Interferon-Inducible MicroRNA miR-128 Modulates HIV-1 Replication by Targeting TNPO3 mRNA

HIV-1 is the causative agent of AIDS. During HIV-1 infection, type I interferons (IFNs) are induced, and their effectors limit HIV-1 replication at multiple steps in its life cycle. However, the cellular targets of INFs are still largely unknown. In this study, we identified the interferon-inducible microRNA (miR) miR-128, a novel antiviral mediator that suppresses the expression of the host gene TNPO3, which is known to modulate HIV-1 replication. Notably, we observe that anti-miR-128 partly neutralizes the IFN-mediated block of HIV-1. Elucidation of the mechanisms through which miR-128 impairs HIV-1 replication may provide novel candidates for the development of therapeutic interventions.

ABSTRACT

The HIV/AIDS pandemic remains an important threat to human health. We have recently demonstrated that a novel microRNA (miR), miR-128, represses retrotransposon long interspaced element 1 (L1) by a dual mechanism, namely, by directly targeting the coding region of the L1 RNA and by repressing a required nuclear import factor (TNPO1). We have further determined that miR-128 represses the expression of all three TNPO proteins (transportins TNPO1, TNPO2, and TNPO3). Here, we establish that miR-128 also influences HIV-1 replication by repressing TNPO3, a factor that regulates HIV-1 nuclear import and viral; replication of TNPO3 is well established to regulate HIV-1 nuclear import and viral replication. Here, we report that type I interferon (IFN)-inducible miR-128 directly targets two sites in the TNPO3 mRNA, significantly downregulating TNPO3 mRNA and protein expression levels. Challenging miR-modulated Jurkat cells or primary CD4⁺ T-cells with wild-type (WT), replication-competent HIV-1 demonstrated that miR-128 reduces viral replication and delays spreading of infection. Manipulation of miR-128 levels in HIV-1 target cell lines and in primary CD4⁺ T-cells by overexpression or knockdown showed that reduction of TNPO3 levels by miR-128 significantly affects HIV-1 replication but not murine leukemia virus (MLV) infection and that miR-128 modulation of HIV-1 replication is reduced with TNPO3-independent HIV-1 virus, suggesting that miR-128-indued TNPO3 repression contributes to the inhibition of HIV-1 replication. Finally, we determine that anti-miR-128 partly neutralizes the IFN-mediated block of HIV-1. Thus, we have established a novel role of miR-128 in antiviral defense in human cells, namely inhibiting HIV-1 replication by altering the cellular milieu through targeting factors that include TNPO3.

IMPORTANCE HIV-1 is the causative agent of AIDS. During HIV-1 infection, type I interferons (IFNs) are induced, and their effectors limit HIV-1 replication at multiple steps in its life cycle. However, the cellular targets of INFs are still largely unknown. In this study, we identified the interferon-inducible microRNA (miR) miR-128, a novel antiviral mediator that suppresses the expression of the host gene TNPO3, which is known to modulate HIV-1 replication. Notably, we observe that anti-miR-128 partly neutralizes the IFN-mediated block of HIV-1. Elucidation of the mechanisms through which miR-128 impairs HIV-1 replication may provide novel candidates for the development of therapeutic interventions.

INTRODUCTION

The HIV/AIDS pandemic remains an important threat to human health. Currently, more than 34 million people are infected with HIV. Unique characteristics make HIV-1 difficult to eradicate. HIV-1 uses and evades the immune system, replicating in and destroying immune cells during active replication. To develop novel and widely applicable strategies, a thorough understanding of the cellular determinants of HIV-1 replication and viral latency is crucial, including epigenetic changes and transcriptional and posttranslational regulation (1–7). Due to its limited genome size (10 kb), HIV-1 has to rely on cellular cofactors to progress through its life cycle. Host factors are not only needed for productive HIV-1 replication but also for latency establishment and reactivation, whereas restriction factors protect the cells against infection and provide innate immunity (8–14).

MicroRNAs and their role as viral restriction factors.

MicroRNA (miR) biogenesis begins in the nucleus with transcription to create the long primary miR (pri-miR), which includes a hairpin that contains the mature sequence. The hairpin is excised by the Microprocessor complex that includes Drosha, an RNase III enzyme, and its cofactor DGCR8, producing the 60- to 70-nucleotide precursor miRs (pre-miR). The pre-miR is exported out of the nucleus by exportin-5, whereas another RNase III enzyme, Dicer, processes the pre-miR into the 21- to 23-nucleotide duplex miR. The strand destined to be the mature sequence is then loaded onto Argonaute (Ago), which along with other proteins forms the miRNA-induced silencing complex (miRISC) (15, 16). Using imperfect base pairing, miRs guide miRISC to specific mRNAs to downregulate their expression by triggering mRNA destabilization or translational repression (17, 18). miR target recognition is often dependent on short 6-nucleotide seed sites, which perfectly complement the 5′ end of the miR (positions 2 to 7) (18). Both miRs and the mRNA binding sites are highly conserved (18, 19). miRs exemplify the emerging view that noncoding RNAs may equal proteins in regulatory importance. The majority of the human transcriptome is predicted to be under miR regulation, positioning this posttranscriptional control pathway within every major genetic cascade (20). miRs can regulate and coordinate development and modulate self-renewal, differentiation, and cell fate establishment (12–24). In addition, Dicer and Drosha knockout experiments have demonstrated that miR pathways repress HIV-1 replication and contribute to the maintenance of latency (25). Furthermore, profiling studies suggests that miRs in general are downregulated upon T-cell activation, suggesting increased target derepression during optimal HIV-1 infection (26–28). Also, miRs have been described to function in complex ways in the intersection between host and pathogen, for example, as an antiviral defense mechanism and/or as a facilitator of latency (29, 30). miRs have been proposed to negatively affect HIV-1 by directly targeting of the viral RNA genome and/or by repressing virus-dependent cellular cofactors; these miRs include miR-29, miR-150, miR-28, miR-125b, miR-223, and miR-382 (31). The complex interplay between miRs and their mRNA targets during HIV-1 infection, replication, and latency needs further investigation.

Work in colleagues’ laboratories and our own established that interferon-induced miRs repress viral replication of hepatitis C virus (HCV) (29, 32–34). As the life cycle of long interspaced element 1 (L1) mimics that of a virus, we next asked the question of whether miRs also regulate L1 activity. L1 retrotransposons and related short interspersed nuclear elements (SINEs) make up approximately 35% of the human genome, with some L1 elements still being active regardless of multiple cellular restriction mechanisms (35–38). We recently performed an anti-miR lentiviral library screen to identify miRs involved in the regulation of de novo L1 retrotransposition. miR-128 was identified as a negative regulator of L1, and further characterization led to the discovery of a novel principle by which miR-128 represses de novo retrotransposition of L1 elements in somatic cells, including cancer cells, cancer stem cells, and induced pluripotent stem cells (iPSCs), by directly binding L1 RNA and targeting it for degradation (39). This finding was surprising, as the most well-understood mechanism by which miRs function relies on the binding and regulation by one miR of multiple cellular mRNAs, whose protein products typically function in a specific pathway within the cell. We therefore explored if miR-128 might also regulate cellular proteins involved in L1 retrotransposition. We used bioinformatics analysis in combination with miR-128 RNA immunopurification (RIP) RNA smart sequencing approaches and identified the mRNA for a nuclear import factor (TNPO1) as a target of miR-128 binding. Our recent results (40) determined that TNPO1 is a functional target for miR-128, by directly targeting TNPO1 for degradation, which results in the L1 ribonucleoprotein (RNP) complex being trapped in the cytoplasm and L1 retrotransposition and genomic integration repressed. Thus, miR-128 represses L1 retrotransposition by the following two independent mechanisms: (i) repression of a key cofactor required by L1 (TNPO1), and (ii) direct binding of L1 RNA (39). The TNPO family of proteins contains, TNPO1, TNPO2, and TNPO3. Interestingly, miR-128 has predicted binding sites in all three TNPO proteins.

Here, we report that the type I interferon-induced microRNA miR-128 regulates TNPO3 mRNA and protein expression levels in cell lines (HeLa, Jurkat, and THP-1 cells) and in primary human CD4⁺ T-cells by directly targeting TNPO3 mRNA. Infection studies using wild-type and TNPO3-independent N74D mutant HIV-1 reporter viruses demonstrate that miR-128 significantly represses HIV-1 replication and viral spreading in cell lines and in CD4⁺ T-cells and that miR-128-induced HIV-1 repression is partly dependent on TNPO3 reduction (N74D HIV-1 showing derepression). The cleavage and polyadenylation specificity factor subunit 6 (CPSF6), one of the serine-rich (SR) proteins, is a conditional restriction factor for HIV-1 and can be transported to the nucleus by TNPO3, the nuclear importer of SR proteins. When TNPO3 is removed from cells, CPSF6 restricts HIV-1 infection (41–43). Finally, analysis of anti-miR-128 CD4⁺ T-cells challenged with HIV-1 in the presence of type I interferon demonstrates that neutralization of endogenously expressed miR-128 partly counteracts the interferon (IFN)-mediated block of HIV-1. Together, these studies support the conclusion that the interferon-inducible microRNA miR-128 functions as a novel antiviral mediator, in part by directly targeting TNPO3 mRNA, resulting in the inhibition of HIV-1 replication and viral spreading.

RESULTS

miR-128 regulates TNPO3 levels.

Three proteins encoded by members of the importin gene family, TNPO1, TNPO2, and TNPO3, contain predicted miR-128 binding sites (Fig. 1A). This is noteworthy, as TNPO3 is required for successful nuclear import of the HIV-1 preintegration complex (PIC) and for viral replication (44–50). As mentioned, CPSF6 is a conditional HIV-1 cofactor, aiding in nuclear steps of virus replication in cells retaining TNPO3 function (49). If TNPO3 is disrupted, CPSF6 accumulates in the cytoplasm and inhibits HIV-1 viral replication (44–50). Of note, we and others have determined that type I interferon induces miR-128 levels, suggesting a role for miR-128 in antiviral defense in human cells (Fig. 1B; see also references 29 and 51). Furthermore, miR-128 is expressed in HIV-1 target cells, including CD4⁺ T-cells and blood-derived monocytes (Fig. 1B; see also references 51, to ,55). Thus, if miR-128 represses expression of TNPO proteins, including that of TNPO3, we would predict that miR-128 represses HIV-1 replication.

FIG 1.

Identification of TNPO3 as a cellular target of miR-128. (A) Predicted miR-128 binding sites in the three different TNPO proteins (TNPO1, TNPO2, and TNPO3) are shown. (B) Primary human CD4⁺ T-cells and blood-derived monocytes were isolated from healthy donors by Ficoll and magnetic-activated cell sorting (MACS) separation versus adhesion separation. Cells were cultured in the presence or absence of type I interferon (IFN-α2, 100 U/ml). Interferon responsiveness of CD4⁺ T-cells was evaluated by tetherin induction (Western blot analysis, left panel). Cells were harvested and miR-specific qPCR was performed for miR-128 and normalized to snoU6. (C) miR-specific qPCR was performed on stable miR-modulated HeLa cells using specific miR-128 RT and qPCR primers and normalized to RNU5A. (D) miR-128, anti-miR-128, and miR control (control)-expressing HeLa cell lines were generated, and Western blot analysis of TNPO3 and α-tubulin amounts was performed. Quantifications of blots are shown (bottom panels). (E) Primary CD4⁺ T-cells were sorted from PBMCs from healthy donors. CD4⁺ T-cells were activated with anti-CD3/anti-CD28 for 2 days and were transduced with miR lentiviruses for 2 days. Level of transduction infection as presented by GFP⁺ transduced cells was determined on day 7. (F) Primary CD4⁺ T-cells were isolated from healthy blood donors by Ficoll and negative MACS separation, and CD4⁺ T-cells were transduced with miR-encoding lentivirus and Western blot analysis of TNPO3 and α-tubulin amounts was performed. Quantifications of blots are shown (bottom panels). (G) Relative levels of TNPO3 to beta-2 microglobulin (B2M) mRNA levels were evaluated by qPCR in HeLa cells, THP-1 cells, and primary CD4⁺ T-cells transduced with miR control, anti-miR-128, or miR-128. Results are shown as mean ± standard error of the mean (SEM) (n = 3 independent biological replicate). *, P < 0.05; **, P < 0.01; and ***, P < 0.001 by two-tailed Student’s t test.

To explore this possibility, we generated stably transduced miR-modulated cell lines (miR-128 overexpressing, anti-miR-128 in which endogenous miR-128 is neutralized, and control miR) of HeLa cells. We validated that miR-128 expression levels were modulated as expected and at physiological relevant levels similar to changes induced by cytokines and growth factors (29, 51) by miR-specific quantitative PCR (qPCR) (miR-128 was induced 4- to 5-fold in miR-128-overexpressing cells, while miR-128 was reduced by 40 to 50% in anti-miR-128-expressing HeLa cells) (Fig. 1C). Next, we evaluated TNPO3 protein levels by Western blot analysis in HeLa cell lines in which miR-128 was reduced or overexpressed. miR-128 significantly reduced TNPO3 protein levels relative to those of miR control HeLa cells (Fig. 1D). In contrast, TNPO3 protein amounts were significantly increased in HeLa cells lines in which endogenously miR-128 levels were reduced (anti-miR-128) (Fig. 1D). Next, we wished to evaluate the effect of miR-128 on TNPO3 protein expression in HIV-1 target cells, namely primary human CD4⁺ T-cells. We obtained peripheral blood mononuclear cells (PBMCs) from healthy donors and isolated CD4⁺ T-cells by Ficoll and negative magnetic-activated cell sorting (MACS) separation. Primary CD4⁺ T-cells were transduced as described for HeLa cells, and the transduction efficiency of primary CD4⁺ T-cells was determined to be 45 to 50% (Fig. 1E). miR-128-overexpressing CD4⁺ T-cells were characterized by a significant decrease in TNPO3 protein levels relative to those of miR-control CD4⁺ T-cells, while TNPO3 protein levels were significantly enhanced in anti-miR-128 CD4⁺ T-cells (Fig. 1F, right panel).

We next evaluated if the significant miR-128-induced regulation of TNPO3 protein levels was correlated with changes in TNPO3 mRNA levels. We generated miR-modulated HeLa, THP-1, and CD4⁺ T-cell (miR-128 overexpression, anti-miR-128, and control miR) lines and evaluated TNPO3 mRNA levels by reverse transcription-quantitative PCR (qRT-PCR). The level of TNPO3 mRNA was significantly reduced in miR-128-overexpressing HeLa cells, THP-1, and CD4⁺ T-cells relative to that in miR control cell lines (Fig. 1G). In contrast, neutralization of miR-128 levels using overexpression of anti-miR-128 resulted in a significant increase in TNPO3 mRNA levels relative to that in miR controls (Fig. 1G). To rule out the possibility that miR-128 overexpression using a lentiviral delivery strategy itself caused reduction in TNPO3 mRNA levels, we also examined the levels of TNPO3 mRNA in HeLa, THP-1, and Jurkat cells after transiently transfecting with miR-128, anti-miR-128, or control miR oligonucleotides (miR mimics and anti-miRs). Forty-eight hours after transfection, significantly enhanced expression of TNPO3 mRNA was observed in HeLa cells transfected with anti-miR-128 oligonucleotide, and significant reduction of TNPO3 mRNA was observed in HeLa cells transfected with the synthetic miR-128 mimic oligonucleotides (data not shown). A similar tendency was observed in Jurkat and THP-1 cells transiently transfected with miR mimics (data not shown). These combined results demonstrate that miR-128 regulates TNPO3 expression levels in multiple cell types, including in the primary HIV-1 target cell type, CD4⁺ T-cells.

miR-128 delays viral infection and replication of HIV-1.

Next, we wished to determine if miR-128-induced repression of TNPO3 causes significant accumulative antiviral effects on HIV-1 viral replication and viral infection. We performed viral infection of miR-modulated Jurkat cells using replication-competent WT HIV-1 virus (HIV-1_NL4-3). We determined that miR-128 significantly reduced viral spreading of wild-type HIV-1 by evaluating the percentage of Gag-positive (Gag⁺) cells by flow cytometry analysis. Level of infection is presented as percent Gag⁺ shown among green fluorescent protein (GFP)-positive transduced cells on days 3, 6, 8, 10, 13, and 15, as previously described (56, 57). In parallel, supernatant was collected in a kinetic manner for p24 assays (Fig. 2A and B).

FIG 2.

miR-128 delays viral spreading of wild-type HIV-1. (A and B) miR-modulated Jurkat cells (miR-128 or control) were generated by stable transduction and spinfected with HIV-1 virus (WT_NL4-3), and spreading was determined by evaluating the % of Gag⁺ cells by flow cytometry analysis at each time point using a fluorescent KC57 antibody. In parallel, supernatant for p24 assays were collected and p24 was assessed using an ELISA. (C) Primary CD4⁺ T-cells were isolated and CD4⁺ T-cells were transduced with miR-encoding lentivirus, supernatant was collected, and p24 assays analysis was performed. The experiments shown are each one of three independent biological experiments (n = 3). p24 standards and samples were run in duplicates therefore no error bars are shown. The R² value for p24 standards is >0.99.

Next, we evaluated the antiviral effect of miR-128 on primary CD4⁺ T-cells. CD4⁺ T cells were isolated from healthy donor PBMCs and sorted by flow cytometry; cells were then activated with anti-CD3/anti-CD28 for 2 days and transduced with miR lentivirus. After 2 days, cells were infected with HIV-1_NL4-3 (100 ng/10⁶ cells) for 8 days. Supernatants were collected on days 1, 2, 3, 6, and 8 for p24 assays. Despite the inefficiency of transduction in the primary cells, CD4⁺ T-cells expressing miR-128 decreased HIV-1 replication over the first week of infection, comparable to what was observed in the Jurkat cell lines. Unfortunately, due to the limited cell viability of the activated primary cells, the time course of infection was shorter than that in the experiment using the Jurkat cells. In summary, these results support the idea that miR-128 delays spreading of viral infection in T cells.

miR-128 directly interacts with TNPO3 mRNA.

We established that overexpression of miR-128 significantly depletes cells of TNPO3 (Fig. 1) and thus, not surprisingly, elicits a significant accumulative antiviral effect on HIV-1 replication and spreading (Fig. 2). Then, we wished to characterize the mechanism by which miR-128 regulates TNPO3 levels. We performed bioinformatics analysis (using TargetScan and miRBase [58, 59]) and identified two predicted miR-128 binding sites in TNPO3 mRNA, one 7-mer seed site in the coding region sequence (CRS) of TNPO3 mRNA (site 1), and a second 7-mer seed site in the 3′ untranslated region (UTR) of TNPO3 mRNA (site 2) (Fig. 3A, top). TNPO3 CRS and 3′ UTR sequences harboring the two potential miR-128 binding sites were cloned into a dual-luciferase (Luc) reporter construct (pEZX-MT05). In addition, a perfect 23-nucleotide miR-128 binding sequence (positive-control) luciferase construct was generated. HeLa cells were transfected with one of the three miR-128 binding site-encoding plasmids (site 1 CDS TNPO3, site 2 3′ UTR TNPO3, or miR-128 perfect binding site) in addition to mature miR-128 or miR control oligonucleotide mimics (Fig. 3A). To determine if the specific miR-128 seed sequences are responsible for the interaction with miR-128, we also measured luciferase activity in HeLa cells transfected with mature miR-128 (or control miR mimics) along with either CRS or 3′-UTR TNPO3 binding sites in which mutations had been introduced (Fig. 3A). While luciferase activity was significantly reduced in cells transfected with miR-128 and encoding the WT binding sites (sites and 2) of TNPO3 as well as with the positive control (Fig. 3B), the decrease in luciferase activity in HeLa cells transfected with mutant TNPO3 and miR-128 was not significant, although it was not completely restored to control levels (Fig. 3C). These results indicate that miR-128 targets TNPO3 mRNA and that both predicted seed sites are responsible for their interaction (Fig. 3C).

FIG 3.

miR-128 represses TNPO3 by binding directly to two sites in TNPO3 mRNA. (A) Schematic of the two predicted miR-128 binding sites in TNPO3 mRNA, which includes a 7-mer seed site (site 1) in the coding reading sequence (CRS) and a 7-mer seed site (site 2) in the 3′ UTR of TNPO3 mRNA (top panel). The predicted base pairing of miR-128 to the seed sequence of the two wild-type (WT) binding sites in TNPO3 mRNA, as well as a representation of mutations that we generated in the seed sequence (mutant), are shown (bottom panels). (B) Relative luciferase activity in HeLa cells transfected with plasmids expressing a Gaussia luciferase gene fused to the one of the two predicted wild-type (WT) binding sites in TNPO3 or a positive-control sequence corresponding to the 22-nucleotide perfect match of miR-128 and cotransfected with control or mature miR-128 mimics were determined 48 h posttransfection (n = 3). (C) Relative luciferase activity in HeLa cells transfected with plasmids expressing the luciferase gene fused to the WT or either mutated binding site (mutant) and cotransfected with control or mature miR-128 mimics were determined 48 h posttransfection (n = 3). (D) Argonaute-RNA immunopurification in HeLa cell lines stably transduced with miR-128 overexpression or miR-128 neutralization (anti-miR-128) was performed. Relative amounts of TNPO3 mRNA normalized to B2M were determined for input samples (left panel, “Input,” n = 3). Relative fraction of TNPO3 transcript amounts associated with immunopurified Ago complexes is shown for immunopurified (IP) samples, TNPO3 fractions normalized to the amount of TNPO3 in input are shown as “corrected” (top right panel, “IP,” n = 3). (E) Relative amounts of GAPDH in the same input and IP samples were determined as a negative control (n = 3). n = 3 independent biological replicates, mean ± SEM. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 by two-tailed Student’s t test.

Furthermore, Argonaute (Ago) complexes containing miRs and target mRNAs were isolated by immunopurification and assessed for relative complex occupancy by the TNPO3 mRNA to determine if miR-128 directly binds TNPO3 mRNA in HeLa cells (cartoon in Fig. 3D), as previously described (39, 60). The relative level of TNPO3 mRNA was significantly lower in cells stably overexpressing miR-128 compared to that of those expressing anti-miR-128 constructs, as expected (Fig. 3D, “Input,” left panel). When correcting for the lower expression level of TNPO3 mRNA (because of lower miR-128 expression levels), which may underestimate the scale of the effect, the relative fraction of immunopurified-Ago-bound TNPO3 mRNA significantly increased when miR-128 was overexpressed, compared to that in the anti-miR-128 control (Fig. 3D, “IP,” right panel). In contrast, miR-128 did not repress GAPDH mRNA expression levels or immunopurified GAPHD mRNA, as expected (Fig. 3E). This result suggests that high levels of miR-128 lead to higher levels of TNPO3 mRNA being bound in Ago complex and that TNPO3 expression is regulated directly by miR-128. These data imply that miR-128 represses TNPO3 expression via a direct interaction with the target sites located in the CRS and 3′ UTR of the TNPO3 mRNA.

miR-128 represses HIV-1 replication.

In order to study if type I interferon-induced miR-128 functions as a novel antiviral mediator during HIV-1 infection in human cells and is specific to restricting HIV-1 replication, we next employed VSV-G pseudotyped HIV-1 vectors that contained either luciferase (NLdELuc) or red fluorescent protein (RFP) gene (HIV-1-RFP) as a reporter gene. HIV-1 env has been deleted in these vectors and the nef gene has been replaced with a reporter gene (49). miR-modulated Jurkat cells expressing miR control or miR-128 were infected with either HIV-1-RFP/VSV-G or murine leukemia virus (MLV)-RFP/VSV-G. Then their infection was examined by fluorescence-activated cell sorter (FACS) analysis for RFP expression after 48 h. These experiments showed that miR-128 significantly inhibits HIV-1-RFP/VSV-G but did not substantially block MLV-RFP/VSV-G in Jurkat cells (Fig. 4A and B). Of note, although the relative miR effect was smaller when using the single-cycle reporter virus than with replication-competent virus (see Fig. 2), we used the single-cycle reporter virus to investigate the mechanism of miR-128-induced HIV-1 inhibition.

FIG 4.

miR-128 inhibits HIV-1 replication of a single-cycle HIV-1 reporter virus. (A and B) Stable miR-modulated Jurkat cells (miR control or miR-128 overexpressing) were spinfected with VSV-G pseudotyped HIV-1 or MLV vectors that contains red fluorescent protein (RFP) gene (HIV-1-RFP or MLV-RFP/VSV-G). The fraction of infected cells was measured by flow cytometry for RFP expression after 48 h, and equal infectivity between samples was verified by quantification of RFP expression. (C to E) miR-modulated HeLa, THP-1, or Jurkat cells (miR-128, anti-miR-128, or miR control) were generated by stable transduction. Alternatively, HeLa cells (F) were transiently manipulated with miR mimics (miR-128, anti-miR-128, or miR control oligonucleotide). (G) Finally, primary human CD4⁺ T-cells were isolated and then activated for 2 days with CD3/CD28 and 30 U/ml interleukin-2 (IL-2). Both stable and transient miR-modulated cells were then spinfected with the HIV-1 reporter virus, and luciferase was measured after 48 h. n = 3 independent biological replicates, mean ± SEM. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 by two-tailed Student’s t test.

Next, we wished to further examine the specificity of the miR-128 effect by comparing miR-128 overexpression versus miR-128 depletion (by anti-miR-128) to miR control-expressing cells. We evaluated the effect of miR-128 on HeLa, Jurkat, and THP-1 cells. Cells were infected with HIV-1 reporter virus (NLdELuc pseudotyped with VSV-G) by spinoculation, and infection was monitored by measuring luciferase activity 48 h after infection. We found that miR-128 significantly inhibited HIV-1 infection, whereas anti-miR-128 significantly enhanced HIV-1 infection relative to that with miR controls in HeLa, Jurkat and THP-1 cells (Fig. 4C to E). Next, primary human CD4⁺ T-cells were tested. The isolated CD4⁺ T-cells were stimulated with 30 IU/ml interleukin-2 (IL-2), activated with CD3/CD28 for 2 days, and transduced with miRs (miR-128, anti-miR-128, and control miR oligonucleotides). miR-modulated CD4⁺ T-cells were then infected with NLdELuc by spinoculation, and luciferase activity was measured after 48 h. miR-128 significantly inhibited HIV-1 infection in primary CD4⁺ T-cells relative to that with miR control cells (Fig. 4F). anti-miR-128 did not affect replication of the HIV-1 infection significantly. Finally, to rule out the possibility that the miR effect was an artifact of lentiviral genomic integration, we tested transient miR-modulated HeLa cells using miR/anti-miR mimic oligonucleotides. We observed that transient miR-128 overexpression significantly inhibited HIV-1 infection and miR-128 neutralization enhanced HIV-1 replication relative to that with miR controls (Fig. 4G). This combined body of data supports the notion that miR-128 specifically and significantly inhibits infection of single-cycle HIV-1 infection in cell lines (HeLa, Jurkat, and THP-1 cells) and in HIV-1 target cells (CD4⁺ T-cells).

miR-128-induced HIV-1 inhibition is partly dependent on TNPO3.

Next, we wished to explore whether miR-128 induced-TNPO3 repression is required for HIV-1 repression. We took advantage of the N74D CA mutant single-cycle HIV-1 reporter virus pseudotyped with VSV-G (N74D NLdELuc). The single mutation at N74 in HIV-1 capsid has been shown to affect the sensitivity of HIV-1 infection to depletion of various nuclear import factors, including TNPO3 (50, 61, 62).

In order to dissect the role of miR-128-induced TNPO3 repression, we generated miR-modulated and TNPO3-modulated (short hairpin-TNPO3 [shTNPO3], TNPO3-overexpressing, or plasmid control) HeLa cell lines. All cell lines were validated for TNPO3 and miR-128 expression levels (Fig. 5A, top middle; Fig. 1A; and Fig. S3). miR-128 and TNPO3-modulated cells were spinoculated with WT or N74D HIV-1 reporter virus. As shown before (Fig. 4C), miR-128 significantly reduced WT HIV-1 replication and anti-miR-128 enhanced WT HIV-1 replication relative to that in miR control HeLa cells (Fig. 5A, left panel). TNPO3 stable knockdown HeLa cells by shRNA transduction (shTNPO3) showed significant HIV-1 inhibition, and overexpression of TNPO3 expression increased WT HIV-1 infection relative to control cells, as expected (Fig. 5A, middle panel). N74D HIV-1 also showed the effect of miR-128-induced inhibition of HIV-1 replication in HeLa cells. However, the potency of inhibitory effect of N74D HIV-1 by miR-128 was less significant compared to that of WT HIV-1 (Fig. 5A, right panel). As previously shown by other groups, we observed that neither TNPO3 knockdown nor overexpression significantly affected N74D HIV-1 infection, verifying that the N74D CA mutant HIV-1 is not dependent on TNPO3 for HIV-1 replication (Fig. S5).

FIG 5.

miR-128-induced HIV-1 repression is partly dependent on TNPO3. (A) miR-modulated HeLa cells (miR-128, anti-miR-128, or control miR) or TNPO3-modulated HeLa cells (shTNPO3, TNPO3-overexpressing, or plasmid controls) were generated by stable transduction, then spinfected with the wild type (WT) or the N74D variant (TNPO3 independent) of the VSV-G- pseudotyped RFP HIV-1 reporter virus; luciferase was measured after 48 h. Equal infection was determined by counting RFP-positive cells. Modulation of TNPO3 protein expression in HeLa cells was verified by Western blot analysis of TNPO3 and α-tubulin (top, middle right). One representative example is shown out of three. (B) miR-modulated THP-1 and Jurkat cells (miR-128, anti-miR-128, or control miR) were generated by stable transduction and then spinfected with either the WT luciferase encoding VSV-G- pseudotyped RFP HIV-1 reporter virus or the N74D mutant HIV-1 reporter virus. Luciferase was measured after 48 h. n = 3 independent biological replicates, mean ± SEM. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 by two-tailed Student’s t test. (C) Model of TNPO3/CPSF6’s effect on HIV-1 replication. CPSF6 functions as a conditional inhibiting factor, acting with low levels of TNPO3. If TNPO3 is disrupted (for example, by miR-128), CPSF6 accumulates in the cytoplasm and impairs viral nuclear entry and viral replication of HIV-1 (as shown in the bottom panel of the cartoon).

Next, we evaluated the antiviral effect of miR-128 on WT and N74D HIV-1 reporter viruses using transformed CD4⁺ cells of monocytic and lymphocytic lineage (THP-1 or Jurkat cells, respectively). We observed that miR-128 induced significant inhibition of WT HIV-1 infection in THP-1 (Fig. 5B) and Jurkat (Fig. 5B) cells, as previously established (Fig. 4D and E). When challenging miR-modulated Jurkat and THP-1 cells with the N74D HIV-1 reporter virus, we found that miR-128-induced inhibition of N74D HIV-1 infection was significantly derepressed compared to repression of WT viral replication. However, miR-128 still significantly reduced N74D HIV-1 reporter activity compared to that in miR control cell lines (Fig. 5B). These studies support the idea that miR-128-induced inhibition of HIV-1 viral replication is partly dependent on reduction of TNPO3 expression in target cells. Based on our findings, we propose a model in which miR-128 modulates the ratio of cytoplasmic TNPO3/CPSF6, which lead to impaired HIV-1 nuclear entry and virus replication (Fig. 5C). Future studies will be needed to evaluate if miR-128 directly affect CPSF6 localization.

Anti-miR-128 counteracts IFN-induced blockage of HIV-1 replication.

Finally, we wished to address the relative importance of miR-128 in the complex antiviral effects of type I interferon. We obtained PBMCs from healthy donors and isolated CD4⁺ T-cells as previously described. CD4⁺ T-cells were transduced with control or anti-miR-128 and then infected with wild-type HIV-1 NL4-3 virus. Cells were cultured in the presence or absence of interferon alpha 2 (IFN-α2) prior to infection at 0.7 μg/ml and were continually treated throughout the experiment. Culture supernatant was collected, and p24 was measured using an enzyme-linked immunosorbent assay (ELISA). As expected, IFN-α2 dramatically diminished HIV-1 replication, as many factors are likely stimulated (some that will be strongly inhibitory) (Fig. 6A). Anti-miR-128 was used to neutralize endogenously expressed miR-128 levels in CD4⁺ cells. The observed degree to which anti-miR-128 counteracts IFN restriction is modest (Fig. 6B; please note the different y axis magnification from that in panel A, zooming in on the miR-128 effect) in experimental settings in which miR-128 levels were reduced by 40 to 50%.

FIG 6.

anti-miR-128 counteracts IFN-induced viral suppression. (A and B) CD4⁺ T-cells were isolated from healthy donors. Some of the CD4⁺ T-cells were transduced with control or anti-miR-128 and then all cells were infected with 100 ng of wild-type HIV NL4-3 virus. The infected CD4⁺ T-cells were cultured in the presence or absence of IFN-α2 prior to infection at 0.7 μg/ml and then continually treated throughout the experiment. Medium was collected and p24 was measured using an ELISA (please note the different y axis magnification from that in panel A, zooming in on the miR-128 effect). The experiment shown is from one donor and is one of three independent biological experiments analyzing cells from three different donors (n = 3). p24 standards and samples were run in duplicates; therefore, no error bars are shown. The R² value for p24 standards is >0.99. (C) Cartoon of proposed miR-128 antiviral effect. miR-128 is induced by type I interferon and directly targets TNPO3 mRNA. Loss of TNPO3 results in accumulation of CPSF6 in the cytoplasm and inhibition of viral nuclear import and HIV-1 replication.

In summary, the body of work described here supports a model in which miR-128 is an interferon-inducible gene and significantly inhibits HIV-1 replication and viral spreading, in part by targeting TNPO3 for degradation (see Fig. 6C).

DISCUSSION

Our present data provide the first evidence that the microRNA miR-128 functions as an antiviral mediator, significantly restricting HIV-1 replication and viral spreading. The finding that miR-128 can be induced by type I interferon in HIV-1 target cells (CD4⁺ T-cells and blood-derived monocytes) and the finding that anti-miR-128 plays a part in counteracting the IFN-induced block of HIV-1 spreading support the conclusion that miR-128 functions as a novel host restriction factor.

Mechanistically, we demonstrate that miR-128 reduces TNPO3 mRNA and protein levels in cell lines (HeLa, Jurkat, and THP-1 cells) and in primary CD4⁺ T-cells, by directly targeting two sites in the TNPO3 mRNA, one in the 3′ UTR and one in the coding region sequence (CRS). Analysis of miR binding sites located in the 3′ UTR of target mRNAs has been the focus of the majority of miR-based studies. However, in-depth analysis by Fang and Rajewsky (63) and our own recent characterization of miR targets (39, 60) have shown that miR target sites in the CRS are of functional importance, are under negative selection, and that target sites in the CRS can enhance regulation mediated by sites in the 3′ UTR. Based on these findings, it is likely that the two miR-128 binding sites in TNPO3 mRNA cooperate for optimal repression of TNPO3 expression.

The finding that TNPO3 is a direct target of miR-128 suggests that miR-128 influences viral replication by interfering with TNPO3/CPSF6-dependent nuclear import of HIV-1 and viral replication. However, infection studies using wild-type and N74D CA mutant HIV-1 reporter viruses suggest that there are other miR-128 targets that can also regulate HIV-1 infection. Our work does suggest that miR-128 functions in a TNPO3-dependent and TNPO3-independent context during HIV-1 infection.

In addition, recent work by Kane et al. (64) demonstrates that HIV-1 exploits numerous nuclear import pathways to target host DNA. The authors establish that heterogeneity is dependent on cyclophilin A (CypA), specific CA sequences, and the representation and stoichiometry of nucleoporins and associated proteins, which can be variable. For example, sensitivity to TNPO3 and/or CPSF6 depletion is shown to be dependent not only on CA sequences but also on cell type and the stage of the cell cycle. For these reasons, further mechanistic studies will help elucidate whether miR-128 directly blocks nuclear import of HIV-1 PIC, where in the viral life cycle the antiviral effect of miR-128 can be observed, and in which cellular contexts this may apply (64).

As mentioned above, we have previously established a novel role for miR-128 in the inhibition of long interspaced element 1 (L1) retrotransposition (39). Interestingly, miR-128-induced L1 restriction takes place by both direct interactions with L1 RNA and indirectly via miR-128-induced depletion of cell host factors, on which L1 is dependent for successful replication (39, 40). From an evolutionary standpoint, miRs have likely evolved to protect us against unwanted L1 mobilization, including cancer-initiating mutagenesis, which could lead to detrimental genomic instability. The network of cellular host factors and pathways which mobile DNA elements are dependent on also restricts the life cycle of retroviruses, and as such, it seems likely that miRs (including miR-128) play a role in the potent arsenal of antiviral defense factors in human cells, also acting against HIV-1 infection. As miRs are known to often function by regulating multiple gene products in the same cellular pathway (65–68), we predict that miR-128 regulates multiple cellular cofactors, on some of which HIV-1 is dependent, and, as such, miR-128 may function as a master restriction factor of RNAs stemming from exogenous pathogens, including HIV-1, in addition to endogenous selfish elements such as L1.

The finding that miR-128 restricts L1 replication by a dual mechanism, namely by regulating cellular cofactors (including TNPO1) and by targeting L1 RNA, raises the question of whether miR-128 may also directly target the HIV-1 genome. However, unlike L1, which is not capable of responding to selective pressure, HIV-1 would be predicted to evolve by mutating the miR-128 binding site(s) and escape miR-128-induced inhibition if miR-128 is potently restricting viral replication.

In summary, our results support a model in which miR-128 is expressed in primary HIV-1 target cells, is a type I IFN response gene, and functions as a novel antiviral defense mechanism during HIV-1 infection, partly by repressing the nuclear import factors TNPO3 and by inhibiting HIV-1 replication and viral spreading (see Fig. 6C). A long-term goal of our mechanistic analysis, which includes network studies, is to add to understanding of the interactome of viral infection, replication, latency, and reactivation, which will enable us to propose novel therapeutic strategies to target specific cofactors that could prevent successful HIV-1 infection, block viral replication, or attenuate establishment/reactivation of latency, all components of a functional cure.

MATERIALS AND METHODS

Cell culture and primary HIV-1 target cell isolation.

All cells were cultured at 37°C and 5% CO₂. HeLa cells (CCL-2), 293T cells (CRL-3216), THP-1 cells (TIB-202), and Jurkat cells (TIB-152) were all obtained from ATCC. Adherent cells were cultured in Eagle’s minimum essential medium (EMEM, catalog no. SH3024401; HyClone) supplemented with 10% heat-inactivated fetal bovine serum (HI-FBS; FB-02; Omega Scientific), 5% GlutaMax (catalog no. 35050-061; Thermo Fisher), and 3% HEPES (15630-080; Thermo Fisher). Jurkat cells were cultured in RPMI 1640 medium supplemented with 10% HI-FBS (FB-02; Omega Scientific).

Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood obtained from anonymous blood donors (New York Blood Center) using a standard Ficoll (Cellgro) procedure. Primary CD4 T-cells isolated using the human CD4 T-cell enrichment kit per the manufacturer’s instruction (Stemcell) were activated using Dynabeads Human T-Activator CD3/CD28 (Life Technologies) and cultured in suspension cell medium supplemented with 30 units per ml of interleukin-2 (IL-2) (PeproTech).

Additionally, primary human blood-derived monocytes were isolated from peripheral blood mononuclear cells after Ficoll density gradient centrifugation. PBMCs were allowed to attach to a 10-cm dish for 2 h and then vigorously washed with phosphate-buffered saline (PBS). Attached cells were subjected to differentiation in RPMI medium supplemented with 10% FBS, penicillin/streptomycin (P/S), 2 mM 2-glutamine, and 100 ng per ml of granulocyte macrophage colony-stimulating factor (GM-CSF) (PeproTech) for 4 days. Some primary CD4 T-cell and primary blood-derived monocyte cultures were stimulated with interferon alpha (100 U/ml) (PeproTech) for 72 to 96 h.

Viral infections.

All HIV-1 reporter vectors encoding fluorescent proteins or firefly luciferase were NL4-3 derived. Reporter particles incorporating VSV-G and replication-competent HIV-1 were derived through transient transfection of HEK293T cells. For some experiments, we took advantage of the N74D CA mutant single-cycle HIV-1 reporter virus pseudotyped with VSV-G (N74D NLdELuc). For some cultures, an MLV-RFP/VSV-G reporter was used to evaluate infection specificity compared to that with HIV-RFP/VSV-G. The percentage of infected cells was measured by FACS analysis of RFP expression.

Additionally, stable miR-modulated Jurkat cells were spinfected with wild-type HIV-1_(NL4-3) virus, and viral replication was determined by p24 ELISAs after 3, 6, 8, 10, 13, and 15 days of infection. Equal infection was confirmed by normalizing to reverse transcriptase (RT) units, or virus titer was determined on GHOST cells and equal infection units were used.

Finally, activated primary CD4⁺ T-cells were infected with HIV-1 NL4-3 (100 ng/10⁶ cells). Supernatant were collected for p24 assays. Infected cells were also harvested and stained for intracellular anti-Gag (clone KC57). Level of infection as presented by % Gag⁺ is shown among GFP⁺-transduced cells.

Transfection and transduction.

Opti-MEM (31985070; Lifetech) and Lipofectamine RNAiMAX (13778075; Lifetech) were used to complex and transfect 20 μM miR-128, anti-miR-128, or control mimics (C-301072-01 and IH-301072-02; Dharmacon) into cells. Opti-MEM and Xtreme Gene HP (06366236001; Roche Lifescience) were used to transfect a pJM101 neomycin L1 reporter plasmid into HeLa cells. Cells were transduced with high-titer virus using Polybrene (sc-134220; Santa Cruz Biotech) and spinoculation (800 × g at 32°C for 30 min). Transduced cells were then selected and maintained using 3 μg/ml puromycin.

shRNA against TNPO3.

shRNA for TNPO3 was designed using the RNAi Consortium (https://www.broadinstitute.org/rnai/public/) using clone TRCN0000235098 and cloned into pLKO.1 puro backbone (catalog no. 8453; Addgene). pLKO shGFP control plasmid was preassembled (catalog no. 30323; Addgene).

Lentiviral packaging.

VSV-G-pseudotyped lentiviral vectors were made by transfecting 0.67 μg of pMD2-G (catalog no. 12259; Addgene), 1.297 μg of pCMV-DR8.74 (catalog no. 8455; Addgene), and 2 μg of mZIP-miR-128, mZIP-anti-miR-128, pLKO-shControl, or pLKO-shHNRNPA1 (transfer plasmid)) into 293T cells using Lipofectamine LTX with Plus reagent (catalog no. 15338030; Thermo Fisher). Virus-containing supernatant was collected 48 h and 96 h posttransfection. Viral supernatants (SUPs) were concentrated using PEG-it virus precipitation solution (LV810A-1) according to the manufacturer’s instructions.

RNA extraction and quantification.

RNA was extracted using Trizol (catalog no. 15596-018; Thermo Fisher) and a Direct-zol RNA isolation kit (R2070; Zymo Research). cDNA was made with a high-capacity cDNA reverse transcription kit (catalog no. 4368813; Thermo Fisher). The amount of TNPO3 mRNA was analyzed by qRT-PCR (sense primer, 5′-AAGCAATTTTGGAGGTGGTG-3′; antisense primer, 5′-atagccaccttggtttcgtg-3′) using Forget-Me-Not qPCR mastermix (Biotium) relative to beta-2-microglobulin (B2M; sense primer, 5′-ATGTCTCGCTCCGTGGCCTTAGCT-3′; antisense primer, 5′-TGGTTCACACGGCAGGCATACTCAT-3′) housekeeping gene and processed using the threshold cycle (ΔΔC_T) method.

Western blotting.

Rabbit anti-TNPO3 antibody (ab109386; Abcam), rabbit anti-CPSF6 antibody (ab99347; Abcam), and mouse anti-tetherin antibody (ab88523; Abcam) were used at 1:2,000. Anti-alpha-tubulin antibody (ab4074; Abcam) was diluted 1:5,000 and used as a loading control; validation of antibodies can be found on the manufacturer websites. Secondary horseradish peroxidase (HRP)-conjugated anti-rat (ab102172; Abcam) or HRP-conjugated anti-rabbit (catalog no. NA934; GE Life Sciences) were used at 1:5,000. Enhanced chemiluminescence (ECL) substrate (catalog no. 32106; Thermo Fisher) was added and visualized on a Bio-Rad ChemiDoc imager.

Argonaute-RNA immunopurification.

Immunopurification of Argonaute (Ago) from HeLa cell extracts was performed using the 4F9 antibody (catalog no. sc-53521; Santa Cruz Biotechnology) as described previously (69, 70).

Briefly, 10-mm plates of 80% confluent cultured cells were washed with buffer A (20 mM Tris-HCl [pH 8.0], 140 mM KCl, and 5 mM EDTA) and lysed in 200 μl of 2X buffer (2XB; 40 mM Tris-HCl [pH 8.0], 280 mM KCl, 10 mM EDTA, 1% NP-40, 0.2% deoxycholate, 2× Halt protease inhibitor cocktail [Pierce], 200 U/ml RNaseout [Life Technologies], and 1 mM dithiothreitol [DTT]). Protein concentration was adjusted across samples with buffer B (20 mM Tris-HCl [pH 8.0], 140 mM KCl, 5 mM EDTA [pH 8.0], 0.5% NP-40, 0.1% deoxycholate, 100 U/ml RNaseOUT [Life Technologies], 1 mM DTT, and 1× Halt protease inhibitor cocktail [Pierce]). Lysates were centrifuged at 16,000 × g for 15 min at 4°C, and supernatants were incubated with 10 to 20 μg of 4F9 antibody conjugated to epoxy magnetic beads (M-270 Dynabeads; Life Technologies) for 2 h at 4°C with gentle rotation (Nutator). The beads, following magnetic separation, were washed 3 times for 5 min with 2 ml of buffer C (20 mM Tris-HCl [pH 8.0], 140 mM KCl, 5 mM EDTA [pH 8.0], 40 U/ml RNaseOUT [Life Technologies], 1 mM DTT, and 1× Halt protease inhibitor cocktail [Pierce]). Following immunopurification, RNA was extracted using miRNeasy kits (Qiagen), following the manufacturer’s recommendations, and qPCR was performed using hnRNPA1 primers designed around the binding site of miR-128 (sense primer, 5′-TCTCCTAAAGAGCCCGAACA-3′; antisense primer, 5′-TTGCATTCATAGCTGCATCC-3′) or GAPDH (sense primer, 5′-GGTGGTCTCCTCTGACTTCAA-3′; antisense primer, 5′-GTTGCTGTAGCCAAATTCGTT-3′) normalized to B2m (sense primer, 5′-ATGTCTCGCTCCGTGGCCTTAGCT-3′; antisense primer 5′-TGGTTCACACGGCAGGCATACTCAT-3′). Results were normalized to their inputs.

Luciferase binding assay.

Wild-type TNPO3 sense primer (5′-AATTCTTGGGTTTGTCACATATGCCACTGTGGAGGAGGTGGATGCAGCTA-3′) and antisense primer (5′-CTAGTAGCTGCATCCACCTCCTCCACAGTGGCATATGTGACAAACCCAA-3′), mutated TNPO3 sense primer (5′-AATTCTTGGGTTTGTCACATATGCCCTTATGGAGGAGGTGGATGCAGCTA-3′) and antisense primer (5′-CTAGTAGCTGCATCCACCTCCTCCATAAGGGCATATGTGACAAACCCAA-3′) or positive-control sense primer (5′-AATTCAAAGAGACCGGTTCACTGTGAA-3′) and antisense primer (5′-CTAGTTCACAGTGAACCGGTCTCTTTG-3′) sequences were cloned into dual-luciferase reporter plasmid (pEZX-MT05; Genecopoeia). HeLa cells (3 × 10⁵) were forward transfected with 0.8 μg of reporter plasmid (WT, mutated, or positive [Pos]) and 20 nM miR-128 mimic (Dharmacon) or control mimic (Dharmacon) using Attractene transfection reagent (catalog no.301005; Qiagen) according to the manufacturer’s instructions. Relative Gaussia luciferase and secreted embryonic alkaline phosphatase (SEAP) was determined using a Secrete-Pair dual luminescence assay kit (SPDA-D010; Genecopoeia). Luminescence was detected by a Tecan Infinite F200 Pro microplate reader.

Statistical analysis.

Student’s t tests were used to calculate two-tailed P values, and data are displayed as mean ± standard error of the mean (SEM) of technical replicates or independent biological replicates (independent biological experiments), with n as indicated.