Purification of general RNA polymerase II transcription factors from mouse for studies of proliferation-specific transcription

Sammanfattning: Accurate initiation of transcription by RNA polymerase II depends on general transcription factors (GTFs), which include the TATA-binding protein (TBP) and the transcription factors (TF) IIB, IIF, IIE and IIH. In order to reconstitute mouse transcription in vitro, we cloned the genes encoding mouse TFIIB, and both subunits of TFIIE and TFIIF from a mouse cDNA library. TBP and TFIIB were expressed in E.coli, while both subunits of TFIIE and the two subunits of TFIIF were expressed in a baculovirus system. All these factors were purified to > 90% homogeneity. The more complex transcription factors, TFIIH and RNA polymerase II, were purified more than 1000-fold and to near homogeneity, respectively, from tissue cultured mouse ascites cells. We have shown that the purified mouse transcription factors are active in a reconstituted RNA polymerase II in vitro transcription assay. The transcription reaction was inhibited by α-amanitine, and dependent on the addition of all the GTFs.Ribonucleotide reductase is a key enzyme in deoxyribonucleotide synthesis. It consists of two subunits, R1 and R2, which are both required for the enzyme activity. Transcription of the R1 and R2 genes is restiricted to the S-phase of the cell cycle, but the mechanisms that control this coordinated expression remain to be identified. We have studied initiation of transcription from the mouse R2 gene using a combination of in vivo reporter gene assays and in vitro transcription assays with crude nuclear extracts or with purified transcription factors. This promoter has an atypical TATA-box and a CCAAT-box that binds the transcription factor NF-Y.We found that a mutation in the R2 CCAAT-box had no effect on the transcription level in in vitro transcription assays reconstituted with pure transcription factors. However, it significantly decreased the level of transcription in similar experiments using crude nuclear extract. We also found that the sequence downstream from the R2 transcription start site (5´-UTR) (from +1 to +17 base pairs relative to transcription start site) is essential for initiation of transcription from this promoter. The presence of the wild type 5´-UTR made the R2 TATA-box redundant. On the other hand, the R2 5´-UTR had a repressing effect on transcription from the mouse R2 promoter. This region contains a palindrome sequence that covers 10 base pairs, and it is partially conserved in the human R2 promoter. Gel shift assays and in vitro transcription experiments using antibodies against mouse TAF4 (=TAF135) demonstrate that TAF4 is a component of the protein complex that interacts with this palindrome region, and suggest involvement of this component of the TFIID complex in negative regulation of the R2 promoter.The Adenovirus Major Late (AdML) promoter is commonly used as a model for studies of transcription initiation and regulation. It is a TATA-box dependent promoter, which also contains an initiator (Inr) element, a CCAAT-box interacting with transcription factor NF-Y, and an E-box binding the upstream stimulatory factor (USF). Using gel shift assays with recombinant NF-Y, USF, and immunopurified human TFIID, we show that binding of USF1 and NF-Y to DNA is not cooperative and that both factors independently facilitate binding of TFIID to the core promoter. The activation domains of NF-Y are expendable for this effect. Negative cofactor (NC2) comprises two subunits, which have a histone-fold structure similar to NF-Y, and represses transcription through formation of an inhibitory complex with TBP. Using an in vitro transcription system based on crude nuclear extracts, we show that NC2 has a negative effect on transcription in the presence of NF-Y or USF1, indicating that the two activators do not act as antirepressors. In vitro transcription using highly purified transcription factors efficiently reproduces repression of transcription by NC2. However, USF1 was inactive and NF-Y had a repressing effect in this system, which suggests that the activator functions of USF and NF-Y depend on cofactors.