Over the past decades, the differentiation of bacterial strains for epidemiological purposes had been based on conventional phenotypic characters. More recently methods studying the directly coded molecules or semantides (nucleic acids or proteins) has allowed, concomitantly with the technical progresses of electrophoresis, the description of stable, discriminant, reproducible markers which were applicable to large series of isolates. Initially applied to study nosocomial infections in industrialized countries, these methods appear to be particularly suitable for an approach of the epidemiology of endemic bacterial infections in sub-Saharan Africa. The fact that these tools remain costly and technically complicated explain that most of these studies are conducted in the laboratories of industrialized countries. This research reveals the epidemiological complexity of most of these infections Thus, the epidemiology of trachoma was studied by the analysis of polymorphism of the major outer membrane protein gene of Chlamydia trachomatis in a village of Gambia. A PCR based technique was used to determine the frequency of infection in symptomatic and clinically negative subjects and to specify the prevalence of the genotypes. The epidemiology of plague was studied by the restriction fragment length polymorphism (RFLP) analysis of the ribosomal RNA genes (ribotyping). Distinct ribotypes differentiated the strains of the first two pandemics from the third. The strains of African origin were particularly heterogeneous, especially in Kenya. This diversity may be explained by the fact that the plague focus is extremely ancient in Central Africa. Bacterial agents of meningitis were also studied The electrophoretic polymorphism of outer membrane proteins of Haemophilus influenzae of b type was used to specify the epidemiology of meningitis in Gambia. The invasive strains exhibited distinct profiles from non-invasive strains. Different types were evidenced in the west, east and central parts of the country The antigenic polymorphism of outer membrane proteins of Neissera meningitidis allowed the differentiation of the mains isolated in Mall according to the period of isolation. Thus the endemic strains of A serotype were distinguished from those belonging to the same serotype which were responsible for the 1994 epidemic. Several molecular methods were applied to the typing of Vibrio cholerae strains, particularly these of the seventh pandemic. The enzyme electrophoretic polymorphism (MLEE), a technique based on RFLP analysis of toxin genes, the arbitrarily primed PCR (AP. PCR) and mainly the ribotyping were applied. This last method revealed that in Africa several clones of V. Cholerae El Tot were responsible of the seventh pandemic. Moreover the technique has evidenced the intercontinental spread of a clone of V. Cholerae isolated in 1993 in Calcutta and identified a year later in Guinae-Bissau. Tuberculosis is at present the first opportunistic infection linked to HIV infection in sub-Saharan Africa. Tuberculosis incidence is particularly high and is expected to increase. Several molecular methods, including IS 6110 RFLP analysis, AP-PCR and spoligotyping were used to study the epidemiology of tuberculosis in various countries: South Africa, Tanzania, Zimbabwe, Kenya and Malawi. The aims of this research varied: prevalence of reactivation and of recently acquired infections, routes of contamination, degree of genetic diversity of the organisms isolated in a given geographic area, urban and rural origins of the infections comparison of isolates from HIV seropositive and HIV seronegative patients. Identical profiles in the strains isolated from several patients could correspond to clusters of infections. However the identification of epidemiological links in mast clusters is hard to obtain. As for nosocomial infections in industrialised countries, these studies, using discriminant molecular tools, demonstrated the epidemiological complexity of these infections in sub-Saharan Africa and have led to the reconsideration of classical models. This research will be useful in the future to optimise prophylactic measures and design new vaccinal strategies.
