Computational Structural Biology 
lation and computer hardware. Nevertheless, 
macromolecular simulation has been successful 
in predicting localized conformations if suffi- 
cient experimental constraints or restraints (e.g., 
in the form of an x-ray structure) are available. It 
is therefore conceivable that other more global 
predictions are possible if appropriate experi- 
mental information is available. We have em- 
barked on trying to predict the association and stabil- 
ity of helices that are believed to form coiled coil 
conformations. Conformational search strategies are 
being employed with empirical energy functions, 
using molecular dynamics and energy minimization 
starting with generic coiled coil-forming a-helices. 
Presently we are applying this approach to the 
family of leucine zipper proteins, which are se- 
quence-specific DNA-binding proteins that regu- 
late gene expression in certain mammalian cells. 
We have predicted the structure of the dimeriza- 
tion domain of GCN4, for which a high-resolu- 
tion x-ray structure will be available shortly (per- 
sonal communication, Thomas Alber, University 
of Utah). Furthermore, we are trying to explain 
why Jun can form a homodimeric protein that 
binds to DNA, whereas Fos is unable to do so and 
only occurs in a Jun-Fos heterodimer. We are also 
applying this approach to glycophorin A and 
other membrane proteins that are believed to 
form predominantly helical structural elements 
crossing the membrane. Interesting results were 
obtained during the past year for vacuum simula- 
tions. Currently we are attempting to incorporate 
the environment (water and lipid molecules) 
into the simulation. A rapidly increasing amount 
of experimental data on mutants of the systems 
studied will allow a thorough comparison of the 
theoretical predictions and the experimental 
results. 
Macromolecular Simulation of 
Free-Energy Differences 
We are involved in a number of projects that 
are aimed at simulating free-energy differences 
between two states of a biological system, using 
the free-energy perturbation technique. Our 
goals are to investigate microscopically the struc- 
ture and stability of protein secondary structural 
elements and protein-peptide complexes and to 
evaluate the reliability of free-energy calcula- 
tions and molecular dynamics simulations as 
tools for this purpose. In particular, we are study- 
ing 1) the complexes of bovine pancreatic ribo- 
nuclease S and a number of mutants of the S pep- 
tide for which x-ray crystal structures, binding 
free energies, and enthalpies have been obtained 
by Frederic Richards, Julian Sturtevant, and their 
colleagues (Yale University) and 2) the com- 
plexes of L-tryptophan and several of its chemical 
analogues with Escherichia coli trp aporepres- 
sor, for which x-ray crystal structures and binding 
free energies have been measured by Paul Sigler 
(HHMI, Yale University) and his colleagues. 
We also plan to carry out free-energy calcula- 
tions on mutants of /3-turns in staphylococcal nu- 
clease, in a joint project with Robert Fox (HHMI, 
Yale University) . High-resolution x-ray structures 
of various mutants and thermodynamic data 
through temperature-dependent NMR studies 
have been obtained. Specifically, we would like 
to simulate the free-energy differences between 
the cis and trans conformations of Pro- 1 1 7 for a 
number of point mutations in the vicinity of the 
proline residue. 
66 
