Staphylococcus lugdunensis (S. lugdunensis) is a coagulase-negative Staphylococcus comprising the normal skin microbiota, primarily colonizing the lower abdomen and extremities.S. lugdunensis has attracted substantial attention in recent years since the discovery of lugdunin by Zipperer et al in 2016. Lugdunin is a secondary metabolite synthesized by non-ribosomal peptide synthetase (NRPS) that inhibits the growth of various gram-positive bacteria, including methicillin-resistant S. aureus (MRSA), suggesting its potential as an antibiotic. To date, 28 complete genome sequences and 11 partial-assembly genome sequences of S. lugdunensis have been published and uploaded to the GenBank database (https://www.ncbi.nlm.nih.gov/genome/browse/#!/prokaryotes/2548/, January 18, 2022). However, there are limited data on the genome sequences of S. lugdunensis from mainland China, and the effects of geographical origin on genetic variations in S. lugdunensis remain unknown.
We analyzed six clinical strains of S. lugdunensis isolated from Beijing, Wuhan, and Fujian, China [Supplementary Table 1, https://links.lww.com/CM9/B273]. The minimum inhibitory concentrations (MICs) of antibiotics were evaluated using the VITEK2 Compact system (bioMerieux, Inc., Durham, USA). Cefoxitin disc dilution (cefoxitin DD), broth microdilution for oxacillin, and oxacillin salt agar methods were used as reference methods for assessing methicillin resistance. The complete genomes of the six S. lugdunensis strains were obtained by next-generation sequencing and Oxford Nanopore sequencing, and then compared with genomes available in the GenBank database. We further characterized the resistome and secondary metabolism gene clusters of these staphylococcal cassette chromosome mec (strains using computational approaches. We identified SCCmec) elements of methicillin-resistant S. lugdunensis (MRSL) and performed collinearity analysis of these strains. Comparisons were performed for the obtained nucleotide sequences of the lug operon with those in the GenBank database using Blastx and Blastp tools against the proteins of the operon. We further explored the relationships of genetic variations of the lug operon with geographical origins and the clonal complex (CC). The detailed methods are reported in the Supplementary Materials, https://links.lww.com/CM9/B273.
The genome sizes of the six strains ranged from 2.59 to 2.71 Mbp, with a GC content of 33.7% to 33.9% [Supplementary Figure 1, https://links.lww.com/CM9/B273]. All strains contained 2405 to 2523 coding sequences, with 60 to 61 tRNAs, 16 to 19 rRNAs, and 51 to 53 sRNAs. Two to three plasmids were identified in five strains (except RMLUG5). Multilocus sequence typing revealed sequence type (ST)3 (RMLUG1, RMLUG3, RMLUG6), ST27 (RMLUG2), ST34 (RMLUG4), and ST6 (RMLUG5).
In antimicrobial susceptibility tests by VITEK2, three of the six strains were resistant to cefoxitin and oxacillin, as confirmed by the oxacillin broth microdilution method and oxacillin salt agar screening. Cefoxitin DD tests showed that strains RMLUG1 and RMLUG6 were susceptible to cefoxitin, although near the breakpoint [Supplementary Table 2, https://links.lww.com/CM9/B273]. The three MRSL strains were susceptible to antibacterial agents showing activity against MRSA, such as linezolid, vancomycin, and teicoplanin [Supplementary Table 3, https://links.lww.com/CM9/B273]). Notably, the methicillin resistance gene mecA was identified in the three MRSL strains (RMLUG1, RMLUG3, and RMLUG6) and in one methicillin-susceptible S. lugdunensis (MSSL) strain (RMLUG2). The SCCmec elements were V(5C2) (RMLUG1, RMLUG2, and RMLUG6) and IVi (2B) (RMLUG3). Only the ccr class 9 (ccrC2 allele 1) gene was identified in the MSSL strain RMLUG4, which is a composite of the SCCmec element.
Supplementary Figure 2, https://links.lww.com/CM9/B273 illustrates the SCCmec structure of the six strains. Strains RMLUG1 and RMLUG6 were consistent with SCCmec type V from S. aureus strain WIS [Supplementary Figure 2A, https://links.lww.com/CM9/B273]. Compared with SCCmec type IVi in S. aureus strain JCSC6668, strain RMLUG3 contained a 1014 bp deletion of the gene encoding ISSep1-like transposase [Supplementary Figure 2B, https://links.lww.com/CM9/B273]. The ST27 strain RMLUG2 contained a complete SCCmec type V, including a type mecA class C2 complex and ccrC1 complex, but in a SCCmec type VII-like order: orfX-J3-ccr-J2-mec-J1. Moreover, the IS431 upstream of mecA had the opposite orientation to that downstream, in contrast with the structure of SCCmec type VII, suggesting a novel SCCmec type V variant [Supplementary Figure 2C, https://links.lww.com/CM9/B273]. The SCCmec elements of RMLUG4 did not carry the mecA gene but contained ccr class 9 (ccrC2) and determinants of resistance to heavy metals such as arsenic and copper. According to SCCmec classification, this element is described as an SCC element, which was designated SCCRMLUG4 [Supplementary Figure 2D, https://links.lww.com/CM9/B273].
MIC tests and genomic analysis indicated that strain RMLUG2 carried the mecA gene but did not exhibit oxacillin resistance. No mutation of the mecA gene was observed among the MRSL strains. C to T substitution was identified at position –33 (i.e., 33 bp upstream of the start codon) within the mecA promoter in strain RMLUG2, which also occurred in RMLUG1 and RMLUG6. A comparison of the sequences of FemXAB family genes between strain RMLUG2 and the MRSL strain JICS135 revealed several mutations in femX (eg, R176K and D341E) and femB (eg, D245N and D259Y). The femX and femB genes of strains RMLUG1, RMLUG3, and RMLUG6 matched those of strain JICS135.
A comparison of the genomes of the six strains to 14 lug operon genes (lugJ to lugM; Supplementary Table 4, https://links.lww.com/CM9/B273) identified the complete lug operon in all strains except RMLUG2, which only contained lugM. There was a nonsense mutation (g.308T > A) in lugM in strain RMLUG2. Six frameshift mutations were identified in all strains in the genes encoding NRPS enzymes. All frameshift mutations resulted from the deletion of one or two nucleotides, leading to premature termination codons. Additionally, an 18 bp deletion was identified at the beginning of the putative regulator lugR gene in our five strains. An insertion segment, g.67_68insATTTTATACAGGAAGAAG, was identified in lugZ of strains RMLUG4 and RMLUG5.
Among all strains, 34 of the 39 genomes in the GenBank database contained 14 complete lug genes, two of which belonged to the same type of strain (NCTC12217) despite different assembly levels from different laboratories. Similar to strain RMLUG2, strains VCU150, VISLISI_25, and C_33 harbored only lugM, but had missense mutations instead of the nonsense mutation. Further analysis of the 38 strains indicated missense mutations in 12 of 14 genes (excluding lugI and lugD), making this the most common mutation type. Missense mutations were detected in lugJ (g.394G > A) in all CC1 isolates. Interestingly, missense mutations were found in lugE (g.331A > G) in 32 of the 38 S. lugdunensis strains, but were not present in any ST1 strain, except strain HKU09-01. Missense mutations in lugG (g.551C > T) were present in all CC6 isolates. In addition, genetic variations in the lug operon were independent of geographical origin.
In the current study, the MIC tests and genomic analysis indicated discordance between genotype and phenotype, similar to the findings of Kao et al There are several possible reasons for this discrepancy. First, the heterogeneity of the mecA gene can suppress methicillin resistance. Gargis et al reported that a frameshift mutation in mecA and lack of the full mecA promoter sequence can cause susceptibility to oxacillin and cefoxitin, respectively. Additionally, Chen et al reported that the –33C to T substitution within the mecA promoter can lower the oxacillin MIC, whereas this mutation was identified in strains RMLUG2(MSSL), RMLUG1(MRSL), and RMLUG6(MRSL). Second, mutations in auxiliary genes may contribute to decreased resistance to methicillin. Giannouli et al reported that the accumulation of amino acid changes in FemXAB family proteins may affect cell wall synthesis, leading to atypical oxacillin responsiveness. Several mutations were detected in the femX, femA, and femB genes between strain RMLUG2 and S. lugdunensis JICS135, which might account for the lowered MIC.
Moreover, we identified CC-dependent genetic variations of the 14 lug genes, indicating that the lug operon may not be conserved in the S. lugdunensis genome. Lebeurre et al detected significant genetic variations independent of CCs in the lug locus. However, they only searched against genes encoding four NRPS enzymes (lugA, lugB, lugC, and lugD) and one regulator gene (lugR), with the results being somewhat consistent with our findings. Our results indicated that not all S. lugdunensis isolates harbor the lug operon. Four strains containing only lugM varied in clinical source and geographical location; thus, no link was evident between S. lugdunensis lug polymorphisms and geographical origin.
There were some limitations to this study. First, we only described the phenotypes and genotypes of the six strains, which could not establish a causal relation. Functional studies are needed to investigate the activity of resistance genes and the lug operon. Second, the GenBank data may be biased because some sequences were obtained only by third-generation sequencing without assembling sequence fragments obtained by other methods. Thus, the accuracy of the sequences should be taken into account. In particular, some nucleotides need to be corrected to confirm the mutations of the lug operon in the future.
Overall, our findings revealed variations in the genomes of S. lugdunensis strains isolated from different cities in China. We report novel SCCmec and SCCRMLUG4 elements, suggesting the need for additional studies on this species. Comparative analysis of the lug operon for all available genomes demonstrated CC-dependent genetic variations; however, further research is needed to elucidate the relationships between genotypes and phenotypes.
The complete genome sequences have been deposited at GenBank under the accessions CP084480-CP084483 (RMLUG1), CP084434-CP084436 (RMLUG2), CP084437-CP084439 (RMLUG3), CP084440-CP084442 (RMLUG4), CP084443 (RMLUG5), and CP084444-CP084446 (RMLUG6).
The authors thank Pengcheng Du (Beijing YuanShengKangTai (ProtoDNA) Genetech Co Ltd.) for the technical help in Oxford Nanopore sequencing. We would like to thank Jing Wu for the collection of patient characteristics. We would like to thank Dr. Ronghua Liu (Microbiology Laboratory of Linfen Central Hospital, Linfen, Shaanxi, China) for antibiotic susceptibility test.
This work was supported by grants from the National Science and Technology Major Project of China (No. 2017ZX10103004-006) and the National Key Research and Development Programme of China (No. 2016YFC0903800).
Conflicts of interest
1. Bieber L, Kahlmeter G. Staphylococcus lugdunensis
in several niches of the normal skin flora. Clin Microbiol Infect 2010;16:385–388. Doi: 10.1111/j.1469-0691.2009.02813.x.
2. Zipperer A, Konnerth MC, Laux C, Berscheid A, Janek D, Weidenmaier C, et al. Human commensals producing a novel antibiotic impair pathogen colonization. Nature 2016;535:511–516. Doi: 10.1038/nature18634.
3. Li DX, Zhai Y, Zhang Z, Guo YT, Wang ZW, He ZL, et al. Genetic factors related to the widespread dissemination of ST11 extensively drug-resistant carbapenemase-producing Klebsiella pneumoniae strains within hospital. Chin Med J 2020;133:2573–2585. Doi: 10.1097/CM9.0000000000001101.
4. International Working Group on the Classification of Staphylococcal Cassette Chromosome Elements (IWG-SCC). Classification of staphylococcal cassette chromosome mec (SCCmec): guidelines for reporting novel SCCmec elements. Antimicrob Agents Chemother 2009;53:4961–4967. Doi: 10.1128/AAC.00579-09.
5. Kao CY, Wu HH, Chang SC, Lin LC, Liu TP, Lu JJ. Accurate detection of oxacillin-resistant Staphylococcus lugdunensis by use of agar dilution. J Microbiol Immunol Infect 2022;55:234–240. Doi: 10.1016/j.jmii.2021.02.009.
6. Gargis AS, Yoo BB, Lonsway DR, Anderson K, Campbell D, Ewing TO, et al. Difficult-to-detect Staphylococcus aureus: mecA-positive isolates associated with oxacillin and cefoxitin false-susceptible results. J Clin Microbiol 2020;58:e2038–19. doi: 10.1128/JCM.02038-19.
7. Chen FJ, Wang CH, Chen CY, Hsu YC, Wang KT. Role of the mecA gene in oxacillin resistance in a Staphylococcus aureus clinical strain with a pvl-positive ST59 genetic background. Antimicrob Agents Chemother 2014;58:1047–1054. Doi: 10.1128/AAC.02045-13.
8. Giannouli S, Labrou M, Kyritsis A, Ikonomidis A, Pournaras S, Stathopoulos C, et al. Detection of mutations in the FemXAB protein family in oxacillin-susceptible mecA-positive Staphylococcus aureus clinical isolates. J A ntimicrob Chemother 2010;65:626–633. Doi: 10.1093/jac/dkq039.
9. Lebeurre J, Dahyot S, Diene S, Paulay A, Aubourg M, Argemi X, et al. Comparative genome analysis of Staphylococcus lugdunensis shows clonal complex-dependent diversity of the putative virulence factor, ess/type VII locus. Front Microbiol 2019;10:2479. Doi: 10.3389/fmicb.2019.02479.