Acute hypoxia increases ventilation. After brief hypoxic exposure, a switchback to room air is accompanied by a ventilatory fall-off in the recovery phase, but ventilation remains above the pre-hypoxic baseline for a time. Post-hypoxic persistent respiratory augmentation (PHRA) is a form of neural plasticity, which is defined as a change in the neural control system based on the memory-like experience (Mitchell and Johnson, 2003). The poststimulus overshoot in ventilatory activity may even go above the stimulus level as is evident in the acute hypoxic ventilatory response (HVR) to static exercise, with the mechanism ascribed to the interaction with the cardiovascular brain control or rapid release of the volitional hypothalamic control over sustained muscle tension (Pokorski et al., 1990). Neural plasticity is essential for stabilizing respiratory control, but the underlying mechanisms are not yet well known (Eldridge and Millhorn, 1986; Dahan et al., 1995; Eldridge, 1996; Powell et al., 1998).

There are plastic interactions in relay circuits of hypoxic stimulus between peripheral chemoreceptors, among which carotid body chemoreceptors are most engaged in creating the HVR, and brain respiratory control pathways (Pamenter and Powell, 2016). The multipronged complexity of PHRA is highlighted by increased carotid chemoreceptor sensitivity due to the withdrawal of the central efferent activity component running down the sinus nerve to the carotid body (Lahiri et al., 1983). That feature has been unraveled in adaptive plasticity to chronic hypoxia but is plausibly also present in repeat acute hypoxic episodes characteristic of sleep apnea syndrome, the disease that distinctly affects brain function and increases chemoreflex sensitivity (Prabhakar, 2016).

Limited understanding of peripheral and central underliers of respiratory plasticity spurred novel lines of research, one of which is the role of transient receptor potential ankyrin 1 (TRPA1) channel. These channels participate in shaping the acute HVR (Pokorski et al., 2014). However, the channels have never been verified in carotid chemoreceptor cells and their effects on the HVR are mediated by mechanisms other than the carotid body (Pokorski et al., 2014). It has been shown that TRPA1 is localized in the chemosensitive parafacial respiratory group (pFRG/RTN) astrocytes in which hypoxia-induced TRPA1 activation facilitates exocytosis of ATP-containing vesicles (Uchiyama et al., 2020). On the basis of these findings, TRPA1 channels in astrocytes have been proposed as an oxygen sensor for respiratory control (Uchiyama et al., 2020). The proposition is in line with studies that show the role of astrocytes in brain synaptic plasticity (De Pittà et al., 2016; Schiera et al., 2020). Astrocytes are also influential for various aspects of respiratory control, including rhythm generation (Okada et al., 2012; Sheikhbahaei et al., 2018) and hypoxic and hypercapnic ventilatory responses (Gourine et al., 2010; Funk et al., 2015; Pokorski et al., 2016; Beltrán-Castillo et al., 2017; Gourine and Funk, 2017; Funk and Gourine, 2018; Sheikhbahaei et al., 2018; Guyenet et al., 2019). It has been reported that astrocytes can detect hypoxia (Tadmouri et al., 2014; Angelova et al., 2015; Fukushi et al., 2016; Onimaru et al., 2021). Therefore, we aimed to test the hypothesis that astrocytes are involved in PHRA and define the role of astrocytes, notably through TRPA1 channels, in the PHRA phenomenon. We used arundic acid (AA) as a pharmacological tool to inhibit astrocytic function in wild-type mice. We also used astrocyte Trpa1 knockout mice to investigate the role of astrocytic TRPA1 channels in PHRA. We found that the presence of active astrocytes is indispensable for the expression of PHRA, but their action is mediated by mechanisms other than TRPA1 channels.

This study investigated the role of astrocytes in the PHRA, representing short-term potentiation of respiration. The findings show that astrocytes mediate PHRA. Pharmacological blockade of astrocyte activation by AA inhibited PHRA. The knockout asTrpa1^−/− mice showed less increase in ventilation in response to hypoxia than Trpa1^f/f mice. However, the magnitude of PHRA was not attenuated in asTrpa1^−/− when compared to Trpa1^f/f mice. Our findings demonstrate the putative role of the astrocyte TRPA1 channels in hypoxia sensing, which confirms the recent findings by Uchiyama et al. (2020). We expanded the role of astrocytes to the mediation of PHRA as well. However, TRPA1 detects mild hypoxia (13%) more closely than severe hypoxia (7%; Takahashi et al., 2011; Pokorski et al., 2014). This suggests that PHRA is more likely to occur under conditions of severe hypoxia. Astrocyte-related action on the short-term PHRA occurs through yet unsettled mechanisms other than TRPA1 channels. The involvement of astrocyte TRPA1 channels has been reported in the hippocampal long-term potentiation in mice (Shigetomi et al., 2013). The contribution of these channels may vary depending on the type of brain plasticity.

In the present study, we used AA, as an inhibitory modulator of astrocyte function. AA inhibits the inflammatory response of astrocytes by reducing GFAP and S100 protein synthesis, increasing the expression of the astroglial glutamate transporter GLAST and releases the glutamate receptor antagonist kynurenic acid from astrocytes (Tateishi et al., 2002; Mori et al., 2004; Asano et al., 2005; Wajima et al., 2013; Yamamura et al., 2013; Yanagisawa et al., 2015). We have previously reported that AA delays the occurrence of seizures and prevents respiratory arrest in severe hypoxia (Fukushi et al., 2020).

The present finding of counteracting the PHRA by AA indicates that astrocytes are influential in shaping respiratory neural plasticity. Hypoxia activates the carotid body, and the information is relayed via the carotid sinus nerve to the medullary solitary tract nucleus, emanating to other respiratory regions in the brainstem and spinal cord (Guyenet, 2014). Astrocytes around the excited neurons are activated via neurotransmitters spilled from neurons. Once activated, they release gliotransmitters that in turn activate respiratory neurons responsible for the sustenance of respiratory potentiation. Of note, the hitherto mechanistic studies on respiratory neural plasticity have been explicitly focused on neurons but not on glial cells. The present study is the first to demonstrate that astrocytes mediate the neural plasticity of respiration.

The short-term potentiation of brain excitability, leading to the continuation of respiratory augmentation after the stimulus cessation, referred to as neural plasticity, has been previously reported (Eldridge, 1973, 1976; Tawadrous and Eldridge, 1974; Eldridge and Gill-Kumar, 1980; Wagner and Eldridge, 1991). The mechanisms of respiratory plasticity are also present in the spinal cord (Feldman et al., 2003; Mitchell and Johnson, 2003; Fuller and Mitchell, 2017). One of the most extensively investigated phenomena in this context is the phrenic long-term facilitation following acute intermittent hypoxia. Regarding the cellular mechanism of facilitation, the Q and S signaling cascades in the phrenic motor nucleus have been proposed, induced by activation of metabotropic receptors coupled to Gq and Gs proteins, respectively, interacting via crosstalk inhibition. The serotonin-dependent Q pathway dominates in the phrenic facilitation during mild-to-moderate hypoxia. In contrast, the S pathway is serotonin-independent and dominates during severe hypoxia (Devinney et al., 2013; Fuller and Mitchell, 2017).

Recent studies have revealed an active role of astrocytes in brain plasticity related to other than respiratory functions, with a notable reference to hippocampal memory (Magistretti, 2006; Ota et al., 2013; Croft et al., 2015; Sims et al., 2015; De Pittà et al., 2016). Astrocytes secrete synapse-modulating gliotransmitters such as glutamate, ATP, d-serine, and GABA (Jourdain et al., 2007; Henneberger et al., 2010; Takata et al., 2011; Kang et al., 2013; Shigetomi et al., 2013; Verkhratsky et al., 2016; Zorec et al., 2018; Santello et al., 2019). The regulation of postsynaptic glutamate receptors, particularly α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors, is dependent on ATP released from astrocytes. The elevation in astrocytic Ca²⁺, occurring slowly in the order of seconds, stimulates glutamate release which activates astrocytic metabotropic glutamate receptors (Agulhon et al., 2012; Navarrete et al., 2012). The classical form of neural plasticity also depends on N-methyl-d-aspartate (NMDA) receptors and Ca²⁺-dependent slow release of d-serine from astrocytes (Henneberger et al., 2010). Further, astrocytes express a variety of receptors such as acetylcholine, ATP, GABA, and endocannabinoids (Porter and McCarthy, 1997; Haydon, 2001; Charles et al., 2003).

There are an increasing number of studies referring to the functional role of astrocytes in respiratory control other than respiratory plasticity. Astrocytes in the brainstem are sensitive to hypoxia and involved in HVR (Tadmouri et al., 2014; Angelova et al., 2015; Marina et al., 2015; Fukushi et al., 2016; Pokorski et al., 2016; Rajani et al., 2018; Uchiyama et al., 2020). Astrocytes in the ventral respiratory network, including the pre-Bötzinger complex, release ATP, which increases respiratory activity during hypoxia, putatively counteracting the depressive effects of hypoxia (Gourine et al., 2005; Marina et al., 2016a; Gourine and Funk, 2017; Funk and Gourine, 2018; Rajani et al., 2018). ATP acts via P2Y₁ receptors in the pre-Bötzinger complex to increase the respiratory burst rate with increases in intracellular Ca²⁺ and glutamate release (Lorier et al., 2007; Huxtable et al., 2010). Astrocytes also are strongly involved in the central control of sympathetic activity and cardiovascular function, including systemic hypertension (Marina et al., 2016b), which are enhanced by acute and particularly repeat hypoxia episodes sensed by carotid chemoreceptors (Prabhakar et al., 2015). There is a biological plausibility that medullary astrocytes, respiratory neurons, and peripheral chemosensing intertwine with each other in shaping PHRA. Alternative study designs are needed to further explore this issue.

In the present study, AA failed to affect V_E, although RR was increased in mice receiving a high dose of AA in room air. This phenomenon suggests that AA can affect breathing, i.e., inhibition of astrocyte activation may alter breathing patterns. In line with this notion, we showed that HVR was attenuated by a high dose of AA. However, AA blunted PHRA much more, suggesting that PHRA is activity-dependent plasticity.

There may be a concern over the time-dependent stability of minute ventilation on the background of a high dose of AA. Our additional investigation revealed that minute ventilation was stable over 240min in this condition (Supplementary Figure 1). Likewise, another set of control investigations showed that hypoxia loadings repeated three times provide close reproducibility (Supplementary Figure 2).

One potential limitation of this study could be a lack of the animal’s temperature control. In the classical “closed chamber” whole-body plethysmography, tidal volume is calculated by measuring the chamber pressure based on the combined gas law stating that the ratio of the product of gas pressure and volume to the absolute gas temperature is equal to a constant (Drorbaugh and Fenn, 1955). The chamber pressure is recorded while the chamber is sealed, and the body temperature weighs in on the result (Mayer et al., 2014; Rourke et al., 2016; Baby et al., 2018). In practice, however, the body temperature changes are so small during the hypoxic challenges of a couple of minutes that they are usually neglected for the sake of simplicity (Onodera et al., 1997). In the present study, we adopted the “open flow” plethysmography in which the chamber gas is continuously suctioned at a constant flow rate during the continuous recording. We calculated the tidal volume by integrating the airflow whose changes are proportional to those in the chamber pressure (Lundblad et al., 2002). In this case, tidal volume is expressed at ambient temperature (25°C), which obviates the need for taking the animal’s body temperature. Another limitation of this study was that we failed to examine metabolic rate in knock-out mice or its potential alterations by AA, which could influence respiration. Metabolic aspects require further exploration using alternative study designs.

The ultimate purpose of this research was to refer to the mechanism of post-hypoxic short-term respiratory plasticity in unanesthetized humans, which is essential to get insights into the pathophysiology of and preventive measures for periodic breathing, e.g., sleep apnea. This purpose stemmed from the studies showing that PHRA is involved in the mitigation of periodic breathing in sleep apnea (Georgopoulus et al., 1992; Mahamed and Mitchell, 2007; Mateika and Syed, 2013; Mateika and Komnenov, 2017) and heart failure (Ahmed et al., 1994); the notion supported in a computer simulation study (Eldridge, 1996). Our results showed that astrocytes, but not the astrocytic TRPA1 channel, were involved in the development of PHRA, suggesting that the TRPA1 is engaged in shaping HVR but not PHRA. The TRPA1 channel likely plays a (patho)physiological role in acute hypoxic conditions such as an attack of bronchial asthma (Shen et al., 2012). In diseases with periodic breathing such as sleep apnea, astrocytes may contribute to its prevention by exerting PHRA. Additionally, the observation that ΔV_ERecovery1/ΔV_EHypoxia tended to be smaller in the female asTrpa1^−/− mice raises the implication of a greater role of astrocytic TRPA1 in the female gender, which requires further exploration.

In conclusion, we have provided novel aspects of PHRA’s role linking it to astrocyte activation and suggesting that this tandem arrangement contributes to respiratory stability and potentially might be influential in the prevention of periodic breathing. However, caution should be exercised in the translation of animal findings to human settings before further exploratory research. We conclude that astrocytes mediate the post-hypoxic persisting respiratory augmentation by mechanisms other than the hitherto recognized role of TRPA1 channels in hypoxia sensing.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by Ethics Committee for Animal Experiments of the Murayama Medical Center.

IF conceived and designed the study, performed the animal experiments, analyzed the data, and drafted the manuscript. KT performed the statistical analysis and drafted the manuscript. MP edited and revised the manuscript. YK, MY, and YH participated in the design of the study. AN and YM provided the animals and revised the manuscript. HO supervised the experiments and revised the manuscript. YO conceived and designed the study, analyzed the data, and revised the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by JSPS KAKENHI (17K08559, 18K17783, 19K17386, 19K17620, 20K19368, and 20K19474) and the Japanese Physical Therapy Association (JPTA2019 and JPTA2020).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.