Pyrimidine benzamide-based thrombopoietin receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2007.07.038

A series of pyrimidine benzamide-based thrombopoietin receptor agonists is described. The lead molecule contains a 2-amino-5-unsubstituted thiazole, a group that has been associated with idiosyncratic toxicity. The potential for metabolic oxidation at C-5 of the thiazole, the likely source of toxic metabolites, was removed by substitution at C-5 or by replacing the thiazole with a thiadiazole. Potency in the series was improved by modifying the substituents on the pyrimidine and/or on the thiazole or thiadiazole pendant aryl ring. In vivo examination revealed that compounds from the series are not highly bioavailable. This is attributed to low solubility and poor permeability.

Full text of DOI:10.1016/j.bmcl.2007.07.038

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5447–5454
Pyrimidine benzamide-based thrombopoietin receptor agonists
Lawrence A. Reiter, Chakrapani Subramanyam, Emilio J. Mangual, Christopher S. Jones,*
Marc I. Smeets, William H. Brissette, Sandra P. McCurdy, Paul D. Lira, Robert G. Linde,
Qifang Li, Fangning Zhang, Amy S. Antipas, Laura C. Blumberg, Jonathan L. Doty,
James P. Driscoll, Michael J. Munchhof, Sharon L. Ripp, Andrei Shavnya,
Richard M. Shepard, Diana Sperger, Lisa M. Thomasco, Kristen A. Trevena,
Lilli A. Wolf-Gouveia and Liling Zhang
Pfizer Global Research & Development, Groton Laboratories, Eastern Point Road, Groton, CT 06340, USA
Received 30 May 2007; revised 10 July 2007; accepted 10 July 2007
Available online 31 July 2007
Abstract—A series of pyrimidine benzamide-based thrombopoietin receptor agonists is described. The lead molecule contains a 2-
amino-5-unsubstituted thiazole, a group that has been associated with idiosyncratic toxicity. The potential for metabolic oxidation
at C-5 of the thiazole, the likely source of toxic metabolites, was removed by substitution at C-5 or by replacing the thiazole with a
thiadiazole. Potency in the series was improved by modifying the substituents on the pyrimidine and/or on the thiazole or thiadiazole
pendant aryl ring. In vivo examination revealed that compounds from the series are not highly bioavailable. This is attributed to low
solubility and poor permeability.
Ó 2007 Elsevier Ltd. All rights reserved.
Platelets, the key cellular component of blood clots, are
produced from megakaryocytes, a large, polyploid cell
with a multi-lobated nucleus.1 Platelet formation, as
well as the differentiation of multipotent hematopoietic
bone marrow stem cells into megakaryocytes, is depen-
dent on the hematopoietic growth factor thrombopoie-
tin (TPO) which is constitutively produced in the liver.
TPO induces this differentiation and the production of
platelets from megakaryocytes through interaction with
the cell surface thrombopoietin receptor (TPOr) which
initiates an intracellular cascade involving the kinase
JAK2 and the transcription factor STAT5. Normal lev-
els of platelets range from 100,000 to 500,000/lL. A per-
son is considered thromobocytopenic when platelet
levels drop below 50,000/lL and at this level, the danger
of uncontrollable bleeding arises. Thrombocytopenia
can occur for a variety of reasons among which are im-
paired liver function, immune thrombocytopenia pur-
pura (ITP), and cancer chemotherapy. Although each
of these conditions can be treated with platelet transfu-
sions, this is not typically done until platelet levels drop
below 20,000/lL owing to the cost as well as the danger
inherent in the transfusion of blood products. These
conditions could, in principle, also be alleviated by treat-
ment with exogenous TPO. However, unlike the other
recombinant hematopoietic growth factors, erythropoie-
tin (EPO) and granulocyte colony-stimulating factor
(GCSF), development of TPO (recombinant or engi-
neered) has not been successful, a key negative clinical
finding being the production of neutralizing antibodies
to an engineered rTPO protein. These antibodies
cross-reacted with native TPO thereby leading to severe
thrombocytopenia. Clearly, one approach to avoiding
this problem would be to identify a low molecular
weight compound that would induce TPOr signaling.
In general, the identification of low molecular weight
compounds that induce signaling of cell surface recep-
tors that are normally stimulated by a protein has not
been particularly fruitful. However, in the case of the
TPO receptor, a number of low molecular weight ago-
nists have been reported.2–7 One such agent has been
studied in the clinic and has been reported to effectively
increase platelet numbers in patients with ITP or receiv-
ing anti-viral treatment for hepatitis.8 Our interest in the
Keywords: Thrombopoietin; Low molecular weight agonist; Small
molecule agonist; Thrombocytopenia.
*
Corresponding author. Tel.: +1 860 715 0516; e-mail:
christopher.s.jones@pfizer.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.038

5448
L. A. Reiter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5447–5454
Table 1. Potency of substituted aryl analogs of 2 (EC50s nM)
F
O
S
S
O
S
X
F
N
S
N
H
N
Ar
N
R
HN
L
4
1
O
3
F
CPD substituent
EC50
clogP
O
S
F
3
4
5
6
7
8
9
H
3440
6850
9000
4.36
4.51
4.83
5.26
4.51
5.08
5.23
5.26
5.49
4.51
5.08
4.86
5.26
4.83
5.49
4.12
4.59
5.23
5.38
5.41
5.54
6.60
6.47
4.40
5.51
5.43
4.66
5.23
5.41
5.55
5.06
5.41
4.48
4.61
4.59
5.41
5.41
5.68
5.31
4.66
4.18
4.66
5.48
5.55
6.19
N
N
N
H
2-F
N
N
H
2
2-Cl
2-CF3
3-F
>10000
>10000
3030
645
Figure 1. Lead molecules.
3-Cl
area was stimulated by the disclosure of a series of thiaz-
olidine thiazoles that possess potent TPOr agonist activ-
ity.9 The low molecular weight, potency, and overall
simplicity made the series (represented by 1, Fig. 1) an
attractive starting point for a drug discovery effort.
There were, however, structural features we wished to
avoid, in particular the thiazolidinedione and the 2-ami-
no-5-unsubstituted thiazole, both of which have been
associated with idiosyncratic toxicity.10
3-Br
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
2
3-CF3
3-OCF 3
4-F
418
284
2560
568
4-Cl
4-CH3
4-CF3
4-OCHF 2
4-OCF 3
4-NO2
2,3-diF
2-F-3-Cl
2-F-3-Br
2-F-3-CF 3
2-F-3-ethyl
2-F-3-butyl
1200
492
>10000
>10000
>10000
415
To initiate our effort, we screened sub-sets of com-
pounds from our files seeking to identify molecules that
contain only one of the undesired groups. This effort led
to the identification of 2, which displayed agonist activ-
ity in a BaF3 cell reporter assay containing the human
TPOr.11 In concordance with the low molecular weight
TPO agonists reported by other groups, 2 does not sig-
nal the murine TPOr or the other related hematopoietic
receptors, EPOr and GCSFr. With the identification of
this TPOr agonist, we began a detailed study of its
SAR in an attempt to improve the potency of the series
(2: EC50 400 nM vs 1: EC50 28 nM) and to eliminate the
second undesired structural feature, the 2-amino-5-
unsubstituted thiazole.
612
100
60
143
53
2-F-3-isobutyl
2-F-3-OCH 3
2-F-3-OCF 3
2,3-diCl
35
3130
32
1340
173
2,4-diF
Owing to the facility with which 4-aryl-2-aminothiazoles
can be synthesized through halogenation of substituted
acetophenones and subsequent condensation with thio-
urea, we initially prepared a wide range of analogs of
2 containing various substituted pendant aryl rings.12
The activity of these spans a range from ꢀ30 nM to
>10 lM (Table 1). Of the monosubstituted analogs (4–
18), the most potent contain a relatively large, lipophilic
group at C-3. A CF3 group at C-4 also confers moderate
potency. Larger lipophilic or polar groups at C-4 do not
induce potent agonism nor does C-2 monosubstitution.
Among the disubstituted analogs (2, 19–41), the 2,3-
disubstitution pattern yields the most potent agonists
when the C-3 group is a relatively large lipophilic group,
a finding consistent with data from the monosubstituted
compounds. 2,4-Disubstitution is not quite as effective
and 2,5-disubstitution does not yield potent agonists.
3,4-Disubstitution can lead to compounds of moderate
potency. A 2,3,4-trisubstituted compound with a lipo-
philic C-3 group (45) is among the most potent analogs.
Although further improvements in potency may have
been attained through additional elaboration of the pen-
dant aryl ring, we did not anticipate identifying signifi-
cantly more potent compounds lying within molecular
2-F-4-Cl
232
2-F-4-CF 3
2,4-diCl
259
510
2,4-diCH 3
2-F-5-CF 3
2-OH-5-Cl
2-OCH 3 -5-Cl
3,4-diF
2260
5190
>10000
>10000
403
37
38
39
40
41
42
43
44
45
46
3-F-4-CF 3
3-CF3 -4-F
3,4-diCl
288
104
347
3,4-diCH 3
2,6-diF
784
>10000
244
2-aza-3-CF 3
2,3,4-triF
84
2,3-diF-4-CF
87
3
2,4-diF-3-CF
33
3
2,6-diCl-4-CF
>10000
3

L. A. Reiter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5447–5454
Table 2. TPOr potency of azine analogs of 22 and related compounds
5449
S
O
X
N
Ar
N
R
L
4
3
CPD
Ar-L
Methoda
A
Attachment point
4
R
H
X
BaF3 EC50 (lM)
47
Benzene
2-F-3-CF3
>10 (0.060)b
2.2 (0.060)b
8.6 (0.060)b
0.40 (0.060)b
7.9 (0.060)b
7.7 (0.060)b
0.32 (0.060)b
>10 (0.060)b
1.6 (0.57)b
NH
NH
N
48
49
50
51
52
53
54
55
56
57
58
59
60
61
2-Aminopyridine
3-Aminopyridine
4-Aminopyridine
2-Aminopyrazine
2-Aminopyrimidine
5-Aminopyrimidine
3-Aminopyridazine
4-Aminopyridazine
4-Aminopyridazine
4-Aminopyridazine
4-Aminopyridazine
2-Amino-1,3,5-triazine
3-Amino-1,2,4-triazine
N-Methylated link
B
4
4
4
4
4
4
4
4
4
4
4
4
4
4
H
H
H
H
H
H
H
H
H
H
H
H
H
H
2-F-3-CF3
2-F-3-CF3
2-F-3-CF3
2-F-3-CF3
2-F-3-CF3
2-F-3-CF3
2-F-3-CF3
4-Cl
N
B
NH
NH
B
N
N
C
C
C
D
D
D
D
D
C
A
E
NH
NH
NH
N
N
N
N
N
N
N
NH
NH
NH
NH
N
N
N
N
2,4-DiF
0.55 (0.17)b
0.12 (0.060)b
0.40 (0.10)b
0.54 (0.060)b
>10 (0.060)b
>10
N
N
2-F-3-CF3
3-CF3-4-F
2-F-3-CF3
2-F-3-CF3
2-F-3-CF3
N
N
NH
NH
N
N
N
N
N
NH
N
N
N
N
CH3
(continued on next page)

5450
L. A. Reiter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5447–5454
Table 2 (continued)
CPD
Ar-L
Methoda
F
Attachment point
4
R
H
X
BaF3 EC50 (lM)
N
N
N
62
63
64
Ether link
2-F-3-CF3
>10
N
O
N-Methyl amide
G
H
4
3
CH3
H
2-F-3-CF3
3,4-DiF
>10
>10
N
N
NH
NH
3-Subst. analog
a See Ref. 13 for synthetic methods.
b Activity of corresponding 4-aminopyrimidine analog.
weight and clogP constraints for drug-like properties.
Notably, the most potent analogs have clogPs > 5.5.
pyrimidine substituent with a 2-F-3-OCF3 phenyl substi-
tuent on the thiazole yielded the most potent analog in
the series, 76 (Fig. 2).
Although the pyrimidine group was not of any partic-
ular concern with regard to potential toxicity or other
issues, we did investigate whether it could be replaced
with other azines (Table 2).13 Of this set (mostly bear-
ing a 2-F-3-CF3-phenyl substituent on the thiazole),
only the 4-aminopyridazine analogs (55–58) approach
the potency of the corresponding 4-aminopyrimidines.
We also observed that alkylating the nitrogen linking
the pyrimidine and benzene rings (61), replacing this
nitrogen with an oxygen (62), methylating the amide
group (63), or altering the point of attachment of
the 4-aminopyrimidine moiety (64) all abolish agonist
activity.
Metabolic oxidation of 2-aminothiazoles is known to
lead to ring opening and production of thiourea deriva-
tives that have the potential for hepatotoxicity unless a
blocking substituent at C-5 is present.10 Accordingly,
we explored two strategies in order to reduce this risk:
(1) introducing a substituent at C-5, and (2) replacing
the thiazole with isomeric thiazoles15 or alternative het-
erocycles: thiadiazoles,16,17 pyrazoles,18,19 isothiazoles20
F
CF3
O
S
N
N
N
H
Introduction of substituents to the 4-aminopyrimidine
revealed that substitution at C-2 is not tolerated while
substituents at C-6 could be varied quite widely yielding
some analogs more potent than 22 (Table 3).14 Notably,
incorporation of hydrophilic groups allowed for potency
improvements without raising the clogP, for example,
the N-azetidin-3-ol analog 75 is both ꢀ6-fold more po-
tent and somewhat less lipophilic. Combining this
N
N
H
22
OH
F
OCF3
N
O
S
N
N
N
H
N
N
H
76
Figure 2. Key thiazole agonists.
Table 3. SAR of substituted 4-aminopyrimdine derivatives
F
CF3
Y
N
O
S
Table 4. SAR of 5-substituted 2-aminothiazoles
R1
N
N
N
R2
H
CF3
Z
Y
O
S
X
N
H
N
N
N
H
Compound
X
Y
EC50 (lM) clogP
N
N
H
22
65
66
67
68
69
70
71
72
73
74
75
H– H–
CH3– H–
0.060
>10
5.41
5.91
5.91
6.40
6.35
6.31
6.75
5.45
5.57
6.34
5.87
5.29
Compound
Y
R1
R2
Z
EC50 (lM)
H–
CH3– CH3–
CH3–
0.043
>10
22
38
69
77
78
79
80
81
82
H–
H–
H–
H–
H–
F–
H–
F–
F–
H–
H–
H–
H–
F–
H–
H–
F–
F–
F–
H–
H–
0.060
0.10
H–
H–
H–
H–
H–
H–
H–
H–
CH3O–
(CH3)2N–
CH3CH2NH–
HOCH2CH2N–
HOCH2CH2N(CH3)–
0.038
0.054
0.13
(CH3)2N–
H–
H–
0.054
0.54
0.18
CH3–
CH3–
CH3–
Cl–
0.065
0.022
(CH3)2N–
(CH3)2N–
H–
0.045
0.011
0.14
CH3OCH2CH2N(CH3)– 0.084
0.023
0.009
–CH2O–
–CH2CH2O–
N-Azetidine
N-3-Azetidinol
H–
0.027

Table 5. Replacements for the 2-amino-4-arylthiazole
O
HN
HET
N
N
A
B
C
D
E
F
G
H
2-Amino-4-
arylthiazoles
2-Amino-
5-aryl-thiazoles
5-Amino
thiadiazoles
3-Amino
3-Amino
pyrazoles
4-Amino
pyrazoles
5-Amino
isothiazole
4-Amino
oxazoles
thiadiazole
BaF3 EC50 (lM)
BaF3 EC50 (lM)
BaF3 EC50 (lM)
BaF3 EC50 (lM)
BaF3 EC50 (lM)
BaF3 EC50 (lM)
BaF3 EC50 (lM)
BaF3 EC50 (lM)
S
N
S
N
S
N
N
S
N
O
R / HET→
↓
HN
N
HN
N
HN
N
HN
N
HN
N
HN
S
HN
N
HN
N
R
R
R
R
R
R
R
R
Ph
3 3.4
—
85 1.6
—
—
—
—
102 >10
—
4-Cl–Ph
13 0.578
29 0.17
2 0.40
—
—
—
—
103 >10
—
—
2,4-DiF–Ph
3,4-DiF–Ph
3,4-DiCl–Ph
2-F-3-CF3–Ph
3-CF3-4-F–Ph
3-F-4-CF3–Ph
3-OCF3–Ph
2-F-3-OCF3–Ph
—
86 1.3
87 0.72
88 4.4
89 1.5
90 0.41
91 >10
92 >10
93 0.97
—
—
—
—
—
94 >10
—
—
—
—
—
39 0.35
22 0.060
38 0.10
37 0.29
11 0.28
27 0.032
—
—
99 >10
100 >10
101 >10
—
—
—
83 >10
84 6.7
—
95 >10
—
97 1.4
98 2.0
—
—
104 7.9
105 2.3
—
—
—
—
—
—
—
—
—
—
—
96 >10
—
—
—
—

5452
L. A. Reiter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5447–5454
or oxazoles.21 Of these only C-5 substituted thiazoles
(Table 4) and 5-amino-3-aryl-[1,2,4]-thiadiazoles (Table
5, Column C) displayed agonist activity on the order of
the parent series. The other replacements proved to be
significantly less active.
Table 7. Cell proliferation of key compounds
Compound
BaF3
reporter
EC50 (lM)
BaF3
proliferation
EC50 (lM)
CFUMEG
proliferation
EC50 (lM)
22
76
0.060
0.007
0.054
0.023
0.009
0.045
0.011
0.027
0.013
0.071
1.4
0.34
—
0.29
8.5
Although introduction of a methyl group at C-5 of the
thiazole ring reduced activity in the 2-F-3-CF3-phenyl
series (77 vs 22), such substitution is tolerated in the
4-F-3-CF3 series yielding an analog only slightly less
potent than the corresponding C-5 unsubstituted deriv-
ative (78 vs 38). The 4-dimethylaminopyrimidine ana-
logs 79and 80 emerged as the most potent agonists in
this series. Fusing a ring between the thiazole and phe-
nyl groups, as in 81 and 82, also served to block C-5
of the thiazole. These compounds both displayed good
agonist activity.
69
74
—
0.71
0.34
1.6
0.13
—
75
79
—
—
80
82
1.1
1.3
—
—
110
111
1.0
1.4
—
ferentiate into a megakaryocyte in the presence of TPOr
signaling in a CFUMEG colony assay.
Within the 5-amino-3-aryl-[1,2,4]-thiadiazoles, we ob-
served that the 2-F-3-CF3 and 2-F-3-OCF3 analogs (89
and 93) are not particularly potent. However, the 4-F-
3-CF3 analog 90 is active and only slightly less potent
than the corresponding 2-amino-4-arylthiazole 38.
Throughout our investigations we noted the low solubil-
ity of the compounds in both organic and aqueous med-
ia. Since low solubility can significantly impact the
absorption of a compound, we assessed a number of
compounds for oral bioavailability in rats. Compound
22, which has thermodynamic solubility of <1 lg/mL,
is only 10% bioavailable after dosing in a standard
methyl cellulose formulation. Other compounds from
the series show even lower bioavailabilities, thiazole 76
only 3% and thiadiazole 109 <2%. Both of these have
thermodynamic solubilities <1 lg/mL. A more soluble
thiadiazole 111, having a thermodynamic solubility of
56 lg/mL, was also found to have low bioavailability,
<1%.
Further optimization of potency and physical properties
(i.e., clogP) in the 5-amino-3-aryl-[1,2,4]-thiadiazole
series was achieved by introduction of C-6 substituents
onto the pyrimidine (Table 6). The most potent analog
is the N-azetidin-3-ol analog 110. Notably, inclusion of
hydrophilic side chains on the pyrimidine lowered the
clogPs of many of these analogs to <5.
Although activity in the BaF3 reporter assay was
deemed to be a good indicator of inherent potency, to
better assess agonist activity, the ability of some of the
more potent analogs to induce BaF3 cell proliferation
was determined (Table 7). All of these compounds in-
duced proliferation with a ꢀ20- to 100-fold shift in
EC50s over the reporter assay, a shift attributed to pro-
tein binding in the proliferation assay. Further substan-
tiation of agonist activity was obtained by the
compounds’ ability to induce proliferation of human
CD34+ cells—a multipotent progenitor cell that will dif-
The latter finding in particular suggested that issues
other than solubility may be responsible for the ob-
served low bioavailability. Attempts to assess the perme-
ability of compounds from the series in in vitro
permeability assays were thwarted by their low solubility
and we therefore resorted to in vivo experiments. Thus,
22 was solubilized in a PEG-400 formulation and dosed
intraduodenally. The bioavailability was 11%, essen-
tially the same as when the compound was dosed orally
in a standard formulation. Additionally, 76 was formu-
lated in a solubility-enhancing spray dried dispersion
that increased its solubility to ꢀ85 lg/mL, but when
dosed orally, it was only slightly more bioavailable
(6%) than when dosed in a standard formulation. These
experiments indicated that in addition to low solubility,
the absorption of compounds from the series was hin-
dered by low permeability. This combination of low sol-
ubility with poor permeability constituted a significant
obstacle, one likely to impede the identification of an or-
ally bioavailable TPOr agonist from this series.
Table 6. SAR of substituted 4-aminopyrimdine thiadiazoles
CF3
N
S
Y
N
O
F
N
N
N
H
N
H
Compound
Y
EC50 (lM) clogP
90
H–
(CH3)2N–
0.41
4.73
5.64
4.78
4.90
5.66
5.19
4.61
4.88
4.62
In conclusion, we have described the identification and
development of a series of pyrimidine benzamides as
low molecular weight agonists of the TPOr. Our goals
for the series pertaining to potency (BaF3
EC50s < 50 nM) and the removal of undesired chemical
features of the lead molecule were achieved. Compounds
from the series were shown to induce proliferation of
human CD34+ progenitor cells into megakaryocytes.
105
106
107
108
109
110
111
112
0.043
0.030
0.22
HOCH2CH2N–
HOCH2CH2N(CH3)–
CH3OCH2CH2N(CH3)–
N-Azetidine
N-3-Azetidinol
0.13
0.081
0.013
0.071
(HOCH2CH2CH2)2N–
HOCH2CH(OH)N(CH3)N– 0.029

L. A. Reiter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5447–5454
5453
Method B—(1) the requisite amine was coupled to tert-
butyl 4-bromobenzoate by the method of Yin et al. Ref.
22; (2) TFA; (3) SOCl2; (4) reaction with requisite
thiazoleamine in pyridine; Method C—(1) the requisite
amine was reacted with tert-butyl 4-fluorobenzoate in the
presence of potassium tert-butoxide in DMF, (2) TFA, (3)
SOCl2, (4) reaction with requisite thiazoleamine in pyri-
dine; Method D—(1) the requisite chloride was coupled to
tert-butyl 4-aminobenzoate by the procedure of Yin et al.
(Ref. 22), (2) TFA, (3) SOCl2, (4) reaction with requisite
thiazoleamine in pyridine; Method E—(1) methyl 4-(pyr-
imidin-4-ylamino)benzoate was methylated with methyl
iodide and sodium hydride, (2) the methylated product
was condensed with the requisite thiazoleamine in a
trimethylaluminum-mediated reaction; Method F—(1)
4,6-dichloropyrimidine was coupled with methyl 4-
hydroxybenzoate in DMF in the presence of cesium
carbonate, (2) the ester was coupled with the requisite
thiazoleamine in a trimethylaluminum-mediated reaction,
(3) H2–Pd/C; Method G—(1) 4-bromo-N-(4-(2-fluoro-3-
(trifluoromethyl)phenyl)thiazol-2-yl)benzamide was alkyl-
ated with methyl iodide and sodium hydride, (2) the
methylated product was coupled to 4-aminopyrimidine by
the procedure of Yin et al. (Ref. 22); Method H—3-
bromo-N-(4-(2-fluoro-3-(trifluoromethyl)phenyl)thiazol-2-
yl)benzamide was coupled to 4-aminopyrimidine by the
procedure of Yin et al. (Ref. 22).
However, the physical properties of compounds in the
series posed
development.
a
significant obstacle to further
References and notes
1. Szilvassy, S. J. Expert Opin. Biol. Ther. 2006, 6, 983.
2. Duffy, K. J.; Darcy, M. G.; Delorme, E.; Dillon, S. B.;
Eppley, D. F.; Erickson-Miller, C. L.; Giampa, L.;
Hopson, C. B.; Huang, Y.; Keenan, R. K.; Lamb, P.;
Leong, L.; Liu, N.; Miller, S. G.; Price, A. T.; Rosen, J.;
Shah, R.; Shaw, A. N.; Smith, H.; Stark, K. C.; Tian,
S.-S.; Tyree, C.; Wiggall, K. J.; Zhang, L.; Luengo, J. I.
J. Med. Chem. 2001, 44, 3730.
3. Duffy, K. J.; Shaw, A. N.; Delorme, E.; Dillon, S. B.;
Erickson-Miller, C.; Giampa, L.; Huang, Y.; Keenan, R.
M.; Lamb, P.; Liu, N.; Miller, S. G.; Price, A. T.; Rosen,
J.; Smith, H.; Wiggall, K. J.; Zhang, L.; Luengo, J. I.
J. Med. Chem. 2002, 45, 3573.
4. Duffy, K. J.; Price, A. T.; Delorme, E.; Dillon, S. B.;
Duquenne, C.; Erickson-Miller, C.; Giampa, L.; Huang,
Y.; Keenan, R. M.; Lamb, P.; Liu, N.; Miller, S. G.;
Rosen, J.; Shaw, A. N.; Smith, H.; Wiggall, K. J.; Zhang,
L.; Luengo, J. I. J. Med. Chem. 2002, 45, 3576.
5. Erickson-Miller, C. L.; DeLorme, E.; Tian, S.-S.; Hopson,
C. B.; Stark, K.; Giampa, L.; Valoret, E. I.; Duffy, K. J.;
Luengo, J. L.; Rosen, J.; Miller, S. G.; Dillon, S. B.;
Lamb, P. Exp. Hematol. 2005, 33, 85.
6. Safonov, I. G.; Heerding, D. A.; Keenan, R. M.; Price, A.
T.; Erickson-Miller, C. L.; Hopson, C. B.; Levin, J. L.;
Lord, K. A.; Tapley, P. M. Bioorg. Med. Chem. Lett. 2006,
16, 1212.
7. Nakamura, T.; Miyakawa, Y.; Miyamura, A.; Yamane,
A.; Suzuki, H.; Ito, M.; Ohnishi, Y.; Ishiwata, N.; Ikeda,
Y.; Tsuruzoe, N. Blood 2006, 107, 4300.
8. Revill, P.; Serradell, N.; Bolos, J. Drugs Future 2006, 31,
767.
9. (a) Takemoto, H.; Takayama, M.; Shiota, T. Eur. Patent 1
207 155; (b) Eur. Patent 1 354 880.
10. Kalgutkar, A. S.; Gardner, I.; Obach, R. S.; Shaffer, C. L.;
Callegari, E.; Henne, K. R.; Mutlib, A. E.; Dalvie, D. K.;
Lee, J. S.; Nakai, Y.; O’Donnell, J. P.; Boer, J.; Harriman,
S. P. Curr. Drug Metab. 2005, 6, 161.
14. C-6 Amino substituted pyrimidines (Table 3) were
prepared in four steps: (1) 4,6-dichloropyrimidine was
coupled with 4-aminobenzoic acid in the presence of concd
hydrochloric acid in acetone, (2) SOCl2, (3) reaction with
requisite thiazoleamine in pyridine, (4) coupling with
requisite amine in NMP at 100 °C.
15. 2-Amino-5-arylthiazoles (Table 5, Column B) were pre-
pared in six steps: (1) 2-aminothiazole was protected with
di-tert-butyl dicarbonate, (2) the thiazole ring was bromi-
nated with bromine in potassium acetate buffered acetic
acid, (3) the carbamate was protected with 4-methoxyb-
enzylchloride in the presence of DBU, (4) the bromide was
coupled with the requisite arylboronic acid using
Pd(Ph3P)4 and Cs2CO3 in dioxane, (5) the amine protec-
tion was removed with neat refluxing TFA, (6) the
aminothiazole was acylated with 4-(pyrimidin-4-ylami-
no)benzoyl chloride.
16. 5-Amino-3-arylthiadiazoles (Table 5, Column C) were
prepared in four steps: (1) the requisite benzonitrile was
converted to the amidine with lithium hexamethyldisilaz-
ane according to the method of Thurkauf et al. (Ref. 23),
(2) the amidine was converted to the aminothiadiazole by
treatment with potassium thiocyanate and bromine
according to the procedure of Goerdeler et al. (Ref. 24),
(3) the aminothiadiazole was acylated in a trimethylalu-
minum-mediated reaction with methyl 4-(pyrimidin-4-
ylamino)benzoate.
17. 3-Amino-5-arylthiadiazoles (Table 5, Column D) were
prepared in four steps: (1) the amino group of 3-amino-5-
methylthio[1,2,4]thiadiazole was protected as the dimeth-
ylformamidine with dimethylformamide dimethylacetal,
(2) the aryl group was introduced using the requisite
arylboronic acid employing the procedure of Liebeskind
et al. (Ref. 25), (3) the amine protection was removed with
p-toluenesulfonic acid in methanol, (4) the aminothiadiaz-
ole was acylated in a trimethylaluminum-mediated reac-
tion with methyl 4-(pyrimidin-4-ylamino)benzoate.
11. Reiter, L. A.; Linde, R. G., II WO 2006033005.
12. Analogs of 2 (Table 1) were prepared by acylation of the
requisite 2-amino-4-arylthiazole with 4-(pyrimidin-4-
ylamino)benzoyl chloride or in a trimethylaluminum-
mediated reaction, with methyl 4-(pyrimidin-4-ylami-
no)benzoate.
O
S
N
X
Ar
+
N
H2N
N
N
H
X = Cl
pyridine
or
S
O
Ar
N
N
N
H
X = OCH3
Al(CH3)3
toluene
N
N
H
Amide formation by these methods was also a key step in
the preparation of other analogs.
13. The anilide and azine analogs (Table 2) were prepared by
the following methods: Method A—(1) N-(4-(2-fluoro-3-
(trifluoromethyl)phenyl)thiazol-2-yl)-4-iodobenzamide was
prepared by acylation of requisite thiazoleamine with 4-
iodobenzoyl chloride; (2) the requisite amine was coupled
to the iodide by the procedure of Yin et al. (Ref. 22);
18. 3-Amino-1-arylpyrazoles (Table 5, Column E) were
prepared in two steps: (1) the aminopyrazoles were
prepared from the requisite hydrazine by the procedure
of Ege et al. (Ref. 26), (2) the aminopyrazoles were
acylated with (pyrimidin-4-ylamino)benzoyl chloride.

5454
L. A. Reiter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5447–5454
19. 4-Amino-1-arylpyrazoles (Table 5, Column F) were
prepared in three steps: (1) the requisite 1-arylpyrazole
was nitrated by the procedure of Khan et al. (Ref. 27), (2)
the nitro group was reduced with ammonium formate in
the presence of Pd/C, (3) the aminopyrazole was acylated
with (pyrimidin-4-ylamino)benzoyl chloride.
with magnesium perchlorate according to the procedure of
Stafford et al. (Ref. 32), (7) the pyrimidine amine was
coupled to the iodide by the procedure of Yin et al. (Ref.
22).
22. Yin, J.; Zhao, M. M.; Huffman, M. A.; McNamara, J. M.
Org. Lett. 2002, 4, 3481.
20. 5-Amino-3-arylisothiazoles (Table 5, Column G) were
prepared in four steps: (1–3) the requisite 5-amino-3-
arylisothiazoles were prepared from the corresponding
benzonitriles by reaction with the anion of acetonitrile,
conversion to the thioamide with hydrogen sulfide and
then oxidation with iodine as described by Heckler (Ref.
28), (4) the aminoisothiazole was acylated with (pyrimidin-
4-ylamino)benzoyl chloride.
21. 4-Amino-2-aryloxazoles (Table 5, Column H) were pre-
pared in seven steps: (1–2) the requisite carboxylic acid
was converted to the corresponding oxazoline and then
oxidized by the method of Phillips et al. (Ref. 29), (3) the
oxazole ester was saponified under standard conditions as
described by Shafer et al. (Ref. 30), (4–5) a Curtius
rearrangement in tert-butanol and subsequent acylation
with 4-iodobenzoyl chloride was effected as described by
Redies et al. (Ref. 31), (6) the BOC group was removed
23. Thurkauf, A.; Hutchison, A.; Peterson, J.; Cornfield, L.;
Meade, R.; Huston, K.; Harris, K.; Ross, R. C.; Gerber,
K.; Ramabhadran, T. V. J. Med. Chem. 1995, 38, 2251.
24. Goerdeler, J.; Wember, K.; Worsch, G. Chem. Ber. 1954,
87, 57.
25. Liebeskind, L. S.; Srogl, J. Org. Lett. 2002, 4, 979.
26. Ege, G.; Hermann, F. J. Het. Chem. 1982, 19, 1265.
27. Khan, M. A.; Lynch, B. M.; Hung, Y.-Y. Can. J. Chem.
1963, 41, 1540.
28. Heckler, R. E. EP 0 129 407.
29. Phillips, A. J.; Uto, Y.; Reno, M. J.; Williams, D. R. Org.
Lett. 2000, 2, 1165.
30. Shafer, C. M.; Molinski, T. F. Heterocycles 2000, 53,
1167.
31. Redies, F.; Redies, B.; Turk, D.; Gille, Ch. DE 2 459 380.
32. Stafford, J. A.; Brackeen, M. F.; Karanewsky, D. S.;
Valvano, N. L. Tetrahedron Lett. 1993, 34, 7873.