Ca Channels May Promote Influx and Trigger Release from Internal Stores

Other Ca influx pathways also likely contribute to Ca accumulation. Experiments in rat optic nerve (Fern et al., 1995a; Brown et al., 2001b) and spinal dorsal columns (Imaizumi et al., 1999) indicate that L- and N-type Ca channels contribute to anoxic injury in these preparations. Although it is likely that these channels allow flux of Ca across plasma membranes, L-type channels may contribute in another role by sensing axonal depolarization and activating Ca release from ryanodine receptors (Ouardouz et al., 2003). Indeed, electron probe studies failed to show a reduction of net Ca accumulation in anoxic optic axons exposed to L-type Ca channel blockers such as nifedipine or nimodipine (Stys and LoPachin, 1998). It is possible that an important role of L-type channels in ischemic axons is activation of ryanodine receptors and release of internal Ca stores, rather than flux of extracellular Ca across the axolemma. Recent experiments in ischemic rat dorsal column (Ouardouz et al., 2003) have shown that unlike anoxia, more prolonged exposure to in vitro ischemia (oxygen-glucose deprivation) results in severe functional injury that cannot be rescued by removal of extracellular Ca. Yet robust protection is afforded by pre-treatment with the high affinity Ca chelator BAPTA as the acetoxymethyl ester to allow penetration across cell membranes. These findings strongly suggest that deleterious Ca is released from an intracellular source, such as endoplas-mic reticulum or mitochondria. Of interest, adding the

L-type Ca channel blocker nimodipine to ischemic dorsal columns already perfused with zero-Ca/EGTA solution increased postischemic recovery of compound action potential amplitudes from 2% in zero-Ca/EGTA alone to 62% with the addition of nimodipine. Clearly, this agent was not neuroprotective as a result of reduction of Ca influx through L-type channels. Instead, it was hypothesized that its interference with voltage sensing by these channels inhibited depolarization-induced release of Ca from endoplasmic reticulum, which proceeded through ryanodine receptors, analogous to Ca-independent "excitation-contraction coupling" in skeletal muscle (FranziniArmstrong and Protasi, 1997). This mechanism is further supported by protective effects of thapsigargin, which depletes Ca stores; ryanodine, which partially blocks release of Ca from endoplasmic reticulum; and either zero-Na perfustate or TTX, which will reduce ischemic depolarization, thereby dampening the change in membrane potential sensed by the Ca channels. Ryanodine receptors (RyR1 and RyR2) and Ca channels (Cav1.2 and Cav1.3) were shown to co-immunoprecipitate (respectively) and be spatially co-localized at the axolemma of dorsal column axons (Fig. 3), further supporting an "excitation-contraction coupling"-like Ca release in these fibers (Ouardouz et al., 2003). Indeed, ryanodine receptor-dependent Ca release may not be the only source of intracellular Ca in ischemic central axons. Inhibition of group I metabotropic glutamate receptors, in addition to external Ca removal and application of nimodipine to block the aforementioned mode of release, allows virtually complete functional recovery of dorsal columns after an hour of severe ischemia (Stys and Ouardouz, 2002). We interpret this additional protection as inhibition of Ca release from IP3 receptors, stimulated by IP3 generated by phospholipase C through mGluR1 activation. These mechanisms may be of great importance because they underscore the argument that control of extracellular Ca influx is necessary, but not sufficient, to protect central axons from more severe ischemic injury.

Pathological states such as inflammatory demyelination may promote inappropriate ectopic insertion of channels into the axon membrane. Lassmann and co-workers showed an excess of N-type Ca channel subunit accumulation at the axolemma in areas of spheroid formation within actively demyelinating lesions in both human brain and EAE animals (Kornek et al., 2001). This raises the interesting possibility that normal axons (at least those not previously exposed to a chronic insult such and inflammatory demyelination) may have very low densities of Ca channels, but chronic pathological states may increase channel densities, thus promoting Ca influx and Ca-dependent injury. What is not yet clear is whether these accumulated channels are functionally inserted into the axolemma of pathological axons where they can impart an increased Ca permeability.

AX

100 nm

Figure 3 Immunolocalization of Cav and RyR in spinal dorsal columns. (A—D) Triple labeled sections (Cav1.3/RyR2/neurofilament) show many Cav and RyR clusters, which are occasionally co-localized. Clusters were found near the surfaces of axon cylinders and within glial structures. Deconvolution (C, D) reveals a more accurate localization of cluster pairs at the surface of an axon cylinder, overlying a neurofilament-poor area, possibly representing an ER cistern. YZ projection (D) shows elongated "fingers" of associated Cav/RyR complexes. Similar distributions of Cav1.2/RyR1 profiles are shown in panels E-J, also associated with axon cylinders and glial regions (I; deconvolved: J). Control sections with primary antibodies omitted showed no signal (K). Double immuno-gold staining (L) using pan-RyR (small grains) and Cav1.3 (large grains) shows close association of both proteins at the axolemma. (Reproduced from Ouardouz et al., 2003, with permission.)

0 0

Post a comment