Chronic intermittent hypoxia alters density of aminergic terminals and receptors in the hypoglossal motor nucleus

Abstract

Rationale: Patients with obstructive sleep apnea (OSA) adapt to the anatomical vulnerability of their upper airway by generating increased activity in upper airway-dilating muscles during wakefulness. Norepinephrine (NE) and serotonin (5-HT) mediate, through α₁-adrenergic and 5-HT₂A receptors, a wake-related excitatory drive to upper airway motoneurons. In patients with OSA, this drive is necessary to maintain their upper airway open. We tested whether chronic intermittent hypoxia (CIH), a major pathogenic factor of OSA, affects aminergic innervation of XII motoneurons that innervate tongue-protruding muscles in a manner that could alter their airway-dilatory action.

Objectives: To determine the impact of CIH on neurochemical markers of NE and 5-HT innervation of the XII nucleus.

Methods: NE and 5-HT terminal varicosities and α₁-adrenergic and 5-HT₂A receptors were immunohistochemically visualized and quantified in the XII nucleus in adult rats exposed to CIH or room air exchanges for 10 h/d for 34 to 40 days.

Measurements and main results: CIH-exposed rats had approximately 40% higher density of NE terminals and approximately 20% higher density of 5-HT terminals in the ventromedial quadrant of the XII nucleus, the region that controls tongue protruder muscles, than sham-treated rats. XII motoneurons expressing α₁-adrenoceptors were also approximately 10% more numerous in CIH rats, whereas 5-HT₂A receptor density tended to be lower in CIH rats.

Conclusions: CIH-elicited increase of NE and 5-HT terminal density and increased expression of α₁-adrenoceptors in the XII nucleus may lead to augmentation of endogenous aminergic excitatory drives to XII motoneurons, thereby contributing to the increased upper airway motor tone in patients with OSA.

Figure 1.

Time course of oxygen level changes in hypoxic chambers during two successive cycles of chronic intermittent hypoxia, as sampled at 13.3-second intervals.

Figure 2.

XII motoneurons retrogradely labeled from the base of the tongue and dopamine β-hydroxylase (DBH)-positive terminals located in the ventromedial quadrant of the XII nucleus in a pair of matched for anteroposterior level brain sections from (A) chronic intermittent hypoxia (CIH)- and (B) sham-exposed rat. (A1, B1) Low-magnification images of the XII nucleus, with rhodamine-labeled XII motoneurons (brown) located mainly in the ventromedial part of the nucleus. A 100 × 100 μm box was placed in this region for counting of DBH-positive terminals (black). (A2, B2) High-magnification images of the counting boxes shown in A1 and B1. Only a small fraction of DBH-positive varicosities contained within the boxes can be seen (those that are within the focal plane of this photograph). (A3, B3) Redrawing of retrogradely labeled XII motoneurons and all DBH-positive terminals contained in the counting boxes shown in panels A2 and B2. Red dots show DBH axonal varicosities found to be closely apposed to labeled XII motoneurons (gray); black dots show all remaining DBH terminals found throughout the depth of these brain sections.

Figure 3.

Dopamine β-hydroxylase (DBH)-positive terminals in the XII nucleus are more numerous in chronic intermittent hypoxia (CIH)- than sham-treated rats. (A) Relationship between the numbers of DBH-positive terminals counted in CIH- versus sham-treated rats in 24 matched for anteroposterior level pairs of brain sections obtained from eight pairs of CIH/sham-treated rats. For all but four pairs of sections, more terminal varicosities were counted in the section from a CIH rat than in the corresponding section from a sham-treated rat (data points above the identity line). (B) The median number of terminals counted in sections from CIH rats (632 per counting box; horizontal line inside the gray box) was significantly higher than for the sham-treated group (455; P = 0.021, paired Mann-Whitney rank sum test; 24 brain sections/group). The gray boxes show 25 to 75% interquartile ranges for the sham (345–671) and CIH rats (497–868); open circles show terminal counts in individual sections.

Figure 4.

5-HT terminal varicosities are more numerous in the XII nucleus in chronic intermittent hypoxia (CIH)- than sham-treated rats. (A, B) High-magnification images showing XII motoneurons retrogradely labeled from the base of the tongue (brown) and terminal fibers and varicosities immunostained for serotonin (5-HT) (black) in a pair of sections matched for anteroposterior (A-P) level from a CIH- and a sham-exposed rat. (C) Relationship between the numbers of 5-HT–positive terminals counted in CIH- versus sham-treated rats for 24 matched for A-P level pairs of brain sections obtained from eight pairs of CIH/sham-treated rats. For most pairs of sections (17/24), more terminal varicosities were counted in the section from a CIH rat than in the corresponding section from a sham-treated rat (data points above the identity line). The average number of 5-HT terminals was significantly higher in CIH- than sham-treated rats (603 ± 37 vs. 503 ± 31 per counting box; P = 0.012, paired t test).

Figure 5.

XII motoneurons immunopositive for α₁-adrenoceptor–like protein are more numerous in chronic intermittent hypoxia (CIH)- than sham-treated rats. (A, B) Examples of α₁-adrenoceptor immunostaining of XII motoneurons in a CIH- and sham-treated rat, respectively. cc = central canal. (C) In the dorsal half of the nucleus, the mean number of α₁-adrenoceptor–positive motoneurons was significantly higher in CIH- than in sham-treated rats (paired t test). The mean number of XII motoneurons positive for α₁-adrenoceptor–like protein also was significantly higher in the ventral than dorsal half of XII nucleus in both CIH- and sham-treated rats (paired t test). In the ventral half of the nucleus, there was only a trend for the mean number of α₁-adrenoceptor–positive motoneurons to be higher in CIH- than sham-treated rats.

Figure 6.

Serotonin (5-HT_2A) receptor-like immunostaining is slightly lower in the XII nucleus in chronic intermittent hypoxia (CIH)- than in sham-treated rats. (A) 5-HT_2A receptor-like immunostaining in the left XII nucleus and the surrounding regions, as converted to a grayscale digital image before densitometric measurements of staining intensity. The continuous white line outlines the entire XII nucleus, whereas the black line ventral to the XII nucleus encircles the reticular formation region used for measurement of background staining (bckg). The white dashed lines encircle the dorsal (XII_d) and ventral (XII_v) halves of the XII nucleus. cc = central canal. (B) With no correction for CIH effect on background staining, the average intensity of 5-HT_2A receptor-like staining intensity within the XII nucleus was not different between CIH- and sham-treated rats (32 pairs of brain sections from eight pairs of CIH/sham-treated rats). In contrast, background staining was significantly higher in CIH- than sham-treated rats. (C) 5-HT_2A receptor-like immunostaining within the XII nucleus was positively correlated with intensity of background staining when analyzed on section-by-section basis. R = correlation coefficient for linear regression. (D) The ratio of 5-HT_2A receptor-like staining intensity within the XII nucleus to background labeling was lower in CIH- than sham-treated rats, suggesting that CIH-treated rats had lower specific expression of 5-HT_2A receptor-like protein in the XII nucleus than sham-treated animals (P = 0.015, paired t test).