Immunohistochemical demonstration of epidermal growth factor (EGF) receptors in normal human placental villi

Placenta(I99O),II, 7-15

Immunohistochemical Demonstration of Epidermal Growth Factor (EGF) Receptors in Normal Human Placental Villi

K O H K A W A G O E a, J U N K O A K I Y A M A b, TOMOYUKI K A W A M O T O c, Y A S U Y K I MORISHITA a & SHIGEO MORI a Department of Obstetrics and Gynaecology, Tokyo University Branch Hospital, Mejirodai 3-28-6, Bunkyoku, Tokyo, Japan. PO. llZ b Department of Obstetrics, Tokyo Metropolitan Tsukiji Sannin Hospital Akashichou 6-24, Tsukiji, Tokyo, Japan. PO. Io4 e Department of Biochemistry, Okayama University Dental School Shikatachou 2-5-i , Okayama City, Okayama Prefecture, Japan. PO. 7oo. d Department of Pathology, Institute of Medical Science, University of Tokyo, Shiroganedai 4-6-1, Minato-Ku, Tokyo, Japan. PO. to8 Paper accepted 8.6.z989

INTRODUCTION The human placenta undergoes dramatic morphological and functional changes during the course of pregnancy. Epidermal growth factor (EGF) stimulates the proliferation of a variety of cells of both epithelial and mesenchymal origins (Branes and Sato, I98o). However, Morrish et al (i987) found that EGF induces differentiation, not proliferation, of placental villi during pregnancy. EGF is considered to mediate cell function, such as the suppression of acidsecretion in gastric parietal cells (Gregory, i975; Sakai et al, I986), and the secretion of human chorionic gonadotropin (hCG) and human placental lactogen (hPL) from trophoblast (Maruo and Mochizuki, 1987; Morrish et al, i987). Serum epidermal growth factor was reported to reach its highest concentration during the early part of pregnancy, but it was detected in the serum throughout the entire course of the gestational period and within the cord blood (Ances, I973). Receptors for EGF, a glycoprotein with a molecular weight of approximately i7o ooo, have tyrosine kinase activity located on the internal side of the cell membrane (Hunter, I984). This paper deals with the tissue and cellular distributions of EGF receptors in normal human placental villi from 6 to 42 weeks gestation. Periodate-lysine-paraformaldehyde fixed sections and frozen sections stained by the avidin-biotin-peroxidase complex (ABC) method (Hsu, Rayhe and Auger, I98x ) were used. Monoclonal antibodies to EGF receptors (Kawamoto et al, 1983) were employed. Special consideration was given to the possible role of EGF in trophoblastic oi43 4oo4/9o/oiooo7 + o9 $o5.oolo

9 199o Bailli~reTindall Ltd

8

Placenta

(i99o),

Vol. I I

tissues, as they related to the characteristic localization of E G F receptors in normal human placental villi.

MATERIAL AND METHODS Mature and immature placental tissues from normal human pregnancies were obtained at various times of gestation. Immature ones were obtained at 6, 7, 8 and 11 weeks of gestation at the time of dilatation and curretage, premature ones at 23, 24, 28 and 31 weeks of gestation (some of which were collected during elective caesarian section), and mature ones at 37, 38, 4o, 4I and 42 weeks of gestation at the time of normal vaginal delivery. Samples from these tissues were excised and dissected into small pieces immediately after delivery or therapeutic abortion. They were washed in saline, and placed in embedding medium (O.C.T. Compound, Miles Laboratory, Inc., Naperville, IL) before being snapfrozen in n-hexane precooled with acetone-dry ice. The delay before freezing, I5min on average, was never more than 3o min. To detect the presence of E G F receptors, a murine monoclonal antibody, 528 IgG (Kawamoto et al, 1983) , was used as the first reagent in the avidin-biotin-peroxidase complex (ABC) staining method (Hsu et al, I98I). This antibody was developed by using E G F receptors from human A-431 epidermoid carcinoma cells as an immunogen followed by purification from ascites by 1:3 saturated ammonium sulphate and ion exchange chromatography. It reacted specifically with the binding domain of E G F receptors (Kawamoto et al, I983). It can fully recognize E G F receptors bound to E G F by its higher affinity to the binding domain. However, this antibody cannot react with any glandular or squamous cells in formalin-fixed and paraffinembedded tissue sections. Cryostat sections, 5 #m thick, were cut and placed on lysine-coated slides. Prior to staining, they were fixed in 4 per cent paraformaldehyde for 15 min and washed three times for 15 min in o.ol M phosphate-buffered saline (PBS). After blocking non-specific reactions with normal horse serum for 3omin at room temperature, the properly diluted 0:4ooo) 5 2 8 ~ G was applied to sections and incubated for 3o min in a moist chamber at 37~ The sections were rinsed in PBS three times during a i5-min period. Then they were incubated in PBS containing Ioo • diluted biotinylated anti-mouse IgG for 3o min, rinsed and incubated with Ioo • diluted third phase reagent of the Vectastain ABC kit (Vector Laboratories Inc.; Burlingame, CA) for 3o min. Each reagent was applied for 3o min in a moist chamber at 37~ followed by washing with PBS three times for 15 min. For the ABC-PO method, the peroxidase activity was visualized using diaminobenzidine (DAB) reagent (3,3-diaminobenzidine tetrahydrochloride, 3 mg in i o ml of PBS with o. i ml of 1 per cent H2Oz) for 5 min. The sections were washed with PBS and rinsed briefly in distilled water, dehydrated in graded alcohol, cleaned in xylene and mounted. For the ABC-GO method, the final reactions were achieved by incubating the sections for 15 3o min with glucose oxidase substratc solution in the dark, then they were washed for 5 min in tap water, cleaned and mounted. For immunoelectronmicroscopic observation, the specimens were fixed with periodatelysine paraformaldehyde (PLP) fixative (McLean and Nakane, 1974) and snap-frozen in n-hexane precooled with acetone-dry ice. Frozen sections 5 #m thick were prepared using a cryostat, and mounted on slide glasses. The slides were then stained following the method mentioned above. After immunostaining, sections were dehydrated in graded alcohol and embedded in Epon 81 z resin (TAAB Laboratories, England). They were detached from glass

Kawago et al: EGF-Receptors in Normal Human Placenta

Figure 1.

9

Immunoreactionfor EGF receptor of immature placentaat 6 weeksof gestation.

slides after the resin polymerized at 6o~ in an oven for 72 h. Semithin and thin sections, cut on an L K B 88oo microtome, were counterstained with lead hydroxide and uranyl acetate. Observations were made using a J E O L iooC electron microscope (Nihon Denshi Co. Ltd., Japan). In each case, two additional sections were prepared; one as a control without the first reagent of 528 IgG, the other for assessing haematoxylin and eosin staining. Resected fresh-frozen tonsils were used as positive controls for the evaluation and titration of antibodies.

RESULTS Intense immunoreaction for E G F receptor was observed as a diffuse linear, or sometimes patchy, pattern on the syncytiotrophoblastic cell surface at various gestational periods (Figures I-3). T h e staining intensity of the syncytiotrophoblastic surface was much stronger at a thickness of approximately 2/~m than that of normal tonsilar squamous epithelia (Figure 4). Though the outer surface of the cytotrophoblasts showed weaker immunoreactivity than the free surface of syncytiotrophoblasts, E G F receptors were detected at the cytotrophoblastic cell surface beneath syncytiotrophoblasts in immature and premature placental villi (Figure 5). Little immunoreaction was observed at the cytotrophoblast boundary in term placenta, probably because there are very few cytotrophoblasts in mature placenta. Judging from the amounts of immunoreactive products where the mononuclear trophoblasts proliferated, more E G F receptors were found at the proximal part of the basal lamina, which separated the trophoblastic layer from the villous stroma, than at the distal part (Figure 6). T h e cytoplasm of both syncytiotrophoblast and cytotrophoblasts was negative for E G F receptors throughout the course of pregnancy. No immunoreactivity was observed in the cell components of the villous stroma.

io

Placenta 099o), Vol. z1

Figure2. Immunoreaction for E G F receptor of placental villi at 28 weeks of gestation.

Figure3. Immunoreaction for E G F receptor of mature placenta at 38 weeks of gestation.

Kawago et al: EGF-Receptors in Normal ttuman Placenta

II

Figure4. Immunoreaction for EGF receptor of tonsilar squamous epithelia as a positive control.

Figure 5. Semi-thin section of normal term placenta showing that the cytotrophoblastic surface beneath the syncytiotrophoblast was positive for EGF receptors as well.

12

Placenta 099o), Vol. I

Figure 6. Immunoreactiveproducts of EGF receptor showingthat more EGF receptors were present at the proximal part of the basal lamina than at the distal part.

By immunoelectronmicroscopy, EGF-receptors were detected on the cell surfaces of both syncytiotrophoblast and cytotrophoblasts. The outer cell membrane of trophoblasts adjacent to the basal lamina was mostly negative for EGF-receptors (Figure 7). At a higher magnification, immunoreactivity was noted in some micropinocytic vesicles just beneath the trophoblastic cell surface, and very rarely in the intracytoplasmic membrane structures, which were probably the endoplasmic reticulum (Figure 8). According to our observations, based on staining intensity, the distribution pattern, localization site, and amount of E G F receptors, were basically the same throughout gestation.

DISCUSSION The present study revealed that placental villi contained a much greater amount of E G F receptor than the control tissues of normal human tonsil. Our findings support those of biochemical analysis of the amount of E G F receptors in placenta (Carson et al, i983; Lai and Guyda, 1984). The localization of E G F receptors, as well as other receptors, has been reported to be membrane-bound, and usually bound to the cell surface (Sakai et al, I986; Mori et al, i987). Maruo and Mochizuki (1987) reported that E G F receptors are localized exclusively in syncytiotrophoblast, paralleling the irnmunohistochemical localization of hCG. They found that the cytoplasm, as well as the free surface of syncytiotrophoblast, was positive for E G F receptors, which decreased during the course of pregnancy. In contrast, our study revealed EGF receptors only on the free surface, not the cytoplasm, of syncytiotrophoblast. In the cytotrophoblastic layer, immunoreactivity was also present at the cell surface, except in the region facing the basal lamina. This distribution pattern of E G F receptors did not change during the gestationai

Kawago et al: EGF-Receptors in Normal Human Placenta

x3

Figure 7. Immunoelectron microscopical localization of EGF receptors in term placenta. Original magnification 2000

•

.

Fzgure 8. Immunoelectron microscopical findings of EGF receptors showing that some micropinocytic vesicles just beneath the trophoblastic cell surface, and very rarely, the intracytoplasmic membrane structures, were positive for EGF receptors.

14

Placenta 099o), Vol. 11

period; furthermore, the amount of EGF receptors was much the same during the course of pregnancy. The discrepancy between our data and the previous report can be explained by differences in the anti-EGF receptor monoclonal antibodies that were employed in each study, as well as by the preparation of tissue sections, frozen ones in our study and formalin-fixed paraffin-embedded ones in the previous study. Our data imply that EGF receptors remain structurally intact and in copious quantities throughout the course of pregnancy. We previously reported that the free surface of syncytiotrophoblast was covered with abundant microvilli, o.4-i.8/am in length (Kawagoe, Kawana and Sakamoto, I98I ). The microvilli of immature trophoblast were greater in number, thinner, longer, and more complicated in structure than those of mature trophoblast (Kawagoe, Kawana and Sakamoto, i98o ). This ultrastructural evidence was considered by the present light and immunoelectronmicroscopic findings. Morrish et al 0987) found that EGF induced large sustained increases in hCG and hPL secretion, in a dose-dependent manner, as well as ultrastructural differentiation of cytotrophoblast to syncytiotrophoblast. They concluded that EGF causes morphological differentiation, but not cell proliferation, in trophoblast, and that the differentiation resulted in increased hCG and hPL secretion from the syncytium (Morrish et al, I987). They speculated that EGF on trophoblast induces the morphological and functional maturation of trophoblast rather than stimulates cytotrophoblastic proliferation. As for cytotrophoblasts in the placental villi, the outer surfaces were positive for EGF receptors. Furthermore, the EGF receptors were more abundant at the proximal part to the basal lamina separating the trophoblastic layer from villous stroma than at the distal part where mononuclear trophoblasts proliferated. These findings may help explain the role of EGF in trophoblastic proliferation, because proliferation of trophoblast only occurs in the cytotrophoblastic layer (Richart, I961; Pierce, Midgley and Beals, i964). Concerning the blood supply of EGF, the EGF receptors are mainly located on the free surface of syncytiotrophoblast facing maternal blood, as well as on the cell surface of cytotrophoblasts that are in close contact to the fetal circulation. We speculate that EGF receptors on syncytiotrophoblast stimulate hCG and hPL secretions, while those on cytotrophoblasts play a role in trophoblast proliferation and maturation, probably mediated by EGF in the fetal circulation.

SUMMARY With an avidin-biotin-peroxidase (or glucose oxidase) complex method using anti-epidermal growth factor receptor monoclonal antibody (528 IgG), the tissue and cellular distribution of the receptors for epidermal growth factors (EGF) in normal human placental villi, from 6 to 42 weeks of gestation, were studied. EGF receptors were mainly localized on the free surface of the syncytiotrophoblast that directly faced to intervillous space of the maternal circulation. The cell surface of cytotrophoblasts, except for the region that was adjacent to the basal lamina, was also positive for EGF receptors. The receptors were in close contact to the fetal vessels in the villous stroma. The EGF receptors on the syncytiotrophoblast were thought to be involved in the production and secretion of human chorionic gonadotropin and placental lactogen, probably under the control of maternal EGF. The receptors on cytotrophoblasts may p!ay a role in trophoblastic proliferation, possibly mediated by EGF in the fertal circulatory system.

Kawago et al: EGF-Receptors in Normal Human Placenta

15

ACKNOWLEDGEMENTS The authors cordially thank Professor Takashi Kawana of Tokyo University Branch Hospital and Dr Mashiko Yanagida, President of Tokyo Metropolitan Tsukiji Sanin Hospital, for their kind advice and help. We would like to express thanks to Mr Tohru Morishita for his valuable technical assistance. This study was supported by a Grant from the Japanese Ministry of Education (No. 6,350,770).

REFERENCES Ances, I. G. (1973) Serum concentrations of epidermal growth factor in human pregnancy. American Journal of Obstetrics and Gynecology, i i5,357-362. Branes, D. & Sato, G. (I98O) Methods for growth of cultured cells in serum-free medium. Analytical Biochemistry, ~o2, 255 570. Carson, S. A., Chase, R., Ulep, E., S c o m m e g n a , A. & Benveniste, R. (i983) Ontogenesis and characteristics of epidermal growth factor receptors in human placenta. American Journal of Obstetrics and Gynecology, 147, 932~939. Cohen, S. (i962) Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the newborn animal. Journal of Biological Chemistry, 37, 1555 1562. Gregory, H. 0975) Isolation and structure of urogastrone and its relationship to epidermal growth factor. Nature (London), 257, 3z5 327 . Hunter, T. 0984) The epidermal growth factor receptor gene and its product. Nature (London), 3 n , 414 416. Hsu, S. M., Rayhe, L. & Anger, H. 0 9 8 0 Use of avidin biotin peroxidase complex (ABC) in immunoperoxidase techniques. Journal of Histochemistry and Cytochemistry, z9, 577 580. Kawagoe, K., K a w a n a , T. & Sakamoto, S. (i98o) Ultrastructural demonstration of the negative surface charge on human trophoblast. Acta Histochemica et C~ytochemica,I3, 25~269. Kawagoe, K., K a w a n a , T. & Sakamoto, S. (I98I) Negative surface charge of abnormal human trophoblast. Acta Histochemica et Cytochemica, 14, 554 560. K a w a m o t o , T., Sato, J. D., Le, A., Polikoff, J., Sato, H. & Mendelsohn, J. (I983) Growth stimulation of A 43I cells by epidermal growth factor: Identification of high-affinity receptors for epidermal growth factor by an antireceptor monoclonal antibody. Proceedings of the National Academy of Science of the United States of America, 8o, I337 I34I. Lai, W. H. & Guyda, H. J. (I984) Characterization and regulation of epidermal growth factor receptors in human placental cell cultures. Journal of Chnical Endocrinology and Metabolism, 58, 344-352. Maruo, T. & Mochizuki, M. (I987) Immunohistochemical localization of epidermal growth factor receptor and myc oncogene product in human placenta: Implication for trophoblast and differentiation. American Journal of Obstetrics and Gynecology, I 5 6 , 721-727 . McLean, I. W. & Nakane, P. K. (i974) Periodate-lysine-paraformaldehyde fixative: a new fixative for immunoelectron microscopy. Journal of Histochemistry and Cytochemistry, 77, Io47-IO5 aMorrish, D. W., Bhardwaj, D., D a b b a g h , L. K., Marusyk, H. & Siy, O. (I987) Epidermal growth factor induces differentiation and secretion of human chorionic gonadotropin and human placental lactogen in normal human placenta. Journal of Ch'nical Endocrinology and Metabolism, 65, 1287 1290. Pierce, G. B., Midgley, A. R. & Beals, T. F. (I964) An ultrastructural study of differentiation and maturation of trophoblast of the monkey. Laboratory Investigation, I3, 451464. Richart, R. (1961) Radioautographic studies of human placenta utilizing tritiated thymidine. Proceedings of the Society for Experimental Biology and Medicine, IO6, 829 83 i. Sakai, K., Mori, S., K a w a m o t o , T., T a n i g u c h i , S., Kobori, O., Morioka, Y., Kuroki, T. & Kano, K. (I986) Expression of epidermal growth factor receptors on normal human gastric epithelia and gastric carcinomas. Journal of the National Cancer Institute, 77, Io47 xo52.