Brain Research, 580 (1992) 351-357 (~) 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
351
BRES 25175
Cholinergically induced REM sleep triggers Fos-like immunoreactivity in dorsolateral pontine regions associated with REM sleep Priyattam J. Shiromani a, Thomas S. Kilduff b, Floyd E. Bloom c and Robert W. McCarley a aVeterans Administration Medical Center and Harvard Medical School, Brockton, MA 02401 (USA), bThomas S. Kilduff, Stanford University, Stanford, CA (USA) and CFloyd E. Bloom, Scripps Clinic and Research Foundation, La Jolla, CA (USA) (Accepted 4 February 1992) Key words: c-los; Immediate-early gene; REM sleep; Brainstem; Acetylcholine
We sought to determine the presence of Fos-like immunoreactive (Fos-LI) cells in the pontine brainstem following cholinergically induced sustained rapid-eye movement (REMc) sleep in cats. Microinjections (0.25/xl) of vehicle (N = 2) or carbachol (2.0/~g/0.25/~1; N = 4) were made into the medial pontine reticular formation. Carbachol produced a state with all the signs of natural REM sleep and with durations of 15.2-57.8 min. Compared with vehicle control animals, carbachol treated animals showed a significantly higher number of Fos-LI cells in pontine regions implicated in REM sleep generation, with longer REMc bouts associated with more Fos-LI cells than the short-duration bout. Regions with REMc-associated Fos-LI increases included: the lateral dorsal tegmental (LDT) and pedunculopontine tegmental (PPT) nuclei, where some Fos-LI cells were immunohistochemically identified as cholinergic; the locus coeruleus, where some of the Fos-LI cells were identifed to be catecholaminergic; the dorsal raphe and the pontine reticular formation. These findings suggest immediate early gene activation is associated with the ubiquitous biological state of REM sleep. It has recently been determined that a class of genes called immediate-early genes ( I E G ) become activated in response to cellular stimulation (reviewed in ref. 22). The proto-oncogene c-los is one of the most frequently activated IEG. It encodes a nuclear protein which dimerizes with other similarly activated proteins via a leucine zipper and the dimer then binds to a D N A regulatory element (AP-1 site) 6. Such gene regulatory proteins might function as third messengers serving to modify the response of the cell to stimulation 15. It has also been suggested that immediate-early gene expression could be useful in identifying neuronal elements subserving behavior 21. Basal levels of Fos are present in the brain and c-fos is transcribed rapidly and transiently in discrete brain regions in response to a variety of experimental manipulations such as pharmacological treatments 5'8'14' 18,19, salt_loading7 and bright light exposure 1'17'2°. In this study we sought to determine whether c-los activation occurred in conjunction with specific sleep states, especially rapid-eye movement (REM) sleep. In mammals, R E M sleep is clearly identified on the basis of several electrophysiological criteria 3°. A sustained R E M like sleep state, which is necessary in order to exclude the possibility of c-fos induction by intervening episodes of n o n - R E M sleep or waking, can be induced by microinjection of cholinergic agonists into the medial pontine
reticular formation 31. We used carbachol, a mixed cholinergic agonist, to produce a sustained R E M sleep episode. Fos-LI cells were observed in the dorsolateral pontine regions that have been implicated in R E M sleep generation 25. The identification of gene activation associated with R E M sleep may be useful in determining the long-term effects of this sleep state and thus, may provide insight into the functional significance of the state. Under Nembutal anaesthesia (35 mg/kg, i.v.), six adult cats weighing 3-5 kg were chronically implanted with a standard set of electrodes for recording the electroencephalogram, electromyogram, electro-oculogram and ponto-geniculo-occipital ( P G O ) spikes. A stainless steel guide cannula (24 gauge) with a stylette (31 gauge) was implanted in the medial pontine reticular formation (medial PRF) area where cholinergic stimulation reliably produces a sustained R E M sleep-like episode (REMc) (target: P --- 3.0, L = 1.8, H = -4.5, 0 = 40 °, according to Berman's stereotaxic atlas 2). Histological localization of the cannula tips in the medial P R F is depicted in the Fig. 1B schematic. Two weeks after recovery from surgery, all cats were given a week of adaptation to the recording conditions and chamber so as to avoid stress associated with handling and unfamiliarity with the sleep recording environment. The adaptation procedure consisted of the exper-
Correspondence: PJ. Shiromani, Veterans Administration Medical Center and Harvard Medical School, 940 Belmont Str. (151C), Brockton, MA 02401, USA. Fax: (1) (508) 586-4527.
352
A
B
I' D
lOmin
F
.
Fig. 1. Schematic representation of the sleep-wake pattern of six cats (panel A) that received microinjection of vehicle or carbachol into the pontine reticular formation. Arrows represent the point where the animals were administered Nembutal and euthanized. The numbers 1-6 represent the six animals and the abbreviations W, S and R refer to waking, slow-wave sleep and R E M sleep, respectively. Panel B depicts the localization and extent of tissue disruption due to the microinjection for each animal. Note that the effectiveness of Fos-LI was not dependent on the size of the microinjection site. Five animals received microinjections in the right pons while one cat, animal 4 in panel A, received a unilateral injection in the left pons. Panels C - E show the distribution of Fos-LI cells at the P2 pontine level in vehicle animal 2 and carbachol animals 3 and 5, respectively. The different symbols in panels C - E represent Fos-LI cells in the LDT (cross), LC (plus), PPT (triangle), medial pons (rectangle) and other pontine areas (filled dots). Abbreviations: mot 5 = motor nucleus of five; bc = brachium conjunctivum" F F G = gigantocellular tegmental field; FTP = paralemniscal tegmental field; IC = inferior colliculus; LC = locus coeruleus; LDT = lateral dorsal tegmental nucleus; PPT = pedunculopontine tegmental nucleus; TD = dorsal tegmental nucleus.
353
Fig. 2. Photomicrographs of Fos-LI cells in the LDT, LC and PPT of vehicle and carbachol treated animals. Panel A represents vehicle animal 2 and arrows point to Fos-L! ceils. Panel B represents Fos-LI cells in carbachol animal 3 that had 15.2 min of REM sleep. Panels C and D depict Fos-LI cells in the LC, LDT and PPT of carbachol animal 5 that had 49 rain of REM sleep. Note that more Fos-LI cells are seen in carbachol treated animals with sustained REMc bouts. Fos is a nuclear protein and therefore the stained nuclei are visible as round to oval shaped. Often the nucleolus is spared. Abbreviations as in Fig. 1.
Fig. 3. Photomicrographs depict Fos-LI cells in the locus coeruleus (LC) and the pedunculopontine tegmental (PPT) nucleus of carbachol animal 4 that had 45.5 min of REMc. Panels A and C were processed for visualization of Fos and tyrosine hydroxylase immunoreactivity while panels B and D represent adjacent sections which were processed for Fos and choline-acetyitransferase immunoreactivity. Panel A denotes Fos-LI cells in the LC and a catecholaminergic neuron is shown to be Fos-LI. Panel B depicts Fos-LI cells in the PPT and one cholinergic neuron is shown to be Fos positive (panel D). The presence of the Fos protein is denoted by the stained nucleus. Abbreviatinn~ as in Fi~,. 1.
354 imenter's placing each cat in a cat-restraining bag, sitting for 5 min with the bag in the lap, withdrawing and reinserting the stylette in the guide cannula and, after an additional minute, releasing the cat from the bag into the sleep recording chamber. This chamber was a well-ventilated, sound attenuated box (62 × 62 × 92 cm) maintained at room-temperature (23°C) with food, water and a litter box. The cats were returned to the vivarium after 3 h in the sleep-recording chamber. All experiments were begun at about 10.00 h. In all cats, vehicle injections of distilled water (0.25/A) were made into the medial PRF. The other four cats were administered carbachol microinjections (2 /~g/0.25 /A, dissolved in distilled water). All injections were made with a Hamilton microliter syringe attached via polyethylene tubing to a 31 gauge stainless-steel cannula. The injection cannula was the same length as the guide cannula so as to minimize tissue damage associated with the cannula insertion. A total volume of 0.25/~1 was slowly and continuously injected. The infusion cannula was left in place for 1 min and then withdrawn and the stylette reinserted. Immediately thereafter, the cats were gently released from the restraining bag into the sleep-recording chamber and polygraphic sleep recordings were made. In all cats, the polygraphic records were monitored for signs of REM sleep which included cortical E E G desynchronization, muscle atonia, PGO spikes and rapid-eye movements. In all cats, the end of REM sleep was defined as the resumption of muscle tone for 2 min or longer. In three animals receiving carbachol, these criteria were used to define the end of REMc and the an-
TABLE I
Counts of Fos-L1 cells in pontine areas implicated in REM sleep at the level of P2 according to Berman's stereotaxic atlas of the cat brainstem The boundaries of LDT, PPT and LC were deduced from our previous histological maps outlining the distribution of catecholaminergic and cholinergic perikarya 27. A schematic distribution
of Fos-LI cells at the P2 level for vehicle animal 2 and carbachol animals 3 and 5 is depicted in Fig. 1C,D and E, respectively. Animal No. Vehicle
Carbachol*
Counts of Fos-Ll cells LDT
LC
PPT
mPRF
1
11
4
15
2
2
7
3
19
3
3 4 5 6
121 532 626 239
34 180 105 105
103 325 242 297
102 199 73 223
* D e n o t e s significantly higher levels of Fos-LI cells in the LDT, LC, PPT and medial P R F of carbachol treated animals compared to vehicle controls (corrected t-test = 4.49; df = 3; P < 0.05).
imals were then euthanized immediately by injection of Nembutal (80 mg/kg, i.p.). The other animals were euthanized by overdose of Nembutal at appropriate intervals after the microinjection (schematically depicted in Fig. 1A). The time from injection of Nembutal to perfusion was about 8 min in all cats. All brains were perfused via the ascending aorta with 300 ml of ice-cold 0.9% saline followed by 2% formaldehyde in 0.1 M phosphate buffer. The brains were blocked and placed overnight in the fixative and then transferred to 15% sucrose/0.1 M phosphate buffer until equilibration. The pontine brainstem was cut at a thickness of 40/~m on a freezing microtome and one-in-four series of sections were processed for Fos-immunohistochemistry using the avidin-biotin procedure with 3,3'diaminobenzidene (DAB) as the chromagen. A polyclonal antibody against the N-terminal peptide 4-17 of the Fos protein was used (sheep anti-Fos, Cambridge Research Biochemicals) at a dilution of 1:2,000 for immunoperoxidase and 1:200 for immunofluorescence. A small quantity of a monoclonal antibody against Fos (mouse anti-Fos, Microbiology Associates) was available and this was tested in one animal. No difference in Fos-LI was noted between the monoclonal and polyclonal antibodies. Some tissue sections from one animal that received carbachol were processed for visualization of both Fos-LI and tyrosine hydroxylase or Fos-LI and choline acetyltransferase. In these double labelling procedures, Fos-LI was detected using the immunoperoxidase procedure with DAB as the substrate and the second primary antibody was detected using a rhodamine-conjugated secondary antibody. The entire pontine brainstem was examined and representative counts of Fos-LI cells were made in the lateral dorsal tegmental (LDT), locus coeruleus (LC), pedunculopontine tegmental (PPT) and medial pontine reticular formation (medial PRF) areas at the level of P2 in Berman's atlas. This level was chosen because the LDT, PPT and LC are adequately represented at this pontine level. The boundaries of the LDT, LC and PPT were determined from our previous histological maps outlining the distribution of catecholaminergic and cholinergic perikarya 27. An independent t-test with a correction for small samples with unequal variances was used to compare mean number of Fos-LI cells -between vehicle and carbachol treated animals (see page 41 in ref. 32). The power of the test was determined to be 79%. Fig. 1A profiles the sleep-wake states in cats receiving vehicle and carbachol microinjections. The two cats that received vehicle injections entered into the first REM sleep bout 37.2 and 37.5 min, respectively, after the injection. One animal was euthanized immediately
355 upon termination of the last R E M sleep bout while the second animal was euthanized 17.5 min after the end of the second R E M sleep episode. In both vehicle control animals, scattered Fos-LI cells were found to be present in the dorsolateral pontine tegmentum which may represent basal levels of Fos-LI in this region. Table I summarizes the counts of Fos-LI cells at the level of P2 in the dorsolateral pons and medial PRF in the vehicle treated animals. An examination of the entire pons revealed no clearly discernible differences in distribution or number of Fos-LI cells between the two control animals. Fig. 1C depicts the distribution of Fos-LI cells in the pontine section of vehicle animal 2 at the level of P2 and a representative photomicrograph of Fos-LI in the L D T and LC is depicted in Fig. 2A. In the four animals that received carbachol, REMc occurred with an average latency of 7.2 min (range: 4.810.8 min) after the microinjection. Fig. 1A summarizes the REMc pattern in these animals. One carbachol treated animal had a REMc episode which lasted 15.2 min. After the end of the REMc bout, the animal remained in quiet wakefulness for 32.7 min at which time Nembutal was administered and the animal euthanized. The other three carbachol treated animals had sustained episodes of REMc lasting 45.5 min, 49.0 min and 57.8 min. Each of these animals was administered Nembutal immediately upon termination of the REMc bout (see Fig. 1A). In the carbachol treated animals, counts of Fos-LI cells in the LDT, LC, PPT and the medial PRF were significantly higher compared to vehicle controls (corrected t = 4.49, df = 3, P < 0.05; see also Table I). Moreover, in the carbachol treated animals, the animal with the shorter REMc bout (15.2 min) had fewer Fos-LI cells in the dorsolateral pons compared with the animals with longer duration REMc bouts (see Table I and Fig. 1D and E and Fig. 2B and C). In the carbachol treated animals, Fos-LI cells were found within the dorsal and median raphe, LDT-PPT, LC and medial pons (Fig. 2C and D). In order to characterize the localization of Fos-LI within neurons, some tissue from one carbachol treated animal with a sustained REMc bout was incubated with the antibodies against tyrosine hydroxylase and choline acetyltransferase. Fig. 3 depicts Fos-LI cells in the LC and PPT and shows that some Fos-LI cells were catecholaminergic or cholinergic. T h e r e were other Fos-LI cells which were not double-labelled. Double-labelled neurons were noted ipsilateral and contralateral to the carbachol microinjection site. Fos-LI cells were noted around the microinjection site in all animals. Higher levels of Fos-LI cells were noted in the ipsilateral injection side in the animal with 15.2 min of REMc (see Fig. 1D), while in the other carba-
chol treated animals, there was an equal distribution of Fos-LI cells in both sides of the dorsolateral pontine brainstem. In contrast to the dorsolateral pontine areas noted above, Fos-LI cells were also noted in the pontine nuclei of the ventral pons, in the periaqueductal grey and in the inferior colliculus. Fos-LI was also noted in the dorsal portions of the dorsal tegmental nucleus of Guddens. All animals, including vehicle controls, showed similar levels of Fos-LI in these areas. We also examined forebrain areas associated with R E M sleep. For instance, the lateral geniculate nucleus is the area from where P G O waves are recorded most easily. Except for scattered Fos-LI in the LGN, we could not detect differences between carbachol and vehicle treated animals even though carbachol microinjections into the medial PRF induced P G O activity. In the hippocampus, where hippocampal 0 rhythm is recorded during REM sleep, we could not discern increased Fos-LI in the carbachol versus vehicle treated animals. These findings indicate that REMc is accompanied by increased Fos-LI in discrete pontine nuclei. Higher levels of Fos-LI were noted in animals with sustained (45.557.8 min) REMc compared with short duration (15.2 min) REMc or vehicle treated animals. Increased Fos-LI was noted in the LDT, PPT, LC, the raphe and the medial P R E These neuronal areas have been implicated in orchestrating the tonic and phasic events related to R E M sleep 25. Both intracellular and extracellular unit studies have shown that pontine neurons exhibit marked changes in firing rates during REM sleep. For instance, medial PRF 9.24 and LDT-PPT 28 neurons increase discharge rates during R E M sleep while dorsal raphe and LC neurons cease firing 12'23. Some of these neurons have been shown to increase discharge in association with REMc 26._ Lydic et al. have also shown higher glucose utilization in discrete pontine regions in association with naturally occurring REM sleep in cats 11. The presence of Fos-LI has been suggested to be a marker of metabolically active neurons21and our study delineates these sites during REMc. Interestingly, we found Fos-LI cells in the LC and dorsal raphe, where neuronal activity typcially stops during REM sleep 12'23. However, as recently shown by Lydic et al., these areas are metabolically active during REM sleep n. In our study, we were able to determine the neurotransmitter identity of Fos-LI neurons and found that some cholinergic neurons and some LC catecholaminergic neurons were Fos positive. Some of the Fos-LI cells could be associated with peptides or other neurotransmitters such as glutamate. For instance, glutaminergic neurons in the dorsolateral pons have been implicated in mediating the atonia associated with REM sleep 1° and some peptides have been shown to modulate REM sleep 16. Peptidergic
356 and glutaminergic neurons are present in the LDT-PPT 4' 29 Pharmacological treatments 5'8'18'19, noxious stimulation3 and pain 13 have been shown to readily induce Fos. In our study, in all animals, Fos-LI was noted in some pontine regions, such as the ventral pontine nuclei, which have not been linked to R E M sleep generation. The presence of Fos-LI cells in these areas could be due to non-sleep related factors such as the microinfusion procedure or han-
sociated with long-duration R E M sleep bouts and not with the total amount of R E M sleep. For instance, One vehicle treated animal (subject 1) and a carbachol treated animal (subject 3) had similar amounts of R E M sleep: However, the vehicle treated animal had fewer number of Fos-LI cells. Another possibility is that the R E M sleep induced by carbachol microinjections may be particularly effective in the induction of Fos-LI. Future studies will have to explore
dling. However, these factors were not sufficient to trigger Fos-LI in the dorsolateral pons because higher Fos-LI was
the relationship between Fos and R E M sleep in more detail. We believe that the use of Fos immunohistochemistry
noted in animals with REMc bouts. Moreover, in animals administered carbachol, higher number of Fos-LI cells oc-
provides a powerful tool with which to identify brain regions where immediate-early genes such as c-los are ex-
curred in conjunction with sustained (45.5-57.8 min) REMc
pressed during R E M sleep. This is a necessary first step towards understanding the changes in gene transcription which accompany and may underlie an important and ubiquitous biological state.
bouts. Therefore, microinjection of carbachol did not produce a non-specific activation of Fos. Moreover, the time from microinjection to euthanasia was comparable in the carbachol-treated animals and even longer in the vehicletreated animals. Thus, the vehicle treated animals had more time within which Fos activation could have occurred. However, compared to the carbachol treated animals the control animals showed only few Fos-LI cells in the dorsolateral pons. Moreover, the amount of Fos-LI may be as-
thank Mr. Michael Gray and Paul Simpson of the VAMC Animal Research Facility for providing care for the animals. Supported by grants from the DVA Medical Research Service (RJ.S., R.W.M.), NIH NS25212 (P.J.S.) and N1H NS30140 (PJ.S.).
1 Aronin, N., Sagar, S.M., Sharp, ER. and Schwartz, W.J., Light regulates expression of a los-related protein in rat suprachiasmatic nuclei, Proc. Natl. Acad. Sci. USA, 87 (1990) 5959-5962. 2 Berman, A.L., The Brainstem of the Cat, The University of Wisconsin Press, Madison, 1968. 3 Bullitt, E., Expression of c-fos like protein as a marker for neuronal activity following noxious stimulation in the rat, J. Comp. Neurol., 296 (1990) 517-530. 4 Clements, J.R. and Grant, S., Glutamate-like immunoreactivity in neurons of the laterodorsal tegmental and pedunculopontine nuclei in the rat, Neurosci. Lett., 120 (1990) 70-73. 5 Dragunow, M., Robertson, G.S., Faull, R.L.M., Robertson, H.A. and Jansen, K., D 2 dopamine receptor antagonists induce fos and related proteins in rat striatal neurons, Neuroscience, 37 (1990) 287-294. 6 Gentz, R., Rauscher, EJ., Abate, C. and Curran, T., Parallel association of Fos and Jun leucine zippers juxtaposes DNA binding domains, Science, 243 (1989) 1695-1699. 7 Giovannelli, L., Shiromani, P.J., Jirikowski, G.E and Bloom, F.E., Oxytocin neurons in the rat hypothalamus exhibit c-los immunoreactivity upon osmotic stress, Brain Res., 531 (1990) 299-303, 8 Graybiel, A.M., Moratalla, R. and Robertson, H.A., Amphetamine and cocaine induce drug-specific activation of the c-fos gene in striosome-matrix compartments and limbic subdivisions of the striatum, Proc. Natl. Acad. Sci. USA, 87 (1990) 69126916. 9 Ito, K. and McCarley, R.W., Alterations in membrane potential and excitability of cat medial poitine reticular formation neurons during changes in naturally occurring sleep-wake states, Brain Res., 292 (1984) 169-175. 10 Lai, Y.Y. and Siegel, J., Medullary regions mediating atonia, J. Neurosci., 8 (1988) 4790-4796. 11 Lydic, R., Baghdoyan, H.A., Hibbard, L., Bonyak, E.V., DeJoseph, M.R. and Hawkins, R.A., Regional brain glucose metabolism is altered during rapid eye movement sleep in the cat: a preliminary study, J. Comp. Neurol., 304 (1991) 517-529.
12 McGinty, D.J. and Harper, R.M., Dorsal raphe neurons: depression of firing during sleep in cats, Brain Res., 101 (1976) 569-575. 13 Menetrey, D., Gannon, A., Levine, J.D. and Basbaum, A.I., Expression of c-los protein in interneurons and projection neurons of the rat spinal cord in response to noxious somatic, articular, and visceral stimulation, J. Comp. Neurol., 285 (1989) 177-195. 14 Morgan, J.I., Cohen, D.R., Hempstead, J.L. and Curran, T., Mapping patterns of c-los expression in the central nervous system after seizure, Science, 237 (1987) 192-197. 15 Morgan, J.I. and Curran, T., Stimulus-transcription coupling in neurons: role of cellular immediate-early genes, TINS, 12 (1989) 459-462. 16 Prospero-Garcia, O., Morales, M., Arankowsky-Sandoval, G. and Drucker-Colin, R., Vasoactive intestinal polypeptide (VIP) and cerebrospinal fluid (CSF) of sleep-deprived cats restores REM sleep in insomniac recipients, Brain Res., 385 (1986) 169173. 17 Rea, M.A., Light increases Fos-related protein immunoreactivity in the rat suprachiasmatic nuclei, Brain Res. Bull., 23 (1989) 577-581. 18 Robertson, G.S., Herrera, D.G., Dragunow, M. and Robertson, H.A., L-DOPA activates c-fos in the striatum ipsilateral to a 6-hydroxydopamine lesion of the substantia nigra, Eur. J. Pharrnacol., 159 (1989) 99-100. 19 Robertson, G.S., Vincent, S.R. and Fibiger, H.C., Striatonigral projection neurons contain D 1 dopamine receptor-activated c-fos, Brain Res., 523 (1990) 288-290. 20 Rusak, B., Robertson, H.A., Wisden, W. and Hunt, S.P., Light pulses that shift rhythms induce gene expression in the suprachiasmatic nucleus, Science, 248 (1990) 1237-1240. 21 Sagar, S.M., Sharp, ER. and Curran, T., Expression of c-fos protein in brain: metabolic mapping at the cellular level, Science, 240 (1987) 1328-1331. 22 Sheng, M. and Greenberg, M.E., The regulation and function of c-fos and other immediate early genes in the nervous system,
357 Neuron, 4 (1990) 477-485. 23 Sheu, Y.S., Nelson, J.P. and Bloom, EE., Discharge patterns of cat raphe neurons during sleep and waking, Brain Res., 73 (1974) 263-276. 24 Shiromani, P., Armstrong, D.M., Bruce, G., Hersh, L.B., Groves, P.M. and Gillin, J.C., Relation of choline acetyltransferase immunoreactive neurons with cells which increase discharge during REM sleep, Brain Res. Bull., 18 (1987) 447-455. 25 Shiromani, P., Gillin, J.C. and Henriksen, S.J., Acetylcholine and the regulation of REM sleep: basic mechanisms and clinical implication for affective illness and narcolepsy, Annu. Rev. Pharmacol. Toxicol., 27 (1987) 137-156. 26 Shiromani, P. and McGinty, D.J., Pontine neuronal response to local cholinergic microinfusion: relation to REM sleep, Brain Res., 386 (1986) 20-31. 27 Shiromani, P.J., Armstrong, D.M., Berkowitz, A., Jeste, D.V. and Gillin, J.C., Distribution of choline acetyltransferase immunoreactive somata in the feline brainstem: implications for REM sleep generation, Sleep, 11 (1988) 1-16.
28 Steriade, M., Datta, S., Par, D., Oakson, G. and Curr6 Dossi, R., Neuronal activities in brain-stem cholinergic nuclei related to tonic activation processes in thalamocortical systems, J. Neurosci., 10 (1990) 2541-2559. 29 Sutin, E.L. and Jacobowitz, D.M., Immunocytochemical localization of peptides and other neurochemicals in the rat laterodorsal tegmental nucleus and adjacent area, J. Comp. Neurol., 270 (1988) 243-270. 30 Ursin, R. and Sterman, M.B., A Manual for Standardized Scoring of Sleep and Waking States in the Adult Cat, Brain Information Service/Brain Research Instite, University of California, Los Angeles, 1981. 31 Vel~lzquez-Moctezuma, J., Gillin, J.C. and Shiromani, P.J., Effect of specific M1,M2 muscarinic receptor agonists on REM sleep generation, Brain Res., 503 (1989) 128-131. 32 Winer, B.J. Statistical Principles in Experimental Design, second edn., McGraw-Hill, New York, 1971.
351
BRES 25175
Cholinergically induced REM sleep triggers Fos-like immunoreactivity in dorsolateral pontine regions associated with REM sleep Priyattam J. Shiromani a, Thomas S. Kilduff b, Floyd E. Bloom c and Robert W. McCarley a aVeterans Administration Medical Center and Harvard Medical School, Brockton, MA 02401 (USA), bThomas S. Kilduff, Stanford University, Stanford, CA (USA) and CFloyd E. Bloom, Scripps Clinic and Research Foundation, La Jolla, CA (USA) (Accepted 4 February 1992) Key words: c-los; Immediate-early gene; REM sleep; Brainstem; Acetylcholine
We sought to determine the presence of Fos-like immunoreactive (Fos-LI) cells in the pontine brainstem following cholinergically induced sustained rapid-eye movement (REMc) sleep in cats. Microinjections (0.25/xl) of vehicle (N = 2) or carbachol (2.0/~g/0.25/~1; N = 4) were made into the medial pontine reticular formation. Carbachol produced a state with all the signs of natural REM sleep and with durations of 15.2-57.8 min. Compared with vehicle control animals, carbachol treated animals showed a significantly higher number of Fos-LI cells in pontine regions implicated in REM sleep generation, with longer REMc bouts associated with more Fos-LI cells than the short-duration bout. Regions with REMc-associated Fos-LI increases included: the lateral dorsal tegmental (LDT) and pedunculopontine tegmental (PPT) nuclei, where some Fos-LI cells were immunohistochemically identified as cholinergic; the locus coeruleus, where some of the Fos-LI cells were identifed to be catecholaminergic; the dorsal raphe and the pontine reticular formation. These findings suggest immediate early gene activation is associated with the ubiquitous biological state of REM sleep. It has recently been determined that a class of genes called immediate-early genes ( I E G ) become activated in response to cellular stimulation (reviewed in ref. 22). The proto-oncogene c-los is one of the most frequently activated IEG. It encodes a nuclear protein which dimerizes with other similarly activated proteins via a leucine zipper and the dimer then binds to a D N A regulatory element (AP-1 site) 6. Such gene regulatory proteins might function as third messengers serving to modify the response of the cell to stimulation 15. It has also been suggested that immediate-early gene expression could be useful in identifying neuronal elements subserving behavior 21. Basal levels of Fos are present in the brain and c-fos is transcribed rapidly and transiently in discrete brain regions in response to a variety of experimental manipulations such as pharmacological treatments 5'8'14' 18,19, salt_loading7 and bright light exposure 1'17'2°. In this study we sought to determine whether c-los activation occurred in conjunction with specific sleep states, especially rapid-eye movement (REM) sleep. In mammals, R E M sleep is clearly identified on the basis of several electrophysiological criteria 3°. A sustained R E M like sleep state, which is necessary in order to exclude the possibility of c-fos induction by intervening episodes of n o n - R E M sleep or waking, can be induced by microinjection of cholinergic agonists into the medial pontine
reticular formation 31. We used carbachol, a mixed cholinergic agonist, to produce a sustained R E M sleep episode. Fos-LI cells were observed in the dorsolateral pontine regions that have been implicated in R E M sleep generation 25. The identification of gene activation associated with R E M sleep may be useful in determining the long-term effects of this sleep state and thus, may provide insight into the functional significance of the state. Under Nembutal anaesthesia (35 mg/kg, i.v.), six adult cats weighing 3-5 kg were chronically implanted with a standard set of electrodes for recording the electroencephalogram, electromyogram, electro-oculogram and ponto-geniculo-occipital ( P G O ) spikes. A stainless steel guide cannula (24 gauge) with a stylette (31 gauge) was implanted in the medial pontine reticular formation (medial PRF) area where cholinergic stimulation reliably produces a sustained R E M sleep-like episode (REMc) (target: P --- 3.0, L = 1.8, H = -4.5, 0 = 40 °, according to Berman's stereotaxic atlas 2). Histological localization of the cannula tips in the medial P R F is depicted in the Fig. 1B schematic. Two weeks after recovery from surgery, all cats were given a week of adaptation to the recording conditions and chamber so as to avoid stress associated with handling and unfamiliarity with the sleep recording environment. The adaptation procedure consisted of the exper-
Correspondence: PJ. Shiromani, Veterans Administration Medical Center and Harvard Medical School, 940 Belmont Str. (151C), Brockton, MA 02401, USA. Fax: (1) (508) 586-4527.
352
A
B
I' D
lOmin
F
.
Fig. 1. Schematic representation of the sleep-wake pattern of six cats (panel A) that received microinjection of vehicle or carbachol into the pontine reticular formation. Arrows represent the point where the animals were administered Nembutal and euthanized. The numbers 1-6 represent the six animals and the abbreviations W, S and R refer to waking, slow-wave sleep and R E M sleep, respectively. Panel B depicts the localization and extent of tissue disruption due to the microinjection for each animal. Note that the effectiveness of Fos-LI was not dependent on the size of the microinjection site. Five animals received microinjections in the right pons while one cat, animal 4 in panel A, received a unilateral injection in the left pons. Panels C - E show the distribution of Fos-LI cells at the P2 pontine level in vehicle animal 2 and carbachol animals 3 and 5, respectively. The different symbols in panels C - E represent Fos-LI cells in the LDT (cross), LC (plus), PPT (triangle), medial pons (rectangle) and other pontine areas (filled dots). Abbreviations: mot 5 = motor nucleus of five; bc = brachium conjunctivum" F F G = gigantocellular tegmental field; FTP = paralemniscal tegmental field; IC = inferior colliculus; LC = locus coeruleus; LDT = lateral dorsal tegmental nucleus; PPT = pedunculopontine tegmental nucleus; TD = dorsal tegmental nucleus.
353
Fig. 2. Photomicrographs of Fos-LI cells in the LDT, LC and PPT of vehicle and carbachol treated animals. Panel A represents vehicle animal 2 and arrows point to Fos-L! ceils. Panel B represents Fos-LI cells in carbachol animal 3 that had 15.2 min of REM sleep. Panels C and D depict Fos-LI cells in the LC, LDT and PPT of carbachol animal 5 that had 49 rain of REM sleep. Note that more Fos-LI cells are seen in carbachol treated animals with sustained REMc bouts. Fos is a nuclear protein and therefore the stained nuclei are visible as round to oval shaped. Often the nucleolus is spared. Abbreviations as in Fig. 1.
Fig. 3. Photomicrographs depict Fos-LI cells in the locus coeruleus (LC) and the pedunculopontine tegmental (PPT) nucleus of carbachol animal 4 that had 45.5 min of REMc. Panels A and C were processed for visualization of Fos and tyrosine hydroxylase immunoreactivity while panels B and D represent adjacent sections which were processed for Fos and choline-acetyitransferase immunoreactivity. Panel A denotes Fos-LI cells in the LC and a catecholaminergic neuron is shown to be Fos-LI. Panel B depicts Fos-LI cells in the PPT and one cholinergic neuron is shown to be Fos positive (panel D). The presence of the Fos protein is denoted by the stained nucleus. Abbreviatinn~ as in Fi~,. 1.
354 imenter's placing each cat in a cat-restraining bag, sitting for 5 min with the bag in the lap, withdrawing and reinserting the stylette in the guide cannula and, after an additional minute, releasing the cat from the bag into the sleep recording chamber. This chamber was a well-ventilated, sound attenuated box (62 × 62 × 92 cm) maintained at room-temperature (23°C) with food, water and a litter box. The cats were returned to the vivarium after 3 h in the sleep-recording chamber. All experiments were begun at about 10.00 h. In all cats, vehicle injections of distilled water (0.25/A) were made into the medial PRF. The other four cats were administered carbachol microinjections (2 /~g/0.25 /A, dissolved in distilled water). All injections were made with a Hamilton microliter syringe attached via polyethylene tubing to a 31 gauge stainless-steel cannula. The injection cannula was the same length as the guide cannula so as to minimize tissue damage associated with the cannula insertion. A total volume of 0.25/~1 was slowly and continuously injected. The infusion cannula was left in place for 1 min and then withdrawn and the stylette reinserted. Immediately thereafter, the cats were gently released from the restraining bag into the sleep-recording chamber and polygraphic sleep recordings were made. In all cats, the polygraphic records were monitored for signs of REM sleep which included cortical E E G desynchronization, muscle atonia, PGO spikes and rapid-eye movements. In all cats, the end of REM sleep was defined as the resumption of muscle tone for 2 min or longer. In three animals receiving carbachol, these criteria were used to define the end of REMc and the an-
TABLE I
Counts of Fos-L1 cells in pontine areas implicated in REM sleep at the level of P2 according to Berman's stereotaxic atlas of the cat brainstem The boundaries of LDT, PPT and LC were deduced from our previous histological maps outlining the distribution of catecholaminergic and cholinergic perikarya 27. A schematic distribution
of Fos-LI cells at the P2 level for vehicle animal 2 and carbachol animals 3 and 5 is depicted in Fig. 1C,D and E, respectively. Animal No. Vehicle
Carbachol*
Counts of Fos-Ll cells LDT
LC
PPT
mPRF
1
11
4
15
2
2
7
3
19
3
3 4 5 6
121 532 626 239
34 180 105 105
103 325 242 297
102 199 73 223
* D e n o t e s significantly higher levels of Fos-LI cells in the LDT, LC, PPT and medial P R F of carbachol treated animals compared to vehicle controls (corrected t-test = 4.49; df = 3; P < 0.05).
imals were then euthanized immediately by injection of Nembutal (80 mg/kg, i.p.). The other animals were euthanized by overdose of Nembutal at appropriate intervals after the microinjection (schematically depicted in Fig. 1A). The time from injection of Nembutal to perfusion was about 8 min in all cats. All brains were perfused via the ascending aorta with 300 ml of ice-cold 0.9% saline followed by 2% formaldehyde in 0.1 M phosphate buffer. The brains were blocked and placed overnight in the fixative and then transferred to 15% sucrose/0.1 M phosphate buffer until equilibration. The pontine brainstem was cut at a thickness of 40/~m on a freezing microtome and one-in-four series of sections were processed for Fos-immunohistochemistry using the avidin-biotin procedure with 3,3'diaminobenzidene (DAB) as the chromagen. A polyclonal antibody against the N-terminal peptide 4-17 of the Fos protein was used (sheep anti-Fos, Cambridge Research Biochemicals) at a dilution of 1:2,000 for immunoperoxidase and 1:200 for immunofluorescence. A small quantity of a monoclonal antibody against Fos (mouse anti-Fos, Microbiology Associates) was available and this was tested in one animal. No difference in Fos-LI was noted between the monoclonal and polyclonal antibodies. Some tissue sections from one animal that received carbachol were processed for visualization of both Fos-LI and tyrosine hydroxylase or Fos-LI and choline acetyltransferase. In these double labelling procedures, Fos-LI was detected using the immunoperoxidase procedure with DAB as the substrate and the second primary antibody was detected using a rhodamine-conjugated secondary antibody. The entire pontine brainstem was examined and representative counts of Fos-LI cells were made in the lateral dorsal tegmental (LDT), locus coeruleus (LC), pedunculopontine tegmental (PPT) and medial pontine reticular formation (medial PRF) areas at the level of P2 in Berman's atlas. This level was chosen because the LDT, PPT and LC are adequately represented at this pontine level. The boundaries of the LDT, LC and PPT were determined from our previous histological maps outlining the distribution of catecholaminergic and cholinergic perikarya 27. An independent t-test with a correction for small samples with unequal variances was used to compare mean number of Fos-LI cells -between vehicle and carbachol treated animals (see page 41 in ref. 32). The power of the test was determined to be 79%. Fig. 1A profiles the sleep-wake states in cats receiving vehicle and carbachol microinjections. The two cats that received vehicle injections entered into the first REM sleep bout 37.2 and 37.5 min, respectively, after the injection. One animal was euthanized immediately
355 upon termination of the last R E M sleep bout while the second animal was euthanized 17.5 min after the end of the second R E M sleep episode. In both vehicle control animals, scattered Fos-LI cells were found to be present in the dorsolateral pontine tegmentum which may represent basal levels of Fos-LI in this region. Table I summarizes the counts of Fos-LI cells at the level of P2 in the dorsolateral pons and medial PRF in the vehicle treated animals. An examination of the entire pons revealed no clearly discernible differences in distribution or number of Fos-LI cells between the two control animals. Fig. 1C depicts the distribution of Fos-LI cells in the pontine section of vehicle animal 2 at the level of P2 and a representative photomicrograph of Fos-LI in the L D T and LC is depicted in Fig. 2A. In the four animals that received carbachol, REMc occurred with an average latency of 7.2 min (range: 4.810.8 min) after the microinjection. Fig. 1A summarizes the REMc pattern in these animals. One carbachol treated animal had a REMc episode which lasted 15.2 min. After the end of the REMc bout, the animal remained in quiet wakefulness for 32.7 min at which time Nembutal was administered and the animal euthanized. The other three carbachol treated animals had sustained episodes of REMc lasting 45.5 min, 49.0 min and 57.8 min. Each of these animals was administered Nembutal immediately upon termination of the REMc bout (see Fig. 1A). In the carbachol treated animals, counts of Fos-LI cells in the LDT, LC, PPT and the medial PRF were significantly higher compared to vehicle controls (corrected t = 4.49, df = 3, P < 0.05; see also Table I). Moreover, in the carbachol treated animals, the animal with the shorter REMc bout (15.2 min) had fewer Fos-LI cells in the dorsolateral pons compared with the animals with longer duration REMc bouts (see Table I and Fig. 1D and E and Fig. 2B and C). In the carbachol treated animals, Fos-LI cells were found within the dorsal and median raphe, LDT-PPT, LC and medial pons (Fig. 2C and D). In order to characterize the localization of Fos-LI within neurons, some tissue from one carbachol treated animal with a sustained REMc bout was incubated with the antibodies against tyrosine hydroxylase and choline acetyltransferase. Fig. 3 depicts Fos-LI cells in the LC and PPT and shows that some Fos-LI cells were catecholaminergic or cholinergic. T h e r e were other Fos-LI cells which were not double-labelled. Double-labelled neurons were noted ipsilateral and contralateral to the carbachol microinjection site. Fos-LI cells were noted around the microinjection site in all animals. Higher levels of Fos-LI cells were noted in the ipsilateral injection side in the animal with 15.2 min of REMc (see Fig. 1D), while in the other carba-
chol treated animals, there was an equal distribution of Fos-LI cells in both sides of the dorsolateral pontine brainstem. In contrast to the dorsolateral pontine areas noted above, Fos-LI cells were also noted in the pontine nuclei of the ventral pons, in the periaqueductal grey and in the inferior colliculus. Fos-LI was also noted in the dorsal portions of the dorsal tegmental nucleus of Guddens. All animals, including vehicle controls, showed similar levels of Fos-LI in these areas. We also examined forebrain areas associated with R E M sleep. For instance, the lateral geniculate nucleus is the area from where P G O waves are recorded most easily. Except for scattered Fos-LI in the LGN, we could not detect differences between carbachol and vehicle treated animals even though carbachol microinjections into the medial PRF induced P G O activity. In the hippocampus, where hippocampal 0 rhythm is recorded during REM sleep, we could not discern increased Fos-LI in the carbachol versus vehicle treated animals. These findings indicate that REMc is accompanied by increased Fos-LI in discrete pontine nuclei. Higher levels of Fos-LI were noted in animals with sustained (45.557.8 min) REMc compared with short duration (15.2 min) REMc or vehicle treated animals. Increased Fos-LI was noted in the LDT, PPT, LC, the raphe and the medial P R E These neuronal areas have been implicated in orchestrating the tonic and phasic events related to R E M sleep 25. Both intracellular and extracellular unit studies have shown that pontine neurons exhibit marked changes in firing rates during REM sleep. For instance, medial PRF 9.24 and LDT-PPT 28 neurons increase discharge rates during R E M sleep while dorsal raphe and LC neurons cease firing 12'23. Some of these neurons have been shown to increase discharge in association with REMc 26._ Lydic et al. have also shown higher glucose utilization in discrete pontine regions in association with naturally occurring REM sleep in cats 11. The presence of Fos-LI has been suggested to be a marker of metabolically active neurons21and our study delineates these sites during REMc. Interestingly, we found Fos-LI cells in the LC and dorsal raphe, where neuronal activity typcially stops during REM sleep 12'23. However, as recently shown by Lydic et al., these areas are metabolically active during REM sleep n. In our study, we were able to determine the neurotransmitter identity of Fos-LI neurons and found that some cholinergic neurons and some LC catecholaminergic neurons were Fos positive. Some of the Fos-LI cells could be associated with peptides or other neurotransmitters such as glutamate. For instance, glutaminergic neurons in the dorsolateral pons have been implicated in mediating the atonia associated with REM sleep 1° and some peptides have been shown to modulate REM sleep 16. Peptidergic
356 and glutaminergic neurons are present in the LDT-PPT 4' 29 Pharmacological treatments 5'8'18'19, noxious stimulation3 and pain 13 have been shown to readily induce Fos. In our study, in all animals, Fos-LI was noted in some pontine regions, such as the ventral pontine nuclei, which have not been linked to R E M sleep generation. The presence of Fos-LI cells in these areas could be due to non-sleep related factors such as the microinfusion procedure or han-
sociated with long-duration R E M sleep bouts and not with the total amount of R E M sleep. For instance, One vehicle treated animal (subject 1) and a carbachol treated animal (subject 3) had similar amounts of R E M sleep: However, the vehicle treated animal had fewer number of Fos-LI cells. Another possibility is that the R E M sleep induced by carbachol microinjections may be particularly effective in the induction of Fos-LI. Future studies will have to explore
dling. However, these factors were not sufficient to trigger Fos-LI in the dorsolateral pons because higher Fos-LI was
the relationship between Fos and R E M sleep in more detail. We believe that the use of Fos immunohistochemistry
noted in animals with REMc bouts. Moreover, in animals administered carbachol, higher number of Fos-LI cells oc-
provides a powerful tool with which to identify brain regions where immediate-early genes such as c-los are ex-
curred in conjunction with sustained (45.5-57.8 min) REMc
pressed during R E M sleep. This is a necessary first step towards understanding the changes in gene transcription which accompany and may underlie an important and ubiquitous biological state.
bouts. Therefore, microinjection of carbachol did not produce a non-specific activation of Fos. Moreover, the time from microinjection to euthanasia was comparable in the carbachol-treated animals and even longer in the vehicletreated animals. Thus, the vehicle treated animals had more time within which Fos activation could have occurred. However, compared to the carbachol treated animals the control animals showed only few Fos-LI cells in the dorsolateral pons. Moreover, the amount of Fos-LI may be as-
thank Mr. Michael Gray and Paul Simpson of the VAMC Animal Research Facility for providing care for the animals. Supported by grants from the DVA Medical Research Service (RJ.S., R.W.M.), NIH NS25212 (P.J.S.) and N1H NS30140 (PJ.S.).
1 Aronin, N., Sagar, S.M., Sharp, ER. and Schwartz, W.J., Light regulates expression of a los-related protein in rat suprachiasmatic nuclei, Proc. Natl. Acad. Sci. USA, 87 (1990) 5959-5962. 2 Berman, A.L., The Brainstem of the Cat, The University of Wisconsin Press, Madison, 1968. 3 Bullitt, E., Expression of c-fos like protein as a marker for neuronal activity following noxious stimulation in the rat, J. Comp. Neurol., 296 (1990) 517-530. 4 Clements, J.R. and Grant, S., Glutamate-like immunoreactivity in neurons of the laterodorsal tegmental and pedunculopontine nuclei in the rat, Neurosci. Lett., 120 (1990) 70-73. 5 Dragunow, M., Robertson, G.S., Faull, R.L.M., Robertson, H.A. and Jansen, K., D 2 dopamine receptor antagonists induce fos and related proteins in rat striatal neurons, Neuroscience, 37 (1990) 287-294. 6 Gentz, R., Rauscher, EJ., Abate, C. and Curran, T., Parallel association of Fos and Jun leucine zippers juxtaposes DNA binding domains, Science, 243 (1989) 1695-1699. 7 Giovannelli, L., Shiromani, P.J., Jirikowski, G.E and Bloom, F.E., Oxytocin neurons in the rat hypothalamus exhibit c-los immunoreactivity upon osmotic stress, Brain Res., 531 (1990) 299-303, 8 Graybiel, A.M., Moratalla, R. and Robertson, H.A., Amphetamine and cocaine induce drug-specific activation of the c-fos gene in striosome-matrix compartments and limbic subdivisions of the striatum, Proc. Natl. Acad. Sci. USA, 87 (1990) 69126916. 9 Ito, K. and McCarley, R.W., Alterations in membrane potential and excitability of cat medial poitine reticular formation neurons during changes in naturally occurring sleep-wake states, Brain Res., 292 (1984) 169-175. 10 Lai, Y.Y. and Siegel, J., Medullary regions mediating atonia, J. Neurosci., 8 (1988) 4790-4796. 11 Lydic, R., Baghdoyan, H.A., Hibbard, L., Bonyak, E.V., DeJoseph, M.R. and Hawkins, R.A., Regional brain glucose metabolism is altered during rapid eye movement sleep in the cat: a preliminary study, J. Comp. Neurol., 304 (1991) 517-529.
12 McGinty, D.J. and Harper, R.M., Dorsal raphe neurons: depression of firing during sleep in cats, Brain Res., 101 (1976) 569-575. 13 Menetrey, D., Gannon, A., Levine, J.D. and Basbaum, A.I., Expression of c-los protein in interneurons and projection neurons of the rat spinal cord in response to noxious somatic, articular, and visceral stimulation, J. Comp. Neurol., 285 (1989) 177-195. 14 Morgan, J.I., Cohen, D.R., Hempstead, J.L. and Curran, T., Mapping patterns of c-los expression in the central nervous system after seizure, Science, 237 (1987) 192-197. 15 Morgan, J.I. and Curran, T., Stimulus-transcription coupling in neurons: role of cellular immediate-early genes, TINS, 12 (1989) 459-462. 16 Prospero-Garcia, O., Morales, M., Arankowsky-Sandoval, G. and Drucker-Colin, R., Vasoactive intestinal polypeptide (VIP) and cerebrospinal fluid (CSF) of sleep-deprived cats restores REM sleep in insomniac recipients, Brain Res., 385 (1986) 169173. 17 Rea, M.A., Light increases Fos-related protein immunoreactivity in the rat suprachiasmatic nuclei, Brain Res. Bull., 23 (1989) 577-581. 18 Robertson, G.S., Herrera, D.G., Dragunow, M. and Robertson, H.A., L-DOPA activates c-fos in the striatum ipsilateral to a 6-hydroxydopamine lesion of the substantia nigra, Eur. J. Pharrnacol., 159 (1989) 99-100. 19 Robertson, G.S., Vincent, S.R. and Fibiger, H.C., Striatonigral projection neurons contain D 1 dopamine receptor-activated c-fos, Brain Res., 523 (1990) 288-290. 20 Rusak, B., Robertson, H.A., Wisden, W. and Hunt, S.P., Light pulses that shift rhythms induce gene expression in the suprachiasmatic nucleus, Science, 248 (1990) 1237-1240. 21 Sagar, S.M., Sharp, ER. and Curran, T., Expression of c-fos protein in brain: metabolic mapping at the cellular level, Science, 240 (1987) 1328-1331. 22 Sheng, M. and Greenberg, M.E., The regulation and function of c-fos and other immediate early genes in the nervous system,
357 Neuron, 4 (1990) 477-485. 23 Sheu, Y.S., Nelson, J.P. and Bloom, EE., Discharge patterns of cat raphe neurons during sleep and waking, Brain Res., 73 (1974) 263-276. 24 Shiromani, P., Armstrong, D.M., Bruce, G., Hersh, L.B., Groves, P.M. and Gillin, J.C., Relation of choline acetyltransferase immunoreactive neurons with cells which increase discharge during REM sleep, Brain Res. Bull., 18 (1987) 447-455. 25 Shiromani, P., Gillin, J.C. and Henriksen, S.J., Acetylcholine and the regulation of REM sleep: basic mechanisms and clinical implication for affective illness and narcolepsy, Annu. Rev. Pharmacol. Toxicol., 27 (1987) 137-156. 26 Shiromani, P. and McGinty, D.J., Pontine neuronal response to local cholinergic microinfusion: relation to REM sleep, Brain Res., 386 (1986) 20-31. 27 Shiromani, P.J., Armstrong, D.M., Berkowitz, A., Jeste, D.V. and Gillin, J.C., Distribution of choline acetyltransferase immunoreactive somata in the feline brainstem: implications for REM sleep generation, Sleep, 11 (1988) 1-16.
28 Steriade, M., Datta, S., Par, D., Oakson, G. and Curr6 Dossi, R., Neuronal activities in brain-stem cholinergic nuclei related to tonic activation processes in thalamocortical systems, J. Neurosci., 10 (1990) 2541-2559. 29 Sutin, E.L. and Jacobowitz, D.M., Immunocytochemical localization of peptides and other neurochemicals in the rat laterodorsal tegmental nucleus and adjacent area, J. Comp. Neurol., 270 (1988) 243-270. 30 Ursin, R. and Sterman, M.B., A Manual for Standardized Scoring of Sleep and Waking States in the Adult Cat, Brain Information Service/Brain Research Instite, University of California, Los Angeles, 1981. 31 Vel~lzquez-Moctezuma, J., Gillin, J.C. and Shiromani, P.J., Effect of specific M1,M2 muscarinic receptor agonists on REM sleep generation, Brain Res., 503 (1989) 128-131. 32 Winer, B.J. Statistical Principles in Experimental Design, second edn., McGraw-Hill, New York, 1971.
