Medullary Raphe Research Articles

Melanin-concentrating hormone-producing neurons in the hypothalamus regulate brown adipose tissue and thus contribute to energy expenditure.

Melanin-concentrating hormone (MCH) neuron-ablated mice exhibit increased energy expenditure and reduced fat weight. Increased brown adipose tissue (BAT) activity and locomotor activity-independent energy expenditure contributed to body weight reduction in MCH neuron-ablated mice. MCH neurons send inhibitory input to the medullary raphe nucleus to modulate BAT activity. Hypothalamic melanin-concentrating hormone (MCH) peptide robustly affects energy homeostasis. However, it is unclear whether and how MCH-producing neurons, which contain and release a variety of neuropeptides/transmitters, regulate energy expenditure in the central nervous system and peripheral tissues. We thus examined the regulation of energy expenditure by MCH neurons, focusing on interscapular brown adipose tissue (BAT) activity. MCH neuron-ablated mice exhibited reduced body weight, increased oxygen consumption, and increased BAT activity, which improved locomotor activity-independent energy expenditure. Trans-neuronal retrograde tracing with the recombinant pseudorabies virus revealed that MCH neurons innervate BAT via the sympathetic premotor region in the medullary raphe nucleus (MRN). MRN neurons were activated by MCH neuron ablation. Therefore, endogenous MCH neuron activity negatively modulates energy expenditure via BAT inhibition. MRN neurons might receive inhibitory input from MCH neurons to suppress BAT activity.

A thermoregulatory role of the medullary raphe in birds.

The brainstem region medullary raphe modulates non-shivering and shivering thermogenesis and cutaneous vasomotion in rodents. Whether the same scenario occurs in the other endothermic group, i.e. birds, is still unknown. Therefore, we hypothesised that the medullary raphe modulates heat gain and loss thermoeffectors in birds. We investigated the effect of glutamatergic and GABAergic inhibitions in this specific region on body temperature (Tb), oxygen consumption (thermogenesis), ventilation (O2 supply in cold, thermal tachypnea in heat) and heat loss index (cutaneous vasomotion) in one-week-old chicken exposed to neutral (31°C), cold (26°C) and heat (36°C) conditions. Intra-medullary raphe antagonism of NMDA glutamate (AP5; 0.5, 5 mM) and GABAA (bicuculline; 0.05, 0.5 mM) receptors reduced Tb of chicks at 31°C and 26oC, due mainly to an O2 consumption decrease. AP5 transiently increased breathing frequency during cold exposure. At 31°C, heat loss index was higher in the bicuculline and AP5 groups (higher doses) than vehicle at the beginning of the Tb reduction. No treatment affected any variable tested at 36oC. The results suggest that glutamatergic and GABAergic excitatory influences on the medullary raphe of chicks modulate thermogenesis and glutamatergic stimulation prevents tachypnea, without having any role in warmth-defence responses. A double excitation influence on the medullary raphe may provide a protective neural mechanism for supporting thermogenesis during early life, when energy expenditure to support growth and homeothermy is high. This novel demonstration of a thermoregulatory role for the raphe in birds suggests a convergent brainstem neurochemical regulation of body temperature in endotherms.

The mechanisms of neuroplasticity during acclimatization to‐ and deacclimatization from chronic hypercapnia are fundamentally different

A hallmark of respiratory-related diseases, such as chronic obstructive pulmonary disease (COPD), is chronic hypercapnia (CH). Common therapeutic intervention for patients experiencing CH is mechanical ventilation to restore blood-gas homeostasis. While many neurophysiological consequences of CH have been characterized, little is known about the effects reversing CH. The primary goal of the present study was to test the hypothesis that abrupt deacclimatization from CH would significantly alter markers of neuroplasticity and tryptophan metabolism, but not levels of excitatory neuromodulators. Utilizing our goat model of increased inspired CO2 (InCO2)-induced CH, adult female goats were exposed to an InCO2 of 6% for 30 days (d) followed by a return to room air for 24 hours (h). Following 24h of recovery, goats were euthanized, and brainstems were rapidly extracted. Tissue punches of key nuclei throughout the brainstem respiratory network (hypoglossal motor nucleus (XII), nucleus tractus solitarius (NTS)/dorsal motor nucleus of the vagus (DMV), ventral respiratory column (VRC), medullary raphe (MR), ventrolateral medulla (VLM), retrotrapezoid nucleus (RTN), and cuneate nucleus (CN)) were obtained and used for either western blot or HPLC analysis. Changes in glutamatergic signaling, neuroinflammation, tryptophan metabolism, and neuromodulators concentration, following recovery from CH, were assessed. Following 24h of recovery, there were no significant differences in AMPA or NMDA receptor expression or phosphorylation, compared to room air control goats, across all nuclei investigated. Similarly, there were no significant differences in the inflammatory cytokine IL1B following 24h recovery across all nuclei investigated, compared to room air control goats. However, there was significantly (P<0.05) lower expression of key enzymes of tryptophan metabolism (Indolamine 2,3-dioxygenase (IDO); Tryptophan Hydroxylase (TPH) and neuronal markers (NeuN) compared to control within the rostral VLM and MR. Within the solitary complex (NTS & DMV), there was significantly (P>0.05) greater concentrations in norepinephrine (+132%±49), serotonin (+348%±142), and dopamine (+86%±53) at 24h of recovery compared to 30d of CH. We conclude that although specific markers of neuroplasticity and neuroinflammation were unchanged from control during deacclimatization, there were brainstem-site dependent changes in excitatory neuromodulators during deacclimatization from CH. These data suggest that the mechanisms of neuroplasticity during acclimatization to- and deacclimatization from- CH are fundamentally different.

Endogenous Opioid Receptor Activation in the Caudal Medullary Raphe Depresses Respiratory Rate in Decerebrate Rabbits

Background The caudal medullary raphe (CMR) is an essential structure in controlling pain, cardiovascular, and respiratory function (1). Partial reversal of systemic opioid-induced respiratory depression was achieved with naloxone injection into the CMR (2). This study investigated the presence of endogenous opioid activity in the CMR. Methods The study was approved by the local Animal Care Committee and conformed to NIH standards. Adult New Zealand White rabbits (3-4kg) were anesthetized, tracheotomized, ventilated (FiO2 0.6, PCO2 45-55 mmHg), decerebrated and vagotomized. Phrenic nerve activity was recorded from the c5 rootlet, time averaged and used to calculate inspiratory (TI) and expiratory (TE) duration, breaths per minute (BPM), and peak phrenic activity (PPA). Using a multibarrel glass micropipette, we performed the following injection protocols into the CMR in the midline between 2 mm caudal and 1.5 mm rostral to the functionally identified preBötzinger Complex (preBC, (3)) and into the middle medullary raphe (middle MR (2), 2.5-3.5mm rostral to the preBC) in separate sets of animals: i) α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA, 50 μM, 70 nl), stepsize 0.5mm rostro-caudal, 0.5mm dorso-ventral. ii) The non-selective opioid antagonist naloxone (1-5 mM, 700 nl), stepsize 0.5mm rostro-caudal, 1st injection at the ventral edge of neuronal activity, 2nd injection 1.5-2mm dorsal; and iii) the µ-opioid antagonist CTAP and δ-opioid antagonist naltrindole (each 1 mM, 350-700 nl), stepsize same as ii). Statistical comparison with Wilcoxon signed rank test. Median (25-75% range). Results i) In 6 animals, AMPA injection into the CMR caudal to and at the level of the preBC increased respiratory rate by up to 59% and PPA by up to 38% at multiple dorso-ventral levels. This was due to a decrease in TI and TE. In the middle MR, AMPA injection depressed respiratory rate, at times to apnea, through an increase in TI and TE, and also depressed PPA. ii) Naloxone injection into the CMR increased respiratory rate from 30 (23-34) to 37 (30-6) BPM (p=0.031, n=6) through a decrease in TI from 0.9 (0.8-1.0) to 0.7 (0.6-0.8) sec (p=0.031) and TE from 1.2 (1.0-1.6) to 0.9 (0.7-1.2) sec (p=0.031). We observed effects with injections caudal to, at the level of and rostral to the preBC. iii) CTAP injection increased respiratory rate from 26 (22-27) to 34 (29-34) BPM, and subsequent naltrindole injection increased respiratory rate to 43 (37-44) BPM (n=3). In the middle MR, CTAP + naltrindole increased respiratory rate between 2-5 BPM. Conclusion Excitation of the CMR increased respiratory rate over an area of several millimeters. During quiet breathing under moderate hyperoxic hypercapnia in our decerebrate rabbit preparation, respiratory raphe neurons were depressed by endogenous opioid activity. It will be important to further investigate the role of the CMR in opioid-induced respiratory depression. (1) Holtman et al., J Neurosci 1986, 6: 1185-93 (2) Zhang et al., Anesthesiology 2007, 107: 288-97 (3) Cook-Snyder et al., Resp Physiol Neurobiol 2019, 260: 37-52

Role of Medulary Raphe in the Control of Thermoeffectors of Precocious Birds

In mammals, caudal brain regions, such as raphe nuclei, are involved in the modulation of thermoeffectors such as peripheral vasodilation/vasoconstriction and nonshivering and shivering thermogenesis. In birds, it is still completely unknown if the same scenario occurs. The aim of the present study was to verify the effect of GABAA and GLU inhibition in the medullary raphe on body temperature (Tb), oxygen consumption, ventilation and heat loss index (HLI) of precocious birds under neutral (31ºC) and cold (26ºC), using of one-week-old chicks as animal models. To this end, the Tb, oxygen consumption (index of thermogenesis), ventilation and heat loss index were measured before and after intra-raphe microinjections of the GABAA receptor antagonist, bicuculline (0.05 mM and 0.5 mM), and the NMDA receptor antagonist, AP5 (0.5mM and 5mM), or vehicle. Protocols were approved by the local Animal Care Committee (CEUA of FCAV/Unesp, nº 013907/17). Bicuculline (0.5 mM) at 31°C significantly decreased (P<0.05) Tb (vehicle: 41.2±0.1°C; Bicuculline 0.05mM: 40.9±0.1°C; Bicuculline 0.5mM: 40.6±0.2°C) and oxygen consumption (13.9±0.7% decrease) of the animals when compared to the vehicle group. Bicuculline (0.05 mM) at 26°C significantly reduced (P<0.05) Tb (vehicle: 41.4±0.1°C; Bicuculline 0.05mM: 40.6±0.1°C) and oxygen consumption (12.6±1.4% of decrease) of the animals when compared to the vehicle group. AP5 (5 mM) at 31°C significantly decreased (P<0.05) Tb (vehicle: 41.3±0.1°C; AP5 0.5mM: 40.9±0.1°C; AP5 5mM: 40.3±0.2°C) and oxygen consumption (14.7±0.6% decrease) of the animals when compared to the vehicle group. AP5 (0.5 mM) at 26°C significantly reduced (P<0.05) Tb (vehicle: 41.2±0.2°C; AP5 0.5mM: 40.7±0.1°C) and oxygen consumption (17.4±3.1% of decrease) of the animals when compared to the vehicle group. Regarding ventilation, only treatment with AP5 transiently increased breathing frequency during cold exposure (30.6±9.8% increase; P <0.05). At 31ºC, the highest doses of bicuculline and AP5 increased HLI at the beginning of the Tb reduction (P<0.05) and at 26°C there was no difference on HLI. Our results suggest that GABAA and NMDA receptors inhibition in raphe neurons show an inhibitory effect on Tb of chicks affecting thermogenesis under neutral (31ºC) and cold (25ºC) conditions.

Naloxone Injections into the Parabrachial Nucleus/ Kölliker‐Fuse Complex, the preBötzinger Complex and the Caudal Medullary Raphe Reverse Remifentanil‐Induced Respiratory Depression

Background We have shown that naloxone microinjections into the Parabrachial Nucleus/ Kölliker-Fuse (PBN/KF) complex and the preBötzinger Complex (preBC) partially reversed remifentanil-induced respiratory rate depression (1). The magnitude of reversal was dose dependent. The goal of this study was to determine whether additional naloxone injection into the caudal medullary raphe (CMR (2)) could completely reverse the opioid effect. Methods The study was approved by the local Animal Care Committee and conformed to NIH standards. Adult New Zealand White rabbits (3-4kg) were anesthetized, tracheotomized, ventilated, decerebrated and vagotomized. Phrenic nerve activity was recorded from the c5 rootlet, time averaged and used to calculate inspiratory and expiratory duration and peak phrenic activity (PPA). Gridwise localized pressure microinjections of AMPA (50 μM, 70 nl) with a multibarrel glass pipette were used to functionally identify the PBN/KF and preBC. The CMR was defined as the area between 2.0mm caudal and 1.5mm rostral to preBC in the midline. We then injected an “apneic” remifentanil IV bolus (12±2mcg), which interrupted inspiratory activity for at least 60 sec. Next, a remifentanil infusion was started at an “analgesic” dose-rate, defined to achieve respiratory rate depression of 50% of control. At steady-state, we sequentially injected naloxone (1mM, 700nl) into the bilateral PBN/KF, the preBC, and the CMR. We repeated the “apneic” remifentanil bolus after naloxone injection into each area. To determine whether injection order affected the naloxone reversal, in separate sets of animals we microinjected naloxone in the reverse order, or solely into the preBC and CMR. Drug effects were compared with Paired t-tests. Results Figure 1 illustrates the effects of sequential naloxone microinjection into the bilateral PBN/KF, preBC, and CMR. At the end of the naloxone injection sequence, respiratory rate depression was completely reversed (A+B) or even increased above baseline (C). Depression of PPA was completely reversed (p=0.756). Conclusion Naloxone injection into the CMR reversed remifentanil-induced respiratory depression in addition to the reversal achieved in the PBN/KF and preBC at “analgesic” and “apneic” remifentanil doses. Further research is needed to determine whether the naloxone effect in the CMR was partially due to reversal of endogenous opioid-receptor activation (see Palkovic B, Endogenous Opioid-Receptor Activation in the Caudal Medullary Raphe Depresses Respiratory Rate in Decerebrate Rabbits, EB 2021). (1) Palkovic et al., Anesthesiology 2020, A3038 (2) Zhang et al., Anesthesiology 2007, 107: 288-97

Impact of ovariectomy and CO2 inhalation on microglia morphology in select brainstem and hypothalamic areas regulating breathing in female rats

The neural network that regulates breathing shows a significant sexual dimorphism. Ovarian hormones contribute to this distinction as, in rats, ovariectomy reduces the ventilatory response to CO2. Microglia are neuroimmune cells that are sensitive to neuroendocrine changes in their environment. When reacting to challenging conditions, these cells show changes in their morphology that reflect an augmented capacity for producing pro- and anti-inflammatory cytokines. Based on evidence suggesting that microglia contribute to sex-based differences in reflexive responses to hypercapnia, we hypothesized that ovariectomy and hypercapnia promote microglial reactivity in selected brain areas that regulate breathing. We used ionized calcium-binding-adapter molecule-1 (Iba1) immunolabeling to compare the density and morphology of microglia in the locus coeruleus (LC), the caudal medullary raphe, the caudal part of the nucleus of the tractus solitarius (cNTS), and the paraventricular nucleus of the hypothalamus (PVN). Tissue was obtained from SHAM (metaestrus) female rats or following ovariectomy. Rats were exposed to normocapnia or hypercapnia (5% CO2, 20 min). Ovariectomy and hypercapnia did not affect microglial density in any of the structures studied. Ovariectomy promoted a reactive phenotype in the cNTS and LC, as indicated by a larger morphological index. In these structures, hypercapnia had a relatively modest opposing effect; the medullary raphe or the PVN were not affected. We conclude that ovarian hormones attenuate microglial reactivity in CO2/H+ sensing structures. These data suggest that microglia may contribute to neurological diseases in which anomalies of respiratory control are associated with cyclic fluctuations of ovarian hormones or menopause.

Blood pressure drives multispectral tuning of inspiration via a linked-loop neural network.

The respiratory motor pattern is coordinated with cardiovascular system regulation. Inspiratory drive and respiratory phase durations are tuned by blood pressure and baroreceptor reflexes. We hypothesized that perturbations of systemic arterial blood pressure modulate inspiratory drive through a raphe-pontomedullary network. In 15 adult decerebrate vagotomized neuromuscular-blocked cats, we used multielectrode arrays to record the activities of 704 neurons within the medullary ventral respiratory column, pons, and raphe areas during baroreceptor-evoked perturbations of breathing, as measured by altered peak activity in integrated efferent phrenic nerve activity and changes in respiratory phase durations. Blood pressure was transiently (30 s) elevated or reduced by inflations of an embolectomy catheter in the descending aorta or inferior vena cava. S-transform time-frequency representations were calculated for multiunit phrenic nerve activity and some spike trains to identify changes in rhythmic activity during perturbations. Altered firing rates in response to either or both conditions were detected for 474 of 704 tested cells. Spike trains of 17,805 neuron pairs were evaluated for short-time scale correlational signatures indicative of functional connectivity with standard cross-correlation analysis, supplemented with gravitational clustering; ∼70% of tested (498 of 704) and responding neurons (333 of 474) were involved in a functional correlation with at least one other cell. Changes in high-frequency oscillations in the spiking of inspiratory neurons and the evocation or resetting of slow quasi-periodic fluctuations in the respiratory motor pattern associated with oscillations of arterial pressure were observed. The results support a linked-loop pontomedullary network architecture for multispectral tuning of inspiration.NEW & NOTEWORTHY The brain network that supports cardiorespiratory coupling remains poorly understood. Using multielectrode arrays, we tested the hypothesis that blood pressure and baroreceptor reflexes "tune" the breathing motor pattern via a raphe-pontomedullary network. Neuron responses to changes in arterial pressure and identified functional connectivity, together with altered high frequency and slow Lundberg B-wave oscillations, support a model with linked recurrent inhibitory loops that stabilize the respiratory network and provide a path for transmission of baroreceptor signals.

Differential modulation of active expiration during hypercapnia by the medullary raphe in unanesthetized rats.

Active expiration represents an important mechanism to improve ventilation in conditions of augmented ventilatory demand, such as hypercapnia. While a rostral ventromedullary region, the parafacial respiratory group (pFRG), has been identified as a conditional expiratory oscillator, little is known about how central chemosensitive sites contribute to modulate active expiration under hypercapnia. In this study, we investigated the influence of the medullary raphe in the emergence of phasic expiratory abdominal activity during hypercapnia in unanesthetized adult male rats, in a state-dependent manner. To do so, reverse microdialysis of muscimol (GABAA receptor agonist, 1mM) or 8-OH-DPAT (5-HT1A agonist, 1mM) was applied in the MR during sleep and wakefulness periods, both in normocapnic (room air) and hypercapnic conditions (7% CO2). Electromyography (EMG) of diaphragm and abdominal muscles was performed to measure inspiratory and expiratory motor outputs. We found that active expiration did not occur in room air exposure during wakefulness or sleep. However, hypercapnia did recruit active expiration, and differential effects were observed with the drug dialyses in the medullary raphe. Muscimol increased the diaphragm inspiratory motor output and also increased the amplitude and frequency of abdominal expiratory rhythmic activity during hypercapnia in wakefulness periods. On the other hand, the microdialysis of 8-OH-DPAT attenuated hypercapnia-induced active expiration in a state-dependent manner. Our data suggest that the medullary raphe can either inhibit or potentiate respiratory motor activity during hypercapnia, and the balance of these inhibitory or excitatory outputs may determine the expression of active expiration.

Mapping of the excitatory, inhibitory, and modulatory afferent projections to the anatomically defined active expiratory oscillator in adult male rats.

The lateral parafacial region (pFL ; which encompasses the parafacial respiratory group, pFRG) is a conditional oscillator that drives active expiration during periods of high respiratory demand, and increases ventilation through the recruitment of expiratory muscles. The pFL activity is highly modulated, and systematic analysis of its afferent projections is required to understand its connectivity and modulatory control. We combined a viral retrograde tracing approach to map direct brainstem projections to the putative location of pFL , with RNAScope and immunofluorescence to identify the neurochemical phenotype of the projecting neurons. Within the medulla, retrogradely-labeled, glutamatergic, glycinergic and GABAergic neurons were found in the ventral respiratory column (Bötzinger and preBötzinger Complex [preBötC], ventral respiratory group, ventral parafacial region [pFV ] and pFL ), nucleus of the solitary tract (NTS), reticular formation (RF), pontine and midbrain vestibular nuclei, and medullary raphe. In the pons and midbrain, retrogradely-labeled neurons of the same phenotypes were found in the Kölliker-Fuse and parabrachial nuclei, periaqueductal gray, pedunculopontine nucleus (PPT) and laterodorsal tegmentum (LDT). We also identified somatostatin-expressing neurons in the preBötC and PHOX2B immunopositive cells in the pFV , NTS, and part of the RF. Surprisingly, we found no catecholaminergic neurons in the NTS, A5 or Locus Coeruleus, no serotoninergic raphe neurons nor any cholinergic neurons in the PPT and LDT that projected to the pFL . Our results indicate that pFL neurons receive extensive excitatory and inhibitory inputs from several respiratory and nonrespiratory related brainstem regions that could contribute to the complex modulation of the conditional pFL oscillator for active expiration.

Quipazine Elicits Swallowing in the Arterially Perfused Rat Preparation: A Role for Medullary Raphe Nuclei?

Pharmacological neuromodulation of swallowing may represent a promising therapeutic option to treat dysphagia. Previous studies suggested a serotonergic control of swallowing, but mechanisms remain poorly understood. Here, we investigated the effects of the serotonergic agonist quipazine on swallowing, using the arterially perfused working heart-brainstem (in situ) preparation in rats. Systemic injection of quipazine produced single swallows with motor patterns and swallow-breathing coordination similar to spontaneous swallows, and increased swallow rate with moderate changes in cardiorespiratory functions. Methysergide, a 5-HT2 receptor antagonist, blocked the excitatory effect of quipazine on swallowing, but had no effect on spontaneous swallow rate. Microinjections of quipazine in the nucleus of the solitary tract were without effect. In contrast, similar injections in caudal medullary raphe nuclei increased swallow rate without changes in cardiorespiratory parameters. Thus, quipazine may exert an excitatory effect on raphe neurons via stimulation of 5-HT2A receptors, leading to increased excitability of the swallowing network. In conclusion, we suggest that pharmacological stimulation of swallowing by quipazine in situ represents a valuable model for experimental studies. This work paves the way for future investigations on brainstem serotonergic modulation, and further identification of neural populations and mechanisms involved in swallowing and/or swallow-breathing interaction.

GABAergic neurons of the medullary raphe regulate active expiration during hypercapnia.

The parafacial respiratory group (pFRG), located in the lateral aspect of the rostroventral lateral medulla, has been described as a conditional expiratory oscillator that emerges mainly in conditions of high metabolic challenges to increase breathing. The convergence of inhibitory and excitatory inputs to pFRG and the generation of active expiration may be more complex than previously thought. We hypothesized that the medullary raphe, a region that has long been described to be involved in breathing activity, is also responsible for the expiratory activity under hypercapnic condition. To test this hypothesis, we performed anatomical and physiological experiments in urethane-anesthetized adult male Wistar rats. Our data showed anatomical projections from serotonergic (5-HT-ergic) and GABAergic neurons of raphe magnus (RMg) and obscurus (ROb) to the pFRG region. Pharmacological inhibition of RMg or ROb with muscimol (60 pmol/30 nL) did not change the frequency or amplitude of diaphragm activity and did not generate active expiration. However, under hypercapnia (9-10% CO2), the inhibition of RMg or ROb increased the amplitude of abdominal activity, without changing the increased amplitude of diaphragm activity. Depletion of serotonergic neurons with saporin anti-SERT injections into ROb and RMg did not increase the amplitude of abdominal activity during hypercapnia. These results show that the presumably GABAergic neurons within the RMg and ROb may be the inhibitory source to modulate the activity of pFRG during hypercapnia condition.NEW & NOTEWORTHY Medullary raphe has been involved in the inspiratory response to central chemoreflex; however, these reports have never addressed the role of raphe neurons on active expiration induced by hypercapnia. Here, we showed that a subset of GABA cells within the medullary raphe directly project to the parafacial respiratory region, modulating active expiration under high levels of CO2.

Evidence of Progressive Brainstem Pathology after Repeated Seizure Exposure in a Novel Rat Model of SUDEP

Mutations in ion channel genes can alter neuronal excitability and have been linked to seizure disorders in humans. We previously established that a knockout rat model of Kcnj16, which encodes potassium channel subunit Kir5.1, on a Dahl SS background (SSKcnj16−/−) exhibits renal and respiratory impairments, in addition to audiogenic reflex epilepsy. While seizures can be considered benign acute events, patients with uncontrolled epilepsy are at high risk for Sudden Unexplained Death in Epilepsy (SUDEP), predicted to be due to cardiorespiratory failure following repeated seizures. The SSKcnj16−/− rat allows intractable epilepsy to be modeled with modifiable SUDEP risk by regulating seizure number and providing a controlled experimental environment with consistent occurrence of seizure‐induced death. During daily seizure exposure for up to 10 days, we observed progressive exacerbation of post‐ictal respiratory rate and heart rate suppression with spontaneous mortality in 33% of SSKcnj16−/− rats within hours of a seizure. We tested the hypothesis that repeated seizures induces neuroinflammation and dysregulation of adenosine (ADO) and serotonin (5HT) within brainstem cardiorespiratory nuclei, which may contribute to progressive cardiorespiratory suppression and subsequent mortalities in SSkcnj16−/− rats. Quantification of brainstem neurochemicals by HPLC revealed increased ADO and decreased 5HT in seizure‐exposed compared to naïve SSkcnj16−/− rats. These findings appeared to be brainstem‐specific as ADO levels in the cortex of seizure exposed rats were decreased. Western blot analyses of whole brainstem lysates showed trends toward decreased expression of ADO kinase (ADK) and A1 receptor (A1R) in seizure‐exposed rats. We used immunofluorescent staining and confocal imaging to label and quantify total 5HT, serotonergic neurons (tryptophan hydroxylase, TPH), microglia (Iba1), astrocytes (GFAP), total neurons (NeuN), and total nuclei (DAPI) in respiratory brainstem nuclei of SSkcnj16−/− rats exposed to 0, 3, or 10 seizures (n=2/group). Consistent with our HPLC data, 5HT staining was decreased in the pre‐Bötzinger Complex (preBötC), Nucleus Ambiguus (NA) and medullary Raphe (MR) of seizure‐exposed rats, which also showed fewer TPH‐positive neurons in MR. Staining also revealed that seizures may induce brainstem inflammation that resulted in transient increases reactive microglia and progressive astrogliosis in these rats. SSkcnj16−/− rats showed increased Iba1 after 3 seizures but no change after 10 seizures in the preBötC, NA, MR and nucleus of the solitary tract (NTS). GFAP progressively increased after 3 and 10 seizures in preBötC, NTS, and MR. Further studies will focus on validating these finding by increasing the number of animals in experimental groups and evaluating additional inflammatory markers. This preliminary data suggests that dysregulation of ADO and 5HT and/or inflammation in respiratory brainstem nuclei may mediate respiratory suppression and subsequent mortality associated with repeated seizure exposure in SSKcnj16−/− rats, which may have central relevance to the pathophysiology of SUDEP which remains poorly understood.Support or Funding InformationR01 HL122358

Raphe‐Derived Neuromodulators Within the Retrotrapezoid Nucleus (RTN) Modulate Ventilatory Chemoreflexes In Vivo.

Brain chemoreceptors, including serotonergic (5‐HT) neurons in the medullary raphe (MR), play a vital role in the control of breathing through feedback control of blood gas homeostasis and modulating breathing to match metabolic demand. Recent in vitro studies have highlighted important contributions of raphe‐derived neuromodulators (via 5‐HT2 and/or 5‐HT7 receptors) to the chemosensitivity of additional putative brain chemoreceptors, the Phox2b‐expressing retrotrapezoid nucleus (RTN) neurons. However, the importance of raphe‐derived neuromodulation of the RTN in the hypoxic and hypercapnic ventilatory chemoreflexes has not been tested in vivo. We hypothesized that unilateral ablation of serotonergic terminals within the RTN will attenuate the ventilatory chemoreflexes in vivo. To test this hypothesis, 8‐week‐old Sprague‐Dawley rats received unilateral stereotaxic injections of the 5‐HT pre‐synaptic terminal toxin 5,7‐Dihydroxytryptamine (5,7‐DHT; 2 mg/ml, 2 μl vol.) targeted to the RTN under isoflurane anesthesia 30 minutes after pretreatment with desipramine (30mg/kg; IP; to limit off target effects on other aminergic neurons/terminals). A control group received an equal volume injection of vehicle. Breathing was measured via whole body plethysmography while exposed to room air, hypercapnia (7% CO2 challenge), or hypoxia (12% O2 challenge) before and 3, 5, 7, 10, and 14 days post‐lesion. Three days after injection the lesion group (n=6) showed significant decreases in the hypoxic (−44%) and hypercapnic (−43%) ventilatory chemoreflexes. The attenuation of both chemoreflexes was still present 7 days post‐lesion but returned to baseline by 14 days post‐lesion. The control group (n=6) showed a similar pattern, with a significant decrease in both the hypoxic and hypercapnic ventilatory chemoreflexes at 3 days post injection but returned to baseline by 7 days post‐injection. Despite both groups showing decreased responses 3 days following surgery, the 5,7‐DHT group showed a 23% lower response to hypoxia compared to control injections (p = .01), and a 20% lower response to hypercapnia than control injections at 3 days (p = .03), indicating a greater effect from neurotoxic lesions than surgery per se. In ongoing studies, fluorescent microbeads were added to injected solutions to mark/confirm injection location, and immunofluorescent staining will be used to determine if: 1) control injections created a lesion, and 2) if 5‐HT terminals are decreased at 3 and 7 days but return by 14 days post‐DHT lesion. Our preliminary data suggest that serotonergic pre‐synaptic terminals within the RTN facilitate both hypoxic and hypercapnic ventilatory chemoreflexes in vivo.Support or Funding InformationFunded by NIH HL122358.

Retrograde projections to the expiratory oscillator, the parafacial respiratory group, in adult male rats

The parafacial respiratory group (pFRG) is a conditional oscillator known to drive active expiration and enhance ventilation during high metabolic and respiratory demands, through the recruitment of abdominal (ABD) expiratory muscles. At rest and under anesthesia, pFRG neurons appear to be silent, but when disinhibited or activated, they trigger active expiration. Furthermore pFRG modulation appears to be critical for recruitment of expiratory ABD activity during REM sleep. There is extensive evidence for functional modulation of pFRG and its role in active expiration, although no systematic analysis of neuroanatomical projections to the pFRG is currently available. Thus, the objectives of this study are: 1) to identify brainstem regions that directly project to pFRG; and 2) define phenotypes of these projections, mainly excitatory and/or inhibitory.To test the objective 1, we used a retrograde virus (herpes simplex virus) injected unilaterally in the pFRG area in adult male rats that expressed the reporter protein eYFP (enhanced yellow fluorescent protein) driven by the neuronal promoter synapsin in order to track neurons projecting to the pFRG. To determine the phenotype of these projections (objective 2), we combined immunofluorescent labeling for eYFP with different neuronal markers: somatostatin (SST), choline acetyltransferase (CHAT), tyrosine hydroxylase (TH), paired Like Homeobox 2 (PHOX2B) and 5‐hydroxytryptamine (5HT); and we used RNAScope assay to target mRNA for Vesicular Glutamate Transporter (Vglut2); Glutamate Decarboxylase 1 (Gad1) and Glycine Transporter 2 (Glyt2).We identified several brainstem structures that directly project to pFRG ipsilaterally and contralaterally: in the medulla, projecting neurons were located in the ventral respiratory column (Bötzinger Complex, preBötzinger Complex, rostral ventral respiratory group, retrotrapezoid nucleus ‐ RTN and pFRG), the nucleus of the solitary tract (NTS), the reticular formation (RF), the vestibular nuclei (extended to the pons and midbrain), the medullary raphe and the nucleus of Roller. Several projecting neurons were also located rostrally in the pons and midbrain within the Kölliker‐Fuse nucleus, the parabrachial nucleus, the periaqueductal grey, the pedunculopontine nucleus (PPT), the laterodorsal tegmental (LDT) and the A5 area. Overall, the ipsilateral distribution was more extensive than in the contralateral side of the virus injection. Phenotype analysis demonstrated the presence of both excitatory (Vglut2+) and inhibitory (Gad1/Glyt2+) cells projecting to pFRG deriving from all the above regions, in addition to PHOX2B+ cells located in the RTN, NTS and part of the RF. Moreover, we observed lack of direct projections to the pFRG area from TH+ cells in the NTS, A5 and Locus coeruleus; 5HT+ cells in the raphe nuclei; SST+ cells in the preBötzinger Complex and CHAT+ cells in the PPT and LDT.The present study provides evidence that pFRG neurons receive excitatory and inhibitory inputs from several breathing‐ and non‐breathing related brainstem regions that could be involved with their conditional activity and the modulation of active expiration in various physiological conditions.Support or Funding InformationCIHR 156338, NSERC 435843; WCHRI Fellowship (VB)

Neurons in the rat superior colliculus translate visual threat into autonomic responses via a relay in the rostral ventromedial medulla

The superior colliculus (SC) is a sensory integration hub in the dorsal brainstem where multimodal information is combined and, depending on the saliency of the competing sensory inputs, appropriate motor commands and supportive autonomic changes initiated. In rodents, the SC is indispensable for initiating behavioral responses to stereotypical visual stimuli that resemble approaching objects, such as looming (an expanding overhead black circle). Here we report that presentation of overhead looming or naturalistic stimuli drove acute surges in blood pressure in telemetered conscious rats, an effect that was replicated by optogenetic stimulation of the deep SC (dSC). dSC stimulation also evoked increases in respiratory rate and tail vasoconstriction in the absence of detectable anxiety‐like behaviors and continued to exert excitatory effects on heart rate, respiratory rate, and sympathetic nerve activity under urethane anesthesia. The objective of the current study was to identify the central pathways responsible for mediating these physiological effects.Anterograde labeling of dSC neurons revealed a previously uncharacterized axonal projections to brainstem cell groups associated with arousal and autonomic control, including the locus coeruleus, A5 group and, most extensively, neurons within a region that spanned the medullary gigantocellular and raphe cell groups, collectively called the rostral ventromedial medulla (RVMM). Optogenetic stimulation of dSC terminals within the RVMM recapitulated some of the sympathetic and respiratory effects evoked by dSC stimulation, and optogenetic dSC activation evoked powerful excitatory effects on extracellular recordings of putative RVMM sympathetic premotor neurons, suggesting that elements of the physiological response dSC stimulation are mediated by direct activation of medullary autonomic neurons.To investigate the contribution of environmental stimuli to the excitability of this pathway we conducted single‐unit recordings of SC neuronal responses to visual and acoustic stimuli using high‐density silicon probes. In addition to responding to stereotypical audio‐visual looming stimuli, we report the presence of SC neurons with higher‐order visual capabilities relating to object detection that differ by subregion and are several orders of magnitude more complex than previously recognized. These tuning properties were also found in subpopulations of opto‐tagged SC neurons that project to the RVMM. Our data suggest that the SC is not only capable of nuanced object recognition, but can translate naturalistic visual cues into fast‐acting autonomic changes via direct medullary projectionsSupport or Funding InformationResearch was supported by the NHMRC and Hillcrest Foundation

Comparison of intrinsic chemosensitivity of acutely dissociated medullary serotonergic and retrotrapezoid nucleus neurons

ObjectiveSerotonergic neurons of the medullary raphe and Phox2b expressing neurons of the retrotrapezoid nucleus (RTN) are both putative respiratory chemoreceptors. Here we compare their intrinsic response to acidosis using patch clamp recordings after acute dissociation in vitro.MethodsThe midline medulla and rostral ventrolateral medulla were dissected from mice (&gt;P10) that expressed fluorescent reporters: 1) Phox2b‐Cre::Floxed‐tdTomato:ChAT‐GFP for RTN, and; 2) ePet‐EYFP for 5‐HT. Cells were treated with papain, mechanically dissociated, plated on CELLocate coverslips and allowed to adhere for a variable period before recordings were made. RTN neurons were identified as those that expressed tdTomato but not GFP. In some cases, immunohistochemistry was used to ensure that RTN neurons did not express tyrosine hydroxylase. Patch clamp recordings were made with standard methods previously used to study chemosensitivity of these neurons in culture and slices. Neurons were considered “chemosensitive” if they increased their firing rate by more than 20% in response to a decrease in pH from 7.4 to 7.2 (due to an increase in PCO2 from 5% to 9%, balance O2). Neurons were chosen for recordings only if they were isolated from other neurons. In some cases a lack of synaptic connections was verified with immunohistochemistry for MAP2 using confocal microscopy. The methods used to dissociate and record 5‐HT and RTN neurons were identical.Results45% of 5‐HT neurons (n=84) were chemosensitive, with a mean increase in firing rate by 135%+/−115% (mean+/−SD). 38% of these neurons increased firing rate by more than 50% (i.e. to 150% of baseline firing rate. In contrast, only 29% of RTN neurons (n=21) increased their firing rate by more than 20%, and none of them increased their firing rate by more than 50%. Chemosensitive RTN neurons increased firing rate by a mean of 29%+/−8%. The mean firing rate of chemosensitive 5‐HT neurons was 1.1+/−0.6 Hz at pH 7.4 and 2.15+/−0.9 Hz at pH 7.2 (n=38; p&lt;0.0001). The mean firing rate of chemosensitive RTN neurons was 3.3+/−2.3 Hz at pH 7.4 and 4.3+/−2.8 Hz at pH 7.2 (n=6; p&lt;0.05). Recordings were performed in a separate cohort of 5‐HT neurons using a change in CO2 from 5% to 9% in 21% O2 (balance N2) and their response was not different than in 95% O2.Conclusions5‐HT neurons had a high degree of intrinsic (cell‐autonomous) chemosensitivity to small changes in pH that was similar to that reported for these neurons in acute brain slices and in cell culture. Intrinsic chemosensitivity of RTN neurons was uncommon. When present it was small, similar to that previously reported in a small percentage of neurons in the neocortex, hippocampus and cerebellum. RTN neurons may play a more important role either as “emergency chemoreceptors” (i.e. responding to very large pH changes) or as relays from other chemoreceptors than as pH sensors themselves.Support or Funding InformationNIH U01 NS090414

Acute Intermittent Hypoxia and Cortical Evoked Potentials in Human Diaphragm

Episodic carotid body activation via acute intermittent hypoxia (AIH) elicits respiratory motor plasticity (known as long‐term facilitation [LTF]) in humans and other mammals. In rodents, AIH‐induced LTF requires serotonergic neuron activation in the medullary raphe nuclei and episodic serotonin release within respiratory motor nuclei, thereby strengthening descending synaptic inputs to spinal respiratory motor neurons. Humans exhibit ventilatory LTF under certain circumstances, although the details of its underlying mechanism are less clear. Recent evidence demonstrates that AIH enhances corticospinal motor evoked potentials (MEP) in a non‐respiratory motor pool innervating the first dorsal interosseous muscle of the hand. It is not known if AIH has similar effects on the phrenic/diaphragm system in humans. Thus, we tested the hypothesis that AIH augments diaphragm MEP amplitude. Twelve healthy young adults were recruited (age = 30 ± 7 years; 5 female). On Day 1, diaphragm MEPs and compound muscle action potentials (CMAPs) were assessed using transcranial (TMS) and cervical (CMS) magnetic stimulation, respectively. The diaphragm electromyogram was recorded using surface electrodes on the chest wall. Reliability of TMS‐ and CMS‐evoked potentials were determined prior to testing. Following baseline measures, subjects were exposed to AIH (15, 1‐minute episodes of 9% O2, with 1‐minute normoxic intervals) via a facemask attached to a hypoxic generator. Cardiovascular responses (heart rate, blood pressure and oxygen saturation) were monitored throughout. Stimulus‐response curves were determined for each subject by delivering stimuli prior to and 30 minutes post AIH. On Day 2 (at least 7 days later), 6 subjects returned for a normoxic (sham) trial. Subjects were blinded to the gas composition delivered on each visit. No significant AIH effects were observed in MEP (change = −1 ± 7%, p= 0.96) or CMAP amplitude (+1 ± 5%, p= 0.97) across the stimulus‐response curve. At an intensity equivalent to 50% of baseline peak amplitude (V50), the mean changes in MEP and CMAP amplitude were +4 ± 14% (p= 0.81) and −5 ± 6% (p= 0.75), respectively. Further, no effect of normoxic sham exposures was found during the control visit (MEP = −8 ± 9%, p= 0.97; CMAP = +7 ± 31%, p= 0.79). Between‐block coefficients of variation for TMS‐ and CMS‐evoked potentials were 22.9% and 6.9%, respectively. Within‐session reproducibility of evoked responses to TMS and CMS were good to excellent (ICC &gt; 0.8). In conclusion, AIH alone does not influence cortical evoked potentials in the diaphragm as it does in a non‐respiratory motor system. We speculate that repetitive daily AIH and/or concomitant hypercapnia during AIH may reveal plasticity in diaphragm cortical evoked potentials in humans.Support or Funding InformationThis work was supported by the UF McKnight Brain Institute, DoD (SCIRP), NIH R01 HL147554 &amp; OT2OD023854, Brooks Rehabilitation, and the Brooks PHHP Research Collaboration.

Serotonin and substance P: Synergy or competition in the control of breathing.

Numerous neurotransmitters identified in the central nervous system play role in ventilatory control. This mini-review focuses on the respiratory effects of two neurotransmitters: serotonin (5-HT) and substance P (SP). We discuss their co-localization in medullary raphe nuclei, expression of proper receptors within the specific regions of respiratory related structures and contribution to respiratory rhythmogenesis.

Thyrotropin releasing hormone (TRH) in the brainstem: Role of Pavlov’s vagally mediated cephalic phase of gastric secretion

Pavlov’s seminal observation that sham feeding in dogs stimulates gastric acid secretion through the vagus nerve pioneered the concept of the cephalic phase of digestion. This has been subsequently extended to a wide range of mammals including humans. In the last decades, experimental evidence in rats established that the three amino acid peptide thyrotropinreleasing hormone (TRH) expressed in the brainstem plays a key role in the vagal stimulation of digestive secretory-motor function. The dorsal motor nucleus of the vagus (DMN) neurons expresses TRH receptor subtype 1 (TRH-R1) and receives input of TRH containing fibers arising from TRH synthesizing neurons in medullary raphe nuclei. The activation of TRHTRH-R1 signaling excites the firing of DMN neurons leading to the activation of vagal efferent discharges and gastric myenteric cholinergic neurons. This results in a vagally mediated and atropine-sensitive stimulation of gastric secretory and propulsive motor function, along with duodenal and pancreatic secretion. Importantly, the blockade of TRH or TRH-R1 in the brainstem inhibits the gastric acid response to sham feeding in rats. Collectively, these convergent data support the physiological relevance of medullary TRHTRH-R1 signaling as the end effector of the vagally mediated stimulation of digestive process in the cephalic phase.