Background Wild boar is an important reservoir of Mycobacterium tuberculosis variant bovis, the main causative agent of bovine tuberculosis (bTB). A proportion of tuberculosis (TB)-affected wild boars shed M tuberculosis by nasal route, favouring the maintenance of bTB in a multihost scenario. The aim of this work was to assess if M tuberculosis nasal excretion is influenced by factors commonly associated with high TB prevalence in wild boar.
Methods TB diagnosis and M tuberculosis isolation were carried out in 112 hunted wild boars from mid-western Spain. The association between the presence of M tuberculosis DNA in nasal secretions and explanatory factors was explored using partial least squares regression (PLSR) approaches.
Results DNA from M tuberculosis was detected in 40.8 per cent nasal secretions of the TB-affected animals. Explanatory factors provided a first significant PLSR X’s component, explaining 25.70 per cent of the variability observed in M tuberculosis nasal shedding. The presence of M tuberculosis in nasal secretions is more probable in animals suffering from generalised TB and mainly coinfected with Metastrongylus species and porcine circovirus type 2, explaining nearly 90 per cent of the total variance of this model.
Conclusion Measures aiming to control these factors could be useful to reduce M tuberculosis shedding in wild boar.
- wildlife management
Statistics from Altmetric.com
Bovine tuberculosis (bTB) is an infectious disease which has led to important economic losses worldwide. bTB is mainly caused by the recently reclassified Mycobacterium tuberculosis variant bovis, formerly called M bovis within the Mycobacterium tuberculosis complex.1 Today, the disease is a significant concern in many countries throughout the world.2 In Spain, the reported prevalence of bTB in cattle from some regions (central and south-western Spain) is notably higher than those required to achieve the officially bTB-free status (bTB herd prevalence <0.1 per cent for 6 consecutive years).3 This has been associated with the interaction with wild reservoirs of M tuberculosis var bovis such as wild boar and red deer.4
The wild boar is considered the main M tuberculosis var bovis wildlife reservoir in Mediterranean ecosystems of southern Portugal and south-western Spain, with a tuberculosis (TB) prevalence higher than 50 per cent.5 The risk factors that can predispose the wild boar to a high prevalence of TB have been widely studied. Factors such as high population density, age or coinfections with other pathogens seem to be related to a higher prevalence of M tuberculosis var bovis infection in wild boar.6–8
In recent studies, it has been shown that a percentage of TB-affected wild boars (36.2–50.4 per cent) shed M tuberculosis DNA (formerly referred as M tuberculosis complex) by different routes (oral, nasal, faecal or urinary).9 10 These animals, especially those called ‘super shedders’, which shed high amount of bacteria by one or more routes,9 raise the probabilities of new infections (through direct and indirect contagions) and favour the maintenance of TB in a multihost scenario.
The reduction in the number of shedders could lessen the horizontal transmission and environmental contamination, which could lead to a decline in TB incidence.9 However, a good knowledge about factors that may favour M tuberculosis shedding in wild boar is necessary to propose effective measures focused on reducing the number of shedders.
The aim of this work was to assess if M tuberculosis nasal excretion is influenced by factors related to high TB prevalence in wild boar, such as TB severity, coinfection with other pathogens, and individual features such as age and sex.
Materials and methods
This study was conducted on 20 wild boar game estates in mid-western Spain where a total of 112 randomly selected hunted wild boars were analysed between October 2011 and February 2013. The sex and age of these animals were determined (when it was possible) following standard procedures.11
Postmortem examination of all animals was performed in the field as previously described,12 with a detailed macroscopic inspection to assess the presence of TB-like visible lesions affecting the lymph nodes (submandibular, retropharyngeal, mediastinal and mesenteric lymph nodes), and thoracic and abdominal organs. Head lymph nodes (submandibular and/or retropharyngeal), lungs (when feasible) and blood samples from each animal were systematically recovered and transported to the laboratory at 4°C. When TB-like visible lesions were detected in other organs, they were also recovered and transported to the laboratory.
To detect the presence of M tuberculosis, a microbiological culture from head lymph nodes, and from a piece of caudal lung lobes (and from other affected organs if they were collected) from each animal, was carried out following standard procedures as previously described.13 Suspicious colonies obtained from microbiological cultures were identified as M tuberculosis by PCR and ‘spoligotyped’ following standard methods.14 15
Depending on microbiological results wild boars were classified into two groups: animals with localised lesion patterns (M tuberculosis was detected in one location, mainly head lymph nodes) and animals with generalised lesion patterns (M tuberculosis was detected in head lymph nodes and other organs, eg, lung, mesenteric lymph nodes and/or spleen).7
Detection of M tuberculosis DNA in nasal samples
To assess the presence of M tuberculosis DNA in nasal secretions, mucous material was recovered from both nasal fossae of each animal using a sterile swab. These nasal swabs were transported in Amies transport medium (Deltalab, Rubí, Catalonia, Spain) at 4°C to the laboratory. Amies transport medium was previously verified as suitable for preserving the viability of Mycobacterium species.16
Nasal swabs were incubated in 1 ml of PBS 1x during 24 hours at 4°C, and DNA from resultant solutions was extracted using a commercial kit (QIAamp DNA Mini Kit, Qiagen, Crawley, West Sussex, UK) following the manufacturer’s protocol. To determine the presence of M tuberculosis DNA, a specific real-time PCR assay targeting the IS1561’ locus was carried out using a pair of primers (forward: 5’- GATCCAGGCCGAGAGAATCTG-3’; reverse: 5’-GGACAAAAGCTTCGCCAAAA-3’) and a probe (5’-FAM ACGGCGTTGATCCGATTCCGC TAMRA-3’) (Eurogentec, Seraing, Belgium), as previously described,17 with an Applied Biosystems 7300 thermocycler (Thermo Fisher Scientific, Waltham, Massachusetts, USA). This insertion sequence is present in all the Mycobacterium tuberculosis complex members except in M tuberculosis var microti.18 19 Duplicate reactions were run for template samples (DNA extracted from nasal swabs), M tuberculosis var bovis (SB0121), and non-template samples were used as controls. Furthermore, in order to discard the presence of PCR inhibitors in nasal samples included in this study, a TaqMan base real-time PCR to detect β-actin gene was carried out using previously described primers, probes and conditions.20
Concomitant pathogens assessment
Contact with a set of pathogens that have been previously related to TB development in wild boar (porcine circovirus type 2 (PCV-2), Metastrongylus species and Aujeszky disease virus (ADV))7 was assessed in studied animals. To detect PCV-2 infections, DNA from head lymph nodes was extracted using a commercial QIAamp DNA Mini Kit (Qiagen) following the manufacturer’s recommendations. A PCV-2-specific PCR was carried out using previously extracted DNA as a template.21 The presence of specific antibodies against ADV was tested using a commercial ELISA kit for pigs and following the manufacturer’s recommendations to differentiate seropositive and seronegative animals (INgezim ADV Total; positive threshold=sample optical density (OD)/positive control OD >0.35). The existence of Metastrongylus species in the lungs was evaluated following previously described procedures.22
The association between the detection of M tuberculosis DNA in nasal fossae and the explanatory factors proposed was assessed by a partial least squares regression (PLSR) approach. This statistical tool is an extension of multiple regression analysis that generates X’s components using the explanatory variables proposed to maximise the variance explained.23 This technique deals with multicollinearity, allowing the proposal of closely related variables as explanatory factors. The optimal number of components used to generate the regression model is obtained using cross-validation principles with Q 2 as an indicator.24
In the present study, the authors ran a model in which the presence (1)/absence (0) of M tuberculosis DNA in nasal samples was considered as a single response variable. On the other hand, TB pattern (0=localised, 1=generalised), sex (male or female), age (measured in years) and contact (negative (0), positive (1)) with other pathogens (PCV-2, ADV and Metastrongylus species) were proposed as explanatory variables. The model was built using TB-affected wild boars with information about all the variables proposed (n=39). Results obtained with PLSR show the relative importance (using relative weight) of each explanatory variable in components generated to explain the response variable (presence/absence of M tuberculosis in nasal secretions). Furthermore, bootstrap resampling techniques were used to generate 95 per cent confidence intervals to elucidate which of the explanatory variables were a statistically significant weight in model components. Finally, the explanatory power of the obtained model could be estimated as the percentage of variance explained. The ‘plspm’ library V.0.3.724 of R V.3.0.3 software was used for these analyses.
M tuberculosis was isolated from 49 animals (43.8 per cent) showing 11 different spoligotype patterns, all of them belonging to the variant bovis of M tuberculosis. According to the M tuberculosis var bovis isolates obtained, 24 of the TB-affected wild boars showed a localised pattern (49 per cent) and 25 animals showed generalised TB lesions (51 per cent).
Individualised information about TB diagnosis (lesions, spoligotypes and so on) can be found in online supplementary file 1. Mean ages and sex rate found in M tuberculosis var bovis infected and non-infected animals are summarised in table 1.
M tuberculosis DNA detection and quantification in nasal samples
M tuberculosis DNA was detected in 20 (40.8 per cent) out of the 49 TB-affected wild boars analysed, whereas the remaining 63 TB-free animals resulted negative to the real-time PCR. Of the animals in which M tuberculosis DNA was detected, 73.6 per cent showed a generalised TB pattern with lesions in head lymph nodes and lungs.
Concomitant pathogens assessment
Whereas PCV-2 DNA was found in 44 wild boars (39.6 per cent), Metastrongylus species was detected in 48 (47.5 per cent) out of the 101 wild boars that could be studied (animals whose complete lungs could be recovered). Furthermore, a total of 84 animals were seropositive against ADV (76.4 per cent).
In the PLSR analysis, all the explanatory variables proposed provided a first statistically significant PLSR X’s component, explaining 25.7 per cent of the observed variability. The first component obtained a value of 0.12, which is considered significant.25 Almost 90 per cent of the total variance explained by this first component was due to explanatory variable showing a significant relative weight (confidence interval avoiding 0): TB severity and infections with Metastrongylus species and PCV-2. The weights of these variables had different signs (table 2). A positive correlation of M tuberculosis DNA detection in nasal samples was found with the severity of TB, infection with PCV-2 and Metastrongylus species, and seropositivity against ADV. Age and sex showed less influence in this component.
The results obtained in this work improve the knowledge about M tuberculosis shedding in wild boar, showing that the probabilities of detecting M tuberculosis DNA in nasal secretions from TB-affected wild boars are influenced by factors related to high prevalence of TB in wild boar populations.
M tuberculosis DNA was detected in nasal secretions from 40.8 per cent of the infected wild boars. This is slightly higher than the results obtained in a recent study which found an 18.8 per cent TB-affected wild boars shedding M tuberculosis DNA by nasal route, using molecular techniques.10 However, the proportion of shedders detected in this work is not an extreme value, since it is between the range of percentages of TB-affected wild boars shedding M tuberculosis DNA by any route (36.2–50.4 per cent).9 10 Furthermore, because the lungs were the most common organ where TB generalised lesions were found (see online supplementary file 1), nasal route is expected to be one of the most important excretion routes in animals studied, representing a high percentage of total shedders.
According to the present study’s PLSR model, M tuberculosis DNA nasal shedding in wild boar could be affected by several factors, explaining more than the 25 per cent of the probabilities that this pathogen can be detected in nasal swabs. Wild boar suffering from generalised TB patterns and coinfected with PCV-2 and Metastrongylus species seems to have a greater likelihood of shedding M tuberculosis DNA by nasal routes.
TB severity has already been positively correlated with M tuberculosis shedding in wild boar10 and other animal species such as cattle and South American camelids.26 27 TB generalised patterns may be related to a temporary or permanent inefficiency of cellular immune responses, with poor contention of pathology progression,28 and hence favouring bacterial dissemination and Mycobacterium species shedding. The results obtained in the PLSR model confirmed that TB severity is a key predisposing factor to wild boars becoming M tuberculosis shedders.
Coinfections with pathogens previously related with TB development in wild boar (Metastrongylus species and PCV-2) also enhanced the probabilities of shedding M tuberculosis DNA by nasal route in wild boar, supposing 48 per cent of the total variance explained by the PLSR model. These pathogens have been previously associated with TB severity and TB prevalence in wild boar7 29 and seem to play a relevant role in the development of TB in this species. Infections with both pathogens could impair the development of an effective Th1 immune response,30–32 promoting more severe TB lesions and mycobacterial shedding. However, this hypothesis needs to be confirmed assessing immunological parameters in coinfected and non-coinfected animals.
Previous studies did not find a statistical relationship between M tuberculosis shedding in wild boar and other factors such as age or gender.9 10 These variables were included as explanatory factors in the PLSR analysis, which confirmed their limited influence on M tuberculosis DNA nasal excretion (see table 2).
The implementation of measures aiming to control factors related with M tuberculosis excretion in wild boar (vaccination against M tuberculosis var bovis to reduce the severity of TB in infected animals,33 vaccination against PCV-234 or deworming35) could be explored as strategies to reduce M tuberculosis shedding in this species.
Correction notice This article has been corrected since it was published Online First. Author affiliations have been amended.
Contributors DR, RM, MB, PFL and JHdM were involved in planning and supervising the work. DR, RM and PG contributed to sample collection. DR, RM, RC, WG, AG and AQ implemented the laboratory experiments. DR, RM, MB, PFL and OB-P performed the analysis of the results. All authors contributed to the writing of the manuscript and provided critical feedback.
Funding This study was supported by Junta de Extremadura Regional Government (GRU10142) and a collaboration agreement between Consejería de Agricultura, Desarrollo Rural, Medio Ambiente y Energía and Extremadura University (FEDER). DR, WL García-Jiménez and AG acknowledge the Junta de Extremadura and the European Social Fund for their research contracts (Ref: PO14024, PO14022 and TA13003, respectively).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.