Effect of Sodium-Alternative Curing Salts on Physicochemical Properties during Salami Manufacture

Refrigerated pork hind leg, beef topside round, and frozen pork backfat were purchased from a domestic meat supplier. After the raw materials were sliced, they stored frozen at –24°C for 2 days. The batters were made with various curing agents in triplicates (12 batches, in total). The mince of salami consisted of 46% pork hind leg, 30% beef topside round, 20% pork back fat and pork skin, 0.5% garlic, 0.1% monosodium glutamate, 0.4% glucose, and 0.2% starter culture. The added meat starter culture (Lyocarni RBL-73, SACCO, Italy) was containing Lactobacillus curvatus and Staphylococcus xylosus with 2.5×10⁶ Log CFU/g and was used at a level of approximately 6 Log CFU/g. Four salami treatments were prepared with different curing salts, NaCl, KCl, or MgCl₂ at 1.9% (w/w meat batter) with sodium nitrite (NaNO₂) at 0.01%. The concentrations of additives were determined by preliminary studies. Only NaCl-added treatment without NaNO₂ was prepared as negative control. Therefore, the treatments were T1 (-control, NaCl), T2 (+control, NaCl+NaNO₂)_, T3 (KCl+NaNO₂), and T4 (MgCl₂+NaNO₂).

Pork, beef, and pork backfat were chopped in a bowl chopper (Fujee, Seoul, Korea), and mixed for 4 min. Ingredients and starter culture were added and mixed in a mixer (Fujee). Using a stuffer (H20E, TALSA, Northampton, EU), the mince was stuffed into fibrous casings (55-mm diameter, Seoul, Korea) using a stuffer (H20E, TALSA, Northampton, EU). Then these were soaked in solution of Aspergillus spp. (obtained from the Korean Agricultural Culture Collection, RDA) for 2 min. The salami samples were streaked onto 18% Glycerol Agar (Dichloran DG18, Kisanbio, Seoul, Korea) and then incubated at 25°C for 7 days. The fungal colonies were diluted with distilled water for producing Aspergillus spp. solution. Samples were dried and aged in a dry-ripening room. The relative humidity and temperature were 80%–98% and 16°C–20°C during the ripening. Samples for physicochemical analyses were obtained from each treatment and after 20 and 40 days of ripening.

The pH of samples was measured by homogenizing a 10 g sample with 90 mL distilled water with a pH meter (PHM201, Radiometer, Villeurbanne, France). A water activity was measured using a water activity meter (Handheld HP23-AW-A, Rotronic AG, Bassersdorf, Switzerland). Volatile basic nitrogen (VBN) was estimated as the method reported by Conway (1950) then reported as mg/100 g. Lipid oxidation was analyzed in triplicate with minor modification of the method of Witte et al. (1970). The absorbance was measured on a spectrophotometer (X-MA 3000, Human, Seoul, Korea) at 530 nm. Water holding capacity (WHC) was measured using the modified method of Grau and Hamm (1953). Briefly, a sample weighing 300 mg was placed on Whatman No. 1 filter paper and compressed for 2 min. WHC was calculated as follows: WHC (%) = (1 − Total meat area / Meat film area) × 100.

Total plate counts, lactic acid bacteria counts, E. coli. counts of the samples were analyzed according to the guidelines specified in the Criteria and Ingredient Standard of Livestock Products (QIA, 2013). Color values were determined using a chromameter (CR-410, Minolta, Tokyo, Japan), at three replicates. A texture analyzer (TA-XT2, Stable Micro Systems, Godalming, UK) equipped with a load cell was (2,500 N) used on texture profile analysis (Bourne, 1978). The samples were reconstructed to 1.5 cm diameter and 1 cm height. Meat sample was compressed to 50% of their height at 1.0 mm/s, with a 50-mm-diameter plunger. The hardness, springiness, cohesiveness, gumminess, and chewiness values were measured. Shear force (kgf) was determined by the method described by the procedure of Bourne (1978).

Meat samples (4 g) were mixed with 20 mL of 0.7 M perchloric acid and homogenized (T25b, Ika Works, Malaysia) for 60 sec at 1,230 g to extract nucleic acids. The extracted nucleic acids were centrifuged (Union 32R, Hanil, Gimpo, Korea) for 14 min at 2,190 g (5°C) and filtered through Whatman No. 4 filter paper (Whatman, Maidstone, UK). The supernatant was then adjusted to pH 7 with 5 N KOH (SevenEasy, Mettler-Toledo Int., Switzerland). The pH-adjusted supernatant was placed in a volumetric flask and adjusted to a volume of 100 mL with 0.7 M perchloric acid (pH 7). After 25 min of cooling, the mixture was centrifuged (Union 32R) at 2,190×g (5°C) and the supernatant was filtered through a 0.2-μm PVDF syringe filter (Whatman). The filtrate was analyzed using a high-performance liquid chromatography (ACME 9000, Young-Lin, Korea) with a Waters-Atlantis C18 RP column (4.6×250 mm, 5 μm particles, Waters, Milford, MA, USA) and a mobile phase of 0.1 M triethylamine in 0.15 M acetonitrile (pH 7.0). The peaks of individual nucleotides were identified using standards of hypoxanthine, inosine, inosine-5-phosphate (IMP) (Sigma Chemical, St. Louis, MO, USA).

The experiment was designed and statistically analyzed in factorial (curing agents×ripening time). All the variables measured was analyzed on their variance using the General Linear Model (GLM) procedure. To determine the differences among the treatment means, the Duncan’s multiple range test was used with significance level of p<0.05. The SAS statistical software package was utilized (SAS, 2002).

The effect of different sodium-alternative curing agents on the physicochemical traits of salamis during ripening is presented in Table 1. After 20 d of ripening of salamis, pH was the highest in T1 but the lowest in T3 (p<0.05). The pH values were higher in KCl treatment and lower in MgCl₂ than that of NaCl treatments during the ripening period (p<0.05). These results are consistent with results of previous report (Kim et al., 2018). This difference was related to the inhibitory action of NaCl substitution by KCl towards the growth of coliforms, which metabolized basic nitrogen compounds leading to pH changes in dry-cured bacons (Alinõ et al., 2010). Gimeno et al. (1999) also reported that NaCl was replaced by a mixture of 44.5% NaCl, 24.5% KCl, 20.6% CaCl₂, and 10.4% MgCl₂, and a more decrease in pH on the mixture than NaCl-only formulation. A pattern similar to this was reported for a partial NaCl decrease by a chloride salts mixture (NaCl 10 g/kg and KCl 5.5 g/kg) (Gimeno et al., 2001). During aging and drying of salamis, their pH significantly declined at 20 and 40 days compared to 0 day. After 20 days of ripening, all treatments had pH below 4.9. This rapid decline in pH is vital for inhibiting pathogenic microorganisms, increasing the safety of fermented sausages (Leroy et al., 2006).

As shown in Table 1, there were no significant differences of water activities among the treatments, even though T4 had higher A_w than the others at 20 and 40 d. Initial A_w was reduced from around 0.98–0.99 to 0.84–0.86 during the process of ripening and drying. A_w gradually declined during the ripening of the salamis (p<0.05). These A_w results coincide with the results in salami (Horita et al., 2014).

During ripening and drying, T3 had significantly lower VBN than the other treatments during ripening (p<0.05) (Table 2). These results may be due to the replacement of NaCl by KCl, since the replacement was found to decrease the quantity of salt-tolerant flora and inhibit the growth of coliforms (Alinõ et al., 2010). During the ripening time, the VBN of salamis continuously increased (p<0.05). Our study showed that T3 samples containing KCl and NaNO₂ had a positive effect on VBN in salamis during aging.

TBA-reactive substance (TBARS) level is used to set an acceptable limit for rancidity for fresh meat (Ockerman, 1985). As presented in Table 2, T3 had significantly lower TBA than the other treatments at 20 and 40 days (p<0.05). On the other hand, T2 had significantly higher TBA at 20 and 40 days. In general, TBA can be affected by salt concentration (Choi et al., 2016). Low salt concentration tended to reduce lipid oxidation and had a less pro-oxidant effect during aging (Andrés et al., 2004). In this case, it appears that KCl induced slower lipid oxidation than the other curing salts. The TBARS levels increased throughout ripening (p<0.05), which is related to the oxidation by the salt that can be favored by the metallic ions contained as impurities in added curing salts (Lorenzo et al., 2015). KCl may have less impurities acerbating lipid oxidation, which could affect the shelf-life of the salami.

WHC of samples continuously decreased during the ripening and drying phases (Table 2). At 20 and 40 days, T1 showed higher WHC than that of the other samples (p<0.05). The result is supported by Kim et al. (2018) finding highest water-holding capacity could be expected at ionic strength 1.0–1.5 with high pH and the substitution by KCl reduced the ionic strength. As the level of NaCl was reduced, the amount of soluble myofibrillar proteins decreased (Gordon and Barbut, 1992). As a results of reducing ionic strength or changing in charge density by using different salts, the water-holding capacity and the gel strength could be changed (Whiting, 1984).

The microbiological characteristics using different curing agents are presented in Table 2. No significant differences among treatments were observed regarding total plate counts, lactic acid bacteria counts (p>0.05). These reports are in agreement with Alinõ et al. (2010) and Aaslyng et al. (2014) who did not find significant differences among the chloride salts used. LAB showed a strong increase above 8.0 Log CFU/g after 20 days without changes until the end of manufacturing. This result was similar to that reported by Campagnol et al. (2011). Coliforms were not detected in the sodium nitrite-added treatment groups in comparison of the only sodium-added treatments at 0 day, which was because the bacteria have low resistance to acidification (González-Fernández et al., 2006). From this result, the growths of total plate count and LAB were not affected by use of sodium-alternative curing salts.

The influence of curing agents on the salami meat color during ripening is shown in Table 3. T3 samples showed a higher redness value (a*) than the other samples, while T1 showed lowest value (p<0.05). Treatments with sodium-nitrite had higher redness value (a*) compared with T1 (only NaCl-added). Thus, sodium nitrite showed more favorable red color in salami.

T1 samples exhibited a higher yellowness (b*) value than the other samples, while T4 showed a significantly lower value. Choi et al. (2016) showed KCl-added sausages were redder than those made with NaCl. The L*, a*, and b* values in all samples continuously declined during the aging (p<0.05), and the results were same as that from another study (Papadima and Bloukas, 1999). A reduction in a* values of the sausages can be attributed to the oxidation of nitrosylmyoglobin to nitrate and brown metmyoglobin (Gøtterup et al., 2008).

T1 samples had a lower shear force than the other samples (p<0.05, Table 4). In addition, T1 showed lower springiness, cohesiveness, and gumminess values, whereas T4 showed a significantly higher values. The higher texture parameters might be partly attributed to the reduction of bound water during aging (Horita et al., 2014). In addition, NaCl reduction in sausages could destroy the texture-profile of sausages due to the reduced ionic strength and charge density, which causes a decrease in the soluble myofibrillar proteins (Toldrá, 1998). Previous researches have indicated that MgCl₂ or CaCl₂ can increase hardness because of negative attribution of divalent salt (Ca²⁺) to water-binding capacity (Horita et al., 2014). The present study showed an similar trend in texture profiles at the end of aging. Campagnol et al. (2011) reported that the NaCl replacement or reduction could decrease the textural properties of salamis. Lücke (1998) attributed this to the ability of NaCl to solubilize and diffuse muscle myofibrillar proteins forming gel between the meat and fat particles, thus favoring slicing and improving the juiciness and texture of the product. All texture parameters in all samples increased continuously and steadily during the ripening (p<0.05), by the lower moisture content. This findings was similar to that reported by Lorenzo et al. (2015) who reported that shear force showed a marked rise during aging.

A further basic taste sensation called ‘umami’ has been expressed as the taste of monosodium-glutamate (MSG), guanosine 5-monophosphate (GMP), and IMP (Mateo et al., 1996). The role of IMP for the generation of meat odor and flavor has been demonstrated in sensory studies (Aristoy and Toldrá, 2009). As shown in Table 5, treatments with sodium-nitrite had lower contents of GMP and IMP compared with T1 (only NaCl-added) (p<0.05). Thus, sodium nitrite showed desirable umami flavor-enhancing activities in salami ripening. The results show that sodium nitrite might possibly affect more directly the formation of nucleotides-related flavor compounds rather than chloride salts. However, little effect was found on the nucleotides and nucleotides degradation products in cured meat products owing to the antioxidant effect of sodium nitrite (Feng et al., 2017).