Previous Article | Next Article ![]()
Journal of Clinical Microbiology, October 2006, p. 3524-3528, Vol. 44, No. 10
0095-1137/06/$08.00+0 doi:10.1128/JCM.00558-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Department of Laboratory Medicine, University of Ulsan and Asan Medical Center,1 Department of Laboratory Medicine, Yonsei University and Severance Hospital,2 Department of Laboratory Medicine, Seoul National University Hospital,3 Department of Laboratory Medicine, Hallym University and Kangnam Sacred Heart Hospital,4 Department of Laboratory Medicine, Catholic University and Kangnam St. Mary's Hospital, Seoul,7 Department of Laboratory Medicine, Kosin University Gospel Hospital, Busan,5 Department of Laboratory Medicine, Ajou University Hospital, Suwon, Republic of Korea,6 Department of Medicine, Harvard University and Beth Israel Deaconess Medical Center, Boston, Massachusetts8
Received 15 March 2006/ Returned for modification 25 April 2006/ Accepted 31 May 2006
|
|
|---|
|
|
|---|
Although no vancomycin-resistant strains have been isolated to date, single therapy with vancomycin was associated with clinical failure in 4 of 11 cases of meningitis due to relatively penicillin-resistant pneumococcal strains (31). Vancomycin-tolerant S. pneumoniae (VTSP) has been described previously and was subsequently linked to a case of recrudescent meningitis (13). Vancomycin tolerance has been proven to play a role in therapeutic failure in an experimental rabbit model of pneumococcal meningitis (17), and patients with meningitis caused by VTSP have been shown to have a poorer survival rate than patients with meningitis caused by nontolerant strains (21). Because tolerance is a precursor to the development of resistance, these findings have important implications for the use of vancomycin for pneumococcal meningitis. The emergence of VTSP would be a serious threat in countries with significant rates of PRSP (29).
We conducted a nationwide surveillance study of antimicrobial resistance among clinical isolates of S. pneumoniae to monitor the emergence of vancomycin-tolerant strains.
|
|
|---|
Antimicrobial susceptibility. All isolates were tested for susceptibility to penicillin (Sigma-Aldrich, St. Louis, Mo.), cefotaxime (Sigma-Aldrich), meropenem (AstraZeneca, Cheshire, United Kingdom), and vancomycin (Sigma-Aldrich). MICs were determined by the broth microdilution method using cation-adjusted Mueller-Hinton broth (BBL) supplemented with 2.5% lysed horse blood (16). Interpretive criteria for susceptibility were those indicated in Clinical and Laboratory Standards Institute (formerly NCCLS) document M100-S16 (3). To confirm vancomycin tolerance, the minimum bactericidal concentration (MBC) of vancomycin was measured using the same media and incubation conditions as those for the MIC tests (15).
Vancomycin-induced lysis rates and definition of tolerance.
Vancomycin-induced lysis rates were measured as described previously (6). Briefly, two or three bacterial colonies from the culture of each isolate were inoculated into 2 ml of semisynthetic casein hydrolysate medium (12) supplemented with 0.1% yeast extract (Difco) (c + y medium) and incubated overnight at 37°C under 5% CO2. When the optical density (OD) of each culture at 620 nm reached 0.20 to 0.30 (5.6 x 107 to 2.7 x 108 CFU/ml), vancomycin was added at a concentration equivalent to 10 times the MIC. Loss of OD was measured hourly for 4 h and then at 20 h after the addition of vancomycin, and those isolates showing a <50% loss of OD were selected as the candidates for measurement of log kill. Cultures of candidate isolates were serially diluted in c + y medium before and 4 h after the addition of vancomycin and were then plated onto blood agar plates. Viable cells were counted after overnight incubation at 37°C under 5% CO2, and log kill counts were determined as log10 decreases in viable cell counts. Log kill counts were performed in duplicate for each candidate. The lysis patterns of R6 and Lyt4-4 were determined from 20 separate experiments in order to establish the tolerance limits. After 4 h of exposure to vancomycin, the mean log kill was 4.0 (standard deviation [SD], ±0.5) and the mean loss of OD was 56.9% (SD, ±10.4%) for R6, whereas the mean log kill was 1.9 (SD, ±0.3) and the mean loss of OD was 9.1% (SD, ±7.1%) for Lyt4-4. The limit that defined vancomycin tolerance was set at 2 SDs above the means for Lyt4-4 (log kill,
2.5; OD loss,
23.3%).
Sequencing and data analysis. VTSP isolates and R6 were subjected to sequencing of vncS, vex2, and pep27. The primers used for PCR and sequencing of vex2, pep27, and vncS have been described by Rodriguez et al. (21). Sequencing was performed using the ABI PRISM dye terminator cycle sequencing ready reaction kit and AmpliTaq DNA polymerase (Perkin-Elmer Applied Biosystems) on a model 373 automated DNA sequencer (Perkin-Elmer, Foster City, CA). The deduced amino acid sequences were aligned with those of reference isolates TIGR4 (GenBank accession no. NC 003028) and R6 and with those of vancomycin-tolerant strains I95, A378, and A43, published elsewhere (6).
Autolysin activity. For vancomycin-tolerant isolates, active autolysin was detected by visible clearing of the suspensions of their cultures in 2% sodium deoxycholate and by the abilities of their cellular extracts to reconstitute the penicillin-induced lysis of the autolysin-defective isolate Lyt4-4 (27).
Serotyping and PFGE. Vancomycin-tolerant isolates were serotyped by the capsular quellung method with commercial antisera (Statens Seruminstitut, Copenhagen, Denmark) as recommended by the manufacturer. The relatedness of vancomycin-tolerant isolates was determined by pulsed-field gel electrophoresis (PFGE) of chromosomal DNA restricted with SmaI and ApaI (7).
|
|
|---|
15 years old. Among all isolates, 29.3%, 62.8%, 42.6%, and 100% were susceptible to penicillin, cefotaxime, meropenem, and vancomycin, respectively. The penicillin MICs at which 50% and 90% of the isolates were inhibited were 2 µg/ml and 4 µg/ml, respectively. Penicillin-susceptible isolates were susceptible to all four drugs. But of the 100 penicillin-resistant isolates, 70 (70%), including 2 CSF isolates, were susceptible to neither cefotaxime nor meropenem. All of the isolates had vancomycin MICs ranging from 0.12 µg/ml to 1 µg/ml.
Vancomycin tolerance.
The mean loss of OD of 188 isolates at 4 h was 56.6% (SD, ±12.3%). Twenty isolates showing
50% losses of OD were counted for viable cells to confirm the vancomycin-induced lysis rate. Most of the 20 clinical isolates showed substantial losses of OD, and viable cell counts similar to those of R6. Only two isolates, S3 and H8, were consistent with the definition of vancomycin tolerance: the losses of OD for S3 and H8 were 0.3% and 19.2%, respectively, and the mean log kills were 1.3 and 1.5, respectively (Fig. 1). The MBC-to-MIC ratios of vancomycin were
64 for both (Fig. 2).
![]() View larger version (12K): [in a new window] |
FIG. 1. Mean percent loss of OD and log kill after addition of vancomycin (at a concentration equivalent to 10 times the MIC) for 20 candidate clinical isolates of pneumococci. The values for R6 and Lyt4-4 are also shown. Isolates S3 and H8 showed rates of killing and lysis similar to those of the autolysin-defective mutant Lyt4-4. Dashed lines indicate the limits of tolerance.
|
![]() View larger version (17K): [in a new window] |
FIG. 2. MBCs of vancomycin for S3, H8, and control strains ATCC 49619, R6, and Lyt4-4. Gray circles, no growth detected on 1-ml cultures.
|
Upon exposure to deoxycholate, H8 underwent rapid lysis, like R6, whereas S3 did not, like Lyt4-4. Moreover, lysates of H8, like those of R6, reconstituted the lysis activity of the autolysin-deficient strain Lyt4-4, whereas lysates of S3 did not, indicating that S3 was autolysin defective (Fig. 3).
![]() View larger version (11K): [in a new window] |
FIG. 3. Functional assay of autolytic activity of vancomycin-tolerant strains by reconstitution of penicillin-induced lysis of Lyt4-4. Crude cellular extracts of S3, H8, R6, or Lyt4-4 were added to cultures of Lyt4-4 together with 0.1 µg/ml penicillin, and then the OD at 620 nm was measured at 0 h, 8 h, and 24 h.
|
![]() View larger version (23K): [in a new window] |
FIG. 4. Amino acid sequences of the C-terminal regions of vncS (A), vex2 (B), and pep27 (C). The deduced sequences from S. pneumoniae R6, S3, and H8 were aligned with those of S. pneumoniae TIGR4 (GenBank accession no. NC 003028). Dashes represent amino acids identical to those of TIGR4. The sequences of vancomycin-tolerant isolates I95, A378, and A43 (6) were aligned for comparison of the C-terminal regions of vncS sequences.
|
|
|
|---|
To our knowledge, S3 was the first clinical isolate of autolysin-defective VTSP. Although the exact mechanism underlying vancomycin tolerance is still unclear, it is apparently related to a defect in autolysis (6, 17). The lack of autolysin alone can explain vancomycin tolerance, as exemplified by the original lytA defective mutant (27). Even though two VTSP isolates emerged from two patients independently and were not clonally related, they shared several genetic characteristics. In this study, we detected novel single-amino-acid substitutions, Q441K of vncS and N25D of vex2, in both S3 and H8. VncS-VncR, Vex, and Pep27 are involved in the signal transduction cascades to activate autolysin (LytA). VncS is a histidine kinase that acts as the sensor of a two-component regulatory system sensing the accumulation of Pep27 and subsequently leading to activation of LytA. Vex is the transporter protein inducing the secretion of Pep27 to the extracellular compartment (14, 17, 18). The V440A substitution in vncS has been proposed to cause tolerance in early VTSP isolates (6). A functional defect in any part of these signal transduction cascades hypothetically results in tolerance (18). However, loss of VncS function alone has not been proven to cause vancomycin tolerance, and the frequencies of V440A were found to be similar in tolerant and nontolerant isolates (20, 21). We did not investigate the effect of the novel amino acid substitutions found in this study. Both isolates S3 and H8 showed a combination of a TIGR4 vex2 allele and an R6 pep27 allele. This finding is consistent with the report of Rodriguez et al. that vancomycin-tolerant isolates show a significant tendency to have this combination (21). As a summary, the vancomycin tolerance found in this study may possibly be attributed to any of three kinds of mechanisms: a defect in autolysis in one isolate, a functional defect in the VncS-VncR system, and the combination of TIGR4 vex2 and R6 pep27 alleles.
Two vancomycin-tolerant strains were highly resistant to penicillin and cefotaxime. Among 188 clinical isolates, the prevalences of penicillin-intermediate and -resistant S. pneumoniae were 17.6% and 53.2%, respectively, findings similar to those of previous reports (9, 10, 25). Published cases of VTSP were all penicillin intermediate or penicillin resistant (2, 6, 8, 13). These findings are consistent with the presumption that selective pressure by antibiotics accelerates the development of tolerance and resistance (21). Due to the very high rate of ß-lactam resistance, a high risk of emergence of VTSP may be anticipated. VTSP emergence could seriously limit the therapeutic options for S. pneumoniae infection in Korea.
Because the patients from whom S3 and H8 was isolated were not being treated for pneumococcal pneumonia, the clinical impact of VTSP could not be assessed in this study. Except for a case of pneumococcal meningitis, the clinical relevance of vancomycin tolerance has not been investigated yet (12, 21). In the previous report investigating lysis-defective strains, a defect in penicillin-induced lysis was shown to affect the course of meningitis adversely, despite appropriate ß-lactam treatment, but not that of bacteremia (30). To establish the clinical relevance of VTSP, a much larger study involving more cases of VTSP infection would be required.
In conclusion, VTSP, with a prevalence of 1.1% in this study, has emerged from PRSP in Korea. Given the high incidence of penicillin resistance, the emergence of VTSP is more likely and furthermore would have a greater impact on therapeutic options for pneumococcal meningitis in Korea. Further monitoring of the prevalence, and further investigation of the clinical relevance, of VTSP is warranted.
We thank E. Tuomanen (St. Jude Children's Research Hospital, Memphis, TN) for kindly donating strains R6 and Lyt4-4 and the primer sequences for vncS, vex2, and pep27 and J. H. Shin (Chonnam National University Hospital), J. Lee (Konyang University Hospital), J. Y. Ahn (Soonchunhyang University Gumi Hospital), Y. Uh (Wonju Christian Hospital), S. G. Hong (Bundang CHA General Hospital), and S. H. Lee (NeoDIN Medical Institute) for collecting clinical isolates of S. pneumoniae. We also thank G. W. Lee (Korean Center for Disease Control and Prevention, Seoul) for serotyping.
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»