Is Slack an Intrinsic Seizure Terminator?

A Brief History of Slick and Slack

Sodium-activated K⁺ currents were initially discovered in guinea pig cardiac cells in the mid-1980s (Kameyama and others 1984), and the presence of a neuronal K⁺ current activated by Na⁺ influx was first described in avian trigeminal ganglion neurons (Bader and others 1985). I_KNa was subsequently found in several other areas, including the mammalian neocortex (Schwindt and others 1989) and olfactory bulb (Egan and others 1992). More recently, it was discovered that sodium-activated potassium (K_Na) channels are encoded by two genes belonging to the Slo family of Ca²⁺-activated K⁺ channels: Slick (Slo2.1) (Bhattacharjee and others 2003) and Slack (“sequence like a Ca²⁺-activated K⁺ channel”; Slo2.2) (Yuan and others 2003), which are expressed widely in central neurons (Bhattacharjee and others 2002; Bhattacharjee and others 2005). The heterogeneity of I_KNa is increased by alternative splicing of Slack into at least five different isoforms (Brown and others 2008) and the formation of heteromeric channels containing Slick and Slack (Chen and others 2009). The roles and properties of the alternative channel configurations are not yet well understood, but channels encoded by Slick and Slack appear to have distinctly different properties. Slick activates rapidly in response to Na⁺ influx, whereas Slack activates slowly with prolonged depolarizations (Bhattacharjee and others 2003; Yuan and others 2003). K⁺ currents through K_Na channels contribute to a large portion of the total outward current in mammalian neurons (Fig. 1) and are therefore thought to play important roles in normal physiology (Budelli and others 2009).

OPEN IN VIEWER

Figure 1.

I_KNa is responsible for a major proportion of the total outward current in central neurons. (A) Effect of the Na⁺ channel blocker tetrodotoxin (TTX) on the total outward current in cultured rat mitral/tufted cells. Current-voltage (IV) relationships were obtained from current responses to voltage steps evoked from a holding potential of −70 mV. Control traces were obtained in calcium-free medium to remove the influence of Ca²⁺ channels (top traces; blue data points). When TTX (1 µM) was added to block Na⁺ channels, a substantial loss of outward K⁺ current occurred (middle traces; green data points). The bottom traces and yellow data points show the TTX-sensitive K⁺ current obtained by subtraction. (B) IV relationship of mitral/tufted cells in the absence (top traces; blue) and presence (middle traces; green) of the I_NaP blocker riluzole (20 µM), and the riluzole-sensitive current (bottom traces; yellow). The IV curves in A and B were constructed using the average of the recorded current at 150 to 250 ms from the beginning of the voltage step. See Budelli and others (2009) for full details. Adapted by permission from Macmillan Publishers Ltd: Nature Neuroscience (12:745-50), copyright (2009).

Afterhyperpolarizations by I_KNa Last Many Seconds

Several observations indicate that I_KNa may be responsible for unusually prolonged afterhyperpolarizations (AHPs), which are triggered under conditions of sustained influx of Na⁺. For instance, I_KNa is activated following bursts in intrinsically bursting neocortical cells (Franceschetti and others 2003) and also generates long-lasting (several seconds) outward currents following ~1 s trains of action potentials in other central neurons (Safronov and Vogel 1996; Zhang and others 2010). In neocortical neurons, a sustained (20 s) depolarization to above the activation threshold for the persistent Na⁺ current (I_NaP) resulted in a gradually increasing I_KNa during the plateau and a very slow decay of I_KNa over tens of seconds upon repolarization (Schwindt and others 1989). Similarly, Na⁺-dependent AHPs lasting up to 15 s were observed in thalamic neurons after intense tonic- and clonic-like activity (Kim and McCormick 1998) (Fig. 2), and high-frequency firing elicited very long-lasting AHPs in songbird brains (Kubota and Saito 1991). Thus, sustained Na⁺-dependent depolarizations appear to strongly activate I_KNa and cause AHPs that are much more prolonged than the fast and medium AHPs caused by I_KCa channels (Wulff and others 2007). These properties make I_KNa theoretically well-suited for mediating a prolonged depressing influence after the sustained depolarizations that accompany many types of seizures.

OPEN IN VIEWER

Figure 2.

Prolonged afterhyperpolarizations (AHPs) during tonic- and clonic-like neuronal activity in thalamic neurons of ferret perigeniculate slices. (A) Rapid action potential firing induced by depolarizing current injection from a membrane potential of approximately −60 mV activated a prolonged AHP that became more pronounced at higher firing frequencies. (B) Action potential bursts elicited by current injections were followed by a slow AHP, which persisted for 5 to 10 s after the end of stimulation. The slow component of the AHP was not Ca²⁺ dependent and could be blocked by Na⁺ channel blockers or by a reduction in extracellular [Na⁺]. See Kim and McCormick (1998) for full details. Adapted with permission from Kim and McCormick, 1998. Journal of Neurophysiology, 80:1222-35.

I_KNa Is Activated by the Persistent Na⁺ Current

Early single-channel studies showed that K_Na channel activation required an intracellular Na⁺ concentration far exceeding physiological levels. Intracellular recordings, in contrast, indicated the presence of Na⁺-dependent outward currents in several brain regions during relatively normal activity (Dryer 1994). Budelli and others (2009) recently offered a possible explanation for this discrepancy. Using whole-cell recording from cultured dissociated neurons, they showed that I_KNa was strongly activated by persistent Na⁺ currents (I_NaP) (Fig. 1B) and that this effect persisted even when neurons were loaded with high intracellular [Na⁺]. They suggested that I_NaP and K_Na channels may cluster in the membrane and that Na⁺ ions may accumulate locally near electrostatic domains of the K_Na channel protein (Budelli and others 2009). I_NaP accounts for a small proportion (<1%) of the total Na⁺ current in many central neurons and has a strong depolarizing effect that amplifies subthreshold inputs and facilitates repetitive action potential firing (Crill 1996). I_NaP is likely to play a major role in epileptic seizures due to its activation during long-lasting depolarizations and repetitive action potential firing (Stafstrom 2007).

Na⁺ Influx During Seizures Is Likely to Activate I_KNa

A common electrographic seizure pattern seen on electroencephalogram and in field potential recordings is a tonic-clonic progression, with a fast regular oscillation (tonic phase) followed by clustered bursts (clonic phase). Recordings of seizure-like events in brain slices have revealed that the tonic phase is accompanied by a sustained direct current shift, generated by simultaneous depolarization of neurons and extrusion of K⁺ ions. At the end of the tonic phase when neurons begin to repolarize, synaptically driven clonic bursts appear and recur at a gradually decreasing frequency until the seizure terminates and the postictal hyperpolarization takes over (Fig. 3). During the tonic phase, neurons typically fire action potentials at high frequencies or depolarize to such a degree that depolarization block of action potential generation occurs (e.g., Anderson and others 1986; Konnerth and others 1986; Walther and others 1986; Aram and Lodge 1988; Jones and Heinemann 1988; Traynelis and Dingledine 1988; Avoli 1990; Dreier and Heinemann 1991; Gloveli and others 1995; Jensen and Yaari 1997; Luhmann and others 2000; Trevelyan and others 2006; Igelstrom and others 2011). Such activity—both rapid firing and sustained depolarizations—is expected to involve Na⁺ currents (Stafstrom 2007).

OPEN IN VIEWER

Figure 3.

Paired intracellular and extracellular recording during seizure-like events in the hippocampal slice bathed in ictogenic high-potassium medium, illustrating typical neuronal activity during seizures. (A) Top trace shows an intracellular recording of a hippocampal CA1 pyramidal neuron, and bottom trace shows a simultaneous field potential recording. The segment indicated with a black bar is expanded in (B). Adapted with permission from Jensen and Yaari, 1997. Journal of Neurophysiology, 77:1224-33.

The exact roles of fast and persistent Na⁺ currents in clinical seizures are not known and may vary between seizure types and brain regions. However, many currently used antiepileptic drugs act at least in part by inhibiting both persistent and fast-inactivating Na⁺ currents (Rogawski and Löscher 2004), and when tested in vitro, Na⁺ channel inhibitors generally block seizure-like events (Khalilov and others 1999; Ohno and Higashima 2002; Kraglund and others 2010; Igelstrom and others 2011). In addition, extracellular [Na⁺] decreases in brain slices during seizure-like events (Yaari and others 1983; Leschinger and others 1993) and during seizures in vivo (Meyer and others 1961). These findings support the importance of Na⁺ influx during seizures. Although intracellular [Na⁺] has not been experimentally measured during seizure activity, it has been shown to increase in hippocampal neurons during trains of action potentials (Rose and others 1999) and during tetanic stimulation of afferents (Rose and Konnerth 2001). This indicates that considerable increases in intracellular Na⁺ ([Na⁺]_i) can accompany rapid firing and sustained depolarizations. Elevations of [Na⁺]_i can also be mediated by Na⁺ entry through N-methyl-D-aspartate receptors (Rose and Konnerth 2001). Seizure-like events induced in brain slices are often dependent on N-methyl-D-aspartate receptors (Walther and others 1986; Jensen and Yaari 1997; Khalilov and others 1999; Quilichini and others 2002; Fujiwara-Tsukamoto and others 2006; Igelstrom and others 2011), supporting the possibility of this additional route of Na⁺ influx during seizures. In summary, Na⁺ influx through several routes seems to be a prominent component of ictal activity. The resulting elevations in [Na⁺]_i would be expected to activate I_KNa. For K_Na channel forms with slow kinetics, like Slack, the long-lasting hyperpolarizing influence may contribute both to seizure termination and the post-ictal hyperpolarization.

Termination In Silico

In silico approaches show great promise for addressing mechanistic aspects of seizures and could be applied to studying the role of I_KNa in seizure termination. Krishnan and Bazhenov (2011) recently constructed a realistic computational model of a cortical network that included ion regulatory mechanisms. Using this model, they showed that seizure-like events were accompanied by accumulation of intracellular Na⁺ and that this in itself led to seizure termination. The exact mechanism of seizure termination in this model was not clear. The model neurons did contain a Na⁺-sensitive K⁺ current, with properties based on dorsal root ganglion neurons (Bischoff and others 1998; Krishnan and Bazhenov 2011), but its role in seizure termination was not discussed. A recent computational model of the entorhinal cortex 4-aminopyridine model achieved seizure termination by adding an activity-dependent “slow depression variable” (Vincent and others 2011). An in silico approach incorporating I_KNa with slow kinetic properties, such as those observed in the neocortex (Schwindt and others 1989) and thalamus (Kim and McCormick 1998), may be useful to test the plausibility of a role of I_KNa in seizure termination.

Time to Take Up the Slack

The preferential activation of I_KNa by depolarizations with durations that are physiologically rare—but typical for seizures—argues for a possible role of I_KNa in seizure termination. I_KNa has long been overlooked in electrophysiology studies because of the common use of Na⁺ channel blockers when studying K⁺ currents. Even following the discovery of Slick and Slack, the study of K_Na channels in native neurons has been hampered by the lack of specific antagonists, and no Slack genetic animals have yet been used. With today’s snowballing developments in molecular biology and pharmacology, it is only a matter of time before these tools become available. In parallel, the ancient conundrum of seizure termination will begin to be addressed with more powerful methods. The use of modern technology in the elucidation of the role of an “old” current in an “old” phenomenon will lead to novel understanding of ictogenesis and potentially even to new treatments.