Abstract
The aim of this study was to investigate the molecular mechanisms of ertapenem resistance among Enterobacteriaceae isolates in a clinical microbiology laboratory at a tertiary university hospital. A total of 40 clinical isolates including 20 resistant and 20 intermediate isolates were collected from August 2012 to July 2013. Ertapenem susceptibility was confirmed by the broth microdilution method. PCR and sequencing analysis of carbapenemase, AmpC β-lactamase, and extended-spectrum β-lactamase (ESBL) genes were performed. Outer membrane proteins (OMPs) were examined by urea-sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Molecular epidemiology studies were performed by pulsed-field gel electrophoresis (PFGE). AmpC β-lactamases and ESBLs were found in 32 (80.0%) and 20 (50.0%) of the 40 isolates with ertapenem non-susceptibility, respectively. Distributions of β-lactamase genes differed among the species. One Citrobacter freundii isolate among the 40 isolates with ertapenem non-susceptibility carrying the blaIMP-1 associated class 1 integron was detected. SDS-PAGE of OMPs showed altered or greatly diminished expression of porins in all isolates of Klebsiella pneumoniae (n=5) and Enterobacter cloacae (n=11) with ertapenem resistance. Porin alterations were less common among the isolates with intermediate susceptibility (4/19). Integration of the results of molecular analysis of β-lactamases and OMP analysis revealed that most of the isolates with ertapenem resistance exhibited β-lactamase activity and porin alteration. PFGE revealed that most isolates were epidemiologically unrelated. Ertapenem resistance in clinical Enterobacteriaceae isolates was associated with β-lactamase activity and porin alteration. Even though carbapenemase-producing Enterobacteriaceae are still rare, continuous monitoring and infection control for carbapenem-resistant Enterobacteriaceae are necessary.
Significance of this study
What is already known about this subject?
There are few data on ertapenem resistance among Enterobacteriaceae isolated from clinical microbiology laboratories.
Ertapenem resistance in Enterobacteriaceae is known to be primarily caused by mechanisms other than carbapenemases.
Ertapenem-resistant Enterobacteriaceae infections are associated with poor prognosis when compared to ertapenem-susceptible Enterobacteriaceae infections.
What are the new findings?
The resistance rate to ertapenem among all Enterobacteriaceae isolates was 2.0% as determined using VITEK2. It was highest in Enterobacter spp. (14.0%), and was 2.5% and 0.4% in Klebsiella pneumoniae and Escherichia coli, respectively. In comparison, the resistance rate to imipenem was 0.1%.
Most of the ertapenem-resistant isolates had porin alteration. A combination of β-lactamase activity and porin alteration tended to result in higher MICs of ertapenem compared to isolates with β-lactamase activity alone. This suggests that ertapenem resistance in Enterobacteriaceae might result from the accumulation of multiple carbapenem-resistance determinants.
PFGE revealed genetic diversity among most of the isolates, suggesting that ertapenem resistance in clinical Enterobacteriaceae isolates was reflected independent of its emergence in different strains. Nonetheless, the phenomenon of the same clone spreading in the same ward during a single period was observed in this study.
How might these results change the focus of research or clinical practice?
It provides useful information about ertapenem resistance and would be helpful for treatment and infection control of erpenem-resistant Enterobacteriaceae.
Introduction
Ertapenem has been widely used since the early 2000s, but it was only recently added to the routine antimicrobial susceptibility tests for Enterobacteriaceae performed by automated systems. Since then, ertapenem-resistant Enterobacteriaceae isolates have been detected. Even though some reports of worldwide surveillance have been published,1–4 there are few data on ertapenem resistance among Enterobacteriaceae isolated from clinical microbiology laboratories.5 ,6
Ertapenem resistance in Enterobacteriaceae is known to be primarily caused by mechanisms other than carbapenemases, the most common expression of β-lactamases such as an AmpC β-lactamase or an extended-spectrum β-lactamase (ESBL) combined with porin loss.1 ,2 ,4 ,5 ,7–9 In general, carbapenems are stable to β-lactamases, but this stability varies between agents, and ertapenem appears to be less stable than other carbapenems: the MICs of ertapenem increased with up to three or four doubling dilutions in Enterobacteriaceae isolates producing ESBLs or AmpC β-lactamases, whereas the MICs of other carbapenems, such as imipenem and meropenem, changed within one or two doubling dilutions.10–12 Imipenem and meropenem often remain moderately active against isolates with low-level ertapenem resistance.5–9 ,13 Ertapenem-resistant Enterobacteriaceae infections are associated with higher mortality rates and poor clinical response rates when compared to ertapenem-susceptible Enterobacteriaceae infections.14 Understanding the underlying resistance mechanisms is important for treatment and infection control of ertapenem-resistant Enterobacteriaceae.
In this study, we investigated the molecular mechanisms of ertapenem resistance for Enterobacteriaceae isolates in a clinical microbiology laboratory at a tertiary university hospital.
Materials and methods
Isolates
Clinical Enterobacteriaceae isolates were collected from the clinical microbiology laboratory of Ewha Womans University Mokdong Hospital. All of the isolates were resistant to ertapenem using the VITEK2 system between August 2012 and July 2013. The resistance rate to ertapenem among all clinical Enterobacteriaceae isolates was 2.0% (97/4876): 14.0%, 2.5%, and 0.4% in Enterobacter spp., Klebsiella pneumoniae, and Escherichia coli, respectively. A total of 72 Enterobacteriaceae isolates were collected. Among them, 40 clinical isolates, including 20 resistant and 20 intermediate isolates as confirmed by broth microdilution, were analyzed: 26 (65.0%) Enterobacter spp., 6 (15.0%) K. pneumoniae, 5 (12.5%) E. coli, 1 (2.5%) Citrobacter freundii, 1 (2.5%) Providencia rettgeri, and 1 (2.5%) Cronobacter sakazakii. Isolates were collected from various specimens. The surveillance study was not performed; this study was studied retrospectively, so the molecular epidemiological relatedness of the isolates had not been recognized at that time. Bacterial identification was performed using VITEK2 with GN card (bioMérieux Inc., Durham, North Carolina, USA). Clinical features of patients were reviewed through electronic medical records. This study was approved by the Institutional Review Board of Ewha Womans University Mokdong Hospital.
Antimicrobial susceptibility test
Antimicrobial susceptibility test for various agents was performed using VITEK2 with AST-N224 card (bioMérieux Inc.), according to the manufacturer's instructions. Tested antimicrobial agents included ertapenem, imipenem, ampicillin, amoxicillin/clavulanic acid, piperacillin/tazobactam, cefazolin, cefotaxime, ceftazidime, cefepime, cefoxitin, aztreonam, amikacin, gentamicin, ciprofloxacin, and trimethoprim/sulfamethoxazole. The antimicrobial susceptibility test for the newer antimicrobial agent was not included. The meropenem susceptibility test was performed using MicroScan (Siemens Healthcare Diagnostics Inc., West Sacramento, California, USA).
The broth microdilution method was performed using 96-well broth microdilution panels according to the Clinical and Laboratory Standards Institute (CLSI) guidelines.15 ,16 The solvent and diluent for preparation of stock solutions of ertapenem were prepared in the cation-adjusted Mueller-Hinton broth (Becton Dickinson, Sparks, Maryland, USA) as described in the CLSI document. The broth microdilution MIC test range was 0.25 μg/mL to 128 μg/mL. Susceptibility results were interpreted using the CLSI guideline recommended in 2013 as follows: ≤0.5 μg/mL for susceptible, 1 μg/mL for intermediate, and ≥2 μg/mL for resistant.15 Enterococcus faecalis ATCC 29212, E. coli ATCC 25922, and Pseudomonas aeruginosa ATCC 27853 were used as controls.
Investigation of β-lactamases
Carbapenemase and ampC β-lactamase screening tests
For carbapenemase detection, a modified Hodge test and a carbapenemase inhibition test (Rosco Diagnostica, Taastrup, Denmark) were performed. Chromogenic agar for screening of carbapenemase-producing isolates was also used (bioMérieux Inc). AmpC β-lactamase detection was performed by a cefoxitin-boronic acid disk synergy test.17 Previously confirmed positive isolates were used as controls.
Molecular analysis of β-lactamase genes
Detection of genes coding for carbapenemases (IMP, SPM, AIM, VIM, OXA, GIM, BIC, SIM, NDM, DIM, and KPC) and AmpC β-lactamases (CMY-1-like, CMY-2-like, DHA-1/2, and MIR-1T/ACT-1) was performed by multiplex PCR as described previously18 ,19 with some modification. An additional primer pair was used to amplify the chromosomal ampC gene from E. coli.20 Detection of genes coding for ESBL (SHV, TEM, CTX-M-1 group, CTX-M-2 group, CTX-M-8 group, and CTX-M-9 group) was performed by PCR as described previously.21 ,22 Carbapenemase in C. freundii was confirmed by PCR and sequencing analysis using primers targeting the blaIMP-1 gene and class 1 integron.23 PCR was performed with PreMix (Bioneer, Daejeon, Korea) containing 1 U of Taq DNA polymerase in a total volume of 20 μL. The total amount of the DNA template in each reaction tube was adjusted to between 50 and 100 ng except for multiplex PCR of the carbapenemase gene, which used 10 ng of the DNA template. The typical PCR program consisted of an initial denaturation step at 94°C for 5 min, followed by 35 cycles of DNA denaturation at 94°C for 30 s, primer annealing at 50°C for 40 s, and primer extension at 72°C for 1 min. Annealing temperatures in multiplex PCR of carbapenemase and AmpC β-lactamase genes were increased to 57°C and 64°C, respectively, to increase stringency. After the last cycle, a final extension step at 72°C for 7 min was performed in all reactions. Five-microliter aliquots of the PCR product were analyzed by gel electrophoresis in 2% agarose. Previously confirmed positive isolates were used as controls.
Outer membrane protein analysis using sodium dodecyl sulfate-polyacrylamide gel electrophoresis
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out to investigate alterations in outer membrane proteins (OMPs) as described previously with some modification.24 Briefly, bacterial cells were disrupted by ultrasonic disintegration and the supernatants were treated with 8 M urea. After incubation for 30 min, OMPs were collected by centrifugation at 25,000 g for 1 h and analyzed by SDS-PAGE on a Mini-PROTEAN 3 Cell apparatus (Bio-Rad, Hercules, California, USA) using 10% (wt/vol) polyacrylamide gels. The gels were stained with Coomassie Brilliant Blue and alteration of porins was determined by comparison of electrophoretic migration patterns between isolates and control strains. OMP analysis was performed for Enterobacter cloacae, K. pneumoniae, and E. coli isolates. E. cloacae ATCC 13047, K. pneumoniae ATCC 13883, and E. coli ATCC 25922 were used as controls.
Molecular epidemiology study using pulsed-field gel electrophoresis
Pulsed-field gel electrophoresis (PFGE) patterns of XbaI-restricted genomic DNA were compared to determine the relatedness of the 72 isolates that were identified as ertapenem-resistant by VITEK2. XbaI-restricted genomic DNA from isolates was separated by PFGE using a CHEF-DR II system (Bio-Rad) according to the manufacturer's protocol. Dendrograms were generated by the unweighted pair group method with the arithmetic average method, and DNA relatedness was calculated on the basis of the criteria suggested by Tenover et al.25
Results
Antimicrobial susceptibilities and clinical features of the isolates
Resistance rates for various antimicrobial agents are shown in table 1. For carbapenems, only two ertapenem-resistant isolates were also resistant to imipenem and one was resistant to meropenem. The isolates showed high resistance rates for almost all β-lactams excluding carbapenems such as ampicillin, amoxicillin/clavulanic acid, third-generation cephalosporin (cefotaxime and ceftazidime), cefoxitin, and aztreonam. Isolates were relatively susceptible to non-β-lactams including amikacin, gentamicin, trimethoprim/sulfamethoxazole and ciprofloxacin.
Clinical characteristics of the patients carrying Enterobacteriaceae with a high MIC for ertapenem (≥8 μg/mL) are shown in table 2. All but one isolate with a high MIC for ertapenem were considered to be possible pathogens. Four patients were aged over 70 years. Three patients had malignant disease and one patient was bedridden. One patient was treated with meropenem and showed clinical improvement.
Investigation of β-lactamases
Detection of ESBL and AmpC β-lactamase
All ertapenem non-susceptible isolates (n=40) were positive in the AmpC β-lactamase screening test. AmpC β-lactamases were found in 32 (80.0%) isolates; 24 isolates with blaMIR/ACT-like, 5 isolates with blaCMY-2-like, and 4 isolates with blaDHA (table 3). ESBL genes were detected in 20 (50.0%) isolates; 13 isolates with blaCTX-M (9 CTX-M-1 group and 4 CTX-M-9 group), 10 isolates with blaSHV, and 12 isolates with blaTEM. Distributions of β-lactamase genes were different among the species. blaMIR/ACT-like genes were almost exclusively detected in Enterobacter spp. Most of the K. pneumoniae isolates had blaDHA and blaSHV, whereas most of the E. coli isolates had blaTEM, blaCTX-M, and blaCMY-2-like genes. blaSHV genes were detected more frequently in resistant isolates (8/20) than intermediate isolates (2/20), and in most of the K. pneumoniae isolates. Otherwise, there were no significant differences in the prevalence of β-lactamases between the resistant and the intermediate isolates.
Detection of carbapenemase
One C. freundii isolate among the 40 isolates with ertapenem non-susceptibility was positive for carbapenemase in the modified Hodge test, carbapenemase inhibition test, and chromogenic agar test for screening of carbapenemase-producing isolates. The MIC of ertapenem determined by broth microdilution was 4 μg/mL. The MIC of imipenem and meropenem determined by automated systems was 0.5 μg/mL and 2 μg/mL, respectively. Multiplex PCR for carbapenemase detected blaIMP. PCR and sequencing analysis using primers targeting the blaIMP-1 gene and class 1 integron confirmed the blaIMP-1-associated class 1 integron.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis for OMP
According to the SDS-PAGE analysis, all of the K. pneumoniae (n=5) and E. cloacae (n=11) isolates with ertapenem resistance showed altered or greatly diminished expression of porins compared to the control strains (table 4). Porin alterations were less common among the ertapenem-intermediate isolates (4/19). Among 14 ertapenem-intermediate isolates of E. cloacae, four had altered porins. All of the ertapenem non-susceptible E. coli isolates conserved porins compared to the control strain.
Mechanisms associated with ertapenem non-susceptibility in Enterobacteriaceae
The results of molecular analysis of β-lactamases and OMP analysis are integrated in table 4. Ertapenem resistance caused by carbapenemase was observed in one isolate. Most of the other isolates with ertapenem resistance were associated with β-lactamase activity and porin alterations. The presence of β-lactamases without porin alteration was observed in most of the intermediate isolates.
Molecular epidemiology
All of the K. pneumoniae and E. coli isolates showed distinct individual pulsotypes and were considered to be epidemiologically unrelated (figure 1 and table 5). Among the E. cloacae isolates, pulsotypes A, B, and C were observed in 4, 2, and 2 different patients, respectively. All four isolates of pulsotype A were isolated from patients in the neonatal intensive care unit during a 3-month period; the two isolates of pulsotype B were isolated from patients in the same ward from the same specimen during a 3-month period; and the two isolates of pulsotype C were isolated from respiratory specimens of patients in the same ward during a 1-week period. The two isolates of pulsotype A with porin alteration showed a relatively high MIC of ertapenem compared to the other two isolates.
Discussion
Increasing resistance to ertapenem among Enterobacteriaceae is becoming a major therapeutic problem. The resistance rates of Enterobacteriaceae to ertapenem have been variously reported, depending on the species. The Study for Monitoring Antimicrobial Resistance Trends (SMART) reported that E. coli isolated in Europe and Latin America during 2008 and 2009 had high susceptibility to ertapenem with resistance rates of 0.2% and 0.3%, respectively.2 ,4 K. pneumoniae, on the other hand, showed a relatively high resistance rate of 6.5% globally.1 Obviously, there are also regional differences. The rates of susceptibility to ertapenem of K. pneumoniae varied between different geographical regions, from 82.3% in the Middle East to 100% in Africa. In Taiwan, the ertapenem susceptibility rate was 92.9%, 80.9%, and 67.9% for E. coli, K. pneumoniae, and E. cloacae, respectively.3
In this study, the resistance rate to ertapenem among all Enterobacteriaceae isolates was 2.0% as determined using VITEK2. It was highest in Enterobacter spp. (14.0%), and was 2.5% and 0.4% in K. pneumoniae and E. coli, respectively. In comparison, the resistance rate to imipenem was 0.1%. Even though there were differences in the antimicrobial susceptibility testing methods and the criteria used for interpretation, in general the resistance rates to ertapenem were higher than those for other carbapenems, and Enterobacter spp. and K. pneumoniae showed higher resistance rates than E. coli.1–4
As reported in previous studies,5–9 ,13 susceptibility to imipenem and meropenem was retained in most of the isolates with ertapenem non-susceptibility. This may reflect relative penetration rates through minor porins, differential susceptibility to efflux, or relative susceptibility to slow hydrolysis by AmpC β-lactamases or ESBLs, and may perhaps be related to the larger size and more negative charge of ertapenem.7 ,13 Further investigation of this differential susceptibility is warranted to determine the optimal antimicrobial therapy.
Previous studies showed that ertapenem resistance in Enterobacteriaceae is mainly due to mechanisms other than carbapenemases, especially expression of β-lactamases, such as an AmpC β-lactamase or an ESBL, combined with porin loss.1 ,2 ,4 ,5 ,7–9 Similarly, in this study, most of the isolates with ertapenem non-susceptibility were associated with β-lactamase activity and/or porin alteration. Most of the ertapenem-resistant isolates had porin alteration, whereas porin alterations were less common among the ertapenem-intermediate isolates. A combination of β-lactamase activity and porin alteration tended to result in higher MICs of ertapenem compared to isolates with β-lactamase activity alone. This suggests that ertapenem resistance in Enterobacteriaceae might result from the accumulation of multiple carbapenem-resistance determinants.
PFGE revealed genetic diversity among most of the isolates, suggesting that ertapenem resistance in clinical Enterobacteriaceae isolates was not the result of dissemination of resistant clones in the hospital but rather reflected independent emergence in different strains. Nonetheless, the phenomenon of the same clone spreading in the same ward during a single period was observed in this study. Interestingly, isolates of the same pulsotype with porin alteration showed relatively high MICs of ertapenem compared to the other isolates, suggesting that the non-susceptibility to ertapenem in Enterobacteriaceae might result from the accumulation of multiple carbapenem-resistance determinants in different isolates.
The patients in this study carrying Enterobacteriaceae with a high MIC of ertapenem (MIC≥8 μg/mL) were older and had underlying diseases. The clinical course and outcome differed from individual to individual. One patient who was treated with meropenem showed clinical improvement, indicating the therapeutic potential of meropenem treatment for ertapenem-resistant Enterobacteriaceae without carbapenemase. Further clinical studies are needed to evaluate the use of other carbapenems for ertapenem-resistant Enterobacteriaceae.
Even though carbapenemase-producing Enterobacteriaceae are still rare in Korea, there has recently been an increase in their detection.26–29 In this study, one C. freundii isolate was confirmed to be a carbapenemase-producing Enterobacteriaceae and screening tests, including the modified Hodge test, carbapenemase inhibition test, and chromogenic agar test, were shown to be helpful for detection. Even though it carried IMP-1, this isolate was not resistant to other carbapenems (imipenem and meropenem) tested using the two automated systems. In this case, ertapenem was only sensitive for detecting carbapenemase-producing isolates. Ertapenem susceptibility has been recommended as a sensitive indicator of KPC,30–33 although its low specificity and positive predictive value have been persistent problems.34–36 In this study, the positive predictive value of ertapenem resistance for detection of carbapenemase-producing isolates was very low (1/20). The low specificity and positive predictive value observed in tests for detection of carbapenemase-producing Enterobacteriaceae were due in part to the low prevalence of these isolates. Therefore, accurate detection of carbapenemase-producing Enterobacteriaceae is still challenging.
In summary, ertapenem resistance in clinical isolates was associated with β-lactamase activity and porin alteration. Even though carbapenemase-producing Enterobacteriaceae are still rare, continuous monitoring and infection control for carbapenem-resistant Enterobacteriaceae is necessary.
Acknowledgments
The authors would like to thank Jongsoo Jeon, Sori Jong, and Hyungsun Kim (Research Institute of Bacterial Resistance, Yonsei University College of Medicine) for their assistance in laboratory work.
Footnotes
This work was presented at the ICAAC/ICC 2015 (Joint 55th Interscience Conference on Antimicrobial Agents and Chemotherapy and 28th International Congress of Chemotherapy Meeting), San Diego, California, USA, September 17–21, 2015 (Poster Board Number: C135).
Contributors H-S Chung carried out the molecular genetic studies and performed the statistical analysis. D Yong carried out the outer membrane protein analysis and molecular epidemiology study. H-S Chung and M Lee participated in the design of the study and drafted the manuscript. All authors read and approved the final manuscript.
Competing interests None declared.
Ethics approval Ewha Womans University Mokdong Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.