Received: 22 June 2020 | Revised: 30 July 2020 | Accepted: 26 August 2020 DOI: 10.1111/tbed.13816 ORIGINAL ARTICLE Phylogenetic relationships investigation of Mycobacterium caprae strains from sympatric wild boar and goats based on whole genome sequencing Giovanna Ciaravino1 | Enric Vidal2,3 | Martí Cortey1 | Maite Martín2,3 | 4 4 5 5 Albert Sanz | Irene Mercader | Claudia Perea | Suelee Robbe-Austerman | 1,2,3 2,3 Alberto Allepuz | Bernat Pérez de Val 1 Departament de Sanitat i Anatomia Animals, Universitat Autònoma de Abstract Barcelona, Bellaterra, Spain Tuberculosis (TB) in wildlife challenges epidemiological surveillance and disease con- 2 IRTA, Centre de Recerca en Sanitat Animal trol. An outbreak of TB was detected in a free-ranging wild boar population of a (CReSA, IRTA-UAB), Bellaterra, Spain 3 OIE Collaborating Centre for the Research Natural Park in Catalonia (Spain) and the outbreak investigation was conducted in and Control of Emerging and Re-emerging the area. During the study period (2015–2020), 278 wild boars were analysed by Swine Diseases in Europe (IRTA-CReSA), Bellaterra, Spain gross pathology, histopathology, mycobacterial culture and DVR-spoligotyping. In 4 Departament d’Agricultura, Ramaderia, addition, all cattle (49) and goat (47) herds of the area were tested with tuberculin Pesca i Alimentació de la Generalitat de skin test. TB compatible lesions were detected in 21 wild boars, and Mycobacterium Catalunya, Barcelona, Spain 5 caprae was isolated in 17 of them with two different spoligotypes: SB0415 (13) and National Veterinary Services Laboratories, United States Department of Agriculture, SB1908 (4). Only two goat herds showed TB positive animals that were subsequently Ames, IA, USA slaughtered. M. caprae with the spoligotypes SB0416 and SB0415 were isolated from Correspondence these animals. To investigate the phylogenetic relationships and the transmission Bernat Pérez de Val, IRTA, Centre de chain of the outbreak, nine strains isolated from six wild boars and three goats of the Recerca en Sanitat Animal, (CReSA, IRTA-UAB), Edifici CReSA, Campus UAB, study area were analysed by whole genome sequencing (WGS) followed by single 0193-Bellaterra, Barcelona, Catalonia, Spain. nucleotide polymorphism (SNP) analysis by maximum likelihood and median-joining Email: bernat.perez@irta.cat network inference methods. Results indicated that infected wild boars maintained M. caprae strains circulation in their own population and have likely transmitted the infection to goats, thus acting as TB reservoirs, compromising the success of live- stock TB eradication campaigns and posing a risk for public health. The results also highlighted the usefulness of WGS followed by SNP analysis in providing relevant epidemiological information when detailed contact data are missing. KEYWORDS epidemiology, goats, Mycobacterium caprae, tuberculosis, whole genome sequencing, wild boar Giovanna Ciaravino and Enric Vidal contributed equally to this article. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2020 The Authors. Transboundary and Emerging Diseases published by Wiley-VCH GmbH 1476 |  wileyonlinelibrary.com/journal/tbed Transbound Emerg Dis. 2021;68:1476–1486. CIARAVINO et al. | 1477 1 | I NTRO D U C TI O N 2 | M ATE R I A L S A N D M E TH O DS Tuberculosis (TB) is a chronic infectious disease caused by 2.1 | Context, study area and period Mycobacterium tuberculosis complex (MTBC) that affects a wide range of mammals, including humans, livestock and wildlife spe- In December 2015, a wild boar with TB suspicious lesions in me- cies. In particular, the control of zoonotic TB has been identified diastinal, tracheobronchial and pre-scapular lymph nodes (Ln) of as a major threat to achieve the goal of human TB eradication the Montseny Natural Park (Catalonia, Spain) was detected during (World Health Organization, 2018). The eradication of TB in cat- inspection in a game meat processing facility. Samples were submit- tle represents one of the leading expenditures on veterinary pro- ted to the laboratory of IRTA-CReSA through the Catalan slaugh- grammes co-funded by the European Union, although its complete terhouse support network—SESC (www.cresa.cat/blogs/​sesc) for eradication has not been achieved yet in many member states diagnostic investigation (Vidal et al., 2016). Histopathology showed (European Commission, 2020). TB compatible lesions (i.e. chronic granulomatous lymphadenitis) Despite many efforts have been focused on bovine TB erad- and M. caprae was isolated. DVR-spoligotyping was performed ac- ication programmes, the epidemiological surveillance of MTBC cording to a previously described method (Kamerbeek et al., 1997) circulation in a non-bovine wildlife–livestock interface is rarely and the spoligotype profile SB0415 was identified (www.Mbovis. addressed in a comprehensive way. It is well-known that both org) (Figure 1). wild boars and goats are reservoirs of MTBC in certain multi- After confirmation of this case, the TB outbreak investiga- host contexts, hindering the eradication of bovine TB (Guta tion was performed in the affected area in both wild boar and et al., 2014; Napp et al., 2013; Naranjo, Gortazar, Vicente, & de ­livestock. The study area (SA) was defined as the whole Natural la Fuente, 2008). Goats are particularly susceptible to TB in- Park and the surrounding area (Figure 1). This is a woodland fection (Crawshaw et al., 2008; Daniel et al., 2009; Pérez de area with coastal Mediterranean climate with a high number of Val et al., 2011) and caprine TB is rapidly spread from infected ­livestock farms, well-preserved wildlife habitats and without­ to susceptible animals within positive herds (Vidal et al., 2017), artificial fencing. Free-ranging wild boars are particularly abun- being a risk of infection for livestock (Cano-Terriza et al., 2018; dant in this area, with an estimated density over 12 individu- Napp et al., 2013; Vidal et al., 2018) and for humans (Rodríguez als/sq. km in a surface of 47,840 ha (Rosell, Pericas, Colomer, & et al., 2009). Mycobacterium caprae, a member of MTBC (Aranaz, Navàs, 2019). Cousins, Mateos, & Domínguez, 2003), is the main causative agent of goat TB in Spain (Rodríguez et al., 2011). On the other hand, infected wild boars cause an important fraction of new bovine 2.2 | Cattle and goat herds testing and sampling TB breakdowns in Spain (Guta et al., 2014; Hermoso de Mendoza et al., 2006). Investigation of all cattle and goat herds of the SA (49 and 47, re- Molecular tools to discriminate MTBC strains, such as spectively) was carried out by the Official Veterinary Services DVR-Spoligotyping (Kamerbeek et al., 1997) and multilocus through tuberculin skin testing (TST) between 2017 and 2018. TST variable number of tandem repeat analysis (MIRU-VNTR), was performed following the procedures and interpretations as pre- have been used worldwide as ancillary tools for animal TB scribed guidelines of the Spanish Ministry of Agriculture, Fisheries epidemiological surveillance and outbreak investigations in and Food (Bezos, Romero, de Juan, Gonzalez, & Sáez, 2019). All posi- multi-host scenarios (Boniotti et al., 2009; Hauer et al., 2015; tive animals were slaughtered and retropharyngeal, tracheobron- Rodriguez-Campos, Gonzalez, et al., 2012). However, these chial and mediastinal Ln or other tissues with visible TB compatible techniques have a rather restricted discriminatory capacity lesions were collected for further pathological and bacteriological since they analyse <1% of MTBC genome (Hauer et al., 2019). confirmation. Thus, they show limitations to detect genomic divergences of MTBC strains belonging to the same regional clonal complex, sometimes making it difficult to trace the origin of a given 2.3 | Wild boars inspection and sampling outbreak (Hauer et al., 2016; Rodriguez-Campos, Schürch, et al., 2012). Alternatively, the use of whole genome se- Intensified sampling and subsequent analysis of hunted wild boars quencing (WGS) has been extended in the last decade due of the SA were conducted during five consecutive hunting sea- to its higher discriminatory power to detect local transmis- sons (i.e. from September to March, between 2015 and 2020). The sion events missed by classical epidemiological investigations main source of tissue samples was game meat processing facili- (Biek et al., 2012). ties either through active sampling conducted by the specialized The aim of this study was to use WGS and epidemiological infor- personnel (official veterinarians, game meat processing facil- mation to investigate the relationships of cattle and goat herds with ity inspectors and forest guard officers) or through the submis- free-ranging wild boars in a TB outbreak area and gain insight into sion of TB suspicious lesions using the SESC platform by meat potential transmission patterns. inspectors (i.e. passive surveillance). Analyses were performed at 1478 | CIARAVINO et al. (a) (b) F I G U R E 1 Study area. (a) Map of Catalonia (North-eastern Spain) that shows the local densities of wild boars estimated for the hunting season 2018–19 (DARP 2019). The brown rectangle shows the outbreak area (14 wild boar/km2). (b) The studied outbreak area includes the Montseny Natural Park (light green) and surrounding areas. The different municipalities are delimited by dotted lines. The silhouette shows the approximate localization of the first detected wild boar with TB lesions and Mycobacterium caprae isolation (the spoligotype profile is indicated within the silhouette) [Colour figure can be viewed at wileyonlinelibrary.com] IRTA-CReSA laboratory. During the study period, 278 wild boars 2.6 | Whole genome sequencing were sampled (mainly head and tracheobronchial Ln), but also other tissues when TB compatible lesions were observed (e.g. A representative subset of M. caprae isolates from wild boar (N = 6) lungs or mesenteric Ln). and livestock (N = 3) of the SA, as well as other 43 M. caprae iso- lates from cattle (N = 18), goats (N = 22), sheep (N = 1) and wild boar (N = 2) in Catalonia (outside the SA) between 2008 and 2018, 2.4 | Pathological examination were sub-cultured in 7H11 plates (BD diagnostics) for 28 days at 37 ± 1°C. Growth was suspended in 1ml PBS in O-ring microtubes All tissues were analysed by gross lesion evaluation performed by and was heat-inactivated at 100°C for 30 min. Inactivated MTBC sectioning the sampled tissues 2–3 mm thick slices under appropri- samples were sent to the National Veterinary Services Laboratories, ate light conditions in a biosafety cabinet. In case of a positive or United States Department of Agriculture (USA) to conduct the non-conclusive result, histopathology (haematoxylin–eosin staining) DNA purification and whole genome sequencing (WGS) as previ- and Ziehl–Neelsen stain were carried out to confirm the diagnosis, ously described (Salvador et al., 2019). Briefly, DNA from bacterial and a fraction of the lesion was frozen for further mycobacterial cultures was collected; libraries were prepared with Nextera XT culture. A lesion was considered compatible with TB when granu- DNA Library Preparation Kit according to manufacturer's instruc- lomatous inflammatory infiltrate with variable areas of necrosis and tions, and sequenced in an Illumina MiSeq device using paired-end mineralization was observed. The presence of Langhan's cells and/or 250 bp segments. Sequences were mapped against the genome of acid fast bacilli was assessed but not indispensable for a TB compat- the reference strain M. bovis AF2122/97 (RefSeq ID NC_002945.4, ible diagnosis. BioProject PRJNA57695, (Malone et al., 2017)) using the Burrows– Wheeler aligner algorithm (BWA, (Li & Durbin, 2010)) and genome analysis toolkit 2.5.2 (GATK, (Auwera et al., 2013)), applying filter- 2.5 | Mycobacterial isolation and identification ing parameters and variant quality score recalibration according to GATK Best Practices recommendations. TB was confirmed by mycobacterial isolation in Löwenstein– Jensen with pyruvate and Coletsos solid media (BD Diagnostics) as previously described (Vidal et al., 2018), and BACTEC MGIT 2.7 | SNP identification and phylogenetic analyses 320 system (BD Diagnostics). Positive growths were stored in brain heart infusion with 20% glycerol at −80°C. Isolates were Single nucleotide polymorphisms (SNPs) that had quality scores then confirmed as MTBC by conventional multiplex PCR (Wilton (QUAL) < 150 and Allele Counts (AC) = 1 were removed, and SNPs & Cousins, 1992), and strains were subsequently typified by that fell into repeat regions (PPE, PE_PGRES gene families, trans- DVR-Spoligotyping. posable elements), or that were in loci with poor read density were CIARAVINO et al. | 1479 (a) (b) (c) (d) (e) F I G U R E 2 TB compatible lesions in wild boars. (a) Wild boar, Lung section. Miliary granulomatous pneumonia. This animal was likely excreting numerous mycobacteria. (b) Wild boar, submandibular lymph node section. Multifocal caseous granulomatous lymphadenitis. (c) Wild boar, lung (same as shown in a). Four neighbouring encapsulated granulomas in the lung parenchyma. H&E. Bar 250 μm. (d) Wild boar, submandibular lymph node. Chronic encapsulated granuloma in the lymph node parenchyma, mostly composed of necrosis, mineralization and minimal macrophage inflammatory infiltrate. H&E. Bar 100 μm. (e): Wild boar, submandibular lymph node. Cluster of multinucleated giant cells (Langhan's cells) diffusely infiltrating the lymph node parenchyma. H&E. Bar 25 μm [Colour figure can be viewed at wileyonlinelibrary.com] removed similarly to previously described (Bryant et al., 2013; Kato- maximum parsimony (MP) post-processing in order to remove all Maeda et al., 2013). SNPs were visually validated using Integrated superfluous median vectors and links that are not contained in Genomics Viewer (IGV, Broad Institute and the Regents of the the shortest trees of the network (Polzin & Daneshmand, 2003). University of California) (Robinson et al., 2011), and those that fell The MP post-processing calculation was available in Network in areas with low coverage, mapping issues or alignment problems Phylogenetic Software (version 4.6; available at www.fluxu​s-en- were manually filtered out. gin​e ering.com/share​n et.htm) which was also used to redraw the The evolutionary relationships among isolates were analysed by network. means of two different phylogenetic analyses; firstly, the concate- nated SNP sequences for all isolates were aligned, and a maximum likelihood phylogeny was constructed using a GTR-CAT model with 3 | R E S U LT S RaxML version 8 (Stamatakis, 2014). The tree was rooted using the genome sequence of M. bovis AF2122/97 reference strain. The con- 3.1 | Findings in livestock fidence of the internal branches of the tree was assessed by means of 1,000 Bootstrap replicates. Two out of the 47 goat herds showed positive animals to the TST, In second place, the whole genomic SNP sequences (i.e. one meat aptitude herd with extensive management (detected in DNA Alignment) were used to generate a median-joining net- December 2017) and one dairy herd with intensive management work (MJN) to further explore the genetic relatedness of iso- (detected in January 2018). The meat herd showed 4/241 (1.7%) lates (Bandelt, Forster, & Rohl, 1999). The MJN was constructed positive animals to TST, three of them were confirmed by culture with SplitsTree4 version 4.15.1 (Huson & Bryant, 2006) using and M. caprae SB0415 was identified. The dairy herd showed 37/255 the default parameters suggested by the program. The network (14.5%) positive animals to TST. All positive animals were slaugh- was also rooted by specifying the genome sequence of M. bovis tered and 14 of them were sampled for post-mortem examination. AF2122/97 as the out-group in order to facilitate its interpre- All examined animals showed TB compatible lesions and the isolated tation as a genealogical history. Thereafter, we computed the M. caprae strains all belonged to the unique spoligotype SB0416. 1480 | CIARAVINO et al. None of the 49 tested cattle herds showed any positive animal to (WB-3). Localizations of the SA strains with WGS and/or spoligotype the TST. information, as well as the localizations of the remaining TB cases detected in the SA, are shown in the map represented in Figure 4. Putative local transmission patterns of M. caprae from the 52 an- 3.2 | Findings in wild boar population alysed strains were inferred from the analysis of the phylogenetic relationships among taxa (Figure 5). The MJN results indicated that Twenty-one out of 278 (7.6%) wild boars showed TB compatible the six strains isolated from wild boars of the SA (from WB-1 to lesions confirmed by histopathology (Figure 2). Of those, 17 were WB-6) were highly related to each other and with the strain from confirmed by culture, identifying two distinct M. caprae spoligo- the meat goat (MG-1). These strains showed genetic relatedness type profiles: SB0415 (13) and SB1908 (4). No other mycobacteria also with other goat and cattle strains isolated in Catalonia between or other pathological agents were isolated from the remaining four 2008 and 2014 (mostly from meat-producing animals) (Figure 5). The samples with histologically TB compatible lesions. The majority of results also revealed the occurrence of transmission events inside cases with TB lesions were obtained during hunting seasons 2016– the SA, between wild boars and from wild boar to the meat goat 2017 and 2017–2018; the pathological and bacteriological annual (at end-point position along the network branch, forming the clade results in wild boar are summarized in Table 1. SA-II marked in red in Figure 5). In particular, the wild boar strain WB-6, isolated in 2015, showed a difference of five SNPs compared with the wild boar strain WB-4 (isolated in 2016) and a difference of 3.3 | Phylogenetic relationships and one SNP with the wild boar strain WB-1 (isolated in 2016); the latter transmission patterns (WB-1) was separated by seven SNPs from the meat goat strain (MG- 1), which was isolated in 2018 (Figure 5). Finally, the two dairy goat Six isolates from wild boars and three from goats (two dairy and one strains, sampled in the same herd, were genetically close to each meat) of the SA were selected for the phylogenetic analyses. Overall, other, differing only in 5 SNPs, but showed a distant genetic relation- the DNA of 52 M. caprae isolates (nine inside and 43 outside the ship to the rest of the strains isolated from the SA (Figure 5). Instead, SA) were sequenced by WGS and the SNP distances between the they were genetically closer to other M. caprae strains (mostly from isolates were analysed. dairy animals of Spanish origin) sampled from outside the SA in pre- The phylogenetic tree showed the separation of the strains in vious years (2009–2012). two major groups (A and B) (Figure 3). All the strains isolated from the SA were included in group A, whereas group B was clearly differ- entiated from group A and was mostly composed by strains isolated 4 | D I S CU S S I O N from fattening cattle imported from Romania. Within group A, three different clusters (A.1, A.2 and A.3) were defined on the basis of The MJN results indicate that seven out of the nine sequences number of SNPs separating each sequenced strain. This clustering from the SA might have a common genealogical origin, and then discriminated the two strains isolated from the dairy goat herd (clus- the infection spread through different animal populations within ter A.2, blue arrows in Figure 3), from the rest of the SA strains that the Natural Park. Even though additional data would be required were included in cluster A.3. A clade composed by the two dairy to establish precise evolutionary relationships among strains from goat strains was defined by 18 specific SNPs (clade SA-I). The spoli- the SA, the results suggest a local transmission of M. caprae from gotype SB0416 was identified in both of them. Within the cluster wild boars to a goat herd of meat aptitude and extensive manage- A.3., the meat goat and three wild boar strains of the SA formed ment. Indeed, the putative evolutionary history inferred from the another clade with genome sequences possessing 32 specific SNPs rooted MJN was congruent with the isolation dates of the strains (clade SA-II, red arrows in Figure 3). The spoligotype SB0415 was forming the clade SA-II (M. caprae strains were first isolated in wild identified in the four strains. Another branch of the cluster A.3 boars and later in the meat goat herd, see Figure 5). As far as we contained the remaining three wild boar strains of the SA (green ar- know, this is the first scientific study reporting infected wild boars rows in Figure 3). The spoligotype SB0415 was identified in two of as the most likely cause of a TB outbreak in goats. Local MTBC them (WB-2 and WB-5), whereas SB1908 was identified in the third hotspots in free-ranging wild boars were previously detected in TA B L E 1 Number of wild boar tissue samples analysed per hunting season in the study area 2019– Season 2015–2016 2016–2017 2017–2018 2018–2019 2020 No. of wild boar tissue samples 23 109 34 66 46 No. samples with TB visible lesions (%) 1 (4%) 13 (12%) 6 (18%) 1 (2%) 0 (0%) Isolated spoligopatterns (No. of isolates) Mycobacterium caprae M. caprae M. caprae M. caprae SB1908 - SB0415 (1) SB0415 (10) SB0415 (2) (1) No. of municipalities affected 1 8 5 1 0 CIARAVINO et al. | 1481 F I G U R E 3 Phylogenetic relationships among Mycobacterium caprae strains genomes. Rooted phylogenetic tree based on maximum likelihood model (GTR-CAT, RAxML) showing a SNP distance of the nine strains of the study area (SA) and 43 strains outside the SA isolated in Catalonia between 2008 and 2018. Root: Mycobacterium bovis AF 2122/97 reference strain sequence. Scale: 10 SNP. Two groups were identified: A, divided in three clusters (A.1, A.2 and A.3) and B. The nine strains of the SA are indicated by arrows as following: dairy goats forming the clade SA-I within the cluster A.2 (blue), wild boars and meat goat forming the clade SA-II within the cluster A.3 (red), and remaining wild boars of the cluster A.3 (green) [Colour figure can be viewed at wileyonlinelibrary.com] Catalonia in relatively restricted natural areas, and were epide- fast animal-to-animal transmission within the herd, not only by di- miologically associated with cattle TB breakdowns (Mentaberre rect contact through aerosols, but also by indirect oral contact by et al., 2014) or positive TST results (Pérez de Val et al., 2019). Now, using shared feeders (Vidal et al., 2017). Therefore, the low number our findings revealed a circulation of MTBC between non-bovine of positive goats detected by TST (four, only three with visible le- species but still posing a risk of infection for cattle herds of the sions and positive culture result) is consistent with a recent exoge- Natural Park that may interact with infected free-ranging wild nous entry of the disease in the herd. This recent transmission chain boars or share pastures with infected goats. could be also supported by the short genetic pairwise distance be- Data on TB status of the meat goat herd were unavailable prior tween strains (i.e. 7 SNPs between WB-1 and MG-1). Pairs differing to the first wild boar TB detection (in 2015), although the very low less than 6 SNPs are considered indicative of recent person-to-per- ratio of TST positive animals found in 2018 (1.7%) was strongly sug- son M. tuberculosis transmission (Nikolayevskyy, Kranzer, Niemann, & gestive of a recent infection, occurring after that date. M. caprae is Drobniewski, 2016), but this threshold to delineate epidemiological particularly adapted to domestic goats (Aranaz et al., 2003), showing links between strains could increase to 12 SNPs within 3 years of 1482 | CIARAVINO et al. F I G U R E 4 Localization of animal TB cases detected in the study area (SA). The map shows the approximate location of the animals from which the strains were isolated (precision level is the municipality, delimited by dotted lines). The number within the animal silhouettes indicates the spoligotype profile (when mycobacterial isolate was obtained) [Colour figure can be viewed at wileyonlinelibrary.com] evolution in human M. tuberculosis outbreaks (Walker et al., 2013) and A.3, Figure 5) across the whole SA, an extensive area with different after 5–6 years in cattle M. bovis outbreaks (Price-Carter et al., 2018). microhabitats, rather than being differentiated by local farm envi- The specific way of entry of the infection into the goat herd ronments (the SA involves almost 100 cattle and goats herds, to- is still unknown, although the indirect contact in contaminated gether). This finding suggests a common origin of M. caprae strains grazing areas, shared by both free-ranging wild boars and goats at a specific time point in the past (probably from infected live- in pasturage, seems the most likely source of the infection. In par- stock) and subsequent maintenance of M. caprae circulation in the ticular, aggregation points such as waterholes and baited places wild boar population of the SA, with mild genetic evolution within are considered hotspots for inter-species transmission of MTBC local sub-populations (i.e. clade SA-II formation). (Barasona, Torres, Aznar, Gortázar, & Vicente, 2017; Payne, The reduced genetic diversity observed among the 42 M. caprae Philipon, Hars, Dufour, & Gilot-Fromont, 2017). On the contrary, strains included in the group A can be explained by the selective WGS results displayed a low genetic relatedness of the two strains pressure carried out after years of bovine TB eradication campaigns of the dairy goat herd with the rest of the strains from the SA, based on test and slaughter of positive animals (Hauer et al., 2015). indicating a separate origin of TB outbreak without any apparent Yet, WGS analysis detected genetic variations of M. caprae strains interaction with wildlife reservoirs of the SA. The origin of this out- of the SA and local transmission patterns that had been missed by break is probably related to other dairy goats from outside of the the DVR-spolgotyping analysis alone, which, in the last two decades, SA (i.e. animal movements). The low SNP variation between the has been the most extended tool for molecular characterization two dairy goat strains (5 SNPs) is consistent with the intra-herd of MTBC isolated from animals worldwide (Haddad, Masselot, & spread of the infection, as previously described in M. caprae-in- Durand, 2004; Milian-Suazo et al., 2008; Rodríguez et al., 2010). fected cattle (Broeckl et al., 2017), which might be facilitated by Interestingly, the WGS also enabled to genetically associate the intensive management of the sampled herd. The persistence strains classified in apparently distant spoligotypes (i.e. SB1908 of MTBC lineages in the livestock–wildlife interface within a local presents deletions in 9 oligonucleotide spacers compared to environment of a farm is common and can be explained by direct or SB0415, www.Mbovis.org database), demonstrating that some indirect contact between farming animals and wildlife hosts (Biek of them were very much closely related in comparison with other et al., 2012). However, in this study, the M. caprae strains isolated strains, despite being classified within the same spoligotype (i.e. from wild boars tended to involve the same genetic lineage (cluster SB0415). These findings evidenced not only the lower discriminatory CIARAVINO et al. | 1483 F I G U R E 5 Evolutionary relationship of Mycobacterium caprae strains based on WGS. (a) Rooted median-joining network analysis performed with the 52 M. caprae strains isolated in Catalonia between 2008 and 2018 (left-upper insert). (b) In the zoomed main central image: Yellow circles show the nine strains of the study area (SA), black circles show the 43 strains isolated outside the SA. Root: Mycobacterium bovis AF2122/97 reference strain. Blue silhouettes indicate the dairy goat strains forming the clade SA-I within the cluster A.2; Red silhouettes indicate the wild boars and meat goat strains forming the clade SA-II within the cluster A.3; Green silhouettes indicate the remaining wild boars strains of the SA within the cluster A.3. The number of small perpendicular lines in each branch represents the SNPs separating sequences. Potential median vectors (hypothetical genotype) which connect the genome sequences are displayed as little circles in light grey. (c) The map (right-bottom insert) represents the suggested evolutionary scenario within the clade SA-II. Silhouettes represent the animal species, the year of M. caprae isolation is indicated below them and arrows indicate the transmission direction [Colour figure can be viewed at wileyonlinelibrary.com] power of conventional molecular characterization tools compared to 2018–19, respectively, (Rosell et al., 2019)). A previous study sug- WGS, but also inaccurate epidemiological associations in outbreak gested a compensatory reproduction effect of wild boars as a re- investigations. sponse to an increased hunting pressure (Mentaberre et al., 2014). Specific non-synonymous SNPs related with virulence were not However, a higher hunting rate could contribute to the removal investigated. Although previous studies showed that wild boars were of animals in advanced stages of the disease from the environ- even more susceptible to M. caprae than M. bovis strains (García- ment, thus limiting the transmission rates of M. caprae within the Jiménez et al., 2013), in the present study, the M. carpae transmission wild boar population. This might explain the reduction of apparent and expansion could be mainly driven by the high wild boar popula- prevalence based on post-mortem examination of visible lesions in tion densities (i.e. increased contact rates) rather than host adap- the two last hunting seasons included in the study. Nevertheless, tation of M. caprae strains. Indeed, Catalonia, and particularly the a previous serological study in a high prevalent area of TB in wild Natural Park of the TB outbreak, shows one of the highest wild boar boars showed that the increase of hunting pressure reduced the densities in Europe (Pittiglio, Khomenko, & Beltran-Alcrudo, 2018; prevalence of TB lesions in adult wild boars but increased the re- Rosell et al., 2019). production rates, resulting in a significant number of seropositive The increase in number of hunted animals in the SA after the piglets although they still had not developed TB visible lesions and outbreak detection did not reduce significantly the wild boar den- were negative to culture (Pérez de Val et al., 2017). Thus, the evo- sity (from 16 to 12 individuals/sq. km in seasons 2015–16 and lution of wild boar TB in the SA should be carefully monitored in 1484 | CIARAVINO et al. the upcoming seasons to confirm the apparent drop in TB preva- lence or its reemergence as a potential risk for livestock herds in Maite Martín https://orcid.org/0000-0002-3588-2838 contact with infected wild boars. Irene Mercader https://orcid.org/0000-0002-5857-8775 In summary, our findings revealed that infected wild boars were Claudia Perea https://orcid.org/0000-0002-4863-6863 able to maintain the mycobacteria in their own population and have Suelee Robbe-Austerman https://orcid. likely transmitted the infection to sympatric goats. The study ev- org/0000-0002-3794-3040 idenced local M. caprae circulation between wild boars and goats Alberto Allepuz https://orcid.org/0000-0003-3518-1991 without an apparent participation of cattle in the epidemiology, re- Bernat Pérez de Val https://orcid.org/0000-0003-3127-9182 inforcing the notion that animal TB should be approached as a multi- host disease which requires holistic control measures. The present REFERENCES study evidences the need of a thorough laboratory diagnosis and ep- Aranaz, A., Cousins, D., Mateos, A., Domínguez, L. Elevation of idemiological follow-up in non-bovine susceptible hosts since these Mycobacterium tuberculosis subsp. caprae Aranaz et al 1999 to spe- cies rank as Mycobacterium caprae comb. nov., sp. nov. International can maintain the infection and compromise the long-term success Journal of Systematic and Evolutionary Microbiology, 53(Pt 6), 1785– of bovine TB eradication programmes. The results also highlighted 1789. https://doi.org/10.1099/ijs.0.02532​- 0 the usefulness of WGS analysis in providing relevant epidemiological Auwera, G. A., Carneiro, M. O., Hartl, C., Poplin, R., del Angel, G., information when detailed contact data are missing or difficult to Levy-Moonshine, A., … DePristo, M. A. (2013). From FastQ data to high-confidence variant calls: The genome analysis toolkit best prac- obtain. tices pipeline. Current Protocols in Bioinformatics, 43(1), 1–11. https:// doi.org/10.1002/04712​50953.bi111​0s43 AC K N OW L E D G E M E N T S Bandelt, H. J., Forster, P., & Rohl, A. (1999). Median-joining networks for This work was supported by the Department of Agriculture, inferring intraspecific phylogenies. Molecular Biology and Evolution, 16(1), 37–48. https://doi.org/10.1093/oxfor​djour​nals.molbev. Livestock, Fisheries and Food (DARP) of the Government of a026036 Catalonia. IRTA is supported by Centres de Recerca de Catalunya Barasona, J. A., Torres, M. J., Aznar, J., Gortázar, C., & Vicente, J. (2017). (CERCA) Program/Generalitat de Catalunya (www.cerca.cat). DNA detection reveals Mycobacterium tuberculosis complex shedding DVR-spoligotyping was performed at VISAVET Health Surveillance routes in its wildlife reservoir the Eurasian wild boar. Transboundary Centre, Universidad Complutense de Madrid, Spain. Whole genome and Emerging Diseases, 64(3), 906–915. https://doi.org/10.1111/ tbed.12458 sequencing was performed at the National Veterinary Services Bezos, J., Romero, B., de Juan, L., Gonzalez, S., & Sáez, J. (2019). Manual Laboratories, United States Department of Agriculture, USA. de Procedimiento Sanitario: Realización de las pruebas de intradermo- We thank the forest guard body of the Generalitat de Catalunya, tuberculinización y gamma-interferón. 2019–2020. Madrid, Spain: Carniques Llorà S.L. and Senglar de Girona S.L., for their contri- Ministerio de Agricultura, Pesca y Alimentación (MAPA) y Centro de Vigilancia Sanitaria Veterinaria (VISAVET). Universidad Complutense bution in wild boar samples collection. We are also grateful to de Madrid. Retrieved from https://www.visav​et.es/data/mapa/man- Zoraida Cervera, Abel Muñoz, Sierra Espinar and Mónica Pérez for ual_proce​dimie​nto_IDTB_IFN_2019.pdf their outstanding technical assistance. Biek, R., O’Hare, A., Wright, D., Mallon, T., McCormick, C., Orton, R. J., … Kao, R. R. (2012). Whole genome sequencing reveals local trans- mission patterns of Mycobacterium bovis in sympatric cattle and C O N FL I C T O F I N T E R E S T badger populations. PLoS Pathogens, 8(11), e1003008. https://doi. The authors declare that they have no competing interests. org/10.1371/journ​al.ppat.1003008 Boniotti, M. B., Goria, M., Loda, D., Garrone, A., Benedetto, A., Mondo, E T H I C A L A P P R OVA L A., … Pacciarini, M. L. (2009). Molecular typing of Mycobacterium bovis strains isolated in Italy from 2000 to 2006 and evaluation of The authors confirm that the ethical policies of the journal, as noted variable-number tandem repeats for geographically optimized geno- on the journal's author guidelines page, have been adhered to. No typing. Journal of Clinical Microbiology, 47(3), 636–644. https://doi. ethical approval was required for this study as all sampling and han- org/10.1128/JCM.01192​- 08 dling procedures were conducted by authorized personnel according Broeckl, S., Krebs, S., Varadharajan, A., Straubinger, R. K., Blum, H., & Buettner, M. (2017). Investigation of intra-herd spread of to Spanish legislation (Royal Decree 2611/1996 and amendments) Mycobacterium caprae in cattle by generation and use of a whole-ge- and European Union laws for protection of animals used for scien- nome sequence. Veterinary Research Communications, 41(2), 113– tific purposes (2010/63/EU). 128. https://doi.org/10.1007/s1125​9-017-9679-8 Bryant, J. M., Harris, S. R., Parkhill, J., Dawson, R., Diacon, A. H., van DATA AVA I L A B I L I T Y S TAT E M E N T Helden, P., … Bentley, S. D. (2013). Whole-genome sequencing to establish relapse or re-infection with Mycobacterium tuberculosis: A The data that support the findings of this study are available from retrospective observational study. The Lancet Respiratory Medicine, the corresponding author upon reasonable request. 1(10), 786–792. https://doi.org/10.1016/S2213​-2600(13)70231​-5 Cano-Terriza, D., Risalde, M. A., Rodríguez-Hernández, P., Napp, S., ORCID Fernández-Morente, M., Moreno, I., … García-Bocanegra, I. (2018). Epidemiological surveillance of Mycobacterium tuberculosis com- Giovanna Ciaravino https://orcid.org/0000-0002-5796-8093 plex in extensively raised pigs in the south of Spain. Preventive Enric Vidal https://orcid.org/0000-0002-4965-3286 Veterinary Medicine, 159, 87–91. https://doi.org/10.1016/J.PREVE​ Martí Cortey https://orcid.org/0000-0003-1126-3423 TMED.2018.08.015 CIARAVINO et al. | 1485 Crawshaw, T., Daniel, R., Clifton-Hadley, R., Clark, J., Evans, H., Rolfe, determine the microevolution of Mycobacterium tuberculosis during S., & de la Rua-Domenech, R. (2008). TB in goats caused by an outbreak. PLoS One, 8(3), e58235. https://doi.org/10.1371/journ​ Mycobacterium bovis. The Veterinary Record, 163(4), 127. https://doi. al.pone.0058235 org/10.1136/vr.163.4.127 Li, H., & Durbin, R. (2010). Fast and accurate long-read alignment with Daniel, R., Evans, H., Rolfe, S., de la Rua-Domenech, R., Crawshaw, T., Burrows-Wheeler transform. Bioinformatics (Oxford, England), 26(5), Higgins, R. J., Clifton-Hadley, R. (2009). Outbreak of tuberculo- 589–595. https://doi.org/10.1093/bioin​forma​tics/btp698 sis caused by Mycobacterium bovis in golden Guernsey goats in Malone, K. M., Farrell, D., Stuber, T. P., Schubert, O. T., Aebersold, R., Great Britain. Veterinary Record, 165(12), 335–342. https://doi. Robbe-Austerman, S., & Gordon, S. V. (2017). Updated reference ge- org/10.1136/vr.165.12.335 nome sequence and annotation of Mycobacterium bovis AF2122/97. European Commission (2020). Outcome of the evaluation procedure of the Genome Announcements, 5(14), e00157-17. https://doi.org/10.1128/ eradication, control and surveillance programmes submitted by Member genom​eA.00157​-17 States for Union financial contribution for 2020: List of the programmes Mentaberre, G., Romero, B., de Juan, L., Navarro-González, N., Velarde, technically approved and amount allocated to each programme. R., Mateos, A., … Serrano, E. (2014). Long-term assessment of wild Brussels, Belgium. Retrieved from https://ec.europa.eu/food/sites/​ boar harvesting and cattle removal for bovine tuberculosis control food/files/​safet​y/docs/cff_animal_vet-progs_guida​nce_progs_erad_ in free ranging populations. PLoS One, 9(2), e88824. https://doi. wd-12728​-2019.pdf org/10.1371/journ​al.pone.0088824 García-Jiménez, W. L., Benítez-Medina, J. M., Fernández-Llario, P., Milian-Suazo, F., Harris, B., Díaz, C. A., Romero Torres, C., Stuber, T., Abecia, J. A., García-Sánchez, A., Martínez, R., … Hermoso de Ojeda, G. A., … Payeur, J. B. (2008). Molecular epidemiology of Mendoza, J. (2013). Comparative pathology of the natural infections Mycobacterium bovis: Usefulness in international trade. Preventive by Mycobacterium bovis and by Mycobacterium caprae in Wild Boar Veterinary Medicine, 87(3–4), 261–271. https://doi.org/10.1016/j. (Sus scrofa). Transboundary and Emerging Diseases, 60(2), 102–109. preve​tmed.2008.04.004 https://doi.org/10.1111/j.1865-1682.2012.01321.x Napp, S., Allepuz, A., Mercader, I., Nofrarías, M., López-Soria, S., Guta, S., Casal, J., Napp, S., Saez, J. L., Garcia-Saenz, A., Perez De Val, Domingo, M., … Pérez de Val, B. (2013). Evidence of goats acting as B., … Allepuz, A. (2014). Epidemiological investigation of bovine domestic reservoirs of bovine tuberculosis. The Veterinary Record, tuberculosis herd breakdowns in Spain 2009/2011. PLoS One, 9(8), 172(25), 663. https://doi.org/10.1136/vr.101347 e104383. https://doi.org/10.1371/journ​al.pone.0104383 Naranjo, V., Gortazar, C., Vicente, J., & de la Fuente, J. (2008). Evidence Haddad, N., Masselot, M., & Durand, B. (2004). Molecular differentia- of the role of European wild boar as a reservoir of Mycobacterium tion of Mycobacterium bovis isolates. Review of main techniques and tuberculosis complex. Veterinary Microbiology, 127(1–2), 1–9. https:// applications. Research in Veterinary Science, 76(1), 1–18. https://doi. doi.org/10.1016/j.vetmic.2007.10.002 org/10.1016/S0034​-5288(03)00078​-X Nikolayevskyy, V., Kranzer, K., Niemann, S., & Drobniewski, F. (2016). Hauer, A., De Cruz, K., Cochard, T., Godreuil, S., Karoui, C., Henault, Whole genome sequencing of Mycobacterium tuberculosis for de- S., … Boschiroli, M. L. (2015). Genetic evolution of Mycobacterium tection of recent transmission and tracing outbreaks: A system- bovis causing tuberculosis in livestock and wildlife in France since atic review. Tuberculosis, 98, 77–85. https://doi.org/10.1016/j. 1978. PLoS One, 10(2), e0117103. https://doi.org/10.1371/journ​ tube.2016.02.009 al.pone.0117103 Payne, A., Philipon, S., Hars, J., Dufour, B., & Gilot-Fromont, E. (2017). Hauer, A., Michelet, L., Cochard, T., Branger, M., Nunez, J., Boschiroli, Wildlife Interactions on baited places and waterholes in a French M.-L., & Biet, F. (2019). Accurate phylogenetic relationships among area infected by bovine tuberculosis. Frontiers in Veterinary Science, Mycobacterium bovis strains circulating in France based on whole 3, 122. https://doi.org/10.3389/fvets.2016.00122 genome sequencing and single nucleotide polymorphism anal- Pérez de Val, B., López-Soria, S., Nofrarías, M., Martín, M., Vordermeier, ysis. Frontiers in Microbiology, 10, 955. https://doi.org/10.3389/ H. M., Villarreal-Ramos, B., … Domingo, M. (2011). Experimental fmicb.2019.00955 model of tuberculosis in the domestic goat after endobronchial in- Hauer, A., Michelet, L., De Cruz, K., Cochard, T., Branger, M., Karoui, C., fection with Mycobacterium caprae. Clinical and Vaccine Immunology, … Boschiroli, M. L. (2016). MIRU-VNTR allelic variability depends 18(11), https://doi.org/10.1128/CVI.05323​-11 on Mycobacterium bovis clonal group identity. Infection, Genetics and Pérez de Val, B., Napp, S., Velarde, R., Lavín, S., Cervera, Z., Singh, M., … Evolution: Journal of Molecular Epidemiology and Evolutionary Genetics Mentaberre, G. (2017). Serological follow-up of tuberculosis in a wild in Infectious Diseases, 45, 165–169. https://doi.org/10.1016/j. boar population in contact with infected cattle. Transboundary and meegid.2016.08.038 Emerging Diseases, 64(1), https://doi.org/10.1111/tbed.12368 Hermoso de Mendoza, J., Parra, A., Tato, A., Alonso, J., Rey, J., Peña, Pérez de Val, B., Sanz, A., Soler, M., Allepuz, A., Michelet, L., Boschiroli, J., Hermoso de Mendoza, M. (2006). Bovine tuberculosis in wild M. L., & Vidal, E. (2019). Mycobacterium microti Infection in free-rang- boar (Sus scrofa), red deer (Cervus elaphus) and cattle (Bos taurus) ing Wild Boar, Spain, 2017–2019. Emerging Infectious Diseases, 25(11), in a Mediterranean ecosystem (1992–2004). Preventive Veterinary 2152–2154. https://doi.org/10.3201/eid25​11.190746 Medicine, 74(2-3), 239–247. https://doi.org/10.1016/j.preve​ Pittiglio, C., Khomenko, S., & Beltran-Alcrudo, D. (2018). Wild boar tmed.2005.10.005 mapping using population-density statistics: From polygons to high Huson, D. H., & Bryant, D. (2006). Application of phylogenetic networks resolution raster maps. PLoS One, 13(5), e0193295. https://doi. in evolutionary studies. Molecular Biology and Evolution, 23(2), 254– org/10.1371/journ​al.pone.0193295 267. https://doi.org/10.1093/molbe​v/msj030 Polzin, T., & Daneshmand, S. V. (2003). On Steiner trees and minimum Kamerbeek, J., Schouls, L., Kolk, A., van Agterveld, M., van spanning trees in hypergraphs. Operations Research Letters, 31(1), Soolingen, D., Kuijper, S., … van Embden, J. (1997). Simultaneous 12–20. https://doi.org/10.1016/S0167​-6377(02)00185​-2 detection and strain differentiation of Mycobacterium tu- Price-Carter, M., Brauning, R., de Lisle, G. W., Livingstone, P., Neill, M., berculosis for diagnosis and epidemiology. Journal of Clinical Sinclair, J., … Collins, D. M. (2018). Whole genome sequencing for Microbiology, 35(4), 907–914. Retrieved from https://jcm-asm- determining the source of Mycobacterium bovis infections in livestock org.are.uab.cat/conte​nt/35/4/907.long https://doi.org/10.1128/ herds and wildlife in New Zealand. Frontiers in Veterinary Science, 5, JCM.35.4.907-914.1997 272. https://doi.org/10.3389/fvets.2018.00272 Kato-Maeda, M., Ho, C., Passarelli, B., Banaei, N., Grinsdale, J., Flores, Robinson, J. T., Thorvaldsdóttir, H., Winckler, W., Guttman, M., Lander, L., … Hopewell, P. C. (2013). Use of whole genome sequencing to E. S., Getz, G., & Mesirov, J. P. (2011). Integrative genomics viewer. 1486 | CIARAVINO et al. Nature Biotechnology, 29(1), 24–26. https://doi.org/10.1038/ USA. Molecular Ecology, 28(9), 2192–2205. https://doi.org/10.1111/ nbt.1754 mec.15061 Rodríguez, E., Sánchez, L. P., Pérez, S., Herrera, L., Jiménez, M. S., Samper, Stamatakis, A. (2014). RAxML version 8: A tool for phylogenetic anal- S., & Iglesias, M. J. (2009). Human tuberculosis due to Mycobacterium ysis and post-analysis of large phylogenies. Bioinformatics (Oxford, bovis and M. caprae in Spain, 2004–2007. The International Journal of England), 30(9), 1312–1313. https://doi.org/10.1093/bioin​forma​tics/ Tuberculosis and Lung Disease: The Official Journal of the International btu033 Union against Tuberculosis and Lung Disease, 13(12), 1536–1541. Vidal, E., Arrieta-Villegas, C., Grasa, M., Mercader, I., Domingo, M., Retrieved from http://www.ncbi.nlm.nih.gov/pubme​d/19919773 & Pérez de Val, B. (2017). Field evaluation of the efficacy of Rodríguez, S., Bezos, J., Romero, B., de Juan, L., Álvarez, J., Castellanos, Mycobacterium bovis BCG vaccine against tuberculosis in goats. E., … Aranaz, A. (2011). Mycobacterium caprae infection in livestock BMC Veterinary Research, 13(1), 252. https://doi.org/10.1186/s1291​ and wildlife, Spain. Emerging Infectious Diseases, 17(3), 532–535. 7-017-1182-5 https://doi.org/10.3201/eid17​03.100618 Vidal, E., Grasa, M., Perálvarez, T., Martín, M., Mercader, I., & Pérez de Rodríguez, S., Romero, B., Bezos, J., de Juan, L., Álvarez, J., Castellanos, Val, B. (2018). Transmission of tuberculosis caused by Mycobacterium E., … Aranaz, A. (2010). High spoligotype diversity within a caprae between dairy sheep and goats. Small Ruminant Research, 158, Mycobacterium bovis population: Clues to understanding the demog- 22–25. https://doi.org/10.1016/j.small​rumres.2017.11.010 raphy of the pathogen in Europe. Veterinary Microbiology, 141(1–2), Vidal, E., Tolosa, E., Espinar, S., de Val, B. P., Nofrarías, M., Alba, A., … 89–95. https://doi.org/10.1016/j.vetmic.2009.08.007 Domingo, M. (2016). Six-year follow-up of slaughterhouse sur- Rodriguez-Campos, S., Gonzalez, S., de Juan, L., Romero, B., Bezos, veillance (2008–2013): The Catalan slaughterhouse support net- J., Casal, C., … Surveillance, S. N. (2012). A database for animal work (SESC). Veterinary Pathology, 53(3), 532–544. https://doi. tuberculosis (mycoDB.es) within the context of the Spanish na- org/10.1177/03009​85815​593125 tional programme for eradication of bovine tuberculosis. Infection, Walker, T. M., Ip, C. L., Harrell, R. H., Evans, J. T., Kapatai, G., Dedicoat, Genetics and Evolution, 12(4), 877–882. https://doi.org/10.1016/j. M. J., … Peto, T. E. (2013). Whole-genome sequencing to delineate meegid.2011.10.008 Mycobacterium tuberculosis outbreaks: A retrospective observational Rodriguez-Campos, S., Schürch, A. C., Dale, J., Lohan, A. J., Cunha, M. study. The Lancet Infectious Diseases, 13(2), 137–146. https://doi. V., Botelho, A., … Smith, N. H. (2012). European 2 – A clonal complex org/10.1016/S1473​-3099(12)70277​-3 of Mycobacterium bovis dominant in the Iberian Peninsula. Infection, Wilton, S., & Cousins, D. (1992). Detection and identification of multi- Genetics and Evolution, 12(4), 866–872. https://doi.org/10.1016/j. ple mycobacterial pathogens by DNA amplification in a single tube. meegid.2011.09.004 Genome Research, 1(4), 269–273. https://doi.org/10.1101/gr.1.4.269 Rosell, C., Pericas, B., Colomer, J., & Navàs, F. (2019). Programa de se- World Health Organization (2018). GLOBAL TUBERCULOSIS REPORT. guiment de les poblacions de senglar a Catalunya. Temporada 2018– Geneva, Switzerland. Retrieved from https://apps.who.int/iris/bitst​ 2019. Barcelona, Spain: Departament d’Agricultura, Ramadria, Pesca ream/handl​e/10665/​27445​3/97892​41565​6 46-eng.pdf?ua=1&ua=1 i Alimentació (DARP) de la Generalitat de Catalunya. Department d’Agricultura, Ramaderia, Pesca i Alimentació. Retrieved from http:// agric​u ltura.gencat.cat/web/.conte​n t/06-medi-natur​a l/caca/enlla​ How to cite this article: Ciaravino G, Vidal E, Cortey M, et al. cos-docum​e nts/infor​m es-tecni​c s/progr​a ma-segui​m ent-pobla​cions​ -sengl​ar-sus-scrof​a/fitxe​rs-binar​is/segui​ment_sengl​ar_cat_2019.pdf Phylogenetic relationships investigation of Mycobacterium Salvador, L. C. M., O’Brien, D. J., Cosgrove, M. K., Stuber, T. P., Schooley, caprae strains from sympatric wild boar and goats based on A. M., Crispell, J., … Kao, R. R. (2019). Disease management at the whole genome sequencing. Transbound Emerg Dis. wildlife-livestock interface: Using whole-genome sequencing to 2021;68:1476–1486. https://doi.org/10.1111/tbed.13816 study the role of elk in Mycobacterium bovis transmission in Michigan,