Effect of concanavalin A dose, unbound concanavalin A, temperature, ca²⁺ and mg²⁺, and vinblastine on capping of concanavalin A receptors of human peripheral blood lymphocytes

It has been observed that ligand receptors are randomly distributed on the surface of the cell membrane. The binding and crosslinking of multivalent ligands by lymphocytes and other cell types can lead to the formation of receptor spots or patches. This can be followed by energy-requiring receptor movement within the plane of the cell membrane resulting in the formation of an aggregation or cap at one pole of the cell. Caps can be endocytosed or shed leaving the cell surface denuded of capped receptors, which are subsequently regenerated (Ault & Unanue, 1974; Bretscher, 1976; DePetris & Raff, 1973; Greaves, Bauminger & Janassy, 1972; Gunther et al. 1973; Loor, Fomi & Pemis, 1972; Pemis, Fomi & Amante, 1970; Taylor, Duffus, Raff & DePetris, 1971; Unanue, Perkins & Kamovsky, 1972). Based on the fluid mosaic model of cell membranes (Singer & Nicolson, 1972) several suggestions concerning the modulation of surface receptors by ligands have been made (Bretscher, 1976; Edelman, 1976; Harris, 1976; Nicolson & Poste, 1976; Raff & DePetris, 1973).

The plant lectin, Concanavalin A (Con A) has been shown to be capable of inducing as well as inhibiting the capping of its own receptors (Yahara & Edelman, 1973b, 1975); and also of inhibiting the capping of immunoglobulin receptors (Unanue & Karnovsky, 1974; Yahara & Edelman, 1972). It has previously been reported using murine lymphocytes that capping of Con A receptors by Con A at 37 °C is inhibited unless the cells are first exposed to low temperatures or to antimitotic agents (Unanue et al. 1972; Yahara & Edelman, 1973 a, 1975). In our studies on the capping of human peripheral blood lymphocyte Con A receptors by Con A, we have found that under appropriate conditions capping of human lymphocyte Con A receptors by Con A can occur at 37 °C without prior low temperature treatment. Further, it was found that the increase in capping of Con A receptors induced by vinblastine is blocked in the presence of unbound Con A.

Heparinized (10 units/ml) venous blood was collected from healthy laboratory personnel. The mononuclear cell fraction was separated using a Ficoll-Hypaque gradient (Boym, 1968) and washed once with Hanks’ balanced salt solution. The cells (5 × 10⁷) were subsequently suspended in 40 ml of phospate-buffered saline (PBS, pH 7·3) with or without Ca²⁺ (CaCl₂, 6·4×10⁻⁴M) and Mg²⁺ (MgCl₂, 6·8 × 10⁻⁴ M) and centrifuged at 200 g for 10 min. The washing procedure was repeated twice more. Cell number was determined using an Accu-Stat blood cell counter and concentration adjusted to 1 × 10⁶ cells/ml. Cell viability was greater than 95% with or without the addition of 100 μg/ml Con A for 1 h using trypan blue dye exclusion test. Haemocytometer counts indicated that over 90% of the mononuclear cell fraction consisted of lymphocytes.

Human peripheral blood lymphocytes at a concentration of 1 × 10⁶/ml were incubated with 2·5–100 μg/ml of Con A (Type IV, Sigma) in PBS with or without Ca^a+ and Mg²⁺. Incubation was carried out at either 37 °C for 1 h or at 4 °C for 15 min, followed by incubation at 37 °C for 45–50 min. Excess Con A was removed after 15 min at either 4 or 37 °C by washing the cells 3 times with PBS with or without Ca²⁺ and Mg²⁺ at the appropriate temperature. In some experiments, the lymphocytes were treated with 10⁻⁴ M vinblastine (Eli Lilly) at room temperature for 30 min prior to incubation at 37 °C for 45–50 min.

Following incubation with Con A, the cells were washed 3 times with PBS and fixed in 1 % glutaraldehyde-PBS for 30 min. The cells were again washed with PBS and incubated with 50 μg/ml horseradish peroxidase (H.RP) (Type VI, Sigma) in PBS for 30 min at room temperature (Bernhard & Avrameas, 1971). Following a further wash, the HRP was localized by the diaminobenzidine reaction (DAB) (3,3′-diaminobenzidine tetrahydrochloride, Sigma) following the procedure of Graham & Kamovsky (1966). Cells were postfixed in 1 % osmium tetroxide at room temperature.

Following osmication, the cells were washed 3 times with PBS, centrifuged at 800 rev/min for 10 min and embedded in 4% aqueous agar. After dehydration with graded ethanols and propylene oxide, the agar blocks were embedded in Epon-Araldite and ultrathin sections cut using a Porter Blum MT-zB ultramicrotome. Sections were examined unstained for analysis of Con A-HRP-DAB labelling, or stained with uranyl acetate and lead citrate to visualize microtubules and microfilaments. Sections were viewed in a Zeiss EM 9S-2 transmission electron microscope.

In each experiment and at each Con A dose, several blocks were sectioned and a minimum of 100 cells were examined in thin sections. Cells were counted as being capped if the HRP label was confined to one half or less of the cell periphery (Fig. 1, p. 35). Care was taken to eliminate occasional granular leukocytes or monocytes from the count as determined by morphological appearance. Results were expressed as the percentage of cells examined which exhibited caps.

Preliminary studies were conducted with 10 μg/ml Con A in PBS containing Ca²⁺ and Mg²⁺ but without the presence of unbound Con A. Lymphocytes were pretreated at 4 °C with Con A and either fixed at this temperature or incubated at 37 °C for various times prior to fixation. When lymphocytes were pretreated with Con A and fixed at 4 °C, the Con A-HRP-DAB label was present around the entire cell periphery. After raising the temperature to 37 °C the label in many cells moved towards one pole of the cell to form a cap. Capping reached its maximum after 30 min with 50% of the cells having capped within 10 min. Further studies using 2·5–100 μg/ml Con A under the same conditions showed that the intensity of labelling was dependent upon Con A concentration; the most heavy labelling occurring at 100 μg/ml. At 2·5 μg/ml Con A, the label was quite faint and it proved difficult to judge the extent of capping at this dose. Consequently, further experimentation was confined to Con A concentrations of 5–100 μg/ml.

The influence of Con A dose, pretreatment with Con A at 4 °C, unbound Con A, and Ca²⁺ and Mg²⁺ on the capping of human lymphocyte Con A receptors following incubation at 37 °C is shown in Table 1. It can be seen that regardless of other variables Con A receptor capping was in general progressively inhibited with an increase in Con A concentration. Capping of Con A receptors at 37 °C in the presence of Ca²⁺ and Mg²⁺ and unbound Con A was markedly evident only at the lowest dose of Con A (Expt. 1). Pretreatment with Con A at 4 °C (Expt. 2) resulted in some increased capping at all Con A doses under the same conditions.

The frequency of capping at 37 °C increased with the removal of unbound Con A (Expt. 3). This increase was most evident at 10 and 20 μg/ml Con A. Pretreatment at 4 °C under the same conditions (Expt. 4), produced essentially the same or a slightly greater effect at the same Con A doses. However, the increased capping associated with lowering the temperature in Expt. 2 was still evident. After removal of Ca²⁺ and Mg²+ (Expt. 5), capping at 37 °C was markedly increased at 10, 20 and 50 μg/ml Con A. Pretreatment at 4 °C (Expt. 6) under these conditions produced essentially the same effect. Again the increased capping seen following incubation at 4 °C was still present.

In Expts. 7 and 8, both unbound Con A and Ca²⁺ and Mg²⁺ were removed. It can be seen that capping at 37 °C was now greater at all Con A doses than was found when unbound Con A or Ca²⁺ and Mg²+ were removed separately. It can also be seen that under these conditions lowering the temperature prior to incubation at 37 °C did not cause any further observable increase in the percentage of capped cells at most Con A doses.

Electron-microscopic examination of at least 100 cells for microtubules and microfilaments in each experiment and at each Con A dose as specified in Table 1 revealed the following: (1) microtubules, but not microfilaments, were absent or reduced in prevalence in the cytoplasm of cells incubated and fixed at 4 °C for 15–30 min (Fig. 2); (2) after increasing the temperature to 37 °C for 30 min, microtubules reappeared; (3) removing unbound Con A and/or Ca²⁺ and Mg²⁺ from the incubation medium had no apparent morphological effects on microtubules or microfilaments in cells incubated at 37 °C (Fig. 3); (4) microtubules and microfilaments were seen in different parts of cytoplasm and adjacent to the cell surface but there was no indication of their direct association with the cell membrane; and (5) although microtubules were often seen oriented towards the capped pole of the cell, they were occasionally found in other areas of the cytoplasm (Figs. 3–5).

Experiments were also performed to investigate Con A-HRP endocytosis and reappearance of Con A receptors in capped cells. Lymphocytes were preincubated in PBS containing Ca²⁺ and Mg²⁺ with 10 μg/ml Con A at 4 °C for 15 min. Excess Con A was removed by washing and the cells incubated for 30 min to 5 h at 37 °C. In capped cells, endocytosis was indicated by the presence of the label within the membrane-bound vesicles in the cytoplasm. These vesicles began to appear within 1 h of incubation at 37 °C and were always located in the cap portion of the cytoplasm. Cells which had not capped also showed evidence of label endocytosis around the entire periphery of the cell membrane. In another set of experiments, following varying periods of incubation at 37 °C, 10 μg/ml Con A was again added to the medium for 15 min. This was followed by washing and fixing of the cells. It was then observed that approximately 35% of the cells did not cap and that it required 4–5 h of incubation before the Con A-HRP label could again be seen around the periphery of 90% of the cells examined.

The effects of vinblastine on capping of Con A receptors are shown in Table 2. Comparison of Expt. 1 in Table 2 with Expt. 1 in Table 1 indicates that the addition of 10⁻⁴ M vinblastine had no effect on the percentage of capped cells in the presence of unbound Con A and Ca²⁺ and Mg²⁺. Electron-microscopic examination of cells treated with vinblastine under these conditions revealed the absence of microtubules along with the presence of paracrystalline structures. These observations indicate that vinblastine entered the cells despite its failure to increase Con A capping (Figs. 6, 7) (Bensch & Malawista, 1969). As seen in Expt. 2, when lymphocytes were first treated with vinblastine they showed a marked increase in capping. However, capping did not increase above 90% at all Con A concentrations unless unbound Con A and Ca²⁺ and Mg²⁺ had been removed (Expt. 3). Expt. 4 suggests that it was the presence of unbound Con A which inhibited the action of vinblastine on cells previously treated with Con A.

Our studies indicate that capping of Con A receptors by Con A could be increased by the pretreatment of cells at 4 °C, or by removing unbound Con A and/or extracellular Ca²⁺ and Mg²⁺. After removal of unbound Con A and Ca²⁺ and Mg²⁺, capping at 37 °C was generally equivalent to that seen following pretreatment of cells at 4 °C under the same condition; i.e. lowering the temperature did not cause any increase in the percentage of capped cells. Several studies have indicated a very low frequency of Con A-capped cells or none at all at 22 or 37 °C even after removal of excess Con A (DePetris & Raff, 1973; Yahara & Edelman, 19736, 1975). However, other studies have reported that Con A capping occurred at 22 or 37 °C in a high percentage of cells (Greaves et al. 1972; Smith & Hollers, 1970; Stackpole et al. 1974). These contradictory observations could reflect different experimental conditions, different strains of animals or different animal species used in these studies.

Our studies on human lymphocytes indicate that under certain conditions pretreatment of cells with Con A at 4 °C increased the capping of Con A receptors at 37 °C. Similar findings have been reported earlier in murine lymphocytes (Yahara & Edelman, 1973b). Although more Con A is bound to the cell at 37 than at 4 °C, this temperature-dependent difference in Con A capping is apparently not due solely to the amount of bound Con A (Yahara & Edelman, 19736). It has been suggested that temperature also affects the mobility of the receptor to which Con A binds. The mobility of Con A and other ligand receptors may in part be controlled by microtubules (Yahara & Edelman, 1973b). The increase in Con A-receptor capping following the pretreatment of cells with Con A at 4 °C has been attributed by some investigators to increased receptor mobility due to microtubule dissociation (Yahara & Edelman, 1973b, 1975). We observed that the percentage of capped cells after removing unbound Con A and extracellular Ca²+ and Mg²⁺ was the same as after pretreating cells with Con A at 4 °C under the same conditions. Further, removing unbound Con A and Ca²⁺ and Mg²⁺ did not result in the dissociation of microtubules.

Our findings that the incidence of Con A capping of Con A receptors increased after removal of extracellular Ca²⁺ and Mg²⁺ run counter to studies on the capping of murine immunoglobulin receptors which were not affected by removing extracellular divalent cations (Schreiner & Unanue, 1976; Taylor et al. 1971). There is evidence which suggests that membrane bound or intracellular calcium may play an important role in ligand capping. It has been shown by the studies of Freedman, Raff & Gomperts (1975) that the binding of Con A to T-lymphocytes results in a transient influx of extracellular calcium into the cell. A study by Schreiner & Unanue (1976) demonstrated that if intracellular Ca²⁺ was increased, capping of immunoglobulin receptors was inhibited and formed caps disrupted. Ryan, Unanue & Karnovsky (1974) reported that local anaesthetics and tranquillizers inhibit Con A- and immuno-globulin-receptor capping. As one explanation for their observations, these authors suggested that the drugs used in their studies prevented the release of intracellular calcium which in turn caused the inhibition of cytoplasmic structures involved in receptor movement. Poste, Papahadjopoulos, Jacobson & Vail (1975) observed that cell aggregation by Con A was increased following the use of local anaesthetics and suggested that this might result from an increase in membrane fluidity due to loss of membrane-bound Ca²⁺. Whether the presence of extracellular Ca²⁺ or Mg²⁺ directly or indirectly influences Con A-receptor movement is unclear. Our studies found that the inhibitory effects of unbound Con A on Con A-receptor capping could be reversed following removal of extracellular Ca²⁺ and Mg²⁺ except at the highest Con A dose. This could indicate that the inhibition of Con A-receptor movement by unbound Con A is dependent upon the presence of extracellular divalent cations except at high Con A doses. Extracellular Ca²⁺ and Mg²⁺ could also influence the interaction of Con A receptors with cytoplasmic proteins which may restrict receptor movement. However, removal of Ca²⁺ and Mg²⁺ may simply reduce the number of Con A molecules bound to the cell surface and thus shift the dose-response curve.

Previous studies have indicated that Con A-receptor capping can be increased by the use of colchicine or vinca alkaloids (Yahara & Edelman, 1973 a). We found that increased Con A capping induced by vinblastine is either partially or completely blocked if unbound Con A is present. This apparently is not because vinblastine was prevented from entering the cell. Our electron-microscopic analysis of vinblastine-treated lymphocytes in the presence of unbound Con A showed that microtubules were dissociated and paracrystalline structures formed. This would indicate that the entry of vinblastine into the cells had occurred (Bensch & Malawista, 1969). It has been suggested that the unbound Con A restricts receptor movement by cross-linking mobile receptors with receptors attached to cytoplasmic proteins sensitive to antimitotic agents. Treatment of cells with colchicine or vinblastine allows greater receptor mobility because few receptors remain in an attached state (Yahara & Edelman, 1973b, 1975). Our results suggest that unbound Con A may also effect receptor mobility through some other mechanism since vinblastine apparently entered the cells and became bound to the cytoplasmic sensitive proteins but did not induce maximal Con A-receptor movement.

This work was supported in part by Grant No. M76.14 from the Population Council, Rockefeller Foundation and in part by NIH Grant No. 2-SO6-RR-0816-07.