Sodium Is Detected by the OVLT to Regulate Sympathetic Tone

Abstract

Hypernatremia is known to elicit a rise in sympathetic tone and blood pressure. In this issue of Neuron, Nomura et al. (2018) now show that this is mediated via the organum vasculosum laminae terminalis (OVLT). Na⁺ activates OVLT neurons via a paracrine mechanism involving sodium channel Na_x expressed by astrocytes and the ependyma.

In normovolemic animals, salt loading via drinking or other means produces hypernatremia and hyperosmolarity that increases sympathetic tone, circulating vasopressin levels, and blood pressure. These neuroendocrine responses may contribute to salt-induced hypertension in susceptible humans (Farquhar et al., 2015).

In this issue of Neuron, Nomura et al. (2018) report that the increase in sympathetic tone and blood pressure elicited by hypernatremia in mice is mediated primarily via the circumventricular organ known as the organum vasculosum laminae terminalis (OVLT). They also demonstrate that this effect of hypernatremia requires the sodium-activated sodium channel Na_x (scn7a) (Hiyama et al., 2002). The foundational observation is that acute salt loading or chronic ingestion of salt increases blood pressure and sympathetic tone in control mice but has virtually no effect in Na_x KO mice.

Using optogenetics and other methods, the authors mapped out the main pathway between the principal output neurons of the OVLT and sympathetic preganglionic neurons (Figure 1). They showed that these OVLT neurons are glutamatergic and innervate the parvocellular portion of the paraventricular nucleus of the hypothalamus (PVH). The latter neurons, which are for the most part also glutamatergic, presumably activate sympathetic preganglionic neurons either directly or by recruiting bulbospinal C1 neurons located in the rostral ventrolateral medulla (RVLM), as shown here using Fos expression as a reporter of neural activity. The existence of a polysynaptic pathway between OVLT and spinal cord with relays in the PVH and RVLM has been suggested before (Stocker et al., 2015), but the present study provides some of the most persuasive evidence to date that this pathway is activated by hypernatremia in vivo. This was done by showing that selective optogenetic activation of the OVLT neurons with PVH projections (henceforth OVLT-PVH neurons) increases blood pressure and, most importantly, that selective optogenetic inhibition of these neurons attenuates the blood pressure rise elicited by infusing sodium into the ventricular space.

Figure 1. Cellular Mechanism and Neural Pathways Contributing to the Pressor Effect of Hypernatremia in Mice.

In blue, hypothetical neuronal input that may also contribute to the activation of OVLT neurons by hypernatremia in vivo. Abbreviations: MCT, monocarboxylate transporter; PVH, paraventricular nucleus of the hypothalamus; RVLM, rostral ventrolateral medulla; SGN, sympathetic ganglionic neurons; SPGN, sympathetic preganglionic neuron.

Na_x is strongly expressed by astrocytes and by the ependymal lining of the OVLT. The way in which activation of Na_x leads to the depolarization of the principal neurons of the OVLT is one of the most intriguing findings of the Nomura paper. Until recently, sodium sensing by OVLT was assumed to be an intrinsic neuronal property (Kinsman et al., 2017b). According to Nomura et al. (2018), OVLT neurons respond to Na⁺ concentration primarily via a paracrine mechanism. The proposed steps include an increase in glucose uptake by the Na_x-expressing astrocytes and ependymal cells, subsequent production of lactic acid, and, finally, release of the latter into the extracellular space via a monocarboxylate transporter (MCT) (Figure 1). Lactate release is a well-described attribute of CNS astrocytes elsewhere, but this phenomenon (the astrocyte-neuron lactate shuttle, ANLS) typically serves to couple synaptic activity and energy delivery to neurons (Magistretti and Allaman, 2018). In the OVLT, lactic acid release seems to ultimately cause the opening of ASIC1A channels expressed by the OVLT-PVH neurons, leading to their depolarization. These conclusions are supported by convincing pharmacological evidence in tissue slices (MCT blockers) and the dramatic reduction in the response to Na produced by downregulating ASIC1A expression selectively within the OVLT. The authors also show that hypernatremia acidifies the extracellular space of the OVLT in normal but not Na_x KO mice. They postulate that this acidification is what leads to the opening of ASIC1 channels and subsequent depolarization of the OVLT-PVH neurons because acidification activates these neurons, whereas lactate application at neutral pH does not. In other words, this evidence suggests that extracellular protons exported by the glia and the ependyma along with lactate could be the ultimate signaling molecule (Figure 1).

The proposed scheme is clearly compatible with the results but perhaps incomplete. Evidence that dissociated OVLT neurons lack a depolarizing response to hypernatremia in either control or Na_x KO mice would help bolster Nomura et al.’s paracrine theory. Also, one could still imagine that the presence of ASIC1A channels is permissive, i.e., somehow necessary to maintain the excitability of the OVLT-PVH neurons, but that these channels are not mediating the effect of sodium per se. Finally, the sodium current elicited by acidification of ASIC1a channels is usually described as rapidly desensitizing, a characteristic that seems ill suited to produce the expected long-lasting effect of hypernatremia on the OVLT-PVH neurons. Elsewhere, lactate has been noted to potentiate the acid-induced ASIC1 cur-rent and even to induce a sustained component of this current. This phenomenon could perhaps be occurring in the OVLT, because the authors show that the response of OVLT neurons to acid is enhanced somewhat by lactate. Nonetheless, the electrophysiological effects of lactate in the OVLT should probably be investigated further as well as the consequences of downregulating or blocking ASIC1 receptors on the intrinsic and synaptic properties of the OVLT-PVH neurons.

The neural pathway between OVLT and sympathetic preganglionic neurons is probably more complex than what Nomura et al. (2018) envision, because hypernatremia changes sympathetic tone in a regionally specific manner. In rats or sheep, where such details can be examined, intracarotic or intracerebroventricular infusion of NaCl produces a relatively limited but dosedependent blood pressure rise just like in mice, but, while lumbar and adrenal SNA increases, cardiac and splanchnic SNA are unaffected and renal SNA decreases (Frithiof et al., 2014; Stocker et al., 2015). Reduced renal nerve activity combined with higher systemic blood pressure may enhance salt excretion by the kidney. The genesis of this pattern is not clarified by the present study.

According to Nomura et al. (2018), around 50% of OVLT-PVH neurons are still activated (i.e., c-Fos positive) following high salt ingestion in Na_x-KO mice. Yet, they report that these neurons no longer responded to hypernatremia in Na_x-KO mice in slices. The substantial residual activation of OVLT-PVH neurons observed after sodium ingestion in Na_x-KO mice is therefore presumably unrelated to local changes in [Na⁺] within the OVLT. Several alternative explanations come to mind. One possibility is that OVLT neurons respond to hyperosmolarity as well as to changes in Na⁺ concentration. This was not detected in the present study but occurs in the rat (Kinsman et al., 2017a). OVLT-PVH neurons could also receive excitatory polysynaptic inputs from peripheral or other central osmo- or sodium sensors (Figure 1), including the subfornical organ. Indeed, the response of OVLT neurons to hypernatremia in vivo seems much larger than in slices (<2 versus ~10 Hz) (Kinsman et al., 2017b). Alternately or in addition, a subset of OVLT-PVH neurons may regulate vasopressin release rather than the sympathetic outflow. In support of the latter possibility, Nomura et al. showed that the BP rise elicited by intracerebroventricular infusion of salt had a vasopressin-dependent component that persisted in Na_x-KO mice.

The OVLT and the subfornical organ have long been suspected to contain sodium sensors, but the effects produced by various classic experiments (lesions, stimulation, use of various neuronal actuators) could also be at least partly explained by postulating that these regions are essential nodal points for wider circuits that control drinking, sodium appetite, or a subset of autonomic responses. For example, thirst-promoting subfornical organ neurons respond to inputs from the oral cavity during eating and drinking (Zimmerman et al., 2016). Clearly, this circumventricular organ is not merely sensing the composition of the blood, hormonal content included, but receives numerous and under-appreciated synaptic inputs. This is most likely also the case with the OVLT.

The nodal point-versussensor debate has been ongoing for years in the field of CO₂ sensing and respiratory chemoreception. There are interesting parallels between the retrotrapezoid nucleus and the OVLT/subfornical organ. The retrotrapezoid nucleus contains a well-defined population of output neurons that convey pH-related information to the respiratory pattern generator (Guyenet et al., 2016). These neurons also receive synaptic input from multiple sources, including peripheral CO₂ sensors such as the carotid bodies (Guyenet et al., 2016). The pluricellular PCO₂/pH sensing mechanism relies both on the intrinsic pH sensitivity of retrotrapezoid nucleus neurons and on paracrine influences from the surrounding astrocytes and vasculature. The nodal point versus sensor controversy can only be settled by a more complete understanding of the molecular and cellular nature of the postulated sensors and of the synaptic connectivity of the principal neurons. The study by Nomura et al. (2018) represents an important step toward understanding how sodium is detected by the OVLT to regulate sympathetic tone. However, the integrative function performed by the principal output neurons (OVLT-PVH) could be just as important to salt-induced hypertension as the sodium sensor itself.