HIV-1 Tat induces the expression of the interleukin-6 (IL6) gene by binding to the IL6 leader RNA and by interacting with CAAT enhancer-binding protein beta (NF-IL6) transcription factors.

Human immunodeficiency virus type 1 (HIV-1) infection is associated with severe psoriasis, B cell lymphoma, and Kaposi's sarcoma. A deregulated production of interleukin-6 (IL6) has been implicated in the pathogenesis of these diseases. The molecular mechanisms underlying the abnormal IL6 secretion of HIV-1-infected cells may include transactivation of the IL6 gene by HIV-1. Here we report the molecular mechanisms of Tat activity on the expression of the IL6 gene. By using 5' deletion mutants of pIL6Pr-CAT and using IL6:HIV-1-LTR hybrid constructs where discrete regions of the IL6 promoter replaced the TAR sequence in HIV-1 LTR, we identified a short sequence of the 5'-untranslated region of the IL6 mRNA that is required for Tat to trans-activate the IL6 promoter. This sequence acquires a stem-loop structure and includes a UCU sequence that binds to Tat and is necessary for full trans-activation. In addition, we provide the evidence that Tat can function by enhancing the CAAT enhancer-binding protein (C/EBP) DNA binding activity and is able to complex with in vitro translated C/EBPbeta, which is a major mediator of IL6 promoter function. By using the yeast two-hybrid system and immunoprecipitation, we observed that the interaction of Tat with C/EBP proteins also occurred in vivo. The data are consistent with the possibility that Tat may function on heterologous genes by interacting with RNA structures possibly present in a large number of cellular and viral genes. In addition, Tat may function by protein-protein interactions, leading to the generation of heterodimers with specific transcription factors.

Human immunodeficiency virus type 1 (HIV-1) infection is associated with severe psoriasis, B cell lymphoma, and Kaposi's sarcoma. A deregulated production of interleukin-6 (IL6) has been implicated in the pathogenesis of these diseases. The molecular mechanisms underlying the abnormal IL6 secretion of HIV-1-infected cells may include transactivation of the IL6 gene by HIV-1. Here we report the molecular mechanisms of Tat activity on the expression of the IL6 gene. By using 5 deletion mutants of pIL6Pr-CAT and using IL6:HIV-1-LTR hybrid constructs where discrete regions of the IL6 promoter replaced the TAR sequence in HIV-1 LTR, we identified a short sequence of the 5-untranslated region of the IL6 mRNA that is required for Tat to trans-activate the IL6 promoter. This sequence acquires a stemloop structure and includes a UCU sequence that binds to Tat and is necessary for full trans-activation. In addition, we provide the evidence that Tat can function by enhancing the CAAT enhancer-binding protein (C/EBP) DNA binding activity and is able to complex with in vitro translated C/EBP␤, which is a major mediator of IL6 promoter function. By using the yeast two-hybrid system and immunoprecipitation, we observed that the interaction of Tat with C/EBP proteins also occurred in vivo. The data are consistent with the possibility that Tat may function on heterologous genes by interacting with RNA structures possibly present in a large number of cellular and viral genes. In addition, Tat may function by protein-protein interactions, leading to the generation of heterodimers with specific transcription factors.
Human immunodeficiency virus type 1 (HIV-1) 1 is the etiologic agent for acquired immunodeficiency syndrome (AIDS) and causes various clinical and immunological abnormalities, including activation of polyclonal B cells that manifests as hypergammaglobulinemia and autoantibody production, lymphadenopathy, Kaposi's sarcoma, and lymphoma of the B cell phenotype (1)(2)(3). Studies on small cohorts of subjects who were exposed to HIV-1 and did not develop HIV-1 infection and individuals who harbored HIV-1 but remained disease-free for long periods (4,5) strongly suggest that the development of AIDS may depend on a dynamic interplay between viral and host cellular gene products. Accordingly, in HIV-1-infected subjects there is a deregulated production of cytokines, including the proinflammatory interleukin-6 (IL6) (6), which affects the growth and differentiation of lymphoid and mesenchymal cells (7) and may contribute to the development of the clinical features of AIDS. Accordingly, IL6 gene transcription is induced in cells infected by HIV-1 (8), and increased levels of IL6 have been reported in serum and cerebral spinal fluid of HIV-1infected patients (9).
The Tat protein of HIV-1 is required for efficient viral gene expression (10 -15). Tat increases the initiation of transcription from the HIV-1 LTR (14) and affects RNA processing and utilization by interacting with a transactivating responsive element (TAR) located between nucleotides ϩ1 and ϩ 44 with respect to the initiation site (ϩ1) of viral transcription (16,17). TAR contains a 6-nucleotide loop and a 3-nucleotide pyrimidine bulge that are essential for Tat activity (18 -21). Tat binds to the bulge and appears to require cellular factors binding to the loop sequence to efficiently transactivate the HIV-1 LTR (22)(23)(24). In addition, Tat interacts with upstream regulatory DNA sequences circumscribed within the NF-B/Sp1 sites of the HIV-1 promoter (25) and with host cell proteins (12,24). The 86-amino acid-long Tat contains a highly conserved cysteinerich region, which mediates the formation of metal-linked dimers in vitro and is essential for Tat function (16 -18). A conserved basic region with 6 arginines and 2 lysines in nine residues, stretching from amino acid 47 to 58, is crucial for nuclear localization, mediates the specific binding of Tat to TAR RNA, and is required for the full activity of Tat (26 -29).
In addition to its role in HIV-1 transcription, Tat may participate in the development of AIDS by modulating the expression of heterologous genes. In support of this possibility, Tat has been shown to increase the expression of cellular genes, such as the IL6 (30) and tumor necrosis factor-␤ genes (31,32), and to activate the life cycle of some AIDS-associated viruses (33). The mechanisms of the Tat-mediated activation of non-HIV-1 genes are obscure. Here, we describe the mechanisms for Tat-mediated induction of the IL6 gene expression. We find that Tat is tethered to the IL6 transcription start site by specific binding to a UCU sequence present in the stem-loop structure of IL6 leader RNA. Tat physically interacts with C/EBP␤ and increases selectively the nuclear pool of C/EBP factors binding to the C/EBP cis sequence in the IL6 promoter. This interaction was confirmed to occur in vivo by immunoprecipitation and by using the yeast two-hybrid system. (34), a HIV-LTR-CAT plasmid was obtained from A. Rabson (MBCL, Piscataway, NJ). The 5Ј deletion mutants of IL6 promoter, pIL6(Ϫ596/ϩ15)-CAT, pIL6(Ϫ225/ ϩ15), and pIL6(Ϫ112/ϩ15)-CAT plasmids, were generated as reported (30). To generate the HIV-1-LTR:(Ϫ112/ϩ15) IL6 promoter fusion plasmid, the TAR-deleted EcoRI-BglII fragment of pILIC-CAT was isolated, filled in, and inserted at the SstI site (filled) of pIL6(Ϫ112/ϩ15). The resulting p⌬ILIC:IL6(Ϫ112/ϩ15)-CAT plasmid carries the IL6 promoter region from Ϫ112 to ϩ15 that substitutes for TAR. To generate p⌬ILIC:IL6(Ϫ112/Ϫ67)-CAT plasmid, the EcoRI fragment of p⌬ILIC: IL6(Ϫ112/ϩ15)-CAT was filled in and digested with SspI. The EcoRI-SspI fragment, containing the TAR-deleted LTR fused to the Ϫ112/Ϫ67 region of IL6 promoter, was cloned in pEMBL-CAT digested with BamHI-HindIII (filled). The SspI-EcoRI fragment, formed by the Ϫ67/ ϩ15 region of IL6 promoter fused to a part of the cat gene, was recovered and HindIII-digested. The Ϫ67/ϩ15 region of the IL6 promoter, the SspI-HindIII fragment, was recovered and cloned in p⌬ILIC: IL6(Ϫ112/ϩ15)-CAT from which the Ϫ112/ϩ15 IL6 fragment was removed by KpnI-HindIII digestion. pIL6(Ϫ596/ϩ15) mutants were produced with the Transformer TM site-directed mutagenesis kit, as instructed by manufacturer (CLONTECH Laboratories, Inc., Palo Alto, CA), with minor modifications. In fact, one oligonucleotide was used to introduce the desired mutations in IL6 promoter and to create the site for PstI. The following oligonucleotides were used (the mutated bases are underlined): 5Ј-CTGAGGCTCATTGGGCCCTCGACCTGCAGGCA-T-3Ј for pIL6(Ϫ596/ϩ15) MI-CAT (bulge mutant); 5Ј-ATTCTGCCCTA-GGCCCGCAGGCATGC-3Ј for pIL6(Ϫ596/ϩ15) M2-CAT (stem mutant); and 5Ј-CTGAGGCTCATTCTGAGCTCGACCTGCAGGCAT-3Ј for pIL6(Ϫ596/ϩ15)M3-CAT (loop mutant).

Plasmids and Cloning Strategies-pILIC-CAT
The pSVT8 and pSVT10 plasmids, expressing the tat gene in a sense or antisense orientation, respectively (35), were obtained from A. Caputo. pCMV-TAT plasmid, expressing the first exon of the tat gene, and pCMV-TAT 49 were a gift of K. T. Jeang (Laboratory of Molecular Microbiology, NIAID, NIH, Bethesda, MD). The pGEX-TAT plasmid was obtained from M. Giacca (International Center for Genetic Engineering and Biotechnology, Trieste, Italy). In this plasmid, the first exon of the tat gene is cloned in pGEX-2T, an isopropyl-1-thio-␤-Dgalactoside-inducible expression vector (Pharmacia, Uppsala, Sweden), which allows the production of GST-Tat fusion proteins. The pBlue610 plasmid expressing the C/EBP␤ (obtained from S. Akira, Institute for Molecular and Cellular Biology, Osaka University, Osaka, Japan) was used for in vitro transcription and translation of C/EBP␤. The pSP6: BSF2.5 plasmid, which allows for the in vitro production of IL6, was obtained from T. Kishimoto (Institute for Molecular and Cellular Biology, Osaka University, Osaka, Japan). The pGAL4-TAT plasmids, used in yeast transfections, were a gift of B. Cullen (36). In these plasmids, the expression of GAL4-Tat fusion sequences, consisting of wild-type or truncated Tat protein fused to GAL4 DNA-binding domain (amino acids 1-117), is directed by the yeast alcohol dehydrogenase promoter, while the yeast selectable marker is HIS3. To generate pGAD424-C/EBP␤, the c/EBP␤ cDNA was cloned downstream the GAL4 activation domain (amino acids 768 -881). The C/EBP␤ cDNA was excised from pBlue610 by SalI-EcoRI digestion and cloned in compatible sites of pGAD-424 vector (37). In this yeast expression vector, the selectable marker is LEU2, while the production of the fusion protein is driven by the alcohol dehydrogenase constitutive promoter. The correct insertion of the plasmids was verified by multiple restriction digestion and by sequencing using the Sanger method (38).
Cell and Transfection Procedures-HeLa-T8 and HeLa-T10, expressing the tat gene in a sense or antisense orientation, respectively, have been described (30). Cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (Flow Laboratories, Milan, Italy), 3 mM glutamine, and 10 mM Hepes buffer, pH 7.2 (Life Technologies, Inc., Milan, Italy). For transient expression experiments, cells were transfected by electroporation using a Bio-Rad apparatus (Bio-Rad, Milan, Italy). 3 ϫ 10 6 cells were resuspended in 0.3 ml of RPMI 1640 supplemented with 20% fetal calf serum and subjected to a double electrical pulse (0.2 V, 960 microfarads) in the presence of the indicated amounts of plasmid DNA. After electroporation, cells were washed and plated in complete medium. Transfection efficiency was monitored by cotransfecting the cells with 5 g of pnls-LacZ plasmid. ␤-Galactosidase activity was assayed using 50 g of protein extracts as described (39).
CAT Assays and Primer Extension Analysis-48 h after transfection, cells were harvested and washed once with PBS. Cell extracts were prepared by three cycles of freeze-thawing in 0.25 M Tris, pH 7.8, and CAT assays were performed as described previously (39). Proteins were measured in each cell extract with the Bio-Rad protein assay kit, and equal amounts of proteins were analyzed for each sample. Each assay contained 50 g of cell extract, 20 l of 4 mM acetyl-coenzyme A (Boehringer Mannheim), 1 l (0.5 Ci) of D-threo-[1,2-14 C]-chloramphenicol (DuPont NEN) in a final volume of 150 l of 0.25 M Tris, pH 7.8. Reactions were incubated for 3 h at 37°C, extracted with ethyl acetate, dried, and spotted on Polygram Sil G silica gel plates (Macherey-Nagel, Dü ren, Germany). Plates were run in a TLC tank containing a mixture of chloroform:methanol (95:5). After a 16-h autoradiography, the TLC plates were cut, and samples were counted in a Beckman LS5000TD scintillation counter.
Electrophoretic Mobility Shift Assays (EMSAs)-Nuclear and cytosolic extracts were isolated as described elsewhere (39 -41). Cells were harvested, washed once in cold PBS, and transferred to 1.7-ml microcentrifuge tubes for a second wash. The supernatant was removed, and the cell pellet was resuspended in lysing buffer (10 mM HEPES, pH 7.9, 1 mM EDTA, 60 mM KCl, 1 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 g/ml aprotinin, 10 g/ml leupeptin, and 0.2% (v/v) Nonidet P-40). After a 5-min incubation on ice, nuclei were collected by centrifugation (500 ϫ g, 5 min). The supernatant (cytosolic proteins) was recovered and stored at Ϫ80°C. Nuclei were rinsed with Nonidet P-40-free lysing buffer, resuspended in 300 l of Nonidet P-40free lysing buffer, and layered on the top of 300 l of the same buffer containing 30% sucrose. After centrifugation at 2,900 ϫ g for 10 min, the pelletted nuclei were resuspended in 150 l of buffer containing 250 mM Tris-HCl, pH 7.8, 60 mM KCl, 1 mM DTT, 1 mM PMSF, 10 g/ml aprotinin, and 10 g/ml leupeptin. Nuclei were then subjected to three cycles of freezing and thawing. The suspension was cleared by centrifugation (135,000 ϫ g, 15 min), and aliquots were immediately tested in a gel retardation assay or stored at Ϫ80°C until use.
In Vitro Tat-C/EBP Protein Interactions-To produce GST-Tat proteins, the plasmid pGEX-TAT was introduced in Escherichia coli strain SF 8 . Bacteria containing the plasmid were grown to 0.4 A 600 and induced with 0.5 mM isopropyl-1-thio-␤-D-galactoside, Inalco, Milan, Italy) for 2 h. Cells were then collected by centrifugation at 4°C (3,000 ϫ g for 15 min) and resuspended in RE buffer containing 50 mM Tris/HCl, pH 8.0, 120 mM NaCl, 0.5% Nonidet P-40, 5 mM DTT, 10 g/ml aprotinin, 10 g/ml leupeptin and 1 mM PMSF. Cells were broken with a French press apparatus, and the lysates were clarified by centrifugation at 4°C and 27,000 ϫ g for 30 min. Proteins were recovered and added to glutathione-Sepharose beads (Pharmacia), previously equilibrated in RE buffer. After an overnight incubation, the beads were extensively washed, and GST-Tat was eluted in RE buffer with 10 mM glutathione. The 35 S-labeled C/EBP␤ proteins were in vitro translated by using TNT TM coupled reticulocyte lysate systems (Promega, Madison, WI) according to the instructions of the manufacturer. For protein interaction studies, 10 g of GST and GST-Tat proteins were incubated with 15 l of translation mixture in buffer A (20 mM Hepes, pH 7.9, 10 mM MgCl 2 , 0.2 mM EDTA, 1 mM DTT, 150 mM NaCl, 5% (v/v) glycerol, and 0.05% Nonidet P-40). The samples were incubated for 2 h at room temperature. At the same time, the glutathione-Sepharose beads were washed, blocked in buffer A with 1 mg/ml BSA for 2 h, and washed again. These beads were added to the samples. After 3 h, the beads were collected by centrifugation (2,000 ϫ g for 10 s) and washed 10 times with buffer A. The pellets were then resuspended in sample buffer (70 mM Tris/HCl, pH 6.8, 7 mM EDTA, 0.01% bromphenol blue, 13% sucrose, 1% SDS, 7 M urea, and 10% (v/v) ␤-mercaptoethanol) and resolved on 12% SDS-polyacrylamide gel. Gels were treated with the Entensify kit (DuPont NEN), dried, and exposed.
Immunoblotting analysis was performed as described (39). Total cell extracts, nuclear proteins, or cytosolic proteins (10 g) were separated by SDS-10% polyacrylamide gel electrophoresis, transferred onto a membrane filter (Cellulosenitrate, Schleicher & Schuell), and incubated with the indicated first antibody in PBS plus 5% dry milk for 2 h at room temperature. Filters were washed three times in PBS and incubated with peroxidase-conjugated goat anti-rabbit or anti-mouse IgG (Boehringer Mannheim) at a 1:2000 dilution for 1 h. The proteins were revealed by using the enhanced chemiluminescence system (ECL) (Amersham).
In Vivo Protein Interaction: Two-hybrid System-Saccaromyces cer-evisiae CTY2 strain was grown in YPD medium (20 g/liter peptone, 10 g/liter yeast extract, 20 g/liter glucose, pH 5.8) and transformed by electroporation. Yeast were inoculated at 3 ϫ 10 6 cells/ml and grown up to 3 ϫ 10 7 cells/ml. Cells were harvested by centrifugation at room temperature (3,000 ϫ g, 15 min) and rinsed with H 2 O. Yeast pellet was resuspended in prepulse buffer (10 mM Tris/HCl, pH 7.5, 1 M sorbitol) and left for 30 min at room temperature. Cells were collected and resuspended in YPD, 1 M sorbitol at 6 ϫ 10 9 cells/ml. 3 ϫ 10 8 cells were added to the DNA mix consisting of 20 g of each plasmid in YPD, 1 M sorbitol to a 10-ml final volume. After 10 min, the mixture was electroporated with a Bio-Rad apparatus set at 1,100 V, 600 ohms, and 25 microfarads in 0.2-cm gap cuvettes. Cells were transferred in 1 ml of 1 M sorbitol and plated on 5-bromo-4-chloro-3-indolyl ␤-D-galactoside (xgal)-selective minimal medium containing 6.7 g of yeast nitrogen base, 20 g of agar, 20 g of glucose, 1 M sorbitol, 1 ϫ amino acids minus leucine and histidine, 0.1 M KPO 4 , pH 7, 20 g/ml of x-gal and H 2 O to 1 liter. Colonies were visible in 3 days and blue-positive after 4 -5 days. The expression of fusion proteins was assayed by immunoblotting, as previously reported (39). For this purpose, 50 g of cell extracts of transfected yeast cells were probed with polyclonal antibodies to Tat or to C/EBP proteins, which were obtained from the National Institutes of Health AIDS Research and Reference Reagent Program (Bethesda, MD) and from Santa Cruz Biotechnology, respectively.

RESULTS
Identification of the Region of the IL6 Promoter Responsive to Tat-We have recently reported that tat expression in epithelial HeLa cells and in MC3 lymphoblastoid cells resulted in the activation of endogenous IL6 gene transcription, as well as in the transcriptional induction of pIL6-CAT plasmid, an IL6 promoter-CAT construct (30). To gain further insight into the molecular mechanisms of the Tat-mediated activation of the IL6 gene, we constructed 5Ј deletion mutants of pIL6-CAT in which the region from Ϫ596 to ϩ15, Ϫ225 to ϩ15, or Ϫ112 to ϩ15 (29,41), was inserted 5Ј to the cat gene (shown in Fig. 1). These plasmids, hereafter referred to as pIL6(Ϫ596/ϩ15)-CAT, pIL6(Ϫ225/ϩ15)-CAT, and pIL6(Ϫ112/ϩ15)-CAT, respectively, were transiently transfected in HeLa cells stably expressing the tat gene in a sense (HeLa-T8) or antisense (HeLa-T10) orientation. Results from these experiments showed that pIL6(Ϫ596/ϩ15)-CAT and pIL6(Ϫ225/ϩ15)-CAT plasmids were efficiently transactivated by Tat, while the pIL6(Ϫ112/ ϩ15)-CAT construct was unresponsive to Tat (Fig. 1). This suggested that Tat-induced activation of the IL6 promoter required a region located between Ϫ225 and Ϫ112 bp. Indeed, this region harbors a C/EBP (NF-IL6) enhancer necessary for efficient IL6 promoter function (30,43). Next, we generated a plasmid where the Ϫ112/ϩ15 base pair region of the IL6 promoter was inserted downstream to a TAR-deleted HIV-1 LTR sequence (p⌬ILIC-CAT). The resulting p⌬ILIC:IL6(Ϫ112/ϩ15)-CAT plasmid (shown in Fig. 1) was transiently expressed in Tat-positive or Tat-negative HeLa cells. The Ϫ112/ϩ15 sequence of the IL6 promoter, (see pIL6(Ϫ112/ϩ15)-CAT in Fig.  1), conferred Tat responsiveness to the TAR-deleted HIV-1 LTR promoter (compare p⌬ILIC-CAT and p⌬ILIC:IL6(Ϫ112/ϩ15)-CAT plasmids in Fig. 1). This indicated that the Ϫ112/ϩ15 region, which was unresponsive to Tat in the context of the IL6 promoter (see pIL6(Ϫ112/ϩ15)-CAT in Fig. 1), could act as a TAR-like element when placed in the context of the HIV-1 promoter.
A primer extension analysis of cat mRNA transcribed from pIL6(Ϫ596/ϩ15)-CAT revealed a protected band of 98 nucleotides ( Fig. 2A), corresponding to the major transcription start site of the IL6 gene (42). The p⌬ILIC:IL6(Ϫ112/ϩ15)-CAT generated a major band of 248 nucleotides, corresponding to the transcription start site of the HIV-1 LTR (shown in Fig. 2A). Moreover, we observed the presence of a 98-nucleotide additional cat band in cells transfected with the p⌬ILIC:IL6(Ϫ112/ ϩ15)-CAT, indicating that the start sites of the IL6 promoter and of the HIV-1 LTR were both functional (Fig. 2B). A densi-tometric analysis of the cat bands confirmed that the HIV-1 LTR start site was preferentially utilized, with a minimal transcription originating from the IL6 promoter start site (not shown). The amount of cat mRNA in tat-expressing cells was 8 -10-fold higher than the cat mRNA transcribed by anti-tattransfected cells. In fact, both the 98-nucleotide cat band generated by transfecting pIL6(Ϫ596/ϩ15)-CAT and the 248-nucleotide cat band generated by the p⌬ILIC:IL6(Ϫ112/ϩ15)-CAT were stronger in Tat-positive than in Tat-negative cells ( Fig.  2A). These results identified the sequence of Ϫ112/ϩ15 as the minimal region of the IL6 promoter required for Tat to transactivate the ⌬TAR HIV-1-LTR. This suggested that the Ϫ112/ ϩ15 base pair region could function as a Tat-responsive sequence, possibly allowing Tat to be directed close to the TATA box of the IL6 promoter.
HIV-1 Tat Interacts with the IL6 Leader RNA-The primer extension results shown in Fig. 2, A and B, indicated that two transcription start sites were active in p⌬ILIC:IL6(Ϫ112/ϩ15)-CAT hybrid plasmid. This allowed the construction of the p⌬ILIC:IL6(Ϫ112/Ϫ67)-CAT plasmid, where the region of Ϫ67/ ϩ15, encompassing both the transcription start site and the 5Ј-untranslated region of the IL6 gene, was deleted (shown in Fig. 1). The resulting p⌬ILIC:IL6(Ϫ112/Ϫ67)-CAT plasmid was transiently transfected in HeLa-T10 (Tat-negative) and HeLa-T8 (Tat-positive) cells to address the question of whether the IL6 leader RNA was required for Tat-induced activation. As shown in Fig. 1, the p⌬ILIC:IL6(Ϫ112/Ϫ67)-CAT plasmid was unresponsive to Tat, indicating that a discrete region of IL6 leader RNA is strictly required for Tat. Accordingly, the Ϫ67/ ϩ15 region, encompassing the transcription start site and the 5Ј-untranslated region of the IL6 gene, restored the responsive-ness of the Tat-deleted p⌬ILIC-CAT plasmid (see p⌬ILIC: IL6(Ϫ67/ϩ15)-CAT in Fig. 1). A secondary structure analysis of this region according to the energy-minimizing algorithm of Zuker (44) defines an RNA stem-loop structure at the 5Ј-untranslated region of the IL6 mRNA (shown in Fig. 3). This RNA contains a UCU stretch that fulfills the sequence requirements for Tat binding to an RNA structure (45) and is potentially able to bind to Tat. To test this possibility, point mutations affecting the secondary RNA structure of the IL6 leader RNA at the bulge, stem, or loop were introduced into pIL6(Ϫ596/ϩ15)-CAT. The resulting mutant plasmids (shown in Fig. 3) were tested for responsiveness to Tat in transient expression experiments. As shown in Fig. 3, mutations that affect the bulge and the stem RNA (mutant M1 and M2, respectively) led to a drastic decrease in Tat responsiveness, while mutations of the loop were ineffective (mutant M3). In these experiments, the pIL6(Ϫ596/ϩ15)M1-CAT and pIL6(Ϫ596/ϩ15)M2-CAT plasmids did show a significant activation in Tat-positive (HeLa-T8) cells, suggesting that Tat can function, albeit at lower efficiency, in the absence of an RNA tethering structure. Indeed, Tat is able to activate the transcription of HIV-1 genes in a TAR-independent way, as recently reported (46 -47).
To test for the physical binding of Tat to the leader IL6 RNA, oligonucleotides corresponding to the wild-type IL6 leader RNA and to the relative mutants M1, M2, and M3 (shown in Fig. 3), were placed under the transcriptional control of the T7 promoter in a sense or antisense orientation and in vitro transcribed. Labeled RNAs were then tested for binding to Tat in an RNA-protein EMSA. Results shown in Fig. 4 indicate that Tat is able to specifically bind to the wild-type IL6 leader RNA. The Tat RNA binding was not displaced by the M1-RNA or M2-RNA  (HeLa-T8) or anti-tat (HeLa-T10) transfected HeLa cells. a, HeLa-T10 or HeLa-T8 were transiently transfected with 10 g of the indicated plasmids. b, CAT activity was determined 48 h after transfection by using 50 g of whole cellular extract. c, expressed as the ratio of the percentages acetylated. Transfection efficiency was monitored by co-transfecting the cells with 5 g of pnls-LacZ plasmid. ␤-Galactosidase activity was assayed using 50 g of protein extracts as described (39). The data are representative of five independent experiments in which two different plasmid preparations were used. Similar results were obtained by transient expression of pSV-T8 and pSV-T10 plasmids.
(affecting the bulge and stem IL6 RNA, respectively), while M3-RNA, affecting the loop IL6 RNA, was able to compete for the binding to Tat. In these experiments, wild-type TAR RNA competed specifically with IL6 RNA for the binding to Tat (Fig.  4A). Accordingly, the binding of Tat to TAR RNA was displaced by wild-type IL6 RNA and by M3-RNA, while M1-RNA and M2-RNA were substantially ineffective (Fig. 4B). In parallel experiments, the antisense RNA sequences were unable to bind to Tat (not shown). Moreover, Tat did not bind to either the single or to the double-stranded oligonucleotides corresponding to the IL6 leader RNA (shown in Fig. 4C).
The Basic Region of Tat Is Required for Tat-mediated Expression of the IL6 Gene-A basic domain of Tat, encompassing the amino acid residues 47-58, has been shown to significantly contribute to the capacity of Tat to bind to HIV-1 TAR RNA through an arginine fork (16,17,28,49). To test whether the arginine-rich domain of Tat is required to transactivate the IL6 promoter, we transfected HeLa cells with plasmids expressing either Tat amino acid residues 1-72 or a truncated form of Tat (residues 1-49, lacking the basic domain), together with p⌬ILIC:IL6(-112/ϩ15)-CAT, p⌬ILIC:IL6(-112/Ϫ67)-CAT, or p⌬ILIC:IL6(Ϫ67/ϩ15)-CAT plasmid. Results from these transient expression experiments showed that the Tat protein lacking the basic domain was unable to significantly transactivate the IL6 promoter (compare the results for p⌬ILIC:IL6(Ϫ112/ ϩ15)-CAT in Table I). Consistent with the results shown in Fig.  1, p⌬ILIC(Ϫ112/Ϫ67)-CAT, lacking the IL6 leader RNA, was unresponsive to both of the tat-expressing plasmids, while p⌬ILIC:IL6(Ϫ67/ϩ15)-CAT was fully responsive to Tat- . It is noteworthy that the residue 1-49-truncated Tat was still able to activate the wild-type HIV-1 LTR, albeit at a lower level than the wild-type Tat, suggesting that the amino-terminal domain of Tat can function as a transcription factor in the absence of TAR binding. Under these circumstances, Tat is possibly tethered to the HIV-1 LTR by a strong interaction with transcription factors binding to HIV-1 LTR cis sequences. This possibility is supported by the observation that Tat cooperates with transcription factors binding upstream regulatory DNA sequences circumscribed within the NF-B/Sp1 region of the HIV-1 promoter and with host cell proteins (12,24,25). Indeed, a binding of Tat to Sp1 factors has been reported (50).

HIV-1 Tat Induces an Increase in C/EBP Binding Activity and Interacts with C/EBP Transcription Factors-
To gain further insights into the molecular mechanisms of the Tat-mediated activation of the IL6 promoter, we tested whether Tat might induce an increased DNA binding activity of C/EBP (NF-IL6) transcription factors, which are major stimulants of the IL6 promoter (43). Nuclear extracts from HeLa cells transfected with pSVT8 (tat-expressing) or with pSVT10 (expressing tat in a antisense orientation) were tested for binding to an oligonucleotide corresponding to the C/EBP cis sequence of IL6 promoter. As shown in Fig. 5A, tat expression leads to a significant increase in C/EBP DNA binding activity. Moreover, an antiserum to C/EBP␤ supershifted the C/EBP complex, while an antiserum to C/EBP␦ was ineffective. In parallel experiments, cytosolic extracts from tat-or anti-tat-transfected cells expressed equal levels of C/EBP DNA binding activity (Fig.  5B). Aliquots of nuclear or cytosolic extracts were assayed for p53 DNA binding activity to monitor for protein concentrations (data not shown).
Immunoblot analysis of cell extracts of HeLa cells transfected with either pSVT8 or pSVT10 plasmid revealed equal amounts of total or cytosolic C/EBP␤ in both tat-and anti-tattransfected cells, while a consistent increase in C/EBP␤ proteins was observed in the nuclear fraction of tat-transfected cells (shown in Fig. 6A). The increase in the nuclear C/EBP␤ was detectable at 36 h post-transfection and declined thereafter (not shown). These data indicate that Tat specifically in-creases the nuclear levels of C/EBP␤ factors, resulting in an enhanced binding activity to C/EBP cis sequence. Under the same conditions, C/EBP␦ proteins were undetectable (data not shown).
To address the question of whether Tat could interact with C/EBP transcription factors, C/EBP␤ was in vitro translated and tested for binding to a GST-Tat fusion protein. As shown in Fig. 7, Tat physically associated with C/EBP␤. Under the same conditions, Tat did not bind to IL6 control protein in vitro produced from pSP6:BSF2.5 plasmid (not shown).
To test the possibility that Tat-C/EBP complexes could form in vivo, HeLa cells were transiently transfected with pSVT8 plasmid and subjected to immunoprecipitation with a Tat-specific monoclonal antibody followed by immunoblotting with antibodies to C/EBP proteins. We observed that Tat was FIG. 4. HIV-1 Tat binds to the IL6 leader RNA sequence. A, doublestranded oligonucleotides corresponding to either wild type IL6 leader RNA or to the relative mutant RNAs (shown in Fig.  3A) were in vitro transcribed in the presence of [ 32 P]UTP and tested for binding to recombinant Tat. B, the HIV-1 TAR sequence (ϩ1/ϩ44) was in vitro transcribed in presence of [ 32 P]-UTP and tested for binding to recombinant Tat. Cross-competitions were tested at 100-fold molar concentrations of unlabeled IL6 or TAR RNAs. C, single-or double-stranded oligonucleotides corresponding to the region of ϩ1 to ϩ15 of the IL6 promoter were labeled with [␥-32 P]ATP and tested for binding to GST or to GST-Tat proteins as detailed under "Materials and Methods." In control experiments, the HIV-1 TAR sequence (ϩ1/ϩ44) was in vitro transcribed in the presence of [ 32 P]UTP and tested for binding to GST or to GST-Tat proteins.
readily revealed in transfected cells, and that C/EBP␤ was specifically detected in immunoprecipitates of tat-expressing cells (Fig. 8, A and B).
To test whether Tat could functionally cooperate with C/EBP factors, we took advantage of the yeast genetic two-hybrid system (37). For this purpose, the C/EBP␤ cDNA was inserted in frame with the sequence of GAL4 coding for the GAL4 activation domain (amino acid residues 768 -881). The result-ing pGAD424-C/EBP␤ plasmid was cotransfected with pGAL4-TAT plasmids carrying the full-length tat sequence or truncated sequences of tat, fused to the GAL4 DNA-binding domain (shown in Fig. 9) in the CTY2 yeast strain, which carries an integrated copy of lacZ gene. Blue colonies grown on x-galselective medium were evaluated as indicative of in vivo interaction between C/EBP␤ and discrete regions of Tat. Results shown in Fig. 9 indicate that Tat interacted in vivo with C/EBP␤ and that this interaction resulted in the transcrip-   Fig. 9 as the interaction of Tat fused to GAL4 binding domain with Tat fused to VP16 activation domain). Moreover, this activation also occurred when Tat-(1-47) was used as a partner of C/EBP␤, indicating that the N-terminal, cysteine-rich, and core regions of Tat represent the minimal region of Tat required for an efficient heterodimerization, while the entire protein is required for a full transcriptional activation. In these experiments, comparable amounts of Tat or C/EBP proteins were produced by transfected yeast cells, as seen by immunoblotting of cell extracts, using antibodies to Tat or to C/EBP␤ (data not shown).

DISCUSSION
Despite the intensive investigation of the immunopathogenesis of AIDS, many questions concerning the molecular mechanisms of HIV-1 primary infection and progression remain unanswered (5,51,52). Recently, the identification of cohorts of HIV-exposed individuals who remain free of infection over a long period of viral exposure (53) as well as the existence of a small subgroup of HIV-1-infected subjects who are long term nonprogressors were described (54). Together with recent reports on viral life cycle (55,56), the above evidence argues that HIV infection and disease progression may ultimately result from a complex interplay between viral and host cellular factors involved in the immunological response to the viral infection and in the clinical evolution of AIDS.
HIV-1 Tat is a potent transactivator of HIV-1 LTR, acting on nascent TAR RNA and promoting full-length gene transcription (10 -13). Accordingly, Tat-defective HIV-1 is not viable (57,58). Emerging evidence shows that, in addition to its role on HIV-1 gene expression, Tat may exert additional functions. Tat is released in some extent extracellularly (20,59) and can function as a cytokine. In fact, Tat promotes the growth of endothelial cells and Kaposi's sarcoma cells directly or synergistically with basic fibroblast growth factor (Ref. 60 and references therein) and enhances cell survival in tat-expressing cells (61). Constitutive expression of tat in transgenic mice results in tumor development, including Kaposi's-like sarcomas and B cell lymphomas (62). Accordingly, stable expression of tat in IL6-dependent cells results in growth factor-independent growth and in tumorigenicity (30). Moreover, data in support of a nontranscriptional function of Tat in virion infectivity has been reported (63). The above evidence strongly suggests that Tat may participate in the establishment of HIV-1 infection and in the development of AIDS clinical fea- tures by promoting the expression of host cellular genes. In support of this possibility, Tat has been shown to activate the expression of the proinflammatory cytokines IL6 and tumor necrosis factor-␤ (30 -32) and to increase interleukin-2 and collagen gene expression (64,65). Tat was also shown to suppress promoter activity of major histocompatibility complex class I genes (66) and to exert immunosuppressive activity on antigen-induced T cell proliferation (67)(68)(69). Moreover, Tat has been shown to promote apoptosis by up-regulating CD95 ligand expression (70) or by activating cyclin-dependent kinases (71).
The mechanisms of Tat function on the expression of heterologous genes are unknown. In this paper, we address in molecular detail the mechanisms of Tat activity on the expression of IL6, a cytokine with a broad biological activity (7, 72) whose expression is deregulated in HIV-infected subjects (6,8,9). By using 5Ј deletion mutants of pIL6-CAT plasmid, and IL6:HIV-1-LTR hybrid plasmids where discrete regions of the IL6 promoter replaced the TAR sequence in HIV-1 LTR, we identified a short sequence of the 5Ј-untranslated region of IL6 mRNA that is required for Tat to transactivate the IL6 promoter. This region can acquire a stem-loop structure including a UCU trinucleotide bulge. Point mutations of the UCU bulge or of the stem resulted in a drastic decrease in Tat responsiveness (shown in Fig. 3) and in the inability of Tat to bind to the IL6 leader RNAs (Fig. 4). The IL6 RNA structure, with an estimated structure energy of Ϫ9.1 kcal/mol, is expected to be less stable than the TAR RNA structure. This suggests that Tat could bind with a low affinity to heterologous RNA sequences and may account for the ability of Tat to regulate the expression of multiple genes. Interestingly, Tat was still able to induce a low but significant activation of the bulge mutant pIL6(Ϫ596/ϩ15)M1-CAT plasmid (shown in Fig. 3), suggesting that Tat can function, albeit at a lower efficiency, without binding to an RNA tethering structure. In this case, Tat could be directed to the transcription start site of IL6 promoter by associating with specific transcription factors. This possibility is supported by the reports showing that Tat may associate with Sp1, TFIID factors, RNA polymerase II, and RNA polymerase II-associated factors (50,(73)(74)(75)(76)(77)(78). In addition, we now provide evidence that Tat can function by cooperating with C/EBP transcription factors. In fact, we observed an increase in the C/EBP DNA binding activity of tat-expressing cells with a selective increase in the amounts of nuclear C/EBP␤ factors (Figs. 5 and 6). This raises the possibility that Tat may increase the nuclear levels of C/EBP transcription factors by inducing post-translational modifications of C/EBP factors through the activation of specific kinases. Indeed, Tat activity on HIV-1 LTR-driven gene expression requires protein kinase C (79). Moreover, specific interaction of Tat with a cellular protein kinase has been reported (80), and serine and threonine phosphorylations of C/EBP␤ are required for IL6 promoter activation (81). The above data are consistent with and extend the recent observation that Tat enhances the tumor necrosis factorinduced activation of NF-B binding activity by possibly inducing protein phosphorylation (82). Since C/EBP and NF-B factors associate as heterodimers (83), which are potent activators of HIV-1 LTR (84), the above data suggest that Tat may also promote HIV-1 gene expression by up-regulating the cellular levels of transcription factors acting on the viral LTR. Moreover, Tat was able to complex with in vitro translated C/EBP␤, which is a major mediator of IL6 promoter function (Fig. 7). By immunoprecipitation and by taking advantage of the yeast two-hybrid system, this interaction was proved to occur also in vivo and to result in transcriptional activation of a reporter lacZ gene (shown in Figs. 8 and 9). The Tat association with C/EBP␤ suggests that Tat may increase the DNA binding ac-tivity of C/EBP dimers by enhancing their affinity for the target DNA. This mechanism accounts for the Tax activity on transcription mediated by bZip proteins (86,87). In the EMSA experiments shown in Fig. 5A, Tat could not be detected in the C/EBP-DNA complexes with an anti-Tat antibody (data not shown). This suggests either that Tat does not directly participate in the C/EBP-DNA complexes or that Tat dissociates from the DNA-binding complexes due to the electrical field of EM-SAs. These possibilities warrant further studies.
The data are consistent with the possibility that Tat may function on heterologous genes by interacting with RNA structures possibly present in a large number of cellular and viral genes, as recently reported (30 -33). In addition, Tat may function by forming heterodimers with specific transcription factors. These possibilities dramatically enhance the capacity of Tat to modulate the expression of heterologous genes and to play a major role in the pathogenesis of HIV-associated diseases.