The discovery of antibiotics was one of the most significant medical achievements of the 20th century, saving many lives and contributing to the control of numerous infectious diseases. The majority of the antibiotics that are used today were discovered in the period referred to as the golden age, i.e., 1940–1960, when at least 20 clinically relevant classes were developed (Katz and Baltz, 2016; O’Neill, 2016; Hutchings et al., 2019). However, the existing classes of antibiotics act selectively against only a few microbial cell targets: cell wall, plasma membrane, synthesis of proteins and metabolites, or DNA transcription and replication, allowing the bacterial species to develop resistance mechanisms during the millions of years of coevolution. Although many international authorities advocate for incentives to encourage the development of new antibiotics,¹ (The Joint Programming Initiative on Antimicrobial Resistance, 2018) the interest of pharmaceutical companies in developing novel antibiotics has dramatically declined. Thus, in the last decades, only two new antibiotic classes (lipopeptides and oxazolidinones) providing coverage only against Gram-positive bacteria have been developed and approved, the other new molecules being analogs of the existing classes, drugs repurposed for the treatment of infectious diseases, non-antibiotics or antibiotic adjuvants (Xie et al., 2022).

The natural ability of bacteria to develop resistance has been accelerated by the selective pressure exerted by the improper use of antibiotics in human therapy, animal husbandry, and agriculture. Between 2000 and 2010, global antibiotic use increased by 36%, reaching 45% in the case of carbapenems, and the amount of all antibiotics used in animal food was estimated to be 200,235 tons in 2030 (Dhingra et al., 2020; Boyd et al., 2021). The acute limitation of currently available therapeutic options leads to increased morbidity and mortality rates, longer treatment duration, and higher hospitalization costs, questioning the efficacy of modern medical practices, which will become very risky because of common infections. Worldwide, it is estimated that 4.95 million deaths were associated with and 1.2 million attributable to bacterial AMR in 2019, mainly from lower respiratory infections and involving six leading pathogens (E. coli, S. aureus, K. pneumoniae, Streptococcus pneumoniae, A. baumannii, and P. aeruginosa) (Murray et al., 2022).

Clinical surveillance programs indicate that the prevalence of human pathogens exhibiting multidrug (MDR), extended drug (XDR), or pan-drug (PDR) resistance is rising to dangerously high levels in all parts of the world (Cheesman et al., 2017). The most threatening resistant pathogens are known under different acronyms, such as ESKAPE (E) [Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, Enterobacter sp. (Escherichia coli), ESCAPE or AmpC-producing SPICE (Serratia, Providencia, indole-positive Proteus, Morganella, Providencia species/Acinetobacter, Citrobacter, Enterobacter species) or are included in different “black” lists (Rice, 2008; Moy and Sharma, 2017)]. In addition, the WHO published the critical/high/medium priority pathogens list for R&D of new antibiotics, including carbapenem-resistant A. baumannii and P. aeruginosa, carbapenem-resistant and extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae, clarithromycin-resistant Helicobacter pylori, fluoroquinolone-resistant Campylobacter spp., Salmonella and Shigella, cephalosporin-resistant, fluoroquinolone-resistant Neisseria gonorrhoeae, ampicillin-resistant Haemophilus influenzae, vancomycin-resistant E. faecium, methicillin-resistant, vancomycin-intermediate, and resistant S. aureus, and penicillin-non-susceptible Streptococcus pneumoniae (WHO, 2017; Asokan et al., 2019). Furthermore, with the increasing use of whole-genome sequencing to analyze antibiotic-resistant pathogens, it has become clear that many of the significant resistance problems are associated with a few successful bacterial clones within a species, with worldwide dissemination in hospitals and, possibly, in the community and the natural environment (Oliver et al., 2015; Andersson and Hughes, 2017; MacLean and San Millan, 2019).

Numerous scientific publications reported the presence in the natural environment of clinically relevant antibiotic-resistant bacteria (ARB) and antibiotic-resistance genes (ARG) (Nnadozie and Odume, 2019; Pazda et al., 2019). Thus, the environment can serve as a source/reservoir of already resistant pathogens or may acquire ARGs from other human / animal-associated bacteria or the environmental resistome. Further, these ARGs could be transferred to the clinic and vice versa (Figure 1).

The potential gene exchange between environmental and clinical resistomes through horizontal gene transfer (HGT) involves mobile genetic elements (MGE) such as plasmids, integrative and conjugative elements (ICE), transposons, and integrons (Singer et al., 2016; Bengtsson-Palme et al., 2018; Larsson and Flach, 2022). The HGT occurs mainly via three mechanisms: conjugation (transfer of DNA between bacteria by direct cell-to-cell contact, considered the main AMR dissemination strategy), natural transformation (uptake of the free DNA from the environment), and transduction (transfer of DNA between bacteria via a bacteriophage) (von Wintersdorff et al., 2016; Chen et al., 2018). The MGE, such as integrons, are frequently associated with multi-drug resistance (MDR), increasing the risk of co-selection and persistence of multiple resistance determinants under the selective pressure of antimicrobial agents (Deng et al., 2015).

Many environmental ARGs have been shown to possess a high sequence similarity to those in human / domesticated animal fecal microbes (Nesme et al., 2014) and human/animal pathogens (Forsberg et al., 2012). For example, the quinolone resistance gene qnrA was first detected in clinical isolates of K. pneumoniae in the USA and then found in the aquatic Gram-negative bacteria Shewanella algae and in Aeromonas spp. isolated from the Seine River (France) (Cattoir et al., 2008; Perry and Wright, 2013). Most of the ARGs with known origins were likely mobilized from their native context by MGEs, such as IS/ISCR elements (Ebmeyer et al., 2021). However, their recent origin is still unknown for the great majority of the ARGs, probably because the original environmental bacterial hosts have not yet been sequenced (Larsson and Flach, 2022).

The “One Health” approach published in 2017 reinforced the importance of addressing the AMR issue outside the clinical sector by encompassing the environmental contribution to the emergence, accumulation, and spread of AMR (Fletcher, 2015; Baker et al., 2018; MacLean and San Millan, 2019). In order to map and understand the transmission routes of ARGs among clinical and environmental bacteria, research on the distribution patterns of ARGs and their temporal dynamics in different microbiomes and studies tracking the acquisition of ARGs in complex communities are required (Andersson and Hughes, 2017; MacLean and San Millan, 2019; Surleac et al., 2020).

There are numerous reports on the detection of wastewater effluents and impacted rivers of a wide variety of clinically significant ARB (Zhang et al., 2009; Giebułtowicz et al., 2018; Limayem et al., 2019).

This review aims to present an update regarding the contribution of WWTPs to the antibiotic resistance reservoir, focusing on ESCAPE pathogens and associated ARGs. We performed a literature search in Pubmed, Scopus, and Google Scholar databases using the combinations of the following keywords: ESCAPE pathogens (searched both as abbreviation and individually for each pathogen) + antibiotic resistance / wastewater treatment plants / antibiotic resistance genes / wastewater / high-risk clones / multidrug resistance. Consequently, a title and abstract screening was performed, and the papers published in the last 10 years, as well as some significant studies published previously, have been considered. The articles duplicated, irrelevant, and without full text were excluded. The remaining studies have been grouped into six categories, corresponding to each ESCAPE pathogen species and one last category, referring to the abundance of ARGs in WWTPs environments and the efficiency of different wastewater treatments in decreasing or removing ARGs.

Wastewater treatment plants receive a continuous discharge of human-associated ARB, ARG, antibiotics, disinfectants, and metals, which are co-locating inside the WWTP, together with the commensal microbiota, generating the perfect premises for antibiotic resistance emergence, accumulation, and dissemination (Bougnom and Piddock, 2017; Karkman et al., 2019; Kraemer et al., 2019; Paulshus et al., 2019; Pärnänen et al., 2019; Reichert et al., 2019).

Although WWTPs reduce the number of total bacteria released into the environment, the current wastewater treatment technologies do not altogether remove and/or biodegrade ARB, ARG, and antibiotics (Jäger et al., 2018; Müller et al., 2018). The sewage’ inner surfaces favor the colonization with wastewater-derived microorganisms that develop biofilms, which are continuously exposed to antibiotic residues and ARB from wastewater (Auguet et al., 2017). An analysis of the biofilm-embedded bacteria from influent and effluent sewage showed a lower susceptibility to antibiotics than bacteria from wastewater (Lépesová et al., 2018). Moreover, biofilms could facilitate the HGT, as shown by an in vitro study, demonstrating an increased transfer rate of the plasmidial CTX-M-15 gene in a K. pneumoniae biofilm compared to planktonic growth conditions (Hennequin et al., 2012). The transfer of conjugative transposons carrying ARG, such as Tn916, might also be responsible for acquiring resistance mechanisms in biofilm bacteria (Hannan et al., 2010). Microfluidic systems combined with laser confocal microscopy, fluorescent labeling of bacteria, and flow cytometry have been developed to investigate the HGT of ARGs in activated sludge biofilm (Qiu et al., 2018). WWTP discharges might also impact the river-streambed biofilm communities that often prevail in natural environments compared to free, planktonic bacteria (Proia et al., 2016). The released WWTPs effluents could impact the river biofilms through the fixation of pharmaceuticals and the persistence of the discharged microorganisms in the streambed biofilms (Aubertheau et al., 2017; Subirats et al., 2017). It has been shown that some ESCAPE pathogens (enterococci, A. baumannii, and P. aeruginosa), as well as associated ARGs (ermB, blaTEM, tetM, and qnrS), are associated with the particulate fraction of the WWTP effluent and are more probably remaining in the riverbed of the receiving water due to sedimentation (Brown et al., 2019). However, little information is available regarding the ARG distribution in the river streambed and how wastewater-associated microorganisms, including potential pathogens, contribute to maintaining the streambed resistome.

Wastewater systems, especially those receiving hospital or livestock wastewater, were demonstrated to be significant sources of significant epidemic pathogens belonging to the so-called high-risk clones: E. coli sequence type (ST) 131 (Finn et al., 2020), E. faecium HiRECC (Sadowy and Luczkiewicz, 2014), Acinetobacter IC2 carrying OXA-23, and IC1 carrying OXA-72 (Goic-Barisic et al., 2017; Higgins et al., 2018), K. pneumoniae ST11, and ST258 (Surleac et al., 2020), and P. aeruginosa ST235, ST111, and ST175. Furthermore, increased correlations of ESCAPE pathogens with clinically relevant ARGs (e.g., blaNDM-1, vanA) were found in WWTP effluents influenced by hospital wastewaters. Furthermore, isolates carrying the mcr-1 gene, providing resistance to the last resort antibiotic colistin, have been reported in WWTP sewage, probably originating from food-producing animals (Ovejero et al., 2017). Additionally, the selective pressure within the wastewater treatment plants (WWTP)/sewer systems, an ecosystem with consistent exposure to antibiotics and other chemical pollutants such as biocides, heavy metals, etc., could contribute to the emergence of new antibiotic-resistant variants, with possible epidemic risk (Flach et al., 2018; Danner et al., 2019; Kraupner et al., 2021). In a study aimed to assess the presence of ESKAPE strains in process water from delivery and unclean areas as well as wastewater from the in-house WWTPs of German poultry slaughterhouses, at least one of the target species was detected in 87.5% of the wastewater samples and 86% of the process water samples, with the following decreasing order of their prevalence: E. coli > A. calcoaceticus-A. baumannii (ACB) complex > S. aureus > K. pneumoniae > Enterobacter spp., Enterococcus spp., and P. aeruginosa (Savin et al., 2020).

In the next subchapters we will present an update of the current knowledge on the ESCAPE pathogens to WWTP relationships.

ESCAPE-related ARGs, encoding every known type of mechanism (target protection, target modification, drug modification, reduced permeability or efflux), have been widely reported in both wastewater and downstream environments across continents (Freeman et al., 2018; Pazda et al., 2019; Surleac et al., 2020; Popa et al., 2021; Gheorghe-Barbu et al., 2022). These ARGs have been detected in bacterial isolates or total genomic DNA samples (Table 1).

Dutch WWTPs were shown to discharge on average 106 copies of ARGs per liter of effluent to the receiving water bodies, the dominant ones being the ARGs to sulphonamides (sul1) and macrolides (ermB), their high occurrence being possibly associated with MGEs or might be due to the prolonged use of these antibiotics in humans and animals (Pallares-Vega et al., 2019). Some ARGs, like the resistance to sulfonamides (sul1, sul2), glycopeptides (vanA), and tetracycline (tetM), were shown to be more abundant in treated wastewater, a fact that might also be explained by a long history of usage of sulphonamide and tetracycline antibiotics (Heuer et al., 2011; Caucci et al., 2016; Pärnänen et al., 2019). Several other ARGs conferring resistance to different classes of antibiotics, including macrolide-lincosamide-streptogramin B (MLSB) (ermA, ermB), beta-lactams (blaSHV-02, blaVEB, blaOXA-10, blaCTX-M-32, blaTEM, and blaCMY-2), glycopeptides (vanA), aminoglycosides (aadA, aadA1, and aadA2) and also multidrug resistance (qacEdelta1) were significantly enriched in effluent wastewater from Chinese and European WWTPs (Hembach et al., 2017). It has been hypothesized that a positive selection or community change during wastewater treatment could enrich vanA (Pärnänen et al., 2019). Worrisome is the detection of the colistin resistance gene mcr-1 in influent samples of seven WWTPs investigated in Germany and its release into the aquatic environment (Hembach et al., 2017).

The absolute abundance of nine ARGs and a class 1 integron-associated integrase gene in the wastewater effluents of 16 WWTPs from ten different European countries could be ranked along the following order: int1 > sul1 > tetM > bla_OXA–58 > bla_TEM > bla_OXA–48 > bla_CTX–M–32 > mcr-1 > bla_CTX–M–15 > bla_KPC–3 (Cacace et al., 2019). Downstream of the plants, these ARGs were quantifiable in many cases, suggesting an impact of the effluent on the receiving waterbody (Cacace et al., 2019). The aph(III)a, bla_OXA, ermB, ermF, sul1, and tetM resistance genes had a wide-spread occurrence in effluents discharged from 12 WWTPs in the Danube River Basin (Romania, Serbia, Hungary, Slovenia, Croatia, Slovakia, Czechia, Austria, Germany); in this study, the highest absolute concentrations (gene copies/mL) were observed for four ARGs (aph(III)a, tetM, vanA, and mecA) at the WWTP Bucharest (Romania), four ARGs (bla_OXA, bla_SHV, qnrS, and tetB) at the WWTP Sabac (Serbia), two ARGs (ermB, sul1) at the WWTP Varazdin (Croatia) and one ARG (ermF) at the WWTP Brno (Czechia) (Alygizakis et al., 2019).

The analysis of the abundance of resistance and virulence genes in Staphylococcus spp. strains from untreated and treated wastewater, an activated sludge (AS) bioreactor, and from surface water collected upstream and downstream the WWTP, revealed that S. aureus was present in 63% of the samples, with 20% of the strains carrying the vanA gene. The hla virulence gene was present in most of the isolates (80%), followed by the PVL gene (Zieliński et al., 2020). The highest abundance of virulence and resistance genes among S. aureus strains was observed in the untreated wastewater, gradually decreasing in treated water and downstream river water, probably due to the synergic effect of wastewater treatment, dilution, as well as influence location-specific abiotic and biotic factors, such as temperature, pH, solar radiation, mineral and organic content, microbial community structure, seasonal influence, competition, temporal variations in the prescription level of antibiotics, etc. (Caucci et al., 2016; Auguet et al., 2017; Zieliński et al., 2020).

The analysis of the water metagenomes sampled immediately downstream, and at 6.4 km from a WWTP effluent released into the surface water of a small river located in the rural area of Knox County, Ohio, showed that the metagenome just downstream of the WWTP effluent was substantially enriched in 15 different ARGs relative to the remote location. Among these enriched ARGs were 6 ARGs commonly associated with A. baumannii, such as msrE, mphE (macrolide resistance), and tet (tetracycline resistance), that persisted 6.4 km downriver. In addition, the samples from directly downstream of the WWTP were enriched in Acinetobacter spp. and gut-associated taxa: Bacteroides and Firmicutes. The ARG levels, taxa composition, and prevalence were independent of the seasonal effluent chlorination and nitrogen and phosphorus concentrations (Murphy et al., 2021).

Using a hiseq-based metagenomic sequencing approach to investigate the core resistome of urban WWTPs persisting through wastewater treatment processes, the influents and effluents of 12 urban WWTPs in Republic of Korea were analyzed. The abundance of some core ARGs such as macrolide-lincosamide-streptogramin–and tetracycline-resistance genes was higher in the influent samples, while others such as sul1, APH(3″)-lb, and RbpA were more abundant in the effluent. Most core ARGs were carried by ESCAPE pathogens, such as A. baumannii, E. cloacae, and P. aeruginosa. Also, the abundance of core ARGs in the effluent was correlated with an abundance of phages, suggesting a role of transduction in disseminating ARGs from WWTP effluent to surface water (Raza et al., 2022).

Generally, conventional WWTP treatments can reduce ARGs’ overall richness, concentration, and relative abundance in variable proportions (Yang et al., 2014; Pallares-Vega et al., 2019; Li et al., 2022). The primary treatment is usually not efficient in removing pathogens and ARB, requiring the use of advanced treatments (e.g., sand and membrane filtration) or disinfection methodologies (e.g., chlorination, UV radiation, and ozonation) after conventional activated sludge treatment (Hazra and Durso, 2022). Many WWTPs only use conventional activated sludge for wastewater treatment, which, however, has been shown to have better removed fecal coliforms, enterococci, and E. coli in comparison to septic tank systems; but, despite this reduction, an increase in the MDR profiles has been observed during wastewater treatment (Barrios-Hernández et al., 2019; Hazra and Durso, 2022). From an analysis of the influent concentrations and the removal of ARGs in a large number of WWTPs in the Netherlands, the researchers concluded that WWTP design parameters such as size, presence of primary clarification, type of phosphorus removal, and operational parameters such as hydraulic retention time, sludge retention time, anaerobic contact time and effluent total suspended solids did not affect the removal of the studied ARGs and MGEs significantly (Pallares-Vega et al., 2019). Disinfection steps, i.e., chlorination (Mao et al., 2015; Yuan et al., 2015), UV irradiation (Sousa et al., 2017), and ozone (Alexander et al., 2016), alone or in combination (UV and chlorination) (Zhang et al., 2015), were applied in WWTPs to assess their efficiency in removing ARB and ARGs. The results showed that they were not able to altogether remove the ARGs. Moreover, some of the cleaning methods could increase the proportion of ARB through the alteration of bacterial cell permeability by the activation of efflux pumps, accumulation of intermediate disinfection by-products, and promotion of HGT, particularly in case of prolonged contact with low chlorine concentrations (Huang et al., 2011). Chlorination could result in the survival of the ARB in more significant percentages due to their different susceptibility to disinfection (Mao et al., 2015). The ozone treatment causes DNA damage but has a reduced effect on guanine-rich DNA (Alexander et al., 2016). UV light is partially absorbed by organic and inorganic matter present in wastewater; thus, the DNA damage depends on the UV dosage; additionally, different sensitivity of ARB to UV radiation was noticed (Hazra and Durso, 2022). A membrane bioreactor has been shown to increase the municipal WWTP efficiency for decreasing the abundance of ARGs and loads of E. coli, enterococci, and P. aeruginosa strains (Ng et al., 2019). Still, the ARGs of 12 antibiotic classes, including aminoglycosides, beta-lactams, macrolides-lincosamides-streptogramins, polymyxins, quinolones, rifamycin, tetracyclines, and multidrug resistance, persisted during treatment. Constructed wetlands (CWs) seem to be a sustainable “green” wastewater treatment option that can be used as an alternative to WWTP or as a component of conventional WWTP, having the ability to remove antibiotics, ARB, and ARGs (Dires et al., 2018; de Oliveira et al., 2019; Riaz et al., 2020; Hazra and Durso, 2022). CWs can be used alone or even after primary treatment, contributing to 2–4 log pathogens reduction (Sleytr et al., 2007; Karimi et al., 2014) and increasing the performance of WWTP when coupled as secondary/tertiary units to primary treatment (Chen et al., 2014).

A recent comprehensive review on the performance of primary, secondary and tertiary treatments of wastewater in removing antibiotics, bacterial pathogens, ARB and ARGs concludes that combinations of different methods are more efficient in the elimination of a large number of ARB (Hazra and Durso, 2022).

In this review, we have focused on the relationships between ESCAPE pathogens and the WWTPs. Different studies performed in different geographical areas report that ESCAPE pathogens and their associated ARGs, including high-risk clones and resistance determinants to last resort antibiotics such as carbapenems and colistin, as well as MDR platforms are present in wastewater. The sources for the enrichment of ESCAPE pathogens are diverse, including clinical settings, animal husbandry facilities, urban and municipal wastewater systems, etc. Some studies are demonstrating the clonal relationships and dissemination of ESCAPE clinical strains, such as A. baumannii, P. aeruginosa and K. pneumoniae into the wastewater via hospital effluents. Also, WWTPs seem to provide a suitable environment for the enrichment of resistance and virulence determinants in ESCAPE bacteria, such as enterococci and S. aureus. Therefore, the efficiency of different wastewater treatment processes and combinations thereof regarding the removal of clinically relevant ARB species and ARGs, as well as the influence of water quality factors on their performance should be explored and monitored. Also, the development of more effective disinfection and treatment methods could contribute to prevent or reduce the release of antibiotics, ARB and ARGs into the natural environment via WWTP effluents. In order to achieve these goals, research is needed to select the most appropriate indicators (ESCAPE bacteria and/or ARGs) and standard methods for assessing them, to facilitate the comparison of resistance levels between countries and different WWTPs (Uluseker et al., 2021). The quantification of known ARGs along the wastewater treatment chain need to be complemented with information regarding the origin of these genes, their genetic context, the type and level of selective pressure, potential for HGT and compatibility with putative pathogenic hosts (Bengtsson-Palme and Larsson, 2015). Also, the possible use of current microbiological indicator species such as enterococci and E. coli in managing the risks associated with the environmental resistome, depending on the total counts as well as the resistance level in these strains, needs to be explored in the future. This knowledge will allow to establish quality threshold limits regarding the AR for point sources and effluents that are being directly discharged to water bodies and agricultural land for irrigation, and ultimately to reinforce the barrier role of WWTPs against environmental and public health threats.

LGM and MCC conceived the review. LGM, MP, IG-B, ICB, GPG, and COV wrote the draft and drown the figure. DR-M, FB, HB, C-FF, MAK, BS, LW, DGJL, DN, KR, AMRH, AW, and HS completed and corrected the drafts. MCC integrated all the comments from the authors and assembled the final form of the review. All authors contributed to the article and approved the submitted version.