THE JOURNAL OF COMPARATIVE NEUROLOGY 306:439-446 (1991)
Nerve Growth Factor mRNA-Containing Cells Are Distributed Within Regions of Cholinergic Neurons in the Rat Basal Forebrain JULIE C. LAUTERBORN, PAUL J. ISACKSON, AND CHRISTINE M. GALL Department of Anatomy and Neurobiology (J.C.L., P.J.I., C.M.G.) and Department of Biological Chemistry (P.J.I.), University of California, Irvine, California 92717
ABSTRACT It has been proposed that nerve growth factor (NGF) provides critical trophic support for the cholinergic neurons of the basal forebrain and that it becomes available to these neurons by retrograde transport from distant forebrain targets. However, neurochemical studies have detected low levels of NGF mRNA within basal forebrain areas of normal and experimental animals, thus suggesting that some NGF synthesis may actually occur within the region of the responsive cholinergic cells. In the present study with in situ hybridization and immunohistochemical techniques, the distribution of cells containing NGF mRNA within basal forebrain was compared with the distribution of cholinergic perikarya. The localization of NGF mRNA was examined by using a 35S-labeledRNA probe complementary to rat preproNGF mRNA and emulsion autoradiography. Hybridization of the NGF cRNA labeled a large number of cells within the anterior olfactory nucleus and the piriform cortex as well as neurons in a continuous zone spanning the lateral aspects of both the horizontal limb of the diagonal band of Broca and the magnocellular preoptic nucleus. In the latter regions, large autoradiographic grain clusters labeled relatively large Nissl-pale nuclei; it did not appear that glial cells were autoradiographically labeled. Comparison of adjacent tissue sections processed for in situ hybridization to NGF mRNA and immunohistochemical localization of choline acetyltransferase ( C U T ) demonstrated overlapping fields of cRNA-labeled neurons and CUT-immunoreactive perikarya in both the horizontal limb of the diagonal band and magnocellular preoptic regions. However, no hybridization of the cRNA probe was observed in other principal cholinergic regions including the medial septum, the vertical limb of the diagonal band, or the nucleus basalis of Meynert. These results provide evidence for the synthesis of NGF mRNA by neurons within select fields of NGF-responsive cholinergic cells and suggest that the generally accepted view of "distant" target-derived neurotrophic support should be reconsidered and broadened. Key words: NGF, diagonal band of Broca, piriform, choline acetyltransferase, in situ hybridization
The basal forebrain cholinergicsystem of the rat consists of acetylcholine-synthesizing neurons distributed across several distinct areas including the medial septal nucleus, the vertical and horizontal limbs of the diagonal band of Broca, the magnocellular preoptic area, substantia innominata, the nucleus basalis of Meynert, and the nucleus of the ansa lenticularis (Paxinos and Butcher, '85 for review). Nerve growth factor (NGF) influences the synthetic activities of these cholinergic neurons and may also provide trophic support that is critical for their survival. Thus the application of exogenous NGF increases both the activity of choline acetyltransferase (ChAT), the enzyme that catalyzes the formation of acetylcholine, (Gnahn et al., '83; Mobley et al., '86; Williams and Rylett, '90) and the levels of o 1991 WILEY-LISS, INC.
mRNAs for ChAT and NGF receptor in basal forebrain neurons (Higgins et al., '89). High-affinity choline uptake and acetylcholine release in forebrain are also enhanced by chronic treatment with NGF (Rylett and Williams, '90; Williams and Rylett, '90). Moreover, infusion of NGF prevents the loss of cholinergic cells that occurs when their projections t o hippocampus are transected (Hefti, '86; Williams et al., '86; Kromer, '87). Other lines of evidence suggest that the basal forebrain cholinergic system is specifically dependent upon distant Accepted December 5,1990. Address reprint requests to Christine Gall, Dept. of Anatomy and Neurobiology, University of California, Irvine, CA 92717.
J.C. LAUTERBORN ET AL.
440
cortical targets as the source of NGF. Cholinergic neurons of the medial septal nucleus and the diagonal band of Broca project predominantly to hippocampus and olfactory forebrain regions (Woolf et al., '84; Zaborszky et al., '86), whereas neurons of the nucleus basalis of Meynert project to neocortex (Paxinos and Butcher, '85 for review). These target fields contain the highest levels of NGF mRNA in the brain (Korsching et al., '85; Shelton and Reichardt, '86; Whittemore et al., '86). The basal forebrain itself contains relatively low levels of NGF mRNA but one of the higher regional levels of NGF protein, thus suggesting that local NGF protein is actually synthesized in regions such as the hippocampus and transported into the basal forebrain (Korsching et al., '85; Whittemore and Seiger, '87 for review). This idea is supported by the observations that exogenous NGF is retrogradely transported from the hippocampus (Schwab et al., '79) and neocortex (Seiler and Schwab, '84) by the basal forebrain cholinergic neurons and that fimbrial transection leads to a build-up of NGF within the hippocampus (Gasser et al., '86; Weskamp et al., '86; Larkfors et al., '87). However, the hypothesis of cortically derived neurotrophic support has recently been questioned by Sofroniew et al. ('901, who showed that uninjured cholinergic neurons of the medial septum atrophy but do not die after their cortical target, the hippocampus, is ablated by the excitotoxin N-methyl-D-aspartate. This observation suggests that if cholinergic cells are critically dependent upon the trophic support of NGF, then there should be other sources of this substance for basal forebrain NGF-responsive neurons. As noted above, there are low levels of NGF mRNA within the basal forebrain itself. Moreover, using the two-site enzyme immunoassay technique, Lorez et al. ('88) found increases in NGF-immunoreactivity within basal forebrain areas following ablation of cortical targets, thereby providing further evidence for local synthesis. These authors suggested that increased NGF immunoreactivity in the lesioned rats may have been due to increased NGF production by glial cells. This postulate is consistent with the evidence that cultured brain astrocytes can synthesize and release NGF (Furukawa et al., '86, '87; Yamakuni et al., '87; Spranger et al., '90) and thus could be an additional source for this trophic factor within the brain. However, recent in situ hybridization studies in several laboratories have not detected NGF mRNA within adult forebrain glia (Rennert and Heinrich, '86; Ayer-LeLievre et al., '88; Whittemore et al., '88; Gall and Isackson, '89). In the present study, in situ hybridization techniques were used to determine the cellular localization of NGF mRNA in the basal forebrain. We found that neurons within several basal forebrain regions express NGF mRNA and that NGF mRNA-containing cells are distributed within fields containing cholinergic neurons. Some of these results have appeared in preliminary form (Lauterborn et al.,'89).
MATERIALS AND METHODS
histochemistry (n = 5) remained in these sohtiOnS for 2 and 24 hours, respectively. In all instances, the whole forebrain was sectioned in the coronal plane at a thickness of 20 pm with a freezing microtome. Sections were collected into cold 4% paraformaldehyde in PB for in situ hybridization or cold PB for immunohistochemistry and acetylcholinesterase histochemistry.
In situ hybridization Tissue sections were processed for the in situ hybridization localization of NGF mRNA using a 35S-labeled970-base RNA probe including a span complementary to 750 bases of the coding region of rat preproNGF (Whittemoreet al., '88). The "anti-sense'' and "sense" RNAs were transcribed in the presence of 35S-UTP from a rat genomic subclone, pBSrNGF, linearized with PvuII, by using T3 and T7 RNA polymerase, respectively. Free-floating tissue sections were transferred sequentially through 0.1 M glycine in PB, proteinase K (1pg/ml) in 0.1 M Tris buffer (pH 8.0) with 50 mM EDTA for 30 minutes at 37"C,0.25% acetic anhydride in 0.1 M triethanolamine (10 min), 2 x saline sodium citrate (SSC) (0.15 M NaCl and 0.015 M NaCitrate, pH 7.0) (30 min), and then incubated in a hybridization solution containing 50% formamide, 10% dextran sulfate, 0.7% ficoll, 0.7% polyvinyl pyrrolidone, 7 mgiml bovine serum albumin, 0.15 mg/ml yeast transfer RNA, 0.33 mg/ml denatured herring sperm DNA, and 20 p M dithiothreitol (DTT) for 1 hour at 60°C. From this prehybridization incubation, the sections were transferred into fresh hybridization solution containing the NGF cRNA probe at a density of 1 x lo6 cpm/lOO p1 with an additional 20 pM DTT and incubated for 30-40 hours at 60°C. Following hybridization, the sections were transferred through 4 x SSC (1hr), ribonuclease A (20 pg/ml) in 10 mM Tris-saline (pH 8.0) with 1mM EDTA for 30 minutes at 45"C,and then descending concentrations of SSC to a final stringency of 0.1 x SSC at 60°C for 1hour. Following an overnight wash in 0.1 x SSC at room temperature, the sections were mounted onto gelatin-coated slides and air dried. The distribution of cRNA hybridization was evaluated with emulsion autoradiography (Kodak NTBZ nuclear track emulsion 1:1 with H,O) and exposures of 4-6 weeks at 4°C. After development of the autoradiograms, the sections were stained with cresyl violet and coverslipped with Permount. Controls for specificity of hybridization included either treatment of sections with ribonuclease A (20 pg/ml) prior to normal hybridization, or hybridization with the labeled "sense" RNA sequence (described above). In neither instance was cellular labeling observed. In addition, the regional distribution of hybridization was consistent with the reported distribution of NGF mRNA (Korsching et al., '85; Shelton and Reichardt, '86; Whittemore et al., '86) in that the greatest number of labeled cells were found in hippocampus and no detectable labeling was found in cerebellum. Finally, as further evidence for specificity, the pattern of hybridization obtained with the present rat preproNGF cRNA was distinct from that obtained with other "S-labeled cRNAs (complementary to mRNAs for brain derived neurotrophic factor (BDNF), Fos, proenkephalin A, and proneuropeptide Y) in all areas examined.
Normal adult male Sprague-Dawley rats (250-300 gms; Simonsen Labs) were used. Animals were overdosed with sodium pentobarbital and intracardially perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) (PB). Choline acetyltransferase Following perfusion, the brains to be processed for in situ immunohistochemistry and ACHE hybridization alone (n = 10)were postfixed in 4% paraformhistochemistry aldehyde in PB for 24 hours and then transferred to 20% sucrose in 4%paraformaldehyde at 4°C for 48 hours; brains In five cases, a spaced series of tissue sections, adjacent to to be processed for both in situ hybridization and immuno- those used for in situ hybridization, were processed for the
NGF mRNA-CONTAINING NEURONS IN BASAL FOREBRAIN localization of choline acetyltransferase (ChAT) according to the avidin-biotin immunohistochemical technique and with 0.1 M Tris buffer (pH 7.4) for rinses and antisera dilution. Sections were incubated for 24 hours at 4°C in either rabbit-anti-ChAT (1:600; Chemicon) or rat-antiC U T (1:20; Boehringer Mannheim) with 0.3% Triton-X 100 and 1% normal goat serum (NGS). The sections were then washed for 30 minutes and incubated in either biotinylated antirabbit IgG or antirat IgG (1:200) containing 0.3% Triton-X 100 and 1% NGS for 1 hour at room temperature, Following a second buffer wash, the sections were transferred to the avidin-biotin complex solution (1:100, Vector Labs.) with 0.3% Triton-X for 1hour at room temperature. After a final wash, the sections were incubated in a saturated solution of 3,3'-diaminobenzidine with 0.01% H,O, in PB, rinsed in PB, mounted onto glass slides, dehydrated and cleared through Histoclear, and coverslipped. An additional series of sections adjacent to those processed for in situ hybridization was mounted onto glass slides and stained with cresyl violet. A final series of sections, adjacent to the ChAT immunohistochemical series, was processed for the histochemical localization of acetyl cholinesterase (Koelle and Friedenwald, '49) and used for further identification of cytoarchitectonic areas. Nomenclature is that of Paxinos and Watson ('86).
441
the lateral septal nucleus where hybridization-labeled neurons were scattered just deep to the islands of Calleja and within the more dorsal ventral pallidal region. Large numbers of labeled neurons were observed in several forebrain regions in sections taken through the level of the body of the septum (Fig. lC,D). Although hybridization was not observed within the medial septal nucleus itself, labeled cells were distributed within the lateral aspect of the horizontal limb of the diagonal band of Broca and extending lateral to this field into what might be considered the deeper ventral pallidum. In progressively more caudal sections, this cluster of labeled cells extended into the lateral aspect of the magnocellular preoptic area (Fig. 2B,E). In addition, a small number of cells situated between the magnocellular preoptic area and the more dorsal anterior commissure were autoradiographically labeled. At the most caudal aspect of the magnocellular preoptic nucleus, only a few labeled neurons were present. No hybridization to NGF mRNA was seen within the more caudal lateral hypothalamic area or within the more dorsal nucleus basalis of Meynert. Comparison of adjacent tissue sections processed for in situ hybridization and the immunohistochemical localization of ChAT demonstrated that in the regions of both the horizontal limb of the diagonal band and the magnocellular preoptic area there was a substantial number of autoradiographically labeled cells in the same areas that contain Analysis of cell size C U T - I R perikarya (Fig. 1). However, these cell groups The sizes of autoradiographically labeled cells and ChAT- were not fully coextensive. Hybridization-labeled cells were immunoreactive (ChAT-IR) cells within the caudal part of most numerous in the lateral aspects of, and extending the horizontal limb of the diagonal band and the magnocel- lateral to, each of these cytoarchitectonic fields, whereas lular preoptic area were measured in two animals. Measure- ChAT-IR cells were distributed throughout each region. In ments were made on a Leitz microscope under brightfield the horizontal limb of the diagonal band in particular, illumination at 4 0 magnification ~ with the aid of a cali- hybridization-labeled cells generally were distributed latbrated eyepiece reticule. Major and minor diameters ( f stan- eral to the very large C U T - I R cells but overlapped the field dard deviation (S.D.)) were measured for the somata of 50 of smaller ChAT-IR cells (Fig. 2). Beyond the horizontal ChAT-IR cells and for both the Nissl-stained nucleus and limb of the diagonal band and the magnocellular preoptic region of highest silver grain density for 50 autoradiograph- area, a smaller number of hybridization-labeled cells were ically labeled cells. The latter measure of the autoradio- observed in regions that contained ChAT-IR perikarya just graphic grain cluster is admittedly subjective, but one could dorsal to these nuclei as well as within the more rostral clearly identify a central region of higher grain density olfactory tubercle and dorsal to the islands of Calleja. (centered over the nucleus) surrounded by a less dense It should be noted that in regions containing both cell scatter of grains. The diameters of the core regions were groups, ChAT-IR perikarya were far more numerous than measured as the only estimate available of the size of the hybridization-labeled cells (Fig. 3). Moreover, as indicated radiolabeled perikarya. above, hybridization-labeled cells were not observed in a number of the principal regions containing CUT-IR neurons. Specifically, in agreement with numerous published RESULTS descriptions, ChAT-IR neurons were observed within the Neurons labeled by hybridization of the 35S-cRNAprobe medial septal nucleus, the vertical limbs of the diagonal were broadly distributed throughout the basal forebrain band, substantia innominata, and the nucleus basalis of extending rostrocaudally from the rostral aspect of the Meynert. Hybridization-labeled cells were not detected in anterior olfactory nucleus through the magnocellular pre- these areas. Within the horizontal limb of the diagonal band and optic area. As indicated in Figure 1, hybridization to NGF mRNA was also prominent in superficial layers of the magnocellular preoptic area, autoradiographic grains were piriform and cingulate cortices and within the claustrum at seen to overlie large pale Nissl-stained nuclei with mean these levels. A detailed description of the distribution of major and minor nuclear diameters of 14.00 2 1.82 (S.D.) hybridization in these and other olfactory and limbic struc- by 11.05 f 1.76 pm (Fig. 4). The average size of the major tures will be presented in a separate report (Lauterborn et density of autoradiographic grains overlying individual labeled neurons measured 27.50 ? 3.88 by 19.45 2 4.41 al., in preparation). At the most rostral planes, hybridization of the cRNA pm, further indicating that the labeled cells are relatively probe labeled cells within the anterior olfactory nucleus large. Smaller cells stained more densely with the Nissl (AON) and within the posterior aspect of the AON lying stain were never seen to be autoradiographically labeled. In between the anterior commissure and the olfactory tuber- contrast to the autoradiographically labeled cells, C U T - I R cle. Occasional labeled neurons were also seen within the cells within the overlapping areas measured on average deeper layers of the olfactory tubercle. The latter cells were 19.40 5 3.77 by 12.25 2 2.38 pm. By comparison, the more numerous in sections cut through the rostral limit of cholinergic cells lying within the medial aspect of the
J.C. LAUTERBORN ET AL.
442
A
B
Fig. 1. Line drawings of a rostral (A) to caudal (F) series of coronal sections through rat forebrain illustrating the distribution of cells labeled by in situ hybridization of the "S-labeled preproNGF cRNA probe and CUT-IR neuronal perikarya (solid circles each indicate 1-5 autoradiographically labeled cells; open triangles and open circles indicate 1-5 and &lo ChAT-IR cells, respectively). Autoradiographically labeled cells and CUT-IR cells are both distributed within (and lateral to) the horizontal limb of the diagonal band (HDB) (C) and within the magnocellular preoptic area (MCP) (C-F), whereas hybridization to NGF mRNA is not observed in fields of ChAT-I cells in the
medial septa1 nucleus (MSN) (C), nucleus basalis of Meynert (B) (E,F), or substantia innominata (SI) (El. Cross-hatching represents relatively diffuse autoradiographic labeling in the claustrum (CI), piriform cortex (PC) (A-F) and cingulate cortex (Cg). The asterisk in F denotes the basolateral nucleus of the amygdaloid complex, which also contains hybridization-labeled cells. Abbreviations: ac, anterior commissure; AP,posterior aspect of the anterior olfactory nucleus; CPu, caudate/ putamen; En, endopiriform nucleus; GP, globus pallidus; ICj, islands of Calleja; ox, optic chiasm; Tu, olfactory tubercle.
NGF mRNA-CONTAINING NEURONS IN BASAL FOREBRAIN
443
Fig. 2. Low magnification photomicrographs of two series of adjacent coronal sections (A-C and D-F) through the basal forebrain of normal adult rats showing a Nissl stain (A, D), autoradiographic localization of in situ hybridization of the 35S-cRNA(B,E; darkfield illumination), ChAT immunoreactivity (C), and acetylcholinesterase (AChE) histochemistry (F). In A-C, cellular hybridization of NGF cRNA (seen as white dots in B) is observed within the field of ChAT-IR neurons in the magnocellular preoptic area (solid arrow indicates a blood vessel in this cytoarchitectonic area in A, B, C). Note the zone of
hybridization within the lateral aspect of panel B overlies layers I1 and 111 of piriform cortex (PC). In E, NGF cRNA-labeled cells are seen within the horizontal limb of the diagonal band (HDB) and the field deep to the olfactory tubercle (Tu). In comparing E and F, one can see that the autoradiographically labeled cells lie in a field of dense AChEhistochemistry (F). Open arrow indicates one blood vessel present in each series. Abbreviations: lot, lateral olfactory tubercle. Bar = 900 pm for A-C; 700 pm for D-F.
horizontal limb of the diagonal band (which does not contain NGF cRNA-labeled neurons) appeared to be larger.
basal forebrain including, most notably, two major regions of cholinergic neurons, the horizontal limb of the diagonal band of Broca and the magnocellular preoptic area. Outside these areas, NGF synthesizing cells were found in the anterior olfactory nucleus, piriform cortex, olfactory tubercle, and aspects of the ventral pallidum; in several of these regions, NGF mRNA-containing cells were distributed
DISCUSSION The data presented here indicate that NGF mRNA is present in neurons distributed across broad fields of the
J.C. LAUTERBORN ET AL.
444
Fig. 3. Photomicrographs of paired adjacent sections (A,B; C,D) through different levels of the magnocellular preoptic area showing the autoradiographic localization of hybridization of the "S-cRNA probe (A,C; darkfield illumination) and ChAT-I (B,D). Open arrows indicate blood vessels, which appear in each of the adjacent sections. Solid arrows in A and C indicate individual hybridization-labeled neurons that can be seen to lie within the field of ChAT-I perikarya. Bar = 340 wm for A, B; 300 pm for C,D.
among cholinergic neurons. Cells containing NGF mRNA, however, were not present in all fields of cholinergic cells. Thus the medial septum/vertical limb of the diagonal band and the nucleus basalis of Meynert, which project primarily to hippocampus and neocortex, respectively, do not contain NGF synthesizing neurons. It has been suggested that glial cells, which synthesize NGF while in culture, might be the local source of NGF in the basal forebrain (Lorez et al., '88). This would accord with the observation that synthesis of NGF in the peripheral nervous system is by non-neuronal cell types including Schwann cells (Bandtlow et al., '87). However, previous in situ hybridization studies strongly suggest that the synthesis of NGF in hippocampus is by neurons (Rennert and Heinrich, '86;Ayer-LeLievre et al., '88; Whittemore et al., '88; Gall and Isackson, '€491, and the present data point to a similar conclusion for the basal forebrain. In particular, labeled cells of the horizontal limb of the diagonal band and the magnocellular preoptic area have large Nissl-pale nuclei indicating that they are most probably neurons. Conversely, smaller more darkly Nissl-stained nuclei, characteristic of some glial populations, were never labeled in our
material. Moreover, hybridization to NGF mRNA labeled large numbers of cells in layers I1 and 111of piriform cortex and it is known that the great majority of cells in these zones are neurons (Switzer et al., '85). The presence of both NGF mRNA-containing neurons and cholinergic cells in certain basal forebrain regions raises the possibility that the growth factor and transmitter are colocalized within individual neurons. Unfortunately, we have not found immunohistochemistry for ChAT to be compatible with the in situ hybridization technique used here and, therefore, have not as yet directly'tested the possibility of colocalization. However, in regard to this point it is noteworthy that the clusters of autoradiographic grains labeling individual basal forebrain neurons had a mean core aggregate size of 27.5 km (major diam.) suggesting NGF mRNA localization in larger neurons than the average 19.4 km diam. ChAT-IR cells in the coextensive field. Moreover, as described here, there are fewer NGF mRNA-containing cells than ChAT-IR cells in these areas and although these populations occupy overlapping fields, their distributions are not fully coextensive. Thus the hybridization-labeled and C U T - I R cell groups are not
NGF mRNA-CONTAINING NEURONS IN BASAL FOREBRAIN
445
specifically, whether they can be sustained by local NGF synthesis following such lesions. Some suggestion that this may be the case is provided by the observation that cholinergic neurons in the medial aspects of the globus pallidus, an area adjacent to the lateral basal forebrain as designated here, are not eliminated by lesions to their target areas in the cingulate cortex (Springer et al., '87). Additional experiments of this type should serve to test the hypothesis that the cholinergic basal forebrain can be differentiated into zones dependent upon NGF of local versus distant origin. One can speculate as to the functional consequences of the arrangements of trophic dependency proposed above. The dependence of some populations of neurons upon locally produced trophic factor might allow for the development and maintenance of "autonomous" regions in which such features as process elaboration and cell survival would be less affected by events occurring in the distant efferent targets (e.g., activity dependent changes in NGF expression) (Gall and Isackson, '89). Dependencies of the type seemingly present in the septohippocampal system would, in contrast, allow influences from the target region t o regulate the neurochemical differentiation and morphologiFig. 4. Brightfield photomicrograph showing t h e autoradiographic cal elaboration of cholinergic afferent neurons. In all, we localization of hybridization of t h e %-cRNA probe (seen as black grains) associated with neurons in t h e magnocellular preoptic area. T h e would expect that the extent to which neurons in a region tissue section was counterstained with cresyl violet to show cell nuclei. are interconnected with their local neighbors might dictate Note that t h e labeled cells (arrows) have relatively large Nissl-pale the degree to which local trophic influences predominate. nuclei in comparison t o t h e surrounding cells. Bar = 50 pm. We do not know the circuit relationships of the NGF producing neurons within the basal forebrain described here. In particular, more detailed information on the identical, but further work is needed to determine whether afferents t o these neurons is needed before we can formuNGF mRNA is localized in a subpopulation of cholinergic late informed hypotheses as to the trophic consequence of cells or within an entirely separate group of non-cholinergic NGF synthesis by these cells. neurons. Previous biochemical studes have shown that dissected ACKNOWLEDGMENTS samples of the septum plus the vertical limb of the diagonal band (Korsching et al., '851, or of this region together with We thank Gary Lynch for comments on the manuscript adjacent areas (Large et al., '86; Shelton and Reichardt, and Scott Whittemore for generously providing the rat '86), contain barely detectable levels of NGF mRNA. Thus NGF clone. This work was supported by NINDS grant it is generally assumed that the NGF protein found in these NS26748 to C.M.G. and NS24747 and a grant from Alzmidline areas arises from retrograde transport from hippo- heimer's Disease and Related Disorders Association to campus and other forebrain targets. The present results are P.J.I. in accord with this hypothesis as it relates to the septal region in particular, but strongly suggest that it should not LITERATURE CITED be generalized to the basal forebrain cholinergic system in its entirety, The lateral aspects of this system, found here to Ayer-LeLiewe, C., L. Olson, T. Ebendal, A. Seiger, and H. Persson (1988) Expression of the p-nerve growth factor gene in hippocampal neurons. contain neurons that express NGF mRNA, had not been Science 240: 1339-1341. specifically evaluated for NGF mRNA content in the earlier subfield dissection studies. In distinction from the more Bandtlow, C.E., R. Heumann, M.E., Schwab, and H. Thoenen (1987) Cellular localization of nerve growth factor synthesisby in situ hyhridizamedial fields and from nucleus basalis of Meynert, the tion. EMBO J. 6:891-899. ascending projections of the horizontal limb of the diagonal Furukawa, S., Y. Furukawa, E. Satayoshi, and K. Hayashi (1986)Synthesis band and the magnocellular preoptic area primarily innerand secretion of nerve growth factor by mouse astroglial cells in culture. Biochem. Biophys. Res. Comm. 13657-63. vate the piriform cortex; the horizontal limb also sends a moderate projection to hippocampus (Woolf et al., '84). Furukawa, S., Y. Furukawa, E. Satayoshi, and K. Hayashi (1987)Synthesis/ secretion of nerve growth factor is associated with cell growth in cultured Although numerous studies have evaluated the dependence mouse astroglial cells. Biochem. Biophys. Res. Comm. 142395-402. of septal neurons on the trophic support provided by distant Gage, F.H., D.M. Armstrong, L.R. Williams, and S.Varon (1988) Morphologtargets in lesion studies (Hefti, '86; Williams et al., '86; ical response of axotomized septal neurons to nerve growth factor. Kromer, '87; Springer et al., '87; Gage et al., '86, '881, J. Comp. Neurol. 289:147-155. considerably less attention has been directed toward this Gage, F.H., K. Wictorin, W. Fischer, L.R. Williams, S. Varon, and A. Bjorklund (1986) Retrograde cell changes in medial septum and issue in regard to these lateral cholinergic groups. As seen diagonal band following fimbria-fornix transection: Quantitative tempohere, the principal forebrain target for these neurons, the ral analysis. Neurosci. 19:241-255. piriform cortex, does contain relatively high levels of NGF C.M., and P.J. Isackson (1989) Limbic seizures increase neuronal mRNA and thus is presumably a major center for NGF Gall,production of messenger RNA for nerve-growth factor. Science245:758synthesis. However, it remains for future work to deter761. mine if the more lateral cholinergic cell groups can survive Gasser, U.E., G. Weskamp, U. Otten, and A.R. Dravid (1986)Time course of the elevation of nerve growth factor (NGF) content in the hippocampus the removal of such potential distant sources of NGF and,
446 and septum following lesions of the septohippocampal pathway in rats. Brain Res. 376.351-356. Gnahn, H., F. Hefti, R. Heumann, M.E. Schwab, and H. Thoenen (1983) NGF-mediated increase of choline acetyltransferase ( C U T ) in the neonatal rat forebrain: Evidence for a physiological role of NGF in the brain? Dev. Brain Res. 9:45-52. Higgins, G.A., K., Sookyong, K.S. Chen, and F.H. Gage (1989) NGF induction of NGF receptor gene expression and cholinergic neuronal hypertrophy within the basal forebrain of the adult rat. Neuron 3247256. Hefti, F. (1986) Nerve growth factor (NGF) promotes survival of septa1 cholinergic neurons after fimbrial transection. J. Neurosci. 6:2155-2162. Koelle, G.B., and J.S. Friedenwald (1949) A histochemical method for localizing cholinesterase activity. Proc. SOC. Exp. Biol. Med. 70;617-622. Korsching, S., G. Auburger, R., Heumann, J. Scott, and H. Thoenen (1985) Levels of nerve growth factor and its mRNA in the central nervous system of the rat correlate with cholinergic innervation. The EMBO J. 4:1389-1393. Kromer, L.F. (1987) Nerve growth factor treatment after brain injury prevents neuronal death. Science 235.214-216. Large, T.H., S.C. Bodary, D.O. Clegg, G. Weskamp, U. Otten, and L.F. Reichardt (1986) Nerve growth factor gene expression in the developing rat brain. Science 234:352-355. Lkkfors, L., I. Stromberg, T. Ebendal, and L. Olson (1987) Nerve growth factor protein level increases in the adult rat hippocampus after specific cholinergic lesion. J. Neurosci. Res. 18;525-531. Lauterborn, J.C., P.J. Isackson, and C.M. Gall (1989) Cellular localization of Neurosci. Abst. 15.864. p-NGF mRNA in rat forebrain. SOC. Lorez, H.P., M. von Frankenberg, G. Weskamp, and U. Otten (1988) Effect of bilateral decortication on nerve growth factor content in the basal nucleus and neostriatum of adult rat. Brain Res. 454:355-360. Mobley, W.C., J.L. Rutkowoski, G.I. Tennekoon, J. Gemski, K. Buchanan, and M.V. Johnson (1986) Nerve growth factor increases choline acetyltransferase activity in developing basal forebrain neurons. Mol. Brain Res. 1.53-62. Paxinos, G., and L.L. Butcher (1985)Organizational principles of the brain as revealed by choline acetyltransferase and acetylcholinesterase distribution and projections. In G. Paxinos (ed): The Rat Nervous System, Vol. 1.Orlando: Academic Press, pp. 487-508. Paxinos, G., and C. Watson (1986)The Rat Brain in Stereotaxic Coordinates. Orlando: Academic Press. Rennert, P.D., and G . Heinrich (1986) Nerve growth factor mRNAin brain: Localization by in situ hybridization. Biochem. Biophys. Res. Comm. 138:813-815. Rylett, R.J., and L.R. Williams (1990) Acetylcholine release following continuous intracerebral administration of nerve growth factor in adult Neurosci. Abst. 16:481. and aged Fisher 344 rats. SOC. Schwab, M.E., U. Otten, Y. Agid, and H. Thoenen (1979) Nerve growth factor (NGF) in the rat CNS: Absence of specific retrograde axonal transport and tyrosine hydroxylase induction in locus coeruieus and substantia nigra. Brain Res. 168;473-483. Seiler, M., and M.E. Schwab (1984) Specific retrograde transport of nerve
J.C. LAUTERBORN ET AL. growth factor (NGF) from neocortex to nucleus basalis in the rat. Brain Res. 300:33-39. Shelton, D.L., and L.F. Reichardt (1986) Studies on the expression of the p nerve growth factor (NGF) gene in the central nervous system: Level and regional distribution of NGF mRNA suggest that NGF functions as a trophic factor for several distinct populations of neurons. Proc. Natl. Acad. Sci. USA83:2714-2718, 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 247;338-342. Spranger, M., D. Lindholm, C. Bandtlow, R. Heumann, H. Gnahn, M. Niiher-NoB and H. Thoenen (1990) Regulation of nerve growth factor (NGF) synthesis in the rat central nervous system: Comparison between the effects of interleukin-1 and various growth factors in astrocyte cultures and in vivo. Europ. J. Neurosci. 269-76. Springer, J.E., S. Koh, M.W. Tayrien, and R. Loy (1987) Basal forebrain magnocellular neurons stain for nerve growth factor: Correlation with cholinergic cell bodies and effects of axotomy. J. Neurosci. Res. I7:111118. Switzer, R.C., J. De Olmos, and L. Heimer (1985) Olfactory system. In G. Paxinos (ed): The Rat Nervous System, Vol. 1. Orlando: Academic Press, pp. 1-36. Weskamp, G., U.E. Gasser, A.R. Dravid, and U. Otten (1986) Fimbria-fornix lesion increases nerve growth factor content in adult rat septum and hippocampus. Neurosci. Lett. 7:121-126. Whittemore, S.R., and A. Seiger (1987) The expression, localization and functional significance of p-nerve growth factor in the central nervous system. Brain Res. Rev. 12:439464. Whittemore, S.R., T. Ebendal, L. Lakfors, L. Olson, A. Seiger, I. Stromberg, and H. Persson (1986) Developmental and regional expression of 0 nerve growth factor messenger RNA and protein in the rat central nervous system. Proc. Nat. Acad. Sci. USA 83:817-821. Whittemore, S.R., P.L. Friedman, D. Larhammar, H. Persson, M. GonzalesCarvajal, and V.R. Holets (1988)Rat p-nerve growth factor sequence and site of synthesis in the adult hippocampus. J. Neurosci. Res. 20:403-410. Williams, L.R., and R.J. Rylett (1990) Exogenous nerve growth factor increases the activity of high-affinity choline uptake and choline acetyltransferase in brain of Fisher 344 male rats. J. Neurochem. 55:10421049. Williams, L.R., S. Varon, G.M. Peterson, K. Wictorin, W. Fischer, A. Bjorklund, and F.H. Gage (1986) Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbria fornix transection. Proc. Natl. Acad. Sci. USA 83:9231-9235. Woolf, N.J., F. Eckenstein, and L.L. Butcher (1984) Cholinergic systems in the rat brain: I. Projections to the limbic telencephalon. Brain Res. Bull. 13:751-784. Yamakuni, T., F. Ozawa, F. Hishinuma, R. Kuwano, Y. Takahashi, and T. Amano (1987) Expression of beta-nerve growth factor mRNA in rat glioma cells and astrocytes from rat brain. FEBS Lett. 223:117-121. Zaborszky, L., J. Carlsen, H.R. Brasher, and L. Heimer (1986) Cholinergic and GABAergic f i e r e n t s 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. Neurol. 243:488-509.
Nerve Growth Factor mRNA-Containing Cells Are Distributed Within Regions of Cholinergic Neurons in the Rat Basal Forebrain JULIE C. LAUTERBORN, PAUL J. ISACKSON, AND CHRISTINE M. GALL Department of Anatomy and Neurobiology (J.C.L., P.J.I., C.M.G.) and Department of Biological Chemistry (P.J.I.), University of California, Irvine, California 92717
ABSTRACT It has been proposed that nerve growth factor (NGF) provides critical trophic support for the cholinergic neurons of the basal forebrain and that it becomes available to these neurons by retrograde transport from distant forebrain targets. However, neurochemical studies have detected low levels of NGF mRNA within basal forebrain areas of normal and experimental animals, thus suggesting that some NGF synthesis may actually occur within the region of the responsive cholinergic cells. In the present study with in situ hybridization and immunohistochemical techniques, the distribution of cells containing NGF mRNA within basal forebrain was compared with the distribution of cholinergic perikarya. The localization of NGF mRNA was examined by using a 35S-labeledRNA probe complementary to rat preproNGF mRNA and emulsion autoradiography. Hybridization of the NGF cRNA labeled a large number of cells within the anterior olfactory nucleus and the piriform cortex as well as neurons in a continuous zone spanning the lateral aspects of both the horizontal limb of the diagonal band of Broca and the magnocellular preoptic nucleus. In the latter regions, large autoradiographic grain clusters labeled relatively large Nissl-pale nuclei; it did not appear that glial cells were autoradiographically labeled. Comparison of adjacent tissue sections processed for in situ hybridization to NGF mRNA and immunohistochemical localization of choline acetyltransferase ( C U T ) demonstrated overlapping fields of cRNA-labeled neurons and CUT-immunoreactive perikarya in both the horizontal limb of the diagonal band and magnocellular preoptic regions. However, no hybridization of the cRNA probe was observed in other principal cholinergic regions including the medial septum, the vertical limb of the diagonal band, or the nucleus basalis of Meynert. These results provide evidence for the synthesis of NGF mRNA by neurons within select fields of NGF-responsive cholinergic cells and suggest that the generally accepted view of "distant" target-derived neurotrophic support should be reconsidered and broadened. Key words: NGF, diagonal band of Broca, piriform, choline acetyltransferase, in situ hybridization
The basal forebrain cholinergicsystem of the rat consists of acetylcholine-synthesizing neurons distributed across several distinct areas including the medial septal nucleus, the vertical and horizontal limbs of the diagonal band of Broca, the magnocellular preoptic area, substantia innominata, the nucleus basalis of Meynert, and the nucleus of the ansa lenticularis (Paxinos and Butcher, '85 for review). Nerve growth factor (NGF) influences the synthetic activities of these cholinergic neurons and may also provide trophic support that is critical for their survival. Thus the application of exogenous NGF increases both the activity of choline acetyltransferase (ChAT), the enzyme that catalyzes the formation of acetylcholine, (Gnahn et al., '83; Mobley et al., '86; Williams and Rylett, '90) and the levels of o 1991 WILEY-LISS, INC.
mRNAs for ChAT and NGF receptor in basal forebrain neurons (Higgins et al., '89). High-affinity choline uptake and acetylcholine release in forebrain are also enhanced by chronic treatment with NGF (Rylett and Williams, '90; Williams and Rylett, '90). Moreover, infusion of NGF prevents the loss of cholinergic cells that occurs when their projections t o hippocampus are transected (Hefti, '86; Williams et al., '86; Kromer, '87). Other lines of evidence suggest that the basal forebrain cholinergic system is specifically dependent upon distant Accepted December 5,1990. Address reprint requests to Christine Gall, Dept. of Anatomy and Neurobiology, University of California, Irvine, CA 92717.
J.C. LAUTERBORN ET AL.
440
cortical targets as the source of NGF. Cholinergic neurons of the medial septal nucleus and the diagonal band of Broca project predominantly to hippocampus and olfactory forebrain regions (Woolf et al., '84; Zaborszky et al., '86), whereas neurons of the nucleus basalis of Meynert project to neocortex (Paxinos and Butcher, '85 for review). These target fields contain the highest levels of NGF mRNA in the brain (Korsching et al., '85; Shelton and Reichardt, '86; Whittemore et al., '86). The basal forebrain itself contains relatively low levels of NGF mRNA but one of the higher regional levels of NGF protein, thus suggesting that local NGF protein is actually synthesized in regions such as the hippocampus and transported into the basal forebrain (Korsching et al., '85; Whittemore and Seiger, '87 for review). This idea is supported by the observations that exogenous NGF is retrogradely transported from the hippocampus (Schwab et al., '79) and neocortex (Seiler and Schwab, '84) by the basal forebrain cholinergic neurons and that fimbrial transection leads to a build-up of NGF within the hippocampus (Gasser et al., '86; Weskamp et al., '86; Larkfors et al., '87). However, the hypothesis of cortically derived neurotrophic support has recently been questioned by Sofroniew et al. ('901, who showed that uninjured cholinergic neurons of the medial septum atrophy but do not die after their cortical target, the hippocampus, is ablated by the excitotoxin N-methyl-D-aspartate. This observation suggests that if cholinergic cells are critically dependent upon the trophic support of NGF, then there should be other sources of this substance for basal forebrain NGF-responsive neurons. As noted above, there are low levels of NGF mRNA within the basal forebrain itself. Moreover, using the two-site enzyme immunoassay technique, Lorez et al. ('88) found increases in NGF-immunoreactivity within basal forebrain areas following ablation of cortical targets, thereby providing further evidence for local synthesis. These authors suggested that increased NGF immunoreactivity in the lesioned rats may have been due to increased NGF production by glial cells. This postulate is consistent with the evidence that cultured brain astrocytes can synthesize and release NGF (Furukawa et al., '86, '87; Yamakuni et al., '87; Spranger et al., '90) and thus could be an additional source for this trophic factor within the brain. However, recent in situ hybridization studies in several laboratories have not detected NGF mRNA within adult forebrain glia (Rennert and Heinrich, '86; Ayer-LeLievre et al., '88; Whittemore et al., '88; Gall and Isackson, '89). In the present study, in situ hybridization techniques were used to determine the cellular localization of NGF mRNA in the basal forebrain. We found that neurons within several basal forebrain regions express NGF mRNA and that NGF mRNA-containing cells are distributed within fields containing cholinergic neurons. Some of these results have appeared in preliminary form (Lauterborn et al.,'89).
MATERIALS AND METHODS
histochemistry (n = 5) remained in these sohtiOnS for 2 and 24 hours, respectively. In all instances, the whole forebrain was sectioned in the coronal plane at a thickness of 20 pm with a freezing microtome. Sections were collected into cold 4% paraformaldehyde in PB for in situ hybridization or cold PB for immunohistochemistry and acetylcholinesterase histochemistry.
In situ hybridization Tissue sections were processed for the in situ hybridization localization of NGF mRNA using a 35S-labeled970-base RNA probe including a span complementary to 750 bases of the coding region of rat preproNGF (Whittemoreet al., '88). The "anti-sense'' and "sense" RNAs were transcribed in the presence of 35S-UTP from a rat genomic subclone, pBSrNGF, linearized with PvuII, by using T3 and T7 RNA polymerase, respectively. Free-floating tissue sections were transferred sequentially through 0.1 M glycine in PB, proteinase K (1pg/ml) in 0.1 M Tris buffer (pH 8.0) with 50 mM EDTA for 30 minutes at 37"C,0.25% acetic anhydride in 0.1 M triethanolamine (10 min), 2 x saline sodium citrate (SSC) (0.15 M NaCl and 0.015 M NaCitrate, pH 7.0) (30 min), and then incubated in a hybridization solution containing 50% formamide, 10% dextran sulfate, 0.7% ficoll, 0.7% polyvinyl pyrrolidone, 7 mgiml bovine serum albumin, 0.15 mg/ml yeast transfer RNA, 0.33 mg/ml denatured herring sperm DNA, and 20 p M dithiothreitol (DTT) for 1 hour at 60°C. From this prehybridization incubation, the sections were transferred into fresh hybridization solution containing the NGF cRNA probe at a density of 1 x lo6 cpm/lOO p1 with an additional 20 pM DTT and incubated for 30-40 hours at 60°C. Following hybridization, the sections were transferred through 4 x SSC (1hr), ribonuclease A (20 pg/ml) in 10 mM Tris-saline (pH 8.0) with 1mM EDTA for 30 minutes at 45"C,and then descending concentrations of SSC to a final stringency of 0.1 x SSC at 60°C for 1hour. Following an overnight wash in 0.1 x SSC at room temperature, the sections were mounted onto gelatin-coated slides and air dried. The distribution of cRNA hybridization was evaluated with emulsion autoradiography (Kodak NTBZ nuclear track emulsion 1:1 with H,O) and exposures of 4-6 weeks at 4°C. After development of the autoradiograms, the sections were stained with cresyl violet and coverslipped with Permount. Controls for specificity of hybridization included either treatment of sections with ribonuclease A (20 pg/ml) prior to normal hybridization, or hybridization with the labeled "sense" RNA sequence (described above). In neither instance was cellular labeling observed. In addition, the regional distribution of hybridization was consistent with the reported distribution of NGF mRNA (Korsching et al., '85; Shelton and Reichardt, '86; Whittemore et al., '86) in that the greatest number of labeled cells were found in hippocampus and no detectable labeling was found in cerebellum. Finally, as further evidence for specificity, the pattern of hybridization obtained with the present rat preproNGF cRNA was distinct from that obtained with other "S-labeled cRNAs (complementary to mRNAs for brain derived neurotrophic factor (BDNF), Fos, proenkephalin A, and proneuropeptide Y) in all areas examined.
Normal adult male Sprague-Dawley rats (250-300 gms; Simonsen Labs) were used. Animals were overdosed with sodium pentobarbital and intracardially perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) (PB). Choline acetyltransferase Following perfusion, the brains to be processed for in situ immunohistochemistry and ACHE hybridization alone (n = 10)were postfixed in 4% paraformhistochemistry aldehyde in PB for 24 hours and then transferred to 20% sucrose in 4%paraformaldehyde at 4°C for 48 hours; brains In five cases, a spaced series of tissue sections, adjacent to to be processed for both in situ hybridization and immuno- those used for in situ hybridization, were processed for the
NGF mRNA-CONTAINING NEURONS IN BASAL FOREBRAIN localization of choline acetyltransferase (ChAT) according to the avidin-biotin immunohistochemical technique and with 0.1 M Tris buffer (pH 7.4) for rinses and antisera dilution. Sections were incubated for 24 hours at 4°C in either rabbit-anti-ChAT (1:600; Chemicon) or rat-antiC U T (1:20; Boehringer Mannheim) with 0.3% Triton-X 100 and 1% normal goat serum (NGS). The sections were then washed for 30 minutes and incubated in either biotinylated antirabbit IgG or antirat IgG (1:200) containing 0.3% Triton-X 100 and 1% NGS for 1 hour at room temperature, Following a second buffer wash, the sections were transferred to the avidin-biotin complex solution (1:100, Vector Labs.) with 0.3% Triton-X for 1hour at room temperature. After a final wash, the sections were incubated in a saturated solution of 3,3'-diaminobenzidine with 0.01% H,O, in PB, rinsed in PB, mounted onto glass slides, dehydrated and cleared through Histoclear, and coverslipped. An additional series of sections adjacent to those processed for in situ hybridization was mounted onto glass slides and stained with cresyl violet. A final series of sections, adjacent to the ChAT immunohistochemical series, was processed for the histochemical localization of acetyl cholinesterase (Koelle and Friedenwald, '49) and used for further identification of cytoarchitectonic areas. Nomenclature is that of Paxinos and Watson ('86).
441
the lateral septal nucleus where hybridization-labeled neurons were scattered just deep to the islands of Calleja and within the more dorsal ventral pallidal region. Large numbers of labeled neurons were observed in several forebrain regions in sections taken through the level of the body of the septum (Fig. lC,D). Although hybridization was not observed within the medial septal nucleus itself, labeled cells were distributed within the lateral aspect of the horizontal limb of the diagonal band of Broca and extending lateral to this field into what might be considered the deeper ventral pallidum. In progressively more caudal sections, this cluster of labeled cells extended into the lateral aspect of the magnocellular preoptic area (Fig. 2B,E). In addition, a small number of cells situated between the magnocellular preoptic area and the more dorsal anterior commissure were autoradiographically labeled. At the most caudal aspect of the magnocellular preoptic nucleus, only a few labeled neurons were present. No hybridization to NGF mRNA was seen within the more caudal lateral hypothalamic area or within the more dorsal nucleus basalis of Meynert. Comparison of adjacent tissue sections processed for in situ hybridization and the immunohistochemical localization of ChAT demonstrated that in the regions of both the horizontal limb of the diagonal band and the magnocellular preoptic area there was a substantial number of autoradiographically labeled cells in the same areas that contain Analysis of cell size C U T - I R perikarya (Fig. 1). However, these cell groups The sizes of autoradiographically labeled cells and ChAT- were not fully coextensive. Hybridization-labeled cells were immunoreactive (ChAT-IR) cells within the caudal part of most numerous in the lateral aspects of, and extending the horizontal limb of the diagonal band and the magnocel- lateral to, each of these cytoarchitectonic fields, whereas lular preoptic area were measured in two animals. Measure- ChAT-IR cells were distributed throughout each region. In ments were made on a Leitz microscope under brightfield the horizontal limb of the diagonal band in particular, illumination at 4 0 magnification ~ with the aid of a cali- hybridization-labeled cells generally were distributed latbrated eyepiece reticule. Major and minor diameters ( f stan- eral to the very large C U T - I R cells but overlapped the field dard deviation (S.D.)) were measured for the somata of 50 of smaller ChAT-IR cells (Fig. 2). Beyond the horizontal ChAT-IR cells and for both the Nissl-stained nucleus and limb of the diagonal band and the magnocellular preoptic region of highest silver grain density for 50 autoradiograph- area, a smaller number of hybridization-labeled cells were ically labeled cells. The latter measure of the autoradio- observed in regions that contained ChAT-IR perikarya just graphic grain cluster is admittedly subjective, but one could dorsal to these nuclei as well as within the more rostral clearly identify a central region of higher grain density olfactory tubercle and dorsal to the islands of Calleja. (centered over the nucleus) surrounded by a less dense It should be noted that in regions containing both cell scatter of grains. The diameters of the core regions were groups, ChAT-IR perikarya were far more numerous than measured as the only estimate available of the size of the hybridization-labeled cells (Fig. 3). Moreover, as indicated radiolabeled perikarya. above, hybridization-labeled cells were not observed in a number of the principal regions containing CUT-IR neurons. Specifically, in agreement with numerous published RESULTS descriptions, ChAT-IR neurons were observed within the Neurons labeled by hybridization of the 35S-cRNAprobe medial septal nucleus, the vertical limbs of the diagonal were broadly distributed throughout the basal forebrain band, substantia innominata, and the nucleus basalis of extending rostrocaudally from the rostral aspect of the Meynert. Hybridization-labeled cells were not detected in anterior olfactory nucleus through the magnocellular pre- these areas. Within the horizontal limb of the diagonal band and optic area. As indicated in Figure 1, hybridization to NGF mRNA was also prominent in superficial layers of the magnocellular preoptic area, autoradiographic grains were piriform and cingulate cortices and within the claustrum at seen to overlie large pale Nissl-stained nuclei with mean these levels. A detailed description of the distribution of major and minor nuclear diameters of 14.00 2 1.82 (S.D.) hybridization in these and other olfactory and limbic struc- by 11.05 f 1.76 pm (Fig. 4). The average size of the major tures will be presented in a separate report (Lauterborn et density of autoradiographic grains overlying individual labeled neurons measured 27.50 ? 3.88 by 19.45 2 4.41 al., in preparation). At the most rostral planes, hybridization of the cRNA pm, further indicating that the labeled cells are relatively probe labeled cells within the anterior olfactory nucleus large. Smaller cells stained more densely with the Nissl (AON) and within the posterior aspect of the AON lying stain were never seen to be autoradiographically labeled. In between the anterior commissure and the olfactory tuber- contrast to the autoradiographically labeled cells, C U T - I R cle. Occasional labeled neurons were also seen within the cells within the overlapping areas measured on average deeper layers of the olfactory tubercle. The latter cells were 19.40 5 3.77 by 12.25 2 2.38 pm. By comparison, the more numerous in sections cut through the rostral limit of cholinergic cells lying within the medial aspect of the
J.C. LAUTERBORN ET AL.
442
A
B
Fig. 1. Line drawings of a rostral (A) to caudal (F) series of coronal sections through rat forebrain illustrating the distribution of cells labeled by in situ hybridization of the "S-labeled preproNGF cRNA probe and CUT-IR neuronal perikarya (solid circles each indicate 1-5 autoradiographically labeled cells; open triangles and open circles indicate 1-5 and &lo ChAT-IR cells, respectively). Autoradiographically labeled cells and CUT-IR cells are both distributed within (and lateral to) the horizontal limb of the diagonal band (HDB) (C) and within the magnocellular preoptic area (MCP) (C-F), whereas hybridization to NGF mRNA is not observed in fields of ChAT-I cells in the
medial septa1 nucleus (MSN) (C), nucleus basalis of Meynert (B) (E,F), or substantia innominata (SI) (El. Cross-hatching represents relatively diffuse autoradiographic labeling in the claustrum (CI), piriform cortex (PC) (A-F) and cingulate cortex (Cg). The asterisk in F denotes the basolateral nucleus of the amygdaloid complex, which also contains hybridization-labeled cells. Abbreviations: ac, anterior commissure; AP,posterior aspect of the anterior olfactory nucleus; CPu, caudate/ putamen; En, endopiriform nucleus; GP, globus pallidus; ICj, islands of Calleja; ox, optic chiasm; Tu, olfactory tubercle.
NGF mRNA-CONTAINING NEURONS IN BASAL FOREBRAIN
443
Fig. 2. Low magnification photomicrographs of two series of adjacent coronal sections (A-C and D-F) through the basal forebrain of normal adult rats showing a Nissl stain (A, D), autoradiographic localization of in situ hybridization of the 35S-cRNA(B,E; darkfield illumination), ChAT immunoreactivity (C), and acetylcholinesterase (AChE) histochemistry (F). In A-C, cellular hybridization of NGF cRNA (seen as white dots in B) is observed within the field of ChAT-IR neurons in the magnocellular preoptic area (solid arrow indicates a blood vessel in this cytoarchitectonic area in A, B, C). Note the zone of
hybridization within the lateral aspect of panel B overlies layers I1 and 111 of piriform cortex (PC). In E, NGF cRNA-labeled cells are seen within the horizontal limb of the diagonal band (HDB) and the field deep to the olfactory tubercle (Tu). In comparing E and F, one can see that the autoradiographically labeled cells lie in a field of dense AChEhistochemistry (F). Open arrow indicates one blood vessel present in each series. Abbreviations: lot, lateral olfactory tubercle. Bar = 900 pm for A-C; 700 pm for D-F.
horizontal limb of the diagonal band (which does not contain NGF cRNA-labeled neurons) appeared to be larger.
basal forebrain including, most notably, two major regions of cholinergic neurons, the horizontal limb of the diagonal band of Broca and the magnocellular preoptic area. Outside these areas, NGF synthesizing cells were found in the anterior olfactory nucleus, piriform cortex, olfactory tubercle, and aspects of the ventral pallidum; in several of these regions, NGF mRNA-containing cells were distributed
DISCUSSION The data presented here indicate that NGF mRNA is present in neurons distributed across broad fields of the
J.C. LAUTERBORN ET AL.
444
Fig. 3. Photomicrographs of paired adjacent sections (A,B; C,D) through different levels of the magnocellular preoptic area showing the autoradiographic localization of hybridization of the "S-cRNA probe (A,C; darkfield illumination) and ChAT-I (B,D). Open arrows indicate blood vessels, which appear in each of the adjacent sections. Solid arrows in A and C indicate individual hybridization-labeled neurons that can be seen to lie within the field of ChAT-I perikarya. Bar = 340 wm for A, B; 300 pm for C,D.
among cholinergic neurons. Cells containing NGF mRNA, however, were not present in all fields of cholinergic cells. Thus the medial septum/vertical limb of the diagonal band and the nucleus basalis of Meynert, which project primarily to hippocampus and neocortex, respectively, do not contain NGF synthesizing neurons. It has been suggested that glial cells, which synthesize NGF while in culture, might be the local source of NGF in the basal forebrain (Lorez et al., '88). This would accord with the observation that synthesis of NGF in the peripheral nervous system is by non-neuronal cell types including Schwann cells (Bandtlow et al., '87). However, previous in situ hybridization studies strongly suggest that the synthesis of NGF in hippocampus is by neurons (Rennert and Heinrich, '86;Ayer-LeLievre et al., '88; Whittemore et al., '88; Gall and Isackson, '€491, and the present data point to a similar conclusion for the basal forebrain. In particular, labeled cells of the horizontal limb of the diagonal band and the magnocellular preoptic area have large Nissl-pale nuclei indicating that they are most probably neurons. Conversely, smaller more darkly Nissl-stained nuclei, characteristic of some glial populations, were never labeled in our
material. Moreover, hybridization to NGF mRNA labeled large numbers of cells in layers I1 and 111of piriform cortex and it is known that the great majority of cells in these zones are neurons (Switzer et al., '85). The presence of both NGF mRNA-containing neurons and cholinergic cells in certain basal forebrain regions raises the possibility that the growth factor and transmitter are colocalized within individual neurons. Unfortunately, we have not found immunohistochemistry for ChAT to be compatible with the in situ hybridization technique used here and, therefore, have not as yet directly'tested the possibility of colocalization. However, in regard to this point it is noteworthy that the clusters of autoradiographic grains labeling individual basal forebrain neurons had a mean core aggregate size of 27.5 km (major diam.) suggesting NGF mRNA localization in larger neurons than the average 19.4 km diam. ChAT-IR cells in the coextensive field. Moreover, as described here, there are fewer NGF mRNA-containing cells than ChAT-IR cells in these areas and although these populations occupy overlapping fields, their distributions are not fully coextensive. Thus the hybridization-labeled and C U T - I R cell groups are not
NGF mRNA-CONTAINING NEURONS IN BASAL FOREBRAIN
445
specifically, whether they can be sustained by local NGF synthesis following such lesions. Some suggestion that this may be the case is provided by the observation that cholinergic neurons in the medial aspects of the globus pallidus, an area adjacent to the lateral basal forebrain as designated here, are not eliminated by lesions to their target areas in the cingulate cortex (Springer et al., '87). Additional experiments of this type should serve to test the hypothesis that the cholinergic basal forebrain can be differentiated into zones dependent upon NGF of local versus distant origin. One can speculate as to the functional consequences of the arrangements of trophic dependency proposed above. The dependence of some populations of neurons upon locally produced trophic factor might allow for the development and maintenance of "autonomous" regions in which such features as process elaboration and cell survival would be less affected by events occurring in the distant efferent targets (e.g., activity dependent changes in NGF expression) (Gall and Isackson, '89). Dependencies of the type seemingly present in the septohippocampal system would, in contrast, allow influences from the target region t o regulate the neurochemical differentiation and morphologiFig. 4. Brightfield photomicrograph showing t h e autoradiographic cal elaboration of cholinergic afferent neurons. In all, we localization of hybridization of t h e %-cRNA probe (seen as black grains) associated with neurons in t h e magnocellular preoptic area. T h e would expect that the extent to which neurons in a region tissue section was counterstained with cresyl violet to show cell nuclei. are interconnected with their local neighbors might dictate Note that t h e labeled cells (arrows) have relatively large Nissl-pale the degree to which local trophic influences predominate. nuclei in comparison t o t h e surrounding cells. Bar = 50 pm. We do not know the circuit relationships of the NGF producing neurons within the basal forebrain described here. In particular, more detailed information on the identical, but further work is needed to determine whether afferents t o these neurons is needed before we can formuNGF mRNA is localized in a subpopulation of cholinergic late informed hypotheses as to the trophic consequence of cells or within an entirely separate group of non-cholinergic NGF synthesis by these cells. neurons. Previous biochemical studes have shown that dissected ACKNOWLEDGMENTS samples of the septum plus the vertical limb of the diagonal band (Korsching et al., '851, or of this region together with We thank Gary Lynch for comments on the manuscript adjacent areas (Large et al., '86; Shelton and Reichardt, and Scott Whittemore for generously providing the rat '86), contain barely detectable levels of NGF mRNA. Thus NGF clone. This work was supported by NINDS grant it is generally assumed that the NGF protein found in these NS26748 to C.M.G. and NS24747 and a grant from Alzmidline areas arises from retrograde transport from hippo- heimer's Disease and Related Disorders Association to campus and other forebrain targets. The present results are P.J.I. in accord with this hypothesis as it relates to the septal region in particular, but strongly suggest that it should not LITERATURE CITED be generalized to the basal forebrain cholinergic system in its entirety, The lateral aspects of this system, found here to Ayer-LeLiewe, C., L. Olson, T. Ebendal, A. Seiger, and H. Persson (1988) Expression of the p-nerve growth factor gene in hippocampal neurons. contain neurons that express NGF mRNA, had not been Science 240: 1339-1341. specifically evaluated for NGF mRNA content in the earlier subfield dissection studies. In distinction from the more Bandtlow, C.E., R. Heumann, M.E., Schwab, and H. Thoenen (1987) Cellular localization of nerve growth factor synthesisby in situ hyhridizamedial fields and from nucleus basalis of Meynert, the tion. EMBO J. 6:891-899. ascending projections of the horizontal limb of the diagonal Furukawa, S., Y. Furukawa, E. Satayoshi, and K. Hayashi (1986)Synthesis band and the magnocellular preoptic area primarily innerand secretion of nerve growth factor by mouse astroglial cells in culture. Biochem. Biophys. Res. Comm. 13657-63. vate the piriform cortex; the horizontal limb also sends a moderate projection to hippocampus (Woolf et al., '84). Furukawa, S., Y. Furukawa, E. Satayoshi, and K. Hayashi (1987)Synthesis/ secretion of nerve growth factor is associated with cell growth in cultured Although numerous studies have evaluated the dependence mouse astroglial cells. Biochem. Biophys. Res. Comm. 142395-402. of septal neurons on the trophic support provided by distant Gage, F.H., D.M. Armstrong, L.R. Williams, and S.Varon (1988) Morphologtargets in lesion studies (Hefti, '86; Williams et al., '86; ical response of axotomized septal neurons to nerve growth factor. Kromer, '87; Springer et al., '87; Gage et al., '86, '881, J. Comp. Neurol. 289:147-155. considerably less attention has been directed toward this Gage, F.H., K. Wictorin, W. Fischer, L.R. Williams, S. Varon, and A. Bjorklund (1986) Retrograde cell changes in medial septum and issue in regard to these lateral cholinergic groups. As seen diagonal band following fimbria-fornix transection: Quantitative tempohere, the principal forebrain target for these neurons, the ral analysis. Neurosci. 19:241-255. piriform cortex, does contain relatively high levels of NGF C.M., and P.J. Isackson (1989) Limbic seizures increase neuronal mRNA and thus is presumably a major center for NGF Gall,production of messenger RNA for nerve-growth factor. Science245:758synthesis. However, it remains for future work to deter761. mine if the more lateral cholinergic cell groups can survive Gasser, U.E., G. Weskamp, U. Otten, and A.R. Dravid (1986)Time course of the elevation of nerve growth factor (NGF) content in the hippocampus the removal of such potential distant sources of NGF and,
446 and septum following lesions of the septohippocampal pathway in rats. Brain Res. 376.351-356. Gnahn, H., F. Hefti, R. Heumann, M.E. Schwab, and H. Thoenen (1983) NGF-mediated increase of choline acetyltransferase ( C U T ) in the neonatal rat forebrain: Evidence for a physiological role of NGF in the brain? Dev. Brain Res. 9:45-52. Higgins, G.A., K., Sookyong, K.S. Chen, and F.H. Gage (1989) NGF induction of NGF receptor gene expression and cholinergic neuronal hypertrophy within the basal forebrain of the adult rat. Neuron 3247256. Hefti, F. (1986) Nerve growth factor (NGF) promotes survival of septa1 cholinergic neurons after fimbrial transection. J. Neurosci. 6:2155-2162. Koelle, G.B., and J.S. Friedenwald (1949) A histochemical method for localizing cholinesterase activity. Proc. SOC. Exp. Biol. Med. 70;617-622. Korsching, S., G. Auburger, R., Heumann, J. Scott, and H. Thoenen (1985) Levels of nerve growth factor and its mRNA in the central nervous system of the rat correlate with cholinergic innervation. The EMBO J. 4:1389-1393. Kromer, L.F. (1987) Nerve growth factor treatment after brain injury prevents neuronal death. Science 235.214-216. Large, T.H., S.C. Bodary, D.O. Clegg, G. Weskamp, U. Otten, and L.F. Reichardt (1986) Nerve growth factor gene expression in the developing rat brain. Science 234:352-355. Lkkfors, L., I. Stromberg, T. Ebendal, and L. Olson (1987) Nerve growth factor protein level increases in the adult rat hippocampus after specific cholinergic lesion. J. Neurosci. Res. 18;525-531. Lauterborn, J.C., P.J. Isackson, and C.M. Gall (1989) Cellular localization of Neurosci. Abst. 15.864. p-NGF mRNA in rat forebrain. SOC. Lorez, H.P., M. von Frankenberg, G. Weskamp, and U. Otten (1988) Effect of bilateral decortication on nerve growth factor content in the basal nucleus and neostriatum of adult rat. Brain Res. 454:355-360. Mobley, W.C., J.L. Rutkowoski, G.I. Tennekoon, J. Gemski, K. Buchanan, and M.V. Johnson (1986) Nerve growth factor increases choline acetyltransferase activity in developing basal forebrain neurons. Mol. Brain Res. 1.53-62. Paxinos, G., and L.L. Butcher (1985)Organizational principles of the brain as revealed by choline acetyltransferase and acetylcholinesterase distribution and projections. In G. Paxinos (ed): The Rat Nervous System, Vol. 1.Orlando: Academic Press, pp. 487-508. Paxinos, G., and C. Watson (1986)The Rat Brain in Stereotaxic Coordinates. Orlando: Academic Press. Rennert, P.D., and G . Heinrich (1986) Nerve growth factor mRNAin brain: Localization by in situ hybridization. Biochem. Biophys. Res. Comm. 138:813-815. Rylett, R.J., and L.R. Williams (1990) Acetylcholine release following continuous intracerebral administration of nerve growth factor in adult Neurosci. Abst. 16:481. and aged Fisher 344 rats. SOC. Schwab, M.E., U. Otten, Y. Agid, and H. Thoenen (1979) Nerve growth factor (NGF) in the rat CNS: Absence of specific retrograde axonal transport and tyrosine hydroxylase induction in locus coeruieus and substantia nigra. Brain Res. 168;473-483. Seiler, M., and M.E. Schwab (1984) Specific retrograde transport of nerve
J.C. LAUTERBORN ET AL. growth factor (NGF) from neocortex to nucleus basalis in the rat. Brain Res. 300:33-39. Shelton, D.L., and L.F. Reichardt (1986) Studies on the expression of the p nerve growth factor (NGF) gene in the central nervous system: Level and regional distribution of NGF mRNA suggest that NGF functions as a trophic factor for several distinct populations of neurons. Proc. Natl. Acad. Sci. USA83:2714-2718, 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 247;338-342. Spranger, M., D. Lindholm, C. Bandtlow, R. Heumann, H. Gnahn, M. Niiher-NoB and H. Thoenen (1990) Regulation of nerve growth factor (NGF) synthesis in the rat central nervous system: Comparison between the effects of interleukin-1 and various growth factors in astrocyte cultures and in vivo. Europ. J. Neurosci. 269-76. Springer, J.E., S. Koh, M.W. Tayrien, and R. Loy (1987) Basal forebrain magnocellular neurons stain for nerve growth factor: Correlation with cholinergic cell bodies and effects of axotomy. J. Neurosci. Res. I7:111118. Switzer, R.C., J. De Olmos, and L. Heimer (1985) Olfactory system. In G. Paxinos (ed): The Rat Nervous System, Vol. 1. Orlando: Academic Press, pp. 1-36. Weskamp, G., U.E. Gasser, A.R. Dravid, and U. Otten (1986) Fimbria-fornix lesion increases nerve growth factor content in adult rat septum and hippocampus. Neurosci. Lett. 7:121-126. Whittemore, S.R., and A. Seiger (1987) The expression, localization and functional significance of p-nerve growth factor in the central nervous system. Brain Res. Rev. 12:439464. Whittemore, S.R., T. Ebendal, L. Lakfors, L. Olson, A. Seiger, I. Stromberg, and H. Persson (1986) Developmental and regional expression of 0 nerve growth factor messenger RNA and protein in the rat central nervous system. Proc. Nat. Acad. Sci. USA 83:817-821. Whittemore, S.R., P.L. Friedman, D. Larhammar, H. Persson, M. GonzalesCarvajal, and V.R. Holets (1988)Rat p-nerve growth factor sequence and site of synthesis in the adult hippocampus. J. Neurosci. Res. 20:403-410. Williams, L.R., and R.J. Rylett (1990) Exogenous nerve growth factor increases the activity of high-affinity choline uptake and choline acetyltransferase in brain of Fisher 344 male rats. J. Neurochem. 55:10421049. Williams, L.R., S. Varon, G.M. Peterson, K. Wictorin, W. Fischer, A. Bjorklund, and F.H. Gage (1986) Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbria fornix transection. Proc. Natl. Acad. Sci. USA 83:9231-9235. Woolf, N.J., F. Eckenstein, and L.L. Butcher (1984) Cholinergic systems in the rat brain: I. Projections to the limbic telencephalon. Brain Res. Bull. 13:751-784. Yamakuni, T., F. Ozawa, F. Hishinuma, R. Kuwano, Y. Takahashi, and T. Amano (1987) Expression of beta-nerve growth factor mRNA in rat glioma cells and astrocytes from rat brain. FEBS Lett. 223:117-121. Zaborszky, L., J. Carlsen, H.R. Brasher, and L. Heimer (1986) Cholinergic and GABAergic f i e r e n t s 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. Neurol. 243:488-509.
