Abstract

Linezolid is an important therapeutic option for treatment of infections caused by glycopeptide- and β-lactam–resistant gram-positive organisms. Linezolid resistance is caused by mutations within the domain V region of the 23S ribosomal RNA (rRNA) gene, which is present in multiple copies in most bacteria. Among clinical Staphylococcus aureus isolates, there has been only 1 reported case of linezolid resistance. In the present study, this isolate was further characterized by determination of the number of mutant 23S rRNA copies, assessment of the stability of the resistant phenotype, and comparison of its growth characteristics with those of linezolid-susceptible S. aureus. All 5 copies of the 23S rRNA gene contained a G2576U mutation in the domain V region. After serial passage on antibiotic-free medium, the isolate maintained resistance to high concentrations of linezolid. Compared with 2 linezolid-susceptible S. aureus isolates, the linezolid-resistant S. aureus isolate demonstrated no significant differences in in vitro growth characteristics

Staphylococcus aureus coagulase-negative staphylococci, and enterococci are important causes of serious hospital-acquired infection. These organisms have shown an increasing incidence of resistance to β-lactams and glycopeptides [1, 2]. Linezolid, by virtue of its activity against these resistant gram-positive pathogens, has emerged as an important therapeutic option. Currently, linezolid is thought to disrupt protein synthesis by binding to the domain V region of 23S rRNA, targeting the ribosomal P site [3–5 ]

The frequency of spontaneous resistance to oxazolidinones among S. aureus has been shown to be <10−9 [6], and this low rate of in vitro mutation has been corroborated by the rarity with which resistant clinical isolates of S. aureus are found. The rRNA operon is present in multiple copies in nearly all bacteria, and mutations in any of a number of sites in the 23S rRNA domain V region have been associated with oxazolidinone resistance. In a study of laboratory-derived linezolid-resistant S. aureus isolates, it was demonstrated that MICs increased in proportion to the number of copies of mutant 23S rRNA genes [7]. Similarly, in clinical isolates of Enterococcus faecium an increase in the number of mutant 23S rRNA copies was shown to correlate with increasing MICs of linezolid [8]

Recently, our group reported the first clinical isolate of linezolid-resistant S. aureus [9]. This isolate (A7817) was recovered from the dialysate of an 85-year-old man undergoing peritoneal dialysis. The linezolid MIC for this methicillin-resistant S. aureus (MRSA) isolate was >32 μg/mL. Sequencing of the domain V region of the 23S rRNA gene revealed the presence of a G2576U mutation (Escherichia coli numbering), and the sequence chromatogram revealed a single peak, suggesting that all copies of the 23S rRNA gene were mutated

The present article reports the methods used to identify individual copies of the 23S rRNA gene, which could then be sequenced to confirm the presence of G2576U in all copies of the ribosomal gene. The number of mutant 23S rRNA copies was also confirmed by a novel method that took advantage of changes in the domain V region restriction enzyme map created by the G2576U mutation. The stability of the resistant phenotype was assessed. Finally, the growth characteristics for this isolate were compared with those of an unrelated MRSA isolate obtained from the same patient (before the emergence of the linezolid-resistant organism) and with S. aureus ATCC 33591

Material and Methods

Bacterial strainsThe characteristics of the strains used in this study are described in table 1

Table 1

Characteristics of strains used in a study of linezolid resistance in methicillin-resistant Staphylococcus aureus (MRSA)

Table 1

Characteristics of strains used in a study of linezolid resistance in methicillin-resistant Staphylococcus aureus (MRSA)

Primers Search of the completed S. aureus N315 genome (GenBank accession no. NC_002745) revealed 5 copies of the 23S rRNA gene. Given the highly conserved nature of the rRNA genes, each copy was aligned (MEGALIGN; DNASTAR), along with DNA flanking each gene (5 kb upstream and 4 kb downstream), to permit examination of regions of sequence variability. Unique primers to amplify the 5 individual copies were identified and optimized using OLIGO (version 5.0) primer analysis software (National Biosciences) (table 2). The third, fourth, and fifth rRNA genes (SArRNA09, SArRNA12 and SArRNA15) had an orientation opposite to that of the first and second genes (SArRNA02 and SArRNA06). All copies were placed in a similar orientation (EDITSEQ; DNASTAR) before alignment

Table 2

Individual rrn copy primers used to amplify copies of the 23S rRNA gene from a linezolid- and methicillin-resistant strain of Staphylococcus aureus

Table 2

Individual rrn copy primers used to amplify copies of the 23S rRNA gene from a linezolid- and methicillin-resistant strain of Staphylococcus aureus

Bacterial DNA extractionSingle colonies of S. aureus A7817, S. aureus A7819, and S. aureus Mu50 were inoculated in 3 mL of brain-heart infusion broth (Becton Dickinson) and incubated overnight with shaking at 37°C. Extraction of genomic DNA with guanidium thiocyanate was performed by methods described elsewhere [12]

Polymerase chain reaction (PCR) amplification of individual 23S rRNA genesIt was predicted that the selected PCR primer sets (table 2) would yield products ranging from 5.5 to 6.5 kb. Given the size of predicted fragments, rTth DNA Polymerase, XL (Applied Biosystems), was used for “long-range” PCR. The template was S. aureus A7817 genomic DNA. The PCR conditions were 1 min at 24°C; denaturation, 3 min at 95°C; 16 denaturing, annealing, and extending cycles of 15 s at 94°C, 30 s at 55°C, and 2.5 min at 72°C; 14 denaturing, annealing, and extending cycles of 15 s at 94°C, 30 s at 55°C, and 2.75 min (increasing by 15 s/cycle through cycle 30) at 72°C; and a final extension of 10 min at 72°C. PCR products were separated by agarose gel electrophoresis, gel-extracted, and purified (QIAquick gel extraction kit; Qiagen)

Domain V PCR amplificationThe domain V region that includes bp 2576 (E. coli numbering) was amplified (hereafter referred to as the “domain V PCR product”) for each of the purified rRNA gene fragments. The primers, optimized using OLIGO version 5.0, were 5′-GCGGTCGCCTCCTAAAAG-3′ (upper primer, corresponding to bases 2280–2297 of the S. aureus 23S rRNA gene; GenBank accession no. X68425) and 5′-ATCCCGGTCCTCTCGTACTA-3′ (lower primer, complementary strand corresponding to bases 2680–2699 of the S. aureus 23S rRNA gene; GenBank accession no. X68425). PCR conditions were denaturation, 5 min at 95°C; 30 denaturing, annealing, and extending cycles of 30 s at 94°C, 30 s at 55°C, and 30 s at 72°C; and a final extension of 10 min at 72°C. Gel electrophoresis was used to separate the PCR product. The 420-bp domain V PCR product was gel-extracted and purified. The purified PCR products then were sequenced using the standard dideoxynucleotide method (Molecular Biology Core Facility, Dana-Farber Cancer Center, Boston). Sequence data were analyzed using CHROMAS 1.45 (Technelysium), EDITSEQ, and MEGALIGN

Southern blot hybridizationA probe for the 23S rRNA gene was produced by amplification of the domain V region of S. aureus A7819, using the domain V region primers described above and similar PCR conditions. The PCR product was gel-extracted and purified and labeled with digoxigenin DNA labeling mix (Roche Diagnostics). Restriction maps of both the 420-bp domain V PCR product and the five 23S rRNA genes (along with flanking DNA) of S. aureus Mu50 were generated using Webcutter (Carolina Biological Supply Company; http://www.carolina.com/webcutter). Genomic DNA from S. aureus A7817 and S. aureus Mu50 was digested with EcoRI (Promega) and, in a separate set of reactions, with NheI and EcoRI (Promega). The digested genomic DNA was separated by gel electrophoresis in 0.7% agarose, run overnight at 35 V at room temperature. Gels were prepared for Southern blot transfer by methods described elsewhere [13]. After overnight transfer to a nylon membrane (Osmonics), DNA was bound to the membrane by baking for 2 h in a vacuum oven at 80°C. Membranes were prehybridized for 4 h and hybridized overnight with the domain V DNA probe, and then immunological detection was performed, according to the manufacturer’s instructions (anti–digoxigenin-AP Fab fragments, blocking reagents, nitro blue tetrazolium chloride, and 5-bromo-4-chloro-3-indolyl phosphate colorizing reagents were from Roche Diagnostics)

Stability of resistanceSingle colonies of S. aureus A7817 were serially passaged 15 times (over the course of a 2-week period) on antibiotic-free Brucella agar with 5% horse blood (Northeast Laboratory). The linezolid MICs for the original S. aureus A7817 isolate and for the serially passaged strain were determined by broth macrodilution testing in 2-mL volumes of brain-heart infusion broth, using a bacterial inoculum of ∼5×105 cfu/mL

Growth curvesSingle colonies of S. aureus A7817, S. aureus A7806, and S. aureus ATCC 33591 were inoculated into 3 mL of Mueller-Hinton II (MHII) broth (Becton Dickinson) and incubated overnight at 37°C with shaking. The following day, each overnight culture was diluted and then inoculated into 20 mL of prewarmed MHII in 250-mL Erlenmeyer flasks to yield a starting inoculum of ∼2.5×104 cfu/mL. This was incubated at 35°C without shaking. Colony counts were determined at 0, 2, 4, 6, 8, 10, and 25 h of incubation

Results

Review of S. aureus N315 gene sequences in GenBank identified 5 copies of the 23S rRNA gene. We designed PCR primers specific for each copy (table 2) and used these to amplify each copy individually. Subsequently, the domain V region of each individual copy was sequenced. All copies of the 23S rRNA gene had the mutation G2576U. We did not detect any of the other mutations previously associated with linezolid resistance, such as G2447U, G2505A, C2512U, G2513U, and C2610G [7, 14]

The restriction map of the domain V PCR product revealed no EcoRI sites. EcoRI digestion of the staphylococcal genomic DNA, followed by hybridization with the domain V probe, therefore, was used as an alternative means of quantifying the copy number of the 23S rRNA. EcoRI digestion of genomic DNA demonstrated 5 discrete bands on Southern blotting for both S. aureus A7817 (approximate molecular weights, 8.7, 7.3, 6.7, 4.1, and 2.0 kb) and S. aureus Mu50 (approximate molecular weights, 10.6, 9.4, 7.6, 6.8, and 4.4 kb) (figure 1)

Figure 1

Southern blot hybridization (using domain V probe) for Staphylococcus aureus Mu50 (lanes 1 and 2) and S. aureus A7817 (lanes 3 and 4) after digestion with EcoRI (lanes 1 and 3) and EcoRI and NheI (lanes 2 and 4)

Figure 1

Southern blot hybridization (using domain V probe) for Staphylococcus aureus Mu50 (lanes 1 and 2) and S. aureus A7817 (lanes 3 and 4) after digestion with EcoRI (lanes 1 and 3) and EcoRI and NheI (lanes 2 and 4)

The restriction map of the mutant domain V PCR product revealed that the G2576U mutation resulted in a new NheI site, which was ∼100 bp proximal to the 3′ end of the 420-bp PCR product. This NheI restriction enzyme site was not present in the wild-type gene, and there were no other NheI sites within the region complementary to the probe. Digestion with EcoRI and NheI again demonstrated 5 bands for S. aureus Mu50 (a linezolid-susceptible strain), and all bands showed comparably intense reactions to the EcoRI digestion (figure 1)

In any 23S rRNA gene that contains the G2576U mutation, the domain V DNA probe would bind to 2 fragments after digestion with NheI (figure 2). One fragment would have ∼100 bases complementary to the probe (the fragment 3′ to the G2576U mutation), and the other fragment would have ∼320 bases complementary to the probe (5′ to the G2576U mutation). Furthermore, the restriction map of the 5 S. aureus Mu50 23S rRNA genes demonstrated that the NheI restriction site that was closest to the region complementary to the domain V DNA probe was ∼110 bp upstream of the 5′ end of the probe region. Digestion of the mutant gene copies with NheI resulted in a fragment of ∼430 bp (figure 1, lane 4). This represented the distance from the NheI site 5′ to the region complementary to the domain V DNA probe to the G2576U mutation site. The 5 larger, fainter bands (figure 1, lane 4; approximate molecular weights, 8.0, 6.5, 5.6, 2.0, and 0.7 kb) represented the DNA fragments containing the remaining 100 bases complementary to the domain V DNA probe. Because 430-bp fragment was complementary to a greater number of labeled bases in the domain V DNA probe, it displayed a more intense reaction than did fragments that were complementary to the remaining 100 bases of the probe. These findings were consistent for all 5 copies of the 23S rRNA gene containing the G2576U mutation

Figure 2

Schematic diagram of a 23S rRNA gene with flanking DNA based on the restriction map of Staphylococcus aureus Mu50. The G2576U mutation, with the resultant NheI restriction site, and its relationship to the domain V region probe is outlined. The closest EcoRI and NheI restriction sites and their approximate distances from the G2576U mutation are diagrammed. The EcoRI restriction site 3′ to the domain V region probe is at a variable distance from the G2576U mutation for each of the five 23S rRNA genes

Figure 2

Schematic diagram of a 23S rRNA gene with flanking DNA based on the restriction map of Staphylococcus aureus Mu50. The G2576U mutation, with the resultant NheI restriction site, and its relationship to the domain V region probe is outlined. The closest EcoRI and NheI restriction sites and their approximate distances from the G2576U mutation are diagrammed. The EcoRI restriction site 3′ to the domain V region probe is at a variable distance from the G2576U mutation for each of the five 23S rRNA genes

After 15 passages of S. aureus A7817 on antibiotic-free medium, macrobroth MIC dilution testing demonstrated that the organism remained resistant to linezolid, with an MIC of 64 μg/mL, which was no different from that of the original isolate

Growth characteristics for S. aureus A7817 were compared with those of 2 other MRSA isolates, S. aureus A7806 and S. aureus ATCC 33591 (figure 3). Over the course of 24 h, growth of S. aureus A7817 was comparable to growth of both S. aureus A7806 and S. aureus ATCC 33591

Figure 3

Growth of linezolid- and methicillin-resistant Staphylococcus aureus strain A7817 and linezolid-susceptible strains S. aureus A7806 and S. aureus ATCC 33591 in Mueller-Hinton II broth

Figure 3

Growth of linezolid- and methicillin-resistant Staphylococcus aureus strain A7817 and linezolid-susceptible strains S. aureus A7806 and S. aureus ATCC 33591 in Mueller-Hinton II broth

Discussion

A number of mutations in the domain V region of the 23S rRNA gene that are associated with oxazolidinone resistance have been described. Linezolid-resistant bacterial isolates generated in the laboratory exhibited the following mutations: in S. aureus G2447U [7]; in E. faecium G2505A; in Enterococcus faecalis G2576U, C2512U, G2513U, and C2610G [14]; and in E. coli G2032A [15]. In the clinical setting, only the G2576U mutation has been described, both in E. faecium [8, 16] and in the strain of S. aureus initially reported by Tsiodras et al. [9], which we have confirmed and further characterized in the present study. This suggests that, although other mutations may occur on in vitro exposure to linezolid, the emergence of G2576U appears to be favored in the clinical setting in both S. aureus and E. faecium

The ∼400-bp intragenic sequence of the 23S rRNA gene that was amplified and sequenced in the present study included the sites of all of mutations that have been described elsewhere, except for G2032A (which has only been reported in E. coli). Therefore, although additional mutations cannot be excluded, we found no evidence that 5 other mutations exist in the domain V region of the 23S rRNA gene reported from laboratory-generated linezolid-resistant strains

By individually sequencing the 23S rRNA genes, we detected 5 copies of the 23S rRNA gene and demonstrated that all 5 copies were mutated in S. aureus A7817. Some studies have described 6 copies of the 23S rRNA gene in certain strains of S. aureus and 5 copies in other S. aureus strains [17, 18]. Review of the GenBank sequences of both S. aureus N315 and S. aureus Mu50 (GenBank accession nos. NC_002745 and NC_002758, respectively) revealed only 5 copies of the 23S rRNA gene. By EcoRI digestion of S. aureus A7817 and S. aureus Mu50, followed by Southern blot hybridization, we demonstrated 5 copies of the 23S rRNA gene in both of these isolates. Furthermore, by taking advantage of the NheI restriction site created by the G2576U mutation, we verified by Southern blot hybridization that all 5 copies of the 23S rRNA gene present in S. aureus A7817 were mutated. Given that the G2576U mutation is the only mutation in the domain V region described as conferring linezolid-resistance among clinical bacterial isolates and that the 23S rRNA gene is highly conserved among bacteria, this restriction enzyme method may be useful in determining the number of mutant copies in other linezolid-resistant bacterial species for which a complete genomic sequence is not available

Kaatz and Seo [6] selected for MRSA isolates that had increased linezolid and eperezolid MICs when they were serially passaged on gradient plates containing each antibiotic. They noted that the increased MICs were stable for up to 3 passages on antibiotic-free medium. Similarly, Murray et al. [19] generated an eperezolid-resistant mutant of S. aureus ATCC 29213 and noted that the resistant phenotype was maintained without any oxazolidinone selection. Prystowsky et al. [14] selected for linezolid resistance among 10 clinical isolates of vancomycin-resistant enterococci by serial passage on successively higher concentrations of linezolid. They assessed the stability of the resistant phenotype by subculturing the isolates on antibiotic-free agar twice weekly for 1 month. Only 1 isolate showed a decrease in MIC, and the decrease reflected only a 2-fold change from the MIC that was measured before passage on the antibiotic-free plates. Our results demonstrating that the linezolid MIC of S. aureus A7817 was unchanged after 15 passages on antibiotic-free medium are consistent with these observations

In our initial publication describing the index patient, S. aureus A7806 (linezolid susceptible) was cultured from peritoneal dialysate before the isolation of the genetically unrelated linezolid-resistant S. aureus A7817 [9]. One hypothesis was that the index patient carried both S. aureus A7817 and S. aureus A7806 but that S. aureus A7806 had more favorable growth characteristics and antibiotic selection pressure permitted the emergence of the previously undetected linezolid-resistant clone. However, the linezolid-resistant organism demonstrated in vitro growth at a rate and to a density similar to those of the comparison strains. This suggested that the G2576U mutation did not significantly impair the metabolic function of the organism and, thus, did not provide any selective disadvantage

Development of linezolid resistance in staphylococci appears to be an infrequent event both in the laboratory [6] and, by inference from the lack of reported cases in the literature, in clinical practice. This infrequency may reflect the redundant nature of the 23S rRNA gene. However, despite the redundancy, we have shown that, in our clinical isolate of a linezolid-resistant S. aureus all 5 copies of the 23S rRNA gene contained the G2576U mutation. The stability of the linezolid-resistant phenotype was confirmed by the maintenance of high-level resistance despite serial passage on antibiotic-free medium. No significant difference in growth characteristics was noted when the linezolid-resistant isolate was compared to linezolid-susceptible S. aureus. Thus, although the generation of linezolid resistance may be a rare event, once it has occurred, there does not appear to be impairment of the organism’s metabolic function, and further dissemination of the resistant organism could occur without continued antibiotic selective pressure. This seems to be true for vancomycin-resistant enterococci; as noted by Herrero et al. [16], once linezolid- and vancomycin-resistant enterococci have been selected for by antibiotic pressure, the organisms spread to patients who were never exposed to linezolid. Judicious use of linezolid may help to reduce the further development and spread of resistance among S. aureus

References

1.
National Nosocomial Infections Surveillance (NNIS) system report, data summary from January 1992–June 2001, issued August 2001
Am J Infect Control
 , 
2001
, vol. 
29
 (pg. 
404
-
21
)
2.
Diekema
DJ
Jones
RN
Oxazolidinone antibiotics
Lancet
 , 
2001
, vol. 
358
 (pg. 
1975
-
82
)
3.
Aoki
H
Ke
L
Poppe
SM
, et al.  . 
Oxazolidinone antibiotics target the P site on Escherichia coli ribosomes
Antimicrob Agents Chemother
 , 
2002
, vol. 
46
 (pg. 
1080
-
5
)
4.
Kloss
P
Xiong
L
Shinabarger
DL
Mankin
AS
Resistance mutations in 23 S rRNA identify the site of action of the protein synthesis inhibitor linezolid in the ribosomal peptidyl transferase center
J Mol Biol
 , 
1999
, vol. 
294
 (pg. 
93
-
101
)
5.
Shinabarger
DL
Mechanism of action of the oxazolidinone antibacterial agents
Expert Opin Investig Drugs
 , 
1999
, vol. 
8
 (pg. 
1195
-
202
)
6.
Kaatz
GW
Seo
SM
In vitro activities of oxazolidinone compounds U100592 and U100766 against Staphylococcus aureus and Staphylococcus epidermidis
Antimicrob Agents Chemother
 , 
1996
, vol. 
40
 (pg. 
799
-
801
)
7.
Swaney
SM
Shinabarger
DL
Schaadt
RD
Bock
JH
Slightom
JL
Zurenko
GE
Oxazolidinone resistance is associated with a mutation in the peptidyl transferase region of 23S rRNA [abstract C‐104]
Abstracts of the 38th Interscience Conference on Antimicrobial Agents and Chemotherapy (San Diego, 1998)
 , 
1998
Washington, DC
American Society of Microbiology
(pg. 
98
-
9
)
8.
Zurenko
GE
Todd
WM
Hafkin
B
, et al.  . 
Development of linezolid‐resistant Enterococcus faecium in two compassionate use program patients treated with linezolid [abstract 848]
Abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy (San Francisco, 1999)
 , 
1999
Washington, DC
American Society of Microbiology
pg. 
118
 
9.
Tsiodras
S
Gold
HS
Sakoulas
G
, et al.  . 
Linezolid resistance in a clinical isolate of Staphylococcus aureus
Lancet
 , 
2001
, vol. 
358
 (pg. 
207
-
8
)
10.
Kuroda
M
Ohta
T
Uchiyama
I
, et al.  . 
Whole genome sequencing of methicillin‐resistant Staphylococcus aureus
Lancet
 , 
2001
, vol. 
357
 (pg. 
1225
-
40
)
11.
Schaefler
S
Perry
W
Jones
D
Methicillin‐resistant strains of Staphylococcus aureus phage type 92
Antimicrob Agents Chemother
 , 
1979
, vol. 
15
 (pg. 
74
-
80
)
12.
Pitcher
DG
Saunders
NA
Owen
RJ
Rapid extraction of bacterial genomic DNA with guanidium thiocyanate
Lett Appl Microbiol
 , 
1989
, vol. 
8
 (pg. 
151
-
6
)
13.
Ausubel
FM
Brent
R
Kingston
RE
, et al.  . 
Short protocols in molecular biology. 3rd ed
 , 
1997
New York
John Wiley & Sons
14.
Prystowsky
J
Siddiqui
F
Chosay
J
, et al.  . 
Resistance to linezolid: characterization of mutations in rRNA and comparison of their occurrences in vancomycin‐resistant enterococci
Antimicrob Agents Chemother
 , 
2001
, vol. 
45
 (pg. 
2154
-
6
)
15.
Xiong
L
Kloss
P
Douthwaite
S
, et al.  . 
Oxazolidinone resistance mutations in 23S rRNA of Escherichia coli reveal the central region of domain V as the primary site of drug action
J Bacteriol
 , 
2000
, vol. 
182
 (pg. 
5325
-
31
)
16.
Herrero
IA
Issa
NC
Patel
R
Nosocomial spread of linezolid‐resistant, vancomycin‐resistant Enterococcus faecium
N Engl J Med
 , 
2002
, vol. 
346
 (pg. 
867
-
9
)
17.
Goering
RV
Black
T
Hare
R
Nolling
J
A comparison of ribosomal RNA operon distribution in methicillin‐resistant Staphylococcus aureus (MRSA) strains currently in genomic sequence analysis [abstract C2‐300]
Abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy (Chicago, 2001)
 , 
2001
Washington, DC
American Society of Microbiology
pg. 
119
 
18.
Wada
A
Ohta
H
Kulthanan
K
Hiramatsu
K
Molecular cloning and mapping of 16S–23S rRNA gene complexes of Staphylococcus aureus
J Bacteriol
 , 
1993
, vol. 
175
 (pg. 
7483
-
7
)
19.
Murray
RW
Schaadt
RD
Zurenko
GE
Marotti
KR
Ribosomes from an oxazolidinone‐resistant mutant confer resistance to eperezolid in a Staphylococcus aureus cell‐free transcription‐translation assay
Antimicrob Agents Chemother
 , 
1998
, vol. 
42
 (pg. 
947
-
50
)