Jonathan M. Boutell

Genomic maps of DNA G-quadruplexes (G4s) can help elucidate the roles that these secondary structures play in various organisms. Herein, we employ an improved version of a G-quadruplex sequencing method (G4-seq) to generate whole genome G4 maps for 12 species that include widely studied model organisms and also pathogens of clinical relevance. We identify G4 structures that form under physiological K+ conditions and also G4s that are stabilized by the G4-targeting small molecule pyridostatin (PDS). We discuss the various structural features of the experimentally observed G-quadruplexes (OQs), highlighting differences in their prevalence and enrichment across species. Our study describes diversity in sequence composition and genomic location for the OQs in the different species and reveals that the enrichment of OQs in gene promoters is particular to mammals such as mouse and human, among the species studied. The multi-species maps have been made publicly available as a resource to the research community. The maps can serve as blueprints for biological experiments in those model organisms, where G4 structures may play a role.

We detected an interaction of the N-terminus of huntingtin (htt171) with the C-terminal region of the nuclear receptor co-repressor (N-CoR) using the yeast two-hybrid system. This interaction was repeat length dependent and specific to htt171; the co-repressor did not interact with the repeat carrying a section of atrophin 1 nor with the androgen receptor or polyglutamine alone. The interaction was confirmed using His-tagged Escherichia coli -expressed C-terminal human and rat co-repressor protein which pulled full-length huntingtin out of homogenized rat brain and in pull-down assays. The N-CoR represses transcription from sequence-specific ligand-activated receptors such as the retinoid X-thyroid hormone receptor dimers and other nuclear receptors including Mad-Max receptor dimers. The mechanism of this repression appears to be through the formation of a complex of repressor proteins including the N-CoR, mSin3 and histone deacetylases. We have used N-CoR and mSin3A antibodies in immunohistochemical studies and find that in Huntington's disease (HD) cortex and caudate, the cellular localization of these proteins is exclusively cytoplasmic whilst in control brain they are localized in the nucleus as well as the cytoplasm; mSin3A immunoreactivity also occurred in a subset of huntingtin positive intranuclear inclusions. The relocalization of repressor proteins in HD brain may alter transcription and be involved in the pathology of the disease.

We have screened a rat brain library to identify proteins which interact with the 5'-end of huntingtin (amino acids 1-171), including the polyglutamine tract, in the yeast two-hybrid system. We detected an interaction with cystathionine beta-synthase (CBS) [L-serine hydrolyase (adding homocysteine), EC 4.2.1.22], which was confirmed in vitro using His-tagged CBS expressed in Escherichia coli , which was able to specifically bind both rat and human full-length huntingtin. Neither normal nor expanded polyglutamine repeat alone interacted with CBS in the yeast two-hybrid system and nor did constructs containing SBMA or DRPLA with normal or expanded polyglutamine tracts. CBS therefore appears to bind specifically to huntingtin. CBS deficiency is associated with homocystinuria, which is known to affect various physiological systems, including the central nervous system. Homocysteine, one of the substrates of CBS, is known to accumulate in homocystinuria and is metabolized to homocysteate and homocysteine sulphinate, both known to be powerful excitotoxic amino acids. It has been suggested that Huntington's disease involves the action of excitotoxic amino acids and this interaction with CBS may suggest a mechanism for such excitotoxic damage.