Inhibition of cell proliferation by an RNA ligand that selectively blocks E2F function.

The control of cell proliferation is of central importance to the proper development of a multicellular organism, the homeostatic maintenance of tissues, and the ability of certain cell types to respond appropriately to environmental cues. Disruption of normal cell growth control underlies many pathological conditions, including endothelial proliferative disorders in cardiovascular disease as well as the development of malignant tumors. Particularly critical for the control of cell growth is the pathway involving the G1 cyclin-dependent kinases that regulate the Rb family of proteins, which in turn control E2F transcription factor activity. Because E2F is critical for regulation of cell proliferation, we sought to identify and to develop specific inhibitors of E2F function that might also be useful in the control of cellular proliferation. Moreover, because the control of E2F activity appears to be the end result of G1 regulatory cascades, the ability to inhibit E2F may be particularly effective in impeding a wide variety of proliferative events. We have used in vitro selection to isolate several unique RNA species from high complexity RNA libraries that avidly bind to the E2F family of proteins. These RNAs also inhibit the DNA binding capacity of the E2F proteins. We also show that an E2F RNA ligand can block the induction of S phase in quiescent cells stimulated by serum addition. As such, these data demonstrate the critical role for E2F activity in cell proliferation and suggest that such RNA molecules may be effective as therapeutic entities to control cellular proliferation.

Addition of poly(A) to nuclear RNA occurs soon after RNA synthesis.

A kinetic analysis of the appearance of [3H]uridine label in RNA sequences that neighbor poly(A), as well as the incorporation of [3H]adenosine label into both the RNA chain and the poly(A) of poly(A)-containing molecules, shows that poly(A) is added within a minute or so after RNA chain synthesis in Chinese hamster ovary cells and HeLa cells. Previous conclusions by several groups (5-7) that poly(A) might be added as long as 20-30 min after RNA synthesis appear to be in error, and the present conclusion seems much more in line with several different types of recent studies with specific mRNAs that suggest prompt poly(A) addition (13-16).

Control of hsp70 RNA levels in human lymphocytes.

The expression of a hsp70 gene in human cells has previously been shown to be related to the growth state of the cells. As an alternative to in vitro synchronization procedures, we have measured steady-state levels of the RNA for a heat-shock protein 70 (hsp70) in human peripheral blood mononuclear cells (PBMC) that are naturally quiescent in a G0 state. The probe used recognized, on RNA blots, one single band. The levels of this hsp70 RNA are elevated in circulating PBMC and decrease when the cells are incubated with serum, or phytohemagglutinin, or simply when they are incubated in culture medium. The levels of hsp70 RNA decrease within 30 min after in vitro culture, and are accompanied by an increase in the levels of c-fos RNA. These findings, together with other recent reports in the literature, suggest a possible role of the hsp70 proteins in the regulation of cell growth.

E2F: a link between the Rb tumor suppressor protein and viral oncoproteins.

The cellular transcription factor E2F, previously identified as a component of early adenovirus transcription, has now been shown to be important in cell proliferation control. E2F appears to be a functional target for the action of the tumor suppressor protein Rb that is encoded by the retinoblastoma susceptibility gene. The disruption of this E2F-Rb interaction, as well as a complex involving E2F in association with the cell cycle-regulated cyclin A-cdk2 kinase complex, may be a common mechanism of action for the oncoproteins encoded by the DNA tumor viruses.

E1A transcription induction: enhanced binding of a factor to upstream promoter sequences.

The adenovirus E1A gene product trans-activates a number of viral and cellular promoters. The mechanism for this transcriptional induction was investigated with an in vivo exoIII mapping technique to assay for proteins that interact with an E1A-inducible promoter. A protein bound to the early E2 promoter was detected in wild-type infected cells. In the absence of E1A induction, specific interactions at the promoter could not be detected, as indicated by the absence of an exoIII-protected fragment. However, if conditions were established that allowed transcription of the E2 gene in the absence of E1A, the same exoIII protection was observed as was found in the presence of E1A. These results suggest a model in which the efficient utilization of the E2 promoter is mediated by a cellular transcription factor. In the absence of E1A, the interaction can take place, but slowly and inefficiently in comparison with the interaction in the presence of E1A.

Transcriptional activation by viral regulatory proteins.

The control of transcription involves the use of many transcriptional regulatory proteins. Viral systems and proteins have been used as models to gain insight into these control processes. These include the adenovirus E1A13S and E1A12S products and the herpes virus VP16 protein. This review examines these diverse mechanisms, but also explores the elements of commonality between them.

Compensation and specificity of function within the E2F family.

Functions encoded by single genes in lower organisms are often represented by multiple related genes in the mammalian genome. An example is the retinoblastoma and E2F families of proteins that regulate transcription during the cell cycle. Analysis of gene function using germline mutations is often confounded by overlapping function resulting in compensation. Indeed, in cells deleted of the E2F1 or E2F3 genes, there is an increase in the expression of the other family member. To avoid complications of compensatory effects, we have used small-interfering RNAs that target individual E2F proteins to generate a temporary loss of E2F function. We find that both E2F1 and E2F3 are required for cells to enter the S phase from a quiescent state, whereas only E2F3 is necessary for the S phase in growing cells. We also find that the acute loss of E2F3 activity affects the expression of genes encoding DNA replication and mitotic activities, whereas loss of E2F1 affects a limited number of genes that are distinct from those regulated by E2F3. We conclude that the long-term loss of E2F activity does lead to compensation by other family members and that the analysis of acute loss of function reveals specific and distinct roles for these proteins.

Transcriptional activation and subsequent control of the human heat shock gene during adenovirus infection.

A cDNA copy of the major human heat shock mRNA was cloned. The clone is complementary to the mRNA encoding the major 70-kilodalton heat shock protein as shown by hybrid arrest translation. We utilized the cloned DNA to measure induction of the gene during adenovirus infection. The mRNA increases in abundance approximately 100-fold during a wild-type adenovirus infection but does not increase more than 2-fold during an infection in which there is no E1A gene function [high multiplicity of infection of an E1A (-) mutant]. Furthermore, by measuring transcription in isolated nuclei, we found that the induction was transcriptional and was mediated by the E1A gene product. The induction was not maintained, however. After a peak level was obtained, transcription returned to preinfection levels. This decline was also reflected in the cytoplasmic mRNA abundance indicating a rapid turnover of the heat shock mRNA. This rapid turnover of the heat shock mRNA appears to be induced by the viral infection since the heat shock mRNA was found to be stable when synthesized in an adenovirus-transformed cell line.

Localization of the adenovirus E1Aa protein, a positive-acting transcriptional factor, in infected cells infected cells.

The function of the adenovirus E1Aa protein (the product of the 13S E1A mRNA) during a productive viral infection is to activate transcription of the six early viral transcription units. To study the mechanism of action of this protein, a peptide which was 13 amino acids long and had a sequence unique to the protein product of the adenovirus 13S E1A mRNA (pE1Aa) was coupled to keyhole limpet hemocyanin and used to raise an antibody in rabbits. The resulting antiserum was specific to this protein and did not react with the protein product of the 12S E1A mRNA, which shares considerable sequence with the E1Aa protein. This antiserum was used to probe for the E1Aa protein in situ by indirect immunofluorescence and in extracts of infected HeLa cells. We found that the protein was associated with large cellular structures both in the nucleus and in the cytoplasm. The nuclear form of the protein was analyzed further and was found to purify with the nuclear matrix.

Adenovirus 5 E2 transcription unit: an E1A-inducible promoter with an essential element that functions independently of position or orientation.

Utilizing deletion mutants of a plasmid containing the adenovirus E2 gene, an E1A-inducible transcription unit, we determined the promoter sequences required for full expression in transient transfection assays. Wild-type expression was obtained from plasmids containing only 79 nucleotides of upstream sequence relative to the transcription initiation site. Removal of an additional nine nucleotides lowered expression 10-fold, and deletion to -59 resulted in near total loss of transcription. Wild-type levels of expression were restored to a -28 deletion mutant by insertion of the sequence from -21 to -262 from the wild-type promoter at the -28 position, in either orientation, even though when inserted in the opposite orientation the relevant sequences were ca. 270 nucleotides upstream from their normal position. Finally, this sequence could be placed at a distance of 4,000 nucleotides from the E2 cap site and still retain near total function. Thus, the E2 promoter element can function independent of orientation and position, properties characteristic of enhancer elements.

Identification of positively and negatively acting elements regulating expression of the E2F2 gene in response to cell growth signals.

Mammalian cell growth is governed by regulatory activities that include the products of genes such as c-myc and ras that act early in G1, as well as the E2F family of transcription factors that accumulate later in G1 to regulate the expression of genes involved in DNA replication. Previous work has shown that the expression of the E2F1, E2F2, and E2F3 gene products is tightly regulated by cell growth. To further explore the mechanisms controlling accumulation of E2F activity, we have isolated genomic sequences flanking the 5' region of the E2F2 coding sequence. Various assays demonstrate promoter activity in this sequence that reproduces the normal control of E2F2 expression during a growth stimulation. Sequence comparison reveals the presence of a variety of known transcription factor binding sites, including E-box elements that are consensus Myc binding sites, as well as E2F binding sites. We demonstrate that the E-box elements, which we show can function as Myc-responsive sites, contribute in a positive fashion to promoter function. We also find that E2F-dependent negative regulation in quiescent cells plays a significant role in the cell growth-dependent control of the promoter, similar to the regulation of the E2F1 gene promoter.

The Rb-related p107 protein can suppress E2F function independently of binding to cyclin A/cdk2.

The interaction of the retinoblastoma susceptibility gene product (Rb)-related p107 protein with the E2F transcription factor in S-phase cells facilitates the formation of a multicomponent complex also containing cyclin A and the p33cdk2 kinase. We have created a series of p107 mutants to assess the ability of p107 to inhibit E2F function and the role of the cyclin A/cdk2 complex in this process. We find that p107 mutants that do not bind to E2F also fail to repress E2F-dependent transcription. Moreover, we find that the ability of p107 to suppress E2F-dependent transcription is not dependent on the ability of p107 to associate with cyclin A/cdk2. Finally, an analysis of the ability of the p107 mutant proteins to suppress cell growth suggests that both E2F-dependent and E2F-independent events correlate with this activity.

Cellular targets for activation by the E2F1 transcription factor include DNA synthesis- and G1/S-regulatory genes.

Although a number of transfection experiments have suggested potential targets for the action of the E2F1 transcription factor, as is the case for many transcriptional regulatory proteins, the actual targets in their normal chromosomal environment have not been demonstrated. We have made use of a recombinant adenovirus containing the E2F1 cDNA to infect quiescent cells and then measure the activation of endogenous cellular genes as a consequence of E2F1 production. We find that many of the genes encoding S-phase-acting proteins previously suspected to be E2F targets, including DNA polymerase alpha, thymidylate synthase, proliferating cell nuclear antigen, and ribonucleotide reductase, are indeed induced by E2F1. Several other candidates, including the dihydrofolate reductase and thymidine kinase genes, were only minimally induced by E2F1. In addition to the S-phase genes, we also find that several genes believed to play regulatory roles in cell cycle progression, such as the cdc2, cyclin A, and B-myb genes, are also induced by E2F1. Moreover, the cyclin E gene is strongly induced by E2F1, thus defining an autoregulatory circuit since cyclin E-dependent kinase activity can stimulate E2F1 transcription, likely through the phosphorylation and inactivation of Rb and Rb family members. Finally, we also demonstrate that a G1 arrest brought about by gamma irradiation is overcome by the overexpression of E2F1 and that this coincides with the enhanced activation of key target genes, including the cyclin A and cyclin E genes.

Molecular analyses of two poly(A) site-processing factors that determine the recognition and efficiency of cleavage of the pre-mRNA.

Poly(A) site processing of a pre-mRNA requires the participation of multiple nuclear factors. Two of these factors recognize specific sequences in the pre-mRNA and form a stable processing complex. Since these initial interactions are likely critical for the recognition of the poly(A) site and the efficiency of poly(A) site use, we have characterized these factors and the nature of their interaction with the pre-mRNA. The AAUAAA specificity factor PF2 is a large, multicomponent complex composed of at least five distinct polypeptides ranging in molecular size from 170 to 42 kDa. The 170-kDa polypeptide appears to mediate interaction with the pre-mRNA. Factor CF1, which provides specificity for the downstream G + U-rich element and stabilizes the PF2 interaction on the RNA, is also a multicomponent complex but is less complex than PF2. CF1 is composed of three polypeptides of molecular sizes 76, 64, and 48 kDa. UV cross-linking assays demonstrate that the 64-kDa polypeptide makes direct contact with the RNA, dependent on the G + U-rich downstream sequence element. Moreover, it is clear that these RNA-protein interactions are influenced by the apparent cooperative interaction involving PF2 and CF1, interactions that contribute to the efficiency of poly(A) site processing.

Functional properties of a Drosophila homolog of the E2F1 gene.

A variety of studies have now implicated the cellular transcription factor E2F as a key participant in transcription control during the cell growth cycle. Although the recent isolation of molecular clones encoding proteins that are components of the E2F activity (E2F1 and DP-1) provides an approach to defining the specific involvement of E2F in these events, definitive experiments remain difficult in the absence of appropriate genetic systems. We have now identified a Drosophila equivalent of E2F1 that we hope will allow an eventual genetic approach to the role of E2F in cellular regulatory events. A cDNA clone was isolated from a Drosophila cDNA library by using a probe containing sequence from the E2F1 DNA binding domain. The sequence of the clone, which we term drosE2F1, demonstrates considerable homology to the human E2F1 sequence, with over 65% identity in the DNA binding region and 50% identity in the region of E2F1 known to interact with the retinoblastoma gene product. A glutathione S-transferase-drosE2F1 fusion protein was capable of binding specifically to an E2F recognition site, and transfection assays demonstrated that the drosE2F1 product was capable of transcription activation, dependent on functional E2F sites as well as sequences within the C terminus of the protein. Finally, we have also identified E2F recognition sequences within the promoter of the Drosophila DNA polymerase alpha gene, and we demonstrate that the drosE2F1 product activates transcription of a test gene under the control of this promoter. We conclude that the drosE2F1 cDNA encodes an activity with extensive structural and functional similarity to the human E2F1 protein.

Identification of distinct roles for separate E1A domains in disruption of E2F complexes.

The adenovirus E1A protein can disrupt protein complexes containing the E2F transcription factor in association with cellular regulatory proteins such as the retinoblastoma gene product (Rb) and the Rb-related p107 protein. Previous experiments have shown that the CR1 and CR2 domains of E1A are required for this activity. We now demonstrate that the CR2 domain is essential for allowing E1A to interact with the E2F-Rb or the E2F-p107-cyclin A-cdk2 complex. Multimeric complexes containing E1A can be detected when the CR1 domain has been rendered inactive by mutation. In addition, the E1A CR1 domain, but not the CR2 domain, is sufficient to prevent the interaction of E2F with Rb or p107. On the basis of these results, we suggest a model whereby the CR2 domain brings E1A to the E2F complexes and then, upon a normal equilibrium dissociation of Rb or p107 from E2F, the E1A CR1 domain is able to block the site of interaction on Rb or p107, thereby preventing the re-formation of the complexes.

A role for a bent DNA structure in E2F-mediated transcription activation.

We examined the role of promoter architecture, as well as that of the DNA-bending capacity of the E2F transcription factor family, in the activation of transcription. DNA phasing analysis revealed that a consensus E2F site in the E2F1 promoter possesses an inherent bend with a net magnitude of 40 +/-2 degrees and with an orientation toward the major groove relative to the center of the E2F site. The inherent DNA bend is reversed upon binding of E2F, generating a net bend with a magnitude of 25 +/- 3 degrees oriented toward the minor groove relative to the center of the E2F site. We also found that three members of the E2F family, in conjunction with the DP1 protein, bend the DNA toward the minor groove, suggesting that DNA bending is a characteristic of the entire E2F family. The Rb-E2F complex, on the other hand, does not reverse the intrinsic DNA bend. Analysis of a series of E2F1 deletion mutants defined E2F1 sequences which are not required for DNA binding but are necessary for the DNA-bending capacity of E2F. An internal region of E2F1, previously termed the marked box, which is highly homologous among E2F family members, was particularly important in DNA bending. We also found that a bent DNA structure can be a contributory component in the activation of the E2F1 promoter but is not critical in the repression of that promoter in quiescent cells. This finding suggests that E2F exhibits characteristics typical of modular transcription factors, with independent DNA-binding and transcriptional activation functions, but also has features of architectural factors that alter DNA structure.

E4F and ATF, two transcription factors that recognize the same site, can be distinguished both physically and functionally: a role for E4F in E1A trans activation.

Previous experiments have identified an element in the adenovirus E4 promoter that is critical for E1A-dependent trans activation and that can confer inducibility to a heterologous promoter. This DNA element is a recognition site for multiple nuclear factors, including ATF, which is likely a family of DNA-binding factors with similar DNA recognition properties. However, ATF activity was found not to be altered in any demonstrable way as a result of adenovirus infection. In contrast, another factor that recognizes this element, termed E4F, was found at only very low levels in uninfected cells but was increased markedly upon adenovirus infection, as measured in DNA-binding assays. Although both the ATF activity and the E4F activity recognized and bound to the same two sites in the E4 promoter, they differed in their sequence recognition of these sites. Furthermore, E4F bound only to a small subset of the ATF recognition sites; for instance, E4F did not recognize the ATF sites in the E2 or E3 promoters. Various E4F and ATF binding sites were inserted into an expression vector and tested by cotransfection assays for responsiveness to E1A. We found that a sequence capable of binding E4F could confer E1A inducibility. In contrast, a sequence that could bind ATF but not E4F did not confer E1A inducibility. We also found that E4F formed a stable complex with the E4 promoter, whereas the ATF DNA complex was unstable and rapidly dissociated. We conclude that the DNA-binding specificity of E4F as well as the alterations in DNA-binding activity of E4F closely correlates with E1A stimulation of the E4 promoter.

Promoter-specific trans-activation by the adenovirus E1A12S product involves separate E1A domains.

Recent studies have shown that the adenovirus E1A12S product can trans-activate transcription by activating the transcription factor E2F. However, E2F cannot be the only target for the E1A12S product, since several cellular promoters have been found to be activated by the E1A12S protein even though they lack E2F sites. Indeed, we now show that activation of the hsp70 promoter by the E1A12S product requires the TATAA sequence. Moreover, activation of the hsp70 promoter requires the N-terminal domain of the E1A protein and does not require the conserved region 2 sequences which are required for the E2F-dependent activation of transcription. We conclude that the targeting of distinct transcription factors, leading to trans-activation of transcription of multiple promoters, involves distinct domains of the E1A proteins that are also required for oncogenic activity.

Identification of an activity in B-cell extracts that selectively impairs the formation of an immunoglobulin mu s poly(A) site processing complex.

The immunoglobulin mu heavy-chain transcription unit is differentially expressed during B-cell development, producing mRNAs that encode secreted (mu s) and membrane-bound (mu m) forms of the heavy-chain polypeptide. Whereas the mu s mRNA and the mu m mRNA are produced in approximately equal abundance in B cells, an increase in the utilization of the mu s poly(A) site contributes to the production of the mu s mRNA as the predominant form in a plasma cell. Previous experiments have demonstrated a correlation between the formation of a stable complex on a poly(A) site and the relative function of the poly(A) site. We have thus investigated the parameters determining the interaction of these factors with the immunoglobulin poly(A) sites. Assays of complex formation involving the two immunoglobulin poly(A) sites by using HeLa cell activities revealed the formation of stable complexes with no apparent difference between the mu s site and the mu m site. In contrast, the mu s-specific complex was markedly less stable when a B-cell extract was used. Fractionation of B-cell extracts has revealed an activity that specifically destabilizes the mu s polyadenylation complex, suggesting that the function of this poly(A) site may be regulated by both positive- and negative-acting factors.