Chemistry Reference
In-Depth Information
unreactive at least to the extent that they reacted, if at all, with rate constants of
<
10
5
M
1
s
1
, the limit of detection for the equipment used. Most surprising was the
observation that glycine displayed a nonlinear pseudo-first-order-decay rate depen-
dence on concentration with significant reactivity at high concentration. This effect
was not explained, but does indicate a problem with relative rate studies of this type
when applied to large biological molecules. If a PAL agent is photoactivated while in
the binding site of an enzyme, the effective concentrations of all amino acid residues
in that binding site will be extremely high, and the reactivity of the encased nitrene or
nitrenium ion much higher that might be expected based on dilute-solution mea-
surements of this type.
Several other interesting aspects of the data in Table 3.4 are that the proteins do not
display reactivities that parallel their reactive amino acid content. This is perhaps not
surprising, since only those amino acid residues on the surface of the proteins will be
available for PAL cross-linking. What strikes this observer as very interesting is that
the order of amino acid reactivity seems to be more consistent with their relative
electron-donating abilities rather than their nucleophilicities. Falvey and Thomas
made note of this same correlation, which is especially obvious for methionine,
which affords nitrenium ion reduction products.
78
Therefore, the second-order rate
constants listed in Table 3.4 are the combined rate constants for both reduction and
nucleophilic substitution, and since electron transfer/nitrenium-ion reduction is
usually not a productive branch of nitrenium ion chemistry in PAL studies, these
rate constants do not necessarily reflect the cross-linking susceptibilities of the
various amino acids listed. Finally, the order of reactivity of nitrenium ions generated
directly, as in this study, does not accurately reflect the order of reactivity of
nitrenium ions generated from azides via nitrenes, since this latter process will
depend on the acidities of the conjugate acids of the nucleophiles that cross-link with
the nitrenium ions, and will proceed through ion pairs incorporating the cross-linking
nucleophiles.
There are a large number of PAL studies that have been conducted with aryl
azides, many or most of which may proceed via the ring expansion ketenimine
branch (Scheme 3.7), while only one class of aryl azides is known with certainty to
proceed via the nitrenium ion branch, 4-azido-2-nitrophenyl amines related to
24b
(Table 3.3). We shall focus this initial discussion on studies that have used these
nitrophenyl azides.
A very interesting comparison of the relative performance of these two classes of
azides has been done in an effort to link together the filaments of the protein F-actin.
F-actin is a protein composed of globular subunits, G-actin containing 375 amino
acid residues, that in conjunction with myosin molecular motors is responsible for
the contraction and relaxation of muscles. The studies in question sought to restrict
this molecular motion between F-actin strands by cross-linking G-actin subunits
together. The two phenyl azide-modified proteins
27
and
28
have been applied in
these studies (Scheme 3.14).
In the case of
, one would expect the ketenimine branch of cross-linking to be
active, and it was noted that the cross-linked peptide chains were susceptible to
hydrolysis during the HPLC separation procedures used in their isolation.
79
27
