Epidermal growth factor receptor expression in human foetal tissues is age-dependent.

Epidermal growth factor receptor (EGFr) is a 170 kD phosphoglycoprotein which spans the cell membrane and mediates the initial response of a wide range of cells to the peptide growth hormone EGF (Carpenter & Cohen, 1979; Cohen, 1983). Transforming growth factor a (TGFa) which is structurally related to EGF, also binds to EGFr to initiate a mitogenic response (Marquardt et al., 1983). Overexpression of EGFr has been associated with malignant transformation of epidermal cells (Ozenne et al., 1986) and with the metastatic potential of breast (Sainsbury et al., 1985) and lung cancer (Veale et al., 1987). The receptor also plays an important role in the regulation of growth and differentiation of epidermal cells both in tissue culture and in vivo. The aim of this study was to characterise EGFr expression in a number of foetal tissues, to determine the suitability of each tissue for further study on the role of EGFr in oncogenesis. Fresh tissue samples were obtained from normal foetuses immediately after prostaglandin-induced terminations. Twenty-six cases of 14-19 weeks estimated gestational age (EGA) were studied. Age was assessed by menstrual history and clinical examination together with foetal weight, crownrump measurement and foot length. Biopsies were snap frozen and 5-7 ,um cryostat sections mounted on gelatincoated slides: these were air-dried and fixed in acetone for 20min at room temperature. The EGFr was identified in the unlabelled peroxidase antiperoxidase (PAP) immunohistochemical method, first described by Sternberger et al. (1970). The primary monoclonals used were MAS 6580 (Sera Lab) which binds to the external portion of the EGFr molecule (Yarden et al., 1985) and EGFR1 (Amersham) that recognizes an antigenic determinant located on the polypeptide chain of the receptor (Mayes & Waterfield, 1984). Sections were incubated at room temperature with primary antibody at a dilution of 1/100 (MAS 6580c) for 60min or at 4°C at a dilution of 1/50 (EGFRI) overnight. After washing these were incubated with rabbit anti-mouse Ig (Dakopatts 1:20) in 20% normal human AB serum for 30 min, followed by a 30 min incubation with 1/100 dilution of PAP complex (Dakopatts). The bound peroxidase was visualized using the diaminobenzidine/H202 reaction. Sections were counterstained in Harris' haematoxylin, dehydrated and mounted in DPX. Controls included the replacement of primary antibody with Tris buffered saline or normal mouse immunoglobulin. In addition normal adult skin and term placenta sections were run as known EGFr positive controls and adult lymphocyte preparations as known EGFr negative controls. The intensity of staining was assessed and graded on a scale from 0 to + +. Control slides were uniformly negative, except known adult skin and term placenta (Figure 1). Cells that were found to be predominantly positive for EGFr are indicated in Table I for each tissue. The staining pattern observed was similar for both EGFr antibodies.

Epidermal growth factor receptor (EGFr) is a 170 kD phosphoglycoprotein which spans the cell membrane and mediates the initial response of a wide range of cells to the peptide growth hormone EGF (Carpenter & Cohen, 1979;Cohen, 1983). Transforming growth factor a (TGFa) which is structurally related to EGF, also binds to EGFr to initiate a mitogenic response (Marquardt et al., 1983). Overexpression of EGFr has been associated with malignant transformation of epidermal cells (Ozenne et al., 1986) and with the metastatic potential of breast (Sainsbury et al., 1985) and lung cancer (Veale et al., 1987). The receptor also plays an important role in the regulation of growth and differentiation of epidermal cells both in tissue culture and in vivo.
The aim of this study was to characterise EGFr expression in a number of foetal tissues, to determine the suitability of each tissue for further study on the role of EGFr in oncogenesis.
Fresh tissue samples were obtained from normal foetuses immediately after prostaglandin-induced terminations. Twenty-six cases of 14-19 weeks estimated gestational age (EGA) were studied. Age was assessed by menstrual history and clinical examination together with foetal weight, crownrump measurement and foot length. Biopsies were snap frozen and 5-7 ,um cryostat sections mounted on gelatincoated slides: these were air-dried and fixed in acetone for 20min at room temperature. The EGFr was identified in the unlabelled peroxidase antiperoxidase (PAP) immunohistochemical method, first described by Sternberger et al. (1970). The primary monoclonals used were MAS 6580 (Sera Lab) which binds to the external portion of the EGFr molecule (Yarden et al., 1985) and EGFR1 (Amersham) that recognizes an antigenic determinant located on the polypeptide chain of the receptor (Mayes & Waterfield, 1984). Sections were incubated at room temperature with primary antibody at a dilution of 1/100 (MAS 6580c) for 60min or at 4°C at a dilution of 1/50 (EGFRI) overnight. After washing these were incubated with rabbit anti-mouse Ig (Dakopatts 1:20) in 20% normal human AB serum for 30 min, followed by a 30 min incubation with 1/100 dilution of PAP complex (Dakopatts). The bound peroxidase was visualized using the diaminobenzidine/H202 reaction. Sections were counterstained in Harris' haematoxylin, dehydrated and mounted in DPX. Controls included the replacement of primary antibody with Tris buffered saline or normal mouse immunoglobulin. In addition normal adult skin and term placenta sections were run as known EGFr positive controls and adult lymphocyte preparations as known EGFr negative controls.
The intensity of staining was assessed and graded on a scale from 0 to + +. Control slides were uniformly negative, except known adult skin and term placenta (Figure 1). Cells that were found to be predominantly positive for EGFr are indicated in Table I     The EGFr was detected on the majority of foetal tissue studied at 17 weeks EGA and above (Table I). However, the receptor was not expressed prior to 17 weeks EGA, except in the skin, gut, liver, thyroid and spleen where weak staining (+) was observed at 16 weeks EGA. By 18 weeks EGA intense staining (+ +) for the receptor was observed on all tissues studied particularly on epithelial cells (Table I, Figures 1 and 2). These results indicate significant EGFr gene expression in the foetus at around 17 weeks EGA.
A study of adult tissue has shown a large number of tissues to be EGFr positive, with the exception of the adrenal and thyroid (Damjanov et al., 1986). This is of interest, as it suggests in the light of the present findings that EGFr gene expression probably becomes repressed in adrenal and thyroid tissues at times beyond those of positive detection in the foetal samples. In the adult, it was shown that actively proliferating epithelia expressed EGFr at the cell surface whereas once cells differentiated into a nonproliferating component, EGFr expression was reduced to undetectable levels or extinguished (Damjanov et al., 1986). Whether other receptors determine growth before 16-17 weeks EGA is at present unknown.
Analysis of DNA from squamous carcinoma cell lines and tumours has revealed that amplification of the gene encoding the EGFr is associated with over-expression of the receptor in these cells (Ozanne et al., 1986). Whether the EGFr gene in foetal tissue is 'switched-on' by some regulatory mechan-ism or simply amplified is open to speculation. Overexpression is not a consequence of proliferation per se, as the receptor is not increased in hyperproliferative skin disorders (Ozanne et al., 1985). Malignant and virally transformed epidermal cells possess 5-50 times more EGFr than normal keratinocytes (Cowley et al., 1986). Therefore, the increased expression of EGFr in epidermal malignancies may be an important component of the malignant phenotype in these tumours. Cancerous tissue in the lung has also been shown to have significantly increased levels of EGFr compared to normal lung (Veale et al., 1987) and amplification of the EGFr gene (Berger et al., 1987). The level of EGFr is also associated with the degree of invasion and poor differentiation of bladder cancer (Neal et al., 1985). It has thus been suggested that the presence of this receptor on squamous cell carcinomas may prove to be of diagnostic value and a suitable target for therapy, yet little is understood about the exact role of the receptor in oncogenesis. We propose to utilize the availability of foetal tissue which is either negative or positive for EGFr to investigate the role of the receptor in malignant cell transformation.