Abstract
Emergence and spread of Klebsiella pneumoniae isolates producing extended-spectrum β-lactamases (ESBLs) present a major threat to public health. In this study, we characterized β-lactam-resistant K. pneumoniae isolates from six wastewater samples obtained from two pharmaceutical industries located in Lagos and Ogun States, Nigeria. Bacteria were isolated by using MacConkey agar; species identification and antibacterial susceptibility testing were performed by Vitek 2. Etest was used for ESBL phenotype confirmation. The presence of β-lactamase genes was investigated by PCR and sequencing. Bacterial strain typing was done by XbaI-macrorestriction and subsequent pulsed-field gel electrophoresis (PFGE) as well as multilocus sequence typing (MLST). Thirty-five bacterial species were isolated from the six samples; among them, we identified seven K. pneumoniae isolates with resistance to β-lactams and co-resistance to fluoroquinolones, aminoglycosides, and folate pathway inhibitors. The ESBL phenotype was confirmed in six K. pneumoniae isolates that harbored ESBL genes blaCTX-M-15 (n = 5), blaSHV-2 (n = 1), and blaSHV-12 (n = 1). PFGE and MLST analysis revealed five clones belonging to four sequence types (ST11, ST15, ST37, ST101), and clone K. pneumoniae-ST101 was present in the wastewater samples from two different pharmaceutical industries. Additionally performed conjugation assays confirmed the location of β-lactamase genes on conjugative plasmids. This is the first confirmation of K. pneumoniae isolates producing CTX-M-15-ESBL from pharmaceutical wastewaters in Nigeria. The co-resistance observed might be a reflection of the different drugs produced by these industries. Continuous surveillance of the environmental reservoirs of multidrug-resistant bacteria is necessary to prevent their further spread.
Introduction
K
There are only a few reports on CTX-M-producing K. pneumoniae in hospitals in Nigeria. The occurrence of blaCTX-M genes in Nigeria was first described in University Teaching Hospitals, Federal Medical Centres, and Specialist Hospitals from six South-Western States of Nigeria in 2006. 6 In that study, 30 multidrug-resistant K. pneumoniae were selected from a random collection of 96 isolates from patients with UTI; the data showed that 17 K. pneumoniae harbored blaCTX-M-15 genes. 6 In further studies, K. pneumoniae isolates with CTX-M-ESBLs were isolated from patients in the intensive care unit of the University of Nsukka Teaching Hospital in Enugu, different hospitals in South-Western Nigeria, and one hospital in North-Eastern Nigeria, respectively.7–10
One of the ways in which bacteria acquire resistance to antibiotics is due to selective pressure as a result of antimicrobial use in humans and animals; water plays an important role in the dissemination of these organisms among humans, animals, and the environment. 11 In a study from the Democratic republic of Congo, K. pneumoniae with blaCTX-M genes were isolated from wastewater; in Nigerian drinking water distribution channels, ESBL-producing K. pneumoniae have been found.12,13 The aim of the present study was to analyze the presence of K. pneumoniae in pharmaceutical wastewaters in Nigeria and to characterize ESBL-producing strains.
Materials and Methods
Sampling and isolation of bacteria
Wastewater samples (n = 6) were collected from discharge points from two pharmaceutical industries in two locations—about 60 km distance—in Lagos (global positioning system [GPS] coordinates: E003° 13′ 21.4″ N06° 36′ 53.5″) and in Ogun States (GPS coordinates: N06° 42′ 19.7″; E003° 13′ 44.3″) in Nigeria. The samples were filled directly into sterile plastic bottles that were preserved on ice to maintain a stable condition of the parameters before analysis and immediately taken to the laboratory for routine physicochemical and microbiological analyses (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/mdr).
Sampling was done three times in 2012 (February, April, and May) at the point where the industrial wastewaters are flowing from the pipe of the production plant into the water passages (gutters) into the environment. Both industries are producers of antibiotics, for example, ciprofloxacin and cotrimoxazole, analgesic agents, and a few other healthcare products. Wastewater samples (10 ml) were added to sterile distilled water (90 ml) and shaken vigorously to ensure uniform distribution of contents. A tenfold serial dilution of suspension was made, and aliquots (0.1 ml) of appropriate dilutions (10−3, 10−5, and 10−8) were plated on MacConkey agar and incubated aerobically for 24 hrs at 37°C. After the overnight incubation, colonies with distinct morphological appearance were further investigated.
Species identification was performed by using standard microbiological tests, such as lactose fermentation test, Oxidase test, and Gram staining. For further confirmation, we used an automated measurement (Vitek 2 GN-ID Card and Vitek 2 card AST-N111; BioMerieux, USA). In addition, a specific PCR for the identification of K. pneumoniae sensu stricto was performed by using primers 50233-F 5′-GCTCTGGGAGATAGACCGCA-3′ and 50233-R 5′-GCGATSGCAGACCAGATGAAT-3′. 14
Antimicrobial susceptibility testing
Antimicrobial susceptibility testing was performed by using the automated system Vitek 2 (BioMerieux, USA) with cards AST-N223 and AST-N248, including 22 antibiotics: ampicillin, ampicillin/sulbactam, piperacillin, piperacillin/tazobactam, aztreonam, cefotaxime, cefpodoxime, ceftazidime, cefuroxime, cefepime, ertapenem, imipenem, meropenem, ciprofloxacin, moxifloxacin, gentamicin, tobramycin, amikacin, tigecycline, colistin, fosfomycin, and trimethroprim/sulfamethoxazole. Results were interpreted according to recommendations of the European Committee on Antimicrobial Susceptibility Testing (EUCAST v5.0).
Phenotypic and genotypic characterization
The Vitek 2 cards included an automated ESBL confirmation test by determining minimum inhibitory concentrations of different cephalosporins (cefepime, cefotaxime, and ceftazidime) alone and in the presence of an ESBL inhibitor (clavulanic acid). In addition, phenotypic confirmation of ESBL production was performed by Etest strips containing cefepime (PM) alone and in combination with clavulanic acid (PML). All K. pneumoniae isolates with any β-lactam resistance and/or a confirmed ESBL phenotype were screened for the presence of different β-lactamase genes (blaTEM-like, blaSHV-like, blaCTX-M-like, blaCMY-like, blaDHA-like, blaOXA-48-like, blaKPC-like, blaVIM-like, blaIMP-like, and blaNDM-like) by PCR and subsequent sequencing.15–17 All gene sequences were compared with data of the GenBank (NCBI) database to identify the exact β-lactamase genotype.
Bacterial strain typing and conjugation assay
The genetic relationships of K. pneumoniae isolates were studied by enzymatic macrorestriction of the total bacterial DNA by XbaI enzyme and subsequent pulsed-field gel electrophoresis (PFGE) following the method of Gaustom et al., 18 and the results were interpreted according to the criteria of Tenover et al. 19 Furthermore, the sequence type (ST) of the K. pneumoniae isolates was determined by multilocus sequence typing (MLST; http://bigsdb.web.pasteur.fr/klebsiella/klebsiella.html). 20
Transfer of resistance was tested in a conjugation assay by using a sodium azide-resistant Escherichia coli K12J53 strain as recipient. 21 Transconjugants were selected on Luria-Bertani agar containing 100 mg/L ampicillin and 200 mg/L sodium azide. Sizes of transferred plasmids were determined by S1-nuclease restriction and PFGE; plasmid replicon types were determined by a multiplex PCR kit (“PBRT KIT”; DIATHEVA, Fano, Italy) as previously described.22,23
Results
Identification of bacteria and antibiotic susceptibility testing
A total of 35 Gram-negative bacterial isolates were recovered on MacConkey agar plates; and these were identified as K. pneumoniae, Proteus mirabilis, Serratia marcescens, Providencia stuartii, Pantoea spp., Pseudomonas aeruginosa, Burkholderia cepacia, Stenotrophomonas maltophilia, Alcaligenes faecalis, and Cronobacter sakazakii. We identified seven K. pneumoniae isolates by confirmatory automated testing and species-specific PCR. Only the K. pneumoniae isolates showed resistances to different antibiotics and were analyzed further in this study.
All seven K. pneumoniae isolates were resistant to ampicillin, piperacillin, ciprofloxacin, gentamicin, and sulfamethoxazole/trimethoprim but they remained susceptible to carbapenems (ertapenem, imipenem, meropenem), amikacin, and colistin. Resistance to piperacillin/tazobactam, cefpodoxime, cefotaxime, and ceftazidime was detected in six isolates (85.7%), and four isolates (57.2%) were resistant to tigecycline (Table 1). The ESBL phenotype was confirmed in six out of the seven K. pneumoniae isolates.
Italics are used to indicate β-lactamases and plasmids that were transferred successfully into Escherichia coli K12J53 in a conjugation experiment. ST indicates multilocus sequence type according to Diancourt et al. (2005).
ACE, cefuroxime/axetil; AMP, ampicillin; ATM, aztreonam; CAZ, ceftazidime; CIP, ciprofloxacin; CPD, cefpodoxime; CTX, cefotaxime; CXM, cefuroxime; ESBLs, extended-spectrum β-lactamases; FEP, cefepime; GEN, gentamicin; MXF, moxifloxacine; PIP, piperacillin; SAM, ampicillin/sulbactam; SXT, trimethoprim/sulfamethoxazole; TGC, tigecycline; TOB, tobramycin; TZP, piperacillin/tazobactam.
Genotypic characterization of K. pneumoniae isolates
PCR screening and sequence analyses revealed that different ESBL genes (blaCTX-M-15, blaSHV-2, and blaSHV-12) were present in the six isolates with the ESBL phenotype; in addition, beta-lactamase genes blaTEM-1, blaSHV-1, blaSHV-11, and blaSHV-28 were identified (Table 1). The isolate without the ESBL phenotype harbored β-lactamase genes blaTEM-1 and blaSHV-11. In the wastewater of the industrial company in Lagos, we found five ESBL-producing K. pneumoniae whereas one ESBL-producing K. pneumoniae was found in the sample from Ogun. The non-ESBL-producing K. pneumoniae was found in Lagos.
Bacterial strain typing and conjugation assay
PFGE typing revealed identical macrorestriction patterns (PFGE-type 1) for three K. pneumoniae isolates, confirming the presence of the same bacterial strain/clone in Ogun (n = 1) and Lagos (n = 2). The remaining isolates were not genetically related (Fig. 1). K. pneumoniae isolates of PFGE-type 1 could be assigned to sequence type (ST101), for the other isolates ST11, ST15, and ST37 were determined.

XbaI-macrorestriction patterns (PFGE) of Klebsiella pneumoniae isolates from pharmaceutical wastewaters, Nigeria. Lane 1, K. pneumoniae-ST101 no. D192 (CTX-M-15, SHV-1); Lane 2, K. pneumoniae-ST101 no. SV12 (CTX-M-15, SHV-1); Lane 3, K. pneumoniae-ST101 no. SV22 (CTX-M-15, SHV-1); Lane 4, K. pneumoniae-ST37 no. SB1 (SHV-11, TEM-1); Lane 5, K. pneumoniae-ST15 no. SB3 (CTX-M-15, SHV-28, TEM-1); Lane 6, K. pneumoniae-ST15 no. SB82 (CTX-M-15, SHV-1, SHV-2, TEM-1); Lane 7, K. pneumoniae-ST11 no. SB102 (SHV-11, SHV-12, TEM-1); Lane M, Molecular marker S. Braenderup H9812. PFGE, pulsed-field gel electrophoresis.
Using a conjugation assay, ampicillin resistance was transferred successfully from two K. pneumoniae strains, SB1 (ST37) and SB82 (ST15), into recipient E. coli K12J53. Antimicrobial susceptibility testing and PCR screening of the resulting transconjugants SB1-TC and SB82-TC revealed that transconjugant SB1-TC was resistant to ampicillin, piperacillin, and gentamicin but remained susceptible to cephalosporins. In contrast, transconjugant SB82-TC was resistant to ampicillin, piperacillin, cefotaxime, ceftazidime, cefpodoxime, and sulfamethoxazole/trimethoprim.
PCR screening confirmed the presence of gene blaTEM-1 in transconjugant SB1-TC and blaCTX-M-15 in transconjugant SB82-TC. S1-nuclease PFGE analysis and plasmid-mediated replicon typing of the K. pneumoniae isolates and the two transconjugants revealed that the CTX-M-15-producing transconjugant SB82-TC harbored one plasmid of ca. 150 kb size (replicon type IncFIIK) but the TEM-1-positive transconjugant SB1-TC harbored two plasmids (ca. 65 and ca. 140 kb). The remaining five K. pneumoniae isolates each harbored two to five plasmids of various sizes (ca. 30–300 kb, Fig. 2), but these could not be transferred by using the simple broth mate conjugation assay.

S1-nuclease restriction patterns (PFGE) of K. pneumoniae donor strains and respective transconjugants. Black arrows label the transferred plasmid in the transconjugants. Lane 1, K. pneumoniae-ST101 no. D192 (CTX-M-15, SHV-1); Lane 2, K. pneumoniae-ST101 no. SV12 (CTX-M-15, SHV-1); Lane 3, K. pneumoniae-ST101 no. SV22 (CTX-M-15, SHV-1); Lane 4, K. pneumoniae-ST37 no. SB1 (SHV-11, TEM-1); Lane 5, E. coli K12J53 transconjugant no. SB1-TC (TEM-1); Lane 6, K. pneumoniae-ST15 no. SB3 (CTXM-15, SHV-28, TEM-1); Lane 7, K. pneumoniae-ST15 no. SB82 (CTX-M-15, SHV-1, SHV-2, TEM-1); Lane 8, Escherichia coli K12J53 transconjugant no. SB82-TC (CTX-M-15); Lane 9, K. pneumoniae-ST11 no. SB102 (SHV-11, SHV-12, TEM-1); Lane M, Molecular marker S. Braenderup H9812 (XbaI restricted).
Discussion
The number of CTX-M-producing Enterobacteriaceae serving as the cause of infection or colonizers of the intestinal tract in humans and animals is increasing worldwide since more than 15 years.4,24 In the present study, we identified CTX-M-15 in five out of six ESBL-producing K. pneumoniae isolates from wastewater samples of pharmaceutical industries in Nigeria. Apart from CTX-M, ESBL-types SHV-12 and SHV-2 were found in two isolates. Several findings on the emergence of CTX-M-producing Enterobacteriaceae causing infections in hospitalized patients or outpatients have been repeatedly reported in different regions of Nigeria since 2006.6–10,25–29 Recently, CTX-M-producing enterobacterial isolates, especially K. pneumoniae, were found in Nigerian livestock farms and in drinking water distribution channels in South-Western Nigeria.13,30
The finding of CTX-M-15 in K. pneumoniae from wastewaters in the present study supports the hypothesis of previous studies that human infections or fecal carriage of ESBL-producing Enterobacteriaceae are a major source of ESBL-producing organisms in wastewaters. 31 However, a detailed genetic comparison of K. pneumoniae isolates from livestock, wastewater, and patients is necessary in future studies to confirm the circulation of distinct clones or distinct resistance gene-carrying plasmids between these settings.
The remnants of drug production in the industrial wastewater may select or co-select ESBL-producing K. pneumoniae showing various additional resistances to antibiotics of different classes in our study (Table 1). Especially, co-resistance to ciprofloxacin was detected in all our K. pneumoniae isolates. Ciprofloxacin is produced in large quantities in the pharmaceutical industries sampled for this study. The resulting selective pressure by even small amounts of ciprofloxacin in the industrial wastewaters might co-select various organisms, including the ESBL-positive K. pneumoniae, which is consistent with studies reporting the presence of ESBL producers in aquatic environments in general.32–35 One limitation of this study is that the presence of antibiotics or antibiotic remnants in the wastewater was not measured.
However, selection pressure by antibiotics is not absolutely necessary when a clone of K. pneumoniae has other properties, resulting in a high adaptability to its environment. The identified clonal lineages of K. pneumoniae in the wastewater samples (ST101, ST11, ST15) are well known as colonizers or the cause of infection in patients. According to the MLST database (http://bigsdb.web.pasteur.fr/klebsiella/klebsiella.html), isolates of ST11, ST15, and ST101 have been reported frequently worldwide from clinical samples, and these isolates produce different ESBLs (e.g., CTX-M-15), AmpC enzymes (e.g., Dhahran Hospital Saudi Arabia), or carbapenemases (e.g., New Delhi Metallo-beta-lactamase, Klebsiella pneumoniae carbapenemase). The clonal dissemination of CTX-M-15-producing Enterobacteriaceae has previously been reported in Nigerian hospitals, but MLST analyses were not performed.6,8,36
Thus, a selection of these multidrug-resistant clonal lineages might already happen in the hospital by antibiotic treatment or other adaption processes of the pathogens. An entrance of these clones into the wastewater and even drinking water due to insufficient wastewater treatment in Nigeria is possible. This could result in a contamination of all water circulation systems and in the survival of these clones even without a selective advantage mediated by antibiotic resistance is conceivable.
In contrast, clonal diversity of ESBL-producing E. coli and K. pneumoniae is well known since the location of ESBL genes on conjugative plasmids facilitates their spread in different bacterial species, and it facilitates the exchange of resistance genes between nonpathogenic strains and their pathogenic counterparts.4,37
In one K. pneumoniae strain of the present study, ESBL gene blaCTX-M-15 was found to be located on a conjugative IncFIIK plasmid. Those IncFIIK plasmids were initially described as virulence plasmids, but another study showed the capability of IncFIIK plasmids to integrate antibiotic resistance genes and persist over a long period.38,39 Further studies from the Czech Republic and Nigeria reported on the transfer of IncF plasmids, for example plasmids, among E. coli strains and Klebsiella spp. isolated from river water and in hospitalized patients, respectively.8,31 However, the role of these plasmids in the spread of ESBL genes in different settings can only be clarified in future by a detailed comparison of whole plasmid sequences.
In the remaining K. pneumoniae isolates of the present study, blaCTX-M-15 could not be transferred, probably due to the location of blaCTX-M-15 on nonconjugative plasmids, the location of blaCTX-M-15 in the chromosome, or limitations of the simple broth mate conjugation system that was used. Antibiotic remnants in wastewaters or antibiotic usage in in- and outpatients or livestock breeding might select all the different ESBL-producing clones that we found in the wastewater. A loss of fitness due to the acquisition of antibiotic resistance was not confirmed for the major clonal lineages of K. pneumoniae in hospitals. 40
This indicates that all selection processes and transfer pathways of these clones are not yet known in detail as well as the extent to which the discharge of these bacteria into the environment contributes to the dissemination of antibiotic resistances, in general, is uncertain. Therefore, studies that compare bacterial isolates and their resistance genes from different sources (wastewater, humans, and animals) in detail are needed to assess the transmission ways of resistant bacteria and their resistance genes.
In conclusion, this study demonstrated the presence of different ESBL-producing K. pneumoniae clones in industrial wastewaters in Nigeria. The most commonly identified ESBL gene blaCTX-M-15 was, in part, found to be located on a conjugative IncFIIK plasmid. Since IncF plasmids with blaCTX-M-15 were previously found in different K. pneumoniae from hospitalized patients in Nigeria, there is a need for detailed comparative analyses of ESBL-positive clones and ESBL gene-carrying plasmids to evaluate the extent of dissemination in our country.
Footnotes
Acknowledgments
The authors thank George A. Jacoby for providing the E. coli K12J53 recipient strain. They extend special thanks to Sibylle Müller-Bertling, Christine Günther from Robert Koch Institute and Anne Kohler, Elsa Baufeld, Dr. Verena, Dr. Lyne, Alexandra, Claudia, Tanya, Imka, Maria, and Julia from Friedrich Loeffler Institute for their excellent technical assistance. They also thank Adebayo Shittu from Obafemi Awolowo University, Nigeria, for his mentorship and contributions. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Disclosure Statement
No competing financial interests exist.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
