|Year : 2012 | Volume
| Issue : 4 | Page : 456-461
Detection of metallo-β-lactamases producing Acinetobacter baumannii using microbiological assay, disc synergy test and PCR
M Purohit1, DK Mendiratta1, VS Deotale1, M Madhan2, A Manoharan2, P Narang1
1 Department of Microbiology, Mahatma Gandhi Institute of Medical Sciences, Sevagram, Wardha, India
2 Department of Medicine Unit I & Infectious Diseases, Prof Benjamin M Pulimood Laboratories for Infection & Immunity, Christian Medical College, Vellore - 632 004, Tamilnadu, India
|Date of Submission||17-Feb-2012|
|Date of Acceptance||11-Apr-2012|
|Date of Web Publication||24-Nov-2012|
D K Mendiratta
Department of Microbiology, Mahatma Gandhi Institute of Medical Sciences, Sevagram, Wardha
Source of Support: None, Conflict of Interest: None
Background: One leading factor responsible for resistance in Acinetobacter baumannii, an important opportunist in health care institutions globally, is the production of carbapenamases like metallo-β-lactamases (MBLs), which hydrolyze a variety of β-lactams including penicillin, cephalosporins and carbapenems. However, neither any standard guidelines are available nor any method has been found to be perfect for their detection. Various methods have shown discordant results, depending upon the employed methodology, β-lactamase substrate and MBL inhibitor used. This study aims to evaluate two phenotypic methods against PCR as gold standard among carbapenem resistant A. baumannii for identifying MBL producers. Materials and Methods: A total of 130 A. baumannii were screened for imipenem and meropenem resistance by Kirby-Bauer disc diffusion method. Phenotypic expression of MBL was detected by EDTA-imipenem-microbiological (EIM) assay and extended EDTA disc synergy (eEDS) test and presence of bla-IMP and bla-VIM was detected by PCR in all the carbapenem resistant isolates. Results: Of the 43 imipenem and/or meropenem resistant A. baumannii isolates, 4 (9.3%) were found to be MBL producers by EIM and 3 (6.97%) by eEDS. Only bla-VIM gene was detected in 7 (16.28%) by PCR. In addition EIM detected 14 (32.56%) carbapenem resistant non-metallo enzyme producers. Conclusion: Of the two MBL genes targeted, bla-VIM was only detected and that too in isolates resistant to both imipenem and meropenem. Further, EIM was useful in differentiating MBL from non-metalloenzymes producers.
Keywords: metallo-β-lactamases, Acinetobacter baumannii, EIM, eEDS, PCR
|How to cite this article:|
Purohit M, Mendiratta D K, Deotale V S, Madhan M, Manoharan A, Narang P. Detection of metallo-β-lactamases producing Acinetobacter baumannii using microbiological assay, disc synergy test and PCR. Indian J Med Microbiol 2012;30:456-61
|How to cite this URL:|
Purohit M, Mendiratta D K, Deotale V S, Madhan M, Manoharan A, Narang P. Detection of metallo-β-lactamases producing Acinetobacter baumannii using microbiological assay, disc synergy test and PCR. Indian J Med Microbiol [serial online] 2012 [cited 2019 Oct 22];30:456-61. Available from: http://www.ijmm.org/text.asp?2012/30/4/456/103770
| ~ Introduction|| |
Acinetobacter baumannii, has emerged as an important opportunistic Gram-negative bacteria in health care institutions globally, as it resists desiccation, is hard to eradicate and has numerous intrinsic and acquired mechanisms of drug resistance. Production of carbapenamases, which hydrolyse carbapenems, the drugs with high efficacy and broad spectrum of activity against this organism, has been a cause of worry to the clinician and the microbiologist. 
Carbapenamases are β-lactamases, which include serine-β-lactamases (KPC, OXA, GES, etc.) and metallo-β-lactamases (MBLs). The latter require metal ion zinc for their activity, which is inhibited by metal chelators like EDTA and thiol-based compounds but not by sulbactam, tazobactam and clavulanic acid.
MBL production is typically associated with resistance to aminoglycosides and fluoroquinolones, further compromising therapeutic options. Among the seven types of MBL genes described throughout the World, bla-IMP and bla-VIM are the most common and have been reported from India too. , The genes responsible for MBL production may be chromosomal or plasmid mediated and pose a threat of horizontal transfer among other Gram-negative bacteria. 
Different phenotypic techniques for detection of MBL producers using different chelators have been described by various authors, ,,, however, no standard guidelines are available and no single method has been found to be perfect  as they have shown discordant results depending on the employed methodology, β-lactam substrate and/or MBL inhibitors used and bacterial genus tested.
The genotypic methods, namely, polymerase chain reaction (PCR) and iso-electric focusing (IEF) have been evaluated in different settings by various workers globally. ,,,,,, From India, however, only two studies have evaluated PCR for detection of MBL encoding gene among A. baumannii. ,
This study aims to evaluate two phenotypic methods for identifying MBL producers among carbapenem resistant A. baumannii using PCR as gold standard.
| ~ Material and Methods|| |
A total of 130 A. baumannii strains isolated by standard methods  from various clinical specimens (sputum, tracheal aspirate, pus, urine, blood, pleural fluid, ascitic fluid, cerebrospinal fluid, wound swabs, etc.) received in the department of Microbiology, from indoor and outdoor patients between July 2009 and June 2011 were included in this study. Positive control used in study was Pseudomonas aeruginosa (bla-VIM and bla-IMP positive) obtained from Dept. Immunología, Microbiología Universidad del País Vasco, Campus de Bizkaia, Leioa Bizkaia, Spain and negative control was A. baumannii obtained from Department of Biotechnology, School of Life Sciences, Puducherry.
Isolates showing resistance to either imipenem or meropenem by the Kirby-Bauer disc diffusion method according to clinical laboratory standards institute guidelines (CLSI, 2009)  were considered as screen test positives and were further tested for confirmation of MBL production as under.
| ~ EDTA-Imipenem Microbiological assay (EIM)|| |
MBL activity was detected in crude cell extract of the bacteria. Crude cell extract was prepared by disrupting bacterial cells by freezing and thawing. The bacteria were collected from the surface of a fresh overnight growth on Mueller-Hinton agar (MHA) and transferred to a pre-weighed microcentrifuge tube until the equivalent of 100 mg of bacterial wet weight was obtained. To this, 1 ml of 50 mm Tris-HCL (pH 8) was added. The microcentrifuge tube was centrifuged at 5000 rpm for 10 min and the pellet thus obtained was frozen at -20°C for 15 min and thawed at 37°C for 10 min. This freezing and thawing procedure was repeated 10 times. Finally, the tube was centrifuged at 10000 rpm for 10 min and the supernatant so obtained was assayed for detection of MBL activity. For detection of MBL activity in crude cell extract, indicator strain, Escherichia More Details coli ATCC 25922 was lawn cultured on MHA plate after adjusting its turbidity to 0.5 McFarland. The seeded plate was kept at room temperature for 5-10 min so as to allow the inoculum to dry. Within 15 min of inoculation, a 10 μg imipenem (oxoid) disc was placed in the centre of the agar surface and four blank filter paper discs were applied around the central imipenem disc in such a manner that these discs had an edge to edge distance of 2 mm from the central disc. The four peripheral filter paper discs were labelled as 'S', 'S/Zn', 'S/E' and 'B'. Disc 'S' was loaded with 20 μl of the prepared crude extract, disc 'S/Zn' with 20 μl of crude extract supplemented with 0.1 mM ZnSO 4 , disc 'S/E' with 20 μl of crude extract supplemented with 20 mM EDTA and disc 'B' with 20 μl of 50 mM Tris-HCL only. The plate was incubated overnight at 37°C [Figure 1]. After overnight incubation growth pattern of indicator strain E. coli within the sensitivity zone of imipenem around the four discs ('S', 'S/Zn', 'S/E' and 'B') was observed. Disc B was considered as negative test control and no growth of indicator strain was observed around this disc. Interpretation: Growth of indicator strain E. coli around 'S' and 'S/Zn' disc only, indicated MBL activity in crude cell extract and considered as MBL positive [Figure 1]a, no growth around 'S', 'S/Zn' and S/E disc indicated there is no β-lactamase activity [Figure 1]b considered as MBL negative and growth around 'S' , 'S/Zn' and 'S/ E' disc [Figure 1]c indicated production of non-metalloenzyme.
|Figure 1: EIM assay: (a) Growth of indicator strain around S and S/Zn disc but no growth around disc S/E indicate MBL production. (b) No growth of indicator strain around S, S/Zn and S/E disc indicate this strain neither produces MBL nor non-metalloenzyme. (c) Growth of indicator strain around S, S/Zn and S/E disc indicate non-metallo enzyme production|
Click here to view
| ~ Extended EDTA-disc synergy (eEDS) test|| |
This test was developed by Patricia et al. using imipenem, meropenem and ceftazidime with two concentrations (0.5 and 0.1 M) of EDTA. We performed this test using imipenem and meropenem only but with both the concentrations of EDTA. Lawn culture of the test, positive control and negative control organisms (0.5 McFarland ) were made on three sets of two plates. Each set was labelled as plate 1 and plate 2. On plate-1, disc of imipenem (10μg) and meropenem (10μg) were placed around a central blank disc in such a manner that the carbapenem discs had an edge to edge distance of 15 mm from the central blank disc. The blank disc was loaded with 10 μl of 0.5 M EDTA after placing the carbapenem disc [Figure 2]. On plate-2 procedure was repeated as described under plate 1 with two modifications. (i) Edge to edge distance between carbapenem disc from blank disc was 10 mm. (ii) The concentration of EDTA was 0.1 M. All the plates were incubated overnight at 37°C. Interpretation: Enhancement of zone of inhibition around imipenem and/or meropenem disc towards the EDTA disc in any of the two plates was considered as positive for MBL production [Figure 2].
| ~ Pcr|| |
Detection of the bla-IMP and bla-VIM genes was carried out using primers as described by Mostachio et al. bla-IMP primers: bla-IMP-F (5′-GAATAGAATGGTTAACTCTC-3′) bla-IMP-R(5′-CCAAACCACTAGGTTATC-3′) and bla-VIM primers: bla-VIM-F (5′- GTTTGGTCGCATATCGCAAC-3′) bla-VIM-R(5′-AATGCGCAGCACCAGGATAG-3′) were used.
Extraction of DNA
A total of 4-5 identical colonies of A. baumannii were re-suspended in 500 μl of sterile saline in 1.5 ml eppendorf tube. This was boiled at 100°C for 10 min, centrifuged at 8000 rpm for 5 min and the supernatant containing DNA was used for further processing.
The PCR mixture used was as follows: 1 μl DNA template in a 49 μl mixture containing 10 mM Tris/HCl (pH 8.8), 50 mM KCl, 4 mM MgCl 2 , 200 μM each dNTP (Fermentas Genetix Biotech Pvt. Ltd., New Delhi), 1 μl of each of the forward and reverse primers (Bangalore genei) and 1 unit Taq DNA polymerase (Fermentas Genetix Biotech Pvt. Ltd., New Delhi).
The PCR conditions included: Initial denaturation at 94°C for 5 min followed by 33 cycles each of 94°C for 25 s, 53°C for 40 s and 72°C for 50 s, followed by a single final elongation step at 72°C for 6 min. The PCR product of 188 bp for bla-IMP and 382 bp for bla-VIM was visualised by 1.5% agarose gel electrophoresis containing ethidium bromide 0.5 μg/ml (Bangalore Genei) [Figure 3].
|Figure 3: PCR for confirmation of MBL genes bla-VIM and bla-IMP. All the four test strains are positive for bla-VIM at 38 2bp. Lane-1. Negative control, Lane-2. Control bla-IMP(188 bp), Lane-3. Control bla-VIM (382 bp), Lane-4. test strain 1 (382 bp), Lane-5. 100– 1500 bp ladder (Fermentas), Lane-6. test strain 2 (382 bp), Lane-7. test strain 3 (382 bp), Lane-8. test strain 4 (382 bp)|
Click here to view
| ~ Results|| |
Out of 130 A. baumannii isolates tested, 43 (33.08%) were found resistant to at least one or both carbapenem tested. Out of these, 36 (27.69%) were resistant to both imipenem and meropenem, 38 (29.23%) to meropenem and 41 (31.54%) to imipenem alone [Table 1].
Of the 43 screen test positives, EIM could confirm 4 (9.3%) as MBL producers and 14 (32.56%) as non-metalllo-enzyme producers, eEDS could confirm 3 (6.97%) isolates using 0.5 M EDTA concentration and 2 (4.65%) using 0.1 M EDTA concentration as MBL producers. PCR detected bla-VIM type of MBL among 7 (16.28%) of the 43 screen test positives. None of the 43 screen test positives showed bla-IMP type MBL gene [Table 2]. The positivity of EIM, eEDS and PCR with respect to carbapenem resistance is given in [Table 2].
|Table 2: Percentage MBL positivity with respect to carbapenem resistant status|
Click here to view
Of the four isolates confirmed for MBL by EIM, two were bla-VIM positive and similarly among three eEDS positives only two showed presence of bla-VIM. None of the 14 non-metalloenzymes producers detected by EIM showed either bla-IMP or bla-VIM.
| ~ Discussion|| |
MBLs have been identified from clinical isolates in members of Enterobacteriaceae, P. aeruginosa and Acinetobacter spp over the past few years. Strains producing these enzymes have been responsible for nosocomial outbreaks that have been accompanied by serious and prolonged infections. MBLs are powerful carbapenemases and can hydrolyze a wide variety of β-lactams, including penicillins, cephalosporins and carbapenems. Since the initial isolation of carbapenem resistant A. baumannii producing bla-IMP-1 and bla-VIM-1 type MBL in Japan and Italy, respectively,  clinical isolates of these strains have been identified worldwide.
The current CLSI document  has no guidelines for detecting MBLs, however, it has recommended modified Hodge test for detection of carbapenamases but in members of Enterobacteriaceae only.
Keeping PCR as gold standard, we found that MBL positivity among isolates selected on the basis of resistance to both imipenem and meropenem was higher (19.44%) than those selected on the basis of resistance to imipenem (17.07%) or meropenem (18.43%) alone [Table 2]. Noyal et al. and Kumar et al. reported MBL positivity as 6.5% and 21%, respectively, among isolates picked up on basis of meropenem resistance only, Singh et al. and Gupta et al. reported 22.22% and 87%, respectively, positivity in IPM resistant, while Amudhan et al. in 46.55% of isolates picked up on the basis of resistance to both IPM and MEM. We could not come across studies comparing carbapenem IPM and/or MEM as screening agents and further confirmation of such positive isolates for MBL production. However, as the PCR positivity was highest (19.44%) among strains that were resistant to both IPM and MEM and the fact that none of the isolates that were resistant to one carbapenem and sensitive to other showed the MBL gene by PCR, we feel that only those isolates of A. baumannii which show resistance to both IPM and MEM should be suspected to be MBL producer and if facilities are available they should be further tested by confirmatory methods.
The EDTA-imipenem-microbiological assay, which differentiates metalloenzymes from non-metalloenzymes, was evaluated for the first time in A. baumanii in our study. By this assay we could confirm 4 isolates (9.3%) as MBL producers and 14 (32.56%) as non-metalloenzyme producers among the 43 screen test positives [Table 2]. However, of the 4 MBL positives only 2 showed presence of bla-VIM and none of the 14 non-metalloenzyme producers showed presence of either bla-IMP or bla-VIM. "False" positivity of EIM (two isolates) against the PCR, could be due to presence of either other uncommon MBL encoding gene like SIM, SPM, GIM, AIM or variants of IMP and VIM.  Very high MBL positivity of 95.2% and 88% in P. aeruginosa using EIM has been reported by Singh et al. and Jain et al.,  respectively. However, Quinones-Falconi et al. using PCR as a gold standard reported only 3.5% MBL positivity in P. aeruginosa using EIM. Galicia et al. have reported 3.4% MBL positivity in P. aeruginosa using EIM with excellent specificity but poor sensitivity of the test. We did not come across any study on EIM in A. baumannii.
It has been emphasized that determination of the correct mechanism of carbapenem resistance is of utmost importance to decide the appropriate therapeutic regimen of the carbapenem resistant non-fermenters. In several studies, intravenous colistin combined with rifampin and imipenem have been recommended for the treatment of carbapenem-resistant isolates lacking MBLs, whereas the combination of colistin and rifampicin (with or without tigecycline) have been recommended for treatment of MBL producing carbapenem-resistant isolates.  In our study 32.6% carbapenem resistant isolates were found to be non-metalloenzyme and 9.3% as MBL producers by EIM, thereby suggesting that EIM could help in deciding the appropriate treatment option in carbapenem resistant isolates.
The eEDS picked up one false positive in our study when higher concentration of EDTA (0.5 M) was used as against none using 0.1 M EDTA, when compared with PCR as gold standard. Similar observations have been made by Patricia et al. and Zerrine et al.,  who advocate addition of lower concentration of EDTA disc also along with higher concentration disc to avoid undesirable or confusing results due to sensitivity of isolates to higher concentration of EDTA.
In our study, we found bla-VIM MBL gene only in 7 (16.28%) of the 43 screen test positive isolates. None of our isolates showed presence of bla-IMP gene. Uma et al. from Puducherry reported only bla-IMP-1 in 42% of A. baumannii, while Amudhan et al. from Chennai reported bla-VIM in 46.55% with and both bla-IMP and bla-VIM in only one isolate of A. baumannii. The non-demonstration of MBL gene in 36 of our 43 screen test positive isolates could be either due to presence of (i) unidentified MBL gene (ii) limitation of the primer set used either with regards to picking up the variant IMP/VIM gene or because of presence of MBL genes other than IMP/VIM (iii) presence of other enzymes [OXA like (Ambler class D) carbapenamases AmpC β-lactamases] or (iv) other mechanism of carbapenem resistance, namely; loss of porins, increase in efflux pump activity alteration in penicillin binding proteins (PBPs), etc.
Finally, we propose that more studies on EIM which differentiates metallo from non-metalloenzyme producers and PCR using universal primers capable of detecting all the prevalent MBL gene need to be undertaken.
| ~ Conclusion|| |
Resistance to both imipenem and meropenem was found to be a better indicator of MBL production. EIM assay appears to be useful in differentiating MBL and non-metalloenzyme producers. Of the two MBL gene (bla-VIM and bla-IMP) targeted, bla-VIM was detected in isolates studied.
| ~ References|| |
|1.||Peleg AY Seifert H, Paterson DL. Acinetobacter baumannii: Emergence of a successful pathogen. Clin Microbiol Rev 2008;21:538-82. |
|2.||Uma Karthika R, Srinivasa Rao R, Sahoo S, Shashikala P, Kanungo R, Jayachandran S, et al. Phenotypic and genotypic assays for detecting the prevalence of metallo-β-lactamases in clinical isolates of Acinetobacter baumannii from a South Indian tertiary care hospital. J Med Microbiol 2009;58:430-5. |
|3.||Amudhan SM, Sekar U, Arunagiri K, Sekar B. OXA β- lactamase mediated carbapenem resistance in Acinetobacter baumannii . Indian J Med Microbiol 2001;29:269-74. |
|4.||Varaiya A, Kulkarni N, Kulkarni M, Bhalekar P, Dogra J. Incidence of metallo-β-lactamase producing Pseudomonas aeruginosa in ICU patients. Indian J Med Res 2008;127:398-402. |
|5.||Franklin C, Liolios L, Peleg AY. Peleg Phenotypic detection of carbapenem-susceptible metallo-β-lactamase producing Gram-negative bacilli in the clinical laboratory. J Clin Microbiol 2006;44:3139-44. |
|6.||Picão RC, Andrade SS, Nicoletti AG, Campana EH, Moraes GC, Mendes RE, et al. Metallo-β-lactamase detection: Comparative evaluation of double-disk synergy versus combined disk tests for IMP-, GIM-, SIM-, SPM-, or VIM- producing isolates. J Clin Microbiol 2008;46:2028-37. |
|7.||Marchiaro P, Mussi MA, Ballerini V, Pasteran F, Viale AM, Vila AJ, et al. Limansky. Sensitive EDTA-based microbiological assays for detection of metallo-β-lactamases in nonfermentative Gram-negative bacteria. J Clin Microbiol 2005;43:5648-52. |
|8.||Jain S, Rai S, Prasad KN, Dhole TN, Maurya A. Detection of carbapenemase enzymes in clinical isolates of Pseudomonas aeruginosa by remodified Hodge test and other phenotypic methods. Eur Soc Clin Microbiol Infect Dis 2011:Abstract P768. |
|9.||Singh SP, Shariff M, Barua T, Thukral SS. Comparative evaluation of phenotypic tests for identification of metallo-β-lactamases producing clinical isolates of Pseudomonas aeruginosa. Indian J Med Res 2009;129:713-5. |
|10.||Mostachio AK, van der Heidjen I, Rossi F, Levin AS, Costa SF. Costa. Multiplex PCR for rapid detection of genes encoding oxacillinases and metallo-β-lactamases in carbapenem resistant Acinetobacter spp. J Med Microbiol 2009;58:1522-4. |
|11.||Lee K, Lee WG, Uh Y, Ha GY, Cho J, Chong Y; Korean Nationwide Surveillance of Antimicrobial Resistance Group. VIM- and IMP-Type Metallo-β-lactamase-Producing Pseudomonas spp. and Acinetobacter spp. in Korean Hospitals. Emerg infect dis 2003;9:868-71. |
|12.||Ellington MJ, Kistler J, Livermore DM, Woodford N. Multiplex PCR for rapid detection of genes encoding acquired metallo-β-lactamases. J Antimicrob Chemother 2007;59:321-2. |
|13.||Aktaº Z, Kayacan CB. Investigation of metallo- β-lactamase producing strains of Pseudomonas aeruginosa and Acinetobacter baumannii by E-test, disk synergy and PCR. Scand J Infect Dis 2008;40:320-5. |
|14.||Murray PR, Baron EJ, Pfaller MA, Jorgensen JH, Landry ML. Acinetobacter. Manual of Clinical Microbiology. 9 th ed., vol. 1:772. Washington DC: American Society for Microbiology; 2007;2.p.772. |
|15.||Clinical Laboratory Standards Institute. Performance standards for antimicrobial disk susceptibility testing. 9 th ed. M2-A9., Wayne, PA: CLSI; 2009. p. 46 |
|16.||Noyal MJ, Menezes GA, Harish BN, Sujatha S, Parija SC. Simple screening tests for detection of carbapenemases in clinical isolates of nonfermentative gram-negative bacteria. Indian J Med Res 2009;129:707-12. |
|17.||Kumar AV, Pillai VS, Dinesh KR, Karim S. The phenotypic detection of carbapenemase in meropenem resistant Acinetobacter calcoaceticus-baumannii complex in a tertiary care hospital in South India. J Clin Diagn Res 2011;5:223-6. |
|18.||Gupta V, Singla N, Chander J. Use of two double disk synergy tests to detect metallo-β-lactamases in nonfermenters. Indian J Med Res 2008;128:671-2. |
|19.||Walsh TR, Toleman MA, Poirel L, Nordmann P. Metallo-β-lactamases: The Quiet before the storm. Clin Microbiol Rev 2005;18:306-25. |
|20.||Quinones-Falconi F, Galicia-Velasco M, Marchiaro P, Mussi MA, Ballerini V, Vila AJ, et al. Emergence of Pseudomonas aeruginosa strains producing metallob-β-lactamases of the IMP-15 and VIM-2 types in Mexico. Clin Microbiol Infect 2010;16:126-31. |
|21.||Galicia M, Bermejo K, Limansky A, Cerón C, Quinones F. Comparison of phenotypic techniques for the detection of metallo-β-lactamases in Pseudomonas aeruginosa from respiratory isolates. Eur J Clin Microbiol Infect Dis 2009:Abstract P2068. |
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2]
|This article has been cited by|
||Phenotypic characterization and colistin susceptibilities of carbapenem-resistant of Pseudomonas aeruginosa and Acinetobacter spp
| ||Mohanty, S., Maurya, V., Gaind, R., Deb, M. |
| ||Journal of Infection in Developing Countries. 2013; 7(11): 880-887 |