As interest in “clean label” foods increases, concerns regarding pathogen inhibition in naturally cured meats have been raised due to nitrite’s antimicrobial efficacy. Although research exists that addresses the safety of naturally cured meat, limited work investigates this concern in fermented and dried products manufactured with nitrite alternatives. Therefore, research was conducted to determine the survival of Escherichia coli O157:H7 (EC), Listeria monocytogenes (LM), and Salmonella spp. (S) in raw, ready-to-eat salami manufactured with various nitrite alternatives. Three independent replications of 4 treatments were evaluated: negative control, positive control (PC; purified/conventional sodium nitrite), Swiss chard powder (SC), and Prosur T-10 (T-10), a commercial dried fruit extract. PC and SC were formulated to 156 ppm. T-10 was formulated to 0.96% of the total formulation. All treatments were formulated using manufacturers’ guidelines for appropriate ingredient utilization. To prepare the salami, pork shoulder butts were ground and inoculated with a 3-strain culture of EC, LM, and S to obtain approximately 7 log10 CFU/g. Dry ingredients and starter culture then were mixed and stuffed into fibrous casings of about 55 mm, fermented (pH <5.0), and dried to a target water activity (aw) of 0.88 in a commercial drying cabinet. After drying, salamis were vacuum packaged and stored (20 ± 0.003°C). Salamis were sampled for pathogen survival, pH, and aw at raw manufacturing, throughout the 3-d fermentation, every week for 7 wk, and on day 118 (n = 9; N = 432). Greater than 2 log10 CFU/g reduction of LM and S was achieved by PC and SC. Processors may use this research as scientific support if they manufacture pork salami with 156 ppm nitrite sourced from purified nitrite or SC, ferment to a pH of less than 5.0, have a final aw of less than 0.88, and have an ingoing salt content of 2.5%.
Salami is a ready-to-eat (RTE), fermented, cured, and dried meat product that typically consists of pork, beef, or a combination of both, mixed with curing agents, salt, sugars, spices, flavorings, and a starter culture (Toldrá, 2002; Maddock, 2015). After mixing meat and nonmeat ingredients together, the meat batter is stuffed into either natural or synthetic casings and hung into an environmental chamber to ferment and dry. Traditional salamis are manufactured without a thermal lethality step. Other meat products rely on cooking to inhibit pathogen growth and survival, the application of hurdle technology. By combining low pH (<5.0), redox potential, and water activity (aw) (<0.90), high salt content (>2%), the addition of competitive microorganisms from a starter culture, and nitrite inclusion permit for the safe production of salami (Leistner and Gorris, 1995).
Nitrite is an essential, multifunctional ingredient in cured, processed meat products. Nitrite is responsible for many cured meat qualities such as color, flavor, and antioxidant activity. The ingredient is also well-documented as an effective inhibitor against pathogenic and spoilage organisms (Sindelar and Milkowski, 2011). Although best acknowledged for its inhibition of Clostridium spp. germination and toxin production, nitrite is also recognized for its control of Escherichia coli O157:H7 (EC), Listeria monocytogenes (LM), Salmonella spp. (S), Staphylococcus aureus, and Bacillus cereus (Milkowski et al., 2010). The effectiveness of nitrite against pathogenic organisms is increased when applied in combination with other intrinsic factors of the meat product such as low pH and aw, high salt concentrations, and other antimicrobial agents (Tompkin, 2005). The mechanisms of pathogen control by nitrite are not well-understood; however, it is primarily thought that nitrite blocks essential compounds for pathogen growth and survival (e.g., oxygen, glucose, and other metabolic enzymes) and penetrates cell membranes, which disturbs the ability of the cell to divide (Yarbrough, et al., 1980; Buchanan et al., 1989; Duffy et al., 1994; Työppönen et al., 2002).
Unfortunately, nitrite and the consumption of cured meat products have been repeatedly associated with an increased risk for some cancer types. Specifically, the American Institute for Cancer Research (AICR) states that there is convincing evidence for an increased risk of colorectal cancer and limited suggestive evidence for an increased risk for lung, pancreatic, stomach, esophageal, and nasopharyngeal cancers due to consuming processed meats manufactured with purified, also known as synthetic or conventional nitrite sources (AICR, 2021). Processed meat products have, therefore, been classified as carcinogenic to humans based on sufficient evidence found by the International Agency for Research on Cancer (IARC), an entity of the World Health Organization (WHO). Furthermore, the IARC has stated that consuming a portion of cured processed meats as small as 50 g increases the risk of colorectal cancer by 18% (WHO-IARC, 2015).
Due to health concerns associated with nitrite consumption, and despite the vast benefits nitrite inclusion has on cured meat products, consumer demands are driving the meat industry away from the application of purified nitrites and toward “clean label” alternatives (Sebranek et al., 2012). Although “clean label” does not have a strict definition, the term is broadly understood as a short, easy to read ingredient statement that is perceived to be nonartificial or nonchemical sounding (Asioli et al., 2017). Generally, consumers have been found to prefer and are willing to pay premiums for clean label products (Grant et al., 2021). Therefore, to meet consumer demands, many meat processors have increased the application of natural nitrite sources in processed meat products (Bizzozero, 2017; Iqbal et al., 2021).
Natural alternatives to nitrite are typically fruits or vegetables that are naturally high in nitrate content, like celery or Swiss chard (SC) (Sebranek et al., 2012). Nitrate content in natural sources can vary due to seasonal, geographic, growing practice, and genetic differences (Kalaycıoğlu and Erim, 2019). The US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS) explains in the Code of Federal Regulations (CFR) 9 §424.21 (c) that it is recommended that natural nitrite sources be formulated to the minimum nitrite content regulated for the product being manufactured and manufactured using the minimum times for fermentation, aging, and drying. However, processors may apply natural sources of nitrite to lower levels of ingoing nitrite content, as these ingredients are not approved as curing agents. Rather, they are considered antimicrobials and flavorings as stated in the USDA-FSIS directive 7120.1, “Safe and Suitable Ingredients Used in the Production of Meat and Poultry Products.”
Furthermore, lower concentrations of ingoing nitrite, approximately 40 ppm, have been shown to achieve desired cured meat quality attributes, whereas higher concentrations, approximately 120 ppm, are needed to prevent pathogen growth and toxin production in products labeled “keep refrigerated” (USDA-FSIS, 1995). Given the lack of regulations for minimum ingoing nitrite in products, like salami, that have been manufactured to ensure shelf stability through a combination of pH, moisture, and appropriate packaging controls (USDA-FSIS, 1995), processors may apply natural nitrite sources in reduced amounts as a method to reduce costs of processing (Sebranek and Bacus, 2007).
Salami and other cured, fermented, and dried meat products have been identified as the causative agent in multiple foodborne outbreaks. An industry changing foodborne outbreak in 1994 was linked to the consumption of EC contaminated dry-cured salami (Centers for Disease Control and Prevention [CDC], 1995). This outbreak led to the USDA-FSIS proposing a rule for a 5 log10 reduction of Shiga toxin-producing EC (STEC) in dry and semidry products containing beef, and S in products containing meat of other species origins (USDA-FSIS, 2001; USDA-FSIS, 2023a). LM is the most frequently isolated pathogen in the USDA-FSIS fermented sausage monitoring program (USDA-FSIS, 2001). Recently, over 69,000 lb of RTE charcuterie products were deemed adulterated because of LM contamination (USDA-FSIS, 2023b). Two outbreaks of S related to naturally cured, RTE, fermented, and dried meat products occurred in late 2021. The USDA-FSIS determined that antipasto Italian-style meat products manufactured with SC as the curing agent contaminated with S were the source of the outbreak that resulted in 12 hospitalizations (USDA-FSIS, 2021; CDC, 2021a). The investigation of this outbreak found that the processor did not have scientific validation of their process achieving a 5 log10 reduction of S. The second outbreak of S was linked to a salami stick manufactured with SC that resulted in 34 salmonellosis infections and 7 hospitalizations across 10 states (CDC, 2021b). It was concluded that this outbreak was a result of the product manufacturer not having sufficient evidence to support the adequate control of S through their fermentation and drying steps (USDA-FSIS, 2022).
Concerns regarding pathogen inhibition in naturally cured processed meat products have been raised due to the variable nitrite content in natural nitrite sources, possibility of reduced ingoing nitrite formulations, and foodborne outbreaks related to naturally cured meat products (Rivera et al., 2019). The concern of natural nitrite sources has been well investigated in other processed meat products like deli-style turkey (Golden et al., 2014), ham (Sullivan et al., 2012), frankfurters (Jackson et al., 2011), and bacon (Gipe, 2012). There remains a gap, however, in scientific literature that investigates the efficacy of natural nitrite sources on pathogen inhibition in RTE, cured, fermented, and dried meat products. Therefore, the objective of this research was to determine the fate of EC, LM, and S during fermentation, drying, and extended reduced oxygen, ambient storage of salami manufactured without a thermal lethality step using various nitrite alternatives.
Materials and MethodsTreatment groups
This research consisted of 4 treatment groups of salami manufactured with or without commercially available curing agents. Treatment groups were no nitrites added (negative control [NC]), purified nitrite standardized at 6.25%, sodium nitrite, and 93.75% sodium chloride (positive control [PC]), SC (Swiss Chard 531, Florida Food Products, LLC; Eustis, FL), and Prosur T-10 (T-10; Prosur Inc.; Naperville, IL), a dried fruit extract comprised of undisclosed fruits and advertised as a replacement to nitrite rather than an alternative source of the ingredient (Table 1). All treatments were performed in 3 independent replications and manufactured according to industry standards for raw RTE salami. Three individual sausages from each treatment and a replicate were taken every sampling day, which resulted in 108 sausages sampled per treatment group at the end of the study (n = 9; N = 432).
Ingredient formulations for each treatment of salami in percentage of total batch.
SC (Swiss Chard 531; Florida Food Products, LLC; Eustis, FL) and T-10 (Prosur; Naperville, IL) were formulated based on manufacturer utilization recommendations.
Culture selection and inoculum preparation
Three strains each of EC, LM, and S were received from the American Type Culture Collection (ATCC; Manassas, VA), CDC (Atlanta, GA), and The Microbiology Culture Collection at the Pennsylvania State University Food Science Department (University Park, PA). EC isolates EDL933 (ATCC 43895; ground beef outbreak), Sakai, and PA-2 (Hartzell, et al., 2011); LM serotypes Scott A, ½a isolate FSL R2-603 (deli meats outbreak), and 4b isolate H3396 (hot dog outbreak); and S serovars Typhimurium (ATCC 14028; chicken organs), Montevideo isolate Smvo13 (CDC), and Derby (ATCC 7378; human isolate) were identified for use in this study.
Cultures of each organism were stored at −80°C prior to use. A loopful of each frozen culture was aseptically transferred to 10 mL tryptic soy broth (TSB; Becton, Dickinson, and Company; BD, Franklin Lakes, NJ) and incubated aerobically at 36°C for 24 h. After incubation, overnight cultures were streaked onto sorbitol MacConkey agar supplemented with cefixime tellurite (CT-SMAC; HiMedia Laboratories, LLC; HiMedia; Kelton, PA), modified Oxford agar (MOX; HiMedia), and xylose lysine deoxycholate agar (XLD; HiMedia) for EC, LM, and S isolation, respectively. CT-SMAC and XLD plates were incubated at 36°C for 24 h, and MOX was incubated at 28°C for 48 h. Resulting isolated colonies of each pathogen strain were confirmed using protein agglutination tests (EC, S, and LM: Microgen® Bioproducts; Hardy Diagnostics; Santa Maria, CA).
To prepare the inoculum, single isolated colonies of each pathogen strain were independently inoculated, in duplicate, into 25 mL of TSB, resulting in 50 mL of overnight culture of each pathogen strain after a 24 h incubation at 36°C. This step was done to achieve an approximate cell concentration of 8 log10 CFU/mL in each overnight culture (adapted from USDA-FSIS, 2012). Overnight culture concentration was determined using methodology detailed in the USDA Microbiological Laboratory Guidebook and through preliminary experimentation. Overnight cultures were centrifuged at approximately 20°C for 5 min at 11,000 × g (Avanti JLA-16.250; Beckman Coulter; Pasadena, CA). After centrifugation, the supernatant was poured off and disposed of, leaving only a concentrated pellet of bacteria. Within pathogen, each pathogen strain was combined using 2.5 mL buffered peptone water (BPW; BD). This process resulted in a 2.5 mL inoculum for each pathogen, EC, LM, and S.
Salami manufacturing and inoculation
Pork shoulder butts (IMPS 406; sourced from local suppliers) were deboned, cubed to approximately 2.54 cm × 2.54 cm, and ground (GMG 180A; Hollymatic Corp.; Countryside, IL) to approximately 5 mm (400 Triumph 3/16” Holes #103421; Speco Inc.; Schiller Park, IL). Ground pork was vacuum sealed (50.8 cm × 71.2 cm bags of 3 mil thickness; Con Yeager Spice Company; Zelienople, PA) as 11.33 kg batches and stored at about 4°C for less than 24 h.
Ground pork was mixed (Hakka 15-Liter Capacity Tank Stainless Steel Manual Meat Mixer; Hakka Brothers; Hayward, CA) with previously described inoculum of EC, LM, and S to obtain an inoculation concentration of approximately 7 log10 CFU/g of each pathogen in the meat batter before the addition of dry ingredients. Uniform distribution was ensured by adding the inocula during mixing. Dry ingredients for each treatment were formulated on a meat block weight basis (Table 1). PC was formulated to 156 ppm ingoing sodium nitrite according to regulated usage rates for comminuted and cured processed meat products (9 CFR §424.21). SC was formulated based on 22,500 ppm sodium nitrite in the ingredient and manufacturers’ recommendations for ingredient application to achieve 156 ppm ingoing nitrite to the product (personal communication, Florida Food Products, February 8, 2022). T-10 was formulated to about 1% of the total formulation according to recommended usage rates provided by the ingredient manufacturer (personal communication, Prosur Inc., March 21, 2022). Ingoing salt amount in NC was adjusted to match the amount in PC. All treatments had a final salt content of 4.64 ± 0.12%. Starter culture (Safepro® B-LC 007; CHR Hansen; Hoersholm, Denmark) combined with sterile deionized (DI) water was added to the meat batter after all dry ingredients were well-distributed. Nitrite content was not measured in the ingredient, meat batter, or final product, as the inclusion of pathogens prevented the transport of products to a nonbiosafety-level-2 laboratory space with the equipment necessary to measure residual nitrite content.
Mixed meat batter was then stuffed (Model MF-15V; Walton’s Incorporated; Wichita, KS) as individual sausages into 55-mm permeable, fibrous casings (Globe Casings; Carlstadt, NJ). Sausages from all treatments were stuffed to an average of 264.71 ± 19.79 g and 52.09 ± 0.36 mm. Stuffed sausages were hung in a drying cabinet (AS50; Impianti Condizionamento Salumifici; Camposanto, Modena, Italy) for fermentation (72 h) to a target pH of 5.0 and drying to a target aw of 0.88 (see Table 2 for drying cabinet program). Throughout drying, sausages were sprayed with distilled white vinegar as needed to prevent surface mold growth (5% acidity; Wegmans Food Markets, Inc.; Rochester, NY). Once meeting the target aw (∼21 d), salamis were removed from the drying cabinet, vacuum sealed (8” × 10” 3 Mil Nylon/Poly; Phoenix Scale & Food Equipment; Dallas, PA), and stored at ambient temperature (20 ± 0.003°C).
Drying cabinet program (AS50; Impianti Condizionamento Salumifici; Camposanto, Modena, Italy).
Drying Cabinet Program
Phase
Min Temp
Max Temp
Min Humidity
Max Humidity
Time (h)
Static cooling
6
8
0
0
5
Hot drip
24
26
0
0
36
Drying
24
26
55
65
12
Drying
22
24
60
70
12
Drying
20
22
65
75
12
Drying
18
20
65
75
12
Seasoning
16
18
65
73
24
Seasoning
14
16
66
73
24
Seasoning
11
13
67
72
0
Temperatures are in °C. Humidity parameters are % relative humidity. Humidity programmed to 0 is at the same relative humidity as the environment. Phase times set to 0 run indefinitely until manually shut off.
Sampling procedure
Individual salamis were randomly selected and evaluated in triplicate on each sampling day: 0 (raw batter), 1, 2, 3, 7, 14, 21, 28, 35, 42, 49, and 118. Day 118 (D118) was added to explore pathogen survival after approximately 4 mo postpackaging. Nine salamis were sampled on each sampling day per treatment (n = 9), and 432 salamis sampled throughout the experiment across all treatments (N = 432). All treatments were sampled equally. The casing was removed aseptically, and 20 g samples from the center of each sausage, totaling 60 g total salami weight at each sampling time for each replicate, were combined then diluted with 240 g BPW, creating a 1:5 dilution in a filtered stomacher bag (BagFilter P; Interscience Laboratories Inc.; St.-Normandy, France). The dilution was stomached at 230 rpm for 30 s (Stomacher® 400 Circulator; Seward Ltd.; West Sussex, UK). After stomaching, the stomachate was serially diluted in 9 mL BPW blanks to an appropriate dilution for the sampling day. Dilutions were pipetted as 100 μL aliquots and spread plated, in duplicate, onto CT-SMAC, MOX, and XLD and incubated at 36°C (CT-SMAC and XLD: 24 h; MOX: 48 h) for the enumeration of EC, LM, and S, respectively. Presumptive colonies were counted based on colony color and morphology respective to each agar used. The lower limit of detection for each agar was 0.40 log10 CFU/g. There were no visible colonies on a plate, a concentration of .39 log10 CFU/g was assigned.
In addition to pathogen enumeration, pH (Testo 206-pH2 pH Meter; Testo, Inc.; Sparta, NJ) and aw (AquaLab Water Activity Meter, Series 4TE; Decagon Devices, Inc.; Pullman, WA) were measured on 3 salamis from each replication on every sampling day. pH was measured from the core of the sausages. aw was measured from a thin slice taken from the center of the sausages. Salt content was measured on D118. To measure salt content, a pulverized salami core from each treatment replication (10 g) was diluted with DI water (90 g) and boiled on a hot plate. The solution was filtered through a filter paper, and a salt strip (Chloride QuanTab® Test Strips; Hach; Loveland, CO) was entered into the filtered liquid. Salt content was analyzed on D118 to determine if formulated, ingoing salt percentages differed from that in the product at the end of storage.
Statistical analysis
Duplicate plate counts were averaged, and populations of EC, LM, and S were converted to log10 CFU/g prior to statistical analysis. Plates with no observed colony growth after incubation were assigned a concentration 0.01 log10 CFU/g, less than the detection limit (0.40 log10 CFU/g) to incorporate the counts into the analysis. Enrichment procedures were not performed because once pathogens reached a concentration below the detection limit, a target 5 log10 reduction would have already been achieved, subsequently making enrichment procedures unnecessary to perform. Average pathogen populations were independently compared within treatment using a general linear model procedure with unique comparisons in Statistical Analysis Software (SAS OnDemand Version 9.4; SAS Institute Inc.; Cary, NC). All results were analyzed using a mixed model procedure in SAS. The model included comparisons across treatments on a sampling day and a treatment group by sampling day interaction. pH and aw were included in the model as fixed effects. Comparisons between pathogens were not made to maintain statistical power. A significance level of P < .05 was assigned to determine statistical significance in both analyses.
ResultspH and a<sub>w</sub>
All salami treatments achieved the target pH of less than 5.0 after the first 24 h of fermentation (average pH of all treatments = 4.79 ± 0.02; Table 3). pH values increased gradually thereafter. The fixed effects of pH did not have a significant effect on pathogen populations in the salami throughout the duration of the study (EC: P = .3562; LM: P = .4861; S: P = .6082).
pH (Testo 206-pH2 pH Meter; Testo, Inc.; Sparta, NJ) and water activity (AquaLab Water Activity Meter, Series 4TE; Decagon Devices, Inc.; Pullman, WA) averages on each sampling day (n = 3; N = 144).
The average aw remained above the lower aw limit for EC and S growth (aw = 0.95) until day 3 (D3) in SC, and day 7 in NC, PC, and T-10 (Table 3). The aw of all treatment groups went below the lower limit for LM growth (aw = 0.92) by day 14. Salamis were vacuum packaged when the target aw of 0.88 was met. The target aw was achieved by all salami treatment groups during week 3 of manufacturing (NC = 0.87; PC = 0.86; SC = 0.84; T-10 = 0.87). Final aw measurements on D118 for NC, PC, SC, and T-10 were 0.83, 0.84, 0.82, and 0.83, respectively. The fixed effects of aw did not have a significant effect on pathogen populations throughout the duration of the study (EC: P = .8364; LM: P = .8861; S: P = .9779).
Bacteria
Curing agent treatment had a significant impact on all pathogen populations for the duration of the study (P < .001). Table 4 shows the EC populations and total reductions for each treatment group throughout the study. An increase in EC populations was observed for NC and T-10 on D3, the end of fermentation (0.08 log10 CFU/g and 0.24 log10 CFU/g, respectively). PC achieved the greatest reduction of EC populations at the end of the drying period on D3 (NC = 1.33; PC = 2.61; SC = 0.78; T-10 = 2.14). EC population reductions between day 0 (D0) and day 21 (D21), when salamis reached target aw and were packaged, were: 1.33, 2.61, 0.78, and 2.14 log10 CFU/g for NC, PC, SC, and T-10, respectively. EC populations on D21 in NC, 5.62 ± 0.11 log10 CFU/g, and SC, 5.44 ± 0.17 log10 CFU/g, were not significantly different (P = .4628). EC populations on D21 in PC and T-10 were 4.28 ± 0.25 log10 CFU/g and 4.37 ± 0.27 log10 CFU/g, respectively, and were not significantly different (P = .7275). Furthermore, EC in NC was significantly different from that in PC (P < .0001) and T-10 (P = .002) on D21. EC in SC on D21 was also significantly different from PC (P = .001) and T-10 (P = .0031). EC populations in NC on day 49 (D49) were significantly different from all other treatment groups (P > .05). Total EC reductions between D0 and D118 for NC, PC, SC, and T-10 were 6.12, 6.50, 5.83, and 6.12 log10 CFU/g, respectively. All comparisons of EC populations between D0 and D118 were significant for each treatment (P < .0001). NC was the only treatment to have EC above the lower limit of detection on D118 (0.83 ± 0.28 log10 CFU/g).
Average Escherichia coli O157:H7 populations and reductions (log10 CFU/g ± standard error).
EC Results
Phase
Sample Day
NC (n = 108)
PC (n = 108)
SC (n = 108)
T-10 (n = 108)
Raw
0
6.95 ± 0.08A,a
6.89 ± 0.22AB,a
6.22 ± 0.20B,a
6.51 ± 0.11AB,a
Fermentation
1
5.04 ± 0.28A,b
6.98 ± 0.15B,a
6.16 ± 0.11C,a
6.57 ± 0.03BC,a
2
6.58 ± 0.26A,a
5.35 ± 0.12B,b
5.26 ± 0.26B,b
6.05 ± 0.15A,b
3
7.07 ± 0.09A,a
6.80 ± 0.07A,a
4.63 ± 0.19B,b
6.75 ± 0.04A,a
Drying
7
6.11 ± 0.12A,b
6.58 ± 0.11A,a
5.94 ± 0.04AB,a
5.39 ± 0.25B,b
14
6.57 ± 0.03A,b
5.07 ± 0.16B,b
4.11 ± 0.39B,b
5.73 ± 0.22C,a
21
5.62 ± 0.11A,c
4.28 ± 0.25B,c
5.44 ± 0.17A,a
4.37 ± 0.27B,b
Packaging
28
4.87 ± 0.49A,d
3.01 ± 0.09B,d
3.60 ± 0.10BC,b
3.92 ± 0.27C,b
35
3.22 ± 0.28A,e
3.01 ± 0.45AB,d
2.42 ± 0.45B,c
3.36 ± 0.19A,c
42
2.52 ± 0.22A,e
2.60 ± 0.36A,d
2.19 ± 0.26A,c
2.19 ± 0.19A,d
49
3.00 ± 0.41A,e
1.51 ± 0.09B,e
1.84 ± 0.71B,c
2.14 ± 0.15B,d
118
0.83 ± 0.28A,f
0.39 ± < 0.01A,f
0.39 ± < 0.01A,d
0.39 ± < 0.01A,e
TR
6.12
6.50
5.83
6.12
D0, day 0; D118, day 118; EC, Escherichia coli O157:H7; NC, negative control; PC, positive control; SC, Swiss chard; T-10, Prosur T-10; TR, total reduction.
Sampling days within a column that have a different lowercase letter than the previous day are significantly different (P < .05).
Populations within a row that do not share an uppercase letter are significantly different (P < .05).
TR is the difference between populations (log10 CFU/g) on D0 and D118.
LM populations in each treatment throughout the study are shown in Table 5. PC and SC achieved the greatest reductions of LM on D3, the last day of the fermentation phase (PC = 1.98 log10 CFU/g; SC = 1.67 log10 CFU/g) compared to NC and T-10. NC, however, had increased from 7.12 ± 0.06 to 7.18 ± 0.09 log10 CFU/g from D0 to D3. LM subjected to T-10 had a 0.39 log10 CFU/g decrease from D0 to D3. Similar trends in LM reductions were observed on D21. LM reductions from D0 to D21, when salami achieved the target aw, were 1.06, 2.35, 2.57, and 1.19 log10 CFU/g for NC, PC, SC, and T-10, respectively. Furthermore, PC differed significantly from NC (P < .0001), SC (P = .0054), and T-10 (P < .0001) on D21. SC was also significantly different from NC (P < .0001) and T-10 (P < .0001). NC was not significantly different from T-10 on D21 (P = .39630). LM populations in NC on D118 were significantly different from PC, SC, and T-10 (P < .0001). A significant difference in LM populations was not seen among days in NC until comparing D21 and day 28 (D28) (P < .0001). Populations of LM subjected to PC, SC, and T-10 did not differ significantly on D118 (P > .05), but NC differed significantly from PC, SC, and T-10 (P < .0001). All reductions from D0 to D118 for all treatments were significant (P < .0001). Total LM reductions for NC, PC, SC, and T-10 were 4.55, 6.89, 6.4, and 6.94 log10 CFU/g, respectively. NC did not achieve a 5 log10 reduction of LM throughout the entirety of the study, including extended ambient storage. Furthermore, LM populations on D118 in T-10 and NC, unlike PC and SC, did not go below the lower limit of detection.
Average Listeria monocytogenes populations and reductions (log10 CFU/g ± standard error).
LM Results
Phase
Sample Day
NC (n = 108)
PC (n = 108)
SC (n = 108)
T-10 (n = 108)
Raw
0
7.12 ± 0.06AB,a
7.28 ± 0.04A,a
6.79 ± 0.13B,a
7.48 ± 0.13A,a
Fermentation
1
7.07 ± 0.11A,a
5.97 ± 0.17B,b
5.61 ± 0.04B,b
6.85 ± 0.07A,b
2
6.96 ± 0.16A,a
5.60 ± 0.17B,b
4.85 ± 0.13B,c
6.83 ± 0.07A,b
3
7.18 ± 0.09A,a
5.30 ± 0.04B,b
5.12 ± 0.14B,b
7.09 ± 0.03A,b
Drying
7
6.83 ± 0.02 A,a
5.26 ± 0.15B,b
4.95 ± 0.26B,b
6.87 ± 0.13A,b
14
6.58 ± 0.1A,a
5.08 ± 0.29B,b
4.62 ± 0.11C,b
6.47 ± 0.09A,c
21
6.54 ± 0.04A,a
5.11 ± 0.07B,b
4.49 ± 0.12C,b
6.75 ± 0.12A,b
Packaging
28
5.11 ± 0.38A,b
3.93 ± 0.06B,c
3.78 ± 0.14B,c
6.22 ± 0.07C,c
35
5.68 ± 0.12A,c
4.17 ± 0.23B,c
3.18 ± 0.09C,d
5.82 ± 0.09A,d
42
5.49 ± 0.17A,b
3.90 ± 0.23B,c
2.99 ± 0.12C,d
5.30 ± 0.19A,e
49
5.00 ± 0.23A,d
2.97 ± 0.24B,d
3.83 ± 0.30C,b
4.43 ± 0.13D,f
118
2.63 ± 0.15A,e
0.39 ± < 0.01B,e
0.39 ± < 0.01B,c
0.54 ± 0.10B,g
TR
4.55
6.89
6.4
6.94
D0, day 0; D118, day 118; LM, Listeria monocytogenes; NC, negative control; PC, positive control; SC, Swiss chard; T-10, Prosur T-10; TR, total reduction.
Sampling days within a column that have a different lowercase letter than the previous day are significantly different (P < .05).
Populations within a row that do not share an uppercase letter are significantly different (P < .05).
TR is the difference between populations (log10 CFU/g) on D0 and D118.
No initial inoculation populations of S differed significantly across treatments (P > .05; Table 6). At the end of fermentation (D3), NC was the only treatment to be observed to have an increase (0.29 log10 CFU/g) in S. NC was significantly different from PC (P = .0001) and SC (P < .0001) on D3. At the end of drying (D21), NC, PC, SC, and T-10 achieved S reductions of 0.58, 2.17, 2.3, and 0.73 log10 CFU/g, respectively. S populations in NC, 5.73 ± 0.18 log10 CFU/g, and T-10, 5.58 ± 0.11 log10 CFU/g, did not differ significantly on D21 (P = .5528). S populations in PC and SC on D21 were 4.63 ± 0.07 and 4.56 ± 0.21 log10 CFU/g, respectively, and were not significantly different (P = .7069). S populations in NC were significantly different from those in PC and SC on D21 (P < .0001). T-10 was also significantly different from PC and SC on D21 (P < .0001). On D49, NC was the only treatment statistically different from any other treatments (P < .0001). All treatments were below the lower limit of detection of S and were not significantly different on D118 (P > .05). All total reductions from D0 to D118 were significantly different (P < .0001). Total reductions of S from D0 to D118 were 6.4, 6.59, 6.74, and 6.39 log10 CFU/g for NC, PC, SC, and T-10, respectively.
Average Salmonella spp. populations and reductions (log10 CFU/g ± standard error).
S Results
Phase
Sample Day
NC (n = 108)
PC (n = 108)
SC (n = 108)
T-10 (n = 108)
Raw
0
6.79 ± 0.06A,a
6.98 ± 0.19A,a
7.13 ± 0.06A,a
6.77 ± 0.05A,a
Fermentation
1
7.22 ± 0.12A,a
6.83 ± 0.19AB,a
6.67 ± 0.19B,a
6.71 ± 0.15B,a
2
7.06 ± 0.15A,a
6.38 ± 0.14BC,a
5.97 ± 0.16C,b
6.61 ± 0.16AB,a
3
7.00 ± 0.15A,a
6.08 ± 0.18BC,a
5.72 ± 0.16C,b
6.53 ± 0.19AB,a
Drying
7
6.38 ± 0.11AC,b
6.10 ± 0.11C,a
5.41 ± 0.11B,b
6.53 ± 0.19A,a
14
6.23 ± 0.26A,b
5.55 ± 0.03B,b
5.37 ± 0.07B,b
5.59 ± 0.12B,b
21
5.73 ± 0.18A,c
4.63 ± 0.07B,c
4.56 ± 0.21B,c
5.58 ± 0.11A,b
Packaging
28
5.41 ± 0.19A,c
3.70 ± 0.13B,d
3.55 ± 0.13B,d
4.84 ± 0.13C,c
35
4.34 ± 0.24A,d
2.11 ± 0.31B,e
2.13 ± 0.14B,e
3.65 ± 0.09C,d
42
3.19 ± 0.22A,f
2.56 ± 0.20B,e
2.27 ± 0.16B,e
2.40 ± 0.26B,d
49
3.27 ± 0.21A,e
1.27 ± 0.16B,f
1.28 ± 0.36B,f
1.61 ± 0.21B,d
118
0.39 ± < 0.01A,g
0.39 ± < 0.01A,g
0.39 ± < 0.01A,g
0.39 ± < 0.01A,d
TR
6.4
6.59
6.74
6.38
D0, day 0; D118, day 118; NC, negative control; PC, positive control; S, Salmonella spp.; SC Swiss chard; T-10, Prosur T-10; TR, total reduction.
Sampling days within a column that have a different lowercase letter than the previous day are significantly different (P < .05).
Populations within a row that do not share an uppercase letter are significantly different (P < .05).
TR is the difference between populations (log10 CFU/g) on D0 and D118 (n = 9; N = 432).
Discussion
The USDA-FSIS (2001) recommends a minimum 5 log10 reduction of S in dry and semidry meat products and at least a 5 log10 reduction of STEC in dry and semidry meat products containing beef. Although a 5 log10 reduction of LM is preferable for a greater margin of safety in fermented meat products, at least a 3 log10 reduction of LM is recommended to be achieved during a lethality treatment of RTE shelf-stable products (USDA, 2023a). Salami is typically packaged for commercial sale once quality and safety parameters are met (D21 of this study); however, no treatment group achieved the desired 5 log10 reduction of any pathogen investigated. Alternatively, using the results of this study combined with “option #5” of the Blue Ribbon Task Force (1996), processors may use the results from this study as scientific validation of a 2 log10 reduction of LM and S in their Hazard Analysis Critical Control Point plan for manufacturing raw, RTE pork salami cured with purified nitrite or SC that includes an analytical method for raw batter testing. Processors are recommended to sample fifteen 25 g samples per lot of product. This research may be used to support the application of “option #5” in the Blue Ribbon Task Force because PC and SC achieved a more than 2 log10 CFU/g reduction of LM and S at the time of packaging. NC and T-10 achieved LM reductions of 1.06 and 1.19 log10 CFU/g, respectively, and S reductions of 0.58 and 0.73 log10 CFU/g, respectively by D21. Although EC is not an adulterant in pork, it is important to acknowledge that PC and T-10 achieved a more than 2 log10 CFU/g reduction of EC at the time of packaging. Furthermore, this study does not serve as scientific validation for a 2 log10 reduction of LM or S in raw, RTE salami manufactured without a curing agent or with T-10. The reductions of EC, LM, and S at D21 display the efficacy of each applying PC, SC, or T-10 since NC did not achieve a more than 2 log10 CFU/g reduction of any pathogen in this study. Although PC and T-10 achieved more than 2 log10 CFU/g reduction of EC on D21, processors should not use this study as sole validation for products containing beef as more research is needed to support the manufacturing procedures in this study for fermented and dried products containing beef.
The successful management of processing controls in this research was exhibited by the fixed effects of pH and aw not having a significant impact on any pathogen population differences among treatments throughout the duration of this study. Furthermore, successful processing controls also were displayed by differences in pathogen populations among treatment groups only being significantly impacted by the treatment itself (P < .0001). The importance of combining hurdle technology is displayed by PC achieving more than 1 log10 CFU/g reduction for all pathogens when compared to NC. All salami treatments were formulated to and manufactured with the same ingoing amount and type of starter culture, which is a mixed culture of microorganisms. The Pediococcus acidilactici strain in the starter culture produces a pediocin byproduct of fermentation known to have bactericidal properties against LM (Nielsen et al., 1990). Despite the equivalence of bacteriocin inclusion and processing control parameters, LM populations in NC were significantly different from PC and SC throughout the entirety of the study, apart from initial inoculation populations. Furthermore, LM populations in T-10 were significantly different from PC and SC throughout the study until D118 (P < .05) and were not significantly different from NC until D28. The manufacturer of T-10 states and has found that there are no, or low, residual nitrates or nitrites (<1 ppm) in the finished product when formulated with the ingredient (Hernández et al., 2021). Therefore, the observed differences between LM populations in salami manufactured with and without nitrite sources may be due to the inhibition sodium nitrite has on LM (Buchanan et al., 1989; Duffy et al., 1994; Ngutter and Donnelly, 2003).
The current study exhibited similar reductions of EC, LM, and S reductions in cured, nonthermal lethality treated pork salami (McKinney et al., 2017) and in LM and S reductions in duck salami (Watson et al., 2021) when formulated to 156 ppm ingoing sodium nitrite. The study presented here is the first of its kind to exhibit the longevity of combining curing agents and other pathogen hurdles to control EC, LM, and S during extended ambient storage of raw, RTE salami. All treatments in this study achieved and exceeded a 5 log10 total reduction of EC and S by D118, exhibiting the efficacy of low oxygen environments as a pathogen inhibitor in combination with other hurdle technologies. NC was the only treatment that did not achieve a 5 log10 reduction of LM (4.55 log10 CFU/g total LM reduction). Therefore, this finding demonstrates the importance of including a curing agent for the control of LM during extended, reduced oxygen storage of salami at ambient temperatures.
Other research has also investigated naturally cured, RTE meat products. Golden et al. (2014) surface inoculated fully cooked, deli-style turkey breasts with LM. The turkey breasts were prepared with various concentrations of purified nitrite or cultured celery powder. The researchers found that the concentration of nitrite, rather than the source, was the most critical factor in inhibiting LM. This knowledge was applied when formulating treatments for this research, as PC and SC were both intentionally formulated to 156 ppm ingoing nitrite. Another study found that LM was better controlled than EC or S in pork bellies injected with brines prepared with natural or purified sources of nitrite (Gipe, 2012). Similarly, when comparing the total reductions of EC, LM, and S in the research detailed here, though not statistically compared, it can be inferred that nitrite sources serve as a bigger hurdle to LM than EC or S in raw, RTE salami.
Conclusions
It is important to acknowledge that this research represents a worst-case-scenario for RTE, fermented, and dried meat products manufactured using various curing agents without a thermal lethality step. A processor may use this experiment as scientific validation for manufacturing raw, RTE salami with purified nitrite or SC formulated to 156 ppm, ferment to a pH of less than 5.0, a final aw of less than .88, and have a total, ingoing salt content of at least 2.5%. Additionally, it is recommended to follow good manufacturing procedures during salami processing, using the methods described in this research. More research is recommended to investigate the safety of naturally cured salami when applying other antimicrobial treatments or ingredients, applying other natural nitrite alternatives, using other types of raw meat materials, and using different style casings or product diameters. Additionally, formulations and processing utilized in this study were designed to emulate those applied by medium, small, and very small processors. Therefore, SC and T-10 were formulated based on supplier recommendations without applying additional analytical testing, specifically residual nitrite analyses. Therefore, analytical measurements of residual nitrite in the product throughout the duration of a challenge study like the one discussed here would add valuable knowledge by providing the ability to correlate nitrite levels to pathogen survival.
Conflict of Interest
The authors declare no conflict of interest regarding the content of this manuscript.
Acknowledgements
This research was supported by HATCH Project #4696.
Author Contributions
Heather Hunt conceptualized the study, conducted the research, collected and analyzed data, and wrote the draft manuscript; Ed Mills edited the manuscript; Cathy Cutter provided laboratory space and edited the manuscript; Jonathan Campbell provided supervision, secured funding, and edited the manuscript.
Literature CitedAmerican Institute for Cancer Research (AICR). 2021. AICR Food facts: processed meat (sausages, ham, bacon, hot dogs, salami). https://aicr.org/cancer-prevention/food-facts/processed-meat/. (Accessed May 5, 2023).Asioli, D., J.Aschemann-Witzel, V.Caputo, R.Vecchio, A.Annunziata, T.Næs, and P.Varela. 2017. Making sense of the “clean label” trends: a review of consumer food choice behavior and discussion of industry implications. Food Res. Int. 99:58–71. https://doi.org/10.1016/j.foodres.2017.07.022.Bizzozero, J.2017. 75% of consumers will pay extra for clean label ingredients. Food & Beverage Insider. https://foodbeverageinsider.com/market-trends-analysis/75-consumers-will-pay-extra-clean-label-ingredients. (Accessed May 1, 2023).Buchanan, R. L., H. G.Stahl, and R. C.Whiting. 1989. Effects and interactions of temperature, pH, atmosphere, sodium chloride, and sodium nitrite on the growth of Listeria monocytogenes. J. Food Protect. 52:1–7. https://doi.org/10.4315/0362-028X-52.12.844.Centers for Disease Control and Prevention (CDC). 1995. Escherichia coli O157:H7 outbreak linked to commercially distributed dry-cured salami—Washington and California, 1994. MMWR-Morbid. Mortal. W. 44:157–160. https://cdc.gov/mmwr/preview/mmwrhtml/00036467.htm. (Accessed February 13, 2023).Centers for Disease Control and Prevention (CDC). 2021a. Salmonella outbreaks linked to Italian-style meats. CDC.gov. https://archive.cdc.gov/www_cdc_gov/salmonella/italian-style-meat-08-21/details.html. (Accessed August 22, 2021).Centers for Disease Control and Prevention (CDC). 2021b. Salmonella outbreak linked to salami sticks. CDC.gov. https://archive.cdc.gov/#/details?url=https://www.cdc.gov/salmonella/i45-10-21/index.html. (Accessed August 22, 2021).Duffy, L. L., P. B.Vanderline, and F. H.Grau. 1994. Growth of Listeria monocytogenes on vacuum-packed cooked meats: effects of PH, aw, nitrite and ascorbate. Int. J. Food Microbiol. 23:377–390. https://doi.org/10.1016/0168-1605(94)90164-3.Gipe, A.2012. Investigation of quality attributes and inhibition of foodborne pathogens in “no-nitrate or nitrite-added” bacon. Ph.D. diss., The Pennsylvania State Univ., University Park, PA.Golden, M. C., L. M.Mcdonnell, V.Sheehan, J. J.Sindelar, and K. A.Glass. 2014. Inhibition of Listeria monocytogenes in deli-style turkey breast formulated with cultured celery powder and/or cultured sugar–vinegar blend during storage at 4°C. J. Food Protect. 77:1787–1793. https://doi.org/10.4315/0362-028X.JFP-14-059.Grant, K. R., R. K.Gallardo, and J. J.McCluskey. 2021. Consumer preferences for foods with clean labels and new food technologies. Agribusiness. 37:764–781. https://doi.org/10.1002/agr.21705.Hartzell, A., C.Chen, C.Lewis, K.Liu, S.Reynolds, and E. G.Dudley. 2011. Escherichia coli O157:H7 of genotype lineage-specific polymorphism assay 211111 and clade 8 are common clinical isolates within Pennsylvania. Foodborne Pathog. Dis. 8:763–768. https://doi.org/10.1089/fpd.2010.0762.Hernández, J. D., A.Castell, N.Arroyo-Manzanares, I.Guillén, P.Vizcaíno, I.López-García, M.Hernández-Córdoba, and P.Viñas. 2021. Toward nitrite-free curing: evaluation of a new approach to distinguish real uncured meat from cured meat made with nitrite. Foods. 10:313. https://doi.org/10.3390/foods10020313.Iqbal, J., D.Yu, M.Zubair, M. I.Rasheed, H. M. U.Khizar, and M.Imran. 2021. Health consciousness, food safety concern, and consumer purchase intentions toward organic food: the role of consumer involvement and ecological motives. SAGE Open. 11:215824402110157. https://doi.org/10.1177/21582440211015727.Jackson, A. L., C.Kulchaiyawat, G. A.Sullivan, J. G.Sebranek, and J. S.Dickson. 2011. Use of natural ingredients to control growth of Clostridium perfringens in naturally cured frankfurters and hams. J. Food Protect. 74:417–424. https://doi.org/10.4315/0362-028x.jfp-10-365.Kalaycıoğlu, Z., and F. B.Erim. 2019. Nitrate and nitrites in foods: worldwide regional distribution in view of their risks and benefits. J. Agr. Food Chem. 67:7205–7222. https://doi.org/10.1021/acs.jafc.9b01194.Leistner, L., and L. G. M.Gorris. 1995. Food preservation by hurdle technology. Trends Food Sci. Tech. 6:41–46. https://doi.org/10.1016/s0924-2244(00)88941-4.Maddock, R.2015. US products: dry sausage. In: F.Toldŕa, Y. H.Hui, I.Astiasarán, J. G.Sebranek, and R.Talon, editors, Handbook of fermented meat and poultry. Wiley Blackwell, Hoboken, NJ. p. 295–298.McKinney, S., C.Cutter, and J.Campbell. 2017. Pathogen reductions during traditional fermentation and drying of pork salamis. Meat and Muscle Biology. 1. https://doi.org/10.22175/rmc2017.122.Milkowski, A., H. K.Garg, J. R.Coughlin, and N. S.Bryan. 2010. Nutritional epidemiology in the context of nitric oxide biology: a risk–benefit evaluation for dietary nitrite and nitrate. Nitric Oxide. 22:110–119. https://doi.org/10.1016/j.niox.2009.08.004.Ngutter, C., and C.Donnelly. 2003. Nitrite-induced injury of Listeria monocytogenes and the effect of selective versus nonselective recovery procedures on its isolation from frankfurters. J. Food Protect. 66:2252–2257. https://doi.org/10.4315/0362-028x-66.12.2252.Nickelson, R., II, J.Luchansky, C.Kaspar, and E.Johnson. 1996. Update on dry fermented sausage Escherichia coli O157:H7 validation research. Research report no. 11-316. National Cattlemen's Beef Association, Chicago, IL.Nielsen, J. W., J. S.Dickson, and J. D.Crouse. 1990. Use of a bacteriocin produced by Pediococcus acidilactici to inhibit Listeria monocytogenes associated with fresh meat. Appl. Environ. Microb. 56:2142–2145. https://doi.org/10.1128/aem.56.7.2142-2145.1990.Rivera, N., M.Bunning, and J.Martin. 2019. Uncured-labeled meat products produced using plant-derived nitrates and nitrites: chemistry, safety, and regulatory considerations. J. Agr. Food Chem. 67:8074–8084. https://doi.org/10.1021/acs.jafc.9b01826.Sebranek, J. G., A. L.Jackson-Davis, K. L.Myers, and N. A.Lavieri. 2012. Beyond celery and starter culture: advances in natural/organic curing processes in the United States. Meat Sci. 92:267–273. https://doi.org/10.1016/j.meatsci.2012.03.002.Sebranek, J. G., and J. N.Bacus. 2007. Cured meat products without direct addition of nitrate or nitrite: what are the issues?Meat Sci. 77:136–147. https://doi.org/10.1016/j.meatsci.2007.03.025.Sindelar, J. J., and A. L.Milkowski. 2011. Sodium nitrite in processed meat and poultry meats: a review of curing and examining the risk/benefit of its use. American Meat Science Association White Papers. https://meatscience.org/docs/default-source/publications-resources/white-papers/2011-11-amsa-nitrite-white-paper.pdf?sfvrsn=4232bbb3_8. (Accessed January 18, 2022).Sullivan, G. A., A. L.Jackson-Davis, S. E.Niebuhr, Y.Xi, K. D.Schrader, J. G.Sebranek, and J. S.Dickson. 2012. Inhibition of Listeria monocytogenes using natural antimicrobials in no-nitrate-or-nitrite-added ham. J. Food Protect. 75:1071–1076. https://doi.org/10.4315/0362-028x.jfp-11-511.Toldrá, F.2002. Dry-cured meat products. Food & Nutrition Press, Inc., Trumbull, CT.Tompkin, R. B.2005. Antimicrobials in food. In: P. M.Davidson, J. N.Sofos, and A. L.Branen, editors, Antimicrobials in food, Taylor & Francis, CRC Press, Boca Raton, FL.Työppönen, S., E.Petäjä, and T.Mattila-Sandholm. 2002. Bioprotectives and probiotics for dry sausages. Int. J. Food Microbiol. 83:233–244. https:/doi.org/10.1016/s0168-1605(02)00379-3.US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 1995. Processing inspectors’ calculations handbook. https://fsis.usda.gov/policy/fsis-directives/7620.3. (Accessed September 30, 2022).US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 2001. Performance standards for the production of processed meat and poultry products. Federal Register. https://federalregister.gov/documents/2001/02/27/01-4420/performance-standards-for-the-production-of-processed-meat-and-poultry-products. (Accessed September 30, 2022).US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 2012. Microbiological laboratory guidebook. https://fsis.usda.gov/news-events/publications/microbiology-laboratory-guidebook. (Accessed September 30, 2022).US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 2021. Fratelli Beretta USA, Inc. recalls ready-to-eat uncured antipasto meat products due to possible Salmonella contamination. https://fsis.usda.gov/recalls-alerts/fratelli-beretta-usa-inc.-recalls-ready-eat-uncured-antipasto-meat-products-due. (Accessed September 30, 2022).US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 2022. Salmonella outbreaks linked to Italian style meats: outbreak investigation after action review. https://fsis.usda.gov/sites/default/files/media_file/2022-04/FSIS-After-Action-Review-2021-09_2022-01.pdf. (Accessed September 30, 2022).US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 2023a. FSIS ready-to-eat fermented, salt-cured, and dried products guideline. https://fsis.usda.gov/guidelines/2023-0002. (Accessed September 30, 2022).US Department of Agriculture, Food Safety and Inspection Service (USDA-FSIS). 2023b. Daniele International LLC recalls ready-to-eat sausage products due to possible Listeria contamination. https://fsis.usda.gov/recalls-alerts/daniele-international-llc-recalls-ready-eat-sausage-products-due-possible-listeria. (Accessed September 30, 2022).Watson, S. C., C. N.Cutter, and J. A.Campbell. 2021. Fate of Salmonella spp., Listeria monocytogenes, and Campylobacter spp. during fermentation and drying of duck salami. Food Protection Trends. 41:423–430. https://doi.org/10.4315/1541-9576-41.4.423.World Health Organization, International Agency for Research on Cancer (WHO-IARC). 2015. IARC monographs evaluate consumption of red meat and processed meat. https://iarc.who.int/wp-content/uploads/2018/07/pr240_E.pdf. (Accessed September 30, 2022).Yarbrough, J. M., J. B.Rake, and R. G.Eagon. 1980. Bacterial inhibitory effects of nitrite: inhibition of active transport, but not of group translocation, and of intracellular enzymes. Appl. Environ. Microb. 39:831–834. https://doi.org/10.1128/aem.39.4.831-834.1980.