cholinergic regulation of hippocampal brain-derived neurotrophic factor mrna expression: evidence...

11
Neuroscience Vol. 52, No. 3, pp. 515-585, 1993 Printed in Great Britain 0306-4522/93 $6.00 + 0.00 Pergamon Press Ltd J? 1993 IBRO CHOLINERGIC REGULATION OF HIPPOCAMPAL BRAIN-DERIVED NEUROTROPHIC FACTOR mRNA EXPRESSION: EVIDENCE FROM LESION AND CHRONTC CHOLINERGIC DRUG TREATMENT STUDIES P. A. LAPCHAK,* D. M. ARAUJO and F. HEFTI Division of Neurogerontology, Andrus Gerontology Center, University of Southern California. University Park, MC-0191, Los Angeles. CA 90089-0191. U.S.A. A~~act-Quantitative in situ hybridi~tion and northern blot analysis techniques were used to determine the effects of removal of the cholinergic input on levels and topographical distribution of brain-derived neurotrophic factor mRNA in the hippocampus of adult rats. First, the effects of partial and full fimbriat transections, which result in partial and near-total cholinergic deafferentation respectively, were compared. Twenty-one days after partial unilateral fimbrial transections, there were significant decreases in brain-derived neurotrophic factor mRNA expression throughout the hippocampal formation. Decreased expression of brain-derived neurotrophic factor mRNA was evident in all areas of localization within the hippocampal formation. The decreases amounted to 22-36% reductions compared with unlesioned control animals. Brain~erived neurotrophic factor mRNA levels were decreased to a greater extent (S&69%) following full unilateral fimbrial transections. Quantitative northern blot analysis indicated that hippocampal BDNF mRNA was decreased by 29 and 68%. three weeks after partial or full unilateral fimbrial transections, respectively. The extent of the reductions in brain-derived neurotrophic factor mRNA levels correlated with reductions in acetylcholinesterase staining density and cholinergic terminal density determined by quantitative autoradiographic analysis of [‘Hlvesamicol binding sites. Second, we found that chronic treatment with atropine (20mg/kg per day for 14 days) decreased (by 54%) brain-derived neurotrophic factor mRNA levels in all areas of localization within the hippocampus. In contrast, chronic treatment with nicotine (1.18 mg/kg per day for 14 days), a treatment known to desensitize nicotinic receptors, did not affect brain-derived neurotrophic factor mRNA expression in the hippocampal formation. The findings provide evidence for cholinergic muscarinic regulation of brain-derived neurotrophic factor mRNA expression in the adult rat hippocampal formation and they suggest the existence of a tonic stimulation of brain-derived neurotrophic factor synthesis by the cholinergic afferents. The synthesis of brain-derived neurotrophic factor (BDNF), a member of the neurotrophin protein family9,34 occurs predominantly in cortical and hippo- campal neurons.‘6+42.s* In contrast to the situation with nerve growth factor (NGF), where the neuro- trophic factor and its receptor are synthesized, respectively, in target and responsive tissues of the septohippocampal and basalo-cortical pathways (for reviews see Refs 32,48; see also Ref. 5 I), BDNF and tyrosine receptor kinase B (trkB) mRNA are expressed within the same cells in the hippocampal fo~~tion.2~.zl,46.47 These findings suggest a functiona role for BDNF different from that of a target-derived neurotrophic factor. In the adult brain, BDNF mRNA levels are elevated by seizures,‘2~‘7~54~55 application of excit- *To whom correspondence should be addressed at Cephalon Inc., 145 Brandywine Parkway, West Chester, PA 19380-4245, U.S.A. Abbreviations: ACh, acetylcholine; AChE, acetylcholin- esterase; BDNF, brain-derived neurotrophic factor; EDTA, ethylenediaminetetra-acetate; NGF, nerve growth factor; PBS, phosphate-buffered saline; rh, recombinant human; SSC, sodium chloride-sodjum citrate buffer; trk, tyrosine receptor kinase. atory amino agonists,‘5 mechanical damage,’ and ischemia,36 suggesting involvement of BDNF in mechanisms of neurodegeneration and/or neural protection. The induction of hippocampal BDNF expression following seizures has given rise to the speculation that BDNF mRNA synthesis is largely regulated by neuronal activity.“~‘5~s4~ss The in situ hybridization studies providing infor- mation on the sites of BDNF mRNA expression in the hippocampus show different levels of expression within distinct sub-regions of the hippo- ~~mpus~20.~l,54*~~ The study by Phillips et aL4’ emphasized the topographical correlation of BDNF mRNA expression with the distribution of cholin- ergic terminals originating from septal cholinergic afferents. More specifically, BDNF mRNA levels are highest in CA2 and CA3 pyramidal cell layers as we11 as in granule neurons of the dentate gyrus and hilar neurons.‘6.4’,so The same sub-regions receive the densest innervation from cholinergic cell bodies located within the medial septal nucleus and the diagonal band of Broca (for reviews, see Refs 18, 53; see also Refs 13, 38, 49). Thus, there exists a potential for interactions between septohippocampa~ cholinergic afferents and hippocampal BDNF 575

Upload: f

Post on 02-Jan-2017

212 views

Category:

Documents


0 download

TRANSCRIPT

Neuroscience Vol. 52, No. 3, pp. 515-585, 1993 Printed in Great Britain

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

J? 1993 IBRO

CHOLINERGIC REGULATION OF HIPPOCAMPAL BRAIN-DERIVED NEUROTROPHIC FACTOR mRNA

EXPRESSION: EVIDENCE FROM LESION AND CHRONTC CHOLINERGIC DRUG TREATMENT STUDIES

P. A. LAPCHAK,* D. M. ARAUJO and F. HEFTI

Division of Neurogerontology, Andrus Gerontology Center, University of Southern California. University Park, MC-0191, Los Angeles. CA 90089-0191. U.S.A.

A~~act-Quantitative in situ hybridi~tion and northern blot analysis techniques were used to determine the effects of removal of the cholinergic input on levels and topographical distribution of brain-derived neurotrophic factor mRNA in the hippocampus of adult rats. First, the effects of partial and full fimbriat transections, which result in partial and near-total cholinergic deafferentation respectively, were compared. Twenty-one days after partial unilateral fimbrial transections, there were significant decreases in brain-derived neurotrophic factor mRNA expression throughout the hippocampal formation. Decreased expression of brain-derived neurotrophic factor mRNA was evident in all areas of localization within the hippocampal formation. The decreases amounted to 22-36% reductions compared with unlesioned control animals. Brain~erived neurotrophic factor mRNA levels were decreased to a greater extent (S&69%) following full unilateral fimbrial transections. Quantitative northern blot analysis indicated that hippocampal BDNF mRNA was decreased by 29 and 68%. three weeks after partial or full unilateral fimbrial transections, respectively. The extent of the reductions in brain-derived neurotrophic factor mRNA levels correlated with reductions in acetylcholinesterase staining density and cholinergic terminal density determined by quantitative autoradiographic analysis of [‘Hlvesamicol binding sites. Second, we found that chronic treatment with atropine (20mg/kg per day for 14 days) decreased (by 54%) brain-derived neurotrophic factor mRNA levels in all areas of localization within the hippocampus. In contrast, chronic treatment with nicotine (1.18 mg/kg per day for 14 days), a treatment known to desensitize nicotinic receptors, did not affect brain-derived neurotrophic factor mRNA expression in the hippocampal formation.

The findings provide evidence for cholinergic muscarinic regulation of brain-derived neurotrophic factor mRNA expression in the adult rat hippocampal formation and they suggest the existence of a tonic stimulation of brain-derived neurotrophic factor synthesis by the cholinergic afferents.

The synthesis of brain-derived neurotrophic factor (BDNF), a member of the neurotrophin protein family9,34 occurs predominantly in cortical and hippo- campal neurons.‘6+42.s* In contrast to the situation with nerve growth factor (NGF), where the neuro- trophic factor and its receptor are synthesized, respectively, in target and responsive tissues of the septohippocampal and basalo-cortical pathways (for reviews see Refs 32,48; see also Ref. 5 I), BDNF and tyrosine receptor kinase B (trkB) mRNA are expressed within the same cells in the hippocampal fo~~tion.2~.zl,46.47 These findings suggest a functiona role for BDNF different from that of a target-derived neurotrophic factor.

In the adult brain, BDNF mRNA levels are elevated by seizures,‘2~‘7~54~55 application of excit-

*To whom correspondence should be addressed at Cephalon Inc., 145 Brandywine Parkway, West Chester, PA 19380-4245, U.S.A.

Abbreviations: ACh, acetylcholine; AChE, acetylcholin- esterase; BDNF, brain-derived neurotrophic factor; EDTA, ethylenediaminetetra-acetate; NGF, nerve growth factor; PBS, phosphate-buffered saline; rh, recombinant human; SSC, sodium chloride-sodjum citrate buffer; trk, tyrosine receptor kinase.

atory amino agonists,‘5 mechanical damage,’ and ischemia,36 suggesting involvement of BDNF in mechanisms of neurodegeneration and/or neural protection. The induction of hippocampal BDNF expression following seizures has given rise to the speculation that BDNF mRNA synthesis is largely regulated by neuronal activity.“~‘5~s4~ss

The in situ hybridization studies providing infor- mation on the sites of BDNF mRNA expression in the hippocampus show different levels of expression within distinct sub-regions of the hippo- ~~mpus~20.~l,54*~~ The study by Phillips et aL4’ emphasized the topographical correlation of BDNF mRNA expression with the distribution of cholin- ergic terminals originating from septal cholinergic afferents. More specifically, BDNF mRNA levels are highest in CA2 and CA3 pyramidal cell layers as we11 as in granule neurons of the dentate gyrus and hilar neurons.‘6.4’,so The same sub-regions receive the densest innervation from cholinergic cell bodies located within the medial septal nucleus and the diagonal band of Broca (for reviews, see Refs 18, 53; see also Refs 13, 38, 49). Thus, there exists a potential for interactions between septohippocampa~ cholinergic afferents and hippocampal BDNF

575

576 P. A. LAI’CHAR <‘I al.

mRNA-producing cells in the CAl, CA2, CA3, and dentate gyrus. In addition, the possibility exists that BDNF may directly affect cholinergic transmission in the hippocampal formation considering the specific localization of BDNF mRNA within the terminal fields of septal cholinergic projections.”

The present study tested whether hippocampal cholinergic neurons are involved in the regulation of BDNF mRNA expression and also whether BDNF directly affects measures of hippocampal cholinergic function. This was accomplished using three differ- ent experimental designs. First, we determined the effects of deafferentation of the hippocampal for- mation by producing either partial or full unilateral fimbrial transections. These knife-cut transections have previously been shown to produce partial or near total loss of cholinergic markers in the hippo- campal formation.‘s~2*~29~3’ Second, we determined whether specific continual antagonism of muscarinic or nicotinic receptors affects hippocampal BDNF mRNA expression. This was accomplished by chron- ically infusing by osmotic minipumps either the muscarinic receptor antagonist atropine or the nicotinic receptor agonist nicotine which results in desensitization-induced blockade of nicotinic receptors.26,27 Third, we determined whether BDNF acutely regulates presynaptic hippocampal cholin- ergic function. This was done by measuring the effects of recombinant human (rh)BDNF on [3H]- acetylcholine (ACh) synthesis in and release from hippocampal slices incubated in vitro.

EXPERIMENTAL PROCEDURES

Adult female Wistar rats (185-210 g) were purchased from Charles River Laboratories. Osmotic minipumps were purchased from Alza Corp. (CA). Atropine sulfate, nicotine bitartrate, physostigmine sulfate, choline chloride, glycyl- glycine, dithiothreitol, ATP disodium salt, choline kinase, acetylcholinesterase (AChE), 3-heptanone, and tetraphenyl- boron sodium salt were purchased from Sigma Chemicals (St Louis, MO). wethyl-3H]choline chloride (86.7 Ci/mmol) was purchased from New England Nuclear (Boston, MA). Molecular biology grade or ultrapure chemicals were purchased from Fisher Scientific, Promega, or GIBCO.

Fimbrial lesions

Fimbrial lesions were performed as described pre- viously.“.29,40 Briefly, rats were anesthetized with equitensin (pentobarbital) and placed in a stereotaxic apparatus. A small slit was cut into the skull on the left side and 2.0mm posterior to bregma. For full transections of the frmbria-fornix connection between septum and hippo- campus, a lancet-shaped knife (8 mm long, 0.7 mm wide, with a sharp tip) was lowered 5mm below dura into the brain at the midline and at 2.0 mm posterior to bregma (corresponding to level A6100 according to the atlas of Koenig and KlippelZ4). The knife was then moved laterally to 5.0 mm lateral to the midline and retrieved at this position. For partial transections of the fimbria-fomix, the knife was inserted 5 mm below the dura into the brain 1.5 mm lateral to the midline and at the same antero- posterior position. The knife was also moved laterally to 5.0 mm lateral to the midline and retrieved at this position. All animals were used 21 days after fimbrial transections.

Chronic cholinergic drug treatment

Cholinergic drugs were chronically administered via Abet osmotic minipumps (Model 2002; delivering 0.5 pi/h for 14 days). Osmotic minipumps were filled with either atropine sulfate to deliver 20 m&kg per day or nicotine bitartrate to deliver 1.18 mg/kg per day, a dose regimen described pre- viously.26,*7 In addition, a control group of animals received osmotic minipumps filled with saline.

Preparation of riboprobes for northern blot anal_ysi.~ and in situ hybridization

The labeled antisense and sense probes used for these studies were prepared from a rat BDNF cDNA construct which was subcloned into the Hind III/Eco Rl sites of pGEM-42.4’ The construct was used for riboprobe synthesis by linearization using either Hind III (sense) or Eco Rl (antisense), as described previously.” The Eco RI linear template was used with T7 RNA polymerase to generate the antisense probe, whereas the Hind III linear template was used with SP6 RNA polymerase to generate the sense probe. [32P]UTP or [“S]UTP were used for riboprobe synthesis for northern blot or in situ hybridization analysis, respectively.

In situ hybridization histochemistrJ

In situ hybridization was carried out as described pre- viously.‘2,30 For in situ hybridization histochemistry, the brains were removed and snap frozen by immersion in 2-methylbutane at -40°C. Brains were subsequently stored at -70°C. Coronal sections were cut at 10pm on a Hacker Instruments cryostat. Sections were thaw-mounted onto gelatin-coated slides and then stored at -70°C. Prior to incubation with labeled riboprobes, thaw-mounted brain sections were postfixed for 30 min with 4% paraformalde- hyde in phosphate-buffered saline (PBS; pH 7.0). Sections were then washed in PBS (PH 7.4; 3 x 10min) followed by a wash in de-ionized water (1 min). Sections were then treated for 1 min with 0.1 M triethanolamine (PH 8.0) fol- lowed by a IO-mm incubation in acetic anhydride/O.l M triethanolamine in order to decrease nonspecific binding. Following a I-min wash in 2 x SSC (1 x SSC is 0.15 M NaCl, 15 mM sodium citrate, pH 7.0) sections were dehy- drated in a series of increasing concentrations of alcohols (30, 50, 70, 85, 95 and 100%). Sections were hybridized for 3 h at 50°C with [‘?$]BDNF cRNA (2 x [email protected]./pg, 0.2-0.4 na cRNAIIII) in the followina hvbridization buffer: 50% for&amide,“4 k SSC, 5 x Denhardt’s solution, 1% sodium dodecyl sulfate, 250 pg/ml transfer RNA, 25 fig/ml poly adenylic acid (5’) 25 pg/ml poly cytidylic acid (53, and 0.1 mM dithiothreitol. Sections were then rinsed in 4 x SSC/20 mM DTT, then 4 x SSC alone. RNAase digestion was carried out for 30 min at 37°C (20 pg/ml RNAse in 0.5 M NaCI, 0.01 M Tris-HCI, 0.001 M EDTA, pH 8.0). Brain sections were then washed at room temperature for 18.-24 h with 2 x SSC containing 0.02 M B-mercantoethanol followed by a l-h high criterion wash with O.l^x SSC at 63°C. The sections were then dehydrated in a series of ethanols containing 0.3 M ammonium acetate to protect against denaturation of the hybrids. They were then air- dried and juxtaposed against Hyperfilm MP (Amersham Inc.). Specific labeling was not observed in adjacent sections incubated with the 35S-labeled sense strand of the BDNF probe or in tissue sections that were pretreated with RNAase (20 fig/ml).

Northern blot analysis

For northern blot analysis, total RNA was extracted using a modification of the method of Chomczynski and Sacchi.” Total RNA was senarated on 1% aaarosei formaldehyde gel and then transferred to Nylon 66 Plus membrane (Hoefer Scientific Instruments, San Francisco, CA) with 4 x SSC (PH 7.0). RNA was cross-linked to the membranes by UV light irradiation using a Stratagene crosslinker. Prior to hybridization with [3ZP]cRNA. the

Cholinergic regulation of BDNF mRNA 571

membrane was prewashed for 30min with 0.1 x SSC, 0.15 M fi-mercaptoethanol, and 1% SDS, and then pre- hybridized for 2 h at 68°C in buffer containing 6 x SSC, 1% SDS, 10% dextran sulfate, lOOpg/ml transfer RNA, 50 pg/ml polyadenylic acid and 100 pgg/ml polycytidylic acid, and 0.5% Carnation powdered milk. Following the prehy- bridization, the buffer was changed and the membranes were hybridized (68”C, overnight) in 10 ml of the same buffer in the presence of [32P]cRNA (5 x lo6 d.p.m./ml of hybridiz- ation medium, - I x lO”d.p.m./ng cRNA). Membranes were then washed once with 5x SSCjlX SDS at 68°C followed by five washes with 2 x SSC/O.l% SDS. The blot was then juxtaposed against X-OMAT autoradiographic film (Kodak) and exposed for four to eight days.

Quuntitative ana!ysi.s of u~turadiogru~hic J%TZS from in situ bybr;djzatio~ and northern blot analysis

Levels of specific hybridization were quantified using a Drexel University Imager analysis program. Quantitation of optical densities was based on calibrated gray scale densities. Densities were quantitated over the entire area of hippocam- pal region and were limited to the CAI, CA2, CA3, and dentate gyrus. For northern blots, the densities were quan- titated over the entire band(s) representing the two BDNF mRNA species.

Adult rats were deeply anesthetized with equitensin and perfused transaortically as described previously.29 Briefly. animals were perfused with a mixture of 4% paraformalde- hyde and 0.2% picric acid in 0.1 M phosphate buffer (pH 7.4). The brains were removed form the skull, postfixed overnight in the same solution, and immersed overnight in a 30% sucrose phosphate-buffered solution. They were snap-frozen by immersion in 2-methylbutane at -40°C and 30-pm sections were cut on a cryostat. Floating sections were preincubated for 30min at room temperature with 30pM tetraisopropyl pyrophosphoramide to inhibit nonspecific cholinesterases. They were then incubated for 15 min at room temperature in a solution of 0.5 mg/ml acetylthiocholine iodide. 3 mM CuSO,, 0.5 mM K,Fe (CN),, 5 mM sodium citrate in 0.1 M Tris-maleate buffer (pH 5.7) and diluted 1:lO with 0.1 M sodium phosphate buffer (pH 7.2) immediately before use. After incubation, the sections were washed and the reaction product visualized by incubating them in a solution containing 4% diaminobenzidine, 0.3% nickel am- monium sulfate, and 0.003% Hz02.

[‘H] Vmumicol autoradiogruphq

Coronal brain sections were prepared as described pre- viously.” Sections were pre-incubated for 30 min at room temperature in 50mM Tris buffer (pH 7.4) that contained 120mM NaCI, 5mM KCI, 2mM CaCI,, and I mM MgC1,. The sections were transferred to and incubated at room temperature in the same buffer containing 20nM f3H]vesamicol (48.0 Ci/mmol; NEN Research Products). Following the incubation, brain sections were rinsed four times for 30 s in Tris buffer at 4”C, followed by a rinse in distilled water (4°C). Sections were then dried by a stream of cool air and apposed to tritium-sensitive film (Hyperfiim 13H]; Amersham Inc.) for five to seven weeks. Specifically bound ligand was determined as the amount bound in the absence as compared to that bound in the presence of 10 PM unlabeled vesamicol; this represented 70-76% of the total binding under the conditions used in this study.

In vitro~/~armac~/ogy : [‘H]acet,ylcholine symhesis and release

Brain slices were prepared as previously described.*Gz9 In experiments testing the effects of rhBDNF on presynaptic cholinergic functions, rhBDNF was added to the tissue slices for a 15-min incubation period before measuring either [‘H]ACh release (basal or evoked) orrH]ACh synthesis.

Newly synthesized i3H]ACh contents and release were measured as described previously.26.27 The lot of rhBDNF used in the present study has previously been characterized and shown to be pharmacologically active.“.‘9 rhBDNF was generously provided by Dr L. E. Burton (Genentech Inc.. South San Francisco. CA).

Statistical analyses

Data-points given correspond to values from hippocampi of individual animals. Results are typically expressed as means + S.E.M. of the number of experiments or animals, as indicated. Statistical significance was assessed using ANOVA followed by linear analysis of contrasts according to Scheffe. The statistical analysis program NCSS (Number Cruncher Statistical System) was used for these tests.

RESULTS

Effect of unilateral jimhrial transections on hippo-

campal brain -derived neurotrophic ,factor m RNA

e~pressiot~

Partial or full unilateral transections of the fimbria

were made using a precisely defined knife-cut. and the distribution of BDNF mRNA was quantitated 21 days after the lesion, at a time-point when the degeneration of lesioned septohippocampal neurons is complete.‘5,29,30 These lesions result in defined reductions of hippocampal cholinergic markers such as choline acetyltransferase activity, high-a~nity choline uptake, and ACh synthesis and release.“x.“,”

In unlesioned control animals, BDNF mRNA was found to be present in the pyramidal cell layers of CAlGCA3 and in the dentate gyrus (Fig. 1A) in both the dorsal and ventral hippocampus. These findings are consistent with those previously reported,‘2,‘h,“,” The effects of partial (B, E) or full (C, F) fimbrial transections on the expression of BDNF mRNA in the hippocampus are shown in Fig. 1. Both types of fimbrial lesions produced significant decreases of BDNF mRNA within the hippocampal formation. specifically within the CA3 pyramidal cell layer and within the dentate gyrus (Fig. 1).

Quantitative analysis of the hybridization signals in the hippocampus following lesions is shown in Table 1 (AJ). The extent of the decreases were dependent on the type of lesion performed. Partial fimbrial transections produced decreases which were in the range of 25-36%, whereas the complete fimbrial transection produced larger decreases in the range of 3840%.

Figure 2 is a representative northern blot of hippo- campal mRNA which shows the typical two species of BDNF mRNA (2.4 and 4.4 kb). Quantitative analysis of the northern blot indicates that partial and full fimbrial transections reduced BDNF mRNA levels by 29 and 68%, respectively (Table 1 A,II).

Correlation qf changes in hippoc~tpa~ hrain -derived neurotroph~c &actor rnRM.4 e.pression with distri- bution qf cholinergic markers: acet?;lcholinestera.w and [‘Hlvesamicol binding sites

Figure 3 shows the distribution pattern for BDNF mRNA (A), AChE staining (D), and [3H]vesamicol

578 P. A. LAPCHAK et al.

8- VIAL Fx

C-FULL Fx

Fig. 1. BDNF mRNA expression in the adult rat brain following partial or full unilateral fimbrial transections. Representative photomicrographs of coronal brain motions through the dorsal (A-C) or ventral (D-F) hippocampus are shown. (A, D) Control levels of [35S]BDNF cRNA hybridixation signal. (B, E) 21 days following a partial fhnbrial transection (partial Fx) shown on the left hand side of the panel. (C, F) Twenty-one days following a full unilateral transection (full Fx) shown on the left-hand side of the panel. Both partial and full 6mbrial transections decrease the hybridization signal in the hippocampal

formation. DG, dentate gyrus.

Cholinergic regulation of BDNF mRNA 519

Table 1. Hippocampal BDNF mRNA levels following fimbrial transections or chronic cholinergic drug treatments

[35S]BDNF cRNA hybridization (per cent control)

A) Fimbrial transections Hippocampal Partial Full region transections transections I) In situ:

1) CA1 75.1 + 8.8%* 40.6 & 16.9%** 2) CA2 70.6 + 5.1%* 62.1 + 4.8%* 3) CA3 78.7 : 11.2% 46.1 ? 5.7%*+ 4) Dentate gyms 64.1 + 8.9%* 50.2 1.8% - z

II) Northern blot: 71.7 + 4.2%* 32.7 + 13.9%** - -

B) Chronic cholinergic drug treatment 1) Atropine 46.3 & 1.4+’ 2) Nicotine 103.4 & 4.8

Measurements are expressed as per cent control and rep- resent a ratio of the optical densities measured on the lesioned side divided by that measured on the unlesioned side x 100. Values given are mean +S.E.M. for five to eight animals per group. Densities were measured over the neuronal fields of the entire area in the dorsal hippocampus for each subregion indicated for animals receiving fimbrial transections. Five to eight coronal brain sections were used for the measurement of den- sities from each group of animals. Significantly different from the contralateral unlesioned side (*P < 0.05, **P < 0.01). For northern blot analysis optical densities were measured for both BDNF mRNA species. For animals chronically treated with atropine or nicotine densities were measured over the dorsal CAl, CA2, and CA3 pyramidal cell layer and over the granule cell layer of the dentate gyrus. Results given are the overall mean & S.E.M. of values from drug-treated animals divided by that for saline-treated x 100. Significantly different from saline-treated control animals (**P < 0.01). Chronic nicotine treatment did not significantly (P > 0.05) alter hippocampal BDNF mRNA levels. Statistical signifi- cance was determined using ANOVA followed by Scheffes assessment of linear contrasts.

binding sites (G), respectively, in the unlesioned control rat hippocampal formation. These panels also show a high density of BDNF mRNA localized to the CA2 and CA3 pyramdial cell layers and in the granule cell layer of the dentate gyrus. The distribution of the presynaptic cholinergic vesicle marker,‘9,44 [3H]vesamicol binding sites, correlates well with the distribution of AChE and BDNF mRNA.

Following partial (Fig. 3 B,E,H) or full (Fig. 3 C,F,I) fimbrial transections there are significant decreases of both cholinergic markers that closely parallel the reduction in BDNF mRNA levels. Quan- titative analysis of [3H]vesamicol binding sites in the hippocampal formation after fimbrial transections indicated that partial and full fimbrial transections significantly decreased the density of [3H]vesamicol binding sites by 24 f 6% (P < 0.05) and 40 f 10% (P < 0.05), respectively.

Eflect of chronic atropine or nicotine administration to adult rats on hippocampal brain -derived neurotrophic factor mRNA expression

The results obtained from the lesion studies sug- gested that the septohippocampal cholinergic sys

Fig. 2. Effect of partial or full fimbrial transections on BDNF mRNA levels quantitated by northern blot analysis. Five micrograms of total RNA extracted from three to four pooled hippocampi were loaded per lane. Partial and full unilateral fimbrial transections reduce the hybridization

signal of both species of BDNF mRNA.

tern was required for the maintenance of normal BDNF mRNA levels within the hippocampal forrna- tion. Thus, this would require the interaction of endo- genous ACh with a postsynaptic receptor population in order to relay the appropriate signal.‘4.‘9 This type of regulation could occur through two different types of cholinergic receptors, either muscarinic or nic- otinic receptors, both of which have been shown to be extensively distributed throughout the hippo- campus.3m6,43,45 Therefore, we determined the effects of muscarinic receptor blockade or nicotinic receptor desensitization on the expression of BDNF mRNA in the hippocampal formation. The muscarinic antagon- ist atropine was chronically administered in order to block muscarinic receptors, whereas chronic nicotine was administered in order to produce desensitization- induced blockade of nicotinic receptors.

The results of these experiments are shown in Fig. 4. The distribution of BDNF mRNA from animals chronically implanted with miniosmotic pumps containing saline is shown in Fig. 4 A, D. The distribution of BDNF mRNA in saline-treated ani- mals is not significantly different from that of con- trol unlesioned animals (Fig. 4). Chronic atropine administration decreases BDNF mRNA levels throughout the hippocampal formation, including CAlCA3, and the dentate gyrus (B, E). The changes represented a 34% decrease. Fig. 4 C and F show that chronic nicotine administration does not affect BDNF mRNA levels. Quantitative data for these experiments are given in Table 1 II.

L

w

Cholinergic regulation of BDNF mRNA

A-SALINE

B-ATROPINE

DG

C-NICOTINE

CA1

DG

E

F

CA3

DG

Fig. 4. Effect of chronic 14-day cholinergic drug treatment on the expression of BDNF mRNA in the hippocampal formation. Representative photomicrographs of corona1 brain sections through the dorsal (AC) or ventral (D-F) hippocampus are shown. (A, D) Control levels of [j5S]BDNF cRNA hybridization signal. (B, E) 14 days following chronic atropine treatment. (C, F) Fourteen days following chronic nicotine treatment. Chronic atropine significantly decreases the hybridization signal in the hippocampus, whereas chronic nicotine does not significantly (P > 0.05) affect BDNF mRNA levels at this time-point.

582 P. A. LAPCHAK et al.

Effect of brain-derived neurotrophic factor on in vitro measures of presynaptic choline@ .fimction

The results described above suggest that, in the hippocampal formation, cholinergic tone is required to maintain steady-state levels of BDNF mRNA. However, it is not known if the converse regu- lation exists, i.e. the regulation of cholinergic func- tion by BDNF. The following set of experiments determined whether acute treatment with BDNF affects presynaptic cholinergic function measured in hippocampal slices in vitro. Spontaneous and evoked release of [3H]ACh release from hippocampal slices and rH]ACh synthesis by hippocampal slices were measured as described previously.26,27

First, we tested the effects of BDNF (2 x lOmy to 2 x 1O-7 M) on the ability of hippocampal slices to synthesize [‘H]ACh from the precursor molecule [‘HIcholine under basal condition (i.e. in normal potassium Krebs medium). BDNF did not signifi- cantly affect this measure of presynaptic cholinergic function (range 93.6 +- 9.2% to 107.9 +_ 10.9% of control values). Second, we determined the effects of BDNF (2 x lo-” to 2 x 10e6 M) on spon- taneous and potassium-evoked (25 mM K’) release of [‘H]ACh from hippocampal slices incubated in vitro. BDNF did not significantly (P > 0.05) affect either spontaneous (range 84.1 f 10.2% to 100 f 10.0% of control values) or evoked (range 86.8 f 5.8% to 124.2 f 13.4% of control values) t3H] ACh release from hippocampal slices (n = 5-7 animals per concentration tested).

DISCUSSION

The present study aimed at determining the role of the hippocampal cholinergic afferents in the regu- lation of hippocampal BDNF expression. Partial and full deafferentation of the hippocampal formation resulted in decreases of BDNF mRNA levels in all sub-regions of the hippocampus, to similar extents as with the choline@ deficit. The cholinergic deficit was demonstrated by a reduction in the density of AChEpositive fibers and presynaptic ACh trans- porter binding sites as determined using [‘HI- vesamicol autoradiography. Furthermore, chronic atropine treatment for 14 days resulted in significant decreases of BDNF mRNA levels in the same hippo- campal subregions. The data suggest that the cholin- ergic system, by acting on muscarinic receptors, may exert a tonic effect on BDNF mRNA synthesizing cells in the hippocampus. In contrast to the cholin- ergic regulation of BDNF expression, our findings failed to provide evidence for the reciprocal regu- lation by BDNF of choline@ neurotransmission in the rat hippocampal formation.

Specific lesions which were previously shown to produce either partial or full unilateral deafferenta- tion of the hippocampal formation were used in the present study (for review, see Ref. 33; also see

Refs 15, 28. 29, 31, 38, 52). These lesions decrease a variety of cholinergic parameters at the level of septal cell bodies and decrease the number of cholinergic terminals within the hippocampal for- mation. Thus, in essence, fimbrial transections destroy a large component of the cholinergic afferen t population. As shown in the present study. 21 days following full fimbrial transections there is a con- siderable loss of AChE staining and [3H]vesamicol binding density throughout the hippocampus. whereas partial fimbrial transections result in a smaller loss of these markers. The decrease of [3H]vesamicol binding sites observed in the present study following hippocampal deafferentation is consistent with a previous study’” which showed that transection of the septohippocampal pathway decreased the density of [3H]vesamicol sites by 30%. Therefore, [3H]vesamicol seems to be a good marker for the degeneration of the septohippocampal pathway in the rat. The reduction in hippocampal cholinergic function following the lesions was correlated with a decrease in the expression of the BDNF mRNA levels. This effect was widespread throughout the hippocampal formation and affected all cellular populations expressing BDNF mRNA. This suggested that the cholinergic component of septal afferents which innervate the hippocampal formation was involved in maintaining the synthesis of BDNF mRNA and that ACh released from cholinergic terminals provides a prominent tonic cholinergic tone which maintains BDNF mRNA expression.

Recently, we have shown that following partial fimbrial transections remaining hippocampal termi- nals have the ability to compensate for the loss of neighboring cholinergic terminals. The compen- sation phenomenon is manifest by an increase in the function of remaining cholinergic terminals when measured in vivo following partial fimbrial transec- tions.29 The present results, showing that partial fimbrial transections reduce BDNF mRNA levels, suggest that these compensatory increases in pre- synaptic choline+ function do not translate into compensation at the postsynaptic level. Similarly, we have found that after partial fimbrial transections, the response of hippocampal Ml muscarinic recep- tors is supersensitive as evidenced by an enhanced production of inositol triphosphate33” following stimulation with agonist (oxotre morine). Thus, these specific changes at the level of cholinergic synapses may account for decreased levels of BDNF mRNA in the hippocampus after partial fimbrial transections.

The hypothesis that cholinergic tone is necessary to sustain hippocampal BDNF mRNA expression was corroborated by pharmacological experiments show- ing that chronic blockade of muscarinic receptors results in decreased BDNF mRNA levels in the hippocampal formation. Moreover, the atropine- induced decrease was greater than that seen in rats with partial fimbrial transections and was observed in

Cholinergic regulation of BDNF mRNA 583

the same regions. The effects of chronic a&opine

treatment on the decrease of BDNF mRNA levels are consistent with the observed effects of fimbrial

transections and suggests that an intact functional cholinergic synapse with ACh interacting with muscarinic receptors is important in maintaining BDNF mRNA synthesis. A similar hypothesis was recently put forward by Lindefors et al.”

They showed that scopolamine pretreatment could partially antagonize the upregulation of hippocampal BDNF mRNA levels, following the activation of the septohippocampal neurons by intraseptal injections of quisqualic acid. In contrast to the muscarinic regulation of BDNF mRNA expression, chronic nicotine administration did not alter the levels of BDNF mRNA expression, even though the chronic nicotine treatment regimen used in the present study produces desensitization-induced blockade of nicotinic receptor populations.26.27.37 It is interesting to note that even though chronic atropine

treatment decreases BDNF mRNA expression, the integrity of the cholinergic component of the septohippocampal pathway apparently remains intact. A previous study2’ showed that a similar treatment regimen with atropine does not signifi- cantly alter hippocampal ChAT activity as com- pared to saline-treated animals. This suggests that endogenous hippocampal-derived BDNF may play a limited role in the maintenance of septohippocampal cholinergic neurons.

Indeed, at present the functional significance of BDNF in the adult septohippocampal pathway is unclear. BDNF mRNA expression by most hippo- campal neurons, together with its trkB receptor,20,” suggests a role different from that of a classical “target-derived neurotrophic factor”. Nevertheless, the findings of the present study are consistent with the view that, under normal steady-state con- ditions, ACh maintains the synthesis of BDNF mRNA possibly to serve as “growth or maintenance factor” for neuronal populations of the septo- hippocampal pathway. The regulation of septal ChAT activity by BDNF in cholinergic neurons in culture,‘.22 and the partial protection of lesioned cholinergic neurons from degeneration by intra- ventricular administration of BDNF,23 supports this notion. In addition, the protective effect of BDNF treatment on adult septal cholinergic cell bodies was only partial and less significant than that produced by NGF.23 Furthermore, BDNF treatment did not prevent lesion-induced decreases of presynaptic cholinergic function in the hippocampus, whereas NGF clearly does.3’ It seems possible that BDNF is involved in the maintenance of only a select population of cholin- ergic neurons of the septohippocampal pathway, which may express the trkB receptor.46.47 The findings, indicating a very limited role in the function of septohippocampal cholinergic neurons, do not preclude a more important role for BDNF as a

growth factor for other populations of hippocampal

neurons. The finding that BDNF mRNA expression can

be regulated by lesions and pharmacological man- ipulations contrasts with the lack of effect of similar manipulations on NGF mRNA levels in the adult brain. Fimbrial transections do not lead to persistent changes of NGF mRNA levels in the hippocampus of adult rats but result in a transitory elevation of NGF protein levels.25,29 It is not clear whether BDNF protein levels are also affected by the lesions or by pharmacological treatment. The differential regulation of mRNA levels of the two neurotrophins suggests different roles in the maintenance of cholinergic afferents, non-cholinergic afferents, and of intrinsic hippo- campal neurons.

Our results invite speculations about the role of

BDNF in neurodegenerative diseases which affect central cholinergic neurons in the hippocampal formation such as Alzheimer’s disease. There is decreased choline acetyltransferase activity, choline uptake and ACh synthesis in the hippocampus in the Alzheimer’s disease brain (for a review see Ref. 11, also see Refs 2, 7). Therefore, these decreases in cholinergic function may diminish the ability of hippocampal neurons to synthesize BDNF mRNA and the protein product BDNF. Decreases in BDNF synthesis may then affect other BDNF responsive neurons and so may lead to deregulation of multiple homeostatic processes in the hippo- campus, which normally maintain neuronal cell

viability and function. Supporting such a concept is the finding that BDNF mRNA levels are sub- stantially decreased in the hippocampus within the granule cell layer of the dentate gyrus of Alzheimer’s disease brains4’

CONCLUSIONS

The findings of the present study show that deafferentation of the hippocampal formation, which produces decreased cholinergic innervation of the hippocampal formation, decreases BDNF mRNA expression in the pyramidal cell layer and granule cell layer of the dentate gyrus. Moreover, we provide evidence for muscarinic, but not nicotinic, receptor-mediated maintenance of hippocampal BDNF mRNA levels. Lastly, we show that BDNF is not involved in the regulation of presynaptic hippocampal cholinergic function as evidenced by the absence of an effect of BDNF on ACh synthesis or release in oitro.

Acknowledgemenfs-This study was supported by grants NS22933, AG10480, the State of California Alzheimer Research Program Grant 91-12965. and the National Parkinson Foundation. PAL was supported by a long- term postdoctoral fellowship from HFSPO (Japan) and a grant/fellowship from the French Foundation for Alzheimer Research (Los Angeles, CA).

584 P. A. LAPWAK et d.

REFERENCES

I. Alderson R. F., Alterman A. L., Barde Y.-A. and Lindsay R. M. (1990) Brain-derived neurotrophic factor Increases survival and differentiated functions of rat septal cholinergic neurons in culture. Neuron 5, 297-306.

la. Altar C. A. and Mar&n M. C. (1988) ‘H-AH5183 binding in brain: autoradiographic distribution, pharmacology. and

2.

3.

4.

5.

6.

7.

8.

9.

10.

11. 12.

13.

14.

15.

16.

17.

18. 19.

20.

21.

22.

23.

24.

25.

26.

21.

28.

29.

effect of cholinergic lesion. Synapse 2, 486493. - __

Araujo D. M., Lapchak P. A., Robitaille Y., Gauthier S. and Quirion R. (1988) Differential alteration of various choline@ markers in cortical and subcortical regions of human brain in Alzheimer’s disease. J. Nrurochem. SO, 19141923. Araujo D. M., Lapchak P. A., Collier B. and Quirion R. (1989) N-[3H]methylcarbamylcholine binding sites in the rat and human brain: relationship to functional nicotinic autoreceptors and alterations in Alzheimer’s disease. frog. Bruin Res. 79, 345-352. Araujo D. M., Lapchak P. A., Regenold W. and Quirion R. (1989) Characterization of [3H]AF-DX 116 binding sites in the rat brain: evidence for heterogeneity of muscarinic-M2 receptor sites. Synapse 4, 106-l 14. Araujo D. M., Lapchak P. A., Meaney M. J., Collier B. and Quirion R. (1990) Effects of aging on nicotinic and muscarinic autoreceptor function in the rat brain: relationship to presynaptic cholinergic markers and binding sites. J. Neurosci. 10, 3069-3078. Araujo D. M., Lapchak P. A. and Quirion R. (1991) Heterogenous binding of [3H]4-DAMP to muscarinic cholinergic sites in the rat brain: evidence from membrane binding and autoradiographic studies. Synapse 9, 165-176. Ball M. J. (1983) Hippocampal histopathology-a critical substrate for dementia of the Alzheimer type. Interdisciplinar.~ Topics Gerontol. 25, 1637. Ballarin M., Emfors P., Lindefors N. and Persson H. (1991) Hippocampal damage and kainic acid injection induce a rapid increase in mRNA for BDNF and NGF in the rat brain. Expl Neural. 114, 3543. Barde Y. A., Edgar D. and Thoenen R. H. (1982) Purification of a new neurotrophic factor from mammalian brain. Eur. molec. Biol. Org. J. 1, 549-553. Chomczynski P. and Sacchi N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol- chloroform extraction. An&t. Biochem. 162, 156-159. Collerton D. (1986) Cholinergic function and intellectual decline in Alzheimer’s disease. Neuroscience 19, l-28. Dugich-Djordjevic 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. Frotscher M. and Leranth C. (1985) Cholinergic innervation pf the rat hippocampus as revealed by choline acetyltransferase immunocytochemistry: a combined light and electron microscopic study. J. camp. Neural. 239, 237-246. Fukuda K., Kubo T., Maeda A., Akiba I., Bujo H., Nakai J., Mishina M., Higashida H., Neher E., Marty A. and Numa S. (1989) Selective effector coupling of muscarinic acetylcholine receptor subtypes. Trends phnrmac. Sci. Suppl. k-10. Hefti F., Dravid A. and Hartikka J. (1984) Chronic intraventricular injections of nerve growth factor elevate hippocampal choline acetyltransferase activity in adult rats with partial septo-hippocampal lesions. Brain Rex 293, 305-3 11. Hofer M., Pagliusi 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. Isackson P. J., Huntsman M. M., Murray K. D. and Gall C. M. (1991) BDNF mRNA expression is increased in adult rat forebrain after limbic seizures: temporal patterns of induction distinct from NGF. Neuron 6, 937-948. Kasa P. (1986) The cholinergic system in brain and spinal cord. Prog. Neurobiol. 26, 21 l-272. Kilboum M. R., Jung Y.-W., Haka M. S., Gildersleeve D. L., Kuhl D. E. and Wieland D. M. (1990) Mouse brain distribution of a carbon-11 labeled vesamicol derivative: presynaptic marker of choline& function. t@ Sci. 47, 1955-1963. Klein R., Martin-Zanca D., Barbacid M. and Parada L. F. (1990) Expression of the tyrosine receptor gene trkB is confined to the murine embryonic and adult nervous system. Development 109, 845-850. Klein R., Nanduri V., Jing S., Lamballe F., Tapley P., Bryant S., Cordon-Cardo C., Jones K. R., Reichardt L. F. and Barbacid M. (1991) The trkB tyrosine protein kinase is a receptor for brain-derived neurotrophic factor and neurotrophin-3. CeN 66, 395-403. Knusel B., Winslow J. W., Rosenthal A., Burton L. E., Seid D. P., Nikolics K. and Hefti F. (1991) Promotion of central cholinergic and dopaminergic neurons differentiation by brain-derived neurotrophic factor but not neurotrophin 3. Proc. natn. Acad. Sci. U.S.A. 88, 961-965. Knusel B., Beck K. D., Winslow J. W., Rosenthal A., Burton L. E., Widmer H. R., Nikolics K. and Hefti F. (1993) Brain-derived neurotrophic factor administration protects basal forebrain cholinergic but not nigral dopaminergic neurons from degenerative changes after axotomy in the adult rat brain. J. Neurosci. (in press). Konig J. F. R. and Klippel R. A. (1963) The Rat Brain. A Stereotaxie Atlas of the Forebrain and Lower Paris of the Brainstem, pp. 22-156. Williams and Wilkins, Baltimore, MD. Korsching S., Heumann R., Thoenen H. and Hefti F. (1986) Cholinergic denervation of the rat hippocampus by fimbrial transections leads to a transient accumulation of nerve growth factor (NGF) without change in mRNA (NGF) content. Neurosci. Left. 66, 175-180. Lapchak P. A., Araujo D. M., Quirion R. and Collier B. (1989) Effect of chronic nicotine treatment on nicotinic autoreceptor function and [3H]methylcarbamylcholine binding sites in rat brain. J. Neurochem. 52, 483-491. Lapchak P. A. and Beaudet B. (1990) Cholinergic regulation of vasoactive intestinal peptide content and release in rat frontal cortex and hippocampus. J. Neurochem. 55, 134C-1345. Lapchak P. A. and Hefti F. (1991) Effect of recombinant human nerve growth factor on presynaptic cholinergic func- tion in rat hippocampal slices following partial septohippocampal lesions: measures of [‘Hlacetylcholine synthesis, [3H&cetylcholine release and choline acetyltransferase activity. Neuroscience 42, 639-649. Lapchak P. A., Jenden D. J. and Hefti F. (1991) Compensatory elevation of acetylcholine synthesis in vivo by cholinergic neurons surviving partial lesions of the septohippocampal pathway. J. Neurosci. 11, 2821 -2828.

Cholinergic regulation of BDNF mRNA 585

30. Lapchak P. A. (1992) A role for interleukin-2 in the regulation of striatal dopaminergic function. .NeuroReport 3, 165-168.

31. Lapchak P. A. and Hefti F. (1992) BDNF and NGF treatment in lesioned rats: effects on cholinergic function and weight gain. NeuraReport 3, 140&143.

32. Lapchak P. A., Araujo D. M. and Hefti F. (1992) Neurotrophins in the central nervous system. f&z*. Neurosci. 3, 1 1 I. 33. Lapchak P. A. (1992) Therapeutic potential for nerve growth factor in Alzheimer’s disease: insights from pharmaco-

logical studies using lesioned central cholinergic neurons, Rev. Neurosci. 3, 109-l 18. 33a. Lapchak P. A., Araujo D. M. and Hefti P. (1993) Regulation of hippocampal muscarinic receptor function by chronic

34.

35.

36.

37.

3x.

39.

40.

41.

42.

43.

44.

4s.

46.

41.

48.

49.

50.

51.

52.

53. 54.

55.

nerve growth factor treatment in adult rats with timbrial transactions. Neu~o.~cje~~e (in press). Leibrock J., 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. Lindefors N.. Ernfors P., Falkenberg T. and Persson H. (1992) Septal cholinergic afferents regulate expression of brain-derived neurotrophic factor and /?-nerve growth factor mRNA in rat hipp~mpus. Expt Bruin Rer. 88, 78 90. Lindvall O., Ernfors P.. Bengzon J.. Kokaia Z., Smith M.-L., Siesio B. K. and Persson H. (1992) Differential regulation of mRNAs for nerve growth factor, brain-derived neurotrophic factor, and neurotrophin-3 in the adult rat brain following cerebral ischemia and hypoglycemia coma. Proc. nntn. Acxzd. Sci. U.S.A. 89, 648-652. Martino-Barrows A. M. and Kellar K. J. (1987) [~H]acetylcholine and [‘HI ( - ) nicotine label the same rc~~~~nition site in rat brain. Moiec. Pharmac. 31, 169-174. Matthews D. A., Salvaterra P. M., Crawford G. D., Houser C. R. and Vaughn J. E. (1987) An immunocytochemtcal study of chdine acetyltransferase containing neurons and axon terminals in normal and partially deafferented hipp[~~ampal formation. Brairr Rex 402, 30-43. McKinney M. and Richelson E. (1984) The coupling of the neuronal muscarinic receptor to responses. A. Ret,. Plrtrmrrrc. Toxicol. 24, 121-,146. Montero C. N. and Hefti F. (1988) Rescue of lesioned septal cholinergic neurons by nerve growth factor: specificity and requirement for chronic treatment. J. Neurosci. 8, 29862999. Philhps H. S., Hains J. M., Laramee G. R., Rosenthal A. and Winslow J. W. (1990) Widespread expression of BDNF but not NT3 by target areas of basal forebrain cholinergic neurons. Science 250, 290-294. Phillips H. S., Hains J. M., Armanini M., Laramee G. R., Johnson S. A. and Winslow J. W. (1991) BDNF mRNA is decreased in the hippocampus of individuals with Alzheimer’s disease. Neuron 7, 6955702. Quirion R. (I 985) Comparative localization of putative pre- and postsynaptic markers of muscarinic cholinergic nerve terminals in brain. Eur. J. Pharmac. 111, 2877289. Quirion R. (1987) Characterization and autoradiographic distribution of hemicholininm-3 high affinity choline uptake sites in mammalian brain. Synapse 1, 2933303. Regenold W.. Araujo D. M. and Quirion R. (1989) Quantitative autoradiographic distribution of [‘H]AF-DX I16 muscarinic-M2 receptor binding sites in rat brain. S,vnap.sc 4, 115-125. Soppet D., Escadon E., Maragos J., Middlemas D. S.. Reid S. W., Blair J., Burton L. E., Stanton 8. R.. Kaplan D. R.. Hunter T.. Nikolics K. and Parada L. F. (1991) The neurotrophic factors BDNF and NT-3 are iigands for the trkB tyrosine kinase receptor. Cell 65, 8955903. Squint0 S. P., Stitt T. N., Aldrich T. H., Davis S., Bianco S. M., Radziejewski C., Glass I>. J., Masiakowski P.. Furth M. E., Valenzuela D. M., DiStefano P. S. and Yancopolous G. D. (1991) trkB encodes a functional receptor For

brain-derived neurotrophic factor and neurotrophin-3 but not nerve growth factor. Cell 65, 885-893. Thoenen H., Bandtlow C. and Heumann R. (1987) The physiological significance of nerve growth factor in the central nervous system: comparison with the periphery. Ret]. Physiol. Biochem. Pharmac. 109, 1455178. Wainer B. H.. Levey A. L., Rye D. B.. Me&am M. M. and Mufson E. J. (1985) Cholinergic and non-cholinergic speto-hippocampal pathways. Neurosci. Left. 54, 45 52. Wetmore C., Ernfors P., Persson H. and Olson L. (1990) Localization of brain-derived neurotrophic factor mRNA to neurons in the brain by 11 sifu hybridization. Expl. Neural. 109, 141~~152. Whittemore S. R., Friedman P. L., Larhammar D., Persson H., Gonzalez-Carvarjarl M. and Holets V. R. f 1988) Rat beta-nerve growth factor sequence and site of synthesis in the hippocampus. J. Neurosci. Res. 20, 403 410. Williams L. R.. Jodelsi K. S. and Donald M. R. (1989) Axotomv-deoendent stimulation of choline acetvltransferase activity by exogenous mouse nerve growth factor in adult rat b&al forebrain. Brain Rex 498, 243 256.. Woolf N. J. (1991) Cholinergic systems in mammalian brain and spinal cord. Prog. Neurobioi. 37, 475-524. Zafra F., Hengerer B., Leibrock J.. Thoenen 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. moles. Biol. Org. J. 9, 3545-3550. Zafra F., Castren E., Thoenen H. and Lindholm D. (1901) Interplay between glutamate and r-aminobutyric acid transmitter systems in the physiological regulation of brain-derived neurotrophic factor and nerve growth factor synthesis in hippocampal neurons. Proc. nam. Acad. Sci. U.S.A. 88, IO.03710,041.

(Accepted 2 September 1992)