Neuroprotective drug riluzole amplifies the heat shock factor 1 (HSF1)- and glutamate transporter 1 (GLT1)-dependent cytoprotective mechanisms for neuronal survival

Neurotoxic injury pathways in differentiated mouse motor neuron-neuroblastoma hybrid (NSC-34D) cells in vitro--limited effect of riluzole on thapsigargin, but not staurosporine, hydrogen peroxide and homocysteine neurotoxicity

The neuroblastoma-spinal motor neuron fusion cell line, NSC-34, in its differentiated form, NSC-34D, permits examining the effects of riluzole, a proven treatment for amyotrophic lateral sclerosis (ALS) on cell death induction by staurosporine (STS), thapsigargin (Thaps), hydrogen peroxide (H(2)O(2)) and homocysteine (HCy). These neurotoxins, applied exogenously, have mechanisms of action related to the various proposed molecular pathogenetic pathways in ALS and are differentiated from endogenous cell death that is associated with cytoplasmic aggregate formation in motor neurons. Nuclear morphology, caspase-3/7 activation and high content imaging were used to assess toxicity of these neurotoxins with and without co-treatment with riluzole, a benzothiazole compound with multiple pharmacological actions. STS was the most potent neurotoxin at killing NSC-34D cells with a toxic concentration at which 50% of maximal cell death is achieved (TC(50)=0.01μM), followed by Thaps (TC(50)=0.9μM) and H(2)O(2) (TC(50)=15μM) with HCy requiring higher concentrations to kill at the same level (TC(50)=2200μM). Riluzole provided neurorescue with a 20% absolute reduction (47.6% relative reduction) in apoptotic cell death against Thaps-induced NSC-34D cell (p≤0.05), but had no effect on STS-, H(2)O(2)- and HCy-induced NSC-34D cell death. This effect of riluzole on Thaps induction of cell death was independent of caspase-3/7 activation. Riluzole mitigated a toxin that can cause intracellular calcium dysregulation associated with endoplasmic reticulum (ER) stress but not toxins associated with other cell death mechanisms.

Riluzole elevates GLT-1 activity and levels in striatal astrocytes

Drugs which upregulate astrocyte glutamate transport may be useful neuroprotective compounds by preventing excitotoxicity. We set up a new system to identify potential neuroprotective drugs which act through GLT-1. Primary mouse striatal astrocytes grown in the presence of the growth-factor supplement G5 express high levels of the functional glutamate transporter, GLT-1 (also known as EAAT2) as assessed by Western blotting and ³H-glutamate uptake assay, and levels decline following growth factor withdrawal. The GLT-1 transcriptional enhancer dexamethasone (0.1 or 1 μM) was able to prevent loss of GLT-1 levels and activity following growth factor withdrawal. In contrast, ceftriaxone, a compound previously reported to enhance GLT-1 expression, failed to regulate GLT-1 in this system. The neuroprotective compound riluzole (100 μM) upregulated GLT-1 levels and activity, through a mechanism that was not dependent on blockade of voltage-sensitive ion channels, since zonasimide (1 mM) did not regulate GLT-1. Finally, CDP-choline (10 μM-1 mM), a compound which promotes association of GLT-1/EAAT2 with lipid rafts was unable to prevent GLT-1 loss under these conditions. This observation extends the known pharmacological actions of riluzole, and suggests that this compound may exert its neuroprotective effects through an astrocyte-dependent mechanism.

Delayed neuroprotection by riluzole against excitotoxic damage evoked by kainate on rat organotypic spinal cord cultures

Kainate-mediated excitotoxicity of organotypic spinal cord cultures is an in vitro model advantageous to investigate basic mechanisms of acute spinal injury and its pharmacological neuroprotection. Using such cultures, the putative neuroprotective agent riluzole applied at 5 μM (plasma therapeutic concentration) was studied for its ability to prevent neurotoxicity evoked by 1 h administration of kainate. We monitored real-time release of glutamate, release of lactate dehydrogenase (LDH) (cell damage marker), occurrence of cell pyknosis, the number of surviving neurons and motoneurons, and cell culture metabolic activity. Co-applied riluzole strongly blocked the kainate-evoked early rise in extracellular glutamate (via calcium dependent or independent processes) and suppressed LDH release (limited to <20% of total). Although there were no significant cell losses within the first h after kainate washout, pyknosis, fewer neurons and motoneurons were observed 24 h later. MTT assay demonstrated that surviving cells were metabolically competent. Co-application of kainate and tetrodotoxin also failed to protect spinal cord slices 24 h later. When riluzole application begun at kainate washout and continued for 24 h, significant neuroprotection was observed for neurons in the central and dorsal regions, while ventral horn cells (including motoneurons) were not protected. Our data suggest that riluzole neuroprotection against excitotoxicity was feasible, although it paradoxically required delayed drug administration, and was not extended to the ventral horn. We propose that riluzole was acting on yet-unidentified processes downstream of glutamate release and receptor activation. Deciphering their identity and role in cell death mechanisms may be an important goal to develop neuroprotection.