Whole genome sequencing of clinical methicillin-resistant Staphylococcus aureus isolates across hospitals in Mwanza, Tanzania
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Methicillin-resistant Staphylococcus aureus (MRSA) is a leading cause of both healthcare-associated and community-acquired infections, characterized by limited treatment options and substantial morbidity and mortality, particularly in low- and middle-income countries such as Tanzania. However, genomic data which provides valuable insights into the pathogen’s genetic landscape, remains scarce. This study utilized whole-genome sequencing (WGS) to examine local molecular epidemiological profiles of 14 MRSA isolates isolated from blood, urine, and pus samples during (n=6; June 2019 – June 2020) and after (n=8; March – August 2023) the implementation of the National Action Plan on Antimicrobial Resistance (NAP-AMR) in Mwanza, Tanzania. All isolates belonged to sequence type 8 (ST8), and 92.9% (13/14) were spa type t1476. Phenotypic and genotypic antimicrobial resistance results were consistent for all antibiotics tested, with the exception of tetracycline. Pairwise single-nucleotide polymorphism (SNP) analysis revealed genetic diversity, with SNP differences ranging from 4 to 140. Maximum likelihood phylogeny analysis based on SNP data identified two closely related MRSA pairs, one from multiple wards and the other confined within the same ward of the same hospital. The WGS of MRSA isolates revealed that all MRSA isolates belonged to ST8 with substantial genetic diversity and phylogenetic analysis revealed two closely related MRSA pairs, suggesting potential transmission events. These findings highlight the importance of genomic surveillance for uncovering MRSA transmission patterns and guiding infection prevention efforts.
Article activity feed
-
-
Dear Authors, thank you for submitting your manuscript to Access Microbiology. It has now been reviewed by two experts in the field, whose comments are attached at the bottom of this email. Both reviewers acknowledge the importance of applying whole-genome sequencing to MRSA isolates from a region with limited genomic surveillance. However, they have raised significant concerns that need to be addressed before the manuscript can be considered for acceptance. These include major methodological and interpretative issues, particularly regarding phylogenetic analysis, clonality, and chromosomal inference of resistance/virulence genes. Unclear sampling frameworks, which limit the strength of the conclusions. Overstatement of impact and lack of contextualisation with existing Tanzanian genomic data. I invite you to address these points in …
Dear Authors, thank you for submitting your manuscript to Access Microbiology. It has now been reviewed by two experts in the field, whose comments are attached at the bottom of this email. Both reviewers acknowledge the importance of applying whole-genome sequencing to MRSA isolates from a region with limited genomic surveillance. However, they have raised significant concerns that need to be addressed before the manuscript can be considered for acceptance. These include major methodological and interpretative issues, particularly regarding phylogenetic analysis, clonality, and chromosomal inference of resistance/virulence genes. Unclear sampling frameworks, which limit the strength of the conclusions. Overstatement of impact and lack of contextualisation with existing Tanzanian genomic data. I invite you to address these points in detail, along with all other reviewer comments, to strengthen the manuscript towards acceptance.
-
Comments to Author
- Clarification of Terminology - Please define ESKAPE pathogens at first mention (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species), even though they are well known, for clarity. - Replace "nosocomial infections" with healthcare-associated infections (HAIs) consistently throughout the manuscript. - Clarify AST as "Antimicrobial Susceptibility Testing" at first mention. - The study reports 410 bloodstream infection (BSI) cases, yet only 14 isolates were sequenced. This raises concerns about representativeness. Please explain the basis for selecting these isolates and whether they are intended to represent the outbreak or a subset of interest (this is major comment). - The stated aim is to clarify K. pneumoniae …
Comments to Author
- Clarification of Terminology - Please define ESKAPE pathogens at first mention (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species), even though they are well known, for clarity. - Replace "nosocomial infections" with healthcare-associated infections (HAIs) consistently throughout the manuscript. - Clarify AST as "Antimicrobial Susceptibility Testing" at first mention. - The study reports 410 bloodstream infection (BSI) cases, yet only 14 isolates were sequenced. This raises concerns about representativeness. Please explain the basis for selecting these isolates and whether they are intended to represent the outbreak or a subset of interest (this is major comment). - The stated aim is to clarify K. pneumoniae in bloodstream infections. However, the supplementary data also include ESBLs in UTIs and RTIs. Please clarify whether the BSIs analyzed were primary infections or secondary bloodstream infections following UTIs/RTIs. - Correct "Marc-April" to "March-April." - Remove stray characters such as "w." at the end of a paragraph. - Ensure consistent formatting of gene names. For example, "blaSHV-187" should be written with "SHV" in capital letters, not "blasHV-187."
Please rate the manuscript for methodological rigour
Satisfactory
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Partially support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
Comments to Author
The study applies Illumina WGS to 14 MRSA isolates from five hospitals in Mwanza, Tanzania, all ST8, and describes their resistome, virulome, and phylogenetic relationships. The setting is under-represented in the literature. However, in its current form the manuscript has major problems in four areas: 1. Study design and sampling are insufficiently described. 2. Several key genomic inferences are overstated or not supported by the methods (especially clonality, "limited" diversity, and chromosomal vs plasmid location of ARGs/VFs). 3. The analysis underuses available contextual data and omits essential phylogenetic quality controls (recombination / mobile element handling, tree support). 4. The conclusions and "Impact" section substantially overstate what can be claimed from 14 isolates. There is a …
Comments to Author
The study applies Illumina WGS to 14 MRSA isolates from five hospitals in Mwanza, Tanzania, all ST8, and describes their resistome, virulome, and phylogenetic relationships. The setting is under-represented in the literature. However, in its current form the manuscript has major problems in four areas: 1. Study design and sampling are insufficiently described. 2. Several key genomic inferences are overstated or not supported by the methods (especially clonality, "limited" diversity, and chromosomal vs plasmid location of ARGs/VFs). 3. The analysis underuses available contextual data and omits essential phylogenetic quality controls (recombination / mobile element handling, tree support). 4. The conclusions and "Impact" section substantially overstate what can be claimed from 14 isolates. There is a need for substantial re-analysis and rewriting. Major comments 1. Overstated claims about clonality and "limited genetic diversity" The abstract and discussion repeatedly describe the population as "highly clonal" with "limited genetic diversity", yet the pairwise SNP distances span 4 to 140 SNPs. * 140 SNPs is not "limited diversity" for a single ST in a hospital-based collection; that range is more consistent with a mix of related and unrelated ST8 backgrounds. * The pangenome analysis also reports 731 accessory genes and 455 "novel" genes across 14 isolates, which is again hard to reconcile with "highly clonal". The "two clones" each consist of only two isolates sampled within 6 to 13 days in the same hospital, with 4 and 8 SNP differences respectively. * These pairs are compatible with recent transmission but do not justify strong language such as "clonal groups at a zonal referral hospital suggesting local transmission" and "reveal clonality and potential cross-transmission" (conclusion). * With only two isolates in each "clone", no recombination filtering described, and no broader time-series, the data support "two closely related isolate pairs" rather than robustly defined clones or transmission chains. 2. Location of ARGs and virulence factors (chromosomal vs plasmid) The abstract, results, discussion, and figure legend all state that "all resistance and virulence genes were chromosomally encoded". Yet, the methods indicate only short-read Illumina sequencing with de novo assembly (Velvet) and use of PlasmidFinder, MobileElementFinder, pLannotate. * Short-read assemblies typically fragment SCCmec and other MGEs. * MobileElementFinder and PlasmidFinder identify plasmid replicons and MGEs, not definitive chromosomal vs plasmid locations for each ARG/VF. * The Figure 3 legend describes a single "plasmid-borne contig" and uses the spatial separation of erm(C) from a rep10 replicon on that contig to infer that all ARGs/VFs are chromosomal, which is not logically valid. * In addition, many staphylococcal ARGs/VFs are associated with transposons, IS elements, or SCCmec cassettes; if such MGEs fall on separate contigs without replicons, their mobility cannot be excluded simply because MobileElementFinder does not annotate them as plasmid-associated in that assembly. 3. Phylogenetic methods: reference choice, recombination/MGEs, and tree support Reference choice: using USA300_FPR3757 ST8 as reference is appropriate and should be clearly highlighted earlier (i.e., same ST). Recombination/MGEs: The manuscript does not mention any masking of recombination, phage regions, or IS elements before SNP calling. Yet the chosen reference contains multiple IS elements and prophages I.e.; ISSep2, position 77993..79553 Incomplete prophage, position 879301-899940 Intact prophage, position 1544651-1603612 ISSau5, position 1630930..1632065 ISSau3, position 1802453..1804455 Intact prophage, position 2084486-2130861 Did you at least exclude any SNPs in these regions (if you didn't filter out regions of recombination)? Without masking these regions, SNP distances and tree topology can be heavily influenced by horizontal events rather than clonal evolution. Tree support and presentation: The text refers to "maximum likelihood" trees, but there is no mention of bootstrap or other support values, and the figure legends for Figures 5 and 6 do not describe a scale bar or support metrics. 4. Study design, sampling frame, and MRSA definition The methods state that this is a cross-sectional hospital-based study among patients with suspected BSI/UTI/SSTI at five facilities. * It is not clear whether the 14 MRSA isolates represent all MRSA recovered in those facilities over the specified periods, or a selected subset (e.g. based on feasibility, resource constraints, or phenotype). * Without a clear sampling frame, it is difficult to interpret statements about "persistence" of lineages or transmission patterns. * Line 96: "Bacterial isolates identified as MRSA were collected…". The criteria and methods for defining MRSA at the point of collection are not described (e.g. cefoxitin/oxacillin AST, PBP2a testing, mecA PCR, or combination). Later, all isolates are re-identified and tested on Vitek MS/Vitek2 with EUCAST interpretation, but the initial diagnostic pathway should be made explicit. 5. Use of existing public genomic data The introduction claims "a significant gap in comprehensive data on MRSA in Tanzania" and the Impact section states this is the "first in-depth genomic insight into MRSA strains using WGS following the implementation of the NAP-AMR in Mwanza". However, later sections acknowledge and include data from several prior Tanzanian MRSA genomics studies (Manyahi et al., Geofrey et al., Omar et al.) and compare ST8 genomes from these works to the current isolates. Additionally, there are publicly available S. aureus genomes from Tanzania (e.g. on platforms such as PathogenWatch https://pathogen.watch/) that could have been used more broadly to contextualize ST8 diversity and to test whether the observed isolates cluster locally or represent a subset of wider regional/global lineages. 6. Overstated impact and conclusions from a sample of 14 isolates Throughout the manuscript, the language often exceeds what can be supported by 14 isolates from 5 hospitals: * Introduction: claims that findings "contribute to the global discourse on combating AMR" and emphasise "the importance of localized genomic research in addressing a worldwide crisis" are very broad for such a small, geographically limited dataset. * Impact section and Conclusion use strong language around "persistence", "stable genetic evolution", "clonality", and "cross-transmission" based on very few isolates and minimal epidemiological data. The "Study limitations" section acknowledges small sample size but then immediately states that it "enriches global AMR surveillance" in a way that feels disproportionate. Minor / specific comments (by section) Abstract 1. Opening sentence ("AMR is a critical global health threat, with MRSA posing a significant challenge"). This is very generic. Consider specifying why MRSA is a challenge (e.g. major cause of HA and CA infections with high morbidity/mortality and limited treatment options), or shorten and move the emphasis to the Tanzanian context to make room for more study-specific content. 2. "All resistance and virulence genes were chromosomally encoded". See Major Comment 2: this should be removed or heavily qualified. 3. "Pairwise SNP analysis revealed limited genetic diversity, with SNP differences ranging from 4 to 140". As above, this is internally contradictory; either delete "limited" or explicitly distinguish between closely related pairs vs broader diversity. 4. "The strong concordance between phenotypic and genotypic resistance…". Generally fine, but you may wish to briefly acknowledge the tetracycline discrepancy even in the abstract (or at least not call it "strong" without nuance). The "Impact of the Study" section currently repeats large portions of the abstract rather than providing a distinct, value-added interpretation. This section should succinctly articulate what this analysis adds beyond existing Tanzanian MRSA genomics studies, and how these findings could inform local AMR surveillance or future work. As written, it does not contribute new insight and should be substantially revised. Introduction 5. Claims about "significant gap in comprehensive data on MRSA in Tanzania" and "contribute to the global discourse" should be toned down and aligned with the existing Tanzanian genomic literature that you already cite later. PathogenWatch contains 107 S. aureus genomes from Tanzania collected between 2005 and 2014 (outside this study period), albeit, I haven't check to confirm if they carry mec gene in silico (i.e., if they're MSSA or MRSA). 8 of these genomes are ST8. Methods 6. Clarify whether the 14 MRSA isolates are all MRSA detected during the periods or a subset; explain the selection criteria. 7. Specify exactly how "bacterial isolates identified as MRSA" were defined at the sentinel sites (e.g. AST, mecA PCR, etc.), and how this relates to the subsequent re-confirmation in Würzburg. 8. Good that an ST8 reference was used to call SNPs; make explicit that this choice was driven by the MLST results and that it minimises mapping bias. 9. The description of MobileElementFinder and pLannotate should clearly state that these tools identify MGEs and annotate plasmid contigs but do not provide definitive chromosomal vs plasmid assignment for all ARGs/VFs in fragmented short-read assemblies. Results 10. "Enrolment" could be confusing in a retrospective WGS analysis; consider rephrasing to "at the time of sampling" or "at the time of isolate collection". 11. Authors report 66 plasmid replicons across 14 isolates; this is interesting and could be briefly discussed in relation to ARG distribution (even if you cannot fully resolve plasmid structures). 12. Figure 3 legend: see Major Comment 2. Wording here is currently misleading and should be revised. 13. For the two low-SNP pairs, please state clearly whether these involve the same patient or different patients (if known). That strongly affects how the reader interprets "clonal clusters". 14. Please define what the scale bars represent in figure 5 and figure 6. Also consider using bootstrapping node support. 15. Figure 6: specify in the manuscript main text how many genomes from each external study were included and how they were selected. Discussion 16. "This study highlights the persistence of ST8 spa type t1476 MRSA strains, a highly virulent and drug-resistant lineage known for its epidemic potential". This over-attributes to the present data; I would reframe as: a. "Our small collection of clinical MRSA isolates in Mwanza was dominated by ST8 spa type t1476, consistent with previous Tanzanian studies reporting widespread dissemination of this lineage" 17. "VF genes and ARGs… were all chromosome-borne". See Major Comment 2; this is not supported by short-read data alone and should be removed or reframed as "we did not detect ARGs/VFs on contigs carrying plasmid replicons". 18. The section on "limited genetic diversity", "two clonal groups suggesting local transmission", and "absence of clonal relationships between MRSA strains from this study and those from previous studies" is internally inconsistent and over-interpreted. I would suggest rewriting this entire paragraph. Distinguish clearly between: low-SNP pairs within this study (compatible with recent transmission), and broader diversity and relationships to historical Tanzanian ST8 isolates (compatible with regional introduction and diversification). Study limitation The small number of isolates and unclear sampling frame are major limitations for any conclusions about persistence, clonality, or transmission. It's not a bad thing. These need to be emphasised more strongly and explicitly linked to the need for cautious interpretation. How were the samples chosen. Were you limited by resources? Was sequencing WGS an issue? Conclusion ""SNP-based phylogenetic analyses reveal clonality and potential cross-transmission of MRSA ST8 strains in this setting." You need to be careful with wording here. A precise, defensible replacement would be: "SNP-based phylogenetic analysis identified two pairs of closely related ST8 isolates, sampled within days at the same hospital, which is compatible with recent transmission but cannot confirm transmission pathways. Given the small sample size and limited epidemiological data, broader conclusions about clonality or cross-transmission across the region cannot be drawn."
Please rate the manuscript for methodological rigour
Poor
Please rate the quality of the presentation and structure of the manuscript
Poor
To what extent are the conclusions supported by the data?
Not at all
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-