- Email: [email protected]
Quantitative data on serotonin nerve terminals in adult rat neocortex
Brain Research, 111 (1976) 301-309 l) Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
301
QUANTITATIVE DATA ON SEROTONIN NERVE TERMINALS 1N A D U L T R A T N E O C O R T E X
ALAIN BEAUDET and LAURENT DESCARRIES Centre de Recherche en Sciences Neurologiques, Universitd de Montrdal, Montreal, H3C 31"8 ( Cana&t)
(Accepted January 5th, 1976)
SUMMARY Serotonin (5-HT) nerve terminals, specifically labeled with [aH]5-HT have been counted in light microscope radioautographs from the fronto-parietal neocortex of adult rats, following prolonged superfusions with relatively high concentrations of tracer (10 a M or 10 4 M), and after 15 or 30 days of radioautographic exposure. Comparative analysis of the results indicated that all 5-HT varicosities did not possess the same capacity to accumulate the exogenous amine. Nevertheless, superfusions with 10-4 M [aH]5-HT provided a sufficient loading of 5-HT boutons to allow their complete detection within any given layer of the neocortex. The labeled varicosities were found to be present within all cortical layers, except layer VI. Their total number was extrapolated to be approximately 1 x 10~/ cu.mm of cortex. This figure represented a mean incidence of one 5-HT nerve terminal/850-1400 cortical synapses (0.07-0.12%). The intralaminar density of 5-HT innervation increased progressively from layer V to layer I, in a distribution pattern suggestive of unspecific afferents. Based on endogenous 5-HT levels of 346 ± 32 ng/g for the fronto-parietal neocortex, the mean 5-HT content per varicosity could be estimated to be 0.35 × 10 a pg, and the concentration 1900 ng/g wet weight of terminal or 0.2~o. In view of their particular ultrastructural features 7, ubiquitous repartition and endogenous amine content, 5-HT nerve endings may be considered as capable of exerting a rather widespread influence in the neocortex.
INTRODUCTION Little information was hitherto available on the innervation of mammalian cerebral cortex by serotonin (5-HT) containing fibers. Histofluorescence studies in the rat have revealed the existence of 'scattered 5-HT terminals present in the super-
302 ticial layers "11, "most of them being found in the outer part of the molecular layer ::;. Further investigations, based on lesion experiments"ahel,e:~, have indicated tha~ these nerve endings originate from cell bodies situated in the raphe nuclei of the mesencephalon (groups B-7, B-8 and B-9 of Dahlstr6m and Fuxe6). High resolution radioautography has recently permitted the identification and ultrastructural characterization of 5-HT fibers in the neocortex of adult ratsL These terminal arborizations were shown to be more numerous than previously suspected, and dispersed throughout layers I V in the fronto-parietal region. The 5-HT axons were described as tenuous, unmyelinated processes, bearing small enlargements free of glial enwrapping. These varicosities were filled with vesicular organelles, bul, even when examined in serial thin sections, were seldom found to be engaged m typical synaptic relationships. Such structural features lead to the postulate that most cortical 5-HT varicosities could be in constant motion along their parent fibers, and might therefore release their biogenic amine in different tissue sites at different moments in time v. The main purpose of the present study was to seek precise quantitative data on the intracortical distribution of 5-HT terminals, and their endogenous amine content, in order to establish a basis for better understanding of the functional significance of 5-HT in cortex. MATERIALS AND METHODS Six adult male Sprague-Dawtey rats (body weight, 300 :~: 60 g), pretreated with a monoamine oxidase inhibitor (/3-phenylisopropyl hydrazine, 5 mg/kg, i.p., 18 and 2 h earlier), were anesthetized with Nembutal and submitted to 3 h cortical superfusion with tritiated serotonin ([aH]5-hydroxytryptamine <:G>- creatinine sulfate, specific activity 11-17.3 Ci/mM, obtained from Amersham-Searle), using a technique previously described in detail v,8. The tracer was applied unilaterally, on a region of the fronto-parietal neocortex corresponding to parts of areas 3, 4, 6 and 10 of Krieg 18. The concentration of [aH]5-HT in the superfusate was 10-~ M in 3 rats, and 10 4 M in the other 3. The cortical superfusions were immediately followed by intra-aortic perfusionfixation with 3.5'~'~,i glutaraldehyde in 0.05 M phosphate buffer 9. The superfused portions of cortex were excised from the cerebral hemispheres, sliced coronally into 1 ram-thick slabs, and post-fixed in osmium tetro×ide. Blocks including all 6 cortical layers were cut out from each slice, dehydrated in ethanol, and flat-embedded in Epon to provide a sectioning plane perpendicular to the pial surface. Two series of l #m-thick sections were obtained from each block, simultaneously processed for light microscope radioautography 17, and examined after 15 and 30 days of exposure, respectively. To obtain quantitative data on the intracortical distribution of serotonin nerve terminals, which have been shown to be specifically labeled in the above conditionsT: the silver grain aggregates present in light microscope radioautographs were systematically counted with the aid of a camera lucida. This was done on 6 sections from
303 each rat, after both durations of radioautographic exposure. Areas for counting were selected by means of a transparent overlay, superimposed on a scale drawing of the sections ( × 170). The labeled sites were charted within 8 squares, drawn on the overlay as a vertical series staggered in the horizontal plane. This preset pattern would be oriented with the upper side of the first square aligned along the pial edge of the section. Each square delineated an area of tissue 150/zm of side. Each of the upper 4 cortical layers was thus randomly sampled over its entire thickness 19 by 1 square, and layer V by 4 squares. Layer VI was not studied, since it was virtually devoid of tracer accumulations. The values were corrected for 1 sq. mm of radioautograph, and their statistical significance was analyzed with a two-tailed t-test. The validity of the sampling technique was verified in 9 radioautographs, and the results obtained by means of the overlay were found to be comparable ( ± 10 ~'/~i)to counts obtained from entire sections. In order to estimate the number of labeled terminals per volumetric unit of cortex, all reactive sites were considered as spheres. It was also assumed that a segment of labeled varicosity 0.1 # m in thickness had to be included within a I /~m-thick section to produce a detectable radioautographic reaction. Since the average diameter of cortical 5-HT varicosities had been measured to be 0.7/zm v, it could be inferred that 1/~m-thick sections spaced at 0.5/~m intervals would sample all reactive terminals contained in a given volume of tissue 22. The number of reactive varicosities being fairly constant in all sections prepared under similar experimental conditions (S.E.M. < 10~,/0, values for 1 cu.mm of cortex could be safely extrapolated by multiplying results for 1 sq. mm of radioautographs by 1000/(1 + 0.5) 666.66. To evaluate the serotonin content of 5-HT varicosities, endogenous levels of the indolamine were measured in cortical regions identical to those prepared for radioautography. The brains from 8 normal rats were rapidly removed after decapitation, frozen to a semi-solid consistency, and sliced coronally from 3 to 9 m m behind the frontal pole. Under a dissecting microscope, strips of the dorso-medial neocortex were excised from each slice, homogenized in 0.1 N HCI, and assayed for serotonin with the radio-enzymatic technique of Saavedra et al. 26. RESULTS Following superfusions of rat neocortex with 10 -4 or 10-5 M [3H]5-HT, accumulations of the tracer were visualized in light microscope radioautographs in the form of small and dense aggregates of silver grains (Fig. 1). These have been shown to designate axonal 5-HT varicosities, filled with vesicular organelles and making only sparse synaptic contacts 7. The intra-cortical distribution of the labeled varicosities is illustrated in Figs. 2 and 3, for the two molarities of [3H]5-HT in the superfusate, after either 15 or 30 days of radioautographic exposure. Fig. 2 emphasizes the influence of the tracer concentration. In every layer, the number of labeled 5-HT varicosities was greater after 10 -4 M than after 10 -5 M superfusions. However, in layers I - l l I , there were only slight differences between counts obtained after 30 days of exposure with 10 -5 M,
304
Fig. 1. Light microscope radioautograph from the neocortex of an adult rai, following 3 h stlpcr.fusion with 10 a M [:~HJ5-HT. Monoamine oxidase inhibited. Small and dense aggregates of >Hxev grains correspond to serotonin nerve terminals - i.e. axonal varicosities - in x~hich the tracer i~ highly concentrated. The diffuse labeling in background presumably represents unbound amine and/or labeled metabolites, Cortical layers I and 11. Exposure time: 2 weeks. Magnification: ~00
a n d those after 15 days with 10-i M [aH]5-HT. Fig. 3 underlines the effect o f ' r a d i o a u t o g r a p h i c exposure. W i t h l 0 :' M superfusions, d o u b l i n g the d u r a t i o n o f exposure p r o d u c e d a c o n s i d e r a b l e increase in the n u m b e r o f reactive sites within all cortical layers. In contrast, with 10 ~ M superfusions, there was no significant difference between counts o b t a i n e d after 15 or 30 days o f exposure. The total n u m b e r o f 5 - H T varicosities per volumetric unit o f cortex was therefore e x t r a p o l a t e d from all counts o b t a i n e d after superfusion with 10 i M [aH]5-HT (Table 1). When the n u m b e r o f labeled varicosities was expressed per cu.mm o f cortical layer ( i n t r a l a m i n a r density), the serotonin nerve terminals a p p e a r e d mostly c o n c e n t r a t e d within the u p p e r 4 layers. This density diminished progressively from layer 1 to layer I11, r e m a i n e d u n c h a n g e d in layer IV, a n d was the lowest in layer V. W h e n the values were expressed as a percentage o f the total n u m b e r o f 5-HT varicosities in cortex (intracortical repartition), layer V was f o u n d to contain the highest p r o p o r t i o n , its thickness being 4 times g r e a t e r than that o f each u p p e r layer. F r o m these d a t a , the overall density o f cortical 5 - H T innervation could be estimated at nearly I 10 ~; varicosities/cu.mm o f cortex. A c c o r d i n g to current
305 xlO~/mm 2
xlO~mm 2 20
15
10
5
0
0
5
10
15
20
25
30
i
E ~////W////////////W~//////A_~ ~q
w~2/////~/
~i ~///////////////~//wA-~ ~r ~////////////W~///h~
lO-SM
~
~ ~,/~1
10-4M
Fig. 2. Radioautographic distribution of 5-HT varicosities in the upper 5 layers of the fronto-parietal neocortex (I-V), following 3 h superfusions with 10 -~ M of 10 4 M [3H]5-HT. Counts obtained from radioautographs developed after 15 days (clear colums) and 30 days (hatched columns) of exposure. Each value represents the mean number (:k S.E.M.) for 18 1 /~m-thick sections from 3 rats, corrected for l sq. mm of radioautograph.
IO-SM
lO-aM
30
"E E
oJ
20
m57
eT'-.. o
x
]]I 5Z
10
o x
S[ 15
30 days
0
15
30 days
Fig. 3. Effect of the duration of radioautographic exposure on the number of axonal varicosities labeled in the upper 5 layers of the fronto-parietal neocortex (I V), following 3 h superfusions with 10 ~ M or 10 -4 M [3H]5-HT. Data collected as in Fig. 2.
TABLE I
Distribution o f 5 - H T nerve terminals in adult rat fronto-parietal neocortex Cortical layer
lntralaminar density (n x 106/eu. mm)
lntracortical repartition ( % o f totalnumber in cortex)
I
1,83
lI Ill IV V
1,47 1,12 1,10 0,57
24 19 14 14 29 *
* Value which takes into account the greater thickness of layer V.
306 evaluations of the total number of synapses in the neocortex of rodents (8.5--14 i0s/ cu.mm) ~,a, this figure represents a mean incidence of one 5-HT varicosity~g50 i400 nerve terminals. The biochemical determinations of endogenous serotonin in the frontoparietal neocortex provided a value of 346 32 ng/g. Since the indolamine is mainly concentrated within axonal varicosities, the average amount of 5-HT per terminal was approximated to be 0.35 10 a pg. Using a spherical volume of 0.18 cu./,~m, and a specific weight of l for 5-HT terminals, the average concentration of serotonin per g wet weight of 5-HT varicosity amounted to 1900 #g or 0.2 i"ii. DISCUSSION An earlier study has already demonstrated that all silver grain aggregates, visible in light microscope radioautographs prepared in the above described conditions, represent 5-HT axonal varicosities specifically labeled with [aH]5-HTV. In order to obtain a reliable quantitative assessment of the cortical serotonin innervation, it remained to be shown that most if not all 5-HT varicosities were amenable to radioautographic detection. This was investigated by varying both the concentration of [aH]5-HT in the superfusate, and the duration of radioautographic exposure: while the molarity of the tracer could affect the amount of [aH]5-HT available for uptake and storage by 5-HT nerve terminals, the duration of exposure would merely influence the intensity of the radioautographic reactions. Comparison of the results obtained after IO -a M and 10 -'~ M superfusions confirmed previous findings indicating that many cortical 5-HT varicosities required high ambient concentration of the tracer for radioautographic visualization 7. This was particularly evident within the deep layers (IV-V), where the local concentration of [aH]5-HT was reduced by a penetration gradient from the surface inward. The increased number of labeled varicosities observed after t0 -4 M superfusions further suggested that, within any layer of the cortex, all 5-HT terminals did not accumulate the same amount of [aH]5-HT. Indeed, after 10 -5 M superfusions, some 5-HT nerve endings could always be detected, whereas others were not. Moreover, numerous 5-HT varicosities, passing unnoticed after 10 -5 M superfusions and short radioautographic exposure, did reveal their presence after doubling the exposure time. This variable capacity to take up and retain [aH]5-HT could possibly be related to different functional states, in terms of endogenous 5-HT content and/or uptake and binding properties. It was however obvious that superfusions with l0 -4 M [aH]5-HT provided a massive loading of all 5-HT varicosities, since in this condition the duration of radioautographic exposure had no significant effect on the number of labeled terminals. This plateau was reached within all cortical layers, indicating that the paucity of reactive varicosities detected in layer VI was not solely due to insufficient penetration of the tracer. The quantitative data gathered after 10-4 M superfusions confirmed the ubiquitous repartition of 5-HT nerve terminals throughout layers I - V in the fronto-parietal
307 neocortex. The gradual increase in intralaminar density of 5-HT varicosities towards the upper layers could merely reflect the progressive axonal arborization of an unspecific afferent system. This pattern of innervation was markedly different from that of other types of monoamine-containing afferents. Within the same region, noradrenergic terminals predominate more clearly in the superficial layers12, 22, whereas dopaminergic terminals appear mainly confined to the deeper layers3,15, 23. The overall density of serotonin innervation, approximated at 1 × 106 varicosities/cu.mm of cortex, was 10 times higher than that estimated for noradrenergic afferents within the fronto-parietal region 22. It corresponds to an incidence of 0.070.12 % of all cortical nerve endings, in accordance with earlier quantitative estimations based on the proportion of synaptosomes labeled with [3H]5-HT in whole forebrain homogenates 2°. The same figure appears relatively low, however, when compared with the proportion of nerve terminals labeled in vitro with tritiated GABA (30~i of all cortical synapses4,~6). The sparsity of the serotonin innervation does not preclude a significant role in cortex, but rather suggests a particular mode of action, especially in view of the rarity of typical synaptic junctions made by the 5-HT varicosities v. Despite the tact that the tissue concentrations of both serotonin and norepinephrine do not greatly differ in the fronto-parietal cortex (346 vs. 228 ng/g7), the higher number of 5-HT terminals implies a much lower endogenous amine content per 5-HT varicosity (0.35 vs. 2.33 × 10-3 pg2Z). However, the intravaricose concentration of both serotonin and norepinephrine is in the same order of magnitude (0.2 vs. 0.3 ~/o), owing to the smaller volume of 5-HT nerve endings (0.18 vs. 0.8 cu./~m). The value of 1900 #g of 5-HT/g wet weight of varicosity falls in the lower range of concentrations showing a linear relationship with the intensity of yellow-fluorescence induced by the Falck and Hillarp technique ~. It is therefore in keeping with the increases in fluorescence intensity reported for cortical 5-HT varicosities after monoamine oxidase inhibition 1° and L-tryptophan pretreatment 21. In the neocortex, endogenous 5-HT can probably be released from all axonal varicosities where it is concentrated, and not only from the small proportion which make typical synaptic contacts 7. It is thus likely that the indolamine does not exert its effect solely upon restricted areas of pc)stsynaptic membrane differentiation, but can diffuse in tissue, to reach relatively distant targets. A similar mode of action has already been proposed for biogenic amines, in both the peripheraF 4 and central nervous system 2s. In cortex, it could account for the presence of a paradoxically high number of presumptive 'postsynaptic' 5-HT receptor sites 2v. The above quantitative data therefore lead to the suggestion that, despite their relative scarcity, 5-HT afferents might exert a rather widespread influence in the neocortex. The entire cortical volume might even be considered to be within reach of the indolamine, if 5-HT can diffuse from the varicosities for a distance of a few microns only. Moreover, the varicosities themselves could possess, as previously suggested, dynamic properties of translocation and/or reshaping along their parent fibers, allowing incessant fluctuation of their functional domains.
308 ACKNOWLEDGEMENTS T h e a u t h o r s are i n d e b t e d to Mrs. S y l v i a G a r c i a a n d M r . K e n n e t h C. W a t k i n s f o r skillful t e c h n i c a l assistance,
and
Messrs.
Robert
Pdloquin,
Daniel Cyr and
E d w a r d R u p n i k f o r g r a p h i c a n d p h o t o g r a p h i c w o r k . T h e y also t h a n k M i s s Lise F a r l e y a n d D r . J a c q u e s de C h a m p l a i n f o r t h e i r h e l p in the b i o c h e m i c a l d e t e r m i n a l i o n s o f c o r t i c a l s e r o t o n i n , a n d are g r a t e f u l to Dr. H e r b e r t H. J a s p e r for his i n s p i r i n g r e v i e w o f the m a n u s c r i p t . T h i s s t u d y was s u p p o r t e d by a F e l l o w s h i p (A.B.) a n d G r a n t M T - 3 5 4 4 (L.[).), f r o m the M e d i c a l R e s e a r c h C o u n c i l o f C a n a d a .
REFERENCES 1 Aghajanian, G. K. and Bloom, F. E., The formation of synaptic junctions in developing brain: a quantitative electron microscopic study, Brain Research, 6 (1967) 716 727. 2 And6n, N.-E., DahlstrSm, A., Fuxe, K., Larsson, K., Olson, L. and Ungerstedt, U., Ascending monoamine neurons to the telencephalon and diencephalon, Acta physiol, stand., 67 (1966) 313 326. 3 Berger, B., Tassin, J. P., Blanc, G., Moyne, M. A. and Thierry, A. M., Histochemical confirmation for dopaminergic innervation of the rat cerebral cortex after destruction of the noradrenergic ascending pathways, Brain Research, 81 (1974) 332-337. 4 Bloom, F. E. and lversen, L. L., Localizing aH-GABA in nerve terminals of rat cerebral cortex by electron microscopic autoradiography, Nature (Lond.), 229 (1971) 628-.630. 5 Cragg, B. G., The density of synapses and neurones in the motor and visual areas of the cerebral cortex, J. Anat. (Lond.), 101 (1967) 639-654. 6 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurons in the central nervous system, l. Demonstration of monoamines in the cell bodies of brain stem neurons, Acta physiol, scand., 62, Suppl. 232 (1964) 1-55. 7 Descarries, L., Beaudet, A. and Watkins, K. C., Serotonin nerve terminals in adult rat neocorte×, Brain Research, 100 (1975) 563-588. 8 Descarries, L. and Lapierre, Y., Noradrenergic axon terminals in the cerebral cortex of rat. I. Radioautographic visualization after topical application of DL-[aH]norepinephrine, Brain Research, 5[ (1973) 141-160. 9 Descarries, L. et Schr6der, J. M., Fixation du tissu nerveux par perfusion '5. grand debih J. Microsc., 7 (1968) 281 286. l0 Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system, ill. The monoamine nerve terminal, Z. Zeltforsch., 65 (1965) 573-596. 11 Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system. IV. Distribution of monoamine nerve terminals in the central nervous system, Acta physiol, scand., 64, Suppl. 247 (1965) 37-85. 12 Fuxe, K., Hamberger, B. and H6kfelt, T., Distribution of noradrenaline nerve terminals in cortical areas of the rat, Brain Research, 8 (1968) 125-131. 13 Fuxe, K., H6kfelt, T. and Ungerstedt, U., Localization of indoiealkylamines in CNS. In S. Garattini and P. A. Shore (Eds.), Advances in Pharmacology, Vol. 6, Part A. Academic Press, New York, [ 968, pp. 235- 25 I. 14 Fuxe, K. and Jonsson, G., Further mapping of central 5ohydroxytryptamine neurons: studies with the neurotoxic dihydroxytryptamines. In E. Costa, G. L. Gessa and M. Sandier (Eds.), Serotonin ~- New Vistas. Advances in Biochemical Psychopharmaeology, Vol. 10, Raven Press. New York, 1974, pp. 1-12. 15 HSkfelt, T., Fuxe, K., Johansson, O. and Ljungdahl, ~,., Pharmaco-histochemical evidence of the existence of dopamine nerve terminals in the limbic cortex, Europ. J. Pharmacol., 25 (1974) 108-112. 16 lversen, L. L. and Bloom, F. E., Studies on the uptake of [aH]GABA and [aH]glycine in slices and homogenates of rat brain and spinal cord by electron microscopic autoradiography, Brabt Research, 41 (1972) 131 -143.
309 17 Kopriwa, B. M. and Leblond, C. P., Improvement in the coating technique of radioautography, J. Histochem. Cytochem., 10 (1962) 269-284. 18 Krieg, W. J. S., Connections of the cerebral cortex. I1. The albino rat. A. Topography of the cortical areas, J. comp. Neurol., 84 (1946) 221 275. 19 Krieg, W. J. S., Connections of the cerebral cortex. II. The albino rat. B. Structure of cortical areas, J. comp. NeuroL, 84 (1946) 277-323. 20 Kuhar, M. J. and Aghajanian, G. K., Selective accumulation of 3H-serotonin by nerve terminals of raphe neurones: an autoradiographic study, Nature (Lond.), 241 (1973) 187 189. 21 Kuhar, M. J., Aghajanian, G. K. and Roth, R. H., Tryptophan hydroxylase activity and synaptosomal uptake of serotonin in discrete brain regions after rnidbrain raphe lesions: correlations with serotonin levels and histochemical fluorescence, Brah~ Research, 44 (1972) 165 176. 22 Lapierre, Y., Beaudet, A., Demianczuk, N. and Descarries, L., Noradrenergic axon terminals in the cerebral cortex of rat. II. Quantitative data revealed by light and electron microscope radioautography of the frontal cortex, Brain Research, 63 (1973) 175 182. 23 Lindvall, O., Bj~3rklund, A., Moore, R. Y. and Stenevi, U., Mesencephalic dopamine neurons projecting to neocortex, Brahl Research, 81 (1974) 325 331. 24 Merrillees, N. C., Burnstock, G. and Holman, M. E., Correlation of fine structure and physiology of the innervation of smooth muscle in the guinea pig vas deferens, J. CellBiol., 19 (1963) 529 550. 25 Ritz6n, M., Quantitative fluorescence rnicrospectrophotometry of 5-hydroxytryptamine formaldehyde products in models and in mast ceils, Exp. Cell Res., 45 (1967) 178 194. 26 Saavedra, J. M., Brownstein, M. and Axelrod, J., A specific and sensitive enzymatic isotopic microassay for serotonin in tissues, J. Pharmacol. exp. Ther., 186 (1973) 508-515. 27 Snyder, S. H. and Bennet, J. P., Biochemical identification of the postsynaptic serotonin receptor in mammalian brain. In E. Usdin and W. E. Bunney (Eds.), Pre- andPostsynaptic Receptors, Dekker, New York, 1975, pp. 191-206. 28 Tennyson, V. M., Heikkila, R., Mytilineou, C., C6t6, L. and Cohen, G., 5-Hydroxydopamine 'tagged' neuronal boutons in rabbit neostriatum: interrelationship between vesicles and axonal membrane, Brain Research, 82 (1974) 341-348. 29 Ungerstedt, U., Stereotaxic mapping of the monoamine pathways in the rat brain, Acta physiol. scand., Suppl. 367 (1971) 1~-8.
301
QUANTITATIVE DATA ON SEROTONIN NERVE TERMINALS 1N A D U L T R A T N E O C O R T E X
ALAIN BEAUDET and LAURENT DESCARRIES Centre de Recherche en Sciences Neurologiques, Universitd de Montrdal, Montreal, H3C 31"8 ( Cana&t)
(Accepted January 5th, 1976)
SUMMARY Serotonin (5-HT) nerve terminals, specifically labeled with [aH]5-HT have been counted in light microscope radioautographs from the fronto-parietal neocortex of adult rats, following prolonged superfusions with relatively high concentrations of tracer (10 a M or 10 4 M), and after 15 or 30 days of radioautographic exposure. Comparative analysis of the results indicated that all 5-HT varicosities did not possess the same capacity to accumulate the exogenous amine. Nevertheless, superfusions with 10-4 M [aH]5-HT provided a sufficient loading of 5-HT boutons to allow their complete detection within any given layer of the neocortex. The labeled varicosities were found to be present within all cortical layers, except layer VI. Their total number was extrapolated to be approximately 1 x 10~/ cu.mm of cortex. This figure represented a mean incidence of one 5-HT nerve terminal/850-1400 cortical synapses (0.07-0.12%). The intralaminar density of 5-HT innervation increased progressively from layer V to layer I, in a distribution pattern suggestive of unspecific afferents. Based on endogenous 5-HT levels of 346 ± 32 ng/g for the fronto-parietal neocortex, the mean 5-HT content per varicosity could be estimated to be 0.35 × 10 a pg, and the concentration 1900 ng/g wet weight of terminal or 0.2~o. In view of their particular ultrastructural features 7, ubiquitous repartition and endogenous amine content, 5-HT nerve endings may be considered as capable of exerting a rather widespread influence in the neocortex.
INTRODUCTION Little information was hitherto available on the innervation of mammalian cerebral cortex by serotonin (5-HT) containing fibers. Histofluorescence studies in the rat have revealed the existence of 'scattered 5-HT terminals present in the super-
302 ticial layers "11, "most of them being found in the outer part of the molecular layer ::;. Further investigations, based on lesion experiments"ahel,e:~, have indicated tha~ these nerve endings originate from cell bodies situated in the raphe nuclei of the mesencephalon (groups B-7, B-8 and B-9 of Dahlstr6m and Fuxe6). High resolution radioautography has recently permitted the identification and ultrastructural characterization of 5-HT fibers in the neocortex of adult ratsL These terminal arborizations were shown to be more numerous than previously suspected, and dispersed throughout layers I V in the fronto-parietal region. The 5-HT axons were described as tenuous, unmyelinated processes, bearing small enlargements free of glial enwrapping. These varicosities were filled with vesicular organelles, bul, even when examined in serial thin sections, were seldom found to be engaged m typical synaptic relationships. Such structural features lead to the postulate that most cortical 5-HT varicosities could be in constant motion along their parent fibers, and might therefore release their biogenic amine in different tissue sites at different moments in time v. The main purpose of the present study was to seek precise quantitative data on the intracortical distribution of 5-HT terminals, and their endogenous amine content, in order to establish a basis for better understanding of the functional significance of 5-HT in cortex. MATERIALS AND METHODS Six adult male Sprague-Dawtey rats (body weight, 300 :~: 60 g), pretreated with a monoamine oxidase inhibitor (/3-phenylisopropyl hydrazine, 5 mg/kg, i.p., 18 and 2 h earlier), were anesthetized with Nembutal and submitted to 3 h cortical superfusion with tritiated serotonin ([aH]5-hydroxytryptamine <:G>- creatinine sulfate, specific activity 11-17.3 Ci/mM, obtained from Amersham-Searle), using a technique previously described in detail v,8. The tracer was applied unilaterally, on a region of the fronto-parietal neocortex corresponding to parts of areas 3, 4, 6 and 10 of Krieg 18. The concentration of [aH]5-HT in the superfusate was 10-~ M in 3 rats, and 10 4 M in the other 3. The cortical superfusions were immediately followed by intra-aortic perfusionfixation with 3.5'~'~,i glutaraldehyde in 0.05 M phosphate buffer 9. The superfused portions of cortex were excised from the cerebral hemispheres, sliced coronally into 1 ram-thick slabs, and post-fixed in osmium tetro×ide. Blocks including all 6 cortical layers were cut out from each slice, dehydrated in ethanol, and flat-embedded in Epon to provide a sectioning plane perpendicular to the pial surface. Two series of l #m-thick sections were obtained from each block, simultaneously processed for light microscope radioautography 17, and examined after 15 and 30 days of exposure, respectively. To obtain quantitative data on the intracortical distribution of serotonin nerve terminals, which have been shown to be specifically labeled in the above conditionsT: the silver grain aggregates present in light microscope radioautographs were systematically counted with the aid of a camera lucida. This was done on 6 sections from
303 each rat, after both durations of radioautographic exposure. Areas for counting were selected by means of a transparent overlay, superimposed on a scale drawing of the sections ( × 170). The labeled sites were charted within 8 squares, drawn on the overlay as a vertical series staggered in the horizontal plane. This preset pattern would be oriented with the upper side of the first square aligned along the pial edge of the section. Each square delineated an area of tissue 150/zm of side. Each of the upper 4 cortical layers was thus randomly sampled over its entire thickness 19 by 1 square, and layer V by 4 squares. Layer VI was not studied, since it was virtually devoid of tracer accumulations. The values were corrected for 1 sq. mm of radioautograph, and their statistical significance was analyzed with a two-tailed t-test. The validity of the sampling technique was verified in 9 radioautographs, and the results obtained by means of the overlay were found to be comparable ( ± 10 ~'/~i)to counts obtained from entire sections. In order to estimate the number of labeled terminals per volumetric unit of cortex, all reactive sites were considered as spheres. It was also assumed that a segment of labeled varicosity 0.1 # m in thickness had to be included within a I /~m-thick section to produce a detectable radioautographic reaction. Since the average diameter of cortical 5-HT varicosities had been measured to be 0.7/zm v, it could be inferred that 1/~m-thick sections spaced at 0.5/~m intervals would sample all reactive terminals contained in a given volume of tissue 22. The number of reactive varicosities being fairly constant in all sections prepared under similar experimental conditions (S.E.M. < 10~,/0, values for 1 cu.mm of cortex could be safely extrapolated by multiplying results for 1 sq. mm of radioautographs by 1000/(1 + 0.5) 666.66. To evaluate the serotonin content of 5-HT varicosities, endogenous levels of the indolamine were measured in cortical regions identical to those prepared for radioautography. The brains from 8 normal rats were rapidly removed after decapitation, frozen to a semi-solid consistency, and sliced coronally from 3 to 9 m m behind the frontal pole. Under a dissecting microscope, strips of the dorso-medial neocortex were excised from each slice, homogenized in 0.1 N HCI, and assayed for serotonin with the radio-enzymatic technique of Saavedra et al. 26. RESULTS Following superfusions of rat neocortex with 10 -4 or 10-5 M [3H]5-HT, accumulations of the tracer were visualized in light microscope radioautographs in the form of small and dense aggregates of silver grains (Fig. 1). These have been shown to designate axonal 5-HT varicosities, filled with vesicular organelles and making only sparse synaptic contacts 7. The intra-cortical distribution of the labeled varicosities is illustrated in Figs. 2 and 3, for the two molarities of [3H]5-HT in the superfusate, after either 15 or 30 days of radioautographic exposure. Fig. 2 emphasizes the influence of the tracer concentration. In every layer, the number of labeled 5-HT varicosities was greater after 10 -4 M than after 10 -5 M superfusions. However, in layers I - l l I , there were only slight differences between counts obtained after 30 days of exposure with 10 -5 M,
304
Fig. 1. Light microscope radioautograph from the neocortex of an adult rai, following 3 h stlpcr.fusion with 10 a M [:~HJ5-HT. Monoamine oxidase inhibited. Small and dense aggregates of >Hxev grains correspond to serotonin nerve terminals - i.e. axonal varicosities - in x~hich the tracer i~ highly concentrated. The diffuse labeling in background presumably represents unbound amine and/or labeled metabolites, Cortical layers I and 11. Exposure time: 2 weeks. Magnification: ~00
a n d those after 15 days with 10-i M [aH]5-HT. Fig. 3 underlines the effect o f ' r a d i o a u t o g r a p h i c exposure. W i t h l 0 :' M superfusions, d o u b l i n g the d u r a t i o n o f exposure p r o d u c e d a c o n s i d e r a b l e increase in the n u m b e r o f reactive sites within all cortical layers. In contrast, with 10 ~ M superfusions, there was no significant difference between counts o b t a i n e d after 15 or 30 days o f exposure. The total n u m b e r o f 5 - H T varicosities per volumetric unit o f cortex was therefore e x t r a p o l a t e d from all counts o b t a i n e d after superfusion with 10 i M [aH]5-HT (Table 1). When the n u m b e r o f labeled varicosities was expressed per cu.mm o f cortical layer ( i n t r a l a m i n a r density), the serotonin nerve terminals a p p e a r e d mostly c o n c e n t r a t e d within the u p p e r 4 layers. This density diminished progressively from layer 1 to layer I11, r e m a i n e d u n c h a n g e d in layer IV, a n d was the lowest in layer V. W h e n the values were expressed as a percentage o f the total n u m b e r o f 5-HT varicosities in cortex (intracortical repartition), layer V was f o u n d to contain the highest p r o p o r t i o n , its thickness being 4 times g r e a t e r than that o f each u p p e r layer. F r o m these d a t a , the overall density o f cortical 5 - H T innervation could be estimated at nearly I 10 ~; varicosities/cu.mm o f cortex. A c c o r d i n g to current
305 xlO~/mm 2
xlO~mm 2 20
15
10
5
0
0
5
10
15
20
25
30
i
E ~////W////////////W~//////A_~ ~q
w~2/////~/
~i ~///////////////~//wA-~ ~r ~////////////W~///h~
lO-SM
~
~ ~,/~1
10-4M
Fig. 2. Radioautographic distribution of 5-HT varicosities in the upper 5 layers of the fronto-parietal neocortex (I-V), following 3 h superfusions with 10 -~ M of 10 4 M [3H]5-HT. Counts obtained from radioautographs developed after 15 days (clear colums) and 30 days (hatched columns) of exposure. Each value represents the mean number (:k S.E.M.) for 18 1 /~m-thick sections from 3 rats, corrected for l sq. mm of radioautograph.
IO-SM
lO-aM
30
"E E
oJ
20
m57
eT'-.. o
x
]]I 5Z
10
o x
S[ 15
30 days
0
15
30 days
Fig. 3. Effect of the duration of radioautographic exposure on the number of axonal varicosities labeled in the upper 5 layers of the fronto-parietal neocortex (I V), following 3 h superfusions with 10 ~ M or 10 -4 M [3H]5-HT. Data collected as in Fig. 2.
TABLE I
Distribution o f 5 - H T nerve terminals in adult rat fronto-parietal neocortex Cortical layer
lntralaminar density (n x 106/eu. mm)
lntracortical repartition ( % o f totalnumber in cortex)
I
1,83
lI Ill IV V
1,47 1,12 1,10 0,57
24 19 14 14 29 *
* Value which takes into account the greater thickness of layer V.
306 evaluations of the total number of synapses in the neocortex of rodents (8.5--14 i0s/ cu.mm) ~,a, this figure represents a mean incidence of one 5-HT varicosity~g50 i400 nerve terminals. The biochemical determinations of endogenous serotonin in the frontoparietal neocortex provided a value of 346 32 ng/g. Since the indolamine is mainly concentrated within axonal varicosities, the average amount of 5-HT per terminal was approximated to be 0.35 10 a pg. Using a spherical volume of 0.18 cu./,~m, and a specific weight of l for 5-HT terminals, the average concentration of serotonin per g wet weight of 5-HT varicosity amounted to 1900 #g or 0.2 i"ii. DISCUSSION An earlier study has already demonstrated that all silver grain aggregates, visible in light microscope radioautographs prepared in the above described conditions, represent 5-HT axonal varicosities specifically labeled with [aH]5-HTV. In order to obtain a reliable quantitative assessment of the cortical serotonin innervation, it remained to be shown that most if not all 5-HT varicosities were amenable to radioautographic detection. This was investigated by varying both the concentration of [aH]5-HT in the superfusate, and the duration of radioautographic exposure: while the molarity of the tracer could affect the amount of [aH]5-HT available for uptake and storage by 5-HT nerve terminals, the duration of exposure would merely influence the intensity of the radioautographic reactions. Comparison of the results obtained after IO -a M and 10 -'~ M superfusions confirmed previous findings indicating that many cortical 5-HT varicosities required high ambient concentration of the tracer for radioautographic visualization 7. This was particularly evident within the deep layers (IV-V), where the local concentration of [aH]5-HT was reduced by a penetration gradient from the surface inward. The increased number of labeled varicosities observed after t0 -4 M superfusions further suggested that, within any layer of the cortex, all 5-HT terminals did not accumulate the same amount of [aH]5-HT. Indeed, after 10 -5 M superfusions, some 5-HT nerve endings could always be detected, whereas others were not. Moreover, numerous 5-HT varicosities, passing unnoticed after 10 -5 M superfusions and short radioautographic exposure, did reveal their presence after doubling the exposure time. This variable capacity to take up and retain [aH]5-HT could possibly be related to different functional states, in terms of endogenous 5-HT content and/or uptake and binding properties. It was however obvious that superfusions with l0 -4 M [aH]5-HT provided a massive loading of all 5-HT varicosities, since in this condition the duration of radioautographic exposure had no significant effect on the number of labeled terminals. This plateau was reached within all cortical layers, indicating that the paucity of reactive varicosities detected in layer VI was not solely due to insufficient penetration of the tracer. The quantitative data gathered after 10-4 M superfusions confirmed the ubiquitous repartition of 5-HT nerve terminals throughout layers I - V in the fronto-parietal
307 neocortex. The gradual increase in intralaminar density of 5-HT varicosities towards the upper layers could merely reflect the progressive axonal arborization of an unspecific afferent system. This pattern of innervation was markedly different from that of other types of monoamine-containing afferents. Within the same region, noradrenergic terminals predominate more clearly in the superficial layers12, 22, whereas dopaminergic terminals appear mainly confined to the deeper layers3,15, 23. The overall density of serotonin innervation, approximated at 1 × 106 varicosities/cu.mm of cortex, was 10 times higher than that estimated for noradrenergic afferents within the fronto-parietal region 22. It corresponds to an incidence of 0.070.12 % of all cortical nerve endings, in accordance with earlier quantitative estimations based on the proportion of synaptosomes labeled with [3H]5-HT in whole forebrain homogenates 2°. The same figure appears relatively low, however, when compared with the proportion of nerve terminals labeled in vitro with tritiated GABA (30~i of all cortical synapses4,~6). The sparsity of the serotonin innervation does not preclude a significant role in cortex, but rather suggests a particular mode of action, especially in view of the rarity of typical synaptic junctions made by the 5-HT varicosities v. Despite the tact that the tissue concentrations of both serotonin and norepinephrine do not greatly differ in the fronto-parietal cortex (346 vs. 228 ng/g7), the higher number of 5-HT terminals implies a much lower endogenous amine content per 5-HT varicosity (0.35 vs. 2.33 × 10-3 pg2Z). However, the intravaricose concentration of both serotonin and norepinephrine is in the same order of magnitude (0.2 vs. 0.3 ~/o), owing to the smaller volume of 5-HT nerve endings (0.18 vs. 0.8 cu./~m). The value of 1900 #g of 5-HT/g wet weight of varicosity falls in the lower range of concentrations showing a linear relationship with the intensity of yellow-fluorescence induced by the Falck and Hillarp technique ~. It is therefore in keeping with the increases in fluorescence intensity reported for cortical 5-HT varicosities after monoamine oxidase inhibition 1° and L-tryptophan pretreatment 21. In the neocortex, endogenous 5-HT can probably be released from all axonal varicosities where it is concentrated, and not only from the small proportion which make typical synaptic contacts 7. It is thus likely that the indolamine does not exert its effect solely upon restricted areas of pc)stsynaptic membrane differentiation, but can diffuse in tissue, to reach relatively distant targets. A similar mode of action has already been proposed for biogenic amines, in both the peripheraF 4 and central nervous system 2s. In cortex, it could account for the presence of a paradoxically high number of presumptive 'postsynaptic' 5-HT receptor sites 2v. The above quantitative data therefore lead to the suggestion that, despite their relative scarcity, 5-HT afferents might exert a rather widespread influence in the neocortex. The entire cortical volume might even be considered to be within reach of the indolamine, if 5-HT can diffuse from the varicosities for a distance of a few microns only. Moreover, the varicosities themselves could possess, as previously suggested, dynamic properties of translocation and/or reshaping along their parent fibers, allowing incessant fluctuation of their functional domains.
308 ACKNOWLEDGEMENTS T h e a u t h o r s are i n d e b t e d to Mrs. S y l v i a G a r c i a a n d M r . K e n n e t h C. W a t k i n s f o r skillful t e c h n i c a l assistance,
and
Messrs.
Robert
Pdloquin,
Daniel Cyr and
E d w a r d R u p n i k f o r g r a p h i c a n d p h o t o g r a p h i c w o r k . T h e y also t h a n k M i s s Lise F a r l e y a n d D r . J a c q u e s de C h a m p l a i n f o r t h e i r h e l p in the b i o c h e m i c a l d e t e r m i n a l i o n s o f c o r t i c a l s e r o t o n i n , a n d are g r a t e f u l to Dr. H e r b e r t H. J a s p e r for his i n s p i r i n g r e v i e w o f the m a n u s c r i p t . T h i s s t u d y was s u p p o r t e d by a F e l l o w s h i p (A.B.) a n d G r a n t M T - 3 5 4 4 (L.[).), f r o m the M e d i c a l R e s e a r c h C o u n c i l o f C a n a d a .
REFERENCES 1 Aghajanian, G. K. and Bloom, F. E., The formation of synaptic junctions in developing brain: a quantitative electron microscopic study, Brain Research, 6 (1967) 716 727. 2 And6n, N.-E., DahlstrSm, A., Fuxe, K., Larsson, K., Olson, L. and Ungerstedt, U., Ascending monoamine neurons to the telencephalon and diencephalon, Acta physiol, stand., 67 (1966) 313 326. 3 Berger, B., Tassin, J. P., Blanc, G., Moyne, M. A. and Thierry, A. M., Histochemical confirmation for dopaminergic innervation of the rat cerebral cortex after destruction of the noradrenergic ascending pathways, Brain Research, 81 (1974) 332-337. 4 Bloom, F. E. and lversen, L. L., Localizing aH-GABA in nerve terminals of rat cerebral cortex by electron microscopic autoradiography, Nature (Lond.), 229 (1971) 628-.630. 5 Cragg, B. G., The density of synapses and neurones in the motor and visual areas of the cerebral cortex, J. Anat. (Lond.), 101 (1967) 639-654. 6 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurons in the central nervous system, l. Demonstration of monoamines in the cell bodies of brain stem neurons, Acta physiol, scand., 62, Suppl. 232 (1964) 1-55. 7 Descarries, L., Beaudet, A. and Watkins, K. C., Serotonin nerve terminals in adult rat neocorte×, Brain Research, 100 (1975) 563-588. 8 Descarries, L. and Lapierre, Y., Noradrenergic axon terminals in the cerebral cortex of rat. I. Radioautographic visualization after topical application of DL-[aH]norepinephrine, Brain Research, 5[ (1973) 141-160. 9 Descarries, L. et Schr6der, J. M., Fixation du tissu nerveux par perfusion '5. grand debih J. Microsc., 7 (1968) 281 286. l0 Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system, ill. The monoamine nerve terminal, Z. Zeltforsch., 65 (1965) 573-596. 11 Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system. IV. Distribution of monoamine nerve terminals in the central nervous system, Acta physiol, scand., 64, Suppl. 247 (1965) 37-85. 12 Fuxe, K., Hamberger, B. and H6kfelt, T., Distribution of noradrenaline nerve terminals in cortical areas of the rat, Brain Research, 8 (1968) 125-131. 13 Fuxe, K., H6kfelt, T. and Ungerstedt, U., Localization of indoiealkylamines in CNS. In S. Garattini and P. A. Shore (Eds.), Advances in Pharmacology, Vol. 6, Part A. Academic Press, New York, [ 968, pp. 235- 25 I. 14 Fuxe, K. and Jonsson, G., Further mapping of central 5ohydroxytryptamine neurons: studies with the neurotoxic dihydroxytryptamines. In E. Costa, G. L. Gessa and M. Sandier (Eds.), Serotonin ~- New Vistas. Advances in Biochemical Psychopharmaeology, Vol. 10, Raven Press. New York, 1974, pp. 1-12. 15 HSkfelt, T., Fuxe, K., Johansson, O. and Ljungdahl, ~,., Pharmaco-histochemical evidence of the existence of dopamine nerve terminals in the limbic cortex, Europ. J. Pharmacol., 25 (1974) 108-112. 16 lversen, L. L. and Bloom, F. E., Studies on the uptake of [aH]GABA and [aH]glycine in slices and homogenates of rat brain and spinal cord by electron microscopic autoradiography, Brabt Research, 41 (1972) 131 -143.
309 17 Kopriwa, B. M. and Leblond, C. P., Improvement in the coating technique of radioautography, J. Histochem. Cytochem., 10 (1962) 269-284. 18 Krieg, W. J. S., Connections of the cerebral cortex. I1. The albino rat. A. Topography of the cortical areas, J. comp. Neurol., 84 (1946) 221 275. 19 Krieg, W. J. S., Connections of the cerebral cortex. II. The albino rat. B. Structure of cortical areas, J. comp. NeuroL, 84 (1946) 277-323. 20 Kuhar, M. J. and Aghajanian, G. K., Selective accumulation of 3H-serotonin by nerve terminals of raphe neurones: an autoradiographic study, Nature (Lond.), 241 (1973) 187 189. 21 Kuhar, M. J., Aghajanian, G. K. and Roth, R. H., Tryptophan hydroxylase activity and synaptosomal uptake of serotonin in discrete brain regions after rnidbrain raphe lesions: correlations with serotonin levels and histochemical fluorescence, Brah~ Research, 44 (1972) 165 176. 22 Lapierre, Y., Beaudet, A., Demianczuk, N. and Descarries, L., Noradrenergic axon terminals in the cerebral cortex of rat. II. Quantitative data revealed by light and electron microscope radioautography of the frontal cortex, Brain Research, 63 (1973) 175 182. 23 Lindvall, O., Bj~3rklund, A., Moore, R. Y. and Stenevi, U., Mesencephalic dopamine neurons projecting to neocortex, Brahl Research, 81 (1974) 325 331. 24 Merrillees, N. C., Burnstock, G. and Holman, M. E., Correlation of fine structure and physiology of the innervation of smooth muscle in the guinea pig vas deferens, J. CellBiol., 19 (1963) 529 550. 25 Ritz6n, M., Quantitative fluorescence rnicrospectrophotometry of 5-hydroxytryptamine formaldehyde products in models and in mast ceils, Exp. Cell Res., 45 (1967) 178 194. 26 Saavedra, J. M., Brownstein, M. and Axelrod, J., A specific and sensitive enzymatic isotopic microassay for serotonin in tissues, J. Pharmacol. exp. Ther., 186 (1973) 508-515. 27 Snyder, S. H. and Bennet, J. P., Biochemical identification of the postsynaptic serotonin receptor in mammalian brain. In E. Usdin and W. E. Bunney (Eds.), Pre- andPostsynaptic Receptors, Dekker, New York, 1975, pp. 191-206. 28 Tennyson, V. M., Heikkila, R., Mytilineou, C., C6t6, L. and Cohen, G., 5-Hydroxydopamine 'tagged' neuronal boutons in rabbit neostriatum: interrelationship between vesicles and axonal membrane, Brain Research, 82 (1974) 341-348. 29 Ungerstedt, U., Stereotaxic mapping of the monoamine pathways in the rat brain, Acta physiol. scand., Suppl. 367 (1971) 1~-8.