The amyotrophic lateral sclerosis protein TDP-43 may exert its toxic effect by uncoupling mitochondria and endoplasmic reticulum, according to a paper in the June 3 Nature Communications. Where these two organelles meet, they form structures called mitochondria-associated ER membranes, aka MAMs, and TDP-43 can disrupt that interaction, report scientists from King’s College London. Senior author Christopher Miller and colleagues discovered this as they were following up on earlier results that another ALS protein, vesicle-associated membrane protein-associated protein-B (VAPB), reaches out from the ER to grab mitochondria. The authors posit that MAM alterations could explain many cellular ALS changes, including calcium dyshomeostasis and mitochondrial maladies.

When ER and mitochondrial membranes come close together, they build transient bridges—the MAMs. Researchers know little about this association, but MAMs have at least two major functions, said Eric Schon of Columbia University Medical Center in New York, who has linked MAMs to Alzheimer’s disease. For one, they manage calcium levels as the ions travel from the ER to the mitochondria. For another, they host synthesis of both cholesterol and phospholipids. Miller’s group previously reported that VAPB in the ER shakes hands with a mitochondrial protein that goes by the moniker protein tyrosine phosphatase-interacting protein 51 (PTPIP51) (De Vos et al., 2012).

Mutations in VAPB cause ALS, and the authors showed in their earlier paper that one such variant binds more tightly to PTPIP51 than wild-type. Like VAPB, TDP-43 contributes to ALS, and it renders mitochondria structurally abnormal and interferes with ER signaling pathways (see Aug 2010 news storyShan et al., 2010). Might TDP-43 act via MAMs, as well?

To find out, co-first authors Radu Stoica and Kurt De Vos, who has moved to the Sheffield Institute for Translational Neuroscience at the University of Sheffield in the United Kingdom, expressed wild-type and mutant TDP-43 in mouse motor neuron NSC-34 cultures. Normally, about 12 percent of outer mitochondrial membranes butt up against the ER in MAMs. Both wild-type and mutant TDP-43 reduced this fraction by a third or more. The same occurred in a mouse model expressing wild-type human TDP-43; the percentage of mitochondrial membranes in the abutting MAMs was about half that in non-transgenic littermates.

Using co-immunoprecipitation, the authors showed that TDP-43 expression reduced binding of VAPB to PTPIP51 in the NSC-34 cells and spinal cords of the model mice. However, TDP-43 itself did not co-immunoprecipitate with VAPB or PTPIP51. The authors concluded that TDP-43 must control their interaction remotely. Researchers recently reported that TDP-43 activates glycogen synthase kinase-3β (GSK-3β) (Ambegoakar and Jackson, 2011), and the Miller group hypothesized that this enzyme could be part of the link. Treating the cells with GSK-3β inhibitors boosted the amount of co-immunoprecipitated PTPIP51 and VAPB, whereas overexpressing GSK-3β diminished this measure. “Thus, TDP-43 activates GSK-3β and GSK-3β regulates the VAPB-PTPIP51 interaction,” the authors wrote. They claim that pathogenic TDP-43 may damage the MAMs, though the steps from TDP-43 to GSK-3β to MAM disruption are not yet understood. The study authors were unavailable to comment for this news article. Other researchers have found that inhibiting GSK-3 protects mouse neurons from the negative effects of the ALS gene superoxide dismutase 1 (see Apr 2013 news story).

Exactly how TDP-43 influences ALS, AD, and frontotemporal dementia is a research hotbed. The protein has been spotted co-localizing with the organelle (see Aug 2010 news storyWang et al., 2013). “However, if TDP-43 regulates the association of mitochondria and ER via activation of GSK-3β, [then] TDP-43 would not necessarily need to be physically associated with these organelles to regulate this pathway,” Brian Freibaum of St. Jude Children’s Research Hospital in Memphis, Tennessee, wrote in an email to Alzforum.

The Calcium Connection
Because MAMs regulate calcium homeostasis, De Vos and Stoica hypothesized that TDP-43 might alter calcium levels in mitochondria and the cytosol. They used a drug to trigger calcium release from the ER in human embryonic kidney HEK293 cells, and tested if a TDP-43 transgene influenced that response. Cultures expressing TDP-43 contained slightly more cytosolic and less mitochondrial calcium than control cells. In fact, Miller and colleagues suggest that disrupting MAMs could lead to a number of problems seen in ALS neurons, including altered calcium homeostasis as well as mitochondrial damage, decreased ATP synthesis, and inappropriate activation of the unfolded protein response in the ER.

“It is unclear at this time how significantly this pathway may contribute to disease pathogenesis, but I think it is definitely worthy of future exploration,” Freibaum wrote. “I would be curious to see if there is evidence of decreased mitochondria-ER association in patient cells.”

ALS might not be the only neurodegenerative condition where MAMs go awry. In the case of Alzheimer’s, Schon’s group reported that presenilin 1 and 2 are found in MAMs, and γ-secretase acts there as well (Area-Gomez et al., 2009). In addition, they discovered that a MAM function—cholesteryl ester and phospholipid synthesis—is overactive in fibroblasts from people with Alzheimer’s (Area-Gomez et al., 2012). “We think that ER and mitochondria communication lies at the heart of AD,” said Schon (reviewed in Schon and Area-Gomez et al., 2012), though few other scientists share this view. Signs point to MAMs in Parkinson’s, too, with recent data that α-synuclein localizes to these points of organelle contact (Guardia-Laguarta et al., 2014). And there are hints MAMs may be involved in frontotemporal dementia, spinal muscular atrophy (see Sep 2010 news story), and Charcot-Marie-Tooth disease (Züchner et al., 2004), Schon said. “I think that MAM is unexplored territory, and its importance is vastly underestimated."—Amber Dance


  1. The finding of a VAPB-PTPIP51 interaction supporting the ER-mitochondria association is interesting and convincing. The effects of TDP-43 overexpression suggest a potential role for TDP-43 in modulating this ER-mitochondrial association and intracellular calcium signaling. However, from results in the paper, the effects of TDP43 appear to be indirect and mediated via GSK-3β. It will be important to determine if impaired ER-mitochondrial association is a major outcome of TDP-43/GSK-3β activation or if it is one of many downstream pathways leading to pathology in TDP-43 ALS. Nevertheless, the potential link between TDP-43 and intracellular calcium signaling is very interesting.

    View all comments by Ilya Bezprozvanny

Make a Comment

To make a comment you must login or register.


News Citations

  1. Paper Alert: Malformed Mitochondria in the Latest TDP-43 Mouse
  2. Stem Cell Screen Points to ALS Disease Target
  3. Latest TDP-43 Mouse Unites ALS and SMA Pathways

Paper Citations

  1. . VAPB interacts with the mitochondrial protein PTPIP51 to regulate calcium homeostasis. Hum Mol Genet. 2011 Dec 13; PubMed.
  2. . Altered distributions of Gemini of coiled bodies and mitochondria in motor neurons of TDP-43 transgenic mice. Proc Natl Acad Sci U S A. 2010 Sep 14;107(37):16325-30. Epub 2010 Aug 24 PubMed.
  3. . Functional genomic screen and network analysis reveal novel modifiers of tauopathy dissociated from tau phosphorylation. Hum Mol Genet. 2011 Dec 15;20(24):4947-77. Epub 2011 Sep 23 PubMed.
  4. . The ALS disease-associated mutant TDP-43 impairs mitochondrial dynamics and function in motor neurons. Hum Mol Genet. 2013 Dec 1;22(23):4706-19. Epub 2013 Jul 4 PubMed.
  5. . Presenilins are enriched in endoplasmic reticulum membranes associated with mitochondria. Am J Pathol. 2009 Nov;175(5):1810-6. PubMed.
  6. . Upregulated function of mitochondria-associated ER membranes in Alzheimer disease. EMBO J. 2012 Nov 5;31(21):4106-23. PubMed.
  7. . Mitochondria-associated ER membranes in Alzheimer disease. Mol Cell Neurosci. 2012 Aug 24; PubMed.
  8. . α-Synuclein is localized to mitochondria-associated ER membranes. J Neurosci. 2014 Jan 1;34(1):249-59. PubMed.
  9. . Mutations in the mitochondrial GTPase mitofusin 2 cause Charcot-Marie-Tooth neuropathy type 2A. Nat Genet. 2004 May;36(5):449-51. Epub 2004 Apr 4 PubMed.

Further Reading


  1. . Close encounter: mitochondria, endoplasmic reticulum and Alzheimer's disease. EMBO J. 2012 Nov 5;31(21):4095-7. PubMed.
  2. . α-Synuclein Controls Mitochondrial Calcium Homeostasis by Enhancing Endoplasmic Reticulum-Mitochondria Interactions. J Biol Chem. 2012 May 25;287(22):17914-29. PubMed.
  3. . Presenilin 2 modulates endoplasmic reticulum (ER)-mitochondria interactions and Ca2+ cross-talk. Proc Natl Acad Sci U S A. 2011 Feb 15;108(7):2777-82. PubMed.
  4. . Modulation of the endoplasmic reticulum-mitochondria interface in Alzheimer's disease and related models. Proc Natl Acad Sci U S A. 2013 May 7;110(19):7916-21. PubMed.
  5. . Familial Alzheimer's disease-associated presenilin-1 alters cerebellar activity and calcium homeostasis. J Clin Invest. 2014 Apr 1;124(4):1552-67. Epub 2014 Feb 24 PubMed.
  6. . Molecular pathways of motor neuron injury in amyotrophic lateral sclerosis. Nat Rev Neurol. 2011 Nov;7(11):616-30. PubMed.
  7. . The GSK3 hypothesis of Alzheimer's disease. J Neurochem. 2008 Mar;104(6):1433-9. PubMed.

Primary Papers

  1. . ER-mitochondria associations are regulated by the VAPB-PTPIP51 interaction and are disrupted by ALS/FTD-associated TDP-43. Nat Commun. 2014 Jun 3;5:3996. PubMed.