Mechanisms of linezolid resistance in staphylococci and enterococci isolated from two teaching hospitals in Shanghai, China

Abstract

Background: Linezolid is one of the most effective treatments against Gram-positive pathogens. However, linezolid-resistant/intermediate strains have recently emerged in worldwide. The purpose of this study was to analyse the prevalence and resistance mechanisms of linezolid-resistant/intermediate staphylococci and enterococci in Shanghai, China.

Results: Thirty-two linezolid-resistant/intermediate strains, including 14 Staphylococcus capitis, three Staphylococcus aureus, 14 Enterococcus faecalis and one Enterococcus faecium clinical isolates, were collected in this study which displayed linezolid MICs of 8 to 512 μg/ml, 8-32 μg/ml, 4-8 μg/ml and 4 μg/ml, respectively. All linezolid-resistant S. capitis isolates had a novel C2131T mutation and a G2603T mutation in the 23S rRNA region, and some had a C316T (Arg106Cys) substitution in protein L4 and/or harboured cfr. Linezolid-resistant S. aureus isolates carried a C389G (Ala130Gly) substitution in protein L3, and/or harboured cfr. The cfr gene was flanked by two copies of the IS256-like element, with a downstream orf1 gene. Linezolid-resistant/intermediate enterococci lacked major resistance mechanisms. The semi-quantitative biofilm assay showed that 14 linezolid-resistant E. faecalis isolates produced a larger biofilm than linezolid-susceptible E. faecalis strains. Transmission electron microscopy showed the cell walls of linezolid-resistant/intermediate strains were thicker than linezolid-susceptible strains.

Conclusion: Our data indicated that major resistance mechanisms, such as mutations in 23S rRNA and ribosomal proteins L3 and L4, along with cfr acquisition, played an important role in linezolid resistance. Secondary resistance mechanisms, such as biofilm formation and cell wall thickness, should also be taken into account.

Figure 1

PFGE of Sma I-digested chromosomal DNA of linezolid-resistant isolates. Lane 17–207, fourteen isolates of LR S. capitis. 17, HS09-17; 24, HS10-24; 44, HS11-44; 49, HS12-49; 51, HS12-51; 53, HS12-53; 55, HS12-55; 58, HS12-58; 60, HS13-60; 201, HS12-201; 203, HS11-203; 204, HS10-204; 206, HS09-206; 207, HS13-207. Lane 302–311, fourteen isolates of LR E. faecalis. 302, HS13-302; 310, HS12-310; 312, HS13-312; 301, HS13-301; 305, HS11-305; 311, HS10-311; 307, HS11-307; 315, RJ13-315; 309, HS12-309; 314, RJ13-314; 303, HS13-303; 308, HS12-308; 313, HS13-313; 304, HS11-304; 311, HS10-311. Lane 56–205, three isolates of LR S. aureus. 56, HS12-56; 202, HS11-202; 205, HS09-205.

Figure 2

Mutations in the 23S rRNA gene and linezolid resistance levels in LR S. capitis isolates. A. The percentages of clones with C2131T and G2603T mutations. B. The percentage of clones with C2131T mutations in isolates with various linezolid resistance levels. C. The percentage of clones with G2603T mutations in isolates with various linezolid resistance levels. Error bars represent the mean ± SEM.

Figure 3

Analysis of plasmids in cfr -positive S. capitis isolates. Plasmids of ca.54 kb were detected in all ten cfr -positive isolates. A. Southern hybridisation of S. capitis isolates with a cfr probe. B. M, E. coli V517 marker; 203, HS11-203; 44, HS11-44; 49, HS12-49; 201, HS12-201; 51, HS12-51; 53, HS12-53; 55, HS12-55; 58, HS12-58; 60, HS13-60; 56, HS12-56.

Figure 4

Schematic representation of the genetic environment of cfr in pHS and pSS-01. The arrows indicate the positions and directions of the transcription of the genes. The regions of homology between pHS and pSS-01 (GenBank accession no. JQ041372) are indicated by dashed lines and grey shading. Δ indicates a truncated gene.

Figure 5

Semi-quantitative biofilm analysis of S. capitis, S. aureus, E. faecalis, E. faecium groups. LS strains were used as controls. Comparisons are vs. LS E. faecalis strains for LR E. faecalis strains. ***P < 0.0001 (one-way ANOVA).

Figure 6

Transmission electron micrographs (TEM) showing cell wall thickness in bacterial isolates. TEM images show LS S. capitis HS12-102 (A), LR S. capitis HS09-17 (MIC 16 μg/ml) grown without (B) and with (C) 4 μg/ml linezolid, LR S. capitis HS12-55 (MIC >256 μg/ml) grown without (D) and with (E) 32 μg/ml linezolid, wild-type ATCC29213 S. aureus (F), LR S. aureus HS11-202 (MIC 16 μg/ml) grown without (G) and with (H) 2 μg/ml linezolid, wild-type ATCC29212 E. faecalis (I), LI E. faecalis HS11-304 (MIC 4 μg/ml) grown without linezolid (J), LR E. faecalis HS12-309 (MIC 8 μg/ml) grown without (K) and with (L) 2 μg/ml linezolid, LS E. faecium HS13-194 (M), LI E. faecium HS11-306 (MIC 4 μg/ml) grown without linezolid (N). Cell wall thicknesses are given in Table 3. Scale bars indicate 200 nm.

Figure 7

Comparison between bacterial cell wall thickness in LS, LR and LI isolates. There are S. capitis (A), S. aureus (B), E. faecalis (C) and E. faecium (D) isolates. 102, HS12-102; 17, HS09-17; 17 + LZD, HS09-17 grown with 4 μg/ml linezolid; 55, HS12-55; 55 + LZD, HS12-55 grown with 32 μg/ml linezolid; ATCC29213, ATCC29213 LS S. aureus; 202, HS11-202; 202 + LZD, HS11-202 grown with 2 μg/ml linezolid; ATCC29212, ATCC29212 LS E. faecalis; 304, HS11-304; 309, HS12-309; 309 + LZD, HS12-309 grown with 2 μg/ml linezolid; 194, HS13-194; 306, HS11-306 (MIC 4 μg/ml). LS reference strains HS12-102 (S. capitis), ATCC29213 (S. aureus), ATCC29212 (E. faecalis) and HS13-194 (E. faecium) were used as controls for the respective groups. 17 versus 102, 55 versus 102, 17 + LZD versus 17, 55 + LZD versus 55, 202 versus ATCC29213, 202 + LZD versus 202, 304 versus ATCC29212, 309 versus ATCC29212, 309 + LZD versus HS12-309 showed significantly differences (**P ≤ 0.001; ***P < 0.0001, two-tailed t-test).