- Email: [email protected]
Anatomical and electrophysiological characterization of presumed dopamine-containing neurons within the supramammillary region of the rat
Brain Research Bulletin, Vol. 20, pp. 307-314. Q Pergamon Press
plc, 1988. Printed in the U.S.A.
0361-9230/88 $3.00 + .OO
Anatomical and Electrophysiological Characterization of Presumed Dopamine-Containing Neurons Within the Supramammillary Region of the Rat PAUL D. SHEPARD,*
GREGORY
A. MIHAILOFFt
AND DWIGHT
C. GERMAN*J:’
*Departments of Physiology, Well Biology and 3Psychiatt-y University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, Received
Dallas, TX 75235
27 August 1986
SHEPARD, P. D., G. A. MIHAILOFF AND D. C. GERMAN. Anatomical and electrophysiological characterization of presumed dopamine-containing neurons within the supramammillary region of the rat. BRAIN RES BULL 20(3) 307-3 14, 1988.-A combination of immunocytochemical, electrophysiological and pharmacological techniques were employed to study the properties of neurons within the supramammillary (SUM) complex of the rat. The SUM region contains a small, but dense, population of tyrosine hydroxylase immunoreactive neurons. Following injection of the orthograde neuroanatomical tracer, Phaseolus Vulgaris leucoagglutinin, into the SUM region, heavy termainal labeling was observed in the lateral septal nucleus, diagonal band of Broca and bed nucleus of the stria terminalis. The electrophysiological and pharmacological properties of antidromicalIy-activated SUM neurons revealed evidence of two neuronal populations. Both groups of neurons exhibited long duration action potentials (>2 msec) and slow conduction velocities (CO.5 m/set). However, cells in one group were characterized by slow and erratic firing rates and insensitivity to dopamine (DA) autoreceptor agonists. Cells in the other group typically exhibited no spontaneous activity but could be induced to discharge by iontophoretic application of glutamate. These latter cells were sensitive to DA autoreceptor stimulation. Of the two populations of mammilloseptal SUM neurons, the silent population exhibited several properties similar to those of midbrain DA neurons. Supramammillary nucleus Lateral septal nucleus Dopamine Electrophysiology Phaseolus Vulgaris leucoagglutinin Tyrosine hydroxylase immunohistochemistry
tyrosine hydroxylase (TH), the rate-limiting enzyme in the synthesis of DA, but not the noradrenergic synethetic dopamine-beta hydroxylase [32]. These THenzyme, positive SUM neurons provide a major input to the lateral septal nucleus [4,31]. In fact, more septal alferents arise from TH-containing cells in the SUM region than in the A10 region. Despite evidence suggesting that neurons within the SUM region provide an important source of dopaminergic afferents to the septal nuclei, there is no information concerning the electrophysiological or pharmacological characteristics of these cells. Of particular interest is whether SUM DA neurons possess rate-moduiating autoreceptors and are thereby subject to the same autoregulatory influences as other mesolimbic DA neurons [I, 10, 34, 351. In this report, we describe experiments in which neuroanatomical, electrophysiological and pharmacological techniques were used to characterize mammilloseptal neurons. Evidence is pro-
THE mesolimbic dopamine (DA) system has been the subject of an intense experimental effort because of its role in locomotion [ 14, 25, 261 as well as its possible involvement in the etiology of schizophrenia [6, 21, 281. Traditionally, the cell bodies of mesolimbic DA neurons have been thought to be confined within the boundaries of the ventral tegmental area of Tsai (VTA) (nucleus A10 of Dahlstrom and Fuxe [7]), where they give rise to fibers which innervate a variety of forebrain areas including the nucleus accumbens, olfactory tubercle, amygdala, anterior limbic cortex and septal regions including the nucleus of the diagonal band, bed nucleus of the stria terminalis and the lateral septal nucleus [2, 3,7, 13, 17-19, 221. Recent neuroanatomical studies have indentified a small collection of catecholamine-containing cells within the supramammillary (SUM) region of the hypothalamus, which may represent a rostral extension of the A10 cell group [31]. These cells appear to be dopaminergic since they contain
‘Requests for reprints should be addressed to Dwight C. German.
307
308
SHEPARD,
vided which suggests that a subpopulation of these cells constitutes a separate nucleus of the mesolimic DA system, distinct from the cells within the VTA. METHOD
Surgical Procedures
Male Sprague-Dawley rats, weighing 200-350 g were used as experimental subjects. Animals were anesthetized with chloral hydrate (400 mglkg, IP) and mounted in a Kopf stereotaxic frame with the incisor bar positioned 2.4 mm below the interaural line. Rats used in electrophysiological experiments were cannulated (femoral vein) for intravenous administration of drugs and supplementary anesthetic. All incision and soft tissue pressure points were infiltrated with a long-acting local anesthetic (mepivacaine HCI, 2%). Body temperature was maintained between 3637°C using an aquatic heating pad. Following a midline incision and reflection of underlying tissue, a burr hole was drilled in the skull overlying the region of the posterior hypothalamus (anterior, 2580-3180 pm; lateral, O-1000 pm, according to the atlas of Konig and Klippel [ 151) and the dura removed. In some experiments, an additional burr hole was drilled over the regions corresponding to the lateral septal nucleus (anterior, 7020-8920 pm, lateral 300-1000 pm) and/or anterior limbic cortex (anterior, 8620-11050 pm, lateral 800-2000 pm) for insertion of stimulating electrodes. Tyrosine Hydroxylase
lmmunocytochemistry
The distribution of putative DA-containing neurons within the SUM complex was characterized using an antise~m to bovine adrenal tyrosine hydroxylase. Four male rats were deeply anesthetized with pentobarbital (60 mg/kg, IP) and perfused transcardially with neutral buffered saline followed by 10% buffered formalin. Following decapitation and removal of the calvaria, the brain was placed in formalin and post-fixed for an additional 7 days after which it was blocked stereotaxically according to the atlas of Konig and Klippel(l51. Sections were cut at 40 pm using a freezing microtome and processed for immunocytochemical localization of TH according to a modification of the perioxidaseantiperoxidase (PAP) method of Sternberger [29] as described by Kozlowski and Nilaver [161. Briefly, free-floating sections were incubated for 14-16 hr at room temperature in a 1: 1000 dilution of rabbit anti-TH antiserum. Specificity of the antiserum used in these experiments is described elsewhere [29]. Following incubation with the primary antiserum, sections were incubated in a solution of sheep anti-rabbit IgG (1: IOO), washed and incubated in a 1:500 dilution of rabbit PAP complex. Sections were developed in a mixture of diaminobenzidine and Ha,. The reaction product was then intensified with nickel according to the method described by Hancock [ 121.
Cell areas were measured using a Leitz microscope which was interfaced, via a drawing tube and digitizing tablet, to a Data General computer. Using a 100x oil immersion objective (total magnification= 1125x), TH-positive somata were identified within the SUM, interfascisular nucleus (IFN) and VTA. A line was drawn around the perimeter of each neuron in which a nucleus was present. After all somata in a section were entered into the computer, mean cell areas were determined.
MIHAILOFF
AND GERMAN
Axonai Tracing
An orthograde tracing method, developed by Gerfen and Sawchenko [9], was used to characterize the distribution of septal afferents arising from neurons within the SUM complex. This method makes it possible to visualize the morphology of axons and their specializations within terminal fields. Moreover, a high degree of regional specificity can be achieved as the label appears not to be transported by fibers of passage. Four animals were prepared as outlined under “Surgical (1.5 mm Procedures. ’ ’ Single barrel glass micropi~ttes o.d.; A-M Systems), containing a small glass capillary, were pulled to a fine taper using a Kopf pipette puller. The tips were broken back to a diameter of 10-15 pm. Electrodes were filled with a 2.5% solution of Phaseolus Vulgaris leucoa~lutinin (PHA-L; Vector Labs, CA) in 0.05 M phosphate buffered saline (PBS) and affixed to a hydraulic microdrive. Electrodes were driven rapidly to a depth of 7.8 mm below the surface of the cortex and a small quantity of PHA-L was iontophoretically delivered using an intermittent (7 set ON; 7 set OFF) application of 5-7 &A of anodal current for 20-30 min. After survival times which ranged from 8-12 days, rats were deeply anesthetized and perfused transcardially with a solution of 4% paraformaldehyde in 0.1 M acetate buffer (pH 6.5) followed by a second fixative consisting of 4% paraformaldehyde and 0.05% glutaraldehyde in 0.1 M borate buffer (pH 9.5). The brain was removed from the cranium, and postfiied for an additional 16 hr in the last fixative. Sections were cut at 30 pm on a Vibratome and free-floating sections were processed for immunocytochemistry according to the avidin-biotin method (Vector Labs). Briefly, sections were incubated in goat anti-PHA-L antiserum for 24-48 hr at 4°C. Following incubation in the primary antiserum, sections were transferred to a solution of biotinylated rabbit anti-goat IgG (Vectastain kit, Vector Labs). The tissue was rinsed several times in saline and placed in the avidin-biotin-HRP complex (ABC) for 45 min. Following a second wash, the tissue was recycled through the IgG and ABC solutions and finally reacted with 0.05% diaminobenzidine and 0.04% H,Os in PBS. The reaction was stopped after 20 min. In some experiments, sections were counterstained with cresyl violet. Single Unit rewording
Animals were prepared as outlined in the section entitled “Surgical Procedures.” Epoxy-coated tungsten (Fredrick Haer) or S-barrel glass (R 62 D Optical) microelectrodes were used for recording extracellular action potentials. in vitro impedance was measured at 135 Hz and ranged from 2.ft4.0 Mfl. Electrodes were stereotaxically positioned over the region of the SUM complex and hydraulically driven to a depth of 6.0 mm below the surface of the cortex and the following 2.0 mm explored for spontaneously active cells. Electrode potentials were ~pl~ed and filtered using a high impedance amplifier (Grass P511, 0.3-3.0 KHz bandwidth) and monitored both visually (Tektronix 51 IS oscilloscope) and aurally (Grass AM8 audio amplifier). Single unit activity was recorded on magnetic tape. Data analysis was performed using a DEC LSI 11/03 mi~computer. Action potentials were disc~mina~d with an oscilioscope triggering circuit which provided a single square wave pulse coincident with each action potential. Rate histograms were compiled by summing action potentials across consecutive 10 set intervals.
CHARACTERIZATION
OF DOPAMINE
NEURONS
Microiontophoresis Standard microiontophoretic techniques were used as previously described [20]. Briefly, electrodes were preloaded with several strands of fine fiberlgas fdaments and pulled to a taper using a Narishige pipette puller. Electrode tips were broken back to a final diameter of 3-6 pm. The center barrel was filled with 2 M NaCl saturated with Fast Green dye and used to record extracellular neuronal signals. One of the four outer barrels was filled with 3 M NaCl and used for current balancing. Each of the remaining barrels was filled with one of the following drug solutions: dopamine-HCl (DA; 0.2 M, pH 4.0), gamma-aminobutyric acid (GABA; 0.05 M, pH 4.0) or l-sodium glutamate (GLU; 0.05 M, pH 8.0). DA and GABA were ejected using anodal current while glutamate was ejected using cathodal current. A retaining current of 10 nA was applied to all drug barrels to prevent leakage. Microiontophoretic retention, ejection and balancing currents were controlled automatically by a constant current apparatus (Medical Systems, Model BH-2). Electrical Stimulation Stainless steel concentric bipolar electrodes (Kopf SNEX-lOO), were used to antidromically activate SUM neurons projecting to the ipsilateral septum, prefrontal or cingulate cortex. Electrical stimulation was provided by a Grass S48 stimulator coupled to constant current and stimulus isolation units. Square wave, cathodal pulses, 500 psec in duration, ranging from 0.1 to 4.5 mA were used. Single units were considered to have been antidromically activated when the following criteria were met: (a) a fixed latency between the onset of the stimulus and the occurrence of the stimulus-induced spike; and (b) no stimulus-induced spike was observed when the stimulus was triggered at a latency equal to or less than the conduction time for the stimulus-induced spike. Localization
of Recording and Stimulating
Sites
At the conclusion of each experiment, the position of the recording electrode was marked by an electrolytic lesion (tungsten electrode; +12 /*A for 45 set) or iontophoretic ejection of Fast Green dye (glass electrode; -20 PA for 1 hr). Stimulation sites were also marked by passing 0.5 mA anodal current through the electrode for 15 sec. Following removal from the stereotaxic appartus the animal was deeply anesthetized with chloral hydrate and perfused transcardially with saline (60 ml) followed by 10% neutral buffered formalin containing 1 g of potassium ferrocyanide. Frozen sections were cut at 50 pm, mounted on subbed slides and stained with cresyl violet for the determination of the position of the recording and stimulating electrodes. RESULTS
Distribution and Morphology
of TH-Positive
309
OF THE RAT
Somata
A small collection of TH-positive neurons was observed within the SUM complex (Fig. IA). The majority of these cells were situated within the medial portion of the SUM complex which is bordered ventrolaterally by the mammillothalmic tract. This dense group of parvocellular THpositive neurons extends in a rostro-caudal direction from the level of the a.rcuate/median eminence to the caudal pole of the mammillary nuclei where they appear to be continuous with midline TH-positive cells in the IFN. At their rostralmost extension, the TH-positive cells are situated medially
within the posterior hypothalamus, ventral to the Al3 DA cell group and dorsal to the caudal-most portion of nucleus A12. At more caudal levels, a few labeled cells were also observed throughout the lateral portion of the SUM complex. The morphology of individual TH-positive cells in the SUM region showed little variation in size or shape (Fig. 1B). Most of the cell bodies were small in size (8-12 pm in diameter) and typically spherical in shape. Nucleoli were eccentrically placed within a large centrally located nucleus. Cells were frequently characterized by 1 or 2 asymmetrically placed proximal dendrites. Distal dendrites arising from the larger diameter processes were relatively short and varicose. The size of the neurons within the SUM, IFN and VTA was determined. The area comprising the soma was estimated and compared among a total of 135 TH-positive neurons within these three regions. Only cells with a clearly defined nucleus were included in the analysis. Comparison of the average soma area occupied by cells in each group revealed significant differences among SUM, VTA and IFN TH-immunoreactive neurons, (means5S.E.M.: SUM=123.3 k6.4 pm’; VTA=204.8-c10.7 pm2; IFN=88.3?4.2 pm2; ANOVA: F=77.79, p
Characteristics
of SUM Neurons
Action potentials obtained from spontaneously active SUM neurons (N=24) were biphasic and of long duration (2.7kO.l msec; Fig. 2A). These cells exhibited slow, and sometimes erratic, fling rates (1.820.5 impulses/set).
310
SHEPARD,
MIHAILOFF
AND GERMAN
CHARACTERIZATION
OF DOPAMINE
NEURONS
SUM neurons could be antidromicahy activated from the anterior limbic cortex (N=9) as well as from the septal nucleus (N=14). In most cases, the antidromic potential consisted of a small initial segment spike (Fig. 2B and C). Antidromic spikes occurred at an average latency of 17.4215 msec, which corresponds to an average estimated conduction velocity of 0.43+0.06 m/set. The estimated conduction velocity of septal-projecting SUM neurons (0.50+0.09 m/set) was on the average somewhat faster than that observed for cortical-projecting SUM neurons (0.34+ 0.11 m/set), however, these differences did not reach statistical significance, Welch’s t(13)=1.54, ~~0.15. In an effort to determine whether SUM neurons possess rate-modulating autoreceptors, autoreceptor-selective doses of the direct-acting DA agonist, apomorphine (5 &kg), were administered intravenously. Of 11 cells tested, only 3 cells exhibited a statistically significant reduction in firing rate. In these cells, apomorphine produced a 25-40% decrease in firing rate for 3-5 min before the firing rate was increased by the subsequent administration of the DA antagonist, haloperidol (0.5 mg/kg). However, administration of the indirect-acting DA agonist, d-amphetamine (0.5-1.0 mg/kg), resulted in a significant decrease (30-50%) in the firing rate of 5 of 6 cells which had previously shown no response to apomorphine (Fig. 3). Both apomorphine and amphetamine-induced inhibitions in fling rate, when observed, were reversed by subsequent administration of haloperidol (0.1-0.5 mg/kg). No differences were found to exist among the discharge properties or estimated conduction velocities associated with apomorphine-sensitive and insensitive cells. To further characterize the sensitivity of SUM neurons to DA autoreceptor agonists, the responsiveness of cells to iontophoretically administered DA was tested. Seven of nine SUM neurons were found to be insensitive to DA at ejection currents (5-10 nA) which have previously been shown to decrease substantia nigra DA neuronal impulse flow by 4060%. These cells were, however, inhibited by iontophoretitally applied GABA. In order to test the possibility that a population of SUM neurons exist in a “silent” or non-active state, the excitatory amino acid gluatamate was ejected continuously while the recording electrode was advanced through the SUM complex. Using this approach twelve cells were studied (Fig. 4). All 12 of the silent SUM neurons were found to be sensitive to the rate-decreasing effects of iontophoretically administered DA (>60% inhibition using 5 nA ejection current). In some cases, it was possible to antidromically activate these SUM neurons from the septal nucleus during glutamate iontophoresis (N=5). These neurons exhibited conduction velocities which were not significantly different from spontaneously active mammilloseptal SUM neurons. DISCUSSION Neuroanatomical
Characteristics
311
OF THE RAT
of SUM Neurons
In agreement with earlier studies [4,3 11, a small collection
A.
0.
FIG. 2. Electrophysiological properties of SUM neurons. (A) An action potential recorded from a SUM neuron. Notice the long spike duration. Time marker=1 msec, voltage marker=100 pV. (B) Antidromic activation of a SUM neuron. A spontaneous spike (s) is present at left and the shock (*) to the septal nucleus elicits an antidromic initial segment spike (a). Time marker=10 msec. (C) When the spontaneous spike precedes the shock by less than the antidromic conduction latency, collision occurs and no antidromic spike is observed. Time marker=10 msec.
of TH-positive neurons exist within the SUM region. The majority of these cells are situated within the medial portion of the SUM complex and extend in the rostral-caudal dimension from the level of the arcuate nucleus to the caudal pole of the mammillary bodies, where they are contiguous with the DA cells in the IFN. Although the antibody used in the present study labels neurons containing other catecholamines, previous work has demonstrated that dopaminebeta-hydroxylase, the enzyme required to convert DA to norepinephrine and epinephrine, is not found within the
FACING PAGE FIG. 1. Immunocytochemical staining of mammilloseptal neurons. (A) TH staining illustrating the relationship between neurons in the medial supramammillary complex (SUMm), the mammillary bodies (MM), the median forebrain bundle (MFB) and the rostral substantia nigra zona compacta (SNc). Arrow points to an electrolytic lesion at a single unit recording site. BAR=200 pm. (B) High power photomicrograph of TH-stained SUM neurons. BAR=60 pm. (C) TH-labeled fibers in the lateral septum. The lateral ventricle can be seen at the upper right portion of the figure. BAR=200 pm. (D) PHA-L injection site in the SUM. BAR=60 pm. (E) PHA-L labeled axons and terminals in the lateral septal nucleus (arrows). Note distribution of PHA-L labeled fibers in comparison to TH-positive fibers in(C). The lateral ventricle can be seen in the upper right comer of the figure. BAR=60 pm.
312
SHEPARD, 8.0
d-AMP 3.0
~
. ..
..
. ..
.
.
.
IO.51 . .. ..
(0.5) . . . . . . . . .._......___............................
APO Y
I
AND GERMAN
r’‘.,.“.........“‘........................................................................................ Glutamate
(11
HALO
15) 15) (IO) / . . . . . . . . . . . . . . . . . . + . . . . . . . . ..t.......... + . . . . . . . . .
10.8
MIHAILOFF
(0.2) (0.1) . . . . . . . .._.......__........... I...............
I
21.5
I
32.3
I
I
Time (min)
43.0
Time (min)
FIG. 3. Rate histogram illustrating the typical response of a spontaneously active SUM neuron to dopaminergic drugs. Intravenous administration of a cumulative dose of 20 /&kg (APO; 5, 5, and 10 pg/kg) had no effect on the activity of this cell. Subsequent administration of d-amphetamine (d-AMP, 0.5, 0.5 mg/kg) caused a 46% inhibition in firing rate. Intravenous administration of haloperidol (HALO, 0.2 mgikg) was effective in reversing the d-AMP-induced reduction in firing rate, and increased cell firing above baseline.
SUM complex [32]. These data suggest that within the SUM complex, anti-TH antibodies serve as a specific marker for DA-containing cells. The SUM neurons do not appear to be a rostra1 extension of the VTA neurons since the cytological characteristics of the TH-positive SUM neurons differ from the cells in the VTA. The perikarya of the SUM cells are spherical, 40% smaller in size than the pyramidal-shaped TH-positive neurons in the VTA, and contain 1 or 2 varicose dendrites as opposed to many. Using both Golgi [24] and fluorescent histochemical staining techniques [23], Phillipson has made similar distinctions between the catecholaminergic VTA and SUM neurons. The results of the present study provide direct evidence for a projection from the SUM complex to limbic and cortical forebrain regions. Injections of the orthograde tracer PHA-L into the SUM complex resulted in significant labeling within the lateral septum, diagonal band of Broca, bed nucleus of the stria terminalis, the olfactory tubercle and tract, prefrontal cortex, nucleus accumbens and hippocampal formation. Although many SUM neurons project to limbic and cortical forebrain sites, those neurons which contain DA are few. The majority of the SUM neurons appear to be nondopaminergic. For example, Swanson [3 1] has reported that over 50% of the VTA neurons labeled following injections of True Blue into the nucleus accumbens, hippocampus, entorhinal cortex and supragenual portions of the anterior limbic cortex were TH-positive, whereas less than 2% of the neurons in the SUM projecting to these regions were THpositive. Of the TH-positive neurons in the SUM region, the greatest percentage (30%) project to the lateral septal nucleus. In summary, these data indicate that dopaminergic neurons within the SUM do not simply constitute a rostral extension of cells within the VTA, but rather, represent an anatomically distinct subpopulation of DA neurons. And unlike the neurons within the VTA which provide a major dopaminergic innervation of several limbic and cortical forebrain regions, the SUM neurons provide only a minor
FIG. 4. Rate histogram illustrating the typical response of a silent SUM neuron to iontophoretically applied dopamine (DA). Prior to the application of glutamate the cell exhibits a complete absence of spontaneous activity. Application of DA at 5-10 nA, during glutamate iontophoresis, inhibited the firing rate of the cell by 48% and 9C%, respectively. GABA was also effective in inhibiting the firing rate of this cell. Haloneridol (HALO: 0.1 ma/kg IV) was administered in order to test whether the cell was “sil&t” because of a tonic dopaminergic inhibition of firing. However, this was not the case since the cell remained silent after the drug was administered.
dopaminergic innervation of these forebrain regions (the most extensive dopaminergic innervation going to the lateral septal nucleus). ElPctrophysiological of SUM Neurons
and Pharmacological
Characteristics
The electrophysiological and pharmacological properties of neurons within the SUM region suggest the existence of two types of septal-projecting neurons. One group of cells was characterized by an absence of spontaneous activity. However, when induced to tire by iontophoretic application of glutamate, these cells exhibited several electrophysiological characteristics similar to VTA DA neurons [5, 10,33,35]. These SUM neurons exhibited long duration action potentials, slow conduction velocities, and a tendency to discharge in bursts. In addition, these neurons were inhibited by iontophoretically applied DA and GABA, as has been previously observed for the DA-containing cells within the VTA [l, 34, 351. These cells may represent mammilloseptal DA-containing neurons which possess impulse-regulating autoreceptors. The other group of cells exhibited fewer of the electrophysiological characteristics previously associated with midbrain DA neurons. Typical of VTA DA neurons, these SUM cells exhibited long duration action potentials, slow conduction velocities, and slow firing rates (0.2 to 4.8 impulses/set). However, following systemic administration of the direct-acting DA agonist, apomorphine, the majority of these cells were found to be refractory to the rate-decreasing effects of the drug. In addition, these cells were insensitive to iontophoretically administered DA. Thus, these cells do not respond to DA autoreceptor stimulation. These cells may, however, have postsynaptic DA receptors since administration of the indirect-acting DA agonist, d-amphetamine, decreased the firing rates of 83% of the cells tested, and this inhibition was reversed by the DA receptor antagonist, haloperidol. Since postsynaptic DA receptors have been reported to have a lower sensitivity to DA than presynaptic
CHARACTERIZATION
OF DOPAMINE
NEURONS
313
OF THE RAT
DA receptors (i.e., autoreceptors) [27], it is possible that d-amphetamine released a higher concentration of DA onto these cells than was applied by microiontophoresis (i.e., low microiontophoretic currents were used, 5-10 nA). Likewise, apomorphine would not be expected to activate postsynaptic DA receptors since it too was given at a presynaptic dosage. Since these SUM neurons appear to lack DA autoreceptors, which are characteristic of the DA neurons in the VTA [5,10], these SUM neurons appear to be non-dopaminergic. In conclusion, immunocytochemical, electrophysiological and pharmacological techniques were used to examine the properties of neurons within the SUM complex. Evidence is presented which indicates that two populations of SUM neurons exist which project to the lateral septal nucleus and other limbic forebrain regions. One population lacks spontaneous activity, and exhibits electrophysiological and pharmacological properties which are similar to the VTA DA neurons. The other population exhibits spontaneous ac-
tivity and appears to be non-dopaminergic. The existence of both dopaminergic and non-dopaminergic mammilloseptal neurons is consistent with evidence previously presented to indicate that there are both dopaminergic and nondopaminergic nigrostriatal neurons [ 111, mesolimbic neurons [lo] and mesocortical neurons [8]. Further studies are necessary to directly characterize the dopaminergic and nondopaminergic nature of these two populations of SUM neurons. ACKNOWLEDGEMENTS
The authors wish to thank Dr. Joachim Raese for kindly providing the TH antiserum used in this study, G. Bamett, S. Askari and K. Bourell for excellent histological work, Ms. L. Boynton for manuscript preparation and Dr. Gerald Kozlowski for helpful discussion of the manuscript. This work was supported in part by United States Public Health Service Grant MH-30546, the BristolMyers Co. and the Upjohn Co.
REFERENCES 1. Aghajanian, G. K. and B. S. Bunney. Dopamine “autoreceptors”: Pharmacological characterization by microiontophoretic single cell recording studies. Naunyn Schmiedebergs Arch Pharmacol297: 1-7, 1977. 2. Beckstead, R. M., V. B. Domesick
and J. H. Nauta. Efferent connections of the substantia nigra and ventral tegmental area in the rat. Brain Res 175: 191-217, 1979. 3. Berger, B., M. Thierry, J. P. Tassin and M. A. Moyne. Dopaminergic innervation of the rat prefrontal cortex-a fluorescence histochemical study. Brain Res 106: 133-145, 1976. 4. Carter, D. A. and H. C. Fibiger. Ascending projections of presumed dopamine-containing neurons in the ventral tegmenturn of the rat as demonstrated by horseradish peroxidase. Neuroscience
2: 569-576,
1977.
5. Chiodo, L. A., M. J. Bannon, A. A. Grace, R. H. Roth and B.
S. Bunney. Evidence for the absence of impulse regulating somatodendritic and synthesis-modulating nerve terminal autoreceptors on subpopulations of mesocortical dopamine neurons. Neuroscience 12: l-16, 1984. 6. Creese, I., D. R. Burt and S. H. Synder. Dopamine receptor binding predicts clinical and pharmacological potencies of antipsychotic drugs. Science 192: 481-483, 1976. 7. Dahlstrom, A. and K. Fuxe. Evidence for the existence of monoamine-containing neurons in the CNS. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acra Physiol Stand 62: l-55, 1964. 8. Deniau, J. M., A. M. Thierry and J. Feger. Electrophysiological
identification of mesencephalic ventromedial tegmental (VMT) neurons projecting to the frontal cortex, septum and nucleus accumbens. Brain Res 189: 315-326, 1980. 9. Gerfen, C. R. and P. E. Sawchenko. An anteriorgrade neuroanatomical tracing method that shows the detailed mortheir axons and terminals: Imphology of neurons, munocytochemical localization of an axonally transported plant lectin, Phaseolus vulgaris leuco-agglutinin (PHA-L). Brain Res 290: 219238,
1984.
10. German, D. C., M. Dalsass and R. S. Kiser. Electrophysiological examination of the ventral tegmental (AlO) area of the rat. Brain Res 181: 191-197, 1980. 11. Guyenet, P. G. and G. K. Aghajanian. Antidromic identification of dopaminergic and other output neurons of the rat substantia nigra. Brain Res 150: 69-84, 1978. 12. Hancock, M. B. Visualization of peptide-immunoreactive processes on serotonin-immunoreactive cells using two-color immunoperoxidase staining. J Histochem Cytochem 32: 311-314, 1984.
13. Hokfelt, T., 0. Johansson, K. Fuxe, M. Goldstein and D. Park. Immunocytochemical studies on the localization and distribution of monoamine neurons in the rat brain-II. Tyrosine hydroxylase in the telecephalon. Med Biol 55: 21-40, 1977. 14. Kelly, P. H., P. W. Seviour and S. D. Iverson. Amphetamine and apomorphine responses in the rat following 6-OHDA lesions of the nucleus accumbens septi and corpus striatum. Brain Res 94: 507-522.
1975.
15. Konig, J. F. R. and R. A. Klippel. The Rat Brain: A Stereotaxic Atlas. Baltimore: Williams and Wilkins, 1963. 16. Kozlowski, G. and G. Nilaver. Immunoelectron microscopy of neuropeptides: Theoretical and technical considerations. In: Current Methods in Cellular Neurobiology, Vol I, Anatomical Techniques. New York: Wiley and Sons,.1983, pp. 133-174.
17. Lindvall, 0. and A. Biorklund. The oraanization of the ascending catecholamine neuron systems in th; rat brain as revealed by the glyoxylic acid fluorescence method. Acta Physiol Stand 412: l-48,
1974.
18. Lindvall, O., A. Bjorklund and I. Divac. Organization of mesencephalic dopamine neurons projecting to the neocortex and septum. Adv Biochem Psychopharmacol 16: 39-46, 1977. 19. Lindvall, 0. and U. Stenevi. Dopamine and noradrenaline neurons projecting to the septal area in the rat. Cell Tissue Res 190: 383-407,
1978.
20. McMillen, B. A., R. T. Matthews, M. K. Sanghera, P. D. Shepard and D. C. German. Dopamine receptor antagonism by the novel antianxiety drug, Buspirone. J Neurosci 3: 733-738, 1983.
21. Meltzer, H. Y. and S. M. Stahl. The dopamine hypothesis of schizophrenia. Schizophr Bull 2: 19-76, 1976. 77 __. Moore, R. Y. Catecholamine innervation of the basal forebrain-I. The septal area. J Comp Neurol 177: 665-684, 1978. 23. Phillipson, 0. T. The cytoarchitecture
of the interfascicular nucleus and ventral tegmental area of Tsai in the rat. J Comp
Neurol 187: 85-98, 1979. 24. Phillipson, 0. T. A golgi study of the ventral tegmental area of Tsai and interfascicular nucleus in the rat. J Comp Neurol 187:
99-116, 1979. 25. Pijnenburg, A. J. J., W. M. M. Honig, J. A. M. Van Der Meyden and J. M. Van Rossum. Effects of chemical stimulation of the mesolimbic dopamine system upon locomotor activity. Eur J Pharmacol 35: 45-58, 1976. 26. Pijnenburg, A. J. J. and J. M. Van Rossum.
Stimulation of locomotor activity following injection of dopamine into the nucleus accumbens. J Pharm Pharmacol25: 1003-1004, 1973.
314
27. Skirboll, L. R., A. A. Grace and B. S. Bunney. Dopamine auto- and postsynaptic receptors: electrophysiological evidence for differential sensitivity to dopamine agonists. Science 206: 80-82, 1979. 28. Snyder, S. H. Dopamine receptors, neuroleptics and schizophrenia. Am J Psychiatry 138: 460-464, 1981. 29. Spencer, S., C. B. Saper, T. Joh, D. J. Reis, M. Goldstein and J. Raese. Distribution of catecholamine-containing neurons in the normal human hypothalamus. Brain Res 328: 73-80, 1985. 30. Stemberger, L. A., P. H. Hardy, J. J. Cuculis and H. G. Meyer. The unlabeled antibody enzyme method of immunocytochemistry. preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidasel and its use in identification of spirochetes. J Hisrochem Cytochem 18: 315-333, 1970.
SHEPARD,
MIHAILOFF
AND GERMAN
31. Swanson, L. W. The projections of the ventral tegmental area and adjacent regions: A combined fluorescent retrograde tracer and immunofluorescence study in the rat. Brain Res Bull 9: 321-353, 1982. 32. Swanson, L. W. and B. K. Hartman. The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-P-hydroxylase as a marker. J Comp Neural 163: 467-506, 1975. 33. Wang, R. Y. Dopaminergic neurons in the rat ventral tegmental area. I. Identification and characterization. Brain Res Rev 3: 123-140, 1981. 34. Wang, R. Y. Dopaminergic neurons in the rat ventral tegmental area. II. Evidence for autoregulation. Brain Res Rev 3: 141-151, 1981. 35. Yim, C. Y. and G. J. Mogenson. Electrophysiological studies of neurons in the ventral tegmental area of Tsai. Brain Res 181: 301-313. 1980.
plc, 1988. Printed in the U.S.A.
0361-9230/88 $3.00 + .OO
Anatomical and Electrophysiological Characterization of Presumed Dopamine-Containing Neurons Within the Supramammillary Region of the Rat PAUL D. SHEPARD,*
GREGORY
A. MIHAILOFFt
AND DWIGHT
C. GERMAN*J:’
*Departments of Physiology, Well Biology and 3Psychiatt-y University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, Received
Dallas, TX 75235
27 August 1986
SHEPARD, P. D., G. A. MIHAILOFF AND D. C. GERMAN. Anatomical and electrophysiological characterization of presumed dopamine-containing neurons within the supramammillary region of the rat. BRAIN RES BULL 20(3) 307-3 14, 1988.-A combination of immunocytochemical, electrophysiological and pharmacological techniques were employed to study the properties of neurons within the supramammillary (SUM) complex of the rat. The SUM region contains a small, but dense, population of tyrosine hydroxylase immunoreactive neurons. Following injection of the orthograde neuroanatomical tracer, Phaseolus Vulgaris leucoagglutinin, into the SUM region, heavy termainal labeling was observed in the lateral septal nucleus, diagonal band of Broca and bed nucleus of the stria terminalis. The electrophysiological and pharmacological properties of antidromicalIy-activated SUM neurons revealed evidence of two neuronal populations. Both groups of neurons exhibited long duration action potentials (>2 msec) and slow conduction velocities (CO.5 m/set). However, cells in one group were characterized by slow and erratic firing rates and insensitivity to dopamine (DA) autoreceptor agonists. Cells in the other group typically exhibited no spontaneous activity but could be induced to discharge by iontophoretic application of glutamate. These latter cells were sensitive to DA autoreceptor stimulation. Of the two populations of mammilloseptal SUM neurons, the silent population exhibited several properties similar to those of midbrain DA neurons. Supramammillary nucleus Lateral septal nucleus Dopamine Electrophysiology Phaseolus Vulgaris leucoagglutinin Tyrosine hydroxylase immunohistochemistry
tyrosine hydroxylase (TH), the rate-limiting enzyme in the synthesis of DA, but not the noradrenergic synethetic dopamine-beta hydroxylase [32]. These THenzyme, positive SUM neurons provide a major input to the lateral septal nucleus [4,31]. In fact, more septal alferents arise from TH-containing cells in the SUM region than in the A10 region. Despite evidence suggesting that neurons within the SUM region provide an important source of dopaminergic afferents to the septal nuclei, there is no information concerning the electrophysiological or pharmacological characteristics of these cells. Of particular interest is whether SUM DA neurons possess rate-moduiating autoreceptors and are thereby subject to the same autoregulatory influences as other mesolimbic DA neurons [I, 10, 34, 351. In this report, we describe experiments in which neuroanatomical, electrophysiological and pharmacological techniques were used to characterize mammilloseptal neurons. Evidence is pro-
THE mesolimbic dopamine (DA) system has been the subject of an intense experimental effort because of its role in locomotion [ 14, 25, 261 as well as its possible involvement in the etiology of schizophrenia [6, 21, 281. Traditionally, the cell bodies of mesolimbic DA neurons have been thought to be confined within the boundaries of the ventral tegmental area of Tsai (VTA) (nucleus A10 of Dahlstrom and Fuxe [7]), where they give rise to fibers which innervate a variety of forebrain areas including the nucleus accumbens, olfactory tubercle, amygdala, anterior limbic cortex and septal regions including the nucleus of the diagonal band, bed nucleus of the stria terminalis and the lateral septal nucleus [2, 3,7, 13, 17-19, 221. Recent neuroanatomical studies have indentified a small collection of catecholamine-containing cells within the supramammillary (SUM) region of the hypothalamus, which may represent a rostral extension of the A10 cell group [31]. These cells appear to be dopaminergic since they contain
‘Requests for reprints should be addressed to Dwight C. German.
307
308
SHEPARD,
vided which suggests that a subpopulation of these cells constitutes a separate nucleus of the mesolimic DA system, distinct from the cells within the VTA. METHOD
Surgical Procedures
Male Sprague-Dawley rats, weighing 200-350 g were used as experimental subjects. Animals were anesthetized with chloral hydrate (400 mglkg, IP) and mounted in a Kopf stereotaxic frame with the incisor bar positioned 2.4 mm below the interaural line. Rats used in electrophysiological experiments were cannulated (femoral vein) for intravenous administration of drugs and supplementary anesthetic. All incision and soft tissue pressure points were infiltrated with a long-acting local anesthetic (mepivacaine HCI, 2%). Body temperature was maintained between 3637°C using an aquatic heating pad. Following a midline incision and reflection of underlying tissue, a burr hole was drilled in the skull overlying the region of the posterior hypothalamus (anterior, 2580-3180 pm; lateral, O-1000 pm, according to the atlas of Konig and Klippel [ 151) and the dura removed. In some experiments, an additional burr hole was drilled over the regions corresponding to the lateral septal nucleus (anterior, 7020-8920 pm, lateral 300-1000 pm) and/or anterior limbic cortex (anterior, 8620-11050 pm, lateral 800-2000 pm) for insertion of stimulating electrodes. Tyrosine Hydroxylase
lmmunocytochemistry
The distribution of putative DA-containing neurons within the SUM complex was characterized using an antise~m to bovine adrenal tyrosine hydroxylase. Four male rats were deeply anesthetized with pentobarbital (60 mg/kg, IP) and perfused transcardially with neutral buffered saline followed by 10% buffered formalin. Following decapitation and removal of the calvaria, the brain was placed in formalin and post-fixed for an additional 7 days after which it was blocked stereotaxically according to the atlas of Konig and Klippel(l51. Sections were cut at 40 pm using a freezing microtome and processed for immunocytochemical localization of TH according to a modification of the perioxidaseantiperoxidase (PAP) method of Sternberger [29] as described by Kozlowski and Nilaver [161. Briefly, free-floating sections were incubated for 14-16 hr at room temperature in a 1: 1000 dilution of rabbit anti-TH antiserum. Specificity of the antiserum used in these experiments is described elsewhere [29]. Following incubation with the primary antiserum, sections were incubated in a solution of sheep anti-rabbit IgG (1: IOO), washed and incubated in a 1:500 dilution of rabbit PAP complex. Sections were developed in a mixture of diaminobenzidine and Ha,. The reaction product was then intensified with nickel according to the method described by Hancock [ 121.
Cell areas were measured using a Leitz microscope which was interfaced, via a drawing tube and digitizing tablet, to a Data General computer. Using a 100x oil immersion objective (total magnification= 1125x), TH-positive somata were identified within the SUM, interfascisular nucleus (IFN) and VTA. A line was drawn around the perimeter of each neuron in which a nucleus was present. After all somata in a section were entered into the computer, mean cell areas were determined.
MIHAILOFF
AND GERMAN
Axonai Tracing
An orthograde tracing method, developed by Gerfen and Sawchenko [9], was used to characterize the distribution of septal afferents arising from neurons within the SUM complex. This method makes it possible to visualize the morphology of axons and their specializations within terminal fields. Moreover, a high degree of regional specificity can be achieved as the label appears not to be transported by fibers of passage. Four animals were prepared as outlined under “Surgical (1.5 mm Procedures. ’ ’ Single barrel glass micropi~ttes o.d.; A-M Systems), containing a small glass capillary, were pulled to a fine taper using a Kopf pipette puller. The tips were broken back to a diameter of 10-15 pm. Electrodes were filled with a 2.5% solution of Phaseolus Vulgaris leucoa~lutinin (PHA-L; Vector Labs, CA) in 0.05 M phosphate buffered saline (PBS) and affixed to a hydraulic microdrive. Electrodes were driven rapidly to a depth of 7.8 mm below the surface of the cortex and a small quantity of PHA-L was iontophoretically delivered using an intermittent (7 set ON; 7 set OFF) application of 5-7 &A of anodal current for 20-30 min. After survival times which ranged from 8-12 days, rats were deeply anesthetized and perfused transcardially with a solution of 4% paraformaldehyde in 0.1 M acetate buffer (pH 6.5) followed by a second fixative consisting of 4% paraformaldehyde and 0.05% glutaraldehyde in 0.1 M borate buffer (pH 9.5). The brain was removed from the cranium, and postfiied for an additional 16 hr in the last fixative. Sections were cut at 30 pm on a Vibratome and free-floating sections were processed for immunocytochemistry according to the avidin-biotin method (Vector Labs). Briefly, sections were incubated in goat anti-PHA-L antiserum for 24-48 hr at 4°C. Following incubation in the primary antiserum, sections were transferred to a solution of biotinylated rabbit anti-goat IgG (Vectastain kit, Vector Labs). The tissue was rinsed several times in saline and placed in the avidin-biotin-HRP complex (ABC) for 45 min. Following a second wash, the tissue was recycled through the IgG and ABC solutions and finally reacted with 0.05% diaminobenzidine and 0.04% H,Os in PBS. The reaction was stopped after 20 min. In some experiments, sections were counterstained with cresyl violet. Single Unit rewording
Animals were prepared as outlined in the section entitled “Surgical Procedures.” Epoxy-coated tungsten (Fredrick Haer) or S-barrel glass (R 62 D Optical) microelectrodes were used for recording extracellular action potentials. in vitro impedance was measured at 135 Hz and ranged from 2.ft4.0 Mfl. Electrodes were stereotaxically positioned over the region of the SUM complex and hydraulically driven to a depth of 6.0 mm below the surface of the cortex and the following 2.0 mm explored for spontaneously active cells. Electrode potentials were ~pl~ed and filtered using a high impedance amplifier (Grass P511, 0.3-3.0 KHz bandwidth) and monitored both visually (Tektronix 51 IS oscilloscope) and aurally (Grass AM8 audio amplifier). Single unit activity was recorded on magnetic tape. Data analysis was performed using a DEC LSI 11/03 mi~computer. Action potentials were disc~mina~d with an oscilioscope triggering circuit which provided a single square wave pulse coincident with each action potential. Rate histograms were compiled by summing action potentials across consecutive 10 set intervals.
CHARACTERIZATION
OF DOPAMINE
NEURONS
Microiontophoresis Standard microiontophoretic techniques were used as previously described [20]. Briefly, electrodes were preloaded with several strands of fine fiberlgas fdaments and pulled to a taper using a Narishige pipette puller. Electrode tips were broken back to a final diameter of 3-6 pm. The center barrel was filled with 2 M NaCl saturated with Fast Green dye and used to record extracellular neuronal signals. One of the four outer barrels was filled with 3 M NaCl and used for current balancing. Each of the remaining barrels was filled with one of the following drug solutions: dopamine-HCl (DA; 0.2 M, pH 4.0), gamma-aminobutyric acid (GABA; 0.05 M, pH 4.0) or l-sodium glutamate (GLU; 0.05 M, pH 8.0). DA and GABA were ejected using anodal current while glutamate was ejected using cathodal current. A retaining current of 10 nA was applied to all drug barrels to prevent leakage. Microiontophoretic retention, ejection and balancing currents were controlled automatically by a constant current apparatus (Medical Systems, Model BH-2). Electrical Stimulation Stainless steel concentric bipolar electrodes (Kopf SNEX-lOO), were used to antidromically activate SUM neurons projecting to the ipsilateral septum, prefrontal or cingulate cortex. Electrical stimulation was provided by a Grass S48 stimulator coupled to constant current and stimulus isolation units. Square wave, cathodal pulses, 500 psec in duration, ranging from 0.1 to 4.5 mA were used. Single units were considered to have been antidromically activated when the following criteria were met: (a) a fixed latency between the onset of the stimulus and the occurrence of the stimulus-induced spike; and (b) no stimulus-induced spike was observed when the stimulus was triggered at a latency equal to or less than the conduction time for the stimulus-induced spike. Localization
of Recording and Stimulating
Sites
At the conclusion of each experiment, the position of the recording electrode was marked by an electrolytic lesion (tungsten electrode; +12 /*A for 45 set) or iontophoretic ejection of Fast Green dye (glass electrode; -20 PA for 1 hr). Stimulation sites were also marked by passing 0.5 mA anodal current through the electrode for 15 sec. Following removal from the stereotaxic appartus the animal was deeply anesthetized with chloral hydrate and perfused transcardially with saline (60 ml) followed by 10% neutral buffered formalin containing 1 g of potassium ferrocyanide. Frozen sections were cut at 50 pm, mounted on subbed slides and stained with cresyl violet for the determination of the position of the recording and stimulating electrodes. RESULTS
Distribution and Morphology
of TH-Positive
309
OF THE RAT
Somata
A small collection of TH-positive neurons was observed within the SUM complex (Fig. IA). The majority of these cells were situated within the medial portion of the SUM complex which is bordered ventrolaterally by the mammillothalmic tract. This dense group of parvocellular THpositive neurons extends in a rostro-caudal direction from the level of the a.rcuate/median eminence to the caudal pole of the mammillary nuclei where they appear to be continuous with midline TH-positive cells in the IFN. At their rostralmost extension, the TH-positive cells are situated medially
within the posterior hypothalamus, ventral to the Al3 DA cell group and dorsal to the caudal-most portion of nucleus A12. At more caudal levels, a few labeled cells were also observed throughout the lateral portion of the SUM complex. The morphology of individual TH-positive cells in the SUM region showed little variation in size or shape (Fig. 1B). Most of the cell bodies were small in size (8-12 pm in diameter) and typically spherical in shape. Nucleoli were eccentrically placed within a large centrally located nucleus. Cells were frequently characterized by 1 or 2 asymmetrically placed proximal dendrites. Distal dendrites arising from the larger diameter processes were relatively short and varicose. The size of the neurons within the SUM, IFN and VTA was determined. The area comprising the soma was estimated and compared among a total of 135 TH-positive neurons within these three regions. Only cells with a clearly defined nucleus were included in the analysis. Comparison of the average soma area occupied by cells in each group revealed significant differences among SUM, VTA and IFN TH-immunoreactive neurons, (means5S.E.M.: SUM=123.3 k6.4 pm’; VTA=204.8-c10.7 pm2; IFN=88.3?4.2 pm2; ANOVA: F=77.79, p
Characteristics
of SUM Neurons
Action potentials obtained from spontaneously active SUM neurons (N=24) were biphasic and of long duration (2.7kO.l msec; Fig. 2A). These cells exhibited slow, and sometimes erratic, fling rates (1.820.5 impulses/set).
310
SHEPARD,
MIHAILOFF
AND GERMAN
CHARACTERIZATION
OF DOPAMINE
NEURONS
SUM neurons could be antidromicahy activated from the anterior limbic cortex (N=9) as well as from the septal nucleus (N=14). In most cases, the antidromic potential consisted of a small initial segment spike (Fig. 2B and C). Antidromic spikes occurred at an average latency of 17.4215 msec, which corresponds to an average estimated conduction velocity of 0.43+0.06 m/set. The estimated conduction velocity of septal-projecting SUM neurons (0.50+0.09 m/set) was on the average somewhat faster than that observed for cortical-projecting SUM neurons (0.34+ 0.11 m/set), however, these differences did not reach statistical significance, Welch’s t(13)=1.54, ~~0.15. In an effort to determine whether SUM neurons possess rate-modulating autoreceptors, autoreceptor-selective doses of the direct-acting DA agonist, apomorphine (5 &kg), were administered intravenously. Of 11 cells tested, only 3 cells exhibited a statistically significant reduction in firing rate. In these cells, apomorphine produced a 25-40% decrease in firing rate for 3-5 min before the firing rate was increased by the subsequent administration of the DA antagonist, haloperidol (0.5 mg/kg). However, administration of the indirect-acting DA agonist, d-amphetamine (0.5-1.0 mg/kg), resulted in a significant decrease (30-50%) in the firing rate of 5 of 6 cells which had previously shown no response to apomorphine (Fig. 3). Both apomorphine and amphetamine-induced inhibitions in fling rate, when observed, were reversed by subsequent administration of haloperidol (0.1-0.5 mg/kg). No differences were found to exist among the discharge properties or estimated conduction velocities associated with apomorphine-sensitive and insensitive cells. To further characterize the sensitivity of SUM neurons to DA autoreceptor agonists, the responsiveness of cells to iontophoretically administered DA was tested. Seven of nine SUM neurons were found to be insensitive to DA at ejection currents (5-10 nA) which have previously been shown to decrease substantia nigra DA neuronal impulse flow by 4060%. These cells were, however, inhibited by iontophoretitally applied GABA. In order to test the possibility that a population of SUM neurons exist in a “silent” or non-active state, the excitatory amino acid gluatamate was ejected continuously while the recording electrode was advanced through the SUM complex. Using this approach twelve cells were studied (Fig. 4). All 12 of the silent SUM neurons were found to be sensitive to the rate-decreasing effects of iontophoretically administered DA (>60% inhibition using 5 nA ejection current). In some cases, it was possible to antidromically activate these SUM neurons from the septal nucleus during glutamate iontophoresis (N=5). These neurons exhibited conduction velocities which were not significantly different from spontaneously active mammilloseptal SUM neurons. DISCUSSION Neuroanatomical
Characteristics
311
OF THE RAT
of SUM Neurons
In agreement with earlier studies [4,3 11, a small collection
A.
0.
FIG. 2. Electrophysiological properties of SUM neurons. (A) An action potential recorded from a SUM neuron. Notice the long spike duration. Time marker=1 msec, voltage marker=100 pV. (B) Antidromic activation of a SUM neuron. A spontaneous spike (s) is present at left and the shock (*) to the septal nucleus elicits an antidromic initial segment spike (a). Time marker=10 msec. (C) When the spontaneous spike precedes the shock by less than the antidromic conduction latency, collision occurs and no antidromic spike is observed. Time marker=10 msec.
of TH-positive neurons exist within the SUM region. The majority of these cells are situated within the medial portion of the SUM complex and extend in the rostral-caudal dimension from the level of the arcuate nucleus to the caudal pole of the mammillary bodies, where they are contiguous with the DA cells in the IFN. Although the antibody used in the present study labels neurons containing other catecholamines, previous work has demonstrated that dopaminebeta-hydroxylase, the enzyme required to convert DA to norepinephrine and epinephrine, is not found within the
FACING PAGE FIG. 1. Immunocytochemical staining of mammilloseptal neurons. (A) TH staining illustrating the relationship between neurons in the medial supramammillary complex (SUMm), the mammillary bodies (MM), the median forebrain bundle (MFB) and the rostral substantia nigra zona compacta (SNc). Arrow points to an electrolytic lesion at a single unit recording site. BAR=200 pm. (B) High power photomicrograph of TH-stained SUM neurons. BAR=60 pm. (C) TH-labeled fibers in the lateral septum. The lateral ventricle can be seen at the upper right portion of the figure. BAR=200 pm. (D) PHA-L injection site in the SUM. BAR=60 pm. (E) PHA-L labeled axons and terminals in the lateral septal nucleus (arrows). Note distribution of PHA-L labeled fibers in comparison to TH-positive fibers in(C). The lateral ventricle can be seen in the upper right comer of the figure. BAR=60 pm.
312
SHEPARD, 8.0
d-AMP 3.0
~
. ..
..
. ..
.
.
.
IO.51 . .. ..
(0.5) . . . . . . . . .._......___............................
APO Y
I
AND GERMAN
r’‘.,.“.........“‘........................................................................................ Glutamate
(11
HALO
15) 15) (IO) / . . . . . . . . . . . . . . . . . . + . . . . . . . . ..t.......... + . . . . . . . . .
10.8
MIHAILOFF
(0.2) (0.1) . . . . . . . .._.......__........... I...............
I
21.5
I
32.3
I
I
Time (min)
43.0
Time (min)
FIG. 3. Rate histogram illustrating the typical response of a spontaneously active SUM neuron to dopaminergic drugs. Intravenous administration of a cumulative dose of 20 /&kg (APO; 5, 5, and 10 pg/kg) had no effect on the activity of this cell. Subsequent administration of d-amphetamine (d-AMP, 0.5, 0.5 mg/kg) caused a 46% inhibition in firing rate. Intravenous administration of haloperidol (HALO, 0.2 mgikg) was effective in reversing the d-AMP-induced reduction in firing rate, and increased cell firing above baseline.
SUM complex [32]. These data suggest that within the SUM complex, anti-TH antibodies serve as a specific marker for DA-containing cells. The SUM neurons do not appear to be a rostra1 extension of the VTA neurons since the cytological characteristics of the TH-positive SUM neurons differ from the cells in the VTA. The perikarya of the SUM cells are spherical, 40% smaller in size than the pyramidal-shaped TH-positive neurons in the VTA, and contain 1 or 2 varicose dendrites as opposed to many. Using both Golgi [24] and fluorescent histochemical staining techniques [23], Phillipson has made similar distinctions between the catecholaminergic VTA and SUM neurons. The results of the present study provide direct evidence for a projection from the SUM complex to limbic and cortical forebrain regions. Injections of the orthograde tracer PHA-L into the SUM complex resulted in significant labeling within the lateral septum, diagonal band of Broca, bed nucleus of the stria terminalis, the olfactory tubercle and tract, prefrontal cortex, nucleus accumbens and hippocampal formation. Although many SUM neurons project to limbic and cortical forebrain sites, those neurons which contain DA are few. The majority of the SUM neurons appear to be nondopaminergic. For example, Swanson [3 1] has reported that over 50% of the VTA neurons labeled following injections of True Blue into the nucleus accumbens, hippocampus, entorhinal cortex and supragenual portions of the anterior limbic cortex were TH-positive, whereas less than 2% of the neurons in the SUM projecting to these regions were THpositive. Of the TH-positive neurons in the SUM region, the greatest percentage (30%) project to the lateral septal nucleus. In summary, these data indicate that dopaminergic neurons within the SUM do not simply constitute a rostral extension of cells within the VTA, but rather, represent an anatomically distinct subpopulation of DA neurons. And unlike the neurons within the VTA which provide a major dopaminergic innervation of several limbic and cortical forebrain regions, the SUM neurons provide only a minor
FIG. 4. Rate histogram illustrating the typical response of a silent SUM neuron to iontophoretically applied dopamine (DA). Prior to the application of glutamate the cell exhibits a complete absence of spontaneous activity. Application of DA at 5-10 nA, during glutamate iontophoresis, inhibited the firing rate of the cell by 48% and 9C%, respectively. GABA was also effective in inhibiting the firing rate of this cell. Haloneridol (HALO: 0.1 ma/kg IV) was administered in order to test whether the cell was “sil&t” because of a tonic dopaminergic inhibition of firing. However, this was not the case since the cell remained silent after the drug was administered.
dopaminergic innervation of these forebrain regions (the most extensive dopaminergic innervation going to the lateral septal nucleus). ElPctrophysiological of SUM Neurons
and Pharmacological
Characteristics
The electrophysiological and pharmacological properties of neurons within the SUM region suggest the existence of two types of septal-projecting neurons. One group of cells was characterized by an absence of spontaneous activity. However, when induced to tire by iontophoretic application of glutamate, these cells exhibited several electrophysiological characteristics similar to VTA DA neurons [5, 10,33,35]. These SUM neurons exhibited long duration action potentials, slow conduction velocities, and a tendency to discharge in bursts. In addition, these neurons were inhibited by iontophoretically applied DA and GABA, as has been previously observed for the DA-containing cells within the VTA [l, 34, 351. These cells may represent mammilloseptal DA-containing neurons which possess impulse-regulating autoreceptors. The other group of cells exhibited fewer of the electrophysiological characteristics previously associated with midbrain DA neurons. Typical of VTA DA neurons, these SUM cells exhibited long duration action potentials, slow conduction velocities, and slow firing rates (0.2 to 4.8 impulses/set). However, following systemic administration of the direct-acting DA agonist, apomorphine, the majority of these cells were found to be refractory to the rate-decreasing effects of the drug. In addition, these cells were insensitive to iontophoretically administered DA. Thus, these cells do not respond to DA autoreceptor stimulation. These cells may, however, have postsynaptic DA receptors since administration of the indirect-acting DA agonist, d-amphetamine, decreased the firing rates of 83% of the cells tested, and this inhibition was reversed by the DA receptor antagonist, haloperidol. Since postsynaptic DA receptors have been reported to have a lower sensitivity to DA than presynaptic
CHARACTERIZATION
OF DOPAMINE
NEURONS
313
OF THE RAT
DA receptors (i.e., autoreceptors) [27], it is possible that d-amphetamine released a higher concentration of DA onto these cells than was applied by microiontophoresis (i.e., low microiontophoretic currents were used, 5-10 nA). Likewise, apomorphine would not be expected to activate postsynaptic DA receptors since it too was given at a presynaptic dosage. Since these SUM neurons appear to lack DA autoreceptors, which are characteristic of the DA neurons in the VTA [5,10], these SUM neurons appear to be non-dopaminergic. In conclusion, immunocytochemical, electrophysiological and pharmacological techniques were used to examine the properties of neurons within the SUM complex. Evidence is presented which indicates that two populations of SUM neurons exist which project to the lateral septal nucleus and other limbic forebrain regions. One population lacks spontaneous activity, and exhibits electrophysiological and pharmacological properties which are similar to the VTA DA neurons. The other population exhibits spontaneous ac-
tivity and appears to be non-dopaminergic. The existence of both dopaminergic and non-dopaminergic mammilloseptal neurons is consistent with evidence previously presented to indicate that there are both dopaminergic and nondopaminergic nigrostriatal neurons [ 111, mesolimbic neurons [lo] and mesocortical neurons [8]. Further studies are necessary to directly characterize the dopaminergic and nondopaminergic nature of these two populations of SUM neurons. ACKNOWLEDGEMENTS
The authors wish to thank Dr. Joachim Raese for kindly providing the TH antiserum used in this study, G. Bamett, S. Askari and K. Bourell for excellent histological work, Ms. L. Boynton for manuscript preparation and Dr. Gerald Kozlowski for helpful discussion of the manuscript. This work was supported in part by United States Public Health Service Grant MH-30546, the BristolMyers Co. and the Upjohn Co.
REFERENCES 1. Aghajanian, G. K. and B. S. Bunney. Dopamine “autoreceptors”: Pharmacological characterization by microiontophoretic single cell recording studies. Naunyn Schmiedebergs Arch Pharmacol297: 1-7, 1977. 2. Beckstead, R. M., V. B. Domesick
and J. H. Nauta. Efferent connections of the substantia nigra and ventral tegmental area in the rat. Brain Res 175: 191-217, 1979. 3. Berger, B., M. Thierry, J. P. Tassin and M. A. Moyne. Dopaminergic innervation of the rat prefrontal cortex-a fluorescence histochemical study. Brain Res 106: 133-145, 1976. 4. Carter, D. A. and H. C. Fibiger. Ascending projections of presumed dopamine-containing neurons in the ventral tegmenturn of the rat as demonstrated by horseradish peroxidase. Neuroscience
2: 569-576,
1977.
5. Chiodo, L. A., M. J. Bannon, A. A. Grace, R. H. Roth and B.
S. Bunney. Evidence for the absence of impulse regulating somatodendritic and synthesis-modulating nerve terminal autoreceptors on subpopulations of mesocortical dopamine neurons. Neuroscience 12: l-16, 1984. 6. Creese, I., D. R. Burt and S. H. Synder. Dopamine receptor binding predicts clinical and pharmacological potencies of antipsychotic drugs. Science 192: 481-483, 1976. 7. Dahlstrom, A. and K. Fuxe. Evidence for the existence of monoamine-containing neurons in the CNS. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acra Physiol Stand 62: l-55, 1964. 8. Deniau, J. M., A. M. Thierry and J. Feger. Electrophysiological
identification of mesencephalic ventromedial tegmental (VMT) neurons projecting to the frontal cortex, septum and nucleus accumbens. Brain Res 189: 315-326, 1980. 9. Gerfen, C. R. and P. E. Sawchenko. An anteriorgrade neuroanatomical tracing method that shows the detailed mortheir axons and terminals: Imphology of neurons, munocytochemical localization of an axonally transported plant lectin, Phaseolus vulgaris leuco-agglutinin (PHA-L). Brain Res 290: 219238,
1984.
10. German, D. C., M. Dalsass and R. S. Kiser. Electrophysiological examination of the ventral tegmental (AlO) area of the rat. Brain Res 181: 191-197, 1980. 11. Guyenet, P. G. and G. K. Aghajanian. Antidromic identification of dopaminergic and other output neurons of the rat substantia nigra. Brain Res 150: 69-84, 1978. 12. Hancock, M. B. Visualization of peptide-immunoreactive processes on serotonin-immunoreactive cells using two-color immunoperoxidase staining. J Histochem Cytochem 32: 311-314, 1984.
13. Hokfelt, T., 0. Johansson, K. Fuxe, M. Goldstein and D. Park. Immunocytochemical studies on the localization and distribution of monoamine neurons in the rat brain-II. Tyrosine hydroxylase in the telecephalon. Med Biol 55: 21-40, 1977. 14. Kelly, P. H., P. W. Seviour and S. D. Iverson. Amphetamine and apomorphine responses in the rat following 6-OHDA lesions of the nucleus accumbens septi and corpus striatum. Brain Res 94: 507-522.
1975.
15. Konig, J. F. R. and R. A. Klippel. The Rat Brain: A Stereotaxic Atlas. Baltimore: Williams and Wilkins, 1963. 16. Kozlowski, G. and G. Nilaver. Immunoelectron microscopy of neuropeptides: Theoretical and technical considerations. In: Current Methods in Cellular Neurobiology, Vol I, Anatomical Techniques. New York: Wiley and Sons,.1983, pp. 133-174.
17. Lindvall, 0. and A. Biorklund. The oraanization of the ascending catecholamine neuron systems in th; rat brain as revealed by the glyoxylic acid fluorescence method. Acta Physiol Stand 412: l-48,
1974.
18. Lindvall, O., A. Bjorklund and I. Divac. Organization of mesencephalic dopamine neurons projecting to the neocortex and septum. Adv Biochem Psychopharmacol 16: 39-46, 1977. 19. Lindvall, 0. and U. Stenevi. Dopamine and noradrenaline neurons projecting to the septal area in the rat. Cell Tissue Res 190: 383-407,
1978.
20. McMillen, B. A., R. T. Matthews, M. K. Sanghera, P. D. Shepard and D. C. German. Dopamine receptor antagonism by the novel antianxiety drug, Buspirone. J Neurosci 3: 733-738, 1983.
21. Meltzer, H. Y. and S. M. Stahl. The dopamine hypothesis of schizophrenia. Schizophr Bull 2: 19-76, 1976. 77 __. Moore, R. Y. Catecholamine innervation of the basal forebrain-I. The septal area. J Comp Neurol 177: 665-684, 1978. 23. Phillipson, 0. T. The cytoarchitecture
of the interfascicular nucleus and ventral tegmental area of Tsai in the rat. J Comp
Neurol 187: 85-98, 1979. 24. Phillipson, 0. T. A golgi study of the ventral tegmental area of Tsai and interfascicular nucleus in the rat. J Comp Neurol 187:
99-116, 1979. 25. Pijnenburg, A. J. J., W. M. M. Honig, J. A. M. Van Der Meyden and J. M. Van Rossum. Effects of chemical stimulation of the mesolimbic dopamine system upon locomotor activity. Eur J Pharmacol 35: 45-58, 1976. 26. Pijnenburg, A. J. J. and J. M. Van Rossum.
Stimulation of locomotor activity following injection of dopamine into the nucleus accumbens. J Pharm Pharmacol25: 1003-1004, 1973.
314
27. Skirboll, L. R., A. A. Grace and B. S. Bunney. Dopamine auto- and postsynaptic receptors: electrophysiological evidence for differential sensitivity to dopamine agonists. Science 206: 80-82, 1979. 28. Snyder, S. H. Dopamine receptors, neuroleptics and schizophrenia. Am J Psychiatry 138: 460-464, 1981. 29. Spencer, S., C. B. Saper, T. Joh, D. J. Reis, M. Goldstein and J. Raese. Distribution of catecholamine-containing neurons in the normal human hypothalamus. Brain Res 328: 73-80, 1985. 30. Stemberger, L. A., P. H. Hardy, J. J. Cuculis and H. G. Meyer. The unlabeled antibody enzyme method of immunocytochemistry. preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidasel and its use in identification of spirochetes. J Hisrochem Cytochem 18: 315-333, 1970.
SHEPARD,
MIHAILOFF
AND GERMAN
31. Swanson, L. W. The projections of the ventral tegmental area and adjacent regions: A combined fluorescent retrograde tracer and immunofluorescence study in the rat. Brain Res Bull 9: 321-353, 1982. 32. Swanson, L. W. and B. K. Hartman. The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-P-hydroxylase as a marker. J Comp Neural 163: 467-506, 1975. 33. Wang, R. Y. Dopaminergic neurons in the rat ventral tegmental area. I. Identification and characterization. Brain Res Rev 3: 123-140, 1981. 34. Wang, R. Y. Dopaminergic neurons in the rat ventral tegmental area. II. Evidence for autoregulation. Brain Res Rev 3: 141-151, 1981. 35. Yim, C. Y. and G. J. Mogenson. Electrophysiological studies of neurons in the ventral tegmental area of Tsai. Brain Res 181: 301-313. 1980.