443 -‘ hydroxylphenyl)-amino-6,7-dimethoxyquinazoline: A ... · Vol. 4, 1405-1414, June 1998 Clinical Cancer Research 1405 443 ‘-‘ hydroxylphenyl)-amino-6,7-dimethoxyquinazoline:

Vol. 4, 1405-1414, June 1998 Clinical Cancer Research 1405

443 ‘ -‘ hydroxylphenyl)-amino-6,7-dimethoxyquinazoline:

A Novel Quinazoline Derivative with Potent Cytotoxic Activity

against Human Glioblastoma Cells

Rama Krishna Narla, Xing-Ping Liu,

Dorothea E. Myers, and Fatih M. Uckun1

Departments of Experimental Oncology [R. K. N., F. M. U.],

Biotherapy [R. K. N., D. E. M., F. M. U.], and Chemistry [X-P. L.],and Drug Discovery Program [X-P. L., D. E. M., F. M. U.], HughesInstitute, St. Paul, Minnesota 55 1 13

ABSTRACT

The novel quinazoline derivative 4-(3’-bromo-4’-

hydroxylphenyl)-amino-6,7-dimethoxyquinazoline (WHI-

P154) exhibited significant cytotoxicity against U373 and

U87 human glioblastoma cell lines, causing apoptotic cell

death at micromolar concentrations. The in vitro antiglio-

blastoma activity of WHI-P154 was amplified >200-fold

and rendered selective by conjugation to recombinant

human epidermal growth factor (EGF). The EGF-P154

conjugate was able to bind to and enter target glioblas-

toma cells within 10-30 mm via receptor (R)-mediated

endocytosis by inducing internalization of the EGF-R

molecules. In vitro treatment with EGF-P154 resulted in

killing of glioblastoma cells at nanomolar concentrations

with an IC50 of 813 ± 139 flM, whereas no cytotoxicity

against EGF-R-negative leukemia cells was observed,

even at concentrations as high as 100 �LM. The in vivo

administration of EGF-P154 resulted in delayed tumor

progression and improved tumor-free survival in a severe

combined immunodeficient mouse glioblastoma xenograft

model. Whereas none of the control mice remained alive

tumor-free beyond 33 days (median tumor-free survival,

19 days) and all control mice had tumors that rapidly

progressed to reach an average size of > 500 mm3 by 58

days, 40% of mice treated for 10 consecutive days with 1

mg/kg/day EGF-P154 remained alive and free of detect-

able tumors for more than 58 days with a median tumor-

free survival of 40 days. The tumors developing in the

remaining 60% of the mice never reached a size >50

mm3. Thus, targeting WHI-P154 to the EGF-R may be

useful in the treatment of glioblastoma multiforme.

Received 2/4/98; revised 4/1/98; accepted 4/3/98.

The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

1 To whom requests for reprints should be addressed, at Hughes Insti-tute, 2665 Long Lake Road, Suite 330, St. Paul, MN 551 13. Phone:(612) 697-9228; Fax: (612) 697-1042.

INTRODUCTION

As the most malignant primary central nervous system

tumors, high-grade anaplastic astrocytoma and glioblastoma

multiforme respond poorly to contemporary multimodality

treatment programs using surgical resection, radiation therapy,

and chemotherapy with a median survival of < 1 year after

initial diagnosis (1-4). Consequently, the development of effec-

tive new agents and novel treatment modalities against these

very poor prognosis brain tumors remains a major focal point in

translational oncology research.

In a systematic effort to identify a cytotoxic agent with

potent antitumor activity against glioblastoma cells, we synthe-

sized several dimethoxy-substituted quinazoline derivatives and

examined their in vitro and in vivo effects on human glioblas-

toma cells. Here, we provide experimental evidence that the

novel quinazoline derivative WHI-P1S42 exhibits potent cyto-

toxic activity against human glioblastoma cells. Notably, target-

ing of WHI-P154 to the surface EGF-R further enhanced its

cytotoxic activity, resulting in rapid apoptotic death of glioblas-

toma cells at nanomolar concentrations in vitro and significantly

improved tumor-free survival in an in vivo SCID mouse

glioblastoma xenograft model.

MATERIALS AND METHODS

Synthesis and Analysis of Quinazoline Derivatives.All chemicals were purchased from the Aldrich Chemical Com-

pany (Milwaukee, WI) and were used directly for synthesis.

Anhydrous solvents such as acetonitrile, methanol, ethanol,

ethyl acetate, tetrahydrofuran, chloroform, and methylene chlo-

ride were obtained from Aldrich as Sure Seal bottles under

nitrogen and were transferred to reaction vessels by cannulation.

All reactions were carried out under a nitrogen atmosphere.

2 The abbreviations used are: WHI-P154, 4-(3’-bromo-4’hydroxylphe-

nyl)-amino-6,7-dimethoxyquinazoline; WHI-P79, 4-(3’-bromophenyl)-amino-6,7-dimethoxyquinazoline; WHI-P97, 4-(3’,5’-dibromo-4’-hy-

droxylphenyl)-amino-6,7-dimethoxyquinazoline: WHI-PI 1 1 . 4-(3’-

bromo-4’-methylphenyl)-amino-6,7-dimethoxyquinazoline: WHI-P131.

4-(4’-hydroxylphenyl)-amino-6,7-dimethoxyquinazoline; WHI-P132,

4-(2’-hydroxylphenyl)-arnino-6,7-dimethoxyquinazoline; WHI-Pl80,

4-(3’-hydroxylphenyl)-amino-6,7-dimethoxyquinazoline; WHI-P 197,

4-(3’-chloro-4’-hydroxylphenyl)-amino-6,7-dimethoxyquinazoline:

WHI-P258, 4-(phenyl)-amino-6,7-dimethoxyquinazoline; SCID, severecombined immunodeficient; EGF, epidermal growth factor; rhEGF, recom-binant human EGF; EGF-R, EGF receptor; NMR, nuclear magnetic reso-nance; 5, d, t, q, m, and br, singlet, doublet, triplet, quartet. multiplet, and

broad, respectively; IR, infrared; GC/MS. gas chromatography/mass spec-

trometiy; m.p., melting point; HPLC, high-performance liquid chromatog-raphy; Sulfo-SANPAH, sulfosuccinimidyl 6-(4’azido-2’-nitrophenylami-no)hexanoate; MU, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyb tetrazoliumbromide; GEN, genistein: TdT, terminal deoxynucleotidyl transferase:VTK, protein tyrosine kinase.

Research. on March 24, 2021. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

Table 1 6,7-Dimethoxyquinazoline derivatives

No R Formular mp(’C) MW�

WHI-P79 3-Br C,,H,4BrN,02 246.0-249.0 360

WHI-P97 3-Br, 5-Br, 4-OH C16H13Br2N,O, >300.0 455

WHI-Pi 1 1 3-Br, 4-CH, C17H16BrN,02 225.0-228.0 374

WHI-P131 4-OH C16H,,N,O, 245.0-248.0 297

WHI-P132 2-OH C16H13N,O, 255.0-258.0 297

WHI-P154 3-Br, 4-OH C,,H14BrN,O, 233.0-233.5 376

WHI-P180 3-OH C16H,,N,O, 256.0-258.0 297

WHI-P197 3-Cl, 4-OH C16H14C1N,O, 245.0(dec) 341

WHI-P258 H C16H,,N,02 258.0-260.0 281

Proton and carbon NMR (‘H and ‘3C NMR) spectra were

recorded on a Mercury 2000 Varian spectrometer operating at 300

and 75 MHz, respectively, using an automatic broad-band probe.

Unless otherwise noted, all NMR spectra were recorded in CDC13

at room temperature. ‘H chemical shifts are quoted in parts per

million (� in ppm) downfield from tetramethyl silane, which was

used as an internal standard at 0 ppm and s, d, t, q, and m. Melting

points were determined using a Fisher-Johns melting apparatus and

are uncorrected. UV spectra were recorded using a Beckman model

DU 7400 UVNis spectrometer with a cell path length of 1 cm.

Methanol was used as the solvent for the UV spectra. Fourier

Transform IR spectra were recorded using an FT-Nicolet model

Protege 460 instrument. The JR spectra of the liquid samples were

run as undiluted liquids using KBr discs. The KBr pellet method

was used for all solid samples. The GC/mass spectrum analysis was

conducted using a Hewlett-Packard GC/mass spectrometer model

6890 equipped with a mass ion detector and Chem Station soft-

ware. The temperature of the oven was steadily increased from

70#{176}Cto 250#{176}C,and the carrier gas was helium.

General Procedure for Synthesis of 6,7-Dimethoxy-

quinazoline Derivatives. The 6,7-dimethoxyquinazoline de-

rivatives for this study were prepared by the condensation of

4-chloro-6,7-dimethoxyquinazoline and the substituted anilines

as outlined in Scheme 1:

CH�Dj�J�N + � ‘ cH�R)fl

R a substituent; n = number

Specifically, a mixture of 4-chloro-6,7-dimethoxyquinazoline

(448 mg, 2 mmol) and the substituted aniline (2.5 mmol) in

ethanol (20 ml) was heated to reflux. Heating was continued for

4-24 h, an excess amount of Et3N was added to the solution,

and the solvent was concentrated to give the crude product,

which was recrystallized from dimethylformamide.

The key starting material, 4-chloro-6,7-dimethoxyquinazo-

line, was prepared using published procedures (5, 6) as outlined

in Scheme 2:

0 0

CH30-(�’jf”0H_(1)SOC�� CH30-�)”#{176}�’NH2 �

CH3O.5�.�.A. (2) NH4OH CH30�I��..k..NO NaBH�

( 1 ) (2)

CH:�1ZI(� HC0�H �

(3) (4)

Specifically, 4,5-dimethoxy-2-nitrobenzoic acid (compound 1)was treated with thionyl chloride, which was directly reduced

with ammonia to yield 4,S-dimethoxy-2-nitrobenzamide (com-

pound 2). Compound 2 was reduced with sodium borohydride in

the presence of catalytic amounts of copper sulfate to give

4,5-dimethoxy-2-aminobenzamide (compound 3), which was

directly refluxed with formic acid to yield 6,7-dimethoxyquina-

zoline-4(3H)-one (compound 4). Compound 4 was refluxed

with phosphorus oxytrichloride to give 4-chloro-6,7-dimethoxy-

quinazoline (compound 5) in good yield.

Quinazoline Derivatives and Their Physical Data. Ta-

ble 1 lists the quinazoline derivatives synthesized for the present

study. Selected analytical data for these compounds and their

precursors are as follows:

4,5-Dimethoxy-2-nitrobenzamide (precursor compound 2).

Yield 88.50%, m.p. 197.0-200.0#{176}C. ‘H NMR (DM50-cU: � 7.60

(s, 2H, �‘�2)’ 7.57 (s, 1H, 6-H), 7.12 (s, 1H, 3-H), 3.90, 3.87 (s, s,

6H, -OCH3). UV (methanol) X,,,..,�(#{128}):206.0, 244.0, 338.0 am. JR

(KBr) u,,�: 3454, 2840, 1670, 15 12, 1274, 1227 cm ‘. GC/MS

m/z 226 (M�, 10.0), 178 (98.5), 163 (100.0), 135 (51.0).

6,7-Dimethoxyquinazoline-4(3H)-one (precursor com-

pound 4). Yield 81.50%, m.p. 295.0-297.0#{176}C. ‘H NMR

(DMSO-d6): S 12.03 (br, 5, 1H, -NH), 7.99 (s, 1H, 2-H), 7.42 (s,

1H, S-H), 7.1 1 (s, lH, 8-H), 3.88, 3.85 (s, s, 6H, -OCH3). UV

(methanol) Xmax(� 212.0, 241.0 nm. JR (KBr) vm�: 3015,

2840, 1648, 1504, 1261, 1070 cm’. GC/MS mlz 206 (M�,

100), 191 (M� -CH3, 31.5), 163 (16.7), 120 (15.2).

4-Chloro-6,7-dimethoxyquinazoline (precursor compound

5). Yield 75.00%, m.p. 259.0-263.0#{176}C. ‘H NMR (DMSO-d6): �

8.75 (s, 1H, 2-H), 7.53 (s, 1H, S-H), 7.25 (s, 1H, 8H), 3.91 (s,

3H, -OCH3), 3.89 (s, 3H, -OCH3); UV (methanol) Xmax(�

207.0, 212.0, 241 .0 nm. IR (KBr) Umax: 2963, 2834, 1880, 1612,

1555, 1503, 1339, 1153, 962cm’. GC/MS m/z 224(M�, 100),

209 (M�-CH3, 9.4), 189 (19.39), 169 (10.55). Anal.

(C,QH9C1N202) C, H, N.

4-(3 ‘-Bromophenyl)-amino-6,7-dimethoxyquinazoline

(WHI-P79). Yield 84.17%, m.p. 246.0-249.0#{176}C. ‘H NMR

(DMSO-d6): S 10.42 (br, s, 1H, NH), 8.68 (s, 1H, 2-H), 8.07-

7.36 (m, SH, 5, 2’, 4’, 5’, 6’-H), 7.24 (s, 1H, 8H), 3.98 (s, 3H,

1406 Activity of WHI-Pl54 against Human Glioblastoma Cells

(5)

CH30-r�’j(�N

CH30-�L.�i

CH3O�J�J(’L)N( A),

CH�0 �



Clinical Cancer Research 1407

was used to generate a UV spectrum for each of the samples to

-OCH3), 3.73 (s, 3H, -OCH3 ); UV (methanol) Xmax(� 217.0,

227.0, 252.0 nm; IR (KBr) Umax 3409, 2836, 1632, 1512, 1443,

1243, l068cm’.GCIMSm/z36l (M� + 1, 61.83), 360 (Mt,

100.00), 359 (M�-l, 63.52), 344 (11.34), 222 (10.87), 140

(13.65). Anal. (C,6H,4BrN3O2) C, H, N.

4-(3 ‘,S ‘-Dibromo-4’-hydroxylphenyl)-amino-6,7-dimeth-

oxyquinazoline (WHI-P97). Yield 72.80%, m.p. >300.0#{176}C. ‘H

NMR (DMSO-d6): � 9.71 (s, lH, -NH), 9.39 (s, lH, -OH), 8.48

(s, lH, 2-H), 8.07 (s, 2H, 2’, 6’-H), 7.76 (s, 1H, 5-H), 7.17 (s,

lH, 8-H), 3.94 (s, 3H, -OCH3), 3.91 (s, 3H, -OCH3). UV

(methanol) �‘max(� 208.0, 210.0, 245.0, 320.0 nm; IR (KBr)

Um� 3504 (br), 3419, 2868, 1627, 1512, 1425, 1250, 1155

cm ‘. GC/MS m/z 456 (M� + 1, 54.40), 455 (M�, 100.00), 454

(M�-l, 78.01), 439 (M�-OH, 7.96), 376 (M�+ 1-Br, 9.76), 375

(Mt-Br, 10.91), 360 (5.23). Anal. (C,6H,3Br2N3O3) C, H, N.

4-(3’-Bromo-4’-methylphenyl)-amino-6,7-dimethoxyquin-

azoline (WHI-Pl 1 1). Yield 82.22%, m.p. 225.0-228#{176}C. ‘H

NMR (DMSO-d6): � 10.23 (s, lH, -NH), 8.62 (s, lH, 2-H), 8.06

(d, lH, f2.,6. = 2.1 Hz, 2’-H), 7.89 (s, 1H, 5-H), 7.71 (dd, lH,

f5..6. = 8.7 Hz, f2.,6. = 2.1 Hz, 6’-H), 7.37 (d, lH, f5.6 8.7

Hz, 5’-H), 7.21 (s, 1H, 8-H), 3.96 (s, 3H, -OCH3), 3.93 (s, 3H,

-OCH3). UV (methanol) �“max(�) 204.0, 228.0, 255.0, 320.0 nm.

IR (KBr) Um� 3431, 3248, 2835, 1633, 1517, 1441, 1281, 115S

cm’. GCIMS m/z 375 (M�+ 1, 76.76), 374 (M�, 100.00), 373

(M�-l, 76.91), 358(M�+ 1-OH, 1 1.15), 357 (1.42), 356 (6.31).

Anal. (C,7H,6BrN3O2.HCI) C, H, N.

4-(4’-Hydroxylphenyl)-amino-6,7-dimethoxyquinazoline

(WHI-P13l). Yield 84.29%, m.p. 245.0-248.0#{176}C. ‘H NMR

(DMSO-d6): � 11.21 (s, 1H, -NH), 9.70 (s, 1H, -OH), 8.74 (s,

lH, 2-H), 8.22 (s, 1H, 5-H), 7.40 (d, 2H, J = 8.9 Hz, 2’,6’-H),

7.29(s, 1H, 8-H), 6.85(d, 2H, J 8.9 Hz, 3’, 5’-H), 3.98 (s, 3H,

-OCH3), 3.97 (s, 3H, -OCH3). UV (methanol) Xmax(#{128}):203.0,

222.0, 251.0, 320.0 nm. JR (KBr)Um�: 3428, 2836, 1635, 1516,

1443, 1234 cm ‘ . GC/MS m/z 298 (M� + 1 , 100.00), 297 (M�,

26.56), 296 (Mtl, 12.46). Anal. (C,6H,5N303.HC1) C, H, N.

4-(2’-Hydroxylphenyl)-amino-6,7-dimethoxyquinazoline

(WHI-Pl32). Yield 82.49%, m.p. 255.0-258.0#{176}C. ‘H NMR

(DMSO-d6): � 9.78 (s, lH, -NH), 9.29 (s, lH, -OH), 8.33 (s, lH,

2-H), 7.85 (s, 1H, 5-H), 7.41-6.83 (m, 4H, 3’, 4’, 5’, 6’-H), 7.16

(s, 1H, 8-H), 3.93 (s, 3H, -OCH3), 3.92 (s, 3H, -OCH3). UV

(methanol) �“max(� 203.0, 224.0, 245.0, 335.0 nm. IR (KBr)

Um� 3500 (br), 3425, 2833, 1625, 1512, 1456, 1251, 1068

cm’. GC/MS m/z 298 (M�+l, 8.91), 297 (M�, 56.64), 281

(M�+l- OH, 23.47), 280 (Mt- OH, 100.00). Anal.

(C,6H,5N303.HC1) C, H, N.

4-(3 ‘-Bromo-4’-hydroxylphenyl)-amino-6,7-dimethoxy-

quinazoline (WHI-P154). Yield 89.90%, m.p. 233.0-233.5#{176}C.

‘H NMR (DMSO-d6): � 10.08 (s, 1H, -NH), 9.38 (s, lH, -OH),

8.40 (s, lH, 2-H), 7.89 (d, 1H, f2.6 = 2.7 Hz, 2’-H), 7.75 (s,

1H, 5-H), 7.55 (dd, lH, f5.6 = 9.0 Hz, f2.6 = 2.7 Hz, 6’-H),

7.14 (s, 1H, 8-H), 6.97 (d, lH, f5.6 9.0 Hz, 5’-H), 3.92 (s,

3H, -OCH3), 3.90 (s, 3H, -OCH3). UV (methanol) �‘max(�

203.0, 222.0, 250.0, 335.0 nm. IR (KBr) Umax 3431 (br), 2841,

1624, 1498, 1423, l244cm’. GCIMS mlz 378(M�+2, 90.68),

377 (M�+ 1, 37.49), 376 (M�, 100.00), 360 (Mt, 3.63), 298

(18.86), 282 (6.65). Anal. (C,6H14BrN3O3.HC1) C, H, N.

4-(3 ‘-Hydroxylphenyl)-amino-6,7-dimethoxyquinazoline

(WHI-P180). Yield 71.55%, m.p. 256.0-258.0#{176}C. ‘H NMR

(DMSO-d6): � 9.41 (s, lH, -NH), 9.36 (s, lH, -OH), 8.46 (s, lH,

2-H), 7.84 (s, lH, 5-H), 7.84-6.50 (m, 4H, 2’, 4’, 5’, 6’-H), 7.20

(s, 1H, 8-H), 3.96 (s, 3H, -OCH3), 3.93 (s, 3H, -OCH3). UV

(methanol) Xmax(� 204.0, 224.0,252.0, 335.0 nm. IR (KBr)

Um�: 3394, 2836, 1626, 1508, 1429, 1251 cm’. GCIMS m/z

297 (M�, 61.89), 296 (Mtl, 100.00), 280 (MtOH, 13.63).

Anal. (C,6H,5N301.HC1) C, H, N.

4-(3 ‘-Chloro-4’-hydroxylphenyl)-amino-6,7-dimethoxy-

quinazoline (WHI-P197). Yield 84.14%, m.p. 245.0#{176}C(dec). ‘H

NMR (DMSO-d6): �l 10.00 (s, 1H, -NH), 9.37 (s, 1H, -OH), 8.41

(s, 1H, 2-H), 7.78 (s, lH, 5-H), 7.49 (d, lH, J2.�,. 2.7 Hz,

2’-H), 7.55 (dd, lH, f56 = 9.0 Hz, f2.6 2.7 Hz, 6’-H), 7.16

(s, lH, 8-H), 6.97 (d, lH, f5.6 9.0 Hz, 5’-H), 3.93 (s, 3H,

-OCH3), 3.91 (s, 3H, -OCH3). UV (methanol) Xmax(#{128}):209.0,

224.0,249.0, 330.0 nm. JR (KBr) Umax: 3448, 2842, 1623, 1506,

1423, 1241 cm’. GC/MS m/z 341 (M�, 100.00), 326 (Mt

CH3, 98.50), 310 (MtOCH3 12.5), 295 (9.0), 189 (13.5), 155

(13.8). Anal. (C,6H14C1N303.HC1) C, H, N.

4-(Phenyl)-amino-6,7-dimethoxyquinazoline (WHI-

P258). Yield 88.26%, m.p. 258.0-260.0#{176}C. ‘H NMR (DM50-

d6): � 11.41 (s, lH, -NH), 8.82 (s, lH, 2-H), 8.32 (s, lH, 5-H),

7.70-7.33 (m, SH, 2’, 3’, 4’, 5’, 6’-H), 7.36 (s, lH, 8H), 4.02

(s,3H, -OCH3), 4.00 (s, 3H, -OCH3). UV (methanol) �‘max(�

210.0, 234.0, 330.0 nm. IR (KBr) Umax: 2852, 1627, 1509, 1434,

1248 cm’. GCIMS m/z 282 (M�+ 1, 10.50), 281 (M�, 55.00),

280 (M�-l, 100.00), 264 (16.00), 207 (8.50). Anal.

(C,6H,5N3O2) C, H, N.

Preparation of the EGF-P154 Conjugate. rhEGF was

produced in Escherichia coli harboring a genetically engineered

plasmid that contains a synthetic gene for human EGF fused at

the NH2 terminus to a hexapeptide leader sequence for optimal

protein expression and folding. rhEGF fusion protein precipi-

tated in the form of inclusion bodies, and the mature protein was

recovered by trypsin cleavage, followed by purification using

ion exchange chromatography and HPLC. rhEGF was 99% pure

by reverse-phase HPLC and SDS-PAGE with an isoelectric

point of 4.6 ± 0.2. The endotoxin level was 0.172 EU/mg. The

recently published photochemical conjugation method using the

hetero-bifunctional photoreactive cross-linking agent, Sulfo-

SANPAH (Pierce Chemical Co., Rockford, IL; Ref. 7), has been

used in the synthesis of the EGF-Pl54 conjugate. Sulfo-

SANPAH-modified rhEGF was mixed with a 10: 1 molar ratio

of WHI-P154 (50 inst solution in DMSO) and then irradiated

with gentle mixing for 10 mm with UV light at wavelengths

254-366 nm with a multiband UV light emitter (model

UVGL-15 Mineralight; UVP, San Gabriel, CA). Photolytic gen-

eration of a reactive singlet nitrene on the other terminus of

EGF-SANPAH in the presence of a 10-fold molar excess of

WHI-P154 resulted in the attachment of WHI-P1S4 to EGF.

Excess WHI-P154 in the reaction mixture was removed by

passage through a prepacked PD-lO column, and l2-kDa EGF-

EGF homoconjugates with or without conjugated WHI-Pl54, as

well as higher molecular weight reaction products, were re-

moved by size-exclusion HPLC. Reverse-phase HPLC using a

Hewlett-Packard 1 100 series HPLC instrument was used for

separation of EGF-P154 from EGF-SANPAH. After the final

purification, analytical HPLC was performed using a Spherisorb

ODS-2 reverse-phase column (250 X 4 mm; Hewlett-Packard).

Prior to the HPLC runs, a Beckman DU 7400 spectrophotometer



1408 Activity of WHI-P154 against Human Glioblastoma Cells

ascertain the Xmax for EGF-Pl54, EGF-SANPAH, and unmod-

ified EGF. Each HPLC chromatogram was subsequently run at

wavelengths of 214, 265, and 480 nm using the multiple wave-

length detector option supplied with the instrument to ensure

optimal detection of the individual peaks in the chromatogram.

Analysis was achieved using a gradient flow consisting of

0-100% eluent in a time interval of 0-30 mm. Five-pA samples

applied to the above column were run using the following

gradient program: 0-5 mm, 0-20% eluent; 5-20 mm, 20-100%

eluent; 25-30 mm, 100% eluent; and 30-35 mm, 100-0%

eluent. The eluent was a mixture of 80% acetonitrile (CH1CN),

20% H2O, and 0. 1% trifluoroacetic acid. Electrospray ionization

mass spectrometry (8, 9) was performed using a triple quadruple

mass spectrometer (PE SCIEX API; Finnigan, Norwalk, CT) to

determine the stoichiometry of P154 and EGF in EGF-P1S4.

Cell Lines. Human glioblastoma cell lines U87 and U373

were used as targets for various cytotoxic agents, including EGF-

P154. These brain tumor cell lines were obtained from American

Type Culture Collection (Rockville, MD) and maintained as con-

tinuous cell lines in DMEM supplemented with 10% fetal bovine

serum and antibiotics. The B-lineage acute lymphoblastic leukemia

cell line Nalm-6 was used as a negative control.

Uptake and Internalization of EGF-P154. Immunoflu-

orescence was used to: (a) examine the surface expression of

EGF-R on brain tumor cells; (b) evaluate the uptake of EGF-Pl54

by brain tumor cells; and (c) examine the morphological features of

EGF-P154-treated brain tumor cells. For analysis of EGF-R ex-

pression and cellular uptake of EGF-P1S4, U87 and U373 glio-

blastoma cells were plated on poly-L-lysine-coated, glass-bot-

tomed, 35-mm Petri dishes (Mattek Corp., Ashland, MA) and

maintained for 48 h. In uptake studies, the culture medium was

replaced with fresh medium containing S �g/mi EGF, EGF-P154,

or unconjugated WHI-Pl54, and cells were incubated at 37#{176}Cfor

5, 10, 15, 30, and 60 mm and 24 h. At the end of the incubation,

cells were washed twice with PBS and fixed in 2% paraformalde-

hyde. The cells were permeabilized, and nonspecific binding sites

were blocked with 2.5% BSA in PBS containing 0.1% Triton

x- 100 for 30 mm. To detect the EGF-R/EGF-P154 complexes,

cells were incubated with a mixture of a monoclonal antibody (I :10

dilution in PBS containing BSA and Triton X-lOO) directed to the

extracellular domain of the human EGF-R (Santa Cruz Biotech-

nobogies, Inc., Santa Cruz, CA) and a polyclonal rabbit anti-Pl54

antibody (dilution, 1:500) for 1 h at room temperature. After rinsing

with PBS, cells were incubated for 1 h with a mixture of a goat

anti-mouse IgG antibody conjugated to FITC (Amersham Corp.,

Arlington Heights, IL) and donkey anti-rabbit IgG conjugated to

Texas Red (Amersham Corp.) at a dilution of 1:40 in PBS. Simi-

larly, tubulin expression was examined by immunofluorescence

using a monocbonal antibody against a-tubulin (Sigma Chemical

Co., St. Louis, MO) at a dilution of 1:1000 and an anti-mouse IgG

conjugated to FITC. Cells were washed in PBS and counterstained

with TOTO-3 (Molecular Probes, Inc., Eugene, OR) for 10 mm at

a dilution of 1: 1000. Cells were washed again with PBS, and the

coverslips were mounted with Vectashield (Vector Labs, Burl-

ingame, CA) and viewed with a confocal microscope (Bio-Rad

MRC 1024) mounted in a Nikon Labhophot upright microscope.

Digital images were saved on a Jaz disc and processed with Adobe

Photoshop software (Adobe Systems, Mountain View, CA).

Cytotoxicity Assay. The cytotoxicity of various com-

pounds against human brain tumor cell lines was performed using

the MU assay (Boehringer Mannheim Corp., Indianapolis, IN).

Briefly, exponentially growing brain tumor cells were seeded into

a 96-well plate at a density of 2.5 X 10” cells/well and incubated

for 36 h at 37#{176}Cbefore drug exposure. On the day of treatment,

culture medium was carefully aspirated from the wells and replaced

with fresh medium containing the quinazoline compounds WHI-

P79, WHI-P97, WHI-Pl3l, and WFII-Pl54, unconjugated EGF, or

EGF-P154 as well as the tyrosine kinase inhibitory isoflavone

GEN, at concentrations ranging from 0.1 to 250 p.M. Triplicate

wells were used for each treatment. The cells were incubated with

the various compounds for 24-36 h at 37#{176}Cin a humidified 5%

CO, atmosphere. To each well, 10 p.1 of MTF (final concentration,

0.5 mg/mI) was added, and the plates were incubated at 37#{176}Cfor

4 h to allow MIT to form formazan crystals by reacting with

metabolically active cells. The formazan crystals were solubilized

overnight at 37#{176}Cin a solution containing 10% SDS in 0.01 M HC1.

The absorbance of each well was measured in a microplate reader

(Labsystems) at 540 am and a reference wavelength of 690 nm. To

translate the A5�, values into the number of live cells in each well,

the A� values were compared with those on standard A5� versus

cell number curves generated for each cell line. The percentage of

survival was calculated using the formula: % survival - Live cell

number[test]/Live cell number [control] X 100. The IC50s were

calculated by nonlinear regression analysis using Graphpad Prism

software version 2.0 (Graphpad Software, Inc., San Diego, CA).

In Situ Detection of Apoptosis. The demonstration of

apoptosis was performed by the in situ nick-end labeling method

using the ApopTag in situ detection kit (Oncor, Gaithersburg,

MD) according to the manufacturer’ s recommendations. Expo-

nentially growing cells were seeded in six-well tissue culture

plates at a density of SO X l0� cells/well and cultured for 36 h

at 37#{176}Cin a humidified 5% CO2 atmosphere. The supernatant

culture medium was carefully aspirated and replaced with fresh

medium containing unconjugated EGF or EGF-Pl54 at concen-

trations of 10, 25, or SO pg/ml. After a 36-h incubation at 37#{176}C

in a humidified 5% CO2 incubator, the supernatants were care-

fully aspirated, and the cells were treated for 1-2 mm with 0.1%

trypsin. The detached cells were collected into a 15-mb centri-

fuge tube, washed with medium, and pelletted by centrifugation

at 1000 rpm for S mm. Cells were resuspended in 50 p.1 of PBS,

transferred to poly-L-lysine-coated coverslips, and allowed to

attach for 15 mm. The cells were washed once with PBS and

incubated with equilibration buffer for 10 mm at room temper-

ature. After removal of the equilibration buffer, cells were

incubated for 1 h at 37#{176}Cwith the reaction mixture containing

TdT and digoxigenin-l l-UTP for labeling of exposed 3’-hy-

droxyl ends of fragmented nuclear DNA. The cells were washed

with PBS and incubated with antidigoxigenin antibody conju-

gated to FITC for 1 h at room temperature to detect the incor-

porated dUTP. After washing the cells with PBS, the coverslips

were mounted onto slides with Vectashield containing pro-

pidium iodide (Vector Labs, Burlingame, CA) and viewed with

a confocal laser scanning microscope. Nonapoptotic cells do not

incorporate significant amounts of dUTP due to lack of exposed

3 ‘-hydroxyl ends and consequently have much less fluorescence

than apoptotic cells, which have an abundance of exposed



150’

125�

-a- DMSO-a-. WHI.P79

WHI.P97

-.- WHI.P131

-0-’ WHI�P154‘-0- GEN

.-�-. WHI�P111

L

-.,- WHI.P197

-0- WH�P258

0

C0

0

0

CC>

C/)

a,0

75,

50�

25

0

10 100 1000 10 100 1000


3 x-P. Liu and F. M. Uckun. Inhibitors of EGF-receptor tyrosine kinase,manuscript in preparation.

Fig. 1 Activity of WHI-Pl54 and other

quinazoline compounds on U373 gliobbas-toma cells. Cells were incubated withWHI-P79, WHI-P97, WHI-PI 1 1, WHI-P131, WHI-Pl32, WHI-Pl54, WHI-Pl80,WHI-Pl97, WHI-P258, or GEN for 24 h in96-well plates, and the cell survival wasdetermined by the MTF assay. Data points

represent the means from three independ-ent experiments; bars, SE. The mean (±SE) IC30s were >250 p.M for WHI-P79,

161.2 ± 22.2 p.M for WHI-P97, >250 p.M

for WHI-P131, 167.4 ± 26.9 p.M for WHI-P154, >250 p.M for GEN, >250 p.M forWHI-Pl 1 1, >250 p.M for WHI-P132,178.1 ± 7.4 p.M for WHI-Pl80, >250 p.M

for WHI-P197, and >250 p.M for WHI-

P258. At 250 p.M, the percent survival val-ues were 69.9 ± 5.1% for WHI-Pl3l and43.9 ± 5.5% forWHI-P154 (P 0.02). At

125 p.M, the % survival values were 75.7 ±

5.2% for WHI-P13l and 53.2 ± 5.5% forWHI-P154 (P = 0.04).

3’-hydroxyl ends. In control reactions, the TdT enzyme was

omitted from the reaction mixture.

Maintenance of SCID Mouse Colony. The SCID mice

were housed in a specific pathogen-free room located in a secure

indoor facility with controlled temperature, humidity, and noise

levels. The SCID mice were housed in microisolater cages, which

were autoclaved with rodent chow. Water was also autoclaved and

supplemented with trimethoprim/sulfomethoxazol 3 days/week.

SCID Mouse Xenograft Model of Human Glioblastoma.

The right hind legs of the CB.17 SCID mice were inoculated s.c.

with 0.5 x 106 U373 human glioblastoma cells in 0.2 ml PBS.

SCID mice challenged with brain tumor cells were treated with

EGF-Pl54 (0.5 mg/kg/dose or 1 mg/kg/dose in 0.2 ml of PBS) as

daily i.p. doses for 10 treatment days, starting the day after inocu-

lation of the glioblastoma cells. Daily treatments with PBS, uncon-

jugated EGF (1 mg/kg/dose), and unconjugated W1-ll-Pl54 (1

mg/kg/dose) were used as controls. Mice were monitored daily for

health status and tumor growth and were sacrificed if they became

moribund, developed tumors that impeded their ability to attain

food or water, or at the end of the 3-month observation period.

Tumors were measured using Vernier calipers twice weekly, and

the tumor volumes were calculated according to the following

formula (10): (width2 X length)/2. For histopathological studies,

tissues were fixed in 10% neutral buffered formalin, dehydrated,

and embedded in paraffin by routine methods. Glass slides with

affixed 6-p.m tissue sections were prepared and stained with H&E.

Primary end points of interest were tumor growth and tumor-free

survival outcome. Estimation of life table outcome and compari-

sons of outcome between groups were done, as reported previously

(11-13).

RESULTS AND DISCUSSION

In Vitro Cytotoxicity of WHI-P154 against Human

Glioblastoma Cells. As shown in Fig. 1, WFII-P154, but not the

Drug Concentration (�.tM)

unsubstituted parent quinazoline compound WFH-P258, exhibited

significant cytotoxicity against the U373 human glioblastoma cell

line in three of three independent experiments with a mean (± SE)

IC50 of 167.4 ± 26.9 p.M and a composite survival curve IC50 of

158.5 p.M. The 4’-hydroxyl substituent on the phenyl ring was

essential for the anti-brain tumor activity of WHI-P154 because

WHI-P79, which differs from WHI-Pl54 only by the lack of the

4’-hydroxyl group on the phenyl ring, failed to cause detectable

cytotoxicity to U373 cells (Fig. 1). Replacement of the 4’-hydroxyl

group with a 4’-methyl group (W1-ll-Plll) resulted in substantial

loss of activity (Fig. 1B). The 3’-bromo substitution on the phenyl

ring likely contributes to the cytotoxicity of WHI-Pl54 because

WHI-Pl3l, lacking this bromo substituent, was less potent than

WHI-P154 (IC50s: >250 p�M versus 167.4 ± 26.9 p.M). Similarly,

2’-hydroxylphenyl (WHI-P132) and 3’-hydroxylphenyl (WH1-

P180) derivatives were less active than WHI-P154. Notably, the

replacement of the 3’-bromo substitution with a 3’-chloro substi-

tution yielded a less active agent (WHI-Pl97). Introduction of a

second bromo group at the 5’ position of the phenyl ring did not

result in altered cytotoxicity; the IC50 of WHI-P97 was 161.2 ±

22.2 p.M, which is virtually identical to that of WHI-Pl54 (Fig. 1).

In experiments not shown in this report, WHI-Pl54 was

found to be a potent inhibitor of the EGF-R tyrosine kinase as

well as Src family tyrosine kinases.3 Therefore, we initially

postulated that the cytotoxicity of WHI-Pl54 against glioblas-

toma cells was due to its tyrosine kinase inhibitory properties.

However, to our surprise, WHI-P79, which is an equally potent

inhibitor of EGF-R and Src family tyrosine kinases (14, 15),

failed to cause any detectable cytotoxicity to U373 glioblastoma



U373 �

,�..�.. .. , � ,.

.� �

4r., �

A

U87

B


Fig. 2 Cell surface expression of EGF receptor on U373 and U87human glioblastoma cells. Cells were fixed in paraformaldehyde, im-munostained with monocbonal antibody to EGF-R (green fluorescence),and counterstained with TOTO-3 (blue fluorescence). The immuno-stained cells were analyzed with a laser scanning confocal microscope.Blue fluorescence represents nuclei.

cells. Thus, the cytotoxicity of WHI-Pl54 to U373 cells cannot

be explained by its tyrosine kinase inhibitory properties alone.

This notion was further supported by the inability of the PTK

inhibitor GEN, which was included as a control, to cause de-

tectable cytotoxicity to U373 cells (IC50, >250 p.M; Fig. 1).

Enhanced Cytotoxicity of EGF-conjugated WHI-P154

against Human Glioblastoma Cells. In contrast to normal

glial cells and neurons, a significant portion of glioblastomas

express the EGF-R at high levels (16-20). Therefore, the

EGF-R could be used as a target for delivering cytotoxic agents

to glioblastoma cells with greater efficiency [reviewed by Men-

delsohn and Baselga (21)]. We confirmed the surface expression

of the EGF-R on the U373 and U87 human glioblastoma cell

lines with immunofluorescence and confocal laser scanning

microscopy using monoclonal antibodies to the extracellular

domain of the EGF-R. Both cell lines showed a diffuse granular

immunoreactivity with the anti-EGF-R antibody (Fig. 2).

In an attempt to enhance the antitumor activity of WHI-

P154 against glioblastoma cells by improving its targeting to

and cellular uptake by glioblastoma cells, we conjugated it to

rhEGF. We first examined the kinetics of uptake and cytotox-

icity of the EGF-P154 conjugate in U373 glioblastoma cells

Control ‘�. , � .

.�,. �t

�

‘�.-) � �:

�‘��i: �

A

r�

A’

WH�154: 30 ��

�.�

B ,

�i.

B’

EGF-P154 1Ormn�

�. . , .,.�

‘�*q�.#

‘�r�

C �

. .i�

� , ,

C’

EGF-P154: 30 mm

!�

� .- , ., i.’-4 �

D,,..

D’Fig. 3 EGF-R-mediated uptake of EGF-P154 by U373 human glio-blastoma cells. U373 cells were incubated with unconjugated WHI-

P154 (B and B’) or EGF-P154 (5 p.g/ml; C, C’, D, and D’) for theindicated times and processed for immunofluorescence to detect inter-nalized EGF-RJEGF-P154 complex using a monoclonal antibody to theEGF-R (green fluorescence) and a polyclonal antibody to WI-H-Pl54(red fluorescence) as described in “Materials and Methods.” Blue flu-

orescence represents the nuclei stained with TOTO-3. A and A’ : inuntreated cells, the EGF-R was localized primarily on the cell surface;no WHI-P154-like immunoreactivity was detected. B and B’: after a30-mm incubation with unconjugated WHI-P154, no change in EGF-Rimmunoreactivity was detected (B). No WHI-P154 immunoreactivity

was detected in the cells (B’). C and C’ : after a 10-mm exposure,

EGF-P154 was bound to the cell surface EGF-R (arrowheads) and

started internalizing into the cytoplasm. D and D’ : by 30 mm, most of

the EGF-R/EGF-Pl54 complexes were internalized and deposited in theperinuclear region (arrows).



WHI-Pi 54:24hr� ..� .

� ��... .s,

A

EGF-P1 54:24hr

�� ; .� �,.

�B B’

..- U87/EGF�P154

..- U87!WHI�P154

.‘- U87/EGF

.6- U87/EGF.GEN

�0- U373IEGF.P154

.0- U373IWHI�P154

-4- U373/EGF

-.. WHI.P154-6- EGF.P154

N

CC>

C/)

ci�025

10 100

B. NALM-6 cells

10001


Fig. 4 Uptake of WHI-P154 and EGF-P154 by U373 human glioblastoma cells.

U373 cells were incubated with unconju-

gated WHI-P154 (A and A’) or EGF-Pl54(5 p.g/ml; B and B’) for 24 h and processedfor immunofluorescence to detect EGF-Rmolecules using a monoclonal antibody tothe EGF-R (green fluorescence) and a poly-clonal antibody to WHI-P154 (red fluores-cence) as described in “Materials andMethods.” Blue fluorescence represents thenuclei stained with TOTO-3.

010 100 1000

Drug Concentration ( �.tM)

Fig. 5 A, cytotoxic activity of EGF-P154 against brain tumor cells. U373 and U87 glioblastoma cells were incubated with EGF, EGF-Pb54,EGF-GEN, or unconjugated WHI-P154 for 36 h in 96-well plates, and the cytotoxicity was determined by the MiT assay. The data points represent

the means from three independent experiments; bars, SE. The mean IC50s for EGF-P154 against U373 and U87 cells were 0.813 ± 0.139 p.M and0.620 ± 0.097 p.M. respectively. The mean IC50s for unconjugated WHI-P154 against U373 and U87 cells were 167.4 ± 26.9 p.M and 178.6 ± 18.5

p.M, respectively. B, cytotoxic activity of unconjugated WHI-Pl54 and EGF-P154 on EGF-R-negative leukemia cells. Nalm-6 cells were incubated

with unconjugated WHI-P154 or EGF-P154 for 36 h in 96-well plates, and the cell survival was determined by the MU assay. Unconjugated

WHI-Pl54 showed considerable cytotoxic activity on Nalm-6 cells (IC50, 9.1 ± 1.2 p.M), whereas EGF-Pl54 showed no cytotoxicity. The data points

represent the means from three independent experiments; bars, SE.



Control EGF

..� p’�..:‘�;:‘�., j...

EGF-P1 54

i,

‘�..�_:4._,�,- ,,�

A B

j’,:

s;��::��

C

EGF-P1 54

‘*�;�

0

fr

:. ,� � ,�,,*

�

, �,. .

.. .

e.

,

�.

,f� -

�,

. -!�, -4*�

‘S#{149}

�

*:�:�

�#{248}

-�

.-i

�,. --

1*! .�H

.� . � “e

�.

.-� � S t��P

-�, 4 --p!�r�’ � �

. �, �i.� .. - � .� -3�, 4:

- . � #{149}�.#{149}*

D� t:

�v �

2,! �. �:fr

5!

*�

- ,�4 :

a.-r� -

.j�

�- ‘ �..


Fig. 6 EGF-Pl54 induced apoptosis in brain tumor cells. A-C, U373 cells were incubated with 25 p.g/mb of EGF or EGF-Pl54 for 24 h and processedfor immunofluorescence using a monoclonal antibody to a-tubulin (green fluorescence). EGF-Pl 54 (but not EGF)-treated cells showed markedshrinkage with disruption of microtubules and lost their ability to adhere to the substratum. Blue fluorescence represents nuclei stained with TOTO-3.D-E, U373 cells were incubated with 25 p.g/ml of EGF-P154 for 36 h, processed for the in situ apoptosis assay, and analyzed with a laser scanningconfocab microscope. When compared with controls treated with EGF (D), several of the cells incubated with EGF-Pl54 (E) showed apoptotic nuclei

(yellow fluorescence). Red fluorescence represents nuclei stained with propidium iodide.

using immunofluorescence and confocal laser microscopy for

the internalized EGF-R and EGF-P154 molecules as well as for

evaluating the morphological changes in treated cells. EGF-

P154, similar to unconjugated EGF (not shown), was able to

bind to and enter target glioblastoma cells via receptor-mediated

endocytosis by inducing internalization of the EGF-R mole-

cubes. Within 10 mm after exposure to EGF-Pl54, the EGF-RJ

EGF-Pl54 complexes began being internalized, as determined

by cobocalization of the EGF-R (detected by anti-EGF-R anti-

body, green fluorescence) and EGF-Pl54 (detected by anti-P154

antibody, red fluorescence) in the cytoplasm of treated cells

(Fig. 3). By 30 mm, the internalized EGF-RIEGF-P154 com-

plexes were detected in the perinuclear region of the treated

glioblastoma cells. In contrast, cells treated with unconjugated

WHI-P154 alone did not reveal any detectable redistribution of

the surface EGF-R or cytoplasmic staining with the anti-P154

antibody (i.e., red fluorescence). By 24 h, WHI-P154 molecules

could also be detected in cells treated with unconjugated WH1-

P154 (Fig. 4). Thus, in accordance with our expectation, con-

jugation of WHI-P154 to EGF resulted in increased and more

rapid uptake of this cytotoxic quinazoline derivative by EGF-

R-positive glioblastoma cells.

We next sought to determine whether the improved deliv-

ery of WHI-Pl54 to glioblastoma cells by conjugation to EGF

results in potentiation of its antitumor activity. To this end, we

compared the cytotoxic activities of EGF-Pl54 and unconju-

gated WHI-Pl54 against U373 and U87 human glioblastoma

cell lines in dose-response studies using in vitro MU assays. As

shown in Fig. SA, EGF-Pl54 killed these glioblastoma cells in

each of three independent experiments at nanomolar concentra-

tions with mean IC50s of 813 ± 139 flM (range, 588-950 nM) for

U373 cells and 620 ± 97 nM (range, 487-761 nM) for U87 cells.



700

600

�50o

EaE� 4000>0

�300

200

100

-.- PBS

-0- EGF-‘.. WHI.P154 (1 mg/kg/d)

-.- EGF.P154 (0.5 mg/kg/d)

-0- EGF’P154(1 mg/kg/d)

A

a>

U)aa

L�.

0E

I-

-0- Control

-4-’WHI.P154 (1 mg/kg/d)

-0- EGF-P154 (0.5 mg/kg/d)

--EGF.P154(1 mg/kg/d)

10 20 30 40 50

Days After Inoculation60


The IC50s derived from the composite cell survival curves were

881 nsi for U373 cells and 601 nr�t for U87 cells. By comparison,

unconjugated WHI-Pl54 killed U373 or U87 cells only at

micromolar concentrations. EGF-P154 was 206-fold more po-

tent than unconjugated WHI-Pl54 against U373 cells (IC50s:

167.4 ± 26.9 p�M versus 81 1 ± 139 nM, P < 0.003) and

288-fold more potent than unconjugated WHI-P154 against U87

cells (IC50s: 178.62 ± 18.46 p.M versus 620 ± 97 nM, P <

0.001; Fig. SA). Unlike WHI-P154, which showed marked cy-

totoxicity against the EGF-R-negative NALM-6 leukemia cells,

EGF-P154 elicited selective cytotoxicity to EGF-R-positive

glioblastoma cell lines only (Fig. 5). Thus, conjugation to EGF

increased the potency of WHI-P154 against human glioblastoma

and at the same time restricted its cytotoxicity to EGF-R-

positive targets. Unlike the EGF-Pl54 conjugate, EGF-GEN, a

potent inhibitor of the EGF-R tyrosine kinase and EGF-R-

associated Src family PTK (22, 23), failed to kill glioblastoma

cells (Fig. SA). Thus, the potent cytotoxicity of EGF-Pl54

cannot be explained by the tyrosine kinase inhibitory properties

of its WFII-Pl54 moiety.

We next used immunofluorescence staining with anti-a-

tubulin antibody and the nuclear dye TOTO-3 in combination

with confocal baser scanning microscopy to examine the mor-

phobogical features of U373 glioma cells treated with either

unconjugated EGF or EGF-P154. After 24 h of exposure to 25

jig/mI EGF-Pl 54 (but not 25 p.g/ml unconjugated EGF), most

of the glioma cells showed an abnormal architecture with com-

plete disruption of microtubules, marked shrinkage, nuclear

fragmentation, and inability to adhere to the substratum (Fig. 6,

A-C). These morphological changes in EGF-P154-treated gli-

oma cells were consistent with apoptosis. To confirm apoptotic

DNA fragmentation in the nuclei of EGF-Pl54-treated glioblas-

toma cells, we used an in situ apoptosis assay that allows the

detection of exposed 3’-hydroxyl groups in fragmented DNA by

TdT-mediated dUTP nick-end labeling. As evidenced by the

confocal laser scanning microscopy images depicted in Fig. 6, D

and E, EGF-Pl54-treated (but not EGF-treated) glioma cells

examined for digoxigenin-dUTP incorporation using FITC-con-

jugated anti-digoxigenin (green fluorescence) and propidium

iodide counterstaining (red fluorescence) showed many apop-

totic yellow nuclei with superimposed green and red fluores-

cence at 36 h after treatment.

In Vivo Antitumor Activity of EGF-P154 in a SCID

Mouse Xenograft Model of Human Glioblastoma. CB.17

SCID mice develop rapidly growing tumors after s.c. inoculation of

0.5 x 106 U373 cells. We examined the in vivo antitumor activity

of EGF-Pl54 in this SCID mouse xenograft model of human

glioblastoma multiforme. EGF-Pl54 significantly improved tumor-

free survival in a dose-dependent fashion, when it was administered

24 h after inoculation oftumor cells. Fig. 7 shows the tumor growth

and tumor-free survival outcome of SCID mice treated with EGF-

P154 (500 �Lg/kg/day X 10 days or 1 mg/kg/day X 10 days),

unconjugated EGF (1 mg/kg/day X 10 days), unconjugated WHI-

P154 (1 mg/kg/day X 10 days), or PBS after inoculation with U373

glioblastoma cells. None of the 15 control mice treated with PBS

(n 5; median tumor-free survival, 19 days), EGF (n = 5; median

tumor-free survival, 23 days), or unconjugated WHI-P154 (n = 5;

median tumor-free survival, 19 days) remained alive tumor-free

beyond 33 days (median tumor-free survival, 19 days; Fig. 7A). All

0 10 20 30 40 50 60 70

Days After Inoculation

Fig. 7 In vivo antitumor activity of EGF-P154 against human glioblas-toma in SCID mice. Twenty-four h after s.c. inoculation of U373

glioblastoma cells, mice were treated with EGF-P154 (0.5 mg/kg/day X

10 days, n - 5, or I .0 mg/kg/day X 10 days, n = 5), EGF (1.0mg/kg/day x 10 days, n = 5), unconjugated WHI-Pl54 (1.0 mg/kg/day X 10 days, n 5), or PBS (n = 5). Mice were monitored for healthstatus and tumor growth. A compares the effect of these treatments on

tumor growth and progression. B depicts the tumor-free survival curves

from SCID mice treated with PBS or EGF (Control: n 10), uncon-

jugated WHI-P154 (n 5), or EGF-P154 (n = S for each of the twodose levels tested). Comparisons of outcome between groups were doneusing the log-rank test. *, no measurable tumors were observed.

of the five mice treated with EGF-P154 at the 500 p.g/kg/day dose

level developed tumors within 40 days with an improved median

tumor-free survival of 33 days (Fig. 7B), and the tumors were much

smaller than in control mice (Fig. 7A). Tumors reached a size of 50

mm3 by 37.5 ± 3.3 days in PBS-treated mice, 34.0 ± 3.0 days in

EGF-treated mice, 36.0 ± 5.1 days in WHI-Pb54-treated mice.

Tumors developing in EGF-Pl54 (500 p.g/kg/day X 10 days)-

treated mice reached the 50-mm3 tumor size -� I 1 days later than

the tumors in control mice treated with PBS, EGF, or WHI-Pl54

(47.4 ± 7. 1 days versus 35.8 ± I .8 days). The average sizes

(mean ± SE) of tumors at 20 and 40 days were 10.2 ± 1.4 mm3

and 92.3 ± 6.0 mm3, respectively, for mice in the control group

(i.e., PBS + EGF groups combined). By comparison, the average

sizes (mean ± SE) of tumors at 20 and 40 days were significantly



1414 Activity of WHI-Pl54 against Human Glioblastoma Cells

smaller at 1 .0 ± 1 . I mm3 (P = 0.002) and 37.6 ± 10.7 mm3 (P =

0.0003) for mice treated with EGF-P154 at the 500-p.g/kg/day dose

level. Notably, 40% of mice treated for 10 consecutive days with 1

mg/kg/day EGF-P154 remained alive and free ofdetectable tumors

for >58 days (PBS + EGF + WHI-Pl54 versus EGF-Pl54, P <

0.00001 by log-rank test). The tumors developing in the remaining

60% of the mice did not reach a size >50 mm3 during the 58-day

observation period. Thus, EGF-Pl54 elicited significant in vivo

antitumor activity at the applied nontoxic dose levels. The inability

of 1 mg/kg/day x 10 days of unconjugated WHI-Pl54 (53.2 nmol)

and unconjugated EGF to confer tumor-free survival in this SCID

mouse model in contrast to the potency of 1 mg/kg/day X 10 days

of EGF-P154 (corresponding to 2.9 nmol of WFII-P154) demon-

strates that: (a) the in vivo antitumor activity of EGF-Pl54 cannot

be attributed to its EGF moiety alone; and (b) conjugation to EGF

enhances the in vivo antitumor activity of WFII-P154 against glio-

blastoma cells by >18-fold.

Taken together, our findings provide unprecedented evi-

dence that WHI-P154 exhibits significant cytotoxicity against

human glioblastoma cells, and its antitumor activity can be

substantially enhanced by conjugation to EGF as a targeting

molecule. Although WHI-P154 is a potent inhibitor of the

EGF-R kinase as well as Src family tyrosine kinases, its cyto-

toxicity in glioblastoma cells cannot be attributed to its tyrosine

kinase inhibitory properties alone, because WHI-P79 with

equally potent PTK inhibitory activity failed to kill WHI-P154-

sensitive glioblastoma cells. Similarly, several PTK inhibitors

capable of killing human leukemia and breast cancer cells

lacked detectable cytotoxicity against glioblastoma cells. Glio-

bbastoma cells exposed to EGF-conjugated WHI-P154 under-

went apoptosis. Future identification of the molecular target for

WHI-P154 may lead to a structure-based design of potentially

more active antiglioblastoma agents. Finally, although we used

EGF to target WHI-P154 to glioblastoma cells in the present

study, other biological agents, including different cytokines

such as insulin-like growth factor and antibodies reactive with

glioblastoma-associated antigens, may be equally effective or

better targeting molecules for this novel quinazoline derivative.

REFERENCES

I . Pardos, M. D., and Wilson, C. B. Neoplasms of the central nervoussystem. in: J. F. Holland, R. C. Bast, Jr., D. L. Morton, E. Frei III, D. W.Kufe, and R. R. Weichselbaum (abs.) Cancer Medicine, Vol. I, pp.1471-1514. Baltimore: Williams and Wilkins, 1997.

2. Brandes, A. A., and Fiorentino, M. V. The role of chemotherapy inrecurrent malignant gliomas: an overview. Cancer Invest., 14: 551-559,

1996.

3. Finlay, J. L. Chemotherapeutic strategies for high grade astrocytoma

of childhood. In: R. J. Packer, W. A. Bleyer, and C. Pochedly (eds.),Pediatric Neuro-Oncology, pp. 278-297. Philadelphia: Harwood Aca-

demics, 1992.

4. Pardos, M. D., and Russo, C. Chemotherapy of brain tumors. Semin.

Surg. Oncol., 14: 88-95, 1998.

5. Nomoto, F., Obase, H., Takai, H., Hirata, T., Teranishi, M., Naka-

mura, J., and Kubo, K. Studies on cardiotic agents. I. Synthesis ofquinazoline derivatives. Chem. Pharm. Bull., 38: 1591-1595, 1990.

6. Thomas, C. L. Catalytic Processes and Proven Catalysts, pp. 1-27.New York: Academic Press, 1970.

7. Uckun, F. M., Evans, W. E., Forsyth, C. J., Waddick, K. G., Tuel-Ahlgren, L., Chelstrom, L. M., Burkhardt, A., Bolen, J., and Myers,D. E. Biotherapy of B-cell precursor leukemia by targeting genistein to

CDI9-associated tyrosine kinase. Science (Washington DC), 267: 886-891, 1995.

8. Feng, R., Konishi, Y., and Bell, A. W. High accuracy molecular

weight determination and variation, characterization of proteins up to 80

KU by ion spray mass spectrometry. J. Am. Soc. Mass Spectrom., 2:

387-401, 1991.

9. Covey, T. R., Bonner, R. S., Shushan, B. I., and Henion, J. D.Determination of protein oligonucleotides and peptides molecularweights by ion spray mass spectometry. Rapid Commun. Mass Spec-trom., 2: 249-256, 1988.

10. Friedman, S. H., Dolan, M. E., Pegg, A. E., Marcelli, S., Keir, S.,Catino, J. J., Binger, D. D., and Schold, S. C., Jr. Activity of temozo-bomide in the treatment of central nervous system tumor xenografts.Cancer Res., 55: 2853-2857, 1995.

11. Waurzyniak, B., Shneider, E. A., Tumer, N., Yanishevski, Y.,Gunther, R., Chelstrom, L., Wendorf, H., Myers, D. E., Irvin, J. D.,Messinger, Y., Ek, 0., Zeren, T., Chandan-Langlie, M., Evans, W. E.,and Uckun, F. M. In vivo toxicity, pharmacokinetics, and antileukemicactivity of TXU (anti-CD7)-pokeweed antivirab protein immunotoxin.Clin. Cancer Res., 3: 881-890, 1997.

12. Anderson, P. M., Meyers, D. E., Hasz, D. E., Covalcuic, K.,

Saltzman, D., Khanna, C., and Uckun, F. M. In vitro and in vivo

cytotoxicity of an anti-osteosarcoma immunotoxin containing pokeweedantiviral protein. Cancer Res., 55: 1321-1327, 1995.

13. Uckun, F. M., Gaynon, P., Sensel, M., Nachman, J., Trigg, M.,Steinherz, P., Hutchinson, R., Bostrom, B., Lange, B., Sather, H., andReaman, G. Clinical features and treatment outcome of childhood T-lineage acute lymphoblastic leukemia according to the apparent matu-rational stage of T-lineage leukemic blasts: a children’s cancer groupstudy. J. Clin. Oncol., 15: 2214-2221, 1997.

14. Bos, M., Mendelsohn, J., Kim, Y., Abbanell, J., Fry, D. W., andBaselga, J. PD153035, a tyrosine kinase inhibitor, prevents epidermalgrowth factor receptor activation and inhibits growth of cancer cells ina receptor number-dependent manner. Clin. Cancer Res., 3: 2099-2106,

1997.

15. Fry, D. W., Kraker, A., McMichael, A., Ambroso, L. A., Nelson,J. M., Leopold, W. R., Connors, R. W., and Bridges, A. L. A specificinhibitor of the epidermal growth factor receptor tyrosine kinase. Sci-

ence (Washington DC), 265: 1093-1095, 1994.

16. Khazaie, K., Schirrmacher, V., and Lichtner, R. B. EGF receptor inneoplasia and metastasis (Review). Cancer Metastasis Rev., 12: 255-

274, 1993.

17. Maruno, M., Kovach, J. S., Kelly, P. J., and Yanagihara, T. Trans-forming growth factor a, epidermal growth factor receptor, and probif-crating potential in benign and malignant gliomas. J. Neurosurg., 75:

97-102, 1991.

18. Yamazaki, H., Fukui, Y., Ueyama, Y., Tamaoki, N., Kawamoto, T.,

Taniguchi, S., and Shibuya, M. Amplification of the structurally andfunctionally altered epidermal growth factor receptor gene (c-erb-B) inhuman brain tumors. Mob. Cell. Biol., 8: 1816-1820, 1988.

19. Torp, S. H., Helseth, E., Dalen, A., and Unsgaard, G. Epidermalgrowth factor receptor expression in human glioma. Cancer lmmunol.Immunother., 33: 61-64, 1991.

20. Hoi Sang, U., Espintu, 0. D., Kelley, P. Y., Klauber, M. R., and

Hatton, J. D. The role of epidermal growth factor receptor in humangliomas: I. The control of cell growth. J. Neurosurg., 82: 841-846, 1995.

21. Mendelsohn, J., and Baselga, J. Antibodies to growth factors and

receptors. Biologic Therapy of Cancer: Principles and Practice, pp.607-623, 1995.

22. Uckun, F. M., Narla, R. K., Jun. X., Zeren, T., Venkatachalam, T.,

Waddick, K. G., Rostostev, A., and Myers, D. E. Cytotoxic activity ofEGF-genistein against breast cancer cells. Clin. Cancer Res., 4: 901-912,1998.

23. Uckun, F. M., Narla, R. K., Zeren, T., Yanishevski, Y., Meyers,

D. E., Waurzyniak, B., Ek, 0., Schneider, E., Messinger, Y., Chebstrom,L., Gunther, R., and Evans, W. In vivo toxicity, pharmacokinetics, and

anti-cancer activity of genistein linked to recombinant human epidermalgrowth factor. Clin. Cancer Res., 4: 1125-1134, 1998.



1998;4:1405-1414. Clin Cancer Res R K Narla, X P Liu, D E Myers, et al. against human glioblastoma cells.a novel quinazoline derivative with potent cytotoxic activity 4-(3'-Bromo-4'hydroxylphenyl)-amino-6,7-dimethoxyquinazoline:

Updated version

http://clincancerres.aacrjournals.org/content/4/6/1405

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/4/6/1405To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

443 -‘ hydroxylphenyl)-amino-6,7-dimethoxyquinazoline: A ... · Vol. 4, 1405-1414, June 1998 Clinical Cancer Research 1405 443 ‘-‘ hydroxylphenyl)-amino-6,7-dimethoxyquinazoline: