Tyrosine kinase inhibitors. 16. 6,5,6-Tricyclic benzothieno[3,2-d]pyrimidines and pyrimido[5,4-b]- and -[4,5- b]indoles as potent inhibitors of the epidermal growth factor receptor tyrosine kinase

Article abstract of DOI:10.1021/jm9903949

Several elaborations of the fundamental anilinopyrimidine pharmacophore have been reported as potent and selective inhibitors of the epidermal growth factor receptor (EGFr) tyrosine kinase. This paper reports on a series of inhibitors whereby some 6,5-bicyclic heteroaromatic systems were fused through their C-2 and C-3 positions to this anilinopyrimidine pharmacophore. Although the resulting tricycles did not produce the enormous potency of some of the (5/6),6,6-bicyclic systems, the best of them had IC₅₀s for the EGFr TK around 1 nM. Investigation of 4-position side chains in the indolopyrimidines confirmed that m-bromoaniline was an optimal substituent for potency. Investigation of substitution within the C-(benzo)ring of benzothienopyrimidines confirmed that introduction of an extra ring can change sharply the effects of substituents when compared to similar bicyclic nuclei, and only two substituents were found which even moderately enhanced inhibitory activity over the parent compound for this series.

Full text of DOI:10.1021/jm9903949

5464
J . Med. Chem. 1999, 42, 5464-5474
Tyr osin e Kin a se In h ibitor s. 16. 6,5,6-Tr icyclic Ben zoth ien o[3,2-d ]p yr im id in es
a n d P yr im id o[5,4-b]- a n d -[4,5-b]in d oles a s P oten t In h ibitor s of th e Ep id er m a l
Gr ow th F a ctor Recep tor Tyr osin e Kin a se
H. D. Hollis Showalter,*,§ Alexander J . Bridges,§ Hairong Zhou,§ Anthony D. Sercel,§ Amy McMichael,# and
David W. Fry#
Departments of Chemistry and Cancer Research, Parke-Davis Pharmaceutical Research, Division of Warner-Lambert
Company, 2800 Plymouth Road, Ann Arbor, Michigan 48105
Received August 4, 1999
Several elaborations of the fundamental anilinopyrimidine pharmacophore have been reported
as potent and selective inhibitors of the epidermal growth factor receptor (EGFr) tyrosine kinase.
This paper reports on a series of inhibitors whereby some 6,5-bicyclic heteroaromatic systems
were fused through their C-2 and C-3 positions to this anilinopyrimidine pharmacophore.
Although the resulting tricycles did not produce the enormous potency of some of the (5/6),6,6-
bicyclic systems, the best of them had IC50s for the EGFr TK around 1 nM. Investigation of
4-position side chains in the indolopyrimidines confirmed that m-bromoaniline was an optimal
substituent for potency. Investigation of substitution within the C-(benzo)ring of benzothieno-
pyrimidines confirmed that introduction of an extra ring can change sharply the effects of
substituents when compared to similar bicyclic nuclei, and only two substituents were found
which even moderately enhanced inhibitory activity over the parent compound for this series.
In tr od u ction
the 5,6-positions of the pyrimidine leads to compounds
which can have good (∼50 nM)17 to truly outstanding
(<10 pM)18 IC50s for the enzyme as exemplified by
compounds 2-6 in Table 1, with examples 3 and 45
illustrating that a five-membered heteroaromatic ring
can be as effective as a phenyl ring.19 For the 6,6-
bicycles the 6- and 7-position substituents are often
highly activating18,20 (see 5 and 6). Furthermore, the
model suggests that bulky 6- and 7-substituents can
protrude from the binding site into the solvent and that
there is room for a further ring fused to the 6- and
7-positions of the bicycle. Such tricycles have been made
and can also be extremely potent and selective inhibi-
tors, as exemplified by the imidazoquinazoline 7a (8
pM),21 the pyrroloquinazoline 7b (440 pM), and the
pyrazoloquinazoline 7c (440 pM).21 Nonlinear tricycles
are considerably less potent than their linear isomers,
e.g., 8a ,b and 9,21 in accord with the bicyclic SAR, which
demonstrated that 6- and 7-position substitution can be
beneficial whereas 5- and 8-substitutions are deactivat-
ing.22
In this paper we wish to report on the SAR of a series
of 6,5,6-fused tricycles as EGFr inhibitors. Such com-
pounds have an inherent bend in the tricyclic nucleus,
but less than that seen in angularly fused (5/6),6,6-
tricycles. In accordance with the previous SAR,13,21 these
inhibitors tend to have inhibitory potencies intermediate
between the linear and angularly fused X,6,6-tricycles.
However, the effects of substitution on the C-benzenoid
ring tend to be quite different from their effects on the
B-phenyl ring of quinazolines.
Among the most common oncogenes that are overex-
pressed in such solid tumors as colon, breast, gastric,
prostatic, and head and neck cancers are those of the
type I receptor tyrosine kinase (RTK) family. These
consist of the epidermal growth factor receptor (EGFr),
erbB-2, erbB-3, and erbB-4 tyrosine kinases.1-6 As the
ability to induce tyrosine phosphorylation is essential
to the oncogenic potential of this family, there has been
an intensive search for inhibitors of its members.5,7-10
Although this field started slowly, the discovery of
highly selective, ATP-competitive inhibitors based on
diarylamines11 has led to the development of agents
with great pharmaceutical potential. The development
of such compounds has led to many compounds showing
oral activity in xenograft models, and at least two agents
have gone through phase I clinical trials.12 The most
explored pharmacophore is a 4-(meta-substituted anili-
no)pyrimidine, which can be modeled into the ATP-
binding site of EGFr in such a way that the pyrimidine
ring orients into the hydrophobic adenine-binding site
to form two backbone H-bonds in the same manner as
adenine.13-15 In the Parke-Davis model13 the aniline
ring forms a considerable dihedral angle to the pyrimi-
dine and occupies a hydrophobic cleft toward the back
of the ATP-binding site, which is unoccupied by sub-
strate. In accord with the model, some anilinopyrim-
idines such as 116 (Chart 1) show low nanomolar IC50s
toward the EGFr TK. In the model, the 5,6-bond of the
pyrimidine faces toward the outside of the catalytic cleft
and there is no particular encumbrance in that region.
Consistent with this, fusing a second aromatic ring to
Ch em istr y
* Correspondence: H. D. H. Showalter. Phone: 734-622-7028. Fax:
734-622-7879. E-mail: hollis.showalter@wl.com.
§ Department of Chemistry.
4-Amino-substituted pyrimido[5,4-b]indole analogues
were prepared as detailed in Scheme 1. Condensation
of 4-chloropyrimido[5,4-b]indole23 (11) was carried out
#
Department of Cancer Research.
10.1021/jm9903949 CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/09/1999

Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26 5465
Ch a r t 1
Sch em e 1a
Sch em e 2a
a
(i) RNH2 in an alcohol/cat. HCl (procedure A) or neat RNH2/
heat (procedure B); (ii) Cs2CO3, dimethyl sulfate, refluxing acetone,
3A sieves.
with a variety of aliphatic and aromatic amines to give
target compounds 34, 35, and 37-41 listed in Table 1.
The preferred conditions were to carry out the reaction
in a refluxing lower alcohol (and on occasion DMA) with
a catalytic amount of HCl (procedure A). In a few
instances, the reaction was run with neat amine at 150
°C (procedure B), although yields tended to be poorer.
Methylation of the N-5 position of 11 with cesium
carbonate in refluxing acetone followed by the addition
of dimethyl sulfate gave the known compound 12.24
Condensation of 12 with 3-bromoaniline then provided
the target analogue 36 of Table 1.
The preparation of a number of 4-amino-substituted
pyrimido[4,5-b]indole analogues is detailed in Scheme
2. Ring closure of either the known 2-amino-1H-indole-
3-carboxylic acid ethyl ester (13a )25 or the 6-methoxy
congener 13b to the corresponding pyrimido[4,5-b]indol-
4-ones 15a ,b, respectively, was performed in hot for-
mamide with sodium methoxide catalysis as previously
described.26 The synthesis of analogues bearing amino
or methyl substituents at the C-2 position was carried
out utilizing literature conditions described to make
similar congeners of intermediates 15c,d .27 Thus, con-
densation of 13a with cyanamide under acidic conditions
provided the guanidine 14c. Reaction of 14c with
refluxing aqueous sodium hydroxide resulted in ring
closure to 2-amino-1,9-dihydro-4H-pyrimido[4,5-b]indol-
a
(i) NH2CN, concd HCl, p-dioxane, reflux (for 14c), CH3CN, dry
HCl, reflux (for 14d ); (ii) aq NaOH, reflux; (iii) formamide, NaOMe,
220 °C (for 15a ,b); (iv) POCl3; (v) RNH2 in an alcohol/cat. HCl or
DMA/cat. HCl/120 °C (procedure A) or neat RNH2/heat (procedure
B); (vi) Cs2CO3, refluxing acetone or 2:1 acetone:DMF, 3A or 4A
sieves, dimethyl sulfate (for 17a ), Cl(CH2)2NEt2‚HCl (for 17b,c);
(vii) CNCH2CO2Et, t-BuOK, THF, reflux; (viii) Zn dust, acetic acid,
55 °C.
4-one (15c). Condensation of 13a with acetonitrile to
give amidine 14d followed by ring closure to 15d
proceeded similarly. Reaction of 15a -d with POCl3
either neat or in refluxing p-dioxane then provided the

5466 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26
Showalter et al.
Sch em e 3a
known 4-chloropyrimido[4,5-b]indole (16a )26 and 16b-
d , respectively. Chloropyrimido[4,5-b]indoles 16a -d
were then condensed with selected amines by the
procedures described above to give the target 4-amino-
substituted analogues (compounds 42-44 and 47-49
in Table 1). Condensation of 16b with 3-bromoaniline
proceeded poorly in refluxing 2-propanol/HCl, primarily
due to its lowered reactivity relative to 16a . However,
upon reaction in DMA/HCl at 120 °C, a modest yield of
target compound 49 (Table 1) was obtained. The 4-chlo-
ropyrimido[4,5-b]indole (16a ) was also N-methylated
under the conditions described above for isomeric com-
pound 11 to give the known 17a .28
To introduce aqueous solubilizing functionality onto
the N-9 position, chloro intermediates 16a ,b were
alkylated with 2-diethylaminoethyl chloride, either in
refluxing acetone or in 2:1 acetone:DMF, using cesium
carbonate as base to provide the corresponding N-
alkylated compounds 17b,c. Compounds 17a -c were
then condensed with 3-bromoaniline in refluxing 2-pro-
panol catalyzed with HCl (procedure A) to provide target
compounds 45, 46, and 50 of Table 1 in excellent yield.
The synthesis of 2-amino-5-methoxy-1H-indole-3-car-
boxylic acid ethyl ester (13b) was carried out utilizing
the same procedure as its 6-protio congener25 and is
outlined in Scheme 2. Condensation of 3-fluoro-4-
nitroanisole (18) with the potassium salt of ethyl cy-
anoacetate gave adduct 19 in quantitative yield. Zinc
dust reduction of 19, utilizing the literature conditions,26
gave a chromatographically separable 1:1 mixture of
13b and the N-hydroxy side product 20 resulting from
incomplete reduction.
The synthesis of benzothieno[3,2-d]pyrimidone 21a,29,30
and its various phenyl ring-substituted derivatives
(Scheme 3), has been described elsewhere.31 Generally
these were converted to the corresponding 4-chlorides
22 in good yields with Vilsmeier’s reagent (oxalyl
chloride/DMF), and the chlorides were then displaced
by amines in a refluxing lower alcohol to give the final
products (procedure A). The two amine-containing tri-
cyclic pyrimidones 21d ,e did not react successfully with
a variety of chlorinating agents. They were activated
by conversion into the corresponding 4-thiones with P2S5
or Lawesson’s reagent32 in diglyme at 115 °C and were
then converted in situ into the corresponding thioethers
23d ,e with methyl iodide and Hunig’s base. The thio-
ethers were surprisingly unreactive and could only be
displaced slowly with 3-bromoaniline under quite dras-
tic conditions in hot ethylene glycol in the presence of
excess aniline hydrochloride21 to give inhibitors 64 and
65 (procedure A). All of the other amine-containing
benzothienopyrimidines were made by Raney nickel
reductions of the corresponding nitro compounds as the
final step of the synthesis (procedure C), which allowed
the side chains to be introduced via the corresponding
chlorides without any problems. Methylation of the
8-amino functionality of 61 was carried out under
Eschweiler-Clarke conditions33 (procedure D) and led
to a mixture of mono- and dimethylation products, 67
and 68, which were separated chromatographically.
a
(i) DMF/oxalyl chloride, 1,2-dichloroethane; (ii) P2S5 or Lawes-
son’s reagent, diglyme, 115 °C, then (i-Pr)2NEt, MeI; (iii) RNH2
in ethylene glycol/cat. HCl, 140 °C (procedure A); (iv) RNH2 in an
alcohol/cat. HCl (procedure A); (v) H2, Raney Ni (procedure C);
(vi) HCO2H, aq HCHO, 90 °C (procedure D); (vii) ethyl thioglyco-
late, NaH, DMSO; (viii) formamide, 190 °C.
Sch em e 4a
a
(i) Methyl bromoacetate, K2CO3, acetone; (ii) NaH, DMSO; (iii)
formamide, 170 °Cl (iv) DMF/oxalyl chloride, 1,2-dichloroethane;
(v) 3-bromoaniline, 2-ethoxyethanol, 135 °C (procedure A).
Both the benzofuranopyrimidone 77 (Scheme 4) and
the pyrido[3,2′:4,5]thieno[3,2-d]pyrimidones 74 and 75
(Scheme 3) were made by a process analogous to the
benzothienopyrimidones. Alkylation of 2-cyanophenol
(29) with methyl bromoacetate followed by treatment
with sodium hydride and then cyclization in DMSO gave
the benzofuran 31 in moderate yield. Niemetowski

Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26 5467
Ta ble 1. Physicochemical and Enzyme Inhibitory Data for 6,5,6-Tricyclic Benzothienopyrimidines and Pyrimidoindoles
b
no.
form
4-X
R
procedure
formula
analysisa
IC50
Reference Compounds
1
2
3
4
5
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
see Figure 1
ref 16
ref 17
ref 5
1
23
11
35
0.025
0.008
0.008
0.39
0.44
29
ref 5
ref 18
ref 20
ref 21
ref 21
ref 21
ref 21
ref 21
ref 21
6
7a
7b
7c
8a
8b
9
1.24
272
Pyrimido[5,4-b]indoles
34
35
36
37
38
39
40
41
I
I
I
I
I
I
I
I
NHPh
H
H
A
A
A
B
B
B
B
B
C
C
C
C
C
C
16H12N4‚1.1HCl
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
2110
72
132
NH(3-BrPh)
NH(3-BrPh)
NHBn
NH(R)CH(Me)Ph
NH(S)CH(Me)Ph
NH-c-C6H11
16H11 BrN4‚1.1HCl
17H13BrN4‚HCl
17H14N4
5-Me
H
H
H
H
H
460
18H16N4‚0.8HCl
419
18H16N4‚0.1HCl
>105
>105
>105
C16H18N4‚0.2H2Oc
C14H17N5‚0.2HCl
NH(CH2)2NMe2
Pyrimido[4,5-b]indoles
42
43
44
45
46
47
48
49
50
II
II
II
II
II
II
II
II
II
NH(CH2)2NMe2
NHPh
H
H
H
9-Me
9-(CH2)2NEt2
2-Me
2-NH2
6-OMe
9-(CH2)2NEt2
6-OMe
B
A
A
A
A
A
A
A
A
C14H17N5‚0.1EtOAc
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
>104
264
31
C16H12N4
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
C
C
16H11BrN4
17H13BrN4‚0.7H2O
742
C22H24BrN5‚2HCl‚1.5H2O
4100
>104
147
1.2
4270
C
17H13BrN4‚1.1HCl
C16H12BrN5‚0.2H2O
17H13BrN4O‚0.7H2O
C23H26BrN5O‚2.1HCl‚0.9H2O
C
Benzothieno[3,2-d]pyrimidines
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
III
NHBn
H
H
H
A
A
A
A
A
A
A
C
C
C
C
A
C
A
A
A
D
D
A
A
C
A
A
C
C
C
17H13N3S
18H15N3S
16H10N3S
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
191
538
1.8
58
11
460
12.3
9.4
2.1
47
0.27
48
0.47
1.0
7.9
80
NH(R)CH(Me)Ph
NH(3-BrPh)
NHPh
NH(3-MePh)
NH(3-CF3Ph)
NH(3-BrPh)
NHPh
NH(3-MePh)
NH(3-CF3Ph)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
NH(3-BrPh)
8-NO2
8-NO2
8-NO2
8-NO2
8-NH2
8-NH2
8-NH2
8-NH2
7-NO2
7-NH2
7-NH2-8-F
7-NHEt-8-F
7-OMe
8-NHMe
8-NMe2
6-OMe
6-NO2
6-NH2
9-OMe
7,8-(OMe)2
C16H10N4O2S‚HCl
C17H12N4O2S‚0.5HCl
C17H9 F3N4O2S‚0.9HCl
C16H10N4O2S‚HCl
C16H12N4S‚0.1HCl‚0.2EtOH
C17H14N4S‚0.1H2O
C17H11F3N4S‚0.3H2O
C16H11BrN4S
C16H9BrN4O2S‚HCl
C16H11BrN4S‚0.5HCl
C16H10BrFN4S
C
C
C
18H14BrFN4S‚0.3EtOH
17H12BrN3OS
17H13BrN4S‚0.5EtOH
1.1
C18H15BrN4S
C
19.8
12.6
158
2.7
17H12BrN3OS‚HCl
C16H9BrN4O2S‚HCl
C16H11BrN4S‚0.3HCl‚0.7MeOH
C
17H12BrN3OS‚HCl
∼20d
444
C18H14BrN3O2S‚HCl
Pyridothieno[3,2-d]pyrimidines
74
75
IV
IV
NH(3-ClPh)
NH(3-BrPh)
H
H
A
A
C
15H9ClN4S
C, H, N
C, H, N
>103
732
C15H9ClN4S
Thiazolothieno[3,2-d]pyrimidines
13H7BrN4S2‚0.1CHCl3‚0.2MeOH
76
V
NH(3-BrPh)
H
A
C
C, H, N
40
Benzofurano[3,2-d]pyrimidines
16H10BrN3O
77
VI
NH(3-BrPh)
H
A
C
C, H, N
740
a
b
The analyses were within (0.4% of the theoretical values. IC50: concentration of compound (nM) to inhibit the phosphorylation of
14-residue fragment of phospholipase C-γ1 by EGFr (prepared from human A431 carcinoma cell vesicles by immunoaffinity
a
chromatography). See Experimental Section for details. Values are the averages from at least two independent dose-response curves;
variation was generally (15%. c N: calcd, 20.75%; found, 20.30%. The low solubility of this compound precluded an accurate IC50
d
determination.

5468 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26
Showalter et al.
synthesis made the desired tricycle 32 in good yield
(Scheme 4). Similarly, treatment of 2-chloronicotinoni-
trile (24) with sodium ethyl thioglycolate followed by
formamide cyclization gave 26. Both were chlorinated
with the oxalyl chloride-derived Vilsmeier reagent and
displaced with 3-bromoaniline without difficulty.
One 5,5,6-tricycle was prepared. 4-Chlorothiazolo-
[2′,3′:4,5]thieno[3,2-d]pyrimidine (28) was prepared by
the method of Iddon,34 and the 4-chloro substituent was
readily displaced with 3-bromoaniline to give the desired
tricycle 76 (Scheme 3).
76 was also unsurprising. The isomeric indolo[2,3-d]-
pyrimidine 44 is twice as potent as 35, suggesting that
this ring fusion may be the more advantageous, which
would be consistent both with the demonstrated greater
tolerance of substituents at the 5- rather than the
8-position in quinazolines and in the angular imidazo-
quinazolines studied previously. Unfortunately, at-
tempts at developing an adaptable synthesis of benzo
ring-substituted benzothieno[2,3-d]pyrimidines35 were
unsuccessful, thus not permitting meaningful explora-
tion of that SAR.
A few compounds confirmed that substitutions in the
A- and B-rings were similar to the quinazolines.22
Substitutions at the 2-position of the indolo[2,3-d]-
pyrimidines 47 and 48 were detrimental with methyl
leading to at least a 300-fold loss of activity and amino
a more modest 5-fold loss. On the same nucleus 9-N
substitution, corresponding to the highly disfavored
8-position of quinazolines, with a methyl group (45) or
a solubilizing diethylaminoethyl side chain (50) led to
15- and 138-fold losses of potency with respect to the
corresponding 9-protio compound 44. In the indolo[3,2-
d]pyrimidine series, N-methylation of the 5-position of
compound 36 led to only a 2-fold loss of activity with
respect to the 5-protio compound 35. In the quinazolines
methoxylation of this position led to about a 25-fold loss
of potency.
Resu lts a n d Discu ssion
The structures and physicochemical properties of the
compounds prepared are given in Table 1. All of the
analogues were evaluated for their ability to inhibit
tyrosine phosphorylation of a polypeptide (a portion of
phospholipase C-γ) by EGF-stimulated full-length EGFr
enzyme isolated from A431 cells. Full dose-response
curves were determined for each compound, and the
resulting IC50s listed in Table 1 are the average of at
least two such determinations.
As can be seen in Table 1 many potent inhibitors of
the EGFr TK can be found in these 6,5,6-tricycles, and
at a first level the SAR is clearly parallel to the previous
series that we have examined.22 Aromatic amine side
chains at the 4-position usually produce sub-micromolar
inhibitors, with many of the better ones being low
nanomolar. Nonaromatic amine side chains lead, as
anticipated, to very poor inhibitors, and in the limited
SAR done m-bromoanilino appears to be the best
substituent, followed by m-methylanilino. The two m-
bromoaniline derivatives with IC50s greater than 1 µM,
47 and 50, contain a 2-methyl and a bulky 9-substituent
(equivalent to the 8-position in a quinazoline), respec-
tively, neither of which would be tolerated in the
corresponding bicyclic SARs. Thus, from what can be
predicted from previous SARs, including the other
tricyclic SAR examined,21 tolerance at these key posi-
tions and the beneficial effect of small lipophilic m-
anilino substituents seem to hold up in these compounds
as well, suggesting that they may well bind in a similar
mode to the well-studied bicyclic compounds. This would
be in accord with molecular modeling, which suggests
that the same binding mode is tolerated for these
tricycles, the X,6,6-tricycles, and the bicycles. However,
there are some differences from the bicyclic SARs, which
will be discussed with the individual tricyclic nuclei
below.
Looking at the aromatic nuclei, the benzothieno[3,2-
d]pyrimidine nucleus is clearly the best of those in this
study. Considered as a bare “parent” nucleus, the
benzothiophene 53 is among the most potent that we
observed and is 40-fold more potent than the corre-
sponding indolo[3,2-d]pyrimidine 35 and surprisingly
400-fold more potent than the corresponding benzofuran
77. As previous SAR studies had shown that the enzyme
appears to tolerate electron richness better than electron
deficiency in the B/C-ring region, and 6/7-substitution
better than 5/8-substitution, the potency of the ben-
zothienopyrimidines is unremarkable. The lack of po-
tency of the more electron-deficient pyridothienopyri-
midines 74 and 75 was not unexpected (see below), and
the reasonable potency of the thiazolothienopyrimidine
A brief examination of the 4-amino substituent con-
firmed that these compounds fit the SAR of the previ-
ously examined anilinopyrimidines. In the indolo[3,2-
d]pyrimidines, two nonaromatic side chains, 40 and 41,
showed no activity at all, whereas a benzyl side chain,
compound 37, gives moderate activity (460 nM). In this
case the (R)-methyl substituent (compound 38) is not
significantly better, but the (S)-methyl substituent
(compound 39) is strongly deactivating, as had been
seen for the quinazolines.36 A simple phenyl substituent
(compound 34) was surprisingly poor in this series (2.11
µM), but the m-bromo substituent (compound 35) pro-
duced the usual activation and, with an IC50 of 72 nM,
was the most potent member of this series. In the
isomeric indolo[2,3-d]pyrimidine series, similar trends
were seen. The m-bromoanilino derivative 44 was about
8.5-fold more potent than the corresponding phenyl
compound 43, and the nonaromatic side chain (com-
pound 42) had no activity. In the benzothieno[3,2-d]-
pyrimidine series, benzyl (compound 51) was a mediocre
inhibitor, and here the (R)-methyl substituent (com-
pound 52) was about 3-fold detrimental, as opposed to
moderately activating in the quinazolines. Here, the
m-bromoanilino side chain of 53 led to excellent activity
(1.8 nM). With the deactivating 8-nitro and the activat-
ing 8-amino substituents on this nucleus, a small series
of m-anilino substitutions were examined and once
again showed similarity with previously described
SARs. The less activated nitro series showed CH3 ∼ Br
> H > CF3 (compounds 54-57) and the more activated
amino series Br > CH3 > H > CF3 (compounds 58-
61), which is very similar to the parent quinazoline
series.
In the benzothieno[3,2-d]pyrimidine series, the effects
of C-ring (benzo) substitution were examined to see if
they are similar to the bicyclic series where electron-
donating substituents in the B-ring are potentiating at

Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26 5469
Ta ble 2. Inhibition of EGFr Autophosphorylation in A431
Cells
the 6- and 7-positions. As these heterocycles are both
longer than the bicycles and have a bent geometry,
which was strongly disfavored in the 5,6,6-tricyclic
series, there was no guarantee that substituent effects
would be similar although it was anticipated that the
usual preference for electron-donating substituents
would be found. A series of methoxy substitutions were
introduced on the phenyl ring, all with the m-bromo-
anilino side chain. Both 6- and 9-methoxy-substituted
analogues 69 and 72 showed a moderate loss (about 10-
fold) in potency compared to 53, but the 7-substituted
compound 66 surprisingly showed a 40-fold loss. The
7,8-dimethoxy derivative 73 showed an even larger loss
of affinity (250-fold). Nitro and amino substitutions at
the 6-, 7-, and 8-positions were also examined, as they
have large effects in the bicyclic series. The 6-nitro
compound 70 was strongly deactivated (88-fold with
respect to 53), the 7-nitro isomer 62 was less so (32-
fold), and the 8-isomer 57 was only 7-fold deactivated.
The same trend showed up in the amines, with the
6-isomer 71 being essentially equipotent to the parent
compound 53, and the 7- and 8-isomers 63 and 61 were
4- and 7-fold, respectively, more potent than 53. With
an IC50 of 270 pM, amine 61 was the most potent
inhibitor that we found in these classes of tricycles.
Removal of the m-bromine led to 5- and 35-fold losses
of activity, respectively, for the corresponding 8-nitro
and 8-amino compounds, 54 and 58. An 8-F substituent
in combination with the 7-amino (compound 64) was
slightly deactivating. Simple alkylation of the amine at
either the 7- or 8-position was moderately deactivating
(65, 67, and 68).
IC50 (nM)a
compd
enzyme
autophos
35
44
45
48
49
53
59
61
64
65
67
72
31
93
630
769
493
13
170
98
86
474
147
1.2
1.8
2.1
0.27
1.0
7.9
1.1
42
85
33
a
Data values represent the average of two independent experi-
ments with a variability of 25% or less.
pyrimidine ring the steric intolerances are less. One
methoxy-substituted indolo[2,3-d]pyrimidine, 49, was
made, and in this case the substitution was highly
favored with a 26-fold improvement in potency giving
a 1.2 nM inhibitor, suggesting the possibility that the
alternative ring fusion might lead to a more bicycle-like
SAR in this ring.
A few aniline substituents, which have proved ad-
vantageous in other series, notably m-methyl and m-
trifluoromethyl, were looked at with the 8-nitro and
8-amino compounds. Methyl proved to be as good as
bromine in the nitro compound 55, but 10-fold poorer
in the amine 59, whereas trifluoromethyl was 5-8-fold
deactivating in the case of both 56 and 60. These results
have parallels elsewhere in the SAR.
As shown in Table 2 several of these compounds had
good activity against cellular autophosphorylation. Com-
pound 49 was the most potent, with only a 10-fold fall
off in potency from that observed in the enzyme assay.
This compound is as potent in cellular assays as the
picomolar bicyclic compounds 5 and 6, but this result
was exceptional. In contrast, the most potent of these
tricycles, compound 61, had a 320-fold fall off in potency
in the cellular assay and had a disappointing 86 nM
IC50, whereas compound 35 was nearly equipotent in
enzyme and cellular assays. There does not appear to
be a good correlation between enzymatic and cellular
data in this subset, although we observed autophos-
phorylation IC50 values of <100 nM for inhibitors with
IC50 values of <10 nM against the enzyme. The great
insolubility of many of these tricycles presumably affects
the cellular data more than the enzyme data and thus
may render the results somewhat ambiguous.
Previous series in this SAR had shown excellent
selectivity for EGFr inhibition over all of the non-erbB
kinases examined, and these tricycles proved to be no
exception, as demonstrated in Table 3. As the selectivity
of this family appears to come from occupancy of a
hydrophobic pocket behind the adenine binding site by
the N-4 aromatic side chain, while the fused rings of
the heteroaromatic nucleus point toward the solvent,
these compounds should have similar poor affinity for
most other kinases, as has been demonstrated for all of
the other inhibitors containing the ar(alk)ylamino-
pyrimidine pharmacophore.
Clearly these results do not parallel the quinazoline
SAR and also do not open themselves up to a simple
electron density explanation. However, low electron
density in this ring does appear to be somewhat
disfavored, with the 8-nitro compound being only 7-fold
deactivated. The 7-nitro isomer 62 shows a deactivation
expected by analogy with the quinazoline SAR. The
6-isomer 70 is more deactivating than expected, but
steric effects could easily play a part at this position.
However, replacing the nitro at this position with an
aza substituent (pyridothienopyrimidine 75) led to even
greater deactivation although results with other het-
erocyclic nuclei suggest that factors other than simple
electron density may come into play when the nucleus
is altered. The 7- and 8-amino substituents suggest at
best a weak activating effect of very electron-rich rings,
with 6-amino being without significant effect. Even
monoalkylation abolishes this effect, as is the case with
the quinazolines. The methoxy substitutions examined
are the most puzzling, with the 6- and 9-positions, as
expected for steric reasons, being moderately deactivat-
ing, but the 7- and 8-position methoxy analogues
examined were highly deactivated. It is difficult to
attribute these results to steric factors, but they are
strongly suggestive that simply increasing the electron
density of the benzo ring is not useful for increasing
affinity and that the modest activation seen with the
7- and 8-amino substituents may be due to some
nonelectronic factor such as hydrogen bonding. The
minor losses of affinity in the 6- and 9-positions relative
to that seen with 5- and 8-substitution in the quinazo-
lines are suggestive that further away from the
Compound 61 was examined in two in vivo models,
using EGFr-transfected fibroblasts and MDA MB-468
human breast tumor cells, respectively, as ip xenografts

5470 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26
Ta ble 3. Selectivity of Inhibitors Against Other Kinases
Showalter et al.
at 25 °C for 36 h. The suspension was filtered over Celite and
the filtrate was concentrated to a solid that was distributed
between CHCl3 and 1% aq NaOH. The organic layer was
washed with H2O (2×), dried, and concentrated to a solid that
was purified by column chromatography eluting with 0% and
then 1% EtOH in CHCl3. The pooled product fractions were
concentrated to a residue that was crystallized to give 12 (278
mg, 35%): mp 153-154 °C (lit.23 mp 145-146 °C).
% inhibition @ 50 µM
IC50 (nM)
compd
EGFr
c-Src
FGFr IR PDGFr PKC
49
51
52
53
59
60
61
66
1.2
191
538
1.8
2.1
47
8 @ 1.25 µM
30
11
20
21
24
17
ND
19
0
0
13
0
14
0
2
10
0
ND
ND
7
18
44
51
0
0
0
0
47
13
ND
22
0
17
12
ND
21
2-Am in o-5-m eth oxy-1H-in d ole-3-ca r boxylic Acid Eth yl
Ester (13b) a n d 2-Am in o-1-h yd r oxy-5-m eth oxy-1H-in -
d ole-3-ca r boxylic Acid Eth yl Ester (20). To an ice-cold
solution of ethyl cyanoacetate (10.9 mL, 102.4 mmol) in
anhydrous THF (170 mL) under N2 was added potassium tert-
butoxide (12.07 g, 107.5 mmol). The formed white suspension
was stirred for 15 min then treated with 3-fluoro-4-nitroanisole
(18)37 (8.86 g, 51.2 mmol). The suspension was heated at reflux
for 1.5 h. The solution was poured into H2O, and the aqueous
mixture was acidified to pH 2 with concentrated HCl. The
mixture was extracted with ether (3×) and then the combined
organic phases were dried and concentrated to give an oil that
was evacuated at 0.3 mm for 2 days. The oil was dissolved in
CH2Cl2 and purified by column chromatography eluting with
CH2Cl2. The product fractions were combined and concentrated
to provide cyano(5-methoxy-2-nitrophenyl)acetic acid ethyl
ester (19) (14.5 g) as a light yellow oil that was ∼93-95%
pure: 1H NMR (CDCl3) δ 8.29 (1H, d, J ) 9.2 Hz), 7.22 (1H,
d, J ) 2.7 Hz), 7.04 (1H, dd, J ) 9.2, 2.7 Hz), 5.69 (1H, s),
4.31 (2H, q, J ) 7.0 Hz), 1.34 (3H, t, J ) 7.2 Hz). The material
was used directly in the next step.
0.27
80
0
0
20
in nude mice. The low solubility of 61 (<30 µg/mL)
proved to be very problematical, and even when dosed
as an ip suspension (50 mg/kg in pH 4 lactate buffer
containing 10% DMA) the compound proved completely
inefficacious. The peak blood levels were measured in
the low nanomolar range, considerably below that
predicted to be required for efficacy from the cellular
data. Since this was so far removed from our target of
acceptable oral bioavailability, and since none of the
other compounds examined in this SAR showed a
significant enough amelioration of physicochemical
properties to expect any better results, no further in vivo
work was carried out in this series.
Con clu sion s
A solution of impure 19 (13.2 g, ∼46.3 mmol) in glacial
AcOH (185 mL) was treated with a single charge of Zn dust
(12.1 g, 185 mmol). The mixture was heated at 55 °C for 45
min, then treated with more Zn (4 g). After heating for another
105 min, the brown mixture was filtered through a pad of SiO2.
The pad was washed well with AcOH and the filtrate was
concentrated to a residue that was distributed between CH2-
Cl2 and H2O. The organic phase was washed with 5% aq
NaHCO3 and concentrated to a residue that showed a ∼1:1
mixture of products by TLC (3:1 CH2Cl2/EtOAc). The residue
was purified by column chromatography eluting sequentially
with 0%, 5%, and 10% EtOAc in CH2Cl2. The fractions
containing the pure higher Rf product were combined and
concentrated to a solid that was sonicated in tert-butyl methyl
ether. The solid was collected to give pure 13b (2.07 g, 19%)
as an off-white solid: mp 149-151 °C. Further chromatogra-
phy of the combined mother liquor and impure fractions
afforded additional 13b (120 mg, 1%): 1H NMR δ 10.44 (1H,
br s, exchanges with D2O), 7.11 (1H, d, J ) 2.2 Hz), 6.98 (1H,
d, J ) 8.4 Hz), 6.61 (2H, br s, exchanges with D2O), 6.48 (1H,
dd, J ) 8.4, 2.7 Hz), 4.20 (2H, q, J ) 7.0 Hz), 3.71 (3H, s),
1.32 (3H, t, J ) 7.2 Hz); CIMS m/z (relative intensity) 235
(MH+, 81), 234 (M+, 100). Anal. (C12H14N2O3) C, H, N.
Processing of the pure cuts from the lower Rf fractions
provided 20 (278 mg) following crystallization from EtOAc: mp
163-164 °C; 1H NMR δ 11.11 (1H, br s, exchanges with D2O),
7.18 (s, 1H), 7.00 (1H, d, J ) 8.4 Hz), 6.75 (2H, br s, exchanges
with D2O), 6.57 (1H, dd, J ) 8.4, 2.4 Hz), 4.21 (2H, q, J ) 7.0
Hz), 3.73 (3H, s), 1.32 (3H, t, J ) 7.1 Hz); CIMS m/z (relative
intensity) 250 (M+, 10). Anal. (C12H14N2O4) C, H, N.
2-[(Am in oim in om eth yl)a m in o]-1H-in d ole-3-ca r boxyl-
ic Acid Eth yl Ester , Hyd r och lor id e (14c). A suspension of
2-amino-1H-indole-3-carboxylic acid ethyl ester (13a )25 (2.04
g, 10.0 mmol), cyanamide (534 mg, 12.7 mmol), and concen-
trated HCl (1 mL) in dioxane (91 mL) was heated at reflux for
48 h. After the reaction mixture had cooled to 25 °C, it was
filtered and the solid was washed well with dry diethyl ether
and then air-dried to give 14c (1.08 g, 38%) as an off-white
solid: mp >250 °C. The compound was used directly in the
next step.
An expansion of the basic ATP-binding site competi-
tive EGFr TK inhibitor pharmacophore of 4-anilino-
pyrimidine to 6,5,6-tricyclic pyrimidines has been
achieved. Generally, the best compounds have potencies
between those of linear and angular X,6,6-tricycles in
the same SAR, as might be expected. The nature of the
heteroatom introduced into the five-membered ring is
surprisingly important. In the [3,2-d] series oxygen was
strongly (27-fold) deactivating with respect to the cor-
responding bromoanilinoquinazoline, whereas nitrogen
was mildly deactivating and sulfur was strongly acti-
vating (15-fold). In the pyrimidoindoles, the only series
where different ring fusions were studied, the [2,3-d]
fusion appears to be about twice as potent as the [3,2-
d] ring fusion. The SAR of anilino substitution is similar
to that of the quinazolines, and substitution at the 2-
and 9-positions ([2,3-d] fusion), corresponding to the
disfavored positions of the quinazoline (2 and 8), is
detrimental, whereas substitution at C-6, corresponding
to C-5 of the quinazolines, is mildly activating. This
suggests that these tricyclic inhibitors can bind in the
same or a very similar mode to the quinazolines to the
EGFr ATP-binding site. No highly activating substitu-
ent was found for the tricyclic nuclei, and although the
best compounds had sub-nanomolar IC50s against EGFr,
this activity was not sufficient to compensate for their
very poor physical properties. Cellular activity was often
much less than would have been predicted from the
bicyclic series, and the one compound examined in vivo
had plasma levels far below predicted efficacy levels
even with ip dosing.
Exp er im en ta l Section
Gen er a l Meth od s. See Supporting Information.
4-Ch lor o-5-m eth yl-5H-pyr im ido[5,4-b]in dole (12). A mix-
ture of 4-chloro-5H-pyrimido[5,4-b]indole hydrochloride23 (11)
(731 mg, 3.6 mmol), dimethyl sulfate (0.41 mL, 4.3 mmol), Cs2-
CO3 (3.5 g, 10.7 mmol), activated 3A molecular sieves (2.7 g),
and acetone (10 mL) was heated at reflux for 6 h, then stirred
2-[(1-Im in oeth yl)a m in o]-1H-in d ole-3-ca r boxylic Acid
Eth yl Ester , Hyd r och lor id e (14d ). Dry HCl was bubbled
into a suspension of 2-amino-1H-indole-3-carboxylic acid ethyl
ester (13a )25 (1.0 g, 4.9 mmol) and CH3CN (65 mL) for 1.5 h.
The mixture was heated at reflux for 2.5 h and then cooled to

Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26 5471
25 °C. The precipitate was collected and dried to give 14d (1.02
g, 74%): mp >300 °C; H NMR δ 12.63 (1H, br s, exchanges
with D2O), 11.89 (1H, br s, exchanges with D2O), 9.95 (1H, br
s, exchanges with D2O), 9.0 (1H, br s, exchanges with D2O),
8.26 (1H, d, J ) 7.2 Hz), 7.49 (1H, d, J ) 7.7 Hz), 7.33-7.24
(2H, m), 4.29 (2H, q, J ) 7.0 Hz), 2.40 (3H, br s), 1.34 (3H, t,
J ) 7.2 Hz). Anal. (C13H15N3O2‚1.2HCl) C, H, N.
4-Ch lor o-1H -p yr im id o[4,5-b]in d ole-2-a m in e H yd r o-
ch lor id e (16c). A suspension of 2-amino-1,9-dihydro-4H-
pyrimido[4,5-b]indol-4-one (15c) (490 mg, 2.5 mmol) and POCl3
(7 mL, 75 mmol) in p-dioxane (13 mL) was heated at reflux
for 4 h, then concentrated. The residue was triturated in EtOH
and the solid was collected to give 16c (170 mg, 27%): mp
>250 °C. The material was used directly in the next step.
1
4-Ch lor o-9-m et h yl-9H -p yr im id o[4,5-b]in d ole (17a ). A
mixture of 4-chloro-1H-pyrimido[4,5-b]indole (16a ) (610 mg,
3 mmol), dimethyl sulfate (0.34 mL, 3.6 mmol), Cs2CO3 (2.93
g, 9 mmol), activated 3A molecular sieves (2.4 g), and acetone
(9 mL) was heated at reflux for 1.5 h. The suspension was
filtered through Celite, and the filtrate was concentrated to a
solid that was dissolved in CHCl3. The solution was washed
with H2O (2×), dried, and concentrated to a solid that was
crystallized from 1:1 EtOAc:petroleum ether. The solid was
collected to give 17a (285 mg, 44%): mp 129-130 °C (lit.28
mp 128-130 °C); 1H NMR (CDCl3) δ 8.80 (1H, s), 8.41 (1H, d,
J ) 8.0 Hz), 7.68-7.64 (1H, m) 7.54 (1H, d, J ) 8.2 Hz), 7.48-
7.44 (1H, m), 1.62 (3H, br s); CIMS m/z (relative intensity)
220 (37Cl, MH+, 34), 219 (37Cl, M+, 24), 218 (35Cl, MH+, 100),
217 (35Cl, M+, 41).
4-Ch lor o-N ,N -d ie t h yl-9H -p yr im id o[4,5-b]in d ole -9-
eth a n a m in e (17b). A suspension of 4-chloro-1H-pyrimido[4,5-
b]indole hydrochloride (16a ) (407 mg, 2 mmol), 2-diethylami-
noethyl chloride hydrochloride (413 mg, 2.4 mmol), Cs2CO3
(1.95 g, 6 mmol), and 4A molecular sieves (1.5 g) in acetone (6
mL) was heated at reflux under N2 for 1.5 h. The mixture was
filtered through Celite, and the filtrate was concentrated to a
viscous oil that was dissolved in CH2Cl2. The solution was
washed with H2O, dried, and concentrated to a residue that
was purified by column chromatography, eluting with 4%
MeOH in CHCl3, to give 17b (495 mg, 82%) as a pale yellow
oil: 1H NMR δ 8.79 (1H, s), 8.41 (1H, d, J ) 8.0 Hz), 7.66-
7.58 (2H, m), 7.46-7.42 (1H, m), 4.57 (2H, t, J ) 6.8 Hz), 2.90
(2H, t, J ) 7.1 Hz), 2.63 (4H, d, J ) 7.0 Hz), 0.99 (6H, t, J )
7.0 Hz); CIMS m/z (relative intensity) 305 (37Cl, MH+, 16), 304
(37Cl, M+, 9), 303 (35Cl, MH+, 47).
4-Ch lor o-N,N-d ieth yl-6-m eth oxy-9H-p yr im id o[4,5-b]in -
dole-9-eth an am in e (17c). Reaction of a suspension of 4-chloro-
6-methoxy-1H-pyrimido[4,5-b]indole (16b) (773 mg, 2.5 mmol,
∼80% pure), 2-diethylaminoethyl chloride hydrochloride (582
mg, 3.4 mmol), Cs2CO3 (2.3 g, 7.1 mmol), 4A molecular sieves
(2.1 g), and 2:1 acetone:DMF (12 mL) was refluxed for 16.5 h
as described for 17b. Similar workup followed by column
chromatography, eluting with 0% and then 2% MeOH in CH2-
Cl2, gave 17c (667 mg, 80%) as a yellow oil: 1H NMR (CDCl3)
δ 8.75 (1H, s), 7.87 (1H, d, J ) 2.4 Hz), 7.47 (1H, d, J ) 8.9
Hz), 7.25 (1H, dd, J ) 8.9, 2.4 Hz), 4.50 (2H, t, J ) 7.2 Hz),
3.96 (3H, s), 2.86 (2H, t, J ) 7.1 Hz), 2.59 (4H, q, J ) 7.1 Hz),
0.96 (6H, t, J ) 7.1 Hz); CIMS m/z (relative intensity) 335
(37Cl, MH+, 7), 333 (35Cl, MH+, 22). Anal. (C17H21N4ClO) C, H,
N.
6-Nitr oben zo[b]th ien o[3,2-d ]p yr im id -4(3H)-on e (21h ).
A mixture of methyl 3-amino-7-nitrobenzothiophene-2-car-
boxylate (242 mg, 0.96 mmol) and formamidine acetate (0.51
g, 4.9 mmol) was heated at 190 °C for 1 h. After cooling, the
mixture was slurried in H2O, and the formed precipitate was
collected and dried to give 21h (162 mg, 68%) as a brown
solid: 1H NMR δ 8.67 (d, 2H, J ) 8.1 Hz, H-7,H-9), 8.39 (s,
1H, H-2), 7.85 (t, 1H, J ) 7.8 Hz, H-8). The material was used
directly in the next step.
1,9-Dih yd r o-6-m e t h oxy-4H -p yr im id o[4,5-b]in d ol-4-
on e (15b). A solution of 2-amino-5-methoxy-1H-indole-3-
carboxylic acid ethyl ester (13b) (2.15 g, 9.2 mmol), NaOMe
(0.5 g, 9.3 mmol), and formamide (200 mL) was heated under
N2 at 220 °C for 1.5 h. The solution was cooled to 25 °C, stored
for 2.5 days, and filtered. The solvent was evaporated by
Kugelrohr distillation at 95 °C/0.8 mmHg to leave a residue
that was triturated in H2O and then heated in 35 mL of boiling
DMF. The suspension was filtered hot over a pad of SiO2, and
the filtrate was concentrated to a solid that was sonicated in
∼30 mL of MeOH. The solid was collected and dried to leave
15b (1.71 g, 72%) that was ∼83% pure: 1H NMR δ 12.16 (1H,
br s, exchanges with D2O), 12.04 (1H, br s, exchanges with
D2O), 8.08 (1H, d, J ) 3.4 Hz, exchanges to s with D2O), 7.46
(1H, d, J ) 1.9 Hz), 7.37 (1H, d, J ) 8.7 Hz), 6.95 (1H, dd, J
) 8.8, 2.5 Hz), 3.81 (s, 3 H); CIMS m/z (relative intensity) 216
(MH+, 100), 215 (M+, 52). The material was used directly in
the next step.
2-Am in o-1,9-d ih yd r o-4H -p yr im id o[4,5-b]in d ol-4-on e
(15c). A mixture 2-[(aminoiminomethyl)amino]-1H-indole-3-
carboxylic acid ethyl ester hydrochloride (14c) (1.0 g, 3.5 mmol)
and NaOH (1.5 g) in H2O (50 mL) was heated at gentle reflux
for 6 h. The mixture was acidified to pH 1 with 5% aq HCl,
and filtered through Celite. The filtrate was extracted with
EtOAc (3×) and then made basic with Na2CO3. The precipitate
that slowly formed was collected and dried to leave 15c (561
mg, 78%) as light tan crystals: mp >275 °C. The material was
used directly in the next step.
1,9-Dih yd r o-2-m et h yl-4H -p yr im id o[4,5-b]in d ol-4-on e
(15d ). A solution of 2-[(1-iminoethyl)amino]-1H-indole-3-car-
boxylic acid ethyl ester hydrochloride (14d ) (1.0 g, 3.6 mmol),
EtOH (300 mL), and 10% aq NaOH (30 mL) was heated at
reflux for 6 h. The cooled solution was concentrated to a
residue that was diluted with 20 mL of H2O. The solution was
adjusted to pH 2 with aq HCl then stored at 25 °C for 30 min.
The precipitated solid was collected and dried to leave 15d
(618 mg, 87%): mp >275 °C; 1H NMR δ 12.11 (1H, br s,
exchanges with D2O), 11.97 (1H, br s, exchanges with D2O),
7,93 (1H, d, J ) 7.7 Hz), 7.43 (1H, d, J ) 7.9 Hz), 7.28 (1H, t,
J ) 7.5 Hz), 7.20 (1H, t, J ) 7.7 Hz), 2.41 (s, 3 H); CIMS m/z
(relative intensity) 200 (MH+, 100), 199 (M+, 85). The material
was used directly in the next step.
4-Ch lor o-1H-pyr im ido[4,5-b]in dole Hydr och lor ide (16a).
A mixture of 9H-pyrimido[4,5-b]indol-4-ol26 (15a ) (5.3 g, 28.6
mmol), POCl3 (50 mL), and p-dioxane (125 mL) was heated at
reflux for 6 h, then stirred at 25 °C for 36 h. The mixture was
filtered and the filtrate was concentrated to an oil that
solidified. The solid was triturated in hot EtOH and collected
to give 16a (3.1 g, 45%) in two crops: mp 253-254 °C (lit.26
1
mp 288-290 °C); H NMR δ 12.84 (1H, br s, exchanges with
D2O), 8.79 (1H, s), 8.29 (1H, d, J ) 8.0 Hz), 7.68-7.42 (3H,
m). The material was used directly in the next step.
4-Ch lor o-6-m eth oxy-1H-p yr im id o[4,5-b]in d ole (16b). A
suspension of 1,9-dihydro-6-methoxy-4H-pyrimido[4,5-b]indol-
4-one (15b) (800 mg, 3.1 mmol, ∼83% pure) and POCl3 (7 mL)
was heated at 90 °C for 6 h. The suspension was concentrated
to a solid that was evacuated at 1 mmHg for 1 h. The solid
was cooled in a -78 °C bath then treated dropwise with cold
H2O. The bath was removed and the frozen solid was allowed
to gradually melt. The solid was filtered, washed well with
cold H2O, and dried to leave 16b (733 mg, 81%) that was ∼80%
pure: 1H NMR δ 12.64 (1H, br s, exchanges with D2O), 8.74
(1H, s), 7.74 (1H, d, J ) 2.4 Hz), 7.57 (1H, d, J ) 8.9 Hz), 7.28
(1H, dd, J ) 8.9, 2.4 Hz), 3.88 (3H, s); CIMS m/z (relative
intensity) 236 (37Cl, MH+, 29), 235 (37Cl, M+, 37), 234 (35Cl,
MH+, 100), 233 (35Cl, M+, 87). The material was used directly
in the next step.
Methyl 3-amino-7-nitrobenzothiophene-2-carboxylate was
prepared as follows: Triethylamine (0.16 mL, 1.15 mmol) was
added dropwise to a solution of 2-chloro-3-nitrobenzonitrile
(191 mg, 1.05 mmol) and methyl thioglycolate (0.10 mL, 1.1
mmol) in DMSO, and the mixture was stirred under N2 at 25
°C. After 30 min the mixture was poured onto H2O, and the
formed precipitate was collected and dried to give methyl
3-amino-7-nitrobenzothiophene-2-carboxylate (244 mg, 92%)
as an orange-red solid: 1H NMR δ 8.60 (dd, 1H, J ) 1.0, 8.1
Hz, H-4), 8.53 (dd, 1H, J ) 0.9, 7.9 Hz, H-6), 7.66 (t, 1H, J )
7.9 Hz, H-5), 7.31 (br s, 2H, NH2), 3.77 (s, 3H, OMe).
2-Chloro-3-nitrobenzonitrile was prepared as follows: DMF

5472 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26
Showalter et al.
(5 drops) was added to a slurry of 2-chloro-3-nitrobenzoic acid
(987 mg, 4.9 mmol) and oxalyl chloride (5.4 mmol) in CH2Cl2
(20 mL) at 25 °C. Following gas evolution, solution was
gradually achieved. After 3 h, the mixture was concentrated
to a yellow residue that was redissolved in CH2Cl2 (2 mL). The
solution was treated rapidly with cold concentrated aq NH3
(2 mL). After 10 min the formed precipitate was collected and
dried. The solid was added to a solution of PPSA (prepared
from P2O5 and hexamethyldisiloxane as described38) in 1,2-
dichloroethane (30 mL), and the mixture was refluxed for 16
h. The reaction mixture was filtered through a pad of SiO2,
which was eluted with hexanes (250 mL) and then 5% MeOH
in CHCl3 (400 mL). The latter washes were concentrated to
give a white solid that was dissolved in a minimum of CHCl3
and eluted through a second pad of SiO2, eluting with 1%
MeOH in CHCl3 (400 mL). Concentration of the eluate gave
the desired nitrile (0.66 g, 74%) as a white solid: 1H NMR δ
8.36 (dd, 1H, J ) 1.6, 8.2 Hz, H-4), 8.27 (dd, 1H, J ) 1.6, 8.0
Hz, H-6), 7.75 (t, 1H, J ) 8.1 Hz, H-5).
(25) (0.92 g, 4.14 mmol) and formamide (10 mL) was heated
at 135 °C for 1 h and then at 190 °C for 4 h. The reaction
mixture was cooled to 25 °C and the solid was collected and
dried to give 26 (0.61 g, 73%) as yellow-brown needles: 1H
NMR δ 13.0 (1H, br s, NH), 8.86 (1H, dd, J ) 4.6, 1.6 Hz, H-7),
8.63 (1H, dd, J ) 8.0, 1.6 Hz, H-9), 8.4 (1H, s, H-2), 7.68 (1H,
dd, J ) 8.1, 4.6 Hz, H-8).
4-Ch lor op yr id o[3′,2′;4,5]th ien o[3,2-d ]p yr im id in e (27).
Following reaction of Vilsmeier’s reagent (15 mmol), made as
described for 22a , with 3H-pyrido[3′,2′;4,5]thieno[3,2-d]-
pyrimid-4-one (26) (0.61 g, 3.0 mmol) in 1,2-dichloroethane (75
mL) at 85 °C for 2 h, the mixture was cooled and distributed
between CHCl3 and H2O. The organic extracts were washed
with H2O and brine, dried, and concentrated to give 27 (0.64
g, 96%) as a yellow solid: 1H NMR δ 9.3 (1H, br s, H-2), 9.0
(1H, dd, J ) 4.7, 1.7 Hz, H-7), 8.9 (1H, dd, J ) 7.3, 1.7 Hz,
H-9), 7.8 (1H, dd, J ) 7.3, 4.7 Hz, H-8).
Meth yl 2-(2-Cya n op h en oxy)eth a n oa te (30). Methyl bro-
moacetate (1.95 mL, 20 mmol) was added dropwise to a stirred
solution of 2-cyanophenol (29) (2.38 g, 20 mmol) and K2CO3
(2.78 g, 20.1 mmol) in acetone (100 mL) at 25 °C. After 24 h,
the mixture was filtered and the filtrate was concentrated to
leave 30 (3.82 g, 100%) as a beige solid: 1H NMR δ 7.76 (1H,
dd, J ) 7.6, 1.7 Hz, H-3), 7.64 (1H, dt, J d ) 1.6 Hz, J t ) 8.0
Hz, H-4), 7.20-7.10 (2H, m, H-5, H-6), 5.04 (2H, s, OCH2-),
3.70 (3H, s, COOMe). The material was used directly in the
next step.
Meth yl 3-Am in oben zo[b]fu r a n -2-ca r boxyla te (31). A
solution of methyl 2-(2-cyanophenoxy)ethanoate (30) (3.82 g,
20 mmol) in DMSO (40 mL) was added dropwise to a stirred
suspension of NaH (0.504 g, 21 mmol) and DMSO (10 mL)
under N2 at 25 °C. After 10 min the mixture was poured onto
ice H2O and extracted with ether. The combined extracts were
washed with H2O and brine, dried, and concentrated to leave
31 (2.15 g, 56%) as a yellow solid: 1H NMR δ 7.95 (1H, d, J )
7.7 Hz, H-4), 7.48 (2H, d, J ) 3.4 Hz, H-6, H-8), 7.29-7.22
(1H, m, H-7), 6.40 (2H, br s, NH2), 3.80 (3H, s, COOCH3).
3H-Ben zofu r a n o[3,2-d ]p yr im id -4-on e (32). A solution of
methyl 3-aminobenzo[b]furan-2-carboxylate (31) (0.28 g, 1.36
mmol) in formamide (5 mL) was heated at 135 °C for 4 h; then
the temperature was raised to 170 °C. After 4 h the reaction
was cooled to 25 °C and a dark purple solid precipitated. The
solid was collected and air-dried to leave 32 (118 mg, 47%):
1H NMR δ 13.0 (1H, br s, NH), 8.25 (1H, s, H-2), 8.05 (1H, d,
J ) 8.1 Hz, H-6), 7.84 (1H, d, J ) 8.3 Hz, H-9), 7.68 (1H, t, J
) 7.7 Hz, H-8), 7.51 (1H, t, J ) 7.7 Hz, H-7).
4-Ch lor oben zoth ien o[3,2-d ]p yr im id in e (22a ). Vilsmei-
er’s reagent was made by the dropwise addition of DMF (0.27
g, 3.5 mmol) to a solution of oxalyl chloride (0.44 g, 3.5 mmol)
in 1,2-dichloroethane (10 mL) with stirring at 25 °C. After
vigorous gas evolution had ceased, benzothieno[3,2-d]-3H-
pyrimid-4-one (21a )31 (337 mg, 1.53 mmol) was added and the
mixture was heated to reflux. After 20 min, the mixture was
cooled and then quenched with saturated aq NaHCO3 (20 mL).
The phases were separated, and the aqueous phase was
extracted with CHCl3 (3×). The combined organic phases were
washed with H2O (2×) and brine, dried, and concentrated to
give 22a (249 mg, 74%) as a light brown solid: 1H NMR
(CDC13) δ 9.09 (1H, s, H-2), 8.53 (1H, dd, J ) 1.8, 7.6 Hz, H-6),
7.95 (1H, d, J ) 7.8 Hz, H-9), 7.73 (1H, dt, J d ) 1.4 Hz, J t
7.7 Hz, H-7), 7.62 (1H, dt, J d ) 1.2 Hz, J t ) 7.5 Hz, H-8).
)
7-Am in o-8-flu or o-4-th iom eth ylben zoth ien o[3,2-d ]p yr i-
m id in e (23d ). A mixture of P2S5 (620 mg, 2.8 mmol) and
7-amino-8-fluoro-3H-[1]benzothieno[3,2-d]pyrimid-4-one (21d )
(620 mg, 2.64 mmol) was stirred at 115 °C in diglyme (13 mL)
under N2 for 1 h and cooled to 25 °C. The mixture was treated
with diisopropylamine (1.29 g, 10 mmol) and MeI (0.31 mL, 5
mmol) and stirred under N2 for 10 min, then filtered. The
filtrate was poured onto stirred ice H2O and the formed
precipitate was collected, triturated in CHCl3, and dried to give
23d (181 mg, 26%): 1H NMR δ 8.96 (1H, s, H-2), 7.91 (1H, d,
J ) 11.2 Hz, H-9), 7.32 (1H, d, J ) 7.6 Hz, H-6), 6.24 (2H, br
s, NH2), 2.74 (3H, s, SCH3).
7-E t h yla m in o-8-flu or o-4-t h iom et h ylb en zo[b]t h ien o-
[3,2-d ]p yr im id in e (23e). A suspension of 7-ethylamino-8-
fluoro-3H-[1]benzothieno[3,2-d]pyrimid-4-one (264 mg, 1.0
mmol) (21e) and Lawesson’s reagent32 (675 mg, 1.65 mmol)
in diglyme (5 mL) was stirred under N2 at 112 °C for 6.5 h.
After cooling, the mixture was concentrated and the residue
was dissolved in DMSO (3 mL). Diisopropylamine (263 mg, 2
mmol) was added, forming an orange-brown solution, followed
by MeI (287 mg, 2 mmol). After 20 min, the mixture was
diluted with H2O. The formed precipitate was collected and
dried to give 23e (266 mg, 77%) as a dark brown solid: 1H
NMR δ 8.93 (1H, s, H-2), 7.87 (1H, d, J ) 11.5 Hz, H-9), 7.32
(1H, d, J ) 7.6 Hz, H-6), 6.48 (1H, br t, NH), 3.24 (2H, p, J )
6.6 Hz, NHCH2), 2.72 (3H, s, SCH3), 1.23 (3H, t, J ) 7.1 Hz,
CH3).
E t h yl 3-Am in op yr id o[3,2-b]t h iop h en e-2-ca r b oxyla t e
(25). Ethyl thioglycolate (0.12 mL, 1.1 mmol) was added to a
stirred mixture of NaH (0.036 g, 1.5 mmol) and DMSO (1 mL)
under N2 at 25 °C. After formation of H2 had ceased, a solution
of 2-chloro-3-cyanopyridine (24) (0.14 g, 1.0 mmol) in DMSO
(2 mL) was added dropwise. After 3 h the mixture was poured
onto stirred ice H2O. The light yellow precipitate was collected
and dried to give 25 (197 mg, 89%): 1H NMR δ 8.68 (1H, dd,
J ) 4.6, 1.6 Hz, H-6), 8.54 (1H, dd, J ) 8.2, 1.6 Hz, H-4), 7.46
(1H, dd, J ) 8.2, 4.5 Hz, H-5), 7.31 (2H, br s, NH2), 4.3 (2H, q,
J ) 7.1 Hz, OCH2-), 1.29 (3H, t, J ) 7.1 Hz, CH3).
4-Ch lor oben zofu r a n o[3,2-d ]p yr im id in e (33). Following
reaction of Vilsmeier’s reagent (3.1 mmol), made as described
for 22a , with 3H-benzofurano[3,2-d]pyrimid-4-one (32) (113,-
mg, 0.61 mmol) in 1,2-dichloroethane (15 mL) at reflux for 1
h, the mixture was cooled and distributed between H2O and
CHCl3. The combined extracts were washed with H2O and then
brine, dried, and concentrated to give 33 (116 mg, 93%) as a
yellow solid: 1H NMR δ 9.08 (1H, s, H-2), 8.30 (1H, d, J ) 8.1
Hz, H-6), 8.02 (1H, d, J ) 8.5 Hz, H-9), 7.90, (1H, dt, J d ) 1.3
Hz, J t ) 7.1 Hz, H-8), 7.64 (1H, dt, J d ) 1.0 Hz, J t ) 7.8 Hz,
H-7).
Gen er a l P r oced u r e A. A stirred solution of the tricyclic
4-chloro- or 4-thiomethylpyrimidine and -aniline (2 molar
excess) in a lower alcohol solvent (0.15-1.0 M solution) was
heated at reflux until TLC showed complete consumption of
the starting pyrimdine. Occasionally, reactions were run in
DMA, ethylene glycol, or 2-methoxyethanol at the indicated
temperatures. All reactions were catalyzed with anhydrous
HCl. After cooling to 25 °C, the mixture was worked up to
provide the desired product. The following compounds were
made by this procedure.
N-P h en yl-5H -p yr im id o[5,4-b]in d ole-4-a m in e H yd r o-
ch lor id e (34). Reaction of 4-chloro-5H-pyrimido[5,4-b]indole
hydrochloride (11)23 (240 mg, 1.0 mmol), aniline (0.273 mL, 3
mmol), and EtOH (1 mL) was followed by dilution of the
suspension with EtOH (4 mL). The solid was collected, washed
with H2O and EtOH, and then recrystallized from DMF/H2O
3H-P yr id o[3′,2′;4,5]th ien o[3,2-d ]p yr im id -4-on e (26). A
mixture of ethyl 3-aminopyrido[3,2-b]thiophene-2-carboxylate
1
to give 34 (82 mg, 27%): mp >340 °C; H NMR δ 12.79 (1H,

Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26 5473
br s), 11.04 (1H, br s), 8.94 (1H, s), 8.27 (1H, d, J ) 8.2 Hz),
7.96 (2H, d, J ) 7.5 Hz), 7.85 (1H, d, J ) 8.4 Hz), 7.71 (1H, t,
J ) 7.7 Hz), 7.49 (2H, t, J ) 8.0 Hz), 7.41 (1H, t, J ) 7.6 Hz),
7.24 (1H, t, J ) 7.4 Hz); CIMS m/z (relative intensity) 261
(MH+, 100), 260 (M+, 57).
Hz), 2.41 (1H, br s); CIMS m/z (relative intensity) 275 (MH+,
100), 274 (M+, 17).
Gen er a l P r oced u r e C. A slurry of the 4-anilinonitrobenzo-
[b]thieno[3,2-d]pyrimidine, Raney Ni, and an appropriate
solvent was hydrogenated at ∼50 psi/25 °C until the theoretical
amount of hydrogen had been taken up. The mixture was
filtered over Celite and the filtrate was concentrated to give a
residue that was further worked up to give the product. The
following compounds were made by this procedure.
8-Am in o-4-an ilin oben zo[b]th ien o[3,2-d]pyr im idin e (58).
Hydrogenation of 4-anilino-8-nitrobenzo[b]thieno[3,2-d]pyri-
midine (54) (234 mg, 0.65 mmol), Raney Ni (0.2 g), and THF
(40 mL)/MeOH (35 mL) for 17 h followed by crystallization
from EtOH gave 58 (59 mg, 30%) as a khaki solid: mp 241-
243 °C; 1H NMR δ 9.54 (1H, br s, NH), 8.67 (1H, s, H-2), 7.78
(2H, dd, J ) 0.9, 8.5 Hz, H-2′, H-6′), 7.75 (1H, d, J ) 8.7 Hz,
H-6), 7.49 (1H, d, J ) 1.9 Hz, H-9), 7.38 (1H, t, J ) 7.9 Hz,
H-3′), 7.12 (1H, dt, J d ) 7.5, J t ) 1.1 Hz, H-4′), 7.10 (1H, dd,
J ) 2.3, 8.7 Hz), 5.45 (2H, br s, NH2); CIMS m/z (relative
intensity) 293 (MH+, 100), 292 (M+, 65).
N-(3-Br om op h en yl)-6-m eth oxy-1H-p yr im id o[4,5-b]in -
d ole-4-a m in e (49). Reaction of 4-chloro-6-methoxy-1H-py-
rimido[4,5-b]indole (16b) (107 mg, 0.37 mmol, 80% pure),
3-bromoaniline (0.15 mL, 1.4 mmol), DMA (1 mL), and 8.5 M
HCl in 2-propanol (1 drop) at 120 °C for 5 h was followed by
concentration to a residue that was triturated in 5% aq
NaHCO3 and collected. The solid was dissolved in a minimum
volume of DMF and purified by preparative TLC (elution with
3:2 CH2Cl2/EtOAc). The product band was extracted from the
SiO2 by sonication in EtOAc and filtration. The filtrate was
concentrated to a solid that was triturated in MeOH, collected,
and dried to give 49 (39 mg, 28%): 1H NMR δ 11.99 (1H, br s,
exchanges with D2O), 8.97 (1H, br s, exchanges with D2O), 8.44
(1H, s), 8.02 (1H, s), 7.91 (1H, d, J ) 2.4 Hz), 7.76 (1H, d, J )
8.0 Hz), 7.42 (1H, d, J ) 8.7 Hz), 7.36-7.24 (2H, m), 7.08 (1H,-
dd, J ) 8.7, 2.2 Hz), 3.87 (3H, s); CIMS m/z (relative intensity)
371 (81Br, MH+, 74), 370 (81Br, M+, 71), 369 (79Br, MH+, 100),
368 (79Br, M+, 65).
P r oced u r e D. 4-(3-Br om oa n ilin o)-8-m eth yla m in oben -
zo[b]th ien o[3,2-d]pyr im idin e (67) an d 4-(3-Br om oan ilin o)-
8-d im eth yla m in oben zo[b]th ien o[3,2-d ]p yr im id in e (68).
To a 0 °C mixture of 8-amino-4-(3-bromoanilino)benzo[b]-
thieno[3,2-d]pyrimidine (61) (363 mg, 0.98 mmol), formic acid
(0.13 g, 88%, 2.5 mmol), and H2O (2 mL) was added dropwise
formaldehyde (0.22 g, 2.6 mmol). The mixture was heated at
90 °C for 24 h then cooled and extracted with CHCl3. The
combined organic phases were washed with H2O and brine,
dried, and concentrated to a solid. Purification by preparative
TLC (elution with 2% MeOH in CHCl3) gave two products. The
upper band was extracted to give 68 (29 mg, 7.4%) as a yellow
4-(3-Br om oa n ilin o)ben zoth ien o[3,2-d ]p yr im id in e (53).
Reaction of 4-chlorobenzothieno[3,2-d]pyrimidine (22a ) (110.1
mg, 0.5 mmol), 3-bromoaniline (107.2 mg, 0.62 mmol), triethyl-
amine (102.8 mg, 1.0 mmol), and ethoxyethanol (2 mL) at 110
°C for 18 h was followed by concentration to an oil that was
purified by preparative TLC (elution with 2% MeOH in CHCl3).
Extraction of the product band gave a solid that was crystal-
lized from EtOH to leave 53 (70 mg, 39%) as pale beige
plates: mp 231-233 °C; 1H NMR (CDCl3) δ 8.88 (1H, s, H-2),
8.49 (1H, dd, J ) 1.7, 7.1 Hz, H-9), 7.96 (1H, t, J ) 1.9 Hz,
1
solid: mp 190-192 °C; H NMR δ 9.69 (1H, br s, NH), 8.69
H-2′), 7.89 (1H, dd, J ) 1.6, 7.0 Hz, H-6), 7.65 (1H, dt, J d
1.5 Hz, J t ) 7 Hz, H-7), 7.60 (1H, dd, J ) 1.5, 7.5 Hz, H-6′),
7.57 (1H, dt, J d ) 1.5 Hz, J t ) 7 Hz, H-8), 7.40 (1H, dt, J d
)
(1H, s, H-2), 8.14 (1H, s, H-2′), 7.91 (1H, d, J ) 9.0 Hz, H-6),
7.79 (1H, d, J ) 7.8 Hz, H-6′), 7.48 (1H, s, H-9), 7.29 (1H, t, J
) 7.6 Hz, H-5′), 7.24-7.20 (2H, m, H-7, H-4′), 2.98 (6H, s, 2 ×
Me); CIMS m/z (relative intensity) 401 (81Br, MH+, 100), 400
(81Br, M+, 95), 399 (79Br, MH+, 99), 398 (79Br, M+, 77).
)
1.7 Hz, J t ) 8 Hz, H-4′), 7.28 (1H, t, J ) 7.8 Hz, H-5′), 6.90
(1H, br s, NH); CIMS m/z (relative intensity) 358 (81Br, MH+,
68), 357 (81Br, M+, 48), 356 (79Br , MH+, 100), 355 (79Br , M+,
41).
The lower band was extracted to give 67 (12 mg, 3%) as a
yellow solid: mp 193-194 °C; 1H NMR δ 9.65 (1H, br s, NH),
8.68 (1H, s, H-2), 8.14 (1H, d, J ) 1.9 Hz, H-2′), 7.79-7.76
(2H, m, H-6, H-6′), 7.29-7.23 (3H, m, H-5′, H-4′, H-9), 6.99
(1H, dd, J ) 2.4, 8.8 Hz, H-7), 6.03 (1H, q, J ) 5.1 Hz, NHCH3),
2.73 (3H, d, J ) 4.9 Hz, NHCH3); CIMS m/z (relative intensity)
387 (81Br, MH+, 92), 386 (81Br, M+, 97), 385 (79Br, MH+, 100),
384 (79Br, M+, 77).
En zym e Assa y. EGFr was prepared from human A431
carcinoma cell shed membrane vesicles by immunoaffinity
chromatography as previously described,39 and the assays were
carried out as reported previously.18 The substrate used was
based on a portion of phospholipase C-γ1, having the sequence
Lys-His-Lys-Lys-Leu-Ala-Glu-Gly-Ser-Ala-Tyr472-Glu-Glu-
Val. The reaction was allowed to proceed for 10 min at room
temperature, then was stopped by the addition of 2 mL of 75
mM phosphoric acid. The solution was then passed through a
2.5-cm phosphocellulose disk which bound the peptide. This
filter was washed with 75 mM phosphoric acid (5×), and
incorporated label was assessed by scintillation counting in
an aqueous fluor. Control activity (no drug) gave a count of
approximately 100 000 cpm. At least two independent dose-
response curves were done and the IC50 values computed. The
reported values are averages; variation was generally (15%.
EGF r Au top h osp h or yla tion in A431 Hu m a n Ep id er -
m oid Ca r cin om a Cells. Cells were grown to confluence in
6-well plates (35-mm diameter) and exposed to serum-free
medium for 18 h. The cells were treated with compound for 2
h and then with 100 ng/mL EGF for 5 min. The monolayers
were lysed in 0.2 mL of boiling Laemlli buffer (2% sodium
dodecyl sulfate, 5% â-mercaptoethanol, 10% glycerol, and 50
mM Tris, pH 6.8), and the lysates were heated to 100 °C for 5
min. Proteins in the lysate were separated by polyacrylamide
gel electrophoresis and electrophoretically transferred to
nitrocellulose. The membrane was washed once in a mixture
of 10 mM Tris, pH 7.2, 150 mM NaCl, and 0.01% azide (TNA),
7-Am in o-4-(3-b r om oa n ilin o)-8-flu or ob en zo[b]t h ien o-
[3,2-d ]p yr im id in e (64). Reaction of 7-amino-8-fluoro-4-thio-
methylbenzothieno[3,2-d]pyrimidine (23d ) (53 mg, 0.2 mmol),
3-bromoaniline (69 mg, 0.4 mmol), 3-bromoaniline hydrochlo-
ride (125 mg, 0.6 mmol), and ethylene glycol (1 mL) at 140 °C
for 38 h was followed by pouring the solution onto ice H2O to
precipitate a solid that was collected. Purification by prepara-
tive TLC (elution with 5% MeOH in CHCl3) followed by
extraction of the product band and crystallization from EtOH
gave 64 (18 mg, 23%) as a mustard yellow solid: mp 244-246
°C; 1H NMR δ 9.62 (1H, s, NH), 8.67 (1H, s, H-2), 8.16 (1H, t,
J ) 1.9 Hz, H-2′), 7.85 (1H, d, J ) 11.1 Hz, H-9), 7.80 (1H,
ddd, J ) 1.0, 1.9, 8.2 Hz, H-6′), 7.32 (1H, t, J ) 8.0 Hz, H-5′),
7.32 (1H, d, J ) 7.7 Hz, H-6), 7.26 (1H, ddd, J ) 1.0, 1.9, 8.0
Hz, H-4′), 6.04 (2H, br s, NH2); CIMS m/z (relative intensity)
391 (81Br, MH+, 77), 390 (81Br, M+, 65), 389 (79Br , MH+, 100),
388 (79Br , M+, 50).
Gen er a l P r oced u r e B. A stirred mixture of the tricyclic
4-chloropyrimidine and the neat amine, usually with a cata-
lytic amount of HCl, was heated at an elevated temperature
until TLC showed complete consumption of starting pyrimi-
dine. After cooling to 25 °C, excess amine was removed under
vacuum and the residue was further processed to give the
desired product. The following compounds were made by this
procedure.
N-(P h en ylm et h yl)-5H -p yr im id o[5,4-b]in d ole-4-a m in e
(37). Reaction of 4-chloro-5H-pyrimido[5,4-b]indole hydrochlo-
ride (11)23 (240 mg, 1 mmol) and benzylamine (1 mL) at 150
°C for 6 h provided a soft solid that was dissolved in EtOAc.
The solution was washed with saturated aq NaHCO3, H2O,
and brine, dried, and concentrated to leave a residue that was
triturated in CH2Cl2 to provide 37 (190 mg, 69%): mp 242-
244 °C; 1H NMR (CDCl3) δ 10.58 (1H, br s), 8.60 (1H, s), 8.08
(1H, d, J ) 8.0 Hz), 7.47-7.14 (8H, m), 4.82 (2H, d, J ) 5.6

5474 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 26
Showalter et al.
(18) Fry, D. W.; Kraker, A. J .; McMichael, A.; Ambroso, L. A.; Nelson,
J . M.; Leopold, W. R.; Connors, R. W.; Bridges, A. J . A specific
inhibitor of the epidermal growth factor receptor tyrosine kinase.
Science 1994, 265, 1093-1095.
and blocked overnight in TNA containing 5% bovime serum
albumin and 1% ovalbumin. The membrane was blotted for 2
h with antiphosphotyrosine antibody (UBI, 1 µg/mL in blocking
buffer) and then washed twice in TNA, once in TNA containing
0.05% Tween-20 and 0.05% nonidet P-40, and twice in TNA.
The membranes were then incubated for 2 h in blocking buffer
containing 0.1 µCi/mL [125I]protein A and then washed again
as above. After the blots were dry they were loaded into a film
cassette and exposed to X-AR X-ray film for 1-7 days. Band
intensities were determined with a Molecular Dynamics laser
densitometer.
(19) Traxler, P. M.; Furet, P.; Mett, H.; Buchdunger, E.; Meyer, T.;
Lydon, N. 4-(Phenylamino)pyrrolopyrimidines: potent and selec-
tive, ATP site directed inhibitors of the EGF-receptor protein
tyrosine kinase. J . Med. Chem. 1996, 39, 2285-2292.
(20) Rewcastle, G. W.; Palmer, B. D.; Thompson, A. M.; Bridges, A.
J .; Cody, D. R.; Zhou, H.; Fry, D. W.; McMichael, A.; Kraker, A.
J .; Denny, W. A. Tyrosine kinase inhibitors. 10. Isomeric 4-[(3-
bromophenyl)amino]pyrido[d]pyrimidines are potent ATP bind-
ing site inhibitors of the tyrosine kinase function of the epider-
mal growth factor receptor. J . Med. Chem. 1996, 39, 1823-1835.
(21) Rewcastle, G. W.; Palmer, B. D.; Bridges, A. J .; Showalter, H.
D. H.; Sun, L.; Nelson, J .; McMichael, A.; Kraker, A. J .; Fry, D.
W.; Denny, W. A. Tyrosine kinase inhibitors. 9. Synthesis and
evaluation of fused tricyclic quinazoline analogues as ATP site
inhibitors of the tyrosine kinase activity of the epidermal growth
factor receptor. J . Med. Chem. 1996, 39, 918-928.
(22) Bridges, A. J .; Zhou, H.; Cody, D. R.; Rewcastle, G. W.; Mc-
Michael, A.; Showalter, H. D. H.; Fry, D. W.; Kraker, A. J .;
Denny, W. A. Tyrosine kinase inhibitors. 8. An unusually steep
structure-activity relationship for analogues of 4-(3-bromo-
anilino)-6,7-dimethoxyquinazoline (PD 153035), a potent inhibi-
tor of the epidermal growth factor receptor. J . Med. Chem. 1996,
39, 267-276.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and spectral data for compounds 22b-j, 35, 36, 38-
48, 50-52, 54-57, 59-63, 65, 66, 69, and 70-77. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Refer en ces
(1) Messa, C.; Russo, F.; Caruso, M. G.; DiLeo, A. EGF, TGF-R, and
EGF-R in human colorectal adenocarcinoma. Acta Oncol. 1998,
37, 285-289.
(2) Reese, D. M.; Slamon, D. J . HER-2/neu signal transduction in
human breast and ovarian cancer. Stem Cells 1997, 15, 1-8.
(3) Revillion, F.; Bonneterre, J .; Peyrat, J . P. ERB-B2 oncogene in
human breast cancer and its clinical significance. Eur. J . Cancer
1998, 34, 791-808.
(4) Hynes, N. E.; Stern, D. F. The biology of erbB-2/neu/HER-2 and
its role in cancer. Biochim. Biophys. Acta-Rev. Cancer 1994,
1198, 165-184.
(5) Bridges, A. J . The epidermal growth factor receptor family of
tyrosine kinases and cancer: can an atypical exemplar be a
sound therapeutic target? Curr. Med. Chem. 1996, 3, 167-196.
(6) Khazaie, K.; Schirrmacher, V.; Lichtner, R. B. EGF receptor in
neoplasia and metastasis. Cancer Metastasis Rev. 1993, 12, 255-
274.
(23) Monge, A.; Palop, J . A.; Goni, T.; Martinez-Crespo, F.; Recalde,
I.; Fernandez-Alvarez, E. Synthesis of 4-hydrazino-5H-pyrimido-
[5,4-b]indole and related compounds. J . Heterocycl. Chem. 1986,
23, 647-649.
(24) Kadushkin, A. V.; Nesterova, I. N.; Golovko, T. V.; Nikolaeva, I.
S.; Pushkina, T. V.; Fomina, A. N.; Sokolova, A. S.; Chernov, V.
A.; Granik, V. G. Condensed pyrrolo[3,2-d]pyrimidines: synthe-
sis and biological activity. Khim.-Farm. Zh. 1990, 24, 18-22;
Chem. Abstr. 1991, 114, 164140.
(25) Grob, C. A.; Weissbach, O. The preparation of oxindole deriva-
tives from o-chloronitrobenzene and cyanoacetic acid or malonic
esters. Helv. Chim. Acta 1961, 44, 1748-1753.
(26) Glushkov, R. G.; Volskova, V. A.; Magidson, O. Y. 2-Aminocar-
boxyethylindole and its derivatives. Khim.-Farm. Zh. 1966, 1,
25-32; Chem. Abstr. 1967, 68, 105143.
(7) Davies, D. E.; Chamberlin, S. G. Targeting the epidermal growth
factor receptor for therapy of carcinomas. Biochem. Pharmacol.
1996, 51, 1101-1110.
(27) Venugopalan, B.; Desai, P. D.; de Souza, N. J . Synthesis of 6,7-
dimethoxypyrimido[4,5-b]indoles as potential antihypertensive
agents. J . Heterocycl. Chem. 1988, 25, 1633-1639.
(8) Gibson, K. H.; Grundy, W.; Godfrey, A. A.; Woodburn, J . R.;
Ashton, S. E.; Curry, B. J .; Scarlett, L.; Barker, A. J .; Brown, D.
S. Epidermal growth factor receptor tyrosine kinase: structure-
activity relationships and antitumour activity of novel quinazo-
lines. Bioorg. Med. Chem Lett. 1997, 7, 2723-2728.
(9) Levitzki, A.; Gazit, A. Tyrosine kinase inhibition: an approach
to drug development. Science 1995, 267, 1782-1788.
(10) Traxler, P. M. Protein tyrosine kinase inhibitors in cancer
treatment. Expert Opin. Ther. Patents 1997, 7, 571-588.
(11) Trinks, U.; Buchdunger, E.; Furet, P.; Kump, W.; Mett, H.;
Meyer, T.; Mueller, M.; Regenass, U.; Rihs, G.; Lydon, N.;
Traxler, P. Dianilinophthalimides: potent and selective, ATP-
competitive inhibitors of the EGF-receptor protein tyrosine
kinase. J . Med. Chem. 1994, 37, 1015-1027.
(28) Portnov, Y. N.; Bulaga, S. N.; Zabrodnyaya, V. G.; Smirnov, L.
D. Synthesis of 9H-pyrimido[4,5-b]indoles from 2-amino-3-
cyanoindoles. Khim. Geterotsikl. Soedin. 1991, 3, 400-402;
Chem. Abstr. 1992, 115, 136045.
(29) Robba, M.; Touzot, P.; Riquelme, R. M. Synthese de benzo(1)-
thieno[2,3-d]pyrimidines et de benzo(1)thieno[3,2-d]pyrimidines.
Tetrahedron Lett. 1972, 44, 4549-4551.
(30) Robba, M.; Touzot, P.; El-Kashef, H. [1]Benzothienopyrimidines.
I. Etude de la 3H-benzothieno[3,2-d]pyrimidone-4. J . Heterocycl.
Chem. 1980, 17, 923-928.
(31) Bridges, A. J .; Zhou, H. Synthetic methodology for the prepara-
tion of benzo[b]thieno[3,2-d]pyrimidines substituted with elec-
tron donating substituents on the benzene ring. J . Heterocycl.
Chem. 1997, 34, 1134-1139.
(32) Scheibye, S.; Pedersen, B. S.; Lawesson, S.-O. The dimer of
p-methoxyphenylthionophosphine sulfide as thiation reagent. A
new route to thiocarboxamides. Bull. Soc. Chim. Belg. 1978, 87,
229-233.
(33) Clarke, H. T.; Gillespie, H. H.; Weishaus, S. Z. The action of
formaldehyde on amines and amino acids. J . Am. Chem. Soc.
1933, 55, 4571.
(34) Athmani, S.; Iddon, B. Thiazolo[4′,5′:4,5]thieno[3,2-d]pyrimidine,
a new heterocyclic system. Tetrahedron 1992, 48, 7689-7702.
(35) Gewald, K.; Neumann, G.; Brettcher, H. Heterocycles from CH-
acidic nitriles: XI. New synthesis of 2-aminothianaphthene.
Chem. Ber. 1966, 99, 261.
(12) Moyer, J . D.; Barbacci, E. G.; Iwata, K. K.; Arnold, L.; Boman,
B.; Cunningham, A.; Diorio, C.; Doty, J .; Morin, M. J .; Moyer,
M. P.; Neveu, M.; Pollack, V. A.; Pustilnik, L. R.; Reynolds, M.
M.; Sloan, D.; Theleman, A.; Miller, P. Induction of apoptosis
and cell cycle arrest by CP-358,774, an inhibitor of epidermal
growth factor receptor tyrosine kinase. Cancer Res. 1997, 57,
4838-4848.
(13) Palmer, B. D.; Trumpp-Kallmeyer, S.; Fry, D. W.; Nelson, J . M.;
Showalter, H. D. H.; Denny, W. A. Tyrosine kinase inhibitors.
11. Soluble analogues of pyrrolo- and pyrazoloquinazolines as
epidermal growth factor receptor inhibitors: synthesis, biological
evaluation, and modeling of the mode of binding. J . Med. Chem.
1997, 40, 1519-1529.
(14) Traxler, P.; Bold, G.; Frei, J .; Lang, M.; Lydon, N.; Mett, H.;
Buchdunger, E.; Meyer, T.; Mueller, M.; Furet, P. Use of a
pharmacophore model for the design of EGF-R tyrosine kinase
inhibitors: 4-(phenylamino)pyrazolo[3,4-d]pyrimidines. J . Med.
Chem. 1997, 40, 3601-3616.
(15) Lydon, N. B.; Mett, H.; Mueller, M.; Becker, M.; Cozens, R. M.;
Stover, D.; Daniels, D.; Traxler, P.; Buchdunger, E. A potent
protein-tyrosine kinase inhibitor which selectively blocks pro-
liferation of epidermal growth factor receptor-expressing tumor
cells in vitro and in vivo. Int. J . Cancer 1998, 76, 154-163.
(16) Thomas, A. P. 4,6-Dianilinopyrimidine derivatives, their prepa-
ration and their use as tyrosine kinase inhibitors. World Pat.
Appl. WO 95/15952, 1995.
(36) Bridges, A. J .; Cody, D. R.; Zhou, H.; McMichael, A.; Fry, D. W.
Enantioselective inhibition of the EGF receptor tyrosine kinase
by 4-(R-phenethylamino)quinazoline derivatives. Bioorg. Med.
Chem. 1995, 3, 1651-1656.
(37) Halfpenny, P. R.; Horwell, D. C.; Hughes, J .; Hunter, J . C.; Rees,
D. C. Highly selective kappa-opioid analgesics. 3. Synthesis and
structure-activity relationships of novel N-[2-(1-pyrrolidinyl)-
4- or -5-substituted-cyclohexyl]arylacetamide derivatives. J .
Med. Chem. 1990, 33, 286-291.
(38) Yokoyama, M.; Yoshida, S.; Imamoto, S. Organic reactions using
trimethylsilyl polyphosphate (PPSE): a convenient synthesis of
nitriles from carboxamides. Synthesis 1982, 591-592.
(39) Gill, G. N.; Weber, W. Purification of functionally active epider-
mal growth factor receptor protein using a competitive antago-
nist monoclonal antibody and competitive elution with epidermal
growth factor. Methods Enzymol. 1987, 146, 82-88.
(17) Ward, W. H. J .; Cook, P. N.; Slater, A. M.; Davies, D. H.;
Holdgate, G. A.; Green, L. R. Epidermal growth factor receptor
tyrosine kinase - investigation of catalytic mechanism, structure-
based searching and discovery of a potent inhibitor. Biochem.
Pharmacol. 1994, 48, 659-666.
J M9903949