Development of potent B-RafV600E inhibitors containing an arylsulfonamide headgroup

Article abstract of DOI:10.1016/j.bmcl.2011.06.021

A potent series of inhibitors against the B-Raf^V600E kinase have been developed that show excellent activity in cellular assays and good oral bioavailability in rats. The key structural features of the series are an arylsulfonamide headgroup, a thiazole core, and a fluorine ortho to the sulfonamide nitrogen. Crown Copyright

Full text of DOI:10.1016/j.bmcl.2011.06.021

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4436–4440
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Development of potent B-RafV600E inhibitors containing
an arylsulfonamide headgroup
⇑
,
John C. Stellwagen a, , George M. Adjabeng a, Marc R. Arnone b, Scott H. Dickerson a, Chao Han b,
Keith R. Hornberger b, Alastair J. King b, Robert A. Mook Jr. a, Kimberly G. Petrov a, Tara R. Rheault a,
Cynthia M. Rominger b, Olivia W. Rossanese b, Kimberly N. Smitheman b, Alex G. Waterson a,
David E. Uehling a, ,à
⇑
a GlaxoSmithKline Oncology CEDD, Five Moore Drive, Research Triangle Park, NC 27709-3398, United States
b GlaxoSmithKline Oncology CEDD, 1250 South Collegeville Road, Collegeville, PA 19426-0989, United States
a r t i c l e i n f o
a b s t r a c t
A potent series of inhibitors against the B-RafV600E kinase have been developed that show excellent activ-
ity in cellular assays and good oral bioavailability in rats. The key structural features of the series are an
arylsulfonamide headgroup, a thiazole core, and a fluorine ortho to the sulfonamide nitrogen.
Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
Article history:
Received 11 January 2011
Revised 5 June 2011
Accepted 7 June 2011
Available online 16 June 2011
Keywords:
B-RafV600E
Sulfonamide
Thiazole
B-Raf is a member of the Raf family of protein kinases, and
serves as a critical component of the MAPK signal-transduction
pathway.1 Mutations in the regulatory domain of B-Raf, particu-
larly the V600E mutation, cause constitutive activation of the ki-
nase function.2 This activity is required to maintain the
proliferation and oncogenic character of cancers possessing mu-
tated B-Raf.3 Selective inhibition of mutant B-RafV600E kinase
should therefore provide clinical benefit to patients with cancers
that commonly express these mutations, particularly melanoma
and colon cancer.4
In an effort to improve the cellular activity of the series, SAR in
the headgroup, core and tail regions of compound 1 were explored
in parallel (Fig. 1).
Headgroup SAR was explored by preparing a set of analogs in
which the amide moiety linking the aryl headgroup to the core
was replaced with a series of other linkers. The general synthetic
route to these analogs is shown in Scheme 1.6 After esterification,
the anilines were reacted with either a benzoyl chloride, an aryl
isocyanate, or an arylsulfonyl chloride to form the amide, urea,
and sulfonamide linker-containing compounds, respectively. The
Compound 1 (Fig. 1), a B-Raf lead identified from our oncology-
directed kinase programs, features an imidazopyridine core and a
large hydrophobic benzamide headgroup. Although this compound
displayed good activity in the B-RafV600E enzyme assay, it had little
activity in either mechanistic (pERK) or anti-proliferative cell-
based assays using the B-RafV600E mutant SKMEL28 cell line5
(Table 1a).
core
F
headgroup
O
N
N
N
F
HN
N
⇑
Corresponding authors. Tel.: +1 617 444 5355; fax: +1 617 577 9822 (J.C.S.), tel.:
+1 647 260 7961 (D.E.U.).
N
N
H
CH3
E-mail addresses: john.stellwagen@amgen.com (J.C. Stellwagen), david.uehlin-
g@oicr.on.ca (D.E. Uehling).
tail
Present address: Amgen, Inc., 360 Binney Street, Cambridge, MA 02142, United
States.
1
à
Present address: Ontario Institute for Cancer Research, 101 College Street, Suite
800, Toronto, ON M5G 0A3, United States.
Figure 1. Early benzamide B-Raf lead.
0960-894X/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.021

J. C. Stellwagen et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4436–4440
4437
Table 1a
Headgroup linker SAR: Imidazopyridine core
N
R
N
N
N
CH3
N
N
H
a
a
b
Entry
R
B-RafV600E IC50 (nM)
pERK EC50 (nM)
SKMEL28 EC50 (nM)
5316
F
H
N
1c
9
>10,000
>20,000
O
F
F
F
F
O
O
9
5
6336
N
H
N
H
F
10c
11
12
>2200
99
NTd
N
H
H
N
132
1109
S
O
O
a
b
c
B-Raf and pERK IC50 values are means of 2–4 experiments; individual values are within 30% of the reported mean value.
SKMEL28 IC50 values are means of two experiments; individual values vary approximately 2-fold.
Compounds tested as HCl salts.
NT = not tested.
d
Table 1b
Alkylthiazole core
H3C
CH3
N
R
S
N
N
CH3
N
N
H
a
a
b
Entry
R
B-RafV600E IC50 (nM)
pERK EC50 (nM)
SKMEL28 EC50 (nM)
F
H
N
12
4
580
52
2753
287
O
S
F
H
N
13
12
O
O
a
B-Raf and pERK (B-RafV600E SKMEL28 cells) IC50 values are means of 2–5 experiments; individual values are within 50% of the reported mean value.
SKMEL28 IC50 values are means of two experiments; individual values are within 5% of the reported mean value.
b
ester was condensed with the lithium anion of 2-chloro-4-methyl-
Replacing the imidazopyridine with a thiazole core, while keep-
ing the benzamide headgroup, also increased potency in cellular
assays (compare 12 with 1). Combining these two observations
generated the potent B-Raf inhibitor 13, showing dramatic
improvements in both pERK potency (52 nM) and anti-prolifera-
tive potency (287 nM) (Table 1b).
The importance of the sulfonamide N–H for B-Raf potency in
the thiazole core series was further investigated and is shown in
Table 2. For this set of analogs, the N-methyl-tetrahydroisoquino-
line tail of 13 was replaced with an N-acetylpiperazinylpyridine
tail to form 14 (Table 2). This tail had been shown to marginally
improve potency in the B-Raf cellular assays.7 The sulfonamide
N–H was then either methylated (15), or replaced with a methy-
lene group (16). The modification or loss of the sulfonamide N–H
in these two analogs caused a dramatic reduction in their in vitro
pyrimidine to generate ketone 4. Bromination of the ketone with
NBS followed by cyclization with either 2-aminopyridine (5) or
2-methylpropylthioamide (6) afforded the desired heterocyclic
core. The aniline ‘tail’ was then added to the chloropyrimidine
using acidic conditions in a microwave to generate the desired ana-
logs (7 and 8).
Evaluation of several different headgroup linkers (1 and 9–
11) revealed that the sulfonamide-containing analog 11 showed
a substantial improvement in cellular potency, particularly in
the pERK mechanistic assay run in B-RafV600E mutant SKMEL28
cells. Interestingly, sulfonamide 11 had similar potency in the
cellular pERK assay and B-Raf biochemical assay, although
activity in the SKMEL28 anti-proliferative assay was still
inadequate.

4438
J. C. Stellwagen et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4436–4440
Ar1
Ar1
A
O
A
O
O
HN
H2N
HN
OH
a,b
c
OMe
N
X
X
X
N
Cl
3
2
4
X = H or F
Ar1
Ar1
A
A
N
N
N
HN
N
HN
f
N
N
X
X
A
NAr2
d
N
N
Cl
e
7
5
Ar1
Ar1
A
HN
N
N
S
S
HN
f
N
N
X
X
NAr2
N
Cl
N
8
6
Scheme 1. General synthetic route for imidazopyridine and thiazole inhibitors. Reagents and conditions: (a) H2SO4, MeOH; (b) for amide linker (A = CO): ArCOCl, TEA, DCM;
for urea linker (A = NHCO): ArNCO, THF; for sulfonamide linker (A = SO2): ArSO2Cl, pyridine, DCM; (c) 2-chloro-4-methylpyrimidine, LiHMDS; (d) NBS, DCM then 2-
aminopyridine, dioxane; (e) NBS, DMF then 2-methyl-propanethioamide; (f) aniline or aminopyridine tail, HCl, 2,2,2-trifluoroethanol, microwave, 180 °C.
Table 2
Sulfonamide N–H substitution
R
O
O
CH3
S
X
N
O
S
N
N
CH3
N
N
N
N
H
B-RafV600E IC50 (nM)
pERK EC50 (nM)
SKMEL28 EC50 (nM)
a
a
b
Entry
X
R
14
15
16
NH
NMe
CH2
CH3
H
H
3.6
1970
1358
7
4731
>20,000
24
8619
11,091
a
B-Raf and pERK (B-RafV600E SK-MEK-28 cells) IC50 values are means of two experiments; individual values are within 50% of the reported mean value.
SKMEL28 IC50 values are means of two experiments; individual values are within 10% of the reported mean value.
b
activity. Thus, the sulfonamide N–H appeared to be a key pharma-
cophore for potent in vitro activity in this series.
The B-Raf potency of this series was further optimized through
an SAR study of the aryl headgroup in sulfonamide 14, summarized
in Table 3. Addition of a fluorine in either the 2- or 3-position of the
aryl ring (compounds 17 and 18) was found to improve both the B-
Raf enzyme activity and the potency in the SKMEL28 proliferation
assay. Fluorine substitution in the 4-position of the ring (19), or a
larger substituent, such as a chlorine (20), led to a moderate reduc-
tion in potency in both the enzyme and cellular assays. The diflu-
orinated analogs 21 and 22 mirrored the high potencies observed
for the mono-fluorinated analogs 17 and 18.
The in vitro intrinsic clearances of compounds in the headgroup
SAR series in rat liver microsomes are also listed in Table 3.
Although the clearances are high for almost all compounds in the
table, fluorination of the headgroup did marginally improve the
metabolic stability of this series. The only analog found to show
a moderate intrinsic clearance in rat microsomes was the 2,6-diflu-
orinated sulfonamide 22. This headgroup provided the best combi-
nation of potency and metabolic stability, and was thus used for
further SAR exploration in other regions of the molecule.
The interactions between the sulfonamide group and the kinase
were evaluated in a model of a compound from the sulfonamide
thiazole series docked into a B-RafV600E crystal structure (Fig. 2).8
We speculate that the arylsulfonamide nitrogen exists as a depro-
tonated species, making hydrogen bonding interactions with the
backbone NH of Asp-594. This is similar to the binding modes ob-
served for the sulfonamide groups in the B-Raf inhibitors PLX47208
and PLX4032.4b This model is also consistent with calculated pKa
data9 showing the sulfonamide NH having a pKa ꢀ7, while the
pKa for the amide and urea analogs ranged from 10 to 12. This dif-
ference in pKa, and thus the ability of the NH to ionize at the pH
that the in vitro assays are conducted at (7.2–7.4), may explain
the large potency differences observed between these linkers. Fi-
nally, the ammonium ion of Lys-483 is in close proximity to the
sulfonamide anion, and can form a hydrogen bond with one of
the sulfonamide oxygens whereas the other sulfonamide oxygen
forms a hydrogen bond with the backbone NH of Phe-595.

J. C. Stellwagen et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4436–4440
4439
Having achieved excellent potency in the enzyme and cellular
assays, our focus turned to further improving the metabolic stabil-
ity and oral exposure of the series (Table 4). For this effort, the
acetylpiperazine moiety in the tail group of 22 was replaced with
a morpholine ring to generate 23, which was found to marginally
enhance the in vitro potency of the series. While 23 maintained
excellent potency in all in vitro assays, it showed high intrinsic
clearance in rat liver microsomes, and poor oral exposure in the
rat. In an attempt to improve the metabolic stability of the series,
potential metabolic sites were targeted. As such, a fluorine atom
was systematically installed around the phenyl ring connecting
the thiazole core to the sulfonamide. Analogs 24 and 25, with a
fluorine atom either ortho to both the thiazole and the sulfonamide
nitrogen or meta to both groups, showed little improvement in oral
exposure in rat despite showing a modest improvement in intrinsic
clearance. However, a fluorine atom para to the thiazole (26) was
found to reduce both the intrinsic clearance and the intravenous
clearance, providing an almost 70-fold increase in the rat oral
dose-normalized AUC compared to 23. In comparing the four-fold
improvement in iv clearance with the 70-fold improvement in oral
Figure 2. Model of sulfonamide inhibitor in the active site of B-RafV600E
.
Table 3
Headgroup SAR
H3C
O
O
CH3
S
HN
N
O
R
S
N
N
CH3
N
N
N
N
H
B-RafV600E IC50 (nM)
pERK EC50 (nM)
SKMEL28 EC50 (nM)
Rat Clic
a
a
b
Entry
R
14
17
18
19
20
21
22
H
2-F
3-F
4-F
3-Cl
2,5-F
2,6-F
3.6
0.6
0.8
9.5
34
7
4
10
9
21
8
10
24
4
8
69
85
8
20
18
NTd
NTd
12
15
7
0.6
1.3
12
a
B-Raf and pERK (B-RafV600E SK-MEK-28 cells) IC50 values are means of 2–5 experiments; individual values are within two-fold of the reported mean value.
SKMEL28 IC50 values are means of two experiments; individual values are within 35% of the reported mean value.
Intrinsic clearance data from liver microsomes, reported in units of mL/min/g liver. Values of <5 are considered low, between 5 and 8 are medium, and above 8 are high.
NT = not tested.
b
c
d
Table 4
Fluorination of the inner-phenyl ring
F
H3C
O
CH3
S
HN
R
O
N
2
5
S
F
O3
N
N
6
N
N
N
H
2
B-RafV600E IC50 (nM)
pERK EC50 (nM)
SKMEL28 EC50 (nM)
Rat Clic
Rat iv Cld,e
Rat po DNAUCd,e
a
a
b
Entry
R
23
24
25
26
H
1.2
0.5
0.9
3.8
7
11
24
23
6
8
14
47
19
10
11
8
NTf
69
48
12
42
87
58
2-F
5-F
6-F
2933
a
b
c
d
e
f
B-Raf and pERK (B-RafV600E SK-MEK-28 cells) IC50 values are means of 2–5 experiments; individual values are within three-fold of the reported mean value.
SKMEL28 IC50 values are means of two experiments; individual values are within two-fold of the reported mean value.
Intrinsic clearance data from liver microsomes, reported in units of mL/min/g liver.
Intravenous clearance reported in units of mL/min/Kg. Oral dose-normalized AUC reported as ng*h/mL/mg/kg.
Data from cassette dosing.
NT = not tested.

4440
J. C. Stellwagen et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4436–4440
3. Michaloglou, C.; Vredeveld, L. C. W.; Mooi, W. J.; Peeper, D. S. Oncogene 2007, 27, 1.
4. (a) Roberts, P. J.; Der, C. J. Oncogene 2007, 26, 3291; (b) Bollag, G.; Hirth, P.; Tsai,
J.; Zhang, J.; Ibrahim, P. N.; Cho, H.; Spevak, W.; Zhang, C.; Zhang, Y.; Habets, G.;
Burton, E. A.; Wong, B.; Tsang, G.; West, B. L.; Powell, B.; Shellooe, R.;
Marimuthu, A.; Nguyen, H.; Zhang, K. Y. J.; Artis, D. R.; Schlessinger, J.; Su, F.;
Higgins, B.; Iyer, R.; D’Andrea, K.; Koehler, A.; Stumm, M.; Lin, P. S.; Lee, R. J.;
Grippo, J.; Puzanov, I.; Kim, K. B.; Ribas, A.; McArthur, G. A.; Sosman, J. A.;
Chapman, P. B.; Flaherty, K. T.; Xu, X.; Nathanson, K. L.; Nolop, K. Nature 2010,
467, 596.
5. (a) Rominger, C.; Schaber, M.; Yang, J.; Gontarek, R.; Weaver, K.; Broderick, T.;
Carter, L.; Copeland, R.; May, E. Arch. Biochem. Biophys. 2007, 464, 130; (b)
Cell-based assays: human melanoma cells (SKMEL28; EMEM) were cultured
in the indicated medium (+10% FBS, 1% penicillin–streptomycin, 1% Sodium
pyruvate) at 37 °C in a humidified, 5% CO2 incubator. Cells were harvested
using trypsin/EDTA (Invitrogen 25,200) and counted using a haemocytometer.
For growth inhibition assays, SKMEL28 cells were plated at 500 cells/well.
Cells were treated with 20, 6.6, 2.2, 0.74, 0.25, 0.08, 0.03, 0.01, 0.003 and
exposure, we believe that the 6-F may be improving both oral
absorption as well as metabolic stability, perhaps by affording an
intramolecular hydrogen bonding interaction with the sulfon-
amide NH. Although the fluorine in this position also caused a
small decrease in the potency of 26 in cellular assays, the enhanced
oral exposure represented a key breakthrough for the series. In cel-
lular anti-proliferation assays, the corresponding HCl salt of com-
pound 26 was found to be highly selective for the B-RafV600E cell
line SK-MEL-28 versus a representative non-B-RafV600E cell line
HFF (52 nM versus 15.6 lM). In order to assess its overall kinase
selectivity profile, compound 26 was tested in a panel of 52 kinase
assays and found to have excellent overall selectivity, showing
IC50’s between 10 nM and 100 nM for only 7 out of 52 kinases
and greater than 100-fold selectivity (typically greater than
1000-fold) against the remainder. Other than B-RafV600E itself, the
most potent additional kinase activities were ActR2B, Alk5, Btk,
ErbB4, Lck, Lyn, and Src1. The combination of excellent enzyme
and cell potencies of 26, along with its promising in vivo oral expo-
sure in rat, represented an excellent platform for further optimiza-
tion into a drug candidate. Notably, higher species PK remained a
major challenge for this series, and efforts within this arylsulfona-
mide headgroup series to improve higher species PK will be re-
ported in due course.
0.001 lM compound, diluted in medium to a final DMSO concentration of
0.2%. After three days treatment time, total ATP was measured (as a surrogate
estimate of cell number) using CellTiter-Glo reagent (Promega G7571). For
the mechanistic in-cell western assay, SKMEL28 cells were plated at 20,000
cells/well and treated with compound as above. After 1 h, compound was
removed and cells were fixed with 3.7% formaldehyde/PBS. Cells were
permeabilized with 0.1% Triton X-100/PBS and blocked with LiCor blocking
buffer. Anti-ERK (Santa Cruz sc-94, 1:100) and anti-phospho-ERK (Santa Cruz
sc-7383, 1:100) primary antibodies in LiCor blocking buffer were incubated
on the cells overnight at 4 °C. Cells were washed 3Â with 0.1% Tween-20/PBS
and anti-rabbit (LiCor IRDye 680, 1:200) and anti-mouse (LiCor IRDye 800CW,
1:800) secondary antibodies in LiCor blocking buffer were added to the cells
for 1 h. Cells were washed 3Â with 0.1% Tween-20/PBS and read on the LiCor
Odyssey instrument. Percent inhibition of cell growth or ERK phosphorylation
was calculated relative to control wells; concentration of compound required
to give 50% inhibition (IC50) was determined using a 4-parameter fit.; (c)
Adjabeng, G. M.; Bifulco, N.; Davis-Ward, R. G.; Dickerson, S. H.; Donaldson, K.
H.; Petrov, K.; Rheault, T. R.; Schaaf,. G. M.; Stellwagen, J. C.; Uehling, D. E.;
Waterson, A. G. PCT Int. Appl. WO2009032667 A1, 2009, pp 278.
In conclusion, a very potent series of inhibitors of the B-RafV600E
mutant kinase has been developed, culminating with the identifi-
cation of 26. This compound combines excellent enzyme, cellular
potency, and good rat oral exposure and served as a key lead for
additional optimization of this series.
6. For detailed synthetic procedures, see Ref 5c.
7. Waterson, A. G.; Petrov, K. G.; Bifulco, N. Y.; Sammond, D. S.; Stellwagen, J. C.
Unpublished results.
Acknowledgments
8. Tsai, J.; Lee, J. T.; Wang, W.; Zhang, J.; Cho, H.; Mamo, S.; Bremer, R.; Gillette, S.;
Kong, J.; Haass, N. K.; Sproesser, K.; Li, L.; Smalley, K. S. M.; Fong, D.; Zhu, Y.-L.;
Marimuthu, A.; Nguyen, H.; Lam, W.; Liu, J.; Cheung, I.; Rice, J.; Suzuki, Y.; Luu,
C.; Settachatgul, C.; Shellooe, R.; Cantwell, J.; Kim, S.-H.; Schlessinger, J.; Zhang,
K. Y. J.; West, B. L.; Powell, B.; Habets, G.; Zhang, C.; Ibrahim, P. N.; Hirth, P.;
Artis, D. R.; Herlyn, M.; Bollag, G. Proc. Natl. Acad. Sci. 2008, 105, 3041.
9. Calculated pKa estimates were generated with ACD v.11.
The authors thank Roseanne Gerding, Neil Bifulco, Kelly Don-
aldson, Douglas Sammond, Ronda Davis-Ward, Philip Harris, Brad-
ley Heidrich, Jessica Ward, Dirk Heerding and Dash Dhanak for
experimental assistance and guidance during the preparation of
this manuscript.
References and notes
1. Hagemann, C.; Rapp, U. R. Exp. Cell Res. 1999, 253, 34.
2. Wan, P. T.; 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. Cell 2004, 116, 855.