Zoonotic potential of escherichia coli and other enterobacteriaceae isolated from poultry meatstudy of antibiotic resistances and definition of clonal groups pathogenic for humans

  1. Díaz Jiménez, Dafne
unter der Leitung von:
  1. Azucena Mora Gutiérrez Doktormutter

Universität der Verteidigung: Universidade de Santiago de Compostela

Fecha de defensa: 01 von Oktober von 2021

  1. María Rosa Caro Vergara Präsident/in
  2. Ana Otero Sekretärin
  3. Ana María Carvajal Urueña Vocal
  1. Departamento de Microbioloxía e Parasitoloxía

Art: Dissertation


COVID-19 has been threatening the world for almost two years now. Fortunately, the care of many researchers has allowed the development of precise combat weapons in the form of vaccines in record time. But this pandemic will leave us many absences, and many consequences, such as those derived from the temporary eclipse of the greatest health challenge: the antimicrobial resistances (AMR). The increase of multidrug-resistant (MDR) bacteria to last-resorts antibiotics (i.e. to colistin, carbapenems, cephalosporins) is one of the most serious public health problems worldwide due to the lack of options for an adequate treatment, the increase of mortality rates and health costs. According to the European Centre for Disease Prevention and Control (ECDC), more than 670.000 bacterial infections can be attributed to MDR bacteria, which causes 33.000 deaths annually in Europe. Considering the risk associated with the antimicrobial use in animals and potential impact on humans, the European Medicines Agency (EMA) has recently proposed a new categorization, including in Category A (“Avoid”) those antibiotics not currently authorized in veterinary medicine in the EU, such as fosfomycin or monobactams; and Category B (“Restrict”) for those antimicrobials that should be restricted in animals to mitigate the risk to public health, namely, quinolones, 3rd- and 4th-generation cephalosporins and polymyxins. Therefore, this is a critical moment when the reduction of antibiotic pressure by different approaches, makes necessary to track bacterial evolution in order to design new strategies. Escherichia coli is part of the commensal microbiota of the digestive system in warm-blood vertebrates that can play diverse roles depending on its virulence profile. While intestinal pathogenic E. coli (InPEC) are accurately distinguished from the commensal gut microbiota based on certain virulence factors, this is not as simple with the extraintestinal pathogenic E. coli (ExPEC) since they behave as opportunistic pathogens that can colonize the intestinal environment without causing harm to the host. Thus, no set of genes can unequivocally define ExPEC strains or the different categories. So far, they have been categorized due to their isolation from infections located outside of the digestive system, and / or based on the presence of genes statistically associated with the extraintestinal pathogenic potential of the strains, which can be used predictively. Besides, certain extraintestinal lineages of E. coli, such as the pandemic ST131, have been worldwide recognized by their implication in human infections, and also, by their role in the spreading of antibiotic resistances. The hypothesis that food, particularly poultry products, can act as a reservoir for human extraintestinal pathogens like E. coli and other Enterobacteriaceae in humans is based on scientific evidence. Certain strains that cause avian pathology (avian pathogenic E. coli, APEC) show a high genetic similarity to those that cause extraintestinal pathology in humans, so several studies report that some human ExPEC strains have evolved from or are common to APEC strains. The evidence suggesting this hypothesis are, among others: A) The geographical and temporal grouping of ExPEC strains isolated from patients with extraintestinal infections, suggesting the appearance of an outbreak or a common source of exposure. B) The global distribution of lineages of identical ExPEC strains, which indicate the global spread of contamination carried through food. C) The detection of identical genotypes of ExPEC isolated from human infections, as well as from food products when strains from the same geographic area were analyzed. D) The disproportionate representation of pandemic or international ExPEC lineages among the hundreds of ST causing extraintestinal infections in all regions of the world, indicating a greater biological or fitness advantage in different reservoirs, production animals or non-production animals (birds wild). E) The relatively recent appearance of the ST69, ST131 and ST393 genotypes as ExPEC, suggesting the recent introduction of these genotypes into the human intestinal niche from external sources. Due to the high plasticity of the E. coli genome hybrid pathotypes are frequent and unpredictably emerging due to the important role played by different mobile genetic elements (MGEs) such as plasmids, bacteriophages, pathogenicity islands, transposons and insertion sequence elements, in the evolution of the bacteria. Furthermore, strains with complex hybrid pathotypes with combinations of two different groups of intestinal pathogenic E. coli (InPEC) (Shiga toxin-producing E. coli; STEC + enteroaggregative E. coli; EAEC) or InPEC and ExPEC (for example, atypical enteropathogenic E. coli; aEPEC + ExPEC and STEC + APEC) are increasingly reported in human clinical cases. Since 2011, when a novel Shiga-toxin-producing E. coli belonging to serotype O104:H4, with virulence features common to the EAEC and CTX-M-15 producer was identified as the one involved in the large German outbreak the concept of pathotype has been questioned and currently, classical and new approaches (WGS), are being used to enhance the understanding of the evolution of this highly adaptable species. The use of antibiotic therapy in food production animals has been accepted as the main cause of the AMR increase, including resistance to colistin. A rapid spread of extended-spectrum β-lactamases (ESBL) has occurred in the last decades, mainly due to their presence in plasmids and expansion through successful clonal groups, such as the pandemic ST131 of E. coli. Presently, there is great concern about the in vivo acquisition of mcr- and blaESBL-bearing plasmids by human E. coli isolates following treatment with colistin, or via animal transmission through direct contact or the food chain. ST131 is the main pandemic clone responsible for the global spread of ESBLs. First identified in 2008, ST131 strains belong to phylogroup B2 and mainly to the serotypes O25b:H4 or, less frequently, O16:H5. Three years after its first isolation, it was already spread, being the bacterial agent involved in more than 50% of cases of UTIs caused by ESBL-producing strains in numerous hospitals in different countries. The prevalence of resistance to first-line oral antibiotics such as trimethoprim-sulfamethoxazole, amoxicillin, and amoxicillin-clavulanate has been steadily increasing during these years, making the treatment of infections very difficult and endangering the lives of patients. Although it is associated with ExPEC infections such as UTI, septicemia, surgical wound infections and meningitis, this clone is also frequently isolated from the digestive tract of healthy humans. That is why, the human intestinal tract was though the main reservoir of ST131. However, the growing scientific community interest towards this ST, found it within diverse sources such as companion, food-production and wild animals, rivers, sewage, even in the Antarctic region. The clades A and C of ST131 are mainly associated with human pathology, while the clade B is determined in strains isolated from different niches such as poultry and pigs, along with humans. An important challenge is to know which determinants make certain clones adapt to a specific host meanwhile others can be transmitted between different species, with jumps as important as between mammals and birds. In the case of ST131, this relationship between the different clades and their presence in different hosts has not be completed understood yet. The present doctoral thesis comprises three studies, “Chicken and turkey meat: Consumer exposure to multidrug-resistant Enterobacteriaceae including mcr-carriers, uropathogenic E. coli and high-risk lineages such as ST131” (Díaz-Jiménez et al., 2020a), “Microbiological risk assessment of turkey and chicken meat for consumer: Significant differences regarding multidrug resistance, mcr or presence of hybrid aEPEC/ExPEC pathotypes of E. coli” (Díaz-Jiménez et al., 2021) and “Genomic Characterization of Escherichia coli Isolates Belonging to a New Hybrid aEPEC/ExPEC Pathotype O153:H10-A-ST10 eae-beta1 Occurred in Meat, Poultry, Wildlife and Human Diarrheagenic Samples” (Díaz-Jiménez et al., 2020b). The aim of the present doctoral thesis, developed in the frame of two national projects (PN AGL2016-79343-R and PID2019-104439RB-C21/AEI/10.13039/501100011033), were to analyse the zoonotic potential of Enterobacteriaceae isolated from poultry, with the characterization of antibiotic resistances and definition of clonal groups pathogenic for humans. Thus, we evaluated the consumer exposure to Enterobacteriaceae with capacity to develop problematic extraintestinal infections, either by their virulence and / or resistance traits, via chicken and turkey meat. The hypothesis of the present thesis was that poultry meat would act as a reservoir, and potentially transmitter, of pathogenic strains that might be implicated in human UTI. To demonstrate this hypothesis, the strategy was to analyze retail poultry meat directly acquired at points of sale with the idea that the final product provides data on what is happening on the farm, at the slaughterhouse, and what goes into the consumer's kitchen. The second strategy was to identify potential uropathogenic clonal groups of E. coli based on specific genetic markers. Finally, we considered “high-risk” strain that with the capacity to develop a serious extraintestinal infection in humans, due to either its virulence potential and / or its antibiotic resistance. The specific goals of the present doctoral thesis were first to design an efficient protocol for the recovery of food-borne E. coli and other pathogenic and / or antimicrobial-resistant Enterobacteriaceae. The second objective was to acquire knowledge on the current situation regarding AMR in poultry farming, paying special attention to antimicrobial categories A and B of EMA. We also aimed to assess the consumer exposure, via poultry meat, to high-risk E. coli and other Enterobacteriaceae isolates with potential to develop severe infections by either bacterial virulence and / or antibiotic resistance traits. Finally, we aimed to explore the food transmission route of specific E. coli clones of human and animal origin through comparative genetic and genomic analysis. We randomly sampled 100 retail fresh meat products (50 of chicken and 50 of turkey) in six Spanish supermarket chains and local butcher located in Lugo (northwest Spain). By conventional culture, 358 different Enterobacteriaceae isolates were recovered (170 isolates recovered from chicken samples and 188 isolates recovered from turkey samples) using MacConkey Lactose, MacConkey Sorbitol with tellurite and cefixime, CHROMID® ESBL and CHROMID®CARBA SMART. Bacterial identification revealed that 323 out of 358 isolates were E. coli, 28 K. pneumoniae, six Serratia fonticola and one Enterobacter cloacae. This collection was fully characterized including: phylogroup, serotype, ST and clonal complex, clonotype, virulence and resistance profile. A second collection was obtained during the period of 2005 to 2015 from different surveillance studies performed at LREC, in Lugo, Spain, which aimed the detection of ESBL-producing E. coli. These studies included samples from chicken, beef and pork meat, as well as poultry farm environment and wildlife. Those isolates conforming the aEPEC pathotype of serotype O153 were further characterized. In our first study we evaluated the consumer exposure via poultry meat to Enterobacteriaceae with capacity to develop severe extraintestinal infections by either bacterial virulence and / or antibiotic resistance traits. The characterization of 256 isolates (84 representative E. coli isolates, 137 ESBL-producing E. coli isolates, 28 ESBL-producing Klebsiella pneumoniae isolates, six ESBL-producing Serratia fonticola isolates and one ESBL-producing Enterobacter cloacae isolate) and the assessment of five parameters showed that 96 out of 100 poultry meat samples acquired in supermarkets of the northwest of Spain posed ≥ one potential risk. Specifically, i) 96% of the samples carried Enterobacteriaceae resistant to antimicrobials of categories A (64% with resistance to monobactams) or B (95% with resistance to cephalosporins of 3rd- and 4rd- generation, quinolones and / or polymyxins) of the new categorization of EMA. ii) More than one ESBL-producing Enterobacteriaceae species were recovered from 29% of samples, mostly E. coli and K. pneumoniae. iii) Characterization of E. coli isolates showed that extraintestinal and / or uropathogenic high-risk clonal groups (ST10, ST23, ST38, ST48, ST58, ST69, ST88, ST93, ST95, ST101, ST115, ST117, ST131, ST141, ST167, ST350, ST345, ST354, ST359, ST410, ST602, ST617, ST641, ST906, ST1485) were present in 62% samples. iv) E. coli isolates recovered from 25% samples conformed the ExPEC status v) E. coli isolates recovered from 17% samples conformed UPEC status. Regarding K. pneumoniae, at least eight of the 11 STs identified in our collection have been reported within human clinic isolates; specifically: ST15, ST45, ST111, ST147, ST307, ST627, ST966 and ST1086 (22 of the 28 K. pneumoniae belonged to these eight STs). The plasmid-mediated colistin resistance mcr-1 gene was determined in 13 E. coli isolates from seven meat samples, however, the eleven K. pneumoniae phenotypically resistant to colistin were negative by PCR for the presence of mcr-1 to mcr-8 genes, probably indicating chromosomic-encoding resistance. In our second study, we assessed the risk for consumers attending only to E. coli isolates, we proposed a laboratory workflow based on six virulence and / or antimicrobial resistance traits and included the development of a duplex PCR for the screening of ExPEC isolates. We characterized 323 isolates recovered from 100 poultry meat samples. This characterization revealed that poultry meat is a rich phylogenetic source of E. coli phylogroups (A to G) and Escherichia clade I. Non-susceptible E. coli isolates to monobactams, 3rd-generation cephalosporins and / or fluoroquinolones, were present in 71% of the samples. Besides, 47% carried ≥2 different E. coli positive for ESBL, pAmpC or mcr genes. Isolates from 78% of the poultry meat exhibited ExPEC status, and 53% were carriers of isolates positive for the UPEC status. The STs identified in 86% of the samples belonged to the so-called ExPEC high-risk lineages, being 73% carriers of clonal groups identified in human infections of the same Health Area. Moreover, different human-associated clones co-occurred in same meat sample: ST131-B2 (CH40-22), ST648-F (CH4-58), ST93-A (CH11-neg) or ST95-B2 (CH38-27), ST354-F (CH88-58), ST155-B1 (CH4-neg). Globally, 84% of the meat samples posed ≥ 3 risks factors, including resistance genes, successful clones and virulence traits. Turkey meat showed significant higher rates concerning mcr-carriage or MDR; while the ExPEC status rate, or the presence of hybrid pathotypes such as the aEPEC/ExPEC O153:H10-A-ST10 (CH11-54), were associated with chicken origin (P < 0.05). In our third study we took as start point the different surveillance studies (2005–2015) in northwest Spain that revealed the presence of eae-positive isolates of E. coli O153:H10 in meat for human consumption, poultry farm, wildlife and human diarrheagenic samples. The aim of this study was to explore the genetic and genomic relatedness between human and animal/meat isolates, as well as the mechanism of its persistence. We also wanted to know whether it was a geographically restricted lineage, or whether it was also reported elsewhere. Conventional typing showed that 32 isolates were O153:H10-A-ST10 fimH54, fimAvMT78, traT and eae-beta1. Amongst these, 21 were CTX-M-32 or SHV-12 producers. The PFGE XbaI - macrorestriction comparison showed high similarity (>85%) between the isolates of the collection. The plasmidome analysis revealed a stable combination of IncF (F2:A-:B-), IncI1 (ST unknown) and IncX1 plasmid types, together with non-conjugative Col-like plasmids. The core genome investigation based on the core genome multilocus sequence typing (cgMLST) scheme from EnteroBase proved close relatedness between isolates of human and animal origin. From our results we concluded that poultry meat microbiota is a source of genetically diverse Enterobacteriaceae, resistant to relevant antimicrobials (categories A and B of EMA) and potentially pathogenic for humans, including hybrid pathotypes of E. coli, high-risk clonal groups of E. coli associated with human extraintestinal and / or uropathogenic pathologies, as well as K. pneumoniae clonal groups of clinical interest. Our results would indicate that the industrial production system for turkey meat seems to exert greater selection pressure of antibiotic resistant strains compared to chicken, which is reflected in significant higher rates of mcr-positive E. coli and MDR isolates, including ESBL-producing K. pneumoniae, in turkey meat. With regard to the methods here investigated, we found that protocols I and II, based on MacConkey Lactose and MacConkey Sorbitol with telurite and cefixime agar incubated at 37 ºC, are the most effective for the recovery of isolates satisfying the ExPEC and UPEC status, as well as the rbfO25b-positive isolates associated with the clonal group STl31. The protocol V (CHROMID® ESBL agar plates 37 ºC) is key for the recovery of ESBL or pAmpC-producing Enterobacteriaceae. The duplex PCR based on iutA and KpsM II genes on MacConkey Lactose and MacConkey Sorbitol with telurite and cefixime agar is essential for the accurate screening of the isolates conforming ExPEC status, as well as for the recovery of those with UPEC status. Finally, we concluded that the microbiological method proposed here (pre-enrichment, enrichment in MacConkey Lactose broth, and inoculation onto MacConkey Lactose agar/ MacConkey Sorbitol with telurite and cefixime agar/CHROMID® ESBL), followed by the screening of six virulence/AMR traits (ExPEC status, UPEC status, ESBL/pAmpC producer, mcr-1 carrier, MDR, rfbO25b), would help to elucidate the role of ExPEC as new extraintestinal food-borne pathogens. Our results prove that a hybrid MDR aEPEC/ExPEC belonging to the clonal group O153:H10-A-ST10 (CH11-54) eae-beta1 is circulating in our region within different hosts, including wildlife. It seems implicated in human diarrhea via food (meat) transmission, and in the spreading of ESBL genes (mainly of CTX-M-32 type). The concomitant presence of IncF (F2:A-:B-), IncI1 and IncX1, together with non-conjugative Col156-like plasmids might be implicated in the successful persistence of this hybrid pathotype.