Bovine tuberculosis (TB) is caused by infection with Mycobacterium bovis and closely related members of the Mycobacterium tuberculosis complex (MTC) [1]. These infections are shared between livestock, wildlife and sporadically, human beings. This re-emerging zoonotic disease is receiving increased attention from scientists and government agencies. Wildlife reservoirs exist worldwide and can interfere with cattle TB eradication efforts [2, 3]. There are three significant wildlife MTC maintenance hosts in Europe, namely the Eurasian badger (Meles meles), mainly in the United Kingdom (UK) and Republic of Ireland (RoI), the Eurasian wild boar (Sus scrofa), mainly in Mediterranean Spain and Portugal, and deer belonging to the subfamily Cervinae such as the red deer (Cervus elaphus) in several regions throughout Europe [4].

Based on habitat, climate features and wildlife population characteristics, Spain can be divided into six bioregions (BR) [5]; the humid and temperate Atlantic Spain would be represented by BR1, including Galicia and Asturias among other regions, along the north coast of the Iberian Peninsula (Figure 1). Continental Mediterranean habitats would be represented by BR3. In Spain, TB in cattle declined significantly from a herd prevalence of 2.24% in 2002 to 1.33% in 2011 [6]. However, the distribution of positive cattle herds in Spain, based on positive single tuberculin skin test confirmed by culture, is not uniform, with higher prevalence (up to 6%) in Mediterranean habitats of the south and west of the Spanish mainland and low (<1%) prevalence elsewhere (Table 1) [6]. In 2011 Asturias and Galicia reported cattle herd prevalences of 0.14 and 0.19%, respectively. However, both regions included hotspots (Local Veterinary Units) where cattle herd prevalence was >4%.

Wildlife monitoring is needed to identify changes in disease occurrence and to measure the impact of interventions [7, 8]. Wildlife Disease Surveillance Schemes have provided the tools for wildlife disease monitoring. These schemes have been in place in Atlantic Spain in Asturias and Galicia since 2001 and 2008, respectively, and have been pivotal for instance for assessing the TB situation of badgers in this region. The recent increase in awareness of TB in wildlife prompted surveillance of badgers in several regions and countries with suitable badger habitats in continental Europe, including Atlantic Spain. Research has estimated prevalence of M. bovis infection to be 12.4% in badgers in northern Spain, with spatial correlation of isolates from cattle and badgers [9, 10].

One recommendation for monitoring diseases in wildlife is selecting the most appropriate wildlife hosts [8]. Wild boar are considered the best TB surveillance target because of their wide habitat range, high abundance and easy accessibility for sampling as a game species. Moreover, wild boar are easy to test for TB because of their lesion distribution [11, 12] and the availability of specific and sensitive antibody tests [13]. MTC infection has been diagnosed in about 50% of wild boar from South-Western Spain [12, 14], and in 11.11% of wild boar from Portugal [15]. Outside the Iberian Peninsula, wild boar TB has been reported from eight European countries [4], including one local survey with a prevalence of 42% in Normandy (France) [16], and prevalences of 3% in northern Italy [17], and 3.6% in Switzerland and Liechtenstein [18]. This demonstrates an increasing importance of this ungulate regarding TB in Europe. However, no systematic studies in wild boar have been carried out in Atlantic Spain, although a seroprevalence between 0.01-5% was established in a recent cross-sectional serosurvey [19], confirming the wild boar as a sensitive host to MTC infection.

There are few reports of Mycobacterium avium complex (MAC) spp. infections in wild boar. M. avium subsp. hominissuis was isolated from two wild boar in Portugal [20] and M. avium subsp. avium from three and four wild boar in Italy and Spain, respectively [21, 22]. A recent study has reported a M. avium prevalence of 11% by culture in Southern Spain [23]. MAC infections in wild boar are of potential relevance due to the possible diagnostic interference with MTC infections by serology or pathology, particularly if culture and identification by polymerase chain reaction (PCR) is not performed.

The aims of the present study were (1) to document the MTC and MAC situation in wildlife of Atlantic Spain, with a focus on wild boar (considered a good target species) using serological, bacteriological, molecular and pathological diagnostic techniques, (2) to identify associations between infected wild boar populations, infected cattle and other infected domestic and wildlife species, and (3) to establish epidemiological and spatial links between MTC strains from domestic animals and wildlife in a particular region.

This is the first cross-sectional survey of wild boar TB in Atlantic habitats combining culture, pathology and serology. The results provide insight into the current status of wild boar as MTC hosts in temperate regions of continental Europe. The main findings were a low TB prevalence, a low proportion of MTC infected wild boar displaying generalized TB lesions, and a higher proportion of MAC infections.

This study provides much needed information on the TB status of wild boar in Atlantic Spain, confirming the presence of mycobacterial infections in these animal populations. The prevalence of MTC infections based on culture (2.59%) and serology (2.08%) was low. This is in agreement with the low prevalence of seropositive wild boar recorded in Atlantic habitats in a previous nation-wide survey (<5%) [19]. In contrast, a 52% MTC infection prevalence using culture was recorded in Doñana National Park in south-western Spain [2], and high infection prevalences have repeatedly been recorded in other regions of Mediterranean Spain and Portugal [4, 36]. The prevalence of antibodies against MTC was estimated at 35% in Mediterranean Spain (BR3) [19]. This large difference between the high prevalence in wild boar from Mediterranean regions and the low prevalence in those from Atlantic ones might be due to differences in the presence of several, not mutually exclusive risk factors for MTC infection in wildlife. First, in Atlantic Spain, the prevalence of TB in cattle herds is very low contrary to the high prevalence of TB in cattle in Mediterranean Spain (Table 1) [6]. Second, wild boar occur at higher densities in Mediterranean habitats than in Atlantic ones in Spain [37, 38]. Third, artificial management of wild boar for hunting purposes, including fencing and feeding, is much more common in Mediterranean habitats [14, 39]. Fourth, dry habitats in Mediterranean Spain facilitate aggregation of animals when watering or feeding (especially during the summer), enabling the transmission of the disease. Such spatial aggregation and intra- or inter-specific contacts at waterholes or feeders are less likely to occur in humid Atlantic habitats.

The proportion of lesions found in Atlantic Spain was markedly lower than in Mediterranean habitats. There may be different explanations for the low proportion of individuals with generalized disease observed in the present study (16.7%). The small sample size available for pathological study could be one of these reasons. This proportion contrasts with the much higher proportion found in Mediterranean Spain (57.8%) [11]. Interestingly, a recent study has demonstrated how the proportion of severe/generalized cases of TB among wild boar increases in dry years in southern Spain, although infection prevalence did not increase [Vicente J, unpublished observations]. This was probably a result of higher food restriction and TB susceptibility, or of augmented aggregation at waterholes. In our study, the prevalence of confirmed TB gross lesions was also low (8.57%) as compared to Mediterranean Spain (82.68%). Lesions were mainly restricted to small size ones in a single organ -the mandibular LN. Thus, it seems that the likelihood of MTC excretion by wild boar in Atlantic Spain is much lower than in Mediterranean habitats, and that most of the infected animals would shed small amounts of bacteria, if any.

Prevalence of histopathological lesions in Asturias was higher (9/70, 12.85%) than prevalence obtained by using culture (2/70, 2.86%). The histopathological findings suggest that an exhaustive pathological study of tissues increases prevalence and that we may have missed infected wild boar by only using culture. The gold standard for TB diagnosis is post-mortem examination with bacteriological confirmation [2]. The number of tissues examined and their pooling may have impaired the sensitivity of culture in this study. However, the average small size of the lesions found in Atlantic Spain has also to be taken into account since it may impair the isolation of mycobacteria. Sampling additional tissues would have increased the sensitivity, as shown for badgers [40].

In this study ELISA results showed a good agreement with culture and this assay has been a pivotal tool for the diagnosis and cost-effective large scale monitoring of TB in this area. Wild boar blood can be easily and cleanly collected [41], and ELISA has proven to be a useful first screening tool. Moreover, ELISA can also be used for detecting areas at risk for targeted surveys based on more sensitive techniques. If this approach is proposed in other Atlantic areas where wild boar TB is understudied, the bigger picture of the epidemiology in the wild boar TB could be revealed.

Wildlife and cattle in the same areas frequently share the same MTC strains [10, 14, 42]. The most frequent spoligotype found in cattle and badger in Spain was SB0121-M. bovis[10]. However, in this study the most frequent spoligotype in wild boar was SB0157-M. caprae (42.42%) and SB0121 was the least frequent. We identified six different M. bovis/M. caprae spoligotypes, four of which (SB0121-M. bovis, SB0130-M. bovis, SB0134-M. bovis and SB0157-M. caprae) had previously been identified in cattle, two (SB0152-M. bovis and SB0157-M. caprae) in goat and sheep and two (SB0121-M. bovis and SB0130-M. bovis) in fox. VNTR analysis contributed with additional information about the strains and demonstrated that cattle isolates with the SB0121, SB0130, SB0134, and SB0157 profiles were geographically associated with wild boar, suggesting an epidemiological link between these species. By contrast VNTR study of the isolates from the wild animal species and cattle with the SB0152-M. bovis did not find any link. The finding of a strain pattern in wild boar (SB0140-M. bovis) that was neither detected in cattle nor in other species during the study period would suggest some cases of independent MTC circulation among wild boar. This was not seen in badgers in previous studies [10]. Also, the identification of SB0157-M. caprae strain in wild boar (42.42%) more frequently than in cattle (0.16%) suggests some differences between the wild and domestic cycles of MTC epidemiology in this study area. However, it must be taken into account that SB0157-M. caprae was identified in one single “hot-spot” area where cattle, goats and sheep also shared this strain.

MAC species (M. avium subsp. avium and M. avium subsp. hominissuis) were identified in wild boar at similar prevalences than MTC species, as also happened with sympatric badgers [9]. This indicates circulation of these bacteria in the region in both species. Interestingly, these mycobacteria have not been identified in cattle in Atlantic Spain in such prevalences (see Table 2). In Mediterranean Spain, wild boar infection by MAC species is always lower than MTC [21, 23, 43]. Since these infections cause gross and histological lesions undistinguishable from TB, MAC infections must be considered when diagnosing MTC in wild boar from Atlantic habitats. As demonstrated in this study, conventional histology supplemented by immunohistochemistry to enhance specificity, would help to detect MTC in wild boar in future studies.

MTC is a multi-host pathogen modulated by host behaviour, abundance, exposure and susceptibility [4]. Whilst in Mediterranean Iberia this multi-host system is based on cattle, wild boar and red deer; in Atlantic Spain the multi-host system would include larger numbers of domestic animals (cattle, sheep and goat) and wildlife species (mainly badger and wild boar), although at lower TB prevalences. Risk factors for TB maintenance in wild boar like aggregation and high densities are not present in Atlantic Spain; but badgers are also infected in this region and could contribute to a ‘mixed-species reservoir’ sensu[44]; nevertheless, overall the existence of a wildlife reservoir such as in Mediterranean areas seems unlikely.

When TB lesions are observed in infected badgers, these occur often in the lungs (suggesting oro-nasal and fecal excretion) and kidneys (suggesting excretion by urine) [40]. Regarding wild boar, studies in high-prevalence regions have shown high proportions of generalized TB with lung involvement, strongly suggesting oro-nasal and fecal shedding [11]. In Atlantic Spain, 0.83% and 0% infected badgers showed visible lung and kidney lesions, respectively [9]. In this study 1.43% wild boar showed visible lung lesions, therefore suggesting a higher risk of transmission.

In any case, the presence of wildlife (wild boar and badger) contributes to the risk associated with cattle TB in farms in northern Spain [Cowie CE, unpublished observations], thus the risk of infection spillback from this species to cattle cannot be excluded. As for badgers, the risk of transmission to cattle might include direct contact on pasture and in buildings, and contamination of pasture, water and cattle feed [45]. Therefore, cautionary measures to prevent contact between cattle and wild boar, and gralloch removal after hunting should be carried out. Also, wild boar population monitoring as well as wild boar TB surveillance should be continued and improved.

This study confirms that (a) wild boar carry MTC mycobacteria at a low prevalence in Atlantic habitats in northern Spain, probably because of climate, management and because of the main reservoir, cattle, also have a low TB prevalence, (b) wild boar from this study area present less generalized lesions than in Mediterranean Spain, possibly due to lower TB prevalence in cattle, different management, habitat and climate conditions, (c) typing reveals similar profiles to those detected in other domestic – cattle, sheep and goat – and wildlife – badger and fox – hosts in this region, suggesting epidemiological links between these species and (d) our results are relevant in the current scenario of increasing wild boar densities [37] and sympatric wildlife TB hosts in Atlantic Spain [10] and elsewhere in continental Europe [18, 46]. To summarize, the wild boar is currently regarded as a spillover TB host in Atlantic Spain, although its status may change based on local or regional risk factors, such as an increase of cattle TB, wild boar abundance, land use changes and changes in cattle and wildlife management practices [14].