Modular Assembly of Allosteric MEK Inhibitor Structural Elements Unravels Potency and Feedback-Modulation Handles

Article abstract of DOI:10.1002/cmdc.201500442

Having recently identified a so-far unexplored area adjacent to the known binding site of allosteric mitogen-activated protein kinase kinase (MEK) inhibitors, we now report an extension of these studies by combining our new side chains with different MEK

Full text of DOI:10.1002/cmdc.201500442

DOI: 10.1002/cmdc.201500442
Full Papers
Modular Assembly of Allosteric MEK Inhibitor Structural
Elements Unravels Potency and Feedback-Modulation
Handles
Ingo V. Hartung,*[a] Florian Pühler,[b] Roland Neuhaus,[c] Arne Scholz,[b] Gerhard Siemeister,[b]
Jens Geisler,[a] Roman C. Hillig,[d] Oliver von Ahsen,[e] and Marion Hitchcock*[a]
Having recently identified a so-far unexplored area adjacent to
the known binding site of allosteric mitogen-activated protein
kinase kinase (MEK) inhibitors, we now report an extension of
these studies by combining our new side chains with different
MEK inhibitor cores in a modular manner. Replacement of the
amide headgroup with inverse sulfonamides resulted in the
identification of new MEK inhibitors with at least 10-fold
higher cellular potency against K-Ras-mutated tumor cells. A
selected inhibitor from this new series retained the favorable
pharmacokinetic profile of its predecessor in rodent and non-
rodent species and displayed significant in vivo efficacy at
once-daily oral doses of 0.25–1 mgkgÀ1 in a K-Ras-mutated
xenograft model. The brain penetration potential of this ana-
logue was significantly attenuated relative to PD325901. In
a second series, the central fluorophenyl core was replaced by
a pyridine moiety which gave rise to a similar boost in cellular
potency. Most notably, analogues from this second series do
not show MEK feedback phosphorylation in K-Ras-mutated
A549 cells. Our results complement recent reports on the
structural intricacies of MEK–Raf feedback interactions.
Introduction
Based on pioneering work by researchers at Pfizer,[1] allosteric
inhibitors of the mitogen-activated protein (MAP) kinase kinas-
es MEK1/2 (MAPKK, MAP2K) have been pursued by many phar-
maceutical companies to target a key node of the canonical
Ras–Raf–MEK–ERK pathway.[2] These substantial efforts resulted
in the approval of trametinib (Mekinist, GSK/Japan Tobacco) in
2013 for the treatment of B-Raf (V600E)-driven metastatic mel-
anoma.[3] The continuously expanding collection of publica-
tions detailing MEK inhibitor optimization programs provides
the scientific community with a comprehensive manual of
strategies and current practice in medicinal chemistry.[4]
by previous generations of MEK inhibitors.[5] Using PD325901
as a starting point, truncation of its hydroxamic ester head-
group was combined with the incorporation of alkyl or aryl
ethers at the neighboring C6 ring position (Figure 1). The pre-
dicted binding mode of our new series with extension of the
C6 side chains into a previously unexplored subpocket was
confirmed by X-ray studies.[5]
We recently described the structural evolution of a novel
and highly potent series of allosteric MEK inhibitors that take
advantage of a binding subpocket which was not addressed
[a] Dr. I. V. Hartung, Dr. J. Geisler, Dr. M. Hitchcock+
Medicinal Chemistry Berlin
Bayer HealthCare AG, 13353 Berlin (Germany)
E-mail: ingo.hartung@bayer.com
marion.hitchcock@bayer.com
[b] Dr. F. Pühler,+ Dr. A. Scholz, Dr. G. Siemeister
Global Therapeutic Research Group Oncology
Bayer HealthCare AG, 13353 Berlin (Germany)
Figure 1. Structural evolution of 6-(aryloxy)benzamide MEK inhibitors.[5]
[c] Dr. R. Neuhaus
After optimization of the capping group, sulfamide 1 was
identified as an MEK inhibitor with nanomolar cell potency
against B-Raf (V600E) as well as Ras-mutated cell lines, high
metabolic stability and resultant long half-lives in rodent and
non-rodent species. Sulfamide 1 was efficacious in B-Raf- and
K-Ras-driven xenograft models and, despite being orally bio-
available, displayed a much lower brain/plasma exposure ratio
than PD325901. The latter finding was highly relevant because
Research Pharmacokinetics
Bayer HealthCare AG, 13353 Berlin (Germany)
[d] Dr. R. C. Hillig
Structural Biology
Bayer HealthCare AG, 13353 Berlin (Germany)
[e] Dr. O. von Ahsen
Global Biomarker Research
Bayer HealthCare AG, 13353 Berlin (Germany)
[+] Former affiliation and address during this project.
ChemMedChem 2015, 10, 2004 – 2013
2004
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
central nervous system (CNS)-mediated adverse effects have
been reported for PD325901.[6]
The C4-fluoro atom of PD-like MEK inhibitors forms an un-
usual hydrogen bond with the amide backbone of Ser212 in
the allosteric binding site.[1] A switch to the pyridine core of
AS703026 was expected to strengthen this key interaction.
We and others have previously explored the structure–activi-
ty relationships (SAR) for the right-side aniline (Figure 1) with-
out identifying beneficial structural changes. Therefore, we de-
cided to keep this part of our compounds unchanged (as a 2-
fluoro-4-iodoaniline side chain).
Having identified a new subpocket adjacent to the known
allosteric binding site, we decided to examine the combination
of our new C6 side chains with different MEK inhibitor cores in
a modular manner, and herein we describe the results of these
studies (Figure 2). We selected the core of RDEA119 (BAY 86-
9766, Ardea Biosciences/Bayer HealthCare)[7] and of AS703026
(Merck-Serono)[8] for this undertaking.
The synthesis and SAR of two new series of highly potent al-
losteric MEK inhibitors, along with pharmacokinetic (PK) and in
vivo efficacy data for one advanced preclinical candidate, are
described herein. In addition, the newly designed MEK inhibi-
tors are compared with their predecessors for their ability to
induce feedback activation of Raf.
Results and Discussion
Synthesis and profiling of compounds containing a core
with an inverse sulfonamide headgroup
All newly synthesized compounds were profiled in an enzymat-
ic COT–MEK cascade assay and in cell proliferation assays with
A375 cells harboring a B-Raf (V600E) mutation as well as
HCT116 and A549 cells harboring K-Ras G13D mutations. All
presented data are average values of at least two independent
measurements. In line with previous publications, A375 cells
were found to be significantly more sensitive to MEK inhibition
than HCT116 or A549 cells.[9] Our goal was to identify MEK in-
hibitors with nanomolar potency in cellular assays with B-Raf-
mutated and Ras-mutated tumor cell lines.
Figure 2. Accessing novel MEK inhibitors by reassembly of MEK inhibitor
The synthesis of MEK inhibitors containing our newly de-
signed C6-phenoxy side chain in combination with an inverse
sulfonamide or sulfamide headgroup is outlined in Scheme 1.
Starting from 2,4,6-trifluoronitrobenzene, both ortho-fluoro
substituents were displaced in a stepwise fashion under basic
conditions. Introduction of the aniline was best accomplished
with lithium hexamethyldisilazide as base, which allowed for
high yield and nearly perfect regioselectivity. The C6-phenoxy
side chain was installed in the presence of cesium carbonate
at slightly increased reaction temperature, which yielded a 7:1
structural elements.
By dismantling the hydroxamic ester headgroup of
PD325901 (Figure 1), we had consciously sacrificed one known
interaction with the MEK enzyme in order to increase metabol-
ic stability.[5] We aimed to reintroduce this interaction option
without compromising metabolic stability by using the inverse
sulfonamide headgroup of RDEA119/BAY 86-9766 (refameti-
nib).
Scheme 1. General synthesis of MEK inhibitors with a sulfonamide or sulfamide headgroup. Reagents and conditions: a) LiHMDS, THF, 08C!RT, 87%;
b) Cs2CO3, DMF, 508C, 50%; c) ClSO2Et, pyridine, RT; d) Na2S2O4 (excess), THF, H2O, 508C, 60% (over 2 steps); e) ClSO2R, pyridine. LiHMDS=lithium hexamethyl-
disilazide.
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2005
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
mixture of regioisomers in favor of the desired C6-phenoxy
product 3. Gratifyingly, protection of the amino group could
be avoided, as the phenol functioned selectively as nucleophile
under our reaction conditions. Capping of the side chain with
ethylsulfonyl chloride was followed by chemoselective nitro re-
duction with sodium dithionite. Finally, the use of various com-
mercial sulfonyl chlorides or related electrophiles allowed us to
efficiently screen various sulfonamide and sulfamide head-
groups. The overall sequence of five steps was easily scaled to
multi-gram product quantities (see Experimental Section below
for details). For a matched-pair SAR exploration of compounds
with a sulfonamide or sulfamide headgroup, we selected com-
pound 6, which contains a C6-phenoxy side chain with an
ethyl sulfonamide capping group, as the benchmark (Table 1).
This compound showed a biochemical IC50 value of 15 nm (al-
ready close to the lower detection limit of this assay), low-
nanomolar potency in the A375 (B-Raf V600E) proliferation
assay, and intermediate nanomolar IC50 values in proliferation
assays with K-Ras-mutated cell lines. Thereby, compound 6
was roughly as potent as PD325901 and compound 1 in these
assays.
Table 1. SAR for compounds containing a core with a sulfonamide or sul-
famide headgroup.
Compd
Headgroup
In vitro IC50 [nm]
MEK1
A375
(B-Raf)
HCT116
(K-Ras)
A549
(K-Ras)
6
7
15
8
<30
287
28
278
39
0.9
8
9
11
9
2
1
32
54
74
93
10
BAY-265
For comparison, compound 1 showed a biochemical IC50
value of 14 nm and inhibited proliferation of A375 cells with an
IC50 of 4 nm, of HCT116 cells with an IC50 of 180 nm and of
A549 cells with an IC50 of 132 nm.
11
0.7
28
26
11
12
9
4
3
206
100
126
69
We were pleased to find that even with the simplest possi-
ble sulfonamide headgroup (methyl sulfonamide 7), a 10-fold
increase in the A375 as well as in the less sensitive HCT116
and A549 proliferation assays was achieved. Matched-pair anal-
ysis with carbamide 6 proves that the inverse sulfonamide is,
per se, a much more potent headgroup than the carbamide
headgroup of PD-like compounds. Incrementally increasing the
size of the appended alkyl group did not lead to further gains
in potency (compounds 8 and 9). This finding demonstrates
that lipophilicity-driven gains in potency cannot be achieved
around the headgroup. Cyclopropyl sulfonamide 10 also
showed the 10-fold potency gain over previous MEK inhibitors
and was superior to compounds with a larger cycloalkyl group
(e.g., cyclobutyl sulfonamide 11). Introduction of a sulfamide
headgroup, as in compound 12, provided no apparent benefit
over compounds with a sulfonamide headgroup.
14
13
14
12
5
63
52
7.5
17
770
736
15
17
0.7
3.3
3.3
Next, we investigated hydroxy-containing alkyl sulfonamides
to probe direct interactions with ATP, as described for Pfizer’s
hydroxamates.[1] Hydroxyalkyl sulfonamides (e.g., 13 and 15)
were well tolerated, with high biochemical and cellular poten-
cy. Interestingly, alkylation of the terminal hydroxy group (as in
methoxyalkyl sulfonamide 14) resulted in diminished potency.
Finally, we employed the fully functionalized RDEA119 head-
group. The resulting compound 16 is one of the most potent
allosteric MEK inhibitors that we have encountered throughout
our optimization program. This finding underscored the excel-
lent design of this specific headgroup. However, due to the
presence of the functionalized C6-phenoxy side chain, the re-
sulting high overall polarity of 16 (total polar surface area of
154.1 Š2) prevented us from further advancing this compound,
as we expected insufficient permeation characteristics for such
a polar compound. In summary, the use of inverse alkyl sulfo-
16[a]
1.5/1.6
0.9/0.6
3/3
3/3
[a] Data for both stereoisomers are listed.
namide headgroups provided us with highly potent MEK inhib-
itors. Cyclopropyl sulfonamide 10 was identified as the most
promising example from this compound series.
Subsequently, we briefly screened variations of the capping
group of the C6-phenoxy side chain to assess the influence of
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2006
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. SAR for compounds containing the sulfonamide core and differ-
Table 3. PK benchmarking of cyclopropyl sulfonamide 10 (BAY-265).
ent capping groups.
Compd
Rat PK[a]
Mouse PK
Brain/plasma
ratio[b]
CLblood
[LkgÀ1 hÀ1
Vss
[LkgÀ1
t1/2
[h]
F
[%]
]
]
PD325901
1
10
0.5
0.03
0.28
1.7
2.0
2.5
5.4
32
9.9
104
62
72
0.11
<0.02
<0.02
[a] Dosage for PD325901: 0.5 mgkgÀ1 i.v./1 mgkgÀ1 p.o.; dosage for 1:
1 mgkgÀ1 i.v./1 mgkgÀ1 p.o.; dosage for 10: 0.5 mgkgÀ1 i.v./1 mgkgÀ1 p.o.
[b] AUCbrain/AUCplasma for 0–3 h after 5 mgkgÀ1 i.v. dosing.
Compd
Capping group
In vitro IC50 [nm]
MEK1
A375
(B-Raf)
HCT116
(K-Ras)
A549
(K-Ras)
10
17
18
11
10
17
0.7
28
55
26
43
data). Clearance in rats was low, resulting in a long half-life of
9.9 h, whereas oral bioavailability was high. Compound 10
showed a similarly favorable PK profile in dogs with low clear-
ance (0.03 LkgÀ1 hÀ1), a long half-life (22 h), and high oral bio-
availability (100%). Most notably, despite high oral bioavailabil-
ity in all investigated species, brain tissue exposure upon i.v.
dosing to mice was below the lower limit of detection, signify-
ing low potential for CNS penetration.
1
0.7
121
107
19
20
10
19
0.3
9
7
6.7
BAY-265 (10) was highly efficacious in various xenograft
models of B-Raf- and K-Ras-driven cancer cell lines. For exam-
ple, an oral dose as low as 0.25 mgkgÀ1 once daily was already
efficacious in an A549 mice xenograft study (Figure 3). Maximal
efficacy (T/C 0.14) was achieved with daily oral doses of
1 mgkgÀ1.
709
790
21
22
13
17
29
6
2080
919
1800
503
our new core on the previously established side-chain SAR
(Table 2).[5] All new analogues were synthesized by an adaption
of the general synthetic route outlined in Scheme 1. The use
of a Boc-protected aminophenol in the second step allowed in-
stallation of the central cyclopropyl sulfonamide by a reduc-
tion–sulfonylation sequence, which was followed by Boc de-
protection and final capping with various sulfonylating, sulfa-
midating, or amidating agents.
The overall SAR observations paralleled those obtained with
our previous carbamide core, as exemplified by compound 1.[5]
Ethyl sulfonamide 10 was found to be slightly superior in po-
tency in the less sensitive cell assays than smaller (17) or larger
(18) analogues. Sulfamide 19 reached single-digit nanomolar
antiproliferative activity against K-Ras-mutated cell lines, al-
though at the expense of increased overall polarity (total polar
Figure 3. A549 xenograft study with cyclopropyl sulfonamide BAY-265 (10).
All treatments were applied once daily by oral gavage. At the end of the ex-
periment the treatment to vehicle control ratios (T/C) were determined:
BAY-265 (10) at 0.25/0.5/1 mgkgÀ1: 0.25/0.20/0.14, compound 1 at
1 mgkgÀ1: 0.15.
2
2
surface area of 139.6 Š vs. 113.6 Š for ethyl sulfonamide 10).
This increase in polarity was undesirable, as it compromises
the chance of achieving high bioavailability. In addition, an ex-
posed sulfamide moiety may lead to off-target activity. Amides
(e.g., propanamide 20 and 2-methylpropanamide 21) turned
out to be significantly less active than the corresponding sulfo-
namides. Finally, by comparing the potency levels of the un-
capped aniline 22 with the most preferred capped sulfona-
mides (e.g., 10), we confirmed that the capping group un-
doubtedly contributes to the overall inhibitory potency.
Synthesis and profiling of compounds containing a pyridine
core
Combining our C6-phenoxy side chains with the pyridyl carba-
mide core derived from AS703026 was investigated with a fo-
cused set of analogues. The route depicted in Scheme 2 was
developed for their synthesis, using 3,5-difluoro-4-pyridinecar-
boxylic acid as starting material. Base-promoted introduction
of the right-side aniline was followed by amidation and nucle-
ophilic displacement of the remaining core fluoro substituent
Cyclopropyl sulfonamide 10 (BAY-265) was characterized in
rodent and non-rodent PK studies (see Table 3 for selected
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2007
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 2. Synthesis of MEK inhibitor 26 with a pyridine core. Reagents and conditions: a) LiHMDS, THF, 08C!RT, 61%; b) 1) CDI (2.3 equiv), DMF, 608C,
90 min; 2) NH3 (25% in H2O), 08C!RT, 60%; c) 1) Cs2CO3, DMF, 508C, 18%; 2) TFA, CH2Cl2, RT, 90%; d) ClSO2Et, pyridine, RT, 84%. CDI=1,1’-carbonyldiimida-
zole, LiHMDS=lithium hexamethyldisilazide, TFA=trifluoroacetic acid.
to install the Boc-protected phenoxy side chain. After depro-
tection, sulfonylation provided the target compound after only
five synthetic steps. This sequence proved to be easily scalable
and rapidly amenable to capping-group diversification.
profiled in our in-house assays. AS703026 displayed a potency
profile similar to those of our initial lead compounds 1 and 6,
with low-nanomolar potency in biochemical and A375 cell pro-
liferation assays, whereas antiproliferative potency against K-
Ras-mutated tumor cells was only moderate (~500 nm). Excit-
ingly, combining the pyridyl carbamide core with our C6-phen-
oxy side chains yielded significantly more potent in-
hibitors, with at least 10-fold lower IC50 values in cell
assays with K-Ras-mutated tumor cells. The exact
nature of the alkyl sulfonamide group had a negligi-
ble influence, with methyl, ethyl, isopropyl, and cy-
clopropyl sulfonamides possessing similar potencies
(compounds 26–29). In contrast, completely omit-
ting the capping group had a detrimental effect on
A selection of newly synthesized analogues with key SAR
data is compiled in Table 4. For comparison, AS703026 was
Table 4. SAR for compounds with a pyridyl carbamide core.
Compd
Side chain
In vitro IC50 [nm]
potency in the less sensitive cell assays (compound
25). As for the first series discussed above, amides
were less potent as capping groups than their re-
spective sulfonamide analogues (compare amides
30 and 31 with sulfonamides 27 and 28). Neverthe-
less, with the much more potent pyridine core, com-
pounds with a greater diversity of side chains still
achieved high potency, as exemplified by benzoxa-
zolone 32.
Ethyl sulfonamide 26 (BAY-672) was advanced to
mouse xenograft studies and proved to be highly ef-
ficacious in vivo at doses as low as 0.5 mgkgÀ1 oral
daily dosing (data not shown). One additional find-
ing from in vitro pharmacological studies is notewor-
thy. It has been reported that activated ERK phos-
phorylates and inhibits CRAF kinase and that the in-
hibition of ERK signaling by first-generation allosteric
MEK inhibitors, such as PD325901, relieves ERK-de-
pendent feedback inhibition of CRAF and induces
MEK phosphorylation in most cells.[10] For example,
treatment of A549 cells with PD325901 or AS703026
induced a significant increase in MEK phosphoryla-
tion (Figure 4).
MEK1
A375
(B-Raf)
HCT116
(K-Ras)
A549
(K-Ras)
AS703026 (see Figure 2)
compound 6 (see Table 1)
7
15
8
<30
432
287
513
278
26
BAY-672
10
8
1
11
18
11
15
23
14
27
28
0.8
0.4
8
29
25
30
6
9
8
1
10
214
195
17
377
335
0.7
5
31
32
8
7
22
2
1120
45
1670
59
In contrast, treatment of A549 cells with our initial
lead compound 1 or the cyclopropyl sulfonamide 10
(BAY-265) induced no, or only a marginal, change in
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2008
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
shown that allosteric MEK inhibitors with a C6-phenoxy side
chain do not induce MEK feedback phosphorylation to a similar
extent as PD325901, with compounds from the pyridine core
series even leading to a decrease in MEK phosphorylation.
Experimental Section
Synthetic procedures
General methods and materials: All air- and moisture-sensitive re-
actions were carried out in oven-dried (1208C) glassware under an
inert atmosphere of argon. All reactive liquid reagents were trans-
ferred by syringe or cannula and were added into the flask
1
through a rubber septum. H and 13C NMR spectra were recorded
Figure 4. Effect of different MEK inhibitors on MEK phosphorylation in A549
cells. Quantified in a MesoScale assay setup with increasing concentrations
of the respective inhibitor. A549 cells were treated for 24 h with the respec-
tive MEK inhibitor or DMSO. The ratio of MEK phosphorylation (Ser217/
Ser221) in the presence of 10 mm MEK inhibitor and in the DMSO control is
shown. Data are shown for one representative experiment. This effect was
observed consistently in a series of related experiments.
at room temperature on Bruker Avance spectrometers operating at
300 or 400 MHz for H NMR and at 75 or 100 MHz for 13C NMR ac-
1
quisitions. NMR signal multiplicities are reported as they appear,
without considering higher-order effects. HRMS data were record-
ed on a Waters Q-TOF Premier mass spectrometer using electro-
spray ionization. Solvents for extraction and chromatography were
reagent grade and used as received. Commercial reagents were
used without purification. The purity of all target compounds was
MEK phosphorylation. Notably, compounds 28 and 29 from
the pyridine core series even decreased the degree of MEK
phosphorylation relative to the dimethyl sulfoxide control.
Recent studies with a new generation of MEK inhibitors from
Genentech and Chugai have revealed that a strong hydrogen
bond to Ser212 in the allosteric MEK binding site may be re-
sponsible for blocking MEK feedback phosphorylation by
Raf.[10] Assessing our SAR for MEK feedback phosphorylation
(Figure 4) leads us to propose that the C6-phenoxy side chain
is another important structural handle for blocking MEK feed-
back phosphorylation, as compound 1 induced significantly
less MEK phosphorylation than PD325901. Introduction of
a stronger hydrogen bond acceptor at the presumed Ser212
interaction site, as in compounds 28 and 29, further potentiat-
ed this effect. In contrast, introduction of the pyridine core
alone (as in AS703026) was not sufficient for blocking MEK
feedback phosphorylation. Suppression of MEK feedback phos-
phorylation has been postulated as a prerequisite for achieving
high efficacy with MEK inhibitors in Ras-mutated tumor enti-
ties.[10] Thus, unraveling the structural intricacies that govern
MEK feedback phosphorylation as outlined for our two new
series of highly potent allosteric MEK inhibitors will be an im-
portant area for further research.
1
>95% as determined by H NMR spectroscopy.
3,5-Difluoro-N-(2-fluoro-4-iodophenyl)-2-nitroaniline (2): 2,4,6-Tri-
fluoronitrobenzene (99 g, 560 mmol) and 2-fluoro-4-iodoaniline
(132 g, 560 mmol) were dissolved in tetrahydrofuran (THF,
1000 mL) and cooled to À108C. LiHMDS (1.0m in THF, 1680 mL,
1680 mmol) was added via a dropping funnel within 60 min while
the internal reaction temperature was kept below 08C. After com-
plete addition, the resulting mixture was allowed to warm to room
temperature, and stirring was continued for 16 h. The reaction mix-
ture was diluted with EtOAc and quenched with 1n HCl. After
phase separation and subsequent washing of the aqueous phase
twice with EtOAc, the combined organic layers were dried and
concentrated. Silica gel chromatography (EtOAc/hexane 2:8) pro-
vided the desired product (192 g, 487 mmol, 87%): 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=8.86 (s, 1H), 7.73 (dd, 1H), 7.55
(dd, 1H), 7.13 (t, 1H), 6.96 (ddd, 1H), 6.42 ppm (dq, 1H).
3-(3-Aminophenoxy)-5-fluoro-N-(2-fluoro-4-iodophenyl)-2-nitroa-
niline (3): 3,5-Difluoro-N-(2-fluoro-4-iodophenyl)-2-nitroaniline (2;
192 g, 486 mmol), Cs2CO3 (238 g, 729 mmol), and 3-aminophenol
(58.3 g, 535 mmol) were suspended in DMF (3500 mL) and stirred
at 508C for 48 h. The reaction mixture was concentrated to an oily
residue, which was resuspended with EtOAc and quenched with
H2O. After separation of layers, the aqueous layer was extracted
twice with EtOAc. The combined organic layers were washed with
a saturated aqueous NaCl solution, dried and concentrated. The
residue was purified by silica gel chromatography (100% hexane,
then hexane/EtOAc 9:1!8:1) to yield the desired product (117 g,
Conclusions
1
242 mmol, 50%): H NMR (300 MHz, [D6]DMSO, 300 K): d=8.45 (s,
We have described two new series of allosteric MEK inhibitors,
which combine our recently identified C6-phenoxy side chains
with cores derived from two MEK inhibitors that are currently
in clinical development. Replacing a carbamide with inverse
sulfonamide headgroups resulted in increased cellular potency
in relevant K-Ras-mutated cell lines by at least 10-fold while
maintaining the favorable PK profile in rodents and non-ro-
dents. A once-daily oral dose of as little as 0.25 mgkgÀ1 was
sufficient to achieve significant efficacy in K-Ras-mutated xeno-
graft models. Use of a pyridyl carbamide core resulted in a simi-
lar improvement in target potency. Furthermore, we have
1H), 7.69 (d, 1H), 7.51 (d, 1H), 7.07 (t, 1H), 7.01 (t, 1H), 6.38 (dd,
1H), 6.29 (d, 1H), 6.14–6.24 (m, 3H), 5.34 ppm (s, 2H).
N-(3-{5-Fluoro-3-[(2-fluoro-4-iodophenyl)amino]-2-nitrophenoxy}-
phenyl)ethanesulfonamide (4): 3-(3-Aminophenoxy)-5-fluoro-N-(2-
fluoro-4-iodophenyl)-2-nitroaniline (3; 117 g, 242 mmol) was dis-
solved in pyridine (980 mL) and treated with ethylsulfonyl chloride
(34.34 mL, 363 mmol) with stirring at room temperature for 48 h.
The reaction mixture was diluted with EtOAc and quenched with
H2O. After separation of layers, the aqueous layer was extracted
twice with EtOAc. The combined organic layers were washed with
a saturated aqueous NaCl solution, dried and concentrated. The
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2009
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
residue was twice taken up in toluene (300 mL) and distilled to
dryness, and used in the following step without further purification
(169.5 g).
N-(3-{2-[(Ethylsulfonyl)amino]-5-fluoro-3-[(2-fluoro-4-iodo-
phenyl)amino]phenoxy}phenyl)ethanesulfonamide (8): 1H NMR
(300 MHz, [D6]DMSO, 300 K): d=9.93 (brs, 1H), 9.08 (brs, 1H), 7.65
(dd, 1H), 7.58 (s, 1H), 7.45 (dd, 1H), 7.34 (t, 1H), 7.10 (t, 1H), 7.02
(dd, 1H), 6.95 (t, 1H), 6.80 (dd, 1H), 6.58 (dd, 1H), 6.17 (dd, 1H),
3.10 (q, 4H), 1.20 (t, 3H), 1.15 ppm (t, 3H).
N-(3-{2-Amino-5-fluoro-3-[(2-fluoro-4-iodophenyl)amino]phen-
oxy}phenyl)ethanesulfonamide (5): N-(3-{5-Fluoro-3-[(2-fluoro-4-
iodophenyl)amino]-2-nitrophenoxy}phenyl)ethanesulfonamide (4,
69.5 g) was dissolved in THF (2600 mL) and warmed to 508C. At
N-(2-{3-[(Ethylsulfonyl)amino]phenoxy}-4-fluoro-6-[(2-fluoro-4-io-
dophenyl)amino]phenyl)propane-2-sulfonamide (9): 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=9.94 (s, 1H), 9.02 (s, 1H), 7.65 (dd,
1H), 7.56 (s, 1H), 7.44 (dd, 1H), 7.34 (t, 1H), 7.08 (t, 1H), 7.02 (dd,
1H), 6.95 (t, 1H), 6.80 (dd, 1H), 6.63 (dd, 1H), 6.20 (dd, 1H), 3.26
(m, 1H), 3.10 (q, 2H), 1.26 (d, 6H), 1.16 ppm (t, 3H).
this temperature
a
solution of sodium dithionite (215 g,
1235 mmol) in H2O (1200 mL) was added within 30 min and stirring
was continued for an additional 150 min, upon which all starting
material was consumed (TLC). The organic layer was separated and
concentrated to an oily residue which was redissolved in EtOAc
and quenched with H2O. After separation of the layers, the aque-
ous layer was extracted twice with EtOAc. The combined organic
layers were washed with a saturated aqueous NaCl solution, dried
and concentrated. The residue was combined with the crude prod-
uct of a second batch and purified by silica gel chromatography
(hexane/EtOAc 2:1) to yield the desired product (79.1 g, 144 mmol,
N-(2-{3-[(Ethylsulfonyl)amino]phenoxy}-4-fluoro-6-[(2-fluoro-4-io-
dophenyl)amino]phenyl)cyclobutanesulfonamide (11): 1H NMR
(300 MHz, [D6]DMSO, 300 K): d=9.94 (brs, 1H), 9.02 (brs, 1H), 7.65
(dd, 1H), 7.58 (s, 1H), 7.45 (dd, 1H), 7.35 (t, 1H), 7.10 (t, 1H), 7.03
(dd, 1H), 6.97 (t, 1H), 6.83 (dd, 1H), 6.59 (dd, 1H), 6.13 (dd, 1H),
3.93 (quint, 1H), 3.10 (q, 2H), 2.08–2.30 (m, 4H), 1.75–1.86 (m, 2H),
1.15 ppm (t, 3H).
1
60% over two steps): H NMR (300 MHz, [D6]DMSO, 300 K): d=9.86
(s, 1H), 7.52 (dd, 1H), 7.42 (s, 1H), 7.31 (d, 1H), 7.25 (t, 1H), 6.90
(dd, 1H), 6.84 (t, 1H), 6.54–6.65 (m, 3H), 6.48 (dd, 1H), 4.47 (s, 2H),
3.08 (q, 2H), 1.14 ppm (t, 3H).
N-(3-{5-Fluoro-3-[(2-fluoro-4-iodophenyl)amino]-2-(sulfamoylami-
no)phenoxy}phenyl)ethanesulfonamide (12): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=9.93 (brs, 1H), 8.59 (brs, 1H), 7.63–7.67 (m,
2H), 7.45 (dd, 1H), 7.32 (t, 1H), 7.14 (t, 1H), 6.97–7.02 (m, 4H), 6.81
(dd, 1H), 6.48 (dd, 1H), 6.00 (dd, 1H), 3.10 (q, 2H), 1.16 ppm (t,
3H).
N-(2-{3-[(Ethylsulfonyl)amino]phenoxy}-4-fluoro-6-[(2-fluoro-4-io-
dophenyl)amino]phenyl)cyclopropanesulfonamide (10): N-(3-{2-
Amino-5-fluoro-3-[(2-fluoro-4-iodophenyl)amino]phenoxy}phenyl)e-
thanesulfonamide (5; 40 g, 73 mmol) was dissolved in pyridine
(600 mL) and cooled to 08C. At this temperature cyclopropylsulfon-
yl chloride (9.7 mL, 95 mmol) was added dropwise, and stirring was
continued overnight at room temperature. Toluene (33 mL) was
added to the reaction mixture and removed by distillation under
reduced pressure. After this procedure was repeated a second
time, the crude product was redissolved in EtOAc and quenched
with H2O. After separation of layers, the aqueous layer was extract-
ed twice with EtOAc. The combined organic layers were washed
with a saturated aqueous NaCl solution, dried and concentrated.
The residue was purified by silica gel chromatography (hexane/
EtOAc 2:1) and recrystallized from methyl tert-butyl ether to yield
the desired product (29.5 g, 45 mmol, 62%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=9.93 (brs, 1H), 9.09 (brs, 1H), 7.64 (dd, 1H),
7.59 (s, 1H), 7.45 (dd, 1H), 7.34 (t, 1H), 7.12 (t, 1H), 7.02 (dd, 1H),
6.96 (t, 1H), 6.81 (dd, 1H), 6.57 (dd, 1H), 6.14 (dd, 1H), 3.10 (q, 2H),
2.62–2.67 (m, 1H), 1.15 (t, 3H), 0.80–0.90 ppm (m, 4H); 13C NMR
(100 MHz, [D6]DMSO, 300 K): d=5.8 (CH2), 8.5 (CH3), 30.8 (CH), 45.7
(CH2), 97.0/97.2 (CH, CÀF coupling), 98.1/98.4 (CH, CÀF coupling),
110.7 (CH), 112.4 (Cq), 115.4 (CH), 115.9 (CH), 122.8 (CH), 124.9/125.1
(CH, CÀF coupling), 129.9/130.0 (Cq, CÀF coupling), 131.3 (CH),
134.3 (CH), 140.6 (Cq), 143.8/143.9 (Cq, CÀF coupling), 152.8/155.3
(Cq, CÀF coupling), 156.3 (Cq), 156.4 (Cq), 156.5 (Cq), 160.7/
163.1 ppm (Cq, CÀF coupling); ESI-HRMS: m/z [M+H]+ calcd for
C23H23F2IN3O5S2: 650.0092, found: 650.0106.
N-(2-{3-[(Ethylsulfonyl)amino]phenoxy}-4-fluoro-6-[(2-fluoro-4-io-
dophenyl)amino]phenyl)-2-hydroxyethanesulfonamide
(13):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.93 (brs, 1H), 9.05 (brs,
1H), 7.81 (s, 1H), 7.64 (dd, 1H), 7.47 (dd, 1H), 7.33 (t, 1H), 7.11 (t,
1H), 7.01 (dd, 1H), 6.94 (t, 1H), 6.78 (dd, 1H), 6.47 (dd, 1H), 6.11
(dd, 1H), 5.08 (brs, 1H), 3.77 (t, 2H), 3.10 (q, 2H), 2.49 (partially ob-
scured by DMSO, 2H), 1.15 ppm (t, 3H).
N-(2-{3-[(Ethylsulfonyl)amino]phenoxy}-4-fluoro-6-[(2-fluoro-4-io-
dophenyl)amino]phenyl)-2-methoxyethanesulfonamide
(14):
1H NMR (300 MHz, [D6]DMSO, 300 K): d=9.93 (brs, 1H), 9.12 (brs,
1H), 7.62–7.68 (m, 2H), 7.46 (dd, 1H), 7.33 (t, 1H), 7.12 (t, 1H), 7.01
(dd, 1H), 6.95 (t, 1H), 6.78 (dd, 1H), 6.50 (dd, 1H), 6.13 (dd, 1H),
3.67 (t, 2H), 3.34 (t, 2H), 3.14 (s, 3H), 3.10 (q, 2H), 1.14 ppm (t, 3H).
N-(2-{3-[(Ethylsulfonyl)amino]phenoxy}-4-fluoro-6-[(2-fluoro-4-io-
dophenyl)amino]phenyl)-1-(2-hydroxyethyl)cyclopropanesulfo-
namide (15): 1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.93 (brs,
1H), 9.08 (brs, 1H), 7.71 (s, 1H), 7.64 (dd, 1H), 7.44 (dd, 1H), 7.34
(t, 1H), 7.10 (t, 1H), 7.03 (dd, 1H), 6.97 (t, 1H), 6.78 (dd, 1H), 6.52
(dd, 1H), 6.09 (dd, 1H), 4.58 (s, 1H), 3.52 (t, 2H), 3.10 (q, 2H), 2.02
(t, 2H), 1.15 (t, 3H), 0.98–1.02 (m, 2H), 0.83–0.88 ppm (m, 2H).
1-(2,3-Dihydroxypropyl)-N-(2-{3-[(ethylsulfonyl)amino]phenoxy}-
4-fluoro-6-[(2-fluoro-4-iodophenyl)amino]phenyl)cyclopropane-
sulfonamide (16): 1H NMR (300 MHz, [D6]DMSO, 300 K): d=9.89
(brs, 1H), 9.86 (brs, 1H), 7.80 (s, 1H), 7.64 (dd, 1H), 7.46 (dd, 1H),
7.33 (t, 1H), 7.10 (t, 1H), 7.01 (dd, 1H), 6.96 (t, 1H), 6.79 (dd, 1H),
6.42 (dd, 1H), 6.05 (dd, 1H), 4.49 (brs, 1H), 3.71 (brs, 1H), 3.16
(brs, 2H), 3.10 (q, 2H), 2.27 (d, 1H), 1.63 (dd, 1H), 1.15 (t, 3H),
0.96–1.07 ppm (m, 4H).
The following compounds 7–16 were prepared analogously to
compound 10 by using other sulfonyl chlorides or isocyanatosul-
fonyl chloride in the final sulfonylation step, followed by standard
transformations where required.
The following compounds 17–22 were prepared by adaption of
the route depicted in Scheme 1 and detailed for compound 10.
N-(3-{5-Fluoro-3-[(2-fluoro-4-iodophenyl)amino]-2-[(methylsulfo-
nyl)amino]phenoxy}phenyl)ethanesulfonamide
(7):
1H NMR
(300 MHz, [D6]DMSO, 300 K): d=9.92 (brs, 1H), 9.13 (brs, 1H), 7.64
(dd, 1H), 7.61 (s, 1H), 7.45 (d, 1H), 7.34 (t, 1H), 7.11 (t, 1H), 7.02
(dd, 1H), 6.95 (t, 1H), 6.81 (dd, 1H), 6.53 (dd, 1H), 6.14 (dd, 1H),
3.51 (s, 3H), 3.10 (q, 2H), 1.15 ppm (t, 3H).
N-(4-Fluoro-2-[(2-fluoro-4-iodophenyl)amino]-6-{3-[(methylsulfo-
nyl)amino]phenoxy}phenyl)cyclopropanesulfonamide
(17):
1H NMR (300 MHz, [D6]DMSO, 300 K): d=7.64 (dd, 1H), 7.59 (s, 1H),
7.44 (d, 1H), 7.34 (t, 1H), 7.11 (t, 1H), 7.01 (dd, 1H), 6.94 (t, 1H),
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2010
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
6.82 (dd, 1H), 6.57 (dd, 1H), 6.15 (dd, 1H), 2.99 (s, 3H), 2.61–2.68
(m, 1H), 0.77–0.90 ppm (m, 4H).
3-(3-Aminophenoxy)-5-[(2-fluoro-4-iodophenyl)amino]isonicoti-
namide (25): 3-Fluoro-5-[(2-fluoro-4-iodophenyl)amino]isonicotina-
mide (24; 5.5 g, 14.6 mmol) and 3-(tert-butoxycarbonylamino)phe-
nol (3.67 g, 17.5 mmol) were dissolved in DMF (180 mL) and treat-
ed with Cs2CO3 (14.3 g, 43.9 mmol), and the mixture was heated at
508C. Stirring was continued at this temperature until complete
consumption of starting material (TLC). The reaction mixture was
diluted with CH2Cl2 and quenched with H2O. After separation of
layers, the aqueous layer was extracted several times with CH2Cl2.
The combined organic layers were washed with a saturated aque-
ous NaCl solution, dried and concentrated. The residue was further
purified by silica gel chromatography to yield the desired product
(2.9 g). This material was dissolved in CH2Cl2 (29 mL) and treated
with TFA (5.8 mL). After the mixture was stirred at room tempera-
ture for 16 h, saturated aqueous NaHCO3 and CH2Cl2 were added.
After separation of the layers, the aqueous layer was extracted sev-
eral times with CH2Cl2. The combined organic layers were washed
with a saturated aqueous NaCl solution, dried and concentrated.
The residue was further purified by silica gel chromatography to
yield the desired product (1.5 g, 3.2 mmol, 18% over two steps):
1H NMR (400 Hz, MHz, [D6]DMSO, 300 K): d=8.24 (s, 1H), 8.10 (s,
1H), 7.91 (s, 2H), 7.71 (s, 1H), 7.62 (dd, 1H), 7.42 (d, 1H), 7.12 (t,
1H), 6.97 (t, 1H), 6.32 (d, 1H), 6.19 (t, 1H), 6.15 (d, 1H), 5.23 ppm
(s, 2H).
N-(4-Fluoro-2-[(2-fluoro-4-iodophenyl)amino]-6-{3-[(isopropylsul-
fonyl)amino]phenoxy}phenyl)cyclopropanesulfonamide
(18):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.88 (brs, 1H), 9.09 (brs,
1H), 7.64 (dd, 1H), 7.60 (s, 1H), 7.44 (d, 1H), 7.32 (t, 1H), 7.12 (t,
1H), 7.03 (dd, 1H), 6.97 (t, 1H), 6.79 (dd, 1H), 6.57 (dd, 1H), 6.12
(dd, 1H), 3.24 (m, 1H), 2.60–2.68 (m, 1H), 1.20 (d, 6H), 0.77–
0.90 ppm (m, 4H).
N-{4-Fluoro-2-[(2-fluoro-4-iodophenyl)amino]-6-[3-(sulfamoylami-
no)phenoxy]phenyl}cyclopropanesulfonamide
(19):
1H NMR
(400 MHz, [D6]DMSO, 300 K): d=9.66 (s, 1H), 9.12 (brs, 1H), 7.64
(dd, 1H), 7.58 (s, 1H), 7.44 (d, 1H), 7.29 (t, 1H), 7.15 (s, 2H), 7.11 (t,
1H), 6.99 (dd, 1H), 6.91 (t, 1H), 6.70 (dd, 1H), 6.56 (dd, 1H), 6.06
(dd, 1H), 2.63–2.70 (m, 1H), 0.82–0.89 ppm (m, 4H).
N-(3-{2-[(Cyclopropylsulfonyl)amino]-5-fluoro-3-[(2-fluoro-4-iodo-
phenyl)amino]phenoxy}phenyl)propanamide
(20):
1H NMR
(400 MHz, [D6]DMSO, 400 K): d=9.97 (s, 1H), 9.11 (brs, 1H), 7.64
(dd, 1H), 7.58 (s, 1H), 7.43–7.46 (m, 2H), 7.37 (d, 1H), 7.31 (t, 1H),
7.12 (t, 1H), 6.78 (d, 1H), 6.56 (dd, 1H), 6.08 (dd, 1H), 2.61–2.68 (m,
1H), 2.27 (q, 2H), 1.03 (t, 3H), 0.78–0.91 ppm (m, 4H).
N-(3-{2-[(Cyclopropylsulfonyl)amino]-5-fluoro-3-[(2-fluoro-4-iodo-
3-{3-[(Ethylsulfonyl)amino]phenoxy}-5-[(2-fluoro-4-iodophenyl)-
amino]isonicotinamide (26): 3-(3-Aminophenoxy)-5-[(2-fluoro-4-io-
dophenyl)amino]isonicotinamide (25; 755 mg, 1.6 mmol) was dis-
solved in pyridine (11 mL) and treated with ethylsulfonyl chloride
(313 mg, 2.44 mmol, dissolved in 0.5 mL pyridine). The reaction
mixture was stirred at room temperature for 16 h, after which
CH2Cl2 and H2O were added. After separation of layers, the aque-
ous layer was extracted twice with CH2Cl2. The combined organic
layers were washed with a saturated aqueous NaCl solution, dried
and concentrated. The residue was triturated with CH2Cl2 to yield
phenyl)amino]phenoxy}phenyl)-2-methylpropanamide
(21):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.93 (s, 1H), 9.11 (brs,
1H), 7.64 (dd, 1H), 7.58 (s, 1H), 7.37–7.47 (m, 3H), 7.30 (t, 1H), 7.11
(t, 1H), 6.78 (dd, 1H), 6.56 (m, 1H), 6.07 (dd, 1H), 2.63–2.71 (m,
1H), 2.52 (m, 1H), 1.04 (d, 6H), 0.82–0.91 ppm (m, 4H).
N-{2-(3-Aminophenoxy)-4-fluoro-6-[(2-fluoro-4-iodophenyl)ami-
no]phenyl}cyclopropanesulfonamide (22): 1H NMR (400 MHz,
[D6]DMSO, 400 K): d=9.08 (brs, 1H), 7.63 (dd, 1H), 7.53 (s, 1H),
7.44 (d, 1H), 7.11 (t, 1H), 7.02 (t, 1H), 6.53 (dd, 1H), 6.38 (dd, 1H),
6.28 (t, 1H), 6.22 (dd, 1H), 6.03 (dd, 1H), 5.28 (s, 2H), 2.60–2.67 (m,
1H), 0.80–0.91 ppm (m, 4H).
1
the desired product (745 mg, 1.34 mmol, 84%): H NMR (300 MHz,
[D6]DMSO, 300 K): d=9.91 (s, 1H), 8.09–8.17 (m, 2H), 8.03 (s, 1H),
7.91 (s, 1H), 7.78 (s, 1H), 7.62 (d, 1H), 7.43 (d, 1H), 7.28 (t, 1H), 7.11
(t, 1H), 6.96 (d, 1H), 6.91 (t, 1H), 6.69 (dd, 1H), 3.09 (q, 2H),
1.15 ppm (t, 3H); 13C NMR (100 MHz, [D6]DMSO, 300 K): d=8.5
(CH3), 45.7 (CH2), 84.6 (Cq), 109.2 (CH), 113.3 (CH), 114.7 (CH), 123.3
(CH), 124.1 (Cq), 124.8/125.0 (CH, CÀF coupling), 130.2/130.3 (Cq, CÀ
F coupling), 131.2 (CH), 133.6 (CH), 134.2 (CH), 135.5 (CH), 138.0
(Cq), 140.6 (Cq), 149.5 (Cq), 152.9/155.3 (Cq, CÀF coupling), 157.5
(Cq), 165.4 ppm (Cq); ESI-HRMS: m/z [M+H]+ calcd for
C20H19FIN4O4S: 557.0156, found: 557.0159.
3-Fluoro-5-[(2-fluoro-4-iodophenyl)amino]isonicotinic acid (23):
3,5-Difluoroisonicotinic acid (6.4 g, 40 mmol) and 2-fluoro-4-iodoa-
niline (9.5 g, 40 mmol) were dissolved in THF (325 mL) and cooled
to 08C. LiHMDS (1.0m in THF, 120 mL, 120 mmol) was added via
a dropping funnel. The resulting mixture was allowed to warm to
room temperature, and stirring was continued for 16 h. The reac-
tion mixture was concentrated and taken up in CH2Cl2 and 2n
aqueous NaOH. After phase separation and subsequent washing of
the organic phase twice with 2n aqueous NaOH, the combined
aqueous layers were acidified with concentrated HCl to pH 3; the
formed precipitate was collected by filtration and dried to yield
the desired product (9.2 g, 24.5 mmol, 61%): 1H NMR (300 MHz,
[D6]DMSO, 300 K): d=8.67 (s, 1H), 8.08–8.14 (m, 2H), 7.66 (d, 1H),
7.47 (d, 1H), 7.15 ppm (t, 1H).
The following compounds 27–32 were prepared analogously to
compound 26 by using other capping groups in the final step or
by replacing the phenol in the substitution reaction.
3-[(2-Fluoro-4-iodophenyl)amino]-5-{3-[(methylsulfonyl)amino]-
phenoxy}isonicotinamide (27): 1H NMR (300 MHz, [D6]DMSO,
300 K): d=9.87 (s, 1H), 8.12–8.15 (m, 2H), 8.03 (s, 1H), 7.92 (s, 1H),
7.78 (s, 1H), 7.62 (d, 1H), 7.42 (d, 1H), 7.30 (t, 1H), 7.11 (t, 1H), 6.95
(d, 1H), 6.91 (t, 1H), 6.71 (dd, 1H), 2.99 ppm (s, 3H).
3-Fluoro-5-[(2-fluoro-4-iodophenyl)amino]isonicotinamide (24):
3-Fluoro-5-[(2-fluoro-4-iodophenyl)amino]isonicotinic acid (23;
9.2 g, 24.5 mmol) and 1,1’-carbonyldiimidazole (9.2 g, 56.6 mmol)
were dissolved in DMF (90 mL), warmed to 608C and stirred for
90 min. The mixture was cooled to 08C, NH3 (25% in H2O, 290 mL)
was added and stirring was continued for 16 h at room tempera-
ture. The resulting slurry was filtered, and the collected precipitate
was washed with H2O and dried at 508C to yield the desired prod-
uct (5.5 g, 14.6 mmol, 60%): 1H NMR (400 MHz, [D6]DMSO, 300 K):
d=8.34 (s, 1H), 8.13 (s, 3H), 8.08 (s, 1H), 7.63 (dd, 1H), 7.44 (d,
1H), 7.13 ppm (t, 1H).
3-[(2-Fluoro-4-iodophenyl)amino]-5-{3-[(isopropylsulfonyl)ami-
no]phenoxy}isonicotinamide (28): 1H NMR (300 MHz, [D6]DMSO,
300 K): d=9.85 (s, 1H), 8.14 (s, 2H), 8.00 (s, 1H), 7.90 (s, 1H), 7.76
(s, 1H), 7.62 (d, 1H), 7.43 (d, 1H), 7.28 (t, 1H), 7.11 (t, 1H), 6.97 (d,
1H), 6.93 (t, 1H), 6.68 (dd, 1H), 3.25 (septet, 1H), 1.20 ppm (d, 6H).
3-{3-[(Cyclopropylsulfonyl)amino]phenoxy}-5-[(2-fluoro-4-iodo-
1
phenyl)amino]isonicotinamide (29): H NMR (400 MHz, [D6]DMSO,
300 K): d=9.83 (s, 1H), 8.13–8.16 (m, 2H), 8.01 (s, 1H), 7.91 (s, 1H),
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2011
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
7.76 (s, 1H), 7.62 (d, 1H), 7.43 (d, 1H), 7.29 (t, 1H), 7.12 (t, 1H), 6.98
(d, 1H), 6.94 (t, 1H), 6.73 (dd, 1H), 2.59–2.66 (m, 1H), 0.88–
0.93 ppm (m, 4H).
DSMZ, Germany) human lung tumor cells at 2000 cells per well,
and HCT116 (ACC-581, DSMZ, Germany) human colorectal carcino-
ma cells at 3000 cells per well in 96-well plates in Dulbecco’s modi-
fied Eagle’s medium (DMEM)/HAMS F12 (Biochrome, Germany)
containing 10% fetal calf serum (FCS) (Biochrome, Germany) and
2 mm l-glutamine. Cells were treated in quadruplicates with serial
dilutions of test compounds for 96 h, followed by quantification of
relative cell numbers upon crystal violet staining.[11] IC50 values
were calculated by means of a four-parameter fit.
3-(3-Acetamidophenoxy)-5-[(2-fluoro-4-iodophenyl)amino]isoni-
1
cotinamide (30): H NMR (300 MHz, [D6]DMSO, 300 K): d=10.00 (s,
1H), 8.10–8.17 (m, 2H), 8.02 (s, 1H), 7.92 (s, 1H), 7.75 (s, 1H), 7.62
(d, 1H), 7.43 (d, 1H), 7.35 (brs, 1H), 7.22–7.29 (m, 2H), 7.12 (t, 1H),
6.68 (ddd, 1H), 1.99 ppm (s, 3H).
3-[(2-Fluoro-4-iodophenyl)amino]-5-[3-(isobutyrylamino)phen-
oxy]isonicotinamide (31): 1H NMR (300 MHz, [D6]DMSO, 300 K):
d=9.90 (s, 1H), 8.13–8.18 (m, 2H), 8.02 (s, 1H), 7.92 (s, 1H), 7.77 (s,
1H), 7.62 (d, 1H), 7.37–7.45 (m, 2H), 7.22–7.33 (m, 2H), 7.12 (t, 1H),
6.68 (ddd, 1H), 2.50 (septet, 1H), 1.03 ppm (d, 6H).
Analysis of MEK phosphorylation
A549 human non-small cell lung cancer cells were grown in
DMEM/HAMS F12 supplemented with 2.5 mm glutamine and 10%
FCS at 378C with 5% CO2 in a humidified incubator. For mechanis-
tic studies, cultured cells were treated with the respective com-
pound dissolved in DMSO or DMSO control for 24 h before har-
vesting cells in MSD lysis buffer (MesoScale Discovery, Rockville,
MD, USA). Compounds were titrated from 0.1 nm up to 10 mm
final concentrations. Total DMSO concentrations in cell culture
were kept at 0.5%. Tumor samples were lysed in a 10-fold excess
of MSD lysis buffer using a tissue lyser (Qiagen, Hilden, Germany).
After determination of the concentration, small aliquots were
frozen and stored at À808C for later analysis. MEK phosphorylation
was analyzed by using 20 mg tumor lysate. Phosphoprotein analysis
on the MSD platform was done using the commercially available
assays according to the manual of the commercially available assay
for phospho-MEK1/2 (Ser217/Ser221) (K151CWD-1, MesoScale Dis-
covery) on a sector 6000 imager (MesoScale Discovery).
3-[(2-Fluoro-4-iodophenyl)amino]-5-[(2-oxo-2,3-dihydro-1,3-ben-
zoxazol-5-yl)oxy]isonicotinamide
(32):
1H NMR
(300 MHz,
[D6]DMSO, 300 K): d=8.08–8.13 (m, 2H), 8.05 (s, 1H), 7.95 (s, 1H),
7.66 (s, 1H), 7.62 (dd, 1H), 7.43 (d, 1H), 7.28 (m, 2H), 7.11 (t, 1H),
6.87 (d, 1H), 6.81 ppm (dd, 1H).
Kinase assay
As a measure of the activation of MEK1, the phosphorylation of
Ser217 and Ser221 of MEK1 (C-terminally His6-tagged GST–MEK1
fusion protein, Millipore cat. no. 14-420) by the recombinant kinase
domain of human Cot1 (residues 30–397, N-terminally His6-tagged,
Millipore cat. no. 14-703) was assessed by using the following TR-
FRET-based assay: A solution of the test compound in DMSO
(50 nL) was pipetted into a black, low-volume 384-well microtiter
plate (Greiner Bio-One, Germany); a solution (3 mL) of 24 nm GST–
MEK1 and 166.7 mm ATP in assay buffer [50 mm Tris·HCl pH 7.5,
10 mm MgCl2, 2 mm dithiothreitol, 0.01% (v/v) Igepal CA-630
(Sigma), 5 mm b-phosphoglycerol] was added, and the mixture
was incubated for 10 min at 228C to allow pre-binding of the test
compounds to GST–MEK1 before the start of the kinase reaction.
The kinase reaction was then started by the addition of a solution
of Cot1 in assay buffer (2 mL), and the resulting mixture was incu-
bated for 20 min at 228C. The concentration of Cot1 in the assay
was adjusted depending on the activity of the enzyme lot to have
the assay in the linear range; a typical enzyme concentration was
~2 ngmLÀ1. The reaction was stopped by the addition of a solution
(5 mL) of TR-FRET detection reagents [13 nm anti-GST-XL665
(#61GSTXLB, Cisbio Bioassays, France), 1 nm europium cryptate la-
beled anti-phospho-MEK1/2 (Ser217/221) (#61P17KAZ, Cisbio Bio-
assays)] in aqueous EDTA solution [100 mm EDTA, 500 mm KF, 0.2%
(w/v) bovine serum albumin in 100 mm HEPES/NaOH pH 7.5]. The
resulting mixture was incubated for 2 h at 228C to allow binding
of the phosphorylated GST–MEK1 to the anti-GST-XL665 and the
europium cryptate labeled anti-phospho-MEK1/2 antibody. Subse-
quently, the amount of Ser217/Ser221-phosphorylated GST–MEK1
was evaluated by measurement of the resonance energy transfer
from the europium cryptate labeled anti-phospho-MEK antibody to
the anti-GST-XL665. Thus, the fluorescence emissions at l 620 and
665 nm after excitation at l 350 nm were measured in a TR-FRET
reader (Pherastar, BMG Labtech, Germany) and the ratio of the
emissions at l 665 and 620 nm was taken as the measure for the
amount of phosphorylated MEK1.
A549 tumor model
A549 cells (1.5”106 cells suspended in 50% Matrigel/50% culture
medium without FCS) were implanted subcutaneously into female
athymic nu/nu mice (Taconic). Tumors were allowed to grow to the
predetermined size of ~25 mm2. Animals were then evenly distrib-
uted between treatment and control groups, and treatment was
started and continued until the untreated control group had to be
euthanized due to large tumor size. Treatment of each animal was
based on individual body weight. Tumor response, by the mea-
surement of tumor area (product of the longest diameter and its
perpendicular) with a caliper, and body weight were determined
twice a week. Housing and handling of animals was in strict com-
pliance with European and German Guidelines for Laboratory
Animal Welfare.
Acknowledgements
The authors thank Franziska Probst, Ingrid Schumann, Sandra
Lange, Bernd Schrçter, Meike Fehder, Andreas Jung, Wilhelm Ebel,
Tanja Jung, Hatu Lam, Petra Kçlsch, Jenny Markert, Kai Dçrre,
Frank Igogeit, Jasmin Breusing, Elke Schmid, Daniel Seifert, Steve
Baethge and Melanie Lerch-Wisniewski for technical support, O.
Schenk for HPLC separations, J. Lorenz and G. Depke for analyti-
cal support, and Kay Greenfield for her contributions to drafting
and editing this publication.
Keywords: feedback activation · inhibitors · kinases · MEK ·
oncology · structure-based drug design
Cell proliferation assays
For proliferation assays, A375 (CRL-1619, ATCC, USA) human mela-
noma cells were seeded at 2500 cells per well, A549 (ACC-107,
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2012
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[1] J. F. Ohren, H. Chen, A. Pavlovsky, C. Whitehead, E. Zhang, P. Kuffa, C.
Yan, P. McConnell, C. Spessard, C. Banotai, W. T. Mueller, A. Delaney, C.
Omer, J. Sebolt-Leopold, D. T. Dudley, I. K. Leung, C. Flamme, J. Warmus,
M. Kaufman, S. Barrett, H. Tecle, C. A. Hasemann, Nat. Struct. Mol. Biol.
2004, 11, 1192.
P. Savy, M. Jones, E. Gancia, B. Burton, R. Newman, J. Boggs, E. Chan, J.
Chan, E. Choo, M. Merchant, P. Rudewicz, M. Ultsch, C. Wiesmann, Q.
Yue, M. Belvin, S. Price, J. Med. Chem. 2012, 55, 4594; i) T. Aoki, I. Hyoh-
doh, N. Furuichi, S. Ozawa, F. Watanabe, M. Matsushita, M. Sakaitani, K.
Ori, K. Takanashi, N. Harada, Y. Tomii, M. Tabo, K. Yoshinari, Y. Aoki, N.
Shimma, H. Iikura, Bioorg. Med. Chem. Lett. 2013, 23, 6223.
[5] I. V. Hartung, M. Hitchcock, F. Pühler, R. Neuhaus, A. Scholz, S. Hammer,
K. Petersen, G. Siemeister, D. Brittain, R. C. Hillig, Bioorg. Med. Chem.
Lett. 2013, 23, 2384.
[6] E. B. Haura, A. D. Ricart, T. G. Larson, P. J. Stella, L. Bazhenova, V. A. Miller,
R. B. Cohen, P. D. Eisenberg, P. Selaru, K. D. Wilner, S. M. Gadgeel, Clin.
Cancer Res. 2010, 16, 2450.
[7] C. Iverson, G. Larson, C. Lai, L.-T. Yeh, C. Dadson, P. Weingarten, T. Apple-
by, T. Vo, A. Maderna, J.-M. Vernier, R. Hamatake, J. N. Miner, B. Quart,
Cancer Res. 2009, 69, 6839.
[2] S. Price, Expert Opin. Ther. Pat. 2008, 18, 603.
[3] a) A. G. Gilmartin, M. R. Bleam, A. Groy, K. G. Moss, E. A. Minthorn, S. G.
Kulkami, C. M. Rominger, S. Erskine, K. E. Fisher, J. Yang, F. Zappacosta,
R. Annan, D. Sutton, S. G. Laquerre, Clin. Cancer Res. 2011, 17, 989; b) H.
Abe, S. Kikuchi, K. Hayakawa, T. Iida, N. Nagahashi, K. Maeda, J. Sakamo-
to, N. Matsumoto, T. Miura, K. Matsumura, N. Seki, T. Inaba, H. Kawasaki,
T. Yamaguchi, R. Kakefuda, T. Nanayama, H. Kurachi, Y. Hori, T. Yoshida, J.
Kakegawa, Y. Watanabe, A. G. Gilmartin, M. C. Richter, K. G. Moss, S. G.
Laquerre, ACS Med. Chem. Lett. 2011, 2, 320.
[4] a) L. Lee, H. Niu, R. Rueger, Y. Igawa, J. Deutsch, N. Ishii, S. Mu, Y. Saka-
moto, R. Busse-Reid, C. Gimmi, P. Goelzer, S. De Schepper, Y. Yoshimura,
J. Barrett, Y. Ishikawa, G. Weissgerber, R. Peck, Clin. Cancer Res. 2009, 15,
7368; b) E. F. Choo, M. Belvin, J. Chan, K. Hoeflich, C. Orr, K. Robarge, X.
Yang, M. Zak, J. Boggs, Xenobiotica 2010, 40, 751; c) Q. Dong, D. R.
Dougan, X. Gong, P. Halkowycz, B. Jin, T. Kanouni, S. M. O’Connell, N.
Scorah, L. Shi, M. B. Wallace, F. Zhou, Bioorg. Med. Chem. Lett. 2011, 21,
1315; d) Y. Isshiki, Y. Kohchi, H. Ikura, Y. Matsubara, K. Asoh, T. Murata,
M. Kohchi, E. Mizuguchi, S. Tsujii, K. Hattori, T. Miura, Y. Yoshimura, S.
Aida, M. Miwa, R. Saitoh, N. Murao, H. Okabe, C. Beluni, C. Janson, C.
Likacs, V. Schück, N. Shimma, Bioorg. Med. Chem. Lett. 2011, 21, 1795;
e) V. E. Laing, D. C. Brookings, R. J. Carbery, J. G. Simorte, M. C. Hutch-
ings, B. J. Langham, M. A. Lowe, R. A. Allen, J. R. Fetterman, J. Turner, C.
Meier, J. Kennedy, M. Merriman, Bioorg. Med. Chem. Lett. 2012, 22, 472;
f) M. E. Adams, M. B. Wallace, T. Kanouni, N. Scorah, S. M. O’Connell, H.
Miyake, L. Shi, P. Halkowycz, L. Zhang, Q. Dong, Bioorg. Med. Chem. Lett.
2012, 22, 2411; g) K. D. Rice, N. Aay, N. K. Anand, C. M. Blazey, O. J.
Bowles, J. Bussenius, S. Costanzo, J. K. Curtis, S. C. Defina, L. Dubenko, S.
Engst, A. A. Joshi, A. R. Kennedy, A. I. Kim, E. S. Koltun, J. C. Lougheed,
J.-C. L. Manalo, J.-F. Martini, J. M. Nuss, C. J. Peto, T. H. Tsang, P. Yu, S.
Johnston, ACS Med. Chem. Lett. 2012, 3, 416; h) R. A. Heald, P. Jackson,
[8] K. Kim, S.-Y. Kong, M. Fulciniti, X. Li, W. Song, S. Nahar, P. Burger, M. J.
Rumizen, K. Podar, D. Chauhan, T. Hideshima, N. C. Munshi, P. Richard-
son, A. Clark, J. Ogden, A. Goutopoulos, L. Rastelli, K. C. Anderson, Y.-T.
Tai, Br. J. Haematol. 2010, 149, 537.
[9] D. B. Solit, L. A. Garraway, C. A. Pratilas, A. Sawai, G. Getz, A. Basso, Q.
Ye, J. M. Lobo, Y. She, I. Osman, T. R. Golub, J. Sebolt-Leopold, W. R. Sell-
ers, N. Rosen, Nature 2006, 439, 358.
[10] a) G. Hatzivassiliou, J. R. Haling, H. Chen, K. Song, S. Price, R. Heald,
J. F. M. Hewitt, M. Zak, A. Peck, C. Orr, M. Merchant, K. P. Hoeflich, J.
Chan, S.-M. Luoh, D. J. Anderson, M. J. C. Ludlam, C. Wiesmann, M.
Ultsch, L. S. Friedman, S. Malek, M. Belvin, Nature 2013, 501, 232; b) N.
Ishii, N. Harada, E. W. Joseph, K. Ohara, T. Miura, H. Sakamoto, Y. Matsu-
da, Y. Tomii, Y. Tachibana-Kondo, H. Iikura, T. Aoki, N. Shimma, M. Arisa-
wa, Y. Sowa, P. I. Poulikakos, N. Rosen, Y. Aoki, T. Sakai, Cancer Res. 2013,
73, 4050.
[11] W. Kueng, E. Silber, U. Eppenberger, Anal. Biochem. 1989, 182, 16.
Received: September 26, 2015
Published online on November 6, 2015
ChemMedChem 2015, 10, 2004 – 2013
www.chemmedchem.org
2013
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim