Transactivation of Naturally Occurring HIV-1 Long Terminal Repeats by the JNK Signaling Pathway

To study the role of MAPK cascades in the regulation of naturally occurring human immunodeficiency virus type 1 long terminal repeats (HIV-1 LTRs), we analyzed several HIV-1 LTRs from patients at different stages of disease progression. One of these naturally occurring HIV-1 LTRs contains an insertion termed the most frequent naturally occurring length polymorphism (MFNLP) and exhibited high inducibility upon T cell activation. We found that the protein kinase mixed lineage kinase 3/src-homology 3 domain-containing proline-rich kinase, a specific activator of the stress-activated protein kinase (SAPK)/JNK signaling pathway in T lymphocytes, induces high transcriptional activation of this promoter. Promoter inducibility is inhibited by the SAPK/JNK inhibitor, the JNK binding domain of the JNK interacting protein 1, and Tam-67 (N-terminal deletion mutant of c-Jun). In electrophoretic mobility shift assay, several protein complexes were found to bind to the MFNLP sequence in T cells. We identified AP-1 factors c-Fos and JunB as MFNLP-binding proteins, whose binding is abolished by introducing point mutations in the 3′-half of the MFNLP sequence. Introduction of these point mutations into the MFNLP containing HIV-1 LTR reducedsrc-homology 3 domain-containing proline-rich kinase -mediated transactivation. These data indicate that the AP-1-like binding site in the MFNLP sequence gives rise to a higher inducibility of natural HIV-LTRs by the SAPK/JNK signaling pathway.

Human immunodeficiency virus type 1 (HIV-1) 1 utilizes cel-lular proteins and transcriptional signals, which are used to regulate T cell functions for virus production (1,2). The sequence in the 5Ј-HIV-1 long terminal repeat (LTR) region is important for the transactivation and regulation of the HIV-1 gene expression (3)(4)(5)(6). It contains binding motifs for DNA binding factors, such as nuclear factor-B, AP-1, and SP-1 (7)(8)(9)(10)(11). The interactions between different factors with the binding motifs on the 5Ј-LTR region determine the transactivation level of the HIV-1 promoters. Therefore mutations or insertions in the LTR region may modulate the transactivation and viral replication (12)(13)(14)(15). Analysis of 500 HIV-1 LTRs from 42 HIV-1-infected patients showed that 38% of the patients harbor a naturally occurring insertion, which is designated as the most frequent naturally occurring length polymorphism (MFNLP) (16,17). The length of the MFNLP sequence can vary between 15 and 34 base pairs with a consensus sequence of 5Ј-ctacacagctgctACAAgaACTGCTGA-3Ј. It locates between positions Ϫ120 and Ϫ121 of the HIVXB2 sequence, which is upstream of the HIV-1 nuclear factor-B binding motifs. Because the MFNLP sequence is not correlated with disease stage, CD4 count, or slope for interpatient or intrapatient MFNLP accumulation frequency, the function of the MFNLP sequence remains unclear. In addition, limited reports documented enhanced (13), decreased (14,15), or no effect on transcriptional activation on HIV-1 (17). Interestingly, most recent results suggest that MFNLPs with a duplicated Ras-responsive binding factor-2 cis element mediate a repressive effect on transcription in activated T cells and monocytes (16).
Binding of extracellular signals to the cell surface receptors leads to the stimulation of different protein kinase cascades including the mitogen-activated protein kinase cascade (18), the stress-induced protein kinase/c-Jun N-terminal kinase (SAPK/JNK), and the p38 cascades (19 -21). The mitogen-activated protein kinase cascade is strongly triggered by agonists such as growth factors and tumor-promoting phorbol esters but is rather weakly responsive to the stress inducers. On the contrary, the SAPK/JNK and p38 protein kinase cascades are more responsive to stress stimuli such as heat and osmotic shock, UV irradiation, DNA-damaging reagents, and proinflammatory cytokines (19,(22)(23)(24)(25). One of the protein serine/ threonine kinases that transactivates the SAPK/JNK signaling pathway is the src-homology 3 domain-containing proline-rich * This work was supported by the Deutsche Forschungsgemeinshaft Grants SFB165 and SFB172 (to U. R. R.) and Lu477/4-1 (to S. L.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

EXPERIMENTAL PROCEDURES
DNA Constructs, Cloning, and Antibodies-All cDNAs were subcloned into the multiple-cloning site of pRSPA (37). The cDNA of SPRK and the corresponding kinase inactive mutant SPRK(KϾA) were kindly provided by K. Gallo and P. Godowski (26). The JNK interacting pro-tein-1 (JIP-1) cDNA utilized in our study consists of the JNK binding domain (JBD), which was kindly provided by R. Davis (38). SPRK, SPRK(KϾA) and JBD were fused to a epitope-tagged Flag. Tam-67 was obtained from J. Birrer (39) and subcloned into the multiple cloning sites of the pRSPA vector. The 5XTRE-luciferase plasmid contains five tandem copies of the TPA-responsive element (40). c-Fos, serum-response element binding factor (SRF), c-Jun, Jun-D, Fra-1, ATF-2, CREB-1, Jun-B, c-Myb antibodies, and anti-Flag antibodies were purchased from Santa Cruz Biotechnology.
Cell Culture, DNA Transfection, and Luciferase Reporter Gene Assay-CD4 ϩ A3.01 T lymphoma cells (T cells) were maintained in RPMI 1640 (Life Technologies, Inc.) supplemented with 10% heat-inactivated fetal calf serum, 2 mM L-glutamine, streptomycin, and penicillin. The cells were cultured routinely to a density of 0.5 ϫ 10 6 to 1.0 ϫ 10 6 cells/ml. A3.01 CD4 ϩ T cells were split 4 ϫ 10 5 cells/ml one day before transfection. 5 ϫ 10 5 to 7 ϫ 10 5 cells/well (6-well plate) were transiently transfected with 0.4 g of HIV-1 LTRs or 0.5 g of 5XTRE, and 2.0 g of pRSPA (KR) expressing diverse kinases by the DMRIE TM -C liposome method (Life Technologies, Inc.). 0.1 g of pKR-HIV-Tat is additionally included for reporter gene assays with the HIV-1 promoters. The cells were incubated for 5 h in an incubator at 37°C under 7% CO 2 in the presence of the DMRIE-C reagent nucleic acid complexes, and 1.5 ml of growth medium was added. For the luciferase reporter assay, the cells were lysed and harvested 24 -38 h posttransfection. 3.01 cells were stimulated with 10 ng/ml or/and 0.5 M ionomycin (Sigma) for 18 h unless otherwise indicated. Cells from each well were harvested in 100 l of lysis buffer (50 mM sodium MES, pH 7.8; 50 mM Tris-HCL, pH 7.8; 10 mM dithiothreitol; 2% Trion X-100). The crude cell lysates were cleared by centrifugation. 50 l of precleared cell extracts was added to 50 l of luciferase assay buffer (125 mM Na-MES, pH 7.8; 125 mM Tris-HCl, pH 7.8; 25 mM magnesium acetate; 2 mg of ATP/ml). The activity was measured after injection of 50 l of 1 mM D-luciferin (AppliChem) in a Berthold luminometer. The luciferase activities were normalized on the ␤-galactosidase activity of co-transfected 1-g Rous sarcoma virus LTR ␤-galactosidase vector. The ␤-galactosidase assay was performed with 20 l of precleared cell lysate according to a standard protocol (76). Results are presented as luciferase units normalized to protein concentration. Each experiment was done in duplicates or triplicates. The mean and standard deviations of at least three independent experiments are shown in the figures.
Immunoblotting-Transfected cells from each well (6 well plate) were lysed in 50 l of lysis buffer (as described above). The crude cell lysates were cleared by centrifugation, 20 l of precleared cell extracts together with 5 l 5ϫ electrophoresis sample buffer (31 mM Tris HCl, pH 6.8; 1% SDS; 5% glycerin; 2.5% mercaptoethanol; 0.05% bromphenol blue) were denatured at 95°C for 5 min. Proteins were separated by SDS-polyacrylamide gel electrophoresis, blotted onto Nitrocellulose BAS-85 membrane (Schleicher & Schuell), and analyzed by Western blot analysis. For Western blot analysis, the membranes were incubated in blocking buffer (5% of nonfat dry milk), Tris-buffered saline, and 0.05% of Tween 20 (TBST) and washed in TBST as described previously (41). Protein A-peroxidase (Amersham Pharmacia Biotech) was used as the secondary antibody. This step was followed by the standard enhanced chemiluminescence reaction (ECL system).
Polymerase Chain Reaction Site-directed Mutagenesis-The BT94t-B1 HIV-1 LTR was subcloned to the pbluescript vector between KpnI and PstI. Polymerase chain reaction site-directed mutagenesis was then carried out as described in the QuikChange TM site-directed mutagenesis instruction manual (Stratagene). Three pairs of primers were designed to introduce mutations in the MFNLP sequence of promoter BT94t-B1: . The mutations were confirmed by sequencing and subsequently recloned back to the luciferase vector pAluc between KpnI and PstI. The mutated HIV-1 promoters (BT-M*, BT-R*, and BT-MR*) were sequenced by using the pAluc primer: 5Ј-CTCTAG-AGGATAGAATGG-3Ј.
Nuclear Extraction and Electrophoretic Gel Mobility Shift Assay-Crude nuclear fractions were extracted as described previously (41). Double-stranded oligonucleotide probes were labeled in a reaction mixture containing 200 ng of double-stranded DNA probe; 50 Ci of [␣-32 P]dCTP, 1 mM dATP, 1 mM dGTP, 1 mM dTTP; 500 mM Tris-HCl, pH 7.5; 100 mM MgCl 2 , and 2 units of klenow fragment. After a 30-min incubation at 37°C, oligonucleotides were separated on a G-25 Sephadex spin column and finally resuspended in Tris-EDTA buffer (30,000 cpm/l). 3 g of nuclear proteins were preincubated on ice with 2 g of poly(dI-dC) (Roche Molecular Biochemicals) and 1 g of bovine serum albumin in bandshift buffer (20 mM Hepes, pH 7.9; 1 mM dithiothreitol, 1 mM EDTA, 50 mM KCL, 4% Ficoll) for 5 min. 60,000 cpm 32 P-labeled oligonucleotide was added in a total volume of 20 l, incubated at room temperature for 15 min, and loaded onto 5% native polyacrylamide gels in 0.5ϫ Tris borate-EDTA buffer. Upon fractionation, gels were dried and exposed for autoradiography. The following oligonucleotides were used as labeled probes and unlabeled competitors, which were optionally added to the DNA protein binding reaction: MFNLP-wt (wild type), ice with 2 g of poly(dI-dC) (Roche Molecular Biochemicals) and 1 g of bovine serum albumin in bandshift buffer for 5 min. 60,000 cpm 32 Plabeled oligonucleotides were added, and incubated on ice for 15 min; 2 g of antibodies were added to the mixture, an we incubated the mixture on ice for 15 min and then at room temperature for another 15 min. the DNA-protein complexes were separated on 5% native polyacrylamide gels as described above.

Stress-induced and Mitogenic
Signaling Pathways Transactivate Naturally Occurring HIV-1 LTRs-We studied the regulation of naturally occurring HIV-1 LTRs by MAPK signaling cascades using six variants together with a laboratory strain HIV-1 LTR NL4 -3. The sequences of these LTRs are presented in Fig. 1A and show distinct point mutations or insertions (Fig.  1A). To investigate the effects of MAPK signaling cascades on the regulation of the naturally occurring HIV-1 LTRs, we used phorbol esters (TPA) to activate the Raf-mitogen-activated protein kinase/extracellular signal-regulated kinase kinase-ERK signaling pathway, and the combination of phorbol esters and ionomycin (TPA/ION) to additionally stimulate the SAPK/ JNK and p38 signaling pathways. As shown in Fig. 1B, the naturally occurring HIV-1 LTR strains as well as the laboratory HIV-1 LTR NL4 -3 strain are stimulated by TPA and TPA/ION. In contrast, no enhancement could be detected with ionomycin stimulation alone. TPA increases the transcriptional activity of the naturally occurring HIV-1 LTRs between 5-and 8-fold, and the combination of TPA/ION activates the HIV-1 promoters between 6-and 32-fold. Interestingly, the HIV-1 promoter, BT94t-B1, containing a MFNLP insertion shows a synergistic transcriptional activation upon double stimulation by TPA/ION, whereas the other naturally occurring and laboratory HIV-1 LTRs only showed additive effects. These results therefore suggest that the transcription of naturally occurring HIV-1 LTRs from various stages of disease progression can be activated by stimuli, which result in the induction of mitogenic and stress-induced signaling cascades.
SPRK Transactivates Naturally Occurring HIV-1 LTRs via SAPK/JNK-The synergistic effect on HIV-1 promoter BT94t-B1 activity by double stimulation with TPA/ION is presumably mediated via stress-induced signaling pathways. The kinase SPRK is a specific upstream activator of SAPK/JNK signaling pathways in A3.01 T cells (32), and we used this to mimic the activation of the SAPK/JNK pathway by TPA/ION. As a negative control we overexpressed kinase dead SPRK-(KϾA), which is an ATP binding site mutant of SPRK. Compared with the empty vector pKR, overexpression of SPRK transactivates different naturally occurring HIV-1 LTRs between 3-and 66-fold. In contrast, the kinase inactive SPRK-(KϾA) did not transactivate any of HIV-1 LTRs (Fig. 2A). Western blot analysis of the protein expression levels showed equal amounts of SPRK and SPRK(KϾA) (Fig. 2B). As observed above, the HIV-1 promoter BT94t-B1 demonstrated significantly higher transactivation by SPRK compared with the wild type promoter NL4 -3 and the other naturally occurring HIV-1 LTRs. That SPRK is acting through SAPK/JNK is shown in Fig. 2C. SAPK/JNK activate AP-1 transcription factors, so we used an AP-1-dependent promoter 5XTRE as a positive control. 5XTRE shows a very high SPRK-induced transactivation, whereas SPRK(KϾA) was not able to stimulate this promoter element, thus SPRK stimulates AP-1-dependent transcription. To determine whether SPRK-induced transactivation of the HIV-1 promoter BT94t-B1 is stimulated via SAPK/JNK signaling pathways, we overexpressed the dominant negative c-Jun (Tam-67) and the JBD of JIP-1. These proteins have been shown previously to inhibit SAPK/JNK pathways (38,42). Expression of both Tam-67 and JBD inhibited the SPRK-induced transactivation of the HIV-1 promoter BT94t-B1 by ϳ50% (Fig. 3A). The corresponding protein expression level of SPRK, Tam-67, and JBD are shown in Fig. 3B. In addition, Tam-67 and JBD reduced SPRK-induced transactivation on the 5XTRE promoter by over 80% (data not shown). Taken together, the above results indicate that SPRK-induced transactivation of HIV-1 promoter BT94t-B1 is mediated at least in part by SAPK/JNK.

AP-1 Factors Bind to the MFNLP Sequence of HIV-1 Promoter BT94t-B1 in T Cells-Compared with the wild type
HIV-1 promoter NL4 -3, the most striking feature of the HIV-1 promoter BT94t-B1 is a 20-base pair-long insertion located in the enhancer region known as MFNLP. To determine whether this MFNLP insertion plays a critical role in the observed behavior of this promoter, we used BT94t-B1-specific MFNLP oligonucleotides in electrophoretic mobility shift assays (EMSA). Five protein complexes (C-1 to C-5) are found to bind to the BT94t-B1-specific MFNLP oligonucleotides from A3.01 T cell nuclear extracts (Fig. 4). Interestingly, in contrast to AP-1 factors (Fig. 5A), the formation of these complexes is not influenced by the duration of TPA or TPA/ION stimulation. This result is supported by a competition assay using the unlabeled MFNLP sequence as shown in Fig. 6.
Analysis of the BT94t-B1-specific MFNLP sequence in the transcription factor data base (GBF bioinformatics) revealed AP-1 and Myb as two potential binding factors. Therefore, several AP-1 antibodies including anti-c-Fos, anti-c-Jun, and anti-JunD as well as Myb (see Fig. 5B) were assayed in EMSA by using MFNLP-and AP-1-specific oligonucleotides. As shown in Fig. 5A, the second binding complex formed on the MFNLP sequence is supershifted by anti-c-Fos antibodies (lane 4), whereas no supershift with anti-c-Jun and anti-JunD antibodies was observed. With respect to the AP-1 oligonucleotides, binding affinity of the protein-DNA complexes is strongly induced after TPA treatment for 3 h (Fig. 5A, lane 7), and supershift analysis of the protein-DNA complexes indicated that c-Fos, c-Jun, and Jun D were part of these complexes (Fig. 5A). As c-Fos does not contain a DNA binding domain, other AP-1 factors must be involved in MFNLP-specific binding. AP-1 antibodies anti-Fra-1, anti-ATF-2, as well as anti-CREB-1, and anti-Myb do not supershift any of the complexes that bind to the MFNLP sequence. Anti-JunB antibody, however, reduces the binding of the second (C2) and the third complex (C3) (Fig.  5B), indicating that JunB and c-Fos bind to the MFNLP sequence.
MFNLP Sequence Shares Similar Factors with Serum-responsive Elements of the c-Fos Promoter-The expression of c-Fos is dependent on the SRE. The factors that bind to this element include the SRF, Elk-1, and Sap-1 (see "Discussion"). We used competition assays to test whether SRE and MFNLP sequences share common binding factors and found that the serum-response element of the c-Fos promoter (SRE c-Fos) and the AP-1 binding motif inhibit the MFNLP sequence nuclearbinding protein complex formation in a dose-dependent fashion (Fig. 6). Because one of the DNA-protein complexes contains c-Fos and JunB, two transcription factors from the AP-1 family, this result supports the results presented in Fig. 5, A and B. However, with respect to the serum-response element of the c-Fos promoter, although this showed very strong competition with several protein-DNA complexes formed in A3.01 nuclear extracts, an anti-SRF antibody only supershifted the SRE c-Fos binding factors not the MFNLP sequence binding factors (Fig. 7).  (Fig. 8B). This result indicates that the 3Ј-half of the MFNLP sequence is the most important region for the MFNLP binding factors. We investigated this further by introducing two mutations to the 3Ј-half of the MFNLP sequence as shown in Fig. 9A. One of the mutations destroyed the potential AP-1 binding site located in the 3Ј-half of the MFNLP sequence, whereas the second mutation was introduced to the Ras-responsive binding factor-II binding motif in the middle of the MFNLP sequence (16). As shown in the bandshift assay, both mutations abolished the binding of the c-Fos-containing complex (Fig. 9B), indicating that both sequences are required for the c-Fos binding. We next determined whether these two mutations modulate the SPRK-induced transactivation of HIV-1 promoter BT94t-B1. Both of these two mutations, as well as a combined double mutation reduced SPRK-induced transactivation between 30 -50% compared with the wild type HIV-1 promoter BT94t-B1 (Fig. 10B). This suggests that the MFNLP sequence of BT94t-B1 plays a distinct role in SPRKinduced transactivation. DISCUSSION In this report, we investigated the regulation of naturally occurring HIV-1 LTRs by MAPK signaling cascades. To date studies of transactivation of HIV-1 LTRs by stress pathways were merely focused on the level of JNK and p38 activation (43)(44)(45)(46). In contrast, we focused on the mechanism of SPRKinduced transactivation; in particular, a naturally occurring HIV-1 promoter that contains a naturally occurring insertion named MFNLP. Using the extracellular stimuli TPA, ionomycin, and TPA plus ionomycin to mimic T cell activation leads to transcriptional activation of naturally occurring HIV-1 LTRs to different levels. In addition, the transactivation of naturally occurring HIV-1 LTRs can be partially inhibited by overexpression of intracellular dominant negative kinases such as ERK-B3, ERK-C3, and C4-HA or by the mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitors PD98059 and UO126 (data not shown). Interestingly, one of the naturally occurring HIV-1 promoters, BT94t-B1, displays a synergistic effect upon double stimulation by TPA/ION. Because the further opening of the (Ca 2ϩ ) inonophore increases transcriptional activity, we analyzed stress-induced protein kinase cascades to see whether they have an additional effect. In this study, overexpression of SPRK, the specific inducer of SAPK/JNK signaling pathway in T cell (32), transactivates HIV-1 promoters to different degrees. This result suggests that the SPRK-activated JNK signaling pathway in A3.01 T cells is involved in the regulation of HIV-1 gene expression.

Point Mutations Introduced to the MFNLP Sequence Abolished Binding of the c-Fos-containing Complexes and Reduced
In accordance with the high inducibility of HIV-1 promoter BT94t-B1 by TPA/ION, this promoter also displayed a much higher SPRK-mediated transcriptional activation in comparison with the wild type promoter NL4 -3 as well as with other naturally occurring HIV-1 promoters employed in our study. BT94t-B1 contains a 20-base pair-long MFNLP insertion, which has been shown to be precisely located at position Ϫ121.  6 and 7), and the AP-1 motif of interleukin 8 promoter (lanes 8 and 9). The EMSA and competition assays were performed as described under "Experimental Procedures."  3, 7, and 8). The EMSA and the competition assay were performed as described under "Experimental Procedures." fp, free probe.
No clinical or transcriptional phenotype that is common to all MFNLP LTRs has been found (16). Although extensive studies have contributed to the identification of cellular transcriptional factors that bind to the 5Ј-LTR, the relevance of different factors to the HIV-1 replication remains unclear. Previous reports have documented the essential function of the LTR sequence for transcription and replication (17,(47)(48)(49). For instance, a 24-base pair insertion with a 5Ј-ACTGC-3Ј motif upstream of nuclear factor-B sites has been shown to be responsible for a 3-fold increase in replication and proviral 5Ј-LTR-driven transcription (13). This insertion was later on named as MFNLP by Estable et al. (17). Different reports argued positive (13), negative (14,15), or no effects on transcription (17). In agreement with the finding of Golub and co-workers (13), we observed a distinct stress-induced enhancement on transactivation of HIV-1 promoter BT94t-B1, which contains a natural MFNLP insertion.
The activation of promoter BT94t-B1 by SPRK can be inhibited by overexpression of JBD and Tam-67, which indicates that JNK and AP-1 factors are the downstream substrates of SPRK in T cells. Tam-67 is an N-terminal deletion (⌬3-122) of c-Jun that lacks the major transactivation domain of c-Jun and inhibits c-Jun-mediated DNA binding (39). Moreover, Tam-67 acts as a potent inhibitor of AP-1-mediated transcriptional activation and transformation. Therefore, we used Tam-67 to inhibit SPRK-induced AP-1 transactivation. JIP-1, a JNK interacting protein-1, contains an N-terminal JBD (50). JIP-1 and JBD have been shown to be the specific inhibitors of JNK signaling, which act as an anchor to induce cytoplasmic retention of JNK (38). However, because Tam-67 and JBD inhibited ϳ50% of SPRK-induced transactivation of promoter BT94t-B1, other signaling pathways may also be involved. We observed up to 50% reduction of SPRK-induced BT94t-B1 activity by overexpression of inhibitor kinase K␤ (KD) (results not shown), a dominant negative kinase of the inhibitor B protein that activates nuclear factor-B through phosphorylation of inhibitor B inhibitor proteins (51)(52)(53). Furthermore, co-expression of both JBD and inhibitor kinase K␤ (KD) further reduced the SPRK-induced BT94t-B1 activity up to 80% (results not shown). Taken together, these data suggest that SPRK activates different downstream substrates, which in turn leads to the transcriptional activation of HIV-1 promoter.
Extracellular signals that activate ERK-and SAPK/JNK-dependent pathways have been shown to stimulate the transcription of c-fos and c-jun (54 -56). The targets of the SAPK/JNK signaling pathway include the transcription factors c-Jun and JunD (33,(57)(58)(59)(60), activating transcription factor (ATF-2), and the Ets domain transcription factors Elk-1 and Sap-1 (57). Five protein complexes in the A3.01 T cell nucleus were found to bind to the BT94t-B1-specific MFNLP sequence in the EMSA. One of the MFNLP binding complexes is identified to contain AP-1 factor c-Fos and JunB. Nevertheless, the formation of these complexes is not significantly influenced by applying different extracellular stimuli nor does the duration of stimulation affect complex formation. Besides, the competition of AP-1-specific oligonucleotides with the MFNLP sequence provides further evidence that AP-1 proteins are one of the factors that bind to the MFNLP sequence. We also revealed the strong competition between the serum-responsive element of the c-Fos promoter (SRE c-Fos) and the MFNLP sequence, but Anti-SRF  1, 6, 11, and 16), TPA-(lanes 2, 3, 7, 8, 12, 13, 17, and 18), or TPA/ION-(lanes 4, 5, 9, 10, 14, 15, 19, and 20)  antibody failed to supershift any of the protein-DNA complexes. SRE usually interacts with complexes consisting of SRF and a member of the ternary complex factor (TCF) family of transcription factors, such as Elk-1 and Sap-1 (61,62). Other factors from the TCF family, or those that induce TCF expression are probably involved in the formation of the protein complexes bound to the MFNLP sequence. Although c-Fos is not phosporylated by SAPK/JNKs but by FRK (a proline-directed MAPK, which is the only known to affect c-Fos activity) (63,64), the SAPK/JNKs are capable of phosphorylating and activating TCF/Elk-1 and Sap-1. Different studies have provided the evidence of induction of c-fos expression through SAPK/JNK-mediated TCF/Elk-1 or Sap-1 phosphorylation. c-Fos regulation by the SAPK/JNK pathway occurs probably only via the phosphorylation of TCF/Elk-1 or Sap-1, which in turn leads to the induction of c-fos (35,57,(65)(66)(67)(68)(69)(70). Increased production of the c-Fos protein is important for AP-1 activity because Jun/Fos heterodimers are more stable than Jun/Jun homodimers (71)(72)(73)(74).
In conclusion, our studies showed the importance of MAPK signaling pathways in the regulation of various naturally occurring HIV-1 promoter activities. The HIV-1 promoter BT94t-B1-specific MFNLP insertion plays a critical role in SPRKinduced transactivation. Nevertheless the precise contribution of SPRK to the induction of HIV-1 LTRs is unknown because of the lack of knowledge about the upstream activator of SPRK. AP-1 factors c-Fos and JunB, together with other factors that form a stable complex on the MFNLP sequence, are involved in the regulatory events. The natural occurrence or selection of an insertion at a certain position in the LTR region implies a role of different insertions that are taken advantages of by the virus for its replication. Identification of JNK signaling in the regulation of HIV LTR and therefore potentially in the control of viral latency highlights these enzymes as molecular targets for the refinement of novel therapies of HIV disease.