Effect of Using Vegetable Powders as Nitrite/Nitrate Sources on the Physicochemical Characteristics of Cooked Pork Products

Chinese cabbages, radish, and spinach grown in the Jeolla-do region of Korea were purchased from local markets, randomly selected, and processed into vegetable powders. The Chinese cabbage was cut into pieces of approximately 3 cm×5 cm after removing the outer leaves and roots. For processing the radish, we first cut and removed 2 cm from the top and the bottom, and then cut into pieces of approximately 2 cm×3 cm×3 cm. The spinach was cut into pieces of 3 cm×5 cm after removing 1 cm from the root. After removing the inedible portions, the cut vegetables were rinsed in distilled, deionized water, and then drained for 1 h in a strainer. Excess water was further removed by pressing 10 pumps using a spinner (OXO Good Grips Salad Spinner, OXO International, Chambersburg, PA, USA). After then, each vegetable was homogenized for 4 min using a food cutter (C6 VV, Sirman, Marsango, Italy). The sides of the food cutter were then scraped, and the vegetables were homogenized for 2 min more. This was repeated until the sample was homogenized for a total of 10 min. The homogenized samples were vacuum-packed in nylon/polyethylene bags (~500 g each) and stored in a freezer (C110AHB, LG Electronics, Changwon, Korea) at −18°C. The frozen samples were then removed from the bags and dried at 60°C in a dryer (EN-FO-3925, Enex Science, Goyang, Korea) for 12 h. The dried samples were pulverized using a blender (51BL30, Waring Commercial, Torrington, CT, USA) for 3 min and passed through a 30-mesh sieve (Test sieve BS0600, Chunggye Sieve, Gunpo, Korea). The vegetable powders were vacuum-packed and stored at −18°C until further use (Table 1). Before testing on meat products, the nitrate content in each of the vegetable powders was adjusted by mixing them with maltodextrin to achieve a target nitrate content of 30,000 ppm (Table 2). A starter culture (CS 299, CHR Hansen, Milwaukee, WI, USA) consisting of Staphylococcus carnosus, sodium nitrite (S2252, Sigma-Aldrich, St. Louis, MO, USA), and sodium ascorbate (#35268, Acros Organics, Geel, Belgium) were obtained from commercial suppliers.

The raw pork ham (M. biceps femoris, M.semitendinosus, and M. semimembranosus) muscles and backfat used in this study were purchased from a local meat processor (Pukyung Pig Farmers Livestock, Kimhae, Korea) at 24–48 h postmortem and prepared as described by Choi et al. (2020) prior to manufacturing the pork products. Ground pork meat and backfat (30 kg/trial) were randomly separated into four groups: traditionally cured control group and three treatment groups cured with vegetable powders (Table 3). All batches also contained some basic ingredients, including 1.5% NaCl, 1.0% dextrose, 0.3% sodium tripolyphosphate, and 0.05% sodium ascorbate, based on the ground meat mixture (75% pork, 15% fat, and 15% ice/water). In the control group, 0.015% sodium nitrite was added to the basic materials. The Chinese cabbage powder (CCP) group was treated with 0.4% CCP and 0.04% nitrate-reducing starter culture (Staphylococcus carnosus). The radish powder (RP) group was treated with 0.4% RP and 0.04% nitrate-reducing starter culture. The spinach powder (SP) group was treated with 0.4% SP and 0.04% nitrate-reducing starter culture. The calculated concentrations of ingoing nitrate from the vegetable powders were 122 ppm for CCP, RP, and SP treatments. For processing the control samples, ground pork meat, pork fat, NaCl, dextrose, sodium tripolyphosphate, sodium nitrite, and sodium ascorbate were added to a mixer (5K5SS, Whirlpool, St. Joseph, MI, USA) with half of the ice/water and then mixed for 5 min. The remaining ice/water was then added, followed by another 5 min of mixing. For the treatment groups, each vegetable powder and starter culture were added to the pork product with the same ingredients and using the same procedure as the control, with the exception of sodium nitrite. Each batch from the control and treatment groups was stuffed into conical tubes (approximately 50 g each) using a stuffer (MOD.5/W Deluxe, Tre Spade, Torino, Italy). The stuffed tubes were centrifuged at 2,000×g for 10 min (Combi R515, Hanil Science Industrial, Incheon, Korea) to remove air pockets. All tubes were then closed with caps and stored in a refrigerator at 10°C for the control or held for 2 h at 40°C in an incubator (C-IB4, Changshin Science, Pocheon, Korea) for the treatment groups. The tubes from each batch were then cooked to an internal temperature of 75°C in a 90°C water bath (MaXturdy 45, Daihan Scientific, Wonju, Korea). The temperature was monitored using a thermometer (Tes-1384, Ketech Scientific Instrument, Kaohsiung, Taiwan). Once cooked, the samples were immediately cooled for 20 min on slurry ice and stored at 2°C–3°C in the dark prior to analysis. All experiments were performed in triplicate.

The moisture content of the hot air dried vegetable powders was determined using the drying method (AOAC, 2016).

A pH meter (Accumet AB150, Thermo Fisher Scientific, Singapore) was used to measure pH on five grams of the vegetable powders or the cooked meat products homogenized in 45 mL of distilled water.

Each meat sample in the conical tube was weighed before cooking, and then weighed again after cooking and cooling. The cooking yield was determined using the following equation:

The Commission Internationale de l’Eclairage (CIE) L* (lightness), a* (redness), and b* (yellowness) values were measured on the freshly cut surfaces of each cooked sample using a colorimeter (Chroma Meter CR-400, illuminant C, 2° standard observer; Konica Minolta Sensing, Osaka, Japan) calibrated with a white plate (L* 94.90, a* –0.39, b* 3.88).

The nitrate ion (NO₃⁻) content in the prepared vegetable powders was determined using the zinc reduction method described by Merino (2009). The obtained results were converted in terms of the concentration of sodium nitrate and expressed as ppm. The residual nitrite content in the cooked meat products was analyzed using the procedures of the AOAC (2016) and reported as ppm.

Nitrosyl hemochrome and total pigment content were determined on cooked cured products after extraction in 80% acetone and acidified acetone using a method established by Hornsey (1956). The nitrosyl hemochrome concentration (ppm) was calculated as A₅₄₀×290. The total pigment concentration (ppm) was calculated as A₆₄₀×680. Finally, curing efficiency (%) was calculated using the following equation:

Malondialdehyde (MDA) content was measured based on 2-thiobarbituric acid reactive substances (TBARS) values using a distillation method reported by Tarladgis et al. (1960). The TBARS values were reported as milligrams of MDA per kilogram of the cooked samples (mg MDA/kg).

The experimental design of this study was a randomized block design with four groups (control and three vegetable powder treatments). The obtained data were statistically analyzed using the Proc GLM (general linear model) procedure of the SAS program (SAS, 2012). If the model revealed a significant difference (p<0.05), the significance of the means was further separated using Duncan’s multiple range test.

After hot air-drying, the pH of CCP, RP, and SP were 6.35, 5.14, and 6.10; the pH of CCP was significantly higher (p<0.05) than those of the others (Table 1). The highest moisture content was observed (p<0.05) in CCP (5.68%), followed by RP (3.74%), while SP had the lowest (p<0.05) moisture content (2.78%). The mean nitrate contents in CCP, RP, and SP were 44,077 ppm, 66,178 ppm, and 120,315 ppm, respectively, with SP showing the highest (p<0.05) nitrate content and CCP the lowest (p<0.05). These results were consistent with the trends of previous studies on fresh Chinese cabbage, radish, and spinach (Chung et al., 2003; Suh et al., 2013). Suh et al. (2013) observed that the nitrate content in Chinese cabbage, radish, and spinach collected from retail markets in Korea averaged 1,059.9 mg/kg, 1,494.0 mg/kg, and 2,123.8 mg/kg, respectively. These results indicated that the nitrate concentration in vegetables was increased by the drying process in our study. Further, CCP, RP, and SP all had considerable higher nitrate contents than commercially available celery juice powders containing approximately 30,000 ppm of nitrate (Sindelar and Houser, 2009; Sebranek et al., 2012). This suggests that prepared vegetable powders may be promising natural alternative nitrate sources that could replace synthetic nitrite for the production of alternatively cured meat products.

Thus, in order to standardize the nitrate concentration in the vegetable powders used, we adjusted the nitrate content in each vegetable powder by adding sufficient maltodextrin to achieve a target nitrate content of 30,000 ppm, which corresponds to the nitrate content in commercially available celery powder (Table 2). The nitrate content in the standardized vegetable powders ranged from 30,407 ppm to 30,514 ppm, showing that the measured contents were very close to the target levels. Similarly, the pH of the standardized vegetable powders also followed the same trend as those of the hot air-dried powders prepared initially. As expected during the manufacturing of the vegetable powders, their colors varied based on their plant origins. RP had significantly higher (p<0.05) CIE L* and CIE a* values than the powders from leafy vegetables, while SP had the lowest CIE L* and CIE a* (p<0.05). CCP had the highest CIE b* values (p<0.05), followed by RP, and SP had the lowest. Sebranek and Bacus (2007) described that vegetable products with low levels of vegetable pigment are ideal for use as natural sources of nitrate for meat processing. Besides, high levels of plant pigments could affect the color of the final meat product (Lavieri et al., 2014; Horsch et al., 2014).

The traditionally cured control and alternatively cured pork products did not have significantly different (p>0.05) cooking yields (Table 4), which was consistent with the results of Krause et al. (2011), who reported that the cooking yields of hams naturally cured with vegetable juice powder and those cured with nitrite were not significantly different. Kim et al. (2019a) and Kim et al. (2019b) also found that cooking loss was not significantly different between pork loins cured with nitrite and those treated with fermented spinach or Swiss chard, which is also consistent with our results. The addition of vegetable powders to cured meat products also affected the pH of the final products. The alternatively cured products (pH range 6.26–6.29) had higher pH (p<0.05) than the traditionally cured control (pH 6.20), although the difference of numerical value is small. However, there were no differences (p>0.05) in the pH of the products with the different vegetable powders, suggesting that the effects of the different vegetable powders on the pH of the cured meat products were hidden, although vegetable powders used had different pH values. This was probably due to the buffering capacity of the meat used or the added phosphates (Krause et al., 2011; Sebranek and Bacus, 2007).

The CIE L* values were highest (p<0.05) for the nitrite-added control and lowest (p<0.05) for the SP treatment (Table 4). Previous studies have shown that powders made from cabbages, Chinese cabbages, radishes, spinaches, and Swiss chard lowered the lightness of alternatively cured meat products (Kim et al., 2019a; Kim et al., 2019b; Ko et al., 2017). Further, our observation of the decrease in lightness after the addition of SP is consistent with the results of Kim et al. (2019a), who found that pork loin marinated with fermented spinach along with organic acid showed lower lightness than those with nitrite and ascorbic acid. In this study, the CIE a* values in alternatively cured products ranged from 8.17 to 10.74 (Table 4), higher than those found in cook pork sausages (0.3–4.8) with 3% fermented spinach, lettuce, celery, or red beet (Hwang et al., 2018). The RP treatment and the control did not show a difference (p>0.05) in their CIE a* values (Table 4), while CCP and SP treatments had lower CIE a* values than the control. Recently, similar results were reported by Ahn et al. (2019), who investigated the effects of dried radish leaves and roots on the quality of pork patties. They found that the combination of powders made from radish leaf and root did not change the redness of pork patties. Ko et al. (2017) reported that cooked sausages with pre-converted nitrite from radish were less red than sausages with sodium nitrite, but significantly redder than sausages treated with cabbage and Chinese cabbage powders, which was partially consistent with our findings. In this study, the lowest CIE a* values shown by the SP treatment (Table 4). According to a report from Bohn et al. (2004), leafy vegetables such as spinach and lettuce have a much higher chlorophyll content than other vegetables. Considered in light of our results, this suggests that the inherent green color in the prepared SP may have affected the CIE a* values and contributed to the decreased redness in the alternatively cured meat products. This is supported by the results of Kim et al. (2019c), who reported that when pork loins were cured with 10% spinach extract and 0.02% starter culture, followed by incubation for 3–12 h, the CIE a* values of the cooked products were significantly lower than those of control products treated with sodium nitrite. In our study, CIE b* values of cooked pork products were increased (p<0.05) by CCP or SP addition (Table 4). However, the RP treatment and the control resulted in similar (p>0.05) CIE b* values, although the RP treatment did show a slight increase. This effect of vegetable-based sources of nitrite/nitrate on the CIE b* values of meat products has been previously reported for hams with celery juice concentrate (Horsch et al., 2014), mortadella-type sausages with parsley extract powder (Riel et al., 2017), and pork loins cured with fermented spinach (Kim et al., 2019a). In many of these previous studies, the authors have speculated that the characteristic color of plant-based ingredients may affect the yellowness of the naturally cured products. Our results suggest that the RP treatment was relatively less affected by the addition of vegetable powders, compared to other treatments with the leafy vegetable powders. RP would therefore be a promising natural replacement for synthetic nitrite for the production of alternatively cured meat products.

The residual nitrite contents of the final products were 75.98, 34.73, 40.08, and 43.46 ppm for the control, CCP, RP, and SP treatments, respectively (Table 5). Many studies have established the fact that the meat undergoes a depletion in nitrite content during thermal processing and chilling (Choi et al., 2020; Honikel, 2008; Xi et al., 2011). Li et al. (2013) reported that only 65% of the initially added nitrite was detected in the product immediately after the production of dry-cured sausages. Xi et al. (2011) found that nitrite was depleted by about 50% following heating, which was similar to the trend shown by the control group in our study. Moreover, the alternatively cured pork products had lower (p<0.05) residual nitrite content than control, while RP treatment were not different from CCP and SP treatments. Other studies have also reported similar trends, with lower residual nitrite content being reported in products cured with vegetable sources and nitrate-reducing starter cultures (Choi et al., 2020; Krause et al., 2011; Riel et al., 2017; Sebranek and Bacus, 2007; Shin et al., 2017). Shin et al. (2017) reported that the pork patties with Swiss chard or celery powder showed a lower residual nitrite content than those with sodium nitrite. In addition, the results obtained in this study suggest that the nitrite contents in the alternatively cured products could be lower than those in the control containing 150 ppm nitrite, because the initial nitrate levels (122 ppm) in the vegetable powders were relatively lower, even if the added nitrate was completely converted to nitrite after incubation.

Cured meats are typically formulated with sodium nitrite or nitrate, which are responsible for the pink cured meat pigment (nitrosyl hemochrome) during subsequent cooking (AMSA, 2012). In naturally cured meat products, nitrite generated from the reduction of nitrate by the starter culture becomes available to participate in normal curing reactions, resulting in cured meat properties similar to nitrite-added meats (Terns et al., 2011). Interestingly, the SP treatment had the highest nitrosyl hemochrome contents (p<0.05) although this treatment had the lowest CIE a* values. The CCP and RP treatments resulted in similar (p>0.05) nitrosyl hemochrome contents to the conventionally cured control, even though products with CCP and RP had a lower residual nitrite content than the control. It is possible that nitrite converted from nitrate in vegetable sources by nitrate-reducing starter culture was already involved in the curing reactions in the reactive form during the meat processing (Sindelar et al., 2007b), which could explain the formation of the cured pigments in the products with vegetable powders.

Total pigment contents in cooked pork products ranged from 49.75 to 62.11 ppm (Table 5). The total pigment contents in the control, CCP, and RP treatments were lower (p<0.05) than those in SP treatment, but were similar (p>0.05) to each other. The general trends in total pigment contents were similar to those observed for the nitrosyl hemochrome contents, suggesting that nitrosyl hemochrome formation may be positively correlated with the total pigment content in meat products. Consistent with our results, Shin et al. (2017) reported that higher nitrosoheme pigment in pork patties with pre-converted Swiss chard powder resulted in higher total pigment content. Sullivan et al. (2012) also found that commercial meat products having the highest cured pigment content had the highest total pigment content.

Curing efficiency is the percentage of total pigment converted to nitroso pigments; the parameter also indicates the degree of cured color fading (AMSA, 2012). In this study, the curing efficiency of the cured pork products ranged from 73.00% to 83.37% (Table 5). The curing efficiency of the RP treatment (83.05%) was not different (p>0.05) from that of the control, and both were higher (p<0.05) than those of the CCP (81.54%) and SP (73.00%) treatments, which were much higher than those reported by Sullivan et al. (2012) in commercially available naturally cured frankfurters (31.3%–46.2%), hams (26.5%–37.3%), and bacons (22.8%–45.6%). According to Pearson and Tauber (1984), the percentage conversion of nitrosoheme to total heme in well-cured meats typically exceeds 80%. In this study, the nitrite-added control, CCP, and RP treatments showed curing efficiencies of over 80%. Shin et al. (2017) obtained similar results upon treating pork patties with 2% pre-converted Swiss chard powder or celery powder. Therefore, CCP and RP may be potentially suitable replacements for sodium nitrite as a natural source of nitrate for curing meat products, in terms of curing efficiency. Surprisingly, SP treatment resulted in the lowest (p<0.05) curing efficiency in this study (Table 5), suggesting that lower levels of nitrosyl hemochrome were converted to total pigment. This may explain the low CIE a* values observed in the SP treatment, despite high nitrosyl hemochrome and total pigment content. Kim et al. (2019b) reported similar results for redness and curing efficiency in cured pork loins treated with fermented Swiss chard.

Nitrite is a highly effective antioxidant that can inhibit lipid oxidation in cooked meat products. As little as 50 ppm nitrite could reduce rancidity in cured meat products (Morrissey and Techivangana, 1985; Sebranek, 2009). The nitrite-added control had lower TBARS values (p<0.05) than the alternatively cured pork products (Table 5). Apparently, the higher TBARS values obtained upon treatments with vegetable powders could have resulted from lower residual nitrite contents obtained in the present study. The alternatively cured products were not expected to have higher TBARS values than the control because vegetables contain natural antioxidants that can reduce lipid oxidation (Septembre-Malaterre et al., 2018; Steinmetz and Potter, 1991). It could be assumed that the addition of 0.4% vegetable powders was not sufficient to compensate for the low residual nitrite in the final products as a result of the loss of natural antioxidants during powdering by the drying process (Kamiloglu et al., 2016; Raikos et al., 2015).

However, among the alternatively cured products, the TBARS values were not different (p>0.05). Tarladgis et al. (1960) suggested 0.5−1.0 mg MDA/kg as a threshold range for detecting off-odor in cooked pork. Thus, our measured TBARS levels (0.083−0.094 mg MDA/kg) in the alternatively cured products were considerably lower than those indicated by Tarladgis et al. (1960). Other studies on naturally cured meat products have shown varying effects of vegetable sources on lipid oxidation, depending on types and added quantity of vegetables (Djeri and Williams, 2014; Kim et al., 2019b; Sindelar et al., 2007b; Sucu and Turp, 2018). Djeri and Williams (2014) reported that turkey bologna with cherry powder and/or celery powder containing pre-generated nitrite had similar TBARS values to the nitrite-added control. However, Kim et al. (2019b) found low TBARS value in the nitrite-added control, compared to pork treated with fermented spinach; they also reported that TBARS values decreased with increasing amounts of fermented spinach extracts. Further, Sindelar et al. (2007b) reported that the nitrite-added control had a significantly lower TBARS value than hams cured with either 0.2% or 0.35% celery juice powder when incubated for 120 min, which was consistent with our results.