Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 13.
Published in final edited form as: Curr Biol. 2023 Feb 6;33(5):827–837.e4. doi: 10.1016/j.cub.2023.01.019

A carotid body-brainstem neural circuit mediates sighing in hypoxia

Yilong Yao 1,2, Jingwen Chen 1, Xingyu Li 1, Zhou-Feng Chen 3, Peng Li 1,4,5,6,7,*
PMCID: PMC10337290  NIHMSID: NIHMS1914917  PMID: 36750092

SUMMARY

Increased ventilation is a critical process that occurs when the body responds to a hypoxic environment. Sighs are long, deep breaths that prevent alveolar collapse, and their frequency is significantly increased by hypoxia. In this study, we first show that sighing is induced by hypoxia as a function of increased hypoxic severity and that hypoxia-induced sighing is capable of increasing the oxygen saturation in a mouse model. We next found that the gastrin-releasing peptide (Grp) expressing neurons in the nucleus of the solitary tract (NTS) are important in mediating hypoxia-induced sighing. Retrograde tracing from these Grp neurons reveals their direct afferent input from the petrosal ganglion neurons that innervate the carotid body, the major peripheral chemoreceptor that senses blood oxygen. Acute hypoxia preferentially activates these Grp neurons in the NTS. Photoactivation of these neurons through their projections in the inspiratory rhythm generator in the ventral medulla induces sighing, whereas genetic ablation or chemogenetic silencing of these neurons specifically diminishes the sighs, but not other respiratory responses, induced by hypoxia. Finally, the mice with reduced sighing in hypoxia exhibit an elevated heart-rate increase, which may compensate for maintaining the blood oxygen level. Therefore, we identified a neural circuit that connects the carotid body to the breathing control center in the ventral medulla with a specific function for hypoxia-induced sighing, which restores the oxygen level.

Graphical Abstract

graphic file with name nihms-1914917-f0001.jpg

In brief

Yao et al. demonstrate that sighing is a preferential ventilatory response to hypoxia in mice that elevates blood oxygen. They identify a neural circuit from the carotid body to the ventral brainstem that specifically controls hypoxia-induced sighing. Diminished sighing in hypoxia after inhibition of this circuit is compensated by an increased heart rate.

INTRODUCTION

Interoception is the sensation of the body’s internal physiological conditions, which is fundamental to body homeostasis.1 Sensing and responding to hypoxia (low oxygen level) is critical for survival, and its failure can cause severe organ damage and even death, including the recent silent hypoxia observed in COVID-19 disease.2,3 Hypoxic ventilatory response (HVR), an increase in ventilation to restore normal blood gases, is one major defense mechanism against hypoxia.4 HVR is controlled by the carotid body, the major peripheral chemoreceptor located at the bifurcation of the carotid artery.5,6,7 When arterial oxygen pressure decreases, the carotid body senses the reduced partial pressure of oxygen in arterial blood and relays a stimulus through the glossopharyngeal nerve to the brain to activate respiratory and cardiovascular responses to acute hypoxia.5,6,7 However, the brain circuit that mediates the hypoxia responses remains elusive.

Sighs are long, deep breaths with a bimodal inspiration that occur spontaneously every several minutes to reverse alveolar collapse (atelectasis) and maintain normal lung function.8–12 Although prior studies have shown that sigh rate is induced by various physiological and emotional conditions, including hypoxia,13–15 little is known about the function and the circuit of hypoxia-induced sighing in vivo.

Similar to eupneic breathing, sighing is believed to be generated by the preBötzinger complex (the preBötC), the inspiratory rhythm generator that contains several thousands of neurons in the ventral medulla.16 We recently identified around 200 preBötC neurons that express receptors for the bombesin-like neuropeptides and are dedicated to controlling sighing.15 These sigh control neurons are essential for not only basal and emotional sighs but also for hypoxia-induced sighing.15,17 Here, we show that the hypoxic responses in sighing and breathing are differentially regulated and that sighs increase the blood oxygen level in both hypoxia and normoxia. By integrating neural circuit tracing, neurogenetic manipulation, and respiratory physiology, we further identify a neural circuit that connects the carotid body to the preBötC and that specifically mediates hypoxia-induced sighing in an in vivo mouse model.

RESULTS

Increased sigh rate is a signature ventilatory response to hypoxia

The HVR is a major defense mechanism against hypoxia in which ventilation increases to restore normal blood gases.4 Although it is known that the sigh rate increases in hypoxia,13–15,18 the correlation between sighing and other breathing phenotypes in hypoxia is not fully characterized. To examine the response of sighing to hypoxia, we placed the mice in a whole-body plethysmograph chamber and exposed them to 15% or 10% O2 for 10 min. Sighing increased immediately after the switch from nor-moxia to hypoxia conditions, from 0.4 ± 0.1 sighs per min in nor-moxia (21% O2), to 1.1 ± 0.1 sighs per min in mild hypoxia (15%O2), and to 2.4 ± 0.2 sighs per min in severe hypoxia (10% O2) (Figures 1A and 1C). In both conditions, the number of sighs plateaued for the first 2–3 min and then continued to be greater than the baseline throughout the hypoxia period, although they slightly decreased toward the end of the hypoxia period (Figures 1A and S1A). The total number of sighs in the 10-min hypoxia phase was significantly elevated as a function of increasing hypoxic severity (Figure 1B). In contrast, the respiratory rate increased only ~15% at the peak value in 15% O2 (Figures 1D and 1E), a much milder effect than the sigh rate (Figure 1C). After the initial ventilatory increase phase, the respiratory rate reached the same level of (or below) the baseline in the second phase of hypoxia (Figures 1E, 1I, S1A, and S1B), known as the ventilatory depression phase.4 In 10% O2, the respiratory rate increases to a similar extent as in 15% O2 (Figures 1E and 1F). As a result, the sigh rate/breathing rate ratio doubled after the switch to 15% O2 (from 1.9 ± 0.5 × 10−3 in nor-moxia to 3.8 ± 0.4 × 10−3 in 15% O2) and further increased to more than 4-fold in 10% O2 (8.1 ± 0.7 × 10−3; Figure S1D), suggesting that sighing is preferentially induced in hypoxia. Furthermore, there was no correlation between the increase in respiratory rate and the increase in sigh rate (Figures 1G and 1H). A similar effect was also observed in other breathing parameters, including the tidal volume (TV) (Figures 1I–1L) and minutes of ventilation (Figures S1E–S1H).

Figure 1. Increased sighing is a signature ventilatory response to hypoxia.

Figure 1.

Open in a new tab

(A) Quantification of sigh numbers (bin 1 min) before, during, and after hypoxia challenges of 15% O2 and 10% O2 (mean ± SEM, n = 20).

(B) The number of sighs in 10 min before, during, and after hypoxia challenges (mean ± SEM, n = 20). **p < 0.01; ns, not significant (repeated-measures one-way ANOVA followed by post hoc t test).

(C) Fold changes of sigh numbers (bin 1 min) before, during, and after hypoxia challenges (mean ± SEM, n = 20).

(D) Fold changes of breathing frequency (bin 1 min) before, during, and after hypoxia challenges (mean ± SEM, n = 20).

(E) Quantification of breathing frequency (frequency, bpm) before, during, and after hypoxia challenges (mean ± SEM, n = 20).

(F) Breathing frequency before, during, and after hypoxia challenges (mean ± SEM, n = 20). **p < 0.01; ns, not significant (repeated-measures one-way ANOVA followed by post hoc t test).

(G) The increased number of sighs (Δ sigh numbers) was not significantly correlated with the increased breathing frequency (Δ frequency) in 15% O2 (Pearson’s correlation, *p = 0.34, r = −0.2240, n = 20).

(H) The increased number of sighs (Δ sigh numbers) was not significantly correlated with the increased breathing frequency (Δ frequency) in 10% O2 (Pearson’s correlation, *p = 0.67, r = 0.1002, n = 20).

(I) Quantification of tidal volume (bin 1 min) before, during, and after hypoxia challenges (mean ± SEM, n = 20).

(J) The tidal volume before, during, and after hypoxia challenges (mean ± SEM, n = 20). **p < 0.01; ns, not significant (repeated-measures one-way ANOVA followed by post hoc t test).

(K) The increased number of sighs (Δ sigh numbers) was not correlated with the increased tidal volume (Δ TV) in 15% O2 (Pearson’s correlation, *p = 0.77, r = 0.0711, n = 20).

(L) The increased number of sighs (Δ sigh numbers) was not correlated with the increased tidal volume (Δ TV) in 10% O2 (Pearson’s correlation, *p = 0.54, r = −0.1471, n = 20).

(M) The oxygen saturation before and after a sigh in normoxia (21% O2) (30 sighs from 3 mice).

(N) The oxygen saturation before and after a sigh in hypoxia (10% O2) (24 sighs from 3 mice).

See also Figures S1 and S2.

The ventilatory response to hypoxia could be affected by the metabolic rate and temperature.19 To study the metabolic rate, we measured the oxygen consumption and carbon dioxide production in hypoxia challenges and calculated the convective ventilation to represent HVR more accurately (Figures S1I–S1N). In mild hypoxia (15% O2), the sigh rate started to exhibit a significant increase to 1.3-fold, but the convective ventilation did not increase significantly (Figures S1O–S1Q). In severe hypoxia (10% O2), the sigh rate increased to 2.6-fold, while the convective ventilation only increased 1.4-fold (Figures S1O–S1Q). These results suggest that sighing is preferentially evoked over HVR in a hypoxia-severity-dependent manner. To study the effect of temperature on the preferential increase of sighing, we first measured the body temperature of the mice in the hypoxia challenges and found a slight, insignificant decrease in body temperature in hypoxia (Figure S2K). To further test whether the temperature is important for the differential responses of sighing and breathing in hypoxia, we conducted the same hypoxia challenges at increased environmental temperatures. The increase in sigh rate remained more dramatic than other respiratory parameters at all tested temperatures (Figures S2A–S2J), supporting the conclusion that sighing is preferentially induced in hypoxia.

In summary, there are three characteristics in the hypoxia-induced sighing that are distinct from other respiratory parameters. First, the sigh rate increased at a much larger magnitude than HVR. Second, from the 15% O2 to the 10% O2 condition, the sigh rate dramatically increased further although the breathing frequency did not change much. Third, the breathing frequency and TV returned toward the baseline faster than the sigh rate in the hypoxia depression phase. Therefore, these results suggest that increased sighing is a robust respiratory response to hypoxia, and the hypoxic responses in sighing and breathing are likely regulated by different mechanisms.

Sigh increases oxygen saturation

Although sighing plays an essential role in various physiological and emotional processes,20,21 its function in regulating blood oxygen has not been elucidated. Because hypoxia strongly increases sighing, we decided to examine the function of sighs in regulating blood oxygen level by recording the breathing phenotype and oxygen saturation (SpO2) of the mouse. Mice were placed in the whole-body plethysmograph chamber with the collar oximeter sensor connected to monitor the real-time SpO2. In normoxia, the SpO2 immediately increases after a sigh (Figure 1M). In the 10% O2 condition, although the baseline SpO2 was dramatically reduced, a sigh is capable of increasing SpO2 to a similar extent as in the normoxia condition (Figure 1N). These results suggest that sighing is an effective breathing pattern variant to temporally increase the SpO2.

NTS Grp neurons receive afferent input of hypoxia from the carotid body

Because sighing and breathing are regulated differently in hypoxia, we next set out to identify the neural circuit controlling the hypoxia-induced sighing. The acute hypoxia response is initiated by the carotid bodies, the major oxygen sensors in the periphery that contain the glomus cells that sense the oxygen level of the blood.5,22 These glomus cells are innervated by the neurons in the petrosal ganglion, which is fused with the nodose ganglion to form the nodose-jugular-petrosal (NJP) ganglion complex in mice.23–25 In hypoxia, the stimulus is transmitted by the ganglion neurons from the carotid bodies to a sensory relay center in the brain—the nucleus of the solitary tract (NTS), where the hypoxic signal is further transmitted to the brain to regulate hypoxic responses, including ventilation.18 We recently identified that the bombesin-like neuropeptide pathways are essential for basal and hypoxia-induced sighing.15 One of the bombesin-like neuropeptides, gastrin-releasing peptide (Grp), is expressed in a subset of the NTS neurons (Figures S3A and S3B). Therefore, we hypothesized that Grp neurons in the NTS may play a role in mediating hypoxia-induced sighing.

We reasoned that if NTS Grp neurons are essential for hypoxia-induced sighing, they must receive afferent inputs from the oxygen chemoreceptors in the carotid bodies. To test this hypothesis, we first retrogradely traced the NJP ganglion neurons that project to the carotid body. A subset of neurons in the NJP ganglion were labeled by a retrograde tracer injected into the carotid body (Figure S3C), which was absent after carotid sinus denervation (Figure S3D), suggesting the specificity of the retrograde labeling. Next, we combined the retrograde tracing from the carotid body with the rabies monosynaptic tracing from the NTS Grp neurons (Figures 2A and 2B). We injected the helper adeno-associated viral vector (AAV) AAV-CMV-FLEX-TVA-mCherry-2A-oG26 into the NTS of Grp-Cre mice to express these helper genes (i.e., TVA receptor and glycoprotein) in a Cre-dependent manner. After 3 weeks of expression of the helper genes, we injected the pseudotyped rabies viruses SADΔG-EGFP (EnvA)27 into the NTS and Alexa Fluor 555-conjugated wheat-germ agglutinin (WGA-555) into the carotid body, respectively (Figures 2A and 2B). Five days later, the brain and ganglion tissue were collected for analysis. Multiplex fluorescent in situ hybridization showed that around 80% Grp neurons were labeled by the helper AAV, and most of the starter neurons (double positive for Gfp and mCherry) were Grp positive (Figures S3E–S3G). Furthermore, a subset of NJP ganglion neurons were labeled by both WGA-555 and GFP, most of which are on the nodose/petrosal side of the NJP ganglion, which is labeled by the expression of the epibranchial placode derivative marker Phox2b (Figure 2C). These results suggest that NTS Grp neurons receive direct afferent inputs from the NJP ganglion neurons that innervate the carotid body.

Figure 2. NTS Grp neurons receive afferent inputs from the carotid body.

Figure 2.

Open in a new tab

(A and B) Scheme for tracing the NJP ganglion neurons that connect the carotid body to NTS Grp neurons.

(C) NJP ganglion slice showing the retrogradely labeled neurons (double positive for WGA555 and EGFP) in the petrosal ganglion. Many of these neurons are labeled by a nodose ganglion marker Phox2b. (C1–C4) Enlarged boxed region from (C). Scale bars, 200 μm (C) and 20 μm (C4).

(D and E) In situ hybridization experiments show the NTS brain region from mice after control (D) and hypoxia conditions (E). The probe for Grp (green) and c-Fos (red). Blue, DAPI; AP, area postrema. Scale bars, 200 μm (D and E) and 50 μm (D′ and E′).

(F) Quantification showing the numbers of c-Fos-positive neurons in the control group and the hypoxia challenge group. n = 3; *p < 0.05 (unpaired t test).

(G) Quantification showing the numbers of c-Fos and Grp double positive neurons inthe control group and the hypoxia challenge group. n = 3; *p < 0.05(unpaired t test). See also Figure S3.

To examine whether this carotid body-NJP ganglion-NTS (Grp neurons) circuit responds to hypoxia, we immunostained mouse brain sections with the immediate early gene product c-Fos after the hypoxia challenge. Under the control condition, only a few Grp neurons expressed c-Fos (Figure 2D). However, 15 min after a 10-min hypoxia challenge with 10% O2, there was a 2-fold increase in the number of Grp neurons that were positive for c-Fos in the NTS (Figures 2E–2G). These results suggest that NTS Grp neurons are activated by the hypoxia challenge.

We therefore conclude that the carotid body sends the hypoxia signal to the NTS Grp neurons through the NJP ganglion neurons.

Activation of NTS Grp neurons induces sighing

To examine the causal relationship between the activation of Grp neurons and induced sighing, we first activated the NTS Grp neurons by optogenetics in anesthetized mice. The AAV vector AAV-EF1α-DIO-hChR2(H134R)-EYFP was injected into the NTS of Grp-Cre mice. After 3–4 weeks to allow the expression of channelrhodopsin-2 (ChR2), the NTS Grp neurons were activated by a blue light laser (Figure 3A). Fluorescent in situ hybridization showed that the viral vector was expressed mainly in the Grp neurons in the NTS (Figures S4K and S4L). Upon laser activation for 5 s, sighs were induced as a bimodal large breath coupled with increased diaphragm muscle activity (Figure 3B), reminiscent of the basal and hypoxia sighs.15 More than one sigh was induced in 67% of the trials, and the first sigh was induced in the first 500 ms in 78% of the trials (Figure 3C). The photoactivation-induced response is specific to induced sighing, with little effect on the breathing frequency (Figure 3B). Because the Grpr neurons in the preBötC are essential for sighing,15 we traced the projections of the NTS Grp neurons and found the eYFP-labeled processes in the preBötC (Figure 3D). Photoactivation of these terminals in the preBötC also induced sighs, with little effect on breathing frequency (Figures 3E and 3F), although the response was slightly milder than the cell body activation (compare Figures 3F and 3C). Such effects were not observed in the Grp-Cre mice injected with the control AAV (Figures S4A–S4F). Therefore, these results suggest that the activation of NTS Grp neurons specifically triggers sighs through their projections to the preBötC.

Figure 3. Activation of Grp neurons in the NTS induces sighing.

Figure 3.

Open in a new tab

(A) Schematic for optogenetic activation of NTS Grp neurons. The AAV-DIO-ChR2-eYFP was stereotactically injected into the NTS of Grp-Cre mice. The optic fiber was implanted above the injection site to activate the neurons. NTS, the nucleus of the solitary tract. Scale bars, 200 μm.

(B) The breathing trace of anesthetized mice across the photostimulation for 5 s (blue line). Please note the sigh induced by the photostimulation. f, breathing frequency; Dia, diaphragm activity; !Dia, integrated diaphragm activity.

(C) Top: raster plots of sigh (tick marks) across photostimulation (blue line, 10-ms duration per activation) in 20 trials. Bottom: the probability bar graph showing the number of induced sighs in the 5-s photostimulation.

(D) Schematic for optogenetic activation of the processes of NTS Grp neurons in the preBötC. The AAV-DIO-ChR2-eYFP was stereotactically injected into the NTS of Grp-Cre mice. The optic fiber was implanted above the preBötC region to activate the processes of Grp neurons. AMB, nucleus ambiguus; preBötC, the preBötzinger complex. Scale bars, 200 μm (50 μm in the inset).

(E) The breathing trace of anesthetized mice across the photostimulation for 5 s (blue line), note the sigh induced by the photostimulation. f, breathing frequency; Dia, diaphragm activity; !Dia, integrated diaphragm activity.

(F) Top: raster plots of sigh (tick marks) across photostimulation (blue line, 10-ms duration per activation) in 20 trials. Bottom: the probability bar graph showing the number of induced sighs in the 5 s of photostimulation.

(G) The breathing trace of a freely moving mouse showing the induced sigh in 5 s of the photostimulation (blue line).

(H) Quantification showing the sigh numbers in 10 min of the Grp-Cre mice injected with AAV-hSyn-DIO-mCherry (mCherry) (n = 9) and AAV-hSyn-DIO-hM3Dq-mChery (hM3Dq) (n = 8), injected into the NTS after the injection of CNO or saline. *p < 0.05; ns, not significant (paired t test). See also Figure S4.

We next examined the effect of NTS Grp neuron activation in freely moving animals. 3 weeks after the injection of AAV-EF1α-DIO-hChR2(H134R)-EYFP and the implantation of fiber optic in the NTS, the Grp-Cre mouse was placed in the whole-body plethysmograph chamber with the optic cable connected. A sigh was induced during photoactivation (Figure 3G), pheno-copying the response in the anesthetized condition. To continuously activate the Grp neurons in freely moving mice, we next used a chemogenetic approach, with designer receptors exclusively activated by designer drugs (DREADDs).28 hM3Dq is an activating DREADD receptor that can turn on neuronal activity upon binding to the otherwise inert ligand, clozapine N-oxide (CNO).28 AAV-hSyn-DIO-hM3Dq was bilaterally injected into the NTS of Grp-Cre mice to express hM3Dq in NTS Grp neurons in a Cre-dependent manner. Three weeks later, the mouse was injected with CNO and placed in the whole-body plethysmo-graph chamber. The sigh rate was significantly increased upon CNO injection (Figure 3H). Thus, activation of NTS Grp neurons is sufficient to induce sighing.

Ablation or inhibition of NTS Grp neurons diminishes hypoxia-induced sighing

To examine whether NTS Grp neurons are required for hypoxia-induced sighing, we first genetically ablated these neurons by stereotactically injecting the AAV that encodes caspase-3 (AAV-DIO-taCasp3)29 to the bilateral NTS of Grp-Cre mice (Figure 4A). Three to four weeks later, around 75% of Grp neurons were ablated by caspase-3 viral vector (Figures S5S–S5V). Mice were exposed to hypoxia (10% O2 balanced by N2) for 10 min and their response was compared with that prior to ablation. The mice with ablation of the Grp neurons exhibited reduced sighing in 10% O2 in comparison with their pre-ablation condition (Figures 4B and 4C), while the mice injected with the control AAV (AAV-hSyn-DIO-mCherry) showed no difference in the hypoxia-induced sighing after the injection (Figures 4C and S5F–S5I). The reduced sighing was pronounced in the initial 5 min of hypoxia but not the second 5 min (Figure 4E). Furthermore, neither the sigh rate in normoxia (Figure 4D) nor other ventilatory responses in hypoxia (Figures 4F and 4G) was affected after the ablation of the NTS Grp neurons. Similar effects were observed when the animals were exposed to 15% O2 (Figures S5A–S5E). These results suggest a specific function of the NTS Grp neurons in mediating hypoxic sighing.

Figure 4. NTS Grp neurons are required for hypoxia-induced sighing.

Figure 4.

Open in a new tab

(A) Schematic for genetic ablation of NTS Grp neurons. AAV-DIO-taCasp3 was stereotactically injected into the bilateral NTS of Grp-Cre mice.

(B) The number of sighs (bin 1 min) in mice before (pre-ab, blue) and after (post-ab, red) the ablation of Grp neurons, before, during, and after a 10-min hypoxia challenge (n = 12). *p < 0.05 (paired t test).

(C) Quantification of sighs in 10-min hypoxia in mice before (pre-op, blue) and after (post-op, red) the injection of control AAV (control, n = 8) or AAV-DIO-taCasp3 (ablation, n = 12). *p < 0.05; ns, not significant (paired t test).

(D) Quantification of sighs in 10-min normoxia in mice (n = 12) before (pre-ab, blue) and after (post-ab, red) the ablation of Grp neurons. ns, not significant (paired t test).

(E) Quantification of sighs in the first and last 5 min of hypoxia in mice (n = 12) before (pre-ab, blue) and after (post-ab, red) the ablation of Grp neurons. *p < 0.05; ns, not significant (paired t test).

(F) Breathing frequency (frequency) in 10 min of normoxia in mice (n = 12) before (pre-ab, blue) and after (post-ab, red) the ablation of Grp neurons. ns, not significant (paired t test).

(G) Tidal volume (TV) in 10 min of normoxia in mice (n = 12) before (pre-ab, blue) and after (post-ab, red) the ablation of Grp neurons. ns, not significant (paired t test).

See also Figure S5.

We next chemogenetically inhibited NTS Grp neurons through a chemogenetic approach. AAV-hSyn-DIO-hM4Di was injected bilaterally into the NTS of Grp-Cre mice (Figure 5A). 90% of Grp neurons were labeled by the AAV construct (Figures S5W and S5X). Four weeks after the viral injection, mice were injected with CNO intraperitoneally to acutely silence the Grp neurons. After CNO-induced silencing of Grp neurons, the hypoxia-induced sighing in mice was significantly reduced in the hypoxia condition when compared with the control injection with saline (Figures 5B and 5C). Distinct from the chronic ablation of Grp neurons (Figure 4E), the diminished sighing was maintained throughout the hypoxia challenge (Figure 5E). Like genetic ablation, acute silencing of Grp neurons had little effect on the basal sighing (Figure 5D) or hypoxia-induced ventilatory responses (Figures 5F and 5G). Similar effects were observed when the animals were exposed to 15% O2 (Figures S5J–S5N), and no difference in any respiratory parameters was observed in the mice injected with the control AAV (Figures 5C and S5O–S5R).

Figure 5. Inhibition of NTS Grp neurons diminishes hypoxia-induced sighing.

Figure 5.

Open in a new tab

(A) Schematic for chemogenetic inhibition of NTS Grp neurons. AAV-hSyn-DIO-hM4Di was stereotactically injected into the bilateral NTS of Grp-Cre mice.

(B) The number of sighs (bin 1 min) in Grp-Cre mice (mean ± SEM, n = 7) injected with AAV-hSyn-DIO-hM4Di into the NTS and treated with saline (blue) or CNO (red) before, during, and after 10-min hypoxia challenge. *p < 0.05 (paired t test).

(C) Quantification of sighs in 10-min hypoxia in mice injected with AAV-hSyn-DIO-hM4Di (hM4Di, n = 7) and control AAV (control, n = 9) and treated with saline (blue) or CNO (red). *p < 0.05; ns, not significant (paired t test).

(D) Quantification of sighs in 10-min normoxia in mice injected with hM4Di expression in Grp neurons (hM4Di, n = 7) treated with saline (blue) or CNO (red). ns, not significant (paired t test).

(E) Quantification of sighs in first and last 5 min of hypoxia in mice (n = 7) with hM4Di expression in Grp neurons treated with saline (blue) or CNO (red). **p < 0.01; ns, not significant (paired t test).

(F) Breathing frequency (frequency) in 10-min hypoxia in mice (n = 7) with hM4Di expression in Grp neurons treated with saline (blue) or CNO (red). ns, not significant (paired t test).

(G) Tidal volume (TV) in 10-min hypoxia in mice (n = 7) with hM4Di expression in Grp neurons treated with saline (blue) or CNO (red). ns, not significant (paired t test).

See also Figure S5.

To test whether the effect is mediated by the projections of NTS Grp neurons to the preBötC, we optogenetically inhibited their terminals using the parapinopsin (PPO), an inhibitory GPCR-based opsin.30 AAV-EF1α-DIO-PPO-Venus was injected bilaterally into the NTS of Grp-Cre mice. Four weeks later, bilateral inhibition of the terminals of NTS Grp neurons in the preBötC was sufficient to diminish the hypoxia-induced sighing (Figures S4G and S4H). This effect was not observed in the Grp-Cre mice injected with the control AAV (Figures S4I and S4J).

Altogether, these results suggest that NTS Grp neurons have a specific function in mediating hypoxia-induced sighing through their projections to the preBötC.

Elevated heart rate compensates for reduced sighing in hypoxia

Because sighing increases the SpO2 level in hypoxia (Figures 1M and 1N), we next tested the SpO2 level of the mouse in hypoxia after sighing is inhibited. To focus on the initial phase of hypoxia, the mice were exposed to 3 min of hypoxia when the NTS Grp neurons were acutely silenced by chemogenetics. To our surprise, there was no change in SpO2 in the hypoxia condition after silencing the NTS Grp neurons (Figure 6A), although the sigh rate was reduced (Figure 5E). We next set out to search for the possible physiological function that could compensate for the reduced sighing. Consistent with our early result (Figure 5C), the breathing frequency was not affected (Figures 6B and 6C), indicating that the SpO2 is not compensated by increased ventilation.

Figure 6. Elevated heart rate compensates for reduced sighing in hypoxia.

Figure 6.

Open in a new tab

(A) The oxygen saturation in normoxia and hypoxia of mice with hM4Di expression in Grp neurons (n = 12) treated with saline (Saline-hM4Di) or CNO (CNO-hM4Di). ns, not significant (paired t test).

(B) Breathing frequency in normoxia and hypoxia of mice with hM4Di expression in Grp neurons (n = 12) treated with saline (Saline-hM4Di) or CNO (CNO-hM4Di). ns, not significant (paired t test).

(C) Hypoxia-induced breathing frequency increase (Δ frequency) of mice with hM4Di expression in Grp neurons (n = 12) treated with saline (Saline-hM4Di) or CNO (CNO-hM4Di). ns, not significant (paired t test).

(D) Heart rate in normoxia and hypoxia of mice with hM4Di expression in Grp neurons (n = 12) treated with saline (Saline-hM4Di) or CNO (CNO-hM4Di). *p < 0.05; ns, not significant (paired t test).

(E) Hypoxia-induced heart rate increase of mice with hM4Di expression in Grp neurons (n = 12) treated with saline (Saline-hM4Di) or CNO (CNO-hM4Di). **p < 0.01 (paired t test).

See also Figure S6.

Acute hypoxia triggers an increased heart rate to increase cardiac output to maintain the blood oxygen level.31 Indeed, the heart rate dramatically increases in a hypoxia challenge (Figure 6D). Importantly, the heart rate increase in hypoxia is significantly higher in mice with CNO-induced silencing of Grp neurons, at 75 ± 6.1 bpm compared with 52 ± 4.6 bpm in the control treatment using saline (Figure 6E). No difference was observed in mice injected with the control virus (Figure S6). These results indicate that the elevated heart rate in hypoxia could help maintain the SpO2, despite the diminished number of sighs.

DISCUSSION

The HVR is a vital function of the body that compensates for the reduced oxygen concentration in the arterial blood. Our results show that elevated sighing is one of the characteristic ventilatory responses to hypoxia, which increases the blood oxygen level. These sighs are controlled by a carotid body-NJP-NTS-preBötC circuit: the NTS Grp neurons receive afferent inputs from the carotid body, and they are activated by acute hypoxia. The activation of these NTS Grp neurons induces sighing via their projections in the preBötC region. Genetic ablation or acute silencing of NTS Grp neurons is sufficient to diminish hypoxia-induced sighing, with little effect on basal sighing or other HVRs. Therefore, our results demonstrate a neural circuit connecting the carotid body to the preBötC for hypoxia-induced sighing, which is an important respiratory function to increase the blood oxygen.

To combat the hypoxic environment, the body has multiple defense mechanisms, including elevated respiration (increased breathing and sighing), increased heart rate and blood pressure, and the redistribution of the blood to the essential organs.32 HVR is a critical function that increases pulmonary ventilation in response to lowered oxygen in the blood.4 Here, we showed that increased sighing is an immediately and preferentially induced ventilatory response to hypoxia, which can increase the blood oxygen level. Furthermore, sighing is more sensitive to hypoxia severity and increases to a higher magnitude than the respiratory rate. This increased sighing also outlasts the other ventilation increase. Therefore, inducing sighing in patients could perhaps be beneficial and efficient in restoring the blood oxygen level. Furthermore, mice exhibit a relatively similar blood oxygen level in hypoxia after the hypoxia-evoked sighs are inhibited. This is likely due to the body having redundant defense mechanisms to maintain blood oxygen because of its vital role in survival. Indeed, the elevated heart rate in hypoxia is greater in the animals with diminished sighing.

Sighing is a critical breathing pattern that occurs spontaneously every several minutes, and these basal sighs are thought to prevent the collapse of alveoli and maintain normal lung function.15,20,21 Because sighing is also associated with a number of behavioral and physiological conditions, it makes perfect sense that sighing has multiple functions in physiology and emotions. For instance, we recently reported that sighing is induced by claustrophobic stress.17 In addition to the capability to increase the SpO2 in hypoxia, as shown here, the hypoxia-induced sigh is also suggested to be important for arousal, the failure of which is thought to be the cause of death in sudden infant death syndrome.33 Therefore, further investigation on the hypoxia sigh and its neural circuit in humans would be impactful to understanding the role of hypoxia-induced sighing in pathological conditions.

Our previous work showed that the bombesin-like peptide receptors (i.e., Grpr and Nmbr) and the preBötC neurons that express these receptors are essential for sighing.15 Based on this, we hypothesize that the Grp neurons in the NTS release GRP peptide in hypoxia to activate the preBötC Grpr neurons to transform eupneic breaths into sighs, although we cannot rule out the contribution of other neurotransmitters in this process. The identification of this hypoxia sigh circuit also sheds new light on the organization and design of neural circuits. This current work, together with our recent discovery of the neural circuit for claustrophobic stress-induced sighing,17 further suggests that distinct inputs converge to the same sigh control neurons in the preBötC15 to regulate sighing in different conditions. Therefore, these works provide an appealing model for a neural circuit design in which various upstream inputs recruit the same neural circuit module or functional unit for a shared motor output.

One intriguing aspect of our findings is that this neural circuit has a specific function in mediating hypoxia-induced sighing, but not the HVR, i.e., the increased breathing frequency and TV. HVR is known to be triggered by the carotid body, in that animals with carotid body denervation do not exhibit increased ventilation.34,35 Although several brain regions are also implicated in the hypoxia response, the neural circuit underlying HVR remains elusive.4 The carotid body to brainstem circuit we identified here is not functional in mediating HVR. This is because the photoactivation of Grp neurons induces sighing with little effect in eupneic breathing, while the ablation or silencing of these neurons diminishes hypoxia-induced sighing but not increased respiratory rate or volume. This is consistent with our observation that hypoxia-induced sighing exhibits different responses in magnitude, duration, and dependence to hypoxia severity in comparison with other ventilatory parameters. The functional specificity of this neural circuit indicates that there could be other defined neuronal populations and neural circuits with specific functions in controlling breathing pattern variants. Acute silencing of NTS Grp neurons diminishes hypoxia-induced sighing in both the initial and later phases of acute hypoxia, but chronic genetic ablation of these neurons results in a more pronounced deficiency only in the initial phase of hypoxia. These results suggest that there may be a redundant mechanism that contributes to hypoxia-induced sighing after the chronic suppression of the circuit identified in this study. Therefore, it would be important to comprehensively define the functional neural circuits for the respiratory responses in hypoxia, including HVR, that connect the carotid body to the brainstem.

NTS is a critical relay center of the brain that receives diverse interoceptive signals from the visceral organs, including the hypoxia stimulus sensed by the carotid body. Experiments on brainstem slice physiology suggest that astrocytes in the preBötC intrinsically sense hypoxia and enhance respiration, including sighing.21 However, in intact animals, perfusion of the carotid bodies with hypoxic fluid or the blood of a hypoxic donor animal induces sighs,36,37 suggesting an important role for the carotid body in hypoxia-induced sighing. Furthermore, carotid body denervation is sufficient to eliminate hypoxia-induced sighing.14 Consistent with these in vivo studies, our results here suggest a critical role for the peripheral afferent pathway via the NTS in regulating sighing in hypoxia. In addition to hypoxia, sighing is also known to be triggered by other peripheral stimuli, such as a change in lung volume, which is sensed by pulmonary mechanoreceptors. Further investigation is needed to reveal where the pulmonary inputs for sighing are connected to the circuitry for sighing. In summary, our work provides a paradigm to further elucidate the pathway and circuit underlying other interoceptive processes, including other aspects of HVR.

In this study, we integrated neural circuit tracing and neurogenetic manipulations of genetically defined neurons to reveal a circuit mechanism for hypoxia-induced sighing. Although this is a powerful strategy to dissect the functional neuronal circuits, there are potential limitations because such experiments strongly depend on the viral targeting approach and the transgenic mouse line to precisely label and manipulate neurons. For instance, we found that increased sighing in hypoxia is not eliminated after inhibition or ablation of the Grp neurons. This could be the consequence of the incomplete transduction efficiency of viral labeling in the transgenic mouse. In addition, Grp mRNA was not detected in some of the neurons labeled by viral transduction. This could be due to the dynamic expression and instability of the neuropeptide transcripts or the imperfect specificity of the transgenic mouse used in this study.

STAR★METHODS

RESOURCE AVAILABILITY

Lead contact

Further information and request for resources and reagents should be directed to and will be fulfilled by the lead contact, Peng Li (penglium@umich.edu).

Materials availability

This study did not generate any new unique reagents or materials to report.

Data and code availability

  • All data reported in this paper will be shared by the lead contact upon request.

  • The code used in this paper will be shared by the lead contact upon request.

  • Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Animals

All procedures were carried out in accordance with the animal care standards in National Institutes of Health (NIH) guidelines and approved by the University Committee on Use and Care of Animals at the University of Michigan. Experiments were performed on 6 to 12-week-old mice maintained in a 12-hour light/dark cycle with free access to food and water. All mice strains utilized were maintained in C57BL/6J genetic background. Both male and female mice were used in the experiments. Grp-Cre is a transgenic mouse line (MMRRC, KH107, Stock Number: 031182-UCD), as previously reported.38

METHOD DETAILS

Breathing data recording

Individual mice were placed in the whole-body plethysmography (WBP) chamber at room temperature (22°C) and with normoxia air (21% O2 balanced with nitrogen) input at 1L/min rate. Mice were allowed to acclimate to the chamber for 30 minutes before the respiratory parameters were recorded by Emka IOX2 software (EMKA Technologies, Paris, France).

For hypoxia challenge, each individual mouse was allowed to acclimate to the chamber and be exposed to normoxia air for 30 minutes, before switched to hypoxia air (10% O2 or 15% O2 balanced by nitrogen) at the same flow rate. The data of the last 10 minutes of the normoxia period was used as the baseline. After 10 minutes in hypoxia, the chamber was changed to normoxia for another 10 minutes. During the switches, the chamber was flushed by the new gas mixture for 1 minute before data collection. The body temperature was tested by the rectal temperature probe TCAT-2LV (Physitemp Instruments, Clifton, NJ, USA) in the home cage, 10 min after normoxia, 5 min and 10 min in hypoxia.

For hypoxia challenge in different temperatures, the WBP chamber and the air were maintained at 26°C, 30°C, or 34°C by adjusting the distance of the heating lamps and the power level of the heating pad.

Stereotactic injection

The adult mice were anesthetized with isoflurane (5% for induction, 1.5% for maintenance), and injected with 5mg/kg of preemptive analgesic carprofen prior to placement in the stereotactic frame (David Kopf Instruments, Model 940) and body temperature was maintained at 36°C using a heating pad (Physitemp, TCAT-2LV). For NTS injection, the following coordinates were used: ±0.4 mm from mid-line, −7.5 mm from posterior to the bregma, and −3.45 mm deep from the brain surface. The virus injected at a rate of 50 nL/min. After injection the glass pipette was left in the brain for 10 min to prevent backflow.

Optogenetic manipulation

For optogenetic activation of the NTS Grp neurons, 400 nL of adenoassociated viral (AAV) vector AAV-EF1α-DIO-hChR2(H134R)-EYFP was stereotactically injected into the NTS of the Grp-Cre mice. AAV-EF1α-DIO-EYFP was injected as the control. For terminal inhibition, AAV-EF1α-DIO-PPO-Venus30 was stereotactically injected into the bilateral NTS of the Grp-Cre mice. Fiber optic cannulas (200 μm diameter) with 1.25 mm ceramic ferrule were implanted 200–500 μm above the virus injection site (for cell body activation) or the preBötC (for terminal activation or inhibition) in the same surgery. The cannulas were stabilized to the skull surface by Metabond (Parkell), and the incision was sealed by absorbable surgical sutures (ETHICON). After 3–4 weeks of recovery, the cannula was connected to a 473 nm laser (10 mW), and the breathing responses to optogenetic activation were tested.

For awake condition, mice were placed in a whole-body plethysmograph chamber (Buxco) and the breathing parameters were recorded by Emka IOX2 software (EMKA Technologies, Paris, France). For the anesthetized condition, mice were injected with 1.6 g/kg urethane (Sigma), before placed on a stereotaxic frame (David Kopf Instruments) with the heating pad (Physitemp, TCAT-2LV). A tracheostomy tube was placed in the trachea through the larynx. Airflow was detected by a flowhead connected to Powerlab system (AD instruments). For the diaphragm electromyogram, wire electrodes (Cooner Wire) were connected to the diaphragm. The sampling rate was 1000 Hz. EMG signal was digitally filtered using a band-pass filter (25–300 Hz) and integrated (time constant of 0.1 s) by Labchart (AD instruments).

Chemogenetic manipulation

For chemogenetic manipulation of NTS Grp neurons, the following viral vectors were used: AAV-hSyn-DIO-hM3Dq-mCherry (activation), AAV-hSyn-DIO-hM4Di-mCherry (inhibition), and AAV-hSyn-DIO-mCherry (control).28 400 nL of AAV was stereotactically injected into bilateral NTS of the isoflurane-anesthetized GRP-Cre mice. After 4 weeks, the clozapine-n-oxide (CNO) was intraperitoneally injected (3 mg/kg) 20 minutes before the behavior test. Saline was injected as a control.

Genetic ablation

Breathing responses of animals to hypoxia challenges were tested before the surgery. To ablate NTS Grp neurons, 400 nL of AAV-Flex-taCasp3-TEVp or AAV-hSyn-DIO-mCherry (control AAV) was stereotactically injected into bilateral NTS of the isoflurane-anesthetized GRP-Cre mice. Four weeks later, the breathing responses to hypoxia were tested again.

The oxygen saturation measurement

The oxygen saturation and the heart rate were detected by the non-invasive MouseOX instrument (STARR Life Sciences, Holliston, Ma, USA). Animals were acclimated with the collar clip for at least 20 minutes before their physiological phenotypes were recorded. The oxygen saturation and heart rate data were integrated with the breathing data by PowerLab (AD instruments).

Measurements of oxygen consumption and carbon dioxide production

O2 consumption, CO2 production and respiratory exchange rate were measured during normoxia and hypoxia in the whole-body plethysmography chamber at room temperature (22°C). Gases were collected from the inflow and outflow of the chamber, and the levels of oxygen (O2) and carbon dioxide (CO2) were analyzed by Field Metabolic System (FMS, Sable Systems, Las Vegas, NV, USA). Breathing and gas composition (O2, CO2, H2O) of chamber air flows were measured during the last 5 min of normoxia and hypoxia. Rates of O2 consumption, CO2 production and water loss were calculated as described39 and normalized to grams of body weight. The convection ventilation was calculated as total ventilation divided by O2 consumption.

Carotid body injection

The mice were anesthetized by isoflurane and placed on a surgical frame in the supine position. Muscles covering the trachea and the area of the carotid bifurcation were exposed through a mid-line neck incision. The sternocleidomastoid muscle was dissected and pulled away from the carotid artery. The bifurcations of the common carotid artery were exposed, and the superior cervical ganglion (SCG), glossopharyngeal nerve, common carotid artery (CCA), external carotid artery (ECA), and the internal carotid artery (ICA) were identified. The glass pipet loaded with tracer Alexa Fluor 555-conjugated wheat-germ agglutinin or Alexa Fluor 488-conjugated wheat-germ agglutinin (Invitrogen, W32464 and W11261) was placed on the carotid body. The tracer was injected at 80 nL/min rate and the pipet was placed in place for 5 minutes after the injection.

Rabies viral monosynaptic tracing

For monosynaptic retrograde tracing of the NTS Grp neurons, the rabies tracing strategy were used. 400 nl of AAV-CMV-FLEX-TVA-mCherry-2A-oG was stereotactically injected into NTS of GRP-Cre mice. Three weeks later, 400 nL of EnvA-pseudotyped rabies-GFP virus (RVdG) was stereotactically injected to the same location. To combine the rabies viral tracing with the carotid body tracing, Alexa Fluor 555-conjugated wheat-germ agglutinin (WGA 555) was injected into the carotid body in the same surgery. 5 days later, the brain and nodose jugular ganglion complex were harvested and analyzed.

Histology and immunostaining

Mice were perfused transcardially with 10 mL of ice-cold PBS followed by 10 mL of 4% ice-cold PFA. Brain and nodose jugular ganglion complex were dissected and post-fixed in 4% PFA at 4°C overnight and then transferred to 30% sucrose in PBS at 4°C for two days. Tissue was embedded in the optimum cutting temperature compound (OCT) and stored at −80°C until use. Tissues were sectioned at 16–30 microns using a Leica CM3050 cryostat and mounted onto Superfrost Plus microscope slides (Thermo Fisher Scientific).

For immunofluorescent staining, the sections were briefly washed by PBS with 0.1% Tween 20 (PBST), permeabilized with 0.3% PBS Triton X-100 for 30 mins, and then blocked by 2% bovine serum albumin (BSA) in PBS for 1 hr. Tissue sections were incubated with primary antibody overnight at 4 °C. The primary antibody used as: goat anti-Phox2b (Santa Cruz, sc-13224,1:150). After washing with PBST, secondary antibodies (donkey anti-goat, Invitrogen, A-21447, 1:500) were applied for 2 hours at room temperature. Then the samples were washed with PBST three times (5 min each time). Sections were stained with 4’,6 -diamidino-2-phenylindole, di-hydrochloride (DAPI, Invitrogen, D1306), before mounted prolong gold antifade reagent (Invitrogen, P36930). Fluorescent images were obtained on a Nikon A1 confocal microscope.

Hypoxia induced c-Fos analysis

The mice were subject to a hypoxia challenge in WBP chamber (30 minutes acclimation in normoxia, 10 minutes hypoxia challenge with 10% O2, and 20 minutes recovery in normoxia), and the brain tissue was immediately harvested and snap-frozen in OCT. Mice placed in WBP chamber without hypoxia challenge were used as control.

Fluorescent in situ hybridization

The brain tissue was sectioned coronally at 16 microns and collected into five sets. Every one out of five sections from the brainstem were used for the following steps. Sections were fixed in 4% paraformaldehyde, dehydrated in an ethanol series, and treated with Pretreatment Reagent (Advanced Cell Diagnostics). Multiplex fluorescent in situ assay (FISH) was performed using proprietary RNA-scope technology with probes: c-Fos, Grp, eYFP, mCherry, and Gfp (Advanced Cell Diagnostics). Fluorescent images were obtained on a Nikon A1 confocal microscope. Six sections covering the caudal NTS region (−7.3 ~ −7.8 mm posterior to the bregma) were used for quantification.

QUANTIFICATION AND STATISTICAL ANALYSIS

All data are presented as mean ± standard error of the mean (SEM). Sample size was specified in the figure legends. Blinding was used in the assays and scoring when possible. For physiological data collection, the investigator was blinded to the information of AAVs injected in the mice. For data analysis, the investigator was blinded to the information of AAVs injected in the mice, and the behavioral conditions. The collected data was first tested for normality using the Shapiro-Wilk test. Data were analyzed with ANOVA or Student’s t tests as indicated in figure legends (GraphPad Prism). P values ≤ 0.05 were considered statistically significant. Statistical significance levels were set at *p < 0.05, **p < 0.01.

Supplementary Material

Yao et al., 2023 Supplemental Information

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
Goat anti-Phox2b Santa Cruz Cat# sc-13224; RRID: AB_2283771
Goat anti-Choline Acetyltransferase Millipore Sigma Cat#AB144P; RRID: AB_2079751
Donkey anti-Goat Alexa Fluor 555 Thermo Fisher Scientific Cat# A-21432; RRID: AB_2535853
Donkey anti-Goat Alexa Fluor 647 Thermo Fisher Scientific Cat# A-21447; RRID: AB_2535864
Bacterial and virus strains
AAV2-Flex-taCasps-TEVP Titer: 4.1 ×1012 UNC N/A
AAV-hSyn-DIO-mcherry Titer: 3.9× 1012 UNC N/A
CMV-Flex-TVAmcherry-2A-Og Titer: 1.4×1013 Salk Cat# 102985
G-Deleted Rabies-eGFP Titer: 3.8×1013 Salk Cat# 32635
AAV-EF1a-DIO-EYFP Titer: 1.3×1013 Addgene Cat# 27056-AAV5
AAV-hSyn-DIO-HM3D-mcherry Titer: 1.8× 1013 Addgene Cat# 44361-AAV2
AAV-EF1a-DIO-hChR2(H134R)-EYFP Titer: 3.0× 1013 Addgene Cat# 20298-AAV5
AAV-hSyn-DIO-HM4D-mcherry Titer: 2.3×1013 Addgene Cat# 44362-AAV2
AAV-Ef1a-DIO-PPO-Venus Titer: 2.5× 1013 Addgene Cat# 139505-AAV9
Chemicals, peptides, and recombinant proteins
RNAscope Probe-Grp Advanced Cell Diagnostics Cat# 317861
RNAscope Probe-c-Fos Advanced Cell Diagnostics Cat# 316921
RNAscope Probe-mCherry Advanced Cell Diagnostics Cat# 431208
RNAscope Probe-EYFP Advanced Cell Diagnostics Cat# 312138
RNAscope Probe-EGFP Advanced Cell Diagnostics Cat# 538858
Wheat Germ Agglutinin, Alexa Fluor 488 Conjugate Invitrogen Cat# W11261
Wheat Germ Agglutinin, Alexa Fluor 555 Conjugate Invitrogen Cat# W32464
Critical commercial assays
RNAscope Multiplex Fluorescent Reagent Kit v2 Advanced Cell Diagnostics Cat# 323100
Experimental models: Organisms/strains
Mouse: Grp-Cre Zhou-Feng Chen (Yu et al.38) N/A
Software and algorithms
ImageJ NIH https://imagej.nih.gov/ij/
Adobe Illustrator CC Adobe https://www.adobe.com/
MouseOx Plus Starr Life Sciences N/A
GraphPad Prism 8.0 GraphPad N/A
Other
Optical fiber RWD Life Science Cat# R-FOC-L200C-22NA
Field Metabolic System Sable Systems N/A
TCAT-2LV animal temperature controller Physitemp Instruments N/A
Open in a new tab

Highlights.

  • Hypoxia preferentially induces sighing in mice, which increases blood oxygen level

  • NTS Grp neurons receive afferent input from carotid body and are activated in hypoxia

  • Hypoxia sighing is controlled by NTS Grp neurons via their projections to the preBötC

  • Elevated heart rate in hypoxia compensates for diminished sighing

ACKNOWLEDGMENTS

We thank C. Phillips for the help with the initial optogenetics experiment and M. Zylinski for copyediting the manuscript. We also thank B. Joos for the help with the Field Metabolic System. This work was supported by NIH grants HL156989 and AT011652, American Thoracic Society Unrestricted Grant, and Parker B. Francis Fellowship (P.L.).

INCLUSION AND DIVERSITY

We support inclusive, diverse, and equitable conduct of research.

Footnotes

SUPPLEMENTAL INFORMATION

Supplemental information can be found online at https://doi.org/10.1016/j.cub.2023.01.019.

DECLARATION OF INTERESTS

The authors declare no competing interests.

REFERENCES

  • 1.Chen WG, Schloesser D, Arensdorf AM, Simmons JM, Cui C, Valentino R, Gnadt JW, Nielsen L, Hillaire-Clarke CS, Spruance V, et al. (2021). The emerging science of interoception: sensing, integrating, interpreting, and regulating signals within the self. Trends Neurosci. 44, 3–16. 10.1016/j.tins.2020.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lutz PL, and Prentice HM (2002). Sensing and responding to hypoxia, molecular and physiological mechanisms. Integr. Comp. Biol 42, 463–468. 10.1093/icb/42.3.463. [DOI] [PubMed] [Google Scholar]
  • 3.Ottestad W, Seim M, and Mæhlen JO (2020). COVID-19 with silent hypoxemia. Tidsskr. Nor. Laegeforen 140, 1–3. 10.4045/tidsskr.20.0299. [DOI] [PubMed] [Google Scholar]
  • 4.Teppema LJ, and Dahan A (2010). The ventilatory response to hypoxia in mammals: mechanisms, measurement, and analysis. Physiol. Rev 90, 675–754. 10.1152/physrev.00012.2009. [DOI] [PubMed] [Google Scholar]
  • 5.Nurse CA, and Piskuric NA (2013). Signal processing at mammalian carotid body chemoreceptors. Semin. Cell Dev. Biol 24, 22–30. 10.1016/j.semcdb.2012.09.006. [DOI] [PubMed] [Google Scholar]
  • 6.Prabhakar NR (2016). Carotid body chemoreflex: a driver of autonomic abnormalities in sleep apnoea. Exp. Physiol 101, 975–985. 10.1113/EP085624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chang AJ, Ortega FE, Riegler J, Madison DV, and Krasnow MA (2015). Oxygen regulation of breathing through an olfactory receptor activated by lactate. Nature 527, 240–244. 10.1038/nature15721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Cherniack NS, von Euler C, G1ogowska M, and Homma I (1981). Characteristics and rate of occurrence of spontaneous and provoked augmented breaths. Acta Physiol. Scand 111, 349–360. 10.1111/j.1748-1716.1981.tb06747.x. [DOI] [PubMed] [Google Scholar]
  • 9.Haldane JS, Meakins JC, and Priestley JG (1919). The effects of shallow breathing. J. Physiol. Pharmacol. Off. J. Pol. Physiol. Soc 1–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Hartland BL, Newell TJ, and Damico N (2014). Alveolar recruitment maneuvers: are your patients missing out? AANA J. 82, 307–314. [PubMed] [Google Scholar]
  • 11.Knowlton GC, and Larrabee MG (1946). A unitary analysis of pulmonary volume receptors. Am. J. Physiol 147, 100–114. 10.1152/ajplegacy.1946.147.1.100. [DOI] [PubMed] [Google Scholar]
  • 12.Mccutcheon FH (1953). Atmospheric respiration and the complex cycles in mammalian breathing mechanisms. J. Cell. Physiol 41, 291–303. [DOI] [PubMed] [Google Scholar]
  • 13.Bartlett D (1971). Origin and regulation of spontaneous deep breaths. Respir. Physiol 12, 230–238. [DOI] [PubMed] [Google Scholar]
  • 14.Basting TM, Abe C, Viar KE, Stornetta RL, and Guyenet PG (2016). Is plasticity within the retrotrapezoid nucleus responsible for the recovery of the PCO2 set-point after carotid body denervation in rats? J. Physiol 594, 3371–3390. 10.1113/JP272046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Li P, Janczewski WA, Yackle K, Kam K, Pagliardini S, Krasnow MA, and Feldman JL (2016). The peptidergic control circuit for sighing. Nature 530, 293–297. 10.1038/nature16964. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Smith JC, Ellenberger HH, Ballanyi K, Richter DW, and Feldman JL (1991). Pre-Bötzinger complex: a brainstem region that may generate respiratory rhythm in mammals. Science 254, 726–729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Li P, Li SB, Wang X, Phillips CD, Schwarz LA, Luo L, de Lecea L, and Krasnow MA (2020). Brain circuit of claustrophobia-like behavior in mice identified by upstream tracing of sighing. Cell Rep. 31, 107779. 10.1016/j.celrep.2020.107779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Alheid GF, and McCrimmon DR (2008). The chemical neuroanatomy of breathing. Respir. Physiol. Neurobiol 164, 3–11. 10.1016/j.resp.2008.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Gautier H (1996). Interactions among metabolic rate, hypoxia, and control of breathing. J. Appl. Physiol (1985) 81, 521–527. 10.1152/jappl.1996.81.2.521. [DOI] [PubMed] [Google Scholar]
  • 20.Li P, and Yackle K (2017). Sighing. Curr. Biol 27, R88–R89. 10.1016/j.cub.2016.09.006. [DOI] [PubMed] [Google Scholar]
  • 21.Ramirez JM (2014). The integrative role of the sigh in psychology, physiology, pathology, and neurobiology. Prog. Brain Res 209, 91–129. 10.1016/B978-0-444-63274-6.00006-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Weir EK, López-Barneo J, Buckler KJ, and Archer SL (2005). Acute oxygen-sensing mechanisms. N. Engl. J. Med 353, 2042–2055. 10.1056/NEJMra050002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Kupari J, Häring M, Agirre E, Castelo-Branco G, and Ernfors P (2019). An atlas of vagal sensory neurons and their molecular specialization. Cell Rep. 27, 2508–2523.e4. 10.1016/j.celrep.2019.04.096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Prescott SL, Umans BD, Williams EK, Brust RD, and Liberles SD (2020). An airway protection program revealed by sweeping genetic control of vagal afferents. Cell 181, 574–589.e14. 10.1016/j.cell.2020.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Walker HK (1990). Cranial nerves IX and X: the glossopharyngeal and vagus nerves. In Clinical Methods: The History, Physical, and Laboratory Examinations, Walker HK, Hall WD, and Hurst JW, eds. (Butterworths). [PubMed] [Google Scholar]
  • 26.Ciabatti E, González-Rueda A, Mariotti L, Morgese F, and Tripodi M (2017). Life-long genetic and functional access to neural circuits using self-inactivating rabies virus. Cell 170, 382–392.e14. 10.1016/j.cell.2017.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Schwarz LA, Miyamichi K, Gao XJ, Beier KT, Weissbourd B, DeLoach KE, Ren J, Ibanes S, Malenka RC, Kremer EJ, et al. (2015). Viral-genetic tracing of the input-output organization of a central noradrenaline circuit. Nature 524, 88–92. 10.1038/nature14600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Krashes MJ, Koda S, Ye C, Rogan SC, Adams AC, Cusher DS, Maratos-Flier E, Roth BL, and Lowell BB (2011). Rapid, reversible activation of AgRP neurons drives feeding behavior in mice. J. Clin. Invest 121, 1424–1428. 10.1172/JCI46229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Yang CF, Chiang MC, Gray DC, Prabhakaran M, Alvarado M, Juntti SA, Unger EK, Wells JA, and Shah NM (2013). Sexually dimorphic neurons in the ventromedial hypothalamus govern mating in both sexes and aggression in males. Cell 153, 896–909. 10.1016/j.cell.2013.04.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Copits BA, Gowrishankar R, O’Neill PR, Li JN, Girven KS, Yoo JJ, Meshik X, Parker KE, Spangler SM, Elerding AJ, et al. (2021). A photoswitchable GPCR-based opsin for presynaptic inhibition. Neuron 109, 1791–1809.e11. 10.1016/j.neuron.2021.04.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Campen MJ, Tagaito Y, Li J, Balbir A, Tankersley CG, Smith P, Schwartz A, and O’Donnell CP (2004). Phenotypic variation in cardiovascular responses to acute hypoxic and hypercapnic exposure in mice. Physiol. Genomics 20, 15–20. 10.1152/physiolgenomics.00197.2003. [DOI] [PubMed] [Google Scholar]
  • 32.Robson JG (1964). The physiology and pathology of acute hypoxia. Br. J. Anaesth 36, 536–541. 10.1093/bja/36.9.536. [DOI] [PubMed] [Google Scholar]
  • 33.Garcia AJ, Koschnitzky JE, and Ramirez JM (2013). The physiological determinants of sudden infant death syndrome. Respir. Physiol. Neurobiol 189, 288–300. 10.1016/j.resp.2013.05.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Fordyce WE (1987). Hypoxic ventilatory control in the awake cat five years after carotid body resection. Respir. Physiol 69, 209–225. 10.1016/0034-5687(87)90028-4. [DOI] [PubMed] [Google Scholar]
  • 35.Long WQ, Giesbrecht GG, and Anthonisen NR (1993). Ventilatory response to moderate hypoxia in awake chemodenervated cats. J. Appl. Physiol (1985) 74, 805–810. 10.1152/jappl.1993.74.2.805. [DOI] [PubMed] [Google Scholar]
  • 36.Schmidt CF (1932). Carotid sinus reflexes to the respiratory center. Am. J. Physiol. Cell Physiol 102, 94–118. [Google Scholar]
  • 37.Winder CV (1937). On the mechanism of stimulation of carotid gland chemoreceptors. Am. J. Physiol 118, 389–398. [Google Scholar]
  • 38.Yu YQ, Barry DM, Hao Y, Liu XT, and Chen ZF (2017). Molecular and neural basis of contagious itch behavior in mice. Science 355, 1072–1076. 10.1126/science.aak9748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lighton JRB (2019). Measuring Metabolic Rates: A Manual for Scientists, Second Edition (Oxford University Press; ). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Yao et al., 2023 Supplemental Information
NIHMS1914917-supplement-Yao_et_al___2023_Supplemental_Information.pdf (22.6MB, pdf)

Data Availability Statement

  • All data reported in this paper will be shared by the lead contact upon request.

  • The code used in this paper will be shared by the lead contact upon request.

  • Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

RESOURCES