Amaranthus Protoplasts'1 - NCBI

Received for publication May 21, 1991 Accepted August 24, 1991

Plant Physiol. (1992) 98, 632-638

0032-0889/92/98/0632/07/$01 .00/0

Mechanism of Aluminum Inhibition of Net 45Ca2+ Uptake by Amaranthus Protoplasts'1 Zdenko Rengel*2 and Daphne C. Elliott School of Biological Sciences, Flinders University of South Australia, GPO Box 2100, Adelaide S.A. 5001, Australia causing a large electrochemical gradient that favors transport of Ca2" into the cell to be established under most growing conditions. Such a gradient may, however, be diminished to a large extent by Al. Under acidic conditions, polyvalent cationic species of Al tend to bind very tightly to the apoplastic negative charges, thus greatly reducing the amount of Ca2+ in the Donnan free space (24 [and references therein]). These physicochemical reactions may at least partly explain the deleterious effect of Al on Ca2" accumulation in plant cells. Transport of Ca2" across the plasma membrane is assumed to occur through Ca2" channel proteins spanning the poorly permeable lipid bilayer (for a review, see ref. 28). No studies that would assess an influence of Al toxicity on transport of Ca2e across plasma membrane appear to have been published yet. Hypothetically, Al may affect Ca2` fluxes through the plasma membrane by affecting Ca2" channels either directly or indirectly by interfering with the action of inositol-1,4,5triphosphate and GTP-binding proteins, as well as by affecting Ca2+-translocating ATPases. Al inhibited inward-rectifying K+ channels in the plasma membrane of guard-cell protoplasts of Viciafaba (26). It also inhibited voltage gating of voltage-dependent anion-selective channel in the mitochondrial outer membrane of Neurospora crassa (8). Effects of Al on Ca2"-transporting channels in plasma membrane of higher plants do not appear to have been tested yet. Inositol- 1,4,5-triphosphate activated Ca2` channels in plant membranes (28 [and references therein]). Al has a high affinity toward phosphate at pH < 7 (4); the formation of an Alphosphate complex is energetically favored at the expense of the physiological Mg-phosphate formation. A high affinity of Al ions toward phosphate may also result in alteration of the function of GTP-proteins that have been shown to stimulate 45Ca2+ fluxes across the plasma membrane of Amaranthus protoplasts (1 1). Based on the inhibitory action of Al ions on Mg2+-ATPase in the plasma membrane of Pisum sativum (20), it may be hypothesized that Al would also disturb the function of the plasma membrane Ca2+-ATPases of Amaranthus protoplasts by binding to and thus depleting the pool of cell ATP that is necessary for fueling Ca2+-translocating ATPases. In addition, Al was found to bind to calmodulin, Ca2+ regulatory cytosolic protein, rendering it functionless (27). Calmodulin was also shown to stimulate plasma membrane Ca2+-ATPases (25). Indirect evidence obtained with Amaranthus protoplasts also suggested calmodulin stimulation of Ca24-ATPase (I l). However, it appears that results of calmodulin activation of Ca2+-

ABSTRACT

Calcium ions serve as a second messenger in signal transduction and metabolic regulation. Effects of Al on calcium homeostasis remain to be elucidated. Short-term net "Ca2+ uptake by Amaranthus tricolor protoplasts was monitored from uptake media prepared to test the influence of pH, Al, and various inhibitors. Accumulation of "Ca2+ increased during the first 3 to 6 minutes and then leveled off or declined. Al and Ca2+ channel blockers (verapamil and bepridil) decreased net 4Ca2+ uptake. This decrease was more pronounced when Al and bepridil were both present in uptake media, but Al did not aggravate verapamilinduced reduction of net "Ca2+ uptake. Erythrosin B and calmidazolium each increased net 45Ca2+ uptake, probably by interfering with Ca2+ efflux. This effect was undetectable in the presence of Al. Mycophenolic acid decreased net "Ca2+ uptake; guanosine alleviated this effect. Al-induced reduction of net "Ca2+ uptake was not aggravated by mycophenolic acid. Net MCa2+ uptake was generally less at pH 4.5 than at 5.5 for all treatments. It is concluded that Al ions affect net "Ca2 uptake by binding to the verapamil-specific channel site that is different from the bepridilspecific one, as well as by interfering with the action of guanosine 5'-triphosphate-binding proteins.

Al toxicity is a major growth-limiting factor in acid soils throughout the world (12). Al-related reduction in root surface area impedes accumulation of most nutrients. In addition, several studies showed that lower internal concentrations of divalent cations, especially Ca and Mg, in Al-stressed plants were due to direct interference of Al with the efficiency of membrane-transport mechanisms (18, 23). Al-stressed plants occasionally contain such low concentrations of Ca that symptoms of Al toxicity resemble those of Ca deficiency (12). Transduction of environmental and hormonal signals to the responsive elements of cell metabolism relies on Ca2` acting as a second messenger (21). Low cytoplasmic Ca2' concentration that is essential for such a role of Ca2e is maintained through its efflux mediated by the energy-dependent plasma membrane Ca2+-ATPases (for a review, see ref. 5) and through sequestration of Ca2e within different cell organelles (9). The calcium concentration in the apoplast is very high (15), ' Financial support from The Finders University Research Budget (to Z.R.). 2Present address: Department of Plant Science, Waite Agricultural Research Institute, University of Adelaide, Glen Osmond, S.A. 5064, Australia.

632

MECHANISM OF Al INHIBITION OF NET 45Ca2+ UPTAKE

ATPase and 45Ca2" transport depend on the experimental system used (5, 22 [and references therein]). The objectives of the present study were to elucidate certain aspects of the physiological mechanism of Al effects on 15Ca2+ fluxes across the plasma membrane of Amaranthus protoplasts by using Ca2" channel blockers as well as inhibitors of Ca2+-ATPase, calmodulin, and GTP-binding proteins. MATERIALS AND METHODS Protoplast Isolation Methods used in this study were similar to those described earlier (10, 1 1). Two grams of Amaranthus tricolor var tricolor seeds (Yates Co., Sydney, Australia) were germinated on foam pads thoroughly soaked with 0.2 mm CaCl2 in the dark at 25 + 1°C for 88 h. Half-seedlings (cotyledons and approximately 5 mm of hypocotyl) were cut off and finely chopped with a razor blade in a small amount of sterile washing solution in a Petri dish (0.5 M sorbitol, 5 mm Mes-KOH, 1 mm CaCl2, 0.5 mM MgCl2, 1 mM KCI, 0.5% (w/v) PVP 3603, 0.05% (w/v) BSA; pH 6.0). The same solution was used to wash the sliced material twice by pressing a Pasteur pipette against the bottom of the dish. The sliced material was then resuspended in 10 mL of enzyme solution (0.4% [w/v] Onozuka Cellulase RS, 0.05% [w/v] pectolyase-Y23, 1% [w/v] BSA; sorbitol, Mes, CaCl2, MgCl2, KCI, and PVP 360 as in washing solution; pH 6.0), evacuated for 3 min, and incubated in the dark at 30°C for 75 min with continuous shaking (60 oscillations/min). Following incubation, the sliced material was filtered through eight layers of cheesecloth, and the filtrate centrifuged at 70g, the pellet was resuspended in 2 mL of washing solution and left on ice. The sliced material was reincubated in fresh 10 mL of enzyme solution for an additional 75 min followed by filtration and centrifugation as before. Combined yield of protoplasts was washed once in washing solution and the second time in the basal medium (0.5 M sorbitol, 5 mM MesKOH, 1 mm CaCl2, 1 mm KCI, 0.1% [w/v] BSA). The pH of the basal medium was adjusted either to 4.63 ± 0.03 (mean + SE) with HCI or to 5.78 ± 0.03 with KOH for the Ca uptake measurements subsequently performed at pH values of 4.5 and 5.5, respectively. Following the second wash, protoplasts were resuspended in the enriched basal medium (concentration of all ingredients 1.08-times higher than in the basal media described above) and kept on ice while subsampling for the uptake experiments. Generally, not more than 2.5 h elapsed between the second wash of protoplasts and completion of all uptake measurements. The two-step preparation process using low enzyme concentrations resulted in protoplast preparations being relatively free ofdebris as revealed by routine microscopic observations. Protoplasts were thus used in the uptake experiments without a gradient purification step (10). Viability of protoplasts was routinely checked by observing either green fluorescence 3 min after staining protoplasts with fluorescein diacetate or dye exclusion 15 min after staining

'Abbreviations: PVP 360, PVP having approximately molecular mass of 360 kD; HSD, Tukey's honestly significant difference.

633

protoplasts with trypan blue. Observations were made on a Leitz Dialux 20EB UV/phase contrast microscope equipped with a mercury lamp. Measurements of Time Course of Net 45Ca2+ Uptake

Uptake experiments were performed in 108 AL of the mixture that was made up of 90 ,L of protoplasts suspended in the 108%-strong basal medium, 10 AL of the same enriched basal medium supplemented with 45Ca2+ to give the final activity of 90 MBq/mL, and 8 ,uL of diluted HCI (approximately 1 mM; pH 3.0) containing Al to give final concentrations in the uptake medium of 2, 10, 50, and 150 ,uM Al. Diluted HCI without added Al was included as a control. After addition of HCl-Al, the final pH values of the uptake media were 4.5 ± 0.03 or 5.5 ± 0.04 (mean ± SE). Measurements of pH values of all the media were performed at the temperature at which uptake of 45Ca2+ was subsequently monitored (30°C). An uptake experiment was initiated on ice by adding HClAl solution to protoplast suspension followed immediately by 45Ca2' addition. Duplicate samples were taken (time zero), and the tube was transferred to a water bath set at 30°C and 60 oscillations/min. Depending on experimental objectives, additional samples were taken at 1.5, 3, 6, 10, and 15 min. At the time of sampling, an aliquot of 20 ,L of uptake medium was withdrawn and layered above a discontinuous gradient consisting of (from the top to the bottom of the microfuge tube) 75 ,L of stopping solution (0.5 M sorbitol, 5 mM Mes-KOH, 4 mm EGTA, 0.1 % (w/v) BSA [pH 6.0]), 70 AL of silicone oil (specific gravity 1.07 at 25°C), and 20,uL of 10% (v/v) HC104. Following 15 s of centrifugation in a Beckman microfuge, tubes were frozen in liquid N2 and cut at the perchloric acid-silicone oil interface. The tube bottom was placed in 0.3 mL of 0.1 % (v/v) Triton X- 100 and vortexed to resuspend protoplasts. A 3.7-mL aliquot of scintillation mixture (0.35% [w/v] 2,5-diphenyloxazole, 0.02% [w/v] pbis[o-methylstyryl]benzene, 25 % [v/v] Triton X-1 14, 75 % [v/ v] xylene) was added, the scintillation tube vigorously shaken, and activity recorded on a liquid scintillation counter (Beckman LS 500OTD). A blank sample handled in the same way but without labeled Ca was included in each experiment; this counting was deducted from all measurements. In addition, counting obtained for a particular treatment at zero time was subtracted from the results obtained at later times for the same treatment. Effects of Various Inhibitory Substances on Net 45Ca2+ Uptake

Protoplasts were preincubated for 20 min in a water bathshaker set up at 30°C and 60 oscillations/min. Preincubation medium was prepared by mixing 100 ,AL of protoplasts suspended in basal medium, 2.5 ,uL of dissolved test substance, and 8,uL of diluted HCI (pH 3.0) supplemented with enough Al to give the final total Al concentration of 0, 10, or 50 AM at pH 4.5 ± 0.03 or 5.5 ± 0.04 (mean ± SE) as measured at 300C. Following preincubation, tubes were allowed to cool on ice for 1 min before 4 ,uL of 45Ca2+ (dissolved in the medium of

634

Plant Physiol. Vol. 98, 1992

RENGEL AND ELLIOTT

the same composition as the preincubation medium) was added (final activity 90 MBq/mL of the uptake medium). Sampling and radioactivity determination were performed as described above. Test substances were dissolved in water (verapamil, bepridil, guanosine), dimethyl sulfoxide (erythrosin B and calmidazolium), or ethanol (mycophenolic acid). Control treatments containing the same concentration of dimethyl sulfoxide or ethanol but no test substance were included in the corresponding experiments. Because calmidazolium may adhere to glass surfaces (manufacturer's note), plastic tubes were used for all preincubation and uptake media throughout this study. Statistics For each uptake experiment, two separate protoplast preparations were combined to create a more representative sample. Dilution of protoplast suspensions was adjusted such that a 20-,gL sample of the uptake medium contained between 30,000 and 35,000 protoplasts. Each experiment was run in duplicate and was repeated three to four times (occasionally as many as eight times). Results reported here represent averages of six to eight replications. Results were analyzed by the multifactorial analysis of variance. Pairwise multiple comparisons among treatment means were performed using the Tukey's HSD procedure (6) that reduced the frequency of comparisonwise type I error. Expression of Results Results were expressed either per count or per surface area of protoplasts. Measurements showed that the average diameter of protoplasts was 18 ± 1 Am (mean ± SE, measured on four groups of at least 34 protoplasts each). The corresponding surface area was 1.018 x 10-9 m2/protoplast. Chemicals Labeled calcium (43Ca2+) was purchased from Amersham Laboratories (Amersham, England), silicone oil (fluid No. 550) from Dow Coming Australia (Blacktown, New South Wales), Mes and Hepes from BDH Chemicals (Poole, England), Pipes from Calbiochem (La Jolla, CA), calmidazolium from Janssen Pharmaceutica (Beerse, Belgium), and Onozuka Cellulase RS from Yakult Honsha (Tokyo, Japan). Other chemicals were from Sigma (St. Louis, MO).

An omission of MgCl2 and KCI from the washing medium as well as omission of KCI from the uptake media did not influence 45Ca2- uptake significantly (data not shown). However, these additions were retained as ingredients of the appropriate media as stated in "Materials and Methods." Chemical speciation of Al was attempted by using the modified GEOCHEM computer program with estimates of mononuclear (i.e. monomeric) Al species as determined by the adapted aluminon method. Depending on buffers tested, aluminon assay recovered between 93 and 99% of added Al (95-97% for Mes) at pH 4.5 and between 82 and 98% at pH 5.5 (82-95% for Mes), thus indicating little formation of solid phases. However, because of a lack of published thermodynamic constants for binding of Al and most other cations to albumin and different buffers used in this study, a number of assumptions had to be introduced, making exact speciation impossible (data not shown). Accumulation of 45Ca2+ increased significantly during the first 6 min (0 Al treatment) or 3 min (2-150,gM Al treatments) and then leveled off or declined (Fig. 1). An increase in 45Ca2" accumulation was more pronounced at higher pH values and lower Al concentrations (interactions time x pH and time x Al were both highly significant, P < 0.0001). A deleterious Al effect was detected already at the first measurement (1.5 mn after the start of uptake measurements) and increased with time (Fig. 1), making the time x Al interaction significant (P < 0.0001).

(A

0

0 0 co 0 0 r-

1000 ci

E

C 0. 0 (UE

E

500

EE c_

°

0 (U

C

ai

-I

c 0

E

m

Q

2500

E

4 2000 C%

.G-

+

'U

U3 U

N

RESULTS Preliminary experiments showed that protoplasts were stable for several hours on ice regardless of the medium in which they were suspended. However, at 30°C, viability of protoplasts as estimated after staining with fluorescein diacetate decreased rapidly reaching 73, 92, 82, and 90% after 20 min in medium containing no buffer, Mes, Pipes, and Hepes, respectively (Al-free media at pH 4.5). These figures reduced to 48, 81, 57, and 70%, respectively, if the medium was supplemented with 150 M Al. The corresponding figures for the media at pH 5.5 were higher, indicating higher viability of protoplasts. Further uptake experiments were performed using Mes buffer only.

1500

CAco

r0

1000

500 6

9

Time (min)

Figure 1. Time course of the effect of Al and pH on 45Ca2+ accumulation measured at 300C and following nominal Al concentrations (in gM): 0 (0), 2 (U), 10 (A), 50 (@), and 150 (E). Tukey's HSDom values for significant treatment factors and interactions were (in nmol/m2): time, 747; Al, 630; pH, 418; time x Al, 1541.

MECHANISM OF Al INHIBITION OF NET 45Ca2+ UPTAKE

16

A. pH 4.5

E2

1218

I-

U

m

7

0 Al 10ijM Al 50 piM Al

N

E 0

7,7; 4

E vZKl

Cu ..O

.

.e0~

-

none

co

L zr

--~

BEPRIDIL

VERAPAMIL

D 16 co

635

In contrast, when combined with verapamil, Al could not reduce net "5Ca2" uptake beyond the reduction already achieved by verapamil alone. As a consequence, the interaction blockers x Al was highly significant (P < 0.0001). The same level of significance was also noted for the interaction blockers x pH because an increase in pH caused an increase in net 45Ca2" uptake, especially if no channel blocker was added. Erythrosin B (2',4',5',7'-tetraiodofluorescein), an inhibitor of plasma membrane Ca2+-ATPase (22), and calmidazolium (R2457 1), a calmodulin inhibitor (13), both greatly increased net 45Ca2" uptake from the uptake media containing no Al (Fig. 3). When Al was added, net 45Ca2+ uptake was reduced to approximately the same level regardless of the presence or absence of inhibitors in the uptake media. However, this reduction was much larger if compared to inhibitor-containing zero-Al uptake medium, making the interaction inhibitors x Al significant (P < 0.0001). Differences in net 45Ca2+ uptake

an( t4.. 12 a)

z

8

4 Cn

0

E

Ca-channel blockers Figure 2. Influence of the relationship between Ca2+ channel blockers and Al on net 45Ca2+ uptake at two pH levels. Bepridil and verapamil were tested at 1 0 AM concentration each. Pretreatment lasted for 20 min. Uptake was measured 1.5 min after the start of the uptake experiment. Tukey's HSDo.O5 values for the significant treatment factors and interactions were (in nmol/m2 .s): blockers, 0.5; Al, 0.5; pH, 0.4; blockers x Al, 0.9; blockers x pH, 0.7.

Z-. 0 E

4..O

ca 0. C

0

none

0.

a

ERYTHROSIN B CALMIDAZOLIUM

B. pH 5.5 X/

40 .

z 30 . An increase in Ca2" activity resulted in an almost linear increase in net 45Ca2+ uptake, whereas elevated Al levels caused a decrease in net 45Ca2+ uptake at all Ca2+ activities (data not shown). However, net 45Ca2+ uptake was less inhibited by Al if the uptake medium contained Ca2' at higher activities (2.0 compared to 0.2 mM), indicating an alleviating effect of Ca2+. In the following experiments, protoplasts were preincubated in the medium containing the specified channel blocker or inhibitor (with or without Al added) for 20 min. To offset for the differences in reduction of protoplast viability during preincubation in various media, measured values of net 45Ca2+ uptake were expressed on the basis of viable protoplasts only. Calcium-channel blockers, bepridil and the phenylalkylamine verapamil, as well as Al ions caused a large decrease in net 45Ca2+ uptake (Fig. 2). The addition of Al to the uptake media containing bepridil resulted in a net 45Ca2+ uptake reduction that was larger than when bepridil was added alone.

20

10 0

none

E RYTHROSIN B CALMIDAZOLIUM

Inhibitors Figure 3. Effects of pH, Al, and inhibitors of plasma membrane Ca2+ATPases on net 45Ca2+ uptake. Concentration of erythrosin B was 1 ,M and that of calmidazolium 20 Mm. Uptake was measured 1.5 min after the start of the experiment that followed the 20-min pretreatment. Tukey's HSDo.o5 values for the significant treatment factors and interactions were (in nmol/m2.s): inhibitors, 1.0; Al, 1.0; pH, 0.8; inhibitors x Al, 1.7; inhibitors x pH, 1.4; Al x pH, 1.4.

636

RENGEL AND ELLIOTT

caused by different pH levels (increased uptake at pH 5.5) were the largest if no Al was added to the uptake media (Al x pH interaction significant, P < 0.0001). Mycophenolic acid, an inhibitor that affects the operation of GTP-binding proteins (19), reduced net 45Ca2" uptake (Fig. 4) but only if no Al was added to the uptake media. The addition of guanosine to the uptake media containing mycophenolic acid restored net 45Ca2+ uptake to the control level. If Al and mycophenolic acid were both added, guanosine not only diminished an inhibitory effect of mycophenolic acid but even increased net 45Ca2` uptake above the level present in the corresponding treatment with Al added only (Fig. 4). DISCUSSION Results of the present study are in accordance with the previously reported effects of calmidazolium, mycophenolic acid, and guanosine on net 45Ca2' uptake (1 1). In addition, Al was shown to significantly reduce net 45Ca2+ uptake. Inhib-

11

en

8

O

4

E

E

ca 0

4

0)

.X +

O-,

16

0o 12

z

8 4

o

Figure 4. Net 45Ca2+ uptake as influenced by pH, Al, mycophenolic acid (MYCOPH.A.), and guanosine (GUAN.). Mycophenolic acid and guanosine were tested at the concentration of 25 and 50 Mm, respectively. Uptake was measured 1.5 min after commencement of the experiment that followed the 20-min pretreatment. Tukey's HSDO.05 values for the significant treatment factors and interactions were (in nmol/m2.s): between treatments, 0.7; Al, 0.7; pH, 0.6; treatment x Al, 1.2.


itory effects of Al were measurable after only 1.5 min exposure, indicating the effect may be much quicker, maybe practically instantaneous (Fig. 1). Bepridil and verapamil each caused a large reduction in net 45Ca2+ uptake by Amaranthus protoplasts (Fig. 2). Similar results were obtained with carrot protoplasts where binding of bepridil and verapamil to a channel receptor site was well correlated with a reduction in Ca2+ influx, with bepridil being more inhibitory than verapamil (14). In our study (Fig. 2), both Ca2+-channel blockers showed approximately the same level of reduction in net 45Ca2+ uptake. However, addition of Al to the bepridil-containing uptake medium increased reduction of net 45Ca2+ uptake. Such an additive effect of Al was not detected in the case of verapamil. These results indicate that Al may bind to the verapamil-binding sites, whereas bepridil appears to bind to different sites. In addition, the additive character of inhibitory effects of Al and bepridil may suggest that more than one binding site is responsible for channel functioning. Verapamil inhibited 45Ca2+ fluxes across the plasma membrane of carrot protoplasts (14), caused a transient reduction in cytosolic Ca2+ activity of tomato and oilseed rape roots (7), but failed to exert significant effect on 45Ca2+ uptake by oat coleoptile cells (30). In addition to occasional differences in responses to verapamil and other channel blockers that have been measured in various experimental systems, a wide range of possible secondary effects may occur after application of different channel blockers (28). In the present study (Fig. 2), a similarly large reduction of net 45Ca2+ uptake caused by verapamil and bepridil, two Ca2`-channel blockers of different chemical structure, suggested the effect of these channel blockers on Ca2+-trinslocating channels in the plasma membrane of Amaranthus protoplasts. Verapamil-binding protein which is likely a Ca2+ channel has been isolated from the corn coleoptiles (16). Because this verapamil-binding protein can be functionally incorporated into a planar bilayer membrane (29) and because Al appears to bind to the verapamil-binding site (the present study, Fig. 2), further detailed studies of the mechanism of Al inhibition of Ca2+ transport may be done by using the techniques of incorporation of channel proteins into bilayer membranes. Al was also shown to inhibit inward-rectifying conductance of K+ channels in the plasma membrane of V. faba guard cells (26). In that study, Schroeder reported that the KD for Al3+ ions was 15 gM; this concentration is within the range of Al concentrations that inhibited 45Ca2+ fluxes across the plasma membrane of Amaranthus protoplasts (between 4 and 30 mm activity of monomeric Al species, data not shown; see also Fig. 2 in the present study). Schroeder (26) attributed inhibitory effects to Al3" ions even though he used quite a complex medium at pH 5.5. Chemical speciation of Al in our study showed that a very small percentage of total monomeric Al species was present as Al3+ (8-9% of total Al at pH 5.5). The questions of what Al species inhibited inward-rectifying conductance of K+ channels in V. faba guard cells (26) as well as Ca2` fluxes through the Ca2`-translocating channels in the plasma membrane of Amaranthus protoplasts (Fig. 2) remain open. The data from Figure 3 indicate the involvement of Ca2+ATPase in Ca2+ efflux from Amaranthus protoplasts, a result

MECHANISM OF Al INHIBITION OF NET 45Ca2l UPTAKE

that was also shown by direct measurement of Ca24-ATPase action in several other experimental systems (5, 9, 22). Efflux of Ca2+ that is driven through the plasma membrane of Amaranthus protoplasts by Ca2`-ATPase appeared to depend on calmodulin activation (Fig. 3). An increase in net Ca2` uptake that came as a result of an abolishing of Ca2+ efflux by either erythrosin B or calmidazolium was almost completely dissipated by the addition of Al, indicating no interaction between Al and calmodulin-dependent Ca24-ATPase. Because Al acted on the Ca2` channel preventing Ca2+ influx into Amaranthus protoplasts (Fig. 2), the inhibition of Ca2` efflux should not result in a large increase in net Ca2` uptake (this is what the data from Fig. 3 suggest). Calmidazolium is a very specific inhibitor of calmodulin, more specific than phenothiazines ( 17 [and references therein]). However, calmidazolium may have a range of side effects, one being an induction of lysis demonstrated on oat protoplasts (at the calmidazolium concentration of 5 ,M [17]). Calmidazolium was tested in the present study at 20 UM. It reduced viability of Amaranthus protoplasts slightly more than other tested substances, but no statistical significance of that difference was detected (data not shown). The data in Figure 4 indicate the involvement of GTP proteins in the regulation of Ca2" fluxes across the plasma membrane ofAmaranthus protoplasts; this result is in accordance with findings from a previously reported study (1 1). It appears that Al ions may interfere with GTP-binding proteins because the presence of mycophenolic acid did not aggravate Al-induced reduction of net 45Ca2+ uptake. In addition, guanosine partly alleviated reduction of net 45Ca2' uptake caused by Al and mycophenolic acid. Al ions may deplete the cell pool of GTP because of their high affinity toward phosphate. As a consequence, providing an additional external supply of guanosine partly dissipated Al inhibitory effects. However, for the interaction between guanosine and Al to occur, Al ions should have entered the protoplast. The mechanism of Al transport into the cytoplasm and the identity of Al species transversing the intact plasma membrane is still more subject to theoretical predictions than experimental measurements (1 [and references therein]). Because protoplasts lack the cell wall whose negative charges strongly bind Al ions (24), Al ions may enter protoplasts much faster (seconds or minutes, see Fig. 1). We suggest that Al ions use the same channelprotein through which Ca2" ions pass into cytoplasm. A competitive relationship of Ca2` and Al ions takes place at the apoplast-facing channel mouth while Ca2` and Al ions are collapsing a large electrochemical gradient (negative inside the cell) by moving into the cytoplasm. It would be premature to make any suggestion concerning what Al species can cross the plasma membrane through a channel-protein because our unpublished data suggest that all monomeric Al ions might have a role in inhibiting net 45Ca2+ uptake. A stunning similarity was demonstrated in the response of maize root cap cells to either deprivation of Ca2+ (2) or addition of Al (3). Both treatments were suggested (but not shown) to reduce the influx of Ca24 into cytosol, the situation that may trigger changes in cellular differentiation and root cap function through changes in the activity and appearance of Golgi apparatus previously linked to Al treatment (3). Such observations may be compared with results presented here

637

and, consequently, extended into a far-reaching hypothesis that Al exerts its toxic effects primarily through alteration of the Ca24 homeostasis in plant cells, disturbance of which could result in a full range of deleterious Al effects on cell metabolism including reduced mitotic activity. This hypothesis will be elaborated upon elsewhere (Z. Rengel, manuscript submitted for publication). Further research that would measure the effects of Al ions on cytoplasmic free Ca24 concentration, instead of on 45Ca2" fluxes across the plasma membrane, is warranted. Measurement of Al effects on Ca24-translocating, verapamil-binding protein functionally incorporated into planar lipid bilayers may be indicated. ACKNOWLEDGMENTS Our thanks are due to Dr. K. Dixon (Flinders University) for letting us use a UV/phase contrast microscope and to Dr. K. Tiller (Commonwealth Scientific and Industrial Research Organization) for the gift of aluminon. Financial support from The Flinders University Research Budget (to Z.R.) is greatly appreciated. LITERATURE CITED 1. Akeson M, Munns D (1990) Uptake of aluminum into root cytoplasm: predicted rates for important solution complexes. J Plant Nutr 13: 467-484 2. Bennet RJ, Breen CM, Bandu VH (1990) A role for Ca2+ in the cellular differentiation of root cap cells: a re-examination of root growth control mechanisms. Environ Exp Bot 30:

515-523 3. Bennet RJ, Breen CM, Fey MV (1987) The effects of aluminium on root cap function and root development in Zea mays L. Environ Exp Bot 27: 91-104 4. Bock JL (1980) The binding of metal ions to ATP: a proton and phosphorus NMR investigation of diamagnetic metal-ATP complexes. J Inorg Biochem 12: 119-130 5. Briskin DP (1990) Ca2+-translocating ATPase of the plant plasma membrane. Plant Physiol 94: 397-400 6. Carmer SG, Walker WM (1985) Pairwise multiple comparisons of treatment means in agronomic research. J Agron Educ 14: 19-26 7. Clarkson DT, Brownlee C, Ayling SM (1988) Cytoplasmic calcium measurements in intact higher plant cells: results from fluorescence ratio imaging of fura-2. J Cell Sci 91: 71-80 8. Dill ET, Holden MJ, Colombini M (1987) Voltage gating in VDAC is markedly inhibited by microsomal quantities of aluminum. J Membr Biol 99:187-196 9. DuPont F, Bush DS, Windle JJ, Jones RL (1990) Calcium and proton transport in membrane vesicles from barley roots. Plant Physiol 94: 179-188 10. Elliott DC, Petkoff HS (1990) Measurement of cytoplasmic free calcium in plant protoplasts. Plant Sci 67: 125-131 11. Elliott DC, Yao Y (1989) Cytokinin and fusicoccin effects on calcium transport in Amaranthus protoplasts. Plant Sci 65: 243-252 12. Foy CD (1988) Plant adaptation to acid, aluminum-toxic soils. Commun Soil Sci Plant Anal 19: 959-987 13. Gietzen K, Wutrich A, Bader H (1981) R24571: a new powerful inhibitor of red blood cell Ca2-transport ATPase and of calmodulin-regulated functions. Biochem Biophys Res Commun 101: 418-425 14. Graziana A, Fosset M, Ranjeva R, Hetheringthon AM, Lazdunski M (1988) Ca24 channel inhibitors that bind to plant cell membranes block Ca24 entry into protoplasts. Biochemistry 27: 764-768 15. Hanson JB (1984) The function of calcium in plant nutrition. Adv Plant Nutr 1: 149-208 16. Harvey HJ, Venis MA, Trewavas AJ (1989) Partial purification

638

17.

18.

19.

20.

21. 22.

RENGEL AND ELLIOTT of a protein from maize (Zea mays) coleoptile membranes binding the Ca2"-channel antagonist verapamil. Biochem J 257: 95-100 Kell A, Donath E (1990) Possible involvement of calmodulin in the maintenance of plasma membrane integrity in oat protoplasts. Biochem Physiol Pflanzen 186: 409-416 Keltjens WG (1988) Short-term effects of Al on nutrient uptake, H+- efflux, root respiration and nitrate reductase activity of two sorghum genotypes differing in Al-susceptibility. Commun Soil Sci Plant Anal 19: 1155-1163 Lee HJ, Pawlak K, Nguyen BT, Robins RK, Sadee V (1985) Biochemical differences among four inosinate dehydrogenase inhibitors, mycophenolic acid, ribavirin, tiazofurin, and selenazofurin studied in mouse lymphoma cell culture. Cancer Res 45: 5512-5520 Matsumoto H, Yamaya Y (1986) Inhibition of potassium uptake and regulation of membrane-associated Mg2+-ATPase activity of pea roots by aluminium. Soil Sci Plant Nutr 32: 179-188 Poovaiah BW, Reddy ASN (1987) Calcium messenger system in plants. CRC Crit Rev Plant Sci 6: 47-103 Rasi-Caldogno F, Pugliarello MC, Olivari C, De Michelis MI (1989) Identification and characterization of the Ca2+-ATPase which drives active transport of Ca2' at the plasma membrane of radish seedlings. Plant Physiol 90: 1429-1434


23. Rengel Z, Robinson DL (1989) Competitive Al3" inhibition of net Mg2" uptake by intact Lolium multiflorum roots. I. Kinetics. Plant Physiol 91: 1407-1413 24. Rengel Z, Robinson DL (1989) Aluminum and plant age effects on adsorption of cations in the Donnan free space of ryegrass roots. Plant Soil 116: 223-227 25. Robinson C, Larsson C, Buckhout TJ (1988) Identification of a calmodulin-stimulated (Ca2" + Mg2+)-ATPase in a plasma membrane fraction isolated from maize (Zea mays) leaves. Physiol Plant 72: 177-184 26. Schroeder JI (1988) K+ transport properties of K+ channels in the plasma membrane of Viciafaba guard cells. J Gen Physiol 92: 667-683 27. Siegel N, Haug A (1983) Aluminum interaction with calmodulin. Evidence for altered structure and function from optical and enzymatic studies. Biochim Biophys Acta 744: 36-45 28. Tester M (1990) Plant ion channels: whole-cell and single-channel studies. New Phytol 114: 305-340 29. Tester M, Harvey HJ (1989) Verapamil-binding fraction forms Ca2' channel in planar lipid bilayers. In J Dainty, MI De Michelis, E Marre, F Rasi-Caldogno, eds, Plant Membrane Transport: The Current Position. Elsevier, Amsterdam, pp 277-278 30. Tretyn A (1987) Influence of red light and acetylcholine on 45Ca2+ uptake by oat coleoptile cells. Cell Biol Int Rep 11: 887-896