Mycobacterial infections in wild boars (Sus scrofa) from Southern Switzerland: Diagnostic improvements, epidemiological situation and zoonotic potential

Abstract The occurrence of mycobacterial infections in different hosts and their implication as obligate or opportunistic pathogens remain mainly unclear. In addition to the well‐known pathogenic members of the Mycobacterium tuberculosis – complex (MTBC), over 180 non‐tuberculous mycobacteria (NTM) species have been described. Although the large majority of the NTM is assumed to be non‐pathogenic to most individuals, an increasing trend in NTM infections has been observed over the last decades. The reasons of such augmentation are probably more than one: improved laboratory diagnostics, an increasing number of immunocompromised patients and individuals with lung damage are some of the possible aspects. Mandibular lymph nodes of 176 hunted wild boars from the pre‐Alpine region of Canton Ticino, Switzerland, were collected. Following gross inspection, each lymph node was subjected to culture and to an IS6110 based real‐time PCR specific for MTBC members. Histology was performed of a selection of lymph nodes (n = 14) presenting gross visible lesions. Moreover, accuracy of matrix‐assisted laser desorption ionization time‐of‐flight mass spectrometry (MALDI‐TOF MS) species identification was compared with sequence analysis of a combination of housekeeping genes. Mycobacteria of the MTBC were detected in 2.8% of the wild boars (n = 5; CI95% 1.2–6.5) and were all confirmed to be Mycobacterium microti by molecular methods. In addition, based on the examined lymph nodes, NTM were detected in 57.4% (n = 101; CI95% 50.0–64.5) of the wild boars originating from the study area. The 111 isolates belonged to 24 known species and three potentially undescribed Mycobacterium species. M. avium subsp. hominissuis thereby predominated (22.5%) and was found in lymph nodes with and without macroscopic changes. Overall, the present findings show that, with the exception of undescribed Mycobacterium species where identification was not possible (3.6%; 4/111), MALDI‐TOF MS had a high concordance rate (90.1%; 100/111 isolates) to the sequence‐based reference method.


| INTRODUC TI ON
Bovine tuberculosis (bTB) is a chronic disease caused by members of the Mycobacterium tuberculosis complex (MTBC; Rodriguez-Campos, Smith, Boniotti, & Aranaz, 2014). MTBC have been isolated from numerous different domestic and wild animal species.
Recent epidemiological investigations have shown the fundamental role played by wildlife in the maintenance of the causal agents of bTB. This results in continuous interspecies transmissions from wild animals to livestock and vice versa, hindering national and international eradication programs (Atkins & Robinson, 2013;Fink et al., 2015;Garcia-Jimenez et al., 2016;Nigsch, Glawischnig, Bago, & Greber, 2018). Badger (Meles meles), free-ranging red deer (Cervus elaphus elaphus) and wild boar (Sus scrofa) are the most relevant known wild animals acting as a reservoir of bTB in Europe.
The ongoing geographic expansion of wild boar populations has raised concerns regarding the monitoring of several infectious diseases, including zoonotic plagues like bTB and hepatitis E Martinelli et al., 2015). Beside the members of the MTBC and Mycobacterium leprae, the agent that causes Hansen's disease, over 180 species of non-tuberculous mycobacteria (NTM) have been described (Gupta, Lo, & Son, 2018). NTM are commonly encountered in the environment and they have been isolated from a variety of sources, including water, feed, soil, dust, aerosol, protozoa and animals (Falkinham, 2015;Ghielmetti et al., 2018). Of these, two species are recognized as true pathogens for humans, namely M. marinum and M. ulcerans (Johansen, Herrmann, & Kremer, 2020). Nevertheless, more than 60 species of NTM are known to be opportunistic pathogenic to humans and other mammals, and infections with these emerging pathogens are now more common than tuberculosis in industrialized countries (Biet & Boschiroli, 2014;Griffith et al., 2007;Tortoli, 2014).
Immunocompromised individuals are highly susceptible to opportunistic NTM infections and improved laboratory diagnostics have enabled more accurate detection of fastidious or extremely slow growing species. Despite the increasing relevance of mycobacterial infections, only restricted information on the occurrence and their diversity in wildlife is available. Moreover, although wild boars are among the most widely distributed large mammals in the world (Oliver, IUCN/SSC Pigs, and Peccaries Specialist Group, & IUCN/SSC Hippo Specialist Group, 1993), the literature concerning this species is mainly focused on the presence of MTBC and the impact of NTM infections on the prevalence of MTBC Chiari et al., 2016;Di Marco et al., 2012;Michelet et al., 2015;Naranjo, Gortazar, Vicente, & de la Fuente, 2008;Richomme, Boschiroli, Hars, Casabianca, & Ducrot, 2010;Santos et al., 2009;Vicente et al., 2006). Canton Ticino is the most southern Canton of Switzerland, and a large proportion of its borders is shared with Italy. The territory encompasses an area of 2,812 km 2 , where the majority of the urban area is concentrated in the flat land and forests cover about one third of the alpine region.
The wild boar presence in the territory has been documented during the XVI century. Thereafter, it disappeared and it is only since 1981 that it has been officially sighted again (Dipartimento Recent studies from Spain, Czech Republic, Brazil and Slovenia comprehensively evaluated the spectrum of NTM species in black pigs using molecular methods (Garcia-Jimenez et al., 2015;Gortazar et al., 2011;Munoz-Mendoza et al., 2013;Pate et al., 2016;Trcka et al., 2006). It is significant to note that, the latter mentioned publications, describe marked differences in the spectrum of species isolated. Such differences may not be exclusively the result of geographic distribution of NTM. The advance in molecular techniques and the progress in mycobacterial characterization led to enormous diagnostic improvements over the past decades (Tortoli, 2014). It is noteworthy that the dissection of the M. avium complex (MAC) in the mentioned publications was performed at different levels, impeding a direct comparison of the isolated mycobacteria. The characterization of NTM from clinical samples is often a challenge for laboratory personal in routine diagnostic.
Because of its rapidness, cost-effectiveness, and high throughput, the matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) technology has nowadays been integrated in the workflow of numerous diagnostic laboratories (Alcaide et al., 2018;Mediavilla-Gradolph et al., 2015;Murugaiyan et al., 2018). However, the accuracy achievable at present with genetic approaches remains superior to the MALDI-TOF MS-based species identification (Tortoli, 2014). Therefore, the accuracy and limitations of this method based on ordinary samples from veterinary origin should be evaluated. The present research used a panel of NTM showing a wide range of species isolated from a common source and verified the consensus grade between sequence analysis and MALDI-TOF MS. This study aimed (a) to determine the occurrence and diversity of mycobacterial species among healthy wild boar hunted in the Canton of Ticino, (b) to identify the geographical distribution of mycobacterial species and (c) to compare

| Collection of samples
Hunting seasons in Switzerland are regulated independently by the Cantonal Veterinary Offices. In total, 1,436 and 1,588 wild boars were hunted in the Canton of Ticino in 2017 and 2018, respectively. Mandibular lymph nodes of 86 and 90 animals were collected during the two hunting periods of the study and the weight of every sampled wild boar was determined. Assuming an estimated population size of 2,500-3,000 animals, approximatively 3% of the population was analysed for two subsequent years.
Sex, age and the exact geographical position of each animal were recorded directly on the field. Age classification was based on tooth eruption patterns: animals <6 months of age were recorded as juveniles, those between 6 months and 2 years of age as yearlings, and adult animals older than 2 years composed the last group. Trained personnel of the Cantonal Veterinary Officeexcised lymph nodes by using sterile dissection knives and scalpels. Each sample was immediately packed into sterile containers and transported to the laboratory under cooled conditions.

| Microbiological procedures
Samples preparation and mycobacterial cultures were performed as described elsewhere (Ghielmetti et al., 2018). Briefly, BBL MGIT liquid media tubes supplemented with Bactec MGIT 960 growth supplement, BBL MGIT PANTA (Polymyxin B, Amphotericin B, Nalidixic acid, Trimethoprim, Azlocillin) antibiotic mixture (Becton, Dickinson, BD) and 50 μg/ml sodium-pyruvate were each inoculated with 0.5 ml of decontaminated and homogenized specimen. In addition, one Löwenstein-Jensen and one Stonebrink agar slants (BD) were inoculated with the same inoculum and incubated up to 8 weeks at 37°C. In order to obtain pure mycobacterial cultures, subcultures on 7H10 agar-plates and on Stonebrink agar slants (BD) were performed at intervals of three to ten days. Simultaneously, 400 μl of culture inoculum was suspended in 100 μl ATL buffer (Qiagen) and transferred onto a Lysing Matrix E tubes (MP Biomedicals).
Genomic DNA was extracted through mechanical cell lysis using a TissueLyser II (Qiagen) and enzymatic digestion with Proteinase K (Qiagen) overnight. Automated DNA purification was performed using the QIAcube instrument in accordance with the QIAamp cador Pathogen Mini Kit protocol (Qiagen). DNA concentration in the final eluate was measured by reading the absorbance at 260 nm using a NanoDrop 2000c Spectrophotometer (Thermo Fisher Scientific), diluted to a maximal concentration of 100 ng/µl and stored at −20°C until use. Purified DNA was used for direct MTBC detection using the qPCR assay targeting insertion sequence IS6110 as described by Reed et al. with slight modifications (Reed et al., 2016). Briefly, the MTB IS6110 probe was double-quenched with iQ500 and BHQ1 instead of ZEN and 3IABkFQ, respectively. Moreover, the qPCR assay internal control was substituted by eGFP as described by Hoffmann et al. and performed on a 7500 Fast real-time PCR system (Applied Biosystems; Hoffmann, Depner, Schirrmeier, & Beer, 2006). DNA from cultured mycobacteria was extracted by inoculating a loop-full of cell materials into 200 μl of chelating ion-exchange resin (InstaGene Matrix) and centrifuged at 13,000 g for 10 min. The supernatant was used in downstream reactions.
Pure cultures that presented acid-fast bacilli (AFBs) by Ziehl-Neelsen (ZN) staining and negative MTBC qPCR results were classified as NTM and further characterized by sequence analysis of a combination of housekeeping genes and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Sanger sequencing of 16S rRNA (Scherrer, Landolt, Carroli, & Stephan, 2018), rpoB (Adékambi, Colson, & Drancourt, 2003) and hsp65 (Telenti et al., 1993) housekeeping genes was performed in duplicates followed by gene homology analyses. For isolates identified as members of the MAC, the complete hsp65 gene was sequenced as proposed by Turenne et al. using primers MAChsp65F and MAChsp65R (Turenne, Semret, Cousins, Collins, & Behr, 2006 real-time PCR, mycobacteria could not be cultivated from any of the described culture media, even after 12 months of incubation. Molecular characterization by mycobacterial interspersed repetitive unit and variable number tandem repeats (MIRU-VNTR) and species determination using spoligotyping were therefore performed using DNA extracts of lymph nodes as previously described (Ghielmetti et al., 2017).

| MALDI-TOF mass spectrometry
Inactivation and preparation of the isolates for MALDI-TOF MS analysis was performed using the Mycobacteria Extraction Method (MycoEX) in accordance with the manufacturer. In order to enable optical evaluation of the tested colonies and because the quality of the spectra obtained from isolates grown on solid media is better than those obtained from liquid media, 7H10 agar plates were chosen as culture medium for MALDI-TOF MS analysis (Kodana et al., 2016;Lotz et al., 2010). A loopful of culture from solid medium was transferred into a 1.5 ml Eppendorf tube with 300 μl of HPLC-water and inactivated for 30 min at 99°C under biosafety level 3 conditions. After a centrifugation step of 2 min at 13,000 g, the supernatant was discharged the pellet was re-suspended in 300 μl of HPLC-water and 900 μl of ethanol.
Thereafter, centrifugation was repeated, and the supernatant was discharged. The tubes were left open enabling the pellets to dry at room temperature. A spatula-tip full of bead suspension (Zirconia/Silica; BioSpec) and 10-50 μl of acetonitrile were added to the pellets, depending on the volume of the pellets.
Mycobacterial cells were disrupted by vortexing at maximal speed for 1 min and 25-50 μl of 70% formic acid were added, depending on the volume of the pellets. In conclusion, the tubes were centrifuged at the same conditions as above and 1 μl of the supernatant was spotted on the MALDI-TOF target plates (MSP 96 target ground steel; Bruker Daltonics) in duplicates. At this point, the target plates were allowed to dry at room temperature and then taken to biosafety level 2 conditions. Thereafter, 1 μl of matrix was added to each spot (HCCA, α-cyano-4-hydroxycinnamic acid).
Peptide mass spectra were acquired in a linear positive ion mode at a maximum laser frequency of 60 Hz across a mass to charge ratio (m/z) of 2,000 to 20,000 Da using the Microflex LT

| Macroscopic and histological examination
After removal of fat and connective tissue, qualified staff inspected the lymph nodes macroscopically and recorded pathological changes. A subset of samples was additionally submitted for histology (n = 14). Selection criteria were focusing on infections with the seven most prevalent Mycobacterium species and presenting macroscopic lesions. In addition, all lymph nodes that tested positive for MTBC by real-time PCR were submitted for histological examination (n = 5). Samples were fixed in 10% buffered formalin and embedded in paraffin. Two to three-micron-thick tissue sections were obtained and stained with haematoxylin and eosin (HE).
Additionally, in cases where lesions consistent with mycobacterial infections were observed after HE staining, an additional ZN staining was performed.

| Geographical distribution and statistical analysis
The geographical distribution of lymph nodes showing growth of mycobacteria and the circulation of the different NTM on the territory were investigated using the free software QGIS Desktop 3.6.1. Based on the collected data regarding sex and age of the animals, statistical analysis was performed using GraphPad Prism 8.2.1 (GraphPad Software). Fisher's exact test was used to evaluate different age groups and the presence of viable NTM isolated from their lymph nodes. Moreover, a possible association between the isolation of MAC and the three age groups was investigated with the same test. Statistical significance was set to p < .05.

| Ethics statement
All animal samples used in this study originated from legally hunted wild boars in accordance with the Swiss legislation (Hunting Law SR 922 and Animal Welfare Act SR 455). An ethical approval or permit for animal experimentation was not applicable.  (Table 1). Three single isolates could not be classified to any known mycobacterial species and may represent new species. M. avium subsp. hominissuis (Mah) predominated with 22.5% of the isolates, followed by M. nonchromogenicum with 21.6% of the isolates.

| RE SULTS
Five samples derived from three juvenile and two adult wild boars were positive by MTBC real-time PCR. Since no growth of mycobacteria over a 12-month incubation period could be achieved, species identification by direct spoligotyping using extracted DNA from the lymph nodes was performed. All five samples presented the same spoligotype signature SB0118, characterized by the presence of spacers 37-38 (www.Mbovis.org). The same signature is also known in the international spoligotyping database SpolDB4 as ST 539 and is indicative for M. microti (Brudey et al., 2006). Mycobacterial co-infections were de- Adult wild boars were more prone to be infected with mycobacteria in comparison with juvenile animals or yearlings (Figure 2a).
Overall, 74.4% of the analysed lymph nodes originating from adult animals showed growth of mycobacteria. Only 54.9% and 45.8% of the juveniles and yearlings presented viable NTM, respectively (p < .05). A correlation between infected animals and their sex was not found, nor a significant association between the isolation of MAC and the three age groups (juveniles vs yearlings p = .388; juveniles vs adults p = .079; yearlings vs adults p = .384; Figure 2b).
The distribution of the different NTM circulating among the analysed animals was investigated based on the geographical data collected.
The hunted wild pig population was divided into seven districts of the study area and, with the exception of the three northern districts where a low number of samples were obtained; a homogeneous distribution of the NTM cultured is shown in the remaining four districts (Figure 1b).

| Macroscopic and histological examination
Of the lymph nodes from the 101 wild boars showing growth of Mycobacterium spp., 25.7% showed macroscopic pathological lesions (Table 2). Macroscopic visible lesions such as single or TA B L E 1 Identification of non-tuberculous mycobacteria isolated from wild boars and human clinical relevance. A total of 111 nontuberculous mycobacteria isolates belonging to 24 known species and three potentially undescribed Mycobacterium species were cultured from 176 wild boar mandibular lymph nodes. Sequence analysis was used as gold standard method. Results of identifications based on MALDI-TOF MS applying two different log scores value cut-offs are shown. Discrepant results intended as (a) no species identification by MALDI-TOF analysis or (b) assignment of a discrepant species compared to sequence analysis are highlighted

F I G U R E 2
Prevalence of non-tuberculous mycobacteria in relation to three different age groups (juvenile, yearling and adult). (a) Adult wild boars were more prone to be infected with non-tuberculous mycobacteria (74.4%) in comparison with juvenile animals (54.9%) or yearlings (45.8%). No significant difference was observed between juvenile and yearlings (p = .3785). **, p ≤ .01; *, p ≤ .05 (compared to all other age groups and as indicated). (b) The prevalence of Mycobacterium avium complex in relation to age groups is shown. No statistically significant correlation between MAC and age of the animals was observed in the present survey which is commonly associated with this species (van Soolingen et al., 1998). Acid-fast bacilli were observed extracellular and within macrophages (Figure 3b).

| Mycobacterium avium complex
The predominant species identified in this study belonged to the MAC (Table 1) (Turenne et al., 2006). M. avium subsp. hominissuis has the broadest host range compared to the other members of the MAC; nevertheless, a clear differentiation between environmental and host-specific members of the MAC is necessary to better understand its distribution, host-adaptation and clinical implications (Turenne et al., 2006). Because NTM infections are not notifiable to public health authorities in most countries, data regarding the incidence and prevalence of diseases caused by these agents are lacking. Identification of potential infection sources however, is of great relevance. Among the two species of NTM that have been most frequently isolated in the present study (M. avium and M. nonchromogenicum), MAC members have been described as the most common cause of mycobacteriosis in human in Northern Europe (Hoefsloot et al., 2013), Japan (Nishiuchi, Iwamoto, & Maruyama, 2017), Korea (Ko et al., 2018) and North America (Boyle, Zembower, Reddy, & Qi, 2015). Recently, possible MAC reservoirs and infection sources for humans and animals including drinking water, bathrooms and hot tubs have been investigated using molecular analyses (Eisenberg et al., 2012;Falkinham, Iseman, de Haas, & van Soolingen, 2008;Hilborn et al., 2008). Noteworthy, a close genetic relatedness between human and swine isolates has been reported (Johansen et al., 2007;Mobius et al., 2006).

| Prevalence of Mycobacteria in relation to age
Adult wild boars were more prone to be infected with mycobacteria in comparison with juvenile animals or yearlings (Figure 2a

| Wild boar and Mycobacteria
Wild boars (Sus scrofa) are among the most widely distributed large mammals worldwide. Their natural range extends from Western Europe and the Mediterranean basin to Eastern Russia, Japan and South-East Asia (Massei et al., 2015). The high mobility associated with the highest reproductive capacity among ungulates enable an annual population growth rate of 250% under favourable food and weather conditions (Ebert, Knauer, Spielberger, Thiele, & Hohmann, 2012;Gethoffer, Sodeikat, & Pohlmeyer, 2007).
Supplemental feeding of wild ungulates is prohibited by law in several Swiss Cantons, including Ticino. Despite this order and the annual population reduction by hunters, the population in the study area is rising, augmenting the animal-to-animal as well as the animalto-human contact probabilities.
Since the hunted animals of the present study were judged to be in an overall healthy condition and suitable for human consumption, the Mah infection rate of 14% observed in the mandibular lymph node is noteworthy and represents a veterinary public health concern. Mandibular lymph nodes have been described to be the preferred entry point for mycobacteria in wild boars (Queiros et al., 2019). This is probably due to their eating habits allowing the intake of environmental contaminations and infected feed sources.
After oronasal infection, viable microorganisms often concentrate in mandibular lymph nodes and from there are eliminated, persist or eventually disseminate throughout other organ systems. Mandibular lymph nodes are the most likely organ for visible lesion caused by MTBC in adult animals and in a relevant proportion of cases this is the only organ affected (Dondo et al., 2007;Martin-Hernando et al., 2007;Munoz-Mendoza et al., 2013).
Only few publications extensively investigated the presence of NTM in wild boar (Garcia-Jimenez et al., 2015;Gortazar et al., 2011;Pate et al., 2016;Trcka et al., 2006).  (Table 1). It is therefore fundamental to implement advanced approaches in order to identify the exact species involved and estimate their relevance as potential infection source for consumers.
In contrast to packs, which tend to use only a small portion of their territory and move to another range at regular intervals, the adult wild boar male has a single territory that can range up to 50 km 2 . These individuals are able to move across the entire territory at daily basis, playing a crucial role in the spread of animal and zoonotic pathogens (Nugent, Gortazar, & Knowles, 2015;Schulz et al., 2019;Spitz, 1992). This high mobility may also be one of the

| CON CLUS ION
In conclusion, a remarkable number of mandibular lymph nodes collected from wild boars presented viable mycobacteria. Although the zoonotic risk of the isolated NTM remains unclear, it must be emphasized that the hunted animals were intended for human consumption and among the isolated species (n = 24), the large majority (n = 18) has been described as human pathogens. In addition, the present findings show that MALDI-TOF MS has a high concordance rate to the reference method and because of his rapidness, cost-effectiveness and high throughput represent a valid diagnostic tool for identification of NTM species in veterinary medicine.

ACK N OWLED G M ENTS
The authors are most indebted to all the hunters and game officers from the Canton of Ticino; without their cooperation, this study would not have been possible. We also gratefully acknowledge Ella Hübschke and Fenja Rademacher for their skilled technical assistance.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.