Dibasic inhibitors of human mast cell tryptase. Part 1: Synthesis and optimization of a novel class of inhibitors

Article abstract of DOI:10.1016/S0960-894X(00)00484-4

The synthesis and optimization of a novel class of reversible and active-site directed dibasic inhibitors of human mast cell tryptase are described. The compounds were shown to be both remarkably potent and selective for tryptase with K(j) values for optimized inhibitors in the picomolar range. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00484-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2357±2360
Dibasic Inhibitors of Human Mast Cell Tryptase. Part 1:
Synthesis and Optimization of a Novel Class of Inhibitors
Kenneth D. Rice,b Anthony R. Ganglo€,a,* Elaine Y.-L. Kuo,a Je€rey M. Dener,c
Vivian R. Wang,a Robert Lum,d William S. Newcomb,e Chris Havel,a
Daun Putnam,a Lynne Cregar,a Martin Wonga and Robert L. Warnef
aDepartments of Medicinal Chemistry and Enzymology, Axys Pharmaceuticals, Inc., 180 Kimball Way,
South San Francisco, CA 94080, USA
bExelixis Pharmaceuticals, Inc., 280 E. Grand Ave., South San Francisco, CA 94080, USA
cChemRx Advanced Technologies, 385 Oyster Point Blvd, South San Francisco, CA 94080, USA
dCV Therapeutics, Inc., 3172 Porter Drive, Palo Alto, CA 94304, USA
eArgonaut Technologies, Inc., 887G Industrial Road, Suite G, San Carlos, CA 94070, USA
fChiron Corporation, 4560 Horton Street, Emeryville, CA 94608, USA
Received 19 May 2000; accepted 11 August 2000
AbstractÐThe synthesis and optimization of a novel class of reversible and active-site directed dibasic inhibitors of human mast cell
tryptase are described. The compounds were shown to be both remarkably potent and selective for tryptase with Ki values for
optimized inhibitors in the picomolar range. # 2000 Published by Elsevier Science Ltd.
Recent advances in cell biology have led to an improved
understanding of the mast cell as a multifunctional, key
e€ector cell in the immune system.1,2 Emerging data
suggest that tryptase, a major mast cell secretory pro-
tease, is a potent mediator of mast cell related allergic
and in¯ammatory pathologies, including most notably
asthma.3,4 Tryptase may function in this context
through several possible mechanisms including the reg-
ulation of potent neuropeptides such as bradykinin,5 the
stimulation of surrounding tissue proliferation and
secretion of secondary cellular mediators by activation
of the protease-activated receptor (PAR) family6 8 and
possibly through direct ampli®cation of the in¯amma-
tory response.9 Phase II clinical trials with our ®rst
generation tryptase inhibitor APC-366 further support
the involvement of tryptase in asthma pathology.10
Herein, we describe the evolution of SAR for a novel
second generation series of highly potent and selective
dibasic inhibitors of human tryptase.
Tidwell,11,12 however, these compounds lack the
potency and selectivity to be of signi®cant biological or
pharmaceutical utility. Recent reports have described
classes of dibasic inhibitors with improved potency and
selectivity.13,14 E€orts in our lab commenced with the
serendipitous discovery of p-xylylenediamine oligomer
8, a very potent, selective, competitive and reversible
tryptase inhibitor. Our goal was the design and optimi-
zation of tryptase inhibitors based on this template
with an aim toward the identi®cation of novel anti-
in¯ammatory drugs suitable for clinical development.
Chemistry
Scheme 1 illustrates the synthetic strategy employed in
the preparation of compounds 7±17. Reaction of a two-
fold excess of the appropriate acid chloride with p-xyly-
lenediamine eciently gave diesters 3±6. Direct forma-
tion of the external amide in the presence of an excess of
p-xylylenediamine or 4-aminobenzylamine at elevated
temperature a€orded the desired diamines 7±11 and 12±
14, respectively. In the latter case, completely selective
acylation of the benzylamine over the aniline nitrogen
was observed. Guanidines 15±17 are prepared by treat-
ment with either cyanamide or pyrazolecarboxamidine.
Several classes of modestly active dibasic inhibitors of
tryptase have been reported earlier by Sommerho€ and
*Corresponding author. Tel.: +1-650-829-1162; fax: +1-650-829-
1123; e-mail: tony_ganglo€@axyspharm.com
0960-894X/00/$ - see front matter # 2000 Published by Elsevier Science Ltd.
PII: S0960-894X(00)00484-4

2358
K. D. Rice et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2357±2360
Scheme 1. Synthesis of the dibasic inhibitors. Reagents: (a) CH3O2C(CH2)nCOCl, DCM, DIPEA; (b) p-xylylenediamine, 40 ꢀC; (c) 4-aminobenzyl-
amine, 40 ꢀC; (d) cyanamide, 4 M HCl in dioxane, 60 ꢀC; (e) N,N0-di(BOC)pyrazolecarboxamidine, DMF; TFA; (f) tert-butyl (1-piperazinecar-
boxylate)-4-carbonyl chloride, DIPEA, DCM, MeOH.
The general synthetic approach employed in the pre-
paration of carbamate linked inhibitors 20±26 was anal-
ogous. The sequence was carried out through reaction
of 1,4-phenylenedimethylene bis-chloroformate15 with
the amino acid of choice under Schotten±Baumann
conditions followed by coupling of the resulting diacid
with a twofold excess of either 4-guanidinobenzylamine
hydrochloride (18)16 or mono-N-(BOC)-p-xylylenedi-
amine17 followed by BOC group removal in the latter
case. Piperazine derived analogue 27 was synthesized by
the acylation of 18 with tert-butyl (1-piperazinecarbox-
ylate)-4-carbonyl chloride18 to give guanidine inter-
mediate 19. Subsequent BOC deprotection and reaction
in twofold excess with cis-1,5-cyclooctylene bis-chloro-
formate19 in aqueous THF at pH 8±9 optimally a€orded
the symmetrical bis-arylguanidine 27.
ylene chain length, although the arylguanidine analogue
was approximately 10-fold less active than the corre-
sponding benzylamine (Ki=3.0 and 40 nM, respec-
tively). In both cases >103-fold selectivity over related
proteases screened was observed. While internal amide
N-methylation is well tolerated in the benzylamine series
(compound 10), N-methylation of the external amide, as
in the case of 11, resulted in a dramatic potency loss.
A second series of dibasic inhibitors was designed to
resist metabolism at the internal amide through repla-
cement with a carbamate linker. Inhibitory potency and
selectivity for this series of analogues is shown in Table
2. Here the glycyl derivatives 20 and 22 were the most
potent. While N-methylation of the carbamate nitrogen
is well tolerated in this series, substitution at the adja-
cent methylene, as in the case of alanine analogues 24
and 25, resulted in a signi®cant potency loss with the
(R,R)-diastereomer demonstrating approximately 10-
fold greater activity than the (S,S)-diastereomer. In the
bis-arylguanidine series replacement of the core sca€old
aromatic ring with alicyclic functionality is well tolerated
and a ®vefold improvement in activity was observed in
the case of cyclooctylene derivative 26. As a further
Results and Discussion
Inhibitory potency versus tryptase and related proteases
for a series of dibasic polyamides is summarized in
Table 1.20 Succinyl derivatives 8 and 16 demonstrated
optimal potency versus tryptase as a function of meth-
Table 1. Amide series SAR
Ki (mM)a
Compound
R1
R2
n
X
Tryptase
Trypsin
Thrombin
Plasmin
7
8
9
10
11
15
16
17
H
H
H
Me
H
H
H
H
H
H
Me
H
1
2
3
2
2
1
2
3
-CH2NH2
-CH2NH2
-CH2NH2
-CH2NH2
-CH2NH2
0.150
0.003
0.009
0.005
2
0.680
0.040
0.100
317
237
274
162
>1000
73
>1000
>1000
>1000
>1000
>1000
>1000
>1000
492
>1000
769
>1000
214
>1000
>1000
>1000
>1000
-NHC(NH)NH2
-NHC(NH)NH2
-NHC(NH)NH2
H
H
H
H
146
70
aData reported is for a single determination. See ref 20 for assay protocols.

K. D. Rice et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2357±2360
2359
Table 2. Carbamate series SAR
Ki (mM)a
Trypsin
Compound
R
X
Y
Tryptase
Thrombin
Plasmin
20
21
22
23
24
25
H
H
H
Me
H
H
-CH2-
-(CH2)2-
-CH2-
-NHC(NH)NH2
-NHC(NH)NH2
-CH2NH2
0.010
0.090
0.0006
0.005
0.031
0.366
56
72
84
50
38
273
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>230
>1000
83
-CH2-
-(R)-CH(CH3)-
-(S)-CH(CH3)-
-CH2NH2
-CH2NH2
-CH2NH2
Ki (mM)a
Compound
X
Z1
Z2
Tryptase
0.002
Trypsin
78
Thrombin
>1000
Plasmin
>1000
26
-NHCH2-
-NHC(NH)NH2
-NHC(NH)NH2
27
28
29
-NHC(NH)NH2
-NHC(NH)NH2
-H
-NHC(NH)NH2
0.00007b
39
182
44
435
>1000
14
494
-H
-H
15
>1000
>1000
101
aData reported is for a single determination. See ref 20 for assay protocols.
bValue reported is the dissociation constant obtained by IC50 determination at variable tryptase and inhibitor concentrations.21
modi®cation of the core sca€old we focused on replace-
ment of the external amide. The piperazine ring was
chosen as a glycyl isostere that would serve to introduce
a urea linkage in place of the external amide. Aryl-
guanidine 27 demonstrated a marked improvement in
potency with a Ki value of 0.07 nM21 and selectivity over
trypsin and thrombin approaching 105. Asymmetric
derivative 28, which di€ers from the optimized lead only
in the absence of a second guanidine, illustrates the
importance of both basic moieties for potent tryptase
inhibition. Removal of the second guanidine a€ords
a nearly complete loss of activity for the intact core
sca€old analogue 29. The dramatic loss in potency that
is observed on sequential removal of the terminal
guanidines from the inhibitor suggests that core sca€old
interactions contribute little to the overall binding
energy of lead 27.
four quasi-equivalent monomers arranged in a ¯at
frame-like structure, with each of the four subunits
contributing a functional catalytic site. A spatially con-
stricted central pore de®nes the active site domain with
directly adjacent S1 pockets of the order of 20 A apart.
A binding mechanism for the dibasic inhibitors descri-
bed herein could be envisioned where the central pore is
spanned with the terminal nitrogen bases docked into
adjacent S1 pockets.
In summary, we have designed and optimized a series of
remarkably potent, selective, competitive and reversible
tryptase inhibitors, exempli®ed by compound 27.
Although limited by poor oral absorption, the optimized
dibasic inhibitors outlined herein represent valuable bio-
logical and clinical tools aimed at the further elucidation
of tryptase pathology in asthma and related allergic and
in¯ammatory diseases. Further SAR development lead-
ing to the selection of an optimized clinical candidate is
described in an additional report following this paper.
The X-ray crystal structure of tryptase has been recently
determined22,23 and revealed a structure comprised of

2360
K. D. Rice et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2357±2360
432.7 mmol) in DCM (100 mL). The mixture was warmed to
Acknowledgements
room temperature over 30 min followed by addition of aq HCl
(0.1 N, 200 mL) then drying the organic layer over anhydrous
magnesium sulfate. Filtration and concentration gave tert-
butyl 1-piperazinecarboxylate-4-carbonyl chloride (45.6 g,
85%) as a pale-yellow solid. 1H NMR (300 MHz, CDCl3) d
3.70 (m, 2H), 3.60 (m, 2H), 3.50 (m, 4H), 1.50 (s, 9H).
The authors wish to thank Mark Dreyer and Liling
Fang for analytical support as well as Heinz Gschwend
and Mike Venuti. We would also like to acknowledge
Bayer AG for their ®nancial support of tryptase
research at Axys Pharmaceuticals, Inc.
19. cis-1,5-Cyclooctanediol (20.2 g, 0.14 mol) and potassium
carbonate (41.4 g, 0.3 mol) were taken into acetonitrile
(250 mL) then cooled to 0 ꢀC followed by dropwise addition
of phosgene (1.9 M in toluene, 220 mL, 0.42 mol) over 1 h
followed by warming to rt and stirring over 12 h. Ether (1 L)
was added and the suspension ®ltered. Concentration followed
by recrystallization from hexane gave cis-1,5-cycloocetylene
bis-chloroformate as a colorless crystalline solid. 1H NMR
(300 MHz, CDCl3) d 5.00±4.85 (m, 2H), 2.20±1.60 (m, 12H).
20. Stock solutions of enzyme (0.06 mg/mL) were prepared
with 10 mM 2-(N-morpholino)ethane sulfonic acid (MES)
containing 2 mM calcium chloride, 20% glycerol and 0.05 mg/
mL heparin. Approximately 1 mg of inhibitor was dissolved in
DMSO (0.2 mL) and diluted 10-fold into bu€er containing
50 mM Tris hydrochloride (pH 8.2), 100 mM sodium chloride,
and 0.05% Tween-20. Seven additional threefold dilutions
were made from the initial dilution into the same bu€er sup-
plemented with 10% DMSO. Aliquots (0.05 mL) from each of
the eight dilutions were transferred to individual wells in a 96-
well U-bottom microtiter plate. Enzyme stock solution (25 mL)
was added to each well and incubated for 1 h at rt. The enzy-
matic reaction was initiated with addition of tosyl-Gly-Pro-
Lys-p-nitroanilide (25 mL, 0.5 mM ®nal concentration) and the
reaction was followed spectrophotometrically at 405 nm.
Initial velocity measurements calculated from the progress
curves by a kinetic analysis program (BatchKi; Peter Kuzmic,
University of Wisconsin, Madison, WI) were used to deter-
mine apparent inhibition constants. The standard error for
single determinations is a factor of 2.
References and Notes
1. Welle, M. J. Leukoc. Biol. 1997, 61, 233.
2. Yong, L. C. J. J. Exp. Toxic Pathol. 1997, 49, 409.
3. Zhang, M. Q.; Timmerman, H. Mediators of In¯ammation
1997, 6, 311.
4. Rossi, G. L.; Olivieri, D. Chest 1997, 112, 523.
5. Kozik, A.; Moore, R. B.; Potempa, J.; Imamura, T.;
Rapala-Kozik, M.; Travis, J. J. Biol. Chem. 1998, 273, 33224.
6. Dery, O.; Bunnett, N. W. Biochem. Soc. Trans. 1999, 27,
246.
7. Mirza, H.; Schmidt, V. A.; Derian, C. K.; Jesty, J.; Bahou,
W. F. Blood 1997, 90, 3914.
8. Steinho€, M.; Vergnolle, N.; Young, S. H.; Tognetto, M.;
Amadesi, S.; Ennes, H. S.; Trevisani, M.; Hollenberg, M. D.;
Wallace, J. L.; Caughey, G. H.; Mitchell, S. E.; Williams, L.
M.; Geppetti, P.; Mayer, E. A.; Bunnett, N. W. Nature Medi-
cine 2000, 6, 151.
9. He, S.; Gaca, M. D. A.; Walls, A. F. J. Pharmacol. Exp.
Ther. 1998, 286, 289.
10. Krishna, M. T.; Chauhan, A. J.; Little, L.; Sampson, K.;
Mant, T. G. K.; Hawksworth, R.; Djukanovic, R.; Lee, T. H.;
Holgate, S. T. Am. J. Respir. Crit. Care Med. 1998, 157, A456.
11. Sturzebecher, J.; Prasa, D.; Sommerho€, C. P. Biol. Chem.
Hoppe-Seyler 1992, 373, 1025.
12. Caughy, G. H.; Raymond, W. W.; Bacci, E.; Lombardy,
R. J.; Tidwell, R. R. J. Pharmacol. Exp. Ther. 1993, 264, 676.
13. Ono, S.; Kuwahara, S.; Takeuchi, M.; Sakashita, H.;
Naito, Y.; Kondo, T. Bioorg. Med. Chem. Lett. 1999, 9, 3285.
14. 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.
15. Richter, R.; Tucker, B. J. Org. Chem. 1983, 48, 2625.
16. Sa®er, S. R.; Kushner, S.; Brancone, L. M.; Subbarrow,
Y. J. Org. Chem. 1948, 13, 924.
17. Krapcho, P.; Kuell, C. S. Synth. Commun. 1990, 20, 2559.
18. To triphosgene (25 g, 84.2 mmol) in DCM (200 mL) at
0 ꢀC was slowly added a solution of tert-butyl 1-piper-
azinecarboxylate (40 g, 214.8 mmol) and pyridine (35 mL,
21. Ki value for the tight binding inhibitor 27 was determined
by the following procedure. Tryptase (0.25, 0.50, 1.0, and
2.0 nM) and inhibitor (40, 110, 340, and 680 pM) were varied
in preincubation mixtures for 1 h at rt followed by addition of
substrate as before. The IC50 of the compound was then
determined at each concentration of tryptase. The y-intercept
of a replot of IC50 (ordinate) versus tryptase concentration
(abscissa) produces a value for the apparent dissociation con-
stant from which the true dissociation constant of the tight
binding competitive inhibitor was calculated.
22. Sommerho€, C. P.; Bode, W.; Matschiner, G.; Bergner,
A.; Fritz, H. Biochim. Biophys. Acta 2000, 1477, 75.
23. Sommerho€, C. P.; Bode, W.; Pereira, P. J. B.; Stubbs, M.
T.; Sturzebecher, J.; Piechottka, G.; Matschiner, G.; Bergner,
A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 10984.