Lecithin) Biosynthesis in Rat Brain Neurons (Phospholipid/N-Methylation/Synaptosomes/Catecholamines/Norepinephrine) CAROL E

Home , Dimethylethanolamine, Lecithin, Phosphatidylcholine

Proc. Natl Acad. Sci. USA Vol. 80, pp. 2063-2066, April 1983 Neurobiology

Dopamine stimulation of phosphatidylcholine (lecithin) biosynthesis in rat brain neurons (phospholipid/N-methylation/synaptosomes/catecholamines/norepinephrine) CAROL E. LEPROHON, JAN K. BLUSZTAJN, AND RICHARD J. WURTMAN* Laboratory of Neuroendocrine Regulation, Department of Nutrition and Food Science, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Communicated by Julius Axelrod, October 18, 1982

ABSTRACT Rat brain synaptosomes contain enzymes, phos- nuclei. The supernatant was layered on top of 4 ml of 1.2 M phatidylethanolamine N-methyltransferase(s) (EC 2.1.1.17), that sucrose and centrifuged at 165,000 X -g for 15 min in a Beck- catalyze the methylation of endogenous phosphatidylethanol- man SW 41 Ti rotor. The resulting pellet contained mito- amine to form its mono-, di-, and trimethyl (i.e., phosphatidyl- chondria; the material at the interface of the gradient (con- choline) derivatives. We observe that the activity of these en- taining the synaptosomes and myelin) was diluted to obtain zymes is enhanced when synaptosomes are incubated with cat- approximately 0.32 M sucrose, layered on top of 4 ml of 0.8 echolamines: 0.1 mM dopamine increases incorporation of [3H]- M sucrose, and centrifuged at 165,000 X g for 15 min (17). methyl groups into monomethylphosphatidylethanolamine, di- The material at the interface, containing myelin, and the pel- methylphosphatidylethanolamine, and phosphatidylcholine by let, shown to contain synaptosomes by electron factors of 1.7, 1.3, and 2.1, respectively, and 0.1 mM norepi- microscopy, nephrine increases [3H]methyl incorporation into monomethyl- were collected. In some cases, the postnuclear supernatant phosphatidylethanolamine and dimethylphosphatidylethanol- was centrifuged at 20,000 x g for 20 min to obtain the P2 pel- amine by factors of 1.6 and 2.1, respectively. Stimulation by let (containing synaptosomes, mitochondria, and myelin). The dopamine, which is observed at concentrations as low as 1 ,IM, post-P2 supernatant was centrifuged at 100,000 X g for 60 min is blocked by haloperidol. to obtain a microsomal pellet. PtdEtnMeTase Assay. PtdEtnMeTase activity was assayed Synaptosomal membranes from rat (1-3) and bovine (4, 5) brain by a modification of the method described previously (4). The contain an enzyme, or enzymes, phosphatidylethanolamine N- standard assay medium was changed to 50 mM Tris HCl, pH methyltransferase(s) (PtdEtnMeTase; EC 2.1.1.17) that en- 7.5/5 mM MgCl2/0.2 mM EDTA/0. 1 mM ATP and tissue ables them to synthesize phosphatidylcholine (PtdCho) by the preparation (0.2-0.4 mg of protein) in a final volume of 120 stepwise methylation of phosphatidylethanolamine (PtdEtn), 1.l. The effect on phospholipid [3Himethyl incorporation of using S-adenosylmethionine (AdoMet) as the methyl donor. omitting various components of this medium was examined Part of this newly synthesized PtdCho is hydrolyzed to lib- (see Table 1). Norepinephrine-HCI (Sigma), dopamineHCl erate free choline (1), which might be available for acetyl- (Sigma), haloperidol lactate (5 mg/ml, containing 1.8 mg of choline synthesis; part may be retained as a constituent of cel- methylparaben and 0.2 mg of propylparaben; a gift from lular membranes. PtdEtnMeTases in various tissues have been McNeil, Fort Washington, PA), were added to some tubes shown to be activated by such receptor agonists as catechol- before incubation. Reactions were started by addition of 2.5 amines [rat reticulocytes (6, 7), C6 glioma astrocytoma (8), jCi of S-adenosyl[methyl-3H]methionine (New England Nu- lymphocytes (9)], lectins [lymphocytes (10), mast cells (11)], clear; 13-15 Ci/mmol; 1 Ci = 37 GBq) to a final AdoMet con- immunoglobulins [mast cells (12), leukemic basophils (13)], centration of 1.2-1.6 ,uM. Samples were incubated for 30 min glucagon [hepatocytes (14)]., vasopressin [pituitary extracts (15), at 37TC. Reactions were stopped by addition of 3 ml of chlo- hepatocytes (16)], and angiotensin [hepatocytes (16)]. This roform/methanol/HCI, 100:50:1 (vol/vol). Water-soluble ra- process is rapid and not dependent on cAMP synthesis. Little dioactive. materials were extracted by washing the incubation information has been available on the control of PtdEtn- mixture with two 2-ml portions of 0.75% KCI in 50% meth- MeTase activity within brain neurons. We report here that anol. catecholamine neurotransmitters, especially dopamine, en- The chloroform phase was taken to dryness in a Savant Speed- hance the activities of phosph6lipid methylating enzymes in Vac lyophilizing centrifuge, and the [3H]phospholipids were rat brain synaptosomes. suspended in 50 p.1 of chloroform/methanol, 1:1 (vol/vol). The [3H]phospholipids, together with phosphatidylmono- METHODS methylethanolamine (PtdMeEtn), phosphatidyldimethyletha- Preparation of Subcellular Fractions. Male Sprague-Daw- nolamine (PtdMe2Etn), and PtdCho standards, were applied ley rats (200-300 g; Charles River Breeding Laboratories) were to silica gel TLC plates (Silica gel G; Absorbosil Plus, Applied housed under a 12-hr/12-hr light/dark schedule (lights on, Science Division, State College, PA) and the chromatogram 0900-2100 hr) and had free access to water and food (Charles was developed in chloroform/propionic acid/n-propanol/water, River chow; 22% protein, 0.018% choline). The animals were 2:2:3:1 (vol/vol) (4). Compounds were visualized with di- decapitated at 1000 hr, and the brains were quickly dissected on ice and homogenized in 10 vol of 0.32 M sucrose in a Pot- Abbreviations: PtdCho, phosphatidylcholine; PtdEtn, phosphatidyl- ter-Elvehjem homogenizer; the homogenate was then cen- ethanolamine; PtdEtnMeTase, phosphatidylethanolamine N-methyltransferase; PtdMe2Etn, phosphatidyldimethylethanolamine; Ptd- trifuged at 1,000 x g for 15 min to remove tissue debris and MeEtn, phosphatidylmonomethylethanolamine; AdoMet, S-adenosylmethionine; Bt2cAMP, N6,02-dibutyryladenosine 3',5'-cyclic mono- The publication costs of this article were defrayed in part by page charge phosphate. payment. This article must therefore be hereby marked "advertise- *To whom reprint requests should be addressed at: Room E25-604, ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact. M.I.T., Cambridge, MA 02139. 2063 Downloaded by guest on October 2, 2021 2064 Neurobiology: Leprohon et al. Proc. Natl. Acad. Sci. USA 80 (1983) phenylhexatriene (10 mg/100 ml of petroleum ether; ref. 18) PtdEtnMeTase activity by dopamine (Table 1); tubes were in- and the silica was then scraped into scintillation vials. The newly cubated with or without dopamine, Mg2+, ATP, or phospho- formed [3H]phospholipids were eluted with 1 ml of methanol, lipid substrates. In most situations, dopamine markedly en- 15 ml of Betafluor (National Diagnostics, Somerville, NJ) was hanced the incorporation of [3H]methyl into PtdMeEtn, added to each vial, and the radioactivity of the samples was PtdMe2Etn, and PtdCho. Mg2+ alone increased both basal and determined. Efficiency of assay for 3H was =50%. Recovery dopamine-stimulated [3H]phospholipid accumulation. ATP, of [3H]phospholipids was =70%. added alone or with Mg2+, lacked a consistent effect on [3H]- To ensure that the methods used for assessing [31H]methyl phospholipid synthesis; in some experiments the combination incorporation correctly identified the various [3H]phospholip- of ATP and Mg2+ enhanced basal Ptd[3H]MeEtn and Ptd[3H]- id fractions, we analyzed both the fractions themselves (from Me2Etn syntheses while in others it did not. Addition of phos- basal and catecholamine-stimulated samples) and their [3H]base pholipid substrates tended to enhance the accumulation of all contents after hydrolysis. The [3H]phospholipids were sepa- three [3H]phospholipids, both in the presence and in the ab- rated by two-dimensional TLC using chloroform/methanol/ sence of dopamine. Since the phospholipids added did not acetic acid, 65:30:5 (vol/vol) in the first dimension and chlo- include PtdEtn, the enhanced accumulation of Ptd[3H]MeEtn roform/methanol/2-propanol/0.25% KCI/triethylamine, 30: probably reflected a slowing in its further metabolism. The 9:25:6:18 (vol/vol) (19) in the second dimension; they were addition of phospholipid substrates to the incubation mixture then scraped from the TLC plate and their radioactivities were did not further enhance determined. Although the triethylamine system results in bet- the effect of dopamine on Ptd- ter separation of the [3H]phospholipids than the propionic acid EtnMeTase. In subsequent experiments, we assayed Ptd- system, the dopamine metabolites homovanillic acid and 4-hy- EtnMeTase activity in the presence of Mg2+ and ATP and droxy-3-methoxyphenylethanol were found to comigrate with without PtdMeEtn and PtdMe2Etn to maximize the dopamine PtdMeEtn and PtdMe2Etn, respectively. Hence, triethylamine effect. was not the TLC system of choice for unidirectional chroma- Both dopamine and norepinephrine stimulated PtdEtn tography. Both these dopamine metabolites migrate with the methylation in a dose-dependent fashion (Table 2). The effects solvent front in the propionic acid system. The bases hydrolyzed of dopamine were detectable at concentrations as low as 1 uM from the [3H]phospholipids (by incubation of phospholipid frac- and, in most studies, could be seen with PtdMeEtn, Ptd- tions, separated by routine TLC, in 6 M HCl at 80'C for 1 hr) Me2Etn, and PtdCho; norepinephrine required higher (10 ,uM) were separated by HPLC on a C18 column and then their ra- concentrations for clear effects and generally did not enhance dioactivities were determined. The mobile phase was 30 mM Ptd[3H]Cho accumulation. sodium phosphate, pH 5/2 mM tetramethylammonium chlo- Subcellular Localization of Dopamine Response. Dopa- ride containing sodium octyl sulfate at 4 mg/liter (20); the flow mine markedly stimulated PtdEtnMeTase activity in the P2 rate was 0.8 ml/min. Choline, monomethylethanolamine, and pellet, microsomes, and synaptosomes (Table 3), while Ptd- dimethylethanolamine eluted from the column between 5 and EtnMeTase activities of myelin and mitochondria were only 6 min. Both methods confirmed the identities of the 3H-labeled marginally affected by dopamine. The fact that the percentage products. increase in PtdEtnMeTase activity in synaptosomes was sim- In each experiment, duplicate samples of each assay mix- ilar to that observed in the P2 pellet suggests that the pres- ture were used, and experiments were repeated three or more ence of a neuronal plasma membrane containing dopamine times. receptors is necessary to mediate this effect of dopamine. The stimulation of microsomal PtdEtnMeTase activity by dopa- RESULTS mine may be due either to the presence of dopamine recep- In a preliminary experiment, we examined the effects of var- tors in the endoplasmic reticulum (from which most of the ious components of the assay mixture on the stimulation of microsomes are derived) or to contamination of our microsomal preparation with neuronal plasma membrane. Indeed Table 1. Effects of Mg2+, ATP, and dopamine on synaptosomal such contamination is suggested by the fact that our micro- PtdEtnMeTase activity somal fraction had somewhat higher specific activity of [3H]Methyl incorporated, (fmol/mg of protein)/30 min PtdMeEtn PtdMe2Etn PtdCho Table 2. Dopamine and norepinephrine stimulation of Dopa- Dopa- Dopa- synaptosomal PtdEtnMeTase activity mine mine mine PtdEtnMeTase activity, (fmol of [3H]methyl Addition(s) Control treated Control treated Control treated Concentration, incorporated/mg of protein)/30 min Without phospholipid substrates gm PtdMeEtn PtdMe2Etn PtdCho None 451 661 265 405 127 280 Mg2+ 630 730 287 493 188 449 Dopamine ATP 513 731 273 312 150 233 - 281 372 157 Mg2+/ATP 449 750 191 488 129 306 0.5 318 361 152 With phospholipid substrates 1 406 406 173 None 675 1,044 1,436 1,469 1,045 1,186 10 459 462 230 Mg2+ 789 1,228 1,254 1,427 849 1,007 100 465 469 240 ATP 614 932 1,596 1,720 1,117 1,167 1,000 590 400 207 680 783 Norepinephrine Mg2+/ATP 1,556 1,368 1,883 1,120 0.5 283 275 125 Synaptosomal PtdEtnMeTase activity was estimated by the accu- 1 280 274 121 mulation of [3H]phospholipids. Where used, dopamine was 10,M, ATP 10 304 325 129 was 100 ,uM, and Mg2e was 5 mM. Phospholipid substrates used were 100 440 328 120 PtdMeEtn and PtdMe2Etn (each at 50 jig per incubation mixture); they 1,000 419 360 104 were dispersed ultrasonically. Results are means of duplicate experiments. Results are means of duplicate experiments. Downloaded by guest on October 2, 2021 Neurobiology: Leprohon et al. Proc. Natl. Acad. Sci. USA 80 (1983) 2065 Table 3. Effect of dopamine on PtdEtnMeTase activity in subcellular fractions of rat brain [3H]Methyl incorporated, (fmol/mg of protein)/30 min PtdMeEtn PtdMe2Etn PtdCho Dopa- Dopa- Dopa- mine S mine % mine % Fraction Control treated change Control treated change Control treated change P2 pellet 351 810 131 477 900 88 321 710 121 Mitochondria 231 268 16 491 530 8 369 441 19 Myelin 88 120 36 466 484 4 281 375 33 Synaptosomes 312 694 122 338 574 70 340 699 106 Microsomes 781 1,181 44 617 985 60 525 979 86 Dopamine was used at 10 ,LM. The amount of protein in each fraction was 0.21-0.25 mg per incubation. Results are means of duplicate experiments. PtdEtnMeTase than the synaptosomes-a finding at variance peridol (Table 4). Since norepinephrine is less potent than with previously reported data (3, 4). dopamine, further studies are needed to determine whether Dopamine Receptor Involvement in Stimulation of Phos- norepinephrine is acting at its own receptor or at a dopami- pholipid Methylation. To establish that dopamine was acting nergic receptor. through receptors to stimulate PtdEtn methylation, its effects Several types of dopamine receptors have tentatively been were examined in the presence of haloperidol. Haloperidol identified in the mammalian brain (21, 26, 27), which may be alone had a slight inhibitory effect on PtdEtn methylation; partially distinguished by their affinities for dopamine or halo- however, in the presence of 10 utM dopamine, haloperidol peridol. The receptor mediating the effect of dopamine on caused a dose-dependent decrease in the stimulation of PtdEtn synaptosomal PtdEtnMeTase activity shares properties with methylation (Table 4), suppressing the accumulation of all three the D1 receptors identified by such binding studies. These [3H]phospholipid fractions. This decrease in [3H]phospholipid receptors display micromolar dissociation constants for do- accumulation was proportionately greater in the dopamine- pamine and haloperidol (21, 26). [They also tend to be as- stimulated group than in the control group. sociated with the adenylate cyclase activation classification ac- cording to Kebabian and Calne (21) or Seeman (26).] We are DISCUSSION at present unable to state whether dopamine is acting on presynaptic or postsynaptic receptors, because our synaptosomal These data show that the catecholamines norepinephrine and preparations contain both presynaptic and postsynaptic mem- dopamine can stimulate PtdEtn methylation in rat brain syn- branes, as visualized by electron microscopy. aptosomes, just as they have previously been shown (6-9) to The activation of PtdEtnMeTase by catecholamines may be act in other tissues. Increased incorporation of [3H]methyl mediated by the interaction of a catecholamine receptor and groups into PtdMeEtn and PtdMe2Etn was observed after ad- the membrane-bound (28) methyltransferase, perhaps chang- dition of 1 ,uM dopamine or 10 ,uM norepinephrine (Table 2). ing the affinity of PtdEtnMeTase for one of its substrates. Al- Dopamine also stimulated the accumulation of [3H]methyl-la- ternatively, receptor activation could increase the permeabil- beled PtdCho, while norepinephrine concentrations as high as ity of the synaptosomal plasma membrane for AdoMet, thus 100 ,uM failed to affect [3H]PtdCho accumulation. The doses providing more substrate for the active sites on the methyl- of dopamine and norepinephrine needed to stimulate Ptd- transferase-synthesizing PtdMeEtn (28). However, neither os- EtnMeTase in our studies were thus similar to those required motically shocking synaptosomes nor prior incubation of them for the in vitro activation of adenylate cyclase in rat brain, with [3H]AdoMet for 16 hr at 0°C altered the responses of dopamine exhibiting greater potency than norepinephrine (22, their PtdEtnMeTases to catecholamine (data not shown). 25). The effect of dopamine on phospholipid methylation ap- The [3H]methyl accumulation was observed mainly in the pears to be receptor mediated, because it is localized in syn- PtdMeEtn and PtdMe2Etn fractions. The fact that the accu- aptosomes (Table 3) and could be-partially blocked by halo- mulation of [3H]methyl-labeled PtdCho was lower may be due to the relatively low AdoMet concentrations present in our Table 4. Effect of haloperidol on basal and dopamine-stimulated assay mixture. In studies by Crews et al. (3), high AdoMet synaptosomal PtdEtnMeTase activity concentrations were required for PtdCho to become the major PtdEtnMeTase activity, (fmol of [3H]methyl methylated phospholipid product. Alternatively, large amounts of [3H]PtdCho might have been formed and then rapidly de- Haloperidol, incorporated/mg of protein)/30 min graded to [3H]choline. We have previously observed that as AuM PtdMeEtn PtdMe2Etn PtdCho much as 30% of the [3H]methyl groups incorporated into syn- Basal aptosomal PtdCho after 30 min of incubation are present in - 347 194 165 the assay mixture as free choline (1). Thus, it is possible that 1 345 184 154 the rates of endogenous brain choline synthesis may also be 10 231 154 135 affected by dopamine. If this synthesis is indeed enhanced by 100 250 136 133 dopamine, this might explain the development of one of the Dopamine stimulated side effects of therapy with haloperidol, tardive dyskinesia. - 620 429 459 Prolonged exposure to dopamine receptor blockers such as 1 565 453 440 haloperidol could inhibit PtdEtnMeTase activity in the basal 10 421 328 314 ganglia, depressing the de novo synthesis of choline and thus 100 346 216 219 diminishing acetylcholine synthesis (29). Indeed, it has been Dopamine was used at 10 ,uM. Results are means of duplicate ex- shown that therapy with choline or lecithin can ameliorate the periments. abnormal movements of tardive dyskinesia (30, 31). Downloaded by guest on October 2, 2021 2066 Neurobiology: Leprohon et al. Proc. Natl. Acad. Sci. USA, 80 (1983) We thank R. W Von Borstel for his participation in some of these 15. Prasad, C. & Edwards, R. M. (1981) Biochem. Biophys. Res. studies. These studies were supported in part by grants from the Na- Commun. 103, 559-564. tional Institute of Mental Health (MH-28783), the National Aeronautics 16. Alemany, S., Varela, I. &. Mato, J. M. (1981) FEBS Lett. 135, 111- and Space Administration (NGR-22-009-627), and the Center for Brain 114. Science and Metabolism Charitable Trust. C.E. L. holds a fellowship 17. Dodd, P. R., Hardy, J. A., Oakley, A. E., Edwardson, J. A., Perry, from the Canadian Medical Research Council. E. K. & Delaunoy, J. P. (1981) Brain Res. 226, 107-118. 18. Hyslop, P. A. & York, D. A. (1980) AnaL Biochem. 101, 75-77. 1. Blusztajn, J. K. & Wurtman, R. J. (1981) Nature (London) 290, 19. Touchstone, J. C., Chen, J. C. & Beaver, K. M. (1980) Lipids 15, 417-418. 61-62. 2. Mozzi, R. & Porcellati, G. (1979) FEBS Lett. 100, 363-366. 20. Potter, P. E., Meek, J. L. & Neff, N. H. (1982) Soc. Neurosci. 8, 3. Crews, F. T., Hirata, F. & Axelrod, J. (1980)J. Neurochem. 34, 143 (abstr). 1494-1498. 21. Kebabian, J. W. & Calne, D. B. (1978) Nature (London) 277, 93- 4. Blusztajn, J. K., Zeisel, S. H. & Wurtman, R. J. (1979) Brain Res. 96. 179, 319-327. 22. Sarau, H. M., Hetzel, J. A., Setler, P. E. & Foley, J. J. (1979) in 5. Percy, A. K., Moore, J. F. & Waechter, C. J. (1982) J. Neuro- Catecholamines: Basic and Clinical Frontiers, eds. Usdin, E., chem. 38, 1404-1412. Kopin, I. J. & Barchas, J. (Pergamon, New York), Vol. 1, pp. 517- 6. Hirata, F., Strittmatter, W J. & Axelrod, J. (1979) Proc. NatL Acad. 519. Sci. USA 76, 368-372. 23. Pritchard, P. H., Pelech, S. L. & Vance, D. E. (1981) Biochim. 7. Bilezikiam, J. P., Spiegel, A. M., Giammon, D. E. & Aurbach, Biophys. Acta 666, 301-306. G. D. (1978) Mol. Pharmacol. 13, 786-795. 24. Seamon, K. B. & Daly, J. W. (1982) J. Cyclic Nucleotide Res. 7, 8. Strittmatter, W. J., Hirata, F., Axelrod, J., Mallorga, P., Tall- 201-224. man, J. F. & Henneberry, R. C. (1979) Nature (London) 282, 857- 25. Daly, J. W. & Partington, C. R. (1979) in Catecholamines: Basic 859. and Clinical Frontiers, eds. Usdin, E., Kopin, I. J. & Barchas, J. 9. LeFur, G., Phan, T., Canton, T., Tur, C. & Uzan, A. (1981) Life (Pergamon, New York), Vol. 1, pp. 520-522. Sci. 29, 2737-2749. 26. Seeman, P. (1980) Pharmacol. Rev. 32, 229-313. 10. Hirata, F., Toyoshima, S., Axelrod, J. & Waxdal, M. J. (1980) Proc. 27. Creese, I. (1982) Trends Neurosci. 5, 40-43. NatL Acad. Sci. USA 77, 862-865. 28. Crews, F. T., Hirata, F. & Axelrod, J. (1980) Neurochem. Res. 5, 11. Hirata, F., Axelrod, J. & Crews, F. T. (1979) Proc. Nati Acad. Sci. 983-991. USA 76, 4813-4816. 29. Wurtman, R. J., Hefti, F. & Melamed, E. (1980) Pharmacol. Rev. 12. Ishizaka, T., Hirata, F., Sterk, A. R., Ishizaka, K. & Axelrod, J. 32, 315-335. (1981) Proc. NatL Acad. Sci. USA 78, 6812-6816. 30. Growdon, J. H., Hirsch, M. J., Wurtman, R. J. & Wiener, W. 13. Ishizaka, T., Hirata, F., Ishizaka, K. & Axelrod, J. (1980) Proc. (1977) N. Engi J. Med. 297, 524-527. NatL Acad. Sci. USA 77, 1903-1906. 31. Davis, K. L., Berger, P. A. & Hollister, L. E. (1975) N. Engl. J. 14. Castano, J. G., Alemany, S., Nicto, A. & Mato, J. M. (1980)J. Biol Med. 293, 152. Chem. 255, 9041-9043. Downloaded by guest on October 2, 2021