Marta Tacão

Resistance to Broad-Spectrum Antibiotics in Aquatic Systems: Anthropogenic Activities Modulate the Dissemination of blaCTX-M-Like Genes

ABSTRACT We compared the resistomes within polluted and unpolluted rivers, focusing on extended-spectrum beta-lactamase (ESBL) genes, in particular bla CTX-M. Twelve rivers from a Portuguese hydrographic basin were sampled. Physicochemical and microbiological parameters of water quality were determined, and the results showed that 9 rivers were classified as unpolluted (UP) and that 3 were classified as polluted (P). Of the 225 cefotaxime-resistant strains isolated, 39 were identified as ESBL-producing strains, with 18 carrying a bla CTX-M gene (15 from P and 3 from UP rivers). Analysis of CTX-M nucleotide sequences showed that 17 isolates produced CTX-M from group 1 (CTX-M-1, -3, -15, and -32) and 1 CTX-M that belonged to group 9 (CTX-M-14). A genetic environment study revealed the presence of different genetic elements previously described for clinical strains. ISEcp1 was found in the upstream regions of all isolates examined. Culture-independent bla CTX-M-like libraries were comprised of 16 CTX-M gene variants, with 14 types in the P library and 4 types in UP library, varying from 68% to 99% similarity between them. Besides the much lower level of diversity among CTX-M-like genes from UP sites, the majority were similar to chromosomal ESBLs such as bla RAHN-1. The results demonstrate that the occurrence and diversity of bla CTX-M genes are clearly different between polluted and unpolluted lotic ecosystems; these findings favor the hypothesis that natural environments are reservoirs of resistant bacteria and resistance genes, where anthropogenic-driven selective pressures may be contributing to the persistence and dissemination of genes usually relevant in clinical environments.

Co-resistance to different classes of antibiotics among ESBL-producers from aquatic systems

In this study we investigated the co-occurrence of resistance to non-beta-lactams among cefotaxime-resistant extended-spectrum beta-lactamase (ESBL) producers (ESBL(+)) versus non-ESBL producers (ESBL(-)), from aquatic environments. Higher prevalence of resistance to tetracycline, fluoroquinolones and aminoglycosides were observed in ESBL(+). Among ESBL(+) resistant to tetracycline (n = 18), tet(A) was detected in 88.9% and tet(B) in 16.7%. Among fluoroquinolone-resistant-ESBL(+) (n = 15), aacA4-cr and qnrVC4 were identified in 26.6% and 40% strains, respectively. The qnrVC4 gene was detected for the first time in Pseudomonas sp. and Escherichia coli. Class 1 integrase genes were detected in 56.41% of ESBL(+) and in 27.67% ESBL(-). Gene cassette arrays identified conferred resistance to aminoglycosides (aadA-type genes and aacA4), trimethoprim (dfrA17), chloramphenicol (catB8), fluoroquinolones (qnrVC4) and beta-lactams (blaOXA-10). Conjugation experiments were performed with CTX-M-producers. Transconjugants showed multiresistance to 3 or more classes of antibiotics, and conjugative plasmids were assigned to IncF, IncK and IncI1 replicons. Results obtained showed that co-selection of resistance to aminoglycosides, quinolones and tetracyclines is prevalent among ESBL-producers and that these features are successfully mobilized by IncF, IncK and IncI1 conjugative plasmids. This study reinforces the importance of natural aquatic systems as reservoir of mobile genetic platforms carrying multiple resistance determinants. Moreover, to the best of our knowledge, this constitutes the first observation of IncK::CTX-M-3 in Aeromonas hydrophila and the first report of IncK plasmids in Portugal.

BOX-PCR is an Adequate Tool for Typing Aeromonas spp.

PCR-based methods of fingerprinting take advantage of the presence of repetitive sequences that are interspersed throughout the genome of diverse bacterial species. They include the repetitive extragenic palindromic (REP) sequence, the enterobacterial repetitive intergenic consensus sequence (ERIC) and the 154-bp BOX element. The combination of the three methods is used for fine discrimination of strains and is designated as rep-polymerase chain reaction (PCR). REP-PCR and ERIC-PCR have been shown to be useful for typing Aeromonas strains. To our knowledge, rep-PCR fingerprinting method using the BOXA1R primer has never been tested on aeromonads. In this study, the BOX-PCR fingerprinting technique was evaluated for the discrimination of strains of some Aeromonas species. All strains were typeable and the majority showed unique banding patterns. Four strains from culture collections were used to investigate the reproducibility of the method. According to our results, BOX-PCR fingerprinting is applicable for typing of Aeromonas strains and can be considered as a useful complementary tool for epidemiological studies of members of this genus.

Environmental Shewanella xiamenensis Strains That Carry blaOXA-48 or blaOXA-204 Genes: Additional Proof for blaOXA-48-Like Gene Origin

The chromosome-encoded beta-lactamases of Shewanella spp. have been recognized as progenitors of bla OXA-48-like genes ([1][1]). The analysis of available genome sequences of Shewanella spp. showed the presence of bla OXA-48-like genes in their chromosome with at least 80% identity to bla OXA-48 ([2

Low Prevalence of Carbapenem-Resistant Bacteria in River Water: Resistance Is Mostly Related to Intrinsic Mechanisms.

Carbapenems are last-resort antibiotics to handle serious infections caused by multiresistant bacteria. The incidence of resistance to these antibiotics has been increasing and new resistance mechanisms have emerged. The dissemination of carbapenem resistance in the environment has been overlooked. The main goal of this research was to assess the prevalence and diversity of carbapenem-resistant bacteria in riverine ecosystems. The presence of frequently reported carbapenemase-encoding genes was inspected. The proportion of imipenem-resistant bacteria was on average 2.24 CFU/ml. Imipenem-resistant strains (n=110) were identified as Pseudomonas spp., Stenotrophomonas maltophilia, Aeromonas spp., Chromobacterium haemolyticum, Shewanella xiamenensis, and members of Enterobacteriaceae. Carbapenem-resistant bacteria were highly resistant to other beta-lactams such as quinolones, aminoglycosides, chloramphenicol, tetracyclines, and sulfamethoxazole/trimethoprim. Carbapenem resistance was mostly associated with intrinsically resistant bacteria. As intrinsic resistance mechanisms, we have identified the blaCphA gene in 77.3% of Aeromonas spp., blaL1 in all S. maltophilia, and blaOXA-48-like in all S. xiamenensis. As acquired resistance mechanisms, we have detected the blaVIM-2 gene in six Pseudomonas spp. (5.45%). Integrons with gene cassettes encoding resistance to aminoglycosides (aacA and aacC genes), trimethoprim (dfrB1b), and carbapenems (blaVIM-2) were found in Pseudomonas spp. Results suggest that carbapenem resistance dissemination in riverine ecosystems is still at an early stage. Nevertheless, monitoring these aquatic compartments for the presence of resistance genes and its host organisms is essential to outline strategies to minimize resistance dissemination.

Genetic diversity and antimicrobial resistance of Escherichia coli from Tagus estuary (Portugal)

Fecal pollution of surface waters is a current world-wide public health concern and may contribute for the dissemination of antibiotic resistance. The Tagus estuary located in the south of Portugal is one of the largest wetlands in the west coast of Europe. In this study, water samples were collected from seven stations with different anthropic pressures along the estuary and evaluated for water quality indicator bacteria. Escherichia coli isolates (n=350) were typed by REP-PCR. Representatives of each REP profile (n=220) were evaluated phenotypically for resistance to 17 antibiotics and characterized in terms of phylogenetic group. Resistant isolates were screened for the presence of antibiotic resistance genes (tet(A), tet(B), sul1, sul2, qnrA, qnrB, qnrS, aacA4-cr, bla(TEM), bla(SHV), bla(CTX-M), bla(CMY-like), bla(IMP), bla(VIM)) and integrase genes (intI1 and intI2). The highest antibiotic resistance prevalence was observed for streptomycin and tetracycline followed by β-lactams and sulphonamides. Among E. coli isolates, 65.16% were resistant to at least one of the 17 antibiotics tested and approximately 19% were multiresistant. In our E. coli population phylo-groups A and D were predominant and characterized by higher prevalence of the antibiotic resistance. intI1 and intI2 genes were found in 12% of the isolates with prevalence of class 1 integrons. A strong correlation between the prevalence of integrons and multiresistance was observed. Differences in terms of antibiotic resistance between phylogenetic groups and between sampling sites were statistically significant. The results demonstrate a high prevalence of antibiotic resistance among E. coli circulating in the Tagus estuary with emphasis on the occurrence of resistance to last-resort antibiotics and on the high incidence of multiresistance.

Antibiotic and metal resistance in a ST395 Pseudomonas aeruginosa environmental isolate: A genomics approach

We analyzed the resistome of Pseudomonas aeruginosa E67, an epiphytic isolate from a metal-contaminated estuary. The aim was to identify genetic determinants of resistance to antibiotics and metals, assessing possible co-selection mechanisms. Identification was based on phylogenetic analysis and average nucleotide identity value calculation. MLST affiliated E67 to ST395, previously described as a high-risk clone. Genome analysis allowed identifying genes probably involved in resistance to antibiotics (e.g. beta-lactams, aminoglycosides and chloramphenicol) and metals (e.g. mercury and copper), consistent with resistance phenotypes. Several genes associated with efflux systems, as well as genetic determinants contributing to gene motility, were identified. Pseudomonas aeruginosa E67 possesses an arsenal of resistance determinants, probably contributing to adaptation to a polluted ecosystem. Association to mobile structures highlights the role of these platforms in multi-drug resistance. Physical links between metal and antibiotic resistance genes were not identified, suggesting a predominance of cross-resistance associated with multidrug efflux pumps.

Co-selection of antibiotic and metal(loid) resistance in gram-negative epiphytic bacteria from contaminated salt marshes.

The goal of this study was to investigate co-selection of antibiotic resistance in gram-negative epiphytic bacteria. Halimione portulacoides samples were collected from metal(loid)-contaminated and non-contaminated salt marshes. Bacterial isolates (n=137) affiliated with Vibrio, Pseudomonas, Shewanella, Comamonas, Aeromonas and with Enterobacteriaceae. Vibrio isolates were more frequent in control site while Pseudomonas was common in contaminated sites. Metal(loid) and antibiotic resistance phenotypes varied significantly according to site contamination, and multiresistance was more frequent in contaminated sites. However, differences among sites were not observed in terms of prevalence or diversity of acquired antibiotic resistance genes, integrons and plasmids. Gene merA, encoding mercury resistance, was only detected in isolates from contaminated sites, most of which were multiresistant to antibiotics. Results indicate that metal(loid) contamination selects for antibiotic resistance in plant surfaces. In salt marshes, antibiotic resistance may be subsequently transferred to other environmental compartments, such as estuarine water or animals, with potential human health risks.

Characterization of antibiotic resistant and pathogenic Escherichia coli in irrigation water and vegetables in household farms

This study aimed to characterize Escherichia coli present in irrigation water and vegetables from 16 household farms. Isolates were obtained from 50% of water (n=210 isolates) and 38% of vegetable samples (n=239). Phylogroups B1 (56% of isolates) and A (22%) were the most prevalent both in water and vegetables. Diarrheagenic strains were detected in vegetables. Irrespective of the source (i.e. water or vegetables), the most common antibiotic resistance was against streptomycin (89% resistant isolates) and tetracycline (24%). Common acquired genes (e.g. blaTEM, tetA, tetB) were found in isolates from both sources. Class I integrons were detected in water (arrays dfrA1-aadA1 and dfr16-blaP1b-aadA2-ereA) and vegetables (unknown arrays). intI2 was detected in water (dfrA1-sat2-aadA1). Plasmids were detected in 14 isolates (IncFIC, IncFIB, IncFrep, IncI1 in both samples; IncY in vegetables). Plasmids from seven isolates were transferrable by conjugation, conferring resistance to antibiotics to the recipient strain. Multidrug-resistant (MDR) strains were isolated from water (12% of the unique isolates) and vegetables (21%). Predominant sequence types (STs) among MDR isolates were ST10, ST297 and ST2522. In some cases, the same STs and identical clones (as showed by rep-PCR typing) were detected in water and vegetables, suggesting cross-contamination. This study identified several risk factors in E. coli isolates from vegetables and irrigation water, raising health concerns. Also, results suggest that irrigation groundwater constitutes a source of E. coli that may enter the food chain through vegetables ingestion.

Molecular assessment of microbiota structure and dynamics along mixed olive oil and winery wastewaters biotreatment

The major parcel of the degradation occurring along wastewater biotreatments is performed either by the native microbiota or by added microbial inocula. The main aim of this study was to apply two fingerprinting methods, temperature gradient gel electrophoresis (TGGE) and length heterogeneity-PCR (LH-PCR) analysis of 16S rRNA gene fragments, in order to assess the microbiota structure and dynamics during mixed olive oil and winery wastewaters aerobic biotreatment performed in a jet-loop reactor (JLR). Sequence homology analysis showed the presence of bacterial genera Gluconacetobacter, Klebsiella, Lactobacillus, Novosphingobium, Pseudomonas, Prevotella, Ralstonia, Sphingobium and Sphingomonas affiliated with five main phylogenetic groups: alpha-, beta- and gamma-Proteobacteria, Firmicutes and Bacteroidetes. LH-PCR analysis distinguished eight predominant DNA fragments correlated with the samples showing highest performance (COD removal rates of 67 up to 75%). Cluster analysis of both TGGE and LH-PCR fingerprinting profiles established five main clusters, with similarity coefficients higher than 79% (TGGE) and 62% (LH-PCR), and related with hydraulic retention time, indicating that this was the main factor responsible for the shifts in the microbiota structure. Canonical correspondence analysis revealed that changes observed on temperature and O2 level were also responsible for shifts in microbiota composition. Community level metabolic profile analysis was used to test metabolic activities in samples. Integrated data revealed that the microbiota structure corresponds to bacterial groups with high degradative potential and good suitability for this type of effluents biotreatments.