PROJECT SUMMARY HIV infection of the CNS is an important source of morbidity and mortality in the chronic phase of disease, despite availability of combination Antiretroviral Therapy (cART). Although early and wide use of cART has reduced the occurrence of the more severe CNS pathologies, milder neurocognitive impairment is documented in 20-50% of infected individuals. HIV invasion of the CNS is thought to occur within the first two weeks of infection, where the virus infects perivascular macrophages and microglia, in addition to CD4+ central memory T cells (TCCM), the primary infected cell in the periphery. The vast majority of the peripheral reservoir exists as a very minor fraction of resting CD4+ TCCM and is thought to be maintained by the clonal expansion of latently infected cells. Thus far, HIV reservoir characterization has focused on either identifying and classifying integration sites or examining the integrity of the integrated proviral genomes, rarely both concurrently. Latency studies are most often performed with patient-derived PBMC and may not adequately represent unique features of tissue-specific reservoirs. In particular, studies of the CNS-specific HIV reservoir are few and limited by sample access. The expanded cell tropism of brain HIV and the potential for low level, ongoing viral replication are suggestive that reservoir dynamics may be dramatically different in the CNS compared to plasma; however, the CNS reservoir remains poorly characterized with regard to proviral genome integrity, integration sites, emergence and reservoir fixation of viral variants and the contribution of clonal expansion to reservoir maintenance. Moreover, while CD4+ T memory cells can proliferate; terminally differentiated myeloid and glial cells, the primary targets of HIV infection in the brain, are long-lived with limited capacity for self- renewal. Standard methods for profiling HIV reservoirs are dependent on short read next generation sequencing (NGS) technologies that require the examination of integration sites to be necessarily divorced from the characterization of their associated proviral genomes. Short read NGS interrogation of integration sites limits the mapability of the resulting data, which may further limit the identification of HIV integration sites; while the use of single genome amplification (SGA) for proviral genome characterization is low throughput and labor intensive. Here we have developed a novel, innovative HIV-specific molecular enrichment approach, combined with single molecule sequencing (“HIV SMRTcap”), which resolves the complete HIV “integron” (flanking integration sites and associated provirus) regardless of genomic context. We propose to apply this technology to the characterization of HIV-infected CNS and lymphoid tissues, provided by the National NeuroAIDS Tissue Consortium and including both viremic (n=3) and cART suppressed (n=3) patients, to define, for the first time, the contribution of clonal expansion t...