Planta
Planta (Berl.) 130, 73-79 (1976)
9 by Springer-Verlag 1976
Ultrastructural Studies of the Binding of Concanavalin A to the Plasmalemma of Higher Plant Protoplasts* J. Burgess and P.J. Linstead John Innes Institute, Colney Lane, Norwich NR4 7UH, U.K.
Summary. The binding of concanavalin A to the plasmalemma of higher plants has been studied using protoplasts of two species. The lectin aggregates both tobacco (Nicotiana tabacum L.) leaf protoplasts and protoplasts prepared from a suspension cell culture of grapevine (Vitis vinifera L.). Differences in lectin binding have been investigated using concanavalin A conjugated to ferritin or bound to colloidal gold. Tobacco protoplasts exhibit continuous and saturated labelling of the plasmalemma surface with gold-concanavalin A mixtures. Vine protoplasts under the same conditions show a discontinuous and patchy distribution of label. These results are discussed in terms of a possible binding mechanism.
(Garrido, 1975), and in their number and mobility with the degree of transformation of cells in culture (Nicolson, 1974; Roos and Temminck, 1975; Rutishauser and Sachs, 1975). The availability of plant protoplasts from a variety of sources means that it may now be possible to investigate some of the parallel properties of the plasmalemma of plant cells. The work to be described here is a study of the ConA binding properties of protoplasts isolated from tobacco leaves and from a tissue culture of Vitis vinifera.
Materials and Methods a) Tobacco Protoplasts
Introduction
It has recently been shown that isolated protoplasts of Daucus carota suspension callus cells are agglutinated by the lectin, concanavalin A (ConA) (Glimelius et al., 1974). The characteristics of this agglutination were broadly similar to that found with certain animal cells (Inbar etal., 1973; Diizgfines, 1975). The reaction of ConA with the plasmalemma of animal cells is presumed to be binding to specific receptor sites; these sites can be visualized by the use of ConA conjugated to ferritin (Nicolson, 1971) or bound to horseradish peroxidase (Temminck et al., 1975). Work on these lines has corroborated the concept of the plasmalemma as a fluid mosaic (Singer and Nicolson, 1972) with mobile ConA receptor sites as one of the exposed outer components (Nicolson and Singer, 1974). Other work has suggested variation in the distribution of these sites with the mitotic cycle * Abbreviations: ConA =concanavalin A ; PBS = Phospholi Buffered Saline; PEG=polyethylene glycol
These were isolated from fully expanded leaves of tobacco, Nicotiana tabacum L , var. White Burley, as previously described (Burgess and Fleming, 1974a).
b) Vine Protoplasts These were isolated from a liquid suspension culture of Vit& vinifera L., cv Mtiller Thurgau, originally established from stem segments in this laboratory in 1974, and maintained by subculture every 48 h in a medium containing salts (Nagata and Takebe, 1971) with 7 rag/1 1-naphthyl-acetic acid and 1 rag/1 benzylaminopurine supplemented with 10% v/v coconut milk. Protoplasts were prepared by pelleting the ceils, resuspending them in an equal volume of 0.6 M mannitol, and after 30 rain to allow plasmolysis to occur, repelleting the cells and resuspending them in 0.6 M mannitol containing 0.5% w/v Macerozyme R-10 (Kinki Yakult Mfg. Co.) and 0.5% w/v potassium dextran sulphate at pH 5.8. This mixture was shaken (60 excursions/min) in a shaking water bath at 25 ~ C for 30 rain. After this time cellulase (Onozuka R-10, Kinki Yakult Mfg. Co.) was added as a solid to give a final concentration of 1% w/v, and the temperature raised to 35~ C. The protoplasts were harvested one hour later by filtering the suspension through eight layers of cheesecloth and centrifugation at 50 g for 5 rain. Protoplasts were washed by repeated centrifugation in 0.6 M mannitol.
74
c) Ferrit& ConA To 1.8 mI of ferritin (Sigma, 100 mg/ml) were added 1.47 g c~methyl-D-mannoside (-MM) (Sigma) and 100 mg ConA (Pharmacia Fine Chemicals) and the mixture made up to 10 ml with 0.5 M sodium chloride buffered to pH 6.8 with 0.05 M sodium phosphate (PBS). The reaction was started by adding 10 gl of purified glutaraldehyde (25% w/v solution, Taab) and terminated after 40 rain at room temperature by dialysis for 1 h at 4~ C against PBS containmg 0.1 M ammonium chloride. The mixture was then dialysed overnight against PBS (2 changes) and the Fe-ConA purified by passing the mixture down a Sepharose 6 B column. The initial ferritin-containing peak (Klein and Adams, 1972) was pooled and used as the label after dialysis against 0.6 M mannitol, and dilution to twice the volume of solution so obtained with 0.6 M mannitol containing 10 mM calcium chloride.
d) Colloidal GoM ConA Colloidal gold was prepared by boiling a 10 a% (w/v) solution of chloroauric acid in water and adding 1% tri-sodium citrate solution at the rate of 1 ml per 50 ml chloroauric acid. This mixture was then boiled until an orange-red colour was fully developed (about three minutes) (Frens, 1971). The colloidal gold thus obtained was spun at 10,000 rpm. for 30 min in a Beckman J-21 centrifuge. This procedure gave a pellet overlaid by a concentrated gold sol with dear solvent above it. The clear supernatant was carefully removed and the concentrated sol pooled. The resultant pool was diluted with water to give a final volume of under 5 ml from 200 ml of starting solution. To this was added with stirring a concentrated aqueous solution of ConA to give a final concentration of 125 gg/ml. One rain later 500 pl of a 1% PEG (MW 20,000) solution was added to stabilize the colloid. Finally calcium chloride was added to a final concentration of 5 mM, and mannitol to 0.6 M, and the volume of the solution made up to exactly 5 ml. This is a modification of the method of Horisberger et al. (1975). In our initial experiments (see text) this 5 ml Au-ConA mixture was made fiom only 50 ml of starting solution.
e) Electron Microscopy This was carried out as previously described (Burgess and Fleming, 1974a) with the modification that low viscosity resin (Spurr, 1969) was used in place of Araldite (see text).
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
rapidly and results in the formation of large clumps of loosely bound protoplasts which sediment readily (Figs. 1, 2). When such preparations are fixed for electron microscopy care has to be taken if the aggregates are not to be broken down. For this reason the low viscosity resin developed by Spurr (1969) has been used throughout this study since the need for centrifuging specimens during the later stages of embedding is thus avoided.
b) Aggregation The characteristics for ConA aggregation are similar to those previously described for polyethylene glycol (PEG) (Burgess and Fleming, 1974a). Adjacent plasma membranes were closely adpressed over considerable distances in section (Fig. 3); the local strong deformation of the membranes observed after PEG treatment (Burgess and Fleming, 1974a) was not found in ConA aggregated protoplasts. The paired membranes were often strikingly parallel to one another (Fig. 3), although the spacing between them was not constant from aggregate to aggregate. Aggregated vine protoplasts, whilst showing the same overall behaviour, in general exhibited a greater spacing between paired membranes (Fig. 4). Fusion of paired membranes was observed only very rarely and only within tobacco protoplast aggregates; the effect was similar to that described for poly-L-lysine induced aggregation (Fig. 5, seealso Burgess and Fleming, 1974a). Aggregation with ConA was totally abolished by pre-incubating the lectin with 0.1 M ~-MM. No ultrastructural changes within the protoplasts themselves were detected following treatment with ConA.
e) Ferritin Conjugated ConA f ) Controls In normal experiments the pellet of washed protoplasts was directly suspended in the ConA mixture and left for 30 min at room temperature. Fixation was carried out either by adding a large excess of fixative to the 5 ml of protoplast suspension or by pelleting the protoplasts, washing them in 0.6 M mannitol, and fixing the pellet from this wash. Controls consisted of adding solid ~-MM to a final concentration of 0.1 M to the Fe-ConA or Au-ConA mixture, and preincubating it for 30 min before use. In addition "Au-BSA" was made by adding bovine serum albumin in place of ConA to the gold sol as a control The gold sol alone without additives other than mannitol was also used as a control.
Results
a) General Observations The aggregation reaction using 250 lag/ml ConA in 0.6 M mannitol with 5 mM calcium chloride proceeds
Treatment of tobacco protoplasts with ConA conjugated to ferritin (Fe-ConA) resulted in the appearance of ferritin particles decorating the outside surface of the plasmalemma (Fig. 6). The distribution of these particles was discontinuous; regions with groups of ferritin particles were separated by regions of blank membrane. Occasionally the spacing between adjacent particles within a group was quite regular (Fig. 7). Vine protoplasts under the same conditions showed almost no ferritin on the membrane surface. Both types of protoplasts exhibited heavy ferritin labelling of amorphous material in the region of the plasmalemma. Areas of undigested wall material entrained within the protoplast pellet were similarly labelled. Pre-treatment of the Fe-ConA with ~-MM abolished the labelling of the plasmalemma of
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
75
Fig. 1. Freshly prepared vine protoplasts. No clumping is apparent • 200 Fig. 2. Vine protoplasts incubated for 30 min at room temperature in 250 gg/ml ConA. The protoplasts are associated into large loose aggregates, x 200 Fig. 3. Junction between two tobacco protoplasts aggregated for 30 min in 250 gg/ml ConA at room temperature. The surface membranes are closely adpressed an parallel for long distances in sections, x 48,000 Fig. 4. Junction between two aggregated vine protoplasts, as Fig. 2. The gap between the parallel surface membranes is generally greater than with tobacco protoplasts, x 23,000
76
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
Fig. 5. A discontinuity across paired surface membranes between aggregated tobacco protoplasts. The cytoplasm of the two protoplasts is in direct contact through a narrow channel. • 37,000 Fig. 6. Part o f the surface o f a tobacco protoplast incubated with Fe-ConA for 30 min at room temperature before fixation. Clusters of ferritin particles are visible on the surface membrane and also associated with amorphous material away from the surface (Arrow). The membrane is not saturated with label, x 50,000 Fig. 7. As Fig. 6. The spacing of the ferritin particles is quite regular within some of the membrane-bound clusters, x 180,000 Fig. 8. Section of a tobacco protoplast labelled with Au-ConA for 30 rain at room temperature. The membrane shows a continuous cover with gold particles. The gold is also associated with material away from the membrane surface, x 16,000 Fig. 9. As Fig. 8. Where loose aggregation occurs, gold particles extend between the paired membranes, x 32,000
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
77
Fig. 10. Surface of a vine protoplast exposed to Au-ConA for 30 min at room temperature prior to fixation. The amount of gold on the membrane surface is much lower than with tobacco under the same conditions, x 32,000 Fig. 11. Surface of a vine protoplast which was briefly pre-fixed in glutaraldehyde before exposure to Au-ConA. The label is evenly distributed and at a much higher density, x 40,000 Fig. 12. Part of the surface of a tobacco protoplast which was incubated overnight in a regeneration medium before being exposed to Au-ConA and fixed. In comparison to the fresh protoplast (Fig. 8) there is less label in close contact with the membrane itself and more label associated with stranded material away from the surface, x 15,000
tobacco protoplasts, but did not eliminate the ferritin labelling of the a m o r p h o u s material or wall fragments.
d) Colloidal Gold ConA Initial experiments with colloidal gold ConA mixtures (Au-ConA) gave similar results to those obtained with F e - C o n A ; tobacco protoplasts showed discontinuous and patchy labelling, whilst vine protoplasts showed almost no labelling at all. However, when the ratio of C o n A to gold in the mixture was reduced by a factor of four (see Methods) a different pattern emerged. Tobacco protoplasts treated with this mixture exhibited heavy and continuous labelling over the entire p l a s m a l e m m a surface (Fig. 8). Where protoplasts were aggregated in the presence of the AuConA, gold particles could be found between the paired p l a s m a l e m m a surfaces (Fig. 9). Vine proto-
plasts under the same conditions did not show saturation with gold particles, but rather a patchy and irregular distribution (Fig. 10). Increased times of incubation o f the vine protoplasts with the A u - C o n A did not change this labelling pattern. Prefixation of protoplasts in glutaraldehyde solutions before treatment with A u - C o n A resulted in an increase in the amount of label seen at the plasmalemma surface in thin sections. This increase in the labelling was most m a r k e d for vine protoplasts, which v~ould normally not show saturation of the surface (cf. Figs. 10, 11). The increase in labelling was greatest when washing after prefixation was minimised. When particular care was taken to wash prefixed protoplasts thoroughly, with m a n y changes of washing solution over 16 h, the increase in labelling due to prefixation was negligible in all cases. Incubation of protoplasts overnight in a nutrient medium known to support the regeneration of a wall around tobacco protoplasts (Burgess and Fleming,
78
1974b) resulted in a change in the pattern of labelling with Au-ConA. Tobacco protoplasts showed a discontinuous distribution of label which was predominantly in the form of clumps rather than single particles at the membrane surface (Fig. 12). Vine protoplasts incubated overnight showed a labelling pattern very similar to freshly prepared tobacco protoplasts; the membrane was more or less saturated with label, with a few clumps of gold particles attached to the protoplasts periphery. Various control experiments were carried out to establish the role of the ConA in the binding reaction. Binding was abolished by preincubating the Au-ConA with e-MM before use. No binding was observed at all when BSA was substituted for ConA. Colloidal gold on its own did not label the membrane surface but was occasionally seen attached to entrained wall fragments.
Discussion
These results confirm and extend a previous report (Glimelius et al., 1974) that ConA binds to the outer surface of the plant plasmalemma in such a way as to cause aggregation and the close approach of paired membranes. This aggregation reaction is abolished by the ConA inhibitor e-methyl-D-mannoside. Although no attempt has been made to measure the extent of aggregation in a quantitative manner, it is clear that tobacco protoplasts are aggregated more readily than are vine protoplasts. This is reflected in more rapid formation of aggregates at a given ConA concentration, and in greater degrees of visual aggregation at reduced ConA concentrations. For electron microscope study 250 gg/ml ConA gives a very high degree of aggregation within 30 rain at room temperature with both type of protoplasts; for visualization of binding patterns 125 gg/m! ConA gives saturating labelling after 30 min at room temperature, at least with tobacco protoplasts. The distribution of binding sites has been found to vary with both types of protoplast and with the label used. Fe-ConA gives a pattern of labelling which at first suggests a patchy distribution of sites such as has been found in a variety of animal cells (Nicolson, 1971 ; Amakawa and Barka, 1975 ; Garrido, 1975). This patchy distribution was also found using AuConA with a relatively high ratio of ConA:Au. Two interpretations of such behaviour are possible; either the distribution is patchy or it is more continuous and the label used to visualize it contains free ConA which masks a proportion of the sites. Decreasing the ratio of ConA:Au by a factor of four whilst maintaining the solution concentration of ConA at
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
125 gg/ml (see Methods) does indeed abolish the patchy labelling on the tobacco protoplast plasmalemma. Under these conditions it may be assumed that the great majority of the ConA molecules present in the Au-ConA mixture are associated with colloidal gold. The binding of the label to the plasmalemma is directly associated with the ConA since it is abolished by e-MM, and by absence of ConA in the mixture. Furthermore the binding is not restored by use of another protein, bovine serum albumin. The preparation of Au-ConA is far simpler and more reproducible than the preparation of Fe-ConA, and for this reason attention has been concentrated on the AuConA technique. Tobacco and vine protoplasts have consistently shown different behaviour towards ConA. Tobacco protoplasts are more easily aggregated by it and show more extensive labelling. This suggests a difference in the chemical environment of the plasmalemma surface of these two species. It is possible that such a difference might reflect the origins of the protoplasts themselves, since one is directly isolated from a normally growing tissue, the leaf, whilst the other derives from an undifferentiated tissue culture. However, there is a simple physiological difference between the two types of protoplasts related to their ability to regenerate a cell wall. Tobacco protoplasts readily and rapidly regenerate an intact wall (Burgess and Fleming, 1974b), whereas under the same conditions, vine protoplasts after 5 days culture show only a very tenuous and atypical wall (Linstead, unpublished). Both Au-ConA and Fe-ConA bind readily to wall fragments remaining after digestion during protoplast formation. It is therefore reasonable to suppose that labelling of the plasmalemma surface of freshly prepared protoplasts represents binding of ConA to nascent wall material. The difference between the two species is possibly therby explained by the relative absence of such nascent wall material from the vine protoplast surface. Incubation of protoplasts overnight produces a change in the labelling pattern which may also be explained in these terms. After overnight incubation, vine protoplasts show overall membrane-bound labelling, corresponding to a slow incipient wall formation. By contrast tobacco protoplasts show patches of heavy labelling with clumps of gold particles corresponding to more extensive wall formation. The absence of labelling from other regions would correspond to sites where wall material had been lost during processing, for electron microscopy, a common observation at such an early stage of wall formation (Burgess and Fleming, 1974b). There seems no need to invoke mobile ConA receptor sites as the underlying mechanism to explain
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
the observed patterns of labelling in this study. Prefixation of cells may be used to immobilise such sites on animal cell surfaces and thus prevent their clumping (Rutishauser and Sachs, 1975). Attempts to reproduce this behaviour with vine protoplasts which show a patchy distribution of label even with the low ConA:Au ratio mixture, have failed. The general and increased labelling which follows prefixation seems to be related to the presence of residual aldehyde fixative. If careful washing procedures are adopted the pattern of labelling is restored to that in unfixed specimens. It is clear that by whatever mechanism binding takes place, labelling with ConA in conjunction with electron dense markers may provide a potentially useful tool to identify the fate of the plasmalemma in a variety of situations. Preliminary experiments show that the label is not removed by subsequent exposure to high concentrations of polyethylene glycol, for example, and this should allow further elucidation of events taking place during and after protoplast fusion.
References Amakawa, T., Barka, T. : Distribution of Concanavalin A binding sites on the surface of dissociated rat submandibular acinar cells. J. Histochem. Cytochem. 23, 607~517 (1975) Burgess, J., Fleming, E.N.: Ultrastructural studies of the aggregation and fusion of plant protoplasts. Planta (Berl.) 118, i83 I93 (1974a) Burgess, J., Fleming, E.N.: Ultrastructural observations of cell wall regeneration around isolated tobacco protoplasts. J. Cell Sci. 14, 439-449 (1974b) Dfizgfines, N.: The Concanavalin A agglutinating system of cell membranes. Biosystems 6, 209-216 (1975) Frens, G.: Controlled nucleation for the regulation of particle
79 size in monodisperse gold suspensions. Nature physical Sci. 241, 20~2 (1971) Garrido, J. : Uttrastructural labelling of cell surface lectin receptors during the cell cycle. Exp. Cell Res. 94, 159 175 (1975) Glimelius, K., Wallin, A., Eriksson, T. : Agglutinating effects of Concanavalin A on isolated protoplasts of Daucus carota. Physiol. Plant. 31, 225-230 (1974) Horisberger, M., Rosset, J., Bauer, H. : Colloidal gold granules as markers for cell surface receptors in the scanning electron microscope. Experientia 31, 1147-1149 (1975) Inbar, M., Huet, C., Oseroff, A.R., Ben-Bassat, H., Sachs, L.: Inhibition of lectin agglutinability by fixation of the cell surface membrane. Biochim. biophys. Acta (Amst.) 311,594-599 (1973) Klein, P.A., Adams, W.R. : Location of ferritin-labelled Concanavalin A binding to influenza virus and tumor celt surfaces. J. Virol. 10, 844-854 (1972) Nagata, T., Takebe, I. : Plating of isolated tobacco mesophyll protoplasts on agar medium. Planta (Berl.) 92, 301-308 (1971) Nicolson, G.L.: Difference in topology of normal and tumour cell membranes shown by different surface distributions of ferritin-conjugated Concanavalin A. Nature new biol. 233, 244-246 (197I) Nicolson, G.L.: Interactions of lectins with animal cell surfaces. Int. Rev. Cytol. 39, 89-190 (1974) Nicolson, G.L., Singer, S.J.: The distribution and assymetry of mammalian cell surface polysaccharides utilising ferritin-conjugated plant agglutinins as specific saccharide stains. J. Cell Biol. 60, 236-248 (1974) Roos, E., Temminck, J.H.M.: Cytochemical comparison between wheat germ agglutinin and Concanavalin A bound to mouse fibroblasts in vitro. Exp. Cell Res. 94, 140-146 (1975) Rutishauser, U., Sachs, L.: Receptor mobility and the binding of cells to lectin coated fibres. J. Cell Biol. 66, 76-85 (1975) Singer, S.J., Nicolson, G.L. : The fluid mosaic model of the structure of cell membranes. Science 175, 720-731 (1972) Spurr, A.R. : A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26, 31-43 (1969) Temminck, J.H.M., Collard, J.G., Spits, H., Roos, E. : A comparative study of four cytochemical detection methods of Concanavalin A binding sites on the cell membrane. Exp. Cell Res. 92, 307-322 (1975)
Received 18 December 1975; accepted 7 January 1976
Planta (Berl.) 130, 73-79 (1976)
9 by Springer-Verlag 1976
Ultrastructural Studies of the Binding of Concanavalin A to the Plasmalemma of Higher Plant Protoplasts* J. Burgess and P.J. Linstead John Innes Institute, Colney Lane, Norwich NR4 7UH, U.K.
Summary. The binding of concanavalin A to the plasmalemma of higher plants has been studied using protoplasts of two species. The lectin aggregates both tobacco (Nicotiana tabacum L.) leaf protoplasts and protoplasts prepared from a suspension cell culture of grapevine (Vitis vinifera L.). Differences in lectin binding have been investigated using concanavalin A conjugated to ferritin or bound to colloidal gold. Tobacco protoplasts exhibit continuous and saturated labelling of the plasmalemma surface with gold-concanavalin A mixtures. Vine protoplasts under the same conditions show a discontinuous and patchy distribution of label. These results are discussed in terms of a possible binding mechanism.
(Garrido, 1975), and in their number and mobility with the degree of transformation of cells in culture (Nicolson, 1974; Roos and Temminck, 1975; Rutishauser and Sachs, 1975). The availability of plant protoplasts from a variety of sources means that it may now be possible to investigate some of the parallel properties of the plasmalemma of plant cells. The work to be described here is a study of the ConA binding properties of protoplasts isolated from tobacco leaves and from a tissue culture of Vitis vinifera.
Materials and Methods a) Tobacco Protoplasts
Introduction
It has recently been shown that isolated protoplasts of Daucus carota suspension callus cells are agglutinated by the lectin, concanavalin A (ConA) (Glimelius et al., 1974). The characteristics of this agglutination were broadly similar to that found with certain animal cells (Inbar etal., 1973; Diizgfines, 1975). The reaction of ConA with the plasmalemma of animal cells is presumed to be binding to specific receptor sites; these sites can be visualized by the use of ConA conjugated to ferritin (Nicolson, 1971) or bound to horseradish peroxidase (Temminck et al., 1975). Work on these lines has corroborated the concept of the plasmalemma as a fluid mosaic (Singer and Nicolson, 1972) with mobile ConA receptor sites as one of the exposed outer components (Nicolson and Singer, 1974). Other work has suggested variation in the distribution of these sites with the mitotic cycle * Abbreviations: ConA =concanavalin A ; PBS = Phospholi Buffered Saline; PEG=polyethylene glycol
These were isolated from fully expanded leaves of tobacco, Nicotiana tabacum L , var. White Burley, as previously described (Burgess and Fleming, 1974a).
b) Vine Protoplasts These were isolated from a liquid suspension culture of Vit& vinifera L., cv Mtiller Thurgau, originally established from stem segments in this laboratory in 1974, and maintained by subculture every 48 h in a medium containing salts (Nagata and Takebe, 1971) with 7 rag/1 1-naphthyl-acetic acid and 1 rag/1 benzylaminopurine supplemented with 10% v/v coconut milk. Protoplasts were prepared by pelleting the ceils, resuspending them in an equal volume of 0.6 M mannitol, and after 30 rain to allow plasmolysis to occur, repelleting the cells and resuspending them in 0.6 M mannitol containing 0.5% w/v Macerozyme R-10 (Kinki Yakult Mfg. Co.) and 0.5% w/v potassium dextran sulphate at pH 5.8. This mixture was shaken (60 excursions/min) in a shaking water bath at 25 ~ C for 30 rain. After this time cellulase (Onozuka R-10, Kinki Yakult Mfg. Co.) was added as a solid to give a final concentration of 1% w/v, and the temperature raised to 35~ C. The protoplasts were harvested one hour later by filtering the suspension through eight layers of cheesecloth and centrifugation at 50 g for 5 rain. Protoplasts were washed by repeated centrifugation in 0.6 M mannitol.
74
c) Ferrit& ConA To 1.8 mI of ferritin (Sigma, 100 mg/ml) were added 1.47 g c~methyl-D-mannoside (-MM) (Sigma) and 100 mg ConA (Pharmacia Fine Chemicals) and the mixture made up to 10 ml with 0.5 M sodium chloride buffered to pH 6.8 with 0.05 M sodium phosphate (PBS). The reaction was started by adding 10 gl of purified glutaraldehyde (25% w/v solution, Taab) and terminated after 40 rain at room temperature by dialysis for 1 h at 4~ C against PBS containmg 0.1 M ammonium chloride. The mixture was then dialysed overnight against PBS (2 changes) and the Fe-ConA purified by passing the mixture down a Sepharose 6 B column. The initial ferritin-containing peak (Klein and Adams, 1972) was pooled and used as the label after dialysis against 0.6 M mannitol, and dilution to twice the volume of solution so obtained with 0.6 M mannitol containing 10 mM calcium chloride.
d) Colloidal GoM ConA Colloidal gold was prepared by boiling a 10 a% (w/v) solution of chloroauric acid in water and adding 1% tri-sodium citrate solution at the rate of 1 ml per 50 ml chloroauric acid. This mixture was then boiled until an orange-red colour was fully developed (about three minutes) (Frens, 1971). The colloidal gold thus obtained was spun at 10,000 rpm. for 30 min in a Beckman J-21 centrifuge. This procedure gave a pellet overlaid by a concentrated gold sol with dear solvent above it. The clear supernatant was carefully removed and the concentrated sol pooled. The resultant pool was diluted with water to give a final volume of under 5 ml from 200 ml of starting solution. To this was added with stirring a concentrated aqueous solution of ConA to give a final concentration of 125 gg/ml. One rain later 500 pl of a 1% PEG (MW 20,000) solution was added to stabilize the colloid. Finally calcium chloride was added to a final concentration of 5 mM, and mannitol to 0.6 M, and the volume of the solution made up to exactly 5 ml. This is a modification of the method of Horisberger et al. (1975). In our initial experiments (see text) this 5 ml Au-ConA mixture was made fiom only 50 ml of starting solution.
e) Electron Microscopy This was carried out as previously described (Burgess and Fleming, 1974a) with the modification that low viscosity resin (Spurr, 1969) was used in place of Araldite (see text).
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
rapidly and results in the formation of large clumps of loosely bound protoplasts which sediment readily (Figs. 1, 2). When such preparations are fixed for electron microscopy care has to be taken if the aggregates are not to be broken down. For this reason the low viscosity resin developed by Spurr (1969) has been used throughout this study since the need for centrifuging specimens during the later stages of embedding is thus avoided.
b) Aggregation The characteristics for ConA aggregation are similar to those previously described for polyethylene glycol (PEG) (Burgess and Fleming, 1974a). Adjacent plasma membranes were closely adpressed over considerable distances in section (Fig. 3); the local strong deformation of the membranes observed after PEG treatment (Burgess and Fleming, 1974a) was not found in ConA aggregated protoplasts. The paired membranes were often strikingly parallel to one another (Fig. 3), although the spacing between them was not constant from aggregate to aggregate. Aggregated vine protoplasts, whilst showing the same overall behaviour, in general exhibited a greater spacing between paired membranes (Fig. 4). Fusion of paired membranes was observed only very rarely and only within tobacco protoplast aggregates; the effect was similar to that described for poly-L-lysine induced aggregation (Fig. 5, seealso Burgess and Fleming, 1974a). Aggregation with ConA was totally abolished by pre-incubating the lectin with 0.1 M ~-MM. No ultrastructural changes within the protoplasts themselves were detected following treatment with ConA.
e) Ferritin Conjugated ConA f ) Controls In normal experiments the pellet of washed protoplasts was directly suspended in the ConA mixture and left for 30 min at room temperature. Fixation was carried out either by adding a large excess of fixative to the 5 ml of protoplast suspension or by pelleting the protoplasts, washing them in 0.6 M mannitol, and fixing the pellet from this wash. Controls consisted of adding solid ~-MM to a final concentration of 0.1 M to the Fe-ConA or Au-ConA mixture, and preincubating it for 30 min before use. In addition "Au-BSA" was made by adding bovine serum albumin in place of ConA to the gold sol as a control The gold sol alone without additives other than mannitol was also used as a control.
Results
a) General Observations The aggregation reaction using 250 lag/ml ConA in 0.6 M mannitol with 5 mM calcium chloride proceeds
Treatment of tobacco protoplasts with ConA conjugated to ferritin (Fe-ConA) resulted in the appearance of ferritin particles decorating the outside surface of the plasmalemma (Fig. 6). The distribution of these particles was discontinuous; regions with groups of ferritin particles were separated by regions of blank membrane. Occasionally the spacing between adjacent particles within a group was quite regular (Fig. 7). Vine protoplasts under the same conditions showed almost no ferritin on the membrane surface. Both types of protoplasts exhibited heavy ferritin labelling of amorphous material in the region of the plasmalemma. Areas of undigested wall material entrained within the protoplast pellet were similarly labelled. Pre-treatment of the Fe-ConA with ~-MM abolished the labelling of the plasmalemma of
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
75
Fig. 1. Freshly prepared vine protoplasts. No clumping is apparent • 200 Fig. 2. Vine protoplasts incubated for 30 min at room temperature in 250 gg/ml ConA. The protoplasts are associated into large loose aggregates, x 200 Fig. 3. Junction between two tobacco protoplasts aggregated for 30 min in 250 gg/ml ConA at room temperature. The surface membranes are closely adpressed an parallel for long distances in sections, x 48,000 Fig. 4. Junction between two aggregated vine protoplasts, as Fig. 2. The gap between the parallel surface membranes is generally greater than with tobacco protoplasts, x 23,000
76
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
Fig. 5. A discontinuity across paired surface membranes between aggregated tobacco protoplasts. The cytoplasm of the two protoplasts is in direct contact through a narrow channel. • 37,000 Fig. 6. Part o f the surface o f a tobacco protoplast incubated with Fe-ConA for 30 min at room temperature before fixation. Clusters of ferritin particles are visible on the surface membrane and also associated with amorphous material away from the surface (Arrow). The membrane is not saturated with label, x 50,000 Fig. 7. As Fig. 6. The spacing of the ferritin particles is quite regular within some of the membrane-bound clusters, x 180,000 Fig. 8. Section of a tobacco protoplast labelled with Au-ConA for 30 rain at room temperature. The membrane shows a continuous cover with gold particles. The gold is also associated with material away from the membrane surface, x 16,000 Fig. 9. As Fig. 8. Where loose aggregation occurs, gold particles extend between the paired membranes, x 32,000
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
77
Fig. 10. Surface of a vine protoplast exposed to Au-ConA for 30 min at room temperature prior to fixation. The amount of gold on the membrane surface is much lower than with tobacco under the same conditions, x 32,000 Fig. 11. Surface of a vine protoplast which was briefly pre-fixed in glutaraldehyde before exposure to Au-ConA. The label is evenly distributed and at a much higher density, x 40,000 Fig. 12. Part of the surface of a tobacco protoplast which was incubated overnight in a regeneration medium before being exposed to Au-ConA and fixed. In comparison to the fresh protoplast (Fig. 8) there is less label in close contact with the membrane itself and more label associated with stranded material away from the surface, x 15,000
tobacco protoplasts, but did not eliminate the ferritin labelling of the a m o r p h o u s material or wall fragments.
d) Colloidal Gold ConA Initial experiments with colloidal gold ConA mixtures (Au-ConA) gave similar results to those obtained with F e - C o n A ; tobacco protoplasts showed discontinuous and patchy labelling, whilst vine protoplasts showed almost no labelling at all. However, when the ratio of C o n A to gold in the mixture was reduced by a factor of four (see Methods) a different pattern emerged. Tobacco protoplasts treated with this mixture exhibited heavy and continuous labelling over the entire p l a s m a l e m m a surface (Fig. 8). Where protoplasts were aggregated in the presence of the AuConA, gold particles could be found between the paired p l a s m a l e m m a surfaces (Fig. 9). Vine proto-
plasts under the same conditions did not show saturation with gold particles, but rather a patchy and irregular distribution (Fig. 10). Increased times of incubation o f the vine protoplasts with the A u - C o n A did not change this labelling pattern. Prefixation of protoplasts in glutaraldehyde solutions before treatment with A u - C o n A resulted in an increase in the amount of label seen at the plasmalemma surface in thin sections. This increase in the labelling was most m a r k e d for vine protoplasts, which v~ould normally not show saturation of the surface (cf. Figs. 10, 11). The increase in labelling was greatest when washing after prefixation was minimised. When particular care was taken to wash prefixed protoplasts thoroughly, with m a n y changes of washing solution over 16 h, the increase in labelling due to prefixation was negligible in all cases. Incubation of protoplasts overnight in a nutrient medium known to support the regeneration of a wall around tobacco protoplasts (Burgess and Fleming,
78
1974b) resulted in a change in the pattern of labelling with Au-ConA. Tobacco protoplasts showed a discontinuous distribution of label which was predominantly in the form of clumps rather than single particles at the membrane surface (Fig. 12). Vine protoplasts incubated overnight showed a labelling pattern very similar to freshly prepared tobacco protoplasts; the membrane was more or less saturated with label, with a few clumps of gold particles attached to the protoplasts periphery. Various control experiments were carried out to establish the role of the ConA in the binding reaction. Binding was abolished by preincubating the Au-ConA with e-MM before use. No binding was observed at all when BSA was substituted for ConA. Colloidal gold on its own did not label the membrane surface but was occasionally seen attached to entrained wall fragments.
Discussion
These results confirm and extend a previous report (Glimelius et al., 1974) that ConA binds to the outer surface of the plant plasmalemma in such a way as to cause aggregation and the close approach of paired membranes. This aggregation reaction is abolished by the ConA inhibitor e-methyl-D-mannoside. Although no attempt has been made to measure the extent of aggregation in a quantitative manner, it is clear that tobacco protoplasts are aggregated more readily than are vine protoplasts. This is reflected in more rapid formation of aggregates at a given ConA concentration, and in greater degrees of visual aggregation at reduced ConA concentrations. For electron microscope study 250 gg/ml ConA gives a very high degree of aggregation within 30 rain at room temperature with both type of protoplasts; for visualization of binding patterns 125 gg/m! ConA gives saturating labelling after 30 min at room temperature, at least with tobacco protoplasts. The distribution of binding sites has been found to vary with both types of protoplast and with the label used. Fe-ConA gives a pattern of labelling which at first suggests a patchy distribution of sites such as has been found in a variety of animal cells (Nicolson, 1971 ; Amakawa and Barka, 1975 ; Garrido, 1975). This patchy distribution was also found using AuConA with a relatively high ratio of ConA:Au. Two interpretations of such behaviour are possible; either the distribution is patchy or it is more continuous and the label used to visualize it contains free ConA which masks a proportion of the sites. Decreasing the ratio of ConA:Au by a factor of four whilst maintaining the solution concentration of ConA at
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
125 gg/ml (see Methods) does indeed abolish the patchy labelling on the tobacco protoplast plasmalemma. Under these conditions it may be assumed that the great majority of the ConA molecules present in the Au-ConA mixture are associated with colloidal gold. The binding of the label to the plasmalemma is directly associated with the ConA since it is abolished by e-MM, and by absence of ConA in the mixture. Furthermore the binding is not restored by use of another protein, bovine serum albumin. The preparation of Au-ConA is far simpler and more reproducible than the preparation of Fe-ConA, and for this reason attention has been concentrated on the AuConA technique. Tobacco and vine protoplasts have consistently shown different behaviour towards ConA. Tobacco protoplasts are more easily aggregated by it and show more extensive labelling. This suggests a difference in the chemical environment of the plasmalemma surface of these two species. It is possible that such a difference might reflect the origins of the protoplasts themselves, since one is directly isolated from a normally growing tissue, the leaf, whilst the other derives from an undifferentiated tissue culture. However, there is a simple physiological difference between the two types of protoplasts related to their ability to regenerate a cell wall. Tobacco protoplasts readily and rapidly regenerate an intact wall (Burgess and Fleming, 1974b), whereas under the same conditions, vine protoplasts after 5 days culture show only a very tenuous and atypical wall (Linstead, unpublished). Both Au-ConA and Fe-ConA bind readily to wall fragments remaining after digestion during protoplast formation. It is therefore reasonable to suppose that labelling of the plasmalemma surface of freshly prepared protoplasts represents binding of ConA to nascent wall material. The difference between the two species is possibly therby explained by the relative absence of such nascent wall material from the vine protoplast surface. Incubation of protoplasts overnight produces a change in the labelling pattern which may also be explained in these terms. After overnight incubation, vine protoplasts show overall membrane-bound labelling, corresponding to a slow incipient wall formation. By contrast tobacco protoplasts show patches of heavy labelling with clumps of gold particles corresponding to more extensive wall formation. The absence of labelling from other regions would correspond to sites where wall material had been lost during processing, for electron microscopy, a common observation at such an early stage of wall formation (Burgess and Fleming, 1974b). There seems no need to invoke mobile ConA receptor sites as the underlying mechanism to explain
J. Burgess and P.J. Linstead: Protoplast Lectin Binding
the observed patterns of labelling in this study. Prefixation of cells may be used to immobilise such sites on animal cell surfaces and thus prevent their clumping (Rutishauser and Sachs, 1975). Attempts to reproduce this behaviour with vine protoplasts which show a patchy distribution of label even with the low ConA:Au ratio mixture, have failed. The general and increased labelling which follows prefixation seems to be related to the presence of residual aldehyde fixative. If careful washing procedures are adopted the pattern of labelling is restored to that in unfixed specimens. It is clear that by whatever mechanism binding takes place, labelling with ConA in conjunction with electron dense markers may provide a potentially useful tool to identify the fate of the plasmalemma in a variety of situations. Preliminary experiments show that the label is not removed by subsequent exposure to high concentrations of polyethylene glycol, for example, and this should allow further elucidation of events taking place during and after protoplast fusion.
References Amakawa, T., Barka, T. : Distribution of Concanavalin A binding sites on the surface of dissociated rat submandibular acinar cells. J. Histochem. Cytochem. 23, 607~517 (1975) Burgess, J., Fleming, E.N.: Ultrastructural studies of the aggregation and fusion of plant protoplasts. Planta (Berl.) 118, i83 I93 (1974a) Burgess, J., Fleming, E.N.: Ultrastructural observations of cell wall regeneration around isolated tobacco protoplasts. J. Cell Sci. 14, 439-449 (1974b) Dfizgfines, N.: The Concanavalin A agglutinating system of cell membranes. Biosystems 6, 209-216 (1975) Frens, G.: Controlled nucleation for the regulation of particle
79 size in monodisperse gold suspensions. Nature physical Sci. 241, 20~2 (1971) Garrido, J. : Uttrastructural labelling of cell surface lectin receptors during the cell cycle. Exp. Cell Res. 94, 159 175 (1975) Glimelius, K., Wallin, A., Eriksson, T. : Agglutinating effects of Concanavalin A on isolated protoplasts of Daucus carota. Physiol. Plant. 31, 225-230 (1974) Horisberger, M., Rosset, J., Bauer, H. : Colloidal gold granules as markers for cell surface receptors in the scanning electron microscope. Experientia 31, 1147-1149 (1975) Inbar, M., Huet, C., Oseroff, A.R., Ben-Bassat, H., Sachs, L.: Inhibition of lectin agglutinability by fixation of the cell surface membrane. Biochim. biophys. Acta (Amst.) 311,594-599 (1973) Klein, P.A., Adams, W.R. : Location of ferritin-labelled Concanavalin A binding to influenza virus and tumor celt surfaces. J. Virol. 10, 844-854 (1972) Nagata, T., Takebe, I. : Plating of isolated tobacco mesophyll protoplasts on agar medium. Planta (Berl.) 92, 301-308 (1971) Nicolson, G.L.: Difference in topology of normal and tumour cell membranes shown by different surface distributions of ferritin-conjugated Concanavalin A. Nature new biol. 233, 244-246 (197I) Nicolson, G.L.: Interactions of lectins with animal cell surfaces. Int. Rev. Cytol. 39, 89-190 (1974) Nicolson, G.L., Singer, S.J.: The distribution and assymetry of mammalian cell surface polysaccharides utilising ferritin-conjugated plant agglutinins as specific saccharide stains. J. Cell Biol. 60, 236-248 (1974) Roos, E., Temminck, J.H.M.: Cytochemical comparison between wheat germ agglutinin and Concanavalin A bound to mouse fibroblasts in vitro. Exp. Cell Res. 94, 140-146 (1975) Rutishauser, U., Sachs, L.: Receptor mobility and the binding of cells to lectin coated fibres. J. Cell Biol. 66, 76-85 (1975) Singer, S.J., Nicolson, G.L. : The fluid mosaic model of the structure of cell membranes. Science 175, 720-731 (1972) Spurr, A.R. : A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26, 31-43 (1969) Temminck, J.H.M., Collard, J.G., Spits, H., Roos, E. : A comparative study of four cytochemical detection methods of Concanavalin A binding sites on the cell membrane. Exp. Cell Res. 92, 307-322 (1975)
Received 18 December 1975; accepted 7 January 1976
