Effects of Chinese pickled and dried mustard on nutritional quality, sensory quality, and shelf life of steamed pork belly

Abstract Steamed pork with pickled and dried mustard (PDM) is a famous Chinese dish. Here, we examined the effects of PDM on nutritional quality, sensory quality, and shelf life of steamed pork belly. Proximate composition, lipid oxidation, fatty acid profiles, protein hydrolysis and oxidation, sensory evaluation, and induction period (IP) of steamed pork belly were determined after addition of different levels (0–100%, WPDM/W pork belly) of PDM. The results demonstrated that PDM could significantly (p < .05) enhance the loss of moisture and fat, increase the ratio of unsaturated to saturated fatty acids, and decrease lipid and protein oxidation in steamed pork belly. Additionally, IP values and steamed pork belly preservation times increased as the amount of added PDM increased. Best sensory quality was achieved when moderate levels of PDM (40%) were added to steamed pork belly. These findings provide insights into the beneficial effects of PDM on steamed pork belly.

Fat and protein are the main nutritional components in pork belly, and their oxidative and hydrolytic reactions in meat are the most important factors affecting quality and shelf life. Ali et al. (2015) reported that deterioration of frozen meat during storage occurs because of lipid and protein degradation. Oxidative reactions occur during storage and processing of meat and meat products, and processing steps such as mincing, cooking, and salt addition, which promote the formation of reactive oxygen species (ROS) and increase the susceptibility of products to oxidation (Nakyinsige et al., 2015). Antioxidants such as gallic and protocatechuic acids present in PDM have the ability to absorb ROS and suppress or inhibit peroxidation and oxidation of lipids and proteins (Harlina et al., 2015).
There is a fine relationship between the degree of oxidation in meat and its shelf life. Therefore, it is necessary to evaluate lipid and protein oxidation, including fatty acid profiles of steamed pork belly, to determine the effects of PDM on meat.
Accordingly, in this study, we aimed to evaluate the effects of Chinese PDM on nutritional quality, sensory quality, and shelf life of steamed pork belly, which may provide useful information for preparation of the Chinese traditional steamed pork with PDM, both for domestic cooking and industrialized production. Additionally, the findings of this study may provide a foundation for expanding the applications of PDM for economic benefits.

| Raw materials and sample preparation
The brand of Xian Heng PDM (Meigan cai) was obtained from Xianheng Foodstuffs Co., Ltd (Shaoxing, China). Pork belly (lean/fat ratio of about 1:2) was purchased from Wal-Mart Supermarket in Hangzhou, Zhejiang Province, China.
Steamed pork with PDM was prepared according to Chinese traditional domestic cooking methods as described by Liu, Jiang, Lian, Zhang, and Ma (1999), with some modifications. Pork belly (about 250 g per replicate in a bowl) was randomly selected and precooked in boiling water (pork:water ratio of 1:4, w/v) for 1 min and then cut into 3 × 4 × 1-cm 3 pieces. Five grams of sugar was added with 0, 50, 100, 150, 200, or 250 g PDM to each bowl, and all materials were then mixed evenly. The entire bowl was then placed in a pressure cooker (Supor, China) for 20 min for high-pressure steaming on an induction cooker (2,300 W), followed by cooling for 5 min. The pressure cooker was then opened, and the pork belly was separated from each bowl. The pork belly was used for texture profile analysis and sensory evaluation after cooling to room temperature or was minced with a mincing machine (Joyoung, China) and then stored at −20°C for further determination of other physical and chemical properties.
Pork belly samples were divided into seven groups as follows: the control group, comprising raw pork belly; steamed pork belly alone; steamed pork belly treated with 20% (50 g) PDM; steamed pork belly treated with 40% (100 g) PDM; steamed pork belly treated with 60% (150 g) PDM; steamed pork belly treated with 80% (200 g) PDM; and steamed pork belly treated with 100% (250 g) PDM. All treatments for each determination were carried out in triplicate.

| Analysis of proximate composition
Moisture content was determined using a halogen moisture detector (DHS-20A; Shanghai Eastsen Analytical Instrument Co., Ltd, China).
Crude protein, ash, and sodium chloride contents were determined according to the National Standard of China. Briefly, the crude protein content was determined using the Kjeldahl procedure with a nitrogen-to-protein conversion factor of 6.25 (GB5009.5-2010).
The ash content was determined by heating 3 g samples in a furnace at 550°C for about 4 hr until a constant weight was obtained (GB5009.4-2010). Sodium chloride content was determined according to the silver nitrate titration method (GB/T 12457-2008).

| Total lipid extraction and content analysis
Total lipids were extracted following the method of Folch, Lees, and Stanley (1957), with some modifications. Briefly, the minced pork belly was homogenized with a 2:1 chloroform-methanol mixture to a final dilution of 10-fold the volume of the sample and extracted in a water bath for 15 min at 60°C, followed by filtration. The extracted lipids were purified by evaporating the solution under vacuum conditions, using a rotary evaporator (RE-52AA; Yarong Biochemical Analysis Co., Ltd, China), and then stored at −80°C for further analysis or dried at 100°C for about 2 hr to a constant weight to determine the total lipid content, expressed as g/100 g fresh sample.
Additionally, chloroform was replaced with dichloromethane in the experiment to reduce toxicity.

| Fatty acid analysis
The fatty acid composition of pork belly was determined by gas chromatography (GC). The methyl ester preparation method was carried out as described by Ichihara, Shibahara, Yamamoto, and Nakayama (1996) and . Briefly, 10 mg of extracted lipids was dissolved in 2 ml hexane and then added to 4 ml of 2 mol/L methanolic KOH. The mixture was then shaken for 30 min at 25°C, and the hexane layer was used for GC analysis, which was performed using a GC7890A gas chromatograph (Agilent Technologies Inc.) equipped with an automatic sampler, a split/splitless injector, a fused silica capillary column (DB-23; 30 m long × 0.32 mm id × 0.25 μm film thickness; Agilent Technologies Inc.), and a flame ionization detector. Nitrogen was used as the carrier gas with the flow rate set at 9 ml/min, the pressure at 6.6016 psi, and a split ratio of 5:1. The temperature of the flame ionization detector was maintained at 280°C, and that of the injector was maintained at 270°C. The column temperature was programmed as follows: maintained at 120°C for 5 min, increased from 120 to 190°C at 5°C/min, maintained at 190°C for 12 min, increased to 210°C at 2.5°C/min, and finally maintained at 210°C for 10 min. A 37 fatty acid methyl ester (FAME) mixture standard (Sigma-Aldrich Co. LLC.) was run under the same conditions.

| Lipid oxidation measurement
Thiobarbituric acid (TBA) and peroxide value (POV) were determined to evaluate the extent of lipid oxidation. TBA measurement was performed as described by Sorensen and Jorgensen (1996), with some modifications. Briefly, the minced sample (10 g) was homogenized (15,000 rpm, 30 s) with 50 ml of a 7.5% trichloroacetic acid (TCA) solution containing 0.1% ethylenediaminetetraacetic acid, using a homogenizer (FSH-2A; Jiangsu Zhenxing Instrument Factory, China). After filtration, 5.0 ml extract was mixed with 5.0 ml of 0.02 mol/L aqueous TBA in stoppered glass tubes and incubated in a water bath at 100°C for 1 hr and then cooled for 10 min in cold water. Absorbance was measured at 532 nm using a UV spectrophotometer (2550; Shimadzu, Japan). The results were calculated from the standard curve of 1,1,3,3-tetraethoxypropane and expressed as milligrams of malondialdehyde (MDA) per kilogram of pork belly sample. POV was assessed according to AOAC method 965.33 and expressed as milliequivalents (meq) of peroxide per kilogram of pork belly sample.

| Soluble protein extraction and concentration determination
The soluble protein of pork belly was extracted with sodium phosphate buffer. Briefly, 2 g minced pork belly was homogenized (15,000 rpm, 30 s) with 10 ml of 20 mmol/L sodium phosphate buffer (pH 6.8), extracted for 1 hr at 4°C, and then filtered with Whatman No. 1 filter paper. The filtrate was preserved at −20°C for further determination. The protein concentration of the filtrate was determined using a BCA kit (Sigma), and bovine serum albumin was used as the standard.

| Protein hydrolysis measurement
The degree of hydrolysis of the protein sample was determined by the trinitrobenzenesulfonic acid (TNBS) method, as described by Adlernissen (1979), with minor modifications. Briefly, 0.020 ml of the protein extraction sample was mixed in a test tube with 0.980 ml sodium phosphate buffer (pH 8.0). One milliliter of 1.0% TNBS solution was added, and the test tubes were shaken and placed in a water bath at 50°C for 60 min. After 60 min, 2.0 ml of 0.1 N HCl was added to terminate the reaction, and the absorbance was then read against water at 340 nm. The degree of hydrolysis of the protein sample was expressed as millimoles of free amino groups per gram of pork belly soluble protein, and glycine was used to construct a standard curve.

| Protein oxidation measurement
Protein carbonyl and sulfhydryl measurements are common methods for determining protein oxidation. The protein carbonyls were determined by reacting proteins with 2,4-dinitrophenylhydrazine (DNPH), as described by Ali et al. (2015), with minor modifications.
Briefly, a 0.7-ml protein sample extract was reacted with 0.3 ml of 10 mmol/L DNPH for 1 hr at room temperature. Proteins were precipitated with 1 ml of 40% TCA and washed three times with 1.0 ml of an ethanol/ethyl acetate mixture (1:1). The protein was dried and dissolved in 3 ml of 6 mol/L guanidine hydrochloride. The absorbance was measured at 370 and 280 nm to determine carbonyl and protein contents, respectively. Results are expressed as the ratio of the absorbance at 370 nm to that at 280 nm. The sulfhydryl content was determined by reacting with 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) according to Srinivasan and Hultin (1997), with some modifications. One milliliter of the protein sample extract was transferred to a tube and dissolved in 2 ml sodium phosphate buffer (pH 8.0).
The mixture was then incubated with 0.5 ml of 10 mmol/L DTNB reagent at room temperature away from light for 1 hr, and the absorbance was measured against phosphate buffer at 412 nm. The results were expressed as millimoles of total free sulfhydryl groups per gram of pork belly soluble protein.

| Texture profile analysis (TPA)
Texture profile analysis was carried out as described by Bourne (1978). A texture analyzer (TA-XT2i 2; Stable Micro Systems, Surrey, UK) with a 5-mm-diameter cylindrical probe was used. The samples were compressed to 50% of their original height in a two-cycle compression using a crosshead speed of 1.0 mm/s. The cubic muscle layer and fat layer (1 × 1 × 1-cm 3 ) cut from pork belly samples were determined five times for every treatment. Hardness, springiness, cohesiveness, gumminess, chewiness, and resilience parameters were calculated for each sample.

| Sensory evaluation
Quantitative descriptive analysis for sensory evaluation was carried out as described by Morita, Kubota, and Aishima (2003), with some modifications. Briefly, 16 panelists (eight men and eight women, ages 20-30 years) with previous experience in food sensory evaluation participated in the study. Each member was randomly given one 10-g pork belly sample from each group. Samples were spread on glass plates, and a glass of water was provided for gargling be-

| OXITEST analysis
Oxidation tests were performed using an OXITEST reactor (Velp Scientifica, Usmate, Milan, Italy), as described by Verardo et al. (2013), with some modifications. Six grams of minced pork belly was used for analysis in each chamber. The conditions were set at 80, 90, and 100°C, with an initial oxygen pressure of 6 bar. Each sample was measured four times for each treatment, and the induction periods (IPs) were calculated using the OXITEST reactor.

| Statistical analysis
All experiments were carried out in triplicate (n = 3). Results were expressed as means ± standard deviations. Treatment effects were analyzed by one-way analysis of variance (ANOVA), using SPSS 20.0 (SPSS Inc., Chicago, IL). Duncan's multiple range tests were performed to determine mean separation. Differences with p values of less than .05 were considered statistically significant and those of less than .01 were considered statistically highly significant.

| Proximate composition
The results for proximate composition of raw and steamed pork belly in the presence of PDM are shown in Table 1. The moisture, crude protein, total lipid, ash, and sodium chloride contents of the raw pork belly sample were 36.92, 20.47, 39.30, 1.10, and 0.20 g/100 g fresh weight, respectively. After steaming, the moisture and crude protein contents of pork belly increased significantly (p < .05), the total lipids decreased significantly (p < .05) to 33.24%, and the ash and sodium chloride contents were not significantly (p > .05) altered. This trend was consistent with that in a report by , who demonstrated that steamed cooking affected the chemical composition of fish fillets.
The increased moisture could be explained by the steaming process, which involves the use of steam from heated water for cooking and may cause samples to hold more moisture.  also observed a decrease in total fat in pork belly samples after cooking and explained this change by the melting and liquefaction of fat during cooking, which may also cause increased moisture and crude protein percentages.
Compared with that in steamed pork belly without PDM, the addition of PDM decreased moisture and total lipid contents. The moisture content of samples after addition of PDM was 37.85-40.26 g/100 g fresh weight and decreased as the amount of PDM increased. When 80% PDM or more was added, the moisture content did not differ significantly (p > .05) from that of the raw sample, that is, addition of 80% or more PDM prevented pork belly samples from holding moisture during steaming. This result could be explained by the physical barrier function and waterabsorbing capacity (diffusion motivated) of PDM, attributed to the physicochemical characteristics of the dehydrated vegetable.
Additionally, PDM also had lipid-binding activity, decreasing the total lipid contents of pork belly from 33.99 to 28.02 g/100 g fresh weight. Thus, PDM, which covered the pork belly during steaming, served as an absorbent medium to lower moisture and fat contents of pork belly, which could explain the extended shelf life of steamed pork.
As shown in Table 1, crude protein and ash contents tended to increase as the amount of PDM increased from 0% to 100% as a result of decreased moisture and total lipid contents. The sodium chloride content increased significantly (p < .05) in samples treated with PDM compared with those in untreated samples; however, this difference was not concentration dependent.
PDM is a type of homemade preserved vegetable, and a small amount of pure, granulated, noniodized pickling salt is added during preparation . Thus, some of the salt may have diffused into the pork belly during the steaming process as no extra salt was added in the experiment. Nevertheless, 20% PDM treatment made the pork belly sufficiently salty (0.76-0.84 g/100 g fresh weight, around four times that of the raw or untreated sample), and higher amounts of added PDM did not have additional effects. Because sodium chloride plays an important role in meat preservation (Verma & Banerjee, 2012), steamed pork with PDM treatment could be stored longer than that without PDM treatment.

| Lipid oxidation
Peroxide value is the primary product of lipid oxidation generated by attack of oxygen on the double bond in fatty acids, and TBA represents the content of secondary lipid oxidation products, primarily aldehydes, which contribute to off-flavors in meat (Cao et al., 2013).
TA B L E 1 Effects of Chinese pickled and dried mustard on the proximate composition of steamed pork belly (g/100 g fresh weight) POV and TBA are commonly used as indicators of lipid oxidation and are extensively used to detect oxidative deterioration of fatcontaining foods (Harlina et al., 2015).
In this study, the TBA ( Figure 1a) and POV (Figure 1b) Huang et al. (2012) and Li et al. (2012), and the relatively low proportion of fat exposed to oxygen (pork belly was covered with PDM, and increased PDM resulted in lower proportion of fat exposed to oxygen). Therefore, PDM could extend the shelf life of steamed pork belly by reducing the degree of lipid oxidation.

| Protein hydrolysis and oxidation
Next, the content of free amino groups was determined to evaluate the degree of protein hydrolysis in pork belly (Figure 2a). The findings showed that the content of free amino groups was 1.94 mmol/g soluble protein in raw pork belly, which was significantly (p < .05) lower than the values measured for other treatments. This could be explained by the observation that high temperatures promote the heat denaturation of proteins (Zhang, Yao, et al., 2013). Increasing the addition level of PDM from 0% to 40% caused the free amino content to increase from 3.40 to 4.30 mmol/g soluble protein (p < .05). Additionally, the free amino content remained steady and did not increase significantly (p > .05) as greater amounts of PDM (40%, 60%, 80%, and 100% PDM) were added. The results showed that PDM promoted the hydrolysis of protein in steamed pork belly and that the degree of protein hydrolysis was dependent on the amount of added PDM within a certain range. This result could be explained by the Maillard reaction in samples, for which glucose from PDM and protein hydrolysates from pork acted as substrates.
The oxidative deterioration of proteins in pork belly treated with PDM was monitored by measuring the amount of carbonyls ( Figure 2b) and free sulfhydryl (SH; Figure 2c) groups. Higher carbonyl contents and lower sulfhydryl contents were indicators of a higher degree of protein oxidation (Ali et al., 2015). Notably, for convenience, the carbonyl values in our study were expressed as the ratio of the absorbance at 370 nm to that at 280 nm, equivalent to the carbonyl content per unit protein in the sample. The carbonyl value increased from 0.06 to 1.00, and free SH decreased from 237.29 to 114.43 mmol/g soluble protein after steaming the raw pork belly, indicating that steaming caused significant (p < .05) protein oxidation in the pork belly sample. This was consistent with the results of Roldan, Antequera, Armenteros, and Ruiz (2014), who showed that TA B L E 2 Effects of Chinese pickled and dried mustard on fatty acid profiles of steamed pork belly (area % of total fatty acids) These results clearly showed that addition of PDM affected protein oxidation in steamed pork belly, consistent with the findings of lipid oxidation in the present study. In fact, previous studies have also reported correlations between lipid and protein oxidation. Soyer, Ozalp, Dalmis, and Bilgin (2010) reported that primary and secondary lipid oxidation products could act as substrates for protein oxidation; thus, once the oxidation of lipids started, the oxidation of proteins would also occur. Similarly, the lower degree of protein oxidation in meat could be a result of the lower degree of lipid oxidation inhibited by PDM. Moreover, the four sensory quality attributes showed the same trend as the total mean score. Thus, the sensory quality of steamed pork belly was the best when 40% PDM was added; in these samples, the highest mean scores were observed for each attribute (color = 18.50, aroma = 20.06, flavor = 21.00, and texture = 21.00). When the amount of PDM was lower, steamed pork belly did not exhibit a sufficient brown color, typical potherb mustard pickled flavor, or acidic taste (Zhao, Tang, & Ding, 2007

| CON CLUS IONS
In the Chinese traditional dish, steamed pork with PDM, PDM plays an essential role in nutritional quality, sensory quality, and shelf life of steamed pork belly. In this study, we found that PDM decreased the moisture and total lipid contents, but increased the crude protein, ash, and sodium chloride contents in steamed pork belly. In addition, lower TBA and POV values as well as higher UFA/SFA ratios in PDMtreated samples were observed, confirming that PDM effectively reduced lipid oxidation in steamed pork belly. PDM also promoted protein hydrolysis and inhibited protein oxidation of steamed pork belly to some degree. Therefore, the OXITEST analysis showed that steamed pork belly had a longer shelf life with higher addition of PDM under the same induced temperature. Furthermore, all of the changes in chemical composition in response to addition of PDM could affect the sensory and nutritional qualities of steamed pork belly. Although 100% PDM treatment could maximize the shelf life of the samples, the optimal sensory qualities of steamed pork belly were achieved by the addition of 40% PDM according to TPA and sensory evaluation.
These results provided a theoretical basis for the preservation mechanism of PDM. However, further studies are required for evaluating the optimal sensory characteristics of the dish prepared with 40% PDM.

ACK N OWLED G M ENTS
This work was supported by Chinese National Science-Technology Support Plan Project (grant no. 2014BAD04B01).

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L R E V I E W
This study does not involve any human or animal testing.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all study participants.