Accessibility of epitopes on human transcription factor IIB in the native protein and in a complex with DNA.

Transcription factor IIB (TFIIB) plays a central role in the assembly of the RNA polymerase II initiation complex. Monoclonal antibodies (mAbs) that react with human TFIIB were prepared and used as probes to identify portions of TFIIB that are accessible when the factor is in solution and when it is contained in a complex with DNA. Seven mAbs were examined and were mapped to three regions of the TFIIB molecule. Only the mAbs that mapped to residues 52-105 inhibited transcription, immunoprecipitated recombinant TFIIB and TFIIB from HeLa cell nuclear extract (NE), and supershifted a complex containing TFIIB, the TATA-binding protein, and DNA. The mAbs that mapped to residues 1-51 and the mAb that mapped to residues 106-316 did not show activity in the functional assays, with the exception of the far N-terminal mAbs (residues 1-51), which immunoprecipitated recombinant TFIIB, but not TFIIB from HeLa cell NE. These data indicate that the region containing residues 52-105 is exposed in solution and when TFIIB is part of the preinitiation complex and that some far N-terminal epitopes are accessible on the purified protein, but become blocked when TFIIB is in HeLa cell NE or in the preinitiation complex.

Identification of cis-acting elements that can obviate a requirement for the C-terminal domain of RNA polymerase II.

We have used an in vitro RNA polymerase II (RNAP II) inhibition-restimulation assay to investigate the inability of a form of RNAP II (RNAP IIB) that lacks the conserved, C-terminal heptapeptide repeat domain (CTD) to transcribe the dihydrofolate reductase (dhfr) promoter. Our previous studies demonstrated promoter-specific responses to RNAP IIB in the inhibition-restimulation assay and suggested the existence of cis-acting elements that alleviate the requirement for the CTD. We have now identified elements from two different classes of promoters that can convert dhfr to a CTD-independent promoter. First, addition of a consensus TATA box to the dhfr promoter resulted in a promoter capable of CTD-independent transcription and increased the promoter's affinity for the general transcription factor TFIID. These results suggest that high affinity for TFIID correlates with an ability to be transcribed by RNAP IIB, supporting a proposed interaction between the CTD and TFIID. Second, transfer of a combination of two elements (located at -25 and +1) from the rep-3b promoter, which does not contain a consensus TATA box but can nonetheless be transcribed by RNAP IIB, into the dhfr promoter also allowed CTD-independent transcription. These elements do not constitute a high affinity binding site for TFIID, indicating that an additional mechanism exists to allow CTD-independent transcription. Thus, elements from two classes of CTD-independent promoters that can obviate a requirement for the CTD appear to function via distinct mechanisms. Our finding that a change in a basal element can affect a requirement for the CTD is consistent with a role for the CTD during the formation of the transcriptional preinitiation complex.

The HIP1 initiator element plays a role in determining the in vitro requirement of the dihydrofolate reductase gene promoter for the C-terminal domain of RNA polymerase II.

We examined the ability of purified RNA polymerase (RNAP) II lacking the carboxy-terminal heptapeptide repeat domain (CTD), called RNAP IIB, to transcribe a variety of promoters in HeLa extracts in which endogenous RNAP II activity was inhibited with anti-CTD monoclonal antibodies. Not all promoters were efficiently transcribed by RNAP IIB, and transcription did not correlate with the in vitro strength of the promoter or with the presence of a consensus TATA box. This was best illustrated by the GC-rich, non-TATA box promoters of the bidirectional dihydrofolate reductase (DHFR)-REP-encoding locus. Whereas the REP promoter was transcribed by RNAP IIB, the DHFR promoter remained inactive after addition of RNAP IIB to the antibody-inhibited reactions. However, both promoters were efficiently transcribed when purified RNAP with an intact CTD was added. We analyzed a series of promoter deletions to identify which cis elements determine the requirement for the CTD of RNAP II. All of the promoter deletions of both DHFR and REP retained the characteristics of their respective full-length promoters, suggesting that the information necessary to specify the requirement for the CTD is contained within approximately 65 bp near the initiation site. Furthermore, a synthetic minimal promoter of DHFR, consisting of a single binding site for Sp1 and a binding site for the HIP1 initiator cloned into a bacterial vector sequence, required RNAP II with an intact CTD for activity in vitro. Since the synthetic minimal promoter of DHFR and the smallest REP promoter deletion are both activated by Sp1, the differential response in this assay does not result from upstream activators. However, the sequences around the start sites of DHFR and REP are not similar and our data suggest that they bind different proteins. Therefore, we propose that specific initiator elements are important for determination of the requirement of some promoters for the CTD.

Purification and characterization of proximal sequence element-binding protein 1, a transcription activating protein related to Ku and TREF that binds the proximal sequence element of the human U1 promoter.

The promoter structure of the known small nuclear RNA (snRNA) genes contains two major effectors of transcriptional activity, a proximal sequence element (PSE) and a distal sequence element (DSE). In previous work, methidiumpropyl-EDTA-Fe(II) footprinting was used to demonstrate the existence in human placental extracts of a protein producing footprints within the PSE and the DSE of the human U1 snRNA gene. This protein (PSE1) has now been purified to homogeneity from both human placental extract and K562 cell nuclear extract. PSE1 consists of two subunits, an alpha subunit with an apparent molecular mass of 83 kDa, and a beta subunit with an apparent molecular mass of 73 kDa in K562 nuclear extracts and 63 kDa in placental extracts. Footprinting and UV cross-linking assays indicate that purified PSE1 binds to the PSE and DSE of the U1 gene. Monoclonal antibodies were prepared which specifically recognize the individual subunits of PSE1. PSE1 is immunologically similar to and shares amino acid sequence with a protein (TREF) which binds the human transferrin receptor (HTFR) promoter. An in vitro transcription system was established for a template consisting of a minimal HTFR promoter placed upstream of the human U1 snRNA-coding region and shown by immunodepletion/addback experiments to specifically require PSE1. Transcription from the adenovirus 2 major late promoter was unaffected in these experiments. This result supports a functional role of PSE1 as a transcriptional activating protein, but its role in transcription of snRNA genes remains to be established. PSE1 also has an immunological relationship to and shares amino acid sequence with the p70 and p86 subunits of the human Ku autoantigen. Ku, PSE1, and TREF may thus be identical proteins or members of a family of heterodimeric proteins consisting of related subunits. Our results support earlier proposals that Ku may be a transcriptional activator.