Background and Summary

The human oral microbiome, a complex and diverse ecosystem composed of bacteria, microeukaryotes, archaea, and viruses, plays a critical role not only in oral health but also in overall systemic health through its interactions with the host’s immune system. Bacteria in the genus Streptococcus are among the earliest colonizers of the oral cavity and are typically acquired shortly after birth1. This bacterial group contributes to the maintenance of health and the development of disease in various ways. Among this bacterial group, we focus on Streptococcus salivarius, a dominant, non-pathogenic species known for its positive effects on oral health.

S. salivarius showcases an impressive ability to curb detrimental microorganisms and subtly orchestrate the host’s immune system, positioning it as a promising candidate for both probiotic and postbiotic roles2,3,4. Yet, as often observed in microbial realms, S. salivarius strains vary widely in their capabilities. Among these, several strains show both preventive and therapeutic potential across diverse applications, with strains K12 and M18 — originating from New Zealand — being the most studied and renowned. Strain K12, in particular, exhibits notable efficacy against respiratory pathogens like Streptococcus pyogenes by producing bacteriocins, including salivaricins A2 and B5.Research by Burton et al.2,6 indicates that S. salivarius K12 can inhabit the oral cavity without toxic side effects, where it further aids in reducing inflammation and enhancing oral health by tempering inflammatory cytokines such as interleukin-6 (IL-6) and interleukin-8 (IL-8), typically heightened by periodontal pathogens4.

Meanwhile, S. salivarius M18, known for its production of salivaricin 9, has shown inhibitory effects on pathogens like Corynebacterium and S. pyogenes7. Beyond these prominent strains, others — such as S. salivarius TOVE-R from the UK and S. salivarius JIM8772 from Germany — have also proven capable of impeding cariogenic and periodontal pathogens, while modulating immune responses through the downregulation of inflammatory pathways8,9,10.

As of October 1, 2024, 401 S. salivarius genome assemblies have been submitted to the NCBI database. However, most of these remain incomplete, with only 21 assemblies reaching the complete genome or chromosome level, while the rest are still at the contig or scaffold stage. This reflects the challenges in generating fully resolved, high-quality genomic data, particularly for S. salivarius, a species whose genetic diversity is shaped by factors such as host ethnicity, dietary patterns, and other environmental influences. Notably, most of the strains analyzed so far have been derived from non-Asian populations, with a focus on strains from Europe and other Western regions. Additionally, it is worth noting that S. salivarius strains from Koreans, who have a unique food culture, have not yet been studied in detail.

To address this gap, we sequenced the complete genomes of 12 S. salivarius strains isolated from Koreans of various ages. These fully assembled, chromosome-level genomes provide a valuable resource for investigating the genetic diversity of S. salivarius in a population that has not been widely studied genomically, namely the Korean population. Along with the genomic data, our dataset includes transcriptome profiles of these strains cultured under standard laboratory conditions.

Of particular interest, we noted marked differences in how these strains interact in vitro with Fusobacterium nucleatum subsp. nucleatum — a bacterium intimately linked to periodontal disease and recently spotlighted in studies of colorectal cancer and colitis11,12,13. This dataset, in this regard, offers a valuable resource for comparative genomics and transcriptomics, potentially unveiling the underlying mechanisms of these microbial interactions. Furthermore, it lays the groundwork for upcoming research into the probiotic potential of S. salivarius. Figure 1 visually summarizes this study, with comprehensive methodologies detailed in the Methods section.

Fig. 1
figure 1

Schematic diagram of the study design.

Full size image

Methods

Bacterial strain isolation and species identification

Buccal mucosa samples were collected from six Korean individuals across various age groups. The sample collection was conducted in 2016 and approved by the Institutional Review Board (IRB) at the Dental Hospital of Kyung Hee University (KHD IRB 1606-5). Written informed consent was obtained from all participants, and for those unable to provide consent, consent was obtained from their legal guardians. The participants included a 10-year-old male (KSS7 and KSS8), a 5-year-old female (KSS9 and KSS10), a 21-year-old female (KSS2 and KSS3), a 9-year-old female (KSS1 and KSS11), a 33-year-old female (KSS4 and KSS5), and a 36-year-old female (KSS6 and KSS12). All participants were systemically healthy, had no history of antibiotic use in the past three months, and presented no active oral diseases, although they had restoratively treated teeth. Samples were collected using sterile swabs and placed into phosphate-buffered saline (PBS) to create bacterial suspensions.

To isolate S. salivarius, 200 µl of each suspension was serially diluted and spread onto Mitis Salivarius agar (MS agar) plates (MB Cell, MB-M0621) supplemented with 1% potassium tellurite (MB Cell, MB-P18452). The plates were incubated at 37 °C in an anaerobic chamber with 10% CO2, 10% H2, and a nitrogen balance for 24 hours. After colony growth, individual colonies were subcultured on BD BACTO™ Brain Heart Infusion (BHI) agar (BD, 237500) to propagate the bacteria further.

Species identification was performed using PCR. DNA was extracted from the colonies using InstaGene™ DNA Purification Matrix (Bio-Rad, 7326030) following the manufacturer’s protocol. The complete 16 S rRNA gene sequence (~ 1.5 kb) was amplified using universal eubacteria primers14. The PCR reaction mixture contained 0.1 µg of template DNA, 0.5 µM of each primer, and 1U of AccuPower® Taq PCR PreMix (Bioneer, K-2606). The concentration of each dNTP was 250 µM, with a total reaction volume of 20 µl. The PCR cycle was as follows: an initial denaturation at 95 °C for 5 minutes, followed by 30 cycles of 95 °C for 20 seconds, 53 °C for 20 seconds, and 72 °C for 1 minute and 20 seconds. The final extension was performed at 72 °C for 5 minutes. PCR products were purified and sequenced and species identification was performed via a BLAST search against the GenBank database.

DNA extraction, library preparation, and whole genome sequencing

S. salivarius strains were cultured in BHI broth until the OD600 reached 0.5-0.7. The cultures were incubated at 37 °C in an anaerobic chamber. After incubation, the bacterial cells were pelleted by centrifugation, and the supernatant was discarded. DNA was extracted from the pellet using the Axen™ Total DNA BYC Mini Kit (Macrogen. MG-P-006-50). For whole genome sequencing of each S. salivarius strain, a Single-Molecule Real-Time (SMRT) bell library was prepared following the manufacturer’s instructions (Pacific Biosciences). Briefly, 4 μg of input genomic DNA was used for library preparation. The Femto Pulse System (Agilent Technologies, Santa Clara, CA, USA) was employed to assess the actual size distribution for all size quality checks, ensuring the library insert sizes were in the optimal range. We sheared the genomic DNA using the Megaruptor® 3 (Diagenode, Liège, Belgium) and purified it using AMPure PB magnetic beads (Pacific Biosciences, Menlo Park, CA, USA) for size-selection. A total of 10 μL of the library was prepared using PacBio SMRTbell prep kit 3.0. SMRTbell templates were annealed using Sequel II Bind Kit 3.2 and Int Ctrl 3.2. Sequencing was performed using the Sequel II Sequencing Kit 2.0 and SMRT cell 8 M Tray, with 15-hour movie capture for each SMRT cell, on the PacBio Sequel IIe (Pacific Biosciences, Menlo Park, CA, USA) platform by Macrogen Inc. (Seoul, Korea). The subsequent steps are based on the PacBio Sample Net-Shared Protocol. The raw base calling data was generated and then the HiFi reads were generated using the CCS algorithm. The HiFi read statistics was summarized in Table 1 and the distributions of HiFi read length and quality was assessed and presented in Fig. 2.

Table 1 HiFi read statistics.
Full size table
Fig. 2
figure 2

HiFi read statistics generated from PacBio Sequel II platform. In each strain, the left panel represents HiFi read length distribution and HiFi read quality distributions is shown in the right panel.

Full size image

Genome assembly and pan genome analysis

To perform de novo assembly for the completely circular genomes for S. Salivarius, the SMRT Link software (v12.0.0.177059) for Sequel II system was used. Briefly, the Microbial Genome Analysis pipeline was loaded from pbcromwell wrapper. The parameter for the analysis was “–task-option ipa2_genome_size = 0–task-option ipa2_downsampled_coverage = 0–task-option microasm_plasmid_contig_len_max = 300000–task-option ipa2_cleanup_intermediate_files = True–task-option filter_min_qv = 20”. The 12 completely circular genomes were successfully generated for all 12 S. Salivarius strains. Among them, KSS1 and KSS11 strains have a ~100 kb plasmid in addition to the bacterial chromosome. The NCBI Prokaryotic Genome Annotation Pipeline was employed for gene annotation process15. All strains have similar size of whole genome and the similar number of genes (Fig. 3 and Table 2). In the process of genome assembly, the method could detect the base modification in the genome. In Table S1, we listed the methylation motifs for each genome. To compare the genomic contents, pan-genomes analysis pipeline (PGAP)16 analysis was performed using gene family method with default parameters (score: 40, e-value: 1 × 10−10, identity: 0.5, and coverage: 0.5). Based on pan-genome profiles, the phylogenetic tree among 13 strains including ATCC 7073 type strain was constructed (Fig. 4a). Interestingly, the strain pairs from the same subject; KSS1 - KSS11, KSS2 - KSS3, KSS4 - KSS5, KSS6 - KSS12, and KSS9 – KSS10 were relatively close each other in the tree. The orthologous gene clusters identified in each strain from the PGAP analysis were categorized into core, dispensable, and strain-specific gene clusters. The pangenome of the twelve S. salivarius strains (Fig. 4b) was composed of 1,446 core gene clusters (54.5%, present in all strains), 771 dispensable gene clusters (29.1%, partially shared among the strains), and 435 strain-specific gene clusters (16.4%).

Fig. 3
figure 3

Circular genome representations for twelve complete S. salivarius strains.

Full size image
Table 2 Genome features of 12 Streptococcus Salivarius strains.
Full size table
Fig. 4
figure 4

Comparative genomic analysis of twelve S. salivarius strains. (a) A phylogenetic tree was constructed based on pan-genome profiles. (b) Pan-genome analysis using PGAP method.

Full size image

RNA extraction, library preparation, and whole transcriptome sequencing (RNA-Seq)

S. salivarius strains were grown as described above. After centrifugation, the supernatant was removed, and the bacterial cells were processed using the AccuPrep® Bacterial RNA Extraction Kit (Bioneer, K-3143), following the manufacturer’s protocol. Total RNA concentration was determined using Quant-IT RiboGreen (Invitrogen, #R11490). RNA integrity was evaluated by running the samples on a TapeStation RNA screentape (Agilent). Only RNA samples with a high RNA Integrity Number (RIN) above 7.0 were selected for library construction. For each sample, RNA libraries were individually prepared using 1 μg of total RNA with the Illumina TruSeq Stranded mRNA Sample Prep Kit (Illumina, Inc., San Diego, CA, USA, #20020595). As the first step, bacterial rRNA was depleted using the NEBNext rRNA Depletion kit (Bacteria) (NEB). After rRNA depletion, the remaining RNA was broken down into smaller fragments through the use of divalent cations at a high temperature. The resulting RNA fragments were then reverse transcribed into first-strand cDNA using SuperScript II reverse transcriptase (Invitrogen) and random primers. Next, second-strand cDNA synthesis was performed using DNA Polymerase I, RNase H and dUTP. The resulting cDNA fragments underwent process of end repair, the addition of a single ‘A’ nucleotide, and adapter ligation. These products were subsequently purified and amplified through PCR to form the final cDNA library. The libraries were quantified using the KAPA Library Quantification kits designed for Illumina Sequencing platforms, following the qPCR Quantification Protocol Guide (KAPA), and further assessed using TapeStation D1000 ScreenTape (Agilent). The indexed libraries were then submitted for paired-end (2 × 101 bp) sequencing on an Illumina NovaSeq. 6000 (Illumina, Inc., San Diego, CA, USA), which was performed by Macrogen Inc. (Seoul, Korea).

RNA-Seq data processing

In each strain, two biological replicates were sequenced and a total of approximately 388 (an average per sample: 16 million pairs of reads) million pairs of read (2 × 101 bp) were generated. The adapter sequences in the raw reads were trimmed by cutadapt software (v 4.9)17. The trimmed fastq files were assessed by FastQC (v 0.10.1) (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and summarized by MultiQC (v 1.25.1) (https://seqera.io/multiqc/) (Fig. 5a). The preprocessed reads were aligned to S. Salivarius ATCC 7073 (type strain) genome using Bowtie2 (v 2.5.4)18 with the parameters, “-k 2–no-discordant”. Around 298 million read pairs (with an average of 12 million pairs per sample) were uniquely mapped and properly paired in the paired-end sequencing mode (Table S2). The uniquely mapped reads were used for calculating read counts for each gene annotated in the S. Salivarius ATCC 7073 RefSeq sequence (GCF_900143035.1). A total 1,914 genes have at least one read mapped on genic regions across all twelve S. salivarius strains. The hierarchical clustering and heatmap based on the normalized counts were generated by DESeq. 2 R package (v 1.30.1)19 (Fig. 5b). As expected, the replicate samples were located to next to each other and similar to the phylogenetic tree based on the genome assembly data, the expression profiles of strain pairs from the same subject were clustered into the same cluster.

Fig. 5
figure 5

RNA-Seq analysis of twelve S. salivarius strains. (a) Evaluation of sequencing quality of RNA-Seq reads assessed by FastQC and MultiQC. (b) A hierarchical clustering and heatmap based on the normalized read counts for the mapped genes.

Full size image

Data Records

The PacBio DNA sequencing data for de novo assembly was deposited to NCBI Sequence Read Archive (SRA) database and the assembled genomes were available in NCBI GenBank; SRP53765120 (CP14585621, CP14585722), SRP53765323 (CP14586024), SRP53765425 (CP14586126), SRP53765527 (CP14586228), SRP53766329 (CP15084730), SRP53765631 (CP14586332), SRP53765733 (CP14586434), SRP53765835 (CP14586536), SRP53765937 (CP14586638), SRP53766039 (CP14586740), SRP53766141 (CP14585842, CP14585943), SRP53766444 (CP14586845). RNA sequencing data was submitted into NCBI Gene Expression Omnibus (GEO) database and the GEO accession number is GSE27879146.

Technical Validation

Assessment of sequencing data quality

All paired-end RNA-Seq reads (101 bp) generated from Illumina platform, were quality checked by FastQC (v 0.10.1). All RNA-Seq samples showed a mean Phred score of 30 or higher in every position in the read (Fig. 5a). There are decent number of read pairs uniquely mapped onto the genome of ATCC 7073 S. salivarius type strain in all 24 samples. Furthermore, the gene expression patterns of S. salivarius strains obtained from the same biological replicate as well as from the same subject were similar to each other, confirming the high quality of the sequencing.

Genome completeness analysis

The de novo assembled genome completeness was checked by BUSCO (v 5.7.1)47 analysis (Fig. 6). The BUSCO is using the predefined bacterial orthologous genes (bacteria_odb10) to evaluate the integrity of the de novo assembled genome. The genome completeness of strains KSS2, KSS5, KSS8, and KSS9 was 98.4%, while the genome completeness of the remaining strains was 99.2%.

Fig. 6
figure 6

BUSCO analysis for the de novo assembled genome completeness.

Full size image