Differential localization of carbachol- and bicuculline-sensitive pontine sites for eliciting REM sleep-like effects in anesthetized rats

Abstract

Carbachol, a cholinergic agonist, and GABA(A) receptor antagonists injected into the pontine dorsomedial reticular formation can trigger rapid eye movement (REM) sleep-like state. Data suggest that GABAergic and cholinergic effects interact to produce this effect but the sites where this occurs have not been delineated. In urethane-anesthetized rats, in which carbachol effectively elicits REM sleep-like episodes (REMSLE), we tested the ability of 10 nL microinjections of carbachol (10 mm) and bicuculline (0.5 or 2 mm) to elicit REMSLE at 47 sites located within the dorsal pontine reticular formation at the levels -8.00 to -10.80 from bregma (B) (Paxinos and Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego, 1997). At rostral levels, most carbachol and some bicuculline injections elicited REMSLE with latencies that gradually decreased from 242 to 12 s for carbachol and from 908 to 38 s for bicuculline for more caudal injection sites. As the latencies decreased, the durations of bicuculline-elicited REMSLE increased from 104 s to over 38 min, and the effect was dose dependent, whereas the duration of carbachol-elicited REMSLE changed little (104-354 s). Plots of REMSLE latency versus the antero-posterior coordinates revealed that both drugs were maximally effective near B-8.80. At levels caudal to B-8.80, carbachol was effective at few sites, whereas bicuculline-elicited REMSLE to at least B-9.30 level. Thus, the bicuculline-sensitive sites extended further caudally than those for carbachol and antagonism of GABA(A) receptors both triggered REMSLE and controlled their duration, whereas carbachol effects on REMSLE duration were small or limited by its concurrent REMSLE-opposing actions.

Figure 1

Examples of short-lasting (mono-phasic) REM sleep-like episodes (REMSLE) elicited by carbachol (a) and 0.5 mM bicuculline (b) injected consecutively into the same pontine site located at the level B-8.72 from bregma. Both drugs elicited REMSLE characterized by similar hippocampal and cortical activations, depressions of XII nerve activity and reductions of the central respiratory rate. XII nerve activity is shown as the moving average (MA) of the signal, with each peak representing the magnitude of inspiratory activity in successive central respiratory cycles. The respiratory rate was calculated in successive 10 s intervals. The latency of the REMSLE elicited by bicuculline was characteristically longer than that elicited by carbachol (bicuculline was injected 240 s prior to the beginning of the trace in b), whereas the durations of the episodes were similar. The vertical lines show the onsets and offsets of both responses, as defined in the Methods. (c, d) Power spectra for cortical EEG determined under baseline conditions (thin line) and around the peak of the responses (thick line) for the carbachol- (c) and bicuculline-induced (d) REMSLE. The insets show the corresponding expanded power spectra (in logarithmic scale) for the 6–12 Hz range, where we observed characteristic increase during REMSLE.

Figure 2

Example of a typical bi-phasic REM sleep-like episode (REMSLE) elicited by 2 mM bicuculline injection made at the level B-8.80 from bregma. Note the characteristic bout of cortical and hippocampal activation at the beginning of the episode followed by a prolonged period of moderately increased cortical and hippocampal powers and continuing depression of XII nerve activity and reduced respiratory rate. The respiratory rate was calculated in successive 15 s intervals. The two vertical lines show the onset and offset of the episode. (b, c) Power spectra for cortical EEG measured under baseline conditions (thin lines) and at two times during the episode (thick lines). The insets show corresponding expanded power spectra (in logarithmic scale) for the 6–12 Hz range in which power of cortical EEG is characteristically increased during REMSLE. The spectrum obtained during the initial phase (b) also shows prominently reduced power in the low frequency range and a prominent peak in the theta-like range (3–5 Hz). The last two changes are in this example relatively larger than in most REMSLE, whereas the increase in the 6–12 Hz range is typical. The power spectrum during the later stage of this REMSLE (c) shows the same features but they are of considerably lower magnitude. The periods from which these spectra were derived are marked by horizontal lines under the EEG traces.

Figure 3

Distribution of all 47 injection sites plotted on the closest standard brainstem cross-sections representing the indicated antero-posterior levels from bregma according to Paxinos and Watson (1997). The effects produced by carbachol are shown on the left-hand side and those elicited from the same sites by bicuculline by the symmetrically located symbols on the opposite side of each cross-section. Filled symbols represent injections that elicited mono-phasic episodes and dotted symbols show bi-phasic REMSLE as indicated (see text for the characteristics of the two types of episodes). Based on the distribution of the effective sites, the rostro-caudal extent of the region where carbachol elicited REMSLE was located rostral relative to the region where bicuculline was effective. The sites at which the long-lasting, bi-phasic REMSLE were elicited by bicuculline were located only caudal to B-8.30 and extended to the level B-9.30, whereas carbachol effectiveness decreased caudal to the level B-8.72. 7, facial motor nucleus; Gi, gigantocellular reticular region; IRt, intermediate reticular region; LC, locus coeruleus; LDT, laterodorsal tegmental nucleus; LDTV, laterodorsal tegmental nucleus, ventral part; Mo5, trigeminal motor nucleus; PC, nucleus pontis caudalis; PO, nucleus pontis oralis; scp, superior cerebellar peduncle; SubCA, nucleus subcoeruleus, alpha; VT, ventral tegmental nucleus; VTx, the site of intersection between the VT and the medial longitudinal fasciculus.

Figure 4

Relationship between the latencies of the REMSLE elicited by carbachol (a) or bicuculline (b) and the antero-posterior coordinates of the injection sites. The plots are vertically divided into the levels corresponding to the standard sections shown in Fig. 3, and the symbols used to designate different effects are the same as in Fig. 3. Injections that did not elicit any effects are shown by open symbols at latencies 5 and 16 min for carbachol and bicuculline, respectively. Both carbachol and bicuculline elicited REMSLE with shortest latencies at the levels B-8.72 and B-8.80; at more caudal levels, carbachol became ineffective whereas the latencies of the REMSLE elicited by bicuculline were still short through the level B-9.30. The site of intersection between the ventral tegmental nucleus and the medial longitudinal fasciculus (VTx) was used as the anatomical reference point.

Figure 5

Relationships between REMSLE latencies and the antero-posterior coordinates of the injection sites could be approximated with separate linear regressions for the episodes elicited by 0.5 and 2 mM bicuculline and carbachol. The plot shows data for effective injections made at, or rostral to, B-8.80. The vertical lines delineate the range of antero-posterior coordinates corresponding to the standard sections shown in Fig. 3, and the symbols used to designate different effects are the same as in Fig. 3. All three regression lines point to level B-8.80 as the region of highest sensitivity for both bicuculline and carbachol, and the differences in the slopes of the regressions are consistent with the diffusion rate being faster for carbachol than bicuculline. The site of intersection between the ventral tegmental nucleus and the medial longitudinal fasciculus (VTx) was used as the anatomical reference point. R, linear regression coefficient.

Figure 6

Relationship between the durations of the REMSLE elicited by carbachol (a) or bicuculline (b) and the antero-posterior coordinates of the injection sites plotted in the same format as in Fig. 4. The injections that did not elicit REMSLE are shown as yielding effects of zero duration (open symbols). Note that the REMSLE elicited by carbachol were 5–10 times shorter than many of those elicited by bicuculline, and that the longest-lasting REMSLE were elicited by either drug from the levels B-8.72 and B-8.80. The site of intersection between the ventral tegmental nucleus and the medial longitudinal fasciculus (VTx) was used as the anatomical reference point.

Figure 7

Relationship between the latencies and durations of REMSLE elicited by carbachol and bicuculline. Linear regression analysis was separately applied to the data for mono-phasic REMSLE elicited by carbachol, REMSLE evoked by 0.5 mM bicuculline (all but one, mono-phasic), and REMSLE elicited by 2 mM bicuculline (all but five, bi-phasic). The symbols used to designate different effects are the same as in Fig. 3. The slopes of the regression lines for 0.5 mM bicuculline and carbachol were similar (the latter was statistically significant). These slopes were nearly flat compared to the regression line for 2 mM bicuculline. The reason for this was that all the REMSLE elicited with latencies shorter than 6 min (thus from the sites located close to the most effective region) were of the much longer-lasting, bi-phasic type. R, linear regression coefficient.

Figure 8

Magnitudes of the changes in XII nerve activity (a), respiratory rate (b), EEG power in the 6–12 Hz range (c) and hippocampal power in the 3–5 Hz range (d) during REMSLE elicited by carbachol and bicuculline shown in relation to the antero-posterior coordinates of the injection sites. Each graph shows combined data for all REMSLE elicited by either carbachol, 0.5 or 2 mM bicuculline. The symbols used to designate different drugs and effects are the same as in Fig. 3. There were only weak trends for the magnitudes of the changes to be larger at the levels B-8.72 and B-8.80, where the latencies were the shortest (see Figs 4 and 5), and for 2 mM bicuculline to produce larger effects than 0.5 mM bicuculline. The site of intersection between the ventral tegmental nucleus and the medial longitudinal fasciculus (VTx) was used as the anatomical reference point.