High level detection of extended spectrum beta-lactamase gene encoding Enterobacteriaceae in public toilets in Abeokuta, Nigeria

Increasing reports of the pandemic nature of extended spectrum β-lactamase (ESBL) producing bacteria in clinical infections called for more surveillance to identify possible sources of infection. This study examined the dissemination of ESBL producing Enterobacteriaceae in hospital and public toilet facilities in Abeokuta. Swab samples were collected from toilet seats, door handles, toilet bowls and flush handles. Detection of ESBL producing isolates was done using double disc test. Presence of ESBL gene was done using multiplex PCR. Overall, ESBL-producing bacteria was detected in 44 (27.5%) out of 160 samples. ESBL-producing bacteria were detected at significantly higher rate (p<0.001) in hospital toilets than in community toilets. Isolates were identified as Klebsiella pneumoniae (n=25), Klebsiella oxytoca (n=2), Citrobacter freundii (n=6), Citrobacter youngae (n=3), Escherichia coli (n=5), Enterobacter cloacae (n=2) and Enterobacter kobei (n=1). . Analysis for the presence of ESBL gene showed that all the isolates harboured the blaCTX-Mgene while blaTEM and blaSHV were detected in six and two isolates respectively. All the ESBL-producing isolates demonstrated resistance to at least three different classes of antimicrobials. The study showed that public toilets could serve as sources for nosocomial and community transmission of ESBL-producing bacteria in the environment.


INTRODUCTION
The extended spectrum β-lactamases (ESBLs) are enzymes capable of hydrolyzing and inactivating beta lactams antimicrobials especially penicillins, monobactams as well as first to fourth generation cephalosporins (Bradford, 2001). Thus, bacteria that produce ESBLs are resistant to many clinically important old and new generation penicillins and cephalosporins (Bradford, 2001). The β-lactam antimicrobials are among the most commonly used antimicrobials in human medicine because of their relatively high efficacy, low toxicity and low cost (Bush and Bradford, 2016). Diseases caused by ESBL-producing bacteria could be very difficult to treat because of nonresponsiveness to antimicrobial therapy and limited choice of antimicrobials (Lautenbach et al., 2001;Hu et al., 2019). Most ESBL strains are not only resistant to β-lactam antimicrobials but display multidrug resistance to other classes of antimicrobials due to co-location and co-selection of resistance genes to other antimicrobial agents alongside genes that mediate synthesis of ESBLs (Ruppé, 2015). The production of ESBLs is particularly high among members of the family Enterobacteriaceae with easy intra-and interspecies sharing of ESBL genes via mobile genetic elements (Ruppé, 2015). The Enterobacteriaceae are widely distributed in nature and are the most commonly isolated bacteria strains in clinical samples. They are associated with hospital-and community-acquired extra-intestinal infections including septicaemia, meningitis and urogenital tract diseases (Vading et al., 2018).
Public toilets could serve as important source of indirect human-to-human transmission of pathogens, which may lead to widespread dissemination of disease during an outbreak (Suen et al., 2019). People spend most of their time outside their homes for many reasons relating to works, social gatherings, travelling and other outdoor activities. Therefore, they need public toilets to relieve themselves. Ready access to clean toilets is essential to maintain hygienic environments and promote public health (Greed, 2006). Public places such as schools, transport stations, hospitals, religious centres and business organizations have toilets facilities, which members of the community may use. In most developing countries, inadequate access to public toilets contribute to environmental pollution and food/water contamination leading to diseases outbreaks (Peprah et al., 2015). High pressure on few available public toilets, poor maintenance and unsanitary conditions make the toilets hot spots for contracting and quick dissemination of pathogens.
Earlier studies have reported the detection of ESBLproducing bacteria from clinical samples in hospitals in Abeokuta (Olowe and Aboderin, 2010). However, there is a dearth of information on the role of public toilets in the dissemination of ESBL-producing Enterobacteriacae among human population in Abeokuta. Hence, the present study investigated the presence and antimicrobial resistance of ESBL-producing Enterobacteriaceae in selected public toilets in Abeokuta, Nigeria.

Collection of samples
One hundred and sixty swab samples were collected from 40 public toilets in Abeokuta. In each toilet, four swab samples were taken from surfaces as follows: toilet seats, toilet bowls, toilet flush handle and door handle. Samples were collected within a period of 10 weeks from April to June 2017. Toilets located within hospitals (n=20) and community (n=20) settings were investigated. Samples were collected from toilets in three public hospitals (hospital setting) as well as from a religious worship centre, a bus station and a tertiary institution (community setting). Samples were properly labelled and preserved in icepack for transportation to the laboratory. Microbiological analysis commenced within 12 hours of sample collection.

Isolation of ESBL-producing isolates from swab samples
Each swab sample was inoculated into nine mililitres of buffer peptone water (BPW) and incubated at 37 o C overnight for pre-enrichment. A loopful of the pre-enrichment broth was streaked on MacConkey agar supplemented with ampicillin (100mg/L). This was incubated at 37 o C for 18 hours. Discrete colonies on MacConkey-ampicillin agar were separately transferred onto MacConkey agar supplemented with cefotaxime (1mg/L) for selective isolation of cefotaxime resistant strains. All cefotaxime resistant isolates were tested for phenotypic detection of ESBL producer using cefpodoxime/cefpodoximeclavulanic acid combination disk test kit (Oxoid DD0029) according to the manufacturer's instruction and in line with recommendations of Clinical and Laboratory Standard Institute (CLSI, 2016). Briefly, test isolate was thoroughly emulsified in normal saline. The suspension was adjusted to an optical density corresponding to 0.5 McFarland standard. This was uniformly spread over Mueller Hinton agar (MHA) plate. Cefpodoxime and cefpodoximeclavulanic acid disks were firmly placed on the surface of the inoculated MHA. The plate was incubated at 35 ± 2 o C for 18 hours. The difference in the diameter of zones of inhibition between cefpodoxime and cefpodoximeclavulanic acid disks was determined. Isolates with a difference of five millimeter or above were confirmed as phenotypic ESBL producers.

Bacterial Identification
Preliminary bacterial identification was based on cultural characteristics on MacConkey agar. Further identification of phenotypic ESBL-producers into genera and species was done based on biochemical characterization using commercial biochemical kit for the identification of Gram negative bacteria (Microbact GNB 24E, Oxoid R Basingstoke, UK) and by the MALDI-TOF MS technique.

Detection of ESBL genes
Genomic DNA was extracted from all phenotypic ESBLproducing isolates by thermolysis as previously describe (Ojo et al., 2016). Amount of DNA was quantified by nanospectrophotometry. The DNA concentration was adjusted to a final working concentration of 100ng/µl. ESBL-determinant genes (bla CTX-M-group1 , bla CTX-M-group9 , bla SHV and bla TEM ) were detected by multiplex (bla CTX-M-group1 , bla CTX-M-group9 and bla SHV ) or simplex (bla TEM ) polymerase chain reaction assay using previously described specific primers and amplification conditions (Olesen et al., 2004;Grobner et al., 2009;Cullik et al., 2010). A second round of PCR assay followed by nucleotide sequencing and sequence analysis were used for further determination of bla CTX-M-group1 gene variants (Ojo et al., 2016;Okpara et al., 2018). Escherichia coli isolates were assigned to phylogenetic groups (A, B1, B2, C, D, E, and F) using a PCR-based classification method (Clermont et al., 2013).

Statistical analysis
The prevalence of ESBL-producing bacteria in toilet facilities in Abeokuta was compared using Chi-squared test and P values of <0.05 were considered statistically significant. All data were analysed using Statistical package for the Social Sciences software (SPSS version 16).
Among the 65 cefotaxime-resistant isolates, 44 were confirmed as phenotypic ESBL-producers. The positive samples represented 27.5% of the total 160 swab samples examined (Table 1). The rate of detection of ESBLproducing bacteria was significantly higher (p > 0.05) in hospital toilets (41.3%) than in community toilets (13.8%). All (100.0%) the hospital toilets tested positive for ESBLproducing bacteria as at least one of the four swab samples collected from each of the 20 hospital toilets yielded an ESBL-producer. In contrast, only five (25.0%) of the 20 community toilets tested positive for ESBL-producing bacteria. The five positive community toilets were located within religious worship centre and tertiary institution. None of the samples from toilets located within the bus station tested positive for ESBL-producing bacteria.
All the ESBL-producing isolates tested positive for the  bla CTX-M-group1 gene. Nucelotide sequence analysis showed that all the isolates haboured the ESBL gene variant bla  . In addition, bla TEM and bla SHV were detected in two and six isolates respectively. The bla TEM was present only in E. coli (n=2) while bla SHV gene was found only in K. pneumoniae (n=6). The E. coli isolates belonged to phylogenetic groups A (n=1), D (n=1) and F (n=3). All the E. coli isolates originated from hospital toilets.
The 44 ESBL-producing isolates were all susceptible to ertapenem but showed 100% resistance to ampicillin, cefotaxime, ceftazidime, sulphomethoxazole/trimethoprim, tetracycline and trimethoprim. In addition, there was 95.5% resistance to ciprofloxacin, 88.6% to nalidixic acid, 86.4% to kanamycin, 84.1% to streptomycin, 81.8% to gentamicin, 75.0% to amoxicillin/clavulanic acid, 56.8% to chloramphenicol, 18.2% to amikacin and 15.9% to cefoxitin (Table 3). All the isolates demonstrated multiple drug resistance trait with resistance to at least an antimicrobial each from more than three different classes of antimicrobials.

DISCUSSION
ESBL-producing bacteria constitute an important group of antimicrobial resistant bacteria with significant challenge to infection management in the healthcare sector. In the present study, ESBL-producing bacteria were detected in high frequency in public toilets within hospital and community settings. All the investigated public toilets within hospitals yielded ESBL-producing bacteria as against 25% of the public toilets within the community. The high level detection of ESBL-producing bacteria in hospital toilets signifies the level of contamination and ease of transmission of these bacteria within hospital settings. These bacteria could be transmitted to in-patients, healthcare workers and visitors to the hospitals. Using public toilets within the hospital is thus more likely to predispose to contracting ESBL-producing bacteria than using toilets within the community. The higher detection of ESBLproducing bacteria in hospital toilets than in community toilets also suggests infectious sources within hospitals as the likely origins of toilets contamination. Extended spectrum β-lactamase producing bacteria has been reported in sewage (Mahato et al., 2019) and air (Solomon et al., 2017) samples from hospitals. Moreover, the aggregation of patients who are carriers of ESBL-producing bacteria in hospital wards could enhance easy dissemination of these bacteria within hospital settings. Earlier authors have reported high isolation rates of ESBL-producing bacteria from clinical specimens from hospitalized patients (Rubio-Perez et al., 2012). It could also be that the use of antimicrobial agents in the hospitals facilitate the selection and proliferation of ESBL-producing bacteria in patients receiving antimicrobial therapy, which in turn serve to contaminate toilets and help the dissemination of the Table 3: Antimicrobial rates of ESBL-producing Enterobacteriaceae from public toilets in Abeokuta, Nigeria.

Antimicrobial agents
Number ( bacteria. Initial observations of resistance to cefotaxime and other new generation cephalosporins in clinical isolates from hospitals was attributed to extensive use of the cephalosporins (Sirot et al., 1987).
The present study showed that the possible transmission of ESBL-producing bacteria through toilets is not limited to toilets within hospitals. Public toilets in the community could also be sources of dissemination of ESBL-producing bacteria within human population in the study area. Contamination of community toilets could be from apparently healthy carriers, hospital outpatients or clinically infected individuals who are not seeking medical attention from hospitals. A previous study identified ESBLproducing Enterobacteriaceae in apparently healthy people in the community with higher prevalence of Klebsiella spp than E. coli (Maharjan et al., 2018). Wherever might be the primary source of contamination, public toilets are likely foci for contracting ESBL-producing bacteria similar to other pathogens. Earlier studies have identified public toilets as sources of possible spread of infections within the community especially during disease outbreaks (Johnson et al., 2013;Suen et al., 2019). The use of public toilets within the community could be a risk factor for insidious transmission of ESBL-producing bacteria to unsuspecting people (Dayan, 2004).
Findings from the present study showed that ESBLproducing trait is present among different species of Enterobacteriaceae being detected in Klebsiella spp, Citrobacter spp, Escherichia coli and Entrobacter spp. Diverse species from the family Enterobacteriaceae including those identified in the present study are known to produce ESBLs (Mesa et al., 2006;Guet-Revillet et al., 2012;Dziri et al., 2016). The production of ESBLs is common among members of the Enterobacteriaceae bacteria due to intra-and inter-species sharing of ESBL traits via transferrable mobile genetic elements (Vaidya, 2011). In this study, Klebsiella spp was the most commonly encountered ESBL-producing species. This finding is similar to earlier reports by other authors who reported higher occurrence of Klebsiella spp among ESBLproducers than other species (Sturenburg and Mack, 2003;Gröbner et al., 2009). Klebsiella spp are dominant ESBL producer especially from clinical samples in hospitals (Gröbner et al., 2009). The CTX-M enzyme, which is the most widely spread ESBL type was first reported in Klebsiella spp from hospital setting (Sirot et al., 1987). All the bacterial species identified in this study are known to produce clinical infection in humans either as major or opportunistic pathogens (Ledeboer et al., 2015). These pathogens have also been implicated in cases of communityacquired and nosocomial infections in hospitalized patients (Ling et al., 2005;Weinstein et al., 2005;Mehrad et al., 2015). Previous studies have identified ESBL-producing Enterobacteriaceae in clinical samples in hospitals within Abeokuta (Olowe and Aboderin, 2010) but none has investigated the possible roles of public toilets in the transmission of these pathogens. The detection of ESBLproducing bacteria in public toilets as observed in this study is of significant public health implication. The involvement of ESBL-producing bacteria in clinical infections has enormous socioeconomic implications. The prolonged hospitalization arising from non-response to antimicrobial therapy could increase the cost of medical care, enhance the spread of infectious agents, precipitate complications and increase the level of anxiety and frustration among family members and caregivers.
In the present study, all the ESBL-producing bacterial isolates harbourbed bla CTX-M-15 ESBL gene variant. Few of the isolates also possessed bla TEM and bla SHV . The bla CTX-M gene is globally disseminated with wide bacterial and plasmid host range (Novais et al., 2007;Woerther et al., 2013). All variants of the bla CTX-M gene mediate production of ESBLs. Only fractions of bla TEM and bla SHV variants mediate production of ESBLs while the rest mediate the production of narrow spectrum β-lactamase. Thus, the ESBL statuses of bla TEM and bla SHV genes detected in this study were not determined because isolates that had bla TEM and bla SHV also had bla CTX-M-15 . The bla CTX-M-15 is a mediator of a very potent ESBL enzyme with pandemic distribution (Woerther et al., 2013). It is the most commonly identified ESBL-gene variant in clinical bacterial isolates (Šeputienė, 2010). In Nigeria, the bla CTX-M-15 gene has been reported in bacteria from clinical samples from humans (Soge et al., 2006) as well as from commensal bacteria from environmental and animal sources (Okpara et al., 2018).
The ecological niche, lifestyle and pathogenicity of E. coli strains vary with phylogenetic origins (Clermont et al. 2015). Studies have shown that E. coli causing extra-intestinal infections are more likely to be members of phylogenetic group B2 and to a lesser extent group D (Clermont et al, 2000;Daga et al., 2019). Three E. coli isolates from the present study belonged to phylogroup F and one isolate each belonged to groups D and A. The phylogroup F is very closely related to the pathogenic phylogenetic group B2 (Clermont et al., 2013). Previous studies have shown that some pathogenic strains of E. coli isolates were members of the phylogenetic group A (Daga et al. 2019). All the E. coli isolates detected in this study were from samples collected within hospital toilets. Thus, they could be pathogenic considering that they might have originated from infectious sources within the hospital.
In addition to being resistant to the β-lactam antimicrobials, the ESBL-producing isolates in this study showed high degrees of resistance to the fluoro/quinolones (ciprofloxacin and nalidixic acid), aminoglycosides (gentamicin, kanamycin, streptomycin), folate inhibitors (sulphamethoxazole, trimethoprim), tetracycline, and phenicol (chloramphenicol). All the isolates demonstrated multidrug resistance trait. Resistance to multiple antimicrobials is a common phenomenom among ESBLproducing bacteria (Ojo et al. 2016). This has been attributed to co-location of resistance genes to different classes of antimicrobials on the same mobile genetic element harbouring ESBL genes (Ojo et al. 2016). Thus, the use of antimicrobials other than the β-lactams could lead to co-selection of ESBL traits. This could probably explain the preponderance of ESBL-producing bacteria in hospital setting where antimicrobial usage is common. Despite the high level of resistance to many of the tested antimicrobials, the isolates in this study were all susceptible to ertapenem and showed considerable susceptibility to cefoxitin and amikacin. Thus, these three antimicrobial agents could be considered in cases of failed therapy to the other antimicrobials. However, to maximize the benefits of antimicrobials and preserve their efficacy, antimicrobials should be used with caution after appropriate microbiological investigations with antimicrobial susceptibility testing. The present study highlighted the possible role of public toilets in the dissemination and transmission of ESBL-producing bacteria in hospital premises and within the community. To minimize the risk of spreading ESBLproducing bacteria through public toilets, more attention should be given to the sanitary conditions of these toilets. In particular, there should be adequate supply of water as well as provision of cleaning materials including detergents and disinfectants. Individual user of public toilets should be aware of the health risks associated with unsanitary toilets and thus must adhere to strict practice of personal hygiene at all times in the overall interest of the general public. Periodic routine microbiological assessment of public toilets for monitoring purposes should be incorporated into the sanitary inspection role of the government. This could be made part of the policies guiding the operations of toilets in public places. This is especially important in hospital settings where toilets could be foci for nosocomial transmission of resistant pathogens to healthy individuals such as hospital workers and visitors or serve to complicate ongoing illnesses in hospitalized patients.