<p>Nevirapine (NVP) is a cheap anti-retroviral drug used in poor countries worldwide, administered to pregnant women at the onset of labour to inhibit HIV enzyme reverse transcriptase. Viruses which may get transmitted to newborns are deficient in this enzyme, and HIV-1 infection cannot be established, thereby preventing mother to child transmission (MTCT). In some cases, babies get infected and positive selection for viruses resistant to nevirapine may be inferred. Positive selection can be inferred from sequence data, when the rate of nonsynonymous substitutions is significantly greater than the rate of synonymous substitutions.</p> <p>Unfortunately, it is found that available positive selection methods should not be used to analyse before- and after- NVP treatment sequence pairs associated with MTCT. Methods which use phylogenetic trees to infer positive selection trace synonymous and nonsynonymous substitutions further back in time than the short time duration during which selection for NVP occurred. The other group of methods for inferring positive selection, the pairwise methods, do not have appreciable power, because they average susbtituion rates over all codons in a sequence pair and not just at single codons. We introduce a simple counting method which we call the Pairwise Homologous Codons (PHoCs) method with which we have inferred positive selection resulting from NVP treatment in longitudinal HIV-1 reverse transcriptase sequences. The PHoCs method estimates rates of substitutions between before- and after- NVP treatment codons, using a simple pairwise method.</p>

<p>Nevirapine (NVP) is a cheap anti-retroviral drug used in poor countries worldwide, administered to pregnant women at the onset of labour to inhibit HIV enzyme reverse transcriptase. Viruses which may get transmitted to newborns are deficient in this enzyme, and HIV-1 infection cannot be established, thereby preventing mother to child transmission (MTCT). In some cases, babies get infected and positive selection for viruses resistant to nevirapine may be inferred. Positive selection can be inferred from sequence data, when the rate of nonsynonymous substitutions is significantly greater than the rate of synonymous substitutions.</p> <p>Unfortunately, it is found that available positive selection methods should not be used to analyse before- and after- NVP treatment sequence pairs associated with MTCT. Methods which use phylogenetic trees to infer positive selection trace synonymous and nonsynonymous substitutions further back in time than the short time duration during which selection for NVP occurred. The other group of methods for inferring positive selection, the pairwise methods, do not have appreciable power, because they average susbtituion rates over all codons in a sequence pair and not just at single codons. We introduce a simple counting method which we call the Pairwise Homologous Codons (PHoCs) method with which we have inferred positive selection resulting from NVP treatment in longitudinal HIV-1 reverse transcriptase sequences. The PHoCs method estimates rates of substitutions between before- and after- NVP treatment codons, using a simple pairwise method.</p>

Integrase catalyzes insertion of a retroviral genome into the host chromosome. Following reverse transcription, integrase binds specifically to the ends of the duplex retroviral DNA, endonucleolytically cleaves two nucleotides from each 3'-end (the processing activity), and inserts these ends into the host DNA (the joining activity) in a concerted manner. Additionally, it has been observed that integrase can catalyze the removal of inserted viral ends (the disintegration activity) in vitro. Presteady-state experiments were performed using synapsed substrates to probe the processing reaction and a disintegration substrate to determine the number of protomers in a functional multimeric complex. In single-turnover studies, a novel "splicing" reaction was observed that revealed complications with accurate quantification of enzymatic activity using the synapsed substrates. The splicing reaction was further used to gain insight into the selection of nucleophiles and electrophiles at the binding site. To reduce the complexity introduced by the integrase-catalyzed splicing reaction, 5'-5' reverse-polarity synapsed substrates were designed that were not susceptible to the splicing reaction and that allowed direct comparison of LTR ends simultaneously bound at the active site. Analysis of the presteady-state assays using these reverse-polarity substrates revealed that the concurrent binding of the biologically relevant U3/U5 combination of viral ends facilitates maximal activity of the processing reaction. A disintegration substrate was used in presteady-state active site titrations to determine a reaction stoichiometry of four integrase protomers per one substrate molecule for the disintegration reaction. A tetrameric active complex was then confirmed using atomic force microscopy to image integrase-DNA complexes during the first catalytic turnover. The observed increase of the tetramer population in the presence of substrate DNA demonstrates that the binding of the disintegration substrate induces assembly of the active tetramer and suggests that tetramer assembly may be an integral and dynamic component of the catalytic pathway. / Graduation date: 2003