Fate of Listeria monocytogenes, Salmonella spp., and Shiga Toxin-Producing Escherichia coli on Slices of an All-Beef Soppressata during Storage

Cells of Listeria monocytogenes, Salmonella spp., or Shiga toxin-producing Escherichia coli (STEC) were inoculated (ca. 4.0 log CFU/slice) onto slices (ca. 4 g each slice) of an all-beef soppressata (ca. pH 5.05 and aw 0.85). The storage of vacuum-sealed slices of inoculated soppressata at 4 °C or 20 °C for 90 days resulted in reductions of all three pathogens by ca. 2.2 to 3.1 or ca. ≥3.3 log CFU/slice, respectively. When pathogen levels decreased to below detection (≤1.18 log CFU/slice) by direct plating, it was possible to recover each of the target pathogens by enrichment, albeit more frequently from slices stored at 4 °C (p < 0.05) compared to 20 °C. In summary, the slices of the commercially produced beef soppressata selected for this study did not provide a favorable environment for either survival or outgrowth of surface-inoculated cells of L. monocytogenes, Salmonella spp., or STEC during storage.


Introduction
Prior to the availability of modern-day preservation technologies, such as refrigeration, freezing, pasteurization, high pressure processing, irradiation, and food-grade antimicrobials, which became mainstream only within the last 200 years or so, human beings were reliant on salting, drying, and/or fermentation to preserve meat [1,2]. Among the most popular of so-called "old-world-style" preserved meats are pork products from Southern Europe dating back to between 100 BC and 700 AD, such as dry-cured ham (e.g., prosciutto) and fermented salume/sausage (e.g., soppressata), respectively [2][3][4][5][6]. However, since pork is the primary protein species comprising these artisanal ready-to-eat (RTE) meats, such products cannot be universally enjoyed by all consumers, notably because some individuals adhere to a restrictive diet [7]. Thus, there has been considerable interest of late to utilize beef as the sole protein source in various specialty/ethnic meats that are traditionally made exclusively with pork. In addition to exploring and improving the taste and texture of artisanal meats made exclusively with beef, further research is also needed to establish and, if needed, to improve upon the safety of these products.
One example of a globally recognized, well-liked, Mediterranean-type salume is soppressata, a dry-cured, often flattened, salami made from coarsely ground leaner cuts of pork [8]. The process of preparing soppressata, namely fermentation and drying, is sufficient to both ensure safety and achieve shelf stability; however, on occasion, pathogens such

Inoculation of Soppressata Slices
Multiple packages (142 g each, single chub per package) from different production lots of a single brand of a commercially produced all-beef soppressata (Brand A) were purchased from an online vendor. The outside of each chub was wiped down with a paper towel moistened with 70% alcohol before the collagen casing was removed with the aid of a sterile scalpel. Next, each chub was sliced (Model 8512, Univex, Salem, NH, USA), and then, a single slice (ca. 3.9 ± 0.34 g each; ca. 3.5 cm diameter; ca. 0.1 cm thick) of soppressata was placed into a cast nylon and polyethylene copolymer bag containing vinyl acetate (Item #185-10; 3 mil standard barrier, 15.2 by 20.3 cm; Associated Bag, Milwaukee, WI, USA). Each face of each slice was separately inoculated with a 25-µL portion (50 µL total; ca. 4.0 log CFU/slice) of overnight-grown (ca. 18-20 h) rifampicin-resistant cells of each pathogen cocktail. The bags containing inoculated slices of soppressata were vacuumsealed to 950 mBar with an Ultravac 550 vacuum packaging unit (Koch Equipment, Kansas City, MO, USA) and then stored at 4 • C or 20 • C for 90 days. In each of the three trials, each of three bags containing an inoculated slice of soppressata was separately analyzed for viable cells of a target pathogen at each sampling interval (N = 3 trials; n = 3 inoculated slices per sampling interval).

Chemical Analyses
Proximate chemical analyses were conducted on a single ca. 400 g composite (representative) sample from each of the two batches/production lots for each beef salume product purchased online between May and August of 2022 (N = 2 trials of each brand of salume; n = 1 sample per each batch/lot of each product). Note, in addition to the brand of salume that we separately inoculated with the three pathogen cocktails (i.e., Brand A), for comparative purposes, we also elaborated the chemical composition of two additional all-beef salume products (Brand B and Brand C) that were not inoculated. Chemical analyses of all three brands of salume were conducted by an independent commercial testing facility using methods approved by the Association of Official Analytical Chemists [31] as follows: fat (AOAC 960.39), salt (AOAC 983.14), ash (AOAC 920.153), moisture (AOAC 950.46Bb), protein (AOAC 991.20.i), carbohydrates (by difference), sodium nitrite (AOAC 973.31), titratable acidity (AOAC 942.15), and water activity (AOAC 978.18).

Microbiological Analyses
Inoculated slices of soppressata were analyzed on days 0, 7, 15, 21, 28, 45, 60, 75, and/or 90 essentially as described [32]. In brief, after opening each bag with the aid of alcohol sterilized scissors, 15 mL of 0.1% sterile peptone water (Difco; Becton, Dickinson Company, Sparks, MD, USA) were added to each bag and the contents were massaged by hand for ca. 1 min. Portions of the resulting rinsate, with and without prior dilution in 0.1% peptone water, were plated onto sorbitol-MacConkey (SMAC; Difco), xylose-lysine-tergitol-4 (XLT-4; Difco), or modified Oxford (MOX; Difco) agar plates containing rifampicin (100 µg/mL) to recover surviving cells of STEC, Salmonella spp., or L. monocytogenes, respectively. The SMAC and XLT-4 agar plates were incubated at 37 • C for 24 h, whereas the MOX agar plates were incubated at 37 • C for 48 h. Colonies typical for each pathogen were enumerated and expressed as log CFU/slice. When pathogen levels decreased to below the detection limit (i.e., ≤1.18 log CFU/slice) by direct plating, pathogen presence was determined via enrichment as previously described [30]. In addition, at each sampling point, the pH of the rinsate was measured using a model 6000P pH/temperature electrode and a model 5500 pH meter (Daigger, Vernon Hills, IL, USA). As received from the online vendor, the 400 g composite salumi samples (see preceding section) were also separately analyzed for the presence/absence of cells of STEC (serogroups O26, O45, O103, O104, O111, O121, O145 (AOAC RI 09130) and serotype O157:H7 (AOAC RI 031002)), L. monocytogenes (AOAC 2003.12), and Salmonella spp. (AOAC 2003.09). The initial total aerobic plate count (APC; AOAC 966.23) and initial levels of lactic acid bacteria (LAB [33]) were also enumerated for each brand of salume.

Statistical Methods
Values for log CFU and pH values were analyzed by fitting a Pathogen (3) × Temperature (2) × Day (9) ANOVA model. The heterogeneity of within-treatment variability for these 54 treatments was incorporated into the model by assigning each treatment to a group. Treatments assigned to the same group had variances within four-times magnitude of all treatments in that group. Groups were specified as a random effect to ensure that the appropriate magnitude of within-treatment variability was associated with each treatment for all statistical tests and comparisons obtained from the ANOVA models. For proximate composition, one-way Sample (3) ANOVA models were fit for each test. All statistical analyses were conducted using SAS PROC MIXED [34] software, with α = 0.05 LSD pairwise means comparisons [35] obtained using SLICE = options in the LSMEANS statement.

Results and Discussion
As elaborated by others [36,37], meat preservation has transitioned from a necessity for survival/subsistence in ancient times to a luxury marked by convenience and unique sensory attributes in more recent times. One of the best examples of a modern-day indulgence is consumption of charcuterie trays comprised of artisanal meats [38][39][40]. The longstanding demand for high-quality and locally sourced artisanal meats and, in more recent years, especially for all-beef products formulated to accommodate individuals on a restrictive diet, have been the catalyst for the ongoing development of new types, tastes, and textures of meat products for consumers to enjoy. It has also served as a justification to scientifically validate the safety and to certify the shelf life of such products. As detailed elsewhere [41][42][43][44][45][46][47][48][49][50][51][52][53][54], our group and other investigators have quantified the potential for various all-beef meats that are dried, salted, cured, cooked, and/or fermented, inclusive of breakfast meats, savory snacks, and salted/dried and/or fermented products, to support the viability or outgrowth of various foodborne bacterial pathogens. Regardless of product categories and characteristics, the physicality and logistics of slicing and packaging, as well as extensive and unavoidable hand manipulation during the preparation and assembly of artisanal foods on charcuterie trays, provide ample opportunities for the introduction of food-borne pathogens onto the surface of RTE artisanal meats [55]. Such conditions also provide adequate justification for collecting additional data related to the occurrence, levels, and control of target pathogens on niche meat products that are fermented/dried, inclusive of all-beef products. Thus, the primary objective of the present study was to determine if an all-beef soppressata would allow for survival or support the (out)growth of L. monocytogenes, Salmonella spp., or STEC during storage.
To our knowledge, there have been no data published on the chemical composition of soppressata prepared with beef as the sole protein source. Thus, as one objective of this study, we elaborated the proximate composition of a commercial all-beef soppressata (Brand A) that was subsequently sliced and inoculated with cells of L. monocytogenes, Salmonella spp., or STEC (Table 2). For comparative purposes, we also quantified the proximate composition of two additional all-beef salumi products (i.e., Brand B and Brand C) that were also purchased online (Table 2). Chemical analyses revealed no significant (p > 0.05) differences in acidity, carbohydrates, fat, moisture, nitrite, protein, or pH levels among Brand A, Brand B, and Brand C. In contrast, chemical analyses revealed Brand C salume as discernibly (p < 0.05) different from Brand A and Brand B products with respect to salt content. Significant differences (p < 0.05) in ash levels among all three brands were also observed. Lastly, chemical analyses revealed that Brand C was similar (p > 0.05) to Brand A with respect to a w , but it had a significantly (p < 0.05) higher a w when compared to Brand B salume. Note, no differences (p > 0.05) were observed between Brand A and Brand B products with respect to a w . In general, on average, the three brands of commercial all-beef salami displayed a slightly low pH (ca. pH 5.0) and low water activity value (ca. a w 0.86), an intermediate moisture level (ca. 36%), and a relatively high fat (ca. 27%) and high salt (ca. 4.9%) content ( Table 2). It should also be noted that there were no significant changes (p > 0.05) in pH of the inoculated slices of the allbeef soppressata (Brand A) after 90 days of storage at 4 • C or 20 • C; the initial pH was pH 4.97 ± 0.09, and regardless of the storage temperature, the final pH of the product after 90 days of storage was pH 5.15 ± 0.03. In comparison to compositional data for the all-beef soppressata analyzed herein, Liguori et al. [56] reported on the moisture (ca. 40%), protein (ca. 32%), and fat (ca. 25%) levels of a traditional pork-based soppressata prepared at an industrial establishment. As another example, Jung and colleagues [57] analyzed salchichón, a soppressata-like, dry-cured pork salume originating from Spain that displayed the following attributes after a 45-day ripening period: moisture (ca. 24%), protein (ca. 44%), ash (ca. 4.7%), fat (ca. 26%), a w (ca. a w 0.80), and pH (ca. pH 6.4). Romeo et al. [58] evaluated the chemical composition of typical Calabrian cured meats made from pork, including soppressata, and reported levels of about 4.1% ash, 27.9% fat, and 29.3% protein, with a water activity level of a w 0.835 and ca. pH 5.6. Lastly, Coppola and colleagues [59] reported levels of about 35.8% moisture, 6.2% ash, 39% protein, 17.2% fat, and ca. pH 6.0 for an all-pork soppressata, as traditionally produced in Southern Italy. The above-mentioned peer-reviewed compositional data for soppressata-type salumi prepared with pork compare favorably with the data reported herein for a soppressata-type salume prepared exclusively with beef. Retrospectively, and based solely on its chemical composition (Table 2), it is not likely that Brand A soppressata would support (appreciable) outgrowth of any of the three pathogens tested herein during cold storage for 90 days. In support of this statement and depending on the menstruum/substrate and temperature evaluated, L. monocytogenes can grow at salt levels of up to ca. 6.5% and at ≥pH 4.4, whereas most strains will not grow at ≤a w 0.93 [60]. Salmonella can also grow at pH 4.0 to pH 4.5, but most strains are not able to grow at salt levels of ≥2.5% or at water activity values of ca. ≤a w 0.93 when held at refrigeration temperatures [61,62]. For STEC, the reported minimum conditions for growth are pH 4.0 to pH 4.5, ≤a w 0.89, and salt levels of ≤6.5% [61,[63][64][65]. For point of reference, Brand A soppressata displayed a salt content of 5.43%, a water activity level of a w 0.85, and an acidity/basicity factor of pH 5.05 (Table 2).
We also conducted a microbiological profile of the three commercial brands of salume as received directly from the attendant online vendors (Table 3). A single soppressata sample from each of two trials/batches from each of the three brands of all-beef salami all tested negative (<0.04 cells/g) by enrichment for L. monocytogenes, Salmonella spp., and STEC. In addition, significantly (p < 0.05) higher levels of LAB were observed for Brand A and Brand C salami compared to Brand B salume; levels of LAB ranged from <1 cell/g for Brand B to ca. 5.1 log CFU/g for Brand A and Brand C salami (Table 3). Although it remains unclear why levels of LAB were appreciably lower for Brand B salume, it is possible that factors such as product age (e.g., use by date), differences in levels and types of autochthonous LAB on raw materials, and/or differences in processing parameters could collectively account for such divergences. Variations were also observed among brands related to results for APC, with appreciably higher levels (p < 0.05) observed for Brand A when compared to Brand B salume; however, no significant (p > 0.05) differences in APC levels were observed between Brand C compared with Brand A or Brand B products; levels of APC ranged from ca. 1.2 to 5.5 log CFU/g (Table 3). It has been well established that L. monocytogenes, Salmonella spp., and STEC are recoverable from a variety of Italian-type sausage. For example, De Cesare and collaborators [66] recovered cells of L. monocytogenes from ca. 15% (36 of 237 samples) of fermented sausage produced in Northern Italy; pathogen levels in positive samples were estimated at <10 MPN/g. Likewise, Quaglia et al. [67] recovered cells of L. monocytogenes from ca. 6.5% (2 of 31 samples) of pork soppressata samples produced in Southern Italy; pathogen levels in positive samples were estimated at ca. 1.5 to 1.8 log CFU/g. Note that these same soppressata samples also tested negative for cells of serotype O157:H7 and serotype O26 strains of E. coli. In another study, Hussein and Bollinger [68] summarized that the prevalence of serotype O157:H7 and non-O157:H7 cells of STEC in raw and fresh sausage worldwide, inclusive of beef sausage, ranged from 0.1 to 4.4% and 17 to 49.2%, respectively; pathogen levels in positive samples were not reported. Regarding Salmonella spp., Levine et al. [69] reported a 3-year cumulative prevalence of 1.43% (10 of 698 samples) for dry and semi-dry sausage collected in the late-1990s in the US from ca. 1800 federally inspected facilities; pathogen levels in positive samples were not reported. As another example, Cabedo and colleagues [70] reported the recovery of cells of Salmonella spp. from 9 of 81 (11.1%) cured/dried pork sausage samples collected between 1998 and 2004 from retail stores or from food processing facilities in Spain; pathogen levels in positive samples were not reported. It has also been well established that cells of L. monocytogenes, Salmonella spp., and STEC have, on occasion, caused human illness from consumption of fermented sausage when/if contaminated with these pathogens by whatever means or routes available. As such, it seems probable that cells of these pathogens, if present on the raw materials or when associated with any of the ingredients used to prepare salami, might remain on the finished product and/or that post-process contamination would exacerbate the risk of food-borne illness. Therefore, from a public health perspective, it is significant that slices of the commercial all-beef soppressata tested herein did not provide a favorable environment for the outgrowth or survival of L. monocytogenes, Salmonella spp., or STEC during extended storage at 4 • C or 20 • C: levels of all three pathogens were reduced by ca. 2.2 to 3.3 log CFU/slice within 90 days (Figures 1 and 2). Error bars represent the standard error of the mean (N = 3, n = 3). The dotted line (-------) denotes the detection limit of ≤1.18 log CFU/slice. When L. monocytogenes, Salmonella spp., or STEC numbers were below the detection limit, no cells of any of the three pathogens tested were recovered via enrichment. For a given storage day, bars/means that have no uppercase letter in common denote statistical differences (α = 0.0001) among pathogens.
When inoculated slices of soppressata were stored at 4 • C, appreciable differences (p > 0.05) in pathogen recovery via direct plating were not observed; initial levels of all three pathogens decreased to ca. 1.4 log CFU/slice after 90 days (Figure 1). However, due to the lower (p < 0.05) initial levels of Salmonella spp. (3.6 log CFU/g) used to the inoculate slices of soppressata compared to the initial levels of L. monocytogenes or STEC (ca. 4.5 log CFU/slice), it was not possible to quantify a 3.0-log reduction for Salmonella by direct plating (detection limit of ≤1.18 log CFU/slice). Regardless, all three pathogens were recovered by direct plating from at least some slices of soppressata at each sampling interval during storage at 4 • C. Overall, although only subtle differences in the levels of the three pathogens were observed during refrigerated storage, there were no significant (p > 0.05) differences in levels of Salmonella, L. monocytogenes, or STEC after 90 days of storage at 4 • C (Figure 1). That being said, when levels of L. monocytogenes decreased to below the detection limit by direct plating beginning on day 75 (6 of 9 samples negative by direct plating) but also on day 90 (1 of 9 samples negative by direct plating), all 7 of these samples testing negative by direct plating subsequently tested negative by enrichment as well (Table 4). For Salmonella spp., when pathogen levels on some slices of soppressata decreased to below detection by direct plating beginning on day 60 (1 sample negative by direct plating) but also on day 75 (2 samples negative by direct plating) and on day 90 (4 samples negative by direct plating), 4 of these 7 (57%) samples testing negative by direct plating subsequently tested positive by enrichment (Table 4). Likewise, for STEC, when pathogen levels on some slices of soppressata tested negative by direct plating beginning on day 60 (1 sample negative by direct plating) but also on day 75 (6 samples negative by direct plating) and on day 90 (4 samples negative by direct plating), 4 of these 11 (36%) samples testing negative by direct plating subsequently tested positive by enrichment ( Table 4). denotes the detection limit of ≤1.18 log CFU/slice. When L. monocytogenes, Salmonella spp., or STEC numbers were below detection limit, no cells of any of the three pathogens tested were recovered via enrichment. For a given storage day, bars/means that have no uppercase letter in common denote statistical differences (α = 0.0001) among pathogens.
There have been only a handful of recalls and illnesses attributed to soppressatatype products, presumably because the intrinsic attributes of such products typically do not provide a favorable environment for persistence or proliferation of L. monocytogenes, Salmonella spp., or STEC over the anticipated refrigerated shelf life [47]. It has also been well established that fermentation and drying alone will typically deliver only a ≤2.0 log reduction in levels of L. monocytogenes, Salmonella spp., or STEC [71][72][73][74][75][76]. Numerous publications have detailed the occurrence and fate of L. monocytogenes, Salmonella spp., or STEC during the manufacture and storage of Italian-type, dry-cured salume containing pork, but relatively few studies are specific to soppressata. As one example of the latter, Luchansky and colleagues [47] monitored the fate of L. monocytogenes or STEC in soupie, an old-world-style, specialty/ethnic pork soppressata, during fermentation, drying, and storage at 20 • C; reductions of ca. ≥5.0 log in levels of both pathogens were observed after 4 and 1 month, respectively, during storage. Regarding the fate of Salmonella during production of Italian salami, Bonilauri and colleagues [77] reported reductions ranging from ca. 1.0 to 4.5 log CFU/g.
The literature cited herein confirms that a traditional pork-based soppressata does not support survival of L. monocytogenes, Salmonella spp., or STEC during manufacture or storage. Similarly, the present study provides the first scientifically sound data confirming that an otherwise similar soppressata made from beef rather than pork does not provide a favorable environment for viability or persistence of these same three pathogens during storage. More specifically, storage of inoculated slices of beef soppressata at 4 • C or 20 • C resulted in reductions of 2.2 to 3.3 log CFU/slice in levels of L. monocytogenes, Salmonella spp., or STEC. Our data also established that when levels of L. monocytogenes, Salmonella spp., or STEC decreased to below detection by direct plating (≤1.18 log CFU/slice), it was more likely on occasion to separately recover all three target pathogens by enrichment from vacuum-packaged slices of beef soppressata following storage at 4 • C (p < 0.05) compared to 20 • C. Nonetheless, our validation of the wholesomeness of (slices of) an all-beef soppressata confirms the safety and broadens the market for this category of drycured, Italian-type salumi and interpolates the food preferences of people adhering to a restrictive diet. Funding: This research received no external funding.

Data Availability Statement:
The data presented in this study are available upon request from the corresponding author. These data are not publicly available due to the proprietary nature of the source/formulation of the substrates evaluated.

Conflicts of Interest:
The authors declare no conflict of interest. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture (USDA). The USDA is an equal opportunity provider and employer.