[Metalloprotein Structure and Design Home page]
[ TSRI Home page]
Project III: Protein Metal Ion Site Design Algorithms
Michael E. Pique,
Victoria A. Roberts,
Homme Hellinga,
Jesus M. Castagnetto
Abstract
The long-term objectives of this project are criteria, algorithms, and tools
for the design of metal binding sites in proteins. Bound metal ions in
proteins perform both catalytic and structural functions. Introducing metal
sites into any desired protein could confer new catalytic functions, improve
stability, or create new diagnostic and therapeutic agents. Currently, each
protein requires a different approach to metal-site design. To allow
testing of specific design hypotheses and to verify our understanding of
metal site requirements, a general, objective predictive algorithm is
required. Given that the basic weakness in rational metalloprotein design
is in the accuracy of the predictions from calculation and analysis, a
separate new research project focused on the development of reliable
criteria, predictive algorithms, and computer-based tools is highly
desirable.
- In Aim 1:
- A database of NMR and X-ray structures will be constructed and
analyzed to categorize sites and devise criteria for design in the form of
three-dimensional stereochemical templates.
- In Aim 2:
- The algorithms for site search and neighborhood side chain
compatibility optimization originally devised by participant Homme
Hellinga will be applied, tested, and improved through the site design and
characterization efforts of the Program Project. The design model takes
into account not only direct metal-ligand geometry and interactions, but
also neighborhood `second shell' side chains whose compatibility with the
site affects folding, binding, and function.
- In Aim 3:
- Computational and interactive graphics tools to shorten the
design-test-redesign cycle will be built in the productive framework of a
high-level Protein Engineering Programming Environment.
The research design is iterative, with design "failures"uated as
design intermediates, whose experimental characterization by
spectroscopy and crystallography will highlight previously unanticipated
aspects of metal site construction. The cross-breeding of empirical data
and computational predictions will allow us to discriminate essential from
accidental properties of sites and enhance scientific understanding of the
construction of metal-binding sites with defined properties. This project
integrates scientists and software developers from
The Scripps Research Institute and
Duke University to focus on computer-aided
metal site design principles in a Program Project context that promotes
their experimental verification.