Acidic fibroblast growth factor mRNA is expressed by basal forebrain and striatal cholinergic neuronsкод для вставкиСкачать
THE JOURNAL OF COMPARATIVE NEUROLOGY 366379-389 (1996) Acidic Fibroblast Growth Factor mRNA is Expressed by Basal Forebrain and Striatal Cholinergic Neurons JENNIFER L. BIZON, JULIE C. LAUTERBORN, PAUL J. ISACKSON, AND CHRISTINE M. GALL Departments of Psychobiology (J.L.B., C.M.G.) and Anatomy and Neurobiology (J.C.L., C.M.G), University of California at Irvine, Irvine, California 92717; Department of Biochemistry and Molecular Biology, The Mayo Clinic-Jacksonville, Jacksonville, Florida 32224 (P.J.I.) ABSTRACT Evidence for the importance of the basal forebrain cholinergic system in the maintenance of cognitive function has stimulated efforts to identify trophic mechanisms that protect this cell population from atrophy and dysfunction associated with aging and disease. Acidic fibroblast growth factor (aFGF) has been reported to support cholinergic neuronal survival and has been localized in basal forebrain with the use of immunohistochemical techniques. Although these data indicate that aFGF is present in regions containing cholinergic cell bodies, the actual site of synthesis of this factor has yet to be determined. In the present study, in situ hybridization techniques were used to evaluate the distribution and possible colocalization of mRNAs for aFGF and the cholinergic neuron marker choline acetyltransferase (ChAT) in basal forebrain and striatum. In single-labeling preparations, aFGF mRNA-containing neurons were found to be codistributed with ChAT mRNA+ cells throughout all fields of basal forebrain, including the medial septumidiagonal band complex and striatum. By using a double-labeling (colormetric and isotopic) technique, high levels of colocalization (over 85%) of aFGF and ChAT mRNAs were observed in the medial septum, the diagonal bands of Broca, the magnocellular preoptic area, and the nucleus basalis of Meynert. The degree of colocalization was lower in the striatum, with 64% of the cholinergic cells in the caudate and 33% in the ventral striatum and olfactory tubercle labeled by the aFGF cRNA. These data demonstrate substantial regionally specific patterns of colocalization and support the hypothesis that, via a n autocrine mechanism, aFGF provides local trophic support for cholinergic neurons in the basal forebrain and the striatum. 1996 Wilcy-Liss, Inc. Indexing terms: in situ hybridization, choline acetyltransferase, neurotrophic factor, caudate, Alzheimer’s disease Basal forebrain cholinergic neurons are believed to be critical for the maintenance of cognitive abilities in humans and in rodents, and they are particularly vulnerable to dysfunction and degeneration with age and disease (Fischer et al., 1989, 1991; Gallagher et al., 1990; Stroessor-Johnson et al., 1992; Armstrong et al., 1993). A large body of literature has correlated the atrophy of these cells and the concomitant decline in cholinergic indices with memory and attentional deficits associated with such disorders as Alzheimer’s disease (Whitehouse et al., 1982; Coyle et al., 1983; Collerton, 1986). Because of this age-related pathology, a great deal of effort has been made to identify endogenous factors that sustain cholinergic neuronal viability. Whereas past research in this area has primarily focused on distant target-derived neurotrophic factors, recent findings suggest 1996 WILEY-LISS. INC. that there are important trophic interactions within the basal forebrain itself. The role of target-derived nerve growth factor (NGF) in the survival of basal forebrain cholinergic cells is well documented. It has been demonstrated that NGF is synthesized in the hippocampus (Gall and Isackson, 1989; Ernfors et al., 1990; Phillips et al., 19901, sequestered by cholinergic afferent axons, and retrogradely transported back to the cell bodies located in the medial septum (Schwab et al., 1979). Following fimbriaifornix transection, which severs cholinergic afferents to the hippocampus and, thereby, Accepted September 14, 1995. Address reprint requests to Christine M. Gall, Department of Anatomy and Neurobioloa, University of California at Irvine, Irvine, CA 92717. E-mail:cgall~~r uci.edu 380 blocks retrograde transport, the majority of the medial septal neurons undergo changes in phenotype, including marked decreases in choline acetyltransferase (ChAT; Gage et al., 1986; Armstronget al., 1987; Sofroniew et al., 1987). In these fields, the number of ChAT-immunoreactive neurons is reportedly reduced by as much as 75% a t 3 weeks following transection (Sofroniew et al., 1987; Peterson et al., 1990). However, the cholinergic cells can be rescued by the infusion of various trophic factors, including not only NGF (Kromer, 1987; Koliatsos et al., 1990; Venero et al., 1994) but also brain-derived neurotrophic factor (BDNF; Morse et al., 1993; Koliatsos et al., 1994; Venero et al., 1994) and fibroblast growth factor (FGF; Anderson et al., 1988; Otto et al., 1989; Gomez-Pinilla et al., 19921, all ofwhich are synthesized normally in the hippocampus (Emoto et al., 1989; Ernfors et al., 1990: Phillips et al., 1990; Gall et al., 1994). Although these findings suggest the importance of targetderived neurotrophic factors in the survival of cholinergic neurons following insult, a growing body of evidence indicates that factors from distant targets may not be an essential determinant of viability under normal circumstances. First, although the cholinergic neurons are particularly susceptible to pathology that is associated with the aging process, a n age-related decline of NGF levels in the hippocampus has not been demonstrated in rat (Hellweg et al., 1990; Crutcher and Weingartner, 1991) nor have postmortem examinations of cholinergic cortical targets in Alzheimer's patients revealed deficits in NGF mRNA (Jette et al., 1994)or protein (Allen et al., 1991).Second, although the medial septal cholinergic cells atrophy following transection of their axonal projections to the hippocampus, these cells survive the removal of all distant trophic support by excitotoxic lesion of hippocampal neurons, which spares the cholinergic axons themselves (Sofroniew et al., 1990, 1993; Kordower et al., 1992). Moreover, transgenic mice that do not express NGF (Crowley et al., 1994) or its high-affinity receptor TrkA (Smeyne et al., 1994) retain cholinergic perikarya in basal forebrain, although the TrkA mutants reportedly fail to maintain cholinergic basal forebrain afferents to the hippocampus (Smeyne et al., 1994). Survival of the cholinergic neurons despite the absence of their target cells, NGF, or the NGF receptor suggests that the cholinergic cells may receive trophic support 1)other than NGF and 2 ) from outside their principal cortical targets. Because very few of the neurons in the medial septum project both to the hippocampus and to another region (Amaral and Kurz, 1985), the idea of local trophic support for these cells becomes compelling. FGFs were noted originally for their ability to promote mitosis in fibroblasts (for review, see Burgess and Maciag, 1989). These broadly acting factors are now known to promote the development and survival of cultured neurons from virtually all brain regions tested (Morrison et al., 1986; Walicke, 1988; Ferrari et al., 1989; Grothe et al., 1989; Garcia-Rill et al., 1991; Kushima et al., 1992).In vivo, they can rescue neurons in basal forebrain (Anderson et al., 1988; Otto et al., 1989; Gomez-Pinilla et al., 1992) and entorhinal cortex (Cummings et al., 1992) from the degenerative effects of axotomy, and they support hippocampal CA1 neurons following cerebral ischemia (Nakata et al., 1993).Acidic and basic FGF (aFGF and bFGF, respectively) act via the same receptors (Asai et al., 1993; Yazaki et al., 19941, and, generally, both are effective in the support of neuronal survival in vitro and in vivo. However, with regard to the viability of cholinergic basal forebrain cells, the .J.L. BIZON ET AL. involvement of aFGF (a.k.a. FGF1) is of particular interest. Recently, immunoreactivity for this FGF family member has been localized in neuronal perikarya within regions of the cholinergic basal forebrain, including the septum and the nucleus basalis of Meynert (Stock et al., 1992),suggesting that this factor may be synthesized locally. However, the distribution of protein immunoreactivity does not settle the issue of the site of synthesis. The possibility exists that aFGF, like other neurotrophic factors that are associated with the cholinergic system, is produced elsewhere and then is transported to the basal forebrain. The goal of the present study was to determine the distribution of those cells that synthesize aFGF mRNA in relation to forebrain neurons expressing the mRNA for ChAT, which is a cholinergic cell marker. In addition, the possibility that aFGF acts as an autocrine neurotrophic factor for the cholinergic forebrain neurons was investigated using an in situ hybridization technique that allowed for the simultaneous localization of mRNAs for ChAT and aFGF. The results demonstrate substantial but regionspecific patterns of colocalization and are consistent with the hypothesis that aFGF acts as an autocrine neurotrophic factor for cholinergic neurons of basal forebrain and striatum. MATERIALS AND METHODS Adult male Sprague Dawley rats ( n = 14) were killed with an overdose of sodium pentobarbital (100 mgikg) and were perfused with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. The brains were removed from the crania, postfixed in perfusate for 24 hours, cryoprotected in 20% sucrose and 4% paraformaldehyde for 48 hours, and then sectioned on a freezing microtome (25 Fm, coronal plane) through the rostra1 forebrain, beginning just caudal to the olfactory bulbs and ending at the midrostral hippocampus. Adjacent series of every ninth section were collected into cold 4%,paraformaldehyde. Free-floating tissue sections were processed either for single-labeling or for double-labeling in situ hybridization as previously described (Lauterborn et al., 1993a,b). For single-labeling experiments, adjacent sections from one group of animals (n = 3 ) were processed for the localization of either ChAT mRNA or aFGF mRNA by using:35SS-labeled cRNA probes. For double-labeling experiments, sections from another group of animals ( n = 11)were processed for the simultaneous detection of ChAT mRNA and aFGF mRNA within individual tissue sections by using a digoxigenin (DIG)-labeledChAT cRNA ("'"ChAT) and :%labeled aFGF cRNA. The ChAT riboprobe was transcribed from a 431 base pair fragment of rat C U T cDNA (Lauterborn et al., 1993b). For the sense and antisense ChAT cRNAs, the constructs were digested with Pvu 11, and transcripts were generated in the presence of 5'-a-I % lthiotriphosphate by using T3 and T7 RNA polymerase. The aFGF :W-labeled riboprobe was transcribed from a 510 base pair fragment (corresponding to nucleotides 245-755 in Goodrich et al., 1989) of rat aFGF cDNA. The aFGF cDNA was obtained by the polymerase chain reaction (PCR) from rat neocortical cDNA 24 hours after the induction of limbic seizures. The oligonucleotide primers (20 bases in length) used for PCR were synthesized based on the published sequence of rat heparin-binding growth factor 1 (Goodrich et al., 1989). The isolated aFGF cDNA was cloned into the EcoRl site of pBS. For antisense aFGF cRNA, the construct was digested aFGF EXPRESSION BY CHOLINERGIC NEURONS 38 1 with P w 11, and transcripts were generated with T3 RNA using sense DIG- or :35S-cRNAprobes or 2) omitting the polymerase in the presence of "S-UTP. An internal Pvu I1 antisense probes. Cellular labeling was not detected in either site exists in the aFGF sequence; therefore, the final result of these conditions. To determine whether the immunohistois a 350-base antisense probe. For sense aFGF cRNA, the chemical procedure produced a nonspecific autoradiographic construct was digested with BamH1, and transcripts were signal, the double-labeling technique was performed as generated with T7 RNA polymerase in the presence of described above, but the 35S-aFGFcRNA was omitted from the initial hybridization incubation. No autoradiographic :35SS-UTP. For both single-labeling and double-labeling experi- grains were detected under these control conditions. ments, free-floating sections were pretreated with proteinAnalysis of colocalization ase K (1 mg/ml in 0.1 M Tris buffer, pH 8.0, with 50 mM EDTA) for 30 minutes, then in 0.25% acetic anhydride in By using a combination of light- and darkfield micros0.1 M triethanolamine for 10 minutes, and then rinsed in copy, tissue processed according to the double-labeling 2 x saline sodium citrate buffer (SSC; 1 x SSC = 0.15 M technique was evaluated to determine the distribution and NaC110.015 M Na-citrate, pH 7.0). Tissue was incubated for the number of cells in specific fields of the rat basal 40-48 hours at 60°C in hybridization buffer (50% for- forebrain that were hybridized with both the D'GChAT mamide, 10% dextran sulfate, 0.7% ficoll, 0.7% polyvinyl cRNA and the "SS-aFGF cRNA (double-labeled) or with pyrrolidone, 0.7% bovine serum albumin, 0.15 mg/ml yeast each cRNA alone (single-labeled). The regions analyzed transfer RNA, 0.33 mgiml denatured herring sperm DNA, included the medial septa1 nucleus, the horizontal and and 40 mM dithiothreitol) containing either the :j5SS-labeled vertical limbs of the diagonal bands of Broca, the caudate/ ChAT cRNA or the 35s-labeled aFGF cRNA separately (for putamen, the olfactory tubercle, the magnocellular preoptic the single-labeling studies) or containing a cocktail of the area, and the nucleus basalis of Meynert. Single- and "I(;ChAT cRNA and the "5S-labeled aFGF cRNA (for the double-labeled cells were counted from six camera lucida double-labeling studies). In all cases, the 35S-labeled cRNA drawings from each of four rats that corresponded to was added to a final density of 1 x lo6 cpmilO0 ~ 1and , the rostrocaudal levels represented in plates 12-17 of Paxinos DIG-labeled cRNA was diluted 1:1,000 in the hybridization and Watson (Paxinos and Watson, 1986). Because of diffibuffer. After hybridization, tissue processed for single- culty in determining cytoarchitectonic boundaries in the labeling was rinsed in 4 x SSC, treated with 20 pgiml RNase-treated tissue, the field designated as ventral striaribonuclease A in 10 mM Tris saline, pH 8.0, with 1 mM tum included the ventral one-third of neostriatum and the EDTA for 30 minutes at 45"C, washed through descending nucleus accumbens. The percentages of DIGCUTcRNA+ concentrations of SSC to a final wash in 0.1 x SSC at 60°C, cells that were also labeled by the :35S-aFGFcRNA and of Y3-aFGF cRNA+ cells that were also labeled by the and then mounted onto gelatin-coated slides. For tissue processed for double-labeling, following hybrid- DrGChATcRNA were determined for select regions. ization, the sections were rinsed in 50% formamidei2 x SSC a t 60"C, rinsed in 2 x SSC at room temperature (RT), RESULTS and treated with ribonuclease A (as above). The tissue was Distribution of aFGF in the basal forebrain then washed in descending concentrations of SSC to a final wash in 0.1 x SSC a t 60"C, rinsed in 50% formamidei2 x The distribution of labeling in tissue sections processed SSC ( 5 minutes) at 60°C, and incubated in 0.1 x SSC for hybridization of either aFGF cRNA or ChAT cRNA alone containing 2% normal sheep serum (NSS) and 0.05% revealed similar patterns of expression in basal ganglia and Triton-X 100 at RT for 1 hour. Following several rinses in in other regions of basal forebrain. Figure 1A,B shows that, Tris-buffered saline (TBS; 100 mM Tris-HCL, 150 mM for both aFGF and ChAT cRNAs, there were large numbers NaCI, pH 7.5), the tissue was incubated with a sheep of heavily labeled cells in the medial septum and the diagonal anti-DIG antiserum (Boehringer Mannheim) diluted 1:2,000 bands of Broca, and there were somewhat fewer densely in TBS containing 2% NSS,0.05% Triton-X 100, and 10 labeled cells scattered across the caudateiputamen and the mM sodium thiosulfate at 37°C for 2 hours. The tissue was olfactory tubercle. Unlike ChAT mRNA, aFGF mRNA then washed in TBS a t RT, rinsed in 0.1 M Tris-HCLi100 expression was also detected throughout the lateral septum mM NaClI50 mM MgC12, pH 9.5 (Tris-MgC12), and incu- (Fig. 1B) and the lateral preoptic area, although, in these bated in Tris-MgClz containing nitroblue tetrazolium (0.34 regions, the autoradiographic grains were diffusely scatmgiml), 5-bromo-4-chloro-3-indolyl phosphate (0.18 mgi tered, and individual labeled neurons were not easily identified. Figure 1C,D shows that, in more caudal regions, cells ml), and levamisole (0.24 mgiml) in the dark for 18-24 hours at RT. Following this colormetric reaction, tissue was containing aFGF and ChAT mRNAs were distributed simirinsed in 10 mM Tris-HCl/l mM EDTA, pH 8.0, and larly within the magnocellular preoptic area and the nucleus basalis of Meynert. At these same levels, several other brain mounted onto gelatin-coated slides. The slide-mounted tissue sections were defatted through regions lacking ChAT mRNA expression contained cells alcohol and Americlear, rehydrated, and then air dried. that were heavily labeled by the aFGF cRNA. Most particuSlides for single-labeled tissue were dipped in Kodak emul- larly, aFGF mRNA expression was high in the anterodorsal, sion (NTBS), and slides for double-labeled tissue were paraventricular, and reuniens nuclei of the thalamus; dipped in Ilford K5.d emulsion. All slides were stored a t 4°C labeling in the anteromedial and anteroventral thalamic for 4-6 weeks, after which the autoradiograms were devel- nuclei was less dense. There was a notable absence of aFGF oped with Kodak D19 (1:l with H 2 0 )and fixed with Kodak cRNA labeling in the paratenial nucleus of thalamus. regular fixer. The slides were then rinsed through an Colocalization of aFGF mRNA with ChAT ascending series of alcohols (50-100%) and Americlear and mRNA in basal forebrain and striatum were coverslipped with Permount. Labeling specificity was assessed by processing tissue In tissue hybridized simultaneously with DIGChATand according to the aforementioned protocol but either 1) 35S-aFGF probes, cells were visualized through a combina- . J . L BIZON ET AL. Fig. I . Darkfield photomicrogmphs showing the autoradiographic localization of Y-choline acetyltransferase t ChAT) cRNA lA,C) and "S-acidic. fibroblast growth factor IaFGF! cRNA IB,D) hybridization in semiadjacent sections through control rat forebrain a t t he level of the medial wptal nucleus tA,B; MS) and the rostra1 thalamus and nucleus basalis of Mevnert 1C.D; nB). A,B: Both cRNAs label cells tha t a re similarly distributed in t he medial septum, the diagonal bands of Broca tion of light- and darkfield microscopy and were examined for double-labeling. Examples of the pattern of cellular labeling are shown in Figure 2, where the colormetrici immunohistochemical label (seen as a blue reaction product in the tissue section) indicates the localization of C U T mRNA, and the autoradiographic silver grains mark aFGF cRNA hybridization. Figure 2A,B shows that some neurons were singly labeled for either aFGF mRNA or C U T mRNA. whereas others cells were found to contain mRNAs for both ChAT and aFGF. Plots of the distributions of cells labeled with ChAT cRNA alone or with both cRNAs are presented in Figure 3 . In all of the forebrain regions analyzed, some degree of aFGF cRNA labeling was observed in cholinergic cells, although the density of labeling with each probe varied with respect to anatomical location. Characteristic of the cholinergic neurons distributed throughout the basal forebrain, the ChAT mRNA+ cells in medial septum had large round perikarya that displayed intense color labeling. Figure 2C and the plots of labeled cells presented in Figure 3A show that all of the ChAT mRNA' (immunohistochemically labeled) cells in this region were also labeled with dense accumulations of autoradiographic grains. The r'i';ChAT cRNA-labeled perikarya in the horizontal and vertical limbs (NDB), the olfactory tubercle (Or,,and the caudatesputamen (<:Pi. Labeling with the a FG F cRNA alone is seen within the lateral septum (LS).C,D: In more caudal sections, ChAT cRNA ( C )and aFGF cRNA (D! both label cells in the mafinocellular preoptic area tMCPAi and the nucleus basalis. A t these levels, the a FG F cRNA also labels the and the intralaminar fields suranterodorsal thalamic niicleus (AD! rounding the paratenial nucleus (asterisk,. Scale bar = 750 p m . of the diagonal bands were reminiscent of the medial septa1 cholinergic neurons, although, within peripheral aspects of these fields, the labeled cells had somewhat smaller somata. Fig. 3. Brighttield photomicrographs showing the cellular localization of ,Y%IFGF cRNA (seen as black autoradiographic g-ainsi and digoxigenin ( DIG,-labelcd ChAT cRNA t"'%hAT cRNA; immunohistochemical color-reaction product, hybridization in basal forebrain and striatum. Labeled cells a re shown in th e olfactory tubercle ( A , . the , medial septa1 nucleus tC1. the vertical ( D ! caudate/putamcn ( B J the and horizontal tE! limbs of the diagonal bands of Broca. the magnocellular preoptic area IF,, and the nucleus basalis of Meynert ( G I .Arrnwheads indicate cells labeled with the a FG F cKNA alone, open arrows indicate cclls labeled with the ChAT cRNA alone, and solid arrows indicate double-labeled neurons. In both the olfactory tubercle (A1 and the caudatc/putanien (HI, S O ~ Ccells a re single-labeled with aFGF cRNA or with ChAT cRNA, whereas other similar-sized neurons are double?-labeled. I n contrast, in the medial septum tC), all labeled cells were large and wvre densely double-labeled by both markers. In the vertical (D! and horizontal t E ) limbs o f t h e diagonal bands of Broca, the &Teat majority of labeled cells were double-labeled; however. as shown here, a few singlc,-lebeled cells were seen a t the lateral edge of each field (in D, the lateral aspect o f t h e field appears on the right). Finally. the large cholinergic colls of the magx)cellular preoptic area IF) and the nucleus basalis tG! were virtually all double-labeled as in the fields shown. Scale bar := 30 ym in A-C,F,G. 35 pn in D.E. aFGF EXPRESSION BY CHOLINERGIC NEURONS 383 Figure 2 .J.L. BIZON ET AL. A c I i Fig. 3 . A-D: Schematic illustrations and bar graphs showing t h e proportion of cholinergic (i.e., ChAT mRNA' 1 cells in basal forebrain and the striatum t h a t also contain a F G F mRNA. A: Drawings show t h e distributions of neurons that express both a F G F mRNA and ChAT mRNA (solid triangles) and those t h a t express ChAT mRNA alone [open circles) in basal forebrain and basal ganglia (small symbols indicate one cell; large symbols indicate four to six cells). Labeling in lateral scptum, cortex, diencephalon, and bed nucleus of stria terminalis IBS'I') was not plotted, nor were cells t h a t were labeled with a F G F cKNA alone. The three schematics, from top to bottom, show plots of t h e rostral. middle, and caudal basal forebrain, respectively. It can be seen here that double-labeled cells were most numerous in medial regions, including t h e medial septum (MS) and t h e diagonal bands of Broca ( N D R I followed by the magnocellular preoptic area IMCPA) and t h e nucleus basalis i nB). Double-labeled cells were fairly evcnly interspersed with single-labelcd cells in t h e caudateiputamen ( C P )and were somewhat less frequently encountered in the olfactory tubercle ( O T ) . B-D: Bar graphs showing counts of single- and double-labeled ChAT mHNA* cells within t h e different cytoarchitectonic fields found in individual tissue sections through thcsc same rostral IB), middle tC). and caudal (D)planes. For each area, the black portion of the bar indicates the number of cells labeled with both a F G F and ChAT cRNAs, whereas t h e stippled portion ( n o t including black s p a n ) shows the number labeled with ChA'T cRNA alone. AAA, anterior amygdaloid area; AcB, nucleus accumbens; aco, anterior commissure: En, endopiriform nucleus; GP, globus pallidus; LHA, lateral hypothalamic area: LS, lateral scptum: opt, optic tract; PIR, piriform cortex; 1'0, preoptic area; S F , septofimbrial nucleus; VP. ventral pallidum; VS, v m t r a l striatum. aFGF EXPRESSION BY CHOLINERGIC NEURONS Moreover, in these cells, the color labeling within the cytoplasm could be seen frequently surrounding a n unlabeled nuclear region. Virtually all of the DIGChATcRNA+ cells in the diagonal bands were double-labeled (Fig. 2D,E), although the density of aFGF cRNA labeling was less than that observed in the medial septum (Fig. 2C). In addition to those cells in the basal forebrain cholinergic system that give rise to cortical efferents, cholinergic cells are distributed across the caudateiputamen, the ventral striatum, and the olfactory tubercle, and previous studies have demonstrated that the latter cells are primarily interneurons (Woolf and Butcher, 1981). In the present material, large DIGChAT-labeledperikarya were found in each of these regions and, as shown in Figure 2A,B and in the plots in Figure 3, a portion of these cells also contained aFGF mRNA. For double-labeled perikarya in the striatum and the olfactory tubercle, the density of aFGF cRNA labeling was less than that observed in other forebrain areas (e.g., the medial septum). Moreover, among these fields, both the number of double-labeled cells (Fig. 3) and the density of W - a F G F cRNA labeling (Fig. 2) appeared greatest in the dorsolateral caudate and decreased toward the more medial aspects of the caudate and toward the ventral striatum and the olfactory tubercle. The ChAT mRNA+ cells in the magnocellular preoptic area were similar to neurons in the horizontal limb of the diagonal band with regard to soma1 size, shape, and incidence of "S-aFGF cRNA labeling. However, in the magnocellular preoptic area (Fig. 2F), the density of the autoradiographic label over individual cells was somewhat lower than in medial septum. The DrGChATcRNA-labeled perikarya in the nucleus basalis of Meynert were very large and elongated. Figure 2G shows that all of the labeled cells in this region were double-labeled; however, the 35SS-aFGF cRNAlabeling was much more diffuse than in other forebrain areas analyzed. Quantification of ChAT/aFGF colocalization The schematic illustrations and bar graphs in Figure 3 show that the number of double-labeled somata varied substantially with respect to anatomical location. There were very high levels of colocalization in regions traditionally associated with the cholinergic basal forebrain. All DIGChATcRNA-labeled neurons in the medial septum (Fig. 3B) and the nucleus basalis of Meynert (Fig. 3D) exhibited dense %-aFGF cRNA labeling. In the horizontal limb of the diagonal band (Fig. 3B) and in the more caudal magnocellular preoptic area (Fig. 3C), over 85% of the ChAT mRNA' cells were double-labeled. Fewer of the probable cholinergic interneurons identified in the regions of the dorsal caudate, the ventral striatum, and the olfactory tubercle contained aFGF mRNA. Figure 3B,C and Table 1show that, among these fields, the highest level of colocalization was detected in the caudate, where 64% of the ChAT mRNA+ cells also contained aFGF mRNA. Fewer than 32% of the DIGChATcRNA+ neurons in the ventral striatum and the olfactory tubercle were also "S-aFGF cRNA+. In these regions, the total number of aFGF mRNA-containing somata were also counted and examined for the presence of ChAT mRNA. In general, there were fewer cells labeled with 3sS-aFGF cRNA alone than there were double-labeled cells. Of the aFGF mRNAcontaining cells, the highest incidence of colocalization was found within the dorsal caudate (770/0), whereas less than 385 TABLE 1. Cell Counts and Percentages of Single- and Double-Labeled Cells in Basal Ganglia and Olfactory Tubercle' Single-labeled cells Region Dorsd caudate Ventralstriaturn Olfactory tubercle Double-labeled cells ChAT cRNA (no.) aFGF cRNA (no.) aFGFiChAT cRNA (no.) ChAT cRNA (%I (%1 23 i 9.6 11 ? 3.4 10 i 2.2 12 2 4.7 38 ? 6.2 10 i 3.4 9 I 2.5 64 2 9.8 32 2 11.7 25 I 6.2 77 i 3.3 43 i 11.6 44 2 9.2 22 t 5.9 27 2 4.9 'Cell counts and pen.entd#?s represent average per section i aFGF cKNA S.D.: analysis of t o u r rats totdl. 43% of the cells in the ventral striatum and the olfactory tubercle were double-labeled. DISCUSSION The present results demonstrate that aFGF mRNA is distributed in all fields of the cholinergic basal forebrain and is colocalized with ChAT mRNA in each, although the extent of colocalization varies with respect to anatomical region. The distribution of aFGF mRNA alone was in good agreement with the immunohistochemical localization of aFGF protein reported by Stock et al. (1992). In each of these studies, labeling was dense in the diagonal bands of Broca, the nucleus basalis of Meynert, and the anterodorsal and paraventricular thalamic nuclei; was moderately dense in the lateral septum and preoptic areas; and was associated with scattered cells in the caudateiputamen and the olfactory tubercle. Together with data on the distributions of aFGF immunoreactivity and mRNA in brainstem and spinal cord (Bean et al., 1991; Elde et al., 1991), these results indicate that aFGF protein generally accumulates within, and serves as a good marker for, cells that produce this factor. Moreover, these data reinforce the conclusion that, in brain, aFGF mRNA is expressed by restricted populations of neurons (Elde et al., 1991) unlike bFGF, which has been shown to be expressed by some neuronal groups in addition to diffusely distributed glial cells (Emoto et al., 1989; Gall et al., 1994). Results ofthe double-labeling analysis demonstrate that aFGF mRNA is expressed by virtually all cholinergic neurons in the medial septum and the nucleus basalis of Meynert. In contrast, in the horizontal limb of the diagonal bands, magnocellular preoptic area, and striatum, there were substantial numbers of doublelabeled neurons codistributed with cells producing aFGF mRNA or ChAT mRNA alone, suggesting a heterogeneity of cell types within both the aFGF+ and the ChAT+ populations. The expression of aFGF within basal forebrain is particularly intriguing in light of evidence that aFGF, like the better-characterized neurotrophins, has the ability to preserve the integrity of cholinergic forebrain neurons. Because bFGF and aFGF act at the same receptor (Asai et al., 1993; Yazaki et al., 19941, the effects of each are indicative of the potential functions for locally produced aFGF. Following exogenous application, both factors promote the differentiation (Burgess and Maciag, 1989; Baird and Klagsbrun, 1991) and increase the ChAT activity (Grothe et al., 1989; Knusel et al., 1990; Kushima et al., 1992; Figueiredo et al., 1993; Yokoyama et al., 1994) of cultured septa1 cholinergic neurons. In vivo infusion of exogenous bFGF after a fimbria/fornix transection (Anderson et al., 1988; Otto et al., 1989; Gomez-Pinilla et al., 1992) or of aFGF after neocortical lesion (Figueiredo et al., 1993) preserves ChAT 386 expression by basal forebrain neurons. Some of these effects may involve interactions with the neurotrophins. The FGFs are known to stimulate NGF expression and release in vitro (On0 et al., 1991; Yoshida and Gage, 19911, and intraventricular bFGF infusion after fimbriaifornix transection maintains the expression of the low-affinity NGF receptor (a.k.a. p75) by basal forebrain cholinergic cells (Gtimez-Pinilla et al., 1992). Thus, there may be synergistic interactions between locally produced aFGF and target-derived neurotrophins in the support of cholinergic neuronal viability. It is well established that the cholinergic neurons distributed across the basal forebrain are not homogeneous but differ with respect to both anatomical connectivity and expression of biochemical markers. The cholinergic neurons in the medial septum and the nucleus basalis of Meynert send axons almost exclusively to one cortical target, the hippocampus and the neocortex, respectively (Woolf et al., 1984; Paxinos and Butcher, 1985). Cells distributed in lateral and ventral fields (i.e., the horizontal limb of the diagonal bands and magnocellular preoptic areas) make more varied and multiple projections across telencephalic structures innervating the entorhinal and cingulate cortices and the olfactory bulb (Woolf et al., 1984; Paxinos and Butcher, 1985; Zaborsky et al., 1986). Within ventral pallidum and substantia innominata, two major subpopulations of cholinergc cells have been identified that can be differentiated on the basis of their predominant efferent projections to cortex or amygdala (Carlsen et al., 1985) and on the basis of their vulnerability to either quisqualic or ibotenic acid, respectively (Boegman et al., 1992). The colocalization of ChAT and aFGF mRNA, as seen here, was greatest in the areas of least cholinergic cell diversity. All cholinergic cells in the medial septum and nucleus basalis and 85%of cholinergic cells in the horizontal limb of the diagonal band and magnocellular preoptic area were double-labeled. Across ventral striatum, ventral pallidum, and lateral substantia innominata, colocalization was less frequently encountered, and major subpopulations were identified that expressed either aFGF or ChAT mRNA alone. The increased diversity of expression patterns in fields containing greater numbers of cholinergic cell types (differentiated by connectional and other biochemical criteria) suggests that the presence of aFGFiChAT expression or of aFGF expression alone may be associated with specific populations of cells. Thus, within the ventral pallidumi substantia innominata, aFGFiChAT colocalization may be a feature of one particular efferent population. Moreover, differences in aFGF mRNA expression across basal forebrain areas could be indicative of distinct trophic requirements across cholinergic populations and, most particularly, could indicate differences in the degree to which these cells depend on distant trophic support. It is well documented that the cholinergic cells of the medial septum and the nucleus of basalis, which exhibit robust aFGF mRNA expression, can survive the ablation of their cortical targets and the removal of putative neurotrophic support from these sources (Sofroniew et al., 1990, 1993; Kordower et al., 1992).It will be interesting to determine whether cholinerg c cells lacking aFGF expression (e.g., those in the olfactory tubercle or in the lateral aspects of the diagonal bands of Broca) are more vulnerable to degeneration following removal of their forebrain targets and of potential distant sources of trophic support. .J.L. BIZON ET AL. An additional consideration with regard to the singlelabeled cholinergic neurons in the horizontal limb of the diagonal band and in magnocellular preoptic areas is that, although these cells do not produce aFGF, they may receive other local neurotrophic support. NGF-synthesizing neurons have been localized in the lateral aspects of the horizontal limb of the diagonal band and in the magnocellular preoptic area, the olfactory tubercle, and the ventral pallidum (Lauterborn et al., 1991; Conner et al., 1992). These intensely NGF mRNA+ cells contain glutamic acid decarboxylase mRNA, indicating that they are GABAergic (Lauterborn et al., 1995). Thus, the cholinergic cells in lateral basal forebrain may receive local trophic support from neighboring cholinergc neurons producing aFGF as well as from closely situated GABAergic neurons producing NGF. In future studies, it will be of interest to determine whether cholinergic cells in these fields have access to local sources of both NGF and aFGF or whether separate neuronal populations benefit from different, locally supplied trophic factors. Confirming previous in situ hybridization investigations (Ibanez et al., 1991; Lauterborn et al., 1993b1, many ChAT mRNA+ neurons were localized throughout the basal ganglia and the olfactory tubercle in the present study. The distribution and size of these cells indicate that they are the well-described large, aspiny neurons thought to have primarily local projection fields (Woolf and Butcher, 1981). Only a portion of the cholinergic striatal neurons expressed aFGF mRNA (i.e., 64% in the dorsal and lateral aspects of the caudate and 32% in more medial and ventral fields). In these regions, a large proportion of the aFGF-cRNA-labeled cells did not express ChAT mRNA (about 40% overall). The single-labeled aFGF neurons were similar to the cholinergic interneurons, in that they were large and were sparsely scattered throughout the striatum. These characteristics suggest that the single-labeled aFGF neurons are not the medium spiny projection neurons that account for over 90% of cells in the striatum, but they represent the second major interneuronal group, i.e., the large aspiny GABAergic interneurons (Heimer et al., 1985).These conclusions are consistent with the immunocytochemical analysis of Stock et al. ( 1992), which suggested that aFGF-immunoreactivity was localized in two distinct populations of interneurons in the neostriatum. The localization of aFGF in striatal interneurons is noteworthy, in that these cells are among the least vulnerable neurons in the striatum, which is demonstrated by their selective survival following ischemia and excitotoxic insult (Beal et al., 1990; Chesselet et al., 1990; Gonzales et al., 1992).This suggests that the local synthesis and actions of aFGF may afford protection to striatal interneurons under these circumstances. Apart from potential autocrine actions, aFGF produced in striatum and basal forebrain may have trophic effects on afferents to these regions. The nigrostriatal afferents are of particular interest in this regard. It has been reported recently that the injection of MPTP', which destroys the majority of the dopamine-producing cells in the substantia nigra, increases aFGF and bFGF mRNA levels in the striatum (Leonard et. al., 1993). Furthermore, infusion of aFGF and bFGF into the striatum after a MPTP+ lesion significantly increases striatal dopamine levels (Date et al., 1990). These findings suggest that aFGF produced in the basal ganglia and the basal forebrain may be transported retrogradely by afferents to these regions, including the dopaminergic somata in the substantia nigra. With regard aFGF EXPRESSION BY CHOLINERGIC NEURONS to this point, C U T + terminals in striatum have been shown to lie in close proximity to tyrosine hydroxylasepositive terminals (Pickel and Chan, 1990; Dimova et al., 19931, allowing for the possibility that FGF could be released by cholinergic processes and then acquired and retrogradely transported by dopaminergic afferents. The dopaminergic neurons of the substantia nigra are known to produce and respond to BDNF, neurotrophin-3, and FGF (Knusel et al., 1990; Bean et al., 1991; Spina et al., 1992; Hyman et al., 1994; Seroogy et al., 1994), suggesting that these factors have a normal autocrine role in dopaminergic neuronal survival. The present findings suggest that, in addition to these locally produced trophic factors, aFGF synthesized in the striatum may provide trophic support for midbrain dopaminergic neurons. For deciphering the role of aFGF in basal forebrain, it will be important to determine which cell populations express the receptor for aFGF and, thus, presumably respond to this factor. Although FGF can rescue forebrain cholinergic cells following injury, it remains unclear whether FGF is acting directly via receptors on the cholinergic cells or through more indirect means, such as promoting the release of other neurotrophic factors from nearby glial cells (Burgess and Maciag, 1989; Yoshida and Gage, 1991). Recently, using sensitive in situ hybridization techniques, Yoshida et al. (1994) reported partial colocalization of mRNA for the flg FGF receptor with CUT-like immunoreactivity in cells of the diagonal bands of Broca and the nucleus basalis but not in cells of the medial septum. Thompson et al. (1991) have reported that the FGFR3 receptor is expressed by neurons throughout basal ganglia, although other investigators have found that this FGF receptor is expressed preferentially by glial cells (Yazaki et al., 19941. An important goal for future studies will be to determine whether these receptors are expressed by the different populations of cholinergic basal forebrain neurons and, thus, might mediate autocrine trophic actions. The present findings reinforce the observation (Stock et al., 1992) that aFGF is expressed prominently by populations of forebrain neurons that are particularly susceptible to degeneration with age and age-related neurological disease. This correspondence gives rise to the hypothesis that the cholinergic neurons normally benefit from aFGF action and become vulnerable if the availability of aFGF fails under certain pathological conditions. It is not known whether aFGF expression is reduced with age or in association with conditions such as Alzheimer’s disease. However, it has been demonstrated that infusion of exogenous aFGF following cortical devascularization prevents subsequent degeneration of nucleus basalis neurons and deficiencies on the Morris water maze task (Figueiredo et al., 1993). This indicates that aFGF is capable of preserving functional as well as chemical and morphological properties of one cholinergic cell group that is particularly susceptible to atrophy in association with pathological aging. Thus, the present data and the available literature are consistent with the idea that aFGF may serve as an autocrine neurotrophic factor for basal forebrain cholinergic neurons, but future studies are needed to determine whether changes in aFGF expression contribute to disease- and age-related neuropathology. ACKNOWLEDGMENTS The authors thank Julie Wong and Zhiquin Sun for their expert technical assistance. This work was supported by 387 NIA grant AGO0538 and by NIMH Research Scientist Development Award MH00974 to C.M.G. J.L.B. was supported by PHS grant MH14599. LITERATURE CITED Allen, S.J., S.H. MacGowan, J.J.S. Treanor, R. Feeney, G.K. Wilcock, and D. Dawborn (19911 Normal p-NGF content in Alzheimer’s disease cerebral cortex and hippocampus. Neurosci. Lett. 131,135-139. Amaral, D.G., and J. Kurz (1985)An analysis of the origins of the cholinergic and noncholinerkic septal projections to the hippocampal formation of the rat. J . Comp. Neurol. 240.37-59. Anderson, K.J., D. Dam, S. Lee, and C.W. Cotman (1988) Basic fibroblast growth factor prevents death of lesioned cholinergic neurons in vivo. Nature 332360-361. Armstrong, D.M., R.D. Terry, R.M. Deteresa, G. Bruce, L.B. Hersh, and F.H. Gage (1987)Response to septal cholinergic neurons to axotomy. J. Comp. Neurol. 264r421-436. Armstrong, D.M., R. Sheffield, G. Buzsaki, K.S. Chen, L.B. Hersh, B. Nearing, and F.H. Gage (19931 Morphologic alterations of choline acetyltransferase-positiveneurons in the basal forebrain of aged behaviorally characterized Fisher 344 rats. Neurobiol. Aging 14,457-470. Asai, T., A. Wanaka, H. Kato, Y. Masana, M. Seo, and M. Tohyama (1993) Differential expression of two members of FGF receptor gene family, FGFR-1 and FGFR-2 mRNA, in the adult rat central nervous system. Mol. Brain Res. 17,174-178. Baird, A,, and M. Klagsbrun (19911 The fibroblast growth factor family. Ann. N.Y. Acad. Sci. 638rxi-xii. Beal, M.F., N.W. Kowall, K.J. Swartz, and R.J. Ferrante (19901Homocysteic acid lesions in rat striatum spare somatostatin-neuropeptide y (NADPHdiaphorase) neurons. Neurosci. Lett. 108:3642. Bean, A.J., R. Elde, Y. Cao, C. Oellig, C. Tamminga, M. Goldstein, R.F. Pettersson, and T. Hokfelt (1991) Expression of acidic and basic fibroblast growth factors in the substantia nigra of rat, monkey, and human. Proc. Natl. Acad. Sci. USA88:10237-10241. Boegman, R.J., J. Cockhill, K. Jhamandas, and R.J. Beninger (1992) Excitotoxic lesions of rat basal forebrain: Differential effects in choline acetyltransferase in the cortex and amygdala. Neuroscience 51,129-135. Burgess, W.H., and T. Maciag (1989) The heparin-binding (fibroblast) growth factor family of proteins. Annu. Rev. Biochem. 58:575-606. Carlsen, T., L. Zaborsky, and L. Heimer (1985)Cholinergic projections from the basal forebrain to the basolateral amygdaloid complex: A combined retrograde fluorescent and immunohistochemical study. J. Comp. Neurol .234: 155- 167. Chesselet, M.F., C. Gonzales, C.S. Lin, K. Polsky, and B.K. Jin (19901 Ischemic damage on the striatum of adult gerbils: Relative sparing of somatostatinergic and cholinergic interneurons contrasts with loss of efferent neurons. Exp. Neurol. 110:209-218. Collerton, D. ( 19861 Cholinergic function and intellectual decline in Alzheimer’s disease. Neuroscience 19:l-28. Conner, J.M., D. Muir, S. Varon, T. Hagg, and M. Manthorpe (1992) The localization of nerve growth factor-like immunoreactivity in the adult rat basal forebrain and hippocampal formation. J. Comp. Neurol. 319,454462. Coyle, J.T., D.L. Price, and M.R. DeLong (19831 Alzheimer’s disease: A disorder of cortical cholinergc innervation. Science 219: 1184-1 190. Crowley, C., S.D. Spencer, M.C. Nishimura, K.S. Chen, S. Pitts-Meek, M.P. Armanini, L.H. Ling, S.B. MacMahon, D.L. Shelton, and A.D. Levinson (1994) Mice lacking nerve growth factor display perinatal loss of sensory and sympathetic neurons yet develop basal forebrain cholinergic neurons. Cell 76:1001-1011. Crutcher, K.A., and J. Weingartner (19911 Hippocampal NGF levels are not reduced in the aged Fisher 344 rat. Neurobiol. Agmg 12r449-454. Cummings, B.J., G.J. Yee, and C.W. Cotman (1992) bFGF promotes the survival of entorhinal layer I1 neurons after perforant path axotomy. Brain Res. 591:271-276. Date, I., M.F.D. Notter, S.Y. Felten, and D.L. Felten (19901 MPTP-treated young mice but not a g n g mice show partial recovery of the nigrostriatal dopaminergc system by stereotaxic injection of acidic fibroblast growth factor (aFGF).Brain Res. 526:156-170. Dimova, R., J. Vuillet, A. Nieoullon, and L. Kerkerian-Le Goff (19931 Ultrastructural features of the choline acetyltransferase-containing neurons and relationships with nigral dopaminergic and cortical afferent pathways in the rat striatum. Neuroscience 53r1059-1071 388 Elde, J.L. BIZON ET AL. R..Y. (ha,A. Cintra. T.C. Brelje, M. Pelto-Huikko, K. Junttila, K. R F. Pettersson, a n d T. Hiikfelt (1991)Prominent expression of Fuxe. acidic fibroblast LTowth factor in motor a n d sensory neurons. Neuron 7::349--9fi4. Emotrr. K..A:M. Gonzilez, P.A. Walicke. E. Wada. D.M. Simmons. S. Shimasaki. and A. Baird ! 19891 Basic fibroblast q o w t h factor IFGFI in the wntr;il nervous system: Identification of specific loci of basic FGF expression in t h e rat brain. Growth Factors 2:21-29. Ernfors. l'.. (:. Wetmore. L. Olson. and H. Persson 11990) Identification of cells in t h c rat brain and peripheral tissues expressing mRNA for meniht~r.:o f t h e nerve p-owth factor family. Neuron .5:511-526. Ferrari. (;., .V.C Minozzi, G. Toffano, A. 1,eon. and S. Skaper (1989) Basic fibrnblmt growth factor promotes t h e survival and t h e development of mttsr~ncephalicneurons in culture. Dev. Biol. 133140-147. Fikweiredrr. I%.(:., P.Piccardo. D. Maysinger. P.B.S. Clarke, and A.C. Cuello I 1998 I Effbcts of acidic fibroblast growth factor on cholinergic neurons of nuc1t:u.i hasalih magmcellularis and in a spatial memory task following corticul d~,vascularization.Neuroscience Fi6:956-963. Fischer. W..F.H. Gage, and A. Bjorklund 11989) Degenerative changes in forebrain cholinerblc nuclei correlatc with cognitive impairments in aged rats. E u r . J . Neurosci. k34-45. Fischer. &'.. K.S. Chen. F.H. Gage. and A. Bjorklund (1991) Progressive declintx in spatial learning and integrity of forebrain cholinergic neurons in rats during a k ~ n gNeurobiol. . A g n g 13:9-23. Gage, F.11.. K. Wictorin, W. Fischer. L.R. Williams, S. Varon, and A . Bjhrklund 11986)Retrograde cell changes in medial septum and diagonal band following fimbria-fornix transection: Quantitative temporal analysis. Ncwroscience 1 9 2 4 1-255. Gall, (:.. and P.J. Isackson ! 1989) Limbic seizures increase neuronal production nf mRNA for nerve growth factor. Science 245,758-760. Gall, C.M , B.Berschauer, and P.J. Isackson 11994)Basic fibroblast growth factor mItNA is increased in forebrain neurons and glia following recurrent limbic seizures. MoI. Brain Res. 2I:190-205. Gallagher. M.. R.D. Burwell, M.H. Kodsi, M. McKinney, S. Southerland, I,. Vella-ltountree, and M.H. Lewis ! 1990) Markers for biogenic amines in the aged rat brain: Relationship to decline in spatial learning ability. Neurohiol. A g n g 11:507-514. Garcia-Rill. E.,D.L. Davies, R.D. Skinner, J.A. Biedermann, and C. McHalffey ! 1991I Fibroblast growth factor-induced increased survival ofcholinerhcc mesopontine neurons in culture. Dev. Brain Res. 60267-270. Gomez-Pinilla. F.. J.W.-K. Lee, and C.W. Cotman 11992) Basic FGF in adult rat brain: Cellular distribution a n d response to entorhinal lesion and tinibria-fornix transection. J . Neurosci. I2:345-3.55. Gonzales. ('.. R.C.Lin, and M.F. Chesselet (1992) Relative sparing of GABAc~gilc interneurnns in t h e striatum nf gerbils with ischemiainduced lesions. Neurosci. Lett. 1.95.53-58. Goodrich. S.P.. G.C. Yan, K. Bahrenburg, and P.E. Mansson (1989) T h e nuclcwtidri sequence of rat heparin binding growth factor 1 (HBGF-1). Nucleic Acids Res. I7:2867. Grothe. c' , If Otto. and K. Unsicker (19891 Basic fibroblast growth factor promotes in vitro survival and cholinergic development of rat septal neuroiis: (:omparisrrn with t h e effect of nerve LTowth factor. Neuroscience %3I:649-661. Iieimer. I... (;.F. Alheid, a n d L. Zaborszky t1985i Basal ganglia. In G. Paxinos !r,dr: The Rat Nervous System: Forebrain and Midbrain. Orlando. Fl.. Academic Press, pp. 37-86. Hellweg, K ,W.Fisher. C. Hock, F.H. Gage, A. B,jBrklund, a n d H. Thoenen ! 19901Nerve Lvowth factor levels and chnline acetyltransferase activity in the hrain of aged rats with spatial memory impairments. Brain Res. 5.37. 12?-l:30. Hyman. C. M.J u h a s z , C. Jackson, P. Wright, N.Y. Ip. and K.M. Lindsay ! 19941 Overlapping a n d distinct actions of the neurotrophins BDNF, N'T-3. and N'T-4'6 o n cultured dnpaminergic and GABAergc neurons of ventral mr,sencephalon. J . Neurosci. 14,335-347. Ibinez. C.F., P. Ernfors, and H. Persson t1991) Development and regional exprc,ssion ofcholine acetyltransferase mKNA in the rat central nervous systrni. d. Neurosci. Res. 29:163-171. ,Jette. N , M.S. (:ole. and M. Fahnestock (19941 N G F mRNA is not decreased in frontal cortex from Alzheimer's di patients. Mol. Brain Res. 25:242-250. Knusel. I3 , P P Michel. J.S. Schwaber, and F.Hefti (1990) Selective and nonselective stimulation of central cholinergic a n d dopaminergic development i n vitro by nerve LTowth factor, basic fibroblast growth factor, epidcrmiil growth factor. insulin and the insulin-like growth factors 1 and 11. .I Neurosci. 10,558-570. Koliatsirs. V.E., H.J.W. Nautd, R.E. Clatterbuck. U.M. Holtzman, W.C. Mobley, and D.1,. Pricc 119901 Mouse nerve LTowth factor prevents degeneration of iutotomized basal forebrain cholinergic neurons in t h e monkey. J . Neurosci. 10:3801-3813. Koliatsos, V.E., D.1,. Price, G . K . Gouras, M . H . Cayouetee, L.E. Burton. and ive effects of nerve ATowth factor. J . W . Winslow I 19941 Highly s rurotrophin-3 on intact and injured brain-derived neu rotrophic fact( basal forebrain magniicellular neurons. J . Comp. Neurnl. 349247-262. Kordower. J.H., M. Hurkr-Watson, J.D. Roback. and B.H. Wainerr (19921 Stability of sept(rhippocanipal neurons following excitotoxic lesions of the rat hippocampus. Exp. Neurol. 11 7:1-16. Kromer, L.F. (19871 Nerve growth prevents neuronal death. Science Kushima, Y.. C. Nishio, 'I' Nonomu nerve growth factor and basic fibroblast growth factur (in survival of cultured septal chnlincrgic neurcrns from adult rats. Brain Res. .598;2fi4270. Lauterhorn, J.. P.J. Isackson. and C. Gall 11991) Nerve LTowth factor mRNA-containing cells art: distributed within reb*ons of cholinergic neurons in t h e rat basal forcbrain. J . Comp. Neurol. 306,439-446. Lauterborn. ,J.(:., P.J. Isackson, R. Mnntalvo, and C.M. Gall (1993a1 In situ hybridization localization ofcholine acetyltransferase mRNA in adult rat brain and spinal cord. Mol. Brain Kes. 17.59-69. Lauterborn. J C . , T. Tran. l'. Isackson. and C.M. Gall l1993bJ NGF niKNA is expressed by GABAerLcc neurons in rat hippocampus. Neuroreport 5:273-276. Lauterborn. J.C.. J.L.. Bizon, ?'.M.D.Tran. and C.M. Gall !I9951NGFmRNA is expressed by GARAergic but not cholinergic neurons in rat basal forebrain. J . Cnmp. Neurol. .?60:454-462. Leonard, S., D. Luthman, J . Logel, J . Luthman. C. Antle. K. Freedman, and B. Iloffer t1993) Acidic and basic fibroblast q o w t h factor mRNAs a r e increased in striatum following MP'I'P-induced dopamine neurofibcr lesion: Assay by quantitative PCH. Mol. Brain Res. 18275-284. Morrison. R.S.. A . Sharma, J . De Vellis, and R.A. Bradshaw (19861 Basic fibroblast, growt.11 factor supports the survival of cerebral cortical neurons in primal? culture. Proc. Natl. Acad. Sci. USA N3:7537-7541 Morse, J.K., S.J. Wiegand. K. Anderson. Y. You, N. Cai, J . Carnahan, J. Miller, P.S. Distefenn, A. Altar, R.M. Lindsay, and R.F. Alderson 11993) Brain-dwivctd neurotrophic factor ( BDNF) prevents the degeneration of medial septal cholinergic neurons following f m b r i a transection. J . Neurosci. 13.4146-4158. Nakata, N., H. Kato, and K. Kokwre (1993) Protective effects of basic fibroblast p o w t h factor against hippocampal neuronal damage following cerebral ischemia in t h c gerbil. Brain Res. 605:354-356. Onn. 'r., H. Saito, 'I'.Kishimoto, T. Okumoto, and K. Miyamoto tl99lI Stimulation of biosynthcsis of nerve growth factor by acidic fibroblast q o w t h factor i n cultured mouse astrocytes. Neurosci. Lett. 126:18-20. Otto, D., M. Frotscher. and K. Unsicker 119891Basic fibroblast growth factor and nerve growth factor administered in gel foam rescue medial septal neurnns aftrtr fimhria fornix transection. J. Neurosci. Res. 2283-91 Paxinos, G., and I,. Butcher I19851 Organizational principles of the hrain a s revealed by chol inc acetyltransfkrase and cholinesterase distribution and prqjections. I n G Paxinos led): T h e Rat Nervous System: Fnrehrain and Midbrain. Orlando, FI,: Academic Press. pp. 487-521. Paxinns, G.. and C Watson ! 19861. The Rat Brain i n Stereotaxic Coordinates. San Diego: Academic Press. Peterson, G.M.. G.W. Lanford, and E.W. Powell 11990)Fate of septohippocampal neurons following fimbria-fornix transection: A time course analysis. Brain 1ir.s. Bull. 25:129-137. Phillips, H.S., J.M. Hains, G.R.Laramee. A. Rosenthal. and J.W. Winslnw ! 1990) Widespread expressinn of BDNF but not NT-3 by targets of basal forebrain cholinergic ncurnns. Science 250;290-294. Pickel, V.M., and J. Chan (1990)Spiny neurons lacking cholin ferase are major targets of cholinerkcc and catecholaminergic terminals in rat striatum. J . Neurnsci. Kes. 2.5:263-280. Schwab, M.E., IT. Otten, Y. Agid, and H. Thoenen (19791 Nerve grnwth factor (NGE'I in the rat CNS: Absence of specific r e t r o p a d c axonal transport and tyrosine hydroxylasc induction in locus coeruleus and substantia nikTa. Brain Kes. 168,473483. Serooky, K.R., K.H. I , u n d q m . T.M.D. 'I'ran, K.M. Gutherie, P.J. Isackson. and C.M. Gall ! 19941 I h p a m i n e r g c neurnns in rat ventral midbrain express brain-dcrived neurotrophic factor and neurotrophin-3 mRNAs. J . Cnmp. Neurol. 542::321-3:34. Smeynti, R.J., R. Klcin, A. Schnapp, and L K . Lnng (19941 Severe senson. and sympathetic neuropatholo@s in mice carrying a disrupted Trk NGF receptor gcnc. Nat ure 3<68:246-249. aFGF EXPRESSION BY CHOLINERGIC NEURONS Sofroniew. M.V., R.C.A. Pearson, and T.P.S. Powell (1987)The cholinergic nuclei of the basal forebrain of the rat: Normal structure, development and experimentally induced degeneration. Brain Res. 411:310-331. Sofroniew, M.V., N.P. Galletly, 0. Isacson, and C.N. Svendsen (1990) Survival of adult basal forebrain cholinergic neurons after loss of target neurons. Science 247r338-342. Sofroniew. M.V., J.D. Cooper, C.N. Svendsen, P. Crossman, N.Y. Ip, R.M. Lindsay, F. Zafra, and D. Lindolm (1993)Atrophy but not death of adult septal cholinergic neurons after ablation of target capacity to produce mRNAs for NGF, BDNF, and NT3. J. Neurosci. 13r5263-5276. Spina, M.B.. S.P. Squinto, J. Miller, R.M. Lindsay, and C. Hyman (1992) Brain-derived neurotrophic factor protects dopamine neurons against 6-hydroxydopamine and N-methyl-4-phenylpyridiniumion toxicity: Involvement of the glutathione system. J. Neurochem. 59r99-106. Stock, A., K. Kuzis, W.R. Woodward, R. Nishi, and F.P. Eckenstein (1992) Localization of acidic fibroblast growth factor in specific subcortical neuronal populations. J. Neurosci. 12r46884700. Stroessor-Johnson, H.M., P.R. Rapp, and D.G. Amaral (1992) Cholinergc cell loss and hypertrophy in the medial septal nucleus of the behaviorally characterized aged rhesus monkey. J . Neurosci. 12r1936-1944. Thompson, L.M., S. Plummer, M. Schalling, M.R. Altherr, J.F. Gusella, D.E. Housman, and J.J. Wasmuth (1991)Agene encodinga fibroblast growth factor receptor isolated from a Huntington disease gene regon of human chromosome 4. Genomics I lr1133-1142. Venero. J.L., B. Knusel, K.D. Beck, and F. Hefti (1994) Expression of neurotrophin and trk receptor genes in adult rats with fimbria transections: Effect of intraventricular nerve growth factor and brain-derived neurotrophic factor administration. Neuroscience 59r797-815. Walickc!. P.A. (1988) Basic and acidic fibroblast growth factors have trophic 389 effects on neurons from multiple CNS regons. J. Neurosci. 8726182627. Whitehouse, P.J., D.L. Price, R.G. Struble, A.W. Clark, J.T. Coyle, and M.R. DeLong (1982)Alzheimer’s disease and senile dimentia: Loss of neurons in basal forebrain. Science 215r1237-1239. Woolf, N.J., and L.L. Butcher (1981) Cholinergic neurons in the caudateputamen complex proper are intrinsically organized: A combined Evans blue and acetylcholinesterase analysis. Brain Res. Bull. 7r487-507. Woolf, N.J., F. Exckstein, and L.L. Butcher (1984) Cholinergic systems in the rat brain: I. Projections to the limbic telencephalon. Brain Res. Bull. 13r751-784. Yazaki, N., Y. Hosoi, K. Kawabata, A. Miyake, M. Minami, M. Satoh, M. Ohta, T. Kawasaki, and N. Itoh (1994) Differential expression patterns of mRNAs for members of the fibroblast growth factor receptor family, FGF1-FGF4, in rat brain. J. Neurosci. Res. 37r445452. Yokoyama, M., R.S. Morrison, I.B. Black, and C.F. Dreyfus (1994) Septa1 neuron cholinergic and GABAergic functions: Differential regulation by basic fibroblast growth factor and epidermal growth factor. Dev. Brain Res. 78:201-209. Yoshida, K., and F.H. Gage (1991)Fibroblast growth factors stimulate nerve growth factor synthesis and secretion by astrocytes. Brain Res. 538: 118126. Yoshida, S., L.P. Lin, Z.L. Chen, K. Momota, K. Kato, T. Tanaka, A. Wanaka, and S. Shiosaka (1994) Basal magnocellular and pontine cholinergc neurons coexpress FGF receptor mRNA. Neurosci. Res. 2Or35-42. Zaborsky, L., J . Carlsen, H.R. Brasher, and L. Heimer (1986) Cholinergc and GABAergc afferents to the olfactory bulb in the rat with special emphasis on the projection neurons in the nucleus of the horizontal limb of the diagonal band. J. Comp. Neural. 243r488-509.