Contagious bovine pleuropneumonia (CBPP) is a bovine bacterial disease of major economic importance in sub-Saharan Africa. Vaccination has been recommended to control the disease in endemic areas such as the Maasai ecosystems of Kenya and Tanzania; however, the currently used live attenuated vaccine has been reported to have poor vaccine safety and efficacy. To compare standard (current) and an improved (buffered) version of the live CBPP-vaccine, several epidemiological studies were carried out in Maasai cattle in Kenya between 2006 and 2008. Specifically, the aims were to estimate CBPP seroprevalence at herd and animal level; to identify risk factors for seroprevalence at both levels; to investigate the spatial distribution of seroprevalence; to compare post vaccination adverse events in cattle vaccinated with a standard and a buffered vaccine, and finally to compare efficacy of the two vaccines to induce seroconversion and to prevent development of clinical signs suggestive of CBPP. A cross-sectional study was carried out in 6872 cattle in 175 randomly selected herds from Loita and Mara divisions. A competitive ELISA revealed that 85% of the herds in the area had at least one seropositive animal and that seropositive herds were harbouring 11% seropositive cattle. A complement fixation test revealed that 46% of the herds had at least one seropositive animal and that seropositive herds were harbouring 4% seropositive cattle. A multivariable logistic regression analysis of the seroprevalence indicated that previous vaccination against CBPP, a history of CBPP outbreaks in the herd, animal age and the location of the herd in the division of Mara were positively correlated to seroprevalence. To investigate the observed difference in herd seroprevalence between the two divisions further, a spatial analysis was conducted. A SatScan test revealed clusters in Mara in areas identified by veterinary personnel as CBPP ‘hot spots’. A logistic regression using spatial information identified that location in the midland agro-ecological zone or close to a river and vaccination were positively associated with seroprevalence. To compare safety and efficacy of a standard and a buffered vaccine, two cohorts of approximately 40,000 cattle were used. The study showed that within 100 days post vaccination, 6.2 cattle per 1000 vaccinates developed adverse events, 4.1 of which were specifically attributable to vaccination and ranging from swelling of the tail to the tail sloughing off. This study revealed a slightly higher incidence of adverse events in cattle vaccinated with the buffered vaccine compared to the standard vaccine. A comparison of the efficacy of the two vaccines revealed that cattle vaccinated with the buffered vaccine had higher odds of seroconversion and lower odds of developing symptoms of CBPP, three and twelve months post vaccination respectively. The epidemiological studies conducted clearly show wide spread seroprevalence in the Maasai cattle. Given the (spatial) heterogeneity observed, control measures should probably be targeted in areas of increased risk (clusters). However, positive association of vaccination and seropositivity call for better diagnostics tests that can differentiate vaccinated from infected animals. Vaccination with buffered vaccine resulted in increased seroconversion, decreased clinical signs indicative of CBPP post vaccination and low seroprevalence post ‘outbreak’. Nevertheless, the increase in adverse events related to the buffered vaccine calls for further research into safer CBPP vaccines.

Deux tests sérologiques (test de fixation de complément (CFT) et l'ELISA de compétition (cELISA)) sont recommandés par l'OIE et utilisés couramment au Laboratoire Central Vétérinaire de Bamako parfois en parallèle dans le diagnostic et le dépistage de la péripneumonie contagieuse bovine (PPCB). La performance de ces tests a été estimée différemment par plusieurs auteurs dans des contextes épidémiologiques différents à partir de méthodes statistiques standards avec un statut sanitaire réel des animaux partiellement ou totalement connu. Dans un milieu où la PPCB est endémique avec différents stades d'évolution de la maladie, sachant que les tests sérologiques sont non parfaits (non gold standards), l'utilisation d'une approche bayésienne semblait appropriée pour une appréciation précise des paramètres de performance de tests qui sont la sensibilité et la spécificité, afin de mieux apprécier la prévalence de la maladie dans le cheptel bovin du delta central du Niger au Mali. Les résultats d'analyse de laboratoire des échantillons de terrain ont servi de bases de données importantes pour une analyse descriptive de la situation épidémiologique par appréciation des patrons de variations des principaux paramètres pouvant exercer une influence majeure sur la propagation de la PPCB. Ceci, dans le but d'aider à la réflexion sur la recherche d'outils et stratégies nouvelles dans le processus de prévention et d'éradication de la PPCB par le développement des modalités d'implantation d'une méthodologie innovante, pratique et efficace comme la qualification sanitaire troupeau concernant la PPCB dans un environnement d'élevage extensif. Cette thèse a permis de mieux définir les corrélations entre les deux tests, d'observer une meilleure sensibilité de cELISA par rapport à CFT permettant de justifier son utilisation seule dans un programme de dépistage à large échelle de la PPCB dans un milieu endémique. La démonstration dans l'étude de l'existence d'agrégation des animaux séropositifs à l'échelle du troupeau et aussi géographique montre qu'un système de qualification sanitaire troupeau pourrait jouer en collaboration avec le réseau national de surveillance épidémiologique vétérinaire, un rôle prépondérant dans la lutte ciblée et la maîtrise de la propagation de la PPCB au Mali. Mots clefs : PPCB-cELISA-CFT-Approche bayésienne-Agrégation-qualification sanitaire-Bovin / Two serological tests (complement fixation test (CFT) and competitive ELISA (cELISA)) are recommended by the OIE and commonly used in Central Veterinary Laboratory of Bamako sometimes in parallel, in the diagnosis and screening for contagious (CBPP). The performance of these tests has been estimated differently by several authors in different epidemiological settings using standard statistical methods with a real status of animals partially or completely known. In an environment where CBPP is endemic and where different stages of disease are available, given that serological tests are not perfect (not gold standard), the use of Bayesian approach seemed appropriate for an accurate assessment of the performance parameters of tests which are the sensitivity, specificity and predictive values to better assess the prevalence of the disease in cattle in the central Niger delta in Mali. The results of laboratory analysis of field samples were used as large database for epidemiological analysis of the geographical distribution of seroprevalence and the influence of major risk factors for the spread of CBPP. This, in order to aid reflection on tools research and new strategies in the process of prevention and eradication of CBPP by developing for implementation of an innovative, practical and effective methodology as sanitary qualification of cattle. This thesis has helped define the correlations between the two tests, observing a better sensitivity of cELISA compared to CFT to justify its use only in a program of widespread testing of CBPP in an endemic environment. In this study, the proof of the existence of aggregation of seropositive animals across herds and geographical level shows that a sanitary qualification system of cattle can play in collaboration with the national network of veterinary epidemiological surveillance a leadership role in targeted control and mastery of the spread of CBPP in Mali.Keys words: CBPP- cELISA - CFT- Bayesian approach -Aggregation- Sanitary qualification –Bovine

Contagious bovine pleuropneumonia (CBPP) is a severe infectious disease caused by Mycoplasma mycoides subsp. mycoides small colony type (M. mycoides SC) and is a vast problem in Africa. Current CBPP prevention is based on attenuated live strain vaccines, but these are limited by factors such as short-term immunity, cold-chain dependence and retained virulence. CBPP can be diagnosed using post-mortem examination, identification of the agent using culture and PCR based methods as well as serological diagnostic methods, but the latter are generally not sensitive enough and there is also demand for an inexpensive, pen side field test.The research presented in this thesis was focused on using recombinantly expressed surface proteins from M. mycoides SC to characterize humoral immune responses to CBPP. Thereby candidate proteins to be used in development of serological diagnostic methods and possibly subunit vaccines could be identified. As a first step, five putative variable surface proteins of M. mycoides SC were expressed and purified from E. coli in Paper I. These proteins were analyzed using immunoblotting techniques and results showed that one protein, MSC_0364, was variably expressed on the surface of M. mycoides SC in vitro. Paper II presents expanded efforts including cloning and expression of 64 recombinant surface proteins and an assay for high throughput analysis of protein-specific IgG, IgA and IgM titers in hundreds of sera using a bead-based screening assay. The assay was evaluated by protein-specific inhibition experiments, comparisons to Western blotting and monitoring of immune responses over time in a study with sera taken from eight animals over 293 days from a previous vaccine trial.Papers III and IV present applications using the recombinant proteins and bead-based screening assay wherein proteins for diagnostic and vaccine development were identified. In Paper III, the assay was used to screen 61 proteins using well-characterized serum samples from cattle with CBPP and healthy controls, resulting in selection of eight proteins suitable for diagnostic use. These proteins were combined and evaluated in a proof-of-concept ELISA with a discriminative power that enabled 96% correct classification of sera from CBPP-affected and CBPP-free bovines. Paper IV reports the results and protein-specific analyses of a vaccine trial using the recombinant putative variable surface proteins presented in Paper I as a subunit vaccine. The vaccine conferred no protection, but a weak vaccine response could not be excluded as the cause of failure. In an effort to identity other protein candidates to be used in a subunit vaccine, protein-specific analysis of humoral immune responses elicited by the currently approved live strain vaccine, T1/44, were investigated. Here, five proteins with high IgG titers associated to immunity were identified: LppQ, MSC_02714, MSC_0136, MSC_0079 and MSC_0431. These proteins may be important in the development of a novel subunit vaccine against CBPP. / QC 20100719