Introduction

Emergence of carbapenem-resistant Enterobacteriaceae (CRE), especially carbapenem-resistant Klebsiella pneumoniae, is a major public health concern worldwide (Munoz-Price et al. 2013). The most effective antibiotics against carbapenem-resistant K. pneumoniae isolates are colistin and tigecycline (Nordmann et al. 2009). Resistance to β-lactam antibiotics in Enterobacteriaceae has been increasingly observed with the production of carbapenemases, extended-spectrum β-lactamases (ESBLs), AmpC β-lactamases, overexpression of efflux pumps, and loss of porins (HU et al. 2014). CRE, particular New Delhi metallo-beta-lactamase (NDM)-producing K. pneumoniae, is one of the most potential gram-negative pathogens in hospital settings (Munoz-Price et al. 2013; Cantón et al. 2012). K. pneumoniae harbors putative efflux pump genes such as oqxA/B, which are originally located on conjugative plasmids, conferring resistance to antibacterial agents, including quinolones and chloramphenicol (Ruiz et al. 2012; Hansen et al. 2007). Previous studies have shown that mutations in DNA gyrase and DNA topoisomerase IV have caused fluoroquinolone resistance (Hansen et al. 2007). Although, plasmid-mediated quinolone resistance (PMQR), harboring qnr and oqxAB genes, has decreased fluoroquinolone susceptibility (Ruiz et al. 2012; Hansen et al. 2007). Furthermore, the qnr genes such as qnrA, qnrS, and qnrB encoding proteins protect topoisomerases from fluoroquinolone antibiotics (Ruiz et al. 2012). Almost all reported aminoglycoside resistance mechanisms are mediated by 16S rRNA methyltransferase (16S-RMTase) among carbapenemases and extended-spectrum β-lactamase (ESBL)-producing gram-negative bacteria (Bueno et al. 2013).

Colistin is an important antibiotic agent in treating infections caused by carbapenem-resistant bacteria. Recently, colistin-resistant isolates have been reported worldwide by intrinsic resistance such as mutations in the two-component regulatory systems PmrAB, PhoPQ and plasmid-mediated transferable colistin resistance gene, mcr-1 (Liu et al. 2016; Quan et al. 2017).

Molecular typing methods are usually used for control and monitor of nosocomial infections in healthcare settings (Messai et al. 2008). One of the most traditional molecular typing systems, providing an overview of microbial source tracking methods, is the enterobacterial repetitive intergenic consensus polymerase chain reaction (ERIC-PCR) (Ben-Hamouda et al. 2003). The method has been introduced as a short consuming and rapid method to predict genetic diversity among multidrug resistance (MDR) isolates in the hospital settings (Wasfi et al. 2016). The aim of this study was to determine the antimicrobial resistance profiles, mechanisms of resistance, and genetic relationship among 30 K. pneumoniae isolates which were collected from hospitalized patients in a burn center in Kerman, Iran.

Material and methods

Bacteria isolates

During this study (October 2015 to February 2016), 30 non-duplicative isolates of K. pneumoniae were recovered from different clinical specimens from hospitalized patients in burn center (Shafa-Hospital), Kerman, Iran. The category of hospitalized patients was in accordance with the current Centers for Disease Control and Prevention (CDC) recommendations (Horan et al. 2008). All of the isolates were identified as K. pneumoniae by microbiological and biochemical standard tests including gram-negative staining, catalase positive, oxidase negative, indole negative, methyl red negative, Voges-Proskauer positive, citrate utilization positive, and urease positive (Mahon et al. 2007).

Antibiotic susceptibility testing

Antibacterial susceptibility test of isolates to different antibiotic disks (ROSCO, Co, Denmark) including cefepime (30 μg), cefotaxime (30 μg), cefoxitin (30 μg), ceftazidime (30 μg), ceftizoxime (30 μg), cefpodoxime (10 μg), imipenem (10 μg), meropenem (10 μg), ertapenem (10 μg), gentamicin (10 μg), amikacin (30 μg), ciprofloxacin (5 μg), norfloxacin (10 μg), and fosfomycin (200 μg) was determined by disk diffusion method on Müller-Hinton agar media (CONDA, Co, Spain) according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) (http://www.eucast.org/clinical_breakpoints/). Minimum inhibitory concentration (MIC) of isolates to cefotaxime, cefepime, imipenem, and colistin was determined by microbroth dilution method according to the Clinical & Laboratory Standards Institute (CLSI 2017). The MIC of isolates to tigecycline was also determined by broth microdilution according to the EUCAST standard breakpoints recommendations. Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were used as standard strains in antibacterial susceptibility testing.

Detection of ESBL-, AmpC-, and carbapenemase-producing isolates

ESBL-producing isolates were determined by double disk method by four disks including ceftazidime (30 μg), cefotaxime (30 μg), cefpodoxime (10 μg), and amoxicillin/clavulanic acid (20 μg/10 μg) according to the British Society for Antimicrobial Chemotherapy (BSAC 2018) (http://www.bsac.org.uk) recommendation and EUCAST.

AmpC-β-lactamase and carbapenemase-producing isolates were determined by AmpC disk test and modified Hodge test (MHT) as described previously, respectively (Black et al. 2005; CLSI 2017). K. pneumoniae ATCC 700603 was used as a positive ESBL control. Enterobacter spp. KEJ-1 with blaNDM (GenBank accession no. KP347135) and Enterobacter spp. KEJ-3 with blaACT (GenBank accession no. KP347137) were used as positive control in MHT and AmpC disk methods, respectively. Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were used as negative control in confirmatory tests for detection of ESBL-, AmpC-, and carbapenemase-producing isolates.

The genomic DNA extraction

The genomic DNA extraction using Exgene Clinic SV (GeneALL, Co, Seoul, Korea, Kat: 106-152) was according to manufacturer’s guidelines.

Detection ofbla, oqxA/Befflux pumps, mcr-1, qnr, and 16S RNA methylase genes was by PCR.

The antibiotic resistance genes including blaTEM, blaSHV, blaCTX-M, blaKPC, blaOXA-48, blaIMP, blaVIM, blaNDM, oqxA/B efflux pumps, mcr-1, qnrA, qnrB, qnrS, rmtA, rmtB, and rmtC were detected by PCR technique. The primers of antibiotic resistance genes that were used in this study are listed in Table 1. PCR amplification was performed in a total volume of 25 μL containing 0.5 μL of each primer (10 pmol/L concentration), 12.5 μL of Taq DNA Polymerase Master Mix RED (AMPLIQON, Inc., Denmark), 1 μL of DNA, and 10.5 μL of DNase and RNase-free water (SinaClon, BioScience, Co, Iran) in a FlexCycler PCR Thermal Cycler (Analytik Jena, Germany) under the following conditions: initial denaturation at 95 °C for 3 min followed by 30 cycles of denaturation at 95 °C for 1 min, annealing at 51–61 °C for 45 s (Table 1), extension at 72 °C for 1 min, and the final extension step was continued for 5 min at 72 °C. The PCR products were electrophoresed on 1.5% agarose gel containing DNA Green Viewer™ dye (Pars tous Biotechnology, Co, Iran) in 0.5× Tris-EDTA-boric acid buffer (TBE). PCR products from blaTEM, blaSHV, blaCTX-M, blaNDM, and rmtC were sequenced in both directions (Macrogen, South Korea). The nucleotide sequences were compared with the GenBank database for homologous nucleotide sequences by BLAST program (www.ncbi.nih.gov/BLAST program). The AmpC beta-lactamase genes including blaFOX, blaMOX, blaEBC, blaACC, blaDHA, and blaCIT were detected using multiplex PCR as previously described (Pérez-Pérez and Hanson 2002) and mcr-1 was detected among isolates with reduced susceptibility to colistin by PCR.

Table 1 Oligonucleotide primers used in this study
Full size table

Molecular typing of isolates by ERIC-PCR

The ERIC-PCR was used for molecular typing of isolates using ERIC2 primer 5′-AAGTAAGTGACTGGGGTGAGC-3′ (Meacham et al. 2003). The ERIC-PCR amplification was carried out in FlexCycler PCR Thermal Cycler (Analytik Jena, Germany) using Taq DNA Polymerase Master Mix RED (AMPLIQON, Inc., Denmark). PCR amplification was performed in a total volume of 25 μL containing 1 μL of primer (10 pmol/L concentration), 12.5 μL of Taq DNA Polymerase Master Mix RED (AMPLIQON, Inc., Denmark), 1 μL of DNA, and 10.5 μL of DNase and RNase-free water (SinaClon, BioScience, Co, Iran) in a FlexCycler PCR Thermal Cycler (Analytik Jena, Germany) under the following conditions: initial denaturation at 95 °C for 5 min followed by 35 cycles of denaturation at 95 °C for 1 min, annealing at 55 °C for 1 min, extension at 72 °C for 2 min, and the final extension step was continued for 10 min at 72 °C. The PCR products were electrophoresed on 1.5% agarose gel containing DNA Green Viewer dye in 0.5× TBE buffer and the results were analyzed in http://insilico.ehu.eus/dice_upgma/. Cut off of 70% was used to discriminate among the isolates.

Results

Clinical samples

Clinical samples were recovered from specimens of burning wounds 4 (13.3%), urine specimens 17 (56.7%), blood 3 (10%), and bronchoalveolar lavage (BAL) 6 (20%).

Antibiotic susceptibility testing

In this study, the rate of resistance to antibiotics was as follows: 15 (50%) of isolates were resistant to cefotaxime and ceftazidime, 16 (53.3%) to cefpodoxime and ceftizoxime, 14 (46.7%) to cefepime, 13 (43.3%) to cefoxitine, 8 (26.7%) to norfloxacin and ciprofloxacin, 14 (46.7%) to amikacin, 10 (33.3%) to gentamicin, 8 (26.7%) to fosfomycin, 8 (26.7%) to meropenem, 7 (23.3%) to imipenem, and 9 (30%) to ertapenem. In this study, the decrease of susceptibility to colistin was observed in 7 (23.3%) of isolates, as follows: one isolate with MIC = 16 μg/mL, four isolates with MIC = 4 μg/mL, and two isolates with MIC = 2 μg/mL, although, the 23 (76.6%) of isolates were shown MIC ≤ 0.5 μg/mL. Our findings showed that all isolates were susceptible to tigecycline with MIC ≤ 1 μg/mL. Antibiotic resistance patterns, MIC to imipenem, cefotaxime, cefepime, colistin, and genetic characterizations of 19 multidrug-resistant isolates have been shown in Table 2.

Table 2 Antimicrobial susceptibility testing and distribution of resistance genes among 19 MDR clinical isolates of K. pneumoniae
Full size table

Phenotypic confirmatory tests

In this study, 19 (63.3%), 5 (16.7%), and 5 (16.7%) of isolates were considered as ESBL-, AmpC-, and carbapenemase-producing by phenotypic methods, respectively. In our findings, among the isolates were resistant to amoxicillin/clavulanate, cefotaxime, ceftazidime, and cefpodoxime the inhibition zone was not observed in 5 (16.7%) isolates in MDDST; these 5 isolates (strains 2, 3, 9, 17, 29, Table 2) showed positive results for AmpC (n = 3) and MHT (n = 5).

Molecular characterization of resistance genes

The ESBLs genes including blaTEM, blaCTX-M, and blaSHV were detected in 11 (57.8%), 16 (84.2%), and 13 (68.4%) of ESBL-producing isolates (n = 19), respectively. blaNDM was detected in the all carbapenem-resistant isolates (n = 9). One of the NDM-producing K. pneumoniae isolates harbored rmtC gene. Sequencing of the full length bla genes PCR products confirmed that the blaNDM, blaTEM, blaCTX-M, and blaSHV genes were 100% identical to blaNDM-1, blaTEM-1, blaCTX-M-15 (n = 11), blaCTX-M-2(n = 5), and blaSHV-12 reported in GenBank. The blaNDM, blaTEM, blaCTX-M, blaSHV, and rmtC sequences were submitted to the GenBank by accession numbers KY856828, MG515601, MG515591, MG515598, and KY849821, respectively. The oqxA, oqxB, qnrB, and qnrS genes were detected in 18 (60%), 13 (43.3%), 9 (30%), and 2 (6.7%) of isolates, respectively. AmpC β-lactamase genes including blaFOX, blaMOX, blaEBC, and blaACC were detected in 1 (3.3%) of isolates, respectively, and blaCIT was detected in 2 (6.7%) of isolates. The genetic characterization of ESBL-, AmpC-, and NDM-producing K. pneumoniae is summarized in Table 2. All isolates were negative for blaVIM, blaIMP, blaKPC, blaOXA-48, blaDHA, rmtA, rmtB, qnrA, and mcr-1 genes.

Molecular typing

The 30 K. pneumoniae isolates were divided into eight clusters (A–H) by ERIC-PCR methods. Five isolates from clusters A and B; two isolates from clusters C, F, G, and H; three isolates from clusters D and E; and five isolates were considered as singleton and one isolate was non-typeable. ERIC-PCR among 30 K. pneumoniae isolates revealed different distinct patterns (Figs. 1 and 2).

Fig. 1
figure 1

Agarose gel electrophoresis of ERIC-PCR products for 30 clinical isolates of Klebsiella pneumoniae

Full size image
Fig. 2
figure 2

Corresponding dendrogram generated with UPGMA clustering method for 30 clinical isolates of Klebsiella pneumoniae

Full size image

Discussion

The continued emergence and dissemination of carbapenemase-producing Enterobacteriaceae is a serious problem among clinicians and public health systems from different parts of the world (Shibl et al. 2013). The important reservoir for carbapenemase-producing isolates might be India and Pakistan, although it has been suggested that the Middle East region is considered as a secondary reservoir for the spread of these bacteria (Shibl et al. 2012; Nordmann et al. 2011). Similar to our findings, Nobari et al. reported a study in which the resistance rates to cephalosporins were as follows: 75, 76.6, and 78.8% isolates were resistant to cefepime, ceftazidime, and cefotaxime, respectively. They showed that imipenem has the highest activity against K. pneumoniae isolates and the percent of resistance to other antimicrobial agents was as follows: ciprofloxacin, 73.3% and amikacin, 22.2% (Nobari et al. 2014). In the current study, colistin and tigecycline have shown the most antibacterial activity against K. pneumoniae isolates. Similar to our findings, Samonis et al. conducted among 65 MDR K. pneumoniae isolates, susceptibility to tigecycline, colistin, carbapenems were 84.6, 75.4, and 21.5%, respectively (Samonis et al. 2012). In our study, despite reduced susceptibility to colistin, mcr-1 gene was not detected. However, resistance to colistin in gram-negative bacteria might be associated to intrinsic resistance such as mutations in the two-component regulatory systems PmrAB and PhoPQ (Quan et al. 2017). One of the most prevalent mechanisms for resistance to colistin is modifications of bacterial lipopolysaccharide, which result in the reduction of colistin affinity (Liu et al. 2016; Quan et al. 2017). The present findings indicated that among cefoxitin-resistant isolates, only 5 (16.7%) of isolates were AmpC positive by AmpC disk test method. Helmy et al. observed the AmpC disk test positive among one out of three cefoxitin-resistant K. pneumoniae (Helmy and Wasfi 2014); therefore, cefoxitin resistance in non-AmpC-producing K. pneumoniae strains in our study may be caused by the development of deficient mutants in porin-encoding genes (Lee et al. 2007). Nevertheless, Hernandez-Alles et al. demonstrated that insertion sequence interruption of porin-encoding genes might also contribute to resistance to cefoxitin (Hernandez-Alles 1999). In the current study, among ESBL-producing isolates, blaCTX-M 16(84.2%) was more prevalent than the other bla genes, which was in agreement with previous findings. In a study in Korea, blaCTX-M (80.6%) was the more prevalent bla gene among ESBL-producing K. pneumoniae (Cha et al. 2018), although Feizabadi et al. and Cheddie et al. showed blaSHV was the more prevalent bla gene among K. pneumoniae isolates (Feizabadi et al. 2010; Cheddie et al. 2017). In other countries, blaVIM and blaIMP carbapenemase-producing Enterobacteriaceae have been reported, although these carbapenemase genes were not reported in this study(Azimi et al. 2014). The most important finding in the present study was high prevalence of blaNDM-1 among carbapenem-resistant isolates.

Within a few years, the rapid global spread of 16S-RMTase producers has been associated with widespread dissemination among members of Enterobacteriaceae that produce NDM-metallo-β-lactamase (Samonis et al. 2012; Helmy and Wasfi 2014), since these resistant determinants are usually harbored on the same conjugative plasmid among Enterobacteriaceae (Lee et al. 2007; Hernández-Allés et al. 1999). In this study, among aminoglycoside resistance genes in clinical isolates, one isolate carried rmtC gene. 16S-RMTase genes, in particular, rmtC, have been so far accumulated on conjugative plasmids, integrons, and other genetic elements that are mainly related to rapid dissemination of β-lactamase genes (Wachino et al. 2006; Toleman and Walsh 2011). At this study, we showed the presence of rmtC and the β-lactamase genes including blaNDM-1, blaCTX-M-15, blaSHV-12, qnrB, oqxA, and oqxB in one isolate simultaneously. However, Jiang et al., Liu et al., and Poirel et al. showed that PMQR proteins including QnrA and QnrB were coproduced with 16S-RMTases (Jiang et al. 2010; Liu et al. 2008; Poirel et al. 2011). In this study, qnr genes were determined as follows: qnrB 9 (30%) and qnrS 2 (6.7%) of which all of them have one or two efflux pumps oqxA/B genes simultaneously. In this work, the prevalence of qnrS (6.7%) gene was significantly lower than that of the qnrB (30%) gene. Similar to our findings, Azadpour et al. reported the prevalence of qnrS (5.55%) was much lower than that of the qnrB (88.9%) (Azadpour et al. 2014). Molecular typing of clinical multidrug-resistant K. pneumoniae isolates is the powerful tool to characterize and prevent the spread of infections in healthcare systems (Wasfi et al. 2016). In the current study, ERIC-PCR revealed 25 distinct patterns, so according to the different antimicrobial susceptibility profiles, resistance gene patterns as well as a wide dissemination of different clones through participant wards in the hospital, confirmed a high diversity and genetic transmission among K. pneumoniae isolates.

Conclusion

In this study, we described one K. pneumoniae isolate co-harboring blaNDM-1 and rmtC genes, which was isolated from urinary tract infection. Epidemiological typing results suggested the rise of different clones of K. pneumoniae, harboring various types of resistance genes, contribute to a wide reservoir of resistance genes among our hospital settings. Therefore, more surveillance on operational infection control policies is essential to prevent the outbreak of these bacteria in healthcare settings.