The brain's metabolism starts to wane decades before Alzheimer's symptoms. Why the energy deficit? Scientists led by Andrés Norambuena and George Bloom at the University of Virginia, Charlottesville, blame failure of a specific type of mitochondrial respiration called nutrient-induced mitochondrial activation. NiMA, they write in a bioRxiv preprint posted on February 4, weakens in mouse models of amyloidosis a year before plaques appear. They believe Aβ oligomers stifle mitochondria by activating GSK3β, a kinase linked to AD. “Disruption of NiMA in the brain could be a very early event in AD pathogenesis in humans,” the authors wrote.

Previously, Norambuena and Bloom described NiMA, a signaling pathway between lysosomes and mitochondria, in cultured human neurons, whereby insulin or a mixture of arginine and leucine activates the master regulatory kinase mTOR on the lysosome surface. This stirs mitochondria to ramp up oxidative phosphorylation; however, soluble Aβ oligomers blocked NiMA (Norambuena et al., 2018).

To look for this in vivo, first author Norambuena studied APPSAA knock-in mice, which express one copy of wild-type mouse amyloid precursor protein and another with a humanized Aβ sequence and three mutations, Swedish, Arctic, and Austrian, to speed plaque formation. These mice have Aβ oligomers in the brain and CSF by 4 months of age, and develop plaques at 16 months. In 2-, 4-, and 6-month-old mice, Norambuena analyzed mitochondrial metabolism and cellular oxygen consumption through a cranial window using two-photon fluorescence lifetime imaging and photoacoustic microscopy, respectively. Both methods measure changes in the intrinsic fluorescence of molecules: the former of two mitochondrial cofactors, the latter of oxygenated hemoglobin.

To stimulate NiMA, Norambuena dripped a solution of arginine and leucine onto the mouse cortex through the cranial window. Thirty minutes later, mitochondrial respiration increased in wild-type mice. The cocktail also jolted mitochondria in 2-month-old APPSAA mice but not in 4- or 6-month-olds (image below). “The simplest explanation for this observation is that the single mutated APP gene drove Aβ production to levels sufficient to block NiMA,” the authors concluded.

Dwindling NiMA. Through a cranial window, two-photon microscopy captured mitochondrial activity up to 80 minutes after stimulation. It rose (blue puncta) in 2-, but not 4- or 6-month-old APPSAA mice. Mitochondrial respiration was measured by the fraction of NAD(P)H bound to mitochondrial respiratory chain enzymes, a proxy for the state of oxidative phosphorylation (a2 percent). [Courtesy of Norambuena et al., bioRxiv, 2024.]

To find molecules that regulate NiMA, Norambuena screened for protein kinases, turning up GSK3β. It hyperphosphorylates tau, promotes Aβ production, and lowers mitochondrial metabolism. A GSK3β inhibitor reportedly increased mitochondrial respiration in mouse hippocampus (Nov 2016 news; Martin et al., 2018).

To see if this might rescue NiMA, Norambuena dripped the GSK3β inhibitor TWS119 through the cranial window. Mitochondrial respiration in the brains of 4-month-old wild-type mice jumped 30 percent, but only 10 percent in 4-month-old APPSAA mice. The GSK3β inhibitor tideglusib did not stimulate NiMA in wild-type mice.

The authors think trouble for NiMA results from an interplay between Aβ oligomers, GSK3β, and the metabolic regulator mTORC1. The latter controls a broad range of cellular processes ranging from protein synthesis to autophagy, and mitochondrial dynamics such as mitophagy. Previously, Bloom's group had reported that Aβ oligomers reduce lysosomal mTORC1 activation, and therefore NiMA, by inhibiting insulin signaling, and that they activate mTORC1 on the plasma membrane to nudge neurons into the cell cycle (Norambuena et al., 2022; Norambuena et al., 2017).

“It is reasonable to speculate that during AD progression, the combination of Aβ production and insulin resistance leads to an imbalance in mTORC1 activity favoring its ectopic and toxic activity at the plasma membrane over its normal physiological functions on lysosomes,” the authors wrote (image below).

Modeling NiMA. In wild-type mice (left), insulin inhibits GSK3β, allowing amino acids to stimulate lysosomal mTORC1 and mitochondrial oxidative phosphorylation (NiMA). In AD (middle), Aβ oligomers release this brake on lysosomal mTORC1, decreasing NiMA, and activate mTORC1 ectopically on the plasma membrane, triggering cell cycle re-entry (CCR). Blocking GSK3β with a drug may restore NiMA (right). [Courtesy of Norambuena et al., bioRxiv, 2024.]

Angelika Harbauer of the Max Planck Institute of Neurobiology in Munich remains unconvinced. “I believe it is too early to claim a causal relationship between APP mutations, GSK3β, mTORC1, and mitochondrial dysfunction in neurons in AD, given the challenges in translating in vitro mechanistic analyses in isolated cell types to the complex cellular environments and interactions in vivo,” she wrote (comment below).—Chelsea Weidman Burke

Comments

  1. Defects in proteostasis and the accumulation of toxic misfolded proteins, as well as mitochondrial dysfunction and a neuroenergetic crisis are intimately linked to each other in the pathogenesis of neurodegenerative diseases. This preprint tries to shed light on the disease progression in Alzheimer’s disease by analyzing mitochondrial function in vivo by NAD/NAD(P)H lifetime imaging and oxygen measurements in the cortices of young APPSAA mice. While no steady-state differences are evident, the mTORC1-driven activation of respiration in response to branched amino acids is impaired in the AD model mice.

    While this finding is intriguing, by nature of the minimally invasive, marker-less imaging techniques, the authors cannot distinguish the contribution of different cell types to this response. The mTORC1 signaling cascade is functional in neurons and certainly regulates aspects of mitochondrial function, but the downstream effects are not always similar to other cell types, e.g., there is no induction of autophagy in response to mTORC1 inhibition in neurons (Maday and Holzbaur, 2016). I believe it is too early to claim a causal relationship between APP mutations/GSK3β/mTORC1/mitochondrial dysfunction in neurons in AD, given the challenges in translating in vitro mechanistical analyses in isolated cell types to the complex cellular environments and interactions in vivo.

    References:

    . Compartment-Specific Regulation of Autophagy in Primary Neurons. J Neurosci. 2016 Jun 1;36(22):5933-45. PubMed.

  2. I’d like to commend the authors on the technical sophistication of the work, for considering the possibility that mitochondria and energy-related metabolism are relevant to AD, and for offering a generally unique and innovative hypothesis.

    The authors report in a mouse model that is engineered to contain a mutated APP under its endogenous promoter, changes in energy and cell-growth related metabolism that manifest prior to plaque deposition. This is consistent with reports from other investigators that, in other mouse models with mutant APP, energy metabolism-related changes substantially precede plaque deposition (Reddy et al, 2004). 

    The current paper speculates that Aβ oligomers are potentially responsible for this study’s observed metabolism changes, which I expect would be the favored hypothesis of those who believe AD is a primary disorder of Aβ oligomers.

    The data in this manuscript are also pertinent to, and potentially consistent with, other data that emphasize functional links between APP and mitochondria, which seem to exist even in the absence of classic APP processing to Aβ (Pope et al., 2021).

    References:

    . Gene expression profiles of transcripts in amyloid precursor protein transgenic mice: up-regulation of mitochondrial metabolism and apoptotic genes is an early cellular change in Alzheimer's disease. Hum Mol Genet. 2004 Jun 15;13(12):1225-40. PubMed.

    . Mutations in the Amyloid-β Protein Precursor Reduce Mitochondrial Function and Alter Gene Expression Independent of 42-Residue Amyloid-β Peptide. J Alzheimers Dis. 2021;83(3):1039-1049. PubMed.

  3. Norambuena and colleagues used an APPSAA/AppNL-G-I knock-in mouse model of AD to demonstrate mitochondrial dysfunction in the very early stage of AD development. It is hard to understand why the authors used a copy model of our AppNL-G-F and other knock-in lines (Saito et al., 2014; Watamura et al., 2022), which are being used by more than 800 groups worldwide (Saido, Alzheimer's Research Guide: Animal Models for Understanding Mechanisms and Medications, 2024, Elsevier, in press).

    Because more than 200 papers using our knock-in mouse lines have been published and because there will be many more, one can make comparisons with past and future reports by using our models. Also, it is sometimes important to utilize the AppNL or ApphAb lines as a negative control.

    References:

    . Single App knock-in mouse models of Alzheimer's disease. Nat Neurosci. 2014 May;17(5):661-3. Epub 2014 Apr 13 PubMed.

    . An isogenic panel of App knock-in mouse models: Profiling β-secretase inhibition and endosomal abnormalities. Sci Adv. 2022 Jun 10;8(23):eabm6155. Epub 2022 Jun 8 PubMed.

  4. We are grateful for the comments that Alzforum readers made to our bioRxiv preprint. Brain hypometabolism is one the earliest biomarkers in AD pathogenesis, but the underlying mechanisms have been difficult to unravel. Previous work by our lab unveiled a novel form of communication between lysosomes and mitochondria, a signaling pathway we named NiMA, for Nutrient-induced Mitochondrial Activity. NiMA involves activation of the mechanistic target of rapamycin complex 1 (mTORC1) by insulin or the amino acids arginine and leucine to induce mitochondrial respiration within minutes in the absence of mitochondrial biogenesis (Norambuena et al., 2018, 2024). Importantly, NiMA was found to be blocked in cultured neurons by soluble Aβ oligomers (AβOs), which activate mTORC1 at the plasma membrane (Norambuena et al., 2017, 2018) and simultaneously inhibit lysosomal mTORC1 (Norambuena et al., 2022). The current bioRxiv paper, which is under peer review, describes NiMA inhibition in vivo in the brains of an APP knock-in, AD mouse model.

    Importantly, we found that NiMA disruption in vivo begins when AβOs are reportedly first detectable in brain, CSF, and plasma, more than a year before any other pathophysiological features are evident in the heterozygous APPSAA/+ mice used for the study. We used these mice simply because they accumulate AβOs sufficiently fast to produce a phenotype, not to argue that NiMA inhibition in vivo requires amyloidogenic APP mutations.

    We agree with Angelika Harbauer that this body of work “cannot distinguish the contribution of different cell types to this response,” but it does provide strong evidence that NiMA inhibition does occur in vivo in conditions mimicking a very early stage of AD pathogenesis.

    We also want to emphasize that NiMA apparently occurs independently of well-known mTORC1-regulated substrates (Norambuena et al., 2018) and processes such as autophagy (unpublished results).

    References:

    . A novel lysosome-to-mitochondria signaling pathway disrupted by amyloid-β oligomers. EMBO J. 2018 Nov 15;37(22) Epub 2018 Oct 22 PubMed.

    . Disrupted mitochondrial response to nutrients is a presymptomatic event in the cortex of the APP SAA knock-in mouse model of Alzheimer's disease. bioRxiv. 2024 Feb 4; PubMed.

    . mTOR and neuronal cell cycle reentry: How impaired brain insulin signaling promotes Alzheimer's disease. Alzheimers Dement. 2017 Feb;13(2):152-167. Epub 2016 Sep 29 PubMed.

    . SOD1 mediates lysosome-to-mitochondria communication and its dysregulation by amyloid-β oligomers. Neurobiol Dis. 2022 Jul;169:105737. Epub 2022 Apr 20 PubMed.

  5. The original paper on the App-SAA knock-in mouse model of Alzheimer’s disease (JAX: 034711) reported that FDG-PET uptake, a surrogate for glucose utilization in brain, revealed cortical hypermetabolism across the studied lifespan of 5-20 months in homozygous App-SAA mice (Fig. 6b, Xia et al., 2022), which Xia et al. contrasted with the hypometabolism reported in the later phase of AD. Heterozygous App-SAA mice were not tested.

    The preprint's discussion starts off by mentioning that brain hypometabolism is one of the earliest signs of AD, but the cortical hypermetabolism of homozygous App-SAA mice isn't mentioned.

    Can the finding of a disrupted mitochondrial response in heterozygous App-SAA mice help explain the cortical hypermetabolism of homozygous App-SAA mice?

    In contrast with App-SAA homozygotes, App-NL-G-F homozygotes do exhibit cerebral glucose hypometabolism (Armstrong et al., 2024).

    References:

    . Novel App knock-in mouse model shows key features of amyloid pathology and reveals profound metabolic dysregulation of microglia. Mol Neurodegener. 2022 Jun 11;17(1):41. PubMed.

    . Protective effect of PDE4B subtype-specific inhibition in an App knock-in mouse model for Alzheimer's disease. Neuropsychopharmacology. 2024 Sep;49(10):1559-1568. Epub 2024 Mar 23 PubMed.

    View all comments by Steven Clapcote

Make a Comment

To make a comment you must login or register.

References

Research Models Citations

  1. AppSAA Knock-in

Mutations Citations

  1. APP K670_M671delinsNL (Swedish)
  2. APP E693G (Arctic)
  3. APP T714I (Austrian)

News Citations

  1. Newfangled Type of Inhibitor Tricks Kinase into Turning It On

Therapeutics Citations

  1. Tideglusib

Paper Citations

  1. . A novel lysosome-to-mitochondria signaling pathway disrupted by amyloid-β oligomers. EMBO J. 2018 Nov 15;37(22) Epub 2018 Oct 22 PubMed.
  2. . GSK3β Regulates Brain Energy Metabolism. Cell Rep. 2018 May 15;23(7):1922-1931.e4. PubMed.
  3. . SOD1 mediates lysosome-to-mitochondria communication and its dysregulation by amyloid-β oligomers. Neurobiol Dis. 2022 Jul;169:105737. Epub 2022 Apr 20 PubMed.
  4. . mTOR and neuronal cell cycle reentry: How impaired brain insulin signaling promotes Alzheimer's disease. Alzheimers Dement. 2017 Feb;13(2):152-167. Epub 2016 Sep 29 PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Disrupted mitochondrial response to nutrients is a presymptomatic event in the cortex of the APPSAA knock-in mouse model of Alzheimer disease. 2024 Feb 04 10.1101/2024.02.02.578668 (version 1) bioRxiv.