Final Report Summary - ECOCHAR (Characterization of widespread uropathogenic multidrug resistant Escherichia coli clones by using ... conventional genotypic methods and high-throughput technology)
Our project focused on the characterisation of E. coli isolates belonging to representative clones associated with hospital and community acquired infections and recently responsible for the dissemination of particular antibiotic resistance mechanisms in order to identify factors contributing to their pathogenicity and persistence. Our results revealed a high inter and intraclonal diversity although particular variants with increased epidemicity and/or pathogenicity potential were identified in different geographic regions and/or settings. We identified features (virulence gene profiles, acquisition of specific antibiotic resistance plasmids, biofilm formation) potentially responsible for the higher epidemicity and persistence of some of these clones although no clear relationship could be established between genotypic virulence profiles and phenotypic biofilm formation, highlighting that other non-explored factors might be involved and the need for further studies. We detected a good discrimination power of particular clinically relevant E. coli clones using different quick and low-cost analytical methods (MALDI-TOF- Matrix-assisted laser desorption / ionisation time of flight mass spectrometry and Fourier transform infrared spectroscopy (FTIR)), which will be further optimised and exploited for potential application at a large-scale basis in a clinical microbiology context. This scientific knowledge is of utmost relevance to develop new low cost, easy and simple methods for identification and discrimination of clinically relevant clones and is of interest for the scientific community.
Summary description of results and impact
Escherichia coli is one of the most frequent bacteria implicated in extraintestinal human and animal infections, and the occurrence of uropathogenic E. coli (UPEC) clones resistant to broad-spectrum cephalosporins, usually caused by production of extended-spectrum ß-lactamases (ESBLs), has been increasingly reported. Epidemiological and essentially genotypic studies have shown the amplification of specific multidrug resistant E. coli clones in wide geographic areas and extended periods of time, which are greatly influencing the outcome of infections and contributing to the dissemination of antibiotic resistance genes. However, genotypic methods impair a rapid and large-scale basis assessment of their prevalence and distribution in different settings, which is critical to define and interrupt transmission pathways. Moreover, the factors contributing to their enhanced epidemicity, pathogenicity and/or persistence are poorly understood. Our goal was to characterise by a multidisciplinary approach the inter and intraclonal diversity of representative E. coli clones associated with hospital and community acquired infections and recently responsible for the dissemination of particular ESBL, and to characterise the factors contributing to their pathogenicity and persistence. Specifically, we aimed:
i) to characterise the virulence profile of clinically relevant E. coli clones;
ii) to investigate the ability of high-throughput FTIR and matrix-assisted laser desorption / ionisation time of flight mass spectrometry (MALDI-TOF) to identify and discriminate clinically relevant UPEC clones, by applying and optimising these methods in comparison with conventional genotypic ones; and
iii) to assess the ability of isolates belonging to these E. coli clones to adhere and form biofilms
For this purpose, we gathered a collection of E. coli isolates belonging to specific clones belonging to phylogenetic groups B2 (ST131), D (ST69, ST393, ST405), A (ST10, ST23) and B1 (ST155, ST359) from diverse geographic regions and settings and spanning a period of 30 years (1980 - 2010), including representative isolates from particular nosocomial or community outbreaks.
We detected a high inter and intraclonal diversity in the collection studied. A higher homology between isolates belonging to B2-ST131, D-ST69, D-ST393 and D-ST405 (69 - 79 %) groups, than among A-ST10, A-ST23, B1-ST155 or B1-ST359 isolates (50 - 54 %), suggesting a higher epidemicity for B2 and D isolates or indicating that they have more recently diverged. Interestingly, despite the high clonal diversity observed for isolates from each clonal group, clusters including particular lineages (related variants exhibiting closely related PFGE-types and virulence gene profiles) from different geographic regions and origins were identified. All isolates were particularly enriched in a diversity of IncFII plasmids (FII, FII+FIA, FII+FIB, FII+FIA+FIB). Moreover, we show that particular widespread lineages (such as ST131 variants) might have been selected and amplified after acquisition of diverse and specific multidrug resistant IncFII plasmids.
A similar frequency between extra-intestinal pathogenic E. coli (ExPEC) (51 %) and non-ExPEC (49 %) isolates was observed, although ExPEC isolates were mostly associated with ST393 (100 %), ST69 (100 %), ST131 (76 %), but also with ST23 complex (43 %), ST359 (29 %), ST405 (18 %), ST155 (14 %) and ST10 (5 %) isolates. Interestingly, non-ExPEC isolates (n = 55; 49 %) were also responsible for extra-intestinal infections (namely UTIs, bacteremia or respiratory infections), highlighting that other non-explored VFs might be influencing extra-intestinal pathogenesis. Higher virulence scores were observed for ST131 (median 11, range 7 - 15), ST393 (median 13, range 7 - 15), ST69 (median 14, range 9 - 15) and ST23 (median 9, range 2 - 12) isolates. fimH (n = 101, 90 %), iutA (n = 86, 77 %) and traT (n = 81, 72 %) were the most frequently virulence genes detected, whereas the distribution of other virulence factors was variable depending on the ST analysed. It is of interest to highlight that common virulence profiles were detected among isolates belonging to the same ST from different geographic regions or origins suggesting the circulation of particular lineages in different niches. Moreover, these results show that the widespread of particular variants belonging to each clonal group might be explained by their higher epidemicity and/or virulence potential.
Most isolates from all STs were considered as weakly adherent strains (n = 83, 71 %), while some were identified as moderately (n = 22, 19 %) or strongly adherent strains (n = 12, 10 %). Moderately or strongly adherent strains were observed mostly among ST10 (n = 12), ST23 (n = 9) isolates, and also ST69 (n = 4), ST393 (n = 3), ST155 (n = 3), ST405 (n = 2), and ST359 (n = 1). Nevertheless, biofilm production was only confirmed in a few of them. Higher adhesion phenotypes were associated with diverse clonal variants exhibiting variable virulence gene profiles preventing to establish a relationship between these genotypic and phenotypic features. It is of interest to highlight that all ST131 isolates were considered as weakly adherent strains suggesting that additional studies are necessary to further investigate the reasons favouring the expansion and persistence of the pandemic ST131 clonal group. Similarly, biofilm production does not seem to explain the epidemicity associated with E. coli clonal groups belonging to the D phylogenetic group (ST69, ST393 and ST405) nor when detected, this feature could be associated with a particular virulence gene or virulence profile. Our results showed that factors other than biofilm formation seem to be influencing the epidemiological success of these E. coli clonal groups and points out the need to further studies to identify factors involved in the selection of particular epidemic variants with enhanced ability to colonise and persist for extended periods of time.
Different high-throughput methods were tested for the ability to identify and discriminate E. coli clones from the different phylogenetic groups studied: FTIR and matrix-assisted laser desorption / ionisation time of flight mass spectrometry (MALDI-TOF). Both methods revealed a reliable and reproducible discrimination of isolates belonging to B2-ST131, D-ST69, D-ST393, D-ST405 clonal groups, and in some instances between B2-ST131 and B2 E. coli isolates belonging to other STs. Although isolates were grouped in different clusters within clone, no correspondence was found between both methods or with clusters defined by classical genotypic procedures (PFGE), preventing from discriminate at the intraclonal level. The prompt identification of clinically relevant E. coli clones by these simple and low cost methodologies suggests their potential future application as diagnostic tools at a large-scale basis to optimise infection control policies and individual patient management in a clinical context. Additional studies are ongoing to optimise these methods for a wider E. coli bacterial population and for potential application to other clinically relevant bacterial species, which could furthermore potentiate their use as rapid, simple and reliable methods for detection of both antibiotic resistant clones in a clinical context.
Our results were partially published in peer-reviewed journals (other manuscripts are being currently reviewed and/or prepared for submission) and presented in the form of poster communications in both international and national congresses. Outreach activities involved lectures directed to undergraduate and graduate health sciences students as well to pubic community. Moreover, proposals based on the results generated in this project provided new funding resources from both national and international organisations, which will allow further developing and deepening this research line.
Summary description of results and impact
Escherichia coli is one of the most frequent bacteria implicated in extraintestinal human and animal infections, and the occurrence of uropathogenic E. coli (UPEC) clones resistant to broad-spectrum cephalosporins, usually caused by production of extended-spectrum ß-lactamases (ESBLs), has been increasingly reported. Epidemiological and essentially genotypic studies have shown the amplification of specific multidrug resistant E. coli clones in wide geographic areas and extended periods of time, which are greatly influencing the outcome of infections and contributing to the dissemination of antibiotic resistance genes. However, genotypic methods impair a rapid and large-scale basis assessment of their prevalence and distribution in different settings, which is critical to define and interrupt transmission pathways. Moreover, the factors contributing to their enhanced epidemicity, pathogenicity and/or persistence are poorly understood. Our goal was to characterise by a multidisciplinary approach the inter and intraclonal diversity of representative E. coli clones associated with hospital and community acquired infections and recently responsible for the dissemination of particular ESBL, and to characterise the factors contributing to their pathogenicity and persistence. Specifically, we aimed:
i) to characterise the virulence profile of clinically relevant E. coli clones;
ii) to investigate the ability of high-throughput FTIR and matrix-assisted laser desorption / ionisation time of flight mass spectrometry (MALDI-TOF) to identify and discriminate clinically relevant UPEC clones, by applying and optimising these methods in comparison with conventional genotypic ones; and
iii) to assess the ability of isolates belonging to these E. coli clones to adhere and form biofilms
For this purpose, we gathered a collection of E. coli isolates belonging to specific clones belonging to phylogenetic groups B2 (ST131), D (ST69, ST393, ST405), A (ST10, ST23) and B1 (ST155, ST359) from diverse geographic regions and settings and spanning a period of 30 years (1980 - 2010), including representative isolates from particular nosocomial or community outbreaks.
We detected a high inter and intraclonal diversity in the collection studied. A higher homology between isolates belonging to B2-ST131, D-ST69, D-ST393 and D-ST405 (69 - 79 %) groups, than among A-ST10, A-ST23, B1-ST155 or B1-ST359 isolates (50 - 54 %), suggesting a higher epidemicity for B2 and D isolates or indicating that they have more recently diverged. Interestingly, despite the high clonal diversity observed for isolates from each clonal group, clusters including particular lineages (related variants exhibiting closely related PFGE-types and virulence gene profiles) from different geographic regions and origins were identified. All isolates were particularly enriched in a diversity of IncFII plasmids (FII, FII+FIA, FII+FIB, FII+FIA+FIB). Moreover, we show that particular widespread lineages (such as ST131 variants) might have been selected and amplified after acquisition of diverse and specific multidrug resistant IncFII plasmids.
A similar frequency between extra-intestinal pathogenic E. coli (ExPEC) (51 %) and non-ExPEC (49 %) isolates was observed, although ExPEC isolates were mostly associated with ST393 (100 %), ST69 (100 %), ST131 (76 %), but also with ST23 complex (43 %), ST359 (29 %), ST405 (18 %), ST155 (14 %) and ST10 (5 %) isolates. Interestingly, non-ExPEC isolates (n = 55; 49 %) were also responsible for extra-intestinal infections (namely UTIs, bacteremia or respiratory infections), highlighting that other non-explored VFs might be influencing extra-intestinal pathogenesis. Higher virulence scores were observed for ST131 (median 11, range 7 - 15), ST393 (median 13, range 7 - 15), ST69 (median 14, range 9 - 15) and ST23 (median 9, range 2 - 12) isolates. fimH (n = 101, 90 %), iutA (n = 86, 77 %) and traT (n = 81, 72 %) were the most frequently virulence genes detected, whereas the distribution of other virulence factors was variable depending on the ST analysed. It is of interest to highlight that common virulence profiles were detected among isolates belonging to the same ST from different geographic regions or origins suggesting the circulation of particular lineages in different niches. Moreover, these results show that the widespread of particular variants belonging to each clonal group might be explained by their higher epidemicity and/or virulence potential.
Most isolates from all STs were considered as weakly adherent strains (n = 83, 71 %), while some were identified as moderately (n = 22, 19 %) or strongly adherent strains (n = 12, 10 %). Moderately or strongly adherent strains were observed mostly among ST10 (n = 12), ST23 (n = 9) isolates, and also ST69 (n = 4), ST393 (n = 3), ST155 (n = 3), ST405 (n = 2), and ST359 (n = 1). Nevertheless, biofilm production was only confirmed in a few of them. Higher adhesion phenotypes were associated with diverse clonal variants exhibiting variable virulence gene profiles preventing to establish a relationship between these genotypic and phenotypic features. It is of interest to highlight that all ST131 isolates were considered as weakly adherent strains suggesting that additional studies are necessary to further investigate the reasons favouring the expansion and persistence of the pandemic ST131 clonal group. Similarly, biofilm production does not seem to explain the epidemicity associated with E. coli clonal groups belonging to the D phylogenetic group (ST69, ST393 and ST405) nor when detected, this feature could be associated with a particular virulence gene or virulence profile. Our results showed that factors other than biofilm formation seem to be influencing the epidemiological success of these E. coli clonal groups and points out the need to further studies to identify factors involved in the selection of particular epidemic variants with enhanced ability to colonise and persist for extended periods of time.
Different high-throughput methods were tested for the ability to identify and discriminate E. coli clones from the different phylogenetic groups studied: FTIR and matrix-assisted laser desorption / ionisation time of flight mass spectrometry (MALDI-TOF). Both methods revealed a reliable and reproducible discrimination of isolates belonging to B2-ST131, D-ST69, D-ST393, D-ST405 clonal groups, and in some instances between B2-ST131 and B2 E. coli isolates belonging to other STs. Although isolates were grouped in different clusters within clone, no correspondence was found between both methods or with clusters defined by classical genotypic procedures (PFGE), preventing from discriminate at the intraclonal level. The prompt identification of clinically relevant E. coli clones by these simple and low cost methodologies suggests their potential future application as diagnostic tools at a large-scale basis to optimise infection control policies and individual patient management in a clinical context. Additional studies are ongoing to optimise these methods for a wider E. coli bacterial population and for potential application to other clinically relevant bacterial species, which could furthermore potentiate their use as rapid, simple and reliable methods for detection of both antibiotic resistant clones in a clinical context.
Our results were partially published in peer-reviewed journals (other manuscripts are being currently reviewed and/or prepared for submission) and presented in the form of poster communications in both international and national congresses. Outreach activities involved lectures directed to undergraduate and graduate health sciences students as well to pubic community. Moreover, proposals based on the results generated in this project provided new funding resources from both national and international organisations, which will allow further developing and deepening this research line.