The effect of specific glycosidases onRicinus communis agglutinin binding to cell surfaces of two tumor sublines

The effects of the sequential application of specific glycosidases on surfaces of living mammalian cells were studied with respect to their ability to bind the β-galactoside-specific lectin,Ricinus communis agglutinin (RCA). Sialidase and β-galactosidases from different sources were tested for their actions on two strains of mouse lymphoma cells differing markedly in their metastatic potential. Binding studies were performed by quantitative flow cytometry with fluorescent RCA, and numbers of specific binding sites and equilibrium association constants for the lectin on living cells were determined before and after the various enzyme treatments. Although the number of binding sites for native and sialidase-treated cells were almost identical for both cell strains, differences in the apparent affinity constants could be detected. Differences between the two strains became even more pronounced, also with respect to the number of binding sites, after treatment with β-galactosidases fromS. pneumoniae and from bovine testis. It is suggested that such combined strategies provide valuable tools for the differentiation of surface carbohydrate moieties on intact living cells, especially for comparative purposes.     

Fluorescence-polarization studies on binding of 4-methylumbelliferyl beta-D-galactopyranoside to Ricinus communis (castor-bean) agglutinin.

The binding of Ricinus communis (castor-bean) agglutinin 1 to saccharides was studied by equilibrium dialysis and fluorescence polarization by using the fluorescently labelled sugar 4-methylumbelliferyl beta-D-galactopyranoside. No appreciable change in ligand fluorescence of 4-methylumbelliferyl beta-D-galactopyranoside was considerably polarized on its binding to the lectin. The association constants obtained by Scatchard analysis of equilibrium-dialysis and fluorescence-polarization data do not differ much from each other, and at 25 degrees C, Ka = 2.4 (+/- 0.2) X 10(4)M-1. These values agree reasonably well with that reported in the literature for Ricinus agglutinin 1. The number of binding sites obtained by the different experimental procedures is 1.94 +/- 0.1 per molecule of 120 000 daltons and is equal to the reported value of 2. The consistency in the values of Ka and number of binding sites indicate the absence of additional subsites on Ricinus agglutinin 1 for its specific sugars. In addition, the excellent agreement between the binding parameters obtained by equilibrium dialysis and fluorescence polarization indicate the potential of ligand-fluorescence-polarization measurements in the investigation of lectin-sugar interactions.     

The primary sequence of Ricinus communis agglutinin. Comparison with ricin

A mixture of synthetic oligonucleotides representing all possible sequences of a peptide present in the ricin B chain has been used to screen a cDNA library constructed using ripening castor bean seed poly(A+) RNA. The eight largest recombinant plasmids selected, by hybridization, a single mRNA species whose translational product was identified as preprolectin by immunoprecipitation. Restriction enzyme analysis of these clones demonstrated that two classes were present representing sequences complementary to two distinct but closely related preprolectin mRNA species. The nucleotide sequence of the cloned cDNA from one of these classes encodes preproricin and has been presented elsewhere (Lamb, F. I., Roberts, L. M., and Lord, J. M., (1985) Eur. J. Biochem. 148, 265-270). The nucleotide sequence of the second class is presented here and shown to represent prepro-Ricinus communis agglutinin. The entire coding sequence was deduced from two overlapping cDNA clones having inserts of 1668 and 1151 base pairs. The coding region defines a preproprotein with a 24-amino acid N-terminal signal sequence preceding the A chain (266 amino acids) which is joined to the B chain (262 amino acids) by a 12-amino acid linking peptide. The protein was confirmed as R. communis agglutinin since the deduced B chain N-terminal sequence corresponds exactly with that determined for purified R. communis agglutinin B chain over a region where several residue differences occur in the ricin B chain. The nucleotide and deduced amino acid sequences of the R. communis agglutinin precursor are compared with those of the ricin precursor.     

Effect of metabolic inhibitors on the agglutination of tumor cells by concanavalin A and Ricinus communis agglutinin

The agglutinations of rat ascites tumor cells by concanavalin A and by $\text{Ricinus communis}$ agglutinin were inhibited by low temperature, 2,4-dinitrophenol and cytochalasin B but not by cycloheximide. These metabolic inhibitors, however, did not inhibit the binding of the agglutinins to the cells. These results suggest that the agglutination was dependent on an active process; probably on a microfilament system responsible for cell surface movement which requires ATP, but not on protein synthesis.     

Cellular energy dependent agglutination of rat ascites tumor cells mediated by concanavalin A and Ricinus communis agglutinin.

Effect of various metabolic inhibitors on the agglutination of rat ascites tumor cells mediated by concanavalin A and Ricinus communis agglutinin was studied using a quantitative assay method for agglutination in which turbidity of cell suspension is measured. Cell agglutination was inhibited by low temperature, cytochalasin B and inhibitors of energy generating systems without affecting lectin binding, and agglutination was not affected by hydroxyurea, actinomycin D or cycloheximide. The inhibitors of energy generating systems decreased the cellular ATP level and inhibited macromolecular synthesis under the conditions where they inhibited the agglutinations. In contrast, cytochalasin B did not depress the cellular ATP level nor inhibit RNA and protein syntheses. These results suggest that the agglutination is associated with cellular energy dependent processes other than macromolecular synthesis; probably with some cellular surface movements participated by microfilament activity.     

Binding of Ricinus communis agglutinin to the mitochondrial inner membrane as an artifact during preparation.

Endosperm from Ricinus communis was homogenized in the presence of 3H-labelled Ricinus communis agglutinin, with or without addition of lactose. In preparations without the binding-specific sugar the subfraction containing the mitochondrial inner membrane contained sufficient labelled agglutinin to account for the agglutinin reported to be associated with this membrane.     

The surface glycoproteins of human skin fibroblasts detected after electrophoresis by the binding of peanut (Arachis hypogaea) agglutinin and Ricinus communis (castor-bean) agglutinin I.

A new methodology was developed to study the cell-surface glycoproteins of cultured human skin fibroblasts. This was based on the binding of a variety of biotinyl-lectins to nitrocellulose electrophoretic transfers of total fibroblast lysates after separation in sodium dodecyl sulphate/polyacrylamide gels, followed by reaction with avidin-biotinyl-peroxidase complexes and detection with 3,3'-diaminobenzidine. The technique proved to be very sensitive and a large number of glycoproteins were detected by binding of concanavalin A and wheat-germ agglutinin. Binding of peanut agglutinin and to a lesser extent of Ricinus communis agglutinin I were found to be dependent on prior removal of sialic acid residues from the glycoproteins. Since by treatment of intact viable cells with neuraminidase only external sialic acid residues were removed, peanut agglutinin and Ricinus communis agglutinin I could thus be utilized for selective detection of cell-surface glycoproteins. Also, because peanut agglutinin was known to bind preferentially to oligosaccharides of the O-glycosidic type, and Ricinus communis agglutinin I to those of the N-glycosidic type, the two lectins were complementary in displaying the surface glycoproteins and in providing information about their oligosaccharide composition.     

Interactions of lectins and monoclonal antibodies with human mononuclear cells. I. Specific inhibition of OKT4 and OKT8 binding by Ricinus communis agglutinin and wheat germ agglutinin

We used flow cytometry to examine effects of lectins on interactions between human lymphocytes and the anti-T cell monoclonal reagents OKT4 (T helper-specific) and OKT8 (T suppressor-specific). Wheat germ agglutinin (WGA) inhibited OKT8 binding to lymphocytes by a mean 77% and Ricinus communis agglutinin (RCA-I) inhibited OKT4 binding by 66%. Inhibition was abolished in each case by appropriate carbohydrate hapten inhibitors of lectin binding, indicating it was mediated by the lectin saccharide combining sites. Neither WGA nor RCA-I inhibited binding of OKT3, a pan-T cell monoclonal reagent. In addition, a group of other lectins with a variety of nominal carbohydrate specificities did not inhibit OKT4 or OKT8 binding. Preincubation experiments and gel filtration indicated that inhibition in each case was due to competition between lectin and monoclonal for binding to cell surfaces, not to direct lectin-monoclonal antibody interactions. Treatment of lymphoid cells with OKT8 and complement reduced OKT8- and WGA-binding cells concurrently, whereas treatment with OKT4 and complement did not reduce percentages of either type of cell. Similarly, specific depletion of OKT8-binding cells abolished the mitogenic response to WGA but not that to PHA. Cell populations enriched for WGA-binding cells prepared by flow cytometry and cell sorting demonstrated parallel enrichment for OKT8-binding and depletion of OKT4-binding cells. Therefore, these data demonstrate specific inhibition of OKT4 and OKT8 binding by the lectins, RCA-I and WGA, respectively. Inhibition was mediated by lectin binding to lymphoid cell surfaces, perhaps directly to the T4 or T8 antigens. The observations indicate that lectins may prove useful for investigating structural features of some immunologic cell surface markers. Furthermore, they provide the possibility that certain in vitro effects of lectins on immune function may result from their interactions with molecules such as the T4 and T8 antigens.     

Quantitative changes of Ricinus communis agglutinin I and Helix pomatia lectin binding sites in the acrosome of rat spermatozoa during epididymal transit.

During passage through the epididymis, spermatozoa undergo a number of changes which result in their acquisition of fertility and motility. Some of the changes that occur include loss of the cytoplasmic droplet and changes in sperm morphology, metabolism and properties of the nucleus and plasma membrane. Changes have also been reported in the acrosomic system of mammalian spermatozoa during their transit through the epididymis. In the present study, the quantitative changes of the glycoconjugate content in the acrosome of rat spermatozoa were examined during their passage through the epididymis using lectin-colloidal gold cytochemistry. Various regions of the epididymis (initial segment, caput, corpus and cauda epididymidis) were fixed by perfusion with 1% or 2% glutaraldehyde buffered in sodium cacodylate (0.1 M), dehydrated in ethanol and embedded without osmication in Lowicryl K4M. Lectin-colloidal gold labeling was performed on thin sections using Ricinus communis agglutinin I (RCA I) or Helix pomatia lectin (HPL) to detect D-galactose- and N-acetyl-D-galactosamine-containing glycoconjugates, respectively. The labeling density over the acrosome of the acrosomic system was evaluated as the number of gold particles per microns 2 of profile area using a Zeiss MOP-3 image analyzer. The overall mean labeling densities over the acrosome of spermatozoa for each lectin was estimated from 4 rats and over the four distinct epididymal regions. The mean labeling density of the acrosome with RCA I and HPL showed a similar pattern along the epididymis, although RCA I revealed approximately twice as many gold particles per epididymal region. In either case, there was a significant decrease in the labeling density of the acrosome of spermatozoa between the initial segment or caput epididymidis and cauda epididymidis (p less than 0.01). A similar decrease was also noted between the initial segment and corpus epididymidis (p less than 0.01). No change was found between the initial segment and caput epididymidis. Controls showed a virtual absence of labeling. These results suggest that in addition to a multitude of changes occurring to spermatozoa during epididymal transit, there are also significant quantitative changes in the glycoconjugate content within the acrosome.     

A quantitative assay for concanavalin A- and Ricinus communis agglutinin-mediated agglutinations of rat ascites hepatoma cells. Relationship between concanavalin A binding and cell agglutination.

A simple quantitative assay method was developed for the agglutination of rat ascites hepatoma cells mediated by Concanavalin A or Ricinus communis agglutinin. This method was based on the principle that the turbidity of a cell suspension is proportional to the sum of the cross-sectional area of cells and aggregatesmas predicted by the theoretical consideration, the turbidity decreased when cells were aggregated and the decrease was a function of the average number of the cells in aggregates. The agglutinability of the cells, judged by this method, showed a maximum value at a certain concentration of the agglutinin. By further addition of the agglutinin, the agglutinability slightly decreased from the maximum. These phenomena were observed both for Concanavalin A and Ricinus communis agglutinin. The binding and the agglutination experiments using [3-H]concanavalin A revealed that the binding to approx;0% of the total receptors caused a maximal agglutination. This suggested that the receptors responsible for the agglutination constitute only a small part of the total receptors on the surface.     

Differential tumor immunogenicity of L1210 and its sublines. I. Effect of an increased antigen density on tumor cell surfaces on primary B cell responses in vitro.

The differential immunogenicity of DBA/2 lymphoma L1210 and three L1210 sublines, each resistant to a different anti-leukemic agent (guanazole, methylglyoxal-bis-guanylhydrazone, and 4,4-diacetyldiphenylurea-bis-guanylhydrazone), was evaluated in vitro. Syngeneic spleen cells from nonimmunized DBA/2 mice were cultured in the presence of graded numbers of irradiated cells of L1210 or its sublines. The stimulation elicited a T-independent primary antibody response in vitro which was measured by determining the number of plaque-forming cells by using the immunizing lymphoma cells as target. Cells of all three sublines exhibited an increased immunogenicity, as compared to that of the parental L1210 cells, in eliciting the response directed to tumor-associated antigens which were common to all sublines. Dose-response experiments showed that high doses of the parental cells did stimulate responses which were detectable with subline cells as target. The results indicated that the differential immunogenicity of L1210 and its sublines, as demonstrated in the present assay system, is primarily quantitative, and was apparently due to increased amount or density of common tumor-associated antigens on the subline cells. The implications of these observations are discussed in relation to the possible mechanisms underlying the emergence of highly immunogenic drug-resistant sublines.     

Effects of nitrogen deprivation on cell division and expansion in leaves of Ricinus communis L.

The effects of nitrogen deprivation on leaf extension, cell numbers and epidermal cell size were followed in leaves of Ricinus communis L. The extent to which reductions in final cell number or final epidermal cell size contributed to the reduction in final leaf size depended on the developmental stage of the leaf at the time of N deprivation. In leaves which already had their full complement of cells (leaf 2), the reduction in final leaf size following nitrogen deprivation was associated with a reduction in final cell size. In leaves that were at earlier stages of development at the onset of N deprivation (leaves 3 and 4), the reduction in final leaf size was greater than in leaf 2. In these younger leaves, the final cell size was even smaller than in leaf 2, but the greatest contribution to reduced final leaf size was a reduction in the number of cells produced. This accounted for approximately 80% of the reduction in final leaf size in leaf 4. During leaf development, the contribution from different tissue layers to the total cell number changed. In the smallest leaf sizes, the contribution from upper and lower epidermis and spongy parenchyma was greater than that from palisade parenchyma. As the leaf size increased, cells in the palisade parenchyma continued to divide for longer than in the other layers. At final leaf size, the contribution from the different tissue layers to total cell number was the same for leaves 2, 3 and 4, irrespective of N treatment. In these final leaf structures, palisade parenchyma contributed 60% of the total cell number. Thus, although nitrogen deprivation affected leaf size variously through cell division and cell expansion, depending on leaf developmental stage at the time of nitrogen deprivation, the ratio of cell numbers and sizes in different tissue layers, at final leaf size, was unaffected.     

Surface carbohydrates of hamster fibroblasts. II. Interaction of hamster NIL cell surfaces with Ricinus communis lectin and concanavalin A as revealed by surface galactosyl label.

When hamster fibroblasts (NIL) were treated with galactose oxidase followed by reduction with tritiated borohydride (GAHMBERG, C. G. AND HAKOMORI, S. (1973) J. Biol. Chem. 248, 4311; STECK, T. L., AND DAWSON, G. (1974) J. Biol. Chem. 249, 2135), two major galactoprotein labels were detected on the cell surface: "galactoprotein a" (apparent molecular weight 200,000) and "galactoprotein b" (apparent molecular weight 130,000). The labeling in galactoprotein a of NIL cells was greatly suppressed by pretreating cells with a high concentration of Ricinus communis lectin or concanavalin A, whereas the label in it was greatly enhanced with a low concentration of the lectins. The label in galactoprotein b of NIL cells was less affected by pretreatment with lectins. In NILpy cells the label in galactoprotein a was absent and the lable in galactoprotein b was enhanced by pretreating cells with lectins at low concentrations, but it was suppressed at high concentrations. The results indicate that NIL cells mainly interact with the lectins through galactoprotein a, whereas NILpy cells interact with the lectins through galacto-protein b. After treatment with lectins, the glycolipids of normal NIL cells, but not NILpy cells, became exposed as evidenced by enhanced labeling, possibly because of "clustering" of glyco-proteins. These results support the view that specific, well defined glycoproteins are the binding sites for lectins, and that these interacting glycoproteins are qualitatively different in normal and transformed cells.