Dr Merridee Wouters

Dr Merridee Wouters, PhD Group Leader, Computational Biology and Bioinformatics Division Senior Lecturer, University of New South Wales Telephone: +61-2-9295 8608 Fax: +61-2-9295 8601 Email: m.wouters@victorchang.edu.au

Research Focus:

Structural Bioinformatics Group:

Proteins are essentially small machines. Understanding how they work is a challenge for structural biologists and structural data mining is a valuable tool in this quest. Protein conformational changes and their role in protein function; and the effect of mutations on protein function can all be studied using structural data mining.

The group is currently undertaking two major projects. The first of these is a computational approach to the prediction of candidate disease genes. The second project is an investigation of the role of disulphides in redox signalling. In addition, several minor projects in collaboration with other research groups at the VCCRI are ongoing. These include projects to predict the molecular basis of disease-causing mutations.

A computational pipeline for the prediction of inherited disease genes:

We are currently developing a computational pipeline to aid in the discovery of disease genes within specific chromosome intervals. The pipeline consists of two major parts:

a)An annotation engine which draws on numerous bioinformatics tools from around the world in order to characterise each gene and its associated product.
b) A search engine that enables investigators to retrieve annotations for specific genes by filtering genes based on intervals, phenotype and structural characteristics.

The pipeline was developed for use by Dr. Diane Fatkin's laboratory to search for candidate disease genes involved in inherited heart diseases such as dilated cardiomyopathy (DCM) and familial atrial fibrillation (fAF). Disease intervals relevant to inherited heart diseases have been reannotated and this information is available locally as searchable webpages. To date beta versions of the annotation and search engines have been implemented and these are available in-house for Dr Fatkin's group. The current version of the pipeline was demonstrated at the 19th Lorne Genome conference 2004.

• To gather as much information as possible and combine it intelligently;
• To reduce the tedious aspects of searching intervals, particularly large ones.

Although the system is already up and running, its ability to predict candidate genes will be greatly increased by further enhancements to the annotation engine in order to provide more comprehensive output.

In the further development of the pipeline we wish to enhance the search function by automating approaches to identifying candidate genes.

Long term plans for the pipeline are to make it available for other researchers and we are currently seeking funding to do so.

Disulphides: structural stabilizers or redox switches?

Disulfides are generally viewed as structurally stabilizing elements in proteins. However, emerging evidence shows that a subclass of disulfides in proteins may act as redox-controlled switches governing conformational changes in proteins. We believe these disulfides are distinguished from the more conventional structural stabilizers by their high torsional energies and their availability or potential availability to solvent.

We first unearthed examples of these high torsional energy disulfides during a structural data mining survey of b-sheets (Wouters and Curmi, 1995) which revealed disulfides linked across the strands. These cross-strand disulfides (CSDs) attracted our notice because they occur in a secondary structure that is already non-covalently linked and because it had previously been asserted that such structures would not exist because of the incompatible conformational constraints of b-sheets and disulfides (Richardson, 1981). Indeed, we found the disulfides and the b-sheets around them were significantly distorted.  

Structural data mining of the PDB done to date suggests that in addition to cell entry mediated by exogeneous organisms, redox controlled nanomachines may contribute to other cellular processes such as signal transduction, amyloid formation and enzyme inhibition. We are currently further investigating these unusual disulphides (Wouters et al, 2004).

G-Coupled Receptor Group:

• Analysis and prediction of non-canonical elements in transmembrane helices;
• Modelling of the adrenergic GPCR subfamily;
• Pattern analysis to define intra- and inter-helical interactions;
• Structural roles for tryptophan.

The helices of polytopic membrane proteins possess unique structural features that are not found in the those of globular proteins. In particular, prolyl residues, very rare in a-helices of globular proteins (excepting at their N and C termini), and thought to be incompatible with a-helical structures, have been found to be abundant in transmembrane helices. An important part of our efforts aims at extending the tools developed for better analysis and definition of helical fine-structure in transmembrane proteins. An examination of the available structures of polytopic proteins reveal that non-a-helical conformations (p, 3₁₀ and kinks) occur frequently and are critical determinants of polytopic protein structure. Non-degenerate motif descriptors have been derived through data mining of training sets of residues taken from the transmembrane-spanning segments of polytopic protein crystal structures. A `search engine' derived from these motif descriptors correctly identifies, and discriminates amongst instances of the above `non-canonical' helical motifs. Our results suggest that deviations from a-helicity are encoded locally in sequence patterns only about 7-9 residues long and can be determined in silico directly from the amino acid sequence.

We are using sequence pattern analysis on the heptahelical GPCR superfamily to further understand the effects of sequence and residue rotamer conformation on helical structure.These findings are being used to refine homology-derived models of the adrenergic receptor subfamily.

The bulky indole ring sidechain of tryptophan presents unique opportunities for binding interactions with adjacent helices when the residue is present in the transmembrane regions of polytopic proteins. The ring can intrude between residues in adjacent turns of the neighbouring helix and form a hydrogen bond with a backbone oxygen. The sidechain can insert between two adjacent helices preventing rotation of the tryptophan-containing helix. Additional interactions with adjacent helices have also been observed. The other amino acids can not provide the same structural and functional roles. The findings indicate an additional role for tryptophan in interhelical interactions beyond concerted sidechain/sidechain movements.

The ultimate goal of the studies in these different areas is to understand the structure and functioning of GPCRs. Through knowledge of how the structure and dynamics of the side chains affect the structure of the receptor, the programmed series of conformational changes that occur upon ligand binding can be followed. Questions about different ground state conformations leading to the activation of different signalling pathways can now start to be answered.

Links:
http://www.gentrepid.org

Co-Investigators:
Samuel Fan
Richard George, PhD
Arthur Liu
Jason Liu
Ignatius Pang
Dr Peter Riek, PhD

Collaborators:
Professor Phil Hogg, UNSW
Dr Isidore Rigoutsos, IBM
Robert Graham, FAA MD FRACP FACP FAHA; VCCRI, Sydney, Australia
Dr Sally Dunwoodie, VCCRI
Dr Diane Fatkin, VCCRI
Dr Siiri Iismaa, VCCRI,
Dr Angela Finch, VCCRI

Publications:
Wouters MA, Curmi PM.  An analysis of side chain interactions and pair correlations within antiparallel beta-sheets: the differences between backbone hydrogen-bonded and non-hydrogen-bonded residue pairs. Proteins 1995; 22:119-31

Wouters MA, Husain A. Changes in zinc ligation promote remodeling of the active site in the zinc hydrolase superfamily. J Mol Biol 2001; 314:1191-207

Liu X, Fernandez M, Wouters MA, Heyberger S, Husain A. Arg(1098) is critical for the chloride dependence of human angiotensin I-converting enzyme C-domain catalytic activity. J Biol Chem 2001; 276:33518-25

Fernandez M, Liu X, Wouters MA, Heyberger S, Husain A.  Angiotensin I-converting enzyme transition state stabilization by HIS1089: evidence for a catalytic mechanism distinct from other gluzincin metalloproteinases. J Biol Chem 2001; 276:4998-5004 

Riek RP, Rigoutsos I, Novotny J, Graham RM. Non-a-helical elements modulate polytopic membrane protein architecture. J Mol Biol 2001; 306:349-362

Chen S, Lin F, Xu M, Riek P, Novotny J, Graham RM. Mutation of a single TMVI residue, Phe282, in the b2 adrenergic receptor results in a structurally distinct activated receptor conformation. Biochemistry 2002; 41:6045-6053

Rigoutsos I, Riek P, Graham RM, Novotny J.  Structural details (kinks and non- a conformations) in transmembrane helices are intrahelically determined and can be predicted by sequence pattern descriptors. Nuc Acids Res 2003; 31:4625-4631

Wouters MA, Liu K, Riek P, Husain A.  A despecialization step underlying evolution of a family of serine proteases. Mol Cell 2003; 12:343-354

Iismaa SE, Holman S, Wouters MA, Lorand L, Graham RM, Husain A. Evolutionary specialization of a tryptophan indole group for transition-state stabilization by eukaryotic transglutaminases. Proc Natl Acad Sci U S A 2003; 100:12636-12641

Whittock NV, Sparrow DB, Wouters MA, Sillence D, Ellard S, Dunwoodie SL, Turnpenny PD.  Mutated MESP2 causes spondylocostal dysostosis in humans. Am J Hum Genet 2004; 74:1249-54

Back to top