Restoration of the Central Chemoreflex by a Progesterone Receptor Agonist After Lesions of the Retrotrapezoid Nucleus

In a recent issue of Acta Physiologica Cardani et al. 1 report that chronic administration of the progesterone receptor agonist etonogestrel (ETO) reverses the chemoreflex deficit caused by lesions of the retrotrapezoid nucleus (RTN, a. k. a. parafacial nucleus) in female rats. This result is notable because desogestrel, a contraceptive precursor of ETO, has been found to restore the hypercapnic ventilatory reflex in a subset of female patients suffering from congenital central hypoventilation syndrome (CCHS), a developmental disease whose adverse consequences on breathing likely result from RTN dysplasia 2, 3. The RTN (Figure 1A) consists of approximately 650 glutamatergic neurons in mice, 5000 in humans, located ventrally in the rostral medulla oblongata 4. RTN neurons are both pH sensors and the keystone of the central chemoreflex which mediates arterial CO2 homeostasis by matching lung ventilation to the metabolic production of CO2 4. In adult rodents, RTN lesions decrease or virtually eliminate the central chemoreflex depending on lesion size and cause severe hypoventilation (elevated arterial PCO2 and lowered arterial PO2 without reduction of metabolic activity) 9. CCHS, whose cardinal signs include hypoventilation, a greatly reduced central chemoreflex and life-threatening apneas during slow-wave (non-REM) sleep, is attributed predominantly to mutations of the transcription factor PHOX2B 5. RTN neuron development in rodents is highly vulnerable to Phox2b mutations 4, 10; it is presumed that RTN lesions also occur in CCHS and are at least partly responsible for the loss of the chemoreflex. Cardani et al. 1 used a rat model of CCHS in which RTN is damaged bilaterally with a toxin, causing a large but incomplete reduction of the chemoreflex. The ability of ETO administration (chronic but not acute) to restore the chemoreflex is attributed to the enhancement of the response of the surviving RTN neurons to CO2. The main supportive evidence is that ETO increases the mRNA level of two acid-sensitive membrane proteins, Task-2 and GPR4 (Figure 1B), whose expression by RTN neurons is required for their activation by CO2 4, 7, 8. The effect of ETO on Task-2 and GPR4 transcripts was detected only in female rats, consistent with the sex-specific effect of this drug on the chemoreflex. In sum, the results of the Cardani study are formally consistent with a sex-specific genomic effect of ETO leading to an improvement of the chemoreflex via restoration of the response of RTN neurons to CO2. This novel theory will of course require further elaboration. Evidence that the response of single RTN neurons to hypercapnia is enhanced by ETO would be highly desirable. ETO treatment could also potentiate the downstream effects of RTN on the breathing network. In addition, the evidence does not prove that ETO acts directly on RTN neurons rather than on their inputs 4. In fact, the authors also uncovered effects of ETO on gene expression within the nucleus of the solitary tract, a region of the brain that regulates breathing partly via inputs to the RTN. Finally, the existence in the literature of seemingly discrepant results relative to how ETO activates breathing needs to be mentioned. Studies on neonatal mouse brain preparations “in vitro” describe a stimulatory effect of acute ETO administration on fictive breathing (the phrenic nerve discharge) at rest or under acidified bath conditions 11 but this effect is not sex-dependent, unlike the effect uncovered by Cardani et al. Besides, the latter authors report that the chemoreflex of rats subjected to RTN lesions is not improved by acute ETO administration. Technical considerations may help explain these seemingly contradictory results. The chemoreflex of unanesthetized rodents is a robust response to respiratory acidosis consisting of an increase in the frequency and amplitude of the phrenic nerve discharge (PND); in intact animals, this response is mediated almost entirely via the RTN 9. In contrast, the response of the PND to metabolic acidosis in neonate brain preparations in vitro is limited to a modest frequency change that is attributed to effects of acid in multiple brain regions including the respiratory rhythm pattern generator. In other words, the mechanisms by which metabolic acidosis activates the breathing network of the neonatal brain in vitro seem different from those that mediate the effect of hypercapnia in intact and mature rodents. The fact that a progesterone receptor agonist might influence each response by different mechanisms is therefore not unexpected. The degree to which RTN is impaired by Phox2b mutations (percent neuronal loss, degree of cellular dysfunction) depends on the type of mutation and is not clearly established in CCHS nor even in animal models. Genetic studies in mice suggest that the loss of RTN neurons caused by Phox2b mutations is less than 100% 6. The presence of a residual contingent of healthy RTN neurons could underlie the partial recovery of the chemoreflex observed during the first few weeks after birth in mice models of CCHS 6. RTN lesion experiments have shown that the intensity of the central chemoreflex is an inverse exponential function of the number of surviving RTN neurons 9; this type of relationship indicates that the survival of a small percentage of intact RTN neurons is enough to maintain a large proportion of the chemoreflex. The present observations raise the interesting possibility that RTN function could be rescued by leveraging the plasticity of this nucleus. The molecular mechanisms that underlie RTN neuron early development and plasticity deserve to be thoroughly investigated. Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.