Discovery of novel allosteric mitogen-activated protein kinase kinase (MEK) 1,2 inhibitors possessing bidentate Ser212 interactions

Article abstract of DOI:10.1021/jm2017094

Using structure-based design, two novel series of highly potent biaryl amine mitogen-activated protein kinase kinase (MEK) inhibitors have been discovered. These series contain an H-bond acceptor, in a shifted position compared with previously disclosed compounds, and an adjacent H-bond donor, resulting in a bidentate interaction with the Ser212 residue of MEK1. The most potent compound identified, 1 (G-894), is orally active in in vivo pharmacodynamic and tumor xenograft models.

Full text of DOI:10.1021/jm2017094

Article
pubs.acs.org/jmc
Discovery of Novel Allosteric Mitogen-Activated Protein Kinase
Kinase (MEK) 1,2 Inhibitors Possessing Bidentate Ser212 Interactions
Robert A. Heald,*,† Philip Jackson,† Pascal Savy,† Mark Jones,† Emanuela Gancia,† Brenda Burton,†
Richard Newman,† Jason Boggs,‡ Emily Chan,‡ Jocelyn Chan,‡ Edna Choo,‡ Mark Merchant,‡
Patrick Rudewicz,‡ Mark Ultsch,‡ Christian Wiesmann,‡ Qin Yue,‡ Marcia Belvin,‡ and Steve Price†
†Argenta, 8/9 Spire Green Centre, Flex Meadow, Harlow, Essex CM19 5TR, U.K.
‡Genentech Inc., 1 DNA Way, South San Francisco, California 94080, United States
S
* Supporting Information
ABSTRACT: Using structure-based design, two novel series of highly potent biaryl amine
mitogen-activated protein kinase kinase (MEK) inhibitors have been discovered. These series
contain an H-bond acceptor, in a shifted position compared with previously disclosed
compounds, and an adjacent H-bond donor, resulting in a bidentate interaction with the
Ser212 residue of MEK1. The most potent compound identified, 1 (G-894), is orally active in
in vivo pharmacodynamic and tumor xenograft models.
INTRODUCTION
■
The Ras-Raf-MEK-ERK mitogen-activated protein kinase
signaling pathway is one of the most promising targets for a
new wave of molecularly targeted antitumor agents.1
Constitutive activation of the extracellular regulated kinase
(ERK) cascade, through oncogenic forms of Ras and mutations
in B-Raf, has been observed in lung, colon, pancreas, kidney,
and ovary primary human tumor samples.2 Similarly, hyper-
activity by upstream receptor tyrosine kinases such as human
epidermal growth factor receptor 2 (HER2) leads to increased
signaling through the ERK pathway.3 Research efforts in recent
years have uncovered a variety of inhibitors of Raf, MEK, and
ERK kinases as potential antitumor agents.4
MEK is a crucial node in the ERK MAPK pathway: the only
known downstream targets of MEK1/2 are ERK1/2. This has
led to particularly intense efforts to discover inhibitors of this
target, and there are now 13 MEK inhibitors at some stage of
Figure 1. Structures of designed analogue 1 with 2, 3, and 4. “A” and
“B” rings of 2 are shown.
clinical evaluation for the treatment of cancer, with two others
having been withdrawn.5,6 The first MEK inhibitors to progress
past phase I clinical trials were 2 (AZD6244/Selumetinib,
Figure 1)7−9 and PD325901,10 both of which were progressed
to phase II studies as single agents for the treatment of a variety
of tumor types. Studies with 2 as a single agent and in
combination with other antitumor agents are continuing.11 The
most advanced MEK inhibitor is GSK1120212,12 which began a
phase III trial in BRAF V600E positive melanoma patients in
late 2010.
synergistic inhibition of tumor growth in basal-like breast
cancer models.13 Additionally, it has been demonstrated that in
K-Ras mutant cell lines, Raf inhibition primes wild-type Raf to
activate the Raf-MEK-ERK pathway, providing a clear rational
for combination therapy with MEK inhibitors.14
Allosteric MEK1/MEK2 inhibitors15 are particularly discrim-
inating kinase inhibitors. These compounds form a tertiary
complex with the MEK enzyme and the ATP binding site and
are thus uncompetitive.16,17 The binding site for so-called type
III kinase inhibitors is proximal to ATP and results in high
Although MEK inhibition alone is expected to provide
significant therapeutic benefit, there is considerable evidence
for synergistic combination with other targeted agents. For
example, concerted MEK and PI3K inhibition provides
Received: December 20, 2011
Published: April 16, 2012
© 2012 American Chemical Society
4594
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
specificity for the target enzyme. It has been postulated that this
allosteric “MEK pocket” may be present in other kinases.18
Examination of structural features of inhibitors disclosed in
the patent literature reveals that the allosteric binding pocket of
MEK1 is highly discriminating in its affinity for drug molecules,
resulting in low structural diversity.15,19 The structures of seven
compounds that have entered the clinic have been disclosed: all
are biaryl amines, by far the most common class of MEK
inhibitor. The potency of these inhibitors is postulated to
depend on three pharmacophoric elements (Figure 2).16 The
of this H-bond acceptor23 and position relative to the aniline
and hydroxamate are determined by the “A-ring” to which these
moieties are attached. It has been proposed that increasing the
strength of the Ser212/H-bond acceptor interaction results in
greater tolerance in the other pharmacophoric elements.24
RESULTS AND DISCUSSION
■
In our search for additional MEK inhibitors, we utilized a
computer-aided drug design (CADD) driven approach
leveraging key information from the X-ray crystal structure
(2.7 Å resolution) of 3 (G-925, Figure 1), a compound from
our first series of novel inhibitors25 which was optimized to give
the orally efficacious analogue 4 (G-573).26 From 2-D overlay
of 2 and 3, it was noticed that the two A ring systems position
the H-bond acceptor differently. Additionally, it was realized
that for a 6,5-heterobicyclic system fusion of the five-membered
ring toward the B-ring, in the opposite sense to 2, together with
movement of the acceptor round one position, placed the key
pharmocophoric elements described above in remarkably
similar position to 3. An H-bond donor was added next to
the acceptor, as it was postulated that this position would be
proximal to the carbonyl of Ser212 resulting in the design of 1
(G-894, Figure 1). This designed compound is fundamentally
different to all six-membered aryl and 6,5-fused biaryl amine
MEK inhibitors disclosed in the literature in that the putative
H-bond acceptor is not positioned on the six-membered ring
meta to the aniline. Modeling of the indazole 1 in the 3/MEK1
protein structure was performed using docking (Figure 3, A)
and manual placement (Figure 3, B), as docking algorithms
underestimate the iodine Val127 interaction. The docked pose
shows bidentate Ser212 interaction, but the aniline ∼1.3 Å out
of the lipophilic pocket compared with 3. In the manual
overlay, the indazole acceptor was still appropriately positioned
near Ser212, with the NH donor shifted slightly from the
Ser212 carbonyl to a distance of 3.0 Å. Although neither pose
of 1 was ideal, the protein movement required to enable all
desired contacts is minimal.
Figure 2. The key interactions of biarylaniline MEK inhibitors: crystal
structure of MEK1 in complex with ATP (green) and PD318088
(orange) (PDB code 1S9J).
first of these is a “B-ring”, which occupies the lipophilic pocket
formed by Leu118, Ile126, Val127, Ile141, Met143, Phe129,
Phe209, and Val211, an important feature of which is para-
substitution with a lipophilic polarizable atom or group capable
of forming an interaction with the carbonyl of Val127. Where
this atom is a halogen, the interaction is termed a halogen
bond.20,21 The second feature is a polar group which can
interact with the terminal phosphate of ATP and Lys97,
typically a hydroxamate, but sometimes an amide, reverse
sulfonamide, or sulfonyl urea.22 The third feature is an H-bond
acceptor, which is suitably juxtaposed to Ser212. The strength
It should be noted that it is not possible to achieve a formal
bidentate Ser212 interaction using monoaryl or 6,5-fused biaryl
MEK inhibitor scaffolds which have been previously disclosed.
The possibility of identifying a MEK inhibitor making such a
bidentate interaction with Ser212, which could lead to
increased affinity, led us to target the synthesis of 1.
Figure 3. (A) Overlay between the docked pose of 1 (orange) and the crystal structure of 3 (cyan, PDB code 3V01). (B) Overlay between 1
manually adjusted in the binding site (orange) and the crystal structure of 3 (cyan).
4595
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthetic Route to Indazole Analogues 1 and 12
a
(a) (COCl)2, DMF, DCM, (ii) tBuOH, pyridine, DCM; (b) (i) LDA, THF, (ii) DMF; (c) N2H4·H2O DME, 90 °C; (d) Boc2O, Et3N, DCM; (e)
Pd2dba3, xantphos, K3PO4, toluene, 90 °C; (f) (i) ICl, DCM or NBS, DCM, (ii) TFA, DCM; (g) EDCI, DIPEA, HOBt, DMF; (h) MeOH, HCl.
The desired 4-phenylamino-5-carboxylic acid core was
synthesized following the route outlined in Scheme 1.
Commercially available 2-bromo-4-fluorobenzoic acid was
converted to the tert-butyl ester 5 under standard conditions.
Regiospecific metalation with LDA followed by DMF quench
gave aldehyde 6. Cyclization of the ortho-fluoro aldehyde 6 to
indazole 7 and subsequent reaction with boc-anhydride
provided the appropriately substituted and protected indazole
core 8. Palladium-catalyzed amination with 2-fluoro-4-trime-
thylsilyl aniline using xantphos as the ligand25 gave anilino ester
9. Regiospecific iodination, directed by the trimethylsilyl group,
followed by simultaneous boc-deprotection and ester hydrol-
ysis, provided 4-(2-fluoro-4-iodo-phenylamino)-1H-indazole-5-
carboxylic acid 10a. Installation of the hydroxamate side-chain
could then be achieved via a coupling reaction with vinyl
hydroxylamine 11a followed by acid-mediated deprotection.
Bromo analogue 12 was prepared similarly from intermediate 9
using NBS rather than iodine monochloride for the
halogentation step. Diol analogue 13 was also prepared from
acid intermediate 10a.
In a biochemical homogeneous time-resolved fluorescence
(HTRF) assay, 1 was found to be ∼2.5-fold more potent than 4
(Table 1), demonstrating that this designed analogue is a
potent inhibitor of MEK. Potency was further assessed in
HCT116 (K-Ras mutant human colorectal carcinoma cell line)
and A375 (BRAF V600E mutant human melanoma cell line)
cell lines, determining both the inhibition of ERK phosphor-
ylation and cell proliferation. In these assays, 1 is consistently
∼10-fold more potent than 4. Bromo analogue 12 is less potent
than 1, as would be expected by reduction in strength of the
Val127 halogen bond. However, 12 is equipotent with 4 in
cellular assays, again demonstrating the excellent potency
facilitated by the indazole scaffold. Results for 4 and 13 show a
direct comparison between the azabenzofuran and indazole
scaffolds. The indazole 13 is marginally more potent in the
Table 1. Activity of Indazole Analogues 1, 12, and 13
Compared with Azabenzofurans 3 and 4
cell proliferation
EC50 (nM)
P-ERK EC50 (nM)
compd MEK HTRF IC50 (nM) HCT116 A375 HCT116 A375
1
13
25
0.2
6.3
4.0
3.4
0.4
0.4
34.0
7.2
16
314
175
193
31
2
264
23
3
4
35
12
13
125
15
6.4
19
3.7
12
HTRF assay, equipotent in A375 cells, but much more potent
in HCT116 cellular assays. The improved potency in cellular
but not biochemical assays of the indazole compared with
azabenzofurans is intriguing. As all of these compounds are
uncompetitive with respect to ATP, it is understandable that
cellular potency is greater than biochemical potency due to high
intracellular concentrations of ATP. As the biochemical assay
used a constitutively activated kinase construct, cellular data
were taken to be more indicative of true potency. Differences in
proliferation values between the two cell lines used are most
likely due to differential sensitivity, as the shift between A375
and HCT116 cell lines is consistently 8−10-fold. 1 shows high
specificity for MEK inhibition with <40% inhibition at 1 μM
against a panel of 100 kinases (Invitrogen SelectScreen, see
Supporting Information), indicating that inclusion of the
indazole was not leading to promiscuous kinase hinge-binding.
An X-ray cocrystal structure (2.7 Å resolution) of 1 with
MEK1 and ATP was obtained which confirmed the allosteric
mode of binding and bidentate interaction with Ser212 (Figure
4, modeled hydrogen atoms shown for clarity) and a structure
consistent with the manual overlay shown above (Figure 3,
panel B). As expected, the iodo-aniline is positioned in the
lipophilic pocket with the iodine atom ∼3.3 Å from the
4596
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
Information). Bioavailability in mouse at 25 mg/kg was higher
perhaps due to saturation of clearance mechanisms. The bromo
analogue 12 was found to have higher clearance in rat than the
iodo analogue despite lower lipophilicity, and bioavailability was
found to be low. Diol hydroxamate 13 was not progressed into
PK as it exhibited low MDCK cell permeability (A:B 2.0, B:A
7.1 × 10−6 cm/s efflux ratio 3.5). The azabenzofuran 4 has
much lower clearance and higher oral exposure than 1 in mice.
Despite low oral exposure in rodents. the high potency of 1
in cellular assays warranted further investigation of this
compound in an efficacy study. Gratifyingly, when dosed orally
in nude HCT-116 tumor-bearing mice, 1 exhibited substantially
greater efficacy at 5 mg/kg than 4 at the same dose (Figure 5),
Figure 4. Crystal structure of 1 in complex with MEK1 (orange, PDB
code 3V04); modeled hydrogens shown for clarity. For comparison,
the crystal structure of MEK1 in complex with 3 is also shown (cyan).
carbonyl of Val127, an appropriate distance for formation of a
halogen bond. With the aniline positioned ideally for lipophilic
contacts and halogen bond, the indazole is able to make a
donor−acceptor interaction with Ser212. The interatom
distance for both the H-bonding atom pairs is 2.58 Å with
angles of 134° and 142° for the indazole N Ser NH and
indazole NH Ser CO interactions. Comparison with the crystal
structure of 3 shows a small but significant induced-fit
movement of Ser212 (when compared with the highly
conserved positions of other residues in the pocket) which
allows the three pharmacophoric elements to interact as
desired: a shift of 0.48 Å is observed for the carbonyl oxygen.
The shift in Ser212 to accommodate the donor−acceptor pair
of 1 indicates that the postulated bidentate interaction exists
and most adequately explains the potency of this compound.
When dosed in rats, the rate of clearance of 1 was found to
be moderate ∼2/3 liver blood flow (Table 2) and oral exposure
and bioavailability were low. Similar results were found in mice
where clearance was approximately equal to liver blood flow
and oral exposure (AUC) was low. Low oral bioavailability in
rat was attributed to high liver extraction ratio and poor
solubility (solubility 0.08 mg/mL (pH 6.5), 0.18 mg/mL
(FASSIF); measured MDCK permeability was moderate to
high with low efflux (A:B 9.9 B:A 13.8 × 10−6 cm/s; ratio: 1.4).
Analysis of plasma samples following oral dosing in rats showed
that glucoronidation and cleavage of the hydroxamate to acid
were the major routes of metabolism (see Supporting
Figure 5. Efficacy in HCT116 colorectal xenograft model. Mice were
randomized and treated with vehicle or 1 at 5, 25, and 75 mg/kg or 4
at 5 mg/kg, orally (PO) every day (QD) for 3 weeks. \ = death or
euthanasia: 2 animals in the vehicle treated group had tumor volume
>2000 mm3; animals euthanized in the compound treated groups had
body weight loss >20%.
with tumor growth inhibition (TGI) approaching stasis (94%).
The 5 mg/kg dose was well tolerated with no loss of body
weight compared with vehicle (data not shown). At the 25 and
75 mg/kg doses, 5 and 10 out of 10 partial responses were
observed but ∼10% loss of body weight was noted at the higher
dose. This improved efficacy compared with 4 can be largely
attributed to increased potency. The free fraction for 1 in mice
is ∼2-fold higher than for 4, which is perhaps within the error
of the experiment, whereas the exposure of 1 is ∼10-fold
higher. Efficacy was found to correlate with inhibition of ERK
phosphorylation in a separate pharmacodynamic experiment: a
single dose of 1 at 25 mg/kg led to maximal inhibition of pERK
Table 2. Rodent PK Data for 1, 4, and 12
a
b
c
d
e
f
compound
species
rat
PPB
dose (mg/kg)/route
Cl (mL/min/kg)
T1/2 (h)
Vdss (L/kg)
AUC (μM·h)
F (%)
1
96.1
97.0
1/IV
36
1.4
2.4
2.0
1.01
1.06
0.41
1.96
5/PO(solution)
1/IV
21
94
mouse
mouse
rat
88
0.6
2.9
25/PO(suspension)
4
98.4
96.5
1/IV
7.7
54
2.3
0.71
4.42
192
25/PO(suspension)
173
12
1/IV
0.36
1.5
0.75
1.29
5/PO(solution)
34
a
b
c
d
e
f
Plasma protein binding. Clearance. Half-life. Volume of distribution at steady state. Area under the curve. Bioavailability. For oral studies, MCT
was used for suspension dosing and 60% PEG 400 for solution.
4597
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
Table 3. Activity Data for Further Indazole Analogues
a
Scheme 2. Synthetic Route to Benzoisothiazole Analogues
a
(a) Boc2O, DMAP, tBuOH, 45 °C; (b) NBS, AIBN, CCl4, reflux; (c) trimethylamine oxide, DCM, DMSO, rt; (d) BnSH, tBuOK, THF, −30 °C;
(e) (i) SOCl2, DCM, rt, (ii) NH3, MeOH, THF, rt; (f) LHMDS, THF, −78 °C; (g) TFA, DCM, rt; (h) RONH2, EDCI, HOBt, DIPEA, THF, rt; (i)
HCl, MeOH, rt.
(∼70% of control values) through 8 h (see Supporting
Information).
Owing to the difficulty in balancing potency, metabolic
stability, and oral bioavailability in the indazole series, our
thoughts turned to other scaffolds that take advantage of our
discovery of a bidentate Ser212 interaction and movement in
the position of the H-bond acceptor. It has been postulated that
the CH of some five-membered heterocycles can act as a weak
H-bond donor.27,28 Additionally, it is likely that our earlier
azabenzofuran template is positioned appropriately to make a
similar interaction. Replacing the fused pyrazole ring of 1 with
an isothiazole would provide a system able to make a C−H···O
donor−acceptor interaction with Ser212 with reduced polarity.
It was considered that the lower energy of H-bond interaction
may be compensated for by a reduced desolvation penalty
compared with the indazole.
In an attempt to optimize the PK/potency balance, a number
of analogues were prepared from acid intermediate 10a that
investigated variation of the hydroxamate group (Table 3).
These analogues were assessed using in vitro or in vivo
experiments as appropriate. Cyclopropylmethyl (14) and ethyl
(15) hydroxamate analogues were found to be significantly less
potent than 1, and both showed no oral exposure when dosed
at 5 mg/kg in rats (data not shown). Replacement of the
hydroxamate with amides was not tolerated: 16 exhibited an
IC50 of >10 μM in the HTRF assay, and the azetidinol analogue
17 possessed very weak activity. In general, it appears that the
indazole scaffold, despite enabling a bidentate donor−acceptor
interaction with Ser212, is less tolerant of changes to the
hydroxamate compared with previously disclosed pyridone
monocyclic inhibitors.24 This is perhaps due to less flexibility in
positioning of the donor−acceptor element of the indazole
compared with the monodentate pyridone.
The synthesis of targeted analogues 18 and 19 (Scheme 2)
depended on preparation of a benzoisothiazole with an
unprecedented substitution pattern. The required intermediate,
tetra-substituted derivative 22, was prepared in two steps from
ester 20 via radical bromination followed by oxidation with
4598
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
trimethylamine oxide and DMSO. Regiospecific SNAr displace-
ment at the 3-position with phenylmethanethiol was achieved
using potassium tert-butoxide at low temperature to furnish the
desired compound 23. The isothiazole ring was formed
following a two-step procedure,29 providing the 6,7-disubsti-
tuted benzoisothiazole scaffold 24. The 2-fluoro-4-iodoaniline
was introduced through SNAr displacement of the remaining
fluorine atom using LHMDS at low temperature, giving the
aniline ester 25. Subsequent steps were carried out as for the
indazole analogues.
Pleasingly, hydroxyethyl hydroxamate analogue 18 showed
excellent potency in all assays and is only ∼2−3-fold less potent
than the direct indazole analogue 1 in P-ERK and cell
proliferation assays (Table 4). These data prove that a formal
permeability the unbound fraction of drugs drives efficacy, it
was concluded that compound 19 does not improve on 1 and it
was not tested further. Compound 18 was not progressed
further, as it is a potent CYP2C9 and CYP2C19 inhibitor.
CONCLUSION
■
In summary, using structure-based design leveraging key
information from a previously disclosed azabenzofuran series,
we have discovered novel allosteric MEK inhibitors which,
partly through strengthened interaction with Ser212, display
improved potency in cellular assays. The most potent
compound discovered, indazole 1, is efficacious at low doses
in a human tumor xenograft model.
The relative contributions to potency of the C−H and N−H
H-bond donors in the benzoisothiazole and indazole scaffolds
cannot be determined from comparison of the biochemical and
cellular data presented here. However, it is clear that the H-
bond acceptor of both these scaffolds is ideally positioned and
facilitates excellent potency in either case. This result is at odds
with previous evidence from the literature which suggests that
the H-bond acceptor should be positioned next to the ring
junction on 6,5-fused scaffolds, as in 2. In this regard, the
topology of MEK inhibitors such as 1 and 19 is unique.
Additionally, taken together, the data on benzofuran, indazole,
and benzoisothiazole biaryl amine allosteric MEK inhibitors
show that the design of the bicyclic A-ring is critical to
achieving a balance of potency and desirable ADME properties.
Further details of related compounds which build on these
discoveries will be disclosed in due course.
Table 4. Activity Data for Benzoisothiazole Analogues
cell proliferation
EC50 (nM)
P-ERK EC50 (nM)
compd MEK HTRF IC50 (nM) HCT116 A375 HCT116 A375
18
19
15
20
0.6
0.9
0.9
1.1
44
62
7
10
H-bond donor on this novel 6,5-fused scaffold is not necessary
for excellent potency. Diol analogue 19 is 2-fold less potent
than indazole analogue 13 in the HTC116 cell proliferation
assay and equipotent in A375 cells.
Calculated physicochemical properties for selected com-
pounds with varying A-rings are shown in Table 5. These data
show that cell shifts for both the benzoisothiazole and indazole
analogues are much reduced compared with azabenzofuran
analogue 3. Azabenzofuran analogue 4, which has a less polar
hydroxamate substituent and an A-ring substituted with
fluorine, has more similar cell shifts to both benzoisothiazoles
and indazole 13. High donor and acceptor count (>11) is the
strongest correlate with increased cell shift; compare 1 and 13
and 3 and 4, presumably due to reduced membrane
permeability. This difference is more marked in A375 cells.
From these data, it is apparent that the indazole and
benzoisothiazole scaffolds achieve increased cell potency by
more optimal positioning of polar functionality.
When dosed in rats, compound 19 showed much reduced
clearance and ∼5-fold higher oral exposure compared with 1
(Table 6). However, as 19 is highly plasma protein bound, the
unbound clearance is no better than for 1 (1875 cf. 923 mL/
min/kg) and unbound oral AUC is very similar (0.046 cf. 0.040
μM·h). As in the absence of active transport or restricted
EXPERIMENTAL SECTION
■
Chemistry. General Experimental Conditions. All solvents and
1
reagents were used as obtained. H NMR spectra were recorded at
ambient temperature using a Varian Unity Inova (400 MHz)
spectrometer with a triple resonance 5 mm probe. Chemical shifts
are expressed in ppm relative to tetramethylsilane.
LC-MS experiments to determine retention times (RT) and
associated mass ions were performed using various methods which
are fully described in the Supporting Information. Compounds 3 and 4
were prepared according to the patented procedures.25 All final
compounds were assessed for purity by LC-MS and found to be ≥95%
purity.
2-Bromo-4-fluoro-benzoic Acid tert-Butyl Ester (5). To a
suspension of 2-bromo-4-fluoro-benzoic acid (28.5 g, 0.13 mol) in
DCM (500 mL) at rt was added oxalyl chloride (11.35 mL, 0.26
mmol) followed by DMF (0.05 mL, catalytic, CARE vigorous gas
evolution), and the reaction mixture stirred for 3 h. The reaction
mixture was concentrated in vacuo and the residue dissolved in DCM
(500 mL) before treatment with a solution of tert-butanol (28.5 g, 0.26
Table 5. Physicochemical Properties for Selected Azabenzofuran, Indazole, And Bezoisothiazole Analogues
a
b
c
d
e
f
compd
scaffold
NAcc
NDon
TPSA
ClogP
HCT116 cell shift
A375 cell shift
3
4
azabenzofuran
8
7
4
3
117
97
2.11
3.42
13
5
11
0.65
1
indazole
7
7
8
4
4
5
99
99
3.32
3.06
2.49
1.2
1.5
2.0
0.15
0.15
0.8
12
13
120
18
19
benzoisothiazole
6
7
3
4
84
3.67
2.84
3.1
0.50
0.47
104
2.9
a
b
c
d
Number of hydrogen bond acceptors. Number of hydrogen bond donors. Topological polar surface area. Calculated using Daylight v4.94.
e
f
(HCT cell proliferation EC50)/(MEK HTRF IC50). (A375 cell proliferation EC50)/(MEK HTRF IC50).
4599
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
Table 6. Rat PK Data for 19
a
b
c
d
e
f
compd
species
rat
PPB
99.2
dose (mg/kg)/route
Cl (mL/min/kg)
T1/2 (h)
Vdss (L/kg)
AUC (μM·h)
F (%)
19
1/IV
15
1.6
0.97
2.2
5.7
5/PO(solution)
51
a
b
c
d
e
f
Plasma protein binding. Clearance. Half-life. Volume of distribution at steady state. Area under the curve. Bioavailability. Vehicle for oral
dosing: 60% PEG 400.
mol) and pyridine (20.5 g, 0.26 mol). The resultant mixture was
stirred at rt for 3 days before it was diluted with DCM and washed (1
M aqueous sodium hydroxide, water, 0.1 M aqueous HCl, water),
dried (Na2SO4), filtered, and concentrated in vacuo to give a yellow
oil. The crude oil was subjected to flash chromatography (Si-PPC,
gradient 0−20% ethyl acetate in cylcohexane) to give the title
4-(2-Fluoro-4-iodophenylamino)-1H-indazole-5-carboxylic
Acid (10a). To a solution of 4-(2-fluoro-4-trimethylsilanylphenylami-
no)-indazole-1,5-dicarboxylic acid di-tert-butyl ester (5.0 g, 10 mmol)
in DCM (20 mL) at 0 °C was added iodine monochloride as a
solution in DCM (20 mL, 1N, 20 mmol). The reaction mixture was
stirred at 0 °C for 15 min and then diluted with saturated aqueous
sodium thiosulfate solution (10 mL) and extracted with DCM. The
combined organic extracts were washed with water, dried (Na2SO4),
and concentrated in vacuo to yield the title compound as a tan foam.
The foam was dissolved in DCM (25 mL) and TFA (15 mL) added.
The reaction mixture was stirred at rt for 2 h before being
concentrated in vacuo. The resultant residue was triturated in
cyclohexane to give the title compound as a gray−pink solid (4.1 g,
quantitative). LC-MS (method B): RT = 3.23 min, [M + H]+ = 398.
1H NMR (DMSO-d6) δ ppm 10.2 (1 H, s), 7.84 (1 H, d, J = 8.9 Hz),
1
compound as a colorless oil (15.2 g, 42%). H NMR (CDCl3, 400
MHz) δ ppm 7.76−7.73 (1H, m), 7.37 (1H, dd, J = 8.4, 2.5 Hz), 7.05
(1H, ddd, J = 8.7, 7.7, 2.5 Hz), 1.60 (9H, s).
2-Bromo-4-fluoro-3-formyl-benzoic Acid tert-Butyl Ester (6).
To a cold (−78 °C) solution of diisopropylamine (11.3 mL, 80.3
mmol) in THF (200 mL) was added n-butyllithium (30 mL, 2.5 M in
hexanes, 75 mmol), and the mixture stirred for 30 min. A solution of
2-bromo-4-fluoro-benzoic acid tert-butyl ester (20.0 g, 73 mmol) in
THF (10 mL) was added dropwise, and the pale-orange solution
stirred at −75 °C for 1.5 h. DMF (10.6 g, 146 mmol) was added, and
the mixture stirred for a further 10 min before quenching with water.
The products were partitioned between diethyl ether and water, and
the organic layer was separated, washed with brine, dried (Na2SO4),
filtered, and concentrated in vacuo to give the title compound as a
7.78 (1 H, d, J = 9.9 Hz), 7.58 (1 H, d, J = 8.3 Hz), 7.16 (1 H, t, J = 8.5
Hz), 7.02 (1 H, s), 6.96 (1 H, d, J = 8.9 Hz).
4-(2-Fluoro-4-iodophenylamino)-1H-indazole-5-carboxylic
Acid (2-vinyloxyethoxy)-amide. To a solution of 4-(2-fluoro-4-
iodophenylamino)-1H-indazole-5-carboxylic acid (2.14 g, 5.39 mmol)
and O-(2-vinyloxy-ethyl)-hydroxylamine (668 mg, 6.50 mmol) in
DMF (50 mL) was added EDCI hydrochloride (1.14 g, 5.93 mmol),
HOBt (0.80 g, 5.93 mmol), and DIPEA (1 mL, 5.93 mmol). The
reaction mixture was stirred at rt for 2 h before being concentrated in
vacuo. The resultant residue was dissolved in ethyl acetate (30 mL),
washed with aqueous saturated sodium bicarbonate solution (300
mL), and the aqueous phase extracted with ethyl acetate (2 × 20 mL).
The combined organic extracts were washed with brine (30 mL), dried
(MgSO4), filtered, and concentrated in vacuo. The resultant residue
was subjected to flash chromatography (SiO2, gradient 0−100%
ethylacetate in cyclohexane) to give the title compound as a pale-
yellow solid (1.85 g, 71%). LC-MS (method B): RT = 3.52 min, [M −
1
yellow solid (20 g, 90%). H NMR (CDCl3, 400 MHz) δ ppm 10.4
(1H, s), 7.80−7.73 (1H, m), 7.17 (1H, t, J = 9.1 Hz), 1.62 (9H, s).
4-Bromo-1H-indazole-5-carboxylic Acid tert-Butyl Ester (7).
A biphasic solution of 2-bromo-4-fluoro-3-formyl-benzoic acid tert-
butyl ester (19.5 g, 65 mmol), DME (80 mL), and hydrazine hydrate
(30 mL) was heated at 90 °C for 1.5 h. After cooling, volatile solvent
was removed in vacuo and the products were partitioned between
ethyl acetate and saturated aqueous ammonium chloride. The organic
extract was dried (Na2SO4), filtered, and concentrated in vacuo to give
the title compound as a tan solid (17.2 g, 89%). LC-MS (method B):
RT = 3.63 min, [M + CH3CN + H]+ = 338/340, [M − H]− = 295/
297. 1H NMR (CDCl3) δ ppm 8.22 (1 H, d, J = 1.0 Hz), 7.82 (1 H, d,
J = 8.7 Hz), 7.44 (1 H, dd, J = 8.7, 1.0 Hz), 1.43 (9 H, s).
1
H]+ = 481. H NMR (CD3OD) δ ppm 7.57 (1 H, dd, J = 10.2, 1.9
Hz), 7.53 (1 H, d, J = 8.7 Hz), 7.45 (1 H, ddd, J = 8.4, 1.9, 1.1 Hz),
7.37 (1 H, s), 7.07 (1 H, d, J = 8.8 Hz), 6.96 (1 H, t, J = 8.6 Hz), 6.50
(1 H, dd, J = 14.3, 6.8 Hz), 4.23 (1 H, dd, J = 14.4, 2.05 Hz), 4.16−
4.15 (2 H, m), 3.99−3.99 (3 H, m).
4-Bromo-indazole-1,5-dicarboxylic Acid di-tert-Butyl Ester
(8). To a solution of 4-bromo-1H-indazole-5-carboxylic acid tert-butyl
ester (12.0 g, 40 mmol) and triethylamine (6.1 mL, 44 mmol) in
DCM (200 mL) was added di-tert-butyl-dicarbonate (9.6 g, 44 mmol),
and the reaction mixture stirred at rt for 18 h. The reaction mixture
was diluted with DCM, washed (saturated aqueous NaHCO3, water),
dried (Na2SO4), filtered, and concentrated in vacuo in the presence of
diatomaceous earth. The crude product was subjected to flash
chromatography (Si-PPC, gradient 0−10% ethyl acetate in cyclo-
hexane) to give the title compound as an off-white solid (8.5 g, 50%).
1H NMR (CDCl3) δ ppm 8.29 (1H, s), 8.18−8.10 (1H, m), 7.92 (1H,
4-(2-Fluoro-4-iodophenylamino)-1H-indazole-5-carboxylic
Acid (2-Hydroxyethoxy)-amide (1). To a solution of 4-(2-fluoro-4-
iodophenylamino)-1H-indazole-5-carboxylic acid (2-vinyloxyethoxy)-
amide (1.85 g, 3.84 mmol) in methanol (40 mL) was added
hydrochloric acid (3 mL, 1N, 3 mmol). The reaction mixture was
stirred at rt for 1 h, during which an off-white solid precipitated. The
reaction mixture was concentrated in vacuo and the residue triturated
with hot methanol/water (10 mL, 1:1). The product was collected by
filtration and dried in vacuo to yield the title compound as an off white
solid (1.26 g, 72%). LC-MS (method A): RT = 8.28 min, [M + H]+ =
457. 1H NMR (DMSO-d6) δ ppm 13.20 (1 H, s), 11.61 (1 H, s), 9.93
(1 H, s), 7.66 (1 H, dd, J = 10.3, 1.9 Hz), 7.46 (1 H, d, J = 8.8 Hz),
7.42 (1 H, dd, J = 8.5, 1.8 Hz), 7.24 (1 H, s), 7.01 (1 H, d, J = 8.8 Hz),
6.91 (1 H, t, J = 8.6 Hz), 4.68 (1 H, s), 3.85 (2 H, dd, J = 5.4, 4.5 Hz),
3.56 (2 H, t, J = 4.8 Hz). HRMS: calculated isotopic Mw (M + Na),
478.9992; observed isotopic Mw (M + Na), 479.0080; confirmed
formula, C16H14FIN4O3; high resolution error, 18.4 ppm.
2,3-Difluoro-4-methyl-benzoic Acid tert-Butyl Ester (20). A
mixture of 2,3-difluoro-4-methyl-benzoic acid (20.0 g, 116 mmol), di-
tert-butyl dicarbonate (25.0 g, 116 mmol), and DMAP (2.0 g, 16.4
mmol) in tert-butanol (500 mL) was stirred at 45 °C for 5 h before
being concentrated in vacuo. The resultant residue was triturated in
diethyl ether and filtered. The filtrate was concentrated in vacuo to
give a residue which was partitioned between ethyl acetate and a 1 M
d, J = 8.7 Hz), 1.70 (9H, s), 1.63 (9H, s).
4-(2-Fluoro-4-trimethylsilanyl-phenylamino)indazole 1,5-Di-
carboxylic Acid di-tert-Butyl Ester (9). A solution of 2-fluoro-4-
trimethylsilanyl-phenylamine (4.4 g, 24 mmol) in toluene (80 mL)
was added to a mixture of 4-bromo-indazole-1,5-dicarboxylic acid di-
tert-butyl ester (8.0 g, 20 mmol), Pd2dba3 (456 mg, 2.5 mol %),
xantphos (576 mg, 5 mol %), and potassium phosphate tribasic (5.96
g, 28 mmol) under nitrogen. The atmosphere was evacuated and
backfilled with nitrogen, and then the reaction mixture heated at 100
°C for 4 h. The cooled reaction mixture was diluted with ethyl acetate,
filtered through Celite, and the filtrate concentrated in vacuo. The
resultant residue was subjected to flash chromatography (Si-PPC, dry
loading on diatomaceous earth, gradient 0−10% ethyl acetate in
cyclohexane) to give a solid which was triturated in methanol to give
1
the title compound as a white solid (7.02 g, 70%). H NMR (CDCl3)
δ ppm 10.03 (1H, s), 8.05 (1H, d, J = 9.0 Hz), 7.57 (1H, dd, J = 9.0,
0.8 Hz), 7.28−7.18 (4H, m), 1.68 (9H, s), 1.63 (9H, s), 0.28 (9H, s).
4600
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
aqueous solution of hydrochloric acid. The organic layer was separated
and washed with a saturated aqueous solution of sodium hydrogen
carbonate followed by brine, dried (Na2SO4), filtered, and
concentrated in vacuo to give the title compound as a colorless oil
(17.3 g, 65%). 1H NMR (CDCl3) δ ppm 7.51 (1 H, ddd, J = 8.3, 6.6,
1.9 Hz), 6.95 (1 H, m), 2.33 (3 H, d, J = 2.3 Hz), 1.59 (9 H, s).
4-Bromomethyl-2,3-difluoro-benzoic Acid tert-Butyl Ester
(21). A solution of 2,3-difluoro-4-methyl-benzoic acid tert-butyl ester
(17.3 g, 75.9 mmol) and N-bromosuccinimide (13.5 g, 75.9 mmol) in
carbon tetrachloride (250 mL) was degassed for 10 min. AIBN (1.2 g,
7.32 mmol) was added, and the reaction mixture was stirred at reflux
for 18 h before being cooled to rt and filtered. The filtrate was
concentrated in vacuo to give a residue which was subjected to flash
chromatography (Si-PPC, gradient 0−10% TBME in cyclohexane) to
afford the title compound as a colorless oil (16.7 g, contaminated by
25% of starting material). 1H NMR (CDCl3) δ ppm 7.61 (1 H, ddd, J
= 8.3, 6.4, 2.1 Hz), 7.18 (1 H, ddd, J = 8.1, 6.4, 1.8 Hz), 4.49 (2 H, d, J
= 1.3 Hz), 1.59 (9 H, s).
saturated aqueous solution of ammonium chloride and extracted with
ethyl acetate. The organic layer was separated and washed with water
followed by brine, dried (Na2SO4), filtered, and evaporated in vacuo.
The residue was subjected to flash chromatography (Si-PPC, gradient
0−10% TBME in cyclohexane) to give the title compound as a yellow
solid (717 mg, 30%). LC-MS (method B): RT = 5.14 min, [M + H]+ =
471. 1H NMR (CDCl3) δ ppm 9.85 (1 H, s), 8.77 (1 H, s), 7.99 (1 H,
dd, J = 8.6, 0.5 Hz), 7.49−7.48 (3 H, m), 6.90 (1 H, t, J = 8.3 Hz), 1.65
(9 H, s).
7-(2-Fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-car-
boxylic Acid (26). To a solution of 7-(2-fluoro-4-iodo-phenylamino)-
benzo[d]isothiazole-6-carboxylic acid tert-butyl ester (710 mg, 1.51
mmol) in DCM (5 mL) were added water (0.15 mL) and TFA (5
mL). The reaction mixture was stirred at rt for 1 h before being
concentrated in vacuo. The resultant residue was azeotroped with
toluene to give the title compound as a yellow solid (571 mg, 91%).
1
LC-MS (method B): RT = 3.95 min, [M + H]+ = 415. H NMR
(CDCl3) δ ppm 9.69 (1 H, s), 8.79 (1 H, s), 8.08 (1 H, d, J = 8.6 Hz),
7.55−7.54 (2 H, m), 7.49 (1 H, d, J = 8.6 Hz), 7.01 (1 H, t, J = 8.3
Hz).
2,3-Difluoro-4-formyl-benzoic Acid tert-Butyl Ester (22). To a
solution of 4-bromomethyl-2,3-difluoro-benzoic acid tert-butyl ester
(12.9 g, 42.1 mmol) in DMSO (80 mL) and DCM (40 mL) at 0 °C
was added trimethylamine N-oxide (3.4 g, 45.3 mmol). The reaction
mixture was stirred at rt for 18 h before being concentrated in vacuo.
The resultant residue was partitioned between iced water and ethyl
acetate. The organic layer was separated and washed with brine twice,
dried (Na2SO4), filtered, and evaporated in vacuo. The resultant
residue was subjected to flash chromatography (Si-PPC, gradient 0−
10% TBME in cyclohexane) to afford the title compound as a white
7-(2-Fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-car-
boxylic Acid (2-Vinyloxy-ethoxy)-amide. To a solution of 7-(2-
fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-carboxylic acid
(328 mg, 0.79 mmol) and DIPEA (0.41 mL, 2.37 mmol) in DMF
(2 mL) were added O-(2-vinyloxy-ethyl)-hydroxylamine (163 mg, 1.58
mmol), EDCI (303 mg, 1.58 mmol), and HOBt (213 mg, 1.58 mmol).
The reaction mixture was stirred for 18 h at rt and then diluted with
ethyl acetate and washed with water, then a saturated aqueous solution
of sodium bicarbonate, and then brine before being dried (Na2SO4),
filtered, and concentrated in vacuo. The resultant residue was
subjected to flash chromatography (Si-PPC, gradient 0−100%
TBME in cyclohexane) to afford the title compound as a yellow
foam (194 mg, 49%). LC-MS (method B): RT = 3.99 min, [M + H]+ =
500. 1H NMR (CDCl3) δ ppm 9.38 (1 H, s), 8.99 (1 H, s), 8.81 (1 H,
s), 8.29−6.73 (2 H, m), 6.82 (1 H, t, J = 8.8 Hz), 6.53 (1 H, dd, J =
14.4, 6.8 Hz), 4.33 (2 H, m), 4.24 (1 H, m), 4.11 (1 H, m), 4.03 (2 H,
m).
1
solid (4.34 g, 43%). H NMR (CDCl3) δ ppm 10.38 (1 H, d, J = 0.8
Hz), 7.71 (1 H, dddd, J = 7.4, 5.6, 1.7, 0.8 Hz), 7.63 (1 H, ddd, J = 7.4,
5.6, 1.5 Hz), 1.61 (9 H, s).
3-Benzylsulfanyl-2-fluoro-4-formyl-benzoic Acid tert-Butyl
Ester (23). To a solution of potassium tert-butoxide (2.0 g, 17.9
mmol) in anhydrous THF (80 mL) was added benzyl mercaptan (2.1
mL, 17.9 mmol). The reaction mixture was stirred at rt for 5 min
before being cooled to −30 °C. A solution of 2,3-difluoro-4-formyl-
benzoic acid tert-butyl ester (4.34 g, 17.9 mmol) in anhydrous THF
(20 mL) was added dropwise over 15 min, and the resultant mixture
was stirred at −30 °C for 30 min before being quenched by addition of
water and extracted with ethyl acetate. The organic layer was
separated, washed with water followed by brine, dried (Na2SO4),
filtered, and concentrated in vacuo to give the title compound as a
yellow oil (6.2 g, 100%). 1H NMR (CDCl3) δ ppm 10.20 (1 H, d, J =
0.6 Hz), 7.84 (1 H, ddd, J = 8.0, 6.8, 0.7 Hz), 7.59 (1 H, dd, J = 8.0, 0.9
Hz), 7.19 (3 H, m), 7.05 (2 H, m), 4.07 (2 H, s), 1.63 (9 H, s).
7-Fluoro-benzo[d]isothiazole-6-carboxylic Acid tert-Butyl
Ester (24). To a solution of 3-benzylsulfanyl-2-fluoro-4-formyl-
benzoic acid tert-butyl ester (6.20 g, 17.9 mmol) in DCM (100 mL)
was added sulfuryl chloride (2.9 mL, 35.8 mmol). The reaction
mixture was stirred at rt for 1 h before being concentrated in vacuo.
The resultant residue was azeotroped with toluene twice and then
taken up in THF (50 mL). The resultant solution was cooled to 0 °C,
and a 2 M solution of ammonia in methanol (100 mL) was added. The
reaction mixture was stirred at rt for 1 h and then concentrated in
vacuo. The resultant residue was partitioned between ethyl acetate and
a saturated aqueous solution of sodium hydrogen carbonate. The
organic layer was separated and washed with water followed by brine,
dried (Na2SO4), filtered, and evaporated in vacuo. The residue was
subjected to flash chromatography (Si-PPC, gradient 0−10% diethyl
ether in pentane) to give the title compound as a yellow solid (2.96 g,
65%). 1H NMR (CDCl3) δ ppm 8.93 (1 H, dd, J = 4.1, 0.5 Hz), 7.91
(1 H, ddd, J = 8.3, 5.8, 0.5 Hz), 7.84 (1 H, d, J = 8.3 Hz), 1.64 (9 H, s).
7-(2-Fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-car-
boxylic Acid tert-Butyl Ester (25). To a solution of 7-fluoro-
benzo[d]isothiazole-6-carboxylic acid tert-butyl ester (1.26 g, 5.0
mmol) and 2-fluoro-4-iodo-phenylamine (1.18 g, 5.0 mmol) in
anhydrous THF (25 mL) at −78 °C was added a 1.0 M solution of
LHMDS in hexanes (10.0 mL, 10.0 mmol) under a nitrogen
atmosphere. The reaction mixture was allowed to warm to rt and
then stirred for 30 min before being quenched by the addition of a
7-(2-Fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-car-
boxylic Acid (2-Hydroxy-ethoxy)-amide (18). To a solution of 7-
(2-fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-carboxylic acid
(2-vinyloxy-ethoxy)-amide (187 mg, 0.37 mmol) in methanol (4
mL) was added a 1.0 M aqueous solution of hydrochloric acid (0.75
mL, 0.75 mmol). The reaction mixture was stirred at rt for 1 h before
being concentrated under reduced pressure. The residue was taken up
in ethyl acetate and washed with a saturated aqueous solution of
sodium bicarbonate and brine, dried (Na2SO4), filtered, and
concentrated in vacuo. The residue was subjected to flash
chromatography (Si-PPC, gradient 0−100%, ethyl acetate in DCM)
to give the title compound as a yellow solid (128 mg, 72%). LC-MS
1
(method A): RT = 9.81 min, [M + H]+ = 474. H NMR (CDCl3) δ
ppm 9.33 (1 H, s), 8.86 (1 H, s), 8.77 (1 H, s), 7.55−7.40 (4 H, m),
6.84 (1 H, t, J = 8.3 Hz), 4.10 (2 H, t, J = 4.2 Hz), 3.91 (1 H, t, J = 6.4
Hz), 3.80 (2 H, t, J = 4.5 Hz).
7-(2-Fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-car-
boxylic Acid ((R)-2,2-Dimethyl-[1,3]dioxolan-4-ylmethoxy)-
amide. To a solution of 7-(2-fluoro-4-iodo-phenylamino)-benzo[d]-
isothiazole-6-carboxylic acid (377 mg, 0.91 mmol) and DIPEA (0.47
mL, 2.73 mmol) in DMF (4 mL) were added O-((R)-2,2-dimethyl-
[1,3]dioxolan-4-ylmethyl)-hydroxylamine (268 mg, 1.82 mmol), EDCI
(349 mg, 1.82 mmol), and HOBt (246 mg, 1.82 mmol). The reaction
mixture was stirred for 18 h at rt before being diluted with ethyl
acetate. The resultant solution was washed with water followed by a
saturated aqueous solution of sodium hydrogen carbonate and then
brine before being dried (Na2SO4), filtered, and concentrated in vacuo.
The resultant residue was subjected to flash chromatography (Si-PPC,
gradient 0−100% TBME in cyclohexane) to give the title compound
as a yellow solid (266 mg, 54%). LC-MS (method B): RT = 4.04 min,
[M + H]+ = 544. 1H NMR (CDCl3) δ ppm 8.88 (1 H, s), 8.63 (1 H,
s), 8.28 (1 H, s), 7.04 (1 H, d, J = 8.3 Hz), 6.98 (1 H, dd, J = 9.5, 1.9
4601
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
Hz), 6.94 (2 H, m), 6.30 (1 H, t, J = 8.3 Hz), 3.96 (1 H, m), 3.62−3.60
(3 H, m), 3.35 (1 H, m), 0.95 (3 H, s), 0.88 (3 H, s).
Molecular Modeling Methods. All molecular modeling experiments
were carried out using software from Schrodinger Inc. The crystal
̈
structure of MEK1 (extracted from the cocrystal with 3, PDB
7-(2-Fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-car-
boxylic Acid ((R)-2,3-Dihydroxy-propoxy)amide (19). To a
solution of 7-(2-fluoro-4-iodo-phenylamino)-benzo[d]isothiazole-6-
carboxylic acid ((R)-2,2-dimethyl-[1,3]dioxolan-4-ylmethoxy)-amide
(263 mg, 0.48 mmol) in methanol (10 mL) was added a 1.0 M
aqueous solution of hydrochloric acid (0.97 mL). The reaction mixture
was stirred at rt for 18 h before being concentrated in vacuo. The
resultant residue was taken up in ethyl acetate, washed with a saturated
aqueous solution of sodium hydrogen carbonate and then brine, dried
(Na2SO4), filtered, and concentrated in vacuo. The resultant residue
was subjected to flash chromatography (Si-PPC, gradient 0−100%
ethyl acetate in DCM) to give the title compound as a yellow solid
(127 mg, 53%). LC-MS (method A): RT = 9.05 min, [M + H]+ = 504.
1H NMR (CD3OD) δ ppm 8.88 (1 H, s), 7.74 (1 H, d, J = 8.4 Hz),
reference 3V01) was prepared using the Protein Preparation Wizard in
Maestro (Protein Preparation Wizard, Schrodinger, LLC, New York,
̈
NY), which assigns bond orders, adds hydrogen atoms, deletes water
molecules, and generates appropriate protonation states. The prepared
protein structure was then minimized using OPLS2001 (until the rmsd
reached 0.3 Å) and used to generate the receptor grids for docking.
The binding site was defined using the native ligand and selecting the
option “docking ligands similar in size”. The ligand structures were
built in Maestro and prepared for docking using LigPrep 2.0 (LigPrep,
Schrodinger, LLC, New York, NY). The docking experiments were
̈
carried out with Glide 4.030 in Extra Precision mode, using the default
parameters setting. The top five docking poses for each ligand were
selected and visually inspected.
X-Ray Crystal Structure Determination Method. Cloning and
Expression. The truncated (residues 63−393)MEK1 was cloned from
a wt MEK1 full length construct via PCR and subcloned into pET24b
(Novagen) as previously described.16 The recombinant MEK1 was
expressed in Escherichia coli (Rosetta 2 pLys S) and grown in Ultra
Yield Flasks (Thomson Instrument Company) TB media/MOPS/
glycerol at 37 °C to an O.D. of 1 (600 nm). The cells were then placed
at 16 °C and induced after 30 min with 0.5 mM IPTG and allowed to
grow an additional 18 h.
Protein Purification and Crystallization. MEK1 was purified using
Cobalt immobilized metal affinity resin (TALON), ion exchange, and
size exclusion chromatography. The protein was concentrated to 15
mg/mL and incubated with 10-fold molar excess inhibitor plus 1 mM
MgAMP-PNP before crystallization. MEK1 crystals grew from hanging
drop vapor diffusion using 12% w/v PEG 8000, 0.4 M NH4H2PO4,
and 0.1 M HEPES pH 6.9 at 18 °C. MEK1 crystals belonging to the
space group P62 and unit cell a = 81 Å and c = 129 Å were previously
described.16
7.61 (1 H, d, J = 8.3 Hz), 7.53 (1 H, dd, J = 10.0, 1.9 Hz), 7.44 (1 H,
ddd, J = 8.4, 1.9, 1.1 Hz), 6.76 (1 H, t, J = 8.5 Hz), 4.05 (1 H, dd, J =
10.0, 3.5 Hz), 3.96−3.83 (2 H, m), 3.63−3.52 (2 H, m). HRMS:
calculated isotopic Mw (M + Na), 525.9710; observed isotopic Mw (M
+ Na), 525.9601; confirmed formula, C17H15FIN3O4S; high resolution
error, −20.07 ppm.
In Vitro Methods for the Determination of MEK Inhibition,
Cellular Levels of Phospho-ERK, and Cell Proliferation. MEK1
Inhibition Assay. A homogeneous time-resolved fluorescence assay
(HTRF) was used to test for inhibition of kinase activity. A
constitutively active MEK1 was obtained from Invitrogen and was
tested at a concentration of 15 nM with 50 μM ATP, 50 nM unactive
ERK1, and the test compound in a 25 mM HEPES buffer, pH 7.5,
containing 10 mM MgCl2, 5 mM β-glycerophosphate, 0.1 mM sodium
orthovanadate, and 0.1 mg/mL Triton X 100. The kinase assay was
carried out for 30 min at 25 °C and was terminated by the addition of
EDTA. HTRF reagents (CisBiol) were added and time-resolved
fluorescence was read using a Wallac Victor V (Perkin-Elmer). IC50
values quoted are the result of at least two separate determinations
where the results were within 2-fold of eachother and were calculated
using the KLfit software package (version 2.0.5).
Assay for Cellular Levels of Phospho-ERK. Cells (HCT116 or
A375) were seeded at 26700/well into collagen-coated 96-well plates
and were incubated overnight under normal tissue culture conditions.
The following day, the cells were exposed to MEK inhibitors for 2 h.
At the end of this time, the cells were fixed and permeabilized by
incubation in 2% formaldehyde for 45 min followed by 100% ice-cold
methanol for 10 min. The fixed and permeabilized cells were washed
with PBS and blocked with 3% BSA in Tris buffered saline plus Tween
20 (TBST) for 1 h at 37 °C. At the end of this time, the blocking
solution was replaced with an antiphospho-ERK antibody raised in
rabbit (New England Biolabs) at a dilution of 1:1000 made in blocking
solution and the cells were incubated overnight at 4 °C, followed by
washing and incubation with antirabbit-Ig− Alexa Flour 488
(Invitrogen) at a dilution of 1:1000 in blocking solution for 2 h at
rt. Following washing, the cells were incubated with 1.5 μM
propridium iodide for 2 h in the dark and read on a high content
imager (Acumen, TTP LabTech) in the red and green channels to
measure P-ERK content and cell number. EC50 values quoted are the
result of at least two separate determinations, where the results were
within 2-fold of eachother and were calculated using the KLfit software
package (version 2.0.5).
Cell Proliferation Assay. Cells (HCT116 or A375) were seeded at
26700/well into collagen-coated 96-well plates and were incubated
overnight under normal tissue culture conditions. The following day,
the cells were exposed to MEK inhibitors for 72 h under normal tissue
culture conditions. At the end of this time, the cells were equilibrated
to rt and mixed 1:1 with Cell Titer Glo reagent prepared as indicated
by the supplier (Promega), followed by a 10 min incubation.
Luminescence was read using a TopCount (Perkin-Elmer). EC50
values quoted are the result of at least two separate determinations,
where the results were within 2-fold of eachother and were calculated
using the KLfit software package (version 2.0.5).
Structure Determination and Refinement. Synchrotron data was
collected at the Advanced Light Source (ALS) beamline 5.0.2.
Diffraction data to 2.7 Å at 100 K was processed using HKL200031
suite of programs. Molecular replacement was performed using
MOLREP in CCP432 suite with PDB accession code 1S9J as the
starting model. Refinement was conducted using REFMAC5.2. The
refinement statistics are summarized in the Supporting Information.
ASSOCIATED CONTENT
* Supporting Information
■
S
Methods and further results for the in vivo pharmacodynamic
and efficacy studies involving 1; experimental procedures for
the preparation of compounds 12−17; NMR spectra for 1 and
19; data collection and refinement statistics for complexes of 3
and 1 with MEK1; kinase panel data for 1; rat in vivo
metabolite identification for 1. This material is available free of
charge via the Internet at http://pubs.acs.org.
Accession Codes
The coordinates for the crystal structures of 1 and 3 with
MEK1 have been deposited with the RCSB Protein Data Bank
under the accession codes 3V04 and 3VO1, respectively.
AUTHOR INFORMATION
Corresponding Author
■
*Phone: +44 (0) 1279 645623. Fax: +44 (0) 1279 645646. E-
mail: robert.heald@glpg.com.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Kirk Robarge, Mark Zak, and Paul Gibbons for
helpful discussions and collection of experimental data.
■
4602
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
Seki, N.; Inaba, T.; Kawasaki, H.; Yamaguchi, T.; Kakefuda, R.;
Nanayama, T.; Kurachi, H.; Hori, Y.; Yoshida, T.; Kakegawa, J.;
Watanabe, Y.; Gilmartin, A. G.; Richter, M. C.; Moss, K. G.; Laquerre,
S. G. Discovery of a Highly Potent and Selective MEK Inhibitor:
GSK1120212 (JTP-74057 DMSO Solvate). ACS Med. Chem. Lett.
2011, 2 (4), 320−324.
(13) Hoeflich, K. P.; O’Brien, C.; Boyd, Z.; Cavet, G.; Guerrero, S.;
Jung, K.; Januario, T.; Savage, H.; Punnoose, E.; Truong, T.; Zhou, W.;
Berry, L.; Murray, L.; Amler, L.; Belvin, M.; Friedman, L. S.; Lackner,
M. R. In Vivo Antitumor Activity of MEK and Phosphatidylinositol 3-
Kinase Inhibitors in Basal-Like Breast Cancer Models. Clin. Cancer Res.
2009, 14, 4649−4664.
(14) Hatzivassiliou1, G.; Song, K.; Yen, I.; Brandhuber, B. J.;
Anderson, D. J.; Alvarado, R.; Ludlam, M. J. C.; Stokoe, D.; Gloor, S.
L.; Vigers, G.; Morales, T.; Aliagas, I.; Liu, B.; Sideris, S.; Hoeflich, K.
P.; Jaiswal, B. J.; Seshagiri, S.; Koeppen, H.; Belvin, M.; Friedman, L.
S.; Malek, S. RAF inhibitors prime wild-type RAF to activate the
MAPK pathway and enhance growth. Nature 2010, 464, 431−435.
(15) Price, S. Putative allosteric MEK1 and MEK2 inhibitors. Expert
Opin. Ther. Pat. 2008, 18, 603−627.
ABBREVIATIONS USED
■
AMP-PNP, adenylyl-imidodiphosphate; DIPEA, diisopropyle-
thylamine; EDCI, 1-ethyl-3-(3′-dimethylaminopropyl)-
carbodiimide hydrochloride; HEPES, 4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid; HOBt, 1-hydroxybenzotriazole;
IPTG, isopropyl-β-D-thiogalactopyranoside; MDCK, Madin−
Darby canine kidney cells; MOPS, 3-(N-morpholino)-
propanesulfonic acid; Pd2dba3, tris-(dibenzylideneacetone)-
dipalladium(0); SEM, standard error of the mean; Si-PPC,
prepacked silica flash chromatography cartridge (Isolute SPE,
Biotage SNAP or ISCO Redisep); TBME, tert-butyl methyl
ether; xantphos, 4,5-bis(diphenylphosphino)-9,9-dimethylxan-
thene
REFERENCES
■
(1) Roberts, P. J.; Der, C. J. Targeting the Raf-MEK-ERK mitogen-
activated protein kinase cascade for the treatment of cancer. Oncogene
2007, 26, 3291−3310.
(16) Ohren, J. F.; Chen, H.; Pavlovsky, A.; Whitehead, C.; Zhang, E.;
Kuffa, P.; Yan, C.; McConnell, P.; Spessard, C.; Banotai, C.; Mueller,
W. T.; Delaney, A.; Omer, C.; Sebolt-Leopold, J. S.; Dudley, D. T.;
Leung, I. K.; Flamme, C.; Warmus, J.; Kaufman, M.; Barrett, S.; Tecle,
H.; Hasemann, C. A. Structures of human MAP kinase kinase 1
(MEK1) and MEK2 describe novel noncompetitive kinase inhibition.
Nature Struct. Mol. Biol. 2004, 11, 1192−1197.
(17) Fischmann, T. O.; Smith, C. K.; Mayhood, T. W.; Myers, J. E.,
Jr.; Reichert, P.; Mannarino, A.; Carr, D.; Zhu, H.; Wong, J.; Yang, R.;
Le, H. V.; Madison, V. S. Crystal structures of MEK1 binary and
ternary complexes with nucleotides and inhibitors. Biochemistry 2009,
48, 2661−2674.
(18) Tecle, H.; Shao, J.; Li, Y.; Kothe, M.; Kazmirski, S.; Penzotti, J.;
Ding, Y.-H.; Ohren, J.; Moshinsky, D.; Coli, R.; Jhawar, N.; Bora, E.;
Jacques-O’Hagan, S.; Wu, J. Beyond the MEK-pocket: Can current
MEK kinase inhibitors be utilized to synthesize novel type III NCKIs?
Does the MEK-pocket exist in kinases other than MEK? Bioorg. Med.
Chem. Lett. 2009, 19, 226−229.
(19) Trujillo, J. I. MEK Inhibitors: a patent review. Expert Opin. Ther.
Pat. 2011, 21 (7), 1045−1069.
(20) Politzer, P.; Lane, P.; Concha, M. C.; Ma, Y.; Murray, J. S. An
overview of halogen bonding. J. Mol. Model. 2007, 13, 305−311.
(21) Voth, A. R.; Ho, P. S. The role of halogen bonding in inhibitor
recognition and binding by protein kinases. Curr. Top. Med. Chem.
2007, 7, 1336−1348.
(22) Iverson, C.; Larson, G.; Lai, C.; Yeh, L.-T.; Dadson, C.;
Weingarten, P.; Appleby, T.; Vo, T.; Maderna, A.; Vernier, J. M.;
Hamatake, R.; Miner, J. N.; Quart, B. RDEA119/BAY 869766: A
potent, selective, allosteric inhibitor of MEK1/2 for the treatment of
cancer. Cancer Res. 2009, 69, 6839−6847.
(23) Laurence, C.; Brameld, K. A.; Graton, J.; Le Questel, J. Y.;
Renault, E. The pKBHX database: toward a better understanding of
hydrogen-bond basicity for medicinal chemists. J. Med. Chem. 2009,
52, 4073−4086.
(24) Wallace, E. M.; Lyssikatos, J.; Blake, J. F.; Seo, J.; Yang, H. W.;
Yeh, T. C.; Perrier, M.; Jarski, H.; Marsh, V.; Poch, G.; Livingston, M.
G.; Otten, J.; Hingorani, G.; Woessner, R.; Lee, P.; Winkler, J.; Koch,
K. Potent and selective mitogen-activated protein kinase kinase
(MEK) 1,2 inhibitors. 1. 4-(4-Bromo-2-fluorophenylamino)-1-meth-
ylpyridin-2(1H)-ones. J. Med. Chem. 2006, 49, 441−444.
(25) Price, S.; Williams, K.; Savy, P. P.; Dyke, H. J.; Montana, J. G.;
Stanley, M. S.; Bao, L. Aza-benzofuranyl compounds and methods of
use. Patent WO2008024725, 2008.
(26) Choo, E. F.; Belvin, M.; Chan, J.; Hoeflich, K.; Orr, C.; Robarge,
K.; Yang, X.; Zak, M.; Boggs, J. Preclinical disposition and
pharmacokinetics−pharmacodynamic modeling of biomarker response
and tumour growth inhibition in xenograft mouse models of G-573, a
MEK inhibitor. Xenobiotica 2010, 11, 751−762.
(2) Hoshino, R.; Chantani, Y.; Yamori, T.; Tsuruo, T.; Oka, H.;
Yoshida, O.; Shimada, Y.; Ari-I, S.; Wada, H.; Fujimoto, J.; Kohno, M.
Constitutive activation of the 41-/43-kDa mitogen-activated protein
kinase signaling pathway in human tumors. Oncogene 1999, 18, 813−
822.
(3) Olayioye, M. A.; Neve, R. M.; Lane, H. A.; Hynes, N. E. The
ErbB signaling network: receptor heterodimerization in development
and cancer. EMBO J. 2000, 19, 3159−3167.
(4) Sebolt-Leopold, J. S. Advances in the development of cancer
therapeutics directed against the RAS-mitogen-activated protein kinase
pathway. Clin. Cancer Res. 2008, 14, 3651−3656.
(5) Thomson Reuters Integrity; Thomson Reuters: East Haven, CT;
http://integrity.prous.com/ (Accessed December 10, 2011).
́
(6) Fremin, C.; Meloche, S. From basic research to clinical
development of MEK1/2 inhibitors for cancer therapy. J. Hematol.
Oncol. 2010, 3, 8.
(7) Yeh, T. C.; Marsh, V.; Bernat, B. A.; Ballard, J.; Colwell, H.;
Evans, R. J.; Parry, J.; Smith, D.; Brandhuber, B. J.; Gross, S.; Marlow,
A.; Hurley, B.; Lyssikatos, J.; Lee, P. A.; Winkler, J. D.; Koch, K.;
Wallace, E. Biological characterization of ARRY-142886 (AZD6244), a
potent, highly selective mitogen-activated protein kinase kinase 1/2
inhibitor. Clin. Cancer Res. 2007, 13, 1576−1583.
(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 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.
(9) 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.
(10) Barrett, S. D.; Bridges, A. J.; Dudley, D. T.; Saltiel, A. R.; Fergus,
J. H.; Flamme, C. M.; Delaney, A. M.; Kaufman, M.; LePage, S.;
Leopold, W. R.; Przybranowski, S. A.; Sebolt-Leopold, J.; Van
Becelaere, K.; Doherty, A. M.; Kennedy, R. M.; Marston, D.;
Howard, W. A., Jr.; Smith, Y.; Warmus, J. S.; Tecle, H. The discovery
of the benzhydroxamate MEK inhibitors CI-1040 and PD0325901.
Bioorg. Med. Chem. Lett. 2008, 18, 6501−6504.
(11) ClinicalTrials.gov; National Institutes of Health: Bethesda, MD;
http://www.clinicaltrials.gov/ct2/results?term=AZD6244 (Accessed
December 10, 2011).
(12) Abe, H.; Kikuchi, S.; Hayakawa, K.; Iida, T.; Nagahashi, N.;
Maeda, K.; Sakamoto, J.; Matsumoto, N.; Miura, T.; Matsumura, K.;
4603
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604

Journal of Medicinal Chemistry
Article
(27) Pierce, A. C.; Sandretto, K. L.; Bemis, G. W. Kinase inhibitors
and the case for CH...O hydrogen bonds in protein−ligand binding.
Proteins 2002, 49, 567−576.
(28) Toth, G.; Bowers, S. G..; Truong, A. P.; Probst, G. The role and
significance of unconventional hydrogen bonds in small molecule
recognition by biological receptors of pharmaceutical relevance. Curr.
Pharm. Des. 2007, 13, 3476−3493.
(29) Fink, D. M.; Strupczewski, J. T. Process and intermediates for
the preparation of 6-fluorobenzoisothiazoles. Can. Pat. Appl. CA
2076556, 1993.
(30) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: A new
approach for rapid, accurate docking and scoring. 1. Method and
assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739−1749.
(31) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data
Collected in Oscillation Mode. Methods Enzymol. 1997, 276, 307−326.
(32) The CCP4 suite: programs for protein crystallography
Collaborative Computational Project, Number 4. Acta Crystallogr.,
Sect. D: Biol. Crystallogr. 1994, D50, 760−763; Vagin, A.; Teplyakov, A.
MOLREP: an Automated Program for Molecular Replacement. J.
Appl. Crystallogr. 1997, 30, 1022−1025.
4604
dx.doi.org/10.1021/jm2017094 | J. Med. Chem. 2012, 55, 4594−4604