trans-dominant mutants of E1A provide genetic evidence that the zinc finger of the trans-activating domain binds a transcription factor

Multiple domains in the 50 kDa form of E4F1 regulate promoter-specific repression and E1A trans-activation

The 50 kDa N-terminal product of the cellular transcription factor E4F1 (p50E4F1) mediates E1A289R trans-activation of the adenovirus E4 gene, and suppresses E1A-mediated transformation by sensitizing cells to cell death. This report shows that while both E1A289R and E1A243R stimulate p50E4F1 DNA binding activity, E1A289R trans-activation, as measured using GAL-p50E4F1 fusion proteins, involves a p50E4F1 transcription regulatory (TR) region that must be promoter-bound and is dependent upon E1A CR3, CR1 and N-terminal domains. Trans-activation is promoter-specific, as GAL-p50E4F1 did not stimulate commonly used artificial promoters and was strongly repressive when competing against GAL-VP16. p50E4F1 and E1A289R stably associate in vivo using the p50E4F1 TR region and E1A CR3, although their association in vitro is indirect and paradoxically disrupted by MAP kinase phosphorylation of E1A289R, which stimulates E4 trans-activation in vivo. Multiple cellular proteins, including TBP, bind the p50E4F1 TR region in vitro. The mechanistic implications for p50E4F1 function are discussed.

The Transcriptional Repressor BS69 is a Conserved Target of the E1A Proteins from Several Human Adenovirus Species

Early region 1A (E1A) is the first viral protein produced upon human adenovirus (HAdV) infection. This multifunctional protein transcriptionally activates other HAdV early genes and reprograms gene expression in host cells to support productive infection. E1A functions by interacting with key cellular regulatory proteins through short linear motifs (SLiMs). In this study, the molecular determinants of interaction between E1A and BS69, a cellular repressor that negatively regulates E1A transactivation, were systematically defined by mutagenesis experiments. We found that a minimal sequence comprised of MPNLVPEV, which contains a conserved PXLXP motif and spans residues 112–119 in HAdV-C5 E1A, was necessary and sufficient in binding to the myeloid, Nervy, and DEAF-1 (MYND) domain of BS69. Our study also identified residues P113 and L115 as critical for this interaction. Furthermore, the HAdV-C5 and -A12 E1A proteins from species C and A bound BS69, but those of HAdV-B3, -E4, -D9, -F40, and -G52 from species B, E, D, F, and G, respectively, did not. In addition, BS69 functioned as a repressor of E1A-mediated transactivation, but only for HAdV-C5 and HAdV-A12 E1A. Thus, the PXLXP motif present in a subset of HAdV E1A proteins confers interaction with BS69, which serves as a negative regulator of E1A mediated transcriptional activation.

Adenovirus E1A Activation Domain Regulates H3 Acetylation Affecting Varied Steps in Transcription at Different Viral Promoters

Despite a wealth of data associating promoter and enhancer region histone N-terminal tail lysine acetylation with transcriptional activity, there are relatively few examples of studies that establish causation between these histone posttranslational modifications and transcription. While hypoacetylation of histone H3 lysines 18 and 27 is associated with repression, the step(s) in the overall process of transcription that is blocked at a hypoacetylated promoter is not clearly established in most instances. Studies presented here confirm that the adenovirus 2 large E1A protein activation domain interacts with p300, as reported previously (P. Pelka, J. N. G. Ablack, J. Torchia, A. S. Turnell, R. J. A. Grand, J. S. Mymryk, Nucleic Acids Res 37:1095–1106, 2009, https://doi.org/10.1093/nar/gkn1057), and that the resulting acetylation of H3K18/27 affects varied steps in transcription at different viral promoters.

How histone acetylation promotes transcription is not clearly understood. Here, we confirm an interaction between p300 and the adenovirus 2 large E1A activation domain (AD) and map the interacting regions in E1A by observing colocalization at an integrated lacO array of fusions of LacI-mCherry to E1A fragments with YFP-p300. Viruses with mutations in E1A subdomains were constructed and analyzed for kinetics of early viral RNA expression and association of acetylated H3K9, K18, K27, TBP, and RNA polymerase II (Pol II) across the viral genome. The results indicate that this E1A interaction with p300 is required for H3K18 and H3K27 acetylation at the E2early, E3, and E4 promoters and is required for TBP and Pol II association with the E2early promoter. In contrast, H3K18/27 acetylation was not required for TBP and Pol II association with the E3 and E4 promoters but was required for E4 transcription at a step subsequent to Pol II preinitiation complex assembly.

IMPORTANCE Despite a wealth of data associating promoter and enhancer region histone N-terminal tail lysine acetylation with transcriptional activity, there are relatively few examples of studies that establish causation between these histone posttranslational modifications and transcription. While hypoacetylation of histone H3 lysines 18 and 27 is associated with repression, the step(s) in the overall process of transcription that is blocked at a hypoacetylated promoter is not clearly established in most instances. Studies presented here confirm that the adenovirus 2 large E1A protein activation domain interacts with p300, as reported previously (P. Pelka, J. N. G. Ablack, J. Torchia, A. S. Turnell, R. J. A. Grand, J. S. Mymryk, Nucleic Acids Res 37:1095–1106, 2009, https://doi.org/10.1093/nar/gkn1057), and that the resulting acetylation of H3K18/27 affects varied steps in transcription at different viral promoters.

Hacking the Cell: Network Intrusion and Exploitation by Adenovirus E1A

As obligate intracellular parasites, viruses are dependent on their infected hosts for survival. Consequently, viruses are under enormous selective pressure to utilize available cellular components and processes to their own advantage. As most, if not all, cellular activities are regulated at some level via protein interactions, host protein interaction networks are particularly vulnerable to viral exploitation. Indeed, viral proteins frequently target highly connected “hub” proteins to “hack” the cellular network, defining the molecular basis for viral control over the host. This widespread and successful strategy of network intrusion and exploitation has evolved convergently among numerous genetically distinct viruses as a result of the endless evolutionary arms race between pathogens and hosts. Here we examine the means by which a particularly well-connected viral hub protein, human adenovirus E1A, compromises and exploits the vulnerabilities of eukaryotic protein interaction networks. Importantly, these interactions identify critical regulatory hubs in the human proteome and help define the molecular basis of their function.

The adenoviral oncogene E1A-13S interacts with a specific isoform of the tumor suppressor PML to enhance viral transcription

PML nuclear bodies (PML NBs), also called ND10, are matrix-bound nuclear structures that have been implicated in a variety of functions, including DNA repair, transcriptional regulation, protein degradation, and tumor suppression. These domains are also known for their potential to mediate an intracellular defense mechanism against many virus types. This is likely why they are targeted and subsequently manipulated by numerous viral proteins. Paradoxically, the genomes of various DNA viruses become associated with PML NBs, and initial sites of viral transcription/replication centers are often juxtaposed to these domains. The question is why viruses start their transcription and replication next to their supposed antagonists. Here, we report that PML NBs are targeted by the adenoviral (Ad) transactivator protein E1A-13S. Alternatively spliced E1A isoforms (E1A-12S and E1A-13S) are the first proteins expressed upon Ad infection. E1A-13S is essential for activating viral transcription in the early phase of infection. Coimmunoprecipitation assays showed that E1A-13S preferentially interacts with only one (PML-II) of at least six nuclear human PML isoforms. Deletion mapping located the interaction site within E1A conserved region 3 (CR3), which was previously described as the transcription factor binding region of E1A-13S. Indeed, cooperation with PML-II enhanced E1A-mediated transcriptional activation, while deleting the SUMO-interacting motif (SIM) of PML proved even more effective. Our results suggest that in contrast to PML NB-associated antiviral defense, PML-II may help transactivate viral gene expression and therefore play a novel role in activating Ad transcription during the early viral life cycle.

Cellular GCN5 is a novel regulator of human adenovirus E1A-conserved region 3 transactivation

The largest isoform of adenovirus early region 1A (E1A) contains a unique region termed conserved region 3 (CR3). This region activates viral gene expression by recruiting cellular transcription machinery to the early viral promoters. Recent studies have suggested that there is an optimal level of E1A-dependent transactivation required by human adenovirus (hAd) during infection and that this may be achieved via functional cross talk between the N termini of E1A and CR3. The N terminus of E1A binds GCN5, a cellular lysine acetyltransferase (KAT). We have identified a second independent interaction of E1A with GCN5 that is mediated by CR3, which requires residues 178 to 188 in hAd5 E1A. GCN5 was recruited to the viral genome during infection in an E1A-dependent manner, and this required both GCN5 interaction sites on E1A. Ectopic expression of GCN5 repressed transactivation by both E1A CR3 and full-length E1A. In contrast, RNA interference (RNAi) depletion of GCN5 or treatment with the KAT inhibitor cyclopentylidene-[4-(4'-chlorophenyl)thiazol-2-yl]hydrazone (CPTH2) resulted in increased E1A CR3 transactivation. Moreover, activation of the adenovirus E4 promoter by E1A was increased during infection of homozygous GCN5 KAT-defective (hat/hat) mouse embryonic fibroblasts (MEFs) compared to wild-type control MEFs. Enhanced histone H3 K9/K14 acetylation at the viral E4 promoter required the newly identified binding site for GCN5 within CR3 and correlated with repression and reduced occupancy by phosphorylated RNA polymerase II. Treatment with CPTH2 during infection also reduced virus yield. These data identify GCN5 as a new negative regulator of transactivation by E1A and suggest that its KAT activity is required for optimal virus replication.

Comparison of E1A CR3-dependent transcriptional activation across six different human adenovirus subgroups

The largest E1A isoform of human adenovirus (Ad) includes a C-4 zinc finger domain within conserved region 3 (CR3) that is largely responsible for activating transcription of the early viral genes. CR3 interacts with multiple cellular factors, but its mechanism of action is modeled primarily on the basis of the mechanism for the prototype E1A protein of human Ad type 5. We expanded this model to include a representative member from each of the six human Ad subgroups. All CR3 domains tested were capable of transactivation. However, there were dramatic differences in their levels of transcriptional activation. Despite these functional variations, the interactions of these representative CR3s with known cellular transcriptional regulators revealed only modest differences. Four common cellular targets of all representative CR3s were identified: the proteasome component human Sug1 (hSug1)/S8, the acetyltransferases p300/CREB binding protein (CBP), the mediator component mediator complex subunit 23 (MED23) protein, and TATA binding protein (TBP). The first three factors appear to be critical for CR3 function. RNA interference against human TBP showed no significant reduction in transactivation by any CR3 tested. These results indicate that the cellular factors previously shown to be important for transactivation by Ad5 CR3 are similarly bound by the E1A proteins of other types. This was confirmed experimentally using a transcriptional squelching assay, which demonstrated that the CR3 regions of each Ad type could compete with Ad5 CR3 for limiting factors. Interestingly, a mutant of Ad5 CR3 (V147L) was capable of squelching wild-type Ad5 CR3, despite its failure to bind TBP, MED23, p300/CBP-associated factor (pCAF), or p300/CBP, suggestive of the possibility that an additional as yet unidentified cellular factor is required for transactivation by E1A CR3.

Requirements for E1A dependent transcription in the yeast Saccharomyces cerevisiae

BACKGROUND

The human adenovirus type 5 early region 1A (E1A) gene encodes proteins that are potent regulators of transcription. E1A does not bind DNA directly, but is recruited to target promoters by the interaction with sequence specific DNA binding proteins. In mammalian systems, E1A has been shown to contain two regions that can independently induce transcription when fused to a heterologous DNA binding domain. When expressed in Saccharomyces cerevisiae, each of these regions of E1A also acts as a strong transcriptional activator. This allows yeast to be used as a model system to study mechanisms by which E1A stimulates transcription.

RESULTS

Using 81 mutant yeast strains, we have evaluated the effect of deleting components of the ADA, COMPASS, CSR, INO80, ISW1, NuA3, NuA4, Mediator, PAF, RSC, SAGA, SAS, SLIK, SWI/SNF and SWR1 transcriptional regulatory complexes on E1A dependent transcription. In addition, we examined the role of histone H2B ubiquitylation by Rad6/Bre1 on transcriptional activation.

CONCLUSION

Our analysis indicates that the two activation domains of E1A function via distinct mechanisms, identify new factors regulating E1A dependent transcription and suggest that yeast can serve as a valid model system for at least some aspects of E1A function.

Transcriptional control by adenovirus E1A conserved region 3 via p300/CBP

The human adenovirus type 5 (HAdV-5) E1A 13S oncoprotein is a potent regulator of gene expression and is used extensively as a model for transcriptional activation. It possesses two independent transcriptional activation domains located in the N-terminus/conserved region (CR) 1 and CR3. The protein acetyltransferase p300 was previously identified by its association with the N-terminus/CR1 portion of E1A and this association is required for oncogenic transformation by E1A. We report here that transcriptional activation by 13S E1A is inhibited by co-expression of sub-stoichiometric amounts of the smaller 12S E1A isoform, which lacks CR3. Transcriptional inhibition by E1A 12S maps to the N-terminus and correlates with the ability to bind p300/CBP, suggesting that E1A 12S is sequestering this limiting factor from 13S E1A. This is supported by the observation that the repressive effect of E1A 12S is reversed by expression of exogenous p300 or CBP, but not by a CBP mutant lacking actyltransferase activity. Furthermore, we show that transcriptional activation by 13S E1A is greatly reduced by siRNA knockdown of p300 and that CR3 binds p300 independently of the well-characterized N-terminal/CR1-binding site. Importantly, CR3 is also required to recruit p300 to the adenovirus E4 promoter during infection. These results identify a new functionally significant interaction between E1A CR3 and the p300/CBP acetyltransferases, expanding our understanding of the mechanism by which this potent transcriptional activator functions.

Identification of a second CtBP binding site in adenovirus type 5 E1A conserved region 3

C-terminal binding protein (CtBP) binds to adenovirus early region 1A (AdE1A) through a highly conserved PXDLS motif close to the C terminus. We now have demonstrated that CtBP1 also interacts directly with the transcriptional activation domain (conserved region 3 [CR3]) of adenovirus type 5 E1A (Ad5E1A) and requires the integrity of the entire CR3 region for optimal binding. The interaction appears to be at least partially mediated through a sequence ((161)RRNTGDP(167)) very similar to a recently characterized novel CtBP binding motif in ZNF217 as well as other regions of CR3. Using reporter assays, we further demonstrated that CtBP1 represses Ad5E1A CR3-dependent transcriptional activation. Ad5E1A also appears to be recruited to the E-cadherin promoter through its interaction with CtBP. Significantly, Ad5E1A, CtBP1, and ZNF217 form a stable complex which requires CR3 and the PLDLS motif. It has been shown that Ad513SE1A, containing the CR3 region, is able to overcome the transcriptional repressor activity of a ZNF217 polypeptide fragment in a GAL4 reporter assay through recruitment of CtBP1. These results suggest a hitherto-unsuspected complexity in the association of Ad5E1A with CtBP, with the interaction resulting in transcriptional activation by recruitment of CR3-bound factors to CtBP1-containing complexes.

Roles for APIS and the 20S proteasome in adenovirus E1A-dependent transcription

We have determined distinct roles for different proteasome complexes in adenovirus (Ad) E1A-dependent transcription. We show that the 19S ATPase, S8, as a component of 19S ATPase proteins independent of 20S (APIS), binds specifically to the E1A transactivation domain, conserved region 3 (CR3). Recruitment of APIS to CR3 enhances the ability of E1A to stimulate transcription from viral early gene promoters during Ad infection of human cells. The ability of CR3 to stimulate transcription in yeast is similarly dependent on the functional integrity of yeast APIS components, Sug1 and Sug2. The 20S proteasome is also recruited to CR3 independently of APIS and the 26S proteasome. Chromatin immunoprecipitation reveals that E1A, S8 and the 20S proteasome are recruited to both Ad early region gene promoters and early region gene sequences during Ad infection, suggesting their requirement in both transcriptional initiation and elongation. We also demonstrate that E1A CR3 transactivation and degradation sequences functionally overlap and that proteasome inhibitors repress E1A transcription. Taken together, these data demonstrate distinct roles for APIS and the 20S proteasome in E1A-dependent transactivation.

Recent lessons in gene expression, cell cycle control, and cell biology from adenovirus

Adenovirus continues to be an important model system for investigating basic aspects of cell biology. Interactions of several cellular proteins with E1A conserved regions (CR) 1 and 2, and inhibition of apoptosis by E1B proteins are required for oncogenic transformation. CR2 binds RB family members, de-repressing E2F transcription factors, thus activating genes required for cell cycling. E1B-19K is a BCL2 homolog that binds and inactivates proapoptotic BAK and BAX. E1B-55K binds p53, inhibiting its transcriptional activation function. In productively infected cells, E1B-55K and E4orf6 assemble a ubiquitin ligase with cellular proteins Elongins B and C, Cullin 5 and RBX1 that polyubiquitinates p53 and one or more subunits of the MRN complex involved in DNA double-strand break repair, directing them to proteosomal degradation. E1A CR3 activates viral transcription by interacting with the MED23 Mediator subunit, stimulating preinitiation complex assembly on early viral promoters and probably also the rate at which they initiate transcription. The viral E1B-55K/E4orf6 ubiquitin ligase is also required for efficient viral late protein synthesis in many cell types, but the mechanism is not understood. E1A CR1 binds several chromatin-modifying complexes, but how this contributes to stimulation of cellular DNA synthesis and transformation is not clear. E1A CR4 binds the CtBP corepressor, but the mechanism by which this modulates the frequency of transformation remains to be determined. Clearly, adenovirus has much left to teach us about fundamental cellular processes.

Activation domain-mediator interactions promote transcription preinitiation complex assembly on promoter DNA

The interaction of activators with mediator has been proposed to stimulate the assembly of RNA polymerase II (Pol II) preinitiation complexes, but there have been few tests of this model. The finding that the major adenovirus E1A and mitogen-activated protein kinase-phosphorylated Elk1 activation domains bind to Sur2 uniquely among the metazoan mediator subunits and the development of transcriptionally active nuclear extracts from WT and sur2-/- embryonic stem cells, reported here, allowed a direct test of the model. We found that whereas VP16, E1A, and phosphorylated Elk1 activation domains each stimulate binding of mediator, Pol II, and general transcription factors to promoter DNA in extracts from WT cells, only VP16 stimulated their binding in extracts from sur2-/- cells. This stimulation of mediator, Pol II, and general transcription factor binding to promoter DNA correlated with transcriptional activation by these activators in WT and mutant extracts. Because the mutant mediator was active in reactions with the VP16 activation domain, the lack of activity in response to the E1A and Elk1 activation domains was not due to loss of a generalized mediator function, but rather the inability of the mutant mediator to be bound by E1A and Elk1. These results directly demonstrate that the interaction of activation domains with mediator stimulates preinitiation complex assembly on promoter DNA.

Molecular anatomy of Tupaia (tree shrew) adenovirus genome; evolution of viral genes and viral phylogeny

Adenoviruses are globally spread and infect species in all five taxons of vertebrates. Outstanding attention is focused on adenoviruses because of their transformation potential, their possible usability as vectors in gene therapy and their applicability in studies dealing with, e.g. cell cycle control, DNA replication, transcription, splicing, virus-host interactions, apoptosis, and viral evolution. The accumulation of genetic data provides the basis for the increase of our knowledge about adenoviruses. The Tupaia adenovirus (TAV) infects members of the genus Tupaiidae that are frequently used as laboratory animals in behavior research dealing with questions about biological and molecular processes of stress in mammals, in neurobiological and physiological studies, and as model organisms for human hepatitis B and C virus infections. In the present study the TAV genome underwent an extensive analysis including determination of codon usage, CG depletion, gene content, gene arrangement, potential splice sites, and phylogeny. The TAV genome has a length of 33,501 bp with a G+C content of 49.96%. The genome termini show a strong CG depletion that could be due to methylation of these genome regions during the viral replication cycle. The analysis of the coding capacity of the complete TAV genome resulted in the identification of 109 open reading frames (ORFs), of which 38 were predicted to be real viral genes. TAV was classified within the genus Mastadenovirus characterized by typical gene content, arrangement, and homology values of 29 conserved ORFs. Phylogenetic trees show that TAV is part of a separate evolutionary lineage and no mastadenovirus species can be considered as the most related. In contrast to other mastadenoviruses a direct ancestor of TAV captured a DUT gene from its mammalian host, presumably controlling local dUTP levels during replication and enhance viral replication in non-dividing host tissues. Furthermore, TAV possesses a second DNA-binding protein gene, that is likely to play a role in the determination of the host range. In view of these data it is conceivable that TAV underwent evolutionary adaptations to its biological environment resulting in the formation of special genomic components that provided TAV with the ability to expand its host range during viral evolution.

In vivo association of adenovirus large E1A protein with the human mediator complex in adenovirus-infected and -transformed cells

The adenovirus large E1A protein activates transcription from early viral promoters by a mechanism that requires a forty amino acid zinc finger activation domain in E1A conserved region 3 (CR3). Recent results indicate that activation by a Gal4 DNA-binding domain-E1A-CR3 fusion requires an interaction between the E1A-CR3 zinc finger and the Sur2 subunit of the mammalian Mediator (of transcription) complex. Although several host proteins have been shown to bind stably to E1A proteins in adenovirus-infected and -transformed cells, an in vivo interaction with Mediator complex subunits has not been described previously. Using immunoprecipitation and gel filtration analyses of nuclear extracts prepared from HeLa cells infected with adenovirus 5 or mutants that express either large or small E1A specifically and from adenovirus 5-transformed cells, we report here that large E1A, but not small E1A, binds to Mediator complex in vivo. Only approximately 1 to 10% of large E1A is bound to Mediator complex at 18 h postinfection and in transformed cells, probably explaining why Mediator complex subunits were not identified among cellular E1A-binding proteins described earlier. Surprisingly, even though extracted Mediator can quantitatively bind to an E1A-CR3 affinity column, only on the order of 1% of cellular Mediator complex is bound by E1A in vivo. Much of the large E1A bound to Mediator in 293 cells is in a stable complex that includes RNA polymerase II, leading us to suggest that the interaction of E1A-CR3 with Mediator stabilizes the interaction of Mediator with the polymerase. This stabilization of the interaction between Mediator and RNA polymerase II may contribute to the mechanism of activation by E1A-CR3.

Comparative sequence analysis of the largest E1A proteins of human and simian adenoviruses

The early region 1A (E1A) gene is the first gene expressed after infection with adenovirus and has been most extensively characterized in human adenovirus type 5 (hAd5). The E1A proteins interact with numerous cellular regulatory proteins, influencing a variety of transcriptional and cell cycle events. For this reason, these multifunctional proteins have been useful as tools for dissecting pathways regulating cell growth and gene expression. Despite the large number of studies using hAd5 E1A, relatively little is known about the function of the E1A proteins of other adenoviruses. In 1985, a comparison of E1A sequences from three human and one simian adenovirus identified three regions with higher overall levels of sequence conservation designated conserved regions (CR) 1, 2, and 3. As expected, these regions are critical for a variety of E1A functions. Since that time, the sequences of several other human and simian adenovirus E1A proteins have been determined. Using these, and two additional sequences that we determined, we report here a detailed comparison of the sequences of 15 E1A proteins representing each of the six hAd subgroups and several simian adenoviruses. These analyses refine the positioning of CR1, 2, and 3; define a fourth CR located near the carboxyl terminus of E1A; and suggest several new functions for E1A.

Activation of the Notch-regulated transcription factor CBF1/RBP-Jkappa through the 13SE1A oncoprotein

Signaling through the Notch pathway controls cell growth and differentiation in metazoans. Following binding of its ligands, the intracellular part of the cell surface Notch1 receptor (Notch1-IC) is released and translocates to the nucleus, where it alters the function of the DNA-binding transcription factor CBF1/RBP-Jkappa. As a result, CBF1/RBP-Jkappa is converted from a repressor to an activator of gene transcription. Similarly, the Epstein Barr viral oncoprotein EBNA2, which is required for B-cell immortalization, activates genes through CBF1. Moreover, the TAN-1 and int-3 oncogenes represent activated versions of Notch1 and Notch4, respectively. Here, we show that the adenoviral oncoprotein 13S E1A also binds to CBF1/RBP-Jkappa, displaces associated corepressor complexes, and activates CBF1/RBP-Jkappa-dependent gene expression. Our results suggest that the central role of the Notch-CBF1/RBP-Jkappa signaling pathway in cell fate decisions renders it susceptible to pathways of viral replication and oncogenic conversion.

The human factors YY1 and LSF repress the human immunodeficiency virus type 1 long terminal repeat via recruitment of histone deacetylase 1

Enigmatic mechanisms restore the resting state in activated lymphocytes following human immunodeficiency virus type 1 (HIV-1) infection, rarely allowing persistent nonproductive infection. We detail a mechanism whereby cellular factors could establish virological latency. The transcription factors YY1 and LSF cooperate in repression of transcription from the HIV-1 long terminal repeat (LTR). LSF recruits YY1 to the LTR via the zinc fingers of YY1. The first two zinc fingers were observed to be sufficient for this interaction in vitro. A mutant of LSF incapable of binding DNA blocked repression. Like other transcriptional repressors, YY1 can function via recruitment of histone deacetylase (HDAC). We find that HDAC1 copurifies with the LTR-binding YY1-LSF repressor complex, the domain of YY1 that interacts with HDAC1 is required to repress the HIV-1 promoter, expression of HDAC1 augments repression of the LTR by YY1, and the deacetylase inhibitor trichostatin A blocks repression mediated by YY1. This novel link between HDAC recruitment and inhibition of HIV-1 expression by YY1 and LSF, in the natural context of a viral promoter integrated into chromosomal DNA, is the first demonstration of a molecular mechanism of repression of HIV-1. YY1 and LSF may establish transcriptional and virological latency of HIV, a state that has recently been recognized in vivo and has significant implications for the long-term treatment of AIDS.

Inhibition of PrKX, a novel protein kinase, and the cyclic AMP-dependent protein kinase PKA by the regulatory proteins of adeno-associated virus type 2

Adeno-associated virus encodes four nonstructural proteins, which are known as Rep78, Rep68, Rep52, and Rep40. Expression of these nonstructural proteins affects cell growth and gene expression through processes that have not yet been characterized. Using a yeast two-hybrid screen, we have demonstrated that a stable interaction occurs between the viral proteins Rep78 and Rep52 and the putative protein kinase PrKX, which is encoded on the X chromosome. The stability and specificity of the Rep-PrKX interaction were confirmed by coimmunoprecipitation of complexes assembled in vitro and in vivo. Overexpressed PrKX, which was purified from cos cells, was shown to phosphorylate a synthetic protein kinase A (PKA) substrate. However, this activity was dramatically inhibited by stoichiometric amounts of Rep52 and weakly inhibited with Rep68, which lacks the carboxy-terminal sequence contained in Rep52. Similarly, a stable interaction was observed with Rep78, which also contains the carboxy-terminal sequence of Rep52. A stable interaction and inhibition were also observed between Rep52 and the catalytic subunit of PKA. By using surface plasmon resonance and kinetic studies, Kis of approximately 300 and 167 nM were calculated for Rep52 with PKA and with PrKX, respectively. Thus, Rep52 but not Rep68 can significantly inhibit the trans- and autophosphorylation activities of these kinases. The biological effects of Rep78-specific inhibition of PKA-responsive genes are illustrated by the reduction of steady-state levels of cyclic AMP-responsive-element-binding protein and cyclin A protein.

AR1 is an integral part of the adenovirus type 2 E1A-CR3 transactivation domain

We have previously shown that the nonconserved carboxy-terminal exon of the adenovirus type 2 E1A-289R protein contains two interchangeable sequence elements, auxiliary region (AR) 1 and AR2, that are required for efficient CR3-mediated transcriptional activation of the viral E4 promoter (M. Bondesson, C. Svensson, S. Linder, and G. Akusjärvi, EMBO J. 11:3347-3354, 1992). Here we show that CR3-mediated transactivation of all adenovirus early promoters and the HSP70 promoter requires the AR1 element. We further show that AR2 can substitute for AR1 only when artificially juxtaposed to CR3. AR1 consists of six tandem glutamic acid-proline (EP) repeats and is positioned immediately downstream of CR3. Genetic dissection of AR1 showed that the number of EP repeats in AR1 is critical for CR3 function. Thus, reducing or increasing the number of EP repeats reduces the CR3 transactivation capacity. Furthermore, the introduction of amino acid substitutions into AR1 suggested that the net negative charge in AR1 is of critical importance for its function as an enhancer of CR3-mediated transcriptional activation. Using an in vitro binding approach, we showed that the AR1 element is not part of the CR3 promoter localization signal mediating contact with the Sp1, ATF-2, or c-Jun upstream-binding transcription factors. Previous studies have suggested that the 49-amino-acid sequence constituting CR3 represents the minimal domain required for E1A-induced activation of viral early promoters. Since AR1 was required for efficient CR3-mediated transcriptional activation of all tested promoters, we suggest that the carboxy-terminal boundary for the CR3 transactivation domain should be extended to include the AR1 element.

A novel genetic system to isolate a dominant negative effector on DNA-binding activity of Oct-2

Recent studies have revealed that interactions between transcription factors play an important role in regulation of gene expression in eukaryotic cells. To isolate cDNA clones that dominantly inhibit the DNA-binding activity of Oct-2, chosen as a representative factor, we have developed a novel screening system. This employs an Escherichia coli tester strain carrying a modified lac operon as a reporter gene, with the lac operator sequence replaced by an octamer sequence. Oct-2 expressed in this tester strain represses the expression of the reporter gene and changes the phenotype of the cell from Lac+to Lac-. Introduction of a cDNA expression library prepared from a human T-cell line into the Oct-2-harboring tester strain allowed selection of three Lac+clones out of 1 x 10(5) transformants. One of them, hT86, encoding a putative zinc finger protein was found to derepress beta-galactosidase activity in the Oct-2-harboring tester strain at the transcriptional level. In gel mobility shift assays, hT86 attenuated the intensity of the retarded band composed of the octamer probe and Oct-2, suggesting a dominant negative effect on the DNA-binding activity of Oct-2. The strategy described here provides a new approach for studying protein-protein interactions that govern the complex regulation of gene expression.

Phosphorylation within the transactivation domain of adenovirus E1A protein by mitogen-activated protein kinase regulates expression of early region 4

A critical role of the 289-residue (289R) E1A protein of human adenovirus type 5 during productive infection is to transactivate expression of all early viral transcription. Sequences within and proximal to conserved region 3 (CR3) promote expression of these viral genes through interactions with a variety of transcription factors requiring the zinc binding motif in CR3 and in some cases a region at the carboxy-terminal end of CR3, including residues 183 to 188. It is known that 3',5' cyclic AMP (cAMP) reduces the level of phosphorylation of the 289R E1A protein through the activation of protein phosphatase 2A by the E4orf4 protein. This study was designed to identify the E1A phosphorylation sites affected by E4orf4 expression and to determine their importance in regulation of E1A activity. We report here that two previously unidentified sites at Ser-185 and Ser-188 are the targets for decreased phosphorylation in response to cAMP. At least one of these sites, presumably Ser-185, is phosphorylated in vitro by purified mitogen-activated protein kinase (MAPK), and both are hyperphosphorylated in cells which express a constitutively active form of MAPK kinase. Analysis of E1A-mediated transactivation activity indicated that elevated phosphorylation at these sites increased expression of the E4 promoter but not that of E3. We have recently shown that one or more E4 products induce cell death due to p53-independent apoptosis, and thus it seems likely that one role of the E4orf4 protein is to limit production of toxic E4 products by limiting expression of the E4 promoter.

Adenovirus E4 open reading frame 4 protein autoregulates E4 transcription by inhibiting E1A transactivation of the E4 promoter

Here we show that the adenovirus early region 4 (E4) open reading frame 4 (ORF4) protein autoregulates its own transcription by inhibiting adenovirus E1A-induced activation of E4 transcription both in transient transfection experiments and during lytic virus growth. The inhibitory activity of E4-ORF4 was selective for E1A-CR3-dependent transactivation and had no effect on CR1 transactivation. The inhibitory activity of E4-ORF4 was relieved by okadaic acid treatment, which inhibits the cellular protein phosphatase 2A (PP2A), suggesting that E4-ORF4 controls the phosphorylated status of transcription factors important for E4 promoter activity. This conclusion agrees with previous demonstrations that E4-ORF4 associates with PP2A and causes a partial dephosphorylation of certain transcription factors, including E1A (U. Müller, T. Kleinberger, and T. Shenk, J. Virol. 66:5869-5878, 1992; T. Kleinberger and T. Shenk, J. Virol. 67:7556-7560, 1993). However, our results indicate that dephosphorylation of E1A itself might not be the primary target for E4-ORF4. Instead, the E4-ORF4-PP2A complex appears to work by dephosphorylation of multiple cellular transcription factors that are involved in E1A transactivation of the E4 promoter.

Mutations in the zinc-binding motif of the reovirus capsid protein delta 3 eliminate its ability to associate with capsid protein mu 1

Reovirus capsid protein delta 3 binds both double-stranded RNA (dsRNA) and zinc. Previous studies have revealed that the amino-terminal zinc finger is not required for the ability of delta 3 to bind dsRNA. We expressed wild-type and mutant delta 3 molecules by in vitro transcription/translation to evaluate the importance of the zinc finger for other functions of delta 3. delta 3 molecules with mutations in the zinc finger did not form complexes with capsid protein mu 1 but bound dsRNA more efficiently than wild-type delta 3 did. In contrast, a dsRNA-binding mutant was unimpaired in its ability to associate with mu 1. Studies with delta 3 fragments support these findings and indicate that sequences critical for delta 3's interaction with mu 1 lie in the amino terminus of the molecule. Our finding that mu 1 and dsRNA do not compete for identical binding sites on delta 3 has implications for its function as a translational regulator in infected cells.

Subregions of the adenovirus E1A transactivation domain target multiple components of the TFIID complex

Transcriptional activation by the adenovirus E1A 289R protein requires direct contacts with the TATA box-binding protein (TBP) and also displays a critical requirement for TBP-associated factors (TAFs) (T.G. Boyer and A. J. Berk, Genes Dev. 7:1810-1823, 1993; J. V. Geisberg, W. S. Lee, A. J. Berk, and R. P. Ricciardi, Proc. Natl. Acad. Sci. USA 91:2488-2492, 1994; W. S. Lee, C. C. Kao, G. O. Bryant, X. Liu, and A. J. Berk, Cell 67:365-376, 1991; and Q. Zhou, P. M. Lieberman, T. G. Boyer, and A. J. Berk, Genes Dev. 6:1964-1974, 1992). In this report, we demonstrate that the activation domain of E1A (CR3) specifically binds to two TAFs, human TAFII250 (hTAFII250) and Drosophila TAFII110 (dTAFII110). These interactions can take place both in vivo and in vitro and require the carboxy-terminal region of CR3; the zinc finger region of CR3, which binds TBP, is not needed to bind these TAFs. We mapped the E1A-binding sites on hTAFII250 to an internal region that contains a number of structural motifs, including an HMG box, a bromodomain, and direct repeats. This represents the first demonstration that hTAFII250 may serve as a target of a transcriptional activator. We also mapped the E1A binding on dTAFII110 to its C-terminal region. This is of significance since, by contrast, Sp1-mediated activation requires binding to the N-terminal domain of dTAFII110. Thus, distinct surfaces of dTAFII110 can serve as target sites for different activators. Our results indicate that E1A may activate transcription, in part, through direct contacts of the CR3 subdomains with selected components of the TFIID complex.

Adenovirus E1A functions as a cofactor for retinoic acid receptor beta (RAR beta) through direct interaction with RAR beta

Transcription regulation by DNA-bound activators is thought to be mediated by a direct interaction between these proteins and TATA-binding protein (TBP), TFIIB, or TBP-associated factors, although occasionally cofactors or adapters are required. For ligand-induced activation by the retinoic acid receptor-retinoid X receptor (RAR-RXR) heterodimer, the RAR beta 2 promoter is dependent on the presence of E1A or E1A-like activity, since this promoter is activated by retinoic acid only in cells expressing such proteins. The mechanism underlying this E1A requirement is largely unknown. We now show that direct interaction between RAR and E1A is a requirement for retinoic acid-induced RAR beta 2 activation. The activity of the hormone-dependent activation function 2 (AF-2) of RAR beta is upregulated by E1A, and an interaction between this region and E1A was observed, but not with AF-1 or AF-2 of RXR alpha. This interaction is dependent on conserved region III (CRIII), the 13S mRNA-specific region of E1A. Deletion analysis within this region indicated that the complete CRIII is needed for activation. The putative zinc finger region is crucial, probably as a consequence of interaction with TBP. Furthermore, the region surrounding amino acid 178, partially overlapping with the TBP binding region, is involved in both binding to and activation by AF-2. We propose that E1A functions as a cofactor by interacting with both TBP and RAR, thereby stabilizing the preinitiation complex.

trans-dominant interference of type 5 adenovirus E1a mutants in cell transformation

Two type 5 adenovirus (Ad5) early region 1a (E1a) mutants, H5in104 and H5dl105, were impaired in viral replication and cell transformation. In addition, these mutants trans dominantly inhibited the frequency with which H5sub309, a phenotypically wild-type mutant, and H5dl520, a high-frequency transformation mutant, transformed CREF cells. Inhibition of transformation varied in proportion to the input ratio of mutant to coinfecting virus. It was found that H5in104, but not H5dl105, could not complement Ad5 E1b mutants that failed to synthesize 19- or 55-kDa E1b product. H5dl105 yielded 10-fold less virus than the wild-type did in 293 cells, which constitutively express E1a and E1b products; similar low yields were also observed with H5in104 and H5dl105 in another E1a- and E1b-expressing transformed cell line, KB16. Marker rescue and DNA sequence analyses, however, indicated that the phenotypes of H5in104 and H5dl105 were the result of their respective E1a mutations. The data presented are the first to demonstrate that mutants of animal viruses can effect dominant interference with the viral function(s) that produce cell transformation.

The zinc finger region of the adenovirus E1A transactivating domain complexes with the TATA box binding protein

The 289R E1A protein of adenovirus transactivates a variety of viral and cellular promoters through protein-protein interactions. In earlier studies, mutational analyses of the E1A transactivating domain identified residues that are critical for transactivation and implied that the zinc finger region of the transactivating domain binds a transcription factor. Also, the E1A activation domain was found to bind to the TATA box binding protein (TBP) in vitro. Here, we tested the significance of the E1A-TBP interaction for E1A transactivation by analyzing the effects of conservative substitutions at each of the 49 residues of the E1A activation domain. Seven of the substitutions significantly diminished TBP binding in vitro. All of these were in the zinc finger region and were defective for transactivation in vivo. The perfect correlation between reduced TBP binding and transactivation argues strongly that a direct interaction between the E1A activation domain and TBP is critical to the mechanism of E1A activation. This genetic analysis leads us to further suggest that another factor, which is limiting, is also necessary for E1A-mediated transactivation.

Modulation of transcriptional activation of the proliferating cell nuclear antigen promoter by the adenovirus E1A 243-residue oncoprotein depends on proximal activators

Previous analyses defined a proliferating cell nuclear antigen (PCNA) E1A-responsive element (PERE) in the PCNA promoter that is essential for transactivation by the 243-residue product of the adenovirus type 2 E1A 12S mRNA (E1A 243R). In this report, we show that the PERE activates a heterologous basal promoter and confers susceptibility to transactivation by E1A 243R, indicating that the PERE is both necessary and sufficient for the response of the PCNA promoter to this oncoprotein. Insertion of linker sequences between the PERE and the site of transcription initiation in the PCNA promoter severely impairs the promoter's response to E1A 243R transactivation. GAL4 sites can replace the function of the PERE in the E1A 243R response of the PCNA basal promoter if transcriptional activators of suitable strength are supplied as GAL4 fusion proteins. Weak transcriptional activators render the PCNA basal promoter subject to transactivation by E1A 243R but do not endow the adenovirus E1B basal promoter with a similar response. Strong transcriptional activators do not support transactivation by E1A 243R, however; instead, E1A reduces the ability of the strong activators to activate both the PCNA and E1B basal promoters. Although other mechanistic differences might determine the response, the data imply a relationship between the activation strength of promoter-proximal effectors and the response of the PCNA basal promoter to E1A 243R. These experiments indicate that the PERE can function autonomously in mediating transactivation by E1A 243R and that the PCNA basal promoter is configured in a manner that permits modulation by E1A 243R of transcriptional activation by promoter-proximal effectors.

Modulation of transcriptional activation of the proliferating cell nuclear antigen promoter by the adenovirus E1A 243-residue oncoprotein depends on proximal activators

Previous analyses defined a proliferating cell nuclear antigen (PCNA) E1A-responsive element (PERE) in the PCNA promoter that is essential for transactivation by the 243-residue product of the adenovirus type 2 E1A 12S mRNA (E1A 243R). In this report, we show that the PERE activates a heterologous basal promoter and confers susceptibility to transactivation by E1A 243R, indicating that the PERE is both necessary and sufficient for the response of the PCNA promoter to this oncoprotein. Insertion of linker sequences between the PERE and the site of transcription initiation in the PCNA promoter severely impairs the promoter's response to E1A 243R transactivation. GAL4 sites can replace the function of the PERE in the E1A 243R response of the PCNA basal promoter if transcriptional activators of suitable strength are supplied as GAL4 fusion proteins. Weak transcriptional activators render the PCNA basal promoter subject to transactivation by E1A 243R but do not endow the adenovirus E1B basal promoter with a similar response. Strong transcriptional activators do not support transactivation by E1A 243R, however; instead, E1A reduces the ability of the strong activators to activate both the PCNA and E1B basal promoters. Although other mechanistic differences might determine the response, the data imply a relationship between the activation strength of promoter-proximal effectors and the response of the PCNA basal promoter to E1A 243R. These experiments indicate that the PERE can function autonomously in mediating transactivation by E1A 243R and that the PCNA basal promoter is configured in a manner that permits modulation by E1A 243R of transcriptional activation by promoter-proximal effectors.

Estradiol-inducible squelching and cell growth arrest by a chimeric VP16-estrogen receptor expressed in Saccharomyces cerevisiae: suppression by an allele of PDR1

We have constructed and characterized a flexible system for analyzing the phenomenon of squelching and estrogen receptor function in the yeast Saccharomyces cerevisiae. The A/B region of the human estrogen receptor was replaced with the transcriptional activating domain of VP16 and expressed in yeast cells from high-copy-number plasmids. Addition of hormone resulted in an immediate inhibition of expression (squelching) of a chromosomally integrated GAL1:lacZ reporter gene and the eventual arrest of cell growth (toxicity). In order to determine whether a relationship exists between toxicity and squelching, mutations were made in this chimeric receptor (VEO) and their effects on transcriptional activation, squelching, and toxicity were compared. A direct correlation was found between mutations in VEO that reduced VP16 transactivation ability in yeast cells and those that reduced both squelching and toxicity. Surprisingly, mutations in the DNA binding domain (DBD) of VEO dramatically reduced squelching and completely relieved toxicity, suggesting a role for the DBD in squelching and strengthening the correlation between squelching and toxicity. To demonstrate the utility of this system for carrying out genetic selection, a plasmid-based yeast genomic bank was screened for genes that can relieve the toxicity of VEO by means of an elevated copy number, resulting in the repeated cloning of an allele of the PDR1 (pleiotropic drug resistance) gene. We present evidence that mutations in PDR1 can modulate the intracellular availability of estradiol by the same mechanism that leads to multiple drug resistance in yeast cells. Taken together, our results provide evidence that cell growth arrest occurs when squelching exceeds a certain threshold and that strong squelching requires both a DBD and a transcriptional activating domain. Furthermore, we show that growth arrest can provide a useful phenotype for carrying out the genetic analysis of both squelching and estrogen receptor function in yeast cells.

Transcriptional activation by the adenovirus larger E1a product is mediated by members of the cellular transcription factor ATF family which can directly associate with E1a

We recently isolated three cDNA clones encoding closely related proteins (ATFa1, ATFa2, and ATFa3) that belong to the activating transcription factor-cyclic AMP-responsive element family of cellular transcription factors. Using cotransfection experiments, we showed that these proteins mediate the transcriptional activation induced by the adenovirus E1a 13S mRNA gene product and that the zinc-binding domains present in both E1a conserved region 3 and the most N-terminal portion of the ATFa proteins play crucial roles in this activity. Reciprocal coimmunoprecipitation experiments demonstrated direct interactions between these proteins. Neither the conserved region 3 domain of E1a nor the N-terminal metal-binding element of ATFa is essential for these interactions. The simultaneous alteration of both the N-terminal and the C-terminal domains of ATFa abolished E1a binding, while either mutation alone failed to impair these interactions.

Estradiol-inducible squelching and cell growth arrest by a chimeric VP16-estrogen receptor expressed in Saccharomyces cerevisiae: suppression by an allele of PDR1

We have constructed and characterized a flexible system for analyzing the phenomenon of squelching and estrogen receptor function in the yeast Saccharomyces cerevisiae. The A/B region of the human estrogen receptor was replaced with the transcriptional activating domain of VP16 and expressed in yeast cells from high-copy-number plasmids. Addition of hormone resulted in an immediate inhibition of expression (squelching) of a chromosomally integrated GAL1:lacZ reporter gene and the eventual arrest of cell growth (toxicity). In order to determine whether a relationship exists between toxicity and squelching, mutations were made in this chimeric receptor (VEO) and their effects on transcriptional activation, squelching, and toxicity were compared. A direct correlation was found between mutations in VEO that reduced VP16 transactivation ability in yeast cells and those that reduced both squelching and toxicity. Surprisingly, mutations in the DNA binding domain (DBD) of VEO dramatically reduced squelching and completely relieved toxicity, suggesting a role for the DBD in squelching and strengthening the correlation between squelching and toxicity. To demonstrate the utility of this system for carrying out genetic selection, a plasmid-based yeast genomic bank was screened for genes that can relieve the toxicity of VEO by means of an elevated copy number, resulting in the repeated cloning of an allele of the PDR1 (pleiotropic drug resistance) gene. We present evidence that mutations in PDR1 can modulate the intracellular availability of estradiol by the same mechanism that leads to multiple drug resistance in yeast cells. Taken together, our results provide evidence that cell growth arrest occurs when squelching exceeds a certain threshold and that strong squelching requires both a DBD and a transcriptional activating domain. Furthermore, we show that growth arrest can provide a useful phenotype for carrying out the genetic analysis of both squelching and estrogen receptor function in yeast cells.

Transcriptional activation by the adenovirus larger E1a product is mediated by members of the cellular transcription factor ATF family which can directly associate with E1a

We recently isolated three cDNA clones encoding closely related proteins (ATFa1, ATFa2, and ATFa3) that belong to the activating transcription factor-cyclic AMP-responsive element family of cellular transcription factors. Using cotransfection experiments, we showed that these proteins mediate the transcriptional activation induced by the adenovirus E1a 13S mRNA gene product and that the zinc-binding domains present in both E1a conserved region 3 and the most N-terminal portion of the ATFa proteins play crucial roles in this activity. Reciprocal coimmunoprecipitation experiments demonstrated direct interactions between these proteins. Neither the conserved region 3 domain of E1a nor the N-terminal metal-binding element of ATFa is essential for these interactions. The simultaneous alteration of both the N-terminal and the C-terminal domains of ATFa abolished E1a binding, while either mutation alone failed to impair these interactions.

Analysis of a developmentally regulated nuclear localization signal in Xenopus

The 289 residue nuclear oncoprotein encoded by the adenovirus 5 Ela gene contains two peptide sequences that behave as nuclear localization signals (NLS). One signal, located at the carboxy terminus, is like many other known NLSs in that it consists of a short stretch of basic residues (KRPRP) and is constitutively active in cells. The second signal resides within an internal 45 residue region of E1a that contains few basic residues or sequences that resemble other known NLSs. Moreover, this internal signal functions in injected Xenopus oocytes, but not in transfected Xenopus A6 cells, suggesting that it could be regulated developmentally (Slavicek et al. 1989. J. Virol. 63:4047). In this study, we show that the activity of this signal is sensitive to ATP depletion in vivo, efficiently directs the import of a 50 kD fusion protein and can compete with the E1a carboxy-terminal NLS for nuclear import. In addition, we have delineated the precise amino acid residues that comprise the second E1a NLS, and have assessed its utilization during Xenopus embryogenesis. Using amino acid deletion and substitution analyses, we show that the signal consists of the sequence FV(X)7-20MXSLXYM(X)4MF. By expressing in Xenopus embryos a truncated E1a protein that contains only the second NLS and by monitoring its cytoplasmic/nuclear distribution during development with indirect immunofluorescence, we find that the second NLS is utilized up to the early neurula stage. In addition, there appears to be a hierarchy among the embryonic germ layers as to when the second NLS becomes nonfunctional. For this reason, we refer to this NLS as the developmentally regulated nuclear localization signal (drNLS). The implications of these findings for early development are discussed.

The hsp70 gene CCAAT-binding factor mediates transcriptional activation by the adenovirus E1a protein

Expression of the human hsp70 gene is cell cycle regulated and is inducible by both serum and the adenovirus E1a protein (K. Milarski and R. Morimoto, Proc. Natl. Acad. Sci. USA 83:9517-9521, 1986; M. C. Simon, K. Kitchener, H.-T. Kao, E. Hickey, L. Weber, R. Voellmy, N. Heintz, and J. R. Nevins, Mol. Cell. Biol. 7:2884-2890, 1987; B. Wu, H. Hurst, N. Jones, and R. Morimoto, Mol. Cell. Biol. 6:2994-2999, 1986; B. Wu and R. Morimoto, Proc. Natl. Acad. Sci. USA 82:6070-6074, 1985). This regulated expression is predominantly controlled by the CCAAT element at position -70 relative to the transcriptional initiation site (G. Williams, T. McClanahan, and R. Morimoto, Mol. Cell. Biol. 9:2574-2587, 1989; B. Wu, H. Hurst, N. Jones, and R. Morimoto, Mol. Cell. Biol. 6:2994-2999, 1986). A corresponding CCAAT-binding factor (CBF) of 999 amino acids has recently been cloned and shown to stimulate transcription selectively from the hsp70 promoter in a CCAAT element-dependent manner (L. Lum, L. Sultzman, R. Kaufman, D. Linzer, and B. Wu, Mol. Cell. Biol. 10:6709-6717, 1990). We report here that the first 192 residues of CBF, when fused to the DNA-binding domain of the heterologous activator GAL-4, are necessary and sufficient to mediate E1a-dependent transcriptional activation. E1a and CBF exhibit complex formation in vitro, suggesting that an in vivo interaction between these proteins may be relevant to the well-characterized E1a-induced transcriptional activation of the hsp70 promoter.

The hsp70 gene CCAAT-binding factor mediates transcriptional activation by the adenovirus E1a protein

Expression of the human hsp70 gene is cell cycle regulated and is inducible by both serum and the adenovirus E1a protein (K. Milarski and R. Morimoto, Proc. Natl. Acad. Sci. USA 83:9517-9521, 1986; M. C. Simon, K. Kitchener, H.-T. Kao, E. Hickey, L. Weber, R. Voellmy, N. Heintz, and J. R. Nevins, Mol. Cell. Biol. 7:2884-2890, 1987; B. Wu, H. Hurst, N. Jones, and R. Morimoto, Mol. Cell. Biol. 6:2994-2999, 1986; B. Wu and R. Morimoto, Proc. Natl. Acad. Sci. USA 82:6070-6074, 1985). This regulated expression is predominantly controlled by the CCAAT element at position -70 relative to the transcriptional initiation site (G. Williams, T. McClanahan, and R. Morimoto, Mol. Cell. Biol. 9:2574-2587, 1989; B. Wu, H. Hurst, N. Jones, and R. Morimoto, Mol. Cell. Biol. 6:2994-2999, 1986). A corresponding CCAAT-binding factor (CBF) of 999 amino acids has recently been cloned and shown to stimulate transcription selectively from the hsp70 promoter in a CCAAT element-dependent manner (L. Lum, L. Sultzman, R. Kaufman, D. Linzer, and B. Wu, Mol. Cell. Biol. 10:6709-6717, 1990). We report here that the first 192 residues of CBF, when fused to the DNA-binding domain of the heterologous activator GAL-4, are necessary and sufficient to mediate E1a-dependent transcriptional activation. E1a and CBF exhibit complex formation in vitro, suggesting that an in vivo interaction between these proteins may be relevant to the well-characterized E1a-induced transcriptional activation of the hsp70 promoter.