Bivalent inhibition of beta-tryptase: distance scan of neighboring subunits by dibasic inhibitors.

Article abstract of DOI:10.1016/S0960-894X(02)00063-X

Based on bifunctional diketopiperazines as templates and m-aminomethyl-phenylalanine as arginine mimetic, we have synthesized a new class of structurally related dibasic tryptase inhibitors with systematically increasing spacer length. These compounds were used to scan the distance between the active sites of two neighboring subunits of the beta-tryptase tetramer. The K(i)-values obtained are a function of the distance between the terminal amino groups and indicate optimal binding of inhibitors with 29-31 bonds between the amino groups. These experimental data are in full agreement with predictions derived from a novel modeling program that allows the docking of bivalent ligands.

Full text of DOI:10.1016/S0960-894X(02)00063-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 985–988
Bivalent Inhibition of ꢀ-Tryptase: Distance Scan of Neighboring
Subunits by Dibasic Inhibitors
Norbert Schaschke,a,* Andreas Dominik,b Gabriele Matschinerc
and Christian P. Sommerhoff c
aMax-Planck-Institut fur Biochemie, D-82152 Martinsried, Germany
¨
bByk Gulden Lomberg Chemische Fabrik GmbH, D-78467 Konstanz, Germany
cAbteilung fur Klinische Chemie und Klinische Biochemie, Klinikumsstandort Innenstadt
¨
der LMU Munchen, D-80336 Munchen, Germany
¨
¨
Received 22 November 2001; revised 14 January 2002; accepted 22 January 2002
Abstract—Based on bifunctional diketopiperazines as templates and m-aminomethyl-phenylalanine as arginine mimetic, we have
synthesized a new class of structurally related dibasic tryptase inhibitors with systematically increasing spacer length. These com-
pounds were used to scan the distance between the active sites of two neighboring subunits of the b-tryptase tetramer. The Ki-values
obtained are a function of the distance between the terminal amino groups and indicate optimal binding of inhibitors with 29–31
bonds between the amino groups. These experimental data are in full agreement with predictions derived from a novel modeling
program that allows the docking of bivalent ligands. # 2002 Elsevier Science Ltd. All rights reserved.
Human b-tryptase is a mast cell-specific serine protease
that exhibits trypsin-like activity by hydrolyzing peptide
In this study, we have systematically investigated the
acceptor properties of b-tryptase by performing a dis-
tance scan using a set of structurally related dibasic
probing compounds. In addition, the ability of the
compounds to bind in a mono- or bivalent fashion was
analyzed using a novel modeling program that allows
the docking of bivalent ligands.
bonds C-terminally of arginine and lysine residues.1,2
A
growing number of biological and immunological data
suggests that tryptase plays a key role in the pathogen-
esis of diverse allergic and inflammatory disorders, most
prominently asthma,3,4 and thus is an interesting thera-
peutic target.5 In particular, the enzyme acts as a (neu-
ro)peptidase and is responsible for enhancing the
contractility of the airway smooth muscle.6 The X-ray
structure of human b-tryptase7,8 revealed that the
enzyme consists of four quasi-identical subunits (A, B,
C, and D) whose active sites are directed towards a
central pore. Therefore, the four negatively charged S1
binding pockets are displayed in a defined spatial
arrangement that should allow inhibition of the enzyme
by dibasic ligands of appropriate length interacting
simultaneously with two neighboring S1subsites as
depicted schematically in Figure 1. Compared to a
monovalent binding, the advantage of such a bivalent
interaction is the gain in affinity and selectivity achieved
by exploiting the entropy effect.9 For human b-tryptase,
a variety of dibasic inhibitors has been described that
take advantage of this particular feature of the
enzyme.10ꢀ15
Due to entropic reasons,9 we have focused our scanning
approach on the shortest distance between neighboring
S1pockets, that is those of subunits A and D (and their
Figure 1. Schematic representation of the b-tryptase tetramer inter-
acting with a bivalent inhibitor. The inter-S1subsite distances are
indicated and the side-chain carboxylate of Asp-189 at the bottom of
each S1pocket is shown.
*Corresponding author. Tel.: +49-89-8578-3910; fax: +49-89-8578-
2847; e-mail: schaschk@biochem.mpg.de
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00063-X

986
N. Schaschke et al. / Bioorg. Med. Chem. Lett. 12 (2002) 985–988
equivalents B and C), comprising 33 A according to the
crystal structure (Fig. 1).7,8 As central scaffold for the
display of two binding heads via spacers of increasing
length diketopiperazines were chosen. In particular,
those derived from aspartic and glutamic acid as well as
mixed ones were utilized to minimize the scaffold’s con-
tribution to the distance between the binding heads.
Diketopiperazines derived from amino acids can be
divided stereochemically into those composed exclu-
sively of l- or d-amino acids with cis-geometry in
respect to the plane of the ring and mixed ones with
trans-geometry. By inspecting the X-ray structure of the
b-tryptase tetramer, diketopiperazines with trans-geo-
metry were expected to allow a better positioning of the
binding heads into the S1subsites of neighboring sub-
units. Among the arginine mimetics, m- and p-amino-
methyl-phenylalanine were investigated as potential
binding heads. Based on the ꢁ15-times higher inhibitory
potency of Ac-dl-Phe(3-H2N-CH2)-OMe (Ki=13.7 mM)
compared to Ac-dl-Phe(4-H2N-CH2)-OMe (Ki=213
mM) for the b-tryptase tetramer, the more potent meta-
derivative was used in the present study, and to adjust
the optimal spacer length, o-amino acids were utilized.
To realize the distance scan, a building block strategy
was applied. The symmetric scaffolds c[d-Asp-l-Asp-]
(5) and c[d-Glu-l-Glu-] (6) were synthesized as outlined
in Scheme 1a. In addition, the asymmetric scaffold c[d-
Asp-l-Glu-] with orthogonal side-chain protection was
obtained from Z-d-Asp(OtBu)-OH and H-l-Glu(OMe)-
OMe following essentially the procedure used for the
symmetric piperazine-2,5-diones (Scheme 1b). Subse-
quently, the side-chain protection pattern of 9 was trans-
formed in three steps to yield 12 as building block suitable
for the final inhibitor assembly. The binding head was
obtained starting from H-dl-Phe(3-CN)-OH in five steps
as outlined in Scheme 2. To vary systematically the dis-
tance between the headgroups, two types of dibasic inhi-
bitors were synthesized: Symmetric ones were obtained as
summarized in Scheme 3a, and an asymmetric inhibitor
was synthesized utilizing the orthogonally protected scaf-
fold 12 to complete the distance scan (Scheme 3b).
The inhibition kinetic data16 for the interaction of the
dibasic distance probes with the b-tryptase tetramer are
summarized in Table 1. The results clearly show a rela-
tion between the distance of the binding heads and the
Scheme 1. (a) Synthesis of the symmetric scaffolds. Reaction conditions: (i) H-l-Xaa(OtBu)-OMe/DIEA/EDC/HOBt, CHCl3; (ii) 10% Pd–C/H2,
MeOH; (iii) T, MeOH; (iv) 95% aq TFA, 0 ꢂC!rt. (b) Synthesis of the asymmetric scaffold. Reaction conditions: (i) H-l-Glu(OMe)-OMe/DIEA/
EDC/HOBt, CHCl3; (ii) 10% Pd–C/H2, MeOH; (iii) T, MeOH; (iv) NaOH, THF/H2O (2:1); (v) (a) Cs2CO3, MeOH; (b) C6H5–CH2–Br, DMF; (vi)
95% aq TFA, 0 ꢂC!rt.
Scheme 2. Synthesis of the binding head and its spacer functionalization. Reaction conditions: (i) MeOH/SOCl2, ꢀ5 ꢂC; (ii) N-ethoxy-
carbonylphthalimide/Na2CO3, dioxane/H2O (1:1); (iii) 10% Pd-C/H2, AcOH; (iv) (Boc)2O/NaHCO3, dioxane/H2O (1:1); (v) H2N–NH2ꢃAcOH,
MeOH, 50 ꢂC; (vi) Z-NH–(CH2)n–COOH/DIEA/EDC/HOBt, CHCl3; (vii) 10% Pd–C/H2, MeOH.

N. Schaschke et al. / Bioorg. Med. Chem. Lett. 12 (2002) 985–988
987
Scheme 3. (a) Synthesis of the symmetric dibasic inhibitors. Reaction conditions: (i) n=1: 18 or 19 or 20; n=2: 17/DIEA/EDC/HOBt, DMF; (ii)
95% aq TFA, 0 ꢂC!rt. (b) Synthesis of the asymmetric dibasic inhibitor. Reaction conditions: (i) 18/DIEA/EDC/HOBt, DMF; (ii) 10% Pd–C/H2,
AcOEt; (iii) 19/DIEA/EDC/HOBt, DMF ; (iv) 95% aq TFA, 0 ꢂC!rt.
affinity of the inhibitors. The dibasic inhibitor 21, that is
that with the shortest distance between the terminal
amino groups, is only about as potent as the headgroup
(2.4 mM vs 13.7 mM, respectively). An increase of the
distance between the amino groups, however, is accom-
panied by a large increase in affinity, reaching a max-
imal cooperativity with compound 23 that is 1370-fold
more potent than the binding head.
b-tryptase tetramer/dibasic ligand of 1:2. The predic-
tions whether the inhibitors bind in a mono- or bivalent
manner are in full agreement with the SAR data derived
from inhibition kinetics. Both diastereomers of inhibitor
21 as well as the R,R-diastereomer of inhibitor 28 were
Table 1. Inhibition of the b-tryptase tetramer by the set of dibasic
distance probes
A novel modeling program17 that allows the docking of
bivalent inhibitors was applied on this set of dibasic
distance probes. For each inhibitor two of its possible
diastereomers (i.e., the S,S- and R,R-compounds; the
absolute stereochemistry refers to the head groups,
respectively) were docked. The result obtained for the
S,S-diastereomer of inhibitor 23 is shown in Figure 2.
The inhibitor can adopt a conformation that allows the
simultaneous interaction of its both positively charged
headgroups with the Asp-189 residues within the S1
pockets of the neighboring subunits A and D (or their
equivalents B and C). According to the docking experi-
ments, the main contribution to the affinity of the diba-
sic inhibitors appears to originate from interactions of
the headgroups within the S1subsites, whereas contacts
between the central spacer region and the enzyme were
not detectable. Furthermore, such modeling confirmed
that there is sufficient space for two dibasic ligands
within one tetramer and thus, allowing a stoichiometry
Compd
m
n
X
Y
Number of bonds
between the
amino groups
Ki (mM)
21
28
22
23
24
2
12
11 Gly
1
114-Abu
2
—
—
25
27
2.4
Gly
—
80.1
08.01
Gly
b-Ala
4-Abu
29
1
b-Ala
0
310.01
33
0.35

988
N. Schaschke et al. / Bioorg. Med. Chem. Lett. 12 (2002) 985–988
7. Barbosa Pereira, P. J.; Bergner, A.; Macedo-Ribeiro, S.;
Huber, R.; Matschiner, G.; Fritz, H.; Sommerhoff, C. P.;
Bode, W. Nature 1998, 392, 306.
8. Sommerhoff, C. P.; Bode, W.; Barbosa Pereira, P. J.;
Stubbs, M. T.; Sturzebecher, J.; Piechottka, G. P.; Matschi-
ner, G.; Bergner, A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
10984.
9. Mammen, M.; Choi, S. K. Whitesides, G. M. Angew.
Chem. Int. Ed. Engl. 1998, 37, 2755.
10. Burgess, L. E.; Newhouse, B. J.; Ibrahim, P.; Rizzi, J.;
Kashem, M. A.; Hartman, A.; Brandhuber, B. J.; Wright,
C. D.; Thomson, D. S.; Vigers, G. P. A.; Koch, K. Proc. Natl.
Acad. Sci. U.S.A. 1999, 96, 8348.
11. Ono, S.; Kuwahara, S.; Takeuchi, M.; Sakashita, H.;
Naito, Y.; Kondo, T. Bioorg. Med. Chem. Lett. 1999, 9, 3285.
12. Rice, K. D.; Gangloff, A. R.; Kuo, E. Y.-L.; Dener, J. M.;
Wang, V. R.; Lum, R.; Newcomb, W. S.; Havel, C.; Putnam,
D.; Cregar, L.; Wong, M.; Warne, R. L. Bioorg. Med. Chem.
Lett. 2000, 10, 2357.
13. Rice, K. D.; Wang, V. R.; Gangloff, A. R.; Kuo, E. Y.-L.;
Dener, J. M.; Newcomb, W. S.; Young, W. B.; Putnam, D.;
Cregar, L.; Wong, M.; Simpson, P. J. Bioorg. Med. Chem.
Lett. 2000, 10, 2361.
14. Schaschke, N.; Matschiner, G.; Zettl, F.; Marquardt, U.;
Bergner, A.; Bode, W.; Sommerhoff, C. P.; Moroder, L. Chem.
Biol. 2001, 8, 313.
15. Dener, J. M.; Rice, K. D.; Newcomb, W. S.; Wang, V. R.;
Young, W. B.; Gangloff, A. R.; Kuo, E. Y.-L.; Cregar, L.; Put-
nam, D.; Wong, M. Bioorg. Med. Chem. Lett. 2001, 11, 1629.
16. Ki-values for the interaction of the inhibitors with the
tryptase tetramer were determined essentially as described
previously.14
Figure 2. Bivalent docking of the S,S-diastereomer of inhibitor 23.
Subunits A and D of the b-tryptase tetramer are shown in ribbon
representation, the inhibitor (colored in yellow) as well as the amino
acid residue Asp-189 (colored in red) of subunits A and D as stick
models: (A) side view, (B) top view.
classified as monovalent binders, whereas all other dia-
stereomers are able to bind in a bivalent manner.
Thus, based on both inhibition kinetics and docking
studies the distance probe 21 is too short to interact
with the tryptase tetramer in a bivalent manner. The
gain in affinity observed with 22 and 23 in comparison
to the binding head together with the results from the
docking unambiguously classify them as bivalent bind-
ers. The decrease in inhibitory potency of 24, a distance
probe still able to interact with b-tryptase in a bivalent
manner (bivalent docking), probably reflects the loss of
conformational entropy upon binding that partially
consumes the entropy effect. In summary, the data
obtained clearly show that the b-tryptase tetramer
recognizes and binds dibasic inhibitors of appropriate
length.
17. Briefly, docking of the ligand is performed as follows: The
bivalent inhibitor is cut into three parts, that is headgroup A,
headgroup D, and spacer fragment. The headgroups are
docked into the S1pockets of subunit A and D, respectively,
using the program FlexX.18 If the headgroups show a binding
mode inside the S1pockets, all fragments of the inhibitor are
reconnected. Tripos force field19 is used to adjust all bond
lengths and bond angles to reasonable values. Subsequently,
two additional geometry optimization steps are applied to the
resulting raw ligand–enzyme complex using the force field
MMFF9420 as implemented in the software package Sybyl.21
First, an optimization step is performed with fixed tryptase
structure but without any constraints on the ligand. For the
final geometry optimization also, parts of the tryptase struc-
ture are allowed to be flexible. Ligands that pass all steps of
the docking algorithm are considered to be bivalent binding
inhibitors.
Acknowledgements
The study was supported by the SFB 469 of the Ludwig-
Maximilians-University of Munich and by Byk-Gulden,
Konstanz, Germany with a postdoctoral grant for N.S.
References and Notes
1. Johnson, D. A. In Handbook of Proteolytic Enzymes; Bar-
rett, A. J.; Rawlings, N. D.; Woessner, J. F., Eds.; Academic:
San Diego, 1998; p 70.
2. Sommerhoff, C. P.; Bode, W.; Matschiner, G.; Bergner, A.;
Fritz, H. Biochim. Biophys. Acta 2000, 1477, 75.
3. Caughey, G. H., Ed. Mast Cell Proteases in Immunology
and Biology; Marcel Dekker: New York, 1995.
4. Caughey, G. H. Am. J. Respir. Cell Mol. Biol. 1997, 16,
621.
18. Kramer, B.; Metz, G.; Raray, M.; Lengauer, T. Med.
Chem. Res. 1999, 9, 463.
19. Clark, M.; Cramer D.; Van Opdenbosch, III N. J. Com-
put. Chem. 1989, 10, 982.
20. Halgren, T. J. Am. Chem. Soc. 1990, 112, 4710.
21. Sybyl 6.7; Triops Ass. Inc.: 2001.
5. Rice, K.; Spencer, J. Exp. Opin. Ther. Pat. 1999, 9, 1537.
6. Sommerhoff, C. P. Am. J. Respir. Crit. Care Med. 2001,
164, 54S.