Discovery of selective and orally available spiro-3-piperidyl ATP-competitive MK2 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.04.016

The identification of a potent, selective, and orally available MK2 inhibitor series is described. The initial absence of oral bioavailability was successfully tackled by moving the basic nitrogen of the spiro-4-piperidyl moiety towards the electron-deficient pyrrolepyridinedione core, thereby reducing the pK_a and improving Caco-2 permeability. The resulting racemic spiro-3-piperidyl analogues were separated by chiral preparative HPLC, and the activity towards MK2 inhibition was shown to reside mostly in the first eluting stereoisomer. This led to the identification of new MK2 inhibitors, such as (S)-23, with low nanomolar biochemical inhibition (EC₅₀ 7.4 nM) and submicromolar cellular target engagement activity (EC₅₀ 0.5 μM).

Full text of DOI:10.1016/j.bmcl.2011.04.016

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3823–3827
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of selective and orally available spiro-3-piperidyl
ATP-competitive MK2 inhibitors
Allard Kaptein, Arthur Oubrie, Edwin de Zwart, Niels Hoogenboom, Joeri de Wit, Bas van de Kar,
Maaike van Hoek, Gerard Vogel, Vera de Kimpe, Carsten Schultz-Fademrecht, Judith Borsboom,
Mario van Zeeland, Judith Versteegh, Bert Kazemier, Jeroen de Roos, Frank Wijnands, John Dulos,
⇑
Martin Jaeger, Paula Leandro-Garcia, Tjeerd Barf
Merck Research Laboratories, MSD, PO Box 20, 5340 BH Oss, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
The identification of a potent, selective, and orally available MK2 inhibitor series is described. The initial
absence of oral bioavailability was successfully tackled by moving the basic nitrogen of the spiro-4-piper-
idyl moiety towards the electron-deficient pyrrolepyridinedione core, thereby reducing the pKa and
improving Caco-2 permeability. The resulting racemic spiro-3-piperidyl analogues were separated by chi-
ral preparative HPLC, and the activity towards MK2 inhibition was shown to reside mostly in the first
eluting stereoisomer. This led to the identification of new MK2 inhibitors, such as (S)-23, with low nano-
molar biochemical inhibition (EC50 7.4 nM) and submicromolar cellular target engagement activity (EC50
Received 11 February 2011
Revised 29 March 2011
Accepted 3 April 2011
Available online 16 April 2011
Keywords:
MAPKAPK2
MK2
0.5 lM).
Ó 2011 Elsevier Ltd. All rights reserved.
Anti-TNF
a
Oral bioavailability
Rheumatoid arthritis
The p38-MK2 signaling pathway is crucial for activation of
pro-inflammatory cytokines in response to stress signals including
pro-inflammatory cytokines, bacterial lipopolysaccharide (LPS)
and hypoxia.1 The pathway plays a central role in pathogenic cel-
lular mechanisms of rheumatoid arthritis (RA) and other related
autoimmune diseases. Drug discovery on p38 has resulted in sev-
eral inhibitors that have progressed to phase 2 in the clinic. Drug
toxicity, lack of efficacy and activation of other inflammatory sig-
naling pathways following p38 inhibition, have hampered progres-
sion to phase 3 clinical trials.2 MK2 is downstream of p38 and an
attractive alternative drug discovery target for the p38-MK2 sig-
4-piperidyl-based MK2 inhibitors suffered from a lack of oral bio-
availability (F = 0.7% (po) in mice). Following iv or sc administra-
tion, compound 1 showed significant inhibition of LPS-induced
TNFa at 12.5 and 30 mg/kg in rats, respectively. Thus, exposure
of 1 was sufficient to achieve in vivo efficacy with these adminis-
tration routes.
With compound 1 as the early lead, we set out to identify ana-
logues that combine MK2 inhibitory activity with favorable ADME
parameters. The general preparation of the inhibitors is outlined in
Scheme 1, and an optimized route as compared to that described in
an earlier report.6 On the commercially Boc-protected piperidine A,
a trimethoxybenzyl group (TMB) was introduced via a reductive
amination. It was advantageous to protect the primary amine func-
tion with a TMB group, since this sterically demanding group facil-
itated the Dieckmann condensation step. It forces the attacking
malonyl carbanion in close proximity to the reacting methyl ester
naling pathway, also given the anti-TNF
a phenotype of MK2 defi-
cient mice.3
Profiles of several ATP-competitive chemotypes have been pub-
lished, and these have been reviewed recently.4 One of the re-
ported scaffolds is the pyrrolopyridinone class of MK2 inhibitors
as exemplified by PH-089 (Fig. 1).5 Proof of principle has been
demonstrated with this orally available MK2 inhibitor, but the
in vitro and in vivo activity is limited for this compound. As previ-
ously reported, introduction of a 4-piperidyl moiety on the pyrro-
O
O
NH
NH
N
N
lopyridinone scaffold gave
a
substantial improvement in
N
N
H
N
F
biochemical and cellular activity as well as in solubility.6 However,
H
O
N
H
O
1
PH-089
⇑
Corresponding author. Tel.: +31 412 661768; fax: +31 412 662546.
E-mail address: tjeerd.barf@merck.com (T. Barf).
Figure 1. Pyrrolopyridinone PH-089 and compound 1.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.04.016

3824
A. Kaptein et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3823–3827
O
the (hetero)-aryl and the pyrimidyl is essential for proper binding
TMB
TMB
to the hinge and front pocket regions. Despite the low nanomolar
biochemical potency of many of these compounds, substantial cel-
lular inhibition was only obtained if the lipophilicity was above a
certain level (c Log P >1). Examples 16–20 are probably not suffi-
ciently lipophilic to produce a decent cellular effect. Introduction
of aliphatic groups, such as a cyclopentyl in 21 (as compared to
unsubstituted 20), restored the cellular activity somewhat. In-
creased lipophilicity did not necessarily correlate with membrane
permeability as determined in our PAMPA and Caco-2 assays,
and permeability properties were disappointingly low for all 4-
piperidyl analogues (Table 2).
O
NH2
O
HN
N
MeO
MeO
O
a
b, c
e
and d
N
N
N
A
B
C
Boc
Boc
Boc
O
N
O
Cl
N
N
TMB
N
NH
N
N
f and g
N
N
N
N
H
H
Cl
Ar
O
Br
N
H
E
D
3 - 21
We reasoned that the lack of permeability could be the prime
cause of both low cellular activity and absence of oral bioavailabil-
ity in this series, as the microsomal and hepatocyte stabilities of
our MK2 inhibitors were generally very good (Table 2). The basicity
of the unsubstituted 4-piperidyl moiety was determined experi-
mentally; the pKa’s of 1 and 3a were 9.3 and 9.1, respectively (Ta-
ble 2). We therefore sought to lower the basicity with the aim to
improve the above mentioned parameters. In general, pKa values
can potentially be reduced via introduction of (electron-withdraw-
ing) aliphatic substituents. Another, less conventional possibility,
was to move the basic nitrogen closer to the electron-deficient pyr-
role ring. By doing so, a net electron-withdrawing effect could low-
er the pKa. Consistent with this hypothesis, the calculated log P
values of 4- and 3-piperidyl analogues with identical molecular
formulas differed on average by 0.3–0.4 log unit (compare e.g., 1
and 23, and 3a and 22 in Table 3).
Preparation of the 3-piperidyl analogues started with iodome-
thylation of commercially available F (Scheme 2). Subsequent
nucleophilic substitution with sodium azide, and reduction to the
primary amine gave intermediate H. Reductive N-alkylation with
trimethoxybenzaldehyde, and construction of the oxolactam ring
according the methods described for the 4-piperidyl derivatives,
furnished intermediate K. Final test compounds were again pre-
pared in analogy of Scheme 1.
Boc
Scheme 1. Reagents and conditions: (a) 2,4,6-trimethoxybenzaldehyde, NaB-
H(OAc)3, EtOH, rt (100%); (b) ClCOCH2COOEt, DMAP, pyridine, DCM, rt (100%); (c)
NaOMe, MeOH, 60 °C (100%); (d) ACN/H2O (1:1) 80 °C (98%); (e) Compd E, NH4OAc,
EtOH (35%); (f) arylboronic acid, PdCl2 (dppf), K3PO4Á7H2O, dioxane, 140 °C, MW; (g)
TFA, rt (4–48% in two steps).
group on the bridgehead resulting in drastically improved isolated
yields. Also, it provided UV-positive intermediates that were easy
to track during the workup and purification stages. Acylation with
ethyl (chloroformyl)acetate, a Dieckmann condensation and decar-
boxylation subsequently rendered intermediate C. Paal–Knorr con-
densation was carried out with chloropyrimidine derivative E (see
insert Scheme 1),6 followed by a Suzuki-coupling to introduce the
various aryl moieties in the front pocket. Final compounds were
generated by a one-step acidic removal of the Boc- and TMB-pro-
tective groups. The N-alkylated analogues 3b and 3c were synthe-
sized via standard reductive amination procedures on the
unsubstituted 4-piperidyl parent molecule 3a.
The basic amine function seems to be a prerequisite for MK2
inhibitory activity, since the direct pyrane analogue
2
demonstrates a 180-fold drop in activity (Table 1).7 Benzofurane
derivative 3a was equipotent to the methylenedioxophenyl
substitution, but the higher cellular potency observed in the TNF
a
Compared to their basic amine containing equivalents, N-alkyl
analogues 3b (Me) and 3c (Et) exhibited higher lipophilicity, lower
pKa values and improved Caco-2 permeability at the expense of
biochemical activity (Table 2). This appeared to only marginally
improve oral exposure in rats, at least within the benzofuranyl ser-
production in THP1 and PBMC cells could not be confirmed in the
Hsp27 phosphorylation target engagement assay (pHsp27). N-
Alkylation of the 4-piperidyl moiety with a methyl (3b) or ethyl
(3c) in the benzofurane series resulted in a 2- and 10-fold loss in
biochemical potency. As for 3a, the cellular THP1 readouts did
not correlate very well. Plotting the target engagement (pHsp27)
ies, when comparing the 4-piperidyl NH (AUC <0.006
(0.078 M h), N-Et (0.154 M h). A substantial improvement in
oral exposure in this series was obtained through the 3-piperidyl
(0.472 M h) analogue. This trend towards improved oral bioavail-
lM h), N-Me
l
l
and cytokine production (TNFa) EC50 values in THP1 cells, the high
overall correlation of these two parameters becomes evident
(Fig. 2). This graph also clearly shows that the benzofurane-derived
MK2 inhibitors belong to the few outliers. There is a 4- to 10-fold
l
ability was also observed in many other 3-piperidyl analogues, for
example, 23 and 25. For these compounds, Caco-2 permeabilities
exceeded 20 nm/s, and the pKa values ranged from 8.2 to 8.4. There
is a fairly good correlation between lower pKa values, increased
permeability and improved oral exposure following rapid rat
pharmacokinetic experiments. The biochemical EC50 values of the
3-piperidyl analogues were somewhat compromised, but this is
more than compensated for by this series’ PK properties: The
difference in EC50 values obtained in the pHsp27 and TNFa assays,
while all other compounds are within a 2-fold difference from equi
potency on both parameters. This suggests that a promiscuous
component aids to the inhibition of LPS-induced cytokine produc-
tion in THP1 cells. Since the benzofuranyl series did not reveal any
cytotoxicity in THP1 cells, as determined with Alamar Blue in con-
cert with cytokine production (>10
lM for 3a–3c; data not shown),
AUC in rats after oral administration of 23 was 3.2 lM h along with
off-target contribution to the overall efficacy on LPS-induced TNF
a
acceptable half lives in human and rat liver microsomes. The oral
production is anticipated for this particular subseries. The poten-
tial to inhibit p38 was routinely monitored for these analogues,
and no substantial inhibition was observed (data not shown).
While screening for additional substituents on the (hetero)-aryl
portion in the front pocket, it became evident that the nature of
properties that are accepted in this region is very diverse (Table
1). Hydrogen bond accepting (4, 15–17) as well as donating sub-
stituents (18–21) are allowed. Also, small lipophilic groups on
the meta position, or bulkier aliphatic groups on the para position
can be beneficial. In sharp contrast, ortho substitution is not
allowed except for fluorine as in 8, showing that co-planarity of
bioavailability of 23 was confirmed in mice and with F = 48%
(AUC = 3.5
(F = 0.7%). Also, the cellular activities as shown by both pHsp27
and TNF inhibition EC50 values are still in the low micromolar
lM h) its oral exposure is superior to that of 1
a
range (Table 3, compounds 23–26) and comparable to 4-piperidyl
analogue 1.
The generation of 3-piperidyl analogues provided another inter-
esting avenue, namely separating and testing of enantiomers. This
often provides increased potency, and could generate additional
proof as to whether MK2 inhibition is indeed correlated with
cellular inhibition of pHsp27 and TNF
a in THP1 cells. A couple of

A. Kaptein et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3823–3827
3825
Table 1
Inhibitory MK2 activity of compounds 1–21
O
O
O
NH
NH
N
N
N
N
N
N
H
H
Ar
Ar
N
2
3 - 21
R1
a
b
c
Compd
Ar
R1
MK2 IMAP EC50 (nM)
TNF EC50
(lM)
pHsp27 EC50 (lM)
PH-089
1
2
3a
3b
3c
—
—
H
—
H
Me
Et
372
4.3
724
5.9
13
9.1
2.7
3,4(–OCH2O–)C6H3
3,4(–OCH2O–)C6H3
Benzofuran-2-yl
Benzofuran-2-yl
Benzofuran-2-yl
0.91
>10
0.28
0.47
1.6
0.62
NTd
1.2
1.6
>10
59
c Log P
1.1
1.1
1.8
3.1
1.8
1.8
2.5
2.5
3.5
2.8
3.1
1.6
0.2
0.7
0
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
3-(COMe)C6H4
4-(COMe)C6H4
3,4(–O(CH2)3O–)C6H3
3-(iPr)C6H4
2-FC6H4
3-FC6H4
3-(CF3)C6H4
4-(CF3)C6H4
4-(tBu)C6H4
Naphth-2-yl
3,5-Di-ClC6H3
Quinolin-3-yl
Pyrid-3-yl
5-MeO-pyrid-3-yl
6-NH2-pyrid-3-yl
6-NHCOEt-pyrid-3-yl
2-NH2pyrimid-5-yl
2-cPentNHpyrimid-5-yl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
11
9.8
10
112
29
8.3
15
6.9
6.2
5.1
4.5
6.3
1.1
6.9
15
6.2
5.0
4.9
>10
4.2
1.4
4.2
6.3
5.9
3.6
6.3
4.8
>10
>10
>10
>10
>10
2.6
3.6
2.1
1.7
0.7
-0.7
1.6
a
Values are means of two independent experiments with duplicates for each dilution per experiment. For experimental details see Ref. 6.
THP1 cells are pre-incubated for 30 min with MK2 inhibitors (duplicates for each dilution) prior to stimulation with LPS. Four hours
b
following LPS induction culture medium is withdrawn from the cells for TNFa measurements. For experimental details see Ref. 6.
c
THP1 cells are pre-incubated for 60 min with MK2 inhibitors (duplicates for each dilution) prior to stimulation with LPS. 10 min following
LPS induction cells are lysed for measurement of Hsp27 and phospho-Hsp27.8
d
NT: not tested.
7.00
mine the absolute configuration of each of the stereoisomers of
23.11 In line with our expectations, the S-configuration could be
traced back to first eluting stereoisomer 1, and the optical antipode
to stereoisomer 2 of the common intermediate. In all chiral pairs
biochemical potency could be attributed to the (S)-stereoisomer,
and for the majority of racemic mixtures and isolated stereoiso-
mers, the biochemical and cellular activities correlate well (Table
3). Compound 24 is an exception, possibly because the cellular po-
tency of the enantiomers and the racemate are in the same range,
but in this case the biochemical potency failed to correlate with
6.50
6.00
5.50
5.00
4.50
pHsp27 and TNFa data. With their sub-micromolar cellular activi-
ties, (S)-23 and (S)-25 have the most attractive overall profile.
Moreover, it is anticipated that (S)-23 will have similar oral bio-
availability as its racemate.
4.50
5.00
5.50
6.00
6.50
7.00
In computational models of MK2 and MK3—which was previ-
ously described as a suitable model for MK2 with a near-identical
active site12—compound (S)-23 is almost perfectly aligned with 1
in X-ray structures of, respectively, MK2:1 and MK3:1,6 with the
obvious exception of the interactions involving the basic amine
(Fig. 3). In the S-configuration of the 3-piperidyl ring, the nitrogen
can be positioned within hydrogen-bond distance of the main
chain carbonyl of Glu190 (Glu170 in MK3 numbering) and close
to the side chain of Asn191/Asn171, and this is not possible for
the (R)-23.
THP1_TNFa (-log EC50 (M))
Figure 2. Correlation Hsp27 phosphorylation (pHsp27) and TNF
a production in
THP1 cells. Solid red line indicates equi-potency and dotted indicate a 2-fold
difference. The red dots represent the benzofuranyl series and the green dot is the
reference compound PH-089.
enantiomerically pure 3-piperidyl analogues were prepared via
separation at the chloropyrimidine intermediate stage by means
of chiral preparative HPLC.10 Finally, enantiomerically pure target
compounds (ee >99%) were prepared as described for the racemic
3-piperidyl inhibitors and as outlined in Scheme 1. Using vibra-
tional circular dichroism (VCD) experiments we were able to deter-
Given the good correlation between the pHsp27 and TNFa data
generated for this chemical series, we inferred that potential
off-target kinase inhibition should not intervene with the observed

3826
A. Kaptein et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3823–3827
Table 2
pKa, Caco-2 and oral exposure relationship
O
O
Ar =
*
*
*
*
NH
NH
N
O
N
Ar
N
MeO
O
N
N
O
N
N
H
H
N
N
Ar
O
O
N
R1
I
II
III
IV
R1
a
b
Compd
Spiro
Ar
R1
MK2 IMAP EC50 (nM)
Caco-2 nm/s
pKa (pH 7.4)
HLM/RLM (t1/2 min)
AUC rat (
l
M h)
c Log P
1
4-Piperidyl
4-Piperidyl
4-Piperidyl
4-Piperidyl
3-Piperidyl
3-Piperidyl
3-Piperidyl
3-Piperidyl
I
H
H
Me
Et
H
H
H
H
4.3
5.9
13
59
11
28
24
21
1
0
10
8
15
34
20
42
9.3
9.1
8.1
8.6
8.4
8.3
8.4
8.2
>60/>60
>60/>60
>60/>60
>60/>59.6
>60/>60
>60/48.5
56.7/42.1
>60/>60
<0.006
<0.006
0.078
0.154
0.472
3.210
NTc
1.7
2.4
2.8
3.4
2.8
2.0
1.0
1.7
3a
3b
3c
22
23
24
25
II
II
II
II
I
III
IV
0.882
a–c
See Table 1.
b
Experimentally determined pKa values at pH 7.4.
Table 3
Inhibitory MK2 activity of compounds 23–26
O
*
*
*
*
O
NH
MeO
O
N
N
N
N
N
N
O
H
O
Ar
NH
23
24
M)
25
26
M)
b
c
b
e
Compd
Stereoisomera
MK2 IMAP EC50 (nM)
pHsp27 EC50
(
l
TNF
1.4
a
-THP1 EC50
(l
TNF
1.1
0.45
3.0
1.0
1.3
>10
0.92
0.43
6.9
3.3
0.93
4.3
a-PBMC EC50
(l
M)
23
Rac.
1
2
Rac.
1
2
Rac.
1
2
Rac.
1
2
28
7.4
91
0.93
0.50
7.7
0.9
1.1
2.3
1.4
0.40
>10
2.1
(S)-23
(R)-23
24
(S)-24
(R)-24
25
(S)-25
(R)-25
26
(S)-26
(R)-26
0.85
8.3
1.4
2.4
2.2
1.5
0.98
>10
3.0
24
9.3
126
21
6.9
129
18
6.5
288
1.2
5.8
1.6
8.7
a
b–d
e
Stereoisomers separated by chiral preparative HPLC on the chloropyrimidine intermediate.
See Table 1.
hPBMC’s are pre-incubated for 30 min with MK2 inhibitors (duplicates for each dilution) prior to stimulation with LPS. Four hours following
LPS induction culture medium is withdrawn from the cells for TNF
a
measurements.9
EtO
O
N
O
I
O
NH2
N
EtO
EtO
a
b and c
N
Boc
Boc
Boc
F
G
H
O
N
TMB
H
TMB
O
N
N
d
e, f
and g
EtO
O
N
Boc
Boc
K
J
Scheme 2. Reagents and conditions: (a) diiodomethane, LDA, THF À78 °C to rt
(87%); (b) NaN3, DMSO, 70 °C (69%); (c) Pd(OH)2, H2(g), EtOH, rt, (97%); (d) 2,4,6-
trimethoxybenzaldehyde, NaBH(OAc)3, EtOH, rt (100%); (e) ClCOCH2COOMe, DMAP,
pyridine, DCM, rt (94%); (f) NaOMe, MeOH, 60 °C (91%); (g) ACN/H2O (1:1) 80 °C
(46%).
Figure 3. Overlay of an X-ray structure of
1 (cyan) in human MK3 and a
inhibition of cytokine production in THP1 cells and PBMCs (Table
3). None of the tested compounds inhibited upstream kinase p38.
Yet, Millipore profiling on 233 kinases revealed a number of other
computational model of the (S)-enantiomer of 3-piperidyl analogue 23 (magenta)
bound to MK3. Numbering is consistent with the sequences of MK2/MK3. Similar
structures and models were generated for 1 and 23 bound to MK2 (not shown).

A. Kaptein et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3823–3827
3827
6. Barf, T.; Kaptein, A.; Wilde, S. de; Heijden, R. van der; Someren, R. van; Demont,
D.; Schultz-Fademrecht, C.; Versteegh, J.; Zeeland, M. van; Seegers, N.;
Kazemier, B.; Kar, B. van de; Hoek, M. van; Roos, J. de; Klop, H.; Smeets, R.;
Hofstra, C.; Hornberg, J.; Oubrie, A. Bioorg. Med. Chem. Lett. 2011. Accepted for
publication.
kinases that were substantially inhibited. At 1
kinases were inhibited P95% by 1. Benzofuranyl derivative 3a
was less selective which aids to the conclusion that its anti-TNF
effects are apparently not only mediated by MK2. This also demon-
strates that the front pocket substitution has significant impact on
overall selectivity observed in kinase panels. 3-Piperidyl analogues
23–25 exhibited much higher selectivity in kinase panels. For in-
stance, compounds 23 and 24 showed P95% inhibition of only
lM, 16 out of 233
a
7. Pyrane analogue 2 was essentially prepared as described for the 4-piperidyl
compounds, starting with methyl 4-aminomethyl-tetrahydro-pyran-4-
carboxylate.
8. THP1 cells are seeded at 1.5 Â 105 cells per well in DMEM F12 (Gibco)
supplemented with Penicillin/Streptomycin (80 U/mL; 80 lg/mL) and 10% FBS
(Fetal Bovine Serum) in 96 well plate. Cells are left to rest for at least 18 h in a
humidified atmosphere at 5–7% CO2 at 37 °C, prior to incubation with test
compounds. Dilutions of test compounds in DMSO (0.1% DMSO final
concentration in assay) are added to the cells. Following a 60-min incubation
two kinases at 1 lM next to potent blockade of MK2, MK3, and
MK5. Inhibition of both MK3 and MK5 will most likely contribute
to the observed cellular effects, since they share common functions
towards downstream events.13 Other kinases that are substantially
LPS is added (final LPS concentration 10 lg/mL). Following a 60 min incubation
cells are harvested and lysed for the measurement of total and phosphorylated
Hsp27 (pHsp27). Total Hsp27 and pHsp27 are quantified using the Multi-Spot
assay from Meso Scale according to protocol.
inhibited (P80% at 1 lM) by all tested compounds in this series are
PIM1&3, DRAK1, ERK1&2, and CHK2. The inhibition of these ki-
nases will need special attention in future optimization cycles.
In summary, simply moving the piperidyl nitrogen from the
four to the three position provided an MK2 inhibitor series that
matched the desired preclinical profile. In doing so, the oral bio-
availability in rodents improved dramatically to acceptable expo-
sure levels. The lower pKa of the basic amine appears to be of
paramount importance to achieve this. Chiral separation and anal-
ysis identified the most active (S)-enantiomer. The in vivo efficacy
following oral administration of these enantiomerically pure 3-
piperidyl analogues in preclinical arthritis models will be the sub-
ject of future investigations.
9. PBMCs are isolated from buffy coats from human blood and are seeded at
2.5 Â 105 cells per well in DMEM F12 (Gibco) supplemented with Penicillin/
Streptomycin (80 U/mL; 80 lg/mL) and 10% FBS (Fetal Bovine Serum) in 96
well plate. Cells are left to rest for at least 1 h, prior to incubation with test
compounds. Dilutions of test compounds in DMSO (0.1% DMSO final
concentration in assay) are added to the cells. Following a 30 min incubation
LPS is added (final LPS concentration 10
5–7% CO2 at 37 °C in humidified atmosphere, culture medium is collected for
the measurement of TNF . The amount of TNF in the culture medium is
quantified by ELISA using HuTNF assay from Cytoset according to protocol.
10. Separation of enantiomers was effected at the chloropyrimidine stage on a
m), eluting with
lg/mL). Following a 4 h incubation at
a
a
a
Chiralpack AD-H column (20 Â 250 mm, particle size
5 l
heptane/IPA (80/20 v/v). Stereoisomer 1 eluted at 15.8 min, and stereoisomer 2
at 27.8 min. Ee’s were >99% as determined with analytical chiral
chromatography.
11. Vibrational Circular Dichroism (VCD) configuration determination of the
stereoisomers of 23: Samples (7.6 mg each) were dissolved in 0.2 mL DMSO-
Acknowledgments
d6 and placed in a cell with 100 lm pathlength and BaF2 windows. IR and VCD
spectra were recorded on a Chiral IR2X™ VCD spectrometer from BioTools, Inc.,
equipped with DualPEM accessories. The instrument was optimized at
We thank Dr. Hans van Eenennaam and Dr. Rob Nelissen for the
valuable discussions regarding the MK2 program. BioTools is
acknowledged for the absolute configuration determination
studies.
1400 cmÀ1
. Spectra were collected for 8 h (stereoisomer 1) and 15 h
(stereoisomer 2) with 4 cmÀ1 resolution. Calculations: A model of the (S)-
configuration was built and a conformational search was carried out with
Hyperchem (Hypercube, Inc., Gainesville, FL, USA) for the entire structure at
the molecular mechanics level. Geometry optimization, frequency, IR and VCD
intensity calculations of 47 conformers, which resulted from the
conformational search, were carried out at the DFT level (B3LYP functional/
6-31G(d) basis set) with GAUSSIAN 09 (Gaussian Inc., Wallingford, CT). The four
lowest-energy conformers were used for calculating the averaged VCD and IR
spectra. The configurations were assigned by spectral comparison of
experimental and calculated spectra.
References and notes
1. Saklatvala, J. Curr. Opin. Pharmacol. 2004, 4, 372.
2. (a) Sweeney, S. E. Nat. Rev. Rheumatol. 2009, 5, 475; (b) Hammaker, D.;
Firestein, G. S. Ann. Rheum. Dis. 2010, 69, i77; (c) Cheung, P. C. F.; Campbell, D.
G.; Nebreda, A. R.; Cohen, P. EMBO J. 2003, 22, 5793.
3. (a) Kotlyarov, A.; Neininger, A.; Schubert, C.; Eckert, R.; Birchmeier, C.; Volk, H.
D.; Gaestel, M. Nat. Cell. Biol. 1999, 1, 94; (b) Hegen, M.; Gaestel, M.; Nickerson-
Nutter, C. L.; Lin, L.-L.; Telliez, J.-B. J. Immunol. 2006, 177, 1913.
4. Schlapbach, A.; Huppertz, C. Future Med. Chem. 2009, 1, 1243.
5. Anderson, D. R.; Meyers, M. J.; Vernier, W. F.; Mahoney, M. W.; Kurumbail, R.
G.; Caspers, N.; Poda, G. I.; Schindler, J. F.; Reitz, D. B.; Mourey, R. J. J. Med. Chem.
2007, 50, 2647.
12. Cheng, R.; Felicetti, B.; Palan, S.; Toogood-Johnson, I.; Scheich, C.; Barker, J.;
Whittaker, M.; Hesterkamp, T. Protein Sci. 2010, 19, 168.
13. (a) Ronkina, N.; Kotlyarov, A.; Dittrich-Breiholz, O.; Kracht, M.; Hitti, E.;
Milarski, K.; Askew, R.; Marusic, S.; Lin, L.-L.; Gaestel, M.; Telliez, J.-B. Mol. Cell.
Biol. 2007, 27, 170; (b) New, L.; Jiang, Y.; Zhao, M.; Liu, K.; Zhu, W.; Flood, L. J.;
Kato, Y.; Parry, G. C. N.; Han, J. EMBO J. 1998, 17, 3372.