Vol. 179, No. 3, 1991 September 30, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1522-1528
ARACHIDONIC ACID AND DIACYLGLYCEROL ACT SYNERGISTICALLY TO ACTIVATE PROTEIN KINASE C IN VITRO AND IN VW0 David S.Lester*, Carlos Collin, Rene Etcheberrigaray and Daniel L.Alkon
National Institutes of Health, Section of Neural Systems, Park 5 Building, Rm.435, Bethesda, MD 20892 Received
August
13,
1991
Using a well-defined model membrane bilayer system, incorporation of both lipid second messengers,l,2-diacylglycerol and arachidonic acid, at submaximal activating concentrations, resulted in a synergistic activation of protein kinase C in a Ca2+/phosphatidylserine-dependent manner as measured by monitoring phosphoryiation of phosphoprotein substrates. The arachidonic acid appears to modulate membrane properties both at the hydrocarbon core and the membrane surface increasing the availability of the diacylglycerol which can bind to and subsequently activate the enzyme. Co-application of these two lipid activators to the Hermissenda photoreceptor reduced K+ channel conductance in a synergistic manner via a PKC-dependent pathway, Thus, these in vivo and in vitro studies suggest that the membrane bilayer properties of these PKC lipid activators interact to specifically regulate the cellular lipid microenvironment resulting in PKC activation. 0 1991 Academic Press,Inc.
It is well established that protein kinase C’ can be activated by either of the lipid second messengers, DAG or AA (1,2).
The concentration of these lipid activators
required for activation of PKC in in vitro studies is often considered to exceed physiological concentrations (3,4). There has been accumulating evidence that DAG and certain fatty acids may interact synergistically to activate PKC (5-8). We have used a model membrane bilayer system (6,9,10,?1) and an established PKC-regulated cellular system, K+ currents in the Hermissenda photoreceptor (12,13), to show that this synergistic activation may be an important cellular mechanism.
*To whom correspondence should be addressed.
Abbreviations: PKC=protein kinase C; DAG= 1&diacylglycerol; AA=arachidonic acid; PS=phosphatidylserine; PC=phosphatidylcholine; MLV=multilamellar vesicle; OAG= loleoyl-2-acetylglycerol; I,=inward K+ current; I,=outward K+ current. 0006-291x/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
1522
Vol.
179, No. 3, 1991
MATERIALS
BIOCHEMICAL
AND BlOPHYSlCAL
RESEARCH COMMUNICATIONS
AND METHODS
PKC was purified from the cytosolic fraction of Wistar rat brains as previously described(l4) and contained the three major isozymes. Mixed phospholipid multimellar lipid vesicles (MLV) were prepared as previously described (8). Phospholipids were obtained from Avanti Polar Lipids, and fatty acids were from NuChek Prep. All lipid products were stored at -80°C under liquid N,. MLVs were prepared and used on the same day. Model membrane bilaver assays. Histone 1 phosohorvlation: Assay conditions were as previously described (8,14). PKC activity is the difference between activity measured in the presence or absence of 10 PM Ca2+. Endoaenous phosohorvlation: An active PKC sample was taken one step before the final purification step of PKC. A number of the proteins that co-purify with PKC are good potential substrates. Based on previous reports, this fraction is free from contamination of other kinases (15). Endogenous phopshorylation was performed under the same conditions as previously described (16). The samples were run on precast polyacrylamide mini-gels (8-16%, Novex) and the dried gel exposed to Kodak X-ray film for 4 hr at -80°C using a DuPont Lightning Plus intensifying screen. Phorbol ester binding Assay conditions using [3H] phorbol dibutyrate (Amersham) were as previously described (8,16). Specific binding is the difference between sample with the lipid mixture plus the activator, minus binding to similar vesicles containing 6 PM phorbol myristate acetate (Calbiochem). Cellular voltaae clamp analyses. Two electrode voltage clamp technique, solutions and standard treatments for Hermissenda neurons have been described (17,18). Lipids were dried under a stream of N, and then under high vacuum overnight. Artificial sea water was added to the dried lipid immediately before use and sonicated (30 s) followed by vortexing (30 s). The suspension was continuously applied to the preparation with bath perfusion (about 5 ml/min). Pulse generation, data acaquisition and analysis was performed using pClamp (Axon Instruments, CA).
RESULTS Arachidonic acid was a relatively poor activator of PKC in this membrane system activating histone 1 phosphorylating activity to a maximal level of only 35% obtained by significantly lower concentrations of diacylglycerol (Figure la). This activation was dependent both on the presence of the acidic phospholipid, phosphatidylserine (PS), and Ca2+ (Figure la), suggesting that we were not selectively activating one specific PKC isozyme. Submaximal activating concentrations of DAG (0.5-l .O PM, lo-25% of maximal DAG-stimulated activity,
Figure 1a) and AA concentrations
stimulated PKC histone 1 (Figure lb) phosphorylating
activity
and endogenous
of 2.5-10 PM
substrate
equivalent to optimal DAG concentrations.
(Figure 2a)
At higher AA
concentrations (11-22 PM), histone 1 and endogenous substrate phosphorylation was increased to activities far above levels obtained with DAG alone (Figures 1b and 2a). Using the [3H] phorbol ester binding assay, interactions with the active phorbol ester/DAG binding site were examined. As previously observed with linoleic acid (8) AA at lower concentrations
(~6 PM) did not compete with [3H] PdBu, but rather 1523
Vol.
779, No. 3, 1991
BIOCHEMICAL
AND
BIOPHYSKAL
RESEARCH
COMMUNICATIONS
2.0 z
1.6
1.6
$1.2 a z 0.8 a g 0.4
0.8
;
L
E 3 0.0
0
5
20
0
5
20
25
FIGURE1 .Arachidonic acid and diacylglycerol synergistically activate protein kinase C histone 1 phosporylating activity in a model membrane system. a) Activation of PKC by increasing concentrations of DAG (m) or AA in the presence (0) or absence (0) of Ca2+. The dependence of the AA upon the acidic phospholipid, PS, was also measured(o). b) The synergistic action of AA and DAG upon PKC activity. The activation was measured at two concentrations of DAG; 0.55 PM (0) or 1 .lO MM (a). The dependence on Ca*+ at the higher concentration of DAG (1 .lO ILM) was also measured (0). Experiments were repeated three times in triplicate. Standard deviations are indicated at each point.
significantly increased the binding in the presence or absence of DAG (Figure 2b). Thus, AA promotes binding of DAG to PKC. At higher AA concentrations, there was a decrease in binding not observed with linoleic acid. This type of biphasic response has been observed with other activators in the same membrane system (11). In order to examine the role of membrane stability (see ref.19 for review) we used a series of fatty acid analogs that modulate AA physical interactions at the hydrocarbon chain (cholesteryl arachidonate) and the charge at the membrane surface
200.0
200.0
160.0
160.0
E" ci
120.0
120.0
3
80.0
80.0
z p"
40.0
40.0
l.l3 ;
b
:
0.0
0.0
E
FIGURE 2. Diacylglycerol and arachidonic acid interact to modulate other protein kinase C activities in a synergistic fashion. a) Arachidonic acid and diacylglycerol interact synergistically to stimulate endogenous phosphorylation. Lane: 1) DAG, 1.1 PM. 2) DAG, 5.4 PM. 3) AA, 22 PM. 4) DAG (1.1 PM) and AA (5.4 PM). 5) DAG (1.1 bM) and AA (22 PM). b) Arachidonic acid does not act at the phorbol ester binding site of protein kinase C. The lipid activators were; DAG (o), AA (m) or AA + 1.1 PM DAG (0).
1524
Vol.
179,
BIOCHEMICAL AND BlOPHYSlCAL RESEARCH COMMUNICATIONS
No. 3, 1991
TABLE 1. Activation of protein kinase C by arachidonic acid analogs and analysis of their diacylglycerol-synergistic potential LIPIDS
histone 1 phosphorylation (pmol “P incorp./mg/min)
PS/PC DAG (1.1 M) DAG (5.5 M) AA DAG + AA ChA DAG + ChA MA DAG + MA EA DAG + EA
0
0.2720.03 0.96+0.04 0.37?0.01 1.8320.17 0.16+0.01 0.3520.05 0.2320.03 0.71kO.08 0.54&0.11 0.7820.16
13H] phorbol ester binding (nmol [3H] PdBu bound/mQ) 90%3 7824 5622 Ill-t7 11329 98?11 8023 10027 92A-4 104+6 11224
The histone 1 phosphorylation and [3H] PdBu binding assays were as described (8,16). DAG concentrations were 1.l or 5.4 PM. The fatty acid analogs were all at 11 PM. Fatty acid analogs used were: AA = arachidonic acid, ChA = cholesteryl arachidonate. MA = methyl arachidonate,EA = ethyl arachidonate.The experimentwas performed three times in triplicate. Data are expressed as result2 S.D..
(methyl and ethyl arachidonate) (see Table 1). The cholesterol analog, which would affect lipid spacing,
was a poor activator and incapable of inducing the synergistic
effect on both histone 1 phosphorylation and phorbol ester binding (Table 1). The ethyl and methyl esters were tess potent than AA in all of their effects which may suggest that the ionizable head group of AA plays some role in the optimal active conformation of the enzyme. Thus, arachidonic acid appears to exert its synergistic effect by modulating both the membrane surface and the hydrocarbon core. To determine whether this potentially new PKC activation pathway applied to a cellular system, we used the established PKC-regulated K+ channel activity in the Type B photoreceptor of the sea snail, Hermissenda, using voltage clamp techniques (12,13). Bath application of the cell permeant DAG analog, l-oleoyl-2-acetylglycerol
(OAG, 5
pg/ml), or AA (5 PM), alone, had no significant effects after 15 min. After 30 min incubation, OAG consistently increased I, by approximately 25%(Figures 3b and e) while AA had no effect (Figures 3a and e). This increase in current was not PKCdependent as the PKC inhibitor, staurosporine (10 PM), had no effect on this OAG enhancement and may be nonspecific, i.e. PKC-independent, as observed in other neuronal systems (20,21). Application of OAG (5 pg/ml) followed by AA (5 PM) 15 min later, produced
a 57 and a 44% reduction of iA and l,(Figures 3c and e),
respectively. These reductions also correlated with a midpoint shift of the steady-state 1525
BIOCHEMICAL
Vol. ‘179, No. 3, 1991
AND BIOPHYSICAL
d
RESEARCH COMMUNICATIONS
OAG-AA
e
FIGURE 3. Synergistic regulation by arachidonic acid and diacylglycerol of K+ currents in Hermissenda Type B photoreceptor. a) Voltage clamp measurements were made of macroscopic outward K+ currents (I) recorded in normal artificial sea water (ASW), and an inward Ca*+ current (I,) recorded with high external K+ ASW in Hermissenda photoreceptors. From a -60 mV holding potential (V,,), these currents were elicited with depolarizaing steps to 0 mV. Current transients were digitized at 500 &point, and leak subtracted 0nline.a) AA: Two superimposedtraces, control (C) and 30 min after addition of 5 MM AA. b) OAG: 5 pg/ml OAG caused an increase in both currents at 30 mint) OAG-AA: 15 min OAG (5 fig/ml) followed by AA 5 PM 15 min 1ater.d) Inward calcium current recorded under high external potassium. The two superimposed traces show that the sequential application of OAG and AA does not change the calcium current, but specifically reduces the tail current that is only carried by a now inward I,. e) Current amplitude changes 30 min after application of different lipids. Bars are meansof at least 5 experiments-+SEM.The statisticalsignificancewas determined using the Students T-test difference between two means.
1, inactivation of about 10 mV to more hyperpolarized membrane potentials, from a typical value of about -58 mV. Treatment of the preparation with staurosporine (10 PM) blocked the reduction in both currents suggesting that the OAG-AA synergistic action 1526
Vol. 179, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
is via a PKC-mediated pathway. The voltage-dependent
Ca2+ currents measured in
medium with elevated K+ (375 mM) and K+ channel blockers (4-AP and TEA) were not sensitive to any of these treatments. I, measured as a tail current under these conditions was also reduced only by the sequential application of OAG followed by AA (Figure 3d).
DISCUSSION An increasing number of studies show that cis-unsaturated fatty acids are capable of interacting synergistically with DAG to activate PKC (5-8). This synergistic activation has recently been shown to be induced in the three predominant PKC isozymes, Types I, II and Ill (22). In this study, we have further examined this phenomenon using DAG and AA in a defined model membrane bilayer and a cellular system.
The synergism is very significant at physiological concentrations of these
activators. The AA-DAG
synergism is far more potent than the linoleic acid-DAG
interaction (8) suggesting a greater specificity for AA. The mechanism by which this is accomplished does not seem to be via modification of the DAG binding site, but rather by alteration of the physical properties (membrane stability and order) of the membrane such that more of the enzyme is activated.
The alterations in the bilayer properties
appear to be both at the membrane surface and the hydrocarbon demonstrated using
core, as
AA analogs. The reduction in the PKC-regulated K+ fluxes in the
Hermissenda photoreceptor
provides the first experimental evidence that this synergy
may occur in biological systems.
Additional preliminary data indicates these two
activators may also act in a synergistic manner in long term potentiation (23) and regulation of Ca*+ channels in hippocampal cell cultures (24). These studies suggest that in neuronal systems the combined activation of phospholipases 4 and C may result in the suitable lipid microenvironment required for optimal PKC activation.
REFERENCES 1.Nishizuka, Y. (1986) Science 233, 305-312. 2. Huang, K.-P. (1989) Trends Neurosci. 12, 425-432. 3. Chauhan, V.P.S., Chauhan, A., Deshmukh, D.S. and Brockerhoff, H. (1990) Life Sci. 47, 981-986. 4.Siegel, D.P., Banschbach, J., Alford, D., Ellens, H., Lis, L.J., Quinn, P.J., Yeagle, P.L. and Bentz, J. (1989) Biochem. 28, 3703-3709. 5. Seifert, R., Schachtele, C. and Schultz, G. (1987) Biochem.Biophys.Res.Comm. 149, 762-768. 6. El Touny, S., Khan, W. and Hannun, Y. (1990) J.Biol.Chem. 265, 16437-16443. 1527
Vol.
179, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
7. Shearman., M.S., Shinomura, T., Oda, T. and Nishizuka, Y. (1991) FEBS Lett. 291, 261-264. 8. Lester, D.S. (1990) Biochim.Biophys.Acta 1054, 297-303. 9. Boni, L.T. and Rando, RR. (1985) J.Biol.Chem. 260, 10819-10825. 10. Molleyres, L.P. and Rando, R.R. (1988) J.Biol.Chem. 263, 14832-14839. 11. Lester, D.S. and Baumann, D.(1991) Eur.J.Pharmacol.Mol.Pharm.206, 301-308. 12. Alkon, D.L., Kubota, M., Neary, J.T., Naito, S., Coulter, D., and Rasmussen, H. (1986) Biochem.Biophys.Res.Comm. 134, 12451253. 13. Alkon, D.L., Naito, S., Kubota, M., Chen, C., Bank, B., Smallwood, J., Gallant, P. and Rasmussen, H. (1988) J.Neurochem. 51, 903-917. 14. Lester, D.S. and Brumfeld, V. (1991) Biophys.Chem. 39, 215224. 15. Huang, K.-P., Chan, K.-F.J., Singh, T.J., Nakabayashi, H. and Huang, F.L. (1986) J.Biol.Chem. 261, 12134-l 2140. 16. Lester, D.S. (1989) J.Neurochem. 52, 1950-1953. 17.Collin, C., Ikeno, H., Harrigan, H.R., Lederhendler, I.and Alkon, D.L. (1988) Bi0phys.J. 54, 955-960. 18. Collin, C., Papageorge, A.G., Sakakibara, M., Huddie, P.L., Lowy, D.R. and Alkon, D.L. (1990) Bi0phys.J. 58, 785-790. 19. Epand, R.M. and Lester, D.S. (1990) Trends Pharm.Sci. 11, 317-320. 20.Doerner, D., Abdel-Latif, M., Rogers, T.B. and Alger, B.E. (1990) J.Neurosci. 10, 1699- 1706. 21. Hockberger, P., Toselli, M., Swandulla, D. and Lux, H.D. (1989) Nature (Lond.) 338, 340-342. 22. Shinomura, T., Asaoka, Y., Oka, M. Yoshida, K. and Nishizuka, Y. (1991) Proc.Nat1.Acad.Sci.U.S.A. 88, 5149-5153. 23. Bramham, C.R., Lester, D.S. and Alkon, D.L. (1991) Soc.Neurosci.Abst. in press. 24. Keyser, D.O. and Alger, B.E. (1991) Neuron, in press.
1528
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1522-1528
ARACHIDONIC ACID AND DIACYLGLYCEROL ACT SYNERGISTICALLY TO ACTIVATE PROTEIN KINASE C IN VITRO AND IN VW0 David S.Lester*, Carlos Collin, Rene Etcheberrigaray and Daniel L.Alkon
National Institutes of Health, Section of Neural Systems, Park 5 Building, Rm.435, Bethesda, MD 20892 Received
August
13,
1991
Using a well-defined model membrane bilayer system, incorporation of both lipid second messengers,l,2-diacylglycerol and arachidonic acid, at submaximal activating concentrations, resulted in a synergistic activation of protein kinase C in a Ca2+/phosphatidylserine-dependent manner as measured by monitoring phosphoryiation of phosphoprotein substrates. The arachidonic acid appears to modulate membrane properties both at the hydrocarbon core and the membrane surface increasing the availability of the diacylglycerol which can bind to and subsequently activate the enzyme. Co-application of these two lipid activators to the Hermissenda photoreceptor reduced K+ channel conductance in a synergistic manner via a PKC-dependent pathway, Thus, these in vivo and in vitro studies suggest that the membrane bilayer properties of these PKC lipid activators interact to specifically regulate the cellular lipid microenvironment resulting in PKC activation. 0 1991 Academic Press,Inc.
It is well established that protein kinase C’ can be activated by either of the lipid second messengers, DAG or AA (1,2).
The concentration of these lipid activators
required for activation of PKC in in vitro studies is often considered to exceed physiological concentrations (3,4). There has been accumulating evidence that DAG and certain fatty acids may interact synergistically to activate PKC (5-8). We have used a model membrane bilayer system (6,9,10,?1) and an established PKC-regulated cellular system, K+ currents in the Hermissenda photoreceptor (12,13), to show that this synergistic activation may be an important cellular mechanism.
*To whom correspondence should be addressed.
Abbreviations: PKC=protein kinase C; DAG= 1&diacylglycerol; AA=arachidonic acid; PS=phosphatidylserine; PC=phosphatidylcholine; MLV=multilamellar vesicle; OAG= loleoyl-2-acetylglycerol; I,=inward K+ current; I,=outward K+ current. 0006-291x/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
1522
Vol.
179, No. 3, 1991
MATERIALS
BIOCHEMICAL
AND BlOPHYSlCAL
RESEARCH COMMUNICATIONS
AND METHODS
PKC was purified from the cytosolic fraction of Wistar rat brains as previously described(l4) and contained the three major isozymes. Mixed phospholipid multimellar lipid vesicles (MLV) were prepared as previously described (8). Phospholipids were obtained from Avanti Polar Lipids, and fatty acids were from NuChek Prep. All lipid products were stored at -80°C under liquid N,. MLVs were prepared and used on the same day. Model membrane bilaver assays. Histone 1 phosohorvlation: Assay conditions were as previously described (8,14). PKC activity is the difference between activity measured in the presence or absence of 10 PM Ca2+. Endoaenous phosohorvlation: An active PKC sample was taken one step before the final purification step of PKC. A number of the proteins that co-purify with PKC are good potential substrates. Based on previous reports, this fraction is free from contamination of other kinases (15). Endogenous phopshorylation was performed under the same conditions as previously described (16). The samples were run on precast polyacrylamide mini-gels (8-16%, Novex) and the dried gel exposed to Kodak X-ray film for 4 hr at -80°C using a DuPont Lightning Plus intensifying screen. Phorbol ester binding Assay conditions using [3H] phorbol dibutyrate (Amersham) were as previously described (8,16). Specific binding is the difference between sample with the lipid mixture plus the activator, minus binding to similar vesicles containing 6 PM phorbol myristate acetate (Calbiochem). Cellular voltaae clamp analyses. Two electrode voltage clamp technique, solutions and standard treatments for Hermissenda neurons have been described (17,18). Lipids were dried under a stream of N, and then under high vacuum overnight. Artificial sea water was added to the dried lipid immediately before use and sonicated (30 s) followed by vortexing (30 s). The suspension was continuously applied to the preparation with bath perfusion (about 5 ml/min). Pulse generation, data acaquisition and analysis was performed using pClamp (Axon Instruments, CA).
RESULTS Arachidonic acid was a relatively poor activator of PKC in this membrane system activating histone 1 phosphorylating activity to a maximal level of only 35% obtained by significantly lower concentrations of diacylglycerol (Figure la). This activation was dependent both on the presence of the acidic phospholipid, phosphatidylserine (PS), and Ca2+ (Figure la), suggesting that we were not selectively activating one specific PKC isozyme. Submaximal activating concentrations of DAG (0.5-l .O PM, lo-25% of maximal DAG-stimulated activity,
Figure 1a) and AA concentrations
stimulated PKC histone 1 (Figure lb) phosphorylating
activity
and endogenous
of 2.5-10 PM
substrate
equivalent to optimal DAG concentrations.
(Figure 2a)
At higher AA
concentrations (11-22 PM), histone 1 and endogenous substrate phosphorylation was increased to activities far above levels obtained with DAG alone (Figures 1b and 2a). Using the [3H] phorbol ester binding assay, interactions with the active phorbol ester/DAG binding site were examined. As previously observed with linoleic acid (8) AA at lower concentrations
(~6 PM) did not compete with [3H] PdBu, but rather 1523
Vol.
779, No. 3, 1991
BIOCHEMICAL
AND
BIOPHYSKAL
RESEARCH
COMMUNICATIONS
2.0 z
1.6
1.6
$1.2 a z 0.8 a g 0.4
0.8
;
L
E 3 0.0
0
5
20
0
5
20
25
FIGURE1 .Arachidonic acid and diacylglycerol synergistically activate protein kinase C histone 1 phosporylating activity in a model membrane system. a) Activation of PKC by increasing concentrations of DAG (m) or AA in the presence (0) or absence (0) of Ca2+. The dependence of the AA upon the acidic phospholipid, PS, was also measured(o). b) The synergistic action of AA and DAG upon PKC activity. The activation was measured at two concentrations of DAG; 0.55 PM (0) or 1 .lO MM (a). The dependence on Ca*+ at the higher concentration of DAG (1 .lO ILM) was also measured (0). Experiments were repeated three times in triplicate. Standard deviations are indicated at each point.
significantly increased the binding in the presence or absence of DAG (Figure 2b). Thus, AA promotes binding of DAG to PKC. At higher AA concentrations, there was a decrease in binding not observed with linoleic acid. This type of biphasic response has been observed with other activators in the same membrane system (11). In order to examine the role of membrane stability (see ref.19 for review) we used a series of fatty acid analogs that modulate AA physical interactions at the hydrocarbon chain (cholesteryl arachidonate) and the charge at the membrane surface
200.0
200.0
160.0
160.0
E" ci
120.0
120.0
3
80.0
80.0
z p"
40.0
40.0
l.l3 ;
b
:
0.0
0.0
E
FIGURE 2. Diacylglycerol and arachidonic acid interact to modulate other protein kinase C activities in a synergistic fashion. a) Arachidonic acid and diacylglycerol interact synergistically to stimulate endogenous phosphorylation. Lane: 1) DAG, 1.1 PM. 2) DAG, 5.4 PM. 3) AA, 22 PM. 4) DAG (1.1 PM) and AA (5.4 PM). 5) DAG (1.1 bM) and AA (22 PM). b) Arachidonic acid does not act at the phorbol ester binding site of protein kinase C. The lipid activators were; DAG (o), AA (m) or AA + 1.1 PM DAG (0).
1524
Vol.
179,
BIOCHEMICAL AND BlOPHYSlCAL RESEARCH COMMUNICATIONS
No. 3, 1991
TABLE 1. Activation of protein kinase C by arachidonic acid analogs and analysis of their diacylglycerol-synergistic potential LIPIDS
histone 1 phosphorylation (pmol “P incorp./mg/min)
PS/PC DAG (1.1 M) DAG (5.5 M) AA DAG + AA ChA DAG + ChA MA DAG + MA EA DAG + EA
0
0.2720.03 0.96+0.04 0.37?0.01 1.8320.17 0.16+0.01 0.3520.05 0.2320.03 0.71kO.08 0.54&0.11 0.7820.16
13H] phorbol ester binding (nmol [3H] PdBu bound/mQ) 90%3 7824 5622 Ill-t7 11329 98?11 8023 10027 92A-4 104+6 11224
The histone 1 phosphorylation and [3H] PdBu binding assays were as described (8,16). DAG concentrations were 1.l or 5.4 PM. The fatty acid analogs were all at 11 PM. Fatty acid analogs used were: AA = arachidonic acid, ChA = cholesteryl arachidonate. MA = methyl arachidonate,EA = ethyl arachidonate.The experimentwas performed three times in triplicate. Data are expressed as result2 S.D..
(methyl and ethyl arachidonate) (see Table 1). The cholesterol analog, which would affect lipid spacing,
was a poor activator and incapable of inducing the synergistic
effect on both histone 1 phosphorylation and phorbol ester binding (Table 1). The ethyl and methyl esters were tess potent than AA in all of their effects which may suggest that the ionizable head group of AA plays some role in the optimal active conformation of the enzyme. Thus, arachidonic acid appears to exert its synergistic effect by modulating both the membrane surface and the hydrocarbon core. To determine whether this potentially new PKC activation pathway applied to a cellular system, we used the established PKC-regulated K+ channel activity in the Type B photoreceptor of the sea snail, Hermissenda, using voltage clamp techniques (12,13). Bath application of the cell permeant DAG analog, l-oleoyl-2-acetylglycerol
(OAG, 5
pg/ml), or AA (5 PM), alone, had no significant effects after 15 min. After 30 min incubation, OAG consistently increased I, by approximately 25%(Figures 3b and e) while AA had no effect (Figures 3a and e). This increase in current was not PKCdependent as the PKC inhibitor, staurosporine (10 PM), had no effect on this OAG enhancement and may be nonspecific, i.e. PKC-independent, as observed in other neuronal systems (20,21). Application of OAG (5 pg/ml) followed by AA (5 PM) 15 min later, produced
a 57 and a 44% reduction of iA and l,(Figures 3c and e),
respectively. These reductions also correlated with a midpoint shift of the steady-state 1525
BIOCHEMICAL
Vol. ‘179, No. 3, 1991
AND BIOPHYSICAL
d
RESEARCH COMMUNICATIONS
OAG-AA
e
FIGURE 3. Synergistic regulation by arachidonic acid and diacylglycerol of K+ currents in Hermissenda Type B photoreceptor. a) Voltage clamp measurements were made of macroscopic outward K+ currents (I) recorded in normal artificial sea water (ASW), and an inward Ca*+ current (I,) recorded with high external K+ ASW in Hermissenda photoreceptors. From a -60 mV holding potential (V,,), these currents were elicited with depolarizaing steps to 0 mV. Current transients were digitized at 500 &point, and leak subtracted 0nline.a) AA: Two superimposedtraces, control (C) and 30 min after addition of 5 MM AA. b) OAG: 5 pg/ml OAG caused an increase in both currents at 30 mint) OAG-AA: 15 min OAG (5 fig/ml) followed by AA 5 PM 15 min 1ater.d) Inward calcium current recorded under high external potassium. The two superimposed traces show that the sequential application of OAG and AA does not change the calcium current, but specifically reduces the tail current that is only carried by a now inward I,. e) Current amplitude changes 30 min after application of different lipids. Bars are meansof at least 5 experiments-+SEM.The statisticalsignificancewas determined using the Students T-test difference between two means.
1, inactivation of about 10 mV to more hyperpolarized membrane potentials, from a typical value of about -58 mV. Treatment of the preparation with staurosporine (10 PM) blocked the reduction in both currents suggesting that the OAG-AA synergistic action 1526
Vol. 179, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
is via a PKC-mediated pathway. The voltage-dependent
Ca2+ currents measured in
medium with elevated K+ (375 mM) and K+ channel blockers (4-AP and TEA) were not sensitive to any of these treatments. I, measured as a tail current under these conditions was also reduced only by the sequential application of OAG followed by AA (Figure 3d).
DISCUSSION An increasing number of studies show that cis-unsaturated fatty acids are capable of interacting synergistically with DAG to activate PKC (5-8). This synergistic activation has recently been shown to be induced in the three predominant PKC isozymes, Types I, II and Ill (22). In this study, we have further examined this phenomenon using DAG and AA in a defined model membrane bilayer and a cellular system.
The synergism is very significant at physiological concentrations of these
activators. The AA-DAG
synergism is far more potent than the linoleic acid-DAG
interaction (8) suggesting a greater specificity for AA. The mechanism by which this is accomplished does not seem to be via modification of the DAG binding site, but rather by alteration of the physical properties (membrane stability and order) of the membrane such that more of the enzyme is activated.
The alterations in the bilayer properties
appear to be both at the membrane surface and the hydrocarbon demonstrated using
core, as
AA analogs. The reduction in the PKC-regulated K+ fluxes in the
Hermissenda photoreceptor
provides the first experimental evidence that this synergy
may occur in biological systems.
Additional preliminary data indicates these two
activators may also act in a synergistic manner in long term potentiation (23) and regulation of Ca*+ channels in hippocampal cell cultures (24). These studies suggest that in neuronal systems the combined activation of phospholipases 4 and C may result in the suitable lipid microenvironment required for optimal PKC activation.
REFERENCES 1.Nishizuka, Y. (1986) Science 233, 305-312. 2. Huang, K.-P. (1989) Trends Neurosci. 12, 425-432. 3. Chauhan, V.P.S., Chauhan, A., Deshmukh, D.S. and Brockerhoff, H. (1990) Life Sci. 47, 981-986. 4.Siegel, D.P., Banschbach, J., Alford, D., Ellens, H., Lis, L.J., Quinn, P.J., Yeagle, P.L. and Bentz, J. (1989) Biochem. 28, 3703-3709. 5. Seifert, R., Schachtele, C. and Schultz, G. (1987) Biochem.Biophys.Res.Comm. 149, 762-768. 6. El Touny, S., Khan, W. and Hannun, Y. (1990) J.Biol.Chem. 265, 16437-16443. 1527
Vol.
179, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
7. Shearman., M.S., Shinomura, T., Oda, T. and Nishizuka, Y. (1991) FEBS Lett. 291, 261-264. 8. Lester, D.S. (1990) Biochim.Biophys.Acta 1054, 297-303. 9. Boni, L.T. and Rando, RR. (1985) J.Biol.Chem. 260, 10819-10825. 10. Molleyres, L.P. and Rando, R.R. (1988) J.Biol.Chem. 263, 14832-14839. 11. Lester, D.S. and Baumann, D.(1991) Eur.J.Pharmacol.Mol.Pharm.206, 301-308. 12. Alkon, D.L., Kubota, M., Neary, J.T., Naito, S., Coulter, D., and Rasmussen, H. (1986) Biochem.Biophys.Res.Comm. 134, 12451253. 13. Alkon, D.L., Naito, S., Kubota, M., Chen, C., Bank, B., Smallwood, J., Gallant, P. and Rasmussen, H. (1988) J.Neurochem. 51, 903-917. 14. Lester, D.S. and Brumfeld, V. (1991) Biophys.Chem. 39, 215224. 15. Huang, K.-P., Chan, K.-F.J., Singh, T.J., Nakabayashi, H. and Huang, F.L. (1986) J.Biol.Chem. 261, 12134-l 2140. 16. Lester, D.S. (1989) J.Neurochem. 52, 1950-1953. 17.Collin, C., Ikeno, H., Harrigan, H.R., Lederhendler, I.and Alkon, D.L. (1988) Bi0phys.J. 54, 955-960. 18. Collin, C., Papageorge, A.G., Sakakibara, M., Huddie, P.L., Lowy, D.R. and Alkon, D.L. (1990) Bi0phys.J. 58, 785-790. 19. Epand, R.M. and Lester, D.S. (1990) Trends Pharm.Sci. 11, 317-320. 20.Doerner, D., Abdel-Latif, M., Rogers, T.B. and Alger, B.E. (1990) J.Neurosci. 10, 1699- 1706. 21. Hockberger, P., Toselli, M., Swandulla, D. and Lux, H.D. (1989) Nature (Lond.) 338, 340-342. 22. Shinomura, T., Asaoka, Y., Oka, M. Yoshida, K. and Nishizuka, Y. (1991) Proc.Nat1.Acad.Sci.U.S.A. 88, 5149-5153. 23. Bramham, C.R., Lester, D.S. and Alkon, D.L. (1991) Soc.Neurosci.Abst. in press. 24. Keyser, D.O. and Alger, B.E. (1991) Neuron, in press.
1528
