2-Aminopyrazolo[1,5-a]pyrimidines as potent and selective inhibitors of JAK2

Article abstract of DOI:10.1016/j.bmcl.2009.10.053

Constitutive activation of the EPO/JAK2 signaling cascade has recently been implicated in a variety of myeloproliferative disorders including polycythemia vera, essential thrombocythemia and myelofibrosis. In an effort to uncover therapeutic potential of blocking the EPO/JAK2 signaling cascade, we sought to discover selective inhibitors that block the kinase activity of JAK2. Herein, we describe the discovery and structure based optimization of a novel series of 2-amino-pyrazolo[1,5-a]pyrimidines that exhibit potent inhibition of JAK2.

Full text of DOI:10.1016/j.bmcl.2009.10.053

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6529–6533
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
2-Aminopyrazolo[1,5-a]pyrimidines as potent and selective inhibitors of JAK2
*
*
Mark W. Ledeboer , Albert C. Pierce , John P. Duffy, Huai Gao, David Messersmith, Francesco G. Salituro ,
Suganthini Nanthakumar, Jon Come, Harmon J. Zuccola, Lora Swenson, Dina Shlyakter, Sudipta Mahajan,
Thomas Hoock, Bin Fan à, Wan-Jung Tsai, Elaine Kolaczkowski, Scott Carrier, James K. Hogan, Richard Zessis,
S. Pazhanisamy, Youssef L. Bennani
Vertex Pharmaceuticals Inc., Cambridge, MA 02139-4242, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Constitutive activation of the EPO/JAK2 signaling cascade has recently been implicated in a variety of
myeloproliferative disorders including polycythemia vera, essential thrombocythemia and myelofibrosis.
In an effort to uncover therapeutic potential of blocking the EPO/JAK2 signaling cascade, we sought to dis-
cover selective inhibitors that block the kinase activity of JAK2. Herein, we describe the discovery and
structure based optimization of a novel series of 2-amino-pyrazolo[1,5-a]pyrimidines that exhibit potent
inhibition of JAK2.
Received 14 August 2009
Revised 8 October 2009
Accepted 12 October 2009
Available online 24 October 2009
Keywords:
Janus kinase 2
Ó 2009 Elsevier Ltd. All rights reserved.
JAK2
Structure based design
2-Amino-pyrazolo[1,5-a]pyrimidines
Polycythemia vera
Kinase inhibitor
Myeloproliferative disorders (MPDs) are a collection of hemato-
logical neoplasms including chronic myelogenous leukemia (CML),
polycythemia vera (PV), essential thrombocythemia (ET), and mye-
lofibrosis, all of which are characterized by the overproduction of
one or more cell types belonging to the myeloid lineage.1 With
the exception of CML, which was shown to be a direct result of
activation of the tyrosine kinase ‘abl’ via a reciprocal translocation
between chromosomes 9 and 22 producing the oncogene bcr-abl,
the cause of the other disorders was poorly understood until re-
cently. JAK2 is a member of the Janus kinase (JAK) family of non-
receptor protein tyrosine kinases, known to play a crucial role in
cytokine mediated signal transduction pathways that also include
JAK1, JAK3 and TYK2.2 In early 2005, several groups independently
discovered a single point mutation in JAK2 (V617F) common in pa-
tients diagnosed with MPDs other than CML. While the mechanism
by which the V617F mutation leads to myeloproliferative disorders
is not fully elucidated, it is apparent that the mutation results in a
gain of function in the JAK2 kinase resulting in pro-proliferative ef-
fects via the JAK2 signaling cascades. Currently, this remains an
area of intense research to understand the exact mechanism by
which the JAK2 kinase is activated via the valine to phenylalanine
substitution at position 617.3,4 In a program directed at discovering
the therapeutic potential of attenuating the EPO/JAK2 signaling
pathway we sought to identify compounds that would block the ki-
nase activity of JAK2 by targeting the ATP binding site.5
Initial screening of our corporate compound collection identi-
fied a small set of submicromolar inhibitors of JAK2 containing
an aminopyrazolopyrimidine core (APP) (9–18, Table 1). Interest-
ingly, compound 12 demonstrated a high affinity for JAK2 and
exhibited 40-fold isotype selectivity versus JAK3. To better under-
stand the binding of these compounds, we initiated protein–ligand
X-ray crystallographic studies of 12 bound to the kinase domain of
JAK2. The co-complex revealed several key binding interactions
responsible for binding this ATP active site.6
Compound 12 forms two hinge hydrogen bonds with its amin-
opyrazolopyrimidine (APP) core (Fig. 1). The hydrogen bonds are to
the backbone NH and carbonyl of Leu-932. An intramolecular
hydrogen bond fixes the pyrimidine conformation, holding it
coplanar with the core ring system while directing its amine sub-
stituent deep into the active site.
The piperidine ring forms hydrophobic contacts with the gly-
cine-rich loop above it and places the amide substituent against
Gly 993 and Asp 994 at the beginning of the activation loop of
JAK2. It is these latter interactions that confer selectivity. The
* Corresponding authors. Tel.: +1 617 444 6100.
E-mail addresses: mark_ledeboer@vrtx.com (M.W. Ledeboer), al_pierce@vrtx.-
com (A.C. Pierce).
Present address: Agios Pharmaceuticals, 38 Sidney Street, Cambridge, MA 02139,
United States.
à
Present address: Nektar Therapeutics, 201 Industrial Road, San Carlos, CA 94070,
United States.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.10.053

6530
M. W. Ledeboer et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6529–6533
Table 1
Activity of APP analogs
R1
N
N
N
R2
H2N
N
N
Compd
R1
H
R2
JAK2 KI (
lM)
JAK3 KI (lM)
9
HN
0.020
0.21
10
11
H
H
N
0.044
0.084
0.46
0.50
N
Figure 1. Crystal structure of Compound 12 bound to JAK2. Dashed lines represent
hydrogen bonds, both intramolecular and to Leu 932 of JAK2. The Ala and Asp
residues of JAK2 that correspond to Gly 993 and Asp 994 are shown in green.
NH2
O
The synthetic approach for this series is outlined in Scheme 1.
Treatment of 4,6-dichloropyrimidine with NaSMe generated the
corresponding 4-chloro-6-thiomethylpyrimidine,8 which was
added to the sodium salt of malononitrile in a Pd-mediated cou-
pling to provide 2.9 Addition of hydrazine hydrate generated diam-
inopyrazole 3 which could be readily cyclized with either 3-
(dimethylamin)acrolein or 3-(dimethylamino)-2-methyl-2-pro-
penal under refluxing conditions to provide the APP core 4a–b.10
Subsequent oxidation of sulfide 4 to the corresponding sulfox-
ide 5 could be achieved with mCPBA at 0 °C to give 6 ready for fur-
ther elaboration. The sulfoxide group could be displaced with both
primary and secondary amines under thermal conditions or more
conveniently on microwave irradiation (NMP, 140–160 °C, typi-
cally 10–20 min).11–13 Compounds of type 7 and 8 were synthe-
sized by sulfoxide displacement of 5a with the corresponding
cyclic piperidine or pyrrolidine carboxylate esters. Hydrolysis of
the esters with NaOH provided the corresponding carboxylic acids,
which were most conveniently isolated as their corresponding so-
dium salts 7. Amide coupling under standard conditions with PyB-
rop provided final amides 8.14
N
N
N
12
13
H
0.017
0.26
0.67
N
N
O
O
Me
2.1
14
15
H
H
0.21
0.88
1.60
4.0
O
N
N
OH
From the SAR data in Table 1, it is clear that a small substitution
at R1 led to dramatic decreases in potency. Comparing 12 with 13,
addition of a methyl group reduced potency by 15-fold.15
This result is consistent with the observed binding mode for 12.
Extending further into the active site leads to a steric clash with the
side-chain of Met-929, resulting in decreased binding affinity
(Fig. 1). We identified a variety of potent compounds by varying
the amine component (R2) to include cyclic, acyclic, primary and
secondary aliphatic amines and by examining a variety of func-
tional groups including amides, alcohols, ethers and basic amines
(see Table 1). Of particular interest was compound 12, since it
exhibited a good potency and isotype selectivity against JAK3.
Selectivity of other amines for JAK2 over JAK3 typically ranged
from 5 to 10-fold. Therefore, we focused our optimization effort
on the SAR of the piperidine carboxamide moiety.
First, we synthesized S-enantiomer 22 in order to confirm the
stereochemical preference suggested by the X-ray co-complex (ta-
ble 2). It was gratifying to see that the binding potency of the S-
enantiomer (22) is slightly better than the corresponding racemate
(12) and maintained the isotype selectivity versus JAK3. Hence, all
piperidine-carboxamides were prepared with the S-configuration
(Table 2). The pyrrolidine containing analogs were prepared as
the racemates (19, 21).
16
H
0.070
0.42
N
N
O
17
18
H
H
0.19
1.80
0.21
NH2
0.025
N
flexibility of the glycine residue of JAK2 allows it to flip into a ‘car-
bonyl-up’ conformation, bringing it and the adjacent Asp 994 into
closer contact with the amide substituent of 12. The ‘carbonyl-
down’ conformation required by the corresponding Ala-Asp (the
green residues in Fig. 1) in JAK3 does not allow the same favorable
contacts.7 Finally, the complex implied a preference for the S-enan-
tiomer of the piperidine carboxamide.
In this Letter, we report the elaboration of the APP series to opti-
mize the potency against JAK2, isotype selectivity versus JAK3, and
to improve pharmacokinetic properties.

M. W. Ledeboer et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6529–6533
6531
NH2
HN
N
CN
d
c
a, b
Cl
SMe
SMe
Cl
NC
H2N
N
N
N
N
N
N
1
2
3
R
R
N
N
R1
N
N
N
N
e, f
N
Y
R2
H2N
H2N
N
N
N
N
6
4 Y = SMe, a R = H, b R = Me,
5 Y = SOMe
e
n
n
N
N
N
R3
N
N
N
N
j
g or h, i
N
N
ONa
R4
4a
H2N
H2N
N
O
N
O
N
N
7
8
n = 0 or 1
n = 0 or 1
Scheme 1. Use Reagents and conditions: (a) NaSMe, THF, 60 °C; (b) malonitrile, NaH, Pd(PPh3)4, DMSO, 100 °C
;
(c) hydrazine hydrate, iPrOH, reflux; (d) 3-
(dimethylamino)acrolein or 3-(dimethylamino)-2-methyl-2-propenal, HOAc, iPrOH, reflux ; (e) mCPBA, DMF, 0 °C; (f) primary or secondary amines, NMP, 140–160 °C, MW,
10–20 min; (g) (S)-methyl pyrrolidine-3-carboxylate, 100 °C; (h) (S)-methyl piperidine-3-carboxylate, 100 °C; (i) NaOH (aq) THF–H2O; (j) 3-dimethylamino-2-methyl-2-
propenal.
Table 2
Table 2 (continued)
Activity of piperidine amide analogs
Compd
R1
n
JAK2 KI
M)
JAK3 KI
M)
TF1-GMCSF
(
l
(
l
IC50
(lM)
n
N
N
CF3
27
1
0.0031
0.067
1.1
N
N
N
N
R1
H2N
N
N
O
Compd
R1
n
JAK2 KI
M)
JAK3 KI
TF1-GMCSF
IC50 (lM)
28
1
0.0046
0.20
0.91
CF3
(
l
(lM)
19
20
–OMe
–OEt
0
1
0.087
0.13
0.60
0.57
—
—
Next, we compared ring size of the cyclic amine. The larger
piperidine carboxamides are typically 10–40-fold more potent
than the corresponding pyrrolidines (21 vs 24). The potency and
selectivity of the intermediate esters (19, 20) was less pronounced,
suggesting that the more rigid and planar amide may confer a con-
formational preference.
21
22
23
0
1
1
0.16
0.007
0.20
2.8
—
N
N
N
0.40
2.1
>20
—
A survey of amide substitution suggested that small aliphatic
tertiary amines provided the best selectivity. Reducing the size of
the N-alkyl groups had a detrimental effect on potency. For exam-
ple, dimethyl amide 23 was 29-fold less potent than diethylamide
22. The loss in potency was accompanied by a loss in selectivity as
well. However, extending one or both of the alkyl groups provided
additional high affinity compounds (24, 25), which maintained 30-
fold selectivity over JAK3 and demonstrated improved cell potency.
However, attempts at further improving the cell potency met with
limited success. Only replacement of one of the alkyl groups with
either a trifluoroethyl (26, 27) or trifluoropropyl (28) further im-
proved potency, while again maintaining isotype selectivity. All
other attempts at improving affinity or cell potency for this piper-
idine carboxamide series were fruitless as the use of larger alkyl
groups, branching of the alkyl groups, monosubstitution, incorpo-
ration of unsaturation or heteroatoms in the alkyl chain all proved
detrimental to potency (data not shown).
24
25
N
1
1
0.0036
0.0047
0.19
0.22
6.5
1.9
N
N
CF3
26
1
0.004
0.17
1.3

6532
M. W. Ledeboer et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6529–6533
Table 3
Activity of benzylamine analogs
N
N
N
R1
H2N
N
N
Compd
R1
JAK2 KI
M)
JAK3 KI
M)
TF1-GMCSF
Compd
R1
JAK2 KI
M)
JAK3 KI
M)
TF1-GMCSF
HT2-IL2
(
l
(
l
IC50
(l
M)
(
l
(
l
IC50
(l
M)
IC50
lM
N
H
29
0.006
0.034
0.054
0.23
>20
38
39
N
H
0.021
0.008
0.29
18
>20
N
N
N
H
30
—
0.081
6.9
11
Cl
F
N
H
N
31
32
33
0.56
4.4
—
>20
—
40
41
42
0.0002
0.052
0.011
0.007
0.24
0.20
0.16
3.2
N
H
0.022
0.036
0.16
0.049
—
—
N
N
H
N
H
N
19
>20
N
H
N
H
N
F
34
35
0.0015
0.0014
0.039
0.025
3.0
1.5
43
44
0.0034
0.0006
0.086
0.012
>20
3.1
F
N
H
0.27
1.9
N
H
Cl
F
F
O
36
37
0.0078
0.0014
0.18
>20
45
46
0.0005
0.0006
0.010
0.012
0.075
0.47
2.4
4.1
N
H
F
F
N
H
0.017
4.7
N
H
In order to overcome the limited cell potency, we sought alter-
native substituents that might allow us to drive down the affinity
further while maintaining suitable isotype selectivity. A survey of
available amines suggested that secondary benzylic amines were
potent compounds and provided opportunity for further optimiza-
tion (see compound 29, Table 3).
In the benzylic amine series, tertiary amines were not as potent.
For example, addition of an N-ethyl group (30) decreased the po-
tency. Replacing the phenyl group with a pyridine group also lead
to loss of potency (31–33). However, incorporation of a 4-halo sub-
stitution such as F or Cl enhanced enzyme potency fourfold and
also improved cell potency (34, 35). Importantly, the 4-F-substitu-
tion on the phenyl ring improved the isotype selectivity against
JAK3 (34) by threefold. Replacing the attachment N atom with oxy-
gen led to a fivefold decrease in potency and an even greater loss of
cell potency (36 vs 34). However, substitution at the benzylic posi-
tion proved fruitful. Addition of a benzylic methyl boosted potency
and further suggested that the S-stereochemistry is preferred (S-37
vs R-38). Combining both 4-F and benzylic methyl groups provided
our most potent compound (40), which was 35-fold selective over
JAK3 and also exhibited submicromolar cellular activity. However,
incorporation of the benzylic methyl group decreased potency

M. W. Ledeboer et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6529–6533
6533
fluorine are significantly more polarized, leading to a more favor-
able interaction with JAK2 that is unavailable in JAK3.
Next, the pharmacokinetic parameters were determined for
compounds with good cellular and enzyme potency. It is evident
from Table 4 that several compounds exhibit a favorable IV PK pro-
file. While compounds 40 and 44 are rapidly cleared as indicated
by high clearances and short T1/2, both compounds 45 and 46 ex-
hibit favorable IV PK profiles with reduced clearance, greater expo-
sure and extended half-lives.
In summary, we have identified potent and selective JAK2
inhibitors with an aminopyrazolopyrimidine core. Specific incor-
poration of the benzylic moiety dramatically enhanced cell po-
tency while maintaining JAK2 isotype selectivity through
a
unique CHÁ Á ÁO hydrogen bond with Gly 993. In addition, several
examples showing promising PK profiles were also identified.
Supplementary data
Figure 2. Crystal structure of compound 40. The CHÁ Á ÁO hydrogen bond between
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.10.053.
the ligand and the carbonyl of Gly 993 is shown in red dashes.
References and notes
Table 4
Pharmacokinetic parameters determined following single IV administration (1 mg/kg)
in male Sprague-Dawley rats
1. Vardiman, J. W.; Harris, N. L.; Brunning, R. D. Blood 2002, 100, 2292.
2. (a) Leonard, W. J.; O’Shea, J. J. Annu. Rev. Immunol. 1998, 16, 293; (b) Wilks, A. F.
Semin. Cell Dev. Biol. 2008, 19, 319.
Compds
Cl (mL/min/kg)
t1/2 (h)
Vss (L/kg)
AUCinf (g h/mL)
3. (a) James, C.; Ugo, V.; Le Couédic, J. P.; Staerk, J.; Delhommeau, F.; Lacout, C.;
Garçon, L.; Raslova, H.; Berger, R.; Bennaceur-Griscelli, A.; Villeval, J. L.;
Constantinescu, S. N.; Casadevall, N.; Vainchenker, W. Nature 2005, 434, 1144;
(b) Levine, R. L.; Wadleigh, M.; Cools, J.; Ebert, B. L.; Wernig, G.; Huntly, B. J. P.;
Boggon, T. J.; Wlodarska, I.; Clark, J. J.; Moore, S.; Adelsperger, J.; Koo, S.; Lee, J.
C.; Gabriel, S.; Mercher, T.; D’Andrea, A.; Froehling, S.; Doehner, K.; Marynen, P.;
Vandenberghe, P.; Mesa, R. A.; Tefferi, A.; Griffin, J. D.; Eck, M. J.; Sellers, W. R.;
Meyerson, M.; Golub, T. R.; Lee, S. J.; Gilliland, D. G. Cancer Cell 2005, 7, 387; (c)
Baxter, E. J.; Scott, L. M.; Campbell, P. J.; East, C.; Fourouclas, N.; Swanton, S.;
Vassiliou, G. S.; Bench, A. J.; Boyd, E. M.; Curtin, N.; Scott, M. A.; Erber, W. N.;
Green, A. R. Lancet 2005, 365, 1054; (d) Kralovics, R.; Passamonti, F.; Buser, A.
S.; Teo, S.-S.; Tiedt, R.; Passweg, J. R.; Tichelli, A.; Cazzola, M.; Skoda, R. C. N. Eng.
J. Med. 2005, 352, 1779.
40
44
45
46
104
111
43.6
34.7
0.9
0.7
3.3
2.4
5.1
4.5
9.5
5.1
0.16
0.15
0.38
0.48
Mean values determined from the concentration (as determined using a specific LC/
MS/MS method) of each compound in rat (n = 2) plasma samples collected prior to
dosing up to 8 h post-dose.
when a 4-Cl phenyl group was present (39), suggesting size limita-
tions of the pocket. Constraining the compounds to further to en-
hance potency and to improve the pharmacokinetic profile of the
compound did not meet with success and in all cases we saw a
drop in enzyme and/or cell potency (41–43). Additional explora-
tion of benzylic substitution identified several other potent com-
pounds that retained much of the isotype selectivity against JAK3
(approx. 20-fold) while still exhibiting submicromolar cellular
activity (44–46).
Next, we assessed the cellular selectivity for JAK2 versus JAK3
mediated STAT-5 phosphorylation. GMCSF stimulation of TF1 cells
leads to STAT-5 phosphorylation by JAK2, while IL2 stimulation of
HT2 cell lines yields phospho-STAT-5 via the JAK3 pathway.16 As
shown in Table 3, for our most potent compounds a good correla-
tion between the enzymatic and cellular selectivity was observed
and cellular selectivities ranged from 8- to 32-fold (40, 44–46).
Compound 45 was found to be the most selective (32-fold).
Screening of selected compounds against our in-house panel of
kinases revealed the series to be quite selective. For example, com-
pound 46 hit only 2 of 24 kinases at Ki <200 nM (FLT3 Ki = 160 nM
and ROCK Ki = 190 nM).
To better understand the effect of F-substitution on selectivity,
we determined the X-ray crystallographic complex of 40 bound to
the kinase domain of JAK2. The selectivity of the 4-F benzylamines
arises from a unique CHÁ Á ÁO hydrogen bond with Gly 993 in JAK2,
as shown in Figure 2.17 In JAK3 this residue is an Ala, and very few
kinases have a Gly in this position as JAK2 does. Because of the flex-
ibility of the glycine side-chain, the protein backbone of JAK2 can
flip up to form a hydrogen bond with the aromatic CH of the ben-
zylamine. While the unsubstituted phenyl does make a weak inter-
action with the carbonyl oxygen, the CH groups adjacent to the
4. For
a recent review on the role of JAK2 mutations in myleoproliferative
disorders see: Morgan, K. J.; Gilliland, D. G. Annu. Rev. Med. 2008, 59, 213.
5. (a) Lucet, I. S.; Fantino, E.; Styles, M.; Bamert, R.; Patel, O.; Broughton, S. E.;
Walter, M.; Burns, C. J.; Treutlein, H.; Wilks, A. F.; Rossjohn, J. Blood 2006, 107,
176; (b) Antonysamy, S.; Hirst, G.; Park, F.; Sprengeler, P.; Stappenbeck, F.;
Steensma, R.; Wilson, M.; Wong, M. Bioorg. Med. Chem. Lett. 2009, 19, 279; (c)
Williams, N. K.; Bamert, R. S.; Patel, O.; Wang, C.; Walden, P. M.; Wilks, A. F.;
Fantino, E.; Rossjohn, J.; Lucet, I. S. J. Mol. Biol. 2009, 387, 219; (d) Manshouri, T.;
Quintás-Cardama, A.; Nussenzveig, R. H.; Gaikwad, A.; Estrov, Z.; Prchal, J.;
Cortes, J. E.; Kantarjian, H. M.; Verstovsek, S. Cancer Sci. 2008, 99, 1265.
6. Crystallographic data for the structures in this paper have been deposited with
the RCSB Protein Data Bank. Compound 12, PDB accession code: 3IO7;
compound 42, PDB accession code: 3IOK.
7. Boggon, T. J.; Li, Y.; Manley, P. W.; Eck, M. J. Blood 2005, 106, 996.
8. (a) Nyberg, W. H.; Cheng, C. C. J. Heterocycl. Chem. 1964, 1, 1; (b) Marchal, L.;
Promel, R.; Martin, R. H.; Cardon, A. Bull. Soc. Chim. Belg. 1960, 69, 177.
9. (a) Schnyder, A.; Indolese, A. F.; Maetzke, T.; Wenger, J.; Blaser, H.-U. Synlett
2006, 3167; (b) Indolese, A. F.; Schynder, A.; Aemmer, T.; Nsenda, T.; Portmann,
R. M.. In Synthetic Methods in Organometallic and Inorganic Chemistry; Hermann,
W. A., Ed.; Thieme: New York, 2002; Vol. 10, p 105.
10. (a) Elnagdi, M. H.; Sallam, M. M. M.; Ilias, M. A. M. Helv. Chim. Acta 1975, 58,
1944; (b) Shaaban, M. R.; Saleh, T. S.; Farag, A. M. Heterocycles 2007, 71, 1765.
11. (a) Sawayama, T.; Yamamoto, R.; Kinugasa, H.; Nishimura, H. Heterocycles
1977, 8, 299; (b) Nagashima, S.; Yokota, M.; Nakai, E.-I.; Kuromitsu, S.; Ohga, K.;
Takeuchi, M.; Tsukamoto, S.-I.; Ohta, M. Bioorg. Med. Chem. 2007, 15, 1044.
12. Chiral benzyl amines were prepared following the methods described in: Liu,
G.; Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc. 1997, 119, 9913.
13. For the convenient preparation of small secondary amines see: Dubowchik, G.
M.; Michne, J. A.; Zuev, D. Bioorg. Med. Chem. Lett. 2004, 14, 3147.
14. (a) Coste, J.; Le-Nguyen, D.; Castro, B. Tetrahedron Lett. 1990, 31, 205; (b) Frerot,
E.; Coste, J.; Pantaloni, A.; Dufour, M. N.; Jouin, P. Tetrahedron 1991, 47, 259.
15. For full details regarding JAK2 and JAK3 Ki determinations refer to
Supplementary data and the following references: (a) Fox, T.; Coll, J. T.; Ford,
P. J.; Germann, U. A.; Porter, M. D.; Pazhanisamy, S.; Fleming, M. A.; Galullo, V.;
Su, M.-S.; Wilson, K. P. Protein Sci. 1998, 7, 2249; (b) Morrison, J. F.; Stone, S. R.
Comments Mol. Cell. Biophys. 1985, 2, 347.
16. For full details regarding IC50 determinations using TF-1 and HT-2 cells refer to
Supplementary data.
17. (a) Taylor, R.; Kennard, O. J. Am. Chem. Soc. 1982, 104, 5063; (b) Pierce, A. C.;
Sandretto, K. L.; Bemis, G. W. Proteins 2002, 49, 567.