ATPases of Oat Roots - Journal of Biological Chemistry

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Society of Biological Chemists, Inc.

Vol. 260, No. 19, Issue of September 5, pp. 10434-10443,1985 Printed in U.S.A.

Similarities and Differences between the Tonoplast-type and the Mitochondrial H’-ATPases of Oat Roots* (Received for publication, February 19, 1985)

Yanzhi WangS and Heven Sze From the Department of Botany, University of Maryland, College Park, Maryland 20742

The native tonoplastandthemitochondrial H+- plast of intact cells (1).These results takentogether with the ATPasesfromoatroots were comparedtodeterpresence of ATPase activities and ATP-dependentH+ pumpmine whether the two enzymes have similar mecha- ing in isolated vacuoles have established that tonoplast memnisms. H+ pumping inlow-density microsomal vesicles branes possess a H+-pumping ATPase (Ref. 3 and references reflected activity from the tonoplast-type ATPase, as therein). ATPase activity andATP-dependent H+ pumping Using isolated vesicles from various plant tissues, several (quinacrinefluorescencequenching)showedsimilar laboratories have shown evidence for electrogenic H+pumping sensitivities to inhibition by N-ethylmaleimide, N,N’- ATPases (Ref. 3 and references therein). The H+-ATPase dicyclohexylcarbodiimide, 4,4’-diisothiocyano-2,2’- associated with low-density microsomal vesicles is referred to stilbene disulfonate,nitrate, quercetin, or 7-chloro-4- as a tonoplast-type H+-ATPase for the following reasons. The nitrobenzo-2-oxa-l,3-diazole. The tonoplast-type H+-pumping ATPase associated with low-density vesicles and ATPase was stimulated by Cl-,Br- > HCO; whereas that associated with tonoplast vesicles are both resistant to the mitochondrial ATPasewas stimulated byHCO; >> Cl-,Br-. Both enzymes hydrolyzed ATP preferentially orthovanadate, stimulated byC1- and inhibited by NO; (3). Furthermore, the tonoplast vesicles have a density of1.10and were inhibited competitively byAMPorADP. 1.12 g/cc similar to the low-density microsomal vesicles. The Apartfrom resistance to azide,thetonoplast-type ATPase was strikingly similar inits inhibitor sensitiv- results strongly suggest the Hf-ATPase activities from the ities to the mitochondrial ATPase. The insensitivity to low-density vesicles are associated with right-side-out tonovanadate of both enzymes suggests the reaction mech- plast vesicles (4). The pH gradient generated (acid inside) in anisms do not involve a covalent phosphoenzyme. In- low-density vesicles can be measured as ATP-dependent[“C] or as quinacrine fluorescence 7-chloro-4-nitrobenzo-2-oxa-1,3-diazolemethylamine distribution hibitionby and N-ethylmaleimide and protection by ATPsuggests quenching (3, 5, 6). The H+ pump is electrogenic as demontyrosine and cysteine residues are in thecatalytic site strated by ATP-dependent [14C]thiocyanatedistribution (5) of the tonoplast ATPase. The mitochondrial ATPase or oxonol absorbance changes (6). This H+-ATPase can be was 100 times more sensitive to N,N‘-dicyclohexyl- clearly distinguished from the H+-pumping ATPase of the carbodiimide inhibition than the tonoplast H*-ATPase. mitochondria because of its insensitivity to oligomycin or These results suggest the tonoplast and the mitochon-azide (3, 7, 8). It also differs from the plasma membrane drial H+-ATPases share common steps in their cataelectrogenic H+-ATPase which is stimulated by K+and inhiblytic and vectorial reaction mechanisms, yet sufficient ited by orthovanadate (3,8). differences exist toindicatetheyaretwodistinct A vanadate-insensitive ATPase could be interpreted as a ATPases. type of phosphohydrolase that does not form a covalent Meristematic cells of higher plants are characterized by numerous small vacuoles that coalesce to form one or more large vacuoles in themature differentiated cells. The vacuoles are important not only for maintaining the cell turgor but also for the transport and storage of numerous ions and metabolites (1).In senescing cells, the vacuoles play a role in intracellular digestion analogous to lysosomes in animal cells (2). Active transport of various solutes (including inorganic cations, anions,amino acids, and sugars) across the tonoplast (the vacuolar membrane) is thought to be driven by a proton motive force. A pH gradient (acid inside relative to the cytoplasm) and sometimes an electrical gradient (positive inside relative to the cytoplasm) can be measured across the tono* This work was supported in part by National Science Foundation Grants PCM 83-10928 and PCM 83-04130 to H. S. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Present address: Biology Department, Wuhan University, Wuhan, Hubei, China.

phosphoenzyme intermediate in the reaction mechanism similar to the FIFo-ATPase of the mitochondria/chloroplast/ bacteria (9). The tonoplast-type ATPase of higher plants may be similar to a vanadate-insensitive H+-pumping ATPase associated with storage granules (10, ll),lysosomes (12-15), endosomes (16), and clathrin-coated vesicles (17, 18) of various animal cells. To understand the transport function, molecular properties, and the mechanism of catalysis of the tonoplast-type ATPase, we are studying the tonoplast-type ATPase from oat roots in more detailand comparing its properties to the mitochondrial ATPase of the same tissue. As the first part of our study, we show here that (i) the H’ transport and enzymatic properties of the native tonoplasttype ATPase arewell correlated and (ii)the native tonoplasttype ATPaseand the native mitochondrial ATPase show striking similarities in their sensitivity to various inhibitors. The results suggest the reaction mechanism of these two enzymes is related, yet sufficient differences exist to indicate they are two distinct ATPases. MATERIALS AND METHODS

Plant Material-Oats (Avena sativa L.var. Lang) were germinated in the dark over an aerated solution of 0.5 m M CaSO,. After 4 or 5

10434

Tonoplast and Mitochondrial H+-ATPases days of growth, the apical section (3-4 cm) of the roots was harvested. Lang oats were generously provided by the Agronomy Department, University of Maryland, College Park (Dr. D. Sammons) and Southern States Cooperative, Baltimore. Preparation of Mitochondria and Law-density Vesicles-The procedure of Churchill and Sze (7) was used with some modifications. All procedures were conducted at 4 "C. Oat roots were homogenized twice with a mortar and pestle in medium containing 30 mM HepesBTP' buffer a t pH 7.4,250 mM D-mannitol, 3 mM EGTA, 1 mM DTT, and0.1% bovine serum albumin using a medium to tissue ratio of 2.5 ml/g, fresh weight. The brei was strained through cheesecloth, and thehomogenate was centrifuged for 10 min at 480 X g to remove unbroken cells and the cell wall fraction. The supernatant was centrifuged at 6000 X g for 15 min; the pellet was resuspended and referred to as the mitochondrial fraction. The supernatant was subsequently centrifuged a t 12,000 X g for 10 min, and the resulting supernatant was further centrifuged a t 60,000 X g for 30 min (SW 28 rotor). The pellet (or the crude microsomal fraction) was gently resuspended in buffer containing 2.5 mM Hepes-BTP, pH 7.4, 250 mM D-mannitOl, 0.5 mM D m , and incubated for 4 min at 35 "c.The latter step facilitates resuspension and may enhance vesicle sealing. The suspension was layered over a 6% dextran (average molecular weight, 79,000) cushion and centrifuged at 70,000 for 2 h (SW 28.1). The turbid interface ontop of the dextran cushion was collected.The H+-ATPase of the 0/6% dextran interface is vanadate insensitive and anion sensitive (5, 7 , 19) similar to an enzyme associated with the vacuolar membrane of various plant tissues (3). Therefore, we refer to the6% dextran interface ATPase as thetonoplast-type ATPase. ATPase Actiuity-ATPase activity was determined by measuring the release of inorganic phosphate from ATP according to a modified Fiske and SubbaRow assay (20). The mitochondrial ATPase activity was measured in a 0.5-ml reaction mixture containing 30 mM BTPHepes at pH 8.0, 3 mM MgSO,, 3 mM ATP-BTP, pH 8.0, 0.1 mM ammonium molybdate, with or without 0.5 mM Na azide and 14-23 pgof membrane protein. The mitochondrial MgATPase was 85% inhibited by 0.5 mM Na azide while the tonoplast-type MgATPase was notinhibited at all (Fig. l)?Specific mitochondrial ATPase activity was determined as the difference in activity in the presence and absence of azide (or azide-sensitive activity). As the ATPase activity in isolated mitochondria was only slightly stimulated (10%) by Triton X-100 or nigericin (not shown), all assays were determined in the absence of the detergent or ionophore. These results suggested that most of the mitochondria were uncoupled and that the substrates appeared to have access to the enzyme-active site. The reaction mixture (0.5 ml) for the tonoplast-type ATPase contained 30 mM Hepes-BTP at pH 7.0, 3 mM MgSO,, 3 mM ATPBTP, pH 7.0, 15-18 pg of membrane protein plus 0.5 mM Na azide, 0.1 mM Na orthovanadate, and 0.1 mM ammonium molybdate to inhibit the mitochondrial ATPase, the plasma membrane ATPase, and acid phosphatases, respectively (21). C1"stimulated activity was the difference between activity in the presence and absence of 50 mM BTP-Cl. ATPase activity was usually determined in the absence of gramicidin, since the properties of the enzyme were not altered by the ionophore (5). Protein concentration was estimated by the Lowry method (22) after precipitation with 10% trichloroacetic acid and using bovine serum albumin as thestandard. ATP-dependent pH Gradient-Relative ApH formation was measured by quinacrine fluorescence quenching using a Farrand System 3 spectrofluorometer a t 495 nm after excitation a t 420 nm (7, 23).

10435 TABLE I

Substrate specificity of the mitochondrial and. tonoplost-type MgATPases from oat roots Numbers in parenthesis representrelative activity in per cent. C1" stimulated activity of the tonoplast-type ATPase (in thepresence or absence of gramicidin) showed the same substrate specificity as the MgATPase activity. MgATPase activity Substrate

Mitochondrial

Tonoolast

pmol PJmg protein. h (%)

ATP-BTP ATP-Na2 AMP-Na2 ADP-Na2 GTP-Na2 PP;-N&

9.34 (100) 8.14 (87) 0.09 (1) 0.01 (0.1) 5.33 (57) -0.11 ( 0 )

9.61 (100) 8.30 (86) 1.15 (12) 1.16 (12) 3.41 (36) 2.71 (28)

MgSO, was added to a 1-ml reaction mixture at 23 "C to initiate H+ pumping. The complete reaction mixture contained 10 mM HepesBTP (pH7.5), 0.33 mM EGTA, 2.5 p~ quinacrine, 190 mM mannitol, 1.5 mM ATP, 1.5 mM MgSO,,50 mM BTP-C1, and 50-80pg of membrane protein. Under these conditions, most of the H+-pumping activity was resistant to vanadate or oligomycin (7).The initial rate of fluorescence quenching (per cent quenching/min) was used as a relative estimate of the rate of H+ pumping (6). Expression of Inhibition Data-We have used the inhibitor concenof either ATPase activity or tration required for 50% inhibition (L) quinacrine fluorescence quenching (relative H+ pumping) as a measure of the relative inhibitory potencies. Similar protein concentrations (pg/ml) were used for the ATPase and H+ pumping assays from one experiment to another. Equivalent amounts of proteins were also used for the mitochondrial ATPase assays. Since wedo not know howmuch of the total protein was contributed by ATPases, the comparison between native tonoplast and native mitochondrial ATPase activities a t this initial stage is semi-quantitative. Under our experimental conditions, when the membrane proteins varied from 28-46 pglml, the inhibitor concentrationsrequired for 50% inhibition were unchanged for each inhibitor tested. RESULTS

Substrate Specificity

Both the mitochondrial and tonoplast-type enzyme hydrolyzed ATP preferentially and GTP partially (Table I). AMP or ADP were not hydrolyzed. Hydrolysis of GTP and ATP could be catalyzed by the same enzyme since the purified F1 component from mitochondria hydrolyzes GTP about 78% as effectively as ATP.3 However, hydrolysis of PPi by the lowdensity vesicle fraction is catalyzed by an enzyme separate from the H+-ATPase. A H+-pumping pyrophosphatase associated with the vacuolar membranes of various plant tissues4 (7,24) has properties differentfrom the H+-pumping ATPase (25, 26). The tonoplast-type ATPase had a higher affinity for the substrate MgATP (Fig. 2a) than the mitochondrial ATPase (Fig. 2b). Hanes-Woolf analyses yielded an apparent K , for The abbreviations used are: Hepes, 4-(2-hydroxyethyl-l-pipera- ATP of 0.25 and 0.8 mM for the tonoplast-type andmitochonzineethanesulfonic acidBTP, bis-trispropaneor1,3-bis[tris(hydrial ATPase, respectively (Fig. 2 , a and b). In the absence of droxymethyl)methylamine]propane; DCCD, N,N'-dicyclohexylcarbodiimide; DIDS, 4,4'-diisothiocyanostilbene-2,2'disulfonate; activating ions, the Vmluof the tonoplast-type and the mitoDTT, dithiothreitol; NBD-C1, 7-chloro-4-nitrobenzo-2-oxa-1,3-dia-chondrial MgATPases were 15.6 and 32 pmol of Pi/mg of zole; NEM; N-ethylmaleimide; quercetin, 3,3',4',5,7-pentahydroxy- protein. h, respectively. BTP-C1 in the absence or presence of flavone; ApH, pH gradient;A$; membrane potential; EGTA, ethylene gramicidin increased the Vmsrof the tonoplast-type ATPase glycol bis(B-aminoethyl ether)-N,N,N'JV'-tetraacetic acid. to 20.8 or 29.4 pmol of Pi/mg of protein. h, respectively, with Figs. 1-5 are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. no change in theK , (Fig. 2a, inset). All subsequent assays for both types of ATPases were conducted with 3 mM MgATP. Full size photocopies are available from the Journal of Biological Though AMP and ADP were not hydrolyzed by the tonoChemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 85M-470, cite the authors, and include a check or money plast-type ATPase, we tested the possibility that they might order for $4.40 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

S. K. Randall and H. Sze, manuscripts in preparation.

'Y. Wang and H. Sze, unpublished results.

Mitochondrial H+-ATPases andTonoplast

10436

TABLEI1 Anion stimuhtion of the mitochondrial and tortoptost-typeATPases from oat roots K-salts when added were 50 mM except for 25 mM K2S0,. ATPase activity Anion

Mitochondria Total

~

~

ASalt

Tonoplast Total

~~

ASak ~

(Fig. 6 4 . We were surprised to find that the mitochondrial ATPase from oat rootswas equally sensitive to NO; (Fig. 65). About 85-90% of the ATPase activities was inhibited by 50 mM NO;. H' pumping in tonoplast vesicles wasalso inhibited by a few mMNOT (Fig. 6c) (5), such that ApH (I50= 1-2 mM) was decreased more than ATPase activity. The tonoplasttype ATPase could also be inhibited by KCIOswith an appar-

~~

pmol P;/mgprotein. h

None KC1 KBr

KI KHC03 K80, KIDA" a

8.91 8.00 18.55 1.44 10.35 1.31 10.22 -0.81 8.10 6.17 15.08 7.93 -0.98 7.72 -1.19

10.55 19.47 16.20 15.25 12.15 14.05

8.92 5.65 4.70 1.60 3.50

KIDA, K-iminodiacetate.

act as competitive inhibitors by binding to the same site as ATP. We had previously found that AMP or ADP inhibited ATP-dependent H' pumping intomicrosomal vesicles of soybean roots.5 Fig. 3, a, b, and c, shows that increasing concentrations of ADP and AMPdecreased the rateof ATP hydrolysis. Lineweaver-Burk analyses showed that AMP or ADP inhibition was competitive, and the Kiof ADP or AMP was about 0.4 mM. Competitive inhibition by AMP orADP of the mitochondrial ATPase was also observed (data not shown). These results indicate that AMP or ADP can bind to the same site(s) as ATP, and the levels of AMP or ADP might modulate the activity of the electrogenic H' pump. pH Optimum The mitochondrial ATPase activity had a distinct pH dependence (Fig. 4). Azide-sensitive ATPase activity was optimal at pH 8.0-9.0 and had almost no activity at pH 6.0-7.0. The ATPase activity associated with low-density vesicles displayed a broad pH dependence with a large peak of activity at pH 7-7.5 and a small peak of activity at pH 8.0-8.5 (Fig. 5). Similar pHdependence was observed whether activity was expressed as (i) gramicidin-stimulated ATPase, (ii) C1--stimulated ATPase, or (iii) NOT-sensitive ATPase (Fig. 5).

Anion Stimulation Although both the tonoplast-type and mitochondrial ATPase were stimulated by anions, the specific anion(s) required for stimulation varied. The mitochondrial ATPase of oat rootswas stimulated optimally by HCO; and very slightly by C1- or Br- (Table 11) as in rat liver mitochondria (27, 28). Sulfate or iminodiacetate were ineffective. However, the tonoplast-type ATPase was maximally stimulated by C1- and Br-. HCQ; and iminodiacetate were partiallystimulatory, though they were ineffective in stimulating pH gradient generation (7). The significance of anion stimulation in either enzyme is not understood. Permeant anions (e.g. C1-) dissipate the membrane potential generated in vesicles (positive inside) by the tonoplast-type ATPase, however, C1- stimulated ATP hydrolysis even in the absence of an electrochemical gradient (5). Thus we and others have interpreted the anion-stimulated ATPase as an enzyme with sites for anion binding which is required for optimal catalytic and vectorial activity (5, 6). Inhibitors Nitrate and DZDS-To our knowledge, the tonoplast-type ATPase is unique among other eukaryotic ATPases in its sensitivity to NO; inhibition at low levels (Iso = 7-8mM) T-H. Chen and H. Sze, unpublished experiments.

Mito

b

-&-

0

0

1

0

'

1

1

'

1

5

1

I

I

1

10

15

20

25

BTP-N03(mM) FIG. 6. Nitrate inhibition of the (a)tonoplast (Tp)-type ATPase, (b) the mitochondrial MgATPase activities, and (c) H+ pumping by the tonoplast ATPase. In the absence of NOT, activities of the tonoplast and mitochondrial ATPase were 8.0 and 14.0 pmol of PJmg of protein. h, respectively. Relative H+ pumping is expressed as per cent of control which is the initial rate of quinacrine fluorescence quenching (%Q/min) in the absence of nitrate. Results are the average of two (H+ pumping) orthree(ATPase) experiments.

Tonoplast and Mitochondrial H+-ATPases

10437

ent I,, of 5 mM (not shown). This result supports the idea a TP that chlorate is a useful analog of nitrate (29). In spiteof the similarity between the tonoplast and mitochondrial ATPases, 'O0I the mode of NO: action is not understood. We showed before that NO, was a noncompetitive inhibitor ofC1- in the tonoplast-type ATPase (5) and have tested whether NO, might compete for the ATP site. The results were ambiguous. NO, appeared to be partly competitive with ATP (data not shown). We have previously shown that DIDS preferentially inhibited the tonoplast-type H'-ATPase (Fig. 7a) andnotthe plasma membrane ATPase activity (5,7 ) . Here we show the mitochondrial ATPase was equally sensitive to DIDS (I5,, = 4-5 PM) as the tonoplast-type ATPase (Fig. 7 , a and b).H+ pumping in low-density vesicles was also inhibited (Fig. 7c) as shown before (5). The mode of action of DIDS on the tonoplast ATPase is not understood yet. DIDS can interact with anion-binding sites and is a specific inhibitor of anion 100 Mito b corn transport (e.g. C1-) inhumanerythrocytes(30)and protoplasts (31). We have previously shown that DIDS may act at anion-binding site(s) of the tonoplast-type ATPase as the presence ofC1- partially protected the inhibitory effects of DIDS (5). We also tested the possibility that DIDS might compete for the ATP site, as these molecules are anions and 50 may associate with common sites. The inhibitory action of DIDS on the native tonoplast-bound ATPase appeared to be noncompetitive (data not shown). DCCD-The tonoplast-type ATPase, expressed as either MgATPase or C1"stimulated ATPase activity, was inhibited 50% by 3-4 WM DCCD (Fig. 8a). H+ pumping into vesicles as determined by quinacrine fluorescence quenching was also I I 1 0 ' inhibited byDCCD with a of about 4-5 PM (Fig. 8b). 0 5 10 15 However, the mitochondrial H'-ATPase from oat roots was about 100 times more sensitive than thetonoplast H+-ATPase to inhibition by DCCD (Fig. 8a). The Im for inhibition of the mitochondrial ATPase was 0.03 pM. At low concentrations, DCCD inhibitsATPase activity and H+ pumping in the mitochondria by binding to a proteolipid in theFo region and thus blocks H' conductance (32). It is possible that DCCD inhibits H+ pumping and ATPase activity in the low-density microsomal vesicles from various plant tissues by a similar mechanism. If so, the differential sensitivity to DCCD would suggest that the properties of the H+ channels from the mitochondrial and tonoplast-type ATPasemay be different. NEM and NBD-C1-N-Ethylmaleimide, a sulfhydryl reagent, inhibitedthe MgATPase activity of the tonoplast more than the MgATPase activity of the mitochondria (Fig. 9, a and b ) . However, C1"stimulatedATPase from the low-density vesicleswas extremely sensitive to NEM inhibition, being 0 5 10 15 20 25 50% inhibited by 2 PM. Similar levels of NEM inhibited ApH DIDS (pM) formation which is routinely determined in the presence of FIG. 7. DIDS inhibition of the (a)tonoplast (!&)-type ATPC1- (Fig. 9c). Though the mitochondria MgATPase activity ase, (b)mitochondrial MgATPase activities, and ( c ) H+ pumpwas relatively resistant to NEM, HC0:-stimulated ATPase ing by the tonoplast ATPase from oat roots. In the absence of activity was 50% inhibited by 3-5 p~ NEM. This inhibition DIDS, activities of the tonoplast-type and mitochondrial ATPase was not caused by indirect effects of NEM on other mito- were 9.5 and 14.4 pmol of Pi/mg of protein. h, respectively. Results of ATPase and H+pumping (quinacrine fluorescence quenching ( 8 ) ) chondrial proteins, as the HCOF-stimulated activity of the are the average of three and two experiments,respectively. purified F1-ATPase is equally sensitive to NEM.3 These results suggest both the mitochondrial and thetonoplast ATPases have sulfhydryl groups required for anion activation and chondrial MgATPase activity to NBD-C1 inhibition. The concentration required to inhibit H+ pumping in tonoplast perhaps anion binding. NBD-Cl has been shown to inhibit the F,Fo-ATPases (mi- vesicles varied between 0.4-1.0 PM (Fig. 1Oc). Whether these tochondrial, chloroplast, or bacterial) by reacting with the results indicate the presence of tyrosine residues in the catahydroxyl group of a tyrosine residue in the catalytic subunit lytic subunit of the tonoplast ATPase has yet to be established. (32). We found that the MgATPase and the C1"stimulated The inhibition of H+ pumping or ATPase activities was ATPase activity were inhibited by NBD-Cl with an I,, of 9 and 2 PM, respectively (Fig. 10, a and b ) . The tonoplast most effective when the tonoplast vesicles were preincubated MgATPase activity is 5-fold more sensitive than the mito- with NEM or NBD-Cl, in the absence of ATP. Figs. 9c and

t

-

2b

'coe 7


10438

lo0l

TP

a

Mito

b

TD 0

;

I

3

I 10

I

1

20

DCCD(vM1

FIG. 8. Differential inhibition by DCCD of (a)the tonoplast

(Tp)-type ATPase and mitochondrial ATPase activities and (b) H+ pumping by the tonoplast ATPase, In the absence of DCCD, MgATPase (0)and CI--stimulated ATPase (A)activities of the tonoplast were 23.5 and 7.1 pmol of Pi/mg of protein. h, respectively. Mitochondrial MgATPase activity (0)in the absence of DCCD was 26.4 pmol of Pi/mg of protein. h. Membranes were preincubated with respective DCCD concentrations for 5-10 min a t room temperature prior to assaying. The results of ATPase and H+ pumping (quinacrine fluorescence quenching ( Q ) )represent the average of five and two experiments, respectively.

the H+ pumping activity, when ATP was added at the same time as the inhibitors. DTT was also effective in protecting against inhibition by NEM or NBD-Cl (Table 111). In thepresence of 1-2 mM DTT, 60-70% of the H+ pumping activity was retained at inhibitor concentrations that decreased 80-90% of H+ pumping activity (Table 111). Under the experimental conditions used, DTT could not reverse the inhibition by NEM; however, it appeared to partially reverse the effects of NBD-Cl. Quercetin-Quercetin inhibits the activity of the soluble and themembrane-bound mitochondrial ATPase from animal tissues (32). We found the MgATPase activity of the tonoplast was more sensitive to quercetininhibition than the C1" stimulatedATPase activity (Fig. 11, a and b), suggesting 1Oc showed that 1.5 mM ATP partiallyprotected

r

' w

\ 4 cL - 0 " l 2

\+ATP

5

10

50

100

NEM (pM) FIG. 9. NEM inhibition of the tonoplast (-)-type and mitochondrial ATPase activities. DTT was removedfrom the membrane preparationsby diluting the tonoplast vesicles or mitochondria in resuspension buffer without DTT (10-fold), pelleting, and resuspending in DTT-free medium. Membranes were preincubated with respective NEM concentrationsfor 5 rnin a t room temperature before initiation of assay with ATP. Results are the average of two experiments. u, tonoplast-type ATPase. In the absence of NEM, MgATPase (0)and C1"stimulated ATPase (A)activities were 18.3 and 10.0 pmol of Pi/mg of protein. h,respectively. 6, mitochondrial ATPase. In the absence of NEM, MgATPase (0)and HCOS-stimulated ATPase (A) activities were 21.3 and 30.2 pmol of Pi/mg of protein. h, respectively. e, H+ pumping by the tonoplast ATPase. Relative quinacrine fluorescence quenching (XQlmin) was determined in vesicles after preincubation with NEM in the absence (0)or presence of 1.5 m M ATP (A).

Mitochondrial H+-ATPases andTonoplast

10439 TABLE 111 Dithiothreitol protection of H + pumping activity in the presence of NEM or NBD-C1 in tonoplast vesicles Concentration of reactants in A are final concentrations in assay mixture. Pretreatment time is indicated in parenthesis.In B, vesicles were pretreated with 17 p~ NEM or 2 mM DTT for a period of time as indicated in parenthesis and diluted 10-fold in the final reaction mixture. H+pumping was initiated by adding ATP andMg. Quinacrine fluorescence quenching

Treatment

A. Control (0.5% ethanol) 3 p~ NBD-Cl(4 min) 33 3 p~ NBD-Cl(4 rnin), 1mM DTT (3min) 3 p~ NBD-C1 1 mM DTT (4 min) B. Control (no ethanol) 100 17 p~ NEM (5 min) 17 PM NEM (5 min), 2 mM DTT (2 min) 2 mM DTT (2 min), 17 p~ NEM (5 min)

+

arbitrary unitslmin

%

17 3.1 5.5

100 19

10

61

18.5 1.8 1.0

10 5

13.0

70

the tonoplast MgATPases were equally sensitive to quercetin (Im = 13-18 PM), the change in sensitivity to increasing quercetin concentration differed (Fig. l l b ) . DISCUSSION

Properties of the Tonoplast-typeH+-pumpingATPaseThis paper describes the first systematic characterization of the inhibitor sensitivities of aplanttonoplast-type H+ATPase. The tonoplast-type ATPase activity and ATP-dependent H’ transport had similar sensitivities to inhibitors, such as NEM, DCCD, DIDS, NBD-C1, and quercetin as summarized in Table IV. These inhibitors probably act directly on specific parts of the enzyme complexand thusinhibit ATP hydrolysis and H+ pumping. Nitrate decreased ApH formation at lower concentrations (I5o= 1-2 mM) than that needed to inhibit the ATPase activity (Im = 7-8 mM). These results could be caused by indirect effects of nitrate on A$ and ApH because of its permeant property. Nitrate collapses the membrane potential 50% at 2 mM (5) apparently by moving into the vesicles down the electrical gradient. It may then dissipate ApH by moving out of the vesicles in parallel with H+ via perhaps a H+/NOT symport system (33). The W differential effects of NBD-C1 on ATP hydrolysis (Iso = 2 or 9 pM) and H+ transport (I6o= 0.4-1.0 PM) is not understood yet. Perhaps, NBD-C1 interacts with tyrosine residues in the -1 3 H’ channel as well as the catalytic part of the ATPase, as I I I t I shown for mitochondrial ATPase from beef heart (34). The 1 10 100 relatively high sensitivity of H+ pumping to quercetin might NBD-CI (PM) also be caused by inhibition of the H+ channel in addition to FIG. 10. NBD-Cl inhibition of the (a)tonoplast (2’’)-type ATPase, (b) mitochondrial ATPase activities, and (c) H+ ATP hydrolysis. The inhibitorstudies combined with the pumping by the tonoplast ATPase from oat roots. In theabsence results of anion stimulation and substrate affinity (Table IV) of NBD-Cl, the MgATPase (0) and C1”stimulated (A) ATPase support the previous conclusion that ATP-dependent H+ activities from a were 15.4 and 7.0 pmol of Pi/mg of protein. h, pumping in low-density vesicles reflect activity from the tonrespectively. The control mitochondrial ATPase activity (0)was 11.1 rmol of PJmg of protein. h. DTT was removed from the membrane oplast-type ATPase (5, 7 , 19). BothNEMand NBD-Cl inhibition of H+ pumping in preparations as described in Fig. 9. Membranes were preincubated with respective NBD-C1 concentrations for 5 min at room tempera- tonoplast vesicles could be protected by preincubation with ture before addition of ATP. In p a n e l c, 0 indicates relative H+ ATP (Figs. 9c and I l c ) , suggesting they may act at a catalytic pumping (quinacrine fluorescence quenching (Q))when membranes site (35). Inhibition by NEM or NBD-C1 was also protected were pretreated with 1 p~ NBD-CI in the presence of 1.5 mM ATP. against by dithiothreitol (Table 111) consistent with the idea Results are theaverage of three experiments. that the active site involves a tyrosine or cysteine residue, or both (17).We are purifying the solubilized ATPase, and our quercetin, unlike NEM, does not interactwith the C1”binding preliminary results suggest a multisubunit enzyme ~ o m p l e x . ~ sites of the enzyme. H+ pumping was inhibited 50% by about Similarities and Differences from the Mitochondrial ATP2 PM quercetin (Fig. l l c ) .Though boththe mitochondrial and ase-Several higher plant tissues have proven to be good ex-

8j ._

L


10440 100

50

0

1

3

10

30

100

QUERCETIN (pM) FIG. 11. Quercetin inhibition of the mitochondrial MgATPase and thetonoplast (Tp)H+-ATPaseactivities. DTT was removed as described in Fig. 9. a, tonoplast-type ATPase. MgATPase (0)and C1"stimulated ATPase (A) activities inthe absence of quercetin were 18.3 and 10.0 pmol Pi/mg of protein. h, respectively. b, differential sensitivity of the mitochondrial (0)and tonoplast MgATPase (0)activities to quercetin. MgATPase of the mitochondria in the absence of quercetin was 21.3 pmol of Pi/mg of pr0tein.h. c , H+ pumping by the tonoplast ATPase expressed as relative quinacrine fluorescence quenching (Q).

perimental systems to study the vacuolar membrane ATPase. The tonoplast-type ATPase can be isolated essentially free from contamination by mitochondrial ATPase because of the low density of the tonoplast vesicles (1.10-1.13 g/cc) (3). This property, plus the insensitivity to azide, were used

to distinguish between the properties of the tonoplast ATPase and the mitochondrial ATPase. The mitochondria H+ATPase from plant tissues has not been as well characterized as that from mammalian mitochondria. Since plant mitochondria may differ from animal mitochondria, we made a systematic comparison of the tonoplast ATPase with the mitochondrial ATPase from the same tissue rather than from another source. Several striking resemblances suggest the tonoplast-type and the mitochondrial ATPases have common steps in their reaction mechanism (Table IV). Using the mitochondrial ATPaseasa model, the insensitivity to vanadate of the tonoplastATPase ( 5 , 7) suggests the reaction mechanism does not involve a covalent phosphoenzyme intermediate (17, 36, 37). The catalytic site of both ATPases mayinvolve tyrosine and cysteine residues based on the similar sensitivity to NBD-C1 and NEM and the protective effects of ATP or DTT. The similarity of the two catalytic sites is illustrated also by the ability to hydrolyze GTP partially as well as the sensitivity to competitive inhibition by ADP or AMP. Though our results do not clarify the site ofNO; or DIDS action, their similar effects on the mitochondrial andtonoplast ATPases suggest they bind to common sites, such as ATP- or anion-binding sites, or both. HCO; stimulation of rat liver mitochondrial ATPase is thought to be caused by enzyme conformational changes after association of HCO, with anion-binding sites (27,28,38).We had interpreted C1- stimulation of the tonoplast H+-ATPase with a similar model ( 5 ) , though we do not understand why C1- is more effective than HCO,. Several reports on mammalian mitochondria have shown that theMgATPase activity is relatively resistant to NEM and mercurials in the absence ofHCO, (38) consistent with our data (Fig.9b). Here we show that HCOL-stimulated ATPase activity is NEM sensitive (Fig. 9b), similar to the ratliver mitochondria (38). Since the anion-stimulated activity of both the tonoplast and mitochondrial ATPases were extremely sensitive to NEM inhibition (Fig. 9, a and b), we suggest the anion-binding sites or the conformation change induced by anions require sulfhydryl groups. The properties of the proton channel of the tonoplast and mitochondrial ATPase may be slightly different. Both ATPases are inhibited by DCCD, but the concentration required for 50% inhibition of the tonoplast ATPase is100-fold higher than that of the mitochondrial ATPase (Fig. 8, Table IV). These results arequalitatively similar to theNeurospora (39) or yeast (40) vacuolar ATPase, the chromaffin granule (10) or platelet granule ATPase ( l l ) , which are less sensitive to DCCD thanthe mitochondrial ATPase. Experiments are underway to identify the DCCD-binding protein and todetermine whether the DCCD-binding protein of the tonoplast ATPase is larger (14-21 kDa) than theone from mitochondria as shown in Neurospora (39) and recently in yeast (40). At present, the only activity that distinguishes the mitochondrial ATPase from the tonoplast ATPase unambiguously is the sensitivity of the mitochondria enzyme to azide or oligomycin (Fig. 1, Ref. 8). Our preliminary results of the purified tonoplast ATPase from oat roots3 are strikingly similar to those from yeast (40) where the purified vacuolar ATPase differed in its subunit composition and molecular weight from those of the mitochondrial ATPase. Relation to Other Tonoplast H+-ATPases and Animal H+ATPases-The inhibitor sensitivities reported here for the tonoplast ATPase from oat roots resemble those of the yeast (40) or Neurospora (39) enzyme, suggesting that the higher plant and fungal H'-ATPases are similar. It is less clear how


10441

TABLEIV Properties of the native mitochondrial and tonoplast H+-ATPasesfrom oat roots Tonoplast Parameter

ATPase

pH optimum Km(ATP) Anion stimulation Inhibitorb Vanadate Nitrate DIDS DCCD NEM Quercetin NBD-Cl Azide

Mitochondria/

H+pumping

ATPase

7.0-7.5 0.25 mM Cl-, Br- > HCOT

8.0” 0.11 mMa C1- > Br- > HCO;“

8.0-9.0 0.8 mM HCO, >> C1-, Br-

Not inhibited“ 8 mMb 5 PM 3.5 p M 35 p~ (MgATPase) 2 p~ (anion stimulated) 13 p M 9 p~ (MgATPase) 2 p M (anion stimulated) Not inhibited

Not inhibited“ 1-2 mM 5-7 pM 4-5 p M 2 fiM

Not inhibited 7 mM 5 fiM 0.035 p M >300 pM 4 PM 18 p M 50 p M ND’