HIV-1 regulates the expression of many host cell genes during infection of human cells . Dysregulation of host genes occurs when viral molecules interact with cell components, disrupt normal cellular pathways, recruit host factors for viral replication or change the endogenous miRNA profile of the cell [50–53]. Here, we report that HIV-1 TAR miRNAs could be another mechanism by which HIV-1 regulates the expression of host genes. We demonstrated that miR-TAR-5p and miR-TAR-3p were incorporated into Ago complexes, which contains the effector components of miRNA-targeted RNA silencing  in human cells  and is required for small RNA species to mediate mRNA regulatory effects . We established stable Jurkat cell lines expressing the HIV-1 TAR miRNAs and we selected four lines that showed variations in TAR miRNA expression levels. Based on the premises that (i) miRNAs regulate mRNAs in a dose-dependent manner , and that (ii) increased miRNA levels would yield a stronger phenotype, the Jurkat TAR-3 cell line was selected for further proteomic analyses and NPM/B23 emerged as a top protein candidate downregulated by the HIV TAR miRNAs. The 2D-gel analysis of a clonal population (i) circumvented a relatively high background associated to poorly infected cells, as previously described , and (ii) allowed us to study a stable latently-infected cell population, in contrast with other early-infected cell studies [56, 57]. This approach, however, also has its intrinsic limitations, the most important of which may be (i) the poor separation of proteins from some areas of the gel, and (ii) a fold change too small to justify mass spectrometry analyses.
A dual luciferase reporter assay was used to confirm the host gene targets we identified in silico and to monitor the 3’UTR-mediated regulation of the genes of interest by the TAR miRNAs. The dual luciferase assay, although powerful, has limitations, all of which can differently influence the results . These include the type of cells studied, the length of the regulatory element (e.g. natural 3’UTR versus 3X binding sites), the binding site accessibility and the architecture of the miRNA:mRNA base pairing. Data should be interpreted with caution, because modest reporter gene downregulation (~20%) has been previously observed in assays where the 3’UTR tested was fused to luciferase genes [31, 59], even when the predicted binding sites had a free energy (ΔG) as low as −30,7 kcal/mol, expecting a much stronger regulation .
Three of the four genes that we selected from our reporter gene activity data were downregulated by TAR miRNAs expressed in the Jurkat cell model. Although the protein levels of Caspase 8, Ikaros and NPM/B23 were downregulated to varying degrees in the Jurkat cells expressing TAR and the J-Lat cell lines, TAR miRNAs did not induce comparable changes at the mRNA level. Therefore, although reduced mRNA levels generally account for most of the decreased protein production , it would be imprudent to rely exclusively on mRNA levels when assessing the effects and importance of TAR miRNAs in regulating host gene expression. Hence, the relevance of our proteomic analysis of Jurkat cells expressing TAR miRNAs that identified NPM/B23 as an mRNA/protein that is targeted by the HIV-1 TAR miRNAs.
The Aiolos gene was upregulated when TAR was expressed at a high level in the Jurkat cells or when the HIV-1 was integrated in the cells as a provirus. The upregulation of a gene upon miRNA binding has been observed with small activating RNAs (saRNAs), which are reported to enhance gene expression [61, 62]. In addition, a given miRNA binding site can mediate repression in some 3’ UTRs, but not others , which could also explain the result. The overall effect of the TAR miRNAs on Aiolos expression may also involve regulation at the transcriptional level and, because Aiolos is a transcription factor, there may be complex feedback loops involved. The TAR miRNA downregulatory effect on Aiolos might be related to other stages of the HIV-1 pathogenesis or could occur in different cells types that have not been tested in this study.
Apoptosis is known to be a significant cause of HIV-1-infected, CD4+ T cell death. However, the molecular mechanism determining the balance between cell survival and cell death remains unclear. Our results suggest that the regulation of selected host mRNAs by TAR miRNAs could influence the choice made by HIV-1 infected cells between survival and apoptosis. This could be the result of a complex interplay involving the up- and down regulation of the levels of pro- and anti-apoptotic proteins by the TAR miRNAs, and the secondary effects on the cellular machinery. Furthermore, because HIV-1 infection is a multi-step process, these cell components may be regulated differently throughout the course of the infection. For instance, up-regulation of the pro-apoptotic Caspase 8, which cleaves and activates other downstream caspases, renders cells susceptible to apoptosis via Fas signaling . Caspase 8 is cleaved by the HIV-1 protease into Casp8p41, a fragment strongly associated with apoptotic, HIV-1 infected CD4+ T cells . HIV-1 could, for example, delay or prevent apoptosis through TAR miRNA-mediated downregulation of Caspase 8 early in the infection cycle, to ensure robust viral replication and packaging.
Another downregulated gene was Ikaros (also called IKZF1), which is the founding member of a family of zinc finger transcription factors that also includes Aiolos (also called IKZF3) . Ikaros increases normal oxidative stress-induced apoptosis in erythroid cells . By reducing the level of Ikaros protein in T cells, TAR miRNAs could decrease apoptotic events that are associated with HIV-1 infection. However, because etoposide treatment can shorten the half-life of Ikaros , we cannot exclude the effect of etoposide on Ikaros as an explanation for the resistance of the etoposide-treated Jurkat TAR and J-Lat cells to apoptosis.
Upregulation of Aiolos expression could also potentially reduce apoptosis of HIV-1 infected cells. Aiolos is a transcription factor whose expression is restricted to lymphoid lineages. Aiolos binds to the Bcl-2 promoter, and also interacts with the Bcl-2 and Bcl-XL proteins [67, 68] to enhance their stability and promote cell survival. Because a high Aiolos mRNA and protein expressions seemed to correlate with TAR miRNA levels and a resistance to apoptosis, it may be related to survival in infected cells.
The effects we have seen, e.g. cell survival and resistance to apoptosis, in TAR-expressing and J-Lat cells lines may also be explained by the downregulation of NPM/B23 protein. NPM/B23 is a major nucleolar, multifunctional protein that has been reported to interact with the HIV-1 proteins Tat and Rev [69, 70] and with many cellular components . A down regulation of NPM/B23 protein expression in the nucleolus could modify protein interactions and disrupt HIV-1 nucleolar Rev localization  to promote its interaction with other cellular or viral components. In the acute phase of HIV-1 infection, NPM/B23 expression is increased  and its acetylated form is recruited, in a Tat-dependant manner, to the HIV-1 LTR to enhance viral transactivation . The targeting of NPM/B23 could be a mechanism used by the TAR miRNAs to limit the replication of HIV-1 thereby facilitating the escape of virus-infected cells from apoptosis and promoting latency stage.
The present study should improve our understanding of the regulation of host genes by viral miRNAs and the potential outcomes. Because hundreds of mRNAs can be targeted by a single miRNA, and miRNAs may act in concert to regulate mRNA translation , we speculate that the TAR miRNAs of HIV-1 have evolved to play a key role in viral pathogenesis, most likely by promoting conditions that favor HIV-1 replication in host cells. Future studies will elucidate the molecular mechanisms by which the HIV-1 miRNAs are produced and how they regulate host gene expression, to determine their relative importance in the HIV-1 replicative cycle and pathogenesis. These studies should provide key insights into how HIV-1 viral miRNAs contribute to shaping the host response to viral infection and identify new, potential targets for development of novel and improved therapeutics treat HIV-1.