|Budget Amount *help
¥2,200,000 (Direct Cost : ¥2,200,000)
Fiscal Year 1995 : ¥600,000 (Direct Cost : ¥600,000)
Fiscal Year 1994 : ¥1,600,000 (Direct Cost : ¥1,600,000)
A quantitative measure of the structural stability of proteins is the difference in free energy DELTAG_d between their native (N) and denatured (D) states. Hence, to identify and estimate the factors contributing to DELTAG_d, it is indispensable to obtain structural information not only on the N state but also on the D atate. It has been established from studies using various structural probes that there exists some intermediate states such as the molten globule (MG) state between the N and the completely unfolded (U) states, and that the D-state structure is very diverse. Complementary to spectroscopic methods, the solution X-ray scattering (SXS) method is a powerful method for studies on the nonnative state which is essentially the multiconformational state, but it cannot determine protein structure by itself. In thisstudy, we developed methods to supplement the above weakness of SXS method, and analyzed the structures of proteins in the nonnative states :
(1) Taking account of the hy
dration of polar groups and the probability distribution of internal rotation angles of peptide chains, we devised a method for generating various conformations in computer.
(2) Extending the above, we developed a method to generate conformations for the multipartite (MP) structure model of the nonnative state, where a polypeptide chain is composed of a segment having the structure identical to that in the N state and segments having totally unfolded structure.
(3) We devised a new algorithm for the SXS simulation, where the X-ray scattering profile is evaluated by explicitly considering the contribution from solvent under the continuum approximation of solvent.
(4) Applying the above method to several natural proteins and unfolded chains, we confirmed its performance and effectiveness, and showed the importance of accurately considering solvent influences.
(5) Applying the MP model to three proteins in the MG state, we compared our prediction with reported data and have confirmed that their SXS profiles, especially that of cytochrome c, can be well explained by this model. However, slight difference has been found for alpha-lactalbumin and myoglobin, indicating necessity to refine the model for them. Less