Rethinking the role of L-type voltage-gated calcium channels in fear memory extinction.

The neurobiological mechanisms underlying extinction of fear memories have attracted considerable experimental interest in recent years, due in part to the clinical implications of this line of research for the treatment of fear-based psychiatric disorders (Milad et al. 2006; Sotres-Bayon et al. 2006). Several years ago, it was discovered that antagonists of the L-type voltage-gated calcium channel (L-VGCC) impair both extinction learning and longterm recall of fear extinction. In those studies, mice given systemic administration of either of the dihydropyridine (DHP) compounds nifedipine or nimodipine were observed to exhibit impairments in fear extinction, as measured by “freezing” behavior, both within a session and also when tested ∼24 h later (Cain et al. 2002, 2005; Suzuki et al. 2004). This finding was of interest given the well-established role of L-VGCCs in synaptic plasticity (Grover and Teyler 1990), in which signaling via the L-VGCC is thought to link activity at the membrane with transcriptional events in the nucleus (Dolmetsch et al. 2001). Further, L-VGCCs are known to play an essential role in synaptic plasticity in the amygdala (Weisskopf et al. 1999; Bauer et al. 2002), suggesting that signaling via L-VGCCs might be a mechanism by which long-term extinction memories are formed and stored. While an initially attractive hypothesis, three new articles appearing in this issue of Learning & Memory force us to re-examine that view. McKinney et al. (2008, this issue) use a genetic and pharmacological approach to examine the role of different L-VGCC isoforms in extinction of context fear memory. L-VGCCs consist of a variety of subunits, of which Cav1.2 and Cav1.3 are the major isoforms expressed in the brain (Striessnig et al. 2006). In a previous study, McKinney and Murphy showed that forebrain deletion of the Cav1.3 isoform impairs consolidation of context fear memory; that is, fear acquisition and short-term memory (STM) are intact, while long-term memory (LTM) is impaired (McKinney and Murphy 2006). However, in that study, deletion of Cav1.3 had no effect on fear extinction; Cav1.3 knockout mice show normal within-session extinction and intact long-term recall of extinction 24 h later (McKinney and Murphy 2006). In the present article, McKinney et al. (2008) examine the role of Cav1.2 in fear memory extinction. A recent study found that deletion of Cav1.2 impairs spatial learning and the L-VGCC–dependent form of LTP in the hippocampus (Moosmang et al. 2005). Surprisingly, however, McKinney et al. (2008) find that mice lacking Cav1.2 in the forebrain have intact acquisition, consolidation, and extinction of fear memory. Thus, using a genetic approach, forebrainexpressed Cav1.3 appears necessary for fear memory consolidation, but neither Cav1.3 nor Cav1.2 appears to be necessary, at least in isolation, for fear memory extinction. Turning to a pharmacological analysis, McKinney et al. (2008) next revisit the question of whether L-VGCC antagonists impair fear extinction, and, if so, why this method of manipulating L-VGCCs is so much more effective than using a molecular genetic approach. They first verify that systemic administration of a dose of nifedipine that has been used in previous experiments (Cain et al. 2002, 2005) impairs fear extinction. In their hands, mice treated with nifedipine exhibit impaired withinsession extinction relative to vehicle-injected mice and impaired extinction recall ∼24 h later. Remarkably, however, this same dose of nifedipine is observed to impair locomotor activity in an open field test and to exhibit aversive properties in its own right. That is, pairing of nifedipine alone with a novel context appears sufficient to condition a long-lasting fear-like response to that context and in a context-specific manner. Effectively, extinction training under the influence of nifedipine appears to amount to retraining using a different US. Thus, the authors suggest, the within-session impairment of fear extinction following systemic injections of L-VGCC antagonists is likely the result of acute toxicity masking as freezing behavior, rapid re-acquisition of fear to the context in which the animal is being extinguished, or both factors. Similarly, impaired long-term “extinction recall” (at 24 h) is likely due, in part, to long-term recall of fear that has been re-acquired as a consequence of pairing the extinction context with systemic L-VGCC blockade. The findings of Busquet et al. (2008, this issue) cast further doubt on the role of L-VGCCs in fear memory extinction. In agreement with previous findings (Cain et al. 2002, 2005; Suzuki et al. 2004), they first show that systemic administration of nifedipine impairs fear extinction learning to both contextual and auditory cues; mice receiving vehicle injections show a characteristic extinction curve, while those receiving nifedipine continue to exhibit “freezing” behavior throughout the extinction session. Using a novel mouse model expressing DHP-insensitive Cav1.2 L-VGCCs, they next show that this effect is completely blocked in Cav1.2DHP / mice, suggesting that the effect of systemically administered nifedipine on fear extinction is mediated through Cav1.2 rather than Cav1.3. Remarkably, however, Busquet et al. (2008) fail to find an effect of multiple doses of nifedipine on fear extinction following intracerebroventricular infusion of the drug, suggesting that the effects of nifedipine on fear extinction are mediated via peripheral Cav1.2 channels. Further, in a test of activity in the open field, they show that mice given systemic injection of nifedipine exhibit marked reductions in exploratory behavior, possibly indicative of a hypotensive effect. This latter finding is consistent with the peripheral distribution of Cav1.2 channels and the documented actions of L-VGCCs on peripheral vasodilation and hypotension (Kubo et al. 1981; Barrett et al. 1988). At first glance, it might appear counterintuitive that a peripherally acting drug can influence the formation of an extinction memory if not by acting in the brain. The report by Waltereit et al. (2008, this issue), however, suggests a mechanism by which L-VGCC antagonists may act peripherally to influence fear memory extinction. In their experiments, Waltereit et al. systematically vary the timing of systemic injections of nifedipine relative to the onset of extinction training. They find that i.p. or s.c. injections of nifedipine are highly effective at impairing extinction of either context or auditory fear memories when given as many as 2 or 4 h prior to extinction training, respectively. Remarkably, however, the authors show that nifedipine peaks in blood serum as quickly as 30 min after administration and is nearly undetectable by 2 h following i.p. injection and by 4 h following 1Corresponding author. E-mail [email protected]; fax (203) 432-7172. Article is online at http://www.learnmem.org/cgi/doi/10.1101/lm.996908. Commentary