Sterol Modulation of Plasma Membrane H+-ATPase Activity ... · soybean (Glycine max 1.) phospholipids and various concentrations of individual sterols using either a freeze-thaw sonication

Plant Physiol. (1 997) 11 3: 163-1 74

Sterol Modulation of the Plasma Membrane H+-ATPase Activity from Corn Roots Reconstituted into Soybean Lipids

Anne Crandmougin-Ferjani’, lsabelle Schuler-Muller2, and Marie-Andrée Hartmann*

lnstitut de Biologie Moléculaire des Plantes, du Centre National de Ia Recherche Scientifique, 28 rue Goethe, 67083 Strasbourg Cedex, France

A partially purified H+-ATPase from the plasma membrane (PM) of corn (Zea mays L.) roots was inserted into vesicles prepared with soybean (Glycine max 1.) phospholipids and various concentrations of individual sterols using either a freeze-thaw sonication or an octylglucoside dilution procedure. Both methods yielded a functional enzyme that retained i t s native characteristics. We have investigated the effects o f typical plant sterols (i.e. sitosterol, stigmasterol, and 24-methylcholesterol) on both ATP hydrolysis and H+ pumping by the reconstituted corn root PM ATPase. We have also checked the influence of cholesterol and of two unusual sterols, 24-methylpollinastanol and 14a,24-dimethylcholest-8-en-3~-ol. Here we present evidence for a sterol modulation of the plant PM H+-ATPase activity. I n particular, cholesterol and stigmasterol were found to stimulate the pump, especially when present at 5 mol%, whereas all of the other sterols tested behaved as inhibitors at any concentration in proteoliposomes. In all situations H+ pumping was shown to be more sensitive to a sterol environment than was ATP hydrolysis. Our results suggest the occurrence of binding sites for sterols on the plant PM H+-ATPase.

The PM H+-ATPase from higher plant cells is a P-type ion-translocating ATPase that produces an inwardly directed proton electrochemical gradient across the PM, thereby providing the driving force for secondary transport of nutrients such as anions, amino acids, sugars, or hormones into the cells (Serrano, 1989; Michelet and Boutry, 1995). The activity of the PM H’-ATPase also contributes to the maintenance of intracellular pH, and regulation of this process has been proposed to mediate a broad range of cellular processes involved in the growth, development, and response of plants to environmental and hormonal signals (Marré and Ballarin-Denti, 1985; Serrano, 1989; Michelet et Boutry, 1995). It is now well established that this enzyme is very sensitive to changes in its lipid environment, and recently much work has been devoted to the requirement of the enzyme for various kinds of lipids (Cooke and Burden, 1990; Palmgren, 1991; Kasamo and Sakakibara, 1995). However, a minimal amount of attention

* Present address: Laboratoire de Cryptogamie et de Phyto- pathologie, Université du Littoral, 50 Rue Ferdinand Buisson, 62228 Calais, France.

Present address: Laboratoire de Pharmacologie et de Toxicolo- gie Fondamentales, 118 Route de Narbonne, 31062 Toulouse, France.

* Corresponding author; e-mail ma.hartmannt3ibmp-ulp.u- strasbg.fr; fax 33-388-35-84-84.

has been paid to sterols, which are the other major components of the PM (Hartmann and Benveniste, 1987) and might also participate in the regulation of the PM Ht- ATPase, as cholesterol does in the case of the Na+,K+- ATPase (Yeagle et al., 1988; Cornelius, 1995) or the sarcoplasmic reticulum Ca*+-ATPase (Simmonds et al., 1982; Ding et al., 1994).

Plant cells contain a mixture of sterols (sitosterol, stigmasterol, and 24-methylcholesterol) instead of the one major sterol-cholesterol or ergosterol-found in mammalian or funga1 cells, respectively. In the present study we have investigated the effects of various individual sterols on ATP hydrolysis and ATP-driven H+ translocation by a partially purified PM H+-ATPase reconstituted into soybean (Glycine max L.) phospholipid vesicles by means of two different methods. We have examined the effects of the typical plant A5-sterols (i.e. sitosterol, stigmasterol, and 24-methylcholesterol) as well as those of cholesterol and of two unusual sterols, 24-methylpollinastanol and 14a,24- dimethylcholest-8-en-3/3-ol (Fig. 1). These compounds were shown to accumulate in plants treated with sterol biosynthesis inhibitors belonging to two important classes of fungicides used in agriculture, morpholines and triazoles (Benveniste, 1986). In that context their effects on the PM H+-ATPase activity were important for us to study.

MATERIALS AND METHODS

Chemicals

Sitosterol (1) and stigmasterol (3) were purchased from Fluka. The structures of the various sterols used are illus- trated in Figure 1. Using GC analysis, sitosterol was found to contain 8% 24-methylcholesterol(2); and in stigmasterol, 3% sitosterol and 1% 24-methylcholesterol were present as sterol impurities. Cholesterol (4) and 24-methylcholesterol were from Sigma. Unusual sterols such as 24- methylpollinastanol (5) and 14a,24-dimethylcholest-8-en- 3/3-01 (6) were isolated as described previously (Schuler et

Abbreviations: ACMA, 9-amino-6-chloro-2-methoxyacridine; cholesterol(4), cholest-5-en-3p-01; DCCD, N,N’-dicyclohexylcarbodiimide; 24-methylcholesterol (2), (24[)-24-methylcholest-5-en-3/3- 01; 24-methylpollinastanol (5), (24~)-24,14a-dimethyl-9p,l9-cyclo- 5a-cholestan-3p-01; PC, phosphatidylcholine; PM, plasma membrane; sitosterol (l), (24R)-24-ethylcholest-5-en-3/3-01; stigmasterol (3), (24S)-24-ethylcholest-5,22E-dien-3p-o1; SW 26, 2,2,2-trichloro- ethyl-3,4-dichlorocarbanilate; Te,, time needed to reach equilib- rium after pH gradient relaxation by hexokinase.

163 www.plantphysiol.orgon March 30, 2020 - Published by Downloaded from

Copyright © 1997 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

164 Grandmougin-Ferjani et al. Plant Physiol. Vol. 11 3 , 1997

R = H, cholesterol(4) R = CH,, 24-methylcholesterol (2) R = C,H,, sitosterol (1) R = C2H, and A*', stigmasterol(3)

'h,

HO & 24-methylpollinastanol (5)

14a,24-dimethylcholest-8-en-3P-ol(6)

Figure 1 . Structures of sterols used in this study.

al., 1990). The phospholipids used in this study were crude soybean (Glycine max L.) phospholipids (L-a-PC, type IV-S) from Sigma. ~-Lyso-3-PC, l-[l-'"C]palmitoyl, and ['"C] DCCD were from Amersham and Commissariat B l'Energie Atomique (Saclay, France), respectively. Octylglucoside was purchased from Boehringer Mannheim. ACMA was obtained from Molecular Probes (Eugene, OR). A11 other biochemicals and reagents were from Sigma, Merck (Darm- stadt, Germany), or Aldrich.

Plant Material

Corn (Zea mays L. cv LG 11 or Pionner) caryopses were allowed to grow in the dark at 25°C. Roots were excised after 6 d.

lsolation of PMs

PM vesicles were isolated by differential and Suc density gradient centrifugation as previously described by either Grandmougin et al. (1989) (procedure 1) or by St. Marty- Fleurence et al. (1988) (procedure 2).

Reconstitution of the PM H+-ATPase by Freeze-Thaw Sonication (Procedure 1)

Solubilization and Partia1 Purification of the PM H+ -A TPase

The ATPase was solubilized in two successive steps. First, membrane proteins other than the ATPase were removed by treatment of the PM vesicles with Triton X-100 and lysoPC at a low concentration. PM vesicles (0.5-1 mg protein mL-') obtained after SUC density gradient centrifugation according to Grandmougin et al. (1989) were diluted (l:l, v /v) with 0.1 M Tris-HC1 (pH 8.0) containing 1 mM PMSF, 13 mM 2-mercaptoethanol, 20% (w,/v) glycerol, Triton X-100 (at a detergent-to-protein ratio of 7 [w/w]), and lysoPC (at a detergent-to-protein ratio of 3 [w/w]). After incubation for 20 min at 4"C, membranes were cen-

trifuged at 100,OOOg for 45 min. The supernatant S, was discarded. The pellet C, was resuspended in 2 mL of 10 mM Tris-HC1 (pH 7.5) with 20% (w/v) glycerol and 1 mM DTT, and the ATPase was extracted from the resulting membranes with a higher lysoPC concentration (lysoPC-to- protein ratio of 7 [w / w]). Following incubation for 10 min at room temperature, the nonsolubilized proteins were eliminated by centrifugation at 100,OOOg for 30 min (pellet C,). The supernatant S, (1 mL) was then applied on the top of a small column (60 X 10 mm) of SM, Bio-Beads (Bio- Rad) to remove lysoPC. After 15 min at 4"C, the column was centrifuged at 250g for 2 min. Control experiments with ['4C]lysoPC demonstrated that with this procedure, 90% of the lysoPC was eliminated. The final eluate containing the partially purified ATPase was used for reconstitution experiments.

Reconstitution

Lipid mixtures were prepared from stock solutions of type IV-S soybean PC (60 mg mL-l in benzene:ethanol[4:1, v/v]) and sterols (10 mg mL-l in chloroform). After drying the solvents under a stream of nitrogen, liposomes were formed by hydration of lipids (phospholipids only or mixed with a sterol at 30 mol%) in 2.5 mM Tris-Mes (pH 6.5) containing 1 mM DTT and 25 mM KCl to give a final lipid concentration of 10.7 mg mL-l. Samples were then sonicated to clarity under argon for 20 min. One volume of the partially purified ATPase (about 100 k g of protein) was mixed with 4 volumes of liposomes to obtain a lipid-to- protein ratio of 180 (w/w). The mixture was rapidly frozen at -70°C. After 20 min, the preparation was thawed in a water bath at room temperature. The suspension was sonicated for 2 min at 4°C and the whole cycle of freeze-thaw sonication was repeated once. Proteoliposomes were stored at 4°C until use.

ATP Hydrolysis Assay

ATPase activity of membrane fractions or proteoliposomes (10-50 pg of protein) was measured in the presence of 20 mM Tris-Mes (pH 6.5) containing 0.25 M SUC, 2.5 mM 2-mercaptoethanol, 3 mM MgSO,, 1 mM NaN,, 25 mM K,SO, 50 mM KC1, 100 mM KNO,, and 0.1 mM sodium molybdate in a total volume of 0.5 mL. NaN,, KNO,, and sodium molybdate were introduced to inhibit mitochondrial ATPase, vacuolar ATPase, and acid phosphatase, respectively. The reaction was initiated by the addition of 3 mM Tris-ATP. A control was run without membranes. A11 assays were carried out in triplicate. After 25 min at 3OoC, the reaction was stopped with 1 mL of 5% (w/v) SDS. Release of Pi was determined according to Chen et al. (1956).

Proton Transport Assa y

ATP-dependent H+ translocation was measured by the initial rale of change in ACMA fluorescence quenching. Proteoliposomes (100 pL, i.e. 7-10 pg of protein) were incubated in 20 mM Tris-Mes (pH 6.5) with 2.5 mM 2-mercaptoethanol, 50 mM KCl, 100 mM KNO,, 3 mM

www.plantphysiol.orgon March 30, 2020 - Published by Downloaded from Copyright © 1997 American Society of Plant Biologists. All rights reserved.


Sterol Effects on the Plasma Membrane H+-ATPase 165

MgSO,, and 1.8 p~ ACMA (in solution in DMSO) in a final volume of 1 mL. After equilibration, the reaction was initiated with 3 mM Na,ATP. The quenching of ACMA fluorescence was monitored at 25°C using an RF 5000 spectrofluorimeter (Shimadzu, Kyoto, Japan) with excitation and emission wavelengths set at 410 and 476 nm, respectively. Fluorescence quenching was reversed by the addition of 5 pg of gramicidin D (in solution in ethanol). Initial rates of H' accumulation were estimated from the initial slopes of ACMA fluorescence quenching and expressed as ( A F I F , x 100) min-' mg-' protein, where F , was the relative fluorescence intensity immediately after ATP addition. Maximum quenching was calculated as the difference between F, and the fluorescence intensity observed at a steady-state level.

Reconstitution by Octylglucoside Dilution (Procedure 2)

ATPase Solubilization

The procedure used here for solubilization and partia1 purification of the ATPase was different from the previous one. PM vesicles of corn roots were isolated according to St. Marty-Fleurence et al. (1988). The enzyme was solubilized in only one step with 2 mg mL-l lysoPC (lysoPC-to-protein ratio of 5 [w/w]). Nonsolubilized proteins were centrifuged at 106,OOOg for 45 min and the ATPase was recovered from the resulting supernatant by precipitation with 45% saturated ammonium sulfate overnight (St. Marty- FIeurence et al., 1988). After centrifugation, the pellet was resuspended in 1 mM Tris-Mes (pH 6.5) containing 20% (w/v) glycerol and 1 mM DTT and the mixture was centrifuged again. After resuspension in the same buffer, the preparation was stored at -70°C without loss of activity for at least a few months.

['4C]DCCD Labeling and SDS-PAGE of the Partially Purified PM H+-ATPase

Before performing reconstitution experiments, the purity of this partially purified PM H+-ATPase from corn roots was checked by SDS-PAGE of the preparation after labeling with [14C]DCCD. The partially purified PM H+-ATPase (220 pg of protein) was diluted in 5 mM Tris-Mes (pH 7.0) containing 250 mM SUC and 1 mM DTT to a total volume of 0.5 mL. [14C]DCCD (0.054 pCi) in ethanol was added to a final concentration of 10 p ~ . After incubation at 4°C for 90 min, the reaction was stopped by addition of 10 mL of the same buffer. The suspension was centrifuged at 100,OOOg for 30 min and the resulting pellet was resuspended in 150 pL of the denaturating buffer used for electrophoresis. For identification of the DCCD-labeled protein, samples (10-15 pg of protein) were heated at 30°C for 3 min and then subjected to SDS-PAGE on 10% acrylamide gels according to Laemmli (1970). Gels were stained with Coomassie bril- liant blue and exposed to x-ray autoradiograph film (Fuji, Tokyo, Japan) at -80°C for 3 weeks.

Reconstitution

PM Hf-ATPase was reconstituted by the octylglucoside dilution method used by Rossignol et al. (1989). Liposomes

were prepared from soybean phospholipids with or without different concentrations of sterols using the same stock solutions described above. Lipids were hydrated in 10 mM Mes-Tris (pH 6.5) containing 0.1 M KCl and 1 mM MgSO,, and sonicated in a bath for 20 min until clarity was reached. The efficiency of the reconstitution depends on the complete solubilization of the different components of the mixture. An important parameter is the detergent-to-lipid molar ratio. For PC:phosphatidylethanolamine (1:L w / w) liposomes, this ratio was found to be 2.9 (Rossignol et al., 1989). Because of the low solubility of sterols in octylglucoside, some modifications were made to this procedure. As a means of visualizing the solubilization process (Pa- ternostre et al., 1988), the turbidity of liposome preparations was measured at 400 nm as a function of the amount of octylglucoside added over a range of lipid (phospholipids plus sterol) concentrations from 1.7 to 10 mM. As ex- pected, transformation of lamellar structures into mixed micelles gave rise to three stage curves (data not shown), allowing a quantitative determination of the effective detergent-to-lipid molar ratio for a complete solubilization of lipids (Paternostre et al., 1988; Rossignol et al., 1989). The value of this ratio was found to be between 17 and 21 for soybean phospholipid vesicles containing 15 to 30 mol% sterol, compared with 2.7 for sterol-free vesicles. A detergent-to-lipid molar ratio of 25 was routinely used. In a typical experiment, 30 pL of liposomes (i.e. 1.5 mg of total lipids) was solubilized in 106 pL of 0.5 M octylglucoside. Reconstitution was performed directly in the cuvette of the spectrofluorimeter by introducing successively an aliquot (6.8 pL) of this solution, 10 pL of ATPase (about 1 pg of protein), and 1.3 mL of a dilution buffer (10 mM Tris-Mes, pH 6.5 containing 100 mM KC1, 1 mM MgSO,, and 0.5 p~ ACMA) to obtain a final octylglucoside concentration of 2 mM and a lipid-to-protein ratio of 75 (w/w) (Rossignol et al., 1989).

Proton Transport Assay

ATP-dependent H+ translocation was measured by the initial rate of change in ACMA fluorescence quenching at 30°C under the conditions described above. After equilibration, H+ pumping was initiated by addition of 1 mM Na,ATP. Four successive and independent measurements were carried out. The pH gradient was discharged by 0.5 p~ nigericin. For pH gradient relaxation experiments, yeast hexokinase (type V, Sigma) (42 units mL-') and 25 mM Glc were added at steady state, and passive H+ efflux was monitored until complete dissipation of the gradient in the presence or absence of 0.5 p~ valinomycin (Szponarski et al., 1991).

Estimation of the lntravesicular Volume of Proteoliposomes

The volume of proteoliposomes was estimated from the fluorescence of trapped calcein within reconstituted membrane vesicles according to the method of Oku et al. (1982). Proteoliposomes were prepared directly in the cuvette as described above by mixing the enzyme with soybean lipids in the absence or presence of 30 mol% cholesterol, sitos-



166

60 h e. E 50

'3 40

8 : 20

2 10

.-

30

.3 .- 3

O

60 h * E

E

50

i; 40 4

v

.- .- .-

30 o .$ 20 -

10

O

2

Grandmougin-Ferjani et ai. Plant Physiol. Vol. 11 3 , 1997

A

PI + PS LysoPC PG PE PC

B

P1 + PS LysoPC PG PE PC

Figure 2. Phospholipid composition of the type IV-S soybean PC from Sigma (A) and of PM vesicles from corn roots (B). Data are means of two distinct experiments (SE 2 15%).

terol, or stigmasterol in the buffer used for reconstitution experiments but including 2 nM calcein. The fluorescence of the suspension was measured before (Ftot) and after addition of CoCl, at a final concentration of 0.27 mM (Fi,,). F,,, represents the fluorescence of a11 calcein present, and Fi, represents the fluorescence of the interna1 compartment plus the unquenched fraction of the externa1 compartment. Subsequently, 9.8 p L of 20% (w/v) Triton X-100 was introduced in the cuvette to destroy the integrity of vesicles, and the resulting fluorescence was measured again (Ftotq). Flu- orescence changes were monitored at 25°C with excitation and emission wavelengths set at 490 and 520 nm, respectively. The volume of the aqueous compartment within the vesicles (percent of total) is given by (Fi, - Ftotq)/(Ftot - F,,,,) x mo.

Protein Determination

Protein concentration was determined by the method of Bradford (1976) with BSA as the standard.

Lipid Analysis

The purity of sterols of commercial origin was checked by GC analysis according to Hartmann and Benveniste (1987). Phospholipids were analyzed as previously described (Ullmann et al., 1987).

1

RESULTS

Reconstitution of the PM H+-ATPase from corn roots was achieved by either freeze-thaw sonication or an octyl-

glucoside dilution method. Both procedures first required the solubilization of the enzyme from PM vesicles and its partia1 purification. In both cases, the phospholipids used to reconstitute the H+-ATPase activity were soybean lipids commercially available from Sigma as the type IV-S soybean PC. Analysis of this crude lipid extract showed that it consisted of 67% (w/w) phospholipids with PC and phosphatidylethanolamine as the major components (81% of total phospholipids) (Fig. 2A); PC accounted for 57%. This type of phospholipid composition is not very different from that of native PM vesicles from maize roots (Fig. 28) (Grandmougin et al., 1989). Phytosterols were also detected and represented (0.01% [w/w]). Among the remaining components of the commercial preparation that have not been analyzed, the occurrence of orange pigments with likely antioxidant properties should be mentioned. Al- though pure soybean phospholipids readily form hy- droperoxides, which can be easily detected by their AZs0 when kept at room temperature for a night, the commercial preparation remains free of these compounds (data not shown). This observation is important when one knows that the membrane permeability of the bilayers is markedly affected by oxidation of unsaturated phospholipid fatty acyl chains (Tanfani and Bertoli, 1989).

Characterizatíon of the PM H+-ATPase Reconstituted According to Procedure 1

The first method used for reconstitution of the PM H+- ATPase from corn roots was the freeze-thawing procedure, the main steps of which are summarized in Table I. Treat- ment of PM vesicles with the mixture of Triton X-100 and a low concentration of IysoPC was shown to solubilize about one-half of the protein, but most of the ATPase activity remained in the pellet C,. Therefore, the supernatant SI was discarded. The subsequent treatment of the pellet with IysoPC solubilized 94% of the ATPase with a 7-fold higher specific activity than that of the PM-bound enzyme. Such an increase probably resulted partly from the reported activating effect of lysoPC on the H+-ATPase activity (Palmgren and Sommarin, 1989). After elimination of lysoPC by filtration on a column of SM, Bio-Beads and reconstitution into soybean lipid vesicles, the specific ac-

Table 1. Summary of ATPase purification and reconstitution by freeze-thaw sonication

ATP-Hydrolyzing Activitv

Main Steps Protein

pmol Pi min- mg- protein

mg

PM vesicles 1.8 0.75 First treatment with

Triton X-1 OO/lysoPC Supernatant (S , ) 1 .o 0.35 Pellet (C,) 0.85 2.3

Second treatment with IysoPC Supernatant (S,) 0.4 5.5 Pellet (C,) 0.5 0.2

After IysoPC removed 0.2 1.6 Reconstituted ATPase 0.2 1.85




tivity of the ATPase was about 2-fold higher than that of PM vesicles, i.e. 1.85 pmol Pi min-’ mg-l protein (Table I).

Both functionalities of the reconstituted ATPase, ATP hydrolysis and H+ translocation, were characterized and compared with those of the native enzyme in PM vesicles. Table I1 shows some properties of the ATP-hydrolyzing activity. The absence of inhibition by NaN,, KNO,, and sodium molybdate indicates no contamination of the reconstituted enzyme by mitochondrial, tonoplastic, or non- specific ATPases. Both enzymes were inhibited to a similar extent by DCCD and Ca2+, but the reconstituted enzyme shows an enhanced sensitivity to vanadate or to the herbi- cide derivative SW 26. After reconstitution, the enzyme retained a pH profile similar to that of native enzyme with a maximum at 6.5 (data not shown) and a net preference for Mg-ATP as the substrate among various nucleotides (Table 111).

H+ pumping by the PM H+-ATPase inserted into soybean phospholipid vesicles was monitored by ACMA fluorescence quenching. The enzyme was found to be able to drive an active Ht translocation in the presence of ATP and thus create an intravesicular acidification. The rapid initial quenching induced by ATP addition was followed by a steady-state (maximum) quench after a few minutes. Initial fluorescence quenching rates were found to be strictly dependent on the presence of Mg2+-ATP and were totally inhibited by 100 p~ orthovanadate (data not shown). Evidence for an electrogenic transport is given by reversion of quenching upon addition of gramicidin or nigericin. H+ pumping was not inhibited by NaN,, KNO,, or sodium molybdate (data not shown). Thus, both functionalities of the reconstituted enzyme exhibited properties similar to those of the PM-bound enzyme.

Characterization of the P M H+-ATPase Reconstituted According to Procedure 2

An alternative procedure was used for reconstitution of the PM H+-ATPase from corn roots. In this case, the start- ing material was a partially purified PM H+-ATPase pre-

Table II. Effect o f inhibitors on ATPase activity after reconstitution by freeze-thaw sonication

ATP hydrolysis was assayed as described in “Materials and Meth- ods.” DCCD and SW 26 were dissolved in ethanol and DMSO, respectively. Solvents were added at a final concentration of 1 %. Each assay was performed in triplicate (SE 2 15%).

lnhibitor Reconstituted ATPase PM Vesicles

% of control” % of controlh

f 1 mM NaN, 99 80 i- 100 p~ Na,M,O, 103 85

+ 100 p~ DCCD 65 65 + 100 p~ Na,VO, 43 50

+ 100 mM KNO, 1 O0 1 O0

+ 100 pM sw 26 40 55 + 300 ~ L M CaCI, 35 33

a Control activity of the reconstituted enzyme: 1.5 pmol Pi min-’ Control activity in PM vesicles: 0.8 pmol Pi mg-’ protein.

min-’ mg-’ protein.

Table 111. Nucleotide-hydrolyzing activity o f the ATPase reconstituted by freeze-thaw sonication

was added at a final concentration of 3 mM. Assays were performed as described in Table I I . Each nucleotide

Substrate

ATP GTP CTP UTP ADP UDP PNPP

Reconstituted ATPase

% of control“

1 O0 O 4

12 17 17 8

PM Enzyme

% of controlb

1 O0 50 20

30 50 20

c -

a Control activity of the reconstituted enzyme: 1.5 pmol Pi min-’ Control activity in PM vesicles: 0.8 @mo1 Pi mg-’ protein.

min-’ mg-’ protein. -, Not determined.

pared according to St. Marty-Fleurence et al. (1988). When proteins from this preparation were separated by SDS- PAGE and stained with Coomassie blue, an intense band corresponding to a polypeptide of about 100 kD was obtained, indicating a clear enrichment of the preparation in ATPase. However, the presence of other bands attested that the preparation was not completely pure. When the electrophoresis was performed after labeling of the preparation with [14C]DCCD, only the 100-kD band was visual- ized on a fluorograph of the gel (data not shown), further supporting the identity of a proton-translocating ATPase (Solioz, 1984).

For reconstitution experiments, we used the octylglucoside dilution method described by Rossignol et al. (1989): after solubilizing the preformed liposomes and the enzyme in octylglucoside, the suspension of mixed micelles was rapidly diluted with a buffer without detergent to obtain a final detergent concentration lower than 20 mM, the critical micellar concentration of octylglucoside. In the case of the enzyme from corn roots, reconstitution conditions for op- timal proton transport rates by the ATPase were obtained for a lipid-to-protein ratio of 75 (w/w) and a final octylglucoside concentration of 2 mM (Rossignol et al., 1989). However, because of the low solubility of sterols in octylglucoside, a detergent-to-lipid molar ratio of 25 instead of 2.9 was used, as indicated in “Materials and Methods,” and, consequently, a higher factor of dilution was required to form proteoliposomes. As depicted further, we could perform ATP-driven H+ translocation measurements with such diluted preparations, but we had some difficulty ac- curately monitoring the ATP hydrolysis activity, mainly because of the lower sensitivity of the colorimetric compared with the fluorometric method. Some properties of the PM H+-ATPase from corn roots reconstituted according to this method are described in papers by St. Marty- Fleurence et al. (1988) and Rossignol et al. (1989).

Effects of Sterols on ATP Hydrolysis and H+ Pumping by the Reconstituted PM H+-ATPase

The partially purified ATPase has been reconstituted by both procedures into soybean lipids in the presence of



168 Grandmougin-Ferjani et al. Plant Physiol. Vol. 113, 1997

various sterols over a range of concentrations between O and 30 mol%. The protein-to-lipid ratio was kept constant. We selected the following sterols: sitosterol (I), 24- methylcholesterol (2), and stigmasterol (3)’ the typical A5- sterols of plant membranes; cholesterol(4), the major sterol of mammalian membranes; and two unusual sterols, 24- methylpollinastanol (5) and 14a,24-dimethylcholest-8-en- 3P-01 (6), which have been shown to accumulate in plants treated with morpholine and triazole fungicides, respectively (Benveniste, 1986) (Fig. 1). It should be noted that the maximum value of 30 mol% sterol in proteoliposomes is similar to the free sterol content (as represented by 55% stigmasterol, 14% sitosterol, 27% 24-methylcholesterol, and 2% cholesterol) of PM vesicles from corn roots (Grandmou- gin et al., 1989).

ATP Hydrolysis

The incorporation of 30 mol% sitosterol, stigmasterol, or 24-methylcholesterol, which is shown in Table IV, did not significantly affect ATP hydrolysis by the ATPase reconstituted according to procedure 1. In contrast, the presence of 24-methylpollinastanol or 14a,24-dimethylcholest-8-en-3P- o1 at the same concentration was found to inhibit this reaction with an efficiency of 60 and 30%, respectively. It is also shown in Table IV that these sterols decrease the sensitivity of the enzyme toward Ca2+ and vanadate.

Effects of sterols on ATP hydrolysis by the ATPase reconstituted by procedure 2 have also been investigated. Despite low activity, the reaction was observed to be linear for 60 min and not significantly affected by sterol content or structure (data not shown). The activity of the ATPase inserted with 15 or 30 mol% sitosterol was tested for its sensitivity to Brij 58, a detergent reported not to modify kinetic properties of the enzyme (Palmgren et al., 1990). No increase in activity was detected over the range of detergent-to-protein ratios (w/w) from 0 to 10 (data not shown). This absence of latency suggests that these sitosterol-containing proteoliposomes likely had the same orientation, with the catalytic site on the outside of the vesicles and thus directly accessible to ATP.

H + Pumping

Figure 3 shows the effects triggered by various sterols present at 30 mol% on the initial rate of H+ uptake by the

ATPase reconstituted according to procedure 1. Sitosterol (1) and 24-methylcholesterol (2) were found to inhibit the activity of the pump with an efficiency of 60%. Stigmasterol (3) was observed to be significantly less inhibitory. In the presence of both unusual sterols, 90% of the activity was inhibited. After sterol incorporation, fluorescence quenching rates were found to remain strictly dependent on Mg*+-ATP and inhibited *by orthovanadate (not shown). Evidence for an electrogenic transport is given by reversion of quenching upon addition of the K + / H t exchanger nigericin.

Similar experiments were carried out on proteoliposomes reconstituted by octylglucoside dilution. As depicted in Figure 4, the incorporation of 15 or 30 mol% sitosterol or 24-methylcholesterol into vesicles results in a strong decrease (34 and 60%, respectively) in the initial rate of ACMA fluorescence quenching. The same percentages of inhibition were observed when the proton transport was monitored in the presence of valinomycin, indicating that the inhibition was not due to the development of a membrane potential (not shown). A significantly lower inhibition of the H+ uptake was observed in stigmasterol- containing proteoliposomes, with no inhibition in the presence of 15 mol% (Fig. 5). When present at 15 or 30 mol%, both unusual sterols, 24-methylpollinastanol(5) and 14a,24-dimethylcholest-8-en-3P-o1 (6), triggered an inhibitory effect on the pump similar to that of typical A5-sterols (Fig. 4). In contrast, cholesterol (4), the major sterol in mammalian membranes but a generally minor sterol in plant membranes, was found to increase the initial rate of ACMA fluorescence quenching. Because of the higher stimulatory effect of cholesterol when present at 15 mol% compared with 30 mol%, it was interesting to monitor initial H’ pumping rates in vesicles less rich in cholesterol. Figure 5 shows that cholesterol induces a significant increase in proton transport when added in amounts as low as 2.5 mol% to reach a maximum at about 5 mol%. Similar experiments were performed with vesicles containing either stigmasterol or sitosterol. For sterol concentrations below 10 mol%, proteoliposomes with stigmasterol exhibit a behavior very similar to those enriched in cholesterol, but for higher concentrations, the activity of the pump is progres- sively inhibited, whereas cholesterol remains an activator. In sitosterol-containing vesicles, the initial rate of H+ up-

Table IV. Effects of various sterols (30 mo/%) on ATP-hydrolyzing activity of the ATPase reconstituted by freeze-thaw sonication and sensitivity toward orthovanadate and CaCI,

Results are exDressed as Dercentages of activity ( ~ S D ) of sterol-free proteoliposomes. Residual Activity

No addition + 300 p~ CaCI, + 100 p~ Na,VO,

%

Sterol

No sterol 1 O0 35 2 4 (6) 43 5 8 (4) Stigmasterol (3) 107 2 3 (3)” 34 t 1 (2) 45 2 1 (2) Sitosterol (1) 100 i 3 (3) 52 (1) 45 (1) 24-Methylcholesterol (2) 86 +- 8 ( 3 ) 40 2 7 ( 3 ) 50 t 7 ( 3 ) 24-Methylpollinastanol (5) 42 2 8 ( 3 ) 3 3 i 1 (2) 46 (1) 1 4cu,24-Dimethylcholest-8-en-3P-ol (6) 70 2 10 (3) 50 2 2 (2) 58 +- 2 (2)

a (n), Number of experiments.



Sterol Effects on the Plasma Membrane H'-ATPase 169

O 1 2 3 5 6

Figure 3. Effects of various sterols (30 mol%) on the initial rate of Hf uptake by the ATPase reconstituted by freeze-thaw sonication. Each assay was performed in triplicate. Data are means of at least three distinct experiments (SD never exceeded 20%) and expressed as percentage of activity of sterol-free proteoliposomes (O). Sterols used: 1, sitosterol; 2, 24-methylcholesterol; 3 , stigmasterol; 5, 24-methylpollinastanol; and 6 , 14a,24-dimethylcholest-8-en-3~-ol.

take is inhibited regardless of the sterol concentration in the vesicles.

Finally, the effects of a mixture of A5-sterols (48% sitosterol, 47.5% stigmasterol, and 4.5% 24-methylcholesterol) on the initial rate of H+ pumping by the PM H'-ATPase was investigated. This mixture was incorporated into proteoliposomes at 15 or 30 mol%. When present at 15 mol% (i.e. about 7 mol% each of sitosterol and stigmasterol), the mixture of sterols was found to stimulate the proton transport by 24%, indicating that the extent of the stimulating effect triggered by stigmasterol (i.e. 40% as shown in Fig. 5 ) was reduced by the inhibitory effect of sitosterol (10Y0). In proteoliposomes containing 30 mol% sterols, the initial rate of H+ uptake was found to be 68% of that of sterol-free vesicles. Under these conditions, only the inhibitory effect of 14 mol% sitosterol was detected. These results demonstrate that effects triggered by both sterols were additive, suggesting that they interact with different parts of the enzyme.

An accurate interpretation of fluorescence quenching rates is not simple. Since the intravesicular volume of proteoliposomes was not precisely measured, the amplitude of pH gradients cannot be calculated. It is assumed that the initial rate of fluorescence quenching largely re- flects H+ pumping, whereas the maximum quenching results from both the pump activity (H' influx) and the membrane permeability to protons (H+ efflux) (Bennett and Spanswick, 1983). The time course and final extent of the maximum quenching is the result of severa1 variables, such as the activity of the H'-ATPase, the size of vesicles, and vesicle passive permeability to protons. Each of these parameters may be affected by changes in the lipid composition of vesicles. To investigate the possible effect of sterol incorporation on the size of vesicles, proteoliposomes containing O or 30 mol% (i.e. the highest sterol concentration used in this work) cholesterol, sitosterol, or stigmasterol were prepared in the presence of the fluores- cent probe calcein. The intravesicular volume of vesicles

was estimated by the measurement of the trapped fluorescence within the vesicles after quenching of the external fluorescence by cobalt ions (Oku et al., 1982). This method is simple and rapid because it does not require the sepa- ration of vesicles from the external medium. In a11 situations the percentage of calcein trapped within vesicles was found to be very similar (between 5 and 6.5%), indicating that the size of vesicles obtained by the octylglucoside dilution procedure was not dependent on the sterol content or molecular species. Thus, the inhibition of initial rates of ACMA fluorescence quenching triggered by sitosterol or other sterols cannot be attributed to an enlargement of the intravesicular volume of proteoliposomes due to the sterol incorporation.

The other important parameter that one has to take into consideration is the potential effect of various sterols on the passive H+ permeability of reconstituted proteoliposomes. This can be done by monitoring relaxation of the pH gradient after addition of hexokinase and Glc (to remove residual ATP) and valinomycin at a steady state, and by measuring Te, (Fig. 6) (Szponarski et al., 1991). Results are presented in Table V. For cholesterol- and sitosterol- containing proteoliposomes, Tes values are significantly higher than those obtained for sterol-free vesicles, indicating that the presence of these sterols decreases the membrane permeability to H+. Such an effect is also indicated by the net increase in the amplitude of maximum quench

+ O 1 2 3 4 5 6 3: Y

h

B g 150 x c ' 5 120

8

4 P l2 s . 0

.3 c

e, 90

60

.g 30

+ O 1 2 3 4 5 6 3:

Figure 4. Effects of various sterols o n the initial rate of H+ uptake hy the ATPase reconstituted by octylglucoside dilution. Each assay was performed in triplicate. Data are means of at least three distinct experiments and expressed as percentage of activity of sterol-free proteoliposomes (O). Sterols were incorporated at either 15 (A) or 30 mol% (B). Numhers of sterols are identical to those in Figure 3. Sterol 4, Cholesterol.



170 Grandmougin-Ferjani et al. Plant Physiol. Vol. 113, 1997

20& O O 10 20 30

mo1 % Figure 5. Changes in the initial rate of H + uptake hy the ATPase reconstituted hy octylglucoside dilution as a function of cholesterol (O), stigmasterol (O), or sitosterol (W) content of proteoliposomes.

induced by these two sterols. By contrast, when stigmasterol or 14a,24-dimethylcholest-8-en-3~-ol is incorporated in vesicles, Te, values are similar to those of vesicles without sterol. Therefore, the inhibitory effect of sitosterol and other sterols on the initial rates of ACMA fluorescence quenching cannot be attributed to an increase in the passive membrane permeability to H+. None of the tested sterols was found to disrupt the membrane structure. When the relaxation of the pH gradient was monitored in the absence of valinomycin, similar Teq values were found for sitosterol- and stigmasterol-containing vesicles, indicating that stigmasterol is able to reduce passive membrane permeability to Kt but not to H+. Taken together, these results strongly suggest that sterols directly or indirectly interfere with the PM H+-ATPase itself.

DI SCUSSION

In plants structural and functional roles of sterols are not well understood. The great structural diversity of lipids and proteins in the PM makes any investigation at the molecular leve1 extremely difficult to achieve. In the present study sterol modulation of the PM H+-ATPase activity from corn roots was assessed after reconstitution of the partially purified enzyme into proteoliposomes by means of either freeze-thaw sonication or octylglucoside dilution. The first procedure, which has been largely used for transport studies, was previously applied to the reconstitution of the PM H'-ATPase from corn roots (St. Marty- Fleurence et al., 1988; Gibrat et al., 1990) and from other plant materials such as tomato roots (Anthon and Spans- wick, 1986), oat roots (Vara and Serrano, 1982; Serrano et al., 1988), or the alga Dunaliella acidophila (Sekler and Pick, 1993). The second method first implies a co-solubilization of the enzyme and lipids in a detergent (octylglucoside or

sodium cholate or deoxycholate or Zwittergent 3-14). Ves- icle formation and reconstitution then occur simultaneously by remova1 of the detergent by dilution, dialysis, or gel filtration. Focusing on plant materials, this method has been used for the reconstitution of the PM H+-ATPase from corn roots (St. Marty-Fleurence et al., 1988; Brauer and Tu, 1989; Rossignol et al., 1989; Hsu et al., 1989), mung bean roots (Kasamo, 1987) and hypocotyls (Kasamo and Yamanishi, 1991), radish seedlings (Cocucci et al., 1985), red beet (ONeill and Spanswick, 1984a, 198413; Singh et al., 1987), and oat roots (Serrano et al., 1988). Other reconstitution processes have also been reported, such as the spontaneous insertion into a preformed bilayer (Simon-Plas et al., 1991) or a planar bilayer system (Briskin et al., 1995). Given the diversity of experimental conditions described in these reports (nature of lipids used, lipid-to-protein ratio, etc.), it is difficult to readily compare the results. Moreover, there are only a few works in which ATP hydrolysis and Ht pumping were conducted with the same preparation and simultaneously.

Here we present evidence that both methods we used yield proteoliposomes capable of ATP-driven H+ transport and ATP hydrolysis, with properties similar to those of other preparations (see refs. cited above). The sensitivity of both functionalities to inhibitors such as vanadate, a specific inhibitor of P-type ATPases, DCCD, and Ca2+ ions, a strong preference for ATP as the substrate, and the maintenance of an optimum pH at 6.5 indicated that the reconstituted enzyme retained its native characteristics. These results demonstrate that crude soybean phospholipids pro- vided an adequate lipidic environment for the enzyme from corn, as already reported (Brauer and Tu, 1989), with no requirement for a specific phospholipid (Serrano et al., 1988; Simon-Plas et al., 1991). Compared with the freeze- thaw procedure, the reconstitution by the octylglucoside dilution procedure, which was conducted directly in the

AT P N iger ic i n

-----. + I..;.--([ H K + ~ 1

T o T, , I , I

I I I I

O 3 6 9 12 (min) Time

Figure 6 . Time course of pH gradient relaxation in the presence of valinomycin after steady state in the case of the ATPase reconstituted hy octylglucoside in sterol-free proteoliposomes. HK, Hex- okinase; Val, valinomycin; T,, time of HK addition; T,, time needed to reach a new steady state. Teq corresponds to the difference between T, and To.



Sterol Effects on the Plasma Membrane H'-ATPase 171

Table V. Effect o f various sterols on init ial and maximum ACMA fluorescence quenching rates and on the relaxation o f the pH gradient in the absence and presence o f valinomycin b y the ATPase reconstituted by octylglucoside dilution

Numbers in brackets correspond to the time required to obtain maximum quench. The p H gradient relaxation was performed after addition of hexokinase and Glc at steady state (maximum quench). Teq = 90 s ? 10 for sterol-free vesicles in the presence or absence of valinomycin. Values are expressed as percentages of control (sterol-free proteoliposomes) and correspond to an average of at least four determinations (mean SD, 10%).

lnitial Q" Maximum Q Teq Sterol Composition

% min-' ma-' Drotein % ma-' Drotein lminl + valb - val

Control 1 O0 100 [51 1 00 1 O0 + Sitosterol (1) (30 moi%) 38 150 [27] 155 165 + Stigmasterol (3) (30 moi%) 70 95 [51 105 160 + 14a,24-DimethylchoIest-8-en-3P-ol (6) (1 5 mol%) 45 120 [20] 105 nd' + Cholesterol (4) (15 mol%) 145 180 [lO] 190 nd

a Q, Quench. val, Valinomycin. nd, Not determined.

cuvette of the spectrofluorometer, was found to be rapid, efficient, sensitive, and highly reproducible. However, because of the high concentration of detergent needed to solubilize sterols, resulting in a high factor of dilution, the use of a more sensitive method to measure ATP hydrolysis is required. Preliminary assays showed that when ATP hydrolysis was assayed in the presence of NADH as well as an ATP-generating system and quantified by monitoring the decrease in NADH fluorescence, the accuracy of measurements was improved.

We investigated the effects of different concentrations (between O and 30 mol%) of various individual sterols on ATP hydrolysis and ATP-driven proton translocation by the partially purified PM H+-ATPase from corn roots. Con- cerning sterol effects on the initial rates of H+ pumping, the two completely different methods of reconstitution gave very similar results (Figs. 4 and 5). Sitosterol (1) and 24-methylcholesterol (2) were found to inhibit H+ pumping, with the percentage of inhibition increasing with the sterol concentration in proteoliposomes. Cholesterol(4) appeared to stimulate the pump over the complete range of concentrations between O and 30 mol%. Stigmasterol (3) exhibited a dual behavior, with a stimulating effect at a low concentration (below 5 mol%) and a moderate inhibitory effect at higher concentrations. Finally, the unusual sterols 24-methylpollinastanol (5) and 14a,24-dimethylcholest-8- en-3/3-01 ( 6 ) displayed an inhibitory effect regardless of their concentration in proteoliposomes. Thus, the sterol modulation of the PM H+-ATPase activity was shown to be dependent on both the sterol concentration and the sterol molecular species. The biphasic effect triggered by stigmasterol on the plant enzyme appears to be very similar to that found for cholesterol for the Na+,K+-ATPase from bovine kidney (Yeagle et al., 1988). It is interesting that the maxi- mal activity of the mammalian enzyme was observed at the native membrane cholesterol content. In PM vesicles from corn roots, cholesterol and stigmasterol are present at 0.6 and 16.5 mol%, respectively, but the effective concentrations of both sterols in the immediate environment of the enzyme might be different. Using an alternative approach consisting of incubations of PM vesicles from oat roots with exogenous sterols, Cooke et al. (1994) reported observations similar to ours, i.e. a stimulation of proton transport following incubation with cholesterol or stigmasterol, and

an inhibition after incubation with a mixture of sitosterol and 24-methylcholesterol. Thus, the activity of the PM H+-ATPase appears to be very sensitive to its sterol environment. For instance, the enzyme is able to discriminate between sitosterol and stigmasterol, two 24-ethylsterols differing only by an additional double bond at C-22 in the side chain of stigmasterol.

Such an effect of structural features of the sterol molecule on the activity of other P-type ATPases has been reported. The activity of the cardiac sarcolemmal Na+,K+-ATPase reconstituted in a PC / phosphatidylserine mixture was shown to be 5- to 6-fold stimulated by cholesterol, 7-dehydrocholesterol, and dihydrocholesterol, but only 2-fold stimulated by 24-methylcholesterol (a 24-methyls- terol); a strong inhibition was observed in the presence of androstenol (absence of a side chain at the C-17 position) or lanosterol (a biosynthetic precursor of cholesterol with two methyl groups in C-4) (Vemuri and Philipson, 1989). It has been discovered that the Ca2+-ATPase of the sarcoplasmic reticulum reconstituted in dimyristoleoylPC was able to distinguish between cholesterol and its 3a-isomer, epicho- lesterol, and among four isomers of cholestanol (Michelan- geli et al., 1989). In the same context, a large body of work has focused not only on a cholesterol modulation but also on a specific sterol requirement of transmembrane ion transport catalyzed by transporters (Michelangeli et al., 1989; Vemuri and Philipson, 1989; Shouffani and Kanner, 1990), channels (Bialecki and Tulenko, 1989; Chang et al., 1995; Popp et al., 1995), or receptors (Fernandez-Ballester et al., 1994).

As already noted above, the interpretation of fluorescence quenching data is complex. Changes in the lipid composition of vesicles could affect both the initial rate and the final extent of quenching via a modification of the size of reconstituted vesicles and/ or alterations in the membrane H+ permeability. As demonstrated by measurements of the percentage of fluorescence of calcein trapped within sterol-free and sterol-containing proteoliposomes, the in- terna1 volume of reconstituted vesicles appeared to be largely independent of the sterol content (between O and 30 mol%) or the sterol structure. Such a result is in agreement with our previous observations concerning large unilamel- lar vesicles made with pure soybean PC and different amounts of various sterols (Schuler et al., 1991) and could



172 Crandmougin-Ferjani et al. Plant Physiol. Vol. 1 1 3 , 1997

result from the high content in polyunsaturated fatty acyl chains of soybean lipids. Therefore, the effects triggered by the various sterols on initial rates of ACMA fluorescence quenching cannot be attributed to changes in the interna1 volume of reconstituted vesicles.

At a steady-state level, the activity of the pump (H+ influx) is counterbalanced by the Hf membrane leakage. As shown in Table V, incorporation of sitosterol or cholesterol into proteoliposomes was found to increase the mag- nitude of maximum quench compared with that observed for sterol-free vesicles in a way closely dependent on the membrane sterol content (data not shown). Such an increase in maximum quench is accompanied by a slowing down of the whole time course of ACMA fluorescence quenching. In particular, in the presence of 30 mol% sitosterol the steady-state level was reached only after 27 min compared with 5 min for sterol-free vesicles. In contrast, other sterols such as stigmasterol and 14a,24-dimethylcholest-8-en-3P-o1 triggered no change in the extent of maximum quench. Experiments were conducted to eval- uate potential effects of the various sterols on the passive membrane permeability to protons. They consisted of add- ing Glc and hexokinase at steady state to block the activity of the pump, and monitoring the time course of relaxation of the pH gradient in the presence or absence of valinomycin. In the presence of the ionophore, recovery of initial ACMA fluorescence occurred less rapidly in proteoliposomes containing sitosterol or cholesterol than in proteoliposomes with no sterol, indicating that these sterols significantly reduced the passive membrane permeability to Ht . For stigmasterol- or 14a,24-dimethylcholest-8-en-3p-o1- containing proteoliposomes, the kinetics of the pH gradient relaxation were found to be similar, indicating that neither compound affected passive membrane permeability to protons. These results closely parallel our previous data concerning the efficiency of the different sterols to reduce water permeability of soybean PC bilayers (Schuler et al., 1991). If we assume that Hf diffusion across a membrane proceeds via associated water molecules (Elamrani and Blume, 1983), then H+/OH- and water permeation are indeed interrelated.

Our results clearly show that the sterol addition to proteoliposomes strongly affects the ATP-driven H+ translocation rate, whereas phosphohydrolase activity was found to be much less sensitive to both sterol content and structure. Such an observation might be viewed as an artifact of the reconstitution procedure. Thus, among the whole pop- ulation of ATPase molecules present in the preparations, only a few might be functionally active. It should be noted that only ATPase molecules readily inserted in the membrane bilayer with the catalytic site oriented toward the outside are able to drive an ATP-dependent H+ translocation, whereas a11 of the molecules of the enzyme, including those just bound on the surface of vesicles, can hydrolyze ATP. In that context, the simultaneous monitoring of both functionalities on the same liposomal preparations as reported by Palmgren (1990) or Briskin et al. (1995) would be informative. However, evidence for differential effects in vitro on ATP hydrolysis and H+ pumping triggered by

various effectors in native PM vesicles has been reported in the literature (Hsu et al., 1992; Cooke et al., 1993, 1994; Faraday and Spanswick, 1995). A change in the coupling ratio of yeast PM H+-ATPase by a metabolite has been mentioned (Venema and Palmgren, 1995), and evidence has been presented for various P-type ATPase mutants in which ATP hydrolysis is uncoupled with ion transport (Portillo and Serrano, 1988; Andersen, 1995). Thus, H+ pumping might be regulated independently of ATP hydrolysis as proposed by Venema and Palmgren (1995).

In conclusion, sterols are likely to modulate the function of the PM H+-ATPase by more than one mechanism. Cur- rently, there are two theories to account for lipid modulation of ATPase activity (see Cooke and Burden, 1990). According to the first theory, changes in lipid composition might induce an alteration of bulk physical properties of the membrane lipid bilayer, thereby indirectly affecting the ability of the enzyme to undergo conformational changes of the enzyme, and thus resulting in changes in its activity. The second hypothesis is based on a direct interaction of lipids with the enzyme as already stated by Simmonds et al., (1982) and Ding et al. (1994) for the sarcoplasmic reticulum Ca2+-ATPase and by Cornelius (1995) for the Na+,K+-ATPase. Using steady-state fluorescence anisotropy measurements (Schuler et al., 1990) and 'H-NMR spectroscopy (Schuler et al., 1991; Krajewski-Bertrand et al., 1992), we have shown that sitosterol and 24- methylcholesterol are able to order fatty acyl chains of soybean PC bilayers with an efficiency similar to that displayed by cholesterol. In contrast, stigmasterol was found to exhibit far less ordering ability. Concerning the unusual sterols, 24-methylpollinastanol behaves like sitosterol, but its isomer 14a,24-dimethylcholest-8-en-3~-01 is less efficient than stigmasterol (Krajewski-Bertrand et al., 1992). The stimulatory or inhibitory effects triggered by the various sterols cannot be unambiguously related to any particular effects of these sterols on membrane fluidity, indicating that sterol modulation of the PM H+-ATPase activity is not a process directly depending on changes in phospholipid fatty acyl chain ordering, as stated in the case of the sarcoplasmic reticulum Ca'+-ATPase (Michelangeli et al., 1989).

The second theory favors a regulation through a direct interaction of the sterol with the protein, which should be highly structurally specific, especially at a low sterol concentration; cholesterol and stigmasterol might play such a role. When given at a concentration of 3 PM to intact corn roots, these two sterols are able to stimulate H+ secretion (Cerana et al., 1984). Because of-its crucial role in the steady-state phosphorylation level of the animal Na+,K+- ATPase, cholesterol is considered to be a direct effector of the enzyme (Yoda and Yoda, 1987; Cornelius, 1995). For the reconstituted shark enzyme, the V,,, for Na+ activation was shown to be 6-fold stimulated at 40 mol% cholesterol and the apparent K , slightly increased (Cornelius, 1995). In our case, similar changes in kinetic properties of the ATP-driven Ht translocation were found to be triggered at 15 mol% cholesterol (B. Kanzler and M. A. Hartmann, unpublished data). For situations in which only inhibi-




tion is observed (i.e. i n the presence of sitosterol, 24- methylcholesterol, 24-methylpollinastanol, and 14a,24- dimethylcholest-8-en-3P-ol), one could postulate that these sterols act as negative effectors by binding to the enzyme. For some of these sterols, the possibility of a mismatch between the hydrophobic span of the protein and the sur- rounding lipid bilayer, which potentially can have functional implications, also must be taken into account (Mouritsen and Bloom, 1993). A n examination of the molecular basis of sterol modulation will be the focus of our future investigations.

ACKNOWLEDCMENTS

We thank Drs. M. Rossignol, V. Szponarski, and G. Vansuyt (Institut National de la Recherche Agronomique [INRA], Mont- pellier, France) for assistance with isolation of the partially purified PM H+-ATPase from corn roots and their advice concerning the reconstitution using the octylglucoside dilution procedure. We also thank Dr. J.P. Blein (INRA, Dijon, France) for the sample of SW 26. Finally, we would like to thank Dr. Marc Lemaire (Centre d’Etudes de Saclay, Gif-sur-Yvette, France) for a critica1 reading of the paper and John Ackerson for grammar improvements.

Received July 5, 1996; accepted September 23, 1996. Copyright Clearance Center: 0032-0889/97/ 113/0163/12.

LITERATURE ClTED

Andersen JP (1995) Functional consequences of alterations to amino acids at the M5S5 boundary of the CaZ+-ATPase of sarcoplasmic reticulum. J Biol Chem 270: 908-914

Anthon GE, Spanswick RM (1986) Purification and properties of the H+-translocating ATPase from the plasma membrane of tomato roots. Plant Physiol 81: 1080-1085

Bennett AB, Spanswick RM (1983) Optical measurements of ApH and A+ in corn root membrane vesicles: kinetic analysis of C1- effects on a proton-translocating ATPase. J Membr Biol71: 95-107

Benveniste P (1986) Sterol biosynthesis. Annu Rev Plant Physiol

Bialecki RA, Tulenko TN (1989) Excess membrane cholesterol alters calcium channels in arterial smooth muscle. Am J Physiol

Bradford MM (1976) A rapid and sensitive method for the quan- titation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254

Brauer D, Tu SI (1989) Phospholipid requirement of the vanadate- sensitive ATPase from maize roots evaluated by two methods. Plant Physiol 89: 867-874

Briskin DP, Basu S, Assmann SM (1995) Characterization of the red beet plasma membrane H+-ATPase reconstituted in a planar bilayer system. Plant Physiol 108: 393-398

Cerana R, Lado P, Anastasia M, Ciuffreda P, Allevi P (1984) Regulating effects of brassino steroids and of sterols on growth and H+ secretion in maize roots. Z Pflanzenphysiolll4: 221-225

Chang HM, Reitstetter R, Mason RP, Gruener R (1995) Attenua- tion of channel kinetics and conductance by cholesterol: an interpretation using structural stress as a unifying concept. J Membr Biol 143: 51-63

Chen PS, Toribara TY, Warner H (1956) Microdetermination of phosphorus. Anal Chem 28: 1756-1758

Cocucci MC, De Michelis MI, Pugliarello MC, Rasi-Caldogno F (1985) Reconstitution of proton pumping activity of a plasma membrane ATPase purified from radish. Plant Sci Lett 37: 189-193

Cooke DT, Burden RS (1990) Lipid modulation of plasma membrane-bound ATPases. Physiol Plant 78: 153-159

36: 275-308

257: C306-C314

Cooke DT, Burden RS, James CS, Seco T, Sierra B (1994) Influ- ente of sterols on plasma membrane proton-pumping ATPase activity and membrane fluidity in oat shoots. Plant Physiol Biochem 32: 769-773

Cooke DT, Ros R, Burden RS, James CS (1993) A comparison of the influence of sterols on the specific activity of the H+- ATPases in isolated plasma membrane vesicles from oat, rye and rice shoots. Physiol Plant 88: 397402

Cornelius F (1995) Cholesterol modulation of molecular activity of reconstituted shark Na+, K+-ATPase. Biochim Biophys Acta 1235: 205-212

Ding J, Starling AP, East JM, Lee AG (1994) Binding sites for cholesterol on CaZ+-ATPase studied by using a cholesterol- containing phospholipid. Biochemistry 33: 49744979

Elamrani K, Blume A (1983) Effect of the lipid phase transition on the kinetics of H+/OH- diffusion across phosphatidic acid bilayers. Biochim Biophys Acta 727: 22-30

Faraday CD, Spanswick RM (1995) Differential stimulation by C1- of H+-transport and ATPase activity in purified plasma membrane vesicles from maize (Zea mays L.) roots. J Exp Bot 46:

Fernandez-Ballester G, Castresana J, Fernandez AM, Arrondo JLR, Ferragut JA, Gonzalez-Ros JM (1994) A role for cholesterol as a structural effector of the nicotinic acetylcholine receptor. Biochemistry 33: 40654071

Gibrat R, Grouzis JP, Rigaud J, Grignon C (1990) Potassium stimulation of corn root plasmalemma ATPase. Plant Physiol93:

Grandmougin A, Bouvier-Nave P, Ullmann P, Benveniste P, Hartmann MA (1989) Cyclopropyl sterol and phospholipid composition of membrane fractions from maize roots treated with fenpropimorph. Plant Physiol 90: 591-597

Hartmann MA, Benveniste P (1987) Plant membrane sterols: isolation, identification and biosynthesis. Methods Enzymol 148:

Hsu AF, Brauer D, Tu SI (1989) Characterization of reconstituted plasma membrane H+-ATPase from maize roots. Physiol Plant

Hsu AF, Rodenbach S, Tu SI (1992) Effects of dicyclohexylcarbodiimide (DCCD) treatment on coupled activities of vanadate- sensitive ATPase from plasma membrane of maize roots. Physiol Plant 85: 17-22

Kasamo K (1987) Reconstitution and characterization of H+- translocating ATPase from the plasma membrane of Phaseolus mungo L. roots. Plant Cell Physiol 28: 19-28

Kasamo K, Sakakibara Y (1995) The plasma membrane H+-ATPase from higher plants: functional reconstitution into liposomes and its regulation by phospholipids. Plant Sci 111: 117-131

Kasamo K, Yamanishi H (1991) Functional reconstitution of plasma membrane H+-ATPase from mung bean (Vigna radiata L.) hypocotyls in liposomes prepared with various molecular species of phospholipids. Plant Cell Physiol 32: 1219-1225

Krajewski-Bertrand AM, Milon A, Hartmann MA (1992) Deuterium-NMR investigation of plant sterol effects on soybean phosphatidylcholine acyl chain ordering. Chem Phys Lipids 63:

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685

Marré E, Ballarin-Denti A (1985) The proton pumps of the plasmalemma and the tonoplast of higher plants. J Bioenerg Bi- omembr 17: 1-21

Michelangeli F, Henderson I, East JM, Lee AG (1989) Interaction of sterols with phospholipid bilayers and the (Ca2+-Mg2+)- ATPase. Biochem SOC Trans 17: 1095-1096

Michelet B, Boutry M (1995) The plasma membrane H+-ATPase. Plant Physiol 108: 1-6

Mouritsen OG, Bloom M (1993) Models of lipid-protein interac- tions in membranes. Annu Rev Biophys Biomol Struct 22:

Oku N, Kendall DA, McDonald RC (1982) A simple procedure for the determination of the trapped volume of liposomes. Biochim Biophys Acta 691: 332-340

1169-1176

1183-1189

632-650

76: 544-550

235-241

145-171



174 Grandmougin-Ferjani et al. Plant Physiol. Vol. 11 3, 1997

O'Neill SD, Spanswick RM (1984a) Solubilization and reconstitution of a vanadate-sensitive H+-ATPase from the plasma membrane of Beta vulgaris. J Membr Biol 79: 231-243

ONeill SD, Spanswick RM (1984b) Characterization of native and reconstituted plasma membrane H+-ATPase from the plasma membrane of Beta vulgaris. J Membr Biol 79: 245-256

Palmgren MG (1990) An Hf-ATPase assay: proton pumping and ATPase activity determined simultaneously in the same sample. Plant Physiol 94: 882-886

Palmgren MG (1991) Regulation of plant plasma membrane Hf- ATPase activity. Physiol Plant 83: 314-323

Palmgren MG, Sommarin M (1989) Lysophosphatidylcholine stimulates ATP dependent proton accumulation in isolated oat root plasma membrane vesicles. Plant Physiol90: 1009-1014

Palmgren MG, Sommarin M, Ulvskov P, Larsson C (1990) Effect of detergents on the H+-ATPase activity of inside-out and right- side-out plant plasma membrane vesicles. Biochim Biophys Acta

Paternostre MT, Roux M, Rigaud JL (1988) Mechanisms of membrane protein insertion into liposomes during reconstitution procedures involving the use of detergents. Biochemistry 27:

Popp B, Schmid A, Benz R (1995) Role of sterols in the functional reconstitution of water-soluble mitochondrial porins from different organisms. Biochemistry 34: 3352-3361

Portillo F, Serrano R (1988) Dissection of functional domains of the yeast proton-pumping ATPase by directed mutagenesis.

Rossignol M, Kervella V, Blein JP (1989) Octylglucoside reconstitution of plasma membrane Ht-translocating ATPase from corn roots. Plant Physiol Biochem 27: 437441

Schuler I, Duportail G, Glasser N, Benveniste P, Hartmann MA (1990) Soybean phosphatidylcholine vesicles containing plant sterols: a fluorescence anisotropy study. Biochim Biophys Acta

Schuler I, Milon A, Nakatani Y, Ourisson G, Albrecht AM, Benveniste P, Hartmann MA (1991) Differential effects of plant sterols on water permeability and on acyl chain ordering of soybean phosphatidylcholine bilayers. Proc Natl Acad Sci USA

Sekler I, Pick U (1993) Purification and properties of a plasma membrane H+-ATPase from the extremely acidophilic alga Du- naliella acidophila. Plant Physiol 101: 1055-1061

Serrano R (1989) Structure and function of plasma membrane ATPase. Annu Rev Plant Physiol Plant Mo1 Biol 40: 61-94

Serrano R, Montesinos C, Sanchez J (1988) Lipid requirements of

1021: 133-140

2668-2677

EMBO J 7: 1793-1798

1028: 82-88

88: 6926-6930

the plasma membrane ATPases from oat roots and yeast. Plant Sci 5 6 117-122

Shouffani A, Kanner BI (1990) Cholesterol is required for the reconstitution of the sodium- and chloride-coupled, y-aminobutyric acid transporter from rat brain. J Biol Chem 265:

Simmonds AC, East JM, Jones OT, Rooney EK, McWhirter J, Lee AG (1982) Annular and non-annular binding sites on the (Ca2+, Mg'+)-ATPase. Biochim Biophys Acta 693: 398-406

Simon-Plas F, Venema K, Grouzis JP, Gibrat R, Rigaud J, Grignon C (1991) Spontaneous insertion of plant plasma membrane (Ht) ATPase into a preformed bilayer. J Membr Biol120:

Singh SP, Sudershan Kesav BV, Briskin DP (1987) Reconstitution and rapid partia1 purification of the red beet plasma membrane ATPase. Physiol Plant 69: 617-626

Solioz M (1984) Dicyclohexylcarbodiimide as a probe for proton translocating enzymes. Trends Biochem Sci 9: 309-312

St. Marty-Fleurence F, Bourdil I, Rossignol M, Blein JP (1988) Active vanadate-sensitive H+-translocation in corn roots membrane vesicles and proteoliposomes. Plant Sci 54: 177-184

Szponarski W, Vansuyt G, Rossignol M (1991) Auxin effects on proton accumulation by reconstituted plasma membrane H+- ATPase. Phytochemistry 30: 1391-1395

Tanfani F, Bertoli E (1989) Permeability of oxidized phosphatidylcholine liposomes. Biochem Biophys Res Commun 163: 241-246

Ullmann P, Bouvier-Navé P, Benveniste P (1987) Regulation by phospholipids and kinetic studies of plant membrane-bound UDP-g1ucose:sterol-P-D-glucose transferase. Plant Physiol 85:

Vara F, Serrano R (1982) Partia1 purification and properties of the proton-translocating ATPase of plant plasma membranes. J Biol Chem 2 5 7 12826-12830

Vemuri R, Philipson KD (1989) Influence of sterols and phospholipids on sarcolemmal and sarcoplasmic reticular cation transporters. J Biol Chem 264 8680-8685

Venema K, Palmgren MG (1995) Metabolic modulation of transport coupling ratio in yeast plasma membrane H+-ATPase. J Biol Chem 270: 19659-19667

Yeagle PL, Young J, Rice D (1988) Effects of cholesterol on (Na',K+)-ATPase ATP hydrolyzing activity in bovine kidney. Biochemistry 2 7 6449-6452

Yoda A, Yoda S (1987) Two different phosphorylation- dephosphorylation cycles of Na,K-ATPase proteoliposomes ac- companying Nat transport in the absence of Kf. J Biol Chem

6002-6008

51-58

51-55

262: 110-115



Documents

Sterol Modulation of Plasma Membrane H+-ATPase Activity ... · soybean (Glycine max 1.) phospholipids and various concentrations of individual sterols using either a freeze-thaw sonication