divergent expression of laminin and fibronectin in … · melanoma and sarcoma cells (terranova,...

J. Cell Sci. 76, 213-223 (1985) 213Printed in Great Britain © Company of Biologists Limited 1985

DIVERGENT EXPRESSION OF LAMININ ANDFIBRONECTIN IN NON-TUMORIGENIC ANDTRANSFORMED LIVER EPITHELIAL CELLS

J. L. JUNKER1 and M. J. WILSON2

'Ultrastructural Studies and 2Nutrition and Metabolism Sections, Laboratory ofComparative Carcinogenesis, National Cancer Institute, FCRF, Frederick, Maryland21701, U.SA.

SUMMARY

The immunocytochemical expression of laminin and fibronectin by non-transformed liverepithelial cells and by transformed cells derived from the same cell line, TRL1215, was examinedin ethionine-transformed cells (ETC), untreated control cells at the same high passage level (HPC),and untreated, low passage cells (LPC). At confluence, a divergent expression of laminin andfibronectin was observed in the three sublines. In flat areas of polygonal cells, LPC showed abundantlaminin staining and sparse fibronectin staining; HPC had intermediate expression of both; andETC had sparse laminin staining and abundant fibronectin. Specifically, the laminin network wasoften found at intercellular junctions, outlining individual LPC, outlining groups of HPC, andhaving a focal expression in ETC. A fine reticulum of laminin was seen in large areas of LPC, inrelatively small areas of HPC, and in multilayered areas of ETC. Fibronectin was visible as thick,matted fibrils, which were sparse in LPC, loosely arranged in HPC and dense in ETC. Prior studieshad shown that the increased fibronectin expression in the transformed cells was associated withincreased expression of actin stress fibres, increased cell spreading and increased numbers of focalcontacts between cell and substrate. The divergent expression of laminin and fibronectin shown hereindicates that these two matrix proteins need not be expressed in a parallel manner during trans-formation, and that increased fibronectin, but not laminin, is associated with maximal cell spreadingand adhesion.

INTRODUCTION

Laminin is a glycoprotein component of basement membranes (Timpl et al. 1979),and is one of a variety of substances that promote cell attachment and growth in vitro(Stenn, Madri, Tinghitella & Terranova, 1983; Couchman, H66k, Rees & Timpl,1983). Laminin specifically mediates the binding of cells of epithelial and endothelialorigin to another basement membrane component, type IV collagen (Kefalides,1973; Timpl et al. 1978; Terranova, Rohrbach & Martin, 1980; Vlodavsky &Gospodarowicz, 1981) in a manner analogous to the binding of fibroblastic cells tocollagen by fibronectin (Kleinman, Klebe & Martin, 1981). In vivo, laminin-containing basement membranes are continuous in normal tissue and benigntumours, whereas laminin structures are disrupted and disorganized in malignanttumours (Barsky, Siegal, Jannotta & Liotta, 1983; Albrechtsen et al. 1981; Burtin,Chavanel, Foidart & Andre, 1983). Cultured cells of various origins have been shown

Key words: liver epithelium, cell culture, chemical carcinogenesis, extracellular matrix, cyto-skeleton.

214 J. L. Junker and M. J. Wilson

to produce laminin as part of their extracellular matrix (Foidart et al. 1980; Leivo etal. 1982; Keski-Oja, Gahmberg & Alitalo, 1982). Moreover, a positive correlationbetween cell surface laminin and metastatic potential has been shown for culturedmelanoma and sarcoma cells (Terranova, Liotta, Russo & Martin, 1982; Malinoff,McCoy, Varani & Wicha, 1984). Virally transformed mouse embryo epithelial cells,however, show no change in the extracellular distribution of laminin (or fibronectin)(Keski-Oja et al. 1982) and normal rat kidney (NRK) cells show a loss of both lamininand fibronectin upon transformation (Hayman, Engvall & Ruoslahti, 1981).

We have been studying a liver epithelial cell line TRL1215 (Brown, Wilson &Poirier, 1983; Idoine, Elliott, Wilson & Weisburger, 1976), which, when trans-formed, shows increased cell—substrate adhesion as shown by increased expression ofcellular fibronectin, increased cell spreading, increased numbers of focal contacts andthe development of a prominent actin stress fibre network (Junker et al. 1985). Yet,the transformed cells are capable of anchorage-independent growth and show multi-layered growth in culture (Heine, Wilson & Munoz, 1984; Brown et al. 1983).

Because of the evidence showing laminin to be the preferred substrate for epithelialcell attachment and growth, the TRL1215 cells were examined to evaluate the associa-tions that might exist between laminin expression, growth rate and degree of trans-formation. In vivo, laminin is not a component of the extracellular matrix of maturehepatocytes (Sell & Ruoslahti, 1982; Martinez-Hernandez, 1984), but it does promoteattachment and growth of foetal hepatocytes (Hirata et al. 1983) and may play a roleduring liver regeneration (Carlsson, Engvall, Freeman & Ruoslahti, 1981; Sell &Ruoslahti, 1982). In this present study, laminin staining was abundant in confluentcultures of low passage level, non-tumorigenic cells. Its amount was reduced in post-confluent cultures of control cells at high passage, and it was scarce in post-confluent,flattened areas of completely transformed cells. Thus, in TRL1215 cells, laminin hasa divergent expression from that of fibronectin, which is sparse in low passage controlcultures and abundant in transformed cultures; and fibronectin, but not laminin, isassociated with a high degree of cell spreading and focal contact formation.

MATERIALS AND METHODS

Cell cultureThe TRL1215 cell line used in this study (Brown et al. 1983) is an epithelial cell line derived from

rat liver. The three populations (sublines) that were studied were designated ETC, HPC and LPC,as defined below. Ethionine-treated TRL1215 cells (ETC) became tumorigenic following exposureto 7-5 mM-DL-ethionine in culture medium for 12 weeks followed by growth for a minimum of 22weeks in carcinogen-free medium. Passage level at the time of examination was approximately P60.High passage cells (HPC) were untreated TRL1215 cells at the same passage level as the ETC. Asa control for the effects of passage level, low passage TRL1215 cells (LPC), P15—P20, were alsoexamined. Cells were grown in Williams' Medium D containing 10% foetal bovine serum (Gibco,Grand Island, NY) as previously described (Brown et al. 1983; Heine et al. 1984).

ItnmunofluorescenceThe specificity of the anti-laminin (BRL, Gaithersburg, MD) and anti-fibronectin (Seragen,

Boston, MA) for their respective antigens was confirmed by Western blot analysis (Towbin,

Laminin in transformed epithelial cells 215

Staehelin & Gordon, 1979) using purified laminin from EHS tumour (BRL), 2M-urea extract ofEHS tumour (a gift from Dr Hynda Kleinman, NIDR, Bethesda, MD), bovine plasma fibronectin(BRL) and normal goat serum (Litton Bionetics, Kensington, MD).

Cells were grown on 22 mm X 22 mm glass coverslips for 3-10 days. The cultures generallyreached confluence 4—6 days following seeding. At the time of staining the cells were rinsed withphosphate-buffered saline (PBS), fixed for lOmin in 4% formaldehyde (TEM grade; Tousimis,Rockville, MD) in PBS, incubated in rabbit anti-mouse laminin or rabbit anti-human fibronectin,1:50 dilution, 30-60min, washed for lOmin with PBS, incubated with RITC-conjugated goatanti-rabbit immunoglobulin G (IgG; Cappel, West Chester, PA), l:4fl dilution, for 30min,washed with PBS for 10min, and mounted with Gelvatol (Monsanto, Springfield, MA). Allincubations were at room temperature. Dilutions were made with PBS containing 1 % bovineserum albumin. In confluent cultures, to permit antibody penetration and to reduce backgroundstaining, cells were extracted for 2 min with 0'2% Triton X-100 in PBS, fixed in 4% formal-dehyde, rinsed with PBS, and blocked for 15 min with 10% normal goat serum before stainingwith anti-laminin. As a control, normal rabbit serum, 1:50 dilution, was used in place of anti-laminin. The cells were examined with a Leitz Ortholux microscope fitted for epifluorescence.Fluorescent images were recorded on Tri-X pan film (Kodak, Rochester, NY) and developed withDiafine (Accufine, Chicago, IL).

Reflection-contrast microscopyCells grown on class coverslips were fixed with 2% glutaraldehyde in 0-1 M-sodium cacodylate

and mounted in a Dvorak-Stotler chamber (Nicholson Precision Instruments, Gaithereburg, MD)before examination with a Leitz Ortholux microscope using a NPL 100/1-32 RK objective.

RESULTS

At early times (1-3 days) after seeding, in subconfluent cultures, laminin stainingof attached and flattened cells appeared as small foci or as larger, variably shapedplaques that were scattered on the (apparently basal) cell surface (Figs 1-3). Overall,there was no obvious organization to the distribution of the laminin, but occasionally

Figs 1-3. Laminin staining of subconfluent cultures, 3 days after seeding. Lamininappears as small foci or larger plaques, generally randomly distributed but occasionallyseen organized in a row near the cell periphery (arrows). Beginnings of the more ex-tensive laminin network that develops after confluence are visible (arrowheads). X210.Fig. 1, LPC: Fig. 2, HPC: Fig. 3, ETC.

216 J.L. Junker and M. J. Wilson

a row of foci or a cluster of plaques was seen near the cell periphery. Laminin stainingpatterns were similar in the three sublines (ETC, HPC and LPC).

At confluence, however, an extended laminin network developed, the appearanceof which varied typically for each of the sublines. The number of days after seeding,perse, was not as important as confluence for this development. The most-organized

Figs 4-9. Laminin staining of confluent cultures.Figs4, 5. LPC, 2 days after confluence. Laminin network frequently outlines individual

cells (Fig. 4) or covers multicellular areas with a fine reticulum (arrow) (Fig. 5). X130.Fig. 6. HPC, 2 days after confluence. Laminin network frequently outlines groups of

cells. Small areas of fine reticulum also visible (arrow). X130.Fig. 7. ETC, 2 days after confluence. Disorganized laminin network partially or com-

pletely surrounds cell clusters in flattened areas. Fine reticulum covers all multilayeredareas. X130.

Fig. 8. ETC, parallel culture to that in Fig. 7, 4 days after confluence. Laminin stainingis sparse in flattened areas, but is plentiful in multilayered areas. X130.

Fig. 9. In areas where Triton extraction removes cells, only foci and plaques of lamininremain on the glass of ETC cultures (cf. Figs 1-3). X190.


Figs 10-13. Phase-contrast/reflection-contrast pairs of subconfluent (Figs 10, 11) andconfluent (Figs 12, 13) ETC show how the area of a cell in contact with the glass decreasesmarkedly with confluence. Also note how overlapping of the cells, typical of ETC, makesit difficult to match nuclei in phase-contrast (Fig. 12) with areas of substratum contact(Fig. 13). Numbers indicate probable match-ups. X860.


laminin system was seen in post-confluent LPC. In these cultures, laminin was seenat intercellular junctions, encircling individual cells, thus forming a small-meshnetwork (Fig. 4). Also frequently observed was a fine reticulum or mat that extendedacross large areas of the monolayer (Fig. 5). In cultures of HPC, thick pronouncedstrands of laminin surrounded groups of cells, forming a large-mesh network. Finereticulum, which could sometimes be seen at intersection points of the network,extended across an area of only a few cells (Fig. 6). The pattern of ETC was moredifficult to categorize as the multilayering of cells gave the network a disorganizedappearance. A few days post-confluence, flattened areas of polygonal ETC had anirregular large-mesh network, whereas multilayered areas had a fine reticulum(Fig. 7). Four days post-confluence, when the distinction between flattened andmultilayered areas was clear, little laminin staining was seen in flattened areas(Fig. 8). The contrast in laminin expression for the three sublines can be seen bestby comparing Figs 4, 6 and 8. However, because of the abundant laminin stainingwhere ETC are multilayered, it cannot be said that the transformed cells have lesslaminin in toto.

Unfixed, confluent ETC were very sensitive to Triton treatment and would

1200

1000

800

s

600

400

200

1 2 3 4 5 6Days

Fig. 14. Mean projected cell area of LPC ( • ) and ETC ( • ) . Measurements were madefrom reflection-contrast micrographs and therefore represent the area of each cell that isin contact with the glass substratum. Cell area increases as cells spread initially, butdecreases as the cultures become confluent. Bars indicate standard error of the mean.Average of n = 51 cells.


Figs 15-17. Fibronectin staining of confluent cultures. The fibronectin network is sparseinLPC(Fig. 15), more fully developed in HPC (Fig. 16) and abundant in ETC (Fig. 17).X130.

completely wash off the coverslip if extraction time was increased from 2 to 5 min orif the Triton concentration was increased from 0-2% to 0-5 %. Under these con-ditions there was some, but never a complete, loss of HPC and LPC. In areas wherecells washed away only small laminin spots remained. No evidence of the extendedlaminin network was left behind by LPC, HPC or ETC (Fig. 9). Reflection-contrastmicroscopy showed that the cell area associated with the substratum decreased withconfluence (Figs 10—13). When the area was measured, it was found that the decreasewas greater for ETC than for LPC (Fig. 14). Such a decrease in cell—substratumcontact area, or some qualitative change in focal contacts with confluence could causeETC to be more susceptible to detergent treatment.

In contrast to the laminin distribution, fibronectin staining was sparse in LPCcultures, moderate in HPC cultures and abundant in ETC cultures (Figs 15-17).

DISCUSSION

In the model of the transformed phenotype that is described most often, developedlargely from studies on fibroblastic cells (see Vasiliev & Gelfand, 1981), transforma-tion is accompanied by a decrease in cell-substrate adhesion, a decrease in cell spread-ing, a reduction in actin stress fibres and a loss of cell surface fibronectin. Exceptionsto this model in epithelial systems have been noted (Smith, Riggs & Mosesson, 1979;Bannikove* al. 1982), and the behaviour of TRL1215 cells is in contrast to the modelin that the tumorigenic, ethionine-treated cells (ETC) spread better on the sub-stratum, have more abundant actin stress fibres and cellular fibronectin, and havemore focal contacts per unit projected cell area than the non-tumorigenic, untreatedlow passage cells (LPC) (Junker et al. 1985). In harmony with the model, however,ETC are capable of anchorage-independent growth, form multilayered arrays in


culture and, by transmission electron microscopy, give indication of reduced numbersand extent of intercellular junctions (zonulae adherentes) (Brown et al. 1983; Heineet al. 1984). The two latter characteristics possibly relate to defects in cell—cell ad-hesion and communication. Untreated high passage cells (HPC) fall between LPCand ETC in the expression of a variety of characteristics (Junker et al. 1985).

Therefore, the TRL1215 cells provide an epithelial cell model in which transforma-tion is not dependent on a decrease in cell-substratum adhesion and, specifically, inwhich the loss of contact inhibition of growth and the acquisition of anchorage-independent growth and tumorgenicity is independent of and even inversely cor-related with a decrease in cell-substratum adhesion.

As part of a continuing investigation of this model system, the expression of lamininby TRL1215 cells has been studied because laminin has been shown to promote theattachment of epithelial cells to plastic and type IV collagen substrata (Kleinman etal. 1984; Terranova et al. 1980; Vlodavsky & Gospodarowicz, 1981). The lamininnetwork formed by these cells is similar to that reported for teratocarcinoma,choriocarcinoma and normal rat kidney cells (Leivo et al. 1982; Alitalo, Keski-Oja &Vaheri, 1981; Hayman et al. 1981) and is more extensive than that produced by avariety of normal and transformed cells (Alitalo et al. 1981; Gospodarowicz, Green-burg, Foidart & Savion, 1981; Keski-Oja et al. 1982) including a tumorigenic liverepithelial cell line, ARL-6 (Foidart et al. 1980).

Our results show that the non-transformed liver cells, which have little cellularfibronectin and generally give evidence of less cell-substrate adhesion, display themore organized laminin network, and that the transformed cells, which have anabundant fibronectin matrix and give evidence of a higher degree of cell-substrateadhesion, have substantial areas of scant laminin expression. This summary of theresults applies to monolayered areas of polygonal cells of LPC, HPC and ETC.

The divergent expression of fibronectin and laminin shown in this study is incontrast to those of other studies comparing non-transformed and transformed cells,which showed a coincident immunocytochemical expression of the two proteins.Hayman et al. (1981) reported a loss of both fibronectin and laminin in transformedfibroblastic rat kidney cells, whereas the epithelial cells examined by Keski-Oja et al.(1982) retained expression of both. Differences in the expression of fibronectin andlaminin seen in TRL1215 cells need not be due to differences in synthesis but, rather,could reflect differences in incorporation of these proteins into the extracellularmatrix, or in degradation of the matrix by the cells (Taylor-Papadimitriou, Burchell& Hurst, 1981; Vaheri & Ruoslahti, 1975; Liotta, Garbisa, Tryggvason & Wicha,1980).

Even though laminin promotes cell attachment, the generalization that laminin (incontrast to fibronectin) does not appear to have a cell spreading activity (Kleinmanet al. 1984) is supported by the results of this study. Although there are reports thatlaminin does support a certain amount of cell spreading of primary hepatocytes,fibroblasts and primary epidermal cells (Johansson, Kjellen, Hook & Timpl, 1981;Kennedy et al. 1983; Stenn et al. 1983), details as to the degree or type of spreading(Yates & Izzard, 1981) are lacking. TRL1215 cells clearly show a direct correlation


of fibronectin expression with cell spreading, and to formation of actin stress fibresand focal contacts (Junker et al. 1985). No such correlation is seen with laminin.Therefore, it is doubtful whether laminin plays a major role in the formation of thosestructures, such as focal contacts (Heath & Dunn, 1978; Bereiter-Hahn, Fox &Thorell, 1979), which are associated with maximal cell spreading and cell—substrateadhesion.

The study of TRL1215 cells emphasizes the complexity of the process of carcino-genesis, and particularly shows that each aspect of a cell's interaction with its sur-roundings is subject to alteration by transformation, and that these alterations are notuniform among all cell types and among all transformation schemes. Since there is nosingle morphological characteristic that has been found to be a unique marker fortransformation, investigations into the interrelationships among a variety of charac-teristics are a necessary part of the study of the cell biology of carcinogenesis.

We thank Dr Ursula Heine for helpful suggestions and critical reading of the manuscript, DrMichelle Cottier-Fox for assistance with reflection-contrast microscopy, Dr Hynda Kleinman forthe gift of EHS tumour extract, and Eliana Munoz and Susan Kenney for dark-room work.

REFERENCES

ALBRECHTSEN, R., NIELSEN, M., WEWER, U., ENGVALL, E. & RUOSLAHTI, E. (1981). Basementmembrane changes in breast cancer detected by immunohistochemical staining for laminin.Cancer Res. 41, S076-S081.

ALITALO, K., KESKI-OJA, J. & VAHERI, A. (1981). Extracellular matrix proteins characterizehuman tumor cell lines. Int.jf. Cancer 27, 755-761.

BANNIKOV, G. A., GUELSTEIN, V. I., MONTESANO, R., T INT, I. S., TOMATIS, L., TROYANOV-SKY, S. M. & VASILIEV, J. M. (1982). Cell shape and organization of cytoskeleton and surfacefibronectin in non-tumorigenic and tumorigenic rat liver cultures. J'. Cell Set. 54, 47—67.

BARSKY, S. H., SIEGAL, G. P., JANNOTTA, F. & LIOTTA, L. A. (1983). Loss of basement mem-brane components by invasive tumors but not their benign counterparts. Lab. Invest. 49,140-147.

BEREITER-HAHN, J., Fox, C. F. & THORELL, B. (1979). Quantitative reflection contrast micro-scopy of living cells. J. Cell Biol. 82, 767-779.

BROWN, J. D., WILSON, M. J. &POIRIER, L. A. (1983). Neoplastic conversion of rat liver epithelialcells in culture by ethionine and S-adenosylethionine. Carcinogenesis 4, 173-177.

BURTIN, P., CHAVANEL, G., FOIDART, J.-M. & ANDRE, J. (1983). Alterations of the basementmembrane and connective tissue antigens in human metastatic lymph nodes. Int. J. Cancer 31,719-726.

CARLSSON, R., ENGVALL, E., FREEMAN, A. & RUOSLAHTI, E. (1981). Laminin and fibronectin incell adhesion: Enhanced adhesion of cells from regenerating liver to laminin. Proc. natn. Acad.Sci. U.SA. 78, 2403-2406.

COUCHMAN, J. R., HOOK, M., REES, D. A. & TIMPL, R. (1983). Adhesion, growth, and matrixproduction by fibroblasts on laminin substrates. J. Cell Biol. 96, 177-183.

FOIDART, J.-M., BERMAN, J. J., PAGLIA, L., RENNARD, S., ABE, S., PERANTONI, A. & MARTIN,G. R. (1980). Synthesis of fibronectin, laminin, and several collagens by a liver-derived epithelialcell line. Lab. Invest. 42, 525-531.

GOSPODAROWICZ, D., GREENBURG, G., FOIDART, J.-M. & SAVION, N. (1981). The productionand localization of laminin in cultured vascular and corneal endothelial cells. J . cell. Physiol. 107,171-183.

HAYMAN, E. G., ENGVALL, E. & RUOSLAHTI, E. (1981). Concomitant loss of cell surface fibro-nectin and laminin from transformed rat kidney cells. J . Cell Biol. 88, 352-357.


HEATH, J. P. & DUNN, G. A. (1978). Cell to substratum contacts of chick fibroblasts and theirrelation to the microfilament system. A correlated interference-reflexion and high-voltageelectron-microscope study. J. Cell Set. 29, 197-212.

HEINE, U. I., WILSON, M. J. & MUNOZ, E. F. (1984). Characterization of rat liver cells trans-formed in culture by DL-ethionine. In Vitro 20, 291-301.

HIRATA, K., YOSHIDA, Y., SHIRAMATSU, K., FREEMAN, A. & HAYASAKA, H. (1983). Effects of

laminin, fibronectin and type IV collagen on liver cell cultures. Expl Cell Biol. 51, 121-129.IDOINE, J. B., ELLIOTT, J. M., WILSON, M. J. & WEISBURGER, E. K. (1976). Rat liver cells in

culture: Effect of storage, long-term culture, and transformation on some enzyme levels. In Vitro12, 541-553.

JOHANSSON, S., KJELLEN, L., HOOK, M. & TIMPL, R. (1981). Substrate adhesion of rathepatocytes: A comparison of laminin and fibronectin as attachment proteins. J . Cell Biol. 90,260-264.

JUNKER, J. L., COTTLER-FOX, M., WILSON, M. J., MUNOZ, E. F. & HEINE, U. I. (1985).

Transformation-associated increase of adhesion, cellular fibronectin, and stress fiber develop-ment in a liver epithelial cell line. J. nat. Cancer Inst. 74, 173-183.

KEFALIDES, N. A. (1973). Structure and biosynthesis of basement membrane. Int. Rev. connect.Tiss. Res. 6, 63-104.

KENNEDY, D. W., ROHRBACH, D. H., MARTIN, G. R., MOMOI, T. & YAMADA, K. M. (1983).The adhesive glycoprotein laminin is an agglutinin. J. cell. Physiol. 114, 257-262.

KESKI-OJA, J., GAHMBERG, C. G. & ALITALO, K. (1982). Pericellular matrix and cell surfaceglycoproteins of virus-transformed mouse epithelial cells. Cancer Res. 42, 1147-1153.

KLEINMAN, H. K., KLEBE, R. J. & MARTIN, G. R. (1981). Role of collagenous matrices in theadhesion and growth of cells. J . Cell Biol. 88, 473-485.

KLEINMAN, H. K., MCGARVEY, M. L., HASSELL, J. R., MARTIN, G. R., VAN EVERCOOREN, A.

B. & DUBOIS-DALCQ, M. (1984). The role of laminin in basement membranes and in the growth,adhesion, and differentiation of cells. In 77K; Role of Extracellular Matrix in Development (ed.R. L. Trelstad), pp. 123-143. New York: Alan R. Liss.

LEIVO, I., ALITALO, K., RISTELI, L., VAHERI, A., TIMPL, R. & WARTIOVAARA, J. (1982). Basal

lamina glycoproteins laminin and type IV collagen are assembled into a fine-fibered matrix incultures of a teratocarcinoma-derived endodermal cell line. Expl Cell Res. 137, 15-23.

LIOTTA, L. A., GARBISA, S., TRYGGVASON, K. & WICHA, M. (1980). Correlation of metastaticbehavior with tumor cell degradation of basement membrane collagen. In Tumor Progression(ed. R. G. Crispen), pp. 49-57. Oxford: Elsevier/North-Holland.

MALINOFF, H. L., MCCOY, J. P., VARANI, J. & WICHA, M. S. (1984). Metastatic potential ofmurine fibrosarcoma cells is influenced by cell surface laminin. Int.J. Cancer 33, 651—655.

MARTINEZ-HERNANDEZ, A. (1984). The hepatic extracellular matrix. I. Electron immuno-histochemical studies in normal rat liver. Lab. Invest. 51, 57-74.

SELL, S. & RUOSLAHTI, E. (1982). Expression of fibronectin and laminin in the rat liver after partialhepatectomy, during carcinogenesis, and in transplantable hepatocellular carcinomas. J. natn.Cancer Inst. 69, 1105-1114.

SMITH, H. S., RIGGS, J. L. & MOSESSON, M. W. (1979). Production of fibronectin by humanepithelial cells in culture. Cancer Res. 39, 4138-4144.

STENN, K. S., MADRI, J. A., TINGHITELLA, T . & TERRANOVA, V. P. (1983). Multiple mechan-isms of dissociated epidermal cell spreading. J. Cell Biol. 96, 63-67.

TAYLOR-PAPADIMITRIOU, J., BURCHELL, J. & HURST, J. (1981). Production of fibronectin bynormal and malignant human mammary epithelial cells. Cancer Res. 41, 2491-2500.

TERRANOVA, V. P., LIOTTA, L. A., Russo, R. G. & MARTIN, G. R. (1982). Role of laminin inthe attachment and metastasis of murine tumor cells. Cancer Res. 42, 2265-2269.

TERRANOVA, V. P., ROHRBACH, D. H. & MARTIN, G. R. (1980). Role of laminin in the attachmentof PAM 212 (epithelial) cells to basement membrane collagen. Cell 22, 719-726.

TIMPL, R., MARTIN, G. R., BRUCKNER, P., WICK, G. & WIEDEMANN, H. (1978). Nature of thecollagenous protein in a tumor basement membrane. Eur.J. Biochem. 84, 43-52.

TIMPL, R., ROHDE, H., ROBEY, P. G., RENNARD, S. I., FOIDART, J.-M. & MARTIN, G. R. (1979).

Laminin - A glycoprotein from basement membranes. J . biol. Chem. 254, 9933-9937.TOWBIN, H., STAEHELIN, T. & GORDON, J. (1979). Electrophoretic transfer of proteins from

polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. natn. Acad.Sci. U.SA. 76, 4350-4354.


VAHERI, A. & RUOSLAHTI, E. (1975). Fibroblast surface antigen produced but not retained byvirus-transformed human cells. J. exp. Med. 142, 530-535.

VASILIEV, J. M. & GELFAND, I. M. (1981). Neoplastic and Normal Cells in Culture, pp. 105-154.Cambridge University Press.

VLODAVSKY, I. & GOSPODAROWICZ, D. (1981). Respective roles of laminin and fibronectin inadhesion of human carcinoma and sarcoma cells. Nature, Lond. 289, 304-306.

YATES, J. R. & IZZARD, C. S. (1981). Cell-to-substrate contacts in an adhesion-defective mutantof Balb/c3T3 cells. J. Cell Set. 52, 183-196.

(Received 19 November 1984 -Accepted 1 February 1985)

divergent expression of laminin and fibronectin in … · melanoma and sarcoma cells (terranova,...

Documents