THERMAL INACTIVATION OF BACTERIAL PATHOGENS UNDER WIDELY CHANGING MOISTURE CONDITIONS IN COOKED BACON AND DRIED APPLE By Narindra Randriamiarintsoa A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Biosystems Engineering – Master of Science 2022 ABSTRACT The overall goal of this thesis was to evaluate phenomenological similarity in bacterial pathogen inactivation under different thermal treatments of two very different food products (cooked bacon and dried apples) with similarly wide changes in moisture during processing. As ready-to-eat (RTE) products, both must comply with specific food safety regulations, under the United States (US) Department of Agriculture – Food Safety Inspection Service and the Food Safety Modernization Act Preventive Controls for Human Foods Rule, respectively. Therefore, there is a need for pathogen inactivation data to validate commercial pathogen control processes for both products. Both conventional and microwave oven cooking of bacon to the required 40% cooking yield achieved >6.5 log reduction of Salmonella. However, when humidity was reduced (dew point ≤25°C), microwave cooking of bacon yielded <6.5 log reduction. When drying apples to a standard moisture content (<24% wet basis), lower Listeria monocytogenes inactivation (1.8 and 2.8 log CFU) was achieved when drying at 60°C, under the studied air velocities (0.7 and 2.1 m/s) (P < 0.05), compared to 80°C, at which Listeria decreased by 5 log reduction by the end of drying. Despite the use of different pathogens, similar inactivation response patterns were observed during both apple drying and bacon cooking, especially microwave cooked bacon under dry conditions, reflecting the simultaneous counter-effects of dynamically increasing product temperature and decreasing product moisture. Therefore, results from this study suggest that it is theoretically possible to develop one model form for bacterial inactivation in widely changing moisture products. ACKNOWLEDGEMENTS First, I would like to express my gratitude to God who has sustained me in this journey and has given me the courage, strength, and health to fulfill my master’s thesis. I am indebted to my advisor, Dr. Bradley Marks for his endless support and guidance throughout my research and my studies at MSU; I could not have asked for a better mentor. I am also grateful for my committee members, Dr. Elliot Ryser and Dr. Sanghyup Jeong, whose dedication and invaluable perspectives have greatly impacted my research skills and critical thinking. This research would not have been possible without Michael James, our lab manager, whose diligence and great organizational skills allowed me to finish all my experiments on time, despite the pandemic. I would like to also thank Ian Hildebrandt for guiding me and helping me with experimental design and data analysis. Additionally, I am thankful for my fellow BAE graduate students and lab mates for their great support in my experiment and for making lab fun, especially Carly, Natasha, Kase, Ian K., Zoe, Andrew, and Christina. Last, not the least, I am grateful for my family and friends for always being by my side and for encouraging me to always do my best, despite the distance. This study was partially funded by the Foundation for Meat and Poultry Research and Education and USDA National Institute for Food and Agriculture (NIFA) Grants No. 2021- 68008-34196 and 2020-68012-31822. iii TABLE OF CONTENTS LIST OF TABLES ......................................................................................................................... vi LIST OF FIGURES ...................................................................................................................... vii CHAPTER 1: INTRODUCTION ................................................................................................... 1 1.1 Foodborne Outbreaks from Ready-To-Eat (RTE) products ................................................. 1 1.2 Salmonella Lethality Requirement for RTE Meat ................................................................ 2 1.3 Food Safety Modernization Act (FSMA) for Other Foods ................................................... 3 1.4 Microbial Safety of Bacon and Dried Apple: Knowledge Gap ............................................ 3 1.5 Goal and Objective ............................................................................................................... 5 CHAPTER 2: LITERATURE REVIEW ........................................................................................ 6 2.1 Bacon .................................................................................................................................... 6 2.1.1 Bacon Processing ........................................................................................................... 6 2.1.2 Microbial Inactivation Studies for Bacon ...................................................................... 6 2.2 Dried Apple ........................................................................................................................... 8 2.2.1 Apple Drying ................................................................................................................. 8 2.2.2 Microbial Inactivation Studies for Dried Apples ........................................................... 9 2.3 Effect of Moisture on Pathogen Inactivation ...................................................................... 10 2.4 Summary ............................................................................................................................. 12 CHAPTER 3: THERMAL INACTIVATION OF SALMONELLA DURING BACON PROCESSING .............................................................................................................................. 14 3.1 Materials and Methods ........................................................................................................ 14 3.1.1 Overall Study Design ................................................................................................... 14 3.1.2 Sample Preparation ...................................................................................................... 14 3.1.3 Culture Preparation and Inoculation ............................................................................ 15 3.1.4 Microwave Oven Treatment ........................................................................................ 15 3.1.5 Impingement Oven Treatment ..................................................................................... 16 3.1.6 Comparison of Salmonella Inactivation in the Lean and Fat Portions of Bacon ......... 17 3.1.7 Effect of Humidity on Salmonella Inactivation during Bacon Microwave Cooking .. 18 3.1.8 Salmonella Enumeration, Moisture Content, and Water Activity Analyses ............... 20 3.1.9 Statistical Analyses ...................................................................................................... 20 3.2 Results and Discussion ....................................................................................................... 21 3.2.1 Microwave Oven Cooking ........................................................................................... 21 3.2.2 Impingement Oven Cooking ........................................................................................ 25 3.2.3 Comparison of Salmonella Inactivation in the Lean and Fat Portions of Bacon ......... 27 3.2.4 Effect of Humidity on Salmonella Inactivation in Microwave Cooked Bacon ........... 29 3.3 Conclusions ......................................................................................................................... 32 CHAPTER 4: LISTERIA MONOCYTOGENES INACTIVATION DURING APPLE DRYING UNDER LOW OR MODERATE TEMPERATURE AND AIR VELOCITY ............................ 33 4.1 Materials and Methods ........................................................................................................ 33 4.1.1 Study Design ................................................................................................................ 33 iv 4.1.2 Sample Preparation ...................................................................................................... 33 4.1.3 Culture preparation and sample inoculation ................................................................ 34 4.1.4 Apple Drying ............................................................................................................... 35 4.1.5 Listeria Enumeration ................................................................................................... 37 4.1.6 Statistical Analyses ...................................................................................................... 38 4.2 Results and Discussion ....................................................................................................... 38 4.2.1 Moisture Content and Water Activity .......................................................................... 38 4.2.2 Colorimetric Results .................................................................................................... 42 4.2.3 Listeria Inactivation ..................................................................................................... 42 4.3 Conclusions ......................................................................................................................... 44 CHAPTER 5: CONCLUSIONS AND FUTURE WORK ............................................................ 45 5.1 Overall Findings.................................................................................................................. 45 5.2 Summary ............................................................................................................................. 47 5.3 Future Work ........................................................................................................................ 49 REFERENCES ............................................................................................................................. 51 APPENDIX A: Collected Data from Bacon Cooking Study ........................................................ 61 APPENDIX B: Collected Data from Apple Drying Study ........................................................... 75 v LIST OF TABLES Table 1: Moisture content, water activity, cooking yield, and Salmonella reduction in bacon slices during microwave cooking.................................................................................................. 22 Table 2: Moisture content, water activity, yield, and Salmonella reduction in bacon slices during impingement cooking.................................................................................................................... 25 Table 3: Moisture content, water activity, yield, and Salmonella reduction in bacon fat and bacon lean during microwave cooking ......................................................................................... 29 Table 4: Moisture content, water activity, yield, and Salmonella reduction during microwave cooking under a dry and less humid conditions ............................................................................ 30 Table 5: Multiple comparisons of the moisture content, water activity, surface temperature, and Listeria population during apple drying across the studied conditions. ....................................... 40 Table 6: Comparison of bacon cooking and apple drying described in this study ...................... 47 Table 7: Collected data from Salmonella inactivation during microwave cooking of bacon ...... 61 Table 8: Collected data from comparison of Salmonella inactivation in bacon fat and bacon lean ................................................................................................................................................ 66 Table 9: Collected data for Salmonella inactivation during bacon microwave cooking where humidity was controlled. ............................................................................................................... 68 Table 10: Collected data for Salmonella inactivation during impingement oven cooking of bacon ............................................................................................................................................. 71 Table 11: Collected data for Listeria monocytogenes inactivation during apple drying ............. 75 vi LIST OF FIGURES Figure 1: Diagram of the humidity control system during microwave cooking of bacon ........... 19 Figure 2: Water activity, moisture content, surface temperature, and Salmonella populations during impingement and microwave cooking of bacon.. .............................................................. 23 Figure 3: Infrared images of bacon slices A) during microwave cooking and B) during impingement oven cooking ........................................................................................................... 24 Figure 4: Surface temperature of bacon cooked in an impingement oven. ................................. 26 Figure 5: Water activity, moisture content, surface temperature, and Salmonella populations in bacon fat and lean portions of bacon and the effect of humidity during microwave cooking of bacon. ............................................................................................................................................ 28 Figure 6: Dew point during microwave cooking of bacon.. ....................................................... 31 Figure 7: Moisture content, water activity, surface temperature, and Listeria populations during apple drying. ................................................................................................................................. 39 vii CHAPTER 1: INTRODUCTION 1.1 Foodborne Outbreaks from Ready-To-Eat (RTE) products Foodborne illness is an ongoing global problem, annually affecting approximately 600 million people worldwide, according to the World Health Organization (108). In the United States (US), the Centers for Disease Control and Prevention (CDC) estimated that 48 million illnesses are annually associated with foodborne agents (15), in which fresh fruits and meat products, such as chicken, pork, beef, and turkey, are the top five food categories associated with outbreak-related illnesses (31). In those foodborne outbreaks, Salmonella and Listeria are among the most well-known bacterial agents, responsible for a total of 210 outbreaks between 2019 and 2020 (26). Salmonella is a gram-negative bacterium naturally found in the intestinal tract of animals, which can contaminate foods through soil, water, and equipment (5, 12, 46, 50, 58, 63, 74), and then grow if conditions such as pH, temperature, and water activity (aw) are favorable (41, 56). Data from the CDC FoodNet show that Salmonella is the bacterial pathogen that causes the most foodborne infections and deaths in the US (30). While Salmonella foodborne outbreaks have been associated with various products, such as fruits and vegetables, low-moisture foods, and meat products, outbreaks related to meat are among the most common, accounting for an overall economic cost of $4.7 billion in pork and chicken alone (67). Foodborne outbreaks linked to Listeria are not as common as Salmonella. However, because of the high mortality rate of listeriosis (104), Listeria causes the second most deaths due to foodborne illnesses in the US (30). Listeria-related outbreaks have been most often traced to RTE meats (27, 28, 29), dairy products (16, 17, 22, 23), fruits, and vegetables (18, 19, 20, 21, 24, 32). Given the persistent issue of foodborne illnesses, federal agencies, such as the US Food and Drug Administration (FDA) and the US Department of Agriculture Food Safety and Inspection 1 Service (USDA FSIS), recognize that food safety is a shared responsibility among stakeholders involved in the food supply chain. Therefore, in addition to recommending safe food handling practices for consumers (91, 95), the FDA and USDA FSIS also require food processers to demonstrate and validate pathogen reductions in various food commodities. These specific regulations are governed by several federal acts, such as the Federal Meat Inspection Act for meat products, the Poultry Products Inspection Act for poultry products, the Egg Products Inspection Act for egg products, and the Food Safety Modernization Act for other food products (89, 96). 1.2 Salmonella Lethality Requirement for RTE Meat Food safety requirements for some RTE meat products have been codified and published in the Federal Register. For example, the Code of Federal Regulation 9 CFR 318.17(a)(1) requires that RTE beef processing, including cooked beef, roast beef, and cooked corned beef, must achieve a 6.5 log reduction of Salmonella (34). Moreover, for uncured meat patties, 9 CFR 318.23 provides a minimum internal temperature/minimum holding time combination under which fully cooked patties must be processed (35), in order to achieve a 5 log reduction of Salmonella and other pathogens, such as shiga toxin-producing E. coli (90). Lastly, for fully cooked poultry products, a 7-log reduction of Salmonella must be met according to 9 CFR 318.150(a)(1) (36). For other RTE meat products, such as cooked bacon, that were not covered by the described regulations, FSIS (90) recommended a 6.5 log reduction for Salmonella or a 5- log reduction with additional support or testing to meet the 9 CFR 417 requirements for hazard analysis and critical control points (33). 2 1.3 Food Safety Modernization Act (FSMA) for Other Foods Adopted in 2011, FSMA codified a more integrated and preventative approach to food safety (93, 96). To achieve this goal, FSMA was implemented by FDA as a series of different rules, such as the Preventive Controls for Human Foods Rule (98). This rule requires food processing facilities to have risk-based food safety plans for pathogen control (98). All food products belonging to categories regulated by FDA are subject to FSMA. These include fresh produce and low-moisture foods, such as dried fruits, but exclude meat, poultry, and fresh eggs, which are regulated by the USDA (102). 1.4 Microbial Safety of Bacon and Dried Apple: Knowledge Gap Cooked bacon and dried apples are two very different products, but they undergo similarly wide moisture changes during processing. Both bacon and dried apple processing sufficiency are standardized based on the endpoint yield or moisture of the product, rather than time-temperature requirements. USDA labeling policy requires that bacon can only be labelled as fully cooked if cooked to a final yield of ≤ 40% (88, 92). Similarly, for dried apples, the moisture content of the final product must not exceed 24% (wet basis (wb)) (82). In both bacon and apple, as long as the standards for yield and moisture content are met, industries can use a range of commercial processing methods and conditions. However, without appropriate process validation, this could result in food safety concerns. For example, for apples, several methods, such as sun drying, microwave drying, or oven drying, are used, in which each processor, either small- or large-scale, uses different parameters (e.g., temperatures and air velocity) depending on their preference and capacity. Even though these various drying methods might result in a similar final moisture content of the apple 3 products, they might not yield similar pathogen reductions (7). As an illustration, when apples were previously dried at 104 or 135°C, >5 log reduction of Salmonella was achieved by the end of drying (45). However, in another study, only a 2.8-log reduction was achieved when apples were dried at 60°C to a similar aw (39). Therefore, because the FSMA Preventative Controls Rule applies to dried apples, evaluating pathogen reductions under those different processing methods and conditions is critical for regulatory compliance. Microwave cooking of bacon has become among the most common processing methods used by industries (51), in addition to moist-air convection-oven cooking. However, given the FSIS recommendation for Salmonella lethality in RTE meat products such as bacon (90), less scientific evidence is currently available regarding pathogen behavior under microwave cooking. Consequently, FSIS has recently identified Salmonella lethality in bacon undergoing microwave cooking as a specific knowledge gap and has encouraged performing challenge studies (90). Another similarity between cooked bacon and dried apples is that they both undergo very wide changes in moisture during processing. During bacon cooking, moisture evaporates and fat is rendered out of the product, decreasing the cook yield, based on total mass. Similarly, in apple drying, moisture is removed as the drying process progresses, via mass convection at and diffusion to the surface. In both cases, the process starts with a high-moisture product and ends with a low-moisture product. As the temperature of the product increases, higher pathogen inactivation would be expected. However, because of the complex coupled heat and mass transfer processes, as the temperature increases, moisture content decreases. Studies have shown that pathogens such as Salmonella become more heat resistant in lower-moisture products (2, 4, 43, 49, 52, 66, 72, 76, 78, 107, 109, 110). Consequently, when the process involves a dynamic decrease in moisture, such as in apple drying and bacon cooking, there is a constant tradeoff of 4 increasing temperature to inactivate pathogens, but also removing moisture and therefore simultaneously increasing the thermal resistance of pathogens (7). The question is how to identify the critical points where pathogen inactivation is sufficient when dealing with dynamic changes in moisture over time in a process. Previous studies working on the effect of moisture content or aw on pathogen inactivation focused more on controlled systems, in which one variable at a time was assessed, such as in isothermal studies (2, 4, 43, 66, 72, 76, 109, 110). Additionally, those studies were mostly done on products that already had low moisture, such as almond kernels, wheat flour, ground cinnamon, and protein powder, which is different from starting with a high-moisture product, such as freshly sliced apples or bacon. Therefore, although the insights from those prior studies are important to understanding pathogen thermal resistance in low-moisture foods, their applicability is limited when dealing with dynamic processes involving very wide changes in both temperature and moisture. 1.5 Goal and Objective The goal of this study was to evaluate the phenomenological similarity between Salmonella and Listeria monocytogenes thermal inactivation across two very different food products (cooked bacon and dried apples) with similarly wide changes in moisture during processing. To achieve this goal, the objectives of this study were: 1) To quantify the thermal inactivation of Salmonella in bacon cooked in an impingement oven under industry-typical conditions. 2) To quantify the thermal inactivation of Salmonella in bacon cooked in a microwave oven to an industry-typical product endpoint. 3) To quantify the impact of air drying temperature and velocity on the inactivation of Listeria monocytogenes during drying of apple slices. 5 CHAPTER 2: LITERATURE REVIEW 2.1 Bacon 2.1.1 Bacon Processing Pork products such as RTE bacon are the third most consumed meat in the US (86). With only two outbreaks reported between 1998 and 2015 (68), bacon has a long record of being a microbiologically safe product. Sliced bacon is cooked using pan frying, conventional oven, or microwave oven, in which the yield must be ≤ 40% to be labeled as fully-cooked (88, 92). For microwave cooking, which is the most used method commercially, James et al. (51) showed that power-output, heating time, and position of the slices in the oven all affect the quality of RTE bacon. However, they assessed quality only subjectively, based on factors such as the degree of doneness and crispiness (51). An industrial microwave oven with a 2450 MHz frequency and 6 kW power output was used, and results showed that by using a 2.2 kg load of sliced bacon, optimal bacon quality was obtained around 115 s (51). Although such information may be useful to the industry in optimizing processes, validation for sufficient Salmonella lethality (6.5 log reduction) is needed, as indicated by scientific gaps identified in the recently revised USDA- FSIS Revised Appendix A (90). 2.1.2 Microbial Inactivation Studies for Bacon Prior studies on bacon safety have used both thermal and non-thermal methods to evaluate pathogen inactivation (8, 10, 11, 37, 70, 77). A methicillin resistant Staphylococcus aureus (MRSA) inactivation study in bacon under home-cooking conditions showed that grilling bacon slices at 177°C for 5 min resulted in ≥6.5 log reduction of MRSA (11). Results from that study can be useful to inform consumer practices on safe bacon handling and cooking. However, 6 from a regulatory perspective, MRSA is not the target pathogen in bacon (90), and a >6.5 log reduction of MRSA may not directly translate to a similar reduction of Salmonella. Additionally, grilling is not a common industrial, large-scale method, which restricts the applicability of the information. Other thermal inactivation studies focused more on bacon slab processing, which is traditionally a partially-cooked product obtained from slow heating and smoking of cured pork bellies (77). Smoked bacon slabs are usually then used to make raw bacon slices (77). Most survival studies on bacon slabs have investigated the behavior, growth, or inhibition of bacterial pathogens such as Clostridium perfringens, Escherichia coli, Listeria monocytogenes, MRSA, Salmonella enterica, and Staphylococcus aureus during cooking, smoking, or cooling (8, 11, 37, 70, 77). Because the endpoint of these processes results in a partially-cooked bacon slab or smoked raw bacon slices, none of these studies answer the question about pathogen reduction during subsequent cooking. A non-thermal approach also has been investigated for its effectiveness on pathogen inactivation during bacon processing. By using non-thermal atmospheric pressure-plasma (NTAP), Calvo et al. (10) obtained only a 1.3 log reduction of Salmonella on sliced bacon after a 15 min treatment. Moreover, NTAP is currently not common in industry. In summary, a significant knowledge gap exists in the literature, with no prior study having systematically evaluated Salmonella thermal reduction during cooking of RTE bacon in commercial-type conditions, which is essential for process validations as identified by the USDA-FSIS. 7 2.2 Dried Apple 2.2.1 Apple Drying Apples are the most available fruit for consumption in the US (85), with an average annual production of 4.4 billion kg (87). Whereas some are distributed directly to consumers as fresh apples, 37% of harvested apples are further processed (87). Data from the US Department of Agriculture Economic Research Service (USDA ERS) show an average production of 113 million kg of dried apples (87), demonstrating the importance of the apple drying industry in the US. According to the US standards of identity, dried apples are defined as apples that undergo some drying process in which the final moisture content must not exceed 24% (wet basis (wb)) (82). Common drying methods include sun drying, conventional hot-air oven drying, microwave drying, and vacuum drying. In most of these methods, factors such as temperature and air velocity affect the rate of drying and therefore likely affect microbiological outcomes. Unlike the regulation of RTE meats, where a specific log reduction of Salmonella is required (42), no specific target pathogen or reduction has been established for dried apples or other fruits. This might be because dried apples have had a long history of being a safe product without reported outbreaks. However, recent outbreaks and recalls in other dried fruits show the need for ensuring pathogen reduction during fruit drying. For example, a Salmonella outbreak associated with dried coconut was reported in 2018 (25), and freeze-dried sliced fruits (100) and dried apricots (103) were recalled due to potential contamination of Salmonella and Listeria monocytogenes, respectively. Additionally, listeriosis outbreaks from caramel apples in 2015 (32) and 2017 (59), along with several recalls of whole (99, 101) or sliced apples (94, 97), show that Listeria monocytogenes is an emerging pathogen of concern in apple products. The Food Safety Modernization Act (FSMA) final rule for Preventive Controls for Human Foods, 8 implemented by the US Food and Drug Administration (FDA), also requires food processing facilities, such as apple drying facilities, to have risk-based food safety plans for their processes (98). Therefore, there is a need to validate commercial practices for apple drying to ensure sufficient inactivation of key pathogens. 2.2.2 Microbial Inactivation Studies for Dried Apples Previous apple drying inactivation studies have focused only on Salmonella and E. coli and did not include Listeria. These studies have shown that conditions such as temperature, pre- treatment method, and apple variety are critical factors in inactivation. When drying apples at 60°C for 6 h, Dipersio et al. (39) reported only up to a 2.8 log reduction of Salmonella. Likewise, drying apples at 57.2°C and 62.8°C for 6 h reduced E. coli population by 2.9 and 3.3 log, respectively (9). Similar E. coli reductions were reported in another study conducted at 62.8°C for 6 h (38). In contrast, at higher drying temperatures of 104 and 135°C, Grasso-Kelley et al. (45) reported > 5 log reduction of Salmonella in apples. In some of these studies, the final aw of the apples was similar, but the log reduction was different because of the drying temperature. For example, in Grasso-Kelley et al. (45), the final aw that corresponded to the > 5 log reduction of Salmonella was 0.247 ± 0.070 at 104°C. In Dipersio et al. (39), the final aw was between 0.229 and 0.257 at 60°C; however, the Salmonella reduction was lower, as illustrated above. Therefore, pathogen inactivation during apple drying depends on the history of drying, such as temperature and drying time, not necessarily just the endpoint state, defined by the standard as moisture content. This means that following a standard moisture content for dried apples may not necessarily translate to sufficient pathogen inactivation. 9 Several studies on apple drying also assessed the use of various pre-treatment methods in which the use of acidic solutions was reported to have enhanced the reduction of Salmonella and E. coli due to their antimicrobial properties (9, 38, 39, 47), whereas methods such as steam blanching were shown to have no effect (9). Only one of these prior studies controlled air velocity, where a low 0.4 m/s air velocity was used during drying. Given that air velocity is an important factor in drying kinetics of foods, especially in fruits and vegetables (55, 57, 79, 106), and that air velocity has been shown to impact the inactivation of pathogens such as E. faecium and Salmonella in low-moisture foods (44, 64), examining the impact of drying air velocity on bacterial survival during apple drying is also needed. 2.3 Effect of Moisture on Pathogen Inactivation In food processing, there are two possible sources of moisture. Moisture can be intrinsic to the food product itself and is usually quantified in terms of aw and/or moisture content. However, moisture can also be introduced during hot-air processing (e.g., water vapor / steam injection) as an extrinsic factor. Previous studies have shown that both forms of moisture can affect pathogen inactivation in food (2, 4, 13, 14, 43, 49, 52, 66, 72, 76, 78, 107, 109, 110); however, for the purpose of this literature review, only the effect of product moisture and/or aw on pathogen inactivation was investigated. Studies on various low-moisture food products have shown that Salmonella becomes more thermally resistant when aw is low. Although the mechanism of such resistance is not entirely understood, Spector and Kenyon (73) hypothesized that Salmonella initiates a stress response mechanism to persist under various environmental conditions, such as desiccation. When wheat flour was isothermally heated at 80°C, Smith et al. (72) reported a D-value of 1.27 ± 0.06 min at 0.582 aw. However, when the aw was reduced to 0.310, the D-value increased to 10.27 ± 0.65 10 min. Similar increased resistance of Salmonella was observed in other studies involving low- moisture foods, such as pistachios, almonds, and protein powder (1, 2, 39, 46, 54, 61, 63, 88, 90, 91). Although Salmonella has been the most reported pathogen to exhibit increased resistance in low-moisture environments, studies have shown that L. monocytogenes can also develop similar resistance in low-moisture foods, such as cocoa powder. Tsai et al. (80) reported D-values of 3.4 ± 0.2 and 11.0 ± 0.5 min at 75°C in cocoa powder, at aw 0.45 and 0.30, respectively. Similarly, another study investigating the thermal resistance of Salmonella, E. coli, and Listeria in pistachios demonstrated that although Salmonella showed the greatest resistance among those pathogens during drying, all of the pathogens showed similar thermal resistance in pistachios when exposed to hot water or hot oil at 80 and 121°C, respectively (60). The effect of aw on pathogen thermal resistance in low-moisture food may depend on other factors, such as the food composition. Jin et al. (110) reported a complex interaction of temperature, food components, and aw on Salmonella thermal resistance in different food matrices. For instance, at aw 0.9 and 52-90°C, D-values were larger in high-protein matrices than in high-fat matrices (110). However, at aw 0.50 and > 77.4°C, the D-value was larger in high-fat matrices than in high-protein matrices (110). This shows product composition is another critical factor in understanding thermal resistance of pathogens in low aw foods. In most of the aforementioned studies, the effect of aw was evaluated under isothermal heating. However, isothermal conditions are rarely consistent with commercial food processes, in which the temperature and moisture of the product may dynamically change over time. Additionally, the effect of aw on pathogen inactivation was often investigated only at discrete aw values and did not include changes in moisture during the entire process. 11 Additionally, most food products used in prior studies were already low-moisture products and did not account for raw products with high moisture content being processed to a low- moisture end-product. Casulli et al. (14) showed that increasing the initial moisture content of pistachios affected Salmonella resistance. In this study, by soaking pistachios in pure water or in 27% NaCl solution to a moisture content of up to 21.3 % (dry basis (db)), thermal treatment of the pistachios achieved a 4-log reduction 55 to 85% faster than when the pistachios contained 6% moisture (db) (14). This shows that initial product moisture content may need to be considered in pathogen inactivation. A subsequent part of that study also modeled Salmonella inactivation in pistachios under different temperature-humidity conditions and incorporated dynamic moisture and aw in their models (13). Although that study included dynamic moisture in inactivation models, applicability may be limited for products with much higher moisture content, such as apples. Therefore, there is a need to evaluate inactivation of pathogens under widely changing moisture conditions in food products with high initial moisture content. 2.4 Summary Although RTE bacon and dried apples are two very different products, the existing literature has shown that both products need process validations, given the food safety regulations established by USDA FSIS Appendix A and FSMA Preventive Controls Rules for Foods, respectively. For bacon, although microwave cooking can provide quality RTE bacon, FSIS has identified Salmonella lethality in bacon microwave cooking, where humidity is not controlled, as a scientific knowledge gap. For apples, recent L. monocytogenes outbreaks from caramel apples have demonstrated that this pathogen could be a potential concern in RTE apple products. However, current studies in apple drying have only focused on inactivation of Salmonella and E. coli. Thus, there is a need to assess L. monocytogenes reduction during apple 12 drying. Additionally, the effect of air velocity should be assessed, because it can affect both the drying and inactivation rate. This literature review also has shown that product moisture can affect the inactivation of pathogens, such as Salmonella and Listeria, in low-moisture foods. As water activity decreases, pathogens become more heat resistant. Prior studies also have indicated that other factors such as the food matrix composition may play a role in this effect. However, in most of these studies, isothermal and iso-moisture conditions were used, which may not realistically reflect most industrial processes, in which moisture changes dynamically. Therefore, there is a need to address this knowledge gap, of pathogen inactivation under widely changing moisture conditions, while assessing the thermal inactivation of Salmonella in commercial bacon processing and L. monocytogenes during apple drying, for purposes related to process validation, regulatory compliance, and an improved understanding of the dynamic moisture effects on pathogen thermal resistance. 13 CHAPTER 3: THERMAL INACTIVATION OF SALMONELLA DURING BACON PROCESSING 3.1 Materials and Methods 3.1.1 Overall Study Design This study consisted of two main treatments: microwave and impingement oven cooking. In each treatment, bacon slices were cooked to a commercial target of 40% yield, and Salmonella lethality was assessed relative to a 6.5 log reduction target. Because large sample-to-sample variability was observed in Salmonella inactivation during microwave cooking, an additional experiment was conducted to determine whether this variability might be attributable to variability in lean and fat content. Therefore, Salmonella inactivation was compared in the lean and fat portions of the bacon under microwave cooking. Additionally, with microwave cooking being the most common industrial method for the production of RTE bacon, and the knowledge gap mentioned in USDA FSIS Revised Appendix A (105) on whether humidity is important in this case, a separate experiment was conducted to determine whether humidity influenced Salmonella inactivation in bacon cooked in a microwave oven. 3.1.2 Sample Preparation Bacon slices 3.0 ± 0.5 mm thick (~ 30 g / slice) were prepared by the Michigan State University Meat Laboratory to replicate raw-smoked commercial bacon. Vacuum-packed bacon slices (0.5 kg per pack) were stored at -20°C and later thawed overnight at 4°C before each experiment. Slices with approximately the same lean and fat content by visual inspection and were selected to minimize variability of the treatments, especially in terms of microwave power distribution during cooking. 14 3.1.3 Culture Preparation and Inoculation Eight strains of Salmonella, identical to the original cultures used to inform USDA FSIS Appendix A (54), were used in this study: Salmonella Typhimurium DT104 H3380, S. Hadar MF60404, S. Copenhagen 109 8457, S. Enteritidis 108 H3527, S. Enteritidis 108 H3502, S. Thompson FSIS 120, S. Montevideo FSIS 051, and S. Heidelberg F5038BG1. The cultures were stored in tryptic soy broth (TSB; Becton Dickinson, Sparks, MD) containing 20% glycerol (v/v) at -80°C. Working cultures were obtained by transferring each strain to TSB, followed by 24 h of incubation at 37°C. One milliliter of each working culture was then spread separately on tryptic soy agar plates (TSA; Difco, BD) for confluent growth. After 24 h of incubation, each strain was harvested in 10 mL of 0.1% buffered peptone water (Difco, BD, Franklin Lakes, NJ), combined in one centrifuge bottle, pelletized (3000 x g for 15 min), and the pellet was resuspended in 100 ml of 0.1% peptone water to obtain the inoculum. Bacon slices were placed on a tray covered with aluminum foil. Each slice was inoculated by spreading 1 mL of inoculum on one side using a L-shaped spreader. After 20 min in a biosafety cabinet, the slices were flipped, and the other side was inoculated similarly. Inoculated slices were placed in sample bags and kept at 4°C until cooking (up to 2 h). 3.1.4 Microwave Oven Treatment A 2450 MHz microwave oven (LMC2075XX , LG Electronics, Denver, CO, USA) with a maximum power output of 1200 W was set to 80% power to produce fully-cooked products (yield < 40%) at ~120 s, which matched a typical commercial-scale microwave cooking time (51). Samples were cooked in batches of 2 slices, for 30, 45, 60, 90, 120, and 150 s, with 6 biological replications and 2 subsamples per replication, in which the 2 subsamples were the 2 15 slices in a cooking batch. Two subsamples were used to quantify spatial variability in heating among the slices. Prior to cooking, each slice was weighed, and 2 slices were placed on a microwave tray (Microware Nordic Ware, Minneapolis, MN, USA) as one batch. Before cooking, each inoculated slice was weighed, and then a 4-cm piece of each slice in a batch was pre-cut for aw and moisture content (MC) analysis later after cooking, and placed back in position with the rest of the full slice prior to cooking. After each time point during cooking, an infrared camera (TiR3FT Fluke, Everett, WA, USA) was used to immediately (< 30 s) capture a thermal image of each slice. The infrared camera was set up on a tripod next to the oven at a vertical distance of ~40 cm above the samples. Images were taken under room lighting 20-30 s after the samples were removed from the oven. Afterwards, the 4-cm pieces of the cooked slices were placed in sample bags for MC and aw analysis, and the remaining portion of each slice was immediately submerged in 50 mL of chilled 0.1% peptone water to stop further Salmonella inactivation. Three inoculated, uncooked slices per run were used as untreated controls. This specific microwave treatment is hereafter referred to as the “humid” condition, as humidity increased significantly inside the oven, due to water vapor leaving the product, and was not controlled or artificially reduced. In all treatments, cooking yield was measured as the mass of the bacon after cooking divided by the initial mass and multiplied by 100. 3.1.5 Impingement Oven Treatment Another set of samples was cooked in a pilot-scale impingement oven (JBT FoodTech, Sandusky, OH), at 60% humidity (v/v) (corresponding to a dew point of 87.4°C), 0.7 m/s exit jet velocity (corresponding to 20% fan speed) and dry bulb temperature of 177°C or 232°C for up to 600 s, with 6 biological replications and 2 slices per replication. The cooking parameters were 16 chosen to approximate commercial oven conditions. Samples were cooked for 60, 120, 180, and 600 s at 177°C, or 60, 90, 120, and 360 s at 232°C. The different cook times at the two temperatures were chosen to achieve similar endpoint cooking yields. Two inoculated bacon slices per batch were put on a metal rack that was conveyed into the oven, with the same procedure as described above for microwave cooking followed for aw and MC analyses. In terms of temperature measurement, one T-type temperature probe (32 gauge) was used to measure oven air temperature, and two T-type thermocouples (36 gauge) were each inserted just below the surface (1-2 mm depth) of the bacon – one into the lean portion, and the other into the fat portion. The thermocouples and probes were connected to a data logger covered by a thermal shield (Multipaq21, Fluke, Salem, NH), and temperature data were collected at a 5 s interval. Infrared images of the slices were also taken immediately (30-40 s) after the samples exited the oven, to compare the endpoint temperatures of the impingement-cooked and microwave-cooked samples. 3.1.6 Comparison of Salmonella Inactivation in the Lean and Fat Portions of Bacon Bacon slices were inoculated as described above. The lean and fat portions were then manually separated using a sterile scalpel, and weighed to obtain ~30 g of each. One lean and one fat portion were placed together on a microwave tray and cooked for 30, 45, 60, and 90 s under similar microwave conditions as the baseline microwave experiment. The cook times 120 and 150 s were excluded from the comparison, because most samples cooked at those times had Salmonella survival below the limit of detection (LOD) of 0.6 log CFU/g. After cooking, a 4-5 g portion of each lean and fat sample was used for aw and MC analysis, with the remaining portions immediately placed in 50 mL of chilled 0.1% peptone water. 17 3.1.7 Effect of Humidity on Salmonella Inactivation during Bacon Microwave Cooking To reduce the humidity during microwave cooking, dry air was forced through a 1-cm diameter hole in the side wall of the oven, above the tray where the bacon slices were placed, and diagonally opposite the magnetron where the microwaves were propagated. Oven air then exited the oven cavity through a vent at the farthest point from the inlet (top and back of the opposite side wall). The flow rate was set at ~0.2 L/s by volume displacement. Humidity of the inlet and exit air was measured using a dew point sensor (Vaisala DMT346, Finland). The inlet air had a humidity ratio of 2 g of water / kg of air, which was equivalent to a relative humidity of 1.2% at ~25.6°C and a dew point of -29.7°C. To measure oven air for humidity, a 1-cm diameter hole was made through the chamber top. A 60-cm siphon tube (0.75 cm i.d.) connected this hole to the inlet of a positive displacement pump (DOL-101-AA, Gast Manufacturing Inc, Harbor, MI), through which air was drawn in and then discharged through an outlet. Another siphon tube, 45-cm long, was used to move the discharged air from the pump outlet to a sealed box, referred to as the collector, containing the aforementioned dew point sensor. When tested, the sensor response time to a change in oven humidity was approximately 2-3 s. The air exiting the collector was passed through a column containing 75% ethanol to eliminate any Salmonella before being released to into the lab environment. The system is illustrated by the simplified diagram in Figure 1. To maintain the dew point below 32°C inside the oven during cooking, both dry air purging as described above and a reduction in sample size was needed, which simultaneously reduced the amount of water vapor being evaporated into the oven air during cooking. Therefore, each bacon slice was divided into 4 pieces weighing ~7.5 g each, with two bacon pieces used per batch. Bacon slices were inoculated using the same procedures as described earlier. Two full- 18 sized inoculated slices were used as in the main microwave cooking experiments; however, instead of cooking the two full slices at once, pairs of quarter-sized slices were cooked separately. After cooking, an infrared thermal image was taken to record the temperature. The first three cooked pieces from each slice were immediately combined and added to 50 mL of chilled 0.1% peptone solution, for subsequent Salmonella enumeration with the fourth piece used for MC and aw analysis. Figure 1: Diagram of the humidity control system during microwave cooking of bacon Preliminary cooking trials again were conducted to determine the appropriate microwave power setting (30%) to achieve the target cooked product yield (< 40%) at a commercially relevant cooking time. Therefore, the experiment was conducted at 30% power for 45, 60, and 150 s. Two main treatments were used: (1) a dry condition with dry air pushed through the microwave at 0.2 L/s during cooking, and (2) a less humid condition using the smaller sized sample but with no dry air forced through the system. In both treatments, the dew point was recorded every 10 s during cooking. Before each experiment, the ambient conditions (dry bulb temperature, relative humidity, and dew point) were measured. The dew point in the collector also was checked by measuring the air inside the oven before each the experiment, with the oven 19 off and closed, using a handheld dew point meter (EXTECH Instruments, HD500, Nashua, NH). On average, the dew point measured by the meter was only 0.7°C higher than that of the air sampling / collector system. 3.1.8 Salmonella Enumeration, Moisture Content, and Water Activity Analyses As mentioned above, when each bacon slice was divided into four equal pieces, three of the pieces were combined for Salmonella enumeration after cooking. The fourth piece, after undergoing each cooking treatment, was used for MC and aw analysis. After each treatment, samples were submerged in 50 mL pre-chilled 0.1% peptone solution, which corresponded to a dilution ratio of ~1:5 based on preliminary experiments. After homogenizing for 2 min (IUL Masticator Silver, 400 ml, IUL S.A., Barcelona, Spain), the samples were serially diluted in 0.1% peptone water and plated on a differential, non-selective medium composed of tryptic soy agar supplemented with 0.05% ammonium citrate, and 0.03% sodium thiosulfate (FTSA; Difco, BD), to enumerate both healthy and sublethally injured cells of Salmonella. All black colonies were counted as Salmonella after 48 h of incubation at 37°C. The limit of detection (LOD) was 0.6 log CFU/g. Duplicate analyses were conducted for aw (Aqualab 4TE, Pullman, WA, USA) and MC following AOAC method 950.46B (3). 3.1.9 Statistical Analyses Temperature data from the infrared images were analyzed over time and across the bacon surface area (SmartView 7.0 software, Fluke Corporation, Everett, WA, USA). The area of each bacon slice was outlined using the visible-wavelength image, and temperature (average, 20 minimum, and maximum) was analyzed within those defined areas using the corresponding infrared image. Assumptions for all data analyses, such as normality and equal variance, were verified (SAS 9.4, SAS Institute Inc., Cary, NC, USA), and ANOVA (α = 0.05) was used to test the effect of the studied conditions (microwave cooking across different time points, impingement oven cooking at two temperatures across different time points, bacon fat vs. lean across different time points, and “humid” vs. “less humid” vs. “dry” conditions) on MC, aw, yield, and Salmonella reduction. Tukey multiple means comparisons were conducted within each treatment across the corresponding cook times. T-tests (α = 0.05) were also used to compare Salmonella lethality to the 6.5 log reduction target (42). In terms of bacterial enumeration, samples that had plates with nondetectable counts were replaced with the limit of detection (0.6 log CFU/g) before proceeding to data analysis. All the statistical analyses were performed in SAS 9.4 using the function PROC MIXED. 3.2 Results and Discussion 3.2.1 Microwave Oven Cooking Moisture content, aw, and yield decreased significantly in the bacon slices during cooking (P < 0.05), as expected (Table 1). Using microwave cooking, the 40% target yield was achieved at 150 s, which was similar to typical commercial processing. A t-test comparison showed that the final yield of 37.3% at 150 s was similar to the yield after 600 s of impingement cooking at 177°C (P > 0.05). However, the MC of the microwave-cooked slices was higher (P < 0.05) than that of the impingement-cooked slices (Table 2) (P < 0.05), due to the forced-air convection drying. 21 The population of the Salmonella inoculum was ~10 log CFU/mL, and after inoculation, 9.1 ± 0.3 log CFU/g initial population was obtained in the bacon slices. Salmonella lethality in the bacon slices increased during microwave cooking (P < 0.05) as expected, with the largest variability at 60 s. By 90 s, Salmonella decreased > 6.5 logs (P < 0.05), at which time the yield was 54% (Table 1), and not yet “fully cooked.” The corresponding surface temperature was 89.6 ± 2.1°C (Figure 2). Table 1: Moisture content, water activity, cooking yield, and Salmonella reduction in bacon slices during microwave cooking Cook Moisture Water Activity Cooking Salmonella Time Content Yield Reductions (s) (% water) (%) (log CFU/g) 0 41.5 ± 7.7** a* 0.974 ± 0.006a - - 30 39.3 ± 9.0 ab 0.957 ± 0.028 b 85.9 ± 1.8 a 1.0 ± 0.2 d 45 35.5 ± 3.9 b 0.940 ± 0.015 c 72.9 ± 2.7 b 3.1 ± 1.1 c 60 36.1 ± 5.9 b 0.933 ± 0.018 c 64.6 ± 3.3 c 6.8 ± 1.7 b 90 32.0 ± 4.9 c 0.909 ± 0.029 d 54.1 ± 3.5 d 8.5 ± 0.6 a 120 25.9 ± 4.1 d 0.829 ± 0.054 e 44.5 ± 3.1 e 8.4 ± 0.6 a 150 18.9 ± 3.9 e 0.758 ± 0.079 f 37.3 ± 3.7 f 8.6 ± 0.1*** *Means followed by the same letter within a column are not significantly different ( = 0.05) ** Mean ± standard deviation *** ≥ 95% of plated samples were below the limit of detection (< 0.6 log CFU/g), therefore the ‘150 s’ time point was excluded from the multiple comparison By the time the microwaved bacon slices were fully cooked (150 s), the temperature was approximately 100°C, as recorded by the infrared camera (Figure 3). This temperature was similar to the temperature of the impingement oven cooked slices taken from the infrared images at the end of cooking (100.0 ± 5.6 °C after 600 s at 177°C cooking temperature and 105.4 ± 2.4 °C after 360 s at 232°C) (Figures 3B and 4). Additionally, heating across the two slices per batch was reasonably homogenous during both microwave and impingement cooking (Figure 3). 22 Impingement oven cooking (177°C, 0.7 m/s, 60% v/v, dew point 87.4°C) 160 aw 1.0 10.0 Salmonella (log CFU/g) MC (%, wb) 120 0.8 8.0 TSurf 0.6 6.0 80 aw Temperature (°C) 0.4 4.0 MC 40 0.2 2.0 0 0.0 0.0 0 150 300 450 600 0 150 300 450 600 Cook Time (s) Cook Time (s) Impingement oven cooking (232°C, 0.7 m/s, 60% v/v, dew point 87.4°C) Salmonella (log CFU/g) 160 aw 1 10.0 MC (%, wb) 120 0.8 8.0 TSurf 0.6 aw 6.0 80 0.4 4.0 Temperature (°C) 40 MC 0.2 2.0 0 0 0.0 0 150 300 450 600 0 150 300 450 600 Cook Time (s) Cook Time (s) Microwave oven cooking (“humid” condition, dew point > 35°C) aw 1.0 10 Salmonella (log CFU/g) 100 MC (%, wb) 0.8 8 80 0.6 6 60 TSurf aw Temperature (°C) 40 0.4 4 20 MC 0.2 2 0 0.0 0 0 30 60 90 120 150 0 30 60 90 120 150 Cook Time (s) Cook Time (s) Figure 2: Water activity, moisture content, surface temperature, and Salmonella populations during impingement and microwave cooking of bacon. The open circles in the Salmonella population curves indicate that one or more samples were below the limit of detection (0.6 log CFU/g). 23 Figure 3A: Infrared images of microwave-cooked bacon slices 30 s 45 s 60 s 90 s 120 s 150 s Figure 3B: Infrared images of bacon slices cooked in the impingement oven 177°C, 60 s 177°C, 120 s 177°C, 180 s 177°C, 600 s 232°C, 60 s 232°C, 90 s 232°C, 120 s 232°C, 360 s Figure 3: Infrared images of bacon slices A) during microwave cooking and B) during impingement oven cooking 24 3.2.2 Impingement Oven Cooking The time to achieve the target cooking yield (< 40%) was, as expected, less at the higher oven temperature (Table 2); however, at similar yields (360 s for 232°C; 600 s for 177°C), the MC and aw were lower for the higher oven temperature. With aw < 0.6, these fully-cooked products met the characteristics of a low-moisture food (40). Overall, yield variability and aw increased over time at both temperatures (Table 2), which could be due to rendering of fat and water evaporation over time. Table 2: Moisture content, water activity, yield, and Salmonella reduction in bacon slices during impingement cooking Cook Cook Moisture Content Water Activity Cooking Salmonella Temp Time Yield Reductions (ᵒC) (s) (% water) (%) (log CFU/g) Control 0 40.8 ± 6.8** a, A* 0.971 ± 0.004 a, A - - 60 36.3 ± 7.3 ab 0.958 ± 0.009 b 84.9 ± 1.3 a > 8.4 (11/12) *** 120 35.4 ± 7.4 b 0.950 ± 0.008 c 77.1 ± 1.6 b > 8.4 (12/12) 177 180 35.4 ± 5.1 b 0.940 ± 0.015 c 70.3 ± 2.1 c > 8.4 (12/12) 600 10.2 ± 4.1 c 0.573 ± 0.152 d 35.6 ± 3.7 d > 8.4 (11/12) 60 37.0 ± 7.3 AB 0.956 ± 0.009 B 80.5 ± 1.8 A > 8.4 (11/12) 90 36.3 ± 5.5 AB 0.944 ± 0.012 C 74.4 ± 2.3 B > 8.3 (11/12) 232 120 35.4 ± 4.9 B 0.932 ± 0.015 C 67.4 ± 3.4 C > 8.3 (11/12) 360 5.5 ± 4.4 C 0.391 ± 0.191 D 32.7 ± 4.9 D > 8.4 (11/12) *Means followed by the same lowercase or uppercase letter within a column for a given cook temperature are not significantly different ( = 0.05) ** Mean ± standard deviation ***(11/12) means 11 of 12 plated samples were below the limit of detection of 0.6 log CFU/g 25 By 60 s, Salmonella decreased > 6.5 logs (P < 0.05) at both cooking temperatures. Such fast inactivation can be explained by the rapid condensation of water vapor on the bacon surface at the beginning of cooking until the bacon surface temperature exceeded the dew point temperature (87.4°C), which significantly enhances surface lethality (49). When compared to the temperature profile from infrared images (Figure 4), the continuous temperature profile obtained using thermocouples (Figure 2) was higher. This difference was due to the 20-30 s time lag necessary to transport the samples out of the oven before the infrared images could be taken. Additionally, the thermal images yielded spatial mean surface temperatures, compared to single point measurements using the thermocouples. Overall, these results demonstrated that under the studied conditions in an impingement oven, the 6.5 log Salmonella lethality performance standard (42) was met before the bacon slices were fully cooked to a commercial yield of < 40%. 180 150 Temperature (ᵒC) 120 90 60 177°C cooking temperature 30 232°C cooking temperature 0 0 100 200 300 400 500 600 Cook time (s) Figure 4: Surface temperature of bacon cooked in an impingement oven, measured using an infrared camera (~20-30 s after cooking as the samples were removed from the oven). 26 3.2.3 Comparison of Salmonella Inactivation in the Lean and Fat Portions of Bacon The fat portion of the bacon had a lower aw and MC (P < 0.05); the final MC at 90 s was 13% and 38% for fat and lean, respectively (Table 3). The initial Salmonella population was higher in the lean than in the fat (P < 0.05), indicating greater uptake in the lean, potentially because of the fibrous characteristics of the lean and the generally hydrophobic characteristic of the fat tissue. The temperature of the fat was consistently higher than the lean from 45 to 90 s (Figure 5). However, lower Salmonella reductions were observed in bacon fat (P < 0.05), even though the statistical interaction of fat vs. lean with cook time was not significant (P > 0.05). Similar observations have been reported in the literature, in which increased fat content was associated with higher bacterial pathogen survival (1, 53). Possible explanations of such increased survival of bacterial pathogens in fat include the lower aw of fat and the commonly reported protective effect of fat on bacteria, potentially due to some underlying physiochemical factors (48). Despite the difference in inactivation, Salmonella decreased > 6.5 logs in both bacon fat and bacon lean by 90 s, which was consistent with the results for microwave cooking of whole bacon slice (Tables 1 and 3). Therefore, the large variation observed for bacon slices after 60 s microwave cooking (Table 1, Figure 2) may be due to the variation in fat and lean content among the slices. 27 Microwave, Bacon Fat 1.0 MC (%, wb) 100 10 aw Salmonella 80 0.8 8 60 TSurf 0.6 6 0.4 aw (log CFU/g) 4 Temperature (°C) 40 20 MC 0.2 2 0 0.0 0 0 30 60 90 120 150 0 30 60 90 120 150 Cook Time (s) Cook Time (s) Microwave, Bacon Lean aw MC (%, wb) 1.0 10 100 TSurf 0.8 8 Salmonella 80 60 0.6 6 aw 4 0.4 Temperature (°C) 40 MC (log CFU/g) 20 0.2 2 0 0.0 0 0 30 60 90 120 150 0 30 60 90 120 150 Cook Time (s) Cook Time (s) Microwave, "Dry" Condition (dew point < 20°C) aw MC (%, wb) 1.0 10 100 Salmonella 80 0.8 8 TSurf 60 0.6 6 aw 40 0.4 4 (log CFU/g) Temperature (°C) 20 MC 0.2 2 0 0.0 0 0 30 60 90 120 150 0 30 60 90 120 150 Cook Time (s) Cook Time (s) Microwave, "Less Humid" Condition (dew point ≤ 25°C) aw MC (%, wb) 1.0 10 100 80 TSurf 0.8 8 60 0.6 6 Salmonella 0.4 aw Temperature (°C) 40 4 (log CFU/g) 20 MC 0.2 2 0 0.0 0 0 30 60 90 120 150 0 30 60 90 120 150 Cook Time (s) Cook Time (s) Figure 5: Water activity, moisture content, surface temperature, and Salmonella populations in bacon fat and lean portions of bacon and the effect of humidity during microwave cooking of bacon. Less humid conditions represent the scenario where no dry air was added into the system when cooking 2 bacon pieces per batch. Humid condition represents the condition of regular microwave cooking of bacon in which humidity was not controlled and 2 whole bacon slices were cooked per batch. Open circles in the Salmonella population curves indicate that one or more samples were below the limit of detection (0.6 log CFU/g). 28 Table 3: Moisture content, water activity, yield, and Salmonella reduction in bacon fat and bacon lean during microwave cooking Cook Moisture Content Water Activity Cooking Salmonella Time Yield Reductions (s) (% water) (%) (log CFU/g) 0 14.3 ± 3.8**e* 0.952 ± 0.014 b - - 30 13.4 ± 3.1 e 0.931 ± 0.011 c 85.4 ± 1.7 b 0.4 ± 0.2 e Bacon 45 13.1 ± 3.2 e 0.917 ± 0.013 c 68.2 ± 3.0 d 1.5 ± 0.3 d fat 60 11.7 ± 2.0 e 0.871 ± 0.026 d 49.7 ± 5.6 f 4.1 ± 2.0 bc 90 13.5 ± 3.9 e 0.791 ± 0.063 e 33.0 ± 2.5 g 7.2 ± 1.6 a 0 59.0 ± 2.6 a 0.971 ± 0.005 a - - 30 54.0 ± 3.2 b 0.960 ± 0.006 ab 89.5 ± 0.9 a 0.9 ± 0.5 ed Bacon 45 50.9 ± 2.7 bc 0.954 ± 0.008 b 83.1 ± 1.5 b 2.8 ± 1.1 c lean 60 47.7 ± 2.6 c 0.945 ± 0.021 bc 76.2 ± 1.3 c 6.2 ± 2.3 ab 90 38.4 ± 3.9 d 0.906 ± 0.018 dc 64.4 ± 2.8 e 8.2 ± 0.7 a *Means followed by the same letter within a column are not significantly different ( = 0.05) ** Mean ± standard deviation 3.2.4 Effect of Humidity on Salmonella Inactivation in Microwave Cooked Bacon On the days of experiments, the ambient environmental conditions in the lab were measured, in which the dew point was 12.2 ± 1.2°C, corresponding to a relative humidity of 48.3 ± 3.3% at a dry bulb temperature of 24.9 ± 0.4°C. Before cooking, the dew point inside the microwave oven was -0.8 ± 0.3°C and 10 ± 1.5°C for the “dry” and “less humid” conditions, respectively. No significant difference in MC, aw, or yield (P > 0.05) was observed between the two main treatments – “dry” and “less humid” (Table 4). Within each treatment, MC and yield decreased with cook time (P < 0.05), as expected. The 40% target yield was achieved by 150 s. The surface temperatures for the “dry” and “less humid” conditions were similar (Figure 5). However, the final temperature at 150 s in both conditions were lower than in the “humid condition” (Figure 2), in which the surface temperature exceeded 100°C at the end of cooking. 29 Table 4: Moisture content, water activity, yield, and Salmonella reduction during microwave cooking under a dry and less humid conditions Cook Moisture Water Activity Cooking Salmonella Time Content Yield Reductions (s) (% water) (%) (log CFU/g) “Dry” 0 51.6 ± 3.7 0.970 ± 0.007 - - condition 45 42.7 ± 6.6** a* 0.958 ± 0.007 a 75.8 ± 2.2 a 1.4 ± 0.4 d (dew point 60 39.7 ± 5.0 a 0.947 ± 0.012 a 68.8 ± 2.1 b 2.2 ± 0.3 bc < 20°C) 150 20.7 ± 3.8 b 0.780 ± 0.064 a 37.5 ± 2.6 c 4.7 ± 0.7 a “Less 0 51.6 ± 3.7 0.970 ± 0.007 - - humid” 45 44.0 ± 5.4 a 0.960 ± 0.007 a 75.2 ± 1.6 a 1.8 ± 0.4 cd condition 60 42.0 ± 4.7 a 0.952 ± 0.007 a 67.8 ± 2.4 b 2.4 ± 0.7 b (dew point 150 19.4 ± 2.3 b 0.771 ± 0.056 a 35.9 ± 3.0 c 5.1 ± 0.7 a ≤ 25°C) *Means followed by the same letter within a column are not significantly different ( = 0.05) ** Mean ± standard deviation The “dry” and “less humid” conditions were not significantly different from each other in terms of Salmonella inactivation (P > 0.05) (Table 4). However, when compared to the baseline microwave cooking of whole bacon slices (i.e., the “humid” condition, dew point > 35°C) (Table 1), both the “dry” and “less humid” conditions resulted in significantly lower Salmonella reductions (P < 0.05). In the “humid” condition, Salmonella decreased > 6.5 logs by 90 s (Table 1). However, only 4.7 and 5.1 log Salmonella reductions were achieved under the “dry” and “less humid” conditions, respectively, at 150 s (Table 4). Thus, humidity affected Salmonella lethality during microwave cooking of bacon. Similar observations were made in previous studies on the effect of humidity on Salmonella inactivation during thermal processing of several foods, including meat products (14, 49, 52, 69). Hildebrandt et al. (49) reported that an oven humidity ≥ 30 % (v/v) significantly increased Salmonella lethality on the surface of various chicken and beef products. Similar results were observed for sesame seeds (69), almonds (52), and pistachios (14), in which increased humidity was associated with higher Salmonella inactivation. For the case of bacon in 30 this study, even though the dew point under the dry condition was lower than that of the less humid condition (Figure 6), the difference was not large enough to detect a significant difference in Salmonella inactivation. The oven dew point was highest when cooking two whole slices at a time (Figure 6, the “humid” condition), in which 6.5 log reductions of Salmonella were achieved before the 40% target yield was achieved. When cooking two full slices (the “humid” condition) the measured dew point was likely lower than the true process value because of the potential introduction of ambient air into the oven when the dew point meter was placed inside; however, the method nevertheless provided a consistent comparison across treatments. Overall, this supplemental experiment showed that humidity did affect Salmonella lethality during microwave cooking of bacon, and future studies should investigate the critical humidity threshold, and the humidity in typical commercial-scale, continuous microwave oven systems, to ensure compliance with USDA FSIS Salmonella lethality requirements. 35 25 Dew point (°C) 15 "Humid" condition "Less humid" condition 5 "Dry" condition -5 0 15 30 45 60 75 90 105 120 135 150 Cook time (s) Figure 6: Dew point during microwave cooking of bacon. Dry condition represents the treatment in which dry air was pushed into the microwave during cooking of 2 bacon pieces. Less humid conditions represent the scenario where no dry air was added into the system when cooking 2 bacon pieces per batch. Humid condition represents the condition of regular microwave cooking of bacon in which humidity was not controlled and 2 whole bacon slices were cooked per batch. 31 3.3 Conclusions In summary, one minute of impingement oven cooking at both the low and high commercial temperatures at a humidity of 60% (v/v) met the USDA-FSIS Salmonella lethality requirement long before the product reached the required yield to be labeled as fully cooked. For microwave cooking, Salmonella inactivation and compliance with USDA FSIS depended on oven humidity. Cooking whole bacon slices at a dew point > 35°C inside the oven achieved > 6.5 log reduction of Salmonella by 90 s, again prior to reaching the required endpoint yield for fully cooked product. However, reducing the humidity in the oven to a maximum dew point of 19°C or 25°C yielded results that did not ensure sufficient Salmonella lethality to meet the target 6.5 log reduction. Therefore, future studies should investigate the humidity threshold needed to meet required food safety standards in bacon processing. 32 CHAPTER 4: LISTERIA MONOCYTOGENES INACTIVATION DURING APPLE DRYING UNDER LOW OR MODERATE TEMPERATURE AND AIR VELOCITY 4.1 Materials and Methods 4.1.1 Study Design To investigate Listeria monocytogenes inactivation during apple drying under conditions similar to industrial processes, this study included a full-factorial experiment (3 replicates) with multiple levels of temperature, air velocity, and drying time (detailed below). In each treatment, the apple slices were dried to a moisture content ≤ 24% (wb), as required by USDA standards defining dried apples (82, 84). All the drying treatments were conducted in a pilot-scale industrial oven (JBT, Sandusky, OH, USA), in which apple surface and oven temperatures were measured over time. Colorimetric analysis was conducted to evaluate the effect of the studied conditions on the browning of the apples (61, 62, 75), which can vary with polyphenol oxidase (PPO) concentration, temperature, and oxygen availability (6). In addition to its effect on marketability of the product, browning can negatively impact the nutritional value (65). 4.1.2 Sample Preparation Organic Gala apples (Rainier Fruit Co, Selah, WA) were acquired from a local retail store and stored at 4°C (up to 21 days) prior to the drying experiments. The apples were labeled “extra fancy,” implying various physical attributes, such as the absence of both internal and external damage based on USDA Agricultural Marketing Service standards (81, 83). After visual inspection for firmness and damage, the unpeeled apples were cored and sliced into 6-mm thick rings similarly to Burnham et al. (9), using a manual corer (19 mm diameter, Cuisipro, Markham, 33 ON, Canada) and a mechanical slicer (55200AN, Nemco, Hicksville, OH, USA). Apple rings with an average mass of 20 g and diameter of ~6 cm were used for the experiments. For each experiment, four slices were used as controls to evaluate the native microflora of the apple slices. Each slice was weighed, diluted in 0.1% phosphate-buffered saline (PBS) (J.T. Baker, Center Valley, PA, USA) at a ratio of 1:5, and homogenized for 2 min (IUL Masticator Silver, 400 ml, IUL S.A., Barcelona, Spain). Appropriate serial dilutions in 0.1% PBS of two of the four slices were plated on a nonselective differential medium Tryptic Soy Agar (TSA) (Difco, Sparks, MD, USA) supplemented with 0.6% yeast extract (Bacto, BD, Sparks, MD, USA), 0.05% ammonium iron citrate (Sigma-Aldrich, St. Louis, MO, USA), and 0.025% esculin hydrate (Across Organics, Morris, NJ, USA) (ETSA). After 48 h of incubation at 37°C, all growing colonies were counted to evaluate the overall background microflora (aerobic mesophiles), and any black colonies would have been counted as presumed Listeria. To further confirm the presence or absence of Listeria in the apple slices before inoculation, the other two slices were plated on Oxford medium base (Difco, BD, Sparks, MD, USA) supplemented with Modified Oxford Antimicrobic Supplement (Difco, BD, Sparks, MD, USA) (MOX). After 48 h of incubation at 37°C, black colonies surrounded by black zone were counted as Listeria. 4.1.3 Culture preparation and sample inoculation Eight strains of L. monocytogenes previously implicated in outbreaks were obtained from Dr. Sophia Kathariou at North Carolina State University: 4b1-GFP (clinical isolate, 1962), CFSAN048782-6 (apples, 2017), CFSAN023957-A10 (mung bean sprouts outbreak, 2014), 2014L-6695-5 (caramel apple outbreak, 2014-2015), 2014L-6680-7 (caramel apple outbreak, 2014-2015), F2365-2 (California cheese outbreak, 1985), 2010L-172304 (celery outbreak, 2010), 34 and H7858-1 (hot dog outbreak, 1998-1999). The methods of Sloniker et al. (71) were used to prepare the Listeria cultures and to obtain the inoculum as follows. All strains were stored at −80°C in tryptic soy broth (TSB; Difco, BD, Sparks, MD, USA) containing 20% glycerol (v/v). Working cultures were obtained by transferring each strain into 100 mL TSB, followed by 24 h incubation at 37°C. Each strain was harvested from TSB broth through centrifugation (3000 x g, 15 min). The pellets from two strains were then combined in one centrifuge bottle, and the mixture was diluted with 100 mL 0.1% PBS solution. After a gentle mixing using a sterile spatula, this mixture of two strains per centrifuge bottle was centrifuged for another 15 min. After centrifugation, all the pellets of all the eight-strains were combined and diluted into 30 mL 0.1% PBS solution to obtain an inoculum with a mean population of 10.0 ± 0.1 log CFU/mL. Apple slices were placed on a sterile metal rack and spread with 0.5 mL of inoculum using a T-shaped spreader. After sitting 15 min in a biosafety cabinet, the slices were flipped, and 0.5 mL of inoculum was spread on the other side of each slice. After another 15 min, the inoculated slices were individually placed in sterile sample bags and kept at 4°C (≤ 1 h) until the start of oven drying. Preliminary work showed that post-inoculated samples were only 0.25 ± 0.01 g heavier than pre-inoculated samples (1.41% mass change), indicating a relatively negligible net moisture addition from the inoculum. 4.1.4 Apple Drying The apple slices were dried at 60 or 80°C dry bulb, and an impingement air velocity of 0.7 or 2.1 m/s (corresponding to a 20 and 60% fan setting on the oven) (JBT, Sandusky, OH, USA). The air flow was perpendicular to the surface of apple slices (from an array of round jet nozzles) with a turbulent regime. The total drying time at 60°C was 180 min (sampling at 30, 60, 35 90, 120, 150, and 180 min), and at 80°C was 150 min (sampling at 30, 60, 90, 120, and 150 min). Different total drying times were used for each temperature to achieve < 24% (wb) moisture content at the end of each treatment. Apples slices were dried in batches of 10 or 12, at 80 and 60°C, respectively. The slices were placed in a single layer on a metallic mesh tray, and arranged in two columns, adjacent to each other, following the position of the air jets inside the oven, with each column containing 5 or 6 apple slices. To better mimic a commercial process with a full load on a dryer belt, ~240 g of additional, non-inoculated apple slices were placed on either side of the two columns of inoculated slices during each drying test. To measure the temperature of the inoculated apple slices during drying, two T-type thermocouples (36 gauge) were inserted just below the surface of two slices that were situated diagonally across from each other on the belt. The experiment was conducted such that these two slices with the thermocouples were the last to be sampled in each drying trial, in order to record the surface temperature during the entire drying process. The thermocouples were connected to a data acquisition system (Multipaq21, Fluke, Salem, NH, USA), which transferred the temperature data to a computer via radio frequency telemetry, at an interval of 5 s. Two thermocouples were used to capture the variability across the slices. A T-type temperature probe (32 gauge) was also attached to the conveyor belt with the thermocouple junction measuring the oven air dry bulb temperature near the samples on the belt over time. At each 30-min sampling point, the tray was briefly removed from of the oven (~1 min), during which one apple slice was randomly taken from each of the two columns, prior to immediately putting the tray back into the oven. One slice was immediately placed in a sample bag, cooled in an ice bath, and used for Listeria enumeration. The other slice was used for colorimetry measurement followed by aw and MC analyses as described below. 36 The colorimetry analysis was performed on one randomly chosen side of the slice using a handheld colorimeter (CR-400, Konica Minolta, Ramsey, NJ, USA). L*a*b* color values were recorded at three different random locations across the slice, in which L* indicated the lightness of the slice, a* the redness or greenness, and b* the yellowness or blueness. The L*a*b* color values of the slice were measured at room temperature in a biosafety hood. Immediately after the measurement, which took approximately 1 to 2 min, the slice was put in a sealed sample bag. The browning index (BI) (61, 62, 75) was computed based on the measured color values and then used to compare the impact of the different drying conditions over time. The same slice that was used for color analysis also was used for aw measurement. The maximum time lag between removal of the sample from the oven and aw analysis was 3-4 h, during which all samples were kept in sealed sample bags at 4°C, except during the colorimetry analysis. To prepare samples for aw measurement, apple slices were cut into small pieces (≤ 50 mm2), then placed in the aw cup (~3-5 g). Water activity was measured twice for each sample at an ambient temperature of ~25°C (Aqualab 4TE, Pullman, WA, USA), and the average of the two values was reported. After aw analysis, AOAC method 950.46B (3) was used to measure the MC of the slice gravimetrically after drying at 100-102°C for ~16 h. 4.1.5 Listeria Enumeration One whole slice from each drying time was diluted 1:6 in 0.1% PBS solution. Diluted samples were then homogenized for 2 min (IUL Masticator Silver, 400 ml, IUL S.A., Barcelona, Spain), serially diluted in 0.1% PBS solution and plated on ETSA. Listeria colonies (black with black halos) were enumerated after 48 h incubation at 37°C. 37 4.1.6 Statistical Analyses After checking for normality and equal variance assumptions using graphical and statistical methods, such as Levene’s test, the effect of temperature and air velocity over time on Listeria inactivation, MC, and aw was evaluated using ANOVA (α = 0.05) (SAS 9.4, SAS Institute Inc., Cary, NC, USA). In order to have the same number of levels for the ‘time’ factor, the 180 min time point for 60°C drying temperature was removed for the ANOVA and mean comparisons. Multiple pair-wise comparisons of the means of the responses over the drying times were then conducted using Tukey’s test (α = 0.05). 4.2 Results and Discussion 4.2.1 Moisture Content and Water Activity Starting at a MC of approximately 85% (wb), the apple slices met the standard of ≤ 24% (wb) MC for dried apples by the end of the studied drying time (Figure 7, Table 5). When dried at 60°C, the target 24% (wb) moisture was achieved after 150 min at an air velocity of 0.7 m/s, compared to 120-150 min at 2.1 m/s. Similarly at 80°C, 24% MC was achieved by 90 min at 0.7 m/s and between 60-90 min at 2.1 m/s. 38 aw 60°C, 0.7 m/s 60°C, 0.7 m/s Listeria (log CFU/g) Temperature (°C) 100 1.0 10 Water activity 80 0.8 8 MC Moisture content (% 60 0.6 6 40 0.4 4 wb) 20 TSurf 0.2 2 0 0.0 0 0 30 60 90 120 150 180 0 30 60 90 120 150 180 Drying time (min) Drying time (min) 100 aw 1 60°C, 2.1 m/s 60°C, 2.1 m/s Temperature (°C) Listeria (log CFU/g) 10 Water activity 80 MC 0.8 8 60 0.6 Moisture content (% 6 40 0.4 TSurf 4 20 0.2 wb) 2 0 0 0 0 30 60 90 120 150 180 Drying time (min) 0 30 60 90 120 150 180 Drying time (min) 100 aw 80°C, 0.7 m/s 1 80°C, 0.7 m/s Listeria (log CFU/g) Temperature (°C) 10 MC Water activity 80 0.8 8 60 0.6 6 40 0.4 4 TSurf Moisture content (% wb) 20 0.2 2 0 0 0 0 30 60 90 120 150 180 0 30 60 90 120 150 180 Drying time (min) Drying time (min) 100 aw 80°C, 2.1 m/s 1 80°C, 2.1 m/s Temperature (°C) Listeria (log CFU/g) MC 10 80 0.8 Water acivity 8 60 0.6 6 40 0.4 TSurf 4 Moisture content (% wb) 20 0.2 2 0 0 0 0 30 60 90 120 150 180 0 30 60 90 120 150 180 Drying time (min) Drying time (min) Figure 7: Moisture content, water activity, surface temperature, and Listeria populations during apple drying. 39 Table 5: Multiple comparisons of the moisture content, water activity, surface temperature, and Listeria population during apple drying across the studied conditions. Temp Air Drying Moisture Water Listeria Velocity Time Content Activity Reductions (°C) (m/s) (min) (% water) (log CFU/g) 0 85.78 ± 0.74** a* 0.990 ± 0.006** a* - 30 79.82 ± 2.14 ab 0.976 ± 0.001 a -0.1 ± 0.3 g 60 72.40 ± 1.82 bcd 0.966 ± 0.005 a 0.4 ± 0.2 fg 0.7 90 56.26 ± 4.74 e 0.922 ± 0.018 a 0.5 ± 0.5 fg 120 37.51 ± 9.84 fg 0.775 ± 0.105 cd 0.6 ± 0.3 f 150 24.29 ± 1.53 ih 0.598 ± 0.067 e 1.4 ± 0.5 cde 180 17.50 ± 3.73 0.473 ± 0.086 1.8 ± 0.3 60 0 85.42 ± 0.37 a 0.989 ± 0.003 a - 30 71.92 ± 6.07 cd 0.962 ± 0.010 a 0.4 ± 0.4 fg 60 67.50 ± 6.45 d 0.927 ± 0.054 a 0.4 ± 0.3 fg 2.1 90 42.74 ± 4.64 f 0.832 ± 0.044 bc 0.8 ± 0.4 ef 120 30.42 ± 2.86 gh 0.700 ± 0.009 d 1.3 ± 0.1 cdef 150 19.34 ± 4.15 ij 0.473 ± 0.055 f 2.3 ± 1.1 c 180 14.40 ± 4.53 0.360 ± 0.021 2.8 ± 0.7 0 85.25 ± 2.80 a 0.991 ± 0.004 a - 30 75.28 ± 1.78 bc 0.968 ± 0.010 a 0.6 ± 0.2 f 60 55.55 ± 3.74 e 0.911 ± 0.014 ab 0.9 ± 0.3 ef 0.7 90 23.24 ± 4.31 ih 0.605 ± 0.087 e 1.7 ± 0.3 cd 120 13.96 ± 3.48 kj 0.316 ± 0.093 g 4.4 ± 0.2 a 150 10.97 ± 2.77 k 0.224 ± 0.025 h 5.2 ± 0.5 a 80 0 84.20 ± 0.87 a 0.991 ± 0.001 a - 30 69.17 ± 1.33 cd 0.954 ± 0.007 a 0.9 ± 0.5 def 60 31.78 ± 11.87 g 0.705 ± 0.141 d 1.6 ± 0.7 cde 2.1 90 14.43 ± 3.25 kj 0.282 ± 0.023 gh 3.6 ± 0.8 b 120 13.44 ± 3.61 kj 0.249 ± 0.027 gh 4.9 ± 0.8 a 150 11.81 ± 2.80 k 0.208 ± 0.062 h 5.2 ± 0.5 a *Means followed by the same letter within each entire column are not significantly different (Tukey’s LSD,  = 0.05) ** Mean ± standard deviation The higher fan speed and drying temperature resulted in faster moisture removal (P < 0.05), as expected, because of increased mass convection and heat transfer. For example, at either drying temperature, the apple surface temperature rose to within ~1-2°C of the air temperature almost 30 min sooner at the higher air velocity (Figure 7). Similarly, the change in apple temperature was faster at 80°C (Figure 7). The brief dips in the temperature profiles 40 (Figure 7) were due to the short period during which the samples tray was removed from the oven for sample collection. Overall, the mean absolute difference (± standard deviation) between the temperatures measured by the two thermocouples on two concurrently dried slices was 1.48 ± 1.80°C, indicating fairly low surface temperature variability during drying. The measured air temperature was, on average, 0.33 ± 3.33°C below the set temperature across all treatments, indicating adequate control of the process temperature. Consistent results also were observed for the change in moisture over time, with the higher drying temperature associated with a faster decrease in moisture beginning as soon as 30 min into the drying period. Air velocity only significantly affected the MC at 30 and 90 min at 60°C and at 60 min at 80°C (P < 0.05), with the lower air velocity associated with slower change in moisture (Figure 7). Previous studies on the drying kinetics of fruits and vegetables confirmed that higher drying temperatures and air velocities result in faster drying rates and moisture diffusion (57, 79, 106), as would be expected. However, if the temperature and air velocity are too high, shrinkage and/or “case hardening” can occur, preventing moisture loss from the food matrix (57, 79). In this study, by visual inspection, shrinkage did not occur until the later phase of drying, after 90 min, depending on the drying temperature and air velocity. In this study, temperature had a greater effect on moisture removal during drying than did air velocity, as shown by the difference in the slopes of the moisture plots. Similar findings were reported in a previous study (57) that examined the drying kinetics of various vegetables. These researchers attributed the lower effect of air velocities to the range of air velocity used, which was considered relatively high (1.5-2.6 m/s) (57). In contrast, our air velocity of 0.7 m/s was much lower, with no apparent shrinkage until later in the drying. Therefore, the lower effect of air velocity on MC may be due to lower heat and mass convection. 41 In terms of aw, the apples slices started at 0.99, and, depending on the drying conditions used, aw only decreased significantly during the later stage of drying (Table 5). At 60°C and 0.7 m/s, aw only changed after 120 min of drying, whereas, at 80°C and 2.1 m/s, a similar change occurred after 60 min (P < 0.05). 4.2.2 Colorimetric Results Apple browning during drying was affected by the time-temperature-air velocity interaction (P < 0.05). Some of the slices had a browning index significantly greater from the raw inoculated slices (P < 0.05). The average browning index of the raw inoculated apple slices was 29.88 ± 3.33, whereas that of the dried apples was 36.95 ± 5.39; however, mean browning was not significantly different across the drying treatments (P > 0.05). 4.2.3 Listeria Inactivation Gala apples were chosen for this study, as this specific variety had been widely used in previous apple drying studies (9, 38, 39, 45). Gurtler et al. (47) reported that survival of Salmonella in Gala apples during drying was higher compared to other varieties, such as Granny Smith, Pink Lady, and Fuji. The initial Listeria population after inoculation was 8.6 ± 0.3 log CFU/g across all the treatments. Among the negative controls that were plated on ETSA, all the samples, except 2, had total counts below the limit of detection (LOD) of 1.7 log CFU/g. Those two samples had an average total plate count of 2.7 log CFU/g, but none of the colonies resembled Listeria. All the negative controls had counts below the LOD on MOX. Temperature and air velocity significantly impacted the inactivation of L. monocytogenes during drying (P < 0.05), with the higher temperature and air velocity resulting in greater 42 reductions. At 60°C and air velocity of 0.7 m/s, the initial Listeria population decreased only 0.6 ± 0.3 log at 120 min, in which the MC was 38% (wb). At that same drying time and temperature, increasing the air velocity to 2.1 m/s decreased Listeria by 1.3 ± 0.1 log, at a MC of 30% (wb). After 180 min of drying, the overall Listeria reduction was 1.8 ± 0.3 and 2.8 ± 0.7 log CFU/g at 0.7 and 2.1 m/s respectively, when the oven was set at 60°C. When the higher drying air temperature of 80°C was used, greater Listeria reduction occurred. By the time the MC was less than 24% (wb) at 90 min, the initial population was reduced by 1.7 ± 0.3 and 3.6 ± 0.8 log CFU/g respectively at 0.7 and 2.1 m/s. At the same drying temperature, both air velocities resulted in similar Listeria reductions of 5.2 ± 0.5 log CFU/g after 150 min. Overall, temperature, air velocity, and drying time each had a significant effect on Listeria inactivation (P < 0.05), with the interaction between temperature, air velocity, and time being significant (P < 0.05). During apple drying in this study, a dynamic change of water activity was observed in all the treatments, even though the rates were different (Table 5). Such different changes in aw are important because they could be important control factors for L. monocytogenes survival in the apple slices. Previous studies have shown that aw has a significant impact on the thermal inactivation of bacteria such as Salmonella in low-moisture products (2, 43, 78, 107, 109, 110). The results from this study are generally consistent with what those from previous investigations of bacterial inactivation under similar drying conditions. When studying Salmonella survival in various apple varieties during drying, Gurtler et al. (47) reported a 2.0 log reduction in Gala apples when no pre-treatment was applied to apples dried at 60°C for 5 h. Similarly, Dipersio et al. (39) achieved a 2.7 to 2.8 log reduction of Salmonella during drying of Gala apples at 60°C for 6 h. In two other studies where Gala apple slices were inoculated with E. 43 coli, the overall reductions were 2.5 log CFU after 6 h at 57.2 °C (9) and 2.5 (38) or 3.3 (9) log CFU after 6 h at 62.8 °C. In all these studies, drying was conducted in a food dehydrator, and the air velocity was not controlled. Nevertheless, the log reductions reported were similar to what was seen in the present study, in which using 60°C drying temperature decreased Listeria by 1.8- 2.8 logs, depending on the air velocity. Therefore, within in the ranges tested, the effect of air velocity appears to be minimal at lower drying temperatures. In another study using higher temperatures (104 and 135°C) and a relatively low air velocity (0.4 m/s), > 5 log reduction of Salmonella was achieved in apple pieces by the end of drying (45). A similar reduction in Listeria was seen in the present study at 80°C at both the lower and higher air velocities. Though the total drying times differed, these two studies show that moderate (80°C) to high drying temperatures (≥104°C) are potentially effective in decreasing both Listeria and Salmonella in apples during drying, compared to 60°C, which is common for home-scale dehydrators. 4.3 Conclusions In summary, temperature and air velocity both affected apple drying, in terms of moisture removal and aw over time, with temperature having a greater impact. Similar effects were also observed for L. monocytogenes inactivation. Drying at 60°C yielded only a 1.8 to 2.8 log reduction in Listeria, whereas a 5 log reduction was achieved at 80°C. These results suggest that drying at 80°C under the studied air velocities might be sufficient for pathogen control in apple drying industry. Although the higher air velocity resulted in greater Listeria reduction, the effect was independent of drying time and temperature. Therefore, future studies should investigate a wider range of temperatures and air velocities to support the validation of various commercial practices for producing dried apple products. 44 CHAPTER 5: CONCLUSIONS AND FUTURE WORK 5.1 Overall Findings This thesis investigated Salmonella and L. monocytogenes inactivation under widely changing moisture conditions in RTE bacon and dried apples. For bacon, the following conclusions were made. 1. During impingement oven cooking, Salmonella decreased > 6.5 logs before the product was fully-cooked (40% yield). 2. During microwave oven cooking, Salmonella decreased > 6.5 logs in the absence of humidity control (dew point ≥ 35°C) before the end of cooking (40% yield). 3. However, when the dew point in the microwave oven was kept below 25°C, Salmonella decreased < 6.5 log, indicating that humidity enhances Salmonella inactivation during microwave cooking of bacon. 4. Additionally, Salmonella was shown to be more resistant in the fat portion of the bacon than in the lean portion during microwave cooking. For apple drying, the following conclusions were drawn: 1. Higher air velocities and higher temperatures contributed to greater L. monocytogenes inactivation. 2. At a drying temperature of 60°C, Listeria decreased < 5 logs by the end of the drying. 3. However, at 80°C, a 5 log reduction of Listeria monocytogenes was achieved, regardless of the air velocity. 45 Despite obvious compositional differences between bacon and apples, and their complex and different coupled heat and mass transfer mechanisms, similar pathogen inactivation trends were seen in both products (Table 6). In both bacon and apple, the MC and aw decreased during drying to a low-moisture state. Similarly, both bacon and apples showed a continually dynamic product surface temperature during processing. Additionally, both product processes yielded generally similar patterns of pathogen reduction over time (nominally “linear”) for the low- humidity cases. For bacon cooked in an impingement oven, the observed rapid inactivation of Salmonella likely resulted from the high process humidity (60% v/v) and subsequent rapid condensation of water vapor on the product surface, creating a significant advantage for pathogen lethality. For apple drying, inactivation of L. monocytogenes was slower because of the countereffects on lethality of simultaneously increasing the temperature and decreasing moisture over time. Additionally, the apple slices were thicker, which impeded moisture removal and lengthened the processing time. 46 Table 6: Comparison of bacon cooking and apple drying described in this study Moisture Maximum Overall Water Content Surface Pathogen Product Treatment Activity (%, wb) Temperature Reduction (°C) (log CFU/g) Initial Final Initial Final Impingement oven cooking: 0.971 0.573 177°C, 60% (v/v) 40.8 10.2 > 8.4 ± ± ~ 138.9 humidity ± 6.8 ± 4.1 (Salmonella) 0.004 0.152 Impingement oven 0.971 0.391 bacon cooking: 40.8 5.5 > 8.4 ± ± ~ 171.2 232°C, 60% (v/v) ± 6.8 ± 4.4 (Salmonella) 0.004 0.191 humidity Microwave cooking: 80% 0.974 0.758 41.5 18.9 > 8.4 power, (1200W ± ± ~ 103.1 ± 7.7 ± 3.9 (Salmonella) max); humid 0.006 0.079 condition Drying: 85.78 17.50 0.990 0.473 1.8 ± 0.3 60°C, 0.7 m/s ± ± ± ± ~ 60 (Listeria) 0.74 3.73 0.006 0.086 Drying: 85.42 14.40 0.989 0.360 2.8 ± 0.7 60°C, 2.1 m/s ± ± ± ± ~ 60 (Listeria) 0.37 4.53 0.003 0.021 apple Drying: 85.25 10.97 0.991 0.224 5.2 ± 0.5 80°C, 0.7 m/s ± ± ± ± ~ 80 (Listeria) 2.80 2.77 0.004 0.025 Drying: 84.20 11.81 0.991 0.208 5.2 ± 0.5 80°C, 2.1 m/s ± ± ± ± ~ 80 (Listeria) 0.87 2.80 0.001 0.062 5.2 Summary Although RTE bacon and dried apples are two completely different products, they both exhibit similar phenomenological changes in MC, aw, and pathogen inactivation (Figures 2, 5, and 7). Under non-isothermal, dynamic temperature conditions, product MC, aw , and pathogen populations all declined. During impingement oven cooking of bacon, Salmonella decreased 47 rapidly very early in the cooking process due to the effect of vapor condensation, which facilitated rapid inactivation before the product started losing moisture. For bacon cooked in the microwave oven where humidity was not controlled, the target lethality also was achieved before obtaining a RTE product. When humidity was controlled (held below 25°C dew point) in the microwave, Salmonella decreased over time, even though the rate of inactivation was slower. In contrast to bacon, the apple drying treatment was conducted under dry conditions. Therefore, product moisture loss over time influenced the inactivation rate of L. monocytogenes. Nevertheless, a gradual Listeria reduction was observed over time. Both dry microwave cooking and apple drying showed similar pseudo-linear bacterial inactivation curves due to the low humidity. These observations demonstrate that despite obvious differences in product composition and processing methods, bacterial inactivation followed a similar trend, with thermal inactivation influenced by the counter-effects of dynamically and simultaneously increasing temperature and decreasing moisture. Thus, it may be possible to represent bacterial pathogen inactivation under such widely changing moisture conditions using one model form that accounts for these multiple factors, and their possible interactions. A previous study modeled Salmonella inactivation in pistachios incorporating dynamic moisture change (Equation 1) (13). Even though the product in that study was already a low-moisture product, their model form could a starting point to predict Salmonella and other pathogen inactivation in food undergoing dynamic wide moisture change over time. 𝑇𝑟𝑒𝑓 −𝑇𝑠(𝑡) (𝑇𝑑,𝑟𝑒𝑓 −𝑇𝑑 )−(𝑇𝑟𝑒𝑓 −𝑇𝑠 (𝑡)) 𝑀𝐶𝑟𝑒𝑓 −𝑀𝐶(𝑡) + + 𝐷𝑇,𝑇𝑑,𝑀𝐶 (𝑡) = 𝐷𝑟𝑒𝑓 × 10 𝑧𝑇 𝑧𝑀 𝑧𝑀𝐶 (Equation 1) (13) − 𝐷𝑇,𝑇𝑑 ,𝑀𝐶 : D-value in function of dynamic temperature and MC over time (min) − Dref : D-value at a specific reference condition (min) 48 − Tref : Reference temperature (°C) − Td,ref : Reference dew point (°C) − Td : Dew point (°C) − Ts(t) : Product surface temperature over time (°C) − MCref : Reference moisture content (% MC, db) − MC(t) : Moisture content over time (% MC, db) − ZT : Parameter defining the effect of product surface temperature changes on 𝐷𝑇,𝑇𝑑 ,𝑀𝐶 (°C) − ZM : Parameter defining the effect of dew point changes on 𝐷𝑇,𝑇𝑑 ,𝑀𝐶 (°C) − ZMC : Parameter defining the effect of product moisture content changes on 𝐷𝑇,𝑇𝑑 ,𝑀𝐶 (% MC, db) 5.3 Future Work Given the existing FSIS and FSMA regulations for bacon and dried apples, respectively, this thesis has shown that more research is needed to ensure full compliance. For bacon, because humidity influenced Salmonella inactivation during microwave cooking, future studies should focus on the threshold humidity needed to meet the FSIS 6.5 log Salmonella reduction requirement in an industrial setting. It is therefore also important to characterize the humidity in actual commercial, continuous microwave systems, to confirm that typical commercial processes (as expected) operate at humidity levels sufficient to ensure compliance with FSIS expectations for process lethality. For apples, this study and previous literature have demonstrated that pathogen lethality depends on the drying conditions used, including drying air temperature and velocity. Given the limited literature on pathogen inactivation during apple drying or fruit drying in general, and the wide range of parameters and processing methods used by apple drying 49 industry, future work is needed to further understand and predict bacterial pathogen inactivation under such widely changing moisture conditions. In this case, potential research ideas include the following. 1. Evaluating pathogen inactivation during apple drying as affected by product geometry, such as thickness. 2. Assessing the effect of a wider range of air velocities (very low to very high) on pathogen inactivation. 3. Investigating the effects of a wider range of temperatures on inactivation, based on the full range of temperature used by the apple drying industry (e.g., 60 and 190°C). 4. Modeling Salmonella and other pathogen inactivation during apple drying under wide moisture changes over time, as a function of product temperature, MC (and/or aw), and perhaps process humidity and air velocity. 5. Evaluating the effect of anti-browning pre-treatment methods in apple drying and integrating the effect in an inactivation model. 6. Combining existing drying technologies, such as microwave-convective drying, and evaluating the hurdle effect on pathogen inactivation. 7. Imitating the industrial practice of multistage drying in apple and evaluating pathogen inactivation during the process. 50 REFERENCES 1. Ahmed, N. M., D. E. Conner, and D. L. Huffman. 1995. Heat-Resistance of Escherichia Coli O157:H7 in Meat and Poultry as Affected by Product Composition. J. Food Sci. John Wiley & Sons, Ltd 60:606–610. 2. Aljarallah, K. M., and M. R. Adams. 2007. Mechanisms of heat inactivation in Salmonella serotype Typhimurium as affected by low water activity at different temperatures. J. Appl. Microbiol. John Wiley & Sons, Ltd 102:153–160. 3. AOAC International. 1996. Official methods of analysis, 16th ed. AOAC International, Gaithersburg, Maryland. 4. Archer, J., E. T. Jervis, J. Bird, and J. E. Gaze. 1998. Heat resistance of Salmonella weltevreden in low-moisture environments. J. Food Prot. International Association for Food Protection 61:969–973. 5. Bailey, J. S. 1993. Control of Salmonella and Campylobacter in Poultry Production. A Summary of Work at Russell Research Center. Poult. Sci. Elsevier 72:1169–1173. 6. Basheer, S. M., S. Chellappan, and A. Sabu. 2022. Enzymes in fruit and vegetable processing, p. 101–110. In Value-Addition in Food Products and Processing Through Enzyme Technology. Academic Press. 7. Bourdoux, S., D. Li, A. Rajkovic, F. Devlieghere, and M. Uyttendaele. 2016. Performance of Drying Technologies to Ensure Microbial Safety of Dried Fruits and Vegetables. Compr. Rev. Food Sci. Food Saf. John Wiley & Sons, Ltd 15:1056–1066. 8. Burnham, G. M., M. A. Fanslau, and S. C. Ingham. 2006. Evaluating microbial safety of slow partial-cooking processes for bacon: Use of a predictive tool based on small-scale isothermal meat inoculation studies. J. Food Prot. IAMFES 69:602–608. 9. Burnham, J. A., P. A. Kendall, and J. N. Sofos. 2001. Ascorbic acid enhances destruction of escherichia coli O157:H7 during home-type drying of apple slices. J. Food Prot. IAMFES 64:1244–1248. 10. Calvo, T., M. Prieto, A. Alvarez-Ordóñez, and M. López. 2020. E ff ect of Non-Thermal Atmospheric Plasma on Foods : Morphological and Physico-Chemical 1–16. 11. Campbell, J. A., J. S. Dickson, J. C. Cordray, D. Olson, A. F. Mendonca, and K. J. Prusa. 2014. Survival of methicillin-resistant Staphylococcus aureus during commercial heat treatment of slab bacon and consumer preparation of sliced bacon. J. Food Prot. 77:83– 86. 12. Carrasco, E., A. Morales-Rueda, and R. M. García-Gimeno. 2012. Cross-contamination and recontamination by Salmonella in foods: A review. Food Res. Int. Elsevier 45:545– 51 556. 13. Casulli, K. E., K. D. Dolan, and B. P. Marks. 2021. Modeling the Effects of Product Temperature, Product Moisture, and Process Humidity on Thermal Inactivation of Salmonella in Pistachios during Hot-Air Heating. J. Food Prot. Allen Press 84:47–57. 14. Casulli, K. E., F. J. Garces-Vega, K. D. Dolan, E. T. Ryser, L. J. Harris, and B. P. Marks. 2018. Impact of Process Temperature, Humidity, and Initial Product Moisture on Thermal Inactivation of Salmonella Enteritidis PT 30 on Pistachios during Hot-Air Heating. J. Food Prot. Allen Press 81:1351–1356. 15. Center for Disease Control and Prevention (CDC). 2018. Burden of Foodborne Illness: Findings . Available at: https://www.cdc.gov/foodborneburden/2011-foodborne- estimates.html. Accessed 2 November 2021. 16. Centers for Disease Control and Prevention. 2022. Listeria Outbreak Linked to Brie and Camembert Cheese . Available at: https://www.cdc.gov/listeria/outbreaks/monocytogenes- 09-22/index.html. Accessed 3 November 2022. 17. Centers for Disease Control and Prevention. 2022. Listeria Outbreak Linked to Ice Cream . Available at: https://www.cdc.gov/listeria/outbreaks/monocytogenes-06-22/index.html. Accessed 3 November 2022. 18. Centers for Disease Control and Prevention. 2022. Listeria Outbreak Linked to Packaged Salads Produced by Dole. Available at: https://www.cdc.gov/listeria/outbreaks/packaged- salad-mix-12-21/index.html. Accessed 3 November 2022. 19. Centers for Disease Control and Prevention. 2022. Listeria Outbreak Linked to Packaged Salads Produced by Fresh Express. Available at: https://www.cdc.gov/listeria/outbreaks/packaged-salad-12-21-b/index.html. Accessed 3 November 2022. 20. Centers for Disease Control and Prevention. 2016. Multistate Outbreak of Listeriosis Linked to Frozen Vegetables . Available at: https://www.cdc.gov/listeria/outbreaks/frozen-vegetables-05-16/index.html. Accessed 3 November 2022. 21. Centers for Disease Control and Prevention. 2016. Multistate Outbreak of Listeriosis Linked to Packaged Salads Produced at Springfield, Ohio Dole Processing Facility. Available at: https://www.cdc.gov/listeria/outbreaks/bagged-salads-01-16/index.html. Accessed 3 November 2022. 22. Centers for Disease Control and Prevention. 2014. Multistate Outbreak of Listeriosis Linked to Roos Foods Dairy Products. Available at: https://www.cdc.gov/listeria/ outbreaks/cheese-02-14/index.html. Accessed 3 November 2022. 52 23. Centers for Disease Control and Prevention. 2017. Multistate Outbreak of Listeriosis Linked to Soft Raw Milk Cheese Made by Vulto Creamery . Available at: https://www.cdc.gov/listeria/outbreaks/soft-cheese-03-17/index.html. Accessed 3 November 2022. 24. Centers for Disease Control and Prevention. 2012. Multistate Outbreak of Listeriosis Linked to Whole Cantaloupes from Jensen Farms, Colorado. Available at: https://www.cdc.gov/listeria/outbreaks/cantaloupes-jensen-farms/index.html. Accessed 3 November 2022. 25. Centers for Disease Control and Prevention. 2018. Multistate Outbreak of Salmonella Typhimurium Infections Linked to Dried Coconut |. Available at: https://www.cdc.gov/salmonella/typhimurium-03-18/index.html. Accessed 1 November 2022. 26. Centers for Disease Control and Prevention. 2022. National Outbreak Reporting System (NORS) Dashboard. Available at: https://wwwn.cdc.gov/norsdashboard/. Accessed 3 November 2022. 27. Centers for Disease Control and Prevention. 2019. Outbreak of Listeria Infections Linked to Deli-Sliced Meats and Cheeses. Available at: https://www.cdc.gov/listeria/outbreaks/deliproducts-04-19/index.html. Accessed 3 November 2022. 28. Centers for Disease Control and Prevention. 2021. Outbreak of Listeria Infections Linked to Deli Meats. Available at: https://www.cdc.gov/listeria/outbreaks/delimeat-10- 20/index.html. Accessed 3 November 2022. 29. Centers for Disease Control and Prevention. 2019. Outbreak of Listeria Infections Linked to Pork Products. Available at: https://www.cdc.gov/listeria/outbreaks/porkproducts-11- 18/index.html. Accessed 3 November 2022. 30. Centers for Disease Control and Prevention (CDC). 2021. FoodNet Fast. Available at: https://wwwn.cdc.gov/foodnetfast/. Accessed 10 December 2021. 31. Centers for Disease Control and Prevention (CDC). 2019. Highlights from the 2017 Surveillance Report | Foodborne Disease Outbreak Surveillance System (FDOSS). Available at: https://www.cdc.gov/fdoss/annual-reports/2017-report-highlights.html. Accessed 10 December 2021. 32. Centers for Disease Control and Prevention (CDC). 2015. Multistate Outbreak of Listeriosis Linked to Commercially Produced, Prepackaged Caramel Apples Made from Bidart Bros. Apples | Listeria | CDC. Available at: https://www.cdc.gov/listeria/outbreaks/caramel-apples-12-14/index.html. Accessed 30 September 2021. 53 33. Code of Federal Regulations. 1996. 9 CFR Part 417 Hazard Analysis and Critical Control Point (HACCP) Systems. National Archives. 34. Code of Federal Regulations. 1999. 9 CFR 318.17 Requirements for the production of cooked beef, roast beef, and cooked corned beef products. 9 CFR 318.17(a)(1). National Archives. 35. Code of Federal Regulations. 1999. 9 CFR 318.23 Heat-processing and stabilization requirements for uncured meat patties. National Archives. 36. Code of Federal Regulations. 1999. 9 CFR 381.150 Requirements for the production of fully cooked poultry products and partially cooked poultry breakfast strips. National Archives. 37. Cruzen, S. M., H. Cetin-Karaca, R. Tarté, J. G. Sebranek, and J. S. Dickson. 2022. Survival of Clostridium perfringens, Staphylococcus aureus and Salmonella enterica in alternatively cured bacon during cooking and process deviations. Meat Sci. Elsevier 184:108687. 38. Derrickson-Tharrington, E., P. A. Kendall, and J. N. Sofos. 2005. Inactivation of Escherichia coli O157:H7 during storage or drying of apple slices pretreated with acidic solutions. Int. J. Food Microbiol. 99:79–89. 39. Dipersio, P. A., P. A. Kendall, M. Calicioglu, and J. N. Sofos. 2003. Inactivation of Salmonella during Drying and Storage of Apple Slices Treated with Acidic or Sodium Metabisull te SolutionsJournal of Food Protection. 40. Fellows, P. J. 2017. Dehydration, p. 661–716. In Food Processing Technology: Principles and Practice, 4th ed. Elsevier. 41. Finn, S., O. Condell, P. McClure, A. Amézquita, and S. Fanning. 2013. Mechanisms of survival, responses, and sources of salmonella in low-moisture environments. Front. Microbiol. Frontiers Media SA 4:331. 42. Food Safety and Inspection Service U.S Department of Agriculture (USDA FSIS). 2017. FSIS Salmonella Compliance Guidelines for Small and Very Small Meat and Poultry Establishments that Produce Ready-to-Eat (RTE) Products and Revised Appendix A. 43. Garces-Vega, F. J., E. T. Ryser, and B. P. Marks. 2019. Relationships of Water Activity and Moisture Content to the Thermal Inactivation Kinetics of Salmonella in Low- Moisture Foods. J. Food Prot. Allen Press 82:963–970. 44. Garcés-vega, F. J. 2017. Quantifying Water Effects on Thermal Inactivation of Salmonella in Low-Moisture Foods. Michigan State University. 45. Grasso-Kelley, E. M., X. Liu, L. A. Halik, and B. Douglas. 2021. Evaluation of Hot-Air 54 Drying To Inactivate Salmonella and Enterococcus faecium on Apple Pieces. J. Food Prot. 84:240–248. 46. Guo, X., J. Chen, R. E. Brackett, and L. R. Beuchat. 2002. Survival of Salmonella on Tomatoes Stored at High Relative Humidity, in Soil, and on Tomatoes in Contact with Soil. J. Food Prot. Allen Press 65:274–279. 47. Gurtler, J. B., S. E. Keller, X. Fan, O. M. Olanya, T. Jin, and M. J. Camp. 2020. Survival of Salmonella during Apple Dehydration as Affected by Apple Cultivar and Antimicrobial Pretreatment. J. Food Prot. 83:902–909. 48. Hansen, N.-H., and H. Riemann. 1963. Factors Affecting the Heat Resistance of Nonsporing Organisms. J. Appl. Bacteriol. John Wiley & Sons, Ltd 26:314–333. 49. Hildebrandt, I. M., N. O. Hall, M. K. James, E. T. Ryser, and B. P. Marks. 2021. Process Humidity Affects Salmonella Lethality at the Surface and Core of Impingement-Cooked Meat and Poultry Products. J. Food Prot. Allen Press 84:1512–1523. 50. Jacobsen, C. S., and T. B. Bech. 2012. Soil survival of Salmonella and transfer to freshwater and fresh produce. Food Res. Int. Elsevier 45:557–566. 51. James, C., K. E. Barlow, S. J. James, and M. J. Swain. 2006. The influence of processing and product factors on the quality of microwave pre-cooked bacon. J. Food Eng. Elsevier 77:835–843. 52. Jeong, S., P. M. Bradley, and O. R. Alicia. 2009. Thermal Inactivation Kinetics for Salmonella Enteritidis PT30 on Almonds Subjected to Moist-Air Convection Heating. J. Food Prot. Allen Press 72:1602–1609. 53. Juneja, V. K., and B. S. Eblen. 2000. Heat inactivation of Salmonella typhimurium DT104 in beef as affected by fat content. Lett. Appl. Microbiol. John Wiley & Sons, Ltd 30:461– 467. 54. Juneja, V. K., B. S. Eblen, and H. M. Marks. 2001. Modeling non-linear survival curves to calculate thermal inactivation of Salmonella in poultry of different fat levels. Int. J. Food Microbiol. Elsevier 70:37–51. 55. Kaya, A., and O. Aydin. 2009. An experimental study on drying kinetics of some herbal leaves. Energy Convers. Manag. Pergamon 50:118–124. 56. Keerthirathne, T. P., K. Ross, H. Fallowfield, and H. Whiley. 2016. A Review of Temperature, pH, and Other Factors that Influence the Survival of Salmonella in Mayonnaise and Other Raw Egg Products. Pathogens. Multidisciplinary Digital Publishing Institute (MDPI) 5. 57. Krokida, M. K., V. T. Karathanos, Z. B. Maroulis, and D. Marinos-Kouris. 2003. Drying 55 kinetics of some vegetables. J. Food Eng. Elsevier 59:391–403. 58. Lee, D., M. Tertuliano, C. Harris, G. Vellidis, K. Levy, and T. Coolong. 2019. Salmonella Survival in Soil and Transfer Onto Produce via Splash Events. J. Food Prot. Allen Press 82:2023–2037. 59. Marus, J. R., S. Bidol, S. M. Altman, O. Oni, N. Parker-Strobe, M. Otto, E. Pereira, A. Buchholz, J. Huffman, A. R. Conrad, and M. E. Wise. 2019. Notes from the Field: Outbreak of Listeriosis Likely Associated with Prepackaged Caramel Apples — United States, 2017MMWR. Morbidity and Mortality Weekly Report. Centers for Disease Control MMWR Office. 60. Moussavi, M., J. C. Frelka, I. M. Hildebrandt, B. P. Marks, and L. J. Harris. 2020. Thermal Resistance of Foodborne Pathogens and Enterococcus faecium NRRL B-2354 on Inoculated Pistachios. J. Food Prot. Allen Press 83:1125–1136. 61. Niamnuy, C., S. Devahastin, S. Soponronnarit, and G. S. Vijaya Raghavan. 2008. Kinetics of astaxanthin degradation and color changes of dried shrimp during storage. J. Food Eng. Elsevier 87:591–600. 62. Palou, E., A. López-Malo, G. V. Barbosa-Cánovas, J. Welti-Chanes, and B. G. Swanson. 1999. Polyphenoloxidase activity and color of blanched and high hydrostatic pressure treated banana puree. J. Food Sci. Institute of Food Technologists 64:42–45. 63. Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott. 2010. Sources and Risk Factors for Contamination, Survival, Persistence, and Heat Resistance of Salmonella in Low-Moisture Foods. J. Food Prot. Allen Press 73:1919–1936. 64. Poirier, D., T. H. Sanders, and J. P. Davis. 2014. Salmonella Surrogate Reduction Using Industrial Peanut Dry Roasting Parameters. Peanut Sci. Allen Press 41:72–84. 65. Queiroz, C., M. L. Mendes Lopes, E. Fialho, and V. L. Valente-Mesquita. 2008. Polyphenol Oxidase: Characteristics and Mechanisms of Browning Control. http://dx.doi.org/10.1080/87559120802089332. Taylor & Francis Group 24:361–375. 66. Santillana Farakos, S. M., J. F. Frank, and D. W. Schaffner. 2013. Modeling the influence of temperature, water activity and water mobility on the persistence of Salmonella in low- moisture foods. Int. J. Food Microbiol. Elsevier B.V. 166:280–293. 67. Scharff, R. L. 2020. Food Attribution and Economic Cost Estimates for Meat-and Poultry- Related Illnesses. J. Food Prot. 83:959–967. 68. Self, J. L., R. E. Luna-Gierke, A. Fothergill, K. G. Holt, and A. R. Vieira. 2017. Outbreaks attributed to pork in the United States, 1998–2015. Epidemiol. Infect. NIH Public Access 145:2980. 56 69. Shrestha, S., W. Birbari, T. Sheehan, and K. A. Glass. 2016. Thermal inactivation of salmonella on sesame-topped bread during baking using high and low oven humidity. Food Prot. Trends 36:116–124. 70. Sindelar, J., K. Glass, R. Hanson, J. G. Sebranek, J. Cordray, and J. S. Dickson. 2019. Validation of lethality processes for products with slow come up time: Bacon and bone-in ham. Food Control. Elsevier Ltd 104:147–151. 71. Sloniker, N. R., O. Raftopoulou, S. Kathariou, R. Beaudry, and E. T. Ryser. 2022. Fate of Listeria monocytogenes Strains on Different Whole Apple Varieties during Long-Term Simulated Commercial Storage, p. 60. In Technical Presentation (T8-07) at the International of Food Protection Annual Meeting 2022, July 31- Agust 3, 2022, Pittsburgh, Pennsylvania, United States. International of Food Protection Annual Meeting 2022 Abstracts. 72. Smith, D. F., I. M. Hildebrandt, K. E. Casulli, K. D. Dolan, and B. P. Marks. 2016. Modeling the Effect of Temperature and Water Activity on the Thermal Resistance of Salmonella Enteritidis PT 30 in Wheat Flour. J. Food Prot. Allen Press 79:2058–2065. 73. Spector, M. P., and W. J. Kenyon. 2012. Resistance and survival strategies of Salmonella enterica to environmental stresses. Food Res. Int. Elsevier 45:455–481. 74. Steele, M., and J. Odumeru. 2004. Irrigation Water as Source of Foodborne Pathogens on Fruit and Vegetables. J. Food Prot. Allen Press 67:2839–2849. 75. Subhashree, S. N., S. Sunoj, J. Xue, and G. C. Bora. 2017. Quantification of browning in apples using colour and textural features by image analysis. Food Qual. Saf. Oxford Academic 1:221–226. 76. Syamaladevi, R. M., R. K. Tadapaneni, J. Xu, R. Villa-Rojas, J. Tang, B. Carter, S. Sablani, and B. Marks. 2016. Water activity change at elevated temperatures and thermal resistance of Salmonella in all purpose wheat flour and peanut butter. Food Res. Int. Elsevier 81:163–170. 77. Taormina, P. J., and G. W. Bartholomew. 2005. Validation of Bacon Processing Conditions To Verify Control of Clostridium perfringens and Staphylococcus aureusJournal of Food Protection. 78. Tapía, M. S., S. M. Alzamora, and J. Chirife. 2020. Effects of Water Activity (a w ) on Microbial Stability as a Hurdle in Food Preservation. Water Act. Foods Fundam. Appl. wiley 323–355. 79. Toǧrul, I. T., and D. Pehlivan. 2003. Modelling of drying kinetics of single apricot. J. Food Eng. Elsevier 58:23–32. 80. Tsai, H. C., M. H. Taylor, X. Song, L. Sheng, J. Tang, and M. J. Zhu. 2019. Thermal 57 resistance of Listeria monocytogenes in natural unsweetened cocoa powder under different water activity. Food Control. Elsevier 102:22–28. 81. United States Department of Agriculture. 2019. Marketing and Regulatory Programs Agricultural Marketing Service Specialty Crops Program Specialty Crops Inspection Division United States Standards for Grades of Apples. 82. United States Department of Agriculture. 1955. United States Standards for Grades of Dried ApplesUnited States Department of Agriculture. 83. United States Department of Agriculture Agricultural Marketing Service. Apple Grades & Standards . Available at: https://www.ams.usda.gov/grades-standards/apple-grades- standards. Accessed 18 October 2022. 84. United States Department of Agriculture Agricultural Marketing Service. Dehydrated Apples Grades and Standards. Available at: https://www.ams.usda.gov/grades- standards/dehydrated-apples-grades-and-standards. Accessed 18 October 2022. 85. United States Department of Agriculture Economic Research Service. 2021. Food Availability (Per Capita) Data System. Available at: https://www.ers.usda.gov/data- products/chart-gallery/gallery/chart-detail/?chartId=58322. Accessed 1 November 2022. 86. United States Department of Agriculture Economic Research Service. 2021. Food Availability and Consumption. Available at: https://www.ers.usda.gov/data-products/ag- and-food-statistics-charting-the-essentials/food-availability-and-consumption/. Accessed 30 September 2021. 87. United States Department of Agriculture Economic Research Service. 2021. Fruit and Tree Nuts Yearbook Tables. Available at: https://www.ers.usda.gov/data-products/fruit- and-tree-nuts-data/fruit-and-tree-nuts-yearbook-tables/#Noncitrus Fruit. Accessed 28 October 2022. 88. United States Department of Agriculture Food Safety and Inspection Service. 2013. Bacon and Food Safety. Available at: https://www.fsis.usda.gov/food-safety/safe-food-handling- and-preparation/meat/bacon-and-food-safety. Accessed 15 October 2022. 89. United States Department of Agriculture Food Safety and Inspection Service. Federal Acts Governing the Work of FSIS. Available at: https://www.fsis.usda.gov/policy. Accessed 3 November 2022. 90. United States Department of Agriculture Food Safety and Inspection Service. 2021. FSIS Cooking Guideline for Meat and Poultry Products (Revised Appendix A). 91. United States Department of Agriculture Food Safety and Inspection Service. 2021. Safe Food Handling and Preparation. Available at: https://www.fsis.usda.gov/food-safety/safe- food-handling-and-preparation. Accessed 3 November 2022. 58 92. United States Department of Agriculture Food Safety and Inspection Service (USDA FSIS). 2005. Food Standards and Labeling Policy Book. 93. United States Food and Drug Administration. 2011. Background on the FDA Food Safety Modernization Act. Available at: https://www.fda.gov/food/food-safety-modernization- act-fsma/background-fda-food-safety-modernization-act-fsma. Accessed 3 November 2022. 94. United States Food and Drug Administration. 2020. Country Fresh Expands Voluntary Recall . Available at: https://www.fda.gov/safety/recalls-market-withdrawals-safety- alerts/country-fresh-expands-voluntary-recall. Accessed 1 November 2022. 95. United States Food and Drug Administration. 2021. Food Safety at Home. Available at: https://www.fda.gov/consumers/free-publications-women/food-safety-home. Accessed 3 November 2022. 96. United States Food and Drug Administration. 2022. Food Safety Modernization Act (FSMA). Available at: https://www.fda.gov/food/guidance-regulation-food-and-dietary- supplements/food-safety-modernization-act-fsma. Accessed 7 July 2022. 97. United States Food and Drug Administration. 2017. Fresh Pak Inc. Recalls Lot Specific Sliced Apple Products Because of Possible Health Risk. Available at: https://www.fda.gov/safety/recalls-market-withdrawals-safety-alerts/fresh-pak-inc-recalls- lot-specific-sliced-apple-products-because-possible-health-risk. Accessed 1 November 2022. 98. United States Food and Drug Administration. 2020. FSMA Final Rule for Preventive Controls for Human Food. Available at: https://www.fda.gov/food/food-safety- modernization-act-fsma/fsma-final-rule-preventive-controls-human-food. Accessed 28 October 2022. 99. United States Food and Drug Administration. 2017. Jack Brown Produce, Inc. Recalls Gala, Fuji, Honeycrisp and Golden Delicious Apples Due to Possible Health Risk. Office of Regulatory Affairs. Available at: http://wayback.archive- it.org/7993/20180125012623/https://www.fda.gov/Safety/Recalls/ucm589722.htm. Accessed 1 November 2022. 100. United States Food and Drug Administration. 2014. Kirkland Signature Real Sliced Fruit Recalled Due to Possible Health Risk. Office of Regulatory Affairs,Office of the Commissioner. Available at: http://wayback.archive- it.org/7993/20170112140101/http://www.fda.gov/Safety/Recalls/ArchiveRecalls/2013/uc m389177.htm. Accessed 1 November 2022. 101. United States Food and Drug Administration. 2019. North Bay Produce Voluntarily Recalls Fresh Apples Because of Possible Health Risk . Available at: https://www.fda.gov/safety/recalls-market-withdrawals-safety-alerts/north-bay-produce- 59 voluntarily-recalls-fresh-apples-because-possible-health-risk. Accessed 1 November 2022. 102. United States Food and Drug Administration. 2019. Producing a Food Product that is Regulated by FDA. Available at: https://www.fda.gov/food/food-industry/producing-food- product-regulated-fda. Accessed 10 December 2021. 103. United States Food and Drug Administration. 2018. Publix Recalls Publix Tropical Medley Mix Due to Possible Contamination with Listeria monocytogenes | FDA. Available at: https://www.fda.gov/safety/recalls-market-withdrawals-safety-alerts/publix- recalls-publix-tropical-medley-mix-due-possible-contamination-listeria-monocytogenes. Accessed 1 November 2022. 104. US Food & Drug Administration (US FDA). 2020. Get the Facts about Listeria | FDA. Available at: https://www.fda.gov/animal-veterinary/animal-health-literacy/get-facts- about-listeria. Accessed 2 November 2021. 105. USDA Food Safety and Inspection Service. Food Safety and Inspection Service. Available at: https://www.fsis.usda.gov/. Accessed 2 November 2021. 106. Vega-Gálvez, A., K. Ah-Hen, M. Chacana, J. Vergara, J. Martínez-Monzó, P. García- Segovia, R. Lemus-Mondaca, and K. Di Scala. 2012. Effect of temperature and air velocity on drying kinetics, antioxidant capacity, total phenolic content, colour, texture and microstructure of apple (var. Granny Smith) slices. Food Chem. Elsevier 132:51–59. 107. Villa-Rojas, R., J. Tang, S. Wang, M. Gao, D. H. Kang, J. H. Mah, P. Gray, M. E. Sosa- Morales, and A. L. Pez-Malo. 2013. Thermal Inactivation of Salmonella Enteritidis PT 30 in Almond Kernels as Influenced by Water Activity. J. Food Prot. Allen Press 76:26–32. 108. World Health Organization (WHO). Estimating the burden of foodborne diseases. Available at: https://www.who.int/activities/estimating-the-burden-of-foodborne-diseases. Accessed 2 March 2022. 109. Xie, Y., T. Cheng, L. Wei, M. J. Zhu, S. S. Sablani, and J. Tang. 2021. Thermal inactivation of Salmonella Enteritidis PT30 in ground cinnamon as influenced by water activity and temperature. Food Control. Elsevier 124:107935. 110. Yuqiao, J. I. N., S. R. Pickens, I. M. Hildebrandt, S. J. Burbick, E. M. Grasso-Kelley, S. E. Keller, and N. M. Anderson. 2018. Thermal Inactivation of Salmonella Agona in Low– Water Activity Foods: Predictive Models for the Combined Effect of Temperature, Water Activity, and Food Component. J. Food Prot. Allen Press 81:1411–1417. 60 APPENDIX A: Collected Data from Bacon Cooking Study Table 7: Collected data from Salmonella inactivation during microwave cooking of bacon Salmonella Cooking Water MC Replication Subsample Population Yield (%) Time (s) Activity (%, wb) (log CFU/ g) 0 1 A 9.1 0.9704 33.21 0 1 B 9.1 0.9732 28.99 0 1 C 9.1 0.9807 16.62 0 1 A 9.6 0.9772 47.76 0 1 B 9.7 0.9725 46.54 0 1 C 9.6 0.9814 42.89 0 1 A 9.1 0.9761 48.58 0 1 B 9.0 0.9778 56.08 0 1 C 9.1 0.9737 45.34 0 1 A 9.2 0.9746 46.37 0 1 B 9.3 0.9605 35.49 0 1 C 9.0 0.9689 42.04 0 1 A 9.0 0.9533 26.92 0 1 B 9.1 0.9707 36.79 0 1 C 9.1 0.9652 38.49 0 1 A 9.1 0.9808 37.40 0 1 B 9.1 0.9769 43.76 0 1 C 9.0 0.9734 44.64 0 2 A 9.6 0.9666 38.41 0 2 B 8.8 0.9690 41.49 0 2 A 8.8 0.9809 47.57 0 3 B 9.7 0.9698 38.68 0 3 A 9.1 0.9617 41.53 0 3 B 9.6 0.9716 34.00 0 4 A 9.2 0.9770 49.01 0 4 B 9.2 0.9779 48.57 0 4 A 9.0 0.9725 34.72 0 5 B 9.4 0.9744 43.77 0 5 A 9.3 0.9697 49.14 0 5 B 9.3 0.9730 49.35 0 6 A 9.3 0.9738 39.15 0 6 B 9.3 0.9693 42.86 0 6 A 9.3 0.9731 43.07 30 1 A 7.7 0.9603 40.28 83.67 61 Table 7 (cont’d) Salmonella Cooking Water MC Replication Subsample Population Yield (%) Time (s) Activity (%, wb) (log CFU/ g) 30 1 B 8.4 0.9635 36.31 88.46 30 1 A 8.4 0.9650 46.35 87.06 30 1 B 8.1 0.9625 36.92 87.08 30 1 A 8.0 0.9647 40.97 84.82 30 1 B 8.4 0.9649 38.26 88.85 30 1 A 8.4 0.9541 46.08 87.47 30 1 B 8.2 0.9597 23.96 86.79 30 1 A 8.4 0.9677 48.47 86.91 30 1 B 8.2 0.9555 35.05 84.73 30 1 A 7.9 0.9749 31.68 85.78 30 1 B 8.2 0.9753 30.42 85.94 30 2 A 8.2 0.9642 38.57 84.24 30 2 B 7.7 0.9598 26.17 85.15 30 3 A 7.8 0.9656 69.33 85.46 30 3 B 8.4 0.8299 37.83 88.24 30 4 A 8.4 0.9559 34.62 88.66 30 4 B 8.0 0.9514 39.09 85.53 30 5 A 7.9 0.9642 44.08 83.11 30 5 B 8.2 0.9524 29.95 85.15 30 6 A 8.4 0.9676 43.39 83.22 30 6 B 8.5 0.9705 44.22 84.16 30 7 A 8.4 0.9638 38.30 84.45 30 7 B 8.5 0.9500 38.07 87.63 45 1 A 4.1 0.9473 39.54 68.84 45 1 B 7.3 0.9308 35.07 75.63 45 1 A 7.0 0.9052 24.86 69.98 45 1 B 5.7 0.9413 33.41 72.97 45 1 A 4.7 0.9430 39.03 70.24 45 1 B 7.0 0.9374 34.81 75.77 45 1 A 4.5 0.9366 38.88 69.67 45 1 B 6.8 0.9368 34.68 76.27 45 1 A 5.8 0.9300 34.80 68.35 45 1 B 7.6 0.9431 39.74 76.43 45 1 A 4.6 0.9422 31.45 73.22 45 1 B 6.0 0.9005 28.46 71.06 45 2 A 5.6 0.9578 38.28 70.35 45 2 B 7.1 0.9466 37.35 72.99 62 Table 7 (cont’d) Salmonella Cooking Water MC Replication Subsample Population Yield (%) Time (s) Activity (%, wb) (log CFU/ g) 45 3 A 7.3 0.9590 35.71 71.43 45 3 B 6.1 0.9430 36.10 75.09 45 4 A 6.7 0.9412 35.77 76.01 45 4 B 5.1 0.9160 34.28 72.79 45 5 A 5.2 0.9582 29.64 69.14 45 5 B 6.5 0.9390 33.69 75.94 45 6 A 5.5 0.9591 41.68 71.41 45 6 B 6.7 0.9528 34.95 76.31 45 7 A 6.5 0.9481 38.94 74.31 45 7 B 7.5 0.9327 34.90 74.64 60 1 A 3.4 0.9376 37.78 66.06 60 1 B 3.8 0.9233 31.20 54.34 60 1 A 1.3 0.9245 31.59 58.28 60 1 B 3.0 0.9502 42.17 66.81 60 1 A 1.1 0.9237 55.59 62.61 60 1 B 1.2 0.9395 35.45 61.01 60 1 A 0.6 0.9245 33.13 65.53 60 1 B 1.1 0.9516 41.86 64.46 60 1 A 0.8 0.9513 45.47 69.55 60 1 B 4.6 0.9358 39.86 64.91 60 1 A 4.1 0.9608 38.83 66.53 60 1 B 1.9 0.9507 40.31 67.39 60 2 A 3.9 0.9520 39.90 61.21 60 2 B 2.9 0.9120 33.29 66.40 60 3 A