Abstract
Objectives
To determine if vancomycin-resistant Enterococcus faecium (VREfm) carriers carry the same VREfm clone after a minimum follow-up of 365 days. For those carrying the same clone, we investigated the genomic evolution per year per genome.
Methods
We used WGS results to assign VREfm clones to each isolate and determine clone shifts. Finally, we calculated distance in core-genome MLST alleles, and the number of SNPs between consecutive VREfm isolates from patients carrying the same VREfm clone.
Results
In total, 44.2% of patients carried the same VREfm clone, and the genomic evolution was 1.8 alleles and 2.6 SNPs per genome per year.
Conclusions
In our population of long-term carriers, we calculated a molecular clock of 2.6 SNPs.
Introduction
Enterococcus faecium is a common nosocomial pathogen, and the cause of both serious invasive infections and colonization.1 In the Capital Region of Denmark we have seen a rise in vancomycin-resistant E. faecium (VREfm) isolates since 2012.2 In this study, we used WGS to examine consecutive VREfm isolates from patients with positive samples taken a minimum of a year apart. In our setting, they would be considered long-term VREfm carriers. However, we suspect some of these patients to have cleared their initial VREfm colonization and been recolonized with a different clone.
Literature on the natural or spontaneous clearance of VREfm colonization is sparse and results vary greatly. One study showed a natural VREfm clearance of only 33% over a 3 year study period.3 In a systematic review, 50% of subjects had cleared VREfm colonization at 25 weeks after initial colonization.4 In a previous clinical trial, investigating the effect of probiotics on eradication of VREfm colonization in an elderly, comorbid and immunocompetent population, the number of participants with spontaneous decolonization was higher than expected: 56% after 4 weeks, and 90% after 24 weeks in the placebo group.5
In this study we aimed to determine how many of our patients considered long-term VREfm carriers were true long-term carriers, that is, still carrying the same VREfm clone for at least a 1 year period. Secondly, we investigated how the VREfm isolates evolved in the intestine of the true long-term carriers.
Methods
Selection of patients
The study is a retrospective study with data from 2012 until September 2022 conducted in the Capital Region of Denmark in collaboration between two Departments of Clinical Microbiology. We cover nine hospitals and all GPs in the region. In the Capital Region of Denmark, we perform active surveillance of VREfm, both from clinical samples and screening samples. The first VREfm isolate per patient per year (clinical or screening sample) is routinely sequenced and kept in our database. An exception applies to patients with both a vanA and a vanB within the same year, in which case, both isolates are sequenced. In hospitals, patients were screened in the case of hospital outbreaks, or upon admission if the patient had had a positive VREfm sample 6 months prior to admission time.6 We did not perform screening of non-hospitalized VREfm carriers. We accessed the database with the objective of identifying patients with repeated VREfm isolates by use of the Danish citizens’ personal identification numbers. Patients were included in the study if they had two or more sequenced VREfm isolates a minimum of 365 days apart. We excluded one patient who had both a vanA and vanB isolate from the same day.
WGS and bioinformatic methods
The VREfm isolates were sequenced on the Illumina MiSeq platform (Illumina Inc., San Diego, USA) as previously described.7 Raw reads were trimmed using BBDuk (https://sourceforge.net/projects/bbmap/) with the parameters ktrim = r, k = 23, mink = 11, hdist = 1, tbo, qtrim = r and minlength = 30, and assembled with SKESA v.2.2 with default settings except inclusion of the parameter –allow_snps. Only assemblies with a genome size in the range 2.7–3.2 Mb, a minimum depth of coverage of 30, and N50 of minimum 10 000 were included in the study. We used SeqSphere+ v.8.2.0 (Ridom GmbH, Munster, Germany; http://www.ridom.de/seqsphere/) with the previously published scheme for E. faecium for the initial bioinformatic analysis.8 Sequencing reads were aligned, and SNPs were called using the NASP pipeline v.1.1.2.9 In the pipeline, duplicated regions in the reference were masked using NUCmer v.3.1, reads were subsequently mapped to the E. faecium Aus0004 (CP003351.1) reference genome with BWA-mem v.0.7.16 and SNPs were called with GATK v.3.8.0.10–13 Consensus bases had a minimum coverage of 10 and a minimum proportion of 0.9 for the called base. Pairwise comparisons between isolates of patients were performed using the consensus base matrix. Recombination regions were detected for every clone group and filtered out using Gubbins v.3.2.1.14
Analysis of long-term carriage
We included two isolates per patient. Most of the patients had only two sequenced VREfm isolates; however, a subgroup of patients had three or more sequenced isolates. Regarding those patients, the first sequenced isolate was included in the study along with the next isolate after a minimum of 365 days. Thus, all patients contributed with only two isolates each for the primary analysis. The SeqSphere+ software assigned core-genome MLST (cgMLST) to each isolate. Subsequently, the isolates were visualized in a minimum spanning tree (MST), and cgMLST clones were defined with a setting of maximum of 20 alleles between isolates. The MSTs are presented in Figures S1, S2 and S3 (available as Supplementary data at JAC-AMR Online). For each patient, we determined if there had been a shift from one clone to another.
Genetic evolution of VREfm isolates over time
For this analysis, patients could contribute with more than two isolates. Concerning the subgroup of patients with three or more sequenced isolates, we included the first isolate as the baseline. The subsequent isolate taken at least 365 days after the baseline isolate was included. Following isolates were included if they were taken a minimum of 365 days after the previous isolate. All included isolates were compared with the baseline isolate in the WGS analysis. An MST was performed as in the primary analysis, and clones were defined for all pairs. We determined whether there had been a shift from one clone to another in each pair of isolates. The pairs without clone shift were identified, and distance matrix was measured to quantify the number of allele differences. Finally, the number of SNPs between pairs were measured, and we calculated the molecular clock for our population.
Statistical analysis
The statistical analysis was performed with R v.4.2.2.15 We used a mixed-effect linear model to estimate the genetic evolution in the secondary analysis, as the observations were not independent.
Ethics
The study was approved by the Danish Data Protection Agency (P-2022-653), and the Danish Health and Medicines Authority (3-3013-1118/1).
Results
Analysis of long-term carriage
Of 6689 patients with at least one positive VREfm sample in the database, 206 patients had at least two sequenced VREfm isolates a minimum of 365 days apart and were included in the analysis. Age and sex for the subgroups are listed in Table 1. For 115 (55.8%) patients the second VREfm isolate belonged to a different clone than the first VREfm isolate, and for 91 (44.2%) both isolates belonged to the same clone. The median number of days between samples was 753.0 (IQR 546.5–1083.5) in the subgroup with clone shift, and 524.0 (IQR 428.5–699.5) in the subgroup without clone shift (Table 1). The distribution of the major STs did not differ between the subgroups, but in the subgroup with clone shift most of the shifts were to an ST80 clone (Table 1).
Table 1.
Summary statistics of included VREfm carriers with and without shift in clones
| With clone shift | Without clone shift | Total | |
|---|---|---|---|
| Patients, n (%) | 115 (55.8) | 91 (44.2) | 206 |
| Female sex, n (%) | 67 (58.3) | 49 (53.8) | 116 (56.3) |
| Age (years), median (IQR) | 74.0 (64.0–80.0) | 72.0 (63.0–77.0) | — |
| Days between samples, median (IQR) | 753.0 (546.5–1083.5) | 524.0 (428.5–699.5) | — |
| MLST | From: To: 80 80 (n = 15; 13.0%) 80 1421 (n = 10; 8.7%) 117 80 (n = 7; 6.0%) 203 80 (n = 8; 7.0%) 203 1421 (n = 12; 10.4%) 1421 80 (n = 22; 19.1%) Other (n = 41; 35.7%) |
MLST observed: 17 (n = 2; 1.1%) 80 (n = 34; 18.7%) 117 (n = 21; 11.5%) 203 (n = 60; 33.0%) 612 (n = 2; 1.1%) 1421 (n = 58; 31.9%) 1478 (n = 1; 0.5%) other (n = 4; 2.2%) |
— |
| vanA, n (%) | 165 (71.7) | 151 (83.0) | 316 (76.7) |
| vanB, n (%) | 63 (27.4) | 30 (16.5) | 93 (22.6) |
| vanA and vanB, n (%) | 2 (0.9) | 1 (0.5) | 3 (0.7) |
Genetic evolution of VREfm isolates over time
A total of 223 pairs of isolates were included in the analysis of the genetic evolution of VREfm isolates over time. Of these, 123 pairs had shifted from one clone to another, and 100 pairs belonged to the same clone. The further analysis of the 100 pairs without clone shift revealed a significant estimate of 1.8 alleles per year per genome (95% CI 1.3–2.3, P < 0.001). The SNP analysis resulted in a significant estimate of 2.6 SNPs per year per genome (95% CI 1.8–3.4, P < 0.001), constituting the molecular clock for the included isolates (Figure S4). The average percentage of compared genome between all isolate pairs was 80.5% (SD = 0.02).
Discussion
In our cohort, 55.8% of the patients had acquired a different VREfm clone when compared with the baseline sample, and 44.2% still carried the same clone at the time of the second sample. The subgroup without clone shift could be long-term carriers of VREfm. Regarding the subgroup with different VREfm clones, recolonization most likely explains long-term carriage of VREfm; however, in vivo horizontal gene transfer from other gut commensals is also possible. Not surprisingly, the median number of days between samples was higher in the subgroup of patients with a clone shift. There was no difference in age between the two subgroups. Rohde and colleagues16 investigated the risk factors of VREfm colonization and found them to be age, recent stay in a long-term care facility, recent hospital admission, recent proton pump inhibitor/antiacid treatment, previous colonization with MDR organisms and antibiotic treatment at the time of VREfm colonization or in the 6 months prior to detection. In the analysis of genetic evolution, we found the genetic evolution to be 1.8 alleles per year, and 2.6 SNPs per year, constituting the molecular clock. In previous studies of multiple individuals, the molecular clock of VREfm has been estimated as between 3.41 and 5 SNPs per year.17,18 In our study we report the molecular clock from a collection of genomes from the same individuals, and we believe this could be the explanation for a lower SNP accumulation.
The study is limited by the retrospective observational study design. We do not have data on duration of hospital stay and comorbidities. In a previous randomized clinical trial we found the VREfm carriers to be a homogeneous group with a high Charlson comorbidity index of 4.5 The patients included in the study have only been retested on hospital readmission. Therefore, we cannot determine the time of decolonization based on these data. Since we observe clonal spread of VREfm, we cannot rule out recolonization with the same clone for the group of patients without clonal shift within the 524 day observation period.19,20 Another limitation is that only one isolate, recovered by picking one random colony, per sample is sequenced. A recent study on immunosuppressed patients suggests they can be carriers of multiple clones at the same time. The authors of that study sequenced 10 colonies per sample and showed that 33.3% of patients carried two VREfm clones.21
The study contributes to the existing knowledge of long-term VREfm carriage and the genetic evolution of the bacteria in carriers. The molecular clock of VREfm should be further investigated in future WGS studies to fully understand the genetic evolution of VREfm.
Conclusions
In conclusion, just over half of the patients with a positive VREfm sample more than a year after their initial sample had acquired a different VREfm suggesting recolonization. We estimated the molecular clock of VREfm in long-term carriers with the same VREfm clone, to be 1.8 alleles per year per genome, and 2.6 SNPs per year per genome.
Supplementary Material
Acknowledgements
We thank statistician Thomas Kallemose from Hvidovre Hospital for his help with the statistical analysis.
Contributor Information
Maja Johanne Søndergaard Knudsen, Department of Clinical Microbiology, Copenhagen University Hospital—Amager and Hvidovre, Hvidovre, Denmark.
Jose Alfredo Samaniego Castruita, Department of Clinical Microbiology, Copenhagen University Hospital—Amager and Hvidovre, Hvidovre, Denmark.
Sarah Mollerup, Department of Clinical Microbiology, Copenhagen University Hospital—Amager and Hvidovre, Hvidovre, Denmark.
Barbara Juliane Holzknecht, Department of Clinical Microbiology, Copenhagen University Hospital—Herlev and Gentofte, Herlev, Denmark; Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
Morten Hoppe, Department of Clinical Microbiology, Copenhagen University Hospital—Herlev and Gentofte, Herlev, Denmark.
Henrik Westh, Department of Clinical Microbiology, Copenhagen University Hospital—Amager and Hvidovre, Hvidovre, Denmark; Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
Mette Pinholt, Department of Clinical Microbiology, Copenhagen University Hospital—Amager and Hvidovre, Hvidovre, Denmark.
Ingrid Maria Cecilia Rubin, Department of Clinical Microbiology, Copenhagen University Hospital—Amager and Hvidovre, Hvidovre, Denmark.
Funding
The WGS was internally funded as a part of the routine work. The project was further funded by the Scandinavian Society for Antimicrobial Chemotherapy (SSAC) Foundation, the Regions' Medical and Treatment Foundation and the Danmarks Frie Forskningsfond.
Transparency declarations
None to declare.
Data availability
All isolates are uploaded in a BioProject. Please refer to Table S1 for BioProject number and metadata for the isolates.
Supplementary data
Figures S1 to S4 and Table S1 are available as Supplementary data at JAC-AMR Online.
References
- 1. Ahmed MO, Baptiste KE. Vancomycin-resistant enterococci: a review of antimicrobial resistance mechanisms and perspectives of human and animal health. Microb Drug Resist 2018; 24: 590–606. 10.1089/mdr.2017.0147 [DOI] [PubMed] [Google Scholar]
- 2. DTU, Statens Serum Institut . DANMAP 2020: Use of Antimicrobial Agents and Occurrence of Antimicrobial Resistance in Bacteria from Food Animals, Food and Humans in Denmark. ISSN 1600-2032. 2021. https://www.ssi.dk/-/media/arkiv/subsites/antibiotikaresistens/danmap_2020_07102021_version-2_low.pdf.
- 3. Roghmann MC, Qaiyumi S, Schwalbe R et al. Natural history of colonization with vancomycin-resistant Enterococcus faecium. Infect Control Hosp Epidemiol 1997; 18: 679–80. 10.1086/647510 [DOI] [PubMed] [Google Scholar]
- 4. Shenoy ES, Paras ML, Noubary F et al. Natural history of colonization with methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE): a systematic review. BMC Infect Dis 2014; 14: 177. 10.1186/1471-2334-14-177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Rubin IMC, Mollerup S, Broholm C et al. No effect of Lactobacillus rhamnosus GG on eradication of colonization by vancomycin-resistant Enterococcus faecium or microbiome diversity in hospitalized adult patients. Microbiol Spectr 2022; 10: e0234821. 10.1128/spectrum.02348-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Knudsen MJS, Rubin IMC, Gisselø K et al. The use of core genome multilocus sequence typing to determine the duration of vancomycin-resistant Enterococcus faecium outbreaks. APMIS 2022; 130: 323–9. 10.1111/apm.13216 [DOI] [PubMed] [Google Scholar]
- 7. Pinholt M, Larner-Svensson H, Littauer P et al. Multiple hospital outbreaks of vanA Enterococcus faecium in Denmark, 2012–13, investigated by WGS, MLST and PFGE. J Antimicrob Chemother 2015; 70: 2474–82. 10.1093/jac/dkv142 [DOI] [PubMed] [Google Scholar]
- 8. de Been M, Pinholt M, Top J et al. Core genome multilocus sequence typing scheme for high-resolution typing of Enterococcus faecium. J Clin Microbiol 2015; 53: 3788–97. 10.1128/JCM.01946-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Sahl JW, Lemmer D, Travis J et al. NASP: an accurate, rapid method for the identification of SNPs in WGS datasets that supports flexible input and output formats. Microb Genomics 2016; 2: e000074. 10.1099/mgen.0.000074 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. McKenna A, Hanna M, Banks E et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res 2010; 20: 1297–303. 10.1101/gr.107524.110 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Delcher AL, Salzberg SL, Phillippy AM. Using MUMmer to identify similar regions in large sequence sets. Curr Protoc Bioinformatics 2003; Chapter 10: Unit 10.3. 10.1002/0471250953.bi1003s00 [DOI] [PubMed] [Google Scholar]
- 12. Li H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv 2013. https://arxiv.org/abs/1303.3997v2.
- 13. DePristo MA, Banks E, Poplin R et al. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 2011; 43: 491–8. 10.1038/ng.806 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Croucher NJ, Page AJ, Connor TR et al. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins. Nucleic Acids Res 2015; 43: e15. 10.1093/nar/gku1196 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. R Core Team . R: A Language and Environment for Statistical Computing. 2021.
- 16. Rohde AM, Walker S, Behnke M et al. Vancomycin-resistant Enterococcus faecium: admission prevalence, sequence types and risk factors—a cross-sectional study in 7 German university hospitals from 2014 to 2018. Clin Microbiol Infect. 2022;. 29: 515–22. 10.1016/j.cmi.2022.11.025 [DOI] [PubMed] [Google Scholar]
- 17. Howden BP, Holt KE, Lam MMC et al. Genomic insights to control the emergence of vancomycin-resistant enterococci. mBio 2013; 4: e00412-13. 10.1128/mBio.00412-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Rios R, Reyes J, Carvajal LP et al. Genomic epidemiology of vancomycin-resistant Enterococcus faecium (VREfm) in Latin America: revisiting the global VRE population structure. Sci Rep 2020; 10: 5636. 10.1038/s41598-020-62371-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hammerum AM, Justesen US, Pinholt M et al. Surveillance of vancomycin-resistant enterococci reveals shift in dominating clones and national spread of a vancomycin-variable vanA Enterococcus faecium ST1421-CT1134 clone, Denmark, 2015 to March 2019. Euro Surveill 2019; 24: 1900503. 10.2807/1560-7917.ES.2019.24.34.1900503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Pinholt M, Bayliss SC, Gumpert H et al. WGS of 1058 Enterococcus faecium from Copenhagen, Denmark, reveals rapid clonal expansion of vancomycin-resistant clone ST80 combined with widespread dissemination of a vanA-containing plasmid and acquisition of a heterogeneous accessory genome. J Antimicrob Chemother 2019; 74: 1776–85. 10.1093/jac/dkz118 [DOI] [PubMed] [Google Scholar]
- 21. Biehl L, Claben AY, Gølz H et al. In-Patient Diversity of Intestinal Vancomycin-Resistant Enterococci in Haematological and Oncological Patients: Results From a Prospective Multicentre Study. ECCMID 2023. Abstract E1072.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All isolates are uploaded in a BioProject. Please refer to Table S1 for BioProject number and metadata for the isolates.