Znt. J . Cancer: 47, 847-852 (1991) 0 1991 Wiley-Liss, Inc.
Publication of the International Union Againa Cancer Publication de I'Union Internationale Contre le Cancer
CHARACTERIZATION OF A NOVEL MONOCLONAL ANTIBODY, 3H- 1 , REACTIVE WITH SQUAMOPROLIFERATIVE LESIONS AND SQUAMOUS-CELL CANCERS Peter G. PARSONS,J. Helen LEONARD, John H. KEARSLEY', Hiroyuki TAKAHASHI, Xu LIN-JIANand Denis J. MOSS Queensland Institute of Medical Research, Bramston Terrace, Herston QLO 4006, Australia. Immunization of mice with membranes from a virustransformed human keratinocyte cell line (KJD-I/SV40) yielded an IgM monoclonal antibody (MAb 3H-I) which reacted with the membrane and cytoplasm of KJD-l/SV40cells and in the perinuclear region of a squamous-cell carcinoma line (Colo- 16). lmmunohistochemistry of formalin-fixed, paraffin-embedded sections using MAb 3H-l gave intense staining of proliferating squamous epithelium in several characteristic patterns. Acanthotic squamous epithelium and well-differentiated squamous-cell carcinomas (SCC) demonstrated a membranousstaining pattern whereas psoriatic skin and undifferentiated SCC of the head and neck exhibited diffuse cytoplasmic, focal cytoplasmic or, in some tumours, no staining. Simple squamous epithelium was unreactive. Western blotting revealed an antigen of 55 kDa. The epitope recognized by MAb 3H-I may be a marker for particular stages of squamous proliferation and differentiation.
kindly provided by Prof. P. Gallimore, University of Birmingham, UK (Brown and Gallimore, 1987). Semi-confluent monolayers were synchronized in Gl/S phase by treatment with 5 m~ hydroxyurea for 18 hr (Kable and Parsons, 1988), then washed and re-fed with control medium. After 1 hr, cells were 3 m ~ washed with phosphate-buffered saline ( 5 m Na2HP0,, ~ KH2P04, 1 4 5 NaCI, ~ ~ ~ 7 . PBS) 2 ; and irradiated in PBS using 1440 J.mP2 of 254 nm UV (10 X 150 cm2 plates) or 144 J.mP2 of 254 nm UV (10 X 150 cm2 plates). The cells were harvested by scraping 4 hr (1440 J.md2 dose) or 20 hr (144 J.m-2 dose) after irradiation, and washed with PBS. After resuspension in 1 ml PBS containing 2 m~ phenylmethylsulfonyl fluoride (PMSF), the cells were lysed with 10 strokes of a Dounce glass homogenizer at O"C, then the 1,OOO g pellet was washed twice in PBS/PMSF and resuspended in 2 ml PBS/PMSF containing 2% Triton X-100. After 2 min at room temperature, the mixture was pelleted (1,000 g, 5 min) and the supematant (solubilized membrane) was stored at - 20°C. The supernatants from the 2 UV treatments were mixed prior to immunization.
Squamous-cell cancers (SCCs) represent a heterogeneous group of common human malignancies which arise from the epithelial lining of skin, oral cavity, pharynx, larynx, oesophagus, lung and female genital tract. Despite a histomorphologic appearance common to all SCCs in these anatomic regions, the Derivation of 3H-1 MAb clinical behaviour of individual tumours can vary widely The above supernatant mixture was emulsified with an equal (Davis, 1985). Furthermore, the diagnosis, classification and biologic potential of various squamous-cell hyperplasias, dys- volume of Freund's complete adjuvant and injected into 3 mice plasias and carcinomas is often difficult to assess because of i.p. (300 p1 per mouse), followed by similar amounts with the subjective nature of currently-used light microscopic crite- adjuvant at 14 days (i.p.) and 28 days (i.v.). After a further 7 ria (Boyd and Reade, 1988). Identification and characterization days, the spleen was harvested and fusion with mouse myof molecular markers associated with SCCs and their precursor eloma NS-1 cells was carried out essentially as described by lesions promise to become an important area of research ac- Kohler and Milstein (1975). Following selection in HAT medium, primary screening of 96 microtitre wells was carried out tivity (Kearsley, 1989). on a mixture of methanol-fixed UV-irradiated (144 J.mP2)and The advent of MAbs to various cellular antigens, oncogenes unirradiated KJD-1ISV40 cells grown in microtitre wells and, and oncoproteins has provided a novel, relatively quick, re- separately, on a mixture of human fibroblasts and MM96 huproducible and simple means of studying various characteris- man melanoma cells. Screening involved incubation of hybridtics of the malignant phenotype (Kohler and Milstein, 1975; oma supernatants with the fixed cells, followed by alkaline Evan et al., 1985). The use of MAbs to functional tissue- phosphatase-linked anti-mouse immunoglobulin and BCIPI specific antigens has led to a better understanding of the patho- NBT substrate solution (McEwan et al., 1988). Hybridomas genetic changes which occur during transformation from a be- which gave a positive reaction on KJD-USV40 (cells stained nign to a malignant state (Stewart et al., 1986), and may shed purple) and negative on the 2 other cell types were subcultured some light on those factors which determine tissue invasion and recloned for further screening. and metastases (Field et a l . , 1989; De Potter et al., 1990). For determination of the periodate sensitivity of the antigen, Herein we describe the characterization and staining pattern cells fixed in microtitre wells were treated with 10 mgiml KIO, of a novel IgM MAb, 3H- 1, which reacts on formalin-fixed for 5 min, washed with PBS and then incubated with 3H-1 tissues with squamo-proliferative lesions and squamous-cell MAb supernatant for the ELISA described above. cancers, but is unreactive with simple squamous epithelia and with most malignancies of non-squamous type. Isotype determination of 3H-I Determination of immunoglobulin isotype was carried out MATERIAL AND METHODS using the Mouse-Typer subisotyping kit (BioRad, Sydney, Australia). Briefly, aliquots (100 p,l) of KJD cell lysate Cell culture (2 X lo7 cells/ml in PBS) were placed in EIA plates (Titertek, The nature and derivation of the human cell lines are listed Flow, Brisbane, Australia) and incubated for 2 hr at 2°C to in Table I. Cells were cultured at 37°C in 5% CO,/air in RPMI allow attachment of antigen. After extensive washings with 1640 medium containing 10% (v/v) foetal calf serum, 100 PBS (PH 7.2), wells were treated with hybridoma supernatant pg/ml streptomycin and 100 I U / d penicillin. Routine assays for Mycoplasma by agar culture or Hoechst dye staining were negative. Immunogen KJD-l/SV40 (Table I) is a continuous line of human keratinocytes transformed by an SV40-adenovirus hybrid and
'To whom correspondence and reprint requests should be addressed. Received: November 5, 1990.
848
PARSONS ET A L .
(MAb 3H-1) for 1 hr at room temperature. Rabbit anti-mouse immunoglobulins (IgG,, IgG,,, IgG,,, IgG,, IgA and IgM) were applied to the wells for 1 hr at room temperature, followed by washings with PBS (PH 7.2). After treatment with goat anti-rabbit IgG horse-radish peroxidase . conjugate, the substrate mixture (hydrogen peroxide and 2,2'-azinodi [3ethylbenzthiazoline substrate]) was added; positive wells turned green. Western blotting Biopsies were dispersed by sonication in 400 ~1 of 2% (wU vol) SDS (sodium dodecylsulphate), 10 mM dithiothreitol and 2 m~ PMSF in 10 m~ TridHCl, PH 7.4. The samples were then placed in a boiling water bath for 2 min, allowed to cool, and centrifuged at 15,000g for 5 min. Samples were stored in aliquots at - 20°C; 2 0 4 samples (150 Fg of protein) were used for electrophoresis. Samples were frozen and thawed once only. Composition and electrophoresis of SDS 5-10% (wt/vol) polyacrylamide slab gels were essentially as reported by Laemmli (1970). Electrophoresis was performed on a minigel system at 60V for 2.5 hr at room temperature. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose paper (Amersham, Little Chalfont, UK) was performed using bicarbonate buffer (PH 9.9) on a transblot apparatus according to the manufacturers' instructions (BioRad). The antigen was detected by incubation with MAb supernatant 3H-1 followed by alkaline phosphatase-linked goat anti-mouse antibody (Silenus, Melbourne, Australia) diluted 1:1,000, followed by development with BCIP/NBT (McEwan et al., 1988). Immunohistochemistry Immunohistochemical staining was performed using the avidin-biotin-peroxidase (ABC) technique (Vector, Burlingame, CA). Briefly, tissue sections were deparaffinized, dehydrated in 100% alcohol and then rehydrated through decreasing alcohol concentrations to water. Sections were treated with 0.3% (wiv) hydrogen peroxide and 20% (wh) methanol in PBS for 10 min to eliminate endogenous peroxidase activity. Sections were then incubated with dilute horse serum for 20 rnin to eliminate non-specific binding of immunoglobulins. After removal of excess horse serum, 3H-1 MAb was added to each section as a supernatant and the slides were incubated for 30 min at room temperature. After several washes in PBS, sections were incubated with diluted biotinylated horse anti-mouse IgM immunoglobulin for 30 rnin at room temperature. The slides were then washed in PBS and incubated for 45 min with Vectastain Reagent (Vector), according to the manufacturer's directions. After a further wash in PBS, sections were incubated for 3-10 min with 3-amino-8-ethylcarbazoleto which 10 drops of 30% hydrogen peroxide had been added. The sections were finally counterstained with haemalum, rinsed and mounted under a coverslip with Glycergel (DAKO, Carpinteria, CA). All immunoperoxidase procedures included both positive and negative control slides. Positive control was performed by using a formalin-fixed section from an SCC tongue which consistently showed strong immunoperoxidase staining with 3H- 1 MAb. Negative controls were incubated with PBS or an irrelevant murine IgM antibody (reactive with formalin-fixed material) instead of 3H-1 antibody in the first step and were then treated as described above. Immunoelectron microscopy Within 15 min of removal from the operating theatre, tissue specimens (less than 1 mm3) were fixed with freshly-prepared 6 , 15 4% formaldehyde in 0.1 M cacodylate buffer, ~ ~ 7 . for min at room temperature. Osmium tetroxide post-fixation was omitted. After several buffer washes, specimens were dehydrated to 70% ethanol and embedded in White acrylic resin
(London Resin, Woking, UK). Resin was polymerized at 50°C for 24 hr. Ultra-thin sections were cut and mounted on nickel grids for immunolabelling . Sections were rinsed on drops of distilled water and nonspecific binding was blocked by a 30-min incubation with 5% BSA in modified Tris buffer (20 m~ Tris, 0.5 M sodium chloride, 20 mM sodium azide and 0.05% Tween 20). After several buffer washes, sections were incubated with a 1:10 dilution of MAb (3H-1 IgM purified ascites) for 2 hr at 37"C, rinsed thoroughly with buffer and then incubated with goat antimouse IgM-colloidal gold (Janssen, Beerse, Belgium, 10 nm particle size) for 1 hr at 37°C. After washing with buffer, sections were contrasted with uranyl acetate and lead citrate and examined in a Jeol 1200 EX transmission electron microscope. RESULTS
Derivation of MAb 3H-1 and reactivity with cultured human cells In an attempt to increase the probability of obtaining MAbs directed against membrane and proliferation-related antigens in SCC of the skin, the transformed keratinocyte line KJDl/SV40 was synchronized in G,/S with hydroxyurea (Kable and Parsons, 1988) and irradiated with UV in early S-phase, then the membrane fraction was employed for immunization. The successful derivation of MAb 3H- 1, the most promising of several similar clones, can perhaps be attributed to the membrane fractionation because the antibody reacted with the cell membrane in the majority of KJD-IISV40 cells (Fig. la). However, the UV irradiation and synchronization of cells used to prepare the immunogen did not appear to have been essential because the 3H-1 antibody reacted with asynchronous KJDI/SV40 cells whether irradiated or not. The isotype of MAb 3H-1 was found to be IgM. MAb 3H-1 reactivity, studied in a wider range of cultured human material (Table I), was restricted to the membrane and cytoplasm of KJD-l/SV40 cells and to the SCC cell line, Colo16. In this line the staining was mainly focal perinuclear and was found in a lower proportion of cells than in the KJD1/SV40 line (Fig. lb). Pretreatment of fixed cells with periodate to destroy carbohydrate epitopes did not abolish reactivity with 3H-I but eliminated the reactivity of melanoma cells with a tyrosinase antibody (McEwan et al., 1988), used in the present work as a positive control. Reactivity with histologic sections of normal tissue One hundred and ninety-six samples from a wide range of normal human tissues were tested for reactivity with MAb 3H-1. As shown in Table 11, staining with MAb 3H-1 was not present in any example of simple squamous epithelium, but was consistently present in epithelial cells of hair follicles, salivary glands, bronchial, stomach and small intestinal mucosae. MAb 3H-1 reactivity was infrequently demonstrated in bladder and prostate mucosae but was absent from all other normal tissues. Reactivity with squamous-cell carcinoma tissues and benign squamo-proliferative lesions As shown in Table 111, MAb 3H-1 reacted with a variety of benign and pre-malignant squamo-proliferative lesions. The staining patterns observed ranged from membranous in acanthotic epidermis and keratocanthoma to focal cytoplasmic in hyperproliferative psoriatic epithelium. The cornified layer above acanthotic squamous epithelium did not react with 3H-1 MAb, and there was often a very distinct disappearance of staining at the junctional region between hyperkeratoses and
MAb A N D PROLIFERATING SQUAMOUS CELLS
849
FIGURE1 - Immunohistochemical localization of the MAb 3H-1 antigen in KJD-I/SVM cells ( a ) and cola-16 cells (b). The secondary antibody was alkaline phosphatase-linked anti-mouse antibody and the substrate was BCIP/NBT. Bar = 50 Fm. FIGURE2 - Immunoreactivity of MAb 3H- 1 with formalin-fixed, paraffin-embedded tissue sections of (a) psoriasis (pennuclear staining pattern), bar = 50 pm; and (b)edge of a hyperkeratosis, showing 3H-1 reactivity in hair follicle and within the hyperplastic epithelium but not within adjacent simple squamous epithelium. Bar = 100 pm.
normal simple (non-staining) epidermis (Fig. 2b). In premalignant skin lesions, all 3 patterns of reactivity were seenfocal cytoplasmic in the immediate suprabasal region transforming to diffusely cytoplasmic/membranous in the spinous and granular layers, interspersed with occasional more superficial cells which demonstrated purely membranous reactivity. In all examples, basal epithelial cells were non-reactive. It is also clear from Table I11 that a large proportion of formalin-fixed sections of SCCs from human tissues react with 3H-1 MAb. These SCCs were derived from lesions of the head and neck, female genital tract, skin, lung and oesophagus. The SCC tissue specimens represent both primary and metastatic tumour sites and the histologic grading ranged from welldifferentiated, low-grade malignancies with the presence of large numbers of keratin whorls to high-grade, poorly differentiated SCCs with very little keratin production and with diffuse invading margins. The 3 staining patterns previously observed-membranous, diffuse cytoplasmic and focal cytoplasmic-were present in varying proportions in the SCC tissues. No staining occurred either in the keratin deposits or in the cells of the surrounding connecting tissue, skeletal muscles, blood vessels or lymphocytes. Less than 10% of poorly differentiated SCCs stained with 3H-1 MAb, whereas a majority of well-differentiated tumours stained in a membranous or membrano-cytoplasmic pattern. The lack of numbers in the poorly differentiated group, however, means that this clinical observation could not be supported statistically. It appeared that a
subgroup of moderately and poorly-differentiated SCCs demonstrated either no staining or a focal cytoplasmic staining pattern. Reactivity of histologic sections of non-squumous-cell carcinoma tissues
In a study of tumours other than SCC (Table IV), MAb 3H-1 only reacts consistently with adenocarcinomas of the stomach and colon. Most importantly, in relation to the potential use of MAb 3H-1 as a diagnostic aid, a consistently negative reaction was noted in virtually all basal-cell carcinomas, malignant melanomas, Merkel cell tumours, non-squamous lung cancers and most other adenocarcinomas. Determination of the molecular weight of the 3H-1 antigen by Western blotting
In repeated experiments, polyacrylamide gel electrophoresis of KJD-USV40 lysates followed by immunoblotting with MAb 3H-1 revealed strong immunoreactivity with a band at 55 kDa (Fig. 3), seen also in the Colo-16 line, tissue from a tongue SCC, and faintly in normal salivary gland. The SCC also showed a band at approximately 210 kDa when the lysate was stored at 4°C for 3 weeks (Fig. 3). No bands were seen in foreskin fibroblasts (Fig. 3) or the MM96 human melanoma cell line (not shown) which served as negative controls. Immunoelectron microscopy
Using 10-nm colloidal gold particles as the electron-dense
850
PARSONS ET AL.
TABLE I - REACTIVITY OF MAb 3H-I WITH CULTURED HUMAN CELLS Cell line
Transformed keratinocytes KJD-1lSV40
Colo 16 Melanoma MM96E
MM 127 MM 170 MM229 MM418 MM540 MM576 A2058 Others HeLa (cervical carcinoma) Foreskin fibroblasts Sch-I (Ewing’s sarcoma) RD-ES (Ewing’s sarcoma) G401 (Wilms’ tumour) NB2 (neuroblastoma)
Derivation of cell line
Brown and Gallimore, 1987 Moore et al., 1975
MAb 3H-1
reactivity localization
+ + (80%)’
membrane cytoplasmic + (30%) perinuclear focal
Whitehead and Little, 1973 Whitehead and Little, 1973 Whitehead and Little, 1973 Maynard and Parsons, 1986 Maynard and Parsons, 1986 Our laboratory’ Our laboratory’ Fabricant et at., 1977 ATCC
P.J. Smith Marshall and Kirchen, 1989 ATCC P.J. Smith
’Percentage of cells with stain.-’Established from lymph-node metastases of human melanoma.
marker, we observed, by irnrnunoelectron microscopy, that 3H-1 antigen was distributed generally within the cytoplasm of the fresh SCC cells examined (Fig. 4). The antigen was not localized to any particular organelle, nor to specific regions of the cytoplasm in the 2 SCC biopsies examined. Later immunohistochemistry showed one of these to have nil staining and the other (specimen seen in Fig. 4), to have mainly cytoplasmic staining. Colloidal gold particles were also present along the inner aspect of the cell membrane. Colo- 16 cells demonstrated very low cytoplasmic labelling, and immunoelectron microscopy of KJD-lISV40 cells was unsuccessful because of a high level of non-specific binding. Normal epidermis showed no labelling. DISCUSSION
This study confirmed the expectation that the virus-transformed human keratinocyte line KJD- 1/SV40 chosen as an immunogen expresses proliferation-related antigens specific to cells of keratinocyte lineage. Since over 80% of asynchronous KJD-USV40 cells reacted with MAb 3H-1, the relevant antigen appeared to be expressed in most phases of the cell cycle of a proliferating cell population. The basis of the different staining patterns observed between SCC lines has yet to be determined. The nature of the antigen or antigens recognized by the 3H- 1 MAb remains to be defined. The predominantly membranous/ interstitial staining reaction resembles the pattern seen with anti-desmosomal antibodies (Cowin and Garrod, 1983; Moll et
TABLE I1 - REACTIVITY OF MAb 3H-1 WITH NORMAL HUMAN TISSUES Normal tissues or cells
Positiveitested
Hair follicle Salivary gland Stomach (mucosa) Small intestine (mucosa) Prostate Bladder Bronchial mucosa Lung Breast Spleen Colon (mucosa) Liver Kidney ovary Lymph nodes Lymphocytes Polymorphonuclear leukocytes Red blood cells Muscle Pancreas Thyroid Pituitary Adrenal Heart Cartilage
20120 15/15 414 313 218 217 111 018 016 015 012 0110 017 016 017 0112 017 0115 0118 011 011 011 011 012 014
TABLE 111 - REACTIVITY OF MAh 3H-I WITH HUMAN SQL’AMOL’S-CELL CARCINOMA TISSCI: AN11 WITH BENIGN 1,ESIONS
Tissues
Squamous-cell carcinoma Oral cavity Larynx Pharynx Skin Lung Ectocervix Oesophagus Unknown primary Bowen’s disease (in-situ SCC) Benign lesions Acanthotic epidermis Keratoacanthoma Psoriasis Intradermal naevi Chromophobe adenoma (pituitary)
Poritiveirerted
24/32 20126 16/20 16/16 518
214 112 111 616 12112 10110 818 018 011
al., 1986), none of which, however, were specific for proliferating squamous epithelia (Moll et al., 1986; Vilela ef al., 1987). The localized juxtanuclear staining seen in our study may represent desmosomal internalization, which occurs in many different types of carcinoma (Garrod and Cowin, 1985; Henderson et al., 1986). The 210-kDa band found in the SCC is similar in molecular weight to the desmosomal protein dpl (normally 215 kDa; Miller et al., 1987) but the 3H-1 epitope was most commonly found strongly expressed in a 55-kDa antigen, suggestive of a cytokeratin-like molecule possibly associated with a range of subcellular structures. The immunoelectron microscopy results do not shed further light on this issue because the distribution of gold-labelled particles was generalized within the cytoplasm of the 2 samples which gave a positive result and the particles were not sufficiently numerous to exclude the possibility of satellite associations as found for dpl (Miller et al., 1987). The strong cytoplasmic association of the antigen in secretory tissue may indicate an involvement in the function of salivary gland and colonic epithelium. The fact that antigenicity was not abolished by pretreatment with periodate suggests that the MAb 3H-1 will be useful in identifying the peptide component of the antigen.
851
MAb AND PROLIFERATING SQUAMOUS CELLS TABLE IV - REACTIVITY OF MAb 3H-1 WITH HUMAN NON-SQUAMOUS MALIGNANCIES
Carcinoma tissue
Positiveitested
Melanoma Basal-cell carcinoma Non-Hodgkin’s lymphoma Ovarian cancer Colon cancer Adenocarcinoma lung Small-cell carcinoma lung Transitional-cell carcinoma bladder Breast cancer Osteosarcoma Thyroid cancer Renal cancer Testis cancer Prostate cancer Hodgkin’s disease Stomach cancer Syringoma Eccrine poroma Cholaneiocarcinoma
0110 0110 013
3/10
414 012
013
3/10
018 014
012 014
016 218 016
313 01 1 012 316
FIGURE4 - Immunoelectron micrograph showing localization of MAb 3H-1 in SCC. Labelling, as indicated by 10 nm colloidal gold particles, is seen within the cytoplasm and along the cell membrane. CM, cell membrane; D, desmosome. Bar = 0.5 nm.
FIGURE3 - Molecular weights of MAb 3H-1 antigens determined by Western blotting of cells and tissues. FS, normal foreskin fibroblasts (negative control); SG, salivary gland ( 5 5 ma);SCC, tongue SCC (210 and 55 kDa, lysate stored at 4°C for 3 weeks); Co, SCC cell line Colo-I6 (55 ma);KJD, transformed keratanocyte cell line KJD1lSV40 (55 m a ) and MWT position of the molecular weight markers
clear pattern similar to that seen in many of the aggressive SCCs. This finding suggests to us that immature, rapidly dividing squamous cells in psoriasis and aggressive SCCs share a common epitope which is localized in the perinuclear region of the disordered squamous cell. There are clearly many other factors which distinguish disordered squamous-cell proliferation in conditions such as psoriasis and hyperkeratoses from that occurring in the case of invasive SCCs. Identification of different cellular phenotypes may be helpful in understanding the mechanism underlying transformation from a benign or proliferative state to a malignant one. Our results suggest that MAb 3H-1 may be a valuable diagnostic and function-related marker for squamo-proliferative lesions, including SCCs. The epitope recognized by the 3H-1 antibody is not expressed at detectable levels in simple keratinizing squamous epithelium and only rarely in non-squamous malignancies, but has been demonstrated in all cases of solar keratoses, Bowen’s disease, keratoacanthomas, psoriasis and a high proportion of SCCs in various localization patterns and molecular weights. ACKNOWLEDGEMENTS
in kDa.
Whatever the function and chemical composition of the antigen recognized by MAb 3H-1, the latter is clearly an IgM antibody resistant to routine histologic processing and therefore has potential for immunodiagnosis of squamous lesions. In addition to being expressed in SCCs, the epitope is also expressed in rapidly dividing, non-malignant squamous cells. In the case of psoriasis, a hyperproliferative disorder of immature squamous epithelial cells, MAb 3H-1 reacts in a focal perinu-
W e thank Mrs. J. Middleton for her assistance in typing the manuscript. This work was supported by the Queensland Cancer Fund and the H.L. Rose Foundation. J.H.K. was in receipt of a Research Fellowship from the Queensland Radium Institute. Prof. P.J. Smith, Oncology Program, University of Queensland, kindly provided cell lines Sch-1 and NB2, and the HeLa and RD-ES cell lines were obtained from the ATCC, Rockville, MD. Immunoelectronmicroscopy was performed by Ms. D. Stenzel, Analytical Electron Microscopy Facility, Queensland University of Technology, Brisbane.
REFERENCES
BOYD,N.M. and READE,P.C., Differences between preneoplastic cells, neoplastic cells and their normal counterparts. J . Oral Path., 17,257-265 (1988). BROWN,K.W. and GALLIMORE, P.H., Malignant progression of an SV,transformed epithelial keratinocyte cell line. Brit. J . Cancer, 56, 545-554 (1987). COWIN,P. and GARROD, D.R., Antibodies to epithelial desmosomes show wide tissue and species cross-reactivity. Nature (Lond.), 302, 148-150 (1983).
DAVIS,R.K., Prognostic variables in head and neck cancer research: an overview. Aust. N . Z . J . Surg., 59, 367-372 (1985). DE POTTER, C., BEGHIN,C., MAKAR,A.P., VANDEKERCHHOVE, D. and ROELS,H.J., The neu-oncogene protein as a predictive factor for haematogenous metastases in breast cancer patients. Int. J . Cancer, 45,55-58 (1990). EVAN,G.I.,LEWIS,G.K., RAMSAY,G . and BISHOP,J.M., Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product. Mol. cell. Biol., 5 , 361&3616 (1985).
852
PARSONS ET AL.
FABRICANT, R.N., DELARCO, J.E. and TODARO, G.J., Nerve growth factor receptors on human melanoma cells in culture. Proc. nut. Acad. Sci. (Wash.), 74, 565-569 (1977). FIELD,J.K., SPANDIDOS, D.A., STELL,P.M., VAUGHAN, E.D., EVAN, G.I. and MOORE,J.P., Elevated expression of the c-myc oncoprotein correlates with poor prognosis in head and neck squamous cell carcinoma. Oncogene, 4, 1463-1468 (1989). GARROD,D.R. and COWIN, P., Desmosome structure and function. In: C.M. Chadwick, (ed.) Receptors in tumour immunology, pp. 95-130, Cambridge University Press, Cambridge ( I 985). HENDERSON, D.W., PAPADIMITRIOU, J.M. and COLEMAN, M., Ultrastructural appearances of cumours, 2nd ed., Churchill-Livingstone, London (1986). KABLE,E.P.W. and PARSONS, P.G., Potency, selectivity and cell cycle dependence of catechols in human tumour cells in vitro. Biochem. Pharmacol., 37, 1711-1715 (1988). KEARSLEY, J.H., Recent initiatives in head and neck cancer: an overview. Aust. N . 2. J . Surg., 59, 367-312 (1989). K~HLER G., and MILSTEIN,C., Continuous cultures of fused cells secreting antibody of predefined specificity. Nature (Lond.), 256, 4 9 5 4 9 7 (1975). LAEMMLI, U.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.), 227, 68G682 (1970). MAYNARD, K. and PARSONS,P.G., Cross sensitivity of methylating
agents, hydroxyurea and methotrexate in human tumour cells of the Mer phenotype. Cancer Res., 46, 5009-5093 (1986). MCEWAN,M . , PARSONS,P.G. and MOSS, D.J., Monoclonal antibody against human tyrosinase and reactive with melanotic and amelanotic melanoma cells. J . invest. Dermatol., YO, 515-519 (1988). MILLER,K., MATTEY,D.L., MEASURES, H., HOPKINS,C. and GARROD, D., Localisation of the protein and glycoprotein component of bovine nasal epithelial desmosomes by electron microscopy. EMBO J . , 6, 885-889 ( 1987). MOLL,R., CORVIN,P., KAPPNELL,H.P. and FRANKE, W.W., Desmosoma1 proteins: new markers for identification and classification of tumours. Lab. Invest., 54, 4 2 4 (1986). MOORE,G.E., MERRICKS . B . , WOODS,L.K. and ARABASZ,N.B., A human squamous cell carcinoma cell line. Cancer Res., 35, 2684-2688 (1975). STEWART.J.. EVAN.(3.1.. WATSON.J.V. and SIKORA.K.. Detection of the c-my' oncogene'product in colonic polyps and carcinomas. Brit. J . Cancer, 53, 1-7 (1986). I
,
VILELA,M.J., PARRISH, E.P., WRIGHT,D.H. and GARROD, D.R., Monoclonal antibody to desmosomal glycoprotein 1-a non-epithelial marker for diagnostic pathology. J . Path., 153, 365-375 (1987). WHITEHEAD, R.H. and LITTLE,J.H., Tissue culture studies on human malignant melanoma. Pigrn. Cell, 1, 383-389 (1973).
Publication of the International Union Againa Cancer Publication de I'Union Internationale Contre le Cancer
CHARACTERIZATION OF A NOVEL MONOCLONAL ANTIBODY, 3H- 1 , REACTIVE WITH SQUAMOPROLIFERATIVE LESIONS AND SQUAMOUS-CELL CANCERS Peter G. PARSONS,J. Helen LEONARD, John H. KEARSLEY', Hiroyuki TAKAHASHI, Xu LIN-JIANand Denis J. MOSS Queensland Institute of Medical Research, Bramston Terrace, Herston QLO 4006, Australia. Immunization of mice with membranes from a virustransformed human keratinocyte cell line (KJD-I/SV40) yielded an IgM monoclonal antibody (MAb 3H-I) which reacted with the membrane and cytoplasm of KJD-l/SV40cells and in the perinuclear region of a squamous-cell carcinoma line (Colo- 16). lmmunohistochemistry of formalin-fixed, paraffin-embedded sections using MAb 3H-l gave intense staining of proliferating squamous epithelium in several characteristic patterns. Acanthotic squamous epithelium and well-differentiated squamous-cell carcinomas (SCC) demonstrated a membranousstaining pattern whereas psoriatic skin and undifferentiated SCC of the head and neck exhibited diffuse cytoplasmic, focal cytoplasmic or, in some tumours, no staining. Simple squamous epithelium was unreactive. Western blotting revealed an antigen of 55 kDa. The epitope recognized by MAb 3H-I may be a marker for particular stages of squamous proliferation and differentiation.
kindly provided by Prof. P. Gallimore, University of Birmingham, UK (Brown and Gallimore, 1987). Semi-confluent monolayers were synchronized in Gl/S phase by treatment with 5 m~ hydroxyurea for 18 hr (Kable and Parsons, 1988), then washed and re-fed with control medium. After 1 hr, cells were 3 m ~ washed with phosphate-buffered saline ( 5 m Na2HP0,, ~ KH2P04, 1 4 5 NaCI, ~ ~ ~ 7 . PBS) 2 ; and irradiated in PBS using 1440 J.mP2 of 254 nm UV (10 X 150 cm2 plates) or 144 J.mP2 of 254 nm UV (10 X 150 cm2 plates). The cells were harvested by scraping 4 hr (1440 J.md2 dose) or 20 hr (144 J.m-2 dose) after irradiation, and washed with PBS. After resuspension in 1 ml PBS containing 2 m~ phenylmethylsulfonyl fluoride (PMSF), the cells were lysed with 10 strokes of a Dounce glass homogenizer at O"C, then the 1,OOO g pellet was washed twice in PBS/PMSF and resuspended in 2 ml PBS/PMSF containing 2% Triton X-100. After 2 min at room temperature, the mixture was pelleted (1,000 g, 5 min) and the supematant (solubilized membrane) was stored at - 20°C. The supernatants from the 2 UV treatments were mixed prior to immunization.
Squamous-cell cancers (SCCs) represent a heterogeneous group of common human malignancies which arise from the epithelial lining of skin, oral cavity, pharynx, larynx, oesophagus, lung and female genital tract. Despite a histomorphologic appearance common to all SCCs in these anatomic regions, the Derivation of 3H-1 MAb clinical behaviour of individual tumours can vary widely The above supernatant mixture was emulsified with an equal (Davis, 1985). Furthermore, the diagnosis, classification and biologic potential of various squamous-cell hyperplasias, dys- volume of Freund's complete adjuvant and injected into 3 mice plasias and carcinomas is often difficult to assess because of i.p. (300 p1 per mouse), followed by similar amounts with the subjective nature of currently-used light microscopic crite- adjuvant at 14 days (i.p.) and 28 days (i.v.). After a further 7 ria (Boyd and Reade, 1988). Identification and characterization days, the spleen was harvested and fusion with mouse myof molecular markers associated with SCCs and their precursor eloma NS-1 cells was carried out essentially as described by lesions promise to become an important area of research ac- Kohler and Milstein (1975). Following selection in HAT medium, primary screening of 96 microtitre wells was carried out tivity (Kearsley, 1989). on a mixture of methanol-fixed UV-irradiated (144 J.mP2)and The advent of MAbs to various cellular antigens, oncogenes unirradiated KJD-1ISV40 cells grown in microtitre wells and, and oncoproteins has provided a novel, relatively quick, re- separately, on a mixture of human fibroblasts and MM96 huproducible and simple means of studying various characteris- man melanoma cells. Screening involved incubation of hybridtics of the malignant phenotype (Kohler and Milstein, 1975; oma supernatants with the fixed cells, followed by alkaline Evan et al., 1985). The use of MAbs to functional tissue- phosphatase-linked anti-mouse immunoglobulin and BCIPI specific antigens has led to a better understanding of the patho- NBT substrate solution (McEwan et al., 1988). Hybridomas genetic changes which occur during transformation from a be- which gave a positive reaction on KJD-USV40 (cells stained nign to a malignant state (Stewart et al., 1986), and may shed purple) and negative on the 2 other cell types were subcultured some light on those factors which determine tissue invasion and recloned for further screening. and metastases (Field et a l . , 1989; De Potter et al., 1990). For determination of the periodate sensitivity of the antigen, Herein we describe the characterization and staining pattern cells fixed in microtitre wells were treated with 10 mgiml KIO, of a novel IgM MAb, 3H- 1, which reacts on formalin-fixed for 5 min, washed with PBS and then incubated with 3H-1 tissues with squamo-proliferative lesions and squamous-cell MAb supernatant for the ELISA described above. cancers, but is unreactive with simple squamous epithelia and with most malignancies of non-squamous type. Isotype determination of 3H-I Determination of immunoglobulin isotype was carried out MATERIAL AND METHODS using the Mouse-Typer subisotyping kit (BioRad, Sydney, Australia). Briefly, aliquots (100 p,l) of KJD cell lysate Cell culture (2 X lo7 cells/ml in PBS) were placed in EIA plates (Titertek, The nature and derivation of the human cell lines are listed Flow, Brisbane, Australia) and incubated for 2 hr at 2°C to in Table I. Cells were cultured at 37°C in 5% CO,/air in RPMI allow attachment of antigen. After extensive washings with 1640 medium containing 10% (v/v) foetal calf serum, 100 PBS (PH 7.2), wells were treated with hybridoma supernatant pg/ml streptomycin and 100 I U / d penicillin. Routine assays for Mycoplasma by agar culture or Hoechst dye staining were negative. Immunogen KJD-l/SV40 (Table I) is a continuous line of human keratinocytes transformed by an SV40-adenovirus hybrid and
'To whom correspondence and reprint requests should be addressed. Received: November 5, 1990.
848
PARSONS ET A L .
(MAb 3H-1) for 1 hr at room temperature. Rabbit anti-mouse immunoglobulins (IgG,, IgG,,, IgG,,, IgG,, IgA and IgM) were applied to the wells for 1 hr at room temperature, followed by washings with PBS (PH 7.2). After treatment with goat anti-rabbit IgG horse-radish peroxidase . conjugate, the substrate mixture (hydrogen peroxide and 2,2'-azinodi [3ethylbenzthiazoline substrate]) was added; positive wells turned green. Western blotting Biopsies were dispersed by sonication in 400 ~1 of 2% (wU vol) SDS (sodium dodecylsulphate), 10 mM dithiothreitol and 2 m~ PMSF in 10 m~ TridHCl, PH 7.4. The samples were then placed in a boiling water bath for 2 min, allowed to cool, and centrifuged at 15,000g for 5 min. Samples were stored in aliquots at - 20°C; 2 0 4 samples (150 Fg of protein) were used for electrophoresis. Samples were frozen and thawed once only. Composition and electrophoresis of SDS 5-10% (wt/vol) polyacrylamide slab gels were essentially as reported by Laemmli (1970). Electrophoresis was performed on a minigel system at 60V for 2.5 hr at room temperature. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose paper (Amersham, Little Chalfont, UK) was performed using bicarbonate buffer (PH 9.9) on a transblot apparatus according to the manufacturers' instructions (BioRad). The antigen was detected by incubation with MAb supernatant 3H-1 followed by alkaline phosphatase-linked goat anti-mouse antibody (Silenus, Melbourne, Australia) diluted 1:1,000, followed by development with BCIP/NBT (McEwan et al., 1988). Immunohistochemistry Immunohistochemical staining was performed using the avidin-biotin-peroxidase (ABC) technique (Vector, Burlingame, CA). Briefly, tissue sections were deparaffinized, dehydrated in 100% alcohol and then rehydrated through decreasing alcohol concentrations to water. Sections were treated with 0.3% (wiv) hydrogen peroxide and 20% (wh) methanol in PBS for 10 min to eliminate endogenous peroxidase activity. Sections were then incubated with dilute horse serum for 20 rnin to eliminate non-specific binding of immunoglobulins. After removal of excess horse serum, 3H-1 MAb was added to each section as a supernatant and the slides were incubated for 30 min at room temperature. After several washes in PBS, sections were incubated with diluted biotinylated horse anti-mouse IgM immunoglobulin for 30 rnin at room temperature. The slides were then washed in PBS and incubated for 45 min with Vectastain Reagent (Vector), according to the manufacturer's directions. After a further wash in PBS, sections were incubated for 3-10 min with 3-amino-8-ethylcarbazoleto which 10 drops of 30% hydrogen peroxide had been added. The sections were finally counterstained with haemalum, rinsed and mounted under a coverslip with Glycergel (DAKO, Carpinteria, CA). All immunoperoxidase procedures included both positive and negative control slides. Positive control was performed by using a formalin-fixed section from an SCC tongue which consistently showed strong immunoperoxidase staining with 3H- 1 MAb. Negative controls were incubated with PBS or an irrelevant murine IgM antibody (reactive with formalin-fixed material) instead of 3H-1 antibody in the first step and were then treated as described above. Immunoelectron microscopy Within 15 min of removal from the operating theatre, tissue specimens (less than 1 mm3) were fixed with freshly-prepared 6 , 15 4% formaldehyde in 0.1 M cacodylate buffer, ~ ~ 7 . for min at room temperature. Osmium tetroxide post-fixation was omitted. After several buffer washes, specimens were dehydrated to 70% ethanol and embedded in White acrylic resin
(London Resin, Woking, UK). Resin was polymerized at 50°C for 24 hr. Ultra-thin sections were cut and mounted on nickel grids for immunolabelling . Sections were rinsed on drops of distilled water and nonspecific binding was blocked by a 30-min incubation with 5% BSA in modified Tris buffer (20 m~ Tris, 0.5 M sodium chloride, 20 mM sodium azide and 0.05% Tween 20). After several buffer washes, sections were incubated with a 1:10 dilution of MAb (3H-1 IgM purified ascites) for 2 hr at 37"C, rinsed thoroughly with buffer and then incubated with goat antimouse IgM-colloidal gold (Janssen, Beerse, Belgium, 10 nm particle size) for 1 hr at 37°C. After washing with buffer, sections were contrasted with uranyl acetate and lead citrate and examined in a Jeol 1200 EX transmission electron microscope. RESULTS
Derivation of MAb 3H-1 and reactivity with cultured human cells In an attempt to increase the probability of obtaining MAbs directed against membrane and proliferation-related antigens in SCC of the skin, the transformed keratinocyte line KJDl/SV40 was synchronized in G,/S with hydroxyurea (Kable and Parsons, 1988) and irradiated with UV in early S-phase, then the membrane fraction was employed for immunization. The successful derivation of MAb 3H- 1, the most promising of several similar clones, can perhaps be attributed to the membrane fractionation because the antibody reacted with the cell membrane in the majority of KJD-IISV40 cells (Fig. la). However, the UV irradiation and synchronization of cells used to prepare the immunogen did not appear to have been essential because the 3H-1 antibody reacted with asynchronous KJDI/SV40 cells whether irradiated or not. The isotype of MAb 3H-1 was found to be IgM. MAb 3H-1 reactivity, studied in a wider range of cultured human material (Table I), was restricted to the membrane and cytoplasm of KJD-l/SV40 cells and to the SCC cell line, Colo16. In this line the staining was mainly focal perinuclear and was found in a lower proportion of cells than in the KJD1/SV40 line (Fig. lb). Pretreatment of fixed cells with periodate to destroy carbohydrate epitopes did not abolish reactivity with 3H-I but eliminated the reactivity of melanoma cells with a tyrosinase antibody (McEwan et al., 1988), used in the present work as a positive control. Reactivity with histologic sections of normal tissue One hundred and ninety-six samples from a wide range of normal human tissues were tested for reactivity with MAb 3H-1. As shown in Table 11, staining with MAb 3H-1 was not present in any example of simple squamous epithelium, but was consistently present in epithelial cells of hair follicles, salivary glands, bronchial, stomach and small intestinal mucosae. MAb 3H-1 reactivity was infrequently demonstrated in bladder and prostate mucosae but was absent from all other normal tissues. Reactivity with squamous-cell carcinoma tissues and benign squamo-proliferative lesions As shown in Table 111, MAb 3H-1 reacted with a variety of benign and pre-malignant squamo-proliferative lesions. The staining patterns observed ranged from membranous in acanthotic epidermis and keratocanthoma to focal cytoplasmic in hyperproliferative psoriatic epithelium. The cornified layer above acanthotic squamous epithelium did not react with 3H-1 MAb, and there was often a very distinct disappearance of staining at the junctional region between hyperkeratoses and
MAb A N D PROLIFERATING SQUAMOUS CELLS
849
FIGURE1 - Immunohistochemical localization of the MAb 3H-1 antigen in KJD-I/SVM cells ( a ) and cola-16 cells (b). The secondary antibody was alkaline phosphatase-linked anti-mouse antibody and the substrate was BCIP/NBT. Bar = 50 Fm. FIGURE2 - Immunoreactivity of MAb 3H- 1 with formalin-fixed, paraffin-embedded tissue sections of (a) psoriasis (pennuclear staining pattern), bar = 50 pm; and (b)edge of a hyperkeratosis, showing 3H-1 reactivity in hair follicle and within the hyperplastic epithelium but not within adjacent simple squamous epithelium. Bar = 100 pm.
normal simple (non-staining) epidermis (Fig. 2b). In premalignant skin lesions, all 3 patterns of reactivity were seenfocal cytoplasmic in the immediate suprabasal region transforming to diffusely cytoplasmic/membranous in the spinous and granular layers, interspersed with occasional more superficial cells which demonstrated purely membranous reactivity. In all examples, basal epithelial cells were non-reactive. It is also clear from Table I11 that a large proportion of formalin-fixed sections of SCCs from human tissues react with 3H-1 MAb. These SCCs were derived from lesions of the head and neck, female genital tract, skin, lung and oesophagus. The SCC tissue specimens represent both primary and metastatic tumour sites and the histologic grading ranged from welldifferentiated, low-grade malignancies with the presence of large numbers of keratin whorls to high-grade, poorly differentiated SCCs with very little keratin production and with diffuse invading margins. The 3 staining patterns previously observed-membranous, diffuse cytoplasmic and focal cytoplasmic-were present in varying proportions in the SCC tissues. No staining occurred either in the keratin deposits or in the cells of the surrounding connecting tissue, skeletal muscles, blood vessels or lymphocytes. Less than 10% of poorly differentiated SCCs stained with 3H-1 MAb, whereas a majority of well-differentiated tumours stained in a membranous or membrano-cytoplasmic pattern. The lack of numbers in the poorly differentiated group, however, means that this clinical observation could not be supported statistically. It appeared that a
subgroup of moderately and poorly-differentiated SCCs demonstrated either no staining or a focal cytoplasmic staining pattern. Reactivity of histologic sections of non-squumous-cell carcinoma tissues
In a study of tumours other than SCC (Table IV), MAb 3H-1 only reacts consistently with adenocarcinomas of the stomach and colon. Most importantly, in relation to the potential use of MAb 3H-1 as a diagnostic aid, a consistently negative reaction was noted in virtually all basal-cell carcinomas, malignant melanomas, Merkel cell tumours, non-squamous lung cancers and most other adenocarcinomas. Determination of the molecular weight of the 3H-1 antigen by Western blotting
In repeated experiments, polyacrylamide gel electrophoresis of KJD-USV40 lysates followed by immunoblotting with MAb 3H-1 revealed strong immunoreactivity with a band at 55 kDa (Fig. 3), seen also in the Colo-16 line, tissue from a tongue SCC, and faintly in normal salivary gland. The SCC also showed a band at approximately 210 kDa when the lysate was stored at 4°C for 3 weeks (Fig. 3). No bands were seen in foreskin fibroblasts (Fig. 3) or the MM96 human melanoma cell line (not shown) which served as negative controls. Immunoelectron microscopy
Using 10-nm colloidal gold particles as the electron-dense
850
PARSONS ET AL.
TABLE I - REACTIVITY OF MAb 3H-I WITH CULTURED HUMAN CELLS Cell line
Transformed keratinocytes KJD-1lSV40
Colo 16 Melanoma MM96E
MM 127 MM 170 MM229 MM418 MM540 MM576 A2058 Others HeLa (cervical carcinoma) Foreskin fibroblasts Sch-I (Ewing’s sarcoma) RD-ES (Ewing’s sarcoma) G401 (Wilms’ tumour) NB2 (neuroblastoma)
Derivation of cell line
Brown and Gallimore, 1987 Moore et al., 1975
MAb 3H-1
reactivity localization
+ + (80%)’
membrane cytoplasmic + (30%) perinuclear focal
Whitehead and Little, 1973 Whitehead and Little, 1973 Whitehead and Little, 1973 Maynard and Parsons, 1986 Maynard and Parsons, 1986 Our laboratory’ Our laboratory’ Fabricant et at., 1977 ATCC
P.J. Smith Marshall and Kirchen, 1989 ATCC P.J. Smith
’Percentage of cells with stain.-’Established from lymph-node metastases of human melanoma.
marker, we observed, by irnrnunoelectron microscopy, that 3H-1 antigen was distributed generally within the cytoplasm of the fresh SCC cells examined (Fig. 4). The antigen was not localized to any particular organelle, nor to specific regions of the cytoplasm in the 2 SCC biopsies examined. Later immunohistochemistry showed one of these to have nil staining and the other (specimen seen in Fig. 4), to have mainly cytoplasmic staining. Colloidal gold particles were also present along the inner aspect of the cell membrane. Colo- 16 cells demonstrated very low cytoplasmic labelling, and immunoelectron microscopy of KJD-lISV40 cells was unsuccessful because of a high level of non-specific binding. Normal epidermis showed no labelling. DISCUSSION
This study confirmed the expectation that the virus-transformed human keratinocyte line KJD- 1/SV40 chosen as an immunogen expresses proliferation-related antigens specific to cells of keratinocyte lineage. Since over 80% of asynchronous KJD-USV40 cells reacted with MAb 3H-1, the relevant antigen appeared to be expressed in most phases of the cell cycle of a proliferating cell population. The basis of the different staining patterns observed between SCC lines has yet to be determined. The nature of the antigen or antigens recognized by the 3H- 1 MAb remains to be defined. The predominantly membranous/ interstitial staining reaction resembles the pattern seen with anti-desmosomal antibodies (Cowin and Garrod, 1983; Moll et
TABLE I1 - REACTIVITY OF MAb 3H-1 WITH NORMAL HUMAN TISSUES Normal tissues or cells
Positiveitested
Hair follicle Salivary gland Stomach (mucosa) Small intestine (mucosa) Prostate Bladder Bronchial mucosa Lung Breast Spleen Colon (mucosa) Liver Kidney ovary Lymph nodes Lymphocytes Polymorphonuclear leukocytes Red blood cells Muscle Pancreas Thyroid Pituitary Adrenal Heart Cartilage
20120 15/15 414 313 218 217 111 018 016 015 012 0110 017 016 017 0112 017 0115 0118 011 011 011 011 012 014
TABLE 111 - REACTIVITY OF MAh 3H-I WITH HUMAN SQL’AMOL’S-CELL CARCINOMA TISSCI: AN11 WITH BENIGN 1,ESIONS
Tissues
Squamous-cell carcinoma Oral cavity Larynx Pharynx Skin Lung Ectocervix Oesophagus Unknown primary Bowen’s disease (in-situ SCC) Benign lesions Acanthotic epidermis Keratoacanthoma Psoriasis Intradermal naevi Chromophobe adenoma (pituitary)
Poritiveirerted
24/32 20126 16/20 16/16 518
214 112 111 616 12112 10110 818 018 011
al., 1986), none of which, however, were specific for proliferating squamous epithelia (Moll et al., 1986; Vilela ef al., 1987). The localized juxtanuclear staining seen in our study may represent desmosomal internalization, which occurs in many different types of carcinoma (Garrod and Cowin, 1985; Henderson et al., 1986). The 210-kDa band found in the SCC is similar in molecular weight to the desmosomal protein dpl (normally 215 kDa; Miller et al., 1987) but the 3H-1 epitope was most commonly found strongly expressed in a 55-kDa antigen, suggestive of a cytokeratin-like molecule possibly associated with a range of subcellular structures. The immunoelectron microscopy results do not shed further light on this issue because the distribution of gold-labelled particles was generalized within the cytoplasm of the 2 samples which gave a positive result and the particles were not sufficiently numerous to exclude the possibility of satellite associations as found for dpl (Miller et al., 1987). The strong cytoplasmic association of the antigen in secretory tissue may indicate an involvement in the function of salivary gland and colonic epithelium. The fact that antigenicity was not abolished by pretreatment with periodate suggests that the MAb 3H-1 will be useful in identifying the peptide component of the antigen.
851
MAb AND PROLIFERATING SQUAMOUS CELLS TABLE IV - REACTIVITY OF MAb 3H-1 WITH HUMAN NON-SQUAMOUS MALIGNANCIES
Carcinoma tissue
Positiveitested
Melanoma Basal-cell carcinoma Non-Hodgkin’s lymphoma Ovarian cancer Colon cancer Adenocarcinoma lung Small-cell carcinoma lung Transitional-cell carcinoma bladder Breast cancer Osteosarcoma Thyroid cancer Renal cancer Testis cancer Prostate cancer Hodgkin’s disease Stomach cancer Syringoma Eccrine poroma Cholaneiocarcinoma
0110 0110 013
3/10
414 012
013
3/10
018 014
012 014
016 218 016
313 01 1 012 316
FIGURE4 - Immunoelectron micrograph showing localization of MAb 3H-1 in SCC. Labelling, as indicated by 10 nm colloidal gold particles, is seen within the cytoplasm and along the cell membrane. CM, cell membrane; D, desmosome. Bar = 0.5 nm.
FIGURE3 - Molecular weights of MAb 3H-1 antigens determined by Western blotting of cells and tissues. FS, normal foreskin fibroblasts (negative control); SG, salivary gland ( 5 5 ma);SCC, tongue SCC (210 and 55 kDa, lysate stored at 4°C for 3 weeks); Co, SCC cell line Colo-I6 (55 ma);KJD, transformed keratanocyte cell line KJD1lSV40 (55 m a ) and MWT position of the molecular weight markers
clear pattern similar to that seen in many of the aggressive SCCs. This finding suggests to us that immature, rapidly dividing squamous cells in psoriasis and aggressive SCCs share a common epitope which is localized in the perinuclear region of the disordered squamous cell. There are clearly many other factors which distinguish disordered squamous-cell proliferation in conditions such as psoriasis and hyperkeratoses from that occurring in the case of invasive SCCs. Identification of different cellular phenotypes may be helpful in understanding the mechanism underlying transformation from a benign or proliferative state to a malignant one. Our results suggest that MAb 3H-1 may be a valuable diagnostic and function-related marker for squamo-proliferative lesions, including SCCs. The epitope recognized by the 3H-1 antibody is not expressed at detectable levels in simple keratinizing squamous epithelium and only rarely in non-squamous malignancies, but has been demonstrated in all cases of solar keratoses, Bowen’s disease, keratoacanthomas, psoriasis and a high proportion of SCCs in various localization patterns and molecular weights. ACKNOWLEDGEMENTS
in kDa.
Whatever the function and chemical composition of the antigen recognized by MAb 3H-1, the latter is clearly an IgM antibody resistant to routine histologic processing and therefore has potential for immunodiagnosis of squamous lesions. In addition to being expressed in SCCs, the epitope is also expressed in rapidly dividing, non-malignant squamous cells. In the case of psoriasis, a hyperproliferative disorder of immature squamous epithelial cells, MAb 3H-1 reacts in a focal perinu-
W e thank Mrs. J. Middleton for her assistance in typing the manuscript. This work was supported by the Queensland Cancer Fund and the H.L. Rose Foundation. J.H.K. was in receipt of a Research Fellowship from the Queensland Radium Institute. Prof. P.J. Smith, Oncology Program, University of Queensland, kindly provided cell lines Sch-1 and NB2, and the HeLa and RD-ES cell lines were obtained from the ATCC, Rockville, MD. Immunoelectronmicroscopy was performed by Ms. D. Stenzel, Analytical Electron Microscopy Facility, Queensland University of Technology, Brisbane.
REFERENCES
BOYD,N.M. and READE,P.C., Differences between preneoplastic cells, neoplastic cells and their normal counterparts. J . Oral Path., 17,257-265 (1988). BROWN,K.W. and GALLIMORE, P.H., Malignant progression of an SV,transformed epithelial keratinocyte cell line. Brit. J . Cancer, 56, 545-554 (1987). COWIN,P. and GARROD, D.R., Antibodies to epithelial desmosomes show wide tissue and species cross-reactivity. Nature (Lond.), 302, 148-150 (1983).
DAVIS,R.K., Prognostic variables in head and neck cancer research: an overview. Aust. N . Z . J . Surg., 59, 367-372 (1985). DE POTTER, C., BEGHIN,C., MAKAR,A.P., VANDEKERCHHOVE, D. and ROELS,H.J., The neu-oncogene protein as a predictive factor for haematogenous metastases in breast cancer patients. Int. J . Cancer, 45,55-58 (1990). EVAN,G.I.,LEWIS,G.K., RAMSAY,G . and BISHOP,J.M., Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product. Mol. cell. Biol., 5 , 361&3616 (1985).
852
PARSONS ET AL.
FABRICANT, R.N., DELARCO, J.E. and TODARO, G.J., Nerve growth factor receptors on human melanoma cells in culture. Proc. nut. Acad. Sci. (Wash.), 74, 565-569 (1977). FIELD,J.K., SPANDIDOS, D.A., STELL,P.M., VAUGHAN, E.D., EVAN, G.I. and MOORE,J.P., Elevated expression of the c-myc oncoprotein correlates with poor prognosis in head and neck squamous cell carcinoma. Oncogene, 4, 1463-1468 (1989). GARROD,D.R. and COWIN, P., Desmosome structure and function. In: C.M. Chadwick, (ed.) Receptors in tumour immunology, pp. 95-130, Cambridge University Press, Cambridge ( I 985). HENDERSON, D.W., PAPADIMITRIOU, J.M. and COLEMAN, M., Ultrastructural appearances of cumours, 2nd ed., Churchill-Livingstone, London (1986). KABLE,E.P.W. and PARSONS, P.G., Potency, selectivity and cell cycle dependence of catechols in human tumour cells in vitro. Biochem. Pharmacol., 37, 1711-1715 (1988). KEARSLEY, J.H., Recent initiatives in head and neck cancer: an overview. Aust. N . 2. J . Surg., 59, 367-312 (1989). K~HLER G., and MILSTEIN,C., Continuous cultures of fused cells secreting antibody of predefined specificity. Nature (Lond.), 256, 4 9 5 4 9 7 (1975). LAEMMLI, U.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.), 227, 68G682 (1970). MAYNARD, K. and PARSONS,P.G., Cross sensitivity of methylating
agents, hydroxyurea and methotrexate in human tumour cells of the Mer phenotype. Cancer Res., 46, 5009-5093 (1986). MCEWAN,M . , PARSONS,P.G. and MOSS, D.J., Monoclonal antibody against human tyrosinase and reactive with melanotic and amelanotic melanoma cells. J . invest. Dermatol., YO, 515-519 (1988). MILLER,K., MATTEY,D.L., MEASURES, H., HOPKINS,C. and GARROD, D., Localisation of the protein and glycoprotein component of bovine nasal epithelial desmosomes by electron microscopy. EMBO J . , 6, 885-889 ( 1987). MOLL,R., CORVIN,P., KAPPNELL,H.P. and FRANKE, W.W., Desmosoma1 proteins: new markers for identification and classification of tumours. Lab. Invest., 54, 4 2 4 (1986). MOORE,G.E., MERRICKS . B . , WOODS,L.K. and ARABASZ,N.B., A human squamous cell carcinoma cell line. Cancer Res., 35, 2684-2688 (1975). STEWART.J.. EVAN.(3.1.. WATSON.J.V. and SIKORA.K.. Detection of the c-my' oncogene'product in colonic polyps and carcinomas. Brit. J . Cancer, 53, 1-7 (1986). I
,
VILELA,M.J., PARRISH, E.P., WRIGHT,D.H. and GARROD, D.R., Monoclonal antibody to desmosomal glycoprotein 1-a non-epithelial marker for diagnostic pathology. J . Path., 153, 365-375 (1987). WHITEHEAD, R.H. and LITTLE,J.H., Tissue culture studies on human malignant melanoma. Pigrn. Cell, 1, 383-389 (1973).
