A case study on microbiological properties of cooked turkey ham with nitrite reduction

Cooked turkey ham is a widely consumed ready-to-eat meat product in which nitrite plays an important role in safety and quality. However, growing health concerns and regulatory constraints have driven efforts to reduce nitrite use. This study evaluated the effect of reducing nitrite levels from 150 to 119 ppm on the microbiological safety and physicochemical stability of cooked turkey ham during refrigerated storage. Microbiological quality was monitored at 7-day intervals, including total viable counts (TVC), lactic acid bacteria (LAB), Enterobacteriaceae, Staphylococcus aureus, and Escherichia coli. TVC and LAB remained below 10 7 CFU/g until day 28 but reached unsatisfactory results on day 35 (10 8 CFU/g), whereas the remaining microbial groups remained below the detection limit throughout storage, thus supporting a commercial shelf life of 28 days. Sulphite-reducing Clostridium spores, Salmonella spp., and Listeria monocytogenes were not detected in any samples. Challenge tests with Listeria monocytogenes and Clostridium sporogenes were conducted under refrigerated storage and temperature abuse conditions (8 °C) for up to 35 and 264 days, respectively. The reduced-nitrite formulation limited the growth of L. monocytogenes, remaining below the 0.5 log CFU/g threshold at 4 °C but showing a transient increase above this threshold at 8 °C, and prevented detectable vegetative outgrowth of C. sporogenes under the tested conditions. Colour parameters showed no significant differences between the standard and reduced-nitrite formulations (P > 0.05). Physicochemical parameters showed small changes during storage time but remained within the acceptable ranges, with pH values ranging from 6.0 to 6.3, water activity (a w) ranging from 0.94 to 0.96. Overall, the results indicate that reducing sodium nitrite to 119 ppm in the tested cooked turkey ham formulation is feasible without compromising microbiological safety or technological functionality during the defined 28-day commercial shelf life under the conditions evaluated.

Turkey ham is widely consumed as a convenient and perceived healthier alternative to traditional processed meats, largely due to its lower fat content and favourable nutritional profile (Hua and Bajinka, 2025). It is particularly popular among consumers seeking lean protein sources, as it provides a significant amount of high-quality protein. In addition to its protein content, its mild flavour, affordability, and ready-to-eat (RTE) nature contribute to its widespread consumption across different populations (Food Safety and Inspection Service, 2011; Melro et al., 2020). However, like other processed meats, Turkey ham often contains curing agents such as nitrites, raising ongoing concerns regarding potential health implications associated with frequent consumption (IARC, 2015).

Among the additives commonly used in turkey ham, sodium nitrite (E250) plays a crucial role in inhibiting microbial growth, particularly the development of Clostridium botulinum. Beyond their antimicrobial function, nitrites also contribute to key technological properties, including cured colour formation, flavour development, and overall product acceptability, which are essential for consumer preference in turkey ham products. However, nitrite can participate in reactions involving nitrosating species and secondary amines, leading to the formation of N-nitrosamines, compounds of toxicological concern due to their potential carcinogenicity (EFSA et al., 2023; Schrenk et al., 2023; French Agency for Food, Environmental and Occupational Health and Safety (ANSES), 2022).

In the European Union, the maximum permitted level of nitrite in cured meat products, including turkey ham, has historically been set at 150 mg/kg. More recently, regulatory discussions and health risk assessments have led to efforts to reduce permitted levels to 80 mg/kg of nitrite in certain products (European Commission, 2023), reflecting increasing concerns highlighted by agencies such as IARC (2015) and French Agency for Food, Environmental and Occupational Health and Safety (ANSES) (2022). Nevertheless, reducing nitrite concentration must balance potential decreases in nitrosamine formation with the antimicrobial protection provided by nitrites, particularly controlling C. botulinum in products such as turkey ham (Ávila et al., 2014; Lebrun et al., 2020; Isci, 2025).

Microbial spoilage is a major contributor to food waste in processed meat products, as it reduces product shelf life and consumer acceptability leading to the disposal of affected products (Karanth et al., 2023; Blanco-Lizarazo et al., 2024). The occurrence of microbial spoilage in turkey ham is influenced by multiple factors, including the microbiological quality of raw poultry meat, storage temperature and duration, hygiene and sanitation of processing equipment, packaging systems, and environmental conditions during production and distribution (Dousset et al., 2016; Hardin, 2016; Karanth et al., 2023). In RTE products, spoilage is caused by a diverse range of microorganisms, whose growth is strongly affected by intrinsic factors, such as pH, water activity, chemical composition, and added preservatives, as well as extrinsic including storage temperature and atmospheric conditions (Raimondi et al., 2019; Zagdoun et al., 2020). Moreover, RTE products are particularly vulnerable to contamination by foodborne pathogens such as Listeria monocytogenes during post-processing or through post-cooking cross-contamination (Nunes et al., 2026a). For this reason, maintaining strict temperature control (0 °C–4 °C) is critical to ensure product safety and quality. However, in large-scale distribution systems, RTE products are frequently exposed to temperature fluctuations and improper handling during transport and storage, which may compromise their microbiological stability.

In this context, the present study aimed to investigate the effect of reduced nitrite concentrations on the microbiological quality of turkey ham, with particular emphasis on the growth behaviour of L. monocytogenes and Clostridium sporogenes strain PA3679, a validated surrogate for proteolytic C. botulinum (Nunes et al., 2026b; Diao et al., 2014). Additionally, key physicochemical parameters were evaluated, including colour, water activity and pH to assess the acceptability of turkey ham formulations containing different nitrite levels.

Sample preparation

The study consisted of the production of three formulations with varying sodium nitrite concentrations. Following the Regulation (EC) No 1333/2008 (European Commission, 2008), the maximum permitted level of added sodium nitrite was 150 ppm, used to define the standard turkey ham formulation. According to regulatory updates, this limit was reduced to 80 ppm (expressed in nitrite ions, corresponding to 119 ppm of sodium nitrite), corresponding to the reduced-nitrite formulation. A third formulation was produced without sodium nitrite. The third formulation (nitrite-free) was included exclusively in the microbial challenge test and was not subjected to physicochemical analyses because it exhibited pronounced technological instability, characterized by excessive syneresis (free water release). The absence of nitrite adversely affected the product’s technological properties, compromising water-holding capacity and overall structural integrity. Similar technological limitations have been reported for nitrite-free cooked ham by Nunes et al. (2026a). The extensive syneresis observed could substantially influence both physicochemical and microbiological measurements, thereby preventing meaningful comparisons with the other formulations.

Turkey ham was produced on a pilot scale. The formulations used are depicted in Table 1. Three independent production runs were performed. In each production run, 60 kg of minced turkey meat from a single raw-material batch were subdivided into three portions, each corresponding to one formulation with a different nitrite concentration. This design ensured that the formulations compared within each production run shared the same raw-material origin and processing conditions, thereby minimizing sources of variability other than nitrite concentration. The entire production procedure was independently repeated three times, providing three independent production replicates for each formulation (n = 3). The production consisted of preparing the brine with all ingredients at a temperature between −2 °C and 2 °C, followed by mixing the turkey meat with the brine using a vacuum mixer operating at constant speed (TECMAQ, Sentmenat, Spain). The resulting batter was stuffed into plastic casings and subjected to a controlled cooking process (high humidity, chamber temperature 77 °C, core temperature 72 °C; Verinox, Italy). After thermal processing, the cooked Turkey ham was cooled, sliced under controlled conditions, packed (150 g per unit, PET EVOH PE) with modified atmosphere (25% CO 2; 75% N 2). The product was then stored under refrigeration (0 °C–5 °C, Testing Lab, Madrid, Spain). The final result consisted of 10 ham units per batch (seven units for microbiological analysis, three units for physico-chemical analysis).

One Turkey ham independent package was analysed (destructive sampling) from each formulation and each batch, at each predefined storage time point. Samples were transported under refrigerated conditions, and analyses were performed on fresh products. Sampling was carried out immediately after chilling (T0) and subsequently at 7-day intervals over a total storage period of 35 days. At each time point, a new unit was opened for analysis.

Microbial assessment

At each sampling time, three cooked Turkey hams (one per batch) per formulation were aseptically opened. From each ham unit, a 25 g composite sample was prepared by pooling subsamples collected from seven different locations into a sterile stomacher bag (BagLight PolySilk 400, Interscience, Saint-Nom-la-Bretèche, France). Subsequently, 225 mL of sterile Buffered Peptone Water (BPW; Biokar Diagnostics, Beauvais, France) was added, and the samples were homogenized for 30 s in a stomacher (BagMixer 400 S, Interscience). Serial decimal dilutions were then prepared in sterile Ringer’s solution (¼ strength; Biokar Diagnostics) for microbiological enumeration. Total viable counts (TVC) were determined on Plate Count Agar (PCA; Biokar Diagnostics) following International Organization for Standardization (2013, 2017c) at 30 °C for 72 h (limit of detection (LOD):10 CFU/g). Lactic acid bacteria (LAB) were enumerated on de Man Rogosa and Sharpe agar (MRS; Biokar Diagnostics) according to International Organization for Standardization (1998) at 30 °C for 72 h (LOD: 10 CFU/g). Enterobacteriaceae were quantified on RAPID Enterobacteriaceae Agar (Bio-Rad, Antibes, France) following International Organization for Standardization (2017b) at 37 °C for 24 h (LOD: 10 CFU/g). Escherichia coli was enumerated on Tryptone Bile X-glucuronide agar (TBX; Biokar Diagnostics) according to Instituto Português da Qualidade (1986b) at 44 °C for 24 h (LOD: 10 CFU/g). Staphylococcus aureus was determined on Baird-Parker Agar (BPA; Biokar Diagnostics) following International Organization for Standardization (2021) at 37 °C for 24 + 24 h (LOD: 10 2 CFU/g). Sulphite-reducing Clostridium spores were detected using a specific medium prepared according to Instituto Português da Qualidade (1986a), with incubation at 37 °C for 5 days. Additionally, the presence of Salmonella spp (ISO 6579, 2017 c) and Listeria monocytogenes (International Organization for Standardization, 2017a) was assessed. All microbiological analyses were performed in duplicate, and results were expressed as log CFU/g.

Challenge testing

Microbiological challenge tests are an essential tool to evaluate the ability of foodborne pathogens such as L. monocytogenes or C. botulinum to grow and/or survive in food products from production to consumption. This process involves deliberately inoculating the product with relevant microorganism(s) and storing it under controlled environmental conditions to assess the risk of foodborne illness and/or determine product stability over time (Betts, 2010).

Microbiological challenge tests were conducted using C. sporogenes inoculated into the raw batter before heat treatment and L. monocytogenes inoculated onto sliced product after processing, following the general methodology described by Nunes et al. (2026b). Samples inoculated with L. monocytogenes were stored at 4 and 8 °C for 35 days, with sampling at 7-day intervals, whereas samples inoculated with C. sporogenes were stored at 4 and 8 °C for 264 days, with sampling at 15-19 days intervals. For each formulation and storage condition, three independent production batches were used as biological replicates, and analytical duplicate measurements were averaged before statistical analysis. At each sampling time, one independently packaged sample from each batch and formulation was analysed as the experimental unit. Results were expressed as log N/N0.

C. sporogenes challenge testing

Spores of C. sporogenes PA3679 were prepared following the procedure described by Nunes et al. (2026a), based on the methodology of Shao and Ramaswamy (2008). The PA3679 strain is widely used as a surrogate in process validation studies due to its genetic and physiological similarity to proteolytic C. botulinum, particularly in terms of thermal resistance, as reported by Diao et al. (2014). The raw batter of turkey ham mixture was inoculated at 1% (v/w) prior to cooking, then stuffed and packaged in plastic casings under anaerobic conditions to simulate whole-piece products with minimal oxygen availability, as described by Couvert et al. (2019). The products were subsequently cooked to a core temperature of 72 °C (oven/bath set at 77 °C), with temperature monitored using a data-logging thermometer. The bacterial counts determined at T0 (2.8 ± 0.2 log CFU/g) reflect the surviving spore population after thermal processing, as a reduction associated with the cooking process is expected. After cooling, samples were stored at 4 °C and 8 °C for a period of 264 days, with sampling conducted every 15 days (until day 30) and every 19 days (from day 30–264). Detection and enumeration of spores were carried out using CSR agar (NP 2262:1986) for spores detection, conducting a heat treatment at 80 °C for 10 min, and TSC agar for vegetative cells enumeration (International Organization for Standardization (2023) LOD: 10 CFU/g), respectively.

L. monocytogenes challenge testing

Following the methodology described by Nunes et al. (2026a), a seven-strain cocktail of L. monocytogenes was prepared. Each strain was first grown overnight in Brain Heart Infusion (BHI) broth at 37 °C, then subjected to cold adaptation at 4 °C for 7 days in accordance with International Organization for Standardization (2019). Subsequently, cultures were centrifuged, washed with Ringer’s solution, combined, and diluted to an initial inoculum of approximately 6.2 log CFU/mL. The 1% (v/w) inoculum was aseptically dispensed at multiple points across the surface of each 10 g turkey ham slice to promote homogeneous distribution. After inoculation, the entire slice was used for microbiological analysis at each sampling time, thereby minimizing potential variability associated with uneven bacterial distribution. The inoculum concentration was verified by plate counting prior to inoculation, and the actual initial bacterial population (T0) was confirmed immediately after inoculation (4.2 ± 0.1 log CFU/g). The samples were then packaged under modified atmosphere conditions (80% N 2 /20% CO 2), representative of industrial charcuterie practices as described by Iacumin et al. (2020) and stored at 4 °C and 8 °C for 35 days. Sampling was carried out at 7-day intervals throughout storage. Microbial enumeration was performed on PALCAM agar, with incubation at 37 °C for 48 h (LOD: 10 2 CFU/g), following the procedure outlined by Nunes et al. (2026b).

The gaseous composition of the headspace in packages containing sliced ham inoculated with Listeria monocytogenes was monitored throughout storage. Measurements were carried out using a Dansensor CheckPoint 3 gas analyser (AMETEK MOCON, Ringsted, Denmark). For each analysis, a syringe was inserted through the packaging material to directly sample the internal atmosphere. This approach allowed the determination of oxygen (O 2) and carbon dioxide (CO 2) levels in the headspace of each formulation at every sampling point, following the methodology described by Raines and Hunt (2010).

Colour analysis

Colour measurements were carried out in accordance with the procedures described by Nunes et al. (2026b), using a CR-400 Chroma Meter (Konica Minolta Sensing Americas, Inc., New Jersey, United States) equipped with a D65 illuminant and an 8-mm aperture. The colour attributes assessed included chroma (C), hue angle (H), and total colour difference (Δ E), with Δ E calculated according to the equation Δ E = [(Δ L) 2 + (Δ a) 2 + (Δ b*) 2 ] 0.5, as outlined by King et al. (2022) and MacDougall (1982), comparing formulation A (150 ppm) with formulation B (119 ppm) at every time point. Measurements were taken at five randomly chosen points on the surface of each sample to ensure a representative evaluation of colour distribution. Prior to analysis, samples were allowed to bloom for 20–30 min to promote colour stabilization and enhance measurement reliability, following the approach described by Škrlep and Čandek-Potokar (2007).

Water activity and pH

All analyses were performed in duplicate in three time points during the product’s shelf life: the beginning (0 days), the midpoint (17 days), and the end of the shelf life (35 days). Water activity (a w) of each sample was determined using a LabMaster-aw Neo instrument (Novasina AG, Lachen, Switzerland). All measurements were carried out at room temperature, ensuring that samples were properly equilibrated prior to analysis to reduce variability associated with both the instrument and sampling procedure.

Subsequently, the pH of each sample was measured using a calibrated benchtop pH meter (Sension + PH3, Hach, Colorado, United States) equipped with a combined glass electrode, following standard procedures for meat product analysis.

Residual nitrite content

Residual nitrite was quantified at the initial sampling time (T0, after slicing) by an accredited external laboratory using the reference method described by the International Organization for Standardization, 1975. Three independent batches were analysed in duplicate, and results were expressed as mg/Kg NO 2 −. Residual nitrite was measured only at T0; therefore, the study does not provide direct information on residual nitrite depletion during storage. This limitation was considered when interpreting the antimicrobial effects observed during shelf-life and challenge testing, since previous studies have shown that residual nitrite concentrations decrease over time (Guéraud et al., 2023; Nunes et al., 2026a; Lebrun et al., 2020).

Statistical analysis

Results are presented as mean ± standard deviation (SD) based on three independent production batches. Statistical analysis was carried out using IBM SPSS Statistics 31.0 (IBM Corp., Chicago, IL, United States). The normality of the data for each experimental condition was assessed using the Shapiro–Wilk test. The influence of the independent variables (storage time and turkey ham formulation) and their interaction was evaluated using a two-way ANOVA (univariate general linear model with fixed factors), considering a 95% confidence level. Because destructive sampling was used at each time point, the same ham unit was not followed repeatedly over storage; therefore, the dataset was not treated as a repeated-measures design. For datasets meeting the normality assumption, differences between means were further examined using Tukey’s post-hoc test. When normality was not satisfied, group comparisons were performed using the non-parametric Kruskal–Wallis test at the same confidence level. For each batch, all microbiological determinations were performed in duplicate and physicochemical determinations were performed in triplicate. Duplicate or triplicate measurements were averaged before statistical analysis and were not treated as independent experimental replicates. Samples with counts below the detection limit were reported as below the detection limit (<10 CFU/g) and assigned a value of 1.0 log 10 CFU/g for graphical representation. Regarding statistical analysis, all analysis had an assigned value of 1.0 log 10 CFU/g, with the exception of S. aureus and L. monocytogenes, corresponding to 2.0 log 10 CFU/g.

Microbial assessment

Sliced turkey ham quality and safety were assessed across formulations A (150 ppm) and B (119 ppm) throughout the 35-day storage period. No Salmonella spp., L. monocytogenes, or Clostridium spores were found in any of the samples tested. The levels of S. aureus, Enterobacteriaceae and E. coli stayed below the detection thresholds of the counting methods used for all samples (10 2 CFU/g for S. aureus; 10 1 CFU/g for Enterobacteriaceae and E. coli), demonstrating minimal or undetectable concentrations throughout the storage duration.

Results for total viable counts (TVC) and lactic acid bacteria (LAB) are depicted in Figures 1, 2 respectively and values below the detection limit were plotted as 1.0 log 10 CFU/g. RTE products such as sliced turkey ham follow strict legislation regarding TVC and LAB criteria, considered unsatisfactory when 7 log CFU/g is reached (European Commission, 2005; FAO & WHO, 2023). Sliced products can be re-contaminated after the cooking process, increasing the chances of earlier spoilage and affecting the shelf life of meat products (Mathias et al., 2013). Although the initial counts for both TVC and LAB are below the detection limit, an increase is expected during the storage time (Menezes et al., 2018; Spampinato et al., 2022), which aligns with the obtained results.

For TVC, a difference between formulations was observed only on day 21 (P = 0.013). The remaining time points presented no significant differences between formulations (P > 0.54). LAB counts presented no significant differences between formulations at any time point (P > 0.71). Standard deviation variability may be explained by microbial distribution and environmental conditions (Pan and Ren, 2022), including both initial bacterial load and storage temperature. Both TVC and LAB counts increased progressively during storage, exceeding 8 log CFU/g by day 35. This increase was probably driven by post-processing contamination, favourable intrinsic product characteristics, and the selective effects of atmosphere modified packaging (Kalschne et al., 2015; Pexara et al., 2002). Although thermal processing eliminates most of the initial microbiota, slicing and handling can reintroduce microorganisms, particularly LAB, from equipment and the processing environment (Pal et al., 2008; Vercammen et al., 2011). These bacteria are well adapted to cooked meat products, which typically exhibit a w > 0.95, near-neutral pH (∼6.0), and abundant nutrients, consistent with the characteristics observed in the present study. Despite the increase in TVC and LAB counts during storage, only minor changes in pH and a w were observed. Furthermore, vacuum or modified atmosphere packaging suppresses aerobic spoilage organisms and favors the growth of facultative anaerobic LAB, allowing continuous, exponential growth throughout storage. As a result, microbial populations can increase by 6–7 log cycles over a typical shelf life of 20–30 days, commonly reaching 10 7 –10 9 CFU/g (Petersen et al., 2010; Slongo et al., 2009; Blanco-Lizarazo et al., 2022), which explains the microbial trends observed in both Figures 1, 2 and spoilage observed at the last time point. According to Premi et al. (2024) and Zhang et al. (2023), nitrite also exerts antimicrobial activity, mainly through the generation of reactive nitrogen compounds, and by lowering the oxidation-reduction potential, thereby affecting enzymes, peptides and membrane structures in sensitive bacteria. Furthermore, the modified atmosphere with carbon dioxide/nitrogen inhibits the growth of aerobic species, favouring the growth of heterofermentative LAB that produce undesired metabolites responsible for odd flavour and bloated package (Marcelli et al., 2024; Shen et al., 2023). This microbial succession may explain why LAB represented an increasing proportion of the total viable count throughout storage, whereas no uncontrolled proliferation of other spoilage microorganisms was observed. Nevertheless, variations in nitrite concentration (150–119 ppm) had no statistically significant effect on either microbial count (P > 0.05). Therefore, the commercial shelf life of both formulations was defined as 28 days, since the final time point was considered unacceptable.

Challenge testing

Microbiological challenge tests are a very important tool that helps to mimic the pathogens behaviour during specific processing, handling and storage conditions (Spanu et al., 2014). Challenge tests are employed in RTE products to evaluate the growth potential of L. monocytogenes, as well as C. botulinum (Lorenzo et al., 2018). Owing to biosafety limitations associated with C. botulinum, C. sporogenes, especially strain PA3679 (ATCC, n.d.), is commonly used as a surrogate microorganism in food process validation studies because of its close genetic and physiological resemblance, notably regarding thermal resistance characteristics (Diao et al., 2014). These challenge studies should be performed under realistic storage conditions, considering both recommended refrigeration temperature (4 °C) and temperature abuse scenarios (8 °C). Sliced cooked turkey ham inoculated with L. monocytogenes was packed under modified atmosphere typical of charcuterie industry practices (Iacumin et al., 2020); when inoculated with C. sporogenes, it was packed in plastic casings to simulate pre-cooking contamination and minimal oxygen conditions (Couvert et al., 2019).

The results of the L. monocytogenes challenge tests are depicted in Figure 3. At 4 °C, significant differences were uncovered between all three formulations at times 21 and 28 days (P = 0.001), with higher values for formulation without nitrite (0 ppm - C), followed by the formulation with 119 ppm (B) and the lowest values were observed in the formulation with 150 ppm (A). At 35 days, values of formulation A and B are equal, and they both differ from formulation C, without nitrite (P = 0.001). At 8 °C, differences were also observed from day 14–35 (0.003 < P < 0.026), where a similarity is noted between hams with 119 ppm and 150 ppm (0.2 < P < 0.8), and these two differing from the turkey ham without nitrite (0.001<P<0.006). Growth of the pathogen was observed along the 35-day period for the cooked turkey ham without nitrite (P<0.001), with a growth of 1 log(N/N0) for 4 °C and a higher growth when stored at 8 °C (approximately 2 log(N/N0)). Statistically, formulation with 119 ppm remained stable over time (0.5 < P < 0.9) at both temperatures, indicating that growth remained limited throughout storage. According to the EURL Lm technical guidance, foods are considered able to support the growth of L. monocytogenes when the growth potential (δ) exceeds 0.5 log CFU/g, whereas values ≤ 0.5 log CFU/g indicate that growth is not supported under the tested conditions (Bergis et al., 2026). The maximum increase in the 119 ppm formulation exceeded 0.5 log CFU/g at day 28 under storage at 8 °C, while remaining below this threshold at 4 °C. The same behavior was observed for formulation A (150 ppm), at times 14 and 28 days, for 8 °C. Therefore, under the tested conditions, both formulations did not support growth according to the 0.5 log CFU/g criterion at 4 °C, whereas under temperature abuse at 8 °C, growth was supported (δ values are depicted in (Supplementary Table S1).

Standard formulation (150 ppm) presented an increase at 4 °C on day 14 (P = 0.013), followed by a decrease until day 35. The same was observed at 8 °C, with a lower value at day 35 (P = 0.021). Nevertheless, these oscillations may be due to sample heterogeneity or the initial inoculum, as described by Everis and Betts (2013), and Nieto et al. (2023). The initial inoculum (∼4.2 log CFU/g) was higher than the level recommended by the EURL Technical Guidance (approximately 2 log CFU/g). This higher inoculum was selected to ensure accurate enumeration throughout storage and to facilitate comparisons between formulations with different nitrite concentrations, rather than to establish regulatory compliance with the EURL classification framework for ready-to-eat foods. Similar inoculum levels have been used in challenge tests designed to evaluate the efficacy of antimicrobial interventions rather than to simulate natural contamination levels (National Advisory Committee on Microbiological Criteria for Foods - NACMCF, 2010; Lopez et al., 2021; Spanu et al., 2014; International Organization for Standardization, 2019; Barcia-Cruz et al., 2026). Nevertheless, because microbial behaviour may vary with inoculum size due to population density effects and increased physiological heterogeneity, the calculated growth potential should be interpreted as reflecting the behaviour of L. monocytogenes under the specific experimental conditions employed rather than as a strict EURL-compliant shelf-life classification.

All samples were packed with modified atmosphere to guarantee similar conditions to the products produced in the charcuterie industry (80% N 2 and 20% CO 2), preserving the product and slowing spoilage (Duthoo et al., 2021; Raimondi et al., 2019). MAP data is shown in Supplementary Table S2, showing that nitrogen and carbon dioxide levels remained stable during the 35-day storage period (P > 0.05).

Nunes et al. (2026a), Bonifacie et al. (2024) demonstrated that nitrite concentrations decrease over time, reducing its inhibitory effect against L. monocytogenes. In parallel, the intrinsic properties of turkey ham, including pH and a w, are conducive to the growth of this pathogen (Bover-Cid et al., 2019; Iacumin et al., 2020; Gowda et al., 2024). Nevertheless, the combination of nitrite with additional hurdles, such as antioxidants or preservatives, can enhance antimicrobial efficacy, consistent with the behaviour observed across formulations in this study (Myers et al., 2013; Sullivan et al., 2012; Nieto et al., 2023; Pal et al., 2008; Torres-Baix et al., 2023). Sodium nitrite can limit the growth of L. monocytogenes through the nitric oxide (NO) formation and other reactive nitrogen species that interfere with enzymes and cellular respiration, causing nitrosative stress (Zhang et al., 2023; Shakil et al., 2022; Isci, 2023). Nitric oxide may continue to form from residual nitrite, contributing to antimicrobial activity even as residual nitrite concentration decreases during storage. The presence of NO also helps reduce the oxidation-reduction potential of the matrix, maintaining an unfavourable intracellular environment for bacteria (Premi et al., 2024; Shakil et al., 2022). A modest reduction in L. monocytogenes (ca. 1 log CFU/g) has also been reported in sliced turkey products (Dalzini et al., 2014). Overall, strict temperature control (≤4 °C) and maintenance of the cold chain remain critical to limit pathogen proliferation in RTE meat products, in line with recent regulatory emphasis on the control of L. monocytogenes throughout shelf life (European Commission, 2024).

The results of the challenge tests with C. sporogenes are presented in Figure 4 and Supplementary Table S3. At 4 °C no detectable vegetative cells were observed in formulations containing 119 ppm or 150 ppm added nitrite from day 15 onwards, with no significant changes over storage time (0.15 < P < 0.99). In contrast, the nitrite-free formulation showed detectable vegetative cells at days 15 and 76, with significant differences relative to the nitrite-containing formulations. A similar trend was observed at 8 °C, where the nitrite-free formulation also differed from the formulations with nitrite at several sampling times (days 15, 76, 168 and 214) (P = 0.003). These findings are consistent with previous studies reporting the antimicrobial effect of nitrite against C. sporogenes and support its role in controlling vegetative cell outgrowth in cooked meat products (Lebrun et al., 2020; Yetim et al., 2006; Majou and Christieans, 2018; Hong et al., 2016). Although some inhibition was also observed in the nitrite-free formulation, this effect was less consistent and may be related to the storage temperature (Dahlsten et al., 2015), nutrient depletion or product composition, including fat content (Bhusal et al., 2021). Therefore, refrigerated storage, sodium nitrite and vacuum packaging likely acted synergistically to limit vegetative outgrowth, thereby contributing to the observed microbiological stability.

Clostridium sporogenes spores remained detectable throughout the 264 days of storage, although their concentration decreased over time (Supplementary Table S3). The initial spore load was detected in 0.01 g of sample in all formulations. In the formulations containing 119 ppm or 150 ppm added nitrite, spores were subsequently detected only at progressively higher sample masses, shifting from 0.1 g at days 15 and 30, to 0.5 g at days 76 and 122, to 1 g at day 168, and to 5 g at days 214 and 264, indicating a gradual reduction in spore concentration during storage. In the nitrite-free formulation, the reduction was less pronounced, with spores remaining detectable in 0.01 g up to day 122, in 0.1 g at day 168, and in 0.5 g at days 214 and 264. These results are in agreement with previous reports showing that Clostridium spores may persist even when vegetative cells are effectively controlled in nitrite-containing products (Bhusal et al., 2021; Lebrun et al., 2020; Nunes et al., 2026a). They also support the view that nitrite contributes to lowering spore load in the product matrix over time (Junqueira Pinelli et al., 2024). Because inoculation was performed before cooking to mimic a realistic contamination from raw materials or the environment, the results indicate that nitrite contributes to the control of C. sporogenes throughout the shelf life of unopened turkey ham. The thermal processing applied during manufacture inactivated vegetative cells, whereas bacterial spores survive because of their highly resistant multilayered structure (Bhusal et al., 2021; Paredes-Sabja et al., 2011). Therefore, Clostridium spores remained in a metabolically dormant state, characterized by minimal water content, low enzymatic activity and high resistance to environmental stresses, including heat and oxidative damage (Wells-Bennik et al., 2016; Berendsen et al., 2016). Although spores can persist throughout refrigerated storage, the synergy of sodium nitrite, heat treatment, cold storage and nutrient depletion probably led to a progressively less recoverable concentration of spores (Wells-Bennik et al., 2016). Although storage at ≤ 4 °C provided better control, the results also highlight the potential safety implications of temperature abuse at 8 °C, particularly for nitrite-free formulations.

A limitation of this study is that microbial challenge testing was performed using C. sporogenes PA3679 as a surrogate organism rather than C. botulinum. Although C. sporogenes is widely used as a non-toxigenic surrogate due to its similar spore resistance and growth characteristics, the present study did not evaluate botulinum toxin production or include proteolytic or non-proteolytic C. botulinum strains (Food Standards Agency, 2020; EFSA et al., 2023; European Food Safety Authority, 2005). Therefore, these findings should be interpreted as surrogate evidence and not as direct confirmation of botulinal safety, particularly for refrigerated ready-to-eat meat products.

Colour analysis

The colour of turkey ham strongly influences consumer acceptance, as it is an important quality attribute. Turkey ham colour parameters are presented in Table 2 over the 35-day storage period for formulations containing 150 or 119 ppm sodium nitrite. Significant changes over storage time were observed for the main colour parameters (P<0.016), particularly by day 35, when both formulations showed lower L *, a *, and b * values than at the beginning of storage. The CIE colour parameters were within the ranges reported for turkey ham by Sinanoglou et al. (2018), Topczewska et al. (2026), namely higher L* values (approximately 65–75), lower a* values (approximately 3–8) than those typically reported for pork ham, and b * values of approximately 8–14, consistent with the lighter and slightly yellow appearance of turkey ham. The total colour difference (Δ E), calculated by comparing formulation A (150 ppm) with formulation B (119 ppm) at each storage time, remained below the perceptibility threshold of 2.0, indicating that colour differences between formulations were not readily perceptible (Ripoll et al., 2023; Altmann et al., 2022; Salueña et al., 2019). Hue values were higher than those typically reported for cooked pork ham, as expected (Iqbal et al., 2010), while chroma values were similar to those reported by Nunes et al. (2026a). Reducing sodium nitrite from 150 ppm to 119 ppm did not significantly affect the overall colour parameters (P > 0.05). The gradual decrease in L, a and b* values observed during storage is likely associated with physicochemical changes occurring in the meat matrix rather than with microbial growth. Although the modified atmosphere limited oxygen availability, residual oxidative reactions and the progressive depletion of residual nitrite may have reduced the stability of cured meat pigments over time, leading to lower redness and yellowness values (Nunes et al., 2026a; Sinanoglou et al., 2018). Simultaneously, minor changes in moisture distribution and protein structure during storage may have altered light scattering properties, contributing to the observed reduction in lightness. Since pH and water activity remained relatively stable throughout storage, these colour changes are more likely attributable to gradual pigment transformations than to microbial metabolism (Honikel, 2008; King et al., 2022).

Water activity and pH

The water activity (a w) and pH values are presented in Table 3. A significant decrease in pH was observed over storage time (P = 0.026), with no significant differences between formulations at any sampling time (P > 0.79). These variations may be associated with microbial activity and strain-specific metabolic changes during storage, as previously reported by Pateiro et al. (2019), Rendueles et al. (2024), and are consistent with the microbial outgrowth observed in the study. Overall, the pH values were consistent with those typically reported for turkey ham, generally ranging from 5.8 to 6.3, depending on formulation and processing conditions (Rendueles et al., 2024; Iqbal et al., 2013).

Water activity, a critical parameter influencing microbial stability and texture in turkey ham, showed variations during storage (P=0.003), with values ranging from 0.94 to 0.96, in agreement with literature data for cooked turkey products (Iqbal et al., 2013). Both formulations showed similar behaviour throughout storage, and no significant differences were found at any time point (P > 0.53). Although statistically significant changes were observed over storage, the magnitude of these changes was limited and the overall physicochemical characteristics of the products were maintained.

Residual nitrite content

Residual nitrite was measured by spectrophotometry and the results are presented in Table 4. No significant differences were observed between formulations (P > 0.50). Previous studies by Nunes et al. (2026a), Nunes et al. (2026b) reported a decrease in residual nitrite concentration over time in Portuguese commercial meat products. Nevertheless, residual nitrite concentration may vary depending on product formulation, ingredients, manufacturing conditions and thermal treatment (with or without smoke) (Isci, 2025; Zhang et al., 2023). Studies by Honikel (2008), Isci (2025) have shown that the depletion of nitrite during storage is associated with several chemical reactions occurring within the meat matrix that progressively consume the compound. Nitrite may be converted into nitric oxide (NO), react with myoglobin, be partially oxidised to nitrate, or bind to proteins, lipids, amino acids and haem compounds (Lebrun et al., 2020; Skibsted, 2011). The low residual nitrite concentrations observed in the present study are consistent with the nitrite instability previously reported by Surace et al. (2025), Nunes et al. (2026a), who reported residual nitrite concentrations of 3.92 ± 0.05 mg/kg and 2.5–4 mg/kg, respectively, in ham products.

Although residual nitrite was not monitored throughout storage in the present study, previous research suggests that antimicrobial activity may be influenced not only by the residual nitrite remaining after processing but also by the initial ingoing nitrite concentration and the early formation of nitric oxide and other reactive nitrogen species (Guéraud et al., 2023). The absence of significant differences between the 150 and 119 ppm formulations in the C. sporogenes challenge test is consistent with this interpretation; however, the present study was not designed to directly assess the relationship between residual nitrite depletion and antimicrobial activity over time. As reported by Guéraud et al. (2023), microbial inhibition may be more strongly influenced by the initial nitrite concentration than by the small amount of free residual nitrite remaining after heat treatment. Furthermore, the residual nitrite concentrations observed in the present study remained within the limits established by Commission Regulation (EU) 2023/2108 (European Commission, 2023), which stipulates that residual nitrite content, expressed as nitrite ion (NO 2 −), shall not exceed 45 mg/kg in products ready for marketing throughout their shelf life. Although reducing sodium nitrite is frequently proposed as a strategy to decrease the potential formation of N-nitrosamines in processed meat products (Guéraud et al., 2023; Zhang et al., 2023), nitrosamine levels were not quantified in the present study. Therefore, no conclusions can be drawn regarding the impact of nitrite reduction on the formation of these compounds. Since nitrite reduction should be evaluated considering both microbiological and chemical safety, future studies should include N-nitrosamine determination and residual nitrite monitoring throughout storage to provide a more comprehensive assessment of the health implications of reduced-nitrite turkey ham formulations.

This study evaluated the microbiological safety and technological properties of cooked turkey ham formulated with sodium nitrite reduced from 150 ppm to 119 ppm. The nitrite-reduced formulation with 119 ppm showed microbial behaviour similar to that of the standard product regarding spoilage microorganisms, with both formulations maintaining a shelf life of 28 days within acceptable limits. An inhibitory effect against C. sporogenes and limited growth of L. monocytogenes were observed, supporting the feasibility of nitrite reduction for this specific turkey ham formulation under the tested conditions, particularly under refrigerated storage. Additionally, technological characteristics such as colour, pH and a w remained consistent with typical values reported for turkey ham, indicating maintenance of the food quality throughout storage. For this specific formulation and processing conditions, the new formulation complies with the new regulatory limits for sodium nitrite, while maintaining product safety and quality. Nevertheless, additional research is needed, mainly on nitrosamine formation, residual nitrite depletion during storage, and microbial community dynamics, to better understand the long-term safety and quality of cooked turkey ham.