Stable RNAs undergo a wide variety of post-transcriptional modifications, that add to the functional repertoire of these molecules. Some of these modifications are catalyzed by stand-alone protein enzymes, while some others are catalyzed by RNA-protein complexes. tRNAs from all domains of life contain many such modifications, that increase their structural stability and refine their decoding properties. Certain regions of tRNAs are more frequently modified than others. Two such regions are the anticodon loop, and the TψC stem. In the halophilic euryarchaeon Haloferax volcanii, tRNATrp and tRNAMet, both of which are transcribed as intron-containing pre-tRNA forms, contain Cm34 and ψ54, in addition to other modifications, in these two regions, respectively. The Cm34 modification in both cases is RNP-mediated: tRNATrp Cm34 formation being guided by its own intron, while that of tRNAMet being guided by a unique guide RNA called sR-tMet. We created genomic deletion of H. volcanii tRNATrp intron by homologous recombination based technique, and showed that this strain is viable, and does not demonstrate any observable growth phenotype. However, the corresponding modifications are absent in this intron-deleted strain. Our structural and functional characterizations of sR-tMet revealed that it is unique in its structural properties and deviates considerably from its homologs in other Archaea. We also identified a novel L7Ae (a core protein associated with archaeal methylation guide RNPs) binding motif in sR-tMet. ψ54, the near universal modification found in TψC stem-loop of archaeal tRNAs is catalyzed by the protein Pus10. An earlier study from our laboratory had shown that Pus10 from two different archaea, Methanocaldococcus jannaschii (MjPus10) and Pyrococcus furiosus (PfuPus10) have differential activities towards ψ54 formation. Using the crystal structure of Human Pus10 as template, we created homology models of MjPus10 and PfuPus10 proteins and identified several residues and motifs that might lead to this difference in activity. By a combination of both in vitro and in vivo mutational approaches, we confirmed several previously unidentified residues/motifs that serve as positive determinants of tRNA ψ54 formation. Finally, as an extension to this study, we have identified a novel tRNA ψ54 forming activity in mammalian nuclear extracts, and attributed this activity to Pus10.

Repetitive sequences, including transposable elements, represent approximately half of the mammalian genome. Epigenetic mechanisms evolved to repress these potentially deleterious mobile elements. However, such elements can be variably silenced between individuals – so called ‘metastable epialleles’. The best known example is the Avy locus where an endogenous retrovirus (ERV) of the intracisternal A-particle (IAP) class was spontaneously inserted upstream of the agouti coat colour gene, resulting in variable IAP promoter DNA methylation, variable expressivity of coat phenotype, and environmentally modulated transgenerational epigenetic inheritance within genetically identical individuals. It is not known whether the behaviour exhibited by the ERV at Avy represents a common occurrence throughout the genome or is unusual. Taking a genetic approach in purified cell populations, I have conducted a systematic genome-wide screen of murine metastable epialleles. I have identified over 100 murine IAPs with properties of metastable epialleles. Like Avy, each exhibits a stable epigenetic state within an individual but epigenetic variability between individuals. Methylation levels are locus-specific within an individual, suggesting cis-acting control. The same screening strategy was applied for identification of metastable epialleles associated with other types of LTR-retroelements. However, many of identified candidates showed no inter-individual methylation variation upon experimental validation. These results suggest that IAPs are the dominant class of ERVs capable of acquiring epigenetic states that are variable between genetically identical individuals. I have conducted an analysis of IAP induced initiation and termination of transcription events using de novo assembled transcriptomes generated for B and T cells. 143 IAPs have been identified to overlap de novo assembled transcripts. 33 IAPs are metastable epialleles. Several of them show an inverse correlation between LTR promoter methylation and adjacent gene expression. In addition, I have shown that metastable epialleles have a characteristic pattern of histone modification and are flanked by the methylation sensitive binding factor CTCF, providing testable hypotheses concerning the establishment and/or maintenance of the variable methylation state. My findings indicate that metastability is, in general, specific to the IAP class of ERVs, that only around 1% of these elements have this unusual epigenetic property and that the ability to impact transcription, such as at agouti in Avy, is not a ubiquitous feature of these loci.

Methylation is a post-translational modification which occurs on lysine and arginine residues. Methylation is difficult to detect due to its low abundance and lack of charge. Our laboratory previously developed a novel enrichment approach, ProMENADe, for lysine and arginine methylation in the human embryonic kidney (HEK) 293T cell line which is coupled with mass spectrometry. Simplifying a lysate with subcellular fractionation prior to enrichment increased the identification of methylation sites by 39.5% while using multiple proteases for digestion increased identification by 27%. Combining these methods yielded a 47.2% increase. Analysis at the 1% methylation level FDR filtered for C-terminal methylation identified 169 sites and further analysis revealed 74 of these sites overlap with the PhosphoSite database. This ProMENADe enrichment strategy yielded 95 novel methylation sites to the field and can be a key tool in the field of methylation allowing for the enrichment and identification of methylated proteins.

DNA methylation is involved in various biological phenomena, and its dysregulation has been demonstrated as being correlated with a number of human disease processes, including cancers, autism, and autoimmune, mental health and neuro-degenerative ones. It has become important and useful in characterising and modelling these biological phenomena in or-der to understand the mechanism of such occurrences, in relation to both health and disease. An attempt has previously been made to map DNA methylation across human tissues, however, the means of distinguishing between methylated, unmethylated and differentially-methylated groups using DNA sequence features remains unclear. The aim of this study is therefore to: firstly, investigate DNA methylation classes and predict these based on DNA sequence features; secondly, to further identify methylation-associated DNA sequence features, and distinguish methylation differences between males and females in relation to both healthy and diseased, sta-tuses. This research is conducted in relation to three samples within nine biological feature sub-sets extracted from DNA sequence patterns (Human genome database). Two samples contain classes (methylated, unmethy-lated and differentially-methylated) within a total of 642 samples with 3,809 attributes driven from four human chromosomes, i.e. chromosomes 6, 20, 21 and 22, and the third sample contains all human chromosomes, which encompasses 1628 individuals, and then 1,505 CpG loci (features) were extracted by using Hierarchical clustering (a process Heatmap), along with pair correlation distance and then applied feature selection methods. From this analysis, author extract 47 features associated with gender and age, with 17 revealing significant methylation differences between males and females. Methylation classes prediction were applied a K-nearest Neighbour classifier, combined with a ten-fold cross- validation, since to some data were severely imbalanced (i.e., existed in sub-classes), and it has been established that direct analysis in machine-learning is biased towards the majority class. Hence, author propose a Modified- Leave-One-Out (MLOO) cross-validation and AdaBoost methods to tackle these issues, with the aim of compositing a balanced outcome and limiting the bias in-terference from inter-differences of the classes involved, which has provided potential predictive accuracies between 75% and 100%, based on the DNA sequence context.

The purpose of this study was to determine whether the methylation of nicotinamide to N-methylnicotinamide could discriminate between differences in methionine nutritional status, and by implication methylation capacity, in healthy humans. As part of this thesis, a highly selective high performance liquid chromatography (HPLC) method for the determination of N-methylnicotinamide (NMN) in urine and plasma was developed and validated. Quantification was by fluorescence detection of the 1,6-naphthyridine derivatives, formed after incubation of NMN with acetophenone in alkaline conditions. Seven volunteers participated in a trial to evaluate the ability of a nicotinamide load test to discriminate between changes in the methylation status of the individual. The methylation status was measured as the time dependent changes in plasma NMN concentrations after a nicotinamide load. A basal nicotinamide load test was performed on each individual. The methylation status was then changed, by means of a methionine load, and the nicotinamide load test was repeated during the enhanced methylation state. The dynamic changes in N-methylnicotinamide levels indicated that the methionine load changed neither the plasma NMN concentrations, nor the rates of NMN formation. The conclusion of this study was that nicotinamide loading could not be used as a dynamic function test to assess biological methylation in healthy humans. / Dissertation (MSc (Chemical Pathology))--University of Pretoria, 2006. / Chemical Pathology / unrestricted

Although structurally unrelated, the human killer cell immunoglobulin-like (KIR) and the rodent lectin-like Ly49 receptors serve similar functional roles in natural killer (NK) cells. Moreover, both gene families display variegated and mostly mono-allelic expression patterns established at the transcriptional level. DNA methylation, but not histone modifications, has recently been shown to play an important role in maintenance of the expression patterns of KIR genes but the potential role of DNA methylation in the expression of Ly49 genes was unknown. My thesis focuses on the role of epigenetic modifications, especially DNA methylation, in the maintenance of mouse Ly49 gene expression. I show that hypomethylation of the region encompassing the main promoter of Ly49a and Ly49c in primary C57BL/6 (B6) mouse NK cells correlates with expression of these genes. Using B6 x BALB/c Fl hybrid mice, I demonstrate that the expressed allele of Ly49a is hypomethylated while the non-expressed allele is heavily methylated, indicating a role for epigenetics in maintaining mono-allelic Ly49 gene expression. Furthermore, the Ly49a promoter region is heavily methylated in fetal NK cells but variably methylated in non-lymphoid tissues. In apparent contrast to the KIR genes, I show that histone acetylation state of the promoter region strictly correlate with Ly49A and Ly49G expression status. Also, the instability of Ly49G expression on some lymphoid cell lines is at least in part due to changes in the level of histone acetylation of the promoter region. As for the activating Ly49 receptors, it seems that although DNA methylation levels of the promoter regions do correlate with the state of expression of these receptors, the pattern of DNA methylation is different from that of the inhibitory Ly49a and c genes. In conclusion, my results support a role for epigenetic mechanisms in the maintenance of Ly49 expression. Moreover, these epigenetic mechanisms appear to vary among the Ly49 genes and also differ from those governing KIR expression. / Medicine, Faculty of / Medical Genetics, Department of / Graduate

The dynamic changes in chromatin play a significant role in lineage commitment and differentiation. These epigenetic modifications control gene expression through recruitment of transcription factors. While the active mark H3K4me3 is present around the transcription start site on the gene, the function of the H3K36me3 mark is unknown. A number of H3K36-specific histone methyltransferases (HMTs) have been identified, however the focus of this study is the HMT Hypb. To elucidate the role of H3K36me3 in mediating expression of developmentally-regulated loci, native chromatin immunoprecipitation (N-ChIP) was performed at a subset of genes. Upon differentiation, we observe that H3K36me3 becomes enriched at the 3’ end of several muscle-specific genes. To further investigate the role of H3K36me3 in myogenesis, a lentiviral-mediated knockdown of the H3K36 HMT Hypb was performed in muscle myoblasts using shRNA. Upon Hypb knockdown, we were surprised to observe enhanced myogenesis. N-ChIP was also performed on differentiated Hypb knockdown cell lines in order to look at H3K36me3 enrichment on genes involved in muscle differentiation. N-ChIP data show a drop in H3K36me3 enrichment levels on myogenin and Ckm genes. The possible occupancy of Hypb on the coding regions of muscle-specific genes was experimentally observed by cross-linked chromatin immunoprecipitation (X-ChIP) on differentiated C2C12 cells and subsequently confirmed by X-ChIP on knockdown lines where the occupancy was lost. A model is proposed that links the observed phenotype with H3K36me3.

Histone H3 lysine 4 (H3K4) methylation is an evolutionarily conserved mark commonly associated with transcription activation in eukaryotes. In mammals, this post-translational modification is deposited by the KMT2 family of H3K4 methyltransferases. Biochemical studies have shown that the enzymatic activity of the KMT2 enzymes is regulated by a core complex of four evolutionarily conserved proteins: WDR5, RbBP5, ASH2L and DPY30, collectively known as WRAD, which are all important for global H3K4 methylation. However, how these proteins interact and regulate the activity of the KMT2 enzymes is not well investigated. During my PhD, I have used structural and biochemical approaches to determine the interactions underlying formation of the core complex and regulation of KMT2 enzymatic activity. My research have shown that 1) WDR5 uses two peptide-binding clefts on opposite sides of its β-propeller domain to bridge the KMT2 enzymes to the regulatory subunit RbBP5, 2) the WDR5 peptidyl-arginine-binding cleft exhibits plasticity to accommodate the binding of all KMT2 enzymes and 3) RbBP5 S350 phosphorylation stimulates formation of the RbBP5-ASH2L complex and H3K4 methylation by the mammalian KMT2 enzymes. Collectively, these studies have provided the structural basis for understanding the important interactions governing KMT2 complex assembly and activity.

UHRF1 is a critical epigenetic regulator which serves as a molecular model for understanding the crosstalk between histone modification and DNA methylation. It is integrated in the process of DNA maintenance methylation through its histone ubiquitylation activity, ultimately functioning as a recruiter of DNA methyltransferase 1 (DNMT1). As the faithful propagation of DNA methylation patterns during cell division is a common molecular phenomenon among vertebrates, understanding the underlying conserved mechanism of UHRF1 for executing such a key process is important. Here, I present a broad-range evolutionary comparison of UHRF1 binding behavior and enzymatic activity of six species spanning across the vertebrata subphylum. According to their distinct binding modes to differentially methylated histone H3, a pattern is emerging which separates between mammalian and nonmammalian orthologs. H. sapiens, P. troglodytes and M. musculus UHRF1 orthologs utilize the functionality of both TTD and PHD domains to interact with histone H3 peptides, while G. gallus, X. laevis, and D. rerio employ either TTD or PHD. Further, UHRF1 allosteric regulation by 16:0 PI5P is a unique case to primate orthologs where H3K9me3 peptide binding is enhanced upon hUHRF1 and pUHRF1 interacting with 16:0 PI5P. This is due to their closed and autoinhibited conformation wherein TTD is blocked by the PBR region in linker 4. 16:0 PI5P outcompetes TTD for PBR binding resulting in a release of TTD blockage, hence, enhanced H3K9me3 binding. However, owing to the lack of phosphatidylinositol binding specificity and reduced sequence conservation of linker 4, the regulatory impact of 16:0 PI5P in avian and lower vertebrate orthologs could not be detected. Additionally, all UHRF1 orthologs exert their ubiquitylation enzymatic activity on histone H3 substrates, supporting the notion that the overall functionality of UHRF1 orthologs is conserved, despite their divergent molecular approaches. Taken together, my findings suggest that UHRF1 orthologs adopt distinct conformational states with a differential response to the allosteric regulators 16:0 PI5P and hemi-methylated DNA.

This study provides information about the kinetics of the hydrolytic ring cleavage of imidazolium compounds and the catalytic effect of micelles in this ring cleavage. 9-Methylcaffeine iodide was synthesized and its base catalyzed ring cleavage was studied. The ring cleavage, via alkaline hydrolysis of the imidazole moiety, showed pseudo-first order kinetics over the pH range of 7.28 to 10.81. Activation parameters were found to be: Ea = 29.4 Kcal mole-1, △H+ = 28.8 Kcal mole-1 , and △S+ = 27.4 cal mole-l deg -1 The kinetics were studied in the presence of micelles and no rate enhancement was observed. The mechanism for ring cleavage was also investigated. 9-Ethylcaffeine iodide was synthesized and its base catalyzed ring cleavage, at 30°C and a pH of 9.42, showed pseudo first order kinetics. 6 -3 -1 The kobs was found to be 1. 5 x 10-3 sec-1. The oxidation kinetics of 8 ,8-dihydro-9-methylcaffeine to 9-methylcaffeinE:· iodide were also examined.