Effects of epidermal growth factor and basic fibroblast growth factor on generation of long-term potentiation in the dentate gyrus of fimbria-fornix-lesioned rats

335

Brain Research, 593 (1992) 335-338 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

BRES 25388

Effects of epidermal growth factor and basic fibroblast growth factor on generation of long-term potentiation in the dentate gyrus of fimbria-fornix-lesioned rats K a z u h o A b e , J u n i c h i Ishiyama a n d Hiroshi Saito Department of ChemicalPharmacology, Facultyof PharmaceuticalSciences, The Unicersityof Tokyo, Tokyo (Japan) (Accepted 14 July 1992)

Key words: Epidermal growth factor; Basic fibroblas~growth factor; Long-term potentiation; Dentate gyrus~Fimbria-fornix lesion

The effects of epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) on long-term potentiation (LTP) in the dentate gyrus in rive were investigated in fimbria-fornix (FF)-Iesioned rats. Transection of FF resulted in decreased frequency of LTP generation. Intracerebroventricular injection of EGF (50 ng) and bFGF (50 ng) significantly facilitated LTP generation in the FF-lesioned rats. These results suggest that EGF and hFGF can promote the hippoeampal LTP impaired by loss of subcortical afferents.

Epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) are well known as potent mitogons for a variety of cell types7'11. However, both have recently been found to promote the survival of primary cultured neurons from various brain regions z'3'~s'|'~'zs. Since presence of EOF and bFGF in the brain has boon demonstrated 9a°'~t', it is possible that they function as neurotrophic factors in the brain. In addition, we have recently found that EGF and bFGF, as acute effects, promote the generation of long-term potentiation (LTP) of evoked potential in the Sohaffer collateraI-CAl pyramidal cell synapses of rat hippocampai slices4's. Furthermore, we have also investigated in vivo effects of EGF and bFGF on the induction of LTP using anesthetized rats and reported that intracerebroventricular (i.c.v.) injection of EGF and bFGF 30 rain before application of totanic stimulation promoted the generation of LTP in the perforant path-dentate granule cell synapses ~4. These findirgs indicate the possibility that they have neuromoduW~tor-like actions also. The LTP in the hippocampus is a form of synaptic plasticity and is considered as the cellular basis of learning and memory 27. The fact that EGF and bFGF have dual effects on brain neurons, i.e. neurotrophic

action and positive modulation of synaptic plasticity, is of great interest to develop novel therapeutic drugs for neurodogenorative disorders such as AIzheimor's disease.

Subcortical afforents to the hippocampus including cholinergic system seem vulnerable to aging 12. Biochemical and pathological studios have shown that the activities of cholinorgic marker enzymes, choline acotyltransforase (CHAT) and acetylcholinestoras¢ (ACHE), decrease in the hippocampus of patients with Alzheimer's disease x'l"~'2'~-2s. Experimental lesions of the fimbria-fornix (FF), a major pathway of subcortical afferents to the dorsal hippocampus, of rats result in marked impairment of spatial learning and memory 17,2°-22,26. Furthermore, Buzsaki and Gage ¢' have recently reported that hippocampal LTP could not be induced in the dentate gyrus of FF-lesioned rats, indicating an essential involvement of subcortical afferents in the LTP generation. Whether EGF and bFGF affect the hippocampal LTP under pathological conditions is a very important subject to consider the treatment of neurodegenerative disorders. Therefore, in the present study, we made a model in which the generation of hippocampal LTP was impaired by bilat-

Correspondence: K. Abe, Department of Chemical Pharmacology, Faculty of Pharmaceutical Sciences, The University of Tokyo, Tokyo 113, Japan.

336 cral FF lesions, and examined the effects of EGF and bFGF on the LTP generation in the dentate gyrus of the impaired model. Male Wistar rats 6 weeks old were anesthetized with a mixture of ketamine and xTlazine (80 and 7 mg/kg, respectively, i.m.) and fixed in a stereotaxie frame. The FF lesion was made bilaterally by inserting a razor blade (width 5.4 mt.~} to a dt:pth of 4.5 m m at a position 1.1 mm posteri~r ~o bregn~a (see Fig. IA). The rats which received only drilling of a skull served as sham-operated group. It has previously been J'~!~orted that FF lesions resulted in dramatic disappearance of AChE-positive fibers in the hippocampus ~''-'~. In order to verify the completeness of the lesion procedure, subgroups of the FF-lesioned and sham-operatcd rats were taken for histochemical evaluation two weeks after the operation. The ra':s were perfused with ice-cold phosphatebuffered saline (pH 7.4) containing 4% paraformaldehyde and 0.2% picric acid. The brains were removed and post-fixed for 24 h in the same fixative and then left in phosphate-buffered saline containing 30% sucrose for 48 h at 4°C. Coronal sections were cut serially at 5(} ~ m thickness with a frozen-stage mierotome and stained for AChE according to the Karnovsky's method I'~. As a result, we clearly found many ACHEpositive fibcr,~ in the hippoeampus of the sham-operated rats, while there were few AChE-positive fibers throughout the dorsal hippoeampus of the FF-Iesioncd rats. This histochemical observation ensures that the subcortical affcrcnts to thc dorsal hippocampus were perfectly removed by the lesion procedure employed in the present study, Two to three weeks after the operation, recording of evoked potential was made as described previously t°. Briefly, rats were anesthetized with urethane (1 g/kg, i.p.) and a-chloralose (25 mg/kg, i.p.), and fixed in a stereotaxic frame. A bipohtr stimulating electrode was placed in the loft entorhinal cortex to stimulate the perforant path and the evoked potential was extraccllularly recorded from the granule cell layer of the ipsilatoral dentate gyrus. A single test stimulus (0.08 ms duration) was applied at intervals of 30 s and the stimulus intensity was set to a level which evoked a population spike of 50% of the maximum, After the response became stable, 5 p,I of the vehicle or the drug was injected into the contralatcral ventricle using a microsyringe (injection time 2.5 rain), The responses to test stimuli were observed for a further 30 rain so that the drug would diffuse thoroughly and to ensure stability of the baseline responses, Brief tetanic stimulation was then applied at the same stimulus intensity and the responses to test stimuli were recorded for 60 rain.

A

- -

l m m

lOO

J| 50

u.

~iS

30

60

100 pulses

60 Hz tehmic stimulation

Fig. 1. A: the diagram shows a coronal view of rat brain at a position 1.1 mm posterior to brcgma. The shaded area was lesioned by inserting a razor blade. B: frequency of LTP generation by application of different conditions of tetanie stimulations (15, 30, 60 and 100 pulses at 6{) Hz) in sham-operated (solid black columns) and FF-lesioned (shaded columns) rats, The number of cases in which LTP was generated by application of tetanus was expressed as a percentage of the total number of cases tested. The number of observations was five in all groups.

LTP was considcred to have occurred if the potentiated spike amplitude remained at a level more than 20% higher than the baseline level 30-60 rain after tetanic stimulation, The amplitude of the evoked potential in the porforant path-dentate granule cell synapses of the FF-lesioned rats was not significantly different from that of the sham-operated, in order to investigate the influence of FF lesion on LTP generation, frequencies of LTP generation by application of tetanic stimulations consisting of 15, 30, 60 or 100 pulses at 60 Hz were compared in sham-operated and FF-lesioned group without injection of drugs. As shown in Fig. 1B, stronger condition of tetanie stimulation was required to generate LTP in FF-lesioned group than in sham-operated group, indicating that LTP became hard to be induced as a result of FF lesion. In other words, this result suggests that subcortical afferents to the hippocampus play a role to facilitate the generation of LTP in normal animals. Difference between the sham-operated group and the FF-iesioned group was most remarkably seen when tetanus of 30 pulses at 60 Hz was applied. Application of 30-pulse, 60-Hz tetanus produced LTP in 4 out of 5 eases in the sham-operated

337 200

A

*

ISO

i

100

200

i -20

i

R

i 0 °*

i

*

i

~ *

i 30

]

i

I

I

i 60

*

1SO

100 I -20

I

I 0

I

I

I SO

I 60

Time ( rain )

Fig. 2. The effects of i.c.v, injection of 50 ng EGF (A, A n = 6) and 50 ng bFGF (B, m, n = 6) on the subthreshold tetanus (30 pulses at 60 Hz)-induced potentiation of evoked potential in the FF-lesioned rats. The data in the group i.e.v, injected with vehicle(e, n = 12) are shown in both A and B to facilitate comparison. The abscissas indicate the time in rain after application of tetanic stimulation.The ordinates indicate population spike amplitude expressed as a percentage of baseline values at 0 rain Oust before tetanic stimulation). The data are represented as the means+ S.E.M. of n observations. Significance of the differences was determined by using Duncan's multiple range test followingamtlysisof variance (ANOVA). Asterisks indicate significantdifferences from the data of vehicle-injected gr(mp (e); * P < 0.115, * * P < O.OI.

group, but in only 1 out 5 cases in the FF-lesioned group. The effects of E G F and b F G F on LTP generation in the FF-lesioned rats were investigated, EOF and b F G F used in the present study were recombinant human E G F (Wakunaga Pharmaceutical Co., Ltd., Osaka, Japan) and an acid-resistant mutein of human b F O F 2 (CS23; Takeda Chemical Industries, Ltd., Osaka, Japan), respectively. They were diluted to desired coneentrations by vehicle (phosphate-buffered saline supplemented with 1 m g / m l bovine serum albumin) just before use. Considering the result shown in Fig. 1B, the stimulation of 30 pulses at 60 Hz was used as a subthreshold tetanus required for generating LTP in FF-lesioned rats. In the vehicle-injected rats of FF-iesioned group, application of the subthreshold tetanus produced a small increase of population spike, but LTP was generated in only 3 out of 12 cases tested. Injection of 50 ng E G F (10/.~g/ml × 5/~1 i.c.v.) 30 min before application of tetanus did not influence the basal amplitude of population spike, but significantly enhanced the potentiation induced by subthreshold tetanus (Fig. 2A). The subthreshold tetanus-induced potentiation in the EGF-injected group resulted in generation of LTP in 5 out of 6 cases tested. Injection of 50 ng b F G F (10 / z g / m l × 5 /xl i.c.v.) also signifi-

cantly enhanced the potentiation induced by subthreshold tetanus, without affecting the basal responses before the tetanus (Fig. 2B). The LTP was generated in 5 out of 6 cases tested in the bFGF-injected group. The result that acute administration of E G F and b F G F promoted the generation of LTP in the dentate gyrus of FF-lesioned rats clearly shows that they can exert the neuromodulatory effects in abnormal cendition as w e l l as in normal condition. Therefore, we might conclude that both E G F and bFGF have potential for application to some neurodegenerative disorder. As described in the introduction, EGF and b F G F may prevent neuronal cell ~ca~l , l lesioned areas and repair the nervous system by their neurotrophic actions. In addition, the present results indicate that E G F and bFGF promote the plasticity of synaptic transmission by their neuromodulatory actions and play a role to compensate the synaptic function impaired by brain damage. Mechanisms underlying the LTP-facilitating effects of E G F and b F G F are not well elucidated, but the present results suggest that subcortical afferents are not always required for E G F and bFGF to exert the LTP-facilitating effects in the hippocampus. It is probable that they directly modulate the glutamatergic neurotransmission mechanisms in the hippocampus but not indirectly through subcortical afferents. We have very recently found that E G F selectively enhances the N M D A receptor-mediated increase of intracellular Ca 2+ concentration in cultured hippocampal neurons Taken together, we have demonstrated that E G F and b F G F promote the generation of hippocampal LTP in vivo impaired by FF lesions. EGF and b F G F might be beneficial in the recovery of hippocampal function impaired by loss of subcortical afferents. Alternatively, even if they are not applicable, further investigations about the relation between long-term (neurotrophic) and short-term (neuremodulatory) actions of E G F and bFGF must give important clues to develop novel therapeutic drugs for neurodegenerative disorders. The authors are grateful to WakunagaPharmaceuticalCo., Ltd. and Takeda Chemical Industries, Ltd. for their generous gifts of drugs. 1 Abe, K. and Saito, H., Epidermal growth factor selectivelyenhances NMDA receptor-mediated increase of intracellular Ca-'+ concentration in rat hippocampal neurons, Brain Res.. 587 (1992) 102-108. 2 Abe, K., Takayanagi, M. and Saito, It., Effects of recombinant human basic fibroblast growth factor and its modified protein, CS23, on survival of primary cultured neurons from various regions of fetal rat brain, Jpn. J. Phannacol., 53 (1990) 221-227. 3 Abe, K., Takayanagi, M. and Saito, H., A comparison of neurotrophie effects of epidermal growth factor and basic fibroblast

338 growth factor in primary cultured neurons from various regions of fetal rat brain, .lpn. J. PharmacoL, 54 (1990) 45-51, 4 Abe, K., Xie. F.-J. and Saito, H., Human basic fibroblast growth factor enhances neuronal survival and modulates synaptic transmission in rat hippocampus, Ear. J. Pharmacol., 183 (19901 2348. 5 Abe, K.. Xie, F.-J. and Saito, H., Epidermal growth factor enhances short-term potentiation and facilitates induction of longterm potentiation in rat hippocampal slices, Brain Res., 547 (1991) 171-174, 6 Buzsaki, G, and Gage, F.H., Absence of long-term potentiation in the subcorlically deafferented dentate gyrus, Brain Res., 484 (1989) 94-101, 7 Carpenter, G. and Cohen, S., Epidermal growth factor, Atom. Rev. Biochem., 48 (19791 193-216. 8 Davies. P. and Maloney, A.J., Selective loss of central cholinergic neurons in AIzhcimer's disease, Lmwet, (1976) 1403. 9 Fallon. J.H., Scroogy, K.B,. Loughlin. S.E.. Morrison. R.S., Bradshaw, R.A., Knauer, D.J. and Cunningham. D.D., Epidermal growth factor immtmoreactive material in the central nervous system: location and aevc?opment. Science, 224 (19841 1107-1109. 10 Gospodarowicz, D.. Ch,'n,.. L, Lui. G.-M.. Baird. A. and Bohlen. P., Isolation of br;fin |ib:~:~ s~ gro~.h filctor by heparin-Sepharos¢ affinity chromatography: identity with pituit,'lry fibroblast growth fi~etor. Proc NaIL Aca~L ScL USA.. 81 (1984)6963-6967. 11 Gospodarowicz, D., Neufeld, G. and Schweiger, L.. Fibl'oblast growth filctor, M¢I Cell. EpuhJcnaol., 4~ (1986) 187-204. 12 Greene, E. and Naranjo. J.N., Degeneration of hippocampal fibers :rod spalial memory deficit in the aged rats. NearobioL Agbrg, 8 ( 19871 35-43. 13 Ilenke. H. and Lang, W., Cholinergic enzymes in neocorlex, hippocampus and basal forebrain of non-neurological and senile dementia of AIzheimer.type patients, Brain Res., 2(~7(19831 281291. 14 lshiyama. J.. Saito, H. :rod Abe. K.. Epidermal growth factor and basic fibroblast growth .k~mrpromote the generation of long-term potentiation in the dentate gyros o[ anaesthetized rats, Neteros<'L Res., 12 (I991) 403~411, 15 Karmwsky. MJ., Th,.' localization of chldinest.ras¢ ,¢tivity in r,t cardiac muscle by electron microscopy. J, Cell Bkd., 23 (1964) 217=232. I¢1 l,akshmanan. J., Wcichscl, M.E. and Fisher, D.A., Epidcrnall p,t'owth factor it~ synaptnsontal fractions nf ntoose cerebral corlc~. J, Neurochem.. 4(1 (1986) I081-1085,

17 Morris, R,G.M., An attempt to dissociate 'spatial mapping" and "working memory' theories of hippocampal function. In W. Seiferl (Ed.L The Neurobiololo, of the Hippocampus, Academic Press, London, 1983, pp. 405-432. 18 Morrison, R.S., Eornblum, HJ., Leslie, F.M. and Bradshaw, R.A., Trophic stimulation of cultured neurons from neonatal rat brain by epidermal growth factor, Science, Z~8 (19~97)72-75. 19 Morrison, R.S., Sharma, A., de Vellis, J. and Bradshaw, R.A., Basic fibroblast growth factor supports the survival of cerebral cortical neurons in primary culture, Proc Natl. Acad. Sci. USA., 83 (19861 7537-7541. 20 Nilsson, O.G., Shapiro, M.L., Gage, F.H., OIton, D.S. and Bj6rklund, A., Spatial learning and memory followingfimbria-fornix transection and grafting of fetal septal neurons to the hippocampus. Exp. Brain Res., 67 (19871 195-215. 21 O'Keefe, J. and Nadel, L., The Hippocampus as a Cognitive Map, Clarendon, Oxford, 1978. 22 OIton, D.S., Bgcker. J.T. and Itandelman, G.E., Hippocampus, space and memory, Behav. Brain Sci., 2 (1979) 313-365. 23 Perry, E.K., Gibson, P.H., Blessed, O., Perry~ R.H. and Tomlin, son, B.E., Neurotransmitter enzyme abnormalities in senile dementia. Choline ace'~,ltransferase ~md glutamic acid decarbox~lase activities in necropsy brain tissue. J. Neurol. Sci., 34 (1977) 247-265. 24 Perry, R.H.. Blessed, G., Perry, E.K. and Tomlinson, B.E., Histochemical observations on cholinesterase activities in the brains of elderly normal and demented (Alzheimer-type) patients, Age Agei,g, 9 (1981119-16. 25 Ransmayr, G.. Cervera, P., Hirsch, E., Ruherg, M., Hersh, L.B., Duyckaerts, C., Hauw, I.-J., Delumeau, C, and Agid, Y., Choline acelyltransferase-like immunoreactivity in the hippoeampal formarion of control subjects and patients with AIzheimer's disease, Neurosckv~ce, 32 ( 19891 701-714. 26 Rawlins. J.N.P., Association across time: the hippoeampus as a temporary memory store, Behm', Brabl Sci., 8 (1985~ 479-497. 27 Teyler, T.J..'rod DiScenna, P., Long.term potentiation, Atom, Rer. Neurosci., 10 (19871 131-161. 28 Walicke, P,, Cowan, W.M., Ueno, N., Baird, A. and Guillemin, R., Fibrohlast growth helot promotes StlWiv|lJ of diss~ciated hippocampal neurons and enhances neurit¢ exlunsio,, Prrw. Nat/. .Ic~M. S('i, US,'I,, 83 (1986) 31112=3l)16,