PDF, English - main document Download (6MB) | Terms of use

Abstract

The infection with HIV-1 nowadays does not represent a condition with a deadly outcome. Due to current therapeutic approaches, the infection with HIV-1 represents a chronic condition in which viral load is kept at undetectable levels, but patients depend on a lifelong therapy without a chance of cure. The eradication of integrated viral DNA still remains the biggest challenge in curing HIV-1. The aim of this work was to contribute to a better understanding and definition of genomic regions and epi-genomic features that HIV-1 targets for integration, and give a detailed description on the importance of chromatin accessibility, as well as the importance of certain genomic features in the process of HIV-1 integration. The first part of this project deals with the importance of HMT G9a activity and H3K9me2 histone mark distribution and deposition in the context of HIV-1 integration in primary CD4+ T cells, which was studied by the application of G9a inhibitor BIX0129, also known as a very potent latency reversing agent. The significance of G9a activity and facultative heterochromatin mark H3K9me2 deposition has previously been shown to affect T cell development and impact shaping of the nuclear architecture. In this work it was demonstrated that the chemical inhibition of G9a and depletion of H3K9me2 by BIX01294 has an increasing effect on HIV-1 integration. The increase in integration was also followed by increased viral transcriptional activity, as well as spatial repositioning of the provirus from the preferred nuclear periphery towards the nuclear center. Similar spatial repositioning has been demonstrated for genes highly and recurrently targeted by HIV-1 for integration (RIGs). However, genic nuclear repositioning upon BIX01294 treatment did not affect transcriptional profiles of HIV-1 RIGs, as demonstrated by RNA microarray analysis, but other groups of genes mainly involved in iron metabolism and inflammatory response were upregulated upon BIX01294 treatment. In addition, HIV-1 integration patterns were shown not to be affected by H3K9me2 depletion, and the virus was still targeting similar genic regions for integration. The analysis of chromatin mark distribution and chromatin binding elements upon BIX01294 treatment on RIGs revealed increased binding profiles of open chromatin mark H3K36me3 which is followed by increased LEDGF/p75 binding upon H3K9me2 depletion. The observed phenomenon might provide an explanation for the observed increased viral integration upon BIX01294 treatment, considering that LEDGF/p75 is a prominent host cell factor involved in the viral integration process. Overall, the first part of this study clearly demonstrated that chromatin accessibility significantly affects HIV-1 integration levels which are directly proportional to viral expression levels and viral activity. The second part of this study deals with the relevance of R-loops, as specific genomic structures, as sites selected for HIV-1 integration in primary CD4+ T cells and macrophages. It was demonstrated that the GFP tagged IN enzyme of HIV-1, in a high occurrence, colocalizes with R-loops in cells, and that for the occurrence of this process a functionally active IN is required. This finding implicated that the observed colocalization is not randomly taking place and that HIV-1 is actively docked to R-loop forming genomic sites. In addition, biochemical as well as computational meta data analysis revealed that HIV-1 RIGs are enriched in R-loops and that R-loop forming sites can accommodate integrated viral DNA. Further on, it was demonstrated that HIV-1 IN has R-loop binding capacity and is also capable of performing the strand transfer reaction on R-loop containing DNA templates. It was also demonstrated that R-loop depletion by RNase H1 overexpression in several cell lines, as well as in primary cells, significantly impairs HIV-1 integration, indicating that R-loop presence is crucial for efficient HIV-1 integration. In line with this result was the finding that RIGs expression was not affected by R-loop removal, indicating that only the presence of R-loops, as structural genomic elements, is more affecting HIV-1 integration compared to gene expression levels. The final finding is also in line with previous work from our lab. In summary, the second part of this study provides strong evidence that R-loops represent structural genomic elements targeted by HIV-1 for integration and also gives new insight into HIV-1 IN functional III features which have not been addressed before.