Non-oxime pyrazole based inhibitors of B-Raf kinase

Article abstract of DOI:10.1016/j.bmcl.2010.12.038

The synthesis and biological evaluation of non-oxime pyrazole based B-Raf inhibitors is reported. Several oxime replacements have been prepared and have shown excellent enzyme activity. Further optimization of fused pyrazole 2a led to compound 38, a selec

Full text of DOI:10.1016/j.bmcl.2010.12.038

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3488–3492
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Non-oxime pyrazole based inhibitors of B-Raf kinase
Bradley J. Newhouse a, , Joshua D. Hansen a, Jonas Grina a, Mike Welch a, George Topalov a, Nicole Littman a,
⇑
Michele Callejo a, Matthew Martinson a, Sarah Galbraith a, Ellen R. Laird a, Barbara J. Brandhuber a,
Guy Vigers a, Tony Morales a, Rich Woessner a, Nikole Randolph a, Joseph Lyssikatos b, Alan Olivero b
a Array BioPharma, 3200 Walnut Street, Boulder, CO 80301, United States
b Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080-4990, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis and biological evaluation of non-oxime pyrazole based B-Raf inhibitors is reported. Several
oxime replacements have been prepared and have shown excellent enzyme activity. Further optimization
of fused pyrazole 2a led to compound 38, a selective and potent B-Raf inhibitor.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 21 October 2010
Revised 2 December 2010
Accepted 6 December 2010
Available online 17 December 2010
Keywords:
B-Raf
Oxime
Pyrazole
Raf kinases, part of the mitogen-activated protein kinase path-
way (MAPK), exist in three isoforms, A-Raf, B-Raf, and C-Raf (Raf-
1), have been shown to be critical for mediating cell proliferation
and survival.1 The Ras-Raf-MEK-ERK pathway has been implicated
in up to30% of humancancers. Muchfocus has been centered around
the serine/threonine kinase B-Raf, which is mutated in 7% of human
cancers. Activating mutations in Raf have been observed in mela-
noma (50–70%), thyroid cancers (40–70%), and ovarian cancer
(35%).2 The most frequent mutation in B-Raf is the valine for glu-
tamic acid substitution (V600E), which is found in the activation
segment of the B-Raf kinase domain.3 This accounts for a 500-fold
increase in the basal rate of MEK phosphorylation over wild-type
B-Raf, essentially rendering V600EB-Raf constitutively active.4 This
makes B-Raf an attractive target for therapeutic intervention.
Significant effort has been ongoing to develop a potent and selec-
tive B-Raf inhibitor. A handful of small molecules that are selective
have entered clinical trials, such as PLX-4032, XL-281, and RAF265,
with others in preclinical development.5 We recently reported on
pyrazole based B-Raf inhibitors that incorporate an oxime moiety
1a (Fig. 1).6 The oxime provides both a donor and acceptor to a
hydrogen bond network that is crucial to the potency and selectivity
of the series (Fig. 2). However, compound 1a showed degradation to
its indane ketone with a half-life of 1.3 hours in a simulated gastric
fluid assay. To address the oxime chemical instability and potential
metabolic liability,7 we sought to identify a suitable compound that
retained potency and selectivity.
In our previous Letter,6 a small number of modified oximes and
oxime replacements were reported. The 2,4-dihydroindeno
[1,2-c]pyrazole (herein referred to as the fused pyrazole) in com-
pound 2a (Fig. 1) showed excellent enzyme potency, encouraging
cellular activity, and no degradation in a simulated gastric fluid as-
say. Although pyrazole 2a is a potent inhibitor, the matched indane
oxime is significantly more active. We determined that the tauto-
mer preference for the pyrazole does favor the desired donor–
acceptor arrangement (Fig. 3a), and that the difference most likely
arises from less optimal distances for the NH::Glu501 contact
(Fig. 3b). Despite the lower activity, the potential for improve-
ments in metabolic stability inspired us to further develop the
fused pyrazole, as well as to explore other compounds that might
provide a similar hydrogen bonding network.
We began by exploring the properties of the fused pyrazole
(Table 1).10 By removing the methylene bridge in the fused pyra-
zole 2a, the pendant pyrazole 3a showed a 30-fold drop in potency
HO
HN
N
N
C4
N1
C3
N
N
N
N
N2
N
N
1a
2a
BRafIC50 = 0.02 nM
BRafIC50 = 2.2 nM
pERKIC50 = 33 nM
pERKIC50 = 1900 nM
⇑
Corresponding author.
E-mail address: bnewhouse@arraybiopharma.com (B.J. Newhouse).
Figure 1. Oxime and non-oxime pyrazole based B-Raf inhibitors.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.12.038

B. J. Newhouse et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3488–3492
3489
Table 1
B-Raf enzyme activity of oxime replacements
R
R
N
N
N
N
OH
N
N
a
b
Compound
R
B-Raf IC50 (nM)
68
N
N
N
N
HN
3a
HN
4a
5a
7
HN
HN
>1000
Figure 2. X-ray crystal structure (2.8 Å resolution) of B-Raf in complex with a
pyridinyl-pyrazole oxime inhibitor.8 The view highlights the hydrogen-bonding
network that involves the oxime moiety. The indane ring occupies the largely
hydrophobic gatekeeper pocket.
6a
7a
23
HN
N
>1000
H
N
HN
N
a
N
N
N
HN
N
8a
11
A
B
HN
N
9a
55
Tautomer
Erel Kcal/mol
ΔΔGsolv Kcal/mol
-11.49
HN
A
B
0
1.49
10a
11a
788
94
-9.57
O
HN
b
O
12a
13a
N
H
647
O
S
O
Figure 3. Stereoelectronic comparison of a simplified indane oxime and a fused
pyrazole. (a) Relative energies and estimated free energies of solvation for the
desired (A) and undesired (B) tautomers of the fused pyrazole suggests that the
solution population of A should be dominant.9 (b) Comparison of the optimized
geometries for the oxime (green) and fused pyrazole (violet) suggests that the ring
constraint of the pyrazole prevents close approach of the proton donor to Glu501 if
the position of the benzo-fusion is to be retained in its hydrophobic pocket; the
oxime hydroxyl also benefits from some rotational flexibility.
H2N
>1000
N
14a
15a
16a
17a
139
N
H
468
HN
N
owing to lost hydrophobic contact of the methylene with Ala481
and Thr529.11 Methyl substitution on the pendant pyrazole (4a) re-
stored a significant amount of this contact, while ethyl substitution
proved sterically excessive (5a). Tethering the ethyl group into a
six-membered fused pyrazole (6a) restored most of the activity;
the 10-fold drop in potency versus 2a is likely caused by residual
steric crowding in this very tight space. Repositioning of the do-
nor–acceptor pair from the 4,5-dihydro-2H-benzo[g]indazole 6a
to the 4,5-dihydro-2H-benzo[e]indazole 7a causes a loss of potency
that underscores the requirement for interaction with Lys483 and
Glu501. Fused triazole 8a and pendant triazole 9a displayed similar
behavior to their pyrazole counterparts. When compared to the
pendant pyrazole 3a and triazole 9a, pyrrole 10a was 4–14-fold
less potent in the enzyme assay, again emphasizing the importance
of the hydrogen bond acceptor for Lys483 (Fig. 2).
H
N
>1000
861
N
H
N
NH
18a
19a
328
OH
1.4
(continued on next page)

3490
B. J. Newhouse et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3488–3492
Table 2
Table 1 (continued)
B-Raf enzyme and pERK cell activity of N1 modified pyrazoles with a fused pyrazole
head group
Compound
R
B-Raf IC50 (nM)
77
HO
20a
21a
HN
N
R1HN
1360
157
N
R
N
HN
N
N
22a
a
R1
1 = @O
NH2
OH
Compound
R
B-Raf IC50 (nM)
pERK IC50 (nM)
230 (9 nM)
886 (10 nM)
262
pKa
23a
24a
25a
R
>1000
>1000
>1000
NH
N
26
27
28
3
9.80
8.45
8
NH
0.57
10.63
Attempts to mimic the oxime or fused pyrazole with various
fused heterocycles, amides, and sulfonamides met with limited
success (11a–18a). Lactam 11a is unable to deliver the hydrogen
bond donor near Glu501, while the trans amide 12a is sterically
encumbered by Leu527 at the back of the gatekeeper pocket. The
fused heterocycles 14a–18a either lack an acceptor, and/or are un-
able to position the donor–acceptor pair adequately.
The lipophilic naphthol 19a was a very potent inhibitor of B-Raf
enzyme at 1.4 nM and had excellent cellular activity (pERK
IC50 = 78 nM). An X-ray crystal structure of a related series showed
that the hydroxyl oxygen superimposes with the center of the
oxime N–O bond, hence the hydroxyl is able to act as both donor
and acceptor.12 However, 19a displayed a high predicted clearance
(17 ml/min/kg) in human hepatocytes. The 2-naphthol regioisomer
20a showed weaker enzyme activity, consistent with its ability to
effectively reach only Glu501.
In an attempt to access the Thr529 and Ala481 residues of the
gatekeeper pocket, we then investigated substitution at the meth-
ylene bridge of the fused pyrazole (compounds 23b–25b).13 Place-
ment of hydrogen bond donors and/or acceptors at this position
provided only weak inhibitors. Their weak potency suggested ste-
ric crowding, which was later confirmed by crystallography (vide
infra).
29
30
3.7
3.2
1671
1540
9.31
NH
NH
10.44
31
32
NH
2
0.66
22.9
396
10.35
10.35
NH2
1516
33
72.5
8164
9.56
N
H
OH
34
35
2
607 (7 nM)
4372
4.6
Numbers in parentheses are from the oxime derivatives.6
Calculated values of R group nitrogen pKa’s using ACD Labs Release 12.00,
Product version 12.00
a
Because of its promising enzyme and cellular activity, we chose
to further optimize compound 2a by varying the N1 tail of the core
pyrazole. The X-ray structure revealed that N1 groups are posi-
tioned under the Gly-rich loop and project toward the exit of the
ATP cleft.
As demonstrated in our earlier report, modifying the N1 posi-
tion with various solubilizing groups can lead to compounds with
excellent enzyme and cell activity. Piperidine substitution at the
N1 position provided one of the most potent compounds in the
oxime series. This created a sound starting point in the fused pyr-
azole series (Table 2), such as compound 26 (Fig. 4).
Lipophilic amines 26–32 (Table 2) showed excellent enzyme
activity, but varied in cellular potency. cis-Amino compound 32
displayed a 35-fold drop in potency relative to its trans counterpart
31. Since the pyrazole core is in an equatorial position on the piper-
idine ring, the amino group of compound 32 would occupy an axial
orientation, placing it closer to the Gly-rich loop of the enzyme.
Removing the conformational restriction of the piperidine ring
(33) resulted in a 25-fold loss in activity relative to 26, probably
due to loss of surface hydrophobic contacts with the Gly-rich loop.
cis-Hydroxycyclohexane 34, whose oxime counterpart is a cell po-
tent compound, proved to be much weaker in the fused pyrazole
series. Although aromatic substitution at the N1 position (R = Ph),
exemplified by 35, gave potent enzyme analogs, acceptable cellular
activity with these increasingly lipophilic compounds was not
obtained.
Figure 4. X-ray crystal structure of 26 (3.6 Å resolution) in complex with B-Raf. The
view is clipped through the N-terminal lobe with a solvent-accessible surface of the
enzyme illustrated in violet; the sidechains of the hinge residues are undisplayed
for clarity. Hydrogen bonds are depicted as dashed yellow lines. Coordinates for
Compound 26 in B-Raf have been deposited in the PDB, accession code 3PSD.
With the N1 substitution data in hand, we chose to further opti-
mize compound 26. Although its predicted hepatic clearance was
8 ml/min/kg in human liver microsomes, a concern was its low

B. J. Newhouse et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3488–3492
3491
Table 3
B-Raf enzyme and pERK cell activity of attenuated piperidine analogs
F
Br
O
O
a,b
c,d
Ts
N
N
N
O
N
H
43
N
OBn
OBn
N
HN
N
44
45
HN
N
N
N
O
i,j
e,f ,g,h
F
N
R
N
Br
N
NH
B
O
N
O
N
46
47
N
b
Compound
R
B-Raf IC50 (nM)
pERK IC50 (nM)
pKa
38
Br
HO
NH
N
36
37
38
4.9
5
4871
619
8.92
8.15
7.83
NH
NH
N
48
O
Scheme 1. Reagents and conditions: (a) N-(benzyloxycarbonyloxy) succinimide,
1:1 dioxane:10% aqueous potassium carbonate (100%); (b) MCPBA, DCM, 0 °C
(100)%; (c) compound 48, cesium carbonate, DMSO, 100 °C (30%); (d) DAST, DCM,
F
F
F
1.4
700
NH
NH
0 °C, (37%); (e) DMF–DMA, tol,
D (93%); (f) hydrazine, cat. AcOH, EtOH, D (22%); (g)
NaH, TsCl, THF, 0 °C (70%); (h) bis(pinacolato)diboron, PdCl2(dppf) DCM adduct,
KOAc, dioxane, 100 °C (97%); (i) compound 45, Pd(PPh3)4, potassium carbonate,
DME/water, 80 °C then 1 M aqueous NaOH, MeOH, 60 °C (51%); (j) Pd(OH)2, MeOH,
balloon of H2 (52%).
39
40
3.8
3.2
407
2119
8.18
iv dosing, and a trend toward greater than proportional increases
in exposure with increasing dose was observed following oral dos-
ing. It should be noted that compound 38 was 547 nM in a CYP3A4
inhibition assay.
Kinase selectivity was next determined for analog 38. This com-
pound was screened against a panel of over 200 kinases, and the
only significant activity observed was against c-Raf (9% POC at
NH
F
F
41a
42a
1.6
532
672
NH
NH
1.25
a
1
lM).
Enantiopure compounds; absolute stereochemistry not determined.
Calculated values of piperazine nitrogen pKa’s using ACD Labs Release 12.00,
b
The synthesis of compound 38 is outlined in Scheme 1. Com-
Product version 12.00.
pound 44 was prepared by Cbz protection of amine 43, followed
by m-CPBA epoxidation gave intermediate 44. Epoxide ring open-
ing by compound 486 gave a mixture of regioisomers (attack at
the C-4 and C-3 of the epoxide, C4 attack predominating).15 Subse-
quent treatment with DAST gave fluoride 45. It is interesting to
note that the product of retention was observed for this fluorina-
tion,16 which was confirmed by 1H NMR analysis. The desired fused
pyrazole coupling partner 47 was prepared from 5-bromoindanone
46 via an enaminoketone intermediate, followed by cyclization,
tosylation (one regioisomer) and boronate ester formation. Cou-
pling of 45 and 47 under standard Suzuki conditions followed by
de-protection afforded compound 38. Similar synthetic protocols
were used to obtain other oxime replacement analogs.
We have prepared several fused pyrazole B-Raf inhibitors that
showed excellent enzyme activity, encouraging cellular potency,
and chemical stability over its oxime counterpart. Optimization
of the fused pyrazole inhibitor led to fluorine substituted piperi-
dine 38 which showed modest oral absorption in mice and good ki-
nase selectivity. Further progress on these inhibitors will be
reported in due course.
Caco-2 permeability (A–B Papp (10À6 cm/s) of 1.88, efflux ra-
tio = 6.17), possibly due to the basicity of the piperidine. A series
of substituted piperidines were then prepared with the intention
of modifying the basicity of the piperidine nitrogen to improve
their physiochemical properties, and their oral absorption (Table
3).
Utilizing established techniques to attenuate amine basicity,14
compounds 36–42 were synthesized. Compounds 37–39, 41 and
42 essentially had similar cell potencies. No strong enantiomeric
(41 and 42) or diastereomeric (38 and 39) preferences were ob-
served. Hydroxypiperidine 36 showed an approximate sevenfold
drop in cell activity relative to its fluoropiperidine counterpart
38, possibly due to its more basic nature. Fluoropiperidine 40 suf-
fered a similar fate likely due to the nitrogen now being one carbon
atom farther away from the fluorine. Caco-2 permeability of com-
pound 38 showed an A–B Papp (10À6 cm/s) of 11.36 (efflux ratio of
1.55), a significant improvement over compound 26 which had an
A–B Papp (10À6 cm/s) of 1.88 (efflux ratio = 6.17). Similar trends
were observed in the oxime series.
We next explored the pharmacokinetic properties of compound
38 (Table 4). CD-1, male mice were dosed (60% PEG400/40% saline
solution) 2.5 mg/kg intravenously (iv) and 25, 50, or 100 mg/kg, or-
ally (PO). High clearance was observed (91.1 ml/min/kg) following
Acknowledgment
The authors thank Andrew Allen for NMR analysis and Susan
Rhodes, Jennifer Otten, and Michelle Livingston for Caco-2, P450,
and solubility determinations.
Table 4
Pharmacokinetics of compound 38 in CD-1 mice
References and notes
iv
po
po
po
1. (a) Hoshino, R.; Chantani, Y.; Yamori, T.; Tsuruo, T.; Oka, H.; Yoshida, O.;
Shimada, Y.; Ari-I, S.; Wada, H.; Fujimoto, J.; Kohno, M. Oncogene 1999, 18, 813–
822; (b) Li, N.; Batt, D.; Warmuth, M. Curr. Opin. Invest. Drugs 2007, 8, 452–456.
2. (a) 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, E.; Ewing, R.;
Floyd, Y.; Gray, K.; Hall, S.; Hawes, R.; Hughes, J.; Kosmidou, V.; Menzles, A.;
Dose (mg/kg)
AUC ( g h/ml)
2.5
25
5.16
—
50
13.4
—
100
38.8
—
l
0.406
91.1
8.74
CL (ml/min/kg)
Vss (l/kg)
—
—
—

3492
B. J. Newhouse et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3488–3492
Mould, C.; Parker, A.; Stevens, C.; Watt, S.; Hooper, S.; Wilson, R.; Jayatilake, H.;
9. (a) Tautomer preferences were computed using the 6-31G(d) basis set. Frisch,
M. J. et al. GAUSSIAN98, Revision A.11, Gaussian: Pittsburgh, PA.; (b) Free energies
of solvation were calculated using SM2.1 Amsol 5.0: Cramer, C. J.; Hawkins, G.
D.; Lynch, G. C.; Giesen, D. J.; Rossi, I; Storer, J. W.; Truhlar, D. G.; Liotard, D. A.
In AMSOL-version 5.0, Quantum Chemistry Program Exchange Program 606, based
in part on AMPAC-version 2.1, Liotard, D. A.; Healy, E. F.; Ruiz, J. M.; Dewar, M. J.
S.
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.; Seigler, H. F.; Darrow, T. L.; Paterson, H.; Marais, R.; Marshall, C. J.;
Wooster, R.; Stratton, M. R.; Futreal, P. A. Nature 2002, 417, 949–954; (b) Cohen,
Y.; Xing, M.; Mambo, E.; Guo, Z.; Wu, G.; Trink, B.; Beller, U.; Westra, W. H.;
Ladenson, P. W.; Sidransky, D. J. Natl. Cancer Inst. 2003, 95(8), 625–627; (c) Xu,
X.; Quiros, R. M.; Gattuso, P.; Ain, K. B.; Prinz, R. A. Cancer Res. 2003, 63(15),
4561–4567.
10. For assay description see: Laird, E.; Lyssikatos, J.; Welch, M.; Grina, J.; Hansen,
J.; Newhouse, B.; Olivero, A.; Topolav, G. WO 2006/084015 A2, 2006.
11. Computed relative energies for the two pyrazole tautomers indicates
a
3. (a) Kalinsky, K.; Haluska, F. G. Expert Rev. Anticancer Ther. 2007, 7, 715–724; (b)
Dhillon, D. S.; Hagan, S.; Rath, O.; Kolch, W. Oncogene 2007, 26, 3279–3290.
4. Wan, P. T.; Garnet, 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,
855–867.
significant population of the undesired tautomer also exists (data not shown).
12. Ren, L.; Wenglowsky, S.; Miknis, G.; Rast, B.; Buckmelter, A. J.; Ely, R. J.;
Schlachter, S.; Laird, E. R.; Randolph, N.; Callejo, M.; Martinson, M.; Galbraith,
S.; Brandhuber, B. J.; Vigers, G.; Morales, T.; Voegtli, W.; Lyssikatos, J. Bioorg
Med. Chem. Lett. 2011, 21, 1243–1247.
5. McCubrey, J. A.; Milella, M.; Tafuri, A.; Martelli, A. M.; Lunghi, P.; Bonai, A.;
Cervello, M.; Lee, J. T.; Steelman, L. S. Curr. Opin. Invest. Drugs 2008, 9, 614–630.
6. Hansen, J. D.; Grina, J.; Newhouse, B.; Welch, M.; Topalov, G.; Littman, N.;
Callejo, M.; Gloor, S.; Martinson, M.; Laird, E.; Brandhuber, B. J.; Vigers, G.;
Morales, T.; Woessner, R.; Randolph, N.; Lyssikatos, J.; Olivero, A. Bioorg. Med.
Chem. Lett. 2008, 18, 4692–4695.
7. (a) Choo, E. F.; Driscoll, J. P.; Feng, J.; Liederer, B.; Plise, E.; Randolph, N.; Shin, Y.;
Wong, S.; Ran, Y. Xenobiotica 2009, 39(09), 700–709; (b) Coutts, R. T.; Dawe, R.;
Dawson, G. W.; Kovach, S. H. Drug Metab. Disp. 1976, 4, 35–39; (c) Heberling, S.;
Girreser, U.; Wolf, S.; Clement, B. Biochem. Pharmacol. 2006, 71, 354–365.
8. PDB accession code 3D4Q.
13. Tao, Z.-F.; Sowin, T.; Lin, N.-H. Tetrahedron Lett. 2005, 46, 7615–7618.
14. (a) Hagmann, W. J. Med. Chem. 2008, 51, 4359–4369; (b) Bohm, H.-J.; Banner,
D.; Bendels, S.; Kansay, M.; Kuhn, B.; Muller, K.; Obst-Sander, U.; Stahl, M.
ChemBioChem 2004, 5, 637–643; (c) van Niel, M. B.; Collins, I.; Beer, M. S.;
Broughton, H. B.; Cheng, S. K. F.; Goodacre, S. C.; Heald, A.; Locker, K. L.;
MacLeod, A. M.; Morrison, D.; Moyes, C. R.; O’Connor, D.; Pike, A.; Rowley, M.;
Russell, M. G. N.; Sohal, B.; Stanton, J. A.; Thomas, S.; Verrier, H.; Watt, A. P.;
Castro, J. J. Med. Chem. 1999, 42, 2087–2104.
15. Determined by 1H NMR.
16. Shellhamer, D. F.; Briggs, A. A.; Miller, B. M.; Prince, J. M.; Scott, D. H.; Heasley,
V. L. J. Chem. Soc., Perkin Trans. 2 1996, 973–977.