Ca2+-dependency and substrate specificity of cholinergic synaptic vesicle ATPase

Neurochem. Int. Vol. 11, No. 1, pp. 107-111, 1987 Printed in Great Britain. All rights reserved

0197-0186/87 $3.00+ 0.00 © 1987 PergamonJournals Ltd

Ca2+-DEPENDENCY A N D SUBSTRATE SPECIFICITY OF CHOLINERGIC SYNAPTIC VESICLE ATPASE HEIKE STELZL, ERNST J. M. GRONDAL and HERBERT ZIMMERMANN* AK Neurochemie, Zoologisches Institut der J. W. Goethe-Universit/it, Siesmayerstr 70, D-6000 Frankfurt am Main, Federal Republic of Germany (Received 27 January 1987; accepted 26 February 1987) AlUraet--The Mg2+-ATPase associated with highly purified cholinergic synaptic vesicles isolated from the Torpedo electric organ can be activated by millimolar concentrations of Ca 2+ only to a maximum of 46%. ATP is hydrolyzed in preference to other tri-nucleotides on Mg2+-activation but not on CaZ+-activation of the enzyme. Analysis of the Ca2+-dependency of vesicle associated ATPase-activities under controlled free Ca2+-concentrations (10 -7 to 10-3M) shows that vesicles do not contain a Mg2+-dependent, CaZ+-stimulated (Ca2+ + Mg2+) ATPase. Our results suggest that cholinergic synaptic vesicles contain a single ATPase preferentially stimulated by Mg:+-ions and that ATP-stimulated vesicular uptake of Ca2+ is not due to the presence of a Mg2+-ATPase stimulated by micromolar Ca2+-concentrations.

Activity of ATPase (ATP hydrolase; EC 3.6.1.3) insensitive to ouabain or oligomycin is associated with fractions of cholinergic synaptic vesicles isolated from the Torpedo electric organ (Breer et al., 1977; Rothlein and Parsons, 1979; Michaelson and Ophir, 1980). The enzyme is modestly stimulated by ACh (Breer et al., 1977) and by bicarbonate (Rothlein and Parsons, 1980, 1982). It has unanimously been reported to be activated by millimolar concentrations of either Ca 2+ or Mg 2+ hut the degree of Ca2+-activation varies between authors (60-90% of Mg 2+-activation). Like for a variety of other vesicular organelles (endocytotic vesicles, chromaflin granules, lysosomes) synaptic vesicle ATPase has been implicated in proton pumping (Anderson et al., 1982; Harlos et al., 1984). The proton electrochemical gradient produced by the vesicular ATPase appears to be directly involved in vesicular uptake of ACh (Anderson et al., 1982). Cholinergic synaptic vesicles also display an ATPstimulated uptake of Ca 2+ (Michaelson et al., 1980; Isra61 et al., 1980; Rephaeli and Parsons, 1982). If this mechanism of uptake was similar to the Ca2+-pump *Address correspondence to: Dr Herbert Zimmermann, AK Neurochemie, Zoologisches Institut der J.W. Goethe-Universit~t, Siesmayerstr. 70, Postfach 11 19 32, D-6000 Frankfurt am Main 11, F.R.G. 107

of the sarcoplasmatic reticulum (Schuurmans Stekhoyen and Bonting, 1981) one might expect--in addition to the proton pumping ATPase---the presence of a second ATPase which is activated by Ca :+ in the micromolar range and involved in Ca2+-transport. A twofold increase of vesicular Mg2+-ATPase activity by addition of 5 0 # M Ca 2+ has been reported (Michaelson et al., 1980). In order to characterize the dependency on Ca2+-ions and the substrate specificity of the vesicle associated ATPase activity(ies) we performed a series of experiments with synaptic vesicle fractions of increasing purity. This was prompted by our observation (Grondal and Zimmermann, 1986) that an ATPase equally activated by millimolar concentrations of either Ca 2+- or Mg2+-ions is associated with the surface of Torpedo synaptosomal plasma membranes and the presumption that this membrane could be contaminant of synaptic vesicle fractions. EXPERIMENTAL PROCEDURES

The electric organ of Torpedo marmorata (supplied by the Institute Universitaire de Biologie Marine, Arcachon, France) was homogenized and fractionated in isotonic (0.4 M) NaCI solution, buffered with HEPES at pH 7.4 essentially as described earlier (Schmidt et al., 1980). Synaptic vesicles were purified using a variety of procedures. Discontinuous gradient: The crude vesicle fraction P3 obtained by differential centrifugation was further separated on a discontinuous isotonic HEPES-buffered NaCl/sucrose

10 N

HEIKE STELZLel a/. Table 1. Activities of ATPase and other markers in synaptic vesicle fractions of varying purity Type of vesicle fraction Discontinuous Continuous Flotation Sepharcryl,1000 colmnn P3 gradient gradient gradient column P~

ATP (nmol x mg protein ]) ATPase (nmol Pi x min] x mg protein ~) Mg2+ (4 mM)

19.0_+8.8 16.3± 10.3

133.4 + 35 37.7__4-4.4

182

1212 -+70.0

102

271 ?.: 12.7

912 +97

43

590± 10.7

36.2

Ca2+ (4 raM) 16.0 __+10.4(99) 28.0 + 5.4 (86) 68.6 (67) 205 ± 9.6 (76) 271 + 9.8 (46) 16.9 Acetylcholinesterase (/~mol x rain-i x mg protein J) ll.8 + 0.5 18.4 + 4.2 17.3 59.8_+ ll.7 42.1 + 19.1 3.6 Values are means _+SD of 3 experiments except for the continuous gradient (single experiment). Values in brackets express activity of Ca2+-ATPase as % of activity of Mg2+ ATPase. gradient as previously described (Schmidt et al., 1980). The vesicle fraction was harvested on top of the 0.4 M sucrose solution. Continuous density gradient: The fractionation was exactly as for the discontinuous gradient except that a continuous NaCl/sucrose gradient (0.2-1.6 M sucrose) was used. The synaptic vesicle peak fraction was identified by its ATP content at a density equivalent to a 0.4 M sucrose solution. Flotation gradient: Flotation of prepurified vesicles was performed as described previously (Carlson et aL, 1978). The synaptic vesicle fraction obtained from the discontinuous sucrose density gradient was centrifuged and the pellet obtained resuspended in HEPES-buffered 0.8 M sucrose solution. The suspension was layered on the bottom of a centrifuge tube and overlayed with lighter isotonic (addition of NaCI) sucrose solutions (0.55, 0.45, 0.2 M). After centrifugation the vesicle peak fraction (at 0.45 M sucrose) was identified by its content in ATP. Column chromatography: This method was performed essentially as described previously (Volknandt and Zimmermann, 1986). Vesicles prepurified by discontinuous density gradient centrifugation were centrifuged to form a pellet and the resuspended vesicle fraction was layered on a column of Sephacryl-1000 (Pharmacia) equilibrated with 0.4 M NaCI (10 mM HEPES-buffer; pH 7.4). Fractions of 1.7ml were collected and the vesicle peak identified by its ATP content. ATP, protein and activities of aeetylcholinesterase (acetylcholine acetylhydrolase; EC 3.1.1.7) and ATPase (deterruination of Pi) were determined as described previously (Grondal and Zimraermann, 1986) except that the concentration of the nucleotide was 0.5 mM instead of I mM. The free Ca2+-concentrations (in the presence of Ca2+-cbelators) in experiments to test for Ca2+-activation of the Mg2+ATPase were calculated from stability constants (Velema et al., 1985). In addition, the actual concentrations of free Ca 2+ in the incubation medium were measured with a Ca2 +-sensitive electrode. RESULTS

Ca2 ÷-dependency A TPase

of

cholinergic

synaptic

vesicle

As judged by the specific activity of A T P (Table 1) the purity of synaptic vesicles isolated by differential density gradient centrifugation (fraction P3) increases on application of additional separation techniques,

with density gradient centrifugation followed by flotation or column chromatography lacing the most effective procedures. The decrease of the specific activity of A T P in the vesicle peak fraction derived from the Scphacryl-1000 column is likely to be due to a loss of A T P from vesicles on passage through the column (Volknandt and Zimmermann, 1986) rather than to a smaller degree of purity. A similar purification pattern is observed for the specific activity of ATPase measured in the presence of M g 2+ (4 mM) except that activity is further increased after column chromatography resulting in a 36 fold purification over fraction P3. The specific activity of 590 nmol Pi x mg protein- 1 × m i n - ~ is by far the highest reported for the Mg2+-ATPase purified with intact cholinergic vesicles. The purification is considerably less pronounced when ATPase activity is assayed in the presence of Ca 2+ (4 mM) where only a 17 fold purification is obtained. This is paralleled by an increase in the ratio of Mg2+-ATPase to Ca2+-ATPase activity on improved purification. Whereas in the crude vesicle fraction P3 stimulation of enzyme activity is equal for Ca 2. and Mg 2+, Ca2+-stimulation falls to 46% of Mg2+-stimulation after column chromatography. Acetylcholinesterase a known constituent of the presynaptic plasma membrane in the Torpedo electric organ (Ref. Grondal and Zimmermann, 1986) is still slightly enriched in the most purified vesicle fraction. The notion that vesicular Ca2+-ATPase activity i s - - a t least in p a r t - - d u e to a contamination is further supported by Fig. I. The void volume of the Sephacryl-1000 column which contains larger membrane fragments exhibits equal activities of Ca2+-ATPase and Mg2+-ATPase. Only activity of Mg2+-ATPase forms a second peak together with synaptic vesicles (main A T P peak) whereas Ca2+-ATPase activity is steadily decaying in the column effluent. In this respect the elution pattern of

Cholinergic vesicle ATPase 20- • 10

40-

-20

109

~ 20-

-~15. Este

15-

se

30 ~

c IO-

E

E

E 5-

0-

5

/ I,

i' ' i 40 50 number of fnaction

30

i 60

i 70

0

Fig. l. Additional purification of synaptic vesicles isolated by sucrose density gradient centrifugation by chromatography on Sephacryl-1000. The following activities were loaded onto the column (% recovery), ATP: 625 nmol (82); Mg2+-ATPase (0.5raM ATP, 4mMMg2+): 657nmol Pi x rain -t (84); Ca2+-ATPase (0.5 mM ATP, 4mM Ca2+): 583nmol Pi x min - ' (75); acetylcholinesterase (esterase): 79 U (86); protein: 1.23 mg (80). Ca2+-ATPase activity is very similar to that of the plasma membrane marker acetylcholinesterase. Acetylcholinesterase has its main peak in the void volume but is also associated with particles of various size, small enough to be retarded by the column and to contaminate the synaptic vesicle fraction. Since the Ca2++ Mg2+-ATPase of sarcoplasmic reticulum may be stimulated at Ca2+-concentrations lower than 1 # M and inhibited at higher concentrations (Schuurmans Stekhoven and Bonting, 1981) we tested the enzyme activity at 2 m M M g 2+concentrations for a wide range of Ca 2+concentrations (Fig. 3). We cannot obtain evidence

a)

10080"

60,'-~40._>

~

0-

c~

~-----C>-----

i

7

i'

6

5

-tog [co2"] [M]

4

Fig. 3. Hydrolysis of ATP by chromatography purified synaptic vesicles at varying Ca2+-concentrations. ( I ) ATP, 0.5mM; Mg2+ 2mM; concentrations of EGTA from 30.7 mM ( 1 0 - 7 M Ca 2+) to 2.5 mM (10 -4 M Ca2+). (17) ATP, 0.5 mM; Mg2+, 2 raM; no EGTA added. (O) ATP, 0.5 raM; no Mg2+; concentrations of EGTA from 27.7 mM (10 -7 M Ca 2+) to 2.5 mM (10 -4 M Ca2+). Typical experiment with triplicate determinations derived from one vesicle preparation. for the presence (in chromatography purified vesicles) of a (Ca 2÷ + Mg:+)-ATPase activity which would be stimulated by micromolar concentrations of Ca 2÷ in the presence of Mg 2+. Interestingly, if high concentrations of EGTA [ethylene glycol bis(fl-aminoethyl ether)-N,N,N',N'-tetraacetic acid] are used as chelators the enzyme activity is increased by about 25%. In the absence of both Mg :+ and EGTA, Ca 2÷-activation of enzyme activity becomes apparent between concentrations of 10 and 100 #M.

Substrate specificity of cholinergic synaptic vesicle A TPase Activity of the synaptosomal plasma membrane Ca 2+ or Mg2+-ATPase has little selectivity for nucleoside triphosphates (Grondal and Zimmermann, 1986). In contrast, vesicular Mg2+-ATPase preferentially hydrolyzes ATP over other nucleotide triphosphates (Fig. 2a). This is not the case for the Ca2+-activated enzyme activity (Fig. 2b). This either indicates a different substrate selectivity of the same enzyme in the presence of Mg 2÷ or Ca 2+ ions or the contribution of a contaminating Ca2+-ATPase to the vesicle fraction. DISCUSSION

20-

Fig. 2. Hydrolysis of nucleotides (all 0.5 mM) by intact chromatography purified synaptic vesicles (comp. Fig. I). (a) 4 mM Mg2+. (b) 4 mM Ca2+. The 100% values correspond to 590 and 271 nmol Pi x min-! x mg protein-' in (a) and (b) respectively. Columns are mean values + SD of three experiments.

Our results suggest that cholinergic synaptic vesicles contain only a single ATPase activated by milEmolar concentrations of Mg2+-ions and to a smaller extent also by millimolar Ca2+-conccntrations. Activities of Ca2+-ATPase exceeding 46% of Mg 2+ATPase represent membrane contamination of the vesicle fraction. This is supported by the finding that,

110

HEIKE STELZLet al.

in the presence of bicarbonate, either Mg2+-ATP or Ca 2+-ATP support (proton gradient dependent) acetylcholine uptake into synaptic vesicles, whereby the effect of CaZ+-ATP is about half of that of Mg 2+-ATP (Parsons et al., 1982). This Ca2+-dependency of the cholinergic vesicle ATPase corresponds to that of the proton translocating ATPase of chromaffin granules. There Ca2+-activation is 20-50% of Mg2+-activation (Flatmark et al., 1985). The substrate specificity represents a further similarity to the chromaffin granule proton pumping ATPase. In accordance to our results on cholinergic vesicles the chromaffin granule Mg2+-ATPase is specific for A T P and uses other nucleotides with considerably reduced activity (Dean et al., 1986). Since our Ca:+-electrode measurements had revealed that an isolated vesicle fraction already has a free concentration of Ca 2÷ ions in the range of 30/~M (unpublished experiments) we found it essential to control the free Ca2+-concentration in our system. This has not been the case in previous studies (Michaelson et al., 1980; Rephaeli and Parsons, 1982; Diebler and Lazareg, 1985). We therefore can exclude that a (Ca2+ + Mg2+) ATPase stimulated (or inhibited) by submicromolar or low micromolar concentrations of Ca 2+ is present in cholinergic synaptic vesicles. A major conclusion from our results is that vesicular uptake of Ca -,+ ions does not appear to be driven by a Mg2+-dependent ATPase activated by micromolar Ca2+-concentrations as has been concluded from experimental evidence available (Whittaker, 1984; Stadler et al., 1985). The role of A T P in stimulating vesicular uptake of Ca 2÷ may be a different one. It has been demonstrated that calmodulin-dependent phosphorylation of vesicular proteins enhances vesicular Ca2+-uptake whereas vesicular ATPase is not stimulated by calmodulin (Rephaeli and Parsons, 1982). It has to be elucidated whether the vesicular (proton pumping) Mg2+-ATPase--by producing ionic gradients---could be indirectly involved in vesicular uptake of Ca 2 ~-ions. Acknowledgements--We would like to thank Dr Irene

Schulz, Max-Planck-Institut fiir Biophysik, Frankfurt, for Ca2+-electrode measurements, Monika Then for technical assistance, and Annemarie Polotzek for typing the manuscript. This work was supported by a grant of the Deutsche Forschungsgemeinschaft (SFB 169, A 10). REFERENCES Anderson D. C., King S. C. and Parsons S. M. (1982) Proton gradient linkage to active uptake of

[~H]acetylcholine by Torpedo electric organ synaptic vesicles. Biochemistry 18, 3037-3043. Breer H., Morris S. J. and Whittaker V. P. (1977) Adenosine triphosphatase activity associated with purified cholinergic synaptic vesicles of Torpedo marmorata. Eur. J. Biochem. 80, 313 318. Carlson S. S., Wagner J. A. and Kelly J. S. (1978) Purification of synaptic vesicles from elasmobranch electric organ and the use of biophysical criteria to demonstrate purity. Biochemistry 17, 1188-1199. Dean G. E., Nelson P. J. and Rudnick G. (1986) Characterization of native and reconstituted hydrogen ion pumping adenosine triphosphatase of chromaffin granules. Biochemistry 25, 4918-4925. Diebler M. F. and Lazereg S. (1985) Mg-ATPase and Torpedo chotinergic synaptic vesicles J. Neurochem. 44, 1633-1641. Flatmark T., Gronberg M., Husebye E. and Berge S. V. (1985) The assignment of the Ca2+ATPase activity of chromaffin granules to the proton translocating ATPase. FEBS Lett. 182, 25-30. Grondal E. J. M. and Zimmermann H. (1986) Ectonucleotidase activities associated with cholinergic synaptosomes isolated from Torpedo electric organ. J. Neurochem. 47, 871-881. Harlos P., Lee D. A. and Stadler H. (1984)Characterization of a Mg2+-ATPase and a proton pump in cholinergic synaptic vesicles from the electric organ of Torpedo marmorata. Eur. J. Biochem. 144, 441-446. lsra61 M., Manaranche R., Marsal J., Meunier F. M., Morel N., Franchon P. and Lesbats B. (1980) ATP-dependent calcium uptake by cholinergic synaptic vesicles isolated from Torpedo electric organ. J. Membrane Biol. 54, 115 126. Michaelson D. M. and Ophir I. (1980) Sideness of (calcium, magnesium) adenosine triphosphatase of purified Torpedo synaptic vesicles. J. Neurochem. 34, 1483-1490. Michaelson D. M., Ophir I. and Angel I. (1980) ATPstimulated Ca 2+ transport into cholinergic Torpedo synaptic vesicles. J. Neurochem. 35, 116-124. Parsons S. M., Carpenter R. S., Koenigsberger R. and Rothlein J. E. (1982) Transport in the cholinergic synaptic vesicle. Fed. Proc. 41, 2765-2768. Rephaeli A. and Parsons S. M. (1982) Calmodulin stimulation of 45Ca2+ transport and protein phosphorylation in cholinergic synaptic vesicles. Proc. Natn. Acad. Sci. U.S.A. 79, 5783-5787. Rothlein J. E. and Parsons S. M. (1979) Specificity of association of a Ca2+/Mg 2+ ATPase with cholinergic synaptic vesicles from Torpedo electric organ. Biochem. Biophys. Res. Commun. 88, 1069-1079. Rothlein J. E. and Parsons S. M. (1980) Bicarbonate stimulation of the Ca2+/Mg 2+ ATPase of Torpedo electric organ synaptic vesicles. Biochem. Biophys, Res. Commun. 95, 1869-1874. Rothlein J. E. and Parsons S. M. (1982) Origin of the bicarbonate stimulation of Torpedo electric organ synaptic vesicle ATPase. J. Neurochem. 39, 1660-1668. Schmidt R., Zimmermann H. and Whittaker V. P. (1980) Metal ion content of cholinergic synaptic vesicles isolated from the electric organ of Torpedo: effect of stimulationinduced transmitter release. Neuroscience 5, 625-638. Schuurmans Stekhoven F. and Bonting L. (1981) Transport triphosphatases: properties and functions. Physiol. Rev. 61, I 76.

Cholinergic vesicle ATPase Stadler H., Kiene L. M., Harlos P. and Welscher U. (1985) Structure and function of cholinergic synaptic vesicles. In: Neurobiochemistry, (Hamprecht B. and Neuhoff V., eds), 36. Colloquium--Mosbach 1985, pp. 55-65. Springer, Berlin, Heidelberg. Velema J., Bloom J. J. and Zaagsma J. (1985) Comparison of (Ca 2+ + Mg2+)-ATPase and Ca2+-ATPase in rat cardial sarcolemma. Int. J. Biochem. 17, 1091-1096.

111

Volknandt W. and Zimmermann H. (1986) Aeetylcholine, ATP, and proteoglycan are common to synaptic vesicles isolated from the electric organs of electric eel and electric catfish as well as from rat diaphragm. J. Neurochem. 47, 1449-1462. Whittaker V. P. (1984) The structure and function of cholinergic synaptic vesicles. Biochem. Soc. Trans. 12, 561-575.