Similarly, physical organic experiments in
which a given chemical reaction was performed
in various solvents, generally ranging from aprotic to protic, showed especially for
many SN2 displacement reactions that the
effect of solvent is to retard the rate relative
to what would be observed for this reaction
under the same conditions in the gas phase.
Extrapolation to the low dielectic typical of
an enzyme’s active-site cavity (18) prompted
early proposals that enzymes act through a
desolvation mechanism (19–21). Alternatively,
the effect may be described as solvent
substitution, with the active-site residues furnishing
a polar framework to replace the
solvating water molecules (22).
Collectively, these experiments suggest that
enzymatic catalysis could be understood in
terms of physical organic principles. One striking
feature, seen repeatedly, is that the catalytic
elements in an active site are precisely positioned
for their function. A beautiful, early
example is furnished by the enzyme triosephosphate
isomerase, which catalyzes the interconversion
of glyceraldehyde 3-phosphate and
dihydroxyacetone phosphate through a cis
enediol intermediate assisted by general acidbase
catalysis (Fig. 2). From kinetic, stereochemical,
chemical modification, and sitespecific
mutagenesis experiments, B had been
identified as Glu165 and HA as His95 (23). Thecrystal structure (Fig.3) with either a substrate or substrate
analog showed that
arrangement of substrate
relative to the
catalyzing residues
was indeed one of remarkable
precision:
The carboxylate is
within 3 Å of enediol
carbons, requiring
minimal motion of the
bidentate carboxylate to shuttle the proton, and
the His95 is within 3 Å of the substrate oxygens,
allowing it to donate a proton (24). [For further
ramifications of the positioning, see (23).] The
above precision facilitates passage through the
transition state and thus provides a more satisfying
picture of what might be meant by
transition-state stabilization.