- Email: [email protected]
Inhibitors of calmodulin and protein kinase C block different phases of hippocampal long-term potentiation
Brain Research, 461 (1988) 388-392
388
Elsevier BRE 23134
Inhibitors of calmodulin and protein kinase C block different phases of hippocampal long-term potentiation Klaus G. Reymann 1, Rudolf Br6demann 2, Hiroshi Kase 3 and Hansjfirgen Matthies 1 1Institute of Neurobiology and Brain Research, Academy of Sciences G.D.R., Magdeburg (G. D.R.), "Instituteof Pharmacology and Toxicology, MedicalAcademy Magdeburg, Magdeburg (G.D.R.) and 3Tokyo Research Laboratories, Kyowa Hakko Kogyo Co., Ltd., Tokyo (Japan) (Accepted 21 June 1988)
Key words: Long-term potentiation; Hippocampus; Synaptic plasticity; Calmodulin; Protein kinase C; Second messenger; Calmidazolium; K-252b; Protein phosphorylation
The effects of a calmodulin (CAM) inhibitor, which does not influence Ca2+ fluxes (calmidazolium, RO-24571), and a new potent inhibitor of protein kinase C (K-252b) on long-term potentiation (LTP) were compared in hippocampal slices. Tetanic stimulation of the stratum radiatum during perfusion of calmidazolium (50 nM) failed to induce the characteristic post-tetanic and long-term increase in the magnitude of CAl-evoked responses. During perfusion with K-252b (50 nM) post-tetanic potentiation and initial LTP is expressed normally, but thereafter declines back to baseline with a 60 rain delay. By themselves, the inhibitors had no significant effect on synaptic transmission in a non-tetanized control input. Our data are in line with current evidence from several laboratories that CaM- and protein kinase C (PKC)-dependent processes are involved in LTP and support the hypothesis that CaM mediates initiation and that PKC mediates mechanisms underlying the maintenance of LTP.
Long-term potentiation (LTP) is a particular type of synaptic plasticity which has several features that make it an interesting m o d e l of information storage 3'3°. Our present knowledge of intracellular signaling systems whereby a short tetanization produces long-term e n h a n c e m e n t of a monosynaptic connection is, however, far from being satisfactory. The purpose of the present study was to investigate the ability of calmodulin (CAM) and protein kinase C (PKC) inhibitors to antagonize either the initiation or the maintenance of LTP. The induction of h i p p o c a m p a l long-term potentiation (LTP) was shown to be correlated by a translocation of the multifunctional, calcium binding regulatory protein calmodulin from cytosolic c o m p a r t m e n t s towards the m e m b r a n e 27 and to be susceptible to disruption by C a M inhibitors s'9'26. Thus, the observed inhibition of LTP by trifluoperazine was interpreted as supporting the role of C a M in mechanisms of synaptic plasticity8'9'26. H o w e v e r , trifluoperazine is also
a potent inhibitor of Ca 2+ influx 12, an ionic event which is crucial for the initiation of LTP 7'14'22. Moreover, others did not find LTP-blocking e f f e c t s of C a M inhibitors is. W e tested here a C a M inhibitor tl,31 which, in low concentration, has no influence on calcium fluxes and P K C activity ~2,t9. Recently, the Ca2+/phospholipid-dependent P K C has been shown also to be involved in mechanisms of LTP 1"4~13'2°'24'2s'29. It seems unlikely that trifluoperazine has exhibited its action in previous experiments 8'9'26 only via inhibition of this kinase. Inhibitors of PKC have been shown to influence LTP in a different way 2°,2s. A l t h o u g h the P K C inhibitor polymyxin B blocked LTP like trifluoperazine, it neither affected the post-tetanic nor the early stage of hippocampal LTP. To substantiate the assumption that PKC actually mediates only the maintenance but not the initiation of LTP we tested the new potent P K C inhibitor K-252b 15, known to be effective in blocking the phorbol ester-induced type of LTP 29. A l t o g e t h e r ,
Correspondence: K.G. Reymann, Institute of Neurobiology and Brain Research, Academy of Sciences G.D.R., Leipziger Strasse 44, DDR-3090 Magdeburg, G.D.R. 11006-8993/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
389 the aim of the present work was to distinguish more precisely the interval in which CaM- and PKC-dependent processes are involved in LTP using an identical testing paradigm for the specific inhibitors. Transverse hippocampal slices were prepared from 7-week-old Wistar rats as previously described 28 with the slices superfused at 35 °C with a medium containing (mM): NaC1 124, KCI 4.9, KH2PO 4 1.2, MgSO4 1.3, CaCI 2 2.5, NaHCO 3 25.6, D-glucose 10. Stimulating and recording electrodes were etched and isolated stainless-steel wires. Extracellular recordings were obtained from the stratum radiatum (population-excitatory postsynaptic potential (EPSP)) and the stratum pyramidale (population spike) of CA1. Biphasic, constant-current pulses (0.2 Hz, 100/~s each half cycle) were delivered to two separate stimulation electrodes positioned on each side of the point of EPSP recording. The electrode on the CA3 side was subsequently used for tetanization (100 Hz/1 s at test stimulation strength), whereas the electrode on the subicular side served as an independent control input 28. Test responses were recorded at a stimulation strength which elicited a population spike (PS) of 30% of its maximum amplitude. Response changes were expressed as the average percentage change of EPSP slope and PS amplitude from the baseline obtained before drug treatment and tetanization. Comparisons between control and drugtreated groups were made with the two-tailed Mann-
Whitney U-test for unpaired observations (P < 0.05). Calmidazolium ll'31 (Boehringer) and K-252b (Na salt) 15 (Kyowa Hakko Kogyo) were first dissolved in dimethylsulfoxide and then added to the perfusion 4030C "¢~
10-Ck O O. (/)
calmidazol.
K-252b
0--10--
0. LU
L~ control [] K - 2 5 2 b o calrnidazoleum
50nM
I 0
I 30
T
100-
tJ
=
i
50-
E 0- ~ -
A I 0
I 30
TIME
/'\. 10.min
90.min
_J Fig. 1. Representative CA1 population spike responses to synaptic activation before and after a 100 Hz/1 s tetanic shock in the presence of the CaM inhibitor calmidazolium (50 nM) or the PKC inhibitor K-252b (50 nM). The computer plots are averages of 4 potentials evoked by low frequency test stimulation. Calibration: 2 mV/5 ms, but 1 mV for the calmidazoliumtreated slice.
I 90 100 Hz
100 Hz
before
[ 60
100 Hz
et control
20-
I 60
I 90 100 Hz
(minutes)
Fig. 2. Changes in the potentiation of population EPSP and population spike from the tetanized input (control n = 8, triangles) seen after 50 nM bath application of calmidazolium and K-252b. In the groups pretreated with calmidazolium (n = 8, circles) no post-tetanic and long-term potentiation occurred. Low concentrations of the PKC-inhibitor K-252b, however, significantly blocked only later stages of LTP (n = 11, squares). The EPSP magnitude was evaluated by measuring the initial, rising slope (mV/ms) of the EPSP between the onset and the maximum. The PS amplitude was evaluated by taking the voltage difference between the peak negativity and the preceding positive peak. The data shown are plotted as an average of the percentage change from baseline responses obtained immediately before the inhibitor application (means across slices, each from a separate animal; S.E.M. is shown at the 30th and the 60th min). Filled symbols indicate a significant change (P < 0.05) in comparison to control experiments. The difference between EPSPs of both drug-treated groups is also significant for 45 min after tetanization.
390 line resulting in a final bath concentration of the solvent <0.01%. Both inhibitors were applied 7 min before the tetanic shock and washed 20 min later. The vehicle alone had no effect on parameters of the present study. A short tetanization led to a long-term enhancement in PS in 8 of 8 slices (EPSP 7 of 8 slices) (Figs. 1 and 2). As was shown previously (for review see refs. 10, 28), the tetanization in one input did not cause any significant change in a separate control input within the first hours after tetanization. During this study, the deviation from baseline in the non-tetanized control input was less than 12 _+ 6% for the EPSP slope and less than 16 _+ 6% for the PS amplitude. Neither compound used here influenced the parameters from this control input. Therefore, the drug effects described below are due to their specific influence on LTP-related processes. The corresponding data from the non-tetanized input are not shown because of the high degree of overlapping of the values from the 3 groups under investigation (maximum deviation between EPSP values 12 _+ 8%; between PS values 32 + 16%). At a higher concentration (500 nM, one slice) calmidazolium reversibly blocked synaptic transmission by about 50%. Preincubation of the slices with 50 nM calmidazolium clearly blocked post-tetanic (1 min) and long-term potentiation (>10 min) in 7 of 8 slices (Figs. 1 and 2). LTP did not recover long after calmidazolium was washed from the chamber. By itself, calmidazolium has only a non-significant late depressive effect on the PS amplitude of the control input (-32 + 16% at 90 min), which cannot explain the clear and permanent prevention of LTP during this study. Some remaining potentiation of the PS may be due to an additional spike potentiating mechanism (E/S left-shift; for explanation see ref. 28), which seems to be not sensitive to calmidazolium. The data presented in Figs. 1 and 2 show that K252b bath-applied at a concentration of 50 nM, which blocks phorbol ester-induced potentiation 29, does not substantially inhibit synaptically evoked post-tetanic and the initial part of long-term potentiation in CA1 hippocampal neurons. An EPSP potentiation >10% was seen in 10 of 11 slices 10 min after tetanization. After some delay, however, LTP of synaptic responses returned to baseline (increase in EPSP >10% in 3 of 11 slices after 90 rain). The PS ampli-
tude was similarly decreased after 60 min. However, as was the case with calmidazolium and polymyxin B2s, some potentiation of the PS remained. In one slice 500 nM K-252b blocked LTP already after 20 min. The small, mostly non-significant early effect of K-252b may be due to an effect on CaM- or cAMPdependent kinases 15 (Kase et al., unpublished data). The effect of both inhibitors was only partially reversible within the duration of the experiment (Fig. 2). A second tetanization at the end of the experiment was followed by a small PS potentiation (calmidazotium group = + 16% at the 1st min with respect to corresponding pretetanic values; K-252b group = +15% and +10% at the 1st and the 15th min after the train, respectively; P < 0.05). In addition, we have measured changes in PS peak latency. The typical shortening of the peak latency following the high-frequency train (0.5 + 0.1 ms decrease up to the 90th min in the control group) was partially antagonized by both inhibitors (calmidazolium: 0.2 + 0.1 ms decrease at the 1st min; K-252b: 0.2 + 0.1 ms decrease at the 90th rain; P < 0.05). The most likely source of calcium necessary for activation of CaM is the influx of extracellular calcium through NMDA-receptor-linked ion channels 7':3'25. Additional sources of free cytosolic calcium might be its influx through ion channels sensitive to Ca 2+ antagonists 14 or its transient release from intracellular stores following the receptor-mediated stimulation of phosphoinositol turnover (cf. ref. 5). The present findings deliver first evidence that a CaM antagonist, which does not influence Ca 2+ fluxes in other neuronal systems 12and which has no affinity to dopamine receptor sites (Van Belle 11) in the concentrations used here, is capable of blocking post-tetanic potentiation and LTP. This is in accordance with similar studies in which less selective CaM antagonists prevented LTP 8"9'26and confirms the hypothesis that besides the Ca 2+ influx also CaM activation is a necessary prerequisite for LTP. Under certain circumstances, post-tetanic potentiation and LTP can independently occur and probably represent separate phenomena 3°. Our data, however, demonstrate the dependence of both phenomena on CaM. There are various CaM-dependent events in preand postsynaptic compartments. Recently it was demonstrated that both tetanization and calmodulin application enhance K+-induced glutamate release 21.
391 Further studies will be of interest to determine the CaM-stimulated target enzymes and their substrates which are important for LTP. In this field, a crucial role can be assumed for the Ca2+/CaM kinase II which is believed to constitute a major fraction of postsynaptic density proteins 16. Another possibility is a 40 kDa protein which is among the proteins phosphorylated during LTP the most susceptible to CaM and its inhibitors 9. It is important to note here that a significant inhibition of LTP by K-252b occurred long after its induction. This confirms a recent multiple phase hypothesis of LTP based on studies in which other types of PKC inhibitors with a different inhibition profile were used 2°'2s. According to our hypothesis 28 an intermediate and late stage of LTP is susceptible to PKC inhibitors and only the late stage is susceptible to the protein synthesis inhibitor anisomycin (>5 h) t°. We referred to these phases as LTP 2 and LTP 3(ref. 28). Because synthesis and dendritic transport of macromolecules are relatively slow, there may be a need for a distinct form of intermediate LTP to bridge the short- and long-term processes, similarly as during memory formation. As far as PKC's role in maintenance of LTP decreases with time after the induction 2s, it seems necessary that PKC and perhaps CaM kinases induce or stabilize late biochemical or even structural changes that have been observed following LTP (for review see ref. 30) rather than mediate the initial production of LTP (LTP 1). Both pre21,24 and postsynaptic 4'13"29PKC-dependent processes have been suggested to be involved. A recent prelimi n a r y observation has demonstrated a delayed increase in sensitivity of quisqualate receptors during LTP 6. The similar time course indicates the interesting possibility that PKC may be involved especially in
1 Akers, R.F., Lovinger, D.M., Colley, P.A., Linden, D.J. and Routtenberg, A., Translocation of protein kinase C activity may mediate hippocampal long-term potentiation, Science, 231 (1986) 587-589. 2 Alkon, D.L., Kubota, M., Neary, J.T., Naito, S., Coulter, D. and Rasmussen, H., C-kinase activation prolongs Ca2+dependent inactivation of K currents, Biochem. Biophys. Res. Commun., 134 (1986) 1245-1253. 3 Andersen, P., Long-term potentiation - - outstanding problems. In J.-P. Changeux and M. Konoshi (Eds.), The Neu-
this mechanism. The relationship between CaM inhibition and late phases of LTP is not yet clear. Although calmidazolium blocks LTP immediately and irreversibly it cannot be ruled out that CaM-dependent processes are also involved in intermediate and late mechanisms. The low concentration of calmidazolium used in this study may exclude any effect on late PKC-dependent processes 19. Following the initial translocation of calmodulin towards the membrane an opposite translocation towards the cytosol occurred 1 h after tetanization in hippocampus/dentate slices under similar experimental conditions as used here 27, indicating the possible existence of a second CaM-dependent process. Therefore, the lack of recovery after washout of calmidazolium could mean that these two distinct second messenger systems (CaM and PKC) converge at some level to enable post-translational modifications (for review see ref. 1) and/or synthesis of LTP-relevant proteins (for review see refs. 10, 17). Such a synergistic interaction of PKC with Ca2+/CaM kinase was described for the Ca2+-mediated K + current reduction occurring during conditioning in Hermissenda. Interestingly, similarly as in the LTP model PKC activation prolongs the effect 2. In line with the interpretation in this communication is the recently described increased hydrolysis of phosphatidylinositol-4,5-biphosphate in a postsynaptic compartment 45 min after tetanization (M.A. Lynch et al., Neurosci. Lett., 84 (1988) 291-296). Further studies will reveal whether both enzymes also share a common substrate during LTP. We thank Karin Schulzeck, Rosemarie Schlichting and Steffi Roth for expert assistance and Reinhard Jork for constructive comments on the manuscript.
ral and Molecular Bases of Learning, Wiley, New York,
1987, pp. 239-269. Andersen, P., Hvalby, O. and Reymann, K., Postsynaptic mechanisms contribute to synaptic potentiation induced by phorbol esters in rat hippocampus pyramidal cells in vitro, J. Physiol. (Lond.), 398 (1988)35P. B/~r, P.R., Wiegant, F,, Lopes da Silva, F.H. and Gispen, W., Tetanic stimulation affects the metabolism of phosphoinositides in hippocampal slices, Brain Research, 321 (1984) 381-385.
392 6 Collingridge, G.L. and Lester, R.A.J., The sensitivity of CA1 neurones to quisqualate following high frequency stimulation in rat hippocampus in vitro, J. Physiol. (Lond.), 398 (1988) 22P. 7 Dingledine, R., N-methylaspartate activates voltage-dependent calcium conductance in rat hippocampal pyramidal cells, J. Physiol. (Lond.), 343 (1983) 385-405. 8 Dunwiddie, T.V., Roberson, N.L. and Worth, T., Modulation of long-term potentiation: effects of adrenergic and neuroleptic drugs, Pharmacol. Biochem. Behav., 17 (1982) 1257-1264. 9 Finn, R.C., Browning, M. and Lynch, G., Trifluoperazine inhibits hippocampal long-term potentiation and the phosphorylation of a 40,000 dalton protein, Neurosci. Lett., 19 (1980) 103-108. 10 Frey, U., Krug, M., Reymann, K. and Matthies, H., Anisomycin, an inhibitor of protein synthesis, blocks late phases of LTP-phenomena in the hippocampal CA1 region in vitro, Brain Research, 452 (1988) 57-65. 11 Gietzen, K., Wiithrich, A. and Bader, H., R 24571: a new powerful inhibitor of red blood cell Ca2÷-transport ATPase and of calmodulin-regulated functions, Biochem. Biophys. Res. Commun., 101 (1981)418-425. 12 Greenberg, D.A., Carpenter, C.L. and Messing, R.O., Interaction of calmodulin inhibitors and protein kinase C inhibitors with voltage-dependent calcium channels, Brain Research, 404 (1987) 401-404. 13 Hu, G.-Y., Hvalby, O., Waalas, S.I., Albert, K.A., Skjef1o, P., Andersen, P. and Greengard, P., Protein kinase C injection into hippocampal pyramidal cells elicits features of long-term potentiation, Nature (Lond.), 328 (1987) 426-429. 14 Izumi, Y., Ito, K., Miyakawa, H. and Kato, H., Requirement of extracellular Ca 2÷ after tetanus for induction of long-term potentiation in guinea pig hippocampal slices, Neurosci. Lett., 77 (1987) 176-180. 15 Kase, H., Iwahashi, K., Nakanishi, S., Matsuda, Y., Yamada, K., Takahashi, M., Murakata, C., Sato, A. and Kaneko, M., K-252 compounds, novel and potent inhibitors of protein kinase C and cyclic nucleotide-dependent protein kinases, Biochem. Biophys. Res. Commun., 142 (1987) 436-440. 16 Kelly, P.T., McGuinness, T.L. and Greengard, P., Evidence that the major postsynaptic density protein is a component of a Ca2+/calmodulin-dependent protein kinase, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 945-949. 17 Krug, M., Loessner, B. and Ott, T., Anisomycine blocks the late phase of long-term potentiation in the dentate gyrus of freely moving rats, Brain Res. Bull., 13 (1984) 39-42. 18 Kuhnt, U. and Mortasawi, A., Are processes activated by
19
20
21
22
23
24
25
26
27
28
29
30 31
calmodulin directly related to the generation of long-term potentiation in the hippocampus?, Neurosci. Lett., 22 (1985) $49. Kuo, J.F., Schatzman, R.C., Turner, R.S. and Mazzei, G.J., Phospholipid-sensitive Ca-dependent protein kinase: a major protein phosphorylation system, Mol. Cell. Endocrinol., 35 (1985) 65-73. Lovinger, D.M., Wong, K.L., Murakami, K. and Routtenberg, A., Protein kinase C inhibitors eliminate hippocampal long-term potentiation, Brain Research, 436 (1987) 177-183. Lynch, M.A. and Bliss, T.V.P., Long-term potentiation of synaptic transmission in the hippocampus of the rat: effect of calmodulin and oleoyl-acetyl-glycerol on release of 3H glutamate, Neurosci. Lett., 65 (1986) 171-176. Lynch, G., Larson, J., Kelso, S., Barrionuevo, G. and Schottler, F., Intracellular injections of EGTA block induction of hippocampal long-term potentiation, Nature (Lond.), 305 (1983) 719-721. MacDermott, A.B., Mayer, M.L., Westbrook, G.L., Smith, S.J. and Barker, J.L., NMDA-receptor activation increases cytoplasmic calcium concentrations in cultured spinal cord neurons, Nature (Lond.), 321 (1986) 519-522. Malenka, R.C., Madison, D.V. and Nicoll, R.A., Potentiation of synaptic transmission in the hippocampus by phorbol esters, Nature (Lond.), 321 (1986) 175-177. Mihaly, A., Kuhnt, U. and Joo, F., NMDA-receptor antagonists inhibit Ca2+-retention in CA1 of the hippocampus during tetanic stimulation, Neurosci. Len., Suppl. 26 (1986) $534. Moody, I., Baimbridge, K.G. and Miller, J.J., Blockade of tetanic- and calcium-induced potentiation in the hippocampal slice preparation by neuroleptics, Neuropharmacology, 23 (1984) 625-631. Popov, N., Reymann, K.G., Schulzeck, K., Schulzeck, S. and Matthies, H., Alterations in calmodulin content in fractions of rat hippocampal slices during tetanic and calcium induced long-term potentiation, Brain Res. Bull., in press. Reymann, K.G., Frey, U., Jork, R. and Matthies, H., Polymyxin B, an inhibitor of protein kinase C, prevents the maintenance of synaptic long-term potentiation in hippocampal CA 1 neurons, Brain Research, 440 (1988) 305-314. Reymann, K.G., Schulzeck, K., Kase, K. and Matthies, H., Phorbol ester-induced hippocampal long-term potentiation is counteracted by inhibitors of protein kinase C, Exp. Brain Res., 71 (1988) 227-230. Teyler, T.J. and DiScenna, P., Long-term potentiation, Annu. Rev. Neurosci., 10 (1987) 131-161. Van Belle, H., R 24571: a potent inhibitor of calmodulin activated enzymes, Cell Calcium, 2 (1981) 483-494.
388
Elsevier BRE 23134
Inhibitors of calmodulin and protein kinase C block different phases of hippocampal long-term potentiation Klaus G. Reymann 1, Rudolf Br6demann 2, Hiroshi Kase 3 and Hansjfirgen Matthies 1 1Institute of Neurobiology and Brain Research, Academy of Sciences G.D.R., Magdeburg (G. D.R.), "Instituteof Pharmacology and Toxicology, MedicalAcademy Magdeburg, Magdeburg (G.D.R.) and 3Tokyo Research Laboratories, Kyowa Hakko Kogyo Co., Ltd., Tokyo (Japan) (Accepted 21 June 1988)
Key words: Long-term potentiation; Hippocampus; Synaptic plasticity; Calmodulin; Protein kinase C; Second messenger; Calmidazolium; K-252b; Protein phosphorylation
The effects of a calmodulin (CAM) inhibitor, which does not influence Ca2+ fluxes (calmidazolium, RO-24571), and a new potent inhibitor of protein kinase C (K-252b) on long-term potentiation (LTP) were compared in hippocampal slices. Tetanic stimulation of the stratum radiatum during perfusion of calmidazolium (50 nM) failed to induce the characteristic post-tetanic and long-term increase in the magnitude of CAl-evoked responses. During perfusion with K-252b (50 nM) post-tetanic potentiation and initial LTP is expressed normally, but thereafter declines back to baseline with a 60 rain delay. By themselves, the inhibitors had no significant effect on synaptic transmission in a non-tetanized control input. Our data are in line with current evidence from several laboratories that CaM- and protein kinase C (PKC)-dependent processes are involved in LTP and support the hypothesis that CaM mediates initiation and that PKC mediates mechanisms underlying the maintenance of LTP.
Long-term potentiation (LTP) is a particular type of synaptic plasticity which has several features that make it an interesting m o d e l of information storage 3'3°. Our present knowledge of intracellular signaling systems whereby a short tetanization produces long-term e n h a n c e m e n t of a monosynaptic connection is, however, far from being satisfactory. The purpose of the present study was to investigate the ability of calmodulin (CAM) and protein kinase C (PKC) inhibitors to antagonize either the initiation or the maintenance of LTP. The induction of h i p p o c a m p a l long-term potentiation (LTP) was shown to be correlated by a translocation of the multifunctional, calcium binding regulatory protein calmodulin from cytosolic c o m p a r t m e n t s towards the m e m b r a n e 27 and to be susceptible to disruption by C a M inhibitors s'9'26. Thus, the observed inhibition of LTP by trifluoperazine was interpreted as supporting the role of C a M in mechanisms of synaptic plasticity8'9'26. H o w e v e r , trifluoperazine is also
a potent inhibitor of Ca 2+ influx 12, an ionic event which is crucial for the initiation of LTP 7'14'22. Moreover, others did not find LTP-blocking e f f e c t s of C a M inhibitors is. W e tested here a C a M inhibitor tl,31 which, in low concentration, has no influence on calcium fluxes and P K C activity ~2,t9. Recently, the Ca2+/phospholipid-dependent P K C has been shown also to be involved in mechanisms of LTP 1"4~13'2°'24'2s'29. It seems unlikely that trifluoperazine has exhibited its action in previous experiments 8'9'26 only via inhibition of this kinase. Inhibitors of PKC have been shown to influence LTP in a different way 2°,2s. A l t h o u g h the P K C inhibitor polymyxin B blocked LTP like trifluoperazine, it neither affected the post-tetanic nor the early stage of hippocampal LTP. To substantiate the assumption that PKC actually mediates only the maintenance but not the initiation of LTP we tested the new potent P K C inhibitor K-252b 15, known to be effective in blocking the phorbol ester-induced type of LTP 29. A l t o g e t h e r ,
Correspondence: K.G. Reymann, Institute of Neurobiology and Brain Research, Academy of Sciences G.D.R., Leipziger Strasse 44, DDR-3090 Magdeburg, G.D.R. 11006-8993/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
389 the aim of the present work was to distinguish more precisely the interval in which CaM- and PKC-dependent processes are involved in LTP using an identical testing paradigm for the specific inhibitors. Transverse hippocampal slices were prepared from 7-week-old Wistar rats as previously described 28 with the slices superfused at 35 °C with a medium containing (mM): NaC1 124, KCI 4.9, KH2PO 4 1.2, MgSO4 1.3, CaCI 2 2.5, NaHCO 3 25.6, D-glucose 10. Stimulating and recording electrodes were etched and isolated stainless-steel wires. Extracellular recordings were obtained from the stratum radiatum (population-excitatory postsynaptic potential (EPSP)) and the stratum pyramidale (population spike) of CA1. Biphasic, constant-current pulses (0.2 Hz, 100/~s each half cycle) were delivered to two separate stimulation electrodes positioned on each side of the point of EPSP recording. The electrode on the CA3 side was subsequently used for tetanization (100 Hz/1 s at test stimulation strength), whereas the electrode on the subicular side served as an independent control input 28. Test responses were recorded at a stimulation strength which elicited a population spike (PS) of 30% of its maximum amplitude. Response changes were expressed as the average percentage change of EPSP slope and PS amplitude from the baseline obtained before drug treatment and tetanization. Comparisons between control and drugtreated groups were made with the two-tailed Mann-
Whitney U-test for unpaired observations (P < 0.05). Calmidazolium ll'31 (Boehringer) and K-252b (Na salt) 15 (Kyowa Hakko Kogyo) were first dissolved in dimethylsulfoxide and then added to the perfusion 4030C "¢~
10-Ck O O. (/)
calmidazol.
K-252b
0--10--
0. LU
L~ control [] K - 2 5 2 b o calrnidazoleum
50nM
I 0
I 30
T
100-
tJ
=
i
50-
E 0- ~ -
A I 0
I 30
TIME
/'\. 10.min
90.min
_J Fig. 1. Representative CA1 population spike responses to synaptic activation before and after a 100 Hz/1 s tetanic shock in the presence of the CaM inhibitor calmidazolium (50 nM) or the PKC inhibitor K-252b (50 nM). The computer plots are averages of 4 potentials evoked by low frequency test stimulation. Calibration: 2 mV/5 ms, but 1 mV for the calmidazoliumtreated slice.
I 90 100 Hz
100 Hz
before
[ 60
100 Hz
et control
20-
I 60
I 90 100 Hz
(minutes)
Fig. 2. Changes in the potentiation of population EPSP and population spike from the tetanized input (control n = 8, triangles) seen after 50 nM bath application of calmidazolium and K-252b. In the groups pretreated with calmidazolium (n = 8, circles) no post-tetanic and long-term potentiation occurred. Low concentrations of the PKC-inhibitor K-252b, however, significantly blocked only later stages of LTP (n = 11, squares). The EPSP magnitude was evaluated by measuring the initial, rising slope (mV/ms) of the EPSP between the onset and the maximum. The PS amplitude was evaluated by taking the voltage difference between the peak negativity and the preceding positive peak. The data shown are plotted as an average of the percentage change from baseline responses obtained immediately before the inhibitor application (means across slices, each from a separate animal; S.E.M. is shown at the 30th and the 60th min). Filled symbols indicate a significant change (P < 0.05) in comparison to control experiments. The difference between EPSPs of both drug-treated groups is also significant for 45 min after tetanization.
390 line resulting in a final bath concentration of the solvent <0.01%. Both inhibitors were applied 7 min before the tetanic shock and washed 20 min later. The vehicle alone had no effect on parameters of the present study. A short tetanization led to a long-term enhancement in PS in 8 of 8 slices (EPSP 7 of 8 slices) (Figs. 1 and 2). As was shown previously (for review see refs. 10, 28), the tetanization in one input did not cause any significant change in a separate control input within the first hours after tetanization. During this study, the deviation from baseline in the non-tetanized control input was less than 12 _+ 6% for the EPSP slope and less than 16 _+ 6% for the PS amplitude. Neither compound used here influenced the parameters from this control input. Therefore, the drug effects described below are due to their specific influence on LTP-related processes. The corresponding data from the non-tetanized input are not shown because of the high degree of overlapping of the values from the 3 groups under investigation (maximum deviation between EPSP values 12 _+ 8%; between PS values 32 + 16%). At a higher concentration (500 nM, one slice) calmidazolium reversibly blocked synaptic transmission by about 50%. Preincubation of the slices with 50 nM calmidazolium clearly blocked post-tetanic (1 min) and long-term potentiation (>10 min) in 7 of 8 slices (Figs. 1 and 2). LTP did not recover long after calmidazolium was washed from the chamber. By itself, calmidazolium has only a non-significant late depressive effect on the PS amplitude of the control input (-32 + 16% at 90 min), which cannot explain the clear and permanent prevention of LTP during this study. Some remaining potentiation of the PS may be due to an additional spike potentiating mechanism (E/S left-shift; for explanation see ref. 28), which seems to be not sensitive to calmidazolium. The data presented in Figs. 1 and 2 show that K252b bath-applied at a concentration of 50 nM, which blocks phorbol ester-induced potentiation 29, does not substantially inhibit synaptically evoked post-tetanic and the initial part of long-term potentiation in CA1 hippocampal neurons. An EPSP potentiation >10% was seen in 10 of 11 slices 10 min after tetanization. After some delay, however, LTP of synaptic responses returned to baseline (increase in EPSP >10% in 3 of 11 slices after 90 rain). The PS ampli-
tude was similarly decreased after 60 min. However, as was the case with calmidazolium and polymyxin B2s, some potentiation of the PS remained. In one slice 500 nM K-252b blocked LTP already after 20 min. The small, mostly non-significant early effect of K-252b may be due to an effect on CaM- or cAMPdependent kinases 15 (Kase et al., unpublished data). The effect of both inhibitors was only partially reversible within the duration of the experiment (Fig. 2). A second tetanization at the end of the experiment was followed by a small PS potentiation (calmidazotium group = + 16% at the 1st min with respect to corresponding pretetanic values; K-252b group = +15% and +10% at the 1st and the 15th min after the train, respectively; P < 0.05). In addition, we have measured changes in PS peak latency. The typical shortening of the peak latency following the high-frequency train (0.5 + 0.1 ms decrease up to the 90th min in the control group) was partially antagonized by both inhibitors (calmidazolium: 0.2 + 0.1 ms decrease at the 1st min; K-252b: 0.2 + 0.1 ms decrease at the 90th rain; P < 0.05). The most likely source of calcium necessary for activation of CaM is the influx of extracellular calcium through NMDA-receptor-linked ion channels 7':3'25. Additional sources of free cytosolic calcium might be its influx through ion channels sensitive to Ca 2+ antagonists 14 or its transient release from intracellular stores following the receptor-mediated stimulation of phosphoinositol turnover (cf. ref. 5). The present findings deliver first evidence that a CaM antagonist, which does not influence Ca 2+ fluxes in other neuronal systems 12and which has no affinity to dopamine receptor sites (Van Belle 11) in the concentrations used here, is capable of blocking post-tetanic potentiation and LTP. This is in accordance with similar studies in which less selective CaM antagonists prevented LTP 8"9'26and confirms the hypothesis that besides the Ca 2+ influx also CaM activation is a necessary prerequisite for LTP. Under certain circumstances, post-tetanic potentiation and LTP can independently occur and probably represent separate phenomena 3°. Our data, however, demonstrate the dependence of both phenomena on CaM. There are various CaM-dependent events in preand postsynaptic compartments. Recently it was demonstrated that both tetanization and calmodulin application enhance K+-induced glutamate release 21.
391 Further studies will be of interest to determine the CaM-stimulated target enzymes and their substrates which are important for LTP. In this field, a crucial role can be assumed for the Ca2+/CaM kinase II which is believed to constitute a major fraction of postsynaptic density proteins 16. Another possibility is a 40 kDa protein which is among the proteins phosphorylated during LTP the most susceptible to CaM and its inhibitors 9. It is important to note here that a significant inhibition of LTP by K-252b occurred long after its induction. This confirms a recent multiple phase hypothesis of LTP based on studies in which other types of PKC inhibitors with a different inhibition profile were used 2°'2s. According to our hypothesis 28 an intermediate and late stage of LTP is susceptible to PKC inhibitors and only the late stage is susceptible to the protein synthesis inhibitor anisomycin (>5 h) t°. We referred to these phases as LTP 2 and LTP 3(ref. 28). Because synthesis and dendritic transport of macromolecules are relatively slow, there may be a need for a distinct form of intermediate LTP to bridge the short- and long-term processes, similarly as during memory formation. As far as PKC's role in maintenance of LTP decreases with time after the induction 2s, it seems necessary that PKC and perhaps CaM kinases induce or stabilize late biochemical or even structural changes that have been observed following LTP (for review see ref. 30) rather than mediate the initial production of LTP (LTP 1). Both pre21,24 and postsynaptic 4'13"29PKC-dependent processes have been suggested to be involved. A recent prelimi n a r y observation has demonstrated a delayed increase in sensitivity of quisqualate receptors during LTP 6. The similar time course indicates the interesting possibility that PKC may be involved especially in
1 Akers, R.F., Lovinger, D.M., Colley, P.A., Linden, D.J. and Routtenberg, A., Translocation of protein kinase C activity may mediate hippocampal long-term potentiation, Science, 231 (1986) 587-589. 2 Alkon, D.L., Kubota, M., Neary, J.T., Naito, S., Coulter, D. and Rasmussen, H., C-kinase activation prolongs Ca2+dependent inactivation of K currents, Biochem. Biophys. Res. Commun., 134 (1986) 1245-1253. 3 Andersen, P., Long-term potentiation - - outstanding problems. In J.-P. Changeux and M. Konoshi (Eds.), The Neu-
this mechanism. The relationship between CaM inhibition and late phases of LTP is not yet clear. Although calmidazolium blocks LTP immediately and irreversibly it cannot be ruled out that CaM-dependent processes are also involved in intermediate and late mechanisms. The low concentration of calmidazolium used in this study may exclude any effect on late PKC-dependent processes 19. Following the initial translocation of calmodulin towards the membrane an opposite translocation towards the cytosol occurred 1 h after tetanization in hippocampus/dentate slices under similar experimental conditions as used here 27, indicating the possible existence of a second CaM-dependent process. Therefore, the lack of recovery after washout of calmidazolium could mean that these two distinct second messenger systems (CaM and PKC) converge at some level to enable post-translational modifications (for review see ref. 1) and/or synthesis of LTP-relevant proteins (for review see refs. 10, 17). Such a synergistic interaction of PKC with Ca2+/CaM kinase was described for the Ca2+-mediated K + current reduction occurring during conditioning in Hermissenda. Interestingly, similarly as in the LTP model PKC activation prolongs the effect 2. In line with the interpretation in this communication is the recently described increased hydrolysis of phosphatidylinositol-4,5-biphosphate in a postsynaptic compartment 45 min after tetanization (M.A. Lynch et al., Neurosci. Lett., 84 (1988) 291-296). Further studies will reveal whether both enzymes also share a common substrate during LTP. We thank Karin Schulzeck, Rosemarie Schlichting and Steffi Roth for expert assistance and Reinhard Jork for constructive comments on the manuscript.
ral and Molecular Bases of Learning, Wiley, New York,
1987, pp. 239-269. Andersen, P., Hvalby, O. and Reymann, K., Postsynaptic mechanisms contribute to synaptic potentiation induced by phorbol esters in rat hippocampus pyramidal cells in vitro, J. Physiol. (Lond.), 398 (1988)35P. B/~r, P.R., Wiegant, F,, Lopes da Silva, F.H. and Gispen, W., Tetanic stimulation affects the metabolism of phosphoinositides in hippocampal slices, Brain Research, 321 (1984) 381-385.
392 6 Collingridge, G.L. and Lester, R.A.J., The sensitivity of CA1 neurones to quisqualate following high frequency stimulation in rat hippocampus in vitro, J. Physiol. (Lond.), 398 (1988) 22P. 7 Dingledine, R., N-methylaspartate activates voltage-dependent calcium conductance in rat hippocampal pyramidal cells, J. Physiol. (Lond.), 343 (1983) 385-405. 8 Dunwiddie, T.V., Roberson, N.L. and Worth, T., Modulation of long-term potentiation: effects of adrenergic and neuroleptic drugs, Pharmacol. Biochem. Behav., 17 (1982) 1257-1264. 9 Finn, R.C., Browning, M. and Lynch, G., Trifluoperazine inhibits hippocampal long-term potentiation and the phosphorylation of a 40,000 dalton protein, Neurosci. Lett., 19 (1980) 103-108. 10 Frey, U., Krug, M., Reymann, K. and Matthies, H., Anisomycin, an inhibitor of protein synthesis, blocks late phases of LTP-phenomena in the hippocampal CA1 region in vitro, Brain Research, 452 (1988) 57-65. 11 Gietzen, K., Wiithrich, A. and Bader, H., R 24571: a new powerful inhibitor of red blood cell Ca2÷-transport ATPase and of calmodulin-regulated functions, Biochem. Biophys. Res. Commun., 101 (1981)418-425. 12 Greenberg, D.A., Carpenter, C.L. and Messing, R.O., Interaction of calmodulin inhibitors and protein kinase C inhibitors with voltage-dependent calcium channels, Brain Research, 404 (1987) 401-404. 13 Hu, G.-Y., Hvalby, O., Waalas, S.I., Albert, K.A., Skjef1o, P., Andersen, P. and Greengard, P., Protein kinase C injection into hippocampal pyramidal cells elicits features of long-term potentiation, Nature (Lond.), 328 (1987) 426-429. 14 Izumi, Y., Ito, K., Miyakawa, H. and Kato, H., Requirement of extracellular Ca 2÷ after tetanus for induction of long-term potentiation in guinea pig hippocampal slices, Neurosci. Lett., 77 (1987) 176-180. 15 Kase, H., Iwahashi, K., Nakanishi, S., Matsuda, Y., Yamada, K., Takahashi, M., Murakata, C., Sato, A. and Kaneko, M., K-252 compounds, novel and potent inhibitors of protein kinase C and cyclic nucleotide-dependent protein kinases, Biochem. Biophys. Res. Commun., 142 (1987) 436-440. 16 Kelly, P.T., McGuinness, T.L. and Greengard, P., Evidence that the major postsynaptic density protein is a component of a Ca2+/calmodulin-dependent protein kinase, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 945-949. 17 Krug, M., Loessner, B. and Ott, T., Anisomycine blocks the late phase of long-term potentiation in the dentate gyrus of freely moving rats, Brain Res. Bull., 13 (1984) 39-42. 18 Kuhnt, U. and Mortasawi, A., Are processes activated by
19
20
21
22
23
24
25
26
27
28
29
30 31
calmodulin directly related to the generation of long-term potentiation in the hippocampus?, Neurosci. Lett., 22 (1985) $49. Kuo, J.F., Schatzman, R.C., Turner, R.S. and Mazzei, G.J., Phospholipid-sensitive Ca-dependent protein kinase: a major protein phosphorylation system, Mol. Cell. Endocrinol., 35 (1985) 65-73. Lovinger, D.M., Wong, K.L., Murakami, K. and Routtenberg, A., Protein kinase C inhibitors eliminate hippocampal long-term potentiation, Brain Research, 436 (1987) 177-183. Lynch, M.A. and Bliss, T.V.P., Long-term potentiation of synaptic transmission in the hippocampus of the rat: effect of calmodulin and oleoyl-acetyl-glycerol on release of 3H glutamate, Neurosci. Lett., 65 (1986) 171-176. Lynch, G., Larson, J., Kelso, S., Barrionuevo, G. and Schottler, F., Intracellular injections of EGTA block induction of hippocampal long-term potentiation, Nature (Lond.), 305 (1983) 719-721. MacDermott, A.B., Mayer, M.L., Westbrook, G.L., Smith, S.J. and Barker, J.L., NMDA-receptor activation increases cytoplasmic calcium concentrations in cultured spinal cord neurons, Nature (Lond.), 321 (1986) 519-522. Malenka, R.C., Madison, D.V. and Nicoll, R.A., Potentiation of synaptic transmission in the hippocampus by phorbol esters, Nature (Lond.), 321 (1986) 175-177. Mihaly, A., Kuhnt, U. and Joo, F., NMDA-receptor antagonists inhibit Ca2+-retention in CA1 of the hippocampus during tetanic stimulation, Neurosci. Len., Suppl. 26 (1986) $534. Moody, I., Baimbridge, K.G. and Miller, J.J., Blockade of tetanic- and calcium-induced potentiation in the hippocampal slice preparation by neuroleptics, Neuropharmacology, 23 (1984) 625-631. Popov, N., Reymann, K.G., Schulzeck, K., Schulzeck, S. and Matthies, H., Alterations in calmodulin content in fractions of rat hippocampal slices during tetanic and calcium induced long-term potentiation, Brain Res. Bull., in press. Reymann, K.G., Frey, U., Jork, R. and Matthies, H., Polymyxin B, an inhibitor of protein kinase C, prevents the maintenance of synaptic long-term potentiation in hippocampal CA 1 neurons, Brain Research, 440 (1988) 305-314. Reymann, K.G., Schulzeck, K., Kase, K. and Matthies, H., Phorbol ester-induced hippocampal long-term potentiation is counteracted by inhibitors of protein kinase C, Exp. Brain Res., 71 (1988) 227-230. Teyler, T.J. and DiScenna, P., Long-term potentiation, Annu. Rev. Neurosci., 10 (1987) 131-161. Van Belle, H., R 24571: a potent inhibitor of calmodulin activated enzymes, Cell Calcium, 2 (1981) 483-494.