Showing papers by "Richard J. A. Wilson published in 2014"

Lamprey breathing when feeding sucks: the respiratory rhythm generator of a parasitic fish

Abstract: Two topics of contention dominate the study of vertebrate respiratory rhythm generation: Are respiratory rhythm generators composed of discrete oscillators or is the circuit distributed? And, how important are intrinsic properties of rhythm generating neurons vs. their synaptic interconnections? In this issue of The Journal of Physiology, Cinelli et al. (2014) report on their ongoing investigation of these issues in the lamprey, the primitive, parasitic jawless fish that patrols near the base of the vertebrate phylogenic tree. The low metabolic rate and hypoxic tolerance of this animal has made its isolated CNS a classic system for studying the neuronal control of locomotion; with renewed interest in how this animal ventilates, it will also likely help unlock the mysteries of respiratory rhythm generation. Unlike most fish, which use a buccal pump to gill ventilate, adult lampreys use muscles associated with their gill slits. Constriction of these muscles causes water expulsion; water inhalation is by passive recoil. This arrangement frees their buccal musculature for parasitic feeding; they can suck, feed and breathe all at the same time (Rovainen, 1996). The rhythmic contractions associated with ventilation are controlled by rhythmic bursts of branchial motor neurons (VIIm–IXm–Xm) that persist with high fidelity when the brainstem is isolated (Kawasaki, 1979). Two brainstem sites have been associated with rhythm generation, an area around the trigeminal motor nucleus (i.e. the paratrigeminal respiratory group; pTRG) and the vagal motor neuron region (VMR). In the study by Cinelli et al. (2014), the authors demonstrate that under ionotropic glutamate receptor antagonists the respiratory rhythm shuts down but can be successfully rescued by bath application of GABA antagonists (application of glycine antagonists had little or no effect). Moreover, the authors were able to replicate this finding by microinjecting GABA antagonists into pTRG. These and other data in the paper (a) substantiate the pTRG as the primary rhythm generating area for gill ventilation in the lamprey brainstem, and (b) show that phasic inhibition within the pTRG is not necessary for bursting. In our opinion the results from this study are compelling, with one small caveat: use of saline more akin to the CSF of a mammal than an aquatic species. Specifically, bicarbonate concentration and CO2 were 20 mm and ∼30 Torr too high, respectively, and pH was ∼0.4 pH units too acidic (Maren & Vogh, 1980). We note that (a) GABA receptors are Cl− and bicarbonate permeable, (b) Cl− and bicarbonate share a membrane transporter, (c) channels, receptors and intracellular signalling pathways are often sensitive to pH (Hubner & Holthoff, 2013), and (d) nervous systems can sometimes produce very similar outputs through very different means. This concern is offset by the historic use of this saline for studying spinal circuits in lamprey and the need for parsimony: varying bicarbonate and Cl− concentrations causes only subtle differences to respiratory motor output (Rovainen, 1983). Notwithstanding this caveat, the demonstration that a discrete pontine oscillator produces gill ventilation in lamprey is remarkable. Firstly, confirmation that the pTRG is necessary and sufficient for rhythm generation allows us to add lamprey to a growing list of vertebrates in which discrete respiratory rhythm generating sites (oscillators) have been identified, including frogs, chicks, rats, mice and goats. Secondly, this architecture is very different to that proposed for gill rhythm generator in other fish (a discrete oscillator for gill ventilation has yet to be located in any other kind of fish, with the most recent data from elasmobranch and teleosts suggesting rhythm generation is distributed along the length of the brainstem; Duchcherer et al. 2010; Taylor et al. 2010). Thirdly, the pTRG is the first confirmed vertebrate respiratory oscillator located in the pons; in all other vertebrates, respiratory oscillators appear to be confined to the medulla (evidence for a pontine oscillator in rat and carp is anecdotal; see Wilson et al. 2006). With regard to rhythmogenic mechanism, Cinelli et al.'s data show that rhythm generation is abolished by ionotropic glutamate receptor blockade. Glutamate receptor blockade also abolishes lung-inflation rhythms produced by the isolated brainstems of turtles and frogs, and the mammalian preBotzinger complex (preBotC). Imaging studies suggest that glutamatergic synapses in the preBotC are necessary for forming the ‘group pacemaker’, amplifying and coordinating otherwise stochastic, incongruent and limited excitatory activity in individual neurons to produce population bursting (Mironov, 2008). In rescuing the pTRG rhythm with GABA antagonists, Cinelli et al. demonstrated that GABA-mediated inhibition only plays a modulatory role in rhythm generation, which in itself is also consistent with a ‘group pacemaker’ mechanism. However, the rescue of rhythm in the presence of glutamate receptor antagonists demonstrates rhythm is not dependent on ionotropic glutamate receptor blockade after all: some other coupling between neurons is sufficient to maintain cohesive output! This coupling, which deserves further investigation, could be via electrical synapses, or it could be via phasic, non-GABA, non-glutamatergic excitatory (or inhibitory) synapses. Finally, we note that the rhythmogenic mechanisms underlying gill ventilation in lamprey and tadpoles are distinct. Unlike the pTRG, the gill oscillator in tadpoles continues in the presence of ionotropic glutamate receptor blockade and is dependent on post-synaptic chloride-mediated inhibition. Along with the difference in location (trigeminal in lamprey, vagal in tadpoles), these functional data suggest gill oscillators in amphibians and lamprey are not homologous. This begs the questions as to the evolutionary relationship (if any) between the pontine pTRG in lamprey and the medullary respiratory oscillators in other species. Breathing when feeding sucks, may be just too difficult without the pons!

Last Word from Richard J. A.

Abstract: VE is additive, hypo-additive interactions are required in frequency or VT .T hus, choice of respiratory variable analysed may be critical (see Younes’ comments). We also advanced a hybrid system, whereby interaction is multifactorial and subject to context-dependent change. This was addressed by several commentaries and echoed by Duffin and Mateika’s discussion of how long-term facilitation transforms an additive to hyper-additive interaction. We illustrated a hybrid system with a model where interaction is hypo-additive below eupnoea and hyper-additive above. In their Rebuttal, Teppema and Smith challenged this scheme citing Nakayama et al (2002). Bain and Ainslie’s comments echoed this sentiment but we respectfully disagree. NotwithstandingthatNakayamaetal.made no attempt to separate central and peripheral chemoreceptors, their data support a redundant and probably hypo-additive system below eupnoea: they show that carotid body stimulant almitrine decreases apnoea likelihood despite causing systemic hypocapnia which reduces central CO2 chemoreceptor activation. In Nakayama et al., systemic hypoxia increased apnoea