Design and synthesis of tricyclic jak3 inhibitors with picomolar affinities as novel molecular probes

Article abstract of DOI:10.1002/cmdc.201300520

The Janus kinase (JAK) signaling pathway is of particular importance in the pathology of inflammatory diseases and oncological disorders, and the inhibition of Janus kinase 3 (JAK3) with small molecules has proven to provide therapeutic immunosuppression. A novel class of tricyclic JAK inhibitors derived from the 3-methyl-1,6-dihydrodipyrrolo[2,3-b:2',3'-d]pyridine scaffold was designed based on the tofacitinib-JAK3 crystal structure by applying a rigidization approach. A convenient synthetic strategy to access the scaffold via an intramolecular Heck reaction was developed, and a small library of inhibitors was prepared and characterized using in vitro biochemical as well as cellular assays. IC50 values as low as 220 pm could be achieved with selectivity for JAK3 over other JAK family members. Both activity and selectivity were confirmed in a cellular STAT phosphorylation assay, providing also first-time data for tofacitinib. Our novel inhibitors may serve as tool compounds and useful probes to explore the role of JAK3 inhibition in pharmacodynamics studies.

Full text of DOI:10.1002/cmdc.201300520

CHEMMEDCHEM
COMMUNICATIONS
DOI: 10.1002/cmdc.201300520
Design and Synthesis of Tricyclic JAK3 Inhibitors with
Picomolar Affinities as Novel Molecular Probes
Matthias Gehringer, Ellen Pfaffenrot, Silke Bauer, and Stefan A. Laufer*[a]
The Janus kinase (JAK) signaling pathway is of particular im-
portance in the pathology of inflammatory diseases and onco-
logical disorders, and the inhibition of Janus kinase 3 (JAK3)
with small molecules has proven to provide therapeutic immu-
nosuppression. A novel class of tricyclic JAK inhibitors derived
from the 3-methyl-1,6-dihydrodipyrrolo[2,3-b:2’,3’-d]pyridine
scaffold was designed based on the tofacitinib–JAK3 crystal
structure by applying a rigidization approach. A convenient
synthetic strategy to access the scaffold via an intramolecular
Heck reaction was developed, and a small library of inhibitors
was prepared and characterized using in vitro biochemical as
well as cellular assays. IC50 values as low as 220 pm could be
achieved with selectivity for JAK3 over other JAK family mem-
bers. Both activity and selectivity were confirmed in a cellular
STAT phosphorylation assay, providing also first-time data for
tofacitinib. Our novel inhibitors may serve as tool compounds
and useful probes to explore the role of JAK3 inhibition in
pharmacodynamics studies.
fects. Pfizer’s pan-JAK inhibitor tofacitinib (1; Figure 1), which
was recently approved by the US Food and Drug Administra-
tion (FDA), had initially been reported as a selective inhibitor
of JAK3.[4] Later it was shown to be poorly selective against
JAK1 and JAK2.[5] It proved to be effective in clinical trials for
the treatment of rheumatoid arthritis (RA), psoriasis, Crohn’s
disease, kidney transplant rejection, and ulcerative colitis, but
side effects remained a serious concern.[6] Thus, the European
Medicines Agency (EMA) voted against its approval in Europe
in April 2013.
While several inhibitors with selectivity towards JAK1 and/or
JAK2 are currently in advanced clinical trials, there is only one
inhibitor currently under clinical investigation claiming selectiv-
ity for JAK3—VX-509 from Vertex Pharmaceuticals.[7] However,
little data justifying its selectivity has been published, and its
structure has only recently been disclosed. Therefore, the cur-
rent gold standard as a chemical probe for JAK3 is still tofaciti-
nib (1) despite its poor selectivity profile within the JAK family.
Recently, the urgent need for high quality probes was high-
JAK3 is an intracellular non-re-
ceptor tyrosine kinase belonging
to the Janus kinase (JAK) family.
The four JAK family members
JAK1, JAK2, JAK3 and TYK2 (tyro-
sine kinase 2) are involved in
various signaling cascades initi-
ated by cytokines and growth
Figure 1. Structures of tofacitinib (1), also known as CP-690550 and marketed under the trade name Xeljanz, a tri-
cyclic JAK inhibitor (2) that was recently published by Roche,[5] and the compound developed in this work (3).
factors mediating immune re-
sponse and proliferation. In con-
trast to the ubiquitous expres-
sion of other JAKs, JAK3 is solely
expressed in hematopoietic tissue restricting its function in cy-
tokine signaling. JAK3 plays a key role in the development of
lymphocytes, such as Tcells, B cells and NK cells. It associates
exclusively with type 1 cytokine receptors employing the
common gamma chain (gc).[1] Receptor activation induces
JAK3/JAK1 heterodimerization and autophosphorylation. The
activated JAK complex phosphorylates signal transducers and
activators of transcription (STAT) proteins, which translocate to
the nucleus and induce gene expression.[2,3] JAK3 is an attrac-
tive target to provide immunosuppression with limited side ef-
lighted by Knapp et al.,[8] underlining the demand for well-
characterized compounds with a distinct selectivity profile.
Consequently, there is an ongoing interest in novel JAK inhibi-
tors to link subtype selectivity to cellular STAT phosphorylation
profiles, and in turn, those profiles to clinical outcome.
The structure–activity relationships (SARs) of the reference
molecule tofacitinib are very narrow. Of the four stereoisomers,
only one is highly active leaving severely restricted possibilities
for modification. Inspired by the crystal structure of the JAK3
kinase domain in complex with 1 (PDB code: 3LXK[2]), we de-
veloped the idea of tricyclic JAK inhibitors. Tofacitinib binds to
the ATP pocket of JAK3 forming two hydrogen bonds towards
the hinge region via the pyrrolo[2,3-d]pyrimidine core scaffold.
As shown in the SAR studies from Pfizer, the chiral methyl
group on the piperidine side chain of tofacitinib plays a crucial
role for potency and kinome selectivity, binding to a small hy-
drophobic pocket at the lower rim of the binding cavity of
[a] M. Gehringer, E. Pfaffenrot, S. Bauer, Prof. Dr. S. A. Laufer
Department of Pharmaceutical & Medicinal Chemistry
Institute of Pharmacy, Eberhard-Karls-University Tuebingen
Auf der Morgenstelle 8, 72076 Tuebingen (Germany)
E-mail: Stefan.Laufer@uni-tuebingen.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201300520.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 277 – 281 277

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
JAK3.[2,4,9] However, the unbound form of tofacitinib allows for
a rotational equilibrium that can shift the side chain 1808 from
the active conformation.
could undergo additional favorable lipophilic interactions with
the hydrophobic front region of JAK3. Encouraged by these
theoretical findings, we developed a novel convergent method
to access 3-methyl-substituted 1,6-dihydrodipyrrolo[2,3-b:2’,3’-
d]pyridines via an intramolecular Heck reaction (Scheme 1).
Racemic amine 4 and heterocyclic precursor 5 were pre-
pared by modified literature procedures (for details, see the
Supporting Information). Nucleophilic aromatic substitution
(SNAr) of the chloride in 5 with amine 4 failed in organic sol-
vents under forcing conditions (e.g., iso-propanol or dioxane,
120–2008C, microwave) using various bases. An alternative
method using a superheated biphasic system with water in
the presence of potassium carbonate also did not lead to the
desired product. Palladium-catalyzed arylamination of 5 as well
as of a nonbrominated precursor applying various catalyst sys-
tems resulted in poor conversion and sluggish reactions. To
our delight, 6 was obtained by carefully melting the halo-het-
erocycle with a slight excess of 4 at 1808C (Scheme 1). Nucleo-
philic substitution of 6 on allyl bromide gave 7 as the final pre-
cursor for Heck coupling. The intramolecular key cyclization
could be accomplished with tetrakis(triphenylphosphine)palla-
dium(0) in N,N-dimethylformamide, using cesium carbonate as
a base. Despite the relatively electron-rich double bond that
generally disfavors Heck reactions, conversion proceeded
smoothly without any N-deallylation under the optimized reac-
tion conditions. In accordance with Baldwin’s rules[17] and in
line with what is commonly observed in intramolecular Heck
transformations, the expected 5-exo-trig cyclization was fa-
vored over the alternative 6-endo-trig pathway. The obtained
crude product carried an exocyclic double bond, and no iso-
merization occurred during the reaction as indicated by the
vinyl methylene group of 8-exo detected as a negative peak at
a d value of 98.1 ppm in the DEPT-135 NMR spectrum. We
found that silica gel catalyzed the isomerization to more ther-
modynamically stable compound 8, while the product bearing
Rigidization is a well-established concept in drug discovery
that our group has previously exploited to improve the poten-
cy and selectivity of p38a MAP kinase inhibitors based on the
pyridinylimidazole and dibenzosuberone scaffold.[10,11] Starting
from our results, researchers from Merck further developed the
approach to obtain selective inhibitors of the c-Met kinase.[12]
We therefore undertook the application of this strategy to
“freeze” the Pfizer compound in its active conformation
(Figure 1). Furthermore, tofacitinib does not address the hydro-
phobic front region of JAK3. We hypothesized that addressing
this region could further enhance the affinity and selectivity of
the compound. Moreover, this hydrophobic cleft is opened to-
wards the bulk solvent, and inhibitor molecules targeting this
region are suited to the attachment of various tags like fluoro-
phores, covalent linkers, or affinity labels, as well as solubilizing
groups.
Our initial attempt was directed towards the synthesis of
1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridines similar to
2
(Figure 1) bearing the tofacitinib side chain.[13] However, stud-
ies from Zak et al. published during the course of this work re-
vealed that this compound class generally shifts the selectivity
towards JAK1.[5,14,15] Nevertheless, this scaffold did not allow
for substitution on the imidazole N3, which we predicted by
molecular modeling to favor JAK3 binding. Thus, we designed
3, a novel tricyclic inhibitor based on a 1,6-dihydrodipyrro-
lo[2,3-b:2’,3’-d]pyridine core structure that allowed for substitu-
tion at the 3-position of the heterocycle.
Docking compound 3 into the JAK3 ATP binding pocket
(PDB code: 3LXK[2]), and analysis of the predicted pharmaco-
phoric interaction pattern using LigandScout[16] revealed a tofa-
citinib-like binding mode (Figure 2). The docking poses sug-
gest that the exocyclic methyl group of the annulated pyrrole
Figure 2. a) Compound 3 docked into the crystal structure of JAK3 (PDB code: 3LXK[2]) using Schrçdinger Glide. b) Overlay of the binding pharmacophores of
tofacitinib bound to JAK3 and the docked conformation of 3. Hydrogen bonds are depicted as red/green arrows, and lipophilic interactions as yellow spheres.
Both graphics were generated using Inte:Ligand LigandScout.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 277 – 281 278

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Scheme 1. Preparation of the tricyclic compounds. Reagents and conditions: a) 1808C, 2.5 h, 63% b) NaH, DMF, 08C, 1 h, then allyl bromide, 08C, 3 h, 73%;
c) Pd(PPh3)4, Cs2CO3, DMF, 758C, 30 min; d) SiO2, 608C, 79%; e) Cs2CO3, THF/MeOH (2:1), 648C, 1 h, 93%; f) H2, Pearlman’s catalyst, MeOH, RT, o/n, 89%;
g) DCC, carboxylic acid, CH2Cl2, RT, 2.5 h, 59–96%; h) acryl nitrile, EtOH, RT, o/n, 96%; i) oxetan-3-ylidene-acetonitrile, EtOH, 708C, 7 d, 54%; j) 2,2,2-trifluoro-
ethanesulfonyl chloride, pyridine, CH2Cl2, À208C, 30 min, then RT, 2 h, 71%.
the exocyclic double bond could be purified by reverse-phase
flash chromatography under neutral conditions. Intermediate 8
was detosylated, and the benzyl group of 9 was subsequently
removed by hydrogenation employing Pearlman’s catalyst.
From 10, a small library of derivatives (3–3h) was prepared
using standard amide coupling or aza-Michael addition. We
were not able to establish a reliable method to separate the
enantiomers of the products on a preparative scale. Thus, all
biological profiling was performed using the racemic products.
To explore the SARs, we performed a preliminary evaluation
of the compounds using our JAK3 ELISA system (see Support-
ing Information). The racemic parent compound 3 with an IC50
value of 18 nm exhibited excellent activity in a similar range as
tofacitinib (7.6 nm), while the free amine 10 was more than
1000 times less potent. Varying the amide substituent led in all
cases to slightly less potent inhibitors (Table 1). Introduction of
a more rigid 2-cyclopropyl cyanoacetamide residue (3a) mar-
ginally decreased potency, while substituents not bearing the
nitrile function generally led to a more pronounced loss in po-
tency, emphasizing the particularly favorable interactions be-
tween the nitrile function and the glycine-rich loop.[2,18] Mim-
icking the nitrile with a negatively polarized trifluoromethyl
group (3c) resulted in a sixfold loss in potency compared with
compound 3, whereas small alkylamide substituents like meth-
ylcyclopropylamide (3d) or propionylamide (3e) proved to be
less favorable in this position. Introduction of a bulkier cyclo-
pentylamide residue (3b) resulted in an IC50 value of 230 nm,
which is in the same potency range as the two aforemen-
tioned inhibitors (compounds 3d,e). Larger substituents were
not probed due to in-house SAR data from similar scaffolds in-
dicating that excessive steric bulk at this position generally di-
minished potency (data not shown). Replacement of the amide
by a sulfonamide bearing a 2,2,2-trifluoroethyl residue (3h) re-
sulted in a twofold loss in potency relative to the correspond-
ing carboxamide (3c).
For probing the influence of conformational flexibility of the
amino substituent, the amide group of 3 was replaced by
a sp3-hybridized carbon atom. Cyanoethyl derivative 3g was
far less potent than the more rigid parent compound. We hy-
pothesize that the drop in potency results from different con-
formational preferences, as well as the lack of preorganization.
To decrease the conformational flexibility, an oxetane residue
was introduced (3 f) to mimic the amide carbonyl.[19] Contrary
to our expectations, this modification proved to be even more
unfavorable, probably because of an altered minimum energy
geometry. Finally, benzylated precursor 9 showed poor inhibi-
tory activity in the low micromolar range.
The most potent compounds were further tested for their
selectivity within the JAK family using a commercial radiomet-
ric assay.[20] JAK3 inhibition observed in this assay qualitatively
correlated with data from our ELISA, but the compounds were
significantly more potent (15- to 100-fold). This discrepancy
can primarily be attributed to the much lower enzyme concen-
tration used in the radiometric filtration binding assay, as well
as to the slightly lower ATP concentration used for testing,
compared with our ELISA method which was optimized to find
highly actives only.
Data obtained from the commercial assay indicated that all
compounds possessed moderate selectivity for JAK3 (Table 2).
Cyanoacetamide 3 proved to be the most potent compound
with an IC50 value of 220 pm. In almost all cases, the best selec-
tivity was observed when compared with TYK2 (12- to 30-fold),
while the least selectivity was obtained against JAK2 (2- to 10-
fold). The most selective compound was 3a, being 23-, 10-
and 18-fold selective against JAK1, JAK2, and TYK2, respective-
ly. Compound 3 demonstrated increased selectivity against
TYK2 (30-fold), while selectivity against JAK1 slightly decreased
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 277 – 281 279

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Table 2. Biological activities of selected compounds against all Janus
Table 1. JAK3 inhibition by the currently investigated JAK inhibitors.
kinase family members.
Compd
IC50 [nm][a]
JAK1
JAK2
JAK3
TYK2
3
3.38
44.7
98.9
50.7
56.1
2.07
19.0
35.0
14.5
26.1
0.22
1.97
15.5
4.60
7.31
6.67
35.3
246
78.8
90.2
3a
3b
3c
3h
Compd
Structure/side chain (R)
IC50 [nm][a]
[a] IC50 values were calculated from the results of a radiometric assay (Km
ATP, 10-dose IC50 with 3-fold serial dilution starting at 10 mm). Data was
obtained from a single determination according to the suppliers stand-
ards.
(R,R)-1
7.57Æ0.59
18.27Æ0.55
3
compounds are potent inhibitors of intracellular STAT phos-
phorylation, and the difference in potency is less pronounced
compared with the raw enzyme data. Tofacitinib inhibits STAT
phosphorylation in the same potency range as our inhibitors
(Table 3). Compared with tofacitinib, compound 3 shows a sig-
nificant shift in inhibitory potency towards STAT5/6 with IC50
values below 3 nm, which was the lowest concentration
tested.
(0.22)
42.6Æ1.53
3a
(1.97)
229Æ13.6
3b
(15.5)
108Æ3.8
3c
(4.60)
Table 3. Inhibition of STAT phosphorylation in HeLa cells by compounds
3, 3a and tofacitinib.
3 d
3e
214Æ17.9
374Æ42.3
Compd
IC50 [nm][a]
STAT3[
STAT1
STAT2
STAT5
STAT6
Tofacitinib
3
3a
25
37
170
170
250
560
13
19
99
9.3
<3.0
<3.0
25
<3.0
34
3 f
672Æ42.3
489Æ38.2
3g
3h
[a] IC50 values for inhibition of STAT phosphorylation after simulation with
IFNa.[21] Data was obtained from a single determination according to the
suppliers standards.
230Æ16.6
(7.31)
9
7664Æ810
5766Æ479
Regarding the physicochemical properties, compound 3 is
well within the Lipinski’s rule of five (calculated logP=2.08;
PSA=77.7 ꢁ2) and slightly more lipophilic compared with tofa-
citinib (calculated logP=1.24; PSA=88.9 ꢁ2) and 2 (calculated
logP=0.76; PSA=90.6 ꢁ2). Compound 3 possesses good solu-
bility in pH 7.4 buffer (>200 mgLÀ1) as determined by HPLC
with UV detection (see Supporting Information).
10
H
[a] IC50 values were calculated from the results of an ELISA. Data repre-
sent the mean ÆSEM of three determinations performed in triplicate.
Values in parentheses are the IC50 values determined using a radiometric
assay (Km ATP, 10-dose IC50 with 3-fold serial dilution starting at 10 mm),
and represent single determinations. All compounds except 1 were
tested as racemic mixtures.
In the current study, we have developed a novel series of tri-
cyclic JAK inhibitors. We applied a “frozen conformer” ap-
proach to the approved drug tofacitinib, shifting its selectivity
towards JAK3 while maintaining potency. A convenient syn-
thetic strategy towards the substituted methyl-1,6-dihydrodi-
pyrrolo[2,3-b:2’,3’-d]pyridine scaffold that employs an intramo-
lecular Heck reaction as the key step for building up the tricy-
clic core was developed. Our methodology could be further
exploited to introduce a variety of substituents to the 3-posi-
tion of the heterocyclic scaffold, including tags or groups that
further improve solubility and selectivity. With an IC50 value of
220 pm, compound 3 is among the most potent JAK3 inhibi-
tors published. It is worth noting that to date, only a few
(15-fold selective). Thus both compounds 3 and 3a offer a su-
perior in vitro selectivity profile compared with tofacitinib. For
the other compounds tested, selectivity, especially against
JAK2, was less pronounced.
Inhibitors 3 and 3a were probed for their capability to inhib-
it STAT phosphorylation in stimulated HeLa cells using a com-
mercial luminex-based multiplex immunoassay.[21] The advant-
age of this type of cell assay is the possibility to assess a STAT
phosphorylation profile of all STATs (excluding STAT4) in one
single cell type. Furthermore, the performance in such an
assay has not yet been described for tofacitinib. Both of our
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 277 – 281 280

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
JAK3-selective compounds have been reported,[7] most of
them being less selective then the examples presented here.
Furthermore, our best inhibitors display druglike properties
and good solubility. Compounds 3 and 3a proved to be highly
active in a cellular STAT phosphorylation assay. The inhibition
of STAT phosphorylation in HeLa cells was significantly shifted
towards STAT5 and STAT6 compared with tofacitinib, whose
profile in this assay is also presented for the first time here.
The results confirm a different intracellular selectivity profile,
making 3 and 3a valuable probes to investigate the role of
JAK3 inhibition in pharmacodynamics studies.
C. A. Hurley, A. Johnson, T. Johnson, J. R. Kenny, P. B. Kohli, R. J. Maxey,
R. Mendonca, K. Mortara, J. Murray, R. Narukulla, S. Shia, M. Steffek, S.
Ubhayakar, M. Ultsch, A. van Abbema, S. I. Ward, B. Waszkowycz, M. Zak,
J. Med. Chem. 2012, 55, 5901–5921.
[6] C. A. Zerbini, A. B. Lomonte, Expert Rev. Clin. Immunol. 2012, 8, 319–
331.
[7] B. W. Dymock, C. S. See, Expert Opin. Ther. Pat. 2013, 23, 449–501.
[8] S. Knapp, P. Arruda, J. Blagg, S. Burley, D. H. Drewry, A. Edwards, D.
Fabbro, P. Gillespie, N. S. Gray, B. Kuster, K E. Lackey, P. Mazzafera,
N. C. O. Tomkinson, T. M. Willson, P. Workman, W. J. Zuercher, Nat. Chem.
Biol. 2013, 9, 3–6.
[9] J. K. Jiang, K. Ghoreschi, F. Deflorian, Z. Chen, M. Perreira, M. Pesu, J.
Smith, D. T. Nguyen, E. H. Liu, W. Leister, S. Costanzi, J. J. O’Shea, C. J.
Thomas, J. Med. Chem. 2008, 51, 8012–8018.
[10] R. Selig, M. Goettert, V. Schattel, D. Schollmeyer, W. Albrech, S. Laufer, J.
Med. Chem. 2012, 55, 8429–8439.
[11] S. A. Laufer, G. M. Ahrens, S. C. Karcher, J. S. Hering, R. Niess, J. Med.
Chem. 2006, 49, 7912–7915.
[12] K. W. Rickert, S. B. Patel, T. J. Allison, N. J. Byrne, P. L. Darke, R. E. Ford,
D. J. Guerin, D. L. Hall, M. Kornienko, J. Lu, S. K. Munshi, J. C. Reid, J. M.
Shipman, E. F. Stanton, K. J. Wilson, J. R. Young, S. M. Soisson, K. J.
Lumb, J. Biol. Chem. 2011, 286, 11218–11225.
Experimental Section
Experimental procedures and characterization of all compounds
are provided in the Supporting Information, available at http://
dx.doi.org/10.1002/cmdc.201300520.
[13] E. Pfaffenrot, D. Schollmeyer, S. Laufer, Acta Crystallogr. Sect. E 2012, 68,
o3052.
Acknowledgements
[14] M. Zak, R. Mendonca, M. Balazs, K. Barrett, P. Bergeron, W. S. Blair, C.
Chang, G. Deshmukh, J. DeVoss, P. S. Dragovich, C. Eigenbrot, N. Ghilar-
di, P. Gibbons, S. Gradl, C. Hamman, E. J. Hanan, E. Harstad, P. R. Hewitt,
C. A. Hurley, T. Jin, A. Johnson, T. Johnson, J. R. Kenny, M. F. T. Koehler,
P. B. Kohli, J. J. Kulagowski, S. Labadie, J. P. Liao, M. Liimatta, Z. H. Lin,
P. J. Lupardus, R. J. Maxey, J. M. Murray, R. Pulk, M. Rodriguez, S. Savage,
S. Shia, M. Steffek, S. Ubhayakar, M. Ultsch, A. van Abbema, S. I. Ward, L.
Xiao, Y. S. Xiao, J. Med. Chem. 2012, 55, 6176–6193.
The authors are grateful to Inte:Ligand Software-Entwicklungs
und Consulting GmbH (Maria Enzersdorf, Austria) for providing
the LigandScout software. The authors thank Daniela Mꢀller for
biological testing, and Michael Forster and Peter Keck for assis-
tance in editing the manuscript and fruitful scientific discussion.
[15] M. Zak, C. A. Hurley, S. I. Ward, P. Bergeron, K. Barrett, M. Balazs, W. S.
Blair, R. Bull, P. Chakravarty, C. Chang, P. Crackett, G. Deshmukh, J.
DeVoss, P. S. Dragovich, C. Eigenbrot, C. Ellwood, S. Gaines, N. Ghilardi,
P. Gibbons, S. Gradl et al., J. Med. Chem. 2013, 56, 4764–4785.
[16] G. Wolber, T. Langer, J. Chem. Inf. Model. 2005, 45, 160–169.
[17] J. E. Baldwin, J. Chem. Soc. Chem. Commun. 1976, 734–736.
[18] N. K. Williams, R. S. Bamert, O. Patel, C. Wang, P. M. Walden, A. F. Wilks,
E. Fantino, J. Rossjohn, I. S. Lucet, J. Mol. Biol. 2009, 387, 219–232.
[19] G. Wuitschik, E. M. Carreira, B. Wagner, H. Fischer, I. Parrilla, F. Schuler,
M. Rogers-Evans, K. Muller, J. Med. Chem. 2010, 53, 3227–3246.
[20] Kinase HotSpot assay, Reaction Biology Corp. (Malvern, PA, USA); http://
www.reactionbiology.com.
Keywords: inflammation · isoform selectivity · janus kinase 3 ·
medicinal chemistry · tricyclic kinase inhibitors
[1] M. G. Cornejo, T. J. Boggon, T. Mercher, Int. J. Biochem. Cell Biol. 2009,
41, 2376–2379.
[2] J. E. Chrencik, A. Patny, I. K. Leung, B. Korniski, T. L. Emmons, T. Hall, R. A.
Weinberg, J. A. Gormley, J. M. Williams, J. E. Day, J. L. Hirsch, J. R. Kiefer,
J. W. Leone, H. D. Fischer, C. D. Sommers, H.-C. Huang, E. J. Jacobsen,
R. E. Tenbrink, A. G. Tomasselli, T. E. Benson, J. Mol. Biol. 2010, 400, 413–
433.
[3] J. J. O’Shea, Ann. Rheum. Dis. 2004, 63 (Suppl. 2), ii67–ii71.
[4] M. E. Flanagan, T. A. Blumenkopf, W. H. Brissette, M. F. Brown, J. M. Casa-
vant, C. Shang-Poa, J. L. Doty, E. A. Elliott, M. B. Fisher, M. Hines, C. Kent,
E. M. Kudlacz, B. M. Lillie, K. S. Magnuson, S. P. McCurdy, M. J. Munchhof,
B. D. Perry, P. S. Sawyer, T. J. Strelevitz, C. Subramanyam, J. M. Sun, D. A.
Whipple, P. S. Changelian, J. Med. Chem. 2010, 53, 8468–8484.
[5] J. J. Kulagowski, W. Bair, R. J. Bull, C. Chang, G. Deshmukh, H. J. Dyke, C.
Eigenbrot, N. Ghilardi, P. Gibbons, T. K. Harrison, P. R. Hewitt, M. Liimatta,
[21] Milliplex MAP STAT luminex-based multiplex immunoassay, EMD Milli-
pore Corp. (Billerica, MA, USA); http://www.millipore.com/LeadDiscov-
ery.
Received: December 11, 2013
Published online on January 8, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 277 – 281 281