Received 22 August 2012; returned 2 November 2012; revised 19 December 2012; accepted 21 December 2012

Transcription

1 J Antimicrob Chemother 03; 68: doi:0.093/jac/dks535 Advance Access publication 8 January 03 Tigecycline susceptibility in Klebsiella pneumoniae and Escherichia coli causing neonatal septicaemia (007 0) and role of an efflux pump in tigecycline non-susceptibility Subhasree Roy, Saswati Datta, Rajlakshmi Viswanathan, Arun K. Singh and Sulagna Basu * Division of Bacteriology, National Institute of Cholera and Enteric Diseases, P33, CIT Road, Scheme XM, Beliaghata, Kolkata 70000, India; Department of Neonatology, Institute of Post-Graduate Medical Education & Research and SSKM Hospital, Kolkata 70000, India *Corresponding author. Tel: , ext Fax: ; supabasu@yahoo.co.in Received August 0; returned November 0; revised 9 December 0; accepted December 0 Objectives: To investigate the trend of tigecycline susceptibility and mechanisms behind tigecycline nonsusceptibility in Klebsiella pneumoniae and Escherichia coli isolates causing neonatal septicaemia (007 0). Methods: MICs of tigecycline for the isolates were determined. The isolates were evaluated for b-lactamases and carbapenemases. Molecular typing of the tigecycline-resistant isolates was performed. Expression of efflux pump genes (acra, acrb and tolc) and regulators (soxs and rama) was examined by real-time RT PCR and western blotting. Sequencing of the rama and ramr genes was carried out to identify mutations within these genes. Results: Tigecycline susceptibility was evaluated in all K. pneumoniae (n¼57) and E. coli (n¼9) blood isolates. The prevalence of extended-spectrum b-lactamase (ESBL)-producing organisms was high, but tigecycline nonsusceptibility remained low in these isolates. Though MIC values of tigecycline remained in the susceptible range, there was a -fold increase in the value of MIC 90 from 007 to 00. Over the 4 year period K. pneumoniae showed higher MIC values of tigecycline in comparison with E. coli. Tigecycline non-susceptibility was not observed among carbapenem-resistant isolates. Only two ESBL-producing clonally distinct K. pneumoniae isolates showed tigecycline resistance with overexpression of rama and the AcrAB TolC pump. No mutations were present within the rama and ramr genes that might enhance the expression of the pump. Conclusions: The study showed for the first time the trend of tigecycline susceptibility in E. coli and K. pneumoniae causing neonatal septicaemia. Tigecycline still has potent antimicrobial effects against most ESBL- or carbapenemase-producing K. pneumoniae and E. coli, but the increasing MIC values make it essential to be vigilant. Keywords: neonates, bloodstream infections, Enterobacteriaceae, tigecycline resistance, carbapenemases, India Introduction Bacterial infections are becoming increasingly daunting due to multiple antibiotic resistance mechanisms that are difficult to combat. The escalating antimicrobial resistance amongst Gram-negative bacteria (GNB) has resulted in a surge of bloodstream infections. 3 GNB, particularly K. pneumoniae and E. coli, areamajor contributor to neonatal sepsis. 4 6 These two pathogens are resistant to multiple antimicrobial agents by different mechanisms, including expression of extended-spectrum b-lactamases (ESBLs), AmpC b-lactamases, 6S rrna methylases, aminoglycosidemodifying enzymes and recently also carbapenemases. 7,8 Moreover, efflux/influx-mediated mechanisms are also being reported in these organisms. 9 Penicillins, cephalosporins and carbapenems are the preferred treatment regimen for many bacterial infections. However, the global emergence of carbapenem resistance, particularly NDM- in Enterobacteriaceae, has further complicated treatment regimens. 0, In general, most NDM- b-lactamase-producing bacteria show resistance to almost all antibiotics except a few such as colistin or tigecycline. 0 It has thus become imperative to be vigilant about the trend in patterns of organism susceptibility to salvage therapies. Tigecycline is an expanded broad-spectrum antibiotic representing a new class called the glycylcyclines. Its structural modification is the addition of a 9-t-butyl-glycylamido side chain to the central skeleton of minocycline. This provides the drug with an expanded spectrum of activity against most # The Author 03. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oup.com 036 Downloaded from on 06 March 08

2 Tigecycline susceptibility in K. pneumoniae and E. coli JAC ESBL- and carbapenemase-producing K. pneumoniae and E. coli. 3,4 Non-susceptibility to tigecycline has been reported previously in K. pneumoniae and E. coli isolates. 5,6 Recently, emergence of non-susceptibility to tigecycline has also been reported in NDM--possessing isolates. 7 However, none of the studies has followed the trend of tigecycline susceptibility in isolates causing bloodstream infection in the neonatal population, particularly during a period when carbapenem resistance due to NDM- emerged in Enterobacteriaceae. This study was carried out in a neonatal intensive care unit (NICU) of a tertiary care hospital to (i) identify the susceptibility pattern to tigecycline in K. pneumoniae and E. coli collected during a 4 year period (when carbapenem resistance emerged in the unit) from blood of septicaemic neonates, and (ii) investigate the role of the AcrAB TolC pump in tigecycline nonsusceptibility during that period. Materials and methods Collection of blood specimens and identification of bacterial isolates The blood specimens were collected prior to antibiotic therapy from neonates admitted to a 0 bed NICU in a tertiary care centre, IPGMER and SSKM Hospital in Kolkata, India, between 007 and 00. One millilitre of blood for culture was drawn with aseptic precautions from a peripheral vein into Peds Plus vials. Blood culture was performed with a BACTEC 9050 system (Becton Dickinson, Sparks, MD, USA) from 008 onwards and manually prior to that. For any culture that flagged positive, Gram staining was performed and subculture was done on appropriate media based on Gram s stain: MacConkey agar (Difco Laboratories, Detroit, MI, USA) and 5% sheep blood agar (Difco Laboratories) for Gram-negative and Gram-positive organisms, respectively. The isolates were identified by different biochemical tests and further verified by using an ID3E or ID3GN system (biomérieux, Marcy l Étoile, France). MIC values and phenotypic detection of ESBLs and carbapenemases MIC values (mg/l) of different groups of antimicrobial agents (see Table ) were evaluated using the Etest method (AB Biodisk, Solna, Sweden) according to the manufacturer s instructions and interpreted according to CLSI criteria. 8 MIC results of tigecycline were interpreted according to EUCAST (EUCAST-0; clinical breakpoints (for tigecycline,.0 mg/l is susceptible,.0 mg/l is intermediate and..0 mg/l is resistant). 4 The MIC 50 and MIC 90 were calculated as the MIC at which 50% and 90%, respectively, of the isolates were inhibited. The MIC of tigecycline was also determined with the efflux pump inhibitor phenylalanine arginine b-naphthylamide (PAbN; 40 mg/l) (Sigma-Aldrich, St Louis, MO, USA) to investigate the role of the AcrAB TolC efflux pump in the tigecycline-resistant and tigecycline-susceptible K. pneumoniae isolates. MICs of other antibiotics (chloramphenicol, nalidixic acid, tetracycline, minocycline and ciprofloxacin) were also determined in the presence of PAbN to investigate the role of the pump in resistance to other substrates of the pump. All isolates were tested for the production of ESBLs, carbapenemases and metallo-b-lactamases using the cephalosporin/clavulanic acid combination disc test, 8 the modified Hodge test and the imipenem EDTA double-disc combination test, respectively. 9 AmpC b-lactamase was also investigated as described in a previous study. 0 PFGE PFGE was performed to check the clonality of the tigecycline-resistant isolates using PulseNet standardized procedures with XbaI (50 U/plug) ( in a CHEF-DR III system (Bio-Rad Laboratories, Hercules, CA, USA). XbaI macrorestriction patterns were compared visually and interpreted according to Tenover et al. DNA amplification of antibiotic resistance genes PCR was carried out for ESBL genes (bla SHV, bla TEM, bla OXA-, and bla CTX-M ), ampc genes (MOX, CIT, DHA, ACC, EBC and FOX), carbapenem resistance genes (bla VIM, bla IMP, bla SPM-, bla GIM-, bla SIM-, bla KPC, bla SME, bla SPM, and bla NDM and bla OXA-3, bla OXA-4, bla OXA-48, bla OXA-5, and bla OXA-58 ) and also for 6S rrna methylase-encoding genes (rmta, rmtb, rmtc, rmtd and arma). Amplified products were further sequenced for the tigecycline-resistant isolates. Gene expression analysis using real-time RT PCR Tigecycline-resistant isolates were examined by real-time RT PCR for the mrna expression level of pump genes (acra, acrb and tolc) and the regulators of the acrab operon (soxs and rama). Primers for acra, rama, soxs and a housekeeping gene (rrse) were used as described previously. For tolc (F-CTACAAACAGGCGGTGGTCT and R-TGTTCAGCTCGTTGATCAGG 5 3 ) and acrb (F-CAATACGGAAGAGTTTGGCA and R-CAGACGAACCTGGGAACC), primers were designed in this study. Overnight bacterial cultures were diluted /00 in Mueller Hinton broth (Difco Laboratories) and grown to mid-logarithmic phase at 378C with vigorous shaking (30 rpm). RNA was harvested using TRIzol reagent (Invitrogen) according to the manufacturer s instructions. RNA preparations were treated with RNase-free DNase (New England Biolabs, USA) and assessed for purity by PCR. RNA yield and quality were determined using a spectrophotometer (Eppendorf, Hamburg, Germany). Two micrograms of the isolated total RNA was reverse transcribed using the high-capacity cdna reverse transcription kit (Applied Biosystems, Warrington, UK) to synthesize the cdna first strand. Thereafter, the amplified cdna first strand was used in the subsequent analysis by real-time quantitative RT PCR. Quantification of the expression of the target genes was performed using Power SYBR Green PCR Master Mix (Applied Biosystems) using the StepOne Plus Real-Time PCR System and software (Applied Biosystems). The reactions were performed with 0.4 mm primer in 0 ml reaction volumes. All samples were analysed in duplicate. The DDC T method was used to perform quantitative relative expression analysis. The expression of each gene was normalized to that of the ribosomal housekeeping gene (rrse). Relative expression of each target gene was then calibrated against the corresponding expression using a tigecyclinesusceptible isolate (expression¼), which served as the control. SDS PAGE and immunodetection of AcrA, TolC, porins and OmpA Whole-cell extracts of the isolates were separated on % SDS polyacrylamide gels and transferred to Immobilon-P filters (Millipore) following standard procedures. AcrA, TolC and outer membrane proteins (OMPs) were detected using polyclonal anti-acra antibody (: dilution), polyclonal anti-tolc antibody (:7000 dilution), polyclonal anti-ompc/f antibody (:3000 dilution) and polyclonal anti-ompa antibody (:3000 dilution), respectively. 3,4 Mutation analysis of rama and ramr The rama gene from the tigecycline-resistant and tigecycline-susceptible isolates was amplified and sequenced using a forward primer (ATTTC Downloaded from on 06 March

3 Roy et al. CGCTCAGGTGATTGA) and a reverse primer (GTGGTTCTCTTTGCGGTAGG) (5 3 ) to identify mutations within the gene that might be responsible for rama overexpression. Previous studies have reported that ramr mutations are directly linked to rama overexpression. 5 PCR amplification and sequencing of the ramr gene and the surrounding genomic region from the tigecycline-resistant and tigecycline-susceptible isolates were performed with the primers and conditions used previously. 5 Results and discussion Trend of tigecycline susceptibility in K. pneumoniae and E. coli K. pneumoniae (n¼57) and E. coli (n¼9) were isolated during the study period and the distribution of isolates is given in Figure. Resistance to tigecycline was observed for 3.5% (/ 57) of the total K. pneumoniae isolates (Table ). The MIC distributions of tigecycline for K. pneumoniae isolates collected during are shown in Figure (a). The range of tigecycline MIC for the K. pneumoniae isolates over the 4 year period was mg/l, with the following values observed in each of the years: mg/l in 007, 3 mg/l in 008, mg/l in 009 and mg/l in 00. The distribution of the MIC values indicated an upward shift of tigecycline MICs among the K. pneumoniae isolates from 007 to 009. The MIC 50 value for K. pneumoniae isolates was mg/l, with values of 0.38,, 0.75 and mg/l in 007, 008, 009 and 00, respectively. The MIC 90 value for K. pneumoniae isolates was.5 mg/l, with values of,.5,.5 and mg/l (a) Number of isolates (b).6.5 MIC (mg/l) Klebsiella pneumoniae MIC (mg/l) values Number of isolates Year No. of K. pneumoniae No. of E. coli Escherichia coli MIC (mg/l) values MIC 90 of Klebsiella pneumoniae MIC 50 of Klebsiella pneumoniae MIC 90 of Escherichia coli MIC 50 of Escherichia coli.5 Figure. Distribution of tigecycline MIC values (a) and MIC 50 and MIC 90 values (b) among 57 isolates of K. pneumoniae and 9 isolates of E. coli as determined by the Etest method. Table. Tigecycline susceptibilities of K. pneumoniae and E. coli isolates obtained from 007 to 00, determined by the Etest method Number of isolates (%) categorized by EUCAST-0 MIC interpretive criteria as: tigecycline susceptible tigecycline resistant K. pneumoniae E. coli K. pneumoniae E. coli During (K. pneumoniae 57/E. coli 9) 55 (96.5%) 9 (00%) (3.5%) 0 ESBL producing (K. pneumoniae 53/E. coli 7) 5 (96.%) 7 (00%) (3.8%) 0 Carbapenem resistant (K. pneumoniae 5/E. coli 3) 5 (00%) 3 (00%) 0 0 None of the isolates was in the intermediate range. 038 Downloaded from on 06 March 08

4 Tigecycline susceptibility in K. pneumoniae and E. coli JAC in 007, 008, 009 and 00, respectively (Figure b). This indicated a -fold increase in the MIC 90 of tigecycline between 007 and 00. However, MIC 50 and MIC 90 values remained within the susceptible range based on the EUCAST breakpoint. The range of tigecycline MIC for the E. coli isolates over the 4 year period was mg/l, with the following values observed in each of the years: mg/l in 007, mg/l in 008, mg/l in 009 and mg/l in 00 (Figure a). The MIC 50 value for E. coli isolates was mg/l, with values of 0.094, 0.064, and 0.9 mg/l in 007, 008, 009 and 00, respectively. The MIC 90 value for E. coli isolates was 0.5 mg/l, with values of 0.9,, 0.5 and mg/l in 007, 008, 009 and 00, respectively (Figure b). The MIC 90 of tigecycline for E. coli isolates again indicated a.-fold increment during However, MIC 50 and MIC 90 values remained within the susceptible range based on the EUCAST breakpoint. All E. coli isolates were susceptible to tigecycline. For both species, the MIC 50 and MIC 90 obtained from small numbers of samples such as in this study should not be overemphasized as a few strains with high MICs may skew the results. In spite of this, incremental changes in MIC values were observed from 007 to 00. In addition, a comparison of the two organisms revealed that MIC 50 and MIC 90 were consistently higher for K. pneumoniae (Figure b). Similar results have also been reported by previous authors, where ESBL-producing E. coli had an MIC 90 of mg/l in comparison with K. pneumoniae, whose MIC 90 was mg/l. 3 Over the period, 9.9% (53/57) and 8.8% (5/57) of K. pneumoniae isolates were found to be ESBL producers and carbapenem resistant, respectively (Table ). Tigecyclineresistant K. pneumoniae were isolated only in 008 (n¼) and in 009 (n¼). Both tigecycline-resistant isolates were ESBL producers, but none of them was carbapenem resistant. The five carbapenem-resistant K. pneumoniae isolates collected in 00 possessed NDM-, but were susceptible to tigecycline (tigecycline MIC range was mg/l). During the study period, 89.5% (7/9) and 5.8% (3/9) of E. coli isolates were found to be ESBL producers and carbapenem resistant, respectively (Table ). The three carbapenem-resistant E. coli isolated in 008, 009 and 00 possessed NDM-, but were susceptible to tigecycline (tigecycline MICs were 0.5 mg/l). Among the K. pneumoniae and E. coli isolates, the incidence of tigecycline resistance was low, with only two tigecyclineresistant K. pneumoniae isolated sporadically during this period. However, there was a doubling of the MIC 90 value in 00 compared with 007 in both species. In spite of the high cephalosporin resistance, resistance to tigecycline remained low, probably because tigecycline was rarely used in the treatment of the neonates. There was a disproportionately higher number of isolates that produced ESBLs (n¼70) compared with the isolates that produced carbapenemases (n¼8) and those that did not produce ESBLs or carbapenemases (n¼6). The tigecycline MICs for the two tigecycline-resistant isolates, KP-3 and KP-9, were 3 and 8 mg/l, respectively (Table ). Both isolates were resistant to multiple antimicrobial agents (Table ). The tigecycline-susceptible K. pneumoniae isolate (KP-S) showed susceptibility to all tested antimicrobials (Table ). Exposure of the tigecycline-resistant clinical isolates (KP-3 and KP-9) to the efflux pump inhibitor PAbN resulted in a decrease (- to.8-fold MIC decreases) in the MICs of tigecycline, chloramphenicol, nalidixic acid, tetracycline and minocycline, confirming the previously reported broad-substrate specificity of the AcrAB TolC pump 5,6 and further suggesting that the AcrAB pump is Table. MIC values of antibiotics in the presence and absence of PAbN and the presence of different b-lactamases and gene expression analysis of efflux pump genes and pump regulators in tigecycline-resistant and tigecycline-susceptible isolates Isolate number, year and month of isolation MIC (mg/l) values of antibiotics MIC (mg/l) values in presence of PAbN Presence of antibiotic resistance genes Fold change in the pump genes and regulators compared with isolate KP-S acra acrb tolc rama soxs KP-3, 008, July AMP,.56; AMK, ; GEN, 64; CIP, 3; CTX,.3; CAZ, 6; IPM, 0.5; ATM, 4; MIN, 8; TGC, 3; TET,.56; CHL,.56; NAL, 4; CST, CIP,.5; CHL, 96; NAL, ; TET,.56; MIN, ; TGC, 0.9 TEM-; OXA-; CTX-M KP-9, 009, January AMP,.56; AMK, 8; GEN, 48; CIP,.3; CTX,.3; CAZ,.56; IPM, 0.5; ATM,.56; MIN,.56; TGC, 8; TET,.56; CHL, 9; NAL,.56; CST, CIP,.3; CHL, ; NAL,.56; TET, 48; MIN, 4; TGC, TEM-; OXA-; CTX-M KP-S, 007, March AMP,.56; AMK, 3; GEN, ; CIP, 0.047; CTX, 0.094; CAZ,.5; IPM, 0.9; ATM, 4; MIN, ; TGC, 0.5; TET,.5; CHL, ; NAL, ; CST, CIP, 0.03; CHL, 0.38; NAL, ; TET, ; MIN, ; TGC, 0.9 no ESBL, AmpC and 6S rrna methylase genes were present AMP, ampicillin; AMK, amikacin; GEN, gentamicin; CIP, ciprofloxacin; CTX, cefotaxime; CAZ, ceftazidime; IPM, imipenem; ATM, aztreonam; MIN, minocycline; TGC, tigecycline; TET, tetracycline; CHL, chloramphenicol; NAL, nalidixic acid; CST, colistin; KP, K. pneumoniae. Downloaded from on 06 March

5 Roy et al. involved in decreased susceptibility of the K. pneumoniae isolates to tigecycline. Exposure of the tigecycline-susceptible isolate (KP-S) to PAbN did not result in substantial decreases in the MICs of the above mentioned antibiotics, suggesting a lower expression level of the AcrAB pump (Table ). Molecular typing of the isolates and presence of other b-lactamases PFGE indicated that the tigecycline-resistant K. pneumoniae isolates (KP-3 and KP-9) and the tigecycline-susceptible isolate (KP-S) were clonally distinct (Figure S, available as Supplementary data at JAC Online). These isolates were included for further gene expression analysis by real-time RT PCR. The tigecycline-resistant isolates possessed ESBL genes, but no AmpC b-lactamases or 6S rrna methylases were detected (Table ). The tigecycline-susceptible isolate was found to be ESBL-negative and AmpC b-lactamases or 6S rrna methylases were not present (Table ) Expression analysis of pump genes and regulators The MICs of tigecycline for the tigecycline-resistant K. pneumoniae isolates KP-3 and KP-9 represent 8-fold and 3-fold increases, respectively, over the MIC for the tigecyclinesusceptible isolate KP-S (MIC 0.5 mg/l). Moreover, phenotypic evidence of expression of efflux pumps had been found in these tigecycline-resistant isolates in the presence of the efflux pump inhibitor PAbN. In this context, we evaluated the role of the AcrAB efflux pump and its regulators (RamA and Sox) in the clonally distinct tigecycline-resistant K. pneumoniae isolates. Approximately and 9.7-fold increased expression of rama was observed in tigecycline-resistant K. pneumoniae isolates KP-3 (MIC 3 mg/l) and KP-9 (MIC 8 mg/l), respectively, KP-S KP-3 AcrA OmpF OmpC KP-9 47 kda 3 kda Positive control KP-S KP-3 Porins TolC 47 kda 3 kda 47 kda OmpC (OmpK36) 3 kda Figure. Western blot analysis to detect the expression levels of AcrA, TolC and porins (OmpC/F) in tigecycline-susceptible (KP-S) and tigecycline-resistant (KP-3 and KP-9) K. pneumoniae isolates. OmpK35 and OmpK36 in K. pneumoniae are homologues of OmpF and OmpC in E. coli. The positive control used in the blot for porins is an E. coli isolate. KP-S KP-9 KP-3 KP-9 compared with the susceptible isolate (KP-S). Quantitative analysis of soxs expression showed no increment compared with the tigecycline-susceptible isolate KP-S (Table ), indicating that it probably has no role in activation of the pump. These data support the hypothesis that increased expression of rama is associated with increased MICs of tigecycline. It was reported previously that rama overexpression in K. pneumoniae was directly linked to elevated expression of the multidrug efflux pump AcrAB. 5 To test whether a similar scenario existed, real-time RT PCR was carried out for the efflux pump genes acra, acrb and tolc. The tigecycline-resistant isolates showed increased expression of the pump genes compared with the tigecycline-susceptible isolate KP-S (Table ). Of note, expression levels of pump genes in KP-3 and KP-9 were in linear proportion to the tigecycline MIC increment (Table ). Further, of the three pump genes, acra had the highest expression level, which was almost equal to rama expression (Table ). The other components of the AcrAB TolC pump (acrb and tolc) wereoverexpressed in the resistant isolates, reaffirming that the three pump proteins are cotranscribed. 6 These results are in agreement with previous published data suggesting that overexpression of the AcrAB pump is involved in decreased tigecycline susceptibility in K. pneumoniae. 5,6 Western blot analysis An additional analysis of AcrA (a component of the major pump in Enterobacteriaceae) and TolC (the OMP of that pump) expression was performed by western blotting to validate the results of real-time RT PCR. The protein levels of AcrA and TolC were found to be higher in KP-9 and in KP-3 than in KP-S, supporting the real-time RT PCR results showing elevated expression levels of AcrA and TolC (Figure ). These data support the hypothesis that increased expression of pump genes is associated with increased MICs of tigecycline. The role of rama in increased expression of the AcrAB pump rama is also involved in the control of porin expression (by decreasing it). 7 Assessment of porin expression revealed the loss of OmpK35 (the homologue of E. coli OmpF) in both the tigecycline-resistant and tigecycline-susceptible K. pneumoniae isolates (Figure ), and thus porin had no role in tigecycline resistance. Mutation analysis of rama and ramr The rama gene was sequenced in the tigecycline-resistant (KP-3 and KP-9) isolates and the tigecycline-susceptible isolate (KP-S). The sequence of the rama region of tigecycline-resistant isolates was identical to that of the susceptible isolate KP-S and the result also corroborated with the sequence of the rama region of tigecycline-susceptible isolate Klebsiella pneumoniae subsp. pneumoniae MGH (GenBank accession no. CP000647). 5 This indicated that no mutations were present within the gene itself that might enhance the overexpression of the gene. In addition, sequencing was also carried out for the negative regulator of rama, i.e. the ramr gene. An earlier study has shown that mutations within the ramr result in rama upregulation. 5 However, in our study transcriptional levels of the rama gene did not appear to be linked to ramr, as mutations within the ramr gene responsible for the multidrug resistance phenotype 040 Downloaded from on 06 March 08

6 Tigecycline susceptibility in K. pneumoniae and E. coli JAC were not observed in KP-3 and KP-9. This result is in accordance with a previous study in which mutations within the ramr gene of tigecycline-non-susceptible K. pneumoniae isolates were not found. 6 RamR-binding site sequences in the rama ramr intergenic region have not been analysed for mutations. Funding The study was supported by institutional funding (NICED, Kolkata). The Collaborative Research Centre of Okayama University for Infectious Diseases in India gave a fellowship to S. R. Conclusions In summary, this study was carried out to investigate the trend of tigecycline susceptibility during a period when carbapenem resistance in Enterobacteriaceae owing to the presence of NDM- emerged in this unit. 0 Recent studies by the same authors have shown that septicaemia in neonates has been complicated by multidrug-resistant organisms. 5,6 In a situation with dwindling options for antibiotic therapy, mainly due to the emergence of carbapenem resistance, tigecycline is an alternative. Resistance to tigecycline remained low in neonatal bloodstream infections during the study period. There were differences in the tigecycline MIC values of E. coli and K. pneumoniae, with K. pneumoniae showing higher MIC values over the 4 year period. Though the presence of ESBLs in both organisms was high, only two ESBL-producing clonally distinct isolates of K. pneumoniae showed tigecycline resistance with overexpression of rama and AcrAB TolC pump proteins. Non-susceptibility to tigecycline was not observed among carbapenem-resistant NDM--possessing isolates in this setup, though such isolates have also been reported recently. 7 Though MIC values of tigecycline remained in the susceptible range, there was an increment from 007 to 00. This increase cannot be attributed to the use of tigecycline as there was restricted use of this antibiotic in the unit. However, it could be indirectly attributed to the use of other antibiotics that are also effluxed through the AcrAB TolC pump. 9 This is probably the first study to assess the trend of tigecycline susceptibility in E. coli and K. pneumoniae causing neonatal bloodstream infections. Our data reveal that tigecycline continued to have potent antimicrobial effects against most K. pneumoniae and E. coli isolates that produced ESBLs or carbapenemases. However, it is imperative to remain vigilant and trends should be continuously monitored. The susceptibility patterns for tigecycline need to be followed in different populations and geographic areas so that a holistic view about the susceptibility patterns can be attained. This information would be the best indicator for the long-term implications of the increasing tigecycline MIC values. Antimicrobial stewardship should be implemented as it helps towards prevention of emergence and cross-transmission of multidrug-resistant organisms. Since we presume that the use of other antibiotics may affect susceptibility to tigecycline, antibiotic stewardship will also help us understand whether this holds true in a real-life situation. Acknowledgements We would like to thank Heinz Schwarz and Helen I. Zgurskaya for the antibodies. We would also like to thank the staff of the Department of Neonatology, SSKM Hospital, who cared for the neonates included in the study, and Mr Subhadeep De for his laboratory assistance. Transparency declarations None to declare. Supplementary data Figure S is available as Supplementary data at JAC Online ( oxfordjournals.org/). References Albrecht SJ, Fishman NO, Kitchen J et al. Reemergence of Gram-negative health care-associated bloodstream infections. Arch Intern Med 006; 66: Mikulska M, Del Bono V, Raiola AM et al. Blood stream infections in allogeneic hematopoietic stem cell transplant recipients: reemergence of Gram-negative rods and increasing antibiotic resistance. Biol Blood Marrow Transplant 009; 5: Wu CJ, Lee HC, Lee NY et al. Predominance of Gram-negative bacilli and increasing antimicrobial resistance in nosocomial bloodstream infections at a university hospital in southern Taiwan, J Microbiol Immunol Infect 006; 39: Zaidi AKM, Thaver D, Ali SA et al. Pathogens associated with sepsis in newborns and young infants in developing countries. Pediatr Infect Dis J 009; 8 Suppl: S Viswanathan R, Singh AK, Mukherjee S et al. Aetiology and antimicrobial resistance of neonatal sepsis at a tertiary care centre in eastern India: a 3 year study. Indian J Pediatr 0; 78: Viswanathan R, Singh AK, Basu S et al. Multi-drug resistant gram negative bacilli causing early neonatal sepsis in India. Arch Dis Child Fetal Neonatal Ed 0; 97: F Livermore DM. Current epidemiology and growing resistance of Gram-negative pathogens. Korean J Intern Med 0; 7: Souli M, Galani I, Giamarellou H. Emergence of extensively drug-resistant and pandrug-resistant Gram-negative bacilli in Europe. Euro Surveill 008; 3: pii¼ Hooper DC. Efflux pumps and nosocomial antibiotic resistance: a primer for hospital epidemiologists. Clin Infect Dis 005; 40: Nordmann P, Poirel L, Walsh TR et al. The emerging NDM carbapenemases. Trends Microbiol 0; 9: Roy S, Singh AK, Viswanathan R et al. Transmission of imipenem resistance determinants during the course of an outbreak of NDM- Escherichia coli in a sick newborn care unit. J Antimicrob Chemother 0; 66: Petersen PJ, Jacobus NV, Weiss WJ et al In vitro and in vivo antibacterial activities of a novel glycylcycline, the 9-t-butylglycylamido derivative of minocycline (GAR-936). Antimicrob Agents Chemother 43: Chen YH, Lu PL, Huang CH et al. Trends in the susceptibility of clinically important resistant bacteria to tigecycline: results from the Tigecycline In Vitro Surveillance in Taiwan study, 006 to 00. Antimicrob Agents Chemother 0; 56: Downloaded from on 06 March 08 04

7 Roy et al. 4 Andrasevic AT, Dowzicky MJ. In vitro activity of tigecycline and comparators against Gram-negative pathogens isolated from blood in Europe ( ). Int J Antimicrob Agents 0; 39: 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 005; 49: Keeney D, Ruzin A, McAleese F et al. MarA-mediated overexpression of the AcrAB efflux pump results in decreased susceptibility to tigecycline in Escherichia coli. J Antimicrob Chemother 008; 6: Berçot B, Poirel L, Dortet L et al. In vitro evaluation of antibiotic synergy for NDM--producing Enterobacteriaceae. J Antimicrob Chemother 0; 66: Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Eighteenth Informational Supplement M00-S8. CLSI, Wayne, PA, USA, Galani I, Rekatsina PD, Hatzaki D et al. Evaluation of different laboratory tests for the detection of metallo-b-lactamase production in Enterobacteriaceae. J Antimicrob Chemother 008; 6: Roy S, Viswanathan R, Singh AK et al. Sepsis in neonates due to imipenem-resistant Klebsiella pneumoniae producing NDM- in India. J Antimicrob Chemother 0; 66: 4 3. Tenover FC, Arbeit RD, Goering RV et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrosphoresis: criteria for bacterial strain typing. JClinMicrobiol995; 33: Källman O, Motakefi A, Wretlind B et al. Cefuroxime non-susceptibility in multidrug-resistant Klebsiella pneumoniae overexpressing rama and acra and expressing ompk35 at reduced levels. J Antimicrob Chemother 008; 6: Gröbner S, Linke D, Schütz W et al. Emergence of carbapenemnon-susceptible extended-spectrum b-lactamase-producing Klebsiella pneumoniae isolates at the university hospital of Tübingen, Germany. J Med Microbiol 009; 58: 9. 4 Ge Q, Yamada Y, Zgurskaya H. The C-terminal domain of AcrA is essential for the assembly and function of the multidrug efflux pump AcrAB-TolC. J Bacteriol 009; 9: Hentschke M, Wolters M, Sobottka I et al. ramr mutations in clinical isolates of Klebsiella pneumoniae with reduced susceptibility to tigecycline. Antimicrob Agents Chemother 00; 54: Rosenblum R, Khan E, Gonzalez G et al. Genetic regulation of the rama locus and its expression in clinical isolates of Klebsiella pneumoniae. Int J Antimicrob Agents 0; 38: Viveiros M, Dupont M, Rodrigues L et al. Antibiotic stress, genetic response and altered permeability of E. coli. PLoS One 007; : e Downloaded from on 06 March 08