Ž .Brain Research 845 1999 97–101www.elsevier.comrlocaterbres

Short communication

Protein kinase C inhibition blocks the early appearance of vestibularcompensation

Carey D. Balaban a,b,) , Maria Freilino a, Guillermo G. Romero c

a Department of Otolaryngology, UniÕersity of Pittsburgh School of Medicine, 203 Lothrop Street, Pittsburgh, PA 15213, USAb Department of Neurobiology, UniÕersity of Pittsburgh School of Medicine, 203 Lothrop Street, Pittsburgh, PA 15213, USA

c Department of Pharmacology, UniÕersity of Pittsburgh School of Medicine, 203 Lothrop Street, Pittsburgh, PA 15213, USA

Accepted 3 August 1999

Abstract

Ž .This study tests the hypothesis that activation of protein kinase C PKC is a critical step for early recovery from spontaneousnystagmus after unilateral ablation of the vestibular periphery. Halothane–NO –O -anesthetized Long–Evans rats received a 5-ml2 2

Ž . w Ž .intracerebroventricular bolus of vehicle distilled water, six rats , PKC inhibitor Iso-H-7 10 mM, four rats; 50 mM, five rats orŽ .x Ž .bisindolemaleimide I Bis-I, 10 mM, six rats , PKG and PKA inhibitor A-3, 1 mM, six rats , or the serine–threonine protein kinase

Ž . Ž .inhibitor H-7 1 mM, five rats; 10 mM, five rats . Surgical unilateral labyrinthectomy UL was completed within 15 min. Sham controlgroups showed no nystagmus. Bis-I and Iso-H-7 significantly retarded the disappearance of spontaneous nystagmus quick phases for 8 h

Ž . Ž .after UL p-0.05 . The effects of Iso-H-7 were dose-dependent: more nystagmus quick phases p-0.05 were present in the 50 mMthan the 10 mM group at 7 and 8 h post-UL. The rats given A-3 showed a delayed retardation of nystagmus loss, which differed

Ž .significantly p-0.05 from controls at 4–8 h after labyrinthectomy. The number of nystagmus quick phases was significantly greaterŽ .than controls p-0.05 in the 10 mM H-7 group at 4, 5, 6 and 48 h post-UL, but only at 6 and 24 h post-UL in the 1 mM H-7 group.

Thus, PKC activation is an important early requirement for vestibular compensation during the acute post-labyrinthectomy period, whilecyclic-nucleotide dependent kinases may be important in a later time frame. q 1999 Elsevier Science B.V. All rights reserved.

Keywords: Vestibular compensation; Protein kinase C; Cerebellum; Labyrinthectomy; Purkinje cell; Signal transduction protein

Ž .Protein kinase C PKC , cAMP-dependent protein ki-Ž . Ž .nase PKA and cGMP-dependent protein kinase PKG

have been implicated in integrative phenomena as diversew xas Aplysia heterosynaptic facilitation 3 , cerebellar long-

w xterm depression 2,5,7 , spinal mechanisms of persistentw xpain and hyperalgesia 18 , and muscarinic receptor-in-

w xduced transmitter release from the urinary bladder 16 .These phenomena appear within distinct temporal domainsand have behavioral consequences. Hence, interactionsbetween multiple kinases may contribute to diverse proper-ties of adaptive, compensatory and learning phenomena.

Damage to the vestibular sensory epithelium or vestibu-lar nerve produces acute vertigo, ocular nystagmus and

) Corresponding author. Department of Otolaryngology, University ofPittsburgh, The Eye and Ear Institute of Pittsburgh, 203 Lothrop Street,Pittsburgh, PA 15213, USA. Fax: q1-412-647-0108; e-mail:cbalaban@vms.cis.pitt.edu

balance dysfunction. The term Õestibular compensationdenotes the spontaneous resolution of these signs over a

Ž .period of days to weeks. Unilateral labyrinthectomy ULin rats produces an immediate asymmetry in neural activityw x w x14 , 2-deoxyglucose uptake 11–13 and c-fos expressionw x4,10 in the vestibular nuclei; the former changes resolvewithin 1 week. Compensation for UL is also accompaniedby transient, regionally selective changes in the distribu-tion of PKC-immunopositive Purkinje cells in the vestibu-locerebellum, which parallel the time course of the disap-

w xpearance of spontaneous nystagmus 1,6 . This study teststhe prediction that central administration of PKC inhibitorswill retard vestibular compensation during the first hoursafter UL.

Adult male Long–Evans rats were anesthetized withhalothane–NO –O . A 5-ml intracerebroventricular bolus2 2

Ž .of either the vehicle distilled water, six rats or a drug wasŽinjected stereotaxically through a burr hole 1 mm caudal

0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.Ž .PII: S0006-8993 99 01958-7

( )C.D. Balaban et al.rBrain Research 845 1999 97–10198

.and 1 mm lateral to bregma . Doses were selected toproduce comparable levels of PKC or PKArPKG inhibi-

Ž .tion re K in the 250 ml volume of rat cerebrospinal fluidiŽ .CSF ; estimated initial CSF concentrations are listed par-enthetically.

Ž . Ž w Ž1 Bisindolemaleimide I HCl Bis-I, 2- 1- 3-dimethyl-. x Ž .aminopropyl -1H-indol-3-yl -3- 1H-indol-3-yl -maleimide-

. w xHCl, Calbiochem 17 is a highly potent PKC inhibitorŽ .K s10 nM , which may inhibit PKA at a considerablyi

Ž .higher concentration K s2 mM . It was administered atiŽa concentration of 10 mM 196 nMs4=K for PKC;i

.0.02=K for PKA in CSF to six rats.iŽ . Ž w x.2 The PKC inhibitor IC s50 mM 15 Iso-H-750

Ž Ž .dihydrochloride 1- 5-isoquinolinesulfonyl -3-methyl-.piperazine HCl was given at concentrations of 10 mM 7

Ž .196 mMs4= IC in CSF to four rats and 50 mM50Ž .0.98 mMs20= IC in CSF to five rats.50

Ž . Ž . Ž3 The potent PKA K s4.3 mM and PKG K s3.8i i. Ž .mM and weak PKC K s47 mM inhibitor A-3 hydro-i

Ž Ž .chloride N- 2-aminoethyl -5-chloronaphthalene-1-sulfo-w x.namide HCl, Calbiochem 9 was given at a concentration

Žof 1 mM 19.6 mMs5=K for PKArPKG and 0.42=Ki i.for PKC to five rats.

Ž .4 The broad-spectrum serine–threonine protein kinaseŽ Ž .inhibitor H-7 dihydrochloride 1- 5-isoquinolinesulfonyl -

. w x2-methylpiperazine HCl; Calbiochem 8 inhibits PKCŽ . Ž . ŽK s6 mM , PKG K s5.8 mM and PKA K s3i i i

. ŽmM . It was given at a concentration of 1 mM approxi-. Žmately 3=K to five rats and 10 mM approximatelyi

.30=K to five rats.i

Surgical UL was completed under halothane–NO –O2 2

anesthesia within 15 min of the intracerebroventricularinjection, according to procedures described in previous

w xpublications 1,6 . Control groups were given a shamŽ w x.operation surgical exposure of middle ear cavity 1,6

after intracerebroventricular injection of the highest doseŽof each drug 50 mM Iso-H-7: two rats; 10 mM Bis-I:

.three rats; 10 mM A-3: three rats; 10 mM H-7: one rat .Rats recovered from the anesthetic within 10 min aftercompletion of the surgery. They were restrained manually

Ž .under a dissecting microscope 16= magnification con-Žnected to a Dage CCD-72 camera DAGE-MTI, Michigan

.City, IN . Spontaneous nystagmus was recorded underambient laboratory illumination on videotape at 1, 2, 3, 4,5, 6, 7, 8, 24, 48 and 72 h after UL for the rats givenIso-H-7, Bis-I and A-3. Rats given H-7 were not tested at7 and 8 h after UL. The number of quick phases ofspontaneous nystagmus was counted from the videotapeŽSony SLV595HF VCR and PVM 1341 or PVM1910

.Sony Trinitron monitor during three successive 20-s peri-ods and averaged. Repeated measures analysis of varianceŽwithin groups factor: post-operative time; between groups

.factor: treatment and post hoc multiple range tests wereused to test the hypothesis that drug treatments affected thefrequency of spontaneous nystagmus after UL. The rats

Žwere euthanized with a pentobarbital overdose 100

.mgrkg, i.p. the UL was verified in paraffin-embeddedw xtemporal bone sections 1,6 for 28r30 UL rats.

w xSince previous studies 1,6 were performed under fen-tanyl-droperidol anesthesia, a second group of animals wasused to determine whether transient PKC expressionchanges in the cerebellum are present after UL underhalothane–NO –O anesthesia. After survival times of 3 h2 2Ž . Ž .four UL, two sham , 6 h four UL, three sham and 24 hŽ .three UL, two sham , the rats were euthanized and per-

w xfused transcardially as described previously 1,6 . Pub-w xlished immunohistochemical procedures 1,6 were used to

visualize PKCa , PKCg, PKCd and PKC´ in alternatesets of frozen sections.

The distribution of PKC immunopositive cells in theflocculonodular lobe after UL was identical to observa-

w xtions under fentanyl-droperiodol anesthesia 6 . Results forPKCd are shown in Fig. 1; the same results were observedfor PKCa , PKCg and PKC´. Three hours after UL, all

Ž .four rats Fig. 1A–D displayed a greater number of PKCimmunopositive Purkinje cells on the side contralateral tothe lesion than on the ipsilateral side. By contrast, asymmetric distribution of PKC immunopositive Purkinjecells was observed in the sham-operated rats at the same

Ž .survival time Fig. 1E . The asymmetric distribution ofŽ .Purkinje cells persisted at the 6-h survival time Fig. 1F ,

and resolved to a symmetric pattern by 24 h after UL.The two relatively specific PKC inhibitors significantly

Ž .retarded the loss of spontaneous nystagmus Fig. 2 . TheŽ .Bis-I dose Fig. 2, upper left panel attenuated the recovery

from spontaneous nystagmus significantly during the firstŽ .8 h after UL p-0.05 , but the animals did not differ

from the vehicle controls at 24 h after surgery. BothŽ .Iso-H-7 doses Fig. 2, upper right panel also attenuated

the decline of spontaneous nystagmus significantly duringthe first 8 h post-UL, and the animals did not differsignificantly from controls at 24 h after UL. The effect ofIso-H-7 was dose-dependent: a higher frequency of nystag-

Ž .mus p-0.05 was present in the 50 mM than the 10 mMgroup at 7 and 8 h post-UL. Since nystagmus was absentin sham control rats given either Bis-I or 50 mM Iso-H-7,these effects do not reflect a transient induction of nystag-mus by the drugs and surgical procedures.

Because PKC inhibitors also inhibit PKA and PKG atw x w xhigh concentrations 8,17 , A-3 9 was administered to

identify effects due to PKG and PKA inhibition. A lessprofound, delayed effect on spontaneous nystagmus was

Ž .observed after A-3 administration Fig. 2, lower left panel .The nystagmus frequency did not differ from the vehiclecontrol group during the first 3 h after UL. Although there

Ž .was a significant effect at 4–8 h after surgery p-0.05 ,the A-3-treated rats displayed significantly less nystagmus

Ž .at these times than rats given either Bis-I p-0.01 or 10Ž .mM Iso-H-7 p-0.05 . The A-3 dose did not signifi-

cantly affect spontaneous nystagmus at 24–72 h after UL.Thus, the early effects of Bis-I and Iso-H-7 indicate a roleof PKC activation in early compensation, while PKA and

()

C.D

.Balaban

etal.r

Brain

Research

8451999

97–

10199

Fig. 1. Photomicrographs illustrating transient asymmetric expression of PKCd in the cerebellar nodulus at 3 and 6 h after UL. Panels A–D show the distribution of Purkinje cells that are immunopositive forŽ .PKCd from four different rats, that survived 3 h after UL. Note the greater number of positive cells on the side contralateral to the UL black arrow . Panel E shows the symmetric distribution of

Ž .PKCd-immunopositive Purkinje cells in a sham-operated control rat 3-h survival . Panel F shows the asymmetric distribution of PKCd-positive Purkinje cells 6 h after UL; the side contralateral to the UL isindicated by the black arrow.

( )C.D. Balaban et al.rBrain Research 845 1999 97–101100

Ž .Fig. 2. Upper left panel: Effects of a single, intracerebroventricular bolus injection 5 ml of the PKC inhibitor bisindolemaleide I on spontaneousŽU . Ž .nystagmus after unilateral labyrinthectomy. An asterisk indicates a significant difference p-0.05 with respect to the vehicle control group at the

same time point. Upper right panel: Effects of a single, intracerebroventricular bolus injection of the PKC inhibitor Iso-H-7 on spontaneous nystagmusŽU . Ž .after unilateral labyrinthectomy. An asterisk indicates a significant difference p-0.05 with respect to the vehicle control at the same time point. A

Ž . Ž .plus sign q indicates a significant difference p-0.05 between the two Iso-H-7-treated groups. Lower left panel: Effects of a single, intracerebroven-Ž . ŽU .tricular bolus injection 5 ml of the PKArPKG inhibitor A-3 on spontaneous nystagmus after unilateral labyrinthectomy. An asterisk indicates aŽ .significant difference p-0.05 with respect to the vehicle control group at the same time point. Lower right panel: Effects of a single,

Ž .intracerebroventricular bolus injection 5 ml of the broad-spectrum protein kinase inhibitor H-7 on spontaneous nystagmus after unilateral labyrinthec-ŽU . Ž . Ž .tomy. An asterisk indicates a significant difference p-0.05 with respect to the vehicle control group at the same time point. A plus sign q

Ž .indicates a significant difference p-0.05 between the two Iso-H-7-treated groups.

PKG activation may contribute to later compensatory pro-cesses.

Administration of H-7 produced a delayed effect similarŽ .to A-3 Fig. 2, lower right panel . No difference was

( )C.D. Balaban et al.rBrain Research 845 1999 97–101 101

observed between H-7-treated rats and controls during thefirst 3 h after UL. Nystagmus frequency was significantly

Ž .greater than controls p-0.05 in the 10 mM H-7 groupat 4–6 and 48 h post-UL. However, the 1 mM H-7 doseretarded the compensatory loss of spontaneous nystagmusonly at 6 h post-UL. Thus, the effects of H-7 are consistentwith PKA and PKG inhibition.

This study has revealed that structurally unrelated PKCinhibitors, Iso-H-7 and Bis-I, significantly retarded thedisappearance of spontaneous nystagmus during the first 8h after UL. This retardation of vestibular compensationwas significant within the first hour after injury. Theeffects of Iso-H-7 were dose-dependent. Since injections ofthese PKC inhibitors in sham-operated rats failed to pro-duce nystagmus, the observed effects cannot be explainedby an interaction between a drug, anesthesia and recoveryfrom surgical exposure of the middle ear cavity. Further-

Ž .more, neither A-3 PKArPKG inhibition nor H-7Ž .equipotent PKC, PKA and PKG inhibition affected thefrequency of nystagmus during the first 3 h after UL.Thus, the blockade of early vestibular compensation byPKC inhibitors suggests that PKC is an essential functionalpathway for early recovery of balance function after dam-age to the vestibular periphery.

It is impossible to pinpoint sites of action of PKCinhibitors administered by the intracerebroventricular route.However, because the early effects of the PKC inhibitorson vestibular compensation are coincident with changes inPKC expression in flocculonodular lobe Purkinje cells, onepotential site of action is the cerebellar cortex. De Zeeuw

w xet al. 5 reported that both cerebellar LTD and vestibulo-ocular reflex adaptation were abolished in transgenic mice

[that express the pseudosubstrate PKC inhibitor PKC 19–]31 . Thus, the blockade of early signs of vestibular com-

pensation by PKC inhibitors is consistent with a transientattenuation of LTD in these regions.

The effects of PKC, PKA and PKG inhibitors areconsistent with an important early signaling role forvestibular compensation. The effects of Bis-I and Iso-H-7provide compelling evidence that PKC activation is essen-tial for vestibular compensation within the first hour afterUL. The effects of A-3, though, indicate that contributionsof PKA and PKG have a delayed time course. These

w xfindings support the concept 3 that interactions betweenkinases, within distinct temporal domains, may be basicmolecular building blocks of neural plasticity.

Acknowledgements

The authors wish to acknowledge the expert technicaland histological assistance of Gloria Limetti, VladimirShikher and Jennifer Matalak. Mark Goto, M.D. assisted inthe initial immunohistochemical studies experiments withhalothane anesthesia. These studies were supported by R01

Ž .DC02556 CDB .

References

w x1 C.D. Balaban, G.G. Romero, A role of climbing fibers in regulationof flocculonodular lobe protein kinase C expression during vestibu-

Ž .lar compensation, Brain Res. 804 1998 253–265.w x2 A.R. Boxall, B. Lancaster, J. Garthwaite, Tyrosine kinase is required

Ž .for long-term depression in the cerebellum, Neuron 16 1996805–813.

w x3 J.H. Byrne, E.R. Kandel, Presynaptic facilitation revisited: state andŽ .time dependence, J. Neurosci. 16 1996 425–435.

w x4 C. Cirelli, M. Pompeiano, P. D’Ascanio, O. Pompeiano, Early c-fosexpression in the rat vestibular and olivocerebellar systems after

Ž .unilateral labyrinthectomy, Arch. Ital. Biol. 131 1993 71–74.w x5 C.I. De Zeeuw, C. Hansel, F. Bian, S.K.E. Koekkoek, A.M.

Vanalphen, D.J. Linden, J. Oberdick, Expression of a protein kinaseC inhibitor in Purkinje cells blocks cerebellar LTD and adaptation of

Ž .the vestibulo-ocular reflex, Neuron 20 1998 493–508.w x6 M.M. Goto, G.G. Romero, C.D. Balaban, Transient changes in

flocculonodular lobe protein kinase C expression during vestibularŽ .compensation, J. Neurosci. 17 1997 4367–4381.

w x7 N.A. Hartell, cGMP acts within cerebellar Purkinje cells to producelong term depression via mechanisms involving PKC and PKG,

Ž .NeuroReport 5 1994 833–836.w x8 H. Hidaka, M. Inagaki, S. Kawamoto, Y. Sasaki, Isoquinolinesulfon-

amides, novel and potent inhibitors of cyclic nucleotide dependentŽ .protein kinase and protein kinase C, Biochemistry 23 1984 5036–

5041.w x9 M. Inagaki, S. Kawamoto, H. Itoh, M. Saitoh, M. Hagiwara, J.

Takahashi, H. Hidaka, Naphthalensulfonamides as calmodulin antag-Ž .onists and protein kinase inhibitors, Mol. Pharmacol. 29 1986

577–581.w x10 G.D. Kaufman, J.H. Anderson, A.J. Beitz, Brainstem fos expression

following acute unilateral labyrinthectomy in the rat, NeuroReport 3Ž .1992 829–832.

w x11 R. Llinas, K. Walton, Vestibular compensation: a distributed prop-erty of the central nervous system, in: H. Asanuma, V.F. WilsonŽ .Eds. , Integration in the Nervous System. Igaku-Shoin, Tokyo,1979, pp. 145–166.

w x12 W.H.M.L. Luyten, F.R. Sharp, A.F. Ryan, Regional differences ofbrain glucose metabolic compensation after unilateral labyrinthec-

w14 x Ž .tomy in rats: a C 2-deoxyglucose study, Brain Res. 373 198668–80.

w x13 J.W. Patrickson, J.H. Bryant, M. Kaderkaro, F.A. Kutyna, A quanti-w14 xtative C -2-deoxy-D-glucose study of brain stem nuclei during

horizontal nystagmus induced by lesioning the lateral crista am-Ž .pullaris of the rat, Exp. Brain Res. 60 1985 227–234.

w x14 W. Precht, H. Shimazu, C.H. Markham, A mechanism of centralcompensation of vestibular function following hemilabyrinthectomy,

Ž .J. Neurophysiol. 29 1966 996–1010.w x15 J. Quick, J.A. Ware, P.E. Driedgerm, The structure and biological

activities of the widely used kinase inhibitor, H7, differ dependingon the commercial source, Biochem. Biophys. Res. Commun. 187Ž .1992 657–663.

w x16 G.T. Somogyi, M. Tanowitz, G. Zernova, W.C. de Groat, M1muscarinic receptor-induced facilitation of ACh and noradrenalinerelease in the rat urinary bladder is mediated by protein kinase C, J.

Ž . Ž .Physiol. London 496.1 1996 245–254.w x17 D. Toullec, P. Pianetti, H. Coste, P. Bellevergue, T. Grand-Perret,

M. Ajakane, V. Baudet, P. Boissin, E. Boursier, F. Loriolle, L.Duhamel, D. Charon, J. Kirilovsky, The bisindolylmaleimide GF109203X is a potent and selective inhibitor of protein kinase C, J.

Ž .Biol. Chem. 266 1991 15771–15781.w x18 K. Yashpal, G.M. Pitcher, A. Parent, R. Quirion, T.J. Coderre,

Noxious thermal and chemical stimulation induce increases in 3H-phorbol 12,13-dibutyrate binding in spinal cord dorsal horn as wellas persistent pain and hyperalgesia, which is reduced by inhibition of

Ž .protein kinase C, J. Neurosci. 15 1995 3263–3272.