Chemistry Reference
In-Depth Information
of Tris the metal ion simply catalyzed the hydration of 2-cyanopyridine to its amide, as
with
22
above. The coordination of both reactants to the same metal ion has produced
selectivity that was induced by the coordination.
In models for carboxypeptidase A we showed the intracomplex catalyzed hydrolysis
of an ester by a metal ion and a carboxylate ion [106], which are the catalytic groups of
carboxypeptidase A. Some mechanistic proposals for the action of carboxypeptidase
involve an anhydride intermediate that then hydrolyzes to the product and the regen-
erated enzyme. Although we later found convincing evidence that the enzyme does not
use the anhydride mechanism in cleaving peptides [96-99], it may well use such a
mechanism with esters. In a mimic of part of this mechanism we showed [107],
but see also Ref. 108, that we could achieve very rapid hydrolysis of an anhydride
by bound Zn
2+
, which is the metal ion in the enzyme. In another model, a carboxylate
ion and a phenolic hydroxyl group, which are in the enzyme active site, could coop-
eratively catalyze the cleavage of an amide by the anhydride mechanism [109].
Catalysis by Zn
2+
is ambiguous, since it is not clear whether the metal ion is coor-
dinated to the carbonyl group of the substrate, as it is in the enzyme. Thus, we exam-
ined the cleavage of an amide by a combination of Co
3+
and a carboxylate group. Since
Co(
III
) is “substitution inert,“ we prepared a complex in which it was directly coordi-
nated to the carbonyl oxygen of a substrate amide, and in which a neighboring car-
boxylate ion or phenol group was a potential second catalytic function [110]. Indeed,
the phenol group was able to assist the hydrolysis by protonating the leaving amino
group, but the carboxylate ion was not effective. The mechanism is intellectually re-
lated to that used by the enzyme [97, 98], but the details differ.
By contrast with the absence of catalysis by an internally attached carboxylate ion in
the above study, an external carboxylate species and, even more, an external phosphate
species were catalysts with the cobalt complexed amide. Thus we examined a process
like that just described in which a phosphonate group was internally attached to the
cobalt complex [111]. In this case it did act as a sequential base/acid catalyst, as
the carboxylate ion does in the enzyme, i.e., it first delivered a hydroxide group to
the complexed carbonyl, acting as a base, and then the proton that it had accepted
was delivered to the leaving group, with the phosphonic acid acting as a general
acid. This sequence is indeed like that in the enzyme. We suggested that the phos-
phonate is more effective because it has a higher pK
a
than the attached carboxylate,
and in the enzyme the carboxylate has an abnormally high pK
a
. The phosphonate is a
better model for the enzyme carboxylate in the artificial enzyme.
These cobalt systems are useful models of enzymatic mechanisms, but they are not
turnover catalysts as enzymes are. To achieve turnover we constructed ligand
25
in
which a metal-coordinating group links two cyclodextrin rings. As its metal complex
it was a good catalyst for the hydrolysis of substrates
26
and
27
that could bind into both
cyclodextrins and stretch across the bound metal ion. (Later we will describe the same
principle applied to selective oxidation reactions.) Ligand
25
as its Cu
2+
complex gave
as much as a10
5
-fold rate acceleration in the ester hydrolysis [112, 113]. With an added
nucleophile that also binds to the Cu
2+
ion, the reaction is accelerated by over 10
7
. The
mechanism deduced (
28
) - in which the metal ion acts as a Lewis acid by coordination
to the substrate carbonyl and also delivers a bound hydroxide ion to the ester carbonyl
