Kazuhisa Ezure

Distribution of medullary respiratory neurons in the rat

In Nembutal-anesthetized and spontaneously breathing rats, a total of 226 respiratory neurons were recorded in the medulla extending from the caudal end of the facial nucleus to 1 mm caudal to the obex. They were classified into inspiratory (I) and expiratory (E) neurons by their temporal relationships to diaphragm EMGs. One hundred and seventeen I and 108 E neurons were identified. I and E neurons were further classified into augmenting, decrementing, and other types based on their firing patterns. Almost all the respiratory neurons recorded were located around the nucleus ambiguus and the nucleus retroambigualis, corresponding to the ventral respiratory group (VRG) of the cat. On the other hand, only a few respiratory neurons were identified around the ventrolateral nucleus of the solitary tract, corresponding to the dorsal respiratory group of the cat. In the VRG, 3 subgroups were distinguished rostrocaudally. One group of E neurons was located ventrally to the rostral part of the nucleus ambiguus, presumably corresponding to the Bötzinger complex defined in the cat. Another group of E neurons extended caudally beyond the obex, from the caudal portion of the nucleus ambiguus through the nucleus retroambigualis. Between these two groups of E neurons, an assembly of predominantly I neurons existed in the vicinity of the nucleus ambiguus. These characteristics of distributions were basically similar to those of the VRG of the cat.

Activity of bulbar respiratory neurons during fictive coughing and swallowing in the decerebrate cat.

1. The behaviour of medullary respiratory neurons was studied during fictive coughing and swallowing evoked by electrical stimulation of the superior laryngeal nerve (SLN) in decerebrate, paralysed and artificially ventilated cats. Fictive coughing, swallowing and respiration were monitored by recording activities of the phrenic, hypoglossal and abdominal nerves. 2. Extracellular recordings were made from respiratory neurons in the ventral respiratory group (VRG) and in the Bötzinger complex (BOT). The neuronal types analysed included decrementing inspiratory neurons (I‐DEC), augmenting expiratory neurons (E‐AUG) and decrementing expiratory neurons (E‐DEC) from the BOT area, and augmenting inspiratory neurons (I‐AUG) and augmenting expiratory neurons (E‐AUG) from the VRG area. 3. During fictive coughing, all the inspiratory and expiratory neurons were active during the inspiratory and expiratory phases of coughing, respectively. The firing of both I‐DEC and I‐AUG neurons was increased and prolonged in association with the augmented inspiratory activity of the phrenic nerve. The activity of E‐AUG neurons of the VRG did not parallel the abdominal nerve activity, suggesting the existence of additional neurons which participate in the generation of abdominal nerve activity during fictive coughing. 4. During fictive swallowing, half of I‐DEC neurons fired transiently at the onset of hypoglossal bursts associated with swallowing; the firing was suppressed during the rest of the hypoglossal bursts. Other I‐DEC neurons were silent during hypoglossal bursts. Some I‐AUG neurons fired during the initial half of hypoglossal bursts, and others were silent. The brief phrenic activity accompanying the swallowing might have originated from this activity in I‐AUG neurons. The discharges of all E‐AUG neurons (BOT and VRG) and the majority of E‐DEC BOT neurons were suppressed during swallowing. 5. We conclude that these five types of respiratory neurons of the BOT and VRG are involved in the generation of the spatiotemporally organized activity of coughing and swallowing, and that at least a part of the neuronal network for respiration is shared by networks for these non‐respiratory activities.

Efferent projections of inspiratory neurons of the ventral respiratory group. A dual labeling study in the rat

The efferent projections of the medullary respiratory neurons of the rat were studied using an anterograde tracer, Phaseolus vulgaris leucoagglutinin (PHA-L). In Nembutal-anesthetized rats, PHA-L was iontophoretically applied to (1) the area of inspiratory neurons of the ventral respiratory group (VRG) around the nucleus ambiguus, or (2) the area ventrolateral to the solitary tract. In addition, a fluorescence retrograde tracer, Fast blue (FB), was injected into the cervical phrenic nerve several days after the PHA-L injection. When PHA-L was injected into the area of predominantly inspiratory neurons of VRG, dense PHA-L-labeled axons were observed bilaterally in the spinal cord: the ipsilateral projections were noticeably denser than the contralateral ones. Fine axonal branches were distributed around a column of the phrenic motoneurons and boutons were observed on the somata of the FB-labeled motoneurons, suggesting monosynaptic connections between VRG inspiratory neurons and phrenic motoneurons. On the other hand, when PHA-L was injected into the area ventrolateral to the solitary tract, only a few descending axons to the spinal cord were seen bilaterally. No contacts between the PHA-L-labeled axons and the FB-labeled phrenic motoneurons were observed. The brainstem projections of the VRG were found bilaterally in the nuclei ambigui, Cajals interstitial nuclei of the solitary nucleus, the solitary nuclei, the hypoglossal nuclei, the Kölliker-Fuses nuclei, and the subcoeruleus areas.

Axonal projections of pulmonary slowly adapting receptor relay neurons in the rat.

We elucidated efferent projections of second‐order relay neurons (P‐cells) activated by afferents originating from slowly adapting pulmonary receptors (SARs) to determine the central pathway of the SAR‐evoked reflexes. Special attention was paid to visualizing the P‐cell projections within the nucleus tractus solitarii (NTS), which may correspond to the inhibitory pathway from P‐cells to second‐order relay neurons (RAR‐cells) of rapidly adapting pulmonary receptors. P‐cells were recorded from the NTS in Nembutal‐anesthetized, paralyzed, and artificially ventilated rats. First, we used electrophysiological methods of antidromic mapping and showed that the majority of the P‐cells examined projected their axons to the caudal NTS and to the dorsolateral pons corresponding to the parabrachial complex. Second, a mixture of HRP and Neurobiotin was injected intracellularly or juxtramembranously into P‐cells. (1) Stained P‐cells (n = 7) were located laterally to the solitary tract and had dendrites extending characteristically along the lateral border of the solitary tract. (2) All P‐cells had stem axons projecting to the ipsilateral medulla. Of these, the axons from five P‐cells projected to the nucleus ambiguus and its vicinity with distributing boutons. Some of these axons further ascended in the ventrolateral medulla, and distributed boutons in the areas ventral or ventrolateral to the nucleus ambiguus. (3) All the P‐cells had axonal branches with boutons in the NTS area. In particular, axons from three P‐cells projected bilaterally to the medial NTS caudal to the obex, i.e., to the area of RAR‐cells. These results show anatomic substrates for the connections implicated in the P‐cell inhibition of RAR‐cells as well as the SAR‐induced respiratory reflexes. J. Comp. Neurol. 446:81–94, 2002.

Swallowing‐related activities of respiratory and non‐respiratory neurons in the nucleus of solitary tract in the rat

Swallowing‐related activity was examined in respiratory (n= 60) and non‐respiratory (n= 82) neurons that were located in and around the nucleus of the solitary tract (NTS) in decerebrated, neuromuscularly blocked and artificially ventilated rats. Neurons that were orthodromically activated by electrical stimulation of the superior laryngeal nerve (SLN) were identified, and fictive swallowing was evoked by SLN stimulation. The pharyngeal phase of swallowing was monitored by hypoglossal nerve activity. Two types of non‐respiratory neurons with swallowing‐related bursts were identified: ‘early’ swallowing neurons (n= 24) fired during periods of hypoglossal bursts, and ‘late’ swallowing neurons (n= 8) fired after the end of hypoglossal bursts. The remaining non‐respiratory neurons were either suppressed (n= 21) or showed no change in activity (n= 29) during swallowing. On the other hand, respiratory neurons with SLN inputs included 56 inspiratory and four expiratory neurons. Inspiratory neurons were classified into two major types: a group of neurons discharged simultaneously with hypoglossal bursts (type 1 neurons, n= 19), while others were silent during bursts but were active during inter‐hypoglossal bursts when swallowing was provoked repetitively (type 2 neurons, n= 34). Three of the expiratory neurons fired during hypoglossal bursts. Many of the swallowing‐related non‐respiratory neurons and the majority of the inspiratory neurons received presumed monosynaptic inputs from the SLN. Details of the distribution and firing patterns of these NTS neurons, which have been revealed for the first time in a fictive swallowing preparation in the rat, suggest their participation in the initiation, pattern formation and mutual inhibition between swallowing and respiration.

Decrementing expiratory neurons of the Bötzinger complex

SummaryIn Nembutal-anesthetized, immobilized, and artificially ventilated cats with intact vagus nerves, extracellularly recorded activities of expiratory (E) neurons whose firing patterns were of decrementing, or the early expiration type (E-DEC neurons) were recorded in the vicinity of the Bötzinger complex (BÖT). A total of 32 E-DEC neurons which were not vagal motoneurons was studied by determining 1) where they were distributed, 2) how their firing was modulated by lung inflation, and 3) if they projected their axons to the respiratory area of the brain stem. E-DEC neurons were located ventromedially to the retrofacial nucleus and were intermingled with E neurons of the augmenting type (E-AUG neurons), which were abundant and representative of neurons in the BÖT. Firing of 25 E-DEC neurons was facilitated by lung inflation, indicating the existence of excitatory input from stretch receptors of the lungs, although the firing of 7 other neurons was not affected. On the other hand, firing of surrounding E-AUG neurons was suppressed by lung inflation. The E-DEC neurons fired in the E phase during a brief stop of the ventilator, indicating that they received central respiratory rhythm. However, they almost never fired during the inspiratory (I) phase even when the lungs were strongly inflated, indicating the existence of strong central inhibition during the I phase. Eight E-DEC neurons were tested for antidromic activation from the contralateral brain stem and the spinal cord by microstimulation. They projected their axons into the contralateral BÖT and the ventral respiratory group (VRG), and these axons extended caudally beyond the obex. None of them were antidromically activated from the C4-5 spinal cord. These E-DEC neurons which receive central respiratory rhythm and peripheral feedback input from the lungs may have wide influence on respiratory neurons of the brain stem.

Excitatory and inhibitory synaptic inputs shape the discharge pattern of pump neurons of the nucleus tractus solitarii in the rat

Abstract The second-order relay neurons of the slowly-adapting pulmonary stretch receptors (SARs) are called pump neurons (P cells) and are located in the nucleus tractus solitarii (NTS). We have shown recently that P cells do not act merely as simple relay neurons of SAR afferents but also receive rhythmic inputs from the central respiratory system. This study aimed to analyze two aspects of the respiratory inputs to P cells: (1) suppression of P cell firing at early inspiration (eI suppression) and (2) facilitation of P cell firing at around the period from late inspiration to early expiration (IE facilitation). This study employed extracellular recordings combined with iontophoretic applications of neuroactive drugs to single P cells, in Nembutal-anesthetized, paralyzed, and artificially ventilated rats. The results showed that several excitatory and inhibitory neurotransmitters were involved in these synaptic events. First, the glycine antagonist strychnine and the GABAA antagonist bicuculline were applied to identify the neurotransmitters acting in eI suppression. Strychnine greatly diminished eI suppression, but bicuculline had little effect. This suggested that eI suppression was elicited by inspiratory neurons that were glycinergic and had a decrementing firing pattern. Second, on the other hand bicuculline markedly enhanced IE facilitation as well as the baseline frequency of P cell firing. The enhancement of IE facilitation was distinctive even when the effects of increased baseline firing on this enhancement were taken into account. Third, IE facilitation was diminished by applications of the NMDA glutamate receptor antagonists D-2-amino-5-phosphonovaleric acid (APV) and dizocilpine (MK-801). These results suggested that glutamatergic synapses on P cells from some unidentified respiratory neurons form excitatory inputs for IE facilitation and GABAA receptor-mediated processes control the strength of IE facilitation, possibly at the presynaptic level. Finally, iontophoretic application of the non-NMDA glutamate receptor antagonist, 6-cyano-7-nitroquinoxaline-2, 3-dione disodium (CNQX), almost completely abolished P cell firing in response to both lung inflation and electrical stimulation of the vagus nerve. This confirmed the previous report that glutamate is the primary neurotransmitter at the synapses between SAR afferents and P cells. We concluded that complicated synaptic inputs involving glycinergic and GABAergic inhibitions, and non-NMDA and NMDA glutamate receptor-mediated excitations form the basic pattern of P cell firing.

Overall distribution of GLYT2 mRNA-containing versus GAD67 mRNA-containing neurons and colocalization of both mRNAs in midbrain, pons, and cerebellum in rats

We aimed to clarify the overall distribution of glycinergic neurons in the midbrain, pons, and cerebellum in rats, using in situ hybridization for mRNA encoding glycine transporter 2 (GLYT2), which reliably detects glycinergic cell bodies. We combined this method with in situ hybridization for mRNA encoding glutamic acid decarboxylase isoform 67 (GAD67), and have presented for the first time global and detailed views of the distribution of glycinergic neurons in relation to GABAergic neurons. In addition to this single-detection study, we performed double-detection of GLYT2 mRNA and GAD67 mRNA to determine the distribution of neurons co-expressing these mRNAs. We have shown that many areas of the brainstem and cerebellum, not only areas where previous immunohistochemical studies have specified, involve double-labeled neurons with GLYT2 and GAD67 mRNAs. In particular, when lightly labeled GLYT2 mRNA-positive neurons were distributed within the area of GAD67 mRNA-positive neurons, almost all such GLYT2 mRNA-positive neurons were GAD67 mRNA-positive. Areas or neuron groups expressing exclusively GLYT2 mRNA or GAD67 mRNA were rather limited, such as the superior colliculus, nucleus of the trapezoid body, and Purkinje cells. The present study suggests that the corelease of glycine and GABA from single neurons is more widespread than has been reported.

Respiration-related afferents to parabrachial pontine regions

The dorsolateral pons around the parabrachial nucleus is an important participant in respiratory control. This area involves various respiration-related neurons, and their respiratory modulation is thought to arise from afferents from medullary respiratory neurons. Today, however, only a limited number of afferent sources have been identified. First, relatively well-characterized afferents to the pons are those originating from two types of the lung stretch receptors, slowly adapting and rapidly adapting receptors. That is, the majority of the second-order relay neurons of these receptors in the nucleus tractus solitarii project to the pons. Second, certain types of respiratory neurons of the medullary respiratory groups are either known to or presumed to project to the pons. For instance, major inhibitory neurons of the Botzinger complex, augmenting and decrementing expiratory neurons, send afferents to the pons. This article overviews such afferents and discusses their connectivity with pontine neurons.

GABA, in some cases together with glycine, is used as the inhibitory transmitter by pump cells in the Hering-Breuer reflex pathway of the rat

The Hering-Breuer reflex is one of the fundamental respiratory reflexes and is mediated by second-order relay neurons of the slowly adapting lung stretch receptors. These neurons, which are called pump cells, are located in the nucleus tractus solitarii and include a population of inhibitory neurons. We aimed to determine which transmitter, GABA or glycine, the inhibitory pump cells use. In addition, we examined whether or not second-order relay neurons of the rapidly-adapting lung stretch receptors (RAR-cells), whose excitatory or inhibitory nature is not known, use these inhibitory neurotransmitters. In Nembutal-anesthetized, neuromuscularly blocked and artificially ventilated rats, we labeled pump cells (n=33) and RAR-cells (n=26) with Neurobiotin and processed the tissues for detection of mRNA encoding either glutamic acid decarboxylase isoform 67 (GAD67) or glycine transporter 2 (GLYT2) using in situ hybridization. The pump cells were located in the interstitial nucleus and its vicinity and the RAR-cells in the commissural subnucleus. The majority (64%) of the pump cells examined for GAD67 mRNA and many (26%) of the pump cells examined for GLYT2 mRNA expressed respective mRNAs. Of the eight pump cells in which both mRNAs were double-detected, three expressed both mRNAs and one expressed GAD67 mRNA but not GLYT2 mRNA, the other four expressing neither mRNAs. On the other hand, RAR-cells expressed neither GAD67 mRNA nor GLYT2 mRNA. The results suggest that the inhibitory pump cells are basically GABAergic and some of them may corelease GABA and glycine, and that RAR-cells are neither GABAergic nor glycinergic. These findings expand our understanding of the networks of lung receptor-mediated reflexes including the Hering-Breuer reflex.