Synthesis and SAR of 4-carboxy-2-azetidinone mechanism-based tryptase inhibitors

Article abstract of DOI:10.1016/S0960-894X(02)00688-1

A series of N1-activated C4-carboxy azetidinones was prepared and tested as inhibitors of human tryptase. The key stereochemical and functional features required for potency, serine protease specificity and aqueous stability were determined. From these studies compound 2, BMS-262084, was identified as a potent and selective tryptase inhibitor which, when dosed intratracheally in ovalbumin-sensitized guinea pigs, reduced allergen-induced bronchoconstriction and inflammatory cell infiltration into the lung.

Full text of DOI:10.1016/S0960-894X(02)00688-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3229–3233
Synthesis and SAR of 4-Carboxy-2-azetidinone Mechanism-Based
Tryptase Inhibitors
James C. Sutton,* Scott A. Bolton, Karen S. Hartl, Ming-Hsing Huang, Glenn Jacobs,
Wei Meng, Martin L. Ogletree, Zulan Pi, William A. Schumacher, Steven M. Seiler,
William A. Slusarchyk, Uwe Treuner, Robert Zahler, Guohua Zhao
and Gregory S. Bisacchi
The Bristol-Myers Squibb Pharmaceutical Research Institute, PO Box 4000, Princeton, NJ 08543-4000, USA
Received 4 January 2002; accepted 26 April 2002
Abstract—A series of N1-activated C4-carboxy azetidinones was prepared and tested as inhibitors of human tryptase. The key
stereochemical and functional features required for potency, serine protease specificity and aqueous stability were determined.
From these studies compound 2, BMS-262084, was identified as a potent and selective tryptase inhibitor which, when dosed intra-
tracheally in ovalbumin-sensitized guinea pigs, reduced allergen-induced bronchoconstriction and inflammatory cell infiltration into
the lung.
# 2002 Elsevier Science Ltd. All rights reserved.
It is estimated that 14 million Americans suffer from
asthma. At the current rate of increased incidence it is
anticipated that 22 million Americans will suffer from
asthma by the end of the decade.1 To lessen the severity
of asthma, therapeutics modulating the underlying
inflammatory processes are being investigated. One tar-
get receiving attention is the trypsin-like serine protease
tryptase, which is produced almost exclusively by mast
cells. Tryptase accounts for a significant 20–25% of the
total lung and skin mast cell protein. Active tryptase is
stored in intracellular granules as a heparin stabilized
tetramer and is released into the extracellular environ-
ment upon mast cell stimulation. Once in the extra-
cellular spaces tryptase modulates inflammatory
processes through a number of pathways (see ref 2 for
reviews).
covalently inhibited enzyme. X-ray crystal structures of
azetidinone tryptase inhibitors related to 1 covalently
bound at the active site of trypsin support this mechan-
ism. At this point we embarked on a program to pre-
pare an optimized analogue of 1, which could be useful
as an inhaled tryptase inhibitor. In this Letter we report
the synthesis of a series of 4-carboxyazetidinone ana-
logues of 1 leading to the identification of the potent
tryptase inhibitor 2, BMS-262084 (tryptase IC50=4
nM), having an improved selectivity profile and showing
efficacy in a guinea pig model of bronchoconstriction and
lung inflammation.
Through directed screening in our laboratories, racemic
azetidinone (1)3 was discovered to be a potent tryptase
inhibitor having modest selectivity against a panel of
other trypsin like serine proteases. Kinetic studies were
consistent with a mechanism-based inhibition of serine
protease activity wherein the N1-activated azetidinone
General methods for the preparation of (3R,4S)-trans
and racemic (3,4)-cis 4-carboxyazetidinones are outlined
in Scheme 1. The (3R,4S)-trans isomers were prepared
*Corresponding author. Tel.: +1-609-252-5719; e-mail: james.sutton
@bms.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00688-1

3230
J. C. Sutton et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3229–3233
Scheme 1. (a) (i) LDA, THF, ꢀ78 to ꢀ20 ꢁC; (ii) 1-chloro-3-iodopropane, ꢀ20 ꢁC; (b) nBu4NI, nBu4NN3, DMF; (c) (i) H2, 10% Pd/C, HOAc,
DMF; (ii) Et3N, N,N0-bis-CBZ-1-guanylpyrazole; (d) BnBr, nBu4NI, NaHCO3, DMF; (e) (i) NaHMDS, THF, ꢀ78 to ꢀ20 ꢁC; (ii) isocyanate, acyl
chloride or sulfonyl chloride; or (iii) isocyanate, Et3N, DMAP, CH2Cl2; (f) H2, 10% Pd/C, dioxane, H2O, HCl.
starting from azetidinone 3, which was prepared from
l-aspartic acid as described by Baldwin.4 Alkylation of
3 with chloroiodopropane followed by azide displace-
ment gave 5 in moderate yield. Conversion to the bis-
Cbz protected guanidine compound 6 was achieved
through hydrogenation of the azide followed by treat-
ment with N,N0-bis-Cbz-1-guanylpyrazole and triethy-
lamine. The acid was benzyl protected to provide
intermediate 7, which was transformed to N1-activated
azetidinones by treatment with base followed by addi-
tion of an isocyanate, acyl chloride or sulfonyl chloride.
Deprotection via hydrogenation provided acids 8–10, 8a
being the active enantiomer of racemic 1.
imine of methylglyoxylate (12), Scheme 2. Subsequent
azide displacement of the intermediate alkylchloride
provided compound 13. Removal of the PMP group
with ceric ammonium nitrate followed with transfor-
mations described above for the trans-azetidinones gave
compound 15. Hydrogenation to remove the Cbz pro-
tecting groups followed by hydrolysis of the methyl
ester under mild conditions in a pH 7.4 aqueous buffer
gave acid 16.
The preparation of a series of N1-piperazine carbonyl
analogues is described in Scheme 3. Reaction of 1-Boc-
piperazine (17) with isocyanates, acyl chlorides or sul-
fonyl chlorides followed by Boc deprotection and treat-
ment with phosgene in CH2Cl2 over solid NaHCO3
provided 4-substituted piperazinecarbonyl chlorides
(18) in moderate to good yields over two steps.
Racemic 3,4-cis azetidinones were prepared via the
Staudinger reaction5 between 5-chlorovaleryl chloride
(11) and the para-methoxyphenyl (PMP) protected
Scheme 2. (a) Et3N, hexane, toluene, 3 A MS; (b) NaN3, nBu4NI, DMF; (c) CAN, CH3CN, H2O; (d) H2, 10% Pd/C, dioxane, 1 N HCl; (e) 1,3-bis-
CBZ-2-methyl-2-thiopseudourea, HgCl2, Et3N, DME, MeOH; (f) (i) NaHMDS, THF, ꢀ78 to ꢀ20 ꢁC; (ii) isocyanate; (g) H2, 10% Pd/C, Dioxane,
H2O, HCl; (h) pH 7.4 phosphate buffer.
Scheme 3. (a) (i) iPr2NEt, DMAP, CH2Cl2; (ii) isocyanate, chloroformate, carbamoylchloride, acyl chloride or sulfonyl chloride; (b) TFA, CH2Cl2;
(c) phosgene, CH2Cl2, NaHCO3; (d) 7, Et3N, DMAP, CH2Cl2; (e) H2, 10% Pd/C, Dioxane, H2O, HCl; (f) primary or secondary amine, BOP, THF;
(g) 4-BOC-1-piperazinecarbonyl chloride, Et3N, DMAP, CH2Cl2.

J. C. Sutton et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3229–3233
3231
Table 1. Tryptase inhibition, selectivity and aqueous stability for compounds 8–10 and 16
Activating group
Entry
R1
Tryptase
Selectivity ratio
Stability, t1/2 at 37 ꢁC (h)
IC50 (nM)
Plasmin
Thrombin
uPA
3
tPA
Trypsin
Buffer pH 7
Buffer pH 9
2.2
Carbamoyl
8a
ent-8a
16a (cis)
8c
Phenyl
Phenyl
Phenyl
Cyclohexyl
Benzyl
Phenyl
6
5380
992
18
71
9
52
>6
1548
>6
117
>6
1
0.3
>48
0.7
34
45
17
12
8
>611
465
142
190
65
2
16
3
5
3270
44
64
465
128
109
1610
4370
1
2
1
3
2
1
>48
>48
9.1
6.8
8d
9a
Acyl
9b
9c
10
2-Thiophene
4-Biphenyl
4-Biphenyl
1
1
1
8.7
11.1
10.6
<0.1
<0.1
<0.1
Sulfonyl
47
7
Table 2. Tryptase inhibition and selectivities for compounds 20–25
Entry
X
R1(R2)
Tryptase
Selectivity ratio
Stability, t1/2 at 37 ꢁC (h)
IC50 (nM)
Plasmin
Thrombin
uPA
tPA
Trypsin
pH 7
pH 9
20a
20b
20c
2(21a)
21b
22a
22b
22c
23
OH
OH
OH
OH
OH
OH
OH
OH
t-Butyl
Methyl
2-Phenethyl
t-Butyl (H)
Ethyl (Ethyl)
t-Butyl
Neopentyl
Phenyl
Phenyl
t-Butyl
t-Butyl
t-Butyl
4
9
0.7
4
5
7
11
5.2
13
0.4
0.7
3
295
210
302
430
227
232
314
332
147
227
172
205
>8250
683
218
70
>8250
>1200
14,246
>8250
>6600
>1500
>1000
>2000
>2500
8300
8
2
8
18
3
3
1.5
1
14
6
3
1
>48
>48
>48
>48
>48
>48
>48
>48
>48
42
6.2
5.6
7
3323
2625
>6600
>1500
>1000
286
926
135
520
154
8
142
43
82
75
8
6.4
2.5
6.5
3.9
1.7
0.3
0.3
0.8
OH
556
105
177
155
25a
25b
25c
NH2
NHCH3
Piperidine
6800
4433
31.8
>48
246
Coupling of carbamoyl chlorides (18) to azetidinone (7)
in the presence of Et3N and DMAPgave adducts 19.
Subsequent hydrogenation then gave the series of com-
pounds 20–23. Amine couplings to acid 6 followed by
condensation with the carbamoyl chloride of N-Boc-
piperazine and subsequent hydrogenation provided a
series of C4-amides, 25a–c.
nyl (10) activating groups were highly potent tryptase
inhibitors (IC50=1–9 nM) with mixed selectivity, but
had relatively poor aqueous stability, likely the result of
the higher degree of activation of the azetidinone.
The superior aqueous stability of the C4-carboxy azeti-
dinones bearing the N1-carbamoyl activating group
focused attention on this series. The data in Table 2
shows the tryptase activity and selectivity profiles of a
series of related compounds. The Boc compound 20a
showed a greatly improved selectivity with the exception
of selectivity against trypsin. Replacement of the Boc
group of 20a provided functionalized carbamates 20b–c,
ureas 21a–b, amides 22a–c, and sulfonamide 23 exhibit-
ing low nanomolar tryptase inhibitory activity
(IC50=0.7–13 nM) and generally good selectivity pro-
files. The C4-amides (25a–c) were very potent tryptase
inhibitors (IC50=0.4–3 nM) but in general had some-
what poorer selectivity against thrombin, compared to
the C4-carboxylic acids. Compound 2 (21a), BMS-
262084, a potent tryptase inhibitor (IC50=4 nM) with a
moderate to good selectivity profile except for trypsin,
and having relatively good aqueous stability was chosen
for in vivo evaluation.
Compounds were screened for activity against human
tryptase and a panel of related serine proteases;6
including trypsin, thrombin, plasmin, urokinase (uPA),
and tissue plasminogen activator (tPA). Stabilities in
aqueous buffer at pH 7 and pH 9 were also evaluated.
Aqueous stabilities of these compounds were found to
mirror stability in human plasma (data not shown).
The data in Table 1 show that the tryptase inhibitory
activity of the C4-carboxy azetidinone series resides
predominantly in the 3R,4S-trans stereoisomer, demon-
strated by comparison of compounds 8a (tryptase
IC50=6 nM), its enantiomer ent-8a (tryptase IC50=5.38
mM) and the racemate of the cis isomer, 16a, (tryptase
IC50=999 nM). However, the inhibitory activity against
most of the other serine proteases also resided in the
3R,4S-trans stereoisomer. Replacement of the N1-phe-
nylcarbamoyl group of 8a with cyclohexyl- or benzyl-
carbamoyl groups provided analogues with diminished
potency but somewhat enhanced aqueous stability, as
assessed at pH 9. Analogues with acyl (9a–9c) or sulfo-
The effect of BMS-262084 on the pulmonary mechanics
of ovalbumin-sensitized guinea pigs7 following allergen-
induced bronchoconstriction were evaluated. Ovalbumin
challenge increased lung resistance (Rlung) approximately

3232
J. C. Sutton et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3229–3233
Figure 1. Effect of BMS-262084, 2, on the pulmonary mechanics in
ovalbumin-sensitized guinea pigs after allergen-induced bronchocon-
striction.
4-fold (Rlung%Á=421ꢂ98, n=20), which reflects con-
striction of the large airways. Ovalbumin also decreased
dynamic lung compliance (Cdyn
)
by two-thirds
(Cdyn%Á=ꢀ69ꢂ3, n=20), which corresponded to
narrowing of the small airways and diminished pul-
monary elasticity. Intratracheal (i.t.) administration of
BMS-262084 (75 mg/kg) 10 min prior to ovalbumin chal-
lenge inhibited the increase in resistance by 79%
(Rlung%Á=89ꢂ26%, n=9) and the decrease in dynamic
lung compliance by 37% (Cdyn%Á=ꢀ43ꢂ6%, n=9),
Figure 1.
Figure 2. BMS-262084, 2, produces a dose-dependent inhibition of the
cellular inflammatory response in ovalbumin-sensitized guinea pigs.
In summary, we have studied the synthesis and SAR of
a series of N1-activated C4-carboxy azetidinone tryp-
tase inhibitors which show moderate selectivity against
related serine proteases with the exception of trypsin.
One of the more potent, selective and aqueous stable
compounds BMS-262084 (2) reduced the bronchocon-
striction and the cellular inflammatory response in
ovalbumin-sensitized guinea pigs.
BMS-262084 also showed a protective effect against
inflammatory cell infiltration in the delayed phase
of the immediate hypersensitivity response to aller-
gen.8 BMS-262084 produced
a
dose-dependent
inhibition of the cellular inflammatory response as
shown in Figure 2. At the 225 mg/kg i.t. dose all cell
counts were reduced by 70–80% to the level observed in
sham-treated animals. BAL protein concentration was
also decreased 22% from 238ꢂ14 mg/mL to 186ꢂ27
mg/mL (p<0.05), a level equal to that observed in sham
control animals (182ꢂ22 mg/mL). Both the cell counts
and the BAL protein concentration following the 225
mg/kg i.t. dose of BMS-262084 and ovalbumin chal-
lenge were not significantly different from those of
ovalbumin-sensitized guinea pigs studied 24 h after
sham operation without ovalbumin challenge. These
results are consistent with compound BMS-262084
exerting a strong protective effect against cellular acti-
vation that accompanies allergen-induced pulmonary
inflammation.
References and Notes
1. Pew Environmental Health Commission Attack Asthma:
Why America Needs a Public Health Defense System to Battle
Environmental Threats, May 2000. Report available at http://
pewenvirohealth.jhsph.edu/html/reports/ PEHCAsthmaReport.
pdf.
2. (a) Gangloff, A. R. Curr. Opin. Investig. Drugs 2000, 1, 79.
(b) Elrod, K. C.; Numerof, R. P. Emerging Ther. Targets 1999,
3, 203. (c) Zhang, M.-Q.; Timmerman, H. Mediators Inflama-
tion 1997, 6, 311.
3. Han, W. T.; Trehan, A. K.; Wright, J. J. K.; Federici,
M. E.; Seiler, S. M.; Meanwell, N. A. Bioorg. Med. Chem.
1995, 3, 1123.
4. Baldwin, J. E.; Adlington, R. M.; Gollins, D. W.; Schofield,
C. J. Tetrahedron 1990, 46, 4733.

J. C. Sutton et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3229–3233
3233
5. For a review see: Holden, K. G. In Chemistry and Biology
of ꢀ-Lactam Antibiotics; Morin, R. B., Gorman, M., Eds;
Academic: New York, 1982; Vol. 2, Chapter 2.
6. Combrink, K. D.; Gulgeze, H. B.; Meanwell, N. A.; Pearce,
B. C.; Pi, Z.; Bisacchi, G. S.; Roberts, D. G. M.; Stanley, P.;
Seiler, S. M. J. Med. Chem. 1998, 41, 4854.
7. Ovalbumin-sensitized guinea pigs were prepared in the fol-
lowing manner; ip injections of ovalbumin (10 mg) with alu-
minum hydroxide (100 mg/kg) given on days 0 and 14,
cyclophosphamide (30 mg/kg) on day 12 and utilized between
day 35 and 65. On the day of the experiment, the animals were
anesthetized with Na-pentobarbital and placed on a mechan-
ical ventilator. Intratracheal flow and pressures were recorded
and processed in real time with a pulmonary mechanics com-
puter (Model PMAG, Buxco, Troy, NY) providing measure-
nistered as an aerosol intratracheal spray using a Model 1A
needle (PenCentury, Philadelphia, PA) producing 25 mm med-
ian particle diameter. The trachea was reconnected to the
ventilator and changes in pulmonary mechanics were followed
over time.
8. In this experiment ovalbumin-sensitized guinea pigs were
anesthetized with isoflurane and the trachea was exposed by a
small skin incision. The Model 1A PenCentury spray injector
was advanced through a small cut between adjacent tracheal
rings up to the bronchial bifurcation. Saline or BMS-262084
(50 mL at 75 or 225 mg/kg) was administered as a spray and
the tracheal incision was closed immediately. Ovalbumin (50
mg/kg) was injected into the jugular veinꢃ6 min after intra-
tracheal dosing. The skin incision was then closed, anesthesia
discontinued and the animals allowed to recover. Twenty-four
h later, the guinea pigs were anesthetized with pentobarbital
and subjected to bronchoaveolar lavage (BAL). Differential
cell counts and protein concentration in the BAL were deter-
mined by a co-worker blinded to the treatment protocol.
ment of airways resistance (Rlung
) and dynamic lung
compliance (Cdyn). Ten min prior to ovalbumin challenge (50
mg/kg, iv) animals were disconnected from the ventilator and
50 mL of BMS-262084 (dose range 2.5–75 mg/kg) was admi-