Tight binding inhibitors of N-acyl amino sugar and N-acyl amino acid deacetylases

Article abstract of DOI:10.1021/ja0600680

Very potent inhibitors were synthesized for the enzymatic deacetylation of N-acetyl-D-glucosamine-6-phosphate (NagA) and N-acetyl-D-glutamate (DGD). The methyl phosphonamidate derivative of D-glucosamine-6-phosphate bound to N-acetyl-D-glucosamine-6-phosp

Full text of DOI:10.1021/ja0600680

Published on Web 03/09/2006
Tight Binding Inhibitors of N-Acyl Amino Sugar and N-Acyl Amino Acid
Deacetylases
Chengfu Xu, Richard Hall, Jennifer Cummings, and Frank M. Raushel*
Department of Chemistry, P.O. Box 30012, Texas A&M UniVersity, College Station, Texas 77842-3012
Received January 4, 2006; E-mail: raushel@tamu.edu
Scheme 2. Methyl Phosphonamidate Analogues of Intermediates
Enzymes within the amidohydrolase superfamily have been
shown to catalyze the hydrolysis of amide and ester bonds to carbon
and phosphorus centers in a wide range of substrates, including
nucleic acids, carbohydrates, and amino acids.1 The hydrolytic water
molecule is activated in these enzymes by direct coordination to a
mononuclear or binuclear metal center in the active site. A subset
of the enzymes in this superfamily catalyzes the hydrolysis of acetyl
groups from N-acyl amino acids or amino sugars. In an effort to
more precisely define the role of the divalent cations in these
systems, we have endeavored to design and synthesize structural
mimics of the putative reaction intermediates. These compounds
should prove useful in the elucidation of enzyme-substrate
interactions and as lead compounds in drug discovery efforts.
The enzymatic hydrolysis of acetyl groups from N-acyl-
substituted substrates may proceed via the direct attack on the amide
bond by an activated water molecule. Compounds that mimic the
structure and geometric constraints of the tetrahedral intermediate
have proven to be useful probes of the reaction mechanism and
tight binding enzyme inhibitors.2 Phosphonamidates are particularly
useful structural surrogates because of the tetrahedral geometry at
the phosphorus center and the similarities in charge and bond
lengths to those predicted to occur during the hydrolysis of amide
bonds.3,4
and similar to that determined previously for D-amino acid
deacetylase from Alcaligenes faecalis DA1.5
Methyl phosphonamidate analogues of the tetrahedral intermedi-
ates for the reactions catalyzed by NagA and DGD were synthe-
sized, and the structures (1 and 2) are presented in Scheme 2. These
compounds are potent and time-dependent inhibitors of NagA and
DGD. The inhibition constant for NagA was determined by
incubating variable amounts of 1 with 10-45 nM of the enzyme
for 30 min in 50 mM phosphate buffer, pH 7.5. The enzyme activity
was determined using 0.5 mM N-acetyl-D-glucosamine-6-phosphate
as the substrate. The residual activity was plotted versus the inhibitor
concentration, and the results for 25 nM NagA are presented in
Figure 1A. A fit of the data to eq 1 for the entire data set gave an
average inhibition constant, Ki, of 34 ( 5 nM.6 In eq 1, Et is the
total enzyme concentration, I is the inhibitor concentration, Vo is
the activity of the enzyme in the absence of inhibitor, and Vi is the
residual activity of the enzyme in the presence of inhibitor.
Scheme 1. Reactions Catalyzed by NagA and DGD
Two examples of deacetylases that belong to the amidohydrolase
superfamily are N-acetyl-D-glucosamine-6-phosphate deacetylase
(NagA) and N-acetyl-D-glutamate deacetylase (DGD). The reactions
catalyzed by these two enzymes are presented in Scheme 1. NagA
from Escherichia coli was purified to homogeneity after expression
of the cloned gene (gi: 16128653) in E. coli. The purified protein
was found to contain 1 equiv of Zn2+ that is likely bound to the
Mâ site. The values of kcat and kcat/Km were determined to be 78
s-1 and 5.7 × 105 M-1 s-1, respectively, at pH 7.5, 30 °C by
following the loss of substrate at 215 nm. The gene for DGD was
cloned from Bordetella bronchiseptica (gi: 33602261) and over-
expressed in E. coli. The purified protein was assayed by following
the formation of formate with formate dehydrogenase at 340 nm
using N-formyl-D-glutamate as the substrate. The values of kcat and
kcat/Km were determined to be 400 s-1 and 1.7 × 106 M-1 s-1 at
pH 7.6 and 30 °C. The enzyme was found to contain 2 equiv of
Zn2+ per subunit, and thus the active site is likely to be binuclear
Figure 1. Inhibition of NagA (panel A) and DGD (panel B) by 1 and 2,
respectively. The kinetic data were fit to eq 1 to obtain a dissociation
constant for 1 with NagA of 34 nM and 460 pM for the binding of 2 with
DGD. Additional details are given in the text.
The time dependence to the onset of inhibition by 1 with NagA
was determined by measuring the rate of bond cleavage as a func-
tion of time. In these experiments, 10 nM NagA was added to a
solution of 0.5 mM substrate containing 0.4-2.4 µM 1. The time
courses for the change in the rate of substrate hydrolysis are shown
in Figure 2A. The time courses were fit to eq 2 to yield the observed
first-order rate constants for the onset of inhibition by 1 with NagA.
In eq 2, P is the product concentration, Vr and Vs are the initial and
final rates of product formation, and kobs is the rate constant for
the onset of the slower rate of product formation.7 The apparent
second-order rate constant for the formation of the inhibited form
9
4244
J. AM. CHEM. SOC. 2006, 128, 4244-4245
10.1021/ja0600680 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
similar instances that the slow step is due to the sluggish dissociation
of a metal-bound water molecule that is necessary before the
inhibitor can make optimal interactions within the active site.3a The
pH dependence and solvent isotope effects for inhibition of
stromelysin by a peptide phosphonamidate suggest that kon is limited
by ligand exchange for a zinc-bound water.8 It is likely that the
small value of k1 for the binding of 1 with NagA is due to a similar
phenomenon.
Scheme 3. Models for Binding of Enzyme Inhibitors
Figure 2. Time dependence for the onset of inhibition by 1 (panel A) with
NagA and 2 (panel B) with DGD. In panel A, the concentrations of 1 are
0.40 µM (O); 0.80 µM (0); 1.6 µM (3); 2.4 µM (4). In panel B, the
concentrations of 2 are 15 nM (O); 25 nM (0); 50 nM (3); 80 nM (4).
The time courses were fit to eq 2 to obtain the first-order rate constants for
the formation of the enzyme-inhibitor complex.
The two phosphonamidate analogues synthesized for this inves-
tigation proved to be extraordinary potent inhibitors of two
deacetylases belonging to the amidohydrolase superfamily. The
inhibition constants for compounds 1 and 2 are 4 × 103- and 5 ×
105-fold smaller than the respective Michaelis constants of their
target enzyme. The Ki values for these compounds are among the
lowest ever measured for the inhibition of deacetylases.9
of NagA was determined by plotting kobs versus the concentration
of 1 as shown in Figure 3A. From the slope of this plot, the apparent
second-order rate constant is 1.3 × 103 M-1 s-1. Correction of the
apparent second-order rate for the concentration of substrate accord-
ing to eq 3 gave a value for k1 of 6.1 × 103 M-1 s-1 (see Scheme 3).
The dissociation constant for the binding of 2 with DGD was
determined by incubating 2.0 nM enzyme with 0.5-60 nM 2 for
75 min at pH 7.6. The residual enzyme activity was then determined
with 3.0 mM N-formyl-D-glutamate as the substrate in 50 mM
HEPES buffer, pH 7.6. The results are shown in Figure 1B and a
fit of the data to eq 1 gives a Ki of 460 ( 70 pM. The time courses
for the onset of inhibition by 2 were determined by measuring the
rate of bond cleavage as a function of time. In these experiments,
1.0 nM DGD was added to a solution of 1.5 mM substrate
containing between 10 and 300 nM of 2 at pH 7.6, and the formation
of formate was determined as a function of time. The time courses
were fit to eq 2 to yield the first-order rate constants for the onset
of inhibition, and the results are presented in Figure 2B. The
apparent first-order rate constants for the onset of inhibition were
plotted versus the concentration of 2, and the results are presented
in Figure 3B.
Vi/Vo ) ([Et] - Ki - [I] + (([I] + Ki - [Et])2 +
(4Ki[Et]))1/2)/(2[Et]) (1)
obst)
P ) Vst + ((Vr - Vs)/kobs)(1 - e(-k
k1 ) appkobs (1 + [S]/[Km])
)
(2)
(3)
k3 ) (kobs - k4) (1 + (app(k2/k1)/[I]))
(4)
Acknowledgment. This work was supported in part by the NIH
(GM 33894 and GM 71790).
Supporting Information Available: The syntheses of compounds
1 and 2 are described. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) (a) Holm, L.; Sander, C. Proteins: Struct., Funct., Genet. 1997, 28, 72-
82. (b) Seibert, C. M.; Raushel, F. M. Biochemistry 2005, 44, 6383-
6391.
(2) Frick, L.; Wolfenden, R. In Design of Enzyme Inhibitors as Drugs; Sandler,
M., Smith, H. J., Eds.; Oxford University Press: Oxford, 1989; pp 19-
48.
(3) (a) Bartlett, P. A.; Marlowe, C. K. Biochemistry 1987, 26, 8553-8561.
(b) Holden, H. M.; Tronrud, D. E.; Monzingo, A. F.; Weaver, L. H.;
Matthews, B. W. Biochemistry 1987, 26, 8542-8553. (c) Morgan, B. P.;
Holland, D. R.; Matthews, B. W.; Bartlett, P. A. J. Am. Chem. Soc. 1994,
116, 3251-3260. (d) Mallari, J. P.; Choy, C. J.; Hu, Y.; Martines, A. R.;
Hosaka, M.; Toriyabe, Y.; Maung, J.; Blecha, J. E.; Pavkovic, S. F.;
Berkman, C. E. Bioorg. Med. Chem. 2004, 12, 6011-6020.
(4) Radkiewicz, J.; McAllister, M. A.; Goldstein, E.; Houk, K. N. J. Org.
Chem 1998, 63, 1419-1428.
Figure 3. Effect of inhibitor concentration on the onset of inhibition by 1
(panel A) with NagA and 2 (panel B) with DGD. The data in panel B were
fit to eq 4. Additional details are provided in the text.
The apparent rate constant for the slow onset of inhibition by 2
with DGD is constant at high levels of inhibitor. These results
suggest that the initial binding event is followed by a subsequent
transformation that enhances the tightness of the interaction between
DGD and 2, as shown in the second model that appears in Scheme
3. A fit of the data from Figure 3B to eq 4 gives values of k3 )
0.017 s-1 and an apparent k2/k1 ) 140 ( 20 nM. The value of k4
was undetermined. Correcting for the concentration of substrate
gives a value for k2/k1 of 19 ( 3 nM. The second step in the binding
of 2 with DGD may be the result of a protein conformational change
that enhances the molecular interactions between the protein and
transition state analogue. Bartlett and Marlowe have proposed for
(5) Liaw, S. H.; Chen, S. J.; Ko, T. P.; Hsu, C. S.; Chen, C. J.; Wang, A. H.
J.; Tsai, Y. C. J. Biol. Chem. 2003, 278, 4957-4962.
(6) (a) Morrison, J. F. Biochim. Biophys. Acta 1969, 185, 269-286. (b) Scully,
M. J.; Morrison, J. F.; Cleland, W. W. Biochim. Biophys. Acta 1996, 1298,
78-86.
(7) Copeland, R. A. EValuation of Enzyme Inhibitors in Drug DiscoVery: A
Guide for Medicinal Chemists and Pharmacologists; John Wiley & Sons
Inc.: Hoboken, NJ, 2005; pp 111-213.
(8) Izquierdo-Martin, M.; Stein, R. L. Bioorg. Med. Chem. 1993 1, 19-26.
(9) (a) McClerren, A. L.; Endsley, S.; Bowman, J. L.; Andersen, N.; Guan,
Z.; Rudolph, J.; Raetz, C. R. H. Biochemistry 2005, 44, 16574-16538.
(b) Glaser, K. B.; Li, J.; Aakre, M. E.; Morgan, D. W.; Sheppard, G.;
Stewart, K. D.; Pollock, J.; Lee, P.; O’Connor, C. Z.; Anderson, S. N.;
Mussatto, D. J.; Wegner, C. W.; Moses, H. L. Mol. Cancer Ther. 2002,
1, 759-768.
JA0600680
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4245