Regulation of concanavalin A agglutination by the extracellular matrix

The agglutinability of rat C₆ glioma cells by concanavalin A (Con A) depends upon cell density. From sparse density to near confluency agglutinability increases as cell density rises. Both the half-maximal concentration and the maximum amplitude of agglutination by Con A are functions of cell density, but are separate cell parameters differing in the extent to which they are affected by density and the point at which they become insensitive to further density increases. Both trypsin and EDTA reduce cell agglutinability. The similarity in recovery kinetics between low density cells and cells dissociated with EDTA or trypsin suggests that low density cells may lose the same surface agglutination component(s) removed by trypsin and EDTA. Density-dependent regulation of Con A agglutinability is anchorage dependent; cells grown in suspension display no such phenomenon. The cooperative cell regulation of agglutinability is mediated by the extracellular matrix, or micro-exudate. The matrix contains two activities: low density cultures produce a matrix inhibitor of Con A agglutinability, while high density cultures produce a matrix promotor.     

Effect of metabolic state on agglutination of human erythrocytes by concanavalin A.

Intact freshly drawn or stored human erythrocytes, which show little agglutination by concanavalin A, become agglutinable by this lectin in the presence of adenosine. alpha-Methylglucose (10 mM) completely inhibits this agglutination. The concanavalin A agglutination shows no sensitivity to vinblastine or cytochalasin B. Resealed membranes preparaed with ATP in lysing and resealing medium give modest agglutinability, while the presence of adenosine in both the lysing and the resealing medium results in a substantial agglutinability of the resealed membranes. Mild trypsin treatment of the erythrocytes causes an enhanced sensitivity to adenosine activation of the concanavalin A agglutination, while extensive trypsin treatment produced highly agglutinable erythrocytes that shown no response to the presence of adenosine in the lectin solution. The extensively treated erythrocytes also show concanavalin A agglutination at temperatures below 37 degrees C, under conditions in which intact or moderately treated erythrocytes do not agglutinate, with or without adenosine present. Results suggest that the adenosine activation of concanavalin A agglutination of intact human erythrocytes is mediated through a metabolic conversion of adenosine to a rapidly turned over metabolite which participates directly in the activation of agglutination. The agglutinability does not appear to depend on whole cell ATP levels, but may involve a particular pool of ATP. The effect of variation of cellular metabolic state and the response of particular systems involved in lectin-mediated agglutinability to cellular metabolism seem to be worth consideration in explaining the frequently large differences in agglutinability of und in cells in different biological states, such as those encountered in normal and transformed cells.     

Glycophorin A interferes in the agglutination of human erythrocytes by concanavalin A. Explanation of the requirement for enzymic predigestion.

Human erythrocytes become agglutinable with concanavalin A (Con A) after treatment with various proteinases or neuraminidase. The extent of agglutinability achieved with different enzymes is, however, different: Pronase, papain, trypsin, neuraminidase and chymotrypsin enhance the agglutinability in decreasing order, the last being barely effective. The actions of the enzymes on band 3, the Con A receptor, do not correlate with their abilities to increase the agglutinability: Pronase, papain and chymotrypsin cleave the protein, but not trypsin or neuraminidase. No significant differences are found in the number of Con A-binding sites or the affinities for the lectin between the normal and trypsin- or Pronase-treated cells. Thus the receptor does not seem to play a role in determining the Con A-agglutinability of erythrocytes. On the other hand, the cleavage of glycophorins, especially glycophorin A, and the release of sialic acid (in the peptide-bound form) are well-correlated with the enhancement in agglutination after the action of proteinases. The release of sialic acid by graded neuraminidase digestion and the increase in Con A-agglutinability show a correlation coefficient of 0.88. The major inhibitory role of glycophorin A in the process is indicated by the agglutination of En(a) heterozygous erythrocytes; the cells, known to bear about 50% glycophorin A molecules in their membrane, are agglutinated approximately half as well without proteolysis as are the trypsin-treated cells. Possible mechanisms by which glycophorin A could affect Con A-mediated agglutination are discussed.     

Studies on the iodinated surface membrane proteins and concanavalin A agglutination of transformed Syrian hamster cells

Chemically transformed Syrian hamster cells exhibit marked agglutination in the presence of the plant lectin, concanavalin A. In this report, we describe conditions which can alter this concanavalin A agglutinability, and compare the surface proteins from transformed cells which express different degrees of agglutinability. Lactoperoxidase-catalyzed iodination of tertiary Syrian hamster cells reveals the major iodinatable protein to be approximately 220 000 daltons. The transformed Syrian hamster cells do not contain this protein in an iodinatable form. Analyses of the transformed cells grown under conditions which decrease the concanavalin A agglutinability do not demonstrate any iodination of the 220 000 mol. wt. protein. These results depict the effects of growth and dibutyryl cyclic AMP on the iodinatable cell surface proteins of transformed cells and indicate that the absence of the 1–220 000 mol. wt. protein is probably not a major determinant of concanavalin A agglutination.     

Studies on the iodinated surface membrane proteins and concanavalin A agglutination of transformed Syrian hamster cells.

Chemically transformed Syrian hamster cells exhibit marked agglutination in the presence of the plant lectin, concanavalin A. In this report, we describe conditions which can alter this concanavalin A agglutinability, and compare the surface proteins from transformed cells which express different degrees of agglutinability. Lactoperoxidase-catalyzed iodination of tertiary Syrian hamster cells reveals the major iodinatable protein to be approximately 220 000 daltons. The transformed Syrian hamster cells do not contain this protein in an iodinatable form. Analyses of the transformed cells grown under conditions which decrease the concanavalin A agglutinability do not demonstrate any iodination of the 220 000 mol. wt. protein. These results depict the effects of growth and dibutyryl cyclic AMP on the iodinatable cell surface proteins of transformed cells and indicate that the absence of the I-220 000 mol. wt. protein is probably not a major determinant of concanavalin A agglutination.     

Effect of metabolic state on agglutination of human erythrocytes by concanavalin

Intact freshly drawn or stored human erythrocytes, which show little agglutination by concanavalin A, become agglutinable by this lectin in the presence of adenosine. α-Methylglucose (10 mM) completely inhibits this agglutination. The concanavalin A agglutination shows no sensitivity to vinblastine or cytochalasin B.Resealed membranes prepared with ATP in lysing and resealing medium give modest agglutinability, while the presence of adenosine in both the lysing and the resealing medium results in a substantial agglutinability of the resealed membranes.Mild trypsin treatment of the erythrocytes causes an enhanced sensitivity to adenosine activation of the concanavalin A agglutination, while extensive trypsin treatment produced highly agglutinable erythrocytes that show no response to the presence of adenosine in the lectin solution. The extensively treated erythrocytes also show concanavalin A agglutination at temperatures below 37°C, under conditions in which intact or moderately treated erythrocytes do not agglutinate, with or without adenosine present.Results suggest that the adenosine activation of concanavalin A agglutination of intact human erythrocytes is mediated through a metabolic conversion of adenosine to a rapidly turned over metabolite which participates directly in the activation of agglutination. The agglutinability does not appear to depend on whole cell ATP levels, but may involve a particular pool of ATP.The effect of variation of cellular metabolic state and the response of particular systems involved in lectin-mediated agglutinability to cellular metabolism seem to be worth consideration in explaining the frequently large differences in agglutinability of und in cells indifferent biological states, such as those encountered in normal and transformed cells.     

Reaction of lectins with human erythrocytes : III. Surface charge density and agglutination

A number of commercially available enzymes were used to modify the cell surface of human erythrocytes to varying degrees. In protease-treated erythrocytes the decrease in surface charge (determined by cell electrophoresis or analysis of sialic acid content) correlates with an increase in agglutinability with concanavalin A (ConA) and wheat germ agglutinin (WGA). On the other hand, treatment with neuraminidase leads to very large decrease in surface charge with only an intermediate increase in agglutinability with both lectins. Subsequent protease treatment of these cells enhances their agglutinability appreciably without further altering their surface charge. It is concluded that the increased agglutinability following protease treatment is due both to a decrease in the net negative charge and a removal of peptides and glycopeptides from the cell surface that may sterically hinder the agglutination reaction.     

Aprotinin induces surface changes in malignant cells in culture: a possible mode of antitumour action.

Treatment of TRES cells with aprotinin (Trasylol) produced marked changes in their surface properties. There was a concentration-dependent increase in the surface charge density of the cells. The agglutinability of the cells was inhibited by the antiproteinase. The cells became less adhesive to concanavalin A-linked plates. The adhesion of treated TRES cells appeared to be mediated by lectin receptors with lower specificity for concanavalin A. These observations and an analysis of the kinetics of adhesion have suggested the possibility that a new class of concanavalin A receptors appears on the surface of aprotinin-treated TRES cells, which might increase immunogenicity of the cells.     

Dynamics of changes in the surface of normal cultivated rat cells after a single exposure to the carcinogen 7,12-dimethylbenz(a)anthracene]

Changes in the cell surface after a single treatment with 7,12-dimethylbenz(a)anthracene (DMBA) of newborn rat carcass in cell culture have been studied by means of the agglutination reaction with concanavalin A. DMBA was shown to cause alterations in the cell surface. At 0.5 mkg/ml of DMBA, the difference in agglutinability of treated and untreated cells persists for 30 days. At 0.1 mkg/ml of DMBA, the agglutinability of drug-treated and control cells was similar on the 4th day after removal of carcinogen. A prolonged culturing of control cells results in an increased agglutinability of cells with concanavalin A, and in 2.5 months it becomes indistinguishable from the agglutinability level of tumor cells with concanavalin A. In 5 months, drastic karyotypic changes are registered in control cultures.     

Interactions of concanavalin A with chick embryo fibroblasts transformed by Rous sarcoma virus. Study with an RSV mutant thermosensitive for transformation.

The interactions between concanavalin A and chick embryo fibroblasts, normal and infected with Rous sarcoma virus (RSV-BH) or its thermosensitive mutant RSV-BH-Ta, have been studied. Normal chick embryo cells and RSV-BH transformed cells showed at 4 and 25 degrees C a similar number of concanavalin A receptors per cell. Analysis of the binding data by the Scatchard relation showed that apparent changes in binding as a function of temperature are due to the thermodynamic properties of the process and not to endocytosis. The lectin receptors on the cell surface of normal and RSV-BH infected cells showed homogeneity in their binding properties. Chick cells infected with RSV-BH-Ta showed a lectin binding behavior that was dependent on the temperature at which the cells were grown. At the permissive temperature for transformation (37 degrees C), the binding process was similar to that observed for normal and RSV-BH infected cells. At the nonpermissive temperature (41 degrees C), the cells showed at least two sets of concanavalin A receptors. The new set of receptors on the cell surface had a lower lectin affinity than those observed in the same cells at 37 degrees C. Chick cells infected with RSV-BH showed an enhanced agglutinability by concanavalin A, as compared with normal cells. Cells infected with RSV-BH-Ta showed a reversal of the correlation between increased concanavalin A agglutinability and the transformed state. At the permissive temperature for transformation, the cells were not agglutinable, whereas at the nonpermissive temperature they presented agglutinability indexes as high as those observed with RSV-BH infected cells. This enhanced agglutinability observed with cells maintained at the nonpermissive temperature for transformation may be related to the new set of low affinity receptors present at 41 degrees C.     

Relationship between cytoagglutination and saturation density of cell growth

The agglutination with concanavalin A and wheat germ agglutinin of the established malignant cells, HEp 2, KB, HeLa, TDB-3, HTC and RV 3T3, and of the putatively normal cells, BHK 21, 3T3 and Wi-38 was examined as a function of their saturation densities in culture. A positive correlation between the saturation density of the cell lines and the capacity to agglutinate was demonstrated. Incubation for 15 minutes with 1.25 mg/ml of trypsin converted non-agglutinating and poorly agglutinating cells into agglutinable ones, while leaving the highly agglutinating lines largely unchanged. The magnitude of change in agglutination after trypsin treatment correlated inversely with saturation density. Although the extent of agglutination varied with the saturation density, the agglutinability of a particular line remained relatively unchanged at different cell densities.     

Differential lectin-mediated agglutinabilities of the embryonic and the first extraembryonic cell line of the early chick embryo

Summary The lectin-mediated agglutinability of cells dissociated from different areas of the gastrulating chick embryo was investigated. Differences in agglutinability were quantified by using a Coulter counter. Cells from the area pellucida (AP) and those from the endoderm of the area opaca (AOEn) are agglutinated by Concanavalin A (Con A), wheat germ agglutinin (WGA) andRicinus communis agglutinin (RCA). In cells from both areas the greatest agglutination response is obtained with RCA. Trypsinization of AOEn cells enhances their agglutinability with Con A, WGA and RCA. The lectin-induced agglutinability of cells from the area pellucida is similar in EDTA-dissociated and trypsinized cells.Cells from the AP are significantly more agglutinable with Con A than those of the AOEn regardless whether the former are obtained by trypsinization or dissociation with EDTA. The higher agglutinability of cells of the area pellucida with Con A, as well as the differential enhancement by trypsin of the agglutinability of AOEn cells with Con A, WGA, and RCA may reflect a difference in the cell surface glycoreceptors between the cells of the are pellucida (predominantly embryonic) and the first extraembryonic (AOEn) cell line. These cells have been shown to sort out from each other at the earliest stages of development.     

Lack of correlation between agglutinability,the surface distribution of con A and post-confluence inhibition of cell division in ten cell lines

Agglutinability by concanavalin A, distribution of surface-bound concanavalin A, and maximal cell density in monolayer culture were examined under similar conditions in parallel cultures of ten established cell lines. The degree of agglutinability of the cell lines did not correlate with the presence or absence of patching of concanavalin A bound to the cell surface, as determined with a hemocyanin marker. Agglutinability was also not always correlated with the loss of post-confluence inhibition of cell division. Two clones of mouse 3T3 fibroblasts that maintained post-confluence inhibition of cell division and low agglutinability differed substantially with respect to the surface distribution of concanavalin A. Patching of concanavalin A binding sites is neither necessary nor sufficient to explain differences in agglutinability between cell lines.     

Interactions of concanavalin A with chick embryo fibroblasts transformed by rous sarcoma virus Study with an RSV mutant thermosensitive for transformation

The interactions between concanvalin A and chick embryo fibroblasts, normal and infected with Rous sarcoma virus (RSV-BH) or its thermosensitive mutant RSV-BH-Ta, have been studied. Normal chick embryo cells and RSV-BH transformed cells showed at 4 and 25 °C a similar number of concanavalin A receptors per cell. Analysis of the binding data by the Scatchard relation showed that apparent changes in binding as a function of temperature are due to the thermodynamic properties of the process and and not to endocytosis. The lectin receptors on the cell surface of normal and RSV-BH infected cells showed homogeneity in their binding properties. Chick cells infected with RSV-BH-Ta showed a lectin binding behavior that was dependent on the temperature at which the cells were grown. At the permissive temperature for transformation (37 °C), the binding process was similar to that observed for normal and RSV-BH infected cells. At the nonpermissive temperature (41 °C), the cells showed at least two sets of concanavalin A receptors. The new set of receptors on the cell surface had a lower lectin affinity than those observed in the same cells at 37 °C.Chick cells infected with RSV-BH showed an enhanced agglutinability by concanavalin A, as compared with normal cells. Cells infected with RSV-BH-Ta showed a reversal of the correlation between increased concanavalin A agglutinability and the transformed state. At the permissive temperature for transformation, the cells were not agglutinable, whereas at the nonpermissive temperature they presented agglutinability indexes as high as those observed with RSV-BH infected cells. This enhanced agglutinability observed with cells maintained at the nonpermissive temperature for transformation may be related to the new set of low affinity receptors present at 41 °C.     

Role of cell membrane galactosyltransferase in concanavalin A agglutination of erythrocytes.

It has been previously observed that rabbit erythrocyte cell surface galactosyltransferase appears to play a role in concanavalin A agglutination of these erythrocytes (Podolsky et al., 1974). Further, a correlation between the occurrence or level of cell surface galactosyltransferase and concanavalin A agglutinability of other cell types has also been observed. The mechanism by which rabbit erythrocyte galactosyltransferase participates in concanavalin A agglutination has now been further defined. The enzyme was solubilized and purified. Characterization of the enzyme properties has shown them to be similar to those reported for other purified galactosyltransferases. Amino acid and carbohydrate analysis showed a high asparagine content and the presence of D-mannose. Specific alpha-mannosidase treatment of the enzyme showed that some of these D-mannose residues were terminal sugars. The purified enzyme also conferred concanavalin A agglutinability to non-agglutinable human erythrocytes. However, the ability to confer concanavalin A agglutinability was unrelated to the enzyme activity per se (as measured with fetuin acceptor) but appeared to be entirely dependent on the presence of terminal alpha-linked D-mannosyl residues in the enzyme structure. These findings suggest that the presence of terminal alpha-mannosidyl residues on cell surface glycoproteins such as galactosyltransferase may be the determining factor in agglutination of cells by concanavalin A.     

Agglutination of Sindbis Virus and of Cells Infected with Sindbis Virus by Plant Lectins

We have examined the agglutination of Sindbis virus and of chick and hamster cells infected with Sindbis virus by two of the plant lectins, concanavalin A and Ricinus communis agglutinin. Both lectins agglutinate the virus by binding to the polysaccharide chains of the envelope glycoproteins. Both chick and hamster cells exhibit increased agglutination by the lectins after infection by Sindbis virus. In the case of chick cells infected with Sindbis virus, this increase in agglutinability occurs between 3 and 5 h after infection. Infected and mock-infected cells bind the same amount of (3)H-labeled concanavalin A, which suggests that the increase in agglutination after infection is due to rearrangements at the cell surface rather than to insertion of new lectin binding sites per se.     

The role of microvilli in the agglutination of cells by concanavalin A

Morphological correlates of lectin agglutinability were examined in eight cell lines of varying sensitivity to agglutination by concanavalin A (ConA). The number of microvilli on the surface of cells growing in monolayers was positively correlated with agglutinability. However, when cells were brought into suspension, they all developed numerous microvilli which persisted when the cells were treated with ConA regardless of whether or not they were agglutinated by the lectin. Treatment of cells with dibutyryl cyclic AMP (db-cAMP) and theophylline caused a parallel decrease in agglutinability and numbers of microvilli in monolayer cultures, but suspended cells from control and treated cultures were identical in appearance in the absence or presence of ConA. The surface morphology of cells agglutinated by ConA was very similar to that of cells that spontaneously agglutinated in the absence of the lectin, and surface bound ConA was rapidly withdrawn from microvilli on all cell types. Neither the morphology of cells nor the surface distribution of ConA can explain observed differences in agglutinability.     

CYCLIC AMP MEDIATES THE CONCANAVALIN A AGGLUTINABILITY OF MOUSE FIBROBLASTS

We have devised a quantitative way to measure the agglutination of cells which utilizes the size discrimination feature of an automatic particle counter. With this method we have studied the agglutinability by concanavalin A of 3T3 cells, a mutant of 3T3 cells (3T3cAMP^tcs) in which cyclic AMP levels fall when the cells are subjected to temperature change or fresh serum, and L929 cells. We find with 3T3cAMP^tcs cells that low levels of cyclic AMP correlate with increased agglutinability and that high levels of cyclic AMP correlate with decreased agglutinability. Prior treatment of these cells with a cyclic AMP phosphodiesterase inhibitor or Bt₂cAMP blocks the increase in agglutinability induced by temperature change. When 3T3 cells are treated with fresh serum, their agglutinability also increases although to a much smaller extent than with 3T3cAMP^tcs cells. Cells change their agglutinability very rapidly. Treatment of L929 cells for 15 min with 1-methyl-3-isobutyl xanthine at 1 mM decreases their agglutinability to the level of normal 3T3 cells. We conclude that in normal and transformed cells the level of cyclic AMP regulates agglutinability.     

Membrane Changes associated with Malignancy

The specific structural and metabolic changes on the surface membrane which result in agglutinability of transformed cells by concanavalin A are associated with cell malignancy in vivo.     

Qualitative and quantitative interactions of lectins with untreated and neuraminidase-treated normal, wild-type, and temperature-sensitive polyoma-transformed fibroblasts.

The lectin receptors of confluently grown hamster BHK, wild type polyoma virus transformed PyBHK, and temperature-sensitive polyoma transformed ts3-PyBHK fibroblasts were investigated using cell agglutination, quantitative (125I)lectin binding, and ferritin-lectin labeling. PyBHK and permissively grown ts3-PyBHK cells agglutinated more strongly with Ricinus communis I agglutinin (RCA-I)compared to BHK and nonpermissively grown ts3-PyBHK, although saturation binding of (125I)RCA-I to these cells at 4 degrees resulted in a twofold difference in lectin-binding sites on BHK and nonpermissively grown ts3-PyBHK cells (1.0-1.3 x 10 7 sites/cell) compared to PyBHK and permissively grown ts3-PyBHK (0.4-0.6 x 10 7 sites/cell). These cells bound equivalent amounts of (125I)concanavalin A (0.8-1 x 10 7 sites/cell) and (125I)wheat germ agglutinin (1-2.2 x 10 7 sites/cell). Under these binding conditions little endocytosis occurred, as judged by the subsequent release of greater than 90% cell-bound (125I)RCA-I by the RCA-I inhibitor lactose and localization of ferritin-RCA-I exclusively to the extracellular plasma membrane surface. However, if the binding is performed at 22 degrees, only 50% of the bound lectin can be removed by lactose, and ferritin-RCA-I is localized inside the cell within endocytotic vesicles. The relative mobility of RCA-I receptors was examined on ts3-PyBHK cells by the ability of ferritin-RCA-I to induce clustering of its receptors at 22 degrees. RCA-I receptors on permissively grown ts3-PyBHK cells appeared to be more mobile than on nonpermissively grown cells. BHK and PyBHK cells were treated with neuraminidase, and the resulting enzyme-treated cells were assayed for lectin agglutinability and quantitative binding of RCA-I, concanavalin A, and wheat germ agglutinin. Neuraminidase treatment resulted in decreased concanavalin A and wheat germ agglutinability and a slight increase in RCA-I agglutinability. The enzyme-treated BHK and PyBHK cells bound less (125I)wheat germ agglutinin (2.8 x 10 6 and 2.2 x 10 6 sites/cell, respectively) and 2.5 and 6.2 times more (125I)RCA-I (2.5-3 x 10 7) and 3.5-4 x 10 7 sites/per cell, respectively). There was no change in the number of concanavalin A binding sites after neuraminidase treatment. The increase in RCA-I binding sites approximated the decrease in wheat germ agglutinin binding sites indicating that the predominant penultimate oligosaccharide residue to sialic acid on these cells is D-Gal.