HIV-1 (human immunodeficiency virus) transcription is primarily controlled by the virally encoded Tat (transactivator of transcription) protein and its interaction with the viral TAR (transcription response element) RNA element. Specifically, binding of a Tat-containing complex to TAR recruits cellular factors that promote elongation of the host RNA polymerase engaging the viral DNA template. Disruption of this interaction halts viral RNA transcription. In the present study, we investigated the effect of pokeweed antiviral protein (PAP), an RNA glycosidase (EC#: 188.8.131.52) synthesized by the pokeweed plant (Phytolacca americana), on transcription of HIV-1 mRNA. We show that co-expression of PAP with a proviral clone in culture cells resulted in a Tat-dependent decrease in viral mRNA levels. PAP reduced HIV-1 transcriptional activity by inhibiting Tat protein synthesis. The effects of PAP expression on host factors AP-1 (activator protein 1), NF-κB (nuclear factor kappa-light-chain-enhancer of activated B-cells) and specificity protein 1, which modulate HIV-1 transcription by binding to the viral LTR (5′-long terminal repeat), were also investigated. Only AP-1 showed a modest JNK pathway-dependent increase in activity in the presence of PAP; however, this activation was not sufficient to significantly enhance transcription from a partial viral LTR containing AP-1 binding sites. Therefore, the primary effect of PAP on HIV-1 transcription is to reduce viral RNA synthesis by decreasing the abundance of Tat. These findings provide a mechanistic explanation for the observed decrease in viral RNAs in cells expressing PAP and contribute to our understanding of the antiviral effects of this plant protein.
Pokeweed antiviral protein (PAP) is an RNA N-glycosidase (EC#: 184.108.40.206) synthesized by the pokeweed plant (Phytolacca americana) that removes purine bases from various RNAs [1,2]. Its canonical substrate is the rRNA, and depurination within the conserved sarcin/ricin stem loop inhibits translation by preventing the binding of elongation factors to the ribosome [3,4]. The antiviral activity of PAP was first considered to be an indirect consequence of this rRNA depurination, whereby inhibition of translation debilitates or kills the cell and limits virus proliferation [5,6]. However, PAP has been shown to directly depurinate some viral RNAs, rendering them defective. For example, adenines are released upon incubation of PAP with HIV-1 (human immunodeficiency virus) mRNAs  and PAP removes purine bases from HIV-1 mRNAs when expressed in human cell culture [8,9]. This depurination correlates with observed decreased HIV-1 protein expression , and is the result of ribosomes stalling at abasic sites in template mRNAs . Therefore, an alternate mechanism for the antiviral activity of PAP is the post-transcriptional regulation of HIV-1 gene expression through the removal of purines from viral mRNAs.
In this report, we investigated whether inhibition of HIV-1 transcription also contributes to lack of virus production observed in cells expressing PAP. HIV-1 transcription is primarily regulated by the virally encoded Tat (transactivator of transcription) protein and the transcription response element (TAR) found within the 5′-long terminal repeat (LTR) of the viral mRNA [12,13]. The LTR is further divided into three regions known as untranslated 3′region (U3), repeat (R) region and untranslated 5′region (U5). Upstream enhancer sequences within U3 recruit cellular activator protein 1 (AP-1), specificity protein 1 (SP-1) and nuclear factor kappa-light-chain-enhancer of activated B-cells (NF-κB) [14–16]. Transcription begins from the R region, which also encodes TAR and forms an mRNA hairpin  in the nascent viral mRNA and interacts with viral Tat protein . Previous data have shown that in the absence of Tat, the level of viral transcripts is greatly decreased and that Tat is essential for viral replication [19,20]. Once the cellular polymerase initiates transcription at the 5′LTR, it stalls shortly after transcribing the TAR mRNA hairpin [21,22]. Subsequent activation of transcription requires phosphorylation of both negative and positive regulators of transcription elongation. Specifically, Tat binds to P-TEFb, which localizes it to TAR and activates the enzyme [23,24]. The kinase component of P-TEFb phosphorylates STP5 and NELF-E, subunits of negative regulators, causing their rearrangement and/or release from the elongation complex [25,26]. In addition, P-TEFb hyperphosphorylates the C-terminal domain of the RNA polymerase II, increasing the processivity of the polymerase and transcription of full-length HIV-1 mRNAs [27,28].
Our previous work with Human T-cell leukemia virus revealed decreased levels of viral mRNAs in cells expressing PAP, with corresponding decreased viral protein levels, including Tax, which is functionally similar to HIV-1 Tat protein . However, we did not investigate whether this decline in mRNA and protein was a specific consequence of Tax RNA depurination by PAP, or some other effect of the enzyme. In addition, we observed a modest activation of c-Jun NH2-terminal kinase 1 (JNK1) and extracellular signal-regulated kinase (ERK1/2) in the presence of PAP [10,30]. The downstream targets of these kinase pathways include the transcription factors AP-1, SP-1 and NF-κB [16,31,32]. Therefore, PAP activation of these host factors could augment the Tat-dependent regulation of transcription. Also, as PAP is active on viral mRNAs, HIV-1 transcription could be modulated by PAP-mediated inactivation of Tat mRNA. Accordingly, the goal of the present study was to determine whether PAP affects HIV-1 transcription and if yes, to characterize the mechanism of this activity.
Materials and methods
Cell culture, plasmids and drugs
HEK 293T cells were cultured in Dubelco's Eagle Modified Medium (DMEM), with 10% fetal bovine serum (FBS) and penicillin/streptomycin (10 mg/ml). Jurkat cells were maintained in RPMI 1640 supplemented with the same level of FBS and antibiotics. The plasmids encoding 3X-FLAG-tagged mature PAP, pcPAP, and the active site mutant (E176V) of PAP, pcPAPx, have been described previously . The pBH10fs plasmid encodes a replication-deficient (reverse transcriptase mutant) form of HIV-1 under the control of a complete HIV-1 promoter within the pHXB2gpt vector backbone. The plasmid contains a frame shift mutation at the 3′-end of the Gag open-reading frame inhibiting the translation of the Pol gene products . pHIVSCAT encodes chloramphenicol acetyltransferase (CAT) from HIV-1 LTR within the pSV-2 vector backbone . The pEF-TAT vector encodes the HIV-1 Tat protein under the control of the elongation factor 1a (EF-1a) promoter within a pEF-BOS backbone . The pAP1-Luc, pSP1-Luc and pNF-Luc expression constructs encode luciferase with AP-1, SP-1 and NF-κB sites, respectively, upstream of the Simian vacuolating virus 40 (SV40) promoter within the pGL4.12 vector backbone. The pLTR-Luc vector encodes luciferase under the control of a partial HIV-1 LTR containing the complete U3 plus 25 bp of the R region in the pGL3 Basic backbone . Actinomycin D (Life Technologies) was used as a specific inhibitor of transcription elongation at a final concentration of 5 µg/ml. PD98059 (Cell Signaling) was used as a specific inhibitor of MEK1/2 to inhibit ERK1/2 MAPK phosphorylation at a final concentration of 20 µM. SP600125 (Cell Signaling) was used as a specific inhibitor of JNK1 at a final concentration of 20 µM.
Transfection, MTT assay and immunoblot analyses
Plasmid DNA was transfected into HEK 293T cells (5 × 105 cells/ml) by the standard calcium phosphate co-precipitation . Briefly, plasmids (including 1.0 µg pcYFP, described below) were added to 1.5 ml microtubes with 405 µl dH2O and 45 µl of 2.5 M CaCl2, and then gently titrated into 15 ml tubes containing 450 µl of 2× HeBS buffer (350 mM NaCl, 60 mM HEPES pH 7.15, 1 mM Na2HPO4). The DNA was incubated at room temperature for 20 min and then added to the cells in a dropwise manner. Cells were incubated for 16 h at 37°C in a 5% CO2 environment, washed with 5 ml of 1× PBS solution, and 10 ml of fresh DMEM media was added and incubation continued for another 24 h prior to harvesting. Jurkat cells were transfected using Lipofectamine 2000 (Invitrogen), following the manufacturer's instructions. Briefly, cells were refed with fresh RPMI 1640 supplemented with 10% FBS without antibiotics and plated at a density of 5 × 105 cells/well in six-well plates prior to transfection. Plasmid DNA (including 1.0 µg pcYFP, described below) and Lipofectamine 2000 reagent were separately mixed with RPMI 1640 without serum or antibiotics, and then combined and incubated for 25 min. The mixture was added dropwise to the cells, and the cells were incubated for 4 h. Following incubation, cells were washed three times with 1× PBS, resuspended in fresh complete media and incubated for an additional 36 h at 37°C. Transfection efficiency was estimated for both cell types by counting cells, following co-transfection with pcYFP, which encodes yellow fluorescent protein from a CMV promoter.
An MTT conversion assay (Roche Applied Science) was carried out as per the manufacturer's instructions. Approximately 40 h after transfection, cells were seeded into a 96-well plate and MTT reagent (20 ml) was added to each well. The plate was incubated at 37°C for 4 h after which 100 ml of Solubilizing buffer (Roche Applied Science) was added to each well and the plate was incubated at 37°C for an additional for 4 h. Absorbance was measured at 595 nm to determine cell viability.
Cell harvesting and immunoblot analysis was carried out as previously described . Briefly, cell pellets were harvested and resuspended in two volumes (∼150–200 µl) of lysis buffer (25 mM HEPES, pH 7.5, 2 mM EGTA, 1 mM DTT, 10% glycerol, 1% Nonidet P-40). The samples were vortexed, incubated on ice for 10 min and then centrifuged at 16 000×g for 10 min at 4°C to pellet cell debris. Supernatant proteins were separated by 12% SDS–PAGE, transferred to nitrocellulose and probed for PAP with monoclonal FLAG antibody (1:2500, # F3165 Sigma), β-actin with monoclonal antibody (1:5000, # F3022 Sigma), YFP with polyclonal antibody (1:5000, #6556, Abcam) or Tat with monoclonal antibody (1:500, NIH Cat# 705).
Northern blot analyses
Cell pellets were suspended in two volumes of lysis buffer (25 mM HEPES, pH 7.5, 2 mM EGTA, 1 mM DTT, 10% glycerol, 1% Nonidet P-40). Samples were vortexed, incubated on ice for 10 min and centrifuged at 16 000×g for 10 min at 4°C to pellet cell debris. RNA was isolated from supernatants with TRIZOL reagent (Molecular Research Center) as per the manufacturer's instructions. Denaturing agarose Northern blot analysis was carried out to determine the expression levels of target mRNAs. Total cellular RNA (15 µg) was denatured at 85°C in formamide buffer (60% formamide, 20% formaldehyde, 20 mM MOPS at pH 7.0, 2 mM NaOAc, 1 mM EDTA) and separated through 0.8% agarose denaturing gel. RNA was then transferred to positively charged nylon and probed for HIV-1 mRNAs with 1 × 106 cpm [α-33P]-radiolabeled antisense riboprobe specific to nucleotides 8881–9060 of HIV-1 RNA (accession # K03455.1). To detect CAT mRNA for the Tat activity assay, the blot was probed with a 1 × 106 cpm [α-33P]-radiolabeled antisense riboprobe specific to nucleotides 220–420 of CAT mRNA. Activity assays for AP-1, SP-1 and NF-κB were probed with a 1 × 106 cpm [α-33P]-radiolabeled antisense riboprobe specific to nucleotides 110–330 of luciferase mRNA. An antisense riboprobe specific for nucleotides 365–680 of human glyceraldehyde 3-phospate dehydrogenase (GAPDH) mRNA was used as a loading control for Northern blots. Radiolabeled RNA was visualized and quantified with a phosphor imager. Densitometry data were quantified using the Quantity One software.
2-kb HIV-1 mRNAs were isolated from cells expressing PAP as described . mRNAs (5 µg) were combined with 1 × 106 cpm [γ-32P]-end labeled reverse primer (5′-CTGTCTCTGTCTCTCTCTCCACCTTCTTCTT-3′), heated denatured for 5 min and placed on ice. Samples were brought to 20 µl in buffer (75 mM KCl, 50 mM Tris–HCl pH 7.9, 10 mM DTT, 3 mM MgCl2, 1 mM dNTPs, 40 units RNase inhibitor, 25 units Superscript III reverse transcriptase) and incubated at 48°C for 90 min. Reactions were stopped by the addition of 10 µl of formamide and samples were separated through a 7 M urea/6% polyacrylamide gel. To identify depurinated nucleotides, dideoxynucleotide sequencing of pBH10fs HIV-1 plasmid was performed with the same reverse primer used for the primer extension assay. Radiolabeled cDNAs were visualized with a phosphor imager.
To compare the effects of PAP expression, relative to PAPx or empty vector, on AP-1 activity and transcription from the viral LTR in the absence of TAR, we performed a one-way ANOVA to compare means across all treatments, followed by Tukey's post hoc test to make pairwise comparisons between groups, for each experiment. P-values ≤0.05 were considered significant.
PAP decreases HIV-1 mRNA levels without causing cell death
To understand how PAP may inhibit HIV-1, cells were co-transfected with the pBH10fs HIV-1 proviral clone and FLAG-tagged pcPAP (wt) or pcPAPx. The latter encodes a catalytically inactive mutant of PAP  which serves as a negative control. Immunoblot analysis confirmed both PAP and PAPx expression in HEK 293T cells (Figure 1A). Expression of PAP reduced levels of HIV-1 RNAs ∼ 4-fold, compared with PAPx and empty vector (pcDNA3) controls (Figure 1B,C). No notable differences in cell viability were observed for the various transfections as monitored by an MTT conversion assay (Figure 1D). Therefore, the decrease in HIV-1 RNAs was not due to toxicity of PAP expression in cells. These data are consistent with our previous results showing PAP-dependent decrease in HIV-1 protein and virion production [8–10] in the absence of detectable cell toxicity.
PAP decreases HIV-1 mRNA levels without decreasing cell viability.
PAP does not alter HIV-1 mRNA stability
To identify whether the decreased levels of HIV-1 RNA were due to reduced synthesis or enhanced degradation, actinomycin D was used to inhibit total cellular transcription in cells co-transfected with the pBH10fs proviral clone, with or without pcPAP. Total cellular RNA was probed for the 3′ region of HIV-1 and quantified relative to GAPDH mRNA over a 24-h time period. Values were normalized to the zero time point. For all three size classes of mRNA, no major differences in stabilities of viral mRNAs were observed in cells transfected with pcPAP versus pcPAPx or pcDNA3 (Figure 2). Based on these data, it is unlikely that the decrease in HIV-1 mRNA levels observed with PAP expression was the result of enhanced degradation.
PAP does not alter HIV-1 mRNA stability.
PAP reduces Tat activity
To test the alternative hypothesis that PAP inhibited HIV-1 transcription, we performed a Tat functional assay, in which the level of CAT mRNA transcribed from the pHIVSCAT reporter is dependent on the activity of the Tat protein (Figure 3A; ). Cells were co-transfected with the pHIVSCAT reporter plasmid and pBH10fs proviral clone, with or without pcPAP, and the level of CAT transcript analyzed via Northern blot. As a positive control, cells transfected with a plasmid expressing Tat under the control of a cellular promoter (pEF-TAT) confirmed the responsiveness of the assay to Tat (Figure 3B). Cells transfected with pcPAP showed an ∼ 4-fold decrease in CAT mRNA, and thus Tat activity, when compared with cells transfected with pcDNA3 or pcPAPx (Figure 3B,C). We chose to work with HEK 293T cells because of their ease of transfection and expression of foreign genes; however, they are not naturally infected with HIV-1. Therefore, we also tested the effect of PAP on Tat activity in T-lymphocytic derived Jurkat cells. Similar to our observations with HEK 293T cells, PAP reduced Tat activity ∼3-fold, as measured by the CAT reporter assay (Figure 3B,C).
PAP inhibits Tat activity.
To confirm that the lack of Tat activity was responsible for the reduced HIV-1 RNA levels in cells expressing PAP, we performed a Tat rescue. Cells were co-transfected with the proviral clone, with or without pcPAP and increasing amounts of pEF-TAT. Transfection with 0.25 µg of pEF-TAT was sufficient to rescue ∼70% of viral RNA levels and excess of full rescue was seen with 1.0 µg of pEF-TAT (Figure 4A,B). Therefore, the addition of a plasmid encoding Tat was able to overcome the inhibitory effect of PAP and restore HIV-1 mRNA expression. Even though we were not able to detect Tat from the proviral clone despite several attempts with different antibodies, we did confirm an increase in Tat expression upon transfection of cells with pEF-TAT that correlated with increased HIV-1 mRNA levels (Figure 4A). Together these results indicate that pcPAP diminishes Tat activity and that this could contribute to the lower levels of viral transcription observed in HEK 293T cells transfected with the pBH10fs proviral clone.
Addition of pEF-TAT restores expression of HIV-1 mRNAs.
Primer extension of the Tat open-reading frame
Given that PAP is an RNA glycosidase that we have shown previously to remove bases from HIV-1 RNA , we searched for missing bases within Tat coding region that could explain the lack of Tat activity in cells expressing PAP. Cells were transfected with the proviral clone and pcPAP, pcPAPx or empty vector, and 2 kb RNAs were used as a template for primer extension with reverse primers specific to the Tat open-reading frame. A resulting cDNA fragment consistently terminated at A5272 (numbering according to BH10 HIV-1 genome) in samples from PAP-expressing cells (Figure 5A). This termination corresponds to a site that is unique to the Tat open-reading frame and does not overlap with any other functional HIV-1 ORF (Figure 5B). Tat expression in cells co-transfected with pEF-TAT and pcPAP was decreased relative to cells transfected with pEF-TAT alone (Figure 5C), and correlates with the termination of cDNA synthesis, which is consistent with depurination of the 2 kb RNA at this site.
PAP-specific termination of primer extension within the Tat open-reading frame of HIV-1 RNA.
PAP activates AP-1
Cellular MAPKs, such as ERK1/2 or JNK1, contribute to regulation of HIV-1 transcription, as the viral LTR contains binding sites for downstream effectors of these proteins [39,40]. We have shown that PAP activates both the ERK and JNK1 pathways [10,30]; therefore, we tested whether transfection with pcPAP would activate downstream transcription via AP-1, SP-1 or NF-κB, as all three have been shown to regulate HIV-1 transcription [16,31,41]. Reporter constructs containing the response elements for these transcription factors upstream of a luciferase gene were transfected into cells, with and without pcPAP. Northern blotting probed for luciferase mRNA and quantified levels relative to GAPDH mRNA. Cells co-transfected with the pAP1-Luc reporter and pcPAP showed modestly increased levels of AP-1 activity when compared with lanes transfected with pcPAPx (P = 0.011) or pcDNA3 (P = 0.028) (Figure 6A,B). This increase was reduced with the addition of SP600125 (JNK1 inhibitor), but not PD98059 (MEK1/2 inhibitor), indicating that PAP activated the AP-1 transcription factor via the JNK1 pathway. In contrast with AP-1, there was no detectable activation of NF-κB or SP-1 reporter constructs upon PAP expression (Figure 6C).
PAP activates AP-1 but not SP-1 or NF-κB.
Effect of PAP on transcription from a partial LTR
To determine whether the observed activation of AP-1 in cells expressing PAP affected HIV-1 promoter activity, we assessed whether PAP could enhance transcription from a partial HIV-1 LTR, containing only the region with host factor binding sites. An LTR-Luc reporter construct containing the U3 region of HIV-1 LTR was transfected into cells with pcPAP. The absence of the downstream R region, with its TAR element, rendered transcription of this construct reliant only on host factors, such as AP-1. A similar trend was observed with PAP expression modestly increasing the level of luciferase mRNA relative to PAPx or empty vector, though the increase was not significant (Figure 7A,B). This increase was not altered by the addition of PD98059 but was reduced with the JNK specific inhibitor SP600125. We conclude that even though PAP activated the JNK pathway to a limited extent through AP-1 (Figure 6), this activation of a host factor is not expected to significantly impact HIV-1 RNA levels. The primary effect of PAP on transcription is to reduce viral RNA levels by decreasing Tat.
Effect of PAP on transcription from a partial LTR.
In the present study, we investigated the mechanism behind the previously observed decrease in HIV-1 protein production in the presence of PAP. Specifically, we tested whether PAP affected transcription of viral mRNAs. We considered both the direct activity of PAP on the virus and its indirect influence via cellular signaling pathways. We show that the negative effect of PAP on HIV-1 transcription is primarily due to a reduction in Tat activity. Primer extension results are consistent with PAP depurination of the Tat open-reading frame, which inhibits its translation and subsequent activation of transcription at the viral LTR. As a result, total viral RNA and protein levels are reduced.
Tat is currently being considered in novel approaches to control HIV-1 levels by maintaining latency. Antiretroviral therapies significantly reduce virus replication and decrease viral titers to below detection. However, an interruption of the antiretroviral drugs results in rebound of virus levels because of reactivation of latent viral reservoirs. Though the control of latency is complicated and not fully understood, the phenomenon has been suggested as a possible form of functional cure . Specifically, deep latency could be achieved through inhibitors of transcription, thereby preventing virus replication in the absence of other antiretroviral therapies, despite integrated provirus in the genome. In this respect, Tat is an attractive target because it has no cellular homologs and is active in the early stages of infection. It was shown that constitutive expression of a mutant Tat in HIV-1 infected cells significantly decreased the level of viral transcription . The transdominant negative mutant Tat contains an amino acid substitution in its arginine-rich motif (TAR-binding domain). This mutant competes with wild-type Tat for interaction with P-TEFb and interferes with transactivation at the viral promoter. More recent data show that this mutant Tat also inhibits virus replication in a TAR-independent manner, by reducing association of RNA polymerase II with the viral LTR, in a mechanism that involves epigenetic changes to HIV-1 chromatin . Using this mutant to induce a form of deep latency would require gene therapy. Alternatively, didehydro-cortistatin A, an analog of a naturally occurring alkaloid, inhibits Tat by binding to its arginine-rich motif and has been suggested as an agent that could prevent reactivation of HIV-1 from latency . Triptolide, a diterpenoid of plant origin, specifically enhances Tat degradation and is currently in phase III trials to test its effect on the size of viral reservoirs in the presence of other antiretroviral therapies . Though we do not plan to test the effect of PAP on latency, our results consistent with Tat RNA depurination leading to significantly reduced activity levels, support ongoing initiatives to achieve long-term latency by controlling Tat in virus-infected individuals.
In addition to Tat, we also considered the potential effects of PAP on cellular factors that modulate transcription of HIV-1. The viral LTR contains binding sites for AP-1, SP-1 and NF-κB, which are downstream effectors of JNK and ERK pathways [16,39,40]. We have shown previously that PAP activates these pathways to a modest level, likely as a consequence of rRNA depurination. Damage to 28S rRNA causes a ribotoxic stress response, which signals from ribosomes via the MAPK cascade [47,48]. Specifically, double-stranded RNA-activated protein kinase (PKR), constitutively bound to ribosomes, is autophosphorylated upon changes in conformation of rRNA caused by some translation inhibitors [49,50]. Activated PKR initiates signaling through the MAPK, p38 and ERK pathways [51,52]. Though ribotoxic stressors often activate pro-apoptotic pathways, this is not the case for PAP . Only 17% of 28S rRNA is depurinated upon PAP expression in cell culture, and this level of ribosome damage is not sufficient to inhibit protein synthesis or cause cytotoxicity [8,29], as shown by the current cell viability assay. Regardless, we tested whether PAP affected transcription in a Tat-independent manner via these cellular factors. Though PAP modestly activated AP-1, this activation was not sufficient to significantly enhance transcription from an LTR lacking a downstream TAR element. Therefore, the primary effect of PAP is to inhibit HIV-1 transcription, through diminished Tat levels.
PAP reduced the accumulation of viral RNAs in the absence of cytotoxicity, which suggests either specificity by PAP for viral RNA over cellular RNAs, or an enhanced sensitivity of viruses to the effects of depurination. We have examined the entire length of GAPDH mRNA for PAP-specific pauses of the reverse transcriptase and found none , though this does not preclude the possibility that other mRNAs are depurinated by PAP. We are currently investigating what sequence or structure motifs are preferentially targeted by PAP, apart from the requirement for a purine and presumably single-stranded RNA; however, it is unlikely that a virus-specific sequence exists. We hypothesize that a purine within a structure similar to the GNRA tetraloop of the sarcin/ricin rRNA loop would be a substrate for PAP. Because incubation of PAP alone with HIV genomic RNA liberates adenines , no cofactor is required for PAP to bind RNA in vitro. This may not be the case in vivo, where RNAs are naturally associated with proteins that determine subcellular localization, and mediate translation and degradation of the RNAs. In the present study, we measured depurination consistent with PAP-specific pauses or stops of the reverse transcriptase. Though we did not distinguish between these two events, it may be possible to treat isolated RNA from cells with an agent such as aniline that would specifically cleave the RNA at abasic nucleotides. Subsequent Northern blotting with a Tat-specific probe may detect the cleaved piece of RNA. We also observed extension bands from RNA isolated from pcDNA cells in the absence of PAP, indicating that there may be regions of Tat RNA that are prone to spontaneous depurination, or regions of secondary structure that cause the reverse transcriptase to pause. We considered the possibility that these regions may be preferentially targeted by PAP, though we did not detect bands in the absence of PAP that were more intense in PAP-expressing cells.
We have also noted differences in depurination sites that appear to be dependent on nucleotide sequence between different proviral clones. For example, in our previous work we detected two PAP-specific termination sites within the Rev open-reading frame and showed that decreased Rev correlated with accumulation of 2 kb viral RNAs . However, we do not observe this accumulation in the present study. Rather, we detected termination of the reverse transcriptase at a single site within the Tat open-reading frame, which we correlate with decline of all viral RNA levels, as would be expected from diminished Tat activity. The two studies are distinguished by the fact that the previous work used the proviral HXB2 clone, whereas the current work used BH10. These clones differ by many nucleotides, several of which exist within the Tat and Rev open-reading frames (accession numbers K03455.1 and M15654.1). We suggest that these changes in sequence, or perhaps resulting structure of the RNA, are responsible for differences we observe in PAP-specific termination sites on the different RNA templates.
Depurinated RNAs cause translating ribosomes to stall at the missing base  and subsequent studies show premature release of the protein from ribosomes . Therefore, depurination of mRNAs negatively impacts protein accumulation. For HIV-1, depurination would also affect cDNA synthesis. The viral reverse transcriptase tends to synthesize cDNA across the lesion, most often incorporating a dATP when encountering a missing base, resulting in increased rates of mutation . We have thus far identified sites of potential depurination in the open-reading frames of vif and rev, in addition to primer extension results in this study [8,9]. Each of these regulatory proteins plays an important role in establishing infection and maintaining virus production. Their loss due to translation inhibition, in combination with mutations of the viral genome, contributes to the antiviral activity of PAP.
activator protein 1
Dubelco's Eagle Modified Medium
extracellular signal-regulated kinase
fetal bovine serum
glyceraldehyde 3-phospate dehydrogenase
human immunodeficiency virus
c-Jun NH2-terminal kinase 1 (JNK1)
5′-long terminal repeat
nuclear factor kappa-light-chain-enhancer of activated B-cells
pokeweed antiviral protein
RNA-activated protein kinase
- R region
specificity protein 1
transcription response element
transactivator of transcription
untranslated 3′ region
Experiments were conceived and designed by M.K. and K.A.H. Experiments were conducted by M.K. Data were analyzed by M.K. and K.A.H. The manuscript was written by M.K. and K.A.H.
This work was supported by a NSERC Discovery Grant to K.A.H. and a NSERC PGS-M Scholarship to M.K.
We are grateful to Dr C. Liang (McGill University) for pBH10fs; Dr B.M. Peterlin (University of California, San Francisco) for pHIVSCAT and pEF-TAT; Dr S. Watanabe (University of Tokyo) for pLTR-Luc (pHIV-BB Luc); Dr J.C. McDermott (York University) for pAP1-Luc and pcYFP. The monoclonal antibody for Tat was obtained through the NIH AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH (catalog no. 705) from Dr B. Cullen (Duke University).
The Authors declare that there are no competing interests associated with the manuscript.