Glycoconjugate and adhesion molecule changes during the cell cycle in a human embryonic epithelial cell line

Glycoconjugate and adhesion molecule changes during the cell cycle in a human embryonic epithelial cell line

Biology of the Cell (1998) 90, 155-159 o Elsevier, Paris 155 Original article Glycoconjugate and adhesion molecule changes during the cell cycle in...

2MB Sizes 0 Downloads 57 Views

Biology of the Cell (1998) 90, 155-159 o Elsevier, Paris

155

Original article

Glycoconjugate and adhesion molecule changes during the cell cycle in a human embryonic epithelial cell line Anna Maria Bolognani Fantin a, Antonella Franchini a, Roberta Malgara b, Barbara Rebecchi a and Anna Maria Fuhrman Conti b aDipartimento di Biologia Animale, Universit2 di Modena, via Berengario, 14, 41100 Modena; bDipartimento di Biologia e Genetica per le Scienze MedicheJniversit6 di Milano, Milan, Italy

The changes in the expression of glycoconjugates and adhesion molecules were studied by selective lectin binding and immunocytochemical reactions in a human embryonic epithelial cell line (EUE cells), synchronized in the cell cycle phases. The results can be summarized as follows: most of the tested lectins display a more diffuse binding for the cytoplasm in G, than S and G, phases; in the S and particularly in G, phases the cytoplasm glycoconjugates are rearranged around the nucleus; cells in mitosis always show a strong binding towards all tested lectins. Cellular fibronectin and its receptor PI integrin are well expressed in G,, but the strongest reaction is observed in the S phase. The immunoreactions for laminin and uvomorulin (L-CAM) are poorly positive in all cell cycle phases. The immunocytochemical reaction for heparan sulfate is positive, with a stronger reaction in S and G, than in G,; on the contrary a diffuse staining with the anti-dermatan sulfate proteoglycan antibody appears unchanged during the cell cycle. (0 Elsevier, Paris) cell cycle / glycoconjugates / adhesion molecules /transformed

INTRODUCTION Cell membrane carbohydrate moieties are involved in cell-cell interactions and are thought to be important in a variety of cell functions including cell-cell and cell-substrate adhesion, cell recognition and cell maturation (Damjanov, 1987; Monsigny et al, 1988). It has been shown that the surface of some monolayer cells undergoes distinct changes during the cell cycle (Rubin and Everhart, 1973; Porter et al, 1973; Hale et al, 1975). Cell-cycle related morphological changes have been described in feline lymphoid cells (Toth, 1981). The cell-cycle timing of the human cell line EUE cells was characterized in kinetic terms (Zippel et al, 1982) on the basis of autoradiography, a standard chemical determination of DNA and Fried’s analysis of DNA distributions. The authors also analyzed celI size distributions under various growth conditions and found that they may furnish information on the Cell cycle and glycoconjugates in EUE cells

cells / immunocytochemistry

proliferative condition of EUE cells (Zippel et ~2, 1982). The expression of a proliferating cell nuclear antigen (PCNA/cyclin), a 36 kDa protein that is upregulated in activated proliferating cells from a variety of tissues and species (Mathews et ~2, 1984), was studied in different cell systems, including EUE synchronized cells (Pellicciari ef al, 1989,199l). Recent advances in the understanding of complex glycoconjugates are based primarily on the development of technical approaches that have allowed bet-

ter characterization, purification and structural analysis of carbohydrate components. Among these newer and improved methods and approaches monoclonal antibodies and lectins have received special attention as the major analytic tools for the study of both soluble and cellular glycoconjugates. In a previous paper we elucidated the nature of the cell coat of EUE cells by histochemical and ultracytochemical procedures and studied the modifications of plasma membrane and cytoplasmic glycoconjugates following the adaptation to hyperBolognani Fantin et al

156

Biology of the Cell I19981 90, 15.5159

osmotic medium (Bolognani Fantin et nl, 1989). Xn the present paper we present a study concerning the changes in glycoconjugates and adhesion molecule expression in the same cell line (EUE, Epiteliali Umane Embrionali) during the cell cycle phases.

MATERIALS AMD &ETWQDS EUE cells were cultured in flasks in Eagle’s medium suyplemented with 12% newborn calf serum. T’o evaluate the progression through the cell cycle phases, the cells lvere synchronized by selective detachment of mitotic cells from asynchronous populations treated with 0.02 mg/mL colcemid (Gibco Labs, USA] for 16 h in the culture medium supplemented with 18% newborn calf serum. Mitotic cells, detached by gently shaking the Petri dishes, were harvested, washed once in cold, serum-free medium to remove colcemid, and re-seeded in Petri dishes on coverslips with the appropriate, pre-warmed medium supplemented with 12% newborn calf serum. The coverslips were removed from culture medium after 3, 14 and 17 h after reseeding (G,, S and G: ph.ases, respectively) and fixed in 496 para-formaldehyde in phosphate buffered saline solution (PBS) pH 7.4 for 15 min. Methods described below were carried out.

Binding with horseradish labeled &tins

poroxidase (HRP)

l~he lectins used were: DBA (Dolickos bz@rusi; SBA (G/y-. ciw maxj; PNA (Arackis kypoguea); WGA (Tritictlm vrtlpreJ; C:on,!I (Canavalia ensiformis), RCA I (Ricinus cornmu. fr~,si. All lectins were obtained from Sigma (Sigma Chemical Co, USA). The optimal concentration was as follows: 0.01 mg/mL WGA; 0.02 mg/mL SBA and DBA; 0.05 mg/mL PNA, RCA I and Con A. Endogenous peroxidase activity was inhibited by incubating the cells in 0.3% hydrogen peroxide/methanol for 30 min before lectin binding. Cells were then washed with PBS and treated at room temperature according to Faraggiana et ui (1982). Negative controls for lectin-sugar binding are provided by treating a cell monolayer with a lectin solution which has been previously incubated with the sugar to be demonstrated. Lectin binding was also performed after an appropriate previous digestion with neuraminidase (type V from Closfridium perfringens, Sigma), a-fucosidase (from bovine epididymis, Sigma) and pgalactosidase (gradeV1 from Esckerickia coli, Sigma) as previously described (Bolognani Fantin et ai, 1989). Controls for enzymatic digestions were performed by incubating the cells in enzyme-free buffer solution for the same time and incubation conditions.

Immunocytochemical

procedures

Celis on coverslips were fixed in 4% para-formaldehyde in PBS for 5 min. After washing in PBS the cells were subjected to indirect immunofluorescence procedure to demonstrate proteoglycans using the following primary anti-, bodies: anti-heparan sulfate (Seikagatu Co, Japan) monoclonal antibody (mAb) (1 :lOO); anti-human dermatan sulfate proteoglycan (Seikagatu Co) mAb (1:lOO):

Cell cycie and glycocoqugates III E&JEcells

anti-proteoglycan adi-6S mAb (Miles Scientific, USA/ (1:50) and anti-vinculin (Boehringer Mannheim, Ger. many) mAb (15). The cells were incubated for 60 min at 37°C rinsed in PBS, incubated in fluorescein isothiocyanate (FIX)-lab&d-anti-mouse IgG (Boehringer Mannheim) (1:30), rinsed in PBS and mounted in 50% glycerol in PBS. The cells were observed and photographed with a .Zeiss Photomikroskop III equipped with fluorescence illumination (band pass filter 485/20 and suppression filter 520 for FITC). An immunocytochemical procedure using avidin-biotin-peroxidase complex technique (Hsu ct al, 1981) was also performed in order to test antichicken fibronectin polyclonal antibody- (pAb) (Chem.icon, USA) (1:500), anti-human fibronectin cell-binding domain mAb (Chemicon) (l:lOO), anti-&-chicken integrin mAb (Sigma) (1:50), anti-human laminin mAb (Sigma) (1:20), and anti-uvomorulin (L-CAM) mAb (Sigmas\ (1:500). The celis were incubated with primary antibody solutions overnight at 4°C The controls of immunocvtochemical reactions were performed by incubating-the cells in a primary antibody free medium.

tmage processing attd analysis The binding patterns of some lectin-HRP conjugates (WGA, Con A and SBA), again after specific glycosidase digestions, and related controls were acquired by a computer-assisted microscope equipped with a CCD camerd (Sony, USA) and analyzed using NIH Image (version 1.59) analysis software. The mean density of staining of 30 cells for each cell cycle phase was measured, and statistical analysis was performed by the Student’s t-test comparing the mean values between the different cell phases.

The results obtained with lectin binding showed a similar behavjour of glycoconjugate components over all phases of cell cycle. The reaction was frequently present both on the plasma membrane-and cytoplasm, but absent after negative controls. Most of the lectins exhibited a more diffuse binding in the 6, phase (figs la, 2a) in comparison to the S and G, phases (figs lb, lc, 2b, 2~). Indeed, a rearrangement of cytoplasmic glycoconjugates in a perinu-. clear triangular area was often observed in the S and particularly in G, phase (fig lc, 2~). Image analysis data performed after ConA, WGA (fig 6) and SBA bindings (specific + aspecific) showed significantly different mean densities between G,, S and G2 (P < O.OOl).The aspecific binding checked on negative

controls

corresponds

to a mean

density

value of 10. Enzymatic digestions affected the lectin binding as previously demonstrated in EUE generai population (BoIognani Fanfzin et al, 1989). In figure 6 the effect of neuraminidase digestion in G, cells was reported as an example. The binding of WGA after neuraminidase treatment in G, ceils was down to about 10 ~100. The cells during mitosis appeared BolognaniFantin et ai

Biology of the Cell (1998) 90, 155-159

lb

2b

3a

2c

3b

Figs l-3.

1. SBA-HRP staining (la, G, phase; lb, S phase; lc, G, phase). G, phase showed the more diffuse binding and G, phase the glycoconjugate rearrangement in perinuclear area. Bar = 10 pm. 2. DBA-HRP staining (2a, G, phase; 2b, S phase; 2c, G, phase). The DBA binding displayed a picture similar to that of SBA binding. Bar = 10 pm. 3. DBA-HRP (a) and WGA-HRP stainings (b) of mitosis phase. The cells appeared round in shape. Bar = 10 pm.

round in shape, without focal contacts for the substrate, and they displayed a very strong binding for all tested lectins (fig 3a, b). The obtained results concern the total (specific + aspecific) lectin bindings as demonstrated by image analysis data. Regarding immunoreactivity for cellular fibronectin, a reaction in dot-like areas on the plasma membrane was observed, with the strongest reactions in the S phase especially as compared to G, Cell cycle and glycoconjugates

in EUE cells

(fig 4a, b, c). The observed nuclear staining should be considered not specific because it is also present in negative control reaction. The same result was obtained using monoclonal and polyclonal antibodies, and a similar picture was found using anti-& integrin receptor for fibronectin. A low immunoreactivity, without differences over the cell cycle phases, was detected for other adhesion molecules (laminin and L-CAM) (not shown). Bolognani Fantin et a/

Biology of the Cell [ 1998) 90, 155-i 59

158

4a

4b

4c

Figs 4, 5. 4. Immunocytochemicat staining with anti-fibronectin pAb (4a, 6, phase; 4b, S phase; 4c, G, phase). The reac. tion in dot-like areas was strong in the S phase. Bar = 10 pm. 5. lmmunocytochemical staining with anti-heparan sulfate mAb (5a, G, phase; 5b, S phase; 5c, G, phase). S and G, phases displayed an increased response. Bar = 10 pm.

100

SO

B e B

80

I B

40

$ *5 20

0 NourQl

S

Q

Fig 6. WGA binding analysis, Each bar represents the mean density value f SD from 30 ceils for each cell cycle phase (specific + aspecific binding). Data were compared using Student’s t-test and significant differences were observed. NeurG,, WGA binding after neuraminidase digestion of cells in G, phase. Cell cycle and glycoconjugates

in EUE cells

With the anti-heparan sulfate antibody, we obtained a weak reaction in G,, an evident response in G, and particularly in the S phase (fig 5a, b, c). In all cases the reaction appeared as small dot bodies on the cell surface and in the nucleus. In a previous paper on characterization of GAG in EUE general popuIation (Bolognani Fantin et al, 1989) we observed that frequently the EUE cell nucleus gave positive results for lectin binding. The previous and present similar results with antibody against heparan sulfate should be related to the presence of nuclear glycoproteins and GAG observed in a large gumber of higher eukaryotic cells (for a review see Stein ef al, 1981). The presence of a polysaccharide fraction associated with chromatin was also showed in several transformed eelis (Stein et al, 1981). The immunoreaction with dermatan. sulfate antibody was poor, diffused throughout the cell and unchanged during the ceil cycle (not shown). Negative result was always observed with anti-prc+ teoglycan Adiantibody. The results demonstrated that the glycoconjugates present in EL&-cells undergo quarititative Bolognani Fafltin et ai

Biology of the Cell (1998) 90, 155-l 59

fluctuations and rearrangements during the cell cycle, particularly in the cytoplasm. Similar changes were observed for expressed adhesion. molecules, particularly fibronectin, and for vinculin-containing focal contacts, visualized by a Leitz Diavert inverted microscope equipped with epifluorescence illumination and interference reflection microscopy optics (IRM) (Bolognani et ~2, 1994). Focal contacts and fibronectin were more present in the S phase, where the cell is particularly increased in size, indicating a greater need to anchor to the substratum and to generate signals via the adhesive receptors that influence cell behaviour. The signals transduced by integrins or by other adhesion receptors play an important role in the regulation of many fundamental biological processes including growth, differentiation and apoptosis (Rosales et ~2, 1995; Juliano, 1996). The pattern of lectin bindings and immunocytochemical reactions for adhesion molecules displayed a different picture so that we can assume that the distribution of reactive molecules is quite different. Our results on focal contact and fibronectin are not in agreement with the literature data on transformed fibroblastic cells and transformed epithelial cells, which appear deficient both in cell-cell adhesion and cell-substratum adhesion molecules (Bannikov ef al, 1982). Lectins are molecules particularly suitable for specifically recognizing glycoconjugates. They were used as probes for the determination of changes in cell surface glycoconjugates of keratinocytes in various stages of differentiation and growth (Ku and Bernstein, 1988; Suter et al, 1991). In EUE synchronized cell cultures an evident redistribution of lectin binding receptors occurs in order to modify the cell surface cues following the different metabolic needs of the cell. A similar redistribution of lectin receptors also occurs on the surface of fibroblasts when the cells change from a spherical, nonattached state, to an adherent state (Swaisgood and Schindler, 1989).

ACKNOWLEDGMENTS The authors are grateful to Marinella Volontb for technical assistencein preparing cell cultures. Researchsupported by a 60% grant from MURST (Italy) to AMBF and AMFC.

REFERENCES Bannikov GA, Guelstein VI, Montesano R, Tint IS, TomatisL, Troyanovski SM and Vasiliev JM (1982) Cell shape and organi-

Cell cycle and glycoconjugates in EUE cells

159

zation of cytoskeleton and surface fibronectin in non-tumori-

genieandtumorigenicrat liver cultures.I CeNSci54,47-67 Bolognani Fantin AM, Menghi G, Franchini A, Bondi AM, Accili D and Fuhrman Conti AM (1989) Characterization of glycoconjugates in an embryonic human epithelial line and changes consequent to adaptation to a hyperosmotic medium. Histochem J 21,79-88 Bolognani L, Bolognani Fantin AM, Franchini A, Volpi N, Venturelli T and Fuhrman Conti AM (1994) Effects of lowpower 632 nm radiation (HeNe laser) on a human cell line: influence on adenylnucleotides and cytoskeletal structures. 1 Photo&m Photobiol B: Biol26,257-264 Damjanov I (1987) Biology of disease. Lectin cytochemistry and histochemistry. Lab Invest 57,5-20 Faraggiana T, Malchiodi F, Prado A and Churg J (1982) Lectinperoxidase conjugate reactivity in normal human kidney. 1 Histochem Cytochem 30,451-458 Hale AH, Winkelhake JL and Weber M (1975) Cell surface changes and Rous sarcoma virus gene expression in synchronized cells. J Cell Biol64,398-407 Hsu SM, Raine L and Fanger H (1981) Use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 29,577-580 Juliano R (1996) Cooperation between soluble factors and integrin-mediated cell anchorage in the control of cell growth and differentiation. BioEssays 18,911-917 Ku WW and Bernstein IA (1988) Lectin binding as a probe of proliferative and differentiative phases in primary monolayer cultures of cutaneous keratinocytes. Exp Cell Res 175,29&316 Mathews MB, Bernstein RM, Franza BR and Garrels JI (1984) Identity of the proliferating cell nuclear antigen and cyclin. Nature 309,374-376 Monsigny M, Roche AC, Kieda C, Midoux P and Obmovitch A (1988) Characterization and biological implications of membrane lectins in tumor lymphoid and myeloid cells. Biochimie 70,1633 Pellicciari C, Mazzini G, Fuhrman Conti AM, De Grada L and Manfredi Romanini MG (1989) Effect of hypertonic medium on human cell growth: III. Changes in cell kinetics of EUE cells. Cell Biol Znt Rep 13,345-356 Pellicciari C, Danova M, Giordano M, Fuhrman Conti AM, Mazzini G, Wang E, Ronchetti E, Riccardi A and Manfredi Romanini MG (1991) Expression of cell cycle related proteins-proliferating cell nuclear antigen (PCNA) and statinduring adaptation and de-adaptation of EUE cells to a hypertonic medium. Cell ProZif24,469-479 Porter KR, Prescott D and Frye J (1973) Changes in surface morphology of CHO cells during the cell cycle. ] Cell Biol 57,815-836 Rosales C, O’Brien, V, Kornberg L and Juliano RL (1995) Signal transduction by cell adhesion receptors. Biochim Biophys Acta 1242,77-98 Rubin RW and Everhart LP (1973) The effect of cell-to-cell contact on the surface morphology of CHO cells. J Cell BioZ57,837-844 Suter MM, Augustin-Voss HG, Pantano DM, Flanders JA and Varvayanis M (1991) Differentiation-dependent expression of lectin binding sites on normal and neoplastic keratinocytes in vivo and in vitro. ] Histochem Cytochem 39,1103-1112 Swaisgood M and Schindler M (1989) Lateral diffusion of lectin receptors in flbroblast membranes as a function of cell shape. Exp CeZIRes 180,515-528 Toth SR (1981) Cell cycle-related morphological changes of feline lymphoid cells as revealed by electron microscopy. Cancer Res 41,4727-4736 Zippel R, Martegani E, Vanoni M, Mazzini G and Alberghina L (1982) Cell cycle analysis in a human cell line (EUE cells). Cytomety 2,426-430 Received 28 April 1997; accepted 23 February 1998

Bolognani Fantin et al