anscriptionally silent. Therefore we have chosen the term “Gag-positive reservoir cell” to define this population. As these 11095475 patients examined for GPR cells in vivo were well controlled on HAART for at least 2 years, the Gag production reflects a portion of the reservoir that is expressed and resistant to anti-retroviral therapy. Regardless of whether or not these cells are latent, GPR cells represent a potential target for the immune system. Our results have clear implications for therapeutic design. As GPR cells can be targeted by CTL, therapeutically enhancing the immune system may allow the clearance of these cells, which would potentially reduce the reservoir. Our findings that an effective CTL response might reduce the size of the reservoir in the absence of HIV replication is broadly consistent with a recent study that found an inverse correlation between mucosal HIV-specific T cells responses and the frequency of HIV DNA-containing PBMC as well as the findings that effective vaccination in non-human primates and therapeutic vaccination in humans SKI-II web appeared to block establishment of or reduce the latent reservoir, respectively. On the whole, our data indicate the importance of CTL in controlling HIV reservoirs. While it seems unlikely that HIV can be cured by enhanced CTL alone, our data suggest it can be dramatically reduced. Thus, efforts to boost the immune response should be pursued not only to control infection but in conjunction with other efforts to enhance HIV expression in order to cure HIV. Experimental Procedures Ethics Statement and study subjects EC and untreated/chronically infected subjects were recruited from the Clinical Reasearch Center, NIH, the Center for Aids Research at the University of Pennsylvania or the SCOPE cohort at the University of California, San Francisco. All participants signed informed consent forms approved by the NIAID IRB, the University of Pennsylvania’s IRB and UCSF’s IRB. The University of Pennsylvania IRB approved the transfer of materials from NIH and UCSF. The EC used for in vitro studies were previously described. The chronically infected noncontrollers were also previously described. PBMC from normal donors were obtained through anonymous 
donation to the University of Pennsylvania’s Human Immunology Core. Isolation of resting CD4+ T cells Frozen PBMC were thawed and resting CD4+ T cells were negatively selected by depletion of lineage markers CD20, CD16, CD14, CD56, CD8, BDCA2, CD11c, and activation markers following the MACS LD depletion protocol. Resting CD4+ T cells were >99% pure. Inoculation and Gag staining Transfection supernatants were prepared by the Penn CFAR using HIV-1NL4-3. Spinoculation, with virus at an MOI of 3, was used for Spreading infection Resting CD4+ T cells from EC isolated as above were cultured for 3 days in RPMI+10% FCS alone or with anti-CD3/ 10 The Visible HIV Reservoir Can Be Cleared by CTL CD28 beads at a concentration of 3 beads/cell + 100U/mL IL-2. Afterwards, cells were washed, spinoculated and DNase treated as above. Half of each sample were treated with 1.25M SQV. Cells were collected. Cells per well were quantified by real-time PCR for the -Globin gene and total HIV DNA was measured by real-time PCR as in. quantitative PCR was used to determine cell quantity as described. Integration and 2-LTR 20573509 copies per cell from cocultured samples were multiplied by 10 to account for the dilution factor of effectors to targets. HLA-B57/27 and non-B57/27 EC measurem