The Gram-positive α-hemolytic Streptococcus suis is a major pathogen in the swine industry and an emerging zoonotic agent that can cause several systemic issues in both pigs and humans. A total of 35 S. suis serotypes (SS) have been identified and genotyped into > 700 sequence types (ST) by multilocus sequence typing (MLST). Eurasian ST1 isolates are the most virulent of all S. suis SS2 strains while North American ST25 and ST28 strains display moderate to low/no virulence phenotypes, respectively. Notably, S. suis 90–1330 is an avirulent Canadian SS2-ST28 isolate producing a lantibiotic bacteriocin with potential prophylactic applications. To investigate the suitability of this strain for such purposes, we sequenced its complete genome using the Illumina and PacBio platforms. The S. suis 90–1330 bacteriocin was found encoded in a locus cargoed in what appears to be an integrative and conjugative element (ICE). This bacteriocin locus was also found to be widely distributed across several streptococcal species and in a few Staphylococcus aureus strains. Because the locus also confers protection from the bacteriocin, the potential prophylactic benefits of using this strain may prove limited due to the spread of the resistance to its effects. Furthermore, the S. suis 90–1330 genome was found to code for genes involved in blood survival, suggesting that strain may not be a benign as previously thought.
Citation: Sun Y, Veseli IA, Vaillancourt K, Frenette M, Grenier D, Pombert J-F (2019) The bacteriocin from the prophylactic candidate Streptococcus suis 90-1330 is widely distributed across S. suis isolates and appears encoded in an integrative and conjugative element. PLoS ONE 14(4): e0216002. https://doi.org/10.1371/journal.pone.0216002
Editor: Axel Cloeckaert, Institut National de la Recherche Agronomique, FRANCE
Received: January 26, 2019; Accepted: April 12, 2019; Published: April 30, 2019
Copyright: © 2019 Sun et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All custom Perl scripts are available at https://github.com/PombertLab/. The S. suis 90-1330 genome has been deposited in GenBank under BioProject PRJNA295196, BioSample SAMN04044079, and accession numbers NZ_CP012731.1 (genome) and KY994647 (plasmid). Illumina and PacBio sequencing reads (including methylation motifs) were deposited in the NCBI Sequence Read Archive under accession number SRP104862: Illumina — SRX2753615 (experiment), SRR5468260 (run); PacBio SRX2755389/SRX2755390 (experiments) and SRR5470805/SRR5470806 (runs). Methylation patterns were included in REBASE (http://rebase.neb.com/cgi-bin/pacbioget?18488). All S. suis strains used in this study are listed in S1 Table.
Funding: This work was supported by funds from the Illinois Institute of Technology (https://web.iit.edu/) to Jean-François Pombert. Iva A. Veseli gratefully acknowledges support by an Undergraduate Research Fellowship from the American Society for Microbiology (https://www.asm.org/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The quick rise of antibiotics resistance in the microbial world is problematic to multiple fields, including the food industry. As such, novel antibiotic compounds and alternate strategies to treat infections and prevent growth of pathogenic bacterial species are increasingly being sought after. One promising avenue is the prophylactic use of bacteriocin-producing commensal or neutral species as probiotics to outcompete virulent ones . Bacteriocins form a wide range of antimicrobial peptides produced by microorganisms . These molecules, frequently short peptides, can affect a narrow to large spectrum of bacteria excluding the species producing the compounds, which are immune to their effects via varying mechanisms . Because of the potentially disruptive effects on the microflora of the host, the ideal species for prophylactic use would produce compounds whose spectrum do not disrupt the commensal organisms inhabiting the targeted environment. Recently, a nisin-related lantibiotic bacteriocin with a membrane permeabilization activity has been found in Streptococcus suis strain 90–1330 . This lanthionine-containing bacteriocin is a killing peptide active against a number of Gram-positive streptococcal and staphylococcal species but shows little to no activity against Gram-negative bacteria .
S. suis is a genetically diverse Gram-positive streptococcal species displaying at least 35 distinct serotypes and categorized in over 700 sequence types by multilocus sequence typing (MLST) . Of these, serotype 2 (SS2) strains are of particular importance to the swine industry, where they can cause a plethora of severe infections in pigs including endocarditis, meningitis, pneumonia and septicemia [6,7]. S. suis SS2 species are also a growing cause of concern in human health and have been associated with several zoonotic infections [7–16]. Not all serotype 2 species display the same levels of virulence, however; sequence type 1 (ST1) isolates are usually the most virulent while sequence types 25 (ST25) and 28 (ST28) show mild to low/no virulence, respectively . Notably, the bacteriocin-producing S. suis 90–1330 (aka 1330) is a SS2-ST28 strain that has been established as avirulent in mice and swine . Considering this lack of virulence and the antibacterial activity of its bacteriocin against Gram-positive swine pathogens, including S. suis SS2-ST1 strains and staphylococcal species Staphylococcus aureus and Staphylococcus hyicus , this bacterium appears to be an excellent candidate for prophylactic use  in the swine industry.
Here we sequenced the complete genome of S. suis 90–1330 using the Illumina and PacBio short and long-read platforms, respectively, to investigate its suitability for prophylactic use and ensure that this strain is indeed devoid of toxin-encoding genes or any other component that could potentially cause problems with its usage as a probiotic. We also explored the origin and distribution of the bacteriocin encoded by S. suis 90–1330 across streptococcal species. In the process, we used the underlying genomic data to identify which S. suis strains are the closest to this SS2-ST28 representative.
Material and methods
Bacterial culture and DNA purification
S. suis isolate 90–1330 (aka 1330), originally isolated from pigs, was acquired from Dr. Marcelo Gottschalk, Faculty of Veterinary Medicine, Université de Montréal . The isolate was cultivated at 37°C in THB medium (IBI Scientific, Peosta, IA, USA) with low agitation (100 rpm). Total genomic DNA was purified using the MasterPure Gram Positive DNA Purification Kit (Epicentre Biotechnologies, Madison, WI, USA) with a lysis incubation time of 3 hours and additional centrifugation steps to remove protein precipitates. DNA thus harvested was quantified with a Qubit 2.0 fluorometer (Life Technologies, Waltham, MA, USA) and its purity assessed electrophoretically on 0.8% agarose gels and spectrophotometrically for the presence of RNA and protein contaminants with the A260/A230 and A260/A280 ratios, respectively, using a Nanodrop 2000c (Thermo Scientific, Waltham, MA, USA).
Genome sequencing was performed using the Pacific Biosciences (PacBio) long read and the Illumina short read platforms. PacBio libraries were prepared and sequenced by the University of Michigan DNA Sequencing Core (Ann Arbor, MI, USA). PacBio SMRTbell libraries were prepared per the PacBio 20 Kb+ protocol with g-TUBE shearing (Covaris, Woburn, Massachusetts, USA) and BluePippin (Sage Science, Beverly, MA, USA) size selection. A total of 108,170 polymerase reads (N50: 19.39 Kb) were sequenced using the P6-C4 chemistry for a total of 1.34 Gbp (420x coverage). Illumina libraries were prepared and sequenced by the Plate-forme d'Analyses Génomiques de l'Université Laval (Quebec City, QC, Canada). DNA for Illumina sequencing was fragmented with a Covaris sonicator and libraries were generated using the Illumina TruSeq chemistry. A total of 663,635 Paired-Ends (PE) 300 bp-long reads (1,327,270 reads total) were sequenced on the Illumina MiSeq platform for a total of 398 Mbp (185x coverage).
PacBio reads were processed and assembled with the RS_HGAP_Assembly.3 pipeline as implemented in SMRT Analysis 2.3.0 [https://github.com/PacificBiosciences/SMRT-Analysis/wiki/SMRT-Analysis-Software-Installation-v2.3.0]. Base calling on the initial genome assembly was verified by mapping the PacBio reads back onto the assembled genome using the RS_Resequencing.1 pipeline from SMRT Portal. Base calling was further verified and corrected by mapping the Illumina reads onto the refined assembly using Geneious R7.1.8 (Biomatters, Auckland, New Zealand) as follows: reads were mapped using the built-in Map to Reference function, divergent positions were highlighted with the Find Variations/SNPs tool, and bases were corrected accordingly. The genome was circularized by detecting overlaps with BLASTN homology searches  and by adjusting the edges with Consed 29.0 . Genome methylation patterns on the final assembly were detected using the RS_Modification_and_Motif_Analysis pipeline in SMRT Portal. The presence of 5-methylcytosine (5mC) was not assessed due to the small kinetic differences involved between methylated and non-methylated bases, as 5mC residues were not converted to 5-carboxylcytosine (5caC) using Tet1-oxidation prior to sequencing.
Genome annotation and comparative genomic analyses
Transfer and ribosomal RNAs were positioned on the S. suis 90–1330 genome with tRNAscan-SE 1.3.1  and RNAmmer 1.2 , respectively. Preliminary protein predictions and annotations were performed with PROKKA 1.11  with products inferred using an E-value threshold of 1e-30 and validated by InterProScan 5.15–54.0  analyses, with conflicting results resolved by manual curation against the UniProt database . Final annotations in the RefSeq genome were performed with the NCBI Prokaryotic Genome Annotation Pipeline (April 2017 release). ComRS competence genes and promoters were further identified by BLAST homology searches using homologous sequences from other S. suis strains as queries [26,27]. GC-skew analyses were performed with GenSkew [http://genskew.csb.univie.ac.at/]. KEGG pathway categories for each putative protein were attributed with BlastKOALA [http://www.kegg.jp/blastkoala/]. Physical maps were plotted with Circos . Comparative alignments between S. suis 90–1330 and other genomes were performed using the progressive Mauve algorithm as implemented in Mauve 2.4.0 . Transposons shared between S. suis 90–1330 and other strains were identified by BLASTP homology searches. Repeated loci were identified using RepeatFinder from the Geneious R9.1.7 package, with ribosomal RNAs filtered out from the results manually.
Genetic diversity inferences
Complete and draft S. suis genomes were downloaded from the NCBI GenBank database with the queryNCBI.pl custom Perl script. Pairwise mutation distances using the MinHash dimensionality-reduction technique were calculated with Mash  and genetic clusters were plotted with t-Distributed Stochastic Neighbor Embedding (t-SNE ) as implemented in the R  Rtsne package  (https://github.com/jkrijthe/Rtsne) with the run_Mash.pl, MashToDistanceCSV.pl and MashR_plotter.pl custom Perl scripts. Single nucleotide polymorphisms (SNPs) from Fig 1C were inferred using read mapping followed by variant calling. To standardize datasets, synthetic Illumina reads (300 bp paired-ends, insert size 500 bp) were generated for every genome investigated at a sequencing depth of 100X using random subsampling with SSRG.pl 1.5 (https://github.com/PombertLab/SNPs/tree/master/SSRG). Briefly, this Perl script deconstructs a genome (string) into reads (substrings) of specified length in both forward and reverse-complemented orientations using random subsampling, producing synthetic reads akin to reads produced by shotgun sequencing methods. Synthetic reads were mapped against genomes in paired end mode (PE) with Bowtie2 22.214.171.124  and pairwise SNPs were calculated with VarScan2 2.4.3 , as implemented in get_SNPs.pl 1.8 (https://github.com/PombertLab/SNPs/tree/master/SSRG) using default parameters. Results from synthetic reads were validated by mapping the S. suis 90–1330 illumina sequencing data against the same genomes.
(A) Structural genome reorganization between S. suis 90–1330 (left) and ST3 (right), an SS3-ST35 representative. Syntenic regions that are conserved between 90–1330 and ST3 are highlighted by ribbons. G+C percentages are plotted under each genotype (90–1330, purple; ST3, green), with values higher and lower than 45% and 38% colored in blue and red, respectively. SNPs between the SS2-ST28 strains 90–1330 and 05HAS68 (left) and the SS3-ST35 strains ST3 and YB51 (right) are highlighted by blue bars under the corresponding genotypes. The putative ICE insertion site in 90–1330 is denoted by an asterisk. (B) Protein orthologs shared between representative SS2-ST28 and SS3-ST35 strains as identified by BLAST bidirectional best hit approaches (E-value cutoff of 1e-10). (C) Numbers of pairwise SNPs between representative SS2-ST28 and SS3-ST35 strains.
Virulence/antimicrobial resistance predictions, prophage and ice analyses
Streptococcal virulence proteins described by Segura et al.  were searched for against the S. suis 90–1330 inferred proteome using BLASTP homology searches with an E-value threshold of 1e-10 . Other putative virulence proteins were downloaded from the core dataset of the virulence factor database (VFDB) (http://www.mgc.ac.cn/VFs/main.htm) and searched against S. suis 90–1330 using BLASTP (E-value cutoff: 1e-10). The presence of antimicrobial resistance proteins in S. suis 90–1330 was assessed using the ResFinder 2.1 online database . Putative prophages were identified using PHASTER .
The presence of a putative ICE in S. suis 90–1330 surrounding the suicin-containing locus was first delimited manually based on the differences highlighted by GC-skew analyses, on the presence of a site-specific integrase, on the presence of ICESa2603-like direct repeats (TTATTTAAGAGTAAC; inferred from ) flanking the corresponding locus, on the presence of conjugation proteins within the locus, and on the presence of a putative ICE oriT site. Tn5252 transposon sequences from Streptococcus pneumoniae were downloaded from GenBank (accessions numbers L29324.1 and AF295925.1; [40,41]) and searched for against the putative S. suis 90–1330 ICE using BLASTP searches (E-value cutoff: 1e-10). Type IV secretion system proteins were then searched for using CONJscan and TXSScan as implemented in MacSyFinder 1.05 [42,43]. Orthologous proteins between the S. suis 90–1330 putative ICE and ICESsuCZ130302 from S. suis CZ130302 (accession number NZ_CP012731.1) were inferred by BLASTP homology searches (E-value cutoff: 1e-10). The presence of the integrated and circular forms of the putative ICE in the S. suis 90–1330 stationary and logarithmic growth phases was assessed by polymerase chain reaction (PCR) with the following primers: GACTCATGCCAAGCCCGAATAG (AN924_RS09900; 1,979,668 to 1,979,689 [forward]), TCTCAGACATAGCCATGCATCC (AN924_RS09905; 1,980,323 to 1,980,339 [reverse complement]), CGCGTAGGCTACCTTAACTTCC (AN924_RS10235; 2,050,558 to 2,050,579 [forward]) and GTTGCATACGCTGTCAAAGCTG (AN924_RS10250; 2,051,504 to 2,051,525 [reverse complement]). PCRs were performed with the following cycles: denaturation (94°C, 1 min), annealing (94°C, 1 min), elongation (72°C, 2 min, 30 cycles), final extension (72°C, 3 min).
A normal streptococcal genome
The S. suis 90–1330 complete genome maps as a single circular molecule of 2,146,151 bp with a GC content of 41.1%. The genome encodes a total of 2,192 genes including 2,120, 56 and 12 genes coding for proteins, transfer RNAs and ribosomal RNAs, respectively. A total of 8 copies of the LSU/SSU ribosomal RNA operon are located on the S. suis 90–1330 chromosome. The genome has a gene density of 1.13 gene per kb and an average length of 111.5 bp for the intergenic spacers. No gene was found interrupted by group I or group II bacterial introns. The genome features a single bidirectional origin of replication, located at position 1,057,979 of the assembled chromosome (S1 Fig), as calculated by GC-skew analyses and by identifying the position of the dnaA gene . A putative mobile genetic element (MGE) featuring a GC content lower by about 10% was also found inserted in the S. suis 90–1330 chromosome, near one of the two junctions where the GC-skew shifts (S1 Fig). The chromosome encodes several transposable elements that are commonly found in other streptococcal species, including S. suis ST28 strain 05HAS68 (S2 Table). Only one type (type II) of clustered regularly interspaced short palindromic repeats (CRISPR) was identified, limited to a single locus composed of eight 36 bp-long repeat units separated by unevenly-sized spacers. This locus—from 189,953 to 190,448 on the chromosome—is found in other Streptococcus species including strains of S. macedonicus and S. thermophilus (from 76% to 89% identity), and displays homology against the genome of Streptococcus phage 128 (E-value: 2E-18) . The 90–1330 genome also features a few additional dispersed repeated elements, mostly from gene duplicates, with a total of 104 distinct motifs of a least 100 nt distributed across 210 loci, with 16 and 32 repeats longer than 1,000 and 500 nt, respectively (excluding the ribosomal RNA copies; S3 Table). Methylation patterns AGCNNNNNGCT and GTAC were found compatible with the type I and type II restriction–modification systems commonly found in streptococcal species ([46,47]; S1A Fig). In addition to the chromosome, S. suis strain 90–1330 also contains a 4,984 bp-long circular plasmid (pSS90-1330) encoding a total of 6 genes involved primarily in recombination and replication (S1B Fig).
Of the 2,120 protein coding genes found on the S. suis 90–1330 chromosome, 1,803 were assigned putative functions by the NCBI PGAP pipeline, with 1,161 assigned KEGG orthologs distributed across common streptococcal metabolic pathways (S1 Fig, S4 Table). The S. suis 90–1330 genome further codes for the type III ComRS competence system typically found in S. suis genomes, including the methylase DpnA required to prevent degradation of foreign DNA acquired via competence ([26,27,46]; S5 Table), and overall the S. suis 90–1330 gene content was found very similar to other S. suis SS2-ST28 genomes, as expected. Interestingly, however, the genotypes of S. suis SS2-ST28 isolates were found to be nearly identical to those from serotype 3 strains: 1) the SS2-ST28 and SS3-ST35 strains display very high levels of synteny, with only a few chromosomal rearrangements between SS2-ST28 strain 90–1330 and SS3-ST35 strain ST3 (Fig 1A), 2) they share a very large number of genes (Fig 1B), and 3) they show very little divergence at the nucleotide level (Fig 1C). This association between SS2-ST28 and SS3-ST35 strains was found compatible with the previously reported phylogenetic analysis of core S. suis genes (see figure S4 from Okura et al. ), in which these strains clustered together. Altogether, these observations suggest that serotype conversion occurred during the evolution of these strains.
The S. suis 90–1330 suicin appears encoded in an integrative and conjugative element
The S. suis 90–1330 bacteriocin AN924_RS09965 (aka suicin ) is encoded in a locus displaying an unexpectedly low GC content that contrasts sharply with the remainder of the chromosome (see Fig 2 and note on S1 Fig). In addition to the suicin cluster, this low GC content region contains a gene coding for a conjugative relaxase (AN924_RS09970) and a plasmid mobilization relaxosome protein (AN924_RS09975), both involved in plasmid conjugation . The locus is flanked on either side by a pseudogenized transposase (AN924_RS10000) and a site-specific integrase (AN924_RS09905). The C-terminal segment of the predicted integrase sequence is homologous (E-value; 1.65e-38) to tyrosine-based site-specific phage recombinases. However, no other viral gene—including those coding for the viral capsid or tail components and necessary for the lytic cycle—was found in the corresponding locus. Genome-wide searches for the presence of phage sequences in S. suis 90–1330 revealed four incomplete and one questionable phages with GC contents averaging that of the chromosome (S1 File), none overlapping with the bacteriocin-containing locus.
A zoom-in of the low-GC locus surrounding the bacteriocin and the predicted map of the excised ICE are shown on the top and bottom of the figure, respectively. Genes and their directionality are indicated by arrows and color-coded by their products: orange–integrase/recombinase; light-purple–suicin-related proteins; deep-blue–conjugative DNA processing proteins; deep-purple–Type IV coupling protein; red–Type IV secretion channel proteins; light green–surface adhesin; yellow–DNA methyltransferases; light-blue–antibiotic resistance proteins; deep-gray–replication initiator protein; deep-green–TXSScan FATA_gbs matches (E-value cutoff 1e-40); light-gray–miscellaneous. The GC content of the zoom-in is shown in the histogram above the schema (sliding window 100 nt, step 100 nt). oriT (origin of transfer); DR (ICESa2603-like direct repeat).
Besides the abovementioned relaxase (AN924_RS09970) and relaxosome protein (AN924_RS09975), further investigation of the genomic surroundings of the bacteriocin-containing locus revealed the presence of genes coding for a Type IV coupling protein (T4CP; AN924_RS10200), for Type IV secretion system (T4SS) proteins, including a Virb4 ATPase (AN924_RS10180) and a murine hydrolase (AN924_RS10175), and for a surface adhesin (AN924_RS10155; Fig 2; S6 Table) common to integrative and conjugative elements (ICE) . In total, 13 genes from the Firmicutes, Actinobacteria, Tenericutes and Archaea (FATA) and Conjugation (CONJ) sets of hidden Markov models  were identified by TXSScan/CONJscan searches (S6 Table). BlastP searches performed in parallel using Tn5252 transposon sequences from Streptococcus pneumoniae [40,41] revealed homology for Tn5252 open reading frames orf4 [E-value 4e-150], orf9 [E-value 5e-47], orf10 [E-value 6e-33], orf25 [E-value 7e-42], orf26 [E-value 0.0] and for fused open reading frames orf20/21 [E-values 3e-19/0.0], orf23/24 [E-values 4e-45/9e-43] and orf27/28 [E-values 5e-133/0.0] (Fig 2). Conserved domain database (CDD) searches on the site-specific integrase also retrieved homology [E-value 1.65e-38] with the C-terminal catalytic domain of integrases from the ICEBs1 family. A putative oriT sequence (TCAGATATTGTGTCCACAATATACCAAAAAT/CATATCGC) was found from 1998605 to 1998636 downstream of the gene coding for the accessory relaxosome protein MobC (Tn5252 orf9). Two direct repeats (TTATTTAAGAGTAAC) identical in sequence to those from several ICESa2603 family members present in S. suis and Streptococcus agalactiae  were found located respectively 45 bp upstream of the 3’end of the integrase, from 1,980,062 to 1,980,076, and downstream of the 3’ end of the gene coding for ribosomal protein L7/L12 (AN924_RS10245), from 2,051,047 to 2,051,061. Notably, the macrolide and tetracycline resistance genes ermB and tetO were also found encoded in some of these ICESa2603 family members.
To assess the functionality of the putative S. suis 90–1330 ICE, PCR were performed to investigate the presence of the excised circular form of the ICE (S2 Fig). Results showed the presence of this circular form, confirming the ability of the ICE to adopt the extrachromosomal form that is required for conjugation. This circular form was mainly detected in stationary phase cells while it was barely detected in cells in the logarithmic growth phase (results not shown). A PCR product corresponding to the junction of the genes contiguous to the ICESa2603-like direct repeats was also detected confirming the absence of the chromosome-integrated ICE in a portion of the cells. This type of population heterogeneity has been previously reported for other ICE (e.g. ICESsuCZ130302 ). Altogether, these results strongly suggest that the bacteriocin cluster is located in an integrative and conjugative element.
Recently, Pan and colleagues characterized an ICE in S. suis strain CZ130302 (ICESsuCZ130302) featuring a bacteriocin . Comparisons between the S. suis 90–1330 suicin locus and that from ICESsuCZ130302 revealed that the encoded products are nearly identical in sequence (S7 Table), with the bacteriocins differing in sequence by a single amino acid. Further comparisons between the S. suis 90–1330 putative ICE and ICESsuCZ130302 revealed that they are also very similar in structure, with most of the differences located in the cargo area of the ICE (S3 Fig).
The bacteriocin-containing locus is found across several divergent S. suis isolates
Initially thought to be rare , the suicin encoded in S. suis 90–1330 (AN924_RS11115) was found at 100% amino acid identity in several strains of S. suis, S. agalactiae, Streptococcus pneumoniae and other streptococcal species, in seven Staphylococcus aureus isolates, and in one strain (F0221F 224) of Enterococcus hirae (Fig 3). When lowering the similarity threshold to 85%, the bacteriocin was also found in streptococcal species Streptococcus pseudopneumoniae, Streptococcus thoraltensis and Streptococcus equinus (93, 91 and 85% aa identity, respectively). Expanding the search to all genes present in the suicin-containing locus and those contained in the S. suis 90–1330 putative ICE, a total of 43 genes were found distributed across most of the bacteriocin-containing strains, including ICESsuCZ130302-encoding S. suis strain CZ130302 (Fig 3). These genes include the T4SS and conjugative DNA processing proteins identified by TXSScan/CONJscan searches, suggesting that this bacteriocin locus is common in streptococcal ICEs. In most loci, the suicin-containing segments were also found to display lowered GC contents compared to the rest of their surroundings (averaging 31.1% vs. 41.1%, respectively). Many strains like S. suis JS14, S. suis LS9N or S. agalactiae C001 feature a different integrase/recombinase than that of S. suis 90–1330 and the S. suis 90–1330 suicin locus was found to be closer to that of Streptococcus pasteurianus ATCC 43144 (99% identity over 100% of its sequence) than to that of the other S. suis strains. Interestingly, the inner section of the S. suis 90–1330 putative ICE appears unique to this genome (Fig 3), with homology searches returning matches covering at best 58% of the region in Streptococcus parauberis. The inner segment of the putative ICE element in S. suis 90–1330 is flanked by a phage terminase and a recombinase, which suggests that the region was added by recombination.
Locus tags/gene names from S. suis 90–1330 are indicated on top (pseudogenes are not shown); genes from Lebel et al.  are highlighted in magenta (sslA; bacteriocin). ICE-related genes identified by TXSScan/CONJscan searches and by Tn5252 BLASTP homology searches are highlighted in blue. The internal cargo-like segment of S. suis 90–1330 is delineated by black bars. Gene copy numbers were inferred by TBLASTN searches with E-value and identity cutoffs of 1e-10 and 75%, respectively. Strains with identical contents were collapsed into single rows to increase readability. S. suis 90–1330 and CZ130302 from S3 Fig are highlighted in orange.
To map the distribution of the bacteriocin-containing locus across S. suis strains in a phylogenetic framework, we performed genetic distances estimations and plotted their presence on a dimensionality reduction graph. The bacteriocin-containing strains were found scattered across several distinct clusters (I, II, III, IV, V, VIII, XII and XIV; Fig 4), including the highly divergent clusters XII and XIV, suggesting a wide distribution across sequenced S. suis isolates.
MASH pairwise genetic distances were clustered by similarity using t-Distributed Stochastic Neighbor Embedding (t-SNE ). Bacteriocin-containing clusters are indicated by asterisks (*). Nodes identified by roman numerals (I—XX) were color-coded by proximity. Complete genomes are indicated by black circles and labelled using the corresponding color palette. Note that because of the multidimensional reduction, the relative distances between the nodes are not quite to scale on the two-dimensional plane.
The S. suis 90–1330 genome codes for genes involved in blood survival
To investigate the suitability of S. suis 90–1330 as a prophylactic strain for the swine industry, we searched for the presence of genes coding for virulence factors and for products involved in antimicrobial resistance.
Because there is no clear consensus on what causes pathogenicity in S. suis strains, we searched the S. suis 90–1330 genome for all putative virulence factors catalogued and discussed by Segura et al.  using a permissive cut-off of 1E-10 (see S8 Table). Unsurprisingly, the S. suis 90–1330 genome codes for capsular polysaccharides (CPS) genes (e.g. cpsC, AN924_RS02950; cpsE, AN924_RS02940) and for capsule regulatory genes (e.g. ccpA, AN924_RS01545), as expected from a serotypable strain. In contrast, of the three genes commonly used as multiplex PCR markers to predict S. suis virulence —sly (hemolysin suilysin; a thiol-activated toxin ), epf (extracellular factor) and mrp (muramidase-released protein)—only mrp (whose product is involved in epithelial cell adherence [54,55]) was found encoded in the genome. This result was expected based on previous PCR-multiplex virulence inquiries  and genomic investigations of various ST28  and ST25  S. suis strains. Expanding the search to genes coding for products potentially involved in bacteremia and/or enhancing S. suis survival in blood revealed the presence of yqfA (AN924_RS10290) coding for hemolysin III, of fbpA (AN924_RS01115) whose product’s N-terminus shows strong homology (9.6E-151) with fibronectin-binding protein A (FbpA) in InterProScan/PFAM searches, and of iga (AN924_RS10410), coding for a proteinase potentially involved in host immune evasion . However, the latter was found upon further inspection to be pseudogenized, with read-mapping over the corresponding locus confirming that the gene is indeed broken and not the result of a sequencing error. Overall, only 24 out of the 140 genes searched against the S. suis 90–1330 genome were not found at the 1E-10 threshold (S8 Table). Additional searches against the virulence factor database (VFDB; http://www.mgc.ac.cn/VFs/main.htm) revealed the presence of more CPS genes and of a gene coding for a TlyC-like (CDD E-value 1e-130) hemolysin C family transporter (AN924_RS07935) potentially involved in virulence (S9 Table). These searches also returned a putative match with a cytolysin activator (cylA; E-value: 8.00E-31) but not for a cytolysin, and upon further inspection the match was found to be spurious, as the corresponding gene (AN924_RS09965) is located on the bacteriocin locus  and codes for a lantiobiotic-specific protease (E-value 2.81e-114).
As expected from previous work on S. suis ST28 strains [56,58], S. suis 90–1330 is not devoid of antimicrobial resistance and the macrolide and tetracycline resistance genes ermB/ermL and tetO were found in the genome with 99.86% and 100% identity against representative sequences from the ResFinder antimicrobial database , respectively, over the full length of their coding sequences. Like other streptococcal strains, S. suis 90–1330 genome codes for several toxin-antitoxin systems (AbiEii/AbiGii; HicA/HicB, RelE/ParE, PezT/PezA, Phd/YefM, Txe/YoeB), and whose overexpression inhibit growth of the bacterium . Interestingly, in addition to the previously reported lantibiotic bacteriocin active against Gram-positive species , the S. suis 90–1330 genome also codes for two Holin (AN924_RS09050, AN924_RS08990) and one Holin-like (AN924_RS10900; pfam16935) toxins with Gram-positive antibacterial properties .
In healthy microbiotas, the presence of commensal and/or neutral microorganisms outcompete pathogenic species for the limited pool of nutrients, resources available and adherence sites, providing a helpful barrier against diseases. This helpful association between host and commensal/neutral microorganisms forms the basis behind the prophylactic use of probiotics. Intuitively, prospective probiotics should be devoid of toxicity and side effects while providing clear positive benefits to the health of the individual(s). Recently, strain 90–1330 from the low virulence/nonvirulent Streptococcus suis serotype 2 –sequence type 28 was shown to express a bacteriocin with a membrane permeabilization effect that inhibits not only the growth of virulent serotype 2 strains of S. suis, but also of other Gram positive swine pathogens–including species hyicus and aureus from the genus Staphylococcus . This suggested that its use as a probiotic could serve as a robust preventive method in the swine industry. Here we determined the complete sequence of the S. suis 90–1330 genome to investigate its suitability as a probiotic. However, our results suggest that the use of this strain for prophylactic purposes may not yield the expected benefits and that despite is apparent lack of virulence [4,18], S. suis 90–1330 may not be entirely harmless.
A mobile bacteriocin locus?
One major argument against the use of an unmodified S. suis 90–1330 strain as a probiotic is the fact that the bacteriocin locus was found encoded within what appears to be an integrative and conjugative element (ICE) shared between different streptococcal species. Mobile genetic elements are common in bacterial genomes and are drivers of genetic diversity between and within lineages . Although derelict/incomplete ICEs are no longer capable of mobility, complete ones can propagate to other loci upon the proper stimuli. The S. suis 90–1330 bacteriocin locus contained in the putative ICE codes not only for the suicin itself but also provides protection against its effects [4,62]. One can easily envision a scenario wherein a mobile bacteriocin-containing ICE could integrate their genetic material in other microorganisms, including the very pathogens whose growth we are trying to inhibit, under sublethal bacteriocin concentrations. This would render the bacteriocin ineffective against them, and while it is not known exactly how prevalent or efficient this process would be, over time there is a non-negligible probability that an active ICE could integrate itself into the genomes of the very organisms that are we are trying to get rid of. Given that a near identical bacteriocin locus was found in a mobile ICE from S. suis strain CZ1303002 , a potent pathogen shown to cause severe meningitis , this scenario likely has occurred in the not so distant past. Furthermore, because Pan and colleagues  successfully transconjugated the S. suis CZ1303002 ICE from suicin+ to suicin- species, it appears that suicin- species can indeed survive long enough to acquire the proper resistance mechanisms under the right conditions.
Here, although the mobility of the S. suis 90–1330 suicin-containing putative ICE has not been tested in vitro, the combined presence of the ICESa2603-like direct repeats, of a putative origin of transfer oriT, of an integrase together with the conjugative DNA processing and Type IV secretion system proteins required for the ICE propagation to ectopic sites and recipient cells  (Fig 2), together with the detected presence of both integrated and excised forms of this element in S. suis 90–1330 (S2 Fig), strongly suggests that this element is genuine. Clearly, more experiments are required to confirm that this element is mobile. However, considering that the suicin-containing locus from the S. suis 90–1330 putative ICE also has been found in the mobile ICESsuCZ130302 from S. suis CZ1303002  and in several other bacterial species also harboring ICE-related genes (Fig 3), it would be prudent to consider the S. suis 90–1330 ICE as capable of mobility unless demonstrated otherwise.
But even if the bacteriocin-containing locus in S. suis 90–1330 is no longer mobile, streptococcal species are naturally competent , and this strain is no exception. All the genes involved in the ComRS competence system and in the methylation of foreign DNA acquired by competence were found in the S. suis genome, which provides this strain with the opportunity to acquire genetic material from bacteria that it kills with its bacteriocin, and one could envision that by killing pathogenic strains, S. suis 90–1330 may acquire/reacquire virulence genes. Considering all the above, engineering the S. suis 90–1330 strain to ensure that the locus is no longer mobile and to knockout the streptococcal competence system prior to its use as a prophylactic appears an essential step to minimize undesirable outcomes.
Would using S. suis 90–1330 as a prophylactic provide tangible benefits to the host?
Intuitively, effectiveness and the absence of side effects are desirable traits of any treatment. One of the major draws of using the S. suis 90–1330 bacteriocin as a substitute for traditional antibiotics resides in its relatively narrow spectrum, which reduces the risk of disturbing the healthy microflora . However, the wide distribution of the S. suis 90–1330 bacteriocin cluster across several streptococcal strains and species (Figs 3 and 4) raises doubt about the overall effectiveness of using this strain as a prophylactic. Because the bacteriocin locus also confers protection against its effect, many strains are already protected against this bacteriocin, including several pathogens. But even for other streptococcal pathogens lacking this locus, because they are naturally competent, merely knocking out mobility of the S. suis 90–1330 bacteriocin-containing locus may not be sufficient to prevent its propagation to other organisms. Any genetic material left in the environment by S. suis 90–1330 could be integrated by pathogens and, therefore, decoupling the bacteriocin from its protectin would help minimize the risks of propagation by preventing the simultaneous acquisition of both genes. Clearly, knocking out the protectin gene from the S. suis 90–1330 genome would render this strain susceptible to its effect, and is therefore not desirable. Considering these issues, administering the bacteriocin alone via other distribution mechanisms may prove more effective than using S. suis 90–1330 as a probiotic.
Is S. suis 90–1330 truly harmless?
Using S. suis 90–1330 and the bacteriocin it produces for prophylactic purposes appeared a very interesting prospect given its apparent lack of virulence . However, our results raise doubts as to whether this strain is indeed devoid of toxicity. While we did not identify the suilysin typically associated with S. suis virulence, we identified several genes whose products are involved in blood survival. Indeed, the presence of genes coding for hemolysin III, for a TlyC-like hemolysin transporter and for platelet-binding protein FbpA all support the ability of this strain to survive if not thrive in the bloodstream. Complementation studies involving TlyC turned a nonhemolytic bacteria into an hemolytic one  and FbpA is considered an important factor in infective endocarditis, with the S. suis FbpA featuring both the N-terminal fibrinogen binding domain and the C-terminal DUF814 domain found in Enterococcus faecalis homologs shown to bind to immobilized fibronectin . While previous infection experiments performed on mice and pigs with S. suis 90–1330 did not lead to infection , one of the three intravenous replicates performed on pigs led to a transient lameness suggesting that this strain might not be entirely benign. The presence of several CPS genes in this serotypable strain also suggests that it may be able to avoid phagocytosis to some extent, and the conditions performed to test the virulence of S. suis 90–1330 cannot, by their nature, reproduce all the possible conditions found in the environment. As discussed recently by Segura et al. , S. suis infections usually originate from the respiratory tract and the intravenous method sidesteps potentially critical early infection stages of this pathogen. But even if S. suis 90–1330 is indeed avirulent, synergies between different strains have been observed in blood infections . Those infections included other supposedly low virulence SS2-ST28 strains, such that under the right conditions, it would be reasonable to expect virulence from the 90–1330 strain.
The intent for this study was to look for the suitability of using S. suis 90–1330 as a prophylactic strain in the swine industry. However, based on our results, we cannot recommend its use as is without further engineering given the apparent mobility of its lantibiotic bacteriocin locus cargoed in what is likely an integrative and conjugative element. But even if this element is no longer mobile, given the wide distribution of this bacteriocin and its resistance mechanism across several streptococcal species, the use of this S. suis isolate as a probiotic may provide only limited protection against virulent strains. Furthermore, based on its genetic paraphernalia, the S. suis 90–1330 strain may not be as benign as previously thought and further testing and/or engineering would be required to ensure the safety of the animals subjected to potential prophylactic treatments involving this strain. Considering all the above, administering the bacteriocin directly as food preservative/supplement using alternate mechanisms that are not reliant on probiotics may prove a better approach to minimize undesirable outcomes, and further work will be required to determine if using this bacteriocin in such a fashion would indeed be effective.
S1 Fig. Physical map of Streptococcus suis 90–1330.
(A) S. suis 90–1330 chromosome. Genes located on the forward strand are indicated by orange boxes whereas those located on the minus strand are highlighted in yellow. Methylation patterns and GC-skew distribution are indicated in the outer and inner rings, respectively. GC-skew patterns (positive and negative values are shown as red and blue, respectively). The putative origin of replication inferred by the GC-skew analyses and the location of dnaA is indicated by an arrow. MGE; bacteriocin-containing mobile genetic element. The sixteen inner rings surrounded by the GC skew plot highlight genes of known function color-coded per KEGG pathways. From outside to inside: amino acid metabolism (orange; 126 genes), carbohydrate metabolism (blue; 246 genes), cellular processes (light green; 14 genes), metabolism of cofactors and vitamins (pink; 57 genes), energy metabolism (violet; 68 genes), environmental information processing (yellow; 151 genes), genetic information processing (pale red; 181 genes), glycan biosynthesis and metabolism (blue; 24 genes), human diseases (green; 50 genes), lipid metabolism (dark cyan, 42 genes), nucleotide metabolism (red; 99 genes), organismal systems (green; 15 genes), metabolism of other amino acids (orange; 26 genes), biosynthesis of other secondary metabolites (pink; 23 genes), metabolism of terpenoids and polyketides (lime green; 24 genes) and xenobiotics biodegradation and metabolism (brown; 22 genes). (B) S. suis 90–1330 plasmid.
S2 Fig. Detection of the circular form of the S. suis 90–1330 ICE.
(A) Diagram of the chromosomic region contiguous to the ICE integration site. The ICESa2603-like direct repeats are represented by yellow boxes. The orientation of the primers used to detect the integrated and circular forms are indicated by thin arrows (see Material and Methods for the primer sequences): 1) AN924_RS09900 –hypothetical protein; 2) AN924_RS09905 –integrase; 3) AN924_RS10235 –replication initiator protein; 4) AN924_RS10235 –ribosomal protein L7/L12. (B) Electrophoretic analysis of the PCR products (0.8% agarose gel); primer pairs are shown above the lanes.
S3 Fig. Structural comparison between the S. suis 90–1330 putative ICE and ICESsuCZ130302 from S. suis CZ130302.
Conserved blocks of genes arrayed in the same order between the two loci are highlighted by alternating colors. Locus tags are derived from accession numbers NZ_CP012731.1 (S. suis 90–1330) and NZ_CP024974.1 (S. suis CZ130302).
S1 File. Putative prophages in the S. suis 90–1330 genomes.
This file is the raw output from PHASTER.
S1 Table. S. suis strains used in this study.
Predicted serotypes are based on BLASTN searches using the cps genes from Liu et al.  against the S. suis accession numbers.
S2 Table. Predicted transposable elements in S. suis 90–1330.
Transposable elements listed in S1 Table were derived from the NCBI PGAP annotations.
S3 Table. S. suis 90–1330 repeat regions.
Repeated loci were identified with RepeatFinder as implemented in Geneious R9.1.7.
S4 Table. Distribution of the S. suis 90–1330 KEGG metabolic pathways.
KEGG metabolic pathways and enzyme commission (EC) numbers were assigned with BlastKOALA.
S5 Table. ComRS competence components in S. suis 90–1330.
ComRS proteins and promoters were identified by BLAST searches of known orthologous streptococcal sequences against the S. suis 90–1330 genome.
S6 Table. ICE proteins identified in the S. suis 90–1330 putative ICE by CONJscan/TXSScan searches.
CONJscan/TXSScan searches were performed with MacSyFinder 1.05. Duplicate hits with lower E-values are grayed out.
S7 Table. Orthologs between the S. suis 90–1330 putative ICE and ICESsuCZ130302.
Orthologs were inferred by BLASTP searches (E-value cutoff: 1e-10). Genes from Lebel et al.  are highlighted in magenta (sslA; bacteriocin). ICE-related genes identified by TXSScan/CONJscan searches and by Tn5252 BLASTP homology searches are highlighted in blue.
S8 Table. Putative virulence factor homologs in S. suis 90–1330.
The virulence factors discussed in Segura et al.  were searched against the S. suis 90–1330 genome using BLAST searches. Genes potentially important in virulence are highlighted in orange; pseudogenes are grayed out.
S9 Table. Putative additional virulence factors in S. suis 90–1330 derived from homology searches against the core dataset of the Virulence Factor Database (VFDB).
S. suis proteins displaying homology with putative VFDB virulence factors (E-value cutoff: 1e-10) are listed here.
This work was supported by funds from the Illinois Institute of Technology to Jean-François Pombert. Iva A. Veseli gratefully acknowledges support by an Undergraduate Research Fellowship from the American Society for Microbiology. The authors would like to thank Anthony Vincent (Université Laval) for help with prophage analyses.
- 1. Yang S-C, Lin C-H, Sung CT, Fang J-Y. Antibacterial activities of bacteriocins: application in foods and pharmaceuticals. Front Microbiol. Frontiers Media SA; 2014;5: 241. pmid:24904554
- 2. Dobson A, Cotter PD, Ross RP, Hill C. Bacteriocin production: a probiotic trait? Appl Environ Microbiol. American Society for Microbiology; 2012;78: 1–6. pmid:22038602
- 3. Cotter PD, Hill C, Ross RP. Food Microbiology: Bacteriocins: developing innate immunity for food. Nat Rev Microbiol. 2005;3: 777–788. pmid:16205711
- 4. LeBel G, Vaillancourt K, Frenette M, Gottschalk M, Grenier D. Suicin 90–1330 from a nonvirulent strain of Streptococcus suis: a nisin-related lantibiotic active on gram-positive swine pathogens. Appl Environ Microbiol. 2014;80: 5484–5492. pmid:24973067
- 5. King SJ, Leigh JA, Heath PJ, Luque I, Tarradas C, Dowson CG, et al. Development of a multilocus sequence typing scheme for the pig pathogen Streptococcus suis: Identification of virulent clones and potential capsular serotype exchange. J Clin Microbiol. 2002;40: 3671–3680. pmid:12354864
- 6. Zimmerman JJ. Diseases of swine. Zimmerman JJ, Karrike LA, Alejandro R, Schwartz KJ, Stevenson GW, editors. Wiley-Blackwell; 2012.
- 7. Domínguez-Punaro MC, Segura M, Plante M-M, Lacouture S, Rivest S, Gottschalk M. Streptococcus suis serotype 2, an important swine and human pathogen, induces strong systemic and cerebral inflammatory responses in a mouse model of infection. J Immunol. 2007;179: 1842–54. pmid:17641051
- 8. Dupas D, Vignon M, Géraut C. Streptococcus suis meningitis. A severe noncompensated occupational disease. J Occup Med. 1992;34: 1102–5. pmid:1432301
- 9. Huong VTL, Long HB, Kinh N V., Ngan TTD, Dung VTV, Nadjm B, et al. Long-term outcomes of patients with Streptococcus suis infection in Viet Nam: A case-control study. J Infect. 2017;
- 10. Loetthong P, Anukul R, Tanburawong N, Samerchea S, Areeratana P, Lekhalula P, et al. Impact of a Food Safety Campaign on Streptococcus suis Infection in Humans in Thailand. Am J Trop Med Hyg. 2017;96: 1370–1377. pmid:28719258
- 11. Callejo R, Zheng H, Du P, Prieto M, Xu J, Zielinski G, et al. Streptococcus suis serotype 2 strains isolated in Argentina (South America) are different from those recovered in North America and present a higher risk for humans. JMM Case Reports. 2016;3: e005066. pmid:28348788
- 12. Mancini F, Adamo F, Creti R, Monaco M, Alfarone G, Pantosti A, et al. A fatal case of streptococcal toxic shock syndrome caused by Streptococcus suis carrying tet (40) and tet (O/W/32/O), Italy. J Infect Chemother. 2016;22: 774–776. pmid:27553071
- 13. Smith TC, Capuano AW, Boese B, Myers KP, Gray GC. Exposure to Streptococcus suis among US swine workers. Emerg Infect Dis. 2008;14: 1925–7. pmid:19046523
- 14. van de Beek D, Spanjaard L, de Gans J. Streptococcus suis meningitis in the Netherlands. J Infect. 2008;57: 158–161. pmid:18538852
- 15. Tramontana AR, Graham M, Sinickas V, Bak N. An Australian case of Streptococcus suis toxic shock syndrome associated with occupational exposure to animal carcasses. Med J Aust. 2008;188: 538–9. pmid:18459929
- 16. Watkins EJ, Brooksby P, Schweiger MS, Enright SM. Septicaemia in a pig-farm worker. Lancet. 2001;357: 38. pmid:11197360
- 17. Fittipaldi N, Xu J, Lacouture S, Tharavichitkul P, Osaki M, Sekizaki T, et al. Lineage and virulence of Streptococcus suis serotype 2 isolates from North America. Emerg Infect Dis. 2011;17: 2239–2244. pmid:22172538
- 18. Quessy S, Dubreuil JD, Caya M, Higgins R. Discrimination of virulent and avirulent Streptococcus suis capsular type 2 isolates from different geographical origins. Infect Immun. American Society for Microbiology (ASM); 1995;63: 1975–9. pmid:7729910
- 19. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10: 421. Artn 421\n pmid:20003500
- 20. Gordon D, Abajian C, Green P. Consed: a graphical tool for sequence finishing. Genome Res. 1998;8: 195–202. pmid:9521923
- 21. Lowe TM, Eddy SR. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 1997;25: 955–964. pmid:9023104
- 22. Lagesen K, Hallin P, Rødland EA, Staerfeldt H-H, Rognes T, Ussery DW. RNAmmer: consistent and rapid annotation of ribosomal RNA genes. Nucleic Acids Res. 2007;35: 3100–3108. pmid:17452365
- 23. Seemann T. Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30: 2068–9. pmid:24642063
- 24. Jones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014;30: 1236–1240. pmid:24451626
- 25. The Uniprot Consortium. Activities at the Universal Protein Resource (UniProt). Nucleic Acids Res. 2014;42: D191–D198. pmid:24253303
- 26. Talagas A, Fontaine L, Ledesma-Garca L, Mignolet J, Li de la Sierra-Gallay I, Lazar N, et al. Structural Insights into Streptococcal Competence Regulation by the Cell-to-Cell Communication System ComRS. PLoS Pathog. 2016;12: e1005980. pmid:27907189
- 27. Zaccaria E, Van Baarlen P, De Greeff A, Morrison DA, Smith H, Wells JM. Control of competence for DNA transformation in Streptococcus suis by genetically transferable pherotypes. PLoS One. 2014;9: e99394. pmid:24968201
- 28. Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. Circos: an information esthetic for comparative genomics. Genome Res. 2009;19: 1639–1645. pmid:19541911
- 29. Darling ACE, Mau B, Blattner FR, Perna NT. Mauve: Multiple Alignment of Conserved Genomic Sequence With Rearrangements. Genome Res. 2004;14: 1394–1403. pmid:15231754
- 30. Ondov BD, Treangen TJ, Mallonee AB, Bergman NH, Koren S, Phillippy AM. Fast genome and metagenome distance estimation using MinHash. Genome Biol. Genome Biology; 2016;17: 132. pmid:27323842
- 31. Van Der Maaten LJP, Hinton GE. Visualizing high-dimensional data using t-SNE. J Mach Learn Res. 2008;9: 2579–2605.
- 32. R Core Team. R: A Language and Environment for Statistical Computing. R Found Stat Comput. 2016;version 3: 3503.
- 33. van der Maaten L. Barnes-Hut-SNE. arxiv.org. 2013; 1301.3342v2.
- 34. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9: 357–359. pmid:22388286
- 35. Koboldt DC, Zhang Q, Larson DE, Shen D, McLellan MD, Lin L, et al. VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res. 2012;22: 568–576. pmid:22300766
- 36. Segura M, Fittipaldi N, Calzas C, Gottschalk M. Critical Streptococcus suis Virulence Factors: Are They All Really Critical? Trends Microbiol. Elsevier Ltd; 2017;25: 585–599.
- 37. Zankari E, Hasman H, Cosentino S, Vestergaard M, Rasmussen S, Lund O, et al. Identification of acquired antimicrobial resistance genes. J Antimicrob Chemother. Oxford University Press; 2012;67: 2640–2644. pmid:22782487
- 38. Arndt D, Grant JR, Marcu A, Sajed T, Pon A, Liang Y, et al. PHASTER: a better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016;44: W16–21. pmid:27141966
- 39. Huang J, Ma J, Shang K, Hu X, Liang Y, Li D, et al. Evolution and Diversity of the Antimicrobial Resistance Associated Mobilome in Streptococcus suis: A Probable Mobile Genetic Elements Reservoir for Other Streptococci. Front Cell Infect Microbiol. 2016;6: 1–14.
- 40. Kiliç AO, Vijayakumar MN, Al-Khaldi SF. Identification and nucleotide sequence analysis of a transfer-related region in the streptococcal conjugative transposon Tn5252. J Bacteriol. 1994;176: 5145–50. Available: http://www.ncbi.nlm.nih.gov/pubmed/8051031 pmid:8051031
- 41. Alarcon-Chaidez F, Sampath J, Srinivas P, Vijayakumar M. TN5252: a model for complex streptococcal conjugative transposons. Adv Exp Med Biol. 1997;418: 1029–32. Available: http://www.ncbi.nlm.nih.gov/pubmed/9331826 pmid:9331826
- 42. Guglielmini J, Néron B, Abby SS, Garcillán-Barcia MP, la Cruz F de, Rocha EPC. Key components of the eight classes of type IV secretion systems involved in bacterial conjugation or protein secretion. Nucleic Acids Res. 2014;42: 5715–5727. pmid:24623814
- 43. Abby SS, Cury J, Guglielmini J, Néron B, Touchon M, Rocha EPC. Identification of protein secretion systems in bacterial genomes. Sci Rep. Nature Publishing Group; 2016;6: 23080. pmid:26979785
- 44. Rajewska M, Wegrzyn K, Konieczny I. AT-rich region and repeated sequences—the essential elements of replication origins of bacterial replicons. FEMS Microbiol Rev. 2012;36: 408–434. pmid:22092310
- 45. Achigar R, Magadán AH, Tremblay DM, Julia Pianzzola M, Moineau S. Phage-host interactions in Streptococcus thermophilus: Genome analysis of phages isolated in Uruguay and ectopic spacer acquisition in CRISPR array. Sci Rep. Nature Publishing Group; 2017;7: 43438. pmid:28262818
- 46. De Ste Croix M, Vacca I, Kwun MJ, Ralph JD, Bentley SD, Haigh R, et al. Phase-variable methylation and epigenetic regulation by type I restriction-modification systems. FEMS Microbiol Rev. 2017;41: S3–S15. pmid:28830092
- 47. Willemse N, Schultsz C. Distribution of Type I Restriction–Modification Systems in Streptococcus suis: An Outlook. Pathogens. 2016;5: 62.
- 48. Okura M, Nozawa T, Watanabe T, Murase K, Nakagawa I, Takamatsu D, et al. A locus encoding variable defense systems against invading DNA identified in Streptococcus suis. Genome Biol Evol. 2017;9: 1000–1012.
- 49. Goessweiner-mohr N, Arends K, Keller W, Grohmann E. Conjugation in Gram-Positive Bacteria. Microbiol Specturm. 2014;2: PLAS-0004-2013.
- 50. Alvarez-Martinez CE, Christie PJ. Biological Diversity of Prokaryotic Type IV Secretion Systems. Microbiol Mol Biol Rev. 2009;73: 775–808. pmid:19946141
- 51. Huang J, Liang Y, Guo D, Shang K, Ge L, Kashif J, et al. Comparative genomic analysis of the ICESa2603 family ICEs and spread of erm(B)- and tet(O)-carrying transferable 89K-subtype ICEs in swine and bovine isolates in China. Front Microbiol. 2016;7: 1–13.
- 52. Pan Z, Liu J, Zhang Y, Chen S, Ma J, Dong W, et al. A novel integrative conjugative element mediates transfer of multi-drug resistance between Streptococcus suis strains of different serotypes. Vet Microbiol. 2019;229: 110–116. pmid:30642585
- 53. Jacobs AA, Loeffen PL, van den Berg AJ, Storm PK. Identification, purification, and characterization of a thiol-activated hemolysin (suilysin) of Streptococcus suis. Infect Immun. 1994;62: 1742–8. pmid:8168935
- 54. Smith HE, Vecht U, Gielkens AL, Smits MA. Cloning and nucleotide sequence of the gene encoding the 136-kilodalton surface protein (muramidase-released protein) of Streptococcus suis type 2. Infect Immun. American Society for Microbiology (ASM); 1992;60: 2361–7. pmid:1587602
- 55. Wang J, Kong D, Zhang S, Jiang H, Zheng Y, Zang Y, et al. Interaction of fibrinogen and muramidase-released protein promotes the development of Streptococcus suis meningitis. Front Microbiol. 2015;6: 1001. pmid:26441928
- 56. Athey TBT, Auger J-P, Teatero S, Dumesnil A, Takamatsu D, Wasserscheid J, et al. Complex population structure and virulence differences among serotype 2 Streptococcus suis strains belonging to sequence type 28. PLoS One. 2015;10: e0137760. pmid:26375680
- 57. Athey TBT, Teatero S, Takamatsu D, Wasserscheid J, Dewar K, Gottschalk M, et al. Population structure and antimicrobial resistance profiles of Streptococcus suis serotype 2 sequence type 25 strains. PLoS One. 2016;11: e0150908. pmid:26954687
- 58. Palmieri C, Magi G, Mingoia M, Bagnarelli P, Ripa S, Varaldo PE, et al. Characterization of a Streptococcus suis tet(O/W/32/O)-carrying element transferable to major streptococcal pathogens. Antimicrob Agents Chemother. 2012;56: 4697–702. pmid:22710115
- 59. Hall AM, Gollan B, Helaine S. Toxin–antitoxin systems: reversible toxicity. Curr Opin Microbiol. 2017;36: 102–110. pmid:28279904
- 60. Rajesh T, Anthony T, Saranya S, Pushpam PL, Gunasekaran P. Functional characterization of a new holin-like antibacterial protein coding gene tmp1 from goat skin surface metagenome. Appl Microbiol Biotechnol. 2011;89: 1061–1073. pmid:20927512
- 61. Johnson CM, Grossman AD. Integrative and Conjugative Elements (ICEs): What They Do and How They Work. Annu Rev Genet. 2015;49: 577–601. Integrative pmid:26473380
- 62. Bierbaum G, Sahl H-G. Lantibiotics: Mode of Action, Biosynthesis and Bioengineering. Curr Pharm Biotechnol. 2009;10: 2–18. pmid:19149587
- 63. Pan Z, Ma J, Dong W, Song W, Wang K, Lu C, et al. Novel variant serotype of Streptococcus suis isolated from piglets with meningitis. Appl Environ Microbiol. American Society for Microbiology; 2015;81: 976–85. pmid:25416757
- 64. Tohya M, Watanabe T, Maruyama F, Arai S, Ota A, Athey TBT, et al. Comparative genome analyses of Streptococcus suis isolates from endocarditis demonstrate persistence of dual phenotypic clones. PLoS One. 2016;11: 1–16.
- 65. Radulovic S, Troyer JM, Beier MS, Lau AOT, Azad AF. Identification and Molecular Analysis of the Gene Encoding Rickettsia typhi Hemolysin. Infect Immun. American Society for Microbiology Journals; 1999;67: 6104–6108. Available: https://iai.asm.org/content/67/11/6104.long
- 66. Singh K V., Rosa SL La, Somarajan SR, Roh JH, Murray BE. The fibronectin-binding protein EfbA contributes to pathogenesis and protects against infective endocarditis caused by Enterococcus faecalis. Infect Immun. 2015;83: 4487–4494. pmid:26351286
- 67. Liu Z, Zheng H, Gottschalk M, Bai X, Lan R, Ji S, et al. Development of Multiplex PCR Assays for the Identification of the 33 Serotypes of Streptococcus suis. PLoS One. 2013;8: 1–11.