About us - an introduction to lab research

The prediction and design of protein folds, and the application of this in bionanotechnology and synthetic biology.

The primary basic research interest of the group is the informational aspect of the protein-folding problem; that is, how does the sequence of a protein determine its active, three-dimensional structure or fold?

We tackle this problem using the following multi-disciplinary approach:

  1. We use bioinformatics to garner sequence-to-structure relationships from protein sequence and structural databases.
  2. We test the relationships ("rules for protein folding") that we find in two ways: (a) through ab initio protein-structure prediction; and (b) via rational protein design, where we engineer natural protein structures, or design new ones completely from scratch (so-called de novo design).
  3. We then test our engineered and design proteins experimentally using biophysical methods. The peptides and proteins are made either by peptide synthesis, or via recombinant DNA methods and the expression of synthetic genes. The products and then characterised using methods including: solution-phase biophysics (CD, FT-IR and fluorescence spectroscopy, AUC and ITC); high-resolution structural biology (NMR spectroscopy and X-ray crystallography); and microscopy (EM, AFM and light microscopy).
  4. Finally, we explore potential applications of some the engineered and designed proteins in the burgeoning fields of bionanotechnology and synthetic biology.

For instance, we are developing self-assembled fibrous materials, including hydrogels, to act as substitutes for the extracellular matrix to foster specific cell growth for applications in tissue engineering; and peptide-based conformational switches as potential components of biosensors.

To help in all of these projects, and to make the design and engineering of biological systems easier in general, we are developing a bank of basic peptide-folding components (Pcomp). We refer to this as a biomolecular design approach in synthetic biology, and see it as one of many different paths through the possible space of engineering and designing new biological systems (see Figure).

Merge sequence for a designed switch between a trimeric coiled coil and a zinc-finger structure

Synthetic-biology space. Various approaches in synthetic biology are resolved according to where they appears in the natural biomolecular hierarchy (y-axis), and by some measure of how synthetic it is (x-axis). The arrows indicate approximate routes through synthetic-biology space taken by studies in any of four different approaches: genome-engineering, biomolecular-engineering, biomolecular-design and artificial protocell design. Blue arrows indicate approaches usually conducted in vivo and orange arrows indicate in vitro approaches. On the left, the various levels in the natural hierarchy and their ranges are described, along with illustrative natural examples from various points in the hierarchy. Taken from references 57 & 59.

More specifically, we are interested in the folding and assembling of a number of protein-folding motifs, including zinc fingers and beta-structured proteins. However, our recent interest and successes have focused the a-helical coiled-coil motif. Coiled coils are protein-folding motifs that direct and cement a wide variety of protein-protein interaction throughout biology. They comprise two or more a-helices that wrap around one another to form helical ropes. Despite their apparent simplicity, these structures are ubiquitous and account for between 5 - 10% of all coding DNA sequence. The bioinformatics challenge is to decipher rules within coiled-coil sequences that determine the different structures that are possible, and discriminate between different coiled-coil partners. It is these motifs and, more importantly, the sequence-to-structure rules that underlie them that the group examines and uses in de novo design.

To help our studies of coiled coils, we have developed an algorithm, SOCKET, to identify coiled-coil domains in the structures of natural proteins, and have used this to create a relational database, CCPLUS, and Periodic Table of Coiled-coil Structures of all the natural coiled-coil domains from the Protein Data Bank (PDB). In turn, we have interrogated the database to garner rules for coiled-coil folding and assembly to guide our de novo coiled-coil designs. Successful designs include, basic multimeric helical "hubs" and "linkers", peptides that "switch" conformational state, and self-assembling fibrous materials.