Porcine pathogens and xenotransplantation: PERV and beyond!

Concerns regarding the transmission of potentially zoonotic porcine viruses via a xenotransplant have prompted a significant number of studies on methods to eliminate or prevent expression and transmission of these viruses. The main focus of these studies, to date, has been the porcine endogenous retrovirus (PERV); PERV is a genetically acquired element and present in the genome of all swine. This situation is problematic as it cannot simply be eliminated from swine by using methods currently employed to exclude exogenous pathogens in barrier facilities. As such, alternative strategies have been sought to circumvent the potential risk of PERV expression and transmission via a xenotransplant, however, there are other existing and emerging pathogens of concern that should be addressed when using this novel technology in vivo. Zoonotic porcine viruses have been identified that require specific diagnostic methods to confirm their absence. Animal husbandry and the exclusion of pathogens from SPF herds for use in xenotransplantation have been widely discussed and a number of organizations have issued guidelines on the screening for infectious agents. Although these recommendations on monitoring protocol and the identification of adventitious agents are clear, there is no comprehensive list of pathogens to be excluded from these animals that can be applied to all centres carrying out xenotransplantation. Currently, SPF animals used for research purposes are monitored for specific pathogens as defined by local guidelines, and may not be tested for all pathogens relevant to xenotransplantation. As recent data has indicated the potential for certain porcine pathogens to cross the species barrier, it is clear that xenotransplantation is a unique situation which may require us to address a more comprehensive panel of microorganisms than is currently recommended for SPF animals. This presentation will discuss data on the presence of pathogens in pigs, other than PERV, that may cause concern during the clinical application of xenotransplantation and the issues regarding the potential transfer of new zoonotic microorganisms.     

Possible risks posed by single‐stranded DNA viruses of pigs associated with xenotransplantation

Routine large‐scale xenotransplantation from pigs to humans is getting closer to clinical reality owing to several state‐of‐the‐art technologies, especially the ability to rapidly engineer genetically defined pigs. However, using pig organs in humans poses risks including unwanted cross‐species transfer of viruses and adaption of these pig viruses to the human organ recipient. Recent developments in the field of virology, including the advent of metagenomic techniques to characterize entire viromes, have led to the identification of a plethora of viruses in many niches. Single‐stranded DNA (ssDNA) viruses are the largest group prevalent in virome studies in mammals. Specifically, the ssDNA viral genomes are characterized by a high rate of nucleotide substitution, which confers a proclivity to adapt to new hosts and cross‐species barriers. Pig‐associated ssDNA viruses include torque teno sus viruses (TTSuV) in the Anelloviridae family, porcine parvoviruses (PPV), and porcine bocaviruses (PBoV) both in the family of Parvoviridae, and porcine circoviruses (PCV) in the Circoviridae family, some of which have been confirmed to be pathogenic to pigs. The risks of these viruses for the human recipient during xenotransplantation procedures are relatively unknown. Based on the scant knowledge available on the prevalence, predilection, and pathogenicity of pig‐associated ssDNA viruses, careful screening and monitoring are required. In the case of positive identification, risk assessments and strategies to eliminate these viruses in xenotransplantation pig stock may be needed.     

Outcomes of Alpha 1,3-GT-knockout Porcine Heart Transplants Into a Preclinical Nonhuman Primate Model

Background

Solid organ xenotransplantation is a potential solution to current organ shortages in allotransplantation. We performed four heart transplantations from alpha1, 3-galactosyltransferase gene-knockout (GT-KO) pigs to cynomolgus monkeys and monitored immunological parameters before and after transplantation.

Methods

After blood typing of the cynomolgus monkeys, we assessed the binding activity of immunoglobulin G (IgG) and IgM of monkey serum and serum toxicity toward porcine peripheral blood mononuclear cells (PBMCs) using flow cytometry. Immunosuppressive protocols consisted of anti-thymocyte globulin (25 mg/kg), rituximab (20 mg/kg), anti-CD154mAb (20 mg/kg), cobra venom factor (0.05 mg/kg), tacrolimus, and steroid. Cynomolgus monkeys with A or AB blood type with the lowest antibody binding and serum toxicity activity on porcine PBMCs were selected as recipients.

Results

Absolute numbers of CD3⁺ T cells, CD20⁺ B cells, and CD3⁺CD95⁺ memory T cells in the peripheral blood were suppressed upto 24 days after transplantation. Interferon gamma production of T cells in response to porcine antigens were also significantly suppressed. Heart xenografts from GT-KO pigs survived for upto 24 days without pathologic evidence of rejection.

Conclusion

We successfully performed 4 heart xenotransplantations using GT-KO pigs. We overcame hyperacute rejection by using GT-KO pigs, and all of the heart xenografts from the GT-KO pigs survived between 11 and 24 days without pathologic evidence of rejection, disseminated intravascular coagulation, or consumptive coagulopathy; however, we need to optimize protocols for immune modulation and postoperative care to attain long-term survival of solid organ xenografts.     

The production of transgenic pigs for potential use in clinical xenotransplantation: microbiological evaluation

Debate over the infection hazards of pig-to-human xenotransplantation has focused mainly on the porcine endogenous retroviruses (PERV). However, hazards of exogenous infectious agents possibly associated with the xenograft have also been evaluated (Xenotransplantation 2000; 7: 143). We report the results of a health monitoring program demonstrating the exclusion of more than 80 potential pathogens from nine cohorts of pigs reared in a high welfare bioexclusion facility as potential xenograft source animals. A dynamic bacterial flora of pigs reared under barrier conditions was characterized, emphasizing the significance of monitoring for multiresistant antimicrobial sensitivity patterns. Evidence was found for exclusion of two commonly residual exogenous viruses, porcine cytomegalovirus and porcine lymphotropic herpesviruses, among a proportion of the cohorts tested. Finally, there was histopathological evidence for low grade pneumonitis among sentinel pigs, likely to have been associated with the use of quaternary ammonium disinfectants during the production process, indicating a need for review of toxicology data for disinfectant agents used in such bioexclusion systems. Intensive health monitoring programs, based upon regularly updated recommendations from the microbiological research community, will enable significant reductions in the potential hazards associated with pig-to-human xenotransplantation.     

Depletion of pulmonary intravascular macrophages prevents hyperacute pulmonary xenograft dysfunction

BACKGROUND: Recent years have brought dramatic progress in the field of xenotransplantation, with the development of transgenic swine and various other means of overcoming the rejection mediated by xenoreactive antibodies. Although progress has been rapid with kidney and heart xenografts, progress with pulmonary xenografts has lagged behind. Recent findings have suggested that donor pulmonary intravascular macrophages may play a critical role in the hyperacute dysfunction of pulmonary xenografts. METHODS: The function of pulmonary xenografts from pigs depleted of pulmonary intravascular macrophages was compared with the function of xenografts from normal pigs. RESULTS: Pulmonary xenografts from pigs from which pulmonary intravascular macrophages were depleted survived (23.5+/-0.9 hours) about five times longer than normal (macrophage sufficient) xenografts (4.4+/-1.41 hours) (P< 0.0001). At 21 hours post-reperfusion, the left pulmonary arterial flow was 225.0+/-34 ml/min in lungs depleted of pulmonary intravascular macrophages, whereas all normal xenografts had failed. CONCLUSIONS: These findings indicate that donor macrophages play a critical role in pulmonary xenograft dysfunction. This finding has broad implications for xenotransplantation, suggesting that porcine macrophages might pose a barrier to the engraftment and function of a variety of porcine organ xenografts.     

Infectious disease risks in xenotransplantation

Hurdles exist to clinical xenotransplantation including potential infectious transmission from nonhuman species to xenograft recipients. In anticipation of clinical trials of xenotransplantation, the associated infectious risks have been investigated. Swine and immunocompromised humans share some potential pathogens. Swine herpesviruses including porcine cytomegalovirus (PCMV) and porcine lymphotropic herpesvirus (PLHV) are largely species‐specific and do not, generally, infect human cells. Human cellular receptors exist for porcine endogenous retrovirus (PERV), which infects certain human‐derived cell lines in vitro. PERV‐inactivated pigs have been produced recently. Human infection due to PERV has not been described. A screening paradigm can be applied to exclude potential human pathogens from “designated pathogen free” breeding colonies. Various microbiological assays have been developed for screening and diagnosis including antibody‐based tests and qualitative and quantitative molecular assays for viruses. Additional assays may be required to diagnose pig‐specific organisms in human xenograft recipients. Significant progress has been made in the evaluation of the potential infectious risks of clinical xenotransplantation. Infectious risk would be amplified by intensive immunosuppression. The available data suggest that risks of xenotransplant‐associated recipient infection are manageable and that clinical trials can be performed safely. Possible infectious risks of xenotransplantation to the community at large are undefined but merit consideration.     

Herpesvirus infections in xenotransplantation: pathogenesis and approaches

Infectious risk remains an important consideration in the clinical application of xenotransplantation. Vascularized xenografts create unique immunological niches in which bidirectional transmission of pathogens between donor and recipient may occur. Enhanced replication of many pathogens is stimulated by the immune responses induced by transplantation and by the immune suppression used to prevent graft rejection. Herpesviruses are the prototype viruses that are activated during immunosuppression. Quantitative diagnostic molecular assays have been developed for the known herpesviruses causing infection in pigs. Recent data suggest that some herpesviral infections, such as porcine cytomegalovirus, may be excluded from swine used as source animals by careful breeding, while others will require novel strategies for control. This review focuses on porcine and baboon herpesviruses in pig-to-non-human primate solid organ xenotransplantation including direct effects (tissue damage), indirect effects (coagulopathy, rejection), and possible approaches to these infections.     

Infectious risk and retroviral restriction factors in xenotransplantation

The clinical application of xenotransplantation evokes immunological and microbiological as well as virological challenges. Porcine pathogens that do not show any symptoms in their natural host could exhibit a risk of fatal infections to humans. The presence of pig infectious agents including zoonotic and dissimilar agents should be reduced by specific pathogen free (spf) breeding of donor animals. However, the genetic information of porcine endogenous retroviruses (PERV) is integrated in the pig genome and can not be eradicated by spf breeding. The concerns about PERV for human xenograft recipients are based on data of in vitro replication of PERV in some human cell lines. So far, viral replication of PERV has been difficult to demonstrate in non‐human primate cell lines and in preclinical studies of non‐human primates receiving porcine xenografts, respectively. In this regard, natural and effective mechanisms of human and porcine cells counteracting productive infections caused by PERV are important to investigate. Intracellular proteins and components of the innate immune system including endogenous “antiretroviral restriction factors” act at various steps in retroviral replication. The cellular front is composed by several constitutively expressed genes which prevent or suppress retroviral infections. Some of these factors such as members of the tripartite motif (TRIM) and the apolipoprotein B mRNA‐editing polypeptide (APOBEC) families as well as tetherin and zinc‐finger antiviral protein (ZAP) could be useful in the management of PERV in xenotransplantation. The risks of infection and the potential role of antiretroviral restriction factors in xenotransplantation are presented in detail.     

Reduction of consumptive coagulopathy using porcine cytomegalovirus-free cardiac porcine grafts in pig-to-primate xenotransplantation

BACKGROUND: Xenotransplantation using pigs as the source species for organs carries a potential risk for transmission and activation of porcine herpesviruses. Activation of porcine cytomegalovirus (PCMV) in pig-to-baboon xenotransplantation is associated with xenograft injury and possibly an increased incidence of consumptive coagulopathy (CC). METHODS: To further investigate the role of PCMV activation in the occurrence of CC, a strategy to exclude PCMV from the donor was developed. To exclude PCMV, piglets were early-weaned and raised separated from other swine. These piglets were used as donors in an experimental protocol of pig-to-baboon heart xenotransplantation. RESULTS: Early weaning of piglets was successful in excluding PCMV. Use of PCMV-free cardiac porcine xenografts in baboons resulted in prolonged graft survival and prevented consumptive coagulopathy in all recipients. CONCLUSIONS: The use of PCMV-free cardiac grafts is beneficial in reducing the direct effects of PCMV activation in the graft (tissue damage) and the indirect effects of PCMV activation in the recipient (consumptive coagulopathy).     

Clostridium Difficile Infection After Total Joint Arthroplasty: Who is at Risk?

Clostridium difficile–associated diarrhea is a recognized postoperative complication. However, the exact risk factors for this condition after total joint arthroplasty (TJA) remain unknown. This case-controlled study intended to identify the predisposing factors for this condition. There were 16 cases of C difficile infections after 9880 TJA (0.16% incidence) between January 2001 and May 2006 at our institution. The cases were matched with 32 controls for month/year of surgery and surgeon. This study suggests that patients with deteriorated physical status or those who receive greater than one antibiotic after surgery are at a higher risk for developing C difficile–associated diarrhea after TJA.     

Clostridium difficile carriage in adult cystic fibrosis (CF); implications for patients with CF and the potential for transmission of nosocomial infection

Clostridium difficile is an anaerobic Gram-positive, spore-forming, toxin-producing bacillus transmitted among humans through the faecal–oral route. Despite increasing carriage rates and the presence of C. difficile toxin in stool, patients with CF rarely appear to develop typical manifestations of C. difficile infection (CDI). In this study, we examined the carriage, toxin production, ribotype distribution and antibiotic susceptibility of C. difficile in a cohort of 60 adult patients with CF who were pre-lung transplant. C. difficile was detected in 50% (30/60) of patients with CF by culturing for the bacteria. C. difficile toxin was detected in 63% (19/30) of C. difficile-positive stool samples. All toxin-positive stool samples contained toxigenic C. difficile strains harbouring toxin genes, tcdA and tcdB. Despite the presence of C. difficile and its toxin in patient stool, no acute gastrointestinal symptoms were reported. Ribotyping of C. difficile strains revealed 16 distinct ribotypes (RT), 11 of which are known to be disease-causing including the hyper-virulent RT078. Additionally, strains RT002, RT014, and RT015, which are common in non-CF nosocomial infection were described. All strains were susceptible to vancomycin, metronidazole, fusidic acid and rifampicin. No correlation was observed between carriage of C. difficile or any characteristics of isolated strains and any recorded clinical parameters or treatment received. We demonstrate a high prevalence of hypervirulent, toxigenic strains of C. difficile in asymptomatic patients with CF. This highlights the potential role of asymptomatic patients with CF in nosocomial transmission of C. difficile.     

Wild griffon vultures (Gyps fulvus) fed at supplementary feeding stations: Potential carriers of pig pathogens and pig‐derived antimicrobial resistance?

The carriage of two important pathogens of pigs, that is enterotoxigenic Escherichia coli (ETEC) and Clostridioides difficile, was investigated in 104 cloacal samples from wild griffon vultures (Gyps fulvus) fed on pig carcasses at supplementary feeding stations (SFS), along with their level of antimicrobial resistance (AMR). E. coli was isolated from 90 (86.5%) samples, but no ETEC was detected, likely because ETEC fimbriae confer the species specificity of the pathogen. Resistance to at least one antimicrobial agent was detected in 89.9% of E. coli isolates, with AMR levels being extremely high (>70%) for tetracycline and streptomycin and very high (>50%) for ampicillin and sulfamethoxazole–trimethoprim. Resistance to other critically important antimicrobials such as colistin and extended‐spectrum cephalosporins was 2.2% and 1.1%, respectively, and was encoded by the mcr‐1 and bla_SHV‐12 genes. Multidrug resistance was displayed by 80% of the resistant E. coli, and bla_SHV‐12 gene shared plasmid with other AMR genes. In general, resistance patterns in E. coli from vultures mirrored those found in pigs. Clostridioides difficile was detected in three samples (2.9%); two of them belonged to PCR ribotype 078 and one to PCR ribotype 126, both commonly found in pigs. All C. difficile isolates were characterized by a moderate‐to‐high level of resistance to fluoroquinolones and macrolides but susceptible to metronidazole or vancomycin, similar to what is usually found in C. difficile isolates from pigs. Thus, vultures may contribute somewhat to the environmental dissemination of some pig pathogens through their acquisition from pig carcasses and, more importantly, of AMR for antibiotics of critical importance for humans. However, the role of vultures would likely be much lesser than that of disposing pig carcasses at the SFS. The monitoring of AMR, and particularly of colistin‐resistant and ESBL‐producing E. coli, should be considered in pig farms used as sources of carcasses for SFS.     

Results of Gal‐Knockout Porcine Thymokidney Xenografts

Clinical transplantation for the treatment of end‐stage organ disease is limited by a shortage of donor organs. Successful xenotransplantation could immediately overcome this limitation. The development of homozygous α1,3‐galactosyltransferase knockout (GalT‐KO) pigs removed hyperacute rejection as the major immunologic hurdle to xenotransplantation. Nevertheless, GalT‐KO organs stimulate robust immunologic responses that are not prevented by immunosuppressive drugs. Murine studies show that recipient thymopoiesis in thymic xenografts induces xenotolerance. We transplanted life‐supporting composite thymokidneys (composite thymus and kidneys) prepared in GalT‐KO miniature swine to baboons in an attempt to induce tolerance in a preclinical xenotransplant model. Here, we report the results of seven xenogenic thymokidney transplants using a steroid‐free immunosuppressive regimen that eliminated whole‐body irradiation in all but one recipient. The regimen resulted in average recipient survival of over 50 days. This was associated with donor‐specific unresponsiveness in vitro and early baboon thymopoiesis in the porcine thymus tissue of these grafts, suggesting the development of T‐cell tolerance. The kidney grafts had no signs of cellular infiltration or deposition of IgG, and no grafts were lost due to rejection. These results show that xenogeneic thymus transplantation can support early primate thymopoiesis, which in turn may induce T‐cell tolerance to solid organ xenografts.     

Arthrite réactionnelle après infection par Clostridium difficile chez deux enfants

In adults, reactive arthritis (ReA) following Clostridium difficile-enterocolitis has been documented. In children, only one case of Clostridium difficile (C. difficile) associated ReA in a child has been reported. We now describe 2 other cases of ReA associated with C. difficile in children. The characteristics of ReA due to C. difficile appear to be similar in adults and children. Both children showed poly-arthritis after an episode of diarrhoea with positive stool cultures for C. difficile. Arthritis was asymmetrical with a self-limiting course. Nonsteroidal anti-inflammatory drug (NSAID) therapy was sufficient. One case is remarkable because of its prolonged course of ReA despite NSAID therapy, and its association with the presence of HLA B27 antigen.     

Seasonality of Clostridium difficile in the natural environment

Clostridium difficile is considered the leading cause of antibiotic‐associated disease worldwide. In the past decade, a large number of studies have focused on identifying the main sources of contamination in order to elucidate the complete life cycle of the infection. Hospitals, animals and retail foods have been considered as potential vectors. However, the prevalence of C. difficile in these types of samples was found to be rather low, suggesting that other contamination routes must exist. This study explores the presence of C. difficile in the natural environment and the seasonal dynamics of the bacterium. C. difficile was isolated from a total of 45 samples out of 112 collected (40.2%) on 56 sampling points. A total of 17 points were positive only during the winter sampling (30.4%), 10 were positive only during the summer sampling (17.9%) and 9 sampling points (16.1%) were positive in both summer sampling and winter sampling. Spore counts in soil samples ranged between 50 and 250 cfu/g for 24.4% of the positive samples, with the highest concentrations detected in samples collected in the forest during winter campaign (200–250 cfu/g). A total of 17 different PCR ribotypes were identified, and 15 of them had the genes coding for toxins A and B. Most of those ribotypes had not previously been found or had been isolated only sporadically (<1% of samples) from hospitals in Belgium. Regarding antimicrobial susceptibility, most of the resistant strains were found during the summer campaign. These findings bear out that C. difficile is present in the natural environment, where the bacterium undergoes seasonal variations.     

<Emphasis Type="Italic">Clostridium difficile</Emphasis> Infection: A Surgical Disease in Evolution

Introduction  Several recent publications suggest an increase in the incidence of Clostridium difficile colitis. However, such studies commonly lack denominators over which to index this rise. There is also concern in the literature that disease virulence is increasing. Methods  Billing, admission, operative, and infection databases at a single tertiary care center identified patients admitted from 1990 to 2006 with a diagnosis of C. difficile infection. Grouped by era, case numbers were indexed against overall hospital, operative, and laboratory volumes. C. difficile colectomy cases were individually examined and analyzed. Results  The number of hospitalized patients diagnosed with C. difficile colitis increased in a linear fashion during the study period (1990, 14 cases; 2006, 927 cases). The colectomy per C. difficile case ratio did not change over the study period (era 1, 0.17%; era 2, 0.20%; era 3, 0.16%). Thirteen patients underwent colectomy with 54% surviving. The increase in patients admitted with a diagnosis of C. difficile was significantly associated with hospital volume (p = 0.04), operative volume (p < 0.001), and lab testing volume (p = 0.008). Conclusion  The number of C. difficile patients admitted to our hospital is rising at an alarming rate. This reflects national trends and urgent action seems warranted to prevent a C. difficile epidemic. Originally Presented at: Digestive Disease Week, May 2007, Washington D.C.     

<Emphasis Type="Italic">Clostridium difficile</Emphasis> Enteritis: An Early Postoperative Complication in Inflammatory Bowel Disease Patients After Colectomy

Clostridium difficile, the leading cause of hospital-acquired diarrhea, is known to cause severe colitis. C. difficile small bowel enteritis is rare (14 case reports) with mortality rates ranging from 60 to 83%. C. difficile has increased in incidence particularly among patients with inflammatory bowel disease. This case series of six patients from 2004 to 2006 is the largest in the literature. All patients received antibiotics before colectomies for ulcerative colitis and developed severe enteritis that was C. difficile toxin positive. Three patients underwent ileal pouch anal anastomosis and loop ileostomy. Four of the six patients had C. difficile colitis before colectomy. Presenting symptoms were high volume watery ileostomy output followed by ileus in five of six patients. Four of the six patients presented with fever and elevated WBC. Five of the six developed complications requiring further surgery or prolonged hospitalization. Patients were treated with intravenous hydration and metronidazole then converted to oral metronidazole and/or vancomycin. None of the patients died. A high suspicion of C. difficile enteritis in patients with inflammatory bowel disease and history of C. difficile colitis may lead to more rapid diagnosis, aggressive treatment, and improved outcomes for patients with C. difficile enteritis.     

The Introduction of Human Heme Oxygenase‐1 and Soluble Tumor Necrosis Factor‐α Receptor Type I With Human IgG1 Fc in Porcine Islets Prolongs Islet Xenograft Survival in Humanized Mice

Apoptosis during engraftment and inflammation induce poor islet xenograft survival. We aimed to determine whether overexpression of human heme oxygenase‐1 (HO‐1) or soluble tumor necrosis factor‐α receptor type I with human IgG₁ Fc (sTNF‐αR‐Fc) in porcine islets could improve islet xenograft survival. Adult porcine islets were transduced with adenovirus containing human HO‐1, sTNF‐αR‐Fc, sTNF‐αR‐Fc/HO‐1 or green fluorescent protein (control). Humanized mice were generated by injecting human cord blood–derived CD34⁺ stem cells into NOD‐scid‐IL‐2Rγ^null mice. Both HO‐1 and sTNF‐αR‐Fc reduced islet apoptosis under in vitro hypoxia or cytokine stimuli and suppressed RANTES induction without compromising insulin secretion. Introduction of either gene into islets prolonged islet xenograft survival in pig‐to‐humanized mice transplantation. The sTNF‐αR‐Fc/HO‐1 group showed the best glucose tolerance. Target genes were successfully expressed in islet xenografts. Perigraft infiltration of macrophages and T cells was suppressed with decreased expression of RANTES, tumor necrosis factor‐α and IL‐6 in treatment groups; however, frequency of pig‐specific interferon‐γ–producing T cells was not decreased, and humoral response was not significant in any group. Early apoptosis of islet cells was suppressed in the treatment groups. In conclusion, overexpression of HO‐1 or sTNF‐αR‐Fc in porcine islets improved islet xenograft survival by suppressing both apoptosis and inflammation. HO‐1 or sTNF‐αR‐Fc transgenic pigs have potential for islet xenotransplantation.     

Experimental Evaluation of Faecal Escherichia coli and Hepatitis E Virus as Biological Indicators of Contacts Between Domestic Pigs and Eurasian Wild Boar

Domestic pigs and Eurasian wild boar (Sus scrofa) share several important viral and bacterial pathogens. Therefore, direct and indirect contacts between domestic pigs and wild boar present a risk of pathogen spillover and can lead to long‐term perpetuation of infection. Biological indicators could be a powerful tool to understand and characterize contacts between wild boar and domestic pigs. Here, faecal Escherichia coli and Hepatitis E virus (HEV) were explored as potential biological indicators under experimental conditions. The data gained in our pilot study suggest that faecal E. coli can be used as biological indicator of contact between wild boar and domestic pig. For HEV, faecal transmission was also confirmed. However, molecular studies on full‐genome basis did not reveal markers that would allow tracing of transmission direction. Based on these promising results, future field studies will especially target the practicability of E. coli microbiome molecular typing as surrogate of contacts at the wildlife–livestock interface.