C O M M U N I C A T I O N S
Scheme 2. Synthesis of the Michael Acceptor-Containing
Pantothenamides 8a-e and CoA Analogues 2a-e
Figure 1. Inhibition of CoADR by CoA analogue 2e. (A) Reaction progress
in the presence of increasing inhibitor concentrations, showing time-
dependent inactivation of CoADR. (B) Plot of the observed rate of
inactivation constants (kobs) versus the concentration of 2e, from which the
second-order rate of inactivation constant was determined. (C) Scheme
showing the two-step mechanism of irreversible inactivation of CoADR
that is in operation in the case of 2e. Results for inhibition by analogues 2a
and 2d are shown in Figures S11 and S12 respectively.
Table 1. Kinetic Parameters for the Inhibition of CoADR by CoA
Analogues
Kia
(µM)
k
inact/KIb
(s-1 ·M-1
CoA analogue
)
2a
2b
2d
2e
0.66 ( 0.12
5.16 ( 0.96
0.30 ( 0.05
0.04 ( 0.01
219.1 ( 45.5
ndc
500.2 ( 89.8
39 690 ( 10 980
The finding that the CoA analogues 2a, 2b, 2d, and 2e bind in
the active site of CoADR suggests that their Michael acceptor
moieties should be ideally positioned to inactivate the enzyme by
conjugate addition as envisaged (Scheme 1B). To determine whether
the formed enzyme-inhibitor complexes result in the irreversible
inhibition, time-dependence analyses were performed on the most
potent inhibitors (i.e., 2a, 2d, and 2e) by using the progress curve
method (Figures 1A, S11A, and S12A).8 Gratifyingly, the progress
curves of all three analogues showed time-dependent irreversible
a For competitive inhibition. b Second-order rate of inactivation
constants. c nd, Not determined.
analogue 2c, although it was successfully prepared by biotransfor-
mation of 8c, decomposed upon purification. The pantothenamides
8a-e and the analogues 2a-e were subsequently assayed for
inhibition of CoADR using a concentration of 200 µM (2c was
tested in crude form). While all five CoA analogues showed
inhibition of CoADR activity, none of the pantothenamides had
any effect, highlighting the essential requirement of CoA’s adenos-
ine and phosphate moieties for recognition and binding.
inactivation of CoADR, with the observed rate of inactivation (kobs
)
also increasing with increasing inhibitor concentration (Figure 1B).
In addition, the hyperbolic shape of this plot confirms that a two-
step inactivation mechanism (Figure 1C) is at play in the case of
2e. This provides further explanation for the competitive inhibition
that is observed in the previous experiment, as the determined Ki
values would refer to the dissociation constant (Ki ) k4/k3) of the
enzyme-inhibitor (EI) encounter complex formed in the reversible
first step. The second-order rate of inactivation constants (Table 1)
determined from these plots of kobs vs [I] showed that, while
analogues 2a and 2d exhibited modest rates of inactivation, the
phenyl sulfone 2e is a much more potent inhibitor with a rate
constant of ∼40 000 s-1 ·M-1. This relative order of inhibition
activity is in agreement with the results of a previous study of the
relative rates of the conjugate addition of 2′-(phenethyl)thiol to
various Michael acceptors, which found the reactivity of a phenyl
vinyl sulfone to be higher than that of the corresponding R,ꢀ-
unsaturated methyl ester.9
To further confirm the irreversibility of inhibition, a CoADR
sample was incubated in the presence of analogue 2e, followed by
gel filtration to remove all unbound small molecules. As expected,
the inhibitor-treated enzyme showed no activity in comparison to
a negative control sample treated in the same manner. Interestingly,
the inhibitor-treated, gel-filtered enzyme did show a very slow return
of activity after ∼10 min of incubation (Figure S14). This suggests
that the enzyme-inhibitor linkage can be broken over time, most
probably by an elimination reaction that regenerates Cys43 and the
inhibitor. However, in light of the progress curve data the rate of
regeneration is seemingly negligible relative to that of inactivation.
In spite of the good CoADR inhibition shown by CoA analogues
such as 2e, these highly polar compounds cannot be used as in
ViVo growth inhibitors since they are not be able to cross the
bacterial cell membrane. However, previous studies have shown
Although the CoA analogues 2a-e were designed to act as
selective, irreversible inhibitors of CoADR by modification of its
active site cysteine, it is also possible that the observed inhibition
can occur by nonspecific reaction of the Michael acceptor moieties
with other enzyme-derived nucleophiles. To demonstrate that these
analogues bind specifically in the active site of CoADR, the
analogues 2a, 2b, 2d, and 2e were therefore evaluated for their
ability to compete with CoAS2 in the CoADR reaction. This was
done by determining the initial rates of the reaction (i.e., without
preincubation) in the presence of increasing substrate concentrations
and various set concentrations of the inhibitor.7 The resulting rate
profiles and corresponding double reciprocal plots (Figures S7-S10)
indicated that the inhibition is indeed competitive in all cases.
Moreover, with one exception the determined Ki values (Table 1)
are all submicromolar, indicating that these CoA analogues are
excellent substrate mimics. In fact, the Ki value of ∼40 nM
exhibited by the most potent inhibitor, the phenyl sulfone 2e, is
nearly 50-fold lower than the Km(CoAS2) of ∼2.0 µM. Based on
this limited set of compounds, a structure-activity relationship
analysis for binding of these analogues in the CoADR active site
suggests that sulfone-based analogues are better suited than their
carboxylic acid ester counterparts and that small substituents (e.g.,
OEt, Me) are preferred over sterically bulky ones (OtBu). In light
of this analysis the excellent inhibition seen for the phenyl sulfone
2e is therefore surprising, although it is possible that this analogue
is uniquely able to form π-stacking interactions with the side chain
of aromatic residues located nearby, such as Tyr361.3 Nonetheless,
these results show that the CoADR active site can accommodate a
variety of substituents on the Michael acceptor and suggest that
their scope and diversity may be expanded in future studies.
9
12854 J. AM. CHEM. SOC. VOL. 132, NO. 37, 2010