It is the million-dollar question in research on amyotrophic lateral sclerosis: Why do only motor neurons fail when the cause of pathology, such as mutant protein, exists in many different cells? In a Nature Neuroscience paper published March 29, scientists from the Friedrich Miescher Institute in Basel, Switzerland, provide a clue—some motor neurons are particularly vulnerable because they are quicker to succumb to endoplasmic reticulum stress. By dissecting out and examining gene expression in neurons that degenerate first in a mouse model of ALS, the authors found that ER stress markers were upregulated in these most sensitive cells. They also discovered that a drug that subdues the ER stress response extended survival in the animals.

Principal investigator Pico Caroni and colleagues at the Friedrich Miescher Institute had previously shown that in a common ALS model—mice overexpressing mutant superoxide dismutase 1—the degeneration of motor neurons follows a distinct order. First to suffer are the largest, fast-fatigable motor neurons, responsible for quick movements such as leaping and sprinting, which lose neuromuscular connections before the animal exhibits visible symptoms. At symptom onset, the fast-fatigue-resistant motor neurons—which also generate quick signals, but are slow to tire—are affected. Slow motor neurons, responsible for prolonged muscle activity like standing and walking, are resistant to the effects of mutant SOD1 and some are still connected to muscle at the time of death (Pun et al., 2006).

In the current study, Caroni, along with first author Smita Saxena and Erik Cabuy, used gene expression data to analyze how pathology progresses in vulnerable, fast-fatigable motor neurons versus resistant motor neurons, encompassing fast-fatigue-resistant and slow neurons. To acquire pure populations of both, they took advantage of the fact that parts of the lateral gastrocnemius are exclusively innervated by vulnerable neurons, while the soleus is only served by resistant ones. They injected rhodamine-labeled dextran as a tracer into those muscles, and the tracer entered the neurons by retrograde transfer. When the scientists sacrificed the mice a few days later, the motor neurons of interest lit up under fluorescent light and the researchers were able to dissect them away from other tissue. “The beauty of the experiment is that we are always comparing the exact same 10-11 vulnerable and 10-11 resistant motoneurons in all our experiments,” Caroni wrote in an e-mail to ARF.

The vulnerable motor neurons of SOD1-G93A mice showed upregulation of a variety of stress genes, in comparison with resistant neurons, as early as 12 days of age. The upregulated genes included players in protein ubiquitination, hypoxia, and NRF2-mediated response to oxidative stress. Later on, around their fifth week, the animals’ vulnerable motor neurons upregulated genes for the ER-mediated unfolded protein response such as ATF4. Vulnerable motor neurons, but not resistant ones, also overexpressed the ER stress protein BiP at day 28. “What [they have] done is unprecedented in terms of detailed analysis of the gene changes,” said Christopher Henderson, co-director of the Motor Neuron Center at Columbia University in New York City. “[They have] analyzed changes in a very small subset of the motor neurons.”

Resistant neurons appeared to undergo similar changes but at a much later time; these cells did not upregulate ER stress markers until 25 to 30 days after the vulnerable cells did. “The results suggest that all motor neurons are particularly sensitive to mutant SOD1, but that subtypes of motor neurons differ in their sensitivities to ER stress, and it is that sensitivity that determines the time course of pathology and denervation,” Caroni wrote.

The scientists then used the drug salubrinal, which maintains the activity of a translation initiation factor that promotes production of stress-relieving proteins, to damp down ER stress in the mutant animals, and found it prolonged life by 25-30 days. These animals typically survive approximately 135 days. “It’s not a blockbuster, but it certainly improved the survival of these animals,” said Serge Przedborski, whose laboratory at Columbia University has found evidence linking ER stress and mSOD1 in mice. However, he noted that salubrinal is a “dirty” drug of uncertain specificity, and it may have effects beyond the ER.

These experiments confirmed the suspected relationship between ER stress and ALS. Previous work had shown ER stress in mSOD1 mice (Kikuchi et al., 2006), but only made the correlation, not a causal link, Przedborski said. Although mSOD1 is responsible for approximately 2 percent of human ALS cases, spinal cord ER stress markers have also been reported in people who died from sporadic ALS (Ilieva et al., 2007). ER stress pathways may be promising therapeutic targets, Caroni wrote, as well as potentially provide biomarkers to monitor disease.

“This builds on earlier reports…that not all motor neurons are equal when faced with the disease,” Henderson said. The problem is a general one, he noted; for example, different parts of the substantia nigra exhibit different pathology in Parkinson disease. That makes it important for scientists to select small neural populations to study, Henderson said, as Saxena and colleagues have done.

Caroni suggested that ER stress response might peak when motor neuron toxins such as mSOD1, compounded by the animal’s age, exceed a particular threshold. “Above the threshold cells would trigger more robust stress responses (unfolded protein response), which can get rid of the protein, but are not compatible with a healthy neuron if they are maintained for too long,” he wrote. However, the research simply leads to another question: Why are certain motor neurons more vulnerable to ER stress? Those million dollars are still up for grabs.—Amber Dance


  1. This paper from Saxena et al. is a very interesting, even outstanding paper. Despite that ER stress has been conceptually linked before to ALS development, the experiments performed here offer a novel view on the chronology of facts before denervation and symptom development in relevant experimental models. It should be useful also for other diseases, where ER stress has been also involved.

    Several findings are really surprising: 1) the clear division between resistant motor neurons (RES) and vulnerable ones (VUL); 2) the predictability on development of the disease that the pathogenic scheme described by authors allows; 3) the dissociation between ubiquitination—often considered a pathological hallmark for this disease and other neurodegenerative diseases—and real axonal pathology; 4) the very early changes at a cellular level (as early as postnatal 5 in some markers) that preclude pathological changes; 5) the presence of novel markers of the disease at an immunological level (such as ATF3, PERK, and similar); 6) the distinctive patterns of expression between RES and VUL neurons; 7) the interplay between growth factor treatment (CTNF) and rescue in ER stress terms; and 8) the positive effect of salubrinal in ALS development and the negative effects of crushing schemes

    The findings fit quite well with some of the "usual suspects" theories of ALS, such as the involvement of mitochondria and glia. Mitochondrial impairment (due to unfolded SOD or to other events) would lead to lower ATP levels or to Ca homeostasis dysregulation, which would affect the ER, increasing the unfolded protein response (UPR). Additionally, and pertinent to our case since we linked ER stress to oxidative stress, ER folding capacities are strongly influenced by oxidative milieu and the findings reported here (including participation of hypoxia and NRf2 dependent pathways) agree with the potentially increased oxidative stress in VUL neurons. Most interestingly, many data (as recently reviewed by Cleveland et al. in the last Cell volume—see Lagier-Tourenne et al., 2009) point out the importance of RNA processing in ALS. The hypothetical interplay between alterations in RNA splicing and ER stress in VUL neurons is in agreement with the high structural and energetical requirements of those cells. It seems that long axons and the structural and functional needs that those "near to pathology" cells (VUL motor neurons) exhibit, make them extremely prone to pathology.

    It is also somewhat surprising that ER stress, which may be considered a logical and physiological consequence of UPR, is followed by cellular demise. It would be also be very interesting, as apoptosis seems not involved, to characterize the distal events of ER stress; i.e., what is the link between ER stress and denervation. This is because although the authors define that salubrinal treatment is useful at preclinical stages, it seems that its efficiency would be much lower at a clinical stage.

    It would be nice to confirm these results in human samples. Though it could be difficult to find VUL and RES motor neurons in samples from human disease specimens, but this would also be useful to extend those findings to the more common form of the disease (sporadic ALS, by far, is commoner than the familial form). Further experiments would have to prove, by in vitro transfection with some of the factors described in the papers, that RESistance to disease is acquired by VULnerable neurons (or vice versa, by using RNAi or similar techniques).


    . Rethinking ALS: the FUS about TDP-43. Cell. 2009 Mar 20;136(6):1001-4. PubMed.

  2. This paper from Eckhard Mandelkow's group seems directly related to the question at hand:

    Ebneth A, Godemann R, Stamer K, Illenberger S, Trinczek B, Mandelkow E. Overexpression of tau protein inhibits kinesin-dependent trafficking of vesicles, mitochondria, and endoplasmic reticulum: implications for Alzheimer's disease. J Cell Biol. 1998 Nov 2;143(3):777-94. Abstract

  3. While appreciating the impressive FALS study by Sexena et al., I could not help frowning on the comment by P.F. Jennings: Why implicate protein tau and axonal transport? Tau-4R transgenic mice develop axonopathy leading to Wallerian degeneration and muscle wasting, but not premature death (Spittaels et al., 1999), as opposed to tau-P301L mice that develop tauopathy and die prematurely (Terwel et al., 2005).

    Both patho-phenotypes are affected by co-expression of GSK3, albeit quite differently: rescue and aggravation of tauopathy, respectively (Spittaels et al., 2000; Terwel et al., 2008).

    Our simplest explanation: excess tau-4R (but not tau-P301L) occupies microtubular binding sites, preventing motor proteins to walk and transport "stuff." GSK3 phosphorylates tau and releases it from the MT to allow transport again, but thereby causes tauopathy at the expense of axonopathy. Why motor neurons are most sensitive to tau-induced degeneration remains an open question. Caroni and co-workers provide possible indications.

    But is protein tau involved in ALS?


    . Prominent axonopathy in the brain and spinal cord of transgenic mice overexpressing four-repeat human tau protein. Am J Pathol. 1999 Dec;155(6):2153-65. PubMed.

    . Glycogen synthase kinase-3beta phosphorylates protein tau and rescues the axonopathy in the central nervous system of human four-repeat tau transgenic mice. J Biol Chem. 2000 Dec 29;275(52):41340-9. PubMed.

    . Changed conformation of mutant Tau-P301L underlies the moribund tauopathy, absent in progressive, nonlethal axonopathy of Tau-4R/2N transgenic mice. J Biol Chem. 2005 Feb 4;280(5):3963-73. PubMed.

    . Amyloid activates GSK-3beta to aggravate neuronal tauopathy in bigenic mice. Am J Pathol. 2008 Mar;172(3):786-98. PubMed.

    View all comments by Fred Van Leuven

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Selective vulnerability and pruning of phasic motoneuron axons in motoneuron disease alleviated by CNTF. Nat Neurosci. 2006 Mar;9(3):408-19. PubMed.
  2. . Spinal cord endoplasmic reticulum stress associated with a microsomal accumulation of mutant superoxide dismutase-1 in an ALS model. Proc Natl Acad Sci U S A. 2006 Apr 11;103(15):6025-30. PubMed.
  3. . Oxidative and endoplasmic reticulum stress interplay in sporadic amyotrophic lateral sclerosis. Brain. 2007 Dec;130(Pt 12):3111-23. PubMed.

Further Reading


  1. . ER stress and unfolded protein response in amyotrophic lateral sclerosis. Mol Neurobiol. 2009 Apr;39(2):81-9. PubMed.
  2. . Increased ER stress during motor neuron degeneration in a transgenic mouse model of amyotrophic lateral sclerosis. Neurol Res. 2007 Dec;29(8):767-71. PubMed.
  3. . Induction of the unfolded protein response in familial amyotrophic lateral sclerosis and association of protein-disulfide isomerase with superoxide dismutase 1. J Biol Chem. 2006 Oct 6;281(40):30152-65. PubMed. RETRACTED

Primary Papers

  1. . A role for motoneuron subtype-selective ER stress in disease manifestations of FALS mice. Nat Neurosci. 2009 May;12(5):627-36. PubMed.