Research Report for Masters degree Faculty of Health Sciences / Since the early 1980's, global trends in dental caries have indicated that 80% of the caries is present in approximately 20% of the population which suggests a variation in susceptibility to the disease. In Sub-Saharan Africa and in South Africa, caries prevalence has shown a downward trend in preschool and school children. The reasons for this decline are obscure and have not been attributed to dietary habit, oral hygiene, the use of fluoride dentrifices or to any public health prevention program. Furthermore, the numbers of mutans streptococci, a group of pathogens associated with dental caries, have remained similar in children with and without caries. This implies that good dental health is possible in the presence of high prevalence of mutans streptococci, but raises speculation that the decrease in dental caries, may be caused by a change in virulence of these strains. It is also unclear if these bacterial strains are acquired through inter-familial transmission or genetically altered by influences from the oral environment. This thesis reports the first studies of gene expression and bacterial virulence in relation to traditional African-foods using clinical isolates from South African children. To establish the source of transmission, the phenotype and genotype of mutans streptococci strains from 31, five-year-old black, and coloured, children and their mothers living in Gauteng were characterized. The children were examined for caries, and plaque and salivary samples collected from both the children and their mothers. Samples were selectively cultured for mutans streptococci, biochemically differentiated and the genetic diversity of these isolates determined by PCR-RFLP of the gtfB and gtfI glucosyltransferase virulence genes. Phenotyping showed that Streptococcus mutans were 90% (155/172) prevalent, but the detection of Streptococcus sobrinus was low and comprised 10% (17/172) of the remaining vii isolates. Twenty-six percent of S. mutans clinical isolates (41/155) did not metabolise melibiose and the gtfA gene encoding for the uptake of this sugar was absent in 26 of the 41 melibiose-negative strains. GtfB gene polymorphisms in S. mutans clinical isolates from the two ethnic populations and from caries-free and caries-active 5-year-old children were similar (Principal Components Analysis). However high genetic diversity was observed in S. mutans isolates, with 23 different gtfB amplitypes shown by PCR-RFLP analysis. Sixteen different gtfI amplitypes were indicated in S. sobrinus clinical strains. The percentage match between gtfB amplitypes (HaeIII enzyme digests) in the children and their mothers ranged from 3% to 9% in caries-free children and caries-active children, respectively. Identical gtfB amplitypes from melibiose-negative phenotypes were shared by four mothers and their children only. To determine the growth and virulence response of the variant mutans streptococci genotypes, six laboratory reference strains (NCTC) and five, clinical isolates were challenged to traditional African staple foods and food combinations. The bacteria were exposed in batch culture for 16 hours to maize, samp, brown bread, maize+milk+sugar, maize+gravy, samp+beans, brown bread+margarine+peanut butter, 3% sucrose and a synthetic complex medium, BHI+3% sucrose. Results showed that growth was slow in maize (5.6 h) and samp (5.7 h) indicated by the long doubling time and the low number of generations (g = 4.2; g = 3.6). Sufficient lactic and acetic acid was produced to drop the pH to below 5.7, the ‘critical’ level for enamel demineralisation during the fermentation of brown bread (5.37), bread+margarine+peanut butter (5.51) and 3% sucrose (5.32). Water-insoluble extracellular polysaccharides produced was significantly lower (P<0.05) in samp and maize and viii maize+milk+sugar, with most residual glucose found in BHI + 3% sucrose (1.61±0.52 mg/mL) and maize+gravy (0.51±0.61 mg/mL). Mean concentrations of extracellular protein ranged from a low of 0.015±0.007 mg/mL in samp+beans to a high of 0.29±0.16 mg/mL in BHI + 3% sucrose. The inherent pH of individual foods closest to neutral was: milk (7.0) beans (6.07), samp (6.20), maize (6.82) and peanut butter (6.90). However, beans (5.7×10 G3 [H%]), milk (5.1 ×10G3 [H ]) and peanut butter (4.0×10 [H ]) showed more efficient buffering, which in combination % G3 % with other food components, raised the inherent buffering capacity of the food. These mixed foods required a larger quantity of acid to be produced by bacterial metabolism to lower the pH to 5.7. A preliminary study on the effect of single foods on S. mutans gtfB gene expression showed that mRNA gtfB transcripts were mostly inhibited in clinical isolates, but not in reference strains. The gtfI gene of all S. sobrinus test strains was expressed in 3% sucrose and BHI+3% sucrose, but the response differed in the remaining foods.The statistically significant association shown between mutans streptococci phenotype, gtf gene expression and the food challenge (P<0.0001), verifies that expressed phenotypic characteristics is dependent on gene expression. The results presented in this thesis show that a high diversity of gtf genes exists in mutans streptococci clinical isolates from South African black African, and coloured, 5-year-old children and their mothers. However, no specific genotype was unique either to dental status or ethnic population, with children acquiring genotypes from other points of contact besidesthe mother. Furthermore, the growth and virulence response (acidogenesis, water-insoluble ECP) of S. mutans and S. sobrinus genotypes and reference strains are subject to the dietary nutrients available. The inherent buffering capacity of maize, samp, maize+milk+sugar, maize+gravy and samp+beans, coupled to a low sucrose content, make these foods non-caries promoting, but the ability of test bacteria to remain viable on these nutrients, indicate that the mutans streptococci have a natural adaptive ability to assimilate other nutrients as a source of carbon when sucrose is limited. The control of gtfB and gtfI gene expression by traditional African foods suggests that virulence of the mutans streptococci are influenced more by the dietary environment than by genotype. Also, the difference in virulence properties between clinical and laboratory reference strains indicate an attenuation in virulence of wild-type strains.