relationships between anionic sites and lectin...

J. Cell Set. 41, 89-104 (1980)Printed in Great Britain © Company of Biologists Limited 1080

RELATIONSHIPS BETWEEN ANIONIC SITES

AND LECTIN RECEPTORS IN THE PLASMA

MEMBRANE OF DICTYOSTELIUM DISCOIDEUM

AND THEIR ROLE IN PHAGOCYTOSIS

RAYMOND HELLIO AND ANTOINETTE RYTERUniti de Microscopic Electronique, Departement de Biologie Moliculaire,Institut Pasteur, 75724 Paris Cedex 15, France

SUMMARY

The disappearance of Wheat Germ Agglutinin (WGA) receptors from the membrane ofyeast-engulfijig-phagocytic cups in Dictyostelium suggested that these receptors could play arole in yeast adsorption or ingestion. This problem was approached by comparing the fate ofWGA, Concanavalin A (Con A) and cationized ferritin (CF) and their effects on the phago-cytosis of yeast, bacteria and latex beads. It can be concluded that CF capped in about 30 minand inhibited phagocytosis of any kind of particles for about 15 min. Con A capped in 20-60min and inhibited phagocytosis of all particles for 1 h 30 min. The time at which phagocytosisstarted to occur corresponded approximately to the moment at which large areas of plasmamembrane were totally devoid of marker. WGA did not cap but induced the formation oflarge and tight aggregates. The surface of the peripheral cells progressively released WGA in 1 h30 min. Afterwards, the cells were able to ingest latex beads and bacteria but did not phago-cytoze yeast. The latter started to be adsorbed onto the cells and to be ingested only 1 h later.

Double labelling experiments showed that CF and Con A receptors were still absent in theplasma membrane, when phagocytosis of any kind of particles started to occur. WGA-labelledcells ingested latex beads and bacteria when their plasma membrane was still devoid of WGAreceptors but were able to ingest yeast only after their regeneration. These observations stronglysuggest that WGA receptors may correspond to specific receptors for yeast phagocytosis.

INTRODUCTION

It has been shown in the accompanying paper (Ryter & Hellio, 1980) that yeastphagocytosis by amoeboid cells of Dictyostelium discoideum led to the disappearanceof Wheat Germ Agglutinin (WGA) receptors from the membrane of the phagocyticcup. The labelling of the phagosome membrane with this lectin also showed that thesereceptors remained absent or undetectable in this membrane throughout yeastdigestion. This phenomenon seemed specific for yeast phagocytosis because it wasobserved neither during ingestion of other particles such as latex beads or bacterianor during endocytosis of axenic medium. It is apparently also peculiar to receptorsof this lectin since it did not occur for Concanavalin A (Con A) receptors nor foranionic sites revealed by cationized ferritin or colloidal iron hydroxide. It thus seemsto correspond to a specific interaction between WGA receptors and yeast.

In the present paper, we have tried to determine to what extent the disappearanceof WGA receptors was actually related to the process of yeast adsorption or ingestion

7 CEL 41

90 R. Hellio and A. Ryter

and whether the presence of these receptors in the plasma membrane was indis-pensable for yeast phagocytosis.

This problem was approached by comparing the ability of cells labelled with WGA,Con A or cationized ferritin (CF) to phagocytoze different particles. By double- andcross-linkings with these markers, it was possible to see whether the different recep-tors comigrated or not, when they regenerated, and whether their presence wasrequired for phagocytosis of yeast, bacteria or latex beads.

MATERIALS AND METHODS

Culture conditions

The strain and culture conditions were those described in the accompanying paper (Ryter& Hellio, 1980). In all experiments, exponentially growing cells were transferred into 17 mMphosphate buffer made in Volvic water (commercial mineral water) when the culture celldensity had reached 2 x 106 cells per ml. The cells were incubated in this medium at 20 °Cunder mild agitation for 15 min to 2 h before being labelled, according to the purpose of theexperiment.

Labelling

Single labelling. The cells received 50 /tg/ml of Con A or WGA. They were incubated at20 °C under mild agitation for different times (15 min-25 h) and then fixed. In some experi-ments, the excess of lectin was removed by transferring the cells into fresh buffer 10 min afterlectin addition. Labelling with CF was performed with 25 /ig/ml for either 5 min at 20 °Cor 30 min at o °C. The cells were washed twice at o °C after different periods of incubationbefore being fixed. In some experiments, the excess of CF was eliminated by transferring thecells into fresh buffer after labelling.

Double labelling. The cells labelled with the first marker and transferred into fresh mediumwere incubated for various periods of time before being fixed with glutaraldehyde. They werewashed for 5 min once with cacodylate buffer, once with o-i % glycine solution and overnightin cacodylate buffer at 4 °C. They were relabelled with one of the markers or with colloidaliron hydroxide (CIH) (Gasic, Berwick & Sorrentino, 1968) as described elsewhere (Ryter& Hellio, 1980). In some experiments, the second labelling with lectins or CF was performedon unfixed cells.

Con A and WGA detection

The fixed cells labelled with Con A and WGA were incubated at 20 °C for 20 min with200 /tg per ml of mannosyl-ferritin conjugate or A^-acetyl-chitobiosyl-ferritin conjugaterespectively (Kieda, Delmotte & Monsigny, 1977). They were washed twice in cacodylatebuffer before being prepared for EM. In some experiments, Con A was revealed with peroxi-

Fig. 1. Thin section of a cell labelled with Con A and incubated for 1 h. Con A wasrevealed with HRP and DAB. Most of the Con A has capped in the uroid region;large contractiles vacuoles (ci>) start to deplete. Precipitates once extracellular poly-saccharide labelled with Con A are visible (arrows), x 8800.Figs. 2-4. Light-microscope pictures of cells labelled with Con A in the presenceof yeast (Figs. 2, 3) or latex beads (Fig. 4). Clusters of particles are seen either on onepole of the cells or in the region of cell adhesion of small cell aggregates, x 5500.Fig. 5. Thin section illustrating the adhesion of 2 yeast cells (y) on the Con A cap reg-ion. This phenomenon was observed between 05 and 1 h only when yeast was addedat the same time as Con A. Con A was revealed with HRP. x 20000.

Anionic sites and lectin receptors 91

J

cv

•A.


dase. The cells were incubated at 20 °C for 30 min with 100 fig per ml horseradish peroxidase(HRP) washed twice and incubated for 15 min in 005 mM cacodylate buffer pH 5-0 containing1 mg per ml DAB and 001 % H2Oa according to Malmgren & Olsson (1977).

Phagocytosis after labelling

The cells starved for 2 h were labelled with Con A, WGA or CF as described above. Eitherlatex beads, yeast or bacteria were added 10 min after the marker or the cells were spun down10 min after the marker addition and transferred into fresh phosphate buffer before receivingthe particles. They were incubated at 20 °C under mild agitation and regularly observed inlight microscopy. When particles started to adsorb on the cells, samples were fixed for electronmicroscopy.

Electron microscopy

Glutaraldehyde-osmium tetroxide fixation, embedding and sectioning were as describedpreviously (Ryter & Hellio, 1980).

Chemicals

CF, lectins and the ferritin-glycosylated conjugates were purchased as indicated in theaccompanying paper (Ryter & Hellio, 1980).

RESULTS

Morphological changes produced by marker addition

The addition of Con A produced cell rounding in a few minutes and the formationof small agglutinates. One to three large vacuoles appeared during the first 15 minand remained visible for about 1 h. They correspond to huge swollen contractilevacuoles (Fig. 1) which started to flatten after 1 h. Con A demonstration with peroxi-dase or ferritin-mannosyl conjugate showed that Con A capping (Fig. 1) started to bevisible 20 to 40 min after its addition, as already observed by other authors in Dictyo-stelium (Gillette, Dengler & Filosa, 1974; Molday, Jaffe & McMahon, 1976; Favard-Se"reno & Livrozet, 1979). The speed of this phenomenon varied, however, with thephysiological state of the cells. It was rather slow (40—60 min) in cells labelled justafter their transfer into phosphate buffer and much faster (15—30 min) after 1 to 5 hof starvation. In addition, polysaccharide material rejected by the cells during starv-ation (de ChasteUier & Ryter, 1977) reacted with Con A and appeared on these sectionsas ferritin-labelled extracellular filamentous material when Con A was revealed withferritin-mannosyl conjugate or formed dense precipitates when Con A was revealedwith peroxidase (Fig. 1). When low Con A concentrations were used (5—10/tg/ml)this material reduced the amount of Con A on the cell surface and also decreased

Figs. 6-8. Cell labelled with WGA and incubated for 10 min (cv, contractile vacuole)(Fig. 6), 60 min (Fig. 7), and 1-5 h (Fig. 8). Ferritin particles are very scarce at 60 minand no longer visible at 15 h. x 72000.

Figs. 9, 10. Double labellings with WGA. The cells were labelled with WGA andincubated for 1-5 h (Fig. 9) or 2-5 h (Fig. 10) before being fixed and relabelled withWGA. x 72000.

Aniomc sites and lectin receptors

CV

93

94 R- Hellio and A. Ryter

agglutination and capping processes. This probably explains why Weeks (1973)observed that agglutination of starved cells required 10 times more Con A thangrowing cells. To prevent Con A adsorption on this material, the cells were spundown and resuspended in fresh phosphate buffer before receiving Con A.

After a 2-h incubation, the cap seemed to detach from the cells as already suggestedby other authors (Gillette et al. 1974; Molday et al. 1976) while agglutinates dissoci-ated themselves. Some cells lysed between 1 and 1-5 h. They probably correspondto those which had not capped rapidly and in which contractile vacuoles had exploded.Observations of cells directly labelled with ferritin-Con A conjugate showed that ConA was very poorly endocytozed contrary to what was observed in macrophages orfibroblasts (Edelson & Cohn, 1974; Storrie & Edelson, 1977).

The cells labelled with WGA presented the same behaviour as those labelled withCon A, but aggregates were much bigger and tighter. WGA could not be revealedinside agglutinates because iV-acetyl-chitobiosyl-ferritin conjugate could not penetratebetween the cells, but the plasma membrane of the cells located at the periphery washomogeneously labelled (Fig. 6). During incubation, no patches or cap were formedbut the number of ferritin particles progressively decreased, corresponding to abouthalf of the initial value after 40 min of incubation. After 60 min, ferritin particleswere very scarce (Fig. 7) and after 1-5 h they were no longer visible (Figs. 8, 18).Some cells lysed between 1 and 2 h and the surviving cells formed long, thin processesemerging from the aggregates or creeping along the surface of neighbouring cells(Figs. 9, 15).

The addition of CF produced the same morphological changes as those observedwith Con A except that agglutinates did not form. When unfixed cells were labelledfor 5 min at 20 °C, their surface was irregularly labelled in contrast to cells labelledafter fixation, thus suggesting that a quick migration of CF sites occurred, as hasalready been found in animal cells (Skutelsky & Danon, 1976; King & Preston, 1977).In order to follow more accurately the capping process, cells were labelled 30 min ato °C, and then incubated at 20 °C. At the outset of the incubation at 20 °C, CFdistribution was almost homogeneous on the whole surface. After 15 min, largepatches were formed and after 40-60 min CF had capped and formed a thick ferritinlayer (Fig. 11) which started to detach from the cell surface or to be endocytozed(Fig. 14). The behaviour of Dictyostelium is thus identical to that observed in theamoeba Naegleria (King & Preston, 1977).

Figs. 11, 12. Cells labelled with CF, incubated for 30 min and relabelled with WGAafter (Fig. n ) or before fixation (Fig. 12). In both cases, WGA receptors werepresent in the cleared membrane. In Fig. 11, part of the CF cap is visible (arrow), itforms a thick layer of ferritin which tends to detach from the plasma membrane,x 72000.

Fig. 13. Cell labelled with CF, incubated for 30 min, fixed and relabelled with Con A.Con A was revealed with ferritin-mannosyl conjugate, x 72000.Fig. 14. Cell labelled with CF and incubated for 60 min. Part of the CF cap has beeningested and can be seen inside pinocytic vacuoles (p). x 44000.


13


Phagocytosis after labelling

All experiments were done on cells which had been starved for 2 h because at thattime uptake is higher (de Chastellier & Ryter, 1977) and more homogeneous among thepopulation than at the beginning of starvation. Generally, cells received yeast orlatex beads and more rarely bacteria because yeast and latex particles were easilyvisualized in light microscopy even after endocytosis, whereas ingested bacteria weremuch more difficult to see inside the cells. Nevertheless, no quantitative measure-ments of particle uptake could be made with yeast or latex beads mainly because ofthe formation of cell aggregates with Con A and WGA in which only peripheral cellscould phagocytoze particles.

Table 1. Effect of CF, Con A or WGA on phagocytosis

ControlParticles without

added CF Con A WGA labelling

YeastLatexBacteria

ISIS

mmmin

1

1•5•5

hh

2I

I

•5•5•5

hhh

2

2

2

minminmin

The cells received the particles 10 min after the addition of the marker and were observedwith the light microscope. The times at which particles started to adsorb on the cells are given.Controls were done either in the absence of marker or by adding a-methyl-mannoside withCon A or AT-acetyl-chitobiose with WGA.

When cells received yeast or latex beads at the same time as Con A, they remaineddevoid of particles for 15-30 min and then started to present clusters of particles inone region (Figs. 2-4). In cell aggregates, particle clusters were located in the regionsof cell contact (Figs. 3, 4). Observations on thin sections showed that no particleswere ingested and that clusters were located in the cap region (Fig. 5). After 1-1-5 hthey detached from the cells. This phenomenon was never observed in untreatedcells or when a-methyl-mannoside was added with Con A. It did not occur eitherwhen cells had been transferred into fresh buffer 10-15 min after Con A addition,before receiving yeast or latex. This particle adhesion was thus due to the linking ofCon A of the cap to Con A adsorbed onto the particles. This phenomenon was notobserved with WGA or CF.

The effects of the marker on the ability of the cells to phagocytoze particles werefollowed by light microscopy but samples were controlled with electron microscopywhen particle-adhesion started to occur. These controls showed that when yeast orbacteria started to adsorb onto the cell surface, they were immediately ingested afterany labelling. In the case of latex, only part of the adsorbed particles were phagocy-tozed as also observed in absence of marker (authors' unpublished observation).

As shown in Table 1, the addition of any of the 3 markers inhibited adsorption andphagocytosis. However, the inhibition period was very different from one marker tothe other. After CF addition, yeast and latex phagocytosis started after about 15 min


whereas after Con A labelling, both particles were ingested only after 1-5 h. AfterWGA labelling, the situation was a little different since latex and yeast ingestion didnot start at the same time. Latex beads were phagocytozed after 1-5 h as with Con A,but yeast was ingested 1 h later. Experiments made with bacteria showed that theywere ingested like latex beads at 1-5 h. In the presence of a-methyl-mannoside or./V-acetyl-chitobiose, phagocytosis occurred immediately and cells did not agglutinate.

These results show that adsorption and ingestion of particles did not take place solong as lectins or CF covered the majority of the cell surface. They started to occuronce Con A or CF capping was well pronounced or when WGA was totally eliminatedfrom the membrane of peripheral cells. However, the fact that after WGA labellingyeast ingestion took place much later than that of latex and bacteria suggested that thecleared membrane surface was not identical after WGA and Con A or CF addition.These observations led us to see what receptors were present on the cell surfaceafter capping with Con A and CF and after elimination of WGA. This study wasmade by performing 2 successive labellings either with the same marker to follow thereceptor regeneration or with another one to study the relationship between the differ-ent receptors.

Multilabellings

The cells were labelled with one of the 3 markers and incubated at 20 °C for 15 minto 2-5 h. They were fixed, carefully washed with glycine solution and relabelled withCon A, WGA, CF or colloidal iron hydroxide (CIH) and then prepared for electronmicroscopy. In some experiments the second marker was also applied to unfixedcells.

Table 2. Results of double labelling

Firstlabelling

CFCon AWGA

CF onfixedcells

CF onunfixed

cells

—

Second labelling

Con Aon

fixedcells

4-

1 4-

Con Aon

unfixedcells

+

WGAon

fixedcells

4- 4- 1

WGAon

unfixedcells

+

CIH onfixedcells

4-4

-4-

The cells were labelled with CF, Con A or WGA, incubated for 30 min after CF or 15 hafter the 2 lectins and then fixed. They were then relabelled with one of these markers or withCIH. The presence of the second marker on the cleared membrane is indicated by + and itsabsence by —.

The distinction on thin sections between the first and the second marker was veryeasy when CF was used as first marker because CF particles rapidly formed aggre-gates whose aspect was quite different from the thin labelling with Con A, WGA orCIH (Fig. 11). When Con A was used as first marker, the large villosities of the uroidregion to which the cap was associated were sufficient to localize the cap (Figs. 1, 5, 21)

15

17

k.

18

R. Hellio and A. Ryter

'f •**. '•4


and see whether or not the rest of the cell surface was labelled with the second marker.In some experiments, Con A was also revealed with peroxidase after the secondlabelling (Figs. 19-21). For WGA-labelled cells, the disappearance of this markerwas checked in parallel on an aliquot of cells which had not been relabelled.

Table 3. Double labelling made 2-5 h after the first one

Firstlabelling

CFCon AWGA

CF onfixedcells

+

Presence of the second marker

Second labelling made at

Con ACF on onunfixed fixed

cells cells

—

Con Aon

unfixedcells

—

on the cleared membrane isabsence by —.

2-5 h

WGAon

fixedcells

+

WGAon

unfixedcells

+indicated by + and its

Table 2 shows that cells first labelled with CF still bore Con A (Fig. 13), WGA(Figs, i i , 12) receptors and anionic sites revealed by CIH on their cleared membrane.This means that these carbohydrate residues did not comigrate with CF sites. Nodifferences in Con A or WGA reaction were observed between cells relabelled beforeor after fixation (Figs. 11, 12). This was not the case in the cells relabelled with CF(Table 2). The labelling was positive on cells labelled after fixation and negative inthose labelled before fixation. The same discrepancy was found in cells first labelledwith Con A or WGA and relabelled with CF (Fig. 15).

Table 2 also shows that the capping of Con A receptors was not accompanied bythat of WGA (Fig. 22) as already observed in scanning electron microscopy by Moldayet al. (1976). Conversely, the disappearance of WGA receptors was not followed bythat of Con A receptors (Fig. 17). Finally, CIH anionic sites were found after anymarker (Figs. 16, 19—21).

All these experiments were performed by relabelling the cells 30 min after CFaddition or 1-5 h after Con A or WGA addition. At these times none of the 3 receptorshad regenerated as indicated in Table 2 and shown in Fig. 9 for WGA except for CFlabelling of fixed cells. When the second labelling was done 2-5 h after the firstlabelling (Table 3), Con A receptors were still absent but WGA ones were present(Fig. 10).

Figs. 15-17. Cell labelled with WGA, incubated for 1-5 h and relabelled after fixationwith CF (Fig. 15), with CIH (Fig. 16), or with Con A and ferritin mannosyl conjugate(Fig. 17). x72000.Fig. 18. Cell labelled with WGA, incubated for 1-5 h and supplied with latex beads.Many latex beads (/) have been ingested and are visible in digestive vacuoles. As al-ready shown in Fig. 8, the plasma membrane is completely devoid of WGA. x 56000.

100

19R. Hellio and A. Ryter

j


DISCUSSION

This study yields new data on the relationship between Con A, WGA, CF and CIHon the effect of these markers on phagocytosis ability and more particularly on theimplication of these receptors in phagocytosis of yeast, latex beads and bacteria.

Single labelling experiments showed that the behaviour of CF, Con A and WGAreceptors was very different. CF receptors look very mobile since they capped within30 min. Con A ones were less mobile, but their mobility seems to depend upon thephysiological state of the cells. The elimination of these 2 markers occurred mostlyby the release of the cap into the extracellular medium and partially also by endo-cytosis of the capped CF. In contrast, WGA receptors did not cap but were progress-ively eliminated from the plasma membrane of the peripheral cells of the aggregates.

Cross-labellings led to the conclusion that the receptors for these 3 markers and CIHanionic sites all corresponded to independent molecules since after elimination of oneof them, the 3 others remained present on the cleared membrane. It is noteworthythat CF and CIH anionic sites were completely independent when they were revealedat pH 6-8 and i-8 respectively, as expected from our previous observations (Ryter &Hellio, 1980) and those of de Bruyn, Michelson & Becker (1978). As observed inEntamoeba histolytica (Pinto da Silva & Martinez-Palomo, 1974), Con A receptorsmoved independently of CIH sites. Also, CF sites and Con A receptors seemed tohave no relationship although both were symmetrically located on both faces of theplasma membrane (Ryter & Hellio, 1980).

The discrepancy observed with CF used as second label on unfixed or fixed cells isnot easy to interpret. As already pointed out in the accompanying paper (Ryter &Hellio, 1980), 2 reasons could explain these contradictory results. Either positiveresults obtained on fixed cells were due to unspecific anionic sites created by glutar-aldehyde fixation and not suppressed by glycine washing, or negative results observedon unfixed cells were due to some masking phenomenon by a soluble cytoplasmiccomponent released by lysed cells. However, the absence of CF anionic sites found onunfixed cells, whatever the marker used first, seems to contradict the fact that Con Aand WGA receptors were perfectly detectable in the cleared membrane of cells firstlabelled with CF. It is thus difficult to admit that CF anionic sites capped independ-ently from Con A and WGA receptors when CF was used first but disappeared from

Figs. 19-21. Cell labelled with Con A and relabelled with CIH after 1-5 h of incu-bation and fixation. Con A was revealed with HKP and DAB after CIH labelling.

Fig. 19. Enlargement of part of the plasma membrane outside the cap region. CIHlabelling is well visible, x 3460x3.

Fig. 20. Enlargement of the cap region showing the villosities strongly labelled withperoxidase. x 34600.

Fig. 21. General view of the cell, x 6500. Areas outlined are those shown enlargedin Figs. 19 and 20.Fig. 22. Part of the cleared membrane in a cell labelled with Con A and relabelledwith WGA after 1-5 h of incubation, x 72000.


the cell surface after Con A and WGA labellings. The assumption of a maskingphenomenon seems thus more plausible.

Double labelling further showed that CF or Con A receptors regenerated moreslowly than WGA ones. One can wonder whether the quick regeneration of WGAreceptors is related to their mode of elimination which did not occur by capping.

If we compare the ability of the cells to phagocytoze after addition of the markersand the results of double-labelling experiments, it can be concluded first that thepresence of any of the these markers on the cell surface completely prevented particleadhesion and ingestion. Our results with Con A are thus in good agreement with theobservations made on Acanthamoeba castellanii (Bowers, 1977) which showed thatendocytosis was inhibited by this lectin. However, contrary to what was proposed byBowers (1977) this inhibition in Dictyostelium does not seem to be due to the inhibitionof membrane fusion only, which normally occurs at the end of the engulfment process,but rather to the prevention of particle adsorption on the cell surface as suggested byBerlin (1972). In contrast, the swelling of the contractile vacuole was probably pro-duced by the impossibility of membrane fusion between the plasma membrane and thecontractile vacuole membrane, which under normal growth conditions leads to theformation of a pore (de Chastellier, Quiviger & Ryter, 1978) and fluid expulsion. Thisblockage of the contractile vacuole is not specific to Con A since it was also observedafter WGA or CF labelling. This means that the presence of any carbohydrate ligandon the plasma membrane prevents the fusion process.

Particle adsorption and ingestion did not occur so long as the markers had not beeneliminated from a large surface area. For latex beads, adsorption and ingestionoccurred 1-5 h after Con A or WGA addition, that is to say before the regenerationof these 2 receptors. It can be thus concluded that none of them are necessary forlatex bead adsorption or ingestion. Yeast ingestion occurred at the same time as thatof latex in cells treated with Con A or CF, on which WGA receptors were present, butwas delayed by 1 h in WGA-treated cells until WGA receptor regeneration. Thisdelay looks specific for yeast since ingestion of bacteria took place at the same time aslatex. Thus WGA receptors seem to be indispensable for yeast phagocytosis but notfor ingestion of latex or bacteria.

This suggests that the disappearance of WGA receptors from the phagocyticmembrane engulfing yeast (Ryter & Hellio, 1980) is not an accidental phenomenonproduced by the presence of yeast only, but corresponds to a process related to theadsorption or ingestion of yeast.

These results lead us to think that Dictyostelium discoideum as macrophages couldhave specific endocytic receptors (Ehlenberger & Nussenzweig, 1977; Griffin, Bianco& Silverstein, 1975; Silverstein, Steinman & Cohn, 1977). The presence of receptorsspecifically involved in yeast phagocytosis might account for previous observationsmade on yeast phagocytosis (Ryter & de Chastellier, 1977) in which it was observedthat Dictyostelium cells growing in axenic medium ingested yeast soon after addition.They rejected the undigested yeast cell wall after about 2-5 h but did not phagocytozenew yeast for several hours, preferring to pinocytoze the axenic medium. In contrast,this selected feeding did not occur in cells starved in phosphate buffer (authors'


personal observations). In this case, they continued to phagocytoze intact yeast until

its complete exhaustion but never endocytozed undigested cell wall rejected into the

medium. This indicates that the cells are able to choose their food. This choice could

be determined by the presence of different receptors which may be more or less

masked in axenic medium. There would be a sort of competition in axenic medium

between yeast and axenic medium components equilibrium of which may be modified

by the release of undigestible material into the culture medium.

REFERENCES

BERLIN, R. D. (1972). Effect of Con A on phagocytosis. Nature, New Biol. 235, 44-45.BOWERS, B. (1977). Comparison of pinocytosis and phagocytosis in Acanthamoeba castellanii.

Expl Cell Res. no , 409-417.BRUYN, P. P. H. DE, MICHELSON, S. & BECKER, R. P. (1978). Non random distribution of

sialic acid over the cell surface of bristle-coated endocytic vesicles of the sinusoidal endo-thelium. J. Cell Biol. 78. 379-389.

CHASTELLIER, C. DE, QUIVIGER, B. & RYTER, A. (1978). Observations on the functioning of thecontractile vacuole of Dictyostelium discoideum with the electron microscope. J. Ultrastnict.Res. 62, 220-227.

CHASTELLIER, C. DE, & RYTER, A. (1977). Changes of the cell surface and of the digestiveapparatus of Dictyostelium discoideum during the starvation period triggering aggregation.J. Cell Biol. 75, 218-236.

EDELSON, P. J. & COHN, Z. A. (1974). Effects of Concanavalin A on mouse peritoneal macro-phages. II . Metabolism of endocytized proteins and reversibility of the effects by mannose.J. exp. Med. 140. 1387-1403.

EHLENDERGER, A. G. & NUSSENZWEIG, V. (1977). The role of membrane receptors in phago-cytosis. J. exp. Med. 145, 357-371-

FAVARD-SERENO, C. & LiVROZET, M. (1979). Plasma membrane structural changes correlatedwith the acquisition of aggregation competence in Dictyostelium discoideum. Biol. cell. 35,45-54-

GASIC, G. J., BERWICK, L. & SORRENTINO, M. (1968). Positive and negative colloidal irons ascell surface electron stains. Lab. Invest. 18, 63-71.

GILLETTE, M. V., DENGLER, R. E. & FILOSA, M. F. (1974). The localization and fate ofCon A in amoebae of the cellular slime mold Dictyostelium discoideum. J. exp. Zool. 190,243-248.

GRIFFIN, F. M. J., BIANCO, C. & SILVERSTEIN, S. C. (1975). Characterization of the macro-phage receptor for complement and demonstration of its functional dependence from rec-eptor for the FC portion of immunoglobulin G. J. exp. Med. 141, 1269—1277.

KIEDA, C , DELMOTTE, F. & MoNSiGNY, M. (1977). Preparation and properties of glycosylatedcytochemical markers. FEBS Letters, Amsterdam 76, 257-261.

KING, C. A. & PRESTON, T. M. (1977). Studies of anionic sites on the cell surface of theamoeba Nacgleriagruberi using cationizedferritin. J'. Cell Set. 28, 133-140.

MALMGREN, L. & OLSSON, Y. (1977). A sensitive histochemical method for light and electronmicroscopic demonstration of horse radish peroxidase. J. Histochem. Cytochem. 25, 1280-1283.

MOLDAY, R., JAFFE, R. & MCMAHON, D. (1976). Concanavalin A and Wheat Germ Agglutininreceptors on Dictyostelium discoideum. J. Cell Biol. 71, 314-322.

PINTO DA SILVA, P. & MARTINEZ-PALOMO, A. (1974). Induced redistribution of membraneparticles, anionic sites and Con A receptors in Entamoeba histolytica. Nature, Lond. 249,170-171.

RYTER, A. & CHASTELLIER, C. DE (1977). Morphometric and cytochemical studies of Dictyostel-ium discoideum in vegetative phase. Digestive system and membrane turnover. J. Cell Biol.75, 200-217.

RYTER, A. & HELLIO, R. (1980). Electron-microscope study of Dictyostelium discoideum plasmamembrane and its modifications during and after phagocytosis. J. Cell Set. 41, 75-88.

104 R- Hellio and A. Ryter

SILVERSTEIN, S. C, STEINMAN, R. M. & COHN, Z. A. (1977). Endocytosis. A. Rev. Biochem.46, 669-722.

SKUTELSKY, E. & DANON, D. (1976). Redistribution of surface anionic sites on the luminal frontof blood vessel endothelium after interaction with polycationic ligand. J. Cell Biol. 71,232-241.

STORRre, B. & EDELSON, P. J. (1977). Distribution of Con A in fibroblasts: direct endocytosisversus surface capping. Cell 11, 707-717.

WEEKS, G. (1973). Agglutination of growing and differentiating cells of Dictyostelium discoideumby Concanavalin A. Expl Cell Res. 76, 467-470.

{Received 13 June 1979)

relationships between anionic sites and lectin...

Documents