- Email: [email protected]
Development of cholinergic markers in mouse forebrain. II. Muscarinic receptor binding in cortex
Developmental Brain Research, 23 (1985) 243- 253 Elsevier
243
BRD 50299
Development of Cholinergic Markers in Mouse Forebrain. II. Muscarinic Receptor Binding in Cortex CHRISTINE F. HOHMANN 1., CANDACE C. PERT2and FORD F. EBNER 1
1Division of Biology and Medicine, Brown University, Providence, R102912 and 2Biol. Psychiatry Branch, NIMH, Bethesda, MD20205 (U.S.A.) (Accepted May 21st, 1985)
Key words: [3H]propylbenzilylcholinemustard - - muscarinic cholinergie ligand - - mouse - - forebrain development
The distribution of muscarinic receptor sites throughout the ontogeny of cerebral cortex in the BALB/c mouse have been labeled, placing special emphasis on binding site development in parietal neocortex and hippocampus. We describe a new procedure for the use of [3H]propylbenzilylcholinemustard as a muscarinic cholinergic ligand in an in vitro binding assay on brain sections. Muscarinic binding sites, as visualized by autoradiography, can be seen in cortex as early as embryonic day 18. They achieve maximal labeling density and adult distribution in neocortex by the end of the first postnatal month. The adult distribution pattern in hippocampus is reached by the second postnatal week, but the maximal density of label is not achieved until 4 weeks of age. Changes in the receptor binding pattern are illustrated at 5 different ages between birth and adulthood. We conclude that muscarinic cholinergic receptors develop late in cortical ontogeny as do other cholinergic markers. The distribution pattern of muscarinic binding sites in mouse cortex is puzzling because it does not correspond to the reported distribution of cells physiologicallyresponsive to applied acetylcholine. These results are compared to the onset of choline acetyltransferase activity and acetylcholine esterase staining. The ontogenesis of the cortical cholingergic system is compared with other features of general cortical morphogenesis.
INTRODUCTION The development of the cortical cholinergic system has recently received renewed attention due in part to its possible involvement in the regulation of cortical plasticity22-25. Different aspects of cholinergic ontogenesis in cerebral cortex have been studied extensively both with biochemical and with histological methods~ ,2,6,12,14,17,18,21,26. In a previous studylO,a~ we began the investigation of cholinergic marker d e v e l o p m e n t in mouse cerebral cortex by describing choline acetyltransferase (CHAT) activity levels and the pattern of acetylcholinesterase (ACHE) histochemistry at various ages between late gestation and maturity. In the present study, we have monitored the ontogeny of another cholinergic marker in mouse cerebral cortex, the muscarinic cholinergic receptor. Muscarinic cholin-
ergic receptors appear to be the physiologically dominant cholinergic receptor species in m a m m a l i a n cerebral cortex~5,16 and thus are important for producing cholinergic effects in cortex. Detailed knowledge of muscarinic receptor site distribution during ontogeny is essential for the understanding of cholinergic influences during the development of cortical circuitry. In the present study we describe an autoradiographic technique which permits the use of the covalently binding muscarinic ligand [3H]Propylbenzilylcholine mustard, in an in vitro binding assay on unfixed brain sections. MATERIALS AND METHODS Female BALB/c mice with timed pregnancies were supplied by the breeding colony of the National Cancer Institute and their pups were sacrificed on
* Present address: The Johns Hopkins University School of Medicine, Department of Child Psychiatry, 725 North Wolfe St., Baltimore, MD 21205, U.S.A. Correspondence: F.F. Ebner, Division of Biology and Medicine, Brown University, Providence, RI 02912, U.S.A.
244
embryonic day 18 (El8) or postnatal day (P) 1,4, N, 14, 31, and 60 (considered adult). El8 was defined as the eighteenth day after vaginal plug detection and P1 was defined as the day of birth. Embryos were removed by Caesarean section and killed by decapitation. Postnatal animals were sacrificed by cervical dislocation prior to decapitation. All brains were removed quickly and frozen in isopentane for 20-30 s at -25 to -35 °C. Four brains were prepared at every age point. Brain sections were processed in 4 separate experiments, each containing one brain from every age group. After the binding assay, two brains at each age point were processed for autoradiography on LKB tritium-sensitive Ultrofilm. The remaining tissue sections were dipped in a solution of nuclear track emulsion (NTB-2 Kodak). Some of the emulsion-dipped slides and all of the slides used for LKB film autoradiography were subsequently counterstained with cresyl violet or with Neutral red. Muscarinic receptor binding was performed with [3H]propylbenzilylcholine mustard ([3H]PrBCM) supplied by New England Nuclear. This irreversibly binding cholinergic ligand has been studied extensively by Burgen et al 4 and has been characterized as a specific muscarinic receptor binding agent. The muscarinic antagonist atropine was used as an inhibitor of radioactive ligand binding to measure non-specific binding. The autoradiographic binding assay was performed according to a method developed by HOhmann and Pert, based on the procedures used by Rotter er al. 20 for muscarinic receptor binding, and by Herkenham and Pert s for opiate receptor binding. Briefly, unfixed frozen tissue was sectioned in a cryostat at -15 °C into 32ktm thick coronal sections. Sections were collected onto cold slides, melted to the slide by briefly touching the back of the slide, dried under vacuum in the cold (5-10 °C) and subsequently stored at -70 °C. For the receptor binding assay, slides were preincubated at room temperature with either Krebs buffer or Krebs buffer containing 10-5 M atropine. The Krebs buffer was freshly prepared for every experiment and gassed with a mixture of 95% CO, and 5% O2 prior to use. The ionic composition of this buffer was: sodium chloride 118 mM, potassium chloride 4.7 mM, calcium chloride 2.6 mM, potassium phosphate monobasic 1.2 mM, magnesium sulfate 1.2 raM, sodium bicarbonate 25 mM and dextrose 11 mM. After 15 min of preincubation
[-~H]PrBCM (previously cyclized a! room temperature in 10 mM phosphate buffer, pH 7.4. a~, a c~mcentration of 10-~ M) was added al a concentration of 2.4 nM and the slides were incubated for 45 rain ~tt room temperature. Subsequent to the incubation, sections were quickly washed in 6 rinses of icc-cokt Krebs buffer, dried well in a cold air stream, fixed with formaldehyde vapors at 8(1 °C, defatted in xylenes and processed for autoradiography (see ref. g). Sections used for scintillation counting were scraped off the slides into scintillation vials, strongly agitated in Aquasol liquid scintillation fluid (NEN) and counted. RESULTS
The binding technique The binding procedure used in this study was modified in several ways from that of Rotter et at. in an attempt to analyze the receptor distribution in untreated, unfixed tissue. Therefore several conditions of the assay had to be reassessed. The [3H]PrBCM concentration of 2.4 nM was adopted unchanged from the PrBCM concentrations used by Rotter et al. 20. However, unlike the Rotter et al. protocol, the tissue used in our study was both unfixed and sectioned at 3 times the thickness (32 ,urn). These changes were likely to have altered the temperature and time-dependency of the binding process. For the determination of optimal assay conditions sections were scraped off the slides after the incubation and counted in scintillation fluid. The optimal incubation conditions at room temperature were achieved at 45 rain, as can be seen in Fig. 1. We ensured a constant high concentration of atropine in the control incubations during this extended incubation time by increasing atropine to concentrations 10 times higher than those reported by Rotter et al. 20. Incubations at temperatures warmer than 30 °C resulted in inferior histological preservation. Incubation at temperatures lower than 20 °C resulted in higher ratios of non-specific to specific binding. Ratios of specific to non-specific binding, when measured in dried, non-defatted sections subsequent to the assay, revealed a specific binding of only approximately 40%. However, a ratio of approximately 70% specific versus non-specific binding can be achieved (see Fig. 1) when the tissue is fixed and defatted after the receptor binding proce-
245 75
¢3 Z
times more measurable cpm than total binding after such t r e a t m e n t of the tissue. Thus more non-specifically b o u n d ligand than specifically b o u n d ligand is removed in the defatting process, indicating that some of the sites for non-specific binding are soluble in the alcohols or in xylenes.
65
LI.
~45 U.I
A utoradiography
CO
Muscarinic receptor sites in mouse cerebral cortex and hippocampus, as visualized by a u t o r a d i o g r a p h y , a p p e a r before birth and increase in density for several weeks after birth. Early in d e v e l o p m e n t silver grains are seen evenly distributed over the densely p a c k e d undifferentiated cells in cortex and hippocampus. With progressing differentiation silver grains are seen p r e d o m i n a n t l y in the neuropil between cells (Fig. 2B). This tendency of the silver grains to a p p e a r in higher density in neuropil areas versus over cell bodies can be seen especially well in the hippocampus (Fig. 2A). The overall pattern of muscarinic receptor sites shows surprisingly little differential distribution at all ages investigated. Neocortex and the h i p p o c a m p a l formation are the cortical areas with the clearest p a t t e r n of binding and we
~ 35
J 15
30
45
6 0 MIN.
Fig. 1. Time-dependency of specific [3H]PrBCM binding to adult brain sections at room temperature. Closed circles show specific binding before defaning of the tissue; open circles show specific binding after defatting. Percent specific binding was calculated by subtracting the cpm of atropine-blocked controls from total cpm and expressing the difference as percent of total counts. This figure shows the results from a typical experiment. Each point represents the mean value for 3 different coronal sections through mouse forebrain. dure ( H 2 0 ; E t O H 70%, 95%, 100%; xylenes 2 × ; E t O H 100%, 95%, 70%; H 2 0 , 5 min each). Total binding before fixation and defatting shows 5 - 1 0
~z~i~ i~~~iiI
Fig. 2. Coronal sections through adult mouse forebrain showing silver grain distribution in the dentate gyrus of the hippocampus (A). The density of silver grains in the dentate molecular layer (upper left cellsparse area) is greater than over the granule cell layer. In layer VI of somatosensory cortex (B) the silver grains are distributed relatively evenly throughout the field of view.
246
Fig. 3. Dark-field micrographs illustrating the postnatal development of muscarmic receptors in hippocampus, cingulatc, ~md medial parietal cortex. The autoradiographs for this figure were prepared by the emulsion dipping technique. The ages shown art:: ,\, P l: t-H P4; C, PS: D, PI4; E, P31; cp, cortical plate; g, granule cell layer: la too, stratum molecularc lacunosum; m, molecular tascr
will focus primarily on these areas in our subsequent description of muscarinic receptor development. Neocortex
Silver grains at levels above background can be detected in all areas of immature neocortex as early as E l 8 , though tissue preservation at this early age is not good enough to determine whether any patterned binding exists. By P1 a moderate n u m b e r of silver grains are distributed homogeneously throughout the cortex. The cortical layer I~ however, shows a noticeably higher grain density, separated from the
rest of cortex by a relatively grain-free zone which occupies lower parts of layer I. This sparsely labeled zone extends slightly into the upper cortical plale (see Fig. 3A). By P4, the grain density increases in subptate layers and in layer I. The latter represents the most intensely labeled part of cortex at this age. A zone of relatively low grain density extends from lower laver 1 throughout the cortical plate. Labeling of receptor sites is at its lowest over the cortical plate (see Fig. 4A, B). At P8 the overall silver grain pattern does not differ very much from the pattern at postnatal
247
Fig. 4. Bright-field micrographs of autoradiographs of [3H]PrBCM binding on LKB Ultrofilm (B,D) compared with the same coronal section after Nissl staining (A,C). A and B show somatosensory cortex of a P4 animal; C and D show somatosensory cortex of a PI4 animal. Note the increase in labeling in the region of layers IV through II from P4 to P14 and the corresponding decrease of labeling in layer I.
248
Fig. 5. Adult somatosensory cortex: bright-field micrograph of an autoradiograph of [3H]PrBCM binding on LKB Uhrolitm (B) compared to the same section after Nissl stain (A). Compare to Fig. 4D and note the relative increase of labeling in cortical laxcrs II and 111 between P14 and adult and the further decrease in silver grains in layer I of cortex. day 4, but the overall intensity of the label has increased. By P14 the double band of silver grains, characteristic of the adult pattern, begins to emerge. The overall intensity of label has markedly increased in all layers of cortex except for a zone of low grain density in upper layer V. The label in supragranular layers, however, still has not reached the same density as labeling in the infragranular layers and is especially low at the border between layer II and 1. The upper tier of layer I remains heavily labeled (see Fig. 4C, D). Towards the end of the fourth postnatal week, the labeling decreases considerably in layer I of neocortex. The grain density in cortical layers IV, lII and II is comparable at this age to layers VI and lower V. The adult mouse neocortex shows two layers of comparable grain density separated by a less densely labeled zone which occupies upper layer V (see Figs. 5 + 6). This relatively grain-sparse zone can be seen in all cortical areas which display a distinct pyramidal cell layer (see Fig. 6 A - C ) . Anteriorly, this zone extends beyond the rhinal fissure and into perirhinal
cortex. More posteriorly, where a distinct pyramidal cell layer cannot be detected lateral to sensory cortex, the sparsely labeled zone only appears in neocortical areas• The layer I band, so prominent at earlier ages, disappears entirely in adulthood and silver grain densities near the pial surface actually decrease to levels below those of deeper layers (see Fig. 5). LKB autoradiographs from adult animals show a lower labeling intensity than autoradiographs from P31 on the same LKB film. Thus overall grain density apparently decreases between the fourth postnatal week and adulthood.
Hippocampus The hippocampus is still very immature at Pl and consists essentially of one row of densely packed immature pyramidal cells. Silver grains can be seen in thin bands lining both sides of this immature pyramidal cell layer (see Fig. 3A). Caudal to the anterior commissure, the hippocampal cell layer still appears continuous with the cingulate pyramidal cell laver and silver grains form two uninterrupted bands
249
Fig. 6. Images of coronal sections from anterior (A) to posterior (D) through the forebrain of an adult mouse. Sections A-C show bright-field autoradiographs on LKB Ultrofilm of total binding (sections incubated with [3H]PrBCMonly). D shows a control section incubated in the presence of atropine. Note the banding pattern in cortex and the differential labeling in various nuclei of the basal forebrain and diencephalon in A-C. The non-specificbinding on the control section (D) is unpatterned.
above and below these cells. By P4, label can be seen in the developing stratum oriens and stratum radiatum and, to a much lesser extent, overlying the pyramidal cell layer. The granule cell layer of the dentate gyrus has become apparent at this age and is free of label, being outlined by the silver grains in the molecular layer and in the hilus region (see Fig. 3B). By the end of the first postnatal week, the intensity of the label has increased in the whole hippocampal formation. The increase of silver grains in the molecular layer of the dentate gyrus and in the stratum lacunosum moleculare is proportionally larger than in the stratum oriens and stratum radiatum. The pyramidal cell layer shows higher grain density than the dentate granule cell layer; both cell layers, however, are far
less intensely labeled than the hippocampal white matter. Bands of relatively lower grain density appear in the polymorph layer and in the stratum moleculare along its border with the stratum radiatum (Fig. 3C). Except for an overall increase in grain density, no differences can be seen between the one- and two-week-old hippocampal formation (Fig. 3D). In the adult animal the hippocampal formation appears more lightly labeled than in the two-week-old animal, but the pattern has not changed qualitatively very much from that of the 8-day-old animal. The lightly labeled pyramidal cell layer is outlined by an intensly labeled stratum oriens and stratum radiatum. Below the stratum radiatum a zone of lower grain density marks the border of the stratum moleculare. Ventrally, in the dentate gyrus, the grain den-
250 sity is increased again, remains high throughout the dentate molecular layer, and drops to almost background levels over the dentate granule cell layer. Grain densities in the stratum lucidum and polymorphe are only slightly higher than grain densities in the pyramidal cell layer (see Figs. 3E and 6B, C). O t h e r cortical areas
Cingulate cortex retains essentially the same labeling pattern throughout development. A band of high density overlies layer I and is continuous with the neocortical layer 1 band. Silver grains are low over the granule cells in layer I1 of cingulate cortex, but below this cell layer grain density is increased and remains uniformly distributed down to the level of the white matter. Grain density below layer II of cingulate cortex always parallels the grain density observed in layer IV of neocortex (Figs. 3 and 6). The only noticeable change in the cingulate labeling pattern is a decrease in intensity of the layer I band which occurs concomitantly with the decrease in neocortical layer I labeling (see Fig. 3). Paleocortical areas show no differentiated silver grain pattern at any point in development. These areas are characterized by a slow increase in labeling intensity to adult levels, at which time their grain density is comparable to layer VI labeling in neocortex (see Figs. 3 and 6). DISCUSSION These results provide a detailed description of muscarinic cholinergic receptor development and distribution in the cerebral cortex of the mouse. They show that some muscarinic receptor binding can be detected in cortex as early as embryonic day 18. A rapid increase in labeling intensity can be seen during the first week of postnatal development in all cortical areas. Noticeable changes in the binding pattern appear in both the neocortex and the hippocampal formation during the first and second week of life. Complete maturity of the binding distribution is reached shortly after the fourth postnatal week. Silver grain densities decrease slightly between 4 weeks and adulthood. While there are clear differences in silver grain density between various cortical layers, silver grains always show a fairly homogeneous tissue distribution if viewed at high magnification. At ages P4
and older the silver grains are distributed cxcnl} throughout the neuropil and over cells, so thai tocali zation over individual cell bodies is not apparent a any age. In the present paper we have introduced a m o d ified technique for radioactive labeling of muscarinic cholinergic receptors on tissue sections. This proce. dure has several advantages over other method,' presently used to visualize muscarinic cholinergic receptors. Tritiated PrBCM is an irreversibly bmdin~ ligand and thus allows extensive post mcubatior washing of the tissue without ligand remowd or diffu: sion within the tissue during autoradiographic processing. The binding properties of PrBCM also allox~ defatting of the brain sections after formalin vapol fixation and prior to autoradiography. This represents a significant improvement over the QNB binding techniques in view of the recent growing awarehess of lipid quenching in autoradiographic processing 9. Tritiated PrBCM has been utilized successful b by Rotter er al. >, 21 to visualize autoradiographic receptor distribution in developing and adult rat brains. The tissue used in these studies was lightly fixed prior to the ligand incubation in order to assure sufficienl tissue preservation, and the animals were anesthetized before brain removal. Using methods recently developed by Herkenham and Pert for opiate receptor binding s, we have been able to perform our binding studies on entirely untreated, unfixed fresh tissue. These conditions have allowed us to detect higher percentage of muscarinic receptors that were present in the live animal than previous studies. 13otlfixation and threatment of the animals with anesthesia could potentially alter the binding properties ol receptors in the brain. We have taken two slightly different approaches t~ monitoring the autoradiographic distribution of muscarinic receptors in cerebral cortical development, Autoradiographs prepared according to the liquic emulsion dipping technique and LKB Ultrofilm autoradiographs complemented each other very well i~ their ability to show developmental changes. While the former technique is indispensable fl:~r the accurate identification of silver grain location, the latter technique allows better tissue preservation, frequently provides sharper images of distribution patterns, and also facilitates low power bright-field visualization of whole brain sections.
251 Our results concerning the distribution and ontogenesis of muscarinic cholinergic binding sites in neocortex are generally in good agreement with previously published results. Kuhar and Yamamura ~3 described an autoradiographic distribution pattern of in vivo [3H]QNB binding in the adult rat neocortex which closely resembles the present findings in the mouse: two strata of similarly high-grain density are separated by a low-density band in upper layer V. Wamsley et al. 27 studied both total and high-versus low-affinity muscarinic receptor patterns in rat brain. Their visualization of total muscarinic binding, performed with [3H]methyl-scopolamine, also shows the above-mentioned pattern in adult neocortex. A number of biochemical in vitro assessments of muscarinic receptor development in mouse and rat cerebral cortex also show the general time-course of muscarinic receptor ontogeny l, 2.6, 14. Thus, in vitro and in vivo binding techniques w!th a variety of different muscarinic cholinergic ligands lead to similar conclusions. Several differences do exist, however, between our results, gained with the procedure outlined above, and those reported by Rotter et al. 2~. To our knowledge these studies report the only other description of muscarinic receptor distribution in rodent cerebral cortex during the early postnatal period. They also illustral~e the transiently heavy labeling in layer I between P1 and P13, and it is noteworthy that the muscarinic receptors are at their densest during the postnatal period when the non-specific thalamocortical fibers are forming their initial synapses in this stratum and the AChE staining is only starting to develop. In our study the grain density at each depth in cortex follows the gradient of neuronal differentiation; i.e. the pattern in layer VI and V develops before the pattern in layers IV, III and II. Finally, the width of the grain sparse band that we see only in the upper half of layer V, extends throughout layers V and IV in the rat 2~. Both the Rotter et al. and the present study used [3H]PrBCM for the receptor binding assay, but the absence of anesthesia and fixation in our study may have preserved more of the receptors. Alternatively, it is possible that the ontogeny of muscarinic receptors is different in mouse and rat. Such a species difference has been noticed in the appearance of AChE in the hippocampus (see refs. 10, 11. The distribution of muscarinic cholinergic binding
sites in the adult rodent neocortex does not correspond precisely with the density profiles of other cholinergic markers. While receptors can be found in relatively high densities in all layers of cortex, except upper layer V and I, AChE is most prominent in upper layer VI and IV and 110-12. The number of cells responsive to iontophoretically applied ACh is largest in deep cortical layers TM 16. Such incongruities between the muscarinic cholinergic receptor pattern and the distribution of other cholinergic properties may become more interpretable once the physiological significance of different receptor subclasses is better understood. Muscarinic cholinergic receptor subclasses do apparently bind antagonists with comparable affinity, but differ in their affinity for agonists and are probably not all of equal physiological importance (see ref. 7 for rewiev). Wamsley et al. 27 provided evidence for the differential distribution of these different receptor subtypes in neocortex. The high agonist affinity muscarinic receptor type, for example, is present in the highest density in cortical layer IV followed by layers V and VI, and much lower levels of this receptor type are present in the supragranular layers. It is unresolved whether all muscarinic receptor subtypes in the CNS are capable of inducing membrane resistance changes and whether all are located on neurons. In conclusion, maturation of muscarinic cholinergic receptors in mouse cerebral cortex occurs late in cortical development. Muscarinic receptor ontogeny thus parallels the late development of other cholinergic markers. A low density of receptors, however, are present in cortex before the appearance of any other cholinergic markers so that receptors are found at times when ChAT activity and presumably ACh levels still are very low. Early postnatal cerebral cortex thus has relatively high levels of AChE and muscarinic receptors but low levels of ACh synthesis 6, 10,11. Since both muscarinic receptors and AChE staining show patterns of differential distribution, this relationship may be conducive to very localized effects of ACh on postsynaptic cells. In areas of high AChE concentration, presumably only receptors very close to the release site would be able to bind ACh released in cortex before it is degraded. This situation appears very different from the adult cerebral cortex where ACh seems capable of diffusion over quite some distance 5,19. It is noteworthy, that both
252 the muscarinic receptor distribution and the A C h E
of the cholinergic system in development may b~
pattern develop from deeper to more superficial cor-
very different from the cholinergic influence on tht
tical layers, yet both show labeling of layer ! almost
adult cerebral cortex, but if so, this function rcmain~
immediately after their first appearance in layer VI.
to be elucidated.
In addition, the time-course of pattern development is remarkably similar for A C h E and muscarinic re-
ACKNOWLEDGEMENTS
ceptors. As m e n t i o n e d in a previous report, these oncorrespond
A preliminary report of the present data has bee~
closely to strata of afferent synapse development 3, ~,
given in Neuroscience Abstracts, Vol. 9, 1983. The
togenic
cholinergic marker
patterns
u. The remarkable changes in patterns in all cholin-
authors would like to thank Dr. W. Bowen for his
ergic markers, which often closely parallel other de-
helpful suggestions in the development of the tech-
velopmental events, suggest that the cholinergic sys-
nique, and Dr. S. Moon-Edley for suggestions thal
tem may play a specific role in cerebral cortical morphogenesis, including synaptogenesis. The function
ported by the N I M H and NIH G r a n t NS13031.
REFERENCES 1 Aronstam, R.S., Kellogg, C. and Abood, L.G., Development of muscarinic cholinergic receptors in inbred strains of mice: Identification of receptor heterogeneity and relation of audiogenic seizure susceptibility, Brain Res., 162 (1979) 231-241. 2 Ban-Barak, J. and Dudai, Y., Cholinergic binding sites in rat hippocampal formation: properties and ontogenesis, Brain Res., 166 (1979) 245-257. 3 Blue, M.E. and Parnavales, J.G., The formation and maturation of synapses in the visual cortex of the rat: It. Quantitative Analysis, J. Neurocytol., 12 (1983) 697-712. 4 Burgen, A.S.V., Hiley, C.R., Young, J.M., The properties of muscarinic receptors in mammalian cerebral cortex, Br. J. Pharmacol., 51 (1974) 279-285. 5 Collier, B. and Mitchell, J.F., The central release of acetylcholine during stimulation of the visual pathway, J. Physiol. (London), 184 (1966) 239-254. 6 Coyle, J.T. and Yamamura, H.I., Neurochemical aspects of the ontogenesis of cholinergic neurons in the rat brain, Brain Res., 118 (1976) 429-440. 7 Ehlert, F.J., Roeske, W.R. and Yamamura, H.I., Muscarinic cholinergic receptor heterogeneity, Trends Neurosci., 5 (1982) 336-339. 8 Herkenham, M. and Pert, C.B., Light microscopic localization of brain opiate receptors: a general autoradiographic method to preserve tissue quality, J. Neurosci., 2 (1982) 1129-1149. 9 Herkenham, M. and Sokoloff, L., Quantitative receptor autoradiography: tissue defatting eliminates differential self-absorbation of tritium radiation in gray and white matter of brain, Brain Res., 321 (1984) 363-368. 10 H6hmann, C.F. and Ebner, F.F., The development of cholinergic markers in normal and transplanted mouse neocortex, Neurosei. Abstr., 8 (1982) 865. 11 H6hmann, C.F. and Ebner, F.F., Development of cholinergic markers in mouse forebrain. I. Choline acetyltransferase enzyme activity and acetylcholinesterase histoehemistry, Dev. Brain Res., 23 (1985) 225-241. 12 Kristt, D.A., Development of neocortical circuitry: histochemical localization of AChE in relation to the cell layers
improved the manuscript. This work has been sup-
of rat somatosensory cortex, J. Comp. Neurol.. 186 (1979) 1-17, 13 Kuhar, M.J. and Yamamura, H.I., Localization of muscarinic receptors in rat brain by light microscopic radioautography, Brain Res., 110 (t976) 229-243. 14 Kuhar, MJ., Birdsall, N.J.M., Burgen, A.S.V. and Hulme, E.C., Ontogeny of muscarinic receptors in rat brain, Brain Res., 184 (1980) 375-383. 15 Krnjevic, K. and Phillis, J.W., Acetylcholine sensitive celb in cerebral cortex, J. Physiol. (London), 166 (19631 296-327. 16 Lamour, Y., Dutar, P. and Jobert, A., Excitatory effect ol acetylcholine on different types of neurons in the first somatosensory neocortex of the rat: laminar distribution and pharmacological characteristics, Neuroscience. 7 (1982) 1483-1494. 17 Matthews, D.A,, Nadler, J.V., Lynch, G.S. and Cotman, C.W., Development of cholinergic innervation in the hip: pocampal formation of the rat. I. Histochemical demonstration of AChE activity, Dev. Biol., 36 (1974) 130-141. 18 Nadler, J.V., Matthews, D.A., Cotman, C.W. and Lynch, G.S., Development of the cholinergic innervation of the hippocampal formation of the rat: II. Quantitative changes in choline acetyltransferase and acetylcholinesterase activities, Dev. Biol., 36 (1974) 142-154. 19 Phillis, J.W., Acetylcholine release from the cerebral cortex: its role in cortical arousal, Brain Res.. 7 (1968) 378-389. 20 Rotter, A., Birdsall, N.J.M., Burgen, A.S.V., Field, P.M., Hulme, E.C. and Raisman, G., Muscarinic receptors in the central nervous system of the rat. I. Technique for autoradiographie localization of the binding of [3H]propylbenzilylcholine mustard and its distribution in the forebrain, Brain Res. Rev., 1 (1979) 141-165. 21 Rotter, A., Field, P.M. and Raisman, G., Muscarinic receptors in the central nervous system of the rat: postnatal development of binding of [3H]propylbenzilylcholinemustard, Brain Res. Rev., 1 (1979) 185-205. 22 Sillito, A.M., Plasticity in the visual cortex. Nature (London), 303 (1983) 477-478. 23 Singer, W., Central core control of visual cortex function. 1. The Neurosciences. Fourth Study Program, MIT Press,
253 Cambridge, MA 1979. 24 Singer, W., Central core control of developmental plasticity in the kitten visual cortex. I. Diencephalic lesions, Exp. Brain Res., 47 (1982) 209-222. 25 Singer, W. and Rauschecker, J.P., Central core control of developmental plasticity in the kitten visual cortex. II. Electrical activation of mesencephalic and diencephalic projections, Exp. Brain Res., 47 (1982) 223-233.
26 Sorimachi, M., Mijamoto, K. and Katakoa, K., Postnatal development of choline uptake by cholinergic nerve terminals, Brain Res., 79 (1974) 343-346. 27 Wamsley, J.K,, Zarbin, M.A., Birdsall, N.J.M and Kuhar, M.J., Muscarinic cholinergic receptors: autoradiographic localization of high and low affinity agonist binding sites, Brain Res., 200 (1980) 1-12.
243
BRD 50299
Development of Cholinergic Markers in Mouse Forebrain. II. Muscarinic Receptor Binding in Cortex CHRISTINE F. HOHMANN 1., CANDACE C. PERT2and FORD F. EBNER 1
1Division of Biology and Medicine, Brown University, Providence, R102912 and 2Biol. Psychiatry Branch, NIMH, Bethesda, MD20205 (U.S.A.) (Accepted May 21st, 1985)
Key words: [3H]propylbenzilylcholinemustard - - muscarinic cholinergie ligand - - mouse - - forebrain development
The distribution of muscarinic receptor sites throughout the ontogeny of cerebral cortex in the BALB/c mouse have been labeled, placing special emphasis on binding site development in parietal neocortex and hippocampus. We describe a new procedure for the use of [3H]propylbenzilylcholinemustard as a muscarinic cholinergic ligand in an in vitro binding assay on brain sections. Muscarinic binding sites, as visualized by autoradiography, can be seen in cortex as early as embryonic day 18. They achieve maximal labeling density and adult distribution in neocortex by the end of the first postnatal month. The adult distribution pattern in hippocampus is reached by the second postnatal week, but the maximal density of label is not achieved until 4 weeks of age. Changes in the receptor binding pattern are illustrated at 5 different ages between birth and adulthood. We conclude that muscarinic cholinergic receptors develop late in cortical ontogeny as do other cholinergic markers. The distribution pattern of muscarinic binding sites in mouse cortex is puzzling because it does not correspond to the reported distribution of cells physiologicallyresponsive to applied acetylcholine. These results are compared to the onset of choline acetyltransferase activity and acetylcholine esterase staining. The ontogenesis of the cortical cholingergic system is compared with other features of general cortical morphogenesis.
INTRODUCTION The development of the cortical cholinergic system has recently received renewed attention due in part to its possible involvement in the regulation of cortical plasticity22-25. Different aspects of cholinergic ontogenesis in cerebral cortex have been studied extensively both with biochemical and with histological methods~ ,2,6,12,14,17,18,21,26. In a previous studylO,a~ we began the investigation of cholinergic marker d e v e l o p m e n t in mouse cerebral cortex by describing choline acetyltransferase (CHAT) activity levels and the pattern of acetylcholinesterase (ACHE) histochemistry at various ages between late gestation and maturity. In the present study, we have monitored the ontogeny of another cholinergic marker in mouse cerebral cortex, the muscarinic cholinergic receptor. Muscarinic cholin-
ergic receptors appear to be the physiologically dominant cholinergic receptor species in m a m m a l i a n cerebral cortex~5,16 and thus are important for producing cholinergic effects in cortex. Detailed knowledge of muscarinic receptor site distribution during ontogeny is essential for the understanding of cholinergic influences during the development of cortical circuitry. In the present study we describe an autoradiographic technique which permits the use of the covalently binding muscarinic ligand [3H]Propylbenzilylcholine mustard, in an in vitro binding assay on unfixed brain sections. MATERIALS AND METHODS Female BALB/c mice with timed pregnancies were supplied by the breeding colony of the National Cancer Institute and their pups were sacrificed on
* Present address: The Johns Hopkins University School of Medicine, Department of Child Psychiatry, 725 North Wolfe St., Baltimore, MD 21205, U.S.A. Correspondence: F.F. Ebner, Division of Biology and Medicine, Brown University, Providence, RI 02912, U.S.A.
244
embryonic day 18 (El8) or postnatal day (P) 1,4, N, 14, 31, and 60 (considered adult). El8 was defined as the eighteenth day after vaginal plug detection and P1 was defined as the day of birth. Embryos were removed by Caesarean section and killed by decapitation. Postnatal animals were sacrificed by cervical dislocation prior to decapitation. All brains were removed quickly and frozen in isopentane for 20-30 s at -25 to -35 °C. Four brains were prepared at every age point. Brain sections were processed in 4 separate experiments, each containing one brain from every age group. After the binding assay, two brains at each age point were processed for autoradiography on LKB tritium-sensitive Ultrofilm. The remaining tissue sections were dipped in a solution of nuclear track emulsion (NTB-2 Kodak). Some of the emulsion-dipped slides and all of the slides used for LKB film autoradiography were subsequently counterstained with cresyl violet or with Neutral red. Muscarinic receptor binding was performed with [3H]propylbenzilylcholine mustard ([3H]PrBCM) supplied by New England Nuclear. This irreversibly binding cholinergic ligand has been studied extensively by Burgen et al 4 and has been characterized as a specific muscarinic receptor binding agent. The muscarinic antagonist atropine was used as an inhibitor of radioactive ligand binding to measure non-specific binding. The autoradiographic binding assay was performed according to a method developed by HOhmann and Pert, based on the procedures used by Rotter er al. 20 for muscarinic receptor binding, and by Herkenham and Pert s for opiate receptor binding. Briefly, unfixed frozen tissue was sectioned in a cryostat at -15 °C into 32ktm thick coronal sections. Sections were collected onto cold slides, melted to the slide by briefly touching the back of the slide, dried under vacuum in the cold (5-10 °C) and subsequently stored at -70 °C. For the receptor binding assay, slides were preincubated at room temperature with either Krebs buffer or Krebs buffer containing 10-5 M atropine. The Krebs buffer was freshly prepared for every experiment and gassed with a mixture of 95% CO, and 5% O2 prior to use. The ionic composition of this buffer was: sodium chloride 118 mM, potassium chloride 4.7 mM, calcium chloride 2.6 mM, potassium phosphate monobasic 1.2 mM, magnesium sulfate 1.2 raM, sodium bicarbonate 25 mM and dextrose 11 mM. After 15 min of preincubation
[-~H]PrBCM (previously cyclized a! room temperature in 10 mM phosphate buffer, pH 7.4. a~, a c~mcentration of 10-~ M) was added al a concentration of 2.4 nM and the slides were incubated for 45 rain ~tt room temperature. Subsequent to the incubation, sections were quickly washed in 6 rinses of icc-cokt Krebs buffer, dried well in a cold air stream, fixed with formaldehyde vapors at 8(1 °C, defatted in xylenes and processed for autoradiography (see ref. g). Sections used for scintillation counting were scraped off the slides into scintillation vials, strongly agitated in Aquasol liquid scintillation fluid (NEN) and counted. RESULTS
The binding technique The binding procedure used in this study was modified in several ways from that of Rotter et at. in an attempt to analyze the receptor distribution in untreated, unfixed tissue. Therefore several conditions of the assay had to be reassessed. The [3H]PrBCM concentration of 2.4 nM was adopted unchanged from the PrBCM concentrations used by Rotter et al. 20. However, unlike the Rotter et al. protocol, the tissue used in our study was both unfixed and sectioned at 3 times the thickness (32 ,urn). These changes were likely to have altered the temperature and time-dependency of the binding process. For the determination of optimal assay conditions sections were scraped off the slides after the incubation and counted in scintillation fluid. The optimal incubation conditions at room temperature were achieved at 45 rain, as can be seen in Fig. 1. We ensured a constant high concentration of atropine in the control incubations during this extended incubation time by increasing atropine to concentrations 10 times higher than those reported by Rotter et al. 20. Incubations at temperatures warmer than 30 °C resulted in inferior histological preservation. Incubation at temperatures lower than 20 °C resulted in higher ratios of non-specific to specific binding. Ratios of specific to non-specific binding, when measured in dried, non-defatted sections subsequent to the assay, revealed a specific binding of only approximately 40%. However, a ratio of approximately 70% specific versus non-specific binding can be achieved (see Fig. 1) when the tissue is fixed and defatted after the receptor binding proce-
245 75
¢3 Z
times more measurable cpm than total binding after such t r e a t m e n t of the tissue. Thus more non-specifically b o u n d ligand than specifically b o u n d ligand is removed in the defatting process, indicating that some of the sites for non-specific binding are soluble in the alcohols or in xylenes.
65
LI.
~45 U.I
A utoradiography
CO
Muscarinic receptor sites in mouse cerebral cortex and hippocampus, as visualized by a u t o r a d i o g r a p h y , a p p e a r before birth and increase in density for several weeks after birth. Early in d e v e l o p m e n t silver grains are seen evenly distributed over the densely p a c k e d undifferentiated cells in cortex and hippocampus. With progressing differentiation silver grains are seen p r e d o m i n a n t l y in the neuropil between cells (Fig. 2B). This tendency of the silver grains to a p p e a r in higher density in neuropil areas versus over cell bodies can be seen especially well in the hippocampus (Fig. 2A). The overall pattern of muscarinic receptor sites shows surprisingly little differential distribution at all ages investigated. Neocortex and the h i p p o c a m p a l formation are the cortical areas with the clearest p a t t e r n of binding and we
~ 35
J 15
30
45
6 0 MIN.
Fig. 1. Time-dependency of specific [3H]PrBCM binding to adult brain sections at room temperature. Closed circles show specific binding before defaning of the tissue; open circles show specific binding after defatting. Percent specific binding was calculated by subtracting the cpm of atropine-blocked controls from total cpm and expressing the difference as percent of total counts. This figure shows the results from a typical experiment. Each point represents the mean value for 3 different coronal sections through mouse forebrain. dure ( H 2 0 ; E t O H 70%, 95%, 100%; xylenes 2 × ; E t O H 100%, 95%, 70%; H 2 0 , 5 min each). Total binding before fixation and defatting shows 5 - 1 0
~z~i~ i~~~iiI
Fig. 2. Coronal sections through adult mouse forebrain showing silver grain distribution in the dentate gyrus of the hippocampus (A). The density of silver grains in the dentate molecular layer (upper left cellsparse area) is greater than over the granule cell layer. In layer VI of somatosensory cortex (B) the silver grains are distributed relatively evenly throughout the field of view.
246
Fig. 3. Dark-field micrographs illustrating the postnatal development of muscarmic receptors in hippocampus, cingulatc, ~md medial parietal cortex. The autoradiographs for this figure were prepared by the emulsion dipping technique. The ages shown art:: ,\, P l: t-H P4; C, PS: D, PI4; E, P31; cp, cortical plate; g, granule cell layer: la too, stratum molecularc lacunosum; m, molecular tascr
will focus primarily on these areas in our subsequent description of muscarinic receptor development. Neocortex
Silver grains at levels above background can be detected in all areas of immature neocortex as early as E l 8 , though tissue preservation at this early age is not good enough to determine whether any patterned binding exists. By P1 a moderate n u m b e r of silver grains are distributed homogeneously throughout the cortex. The cortical layer I~ however, shows a noticeably higher grain density, separated from the
rest of cortex by a relatively grain-free zone which occupies lower parts of layer I. This sparsely labeled zone extends slightly into the upper cortical plale (see Fig. 3A). By P4, the grain density increases in subptate layers and in layer I. The latter represents the most intensely labeled part of cortex at this age. A zone of relatively low grain density extends from lower laver 1 throughout the cortical plate. Labeling of receptor sites is at its lowest over the cortical plate (see Fig. 4A, B). At P8 the overall silver grain pattern does not differ very much from the pattern at postnatal
247
Fig. 4. Bright-field micrographs of autoradiographs of [3H]PrBCM binding on LKB Ultrofilm (B,D) compared with the same coronal section after Nissl staining (A,C). A and B show somatosensory cortex of a P4 animal; C and D show somatosensory cortex of a PI4 animal. Note the increase in labeling in the region of layers IV through II from P4 to P14 and the corresponding decrease of labeling in layer I.
248
Fig. 5. Adult somatosensory cortex: bright-field micrograph of an autoradiograph of [3H]PrBCM binding on LKB Uhrolitm (B) compared to the same section after Nissl stain (A). Compare to Fig. 4D and note the relative increase of labeling in cortical laxcrs II and 111 between P14 and adult and the further decrease in silver grains in layer I of cortex. day 4, but the overall intensity of the label has increased. By P14 the double band of silver grains, characteristic of the adult pattern, begins to emerge. The overall intensity of label has markedly increased in all layers of cortex except for a zone of low grain density in upper layer V. The label in supragranular layers, however, still has not reached the same density as labeling in the infragranular layers and is especially low at the border between layer II and 1. The upper tier of layer I remains heavily labeled (see Fig. 4C, D). Towards the end of the fourth postnatal week, the labeling decreases considerably in layer I of neocortex. The grain density in cortical layers IV, lII and II is comparable at this age to layers VI and lower V. The adult mouse neocortex shows two layers of comparable grain density separated by a less densely labeled zone which occupies upper layer V (see Figs. 5 + 6). This relatively grain-sparse zone can be seen in all cortical areas which display a distinct pyramidal cell layer (see Fig. 6 A - C ) . Anteriorly, this zone extends beyond the rhinal fissure and into perirhinal
cortex. More posteriorly, where a distinct pyramidal cell layer cannot be detected lateral to sensory cortex, the sparsely labeled zone only appears in neocortical areas• The layer I band, so prominent at earlier ages, disappears entirely in adulthood and silver grain densities near the pial surface actually decrease to levels below those of deeper layers (see Fig. 5). LKB autoradiographs from adult animals show a lower labeling intensity than autoradiographs from P31 on the same LKB film. Thus overall grain density apparently decreases between the fourth postnatal week and adulthood.
Hippocampus The hippocampus is still very immature at Pl and consists essentially of one row of densely packed immature pyramidal cells. Silver grains can be seen in thin bands lining both sides of this immature pyramidal cell layer (see Fig. 3A). Caudal to the anterior commissure, the hippocampal cell layer still appears continuous with the cingulate pyramidal cell laver and silver grains form two uninterrupted bands
249
Fig. 6. Images of coronal sections from anterior (A) to posterior (D) through the forebrain of an adult mouse. Sections A-C show bright-field autoradiographs on LKB Ultrofilm of total binding (sections incubated with [3H]PrBCMonly). D shows a control section incubated in the presence of atropine. Note the banding pattern in cortex and the differential labeling in various nuclei of the basal forebrain and diencephalon in A-C. The non-specificbinding on the control section (D) is unpatterned.
above and below these cells. By P4, label can be seen in the developing stratum oriens and stratum radiatum and, to a much lesser extent, overlying the pyramidal cell layer. The granule cell layer of the dentate gyrus has become apparent at this age and is free of label, being outlined by the silver grains in the molecular layer and in the hilus region (see Fig. 3B). By the end of the first postnatal week, the intensity of the label has increased in the whole hippocampal formation. The increase of silver grains in the molecular layer of the dentate gyrus and in the stratum lacunosum moleculare is proportionally larger than in the stratum oriens and stratum radiatum. The pyramidal cell layer shows higher grain density than the dentate granule cell layer; both cell layers, however, are far
less intensely labeled than the hippocampal white matter. Bands of relatively lower grain density appear in the polymorph layer and in the stratum moleculare along its border with the stratum radiatum (Fig. 3C). Except for an overall increase in grain density, no differences can be seen between the one- and two-week-old hippocampal formation (Fig. 3D). In the adult animal the hippocampal formation appears more lightly labeled than in the two-week-old animal, but the pattern has not changed qualitatively very much from that of the 8-day-old animal. The lightly labeled pyramidal cell layer is outlined by an intensly labeled stratum oriens and stratum radiatum. Below the stratum radiatum a zone of lower grain density marks the border of the stratum moleculare. Ventrally, in the dentate gyrus, the grain den-
250 sity is increased again, remains high throughout the dentate molecular layer, and drops to almost background levels over the dentate granule cell layer. Grain densities in the stratum lucidum and polymorphe are only slightly higher than grain densities in the pyramidal cell layer (see Figs. 3E and 6B, C). O t h e r cortical areas
Cingulate cortex retains essentially the same labeling pattern throughout development. A band of high density overlies layer I and is continuous with the neocortical layer 1 band. Silver grains are low over the granule cells in layer I1 of cingulate cortex, but below this cell layer grain density is increased and remains uniformly distributed down to the level of the white matter. Grain density below layer II of cingulate cortex always parallels the grain density observed in layer IV of neocortex (Figs. 3 and 6). The only noticeable change in the cingulate labeling pattern is a decrease in intensity of the layer I band which occurs concomitantly with the decrease in neocortical layer I labeling (see Fig. 3). Paleocortical areas show no differentiated silver grain pattern at any point in development. These areas are characterized by a slow increase in labeling intensity to adult levels, at which time their grain density is comparable to layer VI labeling in neocortex (see Figs. 3 and 6). DISCUSSION These results provide a detailed description of muscarinic cholinergic receptor development and distribution in the cerebral cortex of the mouse. They show that some muscarinic receptor binding can be detected in cortex as early as embryonic day 18. A rapid increase in labeling intensity can be seen during the first week of postnatal development in all cortical areas. Noticeable changes in the binding pattern appear in both the neocortex and the hippocampal formation during the first and second week of life. Complete maturity of the binding distribution is reached shortly after the fourth postnatal week. Silver grain densities decrease slightly between 4 weeks and adulthood. While there are clear differences in silver grain density between various cortical layers, silver grains always show a fairly homogeneous tissue distribution if viewed at high magnification. At ages P4
and older the silver grains are distributed cxcnl} throughout the neuropil and over cells, so thai tocali zation over individual cell bodies is not apparent a any age. In the present paper we have introduced a m o d ified technique for radioactive labeling of muscarinic cholinergic receptors on tissue sections. This proce. dure has several advantages over other method,' presently used to visualize muscarinic cholinergic receptors. Tritiated PrBCM is an irreversibly bmdin~ ligand and thus allows extensive post mcubatior washing of the tissue without ligand remowd or diffu: sion within the tissue during autoradiographic processing. The binding properties of PrBCM also allox~ defatting of the brain sections after formalin vapol fixation and prior to autoradiography. This represents a significant improvement over the QNB binding techniques in view of the recent growing awarehess of lipid quenching in autoradiographic processing 9. Tritiated PrBCM has been utilized successful b by Rotter er al. >, 21 to visualize autoradiographic receptor distribution in developing and adult rat brains. The tissue used in these studies was lightly fixed prior to the ligand incubation in order to assure sufficienl tissue preservation, and the animals were anesthetized before brain removal. Using methods recently developed by Herkenham and Pert for opiate receptor binding s, we have been able to perform our binding studies on entirely untreated, unfixed fresh tissue. These conditions have allowed us to detect higher percentage of muscarinic receptors that were present in the live animal than previous studies. 13otlfixation and threatment of the animals with anesthesia could potentially alter the binding properties ol receptors in the brain. We have taken two slightly different approaches t~ monitoring the autoradiographic distribution of muscarinic receptors in cerebral cortical development, Autoradiographs prepared according to the liquic emulsion dipping technique and LKB Ultrofilm autoradiographs complemented each other very well i~ their ability to show developmental changes. While the former technique is indispensable fl:~r the accurate identification of silver grain location, the latter technique allows better tissue preservation, frequently provides sharper images of distribution patterns, and also facilitates low power bright-field visualization of whole brain sections.
251 Our results concerning the distribution and ontogenesis of muscarinic cholinergic binding sites in neocortex are generally in good agreement with previously published results. Kuhar and Yamamura ~3 described an autoradiographic distribution pattern of in vivo [3H]QNB binding in the adult rat neocortex which closely resembles the present findings in the mouse: two strata of similarly high-grain density are separated by a low-density band in upper layer V. Wamsley et al. 27 studied both total and high-versus low-affinity muscarinic receptor patterns in rat brain. Their visualization of total muscarinic binding, performed with [3H]methyl-scopolamine, also shows the above-mentioned pattern in adult neocortex. A number of biochemical in vitro assessments of muscarinic receptor development in mouse and rat cerebral cortex also show the general time-course of muscarinic receptor ontogeny l, 2.6, 14. Thus, in vitro and in vivo binding techniques w!th a variety of different muscarinic cholinergic ligands lead to similar conclusions. Several differences do exist, however, between our results, gained with the procedure outlined above, and those reported by Rotter et al. 2~. To our knowledge these studies report the only other description of muscarinic receptor distribution in rodent cerebral cortex during the early postnatal period. They also illustral~e the transiently heavy labeling in layer I between P1 and P13, and it is noteworthy that the muscarinic receptors are at their densest during the postnatal period when the non-specific thalamocortical fibers are forming their initial synapses in this stratum and the AChE staining is only starting to develop. In our study the grain density at each depth in cortex follows the gradient of neuronal differentiation; i.e. the pattern in layer VI and V develops before the pattern in layers IV, III and II. Finally, the width of the grain sparse band that we see only in the upper half of layer V, extends throughout layers V and IV in the rat 2~. Both the Rotter et al. and the present study used [3H]PrBCM for the receptor binding assay, but the absence of anesthesia and fixation in our study may have preserved more of the receptors. Alternatively, it is possible that the ontogeny of muscarinic receptors is different in mouse and rat. Such a species difference has been noticed in the appearance of AChE in the hippocampus (see refs. 10, 11. The distribution of muscarinic cholinergic binding
sites in the adult rodent neocortex does not correspond precisely with the density profiles of other cholinergic markers. While receptors can be found in relatively high densities in all layers of cortex, except upper layer V and I, AChE is most prominent in upper layer VI and IV and 110-12. The number of cells responsive to iontophoretically applied ACh is largest in deep cortical layers TM 16. Such incongruities between the muscarinic cholinergic receptor pattern and the distribution of other cholinergic properties may become more interpretable once the physiological significance of different receptor subclasses is better understood. Muscarinic cholinergic receptor subclasses do apparently bind antagonists with comparable affinity, but differ in their affinity for agonists and are probably not all of equal physiological importance (see ref. 7 for rewiev). Wamsley et al. 27 provided evidence for the differential distribution of these different receptor subtypes in neocortex. The high agonist affinity muscarinic receptor type, for example, is present in the highest density in cortical layer IV followed by layers V and VI, and much lower levels of this receptor type are present in the supragranular layers. It is unresolved whether all muscarinic receptor subtypes in the CNS are capable of inducing membrane resistance changes and whether all are located on neurons. In conclusion, maturation of muscarinic cholinergic receptors in mouse cerebral cortex occurs late in cortical development. Muscarinic receptor ontogeny thus parallels the late development of other cholinergic markers. A low density of receptors, however, are present in cortex before the appearance of any other cholinergic markers so that receptors are found at times when ChAT activity and presumably ACh levels still are very low. Early postnatal cerebral cortex thus has relatively high levels of AChE and muscarinic receptors but low levels of ACh synthesis 6, 10,11. Since both muscarinic receptors and AChE staining show patterns of differential distribution, this relationship may be conducive to very localized effects of ACh on postsynaptic cells. In areas of high AChE concentration, presumably only receptors very close to the release site would be able to bind ACh released in cortex before it is degraded. This situation appears very different from the adult cerebral cortex where ACh seems capable of diffusion over quite some distance 5,19. It is noteworthy, that both
252 the muscarinic receptor distribution and the A C h E
of the cholinergic system in development may b~
pattern develop from deeper to more superficial cor-
very different from the cholinergic influence on tht
tical layers, yet both show labeling of layer ! almost
adult cerebral cortex, but if so, this function rcmain~
immediately after their first appearance in layer VI.
to be elucidated.
In addition, the time-course of pattern development is remarkably similar for A C h E and muscarinic re-
ACKNOWLEDGEMENTS
ceptors. As m e n t i o n e d in a previous report, these oncorrespond
A preliminary report of the present data has bee~
closely to strata of afferent synapse development 3, ~,
given in Neuroscience Abstracts, Vol. 9, 1983. The
togenic
cholinergic marker
patterns
u. The remarkable changes in patterns in all cholin-
authors would like to thank Dr. W. Bowen for his
ergic markers, which often closely parallel other de-
helpful suggestions in the development of the tech-
velopmental events, suggest that the cholinergic sys-
nique, and Dr. S. Moon-Edley for suggestions thal
tem may play a specific role in cerebral cortical morphogenesis, including synaptogenesis. The function
ported by the N I M H and NIH G r a n t NS13031.
REFERENCES 1 Aronstam, R.S., Kellogg, C. and Abood, L.G., Development of muscarinic cholinergic receptors in inbred strains of mice: Identification of receptor heterogeneity and relation of audiogenic seizure susceptibility, Brain Res., 162 (1979) 231-241. 2 Ban-Barak, J. and Dudai, Y., Cholinergic binding sites in rat hippocampal formation: properties and ontogenesis, Brain Res., 166 (1979) 245-257. 3 Blue, M.E. and Parnavales, J.G., The formation and maturation of synapses in the visual cortex of the rat: It. Quantitative Analysis, J. Neurocytol., 12 (1983) 697-712. 4 Burgen, A.S.V., Hiley, C.R., Young, J.M., The properties of muscarinic receptors in mammalian cerebral cortex, Br. J. Pharmacol., 51 (1974) 279-285. 5 Collier, B. and Mitchell, J.F., The central release of acetylcholine during stimulation of the visual pathway, J. Physiol. (London), 184 (1966) 239-254. 6 Coyle, J.T. and Yamamura, H.I., Neurochemical aspects of the ontogenesis of cholinergic neurons in the rat brain, Brain Res., 118 (1976) 429-440. 7 Ehlert, F.J., Roeske, W.R. and Yamamura, H.I., Muscarinic cholinergic receptor heterogeneity, Trends Neurosci., 5 (1982) 336-339. 8 Herkenham, M. and Pert, C.B., Light microscopic localization of brain opiate receptors: a general autoradiographic method to preserve tissue quality, J. Neurosci., 2 (1982) 1129-1149. 9 Herkenham, M. and Sokoloff, L., Quantitative receptor autoradiography: tissue defatting eliminates differential self-absorbation of tritium radiation in gray and white matter of brain, Brain Res., 321 (1984) 363-368. 10 H6hmann, C.F. and Ebner, F.F., The development of cholinergic markers in normal and transplanted mouse neocortex, Neurosei. Abstr., 8 (1982) 865. 11 H6hmann, C.F. and Ebner, F.F., Development of cholinergic markers in mouse forebrain. I. Choline acetyltransferase enzyme activity and acetylcholinesterase histoehemistry, Dev. Brain Res., 23 (1985) 225-241. 12 Kristt, D.A., Development of neocortical circuitry: histochemical localization of AChE in relation to the cell layers
improved the manuscript. This work has been sup-
of rat somatosensory cortex, J. Comp. Neurol.. 186 (1979) 1-17, 13 Kuhar, M.J. and Yamamura, H.I., Localization of muscarinic receptors in rat brain by light microscopic radioautography, Brain Res., 110 (t976) 229-243. 14 Kuhar, MJ., Birdsall, N.J.M., Burgen, A.S.V. and Hulme, E.C., Ontogeny of muscarinic receptors in rat brain, Brain Res., 184 (1980) 375-383. 15 Krnjevic, K. and Phillis, J.W., Acetylcholine sensitive celb in cerebral cortex, J. Physiol. (London), 166 (19631 296-327. 16 Lamour, Y., Dutar, P. and Jobert, A., Excitatory effect ol acetylcholine on different types of neurons in the first somatosensory neocortex of the rat: laminar distribution and pharmacological characteristics, Neuroscience. 7 (1982) 1483-1494. 17 Matthews, D.A,, Nadler, J.V., Lynch, G.S. and Cotman, C.W., Development of cholinergic innervation in the hip: pocampal formation of the rat. I. Histochemical demonstration of AChE activity, Dev. Biol., 36 (1974) 130-141. 18 Nadler, J.V., Matthews, D.A., Cotman, C.W. and Lynch, G.S., Development of the cholinergic innervation of the hippocampal formation of the rat: II. Quantitative changes in choline acetyltransferase and acetylcholinesterase activities, Dev. Biol., 36 (1974) 142-154. 19 Phillis, J.W., Acetylcholine release from the cerebral cortex: its role in cortical arousal, Brain Res.. 7 (1968) 378-389. 20 Rotter, A., Birdsall, N.J.M., Burgen, A.S.V., Field, P.M., Hulme, E.C. and Raisman, G., Muscarinic receptors in the central nervous system of the rat. I. Technique for autoradiographie localization of the binding of [3H]propylbenzilylcholine mustard and its distribution in the forebrain, Brain Res. Rev., 1 (1979) 141-165. 21 Rotter, A., Field, P.M. and Raisman, G., Muscarinic receptors in the central nervous system of the rat: postnatal development of binding of [3H]propylbenzilylcholinemustard, Brain Res. Rev., 1 (1979) 185-205. 22 Sillito, A.M., Plasticity in the visual cortex. Nature (London), 303 (1983) 477-478. 23 Singer, W., Central core control of visual cortex function. 1. The Neurosciences. Fourth Study Program, MIT Press,
253 Cambridge, MA 1979. 24 Singer, W., Central core control of developmental plasticity in the kitten visual cortex. I. Diencephalic lesions, Exp. Brain Res., 47 (1982) 209-222. 25 Singer, W. and Rauschecker, J.P., Central core control of developmental plasticity in the kitten visual cortex. II. Electrical activation of mesencephalic and diencephalic projections, Exp. Brain Res., 47 (1982) 223-233.
26 Sorimachi, M., Mijamoto, K. and Katakoa, K., Postnatal development of choline uptake by cholinergic nerve terminals, Brain Res., 79 (1974) 343-346. 27 Wamsley, J.K,, Zarbin, M.A., Birdsall, N.J.M and Kuhar, M.J., Muscarinic cholinergic receptors: autoradiographic localization of high and low affinity agonist binding sites, Brain Res., 200 (1980) 1-12.