ULK1 (Unc-51 like autophagy activating kinase 1)

The apex kinase of macroautophagy initiation — integrates upstream nutrient and energy signals from mTORC1 and AMPK and converts them into autophagosome nucleation. ULK1 is the mammalian ortholog of yeast Atg1, and it sits at the convergence point of the two dominant longevity-relevant signaling axes: mTORC1 inhibition (by rapamycin, caloric restriction, or starvation) releases ULK1 from tonic suppression; AMPK activation (by energy deficit) directly promotes ULK1 activity. Because autophagy declines with age and restoring autophagic flux extends lifespan in multiple model organisms, ULK1 is a central node in the aging biology network.


Identity

FieldValue
UniProtO75385 (ULK1_HUMAN)
NCBI Gene8408
HGNC12558
EnsemblENSG00000177169
Gene symbolULK1
Full nameSerine/threonine-protein kinase ULK1
Chromosomal location12q24.33
Length1050 amino acids
Mouse orthologUlk1 (one-to-one; high sequence conservation in kinase domain)
GenAge entrynull needs-canonical-id — ULK1 was not found in GenAge human subset as of 2026-05-04; check GenAge-models for worm/fly orthologs

Naming note: ULK1 and ATG1 are synonyms. Literature before ~2010 often uses ATG1 (yeast-derived) or hATG1 (human ATG1). UniProt lists both as alternative names. The pathway page autophagy uses ULK1 as the canonical name.


Functional domains

DomainResidues (human)Function
Protein kinase domain16–278ATP-binding catalytic core; site of autophosphorylation and substrate phosphorylation
GABARAP/GABARAPL2 interaction region287–416Mediates binding to LC3 family proteins; important for selective autophagy targeting
Ser-rich low-complexity region295–317Contains Ser317 AMPK phosphorylation site; regulatory hotspot
Extended disordered region393–608Contains Ser556 and Ser638/639 phosphorylation sites
C-terminal domain828–1050Mediates interaction with SESN2; complex assembly scaffold

The disordered linker region between the kinase domain and the C-terminal domain (residues ~300–830) is the primary site of regulatory phosphorylation by both AMPK and mTOR. This region has no known intrinsic fold and is thought to serve as a flexible scaffold for signal integration.


Key post-translational modifications

PTMSite (human)KinaseFunctional consequence
PhosphorylationSer317AMPKActivating; promotes ULK1 kinase activity
PhosphorylationSer556AMPKAMPK-mediated activation site (UniProt O75385); mTOR phosphorylation of human Ser556 is not supported by Kim 2011 or UniProt annotations — remove “also mTOR” claim unsourced
PhosphorylationSer758mTOR (mTORC1)Inhibitory; disrupts ULK1-AMPK interaction and suppresses ULK1 activity
AcetylationLys162KAT5/TIP60Function not fully characterized no-mechanism
AcetylationLys607KAT5/TIP60Function not fully characterized no-mechanism
UbiquitinationLys-63 linkageMultiple E3 ligasesDegradative and non-degradative roles reported

Critical species-numbering caveat: The widely-cited Kim et al. 2011 (Nature Cell Biology) characterizes AMPK-mediated phosphorylation of mouse Ulk1 at Ser317 and Ser777 and mTOR-mediated phosphorylation at Ser757 (mouse numbering) 1. Per the verifier round on ampk, the Ser777 mouse site is NOT conserved in human ULK1 — it does not appear in UniProt O75385. Ser317 (mouse) corresponds to a conserved position in human ULK1 and is annotated as an AMPK substrate in UniProt. The human mTOR site is annotated at Ser758 (one residue offset from the mouse Ser757). Whenever citing site-specific data from Kim 2011 or related papers, confirm whether mouse or human numbering applies.

needs-replication — The full complement of functionally significant phosphorylation sites in human ULK1 under physiological aging conditions has not been systematically mapped.


Core mechanism: autophagy initiation

ULK1 forms the ULK1 initiation complex with three obligate binding partners:

  • ATG13 — scaffold protein; constitutively bound to ULK1; required for ULK1 stability and localization to the phagophore assembly site
  • FIP200 (RB1CC1) — focal adhesion kinase family interacting protein 200 kDa; scaffolds the complex and contributes to phagophore membrane targeting
  • ATG101 — stabilizes ATG13 and is required for the complex’s autophagic activity unsourced — ATG101 is NOT described in Hosokawa 2009 or Ganley 2009 (those papers characterize only ULK1–ATG13–FIP200); ATG101 was identified as a fourth subunit in subsequent work (Mercer et al. 2009, FEBS Lett; and others). Citation 23 removed from this claim — needs correct primary citation.

Under nutrient-replete conditions, mTORC1 associates with the ULK1 complex via RAPTOR and phosphorylates both ULK1 (at Ser758) and ATG13, keeping the complex in a low-activity state. Upon starvation or rapamycin treatment, mTORC1 dissociates from the ULK1 complex; ULK1 dephosphorylates at Ser758 and undergoes autophosphorylation, activating the complex 2.

Once active, ULK1 initiates autophagy by phosphorylating downstream targets:

  1. AMBRA1 — ULK1-phosphorylated AMBRA1 activates the PI3K class III complex (VPS34-Beclin-1-ATG14L), driving PI(3)P production at the ER and phagophore nucleation
  2. Beclin-1 (BECN1) — direct phosphorylation promotes VPS34 complex activity
  3. ATG13 and FIP200 — ULK1 autophosphorylates and transphosphorylates these subunits to amplify and sustain complex activity
  4. SQSTM1/p62 and other cargo receptors — ULK1 phosphorylates selective autophagy receptors to regulate cargo loading

Dual regulation by mTORC1 and AMPK

The opposing phosphorylation model 1:

                    nutrient-rich / high ATP
                    
    mTORC1 ──phospho-Ser758──> ULK1 [INHIBITED]
                                 │
                    starvation / low AMP:ATP ratio
                    
    AMPK ──phospho-Ser317/556──> ULK1 [ACTIVATED]
    mTORC1 dissociates

AMPK and mTORC1 thus exert opposing effects at the same protein, and the balance of their activities determines autophagic flux. This is the mechanistic basis for the well-established observation that nutrient starvation, AMPK-activating compounds (metformin), and mTORC1 inhibitors (rapamycin) all converge on ULK1 activation.

Negative feedback: Once activated, ULK1 phosphorylates AMPK subunits (PRKAA2, PRKAB1) to attenuate AMPK activity, and phosphorylates RAPTOR (a component of mTORC1) to suppress mTORC1 — constituting a negative feedback loop that tempers sustained autophagy induction (UniProt O75385).


Role in aging

Autophagy decline with age

Disabled-macroautophagy is a hallmark of aging. Autophagic flux declines across cell types and tissues with age in multiple model organisms and in humans. ULK1 lies at the apex of this decline — any mechanism that chronically elevates mTORC1 activity or suppresses AMPK activity with age will converge on reduced ULK1 activity and impaired autophagy initiation. no-mechanism — the specific upstream drivers that chronically suppress ULK1 activity in aged mammalian tissues have not been fully characterized; candidate mechanisms include elevated nutrient signaling, reduced AMPK sensitivity, and post-translational modifications that alter the AMPK–mTOR balance.

mTOR inhibition → ULK1 → autophagy → longevity axis

The dominant mechanistic model linking mTOR inhibition to lifespan extension places ULK1 as the required intermediate:

  • Rapamycin extends lifespan in worms, flies, and mice by inhibiting mTORC1
  • mTORC1 inhibition releases ULK1 from Ser758-phosphorylated suppression
  • Active ULK1 initiates autophagy
  • Atg5 RNAi or equivalent loss-of-function abolishes the lifespan-extending effect of rapamycin in C. elegans and Drosophila, demonstrating that autophagy induction is required for the benefit 4 no-fulltext-access

Supporting the sufficiency direction: overexpression of Atg5 (a downstream autophagy gene) in mice activated autophagy and extended median lifespan by 17.2% (both sexes; χ²=17.32, s.e.m., p<0.001) compared to WT littermates, with improved motor function, leanness, increased insulin sensitivity, and reduced age-related pathology 5. needs-human-replication — no equivalent human gain-of-function data exists; the Pyo 2013 finding is mouse-specific.

DimensionStatusNotes
Pathway conserved in humans?yesULK1 complex structure, mTOR inhibitory phosphorylation, and AMPK activating phosphorylation are conserved; key sites confirmed in UniProt O75385
Phenotype conserved in humans?partialAutophagic flux declines with human age (multiple tissue studies); direct ULK1 activity data in aged human tissue is sparse unsourced
Replicated in humans?noNo human lifespan or healthspan trial directly targeting ULK1; indirect evidence via rapamycin/everolimus trials needs-human-replication

ULK1 knockout models

Global Ulk1−/− mice show neonatal lethality during the suckling-to-weaning transition (failure to clear damaged mitochondria in neurons via mitophagy), demonstrating the necessity of ULK1-mediated autophagy in vivo unsourced — neonatal lethality was NOT reported in Hosokawa 2009 (which characterized MEF autophagy defects only); lethality was reported by Kundu et al. 2008 (Blood 112:1493–1502, doi:10.1182/blood-2008-01-132639) — citation 2 removed; correct citation needed. Tissue-specific conditional KOs have been used to study ULK1’s role in specific tissues. The embryonic/neonatal lethality phenotype complicates interpretation of global knockout lifespan studies — ULK1’s contribution to mammalian aging must be inferred primarily from pathway-level evidence and downstream autophagy manipulations. needs-replication — tissue-specific conditional KOs of Ulk1 in aged animals are sparse; aging phenotypes in these models are not well characterized as of 2026.


Pathway membership

  • autophagy — apex initiation kinase; directly activates the phagophore nucleation machinery
  • mtor — direct substrate of mTORC1 (inhibitory phosphorylation at Ser758); ULK1 also feeds back to suppress mTORC1 via RAPTOR phosphorylation
  • ampk — direct substrate of AMPK (activating phosphorylation at Ser317 and Ser556); ULK1 feeds back to attenuate AMPK activity

Key interactors

InteractorInteraction typeFunctional consequence
atg13constitutive complex subunitstabilizes ULK1; required for complex localization to phagophore assembly site
fip200constitutive complex subunit (RB1CC1)scaffold; phagophore membrane targeting
ATG101constitutive complex subunitstabilizes ATG13; required for autophagic activity
mTORC1 (via RAPTOR)nutrient-dependent associationmTOR phosphorylates ULK1-Ser758 to inhibit; dissociates under starvation
ampksignaling kinasephosphorylates ULK1-Ser317/556 to activate
beclin-1downstream substrateULK1 phosphorylates Beclin-1 to activate VPS34 PI3K complex
ambra1downstream substrateULK1-phosphorylated AMBRA1 activates PI3K class III complex at the ER
SESN2C-terminal domain interactionsestrins link oxidative stress and AMPK to ULK1 complex

Interventions that modulate ULK1

  • rapamycin — allosteric mTORC1 inhibitor; releases ULK1 from Ser758-mediated inhibition; best-validated pharmacological autophagy inducer for lifespan extension in model organisms
  • Metformin / AMPK activators — activate AMPK → phospho-Ser317/556 → ULK1 activation; ULK1 is the proximate effector of metformin’s autophagy-inducing effect
  • Caloric restriction — reduces mTORC1 tone and activates AMPK; mechanistic route to ULK1 activation
  • Exercise — activates AMPK in muscle (AMP:ATP ratio increases); ULK1 Ser317 phosphorylation detectable in skeletal muscle post-exercise 6

ULK1-directed small molecules: SBI-0206965 (ULK1/2 inhibitor, tool compound) has been used to confirm ULK1 kinase dependence in cellular autophagy assays. No ULK1 activators have progressed to clinical use as of 2026. long-term-unknown — pharmacological ULK1 activation strategies are preclinical only.


Limitations and open questions

GapTagNotes
Ser777 not conserved in human ULK1Kim 2011 explicitly notes “Ser 777 in the mouse Ulk1 is not conserved in human Ulk1” (Discussion); downstream translational studies must use human-specific site mapping
ATG101 citation requires correctionunsourcedATG101 as a fourth ULK1-complex subunit was NOT described in Hosokawa 2009 or Ganley 2009; needs citation to the correct primary source (Mercer et al. or subsequent work)
Ulk1−/− neonatal lethality citation requires correctionunsourcedLethality NOT reported in Hosokawa 2009 — that paper showed MEF autophagy defects only; lethality is from Kundu et al. 2008 Blood (doi:10.1182/blood-2008-01-132639)
ULK1 activity in aged human tissueunsourcedAutophagic flux known to decline with age; direct ULK1 kinase activity measurements in human aged tissue are sparse
Tissue-specific Ulk1 KO aging studiesneeds-replicationGlobal KO lethal; conditional KO aging phenotypes not well characterized
Human lifespan / healthspan evidenceneeds-human-replicationAll longevity data is model-organism; rapamycin trials address mTOR, not ULK1 specifically
ULK1 acetylation functional roleno-mechanismKAT5/TIP60 acetylation at Lys162 and Lys607 — functional consequences unknown
GenAge entryneeds-canonical-idULK1 not found in GenAge human subset; GenAge-models entry for worm/fly ATG1 orthologs not cross-referenced here

Footnotes

Footnotes

  1. kim-2011-ampk-mtor-ulk1 · doi:10.1038/ncb2152 · n=N/A (biochemical + cell-based) · in-vitro + in-vivo (MEFs and mouse tissue) · model: HEK293, MEFs, mouse skeletal muscle · key finding: AMPK directly phosphorylates Ulk1 at Ser317 and Ser777 (mouse); mTOR phosphorylates Ulk1 at Ser757 (mouse); Ser757 phosphorylation disrupts AMPK-ULK1 interaction; Ser777 site NOT conserved in human ULK1 (see caveat above) · 6,816 citations · locally downloaded (green OA) 2

  2. hosokawa-2009-ulk1-atg13-fip200-mtor · doi:10.1091/mbc.e08-12-1248 · n=N/A (biochemical + cell-based) · in-vitro · model: HEK293T, MEF, HeLa, NIH3T3 cells · key finding: ULK1–Atg13–FIP200 form a stable ~3 MDa complex (ATG101 not described here — identified in subsequent work); mTORC1 associates with the complex in a nutrient-dependent manner via RAPTOR and phosphorylates ULK1 and Atg13; rapamycin/starvation triggers ULK1 dephosphorylation and complex activation; Ulk1−/− MEFs show autophagy defect on starvation (neonatal lethality NOT reported here — that is from Kundu et al. 2008 Blood) · 1,995 citations · locally downloaded (green OA) 2 3

  3. ganley-2009-ulk1-atg13-fip200-essential · doi:10.1074/jbc.m900573200 · n=N/A (biochemical + cell-based) · in-vitro · model: MEF and 293T cells · key finding: ULK1·ATG13·FIP200 triple complex is essential for starvation-induced autophagy; FIP200 and ATG13 individually and additively stimulate ULK1 kinase activity; mTOR directly phosphorylates ULK1 and ATG13; ULK1 kinase activity required for autophagosome formation · 1,512 citations · locally downloaded (hybrid OA)

  4. hansen-2018-autophagy-longevity · doi:10.1038/s41580-018-0033-y · review (Nat Rev Mol Cell Biol) · n=N/A · model: review across C. elegans, Drosophila, mouse data · key claim: Atg5/Atg7 RNAi abolishes lifespan extension by mTOR inhibition in worms and flies, establishing autophagy as a required downstream effector of mTOR inhibition-mediated longevity · 715 citations · no-fulltext-access — DOI lookup failed on two attempts (green OA URL at https://doi.org/10.17863/cam.25782 unreachable by downloader); claims citing this footnote are unverified against the full text

  5. pyo-2013-atg5-lifespan · doi:10.1038/ncomms3300 · n=65 WT + 70 Atg5 Tg (combined sexes, line no. 25; replicated in 3 additional lines) · in-vivo (mouse) · p<0.001 (log-rank) · model: ubiquitous Atg5 overexpressor (C57BL/6 background; pCAGGS promoter) · key finding: Atg5 Tg mice showed 17.2% median lifespan extension (both sexes), elevated autophagy markers, leanness, improved insulin sensitivity, improved motor function, and reduced age-related organ pathology · 689 citations · locally downloaded (green OA)

  6. doi:10.14336/AD.2025.0419 · review (Aging and Disease 2025/2026) · n=N/A · model: narrative review of exercise–AMPK–mTOR–autophagy literature focused on sarcopenic obesity · key claim: exercise-induced ATP depletion activates AMPK, which phosphorylates ULK1 (including at Ser317 and Ser555) to initiate autophagy; review does not present original primary exercise + ULK1 Ser317 muscle data — all claims are syntheses of cited rodent studies · note: caution warranted — this is a secondary review, not primary exercise physiology data · locally downloaded (gold OA)