Acetylcholine and glutamate release during sleep–wakefulness in the pedunculopontine tegmental nucleus and norepinephrine changes regulated by nitric oxide

INTRODUCTION

Acetylcholine (ACh) release in the peri locus coeruleus α (LCα) is known to increase prior to rapid eye movement (REM) sleep.¹ However, the source of this ACh increase is not clear. Recently, some hypotheses were published.^2–4 A common factor in these explanations is that ACh release by cholinergic neurons is somehow regulated by inhibitory inputs to make a REM sleep specific increase of ACh release in the peri LCα. Based on our previous result that nitric oxides (NOx) regulate cholinergic neurons in the laterodorsal tegmental nucleus (LDT) via norepinephrine (NE) release,⁵ we hypothesize that NOx regulates cholinergic neurons in the pedunculopontine tegmental nucleus (PPT) to induce a REM specific increase of ACh. The present study describes first the change of extracellular ACh and glutamate (Glu) release across the sleep–waking cycle in the PPT, then the possibility of the NOx regulation on the PPT.

METHOD

Adult male cats (n = 3) were chronically implanted with standard sleep electrodes and guide cannulae for microdialysis probes under an anaesthetic as described previously.¹ Following recovery from surgery, a microdialysis probe (outside diameter 220 μm; membrane length 1 mm) was inserted through the guide cannula into the PPT. Dialysate sampling with free moving cats (flow rate 2.0 μL/min) was timed to each sleep–waking state under EEG monitoring. Pharmacological studies were performed between 09.00 to 15.00 h in adult Sprague-Dawley rats (n = 12) to investigate the NOx and NE changes produced by applying N-methyl-D-aspartate (NMDA) throughout the sampling period for stimulation via microdialysis probes. Dialysates were collected for 10 min intervals for NOx assay and 15 min intervals for NE assay during the pre-stimulation, stimulation and post-stimulation with NMDA (1 mM), ACh and Glu assay of the samples were performed by the methods described in previous papers.^6,7 The detection limit for ACh and Glu were 60 and 50 fmol, respectively. Norepinephrine in the samples was also measured by HPLC-ECD system (detection limit: 0.1 pg/sample). NOx was separated in a HPLC column (EICOM, NO-PAK) into NO₂ and NO₃ and then measured individually by the Griess method (detection limit: 0.1 pmol).

After the experiment, the animals were given a lethal dose of pentobarbital and sagittal Nissl-sections were made to confirm the loci of the probes.

RESULTS

A total of 150 dialysate samples were obtained from the PPT of three cats. Twenty-seven sets (10, 10, 7 sets from each cat) of dialysates were taken during each of SWS, REM sleep and wakefulness (W) for ACh assay, and 23 sets (10, 7, 6 sets) of dialysates for Glu assay. ANOVA indicates there is a significant change in ACh release across sleep–wakefulness (F = 4.496, d.f. = 2), but not in Glu (F = 0.77). Planned comparisons for each sleep stage using paired t-test indicate that the mean (± S.E.M.) amount of ACh content in dialysates during SWS (96.0 ± 6.3 fmol/min sample) was significantly lower than that during REM sleep (122.0 ± 6.3, P < 0.05) and W (130.0 ± 8.1, P < 0.01), respectively. On the other hand, Glu release during W (268.0 ± 23.5 pmol/min sample) is higher than that during REM sleep (235.5 ± 18.7, P < 0.01) and SWS (242.5 ± 16.5, P < 0.05).

Application of NMDA (1 mM) for 10 min induced a significant increase of NOx (two-way ANOVA: P = 0.005, F = 8.1, d.f. = 6, Newman-Keuls post hoc comparing stimulation and baseline (N–K test): P < 0.05) (Fig. 1). On the other hand, NMDA application (1 mM) for 15 min induced a significant decrease of NE during the 15 min application period (N–K test, P = 0.01). This effect on NE release was blocked by co-application of NO synthase inhibitor (NArg) (Fig. 2).

DISCUSSION

The source of an REM specific ACh release in the peri-LCα is now a major concern in the research of REM sleep generating mechanisms. Cholinergic neurons in the PPT are thought to be one of the candidates that produce the ACh increase prior to and during REM sleep in the peri LCα. This mechanism is not clear, but there are two presumable explanations: first, REM-on neurons are dominant in the PPT/LDT and/or selectively terminate into the peri LCα, and second, that type I cholinergic neurons and inhibitory inputs work synergistically to produce a REM-specific increase of ACh release in the peri LCα. For verification of the first possibility, we investigated Glu and ACh release as a major factor in the regulation of PPT discharge. As no REM-specific increases either in ACh or in Glu release are observed in this work, the first possibility might be denied. Glu increase observed in W is supposedly caused by the abundant inputs from the periphery to the thalamus during W and the changes observed in NE by NMDA stimulation presumably indicate that NE contributes both signal selectivity and quick habituation to the continuous sensory inputs, but the details are not elucidated by this work. A second possibility is more plausible from our results, though no NE increase was observed during NMDA application. We speculate the direct contribution of NOx instead of NE: NOx release induced by increased Glu release during W inhibits ACh release at the axon terminal of PPT neurons to produce a REM-specific increase of ACh release in the peri LCα. Further investigation of the interaction between ACh release and NOx would make the mechanisms more clear.