- Email: [email protected]
The cortical projection of the dorsal raphe nucleus: Some electrophysiological and pharmacological properties
Brain Research, 216 (1981) 61-71 © Elsevier/North-HollandBiomedical Press
61
THE CORTICAL PROJECTION OF THE DORSAL RAPHE NUCLEUS : SOME ELECTROPHYSIOLOGICAL AND P H A R M A C O L O G I C A L PROPERTIES
HANS-RUDOLF OLPE Ciba-Geigy Ltd., Biology Research Laboratory, Pharmaceuticals Division, Basel 4002 (Switzerland}
(Accepted December 4th, 1980) Key words: serotonin -- rat -- cingulate cortex -- rostral cortex -- serotonin antagonists
SUMMARY The effect of repetitive stimulation of the dorsal raphe nucleus (DRN) on the firing of spontaneously active neurons of the rostral and posterior cingulate cortex was investigated in untreated and serotonin-depleted rats under chloral hydrate anesthesia. In the untreated animals, the action of microiontophoreticaUy administered serotonin (5-HT) on cell firing was compared with the transsynaptically elicited effects. In the untreated animals, the main transsynaptic effect on neurons of all cell layers of both the rostral and cingulate cortex was an inhibitory one. In the cingulate cortex 50-70 ~o of the neurons were depressed, and in the rostral cortex 30-60 ~o responded with a temporary arrest of their discharge frequency to D R N stimulation. In contrast, only 5-10 ~o of frontal neurons and 0-5 ~o of the cingulate neurons were activated under these conditions. The inhibitory, transsynaptic effect on cingulate and rostral cortical cells was mimicked in most instances by microiontophoretically administered 5-HT. In the cingulate cortex 92 ~o of the neurons and in the rostral cortex 70 ~o of the neurons inhibited by D R N stimulation were depressed by locally administered 5-HT. The majority of the neurons activated by D R N stimulation were also depressed by microiontophoretically applied 5-HT. Furthermore, some 75~o of the neurons in the cingulate and some 47~ of all neurons tested in the rostral cortex that were not inhibited by D R N stimulation, were also depressed by microiontophoretically applied 5-HT. The peripheral 5-HT antagonists methysergide, cyproheptadine and GP 50 302, administered intraperitoneally, were found to be potent antagonists of transsynaptically elicited inhibitory effects in the cingulate cortex. The depressant action of microiontophoretically administered 5-HT on cingulate cortical neurons was antagonized by all three microiontophoretically administered 5-HT antagonists. In conclusion, the results of the present study are in keeping with recent anatomical observations demonstrating that the entire cortex is densely innervated by
62 5-HT axons reaching all cell layers. It is shown that these fibers exert an inhibitory influence on the actlwty of a high percentage of neurons m d~fferent layers of the rostral and cingulate cortex.
INTRODUCTION The presence of 5-HT-containing fiber systems in the neocortex of the rat brain has been documented in a number of anatomical studiesS,lO,11,19,~1. Until very recently, there appeared to be a striking discrepancy between the density of serotonergic fibers in the cortex as revealed by the aldehyde-induced hlstofluorescence method and the biochemically determined levels of 5-HT 24,27. However, in a recent study of the 5-HT innervation of the rat cerebral cortex, it was shown by means of a highly sensitive immunohistochemical technique that the neocortex is densely innervated by serotonergic fibers. In marked contrast to previous anatomical observations indicating that the sparse 5-HT-fibers are located mainly in the apical layersS,H, 2t, it was shown that the innervation is dense and fairly umform throughout all cortical cell layers 19. On the basis of these findings it was suggested that each neuron in the cortex may be directly under the influence of a 5-HT-containing axon, thus implying that this system may exert a profound influence on cortical activity la. Little is known about the function of this cortical fiber system. At the single cellular level a direct action of 5-HT on neuronal activity has been demonstrated in a number of conflicting microiontophoretic studies showing both excitatory3,14,25,26 and inhibitory 4,~5,~7,x8,~4effects on neurons of different cortical regions. There is an almost complete lack of reformation on the mode of action of 5-HT at a higher organizational level of the cortex. In view of the above-mentioned new anatomical findings on the intracortical distribution of the 5-HT-fiber systems, we investigated the effect of 5-HT on neurons across different cell layers of the cortex. Our main aim was to verify the hypothesis that the 5-HT-system is potentially capable of affecting practically all neurons in the cortex. The effects evoked by D R N stimulation on cortical neuronal activity were compared systematically with the action of microiontophoretically administered 5-HT given m different doses. Serotonin-medlated transsynaptic effects were differentiated from those mediated by other fiber systems by comparing results obtained in untreated rats w~th those obtained in 5-HT-depleted animals. Specific neurotransmitter antagomsts are an almost indispensable tool for the characterization of central receptors. Unfortunately, the efficiency of peripheral 5-HTantagonists has been questioned. Whereas several authors found these drugs active in antagonizing centrally evoked 5-HT-effects in various brain regions6, 2°,zs,~9 others reported some of them to be inactive in this respect in areas densely innervated by 5-HT-fibers 12. Since this controversy could well be partly due to technical problems inherent in microiontophoresis, we have carried out several series of experiments in which we have studied the action of systemically administered 5-HT-antagonists on transsynaptically elicited, inhabitory effects evoked by D R N stimulation. These 5-HT blockers were tested also as antagonists of microiontophoreticaUy administered 5-HT.
63 METHODS
Microiontophoresis and recording The experiments were performed on 66 male rats (Ti :RAI f (SPF)) weighing 280-350 g. They were anesthetized with chloral hydrate (400 mg/kg i.p.) and mounted in a stereotaxic frame. Multibarrel glass micropipettes (3 barrels) were used for recording action potentials of spontaneously active neurons and to administer 5-HT (0.04 M, pH 4.5), methysergide (0.03 M; pH 4), cyproheptadine (0.05 M; pH 4), GP 50302 (0.2 M; pH 4) or ?,-aminobutyric acid (GABA, 0.5 M; pH 3.5). All 5-HT blockers were dissolved in 0.1 ~o NaCI. A fresh solution of each drug was prepared before the beginning of each experiment. One barrel filled with 4 M NaC1 was used for recording action potentials. The two remaining barrels were used either to administer drugs or to balance ejection currents during the 5-HT applications. A retaining current of 10 nA was used throughout. Neurons were recorded in the following two cortical regions: posterior cingulate cortex (area 23 according to Krieg 16) A 5 mm, L 0.5-1.0 mm, V 0-2 mm from the cortical surface; rostral cortex A 10 mm, L 2.0 mm, V 0-2 mm. These coordinates refer to the atlas of De Groot 7. 5-HT was administered to active neurons at intervals of 3-6 min in 'doses' of 20, 40 and 80 nA during periods of 60 sec. In 8 experiments 5-HT was applied near to 80 neurons in both the rostral and cingulate cortex. 5-HT was applied to some 8-12 neurons per rat and brain region. In each experiment neurons were sampled in the course of two to three penetrations of the electrode through the cortex descending perpendicularly from the apical to the basal layers.
Stimulation of the rat DRN and record;ng of transsynaptic effects The area of the D R N was stimulated by means of a modified excentric 'Hess'type electrode consisting of two 0.2-mm thick, insulated wires separated by a distance of 1.5 mm. The free tip of the two electrodes was 1.2 mm long. The electrode was introduced stereotactically using the following coordinates: A 0.1 mm, L 0.75 mm bilaterally, V-1.6 mmT. The D R N was stimulated with a train of rectangular pulses lasting 300 msec at a rate of 25 pps. The individual pulses were 0.3 msec long and the plateau of the resulting current amounted to 1 mA. The effect of D R N stimulation on neurons of the rostral and cingulate cortex was investigated on poststimulus-time histograms (PSTH), which were constructed on a Nicolet 535 Data system. In each brain region, a PSTH from 80 neurons was obtained from 8 untreated control animals and from 8 5-HT-depleted. animals. In the control animals, the effect of microiontophoretically applied 5-HT was studied on each cell as described above. The 5-HTdepleted rats were treated with a daily dose of 300 mg/kg i.p. ofp-chlorophenylalanine (PCPA) on 3 consecutive days prior to the experiment. Each PSTH consisted of 16 sweeps lasting 5 sec. They were plotted by means of an XY-recorder and analyzed according to the criteria described previously~L Briefly, the effect of D R N stimulation was qualified either as inhibitory or excitatory, or as ineffective. The cortex was arbitrarily divided into four zones defined by their distance from the cortical surface, as follows: 0-500/~m, 501-1000 #m, 1001-1500/~m and 1501-2000/~m.
64 The action of the 5-HT-antagonists methysergide, cyproheptadme and G P 50 302, administered systemically, was investigated in 3 series of experiments. In each serms conmsting of 8-10 experiments, one cell showing a pronounced inhibition of cell firing following D R N stimulation was recorded per ammal. If the length of thin inhibition was constant during at least two consecutive control registrations, then the compound was administered systemically m a cumulative manner at intervals of 20 mln. A drug was found to be active if it reduced the period of firing inh]bitlon by at least 50 %. cincjulate cortex PCPA-treated
untreated
depth in cortex p m
n u m b e r of neurons
0-500
n u m b e r of neurons
2°
i
501- 1000
/
1001-1500
20
1501- 2000
20
28
t t t t
24
~e
20
percent of neurons
roetral c~rtex untreated
PCPA-treated
number of neurons
denim
cortex pm 0-500
2O
501-1000
number of ~UI~Drts
24
\
29
% ,
1001-1500
49
I I I I
1501-
percent of neurons
Fig. 1. Profiles through the cm$ulate and rostral cortex of PCPA-treated and control rats showing the percentage of neurons being inhibited (solid line), excited (dotted line), or unaffected (dashext line) by D R N stimulation.
65 At the end of 5 experiments, the brains were removed and fixed in 4 ~o buffered formaline. The location of the stimulation electrode was verified on unstained cryostat sections.
Biochemical determinations Cortical 5-HT and 5-hydroxyindoleacetic acid (5-HIAA) levels were determined in 6 PCPA-treated and two untreated animals according to the method of Curzon and Green 5. RESULTS
Histological verification of the stimulation sites In all brains investigated the stimulation-electrode was found to be located in the area of the D R N .
The effect of the PCPA-treatment on cortical 5-HT and 5-HIAA levels The 5-HT levels in the cortex were reduced from 413 4- 35 ng/g wet weight in control rats to 68.5 4- 7.2 ng/g in PCPA-treated rats ( - - 8 4 ~o). The 5-HIAA levels were reduced from 299 4- 16.5 ng/g wet weight to 35 4- 4.2 ng/g in the 5-HT depleted rats.
Transsynaptically mediated effects on cortical neurons Profiles through the cingulate and rostral cortex of both untreated and treated animals showing the percentages of activated, inhibited and unaffected neurons are depicted in Fig. 1. In the cingulate cortex of untreated rats, some 50-70 ~o of all spontaneously active neurons displayed a temporary arrest of firing in response to cincjulate cortex depth m cortex Pm
0-500
501 - 1000
rostral cortex n u m b e r of
number of neurons
i/ •.', i:
\/
neurons
-
/
'7'\ il
1001 - 1500
l
I I
l
1501 - 2000
l
~ 1~ 2b ab,b 5b ~ ~ 8b 9b~o
~ ~ 2b 3b 4b 5b 8b ~b 8b ~ o p ~ e g n t of nemrotm
Fig. 2. Profiles through the cingulate and rostral cortex depicting the percentage of neurons which are either inhibited (solid line), excited (dotted line) or unaffected (dashed line) by microiontophoretically administered 5-HT. 5-HT was applied during 60 sec with an ejection current of 80 nA.
66 20nA
A
40nA
80nA
B
20-
20nA ~
40nA ~
80nA ~
d 10-
I---I
1 minute
I minute
20n.A C
20n~
40n.._A
40~nA 8On...A
80n._.A
D
i
2o-
2o-,
t
t lO-
,o-
Q. I'--I
I-'--I
I minute
1 minute
Fig. 3. Typical response of rostral cortical neurons to m]croiontophoretically admimstered 5-HT. A: pure, dose-dependent depression of cell firing. B: pure, dose-dependent excitation. C: mixed response, following an ejection with currents of either 40 or 80 nA the period of cell depression is followed by a period of increased cell tiring. D: mixed response; with an ejection current of 80 n A a short pronounced excitatory effect is observed followed by an inhibition.
D R N stimulation. The highest percentage of inhibited cells was noted in the upper 500 /zm thick zone. In PCPA-treated, 5-HT-depleted animals, the percentage of inhibited neurons was strongly reduced in the entire cortex. Only one out of 100 neurons increased its firing rate in response to D R N st]mulation in untreated rats. No excitatory transsynaptic effects were seen in the 5-HT-depleted rats. Exact latencies of inhibition were not determined since the stimulation consisted of a train of pulses. In the rostral cortex some 30-60 ~o of all neurons investigated were inhibited by D R N stimulation. More neurons (50-60 ~o) were affected in the deep cell layers than in the upper (30-35 ~ ) layers. Only 4 out of 99 neurons tested were excited following such stimulation. Again, in the 5-HT depleted rats the number of inhibited neurons was significantly reduced in both the upper and lower cell layers. Two out of the 138 neurons investigated in this area were excited in the treated animals.
The action of microiontophoretically administered 5-HT Profiles through the cingulate and rostral cortex showing the percentages of
67
Control
¢n
15minutes after 3mg/kg i p of Cyproheptadlne
15minutes after l O m g / k g I.p of Cyproheptadine I period of stimulation
I
1 second
Fig. 4. The blocking action of systemicallyadministered cyproheptadine on transsynaptically induced inhibition of a cingulate cortical neuron evoked by stimulation of the DRN is shown.
neurons that were either inhibited or excited, or no affected by microiontophoreticaUy administered 5-HT are presented in Fig. 2. Similarly to the results on transsynaptic effects, inhibition of cell firing was the main action evoked by 5-HT on neurons in both brain regions. Furthermore, on the average, a higher percentage of depressed neurons was seen in the cingulate (80-90 70) than in the rostral cortex (50-60 70). Four different types of response to 5-HT were observed. The most frequent was a pure dosedependent reduction of cell firing with a slow onset of action and a recovery period often lasting up to 5 min (Fig. 3A). A few cells showed a pure, dose-dependent excitatory response characterized by its long recovery time (Fig. 3B). In some rare instances, a mixed type of response was noted. In these cases, cell depression occurred at 'doses' of 20 and 40 nA. However, when ejected with a current of 80 nA an initial phase of cell inhibition was followed by a short period of cell excitation (Fig. 3C). It was difficult to judge whether this was a true dual effect, or whether the excitation was a rebound phenomenon (Fig. 3C). Finally, on a few cells at a dose of 80 nA, there was an initial, brief phase of excitation followed by inhibition (Fig. 3D). The two latter types of neuron were classified as 'inhibited' neurons. Microiontophoretically administered 5-HT depressed 73 70 of the neurons in the cingulate and some 47 % of all neurons in the rostral cortex that were not affected by D R N stimulation. Furthermore, the majority of the transsynaptically activated neurons were equally depressed by 5-HT.
Antagonism of transsynaptic DRN-elicited effects by 5-HT antagonists All 3 compounds tested blocked the transsynaptically elicited depression of cingulate cortical cell firing in a dose-dependent fashion (Table I). Whereas GP 50 302 already displayed effects at a dose of 0.1 mg/kg i.p., methysergide was only active at a dose of 3 mg/kg and crypoheptadine at a dose of 10 mg/kg only (Fig. 4). Recovery could not be observed, since the effect usually outlasted the period during which the
68 TABLE I
The results on the action of three putative 5-HT antagonists on transsynaptically ehcited depression of cingulate cortical neurons The drugs were administered in a cumulattve manner. A drug was considered to be active at a g~ven dose if it reduced the duration o f the DRN-ehclted firing depression by 50 % or more.
GP 50 302
Methysergide
0.03 01 1.0 O. 1 mg/kg mg/kg rng/kg rng/kg
Number of neurons with no effect 2
2
N u m b e r of neurons with transsynapt~c blockade
3
3
Cyproheptadme
l0 30 0 1 rng/kg rng/kg rng/kg
10 rng/kg
3.0 rng/kg
4
3
4
1
I
I
3
5
10 mg/kg
4
neurons could be recorded. With all three compounds the latency of the blocking effect ranged from 10 to 15 min.
The action of microiontophoretically administered 5-HT antagonists on 5-HT elicited neuronal depression All 3 putative 5-HT-antagonists blocked the depressant action of microiontophoretically administered 5-HT on the majority of neurons (Fig. 5). Methysergide and cyproheptadine, when ejected with currents of 40-90 nA, often slightly depressed the activity of the neurons. Methysergide antagonized the 5-HT effect on 5 out of 7 neurons. Cyproheptadine blocked the cell inhibiting effect of 5-HT on 4 out of 6 Cyproheptadme 80nA
8
Methysergide 80nA 5-HT/3OnA
GP 50302 80nA • 5-H..T/3OnA
~
,
1 minute
Fig. 5. The b i l k i n g effect of microiontophoretically administered 5-HT antagomsts on 5-HT ehcited depression o f cingulate cortical neurons is shown.
69 neurons and GP 50 302 was active on 7 out of 8 neurons tested. None of these drugs antagonized the depressant action of GABA when ejected with currents that blocked the depressant action of 5-HT (n = 5) (data not shown). DISCUSSION Stimulation of the DRN and the microiontophoretic administration of 5-HT both resulted in a temporary depression of the firing rate of a fairly high percentage of neurons in apical as well as in basal cell layers. These findings confirm previous studies showing a predominance of inhibitory over excitatory effects elicited by the microiontophoretic administration of 5-HT 4,17,18,2a. In about 360 neurons in both the cingulate and the rostral cortex of rats treated acutely or chronically with various antidepressants, we had obtained similar results23: depression of cell firing was the most common effect and as in the present study, more excitatory responses were noted in the rostral than in the cingulate cortex. It seems unlikely that these excitatory responses represent artefacts since they were very pronounced and mostly outlasted the period of drug application. With the exception of one neuron which was excited transsynaptically as well as by the microiontophoretically administered 5-HT, all other neurons excited by microiontophoretically administered 5-HT were either inhibited or unaffected by DRN stimulation. Thus one may tentatively assume that the excitatory effects obtained upon the microiontophoretic administration of 5-HT may be mediated indirectly through inhibition of an inhibitory interneuron or through presynaptic mechanisms leading to a reduction of an inhibitory input. The dual effects on cell firing seen occasionally when 5-HT was applied with high currents has been reported previously17. They may indicate the interaction of direct and indirect effects. The results of the present study are in keeping with those of a previous one 2s showing that stimulation of the median raphe nucleus causes a reduction in the firing rate of the majority of cerebral cortical neurons tested. The depressant effect of microiontophoretically administered 5-HT and the transsynaptically elicited 5-HT effects were blocked by the 5-HT antagonist metergoline. The present findings lend support to the hypothesis advanced on the basis of anatomical data, that the serotonergic fibers can affect many cells in different cortical cell layerslL The physiological significance of this ubiquitious depressant effect is difficult to evaluate. It may well be that the serotonergic fibers have other additional effects on metabolism or at the information-processing level that are undetectable with the present method. In particular, the interpretation of the results is hampered by the paucity of information on the exact site of 5-HT innervation and its relationship with other afferent systems. The cortical serotonergic innervation originates from neurons of both the dorsal and median raphe nucleil,lZ, 21. To our knowledge, it is not known whether any of the two brain areas investigated is preferentially innervated by one of the two raphe nuclei. Both the noradrenergic fibers deriving from the locus coeruleus and the serotonergic fibers innervate the entire neocortex and, apart from marginal regional variations, it appears that both systems innervate all cortical cell layers. In previous
70 studies we showed that the noradrenerglc system exerts an inhibitory influence on cells throughout all layers of both the cmgulate and the rostral cortexeL Although the number of neurons sampled in these stu&es may have been too low to reflect true innervation density, it is interesting to compare the profiles obtained on the serotonergic and noradrenergic fiber systems. In the rostral cortex we found that both the noradrenergic and the serotonergic fiber system affect more neurons )n deeper than in the upper cell layers. In the cingulate cortex we observed that the actwation of the noradrenergic fibers affects a similar percentage of neurons in different layers 9, whereas the serotonergic fibers appear to affect more neurons m the apical than m the basal cortical layers. The physiological relevance of these findings remains to be elucidated. The three 5-HT antagonists methysergide, crypoheptadine and GP 50 302 transsynaptically blocked the effects on cingulate cortical neurons produced by D R N stimulation in a dose-dependent fashion. GP 50 302 ~s a potent 5-HT antagomst blocking centrally as well as peripherally mediated 5-HT effects (A. Stula, personal communication). On the basis of the present results, it appears that these transsynapt~c effects are mediated by 5-HT-contalning axons. Therefore, the present study shows that these compounds act as 5-HT antagonists in this brain area. This conclusion is coroborated by the microlontophoretic experiments m which we showed that these 5HT antagomsts block the action of 5-HT in the cingulate cortex. Whether or not regional &fferences may account for some of the discrepancies reported in literature remains an open question. It is conceivable that the drugs act at lower doses than those found to be active in the present study, since they were administered in a cumulative manner at relatively short intervals. These intervals may have been too short to observe the maximum effect at a given dose. Methyserglde was found to be about 10 t~mes more potent m v~vo than cyproheptadine. Interestingly, m an m v~tro binding study 2 it was found that methysergide was about 10 times more potent than crypoheptadme in &splacing [aH]5-HT from rat cerebral cortical membranes. ACKNOWLEDGEMENTS I would like to thank Drs. P. C. Waldmeier and B. Fehr for the biochemxcal analysis. I would like to express my gratitude to Dr. V. Baltzer for helpful &scussions and critical reading of the manuscript.
REFERENCES 1 Azmitia, E. C. and Segal, M., An autoradiographlc analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat, J. comp. NeuroL, 179 (1978) 641-668. 2 Bennett, J. P. and Snyder, S. H., Serotonin and lysergic acid ðylamide binding in rat brain membranes" relationship to postsynaptic serotonin receptors, Molec. PharmacoL, 12 (1976) 373-389. 3 Bradshaw, C. M., Roberts, M. H. T. and Szabadi, E., Effect of tncychc antidepressants on monoamine responses of single cortical neurons, Brit. J. Pharmacol., 41 (1971) 394-395. 4 Crawford, J. M. and Curtis, D. R., Pharmacologicalstudies of feline betz cells, J. Physiol. (Lond), 186 (1966) 121-138.
71 5 Curzon, G. and Green, A. R., Rapid method for the determination of 5-hydroxytryptamine and 5-hydroxyindolacetic acid in small regions of the brain, Brit. J. Pharmacol., 39 (1970) 653-655. 6 Davies, J. and Tongroach, P., Neuropharmacological studies on the nigro-striatal and raphestriatal system in the rat, Eurp. J. Pharmacol., 51 (1978) 91-100. 7 De Groot, J., The Rat Forebrain in Stereotaxic Coordinates, North-Holland, Amsterdam, 1972. 8 Descarries, L., Beaudet, A. and Watkins, K. C., Serotonin nerve terminals in adult rat neocortex, Brain Research, 100 (1975) 563-588. 9 Dillier, N., Laszlo, J., Mi~ller, B., Koclla, W. P. and Olpe, H.-R., Activation of an inhibitory noradrenergic pathway projecting from the locus cocruleus to the cingulate cortex of the rat, Brain Research, 154 (1978) 61-68. 10 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) 36-85. 11 Fuxe, K., Hokfeit, T. and Ungerstedt, U., Localization of lndoicalkylamines in CNS. In S. Garattini and P. A. Shore (Eds.), Advances in Pharmacology, Academic Press, New York, 1968, pp. 235-251. 12 Haigler, H. J. and Aghajanian, G. K., Serotonin receptors in the brain, Fed. Proc., 36 (1977) 2159-64. 13 Jacobs, B. L., Foote, S. L. and Bloom, F. E., Differential projections of neurons within the dorsal raphe nucleus of the rat: a horseradish pcroxidase (HRP) study, Brain Research, 147 (1978) 149-153. 14 Johnson, E. S., Roberts, M. H. T. and Straughan, D. W., The responses of cortical neurons to monoamines under differing anaesthetic conditions, J. Physiol. (Lond.), 203 (1969) 261-280. 15 Jordan, L. M., Frederickson, R. C. A., Philhs, J. W. and Lake, N., Microelectophoresis of 5-hydroxytryptamine: a clarification of its action on cerebral cortical neurones, Brain Research, 40 (1972) 552-558. 16 Kreig, W. J. S., Connections of the cerebral cortex. I. The albino rat. J. comp. Neurol., 84 (1946) 221-275. 17 Krnjevic, R. and Phillis J. W., Actions of certain amines on cerebral cortical neurones, Brit. J. PharmacoL, 10 (1963) 471--490. 18 Legge, K. F., Randic, M. and Straughan, D. W., The pharmacology of neurones in the pyriform cortex, Brit. J. PharmacoL, 26 (1966) 87-107. 19 Lidov, H. G. W., Grzanna, R. and Molliver, M. E , The serotonin innervation of the cerebral cortex in the rat - - an lmmunohistochemical analysis, Neuroscience, 5 (1980) 207-227. 20 McCall R. B. and Aghajanian, G. K., Serotonergic facilitation of facial motoneuron excitation, Brain Research, 169 (1979) 11-27. 21 Moore, R. Y., Halaris, A. E. and Jones, B. E., Serotonin neurons of the midbrain raphe: ascending projections, J. comp. Neurol., 180 (1978) 417-438. 22 Olpe, H.-R., Glatt, A., Laszlo, J. and Schellenberg, A., Some electrophyslological and pharmacological properties of the corUcal, noradrenerg~c projection of the locus coeruleus in the rat, Brain Research, 186 (1980) 9-19. 23 Olpe, H.-R. and Schellenberg, A., The sensitivity of cortical neurons to serotonln: effect of chronic treatment with antidepressants, serotonin uptake and monoamine oxldase blocking drugs, J. neural Transm., in press. 24 Phdhps, J. W. and York, D. H., Strychnine block of neural and drug-induced inhibition m the cerebral cortex, Nature (Lond.), 216 (1967) 922-923. 25 Roberts, M. H. T. and Straughan, D. W., Excitation and depression of cortical neurones by 5hydroxytryptamine, J. Physiol. (Lond.), 193 (1967) 269-294. 26 Roberts, M. H. T. and Straughan, D. W., An excitant effect of 5-hydroxytryptamine on single cortical neurones, J. Physiol. (Lond.), 188 (1967). 27 Saavedra, J. M., Brownstein, M. H. and Palkovits, M., Serotonin dlstributton in the hmblc system of the rat, Bram Research, 79 (1974) 437-441. 28 Sastry, B. S. R. and Phillis, J. W , Inhibition of cerebral cortical neurones by a 5-hydroxytryptammergic pathway from median raphe nucleus, Canad. J. Physiol Pharmacol., 55 (1977) 737-743. 29 Segal, M., 5-HT antagomsts in rat hippocampus, Brain Research, 103 (1976) 161-166. 30 Wang, R. Y. and Aghajanian, G. K., Inhibition of neurons in the amygdala by dorsal raphe stimulation: mediation through a direct serotonergic pathway, Brain Research, 120 (1977) 85-102.
61
THE CORTICAL PROJECTION OF THE DORSAL RAPHE NUCLEUS : SOME ELECTROPHYSIOLOGICAL AND P H A R M A C O L O G I C A L PROPERTIES
HANS-RUDOLF OLPE Ciba-Geigy Ltd., Biology Research Laboratory, Pharmaceuticals Division, Basel 4002 (Switzerland}
(Accepted December 4th, 1980) Key words: serotonin -- rat -- cingulate cortex -- rostral cortex -- serotonin antagonists
SUMMARY The effect of repetitive stimulation of the dorsal raphe nucleus (DRN) on the firing of spontaneously active neurons of the rostral and posterior cingulate cortex was investigated in untreated and serotonin-depleted rats under chloral hydrate anesthesia. In the untreated animals, the action of microiontophoreticaUy administered serotonin (5-HT) on cell firing was compared with the transsynaptically elicited effects. In the untreated animals, the main transsynaptic effect on neurons of all cell layers of both the rostral and cingulate cortex was an inhibitory one. In the cingulate cortex 50-70 ~o of the neurons were depressed, and in the rostral cortex 30-60 ~o responded with a temporary arrest of their discharge frequency to D R N stimulation. In contrast, only 5-10 ~o of frontal neurons and 0-5 ~o of the cingulate neurons were activated under these conditions. The inhibitory, transsynaptic effect on cingulate and rostral cortical cells was mimicked in most instances by microiontophoretically administered 5-HT. In the cingulate cortex 92 ~o of the neurons and in the rostral cortex 70 ~o of the neurons inhibited by D R N stimulation were depressed by locally administered 5-HT. The majority of the neurons activated by D R N stimulation were also depressed by microiontophoretically applied 5-HT. Furthermore, some 75~o of the neurons in the cingulate and some 47~ of all neurons tested in the rostral cortex that were not inhibited by D R N stimulation, were also depressed by microiontophoretically applied 5-HT. The peripheral 5-HT antagonists methysergide, cyproheptadine and GP 50 302, administered intraperitoneally, were found to be potent antagonists of transsynaptically elicited inhibitory effects in the cingulate cortex. The depressant action of microiontophoretically administered 5-HT on cingulate cortical neurons was antagonized by all three microiontophoretically administered 5-HT antagonists. In conclusion, the results of the present study are in keeping with recent anatomical observations demonstrating that the entire cortex is densely innervated by
62 5-HT axons reaching all cell layers. It is shown that these fibers exert an inhibitory influence on the actlwty of a high percentage of neurons m d~fferent layers of the rostral and cingulate cortex.
INTRODUCTION The presence of 5-HT-containing fiber systems in the neocortex of the rat brain has been documented in a number of anatomical studiesS,lO,11,19,~1. Until very recently, there appeared to be a striking discrepancy between the density of serotonergic fibers in the cortex as revealed by the aldehyde-induced hlstofluorescence method and the biochemically determined levels of 5-HT 24,27. However, in a recent study of the 5-HT innervation of the rat cerebral cortex, it was shown by means of a highly sensitive immunohistochemical technique that the neocortex is densely innervated by serotonergic fibers. In marked contrast to previous anatomical observations indicating that the sparse 5-HT-fibers are located mainly in the apical layersS,H, 2t, it was shown that the innervation is dense and fairly umform throughout all cortical cell layers 19. On the basis of these findings it was suggested that each neuron in the cortex may be directly under the influence of a 5-HT-containing axon, thus implying that this system may exert a profound influence on cortical activity la. Little is known about the function of this cortical fiber system. At the single cellular level a direct action of 5-HT on neuronal activity has been demonstrated in a number of conflicting microiontophoretic studies showing both excitatory3,14,25,26 and inhibitory 4,~5,~7,x8,~4effects on neurons of different cortical regions. There is an almost complete lack of reformation on the mode of action of 5-HT at a higher organizational level of the cortex. In view of the above-mentioned new anatomical findings on the intracortical distribution of the 5-HT-fiber systems, we investigated the effect of 5-HT on neurons across different cell layers of the cortex. Our main aim was to verify the hypothesis that the 5-HT-system is potentially capable of affecting practically all neurons in the cortex. The effects evoked by D R N stimulation on cortical neuronal activity were compared systematically with the action of microiontophoretically administered 5-HT given m different doses. Serotonin-medlated transsynaptic effects were differentiated from those mediated by other fiber systems by comparing results obtained in untreated rats w~th those obtained in 5-HT-depleted animals. Specific neurotransmitter antagomsts are an almost indispensable tool for the characterization of central receptors. Unfortunately, the efficiency of peripheral 5-HTantagonists has been questioned. Whereas several authors found these drugs active in antagonizing centrally evoked 5-HT-effects in various brain regions6, 2°,zs,~9 others reported some of them to be inactive in this respect in areas densely innervated by 5-HT-fibers 12. Since this controversy could well be partly due to technical problems inherent in microiontophoresis, we have carried out several series of experiments in which we have studied the action of systemically administered 5-HT-antagonists on transsynaptically elicited, inhabitory effects evoked by D R N stimulation. These 5-HT blockers were tested also as antagonists of microiontophoreticaUy administered 5-HT.
63 METHODS
Microiontophoresis and recording The experiments were performed on 66 male rats (Ti :RAI f (SPF)) weighing 280-350 g. They were anesthetized with chloral hydrate (400 mg/kg i.p.) and mounted in a stereotaxic frame. Multibarrel glass micropipettes (3 barrels) were used for recording action potentials of spontaneously active neurons and to administer 5-HT (0.04 M, pH 4.5), methysergide (0.03 M; pH 4), cyproheptadine (0.05 M; pH 4), GP 50302 (0.2 M; pH 4) or ?,-aminobutyric acid (GABA, 0.5 M; pH 3.5). All 5-HT blockers were dissolved in 0.1 ~o NaCI. A fresh solution of each drug was prepared before the beginning of each experiment. One barrel filled with 4 M NaC1 was used for recording action potentials. The two remaining barrels were used either to administer drugs or to balance ejection currents during the 5-HT applications. A retaining current of 10 nA was used throughout. Neurons were recorded in the following two cortical regions: posterior cingulate cortex (area 23 according to Krieg 16) A 5 mm, L 0.5-1.0 mm, V 0-2 mm from the cortical surface; rostral cortex A 10 mm, L 2.0 mm, V 0-2 mm. These coordinates refer to the atlas of De Groot 7. 5-HT was administered to active neurons at intervals of 3-6 min in 'doses' of 20, 40 and 80 nA during periods of 60 sec. In 8 experiments 5-HT was applied near to 80 neurons in both the rostral and cingulate cortex. 5-HT was applied to some 8-12 neurons per rat and brain region. In each experiment neurons were sampled in the course of two to three penetrations of the electrode through the cortex descending perpendicularly from the apical to the basal layers.
Stimulation of the rat DRN and record;ng of transsynaptic effects The area of the D R N was stimulated by means of a modified excentric 'Hess'type electrode consisting of two 0.2-mm thick, insulated wires separated by a distance of 1.5 mm. The free tip of the two electrodes was 1.2 mm long. The electrode was introduced stereotactically using the following coordinates: A 0.1 mm, L 0.75 mm bilaterally, V-1.6 mmT. The D R N was stimulated with a train of rectangular pulses lasting 300 msec at a rate of 25 pps. The individual pulses were 0.3 msec long and the plateau of the resulting current amounted to 1 mA. The effect of D R N stimulation on neurons of the rostral and cingulate cortex was investigated on poststimulus-time histograms (PSTH), which were constructed on a Nicolet 535 Data system. In each brain region, a PSTH from 80 neurons was obtained from 8 untreated control animals and from 8 5-HT-depleted. animals. In the control animals, the effect of microiontophoretically applied 5-HT was studied on each cell as described above. The 5-HTdepleted rats were treated with a daily dose of 300 mg/kg i.p. ofp-chlorophenylalanine (PCPA) on 3 consecutive days prior to the experiment. Each PSTH consisted of 16 sweeps lasting 5 sec. They were plotted by means of an XY-recorder and analyzed according to the criteria described previously~L Briefly, the effect of D R N stimulation was qualified either as inhibitory or excitatory, or as ineffective. The cortex was arbitrarily divided into four zones defined by their distance from the cortical surface, as follows: 0-500/~m, 501-1000 #m, 1001-1500/~m and 1501-2000/~m.
64 The action of the 5-HT-antagonists methysergide, cyproheptadme and G P 50 302, administered systemically, was investigated in 3 series of experiments. In each serms conmsting of 8-10 experiments, one cell showing a pronounced inhibition of cell firing following D R N stimulation was recorded per ammal. If the length of thin inhibition was constant during at least two consecutive control registrations, then the compound was administered systemically m a cumulative manner at intervals of 20 mln. A drug was found to be active if it reduced the period of firing inh]bitlon by at least 50 %. cincjulate cortex PCPA-treated
untreated
depth in cortex p m
n u m b e r of neurons
0-500
n u m b e r of neurons
2°
i
501- 1000
/
1001-1500
20
1501- 2000
20
28
t t t t
24
~e
20
percent of neurons
roetral c~rtex untreated
PCPA-treated
number of neurons
denim
cortex pm 0-500
2O
501-1000
number of ~UI~Drts
24
\
29
% ,
1001-1500
49
I I I I
1501-
percent of neurons
Fig. 1. Profiles through the cm$ulate and rostral cortex of PCPA-treated and control rats showing the percentage of neurons being inhibited (solid line), excited (dotted line), or unaffected (dashext line) by D R N stimulation.
65 At the end of 5 experiments, the brains were removed and fixed in 4 ~o buffered formaline. The location of the stimulation electrode was verified on unstained cryostat sections.
Biochemical determinations Cortical 5-HT and 5-hydroxyindoleacetic acid (5-HIAA) levels were determined in 6 PCPA-treated and two untreated animals according to the method of Curzon and Green 5. RESULTS
Histological verification of the stimulation sites In all brains investigated the stimulation-electrode was found to be located in the area of the D R N .
The effect of the PCPA-treatment on cortical 5-HT and 5-HIAA levels The 5-HT levels in the cortex were reduced from 413 4- 35 ng/g wet weight in control rats to 68.5 4- 7.2 ng/g in PCPA-treated rats ( - - 8 4 ~o). The 5-HIAA levels were reduced from 299 4- 16.5 ng/g wet weight to 35 4- 4.2 ng/g in the 5-HT depleted rats.
Transsynaptically mediated effects on cortical neurons Profiles through the cingulate and rostral cortex of both untreated and treated animals showing the percentages of activated, inhibited and unaffected neurons are depicted in Fig. 1. In the cingulate cortex of untreated rats, some 50-70 ~o of all spontaneously active neurons displayed a temporary arrest of firing in response to cincjulate cortex depth m cortex Pm
0-500
501 - 1000
rostral cortex n u m b e r of
number of neurons
i/ •.', i:
\/
neurons
-
/
'7'\ il
1001 - 1500
l
I I
l
1501 - 2000
l
~ 1~ 2b ab,b 5b ~ ~ 8b 9b~o
~ ~ 2b 3b 4b 5b 8b ~b 8b ~ o p ~ e g n t of nemrotm
Fig. 2. Profiles through the cingulate and rostral cortex depicting the percentage of neurons which are either inhibited (solid line), excited (dotted line) or unaffected (dashed line) by microiontophoretically administered 5-HT. 5-HT was applied during 60 sec with an ejection current of 80 nA.
66 20nA
A
40nA
80nA
B
20-
20nA ~
40nA ~
80nA ~
d 10-
I---I
1 minute
I minute
20n.A C
20n~
40n.._A
40~nA 8On...A
80n._.A
D
i
2o-
2o-,
t
t lO-
,o-
Q. I'--I
I-'--I
I minute
1 minute
Fig. 3. Typical response of rostral cortical neurons to m]croiontophoretically admimstered 5-HT. A: pure, dose-dependent depression of cell firing. B: pure, dose-dependent excitation. C: mixed response, following an ejection with currents of either 40 or 80 nA the period of cell depression is followed by a period of increased cell tiring. D: mixed response; with an ejection current of 80 n A a short pronounced excitatory effect is observed followed by an inhibition.
D R N stimulation. The highest percentage of inhibited cells was noted in the upper 500 /zm thick zone. In PCPA-treated, 5-HT-depleted animals, the percentage of inhibited neurons was strongly reduced in the entire cortex. Only one out of 100 neurons increased its firing rate in response to D R N st]mulation in untreated rats. No excitatory transsynaptic effects were seen in the 5-HT-depleted rats. Exact latencies of inhibition were not determined since the stimulation consisted of a train of pulses. In the rostral cortex some 30-60 ~o of all neurons investigated were inhibited by D R N stimulation. More neurons (50-60 ~o) were affected in the deep cell layers than in the upper (30-35 ~ ) layers. Only 4 out of 99 neurons tested were excited following such stimulation. Again, in the 5-HT depleted rats the number of inhibited neurons was significantly reduced in both the upper and lower cell layers. Two out of the 138 neurons investigated in this area were excited in the treated animals.
The action of microiontophoretically administered 5-HT Profiles through the cingulate and rostral cortex showing the percentages of
67
Control
¢n
15minutes after 3mg/kg i p of Cyproheptadlne
15minutes after l O m g / k g I.p of Cyproheptadine I period of stimulation
I
1 second
Fig. 4. The blocking action of systemicallyadministered cyproheptadine on transsynaptically induced inhibition of a cingulate cortical neuron evoked by stimulation of the DRN is shown.
neurons that were either inhibited or excited, or no affected by microiontophoreticaUy administered 5-HT are presented in Fig. 2. Similarly to the results on transsynaptic effects, inhibition of cell firing was the main action evoked by 5-HT on neurons in both brain regions. Furthermore, on the average, a higher percentage of depressed neurons was seen in the cingulate (80-90 70) than in the rostral cortex (50-60 70). Four different types of response to 5-HT were observed. The most frequent was a pure dosedependent reduction of cell firing with a slow onset of action and a recovery period often lasting up to 5 min (Fig. 3A). A few cells showed a pure, dose-dependent excitatory response characterized by its long recovery time (Fig. 3B). In some rare instances, a mixed type of response was noted. In these cases, cell depression occurred at 'doses' of 20 and 40 nA. However, when ejected with a current of 80 nA an initial phase of cell inhibition was followed by a short period of cell excitation (Fig. 3C). It was difficult to judge whether this was a true dual effect, or whether the excitation was a rebound phenomenon (Fig. 3C). Finally, on a few cells at a dose of 80 nA, there was an initial, brief phase of excitation followed by inhibition (Fig. 3D). The two latter types of neuron were classified as 'inhibited' neurons. Microiontophoretically administered 5-HT depressed 73 70 of the neurons in the cingulate and some 47 % of all neurons in the rostral cortex that were not affected by D R N stimulation. Furthermore, the majority of the transsynaptically activated neurons were equally depressed by 5-HT.
Antagonism of transsynaptic DRN-elicited effects by 5-HT antagonists All 3 compounds tested blocked the transsynaptically elicited depression of cingulate cortical cell firing in a dose-dependent fashion (Table I). Whereas GP 50 302 already displayed effects at a dose of 0.1 mg/kg i.p., methysergide was only active at a dose of 3 mg/kg and crypoheptadine at a dose of 10 mg/kg only (Fig. 4). Recovery could not be observed, since the effect usually outlasted the period during which the
68 TABLE I
The results on the action of three putative 5-HT antagonists on transsynaptically ehcited depression of cingulate cortical neurons The drugs were administered in a cumulattve manner. A drug was considered to be active at a g~ven dose if it reduced the duration o f the DRN-ehclted firing depression by 50 % or more.
GP 50 302
Methysergide
0.03 01 1.0 O. 1 mg/kg mg/kg rng/kg rng/kg
Number of neurons with no effect 2
2
N u m b e r of neurons with transsynapt~c blockade
3
3
Cyproheptadme
l0 30 0 1 rng/kg rng/kg rng/kg
10 rng/kg
3.0 rng/kg
4
3
4
1
I
I
3
5
10 mg/kg
4
neurons could be recorded. With all three compounds the latency of the blocking effect ranged from 10 to 15 min.
The action of microiontophoretically administered 5-HT antagonists on 5-HT elicited neuronal depression All 3 putative 5-HT-antagonists blocked the depressant action of microiontophoretically administered 5-HT on the majority of neurons (Fig. 5). Methysergide and cyproheptadine, when ejected with currents of 40-90 nA, often slightly depressed the activity of the neurons. Methysergide antagonized the 5-HT effect on 5 out of 7 neurons. Cyproheptadine blocked the cell inhibiting effect of 5-HT on 4 out of 6 Cyproheptadme 80nA
8
Methysergide 80nA 5-HT/3OnA
GP 50302 80nA • 5-H..T/3OnA
~
,
1 minute
Fig. 5. The b i l k i n g effect of microiontophoretically administered 5-HT antagomsts on 5-HT ehcited depression o f cingulate cortical neurons is shown.
69 neurons and GP 50 302 was active on 7 out of 8 neurons tested. None of these drugs antagonized the depressant action of GABA when ejected with currents that blocked the depressant action of 5-HT (n = 5) (data not shown). DISCUSSION Stimulation of the DRN and the microiontophoretic administration of 5-HT both resulted in a temporary depression of the firing rate of a fairly high percentage of neurons in apical as well as in basal cell layers. These findings confirm previous studies showing a predominance of inhibitory over excitatory effects elicited by the microiontophoretic administration of 5-HT 4,17,18,2a. In about 360 neurons in both the cingulate and the rostral cortex of rats treated acutely or chronically with various antidepressants, we had obtained similar results23: depression of cell firing was the most common effect and as in the present study, more excitatory responses were noted in the rostral than in the cingulate cortex. It seems unlikely that these excitatory responses represent artefacts since they were very pronounced and mostly outlasted the period of drug application. With the exception of one neuron which was excited transsynaptically as well as by the microiontophoretically administered 5-HT, all other neurons excited by microiontophoretically administered 5-HT were either inhibited or unaffected by DRN stimulation. Thus one may tentatively assume that the excitatory effects obtained upon the microiontophoretic administration of 5-HT may be mediated indirectly through inhibition of an inhibitory interneuron or through presynaptic mechanisms leading to a reduction of an inhibitory input. The dual effects on cell firing seen occasionally when 5-HT was applied with high currents has been reported previously17. They may indicate the interaction of direct and indirect effects. The results of the present study are in keeping with those of a previous one 2s showing that stimulation of the median raphe nucleus causes a reduction in the firing rate of the majority of cerebral cortical neurons tested. The depressant effect of microiontophoretically administered 5-HT and the transsynaptically elicited 5-HT effects were blocked by the 5-HT antagonist metergoline. The present findings lend support to the hypothesis advanced on the basis of anatomical data, that the serotonergic fibers can affect many cells in different cortical cell layerslL The physiological significance of this ubiquitious depressant effect is difficult to evaluate. It may well be that the serotonergic fibers have other additional effects on metabolism or at the information-processing level that are undetectable with the present method. In particular, the interpretation of the results is hampered by the paucity of information on the exact site of 5-HT innervation and its relationship with other afferent systems. The cortical serotonergic innervation originates from neurons of both the dorsal and median raphe nucleil,lZ, 21. To our knowledge, it is not known whether any of the two brain areas investigated is preferentially innervated by one of the two raphe nuclei. Both the noradrenergic fibers deriving from the locus coeruleus and the serotonergic fibers innervate the entire neocortex and, apart from marginal regional variations, it appears that both systems innervate all cortical cell layers. In previous
70 studies we showed that the noradrenerglc system exerts an inhibitory influence on cells throughout all layers of both the cmgulate and the rostral cortexeL Although the number of neurons sampled in these stu&es may have been too low to reflect true innervation density, it is interesting to compare the profiles obtained on the serotonergic and noradrenergic fiber systems. In the rostral cortex we found that both the noradrenergic and the serotonergic fiber system affect more neurons )n deeper than in the upper cell layers. In the cingulate cortex we observed that the actwation of the noradrenergic fibers affects a similar percentage of neurons in different layers 9, whereas the serotonergic fibers appear to affect more neurons m the apical than m the basal cortical layers. The physiological relevance of these findings remains to be elucidated. The three 5-HT antagonists methysergide, crypoheptadine and GP 50 302 transsynaptically blocked the effects on cingulate cortical neurons produced by D R N stimulation in a dose-dependent fashion. GP 50 302 ~s a potent 5-HT antagomst blocking centrally as well as peripherally mediated 5-HT effects (A. Stula, personal communication). On the basis of the present results, it appears that these transsynapt~c effects are mediated by 5-HT-contalning axons. Therefore, the present study shows that these compounds act as 5-HT antagonists in this brain area. This conclusion is coroborated by the microlontophoretic experiments m which we showed that these 5HT antagomsts block the action of 5-HT in the cingulate cortex. Whether or not regional &fferences may account for some of the discrepancies reported in literature remains an open question. It is conceivable that the drugs act at lower doses than those found to be active in the present study, since they were administered in a cumulative manner at relatively short intervals. These intervals may have been too short to observe the maximum effect at a given dose. Methyserglde was found to be about 10 t~mes more potent m v~vo than cyproheptadine. Interestingly, m an m v~tro binding study 2 it was found that methysergide was about 10 times more potent than crypoheptadme in &splacing [aH]5-HT from rat cerebral cortical membranes. ACKNOWLEDGEMENTS I would like to thank Drs. P. C. Waldmeier and B. Fehr for the biochemxcal analysis. I would like to express my gratitude to Dr. V. Baltzer for helpful &scussions and critical reading of the manuscript.
REFERENCES 1 Azmitia, E. C. and Segal, M., An autoradiographlc analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat, J. comp. NeuroL, 179 (1978) 641-668. 2 Bennett, J. P. and Snyder, S. H., Serotonin and lysergic acid ðylamide binding in rat brain membranes" relationship to postsynaptic serotonin receptors, Molec. PharmacoL, 12 (1976) 373-389. 3 Bradshaw, C. M., Roberts, M. H. T. and Szabadi, E., Effect of tncychc antidepressants on monoamine responses of single cortical neurons, Brit. J. Pharmacol., 41 (1971) 394-395. 4 Crawford, J. M. and Curtis, D. R., Pharmacologicalstudies of feline betz cells, J. Physiol. (Lond), 186 (1966) 121-138.
71 5 Curzon, G. and Green, A. R., Rapid method for the determination of 5-hydroxytryptamine and 5-hydroxyindolacetic acid in small regions of the brain, Brit. J. Pharmacol., 39 (1970) 653-655. 6 Davies, J. and Tongroach, P., Neuropharmacological studies on the nigro-striatal and raphestriatal system in the rat, Eurp. J. Pharmacol., 51 (1978) 91-100. 7 De Groot, J., The Rat Forebrain in Stereotaxic Coordinates, North-Holland, Amsterdam, 1972. 8 Descarries, L., Beaudet, A. and Watkins, K. C., Serotonin nerve terminals in adult rat neocortex, Brain Research, 100 (1975) 563-588. 9 Dillier, N., Laszlo, J., Mi~ller, B., Koclla, W. P. and Olpe, H.-R., Activation of an inhibitory noradrenergic pathway projecting from the locus cocruleus to the cingulate cortex of the rat, Brain Research, 154 (1978) 61-68. 10 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) 36-85. 11 Fuxe, K., Hokfeit, T. and Ungerstedt, U., Localization of lndoicalkylamines in CNS. In S. Garattini and P. A. Shore (Eds.), Advances in Pharmacology, Academic Press, New York, 1968, pp. 235-251. 12 Haigler, H. J. and Aghajanian, G. K., Serotonin receptors in the brain, Fed. Proc., 36 (1977) 2159-64. 13 Jacobs, B. L., Foote, S. L. and Bloom, F. E., Differential projections of neurons within the dorsal raphe nucleus of the rat: a horseradish pcroxidase (HRP) study, Brain Research, 147 (1978) 149-153. 14 Johnson, E. S., Roberts, M. H. T. and Straughan, D. W., The responses of cortical neurons to monoamines under differing anaesthetic conditions, J. Physiol. (Lond.), 203 (1969) 261-280. 15 Jordan, L. M., Frederickson, R. C. A., Philhs, J. W. and Lake, N., Microelectophoresis of 5-hydroxytryptamine: a clarification of its action on cerebral cortical neurones, Brain Research, 40 (1972) 552-558. 16 Kreig, W. J. S., Connections of the cerebral cortex. I. The albino rat. J. comp. Neurol., 84 (1946) 221-275. 17 Krnjevic, R. and Phillis J. W., Actions of certain amines on cerebral cortical neurones, Brit. J. PharmacoL, 10 (1963) 471--490. 18 Legge, K. F., Randic, M. and Straughan, D. W., The pharmacology of neurones in the pyriform cortex, Brit. J. PharmacoL, 26 (1966) 87-107. 19 Lidov, H. G. W., Grzanna, R. and Molliver, M. E , The serotonin innervation of the cerebral cortex in the rat - - an lmmunohistochemical analysis, Neuroscience, 5 (1980) 207-227. 20 McCall R. B. and Aghajanian, G. K., Serotonergic facilitation of facial motoneuron excitation, Brain Research, 169 (1979) 11-27. 21 Moore, R. Y., Halaris, A. E. and Jones, B. E., Serotonin neurons of the midbrain raphe: ascending projections, J. comp. Neurol., 180 (1978) 417-438. 22 Olpe, H.-R., Glatt, A., Laszlo, J. and Schellenberg, A., Some electrophyslological and pharmacological properties of the corUcal, noradrenerg~c projection of the locus coeruleus in the rat, Brain Research, 186 (1980) 9-19. 23 Olpe, H.-R. and Schellenberg, A., The sensitivity of cortical neurons to serotonln: effect of chronic treatment with antidepressants, serotonin uptake and monoamine oxldase blocking drugs, J. neural Transm., in press. 24 Phdhps, J. W. and York, D. H., Strychnine block of neural and drug-induced inhibition m the cerebral cortex, Nature (Lond.), 216 (1967) 922-923. 25 Roberts, M. H. T. and Straughan, D. W., Excitation and depression of cortical neurones by 5hydroxytryptamine, J. Physiol. (Lond.), 193 (1967) 269-294. 26 Roberts, M. H. T. and Straughan, D. W., An excitant effect of 5-hydroxytryptamine on single cortical neurones, J. Physiol. (Lond.), 188 (1967). 27 Saavedra, J. M., Brownstein, M. H. and Palkovits, M., Serotonin dlstributton in the hmblc system of the rat, Bram Research, 79 (1974) 437-441. 28 Sastry, B. S. R. and Phillis, J. W , Inhibition of cerebral cortical neurones by a 5-hydroxytryptammergic pathway from median raphe nucleus, Canad. J. Physiol Pharmacol., 55 (1977) 737-743. 29 Segal, M., 5-HT antagomsts in rat hippocampus, Brain Research, 103 (1976) 161-166. 30 Wang, R. Y. and Aghajanian, G. K., Inhibition of neurons in the amygdala by dorsal raphe stimulation: mediation through a direct serotonergic pathway, Brain Research, 120 (1977) 85-102.