The first characterization of Escherichia coli isolates recovered from three wastewater treatment plants in Romania

Antimicrobial resistance (AMR) is an increasingly critical issue in global health, impacting the effectiveness of antibiotics, as well as the management of infectious diseases (Haenni et al. 2022; WHO 2023). There is an increasing concern that the environment is a significant reservoir and spreader of AMR. Factors such as antimicrobial use in healthcare, agriculture, and environmental contamination contribute to the dispersal of antibiotic-resistant bacteria (ARB) and antibiotic-resistant genes (ARGs). These elements spread through various pathways, including wastewaters and agricultural runoff (Samreen et al. 2021; Le et al. 2023).

Wastewater treatment plants (WWTPs) play a crucial role in removing pollutants and pathogens from wastewater before it is discharged into the environment (Le et al. 2023). However, these plants are not always effective in eliminating all ARB, particularly those that have developed resistance to multiple antibiotics (Grehs et al. 2021). Therefore, WWTPs serve as a potential source for ARB due to the constant exposure to antibiotics and diverse bacterial populations (Lupan et al. 2017; Schwermer et al. 2017; Mutuku et al. 2022; Wang et al. 2023). One bacterium that has attracted considerable attention in this context is Escherichia coli (E. coli), a commonly found microorganism present in the intestinal tract of humans and animals and various environments, including wastewater. E. coli, commonly known as an indicator of faecal contamination, plays a significant role in understanding the extent of AMR in aquatic environments (Jang et al. 2017). Furthermore, E. coli is a good potential indicator for monitoring antibiotic resistance in the environment, because its detection is relatively cheap, easy to implement, and allows for comparison of data from humans and animals (Anjum et al. 2021; Dioli et al. 2023).

Moreover, antibiotic-resistant E. coli can be released into rivers, streams, groundwater, and other water sources, potentially exposing humans and animals to these harmful bacteria (Bréchet et al. 2014). While most strains of E. coli are harmless, some can cause serious infections, including urinary tract infections (UTIs), pneumonia, meningitis, appendicitis, bloodstream, and gastrointestinal infections, skin abscesses, intra-amniotic and puerperal infections in pregnant women, and endocarditis (Jang et al. 2017; Price & Wildeboer 2017; Id et al. 2021; Barbu et al. 2023).

The increasing prevalence of antibiotic-resistant E. coli, particularly in wastewater environments, poses a significant public health threat; therefore, appropriately addressing such issues could be useful in designing effective strategies to mitigate the spread of AMR (Lupan et al. 2017; Anjum et al. 2021; Bojar et al. 2021; Wang et al. 2023).

The spread of antibiotic-resistant E. coli can occur through various mechanisms, including direct contact with contaminated water, consumption of contaminated food or water, and contact with contaminated animals. Once an individual is infected with an antibiotic-resistant E. coli strain, the bacteria can subsequently spread to others through close contact, making it difficult to control and contain potential outbreaks (Anjali et al. 2023; Brătfelan et al. 2023).

WWTPs could represent a critical source for monitoring AMR, as they acts as a conduit for diverse bacteria, including E. coli, from different sources such as hospitals and households (Manaia et al. 2016; Lood et al. 2017; Baghal et al. 2021). Investigating the AMR profiles of E. coli isolates obtained from Romanian wastewater can provide valuable insights into the prevalence, distribution, and factors contributing to antibiotic resistance in this specific setting. Moreover, it could also contribute to the elaboration of strategies for monitoring and controlling AMR in the environment.

Therefore, this study aimed to investigate the prevalence and AMR profiles of E. coli isolates recovered from three different WWTPs in Romania, during the warm season of 2023. We used a limited sample size (24) to provide preliminary insights and guide future, more comprehensive studies. Despite the small number of samples, rigorous microbiological methods were employed to ensure the reliability of the data.

Our research specifically targets WWTPs in Romania, a region with limited available data on antibiotic-resistant E. coli. Although the prevalence of antibiotic-resistant E. coli in wastewater from WWTPs has been extensively documented in the literature, by concentrating on this geographic area, our study aims to provide new baseline data that can inform local public health policies and wastewater management strategies. Moreover, we have screened for a wider spectrum of antibiotics, including veterinary antibiotics and those less frequently studied, offering a more comprehensive understanding of resistance patterns.

Wastewater samples were collected simultaneously, during the spring and summer seasons, in April and August 2023 (twice a month), from the influents and effluents of three different WWTPs (A, B, and C) located in the central-west region of Romania. WWTP A is designed to process around 115,000 cubic metres (cbm)/24 h of wastewater from an average of 300,000 inhabitants. WWTP B processes around 5,673 cbm/24 h, from an average of 19,000 inhabitants while WWTP C is currently treating around 4,772 cbm/24 h, from around 31,000 inhabitants. The selected WWTPs (Table 1) are used to treat wastewater originating mainly from households and hospital areas of the cities and the rural areas upstream holding a public sewerage network. The hospitals contributing to the wastewater include a mix of general, specialized, and referral hospitals. In the selected WWTPs, sewage treatment is performed using conventional treatment procedures, which include mechanical, biological, and chemical methods.

The selected WWTPs represent a range of treatment capacities and technological setups, providing insights into the effectiveness of different processes and scales of operation in reducing antibiotic-resistant E. coli. Larger plants like WWTP A, serving a bigger population, may have more advanced treatment stages, potentially leading to lower levels of ARB in their effluents compared with smaller plants like WWTP B and C. Additionally, WWTP A's larger service area results in a higher load of pathogens and antibiotic contaminants, offering unique challenges and opportunities for assessing treatment efficacy.

The samples were collected using continuous 24-h sampling, in 1-L sterile polyethylene bottles, without any preservatives. Continuous 24-h sampling involves collecting wastewater samples over a full 24-h period using an automated machine. The machine typically collects samples at regular intervals, combining them into a composite sample that accurately reflects the average characteristics of the wastewater over the entire day.

The samples were transported at 4 °C in a portable cooler for less than 1 h, then stored at 5 ± 3 °C and analysed within 24 h.

The isolation of the E. coli strains was performed according to the steps mentioned in the protocol (‘SR EN ISO 9308-1:2015/A1:2017’ 2017). The protocol (‘SR EN ISO 9308-1:2015/A1:2017’ 2017) – Water quality. Enumeration of E. coli and coliform bacteria. The membrane filtration method for waters with low bacterial background flora standard aims to detect and enumerate E. coli as an indicator of faecal pollution of water, as well as coliform bacteria. The method is based on the membrane filtration of water samples, followed by the placement of the membrane on the surface of a chromogenic coliform agar plate. After incubation at 36 ± 2 °C for 21–24 h, β-d-galactosidase and β-d-glucuronidase-positive colonies (dark blue to violet) were counted as E. coli, while pink to red colonies (β-d-galactosidase positive) were considered presumptive coliform bacteria other than E. coli (confirmed by a negative reaction to oxidase) (Antohiu & Lapohos 2015).

The typical E. coli colonies were afterward inoculated on TBX medium and incubated at 40 °C. Vitek^® 2 GN cards for identifying a broad range of Gram-negative Enterobacteriaceae (intended for use with the Vitek 2 system) (bioMérieux, Marcy l'Etoile, France) were further used for the biochemical confirmation of the strains.

Antimicrobial susceptibility was assessed by determining the minimum inhibitory concentration (MIC) values, by using the VITEK^® 2 System and the AST-GN96 cards for Gram-negative bacteria (bioMérieux, Marcy l'Etoile, France). The above-mentioned card is essentially considered a miniaturized and abbreviated version of the doubling dilution technique for MICs determined by the microdilution method.

The susceptibility to the following antimicrobials was determined: ampicillin (AMP), amoxicillin/clavulanic acid (AMX/CL), ticarcillin/clavulanic acid (TIC/CL), cefalexin (CFX), cefalotin (CFT), cefoperazone (CFZ), ceftiofur (CFTI), cefquinome (CFQ), imipenem (IMI), gentamicin (GEN), neomycin (NEO), flumequine (FLU), enrofloxacin (ENR), marbofloxacin (MRB), tetracycline (TET), florfenicol (FLR), polymyxin B (PB), and trimethoprim/sulfamethoxazole (TMT/SMX). These antimicrobials cover various classes and mechanisms of action and were chosen based on their clinical relevance and environmental prevalence for testing E. coli in wastewater (Huijbers et al. 2020). Additionally, we also included veterinary antibiotics to demonstrate the potential contamination of agricultural environments and to expand knowledge about veterinary antibiotic resistance in E. coli.Table 2 categorizes the antibiotics according to their exclusive or shared use in human and veterinary medicine.

The results obtained by using the AST-GN96 cards for Gram-negative bacteria (bioMérieux, Marcy l'Etoile, France) were automatically interpreted by the VITEK^® 2 System and categorized as susceptible (S), intermediate (I), and resistant (R). The E. coli isolates, which proved to be resistant to more than three antimicrobial classes, were considered multidrug-resistant (MDR).

The total genomic DNA was extracted following a protocol previously described by Mihaiu et al. (2014). In brief, three specific E. coli isolates were removed from the MacConkey agar plates and afterwards resuspended into 150 μL Chelex solution (Sigma Aldrich, St. Louis, MO, USA). The samples were then subjected to a high-temperature protocol for cell membrane lysis (94 °C for 15 min and 56 °C for 10 min). A Nanodrop ND-1000 spectrophotometer analyser (NanoDrop Technologies, Wilmington, DE, USA) was further used in order to assess the quality and quantity of the extracted DNA.

The presence of AMR genes, namely blaSHV, blaCMY, blaTEM, blaCTX, blaOXA (β-lactamase genes), qnrA (quinolones), aac (gentamicin), sul1 (sulphonamides), and tetA and tetB (tetracyclines) was investigated by PCR multiplex.

The PCR protocol used was previously described by Chirila et al. (2017). Briefly, the PCR reaction mix (25 μL) consisted of the following: 1 × PCR green buffer, 2.5 mM MgCl₂; 5 pmol of each primer, dNTPs each at 200 μM, 2.5 U of TaqDNA polymerase (Promega), and 100 ng of genomic DNA. The analysis was performed under the following conditions: 94 °C for 3 min followed by 35 cycles of 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 1 min, and a final extension step of 73 °C for 5 min. Around 10 μL of the amplified product were loaded into agarose gels (2%). The gels were then stained with EvaGreen (JenaBioscience, Jena, Germany) and electrophoresed (90 W) for 40 min. Visualization was performed under UV light with a Gel Doc XR + Imager (Bio-Rad, Hercules, CA, USA). Strains of MDR E. coli (O157:K88ac: H19, CAPM 5933) were used as positive controls. The primers used to detect the presence of the above-mentioned AMR genes have been previously reported (Chirila et al. 2017).

Following the isolation protocol, a total of 12 E. coli isolates were recovered from the 24 analysed samples (12/24; 50% prevalence).

All recovered E. coli isolates proved to be susceptible to CFZ, CFTI, CFQ, IMI, FLR, and PB. The highest level of resistance found in all the E. coli isolates was to AMP (R-66.66%), TIC/CL (R-50%; I-16.66%), and CFT (R-33.33%; I-33.33%). Strong resistance to TET (R-50%) and TMT/SMX (R-50%) was also observed. The E. coli isolates showed low percentages of resistance to AMX/CL (R-16.66%), GEN (R-16.66%), FLU (R-16.66%), ENR (R-16.66%), and MRB (R-16.66%).

Consistent with their resistance phenotypes, AMR determinants detected among the E. coli isolates included tetA (6/12; 50%), tetB (6/12; 50%), sul1 (6/12; 50%), blaTEM (5/12; 41.66%), blaCTX (3/12; 25%), qnrA (1/12; 8.33%), and aac (1/12; 8.33%) (Table 3).

The study primarily revealed the presence of E. coli in wastewater as well as their AMR patterns, with an emphasis on the releasing of treated wastewater containing MDR E. coli into the environment. The study found a high prevalence of 83.34% MDR E. coli, followed by 50% of the isolates resistant to TET and TMT/SMX, and 41.66% showed resistance to AMP and TIC/CL. This is concerning, as it indicates a significant presence of ARB in the wastewater, potentially posing a threat to public health (Kraemer et al. 2019). Notably, this finding aligns with previous reports from Romania, which have revealed high levels of antibiotic resistance among E. coli isolates from wastewaters (Barbu et al. 2023). For example, one study found that 85.11% of E. coli isolates from hospital wastewater were MDR, meaning they were resistant to three or more classes of antibiotics. Moreover, the study found that 68.09% of E. coli isolates from community wastewater were MDR (Gaşpar et al. 2021). A study by Kwak et al. (2015) reported the prevalence of antibiotic-resistant E. coli in Stockholm's wastewater over one year, finding that 34% of the urban wastewater isolates and 55% of the hospital wastewater isolates were resistant to at least one antibiotic. This suggests that wastewater analysis may represent a potent early warning system for resistance trends, complementing traditional clinical surveillance (Kwak et al. 2015). In our research, the highest resistance levels in the recovered E. coli isolates were observed for AMP (66.66%), TIC/CL (50% resistant and 16.66% intermediate), and CFT (33.33% resistant and 33.33% intermediate). The high resistance rate to AMP and TIC/CL is concerning but not surprising, as β-lactam antibiotics are often reported to have diminished effectiveness against various bacterial pathogens due to widespread use and resultant resistance development. This highlights the importance of exploring alternative treatments and the prudent use of β-lactams (Naqid et al. 2020).

Resistance to TMT/SMX (50%) is particularly troubling as well, this being a combination of antimicrobial agents used to treat UTIs, among others, and raises questions about the available treatment options for common E. coli infections. A study by Mulder et al. (2019) linked prior use of extended-spectrum penicillins and sulphonamides (including trimethoprim itself) with significantly increased trimethoprim resistance in E. coli UTIs. A similar percentage of resistance (50%) was also identified in the case of TET. Lower resistance rates were recorded for AMX/CL, GEN, FLU, ENR, and MRB (16.66%). The findings of the current study are in line with several studies reporting similar resistance rates in European countries (Huijbers et al. 2020; Gregova et al. 2021). Interestingly, we have also observed that antibiotic resistance rates identified in the case of the E. coli isolates recovered from wastewater samples from our study are comparable to or exceed the clinical resistance rates reported by the European Centre for Disease Prevention and Control (ECDC), which indicates a significant environmental reservoir of resistant bacteria that could contribute to the clinical burden of antibiotic resistance (ECDC 2022). As per data from ECDC (Surveillance Atlas of Infectious Diseases), the resistance rates of E. coli isolates in Romania in 2022 were as follows: 62.4% to aminopenicillins, 23.8% to fluoroquinolones, 17.8% to third-generation cephalosporins, and 12% to aminoglycosides (ECDC 2022). The comparison of clinical and environmental isolates underscores the potential for wastewater to serve not only as a reflection of the antibiotic resistance problems faced in clinical settings but also as a contributor to the spread of resistance genes in the environment (Huijbers et al. 2020).

Notably, our study found that E. coli from WWTPs were susceptible to a wider range of antibiotics than expected, (such as CFZ, CFTI, CFQ, IMI, and PB) including some agents typically reserved for severe infections (imipenem). This suggests that these particular antibiotics could be effective against E. coli infections. However, the small sample size, potential differences between environmental and clinical E. coli, and limitations of in vitro studies necessitate further research before using these findings for treatment recommendations. However, the study highlights the importance of effective wastewater treatment and the need for surveillance systems that bridge clinical and environmental monitoring, in order to fully understand and combat the spread of antibiotic resistance (Huijbers et al. 2020; Cao et al. 2021; Gregova et al. 2021).

The detection of E. coli isolates that proved to be resistant to veterinary-specific antibiotics (such as ceftiofur, cefquinome, enrofloxacin, marbofloxacin, flumequine, and florfenicol) implies potential agricultural contamination, notably from the misuse of antibiotics in livestock and improper disposal of animal waste. This scenario suggests that ARB from farms can spread to the environment, particularly through manure used as fertilizer or direct wastewater discharge, exacerbating the risk of resistance transfer and public health threats. It also underscores the need for better waste management in agriculture and better strategies to curb the spread of AMR (Naqid et al. 2020). Several studies highlight that effluents from types of livestock farms significantly affect water quality due to pollutants such as organic matter and microorganisms, including coliform bacteria (Cesoniene et al. 2018; Pham-duc et al. 2020; Cao et al. 2021). The findings from our study, alongside those of other studies, highlight the interconnected nature of antibiotic use in agriculture, the development and spread of AMR, and the need for integrated surveillance systems (Hernando-Amado et al. 2019; Gregova et al. 2021).

Moreover, rainfall plays a crucial role in introducing contaminants into wastewater from agricultural runoff. Data from the Romanian National Meteorological Administration (https://www.meteoromania.ro/, 20 July 2024) for two stations in 2023 indicate significantly higher rainfall (R24) in August (83.5 and 102 mm) compared to April (22 and 31.8 mm). Higher rainfall in August likely increases agricultural runoff, thereby introducing more contaminants into wastewater systems. Rainfall events can elevate nutrient and contaminant runoff from agricultural fields into adjacent water bodies, thus, the introduction of E. coli bacteria and genes from veterinary sources into the wastewater system (Awad et al. 2015; Urase et al. 2020).

These findings raise concerns about the potential for antibiotic-resistant E. coli to spread from wastewater to humans and animals, making it more difficult to treat infections. Also, the research highlights the role of the urban WWTPs as reservoirs and dissemination points of ARB and ARGs (Naqid et al. 2020; Shamsizadeh et al. 2021; Brătfelan et al. 2023).

Also, it is important to mention that three studies previously conducted in the targeted region have indicated that hospital wastewaters do not significantly alter the bacterial diversity or ARG profiles in municipal wastewater systems (Szekeres et al. 2017; Teban-Man et al. 2021, 2022). For example, an analysis revealed that MDR Klebsiella pneumoniae from hospital and non-hospital sources had similar antibiotic susceptibility patterns (Teban-Man et al. 2021). The hospital's specific wastewater treatment processes can vary or may not exist at all, but may include methods such as activated sludge and chlorine disinfection (Szekeres et al. 2017). One hospital that pretreats its wastewater showed moderate efficiency in reducing ARGs, though potentially pathogenic bacteria slightly increased (Szekeres et al. 2017). Overall, the presence of hospital effluents did not significantly impact the bacterial diversity or the pool of carbapenem-resistant genes in municipal wastewater (Teban-Man et al. 2022). The potential dilution effect, as hospitals use more water per capita than domestic sources, likely mitigates the impact of hospital waste on the overall municipal wastewater characteristics (Kumari et al. 2020).

Similar findings were reported in several studies (Marinescu et al. 2015; Bojar et al. 2021; Dioli et al. 2023; Le et al. 2023; Skof et al. 2024). The high prevalence of blaTEM and blaCTX genes in different samples is considered to be responsible for resistance to various β-lactam antibiotics like ampicillin, ticarcillin, and cefotaxime. Moreover, the presence of specific antibiotic resistance determinants, including blaTEM and blaCTX, supports the idea that resistance genes, especially those conferring resistance to β-lactams like ticarcillin/clavulanic acid, can be transferred or maintained within plasmids or other mobile genetic elements among E. coli isolates. The existence of E. coli isolates resistant to AMP, TIC/CL, and first-generation cephalosporins, but susceptible to AMX/CL suggests the presence of certain ARGs that could provide insights into the molecular mechanisms underlying these resistance phenotypes. Certain beta-lactamase genes can hydrolyze these antibiotics but are generally inhibited by clavulanic acid, which could explain the susceptibility to AMX/CL. The blaTEM and blaCTX genes are often located on plasmids, which are mobile genetic elements that can be transferred between bacteria through horizontal gene transfer (HGT) processes, such as conjugation. The blaTEM gene is also known to confer resistance to a wide range of penicillins and has been widely identified in various environmental samples, including wastewater, which might facilitate the spread of resistance genes among bacterial populations in wastewater environments. The ease with which these genes can be transferred and maintained within bacterial communities contributes to the observed patterns of antibiotic resistance (Narciso-Da-Rocha et al. 2014; Mutuku et al. 2022).

The prevalence of tetracycline, sulphonamide, and β-lactam resistance genes highlights the need for targeted strategies to reduce the spread of these resistant bacteria. Interventions could include stricter antibiotic stewardship programmes to minimize the overuse and misuse of these antibiotics, both in clinical settings and in agriculture (https://onehealthejp.eu/about, 17 August 2024). Moreover, continuous surveillance of resistance patterns is essential to adapt treatment guidelines and ensure the effectiveness of existing antibiotics. The findings underscore the importance of a coordinated One Health approach to tackle antibiotic resistance on a global scale (https://www.who.int/news-room/fact-sheets/detail/one-health, 17 August 2024).

This study provided valuable preliminary insights into the prevalence and AMR profiles of E. coli isolates in three WWTPs from the central-west region of Romania, during the spring and summer seasons. While the small sample size limits the generalizability of the findings, the results highlight significant trends and raise important questions about the effectiveness of WWTPs in managing antibiotic resistance pollution.

A high prevalence of MDR E. coli was identified, with 83.34% of isolates exhibiting resistance to multiple antibiotics. Notably, 50% of isolates were resistant to TET and TMT/SMX, and 66.66% showed resistance to AMP. These findings align with other studies highlighting similar resistance patterns in wastewater, emphasizing the environmental reservoir of resistant bacteria. The presence of resistance genes such as blaTEM, tetA, tetB, qnrA, aac, and sul1 was confirmed. These genes, often located on mobile genetic elements like plasmids, facilitate the transfer and maintenance of resistance within bacterial populations. This highlights the ease with which antibiotic resistance can spread, underscoring the need for effective control measures.

The study draws attention to the potential for antibiotic-resistant E. coli to spread from wastewater to natural water bodies, agricultural fields, and subsequently to humans and animals. This emphasizes the interconnectedness through the One Health perspective of environmental and public health, where resistant bacteria from agricultural runoff and wastewater can enter water supplies. Higher rainfall in August likely contributed to increased agricultural runoff, introducing more contaminants into wastewater systems.

The findings also underline the need for more robust surveillance programmes to track the emergence and spread of resistant bacteria and genes, especially in wastewater systems known to act as reservoirs for such resistance. The obtained results also underscore the need for a comprehensive approach that addresses antibiotic use practices in both human and animal sectors, alongside improved wastewater treatment strategies. By tackling these issues, we can work towards mitigating the spread of antibiotic resistance and preserving the effectiveness of these crucial medicines.

Future studies with larger sample sizes and extended sampling periods are necessary to validate these findings and provide a more comprehensive understanding of the environmental impact of E. coli resistance in wastewater. Enhanced wastewater treatment processes and integrated surveillance systems are crucial steps in managing and mitigating the spread of antibiotic resistance.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.