REGULATION OF CYTOPLASMIC STREAMIN IN G ...Cytoplasmic streaming in Vallisneria 387 Table 1. Combination of filters used to obtain monochromatic lights Interference filter KL-45 KL-50

y. Cell Set. 62, 385-405 (1983) 385Printed in Great Britain © The Company of Biologists Limited 1983

REGULATION OF CYTOPLASMIC STREAMING IN

VALLISNERIA MESOPHYLL CELLS

SHINGO TAKAGI* AND REIKO NAGAI't

Department of Biology, Faculty of Science, Osaka University, Toyonaka, Osaka 560,Japan

SUMMARY

Induction and cessation of the rotational cytoplasmic streaming in Vallisneria mesophyll cellscould be controlled by external stimuli. In cells that had been kept in darkness the cytoplasmremained quiescent. However, when the cells were treated in the dark with EGTA solution(10mM or 20mM buffered with 10mM-Tris-maleate at pH7-0), rotational cytoplasmic streamingwas induced. When the cells were transferred again to artificial pond water in the dark, the inducedstreaming was inhibited; that is, only 50 % of the observed cells exhibited active streaming after 2 h.When the cells were irradiated continuously with far-red light (Am̂ K = 750nm, 0'4W/m

2) in thesame external medium, the induced streaming was inhibited almost completely within 2h.

The relative quantum effectiveness of monochromatic light (450—800 nm) in producing cessationof streaming was also investigated. Irradiation with light of 450, 550 and 600 nm was almost aseffective as darkness. Light of 500 and 650 nm was less effective than dark exposure. Only irradiationat 750 nm stopped streaming in almost all cells. But when calcium was excluded from the externalmedium, the effect of far-red light decreased to almost the dark control level. Light of 800 nm alsoinhibited the streaming but the effect was much less than that of far-red light.

Microfilaments in bundles with the long axis parallel to the streaming direction were localized inthe vicinity of the cell membrane. Their configuration, localization and distribution were the samein the present experimental system irrespective of whether the cytoplasm was streaming or quies-cent.

Intracellular calcium was examined by electron microscopic cytochemistry and X-raymicroanalysis. In cells with streaming induced by EGTA, only a small amount of calcium-containing precipitates formed in the cytoplasm in the presence of antimony. A few precipitates werefound in the chloroplasts, the middle lamella of the cell wall and at the border between the cytoplasmand the cell wall. On the other hand, in cells treated with EGTA and subsequently irradiated withfar-red light in artificial pond water, many precipitates were observed in the cytoplasm,chloroplasts, mitochondria and endoplasmic reticulum. The middle lamella was also heavilystained.

On the basis of these observations, it was concluded that rotational cytoplasmic streaming inVallisneria cells can be induced when the free calcium concentration in the cytoplasm decreases andthat the induced streaming is arrested when the free calcium concentration in the cytoplasm in-creases. Far-red light accelerates the increase of calcium in the cytoplasm.

INTRODUCTION

In leaf cells of the higher aquatic plants, Elodea and Vallisneria, rotational stream-ing of the cytoplasm is induced by irradiation with visible light (photodinesis) orapplication of various chemicals (chemodinesis). Thus, this type of streaming is called

•Present address: Department of Biology, General Education, Osaka University, Toyonaka,Osaka 560, Japan.

f Author for correspondence.

386 5. Takagi and R. Nagai

secondary streaming, while the type seen in characean cells is called primary stream-ing, because it persists under normal conditions (Hauptfleisch, 1892; Kamiya, 1959).Seitz (1967, 1972) investigated the nature of the photoreceptor involved in the light-induced movements in Vallisneria and the effect of light on centrifugability ofchloroplasts, cyclosis and phototactic orientational movement of chloroplasts. Heproposed that the primary effect of light is due to regulation of the availability of ATPfrom oxidative and photosynthetic phosphorylation; an increased availability of ATPactivating cytoplasmic streaming and thus inducing chloroplast movement.

The microfilament organization responsible for the generation of motive force hasbeen reported for the epidermal cells of Vallisneria (Yamaguchi & Nagai, 1981).However, the mechanism by which the induction and cessation of streaming isregulated remains obscure.

Basically, the mechanics of cytoplasmic streaming in characean internodal cellsseem to be as follows. Rotational cytoplasmic streaming is caused by unidirectionalsliding of endoplasmic organelles along the bundles of microfilaments that are an-chored on the stationary chloroplast files (Kamitsubo, 1966; Kamiya & Kuroda,1956; Nagai & Rebhun, 1966; Nagai & Hayama, 19796). The microfilaments aremainly composed of F-actin (Palevitz, Ash & Hepler, 1974; Williamson, 1974;Palevitz & Hepler, 1975). The sliding organelles are equipped with myosin-likeprotein (Nagai & Hayama, 1979a,6). Motility requires Mg2+-ATP and Ca2+ at10~7M or less (Williamson, 1975; Williamson & Ashley, 1982; Hayama, Shimmen &Tazawa, 1979; Hayama & Tazawa, 1980; Tominaga & Tazawa, 1981).

In the case of Vallisneria: (1) the microfilament bundles are found at the ectoplas-mic gel layer of the epidermal cells (Yamaguchi & Nagai, 1981); (2) cytochalasin Binhibits streaming in the mesophyll cells (Ishigami & Nagai, 1980); and (3) a myosin-like protein has been extracted and partially purified from Egeiia (Elodea) densa(Ohsuka& Inoue, 1979). These facts are shared by characean cells exhibiting primarystreaming.

This paper aims to show that induction and cessation of cytoplasmic streaming inVallisneria mesophyll cells are controlled by a change in Ca2+ concentration in thecytoplasm and not by a change in microfilament organization.

MATERIALS AND METHODS

PlantVallisneria gigantea Graebn was purchased at a tropical fish store and cultured in a bucket with

soil at the bottom and filled with tap water. The culture was kept under a 12 h light (4000 lux withfluorescent Iamps)/I2h dark regime at room temperature.

Pretreatment of samplesTo exclude possible induction of streaming by cutting and touching the plants, the following

conditioning was done before the experiments.A leaf segment, about 10 cm long, was cut from a healthy plant in the stock culture at the end of

the light period. It was then cut into smaller pieces about 2 mm long, which were then floated onartificial pond water (APW) containing 5 X 10~ 5 M-KC1 , 2 X 10~*M-NaCl, 10"4M-Ca(NO3)2 ,

Cytoplasmic streaming in Vallisneria 387

Table 1. Combination of filters used to obtain monochromatic lights

Interferencefilter

KL-45KL-50KL-55KL-60KL-65KL-70KL-75KL-80

Wavelength(nm)

451-8497-5551-0595-0650-0696-0747-0795-0

Half-band-width (nm)

9-613-59 0

14-017-516-518-017-0

Cut-offfilter

V-Y43V-Y48V-053V-058V-R64V-R68V-R68V-R68

10"4M-Mg(NO3)2 and 2 x 10 5M-Tris-maleate buffer at pH7-0. The air trapped in the inter-cellular space was evacuated, and each piece of leaf was placed in a separate glass vessel with 40 mlof APW and incubated under the original light conditions of the 12h dark and 12h light regime.During incubation, the medium was replaced once at the end of a dark period. After one cycle ofdark and light, each specimen was mounted on a glass slide with a coverslip using a small amountof vaseline at each corner. The air gap between the glass slide and the coverslip was filled with freshmedium. The slide loaded with the specimen and the coverslip was then immersed in a glass vesselfilled with the same medium and kept in the dark for another 12^18h at 20-25 °C. After theseprocedures, the specimens could be used for the experiments without being touched with forceps.The cytoplasm in all cells was stationary at the end of these procedures.

Light sourcesGreen light (Amu = 550 nm, 0-6 W/m2) was used as safe light throughout the experiments,

because green light has no effect on induction and cessation of streaming. The green light wasobtained by combining an interference filter (KL-55, Toshiba, Japan) and a cut-off filter (V-053,Toshiba, Japan). These filters were placed in front of a tungsten lamp. To exclude any heat effect,a heat filter and a glass tank (17 cm X 9 cm X 9 cm) filled with tap water were placed in the light path.Other monochromatic light was prepared similarly with the combination of an interference filter anda cut-off filter (Table 1). The intensity of the monochromatic light was measured with a photodiodedetector (PIN-10DF, United Detector Technology), which was positioned in the focal plane of thelight microscope. The output from the detector was monitored with a voltmeter (SP-H6V, RikenDenshi, Japan) and later converted into a quantum number. This calculation was performed withthe assumption that all of the energy transmitted by each interference filter corresponded to itswavelength maximum.

Irradiation with monochromatic lightThe specimen was irradiated through a condenser lens on the microscope stage with each

monochromatic light at a constant quantum number of 1 -53—1 -58(X 10l8)photon/s.m2 under agiven set of test conditions. When the specimen was irradiated with light of over 700 nm, observa-tions were made using a TV camera (CTC-5600JC, equipped with Newvicon, Ikegami, Japan) anda monitor TV (WV-930, National, Japan).

Application of chemicalsEthvlencglycol-bis(2-aminoethylether)-./V,A',A'',A''-tetraaceticacid (EGTA) (Wako, Japan) was

dissolved in distilled water at 200 mM with potassium hydroxide to adjust the pH to 7-0, then dilutedwith lOniM-Tris-maleate buffer (pH7-0) to give final concentrations of 1, 10 and 20 mM. EGTAwas applied under dark conditions by gentle irrigation between the coverslip and the slide. Theinduction of streaming was examined under green light.


Measurement of effectivenessSpecimens were examined with a Nikon light microscope equipped with 40 X objective and SX

ocular lenses. The effectiveness of irradiation and chemicals on induction or cessation of streamingwas expressed as the ratio of streaming cells to the total of observed cells. Fifty to a hundred cellsfrom four to ten different leaf segments were examined for each test condition. Cells were countedas streaming when their chloroplasts exhibited streaming continuously for at least 5 s.

Electron microscopyThe entire specimen was illuminated and induction or cessation of streaming in all cells were

examined before the specimens were transferred to the fixative.The specimens were prefixed with 2-5 % glutaraldehyde buffered with 25 mM-cacodylate

(pH 7-0) for 2 h, under evacuation for 10 min at the beginning of the fixation. After washing twicewith the buffer, the specimens were post-fixed for 2h with 2% OsO* in the same buffer, underevacuation as before. The irradiated specimens were prefixed in far-red light (Am*, = 750 nm, 0-4 W/m2), then kept in the dark. EGTA-treated specimens were fixed in the dark or under green light.

To examine the intracellular localization of calcium, the specimens were fixed in the same wayas above, but 2% potassium pyroantimonate (Wako, Japan) was added throughout prefixation,postfixation and washing, and the concentration of the buffer solution was reduced to 20 itiM.

After dehydration through a graded series of ethanol, all specimens were embedded in Spurr's(1969) medium, then sectioned. Thin sections were stained with uranyl acetate and lead citrate. Forthe examination for calcium precipitates, the sections were not stained. A JEM 100-C type electronmicroscope was used for observations at 80 kV. For X-ray microanalysis, a Hitachi X-600 typeelectron microscope and KEVEX 7000Q energy analyser were used.

RESULTS

Induction of streaming

A Vallisneria leaf is made up of epidermal cells on each surface and the inner partis occupied by mesophyll cells. A cross-section shows that the middle area of the leafconsists of several layers of mesophyll cells (Fig. 1A) and there are two layers near theleaf edges (Fig. 1B). In all experiments observations were made on mesophyll cellsnear the leaf edges. Cytoplasmic streaming could be induced more easily in mesophyllcells than in epidermal cells and it was desired to keep disturbance of actinic light ata minimum.

In mesophyll cells, no cytoplasmic streaming was observed at the end of the con-ditioning period in darkness, apart from slight movements of some cytoplasmic part-icles. Chloroplasts remained still and were distributed almost evenly in the cell. Thenucleus was anchored at an unspecified locus in the cytoplasm.

When treated with EGTA solution, agitational movements of the cytoplasmicparticles immediately became vigorous and some underwent translatory motions. Intime, some chloroplasts began to follow these local streamlets of cytoplasmic particles,but the movement was haphazard. They moved 10-20/zm at a time and then stopped.This kind of saltatory movement, as named by Rebhun (1964), was repeated in eachchloroplast for a while. At 1—2 min after EGTA had been applied, the chloroplastsgradually formed a line and began to move continuously. Then the cytoplasmicstreaming formed a closed circuit, which rotated along the cell wall unidirectionally,either clockwise or counter-clockwise. The streaming rate was low at the beginning


Fig. 1. Photomicrographs of cross-sections of Vallisneria leaf. A. A middle area whereseveral layers of mesophyll cells (meso) are found between the two layers of epidermal cells(epi). B. An area near the leaf edge where only two layers of mesophyll cells are foundbetween the two layers of epidermal cells, n, nucleus; c, chloroplast; p, cytoplasmicparticle. Bar, 100^m.

but reached a maximum value in the range of 10—20 îm/s within about lOmin. Thenucleus could rarely be observed in the quiescent cell because of its position. Inseveral cases it was observed to be remaining still, but it began to move on applicationof EGTA, as with the chloroplasts, and finally participated in the active rotationalcytoplasmic streaming. Once the cytoplasmic streaming was established, the nucleus


could be observed very clearly moving along the cell wall, together with cytoplasmicparticles and chloroplasts.

To express quantitatively the relationship between EGTA concentration and theinduction of streaming in the dark, the number of cells (iVx , Fig. 2) with cytoplasmicstreaming was counted under green light after appropriate durations of treatment andthe ratio of the streaming cells to the total observed cells (iVtoUi, Fig. 2) was plottedas percentage against time.

In 1 mM-EGTA, the number of streaming cells increased gradually with time. TheArx/A

rtotai ratio reached a plateau at 50% after treatment for 1 h. In 10mM-EGTA,streaming was observed in all cells within 30 min. This active state was maintained for2—3 h. With 20 mM-EGTA, streaming was induced more quickly than with 10mM,but the active state lasted only around 30 min before it started to decline. With thebuffer solution alone, no streaming was induced. Thus, EGTA is clearly responsiblefor the induction in dark conditions.

Cessation of streaming produced by irradiation with light of various wavelengths

Since the actinic light for stopping streaming was not known, the effectiveness ofmonochromatic light of various wavelengths was investigated.

30Time (min)

60

Fig. 2. Induction of streaming in the dark with EGTA. The number of streaming cells(Afx) was counted under green light after various periods of treatment. The ratio of cellsinvolved in streaming to total cells observed (NM,\) is plotted as the percentage againsttime (min). (O O) 1 mM-EGTA (A'IOui. 58; number of specimens iV,, 4);( • • ) 10niM-EGTA (AU.i, 106; N., 8); (A A) 20mM-EGTA (i\\M.\, 53;

t) lOmM-Tris-maleate buffer (ArtOui, 55; Nt, 4), control.

(Jytoplasmic streaming in Vallisneria 391

First, specimens, in which all cells exhibited streaming in the dark due to treatmentwith lOmM-EGTA, were continuously irradiated with far-red (f.r.) light (Amax =750 nm, 0-4 W/m2) after the EGTA solution had been replaced with APW. As shownin Fig. 3 (O O), the number of streaming cells decreased with increase in irr-adiation time. When the specimens were kept in the dark the number of streamingcells also decreased, but much more slowly (Fig. 3, # • ) . Clearly, far-redlight had a definite effect in producing cessation of streaming.

Next, the specimens were treated in the same manner, then were individuallyirradiated continuously with monochromatic light of a different wavelength at thesame quantum number to obtain the time-course of the cessation of streaming. Thevalues of [(Nx/NVM\)iMk — (Nx/Ntoit\)m


interpreted this phenomenon as showing that the microfilament bundles had beencompletely destroyed during the period when streaming ceased and was re-establishedafter removal of the drug, because the streaming direction is known to be determinedby the polarity of F-actin (Kersey, Hepler, Palevitz & Wessells, 1976).

This is also supposed to be the case for irradiation with far-red light. To confirmthis, the configuration, localization and distribution of the microfilaments in cellswere examined and compared at the end of each successive treatment: namely, (a)after the pretreatment in which the specimens had been kept in the dark for 12-18 h;(b) after they were treated with 10 mM-EGTA; and (c) after they were irradiated withfar-red light.

In cells under condition (a), the cytoplasm did not show any sign of streaming. Inthese cells, bundles of microfilaments, which are very similar in appearance to thosein epidermal cells of Vallisneria (Yamaguch & Nagai, 1981), characean internodalcells (Nagai & Rebhun, 1966; Nagai & Hayama, 19796) and higher plants (Parthas-arathy & Miihlethaler, 1972; O'Brien & Thimann, 1966), were localized in the vicin-ity of the cell membrane and arranged parallel to the direction in which the cytoplasmis expected to stream. Fig. 5A shows part of two adjacent mesophyll cells in cross-section at low magnification. Fig. 5B is a magnified picture of Fig. 5A. Cross-sectionsof the microfilaments (mf) appear as 30-50 closely packed electron-dense dots. Thedistance from one bundle to the next varied with the cytoplasmic layer of the cell. Thiscannot be attributed to poor fixation, because the cytoplasm seems to be generally wellpreserved.

Under condition (b), the cytoplasm was actively streaming in all cells. The mor-phology of the microfilament bundles, shown in Fig. 5c, was similar to that undercondition (a). Bundles were found at the site where streaming occurred. In thespecimens under condition (c), streaming was completely inhibited. The microfila-ment bundles also remained unchanged in their appearance, localization anddistribution (Fig. 5D).

From these observations, we conclude that cessation of the streaming in the darkor cessation produced by irradiation with far-red light cannot be explained in termsof changes in the microfilament arrangement.

Role of calcium in the control of streaming

The most plausible factor responsible for controlling the streaming is the Ca2+

concentration in the cytoplasm. Many reports have indicated that various kinds ofmovements in eukaryotic cells are regulated by Ca2+. Also, red and far-red light affectCa2+ movement across the cell membrane (Dreyer& Weisenseel, 1979; Hale&Roux,1980).

From this point of view, we examined the effect of external Ca2+ on cells irradiatedwith far-red light. First, the specimens were treated with EGTA in the dark to inducestreaming in all cells. Some were then transferred to calcium-containing APW andirradiated continuously with far-red light. Other specimens were irradiated incalcium-free APW. Fig. 6 shows the time-course of the ratio of streaming cells to totalcells observed in the presence or absence of Ca2+. Clearly, the induced streaming was


Fig-5

Vyloplasmic streaming in Vallisneria 395

100

80

60

40

20

\ ^^L^\ \VQ

\\

^ 1

\ S N

1

)

20 40 60Time (min)

80 100 120

Fig. 6. Effect of calcium on cessation of streaming. Specimens were treated with EGTAin the dark to induce streaming in all cells (100%). They were then divided into fourgroups according to the experimental conditions: (1) calcium-free APW in the dark( A - - - A ) ; (2) calcium-containing APW in the dark ( • • ) ; (3) irradiated with far-red light in calcium-free APW (A — A) ; and (4) irradiated with far-red light in calcium-containing APW (O O). The ratio A'x/A/l0U| is plotted as percentage against time oftreatment. The length of the vertical bar is the S.E. of each value. A'mui for condition (1)and condition (3) was 87-102 and 40-57, respectively, and A', was 6-7 and 5-7. As forthe S.E., A'toU| and A', for condition (2) and condition (4), see Fig. 3 and its legend.

markedly inhibited only by the combined action of Ca2+ and irradiation with far-redlight. Only 10% of the cells continued streaming after 2h irradiation (O O).In the absence of Ca2+, the inhibition was about the same as that in the dark(A — A). The time-course in the dark was almost the same in the presence or absenceof Ca2+ ( • • , A-- -A) .

The inhibitory effect of Ca2+ was confirmed further by the following experiments.All specimens were treated with EGTA as before. The specimens were then trans-ferred to either calcium-containing or calcium-free APW for irradiation with far-red

Fig. 5. Electron micrographs of microfilament bundles (»«/). About 30—50 closely packeddots, presenting a cross-section of microfilaments, are localized near the cell membrane.The distance from one bundle to the adjacent one varies, A. Part of two adjacent mesophyllcells, kept in the dark for 12— 18 h, in which the cytoplasm did not show streaming. Thefixed cytoplasm seems to be well preserved, B. A higher magnification of A; C, microfila-ment bundles in cells treated with EGTA, in which the cytoplasm was actively streaming;D, microfilament bundles in cells treated with EGTA and subsequently irradiated with far-red light, in which the streaming was completely inhibited. Bars: A, 0-5 fxm\ B -D, 0-1 l-im.


—o

60 120 180

Time (min)240 300

Fig. 7. Repetition of induction and cessation of streaming. Each specimen kept in the darkshowed no streaming. Streaming was induced in all cells by treatment with EGTA in thedark (O O). After removal of EGTA with calcium-free APW (A) or with calcium-containing APW (B) , each specimen was irradiated with far-red light (O — O). When thecells were treated again with EGTA in the dark at the time streaming stopped, theyresumed streaming. These cessation and reactivation cycles could be repeated. Artoui was5 in A and 6 in B.

light. When the cells were treated with EGTA again at the time the cytoplasm ceasedto flow, they resumed streaming. Inhibition and reactivation of streaming could berepeated several times.

In the presence of Ca2+ (Fig. 7B), streaming stopped within 80 min on each irradia-tion. However, in the absence of Ca2+ (Fig. 7A), it took longer, about 160 min, beforestreaming came to a complete standstill. Reactivation of streaming by EGTA couldbe repeated but streaming did not stop completely on the second irradiation, evenafter 3 h.

Thus, streaming stopped more rapidly in the presence of Ca2+ when the cells wereirradiated with far-red light and the effect of the light was completely removed by

Fig. 8. Electron micrographs showing intracellular calcium deposits in cells kept in thedark for 12-18 h before fixation in the presence of potassium pyroantimonate. Precipitatesare formed in the vacuole, cytoplasm (cyt) and the middle lamella (ml) of the cell wall(cw). The sections were not stained, A. Cross-section of cells showing the distribution ofprecipitates at low magnification, B. Cross-section of parts of two adjacent mesophyll cellsat high magnification, epi, epidermal cell; meso, mesophyll cell; chl, chloroplast. Bars:A, 5/im; B, 1 ^m.


epi

"V

meso

8A

#

Fig. 8


• • • >

epi

cw

B

meso

9A

chlml

cyt

' •

K*

Fig. 9


EGTA. These facts may be interpreted as showing that calcium accumulates in thecytoplasm during irradiation and that an increase in Ca2+ concentration in thecytoplasm inhibits streaming.

Calcium in the cytoplasm

To demonstrate the increase in Ca2+ concentration visually, specimens were fixedin the presence of potassium pyroantimonate, which is known to be a fairly specificprecipitant of calcium.

First, we fixed cells just after the pretreatment. In these the cytoplasm did not showany sign of streaming. Precipitates, as shown in Fig. 8, are formed in the cytoplasmand the middle lamella of the cell wall. Precipitates are also abundant in the vacuole.Next, we fixed cells treated with IOITIM-EGTA solution for 25-40min to inducestreaming in all cells. Fig. 9 shows a section of these cells. Only a small amount ofprecipitate is seen in the cytoplasm. A few precipitates are observed in the middlelamella, the chloroplasts and at the border between the cytoplasm and the cell wall.Precipitates in the vacuole are slight.

On the other hand, precipitates are abundantly visible in the cytoplasm in cellstreated previously with EGTA and subsequently irradiated with far-red light in APW(Fig. 10). Precipitates accumulated in the chloroplasts (Fig. 10B), mitochondria andendoplasmic reticulum (data not shown). The middle lamella was heavily stained(Fig. 10B).

The presence of calcium in these precipitates was confirmed using an X-raymicroanalyser. Arrows a, b, c and d in Fig. 11 indicate the precipitates on whichanalysis was done. Four micrographs in the lower portion of this figure show theresults of the analysis alphabetically. Clearly, calcium was present in each precipitate.Osmium comes from the fixative OsO4 and copper from the grid.

These observations reveal that the intracellular calcium concentration is muchlower when the cytoplasm is involved in streaming than when it is immobile.

Fig. 9. Intracellular calcium deposits in cells treated with lOmM-EGTA for 25-40minbefore fixation. Only a small amount of precipitate is observed in the cytoplasm. A fewprecipitates are observed in the middle lamella (ml), chloroplasts (chl) and at the borderbetween the cytoplasm (cyl) and the cell wall (ctv). The sections were not stained, A.Cross-section of cells at low magnification; B, cross-section of cells at high magnification.Bars: A, lO^ni; B, 1 fim.

Fig. 10. Intracellular calcium deposits in cells treated with EGTA and then irradiatedwith far-red light in APW before fixation. Precipitates are abundant in the cytoplasm andchloroplasts, and also in the middle lamella. The sections were not stained, A. Cross-section of cells at low magnification. The asterisk indicates a cell killed by cutting with arazor during the pretreatment of the specimen, B. Cross-section of cells at high magnifica-tion, epi, epidermal cell; meso, mesophyll cell. Bars: A, 10^m; B, l^m.

Fig. 11. X-ray microanalysis of precipitates. The electron micrograph shows parts of twoadjacent mesophyll cells from the specimen shown in Fig. 10. Arrows a, b, c and^/ indicatethe precipitates analysed. The results correspond alphabetically to each micrograph shownin the lower part of the figure. All micrographs show that calcium (CA) is contained in eachanalysed precipitate. Omsium (OS) comes from OsO* and copper (CU) from the grid.

400 S. Takagi and R. Nagai

. 1 . . .

Fig. 10. For legend see p. 399.


1 I . 2 4 K E U * 0 . 0 8 K E U 10.24KEU)

Fig. 11. For legend see p. 399.


DISCUSSION

All the pretreatment procedures used were needed to cause complete stoppage ofstreaming in all cells, in order to obtain a homogeneous preparation of cells as startingmaterial for the experiments. For example, in the case of a single specimen mountedimmediately after cutting and evacuation, omitting the stage of keeping it under theoriginal light conditions of the 12 h dark and 12 h light scheme, led to about 10% ofthe observed cells exhibiting streaming even after 12-18 h dark treatment. In the casein which many pieces, say 20, were floated in a glass vessel under the same conditionsstreaming was active in almost all cells. And even after complete pretreatment, directtouching of a specimen with fingers or forceps could induce streaming in the cells.Therefore, each specimen had to be kept first under the original light condition for24 h after cutting and evacuation, and then mounted on a glass slide before the lastdark treatment. Some unknown substance(s) that induces streaming seems to besecreted from the injured end of the leaf or even upon direct touching of the cells.Further study is needed to identify such a substance(s) and to investigate how stream-ing is induced, i.e. through decreasing calcium concentration in the cytoplasm or byanother mechanism.

The present study clarifies the microfilament organization in Vallisneria mesophyllcells. The microfilaments in Vallisneria are known to be composed mainly of F-actin(Yamaguchi & Nagai, 1981). And because streaming is inhibited by cytochalasin B(Ishigami & Nagai, 1980), they are supposed to act as a motile apparatus for streamingin Vallisneria cells. Their configuration, localization and distribution were the samewhen the cytoplasm was streaming or quiescent. Therefore, morphological changesof the microfilaments are not directly responsible for the induction or cessation ofstreaming.

In the experiments showing that the induction and cessation of streaming could becontrolled alternately with EGTA and irradiation with far-red light (cf. Fig. 7A, B),examination of about 50 cells showed that the direction of the cyclosis, i.e. clockwiseor counter-clockwise, did not change after the second induction with EGTA. Anearlier report (Ishigami & Nagai, 1980) suggested that the microfilaments were ratherlabile but this statement has not been supported by the present observations. This isprobably due to the short duration of the treatment, i.e. 12-18 h in the dark or about2h under irradiation, or else it may be due to the fact that the bundles of microfila-ments or the microfilaments themselves are unstable in the presence of cytochalasinB in Vallisneria mesophyll cells, although in characean cells the microfilament bund-les remain normal after treatment (Bradley, 1973; Williamson, 1978).

The effect of calcium-chelating agents, EDTA or EGTA, on the velocity of stream-ing has been examined. Forde & Steer (1976) observed activation of streaming inElodea leaf cells by 0-lmM-EDTA and its inhibition by higher concentrations,10 mM and 100 min. They confirmed further that active streaming was inhibited byexternal application of Ca2+, and streaming inhibited by a higher concentration ofEDTA returned to normal when Mg2* was applied externally. Yamaguchi & Nagai(1981) reported that streaming in Vallisneria epidermal cells could be induced by the


application of 5-10 mM-EGTA. Although these observations were made under whitelight, they stressed that Ca2+ has a significant effect on secondary streaming. Thepresent study, which was performed with complete elimination of the light effect,confirmed the induction of cytoplasmic streaming by lowering the calcium concentra-tion in the cytoplasm.

The stimulating effect of EGTA on streaming is rapid (Fig. 2), which is readilyplausible if changes in a low Ca2+ concentration in a small cytoplasmic compartmentand a low rate of calcium transport from the vacuole are involved. Though vacuolarCaz+ concentration is supposed to be rather high during treatment with EGTA (cf.Fig. 8), a lack of precipitates in EGTA-treated cells was observed (cf. Fig. 9). Thismay be due to Ca2+ chelation in the free space by EGTA, unavoidably carried overinto the fixative and given access to the cell during fixation, and/or to Ca2+ chelationby EGTA penetrating into the vacuole during the treatment.

The exact concentration of calcium in the cytoplasm involved in streaming couldnot be determined in the present study. According to the observations on characeaninternodal cells, Ca2+ concentration in the streaming cytoplasm is about 10~7M andstreaming is inhibited at a higher concentration (Williamson, 1975; Williamson &Ashley, 1982; Tominaga & Tazawa, 1981). Therefore, it may be reasonable to sup-pose that the Ca2+ concentration in the cytoplasm during normal rotational cytoplas-mic streaming in Vallisneria mesophyll cells is also maintained at about 10~7 M or less.

Inhibition of streaming by irradiation with far-red light occurred more rapidlywhen calcium was present in the external medium than when it was not (Fig. 6). Thisis attributed to the increase in calcium concentration in the cytoplasm. However, littleis known about how such an increase in calcium transport into the cytoplasm isinfluenced by far-red light. Calcium accumulation in cells of Mougeotia has beenreported to be accelerated by irradiation with red light and this effect is cancelled bysubsequent irradiation with far-red light (Dreyer & Weisenseel, 1979). Efflux ofcalcium from oat coleoptiles and their protoplasts is enhanced by irradiation with redlight, and subsequent irradiation with far-red light induces a decrease to levels nearthe dark control (Hale & Roux, 1980). Examining the effect of red light on theinduction of streaming and its relation to that of far-red light may clarify whether apigment such as phytochrome plays a role in calcium transport in Vallisneria.

The precipitant pyroantimonate can penetrate the biological membrane and hencecan be used to fix Ca2+ in biological tissues. It is also useful as an electron-microscopiccytochemical reagent because it forms a dense precipitate with calcium (Caswell,1979). As pyroantimonate can react with Mg2"*" and Na+, as well as Ca2+, to causeprecipitation, deposits are not always precipitates of Ca2+. However, X-raymicroanalysis revealed that calcium was present in each precipitate even though addedmagnesium and/or sodium were coprecipitated. The original X-ray microanalysisdata showed the coexistence of antimony and calcium in the precipitates. Fig. l la ,b, c, d presents the results obtained by computer analysis in which the fractioncontaining antimony was subtracted from that in the original micrograph.

On the basis of these observations, we conclude that rotational cytoplasmic stream-ing in Vallisneria cells can be induced when the calcium concentration in the


cytoplasm decreases and the induced streaming stops when calcium concentration inthe cytpolasm increases.

We thank Professor emeritus N. Kamiya of Osaka University for his valuable criticism, and alsoto Professor H. Shibaoka for his profitable suggestions. We are indebted to Dr N. Matsumoto forlending out the photodiode detector to measure the intensity of light, and to Mr Saitoh (NakaWooks, Hitachi Ltd) for performing X-ray microanalysis.

This work was partly supported by grants-in-aid from the Japanese Ministry of Education,Science and Culture.

REFERENCES

BRADLEY, E. G. (1973). Microfilaments and cytoplasmic streaming: Inhibition of streaming withcytochalasin.J. Cell Sci. 12, 327-343.

CASWELL, A. H. (1979). Methods of measuring intracellular calcium. Int. Rev. Cytol. 56, 145-181.DREYER, E. M. & WEISENSEEL, M. H. (1979). Phytochrome-mediated uptake of calcium in

Mougeotia cells. Planta 146, 31-39.FORDE, J. & STEER, M. W. (1976). Cytoplasmic streaming in Elodea. Can.J.Bot. 54, 2688-2694.HALE, C. C. II & Roux, S. J. (1980). Photoreversible calcium fluxes induced by phytochrome in

oat coleoptile cells. PI. Physiol. 65, 658-662.HAUPTFLEISCH, P. (1892). Untersuchungen iiber die Stromung des Protoplasm in behauteten

Zellen. Jb. wiss. Bot. 24, 173-234.HAYAMA, T., SHIMMEN, T. & TAZAWA, M. (1979). Participation of Ca2+ in cessation of cytoplas-

mic streaming induced by membrane excitation in Characeae internodal cells. Ptvtoplasma 99,305-321.

HAYAMA, T . & TAZAWA, M. (1980). Ca2+ reversibly inhibits active rotation of chloroplasts inisolated cytoplasmic droplets of Cham. Ptvtoplasma 102, 1-9.

ISHIGAMI, M. & NAGAI, R. (1980). Motile apparatus in Vallisneria leaf cells. II. Effects ofcytochalasin B and lead acetate on the rate and direction of streaming. Cell Struct. Fund. 5,13-20.

KAMITSUBO, E. (1966). Motile protoplasmic fibrils of Characeae. II. Linear fibrillar structure andits bearing on protoplasmic streaming. Proc. Japan Acad. 42, 640—643.

KAMIYA, N. (1959). Protoplasmic streaming. Protoplasmatologia 8, 3a.KAMIYA, N. & KURODA, K. (1956). Velocity distribution of protoplasmic streaming in Nitella

cells. Bot. Mag., Tokyo 69, 543-554.KERSEY, Y. M., HEPLER, P. K., PALEVITZ, B. A. & WESSELLS, N. K. (1976). Polarity of actin

filaments in characean algae. Proc. natn. Acad. Sci. U.SA. 73, 165-167.NAGAI, R. & HAYAMA, T. (1979a). Ultraatructure of the endoplasmic factor responsible for

cytoplasmic streaming in Chara internodal cells. J. Cell Set. 36, 121 — 136.NAGAI, R. & HAYAMA, T. (19796). Ultrastructural aspects of cytoplasmic streaming in characean

cells. In Cell Motility: Molecules and Organization (ed. S. Hatano, H. Ishikawa & H. Sato), pp.321-337. Tokyo: University of Tokyo Press.

NAGAI, R. & REBHUN, L. I. (1966). Cytoplasmic microfilaments in streaming Nitella cells. J.Ultrastruct. Res. 14, 571-589.

O'BRIEN, T . P. & THIMANN, K. V. (1966). Intracellular fibers in oat coleoptile cells and theirpossible significance in cytoplasmic streaming. Proc. natn. Acad. Sci. U.SA. 56, 888-894.

OHSUKA, K. & INOUE, A. (1979). Identification of myosin in a flowering plant, Egeria densa.J.Biochem. 85, 375-378.

PALEVITZ, B. A., ASH, J. F. & HEPLER, P. K. (1974). Actin in the green alga, Nitella. Proc. natn.Acad. Sci. U.SA. 71, 363-366.

PALEVITZ, B. A. &HEPLER, P. K. (1975). Identification of actin in situ at the ectoplasm-endoplasminterface of Nitella. Microfilament-chloroplast association. J. Cell Biol. 65, 29-38.

PARTHASARATHY, M. V. & MOHLETHALER, K. (1972). Cytoplasmic microfilaments in plant cells.J. Ultrastruct. Res. 38, 46-62.

REBHUN, L. I. (1964). Saltatory particle movements in cells. In Primitive Motile Systems in CellBiology (ed. R. D. Allen & N. Kamiya), pp. 503-525. New York, London: Academic Press.


SEITZ, K. (1967). Wirkungsspektren fur die Starklichtbewegung der Chloroplasten, diePhotodinese und die lichtabhangige Viskositatsanderung bei Vallisneria spiralis ssp. torta. Z.PflPhysiol. 56, 246-261.

SEITZ, K. (1972). Primary processes controlling the light-induced movement of chloroplasts. AdaProtozool. 11, 225-235.

SPURR, A. R. (1969). A low-viscosity epoxy resin embedding medium for electron microscopy..7.Ultrastruct. Res. 26, 31-43.

TOMINAGA, Y. & TAZAWA, M. (1981). Reversible inhibition of cytoplasmic streaming byintracellular Ca2+ in tonoplast-free cells of Chara australis. Protoplasma 109, 103-111.

WILLIAMSON, R. E. (1974). Actin in the alga, Chara corallina. Nature, Lond. 248, 801-802.WILLIAMSON, R. E. (1975). Cytoplasmic streaming in Chara: A cell model activated by ATP and

inhibited by cytochalasin B.J. Cell Set. 17, 655-668.WILLIAMSON, R. E. (1978). Cytochalasin B stabilizes the subcortical actin bundles of Chara against

a solution of low ionic strength. Cytobiologie 18, 107-113.WILLIAMSON, R. E. & ASHLEY, C. C. (1982). Free Ca2+ and cytoplasmic streaming in the alga

Chara. Nature, Lond. 296, 647-651.YAMAGUCHI, Y. & NAGAI, R. (1981). Motile apparatus in Vallisneria leaf cells. I. Organization of

microfilaments. jf. Cell Sci. 48, 193-205.

(Received 4 January 1983-Accepted 21 February 1983)

Documents

REGULATION OF CYTOPLASMIC STREAMIN IN G ...Cytoplasmic streaming in Vallisneria 387 Table 1. Combination of filters used to obtain monochromatic lights Interference filter KL-45 KL-50