Design and synthesis of o-trifluoromethylbiphenyl substituted 2-amino-nicotinonitriles as inhibitors of farnesyltransferase

Article abstract of DOI:10.1016/j.bmcl.2004.10.014

A non-methionine FT inhibitor lead structure (1) was designed through computer modeling of the peptidomimetic FT inhibitor ABT839. Optimization of this lead resulted in compounds 2e and 2g, with FT IC₅₀ values of 1.3 and 1.8 nM, GGT IC₅₀ of 1400 nM, and EC₅₀ (Ras processing) values of 13 and 11 nM, respectively.

Full text of DOI:10.1016/j.bmcl.2004.10.014

Bioorganic & Medicinal Chemistry Letters 15 (2005) 153–158
Design and synthesis of o-trifluoromethylbiphenyl substituted
2-amino-nicotinonitriles as inhibitors of farnesyltransferase
Gary T. Wang,* Xilu Wang, Weibo Wang, Lisa A. Hasvold, Gerry Sullivan,
Charles W. Hutchins, Steve OꢀConner, Robert Gentiles, Thomas Sowin, Jerry Cohen,
Wen-Zhen Gu, Haiying Zhang, Saul H. Rosenberg and Hing L. Sham
Global Pharmaceutical Research & Development, Abbott Laboratories, Abbott Park, IL 60064, USA
Received 2 September 2004; revised 5 October 2004; accepted 6 October 2004
Available online 28 October 2004
Abstract—A non-methionine FT inhibitor lead structure (1) was designed through computer modeling of the peptidomimetic FT
inhibitor ABT839. Optimization of this lead resulted in compounds 2e and 2g, with FT IC50 values of 1.3 and 1.8nM, GGT
IC50 of 1400nM, and EC50 (Ras processing) values of 13 and 11nM, respectively.
Ó 2004 Elsevier Ltd. All rights reserved.
Inhibition of farnesyltransferase (FT) has been vigor-
ously pursued as a promising target for the treatment
of a broad spectrum of cancers, and several compounds,
such as SCH66336 (SarasarTM) and R115777 (tipifarnib
or ZarnestraTM), are currently in advanced human clini-
cal trials.1–3 FT catalyzes the transfer of a farnesyl moi-
ety from farnesyl pyrophosphate to a cysteine residue
found in the tetrapeptide sequence CAAX (C = Cys,
A = an aliphatic amino acid, X is typically Met) in the
carboxyl terminal of a group of membrane-bound small
G-proteins such as Ras, RhoB, RhoE, lamin A and B,
and transducin. This post-translational processing is
essential for the signal transduction function of these
proteins since the farnesyl moiety is required to anchor
these proteins to the cell membrane. In the early days,
it was believed that farnesylation of Ras protein was
the therapeutic target, since oncogenic mutations of
Ras have been identified in ꢀ30% of all human malig-
nancies and these mutations result in constitutively
active signaling and uncontrolled cell proliferation.4
However, recent data from several pharmacodynamics
studies have revealed a disconnection between Ras farnes-
ylation and anti-neoplastic efficacy, suggesting that
Ras farnesylation should be regarded as a surrogate tar-
get or a biomarker at the best.5–7 An enzyme closely re-
lated to FT is geranylgeranyl transferase (GGT) type I,
which transfers a geranylgeranyl moiety to the CAAX
sequence (X is typically Leu) of the substrate proteins.
Even though it was initially uncertain whether selective
inhibition of FT over GGT was necessary or, perhaps,
simultaneous inhibition of FT and GGT was desired, re-
cent in vivo studies using dual active FT–GGT inhibi-
tors have unequivocally established the necessity for
selective FT inhibitors, since the dual-active inhibitors
were proven to be acutely lethal.8,9
Our FT inhibition program initially resulted in the dis-
covery of ABT-839 (Fig. 1), a first generation clinical
candidate.10–12 ABT-839 is a potent FT inhibitor
(IC50 = 1.1nM) with good selectivity (GGT/FT =
75,000) and cellular activity (EC50 = 16nM in a Ras
processing assay in NIH 3T3 cells), but lacks oral bio-
availability. The poor oral bioavailability of ABT-839
was attributed to the methionine moiety. Toward the
goal of designing templates devoid of the methionine
moiety, the X-ray crystallographic structure of ABT-
839 complexed with FT was compared with the compu-
ter models of Janssenꢀs R115777 and Merckꢀs N-arylpip-
erazinone series of FT inhibitors.13,14 These computer-
modeling efforts led to the designing and subsequent
synthesis of compound 1 (Fig. 1) as our new lead struc-
ture.15 Compound 1, which retains the signature o-tolyl
biphenyl core of ABT-839 but has a cyano group on the
B ring replacing the acyl methionine moiety, showed FT
inhibition IC50 of 84nM, a modest selectivity (GGT
IC50 = 1300nM, GGT/FT = 16) and weak cellular activ-
ity (EC50 = 2700nM). In this paper, we report the first
part of our effort to optimize this lead, which resulted
Keyword: Farnesyltransferase inhibitors.
*
Corresponding author. Tel.: +1 8479372489; fax: +1 8479355165;
e-mail: gary.t.wang@abbott.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.014

154
G. T. Wang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 153–158
C
B
S
F3C
O
NC
CN
CN
A
N
H
CO2H
R1
O
O
N
N
N
N
R2
N
N
N
ABT-839
2
1
Figure 1.
in the discovery of a series of compounds 2 as potent
and selective FT inhibitors with potent activity in cellu-
lar assays.
CN
CN
ArCH2Br
Our attention was first focused on the optimization of
the etheral benzyl (A-ring) position. Thus, a library of
80 analogous ethers 4 was prepared from alcohol 3,15
by Ag2O-catalyzed alkylation with commercial available
benzyl bromides or benzyl iodides prepared in situ from
the benzyl chlorides and NaI (Scheme 1). These com-
pounds were prepared expeditiously using parallel syn-
thesis and were all HPLC-purified and fully characterized.
HO
N
Ar
O
N
Ag2O
CH2Cl2
N
3
N
4
Scheme 1.
The FT inhibition activity of selected 4 is shown in
Table 1. A variety of substituents on the benzyl group
was tolerated and result in augmented potency relative
to 1. However, electron-withdrawing substituents gener-
ally gave more potent inhibitors in comparison to com-
pounds bearing electron-donating groups (4a–j). For
mono-substituted compounds, there was a clear prefer-
ence for para-substitution, while ortho-substitution gave
Table 1. SAR of o-tolyl biphenyls 4
CN
Ar
O
N
N
4
Compd
Ar
FT IC50 (nM)a
Compd
Ar
FT IC50 (nM)a
1
Phenyl
o-Tolyl
m-Tolyl
84
54
4u
4v
o-Cl–phenyl
m-Cl–phenyl
48
19
4a
4b
4c
4d
4e
4f
61
54
4w
4x
p-Cl–phenyl
o-Br–phenyl
11
p-Tolyl
65%b
13
p-iPr–phenyl
p-tBu–phenyl
p-Phenyl–phenyl
1-Naphthyl
2-Naphthyl
8-Quinolyl
51
87
4y
4z
m-Br–phenyl
p-Br–phenyl
12
20
13
4aa
4ab
4ac
4ad
4ae
4af
4ag
4ah
4ai
4aj
m-I–phenyl
p-I–phenyl
4g
4h
4i
17
13
33
76%b
59
o-NO2–phenyl
p-NO2–phenyl
m-CO2CH3–phenyl
p-CO2CH3–phenyl
p-CH3SO2–phenyl
3,4-Cl,Cl–phenyl
3,5-Cl,Cl–phenyl
2,6-Cl,Cl–phenyl
7.5
61
4j
p-Methoxyphenyl
m-F3CO–phenyl
p-F3CO–phenyl
m-F3C–phenyl
p-F3C–phenyl
o-Cyano-phenyl
59
4k
4L
4m
4n
4o
8.3
16
6.1
38
8.1
12
27
19
62
81
N
Cl
Cl
4p
4q
m-Cyano-phenyl
p-Cyano-phenyl
18
4
4ak
4al
15
13
Cl
N
Cl

G. T. Wang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 153–158
155
Table 1 (continued)
Compd
Ar
FT IC50 (nM)a
70
Compd
Ar
FT IC50 (nM)a
29
S
4r
4s
o-F–phenyl
4am
Cl
N
N O
N
m-F–phenyl
p-F–phenyl
38
32
4an
4ao
56%b
50%b
O
4t
O
O
a Bovine FT used.
b Percentage of inhibition at 100nM.
the least potent inhibitor (e.g., compare 4o, 4p, and 4q;
or 4ac and 4ad). Accordingly, the p-cyano-benzyl ether
4q, p-nitro-benzyl ether 4ad and p-CO2CH3–benzyl ether
4af are among the most active compounds, being 10–20-
fold more potent than 1. Additional substitution did not
result in further enhancement in potency (4v vs 4ai, 4w
vs 4ah). For heterocyclic ethers, the pyridine ring of
4ak and 4al and thiazole ring of 4am were tolerated
while the oxadiazole ring (4an) and furan ring (4ao) gave
less active compounds.
2,4-position seemed to give inferior compounds (10f). In
contrast, 2,3-disubstitution, as in 10g, 10h, and 10j, gave
significantly enhanced activity. In particular, the 2-
naphthyl moiety of 10j led to a 20-fold improvement
in FT inhibition and nearly 4-fold improvement in cellu-
lar activity relative to 10a. However, this substitution
rendered 10j also a potent GGT inhibitor (IC50
=
18nM) with no improvement in the FT selectivity as
measured by the GGT/FT ratio.16–18
The ortho-trifluoromethyl substituted 10b stood out as
an attractive compound, with a 6-fold improvement in
FT inhibition and 5-fold improvement in cellular activ-
ity over 10a, as well as a 10-fold improvement in selec-
tivity (GGT/FT ratio of 387 for 10b vs 37.5 for 10a).
Consequently, we decided to combine the o-CF3–biphe-
nyl with other preferred benzyl (A-ring) substitutions
from Table 1. Thus, a set of bromo ethers 11 were pre-
pared from alcohol 8 by alkylation with a selected set of
benzylic bromides. Suzuki coupling of 11 with o-trifluoro-
methylbenzene boronic acid gave the desired biphenyls
12a–d (Scheme 3). Among these compounds (Table 3),
12d provided the best profile. Even though the FT
IC50 decreased by about 3-fold relative to 10b, 12d
showed similar (albeit insufficient) cellular activity and
further improvement in selectivity with a GGT IC50 of
1500nM and GGT/FT ratio of 555.
Having established that p-cyanobenzyl ether is optimal
for A ring, we turned to optimizing the C ring of the
biphenyl unit. As shown in Scheme 2, bromination of
5 gave bromide 6, which was converted to cyano alde-
hyde 7 through cyanidation via diazonium salt and DI-
BAL reduction. Addition of the anion generated from
1-methyl-2-triethylsilylimidazole to 7 gave alcohol 8,16
which was alkylated with p-cyanobenzyl bromide to
give bromide 9, the common intermediate for prepara-
tion of 10a–j (Table 2). Previously, during the optimiza-
tion of the ABT-839 series,10 it was discovered that the
mono-substitution at the ortho-position of the C-ring
was preferred. Therefore, only a relatively small set of
analogs 10 were prepared in the present study. Polar
groups at the ortho-position (10d and 10e) generally re-
sulted in less potent FT inhibitors. Disubstitution at the
Br
CN
Br
Br
CN
Ar
NH2
NH2
HO
N
a
b,c
d
H
EtO2C
EtO2C
Br
O
N
6
8
5
7
NC
CN
NC
CN
e
f
O
O
N
N
N
N
9
10
t
Scheme 2. Reagents and conditions: (a) Br2, DCM/pyridine, À10°C to rt, 24h, 38%. (b) (i) BF3ÆEt2O, BuONO, DCM; (ii) CuCN, NaCN, H2O/
toluene, 48%. (c) DIBAL, DCM, À100°C, 74%. (d) 1-Methyl-2-triethylsilyl-imidazole, t-BuLi, À78°C, 86%. (e) 4-Cyano-benzyl bromide, Ag2O,
DCM, rt, 24h, 58%. (f) ArB(OH)2, Pd(OAc)2, Ph3P, Na2CO3, n-PrOH, H2O, 100°C, 50–80%.

156
G. T. Wang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 153–158
Table 3. SAR of o-trifluoromethyl biphenyls 12
CN
Table 2. SAR of p-cyanobenzyl ethers 10
Ar
NC
CN
Ar
O
N
O
12
F3C
N
N
N
10
Compound AR
IC50 (nM)
EC50 (nM)
FTa GGTb Ras processingc
Compound AR
IC50 (nM)
EC50 (nM)
FTa GGTb Ras processingc
Cl
Cl
12a
1.7
72
NT
10a (4q)
4.8 180
420
CF3
12b
12c
6.1
4.1
160
120
>1000
100
F
F
10b
10c
10d
10e
0.8 310
85
CF3
CF3
CN
1.9 NT
5.5 480
110
CHO
OH
N
N
12d
12d
2.7
9
1500
3900
90
O
O
89% @ 1lm
>1000
83% @ 100nM
a Bovine FT used.
b Bovine GGT used.
14
NT
NT
85
CO2H
c H-ras processing in NIH 3T3 cells.
Cl
In order to further improve the cellular activity of 12d,
we sought to add amino substituents to the 3-cyano-6-
pyridyl ring, leading to structure 2. This idea was based
on the structure of SCH66336, which has a 1-piperidine-
carboxamide interacting with a remote binding pocket
of the FT active site.1,19 Superimposition of the compu-
ter model of a hypothetical (2-dimethylamino-3-cyano-
6-pyridyl)methyl ether with the X-ray structure of
SCH66336 suggested that the R1R2N-residue of 2 can
potentially occupy the same binding site as the terminal
piperidine moiety of SCH66336 (Fig. 2), offering an
opportunity to alter the physical properties of these
molecules.
10f
47
>1000
Cl
Cl
Cl
10g
10h
1.5
240
2.1 130
>1000
10i
10j
160
NT
18
NT
110
S
0.2
a Bovine FT used.
b Bovine GGT used.
c H-ras processing in NIH 3T3 cells.
CN
CN
a
b
8
Ar
O
N
Ar
O
N
Br
F3C
N
N
11
12
Scheme 3. Reagents and conditions: (a) ArCH2Br, Ag2O, DCM, rt,
24h, 40–60%. (b) o-F3CPhB(OH)2, Pd(OAc)2, Ph3P, Na2CO3, n-
PrOH, H2O, 100°C, 40–70%.
Figure 2. Superimposition of the computer model of a hypothetical
(2-dimethylamino-3-cyano-6-pyridyl)methyl ether (in purple) with the
X-ray structure of SCH66336 (in blue).19

G. T. Wang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 153–158
157
F3C
Br
CN
F3C
CN
a
c
b
CN
HO
N
H
H
O
O
N
14
7
13
F3C
F3C
NC
CN
NC
Cl
CN
d
R1
O
N
O
N
N
N
R2
N
N
N
2
15
Scheme 4. Reagents and conditions: (a) o-F3CPhB(OH)2, CsF, Pd(PPh3)4, DME, MeOH, 100°C, 73%. (b) 1-Methyl-2-triethylsilyl-imidazole, t-BuLi,
À78°C, 62%. (c) (3-Chloro-4-cyano-6-pyridyl)methyl bromide, Ag2O, DCM, rt, 24h, 40%. (d) R1R2NH, DMSO, 80°C, 5h. 20–60%.
Table 4. SAR of (2-amino-3-cyano-6-pyridyl)methyl ethers 2
The synthesis of 2 was shown in Scheme 4. Bromoalde-
hyde 7 reacted with o-trifluoromethylbenzene boronic
NC
CN
acid to give 13. Addition of the anion from 1-methyl-
2-triethylsilyl imidazole afforded alcohol 14. Alkylation
of 14 with 2-chloro-3-cyano-6-bromomethyl pyridine,
prepared from bromination of 2-chloro-3-cyano-6-
methyl pyridine, then furnished the key intermediate
15. Displacement of the chloride of 15 with a set of
amines was accomplished by heating in DMSO, leading
to the desired compounds 12a–n (Table 4).
R1
O
N
N
R2
N
F3C
N
2
Compd
R1R2N–
IC50 (nM)
EC50 (nM)
FTa
GGTb
Ras processingc
HO
2a
3.5
970
120
The data in Table 4 indicated that compounds 2 with an
aliphatic cyclic amino substitution (2a–j) all retained the
FT inhibitory potency of 12d, while compounds with
primary amino substituents (2k–n) were 2–3-fold less
potent. In terms of selectivity, the compounds with
six-membered ring substituents (morpholinyl of 2b,
piperidinyl of 2e and 2g, piperazinyl of 2c, 2i, and 2j)
were all more selective than the pyrrolidinyl-substituted
2a, 2d, and 2f. The compounds with six-membered ring
substituents also showed better cellular activity. Particu-
larly, compounds with a six-membered ring cyclic amino
substituents bearing a hydroxy group, such as 2e,g,j,
showed outstanding cellular activity. Most noticeable
is compound 2e, which is a 1.2nM FT inhibitor with a
selectivity over 1000-fold and 13nM cellular activity.
This is likely due to favorable cellular penetration of this
subset of compounds.
N
N
O
2b
2c
1.3
1.2
1200
2600
89
53
N
N
N
2d
2e
2.0
1.3
920
170
13
OH
HO
N
1400
N
2f
3.6
1.8
680
210
11
CO2H
N
2g
1400
In summary, we have succeeded in the transition from
ABT-839 to non-methionine FT inhibitors through
computer-modeling and medicinal chemistry efforts.
Starting from lead compound 1, placing electron-with-
drawing substituents such as cyano group at the para-
position of the A-ring increased the FT inhibition
potency by 15–20-fold. Replacing the o-tolyl of the C-ring
with bicyclic aryl groups such as 1-naphthyl resulted in
very potent FT inhibitors but with very poor selectivity,
indicating very similar volume in this part of the active
HO
H
N
2h
2i
1.7
2.6
1600
1800
330
130
N
Ac
O
N
N
(continued on next page)

158
G. T. Wang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 153–158
9. Lobell, R. B.; Omer, C. A.; Abrams, M. T.; Bhimnath-
Table 4 (continued)
wala, H. G.; Brucker, M. J.; Buser, C. A.; Davide, J. P.;
deSolms, S. J.; Dinsmore, C. J.; Ellis-Hutchings, M. S.;
Kral, A. M.; Liu, D.; Lumma, W. C.; Machotka, S. V.;
Rands, E.; Williams, T. M.; Graham, S. L.; Hartman, G.
D.; Oliff, A. I.; Heimbrook, D. C.; Kohl, N. F. Cancer
Res. 2001, 61, 8758.
Compd R1R2N–
IC50 (nM)
FTa GGTb
EC50 (nM)
Ras processingc
N
2j
1.3
2000
30
N
HO
10. Augeri, D. J.; OꢀConner, S. J.; Janowick, D.; Szczepan-
kiewicz, B.; Sullivan, G.; Larsen, J.; Kalvin, D.; Cohen, J.;
Devine, E.; Zhang, H.; Cherian, S.; Saeed, B.; Ng, S.-C.;
Rosenberg, S. J. Med. Chem. 1998, 41, 4288.
N
2k
7.7
7.5
2200
2400
201
285
H
O
N
H
11. OꢀConner, S. J.; Barr, K. J.; Wang, L.; Sorensen, B. K.;
Tasker, A. S.; Sham, H. L.; Ng, S.-C.; Cohen, J.; Devine,
E.; Cherian, S.; Saeed, B.; Zhang, H.; Lee, J. Y.; Warner,
R.; Tahir, S.; Kovar, P.; Ewing, P.; Alder, J.; Mitten, M.;
Leal, J.; Marsh, K.; Bauch, J.; Hoffman, D. J.; Sebti, S.
M.; Rosenberg, S. H. J. Med. Chem. 1999, 42, 3701.
12. Henry, K. J., Jr.; Wasical, J.; Tasker, A. S.; Cohen, J.;
Ewing, P.; Mitten, M.; Larsen, J. J.; Kalvin, D. M.;
Swenson, R.; Ng, S.-C.; Saeed, B.; Cherian, S.; Sham, H.
L.; Rosenberg, S. H. J. Med. Chem. 1999, 42, 4844.
13. Venet, M. G.; Angibaud, P. R.; Muller, P.; Sanz, G. C.;
W.O. patent 97/21701, 1997.
2l
O
2m
9.1
2300
750
>1000
>1000
N
H
F
N
2n
10
H
a Bovine FT used.
b Bovine GGT used.
c H-ras processing in NIH 3T3 cells.
14. Williams, T. M.; Bergman, J. M.; Brashear, K.; Breslin,
M. J.; Dinsmore, C. J.; Hutchinson, J. H.; MacTough, S.
C.; Stump, C. A.; Wei, D. D.; Zartman, B.; Bogusky, M.
J.; Culberson, C.; Buser-Doepner, C.; Davide, J.; Green-
berg, I. B.; Hamilton, K. A.; Koblan, K. S.; Kohl, N. E.;
Liu, D.; Lobell, R. B.; Mosser, S. D.; OꢀNeill, T. J.; Rands,
E.; Schaber, M. D.; Wilson, F.; Senderak, E.; Motzel, S.
L.; Bibbs, J. B.; Graham, S. L.; Heimbrook, D. C.;
Hartman, G. D.; Oliff, A. I.; Huff, J. R. J. Med. Chem.
1999, 42, 3779.
sites of FT and GGT. Replacing the C-ring with o-tri-
fluoromethylphenyl represents an acceptable compro-
mise. Finally, changing the A-ring into 2-amino-3-
cyano-6-pyridyl resulted in further improvement in
selectivity and cellular activity, leading to compounds
such as 2e and 2g with desirable properties.
15. Hutchins, C., et al. Bioorg. Med. Chem. Lett., in
preparation.
References and notes
16. Wang, L.; Wang, G. T.; Wang, X.; Tong, Y.; Sullivan, G.;
Park, D.; Leonard, N. M.; Li, Q.; Cohen, J.; Gu, W.-Z.;
Zhang, H.; Bauch, J. L.; Jakob, C. G.; Hutchins, C. W.;
Stoll, V. S.; March, K.; Rosenberg, S. H.; Sham, H. L.;
Lin, N.-H. J. Med. Chem. 2004, 47, 612.
1. Bell, I. M. J. Med. Chem. 2004, 47, 1869.
2. Ayral-Kaloustian, S.; Salaski, E. Curr. Med. Chem. 2002,
9, 1003.
3. Adjei, A. A. Drugs Future 2000, 25, 1069.
4. Bos, J. L. Cancer Res. 1989, 49, 4682.
5. Lebowitz, P. F.; Prendergast, G. C. Oncogene 1998, 17,
1439.
6. Prendergast, G. C.; Davide, J. P.; deSolms, S. J.; Giuli-
anin, E. A.; Graham, S. L.; Gibbs, J. B.; Oliff, A.; Kohl,
N. E. Mol. Cell. Biol. 1994, 14, 4183.
7. Prendergast, G. C.; Rane, N. Expert Opin. Invest. Drugs
2001, 10, 2105.
8. deSolms, S. J.; Ciccarone, T. M.; MacTough, S. C.; Shaw,
A. W.; Buser, C. A.; Ellis-Hutchings, M.; Fernandes, C.;
Hamilton, K. A.; Huber, H. E.; Kohl, N. E.; Lobell, R. B.;
Robinson, R. G.; Tsou, N. N.; Graham, S. L.; Beese, L.
S.; Taylor, J. S. J. Med. Chem. 2003, 46, 2973.
17. We have undertaken an effort to further optimize 10j,
primarily by introducing nitrogen into various rings.
Wang, W.-B., et al., in preparation.
18. Hasvold, L. A.; Wang, W.-B.; Gwaltney, S. L.; Rockway,
T. W.; Nelson, L. T. J.; Mantei, R. A.; Fakhoury, S. A.;
Sullivan, G. M.; Li, Q.; Lin, N.-H.; Wang; Le; Zhang, H.;
Cohen, J.; Gu, W.-Z.; Marsh, K.; Bauch, J.; Rosenberg, S.
H.; Sham, H. L. Bioorg. Med. Chem. Lett. 2003, 13, 4001.
19. Strickland, C. L.; Weber, P. C.; Windsor, W. T.; Wu, Z.;
Le, H. V.; Albanese, M. M.; Alvarez, C. S.; Cesarz, D.;
Rosario, J.; Deskus, J.; Mallams, A. K.; Njorge, F. G.;
Piwinski, J. J.; Reminszewski, S.; Rossman, R. R.;
Taveras, A. G.; Vibulbhan, B.; Doll, R. J.; Girijavallab-
han, V. M.; Ganguly, A. K. J. Med. Chem. 1999, 42, 2125.