Malfunctioning mitochondria accumulate in the Alzheimer’s disease brain. Could they be purged? In the February 11 Nature Neuroscience, researchers led by Vilhelm Bohr at the National Institute on Aging, Baltimore, argue that these defective organelles play a key role in disease progression. In both animal and cellular models, revving up their disposal lessened the hallmark pathologies of AD: amyloid-β plaques and phosphorylated tau. In addition, learning and memory behavior bounced back to normal in treated worms and mice. “We believe mitophagy is a central process early in Alzheimer’s disease, and could be a therapeutic target,” Bohr told Alzforum.

  • Mitophagy is compromised in AD brain.
  • Stimulating it reduces Aβ/tau pathology and improves memory in animal models.
  • Phosphorylated tau prevents mitophagy by sequestering parkin.

“This is a remarkable and outstanding paper,” Flint Beal at Weill Cornell Medical College, New York, wrote to Alzforum. “It ties tau phosphorylation and amyloid pathology to mitochondrial dysfunction and defective mitophagy. Aging, the most important risk factor for AD, is also linked to reductions in mitophagy.”

The relationship between mitochondrial damage and AD pathology seems to run both ways, with each worsening the other (Nov 2009 news). Recently, researchers led by Jürgen Götz at the University of Queensland in Brisbane, Australia, elucidated one mechanism by which tau pathology contributes to mitochondrial mismanagement. They report in the February 1 EMBO Journal that cytoplasmic tau binds to the ubiquitin ligase parkin, preventing parkin from reaching damaged mitochondria and triggering their disposal via mitophagy.

“Together with other studies, these papers show clear evidence of a nexus between mitochondrial dysfunction and protein aggregation,” said Russell Swerdlow at the University of Kansas in Kansas City. It remains unclear which phenomenon occurs first. Swerdlow speculated the order might vary in different forms of the disease, with familial AD more likely triggered by protein aggregation, and sporadic AD by aging, dysfunctional mitochondria (Aug 2013 news). 

Mitochondria Aid Phagocytosis. In AD mice treated with a mitophagy enhancer (center and right), more microglia (red) feast on plaques (green) than in untreated AD mice (left). [Courtesy of Fang et al., Nature Neuroscience.]

Previous research has established that neurons in AD brain contain fewer healthy, intact mitochondria and more that are broken, oversized, and misplaced (May 2001 news; Jul 2009 news). Cells maintain their pool of mitochondria mainly via lysosomal degradation of damaged organelles. The kinase PINK1 tags defective mitochondria and recruits parkin, which in turn ubiquitinates mitochondrial proteins, flagging the organelle for disposal in lysosomes. Mutations in either of these genes cause Parkinson’s disease, but have not been linked to AD (Apr 2004 newsApr 2014 news; Sep 2015 news). 

Bohr and colleagues wondered what role mitophagy might play in AD progression. Joint first authors Evandro Fang, Yujun Hou, and Konstantinos Palikaras measured the number of mitochondria associated with lysosomes in postmortem AD hippocampus, finding about half as much mitophagy as in control brain. They found the same thing in human neurons generated from iPS cells from AD patients. One of these lines carried an APP mutation, and a second had two copies of ApoE4. In both, proteins that initiate mitophagy, TBK1 and ULK1, were half as active as in control neurons. TBK1 variants cause familial forms of amyotrophic lateral sclerosis and frontotemporal dementia (Feb 2015 news; Mar 2015 news). 

Would boosting mitophagy ameliorate AD? The authors first screened for compounds that enhanced mitophagy in the roundworm Caenorhabditis elegans. They identified an antibiotic, actinonin, and a plant compound, urolithin A. They fed urolithin A to a worm that expresses human Aβ42, and found it lowered levels of the peptide throughout the body. Notably, these worms treated with either urolithin A or actinonin learned to avoid a noxious chemical as quickly as wild-types did. Treatment likewise restored normal memory to a worm model of tau pathology (Fatouros et al., 2012). In both models, the improvement depended on mitophagy, as it did not occur in PINK1 mutants.

Turning to mice, the authors fed urolithin A or actinonin to APP/PS1 animals for two months. Mitophagy in hippocampal neurons ramped up to normal, while the number of damaged mitochondria fell to control levels. Amyloid plaque burden in the hippocampus fell by two-thirds, and treated animals performed as well as wild-types in the Morris water and Y mazes.

How does mitophagy lower amyloid? Perhaps via the innate immune system. In treated animals, the authors detected more activated microglia surrounding and engulfing plaques than in controls (see image above). Microglia in these mice were more phagocytic than those in untreated mice, as measured by cell shape and protein expression. In addition, these microglia boasted healthy mitochondria similar to those in wild-type mice, while untreated APP/PS1 microglia accumulated threefold more damaged mitochondria than controls did. Possibly, phagocytosis becomes impaired in untreated APP/PS1 mice because this process requires a great deal of cellular energy, and thus needs healthy mitochondria, the authors speculated.

Swerdlow suggested another possibility. He pointed to recent studies that indicate aggregated Aβ and tau can end up inside mitochondria, where they are either cleared by mitochondria themselves, or eliminated when mitophagy grinds up the organelles (Sorrentino et al., 2017; Du et al., 2017). Both Sorrentino et al. and Du et al. found that enhancing mitophagy lowered amyloid aggregation, in agreement with Bohr’s data. “Mitochondria are like trash bags, and mitophagy takes out the trash,” Swerdlow suggested.

Moreover, Shirley ShiDu Yan and colleagues at the University of Kansas in Lawrence reported that stimulating mitophagy with PINK1 improved synaptic plasticity and memory in AD mice. “The present study from Fang et al. supports our discovery,” Yan wrote to Alzforum.

What about tau? Because APP/PS1 mice develop little tau pathology, Bohr and colleagues turned to 3xTg AD mice, which carry a tau mutation on top of APP and PS1 mutations, and form tangles. Treating them with urolithin A for one month inhibited tau phosphorylation and restored memory in the Y maze and object recognition tests. Intriguingly, previous research from Eva Mandelkow at the German Center for Neurodegenerative Diseases in Bonn found that a tau kinase, MARK2, also regulates PINK1 and mitochondrial transport (Matenia et al., 2012). Bohr noted that in AD, tau pathology associates more closely with cognitive decline than amyloid does, suggesting that targeting tau pathology by boosting mitophagy could help patients.

Bohr and colleagues are testing this in a clinical trial run by Steen Hasselbalch at Copenhagen University Hospital, Denmark. AD patients will take nicotinamide riboside, a Vitamin B3 variant that boosts nicotinamide adenine dinucleotide (NAD+). NAD+ precursors are known to stimulate mitophagy. Bohr said an advantage of this dietary supplement is that it has few side effects. NAD+ precursors are beginning to be evaluated for neurodegenerative and other conditions (Feb 2019 news).

Besides testing nicotinamide riboside in people, Bohr plans to further dissect how tau and mitochondria interact in worms. Prior studies have blamed abnormal tau for instigating mitochondrial dysfunction (Manczak and Reddy, 2012; Duboff et al., 2013; Eckert et al., 2014). 

Götz and colleagues added to this literature by showing that cytosolic tau blocks mitophagy. First author Nadia Cummins found that overexpressing either human wild-type or P301L tau in mouse neuroblastoma cells prevented initiation of mitophagy after mitochondria had sustained damage. She traced the cause to a lack of parkin recruitment to these organelles and determined by co-immunoprecipitation that both types of tau directly bound parkin. Testing tau fragments, she found that the amino-terminal end of tau was responsible. Notably, tau did not interact with parkin that was already attached to mitochondria. In short, tau appears to sequester parkin in the cytosol, preventing it from reaching these organelles, the authors concluded. In AD, hyperphosphorylated tau dissociates from microtubules and builds up in cytoplasm.

Bohr noted that the findings dovetail with his. The interaction of tau and parkin strengthens the evidence that mitochondrial dysfunction may crop up early in AD, he said.—Madolyn Bowman Rogers


  1. This is an outstanding study. It not only adds significant evidence supporting a critical role of mitochondrial dysfunction in the pathogenesis of Alzheimer’s disease, but also specifically suggests mitophagy as a promising therapeutic target for AD. 

    Different from the previous neuron-centric study of mitochondrial (dys)function in AD, it is most interesting and significant that this study demonstrated that mitophagy is also impaired in the microglia in AD mouse brain, and that mitophagy stimulation enhances the phagocytic efficiency of microglia to clear amyloid plaques along with mitigation of NLRP3/caspase-1-dependent neuroinflammation.

    While the authors suggested that enhanced phagocytic efficiency in microglia is likely due to the restored supply of required energy by improved mitochondrial homeostasis, detailed mechanisms warrant further investigation. Overall, the study demonstrated the neuroprotective and restorative potential of mitophagy enhancement in AD.

    However, since mitochondrial mass is reduced in pyramidal neurons in AD (Hirai et al., 2001), and current evidence supports the notion of impaired mitochondrial biogenesis in AD (Qin et al., 2009; Sheng et al., 2012), the pursuit of mitophagy enhancers alone as a therapeutic may be limited.


    . Mitochondrial abnormalities in Alzheimer's disease. J Neurosci. 2001 May 1;21(9):3017-23. PubMed.

    . PGC-1alpha expression decreases in the Alzheimer disease brain as a function of dementia. Arch Neurol. 2009 Mar;66(3):352-61. PubMed.

    . Impaired mitochondrial biogenesis contributes to mitochondrial dysfunction in Alzheimer's disease. J Neurochem. 2012 Feb;120(3):419-29. PubMed.

  2. This paper from Fang et al. represents an impressive amount of work examining the interplay between dysfunctional mitophagy, inflammation, and AD pathologies in multiple disease models. These data add to multiple studies implicating dysfunctional mitophagy in PD (reviewed in (reviewed in McWilliams and Muqit, 2017), a growing number of papers demonstrating a number of FTD/ALS genes are involved in the mitophagy process, such as TBK1, SQSTM, OPTN and VCP (Karbowski and Youle, 2011; Heo et al., 2015; Kim et al., 2013), and a few recent studies suggesting a role for mitophagy in AD (Sorrentino, et al., 2017; Du et al., 2017; Checler et al., 2017; Martín-Maestro et al., 2016). Thus, mitophagy is emerging as a crucial mechanism across multiple neurodegenerative diseases.

    Neurons rely quasi-exclusively on mitochondria for ATP production, thus the selective clearance of damaged mitochondria by mitophagy is critical to maintain the bioenergetic integrity of the cell and prevent the accumulation of damaged mitochondria and toxic reactive oxygen species. The reliance on mitochondria also means mitophagy is a rare event, and most studies require overexpression and induction of mitophagy using mitochondrial uncoupling agents that artificially induce mitochondrial depolarization. A strength of this study is the cross-species analysis of basal mitophagy in multiple in vitro and in vivo models. 

    The work raises some interesting questions. There are multiple pathways coordinating mitochondrial integrity, and it will be important to identify the specific mitophagy pathway(s) that is/are dysfunctional in AD. The PINK1-dependent pathway is the most well-characterized mechanism for the induction and execution of mitophagy, however PINK1 knockout mice do not have defective mitophagy (McWilliams et al., 2018), and so it is intriguing that such widespread alterations in basal mitophagy are observed in AD models. The PINK1 pathway could be explored more extensively in these models by using alternative methods, such as mtQC mitophagy mice reporters (McWilliams et al., 2018), and by examining (for example) the levels of phosphorylated ubiquitin at serine 65, which is a substrate of PINK1 precluding the recruitment of Parkin to damaged mitochondria. It would also be important to study mitochondrial biogenesis in more detail, especially as PGC1a levels seem to be reduced in their AD models. Finally, it would be interesting to know whether the defects observed in hippocampus also occur in other brain regions in their disease models.

    A number of studies have suggested an interplay of neuronal and glial cells to orchestrate mitophagy in a transcellular manner (Davis et al., 2014), yet this paper suggests a new role for mitophagy in enhancing a dual role for microglia: the phagocytic clearance of Aβ plaques and the release of pro-inflammatory cytokines. The role of the connected network of neurons and glial cells will need to be studied in more detail in AD and in other neurodegenerative conditions.

    Together, these data add to a growing body of evidence highlighting the importance of mitophagy across multiple diseases. Future mechanistic studies will highlight the potential of this pathway for therapeutic targeting.


    . PINK1 and Parkin: emerging themes in mitochondrial homeostasis. Curr Opin Cell Biol. 2017 Apr;45:83-91. Epub 2017 Apr 22 PubMed.

    . Regulating mitochondrial outer membrane proteins by ubiquitination and proteasomal degradation. Curr Opin Cell Biol. 2011 Aug;23(4):476-82. Epub 2011 Jun 24 PubMed.

    . The PINK1-PARKIN Mitochondrial Ubiquitylation Pathway Drives a Program of OPTN/NDP52 Recruitment and TBK1 Activation to Promote Mitophagy. Mol Cell. 2015 Oct 1;60(1):7-20. Epub 2015 Sep 10 PubMed.

    . VCP is essential for mitochondrial quality control by PINK1/Parkin and this function is impaired by VCP mutations. Neuron. 2013 Apr 10;78(1):65-80. PubMed.

    . Enhancing mitochondrial proteostasis reduces amyloid-β proteotoxicity. Nature. 2017 Dec 14;552(7684):187-193. Epub 2017 Dec 6 PubMed.

    . PINK1 signalling rescues amyloid pathology and mitochondrial dysfunction in Alzheimer's disease. Brain. 2017 Dec 1;140(12):3233-3251. PubMed.

    . Presenilins at the crossroad of a functional interplay between PARK2/PARKIN and PINK1 to control mitophagy: Implication for neurodegenerative diseases. Autophagy. 2017;13(11):2004-2005. Epub 2017 Sep 21 PubMed.

    . PARK2 enhancement is able to compensate mitophagy alterations found in sporadic Alzheimer's disease. Hum Mol Genet. 2016 Feb 15;25(4):792-806. Epub 2015 Dec 31 PubMed.

    . Basal Mitophagy Occurs Independently of PINK1 in Mouse Tissues of High Metabolic Demand. Cell Metab. 2018 Feb 6;27(2):439-449.e5. Epub 2018 Jan 11 PubMed.

    . Transcellular degradation of axonal mitochondria. Proc Natl Acad Sci U S A. 2014 Jul 1;111(26):9633-8. Epub 2014 Jun 16 PubMed.

  3. We agree with Xiongwei Zhu and Wenzhang Wang on significant evidence supporting a critical role of mitochondrial dysfunction in Alzheimer's disease. We have pursued this premise to conduct IRB-approved placebo controlled clinical trials in human LOAD.

    Our clinical experience and published findings on transcranial and intraocular infrared photobiomodulation suggests that increasing ATP activation through stimulating Cytochrome C Oxidase in the mitochondria results in improved cognitive and behavioral functioning in adults with dementia and Parkinson's disease. Recent AD animal model data showed this type of noninvasive stimulation improved praxis memory, motor behavior, reduced Aβ42 and p-tau, confers neuroprotection, and significantly increased VEGF production. Expanded human clinical trials are now underway at Baylor Research Institute (Temple, Texas) and Quietmind Foundation (Elkins Park, Pennsylvania) to evaluate the impact of self-administered, twice-daily, six-minute, transcranial and intraocular exposure to near infrared photobiomodulation.

    Jason Huang also contributed to this comment.


    . Photobiomodulation with Near Infrared Light Helmet in a Pilot, Placebo Controlled Clinical Trial in Dementia Patients Testing Memory and Cognition. J Neurol Neurosci. 2017;8(1) Epub 2017 Feb 28 PubMed.

    . Non-invasive infra-red therapy (1072 nm) reduces β-amyloid protein levels in the brain of an Alzheimer's disease mouse model, TASTPM. J Photochem Photobiol B. 2013 Jun 5;123:13-22. Epub 2013 Mar 22 PubMed.

    . Low-Intensity Light Therapy (1068nm) Protects CAD Neuroblastoma Cells from β-Amyloid-Mediated Cell Death. Biol Med 2014

    . Enhancement of cutaneous immune response to bacterial infection after low-level light therapy with 1072 nm infrared light: a preliminary study. J Photochem Photobiol B. 2011 Dec 2;105(3):175-82. Epub 2011 Sep 6 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. New Triple Transgenic Shows Mitochondrial Damage by Tau, Aβ
  2. Studies Suggest Mitochondria Changes Precede Aging, Alzheimer’s
  3. Mitochondrial Damage in Alzheimer's Disease
  4. Mitochondrial Break-up: Alzheimer’s Alters Fusion, Fission
  5. Pink Mutations Link Parkinson’s Disease to Mitochondria
  6. Novel Ubiquitin Modification Ties Two Risk Genes for Parkinsonism
  7. PINK1 Can Act Alone to Destroy Mitochondria, But Parkin Helps
  8. TANK-Binding Kinase 1 Rumbles in as New ALS Gene
  9. Second Study Salutes TANK-Binding Kinase 1 as ALS Gene
  10. In Small Trial, EH301 Appears to Halt Progression of ALS

Research Models Citations

  1. APPswe/PSEN1dE9
  2. 3xTg

Paper Citations

  1. . Inhibition of tau aggregation in a novel Caenorhabditis elegans model of tauopathy mitigates proteotoxicity. Hum Mol Genet. 2012 Aug 15;21(16):3587-603. PubMed.
  2. . Enhancing mitochondrial proteostasis reduces amyloid-β proteotoxicity. Nature. 2017 Dec 14;552(7684):187-193. Epub 2017 Dec 6 PubMed.
  3. . PINK1 signalling rescues amyloid pathology and mitochondrial dysfunction in Alzheimer's disease. Brain. 2017 Dec 1;140(12):3233-3251. PubMed.
  4. . Microtubule Affinity-regulating Kinase 2 (MARK2) Turns on Phosphatase and Tensin Homolog (PTEN)-induced Kinase 1 (PINK1) at Thr-313, a Mutation Site in Parkinson Disease: EFFECTS ON MITOCHONDRIAL TRANSPORT. J Biol Chem. 2012 Mar 9;287(11):8174-86. PubMed.
  5. . Abnormal interaction between the mitochondrial fission protein Drp1 and hyperphosphorylated tau in Alzheimer's disease neurons: implications for mitochondrial dysfunction and neuronal damage. Hum Mol Genet. 2012 Jun 1;21(11):2538-47. PubMed.
  6. . Why size matters - balancing mitochondrial dynamics in Alzheimer's disease. Trends Neurosci. 2013 Jun;36(6):325-35. PubMed.
  7. . March separate, strike together--role of phosphorylated TAU in mitochondrial dysfunction in Alzheimer's disease. Biochim Biophys Acta. 2014 Aug;1842(8):1258-66. Epub 2013 Sep 17 PubMed.

Further Reading

Primary Papers

  1. . Mitophagy inhibits amyloid-β and tau pathology and reverses cognitive deficits in models of Alzheimer's disease. Nat Neurosci. 2019 Mar;22(3):401-412. Epub 2019 Feb 11 PubMed.
  2. . Disease-associated tau impairs mitophagy by inhibiting Parkin translocation to mitochondria. EMBO J. 2019 Feb 1;38(3) Epub 2018 Dec 11 PubMed.