We used a luciferase reporter construct to test the effect of vFLIP K13 on HIV-1 LTR transcriptional activation. This reporter construct expresses the firefly luciferase gene downstream of the HIV-1 LTR. As shown in Fig. 1A–C transient transfection of vFLIP K13 in 293T and Cos7 cells led to significant (3 and 5 fold, respectively) activation of the HIV-1 LTR where as expression of the vFLIP E8 from the equine herpes virus 2 failed to do so. As HIV-1 LTR is known to be responsive to proinflammatory cytokines, we also carried out a comparative analysis of the HIV-1 LTR activation by K13, TNF-α and IL-1β in 293T cells. As shown in Fig 1C, while K13-induced approximately 3-fold increase in HIV-1 LTR transcriptional activation, treatment with TNF-α(50 ng/ml) and IL-1β (50 ng/ml) resulted in 5–6 fold increase. A possible explanation for this difference lies in the fact that unlike TNF-α and IL-1β, K13 lacks the ability to induce the transcription factor AP1, which is known to activate HIV-1 LTR. We also tested whether vFLIP K13 possesses the ability to activate the HIV-1 LTR in cells naturally infected with HIV-1. As shown in Fig. 1D, transient transfection of K13 in Jurkat cells (human T cell lymphoma cell line) led to modest (2-fold) activation of HIV-1 LTR transcription activity.

We have recently generated point mutants of the vFLIP K13 which differ in their ability to activate the NF-κB pathway 27. In order to test the hypothesis that vFLIP K13 activates the HIV-1 LTR via NF-κB pathway, we carried out a comparative analysis of the ability of wild-type and mutant K13 constructs to activate the HIV-1 LTR reporter construct. In a parallel experiment, we also tested the effect of different K13 constructs on an NF-κB luciferase reporter construct to serve as a positive control. The luciferase expression in the latter construct is driven by four copies of a consensus NF-κB binding-site 28. Consistent with our published results 27, the triple mutant 58AAA demonstrated a complete lack of NF-κB reporter activation while the mutant 67AAA retained partial ability to do so (Fig 2A). Importantly, essentially a similar pattern of reporter activation was obtained when the wild-type and mutant K13 constructs were tested on the HIV-1 LTR reporter construct (Fig 2B). Collectively, the above results suggested the involvement of the NF-κB pathway in vFLIP K13-induced HIV-1 LTR activation.

In order to test the hypothesis that vFLIP K13 activates HIV-1 LTR by inducing the binding of specific transcription factors to the NF-κB binding sites present in the HIV-1 LTR, we used an electrophoretic mobility shift assay (EMSA). As shown in Fig. 3A, nuclear extracts from Jurkat cells expressing vFLIP K13 demonstrated significant DNA-binding activity on radiolabelled oligonucleotides-derived from the NF-κB binding sites present in HIV-1 LTR. In contrast, no HIV-1 LTR DNA-binding activity was observed in nuclear extracts of empty vector-expressing cells (Fig. 3A, compare lanes 1 and 2). The specificity of the complex was demonstrated by its disappearance upon competition with excess cold HIV-1 LTR oligonucleotide duplex and lack of effect upon competition with a non-specific oligonucleotide duplex (Fig. 3A, lanes 3 and 4).

In addition to the classical NF-κB pathway, an alternative (or non-canonical) pathway of NF-κB activation, which involves proteasome-mediated processing of p100/NF-κB2 into p52 subunit, has been described 8. We have recently demonstrated that vFLIP K13 can activate the alternate NF-κB pathway via an IKK1-dependent and NIK- and IKK2-independent process 24. In order to determine the contribution of the classical vs alternate NF-κB pathway to vFLIP K13-induced HIV-1 LTR activation, we used a supershift assay to analyze the nature of the protein complexes bound to HIV-1 LTR from nuclear extracts of vFLIP K13-expressing cells. This assay demonstrated that p50 and c-Rel subunits are the major components of the HIV-1 LTR-bound NF-κB complexes induced by vFLIP K13 with modest contribution from the p65 subunit (Fig. 3B). As the p50, c-Rel and p65 subunits are primarily activated by the classical NF-κB pathway, the above results support the hypothesis that K13 activates the HIV-1 LTR via the classical NF-κB pathway.

We have previously demonstrated that vFLIP K13 activates the classical NF-κB pathway via phosphorylation of IκBα, which leads to its ubiquitination and subsequent degradation via proteasome 22. We used a phosphorylation-resistant mutant of IκBα to test the involvement of the classical NF-κB pathway in vFLIP K13-induced HIV-1 LTR reporter activity. As shown in Fig. 4A, a phosphorylation-resistant mutant of IκBα (IκBαSS32/36AA), in which the two critical N-terminal serine residues have been mutated to alanine, completely blocked vFLIP K13-induced HIV-1 LTR reporter activity.

We used siRNA-mediated downregulation of key subunits of the classical and alternate NF-κB pathways to test their involvement in K13-induced HIV-1 LTR activation. As shown in Fig. 4B, we achieved effective silencing of c-Rel and RelA/p65 expression by siRNA-mediated silencing. Consistent with our supershift assay (Fig. 3B), siRNA-mediated silencing of c-Rel expression led to almost complete suppression of K13-induced HIV-1 LTR activation (Fig. 4C). Similarly, silencing of p65 expression led to significant suppression of HIV-1 LTR activity, although some residual activity was still evident (Fig. 4C). Although p100 acts as a precursor of p52, another important function of p100 is to retain the RelB/p50 and RelB/p52 complexes in the cytoplasm. As such, in order to shut-off the alternate NF-κB pathway, we chose to silence the expression of RelB. As shown in Fig. 4B–C, siRNA-mediated downregulation of RelB, had no significant effect on K13-induced HIV-1 LTR activity. We also failed to observe any effect of p100/p52 silencing on HIV-1 LTR activation (data not shown). Taken together, the above results demonstrate a key role of the c-Rel and p65 subunits of the classical NF-κB pathway in K13-induced HIV-1 LTR reporter activation.

K13 is known to associate with a 700 kDa multi-subunit IKK complex, which consists of two catalytic subunits, IKK1/IKKα and IKK2/IKKβ and a regulatory subunit, NEMO/IKKγ 22. We tested the involvement of the individual components of the IKK complex in vFLIP K13-induced HIV-1 LTR reporter activity by using mouse fibroblast (MEF) cells deficient in IKK1, IKK2 and NEMO, respectively. As shown in Fig. 5A, we observed significant HIV-1 LTR reporter activity by the expression of vFLIP K13 in the wild type MEF cells. In contrast, almost no HIV-1 LTR reporter activity was observed in NEMO-deficient cells. However, some residual HIV-1 LTR reporter activity was observed in IKK1- and IKK2-deficient MEF cells. Collectively, the above results suggest that synergistic action of IKK1, IKK2 and NEMO is required for maximal activation of HIV-1 LTR by K13.

Next we sought to determine whether pharmacological inhibitors of the NF-κB pathway may be used to block vFLIP K13-induced HIV-1 LTR reporter activation. Lactacystin and MG132 are inhibitors of proteasome and block the NF-κB pathway by preventing the degradation of IκB. On the other hand, arsenic acid is believed to block the NF-κB pathway by inhibiting the IKK complex 29. As shown in Fig. 5B, vFLIP K13-induced HIV-1 LTR reporter activation was effectively blocked by MG132, lactacystin and arsenic acid. These results suggest that inhibitors of the NF-κB pathway might have a role in preventing K13-induced HIV-1 LTR reporter activation.

Murr1 is a gene product that has been previously implicated in copper regulation 3031. A recent study demonstrated that Murr1 is highly expressed in CD4+ T cells and serve as a genetic inhibitor factor for HIV-1 replication in the resting lymphocytes 32. Murr1 was shown to block HIV-1 LTR activation and HIV-1 replication by inhibiting the proteasomal degradation of IκB and blocking basal and cytokine-stimulated NF-κB activation 32. Based on the above study demonstrating the importance of Murr1 as an endogenous regulator of HIV-1 LTR activation, we tested its effect on K13-induced HIV-1 LTR activation. As shown in Figure 5C, co-expression of Murr1 led to significant block in K13-induced HIV-1 LTR reporter activity, thereby suggesting that K13-induced activation of HIV-1 replication in resting lymphoid cells may be regulated by Murr1 and K13 may selectively activate HIV-1 replication in activated cells in which expression of Murr1 is known to be down-regulated 32.

HIV-1 Tat is a viral nuclear protein that plays an essential role in HIV-1 gene expression at the transcriptional level 23. Tat has been shown to associate with p300/CBP and P/CAF histone acetyltransferases (HAT) and efficient activation of the integrated HIV-1 LTR is largely dependent on Tat-dependent rearrangement of the nucleosome positioned at the transcription start site 2. HIV-1 LTR is known to bind and respond to HIV Tat protein via a specific Tat-binding site 2. We used deletion mutagenesis of the HIV-1 LTR to test whether vFLIP induced transcriptional activation is dependent on this Tat-binding site. As shown in Figures 6A and 6B, a bulge mutant (containing deletion of nucleotides +23/+25) of HIV-1 LTR, which is defective in Tat activation 33, had no significant effect on vFLIP K13-induced reporter activity. In contrast, vFLIP K13 failed to activate a luciferase report construct containing an HIV-1 LTR in which the NF-κB binding sites had been mutated (Fig. 6C). The above results confirm that vFLIP activates the HIV-1 LTR via the NF-κB binding sites and can do so independent of the Tat-binding site.

Transcriptional activation of genes is usually regulated by multiple transcription factors acting in concert. Thus, while NF-κB has been shown to play a major role in the activation of the HIV-1 LTR, it fails to do so when acting alone 25343536. Along the same lines, the transactivating function of Tat protein requires the presence of NF-κB sites in the HIV-1 LTR and Tat protein is known to cooperate with NF-κB to activate the HIV-1 LTR 13436. We hypothesized that a functional interaction between K13-induced NF-κB and Tat may be particularly important in the early stages of HIV-1 infection when the amount of Tat is limited. To test this hypothesis, we began by performing a dose-response analysis of Tat and selected a dose of Tat (20 ng/ml) which led to sub-maximal activation of HIV-1 LTR activation in 293T cells (Fig. 6D). Next, we analyzed the effect of co-expression of Tat on K13-induced HIV-1 LTR activation. As shown in Figure 6E, while transfection of K13 (250–500 ng/well) led to approximately 2.5–3.5 fold increase in HIV-1 LTR activation, transfection of Tat (20 ng/well) induced 4-fold increase in HIV-1 LTR activity. However, co-expression of K13 with Tat led to a synergistic 12-fold activation of the HIV-1 LTR. These results suggest that K13-induced NF-κB functions synergistically with the Tat protein to activate the HIV-1 LTR.

There is considerable sequence diversity among the HIV-1 isolates that comprise the current global pandemic and these can be grouped into several distinct subtypes or clades 37. In particular, the LTRs of different subtypes show distinct enhancer-promoter configuration and vary in the sequence and number of binding sites for different transcription factor, including NF-κB 3839. Although different HIV-1 LTRs are transcriptionally active, they differ in the level of basal reporter activity 3839. In addition, different HIV-1 LTRs are known to show differential response to TNF-α treatment, which correlates with the number of NF-κB binding sites 3839. Therefore, we sought to determine whether vFLIP K13 will differentially activate luciferase reporter constructs driven by LTRs derived from different HIV strains. Consistent with the published studies 3839, we observed considerable difference in the basal activities of different HIV-1 LTRs promoters when transfected into 293T cells along with an empty vector (Fig. 7A). More importantly, coexpression of vFLIP K13 led to differential activation of luciferase reporter constructs containing LTRs from different subtypes of HIV-1 (Fig. 7A). Thus, subtype C, which possesses three NF-κB binding sites showed the maximum increase in vFLIP-induced HIV-1 LTR reporter activity while subtype E, which possesses only one NF-κB binding site showed the lowest level of basal and vFLIP-induced HIV-1 LTR transcriptional activation (Fig. 7A,B). These results demonstrate that, similar to situation with TNFα, K13 may differentially activate LTRs derived from different strains of HIV-1, which correlate with their NF-κB binding sites.

Plasmids containing pcDNA3-K13-Flag and pcDNA3-E8-Flag, pRSV/LacZ and 293T cells have been described previously 21. An expression construct encoding Murr-1 was generated by RT-PCR using cDNA prepared from H460 cells as a template and subsequently cloned in pcDNA3 vector with a C-terminal HA tag. Luciferase reporter constructs containing LTRs derived from different strains of HIV-1 (pBlue3'LTR-Luc-A-F) were obtained from AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH from Drs. Reink Jeeninga and Ben Berkhout. Wild-type and mutant HIV-1 LTR reporter constructs 61 and expression constructs for HIV Tat were obtained from Dr. Richard Gaynor. The IKKα-/- and IKKβ-/- mouse embryonic fibroblast cells were generated in Dr. Inder Verma's laboratory 6263 and IKKγ/NEMO-/- cells were generated in Dr. Michael Karin's laboratory 64. These cells were kindly provided by Dr. Richard Gaynor and were maintained in DMEM supplemented with 10% FBS. Jurkat cells were cultured in RPMI medium supplemented with 10% FBS and selected in the presence of 1500 μg/ml of G418 (Invitrogen). 293T cells were grown in DMEM with 10% FBS. Arsenic Acid was purchased from Sigma. MG132 and lactacystin were purchased from Calbiochem and Biomol, respectively. Retrovirus constructs containing C-terminal Flag epitope tagged HHV8 vFLIP (K13-Flag) was generated in MSCV neo-based retroviral vector and amphotropic viruses generated and used for infection as described previously 22.

Electrophoretic mobility shift assay was performed essentially as described previously 22, except an HIV LTR oligonucleotide duplex (sense strand, 5' TGC TAC AAG GGA CTT TCC GCT GGG GAC TTT CCA GG 3') was used instead of κB binding oligonucleotide. Nuclear extracts were prepared from Jurkat cells stably expressing an empty vector or vFLIP K13, which have been described previously 23. Antibodies against p50, p65, RelB and c-Rel were purchased from Santa Cruz Biotechnology. An antibody against p52 was purchased from Upstate biotechnology

293 T cells were transfected in duplicate in a 24-well plate with the various test plasmids along with an HIV LTR/luciferase reporter construction (10 ng/well) and a pRSV/LacZ (β-galactosidase) reporter construct (75 ng/well) using calcium phosphate transfection protocol as described previously 21. Cells were lysed 36–48 hours later and extracts were used for the measurement of firefly luciferase and galactosidase activity. Luciferase activity was normalized relative to the galactosidase activity to control for the difference in the transfection efficiency. Cos-7, Jurkat and MEF cells were transiently transfected with empty vector (pCDNA3) or K13 (500 ng/well) along with a HIV/luciferase reporter construct (100 ng/well) and a synthetic Renilla luciferase (phRL-TK; Promega) reporter vector (75 ng/well) by using LIPOFECTAMINE 2000 Reagent (Invitrogen, Carlsbad, CA) according to manufacturer's instruction. Thirty-six hours after transfection, cells lysates used for reporter assays. Luciferase activity was normalized relative to the Renilla luciferase activity to control for the difference in the transfection efficiency. The values shown are averages (Mean ± S.E.) of one representative experiment out of three in which each transfection was performed in duplicate.

siRNA oligonucleotides with two thymidine residues (dTdT) at the 3'-end of the sequence were designed to p65 (sense, 5'-GCCCUAUCCCUUUACGUCAdTdT-3'), c-Rel (sense, 5'-CAACCGUGCUCCAAAUACU dTdT-3'), RelB (sense, AGAUCAUCGACGAGUACAUdTdT-3') and control (sense, 5' GCGCGCUUUGUAGGAUUCGdTdT-3'), along with their corresponding antisense oligonucleotides. The RNA oligonucleotides were synthesized at RNA Oligonucleotide Synthesis Core facility, UT Southwestern Medical center. siRNA oligonucleotides (80 nM) were transfected using calcium phosphate as described previously 65.

Western blot analysis was performed essentially as described previously 22. Primary antibodies used in these experiments were: p65, c-Rel, Rel-B (rabbit polyclonal, Santa Cruz biotechnology) and actin (mouse monoclonal, Sigma).