Novel multidrug-resistant sublineages of Staphylococcus aureus clonal complex 22 discovered in India

ABSTRACT Staphylococcus aureus is a major pathogen in India causing community and nosocomial infections, but little is known about its molecular epidemiology and mechanisms of resistance in hospital settings. Here, we use whole-genome sequencing (WGS) to characterize 478 S. aureus clinical isolates (393 methicillin-resistant Staphylococcus aureus (MRSA) and 85 methicilin-sensitive Staphylococcus aureus (MSSA) collected from 17 sentinel sites across India between 2014 and 2019. Sequencing results confirmed that sequence type 22 (ST22) (142 isolates, 29.7%), ST239 (74 isolates, 15.48%), and ST772 (67 isolates, 14%) were the most common clones. An in-depth analysis of 175 clonal complex (CC) 22 Indian isolates identified two novel ST22 MRSA lineages, both Panton-Valentine leukocidin+, both resistant to fluoroquinolones and aminoglycosides, and one harboring the the gene for toxic shock syndrome toxin 1 (tst). A temporal analysis of 1797 CC22 global isolates from 14 different studies showed that the two Indian ST22 lineages shared a common ancestor in 1984 (95% highest posterior density [HPD]: 1982–1986), as well as evidence of transmission to other parts of the world. Moreover, the study also gives a comprehensive view of ST2371, a sublineage of CC22, as a new emerging lineage in India and describes it in relationship with the other Indian ST22 isolates. In addition, the retrospective identification of a putative outbreak of multidrug-resistant (MDR) ST239 from a single hospital in Bangalore that persisted over a period of 3 years highlights the need for the implementation of routine surveillance and simple infection prevention and control measures to reduce these outbreaks. To our knowledge, this is the first WGS study that characterized CC22 in India and showed that the Indian clones are distinct from the EMRSA-15 clone. Thus, with the improved resolution afforded by WGS, this study substantially contributed to our understanding of the global population of MRSA. IMPORTANCE The study conducted in India between 2014 and 2019 presents novel insights into the prevalence of MRSA in the region. Previous studies have characterized two dominant clones of MRSA in India, ST772 and ST239, using whole-genome sequencing. However, this study is the first to describe the third dominant clone, ST22, using the same approach. The ST22 Indian isolates were analyzed in-depth, leading to the discovery of two new sublineages of hospital-acquired Staphylococcus aureus in India, both carrying antimicrobial resistance genes and mutations, which limit treatment options for patients. One of the newly characterized sublineages, second Indian cluster, carries the tsst-1 virulence gene, increasing the risk of severe infections. The geographic spread of the two novel lineages, both within India and internationally, could pose a global public health threat. The study also sheds light on ST2371 in India, a single-locus variant of ST22. The identification of a putative outbreak of MDR ST239 in a single hospital in Bangalore emphasizes the need for routine surveillance and simple infection prevention and control measures to reduce these outbreaks. Overall, this study significantly contributes to our understanding of the global population of MRSA, thanks to the improved resolution afforded by WGS.


IMPORTANCE
The study conducted in India between 2014 and 2019 presents novel insights into the prevalence of MRSA in the region.Previous studies have character ized two dominant clones of MRSA in India, ST772 and ST239, using whole-genome sequencing.However, this study is the first to describe the third dominant clone, ST22, using the same approach.The ST22 Indian isolates were analyzed in-depth, leading to the discovery of two new sublineages of hospital-acquired Staphylococcus aureus in India, both carrying antimicrobial resistance genes and mutations, which limit treatment options for patients.One of the newly characterized sublineages, second Indian cluster, carries the tsst-1 virulence gene, increasing the risk of severe infections.The geographic spread of the two novel lineages, both within India and internationally, could pose a global public health threat.The study also sheds light on ST2371 in India, a single-locus variant of ST22.The identification of a putative outbreak of MDR ST239 in a single hospital in Bangalore emphasizes the need for routine surveillance and simple infec tion prevention and control measures to reduce these outbreaks.Overall, this study expression of the agr genes, a quorum sensing system that is a key regulator of S. aureus virulence (32).
In the past years, after 2010, another lineage of ST22-MRSA, CA and PVL+, has been reported in countries such as India (33), Kuwait (34), Nepal (35), Japan (36), China (37), and Saudi Arabia (38).Although many studies initially assigned this clone to the European EMRSA-15 ST22 with classic genotyping tools such as multilocus sequence typing (MLST) and pulse-field gel electrophoresis, further analysis of SCCmec subtypes and other molecular markers revealed that it was different from the other known ST22 clones and it most likely originated in Asia (38).To our knowledge, the dominant S. aureus ST22-MRSA identified in Indian hospitals (by references 33,15,13,14) has not yet been characterized using whole-genome sequencng (WGS).
In this study, we characterized the Indian S. aureus population by WGS of a large retrospective collection of 478 clinical isolates from 17 sentinel sites across 10 states in India collected between 2014 and 2019.Moreover, we harnessed the increased resolution of WGS to contextualize the 175 clonal complex (CC) 22 isolates from India with a global public collection of 1624 CC22 isolates and delineated Indian ST22 lineages distinct from those described previously.Although typing studies are helpful for understanding the current makeup of dominant clones, WGS provides a richer resource for surveillance studies and identification of local outbreaks and facilitates a comprehen sive understanding of specific microbes' dynamics.

Population structure of S. aureus across 17 sentinel sites in India
To characterize the population structure of S. aureus in India, 17 sentinel sites were invited to contribute to MRSA isolates collected from patients between 2014 and 2019 (Table S1).Of the 514 isolates received, 508 were confirmed to be S. aureus with Vitek-2 (AST card P628, Biomerieux), and their whole genomes were sequenced.Out of these, 30 genomes did not meet the quality criteria and were further excluded from the study, with the remaining 478 genomes confirmed as S. aureus in silico.Out of these, 85 isolates were later characterized as MSSA using phenotypic and genotypic methods (Table S2), and they were also included in the study.The most frequent specimen source was pus (57%, n = 274), followed by wound (8.9%, n = 43), tracheal (7.7%, n = 37), and blood (3.9%, n = 19; Table S3).The isolates were collected from patients ranging in age from 4 days to 100 years old (median age 42), and 65.89% (n = 315) of the isolates were from males and 34.11% (n = 163) were from females (Table S4).

Antimicrobial resistance (AMR)
In silico AMR profiling was performed on 10 antibiotic classes, and the resistance determinants are shown in more detail in Table S7.One ST239 isolate, sampled from Bangalore in 2015, was found to be resistant to nine classes of antibiotics (G18255029, resistant to all classes tested, except for fusidic acid), while one CC15 isolate, sampled from Pondicherry in 2017, was found to be resistant to only one class of antibiotics (G18255839, resistant to beta-lactams) (Table 2).
The most common resistance gene found in the collection was blaZ, conferring resistance to beta-lactams (present in 465 of the 478 tested isolates), while the least common resistance gene found was mupA, conferring resistance to mupirocin (present in 9 of the 478 tested isolates) (Table 2).

Identification of a potential persistent ST239 outbreak in one Indian sentinel site using WGS
ST239-MRSA is an invasive, highly recombinant, and virulent clone that causes a wide range of life-threatening infections.It has caused hospital epidemics starting from the 1970s throughout the world (39).It is still one of the dominant clones in Asia (40,41) and in Southern India (42,43), although it is slowly being replaced by ST22 and ST772 (15).ST239-MRSA clones were found to carry several types of the SCCmec cassette, including SCCmec types V, III, IV, and I (44).A multidrug-resistant (MDR) ST239-MRSA-III clone exhibiting high resistance to mupirocin and inducible clindamycin resistance has been reported in India since 2010, also carrying the PVL genes (44).
The ST239 isolates were collected from five sentinel sites across the country, and 100%of these were MRSA.All of the ST239 isolates carry a composite version of the SCCmec cassette type III (type III*).We further investigated the structure of the SCCmec type III* cassette in a representative ST239 isolate (G18252308, Fig. S1) with nano pore sequencing.SCCmecFinder v.1.2showed an 83% similarity between the SCCmec type III* cassette and the template sequence AB037671.1,found in strain 85/2082 (https://www.ncbi.nlm.nih.gov/nuccore/AB037671.1/).Unlike the template sequence, the full-length SCCmec III* sequence lacks the mer operon, responsible for resistance to mercury and the pT181 region, but contains extra genes previously found in SCCmec type I, such as the pls gene, which encodes the plasmin-sensitive surface protein, a virulence factor involved in septic arthritis and sepsis.
Forty-four genomes belonging to ST239 and with identical spa types (t030) were aligned to the S. aureus ST239 T0131 reference genome (accession GCF_000204665.1_ASM20466v1), and mean pairwise single-nucleotide polymorphism (SNP) differences were computed between isolates.We identified a cluster of 33 isolates collected from the same sentinel site in Bangalore between 2014 and 2016 (Fig. 2A; https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureuswgs-study#yxiv-st239-outbreak-with-genome-t0131-as-a-reference),where the mean pairwise SNP differences between isolates were seven SNPs.In contrast, the mean pairwise SNP differences between isolates outside the cluster of 33 isolates were 82 SNPs.Our results suggest a nosocomial outbreak that persisted over a period of 3 years.The phylogeny of the 33 isolates is shown in Fig. 2B.One plausible hypothesis is that this outbreak could have a single contamination source, such as a healthcare professional as in the case of the MRSA outbreak in Cambridge (45).However, in the absence of epidemiological information, we cannot state this for sure.

Two novel ST22 MRSA clones, both PVL+ and one harboring the tsst-1 gene
Out of the 175 CC22 isolates from this collection, 138 belong to sequence type 22; 26 belong to ST2371; and 11 belong to other STs.In order to better understand the relationships between the Indian isolates, we contextualized them with a large collection of 1623 CC22 isolates from previously published studies (Fig. 3A).Altogether, 1408 genomes belonged to the EMRSA-15 clone and 105 belonged to the "Gaza clone" (Table 3; Table S10).A temporal analysis of the CC22 genomes with BactDating (47) showed that the Indian isolates form one monophyletic group consisting of two distinct subclusters (first Indian cluster  Microreact view; https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureuswgs-study#me32-cc22-ind-2-isolates-with-one-russian-isolate).The isolate from St.
Petersburg was reported to be the most divergent in the Russian collection and carried tsst-1 and the SCCmec IVa cassette, which were also observed in the Indian isolates of this study.Moreover, this isolate had the same spa type, t005, as 34 isolates belonging to the IND-2 cluster.Due to our limited metadata on the Russian SRR12560298 isolate, we are unable to definitively determine whether this is a case of international transmission or if the isolate originated from an individual from India travelling to Russia.

The CC22 IND-1 cluster
The IND-1 cluster is formed of 107 PVL+ isolates, identified in 9 of the 17 contributing sentinel sites.The average pairwise SNP difference between IND-1 isolates is 86 (0-195), and based on in silico methods, these isolates were predicted to be resistant to betalactams (107), macrolides (50), aminoglycosides (106), quinolones (106), tetracycline (2), and trimethoprim (60).The dominant spa type in this cluster is t852 (82 of 107 isolates).An analysis of the accessory genomes of Indian ST22 isolates using Panaroo (61) revealed that all genomes in the IND-1 cluster carry a version of the SraP protein, which is 111 amino acids shorter than that found in the genomes from the IND-2 cluster, caused by a deletion in an area of the IND-2 SraP with repeated sequences.SraP is a serine-rich adhesin that mediates binding to human platelets, possibly through a receptor-ligand interaction, and is a potential virulence determinant in endovascular infection (62).Clade-specific versions of the SraP proteins were also identified in a study of the CC30-MRSA from Argentina (63).
The SCCmec cassette type IVc found in the IND-1 isolates was 80% similar to the SCCmec cassette from the reference genome of strain 81 of 108 (MR108) (Genbank accession AB096217).Using nanopore long-read sequencing, we obtained the complete sequence of the SCCmec cassette IVc from a representative genome (G18255819), which showed the integration of puB110 plasmid carrying the aadD1 and bleo aminoglycoside resistance genes (see Fig. S2) and the different location of the aacA-aph gene compared to the reference sequence.
A detailed view of the CC22 IND-1 cluster in a global context (see the Microreact view; https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureus-wgs-study#sgfb-cc22-ind-1-isolates-in-global-context)shows that this clone had spread to the UK and Italy.

ST2371, a subclone of the CC22 IND-1 cluster
ST2371 is a single-locus variant of ST22 and a prevalent community-associated MRSA MDR clone.ST2371-MRSA was previously reported in Southern India (16,20) but also sporadically in other parts of the world (64).The 27 ST2371 Indian isolates from this study were collected from four sentinel sites and are closely related to the other ST22 isolates from the IND-1 clade (the mean pairwise SNP difference between the ST22 and the ST2371 genomes is 85
The dominant spa type in the IND-2 cluster is spa t005 (34 of 60 isolates), and the second most dominant one is spa t309 (12 of 60 isolates).
Unlike isolates from clade IND-1, isolates from clade IND-2 carried the SCCmec cassette type IVa; they lack the aadD gene conferring resistance to aminoglycosides and harbor the tsst-1 gene.An analysis of the ST22 pangenome identified the genes entC1, gloB (uncharacterized metallo-beta-lactamase), and tsst as being specific to the genomes in the IND-2 cluster.Enterotoxin type C-1, a product of entC1, is a heat-stable enterotoxin produced by S. aureus that activates the host immune system by binding as unprocessed molecules to major histocompatibility complex class II and T-cell receptor molecules.Enterotoxin type C-1 is often found in processed food from raw milk and can cause food poisoning, and its presence in dairy products has been used as an indicator of food contamination (65).Although absent from the EMRSA-15 reference genome, enterotoxin C (66) was also present in the majority of the ST22 EMRSA-15 genomes (the clade originally called ST22-A [28]).When compared with the reference genome EMRSA-15, the ST22 isolates from the IND-2 cluster have a shorter ComF operon protein one than the reference at position 789971, by 273 nucleotides, due to a premature stop codon.ComF is a competence protein which is thought to play an important role in horizontal gene transfer and furthermore in developing multidrug resistance (67,68).

Conclusions
The present study offers an unprecedented view of MRSA in India from isolates collected between 2014 and 2019 from 17 sentinel sites.Earlier studies have described two of the major dominant clones (ST772 and ST239) using WGS (19,23,69,70,42), and this is the first study that describes the third dominant clone (ST22) using this approach.In addition, the current study also offers details of less studied sequence types, such as ST2371, and provides a clear understanding of phylogenetic relationships between these.An in-depth analysis of the ST22 Indian isolates led to the discovery of two new lineages of hospital-acquired Staphylococcus aureus in India that are both PVL+ and carry resistance genes to fluoroquinolones and aminoglycosides.The presence of the aac(6′)-aph(2″) and aadD resistance genes in these lineages limits treatment options for patients, as these genes confer resistance to commonly used antibiotics.One of the lineages, IND-2, is particularly concerning as it also carries the tsst-1 virulence gene, which increases the risk of severe infections such as toxic shock syndrome and necrotiz ing pneumonia (71).Hospital-acquired infections caused by these lineages can spread rapidly within healthcare settings, posing a risk to vulnerable patients.
Treating infections caused by these lineages may require more expensive antibiot ics, longer hospital stays, and increased use of healthcare resources, leading to higher healthcare costs.Moreover, the geographic spread of these lineages, both within India and internationally, could pose a global public health threat.An analysis of these in a global context found evidence of protential transmission of the two Indian clones to other countries, i.e., the IND-1 cluster to the UK and to Italy and the IND-2 cluster to Russia.
It is important to continue monitoring and characterizing these lineages to inform public health interventions.The samples collected from various body sites and patients of all ages reflect the widespread nature of these lineages.The increased virulence and antimicrobial resistance of these lineages have contributed to their spread in India, making it crucial to identify and address them promptly.In the absence of WGS data, several previous studies have misidentified ST22 isolates as EMRSA-15; however, we were able to show that isolates with similar resistance and virulence profiles belong to the newly characterized Indian lineages.
The study also gives a comprehensive view of the ST2371, a sublineage of CC22, as a new emerging lineage in India and describes it in relationship with the other ST22 isolates.Thus, with the improved resolution afforded by WGS, this study substantially contributed to our understanding of the global population of MRSA.In addition, the retrospective identification of a putative outbreak of MDR ST239 from a single hospital in Bangalore that persisted over the period of 3 years highlights the need for the imple mentation of routine surveillance and simple infection prevention and control measures to reduce these outbreaks (72).
We acknowledge the limitations regarding the clinical information provided and the unknown clinical outcomes associated with the novel sublineages presented in this study, particularly when compared to ST772 or ST239.Furthermore, it is worth noting that the sampling strategies employed in this study resulted in an uneven representation of locations, time points, and infection types.Additionally, it is essential to recognize that the study does not encompass the entirety of India, indicating the need for expanding whole-genome sequencing-based surveillance to include more hospitals and provinces across the country.By doing so, a comprehensive understanding of the epidemiology and dynamics of S. aureus lineages in India can be achieved.This expansion will also facilitate the development of new diagnostic and typing methodologies to identify high-risk clones and enhance infection control measures.

Bacterial isolates
A total of 508 retrospective clinical S. aureus isolates were obtained from 17 sentinel sites across 10 states of the country from the year 2014 to 2019 (Fig. 1; Table 1) through the network of sentinel hospitals established as previously described (73).Thirty genomes did not meet the quality criteria, and the remaining 478 isolates (393 MRSA and 85 MSSA) were included in the current analysis.Epidemiological metadata such as the age and sex of patients and infection type was collected by the sentinel sites.Isolates were confirmed as S. aureus at the reference laboratory morphologically and with the gram-positive ID card on Vitek-2 compact system (Biomerieux).

Whole-genome sequencing, assembly, and annotation
DNA was isolated using the QIAamp DNA mini kit and quantified using Qubit as instructed by the manufacturer.Genome libraries with 450-bp insert size were prepared and sequenced on the Illumina platform with paired-end reads of 150-bp length.The data were assembled using the Spades assembler v.3.14 (74) to generate contigs and annotated with Prokka v.1.5(75).Quality control of sequence data were performed using the GHRU quality control (QC) pipeline based on (i) the basic statistics of raw reads; (ii) the assembly statistics; (iii) contamination due to SNV and sequences from different species; (iv) species prediction using Bactinspector; and (v) overall QC as pass, warning, or fail of each isolate based on these different parameters as described in the pipeline (https://gitlab.com/cgps/ghru/pipelines/dsl2/pipelines/assembly/-/blob/master/qc_conditions_nextera_relaxed.yml).The assembly and other quality metrics of each isolate are provided in Table S11.

Variant detection and phylogenetic analysis
The 478 isolates that passed QC were mapped to the reference genome of S. aureus strain MSSA476 (strain GCF_000011525.1, ST1) using the GHRU-SNP phylogeny pipeline v.1.2.2 (76).The mobile genetic elements (MGEs) were masked in the pseudo genome alignment using MGEmasker (77), and the recombination regions were removed using Gubbins v.2.0.0 (78).The nonrecombinant SNPs were utilized to build a maximum-likeli hood tree using IQ-tree (79) with command line parameters "-czb" to collapse near-zero branches and a general time-reversible model with 1,000 bootstrap replicates.Visualiza tion and phylogeographic analysis were performed on Microreact (80).

Genotyping
ARIBA v.2.14.6 (81) and the PubMLST database were used for multilocus sequence typing of the isolates with the Oxford scheme using the seven housekeeping genes (arcC, aroE, glpF, gmk, pta, tpi, and yqiL) (https://pubmlst.org/organisms/staphylococcus-aureus).The goEBURST algorithm (82) in PHYLOViZ ( 83) software (83) was applied to assign STs to clonal complexes.Those sharing identical alleles at six of seven loci were interpreted as single-locus variants.From the ARIBA results, the isolates with ST designated as novel or ST* were further analyzed using the R package MLSTar (84) as described previously (85).
MLSTar queries the isolate sequence to reference the MLST database and assigns a novel allele sequence or the profiles associated with each housekeeping gene.The novel profile or the novel allele sequences identified by MLSTar were submitted to PubMLST, and new STs were assigned.
The spaTyper tool (86) was used to generate a spa type for the isolates based on the repeat sequences and repeat orders found on http://spaserver2.ridom.de/.
SCCmec typing was performed on the assembled contigs using Staphopia (60).Staphopia includes a primer-based SCCmec typing scheme where the primers are aligned using BLAST.In many cases, Staphopia failed to distinguish between SCCmec type V and SCCmec type VII because the identification of the SCCmec type is based on the ccrC genes.When more than one ccrC gene was present, e.g., both ccrC1 and ccrC2 genes are found, Staphopia reports the sample as "SCCmec V VII." Thus, genomes with uncertain SCCmec types were resolved with an additional analysis with SCCmecFinder (87).

Antimicrobial resistance determinants and virulence genes
Genome data were analyzed for the presence of virulence genes using the Virulence Factor Database (8).Resfinder (89) was used to identify acquired AMR genes in the genomes using the GHRU Resfinder pipeline (90) and mutations using the GHRU AMR pipeline (91) with the PointFinder database (92).All the bioinformatic analyses were performed using custom nextflow pipelines as detailed on protocols.io(93).

Mobile genetic elements
Plasmid replicon types were identified with ARIBA v.2.10.1 (81) and the PlasmidFinder database (94).The 10 most prevalent plasmid replicon types in the entire data set and in the CC22 global collection are shown in the Microreact projects.A detailed list of all plasmid replicon types is shown in Tables S12 to S14.

Nanopore sequencing and assembly
We sequenced representatives of isolates for which the SCCmecfinder gave an alert as "Additional complex(es) was found" and also from the ST772 isolates which were found to carry mecA but no SCCmec cassette, using nanopore Minion-MK1C platform.
The same extracted DNA used for the Illumina sequence was used with the rapid barcoding kit SQK-RBK004 on FLO-MIN106 R9 flow cells to perform long-read sequenc ing.High-accuracy base calling was done with Guppy base caller v.4.3.4 used on MinKNOW v.21.02.20 (available at https://community.nanoporetech.com/downloads) to get fastq files, and the reads below q7 were filtered.Hybrid assembly was performed with both Illumina and nanopore reads using the nf-core Bacas assembly pipeline (95) v.1.1.1 (--assembler "unicycler") to generate a single contig of the complete genome.

CC22 global isolates
The 175 CC22 isolates from the current study were contextualised with 1623 CC22 isolates from previously published studies (Fig. 3A).Altogether, 1,408 genomes belonged to the EMRSA-15 clone and 105 belonged to the Gaza clone (Table 3; Table S5).
The midpoint-rooted phylogenetic tree of the CC22 global isolates was obtained from 44,703 SNPs, after mapping the genomes to the complete genome of strain EMRSA_15_GCF_000695215 and masking regions of recombination and MGEs.
A dated phylogenetic analysis was carried out using BactDating (47), using as an input file the output tree from Gubbins (78).We used the substitution rate as it was estimated for ST22 in a previous study (96).We could not establish the isolation year of two isolates (ERR234855 and ERR234880), which were thus excluded from the temporal analysis.The R code, the input and the output files of our analysis can be found in Supplementary Material 4 provided in Figshare (https://doi.org/10.6084/m9.figshare.21436047.v1).

4 FIG 1
FIG 1 (A) Geographic and temporal distributions and main virulence factors of the STs identified in 478 Indian isolates.The map shows the locations of 17 collecting hospitals across 14 different cities.The markers are proportional in size to the number of contributing hospitals in a particular town.For example, Bengaluru, Karnataka, has three contributing hospitals, i.e., hospitals 12, 14, and 6, and hospitals 16 and 4 in Tiruchirappalli, Tamil Nadu.The pie charts show the proportion of the STs identified at each location.(B) Phylogeny of the 478 Indian S. aureus isolates.The midpoint-rooted phylogenetic tree was inferred from 97,120 informative sites obtained after mapping the genomes to the complete genome of strain Staphylococcus aureus MSSA476 (strain GCF_000011525.1 (ST1) and masking regions of (Continued on next page)

FIG 2 (FIG 1
FIG 2 (A) The timeline shows the distribution of ST239 samples across the outbreak period from March 2014 to October 2016.The bars are colored year-wise.This view is available at https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureus-wgs-study#ffxiv-st239-outbreak.(B) Phylogenetic tree, with linked genotypic data of outbreak isolates at one sentinel site in Bangalore.The maximum-likelihood phylogenetic tree of 43 closely related ST239 isolates was generated using S. aureus ST239 T0131 GCF_000204665.1_ASM20466v1 as a reference genome, identified based on similarity using BactInspector(46).The central cluster of 33 isolates collected from hospital 12 in Bangalore, between 2014 and 2016, is highlighted, together with information regarding the isolates' SCCmec type, spa type, and AMR genetic determinants.The nodes are colored as per the year of collection.The interactive view is available at https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureus-wgs-study#yxiv-st239-outbreak-with-genome-t0131-as-a-reference.

FIG 3 (
FIG 3 (A) Indian CC22 strains in a global context.Phylogenetic tree of the ST22 global collection.The midpoint-rooted phylogenetic tree was obtained from 44,703 SNPs after mapping the genomes to the complete genome of strain EMRSA_15_GCF_000695215 and masking regions of recombination and mobile genetic elements.Tree tips are colored by country of origin as described in the legend.Scale bars represent the number of single-nucleotide polymorphisms per variable site.The interactive views are available at https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureus-wgs-study#olhe-cc22-global-view-rectan gular-tree https://microreact.org/project/2xDvKQhriNveJ4kiVYsmSQ-s-aureus-wgs-study#6w7r-cc22-global-view-radial-tree.(B) Bayesian temporal tree of 1621 ST22 Indian and global isolates with estimated ancestral dates.Tree nodes from major ST22 lineages have been collapsed for clarity.The tree is annotated with the SCCmec types, the presence/absence of the PVL and tsst-1 virulence genes, and the predicted resistance to quinolones and aminoglycosides.The MRCA of IND-1 and IND-2 is marked with blue on the timeline.The data used to generate this tree are available at https://microreact.org/project/2xDvKQhriN veJ4kiVYsmSQ-s-aureus-wgs-study#6w7r-cc22-global-view-radial-tree.

TABLE 1 A
breakdown of the S. aureus Indian collection by clonal complexes a

TABLE 2
Distribution of resistance determinants associated with 10 antibiotic classes across clonal complexes