Selective inhibition of type-I geranylgeranyltransferase in vitro and in whole cells by CAAL peptidomimetics

Article abstract of DOI:10.1016/S0968-0896(97)10040-2

In this paper we describe the synthesis of a family of CAAL peptidomimetics as GGTase-I inhibitors. These inhibitors lack the central dipeptide AA in the key CAAL carboxy terminal sequence of geranylgeranylated proteins and are more selective for GGTase-I over FTase. In whole cells, these compounds are very potent inhibitors of the processing of the geranylgeranylated protein Rap1A without affecting the farnesylated protein H-Ras. One derivative, GGTI-298, inhibited cell division by blocking cells in the G₁ phase of the cell cycle.

Full text of DOI:10.1016/S0968-0896(97)10040-2

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 293±299
Selective Inhibition of Type-I Geranylgeranyltransferase In
Vitro and in Whole Cells by CAAL Peptidomimetics
Yimin Qian, Andreas Vogt, Anil Vasudevan, Saõd M. Sebti*y and
Andrew D. Hamilton*{
Departments of Chemistry and Pharmacology, University of Pittsburgh, Pittsburgh, PA 15260, U.S.A.
Received 20 March 1997; accepted 20 October 1997
AbstractÐIn this paper we describe the synthesis of a family of CAAL peptidomimetics as GGTase-I inhibitors. These
inhibitors lack the central dipeptide AA in the key CAAL carboxy terminal sequence of geranylgeranylated proteins
and are more selective for GGTase-I over FTase. In whole cells, these compounds are very potent inhibitors of the
processing of the geranylgeranylated protein Rap1A without a€ecting the farnesylated protein H-Ras. One derivative,
GGTI-298, inhibited cell division by blocking cells in the G1 phase of the cell cycle. # 1998 Elsevier Science Ltd. All
rights reserved.
Introduction
So far, the major e€ort in designing prenyltransferase
inhibitors has been targeted to FTase, an enzyme that
catalyzes the farnesylation of oncogenic Ras, in the
hope of developing new classes of anti-tumor agents.7
However, the majority of prenylated proteins are ger-
anylgeranylated.8 In cells, the number of geranylger-
anylated proteins is at least 5 to 10 times more than
farnesylated proteins1. Proteins modi®ed by GGTase-I
include Rap1A (CLLL), Rap1B (CQLL), Rac1
(CLLL), Rac2 (CSLL), G25K (CCIF), and RhoA
(CLVL).9 Recently, it was shown that the geranylger-
anylated proteins Rho, Rac, and Cdc42 played an
essential role in cell cycle progression and mitosis.10
Therefore, it is important to obtain speci®c GGTase-I
inhibitors and apply them as probes to study the e€ects
of protein geranylgeranylation in normal and oncogenic
cell growth.
The study of protein isoprenylation has revealed that
either a C15 farnesyl or C20 geranylgeranyl group can be
covalently attached to proteins through the thiol group
of a cysteine side chain near the carboxylate terminus.1
Several small G-proteins require this prenylation for
proper cellular localization and regulation of signal
transduction. Enzymes that catalyze protein prenylation
are farnesyltransferase (FTase), type-I geranylgeranyl-
transferase (GGTase-I), and type-II geranylgeranyl-
transferase (GGTase-II).2±4 FTase catalyzes the
farnesylation of proteins with a CAAX sequence at their
carboxylate terminus where C is cysteine, A is an ali-
phatic amino acid and X is methionine, serine, gluta-
mine, alanine, or cysteine.1 Like FTase, GGTase-I
catalyzes the geranylgeranylation of proteins terminat-
ing with CAAX sequences where X is restricted to leu-
cine, isoleucine or to a lesser extent, phenylalanine.5 In
contrast to FTase and GGTase-I, GGTase-II modi®es
proteins with CC or CXC sequences at the carboxylate
terminus.6
The design of GGTase-I inhibitors can be approached
by mimicking the protein substrate at the C-terminal
sequence (CAAL) or by mimicking the isoprenyl sub-
strate geranylgeranylpyrophosphate (GGPP). GGPP
analogs have been reported recently to inhibit ger-
anylgeranylation of Rap in cell cultures.11 In this paper
we report the design of a novel series of peptidomimetic
inhibitors of GGTase-I and show them to be e€ective
at selectively blocking Rap1A processing in whole
cells.
*Corresponding author.
yCurrent address: H. Lee Mott Cancer Center, Dept. Bio-
chem. Mol. Biol., University of South Florida, Tampa, FL
33612, U.S.A.
{Current address: Department of Chemistry, Yale University,
New Haven, CT 06511, U.S.A.
0968-0896/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(97)10040-2

294
Y. Qian et al./Bioorg. Med. Chem. 6 (1998) 293±299
Synthesis and Biological Assay
and 6 were prepared by using, respectively, the methyl
esters of isoleucine and norleucine in place of leucine
methyl ester. The purity of all ®nal compounds was
checked by analytical HPLC and structures were con-
Peptidomimetics 2a and 2b were synthesized using a
similar approach to that described previously.12 The
coupling of 4-N-Boc-aminobenzoic acid with l-leucine
methyl ester gave the 4-N-Boc-aminobenzoylleucine
methyl ester. Deprotection of the Boc group and sub-
sequent reductive amination with N-Boc-S-trityl-l-
cysteinal gave the fully protected derivatives which were
then deprotected to give 2a and 2b. Compounds 3a and
3b were synthesized from 4-nitro-2-phenylbenzoic
acid.13 The coupling of this carboxylic acid with l-leu-
cine methyl ester in the presence of EDCI and HOBT
provided the corresponding amide. The nitro group was
reduced by hydrogenation and the amine was reacted
with N-Boc-S-trityl-l-cysteinal in the presence of
sodium cyanoborohydride. The fully protected com-
pounds were deprotected by ®rst treatment with aqu-
eous LiOH/THF and then tri¯uoroacetic acid and
triethylsilane. The crude product was puri®ed by pre-
parative HPLC (C-18 column, linear gradient within
forty minutes from 5% acetonitrile to 60% acetonitrile
in aqueous solution containing 0.1% tri¯uroacetic acid)
to give 3a. Compound 3b was prepared by deprotection
of the fully protected compound using tri¯uoroacetic
acid and triethylsilane. Compounds 4a and 4b were
prepared using an analogous route except that the 4-
nitro-2-(1-naphthyl)benzoic acid was used as the spacer.
The proton and carbon NMR data of 4a and 4b showed
two isomers due to the restricted rotation of the aryl±aryl
bond on the NMR time scale. Similarly, compounds 5
1
sistent with H NMR, 13C NMR, high resolution mass
spectra and microanalytical data.
The in vitro inhibition assays of GGTase-I and FTase
were carried out by measuring the [3H]-GGPP and [3H]-
FPP incorporated into H-Ras-CVLL and H-Ras-CVLS,
respectively, using the same procedure described before
(selected data are shown in Figure 2).12 The in vivo
inhibition of geranylgeranylation and farnesylation was
determined based on the level of inhibition of Rap1A
and H-Ras processing, respectively.14 Brie¯y, cells were
treated with various concentrations of peptidomimetics
and the cell lysates were separated on 12.5% SDS-
PAGE. The separated proteins were transferred to
nitrocellulose and immunoblotted using an anti-Ras
antibody (Y-13-258) or an anti-Rap1A antibody (SC-
65). Antibody reactions were visualized using either
peroxidase-conjugated goat anti-rat IgG or goat anti-
rabbit IgG and an enhanced chemiluminescene detec-
tion system.
Results and Discussion
It has been shown that the simple tetrapeptide Cys-Val-
Ile-Leu 1 has a similar potency in inhibiting FTase
(IC50=17 mM) and GGTase-I (IC50=13 mM), and that
HS
H
N
O
H
N
O
H
R
O
HS
H N
N
H N
2
N
H
OH
O
O
2
5: R = L-isoleucine: GGTI-2115.
6: R = L-norleucine: GGTI-2117.
1
2a: R1 = H, R2 = H: GGTI-279.
2b: R1 = H, R2 = CH3: GGTI-280.
3a: R1 = phenyl, R2 = H: GGTI-287.
3b: R1 = phenyl, R2 = CH3: GGTI-286.
4a: R1 = 1-naphthyl, R2 = H: GGTI-297.
4b: R1 = 1-naphthyl, R2 = CH3: GGTI-298.
R
1
H
COOR
2
H
N
HS
N
O
H N
2
Figure 1. Synthesized peptidomimetics as GGTase-I inhibitors.

Y. Qian et al./Bioorg. Med. Chem. 6 (1998) 293±299
295
Table 1 shows that compounds GGTI-279, GGTI-287
and GGTI-297 are more potent inhibitors of GGTase-I
than FTase, with selectivities between three- and four-
fold. Interestingly, when the leucine residue in GGTI-
287 was replaced with an isoleucine or norleucine, the
selectivity for GGTase-I over FTase decreased. The IC50
values for GGTI-279 are 135 nM for GGTase-I and
418 nM for FTase. Hydrophobic functionalization of
the aromatic spacer improved the inhibition potency
signi®cantly. Peptidomimetic GGTI-287 has IC50 values
of 7.3 nM and 21 nM for GGTase-I and FTase, respec-
tively. When compared to tetrapeptide 1, the inhibition
potency of the CAAX peptidomimetics was improved
between 100- and 2600-fold.
Figure 2. GGTase-I inhibition studies based on ability of pep-
tidomimetics to inhibit the transfer of [3H]geranylgeranyl to
Ha-Ras-CVLL (~, 2a; !, 3a; *, 4a; ^, 5).12
These results demonstrated the success of the hydro-
phobic spacer strategy in our peptidomimetic design.
Table 1 also showed that the leucine residue in our
peptidomimetics is critical for selectivity for GGTase-I.
However, this selectivity is not as high as for our pre-
viously reported FTase inhibitors. When the leucine
residue in GGTI-287 was replaced with a methionine,
the compound (FTI-276) became an FTase selective
inhibitor, with IC50 values for FTase and GGTase-I of
0.5 and 50 nM, respectively.13 Nonetheless, the selectiv-
ity for GGTase-I inhibition over FTase is still improved
when compared to the parent tetrapeptide 1. The methyl
esters GGTI-280 and GGTI-286 showed a 50-fold
decrease in GGTase-I inhibition potency when com-
pared to their free carboxylic acid forms GGTI-279 and
GGTI-287. This result suggested that a free carboxylic
acid is required for the potent inhibition of GGTase-I.
Interestingly, GGTI-280 and GGTI-286 showed the
reverse selectivity in the in vitro studies, suggesting that
the requirement of a free carboxylate is more stringent
for GGTase-I as compared to FTase. However, in
whole cells the methyl ester derivatives were more selec-
tive for GGTase-I over FTase.
reduction of the cysteine amide bond in 1 can increase
the selectivity for GGTase-I (IC50=1.9 nM) over FTase
(IC50=550 nM)15. In our earlier studies of FTase inhi-
bitors, we proposed that the central dipeptide AA in the
tetrapeptide CAAX might contribute to the hydro-
phobic interaction in the enzyme substrate binding site
and the amide bond might play no important role16.
Since GGTase-I is closely related to FTase and the two
enzymes share the same a subunit, we applied a similar
strategy to the design of GGTase-I inhibitors. The leu-
cine residue in tetrapeptide 1 was retained to maintain
the selectivity for GGTase-I. The dipeptide AA in 1 was
replaced with a series of aromatic hydrophobic spacers
and the cysteine amide bond was reduced to an amine
(compounds 2±4). Since CAAX tetrapeptides with an
isoleucine at the X position are also preferred substrates
for GGTase-I, compounds 5 and 6 were prepared where
an isoleucine and a norleucine replaced the leucine resi-
due. The structures of the synthesized inhibitors are
shown in Figure 1.
We next determined the ability of these peptidomimetics
to inhibit the processing of Rap1A and H-Ras in
whole cells. Treatment of NIH-3T3 cells that over-
express an activated H-Ras oncoprotein with GGTI-286
resulted in inhibition of processing of the exclusively
We ®rst determined the ability of peptidomimetics 2±6
to inhibit the incorporation of GGPP and FPP into the
geranylgeranylated protein H-Ras-CVLL (Figure 2) and
the farnesylated protein H-Ras-CVLS, respectively.
Table 1. Inhibition of GGTase-I and FTase in vitro by CAAX peptidomimetics
Peptidomimetics
GGTase-I inhibition IC50(nM)‹SEM(n)
FTase inhibition IC50(nM)‹SEM(n)
2a GGTI-279
2b GGTI-280
3a GGTI-287
3b GGTI-286
4a GGTI-297
5 GGTI-2115
6 GGTI-2117
135‹36(4)
6733‹2311(3)
7.3‹2.6(5)
240‹65(3)
56‹20(10)
32‹6(3)
418‹184(4)
897‹183(3)
21‹11(4)
183‹104(3)
203‹71(6)
56‹23(3)
41‹10(3)
47‹13(3)

296
Y. Qian et al./Bioorg. Med. Chem. 6 (1998) 293±299
geranylgeranylated protein Rap1A with an IC50 value of
2 mM.14 In contrast, inhibition of the processing of the
exclusively farnesylated protein H-Ras was less pro-
nounced with an IC50 of 10 mM. Substitution of the
phenyl group in GGTI-286 by a naphthyl group, as in
GGTI-298, increased the selectivity towards GGTase-I
over FTase in whole cells. GGTI-298 inhibited the pro-
cessing of Rap1A with an IC50 of 3 mM and had no
e€ect on the processing of H-Ras at concentrations as
high as 15 mM. At 20 mM of GGTI-298, H-Ras proces-
sing was inhibited by 10%. Higher concentrations of
GGTI-298 were toxic to cells. Even though GGTI-287 is
more potent in vitro than GGTI-297, in whole cells the
corresponding methyl ester derivatives GGTI-286 and
GGTI-298 are equipotent suggesting that GGTI-298 is
more cell permeable or a better substrate for intracel-
lular esterases. Furthermore, in whole cells GGTI-298 is
more selective than GGTI-286 even though in vitro the
corresponding free carboxylate derivatives, GGTI-297
and GGTI-287 have similar selectivities (Table 1). In
addition, GGTI-298 and GGTI-286 are not selective for
GGTase-I over FTase in vitro. These data coupled with
the selectivity results in whole cells suggest that the
methyl groups are cleaved in the cytosol to the active
free carboxylate forms.
Preparative HPLC was performed on a Waters 600E
controller and a Waters 490E multi-wavelength UV
detector with a 25Â10 cm Delta-Pak C-18 300 A car-
tridge column inside a Waters 25Â10 cm Radial Com-
pression Module. Solvents consisting of 20 to 80%
acetonitrile and 0.1% TFA in water were used with a
¯ow rate of 15 mL/min in 40 min.
N-4-[2(R)-Amino-3-mercaptopropyl]aminobenzoyl]-(L)-
leucine hydrochloride (2a). N-Boc-4-aminobenzoic acid
(from Boc protection of 4-aminobenzoic acid) was cou-
pled with l-leucine methyl ester to give N-Boc-4-ami-
nobenzoyl-leucine methyl ester (90%). mp 185±186 ꢀC.
1H NMR (CDCl3) d 7.72 (d, J=8.7 Hz, 2H), 7.43 (d,
J=8.7 Hz, 2H), 6.68 (s, 1H, Boc amide), 6.45 (d,
J=8.2 Hz, 1H, amide), 4.83 (ddd, J=5.0, 8.2 Hz, 1H,
Leu a H), 3.76 (s, 3H, OCH3), 1.63±1.77 (m, 3H, Leu),
1.52 (s, 9H, Boc), 0.97 (dd, J=6.0 Hz, 6H).
The above compound was deprotected to remove the Boc
group by gaseous hydrochloride in methanol and the
resulting hydrochloride salt was treated with N-Boc-S-
trityl-l-cysteinal (1.0 equiv) in the presence of sodium
cyanoborohydride (1.5 equiv) to give a reductive ami-
nation product (68%). mp 86 ꢀC (decomp). 1H NMR
(CDCl3) d 7.62 (d, J=8.6 Hz, 2H, aryl), 7.40±7.44
(m, 6H, trityl), 7.20±7.32 (m, 9H, trityl), 6.47 (d, J=8.7
Hz, 2H, aryl), 6.31 (d, J=8.2 Hz, 1H, amide), 4.84
(ddd, J=5.0, 8.2 Hz, 1H, Leu a H), 4.57 (br d, J=7.5
Hz, 1H, Boc amide), 4.15 (br, 1H, NH), 3.81 (m, 1H,
Cys a H), 3.75 (s, 3H, OCH3), 3.10 (t, J=5.9 Hz, 2H,
CH2N), 2.46 (d, J=5.2 Hz, 2H, CH2S), 1.68±1.75 (m,
2H, Leu), 1.63 (t, J=6.8 Hz, 1H, Leu), 1.43 (s, 9H,
Boc), 0.96 (dd, J=6.0 Hz, 6H); Anal. calcd for
Recently, we have investigated the e€ects of inhibiting
protein geranylgeranylation but not farnesylation on
receptor tyrosine kinase signaling and cell cycle pro-
gression. We found that treatment of NIH-3T3 cells
with GGTI-298 (10 mM) inhibited PDGF- and EGF-
stimulated tyrosine phosphorylation of their receptor
tyrosine kinases.17 Furthermore, GGTI-298 was also
extremely e€ective at blocking cells in the G1 phase of
the cell cycle at concentrations that had little e€ect on
protein farnesylation in whole cells.18 These results
clearly demonstrate the usefulness of CAAL peptidomi-
metics as powerful tools for determining the role of
geranylgeranylated proteins in early events in growth
factor signal transduction as well as later events in the
proliferation process, such as cell cycle progression.
.
C41H49O5N3S 0.4CH2Cl2: C 68.15, H 6.83, N 5.76;
Found C 68.16, H 6.96, N 5.81.
The fully protected compound from above was ®rst
treated with aqueous LiOH-THF and then deprotected
by TFA in the presence of triethylsilane. The resulting
TFA salt was converted to a hydrochloride salt by ®rst
dissolving the salt in 1.7 N HCl in acetic acid and then
adding 3 N HCl in ether to precipitate the product
(58%). mp 145 ꢀC (foaming). [a]25
60.7 (c 0.4,
D
Experimental
1
MeOH). H NMR (CD3OD) d 7.74 (d, J=8.5 Hz, 2H,
aryl), 6.77 (d, J=8.7 Hz, 2H, aryl), 4.63 (dd, J=4.5,
8.5 Hz, 1H, Leu a H), 3.43±3.59 (m, 3H, Cys a H,
CH2N), 2.94 (dd, J=4.6, 14.5 Hz, 1H, CH2SH), 2.83
(dd, J=5.4, 14.5 Hz, 1H, CH2SH), 1.64±1.82 (m, 3H,
Leu), 0.96 (dd, J=6.1 Hz, 6H); LRMS (EI) for
C16H25O3N3S 339 (M+, 15), 263 (80), 133 (100); HRMS
(EI) calcd 339.1616, obsd 339.1602.
Nuclear magnetic resonance spectra were acquired using
1
Bruker AM-300 series spectrometers (300 MHz for H,
75 MHz for 13C). Elemental analysis were performed by
Atlantic Microlabs, Inc., Norcross, GA. Optical rota-
tions [a]25 were measured using a Perkin±Elmer 241
D
polarimeter. All synthesized ®nal compounds (amine
hydrochloride or amine tri¯uoroacetate salts) were
checked for purity by analytical high pressure liquid
chromatography which was performed using a Rainin
HP controller and a Rainin UV-C detector with a
Rainin 250Â4.6 mm, 5 mm Microsorb C18 column.
N-4-[2(R)-Amino-3-mercaptopropyl]aminobenzoyl-(L)-
leucine methyl ester hydrochloride (2b). This compound
was prepared by ®rst treating the fully protected reduc-

Y. Qian et al./Bioorg. Med. Chem. 6 (1998) 293±299
297
tive amination product (406 mg) with mercuric chloride
and then bubbling hydrogen sul®de to precipitate the
mercuric sul®de. Final product was isolated as a hydro-
chloride salt (168 mg, 68%). mp 105 ꢀC (foaming). 1H
NMR (CD3OD) d 7.71 (d, J=8.7 Hz, 2H, aryl), 6.75 (d,
J=8.7 Hz, 2H, aryl), 4.64 (dd, J=4.7, 8.8 Hz, 1H, Leu a
H), 3.71 (s, 3H, OCH3), 3.42±3.58 (m, 3H, Cys a H,
CH2N), 2.95 (dd, J=4.5, 14.5 Hz, 1H, CH2SH), 2.81
(dd, J=5.3, 14.5 Hz, 1H, CH2SH), 1.65±1.81 (m, 3H,
Leu), 0.96 (dd, J=6.1 Hz, 6H); LRMS (EI) for
C17H27N3O3S 353 (M+, 20), 277 (90), 133 (100); HRMS
(EI) calcd 353.1773, obsd 353.1781. Anal. calcd for
1H, Leu a H), 3.41±3.57 (m, 3H, CH2N, Cys a H), 2.94
(dd, J=4.3, 14.5 Hz, 1H, CH2SH), 2.78 (dd, J=5.2,
14.5 Hz, 1H, CH2SH), 1.45 (t, J=6.7 Hz, 2H, Leu),
1.17±1.26 (m, 1H, Leu), 0.78±0.83 (t, J=8.5 Hz, 6H);
13CNMR (CD3OD) d 176.0, 173.0, 150.8, 143.4, 142.4,
131.1, 129.7, 129.4, 128.4, 125.9, 115.3, 111.9, 53.8, 52.5,
44.9, 41.3, 25.6, 25.2, 23.4, 21.7; Anal. calcd for
.
.
C22H29O3N3S CF3COOH H2O: C 52.65, H 5.85, N
7.67, S 5.85; Found C 52.60, H 5.90, N 7.62, S 5.79.
N-4-[2(R)-Amino-3-mercaptopropyl]amino-2-phenylbenz-
oyl-(L)-leucine methyl ester hydrochloride (3b, GGTI-
286). The fully protected precursor (described in the
preparation of GGTI-287) was deprotected by TFA in
the presence of triethylsilane. The resulting TFA salt
was dissolved in 1.7 N gaseous HCl in acetic acid and
then 3 N HCl in ether was added. The ®nal compound
was isolated as a hydrochloride salt. Purity (>95%) was
checked by analytical HPLC at both 220 nm and
.
C17H27O3N3S 1.5HCl: C 50.03, H 6.98, N 10.30, S 7.84;
Found: C 49.73, H 7.16, N 10.27, S 7.72.
N-4-[2(R)-Amino-3-mercaptopropyl]amino-2-phenylbenz-
oyl-(L)-leucine tri¯uoroacetate (3a, GGTI-287). The
coupling of 4-nitro-2-phenylbenzoic acid with l-leucine
methyl ester hydrochloride by using EDCI and HOBT
gave 4-nitro-2-phenylbenzoyl-leucine methyl ester
285 nm. mp 105 ꢀC (foaming). [a]25
22.7 (c 0.1,
D
(96%). mp 101±102 ꢀC. H NMR (CDCl3) d 8.24±8.26
MeOH). H NMR (CD3OD) d 7.42 (d, J=8.5 Hz, 1H,
aryl), 7.31±7.38 (m, 5H, phenyl), 6.76 (d, J=8.5 Hz, 1H,
aryl), 6.68 (s, 1H, aryl), 4.33 (t, J=7.8 Hz, 1H, Leu a
H), 3.67 (s, 3H, OCH3), 3.46±3.55 (m, 3H, Cys ꢀ H,
CH2N), 2.95 (dd, J=4.4, 14.5 Hz, 1H, CH2S), 2.81 (dd,
J=5.1, 14.5 Hz, 1H, CH2S), 1.44 (t, J=7.6 Hz, 2H,
Leu), 1.18±1.25 (m, 1H, Leu), 0.76±0.83 (dd, J=5.1,
6.6 Hz, 6H, 2CH3); 13C NMR (CD3OD) d 174.5, 173.2,
150.8, 143.5, 142.4, 131.1, 129.7, 129.4, 128.5, 125.7,
115.5, 112.1, 53.8, 52.7, 52.6, 44.9, 41.1, 25.6, 25.3, 23.3,
21.7; LRMS (EI) for C23H31O3N3S 429 (M, 12), 353
(80), 285 (55), 209 (95), 180 (100); HRMS (EI) calcd
429.2086, obsd 429.2088.
1
1
(m, 2H), 7.86 (d, J=8.7 Hz, 1H), 7.41±7.46 (m, 5H),
5.71 (d, J=7.4 Hz, 1H, amide), 4.57 (ddd, J=5.0,
7.4 Hz, 1H, Leu a H), 3.67 (s, 3H, OCH3), 1.37±1.46 (m,
1H, Leu), 1.08±1.25 (m, 2H, Leu), 0.78 (dd, J=5.0 Hz,
6H); 13C NMR (CDCl3) d 172.6, 167.1, 148.4, 141.2,
140.7, 137.8, 130.0, 128.9, 128.7, 128.5, 125.1, 122.2,
52.3, 51.0, 41.1, 24.4, 22.6, 21.7.
Reduction of the above nitro compound followed by
reductive amination with N-Boc-S-trityl-l-cysteinal
gave N-[4-[2(R)-tert-butoxycarbonylamino-3-(triphenyl-
methyl)-thiopropyl]amino-2-phenylbenzoyl]-leucine
methyl ester (61.3%). mp 82 ꢀC (decomp). 1H NMR
(CDCl3) d 7.68 (d, J=8.6 Hz, 1H), 7.33±7.41 (m, 11H),
7.17±7.29 (m, 9H), 6.50 (d, J=8.6 Hz, 1H, aryl), 6.31 (s,
1H, aryl), 5.43 (d, J=7.8 Hz, 1H, amide), 4.60 (d,
J=6.1 Hz, 1H, Boc amide), 4.47 (ddd, J=5.0, 8.2 Hz,
1H, Leu a H), 4.19 (br t, 1H, NH), 3.77 (br m, 1H, Cys
N-4-[2(R)-Amino-3-mercaptopropyl]amino-2-naphthyl-
benzoyl-(L)-leucine hydrochloride (4a, GGTI-297). Cou-
pling of 4-nitro-2-naphthylbenzoic acid (prepared from
the coupling of 2-bromo-4-nitrobenzoic acid methyl
ester with 1-naphthylboronic acid and the sponi®cation
of the ester) with l-leucine methyl ester gave 4-nitro-2-
naphthylbenzoyl-leucine methyl ester (93%). mp 144±
a H), 3.62 (s, 3H, OCH ), 3.09 (t, J=5.9 Hz, 2H,
3
CH2N), 2.45 (br m, 2H, CH2S), 1.40 (s, 9H, Boc), 1.27±
1.33 (m, 1H, Leu), 1.03±1.18(m, 2H, Leu), 0.75 (dd,
J=6.0 Hz, 6H); 13CNMR (CDCl3) d 173.2, 168.2, 155.6,
149.4, 144.4, 141.7, 141.2, 131.4, 129.5, 128.8, 128.5,
127.9, 127.6, 126.8, 122.7, 113.6, 111.3, 79.6, 67.1, 51.9,
50.9, 49.5, 47.1, 41.2, 34.3, 28.3, 24.4, 22.7, 21.8; Anal.
145 ꢀC. H NMR (CDCl3) d 8.34±8.39 (m, 1H), 8.25 (s,
1
1H), 8.18 (d, J=8.6 Hz, 0.5H), 8.02 (d, J=8.6 Hz,
0.5H), 7.91±8.00 (m, 2H), 7.62 (t, J=7.0 Hz, 0.5H),
7.48±7.58 (m, 3H), 7.41 (t, J=7.0 Hz, 1.5H), 5.71 (d,
J=7.9 Hz, 0.5H, amide), 5.60 (d, J=7.9 Hz, 0.5H,
amide), 4.29 (m, 1H, Leu a H), 3.57 (s, 1.5H, OCH3),
3.52 (s, 1.5H, OCH3), 1.05±1.11 (m, 0.5H), 0.88±0.97
(m, 0.5H), 0.69±0.78 (m, 0.5H), 0.41±0.59 (m, 7H), 0.19±
0.26 (m, 0.5H). The complex 1H NMR is due to the
restricted aryl±aryl bond rotation.
.
calcd for C47H53O5N3S 0.6CH2Cl2: C 69.48, H 6.59, N
5.11, S 3.89; Found C 69.45, H 6.73, N 5.26, S 3.87.
The above fully protected compound was ®rst depro-
tected by aqueous LiOH-THF then by TFA. The crude
TFA salt was puri®ed by preparative HPLC to give
GGTI-287 (56%). mp 115 ꢀC (decomp). [a]25
0.3, MeOH). H NMR (CD3OD) d 7.42 (d, J=8.5 Hz,
1H, aryl), 7.29±7.38 (m, 5H, phenyl), 6.73 (d, J=8.5 Hz,
1H, aryl), 6.66 (s, 1H, aryl), 4.32 (dd, J=4.3, 5.9 Hz,
1.8 (c
The above compound was ®rst reduced to the amine by
hydrogenation and then treated with N-Boc-S-trityl-l-
cysteinal in the presence of NaB(CN)H3 to give
N-[4-[2(R)-tert-butoxycarbonylamino-3-(triphenylmethyl)-
D
1

298
Y. Qian et al./Bioorg. Med. Chem. 6 (1998) 293±299
thiopropyl]amino-2-naphthylbenzoyl]-leucine
methyl
2.93 (dd, J=4.1, 14.8 Hz, 1H, CH2S), 2.80 (dd, J=4.3,
14.4 Hz, 1H, CH2S), 0.96±1.26 (m, 1.5H), 0.72±0.82 (m,
1H), 0.59 (d, 1.2H), 0.55 (d, 2.8H), 0.34 (d, 1.9H), 0.23±
0.32 (m, 0.6H); LRMS (EI) 479 (M+, 15), 403 (50), 259
(100); HRMS (EI) calcd for C27H33O3N3S 479.2243,
obsd 479.2237. Anal. calcd for C27H33O3N3S.1.8HCl: C
59.48, H 6.38, N 7.71; Found C 59.15, H 6.59, N 7.54.
ester (75%). mp 91 ꢀC (decomp). 1H NMR (CDCl3) d
7.85±8.00 (m, 3H), 7.47±7.67 (m, 4H), 7.39±7.43 (m,
7H), 7.14±7.37 (m, 9H), 6.61 (d, J=8.6 Hz, 1H, aryl),
6.32 (s, 1H, aryl), 5.46 (d, J=7.6 Hz, 0.6H, amide), 5.36
(d, J=7.6 Hz, 0.4H, amide), 4.55 (d, J=7.2 Hz, 1H, Boc
amide), 4.20±4.27 (m, 2H, Leu a H, NH), 3.76 (br, 1H,
OCH3), 3.56 (s, 2H, OCH3), 3.38 (br, 1H, Cys a H), 3.06
(t, J=5.9 Hz, 2H, CH2N), 2.43 (m, 2H, CH2S), 1.36±
1.43 (m, 9H, Boc), 0.81±1.03 (m, 1H), 0.55±0.67 (m,
2.8H), 0.36±0.45 (m, 4.7H), 0.00±0.09 (m, 0.6H); Anal.
N-4-[2(R)-Amino-3-mercaptopropyl]amino-2-phenylbenz-
oyl-(L)-isoleucine tri¯uoroacetate (5). This compound
was prepared in the same way as the preparation of
GGTI-287 except that the leucine was replaced with the
isoleucine. The reductive amination gave the fully pro-
.
calcd for C51H55O5N3S 0.6CH2Cl2: C 71.09, H 6.44, N
4.81; Found C 70.71, H 6.51, N 4.75.
1
tected compound (62%). H NMR (CDCl3) d 7.46 (d,
The above fully protected compound was deprotected as
described before. The ®nal product GGTI-297 was iso-
lated as a hydrochloride salt (77%). mp 160±162 ꢀC
J=8.5 Hz, 1 H, aryl), 7.22±7.16 (m, 11 H), 7.10±6.97 (m,
9 H), 6.32 (d, J=8.6 Hz, 1 H, aryl), 6.12 (s, 1 H, aryl),
5.42 (d, J=8.1 Hz, 1 H, amide NH), 4.33 (m, 1 H, Boc-
amide), 4.25 (dd, J=4.9 Hz, 1 H, Ile-aH), 3.55 (br m, 1
H, cyst-aH), 3.38 (s, 3 H, OCH3), 2.87 (d, J=6.2 Hz, 2
H, CH2N), 2.20 (br m, 3 H, NH and CH2S), 1.23 (m, 1
H, Ile-CH), 1.15 (s, 9 H, Boc), 1.05 (m, 1 H, Ile), 0.92
(m, 1 H, Ile), 0.54 (m, 3 H, Ile-CH3), 0.38 (d, J=6.7 Hz,
3 H, Ile-CH3). The deprotection of this compound gave
the ®nal product (47%). mp 101±105 ꢀC. 1H NMR
(DMSO-d6) d 8.01 (br s, 3 H, NH3), 7.61 (d, J=8.6 Hz,
1 H, aryl), 7.32 (m, 6 H, phenyl and amide), 6.66 (d,
J=8.2 Hz, 1 H, aryl), 6.57 (s, 1 H, aryl), 6.27 (br s, 1 H,
aryl NH), 4.07 (t, J=3.3 Hz, 1 H, Ile-aH), 2.89±2.79 (m,
4 H, CH2N and CH2S), 1.74 (m, 1 H, Ile), 1.26 (m, 1 H,
Ile), 0.89 (m, 1 H, Ile), 0.81 (m, 6 H, Ile); 13C NMR
(DMSO-d6) d 175.3, 170.9, 152.5, 147.4, 139.1, 125.4,
124.7, 123.5, 122.4, 121.4, 110.3, 109.3, 52.0, 51.4, 50.1,
28.7, 27.2, 21.4, 14.3, 13.7.
(decomp). [a]25 +18.9 (c 0.8, MeOH). 1H NMR
D
(CD3OD) d 7.86±7.92 (m, 2H), 7.73 (d, J=8.5 Hz,
0.6H), 7.35±7.67 (m, 5.4H), 6.89 (m, 1H), 6.67 (m, 1H),
4.06 (dd, J=4.6, 5.5 Hz, 0.4H), 4.00 (dd, J=4.6, 5.5 Hz,
0.6H), 3.71±3.82 (m, 0.6H), 3.44±3.58 (m, 2.4H), 2.73±
2.96 (m, 2H), 1.14±1.29 (m, 0.5H), 0.97±1.10 (m, 1H),
0.69±0.78 (m, 1H), 0.59 (d, 1.2H), 0.44±0.49 (m, 2.8H),
0.31 (m, 1.9H), 0.19±0.25 (m, 0.5H); LRMS for
C26H31O3N3S 465 (M+, 8), 389 (65), 259 (90), 246 (100);
HRMS calcd 465.2086, obsd 465.2083.
N-4-[2(R)-Amino-3-mercaptopropyl]amino-2-naphthyl-
benzoyl-(L)-leucine methyl ester hydrochloride (4b,
GGTI-298). The fully protected product from reductive
amination reaction was deprotected by TFA. The TFA
salt was extracted with concentrated NaHCO3 and
methylene chloride. After evaporation of solvents, a
waxy product was obtained (85%). 1H NMR (CDCl3) d
7.98 (d, J=8.5 Hz, 0.6H), 7.84±7.90 (m, 2.4H), 7.65 (d,
J=8.5 Hz, 0.4H), 7.43±7.58 (m, 3.6H), 7.34±7.39 (m,
1H), 6.72 (m, 1H), 6.45 (m, 1H), 5.46 (d, J=7.8 Hz,
0.6H, amide), 5.40 (d, J=7.7 Hz, 0.4H, amide), 4.64 (br,
1H, aryl NH), 4.23 (m, 1H, Leu a H), 3.54 (s, 2H,
OCH3), 3.30 (s, 1H, OCH3), 3.25 (m, 1H, Cys a H),
2.97±3.06 (m, 2H, CH2N), 2.67 (dd, J=4.7, 13.1 Hz, 1H,
CH2S), 2.47 (dd, J=6.5, 13.2 Hz, 1H, CH2S), 1.45±1.65
(br, 2H, NH2), 0.81±1.03 (m, 1H), 0.54±0.67 (m, 3H),
0.36±0.39 (m, 4.3H), 0.00±0.10 (m, 0.7H).
N-4-[2(R)-Amino-3-mercaptopropyl]amino-2-phenylbenz-
oyl-(L)-norleucine tri¯uoroacetate (6). This compound
was prepared in the same way as the preparation of
GGTI-287 except that the leucine was replaced with the
norleucine. The reductive amination gave the fully pro-
1
tected compound (58%). H NMR (CDCl3) d 7.67 (d,
J=8.2 Hz, 1 H, aryl), 7.34±7.30 (m, 11 H, aryl), 7.28±
7.22 (m, 9 H, aryl), 6.93 (d, J=8.1 Hz, 1 H, aryl), 6.45
(s, 1 H, aryl), 5.59 (d, J=8.4 Hz, 1 H, amide NH), 4.41
(br, 1 H, Boc-amide), 4.19 (ddd, J=4.7 Hz, 8.1 Hz, 1 H,
Nleu-a H), 3.37 (s, 3 H, OCH3), 3.21 (br m, 2 H,
cysteine a H and aryl NH), 3.01 (d, J=6.2 Hz, 2 H,
CH2N), 2.71 (m, 2 H, CH2S), 1.97±1.83 (m, 2 H, Nleu
CH2), 1.36±1.27 (m, 4 H, Nleu-CH2), 0.97 (t, J=8.0 Hz,
3 H, CH3). The deprotection of this compound gave the
The above free amine was converted to a hydrochloride
salt by ®rst dissolving the amine in methylene chloride
and then addition of 3 N gaseous HCl in ether to give
compound GGTI-298 (72%). mp 103 ꢀC (foaming).
®nal product (51%). mp 95±98 ꢀC. H NMR (DMSO-
1
[a]25 +11.1 (c 0.4, MeOH). 1H NMR (CD3OD) d
d6) d 12.37 (br s, 1 H, COOH), 8.15 (br s, 3 H, NH3),
7.86 (d, J=7.5 Hz, 1 H, aryl), 7.86±7.29 (m, 6 H, phenyl
and amide), 6.67 (d, J=8.4 Hz, 1 H, aryl), 6.58 (s, 1 H,
aryl), 6.31 (br s, 1 H, aryl NH), 4.09 (ddd, J=8.4 Hz, 3.9
Hz, 1 H, Nleu a H), 3.39 (br s, 2 H, CH2N), 2.92 (m, 1
H, Cyst-aH), 2.78 (m, 2 H, CH2SH), 1.50±1.59 (m, 2 H,
D
7.87±7.93 (m, 2H), 7.70 (d, J=8.5 Hz, 0.6H), 7.62 (m,
1.2H), 7.45±7.59 (m, 3H), 7.34±7.42 (m, 1.2H), 6.88 (m,
1H), 6.67 (m, 1H), 4.11 (dd, J=4.8, 5.4 Hz, 0.4H, Leu a
H), 4.00 (dd, J=4.6, 5.7 Hz, 0.6H, Leu a H), 3.71±3.78
(m, 0.6H), 3.46±3.59 (m, 5.4H, OCH3, CH2N, Cys a H),

Y. Qian et al./Bioorg. Med. Chem. 6 (1998) 293±299
299
NLeu-CH2), 1.20±1.08 (m, 4 H, NLeu-CH2), 0.81 (t,
J=7.5 Hz, 3 H, Nleu-CH3); 13C NMR (DMSO-d6) d
177.0, 171.0, 148.8, 141.0, 138.7, 129.6, 128.2, 127.7,
126.7, 124.3, 113.4, 110.1, 51.9, 51.7, 43.1. 30.2, 27.2,
24.1, 21.4, 12.5.
8. Cox, A. D.; Der, C. J. Critical Reviews in Oncogenesis 1992,
3, 365.
9. Der, C. J.; Cox, A. D. Cancer Cells 1991, 3, 331.
10. Olson, M. F.; Ashworth, A.; Hall, A. Science 1995, 269,
1270.
11. Macchia, M.; Jannitti, N.; Gervas, G.; Danesi, R. J. Med.
Chem. 1996, 39, 1352.
Acknowledgements
12. Vogt, A.; Qian, Y.; Blaskovich, M. A.; Fossum, R. D.;
Hamilton, A. D.; Sebti, S. M. J. Biol. Chem. 1995, 270, 660.
We thank the National Institutes of Health (CA 67771)
for ®nancial support of this work.
13. Lerner, E. C.; Qian, Y.; Blaskovich, M. A.; Fossum, R. D.;
Vogt, A.; Sun, J.; Cox, A. D.; Der, C. J.; Hamilton, A. D.;
Sebti, S. M. J. Biol. Chem. 1995, 270, 26802.
References
14. Lerner, E. C.; Qian, Y.; Hamilton, A. D.; Sebti, S. M. J.
Biol. Chem. 1995, 270, 26770.
1. Casey, P. J. J. Lipid Res. 1992, 33, 1731.
15. Graham, S. L.; deSolms, S. J.; Giuliani, E. A.; Kohl, N. E.;
Mosser, S. D.; Oli€, A. I.; Pompliano, D. L.; Rands, E.; Bre-
slin, M. J.; Deana, A. A.; Garsky, V. M.; Scholz, T. H.; Gibbs,
J. B.; Smith, R. L. J. Med. Chem. 1994, 37, 725.
2. Reiss, Y.; Seabra, M. C.; Goldstein, J. L.; Brown, M. S.
Methods: A Companion to Methods in Enzymology 1990, 1, 241.
3. Moomaw, J. F.; Casey, P. J. J. Biol. Chem. 1992, 267, 17438.
4. Yokoyama, K.; Gelb, M. H. J. Biol. Chem. 1993, 268, 4055.
16. Nigam, M.; Seong, C. M.; Qian, Y.; Blaskovich, M. A.;
Hamilton, A. D.; Sebti, S. M. J. Biol. Chem. 1993, 268, 20695.
5. Casey, P. J.; Thissen, J. A.; Moomaw, J. F. Proc. Natl.
Acad. Sci. U.S.A. 1991, 88, 8631.
17. McGuire, T.; Qian, Y.; Vogt, A.; Hamilton, A. D.; Sebti, S.
M. J. Biol. Chem. 1996, 271, 27402.
6. Casey, P. J.; Seabra, M. C. J. Biol. Chem. 1996, 271, 5289.
18. Vogt, A.; Qian, Y.; McGuire, T.; Hamilton, A. D.; Sebti, S.
M. Oncogene 1996, 13, 1991.
7. Buss, J. E.; Marsters, J. C. Chemistry and Biology 1995, 2,
787.