Inhibitors of farnesyl protein transferase. 4-Amido, 4-carbamoyl, and 4- carboxamido derivatives of 1-(8-chloro-6,11-dihydro-5H-benzo[5,6]- cyclohepta[1,2-b]pyridin-11-yl)piperazine and 1-(3-bromo-8-chloro-6,11- dihydro-5H-benzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)piperazine

Article abstract of DOI:10.1021/jm970462w

The synthesis of a variety of novel 4-amido, 4-carbamoyl and 4- carboxamido derivatives of 1-(8-chloro-6,11-dihydro-5H- benzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)piperazine to explore the SAR of of this series of FPT inhibitors is described. This resulted in the synthesis of the 4-and 3-pyridylacetyl analogues 45a and 50a, respectively, both of which were orally active but were found to be rapidly metabolized in vivo. Identification of the principal metabolites led to the synthesis of a variety of new compounds that would be less readily metabolized, the most interesting of which were the 3- and 4-pyridylacetyl N-oxides 80a and 83a. Novel replacements for the pyridylacetyl moiety were also sought, and this resulted in the discovery of the 4-N-methyl and 4-N-carboxamidopiperidinylacetyl derivatives 135a and 160a, respectively. All of these derivatives exhibited greatly improved pharmacokinetics. The synthesis of the corresponding 3- bromo analogues resulted in the discovery of the 4-pyridylacetyl N-oxides 83b (±) and 85b [11S(-)] and the 4-carboxamidopiperidinylacetamido derivative 160b (±), all of which exhibited potent FPT inhibition in vitro. All three showed excellent oral bioavailability in vivo in nude mice and cynomolgus monkeys and exhibited excellent antitumor efficacy against a series of tumor cell lines when dosed orally in nude mice.

Full text of DOI:10.1021/jm970462w

877
J . Med. Chem. 1998, 41, 877-893
In h ibitor s of F a r n esyl P r otein Tr a n sfer a se. 4-Am id o, 4-Ca r ba m oyl, a n d
4-Ca r boxa m id o Der iva tives of 1-(8-Ch lor o-6,11-d ih yd r o-5H-ben zo[5,6]-
cycloh ep ta [1,2-b]p yr id in -11-yl)p ip er a zin e a n d 1-(3-Br om o-8-ch lor o-6,11-
d ih yd r o-5H-ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)p ip er a zin e1
Alan K. Mallams,* Randall R. Rossman, Ronald J . Doll, Viyyoor M. Girijavallabhan, Ashit K. Ganguly,
J oanne Petrin, Lynn Wang, Robert Patton, W. Robert Bishop, Donna M. Carr, Paul Kirschmeier,
J oseph J . Catino, Matthew S. Bryant, Kwang-J ong Chen, Walter A. Korfmacher, Cymbelene Nardo,
Shiyong Wang, Amin A. Nomeir, Chin-Chung Lin, Zujun Li, J ianping Chen, Suining Lee, J anet Dell,
Philip Lipari, Michael Malkowski, Bodan Yaremko, Ivan King, and Ming Liu
Antiinfectives and Tumor Biology Research, Schering-Plough Research Institute, Kenilworth, New J ersey 07033-0539
Received J uly 15, 1997
The synthesis of a variety of novel 4-amido, 4-carbamoyl and 4-carboxamido derivatives of
1-(8-chloro-6,11-dihydro-5H-benzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)piperazine to explore the
SAR of of this series of FPT inhibitors is described. This resulted in the synthesis of the 4-
and 3-pyridylacetyl analogues 45a and 50a , respectively, both of which were orally active but
were found to be rapidly metabolized in vivo. Identification of the principal metabolites led to
the synthesis of a variety of new compounds that would be less readily metabolized, the most
interesting of which were the 3- and 4-pyridylacetyl N-oxides 80a and 83a . Novel replacements
for the pyridylacetyl moiety were also sought, and this resulted in the discovery of the 4-N-
methyl and 4-N-carboxamidopiperidinylacetyl derivatives 135a and 160a , respectively. All of
these derivatives exhibited greatly improved pharmacokinetics. The synthesis of the corre-
sponding 3-bromo analogues resulted in the discovery of the 4-pyridylacetyl N-oxides 83b (()
and 85b [11S(-)] and the 4-carboxamidopiperidinylacetamido derivative 160b ((), all of which
exhibited potent FPT inhibition in vitro. All three showed excellent oral bioavailability in
vivo in nude mice and cynomolgus monkeys and exhibited excellent antitumor efficacy against
a series of tumor cell lines when dosed orally in nude mice.
In tr od u ction
and pyridyl amides were prepared has been carried out
in these laboratories.8 Additional novel 11-piperidinyl9
and 11-piperazinyl analogues1,9 have been also been
synthesized and found to be potent inhibitors of FPT,
and we wish here to describe the latter series.
Mutations in the ras oncogene have been detected in
a wide variety of human tumors, with the highest
incidence being observed in cancers of the pancreas (ca.
90%), colon (ca. 50%), and lung (ca. 30%).2 The impor-
tant role played by Ras protein in the signal transduc-
tion process involved in cell division is well recognized.2,3
Inhibition of farnesylation of the Ras protein, a key step
in the posttranslational modification of this protein,4,5
continues to be the subject of intense interest as a source
of potential antitumor agents. This has led to the
synthesis of several novel groups of farnesyl protein
transferase (FPT) inhibitors which have recently been
reviewed,6 most of which are peptidic or peptidomimetic
in nature. Most of these are CaaX mimetics, while
others are farnesyl diphosphate analogues, bisubstrate
analogues, or natural products derived from screening
programs. Many of these contain free amino and
sulfhydryl groups in the molecule, which are potentially
metabolically labile.
Recently, this laboratory reported the discovery of a
novel tricyclic FPT inhibitor 1,7 containing the 8-chloro-
5,6-dihydro-11H-benzo[5,6]cyclohepta[1,2-b]pyridine moi-
ety, which is of considerable interest as it is nonpeptidic,
contains no sulfhydryl group, is selective for FPT over
geranylgeranyl protein transferase (GGPT), and com-
petes kinetically with the Ras protein in binding to FPT.
A structure-activity relationship (SAR) study of these
11-piperidinylidene derivatives in which a series of aryl
The synthesis of a variety of 4-amido, 4-carbamoyl,
and 4-carboxamido derivatives of both 1-(8-chloro-6,11-
dihydro-5H-benzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)-
piperazine (5a ) and of 1-(3-bromo-8-chloro-6,11-dihydro-
5H-benzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)pipera-
zine (5b) has been carried out. This led initially to the
preparation of the 4- and 3-pyridylacetamido derivatives
45a and 50a (Table 1). Both were found to be orally
active, but unfortunately they were extensively metabo-
lized. This information, coupled with the knowledge of
the structures of two of the principal metabolites, led
us to synthesize a variety of analogues in this series
having reduced potential for metabolism, the most
S0022-2623(97)00462-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/19/1998

878 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Mallams et al.
Ta ble 1. Structures, Methods of Synthesis, and Antitumor Activity for the 4-Amido, 4-Carbamoyl, 4-Carboxamido, and 4-Alkyl
Derivatives Derived from the Intermediates 5a (Structure A), 46a (Structure B), 48a (Structure C), 90 (Structure D), 98a (Structure
E), 106 (Structure J ), 5b (Structure F), 46b (Structure G), 48b (Structure H), 98b (Structure I), 112 (Structure K), and 114
(Structure L)

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 879
Ta ble 1. (Continued)

880 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Ta ble 1. (Continued)
Mallams et al.
Sch em e 1a
a
(a) NaBH4, MeOH, 0 °C, 2.5 h, 0-25 °C, 1 h; (b) SOCl2, toluene, 0 °C, 2.5 h, 0-25 °C, 0.5 h; (c) piperazine, THF, 25 °C, 19 h.
interesting of which were the N-oxides 80a and 83a .
Novel replacements for the pyridylacetyl moiety were
also sought, and this resulted in the discovery of the
4-N-methyl- and 4-N-carboxamidopiperidinylacetyl de-
rivatives 135a and 160a , respectively. All of the above
derivatives exhibited greatly improved pharmacokinet-
ics. The discovery in these laboratories that introduc-
tion of a 3-bromo substituent resulted in more potent
FPT inhibitors in the 11-ylidene series10 prompted us
to prepare the more potent 4-amido, 4-carbamoyl, and
4-carboxamido derivatives described above, in the 3-bro-
mo series starting from 5b. This resulted in the
synthesis of the 4-pyridylacetyl N-oxides 83b (() and
85b [11S(-)] and the 4-carboxamidopiperidinylaceta-
mido derivative 160b ((), all of which exhibited more
potent FPT inhibition in vitro. All three showed excel-
lent oral bioavailability in vivo in nude mice and
cynomolgus monkeys and exhibited excellent antitumor
efficacy against a series of tumor cell lines when dosed
orally in nude mice.
Ch em istr y
4-Am id o Der iva tives. The tricyclic piperazine de-
rivative 5a was prepared (Scheme 1) essentially as
described previously.11-13 The 3-bromotricyclic ketone14
was synthesized by an alternative route from loratadine,
by a sequence involving nitration, reduction, and bro-
mination developed in these laborartories,10 and was
converted into the corresponding piperazine derivative
5b (Scheme 1). In general the amides described in
Table 1 were prepared by reacting either 5a or 5b with
the appropriate acid in the presence of 1-(3-(dimethy-
lamino)propyl)-3-ethylcarbodiimide hydrochloride, 1-hy-
droxybenzotriazole, and 4-methylmorpholine in DMF
(method A). In this manner, a variety of glycine and
N-substituted analogues 6-28, phenylalanine and N-

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 881
substituted analogues 29-31, and tyrosine and N-
substituted analogues 32-38 were prepared. A series
of aryl 42, 44, arylacetyl 39-41 and 43, and pyridy-
lacetyl analogues 45a ,b, 50a ,b, and 54 were also
prepared (method A). The resolved 11R(+)- and 11S(-
)-piperazine 46a and 48a , respectively, were prepared
as described earlier11 and used to synthesize the 11R(+)
analogues 47 and 51, as well as the 11S(-)-analogues
49 and 52. Reduction of the amide 50a with LiAlH4
afforded the pyridylethyl derivative 53. A series of R,R-
dimethylphenylacetyl 55, R,R-dimethylpyridylacetyl 73,
74a ,b, and R-methylpyridylacetyl 75 analogues were
also prepared using method A. The R-methyl-15 and R,R-
dimethylpyridylacetic16 acids were prepared as de-
scribed in Scheme 6 (Supporting Information). The
3-pyridylacrylicamide 76 was also prepared. The 5- and
7-carboxyquinolines17 were also used to prepare the
quinoline derivatives 77a ,b and 78. The quinolines
77a ,b and 78 constitute rigid analogues of 50a ,b,
respectively, having no possibility of metabolism at the
carbon adjacent to the amide carbonyl.
amides 99-101. The 4,8-dichloropiperazine 106 was
synthesized from the tricyclic N1-oxide of 2a by reacting
it with POCl3 as described in Scheme 9 (Supporting
Information). Reaction of 106 with the appropriate
pyridylacetic acid gave 107 and 108. The 2S-n-butyl-
and 2S-methoxyethyl-substituted piperazines 112 and
114, respectively, were synthesized21-23 as shown in
Scheme 6. The latter were reacted with 4-pyridylacetic
acid N-oxide to give 115 and 116, respectively.
20
The identification of the principal metabolites 79 and
80a formed when 50a was dosed in mice prompted us
to synthesize the 3- and 4-pyridylacetic acid N1-oxides
69-72 (Scheme 6). Using method A, the latter were
reacted with 5a ,b to give the N-oxides 80a ,b, 83a ,b,
93a ,b, and 94a ,b, with 46a ,b to give 81 and 84a ,b and
with 48a ,b to give 82 and 85a ,b, respectively. The rigid
quinoline N-oxide derivative 86 corresponding to 80b
was also prepared using similar techniques.17 The
tricyclicpiperazine N1-oxide 90 was also prepared by the
method outlined in Scheme 7 (Supporting Information)
and reacted with the appropriate pyridylacetic acids
using method A to give the mono- and bis-N-oxides 91
and 92, respectively.
The search for a viable alternative to the pyridylacetyl
moiety that would not be readily metabolized led us to
synthesize a series of piperidine derivatives described
below. Several compounds without the methylene
spacer were synthesized from isonipecotic acid, which
was converted into the derivatives 117a 24 and 117b,25,26
which were in turn reacted with 5a to give 118 and 128,
respectively. Deprotection of 128 afforded 129. The
11R(+)- and 11S(-)-enantiomers 119 and 120 were also
prepared from 46a and 48a , respectively. Nipecotic acid
was converted into the derivative 121a ,27 which was
reacted with 5a to give 122. N-Methylnipecotic acid
To prepare the 5,6-ene analogues, the ketones 2a and
18,19
2b were reacted with SeO2
to give the 5,6-enes,
which were converted into the corresponding pipera-
zines 98a ,b as described in Scheme 8 (Supporting
Information). The latter were used to prepare the

882 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Mallams et al.
Sch em e 2a
121b28 was prepared either by reduction of 126 or by
reductive formylation of nipecotic acid, and it was
reacted with 5a to give 127. D,L-Pipecolinic acid was
converted into the N-acetate 123,29 and the latter was
reacted with 5a to give the diastereoisomers 124 and
125. Reaction of 129 with ethyl chloroformate afforded
130.
To mimic the pyridylacetic acid derivatives, a series
of N-methyl, N-acetyl,30 and N-Boc31 4- and 3-piperidi-
nylacetic acids were synthesized as described in Scheme
11 (Supporting Information). Using method A the
following amides were prepared.The N-methyl acids
were reacted with 5a ,b, 46a ,b, 48a ,b, 98a , and 106 to
give the amides 135a ,b-143. The N-acetyl acids were
reacted with 5a ,b, 46a , and 48a to give 144a -147,
while 144b was prepared by direct acetylation of 150b
as described in method D. The N-Boc acids were reacted
with 5a ,b, 46b, 48b, 98b, and 114 to give the corre-
sponding N-Boc amides including 148a ,b and 149.
These were deprotected using method B to give the
amines 150a ,b-155. Reaction of the amine 150b with
diphenyl carbonate (method E) afforded 156, while
reaction of 150a and 155 with MeCOOCl (method C)
gave 157 and 158. Reductive amination of 150a (method
F) afforded the N-ethyl derivative 159.
a
(a) NH2CH2CH2NHAc, MeOH, 28.8% HCl in EtOH, 3 Å sieves,
NaBH3CN, 60 °C, 236 h; (b) 85% hydrazine hydrate, MeOH, 90
°C, 13 days; (c) method A.
Sch em e 3a
We next prepared a series of N-carboxamido deriva-
tives 160a ,b-166 from the amines 150a ,b-155 and the
piperidine derived from 112, using method G. The
amines 150a ,b and 155 were also reacted with a variety
of isocyanates using method H to give the N-alkylcar-
boxamido derivatives 167a ,b-171a ,b. The amine 150a
was also reacted with MeNCS using method I to give
thiocarboxamido derivative 172. Utilizing method J , the
amines 150a ,b were converted using the appropriate
N-Boc amino acids into a series of amide derivatives
173-176 after deprotection. The amine 150b was also
converted into the cysteinyl amide 177 using methods
K and L.
To prepare rigid analogues of 163, 5-carboxyquino-
line17 was reduced to the 1,2,3,4-tetrahydro derivative32
and to the perhydro derivative,32 and each was con-
verted into the N-Boc derivative and reacted with 5b
to give the N-Boc amides. The latter were deprotected
(method B) to give 178 and 179, which were reacted
with TMSNCO (method G) to give 180 and 181, respec-
tively.
a
(a) SOCl2, toluene, 0 °C, 2 h, 0-25 °C, 1 h; (b) piperazine, THF,
25 °C, 18 h; (c) method A.
An analogue of 45a in which the central piperazine
ring has been replaced by an ethylenediamine moiety
was prepared (Scheme 2) by reductive amination of 2a
with N-acetylethylenediamine to give 190. Hydrazi-
nolysis of the latter gave 191, which was reacted with
4-pyridylacetic acid (method A) to give 192. An ana-
logue of 50a lacking the 5,6-ethano bridge was also
prepared (Scheme 3) by converting the alcohol 193 into
the chloro derivative 194. The latter was reacted with
piperazine to give 195, which was in turn reacted with
3-pyridylacetic acid (method A) to give 196. The regio
isomer 201 was prepared in a similar manner from the
ketone 197 as shown in Scheme 4. An analogue of 196,
lacking the pyridyl ring, was also prepared from mono-
N-Boc-piperazine 202 by reacting it with 3-pyridylacetic
acid to give 203. Deprotection of the latter gave 204
which was acylated to give 205 as described in Scheme
5.
4-Ca r ba m oyl Der iva tives. We prepared a selected
group of carbamoyl derivatives (Table 1) in order to
determine what effect replacement of the methylene
spacer with an oxygen atom would have on the FPT
activity of these inhibitors. Reaction of 5a with an
excess of phosgene afforded 206, which was reacted with
the appropriate phenol to give 207 and 214a (method
M). Using method N, the appropriate chloroformates
were reacted with 5a , or 98a , to give 208, 209, and 218.
The 3-pyridylchloroformate 210 was prepared from the
phenol (method N) and reacted with 5a , 46a , and 48a
to give the carbamoyl derivatives 211-213. The 4-N-
methylpiperidinyl chloroformate 21533 was reacted with
5a ,b to give 216a ,b (method N). An alternative method
for preparing these carbamoyl derivatives involved
Using method A the amine 5b was converted into the
N-Boc-S-Tr-cysteinyl derivative, which was in turn
deprotected to give 184. The N-Boc-cystinyl derivative
was prepared in a similar manner and deprotected to
give 186. The reduced cysteinyl analogues 187-189
were also prepared using simliar procedures to those
described by others.22

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 883
Sch em e 4a
a
(a) NaBH4, MeOH, 0 °C, 1 h; (b) SOCl2, toluene, 0 °C, 3.5 h; (c) piperazine, THF, CH2Cl2, 25 °C, 73 h; (d) method A.
Sch em e 5a
â-chlorocatecholborane as described in method W, mod-
est yields of the carboxamides 251b and 252a ,b were
obtained. Improved yields were obtained when the acyl
azides 235, 238, 241, and 242, prepared as described
in Schemes 13 and 14 respectively35,36 (Supporting
Information) were converted in situ into the isocyanates
236, 239, 243, and 244, which were then reacted with
5a ,b (method U) to give the carboxamides 249a ,b-
252a ,b. The carboxamide 249a was also prepared by
reacting 248 (Scheme 15, Supporting Information) with
the imidazole derivative 219a in the presence of NaH.
When 219a was reacted with 3-amino-N-methylpiperi-
dine and DMAP in DMF (method X), only the carboxa-
mide 253 was isolated, due to the Me2NH present in
the reaction medium. When the reaction was carried
out in the absence of DMF under fusion conditions
(method Y), only the dimer 254 was obtained. The
N-methylpiperazinylcarboxamide 255 was also prepared
by reaction of 206 with mono-N-methylpiperazine as
described in method Z.
a
(a) Method 1; (b) 10% concentrated H2SO4-dioxane (v/v), 0
°C, 0.5 h; (c) 4-chlorobenzoyl chloride, THF, 25 °C, 1 h.
converting 5a ,b into the carbonylimidazole derivatives
219a ,b (Table 1) by reaction with carbonyldiimidazole
(method O). Reaction of 219a ,b with 3-hydroxypyridine
N-oxide or with 3-hydroxy-N-methylpiperidine, in the
presence of ZnBr234 (method P), gave 214b and 217a ,b,
respectively. The carbamoyl derivative 217a was also
prepared by reacting 219a with 3-hydroxy-N-methylpi-
peridine and NaH (method Q).
Biology
F P T a n d GGP T Activity in Vitr o. The in vitro
enzyme assays for measuring FPT and GGPT activity
have been described earlier.7 The results of these
assays for the 4-amido, 4-carbamoyl, 4-carboxamido, and
miscellaneous derivatives are listed in Table 1. The
tricyclicpiperazine 5a lacking any substituents at the
4-position of the piperazine was inactive in the FPT
assay. Of the amides prepared from glycine, phenyla-
lanine, and tyrosine and their N-substituted analogues,
only 6 showed any activity. The remaining amides in
this group were essentially inactive. The amide 6
showed good selectivity for FPT versus GGPT, the IC50
for the latter being >40 µM. The amides 39-41 and
43 derived from phenylacetic acid and substituted
derivatives thereof were moderately active, while the
amides 42 and 44 derived from substituted benzoic acids
were inactive. This confirmed the importance of the
methylene spacer adjacent to the amide carbonyl group
in the piperazine series. Similar observations had been
made in the 11-piperidinylidene series.8,10
4-Ca r boxa m id o Der iva tives. A selected group of
carboxamido derivatives was also prepared in which the
methylene spacer had been replaced by an NH or N-Me
moiety (Table 1). Reaction of 5a with the appropriate
isocyanate using method R afforded the carboxamides
220 and 221. Alternatively 5a ,b, 46a , and 48a were
fused with the appropriate ethyl or methyl cabamoylpy-
ridine derivatives 222-224 and 226 (Scheme 12, Sup-
porting Information), as described in method S, to give
the carboxamido derivatives 227-231 and 250a ,b.
N-Methylation of the carboxamides 227 and 228a using
KOH and MeI (method T) afforded the N-methyl ana-
logues 232 and 233, respectively. When 5a ,b were
reacted with either 4-ethoxycarbonylamino-, or 3-ethox-
ycarbonylamino-N-methylpiperidine in the presence of
Following on the lead structure Sch 44342,8,9 which
contains a 4-pyridylacetamido moiety, we prepared a
series of 4-, 3-, and 2-pyridylacetamides in the pipera-
zine series from racemic 5a ,b as well as from the R(+)
46a and S(-) 48a enantiomers. All of them were found
to be potent FPT inhibitors having good selectivity for
FPT versus GGPT. The 11S(-)-enantiomers 49 and 52
were considerably more potent than the corresponding
11R(+)-enantiomers 47 and 51. The presence of a
3-bromo substituent once again conferred much greater
potency10 on the piperazine 45b relative to the unsub-

884 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Mallams et al.
stituted analogue 45a . The activity of 45a against
K-ras, was less than that observed against H-ras which
was found to be true in general throughout the series.
The 4-pyridylethyl analogue 53, which lacks the amide
carbonyl group of 50a , was devoid of FPT activity (Table
1), indicating the important role that the 4-amido moiety
plays in conferring activity in this series of FPT inhibi-
tors, as well as the 11-piperidinylidene series.10
modest FPT activity. The corresponding N-methyl,
N-Boc, NH, and NCOOEt derivatives 127-130 were
inactive. All of these derivatives lacked the methylene
spacer adjacent to the amide carbonyl. The N-methyl-
4- and -3-piperidinylacetamides 135a ,b-140 on the
other hand, which possess the methylene spacer, were
potent FPT inhibitors. They again showed good selec-
tivity for FPT over GGPT. The corresponding 5,6-ene
derivatives 141 and 142 retained essentially the same
potency as the ethano analogues 135a and 138a ,
respectively. The 4-chloro analogue 143 exhibited a
potency intermediate between that of 135a and 135b.
The N-acetyl-4- and -3-piperidinylacetamides 144a ,b-
147 were less potent than the N-methyl derivatives. The
corresponding N-Boc derivatives in the 3-unsubstituted
series 148a and 149 were inactive, while in the 3-bromo
series the N-Boc derivative 148b showed modest FPT
potency. Removal of the N-Boc group gave the free
piperidinylacetamides 150a ,b-155, which were potent
FPT inhibitors.
The availability of these free piperidines enabled us
to explore the SAR at this site in the molecule to see
what types of functional groups would be tolerated on
the piperidine. The phenyl carbamate 156 in the
3-bromo series was quite active. The methyl carbam-
ates 157 and 158 in the unsubstituted series were less
potent than the corresponding free piperidines. The
N-ethyl derivative 159 showed potency similar to that
of the N-methyl analogue 135a . An alternative neutral
substituent was needed that would be superior to the
N-acetyl and N-carbamoyl substituents, and this was
found to be the N-carboxamido moiety. The N-carboxa-
mido derivatives in both the unsubstituted series, as
well as the 3-bromo series, namely 160a ,b-166, were
found to be potent FPT inhibitors. When 160b was
administered to mice, the principal metabolites were
found to be 256 and 257. A series of N-alkyl-substituted
carboxamides 167a ,b-171a ,b were prepared, and these
analogues were only slightly less potent than the
unsubstituted carboxamides. The methylthiocarboxa-
mide 172 was also a potent FPT inhibitor. A series of
amido derivatives 173-177, derived from various amino
acids, was also prepared, and they were found in general
to be somewhat less potent than the carboxamides. It
was apparent from these studies that a wide variety of
N-substituents could be tolerated in this series of
piperidinylacetyl analogues.
A study of the metabolites formed from 50a when it
was administered to mice, using tandem mass spec-
trometry, led to the identification of the two principal
metabolites 79 and 80a in the serum. Other potential
sites of metabolism were the 3-position in the tricyclic
pyridine ring and the methylene group adjacent to the
amide carbonyl, although these possible metabolites
were not observed from 50a . A series of derivatives
having reduced potential for metabolism was prepared,
and these included the R,R-dimethylphenylacetamido
derivative 55, which was found to be less active than
than either of its 4- or 3-pyridyl counterparts 73 and
74a . The monomethyl analogue 75 was less potent than
the dimethyl derivative 74a , and both the mono- and
dimethyl analogues were less potent than the corre-
sponding unsubstituted analogues 45a and 50a . The
3-bromo derivative 74b was again more potent than the
unsubstituted analogue 74a . The 3-pyridylacrylicamide
76 was also found to possess moderate FPT activity. The
5-carboxamidoquinoline derivatives 77a ,b were pre-
pared as rigid mimics of 50a ,b, but they were found to
be inactive as FPT inhibitors. The 7-carboxamidoquino-
line derivative 78 showed moderate FPT activity. A
series of 3- and 4-pyridylacetamido N4-oxides 80a ,b-
85a ,b and a 5-carboxamidoquinoline N4-oxide 86 were
found to be slightly less potent FPT inhibitors than their
parent pyridine and quinoline counterparts. It is in-
teresting to note that, in the 3-bromo series, the 11R(+)-
enantiomer 84b was equipotent to the 11S(-)-enanti-
omer 85b, in contrast to what was observed in the
3-unsubstituted series where the 11S(-)-enantiomers
were more potent. The great advantage of the N-oxides
80a ,b and 83a ,b-85a ,b, lay in their metabolic stability
and enhanced bioavailability, which will be discussed
later. The N1-oxide 91 was slightly less active than the
pyridyl analogue 50a , while the N1,N4-oxide 92 was less
active than either the corresponding N1-oxide 91 or the
N4-oxide 80a . The R,R-dimethyl derivatives 93a ,b and
94a ,b were as active as their unsubstituted methylene
counterparts 80a ,b and 83a ,b, respectively. The 3-py-
ridylacetamides 99 and 100, having a 5,6-ene moiety18,19
in place of the 5,6-ethano bridge, were less potent than
50a and 80a , while the corresponding 4-pyridylaceta-
mide 101 was equipotent to 84b. The 4-chloro deriva-
tive20 in the 3-pyridyl series 107 was slightly less potent
than the parent compound 50a , while in the 4-pyridyl
N-oxide series 108, it was found to be slightly more
potent than the parent 83a . The effect of 2-substitution
in the piperazine ring21-23 was also investigated, and
the 2S-n-butyl and 2S-methoxyethyl analogues 115 and
116, respectively, showed potency similar to that of the
unsubstituted piperazine 83b.
Four tetra- and perhydroquinoline amides 178-181,
which constitute rigid analogues of 155 and 163, were
prepared, and only 180 and 181 were active as FPT
inhibitors. The cysteinyl and cystinyl amido derivatives
182-186 were also prepared, and the deprotected amino
The search for an alternative, metabolically resistant
group to replace the pyridylacetyl moiety led first to the
synthesis of a series of N-acetyl derivatives 118-120,
122, 124, and 125, of which only 118 and 120 showed

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 885
F igu r e 1. Summary of SAR data.
acid derivatives 184 and 186 showed only modest FPT
activity. The analogues 187-189 in which the carbonyl
group of the cysteine had been replaced by a methylene
group were devoid of FPT activity.
Replacement of the central piperazine ring with an
ethylenediamine moiety as in 192 resulted in a loss of
FPT activity. Similarly, removal of the ethano bridge
as in 196 also gave an inactive compound. The position
isomer 201 and the chlorobenzoyl analogue, lacking the
pyridyl moiety, were also inactive.
The recognition of the importance of the amide
carbonyl together with the adjacent methylene group
led us to synthesize the carbamoyl and carboxamido
derivatives in which the CH2 had been replaced with
an O or NH spacer moiety, respectively (Table 1). The
carboxamido derivatives would be expected to be meta-
bolically stable. The arylcarbamoyl derivatives 207 and
208 were inactive, while 209 and 218 were active as
FPT inhibitors. The corresponding 3-pyridyl analogues
211-213 exhibited FPT activity similar to that of the
corresponding amides described earlier. The N-oxides
214a ,b were again slightly less active than the free
pyridyl analogues. The N-methylpiperidinylcarbamates
216a ,b and 217a ,b were also active as FPT inhibitors.
The imidazole derivative 219a was essentially devoid
of FPT activity, while the 3-bromo analogue 219b was
moderately active. The benzylcarboxamide 220 was
again less active than the phenylcarboxamide 221 in
keeping with what had been observed in the amide
series. The pyridylcarboxamide analogues 227, 228a,b-
231 were all active FPT inhibitors, but were less potent
than their amide counterparts described earlier. Me-
thylation of the carboxamide NH as in 232 and 233
resulted in a considerable loss of potency. The N-oxides
249a ,b and 250a ,b were also active FPT inhibitors,
having slightly lower potency than their corresponding
amide analogues described earlier. In contrast the
N-methylpiperidinylcarboxamides 251a ,b and 252a ,b
were more potent than their corresponding amide
analogues.The carboxamide byproducts 253 and 254
were inactive, as was the piperazinylcarboxamide 255.
The SAR data for three key members of this class of
FPT inhibitors, namely 83b, 135b, and 160b, are
summarized in Figure 1. The recent elucidation of the
X-ray crystal structure of mammalian farnesyl trans-
ferase37,38 afforded for the first time a detailed picture
of the binding pocket. It is clear that the large binding
pocket present in farnesyl transferase can readily
accommodate these novel tricyclic inhibitors.
In view of the fact that Ras farnesylation is an
intracellular event, the effect of these new tricyclic
inhibitors in whole cells was examined in Cos-7 monkey
kidney cells transiently expressing either H-ras-[Val12]-
CVLS, or H-ras-[Val12]CVLL as described earlier.7 The
Cos cell assay is used to measure processing of the H-ras
protein in cells in the presence of compounds that exert
a biochemical inhibition of farnesyl transferase. This
assay measures the ability of a given compound to cross
cellular membranes and inhibit farnesyl transferase in
cells. These tricyclic FPT inhibitors inhibit fanesylation
of Ras-CVLS in cells but do not inhibit Ras-CVLL
processing or morphology,7 indicating that the in vitro
specificity observed with these inhibitors carries over
into the cell-based processing assay. In contrast, Lov-
istatin is equally effective at blocking processing and
morphological changes induced by Ras-CVLS and Ras-
CVLL. The IC50 values are given in Table 1. In general
the Cos IC50s for Ras farnesylation ranged from 0.7 to

886 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Ta ble 2. Pharmacokinetic Parametersa
Mallams et al.
AUC (µg‚h/mL)
dose
(mg/kg)
Cmax (µM)
t1/2 (h)
iv
% oral
bioavailability
time above
IC50 (h) po
no.
iv
po
po
Nude Mice
45a b,c
45bb
50a b
80a b
80b
83b
84a
84b
85a
85b
93b
115
116
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
1.8d
5.3e
0.24d
0.9
0.9
3.9
0.2
0.74
0.2
0.6
1.0
1.2
1.1
1.18
0.8
1.9
0.6
2.0
13
20.8
36
89
100
84
53
100
72
100
47
47
1
6
0.3
4
6
6+
2
7
4
7
4
7+
7
1.1e
0.4e
1.1e
12.3d
24.0e
23.4e
11.3e
27.1e
21.5e
26.1e
25.0d
91.4e
11d
12
34.8e
19.6e
6.0e
45.4
10
3
25
10
24.6
11
11.5
9
1.4
4
30
2
31.9e
15.5e
35.3e
11.7d
43.2e
19.9
135a b
135b
135bb
138b
144b
150b
160b
160bf
167b
4.1d
40.6e
32.9e
14.9e
9.51e
27.2e
33.9e
57.2e
9.9e
1.3d
>1
8.3
8.5
5.7
0.37
2
1.2
1.5
0.5
32
45
100
100
28.8
65
100
96.6
100
4
18.4e
39.0e
17e
24+
24+
24
2
24
8+
7+
4
2.74e
17.6e
70.8e
55.2e
33.3e
3
3
38.8
71.1
80
Cynomolgus Monkeys
50a
50a
25g
25b
5
10
10j
10
10
10
25b
5
48.6h
48.6h
24.4i
1.07h
0.45h
0.29
0.11
1.5
1.5
2.6
2.2
1.4
38.5 ( 9.6
83b
83b
83b
84b
85b
101
135b
160b
160b
9.4 ( 2.3i
35.6 ( 9.5i
16.7 ( 1.5
9.1 ( 1.5i
58.1 ( 11.7i
11.7 ( 1.1i
37.2i
1.2
5.7
3
7
24
24
12+
24+
31
72
7
1.8
8
1.9
1.5
0.5
2
12.3i
4.27 ( 1.09i
12.7 ( 6.3i
2.4
35
10
24
a
Animals were dosed with solutions of the hydrochloride salts of the compounds indicated in the table in 20% HPâCD, unless otherwise
noted. Administered as the hydrochloride salt in saline solution. c ICR mice. AUC (0-4 h). e AUC (0-24 h). f Administered as the
b
d
g
micronized free base in 20% HPâCD. Dosed orally as the micronized free base in corn oil; dosed iv as the hydrochloride salt in saline
h
i
j
solution. AUC (0-8 h). AUC (0-48 h). Dosed po as the hydrochloride salt in 0.4% methyl cellulose.
49 times greater than in the corresponding enzyme
inhibition assays. This is in sharp contrast to many of
the pepidomimetics that have been described in the
literature which, despite their higher intrinsic potency
in some cases and universally poor pharmacokinetics,
exhibit IC50 values that are 200-2000 times greater
than their IC50s in enzymatic assays. The Cos IC50
values seen in these tricyclic inhibitors may well be a
result of their favorable metabolic stability and excellent
cell penetration.
The effects of two of the most important compounds,
namely 83b and 85b, on the anchorage-independent
growth of NIH-H, NIH-K, and DLD-1 tumor cell lines
in soft agar39,40 were determined. The soft agar cloning
assays are used to examine the potential therapeutic
effects of a given farnesyl transferase inhibitor. Com-
pounds that are active in the Cos assay, and therefore
are shown to cross the cell membrane, are then exam-
ined for their potential antitumor activity in the soft
agar cloning assays. Compounds that are inhibitors of
soft agar cloning are candidates for in vivo analysis. The
tumor cell line expressing the H-ras oncogene (NIH-H)
was found to be more sensitive to 83b and 85b than
the cell lines expressing the K-ras oncogene (NIH-K and
DLD-1).
results are given in Table 2. The early pyridylacetyl
lead structures 45a and 50a were shown to be heavily
metabolized, and the low AUCs that were observed for
these compounds reflect this fact. However, it was
encouraging to note that they were both orally bioavail-
able. The 3-bromo analogue 45b showed a slight
improvement over 45a in both AUC, serum half-life, and
duration of action. The introduction of an N-oxide on
the pyridylacetyl moiety in 80a ,b and 83b which ef-
fectively blocked metabolism at this site resulted in a
dramatic increase in the AUCs of these compounds
relative to their pyridylacetyl counterparts. The com-
pounds exhibited good oral absorption in mice with
serum half-lives of 0.6-1.2 h and 84-100% oral bio-
availability. The pharmacokinetics of the pure 11-
enantiomers 84a ,b and 85a ,b were also studied, and
in the 3-unsubstituted series the 11S(-)-enantiomer
85a was found to exhibit superior oral absorption
characteristics to the 11R(+)-enantiomer 84a . In the
3-bromo series, on the other hand, the enantiomers 84b
and 85b exhibited similar AUCs. Thus 85b showed a
po AUC of 35.3 µg‚h/mL, showed a po Cmax of 24.6 µM,
and was 100% orally bioavailability. The serum con-
centrations also remained above the IC50 for 7 h when
dosed orally. The R,R-dimethyl analogue 93b showed
poorer oral absorption characteristics than the unsub-
stituted compound 80b. The introduction of a 2S-n-
butyl substituent in the piperazine ring as in 115
P h a r m a cok in etic Stu d ies. The pharmacokinetic
parameters for a selection of the inhibitors prepared in
this group were determined in nude mice, and the

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 887
F igu r e 2. In vivo antitumor efficacy in mice.
resulted in a higher po AUC, but the oral bioavailability
was lower than that of 83b. The introduction of a 2S-
methoxyethyl substituent in the piperazine ring as in
116 gave a compound with an oral AUC similar to that
of 83b.
AUC than that of the ethano analogue 83b. Both the
N-methyl-4-piperidinylacetyl derivative 135b and the
N-carboxamido-4-piperidinyl derivative 160b exhibited
excellent oral pharmacokinetics.
In Vivo An titu m or Effica cy in Mice. Nude mice
carrying tumors grown from tumor cell lines containing
mutated K-ras (DLD-1 and SW-620) and mutated H-ras
oncogenes (CVLS, CVLL and pT-24) were dosed with
selected inhibitors prepared in this study, and the
results are given in Figure 2. The tumor models include
CVLS (NIH3T3 cells transfected with oncogenic H-ras
) and pT-24 (BALB c/3T3 cells transfected with onco-
genic H-ras) cells, each of which have activated H-ras
containing its native CVLS farnesylation sequence.
Tumor models using human colon adenocarcinoma
DLD-1 and SW-620 cells, which express activated K-ras,
were also used. It should be noted that the SW-620
model is less sensitive to farnesyl transferase inhibitors
than the DLD-1 model. On the other hand, the model
using CVLL (NIH3T3 cells transfected with oncogenic
H-ras) cells in which the terminal sequence of the
activated H-ras is altered to CVLL was used as a control
to direct the prenylation of the protein toward gera-
nylgeranylation rather than farnesylation. As antici-
pated for selective FPT inhibitors, the percent tumor
inhibition in the control CVLL model was considerably
lower than that observed in either the CVLS, or pT-24
models suggesting that these tricyclics were specific Ras
farnesylation inhibitors.41,42 However, the observation
that efficacy was also seen at higher doses (although
significantly reduced) against tumors derived from cells
transformed with a geranylgeranylated form of H-ras
suggests that inhibition of Ras farnesylation may not
be solely responsible for the observed antitumor effect.42
Recent reports have demonstrated that the K- and N-
isoforms of the Ras protein are geranylgeranylated in
cells treated with FPT inhibitors,39,43-45 which may also
be the case in the in vivo setting although this has not
been demonstrated. Whether it is a geranylgeranylated
form of the Ras protein that is driving the transforma-
tion or whether there is a distinct farnesylated protein
lying downstream of Ras in the transformation pathway
that is responsible for the observed activity in the CVLL
model is still unclear.42 K- and N-ras proteins are more
frequently mutated in human cancers than H-ras.
These results suggest a potential mechanism for resis-
tance of tumor cells to FPT inhibitors. However, it has
The N-methyl-4-piperidinylacetyl analogue 135a ex-
hibited better serum levels and bioavailability than the
4-pyridylacetyl analogue 45a . The corresponding 3-bro-
mo analogue 135b showed a comparable po AUC to that
of 83b with a lower Cmax, a longer serum half-life, and
duration above the IC50 but a lower oral bioavailability.
The oral absorption of 135b was significantly higher
when the compound was dosed in saline solution. The
N-methyl-3-piperidinylacetyl derivative 138b also ex-
hibited excellent oral absorption and pharmacokinetics.
In general the N-methylpiperidine analogues exhibited
muscarinic activity (Table 7, Supporting Information).
The N-acetyl-4-piperidinylacetyl analogue 144b on the
other hand exhibited poor oral absorption, while the
N-unsubstituted analogue 150b showed good oral ab-
sorption characteristics. The introduction of an N-
carboxamido substituent as in 160b gave a compound
with excellent oral pharmacokinetics (Table 2) and
negligible muscarinic activity (Table 7). The po AUC
was 70.8 µg‚h/mL with a po Cmax of 38.8 µM and a serum
half-life of 1.2 h. The compound exhibited 100% oral
bioavailability and was present in serum at concentra-
tions above the IC50 for 8+ h. When the free base was
used, a slightly lower AUC was observed than with the
hydrochloride salt. The po AUC of the racemic 160b
was greater than that of the racemic 83b and than that
of the pure 11S(-)-enantiomer 85b. The N-methylcar-
boxamido analogue 167b also showed good oral absorp-
tion, with a po AUC about half that of 160b and a higher
Cmax. It was also present in serum at concentrations
above the IC50 for 4 h.
The pharmacokinetics of the more interesting com-
pounds from the above studies in mice were then
determined in cynomolgus monkeys, and the results are
given in Table 2. As was observed in mice, the 3-py-
ridylacetyl derivative 50a showed poor oral bioavail-
ability due to metabolism. On the other hand, the
4-pyridylacetyl N-oxides 83b, 84b, and 85b all exhibited
excellent oral absorption characteristics with the 11S(-
)-enantiomer 85b being superior to either the racemate
83b or the 11R(+)-enantiomer 84b. Introduction of a
5,6-ene, as in 101, gave a compound with a lower po

888 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Mallams et al.
been shown39,40 that 60-70% of human tumor cell lines
tested were sensitive to growth inhibition by FPT
inhibitors. The mechanism for this sensitivity is not
understood. The effects of FPT inhibitors could be
mediated through Ras if the geranylgeranylated forms
of the protein do not signal as efficiently as the farne-
sylated forms of the proteins. It is also possible that
the effects of FPT inhibitors may be mediated through
farnesylated proteins other than Ras, such as rho B.46
Even though the identity of the proteins mediating the
antitumor effects of FPT inhibitors remains an impor-
tant unanswered question, the effects observed in
human tumor cell lines with several different classes
of these inhibitors suggest broad utility as cancer
chemotherapeutic agents.
with additional tricyclic derivatives from these labora-
tories, as potential orally active antitumor agents.
Exp er im en ta l Section
13C NMR spectra were obtained at 300 MHz on a Varian
Gemini 300 instrument. Chemical shifts are reported in ppm
(δC) using Me4Si as standard. Chemical ionization mass
spectra (CIMS) were obtained on a Hewlett-Packard MS-
engine mass spectrometer using methane as a reagent gas.
The samples were introduced through a particle-beam inter-
face and the source temperature was kept at 250 °C. Fast
atom bombardment mass spectra (FABMS) and liquid second-
ary ion mass spectra (LSIMS) were carried out on double-
focusing mass spectrometers (J EOL J MS-HX-110a, VG-ZAB-
SE, MAT-90) operating at 5-10 kV accelerating voltage.
Samples were introduced into the mass spectrometer in the
condensed phase and desorbed and ionized by bombardment
with fast moving atoms (FAB) or ions (LSIMS). In the case
of FABMS, xenon atoms at 8 kV energy were used, while for
LSIMS, high-energy Cs+ ions at 25-30 kV were applied.
Typically, 1-2 µg of the sample dissolved in 1-2 µL of DMSO
was deposited onto a stainless steel probe tip containing 1-2
µL of matrix (3-NBA, or glycerol/thioglycerol mixture, 1:1),
which was then introduced onto the mass spectrometer.
Rotations were recorded on a Perkin-Elmer 243B polarim-
eter. Microanalyses for C, H, N, and S were determined on a
Fisons EA1108 elemental analyzer, while Br and Cl were run
by Robertson Microlit. All solvents used for reactions were
anhydrous. Products were purified by column chromatography
on 63-200 mesh silica gel (Selecto Scientific). Reactions and
purity of samples were monitored on 250 µm silica gel GF thin
layer plates (Analtech). In general the solvent systems used
in chromatography consisted of the appropriate percentage (ca.
0.25-10%) of a 10% concentrated ammonium hydroxide solu-
tion in methanol, in dichloromethane. Yields are unoptimized.
All reactions were carried out under dry argon.
The amides 39, 45a , and 50a all inhibited tumor
growth to a moderate degree. The racemic 4-pyridy-
lacetyl N-oxide 83b (Sch 59228)41,42 was considerably
more effective in reducing the extent of tumor growth
in all of the cell lines tested. The 11S(-)-enantiomer
85b (Sch 61129) was nearly twice as effective as 83b at
comparable doses. The 5,6-ene analogue 101 was
slightly less effective in reducing tumor volume than
the corresponding ethano analogue 83b. The 4-pip-
eridinylacetyl derivative 150b showed inhibition of
tumor growth similar to that observed with 83b, but it
also showed some toxicity at the higher dose. The
racemic N-carboxamidopiperidinylacetyl derivative 160b
was also almost as effective as the 11S(-)-enantiomer
of the 4-pyridylacetyl N-oxide derivative 85b. The 5,6-
ene analogue 164 was once again less effective in
reducing tumor growth than the corresponding ethano
analogue 160b. The 3-pyridylcarboxamide 228a was
about half as effective in reducing tumor volume as the
corresponding amide 50a . The 2-pyridylcarboxamide
231 showed efficacy similar to that of 228a .
1-(3-B r o m o -8-c h lo r o -6,11-d ih y d r o -5H -b e n zo [5,6]-
cycloh ep ta [1,2-b]p yr id in -11-yl)p ip er a zin e (5b) (Sch em e
1). Using essentially similar procedures to those described
earlier,11 3-bromo-8-chloro-11H-benzo[5,6]cyclohepta[1,2-b]py-
ridin-11-one (2b)11,12 was reduced to the 11-ol 3b (92%): CIMS
m/z 324 (MH+); 13C NMR δC (CH2) 31.3, 29.9; (CH) 145.8, 141.6,
128.6, 127.1, 125.4, 68.8; (C) 154.4, 140.6, 138.6, 134.2, 133.2,
119.6. Anal. (C14H11BrClNO) C, H, Br, Cl, N. The 11-ol 3b
was reacted with thionyl chloride to afford the 11-chloro
derivative 4b which was in turn reacted with an excess of
piperazine to give 5b (Table 1).
These results clearly indicate that these new FPT
inhibitors possess in vivo antitumor activity upon oral
dosing in tumor models expressing activated ras onco-
genes. Particularly noteworthy were the compounds
83b (Sch 59228), 85b (Sch 61129), and 160b (Sch 60677)
which showed excellent in vivo efficacy in mice suggest-
ing broad utility as potential cancer chemotherapeutic
agents.
(+)-1-(3-Br om o-8-ch lor o-6,11-d ih yd r o-5H -b en zo[5,6]-
cycloh ep ta [1,2-b]p yr id in -11(R)-yl)p ip er a zin e (46b) a n d
(-)-1-(3-Br om o-8-ch lor o-6,11-d ih yd r o-5H -b e n zo[5,6]-
cycloh ep ta [1,2-b]p yr id in -11(S)-yl)p ip er a zin e (48b). Ra-
cemic 5b (29.92 g) was resolved by Chiral Technologies on a
CHIRALPAK AD (10 × 50 cm) HPLC column using hexane-
2-propanol-diethylamine (60-40-0.1) as the eluant at 40 °C
using UV 254 nm detection to give 46b (12.9 g, 86.2% recovery,
97.6% ee, 99+% chemical purity) [CIMS m/z 392 (MH+); 13C
NMR δC essentially identical to that reported for 5b (Table
Con clu sion s
A novel group of tricyclicpiperazinyl derivatives has
been discovered which are selective Ras farnesylation
inhibitors. They are selective for FPT over GGPT, are
nonpeptidic in nature, and do not contain sulfhydryl
groups. In the piperazine series an extensive SAR has
been developed within the group of compounds having
amido substituents on the piperazine ring and this was
used to evaluate the merits of the carbamoyl and
carboxamido analogues. All three series gave potent
FPT inhibitors in vitro; however, the amides proved to
be the superior group based on their cellular activity.
Three compounds, namely the racemic 4-pyridylacetyl
N-oxide 83b (Sch 59228), its 11S(-)-enantiomer 85b
(Sch 61129), and the racemic N-carboxamido-4-piperidi-
nylacetyl derivative 160b (Sch 60677) were clearly
superior in view of their excellent pharmacokinetic
profiles and oral activity and were evaluated41,42 along
25
1); [R]D +25.8° (c 0.423, MeOH). Anal. (C18H19BrClN3) C,
H, Cl, N; Br: calcd, 20.35; found, 19.22] followed by 48b (8.5
g, 57.1% recovery, 99.7% ee, 97.7% chemical purity): CIMS
m/z 392 (MH+); 13C NMR δC essentially identical to that
reported for 5b (Table 1); [R]D25 -27.9° (c 0.445, MeOH). Anal.
(C18H19BrClN3) C, H, Cl, N; Br: calcd, 20.35; found, 19.37.
Meth od A: Gen er a l P r oced u r e for th e Syn th esis of th e
Am id es in Ta ble 1. 1-(3-Br om o-8-ch lor o-6,11-d ih yd r o-
5H-ben zo[5,6]cycloh ep ta [1,2-b]pyr id in -11-yl)-4-(4-p yr id y-
la cetyl)p ip er a zin e N4-oxid e (83b). The piperazine deriva-
tive 5b (2.1 g, 5.35 mmol), 4-pyridylacetic acid N-oxide 70 (1.23
g, 8.0 mmol), 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide
hydrochloride (1.54 g, 8.0 mmol), 1-hydroxybenzotriazole (1.08
g, 8.0 mmol), and 4-methylmorpholine (1.18 mL, 8.0 mmol)
were dissolved in DMF (50 mL), and the mixture was stirred
at 25 °C for 18 h. The solution was evaporated to dryness,
and the residue was taken up in CH2Cl2, washed with

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 889
saturated aqueous NaHCO3 and water, and dried (Na2SO4).
The product was chromatographed on a silica gel column using
2.5% (10% concentrated NH4OH in MeOH)-CH2Cl2 to give
83b (Table 1).
Using the above procedure, the various intermediate pip-
erazines 5a ,b, 46a ,b, 48a ,b, 90, 98a ,b, 106, 112, and 114, were
reacted with the appropriate acids to give the products shown
in Table 1 where method A is indicated.
Meth od B: 1-(3-Br om o-8-ch lor o-6,11-dih ydr o-5H-ben zo-
[5,6]cycloh ep t a [1,2-b]p yr id in -11-yl)-4-[(4-p ip er id in yl)-
a cetyl]p ip er a zin e (150b). The N-Boc derivative 148b (4.61
g, 7.5 mmol) was dissolved in MeOH (40 mL), and a 10% (v/v)
concentrated H2SO4 in dioxane solution (100 mL) was added.
The mixture was stirred at 25 °C for 2 h and then basified
with concentrated aqueous NaOH. The usual workup followed
by chromatography on silica gel using 10% (10% concentrated
NH4OH in MeOH)-CH2Cl2 gave 150b (Table 1).
In Table 1 where method G is indicated, the appropriate
amine was reacted with TMSNCO as described above to give
the carboxamide.
Meth od H: 4-[2-[4-(3-Br om o-8-ch lor o-6,11-d ih yd r o-5H-
ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-p ip er a zin yl]-
2-oxoeth yl]-N-m eth yl-1-p ip er id in eca r boxa m id e (167b).
The amine 150b (500 mg, 1.0 mmol) and MeNCO (220.3 mg/
0.228 mL, 4 mmol) were dissolved in CH2Cl2 (5 mL), and the
mixture was stirred at 25 °C for 47 h. Additional MeNCO
(110.2 mg/0.114 mL, 2.0 mmol) was added, and the mixture
was stirred at 25 °C for a total of 144 h and at 74 °C for an
additional 5 h. After an additional 24 h at 25 °C, the reaction
was evaporated to dryness and chromatographed on silica gel
using 2% (10% concentrated NH4OH in MeOH)-CH2Cl2 as the
eluant to give 167b (Table 1).
In Table 1 where method H is indicated, the appropriate
amine was reacted with the isocyanate as described above to
give the N-substituted carboxamide.
In Table 1 where method B is indicated, the appropriate
N-Boc derivative prepared by method A above was deprotected
as above, to give the corresponding free amine.
Meth od I: 4-[2-[4-(8-Ch lor o-6,11-d ih yd r o-5H-ben zo[5,6]-
cycloh epta[1,2-b]pyr idin -11-yl)-1-piper azin yl]-2-oxoeth yl]-
N-m eth yl-1-p ip er id in eca r both ioa m id e (172). The amine
150a (500 mg, 1.1 mmol) and MeNCS (166.6 mg/0.156 mL,
2.2 mmol) were dissolved in CH2Cl2 (5 mL), and the mixture
was stirred at 25 °C for 24 h. Additional MeNCS (166.6 mg/
0.156 mL, 2.2 mmol) was added, and the reaction was allowed
to run for a total of 52 h. The usual workup followed by
chromatography on silica gel using 3% -10% concentrated
NH4OH in MeOH)-CH2Cl2 as the eluant gave 172 (Table 1).
Meth od J : 1-[[1-(2(S)-Am in o-3-h yd r oxy-1-oxop r op yl)-
4-p ip er id in yl]a cetyl]-4-(3-br om o-8-ch lor o-6,11-d ih yd r o-
5H -b e n zo[5,6]cycloh e p t a [1,2-b]p yr id in -11-yl)p ip e r a -
zin e (175b). The amine 150b (500 mg, 1.0 mmol), N-Boc-L-
serine (297.2 mg, 1.5 mmol), 1-(3-(dimethylamino)propyl)-3-
ethylcarbodiimide hydrochloride (277.6 mg, 1.5 mmol),
1-hydroxybenzotriazole (195.7 mg, 1.5 mmol), and 4-methyl-
morpholine (146.5 mg/0.1593 mL, 1.5 mmol) were dissolved
in DMF (25 mL), and the mixture was stirred at 25 °C for 24
h. The product was worked up as described in method A
above, and the resulting N-Boc derivative was taken up in
MeOH (5 mL), and 10% (v/v) concentrated H2SO4 in dioxane
(10 mL) was added. The mixture was stirred at 25 °C for 2 h.
BioRad AG1-X8(OH-) resin was added, and the resin was
filtered off and washed with MeOH. The combined filtrates
were evaporated to dryness and chromatographed on silica gel
using 5% (10% concentrated NH4OH in MeOH)-CH2Cl2 as the
eluant to give 175b (Table 1).
Meth od C: Meth yl 4-[[[4-(8-ch lor o-6,11-d ih yd r o-5H-
ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-p ip er a zin yl]-
ca r b on yl]m et h yl]-1-p ip er id in eca r b oxyla t e (157). The
amine 150a (500 mg, 1.1 mmol) was dissolved in dry THF (5
mL), and MeOCOCl (107.6 mg/0.088 mL, 1.1 mmol) was added.
The mixture was stirred at 25 °C for 1 h and then evaporated
to dryness. The product was worked up in the usual way and
chromatographed on silica gel using 1.5% (10% concentrated
NH4OH in MeOH)-CH2Cl2 as the eluant to give 157 (Table
1).
In Table 1 where method C is indicated, the appropriate
amine was reacted the chloroformate to give the corresponding
carbamoyl derivative.
Meth od D: 1-(3-Br om o-8-ch lor o-6,11-dih ydr o-5H-ben zo-
[5,6]cycloh ep t a [1,2-b]p yr id in -11-yl)-4-[(1-a cet yl-4-p ip e-
r id in yl)a cetyl]p ip er a zin e (144b). The amine 150b (500 mg,
1.0 mmol) and Ac2O (591.4 mg/0.546 mL, 6 mmol) were
dissolved in MeOH (5 mL), and the mixture was stirred at 25
°C for 19 h. The usual workup followed by chromatography
on silica gel using 2% (10% concentrated NH4OH in MeOH)-
CH2Cl2 gave 144b (Table 1).
Meth od E: P h en yl 4-[2-[4-(3-br om o-8-ch lor o-6,11-d i-
h yd r o-5H -b en zo[5,6]cycloh ep t a [1,2-b]p yr id in -11-yl)-1-
p ip er a zin yl]-2-oxoeth yl]-1-p ip er id in eca r boxyla te (156).
The amine 150b (500 mg, 0.1 mmol) and (C6H5O)2CO (1.02 g,
0.2 mmol) were dissolved in 2-propanol (20 mL), and the
mixture was heated at 87 °C for 52.5 h. The usual workup
followed by chromatography on silica gel using EtOAc as the
eluant gave 156 (Table 1).
In Table 1 when method J is indicated, the appropriate
amine was reacted as described above with the N-Boc amino
acid and then deprotected to give the desired compound.
Meth od K: 1-[[1-(2(R)-Am in o-3-m er capto-1-oxopr opyl)-
4-p ip er id in yl]a cetyl]-4-(3-br om o-8-ch lor o-6,11-d ih yd r o-
5H -b e n zo[5,6]cycloh e p t a [1,2-b]p yr id in -11-yl)p ip e r a -
zin e Hyd r och lor id e (177). The amine 150b (1 g, 1.9 mmol),
N-Boc-S-trityl-L-cysteine (1.34 g, 2.85 mmol), 1-(3-(dimethy-
lamino)propyl)-3-ethylcarbodiimide hydrochloride (555.3 mg,
2.85 mmol), 1-hydroxybenzotriazole (391.4 mg, 2.85 mmol), and
4-methylmorpholine (293 mg/0.319 mL, 2.85 mmol) were
dissolved in DMF (50 mL), and the mixture was stirred at 25
°C for 24 h. The product was worked up as described in
method A above and chromatographed on silica gel using 0.4-
0.8% (10% concentrated NH4OH in MeOH)-CH2Cl2 as the
eluant to give the N-Boc-S-trityl derivative (1.66 g, 89%):
CIMS m/z 962.9 (MH+).
Met h od F : 1-(8-Ch lor o-6,11-d ih yd r o-5H -b en zo[5,6]-
cycloh ep ta [1,2-b]p yr id in -11-yl)-4-[(1-eth yl-4-p ip er id in y-
l)a cetyl]p ip er a zin e (159). The amine 150a (500 mg, 1.0
mmol) was dissolved in 0.6 N HCl in CH2Cl2 (10 mL) and
stirred for 5 min. Evaporation to dryness gave the hydrochlo-
ride which was dissolved in MeOH (20 mL). Acetaldehyde
(200.6 mg/0.254 mL, 4.0 mmol), NaBH3CN (85.9 mg, 1.2 mmol)
and 3 Å molecular sieves (500 mg) were added, and the
mixture was heated at 40 °C for 115 h. The usual workup,
followed by chromatography on silica gel using 8% (10%
concentrated NH4OH in MeOH)-CH2Cl2 as the eluant gave
159 (Table 1).
Meth od G: 4-[2-[4-(3-Br om o-8-ch lor o-6,11-d ih yd r o-5H-
ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-p ip er a zin yl]-
2-oxoeth yl]-1-p ip er id in eca r boxa m id e (160b). The amine
150b (11 g, 21.2 mmol) was dissolved in CH2Cl2 (110 mL), and
TMSNCO (14.7 g/17.25 mL, 127.2 mmol) was added dropwise
to the solution. The mixture was stirred at 25 °C for 47 h.
Additional TMSNCO (7.35 g/8.63 mL, 63.6 mmol) was added,
and the mixture was stirred for a total of 166 h. The mixture
was washed with saturated aqueous NaHCO3 and water, dried
(Na2SO4), and evaporated to dryness. Chromatography on
silica gel using 3% (10% concentrated NH4OH in MeOH)-CH2-
Cl2 as the eluant gave 160b (Table 1).
Meth od L. The N-Boc-S-trityl derivative (1.56 g, 1.6 mmol)
was dissolved in CH2Cl2 (15 mL). Et3Si (753 mg/1.05 mL, 6.4
mmol) and TFA (7.5 mL) were added, and the mixture was
stirred at 25 °C for 1 h. The solution was evaporated to
dryness, and the residue was partitioned between hexane and
water. The aqueous layer was again extracted with hexane
(500 mL). The aqueous layer was then passed over a bed of
BioRad AG3 × 4(Cl-) resin (300 mL), and the resin was eluted
with water (800 mL). The aqueous solution was lyophilized
to give 177 hydrochloride (Table 1).

890 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Mallams et al.
In Table 1 when method K is indicated, the appropriate
amine was reacted as described above with the N-Boc-S-trityl
amino acid and then deprotected to give the desired compound.
Meth od R: N-Ben zyl-4-(8-ch lor o-6,11-dih ydr o-5H-ben zo-
[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-p ip er a zin eca r box-
a m id e (220). Benzyl isocyanate (424.3 mg/0.394 mL, 3.19
mmol) was dissolved in CH2Cl2 (1 mL), and a solution of the
piperazine 5a (1 g, 3.19 mmol) was added dropwise at 25 °C
over 20 min. The mixture was stirred at 25 °C for 3 h and
then worked up in the usual way. The product was chromato-
graphed on silica gel using 1% (10% concentrated NH4OH in
MeOH)-CH2Cl2 as the eluant to give 220 (Table 1).
Met h od M: 3-P yr id yl 4-(8-Ch lor o-6,11-d ih yd r o-5H -
ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-p ip er a zin e-
ca r boxyla te N1-Oxid e (214a ). The piperazine 5a (500 mg,
1.6 mmol) and pyridine (0.515 mL, 6.4 mmol) in CH2Cl2 (12
mL) were added dropwise to a stirred solution of 1.93 M
phosgene in toluene (20%) (12.6 mL, 24 mmol) in CH2Cl2 (12
mL) at 0 °C over 45 min. The solution was allowed to warm
to 25 °C over 30 min. Excess phosgene was removed with a
stream of argon, and the solution was evaporated to dryness
to give 206 which was used without purification in the next
step.
The crude 206 (1.6 mmol) above was dissolved in DMF (10
mL) containing pyridine (0.515 mL, 6.4 mmol), and 3-hydroxy-
pyridine N1-oxide (885 mg, 7.97 mmol) was added. The
mixture was stirred at 25 °C for 18 h and worked up in the
usual way. Chromatography on silica gel using 1.5% (10%
concentrated NH4OH in MeOH)-CH2Cl2 as the eluant gave
214a (Table 1).
Met h od N: 3-P yr id yl 4-(8-Ch lor o-6,11-d ih yd r o-5H -
ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-p ip er a zin e-
ca r boxyla te (211). A 1.93 M solution of phosgene in toluene
(20%) (173.5 mL, 334.5 mmol) was added to CH2Cl2 (175 mL),
and the solution was cooled to 0 °C. A solution of 3-hydroxy-
pyridine (7.27 g, 66.9 mmol) and pyridine (7.05 g/7.22 mL, 89.2
mmol) in CH2Cl2 (175 mL) was added dropwise at 0 °C over 1
h. The mixture was stirred and allowed to warm to 25 °C over
2 h. Excess phosgene was removed using a stream of argon
with suitable trapping of the waste stream. The remaining
solution was evaporated to dryness to give 210. The residue
was dissolved in CH2Cl2 (53 mL) and pyridine (83 mL), and
the piperazine 5a (7 g, 1222.3 mmol) was added. The mixture
was stirred at 25 °C for 24 h and worked up in the usual way,
and the product was chromatographed on silica gel using 1%
(10% concentrated NH4OH in MeOH)-CH2Cl2 as the eluant
to give 211 (Table 1).
Met h od O: 1-(8-Ch lor o-6,11-d ih yd r o-5H -b en zo[5,6]-
cycloh ep t a [1,2-b]p yr id in -11-yl)-4-(1H -im id a zol-1-ylca r -
bon yl)p ip er a zin e (219a ). The piperazine 5a (10 g, 31.9
mmol) was dissolved in CH2Cl2 (100 mL) and added to a stirred
solution of carbonyldiimidazole (5.17 g, 31.9 mmol) in CH2Cl2
(150 mL) at 0 °C over 45 min. The mixture was stirred at 0
°C for a total of 2 h and then worked up in the usual way.
Chromatography on silica gel using 2% (10% concentrated
NH4OH in MeOH)-CH2Cl2 as the eluant gave 219a (Table
1).
In Table 1 when method R is cited, the appropriate amine
was reacted with the appropriate isocyanate as described
above to give the desired compound.
Met h od S: 4-(8-Ch lor o-6,11-d ih yd r o-5H -b en zo[5,6]-
cycloh epta[1,2-b]pyr idin -11-yl)-N-(1-m eth yl-3-piper idin yl)-
1-p ip er a zin eca r boxa m id e (252a ). The piperazine 5a (500
mg, 1.6 mmol) and 3-(methoxycarbonylamino)-N-methylpip-
eridine (1.098 g, 6.4 mmol) were fused at 160 °C in a sealed
pressure reaction vessel for 17 h. The product was chromato-
graphed on silica gel using 1.5% (10% concentrated NH4OH
in MeOH)-CH2Cl2 as the eluant to give 252a (Table 1).
In Table 1 when method S is cited, the appropriate amine
was reacted with the appropriate carbamate as described
above to give the desired compound.
Met h od T: 4-(8-Ch lor o-6,11-d ih yd r o-5H -b en zo[5,6]-
cycloh epta[1,2-b]pyr idin -11-yl)-N-m eth yl-N-(3-pyr idin yl)-
1-p ip er a zin eca r boxa m id e (233). The urea 228a (10.03 g,
23.1 mmol) was dissolved in DMSO (37.6 mL), and powdered
KOH (2.62 g, 23.1 mmol) in DMSO (25.1 mL) was added to
the stirred solution at 25 °C over 3 min. MeI (3.31 g/1.4518
mL, 23.1 mmol) was added and the mixture was stirred at 25
°C for 15 min. The usual workup, followed by chromatography
on silica gel using 3-5% (10% concentrated NH4OH in
MeOH)-CH2Cl2 as the eluant afforded 233 (Table 1).
Meth od U: 4-(3-Br om o-8-ch lor o-6,11-dih ydr o-5H-ben zo-
[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-N-(1-m eth yl-4-p ip e-
r id in yl)-1-p ip er a zin eca r boxa m id e (251b). The acyl azide
241 (Scheme 14) (1.92 g, 11.7 mmol) in toluene (100 mL) was
heated at 110 °C for 1 h. The piperazine 5b (500 mg, 1.3
mmol) in CH2Cl2 (100 mL) was added to the cooled solution
at 25 °C, and the mixture was stirred for 85 h. The reaction
was worked up in the usual way, and the product was
chromatographed on silica gel using 2-3% (10% concentrated
NH4OH in MeOH)-CH2Cl2 as the eluant to give 251b (Table
1).
In Table 1 when method U is cited, the appropriate amine
was reacted as described above to give the desired compound.
Met h od V: 4-(8-Ch lor o-6,11-d ih yd r o-5H -b en zo[5,6]-
cycloh ep ta [1,2-b]p yr id in -11-yl)-N-(4-p yr id in yl)-1-p ip er a -
zin eca r boxa m id e N1-Oxid e (249a ). 4-Aminopyridine N1-
oxide 248 (prepared as described in Scheme 15) (87.7 mg, 0.8
mmol) was dissolved in dioxane (6 mL), and a 60% dispersion
of NaH in mineral oil (19.1 mg, 0.6 mmol) was added. The
mixture was stirred at 25 °C for 15 min, and the imidazole
219a (250 mg, 0.61 mmol) was added. The solution was stirred
at 25 °C for 18 h and then worked up in the usual way. The
product was chromatographed on silica gel using 3-5.5% (10%
concentrated NH4OH in MeOH)-CH2Cl2 as the eluant to give
249a (Table 1).
Meth od W: 4-(3-Br om o-8-ch lor o-6,11-dih ydr o-5H-ben zo-
[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-N-(1-m eth yl-4-p ip e-
r id in yl)-1-p ip er a zin eca r boxa m id e (251b). 4-(Ethoxycar-
bonylamino)-N-methylpiperidine (848.5 mg, 4.6 mmol) and
Et3N (553.2 mg/0.762 mL, 5.59 mmol) were dissolved in
toluene (10 mL), and the mixture was heated at 110 °C for
5min. The solution was cooled, â-chlorocatecholborane (843.5
mg, 5.59 mmol) was added, and the mixture was heated at
110 °C for 5 min. The solution was cooled, and a solution of
the piperazine 5b (500 mg, 1.3 mmol) in CH2Cl2 (2 mL) was
added. The mixture was stirred at 25 °C for 65 h and then
worked up in the usual way. The product was chromato-
graphed on silica gel using 3% (10% concentrated NH4OH in
MeOH)-CH2Cl2 as the eluant to give 251b (Table 1).
In Table 1 when method W is cited, the appropriate amine
was reacted as described above to give the desired compound.
Similarly 5b was reacted as above to give 219b (Table 1).
Meth od P : 1-Meth yl-3-piper idin yl 4-(3-Br om o-8-ch lor o-
6,11-d ih yd r o-5H-ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-
yl)-1-p ip er a zin eca r boxyla te (217b). The imidazole deriva-
tive 219b (670.8 mg, 1.6 mmol), ZnBr2 (1.24 g, 6.4 mmol), and
3-hydroxy-N-methylpiperidine (634.9 mg, 6.4 mmol) were
dissolved in DMF (20 mL), and the mixture was heated at 90
°C for 72 h. The mixture was evaporated to dryness and
worked up in the usual way. The product was chromato-
graphed on silica gel using 1.5% (10% concentrated NH4OH
in MeOH)-CH2Cl2 as the eluant to give 217b (Table 1).
In Table 1 when method P is indicated, the appropriate
imidazole derivative was reacted as described above to afford
the desired compound.
Meth od Q: 1-Meth yl-3-p ip er id in yl 4-(8-Ch lor o-6,11-
d ih yd r o-5H-ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-1-
p ip er a zin eca r boxyla te (217a ). 3-Hydroxy-N-methylpipe-
ridine (550 mg, 4.68 mmol) and a 60% dispersion of NaH in
mineral oil (110.5 mg, 3.6 mmol) were added to DMF (10 mL),
and the mixture was stirred at 25 °C for 15 min. The
imidazole derivative 219a (500 mg, 1.2 mmol) was added, and
the mixture was stirred at 90 °C for 96 h. The reaction was
worked up in the usual way and chromatographed on silica
gel using 1.5% (10% concentrated NH4OH in MeOH)-CH2Cl2
as the eluant to give 217a (Table 1).

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 891
1-(8-Ch lor o-6,11-d ih yd r o-5H-ben zo[5,6]cycloh ep ta [1,2-
b]p yr id in -11-yl)-4-[2-(3-p yr id in yl)eth yl]p ip er a zin e (53).
The amide 50a (500 mg, 1.15 mmol) in THF (2 mL) was added
over 3 min to a 1 M LiAlH4 solution in THF (1.154 mL, 1.15
mmol), and the mixture was heated under reflux for 15min.
The mixture was heated at 66 °C for 3 h and then worked up
in the usual way. The product was chromatographed on silica
gel using 3-5.5% (10% concentrated NH4OH in MeOH)-CH2-
Cl2 as the eluant to give 53 (278.6 mg, 58%) (Table 1).
N-[2-(8-Ch lor o-6,11-d ih yd r o-5H-ben zo[5,6]cycloh ep ta -
[1,2-b]p yr id in -11-yl)a m in o]a ceta m id e (190) (Sch em e 2).
Mono-N-acetylethylenediamine (25 g, 244.8 mmol) was dis-
solved in MeOH (250 mL) (Al2O3 dried), and 28.8% (w/w) HCl
in EtOH (10.35 g, 81.6 mmol) and 3 Å molecular sieves were
added. The ketone 2a (9.94 g, 40.8 mmol) and NaBH3CN (3.08
g, 40.8 mmol) were added, and the mixture was heated at 40
°C for 68 h and at 60 °C for a total of 236 h. The reaction was
filtered through Celite and worked up in the usual way. The
product was chromatographed on silica gel using 3-10% (10%
concentrated NH4OH in MeOH)-CH2Cl2 as the eluant to give
190 (2.48 g, 25%) (Table 6).
N-(8-Ch lor o-6,11-dih yd r o-5H-ben zo[5,6]cycloh epta [1,2-
b]p yr id in -11-yl)-1,2-eth a n ed ia m in e (191) (Sch em e 2). The
N-acetate 190 (2.1 g, 6.4 mmol) was dissolved in MeOH (50
mL), and 85% NH2NH2‚H2O (50 mL) was added. The mixture
was heated at 90 °C for 13 days. The mixture was evaporated
to dryness and azeotroped with toluene. The product was
chromatographed on silica gel using 4% (10% concentrated
NH4OH in MeOH)-CH2Cl2 as the eluant to give 191 (2 g, 59%)
(Table 6).
N-[2-(8-Ch lor o-6,11-d ih yd r o-5H-ben zo[5,6]cycloh ep ta -
[1,2-b]p yr id in -11-yl)a m in oet h yl]-4-p yr id in ea cet a m id e
(192) (Sch em e 2). The amine 191 (935 mg, 3.2 mmol) was
reacted with 4-pyridylacetic acid (445.5 mg, 3.2 mmol) as
described in method A above to give 192 (661 mg, 50%) (Table
1).
1-[(4-Ch lor op h en yl)-2-p yr id in ylm et h yl]-4-(3-p yr id in -
yla cetyl)p ip er a zin e (196) (Sch em e 3). The carbinol 193
(3 g, 13.7 mmol) was reacted with thionyl chloride (1.846 mL,
25.3 mmol) under essentially the same conditions as described
earlier11 to give the crude chloro derivative 194. The material
was dissolved in THF (33 mL), and piperazine (5.7 g, 66.4
mmol) in THF (33 mL) was added. The mixture was stirred
at 25 °C for 18 h and then worked up in the usual way.
Chromatography on silica gel using 1-7% (10% concentrated
NH4OH in MeOH)-CH2Cl2 as the eluant gave 195 (252.1 mg,
6%): CIMS m/z 288 (MH+); 13C NMR δC (CH2) 52.8, 52.8, 45.8,
45.8; (CH) 149.2, 136.6, 129.4, 129.4, 128.5, 128.5, 122.0, 122.0,
77.5; (C) 161.2, 139.3, 132.8; HRFABMS calcd for C16H19ClN3
Mr 288.1268 (MH+), found 288.1265.
gel using 2.5% (10% concentrated NH4OH in MeOH)-CH2Cl2
as the eluant to give 201 (182.5 mg, 80%) (Table 1).
1-(4-Ch lor ob en zoyl)-4-(4-p yr id in yla cet yl)p ip er a zin e
(205) (Sch em e 5). 1-N-Boc-piperazine 202 (1 g, 5.37 mmol)
was reacted with 4-pyridylacetic acid (735.5 mg, 5.37 mmol)
as described in method A above. The product was chromato-
graphed on silica gel using 2% (10% concentrated NH4OH in
MeOH)-CH2Cl2 as the eluant to give 203 (1.28 g, 80%): CIMS
m/z 306 (MH+); δC (CH3) 28.1; (CH2) 45.7, 45.7, 43.4, 41.5, 39.9;
(CH) 150.4, 150.4, 124.2, 124.2; (C) 168.3, 154.7, 144.0, 80.4.
Anal. (C16H23N3O3) C, H, N.
Using the same procedure as described above for 112, the
N-Boc derivative 203 (1.23 g, 0.4 mmol) was deprotected, and
the product was chromatographed on silica gel using 3-4%
(10% concentrated NH4OH in MeOH)-CH2Cl2 as the eluant
to give 204 (322.1 mg, 39%): CIMS m/z 206 (MH+); δC (CH2)
47.1, 46.0, 45.5, 42.8, 39.8; (CH) 150.2, 150.2, 124.3, 124.3; (C)
168.1, 144.4. Anal. (C11H15N3O‚0.3H2O) C, H, N.
The piperazine 204 (90 mg, 0.4 mmol) was dissolved in THF
(3 mL), and 4-chlorobenzoyl chloride (0.0557 mL, 0.4 mmol)
was added. The mixture was stirred at 25 °C for 1 h and then
worked up in the usual way. Chromatography on silica gel
using 1-3% (10% concentrated NH4OH in MeOH)-CH2Cl2 as
the eluant gave 205 (134.6 mg, 89%) (Table 1).
1,1-Dim eth yleth yl 2-[1-(3-Br om o-8-ch lor o-6,11-dih ydr o-
5H-ben zo[5,6]cycloh ep ta [1,2-b]p yr id in -11-yl)-4-p ip er a zi-
n y l ]-1 (R )-[[(t r i p h e n y l m e t h y l )t h i o ]m e t h y l ]e t h y l -
ca r ba m a te Isom er s (187 a n d 188). The piperazine 5b (300
mg, 0.76 mmol), powdered 3 Å molecular sieves (200 mg), and
NaBH(OAc)3 (655 mg, 3.04 mmol) were dissolved in dichloro-
ethane (7.8 mL), and the mixture was cooled to 0 °C. 2(R)-
(N-tert-butoxycarbonylamino)-3-triphenylmethylthiopropa-
nal (514.5 mg, 1.15 mmol) in dichloroethane (7.8 mL) was
added dropwise to the stirred solution, and the mixture was
allowed to warm to 25 °C, the reaction being allowed to run
for 21 h. The usual workup followed by chromatography on
silica gel using 3-3.5% acetone in CH2Cl2 gave the less polar
isomer 187 (210 mg, 33%) (Table 6) and the more polar isomer
188 (210 mg, 33%) (Table 6).
4-(3-B r o m o -8-c h lo r o -6,11-d ih y d r o -5H -b e n zo [5,6]-
cycloh ep ta [1,2-b]p yr id in -11-yl)-r(R)-(m er ca p tom eth yl)-
1-p ip er a zin eeth a n a m in e Hyd r och lor id e (189). The com-
bined isomers 187 (384.1 mg) and 188 (640 mg) (total: 1.02 g,
1.2 mmol) were dissolved in CH2Cl2 (10.24 mL), and Et3SiH
(589 mg/0.809 mL, 5.1 mmol) and TFA (1.544 g/1.043 mL, 13.5
mmol) were added.22 The mixture was stirred at 25 °C for 1
h and then evaporated to dryness. The solid was partitioned
between water and hexane, the aqueous layer was passed over
BioRad AG3-X4(Cl-) resin, and the eluate was lyophilized to
give the hydrochloride 189 (530.3 mg, 74%) (Table 1).
The piperazine 195 (250 mg, 0.869 mmol) was reacted with
3-pyridylacetic acid (119 mg, 0.869 mmol) as described in
method A above. The product was chromatographed on silica
gel using 0.25% (10% concentrated NH4OH in MeOH)-CH2-
Cl2 as the eluant to give 196 (261.1 mg, 74%) (Table 1).
1-[(4-Ch lor op h en yl)-4-p yr id in ylm eth yl]-4-(3-p yr id in y-
la cetyl)p ip er a zin e (201) (Sch em e 4). Using essentially the
same procedures as described earlier,11 the ketone 197 (15 g,
68.9 mmol) was reduced to give the carbinol 198 (13.85 g,
91%): CIMS m/z 220 (MH+); δC (CH) 149.1, 149.1, 128.9, 128.9,
128.2, 128.2, 121.7, 121.7, 73.6; (C) 154.0, 141.6, 133.8. Anal.
(C12H10ClNO) C, H, Cl, N. The carbinol 198 (3 g, 13.7 mmol)
was reacted with thionyl chloride (1.84 mL, 25.3 mmol) to give
the chloro derivative 199. The latter was dissolved in THF
(30 mL) and CH2Cl2 (10 mL) and reacted with piperazine (3.53
g, 41.1 mmol) to give, after chromatography on silica gel using
15% EtOAc in hexane as the eluant, the piperazine derivative
200 (679.6 mg, 17%): CIMS m/z 288.25 (MH+); δC (CH2) 52.4,
52.4, 45.8, 45.8; (CH) 150.5, 150.5, 129.6, 129.6, 129.2, 129.2,
123.1, 123.1, 74.8; (C) 151.3, 139.3, 133.7; HRFABMS calcd
for C16H19ClN3 Mr 288.1268 (MH+), found 288.1265.
H-Ra s P r ocessin g in Cos Mon k ey Kid n ey Cells. Pro-
cessing of H-ras in Cos cells was determined as described
previously.7 Briefly, the mammalian expression vector pSV-
Sport (BRL-GIBCO) containing a cDNA encoding the Val12
-
activated form of H-Ras was electroporated into Cos-7 monkey
kidney cells. Cells were then plated into six-well tissue culture
dishes containing 1.5 mL of Dulbecco’s modified Eagle’s
Medium (DMEM) supplemented with 10% fetal calf serum and
the appropriate FPT inhibitors. After 24 h, media was
removed, and fresh media containing the appropriate drugs
was re-added. After an additional 24 h incubation, cells were
washed, harvested, and lysed by homogenization. Debris was
removed by centrifugation. Protein was precipitated using cold
trichloroacetic acid, redissolved in SDS-polyacrylamide gel
electrophoresis sample buffer and electrophoresed on 14%
polyacrylamide gels (Novex, Inc.). Resolved proteins were
electroblotted onto nitrocellulose membranes. Nonspecific
binding to membranes was blocked by preincubation in buffer
containing 2.5% dried milk and 0.5% Tween-20. Ras was
detected by incubation with a Ras-specific monoclonal antibody
(Y13-259; Oncogene Science) in buffer containing 1% fetal calf
serum for 1 h at room temperature. Membranes were then
incubated for 1 h with a 1:5000 dilution of rabbit anti-rat IgG
conjugated to horseradish peroxidase in buffer containing 1%
The piperazine 200 (160.7 mg, 0.56 mmol) was reacted with
3-pyridylacetic acid (76.5 mg, 0.56 mmol) as described in
method A above. The product was chromatographed on silica

892 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6
Mallams et al.
boxamido Derivatives of 1-(3-Bromo-8-chloro-6,11-dihydro-5H-
benzo[5,6]cyclohepta[1,2-b]pyridin-11-yl)piperazine. 214th Na-
tional ACS Meeting, Las Vegas, NV, September 7-11, 1997,
Poster MEDI 189.
fetal calf serum. Processed and unprocessed Ras proteins,
which migrate with distinct electrophoretic mobility, were
visualized using a colorimetric peroxidase reagent.
An ch or a ge-In d ep en d en t Soft Aga r Gr ow th Assa ys.
Logarithmically growing H- or K-ras-transformed mouse NIH-
3T3 fibroblasts were trypsinized and plated in the top layer
of 0.35% low-melting point agarose (LMP) at 10 000 cells/well
of a six-well cluster dish in DMEM (GIBCO) containing 10%
fetal calf serum (FBS) and the appropriate amount of FPT
inhibitor. Bottom layers (0.6% LMP) contained the same
concentration of inhibitor. After 14 days, the clones were
stained with MTT (1 mg/mL in PBS) and then photo-
graphed.The number of soft agar colonies was counted and
inhibition was calculated relative to the number of colonies
in vehicle-treated control wells. When more than one experi-
ment was run, both values are given. The results are sum-
marized below.
(2) Barbacid, M. ras Genes. In Annual Review of Biochemistry;
Richardson, C., Ed.; Annual Reviews Inc.: Palo Alto, CA, 1987;
Vol. 56, pp 779-827.
(3) Hall, A. A Biochemical Function for RassAt Last. Science 1994,
264, 1413-1414.
(4) Casey, P. J .; Solski, P. A.; Der, C. J .; Buss, J . E. p21 is Modified
by a Farnesyl Isoprenoid. Proc. Natl. Acad. Sci. U.S.A. 1989,
86, 8323-8327.
(5) Kato, K.; Cox, A. D.; Hisaka, M. M.; Graham, S. M.; Buss, J . E.;
Der, C. J . Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 6403-6407.
(6) Graham, S. L. Inhibitors of Protein Farnesylation:
a New
Approach to Cancer Chemotherapy. Exp. Opin. Ther. Pat. 1995,
5, 1269-1285.
(7) Bishop, W. R.; Bond, R.; Petrin, J .; Wang, L.; Patton, R.; Doll,
R.; Njoroge, G.; Catino, J .; Schwartz, J .; Windsor, W.; Syto, R.;
Schwartz, J .; Carr, D.; J ames, L.; Kirschmeier, P. Novel Tricyclic
Inhibitors of Farnesyl Protein Transferase. J . Biol. Chem. 1995,
270, 30611-30618.
(8) Njoroge, F. G.; Doll, R. J .; Vibulbhan, B.; Alvarez, C. S.; Bishop,
W. R.; Petrin, J .; Kirschmeier; P.; Carruthers, N. I.; Wong, J .
K.; Albanese, M. M.; Piwinski, J . J .; Catino, J .; Girijavallabhan,
V.; Ganguly, A. K. Discovery of Novel Non-peptide Tricyclic
Inhibitors of Ras Farnesyl Protein Transferase. Bioorg. Med.
Chem. 1997, 5, 101-113.
IC50 (µM)
compound
NIH-H
NIH-K
DLD-1
83b (micronized free base)
85b (micronized free base)
85b (hydrochloride salt)
1.2
1.5
4
6.5
8/8
3/7
(9) Mallams, A. K.; Njoroge, F. G.; Doll, R. J .; Snow, M. E.;
Kaminski, J . J .; Rossman, R. R.; Vibulbhan, B.; Bishop, W. R.;
Kirschmeier, P.; Liu, M.; Bryant, M. S.; Alvarez, C.; Carr, D.;
J ames, L.; King, I.; Li, Z.; Lin, C.-C.; Nardo, C.; Petrin, J .;
Remiszewski, S.; Taveras, A. G.; Wang, S.; Wong, J .; Catino, J .;
Girijavallabhan, V.; Ganguly, A. K. Antitumor 8-Chlorobenzo-
cycloheptapyridines: a New Class of Selective, Non-Peptidic,
Non-Sulfhydryl Inhibitors of Ras Farnesylation. Bioorg. Med.
Chem. 1997, 5, 93-99.
(10) (a) Njoroge, F. G.; Vibulbhan, B.; Rane, D. F.; Bishop, W. R.;
Petrin, J .; Patton, R.; Bryant, M. S.; Chen, K.-J .; Nomeir, A. A.;
Lin, C.-C.; Liu, M.; King, I.; Chen, J .; Lee, S.; Yaremko, B.; Dell,
J .; Lipari, P.; Malkowski, M.; Li, Z.; Catino, J .; Doll, R. J .;
Girijavallabhan, V.; Ganguly, A. K. Structure-Activity Relation-
ship of 3-Substituted N-Pyridinylacetyl-4-(8-chloro-5,6-dihydro-
11H-benzo[5,6]pyridinyl-11-ylidene)piperidine Inhibitors of Far-
nesyl-Protein Transferase: Design and Synthesis of in Vivo
Active Antitumor Compound. J . Med. Chem. 1997, 40, 4290-
4301. (b) Njoroge, F. G.; Vibulbhan, B.; Pinto, P.; Chan, T.-M.;
Osterman, R.; Remiszewski, S.; del Rosario, J .; Doll, R.; Giri-
javallabhan, V.; Ganguly, A. K. Highly Regioselective Nitration
Reactions Provide a Versatile Method of Functionalizing Ben-
zocycloheptapyridine Tricyclic Ring Systems: Application To-
wards Preparation of Nanomolar Inhibitors of Farnesyl Protein
Transferase. J . Org. Chem., in press (J O971100Z).
In Vivo An titu m or Effica cy in Mice (F igu r e 2). Nude
mice were injected sc with the appropriate tumor cells (m ×
106: K-ras DLD-1, m ) 5; K-ras SW-620, m ) 7; H-ras CVLS,
m ) 2; H-ras CVLL, m ) 2; H-ras pT-24, m ) 5) containing a
mutated ras oncogene. Drug treatment was initiated and the
animals (10 each per control and treatment group) were dosed
with the compounds indicated in Figure 2 at 50 and 10 mg/
kg, po, qid for 7 days a week using a solution of the
hydrochloride salt in 20% HPâCD, except for 39, 45a , 50a ,
228a , and 231, which were administered to animals (five each
per control and treatment group) at a dose of 100 mg/kg, po,
qid for 5 days a week using a suspension of the micronized
free base in corn oil. Primary tumor measurements were
determined twice a week for a period of 3-4 weeks. Tumor
volume ) 1/2length × width × height. Efficacy is indicated by
comparing median tumor size of the treated group with that
of the vehicle control group. In the experiment with 150b at
50 mg/kg using DLD-1 tumor cells, 3/10 mice died.
Ack n ow led gm en t. The authors thank their col-
leagues in the Physical and Analytical Group for provid-
ing the spectral and analytical data. We also thank
Vilma Ruperto and Robert McQuade for providing the
muscarinic data.
(11) Piwinski, J . J .; Green, M. J .; Wong, J . U.S. Patent 5,422,351,
J une 6, 1995.
(12) Villani, F. J .; Daniels, P. J . L.; Ellis, C. A.; Mann, T. A.; Wang,
K.-C. Derivatives of 10,11-Dihydro-5H-dibenzo[a,d]cycloheptene
and Related Compounds. III. Azaketones. J . Heterocycl. Chem.
1971, 8, 73-81.
(13) Villani, F. J .; Daniels, P. J . L.; Ellis, C. A.; Mann, T. A.; Wang,
K.-C.; Wefer, E. A. J . Med. Chem. 1972, 15, 750-754.
(14) Wong, J . K.; Piwinski, J . J .; Green, M. J .; Ganguly, A. K.; Anthes,
J . C.; Billah, M. M. Dual Antagonists of Platelet Activating
Factor and Histamine. 2. Pyridine Ring Substitution of N-Acetyl-
4-(8-chloro-5,6-dihydro-11H-benzo[5,6]cyclohepta[1,2-b]pyridin-
11-ylidene)piperidines. Bioorg. Med. Chem. Lett. 1993, 3, 1073-
1078.
Su p p or tin g In for m a tion Ava ila ble: Additional experi-
mental procedures, structures, Schemes 6-15, Tables 3-7,
physical chemical data (yields, analytical data, mass spectral
data, CMR data, specific rotations), antitumor activity, and
statistics (for in vitro and in vivo efficacy studies) for all
synthetic intermediates and final products described in this
study (27 pages). Ordering information is given on any current
masthead page.
(15) Sakamoto, T.; Kondo, Y.; Masumoto, K.; Yamanaka, H. Synthesis
of Methyl 2-(Heteroaryl)propanoates via Palladium-catalyzed
Reaction. Heterocycles 1993, 36, 2509-2512.
(16) Brewster, A. G.; Brown, G. R.; J essup, R.; Smithers, M. J .
Preparation of 2-Pyridylalkyl-4-phenyl-1,3-dioxane Derivatives
as TXA2 Antagonists and Inhibitors of TXA2 Synthase. EP-
288279 A2, October 26, 1988.
(17) Bradford, L.; Elliott, T. J .; Rowe, F. M. The Skraup Reaction
with m-Substituted Anilines. J . Chem. Soc. 1947, 437-445.
(18) Remiszewski, S. W.; del Rosario, J . D.; Doll, R. J .; Bishop, W.
R.; Carr, D.; J ames, L.; Kirschmeier, P.; Petrin, J . Unpublished
observations.
(19) del Rosario, J . D.; Remiszewski, S. W.; Bishop, W. R.; Doll, R.
J .; J ames, L.; Kirschmeier, P.; Nafissi, M.; Njoroge, F. G.; Petrin,
J .; Pinto, P. Benzocycloheptapyridine Farnesyl Protein Trans-
ferase Inhibitors: Effects of Substitution at Position 5 and 6 on
Enzyme Inhibition. 213th National ACS Meeting, San Francisco,
CA, April 13-17, 1997, Poster MEDI 198.
(20) Taveras, A. G.; Cesarz, D. Unpublished observations.
(21) Kiely, J . S.; Priebe, S. R. An Improved Synthesis of (R) and (S)-
2-Methylpiperazine. Org. Prep. Proc. Int. 1990, 22, 761-768.
Refer en ces
(1) (a) Rossman, R. R.; Mallams, A. K.; Doll, R. J ., Girijavallabhan,
V.; Ganguly, A. K.; Petrin, J .; Wang, L.; Patton, R.; Bishop, W.
R.; Kirschmeier, P.; Catino, J .; Bryant, M. S.; Chen, K.-J .;
Korfmacher, W. A.; Nardo, C.; Wang, S.; Nomeir, A. A.; Lin, C.-
C.; Liu, M.; Li, Z.; Chen, S.; King, I. Inhibitors of Farnesyl
Protein Transferase. Part 1. 4-Amido, 4-Carbamoyl and 4-Car-
boxamido Derivatives of 1-(8-Chloro-6,11-dihydro-5H-benzo[5,6]-
cyclohepta[1,2-b]pyridin-11-yl)piperazine. 213th National ACS
Meeting, San Francisco, CA, April 13-17, 1997, Poster MEDI
199. (b) Mallams, A. K.; Rossman, R. R., Doll, R. J ., Girijavall-
abhan, V.; Ganguly, A. K.; Petrin, J .; Wang, L.; Patton, R.;
Bishop, W. R.; Kirschmeier, P.; Catino, J .; Bryant, M. S.; Chen,
K.-J .; Korfmacher, W. A.; Nardo, C.; Wang, S.; Nomeir, A. A.;
Lin, C.-C.; Liu, M.; Li, Z.; Chen, S.; King, I. Inhibitors of Farnesyl
Protein Transferase. Part 2. 4-Amido, 4-Carbamoyl and 4-Car-

Inhibitors of Farnesyl Protein Transferase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 6 893
(22) Graham, S. L.; Williams, T. M. Inhibitors of Farnesyl-Protein
Transferase. WO 95/00497, J anuary 5, 1995.
(36) Pavia, M. R.; Lobbestael, S. J .; Taylor, C. P.; Hershenson, F.
M.; Miskell, D. L. N-Phenyl-N′-pyridinylureas as Anticonvulsant
Agents. J . Med. Chem. 1990, 33, 854-861.
(23) Williams, T. M.; Ciccarone, T. M.; MacTough, S. C.; Bock, R. L.;
Conner, M. W.; Davide, J . P.; Hamilton, K.; Koblan, K. S.; Kohl.
N. E.; Kral, A. M.; Mosser, S. D.; Omer, C. A.; Pompliano, D. L.;
Rands, E.; Schaber, M. D.; Shah, D.; Wilson, F. R.; Gibbs, J . B.;
Graham, S. L.; Hartman, G. D.; Oliff, A. I.; Smith, R. L.
2-Substituted Piperazines as Constrained Amino Acids. Applica-
tion to the Synthesis of Potent, Non Carboxylic Acid Inhibitors
of Farnesyltransferase. J . Med. Chem. 1996, 39, 1345-1348.
(24) Duncan, R. L., J r.; Helsley, G. C.; Welstead, W. J ., J r.; DaVanzo,
J . P.; Funderburk, W. H.; Lunsford, C. D. Aroylpiperidines and
Pyrrolidines. A New Class of Potent Central Nervous System
Depressants. J . Med. Chem. 1970, 13, 1-6.
(25) Crider, A. M.; Tita, T. T.; Wood, J . D.; Hinko, C. N. Esters of
Nipecotic and Isonipecotic Acids as Potential Anticonvulsants.
J . Pharm. Sci. 1982, 71, 1214-1219.
(26) Rowley, M.; Broughton, H. B.; Collins, I.; Baker, R.; Emms, F.;
Marwood, R.; Patel, S.; Patel, S.; Ragan, C. I.; Freedman, S. B.;
Leeson, P. D. 5-(4-Chlorophenyl)-4-methyl-3-(1-(2-phenylethyl)-
piperidin-4-yl)isoxazole: A Potent, Selective Antagonist at Hu-
man Cloned Dopamine D4 Receptors. J . Med. Chem. 1996, 39,
1943-1945.
(37) Park, H.-W.; Boduluri, S. R.; Moomaw, J . F.; Casey, P. J .; Beese,
L. S. Science 1997, 275, 1800-1804.
(38) Strickland, C. Structural Studies of Farnesyl Protein Trans-
ferase. 214th National ACS Meeting, Las Vegas, NV, September
7-11, 1997, COMP 185.
(39) Sepp-Lorenzino, L.; Ma, Z.; Rands, E.; Kohl, N. E.; Gibbs, J . B.;
Oliff, A.; Rosen, N. A Peptidomimetic Inhibitor of Farnesyl:
Protein Transferase Blocks the Anchorage-dependent and -in-
dependent Growth of Human Tumor Cell Lines. Cancer Res.
1995, 55, 5302-5309.
(40) Nagasu, T.; Yoshimatsu, K.; Rowell, C.; Lewis, M. D.; Garcia,
A. M. Inhibition of Human Tumor Xenograft Growth by Treat-
ment with the Farnesyl Transferase Inhibitor B956. Cancer Res.
1995, 55, 5310-5314.
(41) Liu, M.; Chen, P.; Lee, S.; Li, Z.; Bishop, W. R.; Kirschmeier, P.;
Doll, R. J .; Mallams, A. K.; Rossman, R. R.; Dell, J .; Lipari, P.;
Malkowski, M.; Prioli, N.; Catino, J . J . SCH 59228, a Novel
Tricyclic Inhibitor of Farnesyl Protein Transferase, Potently
Blocks the Growth of Ras Transformed Fibroblasts and Human
Tumor Cell Lines in Nude Mice. Proceedings of the American
Assoc. Cancer Res. 1997, 38, Abstract 13357.
(42) Liu, M.; Bryant, M. S.; Chen, J .; Lee, S.; Yaremko, B.; Li, Z.;
Dell, J .; Lipari, P.; Malkowski, M.; Prioli, N.; Rossman, R. R.;
Korfmacher, W. A.; Nomeir, A. A.; Lin, C.-C.; Mallams, A. K.;
Kirschmeier, P.; Doll, R. J .; Catino, J . J .; Girijavallabhan, V.;
Bishop, W. R. Effects of Sch 59228, an Orally Bioavailable
Farnesyl Protein Transferase Inhibitor, on the Growth of On-
cogene-transformed Fibroblasts and a Human Colon Carcinoma
Xenograft in Nude Mice. Cancer Chemother. Pharmacol., manu-
script submitted.
(43) Whyte, D. B.; Kirschmeier, P.; Hockenberry, T. N.; Nunez-Oliva,
I.; J ames, L.; Catino, J . J .; Bishop, W. R.; Pai, J .-K. K- and N-Ras
are Geranylgeranylated in Cells Treated with Farnesyl Protein
Transferase Inhibitors. J . Biol. Chem. 1997, 272, 14459-14464.
(44) Rowell, C. A.; Kowalczyk, J . J .; Lewis, M. D.; Garcia, A. M. Direct
Demonstration of Geranylgeranylation and Farnesylation of Ki-
Ras in Vivo. J . Biol. Chem. 1997, 272, 14093-14097.
(45) Zhang, F. L.; Kirschmeier, P.; Carr, D.; J ames, L.; Bond, R. W.;
Wang, L.; Patton, R.; Windsor, W. T.; Syto, R.; Zhang, R.; Bishop,
W. R. Characterization of Ha-Ras, N-Ras, Ki-Ras4A, and Ki-
Ras4B as in Vitro Substrates for Farnesyl Protein Transferase
and Geranylgeranyl Protein Transferase Type 1. J . Biol. Chem.
1997, 272, 10232-10239.
(27) Zalucky, T. B.; Malspeis, L.; Patel, H.; Hite, G. J . Pharm. Sci.
1965, 54, 687-693.
(28) Nishi, T.; Komatsu, M.; Koga, Y.; Shu, Y., Tamura, K. J P
3148223 A2, J une 25, 1991.
(29) Pizzorno, M. T.; Albonico, S. M. Novel Synthesis of 5,6,7,8-
Tetrahydroindolizines. J . Org. Chem. 1977, 42, 909-910.
(30) Ishihara, Y.; Miyamoto, M.; Nakayama, T.; Goto, G. Central
Colinergic Agents. IV. Synthesis and Acetylcholinesterase In-
hibitory Activities of ω-[N-Ethyl-N-(phenylmethyl)amino]-1-
phenyl-1-alkanones and Their Analogues with Partial Confor-
mational Restriction. Chem. Pharm. Bull. 1993, 41, 529-538.
(31) Egbertson, M. S.; Chang, C. T.-C.; Duggan, M. E.; Gould, R. J .;
Halczenko, W.; Hartman, G. D.; Laswell, W. L.; Lynch, J . J .,
J r.; Lynch, R. J .; Manno, P. D.; Naylor, A. M.; Prugh, J . D.;
Ramjit, D. R.; Sitko, G. R.; Smith, R. S.; Turchi, L. M.; Zhang,
G. Non-Peptide Fibrinogen Receptor Antagonists. 2. Optimiza-
tion of a Tyrosine Template as a Mimic for Arg-Gly-Asp. J . Med.
Chem. 1994, 37, 2537-2551.
(32) Honel, M.; Vierhapper, F. W. Selectivity in the Hydrogenation
of 6- and 8-Substituted-quinolines. J . Chem. Soc., Perkin Trans.
1 1980, 1933-1939.
(33) Engel, W.; Trummlitz, G.; Schmidt, G.; Eberlein, W.; Hammer,
R.; Del Soldato, P. Substituted Dibenzodiazepinones and Drugs
Containing Them. DE 3204158, August 11, 1983.
(34) Ford, M. J .; Ley, S. V. A Simple, One-Pot, Glycosidation
Procedure via (1-Imidazolylcarbonyl) Glycosides and Zinc Bro-
mide. Synlett 1990, No. 5, 255-256.
(35) Ninomiya, K.; Shioiri, T.; Yamada, S. Phosphorus in Organic
Synthesis-VII. Diphenyl phosphorazidate (DPPA). A New Con-
venient Reagent for a Modified Curtius Reaction. Tetrahedron
1974, 30, 2151-2157.
(46) Lebowitz, P. F.; Davide, J . P.; Prendergast, G. C. Evidence That
Farnesyl Transferase Inhibitors Suppress Ras Transformation
by Interfering With Rho Activity. Mol. Cell Biol. 1995, 15,
6613-6622.
J M970462W