A comparative in-vitro evaluation of resistance selection after exposure to teicoplanin, vancomycin, linezolid and quinupristin–dalfopristin in Staphylococcus aureus and Enterococcus spp.

Authors


Corresponding author and reprint requests: L. Drago, Laboratory of Clinical Microbiology, Department of Preclinical Science, LITA Vialba, Via GB Grassi 74, 20157 Milan, Italy
E-mail: lorenzo.drago@unimi.it

Abstract

The ability of breakpoint and serum concentrations of teicoplanin, vancomycin, linezolid and quinupristin–dalfopristin to select resistance was compared for isolates of methicillin-susceptible Staphylococcus aureus (MSSA), methicillin-resistant S. aureus (MRSA), Enterococcus faecalis and Enterococcus faecium. Mutation frequencies were always <10−10, except for two isolates grown in the presence of teicoplanin at the trough serum concentration. After multistep selection, linezolid selected for resistance in staphylococci and enterococci, and serial exposure to certain concentrations of linezolid was more likely to select for stable resistance in MRSA, MSSA and enterococci than was exposure to glycopeptides and quinupristin–dalfopristin.

For many years, glycopeptides were considered to be the cornerstone of therapy for serious methicillin-resistant staphylococcal infections [1]. However, the spread of vancomycin-resistant enterococci is now a concern in hospitals worldwide, with vancomycin-resistant enterococci having become established nosocomial pathogens [2,3]. Following the first report of failure of vancomycin therapy because of a vancomycin-intermediate-resistant Staphylococcus aureus strain in 1997 [4], an increasing number of isolates with decreased susceptibility to glycopeptides has been identified worldwide [5,6]. Moreover, shortly after the introduction of linezolid and quinupristin–dalfopristin, both of which were designed specifically to treat infections caused by Gram-positive bacteria, occasional resistance to both drugs has been reported in clinical isolates of Gram-positive cocci [7–9]. The present study investigated the ability of teicoplanin (Sanofi-Aventis, Milan, Italy), vancomycin (Sigma-Aldrich, St Louis, MO, USA), linezolid (Pfizer, Rome, Italy) and quinupristin–dalfopristin (Sanofi-Aventis) to select for resistance in 20 clinical isolates each of methicillin-resistant S. aureus (MRSA), methicillin-susceptible S. aureus, Enterococcus faecalis and Enterococcus faecium.

Drugs were evaluated at the following concentrations, equal to the resistance breakpoint (BP) and the peak and minimum serum concentrations (Cmax and Cmin, respectively): teicoplanin 32 mg/L (BP), 43.2 mg/L (Cmax), 10 mg/L (Cmin); vancomycin 32 mg/L (BP), 66 mg/L (Cmax), 8 mg/L (Cmin); quinupristin–dalfopristin 4 mg/L (BP), 9.50 mg/L (Cmax); and linezolid 8 mg/L (BP), 15.7 mg/L (Cmax), 3.84 mg/L (Cmin) [10–14]. No data concerning Cmin were available for quinupristin–dalfopristin. The frequency of mutation (number of colonies growing on plates containing antibiotic divided by the total inoculum) was determined by spreading 0.1 mL from a bacterial suspension containing c. 1011 CFU/mL on plates containing each antibiotic at BP, Cmax and Cmin [15]. MICs for colonies grown on antibiotic-containing plates were determined by microdilution according to CLSI recommendations [10].

Selection for resistance was evaluated by serially subculturing bacteria on agar plates containing linear antibiotic concentrations ranging from 0 mg/L to the BP, Cmax and Cmin [16]. An inoculum of 1011 CFU/mL was spread homogeneously on each plate and incubated for 48 h at 37°C. Colonies growing at the highest antibiotic concentration were sampled, grown overnight in antibiotic-free broth, re-inoculated ten times on new antibiotic gradient plates, and then ten times on antibiotic-free plates to evaluate the stability of acquired resistance. MIC values were determined after the first, the fifth and the final passage on antibiotic-containing agar, and after the first, the fifth and the final passage on antibiotic-free agar.

The potential for antibiotics to promote increased rates of mutation has clear implications for the evolution of antibiotic resistance among bacterial pathogens [17]. Although antimicrobial therapy aims to target an infecting organism with lethal or inhibitory concentrations, antibiotic concentrations often fall below the inhibitory level in vivo, or may fail to reach an inhibitory level in some body compartments. For this reason, selection of resistance was evaluated at concentrations commonly achieved during therapy. Frequencies of mutation were low for all the antibiotics tested (<10−10); only two MRSA isolates were able to grow on plates containing teicoplanin at the trough concentration, so that the frequency of mutation ranged from 1.7 × 10−8 to <10−10. However, despite growth, the MICs for these two selected isolates (1 mg/L) did not exceed the resistance breakpoint for teicoplanin.

The use of antibiotic gradients in the multistep selection assay allowed the impact of a wide range of concentrations on resistance development to be explored. The study revealed that Cmax did not select for resistance in staphylococci and E. faecium, even if increases in MICs of teicoplanin and quinupristin–dalfopristin were revealed (Tables 1 and 2). In contrast, high concentrations of linezolid were unable to prevent selection of resistance in E. faecalis, while no changes in susceptibility were observed for glycopeptides (Table 2).

Table 1.   MIC values for Staphylococcus aureus before and after serial passage on antibiotic-gradient agar plates and on antibiotic-free plates
DrugConcentrationaMIC range (mg/L)/no. resistant mutants
BaselineOne stepFive stepsTen stepsTen steps freeb
  1. MRSA, methicillin-resistant S. aureus; MSSA, methicillin-susceptible S. aureus; BP, resistance breakpoint; Cmax, peak serum concentration; Cmin, trough serum concentration.

  2. aValues indicate the highest concentrations in the gradient plates.

  3. bSubcultures on antibiotic-free agar.

MRSA
 TeicoplaninBP (32 mg/L)
Cmax (43.2 mg/L)
Cmin (10 mg/L)
0.125–2
0.125–2
0.125–2
0.25–8/0
0.5–4/0
1–8/0
0.5–8/0
2–8/0
1–8/0
8–32/3
8–16/0
4–32/7
1–16/0
2–16/0
1–16/0
 VancomycinBP (32 mg/L)
Cmax (66 mg/L)
Cmin (8 mg/L)
0.5–1
0.5–1
0.5–1
0.5–2/0
0.5–1/0
0.5–2/0
0.5–4/0
0.5–1/0
0.5–4/0
0.5–32/1
0.5–1/0
2–32/4
0.5–8/0
0.5–1/0
1–8/0
 LinezolidBP (8 mg/L)
Cmax (15.7 mg/L)
Cmin (3.84 mg/L)
1–2
1–2
1–2
1–4/0
1–2/0
1–4/0
1–64/4
1–2/0
1–8/2
2–64/6
1–2/0
2–8/5
2–32/6
1–2/0
2–8/3
 Quinupristin–dalfopristinBP (4 mg/L)
Cmax (9.5 mg/L)
0.125–0.5
0.125–0.5
0.25–1/0
0.125–0.5/0
0.25–1/0
0.5–1/0
1–2/0
0.5–4/0
0.25–1/0
0.25–4/0
MSSA
 TeicoplaninBP (32 mg/L)
Cmax (43.2 mg/L)
Cmin (10 mg/L)
0.125–0.5
0.125–0.5
0.125–0.5
0.125–2/0
0.25–2/0
0.5–8/0
0.125–8/0
0.25–8/0
2–16/0
0.125–16/0
0.25–16/0
2–16/0
0.125–4/0
0.25–8/0
1–8/0
 VancomycinBP (32 mg/L)
Cmax (66 mg/L)
Cmin (8 mg/L)
0.125–0.5
0.125–0.5
0.125–0.5
0.125–0.5/0
0.125–0.5/0
0.125–1/0
0.125–2/0
0.125–0.5/0
1–4/0
0.125–4/0
0.125–0.5/0
1–4/0
0.125–1/0
0.125–0.5/0
0.125–1/0
 LinezolidBP (8 mg/L)
Cmax (15.7 mg/L)
Cmin (3.84 mg/L)
0.5–2
0.5–2
0.5–2
2–4/0
0.5–2/0
2–4/0
4–8/4
0.5–2/0
2–16/6
4–32/6
1–2/0
4–32/9
4–16/4
0.5–2/0
4–32/4
 Quinupristin–dalfopristinBP (4 mg/L)
Cmax (9.5 mg/L)
0.06–0.5
0.06–0.5
0.125–0.5/0
0.06–0.5/0
0.25–0.5/0
0.06–0.5/0
0.125–0.5/0
0.06–0.5/0
0.125–0.5/0
0.06–0.5/0
Table 2.   MIC values for enterococci before and after serial passage on antibiotic-gradient agar plates and on antibiotic-free plates
DrugConcentrationaMIC range (mg/L)/no. of resistant mutants
BaselineOne stepFive stepsTen stepsTen steps freeb
  1. BP, resistance breakpoint; Cmax, peak serum concentration; Cmin, trough serum concentration.

  2. aValues indicate the highest concentrations in the gradient plates.

  3. bSubcultures on antibiotic-free agar.

Enterococcus faecalis
 TeicoplaninBP (32 mg/L)
Cmax (43.2 mg/L)
Cmin (10 mg/L)
0.03–4
0.03–4
0.03–4
0.03–8/0
0.03–4/0
0.03–4/0
0.03–8/0
0.03–4/0
0.03–4/0
0.03–8/0
0.03–4/0
0.03–8/0
0.03–8/0
0.03–4/0
0.03–4/0
 VancomycinBP (32 mg/L)
Cmax (66 mg/L)
Cmin (8 mg/L)
0.5–2
0.5–2
0.5–2
1–2/0
0.5–2/0
1–2/0
1–4/0
0.5–2/0
1–4/0
1–8/0
0.5–2/0
2–8/0
1–8/0
0.5–2/0
2–8/0
 LinezolidBP (8 mg/L)
Cmax (15.7 mg/L)
Cmin (3.84 mg/L)
1–2
1–2
1–2
1–2/0
1–4/0
1–2/0
4–32/13
2–32/14
2–32/14
4–32/18
2–64/17
2–64/17
2–32/16
4–32/15
2–64/16
Enterococcus faecium
 TeicoplaninBP (32 mg/L)
Cmax (43.2 mg/L)
Cmin (10 mg/L)
0.125–0.5
0.125–0.5
0.125–0.5
0.125–0.5/0
0.125–0.5/0
0.25–0.5/0
0.125–0.5/0
0.125–0.5/0
1–2/0
0.125–0.5/0
0.125–0.5/0
0.5–4/0
0.125–0.5/0
0.125–0.5/0
0.125–2/0
 VancomycinBP (32 mg/L)
Cmax (66 mg/L)
Cmin (8 mg/L)
0.5–1
0.5–1
0.5–1
0.5–1/0
0.5–1/0
0.5–1/0
0.5–1/0
0.5–1/0
1–2/0
0.5–1/0
0.5–1/0
1–2/0
0.5–1/0
0.5–1/0
1–2/0
 LinezolidBP (8 mg/L)
Cmax (15.7 mg/L)
Cmin (3.84 mg/L)
1–2
1–2
1–2
2–4/0
1–2/0
1–2/0
2–8/3
1–4/0
2–8/7
2–8/4
1–8/2
4–16/15
2–8/0
1–4/0
2–16/11
 Quinupristin–dalfopristinBP (4 mg/L)
Cmax (9.5 mg/L)
0.25–1
0.25–1
0.25–1/0
0.5–1/0
1–2/0
1–2/0
2–4/3
1–4/2
0.5–2/0
0.5–2/0

Differences in favour of glycopeptides and quinupristin–dalfopristin with respect to linezolid were also observed with BP or Cmin concentrations. Although all the antibiotics selected for an incremental increase in MICs, only the mutants selected by linezolid were stably resistant (Tables 1 and 2). Differences in stability of resistance between glycopeptides and linezolid could be caused by a different degree of fitness in resistant mutants. Indeed, the fitness burden, associated with alterations in cell-wall morphology, is thought to account for the widely observed instability of the resistant phenotype [18].

In 2006, vancomycin breakpoints for S. aureus were changed (from ≤4 to ≤2 mg/mL for ‘susceptible’, from 8–16 to 4–8 mg/mL for ‘intermediate’, and from ≥32 to ≥16 mg/mL for ‘resistant’) to enhance the detection of heterogeneously resistant isolates of S. aureus [19]. According to the new interpretative criteria, four and six MRSA isolates should be reclassified as resistant at the BP and Cmin concentrations, respectively.

A limitation of the present study was that total concentrations of antimicrobial agents were assessed, without any consideration of protein-binding. Since protein-binding varies widely among the agents tested, it might affect the results obtained, particularly for teicoplanin, which is associated with high serum protein-binding. However, the high level of protein-binding may act as an ‘antibiotic reservoir’ that reversibly releases antibiotics from the binding sites as serum and tissue concentrations decrease, provided that serum concentrations are sufficiently high to guarantee effective levels in the blood [20].

In conclusion, these in-vitro studies with concentrations of glycopeptides, quinupristin–dalfopristin and linezolid achievable in vivo suggest that optimal antimicrobial peak concentrations are able to prevent, or at least to limit, the occurrence of resistance. Selecting dosing regimens on the basis of pharmacokinetic and pharmacodynamic properties is of utmost importance in preventing the emergence of resistance. Exposure in vitro to certain concentrations of linezolid seems more likely to select for stable resistance among isolates of MRSA, methicillin-susceptible S. aureus and enterococci than is exposure to glycopeptides and quinupristin–dalfopristin. Care should be taken to use appropriate prescribing of these new agents for the treatment of patients with infections caused by Gram-positive bacteria.

Acknowledgements

These data were presented, in part, at the 16th European Congress of Clinical Microbiology and Infectious Diseases (Nice, 2006). We would like to thank P. Henshaw for assistance in the preparation of this manuscript. This study was supported by a financial grant from Sanofi-Aventis, Milan, Italy. LD has received research funding and has served as a speaker for Sanofi-Aventis. LN and EDV declare that they have no conflicts of interest in relation to this article.

Ancillary