Despite the existence of efficient chemotherapy, tuberculosis remains a leading cause of mortality worldwide. New drugs are urgently needed to reduce the potential impact of the emergence of multidrug-resistant strains of the causative agent Mycobacterium tuberculosis (Mtb). The front-line antibiotic isoniazid (INH), and several other drugs, target the biosynthesis of mycolic acids and especially the Fatty Acid Synthase-II (FAS-II) elongation system. This biosynthetic pathway is essential and specific for mycobacteria and still represents a valuable system for the search of new anti-tuberculous agents. Several data, in the literature, suggest the existence of protein-protein interactions within the FAS-II system. These interactions themselves might serve as targets for a new generation of drugs directed against Mtb. By using an extensive in vivo yeast two-hybrid approach and in vitro co-immunoprecipitation, we have demonstrated the existence of both homotypic and heterotypic interactions between the known components of FAS-II. The condensing enzymes KasA, KasB and mtFabH interact with each other and with the reductases MabA and InhA. Furthermore, we have designed and constructed point mutations of the FAS-II reductase MabA, able to disrupt its homotypic interactions and perturb the interaction pattern of this protein within FAS-II. Finally, we showed by a transdominant genetic approach that these mutants are dominant negative in both non-pathogenic and pathogenic mycobacteria. These data allowed us to draw a dynamic model of the organization of FAS-II. They also represent an important step towards the design of a new generation of anti-tuberculous agents, as being inhibitors of essential protein-protein interactions.