Search

Neonatal breathing in mammals involves multiple neuronal circuits, but its genetic basis remains unclear. Mice deficient for the zinc finger protein Teashirt 3 (TSHZ3) fail to breathe and die at birth. Tshz3 is expressed in multiple areas of the brainstem involved in respiration, including the pre-Bötzinger complex (preBötC), the embryonic parafacial respiratory group (e-pF), and cranial motoneurons that control the upper airways. Tshz3 inactivation led to pronounced cell death of motoneurons in the nucleus ambiguus and induced strong alterations of rhythmogenesis in the e-pF oscillator. In contrast, the preBötC oscillator appeared to be unaffected. These deficits result in impaired upper airway function, abnormal central respiratory rhythm generation, and altered responses to pH changes. Thus, a single gene, Tshz3, controls the development of diverse components of the circuitry required for breathing.

In newborn mice of the control [C3H/HeJ (C3H)] and monoamine oxidase A-deficient (Tg8) strains, in which levels of endogenous serotonin (5-HT) were drastically increased, we investigated how 5-HT system dysregulation affected the maturation of phrenic motoneurons (PhMns), which innervate the diaphragm. First, using immunocytochemistry and confocal microscopy, we observed a 5-HT(2A) receptor (5-HT(2A)-R) expression in PhMns of both C3H and Tg8 neonates at the somatic and dendritic levels, whereas 5-HT(1B) receptor (5-HT(1B)-R) expression was observed only in Tg8 PhMns at the somatic level. We investigated the interactions between 5-HT(2A)-R and 5-HT(1B)-R during maturation by treating pregnant C3H mice with a 5-HT(2A)-R agonist (2,5-dimethoxy-4-iodoamphetamine hydrochloride). This pharmacological overactivation of 5-HT(2A)-R induced a somatic expression of 5-HT(1B)-R in PhMns of their progeny. Conversely, treatment of pregnant Tg8 mice with a 5-HT(2A)-R antagonist (ketanserin) decreased the 5-HT(1B)-R density in PhMns of their progeny. Second, using retrograde transneuronal tracing with rabies virus injected into the diaphragm of Tg8 and C3H neonates, we studied the organization of the premotor network driving PhMns. The interneuronal network monosynaptically connected to PhMns was much more extensive in Tg8 than in C3H neonates. However, treatment of pregnant C3H mice with 2,5-dimethoxy-4-iodoamphetamine hydrochloride switched the premotoneuronal network of their progeny from a C3H- to a Tg8-like pattern. These results show that a prenatal 5-HT excess affects, via the overactivation of 5-HT(2A)-R, the expression of 5-HT(1B)-R in PhMns and the organization of their premotor network.

A retrograde and transneuronal infection with rabies virus was performed in mouse neonates to locate the central nervous structures involved in the motor command of the spinal respiratory motoneurons and to discriminate the location and hierarchical organization of the neurons in and between these infected central nervous structures.

The hyperpolarization-activated cationic current (I(h)) was characterized and its maturation studied on phrenic motoneurons (PMNs), from reduced preparations of foetal (E18 and E21) and newborn (P0-P3) rats, using the whole-cell patch-clamp technique. In voltage-clamp mode, 2-s hyperpolarizing steps (5-mV, -50 to -110 mV) elicited a noninactivating inward current, blocked by external application of Cs+ or ZD 7288. At -110 mV, Ih current density averaged 0.67 +/- 0.41 pA/pF at E18, reached a transient peak at E21 (1.38 +/- 0.11 pA/pF) and decreased at P0-P3 (0.77 +/- 0.22 pA/pF). V1/2 was similar at E18 and E21 (-79 mV) but was significantly hyperpolarized at P0-P3 (-90 mV). The time constant of activation was voltage-dependent, and significantly faster at E21. Reversal potential was similar at all ages when estimated by extrapolation or tail current procedures. It was positively shifted by 25 +/- 6 mV when external potassium was raised from 3 to 10 m M, suggesting a similar sensitivity to K+ from E18 to P0-3. Cs(+) or ZD 7288 applications on PMNs at rest in current-clamp mode, in a partitioned chamber, induced a 10 +/- 2 mV hyperpolarization at E18 and E21, and an 8 +/- 2 mV hyperpolarization at P0-3. The area of the central respiratory drive potential or current was increased by 33 and 31%, respectively, at E21, but was not significantly modified at E18 and P0-3. Our data suggest a critical period during the perinatal maturation of Ih during which it is transiently upregulated and attenuates the influence of the central respiratory drive on PMNs just prior to birth.

In brainstem-spinal cord preparations of neonatal control C3H and transgenic Tg8 mice where deletion of the gene encoding monoamine oxidase-A results in serotonin (5-hydroxytryptamine (HT)) excess, whole cell recordings of identified phrenic motoneurons (Phr Mns) were performed to study the modulation of their activity by 5-HT. In C3H mice, a dual effect was observed: (i) a facilitation via 5-HT(2A) receptors and (ii) a decrease of the transmission of the central inspiratory drive via 5-HT(1B) receptors. In Tg8 mice, the 5-HT(2A)-mediated facilitation was present but the 5-HT(1B)-mediated decrease was lacking. Therefore, the conservation of the 5-HT(2A) response vs. the loss of the 5-HT(1B) one suggest that the two types of receptors respond differently to 5-HT level changes.

In rodent neonates, the neurotransmitter serotonin (5-HT) modulates the activity of both the medullary respiratory rhythm generator and the cervical phrenic motoneurons. To determine whether 5-HT also contributes to the maturation of the respiratory network, experiments were conducted in vitro on the brainstem-spinal cord preparation of neonatal mice originating from the control strain (C3H) and the monoamine oxidase A-deficient strain, which has a brain perinatal 5-HT excess (Tg8). At birth, the Tg8 respiratory network is unable to generate a respiratory pattern as stable as that produced by the C3H network, and the modulation by 5-HT of the network activity present in C3H neonates is lacking in Tg8 neonates. In addition, the morphology of the phrenic motoneurons is altered in Tg8 neonates; the motoneuron dendritic tree loses the C3H bipolar aspect but exhibits an increased number of spines and varicosities. These abnormalities were prevented in Tg8 neonates by treating pregnant Tg8 dams with the 5-HT synthesis inhibitor p-chlorophenylalanine or a 5-HT(2A) receptor antagonist but were induced in wild-type neonates by treating C3H dams with a 5-HT(2A) receptor agonist. We conclude that 5-HT contributes, probably via 5-HT(2A) receptors, to the normal maturation of the respiratory network but alters it when present in excess. Disorders affecting 5-HT metabolism during gestation may therefore have deleterious effects on newborns.

In vivo and in vitro experiments were performed in control (C3H) and monoamine oxidase A (MAOA)-deficient (Tg8) neonatal mice to determine whether MAOA deficiency affected spinal locomotor network maturation. Comparing the swimming behaviors at birth in C3H mice with those in Tg8 mice revealed a delayed role for the hindlimbs in Tg8 swimming, even though adult swimming behavior was acquired at postnatal day 14 (P14) in both strains. Analyzing the locomotor network activity in vitro showed that serotonin (5-HT) induced and modulated locomotor-like discharges in hindlimb ventral roots of C3H but not Tg8 neonates. The Tg8 network began, however, to be affected by 5-HT at P11. Thus both in vivo and in vitro results argue for a transient delay of locomotor network maturation in the Tg8 strain.

Experiments were carried out on the in vitro brainstem-spinal cord preparation of the newborn rat to analyse the effects of substance P (SP) on phrenic motoneuron (PMN) activity. In current-clamp mode, SP significantly depolarized PMNs, increased their input resistance, decreased the rheobase current and shifted the firing frequency-intensity relationships leftwards, but did not affect spike frequency adaptation or single spike configuration. The neurokinin receptor agonist NK1 had SP-mimetic effects, whereas the NK3 and NK2 receptor agonists were less effective and ineffective, respectively. In a tetrodotoxin-containing aCSF, only SP or the NK1 receptor agonist were still active. No depolarization was observed when the NK1 receptor agonist was applied in the presence of muscarine. In voltage-clamp mode, SP or the NK1 receptor agonist produced an inward current (ISP) which was not significantly reduced by extracellular application of tetraethylammonium, Co2+, 4-aminopyridine or Cs+. In aCSF containing tetrodotoxin, Co2+ and Cs+, ISP was blocked by muscarine. No PMN displayed any M-type potassium current but only a current showing no voltage sensitivity over the range -100 to 0 mV, reversing near the expected EK +, hence consistent with a leak current. SP application to the spinal cord only (using a partitioned chamber) significantly increased the phrenic activity. Pretreatment with the NMDA receptor antagonist DL-2-amino-5-phosphonovaleric acid (AP5) decreased the C4 discharge duration and blocked the effect of SP, thus exhibiting an NMDA potentiation by SP. In conclusion, SP modulates postsynaptically the response of phrenic motoneurons to the inspiratory drive through the reduction of a leak conductance and the potentiation of the NMDA component of the synaptic input.

In vitro experiments were conducted on neonatal rat brainstem-spinal cord preparations to test the hypothesis of an inhibitory modulation of phrenic activity by serotonin (5-HT) via non-5-HT2A receptors [Lindsay, A.D. and Feldman, J.L., Modulation of respiratory activity of neonatal rat phrenic motoneurones by serotonin, J. Physiol., 461 (1993) 213-233]. The changes induced by 5-HT and related agents on phrenic root discharges and membrane currents in identified phrenic motoneurons were analysed after blockade of spinal 5-HT2A receptors. Spinal application of 5-HT1B (but not 5-HT1A) receptor agonists depressed the phrenic activity and the effect was prevented by pretreatment with 5-HT1B (but not 5-HT1A, 5-HT2A and 5-HT3) receptor antagonists. Results from phrenic motoneuron whole cell recordings do not reject a presynaptic location of the 5-HT receptors responsible for this depression.

Experiments were performed in vitro on fetal and newborn rat brainstem-spinal cord preparations to analyse the perinatal developmental changes in inspiratory motor output. The amplitude of the inspiratory bursts of the whole C4 ventral root (global extracellular recording), the firing patterns of 80 medullary inspiratory neurons (unitary extracellular recording) and the firing and membrane properties of 71 respiratory neurons in the C4 ventral horn (whole-cell recording) were analysed at embryonic day 18 (E18), 21 (E21) and post natal days 0 to 3 (P0-3). At E18, the amplitude of the C4 bursts was weak and variable from one respiratory cycle to the next, as well as the discharge pattern of most of the medullary inspiratory neurons. C4 motoneurons were immature, very excitable and displaying variable inspiratory discharges, but already able to deliver sustained bursts of potentials when depolarised. At E21 and P0-3, the amplitude of the C4 bursts was increased and stable, most of the medullary inspiratory neurons already were able to generate a stable firing pattern and C4 motoneurons showed maturational changes in terms of the resting potential, spike amplitude and input membrane resistance. This work suggests that the short period extending from E18 to E21 is a critical maturational period for the medullary respiratory network which becomes able to elaborate a stable respiratory motor output.

Previous experiments performed in the in vitro newborn rat brainstem-spinal cord preparation reported that the addition of serotonin (5-HT, 30-50 microM) to the bathing medium induced increases in the respiratory frequency and a large tonic discharge on all the cervical ventral roots. The aim of the present work was to define whether the 5-HT-induced tonic discharge involved respiratory or non-respiratory motoneurones. Intracellular recordings demonstrated that cervical (C2) motoneurones (n = 27) were depolarized by 5-HT but that the 5-HT-induced tonic discharge was mainly due to recruitment of silent motoneurones (n = 18) which fired permanently (15/18; 17 +/- 3 Hz) under 5-HT. The respiratory motoneurones (n = 9) retained a phasic inspiratory discharge (5/9), even if some (4/9) occasionally exhibited a few spikes during expiration. Therefore, it is concluded that the 5-HT-induced tonic discharge is unlikely to have functional significance in respiration.

Newborn rat respiratory activity was recorded on hypoglossal nerve and ventral cervical roots during in vitro experiments performed on superfused brainstem spinal cord preparations. The addition of serotonin (5-HT) to the bathing medium increased the respiratory frequency and selectively depressed the hypoglossal activity. Any decreases in the amplitude of cervical recordings were always restricted and reversible, whereas the hypoglossal activity was abolished. Furthermore, on cervical roots, 5-HT induced a tonic activity superimposed on the respiratory one, which was never observed with the hypoglossal nerve. When 5-HT was applied on isolated hemispinal cord, a tonic activity could still be elicited. These results indicate that serotonin (i) modulates the activity of neurons involved in the generation of respiratory rhythm, (ii) depresses the activity of hypoglossal motoneurons, and (iii) evokes tonic activity in cervical motoneurons, probably as the result of direct spinal effects.

The role of central respiratory drive, muscle spindles and tendon organs in producing respiratory movements has been studied in the cat by recording from motoneurones supplying a set of small axial muscles inserted between each rib and the vertebra immediately rostral, the levatores costae. The levator costae muscles are active during normal inspiration but activity is progressively stronger in the muscles located in more caudal thoracic segments. Intracellular recordings from levator costae motoneurons show a characteristic central respiratory drive potential (c.r.d.p.), comprising phases of depolarization during inspiration alternating with expiratory phased hyperpolarization due to post-synaptic inhibition. Loading or unloading the levator costae muscles increases and decreases, respectively, their normal inspiratory activity. Electrophysiological and histological analysis of levator costae afferents reveals that each muscle contains three to five spindles and two to three tendon organs. By dissecting the levator costae nerve, afferent discharges from muscle spindle primary and secondary endings and those from tendon organs were recorded 'in continuity' and used to trigger an averager for analysing the synaptic potentials they evoke in levator costae motoneurones. Monosynaptic excitation and oligosynaptic inhibition originate from spindle (primary and secondary endings) and from tendon organ afferents respectively. Peculiarly a monosynaptic excitation has been evoked from a tendon organ. The action of the levator costae muscle on the respiratory movements of the ribs in the different thoracic spaces was analysed in relation to their anatomical properties and to the activity of the other respiratory muscles. The levator costae muscles, because of their relatively small size and well-defined population of proprioceptors, appear to provide a favourable preparation for studying the integrative action of the motoneurone.

Central respiratory drive and recruitment order of phrenic motoneurones and inspiratory laryngeal motoneurones were studied in anaesthetized, paralyzed and artificially ventilated cats. Unitary activities of pairs of motoneurones originating from the same population were recorded on thin filaments. Differences of recruitment during inspiration enabled distinction to be made between early (E) and late (L) recruited motoneurones. 'On-line' cross-correlation analyses were performed on either homogeneous pairs (2E or 2L) or heterogeneous pairs (1E and 1L). In 14/30 pairs of phrenic motoneurones and 26/32 pairs of inspiratory laryngeal motoneurones cross-correlation analysis revealed a bell-shaped increase of probability of firing whose characteristics indicate synchronization by central shared excitatory inputs originating from central respiratory drive. For phrenic motoneurones synchronization appeared mainly in homogeneous pairs (2E:7/11; 2L:5/9) whereas motoneurones of heterogeneous pairs rarely shared the same central respiratory drive (2/10). For inspiratory laryngeal motoneurones, cases of synchronization were equally obtained in homogeneous (2E:8/10; 2L:14/16) and heterogeneous (4/6) pairs. These results suggest that differences of recruitment order during inspiration are related: (i) for phrenic motoneurones, partly to differences of excitability but mainly to a dual central respiratory drive which is assumed to divide the population into two components; (ii) for inspiratory laryngeal motoneurones to differences of excitability since all the motoneurones share the same drive.

10 The activity of 107 medullary inspiratory neurones has been recorded extracellularly in anesthetized cats (urethane-chloralose). According to their localization in the medulla and to their axonal pathways (tested by antidromic activation), these neurones were classified as: bulbo-spinal neurones (NBSI) which send their axons to the spinal cord; they are located in the dorsal or the ventral respiratory nucleus; propriobulbar neurons (NPBI) whose axons are probably entirely located within the medulla; they are found only in the ventral nucleus. 20 Summation of gross discharges from phrenic (C5 ventral root) or inspiratory intercostal (from T4 to T10) nerves was triggered by spikes from medullary respiratory neurones. If the studied neurone controls the activity of the recorded motor fibres, after sommation of a great number of sweeps (3,000 to 5,000), the summed nerve activity displays a short delay excitatory wave, indicating that the activity of the motoneurones is increased after the onset of the medullary potential. On the other hand, there is a flat summed nerve activity if the nerve discharge is not directly dependent on the medullary recorded activity. 30 The dorsal nucleus NBSIs send their axon to the contralateral phrenic motoneurones (as it is revealed by the excitatory wave noted on the summed phrenic nerve activity, fig. 1 and 4) and not to the intercostal motoneurones (fig. 4). 40 The ventral nucleus NBSIs control contralateral intercostal motoneurones (fig. 5). Some of them send their axons only to these motoneurones, but the majority also act on phrenic motoneurones (fig. 2 and 5). 50 Connections between NBSIs and phrenic motoneurones are monosynaptic and probably also those between NBSIs and intercostal motoneurones. 60 No excitatory wave has been observed on phrenic or intercostal summed activity when summation was triggered by NPBI spikes (fig. 3). This confirms the accuracy of the "antidromic stimulation test" used to classify medullary respiratory neurons.

1. Intracellular recordings were made from 220 Phrenic Motoneurons (PM) in anaesthetized, spontaneously breathing cats, deafferented from C3 to C7, in order to look for somatic events related to the Recurrent Responses (RR) evoked in PM axons by repetitive stimulation of the phrenic nerve. RR appear sporadically at a constant latency, originate from a spinal nicotinic mechanism and can be evoked in a PM without the presence of an antidromic volley in its axon (Khatib et al. 1986). 2. Using stimuli effective for eliciting RR in axons, we failed to observe intracellularly somatic events corresponding to RR after the occurrence of an antidromic action potential. RR were observed extracellularly in two cases, but in both cases the recording originated from axons. 3. We attempted to elicit somatic RR without a preceding antidromic action potential, using either parathreshold stimulation of the impaled PM, or suprathreshold stimulation of a phrenic strand which excluded the axon of the impaled PM. In both cases, RR-like events, with very stable latencies, appeared sporadically in 4/142 and 2/15 PMs respectively. 4. Parathreshold stimuli or stimulation of a strand were coupled with averaging of the synaptic noise in order to look for small events temporally related to the stimuli. Short latency small depolarizations, looking-like recurrent EPSPs, were revealed in 22/142 and 5/15 PMs respectively. 5. These results confirm the existence of interrelations between PMs, providing for re-excitation and coupling within the phrenic pool, in addition to centrally imposed synchronization.

Interactions between phrenic motoneurons have been analysed in anaesthetized, paralyzed cats after C3 to C7 deafferentation. Effects of electrical stimulation of the C5 phrenic axons have been studied on thin filaments dissected from the stimulated nerve. Repetitive stimulation could elicit, after the primary direct response of the stimulated axons, a secondary response named Recurrent Response, RR. RRs have been obtained in 117/186 phrenic axons. They appear sporadically (mean occurrence: 3.75 RRs elicited by 100 shocks of stimulation) at a constant latency. They originate from a spinal mechanism since they persist after C2 transection and disappear after section of the ventral roots. The mechanism responsible for RR shows spatial and temporal facilitation. The RR probability increases with the number of antidromically invaded motoneurons as revealed by changes either of stimulation intensity or of central respiratory drive. However, RR could be evoked in a motoneuron without an antidromic volley in its axon. Systemic injections of nicotinic blocking drugs such as dihydro-beta-erythroidine or mecamylamine decrease or suppress the occurrence of RR; therefore, cholinergic synapses are involved in the RR generating process. RR are assumed to be due to direct excitatory interactions between homonymous motoneurons. Recurrent axon collaterals impinging directly on neighbouring motoneurons would link together the different motoneurons of the phrenic pool. The functional significance of this phenomenon is discussed.

In anesthetized spontaneously breathing cats (C4-C5 deafferentation), recurrent inhibition of phrenic motoneurons was analyzed by studying either recurrent IPSPs in phrenic motoneurons, or Renshaw cell discharges evoked by C5 phrenic nerve stimulation. Of 90 intracellularly recorded phrenic motoneurons, 7 motoneurons showed evoked recurrent IPSPs with stimulation of C5 phrenic axons subthreshold for eliciting antidromic activation of the motoneuron from which intracellular recording was done. These IPSPs could be reversed by imposed hyperpolarization of the motoneuron, and were of greater amplitude during inspiration than during expiration. Within the phrenic nucleus, interneurons were classified as Renshaw cells if they responded to C5 phrenic axon stimulation with a typical high frequency burst of potentials. Reactivity of these Renshaw cells was related to the respiratory cycle, number of spikes in the burst being greater during inspiration than during expiration. Injection of a nicotinic cholinergic blocker (mecamylamine) decreased responses of Renshaw cells but the respiratory modulation was still present. Some Renshaw cells (18/33) were spontaneously active during inspiration. Their activity was generally maximal during the last third of inspiration. Since: spontaneous activity of Renshaw cells is related to the respiratory drive; persists after C7 spinal transection and after mecamylamine poisoning of the axonal recurrent pathway; and might appear before sustained phrenic activity, the assumption of a central respiratory drive impinging on the Renshaw cells has to be retained.