Upregulation of c-Fos in Activated T Lymphoid and Monocytic Cells by Human Immunodeficiency Virus-1 Tat Protein

THE HUMAN immunodeficiency virus-type 1 (HIV-1) 86-101 amino acid Tat protein is encoded by two exons and comprises five distinct domains: namely the N-terminal, cysteine-rich, core, basic, and C-terminal sequences.¹ The cysteine-rich region is responsible for the formation of intramolecular disulphide bonds,2 whereas the basic region contains nuclear localization signals3 and the binding site for the transactivation response element (TAR).4 The transcriptional activation domain of Tat (amino acids 1 through 48) includes the amino-terminal region, the conserved cysteine-rich and core domains.5 Tat plays a pivotal role for HIV-1 gene expression and replication,1 by interacting with the TAR sequence, located at the 5′ end of all viral mRNAs. However, increasing experimental evidence suggests that Tat can also control viral replication via TAR-independent mechanisms.6,7 

Besides its transactivating function on viral gene expression, Tat affects the survival, growth, and function of a variety of infected and uninfected cells.8-17 These biological effects are at least partially mediated by the upregulation or downregulation of genes involved in the control of cell survival/proliferation^10,18-22 or coding for cytokines.23-29 

Tat can be actively secreted by HIV-1 infected and tat-transfected cells,30 and can easily enter different cell types both in vitro31,32 and in vivo,33 contributing to the transcellular transactivation of HIV-1 long terminal repeat (LTR) promoter in latently infected cells.34-36 Moreover, the addition of an anti-Tat neutralizing antibody in culture blocks an autocrine/paracrine loop of extracellular Tat in tat-transfected cell lines,37 and reduces the levels of gag p24 in the culture supernatant of HIV-1 infected lymphoid T-cell lines and primary peripheral blood mononuclear cells (PBMCs).38,39 These findings add further complexity to the mechanisms of action of Tat protein, strongly suggesting that extracellular Tat may also significantly contribute to the pathogenesis of HIV-1 disease.

It is noteworthy that c-Fos–responsive sequence motifs have been identified in HIV-1 LTR,1 as well as in many promoters of genes encoding for cytokines transactivated by Tat.26,28 In this study, we investigated whether Tat is able to modulate c-Fos protein expression as a part of its mechanism of action. To do this, we selected as model systems the Jurkat human CD4⁺ T and the U937 monocytic cell lines, which are representative of the main targets of HIV-1 infection in vivo.40 Induction of the c-fos gene promoter, c-fos mRNA expression and c-Fos protein synthesis were evaluated in both cell lines, either (1) transiently transfected with tat cDNA, or (2) treated with recombinant Tat protein, whether alone or in combination with various stimuli known to activate c-fos expression. The effect of extracellular Tat on endogenous c-fos gene expression was also evaluated on primary PBMCs. Finally, to evaluate the physiological relevance of these results to HIV-1 infection, we performed experiments in which antisense and sense c-fos oligonucleotides were added to Jurkat cells either transfected with tat cDNA plus constructs containing the LTR region of HIV-1 genome fused with a chloramphenicol acetyltransferase (CAT), or acutely infected with the IIIB strain of HIV-1.

Cell lines and PBMCs.Jurkat CD4⁺ lymphoblastoid T and U937 monocytic cell lines were obtained from the American Type Culture Collection (Rockville, MD) and maintained in RPMI 1640 (GIBCO Laboratories, Grand Island, NY) supplemented with 10% fetal calf serum (FCS; GIBCO) at an optimal cell density of 0.3 to 1.5 × 10⁶ cells/mL.

PBMCs were taken from 11 normal blood donors, who gave their informed consent to this research according to the Helsinki declaration of 1975. Briefly, heparinized (20 U/mL) blood samples were diluted 1:3 with RPMI 1640, layered over Ficoll Histopaque (d = 1.077 g/mL, Pharmacia, Uppsala, Sweden) and centrifuged at 1,500 rpm for 30 minutes. Light-density mononuclear cells were collected, washed twice, counted, and treated with various agonists.

Plasmids, oligonucleotides, and viral proteins.Plasmids pRPneo-C/SL3/tat/S, pRPneo-C/SL3/tat ²²/S, pRPneo-C/SL3/tat/AS containing the HIV-1 tat cDNA, respectively in sense (tat/S), carrying a mutation in the aminoacid cys22 - gly²² (tat²²/S) or in anti-sense (tat/AS) orientation under the control of the LTR region of the SL3 murine leukemia virus were derived from the backbone plasmid, as previously described.29 

Antisense and sense c-fos oligonucleotides were represented by 18 residues starting from and including the start codon for c-fos: gene: ATG ATG TTC TCG GGT TTC AA (sense c-fos, FOS-OLIGO); TTG AAG CCC GAG AAC ATC AT (antisense c-fos, SOF-OLIGO).

The c-fos/CAT plasmid constructs (FC2, FC3, FC4, FC8) containing the bacterial CAT gene under the control of various c-fos promoter sequences from position −1450 (FC2), −711 (FC3), −404 (FC4), −220 (FC8) to position +42 were described by Deschamps et al.41 The wild-type serum responsive element (SRE)/CAT plasmid contained c-fos promoter sequences from −317 to −288 base pairs (bp). The mutated SRE/CAT plasmid contained two point mutations (CC-GG) in positions −309/−308. All the c-fos/CAT plasmids were a generous gift of Dr Sassone-Corsi (IGBMC, Illkirch, France). The β-actin/CAT and the HIV-1 LTR were described previously.29 

Recombinant Tat and p24 proteins, produced in Escherichia coli (ABT, Cambridge, MA), were found free of endotoxin by a Limulus Amebocyte Lysate test (Whittakers, Walkersville, MA).

Transfections.Transient cotransfection experiments were carried out using the DEAE-dextran method as previously described by Zauli et al.37 Briefly, 10 × 10⁶ of exponentially growing Jurkat or U937 cells were maintained in RPMI 1640 plus 1% FCS for 24 hours and then transfected in 3 mL volume with 15 μg of each plasmid DNA in 500 μg DEAE-dextran for 90 minutes. After transfection, cells were seeded again in 10 mL of RPMI 1640 plus 1% FCS for 12 hours. Jurkat cells were supplemented with 15% FCS or 5 μg/mL phytohemagglutinin (PHA; Difco, Detroit, MI) plus 10⁻⁷ mol/L phorbol myristate acetate (PMA, Sigma Chemicals, St Louis, MO), while U937 cells were supplemented with 15% FCS or 10⁻⁷ mol/L PMA for an additional 24 hours. Cells were then lysed and the clarified lysates were assayed for CAT activity using volumes of extract corresponding to equal amounts of proteins as determined by the Bio-Rad protein assay system (Bio-Rad Laboratories, Richmond, CA).

In other experiments, 10 × 10⁶ Jurkat cells were transiently transfected with 15 μg of tat/S plus 5 μg of LTR-CAT plasmids plus either sense c-fos (FOS-OLIGO, 25 μmol/L final concentration) or antisense c-fos (SOF-OLIGO, 25 μmol/L final concentration) oligonucleotides, then treated as described above.

To assay the effect of extracellular Tat on the c-fos promoter activity, 5 × 10⁶ Jurkat or U937 cells were transiently transfected with the FC3 vector as described above. Twelve hours after transfection, 1 μg of recombinant Tat or recombinant p24 plus 15% FCS was added to 5 mL of medium, and incubation was allowed to proceed for an additional 24 hours. In blocking experiments, before its addition to the culture medium, recombinant Tat was preincubated for 2 hours at 37°C with 20 μg/mL of an anti-Tat monoclonal antibody (MoAb, ABT), which recognizes the N-terminus of Tat or with 20 μg/mL of an anti-p24 MoAb (ABT).

RNA analysis.To study the effect of extracellular Tat on endogenous c-fos mRNA expression, 30 × 10⁶ Jurkat were cultured in RPMI plus 1% FCS for 24 hours and then supplemented with 15% FCS or PMA plus or minus 5 μg of recombinant Tat or p24 in 5 mL of medium for 30 minutes. Total RNA was extracted, using the RNAzol kit (Biotecx, Galveston, TX) according to the manufacturer's instructions. 5 μg of poly(A⁺) RNA were then purified from each sample by using the poly(A⁺) RNA purification kit (Qiagen, Hilden, Germany) and processed for Northern blot as previously described.29,c-fos and control probes were represented by the Xho I/Nco I 2,700 bp c-fos fragment42 and by the Xba I/Pst I 780 bp fragment of glyceraldeyde-3-phosphate dehydrogenase gene (GAPDH),29 respectively. Both probes were digoxigenin-labeled using a random primer labeling kit (Boehringer Mannheim, Postfack, Germany), according to the manufacturer's instructions.

For RNase protection analysis, 30 μg of total RNA were processed using a RNase protection kit (Boehringer Mannheim) according to the manufacturer's instructions. A 340 nucleotide complementary digoxigenin-labeled RNA from FC3 was used as a probe for c-fos CAT mRNA.43 

Analysis of c-Fos protein by flow cytometry and confocal microscopy.5 × 10⁶ Jurkat cells or PBMC were cultured in RPMI plus 1% FCS for 24 hours and then supplemented with 15% FCS alone or in combination with Tat or p24 for 45 minutes. Cells were then fixed and permeabilized with cold 70% ethanol for 30 minutes at 4°C, washed twice with phosphate-buffered saline (PBS), and treated for 90 minutes at 37°C with a 1:100 dilution of anti-c-Fos rabbit polyclonal serum (Santa Cruz Biotech, Santa Cruz, CA) in PBS plus 1% bovine serum albumin (BSA). 2.5% of nonspecific blocking guinea pig serum was present during each antibody incubation. Cells were washed twice and incubated for 30 minutes at 37°C with a 1:100 dilution of goat anti-rabbit serum conjugated to fluorescein (GAR-FITC, Sigma) in PBS plus 1% BSA. Negative controls consisted of normal rabbit serum followed by identical second layer labeling as above. The presence and subcellular distribution of c-Fos protein was investigated by FACScan using the Lysis II software (Becton Dickinson, San José, CA). For confocal microscopy, the cells were first spun on coverslips, then fixed and stained with the anti-Fos antibody as described above. The cells were analyzed by means of a MCR-1000 confocal microscopy (Bio-Rad Microscience Ltd, Hemel, Hempstead, UK) equipped with a krypton/argon ion laser emitting at 488 nm. The signal was achieved through an Epidetector filter (passing band 522/35 nm), analyzed by CoMOS software (Bio-Rad) and printed on a Ektachrome 64T-Kodak film (Eastman Kodak, Rochester, NY) by a Focus Imagecorder Plus (Focus Graphics, Inc, Foster City, CA).

Acute HIV-1 infection of Jurkat cells and primary PBMC.Virus stock was represented by the supernatant of H9 lymphoblastoid T cells cultured at optimal cell density (0.5 to 1.5 × 10⁶ cells/mL) and obtained 14 days after infection with HIV-1 (strain IIIB). It contained a reverse transcriptase activity of 1.5 × 10⁶ cpm/mL with an infectivity of 3 × 10⁶ TCID50 (tissue culture infectious dose 50%) equivalents for lymphocytes, determined as previously described.44 One milliliter of HIV-1 was added to either Jurkat cells or primary PBMC for 2 hours at 37°C. Control (mock-treated) cultures were run in parallel by challenging Jurkat cells or PBMC with 1 mL of the supernatant of uninfected H9 lymphoblastoid T cells cultured under optimal conditions. After virus absorption, Jurkat and PBMC were washed and adjusted to the concentration of 1 × 10⁶/mL and seeded in 1 mL of RPMI containing 15% FCS. SOF-OLIGO and FOS-OLIGO were added at a final concentration of 25 μmol/L at the beginning of culture. Every other day, cultures were demi-populated and supplemented with 0.5 mL of fresh medium containing 25 μmol/L of oligonucleotides. The harvested medium was centrifuged and the supernatants were monitored for the presence of gag p24 antigen by means of a commercial immunoplate assay (Du Pont de Nemours, Les Ulis, France), while cell pellet was resuspended and viable cells were counted by Trypan blue dye exclusion.

Statistical analysis.The results are expressed as means ± standard deviations (SDs) of at least three separate experiments performed in duplicate. Two-tailed Student's test for unpaired data was used for statistical comparison.

HIV-1 Tat specifically upregulates c-fos promoter activity in serum-stimulated Jurkat and U937 cells.Since Tat protein is predominantly produced during the early phases of de novo HIV-1 infection or during reactivation from viral latency,1 the effect of Tat on the c-fos promoter expression was first examined in transient transfection experiments. Jurkat and U937 cell lines were cotransfected with the FC3 plasmid, where CAT gene expression is driven by the c-fos promoter, together with an HIV-1 tat expression plasmid (tat/S). Control samples were represented by cells transfected with the FC3 plasmid alone as well as cells cotransfected with FC3 and a tat antisense plasmid (tat/AS), tat carrying a mutation in the amino acid cys22 - gly²² (tat²²/S) or FC3 and the pRPneo-SL3 backbone vector.

When Jurkat or U937 cells were cultured in the presence of low serum concentrations (RPMI + 1% FCS), Tat was unable to upregulate CAT activity over the levels observed in the presence of control vectors (data not shown). The effect of Tat expression was then evaluated in the presence of additional stimuli: 15% FCS or PHA (5 μg/mL) + 10⁻⁷ mol/L PMA for Jurkat, 15% FCS, or 10⁻⁷ mol/L PMA for U937 cells. Under these conditions, CAT activity showed a fivefold to sevenfold increase (P < .01) in both Jurkat (Fig 1A) and U937 (Fig 1B) cells cotransfected with tat/S as compared to cells cotransfected with tat/AS, tat²²/S, pRPneo-SL3 or cells transfected with FC3 alone. The enhancing effect of tat/S on FC3 activity was promoter specific, as Tat did not significantly upregulate CAT expression when the reporter CAT gene was under the transcriptional control of the β-actin promoter (β-actin/CAT) (data not shown).

To provide an additional proof that the increased c-fos promoter activity was specifically mediated by Tat, transient cotransfection assays were performed with tat/S, tat/AS, and FC3 vectors in both Jurkat and U937 cell lines cultured in the presence of 15% FCS. tat/AS significantly (P < .01) reduced the activation of FC3 mediated by tat/S in both Jurkat (Fig 2A) and U937 (Fig 2B) cells from fivefold-sevenfold to twofold-threefold. Moreover, to exclude the possibility that Tat activates the transcription of the CAT gene in an unusual start site, RNase protection analysis was performed using an RNA probe complementary to the CAT gene (Fig 3). After RNase digestion, the protected fragment resulted of 310 nucleotides as expected. An increased level of correctly initiated RNA was observed in Jurkat cells cotransfected with FC3 and tat/S as compared to cells cotransfected with FC3 and tat/AS.

The −317/−288 serum responsive element of the c-fos promoter is required for Tat transactivation in activated Jurkat cells.In the next series of cotransfection assays we used different 5′ deletion mutants of the c-fos promoter linked to the CAT gene (Fig 4A and B) together with tat expressing vectors, to determine which sequences of the c-fos promoter were required for Tat function. Figure 4C shows the CAT analysis in Jurkat cells cultured in 15% FCS after cotransfection with various c-fos promoter/CAT deletion mutants plus tat/S or tat/AS. In the presence of FC2, FC3, and FC4 plasmids, tat was able to significantly enhance CAT activity with respect to tat/AS, whereas tat/S was unable to activate CAT expression from the FC8 mutant over the level achieved in the presence of tat/AS. Therefore, the majority of the Tat-induced activation was mediated by c-fos promoter sequences located between positions −404 and −220 bp.

In this region, different cis-acting regulatory elements exist (Fig 4A) which may be responsible for the transactivation activity mediated by Tat. In fact, this 184 bp region contains multiple promoter elements including the dyad symmetry element that encompasses the SRE, an AP-1 site, the platelet-derived growth factor responsive sequence, a CRE element and an E1A-inducible region.45 Since it has previously been shown that the SRE is essential for the transactivation of c-fos mediated by transactivating Tax protein of human T-cell leukemia virus type 1 (HTLV-1),43 in the next series of experiments, we cotransfected tat/S and tat/AS with plasmids containing the wild-type SRE in front of CAT (Fig 4B and C). Tat significantly activated the SRE/CAT plasmid (a threefold increase of CAT activity with respect to tat/AS), although to a lesser extent than FC3 (a fivefold to sevenfold increase). On the other hand, tat/S was unable to stimulate CAT activity when an SRE mutated was used (data not shown). These data suggest that a functional wild-type SRE was involved in the Tat-mediated transactivation of the c-fos promoter.

Extracellular recombinant Tat is able to upregulate c-fos promoter activity and induce nuclear c-Fos accumulation in both Jurkat cells and primary PBMC.It has been well-documented that Tat can be secreted in a biologically active form either by acutely infected cells or by tat-transfected cell lines, and be taken up by neighboring uninfected cells, thus contributing to the transactivation of the HIV-1 LTR and heterologous cellular gene promoters. To test the effect of extracellular Tat protein on the activity of the c-fos promoter, recombinant Tat or p24 were added to Jurkat and U937 cells 12-hours posttransfection with FC3, and cultures were allowed to procede for an additional 24 hours.

As shown in Fig 5, CAT activity was significantly (P < .01) enhanced by the addition of recombinant Tat but not by recombinant p24. Moreover, Tat activity was specifically neutralized by an anti-Tat MoAb. These results indicate that also in its extracellular form Tat was able to upregulate the activity of the c-fos promoter.

An episomial c-fos promoter/CAT fusion construct differs from the endogenous c-fos gene with respect to the conformation and sequences of the flanking DNA. Therefore, in the next group of experiments we evaluated the effect of extracellular Tat on endogenous c-fos gene expression in Jurkat cells.

Northern blot analysis showed that the endogenous c-fos mRNA expression was markedly higher in Jurkat treated with 15% FCS plus recombinant Tat as compared to cells treated with 15% FCS alone or 15% FCS plus recombinant p24 (Fig 6). The enhancing effect mediated by Tat on c-fos mRNA was specifically blocked by an anti-Tat neutralizing antibody.

In parallel experiments, the amount of c-Fos protein was then evaluated by flow cytometry (Fig 7A through N) in both serum-starved Jurkat and freshly isolated PBMC treated with the same agonists described above for the c-fos mRNA studies. PBMC were examined taking into account that primary T cells and monocytes differ from tumor cell lines in growth control and requirement for growth factors. The flow cytometry approach was chosen, since it has been previously demonstrated that it represents the most sensitive and reliable method for the detection of nuclear oncoproteins, including c-Fos, in hematopoietic cells.46 c-Fos protein was first examined in 24-hour serum–starved Jurkat cells (Fig 7A and D) and freshly isolated PBMC (Fig 7G and L). The level of c-Fos protein significantly increased in both Jurkat and PBMC (Fig 7B and H) after a 45-minute treatment with 15% FCS or PHA plus PMA (Fig 7E and M) and showed a further increase when recombinant Tat was added to 15% FCS or PHA plus PMA (Fig 7C, F, I, and N).

The presence of c-Fos protein was also analyzed in Jurkat cells by indirect immunofluorescence and confocal microscopy (Fig 8A, B, and C). While Jurkat showed weak or absent staining after 24 hours of serum starvation (Fig 8A), the addition of 15% FCS for 45 minutes induced an intense green-yellow positive staining (Fig 8B), which further increased in the presence of Tat (Fig 8C).

Addition of antisense c-fos oligonucleotides in culture diminishes the Tat-mediated transactivation activity on HIV-1 LTR and the gag p24 levels in the culture supernatants of acutely IIIB infected Jurkat cells and PBMC.The experiments described thus far suggested that Tat protein was able to upregulate c-Fos nuclear protein in both T lymphoid and monocytic cells. If this elevation in c-Fos protein is relevant to the ability of Tat to transactivate HIV-1 LTR, preventing this increase in c-fos expression should diminish the expression of HIV-1 LTR and be detrimental for a productive viral infection. Such an approach has been used successfully to analyze the role of c-Fos in regulating the expression of stromelysin by epidermal growth factor,47 and of collagenase by integrin-derived signals.48 Accordingly, we used synthetic DNA oligonucleotides that were complementary to c-fos mRNA sequences (antisense c-fos oligonucleotides, SOF-OLIGO) or identical to c-fos mRNA sequences (sense c-fos oligonucleotides, FOS-OLIGO), introducing them by along with tat/S and an LTR-CAT construct.

When cultures were kept in 15% FCS for the last 24 hours, transfection of Jurkat with SOF-OLIGO but not with FOS-OLIGO markedly decreased the expression of HIV-1 LTR in the presence of tat/S (Fig 9). On the other hand, no significant differences in the levels of LTR-CAT activity were observed when cultures were kept for 36 hours in l% FCS after transfection (data not shown). These results suggested that c-Fos protein is involved in the Tat-mediated regulation of HIV-1 LTR. On the other hand, activation of LTR-CAT in the absence of tat expression was very low and did not allow to establish whether c-Fos was also involved in the basal HIV-1 LTR-CAT transcription.

We then analyzed the levels of gag p24 antigen in the culture supernatants of Jurkat (Fig 10A) or PBMC (Fig 10B) acutely infected with high doses of infectious HIV (IIIB) and continously supplemented with SOF- and FOS-OLIGO. The addition of SOF-OLIGO but not FOS-OLIGO to cultures caused a specific decrease of the gag 24 levels evaluated over a 15-day culture period in either Jurkat cells or PBMC. This effect was not due to a toxic effect of SOF-OLIGO, since the number of viable cells did not significantly change in Jurkat cells that were left untreated with respect to cells treated with SOF- or FOS-OLIGO (data not shown).

The c-Fos protein is a component of the AP-1 transcription factor complex and is a member of a multigene family, including the fos-related (fos B, fra-1, fra-2) and jun-related (c-jun, jun B, jun D) genes. The AP-1 binding site is recognized by dimeric protein complexes composed of Jun homodimers or Fos/Jun heterodimers. The Fos/Jun heterodimer complex shows greater affinity than Jun/Jun homodimer for the AP-1 binding site.49-51 Moreover, c-Fos protein can also heterodimerize with some components of the NF-kB⁵² or ATF/CREB53 families, thus allowing a cross-talk between different transduction pathways, and providing an important route for different intracellular signals to be integrated at the level of transcriptional control.

Our data indicate that the transactivator Tat protein of HIV-1 upregulates c-fos gene expression in both activated Jurkat and U937 cell lines, as well as in freshly isolated activated PBMC. The effects of Tat on promoter activity, mRNA expression and protein synthesis suggested that c-fos upregulation was primarily caused at the transcriptional level. Moreover, the structural integrity of the activation domain of Tat was required for the promotion of the c-fos promoter in both Jurkat and U937 cells, since a single mutation in the aminoacid cys²² was sufficient to completely block the upregulation of c-fos expression. Since the mutation in cys22 results in the complete loss of the transactivating activity of Tat on HIV-1 LTR,44 our findings suggest that Tat-mediated activation of the c-fos promoter and the HIV-1 LTR are likely dependent on the same Tat function. According to previous findings on the transactivation of the IL-2 gene mediated by Tat,28 we found out that the transient expression of tat was unable in itself to enhance c-fos expression in either tumoral or primary unstimulated cells; however, Tat significantly improved the effect of other stimuli such as high serum concentrations.

Experiments performed with c-fos promoter deletion mutants indicated that the responsive region required for Tat activity was set between −404 and −220 bp and that the SRE was critical for Tat transactivation. Interestingly, the c-fos activation mediated by Tat revealed both clear differences and definite similarities with the c-fos transactivation induced by the transactivating Tax protein of HTLV-1.43 Both transactivators synergized with high serum concentrations, PHA or PMA in stimulating c-fos gene expression. Moreover, the analysis of deletion mutants underlined the central role of SRE in the transactivation activity of both Tat and Tax. However, Tax is a stronger inducer of c-fos expression, as it is able to transactivate c-fos promoter even in the presence of a low serum concentration.

The mechanism by which Tat affects c-fos gene expression remains to be fully elucidated. Bulge (AUCUG) and stem-loop (CUGGGA) structures similar to the Tat-binding site on HIV-1 LTR RNA have been identified downstream of the transcriptional start site of the tumor necrosis factor-β⁵⁴ and JC virus⁵⁵ promoters, and have been implicated in the Tat-induced activation of these genes. However, sequence comparison of the c-fos mRNA and TAR revealed no similarities between the TAR region of LTR and c-fos mRNA. Similarly, other authors failed show the presence of stem-loop sequences in different cellular genes activated by Tat.^{26, 28} Therefore, the transactivation of c-fos gene expression may be the consequence of a TAR-independent mechanism of transcriptional activation mediated by Tat. Moreover, the ability of Tat to enhance c-fos transcription only when additional stimuli are present, suggests a primary role in elongation and transcription processivity rather than in the transcription initiation of c-fos gene, as previously described for the HIV-1 LTR.1 

It is remarkable that even in its extracellular form Tat was able to upregulate c-fos expression in activated primary and tumoral cells. The effect of extracellular Tat may imply the uptake and nuclear localization of Tat protein,33 which may act directly on c-fos promoter. Alternatively, extracellular Tat might trigger an intracellular signal transduction pathway after specific membrane interactions with different types of integrin receptors present on those cells sensitive to its biological activity.38,56-58 A similar mechanism has been proposed to explain the Tat-induced proliferation of Kaposi's sarcoma spindle cells and normal endothelial cells preactivated by cytokines.59 If this model is correct, extracellular Tat could mimic the effect of extracellular matrix proteins activating c-fos expression through an intracellular signal cascade.60 

Our data indicate that c-Fos induction by Tat may have a role in the activation of HIV-1 LTR. In fact, SOF-OLIGO significantly inhibited CAT activity in both Jurkat and U937 cells transiently transfected with tat/S and an LTR-CAT reporter plasmid, and were able to specifically decrease the gag p24 levels in the culture supernatants of either Jurkat cells or primary PBMC acutely infected with HIV-1.

The upregulation of c-Fos induced by Tat may influence HIV-1 genome expression and HIV-1 replication in several ways. Firstly, the activation of c-Fos contributes to cell cycle progression of hematopoietic cells, acting throughout all the phases of the cell cycle.61 In this respect, it has been established that HIV-1 infects more efficiently activated cells than the quiescent ones.41 Secondly, transcriptional studies of the HIV-1 genome have shown that the HIV-1 LTR promoter can be subdivided into three functional domains: a basal core promoter, which is made up by the transcriptional start site, a TATA box, and three tandem Sp1 binding sites (GC-boxes); an upstream enhancer element containing two adjacent binding sites for nuclear factor κB (NFκB); a far upstream modulatory region.62 The HIV-1 LTR promoter also contains AP-1 motifs both upstream1 and downstream63 of the transcriptional start site.

These AP-1 sites can function independently of or in concert with the NFκB-binding enhancer to activate HIV-1 transcription. Moreover, a functional and physical interplay of the AP-1 and NF-kB families of transcription factors has recently been described.52 

In conclusion, we have clearly shown that HIV-1 transactivating Tat protein upregulates the expression of c-fos gene and the enhanced c-Fos protein production may represent an important intermediate step for the Tat-dependent HIV-1 genome transactivation. In the light of the positive synergy between NF-κB p65 and c-Fos, the ability of Tat to stimulate both factors has important implications for the promotion of viral replication, and, possibly, for the control of both immune and inflammatory responses through the production of various cytokines. In an evolutionary perspective, it may provide the virus with a mechanism to broaden its LTR response to extracellular stimuli and activate viral gene transcription under a wide variety of cellular conditions. Functional redundancy is a common feature of viral as well as cellular regulatory regions and in this case it may represent a mechanism to increase virus host range and virus replication potentiality.

We are grateful to Dr Paolo Sassone-Corsi for providing all the c-fos/CAT plasmids and to Dr Enzo Lalli for helpfull discussion. This research was supported by “AIDS project” of the Italian Ministry of Health and MURST 40 and 60%.