PhD Thesis defense by Javier Ortega Martín at the VISAVET Centre of the Complutense University of Madrid

June 6^th, 2024

Caprine tuberculosis (TB) is a zoonotic mycobacteriosis caused by members of the Mycobacterium tuberculosis complex (MTBC), mainly M. caprae and M. bovis. Due to its health and economic impact, this disease is subject to surveillance and eradication programs. In Spain, certain regions have been established specific eradication programs for those caprine herds epidemiologically related to cattle herds. In addition, caprine TB eradication programs have been implemented in some Autonomous Communities, also included in a specific manual of the Ministry of Agriculture, Fisheries and Food of Spain in order to harmonize criteria. These programs are mainly based on test and cull strategy and slaughterhouse surveillance. The official ante-mortem diagnosis used in these programs is mainly based on the use of the single and comparative tuberculin tests (SITT and CITT) and, in specific cases, the interferon-gamma release assay (IGRA), both techniques based on the cellular immune response. In addition, there are unofficial humoral-based techniques, which are able to detect specific antibodies to MTBC members which have undergone a major development in the last decade. The diagnostic performance of techniques based on cellular immune response as well as those based on antibody detection can be affected by different factors. The current doctoral thesis is composed of different studies, structured in two chapters, related to immunological diagnostic tests for caprine TB, with a special focus on the factors that may affect their performance.
The first chapter of this thesis consists of three studies focused on optimize the diagnostic techniques based on cellular immune response for caprine TB, especially the intradermal tuberculin test, the cornerstone of eradication programs in goats. The first study of this chapter aimed to evaluate the effect of the Protein Purified Derivatives (PPDs) inoculation site on the diagnostic performance of SITT/CITT in goats. For this purpose, bovine and avian PPDs were administered in cervical and scapular sites of animals belonging to herds with different epidemiological status: high prevalence (n=96), low prevalence (n=94) and TB-free (n=119). In addition, and to gain a better knowledge of the health status of these animals, blood samples were collected in order to obtain plasma and serum for IGRA and an enzyme-linked immunosorbent assay (ELISA), named P22 ELISA, respectively. A significantly (p < 0.01) higher number of reactors to the standard interpretation of CITT was observed on the cervical site compared to the scapular site, as well as a significantly (p < 0.01) higher occurrence of local clinical signs on the cervical site in the high prevalence herd. In contrast, in the low-prevalence herd the number of reactors to the severe interpretation of SITT was significantly (p < 0.01) higher on the scapular site compared to the cervical site. In the TB-free herd, a significantly (p = 0.016) higher number of reactors to the severe interpretation of SITT was observed when the PPDs were administered on a cervical site compared to scapular site. The results of this study contributed to the fact that both locations are recommended by the European Reference Laboratory for Bovine Tuberculosis for the performance of the intradermal tuberculin test for movement to other Member States in caprine, allowing, therefore, the use of a cervical or scapular site for the PPDs administration in goats since the performance of the intradermal tests is not affected.
The second study of this first chapter evaluated the effect of a previous pre-sensitization caused by a intradermal tuberculin test on the results of the diagnostic techniques for caprine TB. For this purpose, goats (n=48) from a high TB prevalence herd were subjected to two successive SITT/CITT with a 3-day separation interval between them. Blood samples for plasma (IGRA) and serum (P22 ELISA) were also collected at days 0 and 3 to evaluate the effect of a previous pre-sensitization on these diagnostic techniques. In this study, a significantly (p = 0.015) higher number of SITT reactors were observed in the pre-sensitized group compared to the control group. However, the pre-sensitization caused by the administration of bovine and avian PPDs had no significant effect on the results of the in vitro diagnostic techniques (IGRA and P22 ELISA). Therefore, the results of this study in goats demonstrated the effect of a recent pre-sensitization (3 days) with PPDs on the results of TB diagnostic techniques leading to a significant increase in the number of reactors to SITT.
Moreover, the last two studies of this first chapter evaluated the effect of topical (at the PPD inoculation site) and parenteral (intramuscular on the neck) application of a corticosteroids (betamethasone and dexamethasone) and a non-steroid anti-inflammatory drug (ketoprofen) 48 hours after the PPDs administration on the results of TB diagnostic techniques in goats. In addition, the potential detection of these substances by High-performance liquid chromatography-Mass spectrometry (HPLC-MS) in hair (betamethasone, dexamethasone and ketoprofen) and serum (dexamethasone and ketoprofen) samples obtained 24 hours after administration of these drugs was evaluated. For this objective, the animals of both studies were subjected to SITT/CITT and blood samples were collected for the IGRA (in plasma) and P22 ELISA (in serum) analysis on day 0 and 3 of the study. Serum and hair samples used for the detection of the administered substances were collected 24 hours after their application (topical or parenteral). In this same sampling, serum samples were also collected to determine the effect of the intramuscular administration of dexamethasone and ketoprofen on serological levels of 15 cytokines and chemokines. In both studies the administration of the anti-inflammatory substances led to a significant (p < 0.05) decrease in the number of SITT reactors. In contrast, there was no significant decrease in the reactivity shown by the other techniques used in this study (CITT, IGRA and P22 ELISA). The detection of the substances by HPLC-MS was different depending on the route of administration, with optimal detection of betamethasone (topically administered) in hair samples, but a poor detection of dexamethasone and ketoprofen in serum and hair samples following intramuscular administration. The results of the third study showed an absence of a significant effect on the serological levels of the different cytokines and chemokines, as no significant differences were observed between the animals in the control group and those treated with dexamethasone and ketoprofen in serum samples obtained 24 hours after administration of the drugs. The results of both studies have contributed to a better understanding of the effect of the administration of anti-inflammatory substances on the diagnosis of TB, as well as to remark the need to develop effective methods for the detection of this activity.
The second chapter of the present doctoral thesis consists of two studies related to the optimization and improvement of the diagnostic performance of caprine TB by using techniques based on the detection of antibodies, increasing the knowledge about the different factors that can affect it. The first study of the second chapter evaluated the effect of three factors on the results of a P22 ELISA in caprine milk samples. The first part of the study determined the effect of the lactation stage on P22 ELISA results. For this purpose, milk samples from 44 animals from a high prevalence herd were collected each month and analyzed during the 6-month lactation period. The second part of the study was focused on the evaluation of the effect of a recent SITT/CITT on P22 ELISA results using milk and serum samples (booster effect). For this objective, serum and milk samples from 37 animals were collected the day of the PPDs administration and 15, 30 and 60 days later and analyzed by P22 ELISA. Finally, the third part of the study showed the effect of successive freeze-thaw cycles in milk samples preserved with azidiol on the P22 ELISA results when analyzing these samples. Milk samples from animals previously positive and negative to P22 ELISA were analyzed and subsequently preserved with azidiol and subjected to successive freeze-thaw cycles. A decrease in the number of reactors to P22 ELISA was observed when analyzing milk samples obtained in the last two months of lactation compared to those collected in the previous samplings. Regarding the booster effect, the evolution of the P22 ELISA results was similar in serum and milk samples obtained 15, 30 and 60 days after the PPDs administration. The highest antibody titres were obtained at 15 days in both samples, maximizing the detection of TB-infected animals. Finally, subjecting milk samples preserved with azidiol to successive freeze-thaw cycles did not result in significant differences (p = 0.99) on the P22 ELISA results in the different analyses performed between each cycle. Thus, the results of this study demonstrated the effect of different factors on the results of the P22 ELISA using milk samples, being of great value for the humoral-based diagnosis of caprine TB using this type of sample.
In the last study of this second chapter, the usefulness of oral fluid samples for humoral diagnosis of TB in goats using the P22 ELISA was evaluated. For this purpose, oral fluid and serum samples from animals from a historically TB-free herd (n=113) and from a high prevalence TB herd (n=197) according to the results of SITT/CITT obtained in previous samplings, were analyzed by P22 ELISA. The results obtained in the TB-free herd were used to calculate the specificity of the P22 ELISA. Besides, and due to the high proportion of reactors observed in the high prevalence herd, all animals were slaughtered and subjected to post-mortem analysis Then, those animals that showed TB compatible lesions and bacteriological isolation (n=64) were included in the study to calculate the sensitivity of the P22 ELISA in oral fluid samples. An excellent specificity [100% (95% CI, 97.4-100)] of the P22 ELISA in oral fluid samples was observed, accompanied by a limited sensitivity [34.4% (95% CI, 22.4-45.6)], which increased when taking into account those animals with the highest lesions score [75% (95% CI, 53.7-96.2)]. In this line, significantly (p = 0.018) larger TB-compatible lung lesions were observed in P22 ELISA-positive animals in oral fluid samples compared to those negatives. However, the sensitivity of P22 ELISA in serum samples was considerably higher, being 87.5% (95% CI, 77.2-93.5) in those animals with TB- compatible lesions and 100% (95% CI, 80.6-100) taking into account only the setting of the animals with severe lesions. According to these results, humoral diagnosis of caprine TB based on oral fluid samples showed a better performance in animals with severe lesions.