Synthesis and evaluation of diphenyl phosphonate esters as inhibitors of the trypsin-like granzymes A and K and mast cell tryptase

Article abstract of DOI:10.1021/jm970543s

Thirty-six new amino acid and peptidyl diphenyl phosphonate esters were synthesized and evaluated to identify potent and selective inhibitors for four trypsin-like proteases: lymphocyte granzymes A and K, human mast cell tryptase, and pancreatic trypsin. Among five Cbz derivatives of Lys and Arg homologues, Z-(4-AmPhe)(P)(OPh)₂ is the most potent inhibitor for granzyme A, and Z-Lys(P)(OPh)₂ is the best inhibitor for granzyme K, mast tryptase, and trypsin. The amidino P1 residue D,L-(4-AmPhGly)(P)(OPh)₂ was utilized in a series of compounds with several different N-protecting groups and systematic substitutions at P2 in Cbz-AA derivatives and at P3 in Cbz-AA-Ala derivatives. Generally, these phosphonates inhibit granzyme A and trypsin more potently than granzyme K and tryptase. The P2 Thr and Ala dipeptide phosphonates, Cbz-AA-(4-AmPhGly)(P)(OPh)₂, are the most potent inhibitors for granzyme A, and Cbz-Thr-(4-AmPhGly)(P)(OPh)₂ (k(obs)[I] = 2220 M^-1 s^{- 1}) was quite specific with much lower inhibition rates for granzyme K and trypsin (k(obs)[I] = 3 and 97 M^-1 s^-1, respectively) and no inhibition with tryptase. The most effective inhibitor of granzyme A was Ph-SO₂-Gly- Pro-(4-AmPhGly)(P)(OPh)₂ with a second-order rate constant of 3650 M^-1 s^{- 1}. The most potent inhibitor for granzyme K was 3,3-diphenylpropanoyl-Pro- (4-AmPhGly)(P)(OPh)₂ with a k(obs)/[I] = 1830 M^-1 s^-1; all other phosphonates inhibited granzyme K weakly (k(obs)/[I] < 60 M^-1 s^-1). Human mast cell tryptase was inhibited slowly by these phosphonates with Cbz- Lys(P)(OPh)₂ as the best inhibitor (k(obs)/[I] = 89 M^-1 s^-1). The overall results suggest that scaffolds of Phe-Thr-(4-AmPhe) and Phe-Pro-Lys will be useful to create selective phosphonate inhibitors for granzymes A and K, respectively, and that P4 substituents offer opportunities to further enhance selectivity and reactivity.

Full text of DOI:10.1021/jm970543s

J . Med. Chem. 1998, 41, 2289-2301
2289
Syn th esis a n d Eva lu a tion of Dip h en yl P h osp h on a te Ester s a s In h ibitor s of th e
Tr yp sin -lik e Gr a n zym es A a n d K a n d Ma st Cell Tr yp ta se
Delwin S. J ackson,† Stephanie A. Fraser,| Li-Ming Ni,† Chih-Min Kam,† Ulrike Winkler,| David A. J ohnson,§
Christopher J . Froelich,‡ Dorothy Hudig,| and J ames C. Powers*,†
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332-0400
Received August 13, 1997
Thirty-six new amino acid and peptidyl diphenyl phosphonate esters were synthesized and
evaluated to identify potent and selective inhibitors for four trypsin-like proteases: lymphocyte
granzymes A and K, human mast cell tryptase, and pancreatic trypsin. Among five Cbz
derivatives of Lys and Arg homologues, Z-(4-AmPhe)P(OPh)2 is the most potent inhibitor for
granzyme A, and Z-LysP(OPh)2 is the best inhibitor for granzyme K, mast tryptase, and trypsin.
The amidino P1 residue D,L-(4-AmPhGly)P(OPh)2 was utilized in a series of compounds with
several different N-protecting groups and systematic substitutions at P2 in Cbz-AA derivatives
and at P3 in Cbz-AA-Ala derivatives. Generally, these phosphonates inhibit granzyme A and
trypsin more potently than granzyme K and tryptase. The P2 Thr and Ala dipeptide
phosphonates, Cbz-AA-(4-AmPhGly)P(OPh)2, are the most potent inhibitors for granzyme A,
and Cbz-Thr-(4-AmPhGly)P(OPh)2 (kobs/[I] ) 2220 M-1 s-1) was quite specific with much lower
inhibition rates for granzyme K and trypsin (kobs/[I] ) 3 and 97 M-1 s-1, respectively) and no
inhibition with tryptase. The most effective inhibitor of granzyme A was Ph-SO2-Gly-Pro-(4-
AmPhGly)P(OPh)2 with a second-order rate constant of 3650 M-1 s-1. The most potent inhibitor
for granzyme K was 3,3-diphenylpropanoyl-Pro-(4-AmPhGly)P(OPh)2 with a kobs/[I] ) 1830 M-1
s-1; all other phosphonates inhibited granzyme K weakly (kobs/[I] < 60 M-1 s-1). Human mast
cell tryptase was inhibited slowly by these phosphonates with Cbz-LysP(OPh)2 as the best
inhibitor (kobs/[I] ) 89 M-1 s-1). The overall results suggest that scaffolds of Phe-Thr-(4-AmPhe)
and Phe-Pro-Lys will be useful to create selective phosphonate inhibitors for granzymes A and
K, respectively, and that P4 substituents offer opportunities to further enhance selectivity and
reactivity.
In tr od u ction
with apoptosis after CTL-mediated killing.11-13
Granzyme A can activate a cellular thrombin receptor14
and will also degrade collagen,15 indicating that it may
have several effects that extend beyond the cytolytic
reaction.
Cytolytic T lymphocytes (CTL)1 and natural killer
(NK) cells play important roles in the cellular immune
system. These immune effector cells recognize and lyse
target cells, such as virus-infected cells, tumor cells, or
the cells of allografted tissues.2,3 The mechanism of
CTL and NK cell-mediated lysis is an issue that has
received a great deal of attention in the past 2 decades.
In the process of target cell killing, proteins present in
cytoplasmic granules of activated T lymphocytes are
released in the proximity of the target cell. The
cytoplasmic granules4,5 contain the pore-forming protein
perforin6,7 and a family of serine proteases termed
granzymes. During the killing process, perforin is
released and creates pores in the target cell membrane,
allowing entry of the granzymes into the cytoplasm of
the target cell. The exact role of the granzymes in this
process still needs definition. These proteases have
been proposed to be involved in killing and/or postkilling
processes.8-10 Granzymes A, B, and K are associated
with the characteristic DNA fragmentation which occurs
Studies with peptide substrates have revealed the
presence of at least seven distinct lymphocyte granule
serine proteases based on their abilities to differentially
hydrolyze synthetic substrates. The hydrolytic activities
are attributable to granzymes (three chymases, two
tryptases, an Asp-ase, and a Met-ase). Murine lympho-
cyte granules contain at least eight serine proteases
(granzymes A to G and M), some of which are identified
only by their gene sequence and have unknown sub-
strate specificity. Related serine proteases have been
discovered in human lymphocyte granules (which in-
clude tryptase, Asp-ase, and chymase activities).9,16-19
Human granzymes A and K cleave after basic residues,
granzyme B cleaves after acidic residues, granzyme M
cleaves after methionine, and an undesignated chymase
cleaves after large aromatic residues.
While suitable substrates have been discovered, very
few papers describe inhibitors for granzymes. Reactive
and specific irreversible inhibitors are needed for the
granzymes. The three types of granzyme inhibitors
reported in the literature are chloromethyl ketones,
4-guanidinobenzoates, and 3-alkoxy-4-chloroisocou-
marins.11,17,20,21 In general, chloromethyl ketones are
* To whom correspondence should be addressed. Tel: (404) 894-
4038. Fax: (404) 894-7452. E-mail: james.powers@chemistry.gatech.edu.
† Georgia Institute of Technology.
| School of Medicine and The College of Agriculture, Howard
Building MC320, University of Nevada, Reno, NV 89557-0046.
‡ Evanston Hospital, Northwestern University Medical School,
Evanston, IL 60201.
§ Department of Biochemistry, College of Medicine, East Tennessee
State University, J ohnson City, TN 37614-0581.
S0022-2623(97)00543-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/03/1998

2290 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
J ackson et al.
Sch em e 1a
a
(a) 145 °C, N2, phthalic anhydride; (b) DMSO, (COCl)2, DIEA, CH2Cl2; (c) P(OPh)3, benzyl carbamate, HOAc, 85-90 °C; (d) NH2-NH2,
i-PrOH; (e) HCl, CHCl3.
potent irreversible inhibitors of serine proteases that
alkylate the active site histidine,22 but a recent report
shows that they only weakly inhibit cytotoxic lympho-
cyte granule serine proteases (kobs/[I] ) 3-77 M-1 s-1
for granzymes A, K, and B).11 The 4-guanidinobenzoate
inhibitors, which are similar to “inverse substrates”, are
potent inhibitors of granzyme A (kobs/[I] ) 540 000 M-1
s-1) but are not specific and inhibit other trypsin-like
enzymes.23,24 3-Alkoxy-4-chloroisocoumarins are potent
acylating agents that react with the active-site serine.
F igu r e 1. Structures of the amino acid diphenyl phosphonate
ester derivatives used to study the primary binding site (S1)
However, the stability of these acyl-enzymes varied,
some of them have short half-lives20,25 and deacylate to
reactivate the granzymes. The limitations of each of
these classes of inhibitors illustrate the need for a
suitable class of granzyme inhibitors.
of trypsin-like enzymes. Cbz-OrnP(OPh)2 (1) is the phosphonate
analogue of ornithine, while Cbz-LysP(OPh)2 (2) and Homo-
LysP(OPh)2 (3) are related to Lys and homolysine, respectively.
Cbz-(4-AmPhGly)P(OPh)2 (4) and Cbz-(AmPhe)P(OPh)2 (5) are
phosphonate derivatives of the arginine analogues 4-amidino-
phenylglycine and 4-amidinophenylalanine, respectively.
Peptidyl (R-aminoalkyl)phosphonate esters are a class
of inhibitors that have been used to study serine
proteases and shown to be very stable and specific
irreversible inhibitors of elastase and various chymo-
trypsin-like enzymes.26-30 Serine proteases such as
trypsin and thrombin are potently inhibited by peptidyl
(R-aminoalkyl)phosphonate diphenyl esters with orni-
thine-, Lys-, and amidino-containing residues at the P1
site.31-34 This paper discusses the synthesis and evalu-
ation of Cbz derivatives of Orn, Lys, HomoLys, AmPh-
Gly, and AmPhe diphenyl phosphonate esters (Figure
1). Additionally, a variety of peptidyl (4-AmPh-
Gly)P(OPh)2 derivatives have been developed to study
the structural preferences at P2 and P3 for both
granzymes A and K and mast cell tryptase. Inhibitors
for granzymes A and K have potential in the treatment
of organ transplant rejection35 and rheumatoid arthri-
tis,36 and inhibitors for the mast cell tryptase could be
used in the treatment of asthma.37
the presence of trifluoroacetic acid.40 This carbamate
and benzyl carbamate were used in the Oleksyszyn
reaction with 4-cyanobenzaldehyde and triphenyl phos-
phite to prepare the diphenyl phosphonate esters 11 and
12.34 Subsequent treatment of the nitrile with HCl in
CHCl3/EtOH produced the imino ester, which was
treated with NH3/MeOH to give compound 13 and the
parent compound Cbz-(4-AmPhGly)P(OPh)2 (4). De-
blocking of the Cbz group from compound 4 was ac-
complished by hydrogenolysis in the presence of HCl to
yield compound 14. The intermediate (4-AmPhGly)P-
(OPh)2 (14) was then coupled with the carboxylic acid
derivatives, N-blocked amino acid derivatives, or blocked
peptide acids using the DCC/HOBt coupling method to
give compounds 16-36, 38, and 40-44. Removal of the
side-chain blocking groups from compounds 38 and 40
with TFA yielded compounds 37 and 39, respectively.
Hydrogenolysis of the N-blocking group from compound
34, followed by treatment with succinic anhydride and
1 equiv of TEA, generated compound 35. In a similar
fashion, 15 was prepared by reaction of succinic anhy-
dride and TEA with compound 14. The naphthyl-
sulfonyl derivative 22 was made by reacting 1-naph-
thylsulfonyl chloride with (4-AmPhGly)P(OPh)2 in the
presence of 1 equiv of TEA at room temperature.
Ch em istr y
The general synthetic pathway for the preparation of
the Cbz derivatives of OrnP, LysP, and HomoLysP is
described in Scheme 1. The amino alcohols 6a -c were
treated with phthalic anhydride followed by Swern
oxidation38 to give the aldehydes 7a -c. The diphenyl
phosphonate esters 8a -c were obtained by treating the
aldehydes (7a -c) with triphenyl phosphite and benzyl
carbamate in the presence of acetic acid at 85-90 °C
using the Oleksyszyn reaction.39 Removal of the pro-
tecting group with hydrazine in i-PrOH solution pro-
duced the side-chain unblocked derivatives 1 (Cbz-
Resu lts a n d Discu ssion
Eva lu a tion of Com p ou n d s. The inhibitory potency
of a series of (R-aminoalkyl)phosphonates with basic
side chains has been measured with four trypsin-like
enzymes: lymphocyte granzymes A and K, mast cell
tryptase, and pancreatic trypsin. The results with five
Cbz derivatives differing in P1 are reported in Table 1,
and the results with various blocking groups or peptide
derivatives containing a P1 4-AmPhGlyP(OPh)2 residue
are shown in Table 2. The phosphonates used for
OrnP(OPh)2),
HomoLysP(OPh)2).
2
(Cbz-LysP(OPh)2), and
3
(Cbz-
Derivatives of (4-AmPhGly)P(OPh)2 (14) were synthe-
sized using the reactions outlined in Scheme 2. The
1-naphthylmethyl carbamate (10) was prepared by
reacting 1-naphthylmethanol with sodium cyanate in

Peptide Phosphonate Protease Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2291
Sch em e 2a
a
(a) TFA; (b) 4-cyanobenzaldehyde, P(OPh)3; (c) HCl(g), CHCl3/EtOH; (d) NH3/MeOH, NH4Cl; (e) H2, Pd/C, HCl; (f) TEA, succinic
anhydride; (g) 1-naphthylsulfonyl chloride, TEA, DMF; (h) RCO2H ) all carboxylic acid derivatives, N-blocked amino acid derivatives,
and blocked peptide acids, DCC or EDC, HOBt, TEA; (i) TFA.
Ta ble 1. Inhibition of Trypsin-like Serine Proteases with Amino Acid Diphenyl Phosphonate Derivativesa
kobs/[I] (M-1 s-1
)
compound
granzyme A
granzyme K
tryptase
trypsin
1, Cbz-(Orn)P(OPh)2
0.6
1210 ( 59
23
1180
2750 ( 92
NIc
23
1.1 ( 0.1
3.8 ( 0.7
2.1 ( 0.2
NIc
89 ( 1
0.7
0.6
17 ( 1.4
NIc
2, Cbz-(Lys)P(OPh)2
6240
21 ( 6
2000b
24b
3, Cbz-(HomoLys)P(OPh)2
4, Cbz-(4-AmPhGly)P(OPh)2
5, Cbz-(4-AmPhe)P(OPh)2
a
Inhibition constants were measured in 0.1 M Hepes, 0.01 M CaCl2, pH 7.5, for trypsin and granzymes A and K or in 0.1 M Hepes,
10% glycerol, 10 mM heparin, pH 7.5, for mast cell tryptase and at 25 °C. The reaction mixtures contained 2% DMSO for trypsin, 7.6%
for granzyme A, and 7.1% for granzymes A and K and tryptase. The substrate was Z-Arg-SBzl for granzymes and trypsin and Z-Arg-SBzl
b
or Z-Lys-SBzl for tryptase in the presence of DTNB. The inhibitor concentration ranged from 1.0 to 450 µM. Previously reported data.34
c NI, no inhibition after 25 min of incubation with enzyme.
inhibition studies are racemic mixtures at the phos-
phonate R-carbon, but only one diastereomer appears
to be effective29 so that true potency will be ∼2-fold
greater than the numbers reported.
interaction of Cbz-(4-AmPhGly)P(OPh)2 with trypsin in
the crystal structure is stabilized by hydrogen bonds
between the amidino nitrogen atoms and the carbox-
ylate oxygens of Asp189 and the oxygens of Gly219,
Ser190 (Oγ), and a “buried” water molecule. The
inhibited derivatives are extremely stable, and trypsin
inhibited by Cbz-(4-AmPhGly)P(OPh)2 recovered no
activity after incubation in a neutral pH buffer for 1
month.
St r u ct u r a l Differ en ces a m on g t h e Gr a n u le
Tr yp ta ses. Granzymes A and K have several struc-
tural features similar to all serine proteases described
to date.42-46 Both granzymes A and K have the char-
acteristic amino-terminal sequence Ile-Ile-Gly-Gly to
allow the amino terminus to form an internal salt bridge
Mech a n ism of In h ibition . Inhibition of serine
proteases by peptidyl phosphonates involves nucleo-
philic substitution on the phosphorus atom by the
oxygen atom of the catalytic Ser195 of the protease
(Figure 2). The substitution proceeds through a penta-
coordinate phosphorus transition state to give a stable
irreversible tetrahedral monophenoxyphosphonyl de-
rivative.34 The recent X-ray structure of Cbz-(4-AmPh-
Gly)P(OPh)2 bound to bovine trypsin is consistent with
this mechanism and also shows that the second phenoxy
group is hydrolyzed during an aging process.41 The

2292 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
J ackson et al.
Ta ble 2. Inhibition of Trypsin-like Serine Proteases by 4-Amidinophenylglycine Phosphonate Diphenyl Ester Derivativesa
kobsd/[I] (M-1 s-1
)
human
granzyme A
rat
rat
mast cell
tryptase
bovine
trypsin
compd
R
granzyme A
granzyme K
4
15
16
17
18
19
20
21
22
13
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Cbz
Suc
1180
857
400 ( 10
1740 ( 41
740 ( 18
5
3.8 ( 0.7
0.6
0.2
2000b
45 ( 3
110 ( 1
360 ( 28
300 ( 22
NI
0.3 ( 0.07
trans-cinnamoyl
3-(2-furyl)acryloyl
3-(2-thienyl)acryloyl
trans-3-(3-pyridyl)acryloyl
3-phenoxybenzoyl
2-phenoxybenzoyl
1-NpSO2
1-naphthylmethoxycarbonyl
Cbz-Ala
Cbz-Val
340 ( 18
1740 ( 237
430 ( 31
1
470 ( 14
340 ( 1
30 ( 1
90 ( 9
1270 ( 85
3
32
4
3
1
1
NI
12
0.3
NIc
1
NI
NI
0.4
10
NI
0.8
21.2 ( 0.2
0.3
1.2
NI
NI
5% inhib
0.9
7% inhib
NI
0.4
2
0.7
1
0.7
3
2
8.5 ( 0.4
4
0.4
2
2
340 ( 53
540 ( 3
330 ( 26
160 ( 2
1670 ( 11
6.7 ( 0.1
6.3 ( 0.7
790 ( 17
2220 ( 40
30 ( 1
50 ( 2
670 ( 53
170 ( 34
30 ( 1
NI
1850 ( 85
41 ( 3
50 ( 1
910 ( 81
97
30 ( 1
17 ( 1
7760 ( 310
4080 ( 32
2660 ( 40
1520 ( 27
3620 ( 588
640 ( 36
240 ( 10
2550 ( 145
2000 ( 24
3140 ( 262
3100 ( 177
3010 ( 318
130b
0.7 ( 0.04
NI
34 ( 4
3
1
13
60 ( 5
NI
Cbz-Leu
Cbz-Pro
Cbz-Thr
Cbz-Lys
15 ( 1
8
Cbz-Phe
14 ( 1
2-phenoxybenzoyl-Pro
3-phenoxybenzoyl-Pro
3,3-diphenylpropanoyl-Pro
3-phenylpropanoyl-Pro
Cbz-Ala-Ala
Suc-Ala-Ala
Cbz-Pro-Ala
Cbz-Asp-Ala
Cbz-Asp(t-Bu)-Ala
Cbz-Lys-Ala
Cbz-Lys(Boc)-Ala
Cbz-Phe-Ala
Boc-D-Phe-Pro
Ph-CH2-SO2-Gly-Pro
Cbz-Ala-Ala-Ala
610 ( 88
330 ( 58
250 ( 8
1510 ( 217
820 ( 29
260 ( 15
380 ( 18
240 ( 5
280 ( 5
90 ( 9
1830 ( 140
50 ( 2
480 ( 27
5
0.3 ( 0.01
5
6
NI
NI
NI
110 ( 5
1770 ( 13
170
3650 ( 180
730 ( 15
4.2 ( 0.3
56 ( 2
87
7
37060 ( 986
1780 ( 33
590 ( 17
a
Inhibition constants were measured in 0.1 M Hepes, 0.01 M CaCl2, pH 7.5, for trypsin and granzymes A and K and in 0.1 M Hepes,
10% glycerol, 10 mM heparin, pH 7.5, for mast cell tryptase and at 25 °C. The reaction mixtures contained 2% DMSO for trypsin, 7.6%
for granzyme A, and 7.1% for granzyme K and tryptase. The substrate was Z-Arg-SBzl for granzymes and trypsin and Z-Arg-SBzl or
b
Z-Lys-SBzl for tryptase in the presence of DTNB. The inhibitor concentration ranged from 1.0 to 450 µM. Previously reported data.34
c NI, no inhibition after 25 min of incubation with enzyme.
with Asp194 and the active-site charge relay system,
Asp102, His57, and Ser195, at homologous positions
flanked by well-conserved peptide segments. Alignment
of amino acid sequences shows that granzymes A43,45
and K44 and mast cell tryptase42,46 are similar to bovine
trypsin in that each of the enzymes has an Asp189 in
the S1 specificity pocket. The molecular model of
granzyme A,43 based on the crystal structure of bovine
trypsin,47 shows a two-amino-acid insertion after Ser217
that extends the entrance to the S1 pocket. Granzyme
K and mast cell tryptase do not have this insertion.42,44
The S2 and S3 subsites derived for granzymes A and K
and mast cell tryptase also have some substitutions
(Table 3). On the basis of these differences, we expected
to observe some differences in the S1, S2, and S3
specificity of the various enzymes. The apparent con-
tributions these substitutions make on the specificity
at three subsites of these enzymes are discussed below.
P r im a r y Sp ecificity P ock et. To look at the P1
specificity, we examined a variety of P1 phosphonates.
Cbz-blocked derivatives of Lys or AmPhe (an Arg
analogue) phosphonates that differed in length by
addition or deletion of a CH2 group were tested (Table
1). The potency of the inhibitors correlates directly with
structural differences in the S1 subsites. For instance
in the case of trypsin, extension of Cbz-(AmPhGly)P-
(OPh)2 with one CH2 group to Cbz-(4-AmPhe)P(OPh)2
decreased the second-order rate constant about 80-fold,
from 2000 to 24 M-1 s-1. The trypsin structure with
Cbz-(4-AmPhGly)P(OPh)2 shows that the distance be-
tween the amidino group and Asp189 is about 3.1 Å,41
which is similar to the 2.8-3.2-Å distance when Arg is
in the S1 pocket (trypsin complexed with Arg15 ana-
logue of the bovine pancreatic trypsin inhibitor).48 The
additional methylene group, as in Cbz-(4-AmPhe)P-
(OPh)2, may create some unfavorable contacts. This
trend is more pronounced when examining Cbz-(Lys)P-
(OPh)2 (kobs ) 6240 M-1 s-1). One-methylene-longer
Cbz-(HomoLys)P(OPh)2 (kobs/[I] ) 21 M-1 s-1) or one-
methylene-shorter Cbz-(Orn)P(OPh)2 (no inhibition) were
both unfavorable. Considering the above data, clearly
the proper extension and size at P1 for trypsin is Lys.

Peptide Phosphonate Protease Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2293
F igu r e 2. Reaction of a 4-amidinophenylglycine phosphonate diphenyl ester derivative (acyl-AmPhGlyP(OPh)2) with a trypsin-
like serine protease. The E‚I complex (left) reacts to form the monophenoxyserine195 ester (center) which then undergoes slow
aging to the tetrahedral monoester (right) in which one phosphonate oxygen is interacting with the oxyanion hole of the serine
protease.
Ta ble 3. Influence of Amino Acid Sequence Differences in the S1 Specificity Pockets and at the Extended Substrate Binding Regions
of Trypsin-like Serine Proteases on Substrate and Inhibitor Specificity
substrate recognition subsitesa
protease
S1
S1
S2
99
Leu
Arg
Glyh
Thr
S3
residue no.b
189
Asp
Asp
Asp
Asp
215
Trp
Phe
Glyg
Trp
216
Gly
Gly
Gly
Asp
217
Ser
Leu
Tyr
Glu
218
c
Glu
218a
Asne
219
Gly
Lys
Lys
Gly
220
Cys
Cys
Cys
Cys
192
Gln
Asn
Lys
Lys
bovine trypsin
granzyme Ad
granzyme Kf
mast cell trypi
a
The subsites refer to those regions of the enzyme which interact with individual residues of a peptide substrate. Thus, the residues
in S1 interact with the P1 residue in a substrate. This P1 residue would be a Lys or Arg in the case of trypsin-like enzymes. The primary
specificity determinate in S1 is Asp189. The S2 residue often is in a position to interact with the P2 residue of a substrate or inhibitor.
b
Residue 192 is near the S1 pocket and can often interact with the P3 residue of a substrate. Chymotrysinogen numbering. There is no
d
residue 218a in chymotrypsinogen. c There is no residue 218 in trypsin. Human granzyme A.43,45 e This insertion extends the entrance
g
to the S1 pocket making a larger opening compared to trypsin. f Rat granzyme K (RNK-Tryp-2).44 This region contains significant
h
i
changes from trypsin. This region also contains a deletion. Human lung mast cell tryptase R.42,46
Also consistent with the predictions based on the
modeled structures, granzyme A prefers larger exten-
sions than trypsin at P1. For instance, Cbz-(4-AmPhe)P-
(OPh)2 (5) with the extra CH2 group (kobs/[I] ) 2750 M-1
s-1) is 2-fold more potent than Cbz-4-(AmPhGly)P(OPh)2
(kobs/[I] ) 1180 M-1 s-1). Moreover, compound 5 is a
much better inhibitor than any of the Cbz-Lys ana-
logues. However, just as with trypsin, the addition of
the CH2 group when going from Cbz-(Lys)P(OPh)2 to
Cbz-(HomoLys)P(OPh)2 results in a significant decrease
(52-fold) in the kobs/[I] value and almost a complete loss
of inhibition is observed with one less CH2 group in Cbz-
(Orn)P(OPh)2 for granzyme A. Collectively, these ob-
servations suggest that the two-amino-acid insertion in
the S1 subsite of granzyme A results in an open and
accommodating S1 pocket, allowing the enzyme to
tolerate bulky extended hydrophobic moieties well.
Distinct structural restrictions also apply to the S1
pocket of granzyme K and human mast cell tryptase.
Like trypsin, granzyme K and tryptase do not have the
two-amino-acid insertion as in the granzyme A in the
S1 pocket. Both granzyme K and tryptase were inhib-
ited best by the P1 Lys derivative 2 (Table 1) and
inhibited weakly by 4-AmPhGly (4) and HomoLys (3)
derivatives. Both enzymes were not inhibited by the
Orn compound 1. Tryptase was also inhibited moder-
ately by the 4-AmPhe derivative 5.
ied, and (2) a series of Cbz-L-AA-D,L-(4-AmPhGly)P-
(OPh)2 analogues were examined. All the trans-acryloyl
derivatives, compounds 16-19, are similar except for
the size of the aromatic ring.
N-Block in g Gr ou p . Notably, P2 substitutions result
in 100-fold differences in the rate constants for these
compounds with Gr A of both species, with (2-furyl)-
acryloyl-(4-AmPhGly)P(OPh)2 being the most effective
(1740 M-1 s-1). Perhaps the potency of this compound
is related to the Arg99 side chain hydrogen bonding with
the furyl group oxygen, or there may be a small
hydrophobic pocket which allows granzyme A to accom-
modate the furyl ring better. The 2-fold decrease in the
second-order rate constants of compounds 18 and 16
versus compounds 17 and 4, respectively, indicates that
the furyl group oxygen and the carbamate oxygen do
contribute to the inhibitors’ binding to the enzyme. The
small 3-arylalkenoyl-containing groups are tolerated
better than the bulky N-blocking groups by granzyme
A. For rat granzyme A and human granzyme A, the
majority of the inhibition constants are the same. The
exception is 1-NpSO2-(4-AmPhGly)P(OPh)2, which is 11-
fold less effective with rat granzyme A.
Using the same group of N-blocked derivatives of (4-
AmPhGly)P(OPh)2 (see Table 2, compounds 4, 13, 15-
22), granzyme K and trypsin were not inhibited as well
as granzyme A. The best granzyme K inhibitor of the
set is (2-furyl)acryloyl-(4-AmPhGly)P(OPh)2 (kobs/[I] ) 32
M-1 s-1) which is not surprising since this enzyme has
the residue Asn99 in the S2 subsite. For trypsin,
P 2 P ock et Sp ecificity. We used two approaches to
assess P2 specificity: (1) the N-blocking group of (4-
AmPhGly)P(OPh)2 compounds was systematically var-

2294 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
J ackson et al.
generally, the 3-arylalkenoyl derivatives did not work
as well as the parent compound 4, and compounds with
bulky N-blocking groups were not effective inhibitors.
All the N-blocked derivatives of (4-AmPhGly)P(OPh)2
inhibited mast cell tryptase poorly.
other flexible P3 compound, Ph-CH2-SO2-Gly-Pro-(4-
AmPhGly)P(OPh)2 (43) (kobs/[I] ) 37 060 M-1 s-1).
The more potent inhibitors for granzyme K contain
bulky P3 hydrophobics and P2 Pro. Examining the
second-order rate constants for the two phenoxybenzoyl-
Pro derivatives, compounds 30 and 31, illustrates the
importance of having the phenyl correctly oriented.
Interestingly, the inhibitor with flexibility (33) inhibits
granzyme K within the same order of magnitude as the
rigid compound 30. Overall, the most potent inactivator
of granzyme K is the bulkiest P3 Pro analogue, 3,3-
P 2 Am in o Acid Der iva tives. To further character-
ize the P2 structural preferences of the tryptase
granzymes, one succinyl and a range of Cbz-L-AA-(4-
AmPhGly)P(OPh)2 were prepared. The P2 pocket for
granzyme A has a distinct preference for small amino
acids (Table 2). The inactivation rate constant of Cbz-
Ala-(4-AmPhGly)P(OPh)2 (kobs/[I] ) 1670 M-1 s-1) is
greater than that of the more restricted analogue Cbz-
Pro-(4-AmPhGly)P(OPh)2 (kobs/[I] ) 790 M-1 s-1), but for
this dipeptide series, optimal inhibition is achieved in
Cbz-Thr-(4-AmPhGly)P(OPh)2 (kobs/[I] ) 2220 M-1 s-1),
when the Arg99 in the S2 subsite has the opportunity
for H-bonding with the side chain of this compound.
Furthermore, Cbz-Thr-(4-AmPhGly)P(OPh)2 had very
little effect against granzyme K and trypsin (kobs/[I] )
3 and 97 M-1 s-1, respectively) and no inhibition against
tryptase, indicating Thr at P2 is selective for granzyme
A.
diphenylpropanoyl-Pro-(4-AmPhGly)P(OPh)2, with kobs
/
[I] ) 1830 M-1 s-1 (32).
P 4 Sp ecificity. Additionally, the P4 and P5 ex-
tended subsite binding of the granzymes was evaluated
by comparing Cbz-Ala-Ala-(4-AmPhGly)P(OPh)2 (34) to
Suc-Ala-Ala-(4-AmPhGly)P(OPh)2 (35) and Cbz-Ala-Ala-
Ala-(4-AmPhGly)P(OPh)2 (44) (Table 2). Examining the
second-order rate constant of inhibitor 34 versus inhibi-
tor 35 for granzyme A, a 3-fold decrease in kobs/[I] is
observed with P4 Suc, while the tetrapeptide derivative,
compound 44 (kobs/[I] ) 730 M-1 s-1), is more compa-
rable with the tripeptide derivative, compound 34 (kobs
/
[I] ) 820 M-1 s-1). Among these tripeptide derivatives
with Cbz P4, Cbz-Phe-Ala-(4-AmPhGly)P(OPh)2 was the
most effective against granzyme A with a kobs/[I] of 1770
For trypsin, the smaller hydrophobic P2 amino acids
provided superior inhibition as they did with granzyme
A. For trypsin the smaller less restricted Cbz-Ala-(4-
AmPhGly)P(OPh)2 was the most effective with a kobs/[I]
) 1850 M-1 s-1. The derivatives with large P2 hydro-
phobic and charged side-chain derivatives behaved as
expected, since these side chains would create spatial
overlap and unfavorable contacts with the side chain
of Leu99. For granzyme K and tryptase, most of these
phosphonates were ineffective. Granzyme K and tryptase
differed in P2 preferences from the others: Cbz-Pro-(4-
AmPhGly)P(OPh)2 was the most effective inhibitor for
granzyme K with a kobs/[I] ) 34 M-1 s-1; Z-Val-(4-
AmPhGly)P(OPh)2 was the best dipeptide inhibitor for
M-1 s-1
.
For trypsin, changing the N-blocking group from Cbz
to Suc in compounds 34 and 35 also resulted in a 6-fold
decrease in inhibitory potency. Extending the peptide
chain to the tetrapeptide, compound 44, produced about
a 2-fold decrease in kobs/[I] for trypsin, while marginally
improving the kinetics for granzyme A. These changes
are understandable since this area of the enzymes has
the amino acid side chain oriented in a direction going
away from the binding pocket. In the tripeptide Cbz
series, all of the analogues were found to be potent
inhibitors of trypsin, except Cbz-Pro-Ala-(4-AmPhGly)P-
(OPh)2 (the one analogue which could possibly be forced
to orient unfavorably). For granzyme K and mast cell
tryptase, the tripeptides were ineffective.
tryptase with a kobs/[I] ) 21 M-1 s-1
.
P 3 Sp ecificity. To examine their P3 structural
preferences, granzymes A and K, mast cell tryptase, and
trypsin were investigated with analogues of RCO-AA-
Ala-D,L-(4-AmPhGly)P(OPh)2. Several N-blocked-Pro-(4-
AmPhGly)P(OPh)2 analogues that were developed as
thrombin inhibitors based on current literature49-51
were found to be excellent inhibitors for the enzymes
in this study. The derivative Ph-CH2-SO2-Gly-Pro-(4-
AmPhGly)P(OPh)2 (43) was the most potent of this series
against granzyme A (kobs/[I] ) 3650 M-1 s-1). The Gly
makes the P3 extension flexible. Another derivative
with a flexible P3 extension, 3-phenylpropanoyl-Pro-(4-
AmPhGly)P(OPh)2 (kobs/[I] ) 1510 M-1 s-1) (33), was not
as potent against granzyme A. In the case of the two
rigid inhibitors, 2-phenoxybenzoyl-Pro-(4-AmPhGly)P-
(OPh)2 (kobs/[I] ) 610 M-1 s-1) was comparable with Cbz-
Pro-(4-AmPhGly)P(OPh)2 (kobs/[I] ) 790 M-1 s-1), while
3-phenoxybenzoyl-Pro-(4-AmPhGly)P(OPh)2 was 2-fold
less effective (kobs/[I] ) 330 M-1 s-1) against granzyme
A.
Com p a r a tive Sp ecificity. Each of these enzymes
was potently inhibited with some degree of specificity.
Specificity is best illustrated with Cbz-Thr-(4-AmPh-
Gly)P(OPh)2 inhibition of granzyme A. This compound
potently inactivates granzyme A (kobs ) 2220 M-1 s-1
)
while weakly inhibiting granzyme K (kobs/[I] ) 3 M-1
s-1) and trypsin (kobs/[I] ) 97 M-1 s-1) and not inhibiting
tryptase. The tripeptide derivative Ph-CH2-SO2-Gly-
Pro-(4-AmPhGly)P(OPh)2 is the best inhibitor for
granzyme A (kobs ) 3650 M-1 s-1) and also has little
activity against granzyme K and tryptase (kobs ) 87 and
2 M-1 s-1, respectively). However, this compound, Ph-
CH2-SO2-Gly-Pro-(4-AmPhGly)P(OPh)2, is also very re-
active with trypsin (kobs ) 37 060 M-1 s-1), making it
not absolutely specific. However, it may be possible to
obtain selectivity for the granzymes vs trypsin in
biological situations since the granzymes are active in
granules and susceptible to inhibition. In contrast,
trypsin is stored in the pancreas as the inactive zy-
mogen, trypsinogen, which would not be inhibited by
phosphonates. When active trypsin is formed in the gut,
it is likely that the concentration of the inhibitors would
be much reduced due to exhange with blood and gut
The trend for flexible vs rigid P3 was also followed,
but to a lesser extent, with trypsin. Moderate inhibition
was also observed with the flexible P3 compounds 32
and 33 (kobs/[I] ) 1520 and 2660 M-1 s-1, respectively).
Optimal inhibition for trypsin was achieved with an-

Peptide Phosphonate Protease Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2295
Syn th esis: Gen er a l Cou p lin g P r oced u r e. The inter-
mediate (4-AmPhGly)P(OPh)2‚HCl was prepared as previously
described.34 All carboxylic acid derivatives, N-blocked amino
acid derivatives, and blocked peptide acids were coupled to
(4-AmPhGly)P(OPh)2‚HCl using the DCC/HOBt coupling
method. Generally, a DMF solution containing 1 mmol of both
the amino acid derivative and (4-AmPhGly)P(OPh)2‚HCl was
cooled to 0 °C. To this DMF solution were added 1.1 mmol of
TEA and 1.2 mmol of HOBt. After the mixture stirred for 15
min at 0 °C, 1.2 mmol of a coupling reagent, DCC or EDC,
was added, and the solution was allowed to stir for an
additional hour at 0 °C and then overnight at 4 °C. Any solid
formed was removed by filtration, and the remaining solution
was evaporated to dryness. The residual oil was purified by
silica gel column chromatography with CHCl3/MeOH/AcOH (8:
2:0.2) as the eluent to purify the final products.
All final products were purified by recrystallization from
MeOH/Et2O or column chromatography and characterized by
NMR, mass spectroscopy (MS), and elemental analysis or high-
resolution MS, unless otherwise indicated. The 1H NMR
spectra indicated that all the derivatives were mixtures of
diastereomers because of the pattern observed for the peak at
δ 5.9-6.2. A doublet of doublets is usually observed which
integrates for 1H, signifying that one doublet comes from one
diastereomer while the other originates from a second dia-
stereomer. No attempts were made to resolve the various D,L-
(4-AmPhGly)P(OPh)2 derivatives. The 4-amidino functional
group is mostly characterized by two multiplets with chemical
shifts in the range of 9.0-9.58 and 9.6-10.5 ppm. Integration
shows the presence of two protons in each multiplet.
fluids, reducing its ability to inhibit trypsin. The best
inhibitor of granzyme K is 3,3-diphenylpropanoyl-Pro-
(4-AmPhGly)P(OPh)2, and relative to the best inhibitors
for the other trypsin-like enzymes examined, this com-
pound is specific for granzyme K.
Con clu sion s
Amino acid and peptidyl diphenyl phosphonate esters
have been prepared for the study of the relationship of
structure to the activities of the granzymes A and K and
tryptase. Like trypsin, granzyme K and tryptase show
a preference for P1 Lys inhibitor residues consistent
with the structure of their narrow hydrophobic S1
pocket. On the other hand, granzyme A is inhibited
better by the larger P1 hydrophobic residues such as
the 4-amidinophenylalanine derivative. This may be
due to a two-amino-acid insertion in the sequence which
may enlarge the S1 pocket. Granzyme A is inhibited
well by small amino acids at the P2 site; the Thr
derivative which has H-bonding character best inhibits
granzyme A. For most cases, granzyme A is better
inhibited by derivatives containing hydrophobic groups
at P3. However, the best overall inhibitor of granzyme
A was Ph-CH2-SO2-Gly-Pro-(4-AmPhGly)P(OPh)2, with
glycine at the P3 position and Ph-CH2-SO2 as an
N-blocking group. The conformationally restricted and
hydrophobic proline at the P2 site best inhibits granzyme
K. The P3 pocket of granzyme K accommodates a large
hydrophobic moiety well, similar to the “aryl binding
pocket” of thrombin. For tryptase, most of these phos-
phonates were ineffective except Cbz-(Lys)P(OPh)2. In
summary, this research provides potent inhibitors for
both granzymes A and K and suggests scaffolds of Phe-
Thr-(4-AmPhGly) and Phe-Pro-Lys for developing better
phosphonate inhibitors for granzymes A and K, respec-
tively. Peptide phosphonates have the advantage of
being stable in water and gave no apparent side effects
when tested in rabbits for 15 min for their anticoagulant
effects (C.-M. Kam and J . C. Powers, unpublished
results).
5-P h th a lim id op en ta n ol. A mixture of phthalic anhydride
(6.46 g, 43.6 mmol) and 5-aminopentanol (4.5 g, 43.6 mmol)
was heated to 145 °C in an open flask for 30 min. A stream
of N2 was applied to expel water vapor. After cooling to room
temperature, the reaction residue was dried on vacuum pump.
1
The product was obtained as a colorless oil: yield 100%; H
NMR (CDCl3) δ 1.4-1.5 (m, 2H), 1.6-1.8 (m, 4H), 3.6 (t, 2H,
J ) 6.5 Hz), 3.7 (t, J ) 7.2 Hz), 7.7 (m, 2H), 7.8-7.9 (m, 2H).
5-P h th a lim id op en ta n a l (7b). The preparation of this
compound begins by adding oxalyl chloride (4.6 mL, 53.2 mmol)
to 60 mL of freshly distilled CH2Cl2 containing DMSO (6.9 mL,
96.0 mmol) at -45 °C and stirring for 5 min. The dropwise
addition (over 20 min) of phthalic alcohol (10.2 g, 43.6 mmol)
in 40 mL of CH2Cl2 was followed by an additional 15 min of
stirring, and then DIEA (22.8 mL, 130.9 mmol) was intro-
duced. At this time the reaction temperature was warmed to
-30 °C, and stirring continued for 30 min. The solvent was
removed, and the residue was taken up in 100 mL of EtOAc,
filtered to remove any DIEA salt, washed with 6% NaHCO3
and water (3 × 40 mL), and dried over Na2SO4. After filtering
the solvent was removed yielding the aldehyde, a yellow oil:
yield 88%; 1H NMR (CDCl3) δ 1.6-1.9 (m, 4H), 2.5 (t, J ) 6.3
Hz, 2H), 3.7 (t, J ) 7.8 Hz, 2H), 7.6-7.9 (m, 4H), 9.7 (s, 1H);
MS (EI+) m/z 231.1.
Exp er im en ta l Section
Ma t er ia ls. Benzyl carbamate, triphenyl phosphite, 4-
cyanobenzaldehyde, 1-hydroxybenzotriazole (HOBt), 1,3-di-
cyclohexylcarbodiimide (DCC), and common reagents and
solvents were purchased from Aldrich Chemical Co., Milwau-
kee, WI. The protected amino acid derivatives Z-Ala-OH,
Z-Phe-OH, and Z-Lys(Boc)-OH were purchased from Calbio-
chem-Novabiochem Corp., San Diego, CA. Preparative thin-
layer chromatography was performed with plates precoated
with 2 mm of silica gel 60 F which were obtained from EM
Separations, Gibbstown, NJ . Hepes was purchased from
Research Organics Inc., Cleveland, OH; heparin was pur-
chased from Sigma Chemical Co., St. Louis, MO; and 4,4′-
dithiodipyridine was purchased from Aldrich Chemical Co.,
Milwaukee, WI. The substrate Cbz-Arg-SBzl was synthesized
as previously described.52 The substrate Cbz-Lys-SBzl was
purchased from Sigma Chemical Co., St. Louis, MO. Melting
points were determined on a Mel-Temp II apparatus and are
uncorrected. All NMR were recorded on a Varian GEMINI
300 instrument. Elemental analyses were performed by
Atlantic Microlab, Inc., Norcross, GA. Analytical HPLC was
performed on a Hewlett-Packard 1090 with a Zorbax Rx-C18
column (4.6 mm × 25 cm). The flow rate was 0.8 mL/min with
CH3CN/H2O containing 0.1% TFA as eluent.
Dip h en yl
N-(Ben zyloxyca r b on yl)a m in o(4-p h t h a l-
im id obu tyl)m eth a n ep h osp h on a te [Cbz-Lys(P h t)P (OP h )2,
8b]. To a mixture of 5-phthalimidopentanal (8.9 g, 38.5 mmol),
triphenyl phosphite (10.1 mL, 38.5 mmol), and benzyl car-
bamate (5.8 g, 38.5 mmol) was added 100 mL of acetic acid.
The solution was stirred at 80-90 °C for 1 h.39 Acetic acid
was removed in vacuo. Crystallizing from methanol yielded
the phosphonate product as a white solid: mp 100-102 °C;
yield 29.1%; 1H NMR (CDCl3) δ 1.5-2.2 (m, 6H), 3.7 (t, 2H, J
) 6.3 Hz), 4.4-4.6 (m, 1H), 5.0-5.3 (m, 2H), 7.0-7.4 (m, 15H),
7.7 (m, 2H), 7.9 (m, 2H). Anal. (C33H31O7N2P) C, H, N. The
analytical data are consistent with reported data.32
Dip h en yl
N-(Ben zyloxyca r b on yl)a m in o(4-a m in o-
bu tyl)m eth a n ep h osp h on a te Hyd r och lor id e [Cbz-LysP -
(OP h )2, 2]. Cbz-Lys(Pht)P(OPh)2 (0.6 g, 1.0 mmol) was dis-
solved in hot i-PrOH (60 °C) followed by addition of hydrazine
(0.1 mL, 3.0 mmol) via a syringe. The reaction mixture was
allowed to stir for 3 h at 60 °C. The solid formed was filtered,
and the solvent was removed. The remaining residue was
taken up in 100 mL of CHCl3, washed with saturated NaCl (4

2296 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
J ackson et al.
× 60 mL), and then dried over Na2SO4. After the CHCl3
solution was cooled to 0 °C, the solution was saturated with
gaseous HCl. Finally, the solvent was removed, and the
residue was triturated in ether to give the product as a white
solid: mp 123-126 °C; yield 75%; 1H NMR (DMSO) δ 1.4-
2.0 (m, 6H), 2.7-2.8 (m, 2H), 4.2-4.4 (m, 1H), 5.0-5.2 (m, 2H),
7.1-7.4 (m, 15H), 7.8-8.1 (m, 3H); MS (FAB+) m/z calcd for
C25H29O5N2P (M + 1) 469.3. Anal. (C25H29O5N2P‚HCl‚0.7H2O)
C, H, N.
the crude material from H2O resulted in the product: mp 126-
1
129 °C; yield 60%; H NMR (DMSO) δ 5.4-5.5 (s, 2H), 6.4-
6.8 (br s, 2H), 7.4-7.6 (m, 4H), 7.9-8.1 (m, 3H); HRMS (EI+)
m/z calcd for C12H11O2N (M + 1) 201.078 97, found 201.079 62.
Dip h en yl N-(1-Na p h th ylm eth yloxyca r bon yl)a m in o(4-
cya n op h en yl)m eth a n ep h osp h on a te (11). Naphthylmethyl
carbamate was used in the Oleksyszyn reaction39 to prepare
1-naphthylmethyloxycarbonyl-(4-CNPhGly)P(OPh)2. In this
procedure, 4-cyanobenzaldehyde (0.88 g, 6.7 mmol) is reacted
with 1-naphthylmethyl carbamate (0.9 g, 4.45 mmol) and
triphenyl phosphite (1.38 g, 4.45 mmol) in acetic acid at 85-
90 °C for 1.5 h. Removal of the solvent, followed by addition
of MeOH and 24 h of cooling at -20 °C gave the product as a
white precipitate: mp 146-149 °C; yield 80%; 1H NMR
(DMSO) δ 5.5-5.66 (dd, 2H, J ) 12.5 Hz), 5.77-5.88 (dd, 1H,
J ) 10.2 Hz), 6.98-7.02 (d, 4H, J ) 8.2 Hz), 7.15-7.25 (m,
2H), 7.25-7.39 (m, 4H), 7.44-7.62 (m, 4H), 7.81-8.08 (m, 7H),
9.0-9.04 (d, 1H, J ) 10.2 Hz); HRMS (FAB+) m/z calcd for
4-P h th a lim id obu ta n ol. This compound, a colorless oil,
was prepared in the same manner as 5-phthalimidopentanol:
1
yield 100%; H NMR (CDCl3) δ 1.6-1.7 (m, 2H), 1.7-1.9 (m,
2H), 3.7 (t, J ) 7.7 Hz, 2H), 4.35 (t, J ) 6.4 Hz, 1H), 7.7-7.8
(m, 2H), 7.8-7.9 (m, 2H), 7.8-7.9 (m, 2H).
4-P h th a lim id obu ta n a l (7a ). This compound, a colorless
oil, was prepared in the same manner as 5-phthalimido-
pentanal: yield 38%; 1H NMR (CDCl3) δ 2.0-2.1 (m, 2H), 2.5-
2.6 (td, J ) 6.0, 1.1 Hz, 2H), 3.7-3.8 (t, J ) 6.8 Hz, 2H), 7.7-
7.8 (m, 2H), 7.8-7.9 (m, 2H), 9.8 (s, 1H); MS (EI+) m/z 217.1.
C
32H25O5N2P (M + 1) 549.1579, found 549.1603. Anal.
(C32H25O5N2P‚0.5H2O) C, H, N.
Dip h en yl
N-(Ben zyloxyca r b on yl)a m in o(3-p h t h a l-
im idopr opyl)m eth an eph osph on ate [Cbz-Or n (P h t)P (OP h )2,
8a ]. This compound was prepared in the same manner as
Cbz-Lys(Pht)P(OPh)2: mp 72-74 °C; yield 47%; 1H NMR
(CDCl3) δ 1.5-2.2 (m, 4H), 3.8 (t, J ) 6.8 Hz, 2H), 4.5-4.7
(m, 1H), 5.0-5.2 (m, 2H), 7.0-7.4 (m, 15H), 7.7 (m, 2H), 7.9
(m, 2H); MS (FAB+) m/z 585.0.
Dip h en yl N-(1-Na p h th ylm eth yloxyca r bon yl)a m in o(4-
a m id in op h en yl)m et h a n ep h osp h on a t e H yd r och lor id e
[Na p h th ylm eth yloxyca r bon yl-(4-Am P h Gly)P (OP h )2, 13].
The preparation of this compound proceeds through the imino
ester derivative of compound 11.34 The synthesis begins by
adding compound 11 (1 g, 1.8 mmol) to CHCl3/MeOH (3:2) and
cooling to 0 °C. Gaseous HCl was allowed to pass through
the solution until saturation was reached, and then the
reaction mixture allowed to stir overnight at 4 °C. Removal
of the solvent and treatment with Et2O gave the imino ester.
After drying for 24 h, the imino ester and 7 equiv of NH4Cl
were taken up in EtOH and cooled to 0 °C. Addition of 1.5
equiv of NH3/MeOH (1.8 mL, 1.48 M) was followed by warming
to room temperature and stirring for 48 h. The solid material
formed was filtered, and the solvent was removed. The residue
was resuspended in CHCl3, gaseous HCl was allowed to pass
through the solution for about 1 min, and the solvent was
concentrated. The final compound was isolated using column
chromatography with CHCl3/MeOH/AcOH (8:2:0.2) as the
Dip h en yl N-(Ben zyloxyca r bon yl)a m in o(3-a m in op r o-
p yl)m eth a n ep h osp h on a te [Cbz-Or n P (OP h )2, 1]. This com-
pound was prepared in the same manner as Cbz-LysP(OPh)2
with the exception that the product was obtained as an acetate
1
salt: 111-113 °C; yield 66%; H NMR (DMSO) δ 1.8 (s, 3H),
1.5-2.1 (m, 4H), 2.6-2.7 (m, 2H), 4.2-4.4 (m, 1H), 5.0-5.2
(m, 2H), 7.1-7.5 (m, 15H), 8.1 (br s, 1H); MS (FAB+) m/z 455.0.
Anal. (C24H27O5N2P‚C2H4O2) C, H, N.
6-P h th a lim id oh exa n ol. This compound was prepared in
the same manner as 5-phthalimidohexanol from 6-aminohex-
anol (3.51 g, 30.0 mmol) and phthalic anhydride (4.44 g, 30.0
mmol) to give the product as a brown oil: yield 91%; 1H NMR
(CDCl3) δ 1.3-1.8 (m, 8H), 3.6-3.7 (m, 4H), 4.3 (t, J ) 6.5
Hz, 1H), 7.7 (m, 2H), 7.8 (m, 2H).
1
eluent: mp 140-145 °C; yield 10%; H NMR (DMSO) δ 5.4-
5.7 (dd, 2H), 5.7-5.9 (m, 1H), 6.9-7.7 (m, 15H), 7.7-8.2 (m,
7H), 9.0-9.6 (m, 2H), 9.6-10.4 (s, 2H); HRMS (FAB+) m/z
calcd for C32H28O5N3P (M + 1) 566.1845, found 566.1900.
Anal. (C32H28O5N3P‚HCl‚3.5H2O) C, H, N.
6-P h th a lim id oh exa n a l (7c). This compound was pre-
pared in the same manner as 5-phthalimidopentanal from
6-phthalimidohexanol (6.73 g, 27.2 mmol), DMSO (4.3 mL, 59.8
mmol), oxalyl chloride (2.9 mL, 32.7 mmol), and DIEA (14.2
mL, 81.7 mmol) to give the product as a light-brown oil: yield
81%; 1H NMR (CDCl3) δ 1.4 (m, 2H), 1.7 (m, 4H), 2.5 (td, J )
7.1, 1.2 Hz, 2H), 3.7 (t, J ) 6.3 Hz, 2H), 7.7-7.9 (m, 4H), 9.8
(s, 1H).
Dip h en yl N-Su ccin yla m in o(4-a m id in op h en yl)m eth a -
n eph osph on ate Hydr och lor ide [Su c-(4-Am P h Gly)P (OP h )2,
15]. Succinic anhydride (33 mg, 0.33 mmol) was added to a
stirred DMF solution at 0 °C that contained (4-AmPhGly)P-
(OPh)2‚HCl (0.15 g, 0.33 mmol) and was stirred for 15 min.
At this time TEA (33.4 mg, 0.33 mmol) was added, and the
solution stirred for an additional 30 min at 0 °C and then
overnight at room temperature. Any solid that formed was
removed by filtration, and the remaining solution was evapo-
rated to dryness. The residual oil was purified by silica gel
column chromatography with CHCl3/MeOH/AcOH (7:3:0.2) as
the eluent. The crude material was treated with xylene and
then toluene after column purification to remove the remaining
DMF. The final product was recrystallized from MeOH/
Et2O: mp 90-95 °C; yield 66%; 1H NMR (DMSO) δ 2.38-2.44
(m, 4H), 5.95-6.1 (m, 1H), 7.0-7.1 (m, 4H), 7.15-7.25 (m, 2H),
7.3-7.4 (m, 4H), 7.8-7.9 (m, 4H), 9.2-9.55 (m, 4H); HRMS
(FAB+) m/z calcd for C24H25O6N3P (M + 1) 482.1481, found
482.1541; reversed-phase HPLC (90% MeCN-H2O) tR ) 2.79
min.
Dip h en yl
N-(Ben zyloxyca r b on yl)a m in o(6-p h t h a l-
im id oh exyl)m eth a n ep h osp h on a te [Cbz-Hom oLys(P h t)P -
(OP h )2, 8c]. This compound was prepared in the same
manner as Cbz-Lys(Pht)P(OPh)2 from 6-phthalimidohexanal
(5.38 g, 22.0 mmol), triphenyl phosphite (5.75 g, 22.0 mmol),
and benzyl carbamate (3.33 g, 22.0 mmol) to give the product
as a light-brown solid: mp 95-96 °C; yield 40%; 1H NMR
(CDCl3) δ 1.3-2.1 (m, 8H), 3.7 (t, J ) 6.6 Hz, 2H), 4.4-4.5
(m, 1H), 5.1-5.2 (m, 2H), 7.0-7.4 (m, 15H), 7.7 (m, 2H), 7.9
(m, 2H). Anal. (C34H33O7N2P) C, H, N.
Dip h en yl
N-(Ben zyloxyca r b on yl)a m in o(6-a m in o-
h exyl)m eth a n ep h osp h on a te [Cbz-Hom oLysP (OP h )2, 3].
This compound was prepared in the same manner as
Cbz-LysP(OPh)2: mp 69-71 °C; yield 59%; 1H NMR (CDCl3) δ
1.2-2.0 (m, 8H), 2.7-2.9 (br s, 2H), 4.4 (m, 1H), 5.0 (m, 2H),
6.0 (d, J ) 9.6 Hz, 1H), 7.0-7.3 (m, 15H), 8.0-8.2 (br s, 3H);
MS (FAB+) m/z 483.1.
Dip h en yl
N-(tr a n s-cin n a m oyl)a m in o(4-a m id in o-
p h en yl)m eth a n ep h osp h on a te h yd r och lor id e [tr a n s-cin -
n a m oyl-(4-Am P h Gly)P (OP h )2, 16]: mp 111-118 °C; yield
18%; 1H NMR (DMSO) δ 5.2-5.4 (m, 1H), 6.1-6.4 (m, 1H),
6.6-6.8 (m, 2H), 6.9-7.65 (m, 16H), 7.65-8.0 (m, 3H), 8.4-
8.6 (m, 1H), 8.7-8.9 (m, 1H), 9.1-9.7 (m, 3H); MS (FAB+) m/z
512.2. Anal. (C29H26O4N3P‚HCl‚H2O) C, H, N.
1-Na p h th ylm eth yl Ca r ba m a te (10). This compound was
prepared using methods previously described.40 Briefly, tri-
fluoroacetic acid (2.22 g, 19.47 mmol) was slowly added to a
cooled (0 °C) benzene solution containing 1-naphthylmethanol
(1.54 g, 9.73 mmol) and sodium cyanate (1.27 g, 19.47 mmol).
The reaction was allowed to go overnight at room temperature.
A 20-mL portion of H2O was added to the mixture, and the
organic layer was separated, dried over MgSO4, filtered, and
concentrated to give a fluffy white solid. Recrystallization of
Dip h en yl
N-(3-(2-fu r yl)a cr yloyl)a m in o(4-a m id in o-
p h en yl)m eth a n ep h osp h on a te h yd r och lor id e [3-(2-fu r yl)-
a cr yloyl-(4-Am P h Gly)P (OP h )2, 17]: mp 97-105 °C; yield

Peptide Phosphonate Protease Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2297
10%; 1H NMR (DMSO) δ 6.1-6.3 (dd, 1H, J ) 9.8 Hz), 6.55-
6.63 (m, 1H), 6.7-6.8 (m, 2H), 6.83-6.84 (d, 1H, J ) 3.3 Hz),
7.0-7.1 (m, 4H), 7.1-7.25 (m, 3H), 7.3-7.4 (m, 4H), 7.45-
7.65 (m, 1H), 7.8-7.95 (m, 4H), 9.2-9.4 (m, 3H), 9.6-9.65 (m,
1H); MS (FAB+) m/z 502.4. Anal. (C27H24O5N3P‚HCl) C, H,
N.
Dip h en yl N-(3-(2-t h ien yl)a cr yloyl)a m in o(4-a m id in o-
p h en yl)m eth a n ep h osp h on a te h yd r och lor id e [3-(2-th ie-
n yl)a cr yloyl-(4-Am P h Gly)P (OP h )2, 18]: mp 114-119 °C;
yield 17%; 1H NMR (DMSO) δ 6.15-6.25 (dd, 1H, J ) 9.0 Hz),
6.65-6.85 (m, 2H), 7.0-7.15 (m, 5H), 7.15-7.25 (m, 2H), 7.3-
7.45 (m, 5H), 7.6-7.71 (m, 2H), 7.8-7.9 (m, 4H), 9.5-9.6 (m,
1H), 9.65-10.1 (br s, 3H); MS (FAB+) m/z 518.0. Anal.
(C27H24O4N3SP‚0.5HCl‚H2O) C, H, N.
Dip h en yl N-(tr a n s-3-(3-p yr id yl)a cr yloyl)a m in o(4-a m i-
d in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e [tr a n s-
3-(3-p yr id yl)a cr yloyl-(4-Am P h Gly)P (OP h )2, 19]: mp 103-
109 °C; yield 17%; 1H NMR (DMSO) δ 6.15-6.3 (m, 1H), 6.7-
6.8 (d, 1H), 6.9-7.6 (m, 14H), 7.7-8.1 (m, 5H), 8.5-8.6 (m,
1H), 8.7-8.9 (m, 1H), 9.1-9.7 (br s, 2H), 9.9-10.0 (m, 1H);
HRMS (FAB+) m/z calcd for C28H25O4N4P (M + 1) 513.1692,
found 513.1730; reversed-phase HPLC (90% MeCN-H2O) tR
) 2.51 min.
1H NMR (DMSO) δ 0.8-0.9 (m, 6H), 1.5-1.8 (m, 3H), 4.3-4.5
(br s, 1H), 5.6-5.8 (m, 1H), 7.0-7.5 (m, 17H), 7.8-8.1 (m, 4H),
9.1-10.0 (br s, 4H); HRMS (FAB+) m/z calcd for C34H37O6N4P
(M + 1) 629.2529, found 629.2545. Anal. (C34H37O6N4P‚-
HCl‚2.5H2O) C, H, N.
Dip h en yl N-(N-ben zyloxyca r bon ylp r olyl)a m in o(4-a m i-
d in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e [Cbz-
P r o-(4-Am P h Gly)P (OP h )2, 26]: mp 87-95 °C; yield 30%; 1H
NMR (DMSO) δ 1.6-1.85 (m, 3H), 2.1-2.3 (m, 1H), 3.25-3.35
(m, 2H), 4.48-4.6 (m, 1H), 4.9-5.1 (m, 2H), 5.9-6.15 (m, 1H),
6.9-7.45 (m, 14H), 7.6-7.64 (d, 1H, J ) 8.4 Hz), 7.75-7.95
(m, 4H), 9.1-9.2 (br s, 2H), 9.3-9.4 (br s, 2H), 9.48-9.65 (m,
1H); HRMS (FAB+) m/z calcd for C33H33O6N4P (M + Cl)
613.2216, found 613.2204. Anal. (C33H33O6N4P‚HCl‚DMF) C,
H, N.
Dip h en yl N-(N-ben zyloxyca r bon ylth r eon yl)a m in o(4-
a m id in op h en yl)m eth a n ep h osp h on a te [Cbz-Th r -(4-Am -
P h Gly)P (OP h )2, 27]: mp 80-105 °C; yield 17%; 1H NMR
(DMSO) δ 1.0-1.2 (2d, 3H), 1.8 (s, 3H), 3.8-3.9 (m, 1H), 4.2-
4.3 (m, 1H), 5.0-5.1 (m, 1H), 5.1 (s, 2H), 6.0-6.1 (m, 1H), 7.0-
7.4 (m, 15H), 7.8-7.9 (s, 4H), 9.3-10.5 (m, 4H); HRMS (FAB+)
m/z calcd for C32H33N4O7P 617.2165, found 617.2171. Anal.
(C32H33N4O7P‚0.5CH3COOH‚2H2O) C, H, N.
Dip h en yl
N-(3-p h en oxyb en zoyl)a m in o(4-a m id in o-
Dip h en yl N-(N-Ben zyloxyca r bon yllysyl)a m in o(4-a m i-
d in op h en yl)m eth a n ep h osp h on a te Hyd r och lor id e [Cbz-
Lys-(4-Am P h Gly)P (OP h )2, 28]. Blocked Cbz-Lys(Boc)-(4-
AmPhGly)2(OPh)2 was prepared and purified using the general
procedure from Cbz-Lys(Boc)-OH (0.84 g, 2 mmol) and (4-
AmPhGly)P(OPh)2‚HCl (1 g, 2 mmol). Removal of the side-
chain protecting group was accomplished by treatment with
TFA for 15-20 min at room temperature. The TFA was
concentrated, and the remaining material was worked up with
Et2O and purified by preparative TLC using CHCl3/MeOH/
AcOH (8:2:0.2) as the solvent: brown oil; yield 20%; 1H NMR
(DMSO) δ 1.2-1.7 (m, 8H), 4.2-4.4 (m, 1H), 4.9-5.1 (m, 2H),
6.0-6.15 (m, 1H), 7.0-7.15 (m, 5H), 7.15-7.4 (m, 12H), 7.5-
7.55 (d, 1H, J ) 8.5 Hz), 7.55-7.75 (m, 1H), 7.8-7.9 (d, 4H, J
) 8.2 Hz), 9.4-9.5 (m, 2H), 10.0-11.0 (br s, 4H); HRMS (FAB+)
m/z calcd for C34H38O6N5P (M + 1) 644.2635, found 644.2607;
reversed-phase HPLC (90% MeCN-H2O) tR ) 2.45 min.
Diph en yl N-(N-ben zyloxycar bon ylph en ylalan yl)am in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e
[Cbz-P h e-(4-Am P h Gly)P (OP h )2, 29]: mp 89-95 °C; yield
p h en yl)m et h a n ep h osp h on a t e h yd r och lor id e [3-p h en -
oxyben zoyl-(4-Am P h Gly)P (OP h )2, 20]: mp 85-93 °C; yield
1
32%; H NMR (DMSO) δ 6.2-6.4 (m, 1H), 7.0-7.3 (m, 10H),
7.3-7.5 (m, 8H), 7.6-7.7 (m, 2H), 7.8-8.0 (m, 4H), 9.2-10 (m,
4H); HRMS (FAB+) m/z (M + Cl) calcd for C33H28O5N3P
578.1845, found 578.1828. Anal. (C33H28O5N3P‚HCl‚0.75DMF)
C, H, N.
Dip h en yl
N-(2-p h en oxyb en zoyl)a m in o(4-a m id in o-
p h en yl)m et h a n ep h osp h on a t e h yd r och lor id e [2-p h en -
oxyben zoyl-(4-Am P h Gly)P (OP h )2, 21]: mp 85-93 °C; yield
28%; 1H NMR (DMSO) δ 6.1-6.3 (m, 1H), 6.9-7.1 (m, 5H),
7.1-7.26 (m, 3H), 7.26-7.4 (m, 7H), 7.45-7.6 (m, 3H), 7.75-
7.9 (m, 6H), 9.5-10.5 (m, 4H); MS (FAB+) m/z 578.0. Anal.
(C33H28O5N3P‚HCl‚1.5H2O) C, H, N.
Dip h en yl N-(1-Na p h t h ylsu lfon yl)a m in o(4-a m id in o-
p h en yl)m eth a n ep h osp h on a te Hyd r och lor id e [1-Np SO2-
(4-Am P h Gly)P (OP h )2, 22]. To prepare this compound 1-naph-
thylsulfonyl chloride (0.15 g, 0.66 mmol) was added to a
stirring DMF solution containing (4-AmPhGly)P(OPh)2‚HCl
(0.3 g, 0.66 mmol) at 0 °C. After the mixture stirred for 15
min, TEA (0.083 g, 0.82 mmol) was added, and the solution
stirred for an additional 30 min at 0 °C and then at 4 °C for
24 h. Any solid that formed was removed by filtration, and
the remaining solution was evaporated to dryness. The
residual oil was purified by silica gel column chromatography
with CHCl3/MeOH/AcOH (9:1:0.1) as the eluent give the
1
18%; H NMR (DMSO) δ 2.6-2.8 (m, 1H), 2.8-2.95 (m, 1H),
4.48-4.7 (m, 1H), 4.85-5.0 (m, 2H), 5.9-6.2 (m, 1H), 6.8-7.5
(m, 20H), 7.5-7.95 (m, 4H), 9.2-9.9 (m, 4H); MS (FAB+) m/z
662.7. Anal. (C37H34O6N4P‚HCl‚0.25H2O) C, H, N.
Dip h en yl N-(N-(2-P h en oxyb en zoylp r olyl))a m in o(4-
a m id in op h en yl)m et h a n ep h osp h on a t e H yd r och lor id e
[2-P h en oxyben zoyl-P r o-(4-Am P h Gly)P (OP h )2, 30]. This
compound was purified by the same method used with
1
product: mp 85-90 °C; yield 16%; H NMR (DMSO) δ 6.6-
1
6.8 (m, 1H), 6.9-7.5 (m, 10H), 7.6-7.8 (m, 7H), 7.9-8.3 (m,
5H), 8.5-9.4 (m, 4H); HRMS (FAB+) m/z calcd C30H26O5N3PS
(M + Cl) 572.1409, found 572.1398. Anal. (C30H26O5N3PS‚
HCl‚0.5H2O) C, H.
compound 15: mp 90-95 °C; yield 43%; H NMR (DMSO) δ
1.55-1.9 (m, 3H), 2.05-2.3 (m, 1H), 3.2-3.6 (m, 2H), 4.5-4.7
(m, 1H), 5.8-6.15 (m, 1H), 6.7-7.6 (m, 19H), 7.7-7.9 (m, 4H),
9.15-9.5 (m, 4H); HRMS (FAB+) m/z calcd for C38H35O6N4P
(M + 1) 675.2372, found 675.2338. Anal. (C38H35O6N4P‚
HCl‚0.5DMF) C, H, N.
Diph en yl N-(N-ben zyloxycar bon ylalan yl)am in o(4-am i-
d in op h en yl)m et h a n ep h osp h on a t e [Cb z-Ala -(4-Am P h -
Gly)P (OP h )2, 23]: mp 146-147 °C; yield 14%; 1H NMR
(DMSO) δ 1.1-1.3 (m, 3H), 4.25-4.45 (m, 1H), 4.9-5.2 (m,
2H), 5.7-6.1 (m, 1H), 6.9-7.1 (m, 4H), 7.15-7.25 (m, 2H),
7.25-7.4 (m, 9H), 7.5-7.65 (m, 1H), 7.75-7.9 (m, 4H), 9.3-
9.5 (m, 1H), 9.9-10.5 (br s, 3H); MS (FAB+) m/z 587.2. Anal.
(C31H31N4O6P‚CH3CO2H‚H2O) C, H, N.
Dip h en yl N-(N-ben zyloxyca r bon ylva lyl)a m in o(4-a m i-
d in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e [Cbz-
Va l-(4-Am P h Gly)P (OP h )2, 24]: mp 135-138 °C; yield 30%;
1H NMR (DMSO) δ 0.7-1.0 (m, 6H), 1.9-2.1 (m, 1H), 4.1-4.3
(m, 1H), 5.0-5.1 (m, 2H), 6.0-6.1 (m, 1H), 6.7-6.8 (m, 1H),
7.0-7.6 (m, 16H), 7.8 (d, J ) 8.5 Hz, 4H), 9.0-10.5 (br d, 4H);
HRMS (FAB+) m/z calcd for C33H35O6N4P (M + 1) 615.2372,
found 615.2434.
Dip h en yl N-(N-(3-P h en oxyb en zoylp r olyl))a m in o(4-
a m id in op h en yl)m et h a n ep h osp h on a t e H yd r och lor id e
[3-P h en oxyben zoyl-P r o-(4-Am P h Gly)P (OP h )2, 31]. Puri-
fication of this compound was similar to that of compound 15:
1
mp 104-110 °C; yield 33%; H NMR (DMSO) δ 1.5-1.9 (m,
3H), 2.1-2.3 (m, 1H), 3.2-3.6 (m, 2H), 4.5-4.8 (m, 1H), 5.9-
6.2 (m, 1H), 6.7-6.9 (m, 1H), 6.9-7.3 (m, 11H), 7.3-7.5 (m,
7H), 7.7-7.9 (m, 4H), 9.1-9.4 (m, 1H), 9.4-9.6 (m, 1H), 9.6-
10.0 (br s, 2H); MS (FAB+) m/z 675.1. Anal. (C38H35O6N4P‚
HCl‚1/3H2O) C, H, N.
Dip h en yl N-(N-(3,3-Dip h en ylp r op a n oylp r olyl))a m in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te Hyd r och lor id e
[3,3-Dip h en ylp r op a n oyl-P r o-(4-Am P h Gly)P (OP h )2, 32].
Purification of this compound was similar to that of compound
15: mp 95-98 °C; yield 15%; 1H NMR (DMSO) δ 1.6-1.9 (m,
3H), 2.2-2.3 (m, 1H), 3.15-3.3 (m, 2H), 3.4-3.6 (m, 2H), 4.33-
4.55 (m, 2H), 5.9-6.1 (m, 1H), 6.7-6.8 (m, 1H), 6.9-7.5 (m,
Diph en yl N-(N-ben zyloxyca r bon ylleu cyl)a m in o(4-a m i-
d in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e [Cbz-
Leu -(4-Am P h Gly)P (OP h )2, 25]: mp 106-108 °C; yield 20%;

2298 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
J ackson et al.
18H), 7.6-7.7 (m, 4H), 7.71-7.9 (m, 2H), 8.95-9.05 (m, 2H),
9.25-9.45 (m, 2H); HRMS (FAB+) m/z calcd for C40H39O5N4P
(M + 1) 687.2736, found 687.2770; reversed-phase HPLC (90%
MeCN-H2O) tR ) 2.77 min.
cation of this compound was similar to that of compound 15:
mp 65-70 °C; yield 60%; 1H NMR (DMSO) δ 1.1-1.3 (m, 3H),
4.3-4.4 (m, 1H), 4.4-4.55 (m, 1H), 4.9-5.1 (m, 2H), 6.0-6.15
(m, 1H), 7.0-7.15 (m, 2H), 7.18-7.5 (m, 10H), 7.5-7.6 (m, 3H),
7.7-7.73 (d, 2H, J ) 8.7 Hz), 7.8-7.9 (m, 2H), 7.96-7.99 (d,
2H, J ) 7.1 Hz), 8.05-8.1 (m, 1H), 9.1 (s, 2H), 9.3-9.45 (m,
2H); HRMS (FAB+) m/z calcd for C35H36O9N5P (M + 1)
702.2329, found 702.2332; reversed-phase HPLC (90% MeCN-
H2O) tR ) 3.09 min.
Dip h en yl N-(N-(3-P h en ylp r op a n oylp r olyl))a m in o(4-
a m id in op h en yl)m et h a n ep h osp h on a t e H yd r och lor id e
[3-P h en ylp r op a n oyl-P r o-(4-Am P h Gly)P (OP h )2, 33]. Pu-
rification of this compound was similar to that of compound
15: mp 61-65 °C; yield 58%; 1H NMR (DMSO) δ 1.6-1.9 (m,
3H), 2.0-2.1 (m, 1H), 2.2-2.3 (m, 1H), 2.7-2.9 (m, 2H), 3.3-
3.55 (m, 3H), 4.5-4.7 (m, 1H), 5.85-6.1 (m, 1H), 6.7-6.9 (m,
1H), 6.9-7.45 (m, 14H), 7.5-7.58 (d, 1H, J ) 8.2 Hz), 7.77-
7.95 (m, 4H), 9.05-9.15 (m, 2H), 9.25-9.5 (m, 2H); HRMS
(FAB+) m/z calcd for C34H35O5N4P (M + 1) 611.2423, found
611.2416; reversed-phase HPLC (90% MeCN-H2O) tR ) 3.14
min.
Dip h en yl N-(N-Ben zyloxyca r bon yl(N-(E-ter t-bu tyloxy-
ca r bon yllysyl))a la n yl)a m in o(4-a m id in op h en yl)m eth a n e-
p h osp h on a te Hyd r och lor id e [Cbz-Lys(Boc)-Ala -(4-Am -
P h Gly)P (OP h )2, 40]. Purification of this compound was
1
similar to that of compound 15: mp 80-86 °C; yield 34%; H
NMR (DMSO) δ 1.1-1.4 (m, 5H), 1.4-1.7 (m, 15H), 3.88-3.98
(m, 1H), 4.45-4.54 (m, 1H), 4.98-5.02 (m, 2H), 6.0-6.12 (dd,
1H, J ) 9.8 Hz), 6.7-6.78 (m, 1H), 6.99-7.02 (d, 1H, J ) 8.0
Hz), 7.07-7.10 (d, 1H, J ) 8.2 Hz), 7.16-7.25 (m, 1H), 7.25-
7.49 (m, 8H), 7.45-7.55 (t, 3H, J ) 7.2 Hz), 7.65-7.72 (d, 3H,
J ) 8.6 Hz), 7.81-7.91 (m, 1H), 7.92-8.0 (d, 3H, J ) 8.4 Hz),
8.04-8.1 (m, 1H), 9.0 (s, 2H), 9.3-9.41 (m, 2H); HRMS (FAB+)
m/z calcd for C42H51O9N6P (M + 1) 815.3533, found 815.3586;
reversed-phase HPLC (90% MeCN-H2O) tR ) 3.09 min.
Diph en yl N-(N-ben zyloxycar bon ylalan ylalan yl)am in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e
[Cbz-Ala -Ala -(4-Am P h Gly)P (OP h )2, 34]: mp 90-95 °C; yield
1
68%; H NMR (DMSO) δ 1.1-1.32 (m, 6H), 4.0-4.2 (m, 1H),
4.4-4.7 (m, 1H), 4.9-5.1 (m, 2H), 5.9-6.1 (m, 1H), 6.9-7.15
(m, 5H), 7.15-7.65 (m, 12H), 7.65-7.7 (m, 1H), 7.8-7.95 (m,
3H), 8.0-8.2 (m, 1H), 9.1-9.6 (m, 4H); MS (FAB+) m/z 658.1.
Anal. (C34H36O7N5P‚0.5HCl) C, H, N.
Dip h en yl N-(N-Ben zyloxyca r bon yllysyla la n yl)a m in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te Hyd r och lor id e
[Cbz-Lys-Ala -(4-Am P h Gly)P (OP h )2, 39]. Purification of this
compound was similar to that of compound 15: mp 106-111
Dip h en yl N-(N-Su ccin yla la n yla la n yl)a m in o(4-a m id i-
n op h en yl)m et h a n ep h osp h on a t e H yd r och lor id e [Su c-
Ala -Ala -(4-Am P h Gly)P (OP h )2, 35]. The tripeptide Cbz-Ala-
Ala-(4-AmPhGly)P(OPh)2 (1.0 g, 0.14 mmol) was deblocked
with 5% Pd/C, H2, and 1 equiv of HCl. The amine hydro-
chloride was added to a stirring DMF solution containing
succinic anhydride (14 mg, 0.14 mmol) and cooled to 0 °C. After
the mixture stirred for 15 min, 1 equiv of TEA (14.1 mg, 0.14
mmol) was added, and the solution stirred an additional 30
min at 0 °C and then overnight at room temperature. Any
solid formed was removed by filtration, and the remaining
solution was evaporated to dryness. The residual oil was
purified by silica gel column chromatography with CHCl3/
MeOH/AcOH (7:3:0.2) as the eluent. The crude material was
treated with xylene and then toluene to remove the remaining
DMF. The final product was recrystallized from MeOH/
1
°C; yield 50%; H NMR (DMSO) δ 1.1-1.4 (m, 6H), 1.4-1.7
(m, 5H), 3.9-4.0 (m, 1H), 4.45-4.55 (m, 1H), 4.98-5.02 (m,
2H), 6.0-6.12 (dd, 1H, J ) 9.8 Hz), 6.99-7.02 (d, 1H, J ) 8.0
Hz), 7.07-7.10 (d, 1H, J ) 8.4 Hz), 7.16-7.25 (m, 1H), 7.25-
7.49 (m, 8H), 7.5-7.58 (t, 3H, J ) 7.5 Hz), 7.62-7.78 (m, 4H),
7.82-7.92 (m, 2H), 7.96-8.0 (d, 3H, J ) 8.6 Hz), 8.04-8.13
(m, 1H), 9.14-9.45 (m, 4H); HRMS (FAB+) m/z calcd for
C
37H43O7N6P (M + 1) 715.3009, found 715.3051; reversed-
phase HPLC (90% MeCN-H2O) tR ) 3.13 min.
Diph en yl N-(N-ben zyloxycar bon ylph en ylalan yl)am in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e
[Cbz-P h e-Ala -(4-Am P h Gly)P (OP h )2, 41]: mp 120-125 °C;
yield 20%; 1H NMR (DMSO) δ 1.1-1.3 (m, 3H), 2.58-2.75 (m,
1H), 2.85-3.0 (m, 1H), 4.15-4.4 (m, 1H), 4.45-4.65 (m, 1H),
4.9-5.0 (m, 2H), 5.9-6.1 (m, 1H), 7.0-7.15 (m, 4H), 7.15-7.4
(m, 15H), 7.4-7.6 (m, 1H), 7.75-7.9 (m, 4H), 8.25-8.4 (m, 2H),
9.35-9.6 (m, 2H), 9.75-10.6 (br s, 2H); HRMS (FAB+) m/z
calcd C40H40O7N5P (M + 1) 734.2744, found 734.2771. Anal.
(C40H40O7N5P‚1/4HCl‚3H2O) C, H, N.
1
Et2O: mp 100-105 °C; yield 50%; H NMR (DMSO) δ 1.1-
1.3 (m, 6H), 2.3-2.44 (m, 4H), 4.2-4.35 (m, 1H), 4.45-4.57
(m, 1H), 5.95-6.1 (m, 1H), 7.0-7.13 (m, 5H), 7.15-7.25 (m,
2H), 7.3-7.4 (m, 1H), 7.8-7.9 (m, 4H), 8.05-8.17 (m, 2H), 9.2-
9.55 (m, 4H); HRMS (FAB+) m/z calcd for C30H35O8N5P
624.2223 (M + 1), found 624.2255; reversed-phase HPLC (90%
MeCN-H2O) tR ) 2.52 min.
Diph en yl N-(P h en ylm eth ylsu lfon ylglycylpr olyl)am in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te Hyd r och lor id e
[P h -CH2-SO2-Gly-P r o-(4-Am P h Gly)P (OP h )2, 43]. Purifica-
tion of this compound was similar to that of compound 15: mp
75-77 °C; yield 25%; 1H NMR (DMSO) δ 1.6-1.8 (m, 2H), 2.0-
2.2 (m, 1H), 3.1-3.2 (m, 1H), 3.7-3.8 (m, 2H), 3.85-3.91 (m,
1H), 4.0-4.15 (m, 1H), 4.25-4.4 (m, 2H), 4.55-4.8 (m, 1H),
5.88-6.15 (m, 1H), 6.9-7.4 (m, 15H), 7.74-7.95 (m, 4H), 9.25-
9.55 (m, 2H), 9.8-10.4 (m, 2H); HRMS (FAB+) m/z calcd for
Diph en yl N-(N-ben zyloxyca r bon ylp r olyla la n yl)a m in o-
(4-a m id in op h en yl)m eth a n ep h osp h on a te h yd r och lor id e
[Cbz-P r o-Ala -(4-Am P h Gly)P (OP h )2, 36]: mp 80-85 °C; yield
1
10%; H NMR (DMSO) δ 1.1-1.4 (m, 4H), 1.7-1.95 (m, 3H),
3.2-3.4 (m, 2H), 4.1-4.4 (m, 2H), 4.95-5.1 (m, 2H), 5.9-6.1
(m, 1H), 6.9-7.12 (m, 1H), 7.17-7.47 (m, 8H), 7.47-7.6 (m,
3H), 7.66-7.7 (d, 4H, J ) 8.6 Hz), 7.85-7.9 (m, 1H), 7.93-
7.96 (d, 4H, J ) 8.2 Hz), 8.15-8.3 (m, 1H), 9.0-9.1 (m, 2H),
9.25-9.4 (m, 2H); HRMS (FAB+) m/z calcd for C36H38O7N5P
(M + 1) 684.2587, found 684.2655; reversed-phase HPLC (90%
MeCN-H2O) tR ) 3.17 min.
C
34H36O7N5PS (M + 1) 690.2151, found 690.2466. Anal.
(C34H36O7N5PS‚HCl‚2H2O) C, H, N.
Diph en yl N-(N-Ben zyloxycar bon ylalan ylalan ylalan yl)-
a m in o(4-a m id in op h en yl)m eth a n ep h osp h on a te Hyd r o-
ch lor id e [Cbz-Ala -Ala -Ala -(4-Am P h Gly)P (OP h )2, 44]. Pu-
rification of this compound was similar to that of compound
15: mp 115-123 °C; yield 30%; 1H NMR (DMSO) δ 1.1-1.32
(m, 9H), 4.0-4.1 (m, 1H), 4.2-4.4 (m, 1H), 4.4-4.6 (m, 1H),
4.95-5.15 (m, 2H), 5.9-6.1 (m, 1H), 6.9-7.15 (m, 5H), 7.2-
7.4 (m, 11H), 7.4-7.6 (m, 2H), 7.8-8.0 (m, 5H), 8.0-8.15 (m,
1H), 9.1-9.5 (m, 4H); HRMS (FAB+) m/z calcd C37H41O8N6P
(M + 1) 729.2802, found 729.2852. Anal. (C37H41O8N6P‚
0.66HCl‚1.33DMF) C, H, N.
En zym e In h ibition Kin etics. 1. En zym es. Human
granzyme A was purified as previously described.53 Bovine
trypsin was obtained from Sigma Chemical Co., St. Louis, MO.
Rat granzymes A and K were prepared from the granules of
RNK-16 leukemia cells.54 Granules were prepared30 and
extracted by freeze-thawing with 1 M NaCl followed by
Dip h en yl N-(N-Ben zyloxyca r b on yl(N-(â-ter t-b u t yl-
a sp a r tyl))a la n yl)a m in o(4-a m id in op h en yl)m eth a n ep h os-
p h on a te Hyd r och lor id e [Cbz-Asp (t-Bu )-Ala -(4-Am P h G-
ly)P (OP h )2, 38]. Purification of this compound was similar
to that of compound 15: mp 105-111 °C; yield 25%; 1H NMR
(DMSO) δ 1.1-1.3 (m, 3H), 1.3-1.45 (m, 9H), 4.3-4.4 (m, 1H),
4.4-4.55 (m, 1H), 4.95-5.1 (m, 2H), 5.95-6.1 (m, 1H), 7.0-
7.1 (m, 2H), 7.17-7.25 (m, 1H), 7.25-7.45 (m, 7H), 7.45-7.55
(m, 4H), 7.65-7.68 (d, 4H, J ) 8.5 Hz), 7.8-7.9 (m, 1H), 7.93-
7.96 (d, 3H, J ) 8.2 Hz), 9.0 (s, 2H), 9.3-9.45 (m, 2H); HRMS
(FAB+) m/z calcd for C39H44O9N5P (M + 1) 758.2955, found
758.2988; reversed-phase HPLC (90% MeCN-H2O) tR ) 3.14
min.
Dip h en yl N-(N-Ben zyloxyca r b on yla sp a r t yla la n yl)-
a m in o(4-a m id in op h en yl)m eth a n ep h osp h on a te Hyd r o-
ch lor id e [Cbz-Asp -Ala -(4-Am P h Gly)P (OP h )2, 37]. Purifi-

Peptide Phosphonate Protease Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2299
removal of debris by centrifugation at 10000g. Three milli-
liters of the extract (1-3 mg of protein) was separated by size-
exclusion chromatography using a 2.5- × 100-cm column of
SuperDex200 (Pharmacia, Piscaway, NJ ) in a buffer of 20 mM
HEPES (Sigma Chemical Co., St. Louis, MO), pH 7.5, with 1
M NaCl, 0.05% (w/v) NaN3, 10% betaine (w/v; Sigma), and
0.1 mM EGTA. This initial step with high salt was required
to dissociate the granzymes from proteoglycan. Then fractions
containing each of the ∼45-kDa Gr A and ∼30-kDa Gr K peaks
were pooled from several SD200 runs, diluted with a MonoS
buffer of 50 mM MES (Sigma), pH 6, containing 0.05% NaN3,
10% betaine (w/v; Sigma), and 20% ethylene glycol (enzyme
grade; Fisher Biotech, Fair Lawn, NJ ) to 0.5 M NaCl, and
separated by cation-exchange chromatography with a Phar-
macia MonoS 0.5- × 5-cm column using a gradient of 0.5-1.0
M NaCl in the MonoS buffer. Enzyme purity was determined
by SDS-PAGE and the identity confirmed by amino acid
sequencing.
K) µL of 8 mM Cbz-Arg-SBzl in DMSO, and 10 µL of 16 mM
DTNB in DMSO.
Pseudo-first-order inactivation rate constants (kobs) were
obtained from plots of ln vt/vo vs time. Each kobs value was
calculated from 5-10 activity determinations which extended
to 2-3 half-lives for that particular inhibition reaction, except
for tryptase. Very slow inhibition against tryptase was
observed so the activity determination was only extended to
1 half-life (more than 1 h). Control experiments were carried
out in the same manner as described above, except DMSO was
added in place of the inhibitor solution in DMSO. For trypsin
three separate kobs values per inhibitor were determined, while
for granzymes A two and for granzyme K one or two kobs values
were determined.
Ack n ow led gm en t. We thank Dr. Sabiha Quazi for
preparing some of the peptide precursors. This work was
supported in part by a National Institutes of Health
Predoctoral Fellowship (1F31GM16378-01), a Graduate
Fellowship from Molecular Design Institute, Georgia
Institute of Technology, and grants from the National
Institutes of Health (HL29307 and GM54401).
Mast cell tryptase was purified from human lung tissue.
Human lung tissue was obtained from the National Disease
Research Institute (Philadelphia). Only the “high-molecular-
weight form” of human lung tryptase as described previously55
was used in these studies because it represents the majority
of the tryptase in lung mast cells and adequate quantities of
the other form were not available. This material is the same
as previously characterized56 and probably the same as puri-
fied by Schwartz.57 The enzyme was stored in 2 M NaCl in
10 mM MES, pH 6.1, 10% glycerol, 0.01% NaN3 at 4 °C. Mast
Refer en ces
(1) Abbreviations: (4-AmPhGly)P(OPh)2, diphenyl amino(4-ami-
dinophenyl)methanephosphonate or diphenyl 4-amidinophenyl-
glycinephosphonate; (4-AmPhe)P(OPh)2, diphenyl 1-amino-2-(4-
amidinophenyl)ethanephosphonate or diphenyl 4-amidino-
phenylalaninephosphonate); Boc, tert-butyloxycarbonyl; Cbz,
benzyloxycarbonyl; (CNPhGly)P(OPh)2, diphenyl amino(4-cy-
anophenyl)methanephosphonate or diphenyl 4-cyanophenylgly-
cinephosphonate; CTL, cytotoxic T lymphocyte; DCC, N,N′-
dicyclohexylcarbodiimide; DIEA, diisopropyl ethylamine; DMF,
N,N-dimethylformamide; DMSO, dimethyl sulfoxide; FAB, fast
atom bombardment; Hepes, N-(2-hydroxyethyl)piperazine-N′-2-
ethanesulfonic acid; HRMS, high-resolution mass spectroscopy;
HOBt, 1-hydroxybenzotriazole; HomoLysP(OPh)2, diphenyl 1,6-
diaminohexanephosphonate; HPLC, high-performance liquid
chromatography; LysP(OPh)2, diphenyl 1,5-diaminopentanephos-
phonate; MES, 2-(4-morpholino)ethanesulfonic acid; MUGB,
4-methylumbelliferyl p-guanidinobenzoate; NK, natural killer;
OrnP(OPh)2, diphenyl 1,4-diaminobutanephosphonate; SBzl,
thiobenzyl ester; Suc, succinyl; TEA, triethylamine; TFA, tri-
fluoroacetic acid; TLC, thin-layer chromatography. The (R-
aminoalkyl)phosphonic acids are analogues of natural R-amino
acids and are designated by the generally accepted three-letter
abbreviations for the amino acid followed by a superscript P.
For example, diphenyl 1-[N-(benzyloxycarbonyl)amino]-2-(4-
amidinophenyl)ethanephosphonate, which is related to 4-ami-
dinophenylalanine, is abbreviated as Cbz-(AmPhe)P(OPh)2.
(2) Berke, G. Cytotoxic T-Lymphocytes: How do they Function?
Immunol. Rev. 1983, 72, 5-38.
(3) Berke, G. Killing Mechanisms of Cytotoxic Lymphocytes. Curr.
Opin. Hematol. 1997, 4, 32-40.
(4) Henkart, P. A.; Millard, P. J .; Reynolds, C. W.; Henkart, M. P.
Cytolytic Activity of Purified Cytoplasmic Granules from Cyto-
toxic Rat Large Granular Lymphocyte Tumors. J . Exp. Med.
1984, 160, 75-93.
(5) Podack, E. R.; Konigsberg, P. J . Cytolytic T Cell Granules.
Isolation, Structural, Biochemical, and Functional Characteriza-
tion. J . Exp. Med. 1984, 160, 695-710.
(6) Masson, D.; Tschopp, J . Isolation of Lytic, Pore-forming Protein
(Perforin) from Cytolytic T-Lymphocyte. J . Biol. Chem. 1985,
260, 9069-9072.
(7) Podack, E. R.; Young, J . D. E.; Cohn, Z. A. Isolation and
Biochemical and Functional Characterization of Perforin 1 from
Cytolytic T-Cell Granules. Prod. Natl. Acad. Sci. U.S.A. 1985,
82, 8629-8633.
(8) Hudig, D.; Ewoldt, G. R.; Woodard, S. L. Protease and Lympho-
cyte Cytotoxic Killing Mechanisms. Curr. Opin. Immunol. 1993,
5, 90-96.
cell tryptase protein concentrations were based on E1%
)
280
28,56 and enzyme active-site concentrations were determined
by MUGB titration.58 Enzyme concentrations and molar ratios
refer to enzyme active sites (monomers). Only enzyme prepa-
rations with greater than 90% active sites were used in this
study. The enzyme stock solutions were determined to be
heparin-free by assessing the presence of heparin with tolui-
dine-blue.59
2. In cu ba tion Meth od . Inhibition of trypsin was initiated
by adding 10 µL of inhibitor (0.055-5.0 mM in DMSO) to 0.49
mL of buffer (0.1 M Hepes, 0.01 M CaCl2, pH 7.5) containing
10 µL of a trypsin stock solution at 25 °C. At different times,
aliquots of 10 µL were withdrawn and added to the assay
mixture, and the residual enzymatic activity was measured
at 410 nm with a Beckman DU-650 spectrometer. The trypsin
stock solution was 5 µM in 1 mM HCl (pH 3) and was stored
at -20 °C prior to use. The assay mixture contained 720 µL
of buffer, 35 µL of 8 mM Cbz-Arg-SBzl in DMSO,52 and 35 µL
of 16 mM DTNB in DMSO.60
The tryptase inhibition reactions were initiated by adding
25 µL of inhibitor (0.5 mM) to 300 µL of buffer (0.1 M Hepes,
10% glycerol, 10 mM heparin, pH 7.5) containing 25 µL of a
stock enzyme solution at 25 °C. Aliquots of 50 µL were
withdrawn at various time intervals and added to the assay
mixture, and the residual enzymatic activity was measured
at 405 nm. The concentration of tryptase stock solution was
7 µM. The enzyme stock solution was prepared in 10 mM
MES, 2 M NaCl, 10% glycerol, and 0.01% NaN3, pH 6.1. The
assay solution contained 1 mL of buffer, 10 µL of 10 mM Cbz-
Lys-SBzl (or Cbz-Arg-SBzl) in DMSO, and 10 µL of 16 mM
DTNB in DMSO.
Human recombinant and rat granzyme A and native rat
granzyme K inhibition reactions were initiated by adding 10
µL of inhibitor (0.014-5.0 mM in DMSO) to 100 µL of buffer
(0.1 M Hepes, 0.01 M CaCl2, pH 7.5), to which 10-30 µL of a
stock enzyme solution had been previously added at 25 °C.
The enzyme stock solutions were 0.3, 0.41, and 0.33 µM in 0.15
M NaCl for human and rat granzymes A and granzyme K,
respectively. All the enzyme stock solutions were stored at
-20 °C prior to use. Aliquots of 20 (granzyme A) or 30
(granzyme K) µL were withdrawn at various time intervals
from the inhibition mixture, and the residual enzymatic
activity was measured spectrophotometrically at 405 nm with
a Molecular Devices microplate reader. The assay solution
contained 200 µL of buffer, 10 (granzyme A) or 20 (granzyme
(9) Meier, M.; Kwong, P. C.; Fregeau, C. J .; Atkinson, E. A.;
Burrington, M.; Ehrman, N.; Sorensen, O.; Lin, C. C.; Wilkins,
J .; Bleackley, R. C. Cloning of a Gene that Encodes a New
Member of the Human Cytotoxic Cell Protease Family. Bio-
chemistry 1990, 29, 4042-4049.
(10) Smyth, M. J .; Trapani, J . A. Granzymes: Exogenous Proteinases
that Induce Target Cell Apoptosis. Immunol. Today 1995, 16,
202-206.

2300 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13
J ackson et al.
(11) Shi, L.; Kam, C.-M.; Powers, J . C.; Aebersold, R.; Greenberg, A.
H. Purification of Three Cytotoxic Lymphocyte Granule Serine
Proteases that Induce Apoptosis through Distinct Substrate and
Target Cell Interactions. J . Exp. Med. 1992, 176, 1521-1529.
(12) Shi, L.; Kraut, R. P.; Aebersold, R.; Greenberg, A. H. A Natural
Killer Cell Granule Protein that Induces DNA Fragmentation
and Apoptosis. J . Exp. Med. 1992, 175, 553-566.
(13) Shiver, J . W.; Su, L.; Henkart, P. A. Cytotoxicity with Target
DNA Breakdown by Rat Basophilic Leukemia Cells Expressing
both Cytolysin and Granzyme A. Cell 1992, 71, 315-322.
(14) Suidan, H. S.; Bouvier, J .; Schaerer, E.; Stone, S. R.; Monard,
D.; Tschopp, J . Granzyme A Released upon Stimulation of
Cytotoxic T. Lymphocytes Activates the Thrombin Receptor on
Neuronal Cells and Astrocytes. Proc. Natl. Acad. Sci. U.S.A.
1994, 91, 8112-8116.
(15) Simon, M. M.; Kramer, M. D.; Prester, M.; Gay, S. Mouse T-cell
Associated Serine Proteinase 1 Degrades Collagen Type IV: A
Structural Basis for the Migration of Lymphocytes Through
Vascular Basement Membranes. Immunology 1991, 73, 117-
119.
(16) Fruth, U.; Sinigaglia, F.; Schlesier, M.; Kilgus, J .; Kramer, M.
D.; Simon, M. M. A Novel Serine Proteinase (HuTSP) Isolated
from a Cloned Human CD8+ Cytolytic T Cell Line is Expressed
and Secreted by Activated CD4+ and CD8+ Lymphocytes. Eur.
J . Immunol. 1987, 17, 1625-1633.
(17) Poe, M.; Wu, J . K.; Blake, J . T.; Zweerink, H. J .; Sigal, N. H.
The Enzymatic Activity of Human Cytotoxic T-Lymphocyte
Granzyme A and Cytolysis Mediated by Cytotoxic T-Lympho-
cytes are Potently Inhibited by a Synthetic Antiprotease, FUT-
175. Arch. Biochem. Biophys. 1991, 284, 215-218.
(18) Poe, M.; Blake, J . T.; Boulton, D. A.; Gammon, M.; Sigal, N. H.;
Wu, J . K.; Zweerink, H. J . Human Cytotoxic Lymphocyte
Granzyme B: Its Purification from Granules and the Charac-
terization of Substrate and Inhibitor Specificity. J . Biol. Chem.
1991, 266, 98-103.
(32) Hamilton, R.; Walker, B. J .; Walker, B. A Convenient Synthesis
of N-Protected Diphenyl Phosphonate Esters Analogues of
Ornithine, Lysine, and Homolysine. Tetrahedron Lett. 1993, 34,
2847-2850.
(33) Hamilton, R.; Kay, G.; Shute, R. E.; Travers, J .; Walker, B. J .;
Walker, B. The Synthesis of Phosphonate Analogues of Amino
Acids and Peptides. Phosphorus Sulfur Silicon 1993, 76, 127-
130.
(34) Oleksyszyn, J .; Boduszek, B.; Kam, C.-M.; Powers, J . C. Novel
Amidine-Containing Peptidyl Phosphonates as Irreversible In-
hibitors for Blood Coagulation and Related Serine Proteases. J .
Med. Chem. 1994, 37, 226-231.
(35) Mueller, C.; Shelby, J .; Weissman, I. L.; Perinat-Frey, T.;
Eichwald, E. J . Expression of the Protease Gene HF as a Marker
in Rejecting Allogeneic Murine Heart Transplants. Transplanta-
tion 1991, 51, 514-517.
(36) Griffiths, G. M.; Alpert, S.; Lambert, E.; McGuire, J .; Weissman,
I. L. Perforin and Granzyme A Expression Identifying Cytolytic
Lymphocytes in Rheumatoid Arthritis. Proc. Natl. Acad. Sci.
U.S.A. 1992, 89, 549-553.
(37) Tanaka, R. D.; Clark, J . M.; Warne, R. L.; Abraham, W. M.;
Moore, W. R. Mast Cell Tryptase: A New Target for Therapeutic
Intervention in Asthma. Int. Arch. Allergy Immunol. 1995, 107,
408-409.
(38) Mancuso, A. J .; Huang, S.-L.; Swern, D. Oxidation of Long-Chain
and Related Alcohols to Carbonyls by Dimethyl Sulfoxide
“Activated” by Oxalyl Chloride. J . Org. Chem. 1978, 43, 2480-
2482.
(39) Oleksyszyn, J .; Subotkowska, L.; Mastalerz, P. Diphenyl 1-Ami-
noalkanephosphonates. Synthesis 1979, 985-986.
(40) Loev, B.; Kormendy, M. F. An Improved Synthesis of Carbam-
ates. J . Org. Chem. 1963, 28, 3421-3426.
(41) Bertrand, J . A.; Oleksyszyn, J .; Kam, C.-M.; Boduszek, B.;
Presnell, S.; Plaskon, R. R.; Suddath, F. L.; Powers, J . C.;
Williams, L. D. Inhibition of Trypsin and Thrombin by Amino-
(4-amidinophenyl)methanephosphonate
Diphenyl
Ester
(19) Reynolds, D. S.; Stevens, R. L.; Lane, W. S.; Carr, W. H.; Austen,
K. F.; Serafin, W. E. Different Mouse Mast Cell Populations
Express Various Combinations of at Least Six Distinct Mast Cell
Serine Proteases. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 3230-
3234.
(20) Odake, S.; Kam, C.-H.; Narasimnah, L.; Poe, M.; Blake, J . T.;
Krahenbuhl, O.; Tschopp, J .; Powers, J . C. Human and Murine
Cytotoxic T Lymphocyte Serine Proteases: Subsite Mapping
with Peptide Thioesters Substrates and Inhibition of Enzyme
Activity and Cytolysis by Isocoumarins. Biochemistry 1991, 30,
2217-2227.
(21) Hudig, D.; Allison, N. J .; Kam, C.-M.; Powers, J . C. Selective
Isocoumarin Serine Protease Inhibitors Block RNK-16 Lympho-
cyte Granule-Mediated Cytolysis. Mol. Immunol. 1989, 26, 793-
798.
(22) Glover, G.; Shaw, E. Purification of Thrombin and Isolation of
a Peptide Containing the Active Center Histidine. J . Biol. Chem.
1971, 246, 4594-4601.
(23) Turner, A. D.; Monroe, D. M.; Roberts, H. R.; Porter, N. A.; Pizzo,
S. V. p-Amidino Esters are Irreversible Inhibitors of Factor IXa
and Xa and Thrombin. Biochemistry 1986, 25, 4929-4935.
(24) Yagasho, T.; Nunomura, S.; Okutome, T.; Nakayama, T.; Ku-
rumi, M.; Sakurai, Y.; Aoyama, T.; Fujii, S. Synthesis and
Structure-Activity Study of Protease Inhibitors. III. Amidi-
nophenols and their Benzyl Esters. Chem. Pharm. Bull. 1984,
32, 2.
(25) Harper, J . W.; Powers, J . C. Reaction of Serine Proteases with
Substituted 3-Alkoxy-4-chloroisocoumarins and 3-Alkoxy-7-
amino-4-chloroisocoumarins. New Reactive Mechanism Based
Inhibitors. Biochemistry 1985, 24, 7200-7213.
(26) Bartlett, P. A.; Lamden, L. A. Inhibition of Chymotrypsin by
Phosphonate and Phosphonamidate Peptide Analogues. Bioorg.
Chem. 1986, 14, 356-377.
(27) Lamden, L. A.; Bartlett, P. A. Aminoalkylphosphonofluoridate
Derivatives: Rapid and Potentially Selective Inactivators of
Serine Peptidases. Biochem. Biophys. Res. Commun. 1983, 112,
1085-1090.
Derivatives: X-ray Structures and Molecular Models. Biochem-
istry 1996, 35, 3147-3155.
(42) J ohnson, D. A.; Barton, G. J . Mast Cell Tryptases: Examination
of Unusual Characteristics by Multiple Sequence Alignment and
Molecular Modeling. Protein Sci. 1992, 1, 370-377.
(43) Murphy, M. E. P.; Moult, J .; Bleackley, R. C.; Gershenfeld, H.;
Weissman, I. L.; J ames, M. N. G. Comparative Molecular Model
Building of Two Serine Proteinases from Cytotoxic T Lympho-
cytes. Proteins Struct. Funct. Genet. 1988, 4, 190-204.
(44) Sayers, T. J .; Wiltrout, T. A.; Smyth, M. J .; Ottawa, K. J .; Pilaro,
A. M.; Sowder, R.; Henderson, L. E.; Sprenger, H.; Lloyed, A. R.
Purification and Cloning of a Novel Serine Protease, RNK-Tryp-
2, from the Granules of a Rat Natural Killer Cell Leukemia. J .
Immunol. 1994, 152, 2289-2297.
(45) Gershenfeld, H. K.; Hershberger, R. J .; Shows, T. B.; Weissman,
I. L. Cloning and Chromosomal Assignment of a Human cDNA
Encoding a T Cell- and Natural Killer Cell-Specific Trypsin-Like
Serine Protease. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 1184-
1188.
(46) Miller, J . S.; Westin, E. H.; Schwartz, L. B. Cloning and
Characterization of Complementary DNA for Human Tryptase.
J . Clin. Invest. 1989, 84, 1188-1195.
(47) Huber, R.; KuKla, D.; Bode, W.; Schwager, P.; Bartels, K.;
Diesenhofer, J .; Steigemann, W. Structure of the Complex
formed by Bovine Trypsin and Bovine Pancreatic Trypsin
Inhibitor. II. Crystallographic refinement at 1.9 Å Resolution.
J . Mol. Biol. 1974, 89, 73-101.
(48) Bode, W.; Walter, J .; Huber, R.; Wenzel, H. R.; Tschesche, H.
The Refined 2.2-Å (0.22-nm) X-ray Crystal Structure of the
Ternary Complex Formed by Bovine Trypsinogen, Valine-valine,
and the Arg15 Analogue of Bovine Trypsin Inhibitor. Eur. J .
Biochem. 1984, 144, 185-190.
(49) Bajusz, S.; Baraba´s, E.; Fauszt, I.; Fehe´r, A.; Horvta´h, G.;
J uha´sz, A.; Szabo´, A. G.; Sze´ll, E. Active Site-Directed Thrombin
Inhibitors: R-Hydroxyacyl-Prolyl-Arginals. New Orally Active
Stable Analogues of D-Phe-Pro-Arg-H. Bioorg. Med. Chem. 1995,
3, 1079-1089.
(50) Cacciola, J .; Fevig, J . M.; Alexander, R. S.; Brittelli, D. R.;
Kettner, C. A.; Knabb, R. M.; Weber, P. C. Synthesis of
Conformationally restricted Boropeptide Thrombin Inhibitors.
Bioorg. Med. Chem. Lett. 1996, 6, 301-306.
(51) Fevig, J . M.; Abelman, M. M.; Brittelli, D. R.; Kettner, C. A.;
Knabb, R. M.; Weber, P. C. Design and Synthesis of Ring-
Constrained Boropeptide Thrombin Inhibitors. Bioorg. Med.
Chem. Lett. 1996, 6, 295-300.
(28) Oleksyszyn, J .; Powers, J . C. Irreversible Inhibition of Serine
Proteases by Peptidyl Derivatives of R-Aminoalkylphosphonate
Diphenyl Esters. Biochem. Biophys. Res. Commun. 1989, 161,
143-149.
(29) Oleksyszyn, J .; Powers, J . C. Irreversible Inhibition of Serine
Proteases by Peptide Derivatives of (R-Aminoalkyl)phosphonate
Diphenyl Esters. Biochemistry 1991, 30, 485-493.
(30) Woodard, S. L.; J ackson, D. S.; Abuelyaman, A. S.; Powers, J .
C.; Winkler, U.; Hudig, D. Chymase-Directed Serine Protease
Inhibitor that Reacts with a Single 30-kDa Granzyme and Blocks
NK-Mediated Cytotoxicity. J . Immunol. 1994, 153, 5016-5025.
(31) Fastrez, J .; J espers, L.; Lison, D.; Renard, M.; Sonveaux, E.
Synthesis of New Phosphonate Inhibitors of Serine Proteases.
Tetrahedron Lett. 1989, 30, 6861-6864.
(52) Cook, R. R.; McRae, B. J .; Powers, J . C. Kinetics of Hydrolysis
of Peptide Thioester Derivatives of Arginine by Human and
Bovine Thrombin. Arch. Biochem. Biophys. 1984, 234, 82-88.
(53) Hanna, W. L.; Zhang, X.; Turbov, J .; Winkler, U.; Hudig, D.;
Froelich, C. J . Rapid Purification of Cationic Granule Pro-
teases: Application to Human Granzymes. Protein Purif. Ex-
press. 1993, 4, 398-402.

Peptide Phosphonate Protease Inhibitors
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 13 2301
(54) Reynolds, C. W.; Bere, J . E. W.; Ward, J . M. Natural Killer
Activity in the Rat. III. Characterization of Transplantable Large
Lymphocyte (LGL) Leukemias in the F344 Rat. J . Immunol.
1984, 132, 534-540.
(58) J ameson, G. W.; Roberts, D. V.; Adams, R. W.; Kyle, W. S. A.;
Elmore, D. T. Determination of the Operational Molarity of
Solutions of Bovine R-Chymotrypsin, Trypsin, Thrombin, and
Factor Xa by Spectrofluorometric Titration. Biochem. J . 1973,
131, 107-117.
(59) Smith, P. K.; Mallia, A. K.; Hermanson, G. T. Colorimetric
Method for the Assay of Heparin Content in Immobilized
Heparin Preparations. Anal. Biochem. 1980, 109, 466-473.
(60) Ellman, G. L.; Courtney, D.; Andres, V.; Featherstone, R. M. A
New and Rapid Colorimetric Determination of Acetylcholin-
esterase. Biochem. Pharmacol. 1961, 7, 88-95.
(55) Little, S.; J ohnson, D. A. Human Mast Cell Tryptase Isoforms:
Separation and Examination of Substrate-Specificity Differences.
Biochem. J . 1995, 307, 341-346.
(56) Smith, T. J .; Houghland, M. W.; J ohnson, D. A. Human Lung
Tryptase. Purification and Characterization. J . Biol. Chem.
1984, 259, 11046-11051.
(57) Schwartz, L. B.; Lewis, R. A.; Austen, K. F. Tryptase from
Human Pulmonary Mast Cells. Purification and Characteriza-
tion. J . Biol. Chem. 1981, 256, 11939-11943.
J M970543S