Computational function assignment for potential drug targets: from single genes to cellular systems.

Biomedical science is currently undergoing an epoch-marking transition from its classical phase to the post-genome era. The outstanding success of world-wide genome sequencing efforts, evidenced by the recent publication of the draft of the human genome, together with the completion of several genomes of eukaryotic model organisms and the availability of microbial genome sequences, is opening up data sources of unprecedented scale for drug discovery. Furthermore, the elucidation of genome expression states through transcriptomic and proteomic techniques is playing a crucial role in the characterisation of disease at the molecular level. At the same time, our still very limited knowledge of the biological functions of genes and proteins at different levels of cellular organisation is preventing full exploitation of the available data. This review will discuss current computational techniques for function prediction based on the sequence-structure-function paradigm. Newly emerging approaches aimed at gaining an expanded understanding of function through integration of data from various sources and modelling of complex cellular systems will also be highlighted.

Can correlated mutations in protein domain families be used for protein design?

Evidence from diverse studies, such as protein design experiments and analysis of the emergence of drug resistance in human immunodeficiency virus 1 (HIV-1), indicates that protein function can be diminished or altered by mutations at positions distant from the classic 'functional' site. Furthermore, results from correlation analysis of the ligand-binding domain of nuclear receptors suggest that mutation events at positions distributed throughout a protein domain may be involved in functional diversification during the evolution of homologous domain families. This review explores potential applications for a protein design procedure based on correlated substitutions.

Prediction of interaction partners for orphan nuclear receptors by prior-based protein sequence profiles.

We present a prior-based profile method for the prediction of protein-protein interaction partners that is here applied to the nuclear receptor superfamily. In this method, the diagnostic features are locally encoded in the physicochemical properties of residues in the interaction surface that are conserved in all proteins belonging to the defining set. The procedure models the positional variation based on that observed in the defining set and a prior-based substitution matrix derived from over 20,000 highly conserved positions in a set of 147 functional protein families. The method clusters sets of nuclear receptors known to interact with retinoid X receptor or corepressor proteins with predictive sets of receptors in C. elegans and higher metazoans. The method effectively reduces the search space of all possible interactions and yields experimentally testable predictions. Applications of this novel approach extend to interaction prediction problems in general, particularly to those that are not amenable to analysis by the rigid-body approximation.

The PWWP domain: a potential protein-protein interaction domain in nuclear proteins influencing differentiation?

Upon characterization of WHSC1, a gene mapping to the Wolf-Hirschhorn syndrome critical region and at its C-terminus similar to the Drosophila ASH1/trithorax group proteins, we identified a novel protein domain designated PWWP domain. To gain insight into its structure, evolutionary conservation and its potential functional role, we performed database searches to identify other PWWP domain-containing proteins. We retrieved 39 proteins, and a multiple alignment shows that the domain spans some 70 amino acids. It is present in proteins of nuclear origin and plays a role in cell growth and differentiation. Due to its position, the composition of amino acids close to the PWWP motif and the pattern of other domains present, we hypothesize that the domain is involved in protein-protein interactions.

Evolutionary constraint networks in ligand-binding domains: an information-theoretic approach.

Ligand-binding sites in homologous protein domains can diverge greatly during evolution. This poses a particularly interesting problem in those cases where the ligand-binding site is situated in, or close to, the domain core, or where ligand-docking induces dramatic conformational changes. These features are present in many receptors and enzymes; the hormone-binding domain of the nuclear receptors for steroids and retinoids, for example, exhibits both characteristics. It is therefore of great interest to determine how binding sites for diverse ligands evolve in core regions of structurally dynamic domains. Are evolutionary changes locally restricted to the ligand-binding site, or are they distributed throughout the domain? We describe here an information-theoretic approach for the study of covariation between ligand-contacting residues and compensatory mutations that preserve the structural integrity and the conformational dynamics of ligand-binding domains. We apply this method to the analysis of the nuclear receptor ligand-binding domain and show that the ligand-contacting residues in the hormone-binding pocket are evolutionarily linked to an extensive network of covarying positions.

v-erbA oncogene initiates ultrastructural changes characteristic of early and intermediate events of meiotic maturation in Xenopus oocytes.

The growth-promoting properties of the retroviral v-erbA oncogene, a highly mutated version of the chicken thyroid hormone receptor (TR) alpha, have so far exclusively been linked to dominant repression of the antimitogenic roles of TR and retinoic acid receptors. Here we show that when expressed in Xenopus oocytes v-ErbA induced ultrastructural changes characteristic of early and intermediate events of meiotic maturation by activating gene transcription. v-ErbA-induced maturation events occurred without activation of the cAMP/maturation-promoting factor signal pathway and were arrested prior to meiotic spindle formation. The effects of v-ErbA were not mimicked by a dominant negative in vitro-generated mutant of human TR, suggesting that v-ErbA can contribute to cell cycle reentry by interference with regulatory pathways distinct from those involving TR. Interestingly, a portion of v-ErbA expressed in oocytes was present at the cytoplasmic fibrils of the nuclear pore complexes, suggesting that in addition to its intranuclear function v-ErbA may modulate nucleocytoplasmic transport.

Constitutive transactivation by the thyroid hormone receptor and a novel pattern of activity of its oncogenic homolog v-ErbA in Xenopus oocytes.

In Xenopus oocytes, the rat thyroid hormone receptor alpha (rTR alpha), but not its oncogenic homolog v-ErbA, constitutively activated thyroid hormone (T3)-responsive reporter genes at four positive thyroid hormone-responsive elements (TREs). At a subset of the positive TREs tested, the addition of T3 resulted in a further enhancement of reporter gene activation. In contrast, both rTR alpha and v-ErbA functioned as constitutive activators when bound to the clone 122 TREs, which are induced by unliganded TR in mammalian cells. Therefore, the responses of the ligand-independent activation domains of TR and v-ErbA to cell-specific and TRE-mediated induction are not equivalent. Coexpression of the human retinoid X receptor alpha (hRXR alpha) enhanced both ligand-dependent and ligand-independent activation functions of rTR alpha and human TR beta (hTR beta) at a palindromic TRE (TREp). An endogenous TR activity mediated T3 induction of TREp, while being repressed by an in vitro-generated dominant negative mutant of TR. T3-mediated gene activation, by exogenous or endogenous TR, was repressed by v-ErbA at three positive TREs, but not at the TRE from the third intron of the rat GH gene (rGH3 TRE). Interestingly, preinjection of nuclear protein extract from anterior pituitary cells converted v-ErbA into a constitutive activator at rGH3 TRE. The pituitary-specific factor Pit-1/GHF-1 or hRXR alpha did not induce activation by v-ErbA at rGH3 TRE, suggesting that the dominant negative phenotype of v-ErbA can be abolished by direct or indirect interactions with other nuclear factors.