Signal Transduction and Transcriptional Regulation

Section: Section of Structural Biology

 

Structural Studies of Groucho/TLE and Complexes

 

LM Pickles, (EM Hemingway), (S Jones), V Good, V Thompson-Vale, LH Pearl; in collaboration with H Clevers, University Hospital, Utrecht; D Ish-Horowicz, ICRF, London; S Stifani, Montreal Neurological Institute

Source of external funding: The Wellcome Trust

Expression of many genes is repressed by recruitment of chromatin modifying enzymes such as histone deacetylases to sequence-specific DNA-binding transcription factors, via interaction with tetrameric Gro(Groucho)/TLE co-repressor proteins. Of particular relevance to cancer, Gro/TLE acts as a co-repressor of the leukaemia-associated AML1, and of Tcf/Lef transcription factors which are the downstream targets of the Wnt-signalling pathway which is inappropriately activated in 80% of colon cancers. Gro/TLEs proteins consist of four main regions; an N-terminal Q-domain which mediates tetramerisation; the CcN domain which is subject to regulatory phosphorylation; a serine-proline rich SP region; and a C-terminal domain containing seven WD40 repeats. Different components of the multi-domain structure offer specific binding sites for the wide range of transcription factors, repression effectors and co-factors with which Gro/TLE proteins interact.

We are using biochemical and structural techniques to gain detailed insight into the basis of these specific interactions, and the assembly of Gro/TLE-mediated repression complexes. Previously we determined the crystal structure of the C-terminal WD40 domain of human TLE1 and have now crystallised the N-terminal tetramerisation and Tcf-binding domain. Gro/TLE complexes transcription factors, histone deacetylases and the UTY/UTX co-factor are being prepared by co-expression in insect cells and will be used in crystallisation trials.

Structural Studies of GSK3β and the β-catenin Turnover-Complex

 

R Dajani, V Good, V Thompson-Vale, LH Pearl; in collaboration with TC Dale, E Fraser, M Yeo, University of Cardiff; L Meijer, CNRS Roscoff.

Source of external funding: Cancer Research UK

 

The Wnt signalling pathway plays a role in developmental processes including, cell proliferation and cell polarity. The Wnt pathway regulates b-catenin levels. Overactivation of the Wnt pathway, which leads to accumulation of β-catenin, is implicated in several human cancers. β-catenin levels are regulated by a "turnover-complex"consisting of scaffold proteins APC, Axin and Diversin, and the protein kinases CK1α, CK1ε and GSK3β. Hyperphosphorylation of the N-terminus of β-catenin by GSK3β (serine/threonine kinase) following "priming" mono-phosphorylation by CK1α or CK1ε flags b-catenin for degradation. Remarkably GSK3β also functions in insulin and growth factor signalling pathways where it is inhibited by phosphorylation by the upstream kinase PKB, but with no apparent cross-talk with Wnt-signalling.

We are using a combination of molecular and structural techniques to determine the basis for assembly and regulation of the β-catenin turnover-complex, and to understand how the associated kinases are specifically recruited and their signals insulated. We have previously solved the crystal structure of GSK3β, defined the structural basis for its specificity for phosphorylated substrates, and determined the mechanism of auto-inhibition following phosphorylation by PKB. In the last year we determined the crystal structure of a complex between GSK3β and its binding site on Axin, revealing the structural basis for its recruitment to the β-catenin turnover-complex and the mechanism of competitive displacement by FRAT. We have also determined the structure of a complex between GSK3β and a novel class of selective protein kinase inhibitors, derived from Tyrian purple indirubins, which may have application in treatment of diabetes and Alzheimer’s. Ongoing work is focussed on structure determination of the GSK3β - Axin - β-catenin ternary complex and on structural studies of the priming component of the turnover-complex involving Diversin, CK1α and CK1&epsilon.

Structural Studies of Protein-Acetylating and Protein-Methylating Co-activators in Transcriptional Regulation

W Yue, LH Pearl; in collaboration with T Kouzarides, Wellcome/Cancer Research UK Institute of Cancer and Developmental Biology, Cambridge

Source of external funding: Cancer Research UK

Acetylases and deacetylases regulate transcription by modifying the acetylation state of histones and other promoter-bound transcription factors. Acetylases have a clear role in cell proliferation and differentiation implying that deregulation of these enzymes has a causative role in cancer, evidence to support this is accumulating. There are four families of acetyltransferases, all of which contain intrinsic acetyltransferase activity, associated with a HAT domain. The majority of acetyltransferases possess a Bromo domain that functions to recognise acetylated lysine residues and as such has a signalling role. Methyltransferases are thought to have a role in transcriptional repression by promoting the formation of heterochromatin through the methylation of histone tails. There are two types of methyltransferase, S-adenosyl-L-methionine-dependent arginine N-methyltransferases (PRMTs) of which there are at least four distinct mammalian family members and the lysine methyltransferases such as Su(Var)3-9. Different histone acetyl- and methyl-transferases modify histone tails at different positions, and are affected by the presence or absence of pre-existing modifications. The combinatorial interplay between these modifications constitutes a complex epigenetic code, whose full meaning is yet to be worked out.

We are applying structural and biochemical techniques to understand the individual specificity and the cooperation between two histone modifying enzymes; the histone acetlytransferase P300/CBP which acetylates lysines 18 and 23 in histone H3, and the arginine methyltransferase CARM1 which methylates arginine 17, but only after P300/CBP has acted. We have defined several soluble active segments from P300/CBP and are developing large-scale purification protocols for crystallization trials. Expression systems for CARM1 are being constructed.