Bacteriocin-Producing Lactic Acid Bacteria Strains with Antimicrobial Activity Screened from Bamei Pig Feces

Lactic acid bacteria (LAB), which are characterized by producing various functional metabolites, including antioxidants, organic acids, and antimicrobial compounds, are widely used in the food industry to improve gut health and prevent the growth of spoilage microorganisms. With the continual incidence of foodborne disease and advocacy of consumers for gut health, LAB have been designated as vital biopreservative agents in recent years. Therefore, LAB with excellent antimicrobial properties and environmental tolerance should be explored further. In this study, we focus on screening the LAB strains from a specialty pig (Bamei pig) feces of the Tibetan plateau region and determine their antimicrobial properties and environmental tolerance to evaluate their potential probiotic values. A total of 116 LAB strains were isolated, from which the LAB strain Qinghai (QP)28-1 was identified as Lactiplantibacillus (L.) plantarum subsp. plantarum using 16S rDNA sequencing and recA amplification, showing the best growth capacity, acid production capacities, environmental tolerance, hydrophobicity, antibiotic susceptibility, and bacteriocin production capacity. Furthermore, this strain inhibited the growth of multiple pathogens by producing organic acids and bacteriocin. These bacteriocin-encoding genes were identified using PCR amplification, including plnS, plnN, and plnW. In conclusion, bacteriocin-producing L. plantarum subsp. plantarum QP28-1 stands out among these 116 LAB strains, and was considered to be a promising strain used for LAB-related food fermentation. Moreover, this study provides a convenient, comprehensive, and shareable profile for screening of superior functional and bacteriocin-producing LAB strains, which can be used in the food industry.


Introduction
Probiotics are defined as "live microorganisms that, when administered in adequate amounts, confer a health benefit on the host." [1][2][3]. As one of the mainly important probiotics, lactic acid bacteria (LAB) have been widely adopted in the food fermentation industry to improve the flavor and taste of products, including cheese, beer, and yogurt. They can also colonize the intestinal tract, promote the digestion and absorption of nutrients, and maintain the stability of the intestinal flora [4]. In recent years, LAB have been shown to produce antimicrobial substances, including organic acids, hydrogen peroxide, reuterin, acetoin, diacetyl, antifungal peptides, and bacteriocin [5][6][7][8], and LAB with broad-spectrum antibacterial properties are widely used for inhibiting spoilage bacteria and foodborne pathogenic bacteria and prolonging food shelf lives [9]. activity, and gas production from glucose, as reported by Zhang et al. [17]. Only milky white, raised, catalase-negative, and Gram-positive isolates were selected for further assays.

Screening of the Strains with Broad-Spectrum Antibacterial Activity
In order to screen for LAB with antibacterial effects on a variety of pathogens, the antimicrobial activity of all the isolates from Bamei pig feces isolates was investigated by the Oxford cup double-layer plate method outlined by Muhammad et al. [5], using the pathogens Escherichia coli ATCC 30105; Micrococcus luteus ATCC 4698; Staphylococcus aureus ATCC 29213; Pseudomonas aeruginosa ATCC 27853; Listeria monocytogenes BAA; Bacillus subtilis ATCC 6633; Salmonella enterica subsp. enterica serovar Enteritidis ATCC 13076; and Salmonella enterica subsp. enterica serovar Typhimurium ATCC 43971 as indicator strains. Firstly, 20 mL of NA medium (NA, QingDao Hopebio Technology Co., Ltd., Qingdao, China) was placed in the plates as a lower layer of medium. Then, the 8 indicator bacteria that were cultured overnight in advance were injected into 5 mL of NA agar and cooled down naturally to about 50 • C with a 3% inoculum rate, mixed thoroughly, and immediately added to the plate to solidify them. Next, the sterilized Oxford cups were positioned on the surface of the NA agar and pressed gently. Finally, 200 µL of cell-free supernatant (CFS) of the isolated LAB was poured into the cups. After incubation at 37 • C for 24 h, the diameter of the inhibition zone was measured; this was repeated three times.

Physiological and Biochemical Characteristics and Tolerance
All the isolates were subjected to physiological and biochemical experiments to evaluate the multiple characteristics of the tested LAB strains, containing assays of the growth curve, acid production capacity, bile salt tolerance, and tolerance to low pH and extreme temperatures. In order to obtain LAB with excellent growth capacity and high resistance to the environment, a number of tolerance tests were performed on the strains obtained from the initial screening. The experimental methods and gradient settings for salt tolerance, extreme temperature tolerance, and acid-base tolerance of LAB strains are reported elsewhere by Zhang et al. [17]. To assess the alkali tolerance, LAB were inoculated with 1% inoculum in MRS liquid with different NaCl concentrations (3.5% and 6.5%) at 37 • C for 2 days; the growth of bacteria was observed or the absorbance at 600 nm was measured and compared with the growth of LAB in MRS liquid without NaCl as the control. Similarly, the temperature tolerance test was set at gradients of 5 • C, 10 • C, 30 • C, 45 • C, and 50 • C, and the acid tolerance test was established with pH gradients of 3, 3.5, 4, 4.5, 5, 8, 9, and 10. Antibiotic susceptibility assays were conducted with the disk diffusion method, acquired from Anisimova and Yarullina [21], with minor modifications. In brief, 10 mL of MRS agar was poured into sterile plates and solidified. Then, 5 mL of the LAB suspension was mixed with MRS agar (50 • C) and poured into the upper layer of the plate, which then solidified. Next, 10 antibiotic discs (Oxoid Ltd., Basingstoke, U.K.), including ampicillin (AMP, 10 µg/disk), erythromycin (E, 15 µg/disk), colistin sulphate (CT, 10 µg/disk), penicillin (P, 10 µg/disk), ciprofloxacin (CIP, 5 µg/disk), gentamicin (CN, 10 µg/disk), chloramphenicol (C, 30 µg/disk), vancomycin (VA, 30 µg/disk), rifampicin (RD, 5 µg/disk), and tetracycline (TE, 30 µg/disk), were added to the surface of the plates. Finally, the diameters of the inhibition halos were measured after 24 h of incubation at 37 • C and interpreted as susceptible (S), moderately susceptible (MS), or resistant (R), based on the breakpoint values specified by the Clinical and Laboratory Standards Institute (CLSI, 2015).
The resistance properties under the simulated gastrointestinal tract (GIT) were assessed following the method of Zhang et al. [18]. For the simulated gastric juice, 0.35 g of pepsin was diluted in 0.2% saline and the pH was adjusted to 2.5 with 1 M HCL. For the simulated intestinal fluid, 0.1 g of trypsin, 1.8 g of bovine bile salt, 1.1 g NaHCO 3 , and 0.2 g NaCL were added to 100 mL sterile water, and the pH was adjusted to 8.0 with 1 M NaOH. The LAB cultured overnight were inoculated into the simulated gastric juice at 5% inoculum, vortex shaken for 30 s, and incubated at 37 • C for 3 h. After 3 h incubation at 37 • C, 100 µL of the 3 h solution was added to 900 µL of simulated intestinal fluid, which was then incubated at 37 • C for 30 s with vortex shaking. The LAB were incubated for 4 h at 37 • C. The viable counts of LAB after 0 h, 3 h, and 7 h were measured by the dilution spread method.

Cell Surface Hydrophobicity and the Auto-Aggregation Assay
The cell surface hydrophobicity and auto-aggregation ability of the screened LAB strains were investigated following the approach outlined by Sirichokchatchawan et al. [22]. Specifically, LAB strains cultured overnight in MRS broth were centrifuged, flushed three times with phosphate-buffered saline (PBS), and then resuspended in PBS until the OD was adjusted to 0.6 at 600 nm (A0). Subsequently, 3 mL of LAB cell suspension was added to 1 mL of xylene and allowed to stand for 40 min at 37 • C. The absorbance (At) of the aqueous phase at 600 nm was measured. The cell surface hydrophobicity (%) was assessed according to the following formula: hydrophobicity = (1 − At/A0) × 100%. The auto-aggregation was estimated as reported by De Souza et al. [23]: 1 mL of LAB suspension was allowed to stand for 2 h at 37 • C, and the absorbance (A1) of its upper liquid layer at 600 nm was determined. Auto-aggregation (%) = 1 − A1/A0 × 100%.

Screening of the LAB Strains Producing Broad-Spectrum Antibacterial Bacteriocin
The antimicrobial activity of the isolates was assessed by the agar diffusion assay [5], with Escherichia coli ATCC 30,105, Micrococcus luteus ATCC 4698, Staphylococcus aureus ATCC 29,213, Pseudomonas aeruginosa ATCC 27,853, Listeria monocytogenes BAA, Bacillus subtilis ATCC 6633, Salmonella enterica subsp. enterica serovar Enteritidis ATCC 13,076, and Salmonella enterica subsp. enterica serovar Typhimurium ATCC 43,971 as the indicator strains. In order to further explore the mechanism of antibacterial activity and to find out which substances are involved in the process of pathogen inhibition by LAB strains, a series of experiments were implemented on the effect of pH, catalase, and protease on the antibacterial ability. The CFS of LAB was adjusted to different pH values using 1 M HCl. The inhibitory ability of the CFS with different pH values against pathogens was determined by the above method. To inhibit the interference of hydrogen peroxide, 3% catalase was dissolved in PBS and added to the CFS at a concentration of 5.0 mg/mL of catalase, and the antibacterial activity of the CFS was detected. The untreated CFS was used as a control. Similarly, proteinase K, pepsin, and trypsin were added to the CFS at a final concentration of 1.0 mg/mL, noting the suitable temperature and pH of the different enzymes. The effect of proteinase K, pepsin, and trypsin on the antibacterial activity of the LAB strains was examined using the agar well diffusion method [24].

Amplification of Bacteriocin-Producing Genes of LAB
Bacteriocin-related gene PCR amplification experiments were performed on the strains, which exhibited antibacterial activity even after acid and hydrogen peroxide exclusion experiments, and were sensitive to protease. The involved genes encoding bacteriocin synthesis found from L. plantarum C11, V90, J23, J51, and WCFS1, and NC8 was identified from the NCBI database. The specific primers were designed (Table 1) for PCR amplification validation and the PCR amplification procedure was referenced in the report by Doulgeraki et al. [25]. The PCR products were detected by 1.5% agarose nucleic acid electrophoresis. The bands matching the expected fragment size were recovered by gum cutting and sent to the Huada Biotech Company Co., Ltd. (Zhengzhou, China); the results were compared with the sequences of the genes encoding these bacteriocin-related genes in the NCBI database.

Statistical Analysis
Principal component analysis (PCA) was performed in Excel combined with the XLSTAT program. Heat map analysis was carried out in ClustVis (https://biit.cs.ut.ee/ clustvis, accessed on 17 September 2021). Statistical analysis was conducted using the SPSS 13.0 software (SPSS, Inc., Chicago, IL, USA). One-way ANOVA tests were applied to analyze the significance of differences (p < 0.05).

Screening of the Broad-Spectrum Antibacterial Active LAB
A total of 116 presumptive LAB isolates was obtained from the fecal samples of 44 Bamei pigs from the original breeding farm of Bamei pigs in Huzhu County, Qinghai Province, China. Eight LAB strains stood out from all the isolates because of their excellent antibacterial activity, which inhibited at least three pathogens (Supplementary Materials Figure S2). By combining 16S rRNA gene sequence analysis with recA gene multiple detection, strains QP3-2, QP19-1, and QP28-1 were identified as L. plantarum subsp. plantarum, and QP4-2 and QP33-2 could be assigned to L. casei ( Figure S2). In the subsequent screening process, Lactococcus lactis QP20-2 and Enterococcus hirae QP22-2 were omitted due to their weak ability to survive in high-salt, high-temperature, or low-pH environments and their low acid-production capacity (Table S3 and Figure S3). L. casei QP4-2 and Enterococcus hirae QP23-1 were also excluded because of their lower survival in GIT, presented in Figure S3C.

Preliminary Validation of the Antimicrobial Substance Produced by LAB
In order to investigate the mechanism of bacterial inhibition of these LAB and whether the acid or the bacteriocin caused the inhibition, some experiments were performed, including assessing the influence of pH, protease, and hydrogen peroxide on the antimicrobial activity. As shown in Figure 1, the antibacterial potential of LAB against all pathogens decreases with elevated pH values, which suggests that the impact of pH on the antimicrobial activity is evident. When the pH of the CFS was 5.0, all four strains showed broad-spectrum inhibition against six of the pathogenic bacteria, with QP28-1 reaching a maximum inhibition circle size of 23.24 mm against M. luteus ATCC 4698. Compared with QP3-2 and 19-1, LAB strains QP28-1 and 33-2 still displayed an inhibition of Staph. aeruginosa ATCC 29213. However, when the pH increased to 5.5, the inhibitory ability of these strains showed a significant decrease; notably, the inhibition of L. monocytogenes BAA completely disappeared, and the original inhibition of Staph. aeruginosa ATCC 29213 by QP28-1 and QP33-2 also disappeared. All four strains still retained a high inhibitory activity against E. coli ATCC 30105, Ps. aeruginosa ATCC 27853, and B. subtilis ATCC 6633, and did not lose performance with the increase in pH, which suggested that these LAB strains of QP3-2, QP28-1, and QP33-2 were good candidates for antimicrobial activity.

Selection of Potentially Probiotic Isolates
In order to assess the similarity and variability between the LAB strains in various probiotic phenotypes, the probiotic characteristics, including bile salt tolerance, gastrointestinal tract (GIT), hydrophobicity, auto-aggregation, and antimicrobial activity of the Considering that LAB may produce hydrogen peroxide (H 2 O 2 ) to inhibit pathogenic bacteria, the bacteriostatic ability of CFS treated by catalase was examined, as shown in Table 2. The inhibitory ability of most of the CFS was maintained after catalase treatment, although it was also noted that the inhibitory ability of four LAB strains against Staph. aeruginosa ATCC 29213 and Salmonella enterica subsp. enterica serovar Enteritidis ATCC 13076 disappeared. Nevertheless, when the CFSs of these four strains were treated with proteinase K, pepsinum, and tryptase, almost all of the inhibitory effects completely disappeared, which implied that the antibacterial substance was sensitive to proteases; thus, it was tentatively determined to be a protein.

Selection of Potentially Probiotic Isolates
In order to assess the similarity and variability between the LAB strains in various probiotic phenotypes, the probiotic characteristics, including bile salt tolerance, gastrointestinal tract (GIT), hydrophobicity, auto-aggregation, and antimicrobial activity of the isolated LAB strains were analyzed through multivariate analysis based on PCA. As illustrated in Figure 2, the first (PC1) and the second (PC2) principal components represented 81.575% and 11.099% of all the variables (bile salt tolerance, GIT, hydrophobicity, auto-aggregation, and antimicrobial activity), indicating significant differences in the probiotic characteristics of the LAB strains. The eight LAB strains were divided in three areas based on PCA. Five of the LAB strains containing QP3-2, QP4-2, QP19-1, QP28-1, and QP33-2 were clustered close to each other, which showed similar probiotic characteristics and exhibited good tolerance and antimicrobial activity. In contrast, QP20-2, QP23-1, and QP22-2 were listed separately, which displayed weak tolerance and inhibition to pathogenic bacteria in the above study results. Furthermore, a heat map was generated to cluster probiotic strains based on their phenotype (bile salt tolerance, GIT, hydrophobicity, auto-aggregation, and antimicrobial activity). As shown in Figure 3, the probiotic phenotype heat map clustered eight LAB strains into two clusters and four sub clusters, A, B, C, and D. Overall, the probiotic properties of cluster A and B and QP33-2 from cluster D were better than the other clusters. The strain QP4-2 from cluster B was excluded because of its low GIT value, which means that it is difficult to survive and colonize the gastrointestinal tract. The strains QP23-1, QP20-2, and QP22-2 were not selected because they displayed lower values for bile salt tolerance, hydrophobicity, auto-aggregation, and antimicrobial activity. The LAB strains QP19-1, QP28-1, QP3-2, and QP33-2 were selected as potentially probiotic isolates because they exhibited excellent probiotic phenotypes, especially in antimicrobial activity.   biotic strains based on their phenotype (bile salt tolerance, GIT, hydrophobicity, auto-aggregation, and antimicrobial activity). As shown in Figure 3, the probiotic phenotype heat map clustered eight LAB strains into two clusters and four sub clusters, A, B, C, and D. Overall, the probiotic properties of cluster A and B and QP33-2 from cluster D were better than the other clusters. The strain QP4-2 from cluster B was excluded because of its low GIT value, which means that it is difficult to survive and colonize the gastrointestinal tract. The strains QP23-1, QP20-2, and QP22-2 were not selected because they displayed lower values for bile salt tolerance, hydrophobicity, auto-aggregation, and antimicrobial activity. The LAB strains QP19-1, QP28-1, QP3-2, and QP33-2 were selected as potentially probiotic isolates because they exhibited excellent probiotic phenotypes, especially in antimicrobial activity.

The Discovery of Bacteriocin-Producing LAB
The analysis of the above results speculated that genes that can encode bacteriocin may be present in these LAB strains; therefore, some of the most common genes in the LAB that encoded bacteriocin were screened and tested. Using the total DNA of the 4 screened strains as templates, PCR amplification was performed using 13 pairs of specific primers. Remarkably, strain QP28-1 was found to possess three genes that could encode bacteriocin. As shown in Figure 4, plnS, plnN, and plnW from QP28-1 all amplified a single band and matched the expected fragment size, which suggested that the LAB strain of QP28-1 could encode these corresponding bacteriocins.

The Discovery of Bacteriocin-Producing LAB
The analysis of the above results speculated that genes that can encode bacteriocin may be present in these LAB strains; therefore, some of the most common genes in the LAB that encoded bacteriocin were screened and tested. Using the total DNA of the 4 screened strains as templates, PCR amplification was performed using 13 pairs of specific primers. Remarkably, strain QP28-1 was found to possess three genes that could encode bacteriocin. As shown in Figure 4, plnS, plnN, and plnW from QP28-1 all amplified a single band and matched the expected fragment size, which suggested that the LAB strain of QP28-1 could encode these corresponding bacteriocins.

The Discovery of Bacteriocin-Producing LAB
The analysis of the above results speculated that genes that can encode bacteriocin may be present in these LAB strains; therefore, some of the most common genes in the LAB that encoded bacteriocin were screened and tested. Using the total DNA of the 4 screened strains as templates, PCR amplification was performed using 13 pairs of specific primers. Remarkably, strain QP28-1 was found to possess three genes that could encode bacteriocin. As shown in Figure 4, plnS, plnN, and plnW from QP28-1 all amplified a single band and matched the expected fragment size, which suggested that the LAB strain of QP28-1 could encode these corresponding bacteriocins.

Discussion
This study focused on screening for LAB superior performance; how to screen for the desired LAB based on the assessments of desired probiotic attributes is a rigorous scientific process. According to the criteria given by the FAO/WHO [1][2][3], the screening of probiotic bacteria should meet the following three basic characteristics: a tolerance of the selective environment of gastrointestinal mucosa; an adherence to the intestinal cells of the host; and the secretion or production of antibacterial substances as breakdown products [26]. The possible mechanism of this probiotic effect is that LAB secrete and produce antimicrobial substances, such as lactic acid, acetic acid, and bacteriocin, which prevent the proliferation of coliforms and other pathogens [27]. On the one hand, LAB strains produce organic acid and lower the pH of the environment through fermentation, making it difficult for pathogenic bacteria to tolerate acidic environments and survive [28]. For example, Cervantes-Elizarrarás et al. found that the organic acid produced by LAB can suppress some Gram-negative pathogens by permeating the cell membrane, thus acidifying the cytoplasm, interrupting its function, and destroying acid-sensitive enzymes [29]. On the other hand, some LAB strains produce bacteriocin, which is a kind of polypeptide substance used to block the growth of some foodborne pathogens and deleterious bacteria, such as Escherichia coli, Staphylococcus, and Salmonella [30]. In our study, the LAB strains isolated and screened from the intestines of local breeding Bamei pigs from the alpine and high-altitude Qinghai-Tibetan Plateau were analyzed for antimicrobial activity and spectrum, and the bacteriocin produced by LAB QP28-1 was detected and identified. The results show that the eight strains of LAB with the best inhibition effects on a variety of pathogens include M. luteus ATCC 4698, P. aeruginosa ATCC 27853, and B. subtilis ATCC 6633, and also have a significant inhibition effect on pathogens E. coli ATCC 30105, S. aureus ATCC29213, L. monocytogenes BAA, and Salmonella enterica subsp. enterica serovar Typhimurium ATCC 43971, indicating their promising application prospects. In previous studies on screening LAB, most of them used one or two pathogenic strains as indicator bacteria, especially Salmonella and E. coli, which are prone to cause diarrhea [31]. In comparison, we used eight kinds of pathogenic bacteria as indicator bacteria and screened LAB with a broad-spectrum antibacterial activity for multiple pathogenic bacteria, which improved the application value. In terms of the isolation source, we isolated the LAB from the intestine of pigs, which means that our strain possessed higher tolerance to acid, bile salts, and the gastrointestinal environment; therefore, it exhibited a higher chance of survival and colonization in potential agricultural production, as well as in the development of feed additive applications. In the process of screening LAB with good bacterial inhibition, it is important to consider not only the level of unilateral inhibition performance, but also some other properties, including high growth capacity, tolerance, colonization capacity, and safety. A higher strain multiplication capacity ensures that more strains grow and multiply, improving the overall antibacterial inhibition effect. The low survival rate in GIT ( Figure S3) and bile salt environments (Table S4) implied that strains QP4-2 and QP23-1 have difficulty surviving in the host, which runs counter to the probiotic definition. Cell surface properties, such as hydrophobicity and aggregation, are essential indicators of probiotic adherence to human intestinal epithelial cells [32,33]; previous studies illustrated that the auto-aggregation of cells increases with the colonization ability [34]. The screened strains were also tested and L. casei QP33-2 exhibited significantly higher hydrophobicity and self-aggregation than other strains ( Figure S4). Furthermore, the principal component analysis (PCA) presented in Figure 2 suggests that there are differences in the probiotic attributes of the screened LAB. However, some of the LAB strains virtually displayed a similarity in the bile salt tolerance, gastrointestinal tract (GIT), hydrophobicity, autoaggregation, and antimicrobial phenotypes of probiotics. Previously, researchers also used PCA to screen for probiotic candidate, investigating the similarities and differences in the probiotic properties of obtained strains [35]. The heat map in Figure 3 clustered the LAB strains based on their phenotypic properties, and all LAB strains were classified into four subtypes. L. plantarum subsp. plantarum QP19-1 and QP28-1 from cluster A were screened due to their significantly better performance than the other strains. The utility of PCA and heat-map analysis in the screening of probiotic isolates with different phenotypes was demonstrated in a previous study [36], which provided sufficient tools in the process of seeking candidate probiotic strains. The hemolytic activity and common virulence, including the factor genes and biogenic amine genes of the eight screened LAB strains, were tested, and no hemolytic activity or virulence gene was found, thus proving that these LAB strains are safe.
In addition to the acid and hydrogen peroxide produced by metabolism of LAB, bacteriocin has important application prospects as a safe and effective bactericidal substance. The agar diffusion method is the most traditional means of validation for identifying bacteriocin-producing strains, but due to the low yield of bacteriocin, the inhibition effect is not significant after the exclusion of acid and hydrogen peroxide, and the accuracy and reproducibility are poor. In recent years, several methods have been established to screen bacteriocin-producing LAB as food preservatives for the food industry. With the ongoing accumulation of biological information, such as encoding genes, the PCR rapid detection method improved the efficiency and reliability in screening bacteriocinproducing strains [37]. Macwana et al. [38] designed 42 pairs of primers through the reported structural genes of different species of lactic acid bacteriocins, and identified 3 novel bacteriocin-producing LAB, Lactococcus lactis RP1, Pediococcus acidilactici Bac3, and Latilactobacillus sakei JD1, through real-time fluorescence quantitative PCR using the DNA of the target strains as templates and then comparing the sequences [38]. Więckowicz et al. [39] categorized the reported bacteriocin sequences into 5 major classes according to homology and designed multiple pairs of specific primers based on the structural characteristics to identify 40 bacteriocin genes encoding class IIa bacteriocins in the genome from cheese. The PCR amplification method was demonstrated to enable high-throughput screening and analysis in samples producing specific bacteriocin. In this study, the CFS of LAB strain QP28-1 retained its antibacterial activity after the successive removal of acids and hydrogen peroxide, indicating that L. plantarum QP28-1 could produce bacteriocin. The PCR analysis was carried out to identify the bacteriocin genes of QP28-1 by using 13 pairs of primers designed to be specific for the genes related to the biosynthesis of bacteriocin. The structural gene plnN/S/W in the genome of strain QP28-1 was successfully amplified by PCR, and was found to be identical to the structural gene plnN/S/W encoding class IIb bacteriocins in L. plantarum C11 after comparison [40].
Strain QP28-1 was identified as a broad-spectrum, highly efficient bacteriostatic L. plantarum strain using the PCR method, which produced class IIb bipeptide bacteriocin. The rapid identification of bacteriocin-related genes using the PCR method accelerated the screening speed and improved the accuracy. However, the rapid detection method also has some limitations, because the PCR method only draws comparisons with the reported bacteriocin genes, resulting in many novel bacteriocins not being detected; therefore, there is still much room for improvements to the method. As such, this experiment was validated by combining the traditional agar diffusion method with the PCR method for bacteriocin production. Further studies on the synthesis and mechanism of action of the L. plantarum QP28-1 bacteriocin gene and the improvement in bacteriocin production are being carried out in our laboratory. Many bacteriocins were found to have tremendous potential as food pretreatment agents in food spoilage preservation [41]. In practice, bacteriocins used as additive preservatives capable of inhibiting pathogens are often implemented as inoculated probiotic themselves or as purified bacteriocins [42]. It is the antimicrobial activity and non-toxic nature of bacteriocins as potential inhibitors of various pathogens that inspired the interest of researchers in bacteriocins [43]. The results show that L. plantarum QP28-1 exhibits a good spectrum of activity towards multiple pathogens and can produce bacteriocin, indicating that it can be used for preservation and inhibiting microbial food spoilage.

Conclusions
In this study, eight LAB strains were isolated from the feces of Bamei pigs with excellent broad-spectrum antibacterial activity. L. plantarum subsp. plantarum QP3-2, QP19-1, QP28-1, and L. casei QP4-2, QP33-2 showed broad antibacterial activity against the pathogenic bacteria E. coli ATCC 30105, M. luteus ATCC 4698, S. aureus ATCC 29213, and Salmonella enterica subsp. enterica serovar Enteritidis ATCC 13076. Combining the accurate screening processes and assessment methods of PCA and heat-map analysis, four strains of LAB, including L. plantarum subsp. Plantarum QP3-2, QP19-1, QP28-1, and L. casei QP33-2, were selected for the further study of the inhibition mechanism. Moreover, these four LAB strains still exhibited antimicrobial activity after acid exclusion and hydrogen peroxide exclusion, suggesting the presence of bacteriocins. The PCR products of QP28-1 bacteriocinrelated genes revealed that QP28-1 could produce bacteriocin. In conclusion, L. plantarum subsp. plantarum QP28-1 exhibited a high potential application value for preventing microbial food spoilage and agricultural applications, which produce bacteriocin. These screened LAB strains could be developed as functional antimicrobial agents in the future, thus warranting a more detailed and in-depth exploration.

Data Availability Statement:
The datasets presented in this study can be found in online repositories. The 16S rRNA gene sequence of LAB strains QP3-2, QP4-2, QP19-1, QP28-1, QP20-2, QP22-2, QP23-1, and QP33-2 used to support the findings of this study were deposited in the GenBank repository with