- Email: [email protected]
Stimulation of 3H-norepinephrine release by trifluoperazine from rat pineal glands
Life Sciences, Vol. 38, pp. 1393-1397 Printed in the U.S.A.
Pergamon Press
STIMULATION OF 3H-NOREPINEPHRINE RELEASEBY TRIFLUOPERAZINE FROMRAT PINEAL GLANDS Karin A. Yurko and Linda F. Quenzer University of Connecticut Health Center Department of Psychiatry Farmington, Connecticut 06032 (Received in final form February 6, 1986)
summar~ T r i f l u o p e r a z i n e (5-200~M) stimulated the release of 3H-NE from isolated whole pineal glands in a dose dependent manner. T r i f l u o peraztne-induced rel ease was not dependent on extracel l u l ar Ca++, whereas 60mMK+-evoked release was attenuated in the presence of EGTA and zero Ca++ Krebs. 60mMK+ and 50~M trifluoperazine produced an additive effect on 3H-NE release. Clonidine (5~M) significantly reduced trifluoperazine-induced release by approximately 50% i n the presence of Ca++, and in its absence, clonidine significantly attenuated the trifluoperazine response by 42%. Thus trifluoperazine may be acting upon the ~2 receptor or intracellular stores of Ca++. These intracellular interactions remain for further study. The importance of Ca2+ to the neurotransmitter release process has been well documented jn several tissues (1). I t is widely accepted that calmodulln, a heaJ; stable CaZ+ binding protein, mediates many of the intracellular effects of CaZ+ (2). Someof the evidence for a significant role for calmodulin in neurotransmitter release has been derived from studies showing inhibition of neurotransmitter release by antagonism of calmodulin with trlfluoperazine (TFP) (3). In addition to being a calmodulin inhibitor, trifluoperazine has also been shown to be a potent inhibitor of protein kinase C which has been demonstrated to evoke such cellular responses as neurotransmltter release (4). However, others have shown a TFP-lnduced increase in release either directly as in the case of catecholamine secretion by chromaffin cells (5,6) or indirectly as in the increase in MEPPfrequency following TFP at the frog neuromuscular junction (7). Our experiments have examined the effect of TFP on the release of ~H-norepinephrine from adrenergic neurons in the intact rat pineal gland. The gland is readily accessible, heavily innervated by catecholamine-contalnlng nerves from the superior cervical ganglion and has been prevlously used to study transmitter release and i t s regulation by presynaptic mechanisms (8,9). Methods Pineal glands were removed from male Sprague-Dawley rats (125-200g) kept in constant light for 24 hrs before sacrifice. For release experiments we used a method described earlier (g). Briefly, the pineals are incubated at 37~ for 15 min. in oxygenated Krebs Ringer solution containing: 120mMNaCl, 4.7~ KCl, 25mM NaHC03, 1.2mM KH2P04, 2.4mM MgSO4, 2.5mM CaC12, l OmMglucose and 60~M ascorbate~ Pairs of glands were incubated in Krebs Ringer containing 2 ~Ci (O.13pM) H-NE (New England Nuclear) for 30 mln. The glands were then washed 0024-3205/86 $3.00 + .00 Copyright (c) 1986 Pergamon Press Ltd.
1394
TFP-induced NE Release
Vol. 38, No. 15, 1986
four times with Krebs Ringer and incubated at 37° in isotope-free Krebs Ringer for consecutive lO min washouts until stable isotope efflux was attained (70 min). OH-NErelease was achieved by the addition of TFP ( g i f t , Smith, Kline and French, Philadelphia, PA) for 5 rain or elevated K+ (60ram KCI with an equivalent reduction in NaCl) for I rain. All blocking drugs and modified Krebs Ringer solutions (containing no CaCl2, ImM EGTA, and 4.gram MgSO4) were applied 20 min before testing with the stimulating agents. At the end of the experiment, tissues were dissolved with 0.5 ml tissue solubilizer (Protosol, New England Nuclear) and counted in lO ml ACS scintillation fluid (Amersham) using a Tracor scintillation spectrometer. Data are expressed as fractional release (FR), i . e . , the ratio of the amount of isotope released into the medium for l min over the total isotope present in pineal glands. Drug effects were assessed by comparing FR before and after drug treatment. In the 3H-NE uptake experiments, individual pineals were incubated in oxygenated Krebs Ringer (37,C) in the absence or presence of either TFP (50pM) or imipramine (lO~M) for 5 min. ~he solutions were replaced by drug free Krebs Ringer containing 2 ~Ci (0.13gM) °H-NE and incubated for 30 min. To terminate the reaction, the bath was aspirated and an aliquot was counted. The pineals were rinsed once briefly with Krebs Ringer and the tissue solubilized and treated as described above. Data are expressed as the fraction of isotope taken up into the p~neal compared to the total isotope available in the i n i t i a l bath (i.e. ~-NE in pineal plus that remaining in the bath). Paired Student-t tests were used for evaluating release experiments in which each pair of pineals was exposed to sequential drug treatments. Uptake experiments were evaluated with the unpaired Student-t tests. Results Following seven lO-min incubations in Krebs Ringer, the resting efflux of isotope into the bath was .0034 * .0003 (n=12) expressed as FR. Addition of 50pM TFP to the Krebs Ringer bathing medium increased the fraction of total tissue isotope released into the bath 3.2 fold to .0112 * .0009 (n=12; p<.01). Under the same conditions, exposure to elevated K+ (60mM) for l min increased the FR 6.7 fold to .0221 * .0035 (n=9; p<.O05). Two consecutive 5 min exposures to TFP (50pM) 30 min apart produced FR of 0.0131 * .0013 and 0.0124 * .0013 (n=15, N.S.). Thus the effect of TFP is reproducible within one experiment at 50uM. Although the efflux during the f i r s t lO min collection period following TFP was greater than baseline, the efflux between TFP pulses returned to normal within 30 min. Thus the effect of TFP was reversible upon washing the tissue. Fig. l shows the dose dependent nature of TFP action on FR in intact ~Hineal glands. The lowest concentration (IgM) of TFP significantly inhibited -NE release into the bath while higher concentrations (5-200pM) produced an increase in release. At 50gM TFP, aH-NE release was 3 times greater than baseline conditions (passive efflux) while higher TFP concentrations produced a 5.5-fold increase in aH-NE found in the bathing medium. WhenTFP (50uM) was added to a bath containing 60~ KCI the stimulatory effects of the two conditions on release were nearly additive (Table I ) . Table I also shows that TFP-i~duces 3H-NE release to the same extent in both normal Krebs Ringer and in Ca¢-free Krebs Ringer containing ImM EGTA. In contrast, as expected, removal of extracellular Ca2+ inhibited K+-induced release. I t has been proposed that presynaptic ~2 receptors in pineal inhibit NE release by an autofeedback mechanism (10). We therefore tested the action of the =2 agonist, clonidine on TFP-induced release. Clonidine (5 pM) signi-
Vol. 38, No, 15, 1986
TFP-induced NE Release
1395
30 A
O O
o qp-
25
x uJ
(n
,~/,, (7)
20
//.~1111
i11
,,-i e, .J <[ Z
o
IU ,<
,~ u.
15
/ 10 i
/
/
/ /
112)
/
(1
5
i 18) 1
/ 2)~ ~(10) / J
. . . . . . . .
i
,
,
......
10
j
,
,
,
100 TFP
.....
m
1000
(pM)
FIG. 1. Dose dependent change in fractional release of 3H-NE following treatment with trifluoperazine (ITP) for 5 min. The arrow shows the mean control (passive efflux) for 59 experiments. Significance (*) was determined with a Student-t test for matched subjects at each drug concentration.
TABLE I KREBS RINGER Passive Efflux (8) 60ram K+ (4) 50gM TFP (8) KCl + 1TP (5)
0.0033 * 0.0286 * 0.0118 * 0.0342 *
0.0002 0.0046 0.0017 0.0050
O-Ca+2 KREBS + EGTA 0.0034 * 0.0003 0.0055 * 0.0005 0.0113 * 0.0014
p NS p<.02 NS
Fractional release of 3H-NE following treatment with elevated rK+ trifluoperazine (ll~P), or both agents in normal Krebs and low ~ 2 + Krebs. Preincubatlon in modified Krebs was for 20 mln. before the addition of the test agent. Significance was determined with a Student-t test for matched subjects. Number of experiments is in paren theses.
1396
TFP-inauced NE Release
Vol. 38, No. 15, 1986
ficantly inhiblted~the response to TFP by 50% (n=11, p<.O05). A 30-40% reduction in KT-evoked °H-NE release following c]onidine in the same preparation has been reported (9). When tested in O-Caz+ (+EGTA) Krebs Ringer, the addition of clonidine (5 ~M) again significantly (n=5; p<.02) attenuated the response to TFP by42%. To determine whether the increase in 3H-NE collected after the addition of TFP was due to inhibition of catecholamine uptake, we compared the effect of TFP and a known uptake inhibitor (imlpramlne) on 3H-NE uptake in pineal glands. Control's mean % uptake was 2.21% (SE=.23), imipramine, 1.07% (SE=.IO) and trifluoperazine, 1.65% (SE=.18). Imipramine (lO ~M) significantly nhibited uptake by 52% (p<.01); TFP (50 ~M) produced a 26% inhibition of H-NE uptake, but this was not s t a t i s t i c a l l y significant (n=§). Further, imipramine, alone, did no~ significantly increase efflux of ~H-NE. Thus, the apparent increase in OH-NE overflow by TFP is not due to uptake inhibition.
i
Discussion Unlike TFP-induced catechol amine rel ease from chromaffin cel I s (5,6), transmitter release from pineal is apparently not due to derangement of cellular ultrastructure or nonspeclfic detergent-like action since release occurred in reproducible fractions and the effect was reversible. In ~ddition, the ~2-agonist clonidine was capable of inhibiting TFP-induced H-NE release by 50% in either the presence or absence of Ca~+. The inhibition by clonidine on TFP-induced release could be simply explained as a direct a f f i n i t y by TFP for the ~2 receptor. Significant inhibition of ~-receptor binding by TFP has been demonstrated in rat brain homogenate with a Ki of 46 nM (11). The TFP-induced release in some respects resembles that of the ~2-antagonist , yohimbine, which increases depolarization-induced efflux of ~H-NE. However, unlike TFP, yohimbine has no apparent direct action on release in the absence of elevated K+ (9). Hence, the probability that TFP is blocking a tonic inhibitory action of norepinephrine to enhance transmitter efflux is less likely. A more complex explanation for TFP's effect on 3H-NE release may be that i t is acting intracellularly to alter stores of Ca++. I t is hypothesized that TFP may be modifying Ca++ binding sites directly to cause an increase TT ++ in intracellular Ca . This would enable Cai to affect other intracellular processes, such as stimulus-secretion coupling for neurotransmitter release. Evidence for a modification of such Ca++ stores has been shown with electron microscopy of pineal nerves (12). Ca++ a f f i n i t y for synaptlc vesicles was significantly decreased or abolished by electrical stimulation, thereby resulting in a loss of Ca++ ions "trapped" by synaptic vesicles and increasing cytoplasmic Ca++ thus contributing to the release of neurotransmitter. This h~othesis would account for TFP's effect in the absence of extracellular Ca and may also provide an explanation for approximately 50% ++ release in the presence of clonidine with or without extracellular Ca . I t is thought that activation of presynaptic a-adrenoceptors decreases the a v a i l a b i l i t y of extrace]lular Ca++ for coupllng (13). Our results with clonidine and TFP, in part, support this idea. Activation of intracellular stores of Ca++ may account for the release of norepinephrine not blocked by clonidine and may also account for similar effects when extracellular Ca+* is removed from the medium. The mechanism of TFP's effect on intracellular processes such as Ca++ stores remains for further study.
Vol. 38, No. 15, 1986
TFP-induced NE Release
Acknowl edBments Research supported by grant NS07540-16 from the National I n s t i t u t e of Neurological and Communicative Disorders and Stroke, National Institutes of Health. References I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.
R.P. RUBIN, Pharmacol. Rev. 22 328-428 (1972). W. CHEUNG, Sci. 207 19-27 (1~rl~O). R.J. DELORENZO,1~e'~eration Proc. 41 2265-2272 (1982). Y. NISHIZUKA, Science 225 1365-137I~ (1984). J.C. BROOKSand S. TREI~, Biochem. Pharmacol. 32 371-373 (1983). J.C. BROOKSand S. TREML, Life Sci. 34 669-674"[1983). S. PUBLICOVER, Naunyn-SchmiedebergAr-~'h. Pharm. 322 83-88 (1983). F. PELAYO, M.L. OUBOCOVICHand S.Z. Langer, Natur-~-(Lond) 274 76-78 ( 1978). L.F. QUENZERand R.L. VOLLE, J. Autonom. Ner. Sys. 8 161-164 (1983). S.Z. LANGER, Pharmacol. Rev. 32 337-362 (1981). S.J. PEROUTKA, D.C. U'PRICHAR~T D.A. GREENBERGand S.H. SNYDER, Neuropharmacol. 16 549-556 (1977). A. PELLEGRINODE-~ALDI and J. CORAZZA, Cell Tissue Res. 216 625-634 (1981). K. STARKE and T. ENDO, Gen. Pharmacol. 7 307-312 (1976).
1397
Pergamon Press
STIMULATION OF 3H-NOREPINEPHRINE RELEASEBY TRIFLUOPERAZINE FROMRAT PINEAL GLANDS Karin A. Yurko and Linda F. Quenzer University of Connecticut Health Center Department of Psychiatry Farmington, Connecticut 06032 (Received in final form February 6, 1986)
summar~ T r i f l u o p e r a z i n e (5-200~M) stimulated the release of 3H-NE from isolated whole pineal glands in a dose dependent manner. T r i f l u o peraztne-induced rel ease was not dependent on extracel l u l ar Ca++, whereas 60mMK+-evoked release was attenuated in the presence of EGTA and zero Ca++ Krebs. 60mMK+ and 50~M trifluoperazine produced an additive effect on 3H-NE release. Clonidine (5~M) significantly reduced trifluoperazine-induced release by approximately 50% i n the presence of Ca++, and in its absence, clonidine significantly attenuated the trifluoperazine response by 42%. Thus trifluoperazine may be acting upon the ~2 receptor or intracellular stores of Ca++. These intracellular interactions remain for further study. The importance of Ca2+ to the neurotransmitter release process has been well documented jn several tissues (1). I t is widely accepted that calmodulln, a heaJ; stable CaZ+ binding protein, mediates many of the intracellular effects of CaZ+ (2). Someof the evidence for a significant role for calmodulin in neurotransmitter release has been derived from studies showing inhibition of neurotransmitter release by antagonism of calmodulin with trlfluoperazine (TFP) (3). In addition to being a calmodulin inhibitor, trifluoperazine has also been shown to be a potent inhibitor of protein kinase C which has been demonstrated to evoke such cellular responses as neurotransmltter release (4). However, others have shown a TFP-lnduced increase in release either directly as in the case of catecholamine secretion by chromaffin cells (5,6) or indirectly as in the increase in MEPPfrequency following TFP at the frog neuromuscular junction (7). Our experiments have examined the effect of TFP on the release of ~H-norepinephrine from adrenergic neurons in the intact rat pineal gland. The gland is readily accessible, heavily innervated by catecholamine-contalnlng nerves from the superior cervical ganglion and has been prevlously used to study transmitter release and i t s regulation by presynaptic mechanisms (8,9). Methods Pineal glands were removed from male Sprague-Dawley rats (125-200g) kept in constant light for 24 hrs before sacrifice. For release experiments we used a method described earlier (g). Briefly, the pineals are incubated at 37~ for 15 min. in oxygenated Krebs Ringer solution containing: 120mMNaCl, 4.7~ KCl, 25mM NaHC03, 1.2mM KH2P04, 2.4mM MgSO4, 2.5mM CaC12, l OmMglucose and 60~M ascorbate~ Pairs of glands were incubated in Krebs Ringer containing 2 ~Ci (O.13pM) H-NE (New England Nuclear) for 30 mln. The glands were then washed 0024-3205/86 $3.00 + .00 Copyright (c) 1986 Pergamon Press Ltd.
1394
TFP-induced NE Release
Vol. 38, No. 15, 1986
four times with Krebs Ringer and incubated at 37° in isotope-free Krebs Ringer for consecutive lO min washouts until stable isotope efflux was attained (70 min). OH-NErelease was achieved by the addition of TFP ( g i f t , Smith, Kline and French, Philadelphia, PA) for 5 rain or elevated K+ (60ram KCI with an equivalent reduction in NaCl) for I rain. All blocking drugs and modified Krebs Ringer solutions (containing no CaCl2, ImM EGTA, and 4.gram MgSO4) were applied 20 min before testing with the stimulating agents. At the end of the experiment, tissues were dissolved with 0.5 ml tissue solubilizer (Protosol, New England Nuclear) and counted in lO ml ACS scintillation fluid (Amersham) using a Tracor scintillation spectrometer. Data are expressed as fractional release (FR), i . e . , the ratio of the amount of isotope released into the medium for l min over the total isotope present in pineal glands. Drug effects were assessed by comparing FR before and after drug treatment. In the 3H-NE uptake experiments, individual pineals were incubated in oxygenated Krebs Ringer (37,C) in the absence or presence of either TFP (50pM) or imipramine (lO~M) for 5 min. ~he solutions were replaced by drug free Krebs Ringer containing 2 ~Ci (0.13gM) °H-NE and incubated for 30 min. To terminate the reaction, the bath was aspirated and an aliquot was counted. The pineals were rinsed once briefly with Krebs Ringer and the tissue solubilized and treated as described above. Data are expressed as the fraction of isotope taken up into the p~neal compared to the total isotope available in the i n i t i a l bath (i.e. ~-NE in pineal plus that remaining in the bath). Paired Student-t tests were used for evaluating release experiments in which each pair of pineals was exposed to sequential drug treatments. Uptake experiments were evaluated with the unpaired Student-t tests. Results Following seven lO-min incubations in Krebs Ringer, the resting efflux of isotope into the bath was .0034 * .0003 (n=12) expressed as FR. Addition of 50pM TFP to the Krebs Ringer bathing medium increased the fraction of total tissue isotope released into the bath 3.2 fold to .0112 * .0009 (n=12; p<.01). Under the same conditions, exposure to elevated K+ (60mM) for l min increased the FR 6.7 fold to .0221 * .0035 (n=9; p<.O05). Two consecutive 5 min exposures to TFP (50pM) 30 min apart produced FR of 0.0131 * .0013 and 0.0124 * .0013 (n=15, N.S.). Thus the effect of TFP is reproducible within one experiment at 50uM. Although the efflux during the f i r s t lO min collection period following TFP was greater than baseline, the efflux between TFP pulses returned to normal within 30 min. Thus the effect of TFP was reversible upon washing the tissue. Fig. l shows the dose dependent nature of TFP action on FR in intact ~Hineal glands. The lowest concentration (IgM) of TFP significantly inhibited -NE release into the bath while higher concentrations (5-200pM) produced an increase in release. At 50gM TFP, aH-NE release was 3 times greater than baseline conditions (passive efflux) while higher TFP concentrations produced a 5.5-fold increase in aH-NE found in the bathing medium. WhenTFP (50uM) was added to a bath containing 60~ KCI the stimulatory effects of the two conditions on release were nearly additive (Table I ) . Table I also shows that TFP-i~duces 3H-NE release to the same extent in both normal Krebs Ringer and in Ca¢-free Krebs Ringer containing ImM EGTA. In contrast, as expected, removal of extracellular Ca2+ inhibited K+-induced release. I t has been proposed that presynaptic ~2 receptors in pineal inhibit NE release by an autofeedback mechanism (10). We therefore tested the action of the =2 agonist, clonidine on TFP-induced release. Clonidine (5 pM) signi-
Vol. 38, No, 15, 1986
TFP-induced NE Release
1395
30 A
O O
o qp-
25
x uJ
(n
,~/,, (7)
20
//.~1111
i11
,,-i e, .J <[ Z
o
IU ,<
,~ u.
15
/ 10 i
/
/
/ /
112)
/
(1
5
i 18) 1
/ 2)~ ~(10) / J
. . . . . . . .
i
,
,
......
10
j
,
,
,
100 TFP
.....
m
1000
(pM)
FIG. 1. Dose dependent change in fractional release of 3H-NE following treatment with trifluoperazine (ITP) for 5 min. The arrow shows the mean control (passive efflux) for 59 experiments. Significance (*) was determined with a Student-t test for matched subjects at each drug concentration.
TABLE I KREBS RINGER Passive Efflux (8) 60ram K+ (4) 50gM TFP (8) KCl + 1TP (5)
0.0033 * 0.0286 * 0.0118 * 0.0342 *
0.0002 0.0046 0.0017 0.0050
O-Ca+2 KREBS + EGTA 0.0034 * 0.0003 0.0055 * 0.0005 0.0113 * 0.0014
p NS p<.02 NS
Fractional release of 3H-NE following treatment with elevated rK+ trifluoperazine (ll~P), or both agents in normal Krebs and low ~ 2 + Krebs. Preincubatlon in modified Krebs was for 20 mln. before the addition of the test agent. Significance was determined with a Student-t test for matched subjects. Number of experiments is in paren theses.
1396
TFP-inauced NE Release
Vol. 38, No. 15, 1986
ficantly inhiblted~the response to TFP by 50% (n=11, p<.O05). A 30-40% reduction in KT-evoked °H-NE release following c]onidine in the same preparation has been reported (9). When tested in O-Caz+ (+EGTA) Krebs Ringer, the addition of clonidine (5 ~M) again significantly (n=5; p<.02) attenuated the response to TFP by42%. To determine whether the increase in 3H-NE collected after the addition of TFP was due to inhibition of catecholamine uptake, we compared the effect of TFP and a known uptake inhibitor (imlpramlne) on 3H-NE uptake in pineal glands. Control's mean % uptake was 2.21% (SE=.23), imipramine, 1.07% (SE=.IO) and trifluoperazine, 1.65% (SE=.18). Imipramine (lO ~M) significantly nhibited uptake by 52% (p<.01); TFP (50 ~M) produced a 26% inhibition of H-NE uptake, but this was not s t a t i s t i c a l l y significant (n=§). Further, imipramine, alone, did no~ significantly increase efflux of ~H-NE. Thus, the apparent increase in OH-NE overflow by TFP is not due to uptake inhibition.
i
Discussion Unlike TFP-induced catechol amine rel ease from chromaffin cel I s (5,6), transmitter release from pineal is apparently not due to derangement of cellular ultrastructure or nonspeclfic detergent-like action since release occurred in reproducible fractions and the effect was reversible. In ~ddition, the ~2-agonist clonidine was capable of inhibiting TFP-induced H-NE release by 50% in either the presence or absence of Ca~+. The inhibition by clonidine on TFP-induced release could be simply explained as a direct a f f i n i t y by TFP for the ~2 receptor. Significant inhibition of ~-receptor binding by TFP has been demonstrated in rat brain homogenate with a Ki of 46 nM (11). The TFP-induced release in some respects resembles that of the ~2-antagonist , yohimbine, which increases depolarization-induced efflux of ~H-NE. However, unlike TFP, yohimbine has no apparent direct action on release in the absence of elevated K+ (9). Hence, the probability that TFP is blocking a tonic inhibitory action of norepinephrine to enhance transmitter efflux is less likely. A more complex explanation for TFP's effect on 3H-NE release may be that i t is acting intracellularly to alter stores of Ca++. I t is hypothesized that TFP may be modifying Ca++ binding sites directly to cause an increase TT ++ in intracellular Ca . This would enable Cai to affect other intracellular processes, such as stimulus-secretion coupling for neurotransmitter release. Evidence for a modification of such Ca++ stores has been shown with electron microscopy of pineal nerves (12). Ca++ a f f i n i t y for synaptlc vesicles was significantly decreased or abolished by electrical stimulation, thereby resulting in a loss of Ca++ ions "trapped" by synaptic vesicles and increasing cytoplasmic Ca++ thus contributing to the release of neurotransmitter. This h~othesis would account for TFP's effect in the absence of extracellular Ca and may also provide an explanation for approximately 50% ++ release in the presence of clonidine with or without extracellular Ca . I t is thought that activation of presynaptic a-adrenoceptors decreases the a v a i l a b i l i t y of extrace]lular Ca++ for coupllng (13). Our results with clonidine and TFP, in part, support this idea. Activation of intracellular stores of Ca++ may account for the release of norepinephrine not blocked by clonidine and may also account for similar effects when extracellular Ca+* is removed from the medium. The mechanism of TFP's effect on intracellular processes such as Ca++ stores remains for further study.
Vol. 38, No. 15, 1986
TFP-induced NE Release
Acknowl edBments Research supported by grant NS07540-16 from the National I n s t i t u t e of Neurological and Communicative Disorders and Stroke, National Institutes of Health. References I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.
R.P. RUBIN, Pharmacol. Rev. 22 328-428 (1972). W. CHEUNG, Sci. 207 19-27 (1~rl~O). R.J. DELORENZO,1~e'~eration Proc. 41 2265-2272 (1982). Y. NISHIZUKA, Science 225 1365-137I~ (1984). J.C. BROOKSand S. TREI~, Biochem. Pharmacol. 32 371-373 (1983). J.C. BROOKSand S. TREML, Life Sci. 34 669-674"[1983). S. PUBLICOVER, Naunyn-SchmiedebergAr-~'h. Pharm. 322 83-88 (1983). F. PELAYO, M.L. OUBOCOVICHand S.Z. Langer, Natur-~-(Lond) 274 76-78 ( 1978). L.F. QUENZERand R.L. VOLLE, J. Autonom. Ner. Sys. 8 161-164 (1983). S.Z. LANGER, Pharmacol. Rev. 32 337-362 (1981). S.J. PEROUTKA, D.C. U'PRICHAR~T D.A. GREENBERGand S.H. SNYDER, Neuropharmacol. 16 549-556 (1977). A. PELLEGRINODE-~ALDI and J. CORAZZA, Cell Tissue Res. 216 625-634 (1981). K. STARKE and T. ENDO, Gen. Pharmacol. 7 307-312 (1976).
1397