This research proposal aims at understanding the fundamental mechanisms of human gene regulation at the initiation step of transcription. Molecular, structural and cell biologists will combined their efforts in a multidiscplinary approach to explore this central process. The enzyme that transcribes human genes into messenger RNA is RNA polymerase II. Transcription by this polymerase is regulated by a plethora of protein factors, including large and complicated machines with many subunits. The first general transcription factor (GTF) to bind gene promoters is TFIID. TFIID is a megadalton-sized multiprotein complex composed of TATA-box binding protein (TBP) and 13 TBP associated factors (TAFs). Despite its crucial role, the detailed molecular architecture and assembly mechanism of TFIID remain elusive. This is partly due to the low abundance and heterogeneity of TFIID in cells, and the difficulties associated with extracting TFIID to study its function. As a consequence, structural analysis of human holo-TFIID has been stalled for many years at moderate resolution (~35 Å) despite major efforts by some of the best laboratories world-wide. At the transcription start site, TFIID interacts with specific nucleosomes carrying defined post-translational modifications (epigenetic marks) that form the local chromatin template. Epigenetic modifications are an intense research focus, owing to the fact that they profoundly influence chromatin function, with direct impact on a large number of human diseases including cancer. TFIID contains reader domains that recognize epigenetic marks. The exact mechanism how TFIID binds to its chromatin template, and what the consequences are for the conformations of TFIID and the nucleosomes, is unknown to date. To activate gene expression, small activator proteins bind upstream of promoters, and several of these communicate with TFIID. Only low resolution data exists on the modalities of TFIID binding to activators, which are essential for properly directing gene expression. Which TAFs exactly may be involved in these interactions is not known, due to the lack of precise structural information on TAF geometries within TFIID, and the paucity of material due to the difficulties to get hold of purified TFIID. How TFIID is assembled in the cell is an unsolved mystery. Recent studies suggest that TFIID is composed of stable modules that may represent assembly intermediates. Unique transcriptional roles have been associated with these assembly intermediates. The identification of such intermediates and of assembly factors that may assist in the formation of holo-TFIID, has remained an unmet challenge. We have recently succeeded in producing, for the first time, fully recombinant, functional human holo-TFIID and its subassemblies in the quality and quantity required for structural and functional analysis. We had developed new expression methods for this purpose. In this proposal, we will exploit this break-through to determine the architecture of this essential complex by hybrid methods combining recombinant production of TFIID and its subcomplexes, cryo-EM, X-ray crystallography, cross-linking mass-spectrometry and multi-constraint modeling. Further, we will dissect the structure of TFIID bound to its epigenetically modified chromatin template, and we will elucidate the molecular determinants of selected transcriptional activators and a repressor binding to recombinant holo-TFIID. We will isolate stable TFIID assembly intermediates and non-canonical TFIID complexes directly from cells, and analyze their composition by using powerful new proteomics approaches. We will produce these assembly intermediates recombinantly and analyze their possible roles in transcription by a variety of means including protein transduction. We will identify putative assembly factors and study their activities in assisting holo-TFIID.