A dissertation submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science in Medicine / Fetal Alcohol Syndrome (F AS), the most severe form of Fetal Alcohol Spectrum Disorder (F ASD), has traditionally been associated with maternal alcohol consumption during pregnancy. However, a number of animal studies have shown an association between paternal preconception alcohol consumption and developmental abnormalities in the offspring that resemble the features of F AS. Dysregulation of epigenetic factors (such as DNA methylation) in the presence of alcohol may provide a plausible mechanism by which paternal alcohol consumption could result in offspring affected with features of F AS. Imprinted genes are expressed in a parentof- origin manner due to DNA methylation at distinct differentially methylated regions (DMRs) and are essential for normal embryonic development. There are only two known paternally methylated DMRs in humans, with an additional one described in mice - associated with Rasgrfl. The first aim of this study was to determine whether the human RASGRFl gene contains a DMR and whether this DMR is paternally methylated. In order to assess the imprint status of RASGRF 1, a number of computational assessments were done to identify key features of imprinted loci. Pyrosequencing analysis was used to assess the methylation status of various CpO islands surrounding RASGRFi in peripheral blood and sperm DNA samples. The RASGRF i-associated CpO regions were not found to exhibit differential methylation in a parent-of-origin manner. The second aIm of the study was to examine the effect of paternal alcohol consumption on the methylation status of the IG-DMR locus in male gametes and to detennine whether alcohol is correlated with methylation in a dose-dependant manner. Methylation assessment was done using the quantitative pyrosequencing technology. While an overall reduction in methylation was noted in males who consumed alcohol after adjusting for confounding variables, the amount of alcohol consumed did not correlate with overall methylation. When analyzed by individual CpG sites, alcohol consumption was found to correlate preferentially with demethylation at CpG 3 while alcohol-dosage preferentially correlated with demethylation at CpG 7. Age was significantly correlated with an increase in the overall methylation at JG-DMR and at individual sites within JG-DMR. In conclusion, these findings support the hypothesis that paternal preconception alcohol consumption can lead to hypomethylation of nonnally hypennethylated DMRs of specific imprinted genes in human spenn. This in tum could have significant implications with regard to the regulation of developmentally significant genes in the zygote and fetus, resulting in developmental, behavioral and neurocognitive disorders.

Alcohol is a major public health problem globally. According to the World Health Organization (WHO) report, alcohol accounts for 2.5 million deaths (4% of total) and 69.4 million (4.5% of total) of Disability-Adjusted Life Years (DALYs), (WHO, 2002, 2011). In South Africa, alcohol was found to be the third highest contributor to death and disability (Parry, 2007/8). Among the many far-reaching consequences of alcohol use in South Africa, Fetal Alcohol Syndrome (FAS) in certain regions in the country, rates among the highest in the world (May et al., 2007). Despite higher comparative levels of FAS in rural areas, almost one third of the population in urban sites (Gauteng) demonstrates that FAS is not exclusively a problem of South African rural areas. This study hypothesized that higher knowledge levels about the effects of alcohol on pregnancy may deter use among women. Employing secondary data analysis from a 2006 cross-sectional household survey, this study explored the prevalence of alcohol use among urban women (18-44 years) in the Tshwane Municipality, in general and in pregnancy. It also examined the relationship between knowledge of alcohol effects on pregnancy and alcohol use. A significant association was found between employment status, pregnancy outcome and general alcohol use among women. An almost significant association was found between knowledge of alcohol effects on pregnancy and past month alcohol use, knowledge levels and alcohol use during pregnancy. Findings partially support the hypothesis. However, knowledge of alcohol effects on pregnancy alone cannot deter women from using alcohol. Multiple influencing factors should be considered in planning prevention programmes for urban women’s alcohol use. Further research with larger sample sizes of pregnant women is suggested to explore the associations conclusively.