Escherichia coli is simultaneously an indicator of water contamination and a human pathogen. This study aimed to characterize the virulence and resistance of E. coli from municipal and hospital wastewater treatment plants (WWTPs) in central Portugal. From a total of 193 isolates showing reduced susceptibility to cefotaxime and/or nalidixic acid, 20 E. coli with genetically distinct fingerprint profiles were selected and characterized. Resistance to antimicrobials was determined using the disc diffusion method. Extended spectrum β-lactamase and plasmid-mediated quinolone resistance genes, phylogroups, pathogenicity islands (PAIs) and virulence genes were screened by polymerase chain reaction (PCR). CTX-M producers were typed by multilocus sequence typing. Resistance to beta-lactams was associated with the presence of blaTEM,blaSHV, blaCTX-M-15 and blaCTX-M-32. Plasmid-mediated quinolone resistance was associated with qnrA, qnrS and aac(6′)-Ib-cr. Aminoglycoside resistance and multidrug-resistant phenotypes were also detected. PAI IV536, PAI IICFT073, PAI II536 and PAI ICFT073, and uropathogenic genes iutA, papAH and sfa/foc were detected. With regard to the clinical ST131 clone, it carried blaCTX-M-15, blaTEM-type, qnrS and aac(6′)-lb-cr; IncF and IncP plasmids, and virulence factors PAI IV536, PAI ICFT073, PAI IICFT073, iutA, sfa/foc and papAH were identified in the effluent of a hospital plant. WWTPs contribute to the dissemination of virulent and resistant bacteria in water ecosystems, constituting an environmental and public health risk.
INTRODUCTION
Escherichia coli is simultaneously a biological indicator of water treatment safety and an important human pathogen responsible for several diseases (Edberg et al. 2000; Kaper et al. 2004). E. coli presents several virulence and antimicrobial resistance genes which contribute to its success as a human pathogen (Pitout 2012). These genes may be disseminated by mobile genetic elements such as pathogenicity islands (PAIs), carriers of virulence factors, or plasmids with genes coding for both resistance and virulence determinants (Hacker et al. 1997; Carattoli 2009). Water constitutes a good matrix for the lateral transfer of mobile genetic elements (Taylor et al. 2011), which are responsible for the dissemination of virulence or resistance traits between bacteria from different sources, contributing to the modification of the natural bacterial ecosystems (Baquero et al. 2008).
Currently, an inverse relationship between antimicrobial resistance and virulence has been the consensus (Moreno et al. 2006). However, recently it has been shown that these two features may co-exist in the same genotype perpetuating the bacterial lineage and highlighting concern because of its dissemination (Dolejska et al. 2011; Colomer-Lluch et al. 2013).
Wastewater treatment plants (WWTPs) are designed to significantly reduce the biological contamination of water. Nevertheless, studies report resistant bacteria in effluents of treated water, and suggest that the conditions in WWTPs favour the proliferation of antibiotic-resistant bacteria and the exchange of genetic elements (Moura et al. 2007; Dolejska et al. 2011; Korzeniewska et al. 2013). The emergence and dissemination of antimicrobial-resistant bacteria has led to increasing concerns about potential environmental and public health risks. Moreover, the carriage of specific virulence genes, especially those located in mobile genetic elements, are important to evaluate the public health risks.
The main objectives of this study were to characterize the virulence and antimicrobial resistance profiles of E. coli collected in waters from municipal and hospital WWTPs from central Portugal and to screen for the presence of mobile genetic elements.
MATERIALS AND METHODS
Bacterial isolates
Between April and May 2011, water samples were collected from four hospitals and three municipal WWTPs located in the central region of Portugal:
University hospital: reference hospital for the central region of Portugal. A large hospital with 1,456 beds, with an extended set of medical specialties and clinical services, as well as a centre of research, serving a population of approximately 430,000 inhabitants.
General hospital: medium-sized hospital with 13 main wards and 350 beds. It serves a population of approximately 369,000 inhabitants.
Pediatric hospital: small reference hospital in central Portugal that supports paediatric units. It is composed of nine main wards and 110 beds serving a population of about 90,000 inhabitants.
Maternity: small hospital with 96 beds and three main wards – gynaecology, obstetrics and neonatology, not including the baby unit. It serves a population of approximately 507,000 women.
Municipal WWTP1: serves a 14,000 population equivalent.
Municipal WWTP2: serves a 213,000 population equivalent. It receives urban wastewaters that include domestic wastewaters and hospital effluents (namely from the four mentioned hospitals).
Municipal WWTP3: serves a 1,500 population equivalent.
Municipal WWTP sampling was performed at the entrance and exit of the station on two occasions and hospital samples were collected on three different dates at the exit of the station. Wastewater samples (250 mL) were collected in amber glass bottles and further vacuum filtered through 1.0 μm glass microfibre filters (GF/C, Whatman, UK), followed by 0.45 μm nylon membrane filters (Whatman, UK). The filters were placed in MacConkey Agar supplemented with 0.5 mg/L of cefotaxime or 10 mg/L of nalidixic acid. A bacterial suspension was prepared with the inoculum and cultured in MacConkey Agar. A maximum of eight presumptive colonies of E. coli per plate were further cultured in Eosin Methylene Blue Agar, and lactose fermenter colonies with a green metallic sheen were selected. The citrate test was used to distinguish E. coli from Citrobacter spp. The identification was confirmed using a polymerase chain reaction (PCR)-based technique with specific primers set targeting uidA gene (Heijnen & Medema 2006).
The genetic relationship was evaluated by BOX-PCR (Versalovic et al. 1994), and only non-duplicate isolates were further analysed.
Susceptibility testing and phenotypic extended spectrum β-lactamase detection
The antimicrobial susceptibility profiles for ampicillin (10 μg), amoxicillin-clavulanic acid (20/10 μg), cefoxitin (30 μg), cefotaxime (30 μg), ceftazidime (30 μg), nalidixic acid (30 μg), ciprofloxacin (10 μg) and gentamicin (10 μg) were determined using a disc diffusion test. Extended spectrum β-lactamase producers were detected with the double disc synergy test (Jarlier et al. 1988). The methods were performed and the results were interpreted based on the Clinical and Laboratory Standards Institute guidelines (CLSI 2010). Multidrug resistance was defined as acquired non-susceptibility to at least one agent in three or more antimicrobial categories (Magiorakos et al. 2012).
Antimicrobial resistance determinants detection
The blaCTX-M, blaTEM and blaSHV genes coding for β-lactamases and plasmid-mediated quinolone resistance (PMQR) determinants qnrA, B and S, and qepA were screened with specific primers by PCR (Cattoir et al. 2007; Mendonça et al. 2007; Ma et al. 2009). For the samples with a positive result for the screening of blaCTX-M, the full gene was further amplified using previously described primers (Conceição et al. 2005) and amplicons were purified with ExoSAP-IT (Affymetrix, USB products). The whole genes were sequenced at Macrogen, Amsterdam, The Netherlands.
aac(6′)-Ib was screened by PCR and isolates positive for the aac(6′)-lb gene were further digested with BtsCI enzyme (New England Biolabs) to identify aac(6′)-lb-cr which lacks the BtsCI restriction site present in the wild-type gene (Park et al. 2006).
Plasmid replicon typing
Plasmid replicon identification was performed according to the PCR-based replicon typing scheme (Carattoli et al. 2005), detecting the main replicon families in Enterobacteriaceae.
Detection of PAIs and other virulence markers
PAI markers were screened according to the Bronowski et al. (2008) scheme, based on the technique first described by Sabaté et al. (2006). This method allows the detection of eight PAIs, encoding several virulence determinants: PAI I536, PAI II536, PAI III536, PAI IV536, PAI IJ96, PAI IIJ96, PAI ICFT073, and finally PAI IICFT073 (Sabaté et al. 2006).
Other virulence genes that may be present in extraintestinal E. coli (EXPEC) such as papAH, papC (P fimbriae structural subunit and assembly), sfa/foc (S and F1C fimbriae), afa/dra (Dr-binding adhesins), iutA (aerobactin receptor), kpsM II (group 2 capsules) and cnf1 (cytotoxic necrotizing factor 1) were screened by PCR (Johnson & Stell 2000), as well as the enterohaemorrhagic E. coli associated virulence genes eaeA (intimin), hlyA (pore-forming cytolysin), stx 1 and 2 (shiga-like toxins) (Ram et al. 2008).
Phylogenetic analysis
The determination of E. coli major phylogroups (A, B1, B2 and D) was performed with a PCR-multiplex detecting chuA, yjaA and DNA fragment tspE4.C2 genes (Clermont et al. 2000; Mendonça et al. 2011).
Multilocus sequence typing (MLST)
MLST of the CTX-M producers was performed based on the PCR amplification and sequencing of seven housekeeping genes, adk, fumC, gyrB, icd, mdh, purA and recA, according to the University College of Cork (Cork, Ireland) scheme for E. coli(http://mlst.ucc.ie/mlst/dbs/Ecoli).
RESULTS
Bacterial isolates
A total of 193 presumed E. coli with reduced susceptibility to cefotaxime and/or nalidixic acid were obtained from WWTPs. The majority of the isolates showed an identical genetic profile and only 20 isolates with distinct profiles were selected (non-duplicate isolates) and further characterized for resistance and virulence profiles (Table 1). Fourteen of the non-duplicate isolates were from municipal WWTPs, while the remaining six were recovered from hospital water samples. The municipal isolates were recovered from WWTP2 (n = 7), followed by WWTP3 (n = 4) and WWTP1 (n = 3). Isolates W4 and W12 were detected in both the influent and effluent of the respective WWTPs, and in addition W12 isolate was detected on two different sampling occasions. From the hospital WWTPs, three strains were recovered from the general hospital, two from the maternity hospital and one from the university hospital. E. coli isolates with reduced susceptibility to CTX or NAL were not detected in the outflow of the paediatric hospital.
Distribution of strains and characterization of phylogeny, virulence and resistance determinants and plasmid incompatibility groups
WWTP . | Strain . | Collection date (day/month) . | Sampling . | Phylogroup . | Virulence determinants . | Resistance profile . | Plasmidic resistance determinants . | Replicon type . | ST . |
---|---|---|---|---|---|---|---|---|---|
Hospital WWTP | |||||||||
Maternity | W1 | 19/4 | Outflow | A | PAI IV536,iutA | NAL | – | F, FIA, FIB,K, I1/Iϒ, P | ND |
W2 | 9/5 | Outflow | D | iutA | AMP, NAL | blaTEM | I1/Iϒ, P | ND | |
University | W3 | 18/4 | Outflow | B2 | PAI IV536, PAI ICFT073, PAI IICFT073,iutA, sfa/foc, papAH | AMP,CAZ, CTX, CN, NAL, CIP | blaCTX-M-15, blaTEM, qnrS, aac(6′)-lb-cr | F, FIB,P | ST131 |
General | W16 | 19/4 | Outflow | D | iutA | AMP, FOX, CAZ,CTX, AMC,GEN, NAL, CIP | blaTEM, qnrA | F | ND |
W17 | 19/4 | Outflow | A | – | AMP, FOX, CTX, NAL, CIP | blaTEM | – | ND | |
W18 | 2/5 | Outflow | A | – | AMP, FOX, CAZ,CTX, AMC, NAL | blaTEM | – | ND | |
Municipal WWTP | ND | ||||||||
WWTP1 | W4 | 19/4 | Inflow/Outflow | D | PAI IV536 | – | – | – | ND |
W5 | 19/4 | Outflow | B1 | – | – | – | – | ND | |
W6 | 19/4 | Outflow | B2 | PAI II536, PAI IV536, PAI IICFT073,iutA | AMP, AMC | blaTEM | F, FIA, I1/Iϒ | ND | |
WWTP2 | W7 | 19/4 | Inflow | B2 | PAI IV536, PAI IICFT073,iutA, papAH | AMP, GEN, NAL, CIP | blaTEM, qnrS | F, FIA | ND |
W8 | 19/4 | Outflow | A | – | AMP, CAZ, CTX, NAL, CIP | blaSHV | I1/Iϒ | ND | |
W9 | 19/4 | Outflow | B1 | PAI IV536,iutA | AMP, FOX, CAZ,CTX, AMC, NAL,CIP | qnrA | F, FIA, FIB, K, I1/Iϒ,P | ND | |
W10 | 19/4 | Outflow | B1 | PAI IV536,iutA | NAL | – | F,K | ND | |
W11 | 19/4 | Outflow | A | PAIIV536,iutA | FOX, AMC, NAL | – | F | ND | |
W12 | 19/4; 5/5 | Inflow/Outflow | B2 | PAI IV536, PAI ICFT073, PAI IICFT073, iutA, eaeA | NAL | – | F | ND | |
W13 | 5/5 | Inflow | D | PAI IV536,iutA | FOX, NAL | – | P | ND | |
WWTP3 | W14 | 19/4 | Outflow | A | iutA | NAL | – | – | ND |
W15 | 5/5 | Inflow | A | PAI IV536 | AMP, CTX, NAL, | blaCTX-M-32 | F,K | ST34 | |
W19 | 5/5 | Inflow | B2 | PAI IV536,,iutA | FOX, NAL | – | F, FIB | ND | |
W20 | 5/5 | Inflow | B1 | PAI IV536 | NAL | qnrA | K | ND |
WWTP . | Strain . | Collection date (day/month) . | Sampling . | Phylogroup . | Virulence determinants . | Resistance profile . | Plasmidic resistance determinants . | Replicon type . | ST . |
---|---|---|---|---|---|---|---|---|---|
Hospital WWTP | |||||||||
Maternity | W1 | 19/4 | Outflow | A | PAI IV536,iutA | NAL | – | F, FIA, FIB,K, I1/Iϒ, P | ND |
W2 | 9/5 | Outflow | D | iutA | AMP, NAL | blaTEM | I1/Iϒ, P | ND | |
University | W3 | 18/4 | Outflow | B2 | PAI IV536, PAI ICFT073, PAI IICFT073,iutA, sfa/foc, papAH | AMP,CAZ, CTX, CN, NAL, CIP | blaCTX-M-15, blaTEM, qnrS, aac(6′)-lb-cr | F, FIB,P | ST131 |
General | W16 | 19/4 | Outflow | D | iutA | AMP, FOX, CAZ,CTX, AMC,GEN, NAL, CIP | blaTEM, qnrA | F | ND |
W17 | 19/4 | Outflow | A | – | AMP, FOX, CTX, NAL, CIP | blaTEM | – | ND | |
W18 | 2/5 | Outflow | A | – | AMP, FOX, CAZ,CTX, AMC, NAL | blaTEM | – | ND | |
Municipal WWTP | ND | ||||||||
WWTP1 | W4 | 19/4 | Inflow/Outflow | D | PAI IV536 | – | – | – | ND |
W5 | 19/4 | Outflow | B1 | – | – | – | – | ND | |
W6 | 19/4 | Outflow | B2 | PAI II536, PAI IV536, PAI IICFT073,iutA | AMP, AMC | blaTEM | F, FIA, I1/Iϒ | ND | |
WWTP2 | W7 | 19/4 | Inflow | B2 | PAI IV536, PAI IICFT073,iutA, papAH | AMP, GEN, NAL, CIP | blaTEM, qnrS | F, FIA | ND |
W8 | 19/4 | Outflow | A | – | AMP, CAZ, CTX, NAL, CIP | blaSHV | I1/Iϒ | ND | |
W9 | 19/4 | Outflow | B1 | PAI IV536,iutA | AMP, FOX, CAZ,CTX, AMC, NAL,CIP | qnrA | F, FIA, FIB, K, I1/Iϒ,P | ND | |
W10 | 19/4 | Outflow | B1 | PAI IV536,iutA | NAL | – | F,K | ND | |
W11 | 19/4 | Outflow | A | PAIIV536,iutA | FOX, AMC, NAL | – | F | ND | |
W12 | 19/4; 5/5 | Inflow/Outflow | B2 | PAI IV536, PAI ICFT073, PAI IICFT073, iutA, eaeA | NAL | – | F | ND | |
W13 | 5/5 | Inflow | D | PAI IV536,iutA | FOX, NAL | – | P | ND | |
WWTP3 | W14 | 19/4 | Outflow | A | iutA | NAL | – | – | ND |
W15 | 5/5 | Inflow | A | PAI IV536 | AMP, CTX, NAL, | blaCTX-M-32 | F,K | ST34 | |
W19 | 5/5 | Inflow | B2 | PAI IV536,,iutA | FOX, NAL | – | F, FIB | ND | |
W20 | 5/5 | Inflow | B1 | PAI IV536 | NAL | qnrA | K | ND |
AMP, ampicillin; CAZ, ceftazidime; FOX, cefoxitin, CTX, cefotaxime; GEN, gentamicin; CIP, ciprofloxacin; NAL, nalidixic acid; AMC, amoxicillin-clavulanic acid.
ND, not determined.
Resistance profile characterization
The majority of the isolates were resistant to nalidixic acid (85%), followed by resistance to ampicillin (50%), amoxicillin-clavulanic acid (35%), cefoxitin (35%), cefotaxime (35%), ciprofloxacin (30%), ceftazidime (25%) and gentamicin (15%). Strains W4 and W5 from municipal WWTP1 were susceptible to all the antibiotics tested, and strains W3 from the university hospital, strain W16 from the general hospital and strain W7 from the municipal WWTP2 were multidrug resistant. Among the antimicrobial resistance determinants screened, blaTEM was detected most (n = 7) followed by qnrA (n = 3), qnrS (n = 2) and blaCTX-M (n = 2). The studied resistance determinants were not detected in nine isolates.
Only two strains carried blaCTX-M genes: blaCTX-M-15 (W3) collected from the university hospital outflow and blaCTX-M-32 (W15) from the municipal WWTP3 inflow water. Isolate W3 was assigned by MLST to ST131 and isolate W15 to ST34. The strain W3 ST131 was multidrug resistant and showed the higher diversity of plasmidic determinants, carrying blaCTX-M-15, blaTEM-type, qnrS and aac(6′)-lb-cr.
The main plasmid groups detected in Enterobacteriaceae family members were also investigated. Four plasmid groups: IncF, IncK, IncI1/Iγ and IncP were detected. IncF was the most prevalent group (n = 11) found in both hospital and municipal WWTPs waters, while in 25% of the isolates no plasmid was identified.
Virulence profile description
The PAIs most frequently detected was PAI IV536 (n = 13) followed by PAI IICFT073 (n = 5), PAI ICFT073 (n = 2) and PAI II536 (n = 1). PAI I536, PAI III536, PAI IJ96 and PAI IIJ96 were not detected. PAI IV536 and PAI IICFT073 were more prevalent in municipal isolates. PAI II536 was exclusively detected in a strain from a municipal WWTP. Different combinations of PAIs were identified (Table 1).
Considering individual virulence genes, the most frequently detected was iutA (n = 13), followed by papAH (n = 2); more common among hospital isolates, sfa/foc and eaeA were less prevalent, each of them being detected in one isolate, the former found in a hospital source and eaeA detected in a municipal WWTP. The genes afa/dra, kpsM II, cnf, hlyA, stx1 and 2 were not detected. The most prevalent phylogenetic group was group A (n = 7) followed by B2 (n = 5), and finally B1 and D (n = 4, each). Strains from phylogroup B2 from both municipal and hospital WWTPs carried more virulence factors, including the ST131 isolate (Table 1). All the other isolates presented virulence determinants regardless of the phylogroup, with the exception of W17 and W8 from phylogroup A, and W5 from phylogroup B1.
DISCUSSION
This study aimed to characterize the virulence and resistance profiles of E. coli selected from municipal and hospital WWTPs from a central region of Portugal, evaluating the possibility of environmental dissemination of pathogenic and/or resistant bacteria from these sources. Several studies indicate the potential dissemination of resistant and/or virulent bacteria from WWTPs into the environment (Jakobsen et al. 2008; Sabaté et al. 2008; Chagas et al. 2011; Dolejska et al. 2011; Colomer-Lluch et al. 2013; Biswal et al. 2014). Nonetheless, only one study concomitantly studied virulence factors and resistance determinants in hospital WWTPs (Jakobsen et al. 2008), and it only focused on gentamicin resistance determinants and in single virulence factors. Here, we extended the study to the identification of PAIs, clusters of virulence genes with the potential to be mobile.
E. coli strains showed resistance to important groups of antibiotics such as beta-lactams, quinolones and aminoglycosides, with multidrug resistance being detected in both municipal and hospital strains, indicating that WWTPs may be responsible for the introduction of multidrug-resistant bacteria into the environment. In the Portuguese Mondego river, where the effluents of the studied WWTPs are discharged, several types of antibiotics, including fluoroquinolones, were recently detected (Santos et al. 2013), which may exert a selective pressure in the dissemination of resistant bacteria in environmental waters (Kummerer & Henninger 2003). Plasmidic resistance determinants are important vehicles of transmission of resistance genes. Several resistance determinants were detected in this study, including blaTEM,blaSHV, blaCTX-M-15 and blaCTX-M-32, responsible for resistance to several beta-lactams, as well as PMQR genes, including qnrA, qnrS and aac(6′)-Ib-cr. Different beta-lactamase genes have already been detected, including blaCTX-M group 1, blaCTX-M group 9, blaSHV and blaTEM genes, in hospital and municipal effluents (Korzeniewska & Harnisz 2013; Korzeniewska et al. 2013). In addition, CTX-M-15 and CTX-M-32 producers have already been detected in river waters in Portugal, with unknown origin, indicating that these determinants may be spreading among water systems (Tacão et al. 2012).
Several virulence factors responsible for enhancing the pathogenic potential of bacteria have been detected in E. coli (Johnson 1991; Johnson & Stell 2000), and some of them are clustered in PAIs, mobile genetic platforms capable of dissemination through horizontal gene transfer (Hacker et al. 1997). Virulence profiles were characterized in the isolates. Results show that PAI IV536 was the most prevalent island, likewise in other studies performed in clinical samples and in water from several origins, but none of them from WWTPs (Sabaté et al. 2006; Mendonça et al. 2012). The association of PAI IV536 to virulence is controversial, as some studies indicate that this island contributes to the virulence of EXPEC (Schubert et al. 2002) but other authors suggest that this is, rather, a fitness element (Oelschlaeger et al. 2002). Several uropathogenic E. coli (UPEC) virulence genes were identified in the isolates, including iutA, involved in the uptake of iron, papAH coding for P fimbriae associated with pyelonephritis (Kallenius et al. 1981; Dowling et al. 1987) and sfa/foc encoding S fimbriae/F1C fimbriae, involved in urinary infections, neonatal sepsis as well as meningitis (Antão et al. 2009). In addition, eae, usually detected in enteropathogenic and enterohaemorrhagic E. coli were also detected in a municipal isolate. This fact may be related to the possible association of animal farms to the municipal WWTP where W12 isolate was detected, as ruminants are known to be important reservoirs of E. coli carrying intimin gene (Blanco et al. 2005). This isolate also carried other virulence determinants and was detected in both the influent and effluent of the WWTP at different collection dates indicating that WWTPs are not only inefficient concerning the elimination of virulent bacteria, but are also contributing to the dissemination of strains carrying virulence-associated genes in the environment.
The phylogenetic background of the strains was studied as an indicator of the virulence potential of the isolates. E. coli strains have been grouped into four different phylogroups (A, B1, B2 and D) according to their virulence features. Virulent extra-intestinal strains belong mainly to group B2 and, to a lesser extent, to group D, while commensal strains belong to groups A and B1 (Clermont et al. 2000). Despite group A and B1 being considered less virulent, strains from these phylogroups harbouring virulence factors were detected in both municipal and hospital isolates. This observation may indicate that even the bacteria considered less virulent may be enhancing their virulent potential, possibly due to horizontal gene transfer of virulence traits.
In this study the international clone E. coli ST131 was detected in the effluent of a hospital. The ST131 isolate carried several pathogenic factors, including PAI IV536, PAI ICFT073, PAI IICFT073, iutA, sfa/foc and papAH as well as resistance determinants blaCTX-M-15, blaTEM, qnrS and aac(6′)-lb-cr and the IncF plasmid, a conjugative plasmid that can easily be spread to other bacteria, and is known for dissemination of blaCTX-M-15 and aac(6′)-Ib-cr (Carattoli 2009; Partridge et al. 2011). ST131 displays both resistance and virulence features which contribute to the success of this international clone, which today is one of the most adapted and efficient human pathogens (Johnson et al. 2010a, b). This clone was previously detected in the effluent of a municipal WWTP in the Czech Republic (Dolejska et al. 2011) and in the influent of a WWTP in Catalonia, Spain (Colomer-Lluch et al. 2013). However, to our knowledge, this is the first finding in hospital WWTPs, highlighting the crucial need for monitoring the efficiency of hospital WWTPs.
The dissemination of bacteria carrying both resistance and virulence determinants, such as ST131, constitutes an important threat to public health and to the environment. Resistant or pathogenic isolates when in contact with autochthonous bacteria may be responsible for the dissemination of resistance and virulence determinants among natural ecosystems by horizontal gene transfer.
CONCLUSIONS
WWTPs constitute a potential mechanism of propagation of resistant and pathogenic bacteria from sewage of diverse origins, into the environment, and may thus contribute to the environmental dissemination of virulence and resistance determinants which constitute an important public health concern.
ACKNOWLEDGEMENTS
All authors contributed equally to this work. N. Mendonça was supported by a grant SFRH/BPD/45815/2008 from Fundação para a Ciência e a Tecnologia, Lisbon, Portugal. This work was supported financially by the European Society of Clinical Microbiology and Infectious Diseases 2010 Research Grant, the PEst-OE/SAU/UI0177/2014 from Fundação para a Ciência e Tecnologia, and by the Center for Pharmaceutical Studies, University of Coimbra, Portugal. No conflict of interest is declared.