Buffered Dry Vinegar and Fermented Onion Juice Enhance Quality and Shelf Life of Fresh Pork Sausage

Sausage ingredients and processing

Fresh pork shoulders and butts from 24 h postmortem pork carcasses were purchased from a regional pork supplier (BOE Farms, Moselle, MS, USA) to manufacture the 4 treatments of breakfast sausage. The study employed a randomized complete block design with 3 replications to evaluate the impact of antimicrobial ingredient technology on the spoilage and shelf life of fresh pork sausage. For each replication, 4 picnic shoulders and 2 Boston butt cuts were randomly selected, yielding approximately 30 kg of meat (22.5 kg pork picnic shoulder and 7.5 kg Boston butt). Following deboning, the meat was then placed into a grinder (Hollymatic Model 300C) and ground through 2.54-cm and 0.95-cm plates to achieve the desired particle size. The ground meat averaged 15.8% ± 0.8% protein, 15.2% ± 1.7% fat, 62.6% ± 1.3% moisture, and 1.7% ± 0.9% collagen, as measured using a near-infrared spectrometer (Food Scan Lab Analyzer, Model 7880, Foss Analytical, Eden Prairie, MN, USA). The ground meat from each replication was further divided into 4 equal batches of 7 kg each, which was randomly assigned to one of 4 treatments: (1) control with no antimicrobial addition (CONTROL), (2) fermented onion juice powder (FOJ) treatment (0.5%, w/w; reFRESH 652; Third Wave Bioactives, Wauwatosa, WI, USA), (3) buffered dry vinegar (BDV) treatment (0.5%, w/w; e(Lm)inate^® V-Dry; Hawkins, Inc. Roseville MN, USA), or (4) a combination (CMB) treatment consisting of both FOJ (0.25%, w/w) and BDV (0.25%, w/w). Each treatment batch was formulated with one of the antimicrobial treatments above, 94% pork, a commercial sausage irradiated seasoning blend (2%, w/w; A.C. LEGG INC, Calera, AL, USA), an antioxidant blend (natural extracts from acerola cherry, green tea, and rosemary, 0.2%, w/w; Hawkins Inc., Roseville, MN, USA), and iced water (3%, w/w). The sausage meat and ingredients were blended for 2 min and then ground through a 4-mm plate and stuffed (Omet Foodtech Continuous Vacuum Filler Model F71: Waltons, KS, USA) into cellulose casings. The final products (2 tubes/treatment) were stored in a freezer (−17.5°C).

Patty preparation

Two days after sausage preparation, Replication 1 was sliced; Replication 2 and Replication 3 were sliced 2 and 4 weeks later, respectively, with all tubes remaining frozen at −17.5°C until their designated processing date. For each replication, 2 sausage tubes were sliced into patties that were 1.27 cm thick and 6 cm in diameter patties and placed into white foam meat trays (21 cm × 15 cm × 1 cm; WebstaurantStore, Lititz, PA, USA), overwrapped with oxygen-permeable PVC film (O₂ Transmission Rate (OTR) =  121 cm³/100 cm²/day, Moisture Vapor Transmission Rate (MVTR) =  2.17 g/cm²/day, Premium-LT 80718; LINPAC Packaging-Filmco Inc., Aurora, OH, USA)), and placed into a meat case (SC-M32E-4-LS Howard/McCray: Philadelphia, PA, USA) at 4 ± 2 °C to thaw out overnight. Samples were stored under fluorescent lighting (F32T8/SPX35/ECO2: Mesquite, TX) at 800–1000 lux to mimic retail conditions. On each day of sampling, 3 trays from each treatment were randomly selected and used for microbiological, instrumental, oxidation, and sensory analysis.

Microbiological analysis

For each treatment (n = 3/treatment/replication), 10 g of meat from a randomly selected sausage patty was aseptically transferred to a Nasco sterile Whirl-Pak bag with an 80 μm pore-sized mesh screen liner (Whirl-Pak, Fort Atkinson, WI, USA), followed by the addition of 90 g of buffered peptone water (Oxoid, Basingstoke, Hants, UK) and 0.2% Polysorbate-20 (Fisher Scientific, Hampton, NH, USA). Each bag was homogenized using a peristaltic stomacher (Neutec Group Inc., Masticator: Barcelona, Spain) at maximum speed for 30s.

The filtrated samples were diluted to obtain readable bacterial counts and plated onto the following: total aerobic bacteria counts (APC) Petrifilm (Petrifilm^® Aerobic Count Plates, Lansing, MI, USA); lactic acid bacteria counts (LAB) Petrifilm (Petrifilm^® Lactic Acid Bacteria Count Plates, Lansing, MI, USA); and Yeast and Mold Counts (Y+M) Petrifilm (Petrifilm^® Yeast and Mold Count Plates, Lansing, MI, USA). For mesophilic bacteria enumeration, APC and LAB Petrifilms were incubated at 30°C for 48 h. For the psychrotrophic bacterial population, a separate set of APC Petrifilms was incubated at 15°C for 120 h. For yeast and mold growth, Y+M Petrifilms were incubated at 15°C for 72 h for yeast counts and 120 h for mold counts. Yeast colonies were differentiated by their small size, defined edges, uniform color ranging from pink-tan to blue-green, raised appearance, and lack of central focus, whereas mold colonies were characterized by their larger size, diffused edges, varied pigment (brown, beige, orange, blue-green), flat appearance, and distinct central focus.

Instrumental analysis

Instrumental measurements of color, pH, and water activity were conducted for each treatment at 4 time points (day 1, day 4, day 7, and day 9), with 3 patties analyzed per treatment at each time point. Color measurements were performed with a spectrophotometer (CM-600d, Konica Minolta Inc., Tokyo, Japan) using standard illuminant D₆₅ (daylight, color temperature: 6505K), an 8° viewing angle, and an 8 mm port. Colors were expressed according to the CIE (Commission Internationale de l’Eclairage) system as L* (lightness), a* (redness), and b* (yellowness). Calibration was conducted using a white calibration cap (CM-A177, Konica Minolta Inc., Tokyo, Japan) before each analysis. The center of each sausage patty was used for each measurement. The pH was measured using a pH meter (HI 98163, Hannah Instruments Inc., Woonsocket, RI, USA) by inserting a penetration probe (FC2323, Hanna Instruments Inc., Woonsocket, RI, USA) into the center of each patty, ensuring the probe was fully covered by the sausage. Before sampling, the pH probe was standardized using calibration buffers at pH 4 and pH 7. Water activity was determined using a calibrated water activity meter (Aqualab 4TE: Pullman, WA, USA). The machine was calibrated using a verification standard with a dew point of 0.984 ± 0.003. Samples were taken from each patty and placed into disposable sample containers before each measurement.

Thiobarbituric acid reactive substances assessment

A thiobarbituric acid reactive substance (TBARS) method was used to quantify lipid oxidation products. At each sampling day (n = 5), 3 patties from each treatment were individually placed into 50 mL centrifuge tubes, flash frozen in liquid nitrogen, vacuum sealed, and stored at −18°C until analysis. The TBARS measurements were conducted following the Cayman Chemical TBARS Assay Kit protocol (Cayman Chemical, Ann Arbor, MI) with modifications. Samples were homogenized, reacted with TBA reagent, heated, cooled, centrifuged, and analyzed at 535 nm using a microplate reader (Biotek Synergy^™ HTX Multi-Mode Microplate Reader, Raleigh, NC, USA). MDA concentrations were calculated using a standard curve.

Descriptive sensory evaluation

Before evaluation, 8 panelists completed eight one-hour training sessions focused on lexicon development and calibration of sensory descriptors. These sessions were led by 2 sensory experts with over 20 years of combined experience evaluating meat products. During training, panelists practiced using the 15-cm line scale to ensure consistent and standardized scoring. Descriptive sensory tests for raw appearance and aroma evaluation were conducted on days 1, 4, 7, and 9 across 3 replications and included 4 to 8 panelists. For each panel, one tray from each treatment was randomly retrieved from the meat case and assigned a random three-digit code. Trays were placed on a white background under LED lights for visual evaluation. Each tray contained 3 sausage patties, and a quarter of one sausage patty per treatment was placed into a vial for olfactory assessment. Panelists evaluated aroma attributes (sourness, oxidation, rancidity, spoilage, and off-notes) and appearance attributes (darkness and sliminess) using a 15-cm line scale (0 = no intensity and 15 = extreme intensity) (Civille and Carr, 2015). Degree of difference (DOD) for both raw aroma and appearance was rated on a 0-5 scale to assess how each sample deviated from a typical fresh breakfast sausage, where 0 indicated no perceptible difference, and 5 represented extreme differences.

For sensory analysis of cooked breakfast sausage, one tray from each treatment was retrieved from the meat case and randomly assigned a three-digit code. Samples were then cooked uncovered in a convection oven (Oster Toaster Oven Model 6293, Wickliffe, OH) at 204°C for 15 min, flipped halfway through, to achieve an internal temperature of 74°C. The cooked sausage patties were then cut into quarters and served to 4 to 8 panelists. Panelists evaluated aroma and flavor attributes using the same 15-cm scale and DOD scale. Panelists also classified samples as fresh, borderline, or spoiled.

Statistical analysis

A randomized complete block design with 3 replications (n = 3) was used to evaluate treatment differences (P < 0.05) in pH, raw color, and water activity. For these variables, the replication-by-treatment (rep × trt) interaction was used as the error term to account for random variation between replications and within-subsample variability. A randomized complete block design with 3 replications (n = 3) was used to evaluate treatment differences (P < 0.05) with respect to microbiological analysis, lipid oxidation, and descriptive sensory attributes, including raw aroma and appearance, as well as cooked aroma and flavor. PROC MIXED (Statistical Analysis Software, Institute, Cary, NC, USA) was used to analyze the effects of treatment and storage day, with replication treated as a random factor. When significant differences among treatments were observed (P < 0.05), Tukey’s HSD test was used to separate means.

Microbiological analysis

On day 1, no differences (P > 0.05) were observed among treatments, with initial counts of approximately 4 Log CFU/g for aerobic and psychrotrophic bacteria and 3 Log CFU/g for LAB and yeast.

On day 4, aerobic bacterial counts for CONTROL and FOJ treatments exceeded 6 Log CFU/g, while BDV and CMB treatments maintained fewer (P < 0.05) counts at between 4.0 and 5.0 Log CFU/g. Similarly, psychrotrophic bacteria counts reached 6-7 Log CFU/g in the CONTROL and FOJ, which were greater (P < 0.05) than the counts of BDV and CMB treatments (approximately 5 Log CFU/g). LAB counts for the CONTROL, FOJ, and BDV increased (P < 0.05) from day 4 to 7, reaching 5–7 Log CFU/g, while CMB maintained stable LAB counts (3–5 Log CFU/g) through day 9 (P > 0.05).

On days 7 and 9 of storage, aerobic and psychrotrophic bacteria counts were lower (P < 0.05) in BDV and CMB treatments than in the CONTROL and FOJ treatments. By day 9, aerobic counts reached 8–9 Log CFU/g in CONTROL and FOJ but remained below 5–6 Log CFU/g in BDV and CMB. Using a spoilage threshold of 7.0 Log CFU/g (Stoops et al., 2015), CONTROL and FOJ were spoiled by day 7, whereas BDV and CMB had not reached spoilage levels by day 9.

Mold counts remained below 1 Log CFU/g throughout the 9-day storage period, with no differences (P = 0.851) among treatments (data not shown). Yeast counts increased over time (P < 0.0001), reaching approximately 5 Log CFU/g by day 4 and exceeding 6–7 Log CFU/g by days 7 and 9 with no differences (P = 0.957) among treatments.

Instrumental analysis

The pH values did not differ (P = 0.630) among treatments when averaged over days, ranging from 5.8 to 6.0 (Table 1). Similarly, no difference (P = 0.403) existed in water activity among treatments, with values ranging from 0.97 to 0.98. When averaged over treatments, pH (P = 0.190) and Aw values (P = 0.403) remained stable throughout storage. Redness (CIE a*) values decreased over storage time (P = 0.001), with a* values decreasing in the CONTROL and BDV treatments over the 9 days (Table 2). On day 1, all treatments showed similar a* values (P > 0.05). By day 9, CONTROL and BDV were less red (6.0 and 5.5, respectively; P < 0.05) than their day 1 samples; FOJ and CMB did not differ from their day 1 values (P > 0.05), though a* values still decreased to 5.6 and 5.2. L* and b* values did not differ among treatments or days (all P > 0.05), with L* values ranging from 46.4 to 51.3 and b* values ranging from 10.7 to 13.8 across all treatment and storage day combinations.

Figure 1.

Microbial counts (Log CFU/g) of fresh pork breakfast sausage patties formulated with traditional sausage ingredients (CONTROL), 0.5% fermented onion juice (FOJ), 0.5% buffered dry vinegar (BDV), and a combination of 0.25% FOJ and 0.25% BDV (CMB) over a 9-day storage period.

^a,b Means within each a given storage day and within the same bacterial count type without a common superscript letter are different (P < 0.05).

^A–C Means that across storage days and within the same bacterial count type, without a common superscript letter, are different (P < 0.05)

*P value due to day × treatment effect.

^†P value due to day effect.

^‡P value due to treatment effect.

SEM: standard error of the mean.

TBARS

No differences existed (P = 0.884) in malondialdehyde (MDA) concentrations among treatments when averaged over storage days (Table 3). MDA values ranged from 0.067 mg/kg (FOJ, day 0) to 0.277 mg/kg (BDV, day 9), with all values remaining below 0.5 mg/kg.

Descriptive sensory analysis of raw sausages

For raw appearance, CONTROL and CMB became darker (P < 0.05) from day 1 to 9, and BDV showed darkening (P < 0.05) by day 7 (Table 4). FOJ maintained its initial redness and did not darken significantly (P > 0.05). Sliminess increased over time (P < 0.0001) for all treatments, with no treatment differences (P = 0.734). By day 9, sliminess scores ranged from 3.5 to 3.6.

Raw aroma attributes (sourness, oxidized, rancid, off-notes, DOD) did not differ among treatments on day 1 (P > 0.05). Treatment effects emerged by day 7, with CONTROL scoring higher (P < 0.05) for oxidized, rancid, off-notes, and DOD than FOJ, BDV, and CMB. By day 9, sourness continued to increase for all treatments (P < 0.05) compared to day 7. On day 9, rancidity and DOD no longer differed (P > 0.05) among treatments, but CMB exhibited lower oxidized aroma and off-notes (P < 0.05) than the CONTROL.

Descriptive sensory analysis of cooked sausages

Cooked aroma attributes did not differ among treatments on days 1 and 4 (P > 0.05). By day 7, CONTROL exhibited higher (P < 0.05) rancidity than BDV and higher spoilage and DOD scores than BDV and CMB (Table 5). The CONTROL also showed higher off–notes than CMB (P < 0.05).

Flavor attributes did not differ among treatments on days 1 and 4 (P > 0.05). Oxidized flavor increased (P < 0.05) for all treatments by day 4 and continued to intensify through day 7 (Table 6). The CONTROL sausage increased in rancidity and DOD by day 4, indicating earlier spoilage–related flavor changes.

By day 7, FOJ was more sour (P < 0.05) than the CONTROL, exhibited more intense off-notes (P < 0.05) than CMB, and a greater (P < 0.05) DOD than BDV. FOJ sausage also increased in off–flavor intensity over time more than any other treatment. The CMB sausage maintained lower off–note intensities (P < 0.05) than the CONTROL, indicating improved flavor stability. Treatments containing BDV (BDV and CMB) were similar to the CONTROL with respect to flavor attributes.

Mechanism of acetic acid in buffered dry vinegar

Buffered vinegar, comprised primarily of acetic acid, sodium acetate, and water, demonstrated potent antimicrobial properties in meat applications (Sengun et al., 2021). The antimicrobial mechanism of buffered vinegar relies on the ability of undissociated acetic acid to penetrate bacterial cell membranes rather than the disruption of the cell membranes (Tan et al., 2015). Once inside the cell, acetic acid dissociates at the more neutral intracellular pH, releasing protons that acidify the cytoplasm and disrupt cellular processes. This acidification inhibits key metabolic enzymes, disrupts the proton motive force across the cell membrane, and depletes cellular energy (ATP) as bacteria attempt to maintain intracellular pH homeostasis (Theron and Lues, 2007).

The BDV treatment demonstrated particularly strong effectiveness against aerobic and psychrotrophic bacteria, which is consistent with the findings of Crist et al. (2014), who demonstrated that a 2.5% vinegar and sodium lactate mixture delayed microbial spoilage in fresh Italian pork sausage, maintaining total plate counts approximately 1 Log CFU/g less than the control and extending shelf life by approximately 4 days. Similarly, Reed and Kumar (2018) reported that 0.7% vinegar in combination with 1% Jasmine tea extract extended the shelf life of fresh chicken sausage to 27 days under refrigerated storage, with aerobic plate counts remaining below 6 Log CFU/g, compared to control samples reaching spoilage levels (>7 Log CFU/g) by day 11. Overall, the limited shelf life (7–9 days) observed in the tested sausages may be attributed to the relatively high initial bacterial load of raw meat and the proliferation of yeasts. This contrasts with studies reporting roughly 2 weeks of shelf life when pre-rigor meat and effective antioxidants, such as rosemary and green tea extracts, were incorporated (Schilling et al., 2018; Schilling et al., 2019).

The buffering of vinegar is critical for maintaining the pH of the sausage above the isoelectric point of myofibrillar muscle proteins, preventing negative effects on texture and water-holding capacity (Kim et al., 2014). The stable pH values observed in this study confirm that BDV maintained antimicrobial efficacy without compromising meat chemistry. However, BDV did not inhibit yeast growth, consistent with reports that yeasts tolerate organic acids due to robust cell wall structures and active acid efflux mechanisms (Jones et al., 2019; Laranjo et al., 2019). The resistance of yeasts to organic acids relates to their more robust cell wall structure (containing β-1,3-glucans and chitin cross-linking), their ability to actively export organic acids via plasma membrane ATPases, and their capacity to tolerate acidic environments through expression of specialized proteins (Frousnoon et al., 2025). BDV also did not prevent color degradation or reduce lipid oxidation, which is expected given its limited antioxidant capacity (Bradley et al., 2011). This finding differs from some lactate-containing antimicrobial systems, which have been shown to preserve color through metmyoglobin-reducing activity (Kim et al., 2006). Bradley et al. (2011) noted that acetic acid functions predominantly as an antimicrobial rather than an antioxidant, which explains the lack of TBARS reduction observed in this study. Despite these limitations, BDV improved cooked sensory attributes by reducing rancidity, spoilage, and off-notes by day 7 without introducing undesirable flavors. These results support BDV as an effective clean-label antimicrobial for fresh pork sausage breakfast patties.

Effectiveness and limitations of fermented onion juice

Fermented onion juice (FOJ) contains a mixture of bioactive compounds, including flavonoids such as quercetin, sulfur-containing thiosulfinates, and fermentation-derived organic acids, that have been associated with antimicrobial and antioxidant activity. These compounds can inhibit microbial growth through membrane disruption, enzyme inhibition, and intracellular acidification (Abdel-Salam et al., 2014; Grzelak-Błaszczyk et al., 2023; Theron and Lues, 2007). However, without a detailed characterization of the FOJ used in this study, these proposed mechanisms remain hypothetical.

In this study, FOJ showed limited antimicrobial effectiveness. Bacterial growth in FOJ-treated sausage closely resembled that of the CONTROL, with aerobic and psychrotrophic counts exceeding 9 Log CFU/g by day 9. This contrasts with some reports of antimicrobial activity in onion-based marinades, although effects in those studies were inconsistent and often dependent on concentration and matrix conditions (Kim et al., 2010).

Several factors may explain FOJ’s limited efficacy at 0.5%. The concentration of active compounds may have been insufficient to suppress the diverse microbial population in fresh sausage, particularly given the high initial load and nutrient-rich environment. Additionally, FOJ’s antimicrobial components may be more effective against specific bacteria that were not predominant in raw pork sausage. Further characterization of FOJ’s active compounds would help clarify its functional potential.

Despite weak antimicrobial performance, FOJ showed a tendency to preserve redness, with slightly higher a* values on day 9 when compared with the CONTROL and BDV, although treatment differences were not significant. This effect may relate to lactate production by fermentation cultures, which can support metmyoglobin-reducing activity (Bradford et al., 1993; Kim et al., 2006), and to the antioxidant properties of onion flavonoids, which can scavenge free radicals (Park et al., 2008). However, these antioxidant effects were insufficient to reduce TBARS values or prevent oxidative sensory notes.

The FOJ’s greatest limitation was its negative impact on sensory quality. By day 7, FOJ samples exhibited higher sourness and off-notes, likely due to organic acids and sulfur compounds (Cheng et al., 2014). Given these sensory drawbacks and minimal antimicrobial activity at 0.5%, FOJ alone is not an effective preservative for fresh sausage. Higher concentrations may improve efficacy but would likely intensify off-flavors. FOJ may be better suited as part of a combination system where its color-protective effects can be utilized at lower, more sensory-acceptable levels.

Complementary effects of combination treatment

The combination treatment (CMB) (0.25% BDV + 0.25% FOJ) showed stronger antimicrobial activity than either ingredient alone, maintaining aerobic counts near 6 Log CFU/g through day 9 and limiting LAB growth to 3–5 Log CFU/g. This enhanced effect aligns with the multiple-hurdle concept: BDV provides acetic acid-driven intracellular pH disruption, while FOJ contributes flavonoids, thiosulfates, and fermentation metabolites that may inhibit enzymes and destabilize membranes. Together, these mechanisms broaden antimicrobial action beyond BDV alone.

The CMB also improved sensory quality. Reducing FOJ to 0.25% minimized sourness and off-notes, and by day 9, CMB samples exhibited less oxidized aroma and less off-note intensity than the CONTROL sausage. In cooked sausage, CMB reduced spoilage aroma, off-notes, and degree of difference by day 7 while matching or exceeding the antimicrobial performance of BDV at twice the concentration. To better understand why these sensory improvements occurred, it is important to consider the relationship between microbial activity and oxidative changes during storage.

Although TBARS values remained low across treatments, they did not fully reflect the sensory oxidation and off-notes that developed during storage. TBARS measures only malondialdehyde, while many other aldehydes, ketones, and alcohols contribute to oxidized aromas and are not captured by this assay; additionally, spoilage bacteria can metabolize MDA into compounds undetected by TBARS (Georgantelis et al., 2007). The superior sensory stability of CMB, despite no TBARS differences, indicates that microbial control, not lipid oxidation, is the primary driver of sensory quality. Overall, CMB offers a clean-label option that extends microbial shelf-life and improves end-of-shelf-life sensory attributes, though yeast growth remains uncontrolled.

Yeast growth

The inability of BDV, FOJ, or CMB to control yeast growth was the major limitation of this study and the primary barrier to extending shelf-life beyond 9 days. Yeast counts increased steadily across all treatments, surpassing 7 Log CFU/g by day 9 with no treatment differences, indicating that organic acid-based antimicrobials are insufficient for yeast control. This reflects fundamental physiological differences between bacteria and yeasts, which possess thicker cell walls and greater acid tolerance.

Yeasts play varied roles in meat systems. Beneficial species such as Debaryomyces hansenii contribute to flavor development and oxidative stability in fermented sausages (Yang et al., 2024), whereas spoilage yeasts, including Candida spp. and Rhodotorula spp., are associated with off-odors, discoloration, and slime formation in fresh meats (Alves Rodrigues et al., 2025). The sliminess observed in all treatments by days 7–9 is consistent with biofilm-forming yeasts such as Candida spp. and S. cerevisiae, which produce extracellular polysaccharides that create slippery surface films (Mohamed et al., 2023; Sidari and Tofalo, 2024).

Several clean-label strategies may help control yeast in future formulations. Increasing salt concentration to 2.5–3.0% can impose osmotic stress that reduces spoilage yeast populations (Barcenilla et al., 2022). Natural yeast-inhibitory compounds such as olive leaf extract, clove or thyme essential oils, and polyphenol-rich plant extracts have also demonstrated antifungal activity in meat systems (Zara et al., 2020). These approaches may provide additional hurdles needed to suppress yeast growth and extend shelf-life beyond 9 days.