Planta

Planta (1989) 177:169-177

9 Springer-Verlag 1989

Characterization of the isolated calcium-binding vesicles from the green alga Mougeotia scMaris, and their relevance to chloroplast movement* Franz Grolig and Gottfried Wagner Botanisches Institut I der Justus-Liebig-Universitfit, Senckenbergstrasse 17 25, D-6300 Giessen, Federal Republic of Germany

Abstract. The calcium-binding vesicles from the green alga Mougeotia scalaris were isolated and characterized after staining in vivo by neutral red or rhodamine B. They were found to possess a protonated group with a pK~ = 9.9, typifying phenolic hydroxyl groups; upon titration, both, phenolic compound(s) and vital dye were concomitantly released from the vesicular matrix. A shift in peak absorbance from 450 nm to 540 nm of the vitally stained vesicles indicated that the neutral form of neutral red was bound to the vesicular matrix as an intermediate form, stabilized via intermolecular hydrogen bonds to the phenolic compound(s). Up to 8.5. I 0 9 dye molecules were calculated to be adsorbed to a mean-size vesicle. Analysis of Langmuir adsorption isotherms indicated that there were two binding sites each for both neutral red and rhodamine B. The isolated vesicles were devoid of calcium, probably because vesicular calcium, bound to the vesicle matrix, was displaced upon dye binding. Dye adsorption to the vesicles in vivo results in substantial inhibition of the reorientational movement of the Mougeotia chloroplast and is explained by dye-mediated disorder of the cellular calcium homoeostasis. Key words: Calcium (vesicles, binding) - Chlorophyta - Chloroplast movement Mougeotia (calcium vesicles) - Phenolic compounds - Vesicle (vital staining, calcium)

Introduction The ribbon-shaped chloroplast in the cylindrical cell of Mougeotia performs, in toto or in part, reor* This paper is part of the Ph.D. thesis of F. Grolig at JustusLiebig-Universitfit Giessen, F R G

Abbreviations: N R = neutral red, RB = rhodamine B, SDS = sodium dodecyl sulfate

ientational movements with respect to the quality and direction of light as perceived by the sensor pigments phytochrome and/or the blue-light pigment (Haupt 1987; Grolig and Wagner 1988). Inhibition of chloroplast movement by calmodulin antagonists and identification of this calcium target protein in Mougeotia (Wagner et al. 1984; Serlin and Roux 1984) indicates that a shift in the intracellular concentration of free calcium is part of the sensory transduction chain in this alga. So, much in analogy to animal nonmuscle cells, current hypotheses claim calcium as a second messenger to transduce the light signal to the actinbased motor apparatus (Wagner et al. 1972; Haupt and Weisenseel 1976; Klein et al. 1980; Roux 1984; Grolig and Wagner 1988). Upon photoactivation of phytochrome and-or the blue-light pigment, Ca 2 + was suggested to enter the cell via the plasma membrane (Dreyer and Weisenseel 1979; Roux 1984) or to be released from numerous cytoplasmic, calcium-containing vesicles, most frequently located at the chloroplast edge (Wagner and Klein 1978; Wagner and Rossbacher 1980). Calcium binding in vivo to these vesicles was indicated by the fluorescent pattern from the calcium-indicator dye chlorotetracycline and was proven in situ by X-ray microanalysis of fixed and of unfixed frozen-hydrated Mougeotia cells (Wagner and Rossbacher 1980). Isolation and purification of these calcium-binding vesicles from Mougeotia was possible only after their in-vivo staining by the vital dyes neutral red (NR) or rhodamine B (RB) (Grolig 1986; Grolig and Wagner 1987). This staining apparently stabilized reversible binding of a phenolic matrix component to a vesicular core. In this paper, the calcium-binding vesicles are characterized both in vivo and in vitro by studying (i) the vesicle structure, as dependent on dye-mediated stabilization of the matrix, and (ii) the consequences of vesicle staining on chloroplast movement. The relevance of the vesicles to cellular calci-

170

F. Grolig and G. Wagner: Isolated calcium-binding vesicles from Mougeotia

um homoeostasis and chloroplast reorientational movement in Mougeotia are discussed.

Material and methods Plant material and isolation of vesicles. Mougeotia scalaris, as classified by means of zygotes of the "Erlangen" strain (Haupt 1970), was grown on a large scale (Russ et al. 1988), and calcium vesicles were isolated after vital staining of algal filaments by N R or RB (Grolig and Wagner 1987). After homogenization of the cells using a nitrogen cell-disruption bomb (Parr Instrument Co., Moline, Ill., USA), vitally stained vesicles were purified from Miracloth-filtered homogenate by differential centrifugation at 700-g for 10 min in a swing-out rotor. Dye binding and stability of the vesicles. Absorption spectra of the vesicles in 0.6 tool. 1- t mannitol, 10 mmol. 1- t glycylglycine-KOH, pH 7.5, were recorded with an integrating sphere (91-00247), connected to an Uvikon 810 spectrophotometer (Kontron, Eching, FRG), 1 h after dilution (1 : 100) of the vesicle stock suspension (10 s vesicles-ml 1). For registration of the absorption spectrum of the suspension-medium under equilibrium conditions, vesicles of the stock suspension were pelleted (8000.g, 3 rain) and the supernatant was recorded after sixfold dilution with suspension-medium (see above). For determination of the total amount of dye bound to the vesicles, I0 gl of vesicle stock suspension (see above) were washed quickly in 1 ml 0.6 mol.1-1 mannitol, 10 mmol.l 1 glycylglycine-KOH, pH 7.5, and pelleted at 8000 .g for 3 rain. Dye was extracted within 30 min by resuspension of the vesicles in 1 ml of 1% (w/v) sodium dodecyl sulfate (SDS) for N R or 1% (w/v) octyl phenoxypolyethoxyethanol (Triton X-100) for RB, dissolved in 10 mmol.l-1 glycylglycine-NaOH/KOH, pH 7.5. Under identical conditions, the extinction coefficients were determined as 2.96.104 1. mol-~, cm-1 for N R at 540 nm and 9.75-104 1 . m o l - l . c m -1 for RB at 553 nm, and used for photometric determination of dye concentration in the centrifuged clear supernatant (11000-g, 3 min). Stability of the calcium-binding vesicles strongly depends on the binding of vital dye (Grolig and Wagner 1987). Consequently, we were able to investigate the parameters of dye binding by photometric measurement of vesicle disintegration, monitored as the decrease in scattered light at wavelengths of no spectral absorbance of the dye (750 nm). For titration, aliquots of the vesicle stock suspension were diluted by 0.6 mol.1-1 mannitol, 50 mmol.l-~ KC1 to result in an apparent absorbance of 1 at 750 nm, equivalent to about 6.106 vesicles-m1-1. Vesicles then were pelleted and quickly washed twice in the dilution medium. After resuspension and equilibration for 1 h, 2 ml of vesicle suspension were titrated in a plastic macrocuvette by stepwise addition of 10 lal each of NaOH or HCl solutions (0.05, 0.1, 0.2, 0.8 and 2.0 N). During stirring for at least 20 s, the pH was measured in the cuvette by an electrode with 3-mm tip diameter (U 402-M3; Ingold, Frankfurt, FRG), followed by measurement of apparent absorbance at 750 nm (Uvicon 810 spectrophotometer). The amount of vital dye and phenolic compound(s) released from the vesicles upon alkalinization, with or without subsequent neutralization, were quantitated in neutralized supernatant after centrifugation (8000.g, 3min; 0.6tool.1 1 mannitol, 50 mmol. 1-1 NaC1). The concentration of N R was photometrically determined as above by mixing 0.5 ml of supernatant with 0.5 ml 2% (w/v) SDS in 20 retool. 1-1 glycylglycineNaOH, pH 7.5 to shift quantitatively the N R to its strongerabsorbing, cationic form. The concentration of phenolic compound(s) was assayed by use of the Folin-Ciocalteau reagent

as described by Grob and Matile (1979): 150 gl supernatant were mixed with 890 ~tl of 2% (w/v) Na2CO3, 90 gl I N NaOH and 10 gl of 2% (w/v) K-Na-tartrate, and incubated at room temperature for 10 rain. Folin-Ciocalteau reagent (90 gl I N) was added to the clear solution, and the absorbance at 750 nm was determined after 30 rain. Absorbance was corrected for N R which gave slight interference in the assay (A e75o = 1.82. 103 mol.1 1.cm-1). To determine the concentration dependence of dye accumulation by Mougeotia cells, algal filaments equivalent to about 15 gg chlorophyll a + b were exposed to dye concentrations ranging from 10-5 to 10-9 mol. 1-1 for N R and RB, respectively, in 200 ml culture medium; pH was adjusted by I mmol-14-(2-hydroxyethyl)-l-piperazine ethane sulfonic acid (Hepes)KOH to 7.5 assuming a similar cytoplasmic pH. After incubation in dye for up to 48 h, to reach equilibrium, the external solution was carefully aspirated and filaments were collected on 20-~tm nylon mesh. The dye content of the algae was determined as described in the next paragraph; algal chlorophyll a + b was extracted in 1 ml methanol and determined according to Senger (1970).

Vital staining and chloroplast movement. The effect of vital staining on reorientational movement was studied as described previously (Wagner et al. 1984). Under dim-green safelight, 0.2 ml of dye solution (10, 40, 80, 120, 160, 200 and 250/xmol-1-1 NR) in growth medium, buffered to pH 6.0 by 10 mmol-1-t KH2PO4, was sucked through the preparation under the microscope coverslip, and cells were stained. Chloroplast reorientational movement was induced by a red light pulse of 60 s passing through a filter consisting of 3 mm of OG 590 nm and 4 mm of K G 1 (Schott, Mainz, FRG), equivalent to a quantum flux of 8.5 g m o l . m - 1 . s - 1 , gained from a double interference line pass filter with 2~,~x= 683 nm (Haupt 1970; Wagner et al. 1984). After 40 min in darkness, the percentage of chloroplasts in a face-on position was evaluated and the viability of the cells was established by subsequent plasmolysis in 0.7 mol. 1-1 sorbitol. To estimate the actual cellular dose of N R or RB, dye taken up by the filaments was extracted within 30 min in 1-3 ml detergent solution (NR: 2% (w/v) SDS, RB: 2% (w/v) Triton X-100, both in 10mmol-1-1 glycylglycine-NaOH or KOH, pH 7.5). Only minimal amounts of chlorophyll escaped through the cell wall, although the cytoplasmic content of the filaments was solubilized by the detergent. After determination of released dye (see above), the filaments were pelleted (11000. g, 3 min) and pipetted onto filter paper, and the amount of chlorophyll a + b was determined as described above. Electron microscopy and X-ray microanalysis. Isolated unfixed, vitally stained vesicles in 10mmol-1-1 glycylglycine-KOH, pH 7.5, kept either untreated or treated with detergent or methanol, were put onto a discharged formvar/carbon-coated copper grid and examined at 80 keV in an electron transmission microscope (EM 302; Philips, Eindhoven, The Netherlands). The circumference of these vesicles was evaluated from electron micrographs by means of a semiautomatic picture analyzer (ASM; Leitz, Wetzlar, FRG). X-ray microanalysis of unfixed vitally stained vesicles (see above) was done up to 10.24 keV by use of a X-ray (Kevex Co., Foster City, Cal., USA) microanalyzer coupled to a Hitachi H-600 scanning electron microscope (Nisseisangyo Co., Tokyo, Japan). Spot analysis (0.~0.5 gm beam diameter) took place for 100 s at 100 kV acceleration voltage with a beam current of 20 rtA. For freeze-etching, purified vesicles in 0.6 mol. 1-1 mannitol as cryoprotectant were fixed in 2% (v/v) glutaraldehyde, 0.15 mol. 1-1 cacodylate-KOH, pH 7.2, for 10 rain at 25 ~ C, fol-

F. Grolig and G. Wagner: Isolated calcium-binding vesicles from Mougeotia lowed by 2 h at 4 ~ C. Small aliquots of vesicle pellet were frozen quickly on gold grids (2 mm diameter) by immersion in thawing nitrogen. After fracture and etching with a Balzers (Liechtenstein) BAF 300, the fracture face was shadowed with platinum at an angle of 35 ~ and the replica stabilized by carbon from above. The organic material was dissolved by 40% chromic acid before viewing the replica in the electron microsope (Moor and Miihlethaler 1963; Volkmann 1981).

I71

LO

| 7

0,8-

.//""

~0.6-

\',.,

04u > 0.2-

Results

Close to neutral pH, vitally stained calcium-binding vesicles, isolated from Mougeotia sca/aris, were stable for weeks provided the equilibrium concentration of N R or RB in the external solution was kept constant. However, when the dye in the suspension was diluted the vesicles started to disintegrate (Fig. 1). A shift of the external pH into the alkaline or acidic range resulted in enhanced disintegration of the calcium vesicles (Fig. 2 a). This effect was most pronounced around pH 10 and, consistently, titration indicated the presence of a protonated group with a pK~ of 9.9 in the vesicular matrix (Fig. 2b). Absorption spectra of both the vesicular suspension and the clear suspension medium in equilibrium with the vesicles showed maximum absorbance at 260 nm; an additional maximum at 540 nm occurred in the presence of the NR-stained vesicles (Fig. 3a). Comparison of the spectra with pH-dependent absorption spectra of pure N R (Fig. 3 b), dissolved in the same medium, indicated that (i) the concentration of N R in equilibrated suspension-medium must have been beyond the limit of spectroscopic detection at 450 nm, and that (ii) ab-

'

:5 '0.30 1

1

m

"0 0.25

' "\,..O.o

0

>

t / 2 =5mm9 - " ~ o - o - , - , - . _ o _ o _ o _ , _ .

0.20

, 0

I

10

I

I

I

20

30

40

time/min Fig. 1. Disintegration of NR-stained calcium-binding vesicles, isolated from Mougeotia scalaris, upon dilution of the equilibrium concentration of N R in the external medium. The decrease in vesicular density was monitored as the decrease in apparent optical density (i.e. turbidity) at 750 nm immediately after 16fold dilution of an equilibrated suspension of 3-10 v vesiclesml - 1

r I

i

2

i

3

~

4

i

~) t0

/

9

[

6 pH

7

i

8

i

9

I

~ 0

111

12

,.~,7,~~

............ ~

pH

i

5

.

.

.

.

.

.

.

.

.

.

p_K_:y?. .

.

.

.

.

.

.

.

.

.

.

!

/

1

0

I

0.1

I

02

p m o l N e O H 9 y e s -1

Fig. 2. a. Disintegration of NR-stained calcium-binding vesicles isolated from Mougeotia scalaris as a function of a stepwise shift in pH of the suspending medium. Arrows at neutral pH indicate the start and direction of vesicle titration for a density of 6. I06 vesicles-m1-1. For further details, see Fig. l . b Buffer capacity in the alkaline range of NR-stained calcium-binding vesicles, isolated from M. scalaris. The titration curve for a 1.5-ml suspension of 107 vesicles.m1-1 runs through a point of inflection (hatched line), indicating a pK, of 9.9

sorption from 250 to 400 nm, as seen in the absence of photometrically detectable amounts of NR, was caused by a component(s) distinct from N R (Fig. 3 a; curve 3). The absorbance maximum at 540 nm of the N R vitally stained vesicles (Fig. 3a; curve i)coincided with that of the red, cationic form (NR +) of N R (Fig. 3b, curve 3); N R binding to polyphenolic acid at pH 7.5 showed a similiar absorbance maximum (Fig. 3 b, curve 2). Vesicular disintegration upon increasing the pH (Fig. 4, Turbidity) was accompanied by release of N R and phenolic compound(s) (Fig. 4, NR, Phe) to the suspension medium. The desorption of both compounds was largely reversible by reestablishing the neutral pH (Fig. 4; NR, Phe), but the vesicular disintegration was irreversible and an amorphous precipitate appeared (not shown).

F. Grolig and G. Wagner: Isolated calcium-binding vesicles from Mougeotia

] 72 3.1nm I

forward titration (o)Ipmo[ NaOH

300

400

500

600

700

800

I

I

I

I

I

I

.10

1.0 N o

0.8

~/nm

> 03D5

10-

/~00

I

500 I

I

600 I

~J

~

E c

o I_o o .Q

0,6

,'

0

E

_o ,

15

20

25

,

,

,

l

Turbidity

\

02

400

10

O 0

o05-

~_/nm

5

0./~

~176176

g

~3;0 ~;o s;0 6;0 7;0 8;0

0 ,

i

,

,

500

,

i

600

~/nrn

Fig. 3. a. Spectral absorption of an equilibrated suspension of NR-stained Mougeotia vesicles (3.106.m1-1; curve 1), of 20 pmol- 1-1 NR (curve 2) and of suspension medium in equilibrium with 3.108 vesicles-ml 1, diluted sixfold (curve3). All spectra were registered in 0.6 m o l d - 1 mannitol, 10 mmol-1-1 glycylglycine-KOH, pH 7.5. A shift in absorbance maximium of NR at pH 7.5 from 450 to 540 nm is clearly seen by comparing curve 1 with curve 2; no NR, but UV-absorbance is detectable in equilibrated suspension medium (curve 3). b Spectral absorption of NR (9 gmol.l-1) in the absence (curve i) and in the presence (curve 2) of 2 mg. m l - ~ tannic acid in the suspension medium containing i0 retool.l- a glycylglycine-KOH, pH 7.5. Curve 3 (control) shows the spectral absorption of neutral red (9 pmol.1 -~) at pH 5.0, in the absence of tannic acid. A shift in absorbance maximum, i.e, from 450 to 540 nm, is clearly seen (curve 1 versus curve 2), as is a change in the extinction coefficient (curve 2 versus curve 3)

Vesicles, isolated in the presence of N R or RB, readily disintegrated in organic solvents such as methanol or in detergents such as Triton X-100 or SDS. There was an inverse correlation between the time course of decrease of apparent optical density, due to vesicular disintegration, and the increase of the cationic form of N R in SDS (Fig. 5 a). Through the discriminative absorbance maxima at 450 nm and 540 nm, respectively, absorption spectra after disintegration of NR-stained vesicles indicate that the dye molecules were incorporated in the neutral form (NR ~ into the neutral Triton X-100-micelles, and in the charged form (NR § into the negatively charged SDS-micelles (Fig. 5b). Dye adsorption of N R or RB by the vesicles in vivo showed pseudosaturation (Fig. 6a). Langmuir adsorption isotherms demonstrated two vesicular binding sites for each dye (Fig. 6 b) with adsorption coefficients of 2.04. ] 0 9 1. m o l - 1 and 2.76. 105 l-tool -1 for NR. The vitally stained vesicles appeared homogenously electron dense in the electron microscope without further fixation (Fig. 7 a). Measurement of

'-o..__.~o

.o

NR

0.8

o ~n t-

0.6

0.4

/IE 1

.C}

o.2

. . . . . . . . . . .

0

0

'he

......

0 .....

"-<3

-~ 0,8 eo

O,g 0.4 .a a

/,""'[ .........

0.2 0

I

I

I

I

l

I

0

5

10

15

20

25

backword titration (e)/pmol HCL

Fig. 4. Disintegration of NR-stained calcium-binding vesicles (Turbidity), isolated from Mougeotia scalaris, upon titration from neutrality into the alkaline range, is accompanied by desorption of both neutral red (NR) and phenolic compound(s) (Phe) from the vesicle core (forward titration; o). Re-adsorption of both, Phe and NR, to the vesicular core is seen upon backward titration to regain neutrality (e), while the vesicular disiutegration turned out to be irreversible, and an amorphous precipitate appeared (not shown)

the vesicular circumference showed a Gaussian distribution (Fig. 7 b) with a mean vesicular circumference of 4.64+ 1.17 pm (SD), which is comparable to similar determinations in situ (Rossbacher 1982). Figure 7c shows an enlarged view of the smooth surface of an NR-stained calcium-binding vesicle when the whole mount is positively stained with 1% (w/v) phosphotungstic acid for 30 s. Isolated, NR-stained calcium-binding vesicles, viewed in the scanning electron microscope without further fixation (see Material and methods), are seen in Fig. 7 d (compare with Fig. 7 a). Treatment with detergent such as SDS resulted in extraction of N R from the now-swelling vesicular matrix and the vesicular periphery showed diminished electron density (Fig. 7 e). Total dye extraction in methanol

F. Grolig and G. Wagner: Isolated calcium-binding vesicles from Mougeotia lO-

CNR. f r e e / i J m o l

@

0

08

NR

06

~

!,o ;, E 8

\

0

05 CRB,free/mmot,

--

g

|

l|

7g a~

0t.

.

m

L.

2O

I

E

r ~06 9

Vu

10

!

" 9

.

I73

2

x~

1 t -I

2 .t~ tr",

0.2 0

o o 02

I

2

3

/-

(dye bound} " (Cdye. free)'~/10 2 t . ( m g chL )-~

o

ve.~

0

o

o

o

i

o

i

i

400

500

~

|

J

00

6

2/rim o

o

o

o

e

o

o

time/rain Fig. 5. a. Disintegration of NR-stained calcium-binding vesicles isolated from Mougeotia scalaris (yes) in 2% (w/v) SDS, and the concomitant appearance of the cationic form of neutral red (NR) in the SDS micelles of the suspending medium, b Spectral absorption of the suspension of NR-stained vesicles, disintegrated in the neutral detergent Triton X-100 (TX-IO0) or the anionic detergent SDS. In the presence of SDS, cationic neutral red (NR +) shows an absorbance maximum at 540 nm; in the presence of Triton X-100, uncharged NR shows an absorbance maximum at 450 nm

resulted in a substantial loss of electron density, but the spherical shape was unchanged (Fig. 7 f). The dense r e m n a n t probably represents the vesicle core after removal of the reversibly b o u n d N R and phenolic constituents of the vesicle matrix. After dye extraction, the vesicles' tendency to aggregate increased. The fracture plane of frozen N R - s t a i n e d vesicles through central parts of the vesicular sphere showed an almost flat cleavage surface after etching (Fig. 8 a). A belt of h o m o g e n e o u s matrix material corresponded with the dense osmiophilic ring seen in isolated vesicles after thin-sectioning (see Fig. 4 in Grolig and Wagner 1987); the central core appeared to be rough and less dense after freezefracturing (Fig. 8 a) and thin sectioning, respectively. A membrane-like structure, resembling the plasm a l e m m a and the tonoplast in fine detail, was discernable in thin sections o f the calcium-binding vesicles in situ, and enclosed the electron-dense vesicular matrix (Fig. 8 d). This structure possibly corresponds with both the s m o o t h surface layer studded with small particles, clearly distinguishable from the rough vesicular matrix after freeze etching

Fig. 6. a. Amount of neutral red (NR) and rhodamine B (RB), taken up into Mougeotia cells, as a function of the extracellular dye concentration in equilibrium (CNR, fr~eor CRB,free)- Note that the concentration ranges are different for NR and RB. b The Langmuir adsorption isotherms from a, replotted as dyebou.d versus (dyebound)"(Cdyee~ee)-1. Two different binding sites are seen for each dye, with adsorption coefficientsof 2.04. 109 1.tool- 1 and 2.76. l0 s 1-tool- 1 for NR

(Fig. 8b), and the matrix-surrounding membranelike structure in Fig. 8 c. In this context we recall that the vesicles, but no other organelle, rapidly started swelling after mechanical rupture of the protoplast in osmotically adjusted suspension medium (Fig. 5 d in Grolig and Wagner 1987). This behaviour is not expected for tightly m e m b r a n e - b o u n d e d organelles, and swelling of the calcium-binding vesicles apparently is based mainly on desorption of reversibly binding matrix components in the absence of vital dye. After vital staining by N R or RB and the subsequent process of isolation, the (unfixed) calcium vesicles were mostly devoid of calcium as revealed by X-ray microanalysis (Fig. 9). A minor peak at 3.690 keV, specific for calcium, cannot be distinguished from the kr for potassium at 3.589 keV, which would be expected to be of this height relative to the k~-peak for potassium at 3.312 keV. The capability for chloroplast m o v e m e n t declined with increasing adsorption of vital dye to the vesicles (Fig. 10). Half-maximal inhibition of chloroplast m o v e m e n t was observed close to 4.10 -6 tool N R . (mg chlorophyll)-1 without a decrease o f viability, as tested by subsequent plasmolysis. Full inhibition o f chloroplast m o v e m e n t was achieved at about 8-10 .6 mol N R . ( m g chlorophyll)- 1, which was about two-thirds o f the maximal loading capacity of the vesicles with N R (see Fig. 6b). At this load, the viability of the cells had

174

F. Grolig and G. Wagner: Isolated calcium=binding vesicles from Mougeotia

6O

5oi >./,o

/

u

g o-30

/

/

//\

\

ca

,op

2o

o b

1./~7

2.71

3.96

5.20

6./-/.

768

circumference/pm

Fig. 7. a. Electron micrograph of unfixed, NR-stained, isolated, calcium-binding vesicles from Mougeotia scalaris, viewed as a whole mount in the transmission electron microscope (TEM). B a r = l txm; x 8000. b Gaussian distribution of vesicle circumference from preparations as in a. e Enlarged view of smooth surface of a single vesicle when positively stained by 1% (w/v) phosphotungstic acid for 30 s. Bar=0.1 Ixm; x 115000. d Whole mount of isolated, NR-stained vesicles viewed in the scanning electron microscope. Bar=0.5 gm; x 11000. e, f. Micrographs showing loss of electron density of vesicles in TEM upon partial dye extraction by SDS (e) or complete dye.extraction by methanol (f). Bars = 1 gm (e), 0.2 gm (f); x 13 000 (e) ; x 45000 (f)

decreased by about one-third. Figure 10, however, shows the impediment of chloroplast movement in viable cells only. Discussion

Through its stabilizing effect, vital staining allowed the isolation and purification of calcium vesicles from Mougeotia (Grolig and Wagner 1987). Studies on the stability of isolated vesicles, as described here, were done to obtain greater insight into the vesicular structure and the mechanism of dye binding. Questions on the mechanism of dye binding are linked to questions concerning (i) the dye-mediated stabilization of the vesicles and (ii) the possi-

ble consequence of their staining in vivo on reorientational movement of the chloroplast. Binding of NR to the isolated vesicles was stable when the ionic strength of the suspension medium was increased (Grolig and Wagner 1987), provided the equilibrium concentration of the neutral form of NR in the suspension medium was kept constant (Fig. 1). Progressive vesicle disintegration upon lowering the pH of the suspension medium (Fig. 2 a) is explained by the shift of the prototropic equilibrium of NR to the cationic forms. Disintegration of vesicles in detergents such as Triton X100 and SDS probably works in a similar way by lowering the concentration of free NR ~ in equilibrium with the vesicle-bound NR ~ as a consequence

F. Grolig and G. Wagner: Isolated calcium-binding vesicles from Mougeotia

175

Fig. 8a--d. Electron micrographs of freeze-fracture replicas of isolated calcium-binding vesicles from Mougeotia scalaris, compared with a cross-sectional view of a calcium-binding vesicle in situ. a Cross-fractured calcium-binding vesicles, typically showing the rough matrix material (rm) in the central part of the vesicles, embedded in and surrounded by a belt of homogeneous material (hm; x 40000). b Calcium-binding vesicle, fractured close to the outer surface. Freeze-etching shows a smooth surface in the center, possibly originating fi'om the seal-like structure (ms, see e), with an outer patch of rough matrix material (rm; x 50000). e Cross-fractured calcium-binding vesicle as in a, but sealed by a membrane-like structure (ms; x 40000). Arrows in a, b and e indicate direction of shadowing; shadows on replicas appear white, d Transmission electron micrograph illustrating in situ a calcium-binding vesicle (cv), bounded by a membrane-like structure (ms), resembling the plasmalemma (pl) and the tonoplast (to); x 36000. chl=chloroptast; cyt= cytoplasm. Bars = 0.2 ~tm (a-e), 0.5 pm (d)

o f the i n c o r p o r a t i o n o f dye molecules ( N R ~ or N R +) into the detergent micelles (Triton X-100 or SDS; Fig. 5). Thus, N R in the vesicular matrix a p p a r e n t l y p r e d o m i n a t e s in the neutral f o r m (NR~ However, N R ~ absorbs m a x i m a l l y at 450 nm, while the N R ~ stained vesicles showed an a b s o r b a n c e m a x i m u m at 540 n m (Fig. 3 a), usually seen for N R +. This a p p a r e n t c o n t r a d i c t i o n is resolved by the r e p o r t e d

f o r m a t i o n o f an intermediate f o r m o f N R ~ which is stabilized in solution by an intramolecular hyd r o g e n b o n d ; this molecule absorbs like N R § but with lower extinction coefficients (Bartels 1956). Observation o f similar a b s o r b a n c e changes u p o n binding o f N R to the p o l y p h e n o l i c tannic acid (Fig. 3 b) strongly supports the view o f N R ~ binding to the vesicular matrix as the intermediate f o r m stabilized via intermolecular h y d r o g e n b o n d -

F, Grolig and G. Wagner: Isolated calcium-binding vesicles from Mougeotia

176

,~ 10 I .~

K

(Cu)

0

0

keY

10.2/,

Fig. 9. X-ray microanalysis of an NR-stained, isolated calciumbinding vesicle from Mougeotia scalaris. Peaks within the spectrum are assigned to elements ranging from potassium (K) via calcium (Ca) and iron (Fe) to copper (Cu), the latter mainly derived from the copper grid. Accelerating voltage=100 kV; beam current = 20 gA

100-

~176" ~ "

80

~" o \ \

0 o

o~ o

o

=

60

o~\

=

40

o

\ o

\o

\ o o

\ \

20

\~.,~

0

'

0

I

2

'

i

4

,

,

i

I

'

6

'

o

'

I

8

,

'

I

10

NRbound/pmot. (rag cht)-I

Fig. 10. Dependence of chloroplast reorientational movement in Mougeotia sealaris on the amount of N R taken up by the calcium-binding vesicles

ing. The hydrogen donors appear to be the phenolic hydroxyl groups. This is demonstrated by the progressive disintegration of the vesicles in alkaline suspension medium, being most pronounced at pH 10. The phenolic hydroxyl groups are abundant enough to serve as ligands for the adsorption of the up to 8.5.10 9 NR molecules per average-size vesicle. The "concentrations" within such a vesicle can be estimated from the titrations (Fig. 2) as about 40tool.1-1 for the phenolic hydroxyl groups, and from the dye extractions as about 8 and 4 mol. 1-1 for the dyes NR and RB, respectively. The combined, reversible release of both NR and the phenolic compound(s) from the isolated

vesicles (Fig. 4) indicates interdependent binding of both compounds to the vesicle core. Neutral red and RB each bind to two different binding sites within the vesicular matrix (Fig. 6). The molar free adsorption energies for these binding sites (to be detailed elsewhere) are sufficient to account for one and two hydrogen bonds for the low-affinity and the high-affinity binding modes, respectively. Presumably, the stabilizing effect of NR and RB results from intermolecular crosslinking of the reversibly bound phenolic compounds via the highaffinity binding mode of these heterocyclic dyes. Titrating the vesicle suspension after prolonged incubation at pH 12 showed irreversible consumption of hydroxyl ions (data not shown). This indicates the hydrolysis of ester linkages and the involvement of phenolic hydroxyl groups. The reaction continues as long as deprotonation of the newly released phenolic hydroxyl groups proceeds (Morrison and Boyd 1973). Here, an ester linkage implies the occurrence of carbonyl oxygen and of oligomeric phenolic compound(s). Considerable amounts of monovalent ions, like potassium (Fig. 9), inside isolated, vitally stained vesicles indicate accompanying negative charges within the vesicular matrix. These negative charges, which cause a Donnan potential, presumably are mainly deprotonated phenolic hydroxyl groups. The latter, together with protonated phenolic hydroxyl groups and the ester-linked carbonyl oxygen provide good candidates for the arrangement of coordination sites for divalent and trivalent cations in the unstained vesicles. Furthermore, the reversible, and hence flexible, binding of phenolic compound(s) to the core, the predominance of oxygen ligands and the capability of C a 2 + t o undergo coordination bonds of various lengths (Levine and Williams 1982; Hepler and Wayne 1985) may provide the molecular basis for the Caselectivity of the coordination sites that was observed in situ (Wagner and Rossbacher 1980). Vesicular stability in vivo probably depends partly on an intermolecular hydrogen-bonding network for the phenolic compound(s). Accumulation of the enormous amounts of vital dye in vesicles (8.5. J 0 9 NR-molecules per average-size vesicle) must cause structural changes in the vesicular matrix and may even lead to an invivo covering membrane being stripped off, a process which, however, does not destabilize the vitally stained matrix (Figs. 7c, 8a). Dye adsorption should have a dramatic effect on vesicular calcium compartmentation as the coordination binding of C a 2 + by the vesicle matrix is disturbed: the C a 2 +b i n d i n g capacity will be lowered by the progressive

F. Grolig and G. Wagner: Isolated calcium-bindingvesicles from Mougeotia

binding of NR or RB to the phenolic hydroxyl groups, thus leading to the observed lack of detectable amounts of calcium in isolated vesicles (Fig. 9). This indicates that the inhibition of chloroplast movement upon vital staining by NR (Fig. 10) is caused by the disturbance of the intracellular Ca 2 +-homoeostasis resulting from the release of unphysiological amounts of Ca 2+ from the vesicular matrix into the cytoplasm, a process which at higher dye loads also impedes viability of the cells. This interpretation is supported by the finding that locally restricted staining of the calcium-binding vesicles in vivo caused local inhibition of chloroplast movement; inhibition of movement was overcome by the cell with time, although the staining of the vesicles persisted in Mougeotia scalaris (Russ et al. 1988). We gratefully acknowledge the use of the freeze-fracture facilities in the laboratory of Professor A. Sievers and the help of Dr. Harald Behrens, Bonn. We thank B. Brfickel and J. Manke for their help in EDAX-analysis at the Institut ffir Arbeitsmedizin, Giessen, and H. Schauer, C. Jung and A. Weisert for contributing technical assistance. We are indebted to Dr. Peter P. Jablonski, Research School of Biological Sciences, A.N.U., Canberra, Australia for improving the English style. With financial support from the Deutsche Forschungsgemeinschaft

References Bartels, P. (1956) Spektralphotomeirische Untersuchungen am Neutralrot (I), Prototropiegleichgewichte und Normaltemperaturspektren im sichtbaren und ultravioletten Spektralbereich. Z. Phys. Chem. N.F. 9, 74-94 Dreyer, E.M., Weisenseel, M.H. (1979) Phytochrome-mediated uptake of calcium in Mougeotia cells. Planta 146, 31 39 Grob, K , Matile, P. (1979) Vacuolar location of glucosinolates in horseradish root cells. Plant Sci. Lett. 14, 327-335 Grolig, F. (1986) Calcium-Vesikel und lichtabh/ingige Chloroplastenreorientierung bei Mougeotia spec. Ph.D. Thesis, Justus-Liebig-University, Giessen, FRG Grolig, F., Wagner, G. (1987) Vital staining permits isolation of calcium vesicles from the green alga Mougeotia. Planta 171,433-437 Grolig, F., Wagner, G. (1988) Light-dependent chloroplast reorientation in Mougeotia and Mesotaenium: Biased by pigment-regulated plasmalemma anchorage sites to actin filaments? Bot. Acta 101, 2-6 Haupt, W. (1970) Hellrot- und Dunkelrot-Wechselwirkungen bei der Chloroplastendrehung yon Mougeotia. Wiss. Z. Ernst-Moritz-Arndt-Universit/it Greifswald, Math. Naturwiss. Reihe 19, 47-54

177

Haupt, W. (1987) Phytochrome control of intracellular movement. In: Phytochrome and photoregulation in plants, pp. 225-237, Furuya, M., ed. Academic Press, London Haupt, W., Weisenseel, M.H. (1976) Physiological evidence and some thoughts on localized responses, intracellular localization and action of phytochrome. In : Light and plant development, pp. 63 74, Smith, H., ed. Butterworths, London Hepler, P.K., Wayne, R.O. (1985) Calcium and plant development. Annu. Rev. Plant Physiol. 36, 397-439 Klein, K., Wagner, G., Blatt, M.R. (1980) Heavy-meromyosindecoration of microfilaments from Mougeotia protoplasts. Planta 150, 354-356 Levine, B.A., Williams, R.J.P. (1982) Calcium binding to proteins and other large biological anion centers. In: Calcium and Cell Function 2, pp. 1-38, Cheung W., ed. Academic Press, London Moor, H., Miihlethaler, K. (1963) Fine structure in frozenetched yeast cells. J. Cell Biol. 17, 27-46 Morrison R.T., Boyd R.N. (1973) Organic chemistry, 3rd ed. Allyn and Bacon, Inc., Boston, Mass., USA Rossbacher, R. (1982) Mikroanalytische Untersuchungen zur Ionenkompartimentierung bei den zygnemalen Gr/inalgen Mougeotia, Spirogyra und Zygnema. Ph.D. Thesis, University of Erlangen-Niirnberg, FRG Roux, S.J. (1984) Ca 2+ and phytochrome action in plants. BioScience 34, 1, 25-29 Russ, U., Grolig F., Wagner, G. (1988) Differentially adsorbed vital dyes inhibit chloroplast movement in Mougeotia scalaris. Protoplasma, in press Senger, H. (1970) Characterisierung einer Synchronkultur von Scenedesmus obliguus, ihrer potentiellen Photosyntheseleistung und des Photosynthese-Quotienten w/ihrend des Entwicklungscyclus. Planta 90, 243-266 Serlin, B.S., Roux, S.J. (1984) Modulation of chloroplast movement in the green alga Mougeotia by the CaZ+-ionophore A 23187, and by calmodulin antagonists. Proc. Natl. Acad. Sci. USA 81, 6368-6372 Volkmann, D. (1981) Structural differentiation of membranes involved in the secretion of polysaccharide slime by root cap cells of cress (Lepidium sativum L.). Planta 151, 180-188 Wagner, G., Haupt, W., Laux, A. (1972) Reversible inhibition of chloroplast movement by cytochalasin B in the green alga Mougeotia. Science 176, 808-809 Wagner, G., Klein, K. (1978) Differential effect of calcium on chloroplast movement in Mougeotia. Photochem. Photobiol. 27, 137 140 Wagner, G., Rossbacher, R. (1980) X-ray microanalysis and chlorotetracycline staining of calcium vesicles in the green alga Mougeotia. Planta 149, 298-305 Wagner, G., Valentin, P., Dieter, P., Marm6, D. (1984) Identification of calmodulin in the green alga Mougeotia and its possible function in chloroplast reorientational movement, Planta 162, 62 67 Received 4 August; accepted 29 September 1988