid smoke and refrigeration for food preservation, which is not sufficient to prevent the growth of L. monocytogenes. ARS scientists at Wyndmoor, Pennsylvania, developed a mathematical model to predict the growth probability of L. monocytogenes in cold smoked salmon as a function of storage temperature and time. The model has a greater than 89% accuracy for predicting no growth of L. monocytogenes. It provides a risk-based approach to manage the shelf-life of cold smoked salmon to manage the growth of L. monocytogenes in cold smoked salmon.

6. Critical control surfaces to prevent growth of Clostridium perfringens in cured meat. Clostridium perfringens (C. perfringens) is a major foodborne pathogen affecting the safety of meat products, including cured meats. To ensure food safety, cured meat products must be properly cooled after cooking to prevent the germination, outgrowth, and multiplication of C. perfringens. Cooling deviation during meat processing may lead to production of C. perfringens enterotoxin (CPE), causing foodborne outbreaks. ARS scientists at Wyndmoor, Pennsylvania, developed and validated a mathematical model to create critical control surfaces using sodium nitrite (curing agent), sodium erythorbate (curing accelerator), sodium tripolyphosphate (water-binding and antimicrobial agent), and salt to prevent the growth of C. perfringens in cured beef even under the optimum growth temperature. The results from this study may be used by the food industry to formulate cured meat products to prevent the growth of C. perfringens and foodborne outbreaks of CPE poisoning.


Review Publications
Huang, L., Jia, Z., Hwang, C. 2021. Growth and no-growth boundary of Listeria monocytogenes in beef – A logistic modeling. Food Research International. 152:110919. https://doi.org/10.1016/j.foodres.2021.110919.
 Huang, L., Hwang, C. 2021. One-step dynamic analysis of growth kinetics of Bacillus cereus from spores in egg fried rice – Model development, validation, and marko chain monte carlo simulation. Food Microbiology. 103. https://doi.org/10.1016/j.fm.2021.103935.
 Chai, H., Hwang, C., Huang, L., Wu, V.C., Sheen, L. 2021. Efficacy of gaseous chlorine dioxide for decontamination of Salmonella, Shiga toxin-producing Escherichia coli, and Listeria monocytogenes on almonds and peppercorns. Food Control. https://doi.org/10.1016/j.foodcont.2021.108556.
 Tan, J.N., Hwang, C., Huang, L., Wu, V.C., Hsiao, H. 2021. A pilot-scale evaluation of using gaseous chlorine dioxide for decontamination of foodborne pathogens on produce and low-moisture foods. Journal of Food Safety. http://doi.org/10.1111/jfs.12937.
 Huang, L., Hwang, C., Sheen, S. 2023. Shelf-life boundaries of listeria monocytogenes in cold smoked salmon during refrigerated storage and temperature abuse. Food Research International. 173:113362. https://doi.org/10.1016/j.foodres.2023.113362.
 Huang, L., Ahmad, N., Juneja, V.K., Stapp-Kamotani, E., Gabiola, J., Minocha, U., Phillips, R., Hooker, M., Walls, I., Cook, K., Lindsay, J. 2023. Growth kinetics of Bacillus cytotoxicus in liquid egg yolk during treatment with phospholipase A2 – A one-step global dynamic analysis. Food Microbiology. 118:104420. https://doi.org/10.1016/j.fm.2023.104420.
 Ahmad, N., Huang, L., Juneja, V.K. 2023. One-step analysis of growth kinetics of mesophilic Bacillus cereus in liquid egg yolk during treatment with Phospholipase A2: Model development and validation. Food Research International. 176:113786. https://doi.org/10.1016/j.foodres.2023.113786.
 Ahmad, N., Huang, L., Hwang, C. 2024. Growth and no-growth boundary of Clostridium perfringens in cooked cured meats – A logistic analysis and development of critical control surfaces using a solid growth medium. Food Research International. 191:114701. https://doi.org/10.1016/j.foodres.2024.114701.
 Sheen, S., Huang, L., Hwang, C. 2024. Numerical simulation of heat transfer during meat ball cooking and microbial food safety enhancement. Journal of Food Science. https://doi.org/10.1111/1750-3841.16949.
 Jiang, T., Guo, F., Fang, T., Hwang, C., Huang, L. 2022. Efficacy of gaseous chlorine dioxide generated by sodium chlorite - carbon dioxide reaction on safety and quality of blueberries, cherry tomatoes, and grapes. Food Control. 143:109288. https://doi.org/10.1016/j.foodcont.2022.109288.
 Hwang, C., Huang, L., Sheen, S. 2024. Modeling the growth probability of Clostridium perfringens in cooked meat as affected by sodium chloride and sodium tripolyphosphate. Microbial Risk Analysis. https://doi.org/10.1016/j.mran.2024.100296.