2,7-Pyrrolo[2,1-f][1,2,4]triazines as JAK2 inhibitors: Modification of target structure to minimize reactive metabolite formation

Article abstract of DOI:10.1016/j.bmcl.2011.10.032

The JAK2/STAT pathway has important roles in hematopoiesis. With the discovery of the JAK2 V617F mutation and its presence in many patients with myeloproliferative neoplasms, research in the JAK2 inhibitor arena has dramatically increased. We report a nov

Full text of DOI:10.1016/j.bmcl.2011.10.032

Bioorganic & Medicinal Chemistry Letters 21 (2011) 7325–7330
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
2,7-Pyrrolo[2,1-f][1,2,4]triazines as JAK2 inhibitors: Modification of target
structure to minimize reactive metabolite formation
⇑
Linda R. Weinberg , Mark S. Albom, Thelma S. Angeles, Henry J. Breslin, Diane E. Gingrich, Zeqi Huang,
Joseph G. Lisko, Jennifer L. Mason, Karen L. Milkiewicz, Tho V. Thieu, Ted L. Underiner, Gregory J. Wells,
Kevin J. Wells-Knecht, Bruce D. Dorsey
Worldwide Discovery Research, Cephalon, Inc. 145 Brandywine Parkway, West Chester, PA 19380-4245, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The JAK2/STAT pathway has important roles in hematopoiesis. With the discovery of the JAK2 V617F
mutation and its presence in many patients with myeloproliferative neoplasms, research in the JAK2
inhibitor arena has dramatically increased. We report a novel series of potent JAK2 inhibitors containing
a 2,7-pyrrolotriazine core. To minimize potential drug-induced toxicity, targets were analyzed for the
ability to form a glutathione adduct. Glutathione adduct formation was decreased by modification of
the aniline substituent at C2.
Received 2 September 2011
Accepted 7 October 2011
Available online 14 October 2011
Keywords:
JAK2
Kinase inhibitors
Pyrrolotriazine
Glutathione
Ó 2011 Elsevier Ltd. All rights reserved.
Janus kinase 2 (JAK2) is one of four members of the JAK family
of non-receptor tyrosine kinases. The JAK family includes JAK1,
JAK2, JAK3, and tyrosine kinase 2 (Tyk2).1 JAKs have a role in trans-
ducing signals for many cytokines, interferons and growth factors.1
JAK2 phosphorylates tyrosine residues on the cytoplasmic side of
the cytokine receptor2,3 leading to subsequent recruitment and
phosphorylation of signal transducers and activators of transcrip-
tion (STATs).1 Phosphorylated STATs form dimers, migrate to the
nucleus, and activate gene transcription.2 STATs bind to specific
gene promoters which regulate proliferation and differentiation.3
Thus, interrupting this signaling could impede growth in tumor
types that rely wholly, or in part, on JAK2 signaling for growth
and/or survival.
A somatic V617F mutation in the JH2 domain of the JAK2 gene
has been linked to incidence of myeloproliferative neoplasms
(MPNs), such as polycythemia vera (PV), essential thrombocythe-
mia (ET) and idiopathic myelofibrosis (IMF).3 Since the discovery
of the V617F mutation, interest in developing JAK2 inhibitors to
treat this defined patient population has increased leading to a
number of patent applications claiming a variety of chemotypes.2
The efficacy of ‘PAN’ JAK inhibitor CP-690550, in clinical trials
for rheumatoid arthritis (RA), is thought to be enhanced by JAK1
and JAK2 inhibition. JAK1 and JAK2 regulate signaling of proinflam-
matory cytokines such as IL-6.4 In a study using small molecule
O
N
N
H
N
N
H
N
N
H
S
O
O
TG-101348 (SAR 30253)
N
N
N N
N
N
N
O
HN
N
H
HN
N
N
H
CP-690550 (Tofacitinib)
INCB-018424 (Ruxolitinib)
Figure 1. JAK inhibitors in clinical trials.4,6
JAK2 inhibitor INCB-018424, JAK/STAT activator IL-6 levels were
suppressed in the plasma of RA patients.5
Some JAK inhibitors currently in clinical trials are shown in
Figure 1.4,6
⇑
Corresponding author.
E-mail addresses: lweinber3@yahoo.com, lweinber@cephalon.com (L.R. Wein-
berg).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.10.032

7326
L. R. Weinberg et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7325–7330
ing or eliminating this possible liability is generally preferable.10
Table 1
JAK2 SAR and % GSH adduct for analogs 1 and 2
One method to assess the propensity of reactive metabolite forma-
tion is to incubate samples with liver microsomes in the presence
of glutathione (GSH) and analyze for adducts by LC–MS/MS.11,12 In
the presence of glutathione, incubation of compound 2 with liver
microsomes formed significant levels of covalent adducts (61%
after 30 min incubation).13 Presumably, CYP induced oxidation of
the C2 aniline generated a reactive quinonediimine and/or quin-
oneimine intermediate that was trapped by GSH (Fig. 2).14
Due to this undesired activity associated with the aniline substi-
tuent, medicinal chemistry was carried out to maximize inherent
JAK2 inhibitory activity, minimize the propensity of glutathione
adduct formation and optimize pharmaceutical properties such as
liver microsome stability and PK.
a
b
Compd
R
JAK2 IC50
ALK
IC50
LMS t1/2
% GSH adducts
a
c
1
2
H
1.14 0.046 30
>40,>40,34,35
>40,>40,>40,>40 61
63
OMe 379 61
9
When the methoxy group on compound 2 was replaced with
a
Average standard deviation (nM) of 3 or more determinations; see Supple-
hydrogen, the resulting compound
1 displayed dramatically
mentary data for assay conditions.
improved JAK2 potency; however, glutathione adduct formation
remained virtually unchanged (63%, Table 1). JAK2 inhibitory activ-
ity was tolerant of a variety of C7-aryl groups ranging from
methoxyphenyl, methoxypyridyl, N-sulfonamidophenyl to ben-
zenesulfonamide (Table 4). However, when less polar groups were
substituted for the piperazinylpropanol group appended on the
C2-aniline, liver microsomal stability (LMS) was reduced (Table 2).
Two strategies were investigated to reduce the reactive metabolite
liability while maintaining JAK2 activity and LMS. The initial effort
involved incorporating electron withdrawing substituents (Cl and
F) on the C2 aniline to reduce its oxidation potential.15 The second
strategy involved replacing the piperazine with piperidine to pre-
vent formation of quinonediimine reactive species.
b
LMS: Liver microsome stability half life (min) in mouse, rat, dog, human.
Relative to parent.
c
As part of our efforts to identify new kinase inhibitor scaffolds,
we elaborated the SAR around the 2,7-pyrrolo[2,1-f][1,2,4]triazine
core. This scaffold was conceptualized as a constrained analog of
the 2,4-diaminopyrimidine core (e.g., TG-101348), a well-recog-
nized kinase inhibitor scaffold.7 From this work, orally bioavailable
pyrrolotriazine analog 2 was identified as an ALK inhibitor with
modest JAK2 activity, and this analog served as a starting point
for
(Table 1).8,9
The human toxicological risks associated with reactive species
a medicinal chemistry effort targeting JAK2 inhibitors
Target compounds were synthesized by functionalization of
7-bromo-2-methylsulfanyl core 3. After a Suzuki coupling, thio-
methyl intermediate 4 was oxidized to either a sulfoxide 5 or a
capable of forming covalent adducts (either through the parent
or its metabolite) are difficult to quantitate preclinically and reduc-
HO
N
HO
N
N+
N
N
N
O
S
O
N
O
N
CYP bioactivation
N
N
N
H
N
N
N
S
O
O
O
2
reactive quinonediimine metabolite
GSH adduct
CYP
H2N
N
O
S
O
N
N
H
N
N
O
CYP bioactivation
HN
O
N
N
H2O
O
S
O
O
N
N
N
N
N
N
N
N
S
O
O
O
reactive quinonediimine metabolite
Figure 2. Proposed mechanism to form a reactive quinoneimine metabolite.

L. R. Weinberg et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7325–7330
7327
Table 2
Table 3
JAK2 SAR and LMS of C2 aniline
JAK2 SAR and % GSH adduct for analogs 17–19
a
Compd Ar
JAK2 IC50
(nM)
% GSH adducts (relative to
parent)
LMS (m,r,d,h)a
7,7,5,7
JAK2 IC50
0.5 0.2
b
Compd
Ar
O
N
O
8
SO2
17
18
1.0 0.2
0.5 0.2
10.6
7.1
O
9
5,8,12,20
5,5,7,5
0.17 0.04
2.1 0.6
N
O
O
O
N
O
S
10
11
19
0.36 0.07
2.7
SO2
HN
N
15,19,19,>40
0.23 0.08
a
Reported as the average standard deviation of 3 or more determinations; see
Supplementary data for assay conditions.
N
N
12
13
<5,11,<5,>40
0.19 0.04
0.47 0.05
Sandmeyer reaction and hydrogenated to afford the 2-chloro-4-
methylpiperazinyl aniline 43 (Scheme 3).
HO
N
Incorporation of a fluorine atom into the C2 aniline led to a ser-
ies of analogs that retained potent JAK2 inhibition and reduced glu-
tathione adduct formation. For example, fluoro-analog 17 had a
comparable JAK2 IC50 value as proto-analog 1 (1 nM) but nearly
6-fold lower glutathione adduct formation (63% ? 11%, Table 3).
Other fluorinated C2 aniline analogs with alternative C7-aryl
groups had similar properties—potent JAK2 inhibition (<1 nM)
with lower, but measurable, reactive metabolite potential (3–7%).
A more dramatic reduction of glutathione adduct formation was
observed in a series of analogs in which a chlorine atom (X = Cl)
was incorporated adjacent to the piperazine nitrogen of the C2 ani-
line (Table 4). In order to expedite syntheses and purification of
analogs in this series, the [piperazin-1-yl]-propan-2-ol group
(e.g., as in analog 1) was replaced by a methyl group (e.g., as in ana-
log 20). In general, liver microsome induced glutathione reactivity
was reduced and potent JAK2 inhibition was maintained. Analogs
with a fluorine ortho to the piperazine nitrogen (21, 24, 27 and
30) were significantly less reactive than unsubstituted analogs
(20, 23, 26 and 29) in the glutathione assay. Analogs with a chlo-
rine ortho to the piperazine (22, 25, 28 and 31) were the same or
slightly less reactive than analogs with a fluorine in that position.
On the other hand, glutathione reactivity was only slightly
attenuated when a chlorine atom was appended adjacent to the
C2 aniline nitrogen (32) rather than adjacent to the piperazine
nitrogen (22, Table 5). Liver microsome induced glutathione reac-
tivity for chloro-analog 32 was only 2-fold less than either proto-
analog 20. This compared with a 25-fold reduction in reactivity
for chloro-analog 22. In the latter case, the chlorine substituent
might exert its influence by forcing the piperazine ring into an ori-
entation perpendicular to the C2 aniline. Such a disruption of con-
jugation between the piperazine nitrogen lone pair and the aniline
would reduce its oxidation potential and diminish formation of a
putative quinoneimine species.
N
>40,36,>40,>40
(S)
14
15
5,6,8,9
0.35 0.05
0.8 0.1
N
O
N
7,7,5,>40
N
N
16
5,6,9,32
0.54 0.06
N
O
(S)
a
LMS: Liver microsome stability half life (min) in mouse (m), rat, (r) dog (d),
human (h).
Average standard deviation (nM) of 3 or more determinations; see Supple-
mentary data for assay conditions.
b
sulfone 6. Aqueous hydrolysis afforded phenol
7 that was
converted to a triflate and then coupled with a substituted aniline
to yield the desired targets (Scheme 1). Analogs with a C7 N-
methyl-N-mesylphenyl group were prepared as described
previously.8,9,16
Substituted anilines were purchased or prepared as previously
described as in Schemes 2 and 3.9 For example, nitrophenyl piper-
azines or piperidines were first N-alkylated, and then reduced to
the corresponding anilines.
An alternative strategy for decreasing reactive metabolite for-
mation involved replacing the piperazine ring with a piperidine
5-fluoro-2-nitrophenyl amine 40 was first reacted with 1-meth-
ylpiperazine, then the amine was converted to a chloride via a

7328
L. R. Weinberg et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7325–7330
N
N
N
b
S
O
N
Ar1
Ar1
a
d
N
N
5a-d
N
N
c
S
N
3
S
N
Br
Ar1
N
N
4a-d
O
S
O
4a, Ar1 = N-tert-butyl-3-benzene
sulfonamide
6a-d
4b, Ar1 = 2-methoxyphenyl
4c, Ar1 = 2-NSO2Me-phenyl
4d, Ar1 = 2-Methoxy-pyridin-5-yl
e, f
N
N
Ar2
N
N
N
N
O
N
Ar1
Ar1
7a-d
8-36
Scheme 1. Target synthesis. Reagents and conditions: (a) aryl boronate (2 equiv), Pd(OAc)2 (0.2 equiv), PPh3 (0.2 equiv), aq Na2CO3, THF, EtOH, 80 °C; (b) m-CPBA (1 equiv),
DCM; (c) sodium tungstate dihydrate (0.05 equiv), 50% H2O2 (3 equiv), MeOH, 60 °C; (d) aq NaOH, 100 °C; (e) N-phenylbis(trifluoromethanesulfonimide) (2 equiv), DIPEA
(2 equiv), 1-methoxy-2-propanol; (f) Ar2NH2 (1.5 equiv), 1-methoxy-2-propanol, 50 °C.
Scheme 2. Synthesis of substituted anilines. Reagents and conditions: (a) (S)-(À)-propylene oxide (1.5 equiv, MeOH; (b) 10% Pd/C, MeOH in H-cube, 10 bar, 30 °C.
Table 4
JAK2 SAR and % GSH adduct for analogs 20–31
a
Compd Ar
X
JAK2 IC50
(nM)
% GSH adducts (relative
to parent)
N
N
N
20
SO2
SO2
SO2
H
1.6 0.3
3.7 0.3
49.6
3
Scheme 3. Synthesis of 2-chloro-4-methylpiperazinyl anilines. Reagents and
conditions: (a) 1-methylpiperazine (4 equiv), 60 °C; (b) H2SO4 (20 equiv), NaNO2
(1.1 equiv), CuCl (2.2 equiv); 12 M aq HCl (20 equiv); (c) Fe (6.7 equiv), acetic acid,
THF.
21
22
F
Cl 1.1 0.2
1.4
ring. Since the oxidation potential of 4-piperidinylaniline is much
higher than 4-piperazinylaniline, the likelihood of forming a reac-
tive species such as a quinonemethide-imine, should be reduced.

L. R. Weinberg et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7325–7330
7329
Table 5
Table 4 (continued)
JAK2 SAR and % GSH adduct for analogs 20, 22 and 32
a
Compd Ar
X
H
F
JAK2 IC50
(nM)
% GSH adducts (relative
to parent)
O
O
O
23
24
25
0.17 0.03 8.3
0.35 0.02
0
0
JAK2 IC50 (nM)a
% GSH adducts (relative to parent)
Cl 0.5 0.2
Compd
X
Y
20
22
32
H
Cl
H
H
H
Cl
1.6 0.3
1.1 0.2
480 185
49.6
1.4
24.3
a
Reported as the average standard deviation of 3 or more determinations; see
26
27
H
0.33 0.09 19.6
SO2
HN
Supplementary data for assay conditions.
Table 6
F
1.0 0.2
1.8
SO2
HN
JAK2 SAR and % GSH adduct for analogs 33–36
28
29
Cl 1.2 0.1
0
SO2
HN
a
Compd Ar
JAK2 IC50
LMS (m,r,d,h)b
29,18,>40,<5
% GSH adductsc
N
33
SO2 0.9 0.3
0
H
F
0.3 0.1
4.9
N
O
O
O
O
34
35
0.6 0.1
7,34,>40,>40
0
0
30
31
0.4 0.1
0
N
1.5 0.1
>40,>40,>40,18
SO2
HN
N
Cl 0.4 0.1
0
36
0.50 0.08 22,>40,>40,>40
0
N
a
Reported as the average standard deviation of 3 or more determinations; see
O
Supplementary data for assay conditions.
a
Average standard deviation (nM) of 3 or more determinations; see Supple-
mentary data for assay conditions.
b
LMS: Liver microsome stability half life (min) in mouse (m), rat, (r) dog (d),
Indeed, no glutathione adduct was detected for any analog in this
series. Although JAK2 activity was maintained (Table 6), some ana-
logs had reduced LMS (e.g., hLMS of 33 vs 1: <5 vs >40 min). In this
piperidinyl series, LMS was dependent on the nature of the C7 aryl
group.17
In summary, starting from a 2,7-pyrrolo[2,1-f][1,2,4]triazine
analog (2) with modest inhibition of JAK2 (IC50 = 417 nM) and high
liver microsome induce glutathione reactivity (>60% relative to
parent), a series of potent JAK2 inhibitors were prepared (IC50
<1 nM) with dramatically reduced glutathione reactivity (0–2%)
and acceptable LMS. These compounds are far less likely to be asso-
human (h).
c
Relative to parent.
ciated with reactive metabolite formation and concomitant unfa-
vorable drug-induced toxicity.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.10.032.

7330
L. R. Weinberg et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7325–7330
form glutathione adducts and are not associated with drug-induced toxicity
References and notes
(i.e. omeprazole), a higher proportion of drugs which are associated with drug-
induced toxicity form such adducts (i.e. acetaminophen).
11. Hop, C. E. C. A.; Kalgutkar, A. S.; Soglia, J. R. Annu. Rep. Med. Chem. 2006, 41, 369.
12. Park, B. K.; Kitteringham, N. R.; Maggs, J. L.; Pirmohamed, M.; Williams, D. P.
Annu. Rev. Pharmacol. Toxicol. 2005, 45, 177.
1. Hitoshi, Y.; Lin, N.; Payan, D. G.; Markovtsov, V. Int. J. Hematol. 2010, 91, 189.
2. Kiss, R.; Sayeski, P. P.; Keseru, G. M. Expert Opin. Ther. Pat. 2010, 20, 471.
3. Quintas-Cardama, A.; Kantarjian, H.; Cortes, J.; Verstovsek, S. Nat. Rev. Drug
Disc. 2011, 10, 127.
13. Wells-Knecht, K. J.; Ott, G. R.; Cheng, M., Wells, G. J.; Breslin, H.; Gingrich, D.;
Weinberg, L.; Mesaros, E.; Huang,, Z.; Yazdanian, M.; Ator, M. A.; Aimone, L.;
Zeigler, K.; Dorsey, B. D. Chem. Res. Toxicol. 2011, submitted for publication.
14. Walgren, J. L.; Mitchell, M. D.; Thompson, D. C. Crit. Rev. Toxicol. 2005, 35, 325.
15. Walker, D. P.; Bi, F. C.; Kalgutkar, A. S.; Bauman, J. N.; Zhao, S. X.; Soglia, J. R.;
Aspnes, G. E.; Kung, D. W.; Klug-McLeod, J.; Zawistoski, M. P.; McGlynn, M. A.;
Oliver, R.; Dunn, M.; Li, J.; Richter, D. T.; Cooper, B. A.; Kath, J. C.; Hulford, C. A.;
Autry, C. L.; Luzzio, M. J.; Ung, E. J.; Roberts, W. G.; Bonnette, P. C.; Buckbinder,
L.; Mistry, A.; Griffor, M. C.; Han, S.; Guzman-Perez, A. Bioorg. Med. Chem. Lett.
2008, 18, 6071.
16. Breslin, H. J.; Chatterjee, S.; Diebold, J. L.; Dorsey, B.; Dunn, D..; Gingrich, D. E.;
Hostetler, G. A.; Hudkins, R. L.; Hunter, R.; Josef, K.; Lisko, J. G.; Mesaros, E. F.;
Milkiewicz, K.; Ott, G. R.; Sundar, B. G.; Theroff, J. P.; Thieu, T.; Tripathy, R.;
Underiner, T. L.; Weinberg, L.; Wells, G. J. International Patent WO
2010071885A1, 2008.
4. Flanagan, M. E.; Blumenkopf, T. A.; Brissette, W. H.; Brown, M. F.; Casavant, J.
M.; Shang-Poa, C.; Doty, J. L.; Elliott, E. A.; Fisher, M. B.; Hines, M.; Kent, C.;
Kudlacz, E. M.; Lillie, B. M.; Magnuson, K. S.; McCurdy, S. P.; Munchhof, M. J.;
Perry, B. D.; Sawyer, P. S.; Strelevitz, T. J.; Subramanyam, C.; Sun, J.; Whipple, D.
A.; Changelian, P. S. J. Med. Chem. 2010, 53, 8468.
5. Malemud, C. J. Pharmaceuticals 2010, 3, 1446.
6. Verstovsek, S. Clin. Cancer Res. 2010, 16, 1988.
7. Ahmed, G.; Bohnstedt, A.; Breslin, H. J.; Burke, J.; Curry, M. A.; Diebold, J. L.;
Dorsey, B.; Dugan, B. J.; Feng, D.; Gingrich, D. E.; Guo, T.; Ho, K.; Learn, K. S.;
Lisko, J. G.; Liu, R.; Mesaros, E. F.; Milkiewicz, K.; Ott, G. R.; Parrish, J.; Theroff, J.
P.; Thieu, T. V.; Tripathy, R.; Underiner, T. L.; Wagner, J. C.; Weinberg, L.; Wells,
G. J.; You, M.; Zificsak, C. A. International Patent WO 2008051547A1, 2008.
8. Thieu, T.; Sclafani, J. A.; Levy, D. V.; McLean, A.; Breslin, H. J.; Ott, G. R.; Bakale,
R.; Dorsey, B. D. Org. Lett. 2011, 13, 4204.
9. Ott, G. R.; Wells, G. J.; Thieu, T. V.; Quail, M. R.; Lisko, J. G.; Mesaros, E. F.;
Gingrich, D. E.; Ghose, A. K.; Wan, W.; Lu, L.; Cheng, M.; Albom, M. S.; Angeles,
T. S.; Huang, Z.; Aimone, L. D.; Ator, M. A.; Ruggeri, B. A.; Dorsey, B. D. J. Med.
Chem. 2011, dx.doi.org/10.1021/jm200758k.
17. Mesaros, E. F.; Thieu, T. V.; Wells, G. J.; Zificsak, C. A.; Wagner, J. C.; Breslin, H. J.;
Tripathy, R.; Diebold, J. L.; McHugh, R. J.; Wohler, A. T.; Quail, M. R.; Wan, W.;
Lu, L.; Huang, Z.; Albom, M. S.; Angeles, T. S.; Wells-Knecht, K.; Aimone, L. D.;
Cheng, M.; Ator, M. A.; Ott, G. R.; Dorsey, B. D. J. Med. Chem. 2011, submitted for
publication.
10. (a) Gan, J.; Ruan, Q.; He, B.; Zhu, M.; Shyu, W. C.; Humphreys, W. G. Chem. Res.
Toxicol. 2009, 22, 690; (b) Although there are some drugs on the market which