To establish retroviral infection, a DNA copy of the viral RNA genome needs to be inserted as a provirus in the host chromatin: a reaction catalysed by the virally encoded enzyme integrase (IN). Integration site selection is influenced by interactions between integrase and host proteins; lentiviral pre-integration complexes (PICs) are targeted to the body of transcription units by LEDGF, whilst BET proteins tether gamma-retroviral PICs to transcription start sites. There are no hot spots of integration of human T-cell lymphotropic virus type 1 (HTLV-1, a delta-retrovirus), but proviruses are enriched in actively transcribed regions of the genome, with a minor but significant proportion of integration sites in proximity to certain transcription factor binding sites.
Previous work from our laboratory identified B56-protein phosphatase 2A (PP2A) as a binding partner and putative host factor for delta-retroviral integration. The work in this thesis builds on the hypothesis that PP2A-B56 is usurped by HTLV-1 during infection from two distinct angles. Firstly, the X-ray crystal structure of B56 in complex with HTLV-1 IN was solved, which highlighted that HTLV-1 mimics a Short Linear Motif found in endogenous binding partners of B56 and informed the design of IN mutants which retain wild-type intrinsic integration activity but do not bind B56 members. These mutants displayed a defect in infection compared to the wild type, as measured by reductions in transactivation of a reporter system and the proviral load. Secondly, a strategy is described for the generation of a B56 KO cell line negative for most B56 members, which will be infected with WT HTLV-1.