Fast inactivation of Nav1.3 channels by FGF14 proteins: An unconventional way to regulate the slow firing of adrenal chromaffin cells

Several groups, including Lingle’s laboratory, have been pioneers in identifying the origins of spontaneous and evoked AP firings in rat and mouse CCs (Lingle et al., 2018). In their columnary arrangement in intact adrenal glands, CCs are effectively regulated by the splanchnic nerve activity (“neurogenic control”). Release of ACh from the sympathetic nerve terminals stimulates nicotinic and muscarinic receptors on CCs, which leads to increased AP firing, opening of voltage-gated calcium channels (Cav), Ca²⁺ entry through open Cav channels, and Ca²⁺-dependent release of CAs into the bloodstream to prepare different body organs for the fight-or-flight response (García et al., 2006). CCs express a multitude of voltage-gated Nav, Cav, and Kv channels that, together with BK, SK, and TASK channels, allow for generating spontaneous “neuron-like” tonic or burst firing patterns even when CCs are isolated from the gland and kept in primary culture (Lingle et al., 2018; see Table 1 in Carbone et al., 2019). Under these conditions, CCs can drive a sustained “nonneurogenic” release of CAs upon stimulation that mimics the body’s response to various secretagogues (histamine, low pHo, hypoxia, and hyperkaliemia; Wallace et al., 2002; Inoue et al., 2008; Salman et al., 2013; Guarina et al., 2017).

Neurogenic and nonneurogenic CA release is regulated by AP firing and are therefore strictly controlled by Nav availability. Yet, the Nav channels of rat and mouse CCs possess interesting unsolved peculiarities. First, rat CCs have been reported to express Nav1.7 (SCN9A) channels (Wada et al., 2008), though the Nav1.3 isoform (SCN3A) is the predominant species expressed in mouse CCs (Vandael et al., 2015), and the Nav currents in rat and mouse CCs possess steady-state inactivation properties consistent with a single Nav component (Lou et al., 2003; Vandael et al., 2015). This suggests that Nav currents in CCs are carried by either Nav type or by a mixture of Nav1.3 and Nav1.7 channels with similar voltage-dependent activation and inactivation gating (Lou et al., 2003; Vandael et al., 2015). The presence of a single type of rapidly inactivating TTX-sensitive Nav current in rodent CCs therefore greatly simplifies the analysis of fast and slow recovery from rapid inactivation undertaken by Martinez-Espinosa et al. (2021a, 2021b) in the earlier issue of JGP. Second, although multiple channels may contribute to AP adaptation in CCs (Vandael et al., 2012; Martinez-Espinosa et al., 2014), use-dependent changes in Nav availability appears as a possible mechanism to regulate frequency, amplitude, and mode of AP firing (Vandael et al., 2015). CCs respond to constant current injection with trains of APs whose frequencies hardly exceed 10–15 Hz (Solaro et al., 1995; Vandael et al., 2012; Martinez-Espinosa et al., 2014; Carbone et al., 2019). During such sustained depolarization, CCs typically exhibit a progressive decline in AP frequency and amplitude (adaptation) due mainly to a cumulative increase of SK currents between APs (Vandael et al., 2012). Block of SK channels by apamin increases AP frequency and AP decline, suggesting that Nav availability changes drastically with AP frequency and contributes to AP adaptation in CCs.

The two articles by Martinez-Espinosa et al. (2021a, 2021b) in the earlier issue shed new light on many of the unsolved questions about CCs’ excitability and bring new evidence concerning the role of Nav channels in three key aspects of the phenomenon. First, what is the main Nav channel isoform expressed in rodent CCs? Second, does the on–off kinetics of long-term inactivation develop independently of fast recovery and play a role on AP firing adaptation? And third, which are the intracellular molecular components that interfere with the endogenous fast inactivation to promote the slow recovery from inactivation? Using rat CCs and properly designed voltage clamp paired-pulse protocols, the authors show first that recovery from Nav channel inactivation exhibits two exponential components of similar magnitude but with short (3–30 ms) and long time constants (50–400 ms; Martinez-Espinosa et al., 2021a). Both processes are complete even after brief depolarizing pulses to 0 mV (5 ms) that fully inactivate the channels. Nav channels enter into the fast or slow recovery paths via largely independent (competing) inactivation mechanisms with similar onset time and independent of voltage (red arrows in Fig. 1 A). During trains of brief depolarization, the two recovery paths are largely independent but develop with time constants that differ by an order of magnitude. The fraction of Nav channels in the fast recovery state recovers fully, whereas the fraction in the slow recovery state recovers only partially during repetitive stimulation. Depending on the frequency of the pulse train (1–10 Hz), the slow recovery pathway will cause Nav channels to accumulate in the long-term inactivated state, reducing in a use-dependent manner the Nav current amplitude (Nav availability) during high-frequency pulse trains (10 Hz). Fig. 1 B, derived from data in Martinez-Espinosa et al. (2021a), shows how inward Nav1.3 currents decrease progressively during trains of 10 pulses at 1, 4, and 10 Hz and how the percentage of slowly recovering Nav channels increases at the end of the train with increasing frequency (52% at 1 Hz, 68% at 4 Hz, and 80% at 10 Hz). In other words, the dual independent inactivation pathways (fast and slow) dynamically regulate Nav availability by promoting use-dependent accumulation in the slow recovery pathway. It appears as if Nav channels effectively “measure” the frequency of APs and convert changes in AP frequency into Nav current availability (up to 80% at 10 Hz at −80 mV holding potential; see Fig. 1 B).

In the second article, taking advantage of the existence of Nav1.3 (Nassar et al., 2006) and FGF14 (Wang et al., 2002) KO mice, Martinez-Espinosa et al. (2021b) identify the molecular components that regulate long-term inactivation of Nav channels in mouse CCs. They first show that nearly half of Nav1.3 KO CCs do not possess TTX-sensitive fast-inactivating Nav currents; the remaining cells exhibit Nav currents of smaller size compared with WT cells. This agrees with quantitative RT-PCR data, which support the existence of a dominant fraction of the SCNA3 α-subunit (Nav1.3) and a minor contribution of the SCNA9 α-subunit (Nav1.7) in mouse CCs (Vandael et al., 2015), proving unambiguously that Nav1.3 is the major isoform contributing to Nav currents in mouse CCs. This challenges the enduring view that Nav1.7 is the dominant sodium channel in neuroendocrine cells. The authors also provide evidence that rat CCs express Nav1.7 and Nav1.3 channels in a ratio of ∼3:1, suggesting a partial contribution of Nav1.3 channels to the total Nav currents also in rat CCs. Thus, the Nav1.3 channel, which is mainly expressed in central and sensory neurons (Cummins et al., 2001; Catterall et al., 2005), appears to also be the main Nav channel responsible of CCs electrical activity. Another central finding of the article is the identification of the molecular entities regulating the slow recovery from fast inactivation. The authors show that deletion of FGF14 proteins in FGF14 null mice causes a marked reduction in the number of mouse CCs exhibiting the slow component of recovery from fast inactivation. Although the onset of fast inactivation is more rapid than in WT CCs, deletion of the slow recovery component preserves the fast recovery process, confirming that the two processes are largely independent. Finally, the authors show that the use-dependence of Nav current reduction during trains of brief stimuli in WT cells is completely abolished in FGF14 null cells, demonstrating the key role of FGF4-mediated inactivation in regulating Nav availability during cell firing.

Together, the two articles show unequivocally that mouse and rat CCs express TTX-sensitive fast inactivating Nav1.3 channels, and that their slow recovery from fast inactivation sets the fractional Nav availability that regulates the frequency and size of spontaneous and evoked AP firing. To accomplish this, Nav1.3, like other Nav isoforms (Goldfarb, 2012; Navarro et al., 2020), interacts with intracellular FGF subunits to drive the channel during fast inactivation into a long-term inactivation state, and the kinetics of this process sets Nav availability and AP firing. The present findings on the recovery of Nav channels from inactivation allow for more precise interpretation of the recent reports on the different firing modes and secretory activity of rodent CCs summarized into two recent reviews (Lingle et al., 2018; Carbone et al., 2019). There are, however, still open questions on how Nav availability may contribute to the transition of CCs from spontaneously firing to slow wave bursting (Wallace et al., 2002; Martinez-Espinosa et al., 2014; Vandael et al., 2015; Guarina et al., 2017; Milman et al., 2021). As suggested by Martinez-Espinosa et al. (2021b), this might be solved using dynamic clamp experiments to evaluate the impact of conventional fast inactivation versus dual pathway fast inactivation on patterns of CC firing.

It is also worth underscoring that these two articles in JGP go beyond the role of the Nav1.3/FGF14 partnership in regulating neuroendocrine cells electrical activity. They provide a general view on how Nav1.3, like other Nav channels (Nav1.2, Nav1.5, and Nav1.6), could interact with various iFGF proteins in central neurons and other excitable cells to act as a molecular “leaky integrator” that detects APs, measures the frequency, and regulates neuronal firing to provide modulatory signals to numerous brain functions (Navarro et al., 2020).

Olaf S. Andersen served as editor.

This work was supported by residual funds of the Telethon Foundation (grant no. GGP15110) to Emilio Carbone.

The author declares no competing financial interests.