Neuronal voltage-gated calcium (Ca2+) channels (VGCCs) are multimeric protein complexes that are localized to the presynaptic membrane of neurons, where they serve several functions. VGCCs are composed of the Cav2.x pore-forming subunit surrounded by β, γ, and α2δ auxiliary subunits, which control channel trafficking and activity (Walker and De Waard, 1998). These channels allow functional coupling between action potentials and Ca2+ entry into nerve terminals, where they control evoked neurotransmitter release. They also play an important role in the control of neurite outgrowth at early developmental stages and are involved in the organization of functional neuromuscular synapses during the development of motor nerve terminals (Nishimune et al., 2004). The latter function is mediated by a direct interaction of the 11th extracellular loop of Cav2.1 (P/Q-type Ca2+ channel) and Cav2.2 (N-type Ca2+ channel) subunits of VGCCs with a leucine-arginine-glutamic acid (LRE) motif of the β2 chain of laminin, a component of the extracellular matrix. In a recent issue of The Journal of Neuroscience, Sann et al. (2008) provide new insights into the functional importance of the interaction between the Cav2.2 subunit and laminin β2 (Sann et al., 2008). They demonstrate in vitro and in vivo in Xenopus laevis embryos that Cav2.2–laminin β2 interaction acts as a stop signal for neurite outgrowth during sensory innervation of the skin, allowing appropriate innervation of the target tissue.
Using immunohistochemistry on tailbud (stage 26) X. laevis embryos, the authors show that Cav2.2 channels are expressed in growth cones of extending commissural and Rohon-Beard sensory axons innervating the skin [Sann et al. (2008), their Fig. 1D,E (http://www.jneurosci.org/cgi/content/full/28/10/2366/F1)], as well as in growth cones of spinal neurons grown in vitro [Sann et al. (2008), their Fig. 1G,H (http://www.jneurosci.org/cgi/content/full/28/10/2366/F1)]. This observation suggests that Cav2.2 channels could be involved in cellular processes other than the control of synaptic activity, such as neurite outgrowth and guidance. In parallel, the authors show that the skin expression pattern of laminin β2 is restricted to hexagonal cells in 2-d-old embryos [Sann et al. (2008), their Fig. 2 (http://www.jneurosci.org/cgi/content/full/28/10/2366/F2)]. Hence, Cav2.2 channels of sensory neurons and laminin β2 are expressed in X. laevis embryos, where they may interact, as previously shown for Cav2.1 channels at the neuromuscular junction (Nishimune et al., 2004).
To investigate the functional significance of the Cav2.2–laminin β2 interaction in neurite outgrowth, X. laevis spinal neurons were grown in vitro, and extending processes were quantified. Whereas neurons had many processes when they were grown on a substrate of laminin-111 (devoid of the β2 chain), the addition of the C-terminal fragment of laminin β2 (containing the LRE domain) specifically inhibited neurite outgrowth [Sann et al. (2008), their Fig. 3A–C (http://www.jneurosci.org/cgi/content/full/28/10/2366/F3)]. Neurite outgrowth was rescued by the addition to the culture medium of a peptide corresponding to the 11th extracellular loop of the Cav2.1 subunit, which also binds to the LRE motif of the laminin β2 [Sann et al. (2008), their Fig. 3D,E (http://www.jneurosci.org/cgi/content/full/28/10/2366/F3)]. These results suggest that inhibition of neurite outgrowth by laminin β2 is mediated by direct interaction of the LRE motif with the 11th extracellular loop of VGCCs.
To further investigate this inhibition pathway and analyze the importance of Ca2+ signaling, neurons were grown in the presence of ω-conotoxin GVIA, a specific inhibitor of N-type Ca2+ channels. In these experimental conditions, laminin β2 did not affect the number of processes, suggesting that Ca2+ influx into neuronal growth cones through Cav2.2 channels is required for laminin-β2-mediated inhibition of neurite outgrowth [Sann et al. (2008), their Fig. 4A–C (http://www.jneurosci.org/cgi/content/full/28/10/2366/F4)].
To examine the involvement of Ca2+ influx in the laminin-β2-mediated stop signal, neurons were grown in vitro on a native or denaturated laminin-β2-coated strip, and the intracellular Ca2+ level ([Ca2+]i) in the growth cones was monitored by confocal microscopy. Fascinatingly, neurites, initially grown on the denaturated substrate, stalled or stopped when their growth cones encountered native laminin β2. This change in neurite growth was associated with a long-lasting increase of [Ca2+]i [Sann et al. (2008), their Fig. 5 (http://www.jneurosci.org/cgi/content/full/28/10/2366/F5)]. Together, these results suggest that Cav2.2–laminin β2 interaction mediates Ca2+ entry into growth cones through N-type Ca2+ channels, initiating an inhibition pathway of neurite outgrowth.
To extend these observations to in vivo neurite outgrowth during sensory innervation of the skin, an agarose bead releasing ω-conotoxin GVIA was implanted in the developing X. laevis embryo from the time of neural tube formation until a late larval stage when sensory terminals have been formed. This method allows efficient delivery of the toxin in vivo, at least 200 μm around the bead. In vivo inhibition of Ca2+ influx through N-type Ca2+ channels resulted in a twofold increase in the number of sensory nerve terminal clusters in the skin [Sann et al. (2008), their Fig. 6 (http://www.jneurosci.org/cgi/content/full/28/10/2366/F6)]. Similar results were obtained using a bead releasing the peptide of the 11th extracellular loop of the Cav2.1 channel, suggesting that the increase in nerve terminal clusters is likely to be mediated by the disruption of Cav2.2–laminin β2 interaction, as observed in vitro [Sann et al. (2008), their Fig. 7 (http://www.jneurosci.org/cgi/content/full/28/10/2366/F7)].
In summary, Sann et al. (2008) provide evidence that Cav2.2–laminin β2 interaction in X. laevis embryos mediates a Ca2+-dependent stop signal for neurite outgrowth during sensory innervation of the skin. The expression pattern of laminin β2, restricted to hexagonal cells, acts as a map for neurite outgrowth, which is read by the N-type Ca2+ channel expressed in the growth cone, ensuring precise innervation of the target tissue. This function is very different from that observed in the case of the mouse neuromuscular junction, where Cav2.1 binding to laminin β2 plays a scaffolding role (and not a Ca2+ signaling function) for precise presynaptic anchoring and enrichment of P/Q-type Ca2+ channels in active zones of motor nerve terminals (Nishimune et al., 2004).
These results leave some questions to be discussed, in particular regarding the molecular mechanism of Ca2+ influx in response to the contact of the growth cone with laminin β2. The authors suggest that laminin β2, by interacting with the Cav2.2 subunit, could induce a stretch activation of N-type Ca2+ channels, allowing Ca2+ influx into the growth cone, which in turn inhibits neurite outgrowth. However, although a mechanosensitivity of N-type Ca2+ currents has been demonstrated (Calabrese et al., 2002), there is no evidence that stretch alone is sufficient to activate N-type voltage-gated Ca2+ channels independently of any change in membrane potential. Hence, to discriminate between a stretch or voltage activation of the Ca2+ influx, it could be interesting to grow neurons on stripes of native and denaturated laminin β2 in the presence of tetrodotoxin (which blocks voltage-activated sodium channels), while monitoring [Ca2+]i in the growth cone. A similar experiment using a soluble laminin β2 C-terminal fragment, which does not interact with the coating of the culture dish (and thus would not induce any stretch of the cell in interacting with the Ca2+ channel), could also provide important information regarding the mechanism of Ca2+ influx into the growth cone.
Another unanswered question concerns the molecular identity of the channel involved in this Ca2+ influx. It was previously shown that Ca2+ entry through stretch-activated Ca2+ channels inhibits neurite outgrowth of X. laevis spinal neurons, in contrast to other influx pathways (such as via voltage-gated Ca2+ channels) that would have opposite effects (Jacques-Fricke et al., 2006). Here, based on the observation that ω-conotoxin GVIA inhibits laminin-β2-mediated inhibition of neurite outgrowth, Sann et al. (2008) suggest that the N-type voltage-activated Ca2+ channel is the channel supporting Ca2+ influx involved in this inhibition. Nevertheless, ω-conotoxin GVIA binds to the 11th extracellular loop of the Cav2.2 subunit (Feng et al., 2003), that is, on the same molecular determinant as laminin β2. Thus, we cannot rule out the possibility that the binding of ω-conotoxin GVIA to the Cav2.2 subunit, in addition to inhibiting the Ca2+ permeability of the channel, could also alter, via a bulky effect, laminin β2 interaction with the channel. Such an effect was proposed (Nishimune et al., 2004) as a mechanism by which antibodies to the 11th loop of the Cav2.1 subunit found in the sera of Lambert-Eaton myasthenic patients could induce the linked neuromuscular disease (Takamori et al., 2000). Coimmunoprecipitation experiments of the Cav2.2 subunit with laminin β2 in the presence of ω-conotoxin GVIA could resolve this question.
Finally, it could be interesting to examine whether a Ca2+ release from intracellular stores is associated with the Ca2+ influx in the laminin-β2-induced neurite outgrowth inhibition. Monitoring neurite outgrowth on native/denaturated stripes in the presence of pharmacological inhibitors of ryanodine and inositol trisphosphate receptors could further detail the nature of the Ca2+ signaling in the control of the sensory innervation.
In conclusion, the study by Sann et al. (2008) provides important insights in the understanding of molecular mechanisms controlling sensory innervation. It also opens an interesting therapeutic strategy: blocking Ca2+ influx into nerve terminals via local administration of specific Ca2+ channel inhibitors could favor the recovery of sensory innervation after a trauma.
Editor's Note: These short, critical reviews of recent papers in the Journal, written exclusively by graduate students or postdoctoral fellows, are intended to summarize the important findings of the paper and provide additional insight and commentary. For more information on the format and purpose of the Journal Club, please see http://www.jneurosci.org/misc/ifa_features.shtml.
- Correspondence should be addressed to Norbert Weiss, Laboratoire Physiologie Intégrative, Cellulaire et Moléculaire, Unité Mixte de Recherche 5123, Centre National de la Recherche Scientifique, Université Claude Bernard Lyon 1, Bâtiment Raphaël Dubois, 43 boulevard du 11 Novembre 1918, 69622 Villeurbanne cedex, France.