Background: Understanding the molecular mechanisms associated with the exceptional quasi-equilibrium between HIV reservoirs and host cells in the model of functional cure represented by Elite Long-Term Nonprogressors (E-LTNPs) should provide clues for developing strategies to purge the HIV reservoirs but remain to be elucidated. Methodology: We compared the whole-genome transcriptional profiling of the 4 major resting (CD25-CD69-DR-) CD4 T cell subsets (naive [TN], centralmemory [TCM], transitional-memory [TTM] and effector-memory [TEM]) sorted from PBMCs of 14 HIV-1 infected individuals including 7 E-LTNPs, 7 viremic LTNPs (Vir-LTNPs) and 7 uninfected individuals (UI). Total HIV DNA was quantified in each T cell subset by real time quantitative polymerase chain reaction (qPCR, Biocentric, Bandol, France). The transcriptomes were analyzed using an Illumina Platform. Statistics were performed using the Limma package of R. Differentially expressed genes (DEGs) were considered significant with a false-discovery rate (FDR) below 0.05. Key molecular signatures were quantified by qPCR in each sorted CD4 T cell subset in parallel to the HIV-1 cellular DNA and unspliced RNA levels. Results: Resting TCM and TTM were the main contributors to the HIV reservoir in LTNPs with a 1-log difference for each subset HIV-DNA content between E-LTNPs and Vir-LTNPs. When comparing each E-LTNP, Vir-LTNP and UI CD4 T-cell subset transcriptome to a single comparator (UI-TN), an increasing gradient of DEGs followed the linear TN to TCM, TTM and TEM differentiation model. Distinct profiles were however observed between groups with the most active and distant transcriptomes in TCM from E-LTNPs but in TEM from Vir-LTNPs. Most DEGs encoded proteins with functions related to intracellular signaling, adherence, cytotoxicity and apoptosis. The E-LTNPs TCM, representing the main CD4 compartment, revealed 2 major specific signatures compared to TCM from Vir-LTNPs and UI: (i) TCR and costimulation signaling genes with 8 overexpressed and 2 down-modulated DEGs, and (ii) type I interferon-related genes with 5 down-modulated and 2 upregulated DEGs. This specific E-LTNPs transcriptome was associated to extremely low levels of HIV-DNA and HIV-RNA transcripts in TCM from E-LTNPs compared to other subsets and to TCM from Vir-LTNPs with significant association between host and HIV genes. Conclusions: Resting CD4 TCM from E-LTNPs display a unique transcriptional profile with key host genome signatures from the TCR/costimulatory signaling and IFN-I pathways associated to low levels of HIV reservoirs and HIV-RNA transcription. These results allow identification of new molecular targets associated with the control of the HIV reservoirs in models of functional cure.