Oxytocin and corticotropin-releasing hormone exaggerate nTS neuronal and synaptic activity following chronic intermittent hypoxia

Abstract

Chronic intermittent hypoxia (CIH) in rodents mimics the hypoxia-induced elevation of blood pressure seen in individuals experiencing episodic breathing. The brainstem nucleus Tractus Solitarii (nTS) is the first site of visceral sensory afferent integration, and thus is critical for cardiorespiratory homeostasis and its adaptation during a variety of stressors. In addition, the paraventricular nucleus of the hypothalamus (PVN), in part through its nTS projections that contain oxytocin (OT) and/or corticotropin-releasing hormone (CRH), contributes to cardiorespiratory regulation. Within the nTS, these PVN-derived neuropeptides alter nTS activity and the cardiorespiratory response to hypoxia. Nevertheless, their contribution to nTS activity after CIH is not fully understood. We hypothesized that OT and CRH would increases nTS activity to a greater extent following CIH, and co-activation of OT+CRH receptors would further magnify nTS activity. Our data shows that compared to their normoxic controls, 10 days CIH exaggerated nTS discharge, excitatory synaptic currents, and Ca²⁺ influx in response to CRH, which were further enhanced by the addition of OT. CIH increased the tonic functional contribution of CRH receptors, which occurred with elevation of mRNA and protein. Together, our data demonstrate that intermittent hypoxia exaggerates the expression and function of neuropeptides on nTS activity.

Graphical Abstract

Following chronic intermittent hypoxia (CIH), the augmentation in neuronal activity and synaptic transmission by oxytocin (OT) and corticotropin-releasing hormone (CRH) within the nucleus Tractus Solitarii (nTS) was enhanced compared to normoxia exposure. nTS slices were analyzed to assess neuronal electrophysiological properties and calcium (Ca²⁺) influx. CIH exaggerated nTS discharge, excitatory synaptic currents, and Ca²⁺ influx specifically in response to CRH, with further amplification observed upon the addition of OT. CIH also increased the tonic functional contribution of CRH and OT receptors to synaptic activity and was accompanied by elevations in mRNA. Together, CIH exaggerates the expression and function of OT and CRH on nTS function, contributing to respiratory and cardiovascular dysregulation in conditions like obstruct sleep apnea.

INTRODUCTION

The initial integration and processing of chemo- and baroafferents critical to the control of breathing and blood pressure occurs within the brainstem nucleus tractus solitarii (nTS). Alterations in nTS synaptic and neuronal activity are critically involved in the physiological [e.g., exercise (Chen et al., 2002)] and pathophysiological [e.g., hypertension (Kline et al., 2007; Accorsi-Mendonca et al., 2016)] control of arterial pressure. The paraventricular nucleus of the hypothalamus (PVN) is reciprocally connected to the nTS (McKellar & Loewy, 1981; Affleck et al., 2012) and plays an important role in the autonomic control of arterial pressure (Li & Pan, 2007; Basting et al., 2018) and reflex responses to brief hypoxia (Ruyle et al., 2019). The PVN sends direct neuronal projections to the nTS to modulate its activity (Ruyle et al., 2019), and PVN neuronal terminations can contain, amongst others, oxytocin (OT) and/or corticotropin-releasing hormone (CRH) (Geerling et al., 2010; Ruyle et al., 2018). These nTS-projecting PVN neurons are also activated by hypoxia (Ruyle et al., 2018). Within the nTS, projections from the PVN augment reflex responses to hypoxia (Ruyle et al., 2023) and oxytocin and CRH independently enhance neuronal or synaptic activity (Lewis et al., 2002; Peters et al., 2008; Wang et al., 2019).

Individuals with obstructive sleep apnea or episodic breathing often develop excessive sympathetic nerve activity, hypertension and cardiac arrhythmias. Sleep apnea is associated with periodic episodes of hypoxia that induce activation of the peripheral chemoreflex. Chronic intermittent hypoxia (CIH) in rodents mimics the episodic hypoxia and the elevation of blood pressure in these individuals (Bosc et al., 2010; Dantzler & Kline, 2020). The central mechanisms by which episodic hypoxia induce pathophysiological responses remain unsolved (Shamsuzzaman et al., 2003) yet available evidence suggests alterations in the nTS and PVN circuitry may contribute. For instance, following CIH, nTS neuronal and synaptic activity (Kline et al., 2007; Almado et al., 2012; Kline et al., 2019) as well as Fos expression is increased (Greenberg et al., 1999b; Knight et al., 2011). In addition, nTS-projections to and their synaptic integration within the PVN are altered (Domingos-Souza et al., 2021), and PVN neuronal activity is elevated (Coleman et al., 2010). In addition, acute bouts of hypoxia increase PVN Fos activation, including nTS-projecting OT and CRH PVN neurons (Ruyle et al., 2018), as well as the production and circulation of OT and CRH (Kelestimur et al., 1997; Dantzler & Kline, 2020). Nevertheless, the influence of OT and/or CRH on the sensory afferent-nTS circuit following CIH requires further study. We have shown that following CIH, OT reduces potassium currents in vagal afferent somas to a greater extent than their normoxic controls (Dantzler & Kline, 2020), and activation of OT neurons can increase the activity of the dorsal motor nucleus of the vagus (DMV) after CIH (Jameson et al., 2016). Nevertheless, the understanding of how OT and CRH act within the brainstem and specifically within the nTS following CIH is relatively sparse.

Given the importance of OT and CRH in the autonomic control of arterial pressure, their influence on nTS circuit activity, and the elevation of these neuropeptides in response to hypoxia, in this study we examined their influence on nTS following CIH. We tested the hypothesis that after CIH, OT and CRH enhance nTS activity to a greater extent than normoxia. Moreover, there will be a greater influence of these neuropeptides to further magnify synaptic transmission after CIH. This hypothesis was examined using electrophysiological, calcium imaging, immunohistochemical and molecular approaches. We show that after CIH conditions, CRH alone and with the addition of OT enhanced nTS activity via elevation of receptor expression to suggest an important contribution to cardiorespiratory function during hypoxia.

METHODS

Ethical Approval.

All animal procedures were performed with the approval of the University of Missouri Animal Care and Use Committee (ACUC number: 42568), and in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and The Journal of Physiology’s policies regarding animal experiments. Three-to-four-week-old Sprague-Dawley male rats were used to compare and extend our previous in vivo and in vitro studies. Rats were purchased from Envigo (Indianapolis, IN, United States) and housed under 12 h light/12 h dark conditions until used. Animals had ad libitum access to food and water.

Chronic intermittent hypoxia (CIH) or normoxia (NORM) exposure.

As performed previously (Kline et al., 2007; Kline, 2010; Kline et al., 2019), rats were exposed to either 10 days of normoxia (NORM, 21% O₂) or CIH (~47 sec of 6% O₂, 10 episodes/h, 8 hr per day) between 0900–1700 using a commercially available system (Biospherix, Richmond NY). Hypoxia was induced by exchanging the oxygen for nitrogen with alternating cycles of 3 min for O₂ and 3 min 15s for N₂; the flow rate of O₂ and N₂ was controlled via a negative feedback system. The O₂ levels decreased from 20.9% to 7% O₂ within 115s and was below 10% for ~150s and 7% for ~90s. Gas remained between 6.5% and 6.2% for ~47s before returning to normoxic conditions within 155s. This exposure protocol is similar to other laboratories (Zoccal et al., 2008; Moraes et al., 2013; Karlen-Amarante et al., 2023). Normoxic groups were exposed to similar air movement and sounds of CIH between 0900–1700, but the rats were only exposed to room air. The oxygen was kept at 21% overnight for both groups. Chamber O₂, CO₂ and temperature were continuously monitored. All experiments were conducted after the last day of CIH.

Sensory afferent labeling.

As described previously (Lima-Silveira et al., 2020; Martinez et al., 2020), ~20 days before the beginning of the NORM or CIH protocol, rats were anesthetized with isoflurane (2–3% in 100% O₂, ~700 ml/min; Covetrus, Dublin, OH) and a surgical plane of anesthesia was confirmed every 15 minutes by lack of palpebral reflex. A small ventral midline incision in the neck region was made to expose the nodose ganglion. The genetically encoded calcium indicator, GCaMP6m (pAAV.Syn.GCaMP6m.WPRE.SV40, RRID:Addgene_100841), was injected (~5 μL) into the ganglion with a Picospritzer (General Valve, Fairfield, NJ). The neck incision was carefully sutured with 4–0 Vicryl suture. Following the surgery, the rats were treated subcutaneously with analgesic (Buprenorphine 0.05 mg/kg, Columbus, OH) and antibiotic (Enrofloxacin 10 mg/kg, Newry, Northern Ireland, UK). Rats were exposed to 10d CIH or NORM beginning ~20 days after AAV injection to allow animal recovery and GCaMP6m expression. The duration of the surgery was 30–40 minutes.

In vitro nTS slice generation.

Brainstem slices containing the nTS were prepared from NORM or 10 days CIH exposed rats, as previously described by us (Martinez et al., 2021; Lima-Silveira et al., 2022). Briefly, rats were deeply anesthetized to effect with isoflurane (5% in 100% O₂, 1000 ml/min) in an induction chamber. Following lack of response to toe pinch, rats were decapitated. The brainstem was removed and placed in an ice-cold NMDG cutting solution containing (in mM): 93 NMDG, 2.5 KCl, 1.2 NaH2PO₄, 10 MgSO₄, 30 NaHCO₃, 20 HEPES, 25 D‐glucose, 5 L-ascorbic acid, 2 thiourea, 3 sodium pyruvate and 0.5 CaCl₂. The pH was adjusted to 7.4 with HCl and aerated with 95% O₂-5% CO₂ with an osmolarity of 295–310 mOsmol. A coronal or horizontal nTS slice (~250 or 280 mm, respectively) was obtained via the use of a vibratome (VT1200S; Leica Microsystems) equipped with a sapphire blade (Delaware Diamond Knives). Slices recovered at 35°C for 12 min in NMDG‐cutting solution before transferring to recording artificial cerebral spinal fluid (aCSF) containing (in mM) 125 NaCl, 3 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 25 NaHCO₃, 10 dextrose, and 2 CaCl₂, bubbled with 95% O₂–5% CO₂ that was used for the remainder of slice experiments. Slices recovered for an additional 20–40 minutes at room temperature.

Multi-electrode array (MEA) recordings.

Coronal nTS slices were positioned on a 6×10 perforated recording array (60pMEA100/30iR-Ti) and held with a nylon mesh on a MEA2100-System headstage (Multichannel Systems GmbH, Germany, RRID:SCR_014809). Following a 25-minute perfusion with aCSF at 33°C to allow the slice’s activity to equilibrate to temperature and flow, the experimental recording protocol was initiated. Spontaneous discharge of neurons from the entire nTS was acquired at a sampling frequency of 25 kHz. Activity was Bessel filtered (high pass at 100 Hz) to remove noise and isolate individual spikes.

MEA data was initially analyzed using Offline Sorter x64 V4 (Plexon, Neurotechnology research Systems version 4.6.2) to isolate neurons within each given channel through K-means clustering waveform isolation and principal components analysis. After this, data was imported into NeuroExplorer (Nex Technologies version 5.411, RRID:SCR_001818) and the frequency of activity was binned in 30s intervals and exported to Microsoft Excel. All neurons that responded to either agonist application, whether an increase or decrease, including those that were silent during initial baseline, were included in analysis.

Patch clamp electrophysiology.

Horizontal nTS slices were placed in a superfusion chamber and held with nylon mesh. Whole-cell recordings from cells directly connected to sensory afferents were conducted on neurons in the medial-commissural nTS. To record excitatory postsynaptic currents (EPSCs), glass recording electrodes were filled with (in mM) 10 NaCl, 130 K+-gluconate, 11 EGTA, 1 CaCl₂, 10 HEPES, 1 MgCl₂, 2 Mg-ATP, and 0.2 Na-GTP, pH 7.3, ~280 mosM, 4–7 MΩ. In voltage clamp experiments, cells were held at −60 mV. Spontaneous EPSCs (sEPSC) were recorded in the absence of any stimulus. Afferent-evoked EPSCs were generated with a concentric bipolar electrode placed on the afferent containing solitary tract (TS, i.e., TS-EPSCs) and evoked via an isolated stimulator (100 μs, 10–200 μA, A.M.P.I., Master-8 and ISO-Flex). The sampling rate for TS-EPSCs evoked at 0.5 Hz was 100–200 kHz, whereas repetitive currents were evoked at 20 Hz and sampled at 20 kHz. Resting membrane potential (RMP) was recorded in current-clamp mode with no holding current (I=0), and action potential (AP) discharge was recorded in current-clamp evoked by depolarizing current steps (0–120 pA, 100 ms). Data were acquired in pClamp 10.7 using a Axopatch 200B or Multiclamp 700B amplifier (Molecular Devices, RRID:SCR_018455), Axon Digidata 1550B (Molecular Devices), and filtered at 2 kHz. For all recordings, the bath temperature was maintained at 33° C (TC-344B, Warner Instruments).

Patch clamp electrophysiological data were analyzed using Clampfit 10 (Molecular Devices, RRID:SCR_011323), Microsoft Excel software and Easy Electrophysiology v2.4.0. Only second-order nTS neurons, identified by their low standard deviation of the synaptic latency [i.e., jitter <0.30 ms] were studied as previously (Matott et al., 2016; Lima-Silveira et al., 2020; Martinez et al., 2020; Lima-Silveira et al., 2022).

Sensory terminal nTS calcium imaging.

Horizontal nTS slices generated from rodents with prior injection of GCaMP6s in the nodose were transferred to a superfusion chamber, held in place via nylon mesh and superfused with aCSF. Changes in Ca²⁺ fluorescence were obtained by illuminating the slice at 488 nm and capturing fluorescence at 520 nm. TS-evoked (10 stimuli at 20 Hz) changes in Ca²⁺ was acquired via a Yokogawa CSU-W1 confocal system (3i, Denver, CO) using 40x water-immersion objective (Olympus LUMPlanFL/IR), Prime 95B sCMOS camera (Photometrics), controlled by Slidebook software (RRID:SCR_014300). Fluorescence was analyzed with Slidebook 6, FIJI ImageJ 1.53s, Python scripts and OriginPro software.

Dissociated nTS neurons and Fura-2 calcium imaging.

As previously performed by us (Ostrowski et al., 2017) the brainstem from 10d NORM and CIH rats was removed from isoflurane-anesthetized rats (5% in 100% O₂, 1000 ml/min to effect; Covetrus, Dublin, OH) following decapitation, and placed in oxygenated ice-cold low calcium-high magnesium aCSF containing, in mM: 124 NaCl, 3 KCl, 1.2 NaH₂PO₄, 1.2 MgSO₄, 25 NaHCO₃, 11 D-glucose, 1 CaCl₂, and 2 MgCl₂ (pH 7.4, ~300 mOsmol). Coronal slices containing the nTS were generated as above. nTS slices were then incubated in aCSF containing papain (10 U/ml, 35 min at 35 °C, Worthington, Lakewood, NJ) that was activated via the addition of EDTA (0.27 mM), mercaptoethanol (0.017 mM), and cysteine-HCl (1.37 mM; Sigma, St. Louis, MO). Following papain digestion, the nTS was microdissected from the brainstem and dorsal motor nucleus of the vagus (DMV) using a razor blade, placed in aCSF solution containing 0.1% DNase (ThermoFisher Scientific/Invitrogen, Eugene, OR) and BSA (0.5%, Sigma, St. Louis, MO) and manually triturated with progressively smaller fire-polished glass pipettes until the tissue was visibly dispersed. Neurobasal-A (GIBCO) was added, supplemented with 2% B-27 (GIBCO), 1% GlutaMAX (GIBCO), 1% insulin-transferrin (GIBCO), and 1% penicillin-streptomycin (GIBCO). The cells were plated on poly-d-lysine (100 μg/ml) coated glass coverslips and allowed to settle and attach for 1 hr in an incubator. Afterward, neurons were submerged with a Neurobasal-A culture medium. All experiments occurred on the same day as the preparation of the cell culture.

nTS neurons were loaded for 30 min with 0.05% Fura-2 AM (50 μg; ThermoFisher Scientific/Invitrogen, Eugene, OR, reconstituted with 50 μl of 20% Pluronic F-127 in DMSO (Molecular Probes)); in 1 mL Neurobasal-A at 37°C and 5% CO₂ in the dark. After Fura-2 incubation, the cells were washed two times in Neurobasal A and allowed to rest for another 30 min. The glass coverslips with Fura-2-loaded nTS cells were then placed in a superfusion chamber on an inverted microscope (IX71; Olympus) and superfused at 500–800 μl/min with a recording solution containing (in mM): 136 NaCl, 5.4 KCl, 0.33 NaH₂PO₄, 10 HEPES, 10 d-glucose, 1.8 CaCl₂, and 1 MgCl₂. Fura-2 was excited at 340 and 380 nm with a Polychrome V monochromator (TILL Photonics) and visualized at ~520 nm using a camera (Retiga Exi Fast 1394; QImaging) and a 20x objective. Images were acquired every 3 s using Micro-Manager software (Open Imaging, RRID:SCR_016865). Dissociated neuron Ca²⁺ fluorescence was analyzed with FIJI ImageJ 1.53s and Excel software.

Protocol and receptor agonists and antagonists used in electrophysiology and imaging.

The following drugs and concentrations were used to examine their influence on EPSCs, nTS discharge, and Ca²⁺ fluorescence: OT (600 nM; Tocris Bioscience, cat 1910) (Dantzler & Kline, 2020), CRH (300nM; Sigma, cat C3042), OT receptor antagonist (OTR, 1 μM, L-368,899 HCl; Tocris Bioscience, cat 2641) (Dantzler & Kline, 2020), and CRH receptor 2 antagonist (CRHR2, 100 nM, K41498; Tocris Bioscience, cat 2070)(Wang et al., 2018). Each drug was applied individually or together for 3–5 minutes interspersed with a 5-minute washout period. Data shown in figures represent cells that received all three drug combinations. Parameters were obtained in the final minutes of each application. OT and CRH have been shown to signal through the OTR and CRHR2 in the nTS (Peters et al., 2008; Wang et al., 2018; Brierley et al., 2021). All other compounds were obtained from ThermoFisher Scientific and Sigma-Aldrich, Inc. A slice/cell was exposed only once to the agonist or antagonist.

Immunohistochemistry of OT and CRH in the PVN and processes in nTS.

NORM and 10d CIH rats were deeply anesthetized with isoflurane (5% in 100% O₂, 1000 ml/min, to effect; Covetrus, Dublin, OH) and transcardially perfused with ice-cold 0.01M phosphate buffered saline (PBS, 125 mL) followed by 4% paraformaldehyde (PFA, 250 mL). Following decapitation, the brain was removed and postfixed in 4% PFA for an additional 2 hours. The tissue was sectioned at 30 μm in a coronal plane on a Vibratome (Leica VT 1000S) and placed in cryoprotectant at −20°C until processed. Heat-induced epitope retrieval and Tyramide Signaling Amplification (TSA) was used to improve fluorescence signal as by us (Ostrowski et al., 2014). Tissue sections containing the PVN and nTS were incubated in mouse monoclonal anti-OT antibody (1:500, Millipore, MAB5296, RRID:AB_11212999) and either rabbit anti-CRH (PVN, 1:5000, T-4037, BMA Biomedicals) or guinea pig anti-CRH (nTS, 1:9000, T-5007, BMA Biomedicals). Antibody specificity has been previously confirmed (Stanic et al., 2010; Kalin et al., 2016; Peng et al., 2017; Ruyle et al., 2018; Jones et al., 2021). Antibodies were placed in PBS-T with 1% normal donkey serum (NDS) and tissue incubated overnight at 4°C. Tissue sections were washed with PBS (3×5 min) and incubated in one or more of the secondary antibodies (all 1:200, Jackson ImmunoResearch): biotinylated donkey anti-mouse, biotinylated donkey anti-rabbit or donkey anti-guinea pig Cy3 or Cy5 with 1% NDS for 2 hours. Following PBS wash (3×5 min), sections were incubated for 1 hour at room temperature in streptavidin-HRP, washed with PBS 3×5 min, and then 10 min Biotin-XX Tyramide working solutions. After adding the reaction stop buffer and 3×5 min washes, the sections were incubated for 2 hours in Streptavidin Cy2 or Cy3 (1:200) and 3% NDS in PBS room temperature. Sections were coverslipped with ProLong Diamond antifade (ThermoFisher Scientific/Invitrogen, Eugene, OR).

The PVN and nTS sections were imaged using an Olympus BX51 widefield microscope equipped with a Hamamatsu Orca Fusion sCMOS camera and X-Cite Mini+ LED illuminator which were unified via Neurolucida imaging software (MBF Bioscience, v2022, Williston, VT, RRID:SCR_001775). Immunohistochemistry was analyzed with FIJI ImageJ. The PVN (NORM, N=5; CIH, N=5) and nTS (NORM, N=6; CIH, N=6) were identified following the coordinates from Paxinos and Watson (2018). The nTS was examined in six sections separated by 180 mm relative to calamus scriptorius (CS), the caudal nTS section where the fourth ventricle closes, and in this paper defined as 0 mm (Bregma, −14.40 mm). The PVN was examined between −1.20 to −2.07 mm relative to Bregma. The portions of PVN were separated in 3 sections, rostral (−1.20 to −1.44 mm), intermediate (−1.44 to −1.92 mm) and caudal (−1.92 to −2.07 mm). The number of OT and CRH immunopositive neurons in the PVN were manually counted by two individuals and averaged. Total immunoreactivity of OT and CRH within the PVN and nTS was obtained after initial background subtraction followed by intensity measurements.

nTS OTR and CRHR2 mRNA via RNAscope.

NORM and 10d CIH rats (N=4 each) were deeply anesthetized with isoflurane (5% in 100% O₂ at 1000 ml/min, to effect; Covetrus, Dublin, OH) and transcardially perfused with ice-cold 0.01 M PBS (125 mL) followed by 4% PFA (250 mL, pH = 7.4; Sigma). Brains were removed and post-fixed overnight in 4% PFA and placed in 20% sucrose for cryoprotection. The brainstems were removed and coronally sectioned at 17 μm using a cryostat (CM1900, Leica). The sections were mounted directly onto slides (Superfrost Plus slides, ThermoFisher Scientific) and stored at −80°C until used. mRNA identification via RNAscope was performed according to Multiplex Fluorescent Reagent Kit v.2 instructions (Advanced Cell Diagnostics, Newark, CA, USA). Briefly, after tissue pretreatment, samples were hybridized with probes specific to mRNA for OTR (Cat. No.: 483671-C2) and CRHR2 (Cat. No.: 417851). Sections were cover-slipped with ProLong Diamond Antifade Mountant with DAPI to identify nuclei (Life Technologies; Eugene, OR). For all samples we simultaneously applied an RNAscope 3-plex positive control probe (Cat. No.: 320881) and 3-plex negative control probes (Cat. No.: 320891). Fluorescent images of the DAPI, OTR, and CRHR2 mRNA probes were acquired via the fluorescent microscope described for immunohistochemistry. The images were analyzed using ImageJ and Cellprofiler 4.2.1 to quantify the mRNA fluorescent signal (“dots”).

Western blot of OTR and CRHR2 protein in nTS.

Brainstem tissue from 10d NORM (N=4) and CIH (N= 4) rats was collected from horizontal slices prepared as in electrophysiology protocols. The nTS was excised from surrounding tissue via scalpel incision medial to the solitary tract and lateral to dorsal motor nucleus of the vagus. nTS tissue was then flash frozen in liquid nitrogen and stored at −80°C until used. The tissue was homogenized in buffer containing (in mM, pH 7.4) 250 sucrose, 10 Tris, 1 EDTA, and protease inhibitors (Complete Mini, EDTA-free tablets, Roche) followed by incubation on ice for an additional 30 min. After centrifuging the homogenate at 13,300 RPM for 10 min at 4°C, the supernatant was collected and the protein concentration was determined by Qubit Protein Assay kit (ThermoFisher Scientific/Invitrogen, Eugene, OR). The concentration of protein used was 20 μg for OTR and 40 μg for CRHR2. Protein sample was mixed with 4x loading dye (Laemmli Sample Buffer, Bio-Rad), boiled at 90°C (5 min), separated on a 4–20% Stain-Free Tris-Glycine extended gel (TGX, Bio-Rad, 4568094). The gel was then imaged with UV light to activate the stain free fluorochrome. Protein was transferred to a polyvinylidene difluoride membrane (PVDF, Bio-Rad). The membrane was pre-blocked in Tris Buffered Saline with 0.1% Tween 20 (TBS-T) containing 5% powdered milk for 1 hr on a shaker at room temperature. The membrane was washed 3×5 min in TBS-T and then incubated in either rabbit anti-OTR (1:300, Abcam, ab217212, overnight, RRID: AB_2904223) or rabbit anti-CRHR2 (1:1000, Abcam, ab236982, 3 nights) antibodies in TBS-T with 5% milk at 4°C on a shaker. Antibody specificity has been previously confirmed (Wang et al., 2023; Wei et al., 2023). Following a TBS-T wash (3×5 min) the membrane was incubated in donkey anti-rabbit antibody conjugated to peroxidase (1:5000, Jackson ImmunoResearch) for 2 hr on a shaker at room temperature. After a TBS-T wash (3×5 min) the membrane was incubated in Clarity Western ECL (Bio-Rad, 1705061) for 5 min. Total protein (via UV-illumination), OTR and CRHR2 expression and molecular markers were imaged on the Bio-Rad ChemiDoc. Blot images were analyzed using ImageLab 6.1 (Bio-Rad) and bands of interest were normalized to the total protein.

Statistical Analysis.

All statistical analyses were performed with GraphPad Prism 9 and 10 (RRID:SCR_002798) or OriginPro (RRID:SCR_014212). The difference in the cellular response to a drug on sEPSC, TS-EPSC, extracellular discharge, membrane potential and action potential discharge in NORM and CIH-exposed cells was tested via two-way repeated measured ANOVA followed by the Fisher LSD post-test. IHC was examined across the nTS or PVN on NORM vs CIH via two-way RM ANOVA. RNAscope and western blot were tested via unpaired t-test. Statistical analysis for IHC, RNAscope and western blot was performed per rat. For patch clamp data all cells were included in the analysis. Further analysis was performed on cells that increased at least 10% and were considered responsive (Browning et al., 2014). The significance level was defined as p < 0.05. Data is presented as mean ± standard deviation (SD). Graphs and figures were made using GraphPad Prism, OriginPro and Adobe Illustrator. “N” denotes number of animals, “n” indicates number of cells or terminals.

RESULTS

Baseline electrophysiological parameters on cells from 10d CIH versus normoxia (NORM) are summarized in Table 1. Following CIH, the amplitude of glutamatergic solitary tract (TS)-evoked excitatory postsynaptic currents (EPSCs) was reduced while the number of asynchronous EPSCs following a 20Hz TS stimulus train was increased. These findings align with those reported in previous nTS studies (Kline et al., 2007; Almado et al., 2012; Kline et al., 2019).

Table 1:

Baseline electrophysiological parameters for NORM and CIH nTS neurons.

	NORM	CIH	p-value
Ext. discharge (events/sec)	17.4 ± 45.0	16.9 ± 37.4	0.1579
Cm (pF)	23.15 ± 8.75	23.81 ± 8.33	0.7837
Latency (ms)	4.44 ± 1.73	4.95 ± 1.66	0.2890
Jitter (ms)	0.20 ± 0.07	0.22 ± 0.06	0.2449
sESPC Freq (Hz).	9.15 ± 4.76	7.91 ± 4.90	0.3583
sEPSC Amp (pA)	17.92 ± 7.50	15.68 ± 4.41	0.1954
TS-ESPC Amp, 0.5Hz (pA)	144.35 ± 98.60	89.54 ± 71.29	0.0258
Asynchronous activity:
--Baseline (events/sec)	40.35 ± 22.36	47.69 ± 20.52	0.2234
--Peak	62.00 ± 24.10	75.38 ± 22.89	0.0436
--Average	55.05 ± 22.37	63.12 ± 20.92	0.1868
RMP (mV)	−66.25 ± 6.84	−67.84 ± 11.10	0.5468

Values are means ± SD. n=26 for patch clamp data, n= 1368–1971 for MEA. Data include agonist, antagonist and vehicle data under Bsl. unpaired t-test

OT and CRH augment overall nTS activity and CIH enhances their effects.

To evaluate the global extent to which OT and CRH inputs alter nTS activity, we utilized multi-electrode array (MEA) recordings to examine nTS neuronal discharge (Figure 1A,B). Each channel captured one or more neuron, which were separated via waveform analysis and principal component analysis (PCA, Figure 1C). OT (600 nM, 5 min), CRH (300 nM, 5 min), and OT+CRH (5 min) were applied. Events were evaluated during the last two minutes of each application during which a steady state concentration was achieved. Vehicle alone applied at similar periods served to evaluate the potential changes in events over the recording period, and thus served as a control.

Figure 1. nTS activity is increased by OT and CRH after CIH.

(A) Representative image of the nTS slice placed on the multi-electrode array (MEA). (B) Representative traces from a normoxia and CIH nTS slice channel during baseline (Bsl) and application of OT, CRH and/or OT+CRH. (C) Isolated waveforms from NORM channel shown in B; each different neuron was detected based on its shape. (D, E) Single unit discharge rates across time in the NORM and CIH cell shown in B&C in response to OT, CRH and OT+CRH. (F) Violin plot with internal box-and-whisker plot showing 25–75% range and median of events recorded during each respective agonist for (N=2, n=968) and CIH (N=3, n=696). p values for Bsl versus respective agonist or group shown in figure. Two-way RM ANOVA, p_drug = <0.0001, p_Hypoxia =<0.0001, p_{drug × hypoxia} =<0.0001) based on cell “n”. LSD multiple comparison p-values shown in figure.

As shown in the representative extracellular discharge from a NORM (Figure 1B,D) and CIH (Figure 1B,E) rat, OT and CRH alone and their co-application increased activity. Mean discharge for all NORM and CIH neurons is quantified in Figure 1F and shows nTS neurons were responsive to OT, CRH and OT+CRH. In both groups, co-application of OT+CRH was greater than either neuropeptide alone (e.g., OT vs OT+CRH, NORM p=0.0034; CIH p < 0.001, 2-way RM ANOVA with LSD). The overall response to OT+CRH was greater in CIH than normoxia. The magnitude (percent) of increase from baseline was compared between groups. The augmentation of discharge was greater after CIH during CRH (2-way RM ANOVA, p_{drug × hypoxia}, NORM, 1.83 ± 5.30% vs CIH, 20.16 ± 290.04%, p=0.0107 with LSD) and OT+CRH (NORM, 3.49 ± 14.93% vs CIH, 34.87 ± 330.65%, p < 0.001 with LSD). Overall, vehicle did not change discharge across time (Table 2). Together, these data demonstrate neuropeptides elevate global neuronal activity and CIH enhances these responses.

Table 2:

aCSF vehicle did not alter synaptic or neuronal properties

Normoxia	Bsl	Vehicle 1	Vehicle 2	Vehicle 3	p-value
MEA events (ev/sec)	2.24 ± 0.33	2.4 ± 0.36	1.82 ± 0.28	2.06 ± 0.29	0.6146
Fura-2 (Ca2+, ΔF/F)	N/A	1.07 ± 0.09	1.07 ± 0.14	1.10 ± 0.19	0.4653
sEPSC Freq. (Hz)	8.03 ± 1.69	7.80 ± 3.42	7.61 ± 2.69	8.03 ± 4.18	0.9734
sESPC Amp. (pA)	22.01 ± 8.09	20.29 ± 7.73	18.86 ± 6.54	19.02 ± 6.33	0.0795
TS-ESPC Amp. (0.5Hz, pA)	165.07 ± 216.66	160.62 ± 204.79	163.58 ± 182.91	142.72 ± 168.02	0.629
TS-ESPC Amp 20 Hz Sum (pA)	744.73 ± 326.51	525.01 ± 260.81	426.97 ± 196.58	406.08 ± 221.85	0.1588
Asynchronous activity:
--Baseline (ev/sec)	46.0 ± 18.88	42.5 ± 15.26	46.75 ± 20.15	61.75 ± 35.67	0.2609
--Peak (ev/sec)	51.25 ± 14.99	63.75 ± 24.95	64.0 ± 19.78	67.75 ± 27.79	0.2752
--Average (ev/sec)	51.5 ± 15.17	55.25 ± 19.13	58.5 ± 21.0	64.25 ± 28.52	0.3283
RMP (mV)	−64.11 ± 10.44	−62.86 ± 3.48	−57.15 ± 7.84	−46.54 ± 15.80	0.2099
CIH	Bsl	Vehicle 1	Vehicle 2	Vehicle 3	p-value
MEA events(s) (ev/sec)	14.45 ± 1.56	23.99 ± 3.57	16.47 ± 3.8	7.95 ± 0.77	0.0007, #
Fura-2 (Ca2+, ΔF/F)	N/A	1.14 ± 0.16	1.10 ± 0.24	1.17 ± 0.33	0.3976
sEPSC Freq. (Hz)	6.74 ± 2.01	7.48 ± 1.84	6.67 ± 1.77	7.24 ± 2.32	0.5911
sESPC Amp. (pA)	13.91 ± 3.96	15.53 ± 5.61	14.45 ± 8.43	12.94 ± 6.47	0.5886
TS-ESPC Amp. (0.5Hz, pA)	135.53 ± 119.27	149.25 ± 120.69	122.37 ± 88.94	151.13 ± 104.62	0.3521
TS-ESPC Amp 20 Hz Sum (pA)	513.17 ± 295.85	548.18 ± 330.67	526.25 ± 363.08	560.02 ± 303.20	0.8043
Asynchronous activity:
--Baseline (ev/sec)	41.6 ± 11.39	49.6 ± 14.17	44.0 ± 7.03	44.4 ± 17.44	0.7588
--Peak (ev/sec)	64 ± 12.96	73.2 ± 6.57	76.8 ± 9.09	69.8 ± 8.16	0.2457
--Average (ev/sec)	54.8 ± 13.68	60.4 ± 9.26	63.4 ± 7.56	56.2 ± 8.22	0.4713
RMP (mV)	−70.92 ± 5.86	−67.25 ± 5.43	−63.01 ± 5.92	−64.91 ± 5.83	0.1709

Values are means ± SD. N=4, n=4 each for patch clamp, Fura-2 N= 4–8, n= 32–30, MEA N=2, n=694–906. P values from 1-way RM ANOVA within each group. N/A in Fura-2 data due as responses were determined only during changing of solutions at a given time point. #MEA LSD multiple comparison: Veh 1 vs Veh3, p=0.0002.

Enhanced elevation of cytosolic Ca²⁺ by OT + CRH in isolated cells after CIH.

To help differentiate the synaptic and somal site of action of OT and CRH on nTS activity, we dissociated the nTS and examined intracellular calcium (Ca²⁺) elevation in isolated neurons via Fura-2 fluorescence (Figure 2A) as another index of neuronal activity. Isolated nTS cells were exposed to OT (600 nM, 3 min), CRH (300 nM, 3 min), and OT+CRH (3 min). At the end of the protocol, the cells were exposed to a high K+ solution (55 mM, 15s) to briefly depolarize neurons and evoke Ca²⁺ entry to confirm cellular viability. As shown in the representative traces (Figure 2B) and the mean data (Figure 2C), intracellular Ca²⁺ increased following OT, CRH and the co-application of OT+CRH in neurons from NORM and CIH-exposed rats compared to the baseline control. Data are quantified in Figure 2C and demonstrate that the increase in Ca²⁺ following CRH and OT+CRH was greater in CIH than NORM. Following CIH, the Ca²⁺ elevation by dual OT+CRH was greater than either agonist alone. To confirm the receptor specificity of neuropeptide effects, we examined the Ca²⁺ response to OT or CRH in NORM neurons in the presence of their respective receptor antagonist. Application of an OT receptor antagonist (OTR-x, L-368,899 HCl, 1 μM) eliminated the Ca²⁺ increase by OT, and this response was less than that of OT alone (Fig 2D). Likewise, a CRH receptor type 2 antagonist (CRHR2-x, K41498, 100 nM) eliminated the Ca²⁺ increase by CRH, which was also attenuated compared to CRH alone (Fig 2E). These data confirm OT and CRH at the concentrations used increase intracellular Ca²⁺ via the OTR and CRHR2, respectively. In addition, the response to brief depolarization (55 mM K) was greater in CIH than NORM (3.08 ± 1.25 vs 2.22 ± 0.97, p= 0.0001, t-test). In comparison to agonist application, aCSF alone did not alter Ca²⁺ across the recording period in either group (Table 2). These data indicate CIH enhances neuropeptide Ca²⁺ responses in isolated nTS neurons.

Figure 2. OT and CRH enhance cytosolic Ca²⁺ after CIH in dissociated nTS neurons.

(A) Example of dissociated nTS neurons under brightfield (top) and 340 nm fluorescence (bottom). Scale = 50 μm. (B) Raw traces of Fura-2 fluorescence illustrating increase in cytosolic Ca²⁺ to 3 min of OT, CRH, OT+CRH and high K⁺. Note greater responses with combined OT+CRH after CIH. (C) Increase in relative fluorescence from baseline in all neurons (NORM N=3, n=53; CIH N=4, n=49). Vehicle control/baseline =”1”. 2-way RM ANOVA, p_drug = <0.0001, p_Hypoxia =0.0001, p_{drug × hypoxia} =<0.0001. LSD multiple comparison NORM vs CIH p-values shown in figure. In panel C, compared to baseline in NORM: ^aBsl vs OT p=<0.0001, ^bBsl vs CRH p=<0.0001, ^cBsl vs OT+CRH p=<0.0001. Compared to baseline in CIH: ^eBsl vs OT p=<0.0001, ^fBsl vs CRH p=<0.0001, ^gBsl vs OT+CRH p=<0.0001). (D, E) In NORM neurons, in the presence of OTR or CRHR2 antagonists (block, “-x”), application of OT (D, N=3, n=24) or CRH (E, N=3, n=16), respectively, did not elevate intracellular Ca²⁺. In addition, the elevation of Ca²⁺ in response to of OT or CRH alone was greater than those in the presence of antagonist. 1-way ANOVA with LSD shown. Data bars shown as mean ± SD. All analysis performed on cell “n”.

CRH and its co-activation with OT enhances afferent integration in the nTS after CIH.

Global nTS activity is enhanced via OT and CRH (i.e., Fig 1 & 2), especially after CIH. Visceral sensory afferents serving the augmented peripheral reflexes observed in CIH initially terminate in the nTS where they are processed. To specifically examine the influence of OT and CRH in sensory afferent integration, we recorded glutamatergic excitatory postsynaptic currents (EPSCs) in second-order (or monosynaptic) nTS neurons from nTS slices via patch clamp whole cell recording (Table 1).

Spontaneous (s) EPSCs in the nTS originate from action potential dependent and independent inputs onto the recorded neurons, including those originating from sensory afferents, interneurons and other central nuclei. As shown in the representative examples in Figure 3A, and quantified in Fig 3BC, acute application of OT, CRH, or their co-application did not alter sEPSC frequency or amplitude in either group. sEPSCs were also not altered by vehicle (Table 2, n=4 cells each group).

Figure 3. Spontaneous (s) EPSC frequency and amplitude are not altered by OT and/or CRH following NORM or CIH.

(A) Representative recording of sEPSCs in second-order nTS neurons from normoxia (black) and CIH (blue) rats. Cells were held at −60 mV. Mean data of (B) frequency and (C) amplitude of sEPSCs. Data bars shown as mean ± SD for NORM (N=7, n=8) and CIH (N=6, n=7) exposed neurons. Also shown are individual neurons with connecting lines to denote the direction of response. Neuropeptide application did not alter frequency (2-way RM ANOVA, p_drug = 0.7807, p_Hypoxia = 0.5863, p_{drug × hypoxia} = 0.3017) nor amplitude (2-way RM ANOVA, p_drug = 0.1562, p_Hypoxia = 0.6909, p_{drug × hypoxia} = 0.2830) of sEPSCs based on cell “n”.

In order to examine the influence of OT and CRH specifically on action potential-evoked synaptic transmission between sensory afferents and second-order nTS neurons, we stimulated the afferent-containing solitary tract (TS) and examined the amplitude of the resulting EPSCs (i.e., TS-EPSCs). As shown in the examples and quantified for currents evoked at 0.5 Hz in all neurons that received all three agonist application (Figure 4AB), in normoxic-exposed cells OT alone did not alter current amplitude but CRH alone significantly increased TS-EPSC amplitude. While the addition of OT+CRH also elevated TS-EPSC amplitude, it did not reach significance (p=0.0796) compared with aCSF baseline. After CIH (Figure 4AB), CRH and its co-application with OT significantly elevated current amplitude. OT, CRH and OT+CRH did not alter TS-EPSC rise, decay or half-width nor holding currents in neurons from either group (Table 3).

Figure 4. CIH augments the afferent-evoked EPSC amplitude responses after CRH and OT+CRH.

(A) Example of TS-EPSCs evoked at 0.5 Hz in NORM (black) and CIH (blue)-exposed neurons. CRH alone and the co-application with OT enhance the evoked TS-EPSC amplitude after CIH. (B) Average data (mean ± SD) of TS-EPSC amplitude showing increase after CRH and its co-application with OT by CIH (NORM N=7, n= 9; CIH N=7, n=7). Also shown are individual neurons with connecting lines to denote the direction of response. 2-way RM ANOVA, p_drug = 0.0031, p_Hypoxia =0.9066, p_{drug × hypoxia} =0.8236). (C) Representative TS-EPSCs during 20 Hz stimulation in NORM and CIH-exposed rats. All events across the stimulus were summed and shown in (D). EPSCs increased with CRH & OT+CRH after CIH (2-way RM ANOVA, p_drug = 0.0269, p_Hypoxia =0.0998, p_{drug × hypoxia} =0.3440; NORM N=6, n=6; CIH N=6, n=7). Mean ± SD shown. Single cells and their connecting lines denote the direction of response are also shown. (E) Average data of TS-EPSC amplitude evoked for 10 stimuli at 20 Hz normalized to the first EPSC showing similar synaptic depression after neuropeptide agonists in NORM (2-way RM ANOVA, p_drug = 0.2109, p_ESPCno =0.0001, p_{drug × ESPCno} =0.3453) and CIH (2-way RM ANOVA, p_drug = 0.1128, p_ESPCno =0.0001, p_{drug × ESPCno} =0.1410). (F) Asynchronous events increase after CIH in presence of CRH and with application of OT+CRH. Top inset, example of spontaneous EPSCs before stimulus (Pre), synchronous 20 Hz TS-EPSCs, and asynchronous events following (peak and average) the stimuli. Within Pre-, Peak and Ave, each stacked bar represents an individual drug application. NORM, N=7, n=8, CIH, N=6, n=7. The height of each bar within each column represents the change in events during each drug. In normoxia (2-way RM ANOVA, p_drug = 0.7073, p_{Async Phase} =<0.0001, p_{drug × Async Phase} =0.5328; N=7, n=8) agonist did not alter the aEPSC. Yet, after CIH there was an increase in aEPSCs (2-way RM ANOVA, p_drug = 0.0129, p_{Async Phase} =0.0002, p_{drug × Async Phase} =0.3958; N=6, n=7). LSD multiple comparisons following a main significant effect are shown in each panel for the indicated comparisons. (G) Live-cell calcium imaging demonstrating individual GCaMP6m pre-labeled terminals in NORM and CIH slices before and during three TS stimulation at 20 Hz (example shown in yellow inset). Scale = 20 μm. (H) Average changes in fluorescence reported as ΔF/F of peak amplitude (2-way RM ANOVA, p_drug = <0.0001, p_Hypoxia =<0.0001, p_{drug × hypoxia} =0.0014; NORM N=2, n=240; CIH N=3, n=109). Analysis for all panels performed on cell “n”.

Table 3:

TS-EPSC and cellular properties in Norm and CIH during OT, CRH and OT+CRH

Normoxia	Bsl	OT	CRH	OT+CRH	p-value
Rise time TS-EPSC (ms)	1.42 ± 0.61	1.48 ± 0.74	1.66 ± 0.98	1.85 ± 1.16	0.2541
Decay time TS-ESPC (ms)	8.54 ± 5.16	6.98 ± 2.29	10.16 ± 8.08	8.29 ± 5.86	0.4534
Half-width TS-EPSC (ms)	4.96 ± 1.82	4.98 ± 2.43	5.27 ± 2.97	5.86 ± 3.11	0.4479
Holding Current (pA)	20.49 ± 27.03	14.7 ± 53.00	1.25 ± 29.71	15.28 ± 23.58	0.6536
R input (MOhm)	769.14 ± 428.99	513.86 ± 369.47a	412.92 ± 224.29b	527.16 ± 243.23c	0.0262
CIH	Bsl	OT	CRH	OT+CRH	p-value
Rise time TS-EPSC (ms)	1.38 ± 0.43	1.39 ± 0.48	1.73 ± 1.25	1.60 ± 1.31	0.6323
Decay time TS-ESPC (ms)	6.09 ± 2.55	7.22 ± 2.68	7.15 ± 2.33	6.57 ± 2.42	0.4174
Half-width TS-EPSC (ms)	4.04 ± 1.37	4.52 ± 1.47	4.86 ± 1.94	4.69 ± 1.91	0.1552
Holding Current (pA)	16.91 ± 28.12	7.27 ± 47.67	14.57 ± 23.62	36.17 ± 37.1	0.2845
R input (MOhm)	968.59 ± 814.14	1017.45 ± 819.98	675.57 ± 550.86	772.55 ± 675.57	0.2247

Values are means ± SD. Norm, n=9; CIH n=7. P values from 1-way RM ANOVA within each group. TS-EPSC kinetics from events recorded at 0.5 Hz. Rinput for NORM:

^aBsl vs OT p= 0.0312;

^bBsl vs CRH p=0.0039, and

^cBsl vs OT+CRH p=0.0402.

Consistent with other studies (Peters et al., 2008; Ho et al., 2014), when we specifically evaluated NORM neurons in which OT increased TS-EPSC amplitude, OT enhanced current amplitude in 4 of 11 cells (36%), including 2 cells that received only OT and are not shown in Fig 4B. Examination of these responsive cells, by paired t-test, indicated that OT elevated TS-EPSC amplitude [94.87 ± 86.72 pA (aCSF) to 192.46 ± 74.13 pA (OT), p=0.0544]. A subset of NORM neurons also increased TS-EPSC amplitude to application of CRH [6/9 cells (67%), aCSF, 134.46 ± 67.53 pA vs. CRH, 174.25 ± 67.17 pA, p=0.0103] and OT+CRH [5/9 cells (56%), aCSF, 146.31 ± 68.04 pA vs. OT+CRH, 198.04 ± 79.94 pA, p=0.0266]. After CIH, a majority of nTS neurons responded to singular and combined neuropeptide application; OT [7 of 10 cells (70%) that includes 2 cells that received only OT that are not shown in Fig 4B; aCSF, 65.48 ± 39.36 pA vs OT, 89.76 ± 50.89 pA, p=0.0158], CRH [5/7 cells (71%), aCSF, 110.34 ± 76.90 pA vs CRH, 170.11 ± 83.31 pA, p=0.0203], and OT + CRH [5/7 cells (71%), aCSF, 114.50 ± 70.82 pA vs OT+CRH, 152.11 ± 68.13 pA, p=0.0446].

To mimic the high-frequency afferent discharge that is induced by sensory activation during CIH, such as hypoxia or elevation in arterial pressure, we increased stimulus frequency to 20 Hz (10 events) and examined EPSCs and terminal Ca²⁺. As shown in the examples in Figure 4C, in normoxia-exposed cells, neither OT, CRH nor their co-application altered overall TS-EPSC amplitude. By contrast, in CIH-exposed neurons, application of CRH alone and co-application of CRH+OT significantly elevated current amplitude during 20 Hz stimulation. Examining the overall current response (sum of 10 events), CRH and OT+CRH had no effect in NORM but elevated synaptic throughput in CIH (Figure 4D). Although CIH and neuropeptide application enhanced EPSC amplitude, their application did not affect the frequency-dependent depression of current amplitude typical of nTS neurons (Figure 4E). In addition to TS-evoked synchronous activity, we determined the influence of OT and CRH on asynchronous (a) EPSCs that occur after the 20 Hz stimulation. aEPSCs are a means to continue information transfer across the synapse following a high frequency stimulus bout and are capable of inducing AP discharge after CIH (Kline et al., 2007). After both normoxia and CIH, aEPSCs were present in the immediate 1 second (peak) and overall, 4 seconds (average) after the stimulus train (example in Fig 4F, inset). In neurons from NORM rats, neither OT, CRH nor their co-application altered aEPSCs. After CIH, OT increased only the average. However, CRH and OT+CRH increased peak and average aEPSCs. To examine the potential presynaptic component of OT and/or CRH application, we examined presynaptic Ca²⁺, essential for neurotransmitter release, in pre-labeled GCaMP6m afferent terminals in response to a 20 Hz TS stimulation. In terminals from both groups, stimulation of the TS increased relative terminal Ca²⁺ under baseline aCSF conditions (example in Fig 4G). In normoxia-exposed slices, OT and CRH increased the Ca²⁺ fluorescence that occurred during TS stimulation (Fig 4H). On the other hand, after CIH, the application of CRH and the co-application of OT+CRH attenuated the TS-evoked elevation of terminal Ca²⁺ relative to aCSF baseline (Fig 4H).

Vehicle alone did not alter any of the TS-EPSC parameters examined in NORM or CIH-exposed rats (N=3, N=4 each, Table 2). Taken together, these data suggest CIH enhances the influence of CRH and OT+CRH on glutamate release and neurotransmission, in addition to its somal influence.

OT and CRH induce spontaneous action potential discharge in second-order neurons following CIH.

The intrinsic properties of nTS second-order neurons were examined in whole cell current-clamp experiments where membrane potential and discharge properties of second-order nTS neurons were determined. As in our other protocols, we applied OT (600 nM, 5 min), CRH (300 nM, 5 min), and OT + CRH (5 min) to neurons from NORM and CIH rats. In both groups, resting membrane potential (RMP) was unaltered by vehicle (Table 2) and neuropeptide agonist application (Figure 5A). In addition, NORM spontaneous action potential discharge in these second-order neurons was unaltered by OT, CRH or OT+CRH. However, following exposure to CIH, 66% of neurons that were quiescent during baseline increased action potential (AP) discharge upon the introduction of OT, CRH, and OT+CRH (Figure 5B&C). The percentage of neurons that began to spontaneously fire was greater in CIH than in NORM (Fig 5C, p = 0.0001, Fisher’s exact test), although there were no differences in discharge among agonists, likely due to the variability of discharge.

Figure 5. Neuropeptides promote spontaneous discharge after CIH.

(A) Resting membrane potential (RMP) of NORM and CIH second-order neurons were unaltered by OT, CRH, and OT+CRH (2-way RM ANOVA, p_drug = 0.2841, p_Hypoxia = 0.4164, p_{drug × hypoxia}=0.6956; NORM N=6, n=7; CIH N=6, n=7). Data bars shown as mean ± SD. Individual cells with connecting lines denote direction of response. (B) Representative membrane voltage (Vm) traces showing firing of spontaneous AP discharge (APd) via neuropeptide application after CIH. (C) Fisher’s exact test showing more spontaneous activity after CIH in the presence of OT, CRH and OT+CRH. Data across agonist are grouped together for NORM and CIH. (D) Representative traces of APd (+40 pA current injection) in a NORM and CIH-exposed second-order neuron. (E, F) Quantification of discharge triggered by each depolarizing current amplitude in NORM (E, N=6, n=7) and CIH (F, N=5, n=6). Normoxia (2-way RM ANOVA, p_drug = 0.0890, p_pA = <0.0001, p_{drug × pA}=0.0.0091). Within OT, 50–100 pA current injection vs Bsl, p values ranged between 0.002–0.0064. Within CRH, 40–100 pA injection vs Bsl, p values were between 0.001–0.0352. OT+CRH at 100 pA vs Bsl, p= 0.0207. CIH (2-way RM ANOVA, p_drug = 0.9671, p_pA = <0.0001, p_{drug × pA}=0.9993. All analysis performed on cell “n”.

To assess the excitability of second-order neurons by OT and CRH in response to current-evoked depolarization, discharge was evoked by current injection (10 pA steps, 0–100 pA, 200 ms) from their resting potential while the total number of evoked action potentials was monitored (Figure 5D). Increasing current injection induced AP discharge in both NORM and CIH groups. AP discharge in response to current injection is shown in Figure 5E&F. In NORM rats, OT and CRH reduced the number of action potentials in response to current, whereas during OT+CRH AP discharge was similar to its baseline (Fig 5E). The reduction in discharge in NORM neurons occurred with concurrent reduction in cell input resistance in the presence of OT, CRH and OT+CRH when compared to Bsl (Table 3). Conversely, after CIH discharge in second-order neurons (Fig 5F) and input resistance (Table 3) were unaltered during neuropeptide application.

CIH reduces OT expression in nTS.

OT and CRH neurons within the PVN were examined via immunofluorescence to evaluate if the increased sensitivity of nTS neurons after CIH was potentially due to alteration in their expression. Figure 6A illustrates OT and CRH immunoreactivity in PVN. While the greatest expression for both neuropeptides was seen in the intermediate PVN, the number of OT and CRH neurons as well as total fluorescence intensity in the PVN were similar between NORM and CIH rats (Fig 6B–E). In addition, a small subset of neurons in each group co-expressed both OT and CRH. Of all CRH neurons that co-expressed OT, there was no differences between NORM and CIH (2.1 ± 1.5% vs 3.6 ± 2.4%, respectively, p=0.297, t-test). Likewise, of all OT neurons that co-expressed CRH, expression was similar between NORM and CIH groups (1.5 ± 1.3% vs 3.0 ± 2.4%, respectively, p=0.257, t-test).

Figure 6. OT fibers are reduced after CIH within the nTS.

(A) Representative images of PVN labeled with CRH (red), OT (green) and their merged image (scale= 200 μm/50 μm) in an intermediate PVN section. (B) The number of OT neurons was not altered after CIH. 2-way RM ANOVA, p_levelPVN =< 0.0001, p_hypoxia = 0.4349, p_{levelPVN × hypoxia}=0.0819; N= 5 each. (C) OT immunofluorescence density was also not altered after CIH. 2-way RM ANOVA, p_levelPVN = 0.0006, p_hypoxia = 0.5591, p_{levelPVN × hypoxia}=0.4045; N= 5 each. (D) The number of CRH neurons were similar between groups. 2-way RM ANOVA, p_levelPVN = < 0.0001, p_hypoxia = 0.3469, p_{levelPVN × hypoxia}=0.2855; N= 5 each. (E) CRH immunofluorescence density was comparable between NORM and CIH. 2-way RM ANOVA, p_levelPVN = 0.0007, p_hypoxia = 0.0622, p_{levelPVN × hypoxia}=0.2290; N= 5 each. (F) Example of CRH and OT fiber immunoreactivity in the NORM and CIH nTS. CIH appeared to reduce overall OT immunoreactivity. Of note, the same animals for the PVN were used for the nTS fibers expression (one PVN series was lost). Yellow arrows indicate co-localization of OT and CRH fibers. Scale = 100 μm/50 μm zoom. (G) Quantification of fiber immunoreactivity in the caudal-rostral nTS showing a decrease in OT staining after CIH. 2-way RM ANOVA, p_levelnTS = 0.9671, p_hypoxia = 0.0476, p_{levelnTS × hypoxia}=0.8938; N= 6). (H) CRH immunoreactivity was not altered by CIH. 2-way RM ANOVA, p_levelnTS = 0.1731, p_hypoxia = 0.6509, p_{levelnTS × hypoxia}=0.3404; N= 6). Main hypoxic p-value shown in figure. All analysis performed on rat “N”.

Next, the density of nTS immunoreactive fibers for OT and CRH was examined after NORM or CIH. OT and CRH immunoreactivity were examined across 6 nTS sections rostral and caudal relative to calamus scriptorius (CS). Figure 6F represents OT (green), CRH (red) and their merged image in both groups. Although not quantified, some processes were immunoreactive for OT and CRH (arrows in Fig 6F). Quantitatively, across the rostral-caudal extent of the nTS, after CIH overall immunoreactivity of OT but not CRH fibers decreased compared with the normoxia rats (Figure 6GH).

OT and CRH receptor expression increases in nTS.

The neurophysiological response to OT and CRH occurs through binding of the neuropeptide to its respective receptor. In response to stressors, OT and CRH receptors expression is not static, but rather can be altered in response to specific stimulus or activation (Cadet et al., 2014; Zanos et al., 2015). We focused on the CRH receptor 2 (CRHR2) based on observations that this receptor is the primary isoform expressed in the nTS and alters nTS activity during hypoxia and hypertension (Wang et al., 2018), whereas there is a single OT receptor (OTR) whose expression and function is increased in CIH within the soma of nTS innervating sensory ganglia (Dantzler & Kline, 2020). In addition, our Ca²⁺ imaging in dissociated neurons (Fig 2) indicated that these receptors mediate the Ca²⁺ influx to OT and CRH application. mRNA expression was examined via RNAscope analysis of nTS slices. An example of otr and crhr2 within the nTS is shown in Figure 7A. When the number of fluorescent puncta or dots was quantified (Figure 7BC), the expression of mRNA for OTR and CRHR2 increased following CIH compared to NORM. Although not quantified, a number of neurons expressed the mRNA for both receptors (arrows in Fig 7A).

Figure 7. OTR and CRHR2 expression increase after CIH.

(A) RNAscope demonstrating otr (green) and crhr2 (red) mRNA expression within nTS rat after normoxia or CIH (scale bars = left, 100μm; right, 50μm). Quantification of RNAscope signal of otr (B) and crhr2 (C) in normoxia and CIH rats. (N=4 each, unpaired t-test). (D, E) Representative bands of OTR (MW ~ 43 kDa) and CRHR2 (MW ~ 48 kDa) protein. Immunoblot quantification of OTR (D) and CRHR2 (E) protein expression normalized to total protein (N=4 each, unpaired t-test), and normoxia expression. All analysis performed on rat “N”.

Protein expression was examined by immunoblot, which was initially normalized to total protein and subsequently NORM expression. Both OTR and CRHR2 were observed in the nTS (inset in Fig 7DE). After CIH, OTR was not altered when compared to NORM (Figure 7D). On the other hand, and similar to mRNA, CRHR2 protein increased (Figure 7E, unpaired t-test). Together, this data suggests that CIH alters the expression of OT within the nTS as well as OT and CRHR2 receptor expression.

Tonic influence of OTR and CRHR2 after CIH.

To evaluate if the OT and CRH receptors are tonically active on second-order nTS neurons, especially after CIH, we blocked these receptors while recording EPSCs. We applied antagonists for OTR (L-368,899 HCl, 1 μM) and CRHR2 (K41498, 100 nM) individually and together for 5 min each as in our agonist protocols. As shown in the representative example of sEPSCs in Figure 8A, the application of OTR and CRHR2 antagonists alone and in combination (i.e., dual block) did not alter sEPSC frequency. Quantitatively (Figure 8B), frequency was unaltered by the antagonists following normoxia and CIH. Likewise, sEPSC amplitude was not altered when examined via 2-way RM ANOVA (p_drug = 0.0546, Fig 8C). When examined only within CIH (1-way RM ANOVA), a decrease in sEPSC amplitude was evident after blocking either peptide alone or together (1-way RM ANOVA, p=0.0075, with LSD vs Bsl in Figure 8C). We also examined the extent to which TS-EPSCs were tonically influenced via OTR and CRHR2 during afferent stimulation. TS-EPSC amplitude was not altered in NORM or CIH-exposed cells by receptor blocker (Figure 8D–G). In addition, frequency dependent depression (e.g., as in Fig 4G) was not altered by receptor block (not shown, 2-way RM ANOVA; NORM, p_drug =0.1575, p_EPSCno < 0.0001, p_{drug × EPSCno} =0.2127. CIH, p_drug =0.5932, p_EPSCno =< 0.0001, p_{drug × EPSCno} =0.6370). Holding current also did not change across receptor block compared to their baseline in either group (2-way RM ANOVA; NORM, p_drug =0.0283, p_hypoxia < 0.8465, p_{drug × hypoxia} =0.9566; CIH, p_drug =0.5932, p_hypoxia < 0.0001, p_{drug × hypoxia} =0.6370). There were no individual differences compared to aCSF (LSD). Altogether, these results suggest a modest tonic influence of the neuropeptides within the nTS circuit that modulates sEPSC amplitude but not afferent evoked synaptic transmission.

Figure 8. Minimal tonic influence by OT and CRH on EPSCs.

(A) Example of sEPSCs from NORM (black) and CIH (purple) rats during receptor block (“-x”). Currents were unaltered by neuropeptide. Mean sEPSC (B) frequency (2-way RM ANOVA, p_drug = 0.3834, p_Hypoxia =0.8477, p_{drug × hypoxia} =0.4318) and (C) amplitude (2-way RM ANOVA, p_drug = 0.0546, p_Hypoxia =0.7542, p_{drug × hypoxia} =0.6680; NORM N=8, n=8; CIH N=8, n=8). When examined only within CIH, OT and CRH receptor block reduced sEPSC amplitude. Shown are the LSD multiple comparison p-values. (D) Example TS-EPSCs (0.5 Hz) in response to neuropeptide receptor block. Note currents are not altered. (E) Average data of TS-EPSC amplitude in presence of OTR + CRHR2 block showing no difference in presence of the antagonist (2-way RM ANOVA, p_drug =0.8225, p_Hypoxia =0.4896, p_{drug × hypoxia} =0.9688). (F) Representative TS-EPSCs during 20 Hz stimulation in normoxia and CIH. All events were summed and shown in (G) to illustrate not change (2-way RM ANOVA, p_drug =0.4695, p_Hypoxia =0.8232, p_{drug × hypoxia} =0.6665; NORM N= 8n=8; CIH N=7, n=7). Data bars show mean ± SD. Also shown are individual neurons with connecting lines to denote the direction of response. All analysis performed on cell “n”.

DISCUSSION

The PVN serves as a critical modulator of cardiorespiratory function, in part through its influence on nTS activity (Mack et al., 2007; Kc & Dick, 2010; Ruyle et al., 2023). We have recently shown that inputs from the PVN to nTS critically contribute to the hypoxic ventilatory response in conscious and anesthetized rats (Ruyle et al., 2019; Ruyle et al., 2023). Acute hypoxia also increases PVN neuronal Fos protein, a marker for persistent neuronal activation, including those of OT and CRH phenotype that project to the nTS (Ruyle et al., 2018). Thus, we reasoned that OT and CRH inputs within the nTS may also modulate neuronal function following CIH, which we and others have shown to induce neuronal activation, sympathoexcitation and hypertension (Greenberg et al., 1999a; Kline et al., 2007; Knight et al., 2011; Dantzler & Kline, 2020). Through a series of electrophysiological, imaging, immunohistochemical and molecular studies we demonstrated that CIH enhanced the excitatory response to neuropeptide exposure. Specifically, global nTS discharge was augmented by OT, CRH and OT+CRH in normoxic and CIH-exposed slices, and the increase by CRH and OT+CRH was greater after CIH. Combined neuropeptide application elevated intracellular Ca²⁺ in dissociated neurons and this response was greater after CIH. In second-order nTS neurons specifically, TS-EPSC amplitude was increased by CRH in both groups, yet OT+CRH elevated current amplitude and asynchronous events in only CIH neurons. Intermittent hypoxia reduced OT immunoreactivity throughout the nTS and upregulated OTR mRNA, while it also increased CRHR2 mRNA and protein. Although the responses to exogenous neuropeptides were enhanced by CIH, likely via increased receptor expression, they generally do not tonically influence in vitro activity. Altogether, our data show that CIH enhances the excitatory influence of CRH and OT+CRH on nTS circuits that critically influence autonomic and cardiorespiratory function.

Oxytocin enhances nTS activity in NORM and after CIH.

Examining OT expression within the PVN via immunohistochemistry, we did not observe an alteration in either the number of OT neurons or overall neuropil immunoreactivity after CIH compared to normoxic controls. PVN^OT neurons innervate the nTS with a moderate density (Peters et al., 2008; Uchoa et al., 2013), and similar observations were noted in the current study. In contrast to the PVN, within the nTS CIH reduced OT immunoreactivity and also increased OT receptor mRNA but not protein. The decrease in nTS OT immunoreactivity following CIH may be due to a reduction in OT production or the result of persistent OT release and thus less to detect. In support of persistent OT production and release, we observed similar OT expression within the PVN after NORM and CIH. In addition, we have reported CIH increases plasma OT (Dantzler & Kline 2020). While the present study did not address the extent that release was increased in the nTS, or if this accounts for the reduced protein expression, other studies examining OT (as well as CRH and vasopressin) have shown a reduction in immunoreactivity correlates with increased release (Sawchenko & Swanson, 1982; Sawchenko, 1987; Berkenbosch & Tilders, 1988; Meister et al., 1990; Sanchez et al., 1998) and, as also seen presently with otr mRNA, gene expression (Yamova et al., 2007). While we did not examine specifically the projections originating from the PVN, they have been shown to possess similar expression patterns and suggested to be the sole source of OT innervation in the nTS (Sawchenko, 1987; Rinaman, 1998). Together, these data indicate that OT influences nTS activity after CIH, perhaps through its enhanced release. It is likely this influence is more prominent in vivo in the background of persistent neuronal activity.

We subsequently examined the functional influence of OT on nTS activity. When analyzed across all recorded second-order neurons, OT application in NORM-exposed nTS slices did not affect spontaneous or asynchronous EPSCs or TS-evoked EPSC amplitude. In contrast, previous studies investigating the impact of OT on nTS activity have indicated an influence on synaptic and neuronal function in a subset of responsive neurons under normal conditions. Notably, OT increased miniature (i.e., action potential independent) and spontaneous EPSC frequency, TS-evoked EPSC amplitude and asynchronous activity, suggesting an influence for OT to elevate presynaptic glutamate release. Interestingly, in these latter studies, less than 50% of nTS neurons were OT-sensitive (Peters et al., 2008; Ho et al., 2014). When we evaluated only neurons that increased by at least 10%, we similarly observed that OT increased TS-EPSC amplitude in ~36% (4/11) of the recorded cells as well as augmented intra-terminal (presynaptic) calcium, consistent with OT enhancing glutamate release in this subset of responsive terminals. The apparent discrepancy between the present results and the previous also may be attributed, in part, to the differences in the concentration of calcium in the recording aCSF. We used 2 mM external Ca²⁺ rather than the previous 1 mM designed to reduce release probability and thus maximize the opportunity to observe an increase in synaptic amplitude should it occur via presynaptic release. Taken together, the present study suggests that OT provides an excitatory influence of synaptic transmission in a portion of sensory afferents.

OT has also been shown to influence nTS activity via postsynaptic mechanisms. For instance, OT induces a depolarizing holding current and decreases potassium conductance in some nTS neurons (Peters et al., 2008; Ho et al., 2014). We did not observe in second-order neurons an alteration in holding current when evaluating either the entire sample or only those neurons that respond to OT by an increase in TS-EPSC amplitude. We also did not detect in response to OT changes in membrane potential or spontaneous discharge in these cells. However, OT application reduced AP discharge in response to increasing current injection in these second-order neurons, which was likely due to the decrease in input resistance that would reduce the effectiveness of current to produce depolarization. In contrast to a reduction in current-evoked APs in second-order neurons, OT increased global spontaneous nTS discharge recorded via array analysis and elevated intracellular Ca²⁺ influx in dissociated cells. The OT-induced increase in spontaneous discharge agrees with previous nTS studies (Henry & Sessle, 1989; Raggenbass et al., 1989; McCann & Rogers, 1990). The differences observed among results in the current study, as well as previous studies, may be severalfold. For instance, the OT elevation in nTS global discharge and somal Ca²⁺ likely represent the excitation of a variety of neurons that exist in the nTS network and are directly or indirectly influenced by the TS. Such divergent responses may be also due to differences in the density and location of OTR receptors on distinct neurons, the intracellular signaling pathways activated, and the underlying ionic currents affected. Further studies examining the specific ion channels influenced in these distinct populations are required. Last, OT may modulate neuronal activity via its binding to vasopressin receptors. However, the OT-mediated increase in Ca²⁺ is ablated via prior block of OTR to confirm OT is working through its respective receptor, which is consistent with previous studies by us (Dantzler & Kline, 2020) and others (Peters et al., 2008). Consequently, it can be inferred that within normoxic conditions, OT significantly modulates baseline nTS activity in a portion of second-order nTS neurons with increased effect in higher-order neurons.

Following CIH, exogenous OT also did not alter sEPSC frequency, TS-EPSC amplitude or visceral terminal Ca²⁺ when all neurons were examined. However, a substantial portion of second-order CIH neurons (7/10, 70%) increased TS-EPSC amplitude in response to OT application. OT also elevated asynchronous release after CIH. Together, these data indicate that CIH enhances the OT modulation of afferent glutamate release. As discussed above, OT fiber immunoreactivity was reduced throughout the nTS, which may suggest OT release is elevated in CIH to tonically influence the nTS network. This concept is supported by the modest reduction in sEPSC amplitude following OTR block only after CIH. An OT influence on nTS network activity is also supported by the excitatory influence of OT on global discharge when recorded via arrays and Ca²⁺ elevation in dissociated neurons, which is similar to the elevated spontaneous discharge by OT seen previously in nTS (Henry & Sessle, 1989; Raggenbass et al., 1989; McCann & Rogers, 1990). In contrast to spontaneous APs or Ca²⁺ elevation, OT did not influence AP discharge in second-order neurons in response to increasing current depolarization, which was in contrast to NORM neurons in which OT blunted the elevation of discharge. The reduced effect by OT on current-induced action potentials may be due either to the lack of change in input resistance which helps sustain discharge, or perhaps an excitatory influence that maintains activity. We have shown that following CIH, visceral afferent (nodose) somas exhibit a reduction in AP firing threshold and Vm depolarization in response to OT due to reduced voltage-gated potassium currents (Dantzler & Kline, 2020). Together, these results and others suggest the influence of OT depends on the neural phenotype, synapse order, and the induced signaling cascades. OTRs influence both central and peripheral neurons (Barberis et al., 1992; Li et al., 2023) that impact cardiovascular regulation, stress responsiveness, and the neuroinflammatory cascade (Szczepanska-Sadowska et al., 2021). Our data extends these results to demonstrate an excitatory influence of OT via pre- and postsynaptic mechanisms within nTS circuits after CIH where the autonomic dysfunction and heightened sympathetic tone are common (Jameson et al., 2016).

CRH augments synaptic transmission after CIH.

Similar to oxytocin, CRH expression within the PVN occurred with the highest density within the intermediate rostrocaudal sections, and CIH did not alter overall CRH somal number or neuropil expression. A portion of these PVN^CRH neurons project to the nTS (Ruyle et al., 2018; Wang et al., 2019). When examined specifically in the nTS of NORM and CIH-exposed rats, in contrast to OT, CRH immunoreactivity was similar between groups while CRHR2 mRNA and protein was elevated in CIH. CRHR1 and CRHR2 receptors are expressed in the nTS (Wang et al., 2018), with CRHR2 shown to have the higher expression and primary influence on neuronal and physiological function (Wang et al., 2018). Thus, the CRHR2 receptor was the primary focus of the present study.

Following normoxia exposure, CRH elevated global AP discharge when recorded via array and augmented Ca²⁺ influx in dissociated nTS neurons. When examined specifically in recorded second-order neurons overall, CRH modestly increased TS-EPSC amplitude and presynaptic Ca²⁺. The former parameters were not under tonic influence of endogenous CRHR2 activation as receptor block did not alter synaptic events. Further examination of these NORM second-order neurons revealed that 67% increased TS-EPSCs amplitude in response to CRH. While the excitatory CRH responses are generally consistent with other nTS studies, our data extend those results to provide a site and mechanism of action, especially after CIH. For instance, Wang et al (2018) demonstrated that ~20% of isolated nTS neurons increased Ca²⁺ in response to CRH in the presence of TTX to eliminate AP discharge. A similar percentage of cells responded in the present study ~22% (13/49) after CIH. When PVN^CRH processes were optogenetically activated in the mouse nTS, blood pressure increased due to release of CRH and glutamate, and CRHR2 receptors contributed (Wang et al., 2018). Acute microinjections of the CRH analogue urocortin in the rat nTS decreased blood pressure and sympathetic activity; responses again mediated by CRHR2 (Nakamura & Sapru, 2009). Within the nTS-adjacent dorsal motor nucleus of the vagus (DMV), CRH through the CRHR2 augmented discharge and nTS-evoked EPSC amplitude (Lewis et al., 2002). In the present study, we show in dissociated nTS neurons that the increase in Ca²⁺ in response to CRH was blocked by prior incubation of a CRHR2 antagonist, indicating this receptor is the primary modulator for Ca²⁺ elevation by CRH in the nTS. Together, these data suggest CRH is an excitatory mediator of cardiorespiratory function with the nTS under normal conditions.

Following CIH, the effects of CRH on nTS neurons were augmented. Notably, the administration of CRH alone increased global discharge and intracellular Ca²⁺ within dissociated nTS cells. Moreover, these responses are greater in CIH than their NORM-exposed neurons. In second-order neurons specifically, CRH also augmented overall afferent-driven (TS) EPSC amplitude, and particularly in 71% of responding neurons, as well as asynchronous release. The increase by CRH on TS-EPSC amplitude suggests it may have presynaptic influences after CIH (as it did in normoxia), with an additional influence on presynaptic-mediated aEPSCs. The elevation in postsynaptic Ca²⁺, AP discharge, CRHR2 mRNA and protein indicate additional postsynaptic contributions. Evidence for both comes from expression studies demonstrating CRHR2 is highly expressed in the nodose ganglion (Egerod et al., 2018; Sapio et al., 2020) and nTS (Tan et al., 2017; Ruyle et al., 2018; Wang et al., 2018). Unilateral nodose ganglionectomy reduces CRHR2 expression by ~65%, confirming a large number of these receptors in the nTS are expressed on vagal afferents (Lawrence et al., 2002). Thus, we cannot rule out the possibility that the increase in CRHR2 protein may be due to elevation of protein in afferent terminals as well as nTS neurons. However, the less robust augmentation in terminal Ca²⁺ to TS stimulation after CIH may suggest that increase in CRHR2 does not occur in sensory afferents but rather nTS neurons. In support of our mechanistic studies, block of CRHR2 within the nTS decreases CIH-mediated elevation of arterial pressure (Wang et al., 2018). CRH receptors are G-protein coupled receptors, and have been shown to elevate L- and T-type voltage gated Ca²⁺ channels and reduce transient and delayed rectifier K⁺ channels (Kratzer et al., 2013). Whether similar mechanisms occur in the nTS, especially after CIH, requires further investigation. Nevertheless, our present findings collectively underscore the significance of CRH as a key modulator of synaptic activity in the nTS, particularly following CIH exposure.

OT and CRH together enhance nTS activity after CIH.

Exogenous OT and CRH individually and together influence nTS activity, particularly after CIH. Their endogenous influence on nTS activity, especially during hypoxic stimuli in vivo, is likely due to the considerable innervation of OT and CRH processes throughout the caudal nTS (present study and Peters et al. (2008); Uchoa et al. (2013); Ruyle et al. (2018); Wang et al. (2019)). In addition, we observed a small percentage of PVN neurons and nTS processes that, although not quantified in the latter, co-express OT and CRH in NORM and CIH. These observations suggest activation of such PVN inputs may co-release OT and CRH to influence nTS neuronal and synaptic activity. Additionally, while crhr2 and otr mRNA were expressed within individual neurons, co-expression was also noted to suggest the ability of a given neuron to respond to both neuropeptides. Thus, these expression studies suggest OT and CRH together may have a mutually beneficial effect on nTS properties.

Consistent with the minimal overall influence of OT and CRH individually in NORM-exposed neurons, the co-application of these neuropeptides also produced negligible effects on EPSC in second-order neurons. Further examination of these NORM neurons revealed that 56% of cells increased TS-EPSC amplitude to co-application of OT+CRH. After CIH, in second-order neurons, OT+CRH co-application increased overall TS-EPSC amplitude, with 71% of neurons responding. Further analysis revealed that only 1 of 9 (11%) NORM neurons responded to all 3 drug combinations whereas 4 of 7 (57%) increased TS-EPSC amplitude after CIH. The increased sensitivity to co-application of OT+CRH after CIH was also observed in the greater responses in global discharge seen in arrays and postsynaptic dissociated cell Ca²⁺ influx that reflect activity from neurons of a variety of phenotypes and synapse order, as well as elevated spontaneous AP discharge specifically in second order neurons. Understanding the factors that promote this cooperation in the nTS could provide insights into the adaptive responses to intermittent hypoxia and other cardiorespiratory stressors. As discussed above, CIH increases mRNA and/or protein expression of OTR and CRHR2. Both receptors are coupled to G-proteins (Kimura et al., 1992; Grammatopoulos, 2012; Mitre et al., 2022), but their signaling pathways are often distinctive. CRH receptors activate protein kinase C (PKC) and utilize Ca²⁺ as a secondary messenger (Dermitzaki et al., 2005), while OT receptors activate protein kinase A (PKA) (Bichet D, 2023). This individuality could explain, in part, how the two neuropeptides act together to elevate nTS activity, especially after CIH. The unique activation of second messengers may also influence specific ion channels to summate cellular activity. This is especially evident in the exaggerated somal Ca²⁺ responses in dissociated neurons to exogenous OT+CRH compared to either peptide alone. However, second messengers may also influence each other. For instance, in the hypothalamic arcuate nucleus, PKA phosphorylates Src family kinase, which reciprocally activate PKC. These cooperative interactions ultimately increase action potential discharge (Sun et al., 2020). Nevertheless, the combined influence of OTR and CRHR2 is likely not a global phenomenon as our electrophysiological, imaging and molecular studies suggest that not all neurons express OTR and CRHR2 individually or together. The influence of OT may become greater upon physiological stressors, including exercise, heart failure (Dyavanapalli et al., 2020), acute hypoxia (Ruyle et al., 2018), and as we show presently, intermittent hypoxia.

Alterations in Ca²⁺ homeostasis can profoundly impact neuronal excitability and neurotransmission (Usher-Smith et al., 2006; Maschio et al., 2012). In the context of sensory processing in the nTS after CIH, the reduced influx of terminal Ca²⁺ to TS stimulation during OT+CRH, respective to its baseline, may be associated with attenuated neurotransmitter release. However, we observed an increase in TS-EPSC amplitude in second-order neurons by OT+CRH after CIH. These are seemingly contradictory results. Several mechanisms may explain these results. For instance, we and others have shown that CIH alters the relationship between terminal Ca²⁺ and glutamate release (Kline, 2010; Almado et al., 2012). Perhaps the activation of the PVN and subsequent OT and CRH release in the nTS serves to promote vesicle release and postsynaptic responses in an effort to preserve or even enhance reflex sensitivity. Alternatively, OT+CRH may influence GABA signaling within the nTS network and sensory terminals. For instance, CRH increases GABAergic inhibitory signaling (Phumsatitpong et al., 2020), including within the DMV (Browning et al., 2014). Interestingly, in these latter DMV studies, the addition of OT attenuates the CRH influence on inhibitory signaling. In addition, perinatal high fat diet elevates the OT decrease of GABAergic signaling due to tonic activation of CRH receptors in the DMV (Carson et al., 2023). Future studies will also be needed to identify the specific sensory modality, projection sites, and cellular mechanisms of these OT+CRH interactions on their activated nTS neurons.

Conclusion.

Elevated sympathetic activity following CIH is a key factor contributing to hypertension (Zoccal et al., 2007; Prabhakar et al., 2012). The activation of CRH neurons has been associated with an increase in blood pressure, while their inhibition is linked to its reduction (Wang et al., 2018). Conversely, OT has been shown to lower blood pressure through DMV neurons (Jameson et al., 2016), and ameliorate upper airway obstruction through activation of hypoglossal motoneurons that project to the tongue (Dergacheva et al., 2023). In addition to these neuronal influences, during and following CIH circulating OT and CRH (with the latter possibly having additional effects through adrenocorticotropic hormone (ACTH) and corticosterone release) likely influences nTS network to promote baseline cardiorespiratory function and reflex function. Our work extends these studies to show that these neuropeptides increase nTS activity through pre- and postsynaptic mechanisms, and these responses may contribute to the elevated sympathetic response and increased blood pressure observed in the context of CIH.

KEY POINTS SUMMARY.

• Episodic breathing and chronic intermittent hypoxia (CIH) are associated with autonomic dysregulation, including elevated sympathetic nervous system activity. Altered nucleus Tractus Solitarii (nTS) activity contributes to this response.

• Neurons originating in the paraventricular nucleus (PVN), including those containing oxytocin (OT) and corticotropin-releasing hormone (CRH), project to the nTS and modulate the cardiorespiratory system. Their role in CIH is unknown.

• In this study, we focused on OT and CRH individually and together on nTS activity from rats exposed to either CIH or normoxia control.

• We show that after CIH, CRH alone and with OT increased to a greater extent overall nTS discharge, neuronal calcium influx, synaptic transmission to second-order nTS neurons, and OT and CRH receptor expression.

• These results provide insights into the underlying circuits and mechanisms contributing to autonomic dysfunction during periods of episodic breathing.

DATA AVAILABILITY

All data within the manuscript are presented in completeness in figures.