Role of the postinspiratory complex in regulating swallow-breathing coordination and other laryngeal behaviors

Abstract

Breathing needs to be tightly coordinated with upper airway behaviors, such as swallowing. Discoordination leads to aspiration pneumonia, the leading cause of death in neurodegenerative disease. Here, we study the role of the postinspiratory complex (PiCo) in coordinating breathing and swallowing. Using optogenetic approaches in freely breathing anesthetized ChATcre:Ai32, Vglut2cre:Ai32 and intersectional recombination of ChATcre:Vglut2FlpO:ChR2 mice reveals PiCo mediates airway protective behaviors. Activation of PiCo during inspiration or the beginning of postinspiration triggers swallow behavior in an all-or-nothing manner, while there is a higher probability for stimulating only laryngeal activation when activated further into expiration. Laryngeal activation is dependent on stimulation duration. Sufficient bilateral PiCo activation is necessary for preserving the physiological swallow motor sequence since activation of only a few PiCo neurons or unilateral activation leads to blurred upper airway behavioral responses. We believe PiCo acts as an interface between the swallow pattern generator and the preBötzinger complex to coordinate swallow and breathing. Investigating PiCo's role in swallow and laryngeal coordination will aid in understanding discoordination with breathing in neurological diseases.

Keywords: control of breathing; mouse; neurocircuitry; neuroscience; swallow function.

Figure 1.. Optogenetic stimulation of postinspiratory complex (PiCo) neurons regulates swallow and laryngeal activation in a phase-specific manner.

(A) Scatter plot of the probability of triggering a swallow (orange) or laryngeal activation (blue) across the respiratory phase (0 start of inspiration, 1 start of next inspiration) in ChATcre:Ai32 mice. * indicates significant increase in the difference between probability of evoking a swallow or laryngeal activation within the first 10% (p=0.02), 70% (p=0.04), and **90% (p=0.005) of the respiratory cycle. (Bi) Scatter plot of the probability of triggering a swallow shows no difference between Vglut2cre:Ai32 (purple), ChATcre:Ai32 (green), and ChATcre:Vglut2FlpO:ChR2 (gold) mice. (Bii) There is no change in the probability of stimulating laryngeal activation between ChATcre:Ai32 and ChATcre:Vglut2FlpO:ChR2 mice. However, there is a significant difference in the probability between Vglut2cre:Ai32 (purple) and ChATcre:Vglut2FlpO:ChR2 (gold) mice at ^**70% (p=0.01) and ^**90% (p=0.003) of the respiratory cycle and Vglut2cre:Ai32 and ChATcre:Ai32 (green) mice at ^##90% (p=0.006) of the respiratory cycle. (C) Representative traces of PiCo-triggered swallow on the left showing the rostrocaudal swallow motor sequence starting with the peak activation of the submental complex and then the laryngeal complex (red arrows), plus swallow-related diaphragm activation known as Schluckatmung. Characterization of laryngeal activation on the right showing only the laryngeal complex is activated in response to the laser in blue.

Figure 1—figure supplement 1.. Stimulation of postinspiratory complex (PiCo) in mice that lack ChR2 or in a region outside of PiCo does not result in activation of PiCo neurons.

(A) Representative traces of PiCo stimulation in Ai32^+/+ mice, not crossed with any genetic cre lines, demonstrates no motor response and no effect on respiratory cycle. (B) Representative trace of PiCo stimulation in ChATcre:Vglut2FlpO mice injected with the pAAV-hSyn Con/Fon hChR2(H134R)-EYFP vector, but did not transfect ChR2 in any neurons, shows no response to PiCo stimulation. (C) Schematic of the in vivo preparation including all nerves and muscles recorded from and ventral optrode placement. Bipolar EMGs were placed in the (1) submental complex, located underneath the chin, recording from the geniohyoid, mylohyoid, and digastric muscles; (2) the laryngeal complex, consisting of the posterior cricoarytenoid; lateral, transverse, and oblique arytenoid, cricothyroid, and thyroarytenoid muscles; and (3) the costal diaphragm. Monopolar suction electrodes were attached to the hypoglossal (XII) and vagus (X) nerves. The blue circles on the surface of the ventral brainstem represents laser location of PiCo activation, corresponding to the representative trace on the right (blue) that triggers a swallow. Moving the lasers medial and slightly caudal (red triangles) does not active PiCo neurons and therefore results in no motor response depicted in the representative trace on the right with the red vertical bar.

Figure 1—figure supplement 2.. Prolonged stimulation of postinspiratory complex (PiCo) does not trigger sequential swallow.

10 s stimulation of ChATcre:Vglut2FlpO:ChR2 neurons reveals PiCo does not trigger multiple swallows but only a single swallow at the beginning. The blue bar indicates laser stimulation.

Figure 2.. Postinspiratory complex (PiCo)-triggered swallows reset the respiratory rhythm, while non-swallows have minimal effect.

Respiratory phase shifts plots were divided into two groups: swallow, PiCo activation that triggered a swallow, or non-swallow, PiCo activation that resulted in laryngeal activation or no motor response. (A) Individual responses in ChATcre:Vglut2FlpO:ChR2 (gold), ChATcre:Ai32 (green), and Vglut2cre:Ai32 (purple) and (B) line of best fit from the above graphs. (C) Representative traces of two examples of swallow (orange star) response on respiratory cycle. On the left, PiCo-triggered swallow inhibits inspiration, resulting in an earlier onset of the next inspiratory breath, and on the right a delay in the next inspiration.

Figure 2—figure supplement 1.. Individual responses in ChATcre:Ai32, Vglut2cre:Ai32, and ChATcre:Vglut2FlpO:ChR2 separated by laser pulse duration.

Respiratory phase shifts plots were divided into two groups: swallow, postinspiratory complex (PiCo) laser activation that triggered a swallow, or non-swallow, PiCo activation that resulted in laryngeal activation or no motor response. Laser pulse duration does not affect respiratory rhythm reset in either swallow or non-swallow responses. This allowed to group all laser pulse durations together seen in Figure 2.

Figure 3.. Effect of postinspiratory complex (PiCo) stimulation duration on swallow behavior and laryngeal activity.

(A) Scatter plot of behavior duration versus laser pulse duration for swallow (orange) and laryngeal activation (blue) only in ChATcre:Vglut2FlpO:ChR2 mice. Each dot represents the average duration per mouse. Data for the laryngeal activation analysis, for all genetic mouse lines, is located in Supplementary file 1A. (B) Representative traces of swallow duration shown by submental complex EMG triggered by 40 ms pulse in orange on the left and 200 ms pulse on the right. Below: representative traces of laryngeal activation, laryngeal complex EMG, duration stimulated by 40 ms pulse in blue on the left, and increases in duration when triggered by 200 ms pulse on the right.

Figure 4.. Swallow-related characteristics in water-triggered swallows and postinspiratory complex (PiCo)-triggered swallows.

(A) Representative trace of a swallow triggered by injection of water into the mouth (blue arrow) on the left and PiCo stimulation (orange) on the right. (B) Histogram of swallows in relation to the onset of inspiration for water swallows (blue, n = 105), ChATcre:Ai32 (green, n = 214), Vglut2cre:Ai32 (purple, n = 369), and ChATcre:Vglut2FlpO:ChR2 (gold, n = 291). There are more swallows in Vglut2cre:Ai32 mice due to a larger N number and a higher probability of triggering a swallow over any other behavior (Figure 1B). (C) Dot plot of each swallow in relation to the inspiratory peak. Swallows triggered by water (blue) or PiCo activation occurred at the same time in relation to inspiratory peak. Data for (B) and (C) are located in Supplementary file 2.

Figure 4—figure supplement 1.. Postinspiratory complex (PiCo)-triggered swallows have a decrease in duration and amplitude compared to water-triggered swallows.

(A) Comparison of durations and (B) amplitude in swallow-related characteristics for swallows triggered by water (Water stim) and swallows triggered by stimulation of PiCo (PiCo stim) in ChATcre:Ai32 (green, N = 10), Vglut2creAi32 (purple, N = 11), and ChATcre:Vglut2FlpO:ChR2 (gold, N = 6). X, vagus nerve; XII, hypoglossal nerve; LC, laryngeal complex.

Figure 5.. Selective transfection of cholinergic/glutamatergic neurons in postinspiratory complex (PiCo) in ChATcre:Vglut2FlpO:ChR2 mice.

(A) Transverse hemisection through Bregma level (–6.9 mm) of the transfected neurons into PiCo bilaterally, left (A1) and right (A2), with the pAAV-hSyn Con/Fon hChR2(H134R)-EYFP vector. (A1a) magnification of the yellow square in (A1) and (A2a) magnification of the yellow square in (A2). (B) Heat map showing the density of neurons transfected by the pAAV-hSyn Con/Fon hChR2(H134R)-EYFP vector from (1) coronal and (2) ventral view of the seven animals used in the functional experiments. X-axis is the transitioning point of compact and semi-compact NAmb. (B3) Rostrocaudal distribution of the total number of transfected neurons counted 1:2 series of 25 µm sections into PiCo. cAmb, nucleus ambiguous pars compacta; scAmb, nucleus ambiguus pars semi-compacta; IO, inferior olive; icp, inferior cerebellar peduncle; Sp5, spinal trigeminal nucleus; VII, facial motor nucleus.

Figure 5—figure supplement 1.. Anatomical characterization of postinspiratory complex (PiCo) region.

(A) Coronal views (Bregma level –6.6 to –7.3 mm) of the ventromedial medulla showing the location of the ChAT neurons (magenta) in PiCo region. (B) Heat map showing the density of ChAT immunoreactive neurons from (B1) coronal and (B2) ventral view of four animals. (B3) Rostrocaudal distribution of the total number of ChAT immunoreactive counted 1:2 series of 25 µm sections into PiCo. Histological analysis of (A, B) was done in C57B6 mice. (C) Coronal views (Bregma level –6.6 to –7.3 mm) of the ventromedial medulla showing the location of the double-conditioned ChATcre:Vglut2FlpO:Ai65 neurons (red) in PiCo region. (D) Heat map showing the density of ChATcre:Vglut2FlpO:Ai65 neurons from (D1) coronal and (D2) ventral view of four animals. (D3) Rostrocaudal distribution of the total number of ChATcre:Vglut2FlpO:Ai65 neurons counted 1:2 series of 25 µm sections into PiCo. Histological analysis of (C, D) was done in ChATcre:Vglut2FlpO:Ai65 mice. X-axis is the transitioning point of compact and semi-compact NAmb. cAmb, nucleus ambiguus pars compacta; scAmb, nucleus ambiguus pars semi-compacta; Amb, nucleus ambiguus pars non-compacta; VII, facial motor nucleus; IO, inferior olive; py, pyramidal tract; Sp5, spinal trigeminal nucleus.

Figure 6.. Missed or low transfection of postinspiratory complex (PiCo) neurons stimulates upper airway responses that cannot unambiguously be characterized as either swallows or laryngeal activation as defined before.

(A) Representative trace of 80 ms activation of ChATcre:Vglut2FlpO:ChR2 neurons at PiCo, resulting in an unknown upper airway activation. The red arrows show the laryngeal complex peak activation occurs before the submental complex peak activation; a reverse order from a typical swallow shown in Figure 1C. (B) Scatter plot of behavior duration versus laser pulse duration for upper airway motor activation. The behavior duration increases as the laser pulse duration increases. Data for this plot is located in Supplementary file 1B. (C) Heat map showing the density of neurons transfected by the pAAV-hSyn Con/Fon hChR2(H134R)-EYFP vector from coronal view of the four ChATcre:Vglut2FlpO:ChR2 mice. Though bilateral transfection, ipsilateral represents the side of the brainstem with the greatest amount of transfection (69 ± 8 neurons and contralateral 34 ± 4 neurons, N = 4). Amb, nucleus ambiguus; IO, inferior olive; py, pyramidal tract; Sp5, spinal trigeminal nucleus.