Novel potent BRAF inhibitors: Toward 1 nM compounds through optimization of the central phenyl ring

Article abstract of DOI:10.1021/jm900242c

BRAF, a serine/threonine specific protein kinase that is part of the MAPK pathway and acts as a downstream effector of RAS, is a potential therapeutic target in melanoma. We have developed a series of small-molecule BRAF inhibitors based on a 1H-imidazo[4,5-b]pyridine-2(3H)-one scaffold (ring A) as the hinge binding moiety and a number of substituted phenyl rings C that interact with the allosteric binding site. The introduction of various groups on the central phenyl ring B combined with appropriate A- and C-ring modifications afford very potent compounds that inhibit V^600EBRAF kinase activity in vitro and oncogenic BRAF signaling in melanoma cells. Substitution on the central phenyl ring of a 3-fluoro, a naphthyl, or a 3-thiomethyl group improves activity to yield compounds with an IC₅₀ of 1 nM for purified V ^600EBRAF and nanomolar activity in cells.

Full text of DOI:10.1021/jm900242c

J. Med. Chem. 2009, 52, 3881–3891
3881
Novel Potent BRAF Inhibitors: Toward 1 nM Compounds through Optimization of the Central
Phenyl Ring
Delphine Me´nard,† Ion Niculescu-Duvaz,† Harmen P. Dijkstra,† Dan Niculescu-Duvaz,† Bart M. J. M. Suijkerbuijk,†
Alfonso Zambon,† Arnaud Nourry,† Esteban Roman,† Lawrence Davies,† Helen A. Manne,† Frank Friedlos,† Ruth Kirk,‡
Steven Whittaker,‡ Adrian Gill,§ Richard D. Taylor,§ Richard Marais,‡ and Caroline J. Springer*,†
The Institute of Cancer Research, UK Centre for Cancer Therapeutics, 15 Cotswold Road, Sutton, Surrey SM2 5NG, United Kingdom,
The Institute of Cancer Research, UK Centre for Cell and Molecular Biology, 237 Fulham Road, London SW3 6JB, United Kingdom, and Astex
Therapeutics Ltd., 436 Cambridge Science Park, Cambridge CB4 0QA, United Kingdom
ReceiVed February 25, 2009
BRAF, a serine/threonine specific protein kinase that is part of the MAPK pathway and acts as a downstream
effector of RAS, is a potential therapeutic target in melanoma. We have developed a series of small-molecule
BRAF inhibitors based on a 1H-imidazo[4,5-b]pyridine-2(3H)-one scaffold (ring A) as the hinge binding moiety
and a number of substituted phenyl rings C that interact with the allosteric binding site. The introduction of
various groups on the central phenyl ring B combined with appropriate A- and C-ring modifications afford very
potent compounds that inhibit V600EBRAF kinase activity in vitro and oncogenic BRAF signaling in melanoma
cells. Substitution on the central phenyl ring of a 3-fluoro, a naphthyl, or a 3-thiomethyl group improves activity
to yield compounds with an IC50 of 1 nM for purified V600EBRAF and nanomolar activity in cells.
Introduction
The RAF family of protein kinases are serine/threonine
specific protein kinases that are key players in the mitogen
activated protein (MAP) kinase pathway. These proteins act
immediately downstream of Ras to conduct extracellular signals
from the cell membrane to the nucleus via a cascade of
phosphorylation events, thereby regulating cell growth, prolif-
eration, and differentiation in response to growth factors,
cytokines, and hormones.1 The v-Raf murine sarcoma viral
oncogene homologue B1 (BRAFa)2 is frequently mutated to
generate active forms in human cancer, with the most frequent
mutation occurring at position 600 as a glutamic acid for valine
Figure 1. Structure of the pyridoimidazolone BRAF inhibitor 1 and
substitution (V600E; 90%).3 V600EBRAF is ∼500-fold more
its proposed binding mode with BRAF.
active than the wild-type protein,4 and in cutaneous malignant
melanoma, these mutations occur with a frequency of 50-70%.
BRAF mutations are also present in a range of other human
BRAF, and PLX40329 and RAF265,10 which target V600EBRAF.
Some of these drugs have progressed to clinical evaluation.11,12
We have previously described the development of new
V600EBRAF inhibitors based on a disubstituted pyrazine
scaffold,13,14 and more recently we reported on a novel series
of BRAF inhibitors, exemplified by 1H-imidazo[4,5-b]pyridine-
2(3H)-one 1 (see Figure 1), which are based on an innovative
hinge-binding fragment.15 The proposed binding mode is shown
in Figure 1 and is based on an analogy to the crystal structure
of BRAF with sorafenib, which was previously resolved by us.4
We demonstrated good activity with these inhibitors against
purified V600EBRAF, but these compounds were only weakly
active at inhibiting cell proliferation, with GI50 values of over
1 µM.15 The compounds we previously reported contain an
unsubstituted phenyl ring B. To improve the potency of these
compounds on both the isolated enzyme and on cell lines driven
by oncogenic BRAF, we anticipated that appropriate function-
alization of the phenyl ring B would lead to the desired gain in
potency in cells. Assuming a binding mode similar to that of
sorafenib, an examination of this cocrystal structure revealed a
space in the lipophilic pocket adjacent to the adenine binding
site formed in the DFG-out conformation (BP-I pocket),16 which
would be available to accommodate these ring B substitutions.
These modifications might simultaneously provide a potential
means of increasing the selectivity of inhibitors toward mutant
BRAF (see Table 1).
cancers, particularly thyroid (30%), colorectal (10%), and
ovarian (35%) cancers.3 Importantly, inhibition of BRAF
signaling blocks cancer cell proliferation and induces apoptosis,
validating V600EBRAF as a therapeutic target5 and offering
opportunities for anticancer drug development.
The interest in developing new drugs targeting the MAP
kinase pathway has increased considerably, and inhibitors are
currently being tested in preclinical and clinical trials.6 Despite
the fact that early detection and improved surgical techniques
has led to increased survival of melanoma patients, the prognosis
of patients with disseminated disease remains disappointing.
Several small molecule inhibitors acting at different levels in
the MAPK pathway have been described, such as PD03259017
and AZD6244,8 which target MEK, the downstream target of
* To whom correspondence should be addressed. Phone: +44 20 8722
4214. Fax: +44 20 8722 4046. E-mail: caroline.springer@icr.ac.uk.
†
The Institute of Cancer Research, UK Centre for Cancer Therapeutics.
The Institute of Cancer Research, UK Centre for Cell and Molecular
‡
Biology.
§
Astex Therapeutics Ltd.
a Abbreviations: BRAF, V-RAF murine sarcoma viral oncogene homo-
logue B1; MAPK, mitogen-activated protein kinase; MEK, MAPK/ERK
kinase; ERK, extracellular regulated kinase; Boc, tert-butoxycarbonyl; TFA,
trifluoroacetic acid; THF, tetrahydrofuran; DCM, dichloromethane; DMF,
dimethylformamide; DMSO, dimethylsulfoxide; atm, atmosphere.
10.1021/jm900242c CCC: $40.75  2009 American Chemical Society
Published on Web 05/27/2009

3882 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Me´nard et al.
Table 1. Effect of Modification of the Central Phenyl Ring of Inhibitor 1

NoVel Potent BRAF Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3883
Scheme 1. Two Synthetic Routes Employed to Generate Novel BRAF Inhibitors
In this paper, we show that several ring B modifications
improve the activity of this class of inhibitor on both purified
V600EBRAF in vitro and on oncogenic BRAF signaling in cells.
phenyl ring B of some intermediates 9, generates final com-
pounds in only three steps from intermediates 5a/f-m.
Whichever route is used, the first step is a coupling reaction
between 2-amino-4-chloro-3-nitropyridine and an appropriately
functionalized 4-aminophenol or 4-amino-1-naphthol. Only
intermediate 4 required 4-chloro-3-nitro-2-N-methylaminopy-
ridine intermediate 17, whose synthesis has already been
described.15 Substituted nitrophenols 6b-d/h,i were com-
mercially available; intermediates 6a, 6f,g and 7j were synthe-
sized. Substitution of the fluorine atom in the 3-fluoro-4-
nitrophenol afforded the corresponding 3-thiomethyl- (6a),
3-methoxy- (6f),17 and 3-dimethylamino-4-nitrophenol (6g)18
intermediates (Scheme 2). 3-Phenyl-4-aminophenol 7j was
synthesized in three steps by Avenova’s method19 via a
Chemistry
The target compounds 21a-y, 22a-f, 23a-c, 24a,b, and
25a,b were obtained via the two routes described in Scheme 1.
In route A, the A-B system is first fully synthesized, whereafter
ring C is attached. Route B relies on the introduction of ring C
before the formation of the 1H-imidazo[4,5-b]pyridine-2(3H)-
one A ring. Route A has the key advantage that it generates
final compounds with different C rings readily and within a few
steps from advanced amino intermediates such as 2a-e, 3d,e,
and 4 summarized in Figure 2. Route B, developed due to the
difficulties encountered in protecting the amino group on the
Figure 2. Key intermediates for the synthesis of BRAF inhibitors.
Scheme 2. Preparation of Substituted Aminophenol Ring B Starting from the Commercially Available 3-Fluoro-4-nitrophenola
a Reagents and conditions: (a) NaSMe, K2CO3, DMF; (b) NaOMe, MeOH, reflux; (c) (CH3)2NH·HCl, K2CO3, DMSO/H2O, 100 °C; (d) Fe, NH4Cl,
EtOH/H2O, reflux; (e) Pd/C, H2, RT, (f) Boc2O, InCl3, 35 °C.

3884 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Scheme 3. Synthesis of Key Intermediates 2a-ea
Me´nard et al.
a Reagents and conditions: (a) 2-amino-4-chloro-3-nitropyridine, tBuOK, DMF, 70 °C; (b) 2-amino-4-chloro-3-nitropyridine, NaH, DMSO, 100 °C; (c)
Boc2O, THF; (d) Pd/C, H2, RT; (e) Fe, NH4Cl, EtOH/H2O for 10a; (f) (Cl3CO)2CO, pyridine, THF; (g) TFA.
Scheme 4. Synthesis of Key Intermediates 3d-ea
a Reagents and conditions: (a) EtOCOCl, pyridine, THF; (b) TFA; (c) MeI, NaH, THF; (d) EtONa, EtOH, microwave condition.
Scheme 5. Synthesis of Key Intermediate 4a
a Reagents and conditions: (a) 4-(N-Boc-amino)naphthol, tBuOK, DMF, 70 °C; (b) H2, Pd/C, EtOH; (c) triphosgene, pyridine, THF; (d) TFA.
Diels-Alder cycloaddition of (Z)-4-benzylidene-2-phenyl-2-
oxazolin-5-one with Danishefsky’s diene.
the expected ether-linked compounds 9a-c/e-m in high yields
and purities. Compounds 10a/d were synthesized in a similar
way.
Substituted p-nitrophenol compounds 6b-d/f-i were reduced
by catalytic hydrogenation over Pd/C to provide p-aminophenols
7b-d/f-i. Reduction of 6a required the use of iron as the
thiomethyl group poisoned Pd/C and Raney nickel catalysts.
Ammonium chloride20 was preferred to acetic acid21 as a proton
source because the latter reaction gave 20% of the undesired
corresponding acetamide. Subsequent protection of the amino
group as required for route A was readily achieved using Boc2O
and catalytic InCl3 (1 mol %)22 at 35 °C to generate 8a and 8d.
Coupling of the functionalized phenols with 2-amino-4-
chloro-3-nitropyridine is presented in Scheme 3. NaH in DMSO
is preferred to tBuOK in dry DMF, as 4-(2-amino-3-nitropyridin-
4-ylamino) substituted phenols were formed as side products
(10 to 100%) in the latter case. The use of NaH provided only
Protection of the amino group of intermediates 9a-c/e was
carried out using Boc2O in THF at room temperature. However,
the presence of bulky groups at the 3-position of the ring B
(see Figure 1 for numbering) and the electronic effects of the
substituents led to slow reactions for compounds 9l and 9m
and a total lack of reactivity for compound 9a. The use of InCl3
(1 mol %) considerably improved the yield to 90% after 4 h at
35 °C for 9a. From this stage on, key intermediates 2a-e, 3d,e,
and 4 were obtained by following a similar synthetic route
described previously by Niculescu-Duvaz et al.15 (Schemes 3,
4, and 5). Reduction of the 3-nitropyridine group of protected
intermediates 10a-e gave intermediates 12a-e. Cyclisation of
the diamino pyridyl derivative with triphosgene, followed by

NoVel Potent BRAF Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3885
Table 2. Combinations of Various Substituted Phenyl Rings C with the Optimal Rings B
a All substitutions occurred in position 3 of the phenyl ring B with the exception of inhibitor 21p substituted in position 2.
deprotection of the ring B amino group with trifluoroacetic acid,
afforded intermediates 2a-e (Scheme 3).
3-trifluoromethylphenyl isocyanate in DCM at room temperature
to form 5a/f-m. The subsequent nitro group reduction was
performed using conditions previously described for the starting
substituted nitrophenol, i.e., Pd/C catalyst, H2, or iron with
NH4Cl (9a) to provide the corresponding 2,3-diaminopyridin-
4-yloxy intermediate 11a/f-m. Finally, formation of the imi-
dazolone ring was performed using triphosgene with pyridine
in THF at room temperature to afford compounds 21a/f-m.
Oxidation of the thiomethyl group of 21a using oxone in
water at 0 °C yielded the corresponding sulfoxide compound
26.
Intermediates 3d,e were obtained in four steps from com-
pounds 12d and 12e.15 First the diamino-intermediates 12d,e
were acylated in position N1 with ethyl chloroformate, after
which the Boc amino group protection was removed with TFA
to give intermediates 15d,e. Selective methylation of the
carbamate nitrogen using MeI and NaH, followed by an
intramolecular cyclization using microwave irradiation produced
imidazolones 3d and 3e, which are methylated at position N1
(Scheme 4).
The synthesis of 3-N-methylated intermediate 4 is similar to
the synthesis described previously for the common intermediates
2a-e, the only difference being the first step coupling of
4-chloro-3-nitro-2-N-methylaminopyridine 1715 and 4-(N-Boc-
aminonaphthol) (Scheme 5).
The final compounds 21b-e/n-y were obtained after cou-
pling key intermediates 2a-e, 3e, and 4 with several substituted
aryl isocyanates in dry THF at room temperature (Scheme 6).
Compounds 24a,b with a methylene amide linker and com-
pounds 25a,b with an amide linker resulting from condensation
of the corresponding substituted benzoyl chloride or pheny-
lacetyl chloride with intermediates 2e and 3e in dioxane in the
presence of triethylamine under an inert atmosphere at 80 °C
for 20 h (Scheme 7).
Results and Discussion
Our aim was to improve the potency and the cellular activity
of 1 by exploring the influence of modifications to ring B. To
compare the biological data from this modified series with the
unsubstituted 1, we kept constant the 1H-imidazo[4,5-b]pyridine-
2(3H)-one ring A, which binds to the hinge region, the urea
linker B-C and the 3-trifluoromethylphenyl substituted ring C,
which interact with the DFG-pocket (BP-III and BP-IV).16
The biological activities of all compounds were tested in vitro
using three assays: (a) the determination of the IC50 against the
isolated V600EBRAF enzyme (IC50, BRAF), (b) the determination
of the growth inhibition by SRB (GI50, SRB) of WM266.4
human melanoma cells that are driven by mutant BRAF, (c)
compounds were also tested for cellular inhibition of the target
pathway (IC50 of ERK phosphorylation in WM266.4 melanoma
cells-IC50, pERK)23 (see Table 1).
As the protection of some 4-(4-substituted aminophenoxy)-
3-nitropyridin-2-amines failed when using the conventional
procedure, route B was explored and the formation of ring A
was performed as the last step (Scheme 8). Thus, the rings C
were introduced by coupling intermediates 9a/f-m with 4-chloro-
Modifications on Ring B. All the modifications of ring B
yield compounds that inhibit oncogenic BRAF except substitu-

3886 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Me´nard et al.
Table 3. Effect of Modification of Ring A and Ring C in the Naphthyl Series
tion in the 3-position with a phenyl (21j) or a dimethylamino
group (21g) that lead to inactive compounds (see Table 1). The
replacement of the phenyl with a naphthyl ring is beneficial.
The introduction of a methyl or a fluoro group at position 3
(21m, 21d) confers greater potency to the compounds than the
same substitution at position 2 (21c, 21k). Despite good activity
against the V600EBRAF enzyme (0.023 and 0.047 µM compared
to 0.037 and 0.008 µM, respectively) a drop of 12- to 16-fold
in potency on pERK is observed. Unexpectedly, the reverse is
seen with a methoxy group (the 2-methoxy compound 21b is
more potent on IC50 BRAF and IC50 pERK than its 3-methoxy
analogue 21f), although no improvement is achieved with
respect to the parent compound 1 with an unsubstituted central
phenyl ring.
In general, hydrophobic groups such as halogens, especially
fluorine, are beneficial for potency on V600EBRAF (BRAF IC50
) 0.023 µM for a chlorine 21l and 0.008 µM for a fluorine
21d). However, inhibition of pERK activity in cells (IC50
ppERK) and inhibition of proliferation (SRB GI50) remain above
1 µM. The presence of a trifluoromethyl group (21i) enhances
the activity on SRB by 4.5-fold to afford a submicromolar
compound at 0.93 µM.
and SRB GI50 at 0.47 µM. The use of the naphthyl system as
ring B (21e), despite having a modest negative effect on BRAF
potency (BRAF IC50 of 30 nM), sharply increases the cellular
activity of these compounds (pERK IC50 of 0.56 µM and SRB
GI50 of 0.47 µM).
Isosteric replacements of the 3-thiomethyl substituent with
other heteroatom groups did not lead to equipotent or more
potent compounds. A methoxy group (21f) decreased drastically
the cell activity to greater than 100 µM, even though it showed
a relatively good BRAF activity at 0.106 µM. A dimethylamino
substituent (21g) is not tolerated. The oxidation of the thiomethyl
group of 21a to a methylsulfonyl group (26) also reduces both
the cellular activity and the activity against BRAF. Similarly,
replacing the naphthyl group by a 2,3-dimethylphenyl ring (21h)
is detrimental.
These results clearly show that the modification of the phenyl
middle ring can greatly enhance the activity of the compounds.
The substitution of the phenyl B ring in position 3 leads to more
potent inhibitors than in position 2. Hydrophobic groups enhance
the activity on cells providing submicromolar compounds in
pERK and SRB assays; the most promising groups are 3-thi-
omethyl, 3-fluorine, or replacement of the phenyl ring with
naphthyl.
Two ring B modifications show considerable overall im-
provements compared to 1. A 3-thiomethyl substituent on phenyl
ring B improves the BRAF potency to an IC50 of 1 nM (21a)
and shows good activity on cells with pERK IC50 at 0.61 µM
Variation of Ring C. We investigated the effect of ring C
when combined with the most beneficially substituted middle
ring B, following our improvements to the potency of 1 against

NoVel Potent BRAF Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3887
Table 4. Compounds with Amide or Methylene Amide Linker B-C
V600EBRAF and also on cells by modification of the phenyl ring
B. We selected C rings that had been identified as active,15 and
we combined them with the 3-fluoro, 3-thiomethylphenyl, and
naphthyl rings B (see Table 2).
Replacement of the chlorine atom of ring C for a hydrogen
atom does not affect the activity against isolated V600EBRAF
for inhibitors with a 3-fluoro- (21q) or a 3-thiomethyl (21n)
substituted phenyl ring B, but the potency on cells decreases
by approximately 1.5-fold. In contrast, the naphthyl compound
21t shows an enhancement on both BRAF IC50 and SRB GI50,
but the inhibition of pERK decreases by more than 3-fold.
Replacement of the trifluoromethyl group with the more
lipophilic trifluoromethylthio group (21s) has no effect on the
potency on the isolated enzyme or on cellular assays. However,
replacing the trifluoromethyl group by a tert-butyl group (21x)
(naphthyl substituted ring B) enhances the potency on all three
assays by almost 2-fold.
Compounds with a 3-trifluoromethyl-6-fluoro phenyl ring C
are less active in cellular potency assays for the 3-fluoro 21r
and 3-thiomethyl 21o compounds despite having a good activity
against V600EBRAF. In contrast, the combination of this ring with
naphthyl ring B (21u) increases the potency on cells. Increasing
the bulk on position 2 by introducing a morpholino group leads
to a decrease of the activity in all three assays.
Among the different rings C investigated, the 3-trifluoro-
methyl-6-fluorophenyl or a 3-tert-butylphenyl group when
combined with a naphthyl middle ring achieve the greatest
potency against V600EBRAF, cellular ERK phosphorylation, and
cell growth inhibition.
N-Methylation of Ring A. We have shown previously that
methylation of the pyridoimidazolone improves the pharmaco-
kinetic (PK) profile of our compounds by eliminating an H-bond
donor while preserving good V600EBRAF and cellular activities.15
We therefore combined the naphthyl ring B with the methylated
pyridoimidazolone ring A and with different rings C.
Of the compounds containing 4-chloro-3-trifluoromethylphe-
nyl 22b, a 3-trifluoro-6-fluoro-methylphenyl 22c or a 3-tert-
Figure 3. Graph showing the linear relationship between pERK IC50
and SRB GI50.

3888 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Me´nard et al.
Figure 5. Selective inhibition of a mutant BRAF-driven human
melanoma cell line displayed with 22c. Growth inhibition of WM266.4
(BRAF mutant) and WM1361 (NRAS mutant) melanoma cell lines
following a 5 day exposure to sorafenib (A) or 22c (B) and analysis
by SRB assay.
values of 29 of the inhibitors synthesized shows a high
correlation (eq 1), as the regression coefficient r is 0.87 (r2 )
0.757) (see Figure 3). This analysis was performed using
TSAR.24
Figure 4. Kinase selectivity of compounds 21n, 21u, and 21x.
Compounds were tested against a panel of 69 kinases at the MRC
National Centre for Protein Kinase Profiling. The color bar represents
percent activity relative to the control.
SRB(GI50) ) 2.41pERK(IC50) - 0.77
n ) 29 s ) 3.80 f ) 81.21
(1)
These data indicate that there is a significant correlation between
the SRB GI50 and the pERK IC50. This finding provides evidence
that the inhibition of growth of the mutant BRAF driven cells
is an on-target effect.
butylphenyl 22d ring C, the methylation on the N1 position is
tolerated with an effectively unchanged activity on V600EBRAF,
despite some detrimental effect on cellular activity compared
to the nonmethylated counterparts 21e, 21u, and 21x (see Table
3). The 3-trifluoromethylphenyl ring C (22e) combined with
the N1-methylated ring A is active with only a slight loss of
potency in the SRB assay. Again, a 3-trifluorothiomethylphenyl
ring C (22f) has no positive effect on cellular assays. Similarly,
a loss of cellular activity is notable for the 3-fluoro-substituted
ring B compound 22a (not shown, BRAF IC50 of 0.01 µM,
pERK IC50 of 21 µM, and SRB GI50 of 100 µM) compared to
its nonmethylated counterpart 21d.
Methylation at the N3 position is detrimental with a significant
drop of potency in all three assays (23a, 23b, and 23c). The
N3 of the imidazolone, facing the hinge backbone,15 does not
tolerate a substitution. These results suggest that the bulkier
naphthyl ring B subtly shifts the position of the hinge-binding
group A with respect to the published unsubstituted compounds
and possibly that hydrogen bonding of NH-3 of the imidazolone
with the hinge occurs to a certain extent for these compounds.
B-C Linker Modifications. To improve the bioavailability
of these BRAF inhibitors, we synthesized compounds with one
less NH-bond donor in the B-C linker (Table 4). Unfortunately,
the replacement of the urea linker for a methylene amide or an
amide leads to inactive compounds (cf 24a vs 21t and 25b vs 22f).
Linear Relationship between pERK IC50 and SRB GI50.
Linear regression analysis of the pERK IC50 and SRB GI50
Selectivity Studies. Our inhibitors were elaborated into the
BPI pocket in an attempt to provide selectivity for mutant BRAF
compared to other similar kinases. In this, we were successful
for the three compounds (21n, 21u, and 21x) we submitted to
the MRC National Centre for Protein Kinase Profiling and which
were assessed at a single concentration of 1 µM. The three
compounds are highly selective, hitting only a few kinases
against the assessed panel of 69 (Figure 4). Src, Lck, and p38
(MAP) kinases are common hits with the three compounds, with
21u showing selectivity toward the ꢀ isoform of p38 with
respect to the R one.
Not surprisingly, the same three proteins have been shown
to be sensitive to ligands that elaborate into the BPI pocket.
This is the case of the recently released cocrystal structure of
nilotinib25 with p38ꢀ, and of the ones of imatinib with Src26
and Lck.27 In all those structures, the methylated central ring
fits into the BPI pocket. Doramapimod, an inhibitor with
similarity to some of our compounds, is a highly selective
inhibitor of p38 MAP kinase28,16 and presents a naphthyl moiety
interacting with the BP1 pocket.
A cell selectivity experiment was also carried out. Growth
inhibition for sorafenib and 22c was performed against the
WM266.4 (BRAF mutant) and WM1361 (NRAS mutant) cell
lines (Figure 5). While no differential was observed in the case

NoVel Potent BRAF Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3889
Scheme 6. Formation of the B-C Urea Linker; Synthesis of Final Compounds via Route A
Scheme 7. Synthesis of Compounds with Amide and Methylene Amide Linker B-C via Route A
of sorafenib, a 6-fold difference is observed between the two
for some of the reported compounds and correlates with the
cell lines for 22c.
cell growth inhibition; this strongly supports the mode of action
of the compounds in this series. We have shown that these
compounds inhibit both the V600EBRAF isolated enzyme and the
enzyme in cellular assays and that this effect translates to potent
inhibition of the growth of a mutant BRAF dependent melanoma
cell line. Several of these compounds achieve over 10-fold
increase in cellular potency compared to the reported BRAF
inhibitor sorafenib. Furthemore, we have shown that elaborating
the phenyl middle ring into the BPI pocket provides protein
kinase and cell selectivities. These promising data warrant
further in vivo investigation of this series.
Conclusions
Following the development of a series of V600EBRAF inhibi-
tors based on a novel hinge-binding fragment,15 we have
extended their scope by modifying the central phenyl ring. Our
investigations show that ring B modification leads to a spectrum
of low nanomolar inhibitors of V600EBRAF. In a number of
examples, notably by ring B functionalization with fluoro,
thiomethyl groups, or replacement with a naphthyl, highly potent
inhibitors are obtained with IC50 values up to an order of
magnitude lower than those for analogues with unsubstituted
phenyl rings B. Seven compounds have an IC50 on V600EBRAF
of less than 10 nM, and compounds 21a and 21n, each
substituted with a thiomethyl group, notably achieve a
V600EBRAF IC50 of 1 nM. Moreover, this 3-thiomethyl substitu-
tion, as for the naphthyl group, increases the potency on cells
greatly with GI50s in the nanomolar range. V600EBRAF inhibition
in cells, measured with the pERK assay, is also submicromolar
Experimental Section
General Information. All starting materials, reagents, and
solvents for reactions were reagent grade and used as purchased.
Chromatography solvents were HPLC grade and were used without
further purification. Reactions were monitored by thin-layer chro-
matography (TLC) analysis using Merck silica gel 60 F-254 thin
layer plates. Flash column chromatography was carried out on

3890 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Me´nard et al.
Scheme 8. Access to Final Compounds 21a/f-m and 26 via Route Ba
a Reagents and conditions: (a) 4-chloro-3-trifluoromethylphenylisocyanate in DCM or THF at room temperature, Ar atm; (b) Pd/C, H2, RT or Fe, NH4Cl
in EtOH/H2O, reflux for 9a; (c)(Cl3CO)2CO, pyridine, THF, room temperature; (d) oxone, 0 °C.
Merck silica gel 60 (0.015-0.040 mm) or in disposable Isolute
Flash Si and Si II silica gel columns. Preparative TLC was
performed on either Macherey-Nagel [809 023] precoated TLC
plates SIL G-25 UV254 or Analtech [2015] precoated preparative
TLC plates, 2000 µm with UV254. LC-MS analyses were performed
on a Micromass LCT/Water’s Alliance 2795 HPLC system with a
Discovery 5 µm, C18, 50 mm × 4.6 mm i.d. column from Supelco
at a temperature of 22 °C using the following solvent system:
solvent A, methanol; solvent B, 0.1% formic acid in water, at a
flow rate of 1 mL/min. Gradient starting with 10% A/90% B from
0 to 0.5 min then 10% A/90% B to 90% A/10% B from 0.5 to 6.5
min and continuing at 90% A/10% B up to 10 min. From 10 to
10.5 min, the gradient reverted back to 10% A/90% B and was
held until 12 min. UV detection was at 254 nm, and ionization
was positive or negative ion electrospray. The molecular weight
scan range was 50-1000. Samples were supplied as 1 mg/mL in
DMSO or methanol with 3 µL injected on a partial loop fill. NMR
spectra were recorded on a Bruker DPX 250 MHz or a Bruker
Advance 500 MHz spectrometer. The signal of the deuterated
solvent was used as internal reference. Chemical shifts (δ) are given
in ppm and are referenced to residual, not fully deuterated solvent
signal (e.g., DMSO-d5). Coupling constants (J) are given in Hz.
The purity of the final compounds was determined by HPLC as
described above and is 95% or higher unless specified otherwise.
Procedure A: 1-(4-(2,3-Dihydro-2-oxo-1H-imidazo[4,5-b]pyri-
din-7-yloxy)-2-(methylthio) phenyl)-3-(4-chloro-3-(trifluorometh-
yl)phenyl)urea (21a). 1-(4-(2,3-Diaminopyridin-4-yloxy)-2-(meth-
ylthio)phenyl)-3-(4-chloro-3-(trifluoromethyl)phenyl)urea 11a (1.52
g, 3.1 mmol) was dissolved in dry THF (41 mL), and pyridine (2
mL) was added and the solution was cooled at 0 °C. Triphosgene
(3.1 mmol) in dry THF (20 mL) was added dropwise. The reaction
mixture was stirred for 15 min more at 0 °C and at room
temperature for 16 h. The solvent was evaporated and the residue
washed with water to provide after filtration the title compound
substituted arylisocyanate (1.1 equiv) in anhydrous THF were stirred
at room temperature for 14 h. The solvent was evaporated and the
solid residue was washed with Et2O to afford after filtration the
1
title compound (28 mg, 98%) as a pale-yellow powder. H NMR
δ: 3.75 (s, 3H, CH3), 6.05 (d, 1H, J ) 5.8, HPy), 7.05 (d, 1H, J )
8.7, Harom), 7.26 (dd, 1H, J ) 8.6 and J ) 1.9, Harom), 7.56 (d, 1H,
J ) 8.8, Harom), 7.60 (d, 1H, J ) 5.8, HPy), 7.65 (d, 1H, J ) 2.1,
Harom), 7.90 (m, 1H, Harom), 8.28 (d, 1H, J ) 2.4, Harom), 10.02 (bs,
2H, NHurea), 10.82 (bs, 1H, NHPy), 10.98 (bs, 1H, NHPy). 13C NMR
δ: 55.7, 104.3, 106.2, 111.0, 112.5, 116.8, 122.2, 122.4, 123.2,
124.0, 126.5, 131.7, 136.2, 138.8, 140.3, 151.2, 151.5, 153.1, 153.8,
156.2, 162.3. LC-MS: m/z 494 ([M + H]+, 100). HRMS: m/z calcd
for C21H15ClF3N5O4 ([M + H]+): 494.0837; found: 494.0836.
Compounds 21c-e/n-y; 22a-f; 23a-c were synthesized in a
similar manner. The data are available in Supporting Information.
Procedure C: N-(4-(2-oxo-2,3-dihydro-1H-imidazo[4,5-b]pyri-
din-7-yl-oxy)naphthalen-1-yl)-aminocarbonylmethylene-(3-triflu-
oromethyl-phenyl) (24a). To 7-(4-aminonaphthalen-1-yloxy)-1H-
imidazo[4,5-b]pyridin-2(3H)-one 2e (50 mg, 0.17 mmol) in anhydrous
dioxane (7 mL), triethylamine (150 µL), and 2-(3-(trifluorometh-
yl)phenyl)acetyl chloride (0.20 mmol) were added under stirring
and nitrogen atmosphere and heated for 20 h at 80 °C. The solvent
was evaporated under vacuum and the solid residue dissolved in
10 mL of EtOAc and washed successively with brine (2 × 10 mL),
a solution of citric acid, and brine again. The organic solution was
evaporated under vacuum, and the solid residue recrystallized from
1
EtOAc to afford the title compound (59 mg, 67%). H NMR δ:
3.97 (s, 2H, CH2), 6.24 (d, 1H, J ) 5.9, HPy), 7.26 (d, 1H, J ) 8.2,
Harom), 7.58-7.74 (m, 7H, Harom + HPy), 7.81 (s, 1H, Harom), 7.98
(d, 1H, J ) 8.2, Harom), 8.12 (d, 1H, J) 8.4, Harom), 10.29 (s, 1H,
NHamide), 11.38 (s, 1H, NHPy), 11.44 (s, 1H, NHPy). LC-MS: m/z
479 ([M + H]+, 100). HRMS: m/z calcd for C25H17F3N4O3 ([M +
H]+) 479.1331; found 479.1318.
Compounds 24b, 25a,b were synthesized in a similar manner.
The data are available in the Supporting Information.
1
(1.55 g, 97%) as a brown powder. H NMR δ: 2.45 (s, 3H, CH3),
6.42 (d, 1H, J smd 4.2, HPy), 6.99 (d, 1H, J smd 8.4, Harom), 7.19
(s, 1H, Harom), 7.63 (s, 2H, Harom), 7.78 (s, 2H, Harom), 8.12 (s, 1H,
1-(4-Chloro-3-(trifluoromethyl)phenyl)-3-(2-(methylsulfonyl)-
4-(2-oxo-2,3-dihydro-1H-imidazo[4,5-b]pyridin-7-yloxy)phenyl)-
urea (26). To a slurry of compound 21a 1-(4-(2,3-dihydro-2-oxo-
1H-imidazo[4,5-b]pyridin-7-yloxy)-2-(methylthio)phenyl)-3-(4-
chloro-3-(trifluoromethyl)phenyl)urea (35 mg, 0.07 mmol) in MeOH
(700 µL) was added slowly a solution of oxone (5 equiv) in water
(900 µL). The mixture maintained at 0 °C was stirred 5 h and then
neutralized with a saturated solution of NaHCO3 to pH 8. The
mixture was diluted in water and then filtered to provide the
H
arom), 8.26 (s, 1H, NHurea), 9.91 (s, 1H, NHurea), 11.25 (s, 1H,
NHPy), 11.44 (bs, 1H, NHPy). 13C NMR δ: 15.7, 105.7, 113.2, 116.4,
116.5, 117.3, 121.7, 122.2, 122.7, 123.8, 124.4, 131.9, 132.0, 133.1,
139.4, 141.1, 145.4, 146.8, 150.5, 152.6, 154.1. LC-MS: m/z 510
([M + H]+, 100). HRMS: m/z calcd for C21H15ClF3N5O3S ([M +
H]+): 510.0611; found: 510.0614.
Compounds 21f-m were synthesized in a similar manner. The
data are available in Supporting Information.
1
expected sulfone as a powder (22 mg, 60%). H NMR δ: 3.32 (s,
Procedure B: 1-(4-(2,3-Dihydro-2-oxo-1H-imidazo[4,5-b]pyridin-
7-yloxy)-3-methoxyphenyl)-3-(4-chloro-3-(trifluoromethyl)phenyl)-
urea (21b). 7-(4-Amino-2-methoxyphenoxy)-1H-imidazo[4,5-b]-
pyridin-2(3H)-one 2b (16 mg, 0.06 mmol) and corresponding
3H, CH3), 6.58 (d, 1H, J smd 5.8, HPy), 7.51 (dd, 1H, J smd 8.9
and J smd 2.7, Harom), 7.54 (d, 1H, J smd 2.7, Harom), 7.64 (d, 1H,
J smd 8.7, Harom), 7.70 (d, 1H, J ) 8.1, Harom), 7.83 (d, 1H, J )
5.8, HPy), 8.11 (m, 1H, Harom), 8.14 (d, 1H, J ) 8.9, Harom), 8.64 (s,

NoVel Potent BRAF Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3891
(11) Sebolt-Leopold, J. S.; Herrera, R. Targeting the mitogen-activated
protein kinase cascade to treat cancer. Nat. ReV. Cancer 2004, 4, 937–
947.
1H, NHurea), 10.29 (s, 1H, NHurea), 11.21 (s, 1H, NHPy), 11.45 (bs,
1H, NHPy). 13C NMR δ: 42.7, 106.7, 113.7, 116.9, 118.5, 121.6,
123.0, 123.8, 125.2, 126.3, 126.6, 130.2, 131.9, 133.3, 139.0, 141.3,
143.9, 147.2, 149.7, 151.9, 154.1. LC-MS: m/z 542 ([M + H]+,
100). HRMS: m/z calcd for C21H15ClF3N5O5S ([M + H]+):
542.0507; found: 542.0503.
(12) Adjei, A. A.; Cohen, R. B.; Franklin, W.; Morris, C.; Wilson, D.;
Molina, J. R.; Hanson, L. J.; Gore, L.; Chow, L.; Leong, S.; Maloney,
L.; Gordon, G.; Simmons, H.; Marlow, A.; Litwiler, K.; Brown, S.;
Poch, G.; Kane, K.; Haney, J.; Eckhardt, S. G. Phase I pharmacokinetic
and pharmacodynamic study of the oral, small-molecule mitogen-
activated protein kinase kinase 1/2 inhibitor AZD6244 (ARRY-
142886) in patients with advanced cancers. J. Clin. Oncol. 2008, 26,
2139–2146.
(13) Niculescu-Duvaz, I.; Roman, E.; Whittaker, S. R.; Friedlos, F.; Kirk,
R.; Scanlon, I. J.; Davies, L. C.; Niculescu-Duvaz, D.; Marais, R.;
Springer, C. J. Novel inhibitors of B-RAF based on a disubstituted
pyrazine scaffold. Generation of a nanomolar lead. J. Med. Chem.
2006, 49, 407–416.
(14) Niculescu-Duvaz, I.; Roman, E.; Whittaker, S. R.; Friedlos, F.; Kirk,
R.; Scanlon, I. J.; Davies, L. C.; Niculescu-Duvaz, D.; Marais, R.;
Springer, C. J. Novel inhibitors of the v-raf murine sarcoma viral
oncogene homologue B1 (BRAF) based on a 2,6-disubstituted pyrazine
scaffold. J. Med. Chem. 2008, 51, 3261–3274.
(15) Niculescu-Duvaz, D.; Gaulon, C.; Dijkstra, H. P.; Niculescu-Duvaz,
I.; Zambon, A.; Me´nard, D.; Suijkerbuijk, B. M. J. M.; Nourry, A.;
Davies, L.; Manne, H.; Friedlos, F.; Ogilvie, L.; Hedley, D.; Whittaker,
S.; Kirk, R.; Gill, A.; Taylor, R. D.; Raynaud, F. I.; Moreno-Farre, J.;
Marais, R.; Springer, C. J. Pyridoimidazolones as Novel Potent
Inhibitors of v-Raf Murine Sarcoma Viral Oncogene Homologue B1
(BRAF). J. Med. Chem. 2009, 52, 2255–2264.
Acknowledgment. This work is supported by Cancer
Research UK (refs: C309/A2187 and C107/A10433), the
Wellcome Trust, the Isle of Mann Anti-Cancer Association, and
The Institute of Cancer Research. We acknowledge NHS
funding to the NIHR Biomedical Research Centre. Professors
Paul Workman and Julian Blagg are acknowledged for their
strong support and helpful reading of the manuscript. Meirion
Richards and Dr. Amin Mirza are thanked for providing
technical assistance.
Supporting Information Available: Experimental protocols and
analytical data for final compounds 21f-m, 21c-e/n-y, 22a-f,
23a-c, 24b, 25a,b, and all intermediates 2a-e, 3d,e, 4, 5a/f-m,
6a/f/g, 7a-d/f-j, 8a/d, 9a-c/e-m, 10a-e, 11a/f-m, 12a-e,
13a-e, 14d,e, 15d,e, 16d,e, 18, 19, 20, and protocols for biological
assays. This material is available free of charge via the Internet at
http://pubs.acs.org.
(16) Liao, J. J. L. Molecular recognition of protein kinase binding pockets
for design of potent and selective kinase inhibitors. J. Med. Chem.
2007, 50, 409–424.
References
(1) Robinson, M. J.; Cobb, M. H. Mitogen-activated protein kinase
pathways. Curr. Opin. Cell Biol. 1997, 9, 180–186.
(2) Cohen, P. Protein kinasessthe major drug targets of the twenty-first
century. Nat. ReV. Drug DiscoVery 2002, 1, 309–315.
(17) Cervera, M.; Marquet, J.; Martin, X. Charge control in the SNAr
reaction. Meta substitution with respect to the activating nitro group
in 3,4-dihalogenonitrobenzenes. Tetrahedron 1996, 52, 2557–2564.
(18) Langer, O.; Dolle, F.; Valette, H.; Halldin, C.; Vaufrey, F.; Fuseau,
C.; Coulon, C.; Ottaviani, M.; Nagren, K.; Bottlaender, M.; Maziere,
B.; Crouzel, C. Synthesis of high-specific-radioactivity 4- and 6-[F-
18]fluorometaraminol-PET tracers for the adrenergic nervous system
of the heart. Bioorg. Med. Chem. 2001, 9, 677–694.
(3) Davies, H.; Bignell, G. R.; Cox, C.; Stephens, P.; Edkins, S.; Clegg,
S.; Teague, J.; Woffendin, H.; Garnett, M. J.; Bottomley, W.; Davis,
N.; Dicks, N.; Ewing, R.; Floyd, Y.; Gray, K.; Hall, S.; Hawes, R.;
Hughes, J.; Kosmidou, V.; Menzies, A.; Mould, C.; Parker, A.;
Stevens, C.; Watt, S.; Hooper, S.; Wilson, R.; Jayatilake, H.; Gusterson,
B. A.; Cooper, C.; Shipley, J.; Hargrave, D.; Pritchard-Jones, K.;
Maitland, N.; Chenevix-Trench, G.; Riggins, G. J.; Bigner, D. D.;
Palmieri, G.; Cossu, A.; Flanagan, A.; Nicholson, A.; Ho, J. W. C.;
Leung, S. Y.; Yuen, S. T.; Weber, B. L.; Siegler, H. F.; Darrow, T. L.;
Paterson, H.; Marais, R.; Marshall, C. J.; Wooster, R.; Stratton, M. R.;
Futreal, P. A. Mutations of the BRAF gene in human cancer. Nature
2002, 417, 949–954.
(4) Wan, P. T. C.; Garnett, M. J.; Roe, S. M.; Lee, S.; Niculescu-Duvaz,
D.; Good, V. M.; Jones, C. M.; Marshall, C. J.; Springer, C. J.; Barford,
D.; Marais, R.; Cancer Genome, P. Mechanism of activation of the
RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell
2004, 116, 855–867.
(19) Avenoza, A.; Busto, J. H.; Cativiela, C.; Peregrina, J. M. New efficient
synthesis of 4-amino-3-arylphenols. Synthesis Stuttgart 1995, 671–
674.
(20) Winn, M.; Reilly, E. B.; Liu, G.; Huth, J. R.; Jae, H. S.; Freeman, J.;
Pei, Z. H.; Xin, Z. L.; Lynch, J.; Kester, J.; von Geldern, T. W.; Leitza,
S.; DeVries, P.; Dickinson, R.; Mussatto, D.; Okasinski, G. F.
Discovery of novel p-arylthio cinnamides as antagonists of leukocyte
function-associated antigen-1/intercellular adhesion molecule-1 interac-
tion. 4. Structure-activity relationship of substituents on the benzene
ring of the cinnamide. J. Med. Chem. 2001, 44, 4393–4403.
(21) Kundu, N. G.; Nandi, B. Palladium-copper catalyzed synthesis of
benzofused heterocycles with two heteroatoms: Novel and highly regio-
and stereoselective syntheses of (E)-2-(2-arylvinyl)-3-tosyl-2,3-dihy-
dro-1,3-benzothiazoles and (E)-2-alkyl(aryl)idene-3,4-dihydro-2H-1,4-
benzothiazines. J. Org. Chem. 2001, 66, 4563–4575.
(5) Karasarides, M.; Chiloeches, A.; Hayward, R.; Niculescu-Duvaz, D.;
Scanlon, I.; Friedlos, F.; Ogilvie, L.; Hedley, D.; Martin, J.; Marshall,
C. J.; Springer, C. J.; Marais, R. B-RAF is a therapeutic target in
melanoma. Oncogene 2004, 23, 6292–6298.
(22) Chankeshwara, S. V.; Chakraborti, A. K. Indium(III) halides as new
and highly efficient catalysts for N-tert-butoxycarbonylation of amines.
Synthesis 2006, 2784–2788.
(6) Friday, B. B.; Adjei, A. A. Advances in targeting the Ras/Raf/MEK/
Erk mitogen-activated protein kinase cascade with MEK inhibitors
for cancer therapy. Clin. Cancer Res. 2008, 14, 342–346.
(7) Brown, A. P.; Carlson, T. C. G.; Loi, C. M.; Graziano, M. J.
Pharmacodynamic and toxicokinetic evaluation of the novel MEK
inhibitor, PD0325901, in the rat following oral and intravenous
administration. Cancer Chemother. Pharmacol. 2007, 59, 671–679.
(8) Davies, B. R.; Logie, A.; McKay, J. S.; Martin, P.; Steele, S.; Jenkins,
R.; Cockerill, M.; Cartlidge, S.; Smith, P. D. AZD6244 (ARRY-
142886), a potent inhibitor of mitogen-activated protein kinase/
extracellular signal-regulated kinase kinase 1/2 kinases: mechanism
of action in vivo, pharmacokinetic/pharmacodynamic relationship, and
potential for combination in preclinical models. Mol. Cancer Ther.
2007, 6, 2209–2219.
(23) The phospho-ERK method is included in the Supporting Information.
(24) TSARTM 3.3; Oxford Molecular Limited: Oxford, UK, 2000.
(25) Filippakopoulos, P., Barr, A., Fedorov, O., Keates, T., Soundararajan,
M., Elkins, J., Salah, E., Burgess-Brown, N., Ugochukwu, E., Pike,
A. C. W., Muniz, J., Roos, A., Chaikuad, A., von Delft, F.,
Arrowsmith, C. H., Edwards, A. M., Weigelt, J., Bountra, C., Knapp,
S. Crystal Structure of Human Mitogen Activated Protein Kinase 11
(p38 beta) in complex with Nilotinib. Structural Genomics Consortium
(SGC) 2009; to be published.
(26) Seeliger, M. A.; Nagar, B.; Frank, F.; Cao, X.; Henderson, M. N.;
Kuriyan, J. c-Src binds to the cancer drug imatinib with an inactive
Abl/c-Kit conformation and a distributed thermodynamic penalty.
Structure 2007, 15, 299–311.
(9) Sala, E.; Mologni, L.; Truffa, S.; Gaetano, C.; Bollag, G. E.;
Gambacorti-Passerini, C. BRAF silencing by short hairpin RNA or
chemical blockade by PLX4032 leads to different responses in
melanoma and thyroid carcinoma cells. Mol. Cancer Res. 2008, 6,
751–759.
(27) Jacobs, M. D.; Caron, P. R.; Hare, B. J. Classifying protein kinase
structures guides use of ligand-selectivity profiles to predict inactive
conformations: Structure of lck/imatinib complex. Proteins: Struct.,
Funct., Bioinf. 2008, 70, 1451–1460.
(10) Ramurthy, S.; Subramanian, S.; Aikawa, M.; Amiri, P.; Costales, A.;
Dove, J.; Fong, S.; Jansen, J. M.; Levine, B.; Ma, S.; McBride, C. M.;
Michaelian, J.; Pick, T.; Poon, D. J.; Girish, S.; Shafer, C. M.; Stuart,
D.; Sung, L.; Renhowe, P. A. Design and Synthesis of Orally
Bioavailable Benzimidazoles as Raf Kinase Inhibitors. J. Med. Chem.
2008, 51, 7049–7052.
(28) Pargellis, C.; Tong, L.; Churchill, L.; Cirillo, P. F.; Gilmore, T.;
Graham, A. G.; Grob, P. M.; Hickey, E. R.; Moss, N.; Pav, S.; Regan,
J. Inhibition of p38 MAP kinase by utilizing a novel allosteric binding
site. Nat. Struct. Biol. 2002, 9, 268–272.
JM900242C