| Nucleic Acids Research | Pages |
ZFX transactivation of the HIV-1 LTR is cell specific and depends on core enhancer and TATA box sequences
Introduction
Materials And Methods
Cells and culture conditions
Construction of the reporter and expression vectors
Site-directed mutagenesis
Gel shift and DMS protection assays
Transfections and reporter assays
RNase protection assay
Results
ZFX575 and ZFX804 transactivate the HIV-1 LTR in several cell lines
The four C-terminal zinc fingers bind three sites within the ScaI-HindIII fragment of the HIV-1 LTR
Contribution of the key elements of the HIV-1 LTR core promoter to ZFX transactivation in TM3 and NCTC2544
Effects of NF[kappa]B, Sp1 and TATA box mutations on ZFX transactivation in CEM and U937 cells
Transactivation by ZFX575 and ZFX804 in the presence of Tat
Discussion
Acknowledgements
References
ZFX transactivation of the HIV-1 LTR is cell specific and depends on core enhancer and TATA box sequences
ABSTRACT
INTRODUCTION
The ZFX gene at Xp21.3-Xp22.1 was identified as a result of its strong homology to ZFY (1), a gene cloned from a region of the human Y chromosome thought to contain the testis-determining factor (2). Further genetic analysis excluded ZFY as the testis-determining factor (3), which later proved to be SRY, a gene lying in the neighbourhood (4). In mammals, the ZFX gene, which is ubiquitously transcribed, is under stronger conservation pressure than ZFY, whose transcription is detected mainly in testis (5-8). The exon-intron structure of Zfx (9) and the two related murine genes, Zfy-1 and Zfy-2 (10), has been established. Four untranslated exons are followed by seven coding exons, the arrangement of which was found to be quite intriguing. Coding exons 5-9 encode proline-, serine- and acid-rich regions and exon 10 encodes a 45 amino acid stretch comprising a nuclear localisation signal (1,9), whereas the last exon encodes 13 zinc fingers of the C2H2 type. Exons 5-10 end in the first nucleotide of a codon while exons 6-11 all begin with the second nucleotide of a codon (9). As a result, each of exons 6-10 can be spliced out, the resulting mRNAs potentially producing proteins that bear the zinc fingers fused to various N-terminal domains.
Alternative splicing has indeed been documented both for 5[prime]-UTR exons and for coding exons (1,7), the functional consequences of which remain to be elucidated. Alternative splicing of some 5[prime]-UTR exons and of exons 5 and 6, documented in human lymphoblastoid cells, was proposed to result in translation of two isoforms of ZFX, ZFX575 and ZFX804 (1).
The first direct evidence of the organismal importance of Zfx was obtained recently by deriving mutant mice deleted of Zfx exons 7-10 (11); some of the mice’s phenotypic features are reminiscent of those of Turner syndrome, which has been proposed as being caused by ZFX/ZFY haplo-insufficiency (12). These knock-out experiments stressed the importance of ZFX in growth during embryonic development and in efficient gametogenesis during adulthood (11). Furthermore, the phenotype indicates that other members of the gene family, although highly similar to Zfx, cannot compensate for all of its actions.
Since establishment of the primary sequences of their cDNAs, ZFX and ZFY were proposed to encode typical transcription factors with an N-terminal transactivation domain, a nuclear localisation signal and a C-terminal DNA-binding domain (1,2,6). It was later demonstrated that the homologous protein (79% identity) encoded by Zfy-1 displays sequence-specific DNA binding (13). Independently, ZFX was re-isolated through its capacity to bind HLA-A11 sequences critical for promoter activity and ZFX575 was shown to transactivate this promoter in the murine Leydig cell line TM3. The context of the sites responsive to ZFX575 has been previously shown to be a key feature in the observed transactivation (14). In order to examine whether this is a common feature of ZFX transactivation, several promoters of other virus or cell genes were tested in co-transfection studies. The HIV-1 LTR was found to be highly up-regulated and the requirements for ZFX transactivation in terms of promoter structure were examined in TM3 and in NCTC2544, a human epithelial cell line which proved also to be an adequate recipient for ZFX transactivation studies, as well as in two other cell types especially relevant for HIV virus biology: CEM, a human lymphoid CD4+ T cell line, and U937, a monocytic cell line. The relative transactivation efficiencies of ZFX575 and ZFX804 were compared and their cooperation with the viral transactivator Tat evaluated in monocytic and lymphoid cell lines.
MATERIALS AND METHODS
Cells and culture conditions
The TM3 cell line was obtained from ATCC and grown in a 1:1 mix of DMEM containing 4.5 g/l glucose with F12 (Gibco Laboratories), 2.5% FBS. The NCTC2544 cell line (ICN Laboratories) was a gift of Mojgan Djavaheri-Mergny and was grown in DMEM containing 4.5 g/l glucose, 5% FBS. The CEM cell line was obtained from O. Bernard, characterised for the SIL-TAL1 rearrangements (15) and maintained in RPMI 1640, 5% FBS, 1 mM pyruvate. The U937 cell line was obtained from ATCC and grown in RPMI 1640, 5% FBS, 1 mM pyruvate. Care was taken to maintain the cell lines in constant exponential growth.
Construction of the reporter and expression vectors
Expression vectors for ZFX575 and ZFX575[Delta]CT have been described previously (14) ZFX804 was constructed by abutting an EcoRI-KpnI cDNA fragment encompassing the complementary domain of ZFX804, beginning with GAATTCGCC ATG(+3)... and ending with ...(+724)AAA GTG GAT GGT ACC to the 5[prime]-end of the ZFX575 cDNA in which the silent mutation GGC ACT/GGT ACC had been introduced at position 736 [numbering according to (1)]. The resulting cDNA was expressed from the EcoRI site of pSG5 (Stratagene); pSG5-ZFX804[Delta]CT was derived from pSG5-ZFX804 by introducing the stop codon TGA in place of the TGT codon corresponding to amino acid 691; this results in truncation of the protein after zinc finger 9. The pRc[setmn]RSV-Tat construct was obtained from R. Galien. The Pgk-1 promoter SalI-HindIII 550 bp fragment was recovered from a gene targeting vector (16) provided by J. C. Bories and cloned in XhoI and HindIII cut p[beta]gal vector (Clontech). The HIV-1 LTR from the HIV-1-pXP2 construct (17), a gift of O. Schwartz, was excised with KpnI and HindIII and introduced into KpnI and HindIII cut pGL3 (Promega). The ScaI-HindIII fragment of the LTR was introduced into SmaI and HindIII cut pGL3 and small synthetic oligonucleotide fragments were phosphorylated, annealed and introduced into BglII and HindIII cut pGL3. The different mutated LTRs described in Figure
Site-directed mutagenesis
The protocol was adapted from the Quickchange kit protocol (Stratagene). An aliquot of 100 ng of the plasmid to be modified was mixed with 200 ng of the two self-complementary mutagenic oligonucleotides (which bear the changed sequence flanked by two arms extending each side of the mutation, each having a Tm of ~56°C) together with 250 µM dNTP and 1 U of Pwo DNA polymerase enzyme (Boehringer) in the regular 1× buffer provided by the supplier. The 50 µl mix was overlaid with mineral oil and the reaction cycled for one round at 94°C for 2 min, 48~52°C for 2 min and 72°C for 7 min and 14 rounds at 94°C for 1 min, 48~52°C for 1 min and 72°C for 7 min. The mix was digested with 10 U DpnI (Biolabs) in 100 µl for 1 h, phenol extracted, ethanol precipitated and the pellet dissolved in 20 µl of water. After agarose gel evaluation between 1/10 and 1/2 of the reaction was again digested with 10 U DpnI for 3 h and introduced into chemically competent DH5[alpha] cells. In each case, three different plasmids bearing the desired mutation were functionally evaluated and constructs were checked by sequencing. The sequences of the mutagenic oligonucleotides are available upon request.
Gel shift and DMS protection assays
The reactions were carried out as described previously using GST-4ZFX(14), except that the labelled DNA probes were added last to the reactions. The sequence reactions were done according to Maxam and Gilbert (18).
Transfections and reporter assays
The TM3 and NCT2544 cells were seeded 18 h before transfection in 24-well plates (500 µl/well) in order to reach 40% confluence at the time of transfection. Each well was transfected with a DNA mix consisting of 0.1 µg luciferase construct, 0.1 µg Pgk-1 promoter-[beta]-galactosidase construct and 0.3 µg expression vector in 10 µl 150 mM NaCl mixed with 10 µl of a PEI solution (1.6 mMeq) in 150 mM NaCl according to the supplier (Euromedex). Ten microlitres of the mix was directly introduced into the culture medium. The medium was removed 3 h later and replaced with 1 ml Dulbecco’s medium, 10% SVF. U937 and CEM cells were rinsed three times with serum-free culture medium and resuspended at 12.5 × 106 cells/ml of Dulbecco’s medium without phenol red (Gibco) supplemented with 1.5% DMSO (19), 2 mM phosphoric acid, 2 mM citric acid and 2 mM glutathione-reduced free acid. An aliquot of 400 µl of the cell suspension was mixed with 20 µg of a DNA solution [1 µg/µl; 2 µg luciferase construct, 2 µg Pgk-1 promoter-[beta]-galactosidase construct and 16 µg expression vector(s)]. The ratios of Tat expression construct and control or ZFX-expressing constructs were 1:4 (which corresponds to the ratio resulting in maximal transactivation). The cells were electroporated at 1 mF, 320 V in 0.4 cm wide cuvettes. After electroporation, cells were immediately transferred to 4 ml RPMI 1640, 1 mM pyruvate, 10% SVF medium. Twenty-four hours later, cells were harvested, rinsed twice with PBS and lysed in Reporter Lysis Buffer (Promega); half of the lysate was taken for [beta]-galactosidase measurements using the [beta]-galactosidase Reporter Gene Assay (Boehringer) following the manufacturer’s instructions. Luciferase measurements were done with the Luciferase Assay System (Promega) following the manufacturer’s instructions. The [beta]-galactosidase values were used to normalise the luciferase assays. Values of three independent experiments were aggregated except for CEM experiments in which five series were taken. Means, standard deviations and graphs were generated with Excel 5.
RNase protection assay
The antisense RNA probes were synthesised from T7 promoter-appended PCR fragments generated with the following oligonuleotides and labelled with [32P]UTP (800 Ci/mmol): T7-[beta]-Gal, GGATCCTAATACGACTCACTATAGGGTCAAAGTAAACGATGGTG, and Pgk-1 -90, GGCTCAGAGGCTGGGAAGG (Pgk-1-[beta]-galactosidase); T7-LUC, GGATCCTAATACGACTCACTATAGGAACCAGGGCGCTATCTC, and HIV-1 -50, GTGGCGAGACCCTCAGATCC (HIV-1-luciferase); T7-LUC, GGATCCTAATACGACTCACTATAGGAACCAGGGCGCTATCTC, and pGL3 4800, GTGCAGGTGCCAGAACATTTCTC (42Sp-luciferase). RNA from 10 million CEM cells transfected with the various constructs were recovered 24 h after electroporation with the RNA NOW reagent (Biogentex) according to the supplier’s instructions and treated with RQ1 DNase (Promega) for 30 min at 37°C, the reaction was split into two equal parts processed further according to the instructions of the HybSpeed RPA kit (Ambion) with gel-purified RNA probes corresponding to the luciferase and [beta]-galactosidase constructs, respectively. Radioactive sequencing reactions were run as markers to estimate the size of the bands on denaturing gels and quantification was with a Molecular Imager or a Biospace 1200 (Biospace, Paris).
RESULTS
In order to gain insight into the mechanisms involved in ZFX transactivation, several promoter-reporter constructs were screened in co-transfection studies in the TM3 cell line. While the HIV-1 LTR was found to be markedly transactived (Fig.
ZFX575 and ZFX804 transactivate the HIV-1 LTR in several cell lines
ZFX575 and ZFX804 transactivation of the HIV-1 LTR were first compared in TM3. As shown in Figure
Figure 1. Transactivation of the HIV-1 LTR in various cell lines. (A) Full-length wild-type LTR transactivation by the two different isoforms, ZFX575 and ZFX804, and their respective mutants devoid of the four C-terminal zinc fingers. Transactivation values are expressed relative to values obtained with the pSG5 expression vector with no insert: 1, pSG5; 2, pSG5-ZFX575; 3, pSG5-ZFX575[Delta]CT; 4, pSG5-ZFX804; 5, pSG5-ZFX804[Delta]CT. The mean values are indicated and the error bars indicate the standard deviations. (B) Increase in HIV-1 LTR-luciferase RNA levels by pSG5-ZFX575 in CEM evaluated by RNase protection assay. The estimated size of the probe and of the protected bands are indicated on the side; the protected bands correspond to the expected position of the HIV-1 LTR (29); 0 corresponds to assays of RNA extracts from cells transfected with pSG5 without the luciferase and [beta]-galactosidase constructs; N.D. indicates the lane of a reaction undigested by RNase and M.W. indicates the lane of an unrelated reaction sequence used to size the RNA. The protected bands corresponding to RNA from the Pgk-1-[beta]-galactosidase used as an internal control are shown on the right . (C) Transactivation of the -141/+82 ScaI-HindIII fragment of the HIV-1 LTR in TM3 (1-5, as in A). Figure 2. DMS protection of the ScaI-HindIII fragment of the HIV-1 LTR by the recombinant fusion protein bearing the four C-terminal zinc fingers, GST-4ZFX. The assay was carried out as previously described (14), except that an 8% denaturing gel was used. Regions of interest are shown; the numbering refers to the main initiation point of HIV-1 RNA (29). B indicates the lane corresponding to the DNA fraction bound by GST-4ZFX; F indicates the free unprotected DNA; AC and T correspond to lanes of sequencing reactions by partial chemical cleavage at the corresponding bases according to Maxam and Gilbert (18). The black dots indicate the protected positions and the arrow indicates a hyper-reactive position.
The four C-terminal zinc fingers bind three sites within the ScaI-HindIII fragment of the HIV-1 LTR
When transactivation was observed, it depended on the four C-terminal zinc fingers in all cell lines. The DNA-binding activity of ZFX to elements of the MHC class I promoter has previously been shown to map to that C-terminal region (14) and the core sequences of the previously identified recognition sites were in accordance with those found for the full-length Zfy-1 protein (13), as well as with those predicted from the amino acid sequence of the three C-terminal zinc fingers. DNA-binding sites of ZFX were therefore sought in the relevant region of the HIV-1 LTR. The DNA-binding sites of GST-4ZFX, a recombinant fusion protein bearing the four C-terminal zinc fingers of ZFX (14), were mapped by DMS protection in the 223 bp ScaI-HindIII fragment. Three sites at which DMS accessibility was markedly changed upon GST-4ZFX binding were detected along this fragment and are shown in Figure
Figure 3. (A) Sequences of the -141/+82 ScaI-HindIII fragment of the wild-type HIV-1 LTR construct and the mutated variants used in this study. TATA box, Sp1 and NF[kappa]B motifs are indicated in bold. The three Sp1 sites are in italic bold letters and GST-4ZFX recognition sites are italicised and underlined. The continuous lines correspond to regions of identity whereas discontinuous lines indicate deletions. The wild-type bases bordering the deleted regions and around the replaced TATA boxes are indicated for clarity. The sequence of the mutated -141/+82 ScaI-HindIII fragment of the HIV-1 LTR (SHm3) devoid of the three specific DNA-binding sites identified by DMS protection assay (Fig. 2) corresponds to the upper line. (B) Oligonucleotide sequences corresponding to the various minimal promoters cloned between BglII and HindIII sites of the pGL3 luciferase reporter. Nucleotides instrumental in the cloning are indicated in lower case letters. Figure 4. (A) Binding of the recombinant fusion protein bearing the four C-terminal zinc fingers, GST-4ZFX, to the mutated (SHm3) and wild-type (SHwt) ScaI-HindIII fragment of the HIV-1 LTR. The assay was carried out as previously described (14), with gel-purified and kinased DNA fragments excised from the respective luciferase reporter constructs. Lane 1, no recombinant protein added; lanes 2-5, GST-4ZFX; lane 3, 2000-fold excess of the SHwt fragment; lane 4, 2000-fold excess of the SHm3 fragment; lane 5, 2000-fold excess of the -282/-224 ZFX-binding site of the HLA-A11 promoter. F indicates the positions of the free probes and B1-B3 indicate the positions of the protein-DNA complexes of increasing molecular ratios: 1:1, 2:1 and 3:1, respectively. (B) Transactivation of the mutated -141/+82 ScaI-HindIII fragment of the HIV-1 LTR (SHm3) by: 1, pSG5; 2, pSG5-ZFX575; 3, pSG5-ZFX575[Delta]CT; 4, pSG5-ZFX804; 5, pSG5-ZFX804[Delta]CT (transactivation results should be compared to those obtained for the corresponding wild-type construct; Fig. 1B). Besides the crucial TAR element lying downstream from the core promoter, three elements of the HIV LTR promoter have been recognised to be of overwhelming importance for transcriptional regulation, namely the duplicated NF[kappa]B sites, the three Sp1 sites and the TATA box (20-26). Deletion of the NF[kappa]B sites within the context of the whole LTR had only limited impact on transactivation in TM3 (Fig. Figure 5. Transactivation of the various mutated LTR and minimal promoters by ZFX575 in TM3 and NCTC2544. (A) Transactivation in TM3. (B) Transactivation in NCTC2544. The shaded bars correspond to co-transfections with pSG5 and open bars to co-transfections with pSG5-ZFX575. Next, mutations of the TATA box were tested in the context of the whole LTR. Deletion of the TATA box increased basal activity of the LTR (~20-fold; not shown) both in TM3 and NCTC2544. On the other hand, transactivation by ZFX575 of the resulting construct, LTR[Delta]TATA, was severely impaired. This suggests that, under the present experimental conditions and in these cell lines, the TATA box region has paradoxical effects as an organiser of repression of the basal transcription level and as a sequence allowing stimulation of transcription. Furthermore, the initiator element (28) seems to be used efficiently only in the absence of the TATA box. Replacing the HIV TATA box with the SV40 TATA box had a similar impact on basal transcription level and transactivation by ZFX575 of the resulting construct, LTRTATA-SV, was only moderately affected (Fig. The effects of various modifications were examined in U937 cells. Deletion or replacement of the HIV-1 TATA box abolished transactivation by ZFX as observed in the other cell lines (Fig. Figure 6. Transactivation of various mutated LTR and minimal promoters by ZFX575 in cell lines of haematopoietic origin. (A) Transactivation in U937. Shaded bars correspond to the basal level obtained with pSG5 and set to 1; open bars indicate transactivation by pSG5-ZFX575. (B) Transactivation in CEM. (C) RNase protection assay. The numbers on the side indicate the estimated size of the probe and of the protected bands. Other indications are similar to Figure 1B. The lower lanes display the protected bands corresponding to RNA from the Pgk-1-[beta]-galactosidase construct used as an internal control. Since Tat is a key element in HIV transcriptional regulation, transactivation by ZFX was examined in its presence. As shown in Figure Figure 7. Transactivation by ZFX in the presence of Tat. Transactivation of the (A) wild-type LTR in CEM, (B) wild-type LTR in U937, (C) LTR[Delta]NF[kappa]B in U937 and (D) mutated truncated promoter SHm3 devoid of TAR in U937 by: 1, pSG 5; 2, pRcTat + pSG5; 3, pRc Tat + pSG5-ZFX575; 4, pRcTat + pSG5-ZFX804. It has previously been shown that ZFX575 activates transcription through specific binding sites of the HLA-A11 promoter (14). These early studies indicated that the precise context of the promoter exerted important effects on the transactivation process, although the mechanisms involved remained elusive. In the hope of gaining insights into such contextual effects and in order to determine whether they are characteristic of ZFX transactivation, screening for new promoters transactivated by ZFX was undertaken. Among many promoters examined, the HIV-1 LTR was transactivated with the largest magnitude. Furthermore, this promoter is one of the most intensively studied in a variety of cell types (reviewed in 29) and was therefore expected to provide an excellent substrate for further analysing the mechanisms of ZFX transactivation. The HIV LTR was not only transactivated in the TM3 murine Leydig cell line, which was previously found to be a useful recipient for ZFX transactivation studies (14), but also in a variety of human cell lines, such as NCTC2544 (an epithelial cell line), CEM (a T lymphoid CD4+ cell line) and U937 (a promonocytic cell line). The present report conveys three main points pertaining to the activity of ZFX and to regulation of the HIV-1 LTR: (i) trans-activation of a promoter by ZFX does not necessarily correlate with the presence of strong specific binding sites recognised by the last four zinc fingers; (ii) the HIV-1 LTR is transactived by ZFX in various cell types, including CD4+ T lymphocytic and monocytic cell lines targeted by the virus; (iii) ZFX575 and ZFX804 have distinct regulatory properties that vary with the cell type. The ZFX DNA-binding site identified at position -70 of the LTR, which is similar to the sites involved in HLA-A11 promoter transactivation by ZFX575 (14) and to Zfy-1-binding sites identified by PCR-SELEX (13), as well as the other variant sites at positions -40 and +7, were found to be unnecessary for ZFX transactivation. However, in all the cell lines tested, the last four C-terminal zinc fingers previously involved in specific DNA binding (14) were found to be critical for ZFX transactivation. Systematic mutagenesis of the 13 zinc fingers indicated that individual impairment of only three of them abolished trans-activation of the HLA-A11 and HIV-1 promoters; these three zinc fingers are located in the C-terminal end of ZFX, shown here to be required for transactivation (C.Gazin, unpublished data). Taken together, these observations suggest that the four C-terminal zinc fingers of ZFX are critical for transactivation of the HIV-1 LTR, because of interaction with cell components other than the specific DNA-binding sites identified in the LTR. Whereas ZFX-binding sites were not required for HIV-1 LTR transactivation, specific TATA box sequences were necessary in all the cell types tested. This was demonstrated either by replacement of the native sequences by the SV40 or HLA-A11 TATA boxes or by deletion of the TATA box, which created a situation in which, most probably, all the transcription complexes were recruited by the initiator of the HIV-1 LTR (28). This suggests that ZFX was able to activate complexes recruited by specific TATA box sequences in all the cell lines providing that either Sp1 or NF[kappa]B were allowed to co-operate. It should be noted that in its native context, the HLA-A11 TATA box permits ZFX transactivation through an upstream region (-205/-273) of that promoter, suggesting that the mechanisms of transcriptional integration involve different, context-dependent elements for the HLA-A11 and HIV-1 promoters. Similarly, transactivation by Tat has been shown to require specific TATA box sequences in the context of the HIV-1 LTR (19-26). Deletions of the Sp1 or NF[kappa]B sites of the LTR had different effects in the various cell types tested. Sp1 sites were found to be necessary and sufficient, along with the native TATA box, in NCTC2544 and TM3, sufficient but not necessary in CEM and insufficient and unnecessary in U937. On the other hand, the NF[kappa]B sites were unnecessary in NCTC2544, TM3 and CEM, but necessary in U937. Remarkably, transactivation by Tat was also severely impaired by deletion of these sites in U937 (Fig. The TATA box, NF[kappa]B and Sp1 sites and their spacing have been shown to be important for TAR-dependent transactivation exerted by Tat (19-26). In the cell lines used here, impairment of ZFX transactivation by mutation of these sites was paralleled by impairment of Tat transactivation. Whether reliance on the same promoter sites for ZFX and Tat transactivation means that ZFX acts in a way somehow similar to Tat remains to be shown. It has been proposed that Tat mainly acts by increasing elongation of transcription complexes through nucleation of a Tat-TAK/TEFb/cyclinT-CDK9 kinase complex believed to be appropriately positioned by interactions with TAR and promoter-bound factors and eventually resulting in efficient phosphorylation of the RNA pol II CTD (30-33) which renders it processive (34). On the other hand, significant effects of Tat on transcription initiation have been reported (35 and references therein) and Tat has been shown recently to recruit p300 and CBP transcriptional integrators (36-38), which may facilitate both initiation and elongation by facilitation of nucleosome displacement through chromatin protein acetylation. As the p300 and CBP co-activators are also involved in the integration of the transcriptional effects of Sp1 (39) and NF[kappa]B (40) in other promoters, it will be worth testing whether these co-activators modulate ZFX transactivation. However, the requirements for ZFX transactivation are more important than those for Tat transactivation. ZFX transactivation can be observed in some cell lines only; furthermore, the relative efficiencies of ZFX575 and ZFX804 display important differences among cell lines in which ZFX transactivation was observed. The permissive or refractory status of several tested cell lines, including those used in the present study, for ZFX575 and/or ZFX804 transactivation could not be correlated to their endogenous expression status for any ZFX isoforms as assessed by semi-quantitative RT-PCR (not shown). In some permissive cell lines, ZFX804 was roughly equipotent to ZFX575 (e.g. in TM3) and completely inactive in others (e.g. in NCTC2544), while the situation was intermediate in CEM and U937. Preliminary studies with ZFX804 deletion variants, including some that deleted most of the ZFX575 transactivation domain, indicated that the N-terminal domain of ZFX804 would essentially act as a negative effector of transcription (C.Gazin, unpublished data). The upstream domain specific to ZFX804 could recruit negative effectors present or missing in different cell types while the domain shared by both isoforms would be essentially competent or not, depending on the cell type, to cooperate with positive effectors of transcription. ZFX/ZFY are candidate genes for Turner syndrome (11,12), a condition which primarily affects gonad maturation and causes subtle alterations in various tissues resulting in an increase in morbidity (12). It has also long been known to include an immunological component and monocytic and T lymphoid cells of Turner syndrome patients have been shown to be affected ex vivo in various aspects of their physiology (41); the responsible gene(s) is therefore expected to usually be active and required in these cell types, among others. However, a definitive proof of ZFX/ZFY involvement in Turner syndrome is still missing. Whatever the case, experimental involvement of ZFX as a co-activator in HIV-1 LTR transactivation opens the possibility of its relevance in HIV virus biology, especially in reactivating the latent provirus. In this respect and in order to better understand the mechanisms of ZFX action, two major points will need to be addressed in the future. First, to define the elusive co-effectors of ZFX and, second, to decipher whether ZFX575 and ZFX804 synthesis, post-translational modification and activity are under the control of specific signalling pathways. I thank Professor F. Sigaux (INSERM U462) for his constant interest in this work and for critically reading the manuscript, Professor L. Degos for encouragement in the early times of this work and Professor F. Morinet for helpful discussions. I thank also J. Boisse, R. Nancel and B. Boursin for photographic work, B. Papp and F. Ferchal for providing various cell lines and discussions, A. Saïb for providing reporter constructs, for useful comments and for critically reading the manuscript andM. H. Stern, P. Paul, O. Schwartz, R. Galien and J. C. Bories for providing various constructs. This work was supported by grant 4041 from the Association pour la Recherche contre le Cancer, by the Ligue Nationale Contre le Cancer and by INSERM.
Contribution of the key elements of the HIV-1 LTR core promoter to ZFX transactivation in TM3 and NCTC2544
Effects of NF[kappa]B, Sp1 and TATA box mutations on ZFX transactivation in CEM and U937 cells
Transactivation by ZFX575 and ZFX804 in the presence of Tat
DISCUSSION
ACKNOWLEDGEMENTS
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification: 27 Apr 1999
Copyright©Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This Article 
Abstract
Print PDF (321K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (3)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Gazin, C.
Search for Related Content
PubMed
PubMed Citation
Articles by Gazin, C.
Social Bookmarking
What's this?