Changes in the microbiological, physicochemical properties of Chinese traditional fermented Suan rou at ripening fermentation

Abstract This study characterized the changes in the microbiological, physicochemical properties of Suan rou during fermentation via three different techniques (Technique A is a traditional production process. Based on technique A, technique B adds a total of 200 g of sucrose to the thinly sliced meat, and technique C changes the amount of salt in the thinly sliced meat to 200 grams.). Compared to batch A, the samples from batches B and C featured more rapid reduction in pH and generated more TA. Myofibrillar proteins in batches B and C showed higher degradation rate, and several low‐molecular‐weight metabolites were determined on the basis of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS‐PAGE) gel lanes. The contents of thiobarbituric acid (TBARS) and total volatile base nitrogen (TVB‐N) and the growth of spoilage bacteria and pathogens were suppressed in the three batches. A relatively compatible acid–salinity proportion was presented in the Suan rou of batches A and B compared with that of batch C. The results show that the Suan rou made by B technology was more palatable acid flavor and abundant nutrition.

generating lactic acid, thus making the product appropriately sour and effectively safe (Zeng et al., 2013a). Staphylococcus, yeast, and mold play important roles in flavor and color enhancement (Nie et al., 2014). Given that pH is reduced close to the isoelectric point of proteins in fermented meat products, a portion of the fermented meat product will be dehydrated, resulting in the reduced water activity (aw) of the products (Yoo et al., 2007). Textual parameters, such as hardness, springiness, and cohesiveness, will change accordingly (Weiss et al., 2010). The inherent endogenous enzymes and portions of the LAB, yeast, and Staphylococcus in naturally fermented meat products exhibit certain proteinase and lipase activities; these enzymes can degrade macromolecules, such as proteins and fats, in meat products and form corresponding small molecular substances, such as polypeptides, free amino acids (FAA), and free fatty acids (FFAs) (Yang et al., 2015). Furthermore, the reaction products from autoxidation of Lipids, which involves inanimate objects or spices, and flavors generated by carbohydrates are also produced in the natural fermentation of meat products (Spaziani et al., 2009). All of these small molecular compounds can notably enhance the palatability and flavor of fermented meat products, thus improving their quality.
Dominant microbe production and complicated physicochemical and substance flavor changes also exist in the natural fermentation of Suan rou. The corresponding dominant microbes and metabolites of biochemical reactions will influence the final sensory property of Suan rou. A more common is the existence of three different formulations of Suan rou in western Hunan, in which the technical A is to add salt, pepper, star anise, cinnamon powder and ginger slices to the meat slices. On the basis of technical A, technical B increases sucrose and corn flour, and technical C is to reduce salt content and add corn flour. Sucrose, as an energy source and a sweetener in a product, has an effect on the physical and chemical parameters of the product.
Salt content also affects the diversity of microorganisms and the flavor of the product. Therefore, there will be some quality differences in the Suan rou produced by the three technologies. However, there are few studies on the quality differences of these three main Suan rou formulations. For this reason, this study aimed to investigate the microbiological, physicochemical, texture, and sensory characteristics of three different techniques of Suan rou during fermentation.
By comparing the microbiological, physicochemical, texture, and sensory characteristics of different formulations of Suan rou, better industrial production formulations can be obtained.

| Sample Preparation
Three different techniques (A, B, and C) were used in this experiment for the production of low-salt Suan rou in western Hunan, China. Technique A was performed as follows. Fresh streaky pork (3-5 cm-width and 3-4 cm-thick) was sliced. For every 10 kg streaky pork, 400 g salt, 10 g Chinese prickly ash, 10 g Illicium verum, 10 g Chinese cinnamon, 100 g dry chili powder, and 400 g ginger slice were added. The materials were mixed, refrigerated, and cured for 24 hr. Then, 75% cured streaky pork (2.6 kg corn flour, 100 g salt, 2.5 g Chinese prickly ash, 2.5 g Illicium verum, and 2.5 g Chinese cinnamon) and stir-fried 25% corn flour (COFCO, Shanghai, China) were evenly mixed and then placed in a jar or glass bottle. Cover the cap. After the edge of the jar was sealed with water, the bottleneck was plugged with straws or corn bran.
Suan rou can be formed after 30 days of natural fermentation.
Technique B was conducted as follows. A total of 200 g saccharose were added to the sliced meat based on technique A. Technique C was applied as follows. The amounts of salt in the sliced meat were changed to 200 g based on Technique A. During fermentation, samples from the three different low-salt Suan rou batches were collected every five days for relevant microbiological and physicochemical analysis.

| Microbiological analysis
Approximately 25 g Suan rou sample was obtained from each group, ground, homogenized (UltraTurrax homogenizer, IKA Labortechnik, Selangor, Malaysia), and then added into 225 ml sterile NaCl water (0.9%). After step-by-step gradient dilution, appropriate decimal dilutions of the three sample solutions were selected. Exactly 0.1 ml solution was then obtained from each sample and poured into different selective media for cultivation to quantify different microorganisms. Three parallel tests were conducted for each sample solution.
Total viable counts were grown aerobically for 3 days at 30°C by using PCA (Oxoid, Basingstoke, England). LAB were cultivated for 2-3 days at 30°C in MRS agar. Staphylococcus were cultivated for 2 days at 37°C in MSA. Yeast was cultivated for 3-4 days at 25°C in PDA, and Enterobacteriaceae were cultivated for 1 day at 37°C inVRBDA. Bacillus spp. were cultured in DTA at 35°C for 2 days and counted, whereas Bacillus spore formers were determined by using 10-fold dilutions of Bacillus spp. incubated in DTA for 2 days at 35°C after pasteurization for 10 min at 80°C. The Pseudomonas incubated in Pseudomonas-Aeromonas selective agar at 26°C for 3 days were counted. Enterococci incubated in KAA at 42°C for 1 day were determined.
The dominant bacteria were isolated from the highest dilution gradient of Suan rou samples at different fermentation stages and then purified in the aforementioned various media. The DNAs of LAB and Staphylococcus were extracted using the method of Benito (Benito et al., 2008), and the specific types of bacteria were analyzed by 16S rRNA gene sequencing. Yeast DNA was extracted using the method of (Kurtzman & Robnett, 1998), and the types of bacteria were analyzed by 26 rRNA gene sequencing.

| Proximate chemical analyses
Protein content was determined in accordance with the Kjeldahl method (AOAC, 1997). A 6.25 factor was used for the conversion of nitrogen content to crude protein. Sarcoplasmic and myofibrillar proteins were extracted according to the methods of Mauriello et al. (2002), and the protein concentrations were measured as described by Lowry et al. (1951).
Crude fat content was determined by Soxhlet method as described in AOAC (1997). Cholesterol levels were determined by enzyme catalytic spectrophotometry according to the manufacturer's instructions. The levels of FFAs were assayed using the method of ACS-ACOD as described by the manufacturer's instructions (Beijing nine Strong Biotechnologies, Beijing, China).
Aw was measured using a digital aw meter (Rotronic Hygroskop DT, Zurich, Switzerland) after stabilization at 25°C for 30 min. Salt content was analyzed using an AOAC procedure (1997).
The pH was determined according to the method of Wang (2000). Approximately 10 g sample was weighed and added with 90 ml deionized water. The pH was determined using a digital pH meter (Mettler Toledo 320-s, Shanghai, China) after the sample was homogenized at 8,000 g for 1 min.
Total acidity (TA) was analyzed in accordance with the AOAC guidelines (1997). Exactly 5 g sample was weighed and added with a small amount of distilled water without carbon dioxide. The sample was homogenized at 8,000 g for 1 min and then transferred to a 250 ml volumetric flask. After heating in water bath at 75°C-80°C for 30 min, the sample was cooled until constant volume and filtered (initial liquid was discarded). The obtained filtrate was titrated with 0.1 mol/L standard NaOH. Finally, the TA was expressed by % lactic acid.
Organic acids were determined according to the method of (Riebroy et al., 2004). Approximately 15 g sample was added with 35 ml distilled water, homogenized at 8,000 g for 1 min, and then centrifuged (Sigma Laborzentrifugen, Model 4K15, Osterode, Germany) at 4,500 g for 15 min. 0.9 ml of supernatant was obtained and evenly mixed with 0.1 ml 5% (w/v) internal-standard butyric acid. The mixture was added with 0.5 M perchloric acid for 5 min. The mixture was then centrifuged at room temperature and 12,000 g for 30 min to remove the protein residue. The supernatant was filtered using 0.45 µm membrane (Minisart RC-15, Satorious, Goettingen, Germany) and injected into the high-performance liquid chromatograph (HPLC) (Agilent1100, USA) with an Ecomsil C18 (4.6 mm × 250 mm). The components and contents of different organic acids in the samples were identified on the basis of the peak area and retention time. The conditions of HPLC were as follows: mobile phase: methyl alcohol:water:phosphoric acid = 5:95:0.05 (v/v), gradient elute. Chromatographic conditions were as follows: 30°C column temperature; 0.8 ml/min flow rate; 210 nm photodiode wavelength.
SDS-PAGE was assessed as described by Laemmli (1970) and implemented in a vertical gel electrophoresis unit (Mini-Protean-3 Cell, Bio-Rad, Richmond, CA, USA) by using a 10% separating gel with 4% stacking gel; 10 µl mixture was loaded in each well. Electrophoresis was employed at constant voltage of 100 V with current limit set to 90 mA for 45 min. After electrophoresis, gel staining and destaining were performed as described by Landeta et al. (2013).

| FAA Determination
The FAA content of Suan rou was determined according to the method of (Aristoy & Toldra, 1991). Approximately 5 g Suan rou was precisely weighed, and 25 ml of 5% perchloric acid was added.
The mixture was homogenized for 5 min and transferred to a 25 ml volumetric flask, and 5% perchloric acid was used to reach a constant volume of 25 ml. The solution was allowed to settle for 2 hr at 4°C. The supernatant was filtered by a double-layer filter paper.
Approximately 5 ml filtrate was centrifuged at 10,000 g for 5 min.

| Determination of FA
Determination of FFA was performed according to the methods of (Bligh & Dyer, 1959). Exactly 5 g sample was weighed, and 10 ml concentrated hydrochloric acid and 20 ml methyl alcohol were added and homogenized for 2 min. Approximately 10 ml concentrated hydrochloric acid was then added and sufficiently homogenized for

| Total Volatile Basic N (TVB-N) Determination
The TVB-N content was determined by the microdiffusion method of Conway (Hughes et al., 2002).

| Textual Determination
The samples were cut into slices with the dimension of 10mm × 30 mm, and then the hardness, elasticity, and chewiness of the fermented sour fish were detected by a texture analyzer (TA-XT2i texture analyzer, Stable Micro Systems Company of Britain).
The testing conditions are as follows: probe diameter: 50 mm; compression speed: 5 mm/s; compression deformation: 50%; time slot for second compression: 1 s; surface induction: 99 g; primary force: 30 g. The determination result was processed by Texture Expert (V1.0).

| Sensory analysis
Because the taste of products of different processes is quite different, it is necessary to pick from three different processes of Suan rou to meet consumer preferences. The differences among the Suan rou made by traditionally natural fermentation were compared by preference analysis. Since we are a food college, the teachers and students have a certain tasting ability. So a total of 50 teachers and postgraduates in food professional in the laboratory were selected to conduct sensory evaluation of Suan rou. The fermented Suan rou was sliced into 5 mm thickness, placed in the tasting plates, and then numbered randomly. The appearance, flavor, color, acidity, salinity, texture, and general acceptability of the samples were scored by a 9-point system: 1 for the worst, 5 for moderate, and 9 for the best (Paglarini et al., 2020).

| Statistical analysis
All experiments were performed in three replicates, and the mean values were calculated. Statistical analysis was performed using the SPSS statistical program (Version 22.0) for Windows (SPSS Inc., Chicago, IL). ANOVA was used to determine the differences between sample batches, and the mean values were evaluated using Duncan's multiple range or least-square difference test (p < .05). Person correlation coefficient was used in bilateral correlation analysis. p < .01 indicates highly significant difference, and p < .05 indicates significant difference.   and P. pentosaceus can sufficiently utilize nutrient substances in the meat and were thus dominant during processing and fermentation of Suan rou. The color, flavor, and texture of the product can be improved through the acid and bacteriocin generated during fermentation; the maturity of the product can be accelerated, and the growth and reproduction of other putrefying and pathogenic bacteria can be restrained, thus improving the quality and ensuring the safety of the product (Lorenzo et al., 2012). Yeast can utilize the metabolites of LAB to proliferate quickly in the acidic environment generated by LAB and simultaneously secrete the metabolites required by LAB growth; thus, these organisms can promote the reproduction and growth of each other (Zeng et al., 2013b).

| Microbiological Analysis
The number of Staphylococcus in Suan rou reached 2.22 -2.57 log cfu.g -1 at the initial period of fermentation. However, with extension of fermentation time, the number of Staphylococcus in batches B and C increased to 4.12 ± 0.13and 4.13 ± 0.12 log cfu.g -1 on the 10th day of fermentation, respectively. With extension of fermentation period, the number of Staphylococcus slowly decreased, and the numbers in the sample groups decreased to 1.12 ± 0.04 and 0.97 ± 0.03 log cfu.g -1 on the 30th day of fermentation. In the batch A samples, the number of Staphylococcus increased to 4.37 ± 0.13 log cfu.g -1 on the 10th day of fermentation. With the further extension of fermentation time, the number of Staphylococcus in batch A samples decreased to 1.43 ± 0.05 log cfu.g -1 on the 30th day of fermentation. This result indicates that the growth of Staphylococcus was evidently restrained in the middle and later periods of fermentation. Similar theories can also be obtained in the analysis of bacterial flora distribution and bacterial characteristics of traditional Chinese Suan yu, suggesting that Staphylococcus is notably restrained in acidic conditions (Zeng et al., 2013b). In a research on the dominant Staphylococcus in salami, (Bonomo et al., 2009) The population of bacilli in the three batches of Suan rou amounted to 2.37 -2.47 log cfu/g at the initial fermentation.
Thereafter, the level of bacilli notably decreased to 0 -0.43 log cfu/g during 30 days of fermentation. The results were similar to the those of Zeng et al. (2013a), who reported that the rapid growth of LAB may cause the suppressed growth of bacilli on fermented meat products.
With mass reproduction of LAB and the reduction of product pH during fermentation, the numbers of these bacteria in the samples of batches B and C were below the limit of detection on the 30th day of fermentation, whereas the numbers of Enterobacteriaceae, Enterococci, and Pseudomonades in the batch A samples decreased to 0.67 ± 0.02, 0.53 ± 0.01, and 0.27 ± 0.01 log cfu/g on the 30th day of fermentation, respectively. This result indicates that compared with batch A, the growth of Enterobacteriaceae, Enterococci, and Pseudomonades, which are suitable for the reproduction of LAB, was inhibited in batches B and C. (Hu et al., 2008) also noted that high levels of LAB grown in silver carp surimi can notably inhibit the growth and reproduction of Enterobacteriaceae, Enterococci, and Pseudomonades. The high correlation (r = .903-.991, p < .05) between the pH and the number of Enterobacteriaceae, Enterococci, and Pseudomonades showed that pH is a key factor restraining the growth of other bacteria. (Lücke, 2000) speculated that the rapidly reduced pH and bacteriocin generated by LAB may be the main factors that restrain putrefying and pathogenic bacteria.

| Physicochemical analysis
The nutritional value and flavor of fermented meat products is largely determined by the degradation degree of proteins in the samples. Compared with batch A, the contents of total, myofibrillar, and sarcoplasmic proteins in batches B and C showed higher protein degradation rates, which respectively decreased from 16.1% ± 0.49%,6.49% ± 0.19%, and 3.44% ± 0.12% to 10.1% ± 0.28%, 2.57% ± 0.09%, and 1.31% ± 0.06% in batch B and from 16.2% ± 0.49%, 6.49% ± 0.19%, and 3.43% ± 0.10% to 9.75% ± 0.27%,2.24% ± 0.09%, and 1.27% ± 0.06% in batch C with extended fermentation time (Table 1). A significant positive correlation was observed between the changes in total protein and myofibrillar and sarcoplasmic proteins (r = .983-.997, p < .01; r = .983-.994, p < .01, respectively). These results indicate that the proteolysis and amounts of small molecules increased significantly and showed certain nutritional significance during fermentation. The muscle proteins of Suan rou were severely hydrolyzed. This finding may be caused mainly by the endogenous cathepsin and microbial protease activity, which cause the formation of small peptides, FAAs, and ammonia N and contribute to the formation of aroma and flavor of fermented products due to protein degradation (Molly et al., 1997). Valyasevi and Rolle (2002)   correlation coefficients between lactic acid bacteria and crude fat and cholesterol in technologies C)respectively). Riebroy et al., 2008 reported that the combined action of acidophilous endogenous enzymes and microbial enzymes can cause the formation of FFAs and reduction in the levels of crude fat and cholesterol, thus contributing to the unique flavor and providing a health factor. In this study, we speculated that the reduction in crude fat and cholesterol and the production of FFAs may be the effect of various microorganisms, such as LAB and Staphylococcus, in Suan rou.
Reasonable aw in fermented food cannot only promotes the proliferation of favorable microorganisms and control the growth of unfavorable microbes but also change the textural parameters of products. In our research, the three groups of low-salt fermented Suan rou showed similar aw (0.91 ± 0.04) (data not shown). According to the research of (Listner, 1975), a pH less than 5.2 and an aw less than 0.95 of fermented meat products are suitable for the extensive proliferation of LAB, and the products can be stored for a long term without cold storage. This result indicates that the aw of Suan rou plays an irreplaceable role in promoting the rapid acid production of the products, improving the safety and texture of the products, and restraining the growth of infectious bacteria. Figure 2 shows the pH and TA of Suan rou during fermentation.
The figure shows that compared with the batch A sample, the samples in batches B and C have shown a faster pH reduction speed and a higher TA generation amount (p < .05). The pH reduced from 6.89 ± 0.06 to 4.34 ± 0.04-3.68 ± 0.03 in batches B and C at the end of fermentation, and the TA values increased from 0.03 ± 0.00 g/ kg to 2.27 ± 0.05 and 2.73 ± 0. 5 g/kg, respectively. The Suan rou in batch A showed a slow acid production speed. The pH of batch A reduced to 4.45 ± 0.06, whereas the TA increased to 2.11 ± 0. 5g/kg at the end of fermentation. In the samples, pH exhibited a high negative correlation with TA (r = -.985 to -.995, p < .01). The rapid generation of LAB in the fermented meat products is strongly related to the reduction of pH value and rise of TA, whereas the fast proliferation of LAB is closely related to nutrient substances, such as glucose and saccharose, added to the product. These saccharides can provide energy for the rapid propagation of LAB. Saccharose was added to the Suan rou of batch B during processing, thus notably promoting the rapid generation of LAB. The 2% salt concentration can promote the rapid proliferation of LAB. Normally, LAB can grow under 4%-5% salt, but the growth rate will be partially restrained.
The salt content added in the Suan rou of batch A during the processing period was between 4% and 5%, which resulted in the low reduction rate of pH and the low generation rate of TA. According to the research of (Riebroy et al., 2004), organic acids such as lactic acid that accumulated in fermented meat products result in low pH and high TA of the products and are an essential factor controlling the growth of other putrefying and pathogenic bacteria, implying the high safety of the products. Furthermore, with the reduction of pH and accumulation of organic acids, the product texture will also be influenced, resulting in the increased hardness and enhanced cohesive force and elasticity.
Organic acid content is also an important factor guaranteeing product safety. In raw meat, lactic and acetic acids are both below the limits of detection. Lactic acids increased to 1.76% ± 0.01%, 2.21% ± 0. 01%, and 2.67% ± 0. 01% in batches A, B, and C at the end of fermentation, respectively (data are not shown). Compared with batch A, batch B, which was added with saccharose, and batch C, which was added with 2% salt, presented higher lactic acid generation amounts (p < .05). Meanwhile, 0.11% -0.12% acetic acid (p > .05) was detected in the three different low-salt fermented Suan rou. This finding indicates that lactic acid is the main organic acid generated in fermentation, whereas a small amount of acetic acid is also generated. Meanwhile, oxalic, succinic, and butyric acids were below the limits of detection. These research results are similar to those of other reports on organic acids in fermented meat products.
For example, the lactic acid generated by the reproduction and metabolism of LAB becomes the main organic acid in the fermented meat products of Som-fug and Salchichón (Riebroy et al., 2004). (Ostergaard et al., 1998) (Astiasarán et al., 1990). Therefore, in our study, the differences in microorganism distribution, acidity, and endogenous enzyme activity caused by the different amounts of sodium chloride and sucrose in the samples may cause the varying degrees of hydrolysis of the Suan rou protein in the different groups, whereas intense partial bands were observed due to the possible comigration of hydrolysis from large proteins.

| FAA analysis
FAA can improve product flavor. Thus, the release of FAA during the fermentation of meat product must be investigated. Table 2 lists the FAA change conditions of the three different groups of low-salt fermented Suan rou. The table shows that compared with the fresh streaky pork before fermentation (BF), the species and quantity of FAAs in the three groups of fermented Suan rou have evidently increased. Batches B and C showed a faster generation rate (p < .05) of FAAs than batch A. However, no evident distinction was observed between the type and amount of the final products in the three groups of Suan rou (p > .05). In the final products of fermented Suan rou, high amounts of essential amino acids (EAAs) for the human body, such as lysine, leucine, and glycine, were detected.
The total amount of FAAs increased from 163 ± 3.61 mg/100 g to 199 ± 5.34, 213 ± 5.79, and 209 ± 5.62 mg/100 g. The evident growth rates of glycine, glutamic acid, aspartic acid, valine, leucine, lysine, and arginine were noted. In the determined FAAs, the content of glutamic acid (35.6 ± 0.27-39.4 ± 0.27 mg/100 g) was the highest at the last stage of fermentation. Compared with the results in existing reports on fermented meat products, the species and quantity of FAAs in Suan rou were notably higher than those in other fermented meat products. According to the research of (Mukherjee et al., 2014), the generation of FAAs in fermented meat products results from the fermentation and maturity process; cathepsin shows Note: Batch A: Suan rou was prepared using Technique A; Batch B: Suan rou was prepared using Technique B; Batch C: Suan rou was prepared using Technique C. a,b,c Values with unlike superscript letters in the same row are significantly different (p < .05) LSD test.
TA B L E 2 Changes in free amino acids of Suan rou during fermentation high activity at pH between 5.0 and 5.5; thus, the proteins are degraded, and the protease of microorganisms (LAB, staphylococci, and yeast) further accelerates the degradation of muscle proteins during fermentation (Qiu et al., 2012). We speculate that the rapid

| FA analysis
Fatty acid degradation plays an important role in the formation of flavor of fermented meat products in the fermentation and ripening process of meat products. In most cases, the release of FFAs occurs as a series of biochemical reactions under the combined action of enzymes of meat products or microorganic lipases, which will form the unique flavor substance of the fermented product (Casaburi et al., 2007).

TA B L E 3
Changes in fatty acids of Suan rou during fermentation palmitoleic, linoleic, and linolenic acids. The content percentage of polyunsaturated fatty acids considerably increased, especially the ω-3 and ω-6 unsaturated fatty acids, which are functional in the human body. Similar conclusions can be found in the work of (Riebroy et al., 2008) on FFAs of Som-fug; the content of saturated fatty acids in Som-fug during fermentation notably decreased, and high amounts of polyunsaturated fatty acids, such as linoleic, linolenic, docosahexaenoic, and eicosapentaenoic acid, were formed.
Verplaetse considered that the increased FFAs may be the result of the combined action of acidophilous endogenous enzymes and microbial enzymes in meat products. neutralization. The TVB-N value was within a controlled range; thus, the freshness of Suan rou can be maintained to a great extent. (Hu et al., 2008) reported that in the fermented silver carp surimi, the group inoculated with a leavening agent can restrain the accumulation of TVB-N. In this research, the TVB-N values for all fermented Suan rou have not exceeded the standard formulated by European Union, that is, the TVB-N value should not be higher than 35 mg TVB-N 100 g -1 in fish. This finding indicates that the three types of low-salt fermented Suan rou all exhibited low TVB-N values.

| Fat oxidation analysis
Thiobarbituric acid reactive substance (TBARS) value has been widely used as an index to predict the fat oxidation degree of meat and meat products. According to previous research, the maximum TBARS value is 5 mg malondialdehyde (MDA)/kg to ensure the high quality of meat (Sikorski et al., 1990). In our research, the initial TBARS for each Suan  With the decreased pH and Aw, the hardness and chewiness of the three different types of low-salt fermented Suan rou evidently strengthened (p < .05) compared with those BF. In batches B and C, which presented faster lactic acid generation rate and pH decrease rate, the hardness and chewiness of the products exhibited more evident increase than those in batch A during the ripening fermentation.

| Textual analysis
No significant difference was detected in the hardness and chewiness of the final products of the three groups of Suan rou (p > .05). In the fermentation of meat products, the decrease in pH close to the isoelectric point of proteins causes the partial dehydration of meat products, induces the gathering of myofibrillar proteins in fermented meat products, and results in acid-induced gel formation (Xu et al., 2012). We speculate that the coordinating influence of these factors has resulted in good hardness and chewiness of the final products of fermentation. (Riebroy et al., 2008) also shared similar research on Som-fug and indicated the notably increased hardness and chewiness of Som-fug.
Cohesiveness can embody the strength of intermolecular forces in fermented meat products. The research of (de Huidobro et al., 2005) indicates that for fermented meat products, a pH below 5.0 implies a classic negative correlation with cohesiveness. The present research has demonstrated that in low-acid-fermented meat products, relevant products showed relatively low cohesive forces given that the moderate-strength acid environment reduced the isoelectric point of proteins. Suan rou is a type of acid-fermented meat product with a pH lower than 4.5. In the fermentation of Suan rou, the pH during part of the fermentation time periods is close to the isoelectric point of protein, resulting in the denaturing of protein gel and therefore the strengthening of intermolecular cohesive force. On the 5th-10th day of fermentation of Suan rou in batches B and C and the 15th-20th day of fermentation in batch A, a high cohesiveness was observed (data are not shown), whereas in the initial and last stages of fermentation, the pH was evidently higher or lower than 5.0. Thus, the cohesive forces of Suan rou in the earlier and later stages of fermentation are relatively low.  A and B is superior to that of batch C in terms of appearance, flavor, taste, mouth feel, and overall acceptability (p < .05). Although the Suan rou of batch C exhibited a faster fermentation speed, its taste was more acidic than those of batches A and B. The coordination between the acidity and salinity of Suan rou is the main factor influencing the acceptance of consumers for this traditional product. (Ostergaard et al., 1998) indicated that consumers would feature a high recognition degree toward low-salt, acid-fermented meat products only when the lactic acid concentration reaches at least 2%, the salt concentration range is between 2.5% and 4.5%, and the pH is lower than 4.5.

| Sensory analysis
Acids were rapidly generated in batch C, and the salt content was 2.13% ± 0.04%. The coordination between the acidity and salinity was reasonable in batches B and A, thus influencing the sensory quality of the product. The hydrolyzation of muscle proteins and fats can also improve the flavor of fermented products. The sensory analysis on fermented silver carp surimi conducted by (Hu et al., 2008) also indicated that high-concentrated small molecular peptides, amino acids, aldehydes, and organic acids can notably improve the acceptability of fermented meat products. Consequently, the high acidity generated in the fermentation process, low salt, hydrolyzation of muscle proteins, and the degradation of fats form the unique flavor of Suan rou.

| CON CLUS ION
Suan rou is a traditional fermented meat product. Compared with those in other similar fermented meat products, the LAB (mainly including L. plantarum and P. pentosaceus) and saccharomycetes (mainly including S. cerevisiae and Hansenula polymorpha) in Suan rou rapidly reproduced during the whole fermentation, and substantial amounts of organic acids such as lactic acid were generated. Batch B, which was added with saccharose, and batch C, which contained a low-salt concentration, exhibited a faster pH reduction rate and TA generation rate compared with batch A, therefore forming the unique and soft sour flavor of Suan rou. Compared with other fermented meat products, Suan rou also features better textual parameters, such as hardness and chewiness. Suan rou contains high levels of FFAs and FAAs that provide this product with rich nutrition and flavor. The sensory analysis of Suan rou indicated that 2% salt addition may result in bad flavor due to high acidity, whereas 4% salt addition amount in batch A and 2% saccharose addition in batch B resulted in a unique fermentation flavor favored by consumers. Combined with the overall indicators, the Suan rou produced by B technology is more easily accepted and loved, and can be used in industrial production.