Optimisation of ITK inhibitors through successive iterative design cycles

Article abstract of DOI:10.1016/j.bmcl.2011.01.035

Based on a hit cluster of compounds inhibiting interleukin-2 inducible T-cell kinase (ITK) in the submicromolar range a series of ITK inhibitor libraries were synthesized. Through iterative design cycles including kinase crystal structure information, indolylindazole libraries were identified which showed low nanomolar activity in enzymatic and cellular assays. The potential of these novel lead series was confirmed through in vivo tests in an anti-CD3-IL2 mouse model. The intravenous administration of highly potent ITK inhibitor 11o resulted in dose-dependent, efficient suppression of IL-2.

Full text of DOI:10.1016/j.bmcl.2011.01.035

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1852–1856
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimisation of ITK inhibitors through successive iterative design cycles
Matthias Herdemann , Alexander Weber ,à, Jérôme Jonveaux, Frank Schwoebel §, Michael Stoeck,
Isabelle Heit
⇑
⇑
Nycomed GmbH, Byk-Gulden-Str. 2, D-78467 Konstanz, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Based on a hit cluster of compounds inhibiting interleukin-2 inducible T-cell kinase (ITK) in the
submicromolar range a series of ITK inhibitor libraries were synthesized. Through iterative design cycles
including kinase crystal structure information, indolylindazole libraries were identified which showed
low nanomolar activity in enzymatic and cellular assays. The potential of these novel lead series was con-
firmed through in vivo tests in an anti-CD3-IL2 mouse model. The intravenous administration of highly
potent ITK inhibitor 11o resulted in dose-dependent, efficient suppression of IL-2.
Received 15 December 2010
Revised 9 January 2011
Accepted 10 January 2011
Available online 14 January 2011
Keywords:
Interleukin-2 inducible T cell kinase
Lead optimisation
Ó 2011 Elsevier Ltd. All rights reserved.
In vivo testing
Anti-CD3 IL-2 mouse model
The non-receptor tyrosine kinase interleukin-2 inducible T cell
kinase (ITK) is mainly expressed in T cells, mast cells and natural
killer cells.1 The protein kinase is activated in T cells through phos-
phorylation in the ITK activation loop by the kinase LCK after T-cell
receptor (TCR) stimulation.2 Phosphorylation of the ITK substrate
O
N
O
N
O
N
PLC-c
13 leads to calcium mobilization, IL-2 production, prolifera-
O
S
tion and differentiation.4 The lack of ITK results in reduced produc-
tion of Th2 cytokines such as IL-4, IL-5 and IL-13.5 ITK has been
shown to play an important role in the development of T cell
dependent late phase responses of allergic asthma. In studies with
ITKÀ/À mice reduced lung inflammation including goblet cell
hyperplasia, mucous production and airway hyper responsiveness
were observed in an ovalbumin-induced allergic asthma model.6
Similar results were reported with a selective ITK inhibitor7 indi-
cating the important role of ITK kinase activity in inflammatory
processes. Therefore, the tyrosine kinase ITK is regarded as a prom-
ising therapeutic target for T cell mediated diseases.
S
N
H
1
N
O
O
N
N
N
S
N
H
O
O
2
NH2
N
O
H
N
N
H
3
N
O
Previously, we reported the identification of potent ITK inhibi-
tors through focused library design wherein structural information
provided by X-ray structures were included.8 Docking studies of
published ITK inhibitors 1 and 2 (Fig. 1) in ITK ATP binding pocket,9
as well as in depth literature analysis led to an initial hit cluster
N
H
N
4a
N
H
Figure 1. Published ITK inhibitors (1, 2) and ITK inhibitors identified after the first
(3) and second (4a) focused library design cycle. Indazole derivatives (3, ITK
IC50 = 5
l
M) were optimised through ring closure to indolylpyrazolopyridine
M).
derivatives (4a, ITK IC50 = 0.06
l
⇑
Corresponding authors. Tel.: +49 7531 84 3130; fax: +49 7531 84 93130 (I.H.).
E-mail addresses: alexander.weber@boehringer-ingelheim.com (A. Weber),
isabelle.heit@nycomed.com (I. Heit).
represented by compound 3 (ITK IC50 = 5
cycles resulted in the identification of indolylpyrazolopyridine 4a
(ITK IC50 = 0.06 M; CD4+T cells/IL-2 release IC50 = 1.4 M) which
l
M).8 Subsequent design
Present address: EPO, Patentlaan 3-9, 2288 EE The Hague, Netherlands.
Present address: Boehringer Ingelheim Pharma GmbH & Co. KG, 88400 Biberach,
Germany.
à
l
l
§
was considered as a promising starting point for a lead finding
Present address: NOXXON Pharma AG, Max-Dohrn-Str. 8-10, 10589 Berlin,
Germany.
project. Therefore, detailed docking studies were performed with
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.035

M. Herdemann et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1852–1856
1853
4a. Analysis of the docked conformation of 4a within the protein
O2N
environment indicated that substituents at position 6 of the indole
moiety might improve the ITK binding affinity by addressing a
small solvent exposed hydrophobic pocket (Ile-369, Leu-379,
Phe-437, Fig. 2). Furthermore, substitutions at position 5 of the
pyrazolopyridine ring system of 4a could address polar amino
acids in the sugar pocket or interact with Asp-500 of the DFG-mo-
tif. Prior to synthesis, chemical accessible modifications at position
6 of the indole motif as well as at position 5 of the pyrazolopyri-
dine moiety of 4a were virtually enumerated and docked in the
ATP binding pocket of ITK.10 Most interesting side chain variations
were subsequently included in the optimisation cycles.
N
N
H
(i), (ii)
R2
I
O2N
N
N
N
BOC
BOC
5
H2N
(iii), (iv)
N
N
The first optimisation cycle was focused on the exploration of
the indole’s position 6, that is, on the replacement of the 6-meth-
oxy group of compound 4a (Fig. 2). The prepared compounds were
first evaluated in an enzymatic assay using an ITK full-length
enzyme. Subsequently, selected compounds were also tested for
their inhibition of anti-CD3/anti-CD28 induced interleukin-2 (IL-
2) release in primary human CD4+T cells. Protocols of both assays
were identical with those reported previously.8 The introduction of
various groups resulted in the identification of several inhibitors
having enzymatic and cellular potency similar to the potency of
compound 4a. However, these efforts did not result in a straight-
forward, significant increase of in vitro potency.14
+
BOC
O
O
7
B
N
R2
R2 = H, Me, OMe
BOC
6
Scheme 1. Reagents and conditions: (i) I2, KOH, DMF, 0 °C to rt, 2.5 h; (ii) Boc2O,
NEt3, DMAP, DMF, 0 °C to rt, 1 h; (iii) Pd(dppf)2Cl2ÁDCM, Cs2CO3, dioxane, 80 °C,
2–4 h, argon; (iv) H2, 10% Pd/C, AcOEt, rt, 15 h.
Thus, further efforts focused on the indazole moiety, in particu-
lar on position 5. As discussed above, substitution at this position
could allow additional interactions with the protein, in particular
with polar amino acids in the sugar pocket or with Asp-500 of
the DFG-motif.9 The approach chosen for in depth investigation
of position 5 is shown in Scheme 1. Starting from commercially
available 5-nitro-indazole the Boc protected iodo-indazole 5 has
been prepared by regioselective iodination under basic conditions
followed by Boc protection in 77% yield. Key intermediates 7 have
been obtained after Suzuki coupling of iodo-indazole 5 with com-
mercially available indol-2-ylboronic acids 6 using a Pd(dppf) cat-
alyst and cesium carbonate as a base, flash chromatography of the
crude products with cyclohexane/ethyl acetate and subsequent
reduction of the nitro group to the amine group with hydrogen
(1 bar) using Pd/C (10%) as catalyst. Yields for the sequence Suzuki
coupling–hydrogenation were in the range of 62–77% (Scheme 1).
The intermediate 7 has then been transformed into the amides 8,
sulfonamides 9, urea compounds 10 and secondary amine deriva-
tives 11 (Scheme 2). Reaction of 7 with acide chlorides in DME in
the presence of triethylamine yielded quantitative conversion to
the corresponding amides 8. The crude product has then directly
been deprotected with potassium carbonate in a mixture of
water/methanol/THF to give the final compounds 8. For the
preparation of sulfonamides 9 best results were obtained with
1,6-lutidine as base in DCM, followed by the same deprotection
procedure as applied for amides 8. Treatment of 7 with isocyanates
in DCM gave the desired urea function of compounds 10. Standard
conditions for reductive amination11 of ketones and aldehydes
with NaBH(OAc)3 in DCM/AcOH were applied for the preparation
of compounds 11. The Boc protecting group at the secondary
amines being stable under the above mentioned deprotection con-
ditions secondary amines such as 11h, 11m and 11o12 have been
isolated after deprotection with TFA in DCM. Intermediates 7 have
also been deprotected directly to give the corresponding amino
compounds 12 (Scheme 2). The synthetic approach which was ap-
plied for amides 8 which were substituted at position 5 of the inda-
zole has also been used for the preparation of amides 13 which
were substituted at position 6 of the indazole (Table 1). These
amides 13 were thus prepared using commercially available 6-ni-
tro-indazole as starting material instead of 5-nitro-indazole
(Scheme 1) which were applied for amides 8 (Scheme 2).
BR
GK
Phe435
PBR
Glu436
O
HR
HN
Phe437
HN
O
N
N
NH
Met438
The amide series 8 have been evaluated while varying at the
same time the substituent R2 at position 6 of the indole moiety.
Acetylamide 8a exhibited an enzymatic IC50 value of 67 nM
(Table 1). This amounts to an about twofold activity increase
compared to 170 nM for the corresponding amino compound 12
wherein R2 represents also a methyl group (Scheme 2). As shown
in Table 1 small substituents are well tolerated (8b and 8c), larger
substituents such as a cyclohexyl moiety or a phenyl ring resulted
in a dramatic decrease of the enzymatic potency (8d and 8f).
However, compounds having a heteroatom in the para-position
of the ring, such as the oxygen of tetrahydropyran (8e) or the cy-
ano substituent (8g) performed better than the corresponding
compounds 8d and 8f. In general, the measured enzymatic po-
tency of compounds which were not substituted at position 6
HN
O
SP
O
SAR
HR = hinge region
GK = gatekeeper
BR = burried region
PBR = phosphate binding region
SP = sugar pocket
SAR = solvent accessible region
Figure 2. Proposed docked binding mode of indolylindazole derivatives 4a in ITK
identified after the second design cycle.8

1854
M. Herdemann et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1852–1856
R2
R2
(v)
NH
N
BOC
H2N
H2N
N
N
7
N
N
H
BOC
12
R2 = H, Me, OMe
(i), (v)
(iii), (v)
(ii), (v)
(iv) and (v) or (vi)
R2
R2
R2
R2
NH
NH
NH
NH
O
H
H
N
H
H
H
N
R3
R3
N
N
N
R3
S
N
N
N
N
R3
O
O
O
R4
N
N
N
N
H
H
H
H
11
8
9
10
Scheme 2. Reagents and conditions: (i) R3COCl, NEt3, DME, 0 °C to rt, 15 h; (ii) R3SO2Cl, 2,6-lutidine, DCM, 0 °C to rt, 15 h; (iii) R3NCO, DCM, 0 °C to rt, 15 h; (iv) R3R4CO,
NaBH(OAc)3, AcOH, DCM, rt, 18 h; (v) 4 M K2CO3aq, MeOH, THF, 60 °C, 3–4 h; (vi) TFA, DCM, 0 °C to rt, 4 h.
Table 1
Table 2
Amide substituents at positions 5 (8) and 6 (13) of the indazole
Sulfonamide and urea derivatives obtained through substitution at position 5 of the
indazole
Compd
R2
R3
IC50 enzymea
M)
IC50 cell,
(
l
IL-2a
(
lM)
Compd R2
R3
IC50 enzymea IC50 cell,
(lM)
IL-2a
(lM)
8a
8b
8c
8d
8e
8f
8g
8h
8i
8k
8l
8m
8n
8o
8p
8q
13a
13b
13c
13d
Me
Me
Me
Me
Me
Me
Me
Me
H
Me
0.067
0.12
0.065
0.99
0.26
2.30
0.83
0.53
0.14
0.16
1.60
1.50
0.087
0.098
0.90
1.00
0.22
0.34
>50
1.0
—
0.6
—
—
—
—
—
—
—
—
—
—
0.5
—
—
—
—
—
Methoxy-methyl
Dimethylamino-methyl
Cyclohexyl
Tetrahydro-2H-pyran-4-yl
Phenyl
4-Cyano-phenyl
Furan-2-yl
Me
9a
9b
9c
9d
9e
9f
9g
9h
9i
9k
9l
9m
10a
10b
10c
10d
10e
10f
Me Me
Me Phenyl
0.036
0.069
0.047
0.045
0.25
0.046
1.41
0.13
0.4
0.4
1.0
0.8
2.9
0.9
0.7
0.5
0.4
0.9
1.0
2.0
0.6
3.2
—
Me 2-Fluoro-phenyl
Me 3-Fluoro-phenyl
Me 4-Fluoro-phenyl
Me 3-Methoxy-phenyl
Me 4-Methoxy-phenyl
Me 3,5-Dimethyl-isoxazol-4-yl
Me 1-Methyl-1H-imidazol-4-yl
Me 1,2-Dimethyl-1H-imidazol-4-yl
Me 1,3,5-Trimethyl-1H-pyrazol-4-yl 0.085
Me 2,4-Dimethyl-thiazol-5-yl
Me Ethyl
Me Cyclohexyl
Me Phenyl
H
H
H
Methoxy-methyl
Cyclohexyl
Phenyl
0.026
0.043
MeO
MeO
MeO
MeO
Me
Me
Me
Me
Me
0.19
0.083
0.78
0.73
0.072
0.75
0.31
Methoxy-methyl
Cyclohexyl
Phenyl
Me
Me 3,5-Dimethyl-isoxazol-4-yl
Me Benzyl
Me Furan-2-yl-methyl
0.3
—
—
Methoxy-methyl
Cyclohexyl
Phenyl
>50
—
a
Means from 2 to 3 independent experiments.
a
Means from 2 to 3 independent experiments.
of the indole moiety (such as 8i–8m) was slightly reduced com-
pared to that of derivatives wherein R2 was either a methyl group
or a methoxy group (8n–8q). As predicted by docking studies,
small substituents were tolerated at position 6 of the indazole
(13a and 13b), but compounds having larger substituents (13c
and 13d) were inactive (Table 1).
para-substituted phenyl groups (9e and 9g). Interestingly, both
imidazole derivatives 9i and 9k showed promising enzymatic po-
tency. The cellular potency observed for 9i was identical to that
of 9a (0.4 lM). The activities within the urea series 10 were similar
to those measured for the amides 8 (Table 2). Small substituents
such as an ethyl group (10a) resulted in a slight increase of enzy-
matic potency. Larger residues such as cyclohexyl and phenyl led
to a significant loss of potency (10b–10c and 10e–10f). Both the
enzymatic and cellular potency of 10d were significantly increased
compared to 10c.
With respect to the variations introduced through reductive
amination at position 5 of the indazole ring, the attachment of
linear or cyclic aliphatic residues reduced the potency (Table 3
11a–11g) compared to 170 nM for the corresponding amino com-
pound 12 wherein R2 represents also a methyl group. However, the
In view of the results obtained for R2 = H, Me and OMe within
the amide series, further series were prepared and tested
exclusively with the Me substituent. The change from the amide
function to a sulfonamide resulted in a significant activity increase
of both enzymatic and cellular potency (Table 2). Methylsulfona-
mide 9a had an IC50 of 36 nM in the enzymatic assay and 0.4 lM
in the cellular assay. Due to the different molecular geometry of
sulfonamides compared to that of amides also larger residues such
as substituted phenyl groups are tolerated (9b–9d and 9f), except

M. Herdemann et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1852–1856
1855
Table 3
say (90 nM). This dramatic activity increase has been observed for
a whole series of basic heterocycles (11h–11k and 11m–11p). In
contrast, N-acetylated compound 11l was far less active than the
basic analogues. No significant difference in enzymatic potency
was observed for either the secondary amine 11h or the tertiary
amines 11i and 11k. The change from the 4-piperidyl ring to a
3-piperidyl ring led to a decrease of the enzymatic potency. 11m
inhibited ITK with an IC50 of 77 nM, tertiary amine 11n with an
IC50 of 66 nM. The enzymatic potencies observed for the five-
membered pyrrolidin compounds 11o (IC50 = 11 nM) and 11p
(IC50 = 30 nM) were in the same range as for the 4-piperidyl inhib-
itors. Moreover, with an IC50 of 20 nM in the cellular assay second-
ary amine 11o was the most potent indolylindazole ITK inhibitor
(Table 3). Ring opening of the pyrrolidine heterocycle resulted in
a minor decrease of the enzymatic ITK inhibition (11q). The signif-
icantly increased binding affinity to the ITK enzyme could be
understood in the light of performed docking studies. Figure 3
shows the binding mode of 11o within the ITK ATP binding pocket.
Main hydrogen-bond interactions were made with backbone
carbonyl oxygen of Glu-436 and backbone nitrogen and oxygen
of Met-438 in the hinge region. The affinity increase of 11o and
related basic compounds compared to parent compound 4a could
be explained with the additional interaction made with Asp-500
(Fig. 3). Unfortunately, crystal structures of 11o or its analogues
with the target protein could not be observed. However, the
proposed binding mode was in line with the obtained SAR informa-
tion for this lead series.
Amine substituents at the position 5 of the indazole
Compd R2
CHR3R4
IC50 enzymea IC50 cell,
(
lM)
IL-2a
(lM)
11a
11b
11c
11d
11e
11f
11g
11h
11i
Me Isopropyl
Me Isobutyl
Me Cyclobutyl
Me Cyclohexyl
Me 4-Phenyl-cyclohexyl
Me Cyclohexyl-methyl
Me Benzyl
0.72
0.86
0.62
1.20
—
—
—
—
12.0
1.70
—
—
0.53
—
Me Piperidin-4-yl
0.016
0.010
0.023
0.37
0.09
0.12
0.50
—
Me N-Methyl-piperidin-4-yl
Me N-Benzyl-piperidin-4-yl
Me N-Acetyl-piperidin-4-yl
Me Piperidin-3-yl
Me N-Ethyl-piperidin-3-yl
Me Pyrrolidin-3-yl
11k
11l
11m
11n
11o
11p
11q
0.077
0.066
0.08
0.03
0.027
—
—
0.02
0.35
—
Me N-Benzyl-pyrrolidin-3-yl
Me 1-Dimethylamino-propan-2-yl
a
Means from 2 to 3 independent experiments.
bioisosteric replacement of one CH2 of the cyclohexyl ring of 11d
by a nitrogen resulted in highly potent compounds. While 11d
showed an IC50 of 1.2 lM, 11h inhibited ITK in the enzymatic assay
with an IC50 of 16 nM and was also highly potent in the cellular as-
BR
Previous selectivity profiling of compound 4a in a panel of
about 30 kinases revealed equipotency against the receptor
tyrosine kinase Flt3.8 Therefore, four representative inhibitors
were selected for profiling in a large panel of 108 kinases.14 Sulfon-
amide 9k (Scheme 2 and Table 2) and the amines 11k and 11o
(Scheme 2 and Table 3) were tested at two different concentrations
(0.1 and 1.0 micromolar). The same screening was also performed
with an alkoxy derivative of 4a wherein the OCH3 substituent was
replaced by a bulkier OCH2–benzoyl residue at the indole’s position
6. The tests revealed that sulfonamide 9k and the alkoxy derivative
of 4a exhibited the best selectivity since only 10 further kinases
out of 108 were inhibited by both compounds at 0.1 micromolar
concentration to a similar extend as ITK. Basic compounds 11k
and 11o showed a medium selectivity inhibiting about 20 further
kinases in a similar range as ITK. All compounds inhibited Aurora
A, Flt3, JAK2, Ret and TrkA equipotently with regard to ITK. Even
though especially compound 9k showed a nicely improved profile,
GK
Phe435
Asp500
Glu436
O
HR
O
H
-
HN
H
N
N
N
O
Phe437
O
N
H
H
NH
+
Met438
HN
PBR
O
SP
3000
2000
SAR
1000
***
***
***
§
0
Figure 4. Level of IL-2 release 2 h after anti-CD3 antibody stimulation after
pretreatment with compound 11o. Data are mean SEM, n = 8 mice per group.
***p 6 0.001 (compared to vehicle; ANOVA followed by Dunnett’s test; § = below
detection limit).
Figure 3. Proposed docked binding mode of 11o in ITK ATP binding pocket.

1856
M. Herdemann et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1852–1856
231; (b) Siliciano, J. D.; Morrow, T. A.; Desiderio, S. V. Proc. Natl. Acad. Sci. U.S.A.
1992, 89, 11194.
2. Heyeck, S. D.; Wilcox, H. M.; Bunnell, S. C.; Berg, L. J. J. Biol. Chem. 1997, 272,
25401.
3. Liu, K. Q.; Bunnell, S. C.; Gurniak, C. B.; Berg, L. J. J. Exp. Med. 1998,
187, 1721.
4. (a) Schaeffer, E. M.; Debnath, J.; Yap, G.; McVicar, D.; Liao, X. C.; Littman, D. R.;
Sher, A.; Varmus, H. E.; Lenardo, M. J.; Schwartzberg, P. L. Science 1999, 284,
638; (b) Fowell, D. J.; Shinkai, K.; Liao, X. C.; Beebe, A. M.; Coffman, R. L.;
Littman, D. R.; Locksley, R. M. Immunity 1999, 11, 399.
the goal of an excellent selectivity profile has not yet been achieved
in this optimisation process.
After further in vitro profiling, compound 11o was selected for
in vivo evaluation in an anti-CD3 antibody-induced IL-2 mouse
model.13 The choice for 11o was made based on its excellent
in vitro potency of 11 nM in the enzymatic and 20 nM in the cellu-
lar assay, combined with acceptable cytotoxicity and favourable
physicochemical properties such as solubility and low plasma
protein binding (data not shown). Intravenous single administra-
tion of 11o resulted in dose-dependent IL-2 suppression (Fig. 4).
An intravenous dose of 10 mg/kg resulted in IL-2 concentration
levels below the detection limit. The ED50 of IL-2 inhibition was
calculated to be 2.1 mg/kg. Moreover, after po administration of
10 mg/kg of 11o the measured IL-2 inhibition was >80% compared
to the vehicle. Intravenous, as well as oral treatment with 11o was
well tolerated at all dose levels and no obvious side effects were
observed compared to the positive control group.
In conclusion, highly potent compounds have been identified
after optimisation within a novel class of ITK inhibitors. Lead
compound 11o has enzymatic and cellular activities in the low
nanomolar range. Dose-dependent, efficient suppression of anti-
CD3 antibody induced IL-2 release was observed in an in vivo
mouse model after intravenous administration. Furthermore, in
the same animal model, oral administration of 10 mg/kg of 11o
resulted in potent inhibition of IL-2 secretion which indicated good
oral bioavailability of the compound that was also well tolerated.
5. Miller, A. T.; Wilcox, H. M.; Lai, Z.; Berg, L. J. Immunity 2004, 21, 67.
6. Mueller, C.; August, A. J. Immunol. 2003, 170, 5056.
7. Lin, T. A.; McIntyre, K. W.; Das, J.; Liu, C.; O’Day, K. D.; Penhallow, B.; Hung, C.
Y.; Whitney, G. S.; Shuster, D. J.; Yang, X.; Townsend, R.; Postelnek, J.; Spergel, S.
H.; Lin, J.; Moquin, R. V.; Furch, J. A.; Kamath, A. V.; Zhang, H.; Marathe, P. H.;
Perez-Villar, J. J.; Doweyko, A.; Killar, L.; Dodd, J. H.; Barrish, J. C.; Wityak, J.;
Kanner, S. B. Biochemistry 2004, 43, 11056.
8. Herdemann, M.; Heit, I.; Bosch, F.-U.; Quintini, G.; Scheipers, C.; Weber, A.
Bioorg. Med. Chem. Lett. 2010, 20, 6998.
9. Brown, K.; Long, J. M.; Vial, S. C.; Dedi, N.; Dunster, N. J.; Renwick, S. B.; Tanner,
A. J.; Frantz, J. D.; Fleming, M. A.; Cheetham, G. M. J. Biol. Chem. 2004, 279,
18727.
10. (a) Gerber, P. R.; Muller, K. M. A. B. J. Comput. Aided Mol. Des. 1995, 9, 251; (b)
www.moloc.ch.
11. (a) Abdel-Magid, A. F.; Maryanoff, C. A.; Carson, K. G. Tetrahedron Lett. 1990, 31,
5595; (b) Abdel-Magid, A. F.; Maryanoff, C. A. Synlett 1990, 111, 537.
12. Preparation of 11o for in vivo testing: Intermediate 7 (R2 = Me; 2.80 mmol)
was dissolved in DCM (25 mL). Acetic acid (0.48 mL) and then 1-Boc-
pyrrolidin-3-one (4.20 mmol) were added. After stirring over 30 min at rt
NaBH(OAc)3 was added and the reaction mixture was stirred overnight.
After addition of H2O (20 mL) and saturated NaHCO3aq (5 mL) the organic
and aqueous layers were separated, the aqueous layer was extracted with
DCM (2 Â 20 mL). The combined organic phases were dried over MgSO4 and
the solvent was removed under reduced pressure. The residue was dissolved
in DCM (15 mL). At 0 °C TFA (10 mL) was added, the ice bath was removed
and the mixture was stirred for 4 h. The reaction mixture was then
concentrated under reduced pressure. Remaining TFA was removed by
addition of toluene (5 mL) and followed by concentration. Subsequently,
DCM (40 mL) and saturated NaHCO3aq (40 mL) were added. After separation
of the organic and aqueous layers the aqueous layer (pH 10) was extracted
with DCM (2 Â 40 mL) and the combined organic phases were dried over
MgSO4. Removal of the solvent under reduced pressure gave the desired
crude product. Purification of 11o was performed by flash chromatgraphy.
After a first chromatography on basic silica gel with DCM/MeOH (95:5) the
product (purity 90%) was submitted to a second flash chromatography with
cyclohexane/AcOEt/MeOH (60:20:20) as eluent to provide 216 mg (23%
overall yield) of pure 11o.
Acknowledgements
We thank J. Wilsberg for skillful synthetic support, M. Beschle
and K.-M. Nagy for LC–MS purifications, Dr. H.-C. Holst and Mr.
T. Kottysch for NMR analyses, G. Prehn, E. Rolser, G. Schaffrath,
J. Schindler and Dr. T. Ciossek for biological in vitro testing,
J. Wegmann, A. Hirrle and A. Haschke for the testing in vivo. We
gratefully thank Professor Dr. G. Baier (University of Innsbruck)
for scientific advice.
13. Anti-CD3 IL-2 mouse model: Female BALB/c mice (ꢀ20 g body weight) received
either one of the three doses of 11o (1, 3, 10 mg/kg body weight) intravenously
10 min or orally (10 mg/kg) 1 h before intravenous injection of a monoclonal
Supplementary data
anti-CD3 stimulating antibody (10 lg/kg body weight, BD PharMingen, clone:
145-2C11, Cat. No.01080D). The positive control group received vehicle
(saline) intravenously and the antibody at the same dose for an equivalent
length of time. Two hours after stimulation mice were finally anaesthetized
and blood samples were taken by heart puncture. Suppression of antibody
induced stimulation was described by the determination of IL-2 levels in
plasma samples. Cytokine concentrations were measured by ELISA (R&D
Systems, Quantikine mouse IL-2 ELISA Kit, Cat. No. M2000).
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.01.035.
References and notes
1. (a) Yamada, N.; Kawakami, Y.; Kimura, H.; Fukamachi, H.; Baier, G.; Altman, A.;
14. Supplementary data.
Kato, T.; Inagaki, Y.; Kawakami, T. Biochem. Biophys. Res. Commun. 1993, 192,