- Email: [email protected]
Antidromic activation of rat dorsomedial hypothalamic neurons from locus coeruleus and median eminence
Brain Research Bulletin, Vol. 18, pp. 291-295.a
0361~9230/$7 $3.00 + .OO
Pergamon Journals Ltd., 1987. Printed in the U.S.A.
Antidromic Activation of Rat Dorsomedial Hypothalamic Neurons From Locus Coeruleus and Median Eminence TETSUYA
SHIROKAWA
AND SHOJI NAKAMURA’
Department of Neurophysiology, Institute of Higher Nervous Activity Osaka University Medical School, Kita-Ku, Osaka 530, Japan Received
9 April 1985
SHIROKAWA, T. AND S. NAKAMURA. Antidromic activation of rat dorsomedial hypothalamic neurons from licus coeruleus and median eminence. BRAIN RES BULL 18(3) 291-295, 1987.~In rats anesthetized with urethane, single unit activity was recorded in the hypothalamic dorsomedial nucleus (DMH) to obtain antidromic response to stimulation of locus coeruleus (LC) and median eminence (ME). Ninety-two cells were activated antidromically from LC and/or ME. Antidromic latencies to LC stimulation ranged from 7 to 39 msec and those to ME stimulation ranged from 5 to 20 msec. Approximately 13% of the neurons recorded revealed antidromic responses simultaneously from LC and ME, and they were found to bifurcate near the soma. The majority of DMH neurons projecting to LC alone were not spontaneously active, while those projecting to ME tended to discharge spontaneously. Dorsomedial hypothalamic
nucleus
Locus coeruleus
Median eminence
THE dorsomedial nucleus of the hypothalamus (DMH) in the rat is one of the major nuclei in the medial hypothalamus. However, little attention has so far been paid to this nucleus, though its size is comparable to that of the ventromedial hypothalamic nucleus which has most vigorously been studied in the medial hypothalamic region. Recent studies using the technique of retrograde transport of horseradish peroxidase (HRP) have indicated that the brain stem and the spinal cord receive afferent fibers from the DMH [3, 9, 191. In particular, it is interesting to note that DMH neurons send their axons to the locus coeruleus (LC) [3] which is composed of noradrenergic neurons whose axons diffusely project to many brain sites [ 1,4]. In addition, a direct connection from the hypothalamus to the LC was suggested from an electron microscopic study showing that lesioning the hypothalamus resulted in the appearance of degenerative synaptic terminals in the LC [ 111. Based on these findings, the present experiments were attempted to obtain electrophysiological evidence for the projection from the DMH to the LC. Furthermore, since the DMH is located in part of the hypothalamic hypophysiotrophic zone, stimulation of the median eminence (ME) was also tested to confirm the presence of DMH neurons projecting to the ME.
181. Body temperature was maintained at 37+ 1°C by a heating pad. Additional anesthetic was injected to qaintain the level of anesthesia during experiments. A tractial tube was inserted, and some animals were respired artiaicially after injection of Flaxidil (50 mgikg, IP). Stimulation
Bipolar stimulating electrodes were implanted into the LC and the ME. For stimulation of the LC, stimulating electrodes were placed within the LC by the method previously described [6, 12, 131. Briefly, the location of the LC was determined by the appearance of field responses evoked by stimulation of the dorsal noradrenergic bundle.(A: 2.0, L: 0.8, depth from the cortical surface: 5.7-6.0). Since the recording site for the DMH was so close to the ME, the stimulating electrodes for the ME were inserted from the site contralateral to the recording side at a lateral angle of 5. The coordinates for the ME were 3.4-3.6 mm anterior to the lambda and 1.0 mm lateral to the midline. The tip of the electrode was placed 0.2-0.5 mm above the brain base. Stimuli applied to the LC and the ME were square pulses of 0.7 msec duration with intensity of 0.1-3.0 mA,-The site of ME stimulation was verified histologically at the end of the experiments. When the stimulating electrode for the ME was found to be misplaced from the ME, only data for LC stimulation were taken for later analysis. In four animals, stimulating electrodes for ME stimulation were not correctly placed within the ME. In these cases, no cells were activated antidromicaliy from these stimulation sites, though the tip of the electrodes was found to be posi-
METHOD
Male and female Sprague-Dawley rats weighing 200-300 g were anesthetized with urethane (1.3 g/kg, IP) and mounted in a stereotaxic apparatus by the Kiinig and Klippel method ‘Present address: Department of Physiology,
Kanazawa University
Antidromic activation
Medical School, Takaramachi
291
13-1, Kanazawa 920, Japan.
FIG. 1. (A) Examples of antidromic responses of DMH neurons to stimulation of the LC and the ME. The cells a and c were activated antidromic~Iy from the LC and the cell b was activated from the ME. In each trace, 5 sweeps were superimposed. Arrows indicate the onset of the stimuli. For the ceH a that was spontaneously active, a peristimulus time histogram was constructed by summing spike discharges during 100 sweeps of the stimuli applied to the LC. (B) The recording sites of the cells a, band c. DMH: dorsomedial hypothalamic nucleus, F: fornix, FMT: fasciculus mamillothalamicus. (C) A photomicrograph showing the recording site c marked by dye deposit (white arrow).
tioned only less than 1 mm apart from the ME. Similarly, the proportion of the cells activated antidromica~Iy from the LC appeared to decrease, when stimulating electrodes for LC stimulation were misplaced from the LC. These observations indicated that, in the present experiment, current spread was restricted to a small area around the stimulating electrodes. Recordings
Extracellular recordings were made by means of a glass micropipette filled with 2 M K acetate containing 45%pontamine sky blue or 3 M KCI. Antidromic responses to LC stimulation were searched for around the site 3.5-3.7 mm anterior to lambda, 1.O mm lateral to the midline and 6.6-8.5 mm ventral to the cortical surface. To miss the sinus, recording electrodes were inserted into the brain at a lateral angle
of 2. Since the majority of cells antidromicalfy activated from the LC were not spontaneously active, we slowly advanced recording electrodes while stimulating the LC at a frequency of 1 Hz. Standard criteria for antidromic activation were as follows: (I) constant response latency, (2) ability to follow stimulation at high frequencies (~200 Hz), and (3) collision with spontaneously occurring spikes when cells were spontaneously active. For statistical test, Student’s E-test or chi-square test was used in the Results section.
RESULTS
Ninety-two cells in the region of the DMH revealed antidromic responses to stimulation of the LC and the ME:
293
DMH-LC PROJECTION / q
Antidromic Latency
;
E
j,’
;c
145’ms ,,I’
. 0.5 ms ,
/,/
”
(1’
. ME Stimulation
i3.5 ms f
c1-)ME
0
10
i0
io
ti msec
Conduction Velocity
{aD 7
in
= 16 8
ME-projecting N=35
mfsec FIG. 2. (A and B) Distribution histograms of antidromic latencies to stimulation of the LC (A) and the ME (B). (C and D) Distribution histograms for conduction velocities of LC- (C) and ME- (D) projecting axons of DMH neurons. Since some of the DMH neurons showing antidromic response to stimulation of the LC were activated antidromically from stimulation of the midbrain sites, the conduction velocities of the LC-projecting axons were calculated by assuming the distance between the LC and the midbrain site to be 4.5 mm. Assuming that the distance from the DMH to the ME was 2 mm, the conduction velocites of DMH neurons projecting to the ME were obtained.
56 (60.9%) from the LC alone, 24 (26.1%) from the ME alone and 12 (13.0%) from both the LC and the ME. Antidromic
Responses
to LC and ME Stimulation
Typical recordings of antidromic responses to stimulation of the LC and the ME are shown in Fig. 1. In this experiment a recording electrode filled with the dye was used to locate the recording site. Using this electrode, antidromic firings evoked by stimulation of either the LC or ME were obtained from 3 cells at sites a, b and c (B). At the end of the experiment, the dye was deposited at site c indicated by the white arrow in (C). The cell a, activated antidromically from the LC (latency, 24 msec) (a in A) was spontaneously active (2 c/set), whereas the cell c showing antidromic activation from the LC (latency, 15 msec) was silent (c in A). The cell b,
FIG. 3. A typical example of a branched neuron projecting to the LC and the ME. In this neuron, antidromic responses to stimulation of the LC (A) and the ME (B) had latencies of I5 and 14 msec, respectively. When LC stimuli were given 31 msec following ME stimuli, antidromic responses to LC stimulation mostly occuned (C). Decreasing the interstimulus interval to 30 msec caused the antidromic response to LC stimulation to disappear(D). The refractory time for this neuron, determined by giving two consecutive stimuli applied to the LC, was 2 msec. From these values, the conduction time between the soma and the branch point was calculated to be 0.5 msec (E).
activated antidromically from the ME (latency, 19 msec) responded mostly with an A-spike and rarely with a full-sized spike (b in A). For the cell a, a peristimulus time histogram (PSTH) compiled by 100 stimuli applied to the LC indicates the occurrence of inhibition lasting approximately 80 msec after antidromic response (a in A). Of the 11 LC-projecting neurons subjected to compilation of PSTHs for LC stimulation, all showed inhibition following antidromic response. The duration of the inhibition ranged from 30 to 220 msec (mean?S.E. = 1042 19.4 msec). Similarly, inhibition was seen after antidromic firings induced by ME stimulation. The mean duration of the inhibition caused by ME stimulation (meanrS.E.=80.0-+20.0 msec, N=3) was not significantly different from that of the LC-induced inhibition. Histological verification done at the end of the experiment indicated that the three cells recorded in this experiment were located within the DMH. In the present experiments, the dye deposition was made for 8 cells projecting to the LC and/or ME. All the marked cells showing the same electrophysiological properties as those of the remaining cells were identified as DMH neurons. Therefore, the units recorded in the present experiments were considered to originate in the DMH, though the possibility could not be excluded that some units were not located in the DMH. Furthermore, all DMH neurons thus identified were found in the dorsal part of the nucleus except one cell located in the ventromedial part. Antidromic
Latency
and Conduction
Velocity
Frequency distributions of antidromic latencies to LC and ME stimulation are shown in Fig. 2A and B. The antidromic latencies of 82 units to LC stimulation ranged from 7 to 39 msec with a mean of 17.8 msec (S.E.=-c0.7 msec) (A), and those of 35 units to ME stimulation ranged from 4.8 to 20 msec (mean?S.E.= 10.720.7 msec) (B). Of those activated antidromically from the LC, 25 DMH
SHIROKAWA
294
n N=49
N=l2
silent cell spontaneously active cell N=24
LC FIG. 4. Percentages of spontaneously active and inactive (silent) cells in the DMH. It is noted that though most LC-projecting cells (left) were silent, the proportion of spontaneously active cells was increased in branched (LC + ME) (middle) and ME-projecting (right) cells.
neurons revealed antidromic firings simultaneously from stimulation of the midbrain site (same coordinates as those for the dorsal noradrenergic bundle). This indicates that at least part of the axons of DMH neurons projecting to the LC may descend near the dorsal noradrenergic bundle. With such neurons projecting to the LC via the midbrain site, conduction velocities of DMH axons projecting to the LC were determined by assuming the distance between the LC and the midbrain site to be 4.5 mm. The conduction velocities so calculated had a mean of 0.51 m/set (S.E.=?0.07 m/set), ranging from 0.16 to 1.5 m/set (C). Similarly, assuming the distance from the DMH to the ME to be 2.0 mm, the conduction velocities of DMH axons projecting to the ME were calculated. They ranged from 0.1 to 0.42 m/set with a mean of 0.22 m/set (S.E.=?0.02 m/set) (D), which was significantly smaller than that of DMH axons projecting to the LC, t(26)=4.438, p
Cell5
As mentioned above, the electrophysiological experiments indicate that about 13% of DMH neurons recorded send their axons simultaneously to the LC and the ME. We will call these neurons “branched cells” hereafter. A typical example of a branched cell showing antidromic responses from both the LC and the ME is shown in Fig. 3. The antidromic latencies of this cell to LC and ME stimulation were 15 (A) and 14 msec (B), respectively. The refractory time, determined by applying two consecutive stimuli to the LC was 2 msec. To determine the time for collision between the two antidromic responses, LC stimulation was given at a certain interval following ME stimulation (C and D). At an interstimulus interval of 3 1 msec, ME-induced response frequently failed to occur (C) and reducing the interval to 30 msec resulted in a complete blocking of the second response (D). From the values obtained, we calculated that the conduction time from the soma to the branch point was 0.5 msec (E). This indicates that the axon of this DMH neuron bifurcated near the soma. The conduction time from the soma to the branch point determined for 11 branched cells ranged from 0 to 6 msec, with a mean of 2.0 msec.
ANI) ~
DMH neurons were classified into three types according to their antidromic responses to stimulation of the Lc’ and the ME: (I) neurons activated from the LX’ alone (I C’ projecting cells), (2) those activated from both the LC and ME (branched cells), and (3) those activated from the ME alone (ME-projecting cells). Concerning spontaneous di\charges of these cells, we found a distinct relation between the spontaneous discharges and the site of projection. Seventy-eight percent (38149) of the LC-projecting cells had no spontaneous discharges, whereas 6m (7:12) of’ the branched cells and 75% (1X/24) of the ME-projecting one\ were spontaneously active (Fig. 4). The incidence of the \ilent cells was significantly higher in the LC-projecting cells than the branched (x2=4.367, p
cording single unit activity from the DMH, and noticed that changes of the firing rate of DMH neurons were not associated with EEG patterns.
Using the HRP technique, Cedarbaum and Aghajanian have shown that the rat LC receives afferent fibers from several hypothalamic nuclei such as the DMH, paraventricular and lateral hypothalamic areas [?I. Consistent with their finding, the present experiment indicated that electrical stimulation of the LC resulted in antidromic activation of DMH neurons. In addition. we found that some DMH neurons projected simultaneously to the LC and the ME with the branch point close to the soma. These results are compatible with the finding of Anschel et al. that there were DMH neurons revealing antidromic activation simultaneously from the midbrain site and the ME 121. These findings further support the evidence that some neurons of hypothalamic areas such as the VMH, anterior hypothalamic areas, paraventricular and arcuate nucleus, project simultaneous to the ME and intra- and extrahypothalamic areas [2, 16-181. Most LC-projecting axons had conduction velocities under 1.O m/set and the axons branching to the ME revealed slower conduction velocities than the LC-projecting axons. The values of conduction velocities of DMH axons are analogous to those reported previously 121. We found that the majority of neurons projecting exclusively to the ME were spontaneously active; approximately 70% of the LC-projecting neurons were silent cells, whereas about the same proportion of ME-projecting neurons showed spontaneous activity. Such a distinct difference in spontaneous discharges between LC- and ME-projecting cells suggests the functional heterogeneity of DMH neurons to the projecting sites. Previous studies using male rhesus monkeys have indicated the participation of the DMH in male sexual behavior. Perachio et ~1. found that electrical stimulation of the DMH produced sexual behavioral responses in freely moving male monkeys [ 1.51.This is consistent with the findings of Oomura et al. that DMH neurons increase the discharges during sexual behavior in partially restrained male monkeys [ 141. On the other hand, Gallo has shown that in ovariectomized rats, electrical stimulation of the DMH suppressed the release of pulsatile luteinizing hormone [5]. These findings suggest that the DMH may be involved in sexual behavior in a manner so as
DMH-LC PROJECTION
295
to activate actual performance of sexual behavior and/or to alter the secretion of hormones relevant to sexual behavior. In this regard, the projection of DMH neurons to the ME may be related to the hormone secretion, while the projection to the LC may play a role in modulating neuronal activity relating to sexual behavior. In particular, since the LC is reported to be associated with emotion, it is possible that the DMH-LC projection may be responsible for changes in emotions occurring in sexual behavior [7,10]. Of interest is the histological finding that DMH neurons project to both the LC and the preganglionic nuclei of the
sympathetic nervous system [19]. Sympathetic neurons innervate various peripheral organs, and noradrenergic neurons arising in the LC also send their axons to many brain sites. Therefore, whatever the function of the DMH, it may be capable of affecting the activites of various peripheral organs and brain cells simultaneously through these peripheral and central noradrenergic neurons. ACKNOWLEDGEMENT We thank Dr. H. Uchiyama for his help in the histological study.
REFERENCES 1. Amaral, D. G. and H. M. Sinnamon. The locus coeruleus: neurobiology of a central noradrenergic nucleus. Prog Neurobiol9: 147-l%, 1977. 2. Anschel, S., M. Alexander and A. A. Perachio. Multiple connections of medial hypothalamic neurons in the rat. Exp Brain Res 46: 383-392, 1982. 3. Cedarbaum, J. M. and G. K. Aghajanian. Afferent projections to the rat locus coeruleus as determined by a retrograde tracing technique. J Comp Neurol 178: 1-16, 1978. 4. Foote, S. L., F. E. Bloom and G. Aston-Jones. Nucleus locus coeruleus: New evidence of anatomical and physiological specificity. Physiol Rev 63: 84s914, 1983. 5. Gallo, R. V. Effect of electrical stimulation of the dorsomedial hypothalamic nucleus on pulsatile LH release in ovariectomized rats. Neuroendocrinology 32: 134-138, 1981. 6. Kayama, Y., T. Negi, M. Sugitani and K. Iwama. Effects of locus coeruleus stimulation on neuronal activities of dorsal latera1 geniculate nucleus and perigeniculate reticular nucleus of the rat. Neuroscience 7: 655-666, 1982. 7. Kety, S. S. The biogenic amines in the central nervous system: their possible roles in arousal, emotion and learning. In: The Neurosciences: Second Study Program, edited by F. 0. Schmitt. New York: Rockefeller University Press, 1970, pp. 324-336. 8. Konig, J. F. and R. A. Klippel. The Rat Brain, A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem. Baltimore: Williams and Wilkins, 1%3. 9. Kuypers, H. G. J. M. and V. A. Maisky. Retrograde axonal transport of horseradish peroxidase from spinal cord to brain stem cell groups in the cat. Neurosci Lett 1: 9-14, 1975. 10. McNaughton, N. and S. T. Mason. The neuropsychology and neuropharmacology of the dorsal ascending noradrenergic bundle-a review. Prog Neurobiol 14: 157-219, 1980.
11. Mizuno, N. and Y. Nakamura. Direct hypothalamic projections to the locus coeruieus. Brain Res 19: 168-163, 1970. 12. Nakamura, S. Some electrophysiological properties of neurons of rat locus coeruleus. J Phvsiol267: 641-658. 1977. 13. Nakamura, S. and K. Iwa&a. Antidromic activation of the rat locus coeruleus neurons from hippocampus, cerebral and cerebellar cortices. Brain Res 99: 372376, 1975. 14. Oomura. Y.. H. Yoshimatsu and S. Aou. Medial nreontic and hypothalamic neuronal activity during sexual behavior of the male monkey. Brain Res 266: 36343, 1983. 15. Perachio, A. A., L. D. Marr and M. Alexander. Sexual behavior in male rhesus monkey elicited by electrical stimulation of preoptic and hypothalamic area. Brain Res 177: 127-144, 1979. 16. Renaud, L. P. Neurosphysiology and neurophamtacology of medial hypothalamic neurons and their extrahypothalamic connections. In: Handbook of the Hypothalamus, Vol 1, edited by P. J. Morgane and J. Panksepo. New York: Marcel Dekker. __ Inc., 1979;~~. 593-693. 17. Renaud, L. P. Tuberoinf’undibular neurons in the basomedial hypothalamus of the rat: electrophysiological evidence for axon cohaterals to hypothalamic and extrahypothalamic areas. Bruin Res 105: 59-72, 1976. 18. Renaud, L. P. and J. B. Martin. Electrophysiological studies of connections of hypothalamic ventromedial nucleus neurons in the rat: evidence for a role in neuroendocrine regulation. Bruin Res 93: 145-151, 1975. 19. Saper, C. B., A. D. Lowey, L. W. Swanson and W. M. Cowan. Direct hypothalamo-autonomic connections. Brain Res 117: 305-312, 1976.
0361~9230/$7 $3.00 + .OO
Pergamon Journals Ltd., 1987. Printed in the U.S.A.
Antidromic Activation of Rat Dorsomedial Hypothalamic Neurons From Locus Coeruleus and Median Eminence TETSUYA
SHIROKAWA
AND SHOJI NAKAMURA’
Department of Neurophysiology, Institute of Higher Nervous Activity Osaka University Medical School, Kita-Ku, Osaka 530, Japan Received
9 April 1985
SHIROKAWA, T. AND S. NAKAMURA. Antidromic activation of rat dorsomedial hypothalamic neurons from licus coeruleus and median eminence. BRAIN RES BULL 18(3) 291-295, 1987.~In rats anesthetized with urethane, single unit activity was recorded in the hypothalamic dorsomedial nucleus (DMH) to obtain antidromic response to stimulation of locus coeruleus (LC) and median eminence (ME). Ninety-two cells were activated antidromically from LC and/or ME. Antidromic latencies to LC stimulation ranged from 7 to 39 msec and those to ME stimulation ranged from 5 to 20 msec. Approximately 13% of the neurons recorded revealed antidromic responses simultaneously from LC and ME, and they were found to bifurcate near the soma. The majority of DMH neurons projecting to LC alone were not spontaneously active, while those projecting to ME tended to discharge spontaneously. Dorsomedial hypothalamic
nucleus
Locus coeruleus
Median eminence
THE dorsomedial nucleus of the hypothalamus (DMH) in the rat is one of the major nuclei in the medial hypothalamus. However, little attention has so far been paid to this nucleus, though its size is comparable to that of the ventromedial hypothalamic nucleus which has most vigorously been studied in the medial hypothalamic region. Recent studies using the technique of retrograde transport of horseradish peroxidase (HRP) have indicated that the brain stem and the spinal cord receive afferent fibers from the DMH [3, 9, 191. In particular, it is interesting to note that DMH neurons send their axons to the locus coeruleus (LC) [3] which is composed of noradrenergic neurons whose axons diffusely project to many brain sites [ 1,4]. In addition, a direct connection from the hypothalamus to the LC was suggested from an electron microscopic study showing that lesioning the hypothalamus resulted in the appearance of degenerative synaptic terminals in the LC [ 111. Based on these findings, the present experiments were attempted to obtain electrophysiological evidence for the projection from the DMH to the LC. Furthermore, since the DMH is located in part of the hypothalamic hypophysiotrophic zone, stimulation of the median eminence (ME) was also tested to confirm the presence of DMH neurons projecting to the ME.
181. Body temperature was maintained at 37+ 1°C by a heating pad. Additional anesthetic was injected to qaintain the level of anesthesia during experiments. A tractial tube was inserted, and some animals were respired artiaicially after injection of Flaxidil (50 mgikg, IP). Stimulation
Bipolar stimulating electrodes were implanted into the LC and the ME. For stimulation of the LC, stimulating electrodes were placed within the LC by the method previously described [6, 12, 131. Briefly, the location of the LC was determined by the appearance of field responses evoked by stimulation of the dorsal noradrenergic bundle.(A: 2.0, L: 0.8, depth from the cortical surface: 5.7-6.0). Since the recording site for the DMH was so close to the ME, the stimulating electrodes for the ME were inserted from the site contralateral to the recording side at a lateral angle of 5. The coordinates for the ME were 3.4-3.6 mm anterior to the lambda and 1.0 mm lateral to the midline. The tip of the electrode was placed 0.2-0.5 mm above the brain base. Stimuli applied to the LC and the ME were square pulses of 0.7 msec duration with intensity of 0.1-3.0 mA,-The site of ME stimulation was verified histologically at the end of the experiments. When the stimulating electrode for the ME was found to be misplaced from the ME, only data for LC stimulation were taken for later analysis. In four animals, stimulating electrodes for ME stimulation were not correctly placed within the ME. In these cases, no cells were activated antidromicaliy from these stimulation sites, though the tip of the electrodes was found to be posi-
METHOD
Male and female Sprague-Dawley rats weighing 200-300 g were anesthetized with urethane (1.3 g/kg, IP) and mounted in a stereotaxic apparatus by the Kiinig and Klippel method ‘Present address: Department of Physiology,
Kanazawa University
Antidromic activation
Medical School, Takaramachi
291
13-1, Kanazawa 920, Japan.
FIG. 1. (A) Examples of antidromic responses of DMH neurons to stimulation of the LC and the ME. The cells a and c were activated antidromic~Iy from the LC and the cell b was activated from the ME. In each trace, 5 sweeps were superimposed. Arrows indicate the onset of the stimuli. For the ceH a that was spontaneously active, a peristimulus time histogram was constructed by summing spike discharges during 100 sweeps of the stimuli applied to the LC. (B) The recording sites of the cells a, band c. DMH: dorsomedial hypothalamic nucleus, F: fornix, FMT: fasciculus mamillothalamicus. (C) A photomicrograph showing the recording site c marked by dye deposit (white arrow).
tioned only less than 1 mm apart from the ME. Similarly, the proportion of the cells activated antidromica~Iy from the LC appeared to decrease, when stimulating electrodes for LC stimulation were misplaced from the LC. These observations indicated that, in the present experiment, current spread was restricted to a small area around the stimulating electrodes. Recordings
Extracellular recordings were made by means of a glass micropipette filled with 2 M K acetate containing 45%pontamine sky blue or 3 M KCI. Antidromic responses to LC stimulation were searched for around the site 3.5-3.7 mm anterior to lambda, 1.O mm lateral to the midline and 6.6-8.5 mm ventral to the cortical surface. To miss the sinus, recording electrodes were inserted into the brain at a lateral angle
of 2. Since the majority of cells antidromicalfy activated from the LC were not spontaneously active, we slowly advanced recording electrodes while stimulating the LC at a frequency of 1 Hz. Standard criteria for antidromic activation were as follows: (I) constant response latency, (2) ability to follow stimulation at high frequencies (~200 Hz), and (3) collision with spontaneously occurring spikes when cells were spontaneously active. For statistical test, Student’s E-test or chi-square test was used in the Results section.
RESULTS
Ninety-two cells in the region of the DMH revealed antidromic responses to stimulation of the LC and the ME:
293
DMH-LC PROJECTION / q
Antidromic Latency
;
E
j,’
;c
145’ms ,,I’
. 0.5 ms ,
/,/
”
(1’
. ME Stimulation
i3.5 ms f
c1-)ME
0
10
i0
io
ti msec
Conduction Velocity
{aD 7
in
= 16 8
ME-projecting N=35
mfsec FIG. 2. (A and B) Distribution histograms of antidromic latencies to stimulation of the LC (A) and the ME (B). (C and D) Distribution histograms for conduction velocities of LC- (C) and ME- (D) projecting axons of DMH neurons. Since some of the DMH neurons showing antidromic response to stimulation of the LC were activated antidromically from stimulation of the midbrain sites, the conduction velocities of the LC-projecting axons were calculated by assuming the distance between the LC and the midbrain site to be 4.5 mm. Assuming that the distance from the DMH to the ME was 2 mm, the conduction velocites of DMH neurons projecting to the ME were obtained.
56 (60.9%) from the LC alone, 24 (26.1%) from the ME alone and 12 (13.0%) from both the LC and the ME. Antidromic
Responses
to LC and ME Stimulation
Typical recordings of antidromic responses to stimulation of the LC and the ME are shown in Fig. 1. In this experiment a recording electrode filled with the dye was used to locate the recording site. Using this electrode, antidromic firings evoked by stimulation of either the LC or ME were obtained from 3 cells at sites a, b and c (B). At the end of the experiment, the dye was deposited at site c indicated by the white arrow in (C). The cell a, activated antidromically from the LC (latency, 24 msec) (a in A) was spontaneously active (2 c/set), whereas the cell c showing antidromic activation from the LC (latency, 15 msec) was silent (c in A). The cell b,
FIG. 3. A typical example of a branched neuron projecting to the LC and the ME. In this neuron, antidromic responses to stimulation of the LC (A) and the ME (B) had latencies of I5 and 14 msec, respectively. When LC stimuli were given 31 msec following ME stimuli, antidromic responses to LC stimulation mostly occuned (C). Decreasing the interstimulus interval to 30 msec caused the antidromic response to LC stimulation to disappear(D). The refractory time for this neuron, determined by giving two consecutive stimuli applied to the LC, was 2 msec. From these values, the conduction time between the soma and the branch point was calculated to be 0.5 msec (E).
activated antidromically from the ME (latency, 19 msec) responded mostly with an A-spike and rarely with a full-sized spike (b in A). For the cell a, a peristimulus time histogram (PSTH) compiled by 100 stimuli applied to the LC indicates the occurrence of inhibition lasting approximately 80 msec after antidromic response (a in A). Of the 11 LC-projecting neurons subjected to compilation of PSTHs for LC stimulation, all showed inhibition following antidromic response. The duration of the inhibition ranged from 30 to 220 msec (mean?S.E. = 1042 19.4 msec). Similarly, inhibition was seen after antidromic firings induced by ME stimulation. The mean duration of the inhibition caused by ME stimulation (meanrS.E.=80.0-+20.0 msec, N=3) was not significantly different from that of the LC-induced inhibition. Histological verification done at the end of the experiment indicated that the three cells recorded in this experiment were located within the DMH. In the present experiments, the dye deposition was made for 8 cells projecting to the LC and/or ME. All the marked cells showing the same electrophysiological properties as those of the remaining cells were identified as DMH neurons. Therefore, the units recorded in the present experiments were considered to originate in the DMH, though the possibility could not be excluded that some units were not located in the DMH. Furthermore, all DMH neurons thus identified were found in the dorsal part of the nucleus except one cell located in the ventromedial part. Antidromic
Latency
and Conduction
Velocity
Frequency distributions of antidromic latencies to LC and ME stimulation are shown in Fig. 2A and B. The antidromic latencies of 82 units to LC stimulation ranged from 7 to 39 msec with a mean of 17.8 msec (S.E.=-c0.7 msec) (A), and those of 35 units to ME stimulation ranged from 4.8 to 20 msec (mean?S.E.= 10.720.7 msec) (B). Of those activated antidromically from the LC, 25 DMH
SHIROKAWA
294
n N=49
N=l2
silent cell spontaneously active cell N=24
LC FIG. 4. Percentages of spontaneously active and inactive (silent) cells in the DMH. It is noted that though most LC-projecting cells (left) were silent, the proportion of spontaneously active cells was increased in branched (LC + ME) (middle) and ME-projecting (right) cells.
neurons revealed antidromic firings simultaneously from stimulation of the midbrain site (same coordinates as those for the dorsal noradrenergic bundle). This indicates that at least part of the axons of DMH neurons projecting to the LC may descend near the dorsal noradrenergic bundle. With such neurons projecting to the LC via the midbrain site, conduction velocities of DMH axons projecting to the LC were determined by assuming the distance between the LC and the midbrain site to be 4.5 mm. The conduction velocities so calculated had a mean of 0.51 m/set (S.E.=?0.07 m/set), ranging from 0.16 to 1.5 m/set (C). Similarly, assuming the distance from the DMH to the ME to be 2.0 mm, the conduction velocities of DMH axons projecting to the ME were calculated. They ranged from 0.1 to 0.42 m/set with a mean of 0.22 m/set (S.E.=?0.02 m/set) (D), which was significantly smaller than that of DMH axons projecting to the LC, t(26)=4.438, p
Cell5
As mentioned above, the electrophysiological experiments indicate that about 13% of DMH neurons recorded send their axons simultaneously to the LC and the ME. We will call these neurons “branched cells” hereafter. A typical example of a branched cell showing antidromic responses from both the LC and the ME is shown in Fig. 3. The antidromic latencies of this cell to LC and ME stimulation were 15 (A) and 14 msec (B), respectively. The refractory time, determined by applying two consecutive stimuli to the LC was 2 msec. To determine the time for collision between the two antidromic responses, LC stimulation was given at a certain interval following ME stimulation (C and D). At an interstimulus interval of 3 1 msec, ME-induced response frequently failed to occur (C) and reducing the interval to 30 msec resulted in a complete blocking of the second response (D). From the values obtained, we calculated that the conduction time from the soma to the branch point was 0.5 msec (E). This indicates that the axon of this DMH neuron bifurcated near the soma. The conduction time from the soma to the branch point determined for 11 branched cells ranged from 0 to 6 msec, with a mean of 2.0 msec.
ANI) ~
DMH neurons were classified into three types according to their antidromic responses to stimulation of the Lc’ and the ME: (I) neurons activated from the LX’ alone (I C’ projecting cells), (2) those activated from both the LC and ME (branched cells), and (3) those activated from the ME alone (ME-projecting cells). Concerning spontaneous di\charges of these cells, we found a distinct relation between the spontaneous discharges and the site of projection. Seventy-eight percent (38149) of the LC-projecting cells had no spontaneous discharges, whereas 6m (7:12) of’ the branched cells and 75% (1X/24) of the ME-projecting one\ were spontaneously active (Fig. 4). The incidence of the \ilent cells was significantly higher in the LC-projecting cells than the branched (x2=4.367, p
cording single unit activity from the DMH, and noticed that changes of the firing rate of DMH neurons were not associated with EEG patterns.
Using the HRP technique, Cedarbaum and Aghajanian have shown that the rat LC receives afferent fibers from several hypothalamic nuclei such as the DMH, paraventricular and lateral hypothalamic areas [?I. Consistent with their finding, the present experiment indicated that electrical stimulation of the LC resulted in antidromic activation of DMH neurons. In addition. we found that some DMH neurons projected simultaneously to the LC and the ME with the branch point close to the soma. These results are compatible with the finding of Anschel et al. that there were DMH neurons revealing antidromic activation simultaneously from the midbrain site and the ME 121. These findings further support the evidence that some neurons of hypothalamic areas such as the VMH, anterior hypothalamic areas, paraventricular and arcuate nucleus, project simultaneous to the ME and intra- and extrahypothalamic areas [2, 16-181. Most LC-projecting axons had conduction velocities under 1.O m/set and the axons branching to the ME revealed slower conduction velocities than the LC-projecting axons. The values of conduction velocities of DMH axons are analogous to those reported previously 121. We found that the majority of neurons projecting exclusively to the ME were spontaneously active; approximately 70% of the LC-projecting neurons were silent cells, whereas about the same proportion of ME-projecting neurons showed spontaneous activity. Such a distinct difference in spontaneous discharges between LC- and ME-projecting cells suggests the functional heterogeneity of DMH neurons to the projecting sites. Previous studies using male rhesus monkeys have indicated the participation of the DMH in male sexual behavior. Perachio et ~1. found that electrical stimulation of the DMH produced sexual behavioral responses in freely moving male monkeys [ 1.51.This is consistent with the findings of Oomura et al. that DMH neurons increase the discharges during sexual behavior in partially restrained male monkeys [ 141. On the other hand, Gallo has shown that in ovariectomized rats, electrical stimulation of the DMH suppressed the release of pulsatile luteinizing hormone [5]. These findings suggest that the DMH may be involved in sexual behavior in a manner so as
DMH-LC PROJECTION
295
to activate actual performance of sexual behavior and/or to alter the secretion of hormones relevant to sexual behavior. In this regard, the projection of DMH neurons to the ME may be related to the hormone secretion, while the projection to the LC may play a role in modulating neuronal activity relating to sexual behavior. In particular, since the LC is reported to be associated with emotion, it is possible that the DMH-LC projection may be responsible for changes in emotions occurring in sexual behavior [7,10]. Of interest is the histological finding that DMH neurons project to both the LC and the preganglionic nuclei of the
sympathetic nervous system [19]. Sympathetic neurons innervate various peripheral organs, and noradrenergic neurons arising in the LC also send their axons to many brain sites. Therefore, whatever the function of the DMH, it may be capable of affecting the activites of various peripheral organs and brain cells simultaneously through these peripheral and central noradrenergic neurons. ACKNOWLEDGEMENT We thank Dr. H. Uchiyama for his help in the histological study.
REFERENCES 1. Amaral, D. G. and H. M. Sinnamon. The locus coeruleus: neurobiology of a central noradrenergic nucleus. Prog Neurobiol9: 147-l%, 1977. 2. Anschel, S., M. Alexander and A. A. Perachio. Multiple connections of medial hypothalamic neurons in the rat. Exp Brain Res 46: 383-392, 1982. 3. Cedarbaum, J. M. and G. K. Aghajanian. Afferent projections to the rat locus coeruleus as determined by a retrograde tracing technique. J Comp Neurol 178: 1-16, 1978. 4. Foote, S. L., F. E. Bloom and G. Aston-Jones. Nucleus locus coeruleus: New evidence of anatomical and physiological specificity. Physiol Rev 63: 84s914, 1983. 5. Gallo, R. V. Effect of electrical stimulation of the dorsomedial hypothalamic nucleus on pulsatile LH release in ovariectomized rats. Neuroendocrinology 32: 134-138, 1981. 6. Kayama, Y., T. Negi, M. Sugitani and K. Iwama. Effects of locus coeruleus stimulation on neuronal activities of dorsal latera1 geniculate nucleus and perigeniculate reticular nucleus of the rat. Neuroscience 7: 655-666, 1982. 7. Kety, S. S. The biogenic amines in the central nervous system: their possible roles in arousal, emotion and learning. In: The Neurosciences: Second Study Program, edited by F. 0. Schmitt. New York: Rockefeller University Press, 1970, pp. 324-336. 8. Konig, J. F. and R. A. Klippel. The Rat Brain, A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem. Baltimore: Williams and Wilkins, 1%3. 9. Kuypers, H. G. J. M. and V. A. Maisky. Retrograde axonal transport of horseradish peroxidase from spinal cord to brain stem cell groups in the cat. Neurosci Lett 1: 9-14, 1975. 10. McNaughton, N. and S. T. Mason. The neuropsychology and neuropharmacology of the dorsal ascending noradrenergic bundle-a review. Prog Neurobiol 14: 157-219, 1980.
11. Mizuno, N. and Y. Nakamura. Direct hypothalamic projections to the locus coeruieus. Brain Res 19: 168-163, 1970. 12. Nakamura, S. Some electrophysiological properties of neurons of rat locus coeruleus. J Phvsiol267: 641-658. 1977. 13. Nakamura, S. and K. Iwa&a. Antidromic activation of the rat locus coeruleus neurons from hippocampus, cerebral and cerebellar cortices. Brain Res 99: 372376, 1975. 14. Oomura. Y.. H. Yoshimatsu and S. Aou. Medial nreontic and hypothalamic neuronal activity during sexual behavior of the male monkey. Brain Res 266: 36343, 1983. 15. Perachio, A. A., L. D. Marr and M. Alexander. Sexual behavior in male rhesus monkey elicited by electrical stimulation of preoptic and hypothalamic area. Brain Res 177: 127-144, 1979. 16. Renaud, L. P. Neurosphysiology and neurophamtacology of medial hypothalamic neurons and their extrahypothalamic connections. In: Handbook of the Hypothalamus, Vol 1, edited by P. J. Morgane and J. Panksepo. New York: Marcel Dekker. __ Inc., 1979;~~. 593-693. 17. Renaud, L. P. Tuberoinf’undibular neurons in the basomedial hypothalamus of the rat: electrophysiological evidence for axon cohaterals to hypothalamic and extrahypothalamic areas. Bruin Res 105: 59-72, 1976. 18. Renaud, L. P. and J. B. Martin. Electrophysiological studies of connections of hypothalamic ventromedial nucleus neurons in the rat: evidence for a role in neuroendocrine regulation. Bruin Res 93: 145-151, 1975. 19. Saper, C. B., A. D. Lowey, L. W. Swanson and W. M. Cowan. Direct hypothalamo-autonomic connections. Brain Res 117: 305-312, 1976.