- Email: [email protected]
Detection of AmpC beta-lactamases using sodium salicylate
Journal of Microbiological Methods 91 (2012) 354–357
Contents lists available at SciVerse ScienceDirect
Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth
Detection of AmpC beta-lactamases using sodium salicylate Mona T. Kashif, Aymen S. Yassin ⁎, Alaa El-Dien M.S. Hosny Department of Microbiology and Immunology, Faculty of Pharmacy, Cairo University, Cairo, 11562, Egypt
a r t i c l e
i n f o
Article history: Received 4 August 2012 Received in revised form 22 September 2012 Accepted 29 September 2012 Available online 8 October 2012 Keywords: AmpC β-lactamases Sodium salicylate
a b s t r a c t AmpC β-lactamases are enzymes that hydrolyze all β-lactam antibiotics except cefipime and imipenem. Currently, there is no standard phenotypic method for detection of such enzymes. This study aims to report the use of sodium salicylate for AmpC β-lactamases detection and to compare its sensitivity and specificity to other commonly known inhibitors. A total of 135 clinical isolates were used to test the effectiveness of sodium salicylate in detection of plasmid- as well as chromosomally encoded AmpC β-lactamases. All isolates were tested by multiplex PCR testing as well as inhibitor-based methods using cloxacillin, phenylboronic acid and sodium salicylate for the detection of AmpC enzymes. Four isolates were confirmed as producers of plasmid-encoded AmpC β-lactamase and a single isolate was confirmed to have both plasmid and chromosomal genes. Cloxacillin and phenylyboronic acid failed to detect most of the plasmid-encoded enzymes. Sodium salicylate was able to detect the Escherichia coli isolates with plasmid-encoded enzymes in addition to few other isolates that were chromosomally mediated. The sensitivity and specificity of sodium salicylate was 50% and 93%, respectively, higher than those of other known inhibitors. We thus conclude that sodium salicylate can be reliably used as an inhibitor in the detection of plasmid-encoded AmpC enzymes in E. coli. © 2012 Elsevier B.V. All rights reserved.
1. Introduction AmpC β-lactamases are cephalosporinases that are active on penicillins and more active on cephalosporins. They can hydrolyze cephamycins such as cefoxitin and cefotetan; oxyiminocephalosporins such as ceftazidime, cefotaxime, and ceftriaxone and monobactams such as aztreonam. However, they have low affinity for cefipime and carbapenem (Jacoby, 2009). They are not well inhibited by clavulanic acid, but are often inhibited by a low concentration of aztreonam or cloxacillin (Bush et al., 1995). They are chromosomally encoded in the majority of Enterobacteriaceae (Livermore, 1995) where they are inducible except in Escherichia coli and Shigella spp. where they are constitutively expressed at very low level that do not cause resistance. On the other hand, Acinetobacter baumannii, Klebsiella pneumoniae, Klebsiella oxytoca, Proteus mirabilis, Salmonella spp. and Stenotrophomonas maltophilia have been found to lack chromosomal AmpC β-lactamase genes (Jacoby, 2009). Recently, plasmid-encoded AmpC β-lactamases have been detected even in species known not to have chromosomally encoded enzymes (Hanson, 2003). Plasmids carrying AmpC β-lactamase genes range in size from 7 to 180 kb (Philippon et al., 2002).
⁎ Corresponding author at: Faculty of Pharmacy, Cairo University, Department of Microbiology and Immunology, Kasr Eleini St., Cairo, 11562, Egypt. Tel.: +20 2 25353100 200 300 400, +20 100 9610341. E-mail address: [email protected] (A.S. Yassin). 0167-7012/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mimet.2012.09.035
Although many phenotypic tests are available for AmpC detection, currently, there is no method for their detection listed by the Clinical Laboratory Standard Institute (CLSI). The available phenotypic tests include enzyme extraction methods, such as three dimensional assay and cefoxitin agar medium based assay (Thomson et al., 1984; Manchanda and Singh, 2003; Nasim et al., 2004). AmpC disk test (Singhal et al., 2005) and tests incorporating AmpC β-lactamase inhibitors such as boronic acid or cloxacillin have been also described (Coudron, 2005; Jeong et al., 2009; Brenwald et al., 2005). Multiplex PCR is also used as a rapid screening tool for detection and differentiation between the six plasmid-encoded AmpC β-lactamase groups (Pérez-Pérez and Hanson, 2002). All these tests are not widely attempted by clinical laboratories either because they are inconvenient and subjective, because they lack sensitivity and specificity, or because they require reagents that are not readily available (Black et al., 2005). Recently, several articles reported that sodium salicylate induces the expression of marRAB operon in several Gram negative bacteria causing an increase in minimum inhibitory concentration (MIC) of various tested antibiotics (Martin et al., 1996; Tavío et al., 2004, 2005). However, in strains with constitutive overexpression of AmpC enzymes, sodium salicylate addition caused reduction in β-lactam antibiotic MIC. This reduction in MIC was found to be due to the repression of AmpC expression by sodium salicylate. In this study, sodium salicylate is tested for its usefulness for detection of AmpC β-lactamase, and its sensitivity and specificity are compared to other commonly known inhibitors (namely, cloxacillin and phenylboronic acid).
M.T. Kashif et al. / Journal of Microbiological Methods 91 (2012) 354–357
2. Materials and methods 2.1. Isolates Isolates used for this study were selected based on their resistance to cefoxitin or to either cefotaxime or ceftazidime (3rd generation cephalosporin) and their lack of extended spectrum beta lactamases (ESBL) expression and consequently, their positive screening for AmpC enzyme (Mirelis et al., 2006). All isolates were identified by Api 20E identification (BioMerieux, France). 2.2. Polymerase chain reaction Plasmid-carried genes encoding AmpC β-lactamases were detected by using PCR according to previously described protocols (Pérez-Pérez and Hanson, 2002). Clinical isolates of Citrobacter freundii, Enterobacter cloacae and Morganella morganii were used as positive controls. A list of primers used and expected product sizes is provided (Table 1). Primers were diluted to a concentration of 1 pmol/μl. PCR was performed in a final volume of 25 μl. Each reaction contained 2.5 μl of 10× PCR buffer, 0.5 μl (100 μM) deoxynucleoside triphosphate mixture, 2 pmol of primers CITM-F and CITM-R, 1.5 pmol of primers DHAM-F, DHAM-R, MOXM-F, MOXM-R, ACCM-F and ACCM-R, 1.25 pmol of primers EBCM-F and EBCM-R, 1 pmol of primers of FOXM-F and FOXM-R and 0.625 U of Taq DNA polymerase. All primers and PCR reagents were supplied by Fermentas (Thermo Fisher Scientific, USA). For template DNA preparation, a single colony of each organism was inoculated into 5 ml of Luria-Bertani broth and incubated for 20 h at 37 °C with shaking. Cells from 1.5 ml of the overnight culture were harvested by centrifugation at 17,310 ×g for 5 min. After the supernatant was decanted, the pellets were re-suspended in 500 μl of distilled water. The cells were lysed by heating at 95 °C for 10 min, and cellular debris was removed by centrifugation at 17,310 ×g for 5 min. One microliter of the supernatant was used as the source of template for amplification. For confirmation of plasmid origin of the multiplex PCR products, plasmids from PCR positive isolates were isolated using GenElute™ plasmid miniprep kit (Sigma-Aldrich, USA) and were then re-analyzed by multiplex PCR. PCR amplification products were purified using PCR purification kit (Qiagen, Germany) and sequenced. Sequences were aligned and analyzed by the BLAST algorithm (Altschul et al., 1990) available as part of the NCBI website (URL: http://blast. ncbi.nlm.nih.gov/Blast.cgi). 2.3. Inhibitor based assays The inhibitory activity of both phenylboronic acid and cloxacillin was tested by disc diffusion method using inoculated Mueller-Hinton agar plates. Discs for testing different inhibitors were prepared by
Table 1 The sequence of primer pairs used for detection of each enzyme family and the expected amplicon size. Enzyme family
Primer
Sequence
Expected amplicon size (bp)
MOX
MOXM-F MOXM-R CITM-F CITM-R DHAM-F DHAM-R ACCM-F ACCM-R EBCM-F EBCM-R FOXM-F FOXM-R
5′-GCT GCT CAA GGA GCA CAG GAT-3′ 5′-CAC ATT GAC ATA GGT GTG GTG C 5′-TGG CCA GAA CTG ACA GGC AAA-3′ 5′-TTT CTC CTG AAC GTG GCT GGC-3′ 5′-AAC TTT CAC AGG TGT GCT GGG T-3′ 5′-CCG TAC GCA TAC TGG CTT TGC-3′ 5′-AAC AGC CTC AGC AGC CGG TTA-3′ 5′-TTC GCC GCA ATC ATC CCT AGC-3′ 5′-TCG GTA AAG CCG ATG TTG CGG-3′ 5′-CTT CCA CTG CGG CTG CCA GTT-3′ 5′-AAC ATG GGG TAT CAG GGA GAT G-3′ 5′-CAA AGC GCG TAA CCG GAT TGG-3′
520
CIT DHA ACC EBC FOX
462 405 346 302 190
355
applying 20 μl of inhibitor solution to each cefoxitin disc (30 μg), discs were allowed to dry for 30 min and then used immediately. Cloxacillin sodium was used as a solution in sterile distilled water (10 μg/μl) and phenylboronic acid was dissolved in DMSO (20 μg/μl). Plates were incubated overnight at 35 °C. An increase of 4 and 5 mm in cefoxitin zone diameter, in presence of cloxacillin and phenylboronic acid, respectively, indicated that the inoculated isolate was an AmpC producer (Coudron, 2005; Tan et al., 2009). 2.4. Testing sodium salicylate The MIC of sodium salicylate was determined first against all the different groups of isolates and a sub inhibitory concentration of (5 mg/ml) in distilled water was used for further testing by broth microdilution method. The MIC of cefoxitin alone and in presence of sodium salicylate was determined and the reduction of MIC, in presence of sodium salicylate was calculated and correlated to the multiplex PCR results and to the presence of inducible chromosomal enzymes. 2.5. Data analysis The performances of various phenotypic test methods, for the detection of plasmid-encoded AmpC β-lactamases, were evaluated by the comparison of their results to that of multiplex PCR method (Lee et al., 2009). Sensitivity and specificity were calculated as follows: Sensitivity ¼ number of true positive isolates by phenotypic test= number of positive isolates by multiplex PCR: Specificity ¼ number of true negative isolates by phenotypic test= number of negative isolates by multiplex PCR: Calculations were carried out as described in the UK standards for Microbiological Investigations: (http://www.hpa.org.uk/webc/HPAwebFile/ HPAweb_C/1317131674973). 3. Results One hundred and thirty five clinical isolates were used to test the effectiveness of sodium salicylate in detection of plasmid — as well as chromosomally encoded AmpC β-lactamases. Among this collection of isolates, 89 were from organisms lacking inducible chromosomal AmpC β-lactamases and they were identified as follows: E. coli (39 isolates), Klebsiella spp. (27 isolates), Acinetobacter spp. (16 isolates), S. maltophilia (five isolates) and Proteus mirabilis (two isolates). Forty-six isolates were from organisms confirmed of having inducible chromosomal AmpC β-lactamase genes and these were identified as follows: Pseudomonas spp. (33 isolates), Citrobacter spp. (five isolates), E. cloacae (four isolates), Serratia marcescens (two isolates) and M. morganii (two isolates). Multiplex PCR showed that only four isolates had the plasmidencoded AmpC β-lactamase and that a single S. marcescens isolate possessed both the plasmid and chromosomal genes (Fig. 1 and Table 2). Sequencing the PCR products showed that the ampC β-lactamase genes of these isolates belonged to the CIT group genes (CMY-4 genes) in case of S. marcescens and one Klebsiella isolate, CIT (CMY-6 gene) in one E. coli isolate and DHA group genes (DHA-1 genes) in one Klebsiella and one E. coli isolates. The inhibitor-based method using cloxacillin and phenylboronic acid was evaluated by the disc diffusion test. The use of each of these inhibitors was associated with high incidence of false positive results. In addition, both reagents failed to detect plasmid- or chromosomally encoded AmpC β-lactamases with few exceptions (Table 2).
356
M.T. Kashif et al. / Journal of Microbiological Methods 91 (2012) 354–357
Fig. 1. Multiplex PCR of the five positive isolates. M: 100 bp marker, CIT, DHA and EBC are positive controls showing the expected sizes. A and C: Klebsiella spp. B: Serratia marcescens, D and E: Escherichia coli.
When sodium salicylate was tested by broth microdilution method, two cut-off values were evaluated: ≥ 1 fold and ≥ 2 folds reduction in MIC values of cefoxitin in presence of sodium salicylate. For both cut off values, in case of E. coli, the two isolates harboring the plasmid-encoded AmpC β-lactamases were detectable; however, none of the isolates with the plasmid-encoded genes were detectable, in case of Klebsiellae spp. and S. marcescens. On the other hand, 28 isolates were recorded as false positive when a cut-off values of ≥1 fold was used while six isolates were recorded as false positive when a cut off value of ≥ 2 fold was used. In addition, out of 46 isolates containing inducible chromosomal AmpC β-lactamases, sodium salicylate detected six isolates when a cut-off value of ≥ 1 reduction in MIC values of cefoxitin was used and four isolates when a cut off value of ≥ 2 fold was used. The specificity value recorded was 59% when a cut-off value of ≥1 reduction in MIC values of cefoxitin was applied and 93% when a cut-off value of ≥2 was applied. Therefore, ≥ 2 fold reduction in cefoxitin MIC was found to be the best cut-off value due to the lower number of false positive isolates detected (higher specificity) and was used for comparative data analysis. When all the inhibitors used for detection of plasmid-encoded AmpC β-lactamases in isolates lacking inducible chromosomal genes were compared, the highest sensitivity (50%) was achieved by the use of sodium salicylate, followed by both cloxacillin and phenylboronic acid (25%).
Specificity values were highest value with cloxacillin (94%) followed by sodium salicylate (93%) and then by phenylboronic acid (61%). 4. Discussion The available phenotypic tests used in the detection of AmpC β-lactamases suffer from low specificity and sensitivity; in addition, they are subjective and inconvenient to be carried out in a routine manner. In this study, sodium salicylate is evaluated for its use in the detection of AmpC β-lactamases based on its inhibitor effect. The evaluation was carried out on clinical isolates that were tested positive for the presence of AmpC β-lactamases. The recorded results showed that the use of sodium salicylate preferably detects AmpC enzymes by broth microdilution method. The best overall sensitivity and specificity (50% and 93%, respectively) were obtained when sodium salicylate was used as inhibitor, with a cut-off value of ≥2 folds reduction in cefoxitin MIC compared to inhibitor based assay using cloxacillin and phenyl boronic acid which are most widely used in AmpC β-lactamase detection (Ingram et al., 2011). In a previous study by Tan and coworkers, both cloxacillin and phenylboronic acid demonstrated 95% specificity (Tan et al., 2009). These findings agree with the specificity recorded in this study for cloxacillin, however, it is much higher than what we observed when
Table 2 Comparison of different tested phenotypic methods for detection of AmpC β-lactamases. Cloxcillin and phenylboronic acid were determined by disc diffusion method, while sodium salicylate was determined by MIC broth dilution method. Genera of screen positive bacteria
AmpC enzyme origin
Multiplex PCR
Detectability of AmpC ß-lactamases using Cloxacillin
Escherichia coli
P/C
Klebsiellae spp.
P
Stenotrophomonas maltophilia Acinetobacter spp. Proteus mirabilis Serratia marcescens
− C − P/IC
Pseudomonas spp. Citrobacter spp. Enterobacter cloacae Morganella morganii
IC IC IC IC
Phenylboronic acid
Sodium salicylate
Result
No. of isolates (total = 135)
+
−
+
−
+
−
+ − + − −
2 37 2 25 5
0 2 1 1 0
2 35 1 24 5
0 16 1 7 0
2 21 1 18 5
2 1 0 3 0
0 36 2 22 5
− − + − − − − −
16 2 1 1 33 5 4 2
1 1 0 0 0 5 4 1
15 1 1 1 33 0 0 1
10 0 0 0 1 5 4 0
6 2 1 1 32 0 0 2
2 0 0 0 1 3 0 0
14 2 1 1 32 2 4 2
P: Have plasmid mediated AmpC enzyme, C: have chromosomal AmpC genes but don't cause resistance, IC: have inducible chromosomal AmpC enzymes.
M.T. Kashif et al. / Journal of Microbiological Methods 91 (2012) 354–357
using phenylboronic acid as an inhibitor. This discrepancy may be due to the inhibitory effect of phenylboronic acid on β-lactamase enzymes other than AmpC (KPC-type) (Doi et al., 2008). Also, the recorded sensitivity for both cloxacillin and phenylboronic acid is much lower than that recorded in the same study by Tan and coworkers (Tan et al., 2009). Similar discrepancies were reported by other groups (Ingram et al., 2011), and those may be due to the presence of other resistance mechanism or to emergence of resistance to cloxacillin as a β-lactam antibiotic. The discrepancies may be also be caused by the failure of phenylboronic acid to detect enzymes from DHA origin (one E. coli and one Klebsiellae spp.) as recorded by Lee and coworkers (Lee et al., 2009). It is noteworthy that all tested inhibitor failed to detect most Pseudomonas isolates and this may be due to that Pseudomonas is less permeable than Enterobacteriaceae and is better able to efflux toxic molecules via MexAB-OprM and other systems as suggested by Mushtaq et al. (2010). The low sensitivity of sodium salicylate when used to detect the enzyme of either plasmid or chromosomal origin is most likely because sodium salicylate failed to detect the majority of inducible chromosomal AmpC β-lactamases (detected four out of 46). We suggest that this low sensitivity with chromosomally encoded enzyme is due to the low level of β-lactamase production in the inducible strains which explains why the effect of salicylate on gene expression did not show significant changes in the MIC values (Tavío et al., 2005). 5. Conclusion Sodium salicylate can be used in the detection of plasmid-encoded and chromosomally overexpressed AmpC β-lactamases in E. coli. To the best of our knowledge, this is the first report suggesting the use of sodium salicylate in detection of AmpC enzymes. It is highly specific for AmpC enzyme and is sensitive in detection of E. coli plasmidencoded AmpC enzymes. It is cheaper and more readily available. Further studies will decipher its mechanism of action as well as testing its sensitivity and specificity on all families of plasmid AmpC enzymes so that it could be used as standard method. Competing interest statement The authors declare that they have no competing interest. Acknowledgments We thank Dr. Ramy Karam Aziz from Cairo University for critically reading the manuscript and for the useful suggestions. References Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215 (3), 403–410 (5). Black, J.A., Thomson, K.S., Buynak, J.D., Pitout, J.D.D., 2005. Evaluation of β-lactamase inhibitors in disc tests for detection of plasmid-mediated AmpC β-lactamases in well-characterized clinical strains of Klebsiella spp. J. Clin. Microbiol. 43, 4168–4171.
357
Brenwald, N.B., Jevons, G., Andrews, J., Ang, L., Fraise, A.P., 2005. Disc methods for detecting AmpC β-lactamase-producing clinical isolates of Escherichia coli and Klebsiella pneumoniae. J. Antimicrob. Chemother. 56, 600–601. Bush, K., Jacoby, G.A., Medeiros, A.A., 1995. A functional classification scheme for blactamases and its correlation with molecular structure. Antimicrob. Agents Chemother. 39, 1211–1233. Coudron, P.E., 2005. Inhibitor-based methods for detection of plasmid-mediated AmpC β-lactamases in Klebsiella spp., Escherichia coli, and Proteus mirabilis. J. Clin. Microbiol. 43, 4163–4167. Doi, Y., Potosk, B.A., Adams-Haduch, J.M., Sidjaba, H.E., Pasculle, A.W., Paterson, D.L., 2008. Simple disc-based method for detection of Klebsiella pneumoniae carbapenemase-Type β-lactamase by use of a boronic acid compound. J. Clin. Microbiol. 46, 4083–4086. Hanson, N.D., 2003. AmpC β-lactamases: what do we need to know for the future? J. Antimicrob. Chemother. 52, 2–4. Ingram, P.R., Inglis, T.J., Vanzetti, T.R., Henderson, B.A., Harnett, G.B., Murray, R.J., 2011. Comparison of methods for AmpC β-lactamase detection in Enterobacteriaceae. J. Med. Microbiol. 60 (Pt 6), 715–721. Jacoby, G.A., 2009. AmpC β-Lactamases. Clin. Microbiol. Rev. 22, 161–182. Jeong, S.H., Song, W., Kim, J.S., Kim, H.S., Lee, K.M., 2009. Broth microdilution method to detect extended-spectrum beta-lactamases and AmpC beta-lactamases in Enterobacteriaceae isolates by use of clavulanic acid and boronic acid as inhibitors. J. Clin. Microbiol. 47, 3409–3412. Lee, W., Jung, B., Hong, S.G., Song, W., Jeong, S.H., Lee, K., Kwak, H., 2009. Comparison of 3 phenotypic-detection methods for identifying plasmid-mediated AmpC βlactamase-producing Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis strains. Korean J. Lab. Med. 29, 448–454. Livermore, D.M., 1995. β -lactamases in laboratory and clinical resistance. Clin. Microbiol. Rev. 8, 557–584. Manchanda, V., Singh, N.P., 2003. Occurrence and detection of AmpC β-lactamases among Gram-negative clinical isolates using a modified three-dimensional test at Guru Tegh Bahadur Hospital, Delhi, India. J. Antimicrob. Chemother. 51, 415–418. Martin, R.G., Jair, K., Wolf, R.E., Rosner, J.L., 1996. Autoactivation of the marRAB multiple antibiotic resistance operon by the MarA transcriptional activator in Escherichia coli. J. Bacteriol. 178, 2216–2223. Mirelis, B., Rivera, A., Miró, E., Mesa, R.J., Navarro, F., Coll, P., 2006. A simple phenotypic method for differentiation between acquired and chromosomal AmpC betalactamases in Escherichia coli. Enferm. Infecc. Microbiol. Clin. 24 (6), 370–372. Mushtaq, S., Warner, M., Livermore, D.M., 2010. In vitro activity of ceftazidime+NXL104 against Pseudomonas aeruginosa and other non-fermenters. J. Antimicrob. Chemother. 65, 2376–2381. Nasim, K., Elsayed, S., Pitout, J.D.D., Conly, J., Church, D.L., Gregson, D.B., 2004. New method for laboratory detection of AmpC β-lactamases in Escherichia coli and Klebsiella pneumoniae. J. Clin. Microbiol. 42, 4799–4802. Pérez-Pérez, F.J., Hanson, N.D., 2002. Detection of plasmid-mediated AmpC βlactamase genes in clinical isolates by using multiplex PCR. J. Clin. Microbiol. 40, 2153–2162. Philippon, A., Arlet, G., Jacoby, G.A., 2002. Plasmid-determined AmpC-Type βlactamases. Antimicrob. Agents Chemother. 46 (1), 1–11. Singhal, S., Mathur, T., Khan, S., Upadhyay, D.J., Chugh, S., Gaind, R., Rattan, A., 2005. Evaluation of methods for AmpC beta-lactamase in gram negative clinical isolates from tertiary care hospitals. Indian J. Med. Microbiol. 23, 120–124. Tan, T.Y., Ng, L.S.Y., He, J., Koh, T.H., Hsu, L.Y., 2009. Evaluation of screening methods to detect plasmid-mediated AmpC in Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis. Antimicrob. Agents Chemother. 53, 146–149. Tavío, M.M., Perilli, M., Vila, J., Becerro, P., Ruiz, J., Amicosante, G., De Anta, M.T.J., 2005. Ciprofloxacin, salicylate, and 2,4-dinitrophenol decrease production of AmpC-type β-lactamase in two Citrobacter freundii clinical isolates. Microb. Drug Resist. 11, 225–231. Tavío, M.M., Perilli, M., Vila, J., Becerro, P., Casanas, L., Amicosante, G., De Anta, M.T.J., 2004. Salicylate decreases production of AmpC type b-lactamases and increases susceptibility to β-lactams in a Morganella morganii clinical isolate. FEMS Microbiol. Lett. 238, 139–144. Thomson, K.S., Mejglo, Z.A., Pearce, G.N., Regan, T.J., 1984. 3-Dimensional susceptibility testing of β-lactam antibiotics. J. Antimicrob. Chemother. 13, 45–54.
Contents lists available at SciVerse ScienceDirect
Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth
Detection of AmpC beta-lactamases using sodium salicylate Mona T. Kashif, Aymen S. Yassin ⁎, Alaa El-Dien M.S. Hosny Department of Microbiology and Immunology, Faculty of Pharmacy, Cairo University, Cairo, 11562, Egypt
a r t i c l e
i n f o
Article history: Received 4 August 2012 Received in revised form 22 September 2012 Accepted 29 September 2012 Available online 8 October 2012 Keywords: AmpC β-lactamases Sodium salicylate
a b s t r a c t AmpC β-lactamases are enzymes that hydrolyze all β-lactam antibiotics except cefipime and imipenem. Currently, there is no standard phenotypic method for detection of such enzymes. This study aims to report the use of sodium salicylate for AmpC β-lactamases detection and to compare its sensitivity and specificity to other commonly known inhibitors. A total of 135 clinical isolates were used to test the effectiveness of sodium salicylate in detection of plasmid- as well as chromosomally encoded AmpC β-lactamases. All isolates were tested by multiplex PCR testing as well as inhibitor-based methods using cloxacillin, phenylboronic acid and sodium salicylate for the detection of AmpC enzymes. Four isolates were confirmed as producers of plasmid-encoded AmpC β-lactamase and a single isolate was confirmed to have both plasmid and chromosomal genes. Cloxacillin and phenylyboronic acid failed to detect most of the plasmid-encoded enzymes. Sodium salicylate was able to detect the Escherichia coli isolates with plasmid-encoded enzymes in addition to few other isolates that were chromosomally mediated. The sensitivity and specificity of sodium salicylate was 50% and 93%, respectively, higher than those of other known inhibitors. We thus conclude that sodium salicylate can be reliably used as an inhibitor in the detection of plasmid-encoded AmpC enzymes in E. coli. © 2012 Elsevier B.V. All rights reserved.
1. Introduction AmpC β-lactamases are cephalosporinases that are active on penicillins and more active on cephalosporins. They can hydrolyze cephamycins such as cefoxitin and cefotetan; oxyiminocephalosporins such as ceftazidime, cefotaxime, and ceftriaxone and monobactams such as aztreonam. However, they have low affinity for cefipime and carbapenem (Jacoby, 2009). They are not well inhibited by clavulanic acid, but are often inhibited by a low concentration of aztreonam or cloxacillin (Bush et al., 1995). They are chromosomally encoded in the majority of Enterobacteriaceae (Livermore, 1995) where they are inducible except in Escherichia coli and Shigella spp. where they are constitutively expressed at very low level that do not cause resistance. On the other hand, Acinetobacter baumannii, Klebsiella pneumoniae, Klebsiella oxytoca, Proteus mirabilis, Salmonella spp. and Stenotrophomonas maltophilia have been found to lack chromosomal AmpC β-lactamase genes (Jacoby, 2009). Recently, plasmid-encoded AmpC β-lactamases have been detected even in species known not to have chromosomally encoded enzymes (Hanson, 2003). Plasmids carrying AmpC β-lactamase genes range in size from 7 to 180 kb (Philippon et al., 2002).
⁎ Corresponding author at: Faculty of Pharmacy, Cairo University, Department of Microbiology and Immunology, Kasr Eleini St., Cairo, 11562, Egypt. Tel.: +20 2 25353100 200 300 400, +20 100 9610341. E-mail address: [email protected] (A.S. Yassin). 0167-7012/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mimet.2012.09.035
Although many phenotypic tests are available for AmpC detection, currently, there is no method for their detection listed by the Clinical Laboratory Standard Institute (CLSI). The available phenotypic tests include enzyme extraction methods, such as three dimensional assay and cefoxitin agar medium based assay (Thomson et al., 1984; Manchanda and Singh, 2003; Nasim et al., 2004). AmpC disk test (Singhal et al., 2005) and tests incorporating AmpC β-lactamase inhibitors such as boronic acid or cloxacillin have been also described (Coudron, 2005; Jeong et al., 2009; Brenwald et al., 2005). Multiplex PCR is also used as a rapid screening tool for detection and differentiation between the six plasmid-encoded AmpC β-lactamase groups (Pérez-Pérez and Hanson, 2002). All these tests are not widely attempted by clinical laboratories either because they are inconvenient and subjective, because they lack sensitivity and specificity, or because they require reagents that are not readily available (Black et al., 2005). Recently, several articles reported that sodium salicylate induces the expression of marRAB operon in several Gram negative bacteria causing an increase in minimum inhibitory concentration (MIC) of various tested antibiotics (Martin et al., 1996; Tavío et al., 2004, 2005). However, in strains with constitutive overexpression of AmpC enzymes, sodium salicylate addition caused reduction in β-lactam antibiotic MIC. This reduction in MIC was found to be due to the repression of AmpC expression by sodium salicylate. In this study, sodium salicylate is tested for its usefulness for detection of AmpC β-lactamase, and its sensitivity and specificity are compared to other commonly known inhibitors (namely, cloxacillin and phenylboronic acid).
M.T. Kashif et al. / Journal of Microbiological Methods 91 (2012) 354–357
2. Materials and methods 2.1. Isolates Isolates used for this study were selected based on their resistance to cefoxitin or to either cefotaxime or ceftazidime (3rd generation cephalosporin) and their lack of extended spectrum beta lactamases (ESBL) expression and consequently, their positive screening for AmpC enzyme (Mirelis et al., 2006). All isolates were identified by Api 20E identification (BioMerieux, France). 2.2. Polymerase chain reaction Plasmid-carried genes encoding AmpC β-lactamases were detected by using PCR according to previously described protocols (Pérez-Pérez and Hanson, 2002). Clinical isolates of Citrobacter freundii, Enterobacter cloacae and Morganella morganii were used as positive controls. A list of primers used and expected product sizes is provided (Table 1). Primers were diluted to a concentration of 1 pmol/μl. PCR was performed in a final volume of 25 μl. Each reaction contained 2.5 μl of 10× PCR buffer, 0.5 μl (100 μM) deoxynucleoside triphosphate mixture, 2 pmol of primers CITM-F and CITM-R, 1.5 pmol of primers DHAM-F, DHAM-R, MOXM-F, MOXM-R, ACCM-F and ACCM-R, 1.25 pmol of primers EBCM-F and EBCM-R, 1 pmol of primers of FOXM-F and FOXM-R and 0.625 U of Taq DNA polymerase. All primers and PCR reagents were supplied by Fermentas (Thermo Fisher Scientific, USA). For template DNA preparation, a single colony of each organism was inoculated into 5 ml of Luria-Bertani broth and incubated for 20 h at 37 °C with shaking. Cells from 1.5 ml of the overnight culture were harvested by centrifugation at 17,310 ×g for 5 min. After the supernatant was decanted, the pellets were re-suspended in 500 μl of distilled water. The cells were lysed by heating at 95 °C for 10 min, and cellular debris was removed by centrifugation at 17,310 ×g for 5 min. One microliter of the supernatant was used as the source of template for amplification. For confirmation of plasmid origin of the multiplex PCR products, plasmids from PCR positive isolates were isolated using GenElute™ plasmid miniprep kit (Sigma-Aldrich, USA) and were then re-analyzed by multiplex PCR. PCR amplification products were purified using PCR purification kit (Qiagen, Germany) and sequenced. Sequences were aligned and analyzed by the BLAST algorithm (Altschul et al., 1990) available as part of the NCBI website (URL: http://blast. ncbi.nlm.nih.gov/Blast.cgi). 2.3. Inhibitor based assays The inhibitory activity of both phenylboronic acid and cloxacillin was tested by disc diffusion method using inoculated Mueller-Hinton agar plates. Discs for testing different inhibitors were prepared by
Table 1 The sequence of primer pairs used for detection of each enzyme family and the expected amplicon size. Enzyme family
Primer
Sequence
Expected amplicon size (bp)
MOX
MOXM-F MOXM-R CITM-F CITM-R DHAM-F DHAM-R ACCM-F ACCM-R EBCM-F EBCM-R FOXM-F FOXM-R
5′-GCT GCT CAA GGA GCA CAG GAT-3′ 5′-CAC ATT GAC ATA GGT GTG GTG C 5′-TGG CCA GAA CTG ACA GGC AAA-3′ 5′-TTT CTC CTG AAC GTG GCT GGC-3′ 5′-AAC TTT CAC AGG TGT GCT GGG T-3′ 5′-CCG TAC GCA TAC TGG CTT TGC-3′ 5′-AAC AGC CTC AGC AGC CGG TTA-3′ 5′-TTC GCC GCA ATC ATC CCT AGC-3′ 5′-TCG GTA AAG CCG ATG TTG CGG-3′ 5′-CTT CCA CTG CGG CTG CCA GTT-3′ 5′-AAC ATG GGG TAT CAG GGA GAT G-3′ 5′-CAA AGC GCG TAA CCG GAT TGG-3′
520
CIT DHA ACC EBC FOX
462 405 346 302 190
355
applying 20 μl of inhibitor solution to each cefoxitin disc (30 μg), discs were allowed to dry for 30 min and then used immediately. Cloxacillin sodium was used as a solution in sterile distilled water (10 μg/μl) and phenylboronic acid was dissolved in DMSO (20 μg/μl). Plates were incubated overnight at 35 °C. An increase of 4 and 5 mm in cefoxitin zone diameter, in presence of cloxacillin and phenylboronic acid, respectively, indicated that the inoculated isolate was an AmpC producer (Coudron, 2005; Tan et al., 2009). 2.4. Testing sodium salicylate The MIC of sodium salicylate was determined first against all the different groups of isolates and a sub inhibitory concentration of (5 mg/ml) in distilled water was used for further testing by broth microdilution method. The MIC of cefoxitin alone and in presence of sodium salicylate was determined and the reduction of MIC, in presence of sodium salicylate was calculated and correlated to the multiplex PCR results and to the presence of inducible chromosomal enzymes. 2.5. Data analysis The performances of various phenotypic test methods, for the detection of plasmid-encoded AmpC β-lactamases, were evaluated by the comparison of their results to that of multiplex PCR method (Lee et al., 2009). Sensitivity and specificity were calculated as follows: Sensitivity ¼ number of true positive isolates by phenotypic test= number of positive isolates by multiplex PCR: Specificity ¼ number of true negative isolates by phenotypic test= number of negative isolates by multiplex PCR: Calculations were carried out as described in the UK standards for Microbiological Investigations: (http://www.hpa.org.uk/webc/HPAwebFile/ HPAweb_C/1317131674973). 3. Results One hundred and thirty five clinical isolates were used to test the effectiveness of sodium salicylate in detection of plasmid — as well as chromosomally encoded AmpC β-lactamases. Among this collection of isolates, 89 were from organisms lacking inducible chromosomal AmpC β-lactamases and they were identified as follows: E. coli (39 isolates), Klebsiella spp. (27 isolates), Acinetobacter spp. (16 isolates), S. maltophilia (five isolates) and Proteus mirabilis (two isolates). Forty-six isolates were from organisms confirmed of having inducible chromosomal AmpC β-lactamase genes and these were identified as follows: Pseudomonas spp. (33 isolates), Citrobacter spp. (five isolates), E. cloacae (four isolates), Serratia marcescens (two isolates) and M. morganii (two isolates). Multiplex PCR showed that only four isolates had the plasmidencoded AmpC β-lactamase and that a single S. marcescens isolate possessed both the plasmid and chromosomal genes (Fig. 1 and Table 2). Sequencing the PCR products showed that the ampC β-lactamase genes of these isolates belonged to the CIT group genes (CMY-4 genes) in case of S. marcescens and one Klebsiella isolate, CIT (CMY-6 gene) in one E. coli isolate and DHA group genes (DHA-1 genes) in one Klebsiella and one E. coli isolates. The inhibitor-based method using cloxacillin and phenylboronic acid was evaluated by the disc diffusion test. The use of each of these inhibitors was associated with high incidence of false positive results. In addition, both reagents failed to detect plasmid- or chromosomally encoded AmpC β-lactamases with few exceptions (Table 2).
356
M.T. Kashif et al. / Journal of Microbiological Methods 91 (2012) 354–357
Fig. 1. Multiplex PCR of the five positive isolates. M: 100 bp marker, CIT, DHA and EBC are positive controls showing the expected sizes. A and C: Klebsiella spp. B: Serratia marcescens, D and E: Escherichia coli.
When sodium salicylate was tested by broth microdilution method, two cut-off values were evaluated: ≥ 1 fold and ≥ 2 folds reduction in MIC values of cefoxitin in presence of sodium salicylate. For both cut off values, in case of E. coli, the two isolates harboring the plasmid-encoded AmpC β-lactamases were detectable; however, none of the isolates with the plasmid-encoded genes were detectable, in case of Klebsiellae spp. and S. marcescens. On the other hand, 28 isolates were recorded as false positive when a cut-off values of ≥1 fold was used while six isolates were recorded as false positive when a cut off value of ≥ 2 fold was used. In addition, out of 46 isolates containing inducible chromosomal AmpC β-lactamases, sodium salicylate detected six isolates when a cut-off value of ≥ 1 reduction in MIC values of cefoxitin was used and four isolates when a cut off value of ≥ 2 fold was used. The specificity value recorded was 59% when a cut-off value of ≥1 reduction in MIC values of cefoxitin was applied and 93% when a cut-off value of ≥2 was applied. Therefore, ≥ 2 fold reduction in cefoxitin MIC was found to be the best cut-off value due to the lower number of false positive isolates detected (higher specificity) and was used for comparative data analysis. When all the inhibitors used for detection of plasmid-encoded AmpC β-lactamases in isolates lacking inducible chromosomal genes were compared, the highest sensitivity (50%) was achieved by the use of sodium salicylate, followed by both cloxacillin and phenylboronic acid (25%).
Specificity values were highest value with cloxacillin (94%) followed by sodium salicylate (93%) and then by phenylboronic acid (61%). 4. Discussion The available phenotypic tests used in the detection of AmpC β-lactamases suffer from low specificity and sensitivity; in addition, they are subjective and inconvenient to be carried out in a routine manner. In this study, sodium salicylate is evaluated for its use in the detection of AmpC β-lactamases based on its inhibitor effect. The evaluation was carried out on clinical isolates that were tested positive for the presence of AmpC β-lactamases. The recorded results showed that the use of sodium salicylate preferably detects AmpC enzymes by broth microdilution method. The best overall sensitivity and specificity (50% and 93%, respectively) were obtained when sodium salicylate was used as inhibitor, with a cut-off value of ≥2 folds reduction in cefoxitin MIC compared to inhibitor based assay using cloxacillin and phenyl boronic acid which are most widely used in AmpC β-lactamase detection (Ingram et al., 2011). In a previous study by Tan and coworkers, both cloxacillin and phenylboronic acid demonstrated 95% specificity (Tan et al., 2009). These findings agree with the specificity recorded in this study for cloxacillin, however, it is much higher than what we observed when
Table 2 Comparison of different tested phenotypic methods for detection of AmpC β-lactamases. Cloxcillin and phenylboronic acid were determined by disc diffusion method, while sodium salicylate was determined by MIC broth dilution method. Genera of screen positive bacteria
AmpC enzyme origin
Multiplex PCR
Detectability of AmpC ß-lactamases using Cloxacillin
Escherichia coli
P/C
Klebsiellae spp.
P
Stenotrophomonas maltophilia Acinetobacter spp. Proteus mirabilis Serratia marcescens
− C − P/IC
Pseudomonas spp. Citrobacter spp. Enterobacter cloacae Morganella morganii
IC IC IC IC
Phenylboronic acid
Sodium salicylate
Result
No. of isolates (total = 135)
+
−
+
−
+
−
+ − + − −
2 37 2 25 5
0 2 1 1 0
2 35 1 24 5
0 16 1 7 0
2 21 1 18 5
2 1 0 3 0
0 36 2 22 5
− − + − − − − −
16 2 1 1 33 5 4 2
1 1 0 0 0 5 4 1
15 1 1 1 33 0 0 1
10 0 0 0 1 5 4 0
6 2 1 1 32 0 0 2
2 0 0 0 1 3 0 0
14 2 1 1 32 2 4 2
P: Have plasmid mediated AmpC enzyme, C: have chromosomal AmpC genes but don't cause resistance, IC: have inducible chromosomal AmpC enzymes.
M.T. Kashif et al. / Journal of Microbiological Methods 91 (2012) 354–357
using phenylboronic acid as an inhibitor. This discrepancy may be due to the inhibitory effect of phenylboronic acid on β-lactamase enzymes other than AmpC (KPC-type) (Doi et al., 2008). Also, the recorded sensitivity for both cloxacillin and phenylboronic acid is much lower than that recorded in the same study by Tan and coworkers (Tan et al., 2009). Similar discrepancies were reported by other groups (Ingram et al., 2011), and those may be due to the presence of other resistance mechanism or to emergence of resistance to cloxacillin as a β-lactam antibiotic. The discrepancies may be also be caused by the failure of phenylboronic acid to detect enzymes from DHA origin (one E. coli and one Klebsiellae spp.) as recorded by Lee and coworkers (Lee et al., 2009). It is noteworthy that all tested inhibitor failed to detect most Pseudomonas isolates and this may be due to that Pseudomonas is less permeable than Enterobacteriaceae and is better able to efflux toxic molecules via MexAB-OprM and other systems as suggested by Mushtaq et al. (2010). The low sensitivity of sodium salicylate when used to detect the enzyme of either plasmid or chromosomal origin is most likely because sodium salicylate failed to detect the majority of inducible chromosomal AmpC β-lactamases (detected four out of 46). We suggest that this low sensitivity with chromosomally encoded enzyme is due to the low level of β-lactamase production in the inducible strains which explains why the effect of salicylate on gene expression did not show significant changes in the MIC values (Tavío et al., 2005). 5. Conclusion Sodium salicylate can be used in the detection of plasmid-encoded and chromosomally overexpressed AmpC β-lactamases in E. coli. To the best of our knowledge, this is the first report suggesting the use of sodium salicylate in detection of AmpC enzymes. It is highly specific for AmpC enzyme and is sensitive in detection of E. coli plasmidencoded AmpC enzymes. It is cheaper and more readily available. Further studies will decipher its mechanism of action as well as testing its sensitivity and specificity on all families of plasmid AmpC enzymes so that it could be used as standard method. Competing interest statement The authors declare that they have no competing interest. Acknowledgments We thank Dr. Ramy Karam Aziz from Cairo University for critically reading the manuscript and for the useful suggestions. References Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215 (3), 403–410 (5). Black, J.A., Thomson, K.S., Buynak, J.D., Pitout, J.D.D., 2005. Evaluation of β-lactamase inhibitors in disc tests for detection of plasmid-mediated AmpC β-lactamases in well-characterized clinical strains of Klebsiella spp. J. Clin. Microbiol. 43, 4168–4171.
357
Brenwald, N.B., Jevons, G., Andrews, J., Ang, L., Fraise, A.P., 2005. Disc methods for detecting AmpC β-lactamase-producing clinical isolates of Escherichia coli and Klebsiella pneumoniae. J. Antimicrob. Chemother. 56, 600–601. Bush, K., Jacoby, G.A., Medeiros, A.A., 1995. A functional classification scheme for blactamases and its correlation with molecular structure. Antimicrob. Agents Chemother. 39, 1211–1233. Coudron, P.E., 2005. Inhibitor-based methods for detection of plasmid-mediated AmpC β-lactamases in Klebsiella spp., Escherichia coli, and Proteus mirabilis. J. Clin. Microbiol. 43, 4163–4167. Doi, Y., Potosk, B.A., Adams-Haduch, J.M., Sidjaba, H.E., Pasculle, A.W., Paterson, D.L., 2008. Simple disc-based method for detection of Klebsiella pneumoniae carbapenemase-Type β-lactamase by use of a boronic acid compound. J. Clin. Microbiol. 46, 4083–4086. Hanson, N.D., 2003. AmpC β-lactamases: what do we need to know for the future? J. Antimicrob. Chemother. 52, 2–4. Ingram, P.R., Inglis, T.J., Vanzetti, T.R., Henderson, B.A., Harnett, G.B., Murray, R.J., 2011. Comparison of methods for AmpC β-lactamase detection in Enterobacteriaceae. J. Med. Microbiol. 60 (Pt 6), 715–721. Jacoby, G.A., 2009. AmpC β-Lactamases. Clin. Microbiol. Rev. 22, 161–182. Jeong, S.H., Song, W., Kim, J.S., Kim, H.S., Lee, K.M., 2009. Broth microdilution method to detect extended-spectrum beta-lactamases and AmpC beta-lactamases in Enterobacteriaceae isolates by use of clavulanic acid and boronic acid as inhibitors. J. Clin. Microbiol. 47, 3409–3412. Lee, W., Jung, B., Hong, S.G., Song, W., Jeong, S.H., Lee, K., Kwak, H., 2009. Comparison of 3 phenotypic-detection methods for identifying plasmid-mediated AmpC βlactamase-producing Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis strains. Korean J. Lab. Med. 29, 448–454. Livermore, D.M., 1995. β -lactamases in laboratory and clinical resistance. Clin. Microbiol. Rev. 8, 557–584. Manchanda, V., Singh, N.P., 2003. Occurrence and detection of AmpC β-lactamases among Gram-negative clinical isolates using a modified three-dimensional test at Guru Tegh Bahadur Hospital, Delhi, India. J. Antimicrob. Chemother. 51, 415–418. Martin, R.G., Jair, K., Wolf, R.E., Rosner, J.L., 1996. Autoactivation of the marRAB multiple antibiotic resistance operon by the MarA transcriptional activator in Escherichia coli. J. Bacteriol. 178, 2216–2223. Mirelis, B., Rivera, A., Miró, E., Mesa, R.J., Navarro, F., Coll, P., 2006. A simple phenotypic method for differentiation between acquired and chromosomal AmpC betalactamases in Escherichia coli. Enferm. Infecc. Microbiol. Clin. 24 (6), 370–372. Mushtaq, S., Warner, M., Livermore, D.M., 2010. In vitro activity of ceftazidime+NXL104 against Pseudomonas aeruginosa and other non-fermenters. J. Antimicrob. Chemother. 65, 2376–2381. Nasim, K., Elsayed, S., Pitout, J.D.D., Conly, J., Church, D.L., Gregson, D.B., 2004. New method for laboratory detection of AmpC β-lactamases in Escherichia coli and Klebsiella pneumoniae. J. Clin. Microbiol. 42, 4799–4802. Pérez-Pérez, F.J., Hanson, N.D., 2002. Detection of plasmid-mediated AmpC βlactamase genes in clinical isolates by using multiplex PCR. J. Clin. Microbiol. 40, 2153–2162. Philippon, A., Arlet, G., Jacoby, G.A., 2002. Plasmid-determined AmpC-Type βlactamases. Antimicrob. Agents Chemother. 46 (1), 1–11. Singhal, S., Mathur, T., Khan, S., Upadhyay, D.J., Chugh, S., Gaind, R., Rattan, A., 2005. Evaluation of methods for AmpC beta-lactamase in gram negative clinical isolates from tertiary care hospitals. Indian J. Med. Microbiol. 23, 120–124. Tan, T.Y., Ng, L.S.Y., He, J., Koh, T.H., Hsu, L.Y., 2009. Evaluation of screening methods to detect plasmid-mediated AmpC in Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis. Antimicrob. Agents Chemother. 53, 146–149. Tavío, M.M., Perilli, M., Vila, J., Becerro, P., Ruiz, J., Amicosante, G., De Anta, M.T.J., 2005. Ciprofloxacin, salicylate, and 2,4-dinitrophenol decrease production of AmpC-type β-lactamase in two Citrobacter freundii clinical isolates. Microb. Drug Resist. 11, 225–231. Tavío, M.M., Perilli, M., Vila, J., Becerro, P., Casanas, L., Amicosante, G., De Anta, M.T.J., 2004. Salicylate decreases production of AmpC type b-lactamases and increases susceptibility to β-lactams in a Morganella morganii clinical isolate. FEMS Microbiol. Lett. 238, 139–144. Thomson, K.S., Mejglo, Z.A., Pearce, G.N., Regan, T.J., 1984. 3-Dimensional susceptibility testing of β-lactam antibiotics. J. Antimicrob. Chemother. 13, 45–54.