Novel indolo[2,1-b]quinazoline analogues as cytostatic agents: synthesis, biological evaluation and structure-activity relationship.

Article abstract of DOI:10.1016/S0960-894X(02)00431-6

In our endeavor to design and synthesize novel anticancer agents, a new series of indoloquinazoline compounds were prepared and tested initially for anticancer activity in vitro against a panel of human cancer cell lines. Most of these compounds exhibited cytotoxic activity in in vitro screens. Compounds were selected and further evaluated using a modified Hollow Fiber Assay for their preliminary in vivo activity against 12 cell lines implanted in the subcutaneous and intraperitoneal compartments in mice. The results indicate that these compounds may constitute a new class of anticancer agents.

Full text of DOI:10.1016/S0960-894X(02)00431-6

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2303–2307
Novel Indolo[2,1-b]quinazoline Analogues as Cytostatic Agents:
Synthesis, Biological Evaluation and Structure–Activity
Relationship
a,
a
a
a
a
Vedula M. Sharma, * P. Prasanna, K. V. Adi Seshu, B. Renuka, C. V. Laxman Rao,
a
a
a
a
G. Sunil Kumar, C. Prasad Narasimhulu, P. Aravind Babu, R. C. Puranik,
a
a
b
b
D. Subramanyam, A. Venkateswarlu, Sriram Rajagopal, K. B. Sunil Kumar,
b
b
b
C. Seshagiri Rao, N. V. S. Rao Mamidi, Dhanvanthri S. Deevi,
b
b
R. Ajaykumar and R. Rajagopalan
aDiscovery Chemistry (Natural Products), Dr. Reddy’s Research Foundation, Miyapur, Hyderabad -500 050, India
bDiscovery Biology, Dr. Reddy’s Research Foundation, Miyapur, Hyderabad -500 050, India
Received 22 March 2002; accepted 6 June 2002
Abstract—In our endeavor to design and synthesize novel anticancer agents, a new series of indoloquinazoline compounds were
prepared and tested initially for anticancer activity in vitro against a panel of human cancer cell lines. Most of these compounds
exhibited cytotoxic activity in in vitro screens. Compounds were selected and further evaluated using a modified Hollow Fiber
Assay for their preliminary in vivo activity against 12 cell lines implanted in the subcutaneous and intraperitoneal compartments in
mice. The results indicate that these compounds may constitute a new class of anticancer agents. # 2002 Elsevier Science Ltd. All
rights reserved.
In our search to identify novel anticancer compounds
from Indian medicinal plants, we screened several crude
plant extracts for their in vitro anticancer activity using
human cancer cell lines. Many of the extracts showed
cytotoxic activity in micromolar concentrations. Among
the crude extracts screened, based on preliminary in
vitro cytotoxic activity, the leafy extract of Wrightia
tinctoria was further fractionated and purified and this
resulted in isolation of a known alkaloid tryptanthrine
(1), which was reported earlier in the literature from
various plant sources including W. tinctoria.
1À5
Com-
pound 1 is reported to possess antibacterial, antifungal
1
and weak cytotoxic activity against B-16 melanoma
6
cells. Although compound 1 is not a potent cytotoxic
compound, it has a structural similarity with a known
7
potent cytotoxic agent batracyclin (1a) (Fig. 1), and
this feature makes it an attractive lead compound for
synthetic and biological studies. Since the yields of
Figure 1. Structures of tryptanthrine (1) and batracyclin (1a).
*
0
Corresponding author. Tel.: +91-40-305439; fax: +91-40-3045438; e-mail: sarmavm@drreddys.com
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00431-6

2304
V. M. Sharma et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2303–2307
compound 1 from the plant are poor, we synthesized a
parent compound tryptanthrine and a series of its new
analogues. These compounds were evaluated initially
for cytotoxic activity in vitro against human cancer cell
lines. Compounds with good GI50 values were further
tested in a rapid preliminary in vivo model using mod-
ified Hollow Fibre Assay (HFA) in Swiss Albino Mice
and known methods in the literature were followed.
Additional derivatives (13–18) containing sulfonyl and
alkanoyl substituents of compound 8 were also made as
shown in Scheme 2 by treating with alkanoylhalides or
alkylsulfonylhalides. To investigate the role of the 12-
keto functionality in the biological activity of these qui-
nazolines, compounds 19–22 were synthesized by
exploiting 12-keto functionality through Grignard reac-
tion. Isatin was reacted with alkylmagnesiumbromides
(SAM), followed by xenograft of HT-29 colon adeno-
carcinoma in nude mice.
ꢀ
in THF at 0 C to obtain 3-hydroxy-3-alkyl-2-inda-
nones. These intermediates were condensed further with
isatoicanhydrides as depicted in Scheme 3.
As outlined in Scheme 1, 6,12-dihydro-6,12-dioxoin-
dolo[2,1-b]quinazoline compounds (1–12) were prepared
8
by condensation of substituted isatoicanhydrides with
substituted isatins
9À11
in toluene by refluxing in the
The final synthetic sequence employed in this study is
shown in Scheme 4 for the preparation of compounds
23–36. Compounds 1, 6 and 7 were reacted with hydro-
xylaminehydrochloride to provide the corresponding
presence of triethylamine. The commercially available
isatin and isatoicanhydride were used to prepare
unsubstituted indoloquinazolines and other analogues,
Scheme 1. (a) Et N (5 equiv), toluene, reflux, 2–4 h, 70–85%.
3
ꢀ
2.
, 0 C, 0.5 h, 95%; (b) SnCl 2H
ꢀ
Scheme 2. (a) Concd HNO
ꢀ
3
/H SO
2
4
2
O (2 equiv), ethanol/concd HCl, 80 C, 2 h, 38%; (c) RCOCl or RSO
2
Cl, Et
3
N,
DCM, 0 C, 0.5–1 h, 65–70%.

V. M. Sharma et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2303–2307
2305
oximes. These E and Z isomeric oximes on treatment
with acid halides and alkyl halides yielded correspond-
ing oxime ester (27) and ethers (24–26, 28–36) respec-
tively. The mixtures of E and Z isomers were evaluated
for their biological activity without separation.
lines, and the active compounds were further screened
rapidly for the in vivo activity using the modified
1
2
HFA. To establish the SAR of this series of com-
pounds, each variable of general structures in Schemes
1, 2 and 4 was systematically modified, and the GI50
values obtained were tabulated in Table 1. Average GI50
values of eight cell lines were used in the following
discussion. The parent compound 1 without any
The indoloquinazolines prepared in this study were
tested in vitro for anticancer activity on eight cancer cell
ꢀ
Scheme 3. (a) RMgBr (2 equiv), THF, 0 C–rt, 1–4 h, 45–50%; (b) isatoicanhydride (2 equiv), Et
3
N (5 equiv), toluene, reflux, 5–10 h, 25–30%.
ꢀ
Scheme 4. (a) (H NOH)
2
2
H
2
SO
4
(2 equiv), pyridine, MeOH, reflux, 4 h, 65%; (b) RX, K
2
CO
3
, D MF, 90 C, 12–20 h, 40–50%.

2306
V. M. Sharma et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2303–2307
Table 1. In vitro cytotoxic activities of indolo [2,1-b] quinazolines
Compd
GI50 values in mM
Breast
MCF7/ADR
CNS
U251
Colon
SW620
Lung
H522
Melanoma
M14
Ovarian
SKOV3
Prostate
DU145
Renal
A498
1
2
5
6
7
8
9
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
30.0
>100
40.0
0.7
1.0
5.0
>100
0.1
100
5.0
>100
60.0
4.0
15.0
>100
15.0
>100
60.0
2.5
>100
40.0
5.0
0.4
>100
60.0
0.15
8.0
0.95
>100
80.0
–4
>100
70.0
>100
4.5
0.7
>100
20.0
>100
60.0
0.7
4.8
5.0
0.8
15.0
5.0
3.0
2.0
3.0
25.0
4.8
25.0
5.0
2.0
–10
1
2
3
4
5
6
7
8–19
0
1–22
3
4
5
6
7
8
9
>100
>100
>100
30.0
10.5
22.0
0.6
10.0
>100
10.0
>100
8.0
20.0
>100
>100
60.0
4.0
>100
0.15
>100
5.0
>100
60.0
>100
>100
>100
>100
20.0
6.0
>100
>100
>100
0.5
20.0
10.0
0.25
10.0
>100
10.0
>100
50.0
3.0
>100
60.0
>100
>100
0.4
20.0
12.0
10.0
18.5
10.0
0.9
0.4
20.0
7.0
0.4
20.0
10.0
20.0
0.06
20.0
>100
20.0
>100
15.0
0.01
0.01
8.0
18.2
18.0
4.0
>100
0.2
>100
>100
12.0
>100
15.0
1.0
>100
8.0
0.05
6.0
15.0
8.0
>100
15.0
>100
>100
8.0
>100
>100
6.0
>100
12.0
>100
60.0
40.0
>100
>100
30.0
20.0
50.0
52.0
6.0
50.0
7.0
80.0
25.0
45.0
0.6
0.7
3.0
4.0
9.0
40.0
30.0
40.0
95.0
90.0
5.0
2.0
2.0
6.0
7.5
30.0
10.0
>100
3.5
8.5
20.0
0.7
0.8
25.0
6.0
7.5
2.5
0.8
0.45
0.095
20.0
10.0
4.0
0.6
5.0
7.5
7.5
0.4
6.0
5.0
4.0
4.0
0.4
2.0
2.0
3.0
4.0
9.0
0.6
2.0
3.0
0.75
0.85
4.5
0.75
0.5
0.75
3.0
3.0
2.5
3.0
5.0
0
1
2
3
4
5
6
7.0
>100
0.09
>100
2.0
2.5
0.4
0.7
2.0
2.0
0.4
2.0
2.0
2.5
0.6
1.5
2.0
0.06
0.3
0.2
0.55
1.0
substituents exhibited cytotoxic activity against B-16
6
22 (Scheme 3). These compounds were found to be
inactive. (ii) 12-Ketoxime ethers/12-ketoxime ester deri-
vatives (24–36) (Scheme 4). It was observed that 12-
ketoxime ethers (24–26, 28–36) are more active than 12-
ketoxime ester (27) with an exception of compounds 25
and 26. Among the ether derivatives, compound 24
containing a terminal hydroxy group in alkyl chain is
less active than side chains with terminal N,N-dialkyl-
amino groups 28–31. Compound 28 having a ter-
minal alkylamino side chain with three carbon atoms
is slightly less active when compared to two carbon
atoms 29, likewise compounds 30, 31 with cyclicamino
side chains are also less active than n-alkylamino side
melanoma cell lines as reported in the literature, how-
ever, in the present in vitro eight panel cancer cell lines,
the compound exhibited cytotoxic activity at 20 mM. We
started our SAR studies from indoloquinazolines bear-
ing different substituents on A ring. Compounds 2–5 did
not have significant anticancer activity in human cancer
cells; however, compounds 6 and 8 with bromo and
amino substituents on the C-10 position showed better
activity. Particularly, compound 6 exhibited cytotoxic
activity at less than 1 mM concentration in four out of
eight cell lines. Compounds 7, 10, 11 and 12 with nitro,
methoxy, methyl and chloro substituents on the same
position exhibited weak activity. To optimize activity of
1
4
chain 28, 29. All these findings are similar even
1
5
8
, different amides 13–17 and sulfonamide 18 were syn-
thesized and tested for anticancer activity. Compound
6 with terminal halo functionality was more potent
with 10-bromo (35, 36) and 10-nitro (32, 33, 34)
derivatives.
1
with an average GI50 of 1 mM. However, when this
compound was tested for its in vitro stability in plasma
of both human and mouse, it was observed to be less
stable, contrary to the parent compound 8 which was
more stable.
Our efforts to improve the cytotoxic activity of com-
pound 1 resulted in compounds 28–36 which showed
promising in vitro cytotoxic activity with GI50 values
ranging from 1 to 5 mM. Compounds 29 and 33 with
good in vitro activity were tested for their in vivo activ-
ity in HFA using two different doses (75 and 150 mg/
1
2,13
At this juncture, additional analogues of 1 were prepared
to know the importance of the 12-keto functionality in
the biological activities of indoloquinazolines. Several
analogues lacking C-12 carbonyl functional group were
synthesized. The alterations that were tested included (i)
kg) in QDÂ4 schedule.
The actively growing differ-
ent types of cancer cells in hollow fibre capsules were
implanted in SAM aspectically under light ether anaes-
thesia in to the subcutaneous (sc) and intraperitonial
(ip) compartments. The anticancer activity in this model
was assessed based on percentage net growth in both the
1
2-hydroxy-12-alkyl/aryl disubstituted derivatives 19–

V. M. Sharma et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2303–2307
2307
Table 2. Results of xenograft studies
Compd MTDin SAM
Schedule
Dose
(mg/kg/day)
Total dose
(mg/kg)
Max. mean%
weight loss (day)
Optimum T/C%
(day)
(mg/kg)
a
2
3
9
3
400
400
QDÂ10
40
40
400
400
14 (24)
14 (26)
75 (24)
23 (17)
a
QDÂ10
a
QDÂ10, single dose for 10 days.
compartments. Each compound was tested against 12
human cancer cell lines. Compounds were selected for
further in vivo testing in xenografts by using the HFA
criteria, that is a %T/C of 50 or less in 10 of the 48
possible test concentrations (12 cell lines  2 sites  2
compound doses). To simplify evaluation, a points sys-
tem has been adopted which allows rapid viewing of the
activity of a given compound. For this a value of 2 is
assigned for each compound dose which results in a
graft studies are planned to assess the anticancer activity
with the sensitive cell lines from a hollow fibre assay.
Acknowledgements
We thank the analytical department for providing the
spectral data.
5
0% or greater reduction in a viable cell mass. The ip
and sc samples were scored separately. Compounds with
a total score ꢁ20 in IP+SC were referred for xenograft
testing. Compounds 29 and 33 scored 30/48, and 36/48,
respectively. These compounds were further evaluated
in HT-29 human colon adeno carcinoma xenografts in
male nu/nu mice (18–24 g, 10–14 weeks old). Tumor-
bearing athymic mice were given 40 mg/kg dose of the
test compounds through the intraperitoneal route in
QDÂ10 schedule. The tumor volume of each mouse was
determined by measuring two dimensions with vernier
calipers using the formula tumor volume=(length Â
References and Notes
1. George, V.; Koshy, A. S.; Singh, O. V.; Nayar, M. N. S.;
Pushpangadan, P. Fitoterapia 1996, LXVII, 553.
. Bergman, J.; Egestad, B.; Lindstrom, J. O. Tetrahedron
Lett. 1977, 30, 2625.
. Honda, G.; Tabata, M. Plantamedica 1979, 36, 85.
. Honda, G.; Tosirisuk, V.; Tabata, M. Plantamedica 1980,
8, 275.
. Honda, G.; Tabata, M.; Tsuda, M. Plantamedica 1979, 37,
72.
. Zou, J.; Huang, L. Yaoxue Xuebao 1985, 20, 45.
2
3
4
3
5
1
6
2
width )/2. The antitumor activity was expressed as the
7. Johnson, R. K.; Hertzberg, R. P. Ann. Rep. Med. Chem.
1990, 25, 129.
8. Coppola, G. M. Synthesis 1980, 505.
optimum T/C% (median tumor volume of control Â
1
00) (Table 2). Compounds showing an optimum T/C%
9
1
1
7
1
. Sandmeyer, T. Helv. Chim. Acta 1919, 2, 234.
0. Stolle, R. J. Prakt. Chem. 1922, 105, 137.
1. Piyasena, H.; Meanwell, N. A. Tetrahedron Lett. 1994, 35,
303.
value of <43% are considered as active. Compound 33
has shown a T/C% of 23 on day 17, this compound is to
be tested in other xenograft models. Compound 33 has
shown 76% bioavailability at 20 mg/kg dose whereas
compound 29 has shown only 14%. Eventhough com-
pound 29 has shown activity in HFA, it did not trans-
late in xenograft studies may be because of its poor
bioavailability.
2. Sriram, R.; Kumar, K. B. S.; Dhanvanthri, S. D.; Shra-
van, K. R.; Krishna, N. S.; Sony, P.; Kumar, R. A.; Rao, C. S.;
Rajagopalan, R. Proc. AACR 2000, 41, 200.
1
3. Melinda, G. H.; Michael, C. A.; Richard, F. C.; Betty,
J. A.; Joseph, G. M.; Lousis, M.; Michael, R. G. Life Sciences
995, 57, 131.
14. Compound 29: mp 152–153 C; IR: 3421, 1689,
1
ꢀ
In conclusion, in this communication we report the
synthesis of a new series of indoloquinazoline deriva-
tives and evaluation of their in vitro and in vivo anti-
cancer activity. Many of the compounds have shown in
vitro activity in the range of 1–5 mM concentration. Two
of the compounds with in vitro cytotoxic activity were
further evaluated in preliminary modified hollow fibre
assay to assess the anticancer property. Based on the
results of hollow fibre assay, compounds 29 and 33 were
tested in nude mice bearing HT-29 colon cancer xenografts
and compound 33 was found to be active. Further xeno-
À1
1
1
642 cm
J=6.0 Hz), 4.72 (2H, t, J=6.0 Hz), 7.36 (1H, t, J=7.6 Hz),
.55 (2H, q, J=7.6 Hz), 7.78 (1H, t, J=7.4 Hz), 7.95 (1H, d,
J=8.0 Hz), 8.30 (1H, d, J=7.6 Hz), 8.40 (1H, d, J=7.4 Hz),
; H NMR (CDCl ) d 2.35 (6H, s), 2.85 (2H, t,
3
7
+
8
1
1
2
.64 (1H, d, J=8.0 Hz); CIMS m/z 335 (MH , 100%).
ꢀ
5. Compound 33: mp 175–176 C; IR: 3419, 2927, 1688,
-1
1
599 CM ; H NMR (CDCl
.54 (2H, t, J=8.0 Hz), 4.76 (2H, t, J=6.0 Hz), 7.59 (1H, t,
3
) d 2.20 (2H, m), 2.30 (6H, s),
J=7.8 Hz), 7.83 (1H, t, J=8.0 Hz), 7.93 (1H, d, J=7.8 Hz),
8.41 (1H, d, J=8.0 Hz), 8.50 (1H, d, J=8.4 Hz), 8.82 (1H, d,
J=9.0 Hz), 9.10 (1H, s); CIMS m/z 394 (MH , 100%).
+