Nonrandom distribution of epidermal growth factor receptors on the plasma membrane of human A431...

Experimental Cell Research 175 (1988) 326-333

Nonrandom Distribution of Epidermal Growth Factor Receptors on the Plasma Membrane of Human A431 Cells

MARIA ROSARIA TORRISI,’ ANTONIO PAVAN, LAVINIA VITTORIA LOTTI, CLAUDIA ZOMPETTA,

ALBERT0 FAGGIONI, and LUIGI FRATI Dipartimento di Medicina Sperimentale, Universitd degli Studi di Roma “La Sapienza,”

Viale Regina Elena, 324, 00161 Roma, Italy

The localization of epidermal growth factor (EGF) receptors over the plasma membranes of human epidermoid carcinoma A43 1 cells was analyzed at the electron microscopic level using surface replica techniques and conventional thin sections, in combination with immunocytochemistry. Immunolabeling was performed using two distinct monoclonal antibodies directed against the extracellular portion of the receptor, followed by protein A-colloidal gold conjugates. Unexpectedly, with the first monoclonal antibody used, the distribution of the receptors in both untixed and glutaraldehyde-fixed cells was clearly regionalized, showing a preferential localization of the immunolabeling at the cell periphery as well as over the areas rich in microvilli and in coated and uncoated pits. A similar pattern of distribution was observed also with the other monoclonal antibody, but only when the cells were fixed with glutaraldehyde before immunolabeling. Treatment with the phorbol ester 12-O-tetradecanoylphorboLl3-acetate modifies this distribution, inducing a more disperse pattern. Our observations suggest that a minor group of EGF receptors. which may represent the high-affinity receptors, presents a regional distribution, similar to that described for typical recycling receptors. @ 1988 Academic PRSS, IK

The distribution of receptors for several ligands over the plasma membranes of motile or spreading cells has been recently ascribed as a consequence of the dynamics of the receptor itself (i.e., its internalization and recycling activity) [l-4]. In fact, recycled receptors and newly synthesized membrane components [5] are preferentially inserted at the periphery or free margin of fibroblasts or epithelial cells growing in monolayer. After insertion, these receptors maintain approximately their localization in those areas of the cell surface, until they are clustered and internalized via coated pits. This uneven distribution has been observed for typical recycling receptors such as those of low-density lipoproteins, ferritin, and transferrin on the surface of A431 human epidermoid carcinoma cells, giant HeLa cells, or fibroblasts [l-4]. On the contrary, epidermal growth factor (EGF) receptors have been described in the majority of reports as uniformly distributed over the surface of several cell types, rapidly clustered only upon binding of the ligand [6-101, and then down-regulated during the internalization process [II, 121. Recently however, evidence for EGF receptor recycling upon EGF withdrawal has been shown [ 131.

To analyze, at electron microscopic level, the distribution of EGF receptors

’ To whom reprint requests should be addressed.

Copyright @ 1988 by Academx Pres, Inc. 326 All rights of reproductmn in any form reserved 0014.4827188 $03.00

Distribution of EGF receptors 327

over the plasma membranes of A431 cells, we used both surface replicas [ 141 and conventional thin sections, in combination with immunocytochemistry. Immuno- labeling was performed with two distinct monoclonal antibodies, directed against the extracellular portion of the EGF receptor, followed by protein A-colloidal gold conjugates. The surface distribution was also analyzed in cells treated with 12-O-tetradecanoylphorbol-13-acetate (TPA), a reagent that inhibits the high- affinity binding of EGF [15-171.

MATERIALS AND METHODS

Cells and Immunolabeling A43 1 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10 % fetal calf

serum plus antibiotics. Before immunolabeling, cells were kept in serum-free medium for 1 h at 37°C. Immunolabeling of EGF receptors was performed on unfixed or fixed cells (2% formaldehyde or 0.5-l-2% glutaraldehyde 30-60 min, 25°C in phosphate-buffered saline (PBS), pH 7.4); the cells were incubated with the following anti-EGF receptor monoclonal antibodies: EGF-RI (Oncor, Gaithes- burg, MD) or 108.1 (a generous gift of Dr. J. Schlessinger, Meloy Laboratories, Rockville, MD) (0.1-0.3 mgiml in PBS for 1 h at 4”C), both directed against the extracellular portion of the receptor. The cells were then washed extensively and labeled for 3 h at 4°C with colloidal gold (prepared by the citrate method) conjugated with protein A (Pharmacia Fine Chemicals, Uppsala, Sweden) [18]. Cells were also treated with TPA (Sigma Chemical Company, St. Louis, MO) (100 ng’ml) at 37°C for 30 min before fixation and immunolabeling. Alternatively, cells were incubated with EGF (Collaborative Research Inc. Lexington, MA) (60 r&ml) at 4°C for 1 h and fixed or warmed to 37°C for 45 min before fixation and immunolabeling. Immunolabeling of HLA antigens was performed on prefixed cells using an anti-human HLA-A,B,C antiserum (1 h at 4°C) (Sera Lab, Crawley Down, Sussex, England) and then labeled with protein A-colloidal gold as above.

Processing for Electron Microscopy Thin sectioning. Labeled isolated cells were postfixed in 1% osmium tetroxide in Verona1 acetate

buffer, pH 7.4, for 2 h at 4°C stained with uranyl acetate (5 mgiml), dehydrated in acetone, and embedded in Epon 812. Thin sections were examined unstained and poststained with uranyl acetate and lead hydroxide.

Surface replication. Labeled cell monolayers on coverslips were dehydrated in a series of ethanol washes and air-dried. Platinum-carbon replicas were obtained in a Balzers BAF 300 freeze-etch apparatus (Balzers AG, Liechtenstein) by shadowing the cell surfaces with platinum-carbon. The replicas were cleaned overnight in household bleach, washed extensively in distilled water, and examined in a Philips EM 400 (Philips, Electron Instruments, Eindhoven, The Netherlands).

RESULTS

Our experiments were performed on A431 cells obtained from two different laboratories. A high variability of the receptor density from cell to cell was observed in all surface replicas and thin sections in accordance with a recent report [19].

The density of the immunolabeling obtained with the monoclonal antibody EGF-RI was lower than expected (approximately 3-4s of the total receptor population, as calculated by counting gold particles/urn* and by extrapolating to the whole cell surface; data not shown). Both density and pattern of distribution were totally comparable in unfixed as well as formaldehyde- and glutaraldehyde- prefixed cells. The distribution of EGF receptor over the cell surface in all thin sections and replicas was highly regionalized: the immunolabeling was preferen-

328 Torrisi et al. 328 Tori

d d

Fig. 1. Surface immunolabeling of A431 cells with monoclonal antibody EGF-RI: nonrandom distribution of EGF receptors over the cell surface (a, c, d, thin sections; b, e, f, surface replicas) with their preferential concentration on the microvilli (g, thin section; h, surface replica) and around coated pits (g, arrow; i, thin sections; j, arrows, surface replica); (a) ~11,850; (b) ~9,770; (c) x 17,750; (d) ~21,805; (e) ~24,000; cf) ~13,025; (g) ~25,160; (h) x7,770: (i) x41.440: (i) x34.040. Bar: 0.5 pm.

tially localized in portions of the plasma membranes, whereas very large areas were completely unlabeled (Figs. 1 a-f). In fact high concentration of the receptors in domains on the surface was observed mainly at the free margin of the cells and in the areas where microvilli (Figs. I a, g, h) and coated and uncoated pits (Fig. 1 g, arrow) were numerous. Even when the labeling was on the pit areas, the


Fig. 2. (a, b) Surface immunolabeling of A431 cells with monoclonal antibody EGF-RI. (a) Drastic reduction of surface EGF receptors, following EGF induced internalization (thin section); (b) following TPA treatment the immunolabeling with EGF-RI appears reduced and more dispersed (surface replica); (c) surface immunolabeling of HLA-A,B,C antigens on A431 cells: random distribution over the cell surface (surface replica). (a) ~26,500; (b) ~20,280; (c) x 14,605. Bar: 0.5 pm.

gold granules appeared unclustered and usually located around but not inside the pits (Figs. 1 i and lj, arrows).

This monoclonal antibody did not compete with EGF for binding to the cells. When A431 cells were incubated with EGF at 4°C for 1 h and then warmed to 37°C for 45 min before fixation, the surface labeling drastically decreased as expected [20], even in the peripheral areas with the pits (Fig. 2a), showing the high specificity of the immunolabeling procedure. Class I HLA antigens, which are known to be randomly distributed on A43 1 cells [ 11, were immunolabeled as a further control of our technical approach. Immunolabeling was performed using a monoclonal antibody anti-human HLA-A,B,C, followed by protein A-colloidal gold as above. As expected, the results confirm previous observations [l] (Fig. 2c).

To verify whether distribution of the receptors is related to their affinity, we next performed experiments with A431 cells treated with the tumor promoter TPA which is known to decrease EGF receptor aftinity [IS171. Treatment for 30 min at 37°C with TPA before fixation and immunolabeling induced approximately 30% decrease of the gold particles (assessed by counting gold granules/urn’ of the labeled areas). TPA treatment appeared to modify the distribution of the receptors: in fact, the gold granules were less concentrated and more evenly dispersed, although maintaining a regional pattern (Fig. 2b).

To confirm these observations we immunolabeled unfixed as well as formaldehyde- or glutaraldehyde-fixed A431 cells using a different monoclonal antibody also directed against the extracellular portion of the receptors, 108.1. In parallel

330 Torrisi et al.

Fig. 3. Surface immunolabeling of A431 cells with monoclonal antibody 108. I as observed in surface replicas: high density and random distribution of EGF receptors on unfixed (a) and formaldehyde-prefixed cells (b); low density and nonrandom distribution of the receptors on glutaraldehyde- prefixed cells (c). (a) x 10,490; (6) x 12,760; (c) x20.640. Bar: 0.5 km.

Distribution of EGF receptors 33 1

a

Fig. 4. Surface immunolabeling of A431 cells with monoclonal antibody 108.1 as observed in thin sections: high density and random distribution of the immunolabeling in unfixed (n) and formaldehyde-prefixed cells (b); low density and nonrandom distribution in glutaraldehyde-fixed cells (c). (a) x 17,000; (b) x 13,090; (C)X 10,200. Bar: 0.5 pm.

experiments with thin sections and surface replicas, a high density and uniform distribution of the labeling was observed in unfixed and formaldehyde-fixed cells (Figs. 3a and 3 b; Figs 4a and 4 b). In glutaraldehyde-fixed cells, however, the number of the receptors appeared reduced; the distribution was again regionalized showing a pattern comparable to that obtained when the monoclonal antibody EGF-Rl was used (Figs. 3c, 4~).

DISCUSSION

The present results show that at least part of EGF receptors are nonrandomly distributed over the surface of A431 cells, with a preferential localization at the cell periphery and in areas rich in microvilli and coated or uncoated pits. This observation conflicts with the majority of reports [6-101, although Haigler et al. have already observed that in A43 1 cells, using a ferritin-EGF conjugate, most of the binding at 4°C occurred at the cell-cell borders [21]. Moreover, in human fibroblasts “‘1-EGF, used as a probe at the electron microscopic level, appeared to bind at 4°C preferentially to coated pits, together with low-density lipoproteins [22]. Our study, due to the high resolution of both gold immunolabeling and

332 Torrid et al.

surface replica approaches, supports the preferential association of the EGF receptors with coated and uncoated pit domains, although it demonstrates that most gold granules appeared to be around but excluded from the pits. Indeed, the presence of the EGF receptors around coated and uncoated pits is in accordance with the involvement of both varieties of invaginations in the process of EGF internalization [23]. Therefore this pattern of distribution of the receptors does not seem peculiar of recycling type of receptors, but a more generalized phenom- enon. A similar association of receptors with microvilli has been also observed for insulin [24, 251, which is known to behave with an uptake process analogous to that of EGF. In addition, the polarized insertion of plasma membrane proteins, as well as the maintenance of a regional distribution, happens not only for different types of receptors, but also for viral proteins in infected cells [5, 26, 271, probably depending upon interactions among the proteins or with the cytoskeleton.

The low density of the immunolabeling observed with the first monoclonal antibody EGF-Rl compared to the total number of EGF receptors in A431 cells (-2x 106), suggests that we label only a minor part of the receptors. An indication that we may observe the high-affinity population derives from the observation that the receptor density appears comparable in unfixed and formaldehyde- or glutaraldehyde-fixed cells. In fact, as recently shown [28], formaldehyde fixation does not seem to affect the binding efficiency to EGF receptors, whereas glutaraldehyde fixation inhibits the binding of EGF or anti-EGF receptor monoclonal antibodies to low-affinity receptors. Glutaraldehyde fixation, however, does not affect the binding to high-affinity receptors ([28 and our unpublished observations), because of their association to the cytoskeleton which has been recently reported [29]. These results are confirmed by the immunolabeling with the second monoclonal antibody, 108.1. In fact, here, we can achieve a drastic reduction of the receptors associated with a regional distribution, similar to that described above, only after fixation of the cells with glutaraldehyde.

A further suggestion that the regional distribution is peculiar to the high-affinity EGF receptors derives from the experiments in which the cells were pretreated with TPA. In fact TPA treatment, which is known to decrease the affinity of the receptors [17], leads to a reduction of our immunolabeling. In addition it appears to modify the receptor distribution as a consequence, as previously suggested [30], of an increase in the lateral mobility of the immobile TPA-susceptible high- affinity receptors.

We thank Dr. J. Schlessinger for the generous gift of the monoclonal antibody 108.1, Mr. G. Lucania for technical assistance, and Mr. S. Valia for excellent photographic work. This work was partially supported by Grant 86.02397.44 from Progetto Finalizzato “Oncologia”, Consiglio Nazio- nale delle Ricerche, Italy.

REFERENCES 1. Hopkins, C. R. (1985) Cell 40, 199. 2. Bretscher, M. S. (1983) Proc. Natl. Acad. Sci. USA 80, 454. 3. Bretscher, M. S., and Thomson, J. N. (1983) EMBO J. 2, 599. 4. Ekblom, P., Thesleff, I., Lehto, V. P., Virtanen, I. (1983) Inr. J. Cancer 31, 111.


5. Bergmann, J. E., Kupfer, A., Singer. S. J. (1983) Proc. Natl. Acad. Sci. USA 80, 1367. 6. Schlessinger, J., Shechter, Y., Willingham, M. C.. and Pastan I. (1978) Proc. Natl. Acad. Sci.

USA 75, 2659. 7. McKanna, J. A., Haigler, H. T., and Cohen, S. (1979) Proc. Nut/. Acad. Sci. USA 76, 5689. 8. Hopkins, C. R., Boothroyd, B., Gregory, H. (1981) Eur. J. Cell Biol. 24, 259. 9. Miller, K., Beardmore, J., Kanety, H., Schlessinger. J.. and Hopkins C. R. (1986) J. Cell Eiol.

102, 500. 10. Schlessinger, J. (1986) J. Cell Biol. 103, 2067. 11. Carpenter, G., and Cohen, S. (1976) J. Cell Biol. 71, 159. 12. Beguinot, L., Lyall. R. M.. Willingham, M. C., and Pastan I. (1984) Proc. Nat/. Acad. Sci. USA

81, 2384. 13. Murthy, U., Basu. M.. Sen-Majumdar, A. and Das, M. (1986) J. Cell Biol. 103. 333. 14. Robenek. H., Hesz, A., and Rassat, J. (1983) J. Ultrastruct. Res. 82. 143. 15. Lee, L. S., and Weinstein, I. B. (1978) Science 202, 313. 16. Shoyab, M., Delarco, J. E., and Todaro, G. J. (1979) Nature (London) 279. 387. 17. Rozengurt, E. (1986) Science 234, 161. 18. Slot, J. W., and Geuze, H. J. (1981) J. Cc/l Eiol. 90, 533. 19. Carpentier, J-L., Rees, A. R., Gregoriou, M.. Kris, R.. Schlessinger, J., and Orci, L. (1986) Exp.

Cell Res. 166, 312. 20. Willingham, M. C. et al. (1983) Exp. Cell Res. 146, 163. 21. Haigler, H. T., McKanna, J. A., and Cohen, S. (1979) J. Cel[ Biol. 81, 382. 22. Carpentier, J-L., et al. (1982) .I. Cell Biol. 95, 73. 23. Hopkins, C. R., Miller, K., and Beardmore, J. M. (1985) J. Cell Sci. Suppl., 3, 173. 24. Carpentier, J-L., Van Obberghen, E., Gorden P., and Orci, L. (1981) J. Cell Biol. 91, 17. 25. Fan, J. Y., et al. (1982) Proc. Natl. Acad. Sci USA 79, 7788. 26. Marcus, P. I. (1962) Cold Spring Harbor Symp. Quant. Eiol. 27. 351. 27. Pavan A., Lotti. L. V., Tonisi, M. R., Migliaccio, G., and Bonatti, S. (1986) Exp. Cell Res. 164, 3. 28. Wiegant, F. A. C., Blok F. J., Defize, L. H. K., Linnemans. W. A. M.., Verkley, A. J., and

Boonstra, J. (1986) J. Cell Biol. 103, 87. 29. Landreth, G. E.. Williams, L. K.. and Rieser, G. D. (1985) J. Cell Biol. 101, 1341. 30. Rees. A. R., Gregoriou, M., Johnson P.. and Garland, P. B. (1984) EMBO J. 3, 1843.

Received December 18. 1986 Revised version received October 17. 1987

Printed in Sweden

Nonrandom distribution of epidermal growth factor receptors on the plasma membrane of human A431...

Documents

Transcript of Nonrandom distribution of epidermal growth factor receptors on the plasma membrane of human A431...