Staphylococcus brunensis sp. nov. isolated from human clinical specimens with a staphylococcal cassette chromosome-related genomic island outside of the rlmH gene bearing the ccrDE recombinase gene complex

ABSTRACT Novel species of coagulase-negative staphylococci, which could serve as reservoirs of virulence and antimicrobial resistance factors for opportunistic pathogens from the genus Staphylococcus, are recognized in human and animal specimens due to advances in diagnostic techniques. Here, we used whole-genome sequencing, extensive biotyping, MALDI-TOF mass spectrometry, and chemotaxonomy to characterize five coagulase-negative strains from the Staphylococcus haemolyticus phylogenetic clade obtained from human ear swabs, wounds, and bile. Based on the results of polyphasic taxonomy, we propose the species Staphylococcus brunensis sp. nov. (type strain NRL/St 16/872T = CCM 9024T = LMG 31872T = DSM 111349T). The genomic analysis revealed numerous variable genomic elements, including staphylococcal cassette chromosome (SCC), prophages, plasmids, and a unique 18.8 kb-long genomic island SbCI ccrDE integrated into the ribosomal protein L7 serine acetyltransferase gene rimL. SbCI ccrDE has a cassette chromosome recombinase (ccr) gene complex with a typical structure found in SCCs. Based on nucleotide and amino acid identity to other known ccr genes and the distinct integration site that differs from the canonical methyltransferase gene rlmH exploited by SCCs, we classified the ccr genes as novel variants, ccrDE. The comparative genomic analysis of SbCI ccrDE with related islands shows that they can accumulate virulence and antimicrobial resistance factors creating novel resistance elements, which reflects the evolution of SCC. The spread of these resistance islands into established pathogens such as Staphylococcus aureus would pose a great threat to the healthcare system. IMPORTANCE The coagulase-negative staphylococci are important opportunistic human pathogens, which cause bloodstream and foreign body infections, mainly in immunocompromised patients. The mobile elements, primarily the staphylococcal cassette chromosome mec, which confers resistance to methicillin, are the key to the successful dissemination of staphylococci into healthcare and community settings. Here, we present a novel species of the Staphylococcus genus isolated from human clinical material. The detailed analysis of its genome revealed a previously undescribed genomic island, which is closely related to the staphylococcal cassette chromosome and has the potential to accumulate and spread virulence and resistance determinants. The island harbors a set of conserved genes required for its mobilization, which we recognized as novel cassette chromosome recombinase genes ccrDE. Similar islands were revealed not only in the genomes of coagulase-negative staphylococci but also in S. aureus. The comparative genomic study contributes substantially to the understanding of the evolution and pathogenesis of staphylococci.


IMPORTANCE
The coagulase-negative staphylococci are important opportunistic human pathogens, which cause bloodstream and foreign body infections, mainly in immunocompromised patients.The mobile elements, primarily the staphylococcal cassette chromosome mec, which confers resistance to methicillin, are the key to the successful dissemination of staphylococci into healthcare and community settings.Here, we present a novel species of the Staphylococcus genus isolated from human clinical material.The detailed analysis of its genome revealed a previously undescribed genomic island, which is closely related to the staphylococcal cassette chromosome and has the potential to accumulate and spread virulence and resistance determinants.The island harbors a set of conserved genes required for its mobilization, which we recognized as novel cassette chromosome recombinase genes ccrDE.Similar islands were revealed not only in the genomes of coagulase-negative staphylococci but also in S. aureus.The comparative genomic study contributes substantially to the understanding of the evolution and pathogenesis of staphylococci.

Phylogenetic relationship of the strains
Five unidentified Staphylococcus sp.strains were collected from various human clinical specimens from both mixed culture and monoculture between 2016 and 2022 (Table 1) and transferred to the National Reference Laboratory for Staphylococci (National Institute of Public Health, Prague) for long-term storage and further study.The phyloge netic analysis of complete 16S rRNA gene sequences consistently placed the five isolates in S. haemolyticus cluster group defined previously (22).The closest relatives were Staphylococcus petrasii, Staphylococcus croceilyticus, and Staphylococcus pragensis, with 16S rRNA gene sequence similarities ranging from 99.80 to 99.59%, while other species were below 99.4% similarity.The topology of the neighbor-joining (NJ) phylogenetic tree constructed with 16S rRNA gene sequences was similar to that of the maximum likelihood (ML) tree (Fig. 1A).Because the 16S rRNA analysis has limited discriminatory power in the genus Staphylococcus (5,6,23), the phylogenetic position of the new isolates was also assessed using the concatenated multilocus sequence data of six routinely used conserved housekeeping genes: rpoB, hsp60, dnaJ, tufA, gap, and sodA for discrimination of staphylococcal species (Fig. 1B).The ML phylogenetic tree for the housekeeping genes had a very similar topology to that of the 16S rRNA gene tree and to the additional phylogenetic trees constructed using the up-to-date bacterial core gene (UBCG) at the nucleotide and protein level (Fig. 1C and D).
The whole-genome phylogenetic distance from the related staphylococcal taxa with an ANI value of <92.3% (Table S3) and digital DNA-DNA hybridization (dDDH) of <45.6% determined the closest species S. petrasii; however, the values were below the species delineation thresholds, which are 95-96% and 70%, respectively (24).The whole-genome phylogeny thus confirmed that the five isolates represent a distinct Staphylococcus species designated Staphylococcus brunensis sp.nov.

Growth, morphological, biochemical, and chemotaxonomical characteriza tion of analyzed isolates
All five isolates exhibited Gram-positive stain, irregular cells ranging in diameter from 433 nm to 1,210 nm (Fig. S1) arranged in pairs, tetrads, and clusters.They grew very well on   S1.
(A) Evolutionary history inferred using neighbor-joining (NJ) method based on complete 16S rRNA gene sequences extracted from whole-genome assemblies.
Filled circles indicate that the corresponding nodes were also identified from analysis by the maximum likelihood (ML) method.The percentage of replicate trees above 50% in which the associated taxa clustered together in the bootstrap test (500 replicates) are shown next to the branches for the NJ and ML methods.S.
aureus DSM 20231 T was used as an outgroup.The evolutionary distances are in the units of the number of base substitutions per site.All ambiguous positions were removed for each sequence pair.There were a total of 1,552 positions in the final data set.(B) Unrooted ML tree based on multilocus sequence analysis of concatenated nucleotide sequences from six loci-rpoB, hsp60, dnaJ, tufA, gap, and sodA-were extracted from whole-genome assemblies.Filled circles indicate that the corresponding nodes were also identified from analysis by the NJ method.There were a total of 3,952 positions in the final data set.Bootstrap probability values (percentages of 500 tree replications) greater than 50% are shown at branch points.The evolutionary distances are given as the number of substitutions per site.(C) Nucleotide sequence-based and (D) protein sequence-based phylogenetic tree of the concatenated alignment of 92 core genes constructed using up-to-date bacterial core gene (UBCG) set.The ML tree was inferred using RAxML software and 100 replicates.The threshold for the gene support index was set to 94% for the nucleotide and 80% for the protein-based tree.Gene support indices are given at branching points (the maximal possible value is 92).The bar indicates the number of substitutions per position.
conventional VP test.Similar results were obtained with the β-glucuronidase test, where all isolates were positive in the STAPHYtest 24 kit, but negative in the API ZYM kit due to a different substrate for enzyme detection.The differentiation of novel staphylococcal isolates from similar and/or closely related staphylococci occurring in human clinical material is shown in Table 2.The species S. petrasii, S. pragensis, and S. haemolyticus were phenotypically the most similar taxa to the aforementioned isolates.Strain-dependent utilization results are specified in Table S4.Complete characteristics of S. brunensis sp.nov.are stated in the protologue given subsequently.Antibiotic susceptibility testing showed that all five strains are susceptible to cefoxitin, clindamycin, gentamicin, chloramphenicol, linezolid, oxacillin, rifampicin, tobramycin, trimethoprim, sulphamethox/trimethoprim (cotrimoxazole), tetracycline, and fusidic acid.Susceptibility to ciprofloxacin and levofloxacin was intermediate.Susceptibility to ampicillin, penicillin G, tigecycline, and erythromycin was strain-dependent (Table S4).
By using cluster analysis of matrix-assisted laser-desorption/ionization-time-of-flight mass spectrometry (MALDI-TOF MS) protein profiles, all five S. brunensis sp.nov.strains were separated into a coherent cluster distinguished from phylogenetically related   (10).

DNA fingerprinting of S. brunensis sp. nov.
Screening of the investigated bacterial group by rep-PCR fingerprinting with primer (GTG) 5 showed their genotypic coherence.All five isolates had visually similar fingerprints, which were grouped into a single cluster and separated from the other entries in the in-house fingerprint database, which includes members of all recognized species of the genus Staphylococcus, including type strains of phylogenetically closely related species (Fig. 3A).The obtained results also suggest that this rapid and easy-to-perform method can be used for the identification of S. brunensis sp.nov., similarly to how we demonstrated for the identification of Staphylococcus spp. in our previous studies (8,23,25).
In contrast to the rep-PCR method, automated ribotyping with the restriction enzyme EcoRI revealed heterogeneity among the five examined strains (Fig. 3B).Three isolates, NRL/St 16/872 T , NRL/St 22/194, and NRL/St 18/288, exhibited unique fingerprint patterns that allowed differentiation at the strain level and indicated that they were not clonally related.The remaining two strains, NRL/St 19/737 and NRL/St 21/187, had visually identical ribotypes, despite being isolated in 2019 and 2021, respectively, at different locations and were therefore not considered clonally related.These results suggest that automated ribotyping with EcoRI can separate isolates of S. brunensis sp.nov. at the strain level, although some strains may have similar ribotype patterns.However, a reliable assessment of the discriminatory power of this technique for typing S. brunensis sp.nov.requires the analysis of a more significant number of strains from different localities and sources.

Genome characterization of S. brunensis sp. nov.
The comparison of sequenced genomes (Table S5) revealed a high degree of similarity between the isolates (Fig. 4).The S. brunensis sp.nov.genomes were 2.5-2.6 Mb long with GC content 33.3-33.4% encoding 2,500-2,700 CDS, 61-62 tRNAs, and 19 rRNAs.The pangenome consists of 2,230 core and 416 accessory, and 569 unique genes in total.The sequenced genomes differ in variable genomic elements, which constitute 5-8% of the genome and are associated with virulence and antimicrobial resistance genes, predominantly located at plasmids.All S. brunensis sp.nov.genomes possess a type IIU CRISPR-Cas complex with 15-20 variable spacers, some of which target siphoviral prophages as determined by blastn search.The core gene kat encoding catalase in the strain NRL/St 19/737 (locus tag MT339_07575) has a 1 bp deletion in homopolymer polyA, which introduces a premature stop codon at the 3' end of the gene.It is possible that the catalase-negative phenotype of the strain is associated with the loss of function of this gene, although a truncated protein without the first 22 amino acids could be produced.
The S. brunensis sp.nov.genomes harbor several large plasmids, which can be grouped by the presence of either rep39 or rep20 gene.Due to high sequence varia bility and mosaicism caused by interspersed IS431, it was not possible to determine a complete consensus sequence for all plasmids.The plasmid contigs comprise betalactam, heavy metal, and disinfectant resistance as well as virulence and saccharide utilization genes as shown in Table 3.All analyzed strains harbor a small rep21 plasmid conferring cadmium resistance.The strain NRL/St 22/194 contains an additional rep10 plasmid harboring the ermC gene similar to the pUSA05-1 plasmid from S. aureus (26).These two small plasmids are integrated into the rep39-type plasmid contig of strain NRL/St 19/737.Furthermore, two cryptic plasmids, one containing the rep13 gene and the other containing rep5b gene were identified (Table 3).
The resistance to penicillin in S. brunensis sp.nov.NRL/St 18/288 is encoded by the blaZ gene located at a 9 kb Tn554-like transposon (Fig. 4) integrated into the chromoso mal isaB gene.The chromosomes of S. brunensis sp.nov.harbor several copies of various full and partial IS elements from the IS3, IS6, IS30, IS1182, IS200/IS605, and ISha1 families (Fig. 4).The IS431 flanked composite transposon localized downstream of the cspC gene contains the gene for a short-chain dehydrogenase associated with survival in stress conditions (27).Each strain carries six to eight ISSha1 copies integrated in six conserved and six variable loci, usually adjacent to rRNA genes.One copy of ISSha1 is inserted in the SCC element of S. brunensis sp.nov.NRL/St 21/287 strain (Fig. 4).
The S. brunensis sp.nov.strains have one or two complete prophages in their genomes integrated into two different 18 bp att sites.The ΦSBR-1 prophage of NRL/St 16/872 T is integrated in the att site AATCCCTTACTTCCCGTT, located in the tRNA-Ser(gga) gene.The other strains encompass one or two CRISPR spacers homologous to the ΦSBR-1, and thus are immune to infection by this phage.The same att site is used by ΦSBR-4 in the genome of NRL/St 19/737.Both prophages are 45.4 kb long and share 79.3% nt identity.The next att site, AATCCCTCCGTTTCCGTT in tRNA-Ser(gct), is occupied by either ΦSBR-2 or ΦSBR-3.Prophage ΦSBR-2, which is 47 kb long, is integrated into the genomes of strains NRL/St 16/872 T , NRL/St 19/288, and NRL/St 21/287.The strain NRL/St 19/737 harbors a 43.5 kb ΦSBR-3 prophage, which shares 79% nt identity with ΦSBR-2 genome along 57% of its length; the difference is mainly in the morphogenesis module.Both ΦSBR-2 and ΦSBR-3 carry a putative accessory virulence factor sialic acid transporter (28) 30), sharing 95.2% nt identity along 75% of their length.The SCC CCM9024 carries genes for the type I restriction-modification (RM) system, sharing high homology with genes encoding a restriction (hsdR) and methylation (hsdM) subunit with the RM system of S. hominis C34847 and S. aureus NTUH-4729 (31).Interestingly, the putP gene, which encodes a sodium proline symporter, was wedged between the ccrB2 gene and a set of three short hypothetical genes conserved in the ccr gene complex, thus disrupting the canonical organization of the complete ccr gene complex.Similarly, the putP gene and the flanking sequence from the ccr gene complex were found in SCC elements of various CoNS, S. epidermidis I1PPP121 (GenBank accession no.MH188479), S. haemolyticus BC05211 (KX181861), S. hominis J6 (LT963442), and S. hominis J11 (LT963438), and in plasmids of S. warneri 16A (CP031267) and S. pasteuri 3C (CP031281), pointing to a high degree of recombination occurring in staphylococci.

Novel genomic island harboring cassette chromosome recombinase genes ccrDE
The strains NRL/St 16/872 T , NRL/St 21/187 and NRL/St 22/194 harbor a mobile element of size 18.8 kb with cassette chromosome recombinase genes, integrated in the rimJ/rimL gene orthologous to the the ribosomal-protein-serine acetyltransferase gene rimL in Escherichia coli and ydaF in Bacillus subtilis (UniProt accession no.P13857 and P96579, respectively).The island designated S. brunensis chromosomal island ccrDE (SbCI ccrDE ) harbors another homologue of rimL with 70% nt identity to the original rimL; thus, it may complement the function of the truncated rimL gene.SbCI ccrDE is bordered by the tetranucleotide motif GAAA.Additionally, the 19 bp direct repeat ATTCCACAATGAAAT CCAT was found on the integration site in rimL and inside the rimL homologue on the island, which suggests a complex recombination event.
SbCI ccrDE possesses a cluster of genes homologous to the ccr gene complex from SCC elements.The core of the ccr complex consists of two ccr genes, similar to SCC elements type IV or II.Additional genes in the ccr complex, that is the putative primase polA, cassette chromosome helicase cch2, three short proteins with the domains SAUGI, DUF960, and DUF1643, and a hypothetical protein (Fig. 5), were homologous to those in SCCmec type V, which has only one ccrC gene in the ccr complex.SbCI ccrDE ccr genes share more than 98% DNA sequence identity to ccrA8B9 recombinases (Table S6) discovered recently in the S. haemolyticus genome (33).The values of nucleotide identity to currently known ccrA1-7, ccrB1-8, and ccrC1-2 genes from SCCs range from 38.0 to 53.3% (Table S6).Although it is slightly above the threshold of 50.0% for the definition of a gene variant according to the rules of IWG-SCC (29), the amino acid (aa) identity level of SbCI ccrDE recombinases to known CcrA, CcrB, and CcrC reaches a maximum of 39.9%, which is substantially lower than aa identity among Ccr allotypes, which range from 50.9 to 92.3%.Therefore, based on the borderline nucleotide identities and low protein identities to CcrA, CcrB, and CcrC allotypes, we propose designating the SbCI ccr recombinases as new allelic types ccrD and ccrE and reclassifying the described ccrA8B9 recombinases accordingly (33).The CcrDE recombinases are able to excise the SbCI ccrDE element from its insertion sequence site, which has been detected by sequence read alignment and PCR analysis (Fig. S2).
Apart from the ccr gene complex, the island SbCI ccrDE carries the 5-methylcytosinespecific restriction enzyme mcrB and specificity subunit mcrC genes for the putative RM system (Fig. 5), commonly found on SCC elements, chromosomal islands, and prophages that help to maintain the mobile element in the genome (34).Approximately 5.8 kbp of the SbCI ccrDE is occupied by a putative transposon bordered by a perfect 17 bp inverted repeat of TGGTTCTGTTGCAAAGT.The transposon has one copy of the insertion sequence from the IS6 family (sharing 95% nt identity to IS431 mec ) and accessory genes encoding putative membrane protein (yeiH), putative transcriptional regulator (cysL), and malate chinonine oxidoreductase (mqo) (Fig. 5), which are frequently found on plasmids.
We surveyed the GenBank database for sequences resembling SbCI ccrDE .In addi tion to S. haemolyticus BC5211 (33), we found related genomic islands with ccrDE in the genomes of S. haemolyticus (GenBank accession nos.CP033814 and CP102568), S. hominis subsp.hominis K1 (CP020618), and S. borealis GDY8P80P (35), and in S. aureus ER04332 and ER11327 (36), and related clones.The island is consistently inserted in the rimL gene, in a locus downstream of the conserved metE gene (Fig. 6A), which is 17-56 kb counterclockwise from the replication origin in CoNS species.However, in S. aureus strains, the rimL gene is located at the end of the oriC environ due to large-scale chromosomal inversion.The variable region of the islands comprises genes for an RM system (Fig. 6A).In S. borealis GDY8P80P, the island harbors many transpo sons with resistance genes to beta-lactam antibiotics, tetracycline, aminoglycosides, and other antimicrobials.The phylogenetic analysis of ccrD and ccrE genes from the related genomic islands (Fig. 6B) and the pairwise nucleotide identity comparison (Table S6) clearly distinguished two allotypes designated ccrD1E1, present in the genomes of S. brunensis sp.nov., S. haemolyticus, and S. aureus, and ccrD2E2, found in the genomes of S. borealis and S. hominis (Fig. 6B; Table S6).

Taxonomic description of S. brunensis sp. nov.
Staphylococcus brunensis (bru.nen′sisL. adj.brunensis from Bruna, the Roman name of the city of Brno, Czech Republic, where this and other staphylococcal species were first described).
The type strain is NRL/St 16/872 T (= CCM 9024 T = DSM 111349 T = LMG 31872 T ).The major respiratory quinone is menaquinone-7.The major fatty acids are C 15:0 anteiso and C 17:0 anteiso .The peptidoglycan type is A3α (A11.2).The DNA G+C content of strain NRL/St 16/872 T is 33.40 mol%, calculated from the whole-genomic sequence.The species description is based on the characterization of five strains isolated from various human clinical materials.Most of the characteristics of the type strain NRL/St 16/872 T agree with the species description.The GenBank/ENA/DDBJ accession number for the 16S rRNA gene is OQ401401.The complete chromosome sequence of the type strain is available under GenBank accession number CP119327.

DISCUSSION
The recent molecular diagnostic methods and polyphasic taxonomic approach, including whole-genome sequencing, allow for more effective species differentiation of CoNS from various sources.Here, we described S. brunensis sp.nov., which occupies similar niches to the closely related species from the S. petrasii phylogenetic complex (10).The available isolates of S. brunensis sp.nov.were associated with the human ear and wound infections, whereas S. petrasii and S. pragensis predominated in wounds, blood, or urinary tract infections (Table S7).However, it is challenging for clinicians to determine whether these CoNS are causative agents of human diseases or suspected contaminants associated with the occurrence of these commensal bacteria on human skin.
The pathogenic potential of CoNS is related to immune evasion, invasion of host tissues, and biofilm formation, allowing them to persist on the surfaces of indwelling medical devices and thus cause chronic infections (1).The intercellular adhesion operon (ica), biofilm-associated protein (bap/bhp), and fibronectin-binding protein genes (fnbA/ fbe) directly associated with biofilm production in S. epidermidis (37) have not been identified in either S. brunensis sp.nov.genomes or the other species from S. petrasii complex.Likewise, the ica operon homologue has not been detected in S. haemolyticus (3) and may be missing in this phylogenetic clade.Biofilm formation in the initial phase is influenced by molecules involved in surface adhesion.In staphylococci, these are microbial surface components recognizing adhesion matrix molecules (MSCRAMMs) (38).Homologs of cell wall-anchored serine-aspartate repeat-containing protein genes sdrC, sdrG, and sdrH have been identified in S. petrasii genomes (10) but not in S. brunensis sp.nov.However, gene homologues for elastin-binding protein (ebp), thermonuclease (nuc), autolysin E (atlE), and gene clusters (cap5 or cap8) involved in the synthesis of capsular polysaccharides have been found in genomes of both S. petrasii (10) and S. brunensis sp.nov.
A comparison of representative genome sequences revealed that the species S. brunensis sp.nov., S. croceilyticus, S. petrasii, and S. pragensis differ only marginally in genomic G+C content, genome size, and the median protein count encoded by the core genome.The major cause of interspecies differences in the S. petrasii complex is the accessory genome and variable genetic elements (Fig. S3).The role of these elements is best described in S. aureus, where HGT contributes to adaptation and evolution into successful lineages (39)(40)(41).
More than 90% of clinical staphylococcal isolates harbor plasmids of various size that can be classified to small multicopy plasmids or larger plasmids carrying several resistance determinants.However, only 5% (42) of staphylococcal plasmids are large multiresistance elements with the ability to mobilize or undergo conjugative transfer.Strains of S. brunensis sp.nov.harbor several large plasmids that encode genes for mobilization (mobA, mobC, mobP) and their spread by HGT is possible.The resistance genes are usually cointegrated between two copies of ISs that promote their spread (43).An example is IS431 previously described in plasmids pSK41 and pGO1 (44), harboring linezolid and high-level resistance to vancomycin (45).It is probable that IS431, identified in S. brunensis plasmid sequences, is responsible for the mosaic structure of the elements and also promotes the spread of resistance genes across the genus Staphylococcus.
The plasmid-borne resistance to antibiotics and disinfectants suggests enhanced survival in healthcare environments.The resistance to beta-lactam (blaZ) and macrolide (ermC) antibiotics in S. brunensis sp.nov.strains correlates with resistance genes on plasmids and the Tn554-like transposon.Resistance to quaternary ammonium com pounds (qacA), copper and heavy metals, and the hexulose utilization operon (hxl) in plasmids of S. brunensis sp.nov.suggests adaptation to various ecological niches similar to nosocomial and community-associated isolates of S. haemolyticus (46,47).
The genomes of S. brunensis sp.nov.contain more insertion sequence elements than the closest relatives from S. petrasii complex (7,8).The copy number variability of IS implies their recent capture and propagation and increases the genome plasticity among the strains.The most abundant ISSha1 elements with 98% similarity to ISSha1 from S. haemolyticus (3) are often located near rRNA genes and may contribute to variable ribotype patterns.
A significant MGE in the genome of S. brunensis sp.nov. is the genomic island SbCI ccrDE with unique recombinase genes ccrD1E1.Xiao et al. (33) reported the existence of these recombinases as ccrA8B9 allotypes.Nevertheless, their amino acid sequences that were highly different from other known ccrAB variants place them among novel ccrDE alleles.This reclassification also made it possible to distinguish another apparent variant ccrD2E2, present in S. hominis and S. borealis.The ccrDE genes in all inspec ted staphylococcal genomes were present on a genomic island related to SbCI ccrDE .Chromosomal islands with ccrDE (CI ccrDE s) share gene structure similar to SCCs, such as conserved ccr complex, presence of variable regions, and flanking by direct repeats.However, unlike SCCs, which are canonicaly inserted in the rlmH (orfX) gene in the oriC environ (48), CI ccrDE s are inserted in the rimL gene.The SCC-like islands with Ccr or closely related large serine recombinase were inserted in the rlmH in other genera of Gram-positive bacteria, that is the genera Mammaliicoccus and Macrococcus, Enterococ cus faecium, Bacillus cereus, Geobacillus vulcani, and Clostridioides difficile (49)(50)(51)(52).Hence, CI ccrDE s might represent a new class of SCC elements integrated into the rimL gene due to the altered specificity of the ccrDE recombinases.
The molecular mechanism of the action of ccrAB recombinases requires specific sequences of around 60-70 bp and the presence of the central dinucleotide GA in the integration site, which are essential for the integration of the cassettes into the highly conserved rlmH gene (20).However, the ccrAB recombinases also act on non-canonical less conserved recombination sites (20).Xiao et al. (33) showed that the ccrDE recombi nases are functional in excision and transfer of an SCC from rlmH, but they do not move together with the cassette.Here, we proved that the CI ccrDE bearing the ccrDE genes is excised from the chromosome.Since CI ccrDE has a similar size as pathogenicity islands and possesses a set of genes required for its replication (53), the element would perfectly fit into the phage capsid to be spread via transduction, which is a common transmission route of MGE in staphylococci (54,55).The high sequence conservation of the ccrDE gene complex in various species corresponds with their spread through horizontal gene transfer.
The proportion and diversity of CI ccrDE was low in the sequenced genomes deposited in the public databases, making it extremely difficult to identify the donor of ccrDE complex responsible for the formation of this element.Since we found ccrD1E1 and ccrD2E2 predominantly in CoNS genomes, it is possible that CoNS species are predis posed to forming this element and it can subsequently be transferred to S. aureus.The variable regions of CI ccrDE are highly diverse in gene content, suggesting that the CI ccrDE element evolved only recently in independent acquisition events, which are also common in the evolution of SCCmec (56) or staphylococcal pathogenicity islands (57).The benefit of CI ccrDE for the bacterial host might be only marginal, so there is no selective pressure to maintain established element and disseminate it to more strains.The only widespread genes identified in all CI ccrDE s that help to preserve the element are the RM system genes that differ in the CI ccrDE s, suggesting that they come from various sources similar to canonical SCC elements (30,31).
With the high selective pressure exerted on staphylococci by the use of antimicro bials, this element is a perfect candidate for the acquisition and spread of resistance and virulence genes.The SCCmec element underlying the successful spread of MRSA clones originated through joining of ccr gene complex with mec gene complex coming from multiple sources.The first signs of the accumulation of resistance genes in CI ccrDE were observed in S. borealis, previously misidentified as S. haemolyticus (35), where the island comprised several transposons.A similar cluster of drug-resistance genes to that in CI ccrDE of S. borealis was also reported in S. aureus on an SCC element (58) and a plasmid (59).We conclude that the gene structure of CI ccrDE indicates its ability to act as a primordial element to accumulate virulence and antimicrobial resistance factors.The spread of the island to established pathogens such as S. aureus would thus represent a new threat to the healthcare system.

Conclusion
The identification of the new species S. brunensis within the Staphylococcus genus expands our understanding of the diversity of coagulase-negative staphylococci.The number of the strains available is still limited, but similar to its closest relative S. petrasii, it can be expected that more strains will be captured as better diagnostic methods are developed.Genome analysis of the new isolates has important implications for studying the role of coagulase-negative staphylococci as a reservoir of transmissible genes that can facilitate improved survival in the environment, resistance to antibiotic treatment, or increased virulence following horizontal transfer.Characterization of a previously unexplored genomic island closely related to the SCC indicates the poten tial for its interspecies transfer enabled by unique ccrDE recombinase genes in both coagulase-negative staphylococci and the more clinically significant S. aureus.Identify ing the new type of MGE thus opens up new possibilities for future research of gene transfer mechanisms in the emergence of multidrug-resistant staphylococcal strains with implications for clinical practice.These findings deepen our understanding of the evolution and pathogenesis of staphylococci, shedding light on how these bacteria acquire and disseminate virulence traits and antibiotic resistance.

Bacterial strains, cultivation and phenotypic identification, and antimicrobial susceptibility testing
Simultaneously analyzed reference strains of Staphylococcus spp.were obtained from the Czech Collection of Microorganisms (CCM, Masaryk University, Brno).The strains grew well in a basic set of staphylococcal media at a temperature of 30-37°C.The morpho logical, biochemical, and physiological characterization was performed as previously mentioned (7,(60)(61)(62).Antimicrobial susceptibility testing by disc diffusion method on Mueller-Hinton agar with adherence to EUCAST guidelines (63) was performed as described previously (60).

Genome sequencing and bioinformatics analyses
The short-read sequencing was conducted for the type strain NRL/St 16/872 T (LGC Genomics, Berlin, Germany).The genomic DNA was isolated with a GenElute Bacte rial Genomic DNA kit (Sigma-Aldrich, St. Louis, MO, USA) from pure culture colonies cultivated on Colombia sheep blood agar (Oxoid).The library was prepared using a Nextera XT DNA Library Preparation Kit (Illumina, San Diego, CA, USA) and sequenced externally by LGC Genomics (Berlin, Germany) on the NextSeq platform with 2 × 150 bp reads (Illumina).Genomes of all novel isolates were sequenced by Oxford Nanopore Technology (ONT).The genomic DNA was extracted as described previously (64).Sequencing libraries were prepared using an SQK-RAD004 rapid barcoding kit and sequenced with a FLO-FLG001 cell in a MinION device and MinKnow v21.10.4 software (Oxford Nanopore Technologies, Oxford, UK).

Phylogenetic and pangenomic analyses
The partial 16S rRNA gene was sequenced by Sanger sequencing in the Eurofins MWG Operon sequencing facility (Ebersberg, Germany) with previously described primers (84).Whole-genome sequences of related staphylococcal species were obtained from the NCBI database (Table S1).The multilocus sequence data of six housekeeping genes (rpoB, groEL, dnaJ, tufA, sodA, and gap) that are commonly used in phylogenetic studies of the Staphylococcaceae were extracted from whole-genome sequence assemblies as follows (gene coordinates of S. aureus): 1420..1974 for rpoB, 270..826 for hsp60, 23..911 for dnaJ, 49..929 for gap, 383..1032 for tufA, and 50..480 for the sodA gene.The phylogenetic analyses were performed with the software MEGA version 11 (85).The UBCG collection of 92 conservative genes (86), the average nucleotide identity (ANI) (87), and digital DNA-DNA hybridization (dDDH) values by the d4 formula using the web-based genometo-genome distance calculator (GGDC) version 3.0 (88) were used for calculations of overall genome relatedness indices.The pangenome was calculated with the Ortho Venn2 pipeline with proteins clustered at the default threshold (89).

DNA fingerprinting
For genotypic characterization of the investigated bacterial group, fingerprinting by repetitive sequence-based PCR (rep-PCR) with the primer (GTG) 5 (90) and automated ribotyping with the restriction enzyme EcoRI were performed.The isolation of DNA for rep-PCR fingerprinting, PCR conditions, and fingerprint analysis were performed as previously described (91).Automated ribotyping was performed using the RiboPrinter microbial characterization system (DuPont Qualicon) according to the manufacturer's instructions.Numerical analysis of rep-PCR fingerprints and EcoRI ribotype patterns was performed using BioNumerics version 7.6 (Applied Maths, Belgium).The ribotype patterns were imported into the BioNumerics software using the load samples import script provided by the manufacturer.

PCR analysis of mobilizable genomic island encoding Ccr recombinases
To determine whether a genomic island is mobilizable, we designed primers spanning the excision site and primers targeting the key genes present in the genomic island (Table S2).The PCR reaction was conducted using Quick-Load 2× master mix with standard buffer (New England Biolabs, Ipswich, MA, USA) and a 200 nM concentration of each primer.The genomic island product was further analyzed by Sanger sequencing (Eurofins Genomics, Germany).

Transmission electron microscopy
A 200-mesh carbon/formvar-coated grid was placed on a drop of suspension of bacteria in water for 20 min.Bacterial cells on the grid were negatively stained with 2% ammo nium molybdate and treated with UV light.A Morgagni 268D Philips (ThermoFisher Scientific, The Netherlands) transmission electron microscope was used to visualize bacterial cells.

MALDI-TOF MS
Protein fingerprinting by means of MALDI-TOF MS using an Ultraflextreme instrument (Bruker Daltonics, Germany) was conducted after a standard extraction protocol (92).MALDI-TOF mass spectra were obtained using an UltrafleXtreme instrument (Bruker Daltonics) operated in linear positive mode using the software FlexControl version 3.4.Signals present in at least seven out of nine independent mass spectra acquired per sample were taken into account.Mass spectra were processed using FlexAnalysis version 3.4 (Bruker Daltonics) and BioTyper version 3.1 software (Bruker Daltonics) supplemented with database version 10.0.0.0 (9,607 entries).

Chemotaxonomic characterization
Respiratory quinones were extracted and analyzed as previously described (93).Identity was confirmed by mass spectrometry, as described by Schumann et al. (94).Analysis of the cellular fatty acid profile was performed using a Microbial Identification Sys tem (MIDI, Newark, DE) according to the Standard Protocol of the Sherlock Microbial Identification System software, version 6.1 (95).The fatty acids were identified using gas chromatography-mass spectrometry (GC-MS) according to the study of Vieira et al. (93).
Isolation and structural analysis of the peptidoglycan was performed according to published protocols with some modifications.Briefly, the amino acid composition of the total hydrolysate (4 N HCl, 100°C for 16 h) of the peptidoglycan was analyzed by GC/MS (protocol 10 by Schumann [96]).The partial hydrolysate (4 N HCl, 100°C, 0.75 h) of the peptidoglycan was analyzed by high-resolution liquid chromatography-mass spectrom etry (LC-MS) as described previously (94,96).Enantiomeric analysis was performed by liquid chromatography as described by reference 97.

FIG 1
FIG 1 Evolutionary analyses of Staphylococcus brunensis sp.nov.and its closest relatives.GenBank accession numbers of used sequences are listed in TableS1.

FIG 2
FIG 2 Dendrogram based on MALDI-TOF MS profiles of Staphylococcus brunensis sp.nov.and other phylogenetically related species.The dendrogram was generated using the correlation distance measure with the average linkage algorithm (UPGMA) settings of the software BioTyper version 3.1 (Bruker Daltonics).

FIG 3
FIG 3 Cluster analysis of rep-PCR fingerprints and ribotype patterns obtained from Staphylococcus brunensis sp.nov.strains and the type strains of phylogenet ically closely related Staphylococcus species.(A) Dendrogram based on (GTG) 5 -PCR fingerprints.(B) Dendrogram based on cluster analysis of EcoRI ribotype patterns obtained using a RiboPrinter system.The dendrograms were calculated with Pearson's correlation coefficients with the UPGMA clustering method (r, expressed as percentage similarity values).

FIG 6
FIG 6 Comparative analysis of chromosomal islands harboring ccrDE cassette chromosome recombinases (CI ccrDE ).(A) Comparison of CI ccrDE and flanking regions from different staphylococcal species.The genomic island CI ccrDE is highlighted with a gray background.Genes are labeled according to known or putative functions, as shown in the legend.Only nucleotide blast hits above 65% identity and longer than 2 kb are shown.The position of the island is provided in parentheses next to the GenBank accession number.(B) Maximum likelihood trees of nucleotide sequences of ccrD and ccrA8 recombinases with ccrA1 as an outgroup, and ccrE and ccrB9 recombinases with ccrB1 as an outgroup, respectively.The trees were constructed using Tamura-Nei model with 500 bootstrap replicates.The evolutionary distances are in the number of base substitutions per site.The protein GenBank accession numbers or locus tags are shown in parentheses.

TABLE 1
Origin of the strains of Staphylococcus brunensis sp.nov.characterized in this study a NRL/St, National Reference Laboratory for Staphylococci, National Institute for Public Health, Prague.bCCM,Czech Collection of Microorganisms.ctryptonesoy agar (TSA), Columbia agar with blood, plate count agar, P agar, and nutrient agar, and did not grow in a thioglycollate medium.The Congo red agar method showed negativity in the production of polysaccharide intercellular adhesin (PIA) associated with biofilm formation.Phenotypic identification based on bacitracin resistance and sensitivity to furazolidone, positive catalase test, growth in the presence of NaCl above 10%, and microscopic morphology assigned five isolates as Staphylococcus sp.In contrast to the main characteristics of staphylococci, isolate NRL/St 19/737 exhibited atypical negative catalase activity.Test-dependent results were also observed for the Voges-Pros kauer (VP) test (acetoin) and for β-glucuronidase.With detection of acetoin production, we only obtained positive results in all five strains when pyruvic acid served as substrate using the commercial VP test (Erba Lachema) instead of glucose as substrate for the

TABLE 2
Differentiation of Staphylococcus brunensis sp.nov.from closely related staphylococci occurring in human clinical material

TABLE 3
Plasmid contigs of Staphylococcus brunensis sp.nov.grouped by replication gene and gene content The genomic island was almost identical in both strains; the only difference is the presence of an insertion sequence ISSha1 in strain NRL/St 21/287.The island is bordered by imperfect 18 bp-long direct repeats GAAGC(A/G)TATCATAA(G/A)TGA and harbor ccrA1B2 genes; thus, according to the rules of the IWG-SCC (29) the element was designated SCC NRL/St 16/872 .The most similar element to SCC NRL/St 16/872 is a composite island in genome of S. hominis C34847 (