Philosophiae Doctor - PhD / The plasticity of single-stranded viral genomes permits the formation of secondary structures through complementary base-pairing of their component nucleotides. Such structures have been shown to regulate a number of biological processes during the viral life-cycle including, replication, translation, transcription, post-transcriptional editing and genome packaging. However, even randomly generated single-stranded nucleotide sequences have the capacity to form stable secondary structures and therefore, amongst the numerous secondary structures formed in large viral genomes only a few of these elements will likely be biologically relevant. While it is possible to identify functional elements through series of laboratory experiments, this is both excessively resource- and time-intensive, and therefore not always feasible. A more efficient approach involves the use of computational comparative analyses methods to study the signals of molecular evolution that are consistent with selection acting to preserve particular structural elements. In this study, I systematically deploy a collection of computationally-based molecular evolution detection methods to analyse the genomes of viruses belonging to a number of ssRNA viral families (Alphaflexiviridae, Arteriviridae, Caliciviridae, Closteroviridae, Coronavirinae, Flaviviridae, Luteoviridae, Picornaviridae, Potyviridae, Togaviridae and Virgaviridae), for evidence of selectively stabilised secondary structures. To identify potentially important structural elements the approach incorporates structure prediction data with signals of natural selection, sequence co-evolution and genetic recombination. In addition, auxiliary computational tools were used to; 1) quantitatively rank the identified structures in order of their likely biological importance, 2) plot co-ordinates of structures onto viral genome maps, and 3) visualise individual structures, overlaid with estimates from the molecular evolution analyses. I show that in many of these viruses purifying selection tends to be stronger at sites that are predicted to be base-paired within secondary structures, in addition to strong associations between base-paired sites and those that are complementarily co-evolving. Lastly, I show that in recombinant genomes breakpoint locations are weakly associated with co-ordinates of secondary structures. Collectively, these findings suggest that natural selection acting to maintain potentially functional secondary structures has been a major theme during the evolution of these ssRNA viruses.

The human immunodeficiency virus (HIV) and acquired immune deficiency syndrome (AIDS) that subsequently develops remain major health concerns even after three decades since the first cases were reported. Successful therapeutic measures to address HIV/AIDS consist mostly of combinations of drugs targeting viral enzymes including reverse transcriptase (RT), protease (PR) and integrase (IN) as well as entry steps of the viral life cycle. The remarkable benefits (e.g. improved quality of life) derived from the use of these agents are unfortunately limited by toxicity to the host and the development of drug resistant viral strains. Drug resistance limits the repertoire of drug combinations available. Unfortunately, because latent forms of the virus exists, therapy has to be life-long and with new infections occurring every day, resistant strains tend to spread. To circumvent these problems, new drugs that inhibit resistant strains or work against new viral targets have to be developed. The history of gold compounds as potential inhibitors of HIV prompted this study in which twenty seven compounds consisting of gold(I), gold(III) and precursors from five classes were tested for drug-likeness, anti-HIV and immunomodulatory effects using wet lab and in silico methodologies. Cytotoxicity determination was done using viability dyes and flow cytometry. Cell proliferation profiles were monitored using the carboxyflourescein succinimidyl ester dye dilution technology and a real time cell analyser for confirming viability dye findings. The compounds’ effects on viral enzymes was determined using direct enzyme assays and in silico molecular modelling techniques. H and P nuclear magnetic resonance spectroscopy studies for determining stability revealed that the backbone chemical shifts of the compounds were relatively unchanged after one week (-20 and 37 ºC) when dissolved in dimethylsulfoxide. Eight of the gold compounds had drug-like properties comparable to clinically available drugs when in silico predictions were performed. The 50% cytotoxic dose of the compounds in human cells was between 1 and 20 μM (clinically relevant concentrations for gold compounds). Three gold(I) compounds inhibited viral infectivity at non-toxic concentrations and two gold(III) compounds did so at cytostatic (anti-proliferative mechanism that is also antiviral) concentrations. In the immunomodulatory assay, cytokine levels were altered by five compounds with one gold(I) and a gold(III) compound significantly reducing the frequency of CD4+ cells (an anti-viral function) from HIV+ donors (p= 0.005 and 0.027 respectively) when multi-parametric flow cytometry was performed. Inhibition of RT activity was predicted in in silico studies to be through interactions with the ribonuclease (RNase) H site although with poor stereochemical orientation while favourable binding predictions with the IN cofactor binding site were observed for some gold(III) complexes. Compounds predicted to interact with the RNase H site of RT and the IN cofactor site require structural modification to improve drug-likeness and binding affinity. The drug-like compound(s) which inhibited viral infectivity and lowered CD4+ cell frequency have potential for incorporation into virostatic cocktails (combination of cytostatic and directly anti-viral agent). Cytostatic agents are known to be less prone to drug resistance and because they lower CD4+ cell frequency, such compounds can potentially limit HIV immune activation. / Thesis (PhD)--University of Pretoria, 2011. / Biochemistry / unrestricted