Pyrrolo[1,2-f]triazines as JAK2 inhibitors: Achieving potency and selectivity for JAK2 over JAK3

Article abstract of DOI:10.1016/j.bmcl.2011.01.022

SAR studies of pyrrolo[1,2-f]triazines as JAK2 inhibitors is presented. Achieving JAK2 inhibition selectively over JAK3 is discussed.

Full text of DOI:10.1016/j.bmcl.2011.01.022

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1425–1428
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrrolo[1,2-f]triazines as JAK2 inhibitors: Achieving potency and selectivity
for JAK2 over JAK3
⇑
Lalgudi S. Harikrishnan , Muthoni G. Kamau, Honghe Wan, Jennifer A. Inghrim, Kurt Zimmermann,
Xiaopeng Sang, Harold A. Mastalerz, Walter L. Johnson, Guifen Zhang, Louis J. Lombardo, Michael A. Poss,
George L. Trainor, John S. Tokarski, Matthew V. Lorenzi, Dan You, Marco M. Gottardis, Kathy F. Baldwin,
Jonathan Lippy, David S. Nirschl, Ruhui Qiu, Arthur V. Miller, Javed Khan, John S. Sack, Ashok V. Purandare
Bristol-Myers Squibb Co. Route 206 & Provinceline Rd., Princeton, NJ 08543-4000, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
SAR studies of pyrrolo[1,2-f]triazines as JAK2 inhibitors is presented. Achieving JAK2 inhibition selec-
tively over JAK3 is discussed.
Received 11 November 2010
Revised 4 January 2011
Accepted 6 January 2011
Available online 11 January 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Kinase
Janus kinase
JAK
JAK2
JAK3
MPD
Myeloproliferative diseases
JAK-STAT
JAK2 is a member of the Janus kinase (JAK) family of non-recep-
tor tyrosine kinases comprising four family members (JAK1, JAK2,
JAK3 and TYK2). Each JAK is composed of two carboxy terminal
kinase domains, JH1 and JH2, with the JH2 domain representing
need exists as the current standard of care is only palliative, and
does not change the course of these diseases.
High sequence homology, however, with other JAK family
members has posed a major challenge to the design of selective
JAK2 inhibitors.5 Given that other JAK family members are involved
in the regulation of immune function, it is important to maintain
selectivity for JAK2 over these family members in order to mitigate
the risks associated with undesired immunosuppression.6 Several
JAK2 inhibitors with varying selectivity profiles are currently being
evaluated in preclinical settings as well as in clinical trials.7 Herein
we report initial ‘hit to lead’ optimization efforts that led to the
identification of the potent and selective JAK2 inhibitor, pyrrolotri-
azine 29.
A subset of an internal compound collection was evaluated
in vitro in JAK2 and JAK3 kinase assays.8 Pyrrolotriazine 1, exhibit-
ing weak JAK2 and JAK3 inhibitory activity, was identified as a hit.
The modular nature of this chemotype made it an attractive start-
ing point to explore SAR. Furthermore, preliminary molecular
modeling studies of the chemotype in an X-ray structure of
JAK29 suggested that the C4 substituent binds to the hinge region
of the JAK2 catalytic domain, while the C2 substituent binds near
the ribose pocket. Since a number of the JAK2 and JAK3 residues
in this region are not conserved and the P-loop conformations
appear to differ between the two, at least when compared to a
a
pseudokinase domain that negatively regulates enzymatic
activity.1 The amino terminal half of the protein contains a FERM
domain (short for band 4.1 ezrin, radixin and moesin) that associ-
ates with cytokine receptors. JAKs are essential components to
cytokine receptor signaling; they are activated upon ligand binding
and subsequently phosphorylate a STAT (signal transducers and
activators of transcription) which is translocated to the nucleus
to initiate gene expression.2 Imbalances in the JAK-STAT pathway
have been implicated in various inflammatory diseases and
cancers.3
The recent discovery of activating mutations in the tyrosine ki-
nase gene, JAK2, and constitutive activation of the JAK2-STAT path-
way in a large number of Philadelphia chromosome negative
myeloproliferative disease patients has ignited considerable inter-
est in these diseases, and has highlighted JAK2 as a therapeutic
intervention point for drug discovery efforts.4 Significant medical
⇑
Corresponding author.
E-mail address: lalgudi.harikrishnan@bms.com (L.S. Harikrishnan).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.022

1426
L. S. Harikrishnan et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1425–1428
N
S
HetAr
Cl
N
HN
c
a or b
1
N
HN
N
4
N
JAK2 IC50 14 µM
JAK3 IC50 18 µM
N
N
N
Cl
N
Cl
N
N
N
N
HetAr
HetAr
HN
HN
O
d
N
N
N
N
N
N
N
N
NH
N
R
O
Scheme 1. Preparation of 2,4-disubstituted pyrrolo[1,2-f]triazines. Reagents and
conditions: (a) HetArNH2, Et3N, iPrOH, 45 °C, 16 h; (b) HetArNH2, NaH, DMF;
(c) piperazine, NMP, 125 °C, 2 h; (d) RCOOH, HATU, DIPEA, DMF, rt, 18 h.
published JAK3 structure,10 compound 1 offered reasonable poten-
tial for achieving selectivity (Fig. 1).
At first, we explored various heteroaromatic amines as C4 sub-
stituents. The results are summarized in Table 1. The poor potency
of lead thiadiazole 1 can be attributed to repulsive interaction be-
tween the lone pair of electrons on N4 of thiadiazole with carbonyl
of Glu-930. When thiadiazole is replaced with an unsubstituted
thiazole (compound 5), this repulsive interaction is replaced with
a weak attractive interaction between the CH at 4 position of thia-
zole and Glu-930. Further variations at C4 revealed that 5-methyl-
pyrazole 7 was more potent than 5-cyclopropylmethylpyrazole 6,
possibly due to small size of the binding pocket. 5-Methyl pyrazole
7 exhibited a good balance between JAK2 potency, selectivity ver-
sus JAK3 and cellular potency (proliferation of SET-2 cell line is
dependent on JAK2 activity11) and hence, was chosen for investi-
gating SAR at the C2-position. No clear trend in selectivity versus
JAK1 was observed with in the compounds disclosed herein.12
The compounds were synthesized via sequential SNAr reactions
of dichloropyrrolotriazine as outlined in Scheme 1. The first SNAr
Figure 1. A binding model of thiadiazole 1 (carbons in magenta) in JAK2 (carbons
and ribbon in green) superposed with an X-ray structure of JAK3 (carbons and
ribbon in cyan). Two key residues that differ between the two kinases are
highlighted, namely JAK2 Ser936 (Cys in JAK3) and Gly993 (Ala in JAK3). Hydrogen
bonds to the hinge are indicated by dotted lines.
Table 1
JAK2 inhibition and selectivity data for various heteroaromatic anilines at C4
HetAr
HN
4
N
N
N
N
Table 2
N
JAK2 inhibition and selectivity data for various piperazine substituents at C2
O
NH
N
Compd
HetAr=
JAK2
JAK3/
JAK2
JAK1/
JAK2
SET-2
HN
IC50
(
lM)
(IC50
, lM)
N
N
N
N
2
N
1
2
14
1.3
19
nd
0.69
N
N
S
N
R
H
N
O
2.7
0.75
1.4
Compd R=
JAK2
IC50
JAK3/JAK2 JAK1/JAK2 SET-2
HN
(
l
M)
(IC50, lM)
8
9
0.031
0.022
15
5
16
nd
nd
H
O
3
4
5
0.19
41
15
37
11
nd
N
N
nd
nd
N
S
N
0.11
1.3
2.6
0.67
1.2
N
10
0.0093
11
5.0
0.059
O
O
O
O
Ph
H
N
11
7
0.0022
0.0028
0.0020
19
19
11
13
0.062
0.37
N
N
6
7
0.022
9.0
19
7.7
4.5
0.98
0.37
Ph
4.5
1.7
H
N
Ph
0.0028
12
0.076
nd = not determined.
nd = not determined.

L. S. Harikrishnan et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1425–1428
1427
Table 3
JAK2 inhibition and selectivity data for various piperazine substituents at C2
NH
N
HN
4
N
N
2
6
N
N
N
n
O
Compd
n
Phenyl-substituent
JAK2
IC50
JAK3/ JAK1/ SET-2
(l
M) JAK2
JAK2
IC50 (lM)
2
4
6
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
Me
0.0030
0.0045
0.0060
0.0048
0.0035
0.0073
0.0021
0.0011
0.0053
0.0011
0.0046
0.0009
0.0012
0.0006
0.0024
0.0010
0.0018
33
3.9
27
1.3
24
22
47
94
9.6
20
17
25
30
16
22
67
69
12
0.16
0.059
0.14
0.0062
0.090
0.071
0.18
0.19
0.12
0.31
0.17
0.054
0.63
0.59
0.054
0.080
0.11
Me
CN
1.1
1.7
0.7
4.5
3.0
9.1
3.9
1.7
2.8
3.8
6.5
7.9
nd
CN
OMe
Cl
Br
SO2Me
Me
Figure 2. X-ray crystal structure of pyrrolotriazine 29 (carbons in magenta) bound
to JAK2 (carbons and ribbon in green). Hydrogen bonds to the hinge are indicated
by dotted lines.
Me
F
F
OMe
Cl
F
Cl
OMe
Leu932, and the pyrazole NH donates a hydrogen bond to the hinge
carbonyl of Glu930. The ortho substituent on phenyl ring binds
near Gly in JAK2 versus Ala in JAK3, which might explain the mod-
ulation in selectivity observed with ortho substituted compounds.
In addition, the benzyl group is tucked under the P-loop which
may also be a source of selectivity between JAK family members
because of subtle packing differences mentioned by authors of an
analysis of JAK3 and TYK2 X-ray structures in comparison with
JAK1 and JAK2 structures.14
In summary, hits to lead optimization established SAR at the
C-2 and C-4 positions of the initial pyrrolotriazine hit, and led to
the discovery of analogs (20, 28 and 29) with significantly
improved in vitro biochemical and cellular potency and JAK3
selectivity. No significant pattern in selectivity versus JAK1 was ob-
served across the set of compounds disclosed in this manuscript.
Unfortunately, most of the compounds suffered from poor
metabolic stability (data not shown). Efforts towards achieving im-
proved metabolic stability will be forthcoming.
F
F
F
2.1
3.8
3.4
F
F
nd = not determined.
displacement reaction at C4 with electron rich heteroaryl amines
proceeded at room temperature with triethylamine as base.
However, electron deficient heteroaryl amines required a stronger
base such as sodium hydride to effect complete reaction at C4. The
second SNAr reaction at C2 required heating the reaction mixture
in NMP up to 125 °C.13
Investigation of N-substitution of the piperazine revealed
improvements in JAK2 potency upon acylation of nitrogen
(Table 2). Further, larger acyl groups containing an aromatic ring
were preferred as exemplified by compounds 11, 7 and 12. Amongst
the acyl substituents, benzoyl compound 11 and phenacetyl com-
pound 7 exhibited the best selectivity versus JAK3. Based upon these
results and the fact that certain residues differ near the C2 binding
pocket of JAK2 and JAK3, compounds 11 and 7 were chosen for fur-
ther fine tuning of the selectivity against JAK3. This SAR is summa-
rized in Table 3.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.01.022.
Amongst the benzoyl compounds (n = 0), substitution at the
4-position of the phenyl ring improved SET-2 potency at the ex-
pense of JAK3 selectivity (e.g., 14 and 16). However, substitution
at the 2-position of the phenyl ring consistently provided greater
selectivity (compared to substitution at 4-position) versus JAK3
(see 13, 15, 17–20) with retention of cellular potency. Further
substitutions such as 2,4- on the phenyl ring did not lead to any
appreciable improvements in selectivity (data not shown). In the
case of phenacetyl compounds (n = 1), electron withdrawing sub-
stituents at the ortho and para positions afforded no significant
improvements in JAK3 selectivity (e.g., 23–27). However 2,4,6-tri-
substitution on the phenyl ring led to the identification of potent
and JAK3 selective (>60-fold) compounds (e.g., 28–29) with excel-
lent antiproliferative activity in the SET-2 cell line. The SAR could
be explained by analysis of an X-ray structure of compound 29
bound to the JAK2 kinase domain (Fig. 2, PDB id code 3Q32) which
was in agreement with the docking-based hypothesis described
earlier. The increased potency of the C4 pyrazole ring analogs
was rationalized by the observation that the pyrazole basic nitro-
gen accepts a hydrogen bond from the amide NH of hinge residue
References and notes
1. (a) Ziemiecki, A.; Harpur, A.; Wilks, A. Trends Cell Biol. 1994, 4, 207; (b)
Saharinen, P.; Vihinen, M.; Silvennoinen, O. Mol. Biol. Cell 2003, 14, 1448.
2. (a) Harpur, A. G.; Andres, A. C.; Ziemiecki, A.; Aston, R. R.; Wilks, A. F. Oncogene
1992, 7, 1347; (b) Yamaoka, K.; Saharinen, P.; Pesu, M.; Holt, V. E. T.;
Silvennoinen, O.; O’Shea, J. J. Genome Biol. 2004, 5, 253.
3. Yu, H.; Pardoll, D.; Jove, R. Nat. Rev. Cancer 2009, 9, 798.
4. (a) James, C.; Ugo, V.; Le Couedic, J.-P.; Staerk, J.; Delhommeau, F.; Lacout, C.;
Garcon, 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)
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; (d) 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; (e) Zhao, R.; Xing, S.; Li, Z.; Fu, X.; Li, Q.; Krantz, S. B.;
Zhao, Z. J. J. Biol. Chem. 2005, 280, 22788; (f) Tono, C.; Xu, G.; Toki, T.; Takahashi,
Y.; Sasaki, S.; Terui, K.; Ito, E. Leukemia 2005, 19, 1843; (g) Tefferi, A.; Lasho, T.

1428
L. S. Harikrishnan et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1425–1428
L.; Schwager, S. M.; Steensma, D. P.; Mesa, R. A.; Li, C.-Y.; Wadleigh, M.;
Gilliland, D. G. Br. J. Haematol. 2005, 131, 320.
9. 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.
10. Boggon, T. J.; Li, Y.; Manley, P. W.; Eck, M. J. Blood 2005, 106, 996.
11. (a) Quentmeier, H.; MacLeod, R. A.; Zaborski, M.; Drexler, H. G. Leukemia 2006,
20, 471; (b) Uozumi, K.; Otsuka, M.; Ohno, N.; Moriyama, T.; Suzuki, S.;
Shimotakahara, S.; Matsumura, I.; Hanada, S.; Arima, T. Leukemia 2000, 14, 142.
12. Since certain JAK2 compounds in the clinic7g are dual JAK2/1 inhibitors,
selectivity (or lack thereof) versus JAK1 is disclosed in this manuscript.
13. For details on experimental procedures see: Purandare, A. V.; Frennesson, D. B.;
Kamau, M. G.; Harikrishnan, L. S.; Saulnier, M. G. WO 2010042684, 2010.
14. Chrencik, J. E.; Patny, A.; Leung, I. K.; Korniski, B.; Emmons, T. L.; Hall, T.;
Weinberg, R. A.; Gormley, J. A.; Williams, J. M.; Day, J. E.; Hirsch, J. L.;
Kiefer, J. R.; Leone, J. W.; Fischer, H. D.; Sommers, C. D.; Huang, H.-C.;
Jacobsen, E. J.; Tenbrink, R. E.; Tomasselli, A. G.; Benson, T. E. J. Mol. Biol.
2010, 400, 413.
5. The catalytic domain of JAK2 is 61% identical to that of JAK3. For residues that
line the ATP binding pocket that are pointing inward, the identity is 91%.
6. (a) Ghoreschi, K.; Laurence, A.; O’Shea, J. J. Immunol. Rev. 2009, 228, 273; (b)
Wilson, L. J. Expert Opin. Ther. Patents 2010, 20, 609.
7. For general reviews on the therapeutic potential of JAK2 inhibitors, see: (a)
Verstovsek, S. Clin. Cancer Res. 2010, 16, 1988; (b) Atallah, E.; Verstovsek, S.
Expert Rev. Anticancer Ther. 2009, 9, 663; (c) Takenaka, K. Ketsueki, Shuyoka
2009, 58, 70; (d) Pardanani, A. Leukemia 2008, 22, P23; (e) Skoda, R. C. Blood
2008, 111, 5419; (f) Morgan, K. J.; Gilliland, D. G. Annu. Rev. Med. 2008, 59, 213;
(g) Purandare, A. P.; Lorenzi, M. V.; Lombardo, L. J. Annu. Rep. Med. Chem. 2010,
45, 211; (h) Kumar, C.; Purandare, A. V.; Lee, F. Y.; Lorenzi, M. V. Oncogene 2009,
28, 2305.
8. For details on biological assays see Supplementary data.