|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 278, Issue 39, 38068-38075, September 26, 2003
Interaction between the Varicella Zoster Virus IE62 Major Transactivator and Cellular Transcription Factor Sp1*![]() From the Department of Microbiology and Witebsky Center for Microbial Pathogenesis and Immunology, University at Buffalo, Buffalo, New York 14214
Received for publication, March 4, 2003 , and in revised form, June 25, 2003.
The varicella zoster virus (VZV) IE62 protein is involved in the activation of expression of all three kinetic classes of VZV proteins. Analysis of the viral promoter for VZV glycoprotein I has shown that the cellular factor Sp1 is involved in or required for the observed IE62 mediated activation. Co-immunoprecipitation experiments show that the two proteins are present in a complex in VZV-infected cells. Protein affinity pull-down assays using recombinant proteins showed that IE62 and Sp1 interact in the absence of any other viral and cellular proteins. Mapping studies using GST-fusion proteins containing truncations of IE62 and Sp1 have delimited the interacting regions to amino acids 612778 in Sp1 and amino acids 226299 in IE62. The region identified in Sp1 is involved in DNA-binding, synergistic Sp1 activation, and Sp1 interaction with cellular transcription factors. The interacting region identified in IE62 overlaps with or borders on sites involved in interactions with the VZV IE4 protein and the cellular factors TBP and TFIIB. Assays using wild-type and mutant promoter elements indicate that Sp1 is involved in recruitment of IE62 to the gI promoter and IE62 enhances Sp1 and TBP binding.
Varicella zoster virus (VZV)1 is a member of the alphaherpesvirinae and the causative agent of chicken pox (varicella) and shingles (zoster). The VZV genome is a linear double-stranded DNA molecule, which encodes approximately seventy proteins (1). The entire complement of VZV genes is believed to be expressed during lytic infection in three broad kinetic classes, immediate early (IE), early (E), and late (L). Transcription of VZV genes is performed by the host cell RNA polymerase II, as is the case with all other herpes viruses. Efficient expression of the VZV genome is driven by a small group of VZV gene products including those encoded by open reading frames (ORFs) 62, 4, 61, 63, and 10 (211). The major viral transactivator is the product of ORF 62 and its complement, ORF 71, which lie within the inverted repeats bracketing the Us region of VZV DNA. This protein is commonly designated IE62 since it is synthesized in the immediate early phase of lytic VZV gene expression. IE62 contains a potent N-terminal acidic transactivation domain and is capable of activating the expression of all three kinetic classes of VZV genes (1214).
While IE62 is involved in transactivation of VZV promoters, careful analysis of a limited number of individual viral promoters has shown that cellular transcription factors acting at sites upstream of the coding regions of the viral genes are also involved in the mechanism of IE62 activation. These proteins include the ubiquitous, sequence specific cellular factor Sp1. Sp1 is the protoype of a family of closely related factors which bind to GC-rich elements including the GC-box (GGGCGG or GGGCGGG) and the related GT/CACCC-box. Sp1 contains five distinct domains, four of which (A, B, C, and D) are involved in various aspects of transcriptional activation as well as a DNA binding region containing three zinc fingers (reviewed in Refs. 1517). Sp1 interacts with the TATA-binding protein (TPB) and the TBP-associated factor TAF 110/130 via the glutamine rich A and B domains (18, 19). Sp1 in addition interacts physically and/or functionally with a variety of other cellular proteins involved in transcription including p53, p300, cJun, RelA, the p107 Rb-related protein, smads 3 and 4, the promyelocytic leukemia protein (PML), the retinoic acid receptor (RAR/RXR), COUP-TF, the von Hippel-Lindau tumor suppressor gene product, YY1 and E2F (2034). Sp1 has also been shown to interact functionally with several viral transcriptional regulatory proteins. These include the E2 protein from bovine and human papilloma viruses (with which a direct physical interaction has been demonstrated), the HIV-1 Vpr protein, the Tax protein of human T cell leukemia virus type 1 (HTLV-1), the parvovirus NS-1 protein, and the Herpes Simplex Virus type 1 (HSV-1) transactivator, ICP4 (3541). Recent studies from this and other laboratories have shown that Sp1 binding sites are important for the regulation of expression of VZV promoters (42, 43). In the work described here we have explored the involvement of Sp1 in IE62-mediated activation of the viral glycoprotein I (gI) promoter using reporter gene assays and site-specific mutations and show that the presence of an Sp1 site is important for this activation. The presence of an Sp1-IE62 complex in infected cell extracts has been demonstrated by co-immunoprecipitation and the regions of these two proteins, which interact have been mapped by protein affinity pull-down assays. Finally, we present data, which indicate that Sp1 is involved in recruiting IE62 to the gI promoter and that the presence of IE62 increases the levels of Sp1 and TBP binding to the promoter.
Cell Lines and VirusesA3.01 cells, a CD4+ continuous human T cell line and MeWo cells, a continuous melanoma cell line, were propagated and maintained as described by Boucaud et al. (44). Virus strain MSP-VZV was propagated, maintained as frozen stocks, and used to infect MeWo cell monolayers as described by Sommer et al. (45). PlasmidsThe gI3.4CAT and gI3CAT reporter plasmids were generated as described by Kantakamalakul et al. (46). The gI3.4CAT plasmid contains a region of the gI promoter up to and including the activating upstream sequence (AUS) and TATA element The gI3CAT plasmid contains the TATA element but not the upstream AUS sequence (47). The Sp1 site mutation was introduced into the gI3.4CAT reporter plasmid by PCR mutagenesis using the following primers: Sp1m(+): GTGTGTGTAAGCTTAGATCTACAGAGTCACGAACCATTACAAGCTTAAGGTTCC and HindIII BglII AUS-Sp1m HindIII Sp1m(-): GGAACCTTAAGCTTGTAATGGTTCGTGACTCTGTAGATCTAAGCTTACACACAC; HindIII AUS-Sp1m BglII HindIII. The HindIII site was used as the cloning site, and the BglII site was used as the selection marker. All oligonucleotides were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA). Correct insertion of the mutated sequences was confirmed by DNA sequencing at the Roswell Park Sequencing Facility (Buffalo, NY). The reporter plasmid containing the Sp1 site mutation was designated Sp1m. The effector plasmid pCMV62, which expresses the IE62 protein has been previously described (12). The GST-Sp1 constructs, originally generated in Dr. Jonathan Horowitz's laboratory (48), were obtained from Dr. Adrian Black (Roswell Park Cancer Institute). These plasmids express full-length Sp1 (amino acids 1778), a C-terminal truncation (amino acids 1612), or an N-terminal truncation (amino acids 612778). Plasmids expressing GST-IE62 C-terminal truncations were those generated by Spengler et al. (49). These plasmids: pGEX-IE62 (1406), pGEX-IE62 (1299), pGEX-IE62 (1226), pGEX-IE62 (1161), and pGEX-IE62 (143) encode the N-terminal 1406, 1299, 1226, 1161, and 143 amino acids of the VZV IE62 protein, respectively. AntibodiesThe anti-IE62 monoclonal (H6) and rabbit polyclonal antibodies used in these studies have been previously described (49). Monclonal antibodies directed against the p32 subunit of human replication protein A (RPA) were purchased from Oncogene Research Products (San Diego, CA). Rabbit polyclonal antibodies directed against the p32 subunit of RPA were obtained from Dr. Thomas Melendy (University at Buffalo). Goat and rabbit polyclonal antibodies directed against Sp1 were obtained from Santa Cruz Biologicals (Santa Cruz, CA). DNA Transfections and Reporter Gene AssaysTransfection of A3.01 cells by electroporation with 10 µg of a specific CAT reporter plasmid and 0.2 µg of pCMV62 effector plasmid was performed as described by Boucaud et al. (44). Cells were harvested 48-h post-transfection and chloramphenicol acetyltransferase (CAT) assays were performed as previously described (1114). The extent of acetylation was quantified by using a Bio-Rad Phosphoimager and Molecular Imaging Software. For transfection experiments aimed at assessing the effect of the presence of IE62 on Sp1 and TBP binding, 175 cm2 flasks containing MeWo cell monolayers at 80% confluency were transfected with 30 µg of pCMV62 plasmid using LipofectAMINE Reagent (Invitrogen) as per the manufacturer's instruction. The transfected cells were harvested 48-h post-transfection, and nuclear extracts were prepared as described below. Preparation of Nuclear and Whole Cell ExtractsPelleted cells were washed once with 30 volumes of phosphate-buffered saline (PBS). Packed cells were resuspended in one packed cell volume of buffer A (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol) at 4 °C and allowed to swell on ice for 15 min. Cells were then lysed by 10 rapid passages through a 25-gauge hypodermic syringe and the homogenate was sedimented briefly at 12,000 x g. The crude nuclear pellet was resuspended in two-thirds of one packed cell volume (determined at the time of cell harvest) of buffer C (20 mM HEPES, pH 7.9, 25% (v/v) glycerol, 0.42 M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol) followed by incubation on ice with stirring for 30 min. The nuclear debris was pelleted by centrifugation for 5 min at 12,000 x g, and the supernatant (nuclear extract) was dialyzed against buffer D (20 mM HEPES, pH 7.9, 20% (v/v) glycerol, 0.1 M KCl, 0.2 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol) for 2 h. The dialyzed extract was then quick-frozen in liquid nitrogen and stored at -70 °C. Magnetic Bead Recruitment AssaysMagnetic bead recruitment assays were performed essentially as described by Lynch et al. (50). A 128-bp biotinylated DNA fragment containing the AUS and TATA elements and 68 bp of downstream sequence was generated from the gI3.4CAT and Sp1m reporter plasmids using biotinylated primers (IDT, Coralville, IA). A 108-bp biotinylated fragment lacking the AUS sequence was derived from the gI3CAT reporter plasmid. 10 pmol of biotinylated DNA in 100 µl of 1x Binding and Washing (B&W) buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 1 M NaCl) were added to 50 µl of M-280 streptavidin-coated magnetic beads (Dynal, Lake Success, NY) which had been prewashed twice with 100 µlof2x B&W buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 2 M NaCl). The mixture was incubated for 15 min at room temperature. The beads were collected using a magnetic particle concentrator (Dynal), washed twice with 200 µl of 1x B&W buffer, and then washed twice with 200 µl of TEN buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 0.1 M NaCl). Next either 250 µg of VZV-infected MeWo nuclear extract or 250 µg of uninfected MeWo or T cell nuclear extract or 250 µg of pCMV62-transfected MeWo cell nuclear extract was incubated with the beads for 1 h at 4 °C. The reactions were resuspended 3x during the 1 h incubation. Next the beads were washed 3x with 400 µl of TEN buffer. To elute specifically bound proteins the beads were incubated with 10 µl of 2x B&W buffer for 5 min at room temperature. The eluted fractions were boiled in 2x SDS-PAGE loading buffer for 10 min and analyzed by 10% SDS-PAGE and immunoblotting. Co-immunoprecipitation of IE62 and Sp1Protein G-Sepharose 4 Fast Flow (Amersham Biosciences) was blocked with a 4% milk/PBS solution for 1 h at 4 °C. Co-immunoprecipitation was initiated by conjugating 50 µg of antibody to 200 µl of protein G-Sepharose for 2 h. The antibodies used were monoclonal anti-IE62 antibody H6 and goat polyclonal anti-SP1 (Santa Cruz Biotechnology). The antibody-conjugated beads were washed with three 500-µl aliquots of 1x PBS/1% Tween 80. The beads were then incubated with MeWo cell nuclear extract (500 µg of protein) for 3 h at 4 °C. Following incubation the reaction mixtures were centrifuged and the beads washed three times with PBS/1% Tween 80. The beads were then resuspended in 2x SDS-PAGE loading buffer and boiled for 10 min. The samples were analyzed on a 7.5% SDS-PAGE gel and transferred to nitrocellulose membranes. The antibodies used for the Western blot were rabbit polyclonal anti-SP1 antibody (Santa Cruz Biotechnology) and rabbit polyclonal anti-IE62 antibody generated in our laboratory. Monoclonal antibodies directed against the RPA32 subunit of RPA and polyclonal rabbit antibodies directed against RPA32, were used in control experiments for either immunoprecipitation or immunoblot detection to determine the specificity of the IE62-Sp1 interaction. Reactive bands were visualized using goat anti-rabbit or goat anti-mouse immunoglobulin G (IgG) conjugated with horseradish peroxidase (Chemicon, Temecula, CA) in conjunction with Supersignal West Pico Chemiluminescence substrate (Pierce, Rockford, IL). Quantification of the relative amounts of IE62 and Sp1 co-immunoprecipitated was performed by densitometry using a Bio-Rad (Hercules, CA) model GS-700 imaging densitometer. Experimental signals were normalized to signals obtained from loading controls or immunoblots of serial dilutions of recombinant IE62 or nuclear extracts from infected cells.
Protein Affinity Pull-down AssaysThe VZV IE62 protein was expressed in recombinant baculovirus and purified from infected cell cultures as previously described (49). In the protein affinity pull-down experiments, GST-Sp1 fusions, GST-IE62 fusions and GST were expressed in Escherichia coli DH5
Sp1 Is Involved in Regulation of the VZV Glycoprotein I PromoterWe have previously identified a 20-base pair element within the VZV gI promoter designated the activating upstream sequence (AUS). This sequence, along with an atypical TATA element (ATAAAA) located 12-base pairs downstream of the AUS are both required for IE62-mediated activation (46, 51) and represent the minimal IE62-responsive gI promoter designated gI3.4. The AUS contains an Sp1 binding site (ACGCCC), which is identical to one of the two Sp1 sites found upstream of the VZV gE promoter (42). This sequence partially overlaps with a predicted atypical binding site for the transcription factor USF and a potential AP-1 binding site. A two base pair mutation was generated in the non-overlapping region of the Sp1 site (Fig. 1A) and the ability of this mutant promoter element (Sp1m) to support IE62-mediated expression of a CAT reporter gene was determined relative to the wild-type promoter and a promoter lacking the entire AUS sequence (gI3). The results are presented in Fig. 1B and show that the Sp1 mutation resulted in expression of CAT activity at levels seen with the AUS-deleted promoter, which is non-responsive to IE62.
Magnetic bead recruitment assays were next performed in order to determine if Sp1 does, in fact, bind to this promoter element and if the introduced mutation alters that binding. In these experiments, oligonucleotides derived from the gI3.4 promoter, Sp1m promoter, and gI3 promoter containing wild-type and mutant Sp1 sites within the AUS or lacking the entire AUS sequence, respectively, were bound to streptavidin-coated magnetic beads. All three oligonucleotides contained the gI promoter TATA element plus 43 nucleotides downstream of the TATA element. The oligonucleotide-bound beads were incubated with MeWo cell nuclear extracts derived from cells transfected with a plasmid expressing the IE62 protein. Bound proteins were eluted with 2 M NaCl separated by SDS-PAGE and analyzed by immunoblotting using anti-Sp1 antibody. The data are presented in Fig. 1C and show that the mutation within the Sp1 site very significantly decreased the binding of Sp1 as compared with the level seen with the wild-type sequence (lanes 1 and 2). Experiments using the promoter element lacking the entire AUS sequence but retaining the TATA box and downstream sequences gave essentially identical results as the mutant Sp1 site (lane 3). Assays performed using uninfected MeWo cell and T cell extracts also showed binding only to the wild-type sequence (data not shown). Thus the results presented above indicate that, in the context of this promoter, Sp1 plays a major role in the mechanism of IE62 activation and this role correlates with the binding of Sp1 to a specific site within the promoter. Co-immunoprecipitation of Sp1 and IE62Based on the fact that mutation of the Sp1 site significantly decreased the level of Sp1 bound to the gI promoter element and also reduced IE62-mediated transactivation of this promoter, we investigated the possibility of an interaction between these two proteins in infected cells. In the first series of experiments a polyclonal goat-anti-Sp1 antibody was used in protein G-enhanced immunoprecipitation from VZV-infected MeWo cell nuclear extracts. Precipitated proteins were separated by SDS-PAGE and analyzed for the presence of Sp1 and IE62 by immunoblot. The results are presented in Fig. 2A and show that the anti-Sp1 antibody precipitated IE62 as well as Sp1. In these experiments, 44% of the total Sp1 was precipitated and 2.4% of the total IE62 was co-precipitated. The converse experiment was also performed using the monoclonal H6 anti-IE62 antibody. The results are presented in Fig. 2B and show that the anti-IE62 monoclonal antibody precipitated both IE62 and Sp1. In these experiments, 21.7% of the total IE62 was precipitated and 4.9% of the total Sp1 was co-precipitated. Two additional repetitions of the immunoprecipitations yielded similar levels of co-precipitation. These results indicate that a fraction of Sp1 and IE62 in VZV-infected cells is present in a protein complex but do not eliminate the possibility that one or more additional proteins or DNA within such a complex act as a bridge between Sp1 and IE62. Control experiments showed that the antibodies used did not co-precipitate the ubiquitous and abundant eukaryotic single-stranded DNA-binding protein, RPA which is involved in DNA replication, repair, and recombination and has been shown to bind to specific cellular promoter regions (52).
Mapping of Sp1 and IE62 Interaction DomainsProtein affinity pull-down assays were performed using recombinant Sp1 and IE62 constructs in order to address the question of a direct physical interaction between these proteins. In a preliminary study, we have recently shown that recombinant IE62 can interact with a GST fusion protein containing full length Sp1 (53). In the work presented here, GST fusion proteins containing full-length Sp1 (amino acids 1778), a C-terminal deletion containing the glutamine rich A and B activation domains and domain C (amino acids 1612) and the complementary N-terminal deletion containing the DNA-binding domain and activation domain D (amino acids 612778) were expressed and bound to glutathione-Sepharose beads. Purified recombinant IE62 derived from baculovirus was then added to the Sp1-GST bound beads. Bound proteins were eluted by boiling and the presence of IE62 in the eluates was detected by immunoblotting. The results are presented in Fig. 3A and show that recombinant IE62 bound to the full-length Sp1-GST fusion protein and to the GST fusion protein containing amino acids 612778 of Sp1. In contrast, IE62 bound at only trace levels to the C-terminal Sp1 truncation (amino acids 1612) and showed no detectable interaction with GST alone. Thus these data indicate that IE62 interacts with the region of Sp1 containing the Sp1 DNA-binding and D domains (Fig. 3B). Furthermore, these data confirm that there is a direct physical interaction since binding occurred between recombinant IE62 and a specific region of Sp1 in the absence of other viral and cellular proteins.
The region of IE62 to which Sp1 binds was identified using a similar strategy. We hypothesized that the region of IE62 which is required for interaction with Sp1 would likely lie within the N-terminal third of IE62. This region of IE62 contains the IE62 acidic activation domain and has been shown to interact with both cell and virus-encoded proteins (14, 49, 54). GST fusion proteins containing C-terminal truncations of IE62 were expressed and bound to glutathione Sepharose. These included GST-fusion proteins containing amino acids 143, 1161, 1226, 1299, and 1406 of IE62. MeWo cell nuclear protein extracts containing Sp1 were then adsorbed to the beads and the presence of Sp1 in eluates from the beads was assessed by immunoblotting. The results are presented in Fig. 4 and show that the region of IE62 required for interaction with Sp1 falls between amino acids 226299 since both the 1406 and 1299 amino acid IE62 fragments bound Sp1 whereas the 1226 amino acid fragment did not. The differences in the levels of binding observed between the two longest constructs most likely reflects the relative amounts of the intact IE62 fragments bound to the beads. The Sp1-binding region (Fig. 4B) borders on, and slightly overlaps with, IE62 amino acid sequences which have previously been shown to be involved in interaction with the cellular transcription factors TBP and TFIIB and significantly overlaps with the region involved in the interaction of IE62 with the viral IE4 protein (49, 54).
Binding of Sp1, IE62, and TBP to the gI3.4 Promoter ElementThe results described above indicating that Sp1 was capable of binding to the wild-type gI3.4 promoter element in the context of uninfected as well as IE62-transfected cell nuclear extracts suggest that binding of Sp1 to the wild-type sequence can occur in uninfected cells yet it is not sufficient for activation of expression from the promoter. We therefore investigated whether the presence of Sp1 is involved in recruitment of IE62 to the gI promoter. Magnetic bead recruitment assays were performed using wild-type and mutant gI3.4 promoter elements and nuclear extracts derived from VZV-infected MeWo cells. The bound proteins were eluted from the beads with high salt, separated by SDS-PAGE and probed with anti-IE62 antibodies or anti-Sp1 antibodies. The resulting immmunoblots are presented in Fig. 5A and show that both IE62 and Sp1 were bound in significant amounts to the beads conjugated with the wild-type sequence. In contrast, little or no binding of either protein was observed to the promoter containing the Sp1m sequence. These results suggest that IE62 is recruited to the gI3.4 promoter element, at least in part, through its interaction with Sp1.
We next wished to investigate if the presence of IE62 influences the binding of Sp1 to the gI3.4 promoter element based on findings with several cellular factors and the HTLV-I tax protein with regard to enhancement or suppression of Sp1-activated expression (24, 26, 28, 33, 34, 39). We also wished to determine if the presence of Sp1 and/or IE62 influences the interaction of TBP/TFIID with the promoter since both proteins can bind TBP and the HSV ICP4 protein has been shown to recruit TBP/TFIID to promoters (55). Magnetic bead recruitment assays were performed with uninfected MeWo cell nuclear extracts, uninfected extracts to which recombinant baculovirus-derived IE62 had been added, and extracts derived from cells transfected with the pCMV62 expression plasmid. The results showed that Sp1 binding to the gI3.4 promoter element was observed in the absence of IE62, but was significantly increased by the presence of either exogenously added recombinant IE62 (Fig. 5B) or by the presence of IE62 expressed from the transfected plasmid (Fig. 5C). An even more striking enhancement was observed with TBP where readily detectable binding was seen only in the presence of IE62 in both cases (Fig. 5, B and C).
In this article we have investigated the functional and physical interaction between the VZV transactivating protein, IE62 and the ubiquitous cellular transcription factor Sp1. IE62 has long been recognized as the major VZV transactivator and is capable of transactivation of promoters representing all three kinetic classes of viral promoters. The mechanism or mechanisms employed by IE62 to achieve transactivation of viral genes, however, remain unknown. Sp1 has been implicated in regulation of VZV gene expression based on work with specific VZV promoters including those controlling expression of the genes encoding glycoproteins I and E (42, 43, 56) and a bidirectional promoter,2 which controls expression of the VZV ORF 28 (DNA polymerase) and ORF 29 (major single-strand DNA binding protein) genes (57). Furthermore, computer analysis has shown that Sp1 binding sites in the form of either the consensus GC-box Sp1 binding site or the ACGCCC binding sequence present in the gI promoter occur in potential control elements for 28 VZV genes including genes from all three kinetic classes (53). This represents a minimum possible number of Sp1 sites since GC and CT-rich sequences which could also be recognized by Sp1 or other members of the Sp-factor family are frequently found in intergenic regions within the viral genome (1). Thus Sp1 likely has a major role in one or more of the mechanisms of IE62-mediated activation of VZV genes.
In the work presented here we have shown that in transient transfection assays the presence of an intact upstream Sp1 site is required for significant IE62-mediated activation of the minimal VZV gI promoter. The integrity of this site has also recently been shown to be important for viral growth in human skin and T cells in a SCID mouse model (43). Thus the in vitro observations presented here correlate with viral infection and pathogenesis in an in vivo setting. In contrast, mutation of the USF and AP-1 sites in the AUS resulted in either a very modest or no effect, respectively, in the mouse model. The requirement for both Sp1 and IE62 for activation of the gI promoter contrasts with findings for the HSV homologue of IE62, ICP4. ICP4 has been shown to be able to functionally substitute for Sp1 at the HSV thymidine kinase promoter, and mutation of the Sp1 site within that promoter had a minor effect on ICP4-induced expression. Moreover, ICP4 inhibits Sp1-mediated activation of its own promoter via formation of a tripartite complex with TBP and TFIIB (41, 58, 59). In contrast, IE62 has been shown to significantly transactivate the ICP4 promoter (13). The observed interplay of IE62 and Sp1 as synergistic co-activators appears therefore, at this point, to be unique to IE62. This implies that the mechanism(s) of activation of IE62, which contains an N-terminal acidic activation domain not found in ICP4, may show important differences (as well as similarities) when compared with those of the HSV transactivator and these differences could be reflected in differences in promoter structure and/or cell tropism between the two viruses. We further showed that IE62 and Sp1 could be co-precipitated from VZV-infected cell extracts using either anti-IE62 or anti-Sp1 antibodies indicating that the two proteins are present within a complex in infected cells and raising the possibility of a direct and specific interaction between them. The regions of Sp1 and IE62 required for direct physical interaction between the proteins were mapped via protein affinity pull-down assays using recombinant GST fusion proteins containing full-length and N-terminal and C-terminal deletion fragments of Sp1, GST fusion proteins containing C-terminal deletion fragments of IE62, and full-length IE62 derived from recombinant baculovirus. These experiments showed that the region of Sp1 required for the interaction encompassed the C-terminal 166 amino acids of the molecule (612778). This region contains the Sp1 DNA-binding domain including the three zinc fingers characteristic of the Sp/XKLF family of transcription factors (15, 17) and the Sp1 D domain, which is involved in synergistic effects between Sp1 molecules (60). Several cellular transcription factors have been shown to directly interact with this region (Fig. 3B) and alter Sp1 activation of specific promoters. These include RelA, PML, Smads 3 and 4, YY1, and E2F (24, 26, 3134). Thus IE62 may share or mimic a mechanism involved in transactivation and transrepression of numerous cellular genes.
The region of IE62 required for interaction with Sp1 falls within the N-terminal third of the protein (amino acids 226299). This region overlaps with and/or borders on sequences involved in interactions with the viral IE4 protein and the cellular factors TBP and TFIIB (Fig. 4B). It is also within Finally we have shown that mutation of the Sp1 site results in little or no binding of either Sp1 or IE62 to the gI3.4 promoter whereas Sp1 binds to the wild-type promoter in the absence of IE62, suggesting that Sp1 aids in recruitment of IE62. Furthermore, the presence of IE62 enhances the binding of Sp1 and appears essential for the recruitment of TBP to the promoter in these assays. The strength of IE62-mediated activation of minimal promoters containing only TATA elements has been shown to be highly dependent on the sequence of these elements (54) and thus the atypical gI TATA-box (ATAAAA) (47, 51) may require the presence of both Sp1 and IE62 for expression. A possible model for activation of the VZV gI minimal promoter based on the data gathered in this study would be one in which IE62 and Sp1 interact independent of other viral and cellular factors. The possible existence of such a complex is supported by protein affinity pull-down assays which showed that interactions were observed at NaCl concentrations as high as 200 mM when the GST-62 (1406) fusion protein was used to capture Sp1 from uninfected cell extracts and as high as 300 mM when GST-Sp1 (612778) was used to capture full-length recombinant IE62 (data not shown). Stable co-immunoprecipitation of IE62 and Sp1 from infected cell extracts was also observed at NaCl concentrations up to 300 mM. Taken together, these data suggest that this tight association requires only the C-terminal portion of Sp1 and full-length IE62. The Sp1-IE62 complex would then bind to the AUS region of the gI promoter via the Sp1 site. Binding of this complex could be further strengthened through the interaction of IE62 with the promoter DNA. Binding of general transcription factor TFIID to the atypical gI TATA element then occurs and is stabilized by interactions between Sp1, IE62, TBP, and TAF 130, ultimately allowing initiation of transcription to occur. Alternatively, IE62 could bind to the TFIID complex, presumably through TBP, as a viral "TAF" and stabilize the virus-modified TFIID binding via interaction with Sp1 already bound to the AUS element. Other factors may aid in this process including, possibly, USF. Either of the above models opens up interesting possibilities regarding how IE62 activation of VZV promoters is controlled during the viral life cycle. As an example, the phosphorylation state of a threonine residue at position 250, which falls within the Sp1 binding site in IE62 (Fig. 4B) has been shown to be important for the interaction of IE62 and a second viral transcriptional activator, IE4. IE4 has been shown to act both synergistically with and independently of IE62 in transactivation of VZV promoters (4, 11, 62, 63). Binding by IE4 occurs almost exclusively to under-phosphorylated forms of IE62 as compared with more heavily phosphorylated forms of IE62 derived either from recombinant baculovirus or infected melanoma cells. Experiments using bacterially expressed fragments of IE62 have implicated phosphorylation of threonine 250 by PKC and/or PKA in this differential binding (49). In contrast, no obvious differences in binding were observed in the work presented here when either recombinant IE62 derived from baculovirus or bacterially expressed fragments of IE62 were used. Therefore Sp1 and IE4 may compete for the same site on IE62 depending on the phosphorylation state of IE62 and thus result in differential IE62 activation of specific viral genes. On a more general note, the phosphorylation state of Sp1 has recently been shown to be altered as the result of infection with HSV (64). It is quite likely that this also occurs upon VZV infection of susceptible cells. Since phosphorylation of Sp1 by a variety of cellular kinases has been shown to be involved in Sp1-mediated activation (reviewed in Ref. 65), it is possible that phosphorylation of Sp1 as a result of viral infection is also an important aspect of the regulation of IE62 activation of VZV gene expression. Both of these possibilities are currently under investigation.
* This work was supported by Grant AI18449 from the NIAID, National Institutes of Health. 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.
1 The abbreviations used are: VZV, varicella zoster virus; HSV-1, herpes simplex virus type 1; IE, immediate early; E, early; L, late; ORF, open reading frame; ICP, infected cell protein; AUS, activating upstream sequence; CAT, chloramphenicol acetyltransferase; PBS, phosphate-buffered saline; GST, glutathione S-transferase; IPTG, isopropyl-
2 M. Yang and W. T. Ruyechan, unpublished observations.
This article has been cited by other articles:
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||