Carcinogenesis vol.12 no.7 pp.1345-1349, 1991
SHORT COMMUNICATION Expression of E-cadherin, P-cadherin and involucrin by normal and neoplastic keratinocytes in culture
Linda J.Nicholson, Xu Fang Pei1 and Fiona M.Watt Keratinocyte Laboratory, Imperial Cancer Research Fund, PO Box 123, Lincoln's Inn Fields, London WC2A 3PX, UK 'Present address: Department of Pathology, Georgetown University Medical School, 3900 Reservoir Road, NW, Washington DC 20007, USA
We have compared expression of involucrin, E-cadherin and P-cadherin in cultures of normal keratinocytes and in five different lines derived from squamous cell carcinomas (SCCs), using Northern analysis and immunofluorescence. In normal keratinocytes there was an inverse correlation between P-cadherin and involucrin expression, whereas E-cadherin was expressed by both basal and terminally differentiating cells. In SCC lines involucrin expression was lower than in normal keratinocytes, and there was variable expression of P- and E-cadherin: E-cadherin mRNA levels tended to be lower in SCC lines than in normal keratinocytes, whereas P-cadherin levels were similar. Our results are consistent with observations of cadherin expression in vivo and suggest that the cultures provide a useful experimental model for investigating the role of cadherins in determining the spatial organization of normal and neoplastic keratinocytes.
E-Cadherin and P-cadherin belong to a family of transmembrane calcium-dependent cell—cell adhesion molecules, the members of which play a fundamental role in establishing and maintaining the multicellular structure of tissues (1). There is striking homology between the mouse and human cadherin sequences, and comparison of P- with E-cadherin also reveals regions of high homology, particularly in the cytoplasmic domains (2-6). In stratified squamous epithelia, such as the epidermis and the lining of the oesophagus and bronchus, E-cadherin is found in all the viable layers, whereas P-cadherin is restricted to the basal layer, sometimes extending into the lower suprabasal layers (7,8). In an immunohistochemical study of human lung carcinomas, P- and E-cadherin were found in all tumours, but P-cadherin was absent from the pearls of terminally differentiating cells found in the squamous cell carcinomas (SCCs*), consistent with the loss of P-cadherin during terminal differentiation in normal tissue (8). Human keratinocytes can be grown in culture under conditions in which they stratify, with proliferating cells restricted to the basal layer and cells undergoing terminal differentiation in the suprabasal layers; all suprabasal cells express involucrin, a major precursor of the cornified envelope (9). Cell lines can be established from SCCs and grown using the same techniques as for normal keratinocytes, thereby providing an in vitro model for comparing the properties of normal and neoplastic keratinocytes (10—12). The aims of the experiments described in this report were to establish whether there was an inverse •Abbreviations: SCC, squamous cell carcinoma; PBSAB, phosphate-buffered saline + 1 mM calcium chloride; PBSABC, phosphate-buffered saline + I mM calcium chloride + 1 mM magnesium chloride. © Oxford University Press
correlation between expression of P-cadherin and involucrin in cultures of normal keratinocytes, as would be predicted from the immunohistochemical studies of stratified squamous epithelia, and whether expression of P- and E-cadherin differed between normal cells and SCC lines. All cells were cultured in the presence of a mitomycin C-treated feeder layer of J2-3T3 cells, in a 1 plus 3 mixture of Ham's F12 and Dulbecco's modified Eagle's medium, supplemented with 1.8 X 10~ 4 M adenine, 10% fetal calf serum, 0.5 /ig/ml hydrocortisone and 5 jtg/ml insulin. The medium of normal, but
J2 0
INV
4 24
E-CAD P-CAD 18S
ff*
Fig. 1. Northern analysis of RNA extracted from J2-3T3 cells or from normal keratinocytes after 0, 4 or 24 h in suspension. Total RNA (15 /jg) was loaded per track and hybridized to various probes: involucrin (INV), Ecadherin (E-CAD), P-cadherin (P-CAD) and 18S rRNA (18S).
INV
E-CAD
P-CAD
12 12 27 F.2 a2
9
4
q
t
18S
Fig. 2. Northern analysis of RNA extracted from SCC lines (as indicated) and from normal keratinocytes (strain a). Total RNA (15 ^tg) was loaded per track and hybridized to the following probes: involucrin (INV), E-cadhenn (E-CAD), P-cadhcrin (P-CAD) and 18S rRNA (18S).
1345
LJ.Nkhoboo, X.F.Pei and F.M.Watt
FIgJ. Double label immunofluorescence for involucrin (a,c,e) and E-cadherin (b,d,f); (a,b) NormaJ keratinocytes. (c,d) SCC-12 F.2. (e,f) SCC-27. Scale bar = 50 jrni.
not SCC, keratinocytes also contained 10 10 M cholera toxin and 10 ng/ml epidermal growth factor (11,13,14). Normal keratinocytes (strains a, s, v, z, passage numbers 3-8) were derived from newborn human foreskin epidermis. SCC-4 and 9 were derived from SCCs of the tongue (11). SCC-12 was from an SCC of facial epidermis; two clones derived from it, F.2 and B.2, have been reported to differ in terminal differentiation capacity and tumorigenicity in nude mice (12). SCC-27 was established from keratinocytes that metastasized to the peritoneal cavity from an SCC of the vulva (J.Rheinwald, personal communication). Total RNA was isolated from cells by guanidine thiocyanate extraction and caesium chloride density gradient centrifugation, 1346
essentially as described by Chirgwin et al. (15). RNA was electrophoresed in 1 % agarose gels containing formaldehyde in buffered formaldehyde solution, then transferred to Hybond-N (Amersham) and probed with the following cDNAs (16): human involucrin, pl-2 (17); mouse E-cadherin, pEM2 (18); human Ecadherin, pv 962 (5); mouse P-cadherin, P-cad/BLSC (18); and a mouse 18S rRNA probe, 100D9 (19). Transcript sizes were as follows: involucrin, 2.1 kb; E-cadherin, - 4 . 5 kb; P-cadherin, - 3 . 2 kb; 18S rRNA, 2.0 kb. The 18S rRNA probe was used as a control for the total amount of RNA loaded in each gel track (Figures 1 and 2). In order to establish whether there was an inverse correlation between the levels of involucrin and P-cadherin mRNA in normal
Cadberin and involucrin expression by keratinocytes
Fig. 4. Double label immunofluorescence for involucrin (a,c,e) and P-cadherin (b,d,f). (a,b) Normal keratinocytes. (c,d) SCC-4. There are no involucrinpositive cells in this field. (e,f) SCC-27. Two cells are marked with a star. The one at the top of the field is involucrin-positive, P-cadherin-positive; the other cell is involucrin-negative, P-cadherin-negative. Scale bar = 50 urn.
keratinocytes, RNA was isolated from cells that had been induced to undergo terminal differentiation by disaggregation and suspension in medium made viscous by addition of methylcellulose (20,21). By 24 h, a 4-fold increase in involucrin mRNA was observed (Figure 1); this correlates with an ~ 3-fold increase in the number of cells expressing involucrin protein, as assessed by immunofluorescence (21). In contrast, mRNA for P-cadherin was present at similar levels at 0 and 4 h in suspension, but decreased markedly between 4 and 24 h (Figure 1). The level of E-cadherin mRNA increased slightly between 0 and 4 h in suspension, a time during which the proportion of cells expressing involucrin does not increase (21), but was
unchanged between 4 and 24 h (Figure 1). As expected, J2-3T3 feeder cells did not express detectable involucrin, P-cadherin or E-cadherin (Figure 1). We next compared mRNA levels for involucrin, E- and P-cadherin in normal keratinocytes and SCC lines (Figure 2). The level of involucrin mRNA was much higher in normal keratinocytes than in any of the SCC lines, as would be predicted from the reduced capacity of the lines for terminal differentiation (10,12). The amount of involucrin mRNA varied between different SCC lines, but the band intensities on the gels were too low to rank the lines according to terminal differentiation capacity on that basis (10,12). E-cadherin was detected in all the SCC 1347
LJ.Nlcholson, X.F.Pd and F.M.Watt
lines; the level varied between lines, but was generally lower than in normal keratinocytes. Of the SCC lines examined, SCC-12 B.2 expressed the highest levels of E-cadherin mRNA. P-Cadherin mRNA was higher in SCC-9, SCC-12 F.2 and SCC-12 B.2 than in normal keratinocytes, whereas the levels in SCC-27 and SCC-4 were equivalent to or slightly lower than the level in normal cells. We also compared cadherin and involucrin expression by double label immunofluoresence staining of normal and SCC keratinocytes (Figures 3 and 4, and results not shown). Cells were fixed in 3.7% formaldehyde in phosphate-buffered saline containing 1 mM calcium chloride (PBSAB) or, in addition, 1 mM magnesium chloride (PBSABC) for 30 min at 4°C, then permeabilized in methanol for 10 min at -20°C. The cells were then incubated for 30—90 min at room temperature in PBSAB or PBSABC containing 10% skimmed milk powder. Cells were incubated first with mouse monoclonal antibodies NCC-CAD-299 (to P-cadherin) or HECD-1 (to E-cadherin) (8), then with fluorescein-conjugated rabbit anti-mouse IgG, then with rabbit antiserum to involucrin (22) and finally with rhodamineconjugated goat anti-rabbit IgG. The same results were obtained when anti-involucrin was added first, followed by the anti-rabbit second antibody, then anti-cadherin, then anti-mouse. Antibodies were diluted in PBSAB or PBSABC, containing 10% skimmed milk powder. Antibody incubations were for 1 - 2 h at room temperature or (NCC-CAD-299 and HECD-1) overnight at 4°C. As reported previously (23), involucrin expression was restricted to suprabasal cells in stratified colonies of normal keratinocytes (Figures 3a and 4a). In the SCC lines, involucrin expression was usually suprabasal (e.g. Figure 3c), but not always (e.g. Figure 4e); and the number of involucrin-positive cells was lower than in normal cultures (Figures 3a,c,e; 4a,c,e). We were unable to rank the SCC lines according to terminal differentiation capacity on the basis of involucrin immunofluorescence, although such a ranking has been reported by involucrin ELJSA (12). E-Cadherin was localized to regions of cell-cell contact (Figure 3b,d,f)- In cultures of normal keratinocytes (Figure 3b) and the SCC lines (Figure 3d,f) both involucrin-positive and involucrin-negative cells expressed E-cadherin. The intensity of staining in most lines was as strong as in normal keratinocytes (cf. Figure 3b,d). However, in SCC-27 there was marked cellto-cell variation in staining intensity (Figure 3f). P-Cadherin was also concentrated at the boundaries between neighbouring cells (Figure 4b,d,f). In colonies of normal keratinocytes, P-cadherin expression was largely confined to the basal layer (Figure 4b). In the SCC lines, involucrin-positive cells tended to be P-cadherin-negative, but involucrin-positive, P-cadherin-positive cells were also observed, as were involucrinnegative, P-cadherin-negative cells (Figure 4f). Within the lines, staining for P-cadherin was more heterogeneous than staining for E-cadherin (cf. Figures 3d and 4d) and SCC-27, in particular, contained a high proportion of unstained cells (cf. Figure 4d,f). In summary, we observed that in cultures of normal keratinocytes E-cadherin was localized to cell—cell boundaries in basal and in suprabasal layers, whereas P-cadherin was restricted to the basal layer. This is in good agreement with the relative distribution of P- and E-cadherin in normal stratified squamous epithelia (7,8). The data from Northern blots of keratinocytes induced to undergo terminal differentiation in suspension are consistent with the immunofluorescence results: between 4 and 24 h in suspension, involucrin mRNA increased in abundance, while P-cadherin decreased and E-cadherin was unchanged. 1348
By immunofluorescence, E-cadherin was observed at intercellular boundaries in the five SCC lines examined; in some cases the staining intensity was equivalent to that in normal keratinocytes—in agreement with the uniform staining of sections of SCCs of the lung that has been reported (8)—but in others staining was weaker and more heterogeneous. Northern analysis of E-cadherin mRNA levels revealed variable expression amongst the lines. Low levels of E-cadherin mRNA and protein have been correlated with invasive and metastatic potential in a number of experimental models (24-26). The SCC lines we examined have yet to be tested for metastatic ability, but it is interesting that SCC-27, which expressed low levels of E-cadherin, was derived from a metastasis rather than a primary tumour. P-Cadherin mRNA levels varied between lines and in some cases were higher than in normal keratinocytes. Whereas in cultures of normal cells involucrin-positive cells were usually P-cadherin-negative, amongst the SCC lines examples of involucrin-positive, P-cadherin-positive and involucrin-negative, P-cadherin-negative cells were often seen, suggesting that the inverse relationship between involucrin and P-cadherin expression had broken down. In conclusion, expression of P- and E-cadherin in cultures of normal keratinocytes mirrors expression in stratified squamous epithelia in vivo. In SCC lines involucrin expression is decreased and there is variable expression of P- and E-cadherin. Thus the cultures provide a useful experimental model with which to test the roles of P- and E-cadherin in determining the normal spatial organization of keratinocytes and the changes that occur on neoplastic transformation. Acknowledgements We are grateful to S.Hirohashi and M.Takeichi for providing antibodies, to D.Edwards, H.Green, M.Takeichi and R.Kemler for providing cDNA probes, and to J.Rbeinwald for providing the SCC lines. We thank W.Senior for expert typing of the manuscript.
References l.Takeichi.M. (1991) Cadherin cell adhesion receptors as a morphogenetic regulator. Science, 251, 1451-1455. 2. NagafuchiA, Shirayoshi.Y., Okazaki.K., YasudaJC. and Takeichi.M. (1987) Transformation of cell adhesion properties by exogenously introduced Ecadherin cDNA. Nature, 329, 341-343. 3. Nose,A., Nagafuchi.A. and Takeichi,M. (1987) Isolation of placenta! cadherin cDNA: identification of a novel gene family of cell-cell adhesion molecules. EMBOJ., 6, 3655-3661. 4. Wheelock.M.J., Buck,C.A., Bechtol.K.B. and Damsky.C.H. (1987) Soluble 80-kd fragment of cell-CAM 120/80 disrupts cell-cell adhesion. J. Cell Biochem., 34, 187-202. 5.Mansouri,A., Spurr,N., Goodfellow.P.N. and Kemler.R. (1988) Characterization arid chromosomal localization of the gene encoding the human cell adhesion molecule uvomorulin. Differentiation, 38, 6 7 - 7 1 . 6. Shimoyama.Y., Yoshida,T., Terada.M., Shimosato.Y., Abe.O. and Hirohashi.S. (1989) Molecular cloning of a human Ca2+-dependent cell-cell adhesion molecule homologous to mouse placental cadherin: hs low expression in human placental tissues. /. Cell Biol., 109, 1787-1794. 7. Nose.A. and Takeichi,M. (1986) A novel cadherin cell adhesion molecule: its expression patterns associated with implantation and organogenesis of mouse embryos. J. Cell Biol., 103, 2649-2658. 8. Shimoyama.Y., Hirohashi,S., Hirano.S., Noguchi.M., Shimosato.Y., Takeichi,M. and Abe,O. (1989) Cadherin cell-adhesion molecules in human epithelial tissues and carcinomas. Cancer Res., 49, 2128—2133. 9. Watt,F.M. (1989) Terminal differentiation of epidermal keratinocytes. Current Opinion Cell Biol., 1, 1107-1115. 10. Rheinwald J.G. and Beckett,M.A. (1980) Defective terminal differentiation in culture as a consistent and selectable character of malignant human keratinocytes. Cell, 22, 629-632. 11. Rheinwald J.G. and Beckett,M.A. (1981) Tumorigenic keratinocyte lines requiring anchorage and fibroblast support cultured from human squamous cell carcinomas. Cancer Res., 41, 1657-1663.
Cadherin and Involucrin expression by keratinocytes 12. RheinwakU.G., Germain.E. and Beckett.M.A. (1983) Expression of keratins and envelope proteins in normal and malignant human keratinocytes and mesothelial cells. In Harris.C.C. and Autrup,H.N. (eds), Human Carcinogenesis. Academic Press, New York, pp. 85-96. 13. RheinwakU.G. and Green,H. (1975) Serial cultivation of strains of human epidermal keratinocytes: die formation of keratinizing colonies from single cells. Cell, 6, 331-344. 14. Morrison.A.I., Keeble.S. and Watt.F.M. (1988) The peanut lectin-binding glycoproteins of human epidermal keratinocytes. Exp. Cell Res., 177, 247-256. 15. Chirgwin,J.M., Przybyla.A.E., MacDonald.RJ. and Rutter.W.J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18, 5294-5299. 16. Maniatis.T., Fritsch.E.F. and SambrookJ. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 17. Eckert.R.L. and Green,H. (1986) Structure and evolution of the human involucrin gene. Cell, 46, 583-589. 18. Nose.A., Nagafuchi,A. and Takeichi.M. (1988) Expressed recombinant cadherins mediate cell sorting in model systems. Cell, 54, 993-1001. 19. Edwards,D.R., Murphy.G., ReynoUsJ.J., Whitham.S.E., Docherty.A.J.P., Angel.P. and Heath.J.K. (1987) Transforming growth factor beta modulates the expression of collagenase and metalloproteinase inhibitor. EMBO J., 6, 1899-1904. 20. Green,H. (1977) Terminal differentiation of cultured human epidermal cells. Cell, 11, 405-416. 21. Adams,J.C. and Watt.F.M. (1989) Fibroncctin inhibits die terminal differentiation of human keratinocytes. Nature, 340, 307-309. 22. Dover.R. and Watt.F.M. (1987) Measurement of the rate of epidermal terminal differentiation: expression of involucrin by S-phase keratinocytes in culture and in psoriatic plaques. J. Invest. Dermatol., 89, 349-352. 23. Banks-Schlegel.S. and Green,H. (1981) Involucrin synthesis and tissue assembly by keratinocytes in natural and cultured human epithelia. J. Cell Biol., 90, 732-737. 24. Hashimoto.M., Niwa.O., Nitta.Y., Takeichi.M. and Yokoro.K. (1989) Unstable expression of E-cadherin adhesion molecules in metastatic ovarian tumor cells. Jpn. J. Cancer Res., 80, 459-463. 25.BehrensJ., Mareel.M.M., Van Roy.F.M. and Birchmeier.W. (1989) Dissecting tumor cell invasion: epithelial cells acquire invasive properties after the loss of uvomorulin-mediated cell—cell adhesion. J. Cell Biol., 108, 2435-2447. 26. Frixen.U.H., Behrens.J., Sachs.M., Eberle.G., Voss.B., Warda.A., Ldchner.D. and Birchmeier.W. (1991) E-Cadherin-mediated cell-cell adhesion prevents invasiveness of human carcinoma cells. J. Cell Biol., 113, 173-185. Received on May 14 1990; revised on April 11, 1991; accepted on April 17, 1991
1349
SHORT COMMUNICATION Expression of E-cadherin, P-cadherin and involucrin by normal and neoplastic keratinocytes in culture
Linda J.Nicholson, Xu Fang Pei1 and Fiona M.Watt Keratinocyte Laboratory, Imperial Cancer Research Fund, PO Box 123, Lincoln's Inn Fields, London WC2A 3PX, UK 'Present address: Department of Pathology, Georgetown University Medical School, 3900 Reservoir Road, NW, Washington DC 20007, USA
We have compared expression of involucrin, E-cadherin and P-cadherin in cultures of normal keratinocytes and in five different lines derived from squamous cell carcinomas (SCCs), using Northern analysis and immunofluorescence. In normal keratinocytes there was an inverse correlation between P-cadherin and involucrin expression, whereas E-cadherin was expressed by both basal and terminally differentiating cells. In SCC lines involucrin expression was lower than in normal keratinocytes, and there was variable expression of P- and E-cadherin: E-cadherin mRNA levels tended to be lower in SCC lines than in normal keratinocytes, whereas P-cadherin levels were similar. Our results are consistent with observations of cadherin expression in vivo and suggest that the cultures provide a useful experimental model for investigating the role of cadherins in determining the spatial organization of normal and neoplastic keratinocytes.
E-Cadherin and P-cadherin belong to a family of transmembrane calcium-dependent cell—cell adhesion molecules, the members of which play a fundamental role in establishing and maintaining the multicellular structure of tissues (1). There is striking homology between the mouse and human cadherin sequences, and comparison of P- with E-cadherin also reveals regions of high homology, particularly in the cytoplasmic domains (2-6). In stratified squamous epithelia, such as the epidermis and the lining of the oesophagus and bronchus, E-cadherin is found in all the viable layers, whereas P-cadherin is restricted to the basal layer, sometimes extending into the lower suprabasal layers (7,8). In an immunohistochemical study of human lung carcinomas, P- and E-cadherin were found in all tumours, but P-cadherin was absent from the pearls of terminally differentiating cells found in the squamous cell carcinomas (SCCs*), consistent with the loss of P-cadherin during terminal differentiation in normal tissue (8). Human keratinocytes can be grown in culture under conditions in which they stratify, with proliferating cells restricted to the basal layer and cells undergoing terminal differentiation in the suprabasal layers; all suprabasal cells express involucrin, a major precursor of the cornified envelope (9). Cell lines can be established from SCCs and grown using the same techniques as for normal keratinocytes, thereby providing an in vitro model for comparing the properties of normal and neoplastic keratinocytes (10—12). The aims of the experiments described in this report were to establish whether there was an inverse •Abbreviations: SCC, squamous cell carcinoma; PBSAB, phosphate-buffered saline + 1 mM calcium chloride; PBSABC, phosphate-buffered saline + I mM calcium chloride + 1 mM magnesium chloride. © Oxford University Press
correlation between expression of P-cadherin and involucrin in cultures of normal keratinocytes, as would be predicted from the immunohistochemical studies of stratified squamous epithelia, and whether expression of P- and E-cadherin differed between normal cells and SCC lines. All cells were cultured in the presence of a mitomycin C-treated feeder layer of J2-3T3 cells, in a 1 plus 3 mixture of Ham's F12 and Dulbecco's modified Eagle's medium, supplemented with 1.8 X 10~ 4 M adenine, 10% fetal calf serum, 0.5 /ig/ml hydrocortisone and 5 jtg/ml insulin. The medium of normal, but
J2 0
INV
4 24
E-CAD P-CAD 18S
ff*
Fig. 1. Northern analysis of RNA extracted from J2-3T3 cells or from normal keratinocytes after 0, 4 or 24 h in suspension. Total RNA (15 /jg) was loaded per track and hybridized to various probes: involucrin (INV), Ecadherin (E-CAD), P-cadherin (P-CAD) and 18S rRNA (18S).
INV
E-CAD
P-CAD
12 12 27 F.2 a2
9
4
q
t
18S
Fig. 2. Northern analysis of RNA extracted from SCC lines (as indicated) and from normal keratinocytes (strain a). Total RNA (15 ^tg) was loaded per track and hybridized to the following probes: involucrin (INV), E-cadhenn (E-CAD), P-cadhcrin (P-CAD) and 18S rRNA (18S).
1345
LJ.Nkhoboo, X.F.Pei and F.M.Watt
FIgJ. Double label immunofluorescence for involucrin (a,c,e) and E-cadherin (b,d,f); (a,b) NormaJ keratinocytes. (c,d) SCC-12 F.2. (e,f) SCC-27. Scale bar = 50 jrni.
not SCC, keratinocytes also contained 10 10 M cholera toxin and 10 ng/ml epidermal growth factor (11,13,14). Normal keratinocytes (strains a, s, v, z, passage numbers 3-8) were derived from newborn human foreskin epidermis. SCC-4 and 9 were derived from SCCs of the tongue (11). SCC-12 was from an SCC of facial epidermis; two clones derived from it, F.2 and B.2, have been reported to differ in terminal differentiation capacity and tumorigenicity in nude mice (12). SCC-27 was established from keratinocytes that metastasized to the peritoneal cavity from an SCC of the vulva (J.Rheinwald, personal communication). Total RNA was isolated from cells by guanidine thiocyanate extraction and caesium chloride density gradient centrifugation, 1346
essentially as described by Chirgwin et al. (15). RNA was electrophoresed in 1 % agarose gels containing formaldehyde in buffered formaldehyde solution, then transferred to Hybond-N (Amersham) and probed with the following cDNAs (16): human involucrin, pl-2 (17); mouse E-cadherin, pEM2 (18); human Ecadherin, pv 962 (5); mouse P-cadherin, P-cad/BLSC (18); and a mouse 18S rRNA probe, 100D9 (19). Transcript sizes were as follows: involucrin, 2.1 kb; E-cadherin, - 4 . 5 kb; P-cadherin, - 3 . 2 kb; 18S rRNA, 2.0 kb. The 18S rRNA probe was used as a control for the total amount of RNA loaded in each gel track (Figures 1 and 2). In order to establish whether there was an inverse correlation between the levels of involucrin and P-cadherin mRNA in normal
Cadberin and involucrin expression by keratinocytes
Fig. 4. Double label immunofluorescence for involucrin (a,c,e) and P-cadherin (b,d,f). (a,b) Normal keratinocytes. (c,d) SCC-4. There are no involucrinpositive cells in this field. (e,f) SCC-27. Two cells are marked with a star. The one at the top of the field is involucrin-positive, P-cadherin-positive; the other cell is involucrin-negative, P-cadherin-negative. Scale bar = 50 urn.
keratinocytes, RNA was isolated from cells that had been induced to undergo terminal differentiation by disaggregation and suspension in medium made viscous by addition of methylcellulose (20,21). By 24 h, a 4-fold increase in involucrin mRNA was observed (Figure 1); this correlates with an ~ 3-fold increase in the number of cells expressing involucrin protein, as assessed by immunofluorescence (21). In contrast, mRNA for P-cadherin was present at similar levels at 0 and 4 h in suspension, but decreased markedly between 4 and 24 h (Figure 1). The level of E-cadherin mRNA increased slightly between 0 and 4 h in suspension, a time during which the proportion of cells expressing involucrin does not increase (21), but was
unchanged between 4 and 24 h (Figure 1). As expected, J2-3T3 feeder cells did not express detectable involucrin, P-cadherin or E-cadherin (Figure 1). We next compared mRNA levels for involucrin, E- and P-cadherin in normal keratinocytes and SCC lines (Figure 2). The level of involucrin mRNA was much higher in normal keratinocytes than in any of the SCC lines, as would be predicted from the reduced capacity of the lines for terminal differentiation (10,12). The amount of involucrin mRNA varied between different SCC lines, but the band intensities on the gels were too low to rank the lines according to terminal differentiation capacity on that basis (10,12). E-cadherin was detected in all the SCC 1347
LJ.Nlcholson, X.F.Pd and F.M.Watt
lines; the level varied between lines, but was generally lower than in normal keratinocytes. Of the SCC lines examined, SCC-12 B.2 expressed the highest levels of E-cadherin mRNA. P-Cadherin mRNA was higher in SCC-9, SCC-12 F.2 and SCC-12 B.2 than in normal keratinocytes, whereas the levels in SCC-27 and SCC-4 were equivalent to or slightly lower than the level in normal cells. We also compared cadherin and involucrin expression by double label immunofluoresence staining of normal and SCC keratinocytes (Figures 3 and 4, and results not shown). Cells were fixed in 3.7% formaldehyde in phosphate-buffered saline containing 1 mM calcium chloride (PBSAB) or, in addition, 1 mM magnesium chloride (PBSABC) for 30 min at 4°C, then permeabilized in methanol for 10 min at -20°C. The cells were then incubated for 30—90 min at room temperature in PBSAB or PBSABC containing 10% skimmed milk powder. Cells were incubated first with mouse monoclonal antibodies NCC-CAD-299 (to P-cadherin) or HECD-1 (to E-cadherin) (8), then with fluorescein-conjugated rabbit anti-mouse IgG, then with rabbit antiserum to involucrin (22) and finally with rhodamineconjugated goat anti-rabbit IgG. The same results were obtained when anti-involucrin was added first, followed by the anti-rabbit second antibody, then anti-cadherin, then anti-mouse. Antibodies were diluted in PBSAB or PBSABC, containing 10% skimmed milk powder. Antibody incubations were for 1 - 2 h at room temperature or (NCC-CAD-299 and HECD-1) overnight at 4°C. As reported previously (23), involucrin expression was restricted to suprabasal cells in stratified colonies of normal keratinocytes (Figures 3a and 4a). In the SCC lines, involucrin expression was usually suprabasal (e.g. Figure 3c), but not always (e.g. Figure 4e); and the number of involucrin-positive cells was lower than in normal cultures (Figures 3a,c,e; 4a,c,e). We were unable to rank the SCC lines according to terminal differentiation capacity on the basis of involucrin immunofluorescence, although such a ranking has been reported by involucrin ELJSA (12). E-Cadherin was localized to regions of cell-cell contact (Figure 3b,d,f)- In cultures of normal keratinocytes (Figure 3b) and the SCC lines (Figure 3d,f) both involucrin-positive and involucrin-negative cells expressed E-cadherin. The intensity of staining in most lines was as strong as in normal keratinocytes (cf. Figure 3b,d). However, in SCC-27 there was marked cellto-cell variation in staining intensity (Figure 3f). P-Cadherin was also concentrated at the boundaries between neighbouring cells (Figure 4b,d,f). In colonies of normal keratinocytes, P-cadherin expression was largely confined to the basal layer (Figure 4b). In the SCC lines, involucrin-positive cells tended to be P-cadherin-negative, but involucrin-positive, P-cadherin-positive cells were also observed, as were involucrinnegative, P-cadherin-negative cells (Figure 4f). Within the lines, staining for P-cadherin was more heterogeneous than staining for E-cadherin (cf. Figures 3d and 4d) and SCC-27, in particular, contained a high proportion of unstained cells (cf. Figure 4d,f). In summary, we observed that in cultures of normal keratinocytes E-cadherin was localized to cell—cell boundaries in basal and in suprabasal layers, whereas P-cadherin was restricted to the basal layer. This is in good agreement with the relative distribution of P- and E-cadherin in normal stratified squamous epithelia (7,8). The data from Northern blots of keratinocytes induced to undergo terminal differentiation in suspension are consistent with the immunofluorescence results: between 4 and 24 h in suspension, involucrin mRNA increased in abundance, while P-cadherin decreased and E-cadherin was unchanged. 1348
By immunofluorescence, E-cadherin was observed at intercellular boundaries in the five SCC lines examined; in some cases the staining intensity was equivalent to that in normal keratinocytes—in agreement with the uniform staining of sections of SCCs of the lung that has been reported (8)—but in others staining was weaker and more heterogeneous. Northern analysis of E-cadherin mRNA levels revealed variable expression amongst the lines. Low levels of E-cadherin mRNA and protein have been correlated with invasive and metastatic potential in a number of experimental models (24-26). The SCC lines we examined have yet to be tested for metastatic ability, but it is interesting that SCC-27, which expressed low levels of E-cadherin, was derived from a metastasis rather than a primary tumour. P-Cadherin mRNA levels varied between lines and in some cases were higher than in normal keratinocytes. Whereas in cultures of normal cells involucrin-positive cells were usually P-cadherin-negative, amongst the SCC lines examples of involucrin-positive, P-cadherin-positive and involucrin-negative, P-cadherin-negative cells were often seen, suggesting that the inverse relationship between involucrin and P-cadherin expression had broken down. In conclusion, expression of P- and E-cadherin in cultures of normal keratinocytes mirrors expression in stratified squamous epithelia in vivo. In SCC lines involucrin expression is decreased and there is variable expression of P- and E-cadherin. Thus the cultures provide a useful experimental model with which to test the roles of P- and E-cadherin in determining the normal spatial organization of keratinocytes and the changes that occur on neoplastic transformation. Acknowledgements We are grateful to S.Hirohashi and M.Takeichi for providing antibodies, to D.Edwards, H.Green, M.Takeichi and R.Kemler for providing cDNA probes, and to J.Rbeinwald for providing the SCC lines. We thank W.Senior for expert typing of the manuscript.
References l.Takeichi.M. (1991) Cadherin cell adhesion receptors as a morphogenetic regulator. Science, 251, 1451-1455. 2. NagafuchiA, Shirayoshi.Y., Okazaki.K., YasudaJC. and Takeichi.M. (1987) Transformation of cell adhesion properties by exogenously introduced Ecadherin cDNA. Nature, 329, 341-343. 3. Nose,A., Nagafuchi.A. and Takeichi,M. (1987) Isolation of placenta! cadherin cDNA: identification of a novel gene family of cell-cell adhesion molecules. EMBOJ., 6, 3655-3661. 4. Wheelock.M.J., Buck,C.A., Bechtol.K.B. and Damsky.C.H. (1987) Soluble 80-kd fragment of cell-CAM 120/80 disrupts cell-cell adhesion. J. Cell Biochem., 34, 187-202. 5.Mansouri,A., Spurr,N., Goodfellow.P.N. and Kemler.R. (1988) Characterization arid chromosomal localization of the gene encoding the human cell adhesion molecule uvomorulin. Differentiation, 38, 6 7 - 7 1 . 6. Shimoyama.Y., Yoshida,T., Terada.M., Shimosato.Y., Abe.O. and Hirohashi.S. (1989) Molecular cloning of a human Ca2+-dependent cell-cell adhesion molecule homologous to mouse placental cadherin: hs low expression in human placental tissues. /. Cell Biol., 109, 1787-1794. 7. Nose.A. and Takeichi,M. (1986) A novel cadherin cell adhesion molecule: its expression patterns associated with implantation and organogenesis of mouse embryos. J. Cell Biol., 103, 2649-2658. 8. Shimoyama.Y., Hirohashi,S., Hirano.S., Noguchi.M., Shimosato.Y., Takeichi,M. and Abe,O. (1989) Cadherin cell-adhesion molecules in human epithelial tissues and carcinomas. Cancer Res., 49, 2128—2133. 9. Watt,F.M. (1989) Terminal differentiation of epidermal keratinocytes. Current Opinion Cell Biol., 1, 1107-1115. 10. Rheinwald J.G. and Beckett,M.A. (1980) Defective terminal differentiation in culture as a consistent and selectable character of malignant human keratinocytes. Cell, 22, 629-632. 11. Rheinwald J.G. and Beckett,M.A. (1981) Tumorigenic keratinocyte lines requiring anchorage and fibroblast support cultured from human squamous cell carcinomas. Cancer Res., 41, 1657-1663.
Cadherin and Involucrin expression by keratinocytes 12. RheinwakU.G., Germain.E. and Beckett.M.A. (1983) Expression of keratins and envelope proteins in normal and malignant human keratinocytes and mesothelial cells. In Harris.C.C. and Autrup,H.N. (eds), Human Carcinogenesis. Academic Press, New York, pp. 85-96. 13. RheinwakU.G. and Green,H. (1975) Serial cultivation of strains of human epidermal keratinocytes: die formation of keratinizing colonies from single cells. Cell, 6, 331-344. 14. Morrison.A.I., Keeble.S. and Watt.F.M. (1988) The peanut lectin-binding glycoproteins of human epidermal keratinocytes. Exp. Cell Res., 177, 247-256. 15. Chirgwin,J.M., Przybyla.A.E., MacDonald.RJ. and Rutter.W.J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18, 5294-5299. 16. Maniatis.T., Fritsch.E.F. and SambrookJ. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 17. Eckert.R.L. and Green,H. (1986) Structure and evolution of the human involucrin gene. Cell, 46, 583-589. 18. Nose.A., Nagafuchi,A. and Takeichi.M. (1988) Expressed recombinant cadherins mediate cell sorting in model systems. Cell, 54, 993-1001. 19. Edwards,D.R., Murphy.G., ReynoUsJ.J., Whitham.S.E., Docherty.A.J.P., Angel.P. and Heath.J.K. (1987) Transforming growth factor beta modulates the expression of collagenase and metalloproteinase inhibitor. EMBO J., 6, 1899-1904. 20. Green,H. (1977) Terminal differentiation of cultured human epidermal cells. Cell, 11, 405-416. 21. Adams,J.C. and Watt.F.M. (1989) Fibroncctin inhibits die terminal differentiation of human keratinocytes. Nature, 340, 307-309. 22. Dover.R. and Watt.F.M. (1987) Measurement of the rate of epidermal terminal differentiation: expression of involucrin by S-phase keratinocytes in culture and in psoriatic plaques. J. Invest. Dermatol., 89, 349-352. 23. Banks-Schlegel.S. and Green,H. (1981) Involucrin synthesis and tissue assembly by keratinocytes in natural and cultured human epithelia. J. Cell Biol., 90, 732-737. 24. Hashimoto.M., Niwa.O., Nitta.Y., Takeichi.M. and Yokoro.K. (1989) Unstable expression of E-cadherin adhesion molecules in metastatic ovarian tumor cells. Jpn. J. Cancer Res., 80, 459-463. 25.BehrensJ., Mareel.M.M., Van Roy.F.M. and Birchmeier.W. (1989) Dissecting tumor cell invasion: epithelial cells acquire invasive properties after the loss of uvomorulin-mediated cell—cell adhesion. J. Cell Biol., 108, 2435-2447. 26. Frixen.U.H., Behrens.J., Sachs.M., Eberle.G., Voss.B., Warda.A., Ldchner.D. and Birchmeier.W. (1991) E-Cadherin-mediated cell-cell adhesion prevents invasiveness of human carcinoma cells. J. Cell Biol., 113, 173-185. Received on May 14 1990; revised on April 11, 1991; accepted on April 17, 1991
1349
