Brain Research 782Ž. 1998 337±342

Short communication Phenidone blocks the increases of proenkephalin and gene expression induced by kainic acid in rat hippocampus: involvement of Fos-related antigene protein

Hyoung-Chun Kim a,), Jeong-Hye Suh a, Je-Seong Won b, Wang-Kee Jhoo a, Dong-Keun Song b, Yung-Hi Kim b, Myung-Bok Wie b, Hong-Won Suh b a College of Pharmacy, Kangwon National UniÕersity, Chunchon 200-701, Kangwon-Do, South Korea b Department of Pharmacology, Institute of Natural Medicine, College of Medicine, Hallym UniÕersity, Chunchon 200-701, Kangwon-Do, South Korea Accepted 11 November 1997

Abstract

To determine the possible role of cyclooxygenaserlipoxygenase pathway in the regulation of proenkephalinŽ. proENK and prodynorphinŽ. proDYN gene expression induced by kainic acid Ž. KA in rat hippocampus, the effects of esculetin, aspirin, or phenidone on the seizure activity, proENK and proDYN mRNA levels, and the level of fos-related antigeneŽ. Fra protein induced by KA in rat hippocampus were studied. EsculetinŽ.Ž 5 mgrkg , aspirin 15 mgrkg . , or phenidone Ž 50 mgrkg . was administered orally five times every 12 h before the injection of KAŽ. 10 mgrkg, i.p. . Seizure activity induced by KA was significantly attenuated by phenidone. However, neither esculetin nor aspirin affected KA-induced seizure activity. The proENK and proDYN mRNA levels were markedly increased 4 and 24 h after KA administration. The elevations of both proENK and proDYN mRNA levels induced by KA were inhibited by pre-administration with phenidone, but not with esculetin and aspirin. ProENK-like protein level increased by KA administration was also inhibited by pre-administration with phenidone, but not with esculetin and aspirin. The increases of proENK and proDYN mRNA levels induced by KA were well correlated with the increases of Fra protein level. Additionally, the induction of Fra protein was inhibited by pre-administration with phenidone, but not with esculetin and aspirin. The results suggest that blockade of both cyclooxygenase and lipoxygenase pathways appears to be responsible for increases of proENK and proDYN mRNA levels induced by KA via inhibiting the induction of Fra protein in rat hippocampus. q 1998 Elsevier Science B.V.

Keywords: Kainic acid; Cyclooxygenase; Lipoxygenase; Proenkephalin; Prodynorphin; Fra protein; Hippocampus

Kainic acidŽ. KA , a unique neuroexcitatoryrneurotoxic cortexwx 13,19 . In addition, the increases of c-Fos, Fra, and substance, produces seizure and neural degeneration in the Jun proteins induced by convulsant dose of KA correlate mammalian CNS, and appears to provide a good model for with AP-1 DNA binding activitywx 18,19 . The neurons of some aspects of human temporal lobe epilepsywx 3,17 . In the hippocampus express both proENK and proDYN the rat hippocampus, KA receptors are concentrated in the mRNA and their protein productswx 11,15 . These proENK CA3rCA4 regions, with moderate levels in the other areas and proDYN mRNA levels are also elevated by KAwx 9 . of the hippocampuswx 8,12 . Both proENK and proDYN genes contain an AP-1-like Several types of proto-oncogene product, such as Fos domain in their promoter regionswx 7,10 . and Jun family proteins, serve as the third messengers in Cis-unsaturated fatty acids can be liberated by phospho- wx the regulation of various types of genes 4,14 . These lipase A2 or diacylglycerol lipase and serve as a precursor, proteins interact with AP-1 domain in the promoter of and subsequently metabolized further via the action of various types of genes, in turn, modulate the transcrip- cyclooxygenase and lipoxygenasewx 5 . Recently, Suh et al. tional activitywx 16 . Convulsant doses of KA induce c-Fos, wx21,22 have reported that arachidonic acid or prosta-

Fra, and Jun proteins in rat hippocampus and entorhinal glandin E2 increases proENK mRNA levels in bovine chromaffin cells. However, the involvement of cyclooxy- genase and lipoxygenase pathways in KA-induced increase ) Corresponding author. Fax: q82-361-55-7865. of mRNA in the hippocampus has not been

0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S0006-8993Ž. 97 01401-7 338 H.-C. Kim et al.rBrain Research 782() 1998 337±342 characterized. In an attempt to examine the role of cyclo- sarcosine, 2% blocking reagent. , and hybridized overnight oxygenaserlipoxygenase in the regulation of proENK and in Dig-labeled proenkephalin and prodynorphin probes proDYN mRNA expression, the effects of phenidone, es- containing prehybridization buffer at 688C. Then mem- culetin, and aspirin on KA-induced proENK and proDYN branes were washed twice in 2= wash solutionŽ 2= SSC, mRNA levels, or proENK and Fra proteins were studied in 0.1% SDS. at room temperature, and further washed in the present study. 0.1= wash solutionŽ. 0.1= SSC, 0.1% SDS at 688C for 1 Male Sprague±Dawley rats weighing 200±280 g were h. After equilibrating in a buffer IŽ 100 mM maleic acid, used. Animals were housed two per group in a room 150 mM NaCl, pH 7.5. for 1 min, the membranes were maintained at 22"58C with an alternating 12 h light±dark gently agitated in buffer IIŽ 1% blocking reagent in buffer cycle. Animals were used only once. EsculetinŽ. 5 mgrkg , I. for 30±60 min and immuno reacted with the diluted aspirinŽ. 15 mgrkg , or phenidone Ž. 50 mgrkg was admin- anti-DIG-alkaline phosphatasew 1: 10,000Ž. 75 mUrmlx in istered orally five times every 12 h. Thirty minutes after a buffer II for 30 min. After washing twice for 15 min per the last administration of each drug, KAŽ. 10 mgrkg was wash in 0.3% Tween 20Ž. in buffer I , the membranes were administered intraperitoneally. Then, animals were sacri- equilibrated in a buffer IIIŽ 100 mM Tris±HCl, pH 9.5, ficed 4 and 24 h after KA administration. 100 mM NaCl, 50 mM MgCl2 . for 2 min and added 1:100 Seizure activity was rated during a 4-h period after the diluted CSPDŽ. Boehringer Mannheim, Germany solution, KA challenge according to the scale devised by Baran et a chemi-luminescent substrate solution, in buffer III. After al.wx 1 . The seizure rating scale were as follows: 0; normal, incubation at 378C for 15±20 min, the membranes were rare wet-dog shakesŽ. WDS , no convulsions, 1; intermedi- exposed to Hyperfilm-MPŽ Amersham, Buckinghamshire, ate numbers of WDS, rare convulsion type A, 2; frequent England. . For re-hybridization, membranes were washed WDS, frequent type A convulsionsŽ no rearing or salva- for 20 min at room temperature in sterile millepore water, tion. , 3; frequent WDS, frequent type A convulsion, ap- then further washed for overnight at 658Cin50mM pearance of type B convulsions with rearingŽ but without Tris±HCl, pH 8.0, 50% dimethylformamide and 1% SDS, falling. , salivation, 4; WDS, type A convulsions, frequent and re-hybridized to Dig-labeled rat cyclophilin cRNA type B convulsions with falling, salivation, 5; continuous probe as a internal loading control. generalized limbic seizures, death within 3 h. Type A All data shown in the present study were repeated at convulsions: focal convulsions affecting head and extremi- least three-times and showed the same tendency of results. ties, starting 30 min after KA injection. Type B convul- No animal received phenidone, esculetin or aspirin alone sions: frequent generalized convulsions usually starting exhibited seizure behaviorŽ. data not shown . The rats 60±90 min after KA injection. Surviving rats expressed receiving KAŽ. 10 mgrkg manifested robust behavioral recurrent seizure behavior. seizures lasting 4 to 6 hŽ. Fig. 1A . Increased seizure scores Animals from each group were anesthetized with 50 induced by KA were significantly attenuated by the pre- mgrkg pentobarbital and then perfused transcardially with treatment with phenidoneŽ. Fig. 1A . However, neither a 140 ml syringe containing salineŽ 40 mlr100 g body esculetin nor aspirin affected KA-induced responseŽ Fig. weight.Ž followed by 4% paraformaldehyde 70 mlr100 g 1A. . body weight. . The brains were then removed, stored in 4% Both hippocampal proENK and proDYN mRNA levels paraformaldehyde overnight and then cut at 40 mm in the were not altered in the animals treated with saline, horizontal plane with a vibratome. Prior to overnight incu- phenidone, esculetin or aspirin aloneŽ. Fig. 1B . Both bation in primary antibody, sections were prewashed in proENK and proDYN mRNA levels were remarkably in- 0.2% Triton X-100 for 15 min, followed by 4% normal creased 4 and 24 h after KA administration. Six hours after goat serum for 20 min. After a 24-h incubation with the KA injection, in all nonseizing rats given KA plus primary Fra antiseraŽ. 1:1000 dilution or proENK antisera phenidone, esculetin or aspirin, the levels of proENK and Ž.1:1000 dilution , sections were then incubated with a proDYN mRNA was sightly increased. However, these secondary biotinylated antiserumŽ. 1:500 dilution for 1 h. levels were returned to the basal level after 24 h after KA Sections were always washed three times with PBSŽ pH administrationŽ. Fig. 1B . Seizing rats pretreated with 7.4. between each incubation step. As the chromogen, phenidone, but not esculetin and aspirin, exhibited de- X 3.3 -diaminobenzidine was used. creases in both proENK and proDYN mRNA levels at all Total cellular RNA and proteins were extracted from time pointsŽ. Fig. 1B . pooled rat hippocampus Ž.ns3rgroup using a rapid Hippocampal proENK-like immunoreactivity or Fra-im- guanidine thiocyanate±water saturated phenolrchloroform munoreactivity was not affected by the treatment with extraction and subsequent precipitation with acidic sodium phenidone, esculetin or aspirin aloneŽ. data not shown . As acetatewx 6 . A total of 10 mg of total RNA were elec- shown in Figs. 2 and 3, KA significantly increased proENK trophoresed and transferred on nylon membrane. After and Fra proteins, especially in the dentate gyrus area in the baking for 1±2 h at 808C, the membranes were pre-hy- hippocampus. The increases of proENK and Fra proteins bridized at 688C for 1 h in pre-hybridization bufferŽ 5= induced by KA was significantly attenuated by the pre- SSC, 50% formamide, 0.02% SDS, 0.1% sodium N-lauroyl treatment with phenidone. However, neither esculetin nor H.-C. Kim et al.rBrain Research 782() 1998 337±342 339

Fig. 1.Ž. A Effects of phenidone, esculetin and aspirin on seizure activity induced by kainic acidŽ.Ž. KA , B proENK and proDYN mRNA levels induced by KA in seizingŽ. S and nonseizing Ž NS . rats. Ž. C Rats were orally administered with either saline, phenidone Ž P, 50 mgrkg . , esculetin Ž E, 10 mgrkg . , or aspirinŽ. A, 15 mgrkg five times every 12 h. Thirty minutes after the last treatment of each compound, either PBS or 10 mgrkg of KA was administered intraperitoneally. Seizure activities were scored according to Baran et al.wx 1 . The statistical significance of the changes in seizure activity was determined by the One-way ANOVA with post-hoc. Each value is the mean"S.E.M. ) p-0.05 compared with group of rats injected with KA alone. nsNumber of animals used in each group. For measuring proENK and proDYN mRNA levels, the animals were sacrificed at indicated time pointsŽ. 4 and 24 h . Total RNA of 10 mg, which were extracted from pooled rat hippocampus Ž.ns3rgroup using a guanidium thiocyanaterphenolrchloroform gradient method, were used for determination of proENK and proDYN mRNA levels. The unregulated mRNA encoding cyclophilin was used as an internal loading control. NonseizingŽ. NS group indicates only the animals with a score of `0'. aspirin affected KA-induced increases of proENK and Fra 4 h and 24 h after KA administration. Previous studies protein levelsŽ. Figs. 2 and 3 . have shown that dentate gyrus of hippocampus expresses The present study demonstrates that KA increases proENK and proDYN gene, or their products, and KA proENK-like protein and proDYN mRNA levels in rat administration elevates both proENK and proDYN mRNA hippocampus. Co-blockade of cyclooxygenase and lipoxy- levels in the dentate gyrus of hippocampuswx 11,15 . The genase by phenidone was required for producing ant- results of the present study were partially in line with agonistic effect against KA-induced responses. Further- previous studies which showed that both proENK and more, inhibition of Fra protein induced by KA by proDYN mRNA levels begin to increase 3 or 4 h after KA phenidone was well correlated with opioid peptides gene administrationwx 9,19 . Previous works have demonstrated expression or proENK-like protein expression, suggesting that convulsant doses of KA causes the inductions of that the blocking effect of phenidone against KA-induced immunoreactivities of Fos and Jun-related proteins in increases of proENK and proDYN mRNA level appears to mouse and rat hippocampuswx 19,20 . In a recent study, we be mediated by reducing the expression of Fra protein, and others have shown that Fra protein was induced after which binds to AP-1 regulatory domain of proENK and KA administrationwx 19,23 . In the present study, we found proDYN genes. that the KA-induced increase of Fra protein is significantly In Northern blot analyses, the levels of proENK and inhibited by the pre-administration with phenidone, but not proDYN mRNA expression were markedly increased after esculetin and aspirin at all time points, suggesting that the 340 H.-C. Kim et al.rBrain Research 782() 1998 337±342

Fig. 2. The representative photomicrographs of immunostained sections with antibody against proENK-like protein 24 h after KA administration. Hippocampal sections of controlŽ. A revealed low level of proENK-like immunostaining. Dense immunostaining for proENK-like protein was visualized in the dentate gyrusŽ. DG after KA injection Ž. B . Seizing KA-treated rats with 50 mgrkg of phenidone showed clearly decreased proENK-like immunostaining in the DGŽ. C . ProENK staining in the hippocampus from seizing KA-treated rats with 50 mgrkg of esculetinŽ. D or aspirin Ž. E was still apparent in the neuronal layer of the DG. Scale bars150 mm. The intensity of proENK-like immunoreactivity was semiquantitatively graded as intense Ž.grade 3 , moderate Ž.Ž. grade 2 , weak grade 1 , very weak Ž grade 0.5 . and not detectable Ž. grade 0 . nsNumber of animals used in each group. Each value is the mean"S.E.M. ) p-0.05 vs. controlŽ. A , ap-0.05 vs. KA-treated rat Ž. B or the seizing rat pretreated with esculetin Ž. D or aspirin Ž. E . H.-C. Kim et al.rBrain Research 782() 1998 337±342 341

Fig. 3. The representative photomicrographs of immunostained sections with antibody against Fra protein 24 h after KA administration. Hippocampal sections of controlŽ. A revealed very little Fra immunostaining. Dense immunostaining for Fra protein was visualized in the dentate gyrusŽ. DG after KA injectionŽ. B . Seizing KA-treated rats with 50 mgrkg of phenidone showed clearly decreased Fra immunostaining in the DGŽ. C . Fra nuclear staining in the hippocampus from seizing KA-treated rats with 50 mgrkg of esculetinŽ. D or aspirin Ž. E was still apparent in the neuronal layer of the DG. Scale bars150 mm. The intensity of Fra immunoreactivity was semiquantitatively graded as intenseŽ. grade 3 , moderate Ž.Ž. grade 2 , weak grade 1 , very weak Ž.grade 0.5 and not detectable Ž. grade 0 . nsNumber of animals used in each group. Each value is the mean"S.E.M. ) p-0.01 vs. controlŽ. A , ap-0.01 vs. KA-treated ratŽ. B or the seizing rat pretreated with esculetin Ž. D or aspirin Ž. E . 342 H.-C. Kim et al.rBrain Research 782() 1998 337±342 products of both cyclooxygenase and lipoxygenase path- wx6 P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by ways might be involved in inducing Fra protein, in turn, acid guanidium thiocyanate±phenol±chloroform extraction, Anal. Biochem. 162Ž. 1987 156±159. causing elevation of proENK and proDYN gene transcrip- wx wx 7 M. Comb, N.C. Birnberg, A. Seasholtz, E. Herbert, H.M. Goodman, tional activity. Won et al. 23 have recently reported that A cyclic AMP- and phorbol ester-inducible DNA element, Nature cycloheximide, a protein synthesis inhibitor, effectively 323Ž. 1986 353±356. inhibits KA-induced increases of proENK and proDYN wx8 C.W. Cotman, D.T. Monaghan, O.P. Ottersen, J. Storm-Mathisen, mRNA, and Fra protein levels, suggesting that on-going Anatomical organization of excitatory receptors and their pathways, Ž. proteinŽ. such as Fra protein synthesis is required for the Trends Neurosci. 10 1987 273±280. wx9 J. Douglass, L. Grimes, J. Shook, P.H.K. Lee, J.S. Hong, Systemic increases of opioid peptides gene expression induced by administration of kainic acid differentially regulates the levels of KA in the hippocampus. prodynorphin and proenkephalin mRNA and peptides in the rat Neither esculetin nor aspirin treated alone affected the hippocampus, Mol. Brain Res. 9Ž. 1991 79±86. wx expressions of proENK and proDYN mRNA levels, or 10 J. Douglass, C.T. McMurry, J.E. Garrett, J.P. Adelman, L. Calavetta, proENK and Fra proteins levels induced by KA, suggest- Characterization of the rat prodynorphin mRNAs in rat spinal cord, Mol. Brain Res. 6Ž. 1989 31±37. ing that the blockade of only one pathway of arachidonic wx11 C. Gall, N. Brecha, H. Karten, K.-J. Chang, Localization of acid cascade may not be enough to exert an inhibitory immunoreactivity to identified axonal and neuronal pop- action against KA-induced responses. The reasons for the ulations of the rat hippocampus, J. Comp. Neurol. 198Ž. 1981 335±350. requirement of blockade of both cyclooxygenase and wx lipoxygenase for the inhibition of KA-induced increases of 12 R. Kerwin, S. Patel, B. Meldrum, Quantitative autoradiographic analysis of glutamate binding sites in the hippocampal formation in proENK and proDYN mRNA levels, or proENK and Fra normal and schizophrenic brain postmortem, Neuroscience 39Ž. 1990 protein levels are not currently clear. Although not compa- 25±32. rable directly, Baran et al. have previously reported that wx13 G.L.G. La Salle, Long-lasting and sequential increase of c-fos blockade of both cyclooxygenase and lipoxygenase, but oncoprotein expression in kainic acid-induced status epilepticus, Ž. not cyclooxygenase or lipoxygenase alone, are also re- Neurosci. Lett. 88 1988 127±130. wx14 W.W. Lamph, P. Wamsley, P. Sassone-Corsi, I.M. Verma, Induction quired to show an antagonistic effect against seizure activ- of proto-oncogene JUNrAP-1 by serum and TPA, Nature 334 ity and neurotoxic effect induced by KA in the hippocam- Ž.1988 629±631. puswx 2 . wx15 J.F. McGinty, S.J. Henriksen, A. Goldstein, L. Terenius, F.E. Bloom, is contained within hippocampal mossy fiber: immuno- histochemical alternations after kainic acid administration and Acknowledgements colchicine-induced cytotoxicity, Proc. Natl. Acad. Sci. USA 80 Ž.1983 589±593. We thank Dr. Jau-Shyong HongŽ National Institute of wx16 J.I. Morgan, T. Curran, Stimulus-transcription coupling in the ner- Environmental Health Sciences, USA. and M.J. Iadarola vous system: involvement of the inducible proto-oncogenes fos and jun, Annu. Rev. Neurosci. 14Ž. 1991 421±451. Ž.National Institute of Dental Research, USA for donating wx Ž.17 J.V. Nadler, Kainic acid as a tool in the study of temporal lobe antiserum against proenkephalin-like protein YGGFMRF epilepsy, Life Sci. 29Ž. 1981 2031±2042. and Fra, respectively. This work was supported by the wx18 K.R. Pennypacker, M.K. McMillian, J. Douglass, J.S. Hong, On- Interdisciplinary Research GrantŽ. 95-0403-19-01-3 from togeny of kainate-induced gene expression in rat hippocampus, J. Korea Science and Engineering FoundationŽ. KOSEF . Neurochem. 62Ž. 1994 438±444. wx19 K.R. Pennypacker, D. Walczack, L. Thai, R. Fannin, E. Mason, J. Douglass, J.S. Hong, Kainate-induced changes in opioid References gene and AP-1 protein expression in rat hippocampus, J. Neu- rochem. 60Ž. 1993 204±211. wx1 H. Baran, H. Lassman, G. Sperk, F. Seiteberger, O. Hornykiewicz, wx20 J.L. Sonnenberg, P.F. Macgregor-Leon, T. Curran, J.I. Morgan, Effect of mannitol treatment on brain neurotransmitter markers in Dynamic alternations occur in the levels and composition of tran- kainic acid-induced epilepsy, Neuroscience 21Ž. 1987 679±684. scription factor AP-1 complexes after seizure, Neuron 3Ž. 1989 wx2 H. Baran, K. Vass, H. Lassmann, O. Hornykiewicz, The cyclooxy- 358±365. genase and lipoxygenase inhibitor BW755C protects rats against wx21 H.H. Suh, P.M. Hudson, R. Fannin, K. Boulom, M.K. McMillian, kainic acid-induced seizures and neurotoxicity, Brain Res. 646Ž. 1994 A.M. Poisner, J.S. Hong, Prolonged stimulation of bovine adrenal

201±206. chromaffin cells with arachidonic acid and E2 in- wx3 Y. Ben-Ari, Limbic seizure and brain damage produce by kainic creases expression of the proenkephalin gene and the secretion of acid: mechanisms and relevance to human temporal lobe epilepsy, wMet5 x enkephalin, J. Pharmacol. Exp. Ther. 263Ž. 1992 527±532. Neuroscience 14Ž. 1983 375±403. wx22 H.H. Suh, P.M. Hudson, J.S. Hong, Expression of the proenkephalin wx4 D. Bohman, T.J. Bos, A. Admon, T. Nishimura, P.K. Vogt, R. A gene andw Met5 x enkephalin secretion induced by arachidonic acid Tjian, Human protooncogene c-jun encodes a DNA-binding protein in bovine adrenal medullary chromaffin cells: involvement of second with structural and functional properties of transcription factor AP-1, messengers, J. Neurochem. 64Ž. 1995 608±613. Science 238Ž. 1987 1386±1392. wx23 J.S. Won, Y.H. Kim, D.K. Song, H.W. Suh, The effect of cyclohex- wx5 R.D. Burgoyne, T.R. Cheek, A.J. O'Sullivan, Receptor activation of imide on the regulation of proenkephalin and prodynorphin gene phospholipase A2 in cellular signalling, Trends Biochem. Sci. 12 expression induced by kainic acid in rat hippocampus, Mol. Brain Ž.1987 332±333. Res. 47Ž. 1997 303±310.