Riluzole disrupts autoresuscitation from hypothermic respiratory arrest in neonatal hamsters but not rats

Abstract

We examined the effect of riluzole on expression of the central respiratory rhythm and the ability of neonates to autoresuscitate from hypothermic respiratory arrest using in vitro brainstem-spinal cord preparations of rats and hamsters. At a constant temperature of 27 °C, riluzole (5–200 μM) decreased the burst amplitude of respiratory-related motor discharge, but had little effect on the fictive respiratory frequency in rat preparations. In contrast, in hamster preparations, riluzole reduced fictive respiratory frequency, but had little effect on burst amplitude. Hamster preparations were more cold-tolerant than rat preparations, with respiratory arrest and autoresuscitation occurring at lower temperatures during cooling of the preparation. This difference was removed by incubation with riluzole (5 μM); riluzole significantly increased the temperature at which fictive respiration arrested and restarted in hamster preparations, but had no effect in rat preparations. The species differences observed in this study may reflect fundamental differences in the relative role of riluzole-sensitive mechanisms in the expression of the respiratory rhythm in early development of an altricial vs. a more precocial species.

Introduction

In mammals, as body temperature falls during progressive hypothermia, the frequency of breathing decreases until respiratory arrest occurs (Adolph, 1951). Evidence suggests that hypothermia-induced respiratory arrest occurs as a result of failure of the central respiratory rhythm generator (Mellen et al., 2002), however, the mechanisms by which this occurs are currently unknown. If intact neonates, or reduced preparations from neonates that have undergone hypothermic respiratory arrest are re-warmed, breathing restarts spontaneously (Adolph, 1951, Mellen et al., 2002, Tattersall and Milsom, 2003, Zimmer and Milsom, 2004). In contrast to neonates, most adult mammals undergo hypothermic respiratory arrest at much warmer temperatures and cannot restart breathing spontaneously upon re-warming (Adolph, 1951). However, in mammals that are capable of hibernation, both cold tolerance and the ability to autoresuscitate are retained into adulthood (Adolph, 1951, Milsom et al., 2001). The bases of these developmental differences between species are currently unknown.

Several recent studies have shown, that the drug, riluzole (RIL) eliminates hypoxia evoked gasping in rats in situ and in rats and mice in vivo (Paton et al., 2006, Pena and Aguileta, 2007, St-John et al., 2007). In addition, autoresuscitation from hypoxic/anoxic respiratory arrest involves gasping that appears to depend on RIL-sensitive mechanisms (Paton et al., 2006, Pena and Aguileta, 2007). Although present data suggest that autoresuscitation from hypothermic respiratory arrest is not initiated by gasping but by prolonged, large, slow breaths (Tattersall and Milsom, 2003, Zimmer and Milsom, 2004), we wondered whether the failure to restart breathing following hypothermic respiratory arrest in adult mammals might be due to a reduction in the ability of RIL-sensitive mechanisms to promote activity in the network at low temperature. Moreover, we wondered whether this might reflect a developmental shift in the mechanisms underlying respiratory rhythm generation in these species.

Riluzole has been used extensively as a tool to investigate the role of the persistent sodium current (I_NaP) in respiratory rhythm generation (Del Negro et al., 2001, Del Negro et al., 2002, Del Negro et al., 2005, Pace et al., 2007, Pena et al., 2004) as it is a potent inhibitor of voltage-dependent I_NaP. RIL at low concentrations (less than 5 μM) acts primarily on I_NaP (Del Negro et al., 2005, Ptak et al., 2005), however, even at this low concentration RIL is not pharmacologically selective and will also affect I_NaT (Ptak et al., 2005), although the effect on I_NaT will unlikely affect respiratory activity. At higher concentrations, RIL has a host of other non-specific effects, including effects on all tetrodotoxin (TTX)-sensitive sodium currents, Ca²⁺ currents and glutamatergic transmission (Doble, 1996, Duprat et al., 2000, Fumagalli et al., 2008, Lamanauskas and Nistri, 2008). Furthermore, depending on the route of administration, RIL will potentially have effects on cells at all levels of the reflex arc, including rhythm generating neurons, upstream regulatory cells (such as the midline raphe; Pace et al., 2007), premotor neurons and motoneurons.

With this in mind, we began addressing the species differences in the ability to autoresuscitate from hypothermic respiratory arrest reported in the literature (Adolph, 1951). We examined neonatal hamsters in addition to neonatal rats as these two species exhibit different developmental profiles. Hamsters are born after only 16 days of gestation while rats are born after 21 days gestation. Furthermore, hamsters, in contrast to rats, are tolerant of cold temperatures and are capable of hibernation (Adolph, 1951). Using these two species allowed us to compare the relative role of RIL-sensitive mechanisms in the respiratory network during early development of an altricial (hamster) vs. a precocial (rat) species, and the role of RIL-sensitive mechanisms in the expression of respiratory rhythm in a cold tolerant (hamster) vs. a non-cold tolerant (rat) species. For these studies we used brainstem-spinal cord preparations from neonates (P0–P4), to test the hypothesis that RIL-sensitive mechanisms are involved in autoresuscitation from hypothermic-respiratory arrest in both species upon re-warming from low temperatures at this stage of development.