Regulation of neurotrophin and traka, trkb and trkc tyrosine kinase receptor messenger RNA expression in kindling

Neuroscience Vol. 53, No. 2, pp. 433-446, 1993

0306-4522/93 $6.00 + 0.00 Pergamon Press Ltd © 1993 IBRO

Printed in Great Britain

REGULATION OF NEUROTROPHIN A N D trkA, trkB A N D trkC TYROSINE KINASE RECEPTOR MESSENGER R N A EXPRESSION IN K I N D L I N G J. BENGZON*~', Z. KOKAIA*, P. ERNFORS~, M. KOKAIA*, G. LEANZA,§ O. G. NILSSON,§ H. PERSSON~and O. LINDVALL* *Restorative Neurology Unit, Department of Neurology, University Hospital, S-221 85 Lund, Sweden :~Department of Medical Chemistry, Laboratory of Molecular Neurobiology, Karolinska Institute, S-104 01 Stockholm, Sweden §Department of Medical Cell Research, Biskopsgatan 5, S-223 62, Lund, Sweden Abstract--Levels of messenger RNA for nerve growth factor, brain-derived neurotrophic factor, neurotrophin-3, and the tyrosine kinase receptors trkA, trkB and trkC have been studied using in situ hybridization in the rat brain 2 h and four weeks after kindling-induced seizures. Epileptiform activity evoked by hippocampal stimulation and exceeding 70 s lead to a concomitant and transient increase of brain- derived neurotrophic factor, nerve growth factor, trkB and trkC messenger RNA expression in dentate granule cells after both focal and generalized seizures. Brain-derived neurotrophic factor messenger RNA levels were also increased bilaterally in the CA IqZA3 regions, amygdala and the piriform, entorhinal, perirhinal, retrosplenial and temporal cortices after generalized seizures. The magnitude of the increases was similar throughout the development of kindling and in the fully kindled brain. No changes of trkA messenger RNA were observed. In amygdalar kindling, elevated brain-derived neurotrophic factor messenger RNA levels developed more rapidly in the amygdala-piriform cortex than after stimulation in the hippocampus but changes in the hippocampal formation were only seen in few animals. Intraventricular 6-hydroxydopamine or a bilateral fimbria-fornix lesion did not alter basal expression or seizure-evoked changes in messenger RNA levels for neurotrophins or trk receptors but increased the number of animals exhibiting elevated levels after the first stimulation, probably due to a prolongation of seizure activity. Both in sham-operated and fimbria-fornix-lesioned rats seizure activity caused a marked reduction of neurotrophin-3 messenger RNA levels in dentate granule cells. The results indicate that activation of the brain-derived neurotrophic factor gene, at least in dentate granule cells, is an "all-or-none" type of response and dependent on the duration but not the severity of seizures or the stage of kindling epileptogenesis. Changes in brain-derived neurotrophic factor, nerve growth factor, neurotrophin-3 and trkB and trkC were observed concomitantly in the dentate gyrus, which suggests that seizure activity sets in motion a cascade of genomic events possibly mediated via a common mechanism. Since altered messenger RNA levels outside hippocampus were detected only for brain-derived neurotrophic factor, neurotrophin and trk gene expression in these regions seems to be regulated differently.

The term kindling refers to a phenomenon whereby repeated administration of a stimulus triggering seizure activity results in progressive intensification of epileptiform responses, culminating in a generalized seizure (e.g. see Racine and Burnham47). Kindling can be induced by electrical stimulation in various sites of the forebrain. Amygdala is a sensitive structure whereas kindling from the hippocampus requires significantly more stimulations. The pattern of development of seizures in kindling is, however, very similar between different stimulation sites within the limbic system and evolves through well-defined stages. When generalized seizures are produced, kindling has been established and even if the animal

tTo whom correspondence should be addressed Abbreviations: AD, afterdischarge; BDNF, brain-derived neurotrophic factor; DTT, dithiotreithol; dATP, deoxyadenosine 5'-triphosphate; EEG, electroencephalogram; FF, fimbria-fornix; NGF, nerve growth factor; SSC, saline-sodium citrate; 6-OHDA, 6-hydroxydopamine. 433

is left unstimulated for a long time it will respond with a generalized seizure to one of the first administered stimuli. 57 Kindling triggered by stimulation in the limbic system has been proposed to be analogous to complex partial epilepsy (also called temporal lobe e p i l e p s y f ) , which is the most frequent type of epilepsy in humans/7 The mechanisms underlying the progressive increase of responsiveness to a constant stimulus during kindling epileptogenesis are not known. Synaptic reorganization of mossy fibres occurs early in the kindling process, progresses with repeated seizures and is then essentially permanent. 8'55 Kindling is facilitated if mossy fibre synaptic reorganization has been induced already before the onset of stimulations and, conversely, is delayed if granule cells and mossy fibres are destroyed prior to kindling. 54 These observations suggest that sprouting of mossy fibres could play an important role in the development of kindling and also in the pathogenesis of human temporal lobe epilepsy. Recent experimental evidence

434

J. BENGZONet al.

has indicated that seizure-evoked changes in the levels of a family of neurotrophic factors, known as the neurotrophins, could be involved in these plastic responses. The prototype for the neurotrophins is nerve growth factor (NGF) 36 and the family also includes brain-derived neurotrophic factor (BDNF) 3,3s neurotrophin-312'23'26'39'5°neurotrophin-42° and neurotrophin 55 which appears to be the mammalian homologue of neurotrophin-424 first isolated from Xenopus laevis. 2° NGF, BDNF and neurotrophin-3 are all expressed in the brain where N G F provides atrophic support for basal forebrain cholinergic neurons2 whereas cells responding in vivo to BDNF and neurotrophin-3 remain to be identified. The highest levels of NGF, BDNF and neurotrophin-3 mRNAs are found in the hippocampus where the factors are expressed in neurons with an overlapping, but distinct, distribution.13'45 Outside the hippocampus, BDNF mRNA has the most widespread distribution, being localized, for example, in the neocortex, piriform cortex and amygdala. Members of the trk family of tyrosine kinase receptors, trkA, trkB and trkC, have been shown to encode essential components of functional high-affinity receptors for NGF, BDNF and neurotrophin-3.21'27'28'3°'31'34'51'52N G F and BDNF mediate their effects via trkA and trkB, respectively, whereas neurotrophin-3 interacts mainly with the trkC receptor but also binds to trkA and trkB receptors. 34 Seizure activity has been reported to induce changes in the levels of mRNA for the neurotrophins in several models of epilepsy. Intraventricular,16 intraperitoneal6° and intrahippocampal~ administration of kainate as well as septal and entorhinal injections of quisqualate37 lead to dramatic increases of N G F and BDNF mRNA levels in hippocampai and cortical neurons. After a dentate gyrus hilar lesion causing seizure activity, elevated N G F mRNA expression was found in dentate gyrus, amygdala and entorhinal, piriform and neocortex. 15 Increases of BDNF mRNA were more widespread and developed more rapidly than the changes in N G F mRNA. 25 In contrast, seizure activity produced by the hilar lesion reduced neurotrophin-3 mRNA levels in dentate granule cells. 49 Using in situ hybridization we have previously shown 1° increases in mRNA for N G F and BDNF after recurrent (every 5 min) seizures evoked by electrical stimulation in the hippocampus, socalled rapid kindling. Forty stimulus-evoked seizures lead to marked and transient increases of BDNF mRNA in the dentate gyms, parietal and piriform cortices, the pyramidal layer of the hippocampus and the amygdaloid complex, l° Elevated levels of N G F mRNA were demonstrated in the dentate gyrus and the parietal and piriform cortices. ~° Moreover, increases spatially and temporally similar to those seen for BDNF mRNA have also been shown for trkB mRNA and the level of trkB protein was found to be elevated in the hippocampus after rapid kindling.43 It has been hypothesized that the changes in neurotrophins and their receptors might be involved either

in kindling-associated neural plasticity or in neuronal protection following epileptic seizures. Some support for an important role of neurotrophic mechanisms in kindling epileptogenesis has been provided by the demonstration that intraventricular injection of antiserum to N G F can delay the development of amygdala kindling.14 In the present study, animals have been stimulated according to the traditional kindling paradigm (once daily) TM with the following main objectives: (i) to measure levels of mRNAs for NGF, BDNF and neurotrophin-3, and for trkA, trkB and trkC in response to seizures at different stages of kindling development and in the fully kindled brain; (ii) to compare changes in mRNA levels for the neurotrophins and their receptors after seizures evoked by stimulation in the hippocampus with those occurring after kindling in the amygdala; (iii) to evaluate the effects on seizure-evoked mRNA levels caused by lesions of two major afferent systems causing hyperexcitability in the hippocampus, i.e. removal of the noradrenergic projection from the locus coeruleus and the fimbria-fornix input.

EXPERIMENTAL PROCEDURES

Subjects Ninety-sixmale Sprague-Dawley rats (ALAB, Stockholm, Sweden) weighing250 g at the start of the experiments were used. The rats were housed under 12h light/12h dark conditions with ad libitum access to food and water. Lesion surgery Twelve rats were put under general equithesin anaesthesia (3.0 ml/kg i.p.), placed in a Kopf stereotaxic frame and 6-hydroxydopamine (6-OHDA; 250/~g of free base in 20 #1 ascorbate-saline) was then infused into the fight lateral ventricle (tooth-bar at 0; 0.6 mm caudal to bregma, 1.5 mm lateral to midline, 4.5 mm ventral to the outer skull surface). Electrodes were implanted after a recovery period of four weeks. In another sixteen rats, bilateral aspirative lesions of the fimhria-fornix (FF) and bilateral removal of the sympathetic superior cervical ganglion were performed in the same session (for details, see e.g. Ref. 53). Eight rats were subjected to a FF-sham operation which consisted of skull and neck incisions. Two weeks' recovery was allowed before electrode implantation. Electrode implantation and kindling Thirty intact rats as well as the previously 6-OHDAtreated, FF-lesioned and FF-sham operated rats were anaesthetized with equithesin, and bipolar stainless steel electrodes (outer diameter 0.25mm; Plastic One Inc., Roanoke, VA, U.S.A.) were implanted bilaterally into the ventral hippocampal CA142A2 region as previously described.33 Electrodes were placed bilaterally in the basolateral amygdala in another 30 intact rats (tooth-bar at -3.3; 2.8 mm caudal to bregma, 5.0 mm lateral to midline, 7.8 mm ventral to the dura). One week later, the rats were electrically stimulated in the left hippocampus or amygdala [100 Hz, 1 ms square wave pulses for 1 s, current set just above threshold for inducing an afterdischarge (AD) of more than 5 s duration] or connected to the stimulatingrecording device without being stimulated. After stimulation, the electroencephalogram (EEG) was continuously recorded from the electrodes until 2 min after cessation of seizure activity. Behavioural seizures were scored according

Neurotrophin, trkA, trkB and trkC mRNA in kindling to Racine. 46 Grades were: 0, normal behaviour, wet dog shakes, arrest; 1, facial twitches (nose, lips, eyes); 2, chewing, head nodding; 3, forelimb clonus; 4, rearing, falling on forelimbs, hindlimb clonus, tail clonus; 5, rearing, falling on side or back, rolling. Animals groups Four intact rats stimulated in the hippocampus and four animals stimulated in the amygdala were decapitated after each of the following treatments: 2 h after one stimulation, 2 h after the first grade 4 seizure, 2 h after the third grade 5 seizure, four weeks after the third grade 5 seizure and 2 h after a stimulation administered to animals which had experienced their third grade 5 seizure four weeks earlier. Two control animals were matched to each of the groups. The 6-OHDA lesioned rats were killed 2 h after either a single stimulation or the third grade 5 seizure (four stimulated and two control rats in each group). FF-lesioned animals were decapitated 2 h after one stimulation or the third grade 5 seizure (six stimulated and two control rats in each group). FF-sham operated rats were subjected to the same treatment (two stimulated and two control rats in the respective groups). After decapitation, the brains were immediately removed and frozen on dry-ice. In situ hybridization Cryostat sections (14/~m thick) were thawed onto polyL-lysine-coated slides (50 mg/ml). After fixation with 10% formalin for 30 min, the sections were processed for in situ hybridization with alpha-[35S]dATP Y-end-labelled probes as previously described, n To detect NGF, BDNF and neurotrophin-3 mRNAs, 50-mer oligonucleotide probes were used. j3 All nine trkB mRNA transcripts presently known to be expressed in the brain ("trkB all") 29,44as well as the fulllength mRNA encoding the functional receptor ("trkB full length") were demonstrated with oligonucleotide probes complementary to nucleotides 1314 1361 and 1360 1407, respectively, in the rat trkB sequence of Middlemas et al. 44 TrkA and trkC oligonucleotide probes were designed using the DNA sequence of previously isolated rat cDNA clones for these mRNAs. 42'43 The nucleotide sequence of the oligonucleotide detecting trkA mRNA was 5'-AAGGTTGAACTCAAAAGGGTTGTCCATGAAGGCAGCCATGATGGAGGC-3'. trkC mRNA was demonstrated with an oligonucleotide with the nucleotide sequence 5'-GAAGTGGCCGTTATGGTCTGGTTGGCTGTGCCCAGGGCATTCTTAGCAAT-Y. The use of these oligonucleotide probes for in situ hybridization, including their specificity, is described in detail elsewhere. 12.42,43Four sections from the same brain were hybridized to each probe. Hybridization to NGF, BDNF, trkB all and trkC mRNA was performed in all animals. Hybridization to the neurotrophin-3 mRNA probe was performed in FF-lesioned and sham-operated rats. trkA mRNA was analysed in intact animals kindled from the hippocampus and amygdala and the trkB fulllength probe was hybridized to sections obtained 2 h after the first grade 4 seizure induced by hippocampal stimulation in intact animals. Hybridization was performed in 50% formamide 4 x saline-sodium-citrate (SSC), 1 × Denhardts solution, 10% dextran sulphate, 0.25 mg/ml yeast tRNA, 0.5 mg/ ml sheared salmon sperm DNA, 1% sarcosyl (N-lauroyl sarcosine), 0.02 M Na2HPO4NaH2PO4 buffer (pH 7.0) and 0.02 M dithiotreithol (DTT) using 107 c.p.m./ml of respective probe. The sections were hybridized at 42°C for 15 h in a humidified chamber with 0.1 ml per slide of hybridization solution. They were subsequently rinsed, washed four times (15 min each) at 55°C in 1 x SSC, and air-dried. The sections were dipped in Kodak NTB-3 photoemulsion (diluted 1:I in water), exposed for five to six weeks at -20°C, developed, fixed, and lightly counterstained with Cresyl Violet. The mRNA levels were quantified by computerized image analysis of autoradiographic films using Image software (Dr

435

Wayne Rasband, NIMH, Bethesda, MD) and a Macintosh Ils computer. Standards were prepared by homogenizing known amounts of alpha-[35S]dATP in a total volume of 5 #1 with 100 mg of brain paste. The isotope containing brain paste was then frozen and 14-/~m sections were cut. Standards and tissue sections were exposed for the appropriate time and were then subjected to image analysis. Control hybridized tissue sections (sense-oligonucleotide probe) were used to define the background hybridization and these values were subtracted from the measured values. Grey levels obtained from standards in the range of 50-50,000 c.p.m./mg were determined and used in a polynomial approximation for construction of a grey level to activity transfer function. The mean value of four measurements taken for each probe, brain and brain area was used for statistical analysis. Statistical procedures Statistical evaluation of differences in kindling parameters and changes in mRNA levels between groups was performed using one-way analysis of variance (ANOVA). Fishers exact test was used to calculate the minimum AD duration for inducing changes in mRNA levels and for statistical analysis of differences between groups in the number of animals displaying changes in mRNA levels. For the statistical analysis, image analysis values from two-electrode-implanted, nonstimulated animals matched to each of the five different stages of hippocampal or amygdala kindling were pooled into control groups of 10 rats. In the 6-OHDA-treated and FF-lesioned groups, non-stimulated rats matched to the first stimulation and the third grade 5 seizure were pooled.

RESULTS Hippocampal kindling in non-lesioned rats Kindling characteristics. In the animals stimulated in the h i p p o c a m p u s the threshold for inducing a n A D was 150 ___39 # A (mean + S.E.M.). The rats subjected to one stimulation exhibited grade 0 seizures with a n A D duration of 46 ___25 s. The n u m b e r of stimulations to reach the first grade 4 a n d the third grade 5 seizure was 16 + 1 a n d 20 ___ i, respectively, a n d the A D durations o f these seizures were 116 + 24 a n d 87 + 7 s, respectively. A single stimulation in rats which h a d experienced their third grade 5 seizure four weeks earlier resulted in grade 4-5 seizures with a n A D d u r a t i o n o f 82 ___24 s. Brain-derived neurotrophic f a c t o r messenger R N A . In agreement with previous studies 12'22'45'58 B D N F m R N A was f o u n d to be moderately expressed in the dentate granule cells a n d in the h i p p o c a m p a l CA3 a n d C A 2 pyramidal n e u r o n s in electrode-implanted, unstimulated control rats. Lower levels of B D N F m R N A were observed over the h i p p o c a m p a l C A I sector, the amygdaloid complex, the piriform, cingulate, a n d e n t o r h i n a l cortices a n d the neocortex (Figs IA, 2A). The first stimulus-evoked seizure gave rise to a m a r k e d increase o f B D N F m R N A bilaterally in the dentate granule cells in one out o f four animals (Fig. 3). T h e increase in this a n i m a l was 8.7-fold c o m p a r e d to control, whereas B D N F m R N A was unaltered in the remaining rats (as defined by a change o f less t h a n 1 s t a n d a r d deviation from the control level; Fig. 3). In addition, bilateral increases in B D N F m R N A were seen in the h i p p o c a m p a l CA3

436

J. BENGZONet al.

D

,6,

B

1E

C

F

Fig. 1. Darklield photomicrographs from autoradiograms showing BDNF mRNA expression at different stages during the development of hippocampal kindling. Coronal sections were prepared from electrode-implanted, non-stimulated control rats (A), from animals killed 2 h after the first stimulation (B), the first grade 4 seizure (C), or the third grade 5 seizure (D) and from rats four weeks after the third grade 5 seizure either before (E) or 2 h after (F) a stimulusevoked generalized seizure. Scale bar = 3.8 mm.

and CA 1 sectors (2.3- and 2.2-fold of control, respectively) in this animal. No change in BDNF mRNA expression could be detected outside the hippocampal formation following the first stimulation (Fig. 1B). After the first generalized (grade 4) seizure, widespread increases of BDNF mRNA were seen in the forebrain of all rats (Fig. 1C). In addition to a 9.3-fold increase in the dentate granule cells (Fig. 3; different compared to control at P < 0.001), BDNF mRNA levels were also markedly elevated in the pyramidal cells of the piriform cortex (7.1-fold of control; P < 0.005). Moderate increases were seen bilaterally in scattered cells in layer II-III of the retrosplenial cortex (4.2-fold of control; P < 0.005) and layers II-III and VI of the perirhinal and temporal cortices as well as in neurons of the basolateral nucleus of the amygdala (4.7-fold of control; P < 0.005) and the dorsal endopiriform nucleus. Elevated BDNF mRNA expression was also detected in the hippocampal CA342AI pyramidal cells (2.8and 2.1-fold of control, respectively; P <0.05 for both). Widespread increases similar to those observed after the first grade 4 seizure were found 2 h after the third grade 5 seizure but had disappeared after four weeks (Figs 1D, E, 2A, B, 3). However, generalized seizures induced by one stimulation in rats which had experienced the third grade 5 seizure four weeks earlier caused increases in BDNF mRNA levels with the same distribution and magnitude as those observed after the first grade 4 and the third grade 5 seizure (Figs 1F, 3).

"~ 1~'

I

Fig. 2. Brightfield photomicrographs of the dentate gyrus granule cell layer from non-stimulated control rats (A, C, E, G, 1) and from animals subjected to the third grade 5 seizure, induced by hippocampal stimulation, 2 h before being killed (B, D, F, H, J). Sections were hybridized to BDNF mRNA (A, B), NGF mRNA (C, D), neurotrophin-3 mRNA (E, F), trkB all mRNA (G, H) and trkC mRNA (I, J). E and F show sections obtained from animals subjected to a sham FF lesion prior to hippocampal kindling. Scale bar = 20/~m.

Neurotrophin, trkA, trkB and trkC mRNA in kindling

Hippocampalklndlin~ 1000"

I /

i-/

I"'I ll~Im

/

1a~4

800" •

/

I'

/ /

4w

IW~ 4 w + I stem

I'

/ /

,

/

I' /

200'

/ / Io /

BDI~

NGF

u:kB

u:kC

Fig. 3. Levels of mRNA for BDNF, NGF, trkB and trkC in the dentate gyrus granule cell layer at different stages during the development of hippocampal kindling as measured by computerized image analysis of X-ray films. Animals were killed 2 h after the first stimulation (1 stim), the first grade 4 seizure (lst gr 4), the third grade 5 seizure (3rd gr 5), four weeks after the third grade 5 seizure (4w) and 2 h after a stimulation applied to animals which had experienced the third grade 5 seizure four weeks earlier (4w+ 1 stim). Means + S.E.M. of four animals in each group. Values are expressed relative to those measured in electrode-implanted control rats connected to the stimulating/recording apparatus without receiving stimulation (n = I0). Open circles denote individual animals with changes less than 1 S.D. from control. Nerve growth factor messenger RNA. The first stimulation gave rise to a 2.4-fold increase in N G F m R N A in the dentate granule cells in one out of four animals (the same animal that displayed an increase in B D N F m R N A expression; Fig. 3). Hybridization to N G F m R N A in the dentate gyrus increased to the same level following the first grade 4 and the third grade 5 seizure (Figs 2C, D, 3; different compared to control at P < 0.005 following the first grade 4 seizure). No change of N G F m R N A levels was detected outside the dentate gyrus after any stimulation. Four weeks after the third grade 5 seizure N G F m R N A expression was similar to control but a generalized seizure in such animals gave rise to an increase of N G F m R N A in the dentate gyrus of the same magnitude as that seen during the development of kindling (Fig. 3). TrkA messenger RNA. In non-stimulated control animals, detectable levels of trkA m R N A were confined to large, presumably cholinergic neurons in the medial septum and the diagonal band of Broca. 42'43'56 No change in trkA m R N A expression was seen in these regions 2 h after either the first stimulus-evoked seizure or generalized seizures induced by hippocampal kindling. TrkB messenger RNA. In agreement with previous studies 32'42'43control animals showed hybridization to the trkB all probe in dentate granule cells, hippocampal CA1 pyramidal neurons, pyramidal cells of the piriform cortex, amygdaloid complex and neo-

437

cortex. The ependymal lining of the ventricles also exhibited an intense labelling. Lower levels of trkB all m R N A were observed over neocortical layers I I - I I I and V-VI, thalamus and hypothalamus. The distribution of trkB full-length m R N A was similar to that of trkB all m R N A except that no labelling of the ependyma was detected. Seizure activity caused elevations in trkB all m R N A levels confined to the dentate granule cells in the same animals that displayed increases in BDNF and N G F m R N A (Figs 2G, H, 3). The magnitude of the increase in trkB all m R N A (varying between 1.8- and 2.2-fold of control levels) was similar after the first stimulation, the first grade 4 and the third grade 5 seizure and after a generalized seizure in animals kindled four weeks earlier (Fig. 3; P < 0.05 following the first grade 4 seizure). The increase in trkB all expression had disappeared four weeks after the third grade 5 seizure (Fig. 3). trkB Full-length m R N A was only analysed 2 h following the first grade 4 seizure. An increase in trkB full-length m R N A expression amounting to 1.5fold of control and confined to the dentate gyrus was seen in all animals. TrkC messenger RNA. In agreement with recent studies42,43 control animals displayed strong trkC m R N A labelling over the dentate granule cells, the hippocampal pyramidal layer, the cingulate cortex, the neocortical layers I I - I I I and the piriform cortex. Moderate labelling was seen over the neocortical layers IV-VI, the amygdaloid complex, the striatum,. the habenular complex, the ventromedial thalamic nucleus, the reticular thalamic nucleus, the hypothalamus and the dorsal and lateral hypothalamic areas. The first stimulation as well as the first grade 4 and third grade 5 seizure evoked a moderate increase of trkC m R N A (to 1.2-1.7-fold of control) in dentate granule cells in the same animals that displayed elevated levels of BDNF, N G F and trkB m R N A (Figs 2I, J, 3; P < 0.05 following the first grade 4 seizure). No changes in trkC m R N A were seen outside the dentate gyrus, trkC m R N A had returned to control levels four weeks after the third grade 5 seizure but stimulation in these animals caused increases similar to those observed after the first grade 4 and third grade 5 seizure (Fig. 3). Amygdalar kindling in non-lesioned rats Kindling characteristics. The AD threshold in the amygdala was 35 + 7/~A. In rats given a single stimulation, grade 0 seizures with an A D duration of 10 + 1 s were elicited, which was significantly shorter than that observed after stimulation in the hippocampus (P < 0.05). Also the number of stimulations needed to reach the first grade 4 (7 + 1) and the third grade 5 seizure (13 + 1) and their durations (74 + 15 and 42 + 11 s, respectively) were significantly less in the amygdala than in the hippocampus (P < 0.05). One stimulation in rats kindled four weeks earlier resulted in grade 5 seizures with an A D duration of 7 4 + 18s.

J. BENGZONet al.

438

A

,...

D

A 1000Amygdala~ ltindllne

T

800-

B

E

200

F

o 0

Amy

Fig. 4. Darkfield photomicrographs of autoradiograms comparing the expression of BDNF mRNA 2 h after the first stimulation (B, E) or the third grade 5 seizure (C, F) induced either by hippocampal (B, C) or amygdalar (E, F) kindling stimulation. Autoradiograms from animals implanted with electrodes in the hippocampus (A) or the amygdala (D) without being stimulated are shown for comparison. Scale bar = 3.8 mm.

B Dentate gyms

r cr jj~, .,-//

300 T

8

// J/ /J f/J

~,jj / J J

2OO i//

Brain-derived neurotrophic factor messenger RNA. Following the first stimulation in the amygdala, increased BDNF mRNA expression was seen ipsilaterally in the perirhinal, temporal, retrosplenial and piriform cortices, the basolateral nucleus of the amygdala and the dorsal endopiriform nucleus in two out of four animals (Fig. 4D). Bilateral increases, similar in magnitude to those seen in response to the first amygdalar stimulation and after hippocampal kindling, were found in these regions in response to the first grade 4 and the third grade 5 seizure in all animals (Figs 4D, 5A). The increases amounted to 3.3-, 8.0- and 3.l-fold of control in the retrosplenial cortex, the piriform cortex and the amygdaloid complex, respectively, after the third grade 5 seizure. In addition, after the first grade 4 and the third grade 5 seizure evoked by amygdalar stimulation, one out of four animals in each group showed bilateral increases of BDNF mRNA in the dentate gyrus and the pyramidal cell layer of the hippocampus. The increases amounted to 7.2-, 2.0- and 1.7-fold of control in the dentate gyrus and the CA3 and CA1 sectors, respectively. BDNF mRNA expression was similar to control four weeks after the third grade 5 seizure but one stimulation at this time-point elicited similar increases of BDNF mRNA levels to those observed after generalized seizures during kindling epileptogenesis. Nerve growth factor, trkA, trkB and trkC messenger RNA. No change in NGF, trkB all and trkC mRNA was seen 2 h after a single stimulation in the amygdala. Following grade 4 and 5 seizures, increased levels of NGF, trkB all and trkC mRNA confined to the dentate gyrus were found in the same animals that

//f

f f f f

J l / / / / f J

/D'/ f / J

lOO O

f f J f f J

NGF

f/,¢ f/.¢ f / J

~lr.B

~rN2

Fig. 5. Levels of mRNA for BDNF (A) and NGF, trkB all and trkC (B) 2 h after the third grade 5 seizure evoked by hippocampal or amygdalar kindling. All groups comprise four animals except rats kindled in amygdala and analysed for trkB all (n = 2) and trkC(n = 3) expression. Means __+ S.E.M. Values are expressed relative to levels in electrode implanted control rats connected to the stimulating/recording apparatus without receiving stimulation. Cx, layer II of retrosplenial cortex; Pir, piriform cortex; Amy, basolateral amygdaloid nucleus. Open circles denote individual animals with changes less than 1 S.D. from control. showed elevated BDNF mRNA levels in the hippocampal formation (Fig. 5B). The magnitude of the increases was comparable to that after hippocampal kindling. No change in trkA mRNA expression was seen.

Hippocampal kindling in 6-hydroxydopamine-treated rats Kindling characteristics. In the rats injected with 6-OHDA four weeks before electrode implantation, the threshold for eliciting AD was 61 _ 21 #A. The AD duration after a single stimulation was 81 _ 12 s and the seizure grade was 0-1;8 + 1 stimulations were required to elicit three grade 5 seizures and the AD duration of the third grade 5 seizure was 98 _ 5 s. The development of generalized (grade 4-5) seizures

Neurotrophin, trkA, trkB and trkC mRNA in kindling

D

B

C

E

F

(G)

439

Effect o f 6 - O H D A lesion

l-"l imcL non-ltlm

t ~ t , 3~ sr 5 te~m. 3rd8r5

*v

200"

NGF trkB trkC Fig. 6. Darkfield photomicrographs of autoradiograms showing BDNF mRNA expression in intact (A~C) and 6-OHDA-treated (D F) animals 2 h after the first stimulation (B, E) or the third grade 5 seizure (C, F) induced by hippocampal stimulation. Non-stimulated (A, D) animals are shown for comparisons. Scale bar = 3.8 mm. (G) Quantification of neurotrophin and trkB all and trkC mRNA expression in the dentate gyrus in non-stimulated 6-OHDA-treated rats (lesion, non-stim) as well as intact and 6-OHDA-treated animals 2 h after the first stimulation (1 stim) or the third grade 5 seizure (3rd gr 5). Means __+S.E.M. of four animals in each group. Values are expressed relative to intact, electrode implanted animals, connected to the stimulating/recording apparatus without being stimulated. Open circles denote individual animals in which changes less than 1 S.D. above the control level were seen. BDNF

in 6-OHDA-treated rats was significantly faster than in intact animals (P < 0.05). Neurotrophin, trkB, and trkC messenger RNA expression. The 6-OHDA lesion did not cause any change in mRNA levels for the neurotrophins or for trkB all and trkC in non-stimulated animals analysed five weeks after the lesion (Fig. 6A, D, G). However, significantly more 6-OHDA-treated animals exhibited increases in BDNF mRNA hybridization in the dentate gyrus in response to the first stimulation compared to intact rats (one out of four in the intact group vs all four 6-OHDA-treated rats; P < 0.01, Fischers exact test). In addition, all lesioned rats displayed bilateral increases in BDNF mRNA hybridization also in the hippocampal CA3 and CAI sectors, amounting to 2.3- and 2.0-fold of control, respectively, following the first stimulation (Fig. 6B, E). The magnitude of the seizure-induced increase of BDNF mRNA in responding animals was not influenced by the lesion (Fig. 6G). No change of BDNF mRNA expression was detected outside the

hippocampal formation. Following the third grade 5 seizure, all animals showed increases in BDNF mRNA identical in distribution and magnitude to those seen in non-lesioned animals (Fig. 6C, F, G). Bilateral increases in NGF, trkB all and trkC mRNA expression were observed in the dentate granule cell layer both after the first stimulation and the third grade 5 seizure in all 6-OHDA-treated animals (Fig. 6G). Thus, compared to intact rats the lesion caused an increased number of animals to respond with elevated levels also of NGF, trkB all and trkC mRNA. Furthermore, increased hybridization to trkB all mRNA was seen in the hippocampal CA3-CA1 sectors in 6-OHDA-treated rats. There was no significant change in the magnitude of the increases after either the first stimulation or the third grade 5 seizure compared to intact animals (Fig. 6G).

Hippocampal kindling in fimbria-fornix lesioned rats Kindling characteristics. The threshold to induce an AD was 89 + 31/~A in rats with F F lesions and

J. BENGZON et al.

440

C

B

D

F-

E

F

Fig. 7. Darkfield photomicrographs of autoradiograms showing the expression of BDNF mRNA in FF-lesioned (D-F) and sham-operated rats (A42) 2 h following the first stimulation (B, E) or the third grade 5 seizure (C, F) evoked by hippocampal stimulation. Non-stimulated controls are shown for comparison (A, B). Scale bar = 3.8 mm.

lpsi]aterai to stimulation

Contralaterai to slimulation

A

B

100(Y

1200-

l---I I~, aca-~n

GEl ~ l l

lO00-

80~

8oo" 601Y

600" 40lY

4Oo201Y

2O0" 0 BDNF

NGF

HT-3

ukB

ukC

BDNF

NGF

NT-3

trkB

C 1200" I""I FF, non-stkn

1000"

o'-A sh~, ~rdSr5 gOZ ~,~dar5

800"

400"

7.00"

BDNF

NGF

NT-3

IrkB

trkC

Fig. 8. Neurotrophin (NT) and t r k m R N A expression in the dentate gyrus in non-stimulated rats with a bilateral FF lesion (FF, non-stim; n = 4) and in sham-operated (sham; n = 2) and FF-lesioned animals (FF; n = 6) 2h after the first stimulation (1 stim; A, B) or the third grade 5 seizure (3rd gr 5; C). Means + S.E.M. Open circles denote individual animals with changes less than 1 S.D. from control level. Values are expressed relative to sham-operated animals implanted with electrodes and connected to the stimulating/recording apparatus without receiving stimulation.

IxkC

Neurotrophin, trkA, trkB and trkC mRNA in kindling 31 _ 4 pA in sham-operated animals. Following the first stimulation the AD duration was 62 + 10 s and 58 + 36 s in the FF-lesioned and sham-operated rats, respectively, and the seizure grade was 0 in both groups. In contrast to sham-operated animals, there was no spread of seizure activity to the contralateral hippocampus after the first stimulation in five out of six animals with F F lesions as evidenced by EEG recordings. The FF-lesioned and sham-operated rats needed 20 + 2 and 24 + 3 stimulations, respectively, to reach the third grade 5 seizure, the duration of which was 104 _+ 5 and 118 _ 6 s, respectively. In the FF-lesioned animals, seizure spread to the contralateral hippocampus was seen after 4.6 + 0.6 stimulations while all sham-operated animals showed an immediate spread in response to the first stimulation (significantly faster at P < 0.005). Neurotrophin, trkB and trkC messenger RNA expression. In non-stimulated rats forebrain levels of BDNF, NGF, neurotrophin-3, trkB all and trkC mRNA were not affected three weeks after the FFlesion (Figs 7A, D, 8A~C). After a single hippocampal stimulation BDNF mRNA expression was increased ipsilaterally in the dentate gyrus in four out of the five animals with FF-lesions displaying unilateral EEG seizure activity (Figs 7B, E, 8A, B). In the remaining FF-lesioned animal with unilateral seizure activity no change in BDNF mRNA was seen. The number of animals showing increased BDNF mRNA levels in the dentate granule cells in response to the first stimulation was significantly higher in the

441

FF-lesioned group than in the group of intact rats (P < 0.01). In the FF-lesioned animal showing contralateral spread of seizure activity after the first stimulation, bilateral increases of hybridization of BDNF mRNA were observed in the dentate gyrus. The magnitude of the increase in BDNF mRNA levels in responding rats (9.2- and 9.8-fold of control in FFlesioned and sham-operated groups, respectively) was not affected by the lesion (Fig. 8A, B). After the third grade 5 seizure, elevated levels of BDNF mRNA were detected in all rats with F F lesions in the same regions (dentate gyrus, hippocampal pyramidal layer, neocortical regions, piriform cortex, amygdaloid complex) as in sham-operated and in intact animals (Figs 7C, F, 8C). The magnitude of the increase did not differ between FF-lesioned and sham-operated rats (Fig. 8C). A significant reduction of neurotrophin-3 mRNA expression compared to non-stimulated rats was found in dentate granule cells both in sham-operated (44% reduction ipsilateral to stimulation) and FFlesioned (22% reduction) animals after the first stimulation (Figs 2E, F, 8A, B). In the four animals with unilateral seizure activity in response to the first stimulation and exhibiting increased BDNF mRNA expression on this side, the neurotrophin-3 mRNA level was decreased ipsilateral to the stimulation (Fig. 8A, B). No change was seen in the contralateral dentate gyms. In contrast, the single animal showing bilateral seizure activity after the first stimulation also exhibited bilateral decreases in neurotrophin-3

F

Fig. 9. Darkfield photomicrographs of autoradiograms showing the concomitant changes in NGF (A42), BDNF (D-F), neurotrophin--3 (G-l), trkBall (J L) and trkC (M43) mRNA expression in the dentate gyrus granule cell layer in a representative single animal 2 h after the third grade 5 seizure with an AD duration of 94 s (C, F, I, L, O) and the lack of any detectable changes compared to control (A, D, G, J, M) in an identically treated animal with an AD duration of 22 s (B, E, H, K, N). Scale bar = 1.6 mm.

442

J. BENGZONet al.

m R N A expression. After the third grade 5 seizure, hybridization to neurotrophin-3 m R N A was reduced in the dentate gyrus in all FF-lesioned and shamoperated animals (Figs 7C, F, 8C, 9). The reductions amounted to 75 and 52% in the lesioned and shamoperated groups, respectively (Fig. 8C; FF-lesioned animals different compared to control at P < 0.05). Uni- or bilateral increases of N G F , trkB all and trkC m R N A levels were found in the same dentate gyri that displayed changes in BDNF and neurotrophin-3 m R N A expression after one stimulation in the FF-lesioned and sham-operated animals (Fig. 8A, B, 9). In rats with F F lesions, the increases ipsilateral to stimulation amounted to 1.9-, 3.5- and 2.4-fold compared to control for N G F , trkB all and trkC respectively (Fig. 8B). Bilateral increases in hybridization to these probes were found in the dentate gyrus in all FF-lesioned rats following the third grade 5 seizure. There was no difference between FF-lesioned and sham-operated animals in the magnitude of the changes of m R N A levels after either the first stimulation or the third grade 5 seizure (Fig. 8C). Correlation between seizure duration and brain -derived neurotrophic factor messenger RNA expression in the dentate gyrus A correlation analysis was performed in order to elucidate the relationship between AD duration and BDNF m R N A level in the dentate gyrus. All image analysis values obtained 2 h post-seizure in intact, 6-OHDA-treated and FF-lesioned animals following the first stimulation and the third grade 5 seizure were pooled. AD durations exceeding 70 s (as determined using Fischers exact test) triggered an increase in B D N F m R N A amounting to 836___ 80% of control (P < 0.001), while seizure-activity of shorter duration 50(~ i

OqP

•

~0

O0

•

7%

•

30~

2000"

11300" o~ •

0

|

20

oo -

J

40

,,o -

i

60

I

| 80

.

i 100

.

• 120

.

i

-

i

140 160 AD duration (s)

Fig. 10. Correlation between AD duration (including secondary ADs) and the level of BDNF mRNA in the dentate gyrus 2 h after hippocampal kindling stimulations. Image analysis and EEG data obtained from intact, 6-OHDA-treated and FF-lesioned animals at different seizure grades were pooled. According to Fischers exact test BDNF mRNA expression is increased when the AD duration exceeds 70 s. BDNF mRNA expression in electrode-implanted, non-stimulated animals (ctrl; means ___S.E.M. of 22 animals) is shown for comparison.

failed to evoke any significant increase (defined as + 1 standard deviation (S.D.) from control) (Fig. 10). DISCUSSION

The present study demonstrates concomitant, transient increases in m R N A expression for N G F , BDNF, trkB and trkC and a decrease in neurotrophin-3 m R N A in dentate granule cells in response to focal and generalized seizures induced by hippocampal kindling stimulation. After generalized seizures, BDNF m R N A was also increased in the hippocampal pyramidal layer, the basolateral amygdala, the piriform cortex, and the perirhinal, retrosplenial and temporal cortices. No change in trkA m R N A was seen. The magnitude of the seizureevoked changes of m R N A levels was similar through the different stages of kindling and in the fully kindled brain. Amygdalar stimulation induced increased BDNF m R N A levels in the amygdala and the piriform and neocortices while altered neurotrophin, trkB and trkC m R N A levels in the dentate gyrus were observed only in a minority of animals. Lesions of the noradrenergic system and the F F had no effect on m R N A levels for the neurotrophins or trk receptors but increased the number of animals showing elevated levels in response to the first stimulation. Neurotrophin, trkA, trkB and trkC messenger RNA expression after hippocampal kindling Marked increases of BDNF m R N A levels were found in the dentate granule cells following generalized seizures, but were only observed in a minority of animals after the first stimulation. This difference in response is most readily explained by the mean duration of epileptiform activity being significantly longer in the groups exhibiting grade 4 and 5 seizures than in the rats subjected to the first stimulation. The correlation analysis indicated that, in the traditional kindling model, seizure duration exceeding 70s causes a very marked elevation of BDNF m R N A expression in the dentate gyrus irrespective of the grade of kindling development. In contrast, following seizure activity with duration less than 70 s, B D N F m R N A expression is unaffected. The magnitude of the seizure-evoked increase of BDNF m R N A was uniform throughout the stages of kindling with little variation between individual, responding animals. In studies using in situ hybridization techniques comparable to those in the present study, increases in BDNF m R N A expression of similar magnitude have been reported after kainate and dentate hilar lesions9'25also indicating that in these epilepsy models, seizure-evoked induction of BDNF m R N A is an "all-or-none"-like response. Generalized seizures also gave rise to increases of BDNF m R N A levels in several regions outside the hippocampus. The localization of these changes coincided with that observed after 40 recurrent seizures in rapid hippocampal kindling, i.e. they were found

Neurotrophin, trkA, trkB and trkC mRNA in kindling bilaterally in the dentate gyrus and pyramidal cell layer of the hippocampus, the basolateral amygdaloid nucleus, dorsal endopiriform nucleus and the piriform, perirhinal, retrosplenial and temporal cortices. In previous studies ~°,25 short periods of seizure activity have been reported to markedly increase B D N F m R N A levels in dentate granule cells, whereas such changes in cortical areas have only been observed after recurrent seizures during several hours produced by kindling or a hilus lesion. The present data indicate, however, that one brief, generalized seizure giving rise to elevated BDNF m R N A expression in the dentate gyrus also induces increased BDNF m R N A levels in other forebrain areas. The same mechanism may thus trigger the increase in BDNF m R N A expression in all responsive regions. Increased hybridization to N G F , trkB all and trkC m R N A was seen in the dentate gyrus of animals which showed elevated BDNF m R N A expression following grades 0, 4 and 5 seizures, indicating similar regulation of transcription for the different genes (Fig. 10). The trkB full-length probe showed the same hybridization pattern as the trkB all probe. No change in the levels of trkA m R N A was demonstrated, which agrees well with our previous findings in rapidly kindled animals. 43 Although increases in N G F , trkB and trkC m R N A levels were smaller than that seen for B D N F mRNA, they were relatively uniform at all seizure grades during kindling development. No changes were detected outside the hippocampus, e.g. in the parietal and piriform cortices, as has been observed for N G F m R N A after repeated seizures in rapid kindling) ° This discrepancy might be due to the present 2 h post-seizure period being too short to observe the changes of N G F m R N A levels in cortical regions. Both in the dentate hilus lesion25 and rapid kindling models t° increases of BDNF m R N A expression occur several hours earlier than those of N G F m R N A in extrahippocampal loci. The effect of seizure activity induced by hippocampal stimulation on neurotrophin-3 m R N A expression was analysed in the F F lesion experiment. Both in the sham-operated and FF-lesioned groups there was a clear reduction of neurotrophin-3 m R N A levels in the dentate granule cells after one stimulusevoked, focal seizure and after the third grade 5 seizure. Reduced levels of neurotrophin-3 m R N A in the dentate gyrus have been reported after recurrent seizures produced by a dentate hilus lesion49 and after cerebral ischemia and hypoglycemic coma. 38 The present findings show that reduced expression of neurotrophin-3 also occurs following kindled seizures. Neurotrophin, trkA, trkB and trkC messenger RNA expression after amygdalar kindling Two main differences in the pattern of changes of BDNF m R N A expression were observed between amygdalar and hippocampal kindling. First, generalized seizures evoked by stimulation in the amygdala

443

gave rise to increased BDNF m R N A expression in dentate granule cells only in single rats whereas increases were always found after hippocampal kindling. The lack of elevated levels of B D N F m R N A in the hippocampus in amygdala-kindled animals indicates that neurotrophic mechanisms in this structure are not critical for the development of amygdalar kindling. This interpretation agrees well with findings that lesions of the hippocampus have no effect on amygdala kindling. 4s Second, already after the first amygdalar stimulation there was a marked increase of BDNF m R N A levels in the amygdala-piriform cortex in two out of four animals. This region has been proposed to play an important role in the development and expression of kindled seizures irrespective of the site of stimulationY With repeated stimulations during kindling development the responsiveness of the neurons in the piriform area is altered, they generate strong burst responses to constant stimuli and recruit many other structures into the discharge, a7 The change in BDNF m R N A levels reported here, and the presumed subsequent increase in BDNF protein may lead to synaptic reorganization thereby providing a possible mechanism underlying the altered responsiveness of the cells in the piriform lobe. If so, the earlier activation of neurotrophic mechanisms in this region after stimulation in the amygdala compared to hippocampus might, at least partly, explain the more rapid development of seizures in amygdalar than in hippocampal kindling. Effects o f 6-hydroxydopamine and fimbria-fornix lesions on neurotrophin, trkB and trkC messenger RNA expression Specific transmitter systems have been proposed to regulate synthesis of BDNF and N G F in hippocampal neurons both under physiological conditions and in response to seizure activity. Glutamatergic mechanisms seem to be importantly involved since blockade of glutamate receptors reduces basal BDNF and N G F m R N A levels59 and attenuates the seizureinduced increase in BDNF and N G F m R N A expression both in vitro 6° and in vivo. 1°'59 Furthermore, a quisqualate injection in the entorhinal cortex, which may activate glutamatergic afferents to the hippocampus, markedly elevates BDNF m R N A levels in hippocampal neurons. 37 Also cholinergic neurons might stimulate BDNF and N G F synthesis since carbachol increases BDNF m R N A levels in vitro 6° and quisqualate injection in the septum, probably activating the cholinergic system, leads to markedly elevated BDNF and N G F m R N A levels in the hippocampus, which can be blocked by scopolamine. 37 The GABAergic system on the other hand, may suppress N G F and BDNF synthesis since administration of GABA agonists to cultured hippocampal neurons decreases BDNF and N G F m R N A levels. 59 In this study we explored whether lesions of afferent systems to the hippocampus, modulating the excitability of hippocampal neurons, influence basal levels and

444

J. BENGZON et al.

seizure-evoked changes of neurotrophin and trk m R N A expression in kindling. The intraventricular 6-OHDA injection permanently depletes the forebrain, including the hippocampal formation, of more than 95 % of its noradrenaline content 6 which leads to a marked facilitation of kindling rate. 4° The F F lesion transects the subcortical efferent projection of the hippocampal formation and removes several afferent systems including cholinergic, GABAergic, noradrenergic and serotonergic projections, as well as minor projections from subcortical nuclei and commisural projections. 7 Although no change of kindling rate was found here, animals with a fimbria-fornix lesion have previously been reported to show chronic interictal spikes and an increased susceptibility to seizures7 probably due to the removal of subcortical inhibitory inputs. The lesions caused no long-term changes of m R N A expression for either the neurotrophins or their receptors in non-stimulated animals. Thus, the different afferent systems severed by the lesions are not critical for maintenance of physiological levels of these molecules in the hippocampus. Significantly more animals showed changes of N G F , BDNF, trkB all and trkC m R N A levels after the first stimulation in the 6-OHDA treated and FF-lesioned groups than in an intact control group. In contrast, after generalized seizures there were no differences in the number of responding animals or in the distribution of mRNA changes. Furthermore, the magnitude of the increases was not affected by the lesions either after focal or generalized seizures. It therefore seems less likely that hippocampal afferents such as the noradrenergic, cholinergic, serotonergic and GABAergic systems directly regulate gene expression in this region. Instead neurotrophin and trkB and trkC receptor synthesis is probably modulated indirectly via an influence of the lesions on the excitability of the region, as evidenced, for example, by a change of seizure duration. The noradrenergic system might act by shortening the duration of A D early in the kindling process, thereby indirectly suppressing the changes of m R N A for the neurotrophins and their receptors and possibly also presumed secondary plastic responses. CONCLUSIONS

The present data indicate that seizure activity sets in motion a constellation of changes in gene expression for the neurotrophins and their receptors, which in dentate granule cells includes increased BDNF, N G F , trkB and trkC m R N A levels and decreased neurotrophin-3 m R N A level whereas trkA m R N A expression is unaltered. The concomitant increased expression of trkB and BDNF m R N A in the dentate gyrus, as previously shown after rapid kindling, cerebral ischemia and hypoglycemic coma 43 may suggest a ligand-induced activation of the receptor, i.e. that increased levels of BDNF protein occurring very rapidly after the insults directly induce

increased levels of trkB mRNA. 19 However, trkC receptor synthesis is regulated differently since elevated levels of trkC m R N A were detected together with reduced neurotrophin-3 m R N A expression in the dentate gyrus. From these data it seems more likely that a common mechanism is responsible for the elevated levels of BDNF, N G F , trkB and trkC mRNAs and the suppression of the rate of transcription of the neurotrophin-3 gene. The correlation analysis indicated that the seizureinduced activation of the B D N F gene is an "all or none" type of response, occurring in the kindling model when epileptiform activity lasts longer than 70 s. In the dentate gyrus probably the entire cascade of genomic events is regulated in the same way, since changes of N G F , neurotrophin-3, trkB and trkC mRNAs were always observed concomitantly with altered BDNF m R N A expression. The regulation of neurotrophin and trk m R N A levels seems to be different in cortical areas outside the hippocampus, where brief periods of generalized seizures gave rise to marked increases of BDNF m R N A levels while hybridization to the other probes was unaffected. The changes in gene expression for the neurotrophins after seizures, cerebral ischemia and hypoglycemia have been proposed to be induced by glutamate release and increased influx of calcium ions. 1°'38'6° It is possible that these changes are triggered when the calcium level in certain cellular compartments has reached a critical threshold value. There is now good evidence that the seizureinduced changes of m R N A expression lead to the presumed alterations of the levels of the neurotrophin and receptor proteins. Following rapid kindling, N G F protein content is increased in the dentate gyrus and parietal and piriform cortices4 and the level of trkB protein is elevated in the hippocampus. 43 Furthermore, a marked increase of tyrosine phosphorylated trkB protein is found in the dentate gyrus after seizures43 indicating an activation of the receptor. The functional consequences of the observed changes of neurotrophins and their receptors during kindling development and in response to seizure activity are still unclear. The parallel increase of BDNF m R N A and activated trkB protein in the dentate granule cells is suggestive of an autocrine or paracrine action locally in the hippocampus. An important role for plastic changes and synaptic reorganization occurring during kindling epileptogenesis seems possible, but neurotrophic mechanisms could also be involved in neuronal protection following epileptic insults. Acknowledgements--We thank Agneta Persson and Bengt Mattsson for help with the illustrations and professor Jan Lanke for expert statistical advice. This work was supported by grants from the Swedish Medical Research Council (14X-8666), the Royal Physiographic Society, the Swedish Natural Science Research Council, the Bank of Sweden Tricentenary Fund, the Swedish Society for Medical Research, the Royal Swedish Academy of Sciences, the European Science Foundation, and the Elsa & Thorsten Segerfalk Foundation.

Neurotrophin, trkA, trkB and trkC mRNA in kindling

445

REFERENCES

I. Ballarin M., Ernfors P., Lindefors N. and Persson H. (1991) Rapid induction of mRNA for brain derived neurotrophic factor after intrahippocampal injection of kainic acid. Expl Neurol. 114, 35-43. 2. Barde Y.-A. (1989) Trophic factors and neuronal survival. Neuron 2, 1525-1534. 3. Barde Y.-A., Edgar D. and Thoenen H. (1982) Purification of a new neurotrophic factor from mammalian brain. Eur. molec. Biol. Org. J. 1, 549-553. 4. Bengzon J., Srderstrrm S., Kokaia Z., Kokaia M., Ernfors P., Persson H., Ebendal T. and Lindvall O. (1992) Widespread increase of nerve growth factor protein in the rat forebrain after kindling-induced seizures. Brain Res. 587, 338-342. 5. Berkemeier L. R., Winslow J. W., Kaplan D. R., Nikolics K., Goeddel D. V. and Rosenthal A. (1991) Neurotrophin-5: a novel neurotrophic factor that activates trk and trkB. Neuron 7, 1031-1041. 6. Bjbrklund A., Nornes H. and Gage F. H. (1986) Cell suspension grafts of noradrenergic locus coeruleus neurons in rat hippocampus and spinal cord: reinnervation and transmitter turnover. Neuroscience lg, 685-698. 7. Buszaki G., Ponomareff G., Bayardo F., Shaw T. and Gage F. G. (1988). Suppression and induction of epileptic activity by neuronal grafts. Proc. natn. Acad. Sci. U.S.A. 85, 9327-9330. 8. Cavazos J. E., Golarai G. and Sutula T. P. (1991)Mossy fiber synaptic reorganization induced by kindling: time-course of development, progression and permanence. Neuroscience 11, 2795-2803. 9. Dugich-Djordjevich M. M., Tocco G., Lapchak P. A., Pasinetti G. M., Najm I., Baudry M. and Hefti F. (1992) Regionally specific and rapid increases in brain-derived neurotrophic factor messenger RNA in the adult rat brain following seizures induced by systemic administration of kainic acid. Neuroscience 47, 303-315. 10. Ernfors P., Bengzon J., Kokaia Z., Persson H. and Lindvall O. (1991) Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron 7, 165-176. 11. Ernfors P., Henschen A., Olson L. and Persson H. (1989) Expression of nerve growth factor receptor messenger RNA is developmentally regulated and increased after axotomy in rat spinal cord motoneurons. Neuron 2, 1605 1613. 12. Ernfors P., Ibanez C. F., Ebendal T., Olson L. and Persson H. (1990) Molecular cloning and neurotrophic activities of a protein with structural similarities to nerve growth factor: developmental and topographical expression in the brain. Proc. natn. Acad. Sci. U.S.A. 87, 5454-5458. 13. Ernfors P., Wetmore C., Olson L. and Persson H. (1990) Identification of cells in rat brain and peripheral tissues expressing mRNA for members of the nerve growth factor family. Neuron 5, 511-526. 14. Funabashi T., Sasak H. and Kimura F. (1988) Intraventricular injection of antiserum to nerve growth factor delays the development of amygdaloid kindling. Brain Res. 458, 132-136. 15. Gall C. and Isackson P. J. (1989) Limbic seizures increase neuronal production of messenger RNA for nerve growth factor. Science 24, 758~61. 16. Gall C., Murray K. and Isackson P. J. (1991) Kanic acid-induced seizures stimulate increased expression of nerve growth factor mRNA in rat hippocampus. Brain Res. 9, 113-123. 17. Gastaut H., Gastaut J. L., Concalves e Silva G. E. and Fernandez Sanches G. R. (1975) Relative frequency of different types of epilepsy: a study employing the classification of the International League Against Epilepsy. Epilepsia 16, 457-461. 18. Goddard G. V. (1969) A permanent change in brain function resulting from daily electrical stimulation. Expl Neurol. 25, 295-330. 19. Hadcock J. R. and Malbon G. (1991) Regulation of receptor expression by antagonists: transcriptional and post-transcriptional controls. Trends Neurosci. 14, 242 247. 20. Hallbrbk F., Ib~ifiez C. F. and H. Persson (1991) Evolutionary studies on the nerve growth family reveal a novel member abundantly expressed in Xeonpus ovary. Neuron 68, 845 858. 21. Hempstead B., Martin-Zanca D., Kaplan D. R., Parada L. F. and Chao M. V. (1991) High-affinity NGF binding requires coexpression of the trk protooncogene and the low-affinity NGF receptor. Nature 350, 678-683. 22. Hofer M., Paglusi S. R., Hohn A., Leibrock J. and Barde Y.-A. (1990) Regional distribution of brain-derived neurotrophic factor mRNA in the adult mouse brain. Eur. molec. Biol. Org. J. 9, 2459-2464. 23. Hohn A., Leibrock J., Bailey K. and Barde Y.-A. (1990) Identification and characterization of a novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature 344, 339-341. 24. Ip N., Ibfinez C., Nye S., McClain J., Jones P., Gies D., Belluscio L., LeBeau M., Espinosa R., Squinto S., Persson H. and Yancopoulos G. (1992) Mammalian n e u r o t r o p h i n ~ : structure, chromosomal localization, tissue distribution, and receptor specificity. Proc. natn. Acad. Sci. U.S.A. 89, 3060-3064. 25. Isackson P. J., Huntsman M., Murray K. and Gall C. (1991) BDNF mRNA expression is increased in adult rat forebrain after limbic seizures: temporal patterns of induction distinct from NGF. Neuron 6, 937-948. 26. Jones K. R. and Reichardt L. F. (1990) Cloning and expression of a eDNA encoding a novel human neurotrophic factor family. Proc. natn. Acad. Sci. U.S.A. 87, 8060-8064. 27. Kaplan D., Hempstead B. L., Martin-Zanca D., Chao M. V. and Parada L. F. (1991) The trk proto-oncogene-product: a signal transducing receptor for nerve growth factor. Science 252, 554-558. 28. Kaplan D., Martin-Zanca D. and Parada L. (1991) Tyrosine phosphorylation and tyrosine kinase activity of the trk proto-oncogene product induced by NGF. Nature 350, 158 160. 29. Klein R., Conway D., Parada L. F. and Barbacid M. (1990) The trk tyrosine protein kinase gene codes for a second neurogenic receptor that lacks the catalytic kinase domain. Cell 61, 647-656. 30. Klein R., Jing S. Q., Nanduri V., ORourke E. and Barbacid M. (1991) The trk proto-oncogene encodes a receptor for nerve growth factor. Cell 65, 189-197. 31. Klein R., Nanduri V., Jing S., Lamballe F., Tapely P., Bryant S., Cordon-Cardo C., Jones K. R., Reichart L. F. and Barbacid M. (1991) The trkB tyrosine proteine kinase is a receptor for brain-derived neurotrophic factor and neurotrophin--3. Cell 66, 395-403. 32. Klein R., Parada L. F., Coulier F. and Barbacid M. (1989) TrkB, a novel tyrosine proteine protein kinase receptor expressed during mouse neural development. Eur. molec. Biol. Org. J. 8, 3701-3709. 33. Kokaia M., Kal~n P., Bengzon J. and Lindvall O. (1989) Noradrenaline and 5-hydroxytryptamine release in the hippocampus during seizures induced by hippocampal kindling stimulation: an in vivo microdialysis study. Neuroscience 32, 647-656.

446

J. BENGZON et al.

34. Lamballe F., Klein R. and Barbacid M. (1991) TrkC a member of the trk family of tyrosine protein kinases in a receptor for neurotrophin--3. Cell 66, 967-979. 35. Leibrock J. L., Lottspeich F., Hohn A., Hofer M., Hengerer B., Masiakowski P., Thoenen H. and Barde Y.-A. (1989) Molecular cloning and expression of brain-derived neurotrophic factor. Nature 341, 149-152. 36. Levi-Montalcini R. (1987) The nerve growth factor 35 years later. Science 237, 1154-1162. 37. Lindefors N., Ernfors P., Falkenberg T. and Persson H. (1991) Septal cholinergic afferents regulate expression of brain-derived neurotrophic factor and b-nerve growth factor mRNA in rat hippocampus. Expl Brain Res. 88, 78-90. 38. Lindvall O., Ernfors P., Bengzon J., Kokaia Z., Smith M.-L., Siesj6 B. K. and Persson H. (1992) Differential regulation of mRNAs for nerve neurotrophin-3 in the adult rat brain following cerebral ischemia and hypoglycemic coma. Proc. natn. Acad. Sci. U.S.A. 89, 648-652. 39. Maisonpierre P. C., Belluscio L., Squinto S., Ip N. Y., Furth M. E., Lindsay M. and Yancopolous G. D. (1990) Neurotrophin--3: a neurotrophic factor related to NGF and BDNF. Science 247, 1446-1451. 40. Mclntyre D. C., Saari S. M. and Pappas B. A. (1979) Potentiation of amygdala kindling in adult or infant rats by injections of 6-hydroxydopamine. Expl Neurol. 63, 527-544. 41. McNamara J. O. (1984) Kindling: an animal model of complex partial epilepsy. Ann. Neurol. 16, 62-76. 42. Merlio J.-P., Ernfors P., Jaber M. and Persson H. (1993) Molecular cloning and identification of cells expressing mRNAs for members of the trk family in the rat central nervous system. Neuroscience (submitted). 43. Merlio J.-P., Ernfors P., Kokaia Z., Middlemas D. S., Bengzon J., Kokaia M., Smith M.-L., Siesj6 B. K., Hunter T., Lindvall O. and Persson H. (1993) Increased production of the trkB protein-tyrosine kinase receptor after brain insults. Neuron (in press). 44. Middlemas D. S., Lindberg R. A. and Hunter T. (1991) TrkB, a neural receptor protein-tyrosine kinase: evidence for a full-length and two truncated receptors. Molec. Cell Biol. 11, 143 153. 45. Phillips H. S., Hains J. M., Laramee G. R., Rosenthal A. and Winslow J. W. (1990). Widespread expression of BDNF but not NT-3 by target areas of basal forebrain cholinergic neurons. Science 250, 290594. 46. Racine R. J. (1972) Modification of seizure activity by electrical stimulation: II. Motor seizure. Electroenceph, clin. NeurophysioL 32, 281-294. 47. Racine R. J. and Burnham W. M. (1984) The kindling model. In Electrophysiology of Epilepsy (eds Schwartzkroin P. A. and Wheal H.), pp. 153-171, Academic Press, London. 48. Racine R. J., Paxinos G., Mosher J. M. and Kariss E. W. (1988) The effects of various lesions and knife-cuts on septal and amygdala kindling in the rat. Brain Res. 454, 264-272. 49. Rocamora N., Palacios J. M. and Mengold G. (1992) Limbic seizures induce a differential regulation of the expression of nerve growth factor, brain-derived neurotrophic factor and neurotrophin-3, in the rat hippocampus. Brain Res. 13, 27-33. 50. Rosenthal A., Goeddel D. V., Nguyen T., Lewis M., Shih A., Laramee G. R., Nikolics K. and Winslow J. W. (1990) Primary structure and biological activity of a novel human neurotrophic factor. Neuron 4, 767-773. 51. Soppet D., Escandon E., Maragos J., Middlemas D. S., Reid S. W., Blair J., Burton L. E., Stanton B. R., Kaplan D. R., Hunter T., Nikolics K. and Parada U F. (1991) The neurotrophic factors brain-derived neurotrophic factor and neurotrophin-3 are ligands for the trkB tyrosine kinase receptor. Cell 65, 895 903. 52. Squinto S., Stitt T., Aldrich T., Davis S., Bianco S., Radieziejewski C., Glass D., Masiakowski P., Furth M., Valenzuela D., DiStefano P. and Yancopoulos G. (1991) TrkB encodes a functional receptor for brain-derived neurotrophic factor and neurotrophin-3 but not nerve growth factor. Cell 65, 885-553. 53. Stenevi U., Bj6rklund A. and Svengaard N.-A. (1976) Transplantation of central and peripheral monoamine neurons to the adult brain: techniques and conditions for survival. Brain Res. 114, 1 20. 54. Sutula T. P. (1990) Experimental models of temporal lobe epilepsy: new insights from the study of kindling and synaptic reorganization. Epilepsia 31, $45 $54. 55. Sutula T. P., He X. X., Cavazos J. and Scott G. (1988) Synaptic reorganization in the hippocampus induced by abnormal functional activity. Science 239, 1147-1150. 56. Vazquez M. and Ebendal T. (1991) Messenger RNAs for trk and the low-affinity NGF receptor in rat basal forebrain. NeuroReport 2, 593 596. 57. Wada J. A., Sato M. and Corcoran M. E. (1974), Persistent seizure susceptibility and recurrent spontaneous seizures in kindled cats. Epilepsia 15, 465-478. 58. Wetmore C., Ernfors P., Persson H. and Olson L. (1990), Localization of brain-derived neurotrophic factor mRNA to neurons in the brain by in situ hybridization. Expl Neurol. 10, 141 152. 59. Zafra F., Castr6n E., Thoenen H. and Lindholm D. (1991) Interplay between glutamate and 7-aminobutyric acid transmitter systems in the physiological regulation of brain-derived neurotrophic factor and nerve growth factor synthesis in hippocampal neurons. Proc. natn. Acad. Sci. U.S.A. 88, 10,037-10,041. 60. Zafra F., Hengerer B., Leibrock J., Thonen H. and Lindholm D. (1990) Activity dependent regulation of BDNF and NGF mRNAs in the rat hippocampus is mediated by non-NMDA glutamate receptors. Eur. molec. Biol. Org. J. 9, 3545-3550. (Accepted 16 November 1992)