HYBRIDOMA Volume 9, Number 4, 1990 Mary Ann Liebert, Inc., Publishers
Production of a Monoclonal Antibody as Immunohistochemical Marker on Paraffin Embedded Tissues Using a New Immunization Method G.F.
PANCINO,1 E. OSINAGA,12 W. VORAUHER,3 A. KAKOUCHE,4 D. MISTRO,23 C. CHARPIN,5 and A. ROSETO1 'Division d'Immunocytologie Appliquée, (DICA) Université de Compiégne, BP. 649-60206 Compiégne Cedex, and UPR0043 CNRS Paris, France 2Departamento de Bioquímica, Facultad de Medicina, Montevideo, Uruguay 1 Laboratoire d Anatomopathologie de Compiégne, rue St. Joseph, Compiégne, France 4DICA Université de Compiégne, Compiégne, France 5Laboratoire d'Anatomie Pathologique, Faculté de Médecine, Marseille
ABSTRACT
This report describes a method for the production of murine monoclonal antibodies (MAbs) against cellular antigens preserved during formol fixation and paraffin embedding of human tissues in an attempt to select markers that would be useful in immunopathology. Hybridomas were prepared using spleen cells from mice immunized with cell suspensions obtained from formalin-fixed paraffin block sections of a human breast carcinoma. A monoclonal antibody 83 D. was selected, which was reactive with paraffin embedded breast carcinoma tissues, but not with normal breast. The reactive antigen has a high molecular weight ( 400-1000 kD) and was detected on the cell surface of live human breast cancer cell lines and on frozen tissues sections. These results demonstrate that the MAb 83 D. identifies a native breast tumor associated epitope conserved during tissue fixation and embedding and could be used as an immunohistochemical marker. INTRODUCTION monoclonal antibodies (MAbs) into immunohistological human tissue represents a useful tool for complementing classical morphological diagnosis of malignancies. (1-5). Immunohistochemical studies, eg. immunoperoxidase, can be performed either on frozen sections or on fixed and paraffin-embedded biopsy specimens. Although immunoenzymatic techniques on paraffin sections are simple methods which allow good morphologic analysis, not all MAbs are suitable for this purpose, since reactive antigens can be destroyed or modified by fixation and embedding (6,7). Membrane antigens are especially sensitive to these procedures, but some cytoplasmic and nuclear antigens likewise do not resist fixation and embedding. In immunohistochemistry studies of carcinoma of the breast, several examples exist concerning loss of reactivity after fixation and embedding ; these include nuclear antigens (8), cytoplasmic antigens (9), and membrane antigens (10-12). In the present paper, we report an immunization procedure using fixed and paraffin-embedded tissues as immunogens, with the aim of generating MAbs directed against antigens preserved during fixation and embedding procedures. We describe the generation of the MAb 83D., which identifies a breast carcinoma associated antigen not detectable on normal breast. The
introduction
of
procedures for studying
389
MATERIALS AND METHODS
-
Immunization
:
Human breast carcinoma tissues were fixed in 10 % formaldehyde buffer and embedded in paraffin. Diagnosis of poorly differentiated breast adenocarcinoma was assessed by a pathologist. Ten micron paraffin sections were cut from paraffin blocks and stored in sealed containers. Section deparaffinization was performed in tubes by two 10 min baths of toluene 100 % followed by two 10 min baths of ethanol 100 %, a procedure routinely used for immunohistochemical staining. At each step, supernatant was removed after centrifugation. Sections were washed three times with PBS and then minced with scissors. The cell suspensions were aspirated in a 1 ml syringe, leaving large fragments in the bottom, and BALB/c mice were immunized by intraperitoneal (i.p.) inoculation. Each immunization corresponded to about 7 paraffin sections. 60 days later mice were boosted with an i.p. inoculation of immunogen prepared from 7 paraffin sections. Mice were boosted i.p. twice more at fourteen-day intervals. Seven days after each boost, blood samples were collected from mice by retroorbital sinus bleeding and serum titre tested by immunoperoxidase on slides with deparaffinized tissue sections from the same breast carcinoma, used as immunogen. Spleens were removed for cell fusions four days after the last inoculation.
Hybridoma methodology
and
screening
Splenic lymphocytes of immunized mice were fused with the Sp20 murine myeloma cell line, at a rate of 4:1, using the method of Köhler and Milstein (13) with some modifications (14). Briefly cell suspensions of mice spleens and SP-0 were fused with 50 % (W/V) polyethylene glycol (Mr4000) in serum-free RPMI 1640. Fused cells were plated in Linbro 24-well plates and grown in selective medium (hypoxanthin-azaserin). Supernatants were screened on deparaffinized tissue sections from the breast carcinoma used immunization by the ABC for immunoperoxidase method (Vector Laboratories Inc. Burlingan). Eight-section slides, prepared mounting 2u tissue sections with 2% albumin, were deparaffinized and incubated for 30 min with normal horse serum and overnight at 4°C with supernatants. The subsequent steps were performed according to manufacturer's instructions and immunostaining was developed with 3-amino 9 ethyl carbazole. Negative controls (PBS or an irrelevant immune MAb) were performed for each experiment. Selected hybrids were cloned by limiting dilution. The subclass of monoclonal antibodies (MAbs) was determined by Ouchterlony immunodiffusion against antisera specific for the mouse lg classes (Nordic). Immunofluorescence
on cancer
cell lines
-
Cancer cell lines T47D (15), (16) and H466B (17), and MRC5 fibroblasts were grown on "Labtek" slides (MILES) for 24 h, washed with PBS and fixed in acetone for 10 min at 4°C. Slides were incubated with hybridoma supernatants for 1h at room temperature, washed in PBS and then incubated with FITC-labeled goat anti-mouse lg serum (Biosys) for 1 h. Slides were washed with PBS and mounted with 20 % glycerol in PBS. Alternatively, 100 ul of live cell suspension (10 /ml) in PBS, BSA 1%, NaN3 0.1% were incubated with hybridoma supernatant for 30 min on ice. After two washes, cells were resuspended in 200 ul of goat-anti-mouse FITC-labeled serum in PBS/BSA/azide and incubated for 30 min on ice. Cells were mounted in 20 % glycerol in PBS. Controls with an irrelevant immune IgM were performed for each experiment.
MCFy
Immunoblotting
MCF?
membranes NP-40 extracts were prepared as described previously (18). Sodium dodecyl sulfate gel electrophoresis was carried out in 3-10 % gradient gels and the proteins were then transferred to 0.2um nitrocellulose membranes at 40V overnight using a Biorad electroblot system with a 25 mM Tris-HCl (pH 8.3), 192 mM glycine 10 % methanol transfer buffer. Immunological reaction was performed as
described (19).
390
RESULTS
Immunization serum titer elicited by mouse immunization was evaluated after each booster in comparison with mouse sera taken before immunization, using immunoperoxidase staining on breast carcinoma deparaffinized sections. After the first injection, immunized mouse sera stained the majority of epithelial cells, without apparent staining of collagen or adipose tissue. The serum titer augmented with successive boosters, but collagen staining also appeared. After the third booster, cancer cell staining was found until a dilution of 1:2000. Collagen staining was not seen at dilutions over 1:1000, leaving only tumor cell staining (fig. 1A).
The
Iff t
—
*
i
**
k.
*
Production of a Monoclonal Antibody as Immunohistochemical Marker on Paraffin Embedded Tissues Using a New Immunization Method G.F.
PANCINO,1 E. OSINAGA,12 W. VORAUHER,3 A. KAKOUCHE,4 D. MISTRO,23 C. CHARPIN,5 and A. ROSETO1 'Division d'Immunocytologie Appliquée, (DICA) Université de Compiégne, BP. 649-60206 Compiégne Cedex, and UPR0043 CNRS Paris, France 2Departamento de Bioquímica, Facultad de Medicina, Montevideo, Uruguay 1 Laboratoire d Anatomopathologie de Compiégne, rue St. Joseph, Compiégne, France 4DICA Université de Compiégne, Compiégne, France 5Laboratoire d'Anatomie Pathologique, Faculté de Médecine, Marseille
ABSTRACT
This report describes a method for the production of murine monoclonal antibodies (MAbs) against cellular antigens preserved during formol fixation and paraffin embedding of human tissues in an attempt to select markers that would be useful in immunopathology. Hybridomas were prepared using spleen cells from mice immunized with cell suspensions obtained from formalin-fixed paraffin block sections of a human breast carcinoma. A monoclonal antibody 83 D. was selected, which was reactive with paraffin embedded breast carcinoma tissues, but not with normal breast. The reactive antigen has a high molecular weight ( 400-1000 kD) and was detected on the cell surface of live human breast cancer cell lines and on frozen tissues sections. These results demonstrate that the MAb 83 D. identifies a native breast tumor associated epitope conserved during tissue fixation and embedding and could be used as an immunohistochemical marker. INTRODUCTION monoclonal antibodies (MAbs) into immunohistological human tissue represents a useful tool for complementing classical morphological diagnosis of malignancies. (1-5). Immunohistochemical studies, eg. immunoperoxidase, can be performed either on frozen sections or on fixed and paraffin-embedded biopsy specimens. Although immunoenzymatic techniques on paraffin sections are simple methods which allow good morphologic analysis, not all MAbs are suitable for this purpose, since reactive antigens can be destroyed or modified by fixation and embedding (6,7). Membrane antigens are especially sensitive to these procedures, but some cytoplasmic and nuclear antigens likewise do not resist fixation and embedding. In immunohistochemistry studies of carcinoma of the breast, several examples exist concerning loss of reactivity after fixation and embedding ; these include nuclear antigens (8), cytoplasmic antigens (9), and membrane antigens (10-12). In the present paper, we report an immunization procedure using fixed and paraffin-embedded tissues as immunogens, with the aim of generating MAbs directed against antigens preserved during fixation and embedding procedures. We describe the generation of the MAb 83D., which identifies a breast carcinoma associated antigen not detectable on normal breast. The
introduction
of
procedures for studying
389
MATERIALS AND METHODS
-
Immunization
:
Human breast carcinoma tissues were fixed in 10 % formaldehyde buffer and embedded in paraffin. Diagnosis of poorly differentiated breast adenocarcinoma was assessed by a pathologist. Ten micron paraffin sections were cut from paraffin blocks and stored in sealed containers. Section deparaffinization was performed in tubes by two 10 min baths of toluene 100 % followed by two 10 min baths of ethanol 100 %, a procedure routinely used for immunohistochemical staining. At each step, supernatant was removed after centrifugation. Sections were washed three times with PBS and then minced with scissors. The cell suspensions were aspirated in a 1 ml syringe, leaving large fragments in the bottom, and BALB/c mice were immunized by intraperitoneal (i.p.) inoculation. Each immunization corresponded to about 7 paraffin sections. 60 days later mice were boosted with an i.p. inoculation of immunogen prepared from 7 paraffin sections. Mice were boosted i.p. twice more at fourteen-day intervals. Seven days after each boost, blood samples were collected from mice by retroorbital sinus bleeding and serum titre tested by immunoperoxidase on slides with deparaffinized tissue sections from the same breast carcinoma, used as immunogen. Spleens were removed for cell fusions four days after the last inoculation.
Hybridoma methodology
and
screening
Splenic lymphocytes of immunized mice were fused with the Sp20 murine myeloma cell line, at a rate of 4:1, using the method of Köhler and Milstein (13) with some modifications (14). Briefly cell suspensions of mice spleens and SP-0 were fused with 50 % (W/V) polyethylene glycol (Mr4000) in serum-free RPMI 1640. Fused cells were plated in Linbro 24-well plates and grown in selective medium (hypoxanthin-azaserin). Supernatants were screened on deparaffinized tissue sections from the breast carcinoma used immunization by the ABC for immunoperoxidase method (Vector Laboratories Inc. Burlingan). Eight-section slides, prepared mounting 2u tissue sections with 2% albumin, were deparaffinized and incubated for 30 min with normal horse serum and overnight at 4°C with supernatants. The subsequent steps were performed according to manufacturer's instructions and immunostaining was developed with 3-amino 9 ethyl carbazole. Negative controls (PBS or an irrelevant immune MAb) were performed for each experiment. Selected hybrids were cloned by limiting dilution. The subclass of monoclonal antibodies (MAbs) was determined by Ouchterlony immunodiffusion against antisera specific for the mouse lg classes (Nordic). Immunofluorescence
on cancer
cell lines
-
Cancer cell lines T47D (15), (16) and H466B (17), and MRC5 fibroblasts were grown on "Labtek" slides (MILES) for 24 h, washed with PBS and fixed in acetone for 10 min at 4°C. Slides were incubated with hybridoma supernatants for 1h at room temperature, washed in PBS and then incubated with FITC-labeled goat anti-mouse lg serum (Biosys) for 1 h. Slides were washed with PBS and mounted with 20 % glycerol in PBS. Alternatively, 100 ul of live cell suspension (10 /ml) in PBS, BSA 1%, NaN3 0.1% were incubated with hybridoma supernatant for 30 min on ice. After two washes, cells were resuspended in 200 ul of goat-anti-mouse FITC-labeled serum in PBS/BSA/azide and incubated for 30 min on ice. Cells were mounted in 20 % glycerol in PBS. Controls with an irrelevant immune IgM were performed for each experiment.
MCFy
Immunoblotting
MCF?
membranes NP-40 extracts were prepared as described previously (18). Sodium dodecyl sulfate gel electrophoresis was carried out in 3-10 % gradient gels and the proteins were then transferred to 0.2um nitrocellulose membranes at 40V overnight using a Biorad electroblot system with a 25 mM Tris-HCl (pH 8.3), 192 mM glycine 10 % methanol transfer buffer. Immunological reaction was performed as
described (19).
390
RESULTS
Immunization serum titer elicited by mouse immunization was evaluated after each booster in comparison with mouse sera taken before immunization, using immunoperoxidase staining on breast carcinoma deparaffinized sections. After the first injection, immunized mouse sera stained the majority of epithelial cells, without apparent staining of collagen or adipose tissue. The serum titer augmented with successive boosters, but collagen staining also appeared. After the third booster, cancer cell staining was found until a dilution of 1:2000. Collagen staining was not seen at dilutions over 1:1000, leaving only tumor cell staining (fig. 1A).
The
Iff t
—
*
i
**
k.
*
