K. Olsen, S. Fischer, M. Karplus.
Using
the Conjugate Peak Refinement method (Fischer & Karplus, Chem. Phys.
Let. (1992) 194:252), we have calculated the first continuous and
energetically plausible path for the allosteric transition of hemoglobin.This
simulation gives a sequence of events linking the T- and R-states in atomic
detail, which can be compared to the stereochemical mechanism proposed
by Perutz (Nature (1970) 228:726).We
find that the important salt bridges break early before the HC2 Tyr residues
move out of their pockets and before the major quaternary change, in which
the b1His97
moves past the a2Val41
(the switch).This switch at the a1b2
interface is synchronous with the corresponding switch at the a2b1
interface.While these changes occur
as predicted by Perutz, there are several aspects of the transition that
could not have been anticipated by just comparing the crystallographic
end-states.The distal His of the b-chains
move out of the oxygen binding pocket and form a hydrogen bond to the mainchain
carbonyl of bLys59
early in the path and then return to their original positions after the
switch.The global C2 symmetry between
the two ab
dimers is retained during the entire transition, while most of the quaternary
changes in the two a-chains
occur significantly before the quaternary changes in the two b-chains.Visual
inspection of the transition reveals that the G-helix of each a-chain
serves as the rotation axis for its ab
dimers.