Protein Kinase D3 Is Essential for Prostratin-Activated Transcription of Integrated HIV-1 Provirus Promoter via NF-κB Signaling Pathway

Prostratin has been proposed as a promising reagent for eradicating the latent HIV-1 provirus by inducing HIV-1 transcription activation. The molecular mechanism of this activation, however, is far from clear. Here, we show that the protein kinase D3 (PKD3) is essential for prostratin-induced transcription activation of latent HIV-1 provirus. First, silencing PKD3, but not the other members of PKD family, blocked prostratin-induced transcription of HIV-1. Second, overexpressing the constitutively active form of PKD3, but not the wild-type or kinase-dead form of PKD3, augmented the expression of HIV-1. Consistent with this observation, we found that prostratin could trigger PKD3 activation by inducing the phosphorylation of its activation loop. In addition, we identified PKCε of the novel PKC subfamily as the upstream kinase for this phosphorylation. Finally, the activation effect of PKD3 on HIV-1 transcription was shown to depend on the presence of κB element and the prostratin-induced activation of NF-κB, as indicated by the fact that silencing PKD3 blocked prostratin-induced NF-κB activation and NF-κB-dependent HIV-1 transcription. Therefore, for the first time, PKD3 is implicated in the transcription activation of latent HIV-1 provirus, and our results revealed a molecular mechanism of prostratin-induced HIV-1 transcription via PKCε/PKD3/NF-κB signaling pathway.


Introduction
The standard treatment for HIV-1 infected individuals currently is the administration of a combination of antiviral agents that are collectively termed highly active antiretroviral therapy (HAART) [1]. HAART is able to reduce plasma viremia to undetectable level and thereby slow down disease progression. HAART, however, fails to eradicate the latent HIV-1 in the quiescent T-cell reservoirs, which contains integrated but transcriptionally dormant provirus [1,2]. One strategy for eliminating the latent pool of HIV-1 is to awaken the transcription of provirus in these cells in the presence of HAART, thereby purging the cells by viral toxicity and cellular immune system [3][4][5].
Prostratin, a nontumorigenic phorbol ester, has been shown to be capable of promoting the transcription activation of latent HIV-1 [6,7]. Although prostratin and its recently developed analogs have been proposed as promising candidates for eradicating the latent HIV-1 provirus [8,9], the molecular mechanism of inducing the transcription activation of HIV-1 provirus is still far from clear. A study in J-Lat T cells with integrated HIV-1 reporter gene identified that the novel PKC (nPKC) subfamily, but not the classic PKC (cPKC) or atypical PKC (aPKC) subfamily, is critical for prostratin-induced transcription activation of latent HIV [10]. It was found that prostratin enables the plasma membrane translocation of PKC , and of cPKC, and of nPKC, and of aPKC. By analyzing the differential effects of PKC inhibitors, it is suggested that nPKCs, but not cPKCs or aPKCs, are involved in the transcription activation of latent HIV-1 provirus via activating NF-B signaling pathway [10].
Protein kinase D (PKD) is a family of serine/threonine protein kinases that has been implicated as the downstream effectors of nPKC subfamily [11][12][13]. Initially designated as 2 BioMed Research International a member of the nPKC subfamily [14], PKD was later found to be homologous with Ca 2+ /calmodulin-dependent kinase (CaMK) superfamily and thereby reassigned as an independent kinase family consisting of PKD1, PKD2, and PKD3 [15,16]. Accumulating evidence has implicated the role of PKD family in diverse cellular processes, such as the mediation of the cellular effect of multiple hormones and growth factors, the fission of vesicles from trans-Golgi network, the oxidative stress response, and the activation of T and B cells (for review, see [15,16]). The role of PKD in HIV-1 transcription, however, is still unknown.
In this study, we examined the function of PKD in prostratin-stimulated transcription of HIV-1 by using HeLa cells with an integrated HIV-LTR-luciferase gene and Jurkatbased J-Lat clone 2D10 cells containing transcriptionally latent HIV-1 provirus with eGFP in place of Nef as reporter systems. Our data showed that prostratin mainly activates the transcription initiation of HIV-1 provirus. PKD3, but not other PKDs, is essential for this transcription activation. Moreover, we identified that PKC of the nPKC family is able to activate PKD3, which in turn enhances NF-B nuclear translocation for the expression of HIV-1. Thus, our data revealed a mechanism of prostratin-stimulated HIV-1 transcription via PKC /PKD3/NF-B signaling pathway.

Chemicals.
Prostratin is from LC Laboratories. The PKC inhibitors Gö6976, Gö6983, and Gö6850 (bisindolylmaleimide I) are from Calbiochem. BAY-11-7085 (BAY) is from Santa Cruz. All other chemicals are from Amresco or Sigma.

Treatment of Cells with Various Pharmacological
Compounds. Cells at 50% confluence were pretreated with various inhibitors for 1 hr, followed by 2 M prostratin treatment for 6 hrs. For J-Lat 2D10 cells, cells were cotreated with 0.5 M prostratin plus indicated inhibitors for 16 hrs. The final concentration of inhibitor was as follows: 2 M for Gö6976, 1 M for Gö6983, 2 M for Gö6850, and 10 M for BAY-11-7085 (BAY). [25,26] or the HeLa ells transfected with HIV-LTR-luciferase or NF-B-luciferase reporter gene were incubated with indicated pharmacological compounds in six-well plates. Cell lysates were prepared to measure luciferase activity as previously described [21]. The error bars were calculated based on three independent experiments.

Flow Cytometry
Analysis. 2D10 cells were measured for GFP expression on flow cytometer. 10,000 events were collected and analyzed. The percentage of GFP-positive cells was presented to reflect the transcriptional activation of the latent HIV-1 provirus. The error bars were calculated based on three independent experiments.

Quantitative RT-PCR (qRT-PCR).
qRT-PCR was performed as previously described [21]. In brief, total RNA was isolated from cells with Trizol (Invitrogen), and 1 g of the RNA was reverse-transcribed with ReverTra Ace qPCR RT Kit (Toyobo) in a total volume of 20 L. For real-time PCR amplification, 2 L of cDNA was used as template. The PCR amplification was performed on Eppendorf Mastercycler ep realplex 2 with the following program: 95 ∘ C for 2 min followed by 35 cycles of 10 sec at 95 ∘ C, 15 sec at 60 ∘ C for annealing, and 20 sec at 70 ∘ C for extension. The expression levels were normalized to that of -actin. The error bars were calculated based on three independent experiments. The primers were as follows: PKD1 (forward 2.9. Chromatin Immunoprecipitation (ChIP). ChIP was carried out with HIV-LTR-Luc integrated HeLa cells as described previously [21]. Immunoprecipitated DNA was analyzed by real-time PCR with SYBR Green Realtime PCR Master Mix Plus (Toyobo) and primers in HIV-LTR promoter region (−111 to −31, forward: 5 -GCT ACA AGG GAC TTT CCG CTG G, reverse: 5 -AGG ATC TGA GGG CTC GCC ACT). The PCR amplification was performed on Eppendorf Mastercycler ep realplex 2 with the following program: 95 ∘ C for 2 min followed by 35 cycles of 10 sec at 95 ∘ C, 15 sec at 60 ∘ C for annealing, and 20 sec at 70 ∘ C for extension. The results from two independent experiments were averaged and plotted as percentage of input.

Immunofluorescence (IF).
HeLa cells with or without transfection were cultured on glass coverslips and treated as indicated. After PBS wash, the cells were subjected to immunostaining and visualized with confocal microscopy as previously described [25,27].

Prostratin Activates the Transcriptional Initiation of Latent HIV Provirus.
To study the molecular mechanism of HIV-1 transcription activation by prostratin, we employed a HeLa-based cell line with an integrated HIV-LTR-luciferase reporter gene (HIV-LTR-Luc) and a Jurkat-based cell line containing a transcriptionally latent HIV-1 provirus with eGFP in place of Nef (J-Lat clone 2D10) [24] as model systems. In HIV-LTR-Luc cells, 2 M prostratin treatment could progressively induce the luciferase activity to about 20fold by 6 hrs as indicated by luciferase assay (Figure 1(a)). In J-Lat 2D10 cells, prostratin treatment for 16 hrs induced a concentration-dependent increase of the GFP-positive cells to 54% in 0.5 M prostratin treated cells, as indicated by flow cytometry analysis (Figure 1(b)). Hence, we used 2 M prostratin treating for 6 hrs for HeLa cells with HIV-LTR-Luc reporter gene and 0.5 M prostratin for 16 hrs for 2D10 cells.
To investigate whether the prostratin-induced expression of latent HIV-1 occurs at transcription initiation or elongation step, we utilized a real-time RT-PCR (qRT-PCR) assay with the primers targeting distinct regions of the transcript. The amplification of TAR region (12-59) represents transcription initiation, and that of the luciferase gene represents transcription elongation (Figure 1(c), top panel). As indicated in the bottom panel of Figure 1(c), prostratin treatment increased the products corresponding to initiation as well as elongation to a similar extent, suggesting that prostratin was able to activate transcription initiation with no apparent effect on transcription elongation.

Knocking Down of PKD3 Blocks Prostratin-Activated
HIV-1 Transcription. PKDs have been implicated as the downstream effectors of novel PKC (nPKC) subfamily [11][12][13], which has been shown to be involved in prostratininduced transcription activation of HIV-1. There are three isoforms of PKD with different expression profiles in various human tissues and cell lines. To explore the role of PKDs in prostratin-induced activation of latent HIV-1, we first examined the expression profile of PKDs in HeLa and Jurkat cells. qRT-PCR analysis indicated that PKD2 and PKD3, but not PKD1, are expressed in these two cell lines (Figure 2(a)). Hence, PKD1 was ruled out from the list of candidates.

Prostratin Activates PKD3 to Enhance HIV-1 Transcription.
It has been reported that the kinase activity of PKD3 relies on the phosphorylation of its activation loop at Serine 731 and 735 (phospho-S731/735) [12]. Hence, we tested whether prostratin could induce the phosphorylation of PKD3's activation loop. As the commercially available antibody reacts with the phosphorylated activation loops of all 3 PKDs that have similar molecular weights, we used GFP-PKD3 construct to distinguish it from the endogenous PKDs by size. HeLa cells expressing GFP-PKD3 were treated with prostratin, and the levels of phospho-S731/735 of GFP-PKD3 at different time points were analyzed by Western blotting (Figure 2(d)). The phosphorylated S731/735 of GFP-PKD3 was undetectable in the untreated cells but was dramatically increased after 0.5 hrs of prostratin treatment and reached high levels at 2 hrs, suggesting that prostratin is able to activate PKD3 by inducing the phosphorylation of its activation loop.
Next, we tested whether the HIV-1 expression relies on the phosphorylation of PKD3's activation loop. As shown in Figure 2(e), in the absence of prostratin treatment, the overexpression of wild-type PKD3 (PKD3-WT) did not induce the expression of HIV-1. Remarkably, the PKD3-CA, a mutant PKD3 containing S731E/S735E mutation with the acidic glutamates to mimic phosphorylated activation loop and thereby rendering it constitutively active, enhanced HIV-1 expression (Figure 2(e), CA). As expected, overexpressing of the kinase-dead form of PKD3 (PKD3-KD), which contains S731A/735A mutations that prevent it from phosphorylation and therefore lacks kinase activity [19,28], failed to induce HIV-1 expression (Figure 2(e), KD). Taken together, these results indicated that the prostratininduced HIV-1 expression relies on prostratin-induced activation of PKD3.

Novel PKCs Are Required for Prostratin-Induced Activation of PKD3.
To address the role of PKC in activating PKD3 during prostratin-induced transcription activation of latent HIV-1 provirus, both HIV-LTR-Luc and J-Lat 2D10 cells were pretreated with various PKC inhibitors followed by prostratin incubation (Figures 3(a) and 3(b)). Gö6983 which inhibits all three subfamilies of PKC but not PKD1 [29] and Gö6850 which inhibits classic PKC (cPKC) and novel PKC (nPKC) but not atypical PKC (aPKC) subfamily [30] blocked prostratin-induced HIV-1 expression, suggesting that aPKC subfamily is not required for this process. Consistent with the fact that both HeLa and Jurkat cells lack PKD1 expression (Figure 2(a)), the inhibitor Gö6976 which inhibits cPKC and PKD1, but not nPKC [30], did not block prostratin-induced HIV-1 expression but enhanced HIV-1 expression. These data indicated that nPKC subfamily is required for prostratinstimulated HIV-1 expression.
Since both nPKC and PKD3 (Figure 2) are required for prostratin-induced HIV-1 expression, we asked whether the activation of PKD3 is regulated by nPKC. The effect of PKC inhibitors on the prostratin-induced phosphorylation of PKD3 was tested. In line with their effect on HIV-1 expression (Figures 3(a) and 3(b)), Gö6983 and Gö6850, but not Gö6976, remarkably blocked prostratin-induced phosphorylation of GFP-PKD3 at its activation loop (Figure 3(c)), indicating that nPKC is required for PKD3 activation.

Prostratin-Induced PKD3 Activation Is Mediated by PKC .
There are 4 members of nPKC subfamily that can be classified as two subgroups, namely, PKC and closely related PKC , as well as PKC and closely related PKC [31]. To test whether the two subgroups of nPKC are involved in prostratininduced HIV-1 expression, we constructed shRNAs targeting PKC and PKC , respectively ( Figure S1B). As shown in Figure 3(d), cotransfecting HIV-LTR-luciferase reporter construct with shPKC , but not shPKC , reduced prostratininduced HIV-1 expression, suggesting that PKC is required for this process.
Next, we asked whether the PKD3 is the downstream effector of PKC . To this end, we cotransfected GFP-PKD3 together with constitutively active form (CA) of PKC , PKC , or PKC and tested the phosphorylation levels of S731/S735 of GFP-PKD3 (Figure 3(e)). In line with the data of knockdown experiment (Figure 3(d)), overexpressing CA-PKC , but not CA-PKC or CA-PKC , induced significant phosphorylation of GFP-PKD3 (Figure 3(e)), indicating that PKC but not PKC plays a major role in activating PKD3. Moreover, although overexpressing either CA-PKC or CA-PKC induced the expression of HIV-1, silencing PKD3 only blocked HIV-1 expression induced by CA-PKC , but not that by PKC (Figure 3(f)), suggesting that PKC may activate HIV-1 expression independent of PKD3.
Taken together, these data indicated that PKC , but not PKC , functions upstream of PKD3 in prostratin-stimulated transcriptional activation of latent HIV-1 provirus. Although both PKC and PKC are novel PKCs, their sequence divergence at the N-terminus may render them unique functions.

Prostratin Enhances Promoter Binding of NF-B to
Augment HIV-1 Transcription. A previous report showed that prostratin enhanced nuclear translocation of NF-B to activate HIV-1 transcription [10]. Since both PKD1 and PKD2 were shown to activate NF-B [32,33] and recently PKD2 and PKD3 were reported to activate NF-B pathway in prostate cancer cells [34], we suspected that prostratin may enhance HIV-1 expression through PKD3's activation effect on NF-B pathway. To test this, we first examined whether the prostratin's effect relies on the cis-element B of HIV-1 promoter. Using COP mutagenesis methods [22], the following HIV-1 promoter mutant constructs were made: dSp1 (without Sp1 binding sites), dEnh (without B enhancer element), dTAR (without TAR sequences), or dSp1/dEnh (Figure 4(a)). These constructs were transiently transfected into HeLa cells followed by the treatment with 2 M prostratin for 6 hrs, and the cells were harvested for luciferase assay. As shown in Figure 4(b), the deletion of either Sp1 or TAR element did not interfere with the activation effect of prostratin, whereas the deletion of B enhancer element (dEnh) or the combined deletion of Sp1 and B enhancer (dSp1/dEnh) reduced the activation effect of prostratin. Hence, the activation effect of prostratin depends on the presence of B enhancer element of the HIV-1 promoter.
Next, we performed chromatin immunoprecipitation to examine whether prostratin enhances the binding of NF-B and Sp1 to HIV-1 promoter. Upon prostratin treatment, the binding of RelA subunit of NF-B increased, whereas the binding of Sp1 was not affected (Figure 4(c)). Moreover, in the presence of NF-B inhibitor BAY-11-7085 (BAY), which blocks the nuclear translocation of NF-B [35], the prostratin-induced transcription initiation was greatly reduced (Figure 4(d)).
Therefore, these data showed that prostratin enhanced transcription initiation by increasing the promoter recruitment of NF-B.

PKD3 Is Required for B-Dependent Transcription Activation of HIV-1 by Prostratin.
To address the role of PKD3 on NF-B-dependent transcription activation of HIV-1, we utilized a luciferase reporter construct that contains 5 tandem copies of B enhancer (NF-B-Luc). Remarkably, PKD3 silencing blocked the stimulatory effect of prostratin on the reporter expression ( Figure 5(a)), indicating that PKD3 is required for this process. Moreover, the overexpression of either wild-type PKD3 or the constitutive active form of PKD3 (PKD3 CA) enhanced the expression of NF-B-Luc, whereas the kinase-dead mutant (PKD3-KD) did not ( Figure 5(b)), indicating that the kinase activity of PKD3 is required for activating NF-B signaling pathway.
If PKD3 activates HIV-1 transcription through NF-B pathway, one would expect that transcription activation should depend on both the PKD3 kinase activity and the B enhancer element. As shown in Figure 5(c), the overexpression of the PKD3-CA enhanced the expression of wild-type and dTAR luciferase reporter gene, but not the dEnh reporter, whereas the PKD3-KD failed to enhance the expression of all three reporters. These observations indicated that PKD3 activates transcription through NF-B pathway.

PKD3 Is Required for Prostratin-Induced Nuclear
Translocation of NF-B. As a transcription factor, NF-B must translocate to nucleus to activate gene transcription. A previous study showed that prostratin could enhance the nuclear translocation of NF-B [10]. Here, we tested whether PKD3 is required for this process ( Figure 5(d)). NF-B was present in cytoplasm in untreated cells and translocated into nucleus upon prostratin treatment in cells transfected with a control shRNA construct. Notably, in cells transfected with PKD3 shRNA, prostratin failed to induce the nuclear translocation of NF-B, indicating that PKD3 is essential for prostratininduced nuclear translocation of NF-B.

Conclusions
In this study, we investigated the molecular mechanism of how prostratin induces the transcriptional activation of latent HIV-1 provirus and propose the following model ( Figure 5(e)): prostratin works through PKC to induce the phosphorylation of PKD3's activation loop at S731/S735, thereby activating PKD3. Subsequently, the activated PKD3 enhances the nuclear localization of NF-B, which binds to B element at HIV-1 promoter and thereby enhances transcription initiation of the latent HIV-1 provirus.
Intriguingly, although both PKD2 and PKD3 are expressed in HeLa and 2D10 cells, only PKD3 was shown to be involved in prostratin-induced HIV-1 transcription activation (Figures 2 and 3). PKD3 has been shown to be structurally distinct from PKD2 [15,16]. PKD2 but not PKD3 contains an N-terminal hydrophobic domain and a C-terminal PDZ binding motif [36]. The PH (pleckstrin homology) and C1 domains of PKD3 are also more divergent than those of PKD1 and PKD2 [36,37]. These structural differences may enable PKD3, but not PKD2, to be the downstream effector of PKC . Moreover, PKD3, but not PKD2, was found to translocate to nucleus upon stimulation and this process is dependent on its unique C-terminal sequence [38,39]. Hence, PKD3 may be well suited for a nuclear function such as transcription activation.
PKD1 and PKD2 were known to activate NF-B pathway about 10 years ago. PKD1 was found to be essential for the activation of NF-B pathway during oxidative stress [40]. Similarly, PKD2 was reported to mediate the activation of NF-B pathway in chronic myeloid leukemia cells [32]. The role of PKD3 in NF-B pathway, however, was not known until most recently. In a study of prostate cancer cell lines, PKD3 was implicated in the activation of NF-B pathway to enhance uPA gene expression [34]. Our study provides a second example for PKD3's function and strengthens the notion that PKD3 is an important upstream factor for activating NF-B pathway to regulate gene transcription.
In summary, our study revealed for the first time the essential role of PKD3 in prostratin-induced HIV-1 transcription and a mechanism of prostratin-stimulated HIV-1 transcription via PKC /PKD3/NF-B signaling pathway. These results should provide a foundation for further studies in cells derived from individuals with latent HIV infection.