- Email: [email protected]
Morphological features and electrophysiological properties of serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus
Brain Research 900 (2001) 110–118 www.elsevier.com / locate / bres
Research report
Morphological features and electrophysiological properties of serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus An intracellular recording and labeling study in rat brain slices a a b c, Yun-Qing Li , Hui Li , Takeshi Kaneko , Noboru Mizuno * a
Department of Anatomy and K.K. Leung Brain Research Centre, The Fourth Military Medical University, Xi’ an 710032, People’ s Republic of China b Department of Morphological Brain Science, Graduate School of Medicine, Kyoto University, Kyoto 606 -8501, Japan c Tokyo Metropolitan Institute for Neuroscience, Fuchu, Tokyo 183 -8526, Japan Accepted 6 February 2001
Abstract The morphology and electrophysiological properties of serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus (DRN) of the rat were examined in frontal brain slices. Biocytin was injected intracellularly into the intracellularly recorded neurons. Then the morphology of the recorded neurons was observed after histochemical visualization of biocytin. The recorded neurons extending their main axons outside the DRN were considered as projection neurons. Subsequently, serotonergic nature of the neurons was examined by serotonin (5-HT) immunohistochemistry. The general form of the dendritic trees is radiant and poorly branching in both 5-HT- and non-5-HT neurons. However, the dendrites of the 5-HT neurons were spiny, whereas those of the non-5-HT neurons were aspiny. The main axons of both 5-HT- and non-5-HT neurons were observed to send richly branching axon collaterals to the DRN, ventrolateral part of the periaqueductal gray and the midbrain tegmentum. In response to weak, long depolarizing current pulses, the 5-HT neurons displayed a slow and regular firing activity. The non-5-HT neurons fired at higher frequencies even when stronger current was injected. Some other differences in electrophysiological properties were also observed between the 5-HT-immunoreactive spiny projection neurons and the 5-HT-immunonegative aspiny projection neurons. 2001 Elsevier Science B.V. All rights reserved. Theme: Other systems of the CNS Topic: Comparative neuroanatomy Keywords: Dorsal raphe nucleus; Serotonin; Intracellular recording; Intracellular staining; Immunohistochemistry; Rat
1. Introduction The serotonin (5-hydroxytryptamine; 5-HT) system, which has diverse projections throughout the central nervous system and exerts a tonic modulatory influence on its target, is implicated in a vast array of physiological and behavioral processes (for review see Refs. [14,17]). The
*Corresponding author. Office of the Director, Tokyo Metropolitan Institute for Neuroscience, Fuchu, Tokyo 183-0042, Japan. Tel.: 181423-253881 ext. 4417; fax: 181-423-218678. E-mail address: [email protected] (N. Mizuno).
dorsal raphe nucleus (DRN), the largest cluster of 5-HTcontaining neurons in the rat brain (for review see Ref. [17]), also sends many ascending and descending projection fibers to the brain regions ([28,29]; for further review, see Ref. [14]). The DRN of the rat, however, has been indicated to contain a substantial number of non-serotonergic neurons ([7]; for further review see Refs. [14,17]); some of these non-serotonergic DRN neurons appear to be projection neurons ([6,26]; for further review see Ref. [14]). The general morphology of rat DRN neurons, including the general form of the dendritic tree and axonal arborization, was examined by the Golgi method [5,8,16]. Intracel-
0006-8993 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 01 )02272-7
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
lular horseradish peroxidase (HRP)-labeling combined with intracellular recording also revealed the general morphology of rat DRN neurons with some particular electrophysiological properties [2,24,25]. In these studies, serotonergic nature of the recorded neurons were often presumed on the basis of their electrophysiological and pharmacological characteristics (see also Refs. [10–13]), but the neurochemical nature of the neurons was not directly examined. Electrophysiological characterization of serotonergic DRN neurons have been attempted by using either extracellular or intracellular techniques in combination with selective destruction of serotonergic neurons [4], visualization of 5-HT-containing histofluorescent neurons in the immediate vicinity of the recording electrode [1], or intracellular injection of a red fluorescing dye (ethidium bromide) in combination with the subsequent visualization of both the injected dye and yellow fluorescence of 5-HT inside the recorded neurons [3]. These studies, however, did not sufficiently reveal the general morphology of the DRN neurons examined. Thus, it is still incompletely known whether the 5-HT expression might be linked to distinct morphological features and / or electrophysiological properties of projection neurons in the DRN. The present study was done in an attempt to shed more light on this issue: electrophysiological properties of rat DRN neurons were examined intracellularly in frontal brain slices maintained in vitro, and then biocytin was injected intracellularly into the recorded neurons in order to observe the morphology of the recorded neurons by histochemical visualization of biocytin. The recorded neurons extending their axons outside the DRN were considered as projection neurons. Serotonergic nature of the neurons was examined by 5-HT immunohistochemistry.
2. Materials and methods Twenty-six adult male albino rats (Wistar; Oriental Bioservice, Kyoto, Japan) weighing between 150 and 190 g were used in the present study. All procedures of the experiments were approved by the Animal Care and Use Committees at The Fourth Military Medical University (Xi’an, People’s Republic of China) and at the Graduate School of Medicine, Kyoto University (Kyoto, Japan). After decapitation under ether anesthesia, the lower brainstems were quickly removed. The mesopontine levels of the brainstems were cut frontally into 500-mm-thick slices on a Microslicer (Dosaka EM, Kyoto, Japan). The slices were pre-incubated at 208C for 1–8 h in artificial cerebrospinal fluid that were saturated with 95% O 2 and 5% CO 2 . The artificial cerebrospinal fluid was composed of (in mM) 124 NaCl, 3.3 KCl, 26 NaHCO 3 , 1.3 KH 2 PO 4 , 2.5 CaCl 2 , 1 MgSO 4 , and 10 D-glucose. The slices were then
111
transferred to a chamber of interface type (BSC-Haas Top; Medical Systems, Greenvale, NY), where the temperature was maintained at 34–358C. Glass micropipettes were prepared with a puller (P-87; Sutter, Novato, CA) and were filled with 3% biocytin (Sigma, St. Louis, MO) which was dissolved in 50 mM Tris–HCl (pH 7.4) containing 2 M potassium methylsulfate. The glass micropipettes were directed toward the DRN just beneath and within the ventral part of the central gray in the caudal midbrain levels and the upper pontine levels. The electrode resistance was 100–200 MV for intracellular recording and labeling. The input signal was fed into a high-input impedance DC amplifier with an active bridge circuit (IR-283; NeuroData, New York, NY) and stored in a computer through an analog–digital converter (MacLab; AD Instruments, Castle Hill, Australia). Biocytin was injected with depolarizing current pulses of 200 ms duration at 1 Hz for 15–30 min. For a particular cell recorded intracellularly, the electrical membrane property, such as adaptation of spike frequency, depolarizing sag, inward rectification, rebound hump, fast depolarization, or slow depolarization (Table 1), was considered to be present only when more than three of the four authors were of one accord in admitting the presence of the property. At the end of recording, the slices were placed in 0.1 M phosphate buffer (pH 7.3) containing 4% formaldehyde and 0.05% glutaraldehyde for 20 h at room temperature. After cryoprotection with 30% sucrose in 10 mM phosphate-buffered 0.85% saline (PBS; pH 7.3), the slices were further cut serially into 30-mm-thick sections on a freezing microtome; each of the sections were placed serially in a well containing PBS. In the procedures described below, the serial sections, each of which were placed separately in a well, were incubated at room temperature; after each incubation, they were rinsed with PBS containing 0.3% Triton X-100 and 0.02% sodium methiolate (PBS-X). The sections were first placed in PBS containing 2% H 2 O 2 for 30 min to suppress endogenous peroxidase activity, and then incubated with PBS-X containing 10% normal donkey serum for 30 min, and further incubated for 18–20 h with PBS-X containing 5 mg / ml aminomethylcoumarin acetic acid (AMCA)conjugated avidin D (Vector Laboratories, Burlingame, CA), 1 mg / ml rabbit anti-5-HT immunoglobulin G (IgG) antibody (Incstar, Stillwater, MN), and avidin–biotinylated peroxidase complex (ABC Elite; Vector), and finally incubated for 4–6 h with PBS-X containing 20 mg / ml biotinylated horseradish peroxidase (Vector) and 10 mg / ml dichlorotriazinyl aminofluorescein (DTAF)-conjugated donkey anti-rabbit IgG (Chemicon, Temecula, CA). The sections were mounted onto clean glass slides, coverslipped with 0.05 M PBS containing 50% (v / v) glycerin and 2.5% (w / v) triethylene diamine (anti-fading agent), and then observed with an epifluorescence microscope (Axiophot; Zeiss, Oberkochen, Germany) under appropriate
112
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
filters for AMCA (excitation 359–371 nm; emission 397– 490 nm) and DTAF (excitation 450–490 nm; emission 515–565 nm) to find out neuronal cell bodies which were labeled with biocytin injected intracellularly and to examine whether these neurons might exhibit immunoreactivity for 5-HT. After the epifluorescence microscopy for AMCA and DTAF, the sections were detached from the glass slides and washed with PBS. Subsequently, the sections were placed in 50 mM Tris–HCl (pH 7.6) containing 0.02% diaminobenzidine–4HCl (Dojin, Kumamoto, Japan), 10 mM nickel ammonium sulfate, and 0.0001% H 2 O 2 for 30–60 min at room temperature to visualize peroxidase bound to biocytin. The sections were then mounted on a gelatin-coated glass slide, air-dried, counterstained lightly with 3% neutral red, dehydrated in ethanol series, cleared in xylene, and coverslipped. The biocytin-labeled neurons in the DRN were reconstructed onto a two-dimensional plane by camera lucida with the aid of a drawing tube attached to a light microscope.
3. Results A total of 19 neurons from 26 rats were recorded stably for 30–60 min and then injected with biocytin intracellularly by using depolarizing current pulses. The cell bodies of the encountered neurons were located 120–250 mm deep from the surface of the 500-mm-thick frontal slices of the caudal midbrain and the rostral pons. Camera lucida drawing of these neurons revealed that 15 of the 19 neurons recorded intracellularly had their cell bodies within the DRN. All of these DRN neurons extended their axons outside the DRN; seven of them displayed 5-HT immunoreactivity (Fig. 1a,a9),whereas the remaining eight showed no 5-HT immunoreactivity (Fig. 1b,b9).
3.1. Morphological features The 5-HT-immunopositive projection neurons in the DRN had fusiform, triangular or multipolar cell bodies, the major diameter of which ranged from 15 to 35 mm. The
Fig. 1. Fluorescence photomicrographs of 5-HT-immunopositive and 5-HT-immunonegative projection neurons in the DRN which were intracellularly labeled with biocytin. The field of (a) is identical to that of (a9), while the field of (b) is identical to that of (b9). The fields (a) and (b) are taken with an appropriate filter for AMCA (for biocytin labeling), whereas the fields (a9) and (b9) are taken with an appropriate filter for DTAF (for 5-HT immunoreactivity). Arrows in (a) and (b) indicate biocytin-labeled neurons, while arrowheads in (a9) and (b9) indicate 5-HT-immunopositive neurons. The biocytin-labeled neuron in (a) is 5-HT-immunopositive in (a9), whereas the biocytin-labeled neuron in (b) shows no 5-HT immunoreactivity in (b9). Scale bar540 mm.
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
dendrites displayed a poorly branched radiating pattern of arborization. Two to three primary dendrites arose from fusiform or triangular cell bodies (Figs. 2a9 and 3a), while 4–7 primary dendrites originated from multipolar cell bodies. The primary dendrites gave rise to secondary branches after running a short distance (30–100 mm). The dendritic spines were rather sparse on the primary and the secondary dendrites, but were progressively more dense on the higher order dendrites (Figs. 2a9 and 3b). The dendritic fields were oval or triangular in shape and had the major diameter ranging from 200 to 750 mm. The majority of them remained within the DRN, although a few dendrites of the multipolar neurons were observed to extend into the ventrolateral division of the periaqueductal gray. The main axons arise from the cell bodies or the primary dendrites. Axonal branches with prominent varicosities were distributed widely in the DRN and also in the periaqueductal gray, especially in its ventrolateral division (Figs. 2a0 and 3c). Many axonal branches were further traced into the midbrain reticular formation; some of them appeared to terminate on the cell bodies of the mesencephalic trigeminal nucleus neurons (Figs. 2a0 and 3d). Axonal branches often showed recurrent courses (Fig. 2a0). The target of the main axons could not be determined in the present study. The somatodendritic domain of the 5-HT-immunonegative projection neurons in the DRN had essentially the same general morphology features with those of the 5-HTimmunopositive projection neurons in the DRN, except that the dendritic spines were rarely seen in the 5-HTnegative neurons (Compare Figs. 4 and 5 with Figs. 2 and 3). Many axonal branches of the 5-HT-immunonegative projection neurons of the DRN were also observed to be distributed in the ventrolateral part of the periaqueductal gray and the midbrain reticular formation as well as in the DRN, although no targets of the main axons of these neurons could be confirmed (Fig. 4). No presumed axon terminals of the 5-HT-immunonegative projection neurons of the DRN were found on the neuronal cell bodies in the mesencephalic trigeminal nucleus. The main axons could not be traced to their target.
3.2. Electrophysiological properties The data obtained from morphologically defined 15 projection neurons in the DRN (seven 5-HT-immunopositive and eight 5-HT-immunonegative projection neurons) were summarized in Table 1. All of these neurons had been examined electrophysiologically by injecting depolarizing and hyperpolarizing current pulses of 120 ms and 500 ms; the intracellular penetration stably lasted for 30–60 min. After the intracellular recording study, the recorded neurons were intracellularly labeled successfully with the biocytin method and confirmed to be projection neurons sending their main axons outside the DRN.
113
Fig. 2. Camera lucida reconstruction of a single exemplary 5-HT-immunopositive projection neuron, the morphology of which was visualized by the intracellular biocytin injection. The location, the features of the somatodendritic domains, and the arborization of the main axons are shown in (a), (a9), and (a0), respectively. Arrows point to the cell body of the neuron. Open arrow indicates the main axon. Dotted lines indicate the border of the DRN or the periaqueductal gray (PAG). Aq, mesencephalic aqueduct; Vme, mesencephalic trigeminal nucleus neuron.
Some electrical membrane properties were different statistically between the 5-HT-immunopositive and the 5-HT-immunonegative projection neurons: the membrane time constant was statistically larger in the 5-HT neurons than in the non-5-HT neurons, whereas no statistical differences were detected in the resting membrane potential, the input resistance, the first time constant, or the electrotonic length (Table 1). Weak depolarizing current injection elicited action potential, the characteristics of which were also different statistically between the 5-HT- and the non-5-HT neurons: The spike height and the spike width were statistically larger in the 5-HT neurons than in the non-5-HT neurons, whereas no statistical difference was found in the threshold for spikes (Table 1 and Fig. 6). In response to stronger depolarizing current pulses, the neurons responded with a train of action potentials (Fig. 6). In response to weak, long depolarizing current pulses, the 5-HT neurons displayed a slow and regular firing activity (662 Hz). On the other hand, the non-5-HT neurons fired at higher frequencies (2068 Hz) even when stronger current was injected. Analysis of interspike intervals revealed some degree of adaptation of spike frequency in one of the seven 5-HT neurons and in seven of the eight non-5-HT neurons (Table 1); in Fig. 7, adaptation of spike frequency is not so
114
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
Fig. 3. Photomicrographs of the 5-HT-immunopositive projection neuron that is depicted in Fig. 2. The neuron in (a) emits three primary dendrites. A part of a dendrite, which is demarcated by a rectangle in (a), is enlarged in (b), where the dendritic spines are pointed by arrowheads. Arrows in (c) indicate varicosities on an axonal branch. In (d), some of the axonal branches with presumed axon terminals of this 5-HT-immunopositive neuron are seen to constitute the pericellular basket around a cell body (asterisk) of a mesencephalic trigeminal nucleus neuron (Compare with Fig. 1a9). Abbreviations are as in Fig. 1.
evident in the 5-HT neuron (Fig. 7a) as in the non-5-HT neuron (Fig. 7b). When hyperpolarizing current pulses were injected, time-dependent depolarization (depolarizing sag), which was indicative of the well-known hyperpolarization-activated cationic current (h current), was seen in six of the eight non-5-HT neurons (b 1 in Fig. 6), but not in the 5-HT neurons. The overall inward rectification, which was measured at the end of 500-ms hyperpolarizing current pulses (arrows in a 1 ,b 1 of Fig. 6), was observed in four of the eight non-5-HT neurons, but not in the 5-HT neurons. At the offset of the hyperpolarizing current pulses, rebound
depolarization hump were observed in six of the non-5-HT neurons; these rebound humps were not accompanied with spikes (b 1 in Fig. 6). No rebound depolarizing hump were observed in the 5-HT neurons. When weak long depolarizing current pulses were injected, the 5-HT neurons displayed a slow regular discharge pattern with gradual interspike depolarization and large fast afterhyperpolarization (a 1 ,a 2 in Fig. 6). On the other hand, action potentials induced in the non-5-HT neurons were often followed by a small and rapid depolarization which led into a hyperpolarization (b 1 in Fig. 6). The same neurons further exhibited long-lasting depolar-
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
115
Fig. 4. Camera lucida reconstruction of a single exemplary 5-HT-immunonegative projection neuron, the morphology of which was visualized by the intracellular biocytin injection (Fig. 1b,b9). The location, the features of the somatodendritic domains, and the arborization of the main axons are shown in (a), (a9), and (a0), respectively, as in Fig. 2.
Table 1 Electrical membrane properties of the serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus a Property
Serotonergic neurons (n57)
Non-serotonergic neurons (n58)
Resting membrane potential (mV) Input resistance (MV)c Membrane time constant (t0 , ms)d First time constant (t1 , ms)d Electrotonic length e Threshold for spikes (mV) Spike height (mV) Spike width at base (ms) Spike width at half height (ms) Adaptation of spike frequency Depolarizing sag Inward rectification b Rebound hump Fast afterdepolarization Slow afterdepolarization
68613 b 174623 7.461.4 1.361.2 2.060.7 275611 126617 3.661.2 1.260.4 1f 0 0 0 0 0
70610 158620 5.561.3* 1.160.7 1.860.9 27068 104616* 0.960.3** 0.660.2** 7** 6** 4** 6** 5** 5**
a
Statistical significance was evaluated by using the two-tailed, unpaired U-test or Fisher’s exact probability test. *P,0.05 compared with nonserotonergic projection neurons in the DRN. **P,0.01 compared with non-serotonergic projection neurons in the DRN. b Mean6S.D. c Input resistance was determined by hyperpolarizing current pulses, producing voltage shifts of 6–15 mV negative to the rest. d Time constants were determined by injecting a short (2 ms) depolarizing current pulse. e The electrotonic length of an equivalent sealed-end cylinder was ]] estimated by the expression pœt0 /t1 2 1. f Numbers of cases. g Determined at the end of 500-ms current injection (see Fig. 6, arrows).
Fig. 5. Photomicrographs of the 5-HT-immunonegative projection neuron that is depicted in Fig. 4 (also see Fig. 1b,b9). The initial part of the main axon (AX) and three primary dendrites are seen. A part of a dendrite (DEN), which is demarcated by a rectangle in (a), is enlarged in (b); no spines are seen on the dendrites. Arrows in (c) and (d) indicate varicosities on axonal branches within the DRN (c) and the periaqueductal gray (d). Scale bars: a540 mm; b58 mm; c,d510 mm.
116
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
Fig. 6. Intracellular recordings of the 5-HT- (a 1 ,a 2 ) and non-5-HT neurons (b 1 ,b 2 ) in DRN. The morphology of the 5-HT neuron is shown in Figs. 2 and 3, while that of the non-5-HT neuron is shown in Figs. 4 and 5. Long (500 ms) depolarizing and hyperpolarizing current pulses were injected into each neuron. Arrows indicate the points where the input resistance and overall inward rectification were determined.
ization after the offset of depolarizing current pulses (b 1 in Fig. 6). Such small rapid depolarization has been reported as the depolarizing afterpotential (DAP) [30] or the fast afterdepolarization (fADP) [9]. The same neurons further exhibited long-lasting depolarization after the offset of depolarizing current pulses (b 1 in Fig. 6). Such slow return of membrane potential after depolarization has been reported as the slow afterdepolarization potential (sADP) [15,27].
4. Discussion The morphology and electrophysiological properties of 5-HT-immunopositive and 5-HT-immunonegative projection neurons in the rat DRN were investigated in the present study: Intracellular recording and then intracellular labeling with the biocytin were performed in frontal brain slices of 500 mm thickness. The biocytin-labeled neurons were visualized in the serial sections of the brain slices. After observation of the morphology of the biocytinlabeled neurons, 5-HT immunoreactivity of the neurons were examined immunohistochemically. It was possible that the present observation by twodimensional camera lucida drawings might fail to detect
some features of neuronal geometry in the third dimension. However, the axons and dendrites of the sampled neurons were visualized clearly to the peripheral portions with axonal and dendritic branches with very fine diameters. Both 5-HT-positive and 5-HT-negative neurons were considered as projection neurons when their main axons were traced to the tegmental regions of the midbrain outside the DRN, although it was impossible to trace the whole extent of the courses of the main axons. The previous study indicated that axon terminals of serotonergic DRN neurons were in symptomatic synaptic contacts upon mesencephalic trigeminal nucleus neurons [21]. This was also indicated in the present study (Figs. 2a0 and 3d). The DRN of the rat has been indicated to contain a substantial number of non-serotonergic cells ([7]; for further review see Refs. [14,17]). The neurochemical nature of 5-HT-immunonegative neurons observed in the present study remains to be determined. It could also be emphasized that one of the problems inherent to the immunohistochemical technique is the interpretation of false-negative cells. In the present study, however, the dendrites of all 5-HT-immunopositive projection neurons sampled were spiny, whereas those of all 5-HT-immunonegative projection neurons examined were aspiny; somatic spines, which have been reported in some DRN
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
Fig. 7. Spike frequency adaptation of a 5-HT- (a) and a non-5-HT neuron (b) in response to current injection of a long depolarizing current pulse. The morphology and electrophysiological properties of the 5-HT neuron are shown in Figs. 2, 3 and 6a 1 ,a 2 , while those of the non-5-HT neuron are shown in Figs. 4, 5 and 6b 1 ,b 2 . Interspike intervals are plotted against the time after the onset of 500-ms depolarizing current pulses. Adaptation of spike frequency (slope) is not so evident in the 5-HT neuron (a) than in the non-5-HT neuron (b).
neurons in the previous studies [2,5,24,25], were not found in the present study. No other particular differences were detected in the general morphology between the 5-HTimmunopositive and the 5-HT-immunonegative projection neurons in the present study. Both the 5-HT-immunopositive and the 5-HT-immunonegative projection neurons observed in the present study were observed to send many axonal branches within the DRN. These neurons in the DRN were considered to play roles not only as projection neurons but also as intrinsic neurons. The intracellular recordings in the present study were performed in the bridge-balance mode with high-impedance recording electrodes (100–200 MV). This may have led to inaccurate measurements due to shortcomings inherent in the technique. During the present experiments, the diode characteristics of the electrodes were tested by injecting 500-ms depolarizing or hyperpolarizing current pulses with the bridge-balanced mode turned off (data not shown). When the injected depolarizing or hyperpolarizing current pulses ranged from 10.5 nA to 20.5 nA, the difference from the linear current–voltage relationship was less than 5%. The input resistance and the measured time
117
constants of the sampled neurons in the present study apparently might be inaccurate (smaller) due to shunt current around the sharp electrodes. It was considered, however, that the comparison of electrical membrane properties between the 5-HT-immunopositive and the 5HT-negative projection neurons could be made on the basis of the measurements obtained in the present study. Some differences were observed in the electrophysiological properties between the 5-HT-immunopositive and the 5-HT-immunonegative projection neurons sampled in the present study (Table 1; Figs. 6 and 7). Serotonergic neurons in the rat DRN have been considered to exhibit the slow rate of spontaneous firing with a regular rhythm ([2–4,26]; for further review see Refs. [17,18]). The 5-HTimmunopositive neurons sampled in the present study, also showed a slow rhythmical firing when depolarizing current pulses were injected. The gradual interspike depolarization which was associated with the slow regular discharge pattern of the 5-HT-immunopositive neurons observed in the present study (a 1 ,a 2 in Fig. 6) has also been observed in the previous studies [2,3]. On the other hand, some neurons in the rat DRN have been reported to fire spikes frequently in short bursts [10–13]. A high proportion of serotonergic mesopontine neurons in microculture have also been reported to discharge in bursts in response to a single intracellular depolarizing current pulse [19]. Such burst-firing neurons, however, were not detected in the 5-HT-immunopositive projection neurons in the present study. On the other hand, non-serotonergic neurons with various electrophysiological properties have been reported in the rat DRN. This heterogeneous population of neurons showed a less regular or irregular firing pattern, with a firing rate ranging from very slow to fast activity (0.1–30 Hz) [2,3,13,24,26]. In the present study, the 5-HT-immuonegative neurons often showed the depolarizing afterpotential (DAP) [30] or the fast afterdepolarization (fADP) [9], as well as the slow afterdepolarizing potential (sADP) [15,27], when depolarizing current pulses were injected. The fADP or DAP, which were often observed after an action potential in the pyramidal neurons of layer VI of the primary motor and somatosensory cortices and in layers V and VI of the prefrontal cortex in the rat, has been suggested to be an important determinant of the firing pattern of the pyramidal neurons in the cerebral cortex [9,20,30]. On the other hand, the sADP, which was reported in the pyramidal neurons of layer V in the cat sensorimotor cortex [27] and in neurons of the rat dorsolateral septal nucleus [15], has been indicated to be associated with an increased membrane conductance and to trigger action potentials; this could induce or potentiate the bursting activity of neurons in the absence of further stimulation [15,27]. Such sADP was also observed in the intrinsic neurons of the rat medullary dorsal horn [22,23]. In summary, the present study indicated that projection neurons in the rat DRN were divided into the serotonergic
118
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
spiny and the non-serotonergic aspiny neurons. Some differences in electrophysiological properties were also revealed between these two populations of projection neurons. Both of these projection neurons were considered to be involved in the local circuit mechanism within the DRN through their richly branched axon collaterals.
Acknowledgements The authors thank Mr Akira Uesugi and Miss Keiko Okamoto of Kyoto University, Ms Yue-Pin Yuan in The Fourth Military Medical University, and Mr Nobuyuki Kobayashi and Mr Hideki Itabashi in Tokyo Metropolitan Institute for Neuroscience for their photographic help. This work was supported in part Grants-in-Aid from the National Natural Science Foundation of China (39625011, 39970239) and from the Ministry of Education, Science, Sports and Culture of Japan (12680743).
References [1] G.K. Aghajanian, H.J. Haigler, L-tryptophan as a selective histochemical marker for serotonergic neurons in single-cell recording studies, Brain Res. 81 (1974) 364–372. [2] G.K. Aghajanian, C.P. VanderMaelen, Intracellular recordings from serotonergic dorsal raphe neurons: pacemaker potentials and the effect of LSD, Brain Res. 238 (1982) 463–469. [3] G.K. Aghajanian, C.P. VanderMaelen, Intracellular identification of central noradrenergic and serotonergic neurons by a new double labeling procedure, J. Neurosci. 2 (1982) 1786–1792. [4] G.K. Aghajanian, R.Y. Wang, J.B. Baraban, Serotonergic and nonserotonergic neurons of the dorsal raphe: reciprocal changes in firing induced by peripheral nerve stimulation, Brain Res. 153 (1978) 169–175. [5] H. Danner, C. Pfister, Untersuchungen und Zytoarchitektonik des Nucleus raphe dorsalis der Ratte, J. Hirnforsch. 21 (1980) 655–664. [6] F. Datiche, P.-H. Luppi, M. Cattarelli, Serotonergic and nonserotonergic projections from the raphe nuclei to the piriform cortex in the rat: a cholera toxin B subunit (CTb) and 5-HT immunohistochemical study, Brain Res. 671 (1995) 27–37. [7] L. Descarries, K.C. Watkins, S. Garcia, A. Beaudet, The serotonin neurons in nucleus raphe dorsalis of adult rat: a light and electron microscope radioautographic study, J. Comp. Neurol. 207 (1982) 239–254. [8] S. Diaz-Cintra, L. Cintra, T. Kemper, O. Resnick, P.J. Morgane, Nucleus raphe dorsalis: a morphometric Golgi study in rats of three age groups, Brain Res. 207 (1981) 1–16. [9] S. Haj-Dahmane, R. Andrade, Calcium-activated cation nonselective current contributes to the fast afterdepolarization in rat prefrontal cortex neurons, J. Neurophysiol. 78 (1997) 1983–1989. ´ S.E. Gartside, A.E.P. Villa, T. Sharp, Evidence for a [10] M. Hajos, repetitive (burst) firing pattern in a sub-population of 5-hydroxytryptamine neurons in the dorsal and median raphe nuclei of the rat, Neuroscience 69 (1995) 189–197. ´ T. Sharp, Burst-firing activity of presumed 5-HT neurones [11] M. Hajos, of the rat dorsal raphe nucleus: electrophysiological analysis by antidromic stimulation, Brain Res. 740 (1996) 162–168.
´ T. Sharp, A 5-hydroxytryptamine lesion markedly reduces [12] M. Hajos, the incidence of burst-firing dorsal raphe neurones in the rat, Neurosci. Lett. 204 (1996) 161–164. ´ T. Sharp, N.R. Newberry, Intracellular recordings from [13] M. Hajos, burst-firing presumed serotonergic neurones in the rat dorsal raphe nucleus in vivo, Brain Res. 737 (1996) 308–312. [14] G. Halliday, A. Harding, G. Paxinos, Serotonin and tachykinin systems, in: G. Paxinos (Ed.), The Rat Nervous System, Academic Press, San Diego, CA, 1995, pp. 929–974. [15] H. Hasuo, K.D. Phelan, M.J. Twery, J.P. Gallagher, A calciumdependent slow afterdepolarization recorded in rat dorsolateral septal nucleus neurons in vitro, J. Neurophysiol. 64 (1990) 1838– 1846. [16] B. Hoelzel, C. Pfister, Morphometrischer Vergleich der Neurontypen des Nucleus raphe dorsalis und des Nucleus centralis superior (Nc. raphe medianus) der Ratte, Z. Mikroskop.-Anat. Forsch. 97 (1983) 529–534. [17] B.L. Jacobs, E.C. Azmitia, Structure and function of the brain serotonin system, Physiol. Rev. 72 (1992) 165–229. [18] B.L. Jacobs, C.A. Fornal, Activity of brain serotonergic neurons in the behaving animal, Pharmacol. Rev. 43 (1991) 563–578. [19] M.D. Johnson, Synaptic glutamate release by postnatal rat serotonergic neurons in microculture, Neuron 12 (1994) 433–442. [20] T. Kaneko, Y. Kang, N. Mizuno, Glutaminase-positive and glutaminase-negative pyramidal cells in layer VI of the primary motor and somatosensory cortices: a combined analysis by intracellular staining and immunocytochemistry in the rat, J. Neurosci. 15 (1995) 8362–8377. [21] J.-L. Li, K.-H. Xiong, Y.-Q. Li, T. Kaneko, N. Mizuno, Serotonergic innervation of mesencephalic trigeminal nucleus neurons: a light and electron microscopic study in the rat, Neurosci. Res. 37 (2000) 127–140. [22] Y.-Q. Li, H. LI, T. Kaneko, N. Mizuno, Substantia gelatinosa neurons in the medullary dorsal horn: An intracellular labeling study in the rat, J. Comp. Neurol. 411 (1999) 399–412. [23] Y.-Q. Li, H. Li, K. Yang, Z.-M. Wang, T. Kaneko, N. Mizuno, Intracellular labeling study of neurons in the superficial part of the magnocellular layer of the medullary dorsal horn of the rat, J. Comp. Neurol. 428 (2001) 641–655. [24] M.R. Park, Intracellular horseradish peroxidase labeling of rapidly firing dorsal raphe projection neurons, Brain Res. 402 (1987) 117– 130. [25] M.R. Park, H. Imai, S.T. Kitai, Morphology and intracellular responses of an identified dorsal raphe projection neuron, Brain Res. 240 (1982) 321–326. [26] S. Sawyer, J.M. Tepper, S.J. Young, P.M. Groves, Antidromic activation of dorsal raphe neurons from neostriatum: physiological characterization and effects of terminal autoreceptor activation, Brain Res. 332 (1985) 15–28. [27] P.C. Schwindt, W.J. Spain, R.C. Foehring, M.C. Chubb, W.E. Crill, Slow conductances in neurons from cat sensorimotor cortex in vitro and their role in slow excitability changes, J. Neurophysiol. 59 (1988) 450–467. [28] R.P. Vertes, A PHA-L analysis of ascending projections of the dorsal raphe nucleus in the rat, J. Comp. Neurol. 313 (1991) 643–668. [29] R.P. Vertes, B. Kocsis, Projections of the dorsal raphe nucleus to the brainstem: PHA-L analysis in the rat, J. Comp. Neurol. 340 (1994) 11–26. [30] C.R. Yang, J.K. Seamans, N. Gorelova, Electrophysiological and morphological properties of layers V–VI principal pyramidal cells in rat prefrontal cortex in vitro, J. Neurosci. 16 (1996) 1904–1921.
Research report
Morphological features and electrophysiological properties of serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus An intracellular recording and labeling study in rat brain slices a a b c, Yun-Qing Li , Hui Li , Takeshi Kaneko , Noboru Mizuno * a
Department of Anatomy and K.K. Leung Brain Research Centre, The Fourth Military Medical University, Xi’ an 710032, People’ s Republic of China b Department of Morphological Brain Science, Graduate School of Medicine, Kyoto University, Kyoto 606 -8501, Japan c Tokyo Metropolitan Institute for Neuroscience, Fuchu, Tokyo 183 -8526, Japan Accepted 6 February 2001
Abstract The morphology and electrophysiological properties of serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus (DRN) of the rat were examined in frontal brain slices. Biocytin was injected intracellularly into the intracellularly recorded neurons. Then the morphology of the recorded neurons was observed after histochemical visualization of biocytin. The recorded neurons extending their main axons outside the DRN were considered as projection neurons. Subsequently, serotonergic nature of the neurons was examined by serotonin (5-HT) immunohistochemistry. The general form of the dendritic trees is radiant and poorly branching in both 5-HT- and non-5-HT neurons. However, the dendrites of the 5-HT neurons were spiny, whereas those of the non-5-HT neurons were aspiny. The main axons of both 5-HT- and non-5-HT neurons were observed to send richly branching axon collaterals to the DRN, ventrolateral part of the periaqueductal gray and the midbrain tegmentum. In response to weak, long depolarizing current pulses, the 5-HT neurons displayed a slow and regular firing activity. The non-5-HT neurons fired at higher frequencies even when stronger current was injected. Some other differences in electrophysiological properties were also observed between the 5-HT-immunoreactive spiny projection neurons and the 5-HT-immunonegative aspiny projection neurons. 2001 Elsevier Science B.V. All rights reserved. Theme: Other systems of the CNS Topic: Comparative neuroanatomy Keywords: Dorsal raphe nucleus; Serotonin; Intracellular recording; Intracellular staining; Immunohistochemistry; Rat
1. Introduction The serotonin (5-hydroxytryptamine; 5-HT) system, which has diverse projections throughout the central nervous system and exerts a tonic modulatory influence on its target, is implicated in a vast array of physiological and behavioral processes (for review see Refs. [14,17]). The
*Corresponding author. Office of the Director, Tokyo Metropolitan Institute for Neuroscience, Fuchu, Tokyo 183-0042, Japan. Tel.: 181423-253881 ext. 4417; fax: 181-423-218678. E-mail address: [email protected] (N. Mizuno).
dorsal raphe nucleus (DRN), the largest cluster of 5-HTcontaining neurons in the rat brain (for review see Ref. [17]), also sends many ascending and descending projection fibers to the brain regions ([28,29]; for further review, see Ref. [14]). The DRN of the rat, however, has been indicated to contain a substantial number of non-serotonergic neurons ([7]; for further review see Refs. [14,17]); some of these non-serotonergic DRN neurons appear to be projection neurons ([6,26]; for further review see Ref. [14]). The general morphology of rat DRN neurons, including the general form of the dendritic tree and axonal arborization, was examined by the Golgi method [5,8,16]. Intracel-
0006-8993 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 01 )02272-7
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
lular horseradish peroxidase (HRP)-labeling combined with intracellular recording also revealed the general morphology of rat DRN neurons with some particular electrophysiological properties [2,24,25]. In these studies, serotonergic nature of the recorded neurons were often presumed on the basis of their electrophysiological and pharmacological characteristics (see also Refs. [10–13]), but the neurochemical nature of the neurons was not directly examined. Electrophysiological characterization of serotonergic DRN neurons have been attempted by using either extracellular or intracellular techniques in combination with selective destruction of serotonergic neurons [4], visualization of 5-HT-containing histofluorescent neurons in the immediate vicinity of the recording electrode [1], or intracellular injection of a red fluorescing dye (ethidium bromide) in combination with the subsequent visualization of both the injected dye and yellow fluorescence of 5-HT inside the recorded neurons [3]. These studies, however, did not sufficiently reveal the general morphology of the DRN neurons examined. Thus, it is still incompletely known whether the 5-HT expression might be linked to distinct morphological features and / or electrophysiological properties of projection neurons in the DRN. The present study was done in an attempt to shed more light on this issue: electrophysiological properties of rat DRN neurons were examined intracellularly in frontal brain slices maintained in vitro, and then biocytin was injected intracellularly into the recorded neurons in order to observe the morphology of the recorded neurons by histochemical visualization of biocytin. The recorded neurons extending their axons outside the DRN were considered as projection neurons. Serotonergic nature of the neurons was examined by 5-HT immunohistochemistry.
2. Materials and methods Twenty-six adult male albino rats (Wistar; Oriental Bioservice, Kyoto, Japan) weighing between 150 and 190 g were used in the present study. All procedures of the experiments were approved by the Animal Care and Use Committees at The Fourth Military Medical University (Xi’an, People’s Republic of China) and at the Graduate School of Medicine, Kyoto University (Kyoto, Japan). After decapitation under ether anesthesia, the lower brainstems were quickly removed. The mesopontine levels of the brainstems were cut frontally into 500-mm-thick slices on a Microslicer (Dosaka EM, Kyoto, Japan). The slices were pre-incubated at 208C for 1–8 h in artificial cerebrospinal fluid that were saturated with 95% O 2 and 5% CO 2 . The artificial cerebrospinal fluid was composed of (in mM) 124 NaCl, 3.3 KCl, 26 NaHCO 3 , 1.3 KH 2 PO 4 , 2.5 CaCl 2 , 1 MgSO 4 , and 10 D-glucose. The slices were then
111
transferred to a chamber of interface type (BSC-Haas Top; Medical Systems, Greenvale, NY), where the temperature was maintained at 34–358C. Glass micropipettes were prepared with a puller (P-87; Sutter, Novato, CA) and were filled with 3% biocytin (Sigma, St. Louis, MO) which was dissolved in 50 mM Tris–HCl (pH 7.4) containing 2 M potassium methylsulfate. The glass micropipettes were directed toward the DRN just beneath and within the ventral part of the central gray in the caudal midbrain levels and the upper pontine levels. The electrode resistance was 100–200 MV for intracellular recording and labeling. The input signal was fed into a high-input impedance DC amplifier with an active bridge circuit (IR-283; NeuroData, New York, NY) and stored in a computer through an analog–digital converter (MacLab; AD Instruments, Castle Hill, Australia). Biocytin was injected with depolarizing current pulses of 200 ms duration at 1 Hz for 15–30 min. For a particular cell recorded intracellularly, the electrical membrane property, such as adaptation of spike frequency, depolarizing sag, inward rectification, rebound hump, fast depolarization, or slow depolarization (Table 1), was considered to be present only when more than three of the four authors were of one accord in admitting the presence of the property. At the end of recording, the slices were placed in 0.1 M phosphate buffer (pH 7.3) containing 4% formaldehyde and 0.05% glutaraldehyde for 20 h at room temperature. After cryoprotection with 30% sucrose in 10 mM phosphate-buffered 0.85% saline (PBS; pH 7.3), the slices were further cut serially into 30-mm-thick sections on a freezing microtome; each of the sections were placed serially in a well containing PBS. In the procedures described below, the serial sections, each of which were placed separately in a well, were incubated at room temperature; after each incubation, they were rinsed with PBS containing 0.3% Triton X-100 and 0.02% sodium methiolate (PBS-X). The sections were first placed in PBS containing 2% H 2 O 2 for 30 min to suppress endogenous peroxidase activity, and then incubated with PBS-X containing 10% normal donkey serum for 30 min, and further incubated for 18–20 h with PBS-X containing 5 mg / ml aminomethylcoumarin acetic acid (AMCA)conjugated avidin D (Vector Laboratories, Burlingame, CA), 1 mg / ml rabbit anti-5-HT immunoglobulin G (IgG) antibody (Incstar, Stillwater, MN), and avidin–biotinylated peroxidase complex (ABC Elite; Vector), and finally incubated for 4–6 h with PBS-X containing 20 mg / ml biotinylated horseradish peroxidase (Vector) and 10 mg / ml dichlorotriazinyl aminofluorescein (DTAF)-conjugated donkey anti-rabbit IgG (Chemicon, Temecula, CA). The sections were mounted onto clean glass slides, coverslipped with 0.05 M PBS containing 50% (v / v) glycerin and 2.5% (w / v) triethylene diamine (anti-fading agent), and then observed with an epifluorescence microscope (Axiophot; Zeiss, Oberkochen, Germany) under appropriate
112
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
filters for AMCA (excitation 359–371 nm; emission 397– 490 nm) and DTAF (excitation 450–490 nm; emission 515–565 nm) to find out neuronal cell bodies which were labeled with biocytin injected intracellularly and to examine whether these neurons might exhibit immunoreactivity for 5-HT. After the epifluorescence microscopy for AMCA and DTAF, the sections were detached from the glass slides and washed with PBS. Subsequently, the sections were placed in 50 mM Tris–HCl (pH 7.6) containing 0.02% diaminobenzidine–4HCl (Dojin, Kumamoto, Japan), 10 mM nickel ammonium sulfate, and 0.0001% H 2 O 2 for 30–60 min at room temperature to visualize peroxidase bound to biocytin. The sections were then mounted on a gelatin-coated glass slide, air-dried, counterstained lightly with 3% neutral red, dehydrated in ethanol series, cleared in xylene, and coverslipped. The biocytin-labeled neurons in the DRN were reconstructed onto a two-dimensional plane by camera lucida with the aid of a drawing tube attached to a light microscope.
3. Results A total of 19 neurons from 26 rats were recorded stably for 30–60 min and then injected with biocytin intracellularly by using depolarizing current pulses. The cell bodies of the encountered neurons were located 120–250 mm deep from the surface of the 500-mm-thick frontal slices of the caudal midbrain and the rostral pons. Camera lucida drawing of these neurons revealed that 15 of the 19 neurons recorded intracellularly had their cell bodies within the DRN. All of these DRN neurons extended their axons outside the DRN; seven of them displayed 5-HT immunoreactivity (Fig. 1a,a9),whereas the remaining eight showed no 5-HT immunoreactivity (Fig. 1b,b9).
3.1. Morphological features The 5-HT-immunopositive projection neurons in the DRN had fusiform, triangular or multipolar cell bodies, the major diameter of which ranged from 15 to 35 mm. The
Fig. 1. Fluorescence photomicrographs of 5-HT-immunopositive and 5-HT-immunonegative projection neurons in the DRN which were intracellularly labeled with biocytin. The field of (a) is identical to that of (a9), while the field of (b) is identical to that of (b9). The fields (a) and (b) are taken with an appropriate filter for AMCA (for biocytin labeling), whereas the fields (a9) and (b9) are taken with an appropriate filter for DTAF (for 5-HT immunoreactivity). Arrows in (a) and (b) indicate biocytin-labeled neurons, while arrowheads in (a9) and (b9) indicate 5-HT-immunopositive neurons. The biocytin-labeled neuron in (a) is 5-HT-immunopositive in (a9), whereas the biocytin-labeled neuron in (b) shows no 5-HT immunoreactivity in (b9). Scale bar540 mm.
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
dendrites displayed a poorly branched radiating pattern of arborization. Two to three primary dendrites arose from fusiform or triangular cell bodies (Figs. 2a9 and 3a), while 4–7 primary dendrites originated from multipolar cell bodies. The primary dendrites gave rise to secondary branches after running a short distance (30–100 mm). The dendritic spines were rather sparse on the primary and the secondary dendrites, but were progressively more dense on the higher order dendrites (Figs. 2a9 and 3b). The dendritic fields were oval or triangular in shape and had the major diameter ranging from 200 to 750 mm. The majority of them remained within the DRN, although a few dendrites of the multipolar neurons were observed to extend into the ventrolateral division of the periaqueductal gray. The main axons arise from the cell bodies or the primary dendrites. Axonal branches with prominent varicosities were distributed widely in the DRN and also in the periaqueductal gray, especially in its ventrolateral division (Figs. 2a0 and 3c). Many axonal branches were further traced into the midbrain reticular formation; some of them appeared to terminate on the cell bodies of the mesencephalic trigeminal nucleus neurons (Figs. 2a0 and 3d). Axonal branches often showed recurrent courses (Fig. 2a0). The target of the main axons could not be determined in the present study. The somatodendritic domain of the 5-HT-immunonegative projection neurons in the DRN had essentially the same general morphology features with those of the 5-HTimmunopositive projection neurons in the DRN, except that the dendritic spines were rarely seen in the 5-HTnegative neurons (Compare Figs. 4 and 5 with Figs. 2 and 3). Many axonal branches of the 5-HT-immunonegative projection neurons of the DRN were also observed to be distributed in the ventrolateral part of the periaqueductal gray and the midbrain reticular formation as well as in the DRN, although no targets of the main axons of these neurons could be confirmed (Fig. 4). No presumed axon terminals of the 5-HT-immunonegative projection neurons of the DRN were found on the neuronal cell bodies in the mesencephalic trigeminal nucleus. The main axons could not be traced to their target.
3.2. Electrophysiological properties The data obtained from morphologically defined 15 projection neurons in the DRN (seven 5-HT-immunopositive and eight 5-HT-immunonegative projection neurons) were summarized in Table 1. All of these neurons had been examined electrophysiologically by injecting depolarizing and hyperpolarizing current pulses of 120 ms and 500 ms; the intracellular penetration stably lasted for 30–60 min. After the intracellular recording study, the recorded neurons were intracellularly labeled successfully with the biocytin method and confirmed to be projection neurons sending their main axons outside the DRN.
113
Fig. 2. Camera lucida reconstruction of a single exemplary 5-HT-immunopositive projection neuron, the morphology of which was visualized by the intracellular biocytin injection. The location, the features of the somatodendritic domains, and the arborization of the main axons are shown in (a), (a9), and (a0), respectively. Arrows point to the cell body of the neuron. Open arrow indicates the main axon. Dotted lines indicate the border of the DRN or the periaqueductal gray (PAG). Aq, mesencephalic aqueduct; Vme, mesencephalic trigeminal nucleus neuron.
Some electrical membrane properties were different statistically between the 5-HT-immunopositive and the 5-HT-immunonegative projection neurons: the membrane time constant was statistically larger in the 5-HT neurons than in the non-5-HT neurons, whereas no statistical differences were detected in the resting membrane potential, the input resistance, the first time constant, or the electrotonic length (Table 1). Weak depolarizing current injection elicited action potential, the characteristics of which were also different statistically between the 5-HT- and the non-5-HT neurons: The spike height and the spike width were statistically larger in the 5-HT neurons than in the non-5-HT neurons, whereas no statistical difference was found in the threshold for spikes (Table 1 and Fig. 6). In response to stronger depolarizing current pulses, the neurons responded with a train of action potentials (Fig. 6). In response to weak, long depolarizing current pulses, the 5-HT neurons displayed a slow and regular firing activity (662 Hz). On the other hand, the non-5-HT neurons fired at higher frequencies (2068 Hz) even when stronger current was injected. Analysis of interspike intervals revealed some degree of adaptation of spike frequency in one of the seven 5-HT neurons and in seven of the eight non-5-HT neurons (Table 1); in Fig. 7, adaptation of spike frequency is not so
114
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
Fig. 3. Photomicrographs of the 5-HT-immunopositive projection neuron that is depicted in Fig. 2. The neuron in (a) emits three primary dendrites. A part of a dendrite, which is demarcated by a rectangle in (a), is enlarged in (b), where the dendritic spines are pointed by arrowheads. Arrows in (c) indicate varicosities on an axonal branch. In (d), some of the axonal branches with presumed axon terminals of this 5-HT-immunopositive neuron are seen to constitute the pericellular basket around a cell body (asterisk) of a mesencephalic trigeminal nucleus neuron (Compare with Fig. 1a9). Abbreviations are as in Fig. 1.
evident in the 5-HT neuron (Fig. 7a) as in the non-5-HT neuron (Fig. 7b). When hyperpolarizing current pulses were injected, time-dependent depolarization (depolarizing sag), which was indicative of the well-known hyperpolarization-activated cationic current (h current), was seen in six of the eight non-5-HT neurons (b 1 in Fig. 6), but not in the 5-HT neurons. The overall inward rectification, which was measured at the end of 500-ms hyperpolarizing current pulses (arrows in a 1 ,b 1 of Fig. 6), was observed in four of the eight non-5-HT neurons, but not in the 5-HT neurons. At the offset of the hyperpolarizing current pulses, rebound
depolarization hump were observed in six of the non-5-HT neurons; these rebound humps were not accompanied with spikes (b 1 in Fig. 6). No rebound depolarizing hump were observed in the 5-HT neurons. When weak long depolarizing current pulses were injected, the 5-HT neurons displayed a slow regular discharge pattern with gradual interspike depolarization and large fast afterhyperpolarization (a 1 ,a 2 in Fig. 6). On the other hand, action potentials induced in the non-5-HT neurons were often followed by a small and rapid depolarization which led into a hyperpolarization (b 1 in Fig. 6). The same neurons further exhibited long-lasting depolar-
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
115
Fig. 4. Camera lucida reconstruction of a single exemplary 5-HT-immunonegative projection neuron, the morphology of which was visualized by the intracellular biocytin injection (Fig. 1b,b9). The location, the features of the somatodendritic domains, and the arborization of the main axons are shown in (a), (a9), and (a0), respectively, as in Fig. 2.
Table 1 Electrical membrane properties of the serotonergic and non-serotonergic projection neurons in the dorsal raphe nucleus a Property
Serotonergic neurons (n57)
Non-serotonergic neurons (n58)
Resting membrane potential (mV) Input resistance (MV)c Membrane time constant (t0 , ms)d First time constant (t1 , ms)d Electrotonic length e Threshold for spikes (mV) Spike height (mV) Spike width at base (ms) Spike width at half height (ms) Adaptation of spike frequency Depolarizing sag Inward rectification b Rebound hump Fast afterdepolarization Slow afterdepolarization
68613 b 174623 7.461.4 1.361.2 2.060.7 275611 126617 3.661.2 1.260.4 1f 0 0 0 0 0
70610 158620 5.561.3* 1.160.7 1.860.9 27068 104616* 0.960.3** 0.660.2** 7** 6** 4** 6** 5** 5**
a
Statistical significance was evaluated by using the two-tailed, unpaired U-test or Fisher’s exact probability test. *P,0.05 compared with nonserotonergic projection neurons in the DRN. **P,0.01 compared with non-serotonergic projection neurons in the DRN. b Mean6S.D. c Input resistance was determined by hyperpolarizing current pulses, producing voltage shifts of 6–15 mV negative to the rest. d Time constants were determined by injecting a short (2 ms) depolarizing current pulse. e The electrotonic length of an equivalent sealed-end cylinder was ]] estimated by the expression pœt0 /t1 2 1. f Numbers of cases. g Determined at the end of 500-ms current injection (see Fig. 6, arrows).
Fig. 5. Photomicrographs of the 5-HT-immunonegative projection neuron that is depicted in Fig. 4 (also see Fig. 1b,b9). The initial part of the main axon (AX) and three primary dendrites are seen. A part of a dendrite (DEN), which is demarcated by a rectangle in (a), is enlarged in (b); no spines are seen on the dendrites. Arrows in (c) and (d) indicate varicosities on axonal branches within the DRN (c) and the periaqueductal gray (d). Scale bars: a540 mm; b58 mm; c,d510 mm.
116
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
Fig. 6. Intracellular recordings of the 5-HT- (a 1 ,a 2 ) and non-5-HT neurons (b 1 ,b 2 ) in DRN. The morphology of the 5-HT neuron is shown in Figs. 2 and 3, while that of the non-5-HT neuron is shown in Figs. 4 and 5. Long (500 ms) depolarizing and hyperpolarizing current pulses were injected into each neuron. Arrows indicate the points where the input resistance and overall inward rectification were determined.
ization after the offset of depolarizing current pulses (b 1 in Fig. 6). Such small rapid depolarization has been reported as the depolarizing afterpotential (DAP) [30] or the fast afterdepolarization (fADP) [9]. The same neurons further exhibited long-lasting depolarization after the offset of depolarizing current pulses (b 1 in Fig. 6). Such slow return of membrane potential after depolarization has been reported as the slow afterdepolarization potential (sADP) [15,27].
4. Discussion The morphology and electrophysiological properties of 5-HT-immunopositive and 5-HT-immunonegative projection neurons in the rat DRN were investigated in the present study: Intracellular recording and then intracellular labeling with the biocytin were performed in frontal brain slices of 500 mm thickness. The biocytin-labeled neurons were visualized in the serial sections of the brain slices. After observation of the morphology of the biocytinlabeled neurons, 5-HT immunoreactivity of the neurons were examined immunohistochemically. It was possible that the present observation by twodimensional camera lucida drawings might fail to detect
some features of neuronal geometry in the third dimension. However, the axons and dendrites of the sampled neurons were visualized clearly to the peripheral portions with axonal and dendritic branches with very fine diameters. Both 5-HT-positive and 5-HT-negative neurons were considered as projection neurons when their main axons were traced to the tegmental regions of the midbrain outside the DRN, although it was impossible to trace the whole extent of the courses of the main axons. The previous study indicated that axon terminals of serotonergic DRN neurons were in symptomatic synaptic contacts upon mesencephalic trigeminal nucleus neurons [21]. This was also indicated in the present study (Figs. 2a0 and 3d). The DRN of the rat has been indicated to contain a substantial number of non-serotonergic cells ([7]; for further review see Refs. [14,17]). The neurochemical nature of 5-HT-immunonegative neurons observed in the present study remains to be determined. It could also be emphasized that one of the problems inherent to the immunohistochemical technique is the interpretation of false-negative cells. In the present study, however, the dendrites of all 5-HT-immunopositive projection neurons sampled were spiny, whereas those of all 5-HT-immunonegative projection neurons examined were aspiny; somatic spines, which have been reported in some DRN
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
Fig. 7. Spike frequency adaptation of a 5-HT- (a) and a non-5-HT neuron (b) in response to current injection of a long depolarizing current pulse. The morphology and electrophysiological properties of the 5-HT neuron are shown in Figs. 2, 3 and 6a 1 ,a 2 , while those of the non-5-HT neuron are shown in Figs. 4, 5 and 6b 1 ,b 2 . Interspike intervals are plotted against the time after the onset of 500-ms depolarizing current pulses. Adaptation of spike frequency (slope) is not so evident in the 5-HT neuron (a) than in the non-5-HT neuron (b).
neurons in the previous studies [2,5,24,25], were not found in the present study. No other particular differences were detected in the general morphology between the 5-HTimmunopositive and the 5-HT-immunonegative projection neurons in the present study. Both the 5-HT-immunopositive and the 5-HT-immunonegative projection neurons observed in the present study were observed to send many axonal branches within the DRN. These neurons in the DRN were considered to play roles not only as projection neurons but also as intrinsic neurons. The intracellular recordings in the present study were performed in the bridge-balance mode with high-impedance recording electrodes (100–200 MV). This may have led to inaccurate measurements due to shortcomings inherent in the technique. During the present experiments, the diode characteristics of the electrodes were tested by injecting 500-ms depolarizing or hyperpolarizing current pulses with the bridge-balanced mode turned off (data not shown). When the injected depolarizing or hyperpolarizing current pulses ranged from 10.5 nA to 20.5 nA, the difference from the linear current–voltage relationship was less than 5%. The input resistance and the measured time
117
constants of the sampled neurons in the present study apparently might be inaccurate (smaller) due to shunt current around the sharp electrodes. It was considered, however, that the comparison of electrical membrane properties between the 5-HT-immunopositive and the 5HT-negative projection neurons could be made on the basis of the measurements obtained in the present study. Some differences were observed in the electrophysiological properties between the 5-HT-immunopositive and the 5-HT-immunonegative projection neurons sampled in the present study (Table 1; Figs. 6 and 7). Serotonergic neurons in the rat DRN have been considered to exhibit the slow rate of spontaneous firing with a regular rhythm ([2–4,26]; for further review see Refs. [17,18]). The 5-HTimmunopositive neurons sampled in the present study, also showed a slow rhythmical firing when depolarizing current pulses were injected. The gradual interspike depolarization which was associated with the slow regular discharge pattern of the 5-HT-immunopositive neurons observed in the present study (a 1 ,a 2 in Fig. 6) has also been observed in the previous studies [2,3]. On the other hand, some neurons in the rat DRN have been reported to fire spikes frequently in short bursts [10–13]. A high proportion of serotonergic mesopontine neurons in microculture have also been reported to discharge in bursts in response to a single intracellular depolarizing current pulse [19]. Such burst-firing neurons, however, were not detected in the 5-HT-immunopositive projection neurons in the present study. On the other hand, non-serotonergic neurons with various electrophysiological properties have been reported in the rat DRN. This heterogeneous population of neurons showed a less regular or irregular firing pattern, with a firing rate ranging from very slow to fast activity (0.1–30 Hz) [2,3,13,24,26]. In the present study, the 5-HT-immuonegative neurons often showed the depolarizing afterpotential (DAP) [30] or the fast afterdepolarization (fADP) [9], as well as the slow afterdepolarizing potential (sADP) [15,27], when depolarizing current pulses were injected. The fADP or DAP, which were often observed after an action potential in the pyramidal neurons of layer VI of the primary motor and somatosensory cortices and in layers V and VI of the prefrontal cortex in the rat, has been suggested to be an important determinant of the firing pattern of the pyramidal neurons in the cerebral cortex [9,20,30]. On the other hand, the sADP, which was reported in the pyramidal neurons of layer V in the cat sensorimotor cortex [27] and in neurons of the rat dorsolateral septal nucleus [15], has been indicated to be associated with an increased membrane conductance and to trigger action potentials; this could induce or potentiate the bursting activity of neurons in the absence of further stimulation [15,27]. Such sADP was also observed in the intrinsic neurons of the rat medullary dorsal horn [22,23]. In summary, the present study indicated that projection neurons in the rat DRN were divided into the serotonergic
118
Y.-Q. Li et al. / Brain Research 900 (2001) 110 – 118
spiny and the non-serotonergic aspiny neurons. Some differences in electrophysiological properties were also revealed between these two populations of projection neurons. Both of these projection neurons were considered to be involved in the local circuit mechanism within the DRN through their richly branched axon collaterals.
Acknowledgements The authors thank Mr Akira Uesugi and Miss Keiko Okamoto of Kyoto University, Ms Yue-Pin Yuan in The Fourth Military Medical University, and Mr Nobuyuki Kobayashi and Mr Hideki Itabashi in Tokyo Metropolitan Institute for Neuroscience for their photographic help. This work was supported in part Grants-in-Aid from the National Natural Science Foundation of China (39625011, 39970239) and from the Ministry of Education, Science, Sports and Culture of Japan (12680743).
References [1] G.K. Aghajanian, H.J. Haigler, L-tryptophan as a selective histochemical marker for serotonergic neurons in single-cell recording studies, Brain Res. 81 (1974) 364–372. [2] G.K. Aghajanian, C.P. VanderMaelen, Intracellular recordings from serotonergic dorsal raphe neurons: pacemaker potentials and the effect of LSD, Brain Res. 238 (1982) 463–469. [3] G.K. Aghajanian, C.P. VanderMaelen, Intracellular identification of central noradrenergic and serotonergic neurons by a new double labeling procedure, J. Neurosci. 2 (1982) 1786–1792. [4] G.K. Aghajanian, R.Y. Wang, J.B. Baraban, Serotonergic and nonserotonergic neurons of the dorsal raphe: reciprocal changes in firing induced by peripheral nerve stimulation, Brain Res. 153 (1978) 169–175. [5] H. Danner, C. Pfister, Untersuchungen und Zytoarchitektonik des Nucleus raphe dorsalis der Ratte, J. Hirnforsch. 21 (1980) 655–664. [6] F. Datiche, P.-H. Luppi, M. Cattarelli, Serotonergic and nonserotonergic projections from the raphe nuclei to the piriform cortex in the rat: a cholera toxin B subunit (CTb) and 5-HT immunohistochemical study, Brain Res. 671 (1995) 27–37. [7] L. Descarries, K.C. Watkins, S. Garcia, A. Beaudet, The serotonin neurons in nucleus raphe dorsalis of adult rat: a light and electron microscope radioautographic study, J. Comp. Neurol. 207 (1982) 239–254. [8] S. Diaz-Cintra, L. Cintra, T. Kemper, O. Resnick, P.J. Morgane, Nucleus raphe dorsalis: a morphometric Golgi study in rats of three age groups, Brain Res. 207 (1981) 1–16. [9] S. Haj-Dahmane, R. Andrade, Calcium-activated cation nonselective current contributes to the fast afterdepolarization in rat prefrontal cortex neurons, J. Neurophysiol. 78 (1997) 1983–1989. ´ S.E. Gartside, A.E.P. Villa, T. Sharp, Evidence for a [10] M. Hajos, repetitive (burst) firing pattern in a sub-population of 5-hydroxytryptamine neurons in the dorsal and median raphe nuclei of the rat, Neuroscience 69 (1995) 189–197. ´ T. Sharp, Burst-firing activity of presumed 5-HT neurones [11] M. Hajos, of the rat dorsal raphe nucleus: electrophysiological analysis by antidromic stimulation, Brain Res. 740 (1996) 162–168.
´ T. Sharp, A 5-hydroxytryptamine lesion markedly reduces [12] M. Hajos, the incidence of burst-firing dorsal raphe neurones in the rat, Neurosci. Lett. 204 (1996) 161–164. ´ T. Sharp, N.R. Newberry, Intracellular recordings from [13] M. Hajos, burst-firing presumed serotonergic neurones in the rat dorsal raphe nucleus in vivo, Brain Res. 737 (1996) 308–312. [14] G. Halliday, A. Harding, G. Paxinos, Serotonin and tachykinin systems, in: G. Paxinos (Ed.), The Rat Nervous System, Academic Press, San Diego, CA, 1995, pp. 929–974. [15] H. Hasuo, K.D. Phelan, M.J. Twery, J.P. Gallagher, A calciumdependent slow afterdepolarization recorded in rat dorsolateral septal nucleus neurons in vitro, J. Neurophysiol. 64 (1990) 1838– 1846. [16] B. Hoelzel, C. Pfister, Morphometrischer Vergleich der Neurontypen des Nucleus raphe dorsalis und des Nucleus centralis superior (Nc. raphe medianus) der Ratte, Z. Mikroskop.-Anat. Forsch. 97 (1983) 529–534. [17] B.L. Jacobs, E.C. Azmitia, Structure and function of the brain serotonin system, Physiol. Rev. 72 (1992) 165–229. [18] B.L. Jacobs, C.A. Fornal, Activity of brain serotonergic neurons in the behaving animal, Pharmacol. Rev. 43 (1991) 563–578. [19] M.D. Johnson, Synaptic glutamate release by postnatal rat serotonergic neurons in microculture, Neuron 12 (1994) 433–442. [20] T. Kaneko, Y. Kang, N. Mizuno, Glutaminase-positive and glutaminase-negative pyramidal cells in layer VI of the primary motor and somatosensory cortices: a combined analysis by intracellular staining and immunocytochemistry in the rat, J. Neurosci. 15 (1995) 8362–8377. [21] J.-L. Li, K.-H. Xiong, Y.-Q. Li, T. Kaneko, N. Mizuno, Serotonergic innervation of mesencephalic trigeminal nucleus neurons: a light and electron microscopic study in the rat, Neurosci. Res. 37 (2000) 127–140. [22] Y.-Q. Li, H. LI, T. Kaneko, N. Mizuno, Substantia gelatinosa neurons in the medullary dorsal horn: An intracellular labeling study in the rat, J. Comp. Neurol. 411 (1999) 399–412. [23] Y.-Q. Li, H. Li, K. Yang, Z.-M. Wang, T. Kaneko, N. Mizuno, Intracellular labeling study of neurons in the superficial part of the magnocellular layer of the medullary dorsal horn of the rat, J. Comp. Neurol. 428 (2001) 641–655. [24] M.R. Park, Intracellular horseradish peroxidase labeling of rapidly firing dorsal raphe projection neurons, Brain Res. 402 (1987) 117– 130. [25] M.R. Park, H. Imai, S.T. Kitai, Morphology and intracellular responses of an identified dorsal raphe projection neuron, Brain Res. 240 (1982) 321–326. [26] S. Sawyer, J.M. Tepper, S.J. Young, P.M. Groves, Antidromic activation of dorsal raphe neurons from neostriatum: physiological characterization and effects of terminal autoreceptor activation, Brain Res. 332 (1985) 15–28. [27] P.C. Schwindt, W.J. Spain, R.C. Foehring, M.C. Chubb, W.E. Crill, Slow conductances in neurons from cat sensorimotor cortex in vitro and their role in slow excitability changes, J. Neurophysiol. 59 (1988) 450–467. [28] R.P. Vertes, A PHA-L analysis of ascending projections of the dorsal raphe nucleus in the rat, J. Comp. Neurol. 313 (1991) 643–668. [29] R.P. Vertes, B. Kocsis, Projections of the dorsal raphe nucleus to the brainstem: PHA-L analysis in the rat, J. Comp. Neurol. 340 (1994) 11–26. [30] C.R. Yang, J.K. Seamans, N. Gorelova, Electrophysiological and morphological properties of layers V–VI principal pyramidal cells in rat prefrontal cortex in vitro, J. Neurosci. 16 (1996) 1904–1921.