https://orcid.org/0000-0002-2621-3720
Figures
Abstract
Wastewater discharge into the environment in resource-poor countries poses a threat to public health. Studies in this area within these countries are limited, and the use of high-throughput whole-genome sequencing technologies is lacking. Therefore, understanding of environmental impacts is inadequate. The present study investigated the antibiotic resistance profiles and diversity of beta-lactamases in Escherichia coli strains isolated from environmental water sources in Accra, Ghana. Microbiological analyses were conducted on wastewater samples from three hospitals, a sewage and wastewater treatment plant, and water samples from two urban surface water bodies. Confirmed isolates (N = 57) were selected for phenotypic antibiotic resistance profiles. Multi-drug-resistant isolates (n = 25) were genome sequenced using Illumina MiSeq sequencing technology and screened for sequence types, antibiotic resistance, virulence and beta-lactamase genes, and mobile genetic elements. Isolates were frequently resistant to ampicillin (63%), meropenem (47%), azithromycin (46%), and sulfamethoxazole-trimethoprim (42%). Twenty different sequence types (STs) were identified, including clinically relevant ones such as ST167 and ST21. Five isolates were assigned to novel STs: ST14531 (n = 2), ST14536, ST14537, and ST14538. The isolates belonged to phylogroups A (52%), B1 (44%), and B2 (4%) and carried β-lactamase (TEM-1B, TEM-1C, CTX-M-15, and blaDHA-1) and carbapenemase (OXA-1, OXA-181) resistance genes. Dominant plasmid replicons included Col440I (10.2%) and IncFIB (AP001918) (6.8%). Polluted urban environments in Accra are reservoirs for antibiotic-resistant bacteria, posing a substantial public health risk. The findings underscore the need for targeted public health interventions to mitigate the spread of antibiotic-resistant bacteria and protect public health.
Citation: Tettey R, Egyir B, Tettey P, Arko-Mensah J, Addo SO, Owusu-Nyantakyi C, et al. (2024) Genomic analysis of multidrug-resistant Escherichia coli from Urban Environmental water sources in Accra, Ghana, Provides Insights into public health implications. PLoS ONE 19(5): e0301531. https://doi.org/10.1371/journal.pone.0301531
Editor: Mabel Kamweli Aworh, North Carolina State University, UNITED STATES
Received: December 27, 2023; Accepted: March 18, 2024; Published: May 24, 2024
Copyright: © 2024 Tettey et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets generated for this study can be found at: [https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA900446]. (Accession numbers JAPJST000000000; JAPJSS000000000; JAPJTO000000000; JAPJTN000000000; JAPJTP000000000; JAPJTM000000000; JAPJTL000000000; JAPJTK000000000; JAPJTJ000000000; JAPJTG000000000; JAPJTI000000000; JAPJTH000000000; JAPJTF000000000; JAPJTE000000000; JAPJTD000000000; JAPJTC000000000; JAPJTA000000000; JAPJTB000000000; JAPJSY000000000; JAPJSU000000000; JAPJSX000000000; JAPJSW000000000; JAPJSV000000000; JAQZRM000000000; JAPJSZ000000000). Escherichia coli isolates from human, poultry, pig from Ghana were curated from the BB-BRC 3.30.19a website (https://www.bv-brc.org/view/Taxonomy/561#view_tab=genomes) for the phylogenetic analysis. All other relevant other data are within the paper.
Funding: This work was partially supported by the BANGA Africa Project (Funded by the Carnegie Fund) awarded to RT, the CAPREX project (RY89) and the SeqAfrica project awarded to BE. Identification of bacteria and antimicrobial susceptibility testing was supported by CAPREX project (RY89). Whole-genome sequencing of E. coli was funded by the SeqAfrica project. The SeqAfrica project is funded by the Department of Health and Social Care’s Fleming Fund using UK aid. The sources of funding were not involved in the study design, sample collection, data analysis, or interpretation.
Competing interests: I have read the journal’s policy and the authors of this manuscript have no competing interests.
Abbreviations: ABR, antibiotic-resistant bacteria; AMR, Antimicrobial Resistance; AR, Antibiotic resistance; AST, antibiotic susceptibility testing; CGE, Center for Genomic Epidemiology; E. coli, Escherichia coli; ESBL, Extended spectrum β-lactamase; e-waste, electronic waste; ExPEC, extraintestinal pathogenic E. coli; HE, hospital effluent; HGT, horizontal gene transfer; KL, Korle Lagoon; LMICs, low-and middle income countries; MALDITOF MS, matrix-assisted laser desorption/ionization-time of flight mass spectrometry; MAR, Multiple Antibiotic Resistance; MDR, multidrug resistant; MGEs, mobile genetic elements; MLST, multilocus sequence typing; OD, Odaw River; PCR, polymerase chain reaction; qPCR, quantitative PCR; SE, sewage effluents; SNPs, single-nucleotide polymorphisms; ST, sequence types; UPEC, uropathogenic E. coli; VAGs, virulence-associated genes; WGS, whole genome sequencing; WHO, World Health Organization; WWTP, wastewater treatment plant
Introduction
Antibiotic resistance (AR) poses a threat human and animal health, casting uncertainty on the attainment of sustainable development goals– 1, 2, 3, 6, 10, 14, and 15–[1]. Whereas AR was previously believed to be associated with the misuse of antibiotics in healthcare and animal husbandry, the current pervasiveness of antibiotics (parent compounds and functional metabolites) in the natural environment [2,3] implies that other sources are involved [4,5]. There are reports that the increased levels of AR in urban environments in low- and middle-income countries (LMICs) may be due to the lack of proper waste management systems, a major concern in the fight against antimicrobial resistance (AMR) [6,7]. Additionally, untreated or partially treated waste containing antibiotic-resistant bacteria (ABR) and antibiotic-resistance genes (ARGs) is discharged into the environment together with other micropollutants, including heavy or toxic metals from the use of personal care products and other anthropogenic activities. The discharge of these chemical compounds and biological agents transforms the environment into hotspots for the acquisition, transfer, and dissemination of ARGs between pathogenic and environmental bacteria via horizontal gene transfer (HGT) [5,8]. These facts emphasize the need to investigate potential sources of ARB and emerging ARGs of clinical relevance circulating in environments where they could easily spread to humans and animals directly or indirectly [9].
Aquatic ecosystems are recognized as hotspots for the development and dissemination of AR [10,11]. These environments may pose a health risk because of their proximity or the benefits derived by nearby communities from them. Antibiotic-resistant bacteria can spread from these environments to recreational waters, drinking water, and farm produce through rainwater run-off and flooding. For example, multidrug-resistant (MDR) pathogenic E. coli causing human and animal infections have been documented in the environment in Ghana and elsewhere [6,12–14]. However, studies utilizing sequencing technologies to study environmental E. coli or other bacteria are limited in Africa [14]. Ghana, in particular, currently lacks these data because studies employing whole genome sequencing (WGS) technologies to characterise ARGs in environmental bacteria are rare. Most studies typically report phenotypic AMR data and polymerase chain reaction (PCR) analyses [12,15,16]. Consequently, these studies may be limited in their scope of ARGs investigated and the complexity of their relationships [17]. As a result, there is insufficient information about potential or emerging pathogens and ARGs from the environment that could spread to clinically important pathogens, leading to infections that are difficult to treat. The World Health Organization (WHO) has compiled a list of antibiotic-resistant "priority pathogens" with the aim of facilitating and advancing research and development of new antibiotics. This list specifically emphasizes the significant threat posed by Gram-negative bacteria exhibiting MDR. The most critical group includes clinically important MDR bacteria belonging to the Enterobacteriaceae family, such as E. coli [18,19].
Escherichia coli is often used as an indicator of fecal contamination of water [20] because it is a non-pathogenic bacterium that colonizes the gastrointestinal tract of humans and other animals. It becomes pathogenic by acquiring specific virulence factors [21]. Pathotypes causing intestinal or extraintestinal infections in humans pose a public health risk. A biphasic lifestyle, with host-associated and host-independent phases, is an adaptive feature that enables E. coli to survive and contribute to the spread of ARGs in natural environments [22]. Furthermore, there are reports of its high resistance level to older human and veterinary antibiotics, such as tetracycline, ampicillin, and streptomycin, and an increasing level of resistance to newer antibiotics, such as cephalosporins and fluoroquinolones [23]. Multidrug-resistant E. coli isolates are those that are resistant to at least one agent in three or more antibiotic classes [24]. Escherichia coli can become resistant to antibiotics through chromosomal mutations or by acquiring resistance factors such as mobile genetic elements (MGEs) from closely related or unrelated bacterial species. For instance, the tetracycline resistance gene, tet(M), primarily found on transposons in Gram-positive bacteria, was also observed on plasmids in E. coli from a natural river basin [25]. Additionally, the most common Enterobacteriaceae β-lactamases, blaSHV, blaTEM, and blaCTX-M, are also common with E. coli isolates [26].
The aim of this study was to comprehensively analyze the genomics of E. coli strains isolated from polluted environmental sources to assess the potential risks posed to public health and safety. In this study, we examined AR patterns, phylogenic diversity, VAGs, and their association with MGEs of E. coli in wastewater samples collected from hospitals and a wastewater treatment plant (WWTP) that discharges into one of the water bodies included in this study, as well as surface water from two water bodies. Additionally, previous reports [12,27] have shown that environmental water sources heavily impacted by human activities have the potential to harbor and transmit ARB. Hence, we hypothesize that E. coli isolated from environmental sources will show a greater prevalence of resistance to the antibiotics compared with those isolated from a less impacted environmental water source.
Materials and methods
Ethical approval
Ethical approval was not required for this study because only environmental samples were used; no human participants or animal models were involved.
Study site description
Samples were collected from the Odaw River (OD) and the Korle Lagoon (KL), effluents from a sewage treatment plant, and three hospitals in Accra, Ghana’s capital city. Additionally, samples were collected from a pristine site known as the Wli waterfalls, which is situated approximately 189.2 km away from Accra, to serve as a reference site. The waterfall serves as a conservation and tourism attraction site and is less impacted by intense human activity compared to the other sampling sites.
The hospitals comprise a primary and two tertiary healthcare facilities with bed capacities of 2000, 400, and 130, respectively, in Accra. The OD, a major river with other tributaries, originates in the Eastern Region of Ghana and flows through Accra. It is marked by intense human activity along its course: industries, agricultural areas (vegetable farms and animal ranches), transport terminals with shopping centers, open markets, and health facilities. Along its course are slums known for open defaecation and indiscriminate disposal of refuse, urban areas, and health centers. Agbogbloshie, popularly known for informal electronic waste recycling, is a major landmark in its catchment area. Located a short distance from the riverbanks lies a previously abandoned landfill site that has undergone reclamation efforts and has been transformed into a thriving vegetable garden by the urban underprivileged community. This community uses water sourced from the river to irrigate the cultivated vegetables.
The OD flows into the KL, ultimately draining into the Gulf of Guinea. A sewage and wastewater treatment plant (WWTP) and a drainage system, gathering water from nearby communities within the KL catchment area, discharge midway along the KL stream. These two water bodies were previously significant providers of fish and opportunities for recreational activities. However, they have undergone severe pollution and are presently unsuitable for human use. They now sustain only a limited number of fish species, namely Clarias gariepinus and Oreochromis niloticus. These water bodies could serve as hotspots for the development and spread of ARB in these settings and beyond. Therefore, it is imperative to conduct studies that evaluate the impact of waste on the dissemination of ARB and ARG within these ecosystems.
Study design and sample collection
A two-time study was undertaken in Accra, Ghana during the dry (January) and wet (July) seasons of 2021. A total of 150 samples comprising wastewater samples (n = 12) collected at the points of discharge from the three hospitals, and from a municipal sewage and WWTP (n = 8) at the point of discharge into the KL.
The selection of hospitals was guided by the understanding that hospital wastewater serves as a significant reservoir for MDR pathogens, subsequently discharged directly into drains that ultimately feed into the two water bodies. Different levels of health facilities were chosen to showcase the varying tiers of healthcare available, all of which discharge into the aforementioned water bodies.
Manual grab water and sediments samples were collected from the WWTP and hospital waste water because they discharge into the OD and the KL, which are two main water bodies that flow through Accra Central. Sampling from the WWTP and hospital effluents (HE) aimed at source tracking to determine the occurrence of the target organisms before their release into the receiving water bodies. Sampling was conducted across multiple points within the same source to obtain representative samples.
Surface water and sediment samples were collected because reports [28,29] indicate that during regular flow conditions, approximately 20% to 35% of indicator microorganisms in the aquatic environment are linked to suspended particles that have the potential to settle. During storm events, these values experience an increase ranging from 30 to 55%. Hence, failure to consider sediment quality when monitoring surface water quality in the aquatic ecosystem may result in an underestimation of the true microbial load at sampling sites. This underestimation could pose a potential human health risk when sediment resuspension occurs, releasing bacteria into the surface water [30].
Surface water (n = 18) and sediment (n = 18) samples were also collected from the OD and the KL (S1 Table; Fig 1). Water and sediment samples were collected from the Wli waterfalls (n = 18). Water samples for microbiological analysis were collected with a stainless-steel bucket into 500-mL sterile borosilicate bottles and capped [31]. Sediment samples (about 50g each) were collected alongside the water samples by pushing down a hollow cylindrical tube to a depth of 0–10 cm. Sediment samples were homogenized and transferred into sterile sample bags (Whirl-Pak®, Merck, Germany). Samples were appropriately labelled and documented in a field notebook as well as a sampling form before leaving the sampling areas.
Yellow markings indicate a few of the sampling sites.
Samples were placed were chilled at 4°C in a portable icebox and transported to the laboratory. They were stored at 4°C prior to analysis and transported to the laboratory within six hours after collection in order to maintain their temperature at 4°C.
Isolation and identification of Escherichia coli
Standard microbiological methods were used for E. coli isolation and identification. Water samples were mixed thoroughly and serially diluted by transferring 1 ml of the sample to 9 ml of 0.9% sterile saline, followed by subsequent serial dilutions (10−1 to 10−6). Chromogenic coliform agar (CCA; Oxoid Limited, UK) plates were inoculated with 0.1 ml each of 10−5 and 10−6 dilutions, followed by surface spreading and incubation at 37°C for 24 h. Additionally, 10g of sediment was transferred to 90 ml of sterile saline, mixed vigorously by vortexing for 30 s, and allowed to settle. The supernatant was serially diluted, followed by inoculation and incubation of CCA plates as described above. Characteristic blue/purple colonies indicative of E. coli colonies were Gram stained and Gram-negative colonies were further confirmed by subculturing on Eosin Methylene Blue Agar (EMBA, Oxoid Limited, UK) at 37°C for 24 h. The growth of colonies with a green metallic sheen was presumptive of E. coli. Secondary identification was performed using pure E. coli colonies using matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDITOF MS, Bruker Daltonik GmbH, Leipzig, Germany). Briefly, the surface of the colony to be identified was touched with a sterile loop to make a smear on the steel target and air dried for 15 min. The resulting smears were overlaid with 1 μl of 70% formic acid and air dried. Subsequently, they were overlaid again with 1μl of matrix solution (containing cyano-4- hydroxy-cinnamic acid in 50% acetonitrile with 2.5% trifluoroacetic acid) and air dried for 15 min. The steel target plate was placed in the high-vacuum source chamber of the Microflex LT benchtop instrument, which was operated using FlexControl software (Bruker Daltonik GmbH, Leipzig, Germany). As an internal control, a 0.5 L Bacterial Test Standard (Bruker Daltonics Inc Billerica, MA) was used. The results were saved and printed. Protein identities were assigned using a CSU-PMF internal library, where a 70% match of total mass-spectral peaks (score of >2.0) identified an isolate as E. coli [32]. Isolates confirmed as E. coli were subjected to further downstream analyses.
Antibiotic susceptibility testing
The antibiotic susceptibility testing (AST) of the identified E. coli isolates was conducted with a panel of 12 antibiotic discs (Oxoid, UK), including chloramphenicol (C30), ampicillin (AMP10), ciprofloxacin (CIP5), trimethoprim-sulfamethodazole (STX1.25/23.75), cefotaxime (CTX30), ceftazidime (CAZ30), ceftriaxone (CRO30), cefuroxime (CXM30), meropenem (MEM10), fosfomycin (FOF200), azithromycin (AZM15), and amikacin (AK30). These antibiotics were chosen based on the Clinical and Laboratory Standards Institute (CLSI) guidelines for antibiotic susceptibility testing (AST) of Enterobacterales [33] and because they have been frequently used in the treatment of infections caused by Gram-negative bacteria in human and veterinary medicine in Ghana. The disc diffusion method was employed. To increase the reliability of the results, E. coli ATCC 25922 was used as a positive control. The CLSI breakpoint charts were followed for the interpretation of minimum inhibitory concentrations (MICs) and zone diameters [33]. Colonies showing resistance to three or more antibiotic discs, considered MDR, were selected and prepared for sequencing [24].
Confirmation of extended spectrum beta Lactamase-E. coli
Escherichia coli isolates that exhibited resistance to multiple cephalosporins were suspected of being Extended Spectrum Beta Lactamase (ESBL) producers [zone diameter of ≤27 mm for cefotaxime, ≤22 mm for ceftazidime, and ≤25 mm for ceftriaxone around the discs]. In accordance with the CLSI guidelines [33], the double-disc synergy test (DDST) was employed to validate the ESBL production of suspected colonies. Combination discs containing cefotaxime (30 μg) and ceftazidime (30 μg)/clavulanic acid (10 μg) were positioned on the seeded MH plates, approximately 24 mm away from their corresponding single discs (ceftazidime and cefotaxime, each at 30 μg). The MH plates were subsequently incubated for 16–18 hours at 37°C. To substantiate the test results, the E. coli ATCC 25922 reference strain, which lacked ESBL activity, was incubated concurrently with the experiment. Extended Spectrum Beta Lactamase production was deemed to have occurred when the diameter of the combination discs increased by more than 5 mm in comparison to their individual single discs.
Genomic DNA extraction and whole genome sequencing
Of the 43 E. coli isolates obtained, 25 (58%) were MDR and were selected for WGS comprised six (24%), four (16%), 12 (48%), and three (12%) were selected from HE, KL, OD, and WWTP, respectively. DNA extraction and WGS were performed following methods described previously [34]. Briefly, DNA was extracted using the QIAamp DNA Mini Kit (Qiagen Inc., GmbH, Holden, Germany) and quantified using a Qubit TM 4.0 Fluorometer (Invitrogen, Thermo Fischer Scientific, Singapore). The library was prepared using the Nextera DNA flex library preparation kit (Illumina, San Diego, CA, USA). An Agilent 2100 bioanalyzer system (Agilent, Santa Clara, CA, USA) and quantitative PCR (qPCR) using the Kapa Sybr Fast qPCR kit (Kapa Biosystems, USA) were used to measure the quality and concentration of fragmented DNA amplified libraries, respectively. Libraries were pooled and loaded on the Illumina 2 x 300 cycle for sequencing on the Mi-Seq machine (Illumina Inc., San Diego, CA, USA).
Bioinformatic analyses
The raw sequenced reads were subjected to quality filtering and trimming using FASTQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and Trimmomatic V0.39 [35]. The minimum quality threshold was set at Q20 [36]. Trimmed reads underwent de novo assembly using the Unicycler V0.4.8 assembler with default parameters. Only contigs with a length greater than 200 bp were retained for subsequent analysis.
All contiguous sequences were subsequently submitted to GenBank and assigned accession numbers under BioProject PRJNA900446. The assembled genomes were analysed for multilocus sequence typing (MLST) sequence types (STs) on the MLST 2.0 database hosted by the Center for Genomic Epidemiology (CGE) (http://cge.cbs.dtu.dk/services/MLST/) using default parameters. Isolates without STs were proffered to the EnteroBase Escherichia/Shigella database (https://enterobase.warwick.ac.uk/species/index/ecoli) and assigned novel STs. The acquired resistance gene types or chromosomal mutation mediating resistance were analysed using ResFinder 4.1 hosted by CGE (https://cge.food.dtu.dk/services/ResFinder/) using default parameters. Insertion sequences were determined using ISFinder. FimH and FumC types were assigned using CHTyper 1.0 (https://cge.food.dtu.dk/services/CHTyper/) using default parameters. The phylogroups of the isolates were determined using the ClermonTyping database (http://clermontyping.iame-research.center/) using default parameters.
Phylogenetic analyses
Whole genome sequences of isolates were uploaded and analysed using REALPHY 1.13, hosted by the Swiss Institute of Bioinformatics and the Universität Basel Center for Molecular Life Sciences (https://realphy.unibas.ch/realphy/). Salmonella enterica subsp. enterica serovar Typhimurium str. The LT2 genome (Accession number: PRJNA57799) was used as an outgroup to root the tree, allowing the phylogenetic distance between isolates and branches to be easily configured. To provide insights into the generated tree, the phylogeny was visualized and annotated based on the metadata using iTOL V6 (https://itol.embl.de/tree/). Furthermore, a phylogenetic analysis generated on the basis of single-nucleotide polymorphisms (SNPs) of the core genes of E. coli isolates from this study and 48 isolates (S2A and S2B Tables) comprising those from human, poultry, and pigs from previous studies in Ghana, curated at the BB-BRC 3.30.19a website (https://www.bv-brc.org/view/Taxonomy/561#view_tab=genomes) was performed to understand the clonal relationship between these isolates as well as the epidemiological significance of isolates from this study.
Results
Distribution of Escherichia coli isolates across all studied sources
A total of 150 water and sediment samples were cultured, resulting in 57 E. coli isolates [61.4% (35/57) from OD, 18% (10/57) from KL, 9% (5/57) from WWTP, and 12.3% (7/57) from HE] Escherichia coli was not isolated from samples from the pristine site; instead, the isolates were observed to be Citrobacter freundii and were not included in further downstream analysis.
Antibiotic susceptibility profiles and characterization of the ESBLs
In general, E. coli isolates demonstrated varying levels of resistance to the antibiotics against which they were tested. Of the 57 E. coli isolates tested, 75.4% (43/57) showed resistance to at least one of the 12 antibiotics they were tested against. A total of 19%, 12%, 9%, 8% and 7% of the isolates observed to be MDRs were resistant to three, five, nine, nine and six antibiotics, respectively (S1 Fig). Three isolates from OD and one from hospital effluent (HE) were observed to be resistant to nine antibiotics. One isolate each from HE and OD, and two from KL were observed to be resistant to eight antibiotics (S2 Fig). Generally, resistance ranged from 8.8% (CAZ) to 63.2% (AMP). Most isolates (over 40%) were resistant to AMP 63.2%(36/57), MEM 47.4% (27/57), AZM 45.6%(26/57), and SXT 42.1%(24/57). Some of the isolates (below 40%) showed resistance to CIP 29.8%(17/57), CAZ 8.8%(5/57), CRO 14%(8/57), C 17.5%(10/57), CTX 22.8%(14/57), AK 21.1%(12/57), and CXM 15.8%(10/57). All isolates were susceptible to fosfomycin (100%) (S3 Table; Fig 2). Isolates from OD showed high resistance rates ranging from 9.3% to 41.9%, followed by isolates from HE, with rates ranging from 2.3% to 16.3%; and from 2.3% to 9.3% for KL and WWTP samples (S4 Table; S3 Fig). A significant difference in antibiotic resistance of the isolates was observed between the sites (p < 1.72E-4; 95% CI).
Percentages of resistant, susceptible and intermediate isolates are shown in red, blue, and grey color respectively.
The resistance profiles AMP-SXT, AMP-MEM, AMP-SXT-CIP, AZM-MEM, and AZM-MEM-AMP-C-SXT were observed in both seasons (Table 1). Some isolates were resistant to a maximum of nine and eight antibiotics with the following observed resistance patterns: CIP-CAZ-AZM-AMP-CRO-CTX-MEM-AK-SXT, CIP-AZM-CRO-AMP-CTX-MEM-C-AK-SXT, CIP-AZM-CXM-CRO-AMP-MEM-C-AK, and CIP-AZM-CRO-AMP-CXM-CTX-MEM-C (S4 Fig). Of the 43 E. coli isolates (43/57), 36(84%) were resistant to beta-lactam antibiotics, whereas 14 (33%) were ESBL producers. The resistant isolates were mostly found in water sources 49%(28/57) versus sediment sources 26%(15/57). Table 1 shows the season, source, resistance profiles, and ESBL-producing status of the E. coli isolates.
Determination of the multiple antibiotic resistance index
The Multiple Antibiotic Resistance (MAR) index is a means of source tracking antibiotic-resistant isolates. The MAR index was determined by dividing the number of antibiotics to which an isolate is resistant to (a) by the total number of the antibiotics used in the study (b): a/b (S4 Table). The MAR indexes of the 43 E. coli isolates ranged from 0.08 to 0.8 with the predominant MAR index being 0.2 in 14 (33%) isolates. Out of the total isolates, 40 (93%) had an MAR index ≥0.2.
Genome characteristics of sequenced Escherichia coli isolates
A total of 25 E. coli isolates (25 of 43) showing resistance to three or more antibiotics [24], representing 44% (25/57) of isolates collected from all five sampling sites, were sequenced. The genome characteristics of the sequenced E. coli isolates, including the total assembled genome, the GC content, N50, and total number of contigs, are also shown in S5 Table.
Sequence types and phylogroups
The 25 isolates were classified into twenty sequence types (STs; Table 2, S5 Fig). All sample sources exhibited a diversity of STs: ST1662 (n = 2, 8%), ST21 (n = 2, 8%), ST167 (n = 2, 8%), and ST866 (n = 2, 8%). Five isolates were assigned to novel STs: ST14531 (n = 2) belonging to the ST446 clonal complex (CC), ST14536, ST14537, and ST14538 (ST226 clonal complex). Fifteen sequence types were singletons (Table 2).
The most common phylogroup identified among the collection was A (13; 52%), followed by B1 (11; 44%), and B2 (1; 4%) (S6 Fig). Isolates from OD belonged to phylogroups A (4; 16%) and B1 (8; 32%). Meanwhile, phylogroup A was the most common isolate from KL (3; 12%), WWTP (4; 16%), and HE (2; 8%) (S7 Fig). Interestingly, only one isolate from HE belonged to phylogroup B2, which is usually associated with pathogenicity (Table 2). Of the 14 EBSL isolates observed, eight (57%) belonged to phylogroup A, five (36%) belonged to B1, and one (7%) belonged to B2, and were mostly MDR.
Antibiotic-resistant genes of Escherichia coli isolates
Thirty-seven different types of ARGs, totaling 128 ARGs in all, were found among the E. coli isolates (21/25) observed across all studied sources (Table 2; S8 Fig). The dominant ARGs identified comprised the sulfonamide resistance gene sul2 (12/128; 9.4%), aminoglycoside resistance gene aph(6)-Id and tetracycline resistance gene tetB (9/128; 7% each), chloramphenicol resistance gene catA1 (8/128; 6.3%), sulfonamide resistance gene sul1 and folate pathway antagonist dfrA14 (7/128; 5.4% each), quinolone resistance gene qnrS1, and macrolide resistance gene mph(A) (6/128; 4.7% each). Additional genes included aac(3)-IId, ant(3’’)-Ia, aph(3’)-Ia, aadA1, aadA2 (aminoglycosides), qepA4, qnrB1, qnrS1, qnrB19 (quinolones), dfrA7, dfrA12 (folate pathway antagonist), floR (amphenicols), qacE (biocide resistance), sitABCD (hydrogen peroxide), and erm(B) (macrolides, lincosamides, and streptogramin b).
No ARGs included in the ResFinder databases were identified in four isolates (16%), two from OD and two from the KL. The fosA7 gene, which mediates resistance to fosfomycin, was identified in one E. coli isolate, but all E. coli isolates were susceptible to fosfomycin. Further genomic investigations are required to understand the resistance mechanisms in these environmental isolates.
The β-lactamase genes identified comprised the class A (Ambler classification) EBSL genes, blaTEM-1B 7.8% (10/128), blaCTX-M-15 2.3% (3/128), blaTEM-1C 1.7% (2/128), and blaTEM-35 genes. Additionally, singletons such as the class D oxacillinase genes blaOXA-1, -181 and the class B ampC β-lactamase gene: blaDHA-1 were identified. One isolate (C7) phenotypically showed ESBL activity but, not harbour ESBL genes. Further genomic analysis is needed to understand the resistance mechanism employed by this strain. ESBL-producing activity was observed among 12 STs (48%; 12/25), which demonstrated a wide range of resistance to a β-lactam antibiotic, cephalosporins, and other antibiotic classes such as quinolones, and carbapenems (Table 2). The dominant β-lactamase genes were observed in isolates belonging to STs ST866 21%(4/19), ST700 11% (2/19), and ST 167 11%(2/19).
Mutations in the quinolone resistance-determining regions (QRDRs) in chromosomal DNA gyrase (gyrA and gyrB) and DNA topoisomerase IV (parC and parE) genes were investigated for all isolates. The gyrA gene (S83L, D87N), parC gene (S80I), and parE gene (S458A) all exhibited mutations in isolate D2_S40_L001, conferring chromosomal mutations associated withfluoroquinolones resistance. The isolate (E2_S39_L001) had a mutation in the gyrA gene (S83L). Isolate D7 had three mutations in pmrA (E147K), folP (V263L), and parC (E62K) genes; however, the mutations are not known. Isolate E12 had a mutation in parC (E62K).
Genetic environment of ARGs of beta-lactamase-producing isolates
The majority of β-lactamase-producing strains were observed to possess ARGs associated with MGEs, including class 1 integrons, plasmids, insertion sequences, and transposons. These results highlight the genomic adaptability of the isolated strains (Table 3). The blaTEM gene exhibited a high frequency of association with a protein belonging to the recombinase family. The Tn3 family transposon was identified as the most prevalent transposon and was primarily linked to quinolone, sulfonamide, tetracycline, and trimethoprim. Furthermore, the presence of the IS91 insertion sequence was commonly detected and found to be correlated with resistance to sulfonamide, trimethoprim, and quinolone antibiotics. The presence of the qacEΔ1 gene, which confers resistance to disinfectants, was linked with the sulfonamide resistance gene sul1 and the aminoglycoside resistance genes aadA2 (specifically for streptomycin and spectinomycin) in four isolates (F7, E2, F9, and D2) and only the sul1 gene in one isolate (F8) from HE, WWTP, and OD. Additionally, qacEΔ1 gene was associated with a PMQR gene (qepA4) encoding a 14-transmembrane-segment efflux pump belonging to the major facilitator superfamily (MFS), associated with resistance to ciprofloxacin. The ARGs and MGEs found in the E. coli isolates exhibited a high degree of similarity (ranging from 98% to 100%) with the target sequences stored in the GenBank database. Most hits corresponded to various E. coli plasmids, along with plasmids linked to other members of the Enterobacteriaceae family. One isolate (D12) contained the plasmid pRiA4b ORF-3 family protein, which exhibited similarity to the protein encoded by ORF-3 (Q44206) present on the plasmid pRiA4 in the bacterium Agrobacterium rhizogenes. This plasmid has been reported to play a significant role in the development of tumours at wound sites in plants infected by the bacterium [37].
Virulence-associated genes
Virulome analysis showed a total of 178 VAGs comprising 41 different genes across the 25 E. coli isolates. All isolates carried between three to 23 VAGs. Seventy (39%) VAGs were observed in phylogroup A, followed by B1 85 (48%) and B2 23 (13%). Among the individual isolates, the isolate in phylogroup B2 exhibited the highest VAGs compared to the individual isolates in phylogroups A and B1. The most dominant VAGs observed were terC 14% (25/178), followed by gad and hylE 13% (24/178 each,) and IpfA 7% (12/178) [Table 2; S9 Fig]. Isolates from water recorded the highest number of VAGs 63%(113/178) compared with those from sediment 37%(65/178). In addition, isolates from the OD recorded the highest VAGs 49%(87/178), followed by isolates from hospital effluent 33%(58/178), sewage 10%(18/178), and the KL 8%(15/178). Most VAGs were characteristic of the extraintestinal pathogenic E. coli (ExPEC) pathotype, including different genes encoding adhesins (fimH, afaA, afaB, afaC, afaD, afaE), iron acquisition systems (fyuA, sitA, irp2), toxins (hlyE, sat, usp, vat) and protectins (traT, ompT, iss).
Mobile genetic elements (plasmids, insertion sequences, and integrons)
A total of 59 plasmid replicons comprising 27 different types were recorded in isolates from all studied sources. The predominant plasmid replicons recorded were Col440I (10.2%; 6/59), followed by IncFIB(AP001918) and IncY (6.8%; 4/59) (Table 4; S10 Fig). Nineteen incompatibility (Inc) plasmid replicons were identified with different frequencies, including IncFIA, IncFIA(HI1), IncFIB(AP001918), IncR, and IncY. The majority of isolates (18/59; 31%) harboured the IncF (FIA, FIB, FIC and FII), IncH (9/59; 15%), IncI (5/598.5%) and IncY (4/59; 7%). Two isolates harbouring the IncF groups also harboured the Incl and IncH groups. Plasmid replicons (19/59, 32%; 11/59, 19%; 10/59, 17%) comprising different frequencies of the Inc (F, H, Y and R), Col and p0111 groups were identified in isolates from the OD, HE, SE, and KL, respectively. However, no plasmids were observed in four isolates; three were from OD and one from KL. The majority of isolates (60%; 15/25) had more than one plasmid replicon. A total of 56 different types of ISs and transposons, totaling 228 in all, were identified in this study. A total of 47 IS families of insertion sequences were identified, followed by the cn, MITEEc1, and ICEE families with three and one, respectively. MITEEc1 (10.5%), IS3 (7.5%), IS609 (5.3%), ISEc30 (4.8%), ISEc1 (4.4%), IS102, ISEc38, ISKpn8 (3.9%), IS26, IS629 (3.1%), IS5075, IS30, Iskpn19, Tn2 (2.6%), and ISSen4 (2.2%) were the most dominant insertion sequences and transposons identified in this study. All isolates harbored more than one mobile genetic element. There was no distinct pattern in terms of the replicon type, ST, or source of isolation.
Phylogenetic analysis of Escherichia coli isolates
Whole genome phylogenetic analysis, in conjunction with the metadata, revealed that isolates clustered into two major clusters, each with subclusters. It was observed that isolates of the same phylogroup clustered together with strains clustering into A, B1, and B2 groups. Clustering was also observed among isolates from the same sample type (water and sediments). Similarly, isolates belonging to the same Fum and Fim types clustered together (Fig 3). It was observed that the E. coli strains from this study clustered with E. coli strains from humans, poultry, and pigs. Escherichia coli clustered together based on similarities within the phylogroups. Escherichia coli strains from this study also clustered with other non-environmental strains based on their STs, although a great deal of diversity was observed across all the sample sources (Fig 4).
The squares, stars, and circles at the tip the location, sample type, sequence types, fumC and fimH allelesE, and plasmids, respectively.
Discussion
This present study investigated the genomic characteristics of E. coli isolated from both wastewater and surface water sources in Accra, Ghana. Escherichia coli identified in different water sources exhibited diverse genomic characteristics, including ARGs, beta-lactamases, carbapenemases, and MGEs, displaying both diversity and complexity. These genomic features can be horizontally transferred between different bacterial species, thereby facilitating the dissemination and emergence of resistance genes as they converge within the environment. Consequently, the contaminated urban environment serves as an abundant and ever-changing reservoir of ARBs and ARGs. To the best of our knowledge, this study represents one of the most comprehensive descriptive studies conducted in Ghana, in which the genomics of E. coli originating from these distinct ecological contexts have been extensively examined using WGS.
The study observed the prevalence of resistance to ciprofloxacin, azithromycin, ampicillin, cefuroxime, cefotaxime, meropenem, chloramphenicol, and sulfamethoxazole-trimethoprim in all sources examined. High resistance were particularly noted for ampicillin, meropenem, azithromycin, and sulfamethoxazole-trimethoprim. This finding is consistent with the results of a study conducted in Ghana by Addae-Nuku et al. (2022) [15], which observed a high prevalence of resistance among isolates from hospital wastewater to sulfamethoxazole-trimethoprim, ampicillin, ceftriaxone, and ceftazidime. The results of this study also align with previous research in South Africa [14], documenting predominant resistance to ampicillin (83.3%) and sulfamethoxazole-trimethoprim (75%). Gaşpar et al. (2021) [38] reported relatively elevated resistance rates in Romania, specifically to fosfomycin (63%), ampicillin (72%), sulfamethoxazole-trimethoprim (54%), ceftazidime (58%), cefotaxime (28%), ciprofloxacin (36%), and meropenem (4%). An interesting observation in their study is the varying resistance levels to fosfomycin and meropenem compared with 100% susceptibility to fosfomycin and the high resistance to meropenem observed in this study.
The high prevalence of resistance observed in this study may be attributed to several factors, with the primary one being the inadequacy of waste management infrastructure to treat waste before discharge into various locations such as the OD and the KL [15,39]. This results in the contamination of these sources with ARBs, which may outcompete native bacteria over time and disseminate ARGs to other bacterial species through HGT. Other possible factors contributing to the high prevalence observed could be the intense usage of these antibiotics in healthcare settings and the subsequent development of resistance, as previously reported [40–42]. Furthermore, self-medication practices [43] and use of antibiotics in animal husbandry [44] could contribute to the development and spread of ARB in the community. The presence of these factors, coupled with the current challenge of improper waste handling facing the city of Accra, could result in the discharge of resistant bacteria and antibiotic residues into the environment. Moreover, the documented presence of residues from β-lactams, quinolones, cephalosporins, and sulfonamides in environmental sources in Ghana in previous studies [3,45,46] may be associated with the prevalence of MDR E. coli isolates observed in the present study.
A notably high level of antibiotic resistance was observed across all the antibiotics investigated in this study, with a particularly high prevalence in isolates originating from the OD, followed by the HE compared to the KL and SW. This observed pattern of resistance could potentially be attributed to exposure of these bacterial isolates to heavy metals particularly from the OD, which used to be the hub of informal e-waste recycling. Elevated concentrations of heavy metals, including lead (Pb), cadmium (Cd), copper (Cu), nickel (Ni), iron (Fe), and chromium (Cr), exceeding internationally accepted limits [47] have been measured in the water bodies (OD and KL) and soil adjacent to their banks [48–50]. There have been documented instances of bacteria developing resistance to antibiotics as a result of exposure to heavy metals [51,52]. Due to their non-biodegradable nature and tendency to persist in the environment, heavy metals create a persistent reservoir of chemicals, contributing to the co-selection of bacteria resistant to both heavy metals and antibiotics in these environments [53]. This poses a significant public health concern, as individuals residing in close proximity to such environments may be at risk of contracting or becoming carriers of MDR bacteria, presenting challenges in terms of effective treatment. Additionally, drinking water sources and recreational waters [54], farm produce irrigated with contaminated water, and foliage consumed by ruminants could become contaminated and act as reservoirs of ARBs and ARGs. Previous studies have reported that drinking water sources have shown limited effectiveness in the removal of ARGs [55]. The potential consequences of this situation may pose a risk to the integrity of food safety measures and impede progress towards the attainment of the SDGs.
The level of resistance to meropenem, an important carbapenem used for the management of bacterial infections within healthcare facilities in Ghana, is a matter of considerable concern [56]. This is concerning because resistance could potentially be transferred to the clinic, making treatment of infections challenging. This finding aligns with published data from Ghana as reported by Odonkor et al., (2022) [54], albeit exhibiting a slightly higher prevalence. Additionally, all E. coli isolates were completely susceptible to fosfomycin. To date, there is a dearth of reports documenting the presence of fosfomycin resistance in both environmental and clinical sources within the geographical context of Ghana [57,58]. The potential explanation for this phenomenon could be attributed to the high cost of the antibiotic in question and its comparatively reduced usage compared with alternative antibiotics [58]. The results of this study inconvertibly show that there is prevalence of resistance to commonly used antibiotics in patients, thus indicating the potential for this resistance to extend to newer antibiotics such as fosfomycin, which thus far did not show any resistance among all E. coli isolates. Phenotypically observed isolates categorized as intermediates (neither resistant nor susceptible) can develop resistance over time, either by acquiring ARGs or through chromosomal mutations.
In this study, it was observed that more than 90% of the E. coli isolates exhibited a minimum MAR index of 0.2 or higher. This finding implies that these isolates likely originated from environments in which they were potentially subjected to the influence of various antibiotics. Therefore, these environmental sources pose a significant risk of the transmission of MDR bacteria among communities residing in close proximity. We recorded MDR isolates that exhibited resistance to a maximum of nine antibiotics in the OD and HE samples, and eight antibiotics in the KL sample. The elevated MAR index may also be attributed to horizontal gene transfer events involving plasmids among various bacterial species within the surrounding ecosystem. The above-mentioned observation aligns with the findings reported by Krumperman (1983) [59] and Grabow and Prozesky (1973) [60], indicating plasmid exchanges between E. coli and other coliforms in stagnant sections of wastewater systems. An observed MAR index of 0.08 for certain isolates located upstream of the OD and midstream indicates that these isolates may be native to the environment and have not been subjected to extensive antibiotic exposure or acquired plasmids that contribute to MDR. This observation indicates that the pollution of these water bodies may be attributed to external sources, which is indeed the prevailing situation. It is imperative for Ghana to effectively implement policies of recycling of solid waste and the treatment of wastewater before its discharge into the environment. This measure will be crucial for mitigating the escalation of environmental antimicrobial resistance and the subsequent health hazards it poses.
Majority of E. coli strains isolated from all the sources were primarily associated with phylogroups A and B1, with phylogroup A showing the highest prevalence. A single isolate belonging to phylogroup B2 was isolated from HE. The higher occurrence of phylogroups A and B1, as opposed to B2, aligns with previous research conducted in different environmental settings [61,62]. However, this particular pattern has only been documented in clinical isolates within the Ghanaian context [63]. The elevated occurrence of groups A and B1 can be ascribed to the historically substantial quantities of untreated waste derived from humans and animals that have been discharged into these ecosystems. The co-occurrence of these two groups has frequently been documented in previous studies [61], and their ability to endure environmental stresses in aquatic ecosystems has been reported [64]. E. coli strains that fall within the A and B1 phylogroups are commensal bacteria found in the gastrointestinal microbiota of vertebrates and aquatic habitats. In contrast, the B2 phylogroup, which consists of pathogenic strains, has been observed in herbivorous and omnivorous mammals [61,65–67].
The E. coli isolates investigated in this study, classified as part of the B2 group, showed a collective presence of 23 virulence genes. Phylogroup B2 is known to be pathogenic and is responsible for causing extraintestinal E. coli infections, specifically in humans [61,65,68,69]. Furthermore, it has been demonstrated that B2 strains exhibit a greater ability to survive for longer periods in infants than other strains [70]. In contrast, groups A and B1 contained virulence genes ranging from four to seventeen (4–17). This is consistent with findings documented in previous studies [68,69]. Previous studies have indicated that phylogroups A and B1 exhibit reduced virulence, yet they are still capable of causing intestinal infections in mammals. According to Ochman, Lawrence and Grolsman (2000) [71], it is possible for phylogroups A and B1 to acquire virulence genes, leading to the occurrence of infections. There are reports of the association of phylogroup A with urinary tract infections [72], and A, B1, and B2 [73] with diarrheagenic infections. The presence of these phylogroups possessing the capacity to acquire virulence genes and their circulation within the environment is a matter of significant public health concern because, they can induce severe complications in vulnerable populations, including immunocompromised individuals such as the elderly, children, and convalescents.
Various studies have reported the link between phylogroups and antibiotic resistance [62,65,74]. In this study, phylogroups A and B1 exhibited MAR indices ranging from 0.3 to 0.8, constituting 52% and 44% of the recorded data, respectively. While only one isolate was identified as belonging to phylogroup B2, its MAR index was significantly elevated at 0.6 (1%). This aligns with the findings of previous studies conducted by Mosquito and colleagues [74], which a documented a higher percentage of AR in the B2 group compared to the A and B1 groups. Moreover, several studies [62,65,75] have indicated that the A and B1 groups could be MDR, as observed in this study. This phenomenon may be attributed to HGT [75]. Considering the predisposition of groups, A and B1 to induce gastrointestinal and extra-intestinal infections, the noteworthy prevalence of AR exhibited by these isolates may lead to challenging therapeutic scenarios.
The ESBL-Ec isolates identified from all studied sources, except KL, were found to be present in eight clonal lineages: ST866 (n = 2), ST700, ST390, ST342, ST218, ST202, ST167, and ST21 as well as three novel lineages, namely ST14536, ST14537, and ST14538. These STs are uncommon in other studies from Ghana and Africa. In a different context, the ST21 clone, a subgroup of the Shiga toxin-producing E. coli clone O26:H11/H-, was identified as the main cause of haemolytic uraemic syndrome (HUS) and bloody diarrhoea [76]. Additionally, an MDR-high-risk clone, ST167, was identified. The presence of ST167 has been reported in poultry and humans in Ghana [44]. Although the two isolates assigned to this ST did not demonstrate any carbapenemase activity, they were identified as ESBLs, suggesting the potential for the development of to carbapenems resistance over time. The potential dissemination of ST21 and ST167 within the environment poses a significant risk to public health, necessitating vigilant monitoring measures to ensure public health protection.
Our study also observed that the ESBL-Ec strains were associated with various β-lactamase genes, including CTX-M-15, TEM-1, -35, -1B, -1C (ESBLs), DHA-1 (ampC BLs), and OXA-1, -181 (oxacillinases). This observation is consistent with the findings reported in a study conducted by Delgado-Blas et al. (2022) [18]. In their research, the authors detected the presence of comparable genes, such as CTX-M-15, TEM-1, -1B, -1C, DHA-1, and OXA-1, in Citrobacter spp. as well as TEM-1C in E. coli, isolated from hospital effluents in the northern region of Ghana. The aforementioned results consistent with similar findings reported in Africa and other regions [14,27,77]. Despite the limited amount of information on the presence of these genes in environmental sources in Ghana, their existence has been recorded in alternative sources such as poultry, animal products, and the surrounding ecosystem [42,44,58,78,79]. The presence of the CTX-M-15 gene, among the most prevalent CTX-M-type genes, raises a significant public health concern due to its association with the dissemination of MDR bacterial strains and its substantial contribution to nosocomial infections globally [58,80]. Recently, the presence of the OXA-181 gene, which is a variant of the OXA-48 carbapenemase-producing enterobacteriaceae (CPE), was also documented in a clinical setting in Ghana [79,81]. The ubiquity of these β-lactamase genes in the environment implies that their spread may be extensive and can contribute to the rise of AR in healthcare settings. For that reason, its presence in the environment should be closely monitored to control its spread in the community.
The aforementioned E. coli β-lactamase genes and other ARGs are often disseminated by plasmids. In this study, the Inc plasmid replicon groups were the dominant types with 19 different types of plasmids (including Incl2, IncH1, IncL/M, IncY, IncR, IncQ, IncN and IncF groups). IncF replicons (FIA, FIB, FIC and FII) were the most dominant types (31%), and were found to be resistant to multiple classes of antibiotics, including cephalosporins, sulfamethoxazalole-trimethoprim, carbapenem, and quinolone. These data are consistent with previous reports from environmental sources in Africa and other parts of the world [27,82]. The IncF (IA-II-IB) group of plasmids is frequently found in bacteria isolated from humans, animals, and the environment [83] and is often the carrier of the CTX-M genes in E. coli [27], as evidenced in this study. These groups of plasmids carry resistance genes for broad-spectrum cephalosporins and other classes of antibiotics including aminoglycosides, trimethoprim, sulphonamides, tetracyclines, and chloramphenicol [84]. Given their presence in Enterobacteriaceae, a bacterial family known for its broad host range, these plasmids can exert a substantial influence on the persistent dissemination of ARGs among humans, animals, and the environment.
This study identified a wide range of VAGs commonly linked to pathogenic E. coli strains including those that cause diarrheagenic and extraintestinal diseases. The findings of our study are consistent with other studies that have identified similar virulence genes in environmental E. coli isolates [14,62,85,86]. The phylogroup B1 was observed to carry the highest number of virulent genes, followed by A and B2, respectively. While phylogroups A and B1 are considered commensal strains, B2 is known as a pathogenic strain. Notably phylogroups A and B1 have been associated with urinary tract infections [87], suggesting that commensal strains can become pathogenic by acquiring virulence factors from pathogenic strains in the environment. Virulence genes were associated with specific activities such as colonization (fimH: type 1 fimbriae, csgA: curli fimbriae) in isolates D7_S154 and C7_S158, fitness (fyuA: ferric yersiniabactin uptake) in isolates E2_S39 and D9_S44, toxins (hlyE: alpha haemolysin toxin, astA: enteroaggregative heat-stable toxin, secretory diarrhoea) in isolates F3_S150, D2_S40 and F7_S155. Escherichia coli strains in phylogroup B2 are known to be pathogenic, as observed in this study where the isolate originated from a hospital effluent, indicating a human origin. This isolate was found to possess virulence factors associated with serum survival protein (iss), adherence and biofilm formation (afaA-afaD), and the vacuolating autotransporter protein (vat) gene, which codes for a cytotoxin. The presence of both VAGs and ARGs observed across several of the isolates except four isolates (C5, C7, B9 and D1) that showed no ARGs, agrees with other studies where E. coli was isolated from the environment [88]. This finding contrasts with a study suggesting a trade-off between resistance and virulence of E. coli [89].
Phylogenomic analyses of isolates sharing the same STs, and phylogenetic groups showed some similarities, suggesting a potential relationship and possible dissemination by both clonal and horizontal spread. This poses a significant risk to public health, as MDR strains may easily propagate across different hosts within a One Health continuum.
Temperature is the primary determinant affecting the survival and growth of E. coli in the environment [88]. Temperatures recorded for wet and dry seasons ranged between 26.9–31.2°C and there was no observed impact of seasonality on E. coli strains isolated. Escherichia coli grows at an optimal temperature of 37–44°C. This implies that in tropical regions characterized by consistently high temperatures, E. coli has the potential to persist in contaminated environments throughout the year. Overall, understanding the complex interplay between seasonality and environmental E. coli is crucial for assessing water quality and potential health risks, implementing effective management strategies to control E. coli levels, and predicting changes in E. coli concentrations under future climate scenarios. The research on seasonality and environmental E. coli is ongoing, and more data is needed to fully understand the intricate interactions across diverse ecosystems. By continuing to investigate these dynamics, we can better protect the health of our environment and communities.
Conclusions
The findings of our study highlight the intricate and varied genomic attributes exhibited by MDR E. coli isolated from urban polluted environmental sources. The presence of strains exhibiting ESBL activity and carrying the blaCTX-M-15 gene has implications for clinical practice and a matter of significant public health concern. This gene has been linked to the extensive spread of MDR bacteria on a global scale. It is noteworthy that the presence of ST21 has not been documented in clinical isolates within the Ghanaian context. Furthermore, the detection of IncF plasmids and MGEs, which play a crucial role in promoting the dissemination of these ARGs, was also noted. The findings indicate that public health is at risk and requires urgent attention. To address this issue, it is imperative to develop and implement effective strategies pertaining to waste management. By doing so, we can effectively mitigate the discharge of ARB into the environment and, curtail the dissemination of ARGs. The results of this study provide evidence for the necessity of conducting AMR surveillance studies that encompass the One Health approach. Usage of high-throughput technologies such as WGS is crucial to gain genomic insights on the AMR status in Ghana, with the goal of protecting public health. The present study contributes to the expansion of our understanding of the resistance mechanisms and pathogenic attributes of E. coli, which facilitate its proliferation and dissemination within environmental niches.
Policy implications
The implications of this study are profound for both policy formulation and practical interventions. The evident insufficiency of proper waste treatment infrastructure in Ghana emerges as a key determinant of AMR proliferation, demanding immediate attention and resolution. Addressing the issue of inadequate waste management is intricate, involving various sectors and requiring the concerted efforts of multiple stakeholders, prominently the government. A critical component of this effort is the implementation of policies and actions, with a particular focus on providing robust Water, Sanitation, and Hygiene (WASH) facilities, especially in urban slums.
Recognizing the pivotal role of the media in disseminating information, it becomes essential for effective public education aimed at fostering behavioral change regarding waste management practices. Concurrently, the establishment of comprehensive public health surveillance is imperative to monitor AMR. Environmental officers hold the responsibility of ensuring compliance with waste management regulations, adding another layer to the multifaceted solution required.
Moreover, adopting a One Health approach is indispensable for the surveillance of antibiotic-resistant bacteria and genes. This approach facilitates evidence-based decision-making, particularly crucial in resource-poor settings like Ghana. By integrating insights from human, animal, and environmental health, a holistic understanding of the extent of the problem can be achieved, enabling more effective strategies to combat antimicrobial resistance.
Supporting information
S1 Fig. Frequency of Escherichia coli isolates showing number of antibiotic resistances.
https://doi.org/10.1371/journal.pone.0301531.s001
(TIF)
S2 Fig. Frequency of Escherichia coli isolates from sampling sites showing number of antibiotic resistances.
https://doi.org/10.1371/journal.pone.0301531.s002
(TIF)
S3 Fig. Percentage of Escherichia coli isolates from the sampling sites and their phenotypic resistance to 12 antibiotics.
Odaw River (OD), Korle Lagoon, sewage (SW) and hospitals effluents (HE), AMP—Ampicillin, CXM—Cefuroxime, CTX—Cefotaxime, CAZ—Ceftazidime, CRO—Ceftriaxone, CIP—Ciprofloxacin, AZM—Azithromycin, AK—Amikacin, MEM—Meropenem, C—Chloramphenicol, AK—Amikacin, SXT-Sulfamethoxazole-Trimethoprim.
https://doi.org/10.1371/journal.pone.0301531.s003
(TIF)
S4 Fig. Resistance patterns of Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s004
(TIF)
S5 Fig. Distribution of sequence types (STs) among 25 Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s005
(TIF)
S6 Fig. Distribution of Phylogroups among 25 Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s006
(TIF)
S7 Fig. Distribution of phylogroups among Escherichia coli isolates from sample sources.
https://doi.org/10.1371/journal.pone.0301531.s007
(TIF)
S8 Fig. Distribution of acquired resistance genes amongst Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s008
(TIF)
S9 Fig. Distribution of virulence genes across E. coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s009
(TIF)
S10 Fig. Distribution of plasmids amongst Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s010
(TIF)
S1 Table. Source and co-ordinates of sampling locations.
https://doi.org/10.1371/journal.pone.0301531.s011
(DOCX)
S2A Table. Genomic characteristics of E. coli isolates downloaded from the BVBR 3.30.19a.
https://doi.org/10.1371/journal.pone.0301531.s012
(DOCX)
S2B Table. Metadata of E. coli isolates downloaded from the BVBR 3.30.19a.
https://doi.org/10.1371/journal.pone.0301531.s013
(XLSX)
S3 Table. Antibiotic resistance and susceptibility values of Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s014
(DOCX)
S4 Table. Source and resistant profiles of Escherichia coli isolates.
https://doi.org/10.1371/journal.pone.0301531.s015
(DOCX)
S5 Table. Genomic characteristics of E. coli isolate.
https://doi.org/10.1371/journal.pone.0301531.s016
(DOCX)
Acknowledgments
The authors will like to thank, all members of Dr. Beverly Egyir’s Group, especially Felicia Owusu and Grebstad Rabbi Amuasi at the Bacteriology Department of the Noguchi Memorial Institute for Medical Research for their assistance in diverse ways.
References
- 1.
Heijden van der, Maarten, Sandgren, Andreas Pränting, Maria Karvanen, Matti Aagaard H, et al. New ReAct Report: When the Drugs Don’t Work—Antibiotic Resistance as a Global Development Problem. In: NEWS AND OPINIONS [Internet]. 2019 [cited 20 Jul 2020]. http://www.reactgroup.org.
- 2. Azanu D, Adu-Poku D, Saah SA, Appaw WO. Prevalence of Pharmaceuticals in Surface Water Samples in Ghana. J Chem. 2021;2021.
- 3. Borquaye LS, Ekuadzi E, Darko G, Ahor HS, Nsiah ST, Lartey JA, et al. Occurrence of antibiotics and antibiotic-resistant bacteria in landfill sites in Kumasi, Ghana. J Chem. 2019;2019.
- 4. Baquero F, Martínez JL, Cantón R. Antibiotics and antibiotic resistance in water environments. Curr Opin Biotechnol. 2008;19: 260–265. pmid:18534838
- 5. Kümmerer K. Resistance in the environment. J Antimicrob Chemother. 2004;54: 311–320. pmid:15215223
- 6. Adelowo OO, Caucci S, Banjo OA, Nnanna OC, Awotipe EO, Peters FB, et al. Extended Spectrum Beta-Lactamase (ESBL)-producing bacteria isolated from hospital wastewaters, rivers and aquaculture sources in Nigeria. Environ Sci Pollut Res. 2018;25: 2744–2755. pmid:29139076
- 7. Cantón R. Antibiotic resistance genes from the environment: A perspective through newly identified antibiotic resistance mechanisms in the clinical setting. Clin Microbiol Infect. 2009;15: 20–25. pmid:19220348
- 8. Aminov RI. Horizontal gene exchange in environmental microbiota. Front Microbiol. 2011;2: 1–19. pmid:21845185
- 9. Rosewarne CP, Pettigrove V, Stokes HW, Parsons YM. Class 1 integrons in benthic bacterial communities: Abundance, association with Tn402-like transposition modules and evidence for coselection with heavy-metal resistance. FEMS Microbiol Ecol. 2010;72: 35–46. pmid:20132306
- 10. Alves MS, Pereira A, Araújo SM, Castro BB, Correia ACM, Henriques I. Seawater is a reservoir of multi-resistant Escherichia coli, including strains hosting plasmid-mediated quinolones resistance and extended-spectrum beta-lactamases genes. Front Microbiol. 2014;5: 1–10. pmid:25191308
- 11. Czekalski N, Díez EG, Bürgmann H. Wastewater as a point source of antibiotic-resistance genes in the sediment of a freshwater lake. ISME J. 2014;8: 1381–1390. pmid:24599073
- 12. Banu RA, Alvarez JM, Reid AJ, Enbiale W, Labi AK, Ansa EDO, et al. Extended spectrum beta-lactamase escherichia coli in river waters collected from two cities in Ghana, 2018–2020. Trop Med Infect Dis. 2021;6: 2018–2020. pmid:34203078
- 13. Raven KE, Ludden C, Gouliouris T, Blane B, Naydenova P, Brown NM, et al. Genomic surveillance of Escherichia coli in municipal wastewater treatment plants as an indicator of clinically relevant pathogens and their resistance genes. Microb genomics. 2019;5. pmid:31107200
- 14. Mbanga J, Amoako DG, Abia ALK, Allam M, Ismail A, Essack SY. Genomic Insights of Multidrug-Resistant Escherichia coli From Wastewater Sources and Their Association With Clinical Pathogens in South Africa. Front Vet Sci. 2021;8: 1–13. pmid:33718473
- 15. Addae-Nuku DS, Kotey FCN, Dayie NTKD, Osei MM, Tette EMA, Debrah P, et al. Multidrug-Resistant Bacteria in Hospital Wastewater of the Korle Bu Teaching Hospital in Accra, Ghana. Environ Health Insights. 2022;16. pmid:36311334
- 16. Kichana E, Addy F, Dufailu OA. Genetic characterization and antimicrobial susceptibility of Escherichia coli isolated from household water sources in northern Ghana. J Water Health. 2022;20: 770–780. pmid:35635771
- 17. Massella E, Reid CJ, Cummins ML, Anantanawat K, Zingali T, Serraino A, et al. Snapshot study of whole genome sequences of escherichia coli from healthy companion animals, livestock, wildlife, humans and food in italy. Antibiotics. 2020;9: 1–22. pmid:33172096
- 18. Delgado-Blas JF, Valenzuela Agüi C, Marin Rodriguez E, Serna C, Montero N, Saba CKS, et al. Dissemination Routes of Carbapenem and Pan-Aminoglycoside Resistance Mechanisms in Hospital and Urban Wastewater Canalizations of Ghana. mSystems. 2022;7. pmid:35103490
- 19.
World Health Organization. Global priority list of antibiotic-resistant bacteria to guide research, discovery, and development of new antibiotics. 2017 [cited 11 Aug 2023]. https://www.who.int/medicines/publications/WHO-PPL-Short_Summary_25Feb-ET_NM_WHO.pdf.
- 20. Kappell AD, De Nies MS, Ahuja NH, Ledeboer NA, Newton RJ, Hristova KR. Detection of multi-drug resistant Escherichia coli in the urban waterways of Milwaukee, WI. Front Microbiol. 2015;6: 1–12. pmid:25972844
- 21. Kaper JB, Nataro JP, Mobley HLT. Pathogenic Escherichia coli. Nat Rev Microbiol. 2004;2: 123–140. pmid:15040260
- 22. Van Elsas JD, Semenov A V., Costa R, Trevors JT. Survival of Escherichia coli in the environment: Fundamental and public health aspects. ISME J. 2011;5: 173–183. pmid:20574458
- 23. Tadesse DA, Zhao S, Tong E, Ayers S, Singh A, Bartholomew MJ. Antimicrobial drug resistance in Escherichia coli from humans and food animals, United States, 1950–2002. Emerg Infect Dis. 2012;18: 741–749. pmid:22515968
- 24. Magiorakos A-P, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect. 2012;18: 268–281. pmid:21793988
- 25. Jianying HU, Jiachen SHI, Chang H, Dong LI, Yang MIN, Kamagata Y. Phenotyping and genotyping of antibiotic-resistant Escherichia coli isolated from a natural river basin. Environ Sci Technol. 2008;42: 3415–3420. pmid:18522127
- 26. Awosile BB, Agbaje M, Adebowale O, Kehinde O, Omoshaba E. Beta-lactamase resistance genes in Enterobacteriaceae from Nigeria. Afr J Lab Med. 2022;11: 1–10. pmid:35282396
- 27. Adelowo OO, Ikhimiukor OO, Knecht C, Vollmers J, Bhatia M, Kaster AK, et al. A survey of extended-spectrum beta-lactamase-producing Enterobacteriaceae in urban wetlands in southwestern Nigeria as a step towards generating prevalence maps of antimicrobial resistance. PLoS One. 2020;15: e0229451. Available: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7055906/pdf/pone.0229451.pdf. pmid:32130234
- 28. Characklis GW, Dilts MJ, Simmons OD, Likirdopulos CA, Krometis LAH, Sobsey MD. Microbial partitioning to settleable particles in stormwater. Water Res. 2005;39: 1773–1782. pmid:15899275
- 29. Krometis LAH, Characklis GW, Simmons OD, Dilts MJ, Likirdopulos CA, Sobsey MD. Intra-storm variability in microbial partitioning and microbial loading rates. Water Res. 2007;41: 506–516. pmid:17141293
- 30. Donovan EP, Staskal DF, Unice KM, Roberts JD, Haws LC, Finley BL, et al. Risk of gastrointestinal disease associated with exposure to pathogens in the sediments of the Lower Passaic River. Appl Environ Microbiol. 2008;74: 1004–1018. pmid:18156335
- 31. Kalkhajeh YK, Amiri BJ, Huang B, Khalyani AH, Wenyou H, Gao H, et al. Methods for Sample Collection, Storage, and Analysis. Water. 2019;11: 1–24.
- 32. Haberecht HB, Nealon NJ, Gilliland JR, Holder A V, Runyan C, Oppel RC, et al. Antimicrobial-Resistant Escherichia coli from Environmental Waters in Northern Colorado. 2019;2019. pmid:30906330
- 33.
CLSI. Performance Standards for Antimicrobial Susceptibility Testing. CLSI Supplement M100. Clinical and Laboratory Standards Institute. 31st ed. 2021.
- 34. Egyir B, Dsani E, Owusu-Nyantakyi C, Amuasi GR, Owusu FA, Allegye-Cudjoe E, et al. Antimicrobial resistance and genomic analysis of staphylococci isolated from livestock and farm attendants in Northern Ghana. BMC Microbiol. 2022;22: 1–10. pmid:35864456
- 35. Bolger A, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30: 2114–20. pmid:24695404
- 36.
Andrews S. FastQC: a quality control tool for high throughput sequence data. 2010.
- 37. Endoh H, Hirayama T, Aoyama T, Oka A. Characterization of the virA gene of the agropine-type plasmid pRiA4 of Agrobacterium rhizogenes. FEBS Lett. 1990;271: 28–32. pmid:2226811
- 38. Gaşpar C-M, Cziszter LT, Lăzărescu CF, Ţibru I, Pentea M, Butnariu M. Antibiotic Resistance among Escherichia coli Isolates from Hospital Wastewater Compared to Community Wastewater. Water. 2021;13: 3449.
- 39. Adomako LAB, Yirenya-Tawiah D, Nukpezah D, Abrahamya A, Labi AK, Grigoryan R, et al. Reduced bacterial counts from a sewage treatment plant but increased counts and antibiotic resistance in the recipient stream in accra, ghana—a cross-sectional study. Trop Med Infect Dis. 2021;6. pmid:34068850
- 40. Newman MJ, Frimpong E, S DE, Opintan JA, Asamoah-Adu A. Resistance to antimicrobial drugs in Ghana. Infect Drug Resist. 2011;4: 215–220. pmid:22259250
- 41. Obeng-Nkrumah N, Twum-Danso K, Krogfelt KA, Newman MJ. High levels of extended-spectrum beta-lactamases in a major teaching hospital in Ghana: The need for regular monitoring and evaluation of antibiotic resistance. Am J Trop Med Hyg. 2013;89: 960–964. pmid:24043693
- 42. Dela H, Egyir B, Majekodunmi AO, Behene E, Yeboah C, Ackah D, et al. Diarrhoeagenic E. coli occurrence and antimicrobial resistance of Extended Spectrum Beta-Lactamases isolated from diarrhoea patients attending health facilities in Accra, Ghana. PLoS One. 2022;17: 1–15. pmid:35617316
- 43. Donkor ES, Tetteh-Quarcoo PB, Nartey P, Agyeman IO. Self-medication practices with antibiotics among tertiary level students in Accra, Ghana: A cross-sectional study. Int J Environ Res Public Health. 2012;9: 3519–3529. pmid:23202760
- 44. Falgenhauer L, Imirzalioglu C, Oppong K, Akenten CW, Hogan B, Krumkamp R, et al. Detection and characterization of ESBL-producing Escherichia coli from humans and poultry in Ghana. Front Microbiol. 2019;10: 1–8. pmid:30697208
- 45. Gyesi JN, Nyaaba BA, Darko G, Mills-Robertson FC, Miezah K, Acheampong NA, et al. Occurrence of Pharmaceutical Residues and Antibiotic-Resistant Bacteria in Water and Sediments from Major Reservoirs (Owabi and Barekese Dams) in Ghana. J Chem. 2022;2022.
- 46. Azanu D, Styrishave B, Darko G, Weisser JJ, Abaidoo RC. Occurrence and risk assessment of antibiotics in water and lettuce in Ghana. Sci Total Environ. 2018;622–623: 293–305. pmid:29216470
- 47. Dodd M, Amponsah LO, Grundy S, Darko G. Human health risk associated with metal exposure at Agbogbloshie e-waste site and the surrounding neighbourhood in Accra, Ghana. Environ Geochem Health. 2023;45: 4515–4531. pmid:36853522
- 48. Caravanos J, Clark E, Fuller R, Lambertson C. Assessing Worker and Environmental Chemical Exposure Risks at an e-Waste Recycling and Disposal Site in Accra, Ghana. J Heal Pollut. 2011;1: 16–25.
- 49. Fosu-Mensah BY, Addae E, Yirenya-Tawiah D, Nyame F. Heavy metals concentration and distribution in soils and vegetation at Korle Lagoon area in Accra, Ghana. Cogent Environ Sci. 2017;3.
- 50. Chama M, Amankwa E, Oteng-Ababio M. Trace metal levels of the Odaw river sediments at the Agbogbloshie e-waste recycling site. J Sci Technol. 2014;34: 1.
- 51. Fuentes AM, Ama CF. Mercury induces multiple antibiotic resistance in. 1997;154: 385–388.
- 52. Amachawadi RG, Scott HM, Vinasco J, Tokach MD, Dritz SS, Nelssen JL, et al. Effects of In-Feed Copper, Chlortetracycline, and Tylosin on the Prevalence of Transferable Copper Resistance Gene, tcrB, among Fecal Enterococci of Weaned Piglets. Foodborne Pathog Dis. 2015;12: 670–678. pmid:26258261
- 53. Baker-Austin C, Wright MS, Stepanauskas R, McArthur J V. Co-selection of antibiotic and metal resistance. Trends Microbiol. 2006;14: 176–182. pmid:16537105
- 54. Odonkor ST, Simpson SV, Morales Medina WR, Fahrenfeld NL. Antibiotic-Resistant Bacteria and Resistance Genes in Isolates from Ghanaian Drinking Water Sources. J Environ Public Health. 2022;2022. pmid:36246472
- 55. Pruden A, Pei R, Storteboom H, Carlson KH. Antibiotic resistance genes as emerging contaminants: studies in northern Colorado. Environ Sci Technol. 2006;40: 7445–7450. Available: http://www.academia.edu/download/46401665/Antibiotic_Resistance_Genes_as_Emerging_20160611-22969-c3isqf.pdf. pmid:17181002
- 56. Labi AK, Obeng-Nkrumah N, Sunkwa-Mills G, Bediako-Bowan A, Akufo C, Bjerrum S, et al. Antibiotic prescribing in paediatric inpatients in Ghana: A multi-centre point prevalence survey. BMC Pediatr. 2018;18: 1–9. pmid:30572851
- 57. García-Vello P, González-Zorn B, Saba CKS. Antibiotic resistance patterns in human, animal, food and environmental isolates in ghana: A review. Pan Afr Med J. 2020;35: 1–15. pmid:32499853
- 58. Mahazu S, Sato W, Ayibieke A, Prah I, Hayashi T, Suzuki T, et al. Insights and genetic features of extended-spectrum beta-lactamase producing Escherichia coli isolates from two hospitals in Ghana. Sci Rep. 2022;12: 1–11. pmid:35115628
- 59. Krumperman P. Multiple antibiotic resistance indexing of Escherichia coli to identify high-risk sources of faecal contamination of water. Appl Environ Microbiol. 1983;46: 165–170. pmid:25779106
- 60. Grabow WO, Prozesky OW. Drug resistance of coliform bacteria in hospital and city sewage. Antimicrob Agents Chemother. 1973;3: 175–180. pmid:4597713
- 61. Figueira V, Serra E, Manaia CM. Differential patterns of antimicrobial resistance in population subsets of Escherichia coli isolated from waste- and surface waters. Sci Total Environ. 2011;409: 1017–1023. pmid:21215425
- 62. Bong CW, Low KY, Chai LC, Lee CW. Prevalence and Diversity of Antibiotic Resistant Escherichia coli From Anthropogenic-Impacted Larut River. Front Public Heal. 2022;10: 1–15. pmid:35356018
- 63. Deku JG, Duedu KO, Kinanyok S, Kpene GE, Feglo PK. Phylogenicity and Virulence Profiles of Clinical Escherichia coli Isolates in the Ho Teaching Hospital of Ghana. Biomed Res Int. 2022;2022.
- 64. Walk ST, Alm EW, Calhoun LM, Mladonicky JM, Whittam TS. Genetic diversity and population structure of Escherichia coli isolated from freshwater beaches. Environ Microbiol. 2007;9: 2274–2288. pmid:17686024
- 65. Ghaderpour A, Ho WS, Chew LL, Bong CW, Chong VC, Thong KL, et al. Diverse and abundant multi-drug resistant E. coli in Matang mangrove estuaries, Malaysia. Front Microbiol. 2015;6. pmid:26483759
- 66. Bingen E, Picard B, Brahimi N, Mathy S, Desjardins P, Elion J, et al. Phylogenetic analysis of Escherichia coli strains causing neonatal meningitis suggests horizontal gene transfer from a predominant pool of highly virulent B2 group strains. J Infect Dis. 1998;177: 642–650. pmid:9498443
- 67. Gordon DM, Cowling A. The distribution and genetic structure of Escherichia coli in Australian vertebrates: Host and geographic effects. Microbiology. 2003;149: 3575–3586. pmid:14663089
- 68. Rolland K, Lambert-Zechovsky N, Picard B, Denamur E. Shigella and enteroinvasive Escherichia coli strains are derived from distinct ancestral strains of E. coli. Microbiology. 1998;144: 2667–2672. pmid:9782516
- 69. Picard B, Garcia JS, Gouriou S, Duriez P, Brahimi N, Bingen E, et al. The link between phylogeny and virulence in Escherichia coli extraintestinal infection? Infect Immun. 1999;67: 546–553. pmid:9916057
- 70. Nowrouzian FL, Adlerberth I, Wold AE. Enhanced persistence in the colonic microbiota of Escherichia coli strains belonging to phylogenetic group B2: role of virulence factors and adherence to colonic cells. Microbes Infect. 2006;8: 834–840. pmid:16483819
- 71. Ochman H, Lawrence JG, Grolsman EA. Lateral gene transfer and the nature of bacterial innovation. Nature. 2000;405: 299–304. pmid:10830951
- 72. Clermont O, Bonacorsi S, Bingen E. Rapid and simple determination of the Escherichia coli phylogenetic group. Appl Environ Microbiol. 2000;66: 4555–4558. pmid:11010916
- 73. Pérez C, Gómez-Duarte OG, Arias ML. Diarrheagenic Escherichia coli in children from Costa Rica. Am J Trop Med Hyg. 2010;83: 292–297. pmid:20682870
- 74. Mosquito S, Pons MJ, Riveros M, Ruiz J, Ochoa TJ. Diarrheagenic escherichia coli phylogroups are associated with antibiotic resistance and duration of diarrheal episode. Sci World J. 2015;2015. pmid:25811044
- 75. Founou LL, Founou RC, Allam M, Ismail A, Essack SY. Genome Analysis of ESBL-Producing Escherichia coli Isolated from Pigs. Pathogens. 2022;11: 1–18. pmid:35890020
- 76. Zweifel C, Cernela N, Stephan R. Detection of the emerging Shiga toxin-producing Escherichia coli O26: H11/H- sequence type 29 (ST29) clone in human patients and healthy cattle in Switzerland. Appl Environ Microbiol. 2013;79: 5411–5413. pmid:23811503
- 77. Zarfel G, Lipp M, Gürtl E, Folli B, Baumert R, Kittinger C. Troubled water under the bridge: Screening of River Mur water reveals dominance of CTX-M harboring Escherichia coli and for the first time an environmental VIM-1 producer in Austria. Sci Total Environ. 2017;593–594: 399–405. pmid:28351808
- 78. Deku JG, Duedu KO, Ativi E, Kpene GE, Feglo PK. Occurrence and distribution of extended-spectrum β-lactamase in clinical Escherichia coli isolates at Ho Teaching Hospital in Ghana. Ghana Med J. 2021;55: 298–307. pmid:35957938
- 79. Prah I, Ayibieke A, Mahazu S, Sassa CT, Hayashi T, Yamaoka S, et al. Emergence of oxacillinase-181 carbapenemase-producing diarrheagenic Escherichia coli in Ghana. Emerg Microbes Infect. 2021;10: 865–873. pmid:33879019
- 80. Pitout JDD, Nordmann P, Laupland KB, Poirel L. Emergence of Enterobacteriaceae producing extended-spectrum β-lactamases (ESBLs) in the community. J Antimicrob Chemother. 2005;56: 52–59. pmid:15917288
- 81. Mahazu S, Prah I, Ayibieke A, Sato W, Hayashi T, Suzuki T, et al. Possible Dissemination of Escherichia coli Sequence Type 410 Closely Related to B4/H24RxC in Ghana. Front Microbiol. 2021;12: 1–13. pmid:34925277
- 82. Wang J, Stephan R, Karczmarczyk M, Yan Q, Hächler H, Fanning S. Molecular characterization of blaESBL-harboring conjugative plasmids identified in multi-drug resistant Escherichia coli isolated from food-producing animals and healthy humans. Front Microbiol. 2013;4: 1–9. pmid:23874325
- 83. Yang QE, Sun J, Li L, Deng H, Liu BT, Fang LX, et al. IncF plasmid diversity in multi-drug resistant Escherichia coli strains from animals in China. Front Microbiol. 2015;6. pmid:26441898
- 84. Paterson DL. Recommendation for treatment of severe infections caused by Enterobacteriaceae producing extended-spectrum β-lactamases (ESBLs). Clin Microbiol Infect. 2000;6: 460–463. pmid:11168179
- 85. Hamilton MJ, Hadi AZ, Griffith JF, Ishii S, Sadowsky MJ. Large scale analysis of virulence genes in Escherichia coli strains isolated from Avalon Bay, CA. Water Res. 2010;44: 5463–5473. pmid:20643468
- 86. Kuhnert P, Boerlin P, Frey J. Target genes for virulence assessment of Escherichia coli isolates from water, food and the environment. FEMS Microbiol Rev. 2000;24: 107–117. pmid:10640601
- 87. López-Banda DA, Carrillo-Casas EM, Leyva-Leyva M, Orozco-Hoyuela G, Manjarrez-Hernández ÁH, Arroyo-Escalante S, et al. Identification of virulence factors genes in escherichia coli isolates from women with urinary tract infection in Mexico. Biomed Res Int. 2014;2014. pmid:24895634
- 88. Jang J, Hur HG, Sadowsky MJ, Byappanahalli MN, Yan T, Ishii S. Environmental Escherichia coli: ecology and public health implications—a review. J Appl Microbiol. 2017;123: 570–581. pmid:28383815
- 89. Vila J, Simon K, Ruiz J, Horcajada JP, Velasco M, Barranco M, et al. Are quinolone-resistant uropathogenic Escherichia coli less virulent? J Infect Dis. 2002;186: 1039–1042. pmid:12232848