Developmental profile of NGF immunoreactivity in the rat brain: a possible role of NGF in the establishment of cholinergic terminal fields in the hippocampus and cortex

Developmental Brain Research 101 Ž1997. 67–79

Research report

Developmental profile of NGF immunoreactivity in the rat brain: a possible role of NGF in the establishment of cholinergic terminal fields in the hippocampus and cortex James M. Conner ) , Silvio Varon Department of Biology, 0506, UniÕersity of California, San Diego, La Jolla, CA 92093, USA Accepted 25 February 1997

Abstract In the current investigation, we have examined the developmental profile of nerve growth factor immunoreactivity ŽNGF-ir. in the postnatal rat. During the first 3 weeks after birth, NGF-ir was observed within the hippocampal mossy fiber region, where it persists throughout adulthood and appeared transiently within three additional zones – the dentate gyrus supragranular zone, the tenia tectarintermediate lateral septum, and the cingulaterretrosplenial cortex. In all cases, the appearance of NGF-ir progressed in a rostrocaudal pattern over time. A strong correlation was seen between the pattern of NGF-ir and cholinergic innervation in the dentate gyrus supragranular zone, both spatially and temporally, suggesting that NGF may direct the innervation of cholinergic afferents to this region. A spatial correlation was also observed between NGF-ir and cholinergic innervation within the retrosplenial cortex and tenia tecta. With our current techniques, however, we were unable to determine at what point during development the adult-like pattern of cholinergic terminal innervation in these regions occurred and, thus, were not able establish a temporal correlation in these regions. Within the cingulate cortex, there was no evidence suggesting that the developmental appearance of NGF-ir in this region was associated with a specific enhancement of cholinergic innervation. Thus, the results of the current investigation clearly identify the presence of transiently occurring zones of NGF-ir during postnatal CNS development, although defining their exact functional role will require additional investigation. q 1997 Elsevier Science B.V. Keywords: Nerve growth factor; Basal forebrain; Synaptogenesis; Immunohistochemistry

1. Introduction Nerve growth factor ŽNGF. has been postulated for many years to serve as a target-derived retrograde signal for specific populations of neurons in the peripheral nervous system where it influences neuronal survival and differentiation during critical periods of development. Within the CNS, a similar role of NGF might be inferred for specific populations of neurons in the basal forebrain based upon several lines of evidence. NGF is present in highest concentrations within the innervation targets of basal forebrain cholinergic neurons w28,36x, and cholinergic neurons within the basal forebrain possess both high and low affinity receptors for NGF w4,20,25x. NGF is specifically bound to these receptors and retrogradely

) Corresponding author. Fax: q1 Ž619. 534-0128; e-mail: [email protected]

0165-3806r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 3 8 0 6 Ž 9 7 . 0 0 0 5 1 - 5

transported from the target territories to the basal forebrain w9,35x. The possibility that NGF might act as a cholinergic differentiation factor has been suggested based on the demonstration that Ži. NGF expression in basal forebrain target territories is correlated with choline acetyltransferase ŽChAT. activity during postnatal development w29x, and Žii. NGF has been shown capable of stimulating ChAT activity in perinatal basal forebrain cholinergic neurons, both in vitro and in vivo w19,33x. Speculation that NGF might also serve as a target-derived chemoattractant for guiding axonal outgrowth from basal forebrain cholinergic neurons to their appropriate targets has also been offered based upon Ža. in vitro slice culture studies demonstrating that basal forebrain explants selectively send out AChE-positive fibers toward cortical or hippocampal slices but do not respond similarly to non-target slices w17,18x, and Žb. in vivo adult rat regeneration models, whereby NGF both stimulates and directs cholinergic axonal regrowth w21,22x. The ability of NGF to rescue injured basal forebrain

68

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

cholinergic neurons in selected animal models of neurodegeneration w24,40x has also led to the suggestion that NGF may act as a developmental survival factor for these

neurons in much the same way NGF acts upon peripheral sensory and sympathetic neuronal cell populations. Most of these proposed roles for NGF in the developing

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

basal forebrain cholinergic system, however, have not been supported by recent investigations using null mutations for either the NGF or the high affinity NGF-receptor ŽTrkA.

69

genes w14,37x. These studies have indicated that Ži. NGF is not essential for the survival of central cholinergic neurons of the basal forebrain during development, Žii. basal fore-

70

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

brain neurons develop a cholinergic phenotype in the absence of NGF Žalthough reduced ChAT levels in these animals may suggest cholinergic activity is modulated to some extent by NGF availability., and Žiii. basal forebrain cholinergic neurons extend processes toward, and ramify within, their normal cortical and hippocampal targets without NGF being present. Furthermore, studies comparing the developmental onset of NGF gene expression Žand protein production. within CNS innervation targets and the time when these targets become innervated by NGF-sensitive fibers have further indicated that target innervation has occurred well before NGF expression has begun w27,32,39x. This is in marked contrast to what is known for the situation in the developing peripheral nervous system, where NGF expression in a peripheral tissue occurs either before or at the time of its innervation by NGF-sensitive afferents w15,16x. In the hippocampal formation, however, the onset of NGF expression is most tightly correlated with the onset of synaptogenesis w13,39x, raising the possibility that NGF plays a major role in the establishment of synaptic connections. In previous studies, we have used immunohistochemical techniques to demonstrate the presence of NGF in restricted subregions of the hippocampal formation w8x. In addition, a defined model of sympathetic sprouting into the hippocampal formation has provided evidence for a strict correlation between the tissue regions where NGF is localized and those which the sprouting sympathetic axons will innervate w10,41x. Together, such findings have led to the idea that NGF may topographically define the terminal fields for NGF-sensitive afferents during normal, as well as experimentally induced sprouting. We have extended the use of the sympathetic sprouting model to show that the pattern of experimental sympathetic innervation can be deliberately altered by either additional lesion-induced regions of NGF immunoreactivity w10x or exogenously added NGF w11x, thereby confirming that the original distribution of endogenous NGF was critical for defining where the sympathetic fibers would innervate.

Using another experimental model, previous studies have shown that the interruption of entorhinal afferents to the hippocampal formation causes the endogenous cholinergic afferents to undergo synaptic remodeling w12,30x. Using this model, we have demonstrated that concurrent changes occur in the distribution of NGF-ir within the hippocampal formation, again encouraging the view that NGF may be involved in determining where these sprouting fibers will go w7x. Furthermore, blocking the actions of endogenous NGF completely blocks cholinergic sprouting in this model w38x. An interesting facet concerning sprouting of the cholinergic terminals in this model is that the lesion-induced appearance of NGF-ir in the region where sprouting takes place was transient. The appearance of NGF-ir in the deafferented region, which preceded the onset of cholinergic sprouting, was maximal during the time when cholinergic sprouting has been reported to be greatest and declined at later times when cholinergic sprouting has reportedly slowed w7x. Moreover, the lesioninduced NGF-ir would occur even if the cholinergic sprouting was prevented Žby removing the cholinergic afferents from the hippocampal formation.. This suggested that the NGF is required to establish a new terminal field for the sprouting cholinergic fibers, but may not be required once reorganization is complete. The apparent ability of topographically restricted regions of NGF Ževen transiently occurring ones. to establish terminal fields for incoming NGF-sensitive axons has raised the question whether such a mechanism may be directing the distribution of normal endogenous basal forebrain cholinergic afferents within their targets during development. To begin investigating such a question, we have examined the developmental profile of NGF-ir in the postnatal rat. 2. Materials and methods Sprague–Dawley rats ŽHarlan, San Diego, CA. at various postnatal ages Žpostnatal day ŽP. 0, 1, 3, 5, 6, 7, 8, 9,

Fig. 1. Ž On p. 68 . Developmental appearance of NGF immunoreactivity in the cortex. At P0, moderate staining for NGF was seen overlying the cell layer of the tenia tecta ŽTT. and extending out into the intermediate lateral septum ŽLSI. ŽA.. Heavy labeling was also seen at P0 within the cingulate cortex ŽCg. ŽB. and extending back to the rostral level of the retrosplenial cortex ŽRS. ŽC.. Staining was reduced at more caudal levels of the retrosplenial cortex Žarrow in D.. By P5, staining in the cingulate cortex was markedly diminished Žarrow in E. but remained high in the rostral retrosplenial cortex ŽF.. NGF staining also increased in the retrosplenial cortex at more caudal levels ŽG.. By P7, NGF staining was completely absent from the cingulate cortex Žarrow in H.. NGF staining still persisted within the rostral retrosplenial cortex ŽI. and was at its maximum within the caudal retrosplenial cortex ŽJ.. By P12, all NGF staining within the cortex was below the threshold of detection ŽK–M.. Scale bar in A s 250 m m; bar in D s 300 m m and applies to B–M. gcc, genu corpus callosum; HF, hippocampal formation. Fig. 2. Ž On p. 69 . NGF immunoreactivity in the developing hippocampal formation. At P0, very little NGF immunostaining was present in the hippocampal formation. At the most rostral pole of the hippocampus, moderate staining was observed in the medial portion of CA1 and overlying the dentate gyrus ŽDG. granule cell layer Žarrow in A.. Staining was completely absent from more caudal sections ŽB., although some specific labeling was observed within the subiculum ŽS. Žarrow in B.. By P3, some NGF-ir was detected in the mossy fiber region in more rostral hippocampal sections Žarrow in C., but staining was still absent at more caudal hippocampal levels ŽD.. By P7, heavy staining was seen within the rostral hippocampal formation and the distinction of the mossy fiber region was obvious ŽE.. In addition, a heavy band of NGF staining was also observed along the supragranular zone, primarily overlying cells in the outer one-third of the granule cell layer Žarrow in E.. At more caudal levels of the hippocampal formation, NGF staining was just becoming apparent within the mossy fiber region at P7 ŽF.. By P21, NGF staining was equally robust throughout the entire rostrocaudal extent of the hippocampal formation ŽG, H.. In addition, staining within the dentate gyrus supragranular zone was markedly reduced by P21. Scale bar in H s 500 m m and applies to A–H.

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

10, 11, 12, 14, 17, 21, and adult Ž; 3 monthsr200–225 g female.. were obtained from four separate litters. In most cases, four rats were sacrificed at each timepoint, with the

71

exception that five animals were used at P0 and only three animals were used at P17 and P21. For most of the timepoints examined, one animal from each litter was used. Animals less than 48 h old were anesthetized by hypothermia and those older than 48 h were anesthetized using ketamine hydrochloriderxylazineracepromazine. While under deep anesthesia, animals were perfused transcardially with phosphate-buffered saline followed with 2% paraformaldehyde q0.2% parabenzoquinone in 0.07 M phosphate buffer. Their brains were removed, post-fixed for 2 h in the same fixative, and cryoprotected overnight in 30% sucrose in 0.1 M phosphate buffer. Specimens were cut in the coronal plane on a sliding microtome Ž40–90 m m thickness. and processed for NGF, p75, or TrkA immunoreactivity as has been described w8x. In brief, sections were washed in Tris-buffered saline ŽTBS., incubated in TBS containing 0.25% Triton X-100, and blocked in TBS q2% bovine serum albumin q5% normal goat serum or normal horse serum. Staining was performed by incubating sections with primary antibodies Ž0.05 m grml anti-NGF w8,9x, 0.1 m grml anti-p75 ŽIgG192, Boehringer Mannheim, Indianapolis, IN. or 1:4000 dil anti-trkA w5x. for 48 h at 48C. This was followed by incubating sections with secondary antibodies Ž1.5 m grml either biotinylated goat anti-rabbit or horse anti-mouse ŽVector Labs, Burlingame, CA. for 3 h at room temperature, and with an avidinrbiotinrperoxidase reagent Ž1:250 dilution ABC Elite ŽVector Labs.. for 90 min at room temperature. Sections were then reacted in a Tris-HCl solution containing 0.04% diaminobenzidine tetrahydrochloride, 0.06% nickel chloride, and 0.06% H 2 O 2 in Tris-HCl buffer. Additional sections were processed for AChE histochemistry as described w34x.

3. Results 3.1. NGF immunoreactiÕity In the adult animal, NGF-ir was only seen within the hippocampal formation and basal forebrain as has been described w8x. An intense zone of NGF-ir was present in the hippocampal formation in an area corresponding to the region occupied by axonal processes from the dentate gyrus granule cells Žmossy fiber region.. Within the basal forebrain, NGF staining appeared as discrete puncta con-

Fig. 3. NGF staining in developing basal forebrain neurons. Even at P0, weak staining for NGF could be detected in cells of the basal forebrain Žarrows in A., particularly those of the nucleus basalis magnocellularis ŽA.. As was observed in the adult, staining appeared as discrete puncta contained within the cell’s cytoplasm. Staining intensified and became apparent in more basal forebrain neurons over time ŽP3 ŽB.; P7 ŽC.. until a maximum level of staining was reached at P12 ŽD.. Scale bar in Ds 25 m m and applies to A–D.

72

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

tained within cell bodies, which have been identified previously as cholinergic neurons w8x. Developmentally, moderate NGF labeling was seen

within the tenia tecta and indusium griseum at P0 ŽFig. 1A., began to fade by P8, and was absent by P12. Within the developing cortex, NGF-ir was detected at P0 over cell

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

bodies throughout the cingulate cortex and extending to the most rostral portion of the retrosplenial cortex ŽFig. 1B,C.. At this time, very weak staining was seen in more caudal portions of the retrosplenial cortex ŽFig. 1D.. By P5, NGF-ir within the most rostral portion of the cingulate cortex had begun to fade and NGF-ir within the retrosplenial cortex had extended more caudally ŽFig. 1E–G.. At P7, NGF-ir was completely absent from within the cingulate cortex, while robust labeling within the retrosplenial cortex extended caudally to a level where the posteriormost extent of the hippocampal formation occurs ŽFig. 1H–J.. Staining within the retrosplenial cortex gradually faded, beginning at the most rostral level and progressing caudally, until by P12 all cortical labeling for NGF had faded below a threshold of detectability ŽFig. 1K–M.. In the developing hippocampal formation, faint immunoreactivity for NGF was seen at P0 over cells within the most rostral portion of the dentate gyrus granule cell layer ŽFig. 2A.. By P3, the intensity of the staining over the granule cells had increased, and it was noted that labeling was found almost exclusively over granule cells within the outer portion of the granule cell layer. At P3, some faint NGF-ir was also present within the mossy fiber region at the most rostral level of the hippocampal formation ŽFig. 2C., although its similarity to the adult pattern was not readily apparent until P5–6. By P7 the NGF-ir overlying the outermost granule cell layer was intensified at more rostral levels and had extended caudally through the rostral half of the hippocampal formation Žthe caudal hippocampal formation had only weak NGF staining at this time; Fig. 2E,F.. NGF-ir over the granule cells began to fade by P12 but did not quite reduce to adult levels by P21 ŽFig. 2G.. Staining within the mossy fiber region also became more defined, as well as intensified, with increasing postnatal times and appeared to reach an adult-like pattern throughout the hippocampal formation by P21 ŽFig. 2G,H.. By P17–21, lightly stained NGF-ir cells could also be observed scattered within the CA1 Žprimarily within the pyramidal cell layer. and CA2rCA3 Žwithin the pyramidal cell layer and stratum lucidum. hippocampal regions. Within the basal forebrain, punctate NGF-ir was detected inside of cell bodies in the nucleus basalis, medial septum and diagonal band as early as P0 ŽFig. 3A..

73

Staining in these neurons continued to increase in intensity until about P12 ŽFig. 3B–D., after which there was a slight decrease in staining until adult levels were reached at P21. 3.2. NGF-receptor immunoreactiÕity The choice of the p75 marker for demonstrating cholinergic innervation of the hippocampal formation and cortex was made because Ži. the fixation protocol necessary for preserving the NGF antigen has thus far proven to be incompatible with immunohistochemical staining for ChAT, and Žii. the p75 immunohistochemical procedure was much more reliable in our hands than was the AChE histochemistry. Justification for this choice was made based upon the following evidence: first, previous anatomical studies have indicated that cholinergic innervation of the hippocampal formation and cortex arises primarily Žif not exclusively. from the cholinergic neurons within the basal forebrain w26,31x. Other studies w1x have indicated that basal forebrain cholinergic neurons colocalize for the p75 antigen almost exclusively. Finally, specific immunotoxic lesions of the p75 containing cells within the basal forebrain completely eliminate cholinergic innervation within the hippocampal formation and cortex, further confirming that this cell population provides exclusive cholinergic innervation of these regions w2,3,23x. Nevertheless, confirmation of the cholinergic nature of the cortical and hippocampal innervation was made in most instances by processing adjacent sections for AChE histochemistry. It is important to point out that, although p75-immunoreactive fibers are located throughout the adult CNS, we have limited our evaluation of them to the primary innervation targets for basal forebrain cholinergic neurons – namely the cortex and hippocampal formation. In the adult animal, a distinctive pattern of lamination was observed within the hippocampal formation ŽFig. 4A,C.. p75-ir fibers were most heavily concentrated in the supragranular zone of the dentate gyrus, overlying the outer one-third of the dentate gyrus granule cell layer. Moderate innervation was found in CA1 stratum oriens and within the one-third of stratum radiatum adjacent to the pyramidal cell layer. Fibers in the remaining two-thirds of the CA1 stratum radiatum and throughout lacunosum

Fig. 4. Cholinergic basal forebrain innervation in the adult rat hippocampal formation and cortex. Cholinergic terminal innervation in the adult hippocampal formation revealed by p75 immunoreactivity ŽA. or AChE histochemistry ŽC.. Specific patterns of laminar innervation include: Ži. a heavy band of innervation on either side of the pyramidal cell layer of CA1 Žsolid arrow in A.; Žii. a distinct laminar pattern in the dentate gyrus ŽDG., whereby a heavy band of staining occurred in the supragranular zone Žopen arrow in A.; Žiii. an almost clear zone in the inner one-third of the molecular layer Žmol.; and Živ. moderate innervation in the outer two-thirds of the molecular layer. Robust cholinergic innervation was also seen within the tenia tecta ŽTT.rintermediate lateral septum ŽLSI. in the adult animal ŽB, D.. Within the cingulate cortex ŽCg., p75-ir fibers were distributed fairly uniformly across the various cortical layers, with no apparent laminar pattern of innervation ŽE.. In the retrosplenial cortex ŽRS. ŽF., a distinctive pattern of cholinergic innervation was seen, with heavier p75 innervation occurring within the region immediately adjacent to layer 2 neurons Žarrow in F.. In the remaining cortical subfields Žparietal cortex ŽPar. shown in G., cholinergic innervation was rather uniform across cortical layers, with p75-ir fibers running parallel and perpendicular to the cortical surface. Confirmation of the cholinergic nature of these afferents was made by processing similar sections for either p75 immunohistochemistry ŽA, B, E, F, G. or AChE histochemistry ŽC, D, H, I, J.. Scale bar in A s C s 500 m m; bar in B s D s 250 m m; bar in E s 500 m m and applies to E–J. gcc, genu corpus callosum; lv, lateral ventricle; Fr, frontal cortex; cc, corpus callosum; cg, cingulum.

74

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

moleculare were much more sparse. A moderate density of p75-ir fibers was also seen in the dentate gyrus molecular layer and within the hilus. Also, moderate fiber density

was observed in CA3 and CA2 stratum oriens and lucidum. Within the tenia tecta and intermediate lateral septum ŽFig. 4B,D., cholinergic fibers were concentrated

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

on either side of the cell layer, with the heaviest labeling seen within the lateral septum, adjacent the tenia tecta. p75-ir fibers were distributed throughout all regions of the adult cortex and across all cortical layers ŽFig. 4E–G.. Fibers were oriented primarily either parallel or perpendicular to the cortical surface. In nearly all cortical regions, fibers tended to be distributed rather uniformly across the cortical layers, except within the retrosplenial cortex where a distinctive pattern of lamination occurred. In the retrosplenial cortex, p75-positive fibers tended to be more heavily concentrated in the region just lateral to the layer 2 neurons ŽFig. 4F.. The pattern of fiber staining seen in the hippocampal formation and cortex was similar, regardless of whether p75 immunohistochemistry or AChE histochemistry was used for identifying cholinergic processes. In the developing cortex, p75-ir fibers were present at P0 in nearly all subregions. In the frontal cortex ŽFig. 5A., fibers appeared to begin normal patterning, most noticeable in deeper cortical layers, with fibers forming a meshwork perpendicular and parallel to the cortical surface. At more caudal levels Žwhere the hippocampal formation begins., p75-ir fibers within the cortex were almost exclusively oriented in a radial fashion Ži.e., perpendicular to the cortical surface; Fig. 5B.. By P5, cortical fiber labeling for p75 appeared to be organized similar to the adult in the frontal cortex and extending as far back as the level of the septum, but was still radially oriented in more caudal sections ŽFig. 5E.. In more rostral sections, fibers within frontal and parietal cortex seem to organize first, while those in the insular cortex remained more radially oriented. By P9, p75-ir fiber staining was well organized in most cortical regions with the exception of the insular cortex, where some radial staining persisted. By P12 the entire cortex appeared to have an adult-like distribution of p75-ir fibers. As early as P0, p75-ir fibers were present throughout the rostrocaudal extent of the hippocampal formation ŽFig. 6A,B.. The estimated number of p75-ir fibers in the hippocampal formation was low at this timepoint and they tended to be distributed rather evenly across the various hippocampal subfields Žno evidence of a discernible pattern was seen.. By P5–P7, p75-ir fibers had begun to accumulate over the cells in the outer one-third of the granule cell layer, giving the first hint of any type of organization of cholinergic processes within the hippocam-

75

pal formation ŽFig. 7.. At P7, the first evidence of lamination of p75-ir fibers was also seen within CA1, where fibers tended to accumulate on either side of the pyramidal cell layer and ‘‘thin out’’ from within the stratum lacunosum moleculare. Lamination of p75-ir fibers throughout the rostrocaudal extent of the hippocampal formation appeared to reach an ‘‘adult-like’’ level of organization by P14. Intense p75 and TrkA immunoreactivity was present within neurons in all three basal forebrain regions at P0 and remained high throughout the developmental timecourse Ždata not shown.. It is also interesting to note that specific p75 immunoreactivity was seen in association with blood vessels in the brain. This vascular staining was most pronounced during the second and third postnatal weeks and then faded significantly as animals approached adulthood. The source of the p75-ir associated with vascular elements in the brain has yet to be determined.

4. Discussion Previous studies investigating lesion-induced sprouting have led to the speculation that NGF may become topographically restricted in certain CNS regions for the purpose of guiding innervation into those regions by NGFsensitive neuronal processes w7,10,11,41x. An extension of this hypothesis might propose that a similar mechanism is used to direct target innervation by NGF-sensitive cell populations during development. Basal forebrain neurons may initially innervate their cortical and hippocampal targets independent of NGF w14x and establish a uniform distribution of cholinergic terminals throughout these regions w27x. At some later stage during their postnatal development, however, basal forebrain cholinergic fibers may become dependent upon NGF for their maintenance, thereby allowing innervation to persist only in the restricted areas where NGF is made available Žperhaps even maintaining innervation in proportion to the magnitude of the neurotrophic stimulus encountered.. Thus, discrete and laminar patterns of NGF, presented within target tissues during development, may lead to the definition of laminar patterns of innervation by NGF-sensitive cholinergic afferents. In the present study, we have explored this possibility by evaluating NGF-ir within innervation targets of NGF-

Fig. 5. Cholinergic innervation within the developing cortex. At P0 ŽA–C., p75-positive cholinergic fibers are observed in most cortical regions. Within the frontal cortex ŽFr. ŽA., fibers in the deeper cortical layers appeared to organize similar to innervation seen in the adult, oriented parallel and perpendicular to the cortical surface. In other cortical regions, such as the temporal cortex ŽTe. ŽB., innervation was almost exclusively oriented in a radial pattern, extending from the corpus callosum Žcc. towards the cortical surface Žindicated by arrow in B.. p75-ir fibers were only weakly observed in the insular cortex ŽIns. at this time ŽC., although heavy p75 innervation of the pyriform and olfactory cortex was noted Žopen arrow in C.. By P5 ŽD–F., cholinergic innervation of the frontal cortex ŽD. was comparable to the pattern seen in the adult animal although staining in the temporal ŽE. and insular cortex ŽF. were still oriented primarily in a radial fashion. By P9, cholinergic innervation in most cortical regions was comparable to that seen in the adult Žfrontal cortex ŽG. and temporal cortex ŽH.., although fibers in the insular cortex ŽI. remained oriented in a radial fashion and appeared more dense than in the adult ŽJ.. By P12 cholinergic innervation of all cortical regions appeared similar to innervation seen in the adult animal. Scale bar in A s 200 m m and applies to A, B, D, E, G, H; bar in C s 500 m m and applies to C, F, I, J. CPu, caudate putamen; Par, parietal cortex.

76

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

sensitive basal forebrain cholinergic neurons in the rat during the first 3 weeks of postnatal life, a time when cholinergic terminal fields are established and synaptogen-

esis takes place w13x. It is important to bear in mind that the immunohistochemical and histochemical techniques used to visualize NGF-ir and cholinergic fibers each have

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

Fig. 7. Correlation of transient NGF staining and cholinergic innervation within the dentate gyrus supragranular zone. At P7, a heavy band of NGF immunoreactivity was seen in the dentate gyrus supragranular zone Žarrows in A., primarily overlying cell bodies in the outer one-third of the granule cell layer ŽA.. At this time during development, a noticeably heavier band of p75-positive cholinergic terminals also appeared along the dentate gyrus supragranular zone Žarrows in B.. Scale bar s 75 m m. GrDG, granule layer of dentate gyrus.

a threshold of sensitivity associated with them. It is possible, therefore, that the markers identified by these techniques are present in CNS regions at levels which escape detection by our present methods. Using the sensitive immunohistochemical protocol previously described for detecting NGF-ir w8x, we have determined the timecourse for the appearance of NGF-ir within the mossy fiber region of the hippocampal formation, which is seen in the adult, and have identified three additional zones of NGF-ir appearing transiently during postnatal development. One such region is within the

77

dentate gyrus supragranular zone, one within the tenia tectarlateral septum, and one within the cingulaterretrosplenial cortex. Within each of these regions, the appearance of NGF-ir evolved in a rostrocaudal pattern over time. In most cases, staining in rostral regions began to fade as more caudal staining appeared, until at a given time Ž; P12. all staining had disappeared. In the case of the hippocampal mossy fiber region, NGF staining still progressed along a rostrocaudal axis but the entire extent of staining persisted throughout the adult life of the animal. An attempt was made to determine if the transiently appearing zones of NGF-ir occurring during development correlated with innervation by NGF-sensitive basal forebrain cholinergic fibers. In the present study, cholinergic innervation within the hippocampal formation and cortex was monitored primarily by the immunohistochemical detection of the p75 antigen, but was confirmed in most instances with AChE histochemistry. Our results on the developmental timecourse of cholinergic innervation of the hippocampal and cortical targets are almost identical to the findings of Koh and Loy w27x, although no attempt was made to correlate specific patterns of cholinergic innervation with NGF distributions in this previous study. In the present study, several pieces of evidence suggested that the appearance of NGF within the dentate gyrus supragranular zone may be guiding NGF-sensitive basal forebrain axons to their appropriate terminal field: Ži. the location of NGF within the supragranular zone was precisely correlated with the region innervated by cholinergic afferents; Žii. the time when the NGF patches appeared in the supragranular zone during development was associated with, and slightly preceded, the time when NGF-sensitive cholinergic fibers accumulated within this region; and Žiii. the spatio-temporal manifestation of the patches, in a rostrocaudal fashion, was in agreement with the progression of establishing cholinergic terminal fields. The fact that this zone of NGF-ir was only present transiently during development, while the distinct laminar pattern of cholinergic terminals remained throughout adulthood, may suggest that the NGF is necessary only for establishing terminal fields and is not required to maintain them. Within the cingulate cortex, no such correlation was seen. Although robust NGF-ir was observed in this region between birth and P5, no distinct accumulation of cholinergic terminals appeared, even during adult life. In the

Fig. 6. Cholinergic innervation of the developing hippocampal formation. p75-positive cholinergic fibers were observed throughout the rostrocaudal extent of the hippocampal formation at P0, although no particular laminar pattern of fiber organization was noted ŽA, B.. At P5 ŽC, D., the density of cholinergic innervation had increased above that observed at P0 and some evidence of lamination was noted within the supragranular zone Žarrow in C.. Within other regions of the hippocampal formation, fibers tended to be evenly distributed across layers and no evidence of lamination was apparent. By P10 ŽE, F., evidence of specific lamination was observed in many regions. In addition to the heavy band of fibers over the dentate gyrus supragranular zone, a band of cholinergic fibers began to form on either side of the CA1 pyramidal cell layer Žopen arrow in F., while fibers appeared to vacate much of the stratum radiatum Žrad. and lacunosum moleculare Žlmol.. At this time, innervation within the dentate gyrus molecular layer Žarrows in E and F. still appeared somewhat disorganized compared to the even pattern seen in the adult Žsee Fig. 4A.. By P14 ŽG, H., however, cholinergic innervation within all portions of the hippocampal formation appeared similar to that seen in the adult animal. Scale bar in A s 200 m m and applies to A and B; bar in H s 500 m m and applies to C–H. DG, dentate gyrus.

78

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79

retrosplenial cortex and tenia tectarlateral septum, the association between transiently appearing NGF-ir and the accumulation of cholinergic fibers was less clear. The location of the transient zone of NGF-ir in both of these regions was very similar to regions where cholinergic terminals were concentrated in the adult animal. In our investigation, however, we were not able to determine when these areas of dense cholinergic innervation formed. It is possible that the accumulation of cholinergic terminals in these regions occurs after P21 Žthe latest developmental timepoint investigated in this study., in which case it is unlikely that the NGF appearing early in development was associated with establishing cholinergic terminal fields. Alternatively, the levels of cholinergic markers Žp75 and AChE. in these regions may have been less than optimal during development, making it difficult to detect all cholinergic fibers in the region. It is important to point out that many basal forebrain cholinergic target territories become innervated Žand even laminated. without patches of NGF being detected in them during development. The entire cortex is occupied by cholinergic fibers arising from neurons in the nucleus basalis magnocellularis, even though NGF-ir is not detected within the cortex Žother than cingulate and retrosplenial cortex. at any time during development. Similarly, many hippocampal areas, such as the dentate gyrus molecular layer and the CA1–CA3 stratum oriens, become moderately innervated by cholinergic process from the medial septum without the occurrence of detectable NGF patches. The possibility exists that cholinergic innervation to these regions occurs independent of NGF and that NGF is only required in regions Žsuch as tenia tecta, retrosplenial cortex and dentate gyrus supragranular zone. where additional, heavy innervation will take place. Alternatively, NGF-ir associated with the establishment of cholinergic innervation in most regions may simply not be detected using our current immunohistochemical protocol and only in those regions where high levels of NGF are directing heavy cholinergic innervation can the NGF-ir be seen. Support for this latter explanation comes from various studies demonstrating low, but detectable levels of extractable NGF in various cortical and hippocampal regions where NGF-ir is not observed w6,28x.

w2x

w3x

w4x

w5x

w6x

w7x

w8x

w9x

w10x

w11x

w12x

w13x

w14x

w15x

w16x

Acknowledgements We wish to thank Dr. L.F. Reichardt and Dr. D.O. Clary for the gift of the TrkA antibody. The work described here was largely supported by NINDS NS-16349.

w17x

w18x

References w1x P.E. Batchelor, D.M. Armstrong, S.N. Blaker, F.H. Gage, Nerve growth factor receptor and choline acetyltransferase colocalization in

w19x

neurons within the rat forebrain: response to fimbria fornix transection, J. Comp. Neurol. 284 Ž1989. 187–204. J. Berger-Sweeney, S. Heckers, M.-M. Mesulam, R.G. Wiley, D.A. Lappi, M. Sharma, Differential effects on spatial navigation of immunotoxin-induced cholinergic lesions of the medial septal area and nucleus basalis magnocellularis, J. Neurosci. 14 Ž1994. 4507– 4519. A.A. Book, R.G. Wiley, J.B. Schweitzer, Specificity of 192 IgGsaporin for NGF receptor-positive cholinergic basal forebrain neurons in the rat, Brain Res. 590 Ž1992. 350–355. C.R. Buck, H.M. Martinez, M.V. Chao, I.B. Black, Differential expression of the nerve growth factor receptor gene in multiple brain areas, Dev. Brain Res. 44 Ž1988. 259–268. D.O. Clary, G. Weskamp, L.R. Austin, L.F. Reichart, TrkA crosslinking mimics neuronal responses to nerve growth factor, Mol. Biol. Cell 5 Ž1994. 549–563. F. Collins, K.A. Crutcher, Neurotrophic activity in the adult rat hippocampal formation: regional distribution and increase after septal lesion, J. Neurosci. 5 Ž1985. 2809–2814. J.M. Conner, B. Fass-Holmes, S. Varon, Changes in nerve growth factor immunoreactivity following entorhinal cortex lesion – a possible molecular mechanism regulatine cholinergic sprouting, J. Comp. Neurol. 345 Ž1994. 409–418. J.M. Conner, D. Muir, S. Varon, T. Hagg, M. Manthorpe, The localization of nerve growth factor-like immunoreactivity in the adult rat basal forebrain and hippocampal formation, J. Comp. Neurol. 319 Ž1992. 454–462. J.M. Conner, S. Varon, Distribution of nerve growth factor-like immunoreactive neurons in the adult rat brain following colchicine treatment, J. Comp. Neurol. 326 Ž1992. 347–362. J.M. Conner, S. Varon, Nerve growth factor influences the distribution of sprouting sympathetic terminals in the hippocampal formation following cholinergic deafferentation, Exp. Neurol. 130 Ž1994. 15–23. J.M. Conner, S. Varon, Effects of exogenous nerve growth factor upon sympathetic sprouting into the hippocampal formation, Exp. Neurol. 136 Ž1995. 123–135. C.W. Cotman, D.A. Matthews, D. Taylor, G.S. Lynch, Synaptic rearrangemtent in the dentate gyrus: histochemical evidence of adjustments after lesions in immature and adult rats, Proc. Natl. Acad. Sci. USA 70 Ž1973. 3473–3477. B. Crain, C. Cotman, D. Taylor, G.A. Lynch, A quantitative electron microscopic study of synaptogenesis in the dentate gyrus of the rat, Brain Res. 63 Ž1973. 195–204. C. Crowley, S.D. Spencer, M.C. Nishimura, K.S. Chen, S. PittsMeek, M.P. Armanini, L.H. Ling, S.B. McMahon, D.L. Shelton, A.D. Levinson, H.S. Phillips, Mice lacking nerve growth factor display perinatal loss of sensory and sympathetic neurons yet develop basal forebrain cholinergic neurons, Neuron 76 Ž1994. 1001– 1011. A.M. Davies, Molecular and cellular aspects of patterning sensory neurone connections in the vertebrate nervous system, Development 101 Ž1987. 185–208. A.M. Davies, C. Bandtlow, R. Heumann, S. Korsching, H. Rohrer, H. Thoenen, Timing and site of nerve growth factor synthesis in developing skin in relation to innervation and expression of the receptor, Nature 326 Ž1987. 353–358. B.H. Gahwiler, A. Enz, F. Hefti, Nerve growth factor promotes development of the rat septohippocampal cholinergic projection in vitro, Neurosci. Lett. 75 Ž1987. 6–10. B.H. Gahwiler, F. Hefti, Guidance of acetylcholinesterase-containing fibers by target tissue in co-cultured brain slices, Neuroscience 13 Ž1984. 681–689. H. Gnahn, F. Hefti, R. Heumann, M. Schwab, H. Thoenen, NGFmediated increase of choline acetyltransferase ŽChAT. in the neonatal forebrain: evidence for a physiological role of NGF in the brain?, Dev. Brain Res. 9 Ž1983. 45–52.

J.M. Conner, S. Varonr DeÕelopmental Brain Research 101 (1997) 67–79 w20x F. Gomez-Pinilla, C.W. Cotman, M. Nieto-Sampedro, NGF receptor immunoreactivity in the rat brain: topographic distribution and response to entorhinal ablation, Neurosci. Lett. 82 Ž1987. 260–266. w21x T. Hagg, H.L. Vahlsing, M. Manthorpe, S. Varon, Nerve growth factor infusion into the denervated adult rat hippocampal formation promotes its cholinergic reinnervation, J. Neurosci. 10 Ž1990. 3087– 3092. w22x T. Hagg, S. Varon, Neurotropism of nerve growth factor for adult rat septal cholinergic axons in vivo, Exp. Neurol. 119 Ž1993. 37–45. w23x S. Heckers, T. Ohtake, R.G. Wiley, D.A. Lappi, C. Geula, M.-M. Mesulam, Complete and selective cholinergic denervation of rat neocortex and hippocampus but not amygdala by an immunotoxin against the p75 NGF receptor, J. Neurosci. 14 Ž1994. 1271–1289. w24x F. Hefti, Nerve growth factor promotes survival of septal cholinergic neurons after fimbrial transections, J. Neurosci. 6 Ž1986. 2155–2162. w25x D.M. Holtzman, J. Kilbridge, Y. Li, E.T. Cunningham, N.J. Lenn, D.O. Clary, TrkA expression in the CNS: evidence for the existence of several novel NGF-responsive CNS neurons, J. Neurosci. 15 Ž1995. 1567–1576. w26x P. Kasa, The cholinergic systems in brain and spinal cord, Prog. Neurobiol. 26 Ž1986. 211–272. w27x S. Koh, R. Loy, Localization and development of nerve growth factor-sensitive rat basal forebrain neurons and their afferent projections to hippocampus and neocortex, J. Neurosci. 9 Ž1989. 2999– 3018. w28x S. Korsching, G. Auburger, R. Heumann, J. Scott, H. Thoenen, Levels of nerve growth factor and its mRNA in the central nervous system of the rat correlate with cholinergic innervation, EMBO J. 4 Ž1985. 1389–1393. w29x T.H. Large, S.C. Bodary, D.O. Clegg, G. Weskamp, U. Otten, L.F. Reichardt, Nerve growth factor gene expression in the developing rat brain, Science 234 Ž1986. 352–355. w30x G.S. Lynch, D.A. Matthews, S. Mosko, T. Parks, C.W. Cotman, Induced acetylcholinesterase-rich layer in the dentate gyrus following entorhinal lesions, Brain Res. 42 Ž1972. 311–318. w31x P.L. McGeer, E.G. McGeer, Cholinergic systems and cholinergic pathology, in: A. Lajtha ŽEd.., Handbook of Neurochemistry. Vol. 6. Receptors in the Nervous System, Plenum Press, New York, 1984, pp. 379–410.

79

w32x T.A. Milner, R. Loy, D.G. Amaral, An anatomical study of the development of the septo-hippocampal projection in the rat, Dev. Brain Res. 8 Ž1983. 343–371. w33x W.C. Mobley, J.L. Rutkowski, G.I. Tennekoon, J. Gemski, K. Buchanan, M.V. Johnston, Nerve growth factor increases choline acetyltransferase activity in developing basal forebrain neurons, Mol. Brain Res. 1 Ž1986. 53–62. w34x N.T. Naik, Technical variations in Koelle’s histochemical method for demonstrating cholinesterase activity, Quart. J. Microsc. Sci. 104 Ž1963. 89–100. w35x M. Seiler, M.E. Schwab, Specific retrograde transport of nerve growth factor ŽNGF. from neocortex to nucleus basalis in the rat, Brain Res. 300 Ž1984. 33–39. w36x D.L. Shelton, L.F. Reichardt, Studies on the expression of the b-nerve growth factor ŽNGF. 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. USA 83 Ž1986. 2714–2718. w37x R.J. Smeyne, R. Klein, A. Schnapp, L.K. Long, S. Bryant, A. Lewin, S.A. Lira, M. Barbacid, Severe sensory and sympathetic neuropathies in mice carrying a disrupted TrkrNGF receptor gene, Nature 368 Ž1994. 246–249. w38x E.E.M. Van der Zee, J. Fawcett, J. Stanisz, R. Racine, J. Diamond, Anti-NGF treatment blocks the collateral sprouting of cholinergic fibers in the hippocampus, J. Comp. Neurol. 326 Ž1992. 91–100. w39x S.R. Whittemore, T. Ebendal, L. Larkfors, L. Olson, A. Seiger, I. Stromberg, H. Persson, Developmental and regional expression of b-nerve growth factor messanger RNA and protein in the rat central nervous system, Proc. Natl. Acad. Sci. USA 83 Ž1986. 817–821. w40x L.R. Williams, K.S. Jodelis, M.R. Donald, Axotomy-dependent stimulation of choline acetyltransferase activity by exogenous mouse nerve growth factor in adult rat basal forebrain, Brain Res. 488 Ž1989. 243–266. w41x C.S. Yu, K.A. Crutcher, Nerve growth factor immunoreactivity and sympathetic sprouting in the rat hippocampal formation, Brain Res. 672 Ž1995. 55–67.