Tel.+49 30 83 8-518 40/518 43 Fax.+49 30 838 45 18 51
Over the past decades multi-resistant Enterobacteriaceae like Extended-spectrum beta-lactamases (ESBL)-producing Escherichia (E.) coli have become a major challenge to infection control both in human and veterinary medicine. The spread of antimicrobial resistance occurs mainly by the acquisition of mobile genetic elements like resistance plasmids or the clonal spread of multi-resistant lineages. Most of the resistance genes in pathogens have evolved originally in long periods of evolution in environmental bacteria like the ESBL-enzyme family blaCTX-M which presumably originates from a soil Kluyvera species. Apparently within a half century of usage of antimicrobials in human and veterinarian clinics, the environmental resistome has made its way into bacteria of clinical importance. However, within the last years it became obvious that we have to consider the other side of the medal of this development as well: the transmission of pathogenic and now multi-resistant bacteria and/or their resistance genes back to the environment and subsequently to wildlife. My habilitation thesis focuses on multi-resistant E. coli as a prototype species for the spread of antimicrobial resistance into wildlife. E. coli represents a commensal of the gut of many birds and mammals including humans. Due to its omnipresence in faeces it is distributed to the environment where it can survive as well. For these reasons it has a long tradition as indicator bug of faecal pollution. Despite its commensal character E. coli is frequently implicated in intestinal and extra-intestinal infectious diseases the treatment of which requires the use of anti-infectives. Furthermore multi-resistant E. coli especially ESBL-producers are among the “super bugs” with pose a major threat to public health due to limited treatment options in case of infectious diseases. Summing up, this makes E. coli an ideal paradigmatic candidate for my research. In contrast to the wealth of studies dealing with ESBL-producing E. coli, be it in human, veterinary medicine or livestock breeding their presence and impact on the microbiota of wildlife has been addressed rarely. Nevertheless, due to the work of a small number of groups including my own wildlife has gained more attention in the last years as the occurrence of ESBL-producing E. coli in wildlife could implicate consequences like new reservoir functions and transmission pathways with impact on human and animal health due to the zoonotic potential of E. coli. My initial studies aimed at gaining detailed information on the host distribution of multi-resistant E. coli in avian and small mammal wildlife species. Two avian groups were identified as highly prevalent carriers of multi-resistant E. coli, namely birds of prey and waterfowl. But as we also observed passerines and other avian groups so the carriage of multi-resistant E. coli does not seem to be restricted, pointing towards the absence of host species dependence. The resistance patterns of the avian isolates are comparable to the ones that have been reported for livestock in Europe. Interestingly, in our study on rural rodents we found lower numbers of multi-resistant E. coli compared to rural birds (2% vs. 5%). Nevertheless, similar to the results of the avian study the resistance pattern of the rodent-borne E. coli was comparable to the common antimicrobial resistances observed among E. coli from swine, poultry or cattle. Additionally we found a correlation between antimicrobial resistance of E. coli isolates in wild mice and livestock densities for Germany. This corroborates the idea of some kind of environmental pollution with multi-resistant E. coli by farming and livestock breeding practises resulting in higher rates of multi-resistant bacteria in rodents. Our initial screening study on urban rats supported the theory of an influence of synanthropism on carriage rates of multi-resistant bacteria in wildlife as 13.6% of the rat population carried multi-resistant E. coli. The initial screening studies as the first part of my work revealed that a broad range of mammal and avian species can carry multi-resistant E. coli. Furthermore, the position of the host in the food chain and its general feeding behaviour seem to be influential factors. Moreover among the multi-resistant isolates we observed a frequent combination of resistance with phylogenetic backgrounds related to virulence and the possession of genes associated with extra-intestinal virulence. During the second part of my work I concentrated on ESBL-producing E. coli. This enabled a deep characterization and comparison to isolates of human and animal origin. Among the isolates from avian hosts we detected a clonal lineage of E. coli of the sequence type ST648. This sequence type seems to be associated with ESBL-producing E. coli in human and veterinary medicine worldwide. Additionally, we described an ESBL-producing C. freundii strain from a Tawny owl, which was the first description of an ESBL-producing Citrobacter in wildlife. The detailed characterization of several rat borne ESBL-producing E. coli revealed the presence of a pandemic ESBL-producing E. coli lineage with high extraintestinal virulence. At this time point our work was the first to describe the B2-O25b-ST131 pandemic ESBL-producing E. coli lineage in urban rats and furthermore in wild mammals in general. Another ESBL-producing E. coli obtained from a rat presented a hybrid strain that paradigmatically combined multi-resistance and high extraintestinal pathogenicity. The isolate belonged to one of the most virulent ExPEC (extraintestinal pathogenic E. coli) lineages (ST95) and proofed its pathogenicity in an animal infection model. These finding gave initial evidence of a possible role of urban rodents as hosts of ESBL-producing E. coli with a high extraintestinal virulence potential. The types of ESBL enzymes as well as the phylogenetic background of the ESBL-producing E. coli characterized from wildlife overlap largely with those ESBL enzymes dominant in isolates from human and veterinary patients. As shown for ST131 even identical clonal lineages are present underlining the zoonotic character of ESBL-producing E. coli. In the last part of my work I used selective isolation methods to verify the carriage rates of ESBL-producing E. coli in wildlife. In one study comparing birds of prey from remote and human influenced areas we obtained a carriage rate of ESBL-producing E. coli of 5 % independent from where the isolates originated from, namely Germany or Mongolia. The Mongolian birds were sampled in the Gobi desert, an area with no considerable human and livestock populations and absence of agriculture. These findings point towards a contribution of avian migration to the transmission of multi-resistant bacteria in this remote area as it is very unlikely that the birds picked up the strain locally by the time of sampling. The characterization of ESBL-producing E. coli from birds of prey isolated in Mongolia and Germany as well as the ones from urban brown rats (R. norvegicus) identified clonal lineages that represent clinical relevant isolates with a clear zoonotic potential. In urban rats of the species R. norvegicus we furthermore found alarming high rates with of 16% of all animals carrying an ESBL-producing E. coli. These rates exceeded those which have recently been reported for healthy individuals from comparable urban settings (5% - 8%), but were similar to the ESBL-producing E. coli rates reported in hospitalized patients or their household contacts (12% - 16%). Urban rats might therefore present a sentinel and a possible vector within urban transmission cycles and could become a permanent environmental source of zoonotic and multi-resistant E. coli. Both the clear influence of synanthropism as demonstrated by the high rates of ESBL-producing E. coli in R. norvegicus, as well as still substantial carriage rates in remote areas in the birds underlined the need for holistic approaches, comprising humans, animals and the environment to explore putative transmission cycles of multi-resistant ESBL-producing E. coli. In addition it perfectly depicts the importance of the “One Health” initiative.
As a first step in this direction we comparatively characterized ESBL-producing E. coli of avian origin and belonging to phylogenetic group D-and sequence type (ST) 648 producing CTX-M-type ESBLs together with isolates from companion animals, livestock and humans. We were able to proof a pandemic occurrence of the same clonal lineages of D-ST648-ESBL-producing E. coli independent from the host. This finding highlights the possibility of interspecies transmission notably from human to companion and wild animals and vice versa. Besides these public health issues another dimension should be kept in mind. The widespread occurrence of ESBL-producing E. coli in wildlife, even though these animals have never been exposed continuously to antimicrobials, weakens the presumption of a decline of resistance with the absence of antimicrobial selection pressure alone. The theory of a “burden of resistance” might be of overestimated importance at least for some specific multi-resistant E. coli lineages like ST131 or ST648. The frequent observation of a combination of multi-resistance with the possession of genes associated with extraintestinal virulence in the wildlife isolates could be one explanatory approach for the success of ESBL- producing E. coli in wildlife and other non-resistance factors might be of equal importance. Based on my data it seems reasonable that future research in this field should focus on two areas. First of all detailed epidemiological approaches are needed as most of the current data is not representative, enabling only preliminary insights. The widespread occurrence of ESBL- producing E. coli in wildlife hosts has been proven by my work and others showing that wildlife might serve as sentinel for the spread of antimicrobial resistance in the environment. Future epidemiological studies should therefore concentrate on holistic approaches either on (I) explicit synanthropic species like rats and possible urban transmission pathways as the chances of transmission are higher due to spatial proximity of wildlife and humans or (II) on a global scale on intercontinental migratory birds as their mobility range and numbers is comparable to international travel. Additionally it will be important to screen for the arrival of AmpC- and or carbapenemase-producing E. coli in wildlife populations, as their importance in the clinical field is rising and their appearance in the environment seems only a question of time. Secondly we need a detailed molecular characterization of ESBL-producing E. coli in wildlife to understand the success of multi-resistant bacteria in non-clinical settings which might have implications for the treatment of multi-resistant bacteria in the clinics as well. Studies in this context should verify (I) basic questions of host-pathogen interaction of ESBL-producing E. coli like colonization and shedding and (II) elucidate the influence of non-resistance and/or fitness factors like metabolism, adhesion or motility on the success of ESBL-producing E. coli in the environment. As first studies clearly point toward the existence of such contributing factors my future work will focus on these areas.