HYBRIDOMA Volume U, Number 3, 1992 Mary Ann Liebert, Inc., Publishers

185 HER2/neu Epitope Mapping with Murine Monoclonal Antibodies FILIPPO CENTIS, *ELDA TAGLIABUE, SRINIVAS UPPUGUNDURI, RITA PELLEGRINI, STEFANIA MARTIGNONE, ANTONIO MASTROIANNI, SYLVIE MÉNARD, and MARIA I. COLNAGHI Istituto Nazionale per lo Studio e la Cura dei Tumori, Via Venezian 1, 20133 Milan, Italy *Present address: Divisione Ematologica e Centro Trapianto Midollo Osseo di Muraglia, U.S.L. Tre, Ospedale di Pesaro, Italy

Division

of Experimental Oncology E,

ABSTRACT In order to obtain further information on the biological role of the HER2/neu oncoprotein, 7 new monoclonal antibodies (MAbs) were produced against the pl85HER2 extracellular domain. These MAbs, together with two others previously produced, were used to investigate the pl85HER2 expression in breast carcinomas and compare the recognized antigenic determinants. The 7 reagents (MGR4,5,6,7,8,9 and 10), were shown to define five distinct epitopes. Three of these MAbs (MGR5,7,10), as well as one previously produced (MGR2), recognize the same epitope (Epitope-1) which seems, therefore, to be highly immunogenic for the murine immune system. Epitope-2 recognized by the MGR4 MAb, appears to be closely related to epitope-1 due to a cross inhibition between MGR4 and MGR 10, but not MGR2. Epitope-2 is the only one of the 5 also present on the product of the neu oncogene, the rat analogue of the human HER2/neu gene. None of the reagents against epitope-1 and epitope-2 were found to mediate receptor internalization, whereas MGR6 as well as a previously produced MAb (MGR3), both of which define epitope-3 and MGR8 which defines epitope-4, were found to do so. Epitope-4 like the neu-specific peptide recognized by the reference c-neu Ab3 MAb, was detectable on all pl85HER2 positive breast cancer, independently from the quantitative content of the oncoprotein, at variance with the other 4 epitopes whose availability on pl85HER2 for the relevant MAbs varied with the degree of overexpression. Epitope-5, recognized by the MGR9 MAb, on the contrary to the other epitopes, was prevalently localized at the basal membrane level of the tumor nodule.

INTRODUCTION a normal cellular gene present on chromosome 17 (1), encoding a protein which displays a close sequence homology with, but is distinct from, the EGF-R. This gene is the human analogue of the transforming neu gene, found originally in a rat neuroblastoma cell line (2)· The product of the c-erbB2 protooncogene is a transmembrane glycoprotein of 185KDa which, due to its structure (3) and localization on the membrane, seems to be a receptor for a new growth factor (4). In normal breast cells, c-erbB2 is present as a single gene copy, whereas its amplification and

C-erbB2 is

267

consequent overexpression have been found in 25-30% (5,6) of primary human breast cancer. This evidence stimulated a considerable clinical interest in the role played by the c-erbB2 gene in breast cancer. Following the report from Slamon's group that amplification of this gene is related to a poor prognosis (6), several other groups have confirmed the high incidence of amplification of the gene in breast cancer and the correlation between amplification and poor prognosis (7-11). Since different studies (12,13) demonstrated that, when overexpressed, the cerbB2 gene product represents a tumor-restricted marker, the use of monoclonal antibodies (MAbs) which recognize the extracellular domain of this protein in immunohistochemistry, could have diagnostic or prognostic implications. Moreover, the availability of several reagents which recognize different epitopes of the same molecule might be useful for a better understanding of the role of the oncoprotein and even for therapeutical approaches in breast cancer

patients.

In this paper we describe the production and characterization of 7 murine MAbs which all recognize the extracellular domain of the c-erbB2 gene product and compare them with two other MAbs previously produced (14). MATERIALS AND METHODS

Cell lines Human cell lines including lung adenocarcinoma Calu-3, breast carcinomas MCF-7, SK-BR-3, HBL100, colon carcinoma HT29 and vulvar epidermoid carcinoma A431 were provided by the ATCC (Rockville, MD). The ovary carcinoma cell line IGROV1 and the small cell lung carcinoma cell line POVD were kindly provided by Dr. J. Bénard (Institute G. Roussy, Villejuif), and Dr. G. Pratesi (Istituto Nazionale Tumori, Milan) respectively. Peripheral blood lymphocytes were obtained from healthy donors by fractionation on Ficoll-Hypaque

(Pharmacia, Uppsala, Sweden).

were all maintained in RPMI-1640 medium (Microbiological Associates, Walkersville, MD) supplemented with 10% fetal bovine serum (FBS), penicillin (lOO/ig/ml) and

The cell lines

streptomycin (100/ig/ml).

The murine mammary carcinoma cells were obtained from two transgenic mice kindly provided by Dr. P. Vezzoni (Istituto Tecnologie Biomediche Avanzate, CNR, Milan), in which the rat neu gene under the control of the murine mammary tumor virus promoter, was inserted.

Production of monoclonal antibodies BALB/c mouse received 3 s.c. and i.p. injections, at five-day intervals, of 2-4X106 Calu-3 live cells with incomplete Freund Adjuvant. Three days before killing the animal was recalled daily with an i.p. and i.v. injection of the immunizing material. The splenocytes were fused with the mouse myeloma line NSO (15) and the hybridoma supernatants were screened on a pool of different tumor cells including Calu-3, MCF-7, HT29, A431 cell lines and fresh lymphocytes. Calu-3 cells were previously stained by diamino-benzidine incorporation. The hybridomas, selected for reactivity of their supernatants with Calu-3 cells only, were repeatedly cloned and maintained in pristane pretreated BALB/c mice. To determine the isotype of the selected MAbs, an indirect immunofluorescence (IF) test on Calu-3 live cells was performed with fluoresceinated goat anti-mouse heavy-chain A

immunoglobulin (Meloy, Springfield, VA). Immunohistochemistry Live cell suspensions and cryosections were tested respectively by indirect IF and immunoperoxidase (IP) technique, using as a positive control the c-neu Ab3 MAb (Oncogene Science, Inc. Manhasset, NY), directed against an epitope located on the intracellular domain of pl85HER2 (16), at the concentration of 10/ig/ml. The indirect IF and IP techniques were respectively performed with fluorescein isotyocianateconjugated goat anti-mouse Ig (Meloy Laboratories Inc. Springfield, VA) and with Avidin biotin peroxidase complex (ABC) kits (Vector Burlingham, CA) by using MGR MAbs as purified antibody at the concentration of \0µg/ ïú. The results of both tests were scored according to the staining intensity: Strong (+ + +), Intermediate (+ +), Weak ( + ), Negative (-). Purification and radioisotope labelling of MAbs The antibodies of IgG isotype were purified from the ascitic fluid by affinity chromatography on a column prepacked with Protein G Sepharose 4 Fast Flow belonging to the MAb Trap G Kit (17). The purified MAbs were labelled by lactoperoxidase-catalyzed iodination. To 50 µ\ of each

268

MGR MAb were added 1 mCi of 125I Na (Amersham International, Little Chalfont UK), 3µg of 10 ul of 0.03% H20.¡. After 10 min. at room temperature, the samples were desalted on a Bio-Gel P-6 column (Biorad, Richmond CA).

lactoperoxidase, and

Radioisotope labelling and detergent solubilization of cells suspensions from Calu-3 and A431 lines were used for membrane lactoperoxidasecatalyzed iodination as previously described (14). Solubilization was carried out at 0°C for 45 min. by adding 1-2 ml of Tris-HCl 50 mM pH7.4 containing 1% NP40, 1% Antagosan (Boehring, l'Aquila, Italy) and 0.001M phenylmethylsulphonyl-fluoride (PMSF) (Sigma, St. Louis, MO). The lysates were centrifuged (15 min. at 10,000 g) and the soluble extracts were then precleared before immunoprecipitation. Immunoprecipitation Aliquots of the two precleared extracts, containing about 3x10e cpm were incubated with 50 µ\ of the different Seph.-MGR MAbs and immunoprecipitated using established procedures (14). Binding of the MAbs to Protein A-Sepharose CL4B was performed as follows: aliquots of 50 µ\ of protein-A Sepharose CL4B, coated with rabbit anti-mouse IgG heavy and light chains (100 /¿g/ml) were incubated overnight at 0°C with 500 µ\ of the different MAbs diluted 1:10. Competitive binding Competitive binding among MGR MAbs was carried out in 96-well plates using live Calu-3 cell monolayers. Different MGR MAbs or an unrelated MAb as control were used as ascitic fluid diluted 1:25, incubated for 45 min. at 0°C in the presence of a fixed amount of labelled MGRs. The plates were then washed and each well was refilled with µ of 2N Na OH for 20 min. at room temperature.The content of each well was then harvested and the bound radioactivity was determined in a gamma counter. As 100% binding the bound cpm in the presence of the unrelated MAb was used. No more than 10% variation was found as compared with the bound Live cell

cpm of the MAb alone.

Internalization

As regards the indirect uptake assay, unfixed monolayer-grown Calu-3 cells were incubated with the MGR MAb as hybridoma culture supernatant at 4° C. After lh the cells were washed twice with cold PBS added with 0.03% of BSA and then incubated for 3 h at 0 or 37°C with BSA-PBS. The cells were then washed, further incubated with 125I-labelled sheep anti-mouse whole antibody (lxlO5 cpm/well) in 50 µ\ volume for lh at 0°C, washed again twice, and then incubated for 20 min. at room temperature with 100^1 of NaOH 2N. Finally, the cell-bound residual radioactivity was counted. The non-specific binding of 125I-labelled sheep anti-mouse whole antibody to the cells was determined by treating control Calu-3 cells without MAbs in exactly the same manner as mentioned above. The controls for 100% residual cpm was the bound cpm before the 3 h incubation and for 0% the bound cpm obtained without the first MAb. In the direct uptake assay live Calu-3 cells (100,000/sample) were incubated at 0°C for 30 min. with the radiolabelled antibody (200,000 cpm/sample). After the first incubation period the cells were washed twice with PBS containing 0.03% bovine serum albumin and the total bound radioactivity was counted in a gamma counter. The cells were then suspended in PBS and incubated at 37°C for 3h while controls were incubated at 0°C. After the incubation period 500µ1 of PBS was added and the cells were pelleted. The cells were washed once and the amount of radioactivity still associated with the cell was measured. The cells were then resuspended in 0.2 ml of glycine buffer (0.05 M glycine-Hcl, 0.1 M Nací, pH2.8), kept at room temperature for 20 min., washed twice and the amount of radioactivity not dissociated by low pH was measured.

RESULTS

lung adenocarcinoma Calu-3 cell line was chosen to immunize mice due to pl85HER2 overexpression on the cell surface membrane. From the fusion of NSO myeloma cells with spleen cells from an immunized BALB/c mouse 800 hybrid colonies were obtained. After the 1st screening carried out as described in the Materials and Methods, among the 800 hybridomas tested, 31 colonies were selected for their specific reactivity with Calu-3 cells. The 2nd screening was performed on frozen acetone-fixed The human

269

I.F.

reactivity

TABLE 1. of 7 MGR MAbs

on

Reactivity

MAbs

with

tumor cell lines

with cell lines

pl85HER2

pl85HER2 overexpression*

without

overexpression* Calu 3

(lung ca.) MGR4 MGR5 MGR6 MGR7 MGR8 MGR9 MGR10

++ ++ ++ ++ ++ + +++

SK-BR-3

IGR0V1

(breast ca.)

HBL100

POVD

(ovary ca.) (breast ca.) (SCLC)

++ -

++ + ++ ++ + ++

-

-

-

-

-

-

Controlled by evaluating the reactivity with the c-neu Ab3 MAb Score-system: strong (+++); intermediate (++); weak (+); absent (-) *

sections of breast carcinoma previously selected for strong positivity or negativity with the c-neu Ab3 MAb. Seven supernatants which displayed c-neu Ab3 superimposable distribution of reactivity were selected and the relevant MAbs were designated as MGR4,5,6,7,8,9,10. According to an indirect IF test on Calu-3 cells using fluorescein-labelled goat anti-mouse heavy-chain immunoglobulin, one of the MAbs was of IgG2 class (MGR6), one of IgM class (MGR4), whereas the other 5 were of IgGl class. A further characterization of these MAbs was carried out on another panel of human cell lines with a different content of pl85HER2 oncoprotein, as evaluated by c-neu Ab3 reactivity. The results showed that all 7 MGR MAbs only reacted with 100% of Calu-3 and SK-BR-3 cells, i.e. cells from the two human cell lines which overexpressed pl85HER2 on the surface. On the contrary, they did not react with the other 3 neu-negative cell lines (Table 1). The antigens recognized by these 7 MAbs were analyzed by immunoprecipitation. To this aim the diluted ascitic fluid containing the MGR MAb to be evaluated was bound to SepharoseProtein -Rabbit anti-mouse Ig and used to immunoprecipitate 125I-labeIled surface proteins extracted from Calu-3. Two previously described MAbs (14) directed against pl85HER2 (MGR2 and MGR3) were also included as positive controls. The results in Fig.l showed that all the 7 new MAbs (lane a-g) specifically precipitated the same molecule recognized by MGR2 and MGR3 MAbs (lanes h, i). None of these reagents were found to precipitate the EGF-R, solubilized from the labelled A431 cells, which was precipitated by the control anti-EGF-R-MAb (data not shown). In order to determine whether the MGR MAbs recognized the same or different epitopes on the pl85HER2 molecule, competitive binding experiments were performed (Table 2). MGR2,5,7 and 10 were found to recognize the same epitope since the binding of either MGR2 or MGR10 iodinated MAbs was inhibited by either the MGR2, MGR5, MGR7, or MGR10 unlabelled MAbs. MGR4 was found to recognize a closely related epitope since it inhibited

f

p185—

g

h

s

Autoradiographic analysis of 12SI-labelled Calu-3 cell extract after radioimmunoprecipitation and SDS-PAGE. Material immunoprecipitated by: MGR4 (lane a), MGR5 (lane b), MGR6 (lane c), MGR7 (lane d), MGR8 (lane e), MGR9 (lane f), MGR10 (lane g), MGR2 (lane h), MGR3 (lane i). Fig.

1:

270

TABLE 2.

Competitive binding analysis

the Calu-3 cell line

Epitope

125I-MAbs

Unlabelled MAbs MGR2 MGR2 MGR5 MGR7 MGR10

on

++ + + +

MGR 10

MGR3

MGR8

MGR9

+ + + ++

MGR4 MGR3 MGR6

++ +

MGR8

++

MGR9

binding inhibition: inhibition; -: 60% inhibition;

+:

30-60%

MGR10 but not MGR2 binding. The labelled MGR3 cross-competed with itself and unlabelled MGR6. These two MAbs therefore defined a third epitope, whereas MGR8 and MGR9 were unique since they were unable to cross-compete with any of the other cold MGR MAbs. The nine tested MAbs, therefore, defined 5 different epitopes on the extracellular domain of

pl85HER2.

To evaluate the capacity of MGR MAbs to induce pl85HER2 internalization on Calu-3, an indirect uptake test, which measured the disappearance of the MAb from the Calu-3 cell surface, after incubation at 37 or 0°C, was performed. The results shown in Fig. 2 indicated that only those MAbs which recognized epitopes 3 and 4 were able to induce internalization. In fact, MGR3, MGR6 and MGR8 MAbs displayed a binding decrease which exceeded 50% on the Calu-3 cell surface after 3 h of incubation at 37°, but not at 0°C, whereas all the other MAbs were found to be stable on the membrane after incubation at either 0 or 37°C. To further

Fig. 2: Analysis of MGR MAb internalization on live Calu-3 cells by an indirect uptake test, which measured the % of residual cpm of the MAb on the cell surface after incubation at 37°C (Solid bar) or at 0°C (Hatched bar). 271

TABLE 3. MGR MAB-INDUCED INTERNALIZATION ON LIVE Calu-3 MEASURED WITH DIRECT TEST

% residual

MAb

binding

at 37°C

MGR2 MGR3 MGR6 MGR9

at OT

3

7 30 28 17

4 3 4

control the internalization, a direct test using labelled MAbs was performed on the Calu-3 cells. In keeping with the results of the indirect test, a 28 and 30 % of internalization was found with MGR6 and MGR3 respectively whereas MGR9 gave an intermediate value of 17% and MGR2 6% which is considered negative. To investigate whether the five determinants were specific for the human pl85HER2 oncoprotein all the MAbs were tested by membrane I.F. on mammary carcinoma cells obtained from two transgenic mice in which the rat neu-activated gene was inserted. Only MGR4, which recognized epitope-2, scored positive on these cells. To study the expression of the 5 different epitopes on human cancer, the nine MGR MAbs were tested on frozen sections of carcinomas previously selected for reactivity with the c-neu Ab3 MAb and divided into four groups depending on the staining intensity (strong, intermediate, weak or negative). The reactivity of all MGR MAbs, except MGR8, gradually decreased with the decrease of the pl85HER2 overexpression on the cellular membrane (Table 4). All the tested samples showing a strong c-neu Ab3 staining scored positive with all MGR MAbs, whereas a partial reactivity was Immunohistochemical

analysis

TABLE 4. of HER-2/neu

protein overexpression cryostatic sections positive cases/N" of tested cases

in

human breast carcinoma

Epitope

MAbs

N° +++*

1

Global 2

Global 3

Global 4

Global 5

MGR2 MGR5 MGR7 MGR10

positivity MGR4

4/4 4/4 4/4 4/4

100%

4/4

of

++

5/5 5/5 3/5 2/5

2/5 2/5 2/5 1/5

0/6 0/6 0/6 0/6

5/5

2/5

0/6

4/5 3/5

0/6 0/6

75%

positivity

100%

100%

MGR3 MGR6

4/4 4/4

5/5 4/5

positivity MGR8

positivity MGR9

100%

4/4

+

90%

5/5

35%

40%

80%

0%

0%

0%

100%

100%

100%

5/5

0/6

4/4

2/5

1/5

0/6

0%

100% 40% 20% 0% positivity *Level of measured pl85HER2 overexpression according to reactivity of the cryostatic sections with the c-neu Ab3 MAb. Global

the

Score-system: strong (+++); intermediate (++); weak (+); absent (-). 272

Fig. 3: Immunohistochemical test on frozen sections of ductal carcinoma, a) MGR6: homogeneous staining of the tumor cells. ( x300 )

Fig. 3: Immunohistochemical test on frozen sections of ductal carcinoma. b) MGR9: strong reactivity at the basal membrane level and weak staining (x300)

of the tumor cells.

observed on samples with intermediate and weak c-neu Ab3 staining, except for MGR8 which also reacted with the total number of these samples. All MGR MAbs except one (MGR9) showed a homogeneous staining of the tumor cell membrane (Fig. 3a), whereas normal cells present on the tissue sections were negative. On the contrary, MGR9 showed a peculiar staining pattern consisting of a strong reactivity at the basal membrane level in all the tested c-neu Ab3 positive cases and a weak labelling of the tumor cells only on sections which displayed c-neu Ab3 strong reactivity (Fig. 3b). 273

All nine MAbs

were unable to react on paraffin-embedded sections of breast carcinomas overexpressing pl85HER2 and gave negative staining in the Western blot analysis of the NP40 extract of Calu-3 (data not shown).

DISCUSSION

By using the human lung adenocarcinoma cell line Calu-3 as immunogen, which overexpresses the oncoprotein pl85HER2 on the cell surface, we obtained 7 new anti-HER2/neu MAbs. The results of immunohistochemistry tests on different tumors cell lines, together with those of immunoprecipitation experiments, show that these MAbs are directed against the extracellular domain of pl85HER2 and do not cross-react with EGF-R. These MAbs and two we previously produced (14) were used to analyze in depth the antigenic characteristics of the pl85HER2. In competitive binding experiments the nine reagents were shown to define five epitopes which display different characteristics. Four cross-reacting MAbs (MGR2, 5, 7 and 10) were obtained against epitope-1, which, therefore, seems to be highly immunogenic in mice. A low degree of conservation throughout the evolution of the relevant extracellular region or, alternatively, the location of this epitope in a region usually hidden or scarcely available on the cellular membrane could explain this result. The low reactivity of MGR2, 5, 7 and 10 MAbs on cryostatic sections of breast carcinomas with a weak pl85HER2 overexpression could support this second hypothesis. The MGR4-recognized epitope (epitope-2) appears to be closely related but not identical to epitope-1. In fact, MGR4 inhibited the binding of only one (MGR10) of the two tested MAbs which define epitope-1, which suggests a steric inhibition due to a close association of the two epitopes. Epitope-2 is the only one also present on the product of the rat neu oncogene, the analogue of the human HER2/neu.

None of the reagents against epitope-1 and epitope-2 were found to mediate receptor internalization. In our experiments, internalization has been evaluated as the disappearance from the membrane of the bound MAb after incubation at 37°C. This technique offers the advantage of not requiring purified antibodies and can, therefore, be used at a screening level. In the 4 cases tested in parallel, the indirect test was found to give results similar to those obtained with the direct test in which internalization is evaluated as the fraction of MAbs which cannot be acid-desorbded from the membrane. In addition ,the disappearance of the MAb from the membrane cannot be the result of the release of the complex pl85HER2-MAb, since the amount of soluble pl85HER2 found in the culture supernatant after incubation represents no more than 1% of the membrane pl85HER2 oncoprotein (data not shown). MGR6, which recognizes the same epitope-3 as the already described MGR3 MAb (14), was shown to induce a similar level of internalization. This observation suggests that internalization is epitope-dependent rather than MAb-dependent and that epitope-3 might be located in the ligand binding site. Nevertheless, another MAb designated MGR8, directed against epitope-4, was also found to mediate internalization. Since it is rather unlikely that pl85HER2 displays two different ligandbinding regions, probably the receptor internalization can be induced by different mechanisms one of which mimics the ligand binding effect. A second mechanism could be a receptor dimerization induced by the bivalent binding of some MAbs. In agreement, the results obtained with anti-EGF-R MAbs demonstrated that reagents either able (18) or unable (19) to block EGF binding to receptor mediated internalization. The peculiarity of epitope-5, recognized by the MGR9 MAb, is its predominant localization at the basal membrane level of the tumor nodule. This finding suggests a possible interaction between the oncogene product and some components of the basal membrane. Other MAbs produced against this oncoprotein (20) were used to define different epitopes of the extracellular domain of pl85HER2. Cross-competition between MGR MAbs and these reagents will be of value to investigate whether more epitopes than the five described here are present in the extracellular part of the receptor. Due to their capability to discriminate the quantitative expression of pl85HER2, MGR MAbs could be useful to define the overexpression level of the oncoprotein on tumoral cell membranes in diagnostic and/or prognostic immunohistochemical screening tests easy to carry out.

In addition, since different studies (12,13) demonstrated that overexpressed pl85HER2 appears to be a specific tumor marker, some of the MAbs here described could be possible candidates for in vivo radioimaging or therapeutic approaches.

274

ACKNOWLEDGMENTS The research activities leading to this work were developed in relation to a contract of the National Program of Research for Advanced Biotechnologies (Rif. 1146/193-07-8602), by Sudbiotec for the Italian Ministry of University and Scientific and Technological Research. Filippo Centis was supported by a grant from the Istituto Oncologico Pesarese "Raffaele

Antoniello". We thank Mrs. M. Mazzi and Mrs. P. Aiello for their excellent technical assistance, Ms. L. Mameli and Ms. M. Hatton for manuscript preparation and Mr. M. Azzini for photographic

reproductions.

REFERENCES

1.

Fukushige, S.I., Matsubara, K.I., Yoshida, M., Sasaki, M., Suzuki, T., Sembak, K., Toyoshima, K., Yamamoto, T. (1986) Localization of a novel v-erb B-related gene, C-erbB-2, on human chromosome 17 and its amplification in a gastric cancer cell line. Mol. Cell Biol. 6: 955-958. 2. Padhy, L.C., Shih, C, Lowing, D, Filkelstein, R., Weinberg, R.A. (1982) Identification of a phosphoprotein specifically induced by the transforming DNA of rat neuroblastomas. Cell 28: 865-871.

3.

Akiyama, T., Sudo, C, Ogawara, M., Toyoshima, ., Yamamoto, T. (1986) The product of the c-erb B-2 gene: a 185-Kilodalton glycoprotein with tyrosine kinase activity. Science 232: 1644-1646.

4.

Gullick, W.J., Berger, M.S., Bennett, P.L.P., Rothbard, J.G., Waterfield, M.D. (1987) Expression of the c-erb B-2 protein in normal and transformed cells. Int. J. Cancer 40: 246254.

5. Van de

6.

Vijver, M.J., Mooi, W.J., Peterse, J.L., Nüsse, R. (1988) Amplification and overexpression of the neu oncogene in human breast carcinomas. Eur. J. Surg. Oncol. 14 (2): 111-114. Slamon, D.J., Clark, G.M., Wong, S.G., Levin, W.J., Ullrich, ., McGuire, W.L. (1987) Human breast cancer correlation of relapse and survival with amplification of the HER-

2/neu oncogene. Science 235:

7.

8.

177-182.

Berger, M.S., Locher, G.W., Saurer, S., Gullick, W.J., Waterfild, M.D., Groner, B., Hynes, M.E. (1988) Correlation of c-erb-B2 gene amplification and protein expression in human breast carcinoma with nodal status and nuclear grading. Cancer Res. 48: 1238-1243. Slamon, D.J., Godolphin, W., Jones, L.A., Holt, J.A., Wong, S.G., Keith, D.E., Levin, W.J., Stuart, S.G, Vdove, S., Ullrich, A, Press, M.F. (1989) Studies of the HER-2/neu proto¬ oncogene in human breast and ovarian

9.

cancer.

Science 244: 707-712.

Paik, S., Hazan, R., Fisher, E.R., Sass, R.E., Fisher, B., Redmond, C, Schlessinger, J., Lippman, M.E., King, R. (1989) Pathologic findings from the national surgical adjuvant breast and bowel project prognostic significance of erbB-2 primary breast cancer. J. Clin. Oncol. 8: 103-112.

10.

in

Wright, C, Angus, B, Nicholson, S., Sinsbury, J.R.C., Cairns, J, Gullick, W.J., Kelly, P., Harris, A.L., Home, W. (1989) Expression of c-erbB-2 oncoprotein: a prognostic indicator in human breast

11.

protein overexpression

cancer.

Cancer Res. 49: 2087-2090.

Rilke, F., Colnaghi, MX, Cascinelli, N., Andreola, S., Baldini, M.T., Bufalino, R., Della Porta, G., Ménard, S., Pierotti, M.A., Testori, A. (1991) Prognostic significance of HER2/neu expression in breast cancer and its relationship to other prognostic factors. Int. J. Cancer 49: 44-49.

12. De Potter, C.R., Van Deale, S., Van De

Vijver, M.J., Pauwels, C, Maertens, G., De Boever,

275

J., Vandekerckhove, D., Roéis, H. (1989) The expression of the

neu oncogene product in Histopathology 15: 351-362. Natali, P.G., Nicotra, M.R., Bigotti, ., Venturo, I., Slamon, DJ., Fendly, B.M., Ullrich, A. (1990) Expression of the pl85 encoded by HER2 oncogene in normal and transformed

breast lesion and in normal fetal and adult human tissues.

13.

human tissues. Int. J. Cancer 45: 457-461.

Tagliabue, E., Centis, F., Campiglio, M., Mastroianni, ., Martignone, S., Pellegrini, R., Casalini, P., Lanzi, C, Ménardr S., Colnaghi, M.I. (1990) Selection of monoclonal antibodies which induce pl85HER2 internalization, phosphorylation and growth inhibition of cells with HER2/neu gene amplification. Int. J. Cancer 47: 933-937. 15. Kearney, J.F., Radbruch, ., Liesegang, ., Rajewsky, . (1979) A new mouse myeloma cell line that has lost immunoglobulin expression but permits the constitution of antibodysecreting hybridoma cell lines. J. Immunol. 123: 1548-1550. 16. Van De Vijver, M.J., Peterse, J.L., Mooi, W.J., Wisman, P., Lomans, J., Dalesio, O., Nusse, R. (1988) Neu-protein overexpression in breast cancer: association with comedo-type ductal carcinoma in situ and limited prognostic value in stage II breast cancer. New Engl. J. Med.

14.

319: 1239-1245.

17.

Nygren, P.A., Eliasson, M., Palmcrantz, E., Abrahamsen, L., Uhlen, M. (1988) Analysis and uses of serum albumin binding domain of streptococcal Protein G. J. Mol. Recog. 1: 540543.

18.

J. (1987) Toxicity of ligand and antibodyconjugates recognizing the epidermal growth factor receptor. J. Cell

Vollmar, A.M., Banker, D.E., Mendelsohn, directed ricin A Chain Phys. 131: 418-425.

Mayes EL, Stroobant P, et al. (1982) A monoclonal antibody to the human epidermal growth factor receptor. J. Cell Biochem. 20: 149.

19. Waterfield MD,

20.

Fendly, B.M., Winget, M., Hudziak, R.M., Lipari, M.T., Napier, M.A., Ullrich, A. (1990) Characterization of murine monoclonal antibodies reactive to either the human epidermal growth factor receptor or HER2/neu gene product. Cancer Res. 50: 1550-1558. to: Correspondence Dr. M.I.

Colnaghi

Division of Experimental Oncology E Istituto Nazionale Tumori Via G. Venezian 1 20133 Milan, Italy

Received for publication: 1/13/92 Accepted after revision: 2/6/92

276