Evaluation of new vaccines against tuberculosis in goats and advances in their immunological diagnosis

PhD Thesis defense by Álvaro Roy Cordero at the VISAVET Centre of the Complutense University of Madrid

December 11^st, 2020

Tuberculosis in goats is a zoonotic mycobacteriosis, mainly caused by M. caprae and M. bovis, which leads to important economic losses for the goat producers and may represent a reservoir of tuberculosis for other domestic and wild animal species, as well as humans. At the European level, the active diagnosis of tuberculosis in goats is limited to goat herds coexisting or having epidemiological links with infected cattle or when raw milk and products made from raw milk are intended for human consumption (European Regulation EC 853/2004). However, specific eradication programs have been implemented in goats in some regions of Spain with an important goat population. These programs are based on the use of intradermal tests and, occasionally, the interferon-gamma (IFN-γ) release assay (IGRA) in infected farms. In addition, there are some commercial antibody tests and others in an experimental phase.

An increasing number of studies on the performance of these diagnostic techniques has been published in the last 20 years, but to date there is no systematic review and meta-analysis of the accuracy of these techniques in goats. In this regard, the objective of the first study of this thesis was to conduct a systematic review and meta-analysis of the accuracy of the ante-mortem diagnostic techniques of tuberculosis in small ruminants. Most of the reviewed studies were carried out in Spain (41%), in goats (76%), and the single intradermal tuberculin (SIT) test was the most widely reported technique (89%). In this sense, the procedures and interpretation criteria described in most of the studies were equal or similar to those used in cattle, although this circumstance did not prevent the high between-study heterogeneity observed. The information reported in the included studies showed some shortcomings in fields that may affect the accuracy of the diagnostic tests (for example, in the case of the SIT test, factors such as the tuberculin potency, the tuberculin manufacturer or the inoculation site). In our study, the median pooled sensitivity estimates of the SIT test (ranged from 51% to 59% depending on the interpretation criteria) and the comparative intradermal tuberculin (CIT) test (ranged from 30% to 50% depending on the interpretation criteria) were lower than that previously described in cattle. Nevertheless, the specificity estimates were high for both tests (equal to or greater than 95%), except for the SIT test in trials that reported the vaccination against M. avium subsp. paratuberculosis (MAP) (78% and 90% depending on the interpretation criteria). The estimated sensitivity of the IGRA was slightly higher than that of the SIT test (66% and 72%, 0.1 and 0.05 cut-offs, respectively), but lower than previously reported in cattle. However, the median pooled specificity estimates of the IGRA were high, regardless the interpretation criteria (ranged from 98% to 99%).

Unfortunately, the high between-study heterogeneity found hampered the conclusive interpretation of pooled sensitivity and specificity estimates. However, based on the low sensitivity of the intradermal tests found in our study, it would be recommended the use of the IGRA or serological tests in parallel to maximize the sensitivity in infected herds. Our results also highlight the need for further studies of diagnostic techniques optimization in goats, with a special focus on the evaluation of new serological techniques for their potential cost-effectiveness.

The second study of the present thesis tackles the interference problem caused by the vaccination against MAP on the diagnostic tests for tuberculosis in goats, as described in the previous study on the SIT test. More specifically, the aim of the study was to evaluate the temporal trend of this interference on the intradermal tests, IGRA and two antibody tests (P22 ELISA and DR-ELISA). In this sense, a cohort of 99 goats from a herd with no previous history of tuberculosis was vaccinated with Gudair at seven months of age. They were subjected to consecutive intradermal tests, IGRA, P22 ELISA and DR-ELISA every three months, until the interference disappeared. When using the SIT test, a variable number of positive reactors were observed at three months (32.3%, 95% CI 23.9–42.1), six months (11.5%, 95% CI 6.5–19.4), nine months (6.4%, 95% CI 3.0–13.2) and twelve months (0%, 95% CI 0–4.0) post-vaccination. In contrast, the CIT test had a specificity of 100% (95%, CI 96.0–100), regardless of the time post-vaccination. The IGRA also obtained high specificity values throughout the study period (≥ 97%, 95% CI 91.5-100). No significant interference on the serological tests was recorded at three months post-vaccination [P22 ELISA, specificity = 96% (95% CI 90.1–98.4); DR-ELISA, specificity = 98% (95% CI 92.9–99.4)], although an increase in the antibody titers was observed in the following herd testing events.

In conclusion, the use of the SIT test causes the onset of false-positive reactors if applied before 12 months post-vaccination in a tuberculosis-free and paratuberculosis-vaccinated herd. Nevertheless, the CIT test and IGRA obtained high specificity values under this epidemiological scenario. The antibody tests were also highly specific, although their specificity decreased significantly after several intradermal tests.
In the last study of this first chapter on diagnostic tests, it was also evaluated the performance of the experimental ELISA based on the P22 protein complex (P22 ELISA) for the detection of antibodies against Mycobacterium tuberculosis complex (CMTB) in goat milk. In this study, serum and milk samples were collected from two goat herds with different apparent prevalence of tuberculosis (72.2% and 7.3%, parallel interpretation of the SIT test and IGRA). Milk samples were collected from all dairy animals and from the bulk tank milk of both herds. Furthermore, in order to evaluate the stability of antibodies in milk under different preservation and storage procedures, milk samples from 20 goats of the high prevalence herd were subjected to different storage procedures: i) refrigerated without preservatives and assayed within the day of collection; ii) frozen the day of collection and tested one week later; and iii) kept refrigerated at 4 °C with added azidiol for one week before processing.

The correlation between milk and serum antibody titers was very strong in the high prevalence herd (rs = 0.91) and moderate in the low prevalence herd (rs = 0.46). It was also concluded that the preservation of the milk samples refrigerated in azidiol for one week did not affect the qualitative or quantitative result of the P22 ELISA. Among the 21 slaughtered animals at the end of the study, the P22 ELISA detected the highest proportion of lesion and/or culture positive animals (61.9% and 66.7% using milk and serum samples, respectively). Our results also indicated that the parallel use of the SIT test and P22 ELISA in infected herds, either using serum or milk samples, can be an efficient and economically viable diagnostic tool to maximize the detection of infected animals and accelerate the control of outbreaks. On the other hand, bulk tank milk samples might be a screening tool at the herd level in regions without eradication programs implemented or where it is not feasible to carry out an individual diagnosis in all herds. Our preliminary results showed that milk samples from the bulk tank milk were positive only in the high prevalence herd and, therefore, more studies are needed in herds with different prevalence and number of lactating animals in order to determine the minimum level of detection.

The second chapter includes four studies on the evaluation of new vaccines against tuberculosis and the preventive or therapeutic effect of vitamin D supplementation in goats exposed to a natural infection of tuberculosis. The immunogenicity and efficacy of a heat-inactivated M. bovis (HIMB) vaccine was assessed in the first two studies. In the first study, HIMB was administered orally and intramuscularly using a vaccination and revaccination protocol. No positive animals to the SIT test, IGRA and an experimental ELISA were observed in orally vaccinated goats two months after the boost, regardless of the interpretation criteria applied. However, the vaccination via the intramuscular route caused interference on the SIT test and IGRA in most of the animals (68.8% and 62.5%, respectively). The efficacy of the intramuscular route was evaluated in the second study of this chapter. In this study, intramuscular vaccination with HIMB conferred partial protection against a natural exposure to M. caprae, specifically by reducing the number of cranial lung lobes with lesions in vaccinated animals compared to the control group after ten months of exposure. However, the reduction in the lesion scores and the severity of lesions was not significantly lower in the HIMB vaccinated group.

The last two studies have evaluated the immunogenicity and efficacy of BCG and MTBVAC vaccines and the effect of vitamin D supplementation in goats using the same model of natural transmission used in the previous study. The results of these studies have two scopes: i) the direct applicability of MTBVAC and vitamin D supplementation in goats or other related animal hosts of tuberculosis, and ii) provide preclinical data on the efficacy of MTBVAC and vitamin D supplementation against a natural MTBC infection. MTBVAC was developed from the rational attenuation of an M. tuberculosis clinical isolate by two independent genetic deletions in the genes phoP and fadD26, which encode two major virulence factors. It is the only live M. tuberculosis vaccine in clinical trials, currently in clinical phase II, and constitute a promising candidate to replace BCG in neonates or be used as a booster of BCG in adolescents. In this sense, in the sixth study we assessed the immunogenicity and efficacy of BCG and MTBVAC vaccines against a natural M. caprae infection in goats. To achieve this aim, a group vaccinated with MTBVAC, a group vaccinated with BCG and a non-vaccinated group were exposed to a group of infected animals for nine months. With regard to the immunogenicity results, MTBVAC showed a greater bovine PPD-specific IFN-γ response in the IGRA and, therefore, a higher number of positive animals compared to the BCG group one month post-vaccination. It was also observed certain reactivity to ESAT-6 and CFP-10 antigens in a proportion of animals vaccinated with MTBVAC. This higher immunogenicity in terms of IFN-γ response in the IGRA was not correlated with protection and highlights the need to develop new specific DIVA antigens to distinguish infected from vaccinated with MTBVAC. On the other hand, the severity of gross lesions was lower in the pulmonary lymph nodes and in the total score of organs examined in the MTBVAC and BCG groups. Furthermore, MTBVAC effectively reduced the frequency of animals with extra-pulmonary lesions, while BCG reduced the pathology severity in the lungs, but no differences were found between both vaccinated groups.

In recent years, some studies have suggested the vitamin D supplementation in human tuberculosis, within the framework of the so-called host-directed therapies (HTDs), as a complement to antibiotic therapy aiming to improve the clinical outcome or reduce the duration of treatment on the one hand, and, on the other, as an immunomodulatory agent to boost the immune system and prevent infection or disease progression. Nevertheless, there are few studies on the immunomodulatory effect of vitamin D and its potential role against animal tuberculosis in the scientific literature, and only one study conducted in deer and wild boar has reported the effect of vitamin D supplementation on the pathology severity. The last study of the present thesis aims to study the prophylactic and/or therapeutic effect of vitamin D3 supplementation in goats exposed to a group of naturally M. caprae-infected goats during ten months. In this final study, the number of animals that were positive to tuberculosis diagnostic tests did not show significant differences between vitamin D-supplemented and control goats, and no inhibition of IFN-γ production after stimulation with bovine PPD was observed in the vitamin D-supplemented group. On the contrary, positive reactors to cell-based and antibody-based techniques were detected earlier in the vitamin D-supplemented group. Furthermore, our results indicated that high vitamin D supplementation in goats every two months did not reduce tuberculosis infection risk nor the diffusion and severity of tuberculosis lesions. In addition, goats supplemented with vitamin D presented hyperphosphataemia and renal injury with calcifications suggestive of vitamin D intoxication.