Arousal from sleep in response to intermittent hypoxia in rat pups is modulated by medullary raphe GABAergic mechanisms

Abstract

Arousal is an important defense against hypoxia during sleep. Rat pups exhibit progressive arousal impairment (habituation) with multiple hypoxia exposures. The mechanisms are unknown. The medullary raphe (MR) is involved in autonomic functions, including sleep, and receives abundant GABAergic inputs. We hypothesized that inhibiting MR neurons with muscimol, a GABA(A) receptor agonist, or preventing GABA reuptake with nipecotic acid, would impair arousal and enhance arousal habituation and that blocking GABA(A) receptors with bicuculline would enhance arousal and attenuate habituation. Postnatal day 15 (P15) to P25 rat pups were briefly anesthetized, and microinjections with aCSF, muscimol, bicuculline, or nipecotic acid were made into the MR. After a ∼30-min recovery, pups were exposed to four 3-min episodes of hypoxia separated by 6 min of normoxia. The time to arousal from the onset of hypoxia (latency) was determined for each trial. Latency progressively increased across trials (habituation) in all groups. The overall latency was greater after muscimol and nipecotic acid compared with aCSF, bicuculline, or noninjected controls. Arousal habituation was reduced after bicuculline compared with aCSF, muscimol, nipecotic acid, or noninjected pups. Increases in latency were mirrored by decreases in chamber [O2] and oxyhemoglobin saturation. Heart rate increased during hypoxia and was greatest in muscimol-injected pups. Our results indicate that the MR plays an important, not previously described, role in arousal and arousal habituation during hypoxia and that these phenomena are modulated by GABAergic mechanisms. Arousal habituation may contribute to sudden infant death syndrome, which is associated with MR serotonergic and GABAergic receptor dysfunction.

Fig. 1.

An example of an experiment after a medullary microinjection of artificial cerebrospinal fluid (aCSF) illustrating the period from the onset of hypoxia to the time of arousal. Note that there is an increase in both heart rate and respiratory rate (f_R) prior to arousal, with the suggestion that respiratory rate starts to increase prior to heart rate. The onset of hypoxia and the time of arousal are indicated by vertical dashed lines. Arousal is heralded by an abrupt increase in body movement. MOVE, movement, CHO₂, chamber [O₂].

Fig. 2.

A and B: schematic diagram showing the locations of microinjections. Boxes have been drawn to indicate the boundaries to determine the accuracy of the microinjections. Dimensions for the postnatal day 25 (P25) animals were scaled so that both postnatal day 15 (P15) and P25 data could be displayed on the same diagram. C: unstained section showing an example of the determination of lesion foci with fluorescent microbeads. The arrow points to the center of the injection. The facial nuclei (nVII) have been identified for reference.

Fig. 3.

Arousal latencies in response to four trials of hypoxia. Microinjections of muscimol and nipecotic acid into the medullary raphe resulted in significantly longer mean latencies than microinjections of aCSF and bicuculline or in pups that were not anesthetized, operated on, or microinjected (*) but did not increase habituation. In contrast, microinjection of bicuculline eliminated habituation (+). Note that the pattern of arousal latencies for pups microinjected with aCSF was not different from pups in which there was no anesthesia, surgery, or microinjection (no injection control). All values are expressed as means ± SE.

Fig. 4.

Comparison of the mean latency, averaged across trials, for the air control experiments carried out in 18 P15 pups after aCSF, bicuculline, or muscimol microinjections into the medullary raphe with P15 pups exposed to hypoxia. Pups microinjected with muscimol had longer arousal latencies during air breathing (more frequent spontaneous arousals) compared with arousal latencies of pups microinjected with either aCSF (P = 0.022) or bicuculline (P = 0.001) (+). Arousal latencies during air breathing were significantly longer than arousal latencies during hypoxia for pups microinjected with muscimol (P < 0.001) or aCSF (P = 0.010), but not bicuculline (*). All values are expressed as the means ± SE.

Fig. 5.

A: chamber [O₂] at the time of arousal over four trials of hypoxia after medullary raphe microinjection of aCSF, muscimol, bicuculline, and nipecotic acid. Mean chamber [O₂] at arousal was lower after microinjection of muscimol than after microinjections of aCSF and bicuculline and was higher after microinjection of bicuculline than after microinjections with muscimol or nipecotic acid. By trial 4, pups microinjected with muscimol aroused at lower chamber [O₂] compared with those microinjected with aCSF (P = 0.011) or bicuculline (P = 0.001) (*). In contrast, pups microinjected with bicuculline aroused at higher chamber [O₂] than those injected with muscimol (P = 0.001) or nipecotic acid (P = 0.019) (+). B: HbO₂Sat at the time of arousal over four trials of hypoxia after medullary raphe microinjection of aCSF, muscimol, bicuculline, and nipecotic acid. Similar to chamber [O₂], mean HbO₂Sat (averaged over all four trials) at the time or arousal was lowest after microinjections of muscimol compared with those microinjected with aCSF, bicuculline, or nipecotic acid. Pups microinjected with bicuculline aroused at higher mean HbO₂Sats compared with those microinjected with nipecotic acid or muscimol. By trial 4, pups injected with muscimol aroused at significantly lower HbO₂Sats compared with those injected with aCSF (P = 0.001) or bicuculline (P = 0.002) (*), but HbO₂Sats at arousal were not different than in pups microinjected with nipecotic acid. Similarly, pups injected with nipecotic acid aroused at lower HbO₂Sats than those injected with aCSF. In contrast, pups injected with bicuculline aroused at significantly higher HbO₂Sats than pups injected with either muscimol (P = 0.002) or nipecotic acid (P = 0.014) (+). All values are expressed as the means ± SE.

Fig. 6.

Body temperatures before trial 1 and trial 4 of hypoxia. Initial body temperatures were greater in pups microinjected with bicuculline compared with the other groups (*), and temperatures were lowest after microinjections of muscimol (+) and nipecotic acid. By the last trial, there were no differences in body temperature between the groups. Values are expressed as the means ± SE.