vesicles (acetylcholine uptake/torpedo)

'Proc. Nati. Acad. Sci. USA Vol. 78, No. 4, pp., 2048-2052, April 1981 Biochemistry Saturable acetylcholine transport into purified cholinergic synaptic vesicles (acetylcholine uptake/torpedo) DANIEL M. MICHAELSON AND ITZCHAK ANGEL The George S. Wise Faculty of Life Sciences, Department of Biochemistry, Tel-Aviv University, Tel-Aviv, Israel Communicated by Julius Axelrod, November 24, 1980 ABSTRACT The uptake of [3H]acetylcholine ([3H]AcCho) into cholingeric synaptic vesicle ghosts purified from Torpedo electric organ was studied at concentrations of [3H]AcCho ranging from 0.1 to 10 mm. The accumulated [3H]AcCho can be released either by hypoosmotic buffer or by low levels of the detergent Triton X- 100. Kinetic analysis ofthe initial rate of [3H]AcCho uptake reveals temperature-dependent saturation kinetics which are best fitted by high-affinity (KTh 0.3 mm) and = low-affinity (KT1 10 mm) vesicular [3H]AcCho transport systems. Several lines of evidence suggest that [3H]AcCho transport is mediated by vesicle-associated transport systems and not by a contaminant of other subcellular moieties such as the plasma membrane choline transport system. (i) The specific activity of the [3H]AcCho transport systems is higher in the purest vesicular fraction than in the less-pure fractions. (ii) Ghosts prepared from isolated synaptosomes manifest only low levels of low-affimity [3H]AcCho transport and no highaffinity [3H]AcCho transport. (iii) The vesicular AcCho transport systems lack some of the typical characteristics of synaptosomal choline transport, such as Na' activation. (iv) The ratio of uptakes of [3H]AcCho and [3H]choline (10,uM) is about 5-fold higher in the pure vesicles than in isolated synaptosomal membranes. Addition of Mg2a-ATP decreases the rate of vesicular [3H]AcCho uptake by about 50%. The simultaneous addition of NaHCO3 and Mg2a-ATP results in activation of [3H]AcCho uptake to about 125% (relative to control), which is a 2.54fold enhancement relative to the rate observed with Mg2+-ATP. The present findings demonstrate'the presence of novel vesicle-associated AcCho transport systems. Their physiological role in the life cycle of the cholinergic synaptic vesicle and nerve terminal are discussed. Acetylcholine (AcCho) is stored in synaptic vesicles at a concentration of ca. 0.15-0.5 M (1-3) which is about 1 order of magnitude higher than in the presynaptic cytoplasm (=='30 mm) (4). AcCho is believed to be synthesized in the cytoplasm (5, 6). Hence, a carrier and a mechanism to provide the energy for its active transport into the synaptic vesicles are essential. In catecholaminergic synaptic vesicles and storage granules, for example, it has been shown that the biogenic amines are transported into the vesicles by a specific, reserpine-inhibited carrier and that the energy needed for concentrating the catecholamines in the vesicles-is provided by a vesicle-associated proton ATPase (for review, see refs. 7 and 8). Previous attempts have failed to demonstrate accumulative AcCho uptake in vitro into coritcal slices (9), sympathetic ganglion (10), or Torpedo cholinergic vesicles (11). Recently, Carpenter and Parsons (12) have demonstrated electrogenic AcCho uptake into hyperpolarized (interior negative) Torpedo vesicles. We have recently shown (13, 14) that the interior of isolated intact Torpedo synaptic vesicles is acidic, and that dissipation of the vesicular proton gradient results in AcCho release from The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. 1734 solely to indicate this fact. 2048 the vesicles. These findings suggest that uptake and storage of AcCho in cholinergic synaptic vesicles are affected by the internal ph and by the electric potential across the vesicular membrane. The presence of saturable AcCho transport has not yet been reported. In this communication we report that purified cholinergic Torpedo synaptic- vesicles take up [3H]AcCho and that the rate of [3H]AcCho transport follows saturation kinetics which are best fitted by two transport systems with affinity constants of ca. 0.3 mm and ca. 10 mm. EXPERIMENTAL Materials. Live Torpedo ocellata were caught off the coast of Tel-Aviv. Acetylcholine, choline, ethylene glycol bis(,b-ami noethyl ether)-n,n.n',n'-tetraacetic acid (EGTA), and ATP were from Sigma. Phospholine iodide was from Ayerst Laboratories. [3H]AcCho (90.0 mci/mmol; 1 Ci = 3.7 X 10' becquerels) was from New England Nuclear. [3H]Choline (8.3 Ci/mmol) was from The Radiochemical Centre (Amersham, England). All other chemicals were of reagent grade. Purification of Synaptic Vesicles. The electric organs were excised from chilled live T. ocellata. The excised tissue, 10-15% (wt/vol) in 0.8 M glycine/1 mm EGTA, ph 6.6, was homogenized and fractionated by differential centrifugation as described (15, 16). The S2 supernatant was then centrifuged at '250,000 X g for 1 hr; the resulting pellet (P3) which contained synaptic vesicles, was resuspended on 0.1 M sucrose/0.35 M NaCl/0.5 mm EGTA, ph 7.4, and loaded on a discontinuous density gradient as described (17). Density gradient centrifugation yielded a vesicle fraction (12) that was further purified (fraction SV) by gel filtration chromatography through a column packed with controlled-pore glass beads (Sigma, CPG-2500 200). Vesicle purity was determined both biochemically and morphologically (17). AcCho and ATP were present in a ratio of 3:1 and at a specific concentration of 1010 nmol of AcCho per mg of SV membrane phospholipid. Preparation of Vesicular Ghosts. Synaptic vesicle ghosts were prepared from the P3, 12, and SV fractions by hypoosmotic treatment. The crude vesicular fraction (P3) was resuspended in 100 mm sucrose/350 mm NaCl/0.5 mm EGTA, ph 7.4, lysed by a 1:10 dilution with 10 mm Tris-HCl (ph 6.8), pelleted (250,000 X g for 1 hr), and resuspended in 0.8 M glycine/10 mm Tris-HCl, ph 6.8, at a concentration of 5-10 mg of protein The vesicle fraction 12 was lysed by a 1:10 dilution with 10 mm Tris'HCl (ph 6.8), pelleted, and resuspended at a concentration of0.5-1.0 mg ofprotein per ml as described above. Fraction SV, after centrifugation and resuspension in 0.8 M glycine/ Abbreviations: AcCho, acetylcholine; EGTA, ethylene glycol bis(,1- aminoethyl ether)-nn,n',n'-tetraacetic acid.

Biochemistry: Michaelson and Angel 10 mm Tris1HCl, ph 6.8, was lysed by a 1:10 dilution with 10 mm Tris HCl (ph 6.8); the lysed vesicles were then pelleted and resuspended at a concentration of 0.05-0.15 mg of protein All vesicular ghosts were stored in liquid nitrogen. Prior to use, they were thawed and washed by centrifugation and resuspension in 0.8 M glycine/10 mm Tris HCI, ph 6.8, at the concentrations indicated above. Preparation of Torpedo Synaptosomes and Synaptosomal Ghosts. Torpedo synaptosomes were prepared as described (15, 16). The synaptosomes (fraction a2j were pelleted (20,000 X g for 30 min) and resuspended in 0.8 M glycine/ 10mM Tris HCI, ph 6.8, at a concentration of 2 mg of protein They were lysed by a 1:10 dilution with 10 mm Tris HCI (ph 6.8), and the resulting synaptosomal ghosts were pelleted. The pellet was washed twice by resuspension in 0.8 M glycine/10 mm Tris HCI, ph 6.8, and centrifugation. It was finally suspended at a concentration of 0. 5-1.0 mg of protein per ml in 0.8 M glycine/10 mm Tris HCI, ph 6.8. [3H]AcCho and [3H]Choline Uptake. Synaptic vesicles and synaptosomal ghosts, 0.7 ml at the concentrations depicted above, were preincubated for 20 min at 25 C and at least 15 min at 4 C with 80,umol of the acetylcholinesterase inhibitor phospholine iodide. [3H]AcCho uptake was routinely studied at 25 C. It was initiated by a 1:1 dilution ofthe membrane fractions with 0.8 M glycine/10 mm Tris-HCl, ph 6.8, containing [3H]AcCho (=3 X 106 cpm/ml) at the specified concentration. At various time intervals, aliquots (0.1 ml) were placed in duplicate on GF/C filters (Whatman) which were rapidly washed (five times with 2-ml portions)- with ice-cold 0.8 M glycine/10 mm Tris HCl, ph 6.8. The filters were placed in vials containing 5 ml of Hydroluma (Lumuc) scintillation fluid and maintained at 25 C for 30 min. The radioactivity was then assayed by liquid scintillation spectrometry (Packard Prias, model PL). Standard tritiated water (Packard) was used to establish the efficiency of counting (40-45%). In order to reduce nonspecific adsorption of [3H]AcCho, prior to the experiment the GF/C filters were treated for 10 min by immersion in 0.8 M glycine/ 10 mm Tris1HCI, ph 6.8, containing 50 jig ofpolylysine (Sigma) Under these conditions the amount of [3H]AcCho taken up by the vesicular ghosts (at t =- 2 min) was about 3-fold higher than the blank measured at t = 0. When the effect of buffer composition on [3H]AcCho uptake was examined, the reaction was initiated by diluting the vesicles with the appropriate isoosmotic buffer (1:1) to yield the desired final concentrations. The validity of the filtration method was ascertained by control experiments in which intact vesicles were filtered on GF/C filters. After filtration, the AcCho retained on the filters was extracted and measured by bioassay (18). This control revealed that 95 ± 5% of the vesicular AcCho was retained by the filter, thus demonstrating that the vesicles are trapped by the filter and that they are not ruptured by the filtration procedure. [3H]Choline uptake was similarly assayed except that the reaction mixture contained [3H]choline (-3 x 10' cpm/ml) at the desired concentration and no AcCho. The reaction was initiated by a 1:1 dilution ofthe membrane fractions with 200 mm NaCl/ 6 mm KCI/400 mm glycine/10 mm Tris HCl, ph 7.0, for [3H]choline uptake in the presence ofna+ or with 0.8 M glycine/ 10 mm Tris HCl, ph 6.8 for [3H]choline uptake in the absence of Na+. Control experiments in which [3H]choline was added in trace amounts to the vesicles and synaptosomes prior to lysis, coupled with direct measurements ofthe AcCho contents ofthe ghosts (18), revealed that less than 0.1% of the original choline and AcCho were retained after lysis and washing. Thus, the concentrations of external choline derived from the membranes Proc. Natl. Acad. Sci. USA 78 (1981) 2049 was less than 0.1 pm in the reaction mixture and therefore did not affect the results. Measurement of Internal Volume. The internal volume of purified synaptic vesicles was measured as described (14), except that protein was determined according to Bradford (19). RESULTS Incubation of ghosts prepared from the I2 synaptic vesicle fraction and suspended in 0.8 M glycine/10 mm Tris HCl, ph 6.8, with 0.2 mm [3H]AcCho resulted in accumulation of radiolabel in the vesicles. Uptake was linear for 2 min and then leveled off to a plateau (Fig. 1). Similar results were observed with the P3 and SV fractions. The greatest extent of [3H]AcCho accumulation, as measured at the plateau, was seen in the pure vesicles. Incubation of the pure vesicles with 0.2 mm [3H]AcCho resulted in uptake of ca. 12.1 nmol of [3H]AcCho per mg of protein, whereas the less pure (12 fraction) (Fig. 1) and the crude P3 vesicle fractions took up only ca. 1 nmol of [3H]AcCho per mg of protein. All uptake experiments were performed in the presence of 40,uM phospholine iodide, which completely inhibits the trace acetylcholinesterase activity associated with the synaptic vesicle fractions (17). Furthermore, deliberate hydrolysis of [3H]AcCho prior to the experiment resulted in complete abolition of uptake, confirming that the radiolabel accumulated in the vesicles was [3H]AcCho and not any of its breakdown products. Dilution of the vesicles with hypoosmotic buffer resulted in loss ofthe vesicle-bound [3H]AcCho (t,12 2 min) (Fig. 1). Efflux of[3h]accho was likewise observed after dilution ofthe vesicles with isoosmotic buffer containing unlabeled AcCho or no AcCho (Fig. 1). A similar effect was achieved by treating the vesicles with the nonionic detergent Triton X-100 at 0.1% (not shown) except that, under these conditions, the rate of liberation of the vesicle-bound [3H]AcCho was faster (t/2 0.5 min). Measurements of the rate of vesicular [3H]AcCho uptake as a function of [3H]AcCho concentration revealed that, within the 0 1.2-0~~~~~~~ ~0.4f I 0 ~ ~ ~ 0 0.4 -~~~~~~~~ 2 4 6 8 10 Time, min FIG. 1. Time course of [3H]AcCho uptake by isolated cholinergic synaptic vesicles ghosts (o) at 25 C. At the indicated (vertical arrow) time the vesicles were diluted 1:10 with either water (o) or a reaction buffer that contained no [3H]AcCho (e). The ghosts were prepared from vesicles purified by density gradient centrifugation (fraction I2). Results presented are mean ± SD of three experiments.

2050 Biochemistry: Michaelson and Angel Proc. Natl. Acad - Sci - USA 78 (1981) 1/[3H]AcCho, mm-1 V FIG. 2. (A) Double-reciprocal plot of initial rate of [3H]AcCho uptake by isolated synaptic vesicles ghosts. Data are best described by two straight lines with.positive ordinate intercepts, indicating a two-component system. The lines were calculated by using linear regression analysis. The experiments were performed in duplicate at 25TC, and.the ghosts were preparedfom vesicles purified by density gradient centrifugation (fraction I2). The velocity (V) is expressed as nmol of [3H]AcCho accumulated per mg ofprotein per min. (B).Scatchard (20) plot of the data in A. The presynaptic membrane ofcholinergic nerve endings contains a Na'-dependent high-affinity choline carrier (21, 22). Hence, it may be argued that the results presented above are due to transport of [3H]AcCho by the choline carrier and not via the novel vesicular [3H]AcCho transport system described here. As shown in Fig. 4, membrane ghosts prepared from iso-.lated Torpedo synaptosomes did accumulate [3H]AcCho. However, careful analysis revealed that the data are best fitted by a-single transport system (r = 0.99) with an affinity of KT = 5.5 ± 0.6 mm and V..s about one-seventh that ofthe synaptic vesicle ghosts (Fig. 3). We further examined the possibility that [3H]AcCho uptake by the synaptic vesicles.may be due to the high-affinity choline carrier in the plasma membrane by measuring the rate of [3H]choline uptake by the various subcellular fractions. In the presence of 100 mm Na+, the initial rate of [3H]choline uptake (measured at 10,uM external [3H]choline) by the synaptosomal membranes (0. 106 ± 0.03 nmol of [3H]choline per mg ofprotein per min; mean ± SD of four experiments) was similar to that in the SV fraction (0.098 ± 0.047 nmol; three experiments). A B range investigated (0..1-10 mm), uptake was.linear for at least 2 min. Analysis of the initial rate of the vesicular [3H]AcCho uptake (determined from the slopes ofthe uptake curves during the first 2 min) as a function of ligand concentration revealed that [3H]AcCho transport follows saturation kinetics which, in the [3H]AcCho concentration range of 0.1 to 10 mm, is best fitted by two transport systems (Fig. 2). Regression analysis revealed that a fit of the data to a two transport systems model has a higher correlation coefficient (r = 0.95) than that of a single transport system (r = 0.70). -In the I2 vesicles, the high-affinity [3H]AcCho transport had an affinity constant KTh of0. 32 ± 0.04mM and a specific activity.of 2.17 ± 0.96 nmol of [3H]AcCho per mg of protein per min (mean ± SD of three determinations). The KTh of the pure (SV) and of the crude (P3) vesicles was similar to that of I2. The specific activity ofthe high-affinity transport of the SV fraction was more than 2-fold higher than that of the 12 fraction, whereas the rate of uptake by the P3 vesicles was lower than that in the I2 vesicles (Fig. 3). Similarly, the specific activity of the low-affinity [3H]AcCho transport was highest in the pure SV fraction (Fig. 3). The affinity constant (KT1) of the low-affinity transport system in the I2 and P3 vesicles was ca. 10 mm (Fig. 2), whereas that in the SV is about 2-fold larger. 56-120- >4-80- <_IlK 2 Ti 40- JrLLI P3 12 SV a2 P3 I2 SV a2 FIG. 3. V.. values of the high-affinity (A) and low-affinity (B) [3H]AcCho transports in the synaptic vesicle and synaptosomal fractions.'the experiments were performed at 25TC. Values presented were obtained by'linear regression analysis of double-reciprocal plots ofthe initial rate ofj5h]accho uptake, determined over an [3H]AcCho concentration range of 0.1 to 10 mm. Values presented are mean + SD of at least three determinations. (For definition of subcellular fractions, see text.) -4-2 0 2 4 6 8 10 1/[3H]AcCho, mm-' FIG. 4. Double-reciprocal plot of the initial rate of [3H]AcCho uptake by Torpedo synaptosomal ghosts. Data are best described by a single straight line with positive ordinate intercept, indicating a single-component system. Each point is the mean of duplicate determinations. The experiments were performed at 250C, and the velocity (V) is expressed as nmol of [3HlAcCho accumulated per mg of protein per min.

Biochemistry: Michaelson and Angel Table 1. by Torpedo synaptic vesicles [3H]AcCho uptake, Addition % of control Control 100 Mg2+ +ATP 57 ± 26 (4) Ca2+ + Mg2+ +ATP 74 ± 34 (2) Me+ + ATP + NaHCO3 125 ± 18 (3) The rates of high-affinity [3H]AcCho transport were measured at [3H]AcCho concentrations of 0.1-0.2 mm. Results are presented as a percentage of the control rate of [3H]AcCho uptake by either 12 or SV vesicles in the absence of additions. Concentrations were: ATP, 5 mm; Mg2+, 4 mm; Ca2+, 0.1 mm; and NaHCO3, 10 mm. Values are means ± SD. The number of determinations is indicated in parentheses. Effect of Mg2e-ATP and NaHCO3 on [3H]AcCho uptake However, in the absence of Na+ there was a 50% inhibition of the synaptosomal [3H]choline uptake whereas the rate of [3H]choline uptake by the pure (SV) vesicles was somewhat increased. The finding that the vesicular [3H]choline transport was not activated by Na+ demonstrates that it is not due to a contaminant of presynaptic plasma membranes and that it may therefore be mediated by the AcCho transport systems. Cholinergic synaptic vesicles contain a Ca2+, Mg2+ ATPase (17, 23, 24) which is outwardly oriented (25). We therefore examined the effect of ATP on the accumulation of [3H]AcCho in the vesicles. The rate of [3H]AcCho transport (measured at external [3H]AcCho of 0.1-0.2 mm) was slightly decreased in the presence of Mg2+ (4 mm) and ATP (3 mm) as well as by Ca2+ (0.1 mm) and similar Mg2e and ATP concentrations (Table 1). By contrast, addition of Mg2+ and ATP in the presence of NaHCO3 (10 mm) resulted in an increase in the rate of [3H]AcCho uptake. Under these conditions, NaHCO3 by itself had no effect. These findings are similar to those recently reported by Koenigsberger and Parsons (26) and suggest that [3H]AcCho accumulation by the vesicles may be affected by ATP and the appropriate ions. DISCUSSION The findings presented in this communication demonstrate that purified cholinergic synaptic vesicles take up [3H]AcCho and that the rate of [3H]AcCho transport into the vesicles follows saturation kinetics which are best fitted by two transport systems having affinity constants in the range of 0.3 mm and 10 mm, respectively. The finding that [3H]AcCho is released from the vesicles by hypoosmotic treatment and by low levels of detergent suggests that the accumulated AcCho is indeed transported into the vesicles. It should be noted that the affinity constants reported here are lower than the estimated concentration of cytoplasmic AcCho in Torpedo nerve endings (-30 mm) (4) and thus are compatible with the concept of AcCho transport in vivo from the presynaptic cytoplasm into the synaptic vesicles. Three criteria were used to demonstrate that AcCho is indeed taken up by synaptic vesicles and not by nonvesicular contaminants such as the plasma membrane high-affinity choline carrier. (i) The specific activity of both the high-affinity and low-affinity AcCho transport systems is higher in the pure synaptic vesicles than in the less-pure vesicle fractions. Furthermore, during the purification procedure the increase in specific rates of [3H]AcCho transport by the vesicular ghosts correlates well with the AcCho content ofthe intact vesicles. The maximal rates of the high-affinity and low-affinity AcCho transport systems in the pure vesicles (SV) (4.7 ± 1.9 and 138 ± 33 nmol of AcCho Proc. Natl. Acad. Sci. USA 78 (1981) 2051 per mg protein per min, respectively) were 3- and 4.5-fold higher than those of the crude (P3) vesicles fractions, whereas the AcCho content of the intact SV vesicles (1010 nmol ofaccho per mg ofmembrane phospholipid) (17) is 5-fold higher than that of P3. (ii) The rate of [3H]AcCho transport in Torpedo synaptosomal membranes is much lower than that in the synaptic vesicles. Kinetic analysis of synaptosomal [3H]AcCho transport revealed low-affinity [3H]AcCho transport with a Vm about one-seventh that of the vesicles. Because the intact synaptosomes contain synaptic vesicles (27, 28), it is possible that [3H]AcCho transport by the synaptosomal preparation is in fact due to the presence of vesicular membranes. (iii) Examination of Na'-dependent [3H]choline uptake revealed no such activity in the pure synaptic vesicles, whereas, consistent with previous reports (20, 21, 29, 30), it does occur in the plasma membrane and to a fairly marked extent. The pharmacological properties ofthe AcCho transport systems described here are not yet known. The finding that the vesciles take up [3H]choline may be interpreted to suggest that [3H]choline transport is mediated by the AcCho transporters. The ratio of the Vm. values of the low- and high-affinity AcCho transport systems in the SV fraction, is ca. 30 and differs from that of the less pure 12 and P3 vesicular fractions (ca. 7 and 18, respectively). This suggests that not all the vesicles contain the two transport systems to the same extent. It should be noted that the purification procedure (17) used in this study was designed for obtaining pure vesicles, taking as a criterion of purity the highest possible concentration ofaccho relative to vesicular protein and membrane phospholipid. Hence, it is possible that the low-affinity transport system is present mainly in the fully mature vesicles and the high-affinity system is more abundant in vesicles at a different stage of their life cycle. Heterogeneity in cholinergic synaptic vesicles has been demonstrated (31, 32) and two populations of synaptic vesicles, differing in size and turnover oftheir AcCho and ATP contents, have been identified and partially purified. It may be ofinterest to examine the extent of the high- and low-affinity AcCho transport systems in these vesicular populations; in addition, it may be possible to use AcCho transport as an assay for the isolation offunctionally different vesicular subpopulations. The molecular properties and the mechanisms of AcCho transport by the high- and low-affinity systems are not known. We have recently found that the AcCho affinities of both systems may be modified experimentally (unpublished data). This finding as well as the variability of the affinity ofthe low-affinity AcCho transport system among various subcellular fractions, suggests that the two systems might in fact represent two different states of a single moiety. Utilizing the known amount of [3H]AcCho accumulated within the vesicles (12.1 nmol of [3H]AcCho per mg of SV protein at 0.2 mm external [3H]AcCho) and the measured intravesicular volume (3.8,ul/mg of SV protein) (14), we can calculate that under these conditions the average intravesicular concentration of [3H]AcCho is about 3 mm-i. e., more than 10- fold higher than the external [3H]AcCho concentration. Assuming that the vesicle ghosts contained no AcCho prior to [3H]AcCho uptake, this finding would suggest that the vesicle ghosts are capable of concentrating AcCho against a concentration gradient. We have recently shown (14) that in the intact vesicles an acidic storage complex may provide part of the driving force for the retention ofaccho within the vesicles. Hence, it is possible that the complex-forming apparatus is retained in the vesicular ghosts and thus, by removing AcCho from the intravesicular aqueous compartment, it provides an apparent driving force for the accumulation ofaccho within the vesicular

2052 Biochemistry: Michaelson and Angel ghosts. The finding that the release of the accumulated [3H]AcCho by hypoosmotic treatment is relatively slow (t,,2 2 min) may also be interpreted in terms of a slow dissociation ofan intravesicular AcCho complex. An alternative explanation is that some AcCho (less than 0.1% of the total AcCho content of the intact vesicles as determined by bioassay) is retained within the vesicular ghosts and that the carrier-mediated [3H]AcCho transport results in the labeling ofthis intravesicular AcCho pool. In the intact vesicles the concentration of membrane-bound AcCho has been estimated at 0.15-0.7 M (1-3), whereas the concentration of free cytoplasmic AcCho is about 1 order of magnitude lower (4, 33). Hence, an energy source must be required for the accumulation ofaccho in the vesicles. The finding, originally reported by Koenigsberger and Parsons (26), that Mg2e and ATP in the presence of HC03- enhance [3H]AcCho uptake, suggests that under the appropriate conditions ATP may provide this energy source for AcCho accumulation. We also found (Table 1) that Mg2+-ATP did not activate, but rather inhibited, transport and that, in the presence of HCO3-, Mg2- ATP-activated AcCho tsport Thus, mere activation of the vesicle-associated Ca2,Mg2 -ATPase (17, 23-25) is not sufficient for driving AcCho uptake. Carpenter et al. (12, 34) have demonstrated that hyperpolarization (interior negative) of Torpedo synaptic vesicles leads to accumulation ofaccho. Hence, it is possible, that, under suitable ionic conditions, ATP may bring about a change in the electrochemical and ionic gradients across the vesicles, which in turn will provide the energy for AcCho accumulation. Further studies in which the intra- and extravesicular ionic contents are systematically altered will help in unraveling the role ofatp, ionic gradients, and the vesicular Ca2+,Mg2+-ATPase in AcCho accumulation. We thank Prof. M. Sokolovsky for his assistance and for helpful discussions. This work was supported in part by a grant from the Muscular Dystrophy Association. 1. Whittaker, V. P. & Sheridan, M. W. (1965) J. Neurochem. 12, 363-372. 2. Breer, H., Morris, S. J. & Whittaker, V. P. (1978) Biochemistry 87, 453-454. 3. Ohsawa, K., Dow, G. H. C., Morris, S. J. & Whittaker, V. P. (1979) Brain Res. 161, 447-457. 4. Israel, M., Dunant, Y. & Manaranche, R. (1979) Prog. Neurobiol. 13, 237-275. Proc. Natl. Acad - Sci USA 78 (1981) 5. Fonnum, F. (1967) Biochem. J. 103, 262-270. 6. Fonnum, F. (1968) Biochem. J. 109, 389-398. 7. Winkler, A. (1977) Neuroscience 2, 657-683. 8. Njus, D. & Radda, G. K. (1978) Biochim. Biophys. Acta 463, 219-244. 9. Kuhar, M. J. & Simon, J. R. (1974)J. Neurochem. 22, 1135-1137. 10. Marchbanks, R. M. (1968) Biochem. J. 106, 87-95. 11. Suszkiw, J. B. (1976)J. Neurochem. 27, 853-857. 12. Carpenter, R. S. & Parsons, S. M. (1978) J. Biol. Chem. 253, 326-329. 13. Michaelson, D. M., Pinchasi, I., Angel, I., Ophir, I. & Rudnik, G. (1979) in Molecular Mechanisms of Biological Recognition, ed. Balaban, M. (Elsevier/North-Holland, Amsterdam), pp. 361-372. 14. Michaelson, D. M. & Angel, I. (1980) Life Sci. 27, 39-44. 15. Michaelson, D. M. & Sokolovsky, M. (1976) Biochem. Biophys. Res. Commun. 73, 25-31. 16. Michaelson, D. M. & Sokolovsky, M. (1978)J. Neurochem. 30, 217-230. 17. Michaelson, D. M. & Ophir, I. (1980) in Neurobiology ofcholinergic and Adrenergic Transmitters, Monographs in Neural Sciences, ed. Cohen M. M. (Karger, Basel), Vol. 7, pp. 19-29. 18. The Edinburgh Staff (1970) Pharmacological Experiments on Isolated Preparations (Livingston, London). 19. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 20. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672. 21. Dowdall, M. J. & Simon, E. J. (1973)J. Neurochem. 21, 969-982. 22. Jope, J. S. (1979) Brain Res. Rev. 1, 313-344. 23. Breer, H., Morris, S. J. & Whittaker, V. P. (1977) Eur. J. Biochem. 80, 313-318. 24. Rothlein, J. E. & Parsons, S. M. (1979) Biochem. Biophys. Res. Commun. 88, 1069-1076. 25. Michaelson, D. M. & Ophir, I. (1980) J. Neurochem. 34, 1483-1490. 26. Koenigsberger, R. & Parsons, S. M. (1980) Biochem. Biophys. Res. Commun. 94, 305-312. 27. Israel, M., Manaranche, R., Mastour-Franchon, P. & Morel, N. (1976) Biochem. J. 160, 113-115. 28. Michaelson, D. M., Bilen, J. & Volsky, D. (1978) Brain Res. 154, 409-414. 29. Yamamura, H. J. & Snyder, S. H. (1973) J. Neurochem. 21, 1355-1374. 30. Wheeler, D. D. (1979)J. Neurochem. 32, 1197-1213. 31. Zimmermann, H. & Denston, C. R. (1977) Neuroscience 2, 715-730. 32. Suszkiw, J. B., Zimmermann, H. & Whittaker, V. P. (1978) J. Neurochem. 30, 1269-1280. 33. Morel, N., Israel, M. & Manaranche, R. (1978) J. Neurochem. 30, 1553-1557. 34. Carpenter, R. S., Koenigsberger, R. & Parsons, S. M. (1980) Biochemistry, 15, 43734379.