ATG101

ATG101 is a metazoan-specific autophagy protein and obligate stoichiometric member of the ULK1 initiation complex. It has no yeast ortholog, marking the vertebrate ULK1 complex as a four-subunit assembly (ULK1/ULK2 + ATG13 + FIP200 + ATG101) rather than the trimeric yeast Atg1–Atg13–Atg17 system. ATG101’s primary biochemical role is to stabilize ATG13 against proteasomal degradation; without it, ATG13 levels drop and the entire ULK1 complex becomes non-functional, blocking starvation-induced autophagy. Its HORMA domain forms a heterodimer with the ATG13 HORMA domain, and a surface-exposed “WF finger” on ATG101 is required for recruiting downstream autophagy factors.

Identity

Field	Value
UniProt	Q9BSB4 (ATG101_HUMAN)
NCBI Gene	60673
HGNC	25758
Gene symbol	ATG101
Aliases	C12orf44, PP894
Length	218 amino acids
Mass	~25 kDa
Chromosome	12q13.13
Mouse ortholog	Atg101 (one-to-one)
Yeast ortholog	None (metazoan-specific)

ATG101 is encoded by ATG101 (formerly annotated C12orf44 as an open reading frame of unknown function before 2009). Protein-level evidence is confirmed in UniProt (Swiss-Prot entry, manually reviewed).

Evolutionary context — a metazoan-specific component

In Saccharomyces cerevisiae the autophagy initiation kinase complex has three core subunits: Atg1 (kinase), Atg13, and the Atg17–Atg31–Atg29 scaffold. There is no yeast Atg101. In mammals, FIP200 (RB1CC1) replaces the yeast Atg17–Atg31–Atg29 scaffold, and ATG101 — which has no yeast counterpart — joins as a fourth subunit ¹².

This evolutionary distinction has two implications for aging biology:

Results from yeast autophagy genetics do not directly apply to the mammalian ULK1 complex architecture.
ATG101 represents a metazoan innovation that may tune autophagy to cell-type or developmental signals not present in yeast.

Dimension	Status
Pathway conserved in humans?	yes — ATG101 protein well-characterized in human cells
Phenotype conserved in humans?	partial — RNAi data from HeLa; no germline KO mouse lifespan study
Replicated in humans?	no — no human genetics or clinical data yet

needs-human-replication

Domain architecture

ATG101 contains a single recognizable domain spanning most of its 218-residue length: a HORMA domain (aa 1–198 by Qi 2015 structural boundaries; residues 199–218 form the C-terminal safety belt, which appeared completely unstructured in the crystal). HORMA (Hop1, Rev7, Mad2) domains adopt a characteristic fold with a “safety-belt” loop that can adopt open or closed conformations.

Key structural features:

O-conformation HORMA — with regulated WF finger. ATG101 HORMA adopts the O (open) conformation, dimerizing with ATG13 HORMA (which is in the C conformation) in the same architectural orientation as the C-Mad2/O-Mad2 heterodimer ³⁴. However, in the human Atg13-Atg101 crystal structure (1.6 Å), the WF finger of ATG101 is folded back onto the HORMA domain in a closed, inactive (“off”) conformation: Trp110 and Phe112 pack against Ile117, and the aliphatic parts of Gln104, Lys107, and Arg109 bury the motif in a hydrophobic pocket ³. This is distinct from the S. pombe structure (3.0 Å), in which the WF finger projects outward. Qi et al. explicitly note that “the WF finger is not always in a position to bind its as yet un-identified interaction partners” — suggesting the exposure of this motif is regulated rather than constitutive ³. no-mechanism — the molecular trigger that switches the WF finger from closed to open conformation in human cells is unknown.
WF finger. A surface-exposed motif on ATG101 (named for conserved Trp and Phe residues) that is conserved across metazoans but absent in budding yeast. Qi et al. identified this motif via high solvent accessibility in the S. pombe structure (Suzuki 2015) and confirmed its hydrophobic pockets in the human structure; the WF finger residues that form one of five conserved benzamidine-binding pockets suggest a functional binding surface ³. The specific binding partners of the human WF finger “remain to be determined” per Qi et al. ³ — the claim that it recruits GABARAPL1, GABARAPL2, and LC3C derives from Suzuki 2015 ⁴, which is closed-access and cannot be independently verified here. no-fulltext-access (Suzuki 2015 WF-finger/Atg8-family binding claim). needs-replication — WF-finger mutant functional validation in mammalian knock-in models is lacking.
HORMA–HORMA heterodimer interface. The ATG101 HORMA and ATG13 HORMA make extensive contacts in the C-Mad2/O-Mad2 conformational architecture. In the Qi 2015 human structure (1.6 Å), key contacts include: ATG13 Ser127 hydrogen bond to ATG101 His31 (2.8 Å); a buried salt bridge between ATG13 αC Arg133 and ATG101 Asp54 (shortest heteroatom distance 2.64 Å), which is unique to metazoan Atg13-Atg101 complexes; and packing of ATG13 β2′ against ATG101 β2. The interface buries 1,523 Å² of surface area ³. The interface is also the basis for mutual stabilization: Atg13 expressed alone cannot be isolated in the Atg13 HORMA domain construct, indicating Atg101 is needed for its ordered expression ³. Interface integrity was probed by 17 mutants assayed in HEK293 pull-downs ³.
Benzamidine-marked hydrophobic pockets. Qi 2015 identified five hydrophobic pockets on the Atg13-Atg101 heterodimer surface using crystallographic benzamidine soaking. Two (sites 1 and 2) are formed at the Atg13-Atg101 interface and are lined by residues conserved in animals but largely absent from fission yeast — suggesting animal-specific interaction surfaces beyond the WF finger, making the heterodimer an “interaction hub for multiple partners” ³. no-mechanism — the endogenous protein(s) binding these pockets in living cells remain unidentified.

ATG101 lacks catalytic residues; it is a scaffold and adaptor, not an enzyme.

Function

ATG13 stabilization

The best-characterized molecular function of ATG101 is preventing proteasomal degradation of ATG13. When ATG101 is knocked down by RNAi in HeLa cells, ATG13 protein levels drop markedly without a corresponding reduction in ATG13 mRNA, indicating the loss is post-translational ¹². ULK1 levels are also modestly reduced in the same experiments, consistent with ATG13 serving as the scaffold that bridges ATG101 to ULK1. The ULK1 complex thus collapses in the absence of ATG101.

Autophagy flux requirement

ATG101 RNAi in HeLa cells prevents starvation-induced autophagy flux, as measured by LC3-II accumulation and GFP-LC3 puncta formation under amino acid withdrawal ¹². Both 2009 discovery papers independently report this phenotype, providing parallel confirmation. needs-replication — no mammalian germline Atg101 KO lifespan study has been characterized in the original papers; conditional-KO and tissue-specific data remain to be established.

Recruitment of Atg8-family proteins to the phagophore

Via its WF finger, ATG101 is proposed to bind members of the mammalian Atg8 family (GABARAPL1, GABARAPL2, and MAP1LC3C) to recruit them to the pre-autophagosomal structure early in the initiation stage, upstream of LC3 lipidation ⁴. This claim derives exclusively from Suzuki 2015 (fission yeast structure + fission yeast functional data), which is closed-access and cannot be verified here. no-fulltext-access (Suzuki 2015). Qi 2015 confirms that the WF finger forms a conserved hydrophobic surface identified by benzamidine binding but states that “what the WF finger binds to…remains to be determined” in human proteins ³. Whether the WF finger function is separable from ATG13 binding in mammalian knock-in models has not been established.

The ULK1 initiation complex — structural context

ATG101 is one of four obligate subunits in the mammalian pre-initiation complex:

Subunit	Role
ulk1 / ULK2	Ser/Thr kinase; integrates mTORC1 and AMPK signals
atg13	Scaffold; binds ULK1 C-terminus; HORMA domain docks ATG101
fip200	RB1CC1; FIP200/FAT-C domains; recruits downstream ATG proteins
ATG101	Stabilizes ATG13; WF finger recruits Atg8-family members

The complex constitutively assembles under nutrient-replete conditions but is held inactive by mTORC1-mediated phosphorylation of ULK1 (primarily Ser758 in human ULK1) and ATG13. Upon mTORC1 inhibition (starvation, rapamycin), or AMPK activation (energy deficit), ULK1 kinase activity is released and autophosphorylates complex members to nucleate the phagophore.

ATG101 does not carry known regulatory phosphorylation sites analogous to the mTORC1 or AMPK sites on ULK1 and ATG13. Its regulation appears to operate through the complex rather than through independent PTM control. unsourced — confirm no PTMs on ATG101 from PhosphoSitePlus during next lint pass.

Discovery

ATG101 was identified independently in 2009 by two groups:

Mercer, Kaliappan & Dennis (2009) — identified the protein as a novel ATG13-binding partner in a biochemical screen. Named it Atg101. Showed it interacts with both ATG13 and ULK1, demonstrated ATG101 RNAi destabilizes ATG13 and blocks starvation-induced autophagy in HeLa cells ¹.
Hosokawa, Hara, Kaizuka et al. (2009, Autophagy) — independently identified the same protein interacting with ATG13 in a mammalian context. Provided corroborating evidence for the ATG13-stabilization function and complex membership ².

Note: Hosokawa et al. also published a separate 2009 paper (Hosokawa et al. 2009, Mol Biol Cell, 10.1091/mbc.e08-12-1248) describing the mammalian ULK1–ATG13–FIP200 trimer — that earlier paper does not include ATG101, which was added in the parallel Autophagy paper. The ulk1 page explicitly reflects this attribution.

The structural basis for ATG101–ATG13 association was resolved in 2015 by two crystal structures:

Suzuki, Kaizuka, Mizushima & Noda (2015, Nat Struct Mol Biol) — solved the S. pombe Atg101–Atg13 complex at 3.0 Å; revealed HORMA–HORMA heterodimer and identified the WF finger as a solvent-exposed functional motif; reported WF-finger binding to Atg8-family members in fission yeast ⁴. Note: closed-access (#gap/no-fulltext-access); details of Atg8-family binding claim unverified.
Qi, Kim, Stjepanovic & Hurley (2015, Structure) — solved the human ATG13(12–200)–ATG101(1–198) HORMA heterodimer at 1.6 Å (PDB: 5C50); confirmed C-Mad2/O-Mad2 architecture; showed WF finger is in a closed/inactive conformation in the human heterodimer (sequestered in hydrophobic pocket); identified five benzamidine-binding hydrophobic pockets including two animal-specific sites suggesting an interaction hub role; mutational analysis (17 variants) in HEK293 pull-downs ³.

Aging relevance

ATG101 is not itself an aging gene in the direct sense (no lifespan data from its manipulation), but it is a required component of the ULK1 initiation complex, whose activity is essential for autophagy flux. Because:

autophagy declines with age across model organisms and humans;
Restoring ULK1 complex activity (e.g., via mTORC1 inhibition by rapamycin, or AMPK activation) extends lifespan in multiple organisms;
ATG101 loss is sufficient to collapse the ULK1 complex and block autophagy;

…ATG101 is a limiting-factor node for age-related autophagy decline. Whether ATG101 expression or complex assembly changes during aging has not been systematically characterized in aged tissue. unsourced needs-replication

The vertebrate-specific HORMA–HORMA heterodimer interface also means that therapeutic strategies targeting the ULK1 complex must account for ATG101’s structural role — small molecules designed around the yeast Atg1–Atg13 binary interaction will not map directly onto the mammalian complex geometry.

Subcellular localization

Basal conditions: Diffuse cytoplasmic distribution.
Starvation: Recruited to punctate pre-autophagosomal structures (isolation membranes); co-localizes with ATG13 and ULK1 at these sites.

Translocation to the phagophore assembly site is dependent on intact HORMA–HORMA dimerization with ATG13 and on ULK1 kinase activity.

Key interactors

Interactor	Interaction type	Notes
atg13	Direct; HORMA–HORMA heterodimer	Primary binding; ATG101 stabilizes ATG13
ulk1	Indirect; via ATG13 scaffold	No reported direct ULK1–ATG101 binding surface
fip200	Complex co-membership	Co-precipitates in four-subunit complex pull-downs
GABARAPL1	Direct; WF finger (Suzuki 2015)	Atg8-family recruitment to phagophore; Suzuki 2015 claim, closed-access ⁴
GABARAPL2 (GATE-16)	Direct; WF finger (Suzuki 2015)	Atg8-family recruitment; Suzuki 2015 claim, closed-access ⁴
MAP1LC3C	Direct; WF finger (Suzuki 2015)	Atg8-family recruitment; Suzuki 2015 claim, closed-access ⁴

Limitations and gaps

No mammalian germline Atg101 KO phenotype characterized in the discovery papers; conditional KO data would establish its in vivo requirement and tissue-specific roles. needs-replication
No aging-specific expression data — whether ATG101 protein levels or complex assembly changes in aged tissues is unknown. unsourced
PTM landscape essentially uncharacterized — no regulatory phosphorylation sites on ATG101 are established; PhosphoSitePlus cross-check needed. unsourced
WF finger functional data in mammalian cells — Qi 2015 mutational interface analysis used co-transfected HORMA fragments in HEK293 cells (pull-down assay); Suzuki 2015 (closed-access) characterizes WF-finger binding to Atg8-family proteins. Knock-in point mutant validation in mouse or primary mammalian cells is lacking. needs-replication no-fulltext-access (Suzuki 2015)
HGNC ID 25758 and Ensembl ID should be verified against live HGNC and Ensembl databases on next lint pass. needs-canonical-id (Ensembl null in frontmatter)
No human genetic disease directly attributable to ATG101 loss-of-function as of 2026; MIM 615089 is listed in UniProt but the link is to ATG13-related entries. unsourced
Benzamidine pockets are unoccupied in vivo — five hydrophobic pockets on the heterodimer surface (Qi 2015) are conserved and animal-specific, but endogenous binding partners have not been identified. no-mechanism
Mercer 2009 and Hosokawa 2009 full-text unverified — Taylor & Francis paywall blocked PDF download. Quantitative RNAi phenotype data (ATG13 protein level reduction, LC3-II accumulation, GFP-LC3 puncta counts) derive from these papers but cannot be cross-checked against source values. no-fulltext-access

Footnotes

doi:10.4161/auto.5.5.8249 · mercer-2009-atg101-discovery · in-vitro · model: HeLa cells (RNAi knockdown); parallel discovery of ATG101 as ATG13-binding partner; Autophagy 5(5):649–662 (2009); authors: Mercer CA, Kaliappan A, Dennis PB; 417 citations (impact_score 0.80, FWCI 16.7). PDF download failed (Taylor & Francis paywall); quantitative RNAi phenotype claims not verified against full text. no-fulltext-access ↩ ↩² ↩³ ↩⁴
doi:10.4161/auto.5.7.9296 · hosokawa-2009-atg101 · in-vitro · model: mammalian cells; independent parallel ATG101 discovery; Autophagy 5(7):973–979 (2009); authors: Hosokawa N, Sasaki T, Iemura S, Natsume T, Hara T, Mizushima N; 458 citations (impact_score 0.79, FWCI 15.9). PDF download failed (Taylor & Francis paywall); quantitative claims not verified against full text. no-fulltext-access. Distinct from Hosokawa et al. 2009 Mol Biol Cell (10.1091/mbc.e08-12-1248) which describes the ATG13–ULK1–FIP200 trimer without ATG101. ↩ ↩² ↩³ ↩⁴
doi:10.1016/j.str.2015.07.011 · qi-2015-human-atg13-atg101-horma · in-vitro (crystal structure at 1.6 Å, human proteins; pull-down mutational analysis in HEK293 cells) · Structure 23:1848–1857 (2015); authors: Qi S, Kim DJ, Stjepanovic G, Hurley JH; PDB: 5C50; 96 citations; OA PDF available (bronze; local copy confirmed). ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹
doi:10.1038/nsmb.3036 · suzuki-2015-atg101-atg13-horma · in-vitro (crystal structure, S. pombe proteins, 3.0 Å) · Nat Struct Mol Biol 22:572–580 (2015); authors: Suzuki H, Kaizuka T, Mizushima N, Noda NN; 113 citations. Not OA; no local PDF (closed-access). no-fulltext-access ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

🧬 Aging Wiki

Explorer

atg101