13 April 2012. Neurons rely on calcium to mediate between electrical and chemical signaling, but they must handle it with care, as too much of the ion can do damage. A paper in the April 4 Journal of Neuroscience supports an emerging theory that abundant calcium makes neurons vulnerable. The authors, from the Mediterranean Institute of Neurobiology in Marseille, France, found that the endoplasmic reticulum-based calcium buffer calreticulin dwindles in those motor neurons most likely to succumb to amyotrophic lateral sclerosis. Calreticulin, also an ER chaperone, sponges up excess calcium to protect the organelle and the cell. Loss of calreticulin in neuron models of ALS not only elevates cellular concentrations of the cation, but also appears to exacerbate a feedback loop that dooms the neurons, the researchers found. Boosting calreticulin as a potential therapy would require a neuron-specific approach, since simply overexpressing calreticulin alters the heartbeat and eventually kills mice (Nakamura et al., 2001).
Only motor neurons degenerate and die in ALS, but certain subsets are more vulnerable than others. The discovery of calreticulin loss in ALS model cells and mice “makes a new point in understanding why certain motor neurons die in ALS,” said senior author Brigitte Pettmann, who led the work with first author Nathalie Bernard-Marissal. The team examined the selective vulnerability of motor neurons to a specific cell-death pathway mediated by the Fas death receptor and nitric oxide. Previously, Pettmann’s group showed that motor neurons from mice overexpressing the mutant human superoxide dismutase 1 (SOD1) that causes ALS are particularly sensitive to the effects of Fas and NO (see ARF related news story on Raoul et al., 2002). In these neurons, activation of Fas turns up NO production; NO then boosts the expression of the Fas ligand FasL, amplifying a vicious cycle. Previously, the group screened for proteins up- or downregulated in mSOD1 motor neurons affected by Fas signaling, and a halving of calreticulin levels was among the few changes they observed (Duplan et al., 2010).
To examine the importance of calreticulin in cell death, the researchers artificially heightened and lowered its levels. In cultures of wild-type NSC34 cells, reducing calreticulin with RNA interference killed the neurons. Overexpressing calreticulin in embryonic motor neurons from SOD1-G93A mice saved 60 percent of them, even though FasL treatment normally kills these cells. Bernard-Marissal and colleagues concluded that loss of calreticulin is involved in the death of mSOD neurons susceptible to Fas and NO.
Next, the team went in vivo with the mSOD1 model mice. Staining the spinal cord for calreticulin, they observed that the intensity dropped by half, compared to wild-type mice, starting at 38 days of age. Importantly, calreticulin did not fade in all motor neurons, just the ALS-vulnerable ones. The scientists applied retrograde labeling to identify susceptible neurons innervating the tibialis anterior, and resistant ones innervating the soleus muscles. At 38 days, only the tibialis-linked neurons lost calreticulin.
Returning to embryonic cultures, the researchers determined calcium concentration with calcium-sensitive dyes. They saw that mutant SOD1 motor neurons sported elevated basal calcium levels and, following FasL treatment, calcium spiked twice as much as in wild-type cells. This suggests loss of calcium homeostasis in the mutant neurons, presumably because the ER cannot sop up excess calcium without calreticulin.
The ER became stressed in the vulnerable cells. Upon treatment with FasL, mSOD1 neurons manufactured more of the ER stress markers p-eIF2α and CHOP than wild-type cultures. The drug salubrinal, which blocks ER stress, protected mSOD1 neurons from Fas-mediated death. Salubrinal is unsuitable as a therapeutic, Bernard-Marissal noted, because it causes side effects such as long-term memory loss. The drug inhibits p-eIF2α phosphatases (see ARF related news story), leaving the initiation factor free to promote translation of the β-secretase that generates amyloid-β (see ARF related news story). The authors believe ER stress occurs downstream of calreticulin loss, in part because salubrinal prevented cell death in cultures treated with RNAi to silence calreticulin.
The researchers propose a pathway by which the production of NO leads to calreticulin tapering. This would induce not only synthesis of FasL, amplifying the feedback loop creating more NO, but also ER stress, perhaps due to increased calcium levels. The precise steps remain incomplete, but in essence, Pettmann believes that Fas/NO-vulnerable motor neurons expressing mSOD1 are unable to handle these insults, so they die. Pettmann only observed calreticulin loss and degeneration in the vulnerable mSOD1 motor neurons; it did not occur in control cells.
Calreticulin’s protein-folding function could also be involved in neurodegeneration, noted Marek Michalak of the University of Alberta in Edmonton, Canada, who was not involved in the study. “The simplest explanation is that less of the molecular chaperone is available and, therefore, protein-folding quality control will be affected,” he said. However, he opined that most likely a combination of ER stress and misfolded proteins interact to promote motor neuron degeneration. Michalak expressed surprise that calreticulin—normally a marker for ER stress—dropped in the study model. Somehow, the neurons in question must deal with stress differently from most cell types, he speculated.
Pettmann’s work follows other studies indicating that excess calcium poisons motor neurons in ALS (Appel et al., 2001). The twist is that she studied an ER, instead of cytoplasmic, calcium buffer, said Ilya Bezprozvanny of the University of Texas Southwestern Medical Center at Dallas, who was not part of the study team. Neurons, compared to other cell types, rely strongly on calcium during action potential firing, he said. With age, the neuron’s ability to buffer calcium levels abates, and if calreticulin levels drop, “it weakens the ability of the ER to handle the calcium, effectively meaning that the mitochondria have to do more,” Bezprozvanny said. Calcium ions stimulate mitochondria, yielding reactive oxygen species that damage the organelles and ultimately the entire cell, he suggested.
Pettmann’s “findings may have significance beyond ALS, since both increased ER stress and decreases in calreticulin have been detected in other major neurodegenerative diseases,” agreed Pico Caroni of the Friedrich Miescher Institute for Biomedical Research in Basel, Switzerland, in an e-mail to ARF. Caroni studies selective motor neuron vulnerability. For example, calreticulin expression ebbs in the brains of people with Alzheimer’s (Taguchi et al., 2000) and in rat models for Parkinson’s (Lessner et al., 2010). The calcium-binding protein calbindin also disappears from neurons vulnerable to AD (Riascos et al., 2011). Pettmann believes that calreticulin loss could be associated with neurodegeneration in general. Any time calreticulin melts away, it becomes harder for neurons to manage the ER stress that is common to all neurodegenerative conditions, she said.
Designing a treatment to influence the calreticulin pathway in people could be tricky. “Calcium homeostasis, calreticulin levels, and ER stress are such universal cellular components and processes,” Caroni noted. “Nevertheless, calreticulin regulation seems to provide a specific entry point.”—Amber Dance.
Bernard-Marissal C, Moumen A, Sunyach C, Pellegrino C, Dudley K, Henderson CE, Raoul C, Pettmann B. Reduced calreticulin levels link endoplasmic reticulum stress and Fas-triggered cell death in motoneurons vulnerable to ALS. J. Neurosci. 2012 Apr 4;32(14):4901-12. Abstract