Validation of brownie baking step for controlling Salmonella and Listeria monocytogenes

Abstract Pathogens, such as Salmonella and Listeria monocytogenes, can survive under the dry environment of flour for extended periods of time and could multiply when flour is hydrated to prepare batter or dough. Therefore, inactivation of these pathogens during the cooking/baking step is vital to ensure the microbiological safety of bakery products such as brownies. The aim of this research was to validate a simulated commercial baking process as a kill‐step for controlling Salmonella and L. monocytogenes in brownies and to determine thermal inactivation parameters of these pathogens in brownie batter. Independent studies were conducted in a completely randomized design for each pathogen. All‐purpose flour was inoculated with a 5‐serovar Salmonella and 3‐strain L. monocytogenes cocktails. For baking validation, brownie batters were prepared from inoculated flour, and cooked in the oven set at 350°F (176.7°C) for 40 min followed by 15 min of ambient air cooling. For calculating D‐values, brownie batter was transferred into thermal‐death‐time disks, sealed, and placed in hot‐water baths. The samples were held for pre‐determined time intervals in hot‐water baths and immediately transferred to cold‐water baths. Microbial populations were enumerated using injury‐recovery media. At the end of baking, Salmonella and L. monocytogenes populations decreased by 6.3 and 5.9 log CFU/g, respectively. D‐values of Salmonella and L. monocytogenes cocktails were 53.4 and 37.5 min at 64°C; 27.2 and 16.9 min at 68°C; 10.7 and 9.1 min at 72°C; and 4.6 and 7.3 min at 76°C; respectively. The z‐values of Salmonella and L. monocytogenes cocktails were 11.1 and 16.4°C, respectively. This study can be used as a supporting document for the validation of similar brownie baking processes to control Salmonella and L. monocytogenes. The data from this study can also be employed for developing basic prediction models for the survival and thermal resistance of these pathogens during brownie baking step.


| INTRODUC TI ON
The U.S. Food and Drug Administration's Food Safety Modernization Act (FSMA) is an important step taken toward making the food supply chain in the United States safer by striving to prevent foodborne illnesses rather than responding to the foodborne illness outbreaks (FDA, 2018a). The FSMA requires that food processor should have science-based food safety as well as food defense plans to prevent accidental and intentional contamination (FDA, 2018a;Kennedy et al., 2014). According to the FSMA, the preventive control processing steps should be validated on the basis of scientific studies to ensure that identified hazards are controlled or eliminated (FDA, 2018b).
Although thermal processing steps during food preparation are considered effective to control or eradicate foodborne pathogens, these thermal kill-step steps need to be validated to ensure that every food particle is exposed to the minimum required temperature for the minimum required time to effectively control the identified microbiological hazard. Contamination in foods can be introduced from various sources including raw ingredients, processing environment, employees, and/or post-processing steps. Foodborne pathogens, such as Salmonella and Listeria monocytogenes, cannot grow in low water activity (a w ) food ingredients (such as flour, powders, and spices), but they can survive for months to years in dry environments (Ballom et al., 2020;Beuchat et al., 2013;Taylor et al., 2018).
Pathogens dormant in dry ingredients could start multiplying when rehydrated to prepare batter or dough, and could cause illnesses if appropriate preventive control steps are not effective. Therefore, it is vital for the food industry to validate all implemented preventive controls to ensure the safety of the finished food products.
The main objective of this research was to validate a simulated commercial brownie baking process to control Salmonella and L. monocytogenes in brownies using a conventional kitchen oven.
Dry ingredients such as all-purpose flour used in the production of brownies could introduce microbial contamination to the brownie batter as flour does not have any post-process lethality treatment (Myoda et al., 2019). Salmonella and L. monocytogenes were used in this validation study because they are two of the most common nonspore former foodborne pathogens that can survive in dry environments for prolonged periods of time. Moreover, these pathogens are also two of the top five pathogens responsible for foodborne-related illnesses and/or deaths in the United States (Scallan et al., 2011).
Listeria monocytogenes represented Gram positive bacterial group, whereas Salmonella represented Gram negative group.

| Experimental and statistical design
This research consisted of two independent studies: (a) brownie baking process validation: consisted of two individual validation experiments using artificially inoculated flour to control Salmonella and L. monocytogenes in brownies during baking, and product water activity (a w ) and pH were monitored during the baking as well; (b) D-and z-values determination: consisted of two individual experiments to determine thermal inactivation parameters of Salmonella and L. monocytogenes in brownie batter prepared from artificially inoculated flour.
For the validation studies, brownie batter was prepared from inoculated flour, baked, aseptically sampled, and enumerated for the respective pathogen populations. These studies were conducted as completely randomized design with ten treatments (pre-baking; 5, 10, 15, 20, 25, 30, 35, and 40 min sampling times during baking, and 15 min of ambient cooling post-baking) and three replications. Oneway ANOVA using Minitab® 19 was used to determine any statistical differences (α = 0.05) in the microbial populations, a w and pH. The D-

| Microbial cultures
All microbial cultures were obtained from the American Type Culture Collection (ATCC®; Manassas, VA) in freeze-dried forms. Five serovars of Salmonella and three strains of L. monocytogenes were used in the current research (Table 1). These cultures were selected based on their isolation from various food matrices and/or association with foodborne illnesses. All cultures were propagated according to the manufacturer's instructions, and working cultures were stored in 10 ml of Brain Heart Infusion (BHI; Difco™, Becton, Dickinson and Company) broth at 4°C.

| Master inoculation preparation
For each replication, individual serovars/strains of Salmonella and L. monocytogenes were streaked from BHI broth onto BHI agar as lawns using sterile swabs and incubated at 37°C for 24 hr. Lawns of individual strains were harvested using 1 ml of 0.1% peptone solution (Bacto™, Becton, Dickinson and Company) twice and dislodging bacterial cells using sterile plastic L-spreaders . The harvested individual bacterial solutions were then mixed in equal proportion to obtain a 5-serovar cocktail of Salmonella or 3-strain cocktail of L. monocytogenes, transferred into 50 ml sterile plastic tubes connected to spray nozzles, and used as master inoculum for inoculating the flour. The spray nozzles were calibrated to discharge ~1 ml of master inoculum per squirt. The lawn method was used for the master inoculum preparation because it produced highly concentrated bacterial cell solution, which resulted in less volume of inoculum required for mist inoculating the flour.

| Flour inoculation
Two hundred grams of all-purpose wheat flour was weighed in a sanitized sealable plastic tub (~38 × 26 × 7 cm, Rubbermaid Inc.) and spread evenly to get a uniform layer. The flour in tub was then transferred inside a biosafety cabinet and mist inoculated by spraying ~4 ml of master inoculum in four squirts. Inoculated flour was then dried back to the original pre-inoculation weight by incubating with open lids at 37°C for ~4 hr to achieve the a w of 0.305 ± 0.01. The dried, inoculated flour was mixed well using a sanitized spatula, sealed with air-tight lids, and stored at room temperature (~20°C) until used within 2 days.

| Brownie batter preparation
All of the ingredients were purchased from a local store in Pullman, WA. The Brownie recipe including all the baking parameters used in the study was based on a general recipe used for commercial bakery brownies. Inoculated flour was weighed in a sanitized mixing bowl followed by the addition of granulated sugar, powdered sugar, cocoa powder, salt, soybean oil, water, fresh liquid whole egg, and vanilla extract ( Table 2). The mixing bowl with ingredients was then attached to a kitchen mixer (Classic KitcehnAid®) and mixed for 1 min at speed-2 with scrapping after first 30 s.

| Brownie baking
Brownie batter (~1,690 g) was evenly spread [~1.25 cm height] using a sanitized spatula into a 12 × 12 inch (30.48 × 30.48 cm) pan and placed into a conventional oven preheat to 350°F (176.7°C) ( Figure 1). Five T40 fine-gauge type-K thermocouples (Thermo-Electra) were placed in the brownie batter, and one was placed inside the oven. The thermocouples in brownie batter were placed in 5 different spots throughout the pan: at the geometric center, at the center depth halfway from the edge of the pan to the center, at the center depth in the corner, at the top surface of brownies, and at the bottom of the brownies ( Figure 2). All six thermocouples were connected to a data logger (USB-TC Measurement Computing™) and a computer. The timer was started as soon brownie batter was placed in the oven, and the brownies were sampled every 5 min. After 40 min of baking, baked brownies were removed from the oven, followed by 15 min of ambient air cooling. During the preliminary work, it was confirmed that opening the oven door for <10 s for sampling did not affect the product temperature. However, although the oven air temperature did decrease during the sampling process, this decreased oven air temperatures reached back to the pre-sampling temperatures in <45 s.

| a w and pH determination
The a w of samples was measured at 25°C using a calibrated a w meter (Novasina Labswift Portable a w meter). The a w was measured for the batter prepared from inoculated flour at 0,5,10,15,20,25,30,35,40, and after a 15 min ambient air cooling. For measuring a w during

Soybean oil 200
Whole egg 300 Vanilla extract 20 baking, one sample was taken at each sampling time. Until 20 min, a w was measured for the whole brownie batter sample. However, from 20 min, there was a noticeable difference in appearance of the crust and crumb; therefore, the crumb and crust were separated and used to measure a w separately. Samples were aseptically transferred into a w cups, sealed, and allowed to cool to ~25°C before measuring a w .
After measuring a w , sample pH was also measured using a calibrated surface pH meter (ExStik® Waterproof pH Meter, Extech) at 25°C. Although the actual baking time to achieve best quality brownies was baking for 40 min, samples were taken at 5 min intervals during baking to account for worst-case scenarios where brownies might be undercooked because of cooking process malfunctioning. As microbial kill in brownies can continue during ambient cooling, a 15-min ambient cooling treatment was also included in the studies.

| Brownie baking validation
The average temperature of the brownies during baking is presented in Figure 4. After 10 min of heating, the average brownie tempera- baking, the brownie temperature had increased to 91.9 ± 3.23°C.
After 40 min of baking, the complete bake time, the temperature was 106.9 ± 2.54°C. At the end of 15 min of ambient air cooling, the brownie temperature decreased to 64.5 ± 3.12°C. The average pH and a w throughout the baking process are presented in Figures 5 and 6, respectively. Overall, pH of brownies did not change during the baking process. The pH of brownie batter at the start of baking was 5.7 ± 0.078; whereas, at the end of ambient cooling, brownie pH was 5.6 ± 0.094 and 5.8 ± 0.060 in crust and crumb, respectively.
The a w of brownie batter at the start of baking was 0.822 ± 0.007, which significantly decreased after ambient air cooling, with a w of 0.658 ± 0.013 in brownie crumb and 0.532 ± 0.027 in brownie crust. during the baking process. Channaiah et al. (2016) concluded that as E. faecium took longer to achieve >6 log reduction and exhibited greater thermal resistance compared to Salmonella, E. faecium can be used as a surrogate for Salmonella for in-plant validation studies. In another validation study, Channaiah et al. (2018) reported that frying donuts in oil at 190.6°C for 2 min (1 min on each side) resulted in >7 log reductions in 7-serovar Salmonella cocktail population (Newport, Typhimurium, Senftenberg, Tennessee and three dry food isolates) and the internal donut temperature reached to ~119°C.  Channaiah et al. (2016Channaiah et al. ( , 2017Channaiah et al. ( , 2018 can be attributed to the different Salmonella serovars used and the differences in the chemical/ nutrient composition of the products used in these studies. In general, with the increase in fat, protein and sugar content of products, the thermal resistance of microorganisms also increases. Fat,  protein, and sugar molecules have the capability to form a protective coating around the bacterial cell (Juneja et al., 2001;Mattick et al., 2001;Smith et al., 2001). This protective coating acts as a sheath against the heat effect and subsequently leading to elevated thermal resistance values. For example, Sekhon et al. (2020) determined the thermal resistance of Salmonella cocktail (similar to this study) in nonfat dry milk and whole milk powder during the storage period of 6 months. Sekhon et al. (2020) reported that although the thermal resistance of Salmonella cocktail in whole milk powder (higher fat content) and nonfat dry milk was similar at the start of the storage, the thermal resistance of Salmonella increased during the storage period in whole milk powder but stayed similar in nonfat dry milk.

| CON CLUS IONS
The current research showed that a typical commercial brownie baking process with >100°C internal temperature for at least 40 min will result in ≥5-log reductions in Salmonella and L. monocytogenes. It should be noted that validation studies are specific to the individual pathogens in specific food formulation and process.
Therefore, independent validation studies should be conducted for other pathogens or if significant changes are made to the brownie recipe used in the current study. The thermal resistance of bacteria immensely depends on the pH and a w of the food matrix; therefore, the continuous determination of pH and a w values during the baking process is critical in studying the survivability of respective pathogens. The D-values calculated in the current study can be used as the bases to develop the thermal inactivation predictive models for the inactivation of L. monocytogenes and Salmonella in brownies. However, additional D-and z-value data should be calculated to develop strong and robust inactivation models.

ACK N OWLED G EM ENT
The authors would like to acknowledge AIB International, Inc., Manhattan, KS for funding this project.

CO N FLI C T O F I NTE R E S T
The authors do not have any conflict of interest.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.