Skip to main content

Structure and function of SMAT: a putative cellular decision-maker for transcription factor assembly

Final Report Summary - SMAT (Structure and function of SMAT: a putative cellular decision-maker for transcription factor assembly)

In this project we investigated the assembly and architecture of a large multi-protein complex, the general transcription factor IID (TFIID), which is a key player in regulated transcription of eukaryotic genes by RNA polymerase II. The scaffold of TFIID is composed of a number of TATA-box binding protein-associated factors (TAFs), which are also found in other multimeric transcription factors, e.g. Spt-Ada-Gcn5 acetyltransferase complex (SAGA). An elaborate system of decision-making therefore must occur that directs the assembly of TAFs into either TFIID or SAGA or other TAF-containing complexes.

Recently, a multimeric 400 kDa small TAF-containing complex, SMAT, was discovered. SMAT is proposed to be a potent regulator of committed TFIID and SAGA assembly, since it is composed of a TFIID-specific protein, TAF8, a SAGA-specific protein, SPT7L, and a protein, that is present in both complexes, TAF10. Interestingly, TAF10 interacts also with another TFIID-specific protein, TAF3.

The objective of this project was to elucidate the role of SMAT in eukaryotic gene regulation by means of an integrated, interdisciplinary approach combining structural, biochemical and in vivo cell biology methods. We also investigated the molecular basis for TAF10 promiscuity by X-ray crystallography to be able to define structure-based point mutations, which will abolish TFIID and/or SAGA assembly in vitro and in vivo.

Subcomplexes of TFIID and SAGA were produced using the MultiBac system, a baculoviral expression vector system for protein production in insect cells, and purified to homogeneity. Large amounts of TAF8/TAF10 and SPT7L/TAF10 were obtained in high quality and used to reconstitute SMAT in mixing experiments. However, formation of the complex could not be detected. Also, balanced co-expression of TAF8 (one copy), TAF10 (two copies) and SPT7L (one copy) using a polyprotein approach in insect cells did not yield SMAT. These results suggest that either a yet unidentified partner is missing in SMAT preparations or specific PTMs, which are not provided by the expression host, are required for SMAT formation.

In order to characterize the hierarchical assembly of TFIID, binding studies of TAF10-containing complexes with other components of TFIID were performed and identified new interactions within a core TFIID complex. These yet unknown interactions between TAFs were mapped on the molecular level by various biochemical and biophysical methods. Crystals of a complex of TAF8/TAF10 were obtained by an iterative approach of limited proteolysis, construct variation and high-throughput crystallization screening and the structure was solved at a resolution of 2.4 A by the Sulfur-SAD method. Crystallization and structure determination of the newly identified complexes is ongoing.

It has been demonstrated that TAF10, which doesn't have a nuclear localization sequence (NLS), hitchhikes on the NLS-containing TAF8 to the nucleus. This co-import strategy, which depends on the molecule Importin alpha1, might have interesting implications for TFIID assembly. To understand the co-import mechanism on the molecular level, the crystal structure of Importin alpha1 bound to a peptide of the nuclear localization sequence of TAF8 was solved at a resolution of 1.8 A. A potential nuclear import particle (NIP) comprising TAFs and Importin alpha1 was reconstituted from individual purified components. Composition and stoichiometries of NIP were analyzed by analytical ultracentrifugation and native mass spectrometry experiments. The import of NIP is being analyzed in vivo in mammalian cells using fluorescently labelled fusion proteins.

The findings from this project opened up exciting insights into the architecture and assembly of TFIID with implications for the assembly of SAGA. Furthermore, the project strengthened collaborations between the host institute and institutes in Europe, notably at IGBMC in Strasbourg, France. The results will be made available to the scientific community upon publication.