JAC Advance Access originally published online on July 21, 2008
Journal of Antimicrobial Chemotherapy 2008 62(5):986-990; doi:10.1093/jac/dkn296
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original research |
Cefuroxime non-susceptibility in multidrug-resistant Klebsiella pneumoniae overexpressing ramA and acrA and expressing ompK35 at reduced levels
1 Clinical Microbiology, MTC, Karolinska Institutet, Karolinska University Hospital, Solna, Stockholm, Sweden 2 Department of Laboratory Medicine, Karolinska Institutet, Karolinska University Hospital, Huddinge, Sweden 3 Unit of Infectious Diseases, Department of Medicine, Karolinska Institutet, Karolinska University Hospital, Solna, Stockholm, Sweden 4 Division of Antibiotics and Infection Control, Department of Bacteriology, Swedish Institute for Infectious Disease Control, Solna, Sweden
* Correspondence address. Clinical Microbiology L2:02, Karolinska University Laboratory, Karolinska University Hospital, Solna, 171 76 Stockholm, Sweden. Tel: +46-8-51779642; Fax: +46-8-308099; E-mail: owe.kallman{at}karolinska.se
Received 15 February 2008; returned 21 March 2008; revised 8 May 2008; accepted 30 June 2008
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
Objectives: The aims were to study if efflux and down-regulation of porins contribute to cefuroxime resistance in Klebsiella pneumoniae and to co-resistance to unrelated antibiotics.
Methods: Ten cefuroxime-non-susceptible but cefotaxime-susceptible blood culture isolates of K. pneumoniae and one multiply antibiotic-resistant (MAR) laboratory strain (selected by chloramphenicol) were examined. Transcription of the genes acrA, ompK35, ramA, marA and soxS was determined with quantitative RT–PCR.
Results: All clinical isolates and the MAR laboratory strain had similar antibiograms with non-susceptibility to cefuroxime, tigecycline, chloramphenicol and nalidixic acid. Phenylalanine arginine β-naphthylamide (PAβN) increased susceptibility to tigecycline, chloramphenicol and nalidixic acid, but not to cefuroxime. Increased acrA transcription and decreased ompK35 transcription was seen in all strains. Increased ramA transcription was seen in all strains except one clinical isolate.
Conclusions: This multidrug-resistant phenotype of K. pneumoniae is associated with increased acrA and ramA transcription and decreased ompK35 transcription. Since the cefuroxime resistance was not reversed by PAβN, it was probably attributable to decreased levels of OmpK35, rather than to efflux.
Keywords: porins , efflux , drug resistance , bacterial , cephalosporins , MDR
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
Acknowledged cefuroxime resistance mechanisms in Klebsiella pneumoniae are production of extended-spectrum β-lactamases (ESBLs) and down-regulation of porins, especially OmpK35.1,2 In Escherichia coli, active efflux has been shown as a possible cause of cefuroxime resistance.3 As in E. coli, the efflux pump AcrAB seems to be important for efflux-mediated resistance in K. pneumoniae.4 Mutations in the local regulator gene acrR have been shown to be associated with increased expression of acrA in K. pneumoniae.4 RamA, MarA and SoxS are global regulatory proteins involved in multidrug resistance (MDR).4,5 RamA is a transcriptional activator in K. pneumoniae involved in the regulation of acrAB expression and maybe also in the expression of other efflux pumps and porins.4–6 MarA and SoxS cause decreased porin expression and increased efflux pump expression in different species of Enterobacteriaceae.4
In this study, cefuroxime-resistant strains of K. pneumoniae were examined for the putative resistance mechanisms efflux, porin loss and ESBL production.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
Bacterial strains
All clinical blood culture isolates of K. pneumoniae from 2004 to 2006 (from the clinical laboratory at Karolinska University Hospital, Solna, Stockholm) that were cefuroxime-non-susceptible but cefotaxime-susceptible were examined (n = 10). K. pneumoniae ATCC 25955 was used as a control strain.
Selection of multiply antibiotic-resistant (MAR) laboratory strain
K. pneumoniae ATCC 25955 was grown on Iso-Sensitest agar (ISA) (Oxoid, Basingstoke, UK) containing chloramphenicol in increasing concentrations (3–4 cycles), and a strain with chloramphenicol MIC > 256 mg/L was selected. It quite easily reverted phenotypically to the original strain and was therefore grown in media containing 64 mg/L chloramphenicol. Similar methods have been used before.7
Antimicrobial susceptibility testing
MICs of cefuroxime, cefotaxime, ceftazidime, chloramphenicol, tigecycline, tetracycline, ciprofloxacin and trimethoprim/sulfamethoxazole were determined by Etest (Biodisk, Solna, Sweden) (www.srga.org). Susceptibility to nalidixic acid (30 µg discs) was determined by the disc diffusion method (Oxoid), (www.srga.org). Screening for ESBLs was performed using Etest ESBL (cefotaxime/cefotaxime + clavulanic acid and ceftazidime/ceftazidime + clavulanic acid) (Biodisk) (www.srga.org). MICs and inhibition zones, respectively, of the above-mentioned antibiotics were also determined on ISA plates containing the efflux pump inhibitor phenylalanine arginine β-naphthylamide (PAβN) (Sigma-Aldrich, St Louis, MO, USA) at 40 mg/L and for cefuroxime also on ISA plates containing clavulanic acid (Sigma-Aldrich) at 2 mg/L.
Real-time RT–PCR for quantification of mRNA encoded by acrA, ompK35, ramA, marA and soxS
Real-time RT–PCR was used to analyse the mRNA (transcription) levels of the genes acrA, ompK35, ramA, marA and soxS. As a reference, gene rrsE was used for normalizing the transcription levels of target genes.6 The method has previously been described for Pseudomonas aeruginosa8 and was used with some minor modifications. Total RNA was extracted (High Pure RNA Isolation Kit, Roche, Mannheim, Germany) and synthesis of cDNA from 20 ng of extracted RNA for each reaction was performed using the 1st strand cDNA synthesis kit for RT–PCR (Roche). With the Rotor-Gene 3000 real-time PCR apparatus (Corbett Research, Morlake, Australia) using SYBR Green (Qiagen, Hilden, Germany) for DNA detection and specific primers (Cybergene, Huddinge, Sweden), real-time PCR (40 cycles, 20 s each per cycle of 95°, 52° and 72°) was performed to quantify cDNA. Primers used (5'–3') for acrA were ATGTGACGATAAACCGGCTC and CTGGCAGTTCGGTGGTTATT, for ompK35 GAAGGTTCCCAGACCACAAA and ACGGCCATAGTCGAATGAAC, for ramA6 GCATCAACCGCTGCGTATT and CGTTGCAGATGCCATTTCG, for marA GAATGGCCGGTCTCTTTCTT and CCTGTCGCTGGAAAAAGTGT, for soxS GCAGGCGGCGCTGGCGAATA and AGTCGCCAGAAAGTCAGGAT and for rrsE6 TTGACGTTACCCGCAGAAGAA and GCTTGCACCCTCCGTATTACC. The specificity of the generated product was tested by melting-point analysis. Cycle threshold (CT) values of the target genes were directly compared with the CT values of the reference gene, rendering the normalized expression (transcription) value. Amplification was performed in triplicate from three different cDNA preparations. The mean of the normalized expression (transcription) values was calculated for acrA, ramA, marA, soxS and ompK35 for each strain, using the Q-gene software.8 Statistical analysis was performed with t-test for independent samples.
The acrR gene was amplified using a forward primer located upstream (CGTAACCTCTGTAAAGTCAT) and a reverse primer at the end of the gene (GCTGACAAGCTCTCCGGGC) (5'–3', Cybergene). AmpliTaqGold (Applied Biosystems, Stockholm, Sweden) was used for the PCR, and the cycling programme comprised 10 min at 95°C, followed by 30 cycles of 45 s at 94°C, 45 s at 52°C and 2 min at 72°C, with a final extension for 10 min at 72°C. PCR templates were purified with the Jetquick Spin Column Kit (Genomed; Saveen Werner AB, Malmoe, Sweden). DNA sequencing of PCR products was performed with the dideoxy chain-termination method. The amplification primers were used for the sequence reactions along with an additional internal forward primer (AAGTGTGAGTTCGTCGGTGA) (5'–3', Cybergene). Sequence reactions were performed using an ABI Prism Big Dye Terminator Cycle Sequencing kit v.3.1 (Applied Biosystems) according to the manufacturer's instructions. Sequence analysis was performed on an ABI Prism 3100 Genetic Analyzer (Applied Biosystems) using Seqscape software v.2.5.0 (Applied Biosystems).
In order to confirm that the clinical isolates did not represent one clone of K. pneumoniae with the same antibiotic susceptibility pattern, a semi-automated system for biochemical fingerprinting (the PhenePlate or PhP system) was used for the clinical isolates (PhPlate AB, Stockholm Sweden).9 The correlation coefficient (r) between each pair of isolates was calculated and isolates with r 
0.97 were considered to belong to the same PhP type, and hence to be potentially clonally related.9 Only two of the clinical isolates (KP-7 and KP-10) were found to belong to the same PhP-type.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
Antibiotic susceptibility
MICs and inhibition zones without and with PAβN are shown in Table 1. The MAR laboratory strain and the clinical isolates showed similar antibiograms, with non-susceptibility to cefuroxime, chloramphenicol, nalidixic acid and tigecycline. However, while the clinical isolates were phenotypically stable, the MAR laboratory strain was unstable, suggesting that the strain was phenotypically adapted to high antibiotic concentrations. Several isolates showed decreased susceptibility to ceftazidime and though they were cefotaxime-susceptible, MICs were higher than for the ATCC strain and for the wild-type population (www.eucast.org). All clinical isolates were ESBL-negative and clavulanic acid did not decrease the cefuroxime MIC significantly (4-fold) in any of the strains, which makes overexpression of SHV-1 unlikely (data not shown). PAβN increased susceptibility significantly (4-fold MIC decrease or 5 mm increase in inhibition zone diameter) to chloramphenicol, nalidixic acid, tigecycline and trimethoprim/sulfamethoxazole in the MAR laboratory strain and in most clinical isolates, while cephalosporin susceptibility was not affected significantly. This suggests that the decreased susceptibility to chloramphenicol, nalidixic acid and tigecycline is at least partly explained by efflux, but that another mechanism (probably lack of ompK35) is the cause of the cefuroxime resistance. K. pneumoniae strains with a similar MDR pattern with resistance to nalidixic acid, chloramphenicol and trimethoprim have been described before, but cefuroxime susceptibility was not determined in these studies.6,10 Proposed mechanisms explaining this phenotype pattern have been down-regulation of porins and increased efflux due to increased expression of the AcrAB efflux pump.6,10 In this study we found that this MDR phenotype is also associated with cefuroxime resistance and that all clinical isolates in this study, selected primarily because they were cefuroxime-non-susceptible but cefotaxime-susceptible, were also non-susceptible to chloramphenicol, tigecycline and nalidixic acid. In a study of antibiotic cross-resistance in K. pneumoniae, 91% of isolates with decreased cefuroxime susceptibility were also chloramphenicol-non-susceptible, compared with 7% of the cefuroxime-susceptible isolates.11 This provides further evidence for the existence of co-resistance between these chemically unrelated antimicrobial agents.
|
Transcription of acrA, ompK35, ramA, marA and soxS and acrR sequencing
The relative transcription levels (compared with the ATCC strain) for each strain for the examined genes are shown in Table 2. All strains had increased acrA transcription levels and decreased ompK35 transcription levels. Further, all strains except one (KP-4) had increased ramA transcription levels. Four of the clinical isolates and the MAR laboratory strain had significantly increased transcription levels of marA. For soxS, only the MAR laboratory strain had increased transcription levels. Thus, ramA seems to be the most important transcriptional activator, rendering the MDR phenotype in the clinical isolates and the laboratory strain, though marA and soxS might also be of importance in the laboratory strain. Overexpression of ramA (but not of marA and soxS) has previously been associated with increased acrAB expression in clinical isolates of K. pneumoniae.4 The sequences of the acrR regulatory gene were compared with a published GenBank sequence (NC_009648 [GenBank] ). Except for some silent mutations, only one single amino acid change of uncertain relevance was found: isolate KP-3 featuring a Glu190Lys change (Table 2).
|
Conclusions
Current evidence suggests that increased expression of ramA, resulting in increased acrAB expression and decreased ompK35 expression contributes to decreased susceptibility to several unrelated antibiotics: cefuroxime, tigecycline, chloramphenicol, nalidixic acid and, in most isolates, trimethoprim/sulfamethoxazole. The cefuroxime resistance was not reversed by the efflux pump inhibitor PAβN, implying loss of OmpK35 as the probable cause of cefuroxime resistance, while efflux of the AcrAB pump at least partially explains the non-susceptibility to tigecycline, chloramphenicol, nalidixic acid and trimethoprim/sulfamethoxazole. The finding of this MDR profile in multiple blood culture isolates of K. pneumoniae, all of them employing the same mechanisms of resistance, underscores the clinical importance of this phenotype.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
No specific external funding has been received.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We would like to thank Aina Iversen for excellent assistance with the PhP typing and analysis.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsReferences |
|---|
1 Poole K. Resistance to β-lactam antibiotics. Cell Mol Life Sci (2004) 61:2200–23.[Web of Science][Medline]
2
Lee CH, Chia JH, Chu C, et al. In vivo selection of OmpK35-deficient mutant after cefuroxime therapy for primary liver abscess caused by Klebsiella pneumoniae. J Antimicrob Chemother (2006) 58:857–60.
3 Källman O, Fendukly F, Karlsson I, et al. Contribution of efflux to cefuroxime resistance in clinical isolates of Escherichia coli. Scand J Infect Dis (2003) 35:464–70.[CrossRef][Web of Science][Medline]
4
Schneiders T, Amyes SG, Levy SB. Role of AcrR and RamA in fluoroquinolone resistance in clinical Klebsiella pneumoniae isolates from Singapore. Antimicrob Agents Chemother (2003) 47:2831–7.
5
George AM, Hall RM, Stokes HW. Multidrug resistance in Klebsiella pneumoniae: a novel gene, ramA, confers a multidrug resistance phenotype in Escherichia coli. Microbiology (1995) 141:1909–20.
6
Ruzin A, Visalli MA, Keeney D, et al. Influence of transcriptional activator RamA on expression of multidrug efflux pump AcrAB and tigecycline susceptibility in Klebsiella pneumoniae. Antimicrob Agents Chemother (2005) 49:1017–22.
7
George AM, Levy SB. Amplifiable resistance to tetracycline, chloramphenicol, and other antibiotics in Escherichia coli: involvement of a non-plasmid-determined efflux of tetracycline. J Bacteriol (1983) 155:531–40.
8 El Amin N, Giske CG, Jalal S, et al. Carbapenem resistance mechanisms in Pseudomonas aeruginosa: alterations of porin OprD and efflux proteins do not fully explain resistance patterns observed in clinical isolates. APMIS (2005) 113:187–96.[CrossRef][Web of Science][Medline]
9 Tullus K, Burman LG, Haeggman S, et al. Epidemiologic typing of international collections of Klebsiella spp.: computerized biochemical fingerprinting compared with serotyping, phage typing and pulsed-field gel electrophoresis. Clin Microbiol Infect (1999) 5:184–9.[Medline]
10 Gutmann L, Williamson R, Moreau N, et al. Cross-resistance to nalidixic acid, trimethoprim, and chloramphenicol associated with alterations in outer membrane proteins of Klebsiella, Enterobacter, and Serratia. J Infect Dis (1985) 151:501–7.[Web of Science][Medline]
11
Schumacher H, Scheibel J, Moller JK. Cross-resistance patterns among clinical isolates of Klebsiella pneumoniae with decreased susceptibility to cefuroxime. J Antimicrob Chemother (2000) 46:215–21.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||