In 2015, Lactiplantibacillus plantarum K25 (L. plantarum K25) was obtained from Tibetan Kefir grains and subsequently preserved as frozen stocks at a temperature of −9°C in MRS broth that was supplemented with 20% (v/v) glycerol. The identification of these strains was primarily based on Gram staining, catalase testing, and cellular morphology. The identification of strains was carried out using the API 50 CHL test (bioMerieux, Marcy-l'Étoile, France), and 16S rDNA sequencing analysis was carried out as previously described (Wang et al., 2015; Zhang et al., 2017, 2020; Yunyun et al., 2018; Aziz et al., 2020a,b,c, 2021, 2022a,b).

The Wizard^® Genomic DNA Purification Kit from Promega, Madison, WI, USA was utilized to extract the genomic DNA, which was subsequently quantified using a TBS-380 fluorometer (Turner Bio Systems Inc., Sunnyvale, CA, USA). The insert size for the quantification was 15 kb. For further analysis, DNA of superior quality was utilized, with an OD260/280 ratio ranging from 1.8 to 2.0 and a quantity exceeding 20 μg. The sheared fragments were utilized to prepare Illumina sequencing libraries using the NEXTflex™ Rapid DNA-Seq Kit. In brief, a total of 50 prime ends underwent initial end-repair and phosphorylation. Subsequently, 30 termini were subjected to A-tailing and ligation with sequencing adapters. The third phase involved the amplification of the adapters-ligated products through PCR for enrichment purposes. The libraries that had been prepared were subjected to paired-end Illumina sequencing, with a read length of 150 bp for each end, utilizing an Illumina HiSeq X Ten platform. The genome of the chosen strain, L. plantarum K25, was sequenced using both single molecule real-time (SMRT) technology and Illumina sequencing platforms. The resulting sequence was assigned the accession number GCA_003020005.1. The complexity of the genome was assessed utilizing the Illumina data (Churro et al., 2012). The genomic DNA was isolated in accordance with the manufacturer's protocol, utilizing the Qiagen DNA extraction kit. The isolation process was carried out from March to May 2019.

On April 14, 2023, a customized bash script was used to retrieve 10 genomes of the Lactiplantibacillus plantarum that were accessible from the NCBI RefSeq database (https://www.ncbi.nlm.nih.gov). Our sequenced genomes were the Lactiplantibacillus plantarum K25 strain that was earlier submitted to NCBI with the specifically allocated accession number GCA_003020005.1. To make it simple to distinguish between various genomes, the genomes were changed from assembly names to strain names.

MetaGeneMark (available at http://opal.biology.gatech.edu/GeneMark), was utilized to perform gene coordinates prediction, protein sequence, and nucleotide sequence prediction (Gemayel et al., 2022). Additionally, Prokka v1.12 was also employed to develop pipelines for tRNA, tmRNA, rRNA, and CRISPR sequence prediction using Aragorn (available at http://www.ansikte.se/ARAGORN), and the genome sequence was submitted in the FASTA DNA file to predict the tRNA, tmRNA, and CRISPR sequences present in the genome (Gemayel et al., 2022).

To investigate the taxonomic boundaries of the genomes, average nucleotide identity (ANI) was determined using a genome-to-genome distance calculator (GGDC) (accessible at http://ggdc.dsmz.de) that uses BLAST + as the similarity search program. All of the 10 retrieved genomes along with our reference genome were submitted to the server for analysis (Auch et al., 2010).

For the analysis of genome-based taxonomy, the Type (Strain) Genome Server (TYGS) was utilized, which is freely accessible (https://tygs.dsmz.de). The whole-genome sequence was submitted in the FASTA DNA file for analysis, and the results were further analyzed to evaluate the genome-based taxonomy (Meier-Kolthoff and Göker, 2019).

The pan-genome analysis was performed using the PanX: Pan-Genome Analysis and Exploration online server that is freely available (https://pangenome.org). The species was selected as Lactiplantibacillus plantarum for the analysis, and the core genes, distribution ranking, and strain trees were obtained (Ding et al., 2018).

The analysis of gene plasticity is a valuable tool in identifying drug targets and developing innovative strategies to combat bacterial infections. Understanding the evolution of bacterial genomes and how they adapt to new environments is crucial in this process. By identifying the genomic regions involved in gene plasticity, we can gain insights into the diversity of bacterial populations, the distribution of virulence factors, and the spread of antibiotic-resistance genes. The genomic plasticity analysis was performed by IslandViewer 4 (freely accessible at https://www.pathogenomics.sfu.ca). The genome of L. plantarum K25 was selected for analysis, and the results were interpreted (Dhillon et al., 2013).

The prediction of prophage regions in the genome of the L. plantarum K25 strain was performed using the PHAge Search Tool Enhanced Release (PHASTER), which is freely accessible (https://phaster.ca). The prophage start-end regions, CDS, and percentage GC were predicted as the output that was further interpreted (Arndt et al., 2016). The identification of Bacteriocins in the K25 genome was analyzed using the BAGEL4 online server (freely accessible at http://bagel4.molgenrug.nl). The FASTA DNA file of the genome was given as the input, and the results were interpreted to determine the bacteriocin-producing genes in the genome under study (van Heel et al., 2018).

The retrieved genomic data were saved in the FASTA format for further analysis. There were a total of 10 genomes retrieved from the NCBI RefSeq database. The retrieved reference genomes are shown in Table 1.

For the microbial genome of L. plantarum K25, MetaGeneMark was used to predict genes. A total of 3,199 genes with an average length of 800–1,000 bp were predicted. With an average intergenic distance of 1,000 bp, the predicted genes were dispersed throughout the genome.

Various metrics, including sensitivity, specificity, and positive predictive value (PPV), were used to assess the gene prediction accuracy. The number of genuine genes that were accurately identified was 3,120, which represents the sensitivity of the gene predictions. The number of anticipated genes that were true positives, or specificity, was 1,559 throughout the genome.

Our analysis revealed a wide range of tRNA, tmRNA, and rRNA sequences in the genomes of the organisms studied. For tRNA, we found a variety of types, including tRNA-Val, tRNA-Met, tRNA-Ala, and others. The predicted secondary structures of the tRNA molecules were also analyzed, and we observed variations in the anticodon loops and stems.

In addition to tRNA, we also predicted the presence of tmRNA in the genomes analyzed. tmRNA is involved in the degradation of aberrant proteins during translation, and its prediction can provide insight into protein quality control mechanisms. The total number of tRNA genes was observed to be 69. We found a diverse range of tmRNA sequences, with variations in the size, structure, and location within the genome.

tRNA codon and tRNA anticodon frequencies are shown in Table 2.

Our analysis revealed a wide range of ANI values among the genomes studied. ANI values ranged from 77.4 to 84.8%, indicating significant differences in genomic similarity among the organisms. We observed that organisms within the same genus had higher ANI values, indicating a higher degree of genomic similarity, while organisms from different genera had lower ANI values, indicating a lower degree of genomic similarity.

We also observed that ANI values varied within the same species. In some cases, ANI values were close to 80%, indicating a high degree of genomic similarity, while in other cases, ANI values were lower, indicating a lower degree of genomic similarity. This variability may be due to differences in genomic content or evolution within the same species. Table 3 shows the detailed results of the ANI analysis.

According to our findings, the pan-genome consists of thousands of genes and is quite big and diversified. We found that there are various types of genomes, one consisting of genes shared by all creatures and the other consisting of genes shared by only a subset of those organisms. Different prokaryotes had different-sized auxiliary genomes, which we interpret to mean they had different genomic contents and had evolved in different ways.

The pan-genome's functional annotation uncovered numerous pathways involved in metabolism, transport, and regulation. The presence of genes involved in virulence and antibiotic resistance also varied across the species we examined.

Insights into the evolutionary links among the creatures were gained by building a phylogenetic strain tree based on the pan-genome. Within the strain tree, our data showed separate groups, or clades, which we interpret to represent unique patterns of genomic diversity and evolutionary history. With a few notable exceptions, we found that the evolutionary relationships among the creatures matched their taxonomic classification. The gene count rank distribution ranged from 35 genes with a length of ~6,000 bp. The strain count was observed to be 80. The gene count rank distribution and strain count rank distribution are shown in Figure 1, and the strain tree of the relevant genomes is shown in Figure 2.

Our research showed that the genomes of the creatures we analyzed contained a wide variety of genomic islands and mobile genetic components. Pathogenicity islands, prophages, and integrative conjugative elements are only a few examples of the genomic islands we found. We also found several different kinds of transposons and insertion sequences, which are mobile genomic elements. Genomic islands and MGEs were annotated functionally, revealing a vast variety of genes involved in virulence, antibiotic resistance, and other cellular activities. Differences in the genomic island and MGE presence across studied taxa suggest distinct genomic content and evolutionary histories. The genome island prediction map is shown in Figure 3.

Our research also found evidence of genetic flexibility in locations that are specific to individual or clade-specific species. Additional research into these regions may be worthwhile because they may house genes with crucial roles in the development of certain phenotypes or functions. Table 4 shows the predicted gene names along with their products.

Based on our research, we determined that the genome of the L. plantarum K25 strain has four prophage regions. Different prophage families, such as the Siphoviridae and the Myoviridae, were found dispersed throughout the genome. As a further indicator of changes in genomic content and evolution, we found that the prophage areas varied among the taxa we examined. The prophage regions starting from the genes implicated in virulence, antibiotic resistance, and other cellular functions were among 2111511 to 2153545, containing a total of 24 CDS with a GC content of 43.38%. Several functional categories were discovered by functional annotation of the prophage areas. We found that several prophage areas had genes involved in stress response and DNA repair, which raises the possibility that these prophages help protect the organism from environmental challenges. Figure 4 shows the genomic map of the prophage regions present in the genome.

Moreover, our investigation into the genome of the L. plantarum K25 strain uncovered prophage remnants, suggesting that the organism may have experienced prophage loss or decay. Insights concerning the organism's evolution and its interactions with other bacteria might be gained by studying these prophage remains. Our research showed that the genome of the L. plantarum K25 strain contains numerous clusters of bacteriocin genes. Class I and class II bacteriocins were among the many kinds we found. Additionally, we found that bacteriocin gene clusters varied between the organisms we analyzed, suggesting a spectrum of genomic diversity and evolutionary history.

Genes involved in biosynthesis, regulation, and immunity were only some of the many functional categories discovered by the functional annotation of the bacteriocin gene clusters. We found that certain clusters of bacteriocin genes including Enterocin and Plantaricin and antibiotic-resistance genes raise the possibility that bacteriocins function to defend the organism against both environmental challenges and rival germs. Figure 5 shows the map of the bacteriocin genes predicted.

Our research also found several structural variations of bacteriocins in the genome of the L. plantarum K25 strain. It may be worthwhile to investigate whether or not these variations have unique antibacterial activities and characteristics.