5-HT2A receptor-like protein is present in small neurons located in rat mesopontine cholinergic nuclei, but absent from cholinergic neurons

Neuroscience Letters 314 (2001) 77–81 www.elsevier.com/locate/neulet

5-HT2A receptor-like protein is present in small neurons located in rat mesopontine cholinergic nuclei, but absent from cholinergic neurons Richard Fay a,b, Leszek Kubin a,b,* a

Department of Animal Biology 205ED/VET, School of Veterinary Medicine, University of Pennsylvania, 3800 Spruce Street, Philadelphia, PA 19104-4283, USA b Center for Sleep & Respiratory Neurobiology, University of Pennsylvania, Philadelphia, PA 19104-6046, USA Received 5 April 2001; received in revised form 9 August 2001; accepted 13 August 2001

Abstract The cholinergic neurons of the pedunculopontine and laterodorsal tegmental nuclei (PPN and LDN) increase their activity during wakefulness and REM sleep and interact with brainstem neurons containing serotonin (5-HT) and other amines. To determine whether mesopontine neurons that contain nitric oxide synthase (NOS), a marker for cholinergic cells, express 5-HT2A receptors, dual immunostaining for 5-HT2A receptor-like protein and NOS was employed with either peroxidase or fluorescent secondary probes. Within the PPN and LDN, different cells expressed 5-HT2A receptors and NOS. In addition to the lack of co-localization, the 5-HT2A receptor-expressing cells were smaller and less numerous than the adjacent NOS neurons. We propose that 5-HT2A receptor-expressing cells are local inhibitory interneurons whose one function is to ensure the reciprocal patterns of activity in subpopulations of mesopontine cholinergic and aminergic neurons. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Arousal; Immunohistochemistry; Motor control; Nitric oxide synthase; Pons; Serotonin receptors; Sleep

Reciprocal inhibitory interactions between acetylcholine (ACh)- and serotonin (5-HT)- containing mesopontine neurons represent an important component of the model explaining the generation of the sleep-wake cycle [12]. According to the model, 5-HT neurons have the highest activity during wakefulness and inhibit the discharge of a subpopulation of mesopontine ACh neurons of the pedunculopontine and laterodorsal tegmental nuclei (PPN and LDN). A reduction of 5-HT neuronal activity during quiet sleep weakens this inhibitory influence, thereby facilitating the transition into the rapid eye movement (REM) sleep. When the latter is fully established, 5-HT neurons are silent while ACh cells reach high levels of activity. Recordings from ACh and 5-HT cells across the sleep-wake cycle support this concept [5,13,17]. Also consistent with the model, 5-HT inhibits mesopontine cholinergic neurons [6,8,10], and ACh inhibits most serotonergic cells [6]. More recent data indicate, however, that the reciprocal interaction between ACh and 5-HT neurons is not solely * Corresponding author. Tel.: 11-215-898-1893; fax: 1215-5735186. E-mail address: [email protected] (L. Kubin).

based on direct mutual inhibitory interactions. First, postsynaptic inhibition mediated by gamma amino butyric acid (GABA) also contributes to the suppression of 5-HT cell activity [4]. Second, serotonergic inhibition does not uniformly affect all mesopontine ACh neurons [18]. Third, serotonergic excitation also may play a role in shaping the activity of ACh cells, as immunohistochemical and in situ hybridization studies suggest that mesopontine ACh cells express the excitatory 5-HT2A receptors [14,15]. The latter possibility has not been incorporated into the current models of interactions between aminergic and cholinergic neurons. We recently found neurons expressing 5-HT2A receptors in the pontine regions corresponding to the PPN and LDN [3]. However, the cells were not present throughout these nuclei, and their total number appeared to be smaller than expected from the studies describing the distribution of ACh neurons [1,16]. Since this was not consistent with the proposed widespread expression of 5-HT2A receptors by pontine ACh neurons [14], we investigated the presence of 5-HT2A receptor-like protein in the mesopontine cells containing nitric oxide synthase (NOS), a marker for ACh neurons [20]. An abstract has been published [2]. Five Sprague–Dawley rats (Charles River, MA) were

0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 20 8- X

78

R. Fay, L. Kubin / Neuroscience Letters 314 (2001) 77–81

used. All procedures were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania. The animals were anesthetized with pentobarbital sodium (150 mg/kg intraperitoneally (i.p.); Nembutal, Abbott) and perfused with phosphate buffered saline (PBS; pH 7.4) containing 5000 U heparin and 0.003% lidocaine followed by cold 4% paraformaldehyde and 5% dimethyl sulfoxide in PBS. The brains were postfixed, cryoprotected in 30% sucrose, and cut into five series of 30 mm sections in the coronal (four animals) or sagittal (one animal) plane. All sections were successively incubated in 1% sodium borohydride in PBS and 70% methanol with 0.1% H2O2. They were then incubated in rabbit 5-HT2A receptor antiserum (1:750; Diasorin, Stillwater, MI) in 5% goat serum and 0.4% Triton X-100 in PBS for 36–48 h at 48C, then in biotinylated antirabbit antibodies for 1 h, and then submitted to avidin-biotinhorseradish peroxidase (HRP) process (ABC kit; Vector). HRP was visualized using 0.03% diaminobenzidine (DAB), 0.05% Ni-ammonium sulfate, and 0.0012% H2O2 in 50 mM Tris buffer (pH 7.4). In order to block any unbound biotin and avidin remaining from the 5-HT2A receptor labeling procedures, the sections selected for labeling with a second antibody were incubated in saturated avidin followed by a saturated biotin solution for 30 min each (Avidin/Biotin blocking kit, Vector). They were then incubated in 5% horse serum containing 0.4% Triton X-100 and a mouse monoclonal anti-NOS antiserum (1:3000; RBI, Natick, MA) for 36–48 h at 48C. NOS labeling was visualized with a biotinylated anti-mouse antibody (Vector) and the ABC/DAB procedure without Ni ions. In two animals, we used a fluorescent tag to detect NOS. For this, following incubation with anti-NOS antiserum (1:1500) and biotinylated secondary antibody, the sections were incubated for 1.5 h with egg white-indocarbocyanine (Cy3;1:1000; Jackson, PA). Cells (nucleus with at least one dendrite present) immunopositive for the 5-HT2A receptor were identified by the black granular reaction product characteristic of Ni-intensified DAB- HRP reaction. Cells labeled for NOS were identified by the diffuse brown staining, typical of DAB-HRP reaction without Ni ions. Cell counts were made from HRPlabeled sections, whereas the data about co-localization of markers was derived from both dual labeling procedures, HRP and fluorescent. Digital photomicrographs were taken with a Polaroid DMCle camera and imported to Photoshop software (Adobe). The distribution of ACh neurons revealed by NOS immunoreactivity was consistent with previous reports [1,16,20], and the location of neurons expressing 5-HT2A receptors with our previous study [3]. In control experiments, omission of NOS antisera following the staining for the 5-HT2A receptors resulted in the absence of all subsequent labeling, indicating the lack of any crossreactivity between the antibodies used to visualize 5-HT2A receptors and secondary antibodies for NOS. The specificity of the anti-5-HT2A antisera was verified previously [3]. Following single or double-labeling for NOS and/or 5-

HT2A receptors, the number of NOS-containing neurons in the PPN region far exceeded the number of cells immunoreactive for 5-HT2A receptor. Within the middle two thirds of the rostro-caudal extent of the PPN, the average number of 5-HT2A receptor-positive cells per section and per side was 12.5 ^ 0.6 (SE) (range: 4–27; counted in 30 sections from three rats). In the same sections, there were 54 ^ 2.3 NOS cells per section per side (range: 23–95). The 5-HT2A receptor-expressing neurons were intermingled among NOS neurons, with a tendency to aggregate dorsolaterally and ventromedially to the superior cerebellar peduncle (Fig. 1A,B). In the LDN, many 5-HT2A receptor cells were located near the dorsal and dorsolateral border of the LDN (Fig. 1A,C). Additional small cells positive for 5-HT2A receptors were commingled with NOS neurons within the LDN (Fig. 1D). In both the PPN and LDN, most 5-HT2A receptor cells were spindle-shaped and smaller than NOS cells. The average long-axis somatic diameters were significantly larger for the NOS cells (21.9 ^ 0.63 mm(SE), n ¼ 142) than for the 5-HT2A cells present in the same area (12.1 ^ 1.17 mm, n ¼ 51; P , 0:001, t-test). In sections double-stained with HRP, 5-HT2A-positive processes were often present adjacent to, or crossed over NOS-positive neurons, and NOS-stained processes were often in a very close proximity to 5-HT2A receptor-positive cells (Fig. 1B). In all such instances, however, it was possible to verify that the two chromogens were present in separate cellular elements by observation with high magnification immersion objectives (Fig. 1B,D). Importantly, there was no evidence of NOS co-localization with 5-HT2A receptors in either the PPN or LDN. With the exception of a small fraction of particularly dark-brown stained NOS neurons, most NOS cells could be clearly distinguished from blackstained 5-HT2A receptor- expressing neurons. Similarly, in sections double-stained using fluorescent labeling for NOS, we found no evidence for co-localization of the two proteins within the same cell or its proximal dendrites anywhere in the PPN and LDN region (Fig. 2). Interestingly, some 5-HT2A receptor-positive cell bodies were very closely apposed to NOS-positive cells (Fig. 2A). To verify that our dual immunohistochemical procedures can visualize two markers in a single cell, sections containing the dorsal and median raphe´ nuclei were subjected to staining using anti-5-HT (1:400,000 made in rabbit; Sigma) and anti-NOS antisera, the latter because cells in raphe´ nuclei also contain NOS [7,20]. 5-HT was visualized with Cy3, and NOS with a secondary antibody conjugated to fluoresceine (FITC) (1:100; Vector). In contrast to the absence of co-localization of NOS and 5-HT2A receptors in the PPN and LDN, this dual labeling confirmed that NOS and 5-HT are co-localized in a subpopulation of raphe´ neurons (Fig. 2C). Our results demonstrate that 5-HT2A receptor-like protein is expressed in a distinct population of small cells in the PPN/LDN that do not express NOS and, therefore, are not

R. Fay, L. Kubin / Neuroscience Letters 314 (2001) 77–81

79

Fig. 1. Distribution and morphological features of 5-HT2A receptor- and NOS-expressing neurons in the PPN and LDN. (A) The distribution of NOS (black) and 5-HT2A (red) cells in the PPN and LDN regions redrawn from one double-labeled coronal section. NOS cells at locations other than PPN and LDN (e.g. raphe´ nuclei [7,20]), and 5-HT2A receptor cells located farther than 25 mm away from NOS-stained profiles are not shown. Abbreviations: 4n, trochlear nerve; Me5, mesencephalic trigeminal nucleus; mlf, medial longitudinal fasciculus; scp, superior cerebellar peduncle. (B) Two 5-HT2A receptor-positive PPN cells (black; shown by arrows) located near a NOS-positive neuron (brown). Although cell bodies and processes positive for one or the other marker were often closely apposed, the two labels were not co-localized in the same cells. Orientation markers: D, dorsal; L, lateral. (C,D) The relative size and position of NOS- and 5-HT2A receptor-positive neurons in the LDN. The 5-HT2A receptor-positive cells are less numerous and smaller than NOS-positive neurons. They are often located in the dorsal portion of the LDN (arrows in C) or intermingled among NOS cells within the main body of the nucleus (e.g. the cell framed in C and shown at a higher magnification in D).

cholinergic. In contrast, one immunohistochemical study reported a nearly 100% co- localization of choline acetyltransferase with 5-HT2A receptors in mesopontine neurons [14]. The study used 5-HT2A receptor antiserum generated against the same portion of the receptor protein as the anti-

serum that we used but the methodical differences between the two studies were many, including tissue treatment, section thickness, choice of secondary antibodies, and marker for ACh cells. Notably, we used the anti-5-HT2A receptor antibody at 30 times lower concentration than the

Fig. 2. Dual-staining for NOS (A) and 5-HT2A receptor-like protein (B) using a fluorescent marker (Cy3) and Ni-intensified HRP, respectively, did not reveal any co-localization of the two proteins. The white outlines in (A) represent the three 5-HT2A receptor-positive cells marked by arrows in (B). Note the close apposition between the 5-HT2A receptor-positive cell on the right and a NOS-positive neuron. Orientation markers: D, dorsal; M, medial. (C) Superimposed confocal images of one optical cross-section through the dorsal raphe´ nucleus double-stained for 5-HT (red) and NOS (green), with the two markers co-localized in one cell (arrow).

80

R. Fay, L. Kubin / Neuroscience Letters 314 (2001) 77–81

other study and found strong labeling in all orofacial motoneurons that are well known to possess 5-HT2A receptors [3], whereas the other study did not detect 5-HT2A receptor labeling in trigeminal motoneurons [14]. Moreover, the 5HT2A receptor labeling in large PPN/LDN cells was faint in the latter study, and no small cells were observed. In contrast, we found very distinct 5-HT2A labeling in small cells and not a trace of it in large cells of the PPN/LDN. Thus, while the other study yielded images suggesting a widespread expression of 5-HT2A receptors by mesopontine ACh cells, our data show that 5-HT2A receptor-expressing cells and ACh cells represent two distinct populations. Since postsynaptic effects mediated by 5-HT2A receptors are excitatory, the presence of 5-HT2A receptors would imply that 5HT can excite ACh neurons. However, pharmacological studies do not yield any evidence of this; rather, 5-HT is inhibitory to ACh neurons, with the effect being mediated by 5-HT1A receptors [6,8,10,18]. These neuropharmacological data are consistent with our findings. The function of the small neurons intermingled among ACh neurons and expressing 5-HT2A receptors remains to be determined, but their localization and morphology suggest that they are local inhibitory interneurons. Indeed, the size and distribution of GABA-ergic neurons scattered among mesopontine ACh neurons are similar to the 5-HT2A receptor-expressing cells identified in our study [4,11]. Local GABA-containing neurons of the PPN and LDN may play an important role in sleep- wake regulation, as suggested by the increased c-fos expression in GABAergic neurons in these nuclei following REM sleep [11,19]. In particular, some inhibitory cells that express 5-HT2A receptors may, by means of disinhibition, contribute to the increase in the activity of those ACh neurons that are selectively activated during REM sleep. Alternatively, if the 5HT2A receptor- expressing neurons are excitatory, they may shape the activity of those ACh neurons that, parallel to 5HT neurons [4,13], selectively increase activity during wakefulness. The close apposition between some 5-HT2A receptor and NOS neurons is similar to the previous ultrastructural evidence for such close contacts between mesopontine ACh neurons and small non-ACh cells [16]. Physiological and pharmacological data also support the possibility that some PPN/LDN cells that express 5-HT2A receptors are GABAergic and mediate a negative feedback to 5-HT neurons of the dorsal raphe´ . In those studies, inhibitory postsynaptic currents produced in dorsal raphe´ cells were blocked by either 5-HT2A receptor antagonist or GABAA receptor antagonists [9]. Importantly, many GABAergic cells innervating the dorsal raphe´ nucleus are located in LDN and PPN regions identical to those where we found 5-HT2A receptor- expressing neurons [4]. Thus, our results fit well with those earlier findings if the 5-HT2A receptor expression occurred in GABAergic interneurons. In summary, we determined that mesopontine ACh neurons do not express 5-HT2A receptors; rather the LDN/ PPN cells expressing these receptors represent a distinct

population. The distribution and morphology of these cells suggests that they are inhibitory interneurons whose function is to ensure reciprocal interactions between 5-HT and ACh neurons. This study was supported by grants HL-47600 and HL60287. R. Fay was a Parker B. Francis Fellow in Pulmonary Research.

[1] Armstrong, D.M., Saper, C.B., Levey, A.I., Wainer, B.H. and Terry, R.D., Distribution of cholinergic neurons in rat brain: demonstrated by the immunocytochemical localization of choline acetyltransferase, J. Comp. Neurol., 216 (1983) 53– 68. [2] Fay, R. and Kubin, L., 5-HT2A receptor-like immunoreactivity is present in cells and cellular processes adjacent to mesopontine nitric oxide synthase-containing neurons, Sleep, 23(Suppl. 2) (2000) A108. [3] Fay, R. and Kubin, L., Pontomedullary distribution of 5-HT2A receptor-like protein in the rat, J. Comp. Neurol., 418 (2000) 323–345. [4] Gervasoni, D., Peyron, C., Rampon, C., Barbagli, B., Chouvet, G., Urbain, N., Fort, P. and Luppi, P.-H., Role and origin of the GABAergic innervation of dorsal raphe serotonergic neurons, J. Neurosci., 20 (2000) 4217–4225. [5] Kayama, Y., Ohta, M. and Jodo, E., Firing of ‘possibly’ cholinergic neurons in the rat laterodorsal tegmental nucleus during sleep and wakefulness, Brain Res., 569 (1992) 210– 220. [6] Koyama, Y. and Kayama, Y., Mutual interactions among cholinergic, noradrenergic and serotonergic neurons studied by ionophoresis of these transmitters in rat brainstem nuclei, Neuroscience, 55 (1993) 1117–1126. [7] Le´ ger, L., Charnay, Y., Burlet, S., Gay, N., Schaad, N., Bouras, C. and Cespuglio, R., Comparative distribution of nitric oxide synthase- and serotonin-containing neurons in the raphe nuclei of four mammalian species, Histochem. Cell Biol., 110 (1998) 517–525. [8] Leonard, C.S. and Llina´ s, R., Serotonergic and cholinergic inhibition of mesopontine cholinergic neurons controlling REM sleep: an in vitro electrophysiological study, Neuroscience, 59 (1994) 309–330. [9] Liu, R., Jolas, T. and Aghajanian, G., Serotonin 5-HT2 receptors activate local GABA inhibitory inputs to serotonergic neurons of the dorsal raphe nucleus, Brain Res., 873 (2000) 34–45. [10] Luebke, J.I., Greene, R.W., Semba, K., Kamondi, A., McCarley, R.W. and Reiner, P.B., Serotonin hyperpolarizes cholinergic low-threshold burst neurons in the rat laterodorsal tegmental nucleus in vitro, Proc. Natl. Acad. Sci. USA, 89 (1992) 743–747. [11] Maloney, K.J., Mainville, L. and Jones, B.E., Differential cfos expression in cholinergic, monoaminergic and GABAergic cell groups of the pontomesencephalic tegmentum after paradoxical sleep deprivation and recovery, J. Neurosci., 19 (1999) 3057–3072. [12] McCarley, R.W., Greene, R.W., Rainnie, D. and Portas, C.M., Brainstem neuromodulation and REM sleep, Semin. Neurosci., 7 (1995) 341–354. [13] McGinty, D.J. and Harper, R.M., Dorsal raphe neurons: depression of firing during sleep in cats, Brain Res., 101 (1976) 569–575. [14] Morilak, D.A. and Ciaranello, R.D., 5-HT2 receptor immunoreactivity on cholinergic neurons of the pontomesencepha-

R. Fay, L. Kubin / Neuroscience Letters 314 (2001) 77–81 lic tegmentum shown by double immunofluorescence, Brain Res., 627 (1993) 49–54. [15] Pompeiano, M., Palacios, J.M. and Mengod, G., Distribution of the serotonin 5-HT2 receptor family mRNAs: comparison between 5-HT2A and 5-HT2C receptors, Mol. Brain Res., 23 (1994) 163–178. [16] Spann, B.M. and Grofova, I., Cholinergic and non-cholinergic neurons in the rat pedunculopontine tegmental nucleus, Anat. Embryol., 186 (1992) 215–227. [17] Steriade, M., Datta, S., Pare´ , D., Oakson, G. and Curro´ Dossi, R., Neuronal activities in brain-stem cholinergic nuclei related to tonic activation processes in thalamocortical systems, J. Neurosci., 10 (1990) 2541–2559.

81

[18] Thakkar, M.M., Strecker, R.E. and McCarley, R.W., Behavioral state control through differential serotonergic inhibition in the mesopontine cholinergic nuclei: a simultaneous unit recording and microdialysis study, J. Neurosci., 18 (1998) 5490–5497. [19] Torterolo, P., Yamuy, J., Sampogna, S., Morales, F.R. and Chase, M.H., GABAergic neurons of the laterodorsal and pedunculopontine tegmental nuclei of the cat express cfos during carbachol-induced active sleep, Brain Res., 892 (2001) 309–319. [20] Vincent, S.R. and Kimura, H., Histochemical mapping of nitric oxide synthase in the rat brain, Neuroscience, 46 (1992) 755–784.