Autophagy (macroautophagy)

The cellular bulk-degradation pathway by which damaged organelles, protein aggregates, and other long-lived components are sequestered into double-membraned autophagosomes and delivered to the lysosome for hydrolytic degradation. Declines with age across organisms from yeast to mammals and is now recognized as a primary hallmark in its own right (disabled-macroautophagy, added in the 2023 hallmarks update). Restoring autophagy — genetically or pharmacologically — extends lifespan in multiple models.

This page covers macroautophagy specifically. Selective variants (mitophagy, lipophagy, chaperone-mediated-autophagy) and other catabolic pathways (ubiquitin-proteasome-system) are separate.

Mechanism — the autophagy machinery

Autophagosome formation proceeds in five conceptual stages, each governed by a distinct ATG-protein subset:

StageKey complexFunction
Initiationulk1 (ULK1/ULK2 + ATG13 + FIP200 + ATG101)mTORC1 inhibits ULK1 directly; AMPK activates it. ULK1 phosphorylates downstream Atg proteins to nucleate the phagophore.
NucleationClass III PI3K complex (Beclin-1 + VPS34 + VPS15 + ATG14L)Generates PI3P at the phagophore membrane; recruits WIPI proteins.
ElongationATG12–ATG5–ATG16L1 conjugation system + lc3–PE conjugation systemTwo ubiquitin-like conjugation cascades; LC3-II decorates the autophagosome membrane and serves as the canonical autophagy marker.
Cargo recognitionReceptors: p62/SQSTM1, NBR1, OPTN, NDP52, TAX1BP1Bridge ubiquitinated cargo to LC3 on the inner autophagosome membrane.
Fusion + degradationSNARE machinery (STX17, SNAP29, VAMP8) + lysosomal hydrolasesAutophagosome fuses with lysosome → autolysosome; cargo digested; building blocks recycled.

A complete autophagy flux measurement requires assessing lysosomal degradation, not just autophagosome accumulation — a static increase in LC3-II can mean either elevated formation OR blocked clearance. Standard practice uses bafilomycin A1 (lysosomal inhibitor) to distinguish.

Regulation

SignalEffect on autophagyMechanism
Nutrient excess (amino acids, glucose)↓mTORC1 active → phosphorylates ULK1 + ATG13 → autophagy block
Energy stress (low ATP)↑ampk active → phosphorylates ULK1 directly (activating); also suppresses mTORC1 via TSC2 and Raptor phosphorylation unsourced — TSC2/Raptor detail not in Hansen 2018; cite dedicated AMPK-mTOR literature
ER stress (UPR)↑Activated via PERK + ATF4
Hypoxia↑HIF-1α-dependent + AMPK-dependent
DNA damage↑ (acutely)p53-mediated transcriptional induction
Glucagon↑Cyclic AMP → AMPK

Master transcriptional regulator: TFEB (“master regulator of lysosomal biogenesis and autophagy”) — phosphorylated and cytoplasm-retained by mTORC1; nuclear-translocated upon mTOR inhibition or starvation, driving expression of many lysosomal/autophagy genes (the CLEAR network). unsourced — the frequently-cited “~500 genes” figure does not appear in Hansen 2018 or Settembre 2011; the exact number of CLEAR network targets needs a primary citation (likely Palmieri 2011 or a later TFEB ChIP-seq study).

Selective autophagy

Macroautophagy can be cargo-specific via dedicated receptors:

  • mitophagy — selective autophagy of damaged mitochondria. PINK1/Parkin pathway is the canonical induction route. Major cardio- and neuro-aging relevance.
  • lipophagy — degradation of lipid droplets. Hepatic/metabolic relevance.
  • Aggrephagy — clearance of protein aggregates (p62-bridged). Neurodegeneration relevance.
  • Xenophagy — clearance of intracellular pathogens.
  • Pexophagy / ER-phagy / ribophagy — peroxisome / ER / ribosome turnover.
  • chaperone-mediated-autophagy is not macroautophagy — it uses LAMP2A-mediated direct translocation of soluble proteins into the lysosome, bypassing the autophagosome.

Role in aging

Decline with age

Autophagic flux declines progressively in nearly every tissue measured across model organisms. Mechanisms include:

  • Reduced expression of ATG proteins (ATG5, ATG7, BECN1) in aged tissues
  • Lysosomal dysfunction — accumulation of lipofuscin reduces hydrolase activity
  • Disrupted membrane trafficking (autophagosome–lysosome fusion impairment)
  • Chronic mTOR over-activation in aged tissues (the “hyperfunction” angle)

Genetic evidence — autophagy is required for lifespan extension

The longevity benefits of multiple interventions (caloric restriction, mTOR inhibition, IIS reduction, sirtuin activation) are abolished by autophagy gene knockouts in worms and flies — establishing autophagy as a downstream convergence point for many longevity pathways 1.

Interventions that induce autophagy

InterventionMechanismLifespan evidence
caloric-restrictionReduced amino acid input → mTOR ↓Extensive across organisms
Intermittent fastingPeriodic mTOR ↓ + AMPK ↑Mouse + emerging human data
rapamycinmTORC1 inhibition → ULK1 disinhibitionMouse (NIA ITP)
metforminAMPK activationMouse + observational human
spermidineEP300 inhibitor (specifically narrowed by Pietrocola 2015) — de-represses ATG5/ATG7/ATG8/ATG12 acetylation and reduces H3K56ac; the original “broad HAT inhibitor” framing from Eisenberg 2009 was refined to EP300-selective by Pietrocola 2015 (siRNA screen of 43 acetyltransferases identified EP300 as the unique autophagy-repressor candidate; C646 EP300-specific inhibitor phenocopied the spermidine effect) 2Yeast, worms, flies, human immune cells 2; oxidative stress inhibition (not lifespan) in mice; observational human associations
ExerciseAMPK + LC3 lipidation increase in muscleAcute autophagy induction confirmed in human muscle biopsies

Methods notes

Common autophagy readouts (each with caveats):

  • LC3-II/LC3-I ratio (Western) — increased lipidation indicates more autophagosomes; doesn’t distinguish formation from blocked clearance.
  • p62 levels — accumulates when autophagy is blocked; depleted when flux is high. Confounded by transcriptional regulation.
  • GFP-LC3 / mCherry-GFP-LC3 puncta — gold standard for flux when paired with lysosomal inhibitor controls.
  • Electron microscopy — direct visualization of autophagosomes; labor-intensive.

Limitations and gaps

  • Quantitative autophagy flux in human tissues is technically difficult; most human aging data is from peripheral cells (PBMCs, fibroblasts) rather than the most-affected tissues (brain, heart). needs-human-replication
  • Whether boosting autophagy beyond youthful levels has a hormetic dose-response (too much = bad) is unsettled. dose-response-unclear
  • Cell-type-specific roles of autophagy in aging are not uniform — neurons and cardiomyocytes appear most-dependent; some immune cell contexts may benefit from less autophagy.

Footnotes

Footnotes

  1. doi:10.1038/s41580-018-0033-y · review · multi-organism · key reference for autophagy’s role in aging across model systems ↩

  2. doi:10.1038/ncb1975 · in-vivo (yeast, worms, flies) + human immune cells · multi-organism · spermidine-induced autophagy extends lifespan; mice: oxidative stress inhibition only (not lifespan); mechanism: HAT inhibition → H3 deacetylation → autophagy transcript upregulation ↩ ↩2