3 potassium channel double knockout mice

Accepted Manuscript Title: Abnormal respiration under hyperoxia in TASK-1/3 potassium channel double knockout mice Authors: Philipp K. Buehler, Doris Bleiler, Ines Tegtmeier, Dirk Heitzmann, Christian Both, Michael Georgieff, Richard Warth, J¨org Thomas PII: DOI: Reference:

S1569-9048(17)30076-9 http://dx.doi.org/doi:10.1016/j.resp.2017.06.009 RESPNB 2827

To appear in:

Respiratory Physiology & Neurobiology

Received date: Revised date: Accepted date:

8-3-2017 23-6-2017 24-6-2017

Please cite this article as: Buehler, Philipp K., Bleiler, Doris, Tegtmeier, Ines, Heitzmann, Dirk, Both, Christian, Georgieff, Michael, Warth, Richard, Thomas, J¨org, Abnormal respiration under hyperoxia in TASK-1/3 potassium channel double knockout mice.Respiratory Physiology and Neurobiology http://dx.doi.org/10.1016/j.resp.2017.06.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Abnormal respiration under hyperoxia in TASK-1/3 potassium channel double knockout mice Philipp K. Buehler1#, Doris Bleiler2,3#, Ines Tegtmeier2, Dirk Heitzmann2,4, Christian Both1, Michael Georgieff5, Richard Warth2, Jörg Thomas1* 1 University children’s hospital, Steinwiesstr. 75, CH-8032 Zürich, Switzerland Institute of Physiology, University of Regensburg, D-93053 Regensburg, Germany 3 Department of Anaesthesia, University Hospital Regensburg, 93042 Regensburg, Germany 4 Universitätsmedizin Mannheim, V. Medizinische Klinik, Theodor-Kutzer-Ufer 1-3, 68167 Mannheim, Germany 5 Institute of Anesthesiology, University of Ulm, D-89081 Ulm, Germany 2

*Corresponding Author: Dr. Jörg Thomas University children’s hospital Steinwiesstr. 75 CH-8032 Zürich Switzerland [email protected] #

Contributed equally to this work

1

Abstract Despite intensive research, the exact function of TASK potassium channels in central and peripheral chemoreception is still under debate. In this study, we investigated the respiration of unrestrained TASK-3 (TASK-3-/-) and TASK-1/TASK-3 double knockout (TASK-1/3-/-) adult male mice in vivo using a plethysmographic device. Ventilation parameters of TASK-3-/- mice were normal under control condition (21% O2) and upon hypoxia and hypercapnia they displayed the physiological increase of ventilation. TASK-1/3-/- mice showed increased ventilation under control conditions. This increase of ventilation was caused by increased tidal volumes (VT), a phenomenon similarly observed in TASK-1-/- mice. Under acute hypoxia, TASK-1/3-/- mice displayed the physiological increase of the minute volume. Interestingly, this increase was not related to an increase of the respiratory frequency (fR), as observed in wild-type mice, but was caused by a strong increase of VT. This particular respiratory phenotype is reminiscent of the respiratory phenotype of carotid body-denervated rodents in the compensated state. Acute hypercapnia (5% CO2) stimulated ventilation in TASK-1/3-/- and wild-type mice to a similar extent; however, at higher CO2 concentrations (> 5% CO2) the stimulation of ventilation was more pronounced in TASK-1/3-/- mice. At hyperoxia (100% O2), TASK-1-/-, TASK-3-/- and wild-type mice showed the physiological small decrease of ventilation. In sharp contrast, TASK-1/3-/- mice exhibited an abnormal increase of ventilation under hyperoxia. In summary, these measurements showed a grossly normal respiration of TASK-3-/- mice and a respiratory phenotype of TASK-1/3-/- mice that was characterized by a markedly enhanced tidal volume, similar to the one observed in TASK-1-/- mice. The abnormal hyperoxia response, exclusively found in TASK-1/3-/- double mutant mice, indicates that both TASK-1 and TASK-3 are essential for the hyperoxia-induced hypoventilation. The peculiar respiratory phenotype of TASK-1/3 knockout mice is reminiscent of the respiration of animals with long-term carotid body dysfunction. Taken together, TASK-1 and TASK-3 appear to serve specific and distinct roles in the complex processes underlying chemoreception and respiratory control. Keywords: TASK potassium channels; ; ; , chemoreception, whole body plethysmograph, hyperoxia

2

1.

Introduction

1.1 General aspects of background potassium channels in chemo-sensing It is essential for mammals to sense changes in blood gases (i.e. O2/CO2/H+) to ensure the appropriate respiratory response to metabolic challenges (e.g. exercising) or different environments (e.g. high altitude). The so-called respiratory chemoreception can be divided into central and peripheral components: Whereas the main peripheral chemoreceptors are the carotid bodies located in the in vicinity of the two carotid arteries (Buckler, 2015), neurons of the retrotrapezoid nucleus (RTN) located in the ventrolateral medullary surface (VSM) of the brainstem were identified as essential component for central CO2 sensitivity (Guyenet, 2014). It is commonly accepted, that the central respiratory chemoreflex stimulates breathing by an increase of the brain CO2, whereas the peripheral chemoreflex activates ventilation by hypoxia in a pH-dependent manner (Kumar and Prabhakar, 2012). In this regard, the importance of the TASK background potassium channels (two-pore-domain potassium channels, K2P channels, KCNK gene family) in central and peripheral respiratory chemoreception has been demonstrated (Bayliss et al., 2015; Buckler, 2015; Kumar et al., 2015). Two-pore-domain potassium channels are almost voltage-independent and sensitive to various neurotransmitters, pharmaceutical compounds (i.e. volatile anesthetics), and physicochemical factors (pH, redox potential, O2, CO2/H+) (Goldstein et al., 2001). For some of them, namely the TASK channels (Twikrelated acid-sensitive K+ channels) a key function in central and peripheral chemo-sensitivity has been proposed (Bayliss et al., 2015; Buckler, 2015). Especially for three TASK subunits (TASK-1; TASK-2 and TASK-3), an important role in pH-dependent homeostatic reflexes like the respiratory response to changes in CO2/H+ or O2 has been shown (Gestreau et al., 2010; Kim et al., 2009; Trapp et al., 2008). 1.2 Role of TASK channels in central chemoreception In the retrotrapezoid nucleus (RTN), a cluster of Phox2b (Paired-like homeobox 2b) expressing neurons is responsible for the CO2 chemosensory function (Dubreuil et al., 2008) and some of these RTN neurons express TASK-2 channels (Gestreau et al., 2010; Wang et al., 2013). In TASK-2 knockout mice, the respiratory answer to hypercapnia is disturbed but not totally abolished (Gestreau et al., 2010). In this context, G protein activated receptor 4 (GPR4) is important for respiratory control by CO2 and mice with a double knockout for GPR4 and TASK-2 showed a nearly completely abolished CO2 response (Kumar et al., 2015). Whereas TASK-2 expression is restricted to a few groups of neurons in the mouse brainstem, TASK-1 and TASK-3 expression was found in a variety of putative respiratory neurons in the brainstem e.g. in respiratory motoneurons of the brainstem and the spinal cord (Karschin et al., 2001), in cells of the respiratory rhythm-generating pre-Bötzinger nucleus (pre-BötC), cells of the rostral ventral respiratory group (rVRG) (Washburn et al., 2002), in respiratory hypoglossal motoneurones and in the putatively chemoreceptive neurons of the locus coeruleus and serotonergic neurons of the

3

raphe nuclei (Sirois et al., 2000). Investigations in neonatal serotonergic raphe neurons in vitro suggested that TASK-1 and TASK-3 play an important role in central chemoreception (Lazarenko et al., 2010). Surprisingly, the first in vivo study showed that central chemoreception is largely preserved in TASK knockout mice (Bayliss et al., 2015). 1.3 Peripheral chemoreception by TASK channels The carotid body is composed of sensory cells (type I glomus cells) and sustentacular cells (type II). The type I glomus cells respond to hypoxia in pH-dependent way with depolarization of the cell membrane and activation of voltage-gated Ca2+ channels. With the increase of the intracellular Ca2+ in type I cells neurosecretion of mainly ATP takes place with excitation of the adjacent afferent nerves via purinergic receptors (Rong et al., 2003). The background potassium conductance of type I cells is mainly carried by TASK-1, TASK-3 and heterodimers of these two TASK channels (Kim et al., 2009; Ortega-Saenz et al., 2010). An acidosis- and/or hypoxia-induced inhibition of these TASK channels is thought to be the initial step in the signaling cascade underlying peripheral chemo-sensation (Buckler, 2015; Buckler et al., 2000). The mitochondrial complex I (MCI) dependent signals appear necessary to initiate this potassium channel inhibition. In fact, plethysmographic studies showed a slightly impaired hypoxia and hypercapnia response in TASK-1 and TASK-1/3 double knockout mice (Trapp et al., 2008) and a completely missing hypoxia response in MHI-knockout mice (Fernandez-Aguera et al., 2015). The precise role of TASK-1 and TASK-3 channels for peripheral chemoreception, however, is still a matter of debate. 1.4 Effects of oxygen on carotid body function It is widely accepted, that acute hyperoxia transiently inhibits ventilation in mammals, probably via suppressing carotid body function (Bouferrache et al., 2000). Chronic application of high inspiratory oxygen in neonatal mammals impairs the carotid body response to hypoxia or hyperoxia even throughout life (Bates et al., 2014; Bavis et al., 2013). The molecular mechanisms underlying the hyperoxia-induced impairment of carotid body function is still not fully understood, but ion channels appear to be implicated: TASK-1, TASK-3 and L-type Ca2+ channel mRNA expression was significantly down-regulated under chronic hyperoxia suggesting a potential role of these proteins in the adaptation to changes of oxygen (Kim et al., 2012). The present study aimed at investigating the effects of hypoxia, hyperoxia and hypercapnia on the respiration of TASK-1-/-, TASK-3-/- and TASK-1/3-/- mice in order to determine the in vivo relevance of these potassium channels for chemo-sensation.

4

2.

Materials and Methods

TASK-3 and TASK-1/3 potassium channel knockout mouse (TASK-3-/-/TASK-1/3-/-) The TASK-3-/- mice have been generated as described earlier (Guyon et al., 2009) and were backcrossed for seven generations into the C57Bl/6J genetic background. The TASK-3+/+ and TASK-3-/- mice were derived from C57Bl/6J heterozygous breeding pairs. For genotyping, tail biopsies were performed; DNA was extracted and tested for the presence of wild-type and mutant alleles. TASK-1/3-mice were generated by crossing TASK-3-/- and TASK-1-/- mice (Aller et al., 2005). Resulting TASK-3+/- TASK-1+/animals were further crossed, and double knockout progeny (TASK-1/3-/-) was identified by analyzing tail DNA (Guyon et al., 2009). The mice were kept on a 12:12 h light/dark cycle beginning at 8 am, and had free access to chow (standard diet) and water. The experiments were done during the light cycle (from 8 am until 8 pm). All animal experiments were performed according to the guidelines for the care and use of laboratory animals published by the US National Institutes of Health and were approved by the local councils for animal care according to the German law for animal care. For the experiments in TASK-1/3-/- mice, only adult male mice of similar age were analyzed. For the experiments in TASK-3-/- mice, adult female and male mice were investigated. The body weights of knockout mice (TASK-3-/- and TASK-1/3-/-) were similar to those of wild-type mice. Whole body plethysmography The respiratory data were obtained using whole body plethysmography for unrestrained animals (emka technologies, Paris, France) as described previously (Jungbauer et al., 2016). Briefly, the animals were free to move, had free access to water and were not anaesthetized. A constant flow pump connected to the animal chamber ensured continuous flow in the range of 0.5 ± 0.05 l/min of fresh air thereby preventing CO2 accumulation. The experiments were performed at room temperature (22 ± 1,5 C). Hypoxic and hyperoxic-hypercapnic gas mixtures were applied into the animal chamber by the above-mentioned flow pump. The O2 content was controlled by an oximeter (Oxydig, Drägerwerk, Lübeck, Germany). Prior to the experiments, animals were placed into the plethysmography chambers for a minimum of one hour to allow acclimatization. Sniffing periods, which are mostly accompanied by body movements during the experiment, were excluded from analysis by applying a cut-off for the respiratory frequency of 350 beats/min. The respiration patterns of the mice were detected by a pressure transducer coupled to an amplifier (emka technologies, France).

5

The signals were recorded and analyzed with IOX2 software (emka technologies, France). The following respiratory parameters were registered: respiratory frequency (fR), tidal volume (VT), minute ventilation (MV), peak expiratory flow (PEF), inspiratory flow (PIF), expiration time (TE), inspiratory time (TI). Volume-related respiratory parameters (VT, MV, PEF, PIF) were normalized to the body weight of the individual mice. The relaxing time (RT) represents the time for the expiratory area to decline to 36% of the total expiratory area. Calibration of the system for every chamber was performed before every experiment according to the manufacturer’s protocol. All experimental series started with a control period with room air (21 % O2) for at least 30 minutes. In the hyperoxic-hypercapnic experimental series, the gas was changed to 100 % O2. Then, the CO2 content of gas mixture was changed to 5 % CO2/95 % O2 for a period of 15 – 30 min. In the stepwise hypercapnic series, CO2 content was increased every 15 min by about 1 %. In the hypoxia series, after a control period (21% O2) for 15 - 60 min the mice were exposed to a hypoxic gas mixture (8 or 10% O2). The hypoxic gas mixture was created by adding nitrogen to room air. Data for a more detailed analysis of the hyperoxia-induced effect on respiration in TASK-1-/-, TASK-3-/- and TASK-1/3-/- mice were extracted from the hyperoxia-hypercapnia experimental series. Statistics: Analysis for parametric and nonparametric data was performed by Shapiro-Wilk test. Descriptive statistics with differences between the two genotypes are shown as median and interquartile range (IQR) or box plots with min/max-whiskers in the figures. The median (VT, MV, Ti, TE, PIF, PEF and fR) of two groups (between wild-type and TASK-3-/--, TASK-1/3-/-- or TASK-1-/- mice) were compared by ManWhitney test (unpaired). To compare respiratory parameters during control and experimental periods within one group, a repeated measures Analysis of Variance (ANOVA) or Wilcoxon Test was used (paired). A p < 0.05 was considered to be significant. Statistical analysis was performed using Prism 7.0b (GraphPad software, La Jolla, USA) and Microsoft Excel (Microsoft Office Professional Plus 2013, Redmond, USA). In the figures, the following symbols were used to indicate statistical significance: "" (p<0.05) “” (p<0.001) between wild-type and TASK-3-/- or TASK-1/3-/- mice (unpaired); “#” (p<0.05) “##” (p<0.001) significant differences between control and experimental period within a given genotype (paired).

6

3. Results 3.1 Increased tidal volume in TASK-1/3 double knockout mice The respiration of unrestrained wild-type, TASK-3-/- and TASK-1/3-/- mice was measured using a plethysmography device. Respiration in male and female TASK-3-/- mice under control conditions (21% O2) appeared normal (Fig. 1A). Male TASK-1/3-/- mice exhibited an enhanced tidal volume (VT) and increased ventilation (minute volume (MV)) under control conditions (21% O2) (Fig. 1B). Furthermore, TASK-1/3-/- mice showed decreased inspiration time (Ti) paralleled by higher expiratory peak flow (PEF) and inspiratory peak flow (PIF) indicating a modified respiratory pattern (Fig. 1B). 3.2 TASK-3-/- mice ventilation under hypoxia and hypercapnia conditions TASK-3-/- mice of both sexes showed a grossly normal response of the ventilation upon hypoxia (10% O2) (Fig. 2A/2B). A more detailed analysis revealed that TASK-3-/- mice displayed, in contrast to wildtype mice, no shorting of the expiration time (TE) during the first minutes of acute hypoxia. Moreover, the hypoxia-induced respiratory depression (after 1 hour hypoxia) seemed enhanced in TASK-3-/- mice as evidenced by a significantly prolonged TE and a lower respiratory frequency (Fig. 2). The response of TASK-3-/- mice to acute hyperoxic hypercapnia (5 % CO2), including the shortening of TE and Ti (data not shown) was normal (Fig. 3). Taken together, these data indicate that the respiratory responses of TASK-3-/- mice to hypoxia and hypercapnia were grossly normal and chemo-sensation appeared normal. 3.3 Respiratory response of TASK-1/3-/- mice under acute hypoxia Male TASK-1/3-/- mice showed a physiological small increase of the minute volume (MV) under acute hypoxia (8% O2) (Fig. 4). However, in contrast to wild-type mice this increase in ventilation was almost exclusively caused by an enhanced tidal volume (VT) and not by an increase of the respiratory frequency. Similar to TASK-3-/- mice, the expiration time (TE) of TASK-1/3-/- mice was not shortened in the first minutes of acute hypoxia (Fig. 2A/B and 4). 3.4 TASK-1/3-/- mice showed enhanced ventilation under pronounced hypercapnia Under acute hyperoxic-hypercapnia (95% O2/5% CO2) TASK-1/3-/- and wild-type mice showed a similar increase of ventilation (Fig. 5). The inspiration and expiration times of TASK-1/3-/- mice were shortened by acute hypercapnia to an extent similar to wild-type mice (data not shown). Interestingly, during a protocol with step-wise increasing hypercapnia TASK-1/3-/- mice exhibited enhanced ventilation at higher CO2 concentrations (6 % CO2) compared to wild-type mice (Fig. 6). In summary, these data are suggestive for a modified tuning curve of the CO2 response in TASK-1/3-/- mice. 3.5 TASK-1/3-/- mice exhibited a non-physiological response to hyperoxia

7

In hyperoxic-hypercapnia experiments, TASK-1/3-/- mice showed an unexpected stimulation of respiration under hyperoxia, whereas the wild-type mice exhibited a physiological transient decrease of ventilation (Fig. 5 and 6). Mean values for the experimental periods (first 15 min) summarizing the responses of TASK-1-/-, TASK-3-/- or TASK-1/3-/- mice with their wild-type counterparts are depicted in Supplementary Fig. A1/Fig. A2. The hyperoxia-induced activation of ventilation in TASK-1/3-/- mice was caused by an increased respiratory frequency (fR) and enhanced tidal volume (Supplementary Fig. A1B). By contrast, the single knockout of TASK-3 (Supplementary Fig. A1A) and TASK-1 (Supplementary Fig. A2) did not affect the global respiratory response to hyperoxia.

8

4. Discussion TASK channels have been proposed to play a role in peripheral and central chemoreception. Despite compelling in vitro evidence, the in vivo function of TASK potassium channels in the control of respiration is still a matter of debate. Here, we investigated the respiration of unrestrained TASK-3 (TASK-3-/-) and TASK-1/TASK-3 double knockout (TASK-1/3-/-) adult male mice in vivo using a plethysmographic device. 4.1 Respiratory phenotype of TASK-3-/- mice TASK-3 channels are expressed in regions of the brain stem that are involved in the generation of the respiratory rhythm (Bayliss et al., 2015). However, basal and stimulated respiration of TASK-3-/- mice was largely normal, similar to a previous study (Trapp et al., 2008). Only subtle changes were observed such as a missing reduction of TE in the first minutes of hypoxia and an enhanced respiratory depression at longer periods of hypoxia. In carotid bodies, TASK-1/3 heteromers provide the major part of the oxygen-sensitive background K+ conductance (Kim et al., 2009). The deletion of TASK-3 or TASK-1 channels appeared to have no or little effect on the hypoxia-induced neurosecretion in type I cells in vitro. This finding can be explained by the fact that homomeric forms of either TASK-1 or TASK-3 are also oxygen-sensitive (Buckler, 2015; Ortega-Saenz et al., 2010; Turner and Buckler, 2013). Additionally, an increased TASK-1 expression may mask the respiratory phenotype of TASK-3-/- mice, which has been suggested for neuronal cells of TASK-3-/- mice (Brickley et al., 2007). Therefore, a conditional knockout of the TASK-3 gene may help to decipher the distinct function of TASK-3 channels in chemoreception in vivo. In TASK-1-/- mice, a modified respiration was only observed in male mice (Jungbauer et al., 2016). In TASK-3-/- mice, however, the discrete changes of respiration were observed in both sexes. The absence of a sex-specific phenotype is in agreement with previous findings demonstrating reduced sensitivity to inhalation anesthetics and exaggerated nocturnal activity in TASK-3-/- mice of both sexes (Linden et al., 2007). 4.2 Increased tidal volume in TASK-1/3-/- mice Male TASK-1/3-/- mice exhibited increased basal ventilation caused by an enlarged tidal volume, increased peak expiratory and peak inspiratory flow, a phenotype similar to the one of TASK-1-/- mice (Jungbauer et al., 2016). Together with the subtle respiratory phenotype of TASK-3-/- mice, this suggests that the TASK-1 gene deletion is mainly causative for the phenotype of TASK-1/3-/- mice. What are possible explanations for the increased ventilation in TASK-1/3-/- mice?

9

(i) Enhanced ventilation is a consequence of increased need for oxygen, e.g. during increased activity due to increased sympathetic tone. However, there is yet no evidence for increased oxygen consumption. (ii) Impaired gas exchange in the lungs could require increased alveolar ventilation. In male TASK-1 or TASK-1/3 knock out mice lung histology, pulmonary arterial pressure and cardiovascular system seemed largely unaffected (Heitzmann et al., 2008; Manoury et al., 2011). However, in humans TASK1 mutations have been linked to pulmonary hypertension (Ma et al., 2013). Very recently, it has been reported that in contrast to humans, the TASK-1 potassium channel is not critically involved in mediating hypoxic pulmonary vasoconstriction (Murtaza et al., 2017). On the other hand, in TASK-1-/mice and TASK-1/3-/- mice the relaxing time (representing the time for the expiratory area to decline to 36% of the total expiratory area) is prolonged, which may reflect increased respiratory resistance and, thereby, impaired lung function (Jungbauer et al., 2016). Further studies are needed to obtain a more detailed picture of the role of TASK channels for lung function. (iii) Modified and adapted neuronal control of respiration. After 45 d, carotid body-denerved rats exhibited a respiratory phenotype similar to the one of TASK-1/3-/- and TASK-1-/- mice (Jungbauer et al., 2016), namely an increased ventilation induced by an enlarged tidal volume (Roux et al., 2000). Given the fact that neonatal TASK-1-/- mice display a 30-40% diminished ventilation under control (like carotid body-denerved animals) (Jungbauer et al., 2016), it is very likely that the increase of VT reflects a central compensation mechanism induced by disturbed peripheral chemoreception. 4.3 Hypoxia in TASK-1/3-/- mice: In previous reports, impaired hypoxia-induced ventilation was found in TASK-1 knockout and TASK-1/3 double knockout mice (Jungbauer et al., 2016; Trapp et al., 2008). Here, we also observed an impaired increase of the respiration frequency (fR) together with a missing shortening of the expiration time (TE) in these knockout models. However, and in contrast to Trapp et al. (Trapp et al., 2008), we found a normal hypoxia-induced increase of global ventilation (minute volume) in TASK-1/3-/- mice due to a pronounced rise in tidal volume. Like male TASK-1/3-/- mice, 45d after carotid body denervation rats showed the meticulous respiratory phenotype (adequate MV response by increased VT, very small fR increase, diminished TE shortening) under hypoxia (Roux et al., 2000). As already suggested by others, TASK-1/TASK-3 deletion in carotid bodies are probably compensated by yet unidentified mechanisms (Buckler, 2015; Ortega-Saenz et al., 2010; Turner and Buckler, 2013). The results of studies in animals with carotid body denervation suggest that the increased tidal volume in male TASK-1/3-/- mice under hypoxia represents one compensatory mechanism, which is most likely of central origin. Additionally,

TASK-1 and TASK-3 seem to be essential for the shortening of the expiration time (TE) under

10

hypoxia, because all three knock-out lines (TASK-1-/-, TASK-3-/- and TASK-1/3-/-) were unable to shorten TE under hypoxia.

4.4 Hypercapnia-stimulated respiration in TASK-1/3-/- mice The literature on the role of TASK-1 and TASK-3 channels for the respiratory response to hypercapnia is somehow controversial. Some previous studies provided evidence that neither TASK-1 nor TASK-3 channels are involved in central CO2 chemo-sensing (Bayliss et al., 2015; Mulkey et al., 2007). By contrast, Trapp et al. observed an impaired CO2-mediated stimulation of ventilation in TASK-1-/- and TASK-1/3-/- mice (Trapp et al., 2008). In the present study, TASK-1/3-/- mice exposed to 5 % CO2 responded normally. When exposed to stepwise increase in CO2, TASK-1/3-/- mice responded with a rather small relative increase at low CO2 concentrations (1-4%) but a very strong increase at high concentrations (5-6% CO2). The variable findings with regard to CO2 sensitivity might be explained by (i) different protocols for hypercapnia (hyperoxic (Mulkey et al., 2007) vs. normoxic (Trapp et al., 2008) hypercapnia) and by (ii) a mildly modified "tuning curve" for CO2 of TASK-1/3-/- mice with an attenuated response at lower CO2 concentrations and a more pronounced response at higher CO2 concentrations. One month after carotid body denervation these animals responded with an enhanced ventilation upon high CO2 concentrations (Klein et al., 1982). A crosstalk between the carotid bodies and the retrotrapezoid nucleus has been suggested (Fiamma et al., 2013). One possible explanation could be that in TASK-1/3-/- mice the central CO2 response has adapted to an impaired carotid body input. 4.5 Effect of hyperoxia in TASK knockout mouse models Upon acute hyperoxia (100% O2), male TASK-1-/- and TASK-3-/- mice responded normally and displayed a small and transient depression of respiration. Unexpectedly, TASK-1/3-/- mice exhibited stimulated ventilation under acute hyperoxia. Interestingly, Trapp et al. observed in in vitro experiments on carotid bodies of TASK-1/3-/- mice that the firing rate of single chemoafferent fibers, and the activity of the carotid sinus nerve were significantly increased at 100 % O2 compared to wild-type mice or mice with a single TASK channel knockout (Trapp et al., 2008). Based on the in vitro data of Trapp et al. (Trapp et al., 2008) and our present in vivo data, it is tempting to speculate that disturbed carotid body function due to the combined knockout of TASK-1 and TASK-3 is related to the hyperoxia-induced respiratory stimulation. The “silencing” of the carotid bodies under acute hyperoxia leading to transient hypoventilation in mammals is still not fully understood. It is conceivable, that TASK channels in the carotid bodies show a high open probability under hyperoxia and, thereby, decrease the carotid body output (Buckler et al., 2000).

11

Interestingly, in cats with deafferention of carotid bodies and rats with excision of carotid bodies, acute hyperoxia also stimulated ventilation (Miller and Tenney, 1975; Olson et al., 1988). These results suggest that under conditions with disturbed carotid body function, hyperoxia might activate respiration directly via central mechanisms. Therefore, experiments in isolated brainstem preparation in TASK-1/3-/- mice may answer the question, whether central or peripheral mechanisms are responsible for the unphysiologically stimulated respiration under hyperoxia in these mice. Most of the brainstem preparation, however work only at “control” under high oxygen concentration, which probably makes it nearly impossible to show genotype-related differences under hyperoxia. 5. Conclusion TASK-1 and TASK-3 knockout mice as well as TASK-1/3 double knockout mice are vital and display a grossly normal regulation of ventilation. A detailed analysis of the respiratory responses to hypercapnia, hypoxia and hyperoxia disclosed modified breathing patterns, especially in TASK-1 and TASK-1/3 knockout mice. The modified breathing pattern in these knock-out mice under control and hypoxia are strikingly identical to the compensated breathing pattern in rats with carotid body denervation, thus supporting the hypothesis of a disturbed carotid body function in these mice. In the future, conditional knockouts of these TASK channels may help to possibly elude such compensation mechanisms and to obtain a more detailed picture of the in vivo function of these background K+ channels in the regulation of respiration. TASK-1 and TASK-3 channels are indispensable component

for TE shortening under hypoxia since TASK-1-/-, TASK-3-/- and TASK-1/3-/- mice were unable to shorten the expiration time under hypoxia. Unexpectedly, TASK-1/3 knockout mice responded with a stimulation of ventilation upon hyperoxia mimicking the situation of animals with abrogated carotid body function. These data suggest that TASK-1 and TASK-3 channels are possibly the molecular determinants in carotid bodies inducing hypoventilation during hyperoxia. Future studies will be necessary to test if impaired chemo-sensation after chronic oxygen treatment of human neonates is at least in part - caused by disturbed TASK channel function. Conflict of interest The authors declare no conflict of interests. Acknowledgements: The authors thank Drs. Jacques Barhanin and Christain Gestreau for fruitful discussion and providing the TASK mouse models, Dr. Sally Hynes for proofreading the manuscript and Christina Sterner for technical support. The study was supported by the Deutsche Forschungsgemeinschaft (FOR1086 to RW).

12

References Aller, M.I., Veale, E.L., Linden, A.M., Sandu, C., Schwaninger, M., Evans, L.J., Korpi, E.R., Mathie, A., Wisden, W., Brickley, S.G., 2005. Modifying the subunit composition of TASK channels alters the modulation of a leak conductance in cerebellar granule neurons. J Neurosci 25, 11455-11467. Bates, M.L., Farrell, E.T., Eldridge, M.W., 2014. Abnormal ventilatory responses in adults born prematurely. N Engl J Med 370, 584-585. Bavis, R.W., Fallon, S.C., Dmitrieff, E.F., 2013. Chronic hyperoxia and the development of the carotid body. Respir Physiol Neurobiol 185, 94-104. Bayliss, D.A., Barhanin, J., Gestreau, C., Guyenet, P.G., 2015. The role of pH-sensitive TASK channels in central respiratory chemoreception. Pflugers Arch 467, 917-929. Bouferrache, B., Filtchev, S., Leke, A., Marbaix-Li, Q., Freville, M., Gaultier, C., 2000. The hyperoxic test in infants reinvestigated. Am J Respir Crit Care Med 161, 160-165. Brickley, S.G., Aller, M.I., Sandu, C., Veale, E.L., Alder, F.G., Sambi, H., Mathie, A., Wisden, W., 2007. TASK-3 two-pore domain potassium channels enable sustained high-frequency firing in cerebellar granule neurons. J Neurosci 27, 9329-9340. Buckler, K.J., 2015. TASK channels in arterial chemoreceptors and their role in oxygen and acid sensing. Pflugers Arch 467, 1013-1025. Buckler, K.J., Williams, B.A., Honore, E., 2000. An oxygen-, acid- and anaesthetic-sensitive TASK-like background potassium channel in rat arterial chemoreceptor cells. J Physiol 525 Pt 1, 135-142. Dubreuil, V., Ramanantsoa, N., Trochet, D., Vaubourg, V., Amiel, J., Gallego, J., Brunet, J.F., Goridis, C., 2008. A human mutation in Phox2b causes lack of CO2 chemosensitivity, fatal central apnea, and specific loss of parafacial neurons. Proc Natl Acad Sci U S A 105, 1067-1072. Fernandez-Aguera, M.C., Gao, L., Gonzalez-Rodriguez, P., Pintado, C.O., Arias-Mayenco, I., GarciaFlores, P., Garcia-Perganeda, A., Pascual, A., Ortega-Saenz, P., Lopez-Barneo, J., 2015. Oxygen Sensing by Arterial Chemoreceptors Depends on Mitochondrial Complex I Signaling. Cell Metab 22, 825-837. Fiamma, M.N., O'Connor, E.T., Roy, A., Zuna, I., Wilson, R.J., 2013. The essential role of peripheral respiratory chemoreceptor inputs in maintaining breathing revealed when CO2 stimulation of central chemoreceptors is diminished. J Physiol 591, 1507-1521. Gestreau, C., Heitzmann, D., Thomas, J., Dubreuil, V., Bandulik, S., Reichold, M., Bendahhou, S., Pierson, P., Sterner, C., Peyronnet-Roux, J., Benfriha, C., Tegtmeier, I., Ehnes, H., Georgieff, M., Lesage, F., Brunet, J.F., Goridis, C., Warth, R., Barhanin, J., 2010. Task2 potassium channels set central respiratory CO2 and O2 sensitivity. Proc Natl Acad Sci U S A 107, 2325-2330. Goldstein, S.A., Bockenhauer, D., O'Kelly, I., Zilberberg, N., 2001. Potassium leak channels and the KCNK family of two-P-domain subunits. Nat Rev Neurosci 2, 175-184.

13

Guyenet, P.G., 2014. Regulation of breathing and autonomic outflows by chemoreceptors. Compr Physiol 4, 1511-1562. Guyon, A., Tardy, M.P., Rovere, C., Nahon, J.L., Barhanin, J., Lesage, F., 2009. Glucose inhibition persists in hypothalamic neurons lacking tandem-pore K+ channels. J Neurosci 29, 2528-2533. Heitzmann, D., Derand, R., Jungbauer, S., Bandulik, S., Sterner, C., Schweda, F., El Wakil, A., Lalli, E., Guy, N., Mengual, R., Reichold, M., Tegtmeier, I., Bendahhou, S., Gomez-Sanchez, C.E., Aller, M.I., Wisden, W., Weber, A., Lesage, F., Warth, R., Barhanin, J., 2008. Invalidation of TASK1 potassium channels disrupts adrenal gland zonation and mineralocorticoid homeostasis. EMBO J 27, 179-187. Jungbauer, S., Buehler, P.K., Neubauer, J., Haas, C., Heitzmann, D., Tegtmeier, I., Sterner, C., Barhanin, J., Georgieff, M., Warth, R., Thomas, J., 2016. Sex-dependent differences in the in vivo respiratory phenotype of the TASK-1 potassium channel knockout mouse. Respir Physiol Neurobiol. Karschin, C., Wischmeyer, E., Preisig-Muller, R., Rajan, S., Derst, C., Grzeschik, K.H., Daut, J., Karschin, A., 2001. Expression pattern in brain of TASK-1, TASK-3, and a tandem pore domain K(+) channel subunit, TASK-5, associated with the central auditory nervous system. Mol Cell Neurosci 18, 632-648. Kim, D., Cavanaugh, E.J., Kim, I., Carroll, J.L., 2009. Heteromeric TASK-1/TASK-3 is the major oxygensensitive background K+ channel in rat carotid body glomus cells. J Physiol 587, 2963-2975. Kim, I., Donnelly, D.F., Carroll, J.L., 2012. Postnatal hyperoxia impairs acute oxygen sensing of rat glomus cells by reduced membrane depolarization. Adv Exp Med Biol 758, 49-54. Klein, J.P., Forster, H.V., Bisgard, G.E., Kaminski, R.P., Pan, L.G., Hamilton, L.H., 1982. Ventilatory response to inspired CO2 in normal and carotid body-denervated ponies. J Appl Physiol Respir Environ Exerc Physiol 52, 1614-1622. Kumar, N.N., Velic, A., Soliz, J., Shi, Y., Li, K., Wang, S., Weaver, J.L., Sen, J., Abbott, S.B., Lazarenko, R.M., Ludwig, M.G., Perez-Reyes, E., Mohebbi, N., Bettoni, C., Gassmann, M., Suply, T., Seuwen, K., Guyenet, P.G., Wagner, C.A., Bayliss, D.A., 2015. PHYSIOLOGY. Regulation of breathing by CO(2) requires the proton-activated receptor GPR4 in retrotrapezoid nucleus neurons. Science 348, 12551260. Kumar, P., Prabhakar, N.R., 2012. Peripheral chemoreceptors: function and plasticity of the carotid body. Compr Physiol 2, 141-219. Lazarenko, R.M., Willcox, S.C., Shu, S., Berg, A.P., Jevtovic-Todorovic, V., Talley, E.M., Chen, X., Bayliss, D.A., 2010. Motoneuronal TASK channels contribute to immobilizing effects of inhalational general anesthetics. J Neurosci 30, 7691-7704. Linden, A.M., Sandu, C., Aller, M.I., Vekovischeva, O.Y., Rosenberg, P.H., Wisden, W., Korpi, E.R., 2007. TASK-3 knockout mice exhibit exaggerated nocturnal activity, impairments in cognitive functions, and reduced sensitivity to inhalation anesthetics. J Pharmacol Exp Ther 323, 924-934. Ma, L., Roman-Campos, D., Austin, E.D., Eyries, M., Sampson, K.S., Soubrier, F., Germain, M., Tregouet, D.A., Borczuk, A., Rosenzweig, E.B., Girerd, B., Montani, D., Humbert, M., Loyd, J.E., Kass,

14

R.S., Chung, W.K., 2013. A novel channelopathy in pulmonary arterial hypertension. N Engl J Med 369, 351-361. Manoury, B., Lamalle, C., Oliveira, R., Reid, J., Gurney, A.M., 2011. Contractile and electrophysiological properties of pulmonary artery smooth muscle are not altered in TASK-1 knockout mice. J Physiol 589, 3231-3246. Miller, M.J., Tenney, S.M., 1975. Hyperoxic hyperventilation in carotid-deafferented cats. Respir Physiol 23, 23-30. Mulkey, D.K., Talley, E.M., Stornetta, R.L., Siegel, A.R., West, G.H., Chen, X., Sen, N., Mistry, A.M., Guyenet, P.G., Bayliss, D.A., 2007. TASK channels determine pH sensitivity in select respiratory neurons but do not contribute to central respiratory chemosensitivity. J Neurosci 27, 14049-14058. Murtaza, G., Mermer, P., Goldenberg, A., Pfeil, U., Paddenberg, R., Weissmann, N., Lochnit, G., Kummer, W., 2017. TASK-1 potassium channel is not critically involved in mediating hypoxic pulmonary vasoconstriction of murine intra-pulmonary arteries. PLoS One 12, e0174071. Olson, E.B., Jr., Vidruk, E.H., Dempsey, J.A., 1988. Carotid body excision significantly changes ventilatory control in awake rats. J Appl Physiol (1985) 64, 666-671. Ortega-Saenz, P., Levitsky, K.L., Marcos-Almaraz, M.T., Bonilla-Henao, V., Pascual, A., Lopez-Barneo, J., 2010. Carotid body chemosensory responses in mice deficient of TASK channels. J Gen Physiol 135, 379-392. Rong, W., Gourine, A.V., Cockayne, D.A., Xiang, Z., Ford, A.P., Spyer, K.M., Burnstock, G., 2003. Pivotal role of nucleotide P2X2 receptor subunit of the ATP-gated ion channel mediating ventilatory responses to hypoxia. J Neurosci 23, 11315-11321. Roux, J.C., Peyronnet, J., Pascual, O., Dalmaz, Y., Pequignot, J.M., 2000. Ventilatory and central neurochemical reorganisation of O2 chemoreflex after carotid sinus nerve transection in rat. J Physiol 522 Pt 3, 493-501. Sirois, J.E., Lei, Q., Talley, E.M., Lynch, C., 3rd, Bayliss, D.A., 2000. The TASK-1 two-pore domain K+ channel is a molecular substrate for neuronal effects of inhalation anesthetics. J Neurosci 20, 63476354. Trapp, S., Aller, M.I., Wisden, W., Gourine, A.V., 2008. A role for TASK-1 (KCNK3) channels in the chemosensory control of breathing. J Neurosci 28, 8844-8850. Turner, P.J., Buckler, K.J., 2013. Oxygen and mitochondrial inhibitors modulate both monomeric and heteromeric TASK-1 and TASK-3 channels in mouse carotid body type-1 cells. J Physiol 591, 59775998. Wang, S., Benamer, N., Zanella, S., Kumar, N.N., Shi, Y., Bevengut, M., Penton, D., Guyenet, P.G., Lesage, F., Gestreau, C., Barhanin, J., Bayliss, D.A., 2013. TASK-2 channels contribute to pH sensitivity of retrotrapezoid nucleus chemoreceptor neurons. J Neurosci 33, 16033-16044.

15

Washburn, C.P., Sirois, J.E., Talley, E.M., Guyenet, P.G., Bayliss, D.A., 2002. Serotonergic raphe neurons express TASK channel transcripts and a TASK-like pH- and halothane-sensitive K+ conductance. J Neurosci 22, 1256-1265. 6. Figure legends: Fig. 1: Basal respiration of male and female TASK-3-/- and male TASK-1/3-/- mice. (A) Respiratory parameters of TASK-3 mice under control condition (21% O2): Tidal volume (VT); respiratory frequency (fR); minute volume (MV); peak expiratory/inspiratory flow (PEF/PIF); expiration/inspiration time (TE/Ti); relaxing time (RT). Data of wild-type mice are shown in black (male n=21/female n=22); data of TASK-3-/- mice are shown in red (male n=19/ female n=22). Data are shown as box plots with min/max whiskers. Significant differences between wild-type (TASK-3+/+) and TASK-3/-

mice (P < 0.05) are indicated by “”.

(B) Respiratory parameters of male TASK-1/3 mice under control condition (21% O2). Data of wild-type mice are shown in black (n=11); data of TASK-1/3-/- mice are shown in red (n=12). Data are shown as box plots with min/max whiskers. Significant differences between wild-type (TASK-1/3+/+) and TASK1/3-/- mice (P < 0.05) are indicated by “”. Fig. 2A: Effects of hypoxia in male TASK-3-/- mice. Respiratory response to hypoxia in wild-type mice (n=8) and TASK-3-/- mice (n=8). Changes of respiratory frequency (fR), tidal volume (VT), minute volume (MV), and expiration time (TE) under hypoxia are depicted. In TASK-3-/- mice, beside a missing physiological decrease of TE in the acute phase of hypoxia, no obvious changes of the respiratory hypoxia response were observed. Data are shown as median ± interquartile range (IQR). “” indicates significant differences between TASK-3+/+ and TASK3-/- mice (P < 0.05; unpaired). Fig. 2B Effects of hypoxia in female TASK-3-/- mice Respiratory response to hypoxia in wild-type mice (n=4) and TASK-3-/- mice (n=4). Changes of respiratory frequency (fR), tidal volume (VT), minute volume (MV), and expiration time (TE) under hypoxia are depicted. Female TASK-3-/- mice responded identical to hypoxia as the male TASK-3-/- mice. Data are shown as median ± interquartile range (IQR). “” indicates significant differences between TASK-3+/+ and TASK-3-/- mice (P < 0.05; unpaired). Fig. 3: Effects of hypercapnia in male and female TASK-3-/- mice. Respiratory response of male wild-type (male: n=8/female: n=4) and TASK-3-/- mice (male: n=8/female: n=4) under acute hypercapnia (5% CO2/95% O2; 30 min). Data are shown as median ± interquartile

16

range (IQR). “#“ indicates significant differences (p < 0.05) between control and experimental period within the same group of mice. “##” indicates significant differences (p < 0.001) between control and experimental period within the same group of mice. No significant differences were observed between the genotypes. Fig. 4: Fig. 2: Effects of hypoxia in TASK-1/3-/- mice. The effects of acute hypoxia were measured in wild-type (n=4) and TASK-1/3-/- mice (n=4). Average changes of respiratory frequency (fR), tidal volume (VT), minute volume (MV) and expiration time (TE) are depicted. TASK-1/3-/- mice showed under acute hypoxia no increase of the respiratory frequency but an adequate increase of minute volume compared to wild-type mice due to an enhanced tidal volume. Data are shown as median ± interquartile range (IQR). “” indicates significant difference between the genotypes (P < 0.05; unpaired). Fig. 5: Effects of acute hypercapnia in TASK-1/3-/- mice. The effects of hyperoxic hypercapnia were measured in wild-type (n=8) and TASK-1/3-/- mice (n=8). Respiratory frequency (fR), tidal volume (VT) and minute volume (MV) are shown. Acute hyperoxic hypercapnia (5% CO2/95% O2) induced similar stimulation of respiration in both genotypes. Interestingly, hyperoxia alone (100% O2) activated fR of TASK-1/3-/- mice, whereas fR of wild-type mice was transiently inhibited. Data are shown as median ± interquartile range (IQR) and significant differences (p < 0.05; unpaired) between the genotypes are indicated by “”. Fig. 6: Effects of stepwise increases in CO2 on respiration of TASK-1/3-/- mice. The effects of stepwise increases in CO2 were measured in wild-type (n=8) and TASK-1/3-/- mice (n=8). Please note the increased minute volume of TASK-1/3-/- mice under hyperoxia (100%) and 6% CO2. Data are shown as median ± interquartile range (IQR) and significant differences (p < 0.05) between the genotypes are indicated by “”.

17

Figr-1

18

19

20

21

22