Background: Parkinson’s disease (PD) is a complex disease with a multi-factorial aetiology, comprising both genetic and environmental risk factors. The disease pathology is progressive and neurodegenerative where dopaminergic nerve cell death occurs predominantly in the substantia nigra pars compacta (SNpc) with the subsequent loss of the dopamine neurotransmitter in the basal ganglia. The most significant risk factors for PD include an advancing age and a family history of the disease, while environmental and lifestyle risk factors such as pesticide exposure and smoking are widely accepted as risk altering exposures. Currently up to 10% of PD is attributed to Mendelian inherited PD at one of 13 PARK loci in 9 genes. The pursuit of common susceptibility alleles for idiopathic PD has proven challenging with only a few loci reproducibility associated with an altered risk. The aim of this thesis is to study, using a candidate gene case-control design, the potential role of genetic variants in PD. The APOE candidate gene was hypothesized to modify the risk of PD as it is a proven modifier of Alzheimer’s disease (AD). The common pathological finding in PD of elevated levels of iron within the SNpc is proposed to increase the oxidative state of the nerve cells and predispose the dopaminergic neurons to apoptosis. Therefore, susceptibility alleles within the candidate genes that regulate iron metabolism and homeostasis are hypothesized to alter iron metabolism and predispose to iron-induced neurodegeneration in PD. Missense variants and common “tagging” SNPs with the HFE, Transferrin and Transferrin Receptor genes are investigated extensively in this thesis. Finally, autosomal recessively inherited PD can result from mutations in the parkin gene at the PARK2 locus. The final hypothesis explored in this thesis suggests that non-deleterious missense variants in the parkin gene modify the risk for developing sporadic PD. Further genetic variation in the parkin gene such as exon rearrangements is a frequently reported mutation where heterozygosity for these rearrangements may increase the risk of PD. Heterozygous deletions or duplications of exons in the parkin gene provide technical challenges for their detection. In this thesis a novel assay for the detection of these mutations is investigated. Methods: Genotyping was performed using PCR-RFLP for genetic variants in the APOE (E2 and E4 alleles), HFE (C282Y, H63D and S65C), Transferrin receptor (TfR; S142G), Transferrin (Tfn; P570S and G258S), IREB2 genes (L159V) and the parkin gene (S167N, R366W and V380L) in a cohort of 425 PD cases and 387 controls recruited from throughout Queensland, Australia. A tagged SNP high-throughput genotyping approach was then employed to try to replicate single SNP associations in 6 iron-related genes using a cohort of 1034 PD cases and 774 controls. These genetic variants were analysed for direct association with PD risk, age of onset effects as well as potential gene x gene (GxG) and gene x environment (GxE) interactions. Additionally, a quantitative PCR assay was developed to detect heterozygous deletions and duplications within the parkin gene and utilised to screen 43 YOPD cases for these mutations. Results: The initial study of the HFE C282Y variant revealed a significant protective association with PD in the two independent cohorts studied. Further study did not reveal significant associations with PD for the other HFE variants or missense variants within the Tfn and TfR genes. When analysed for GxE interactions, the C282Y, P589S and G277S variants showed evidence for an increased risk of PD in synergy with pesticide and herbicide exposure. Carriers of the risk variant and with toxin exposure were at two-fold increased risk of PD, although the number of individuals in this category was small. A further investigation of the role of common genetic polymorphisms in iron genes revealed only one of the 20 SNPs genotyped using high-throughput multiplex methods, remained significantly associated with PD after correction for age and sex. The rs198855 SNP is downstream of the HFE gene and further implicates a role for HFE in PD. The APOE E4 allele demonstrated modifying effects for the age of PD onset, restricted to the female cases. Analysis of the parkin missense variants also demonstrated a modifying effect on the age of PD onset in carriers of the S167N variant, with putative interactions between the APOE E4 allele, a family history of PD and toxin exposure that further reduced the age of onset. Twenty individuals of the 43 YOPD cases screened demonstrated heterozygous parkin exon rearrangements using the novel qPCR method. Conclusions: Non-synonymous variants within iron-related genes or the parkin gene putatively interact with herbicide and pesticide exposure to increase the risk of PD or modify the phenotype, highlighting the need for future studies to address the multi-factorial aetiology of PD in their study design and analysis. This thesis provides evidence for the association between genetic variation within the HFE locus and PD and for the APOE E4 allele as a modifier of PD.

<p>Advancements in human genomic technology have helped to improve our understanding of how genetic variation plays a central role in the mechanism of disease susceptibility. However, the very high dimensional nature of the data generated from large-scale genetic association studies has limited our ability to thoroughly examine genetic interactions. A prioritization scheme – Variance Prioritization (VP) – has been developed to select genetic variants based on differences in the quantitative trait variance between the possible genotypes using Levene’s test (Pare et al., 2010). Genetic variants with Levene’s test p-values lower than a pre-determined level of significance are selected to test for interactions using linear regression models. Under a variety of scenarios, VP has increased power to detect interactions over an exhaustive search as a result of reduced search space. Nevertheless, the use of Levene’s test does not take into account that the variance will either monotonically increase or decrease with the number of minor alleles when interactions are present. To address this issue, I propose a maximum likelihood approach to test for trends in variance between the genotypes, and derive a closed-form representation of the likelihood ratio test (LRT) statistic. Using simulations, I examine the performance of LRT in assessing the inequality of quantitative traits variance stratified by genotypes, and subsequently in identifying potentially interacting genetic variants. LRT is also used in an empirical dataset of 2,161 individuals to prioritize genetic variants for gene-environment interactions. The interaction p-values of the prioritized genetic variants are consistently lower than expected by chance compared to the non-prioritized, suggesting improved statistical power to detect interactions in the set of prioritized genetic variants. This new statistical test is expected to complement the existing VP framework and accelerate the process of genetic interaction discovery in future genome-wide studies and meta-analyses.</p> / Master of Health Sciences (MSc)

Les gènes du métabolisme des xénobiotiques (MX) impliqués dans l’activation et l’élimination des cancérogènes environnementaux pourraient moduler le risque de cancer du sein, mais leurs effets dans ce cancer ont été peu étudiés et sont mal connus. Les objectifs de la thèse étaient d’étudier le rôle des gènes MX dans le cancer du sein d’une part, et dans les niveaux biologiques de cancérogènes mammaires suspectés, d’autre part. Les analyses ont porté sur les données d’une étude cas-témoins en population sur les cancers du sein (étude CECILE). L’association avec le cancer du sein a été étudiée (1) avec les variants du gène NAT2 qui déterminent le type d’acétyleur lent ou rapide de chaque individu ; (2) les polymorphismes des gènes MX étudiés conjointement au niveau de chacun des gènes et au niveau de l’ensemble du pathway à l’aide d’une méthode exploratoire de type « gene-set ». Dans chacune de ces approches, les interactions avec la consommation de tabac ont été étudiées. Dans une dernière partie, nous avons cherché à identifier les polymorphismes des gènes MX prédictifs des concentrations sanguines de polluants organochlorés persistants (p,p’-DDE et PCB153) chez les témoins de l’étude CECILE. Le risque de cancer du sein était augmenté chez les femmes ayant un profil génétique NAT2 d’acétyleuses rapides par rapport aux femmes ayant un profil d’acétyleuses lentes. Parmi les acétyleuses lentes, les femmes fumeuses avaient un risque de cancer du sein augmenté par rapport aux non fumeuses indiquant l’existence d’une interaction tabac-NAT2. L’approche « gene-set » montrait que les polymorphismes au niveau de plusieurs gènes MX et au niveau de l’ensemble du pathway étaient associés collectivement au cancer du sein. L’association entre le cancer du sein et l’ensemble des polymorphismes du pathway XM était observée chez les fumeuses, indiquant le rôle de la consommation de tabac dans cette association. Enfin, nous avons montré l’effet du gène CYP2B6 en tant que déterminant des niveaux sanguins de p,p’-DDE et PCB153. Nos résultats mettent en évidence un rôle des gènes XM dans le cancer du sein qui peut être expliqué par leur fonction dans le métabolisme et l’élimination des cancérogènes environnementaux. / The xenobiotic metabolism (XM) genes involved in the activation and elimination of environmental carcinogens may modulate breast cancer risk, but their effects in breast cancer have been little studied and are poorly understood. The objectives of the PhD were to study the role of XM genes in breast cancer on the one hand, and in the biological levels of suspected breast carcinogens on the other. The analyses were based on a population-based case-control study on breast cancer (CECILE study). We investigated the association of breast cancer (1) with NAT2 gene variants that determine the type of slow or rapid acetylator in each individual; (2) with polymorphisms of XM genes that were studied jointly at the gene and at the XM pathway level using a gene set method. In each of these approaches, interactions with tobacco consumption were studied. In a final section, we sought to identify polymorphisms of XM genes that predict blood concentrations of persistent organochlorine pollutants (p,p'-DDE and PCB153) among the controls of the CECILE study.The risk of breast cancer was increased in women with a NAT2 genetic profile of rapid acetylators compared to women with a profile of slow acetylators. Among slow acetylators, current smokers had an increased risk of breast cancer compared to non-smokers, indicating an interaction between tobacco smoking and NAT2 genotype. The gene set approach showed that polymorphisms at the level of some XM genes and at the level of the entire pathway were collectively associated with breast cancer. The association between breast cancer and all pathway XM polymorphisms was observed in female smokers, indicating a role for tobacco smoking in this association. Finally, we have shown that CYP2B6 gene was a determinant of blood levels of p,p'-DDE and PCB153. Our results highlight a role of XM genes in breast cancer that is explained by their function in the metabolism and elimination of environmental carcinogens.

INTRODUCTION AND AIM - Cardiovascular disease is globally a major risk factor for morbidity and mortality. It is caused by various factors, one of which is an abnormal haemostatic process. Fibrinogen is a haemostatic factor that is considered to be an independent risk factor for cardiovascular disease. Elevated fibrinogen can be caused by environmental and genetic factors which increase the risk of the occurrence of thrombosis. The fibrinogen y' chain, which is one of the three chains of fibrinogen, has two different variants, the yA and y’. The presence of the fibrinogen y’ chain has been associated with thrombotic disorders. Many studies have investigated the fibrinogen variables in Caucasian individuals, but only a few such studies have been conducted on non-Caucasian individuals. The genetic diversity of ethnic groups differs and could cause differences in the fibrinogen variables between these groups. Fibrinogen is known to increase with age; therefore to explain changes over time in fibrinogen concentrations it was also important to investigate whether genetic determinants and possible gene–environment interactions influenced fibrinogen over time. In this study the main aim was to determine the change in the fibrinogen variables over a five-year period within a black South African cohort subdivided according to genotypes associated with fibrinogen variables, and to determine whether the observed changes were modulated by environmental factors. PARTICIPANTS AND METHODS - Data [baseline (n=2010) and follow-up (n=1288)] were collected in the Prospective Urban and Rural Epidemiology (PURE) study during 2005 and 2010 from apparently healthy black men and women aged between 35 and 65 years and residing in rural or urban settlements. Experimental methods included analysis of fibrinogen and fibrinogen y’ concentrations, single nucleotide polymorphisms (SNPs) and determination of environmental factors associated with the fibrinogen variables. RESULTS - The fibrinogen variables increased significantly from 2005 to 2010 in both the rural and urban participants, as well as in both men and women. The major environmental factors that affected the fibrinogen variables were C-reactive protein (CRP), interleukin-6 (IL-6), body mass index (BMI), glycated haemoglobin (HbA1c), age, blood lipids, human immunodeficiency virus (HIV) and tobacco use. Fibrinogen increased consistently from 2005 to 2010 in the respective genotypes of all SNPs analysed, except in the FGG 9340 T>C homozygous mutant carriers. Fibrinogen y’ also increased in general in most genotypes from 2005 to 2010, except in the FGG 10034 C>T mutant allele carriers, where a decrease was observed. It was determined that CRP was the only environmental factor that influenced the change in fibrinogen over time and that FGG 10034 C>T was the only SNP that influenced the change in fibrinogen y’ over the five years. Four gene–environment interactions also influenced fibrinogen on a cross-sectional level, i.e. FGA 2224 G>A with age, FGB Arg448Lys with HIV status, FGB 1643 C>T with urbanisation and FGB 1038 G>A with HbA1c. Only the FGG 9340 T>C with HbA1c interaction was found to predict change in fibrinogen concentrations over the five years. CONCLUSION - Both environmental and genetic factors significantly influenced the fibrinogen variables cross-sectionally as well as prospectively. It was clear that the influence of the environmental factors was mediated by genetic polymorphisms and vice versa, as can be seen by the gene–environment interactions found in this study. An important finding of this study was that the interaction of HbA1c with two SNPs on fibrinogen variables may explain the known inconsistent relationship found between fibrinogen concentrations and diabetes. / MSc (Dietetics), North-West University, Potchefstroom Campus, 2014

INTRODUCTION AND AIM - Cardiovascular disease is globally a major risk factor for morbidity and mortality. It is caused by various factors, one of which is an abnormal haemostatic process. Fibrinogen is a haemostatic factor that is considered to be an independent risk factor for cardiovascular disease. Elevated fibrinogen can be caused by environmental and genetic factors which increase the risk of the occurrence of thrombosis. The fibrinogen y' chain, which is one of the three chains of fibrinogen, has two different variants, the yA and y’. The presence of the fibrinogen y’ chain has been associated with thrombotic disorders. Many studies have investigated the fibrinogen variables in Caucasian individuals, but only a few such studies have been conducted on non-Caucasian individuals. The genetic diversity of ethnic groups differs and could cause differences in the fibrinogen variables between these groups. Fibrinogen is known to increase with age; therefore to explain changes over time in fibrinogen concentrations it was also important to investigate whether genetic determinants and possible gene–environment interactions influenced fibrinogen over time. In this study the main aim was to determine the change in the fibrinogen variables over a five-year period within a black South African cohort subdivided according to genotypes associated with fibrinogen variables, and to determine whether the observed changes were modulated by environmental factors. PARTICIPANTS AND METHODS - Data [baseline (n=2010) and follow-up (n=1288)] were collected in the Prospective Urban and Rural Epidemiology (PURE) study during 2005 and 2010 from apparently healthy black men and women aged between 35 and 65 years and residing in rural or urban settlements. Experimental methods included analysis of fibrinogen and fibrinogen y’ concentrations, single nucleotide polymorphisms (SNPs) and determination of environmental factors associated with the fibrinogen variables. RESULTS - The fibrinogen variables increased significantly from 2005 to 2010 in both the rural and urban participants, as well as in both men and women. The major environmental factors that affected the fibrinogen variables were C-reactive protein (CRP), interleukin-6 (IL-6), body mass index (BMI), glycated haemoglobin (HbA1c), age, blood lipids, human immunodeficiency virus (HIV) and tobacco use. Fibrinogen increased consistently from 2005 to 2010 in the respective genotypes of all SNPs analysed, except in the FGG 9340 T>C homozygous mutant carriers. Fibrinogen y’ also increased in general in most genotypes from 2005 to 2010, except in the FGG 10034 C>T mutant allele carriers, where a decrease was observed. It was determined that CRP was the only environmental factor that influenced the change in fibrinogen over time and that FGG 10034 C>T was the only SNP that influenced the change in fibrinogen y’ over the five years. Four gene–environment interactions also influenced fibrinogen on a cross-sectional level, i.e. FGA 2224 G>A with age, FGB Arg448Lys with HIV status, FGB 1643 C>T with urbanisation and FGB 1038 G>A with HbA1c. Only the FGG 9340 T>C with HbA1c interaction was found to predict change in fibrinogen concentrations over the five years. CONCLUSION - Both environmental and genetic factors significantly influenced the fibrinogen variables cross-sectionally as well as prospectively. It was clear that the influence of the environmental factors was mediated by genetic polymorphisms and vice versa, as can be seen by the gene–environment interactions found in this study. An important finding of this study was that the interaction of HbA1c with two SNPs on fibrinogen variables may explain the known inconsistent relationship found between fibrinogen concentrations and diabetes. / MSc (Dietetics), North-West University, Potchefstroom Campus, 2014