Volume 19, Number 1—January 2013
Linezolid Dependence in Staphylococcus epidermidis Bloodstream Isolates
We document linezolid dependence among 5 highly linezolid-resistant (LRSE) Staphylococcus epidermidis bloodstream isolates that grew substantially faster at 32 µg/mL linezolid presence. These isolates carried the mutations T2504A and C2534T in multiple 23S rRNA copies and 2 mutations leading to relevant amino acid substitutions in L3 protein. Linezolid dependence could account for increasing LRSE emergence.
Linezolid is highly effective against Staphylococcus epidermidis (1). Linezolid-resistant S. epidermidis (LRSE) isolates are limited worldwide (2), and few LRSE outbreaks have occurred (3,4). Linezolid resistance in S. epidermidis has been attributed to specific 23S rRNA mutations (G2576U, G2447U, U2504A, C2534U, and G2631U) (5,6), cfr gene (7), or mutations in ribosomal proteins L3, L4, and L22 (7).
Dependence on linezolid for bacterial growth has not been reported but has been described for other antimicrobial drugs (8–10). We report the characteristics of partially linezolid-dependent LRSE causing bloodstream infections (BSIs).
Twenty-seven LRSE isolates were randomly selected for study among the 46 single-patient LRSE isolates recovered from BSIs in Tzaneio General Hospital (Piraeus, Greece) during 2008–2010. Isolates were identified by Vitek 2 (bioMérieux, Marcy l’Etoile, France). Chloramphenicol and clindamycin MIC was determined by E-test (bioMerieux) and linezolid MIC by using broth microdilution (11).
The 27 LRSE isolates were tested by pulsed-field gel electrophoresis (PFGE) as described (12) and screened for cfr gene (7). Mutations in the peptidyl-transferase center were identified for each separate 23S rRNA copy as described (13).
In 8 LRSE isolates representing all PFGE types, genes encoding the L3, L4, and L22 ribosomal proteins that factor in ribosome assembly were sequenced to identify mutations conferring linezolid resistance (6). Nucleotide and amino acid sequences were analyzed by using Lasergene software (DNASTAR, Madison, WI, USA) and compared with those of the linezolid-susceptible S. epidermidis (LSSE) strain ATCC12228 (GenBank accession no. AE015929).
Growth curves were conducted in the presence and absence of linezolid for the above 8 LRSE isolates, 1 clinical LSSE isolate (A1521, linezolid MIC 2 μg/mL), and the ATCC 29213 S. aureus strain (linezolid MIC 0.5 μg/mL) as controls. Linezolid concentrations tested were half-MIC for controls and 3 LRSE isolates with low MIC (16–32 μg/mL) and 8, 16, 32, 64, and 128 μg/mL for 5 LRSE isolates with MIC >256 μg/mL. Growth curves were performed in triplicate by diluting 20 μL Mueller-Hinton broth culture in 2 mL broth, followed by incubation at 37°C under constant shaking; turbidity of cultures (McFarland scale) was measured every 6 h for 36 h. We statistically compared isolate growth at each time point using the paired t test and Minitab software version 13.31 (www.minitab. com); p<0.05 indicated statistical significance.
We retrospectively examined medical records (anonymized demographic data, clinical characteristics, comorbidities, prior linezolid treatment for >3 days, and in-hospital deaths) of the 27 patients harboring LRSE to ascertain factors influencing resistance acquisition and outbreak persistence. Each of the 27 patients yielding LRSE had prolonged hospitalization and carried a central venous catheter. Twenty-one were mechanically ventilated, and 25 received linezolid treatment (Table 1).
Linezolid MICs were >256 μg/mL for 23 LRSE isolates and 8–32 μg/mL for 4 LRSE isolates. All isolates were co-resistant to clindamycin and chloramphenicol, but the cfr gene was not detected by PCR in any isolate (7). Three PFGE types were identified. PFGE type I comprised the 23 highly LRSE isolates, which all carried mutations T2504A and C2534T; 3 LRSE isolates were related to each other (type II) and carried the mutations G2576T and C2534T; and 1 LRSE isolate was unique (type III) and carried G2576T along with novel mutations C2356T or T2334C in different 23S rRNA copies each. All isolates had mutations in 3–6 copies of 23S rRNA. The cfr gene was not detected in any isolate.
Characteristics of the 8 LRSE isolates tested by growth analysis are shown in Table 2; curves of the 5 highly LRSE isolates at 0, 32, and 128 μg/mL linezolid and of the 3 low-level LRSE and controls at half-MIC linezolid are shown in Figures 1 and 2. The growth of all 8 LRSE isolates was significantly slower than for the S. aureus control (p<0.05 at 24 h and at 36 h incubation for all isolates). Exposure to 8 μg/mL linezolid did not affect growth of the 5 highly LRSE isolates (p>0.05 for all isolates; data not shown). The 3 low-level LRSE isolates and the LSSE control showed moderately slower growth (p>0.05 at 24 h and 36 h) and the S. aureus control showed significantly slower growth (p<0.05 at 24 h and 36 h) at half-MIC linezolid than without linezolid. However, exposure of the 5 highly LRSE isolates to 32 and 128 μg/mL linezolid resulted in significantly faster growth compared with linezolid absence (p<0.05 at 24 and 36 h with 32 μg/mL linezolid and p<0.01 at 24 and 36 h with 128 μg/mL linezolid for all 5 isolates), suggesting partial linezolid dependence. Remarkably, all 5 linezolid-dependent LRSE isolates grew significantly faster with 128 μg/mL linezolid than did the 3 low-level LRSE isolates and the LSSE control with half-MIC and without linezolid (p<0.05 at 24 h and 36 h). Furthermore, 3 linezolid-dependent LRSE isolates (A2864, A2562, 217) grew significantly faster with 128 μg/mL linezolid than did the S. aureus control without linezolid (p<0.05 at 24 h and 36 h).
The 5 linezolid-dependent LRSE isolates had 2 potentially relevant amino acid substitutions, G152D (shift from a small amino acid to a negative hydrophilic) and D159Y (shift of hydrophilic to hydrophobic amino acid), and a less significant one (L101V) in L3 protein. No amino acid changes were observed in the remaining 3 isolates tested for proteins L3, L4, and L22 or in proteins L4 and L22 for any isolate tested.
All study isolates were recovered from patients with BSIs, indicating relatively high infectivity. Most of the LRSE isolates were clonally related, but 3 distinct PFGE types were detected, implying that linezolid resistance emerged in at least 3 different strains, which subsequently spread between patients. However, all linezolid-dependent isolates were clonal, implying that dependence possibly emerged once or on few occasions.
Antimicrobial drug resistance associated with dependence has been described for streptomycin and vancomycin (8–10). While investigating linezolid resistance in 8 LRSE isolates, we observed slower growth without linezolid than in controls, possibly resulting from mutations conferring resistance-exerted fitness cost. Surprisingly, linezolid concentrations at >32 μg/mL caused impressive growth acceleration in all 5 highly LRSE isolates, rendering significantly faster growth than without linezolid. Linezolid dependence is evident starting from relatively low linezolid concentrations, against which LRSE may be exposed in vivo during linezolid treatment. In fact, most of these 27 patients, including all 5 harboring linezolid-dependent LRSE, had prolonged linezolid treatment before yielding LRSE. This exposure also may have fostered the transition from resistance to dependence as suggested previously in vancomycin-dependent enterococci (8). Therefore, the high intrahospital linezolid consumption may favor not only LRSE selection but also their competitive survival. Should linezolid dependence prove common in highly LRSE isolates, it could explain their increasing clinical occurrence and the emergence of LRSE outbreaks (3,4,13). To support this hypothesis, growth with and without linezolid needs to be tested on larger collections of LRSE isolates. Growth characteristics of LRSE isolates reported previously should also be studied.
The underlying mechanism by which linezolid binding to the mutated ribosomal subunits enhances growth may be complex. All 5 linezolid-dependent isolates harbored mutations T2504A combined with C2534T, whereas the linezolid-nondependent isolates harbored other mutations in 23S rRNA genes (Table 2). Also, only the linezolid-dependent isolates carried mutations in the ribosomal protein L3, known to stimulate ribosome assembly. The coupling of rRNA synthesis from precursor RNA molecules and ribosome assembly possibly affects the overall rate of protein synthesis in vivo (14). Linezolid may interfere in this interaction, thus affecting the ribosomal assembly and enabling interactions with precursor forms of the 50S subunit, as demonstrated for erythromycin (15). We speculate that linezolid-dependent cells may possess linezolid-dependent ribosomal precursor particles exhibiting different structural conformation, which favors a faster rate of the overall protein synthesis recovery. This feature might explain the linezolid-dependent growth of the isolated strains. Further functional ribosomal characterization is required to elucidate linezolid dependence.
Dr Pournaras is assistant professor of microbiology at the Medical School of the University of Thessaly, Greece, and associate professor of microbiology at the Medical School of the University of Athens, Greece. His research interests include detection methods, surveillance, and molecular mechanisms of antimicrobial resistance and molecular epidemiology of multidrug-resistant bacteria.
This work was supported in part by grants from the Research Committee of University of Thessaly, Greece.
- Brickner SJ, Hutchinson DK, Barbachyn MR, Manninen PR, Ulanowicz DA, Garmon SA, Synthesis and antibacterial activity of U-100592 and U-100766, two oxazolidinone antibacterial agents for the potential treatment of multidrug-resistant gram-positive bacterial infections. J Med Chem. 1996;39:673–9 and. DOIPubMedGoogle Scholar
- Biedenbach DJ, Farrell DJ, Mendes RE, Ross JE, Jones RN. Stability of linezolid activity in an era of mobile oxazolidinone resistance determinants: results from the 2009 Zyvox® Annual Appraisal of Potency and Spectrum program. Diagn Microbiol Infect Dis. 2010;68:459–67 and. DOIPubMedGoogle Scholar
- Bonilla H, Huband MD, Seidel J, Schmidt H, Lescoe M, McCurdy SP, Multicity outbreak of linezolid-resistant Staphylococcus epidermidis associated with clonal spread of a cfr-containing strain. Clin Infect Dis. 2010;51:796–800 and. DOIPubMedGoogle Scholar
- Mulanovich VE, Huband MD, McCurdy SP, Lemmon MM, Lescoe M, Jiang Y, Emergence of linezolid-resistant coagulase-negative Staphylococcus in a cancer centre linked to increased linezolid utilization. J Antimicrob Chemother. 2010;65:2001–4 and. DOIPubMedGoogle Scholar
- Wong A, Reddy SP, Smyth DS, Aguero-Rosenfeld ME, Sakoulas G, Robinson DA. Polyphyletic emergence of linezolid-resistant staphylococci in the United States. Antimicrob Agents Chemother. 2010;54:742–8 and. DOIPubMedGoogle Scholar
- Zhu W, Tenover FC, Limor J, Lonsway D, Prince D, Dunne WM Jr, Use of pyrosequencing to identify point mutations in domain V of 23S rRNA genes of linezolid-resistant Staphylococcus aureus and Staphylococcus epidermidis. Eur J Clin Microbiol Infect Dis. 2007;26:161–5 and. DOIPubMedGoogle Scholar
- Mendes RE, Deshpande LM, Farrell DJ, Spanu T, Fadda G, Jones RN. Assessment of linezolid resistance mechanisms among Staphylococcus epidermidis causing bacteraemia in Rome, Italy. J Antimicrob Chemother. 2010;65:2329–35 and. DOIPubMedGoogle Scholar
- Tambyah PA, Marx JA, Maki DG. Nosocomial infection with vancomycin-dependent enterococci. Emerg Infect Dis. 2004;10:1277–81 and. DOIPubMedGoogle Scholar
- Moubareck C, Meziane-Cherif D, Courvalin P, Périchon B. VanA-type Staphylococcus aureus strain VRSA-7 is partially dependent on vancomycin for growth. Antimicrob Agents Chemother. 2009;53:3657–63 and. DOIPubMedGoogle Scholar
- Honoré N, Marchal G, Cole ST. Novel mutation in 16S rRNA associated with streptomycin dependence in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 1995;39:769–70 and. DOIPubMedGoogle Scholar
- Clinical and Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing, 20th informational supplement. CLSI document M100–S20. Wayne (PA): The Institute; 2010.
- Ikonomidis A, Michail G, Vasdeki A, Labrou M, Karavasilis V, Stathopoulos C, In vitro and in vivo evaluations of oxacillin efficiency against mecA-positive oxacillin-susceptible Staphylococcus aureus. Antimicrob Agents Chemother. 2008;52:3905–8 and. DOIPubMedGoogle Scholar
- Liakopoulos A, Spiliopoulou I, Damani A, Kanellopoulou M, Schoina S, Papafragas E, Dissemination of two international linezolid-resistant Staphylococcus epidermidis clones in Greek hospitals. J Antimicrob Chemother. 2010;65:1070–1 and. DOIPubMedGoogle Scholar
- Nierhaus KH. Ribosome assembly. In: Nierhaus KH, Wilson DN, editors. Protein synthesis and ribosome structure. Weinheim (Germany): Wiley-VCH & Co. KGaA; 2004. p. 85–95.
- Wilson DN. The A–Z of bacterial translation inhibitors. Crit Rev Biochem Mol Biol. 2009;44:393–433 and. DOIPubMedGoogle Scholar
TablesCite This Article
Table of Contents – Volume 19, Number 1—January 2013
|EID Search Options|
|Advanced Article Search – Search articles by author and/or keyword.|
|Articles by Country Search – Search articles by the topic country.|
|Article Type Search – Search articles by article type and issue.|
Please use the form below to submit correspondence to the authors or contact them at the following address:
Spyros Pournaras, Department of Microbiology, Medical School, University of Thessaly, Viopolis, PC 41110, Larissa, Greece