ATG7 - 3D Mechanism Simulator

Loading structural data… (3VH4)

References & Databases

START HERE — how to read Atg8 + ATP → E1~Atg8 + AMP + PPi A two-half-reaction E1 cartoon on the yeast Atg7 frame that actually exists: 3VH4 (2.65 Å) S. cerevisiae Atg7 C-terminal domain with crystal Atg8 + MgATP + Zn. Human ATG7 (O95352) has no PDB — species hedge. Yeast Cys507 ≈ human Cys572. Teaching cargo is LC3; this fold is yeast Atg8. Crystal occupancy is adenylation-like MgATP·Gly116, not E1~Atg8 thioester. Cys507 SG stands ~14 Å from Gly116 C on this snapshot, so the catalytic Cys507 C-domain must swing in ~14 Å to the adenylate site — the E1~Atg8 thioester is drawn [schematic] at the active site. AMP, PPi, Atg8-AMP analog, and the thioester are [schematic]. Cys507 ★ is the compiled E1 thiol. Zn knuckle is named occupancy, not the nucleophile. White is nucleotidyl P-attack (Gly-OXT → ATP PA). Blue is Cys-S nucleophile, not hydride. Measured Å are deposited 3VH4 pairs only. ATG12 is not this page. ATG3 next. 3T7H NTD is a companion, not this mesh. Atg8 K26P is an engineered mutation, named. Why no PE: PE (phosphatidylethanolamine) is the membrane-lipid anchor that LC3 is ultimately conjugated to; it is downstream of this E1 step and has no coordinates on 3VH4 or any ATG page (atg3/atg4/atg5) — named on the pathway, never drawn. Not a pure A-page (covalent E~Atg8).
Crystal structures PDB 3VH4 — yeast Atg7 CTD (UNP 295–630) + Atg8 + MgATP, 2.65 Å (Noda 2011). Coordinates are public. 3T7H is the Atg7 N-terminal domain companion — not this mesh. Human ATG7 UniProt O95352 has no PDB. Atg8 on this deposit carries engineered K26P (SEQADV).
Mechanism — two E1 half-reactions Atg7 is the autophagy E1: Atg8/LC3 + ATP → E1~Atg8 thioester + AMP + PPi. (1) Atg8 C-terminus attacks ATP α-P → Atg8-AMP + PPi. (2) Cys507-SH attacks that carbonyl → E1~Atg8 + AMP. Noda et al., Mol. Cell 2011, PMID:22055191. No IUBMB EC is assigned (UBA-like). ATG12 activation is a different cargo and is not this page. Source: Noda 2011 · UniProt P38862 (yeast) · O95352 (human, no PDB)
0 — Setup: Mg and Zn present, pocket shown empty 3VH4 holds MgATP + Atg8 Gly116. The pocket is shown empty so ordered binding can start. Mg and Zn stay as named occupants. Opaque, run:null (no autoplay). Human O95352 has no PDB; this mesh is yeast. Cys507 ★ is the compiled E1 thiol (human Cys572).Source: PDB 3VH4 (2.65 Å)
① — Atg8 C-terminus binds Crystal Atg8 Gly116 docks as the ubiquitin-like C-terminus — real 3VH4 occupancy. Teaching human cargo is LC3, not this yeast Atg8 fold. Carboxylate faces the ATP α-phosphate / Mg. ATG12 is not this page.Source: 3VH4 Atg8 chain B Gly116
② — ATP binds Crystal ATP docks as deposited 3VH4 MgATP. Gly116 OXT ⋯ ATP PA = 2.83 Å. Mg ⋯ ATP O2B = 2.26 Å. This is not AMP and not the thioester. Cys507 is still far from Gly116 C on this snapshot.Source: 3VH4 ATP-A702 / MG-A703
③ — Gly-OXT attacks ATP α-P → Atg8-AMP + PPi (hypothetical, lit.-based) White phosphoryl rides from Atg8 C-terminal OXT to crystal ATP PA. Gly116 + ATP fade in place to schematic Atg8-AMP analog and schematic PPi. The deposited crystal is not the mixed anhydride — 3VH4 holds ATP + free Gly116; the Atg8-AMP analog is schematic. Cys507 is not yet the nucleophile. Slow-motion on purpose. White = nucleotidyl P-attack, not hydride.Source: compiled E1 adenylation · Noda 2011
④ — PPi leaves [schematic] PPi departs. The Atg8-AMP analog stays as teaching occupancy. Asp438 OD2 ⋯ Mg = 2.62 Å is the deposited metal ligand. Cys507 still stands ~14 Å from Gly116 C on this crystal — the next attack is [schematic], not this snapshot.Source: 3VH4 Asp438 / MG; schematic PPi
⑤ — Cys507-S attacks carbonyl → E1~Atg8 + AMP (hypothetical, lit.-based) Blue nucleophile glides in from the Cys side and seats beside the Atg8-AMP analog carbonyl — the catalytic Cys507 C-domain swings in ~14 Å to the adenylate site, so the attack is local (short blue arrow), not a 14 Å reach. Atg8-AMP fades in place to schematic E1~Atg8 thioester (now at the active site, S-bonded to Cys507) and schematic AMP at the nucleotide seat. On this crystal Cys507 SG is ~14 Å from Gly116 C — thioester geometry is [schematic], not deposited. Second E1 half-reaction. Not a pure A-page. Slow-motion on purpose. Blue = sulfur nucleophile, not hydride.Source: compiled E1 thioesterification · UniProt P38862 Cys507
⑥ — AMP leaves — ATG3 next [schematic AMP] AMP departs. Schematic E1~Atg8 thioester stays at the active site (Cys507 swung in). Mg and Zn stay. PPi already left with the first half-reaction. The thioester is the substrate of ATG3. This page does not transfer to E2.Source: autophagy E1→E2 bookkeeping · Noda 2011

How to read this page — a first-pass primer First pass: Setup (Mg+Zn present, pocket empty, Cys507 ★, yeast mesh) → ① Gly116 docks → ② crystal ATP docks → ③ white Gly-OXT → ATP PA, fade to Atg8-AMP analog + PPi → ④ PPi leaves → ⑤ blue Cys507-S swings in → carbonyl (local attack), fade to E1~Atg8 + AMP at the active site → ⑥ AMP leaves; thioester stays; ATG3 next. Second pass: 2.83 / 2.26 / 2.62 Å are crystal facts on 3VH4; AMP, PPi, Atg8-AMP, and the thioester are [schematic]. White = nucleotidyl P-attack. Blue = Cys-S. Third pass: open “Disagreements” for species hedge, LC3 vs Atg8, adenylation occupancy vs thioester, Cys507 ~14 Å, and ATG12 not this page.
Chemistry primer — E1 adenylation then thioester ATG7 activates Atg8/LC3 in two half-reactions, like other ubiquitin-like E1 enzymes. The C-terminal carboxylate attacks ATP α-P; PPi leaves and an Atg8-AMP mixed anhydride forms. Cys507-SH then attacks that carbonyl; AMP leaves and the product is E1~Atg8 thioester, the substrate of ATG3. On 3VH4 the Cys507 SG stands ~14 Å from that carbonyl, so the Cys507 C-domain swings in to the adenylate site and the thioester is drawn [schematic] there (not out at the far Cys). Gold is unused on this page (no catalytic H⁺ sphere). White = nucleotidyl P-attack (Gly-OXT → ATP PA). Blue = sulfur nucleophile (Cys-S → carbonyl), not hydride. This is not ATG3 (E2 transfer) and not ATG12 cargo. PE (phosphatidylethanolamine) — the membrane-lipid anchor LC3 is finally conjugated to — is downstream and absent from 3VH4, so it is named, never drawn.
Words used on screen — species hedge · LC3 vs Atg8 · E1 Species hedge — the mesh is yeast Atg7 CTD; human ATG7 has no PDB. Yeast Cys507 ≈ human Cys572. LC3 vs Atg8 — teaching human cargo is LC3; this fold is yeast Atg8. E1 — ubiquitin-like activating enzyme (two half-reactions). Adenylation occupancy — deposited MgATP + free Gly116, not intact Atg8-AMP and not E1~Atg8.

Legend & Fidelity

Mechanism fidelity — measured vs. inferred The measured layer: 3VH4 (2.65 Å) yeast Atg7 CTD with Atg8, ATP, MG, ZN; Gly116 OXT ⋯ ATP PA = 2.83 Å; Gly116 C ⋯ ATP PA = 3.74 Å; Gly116 OXT ⋯ ATP O1A = 2.36 Å; MG ⋯ ATP O2B = 2.26 Å; MG ⋯ ATP O3G = 2.40 Å; Asp438 OD2 ⋯ MG = 2.62 Å; Asp438 OD2 ⋯ ATP O2A = 2.82 Å. The inferred layer ([schematic]): PPi, Atg8-AMP analog, AMP, E1~Atg8 thioester, transferring nucleophile spheres, product leave. Cys507 SG ⋯ Gly116 C ≈ 14 Å is deposited — the catalytic Cys507 C-domain must swing in ~14 Å to the adenylate site, so the ⑤ attack is drawn locally (short arrow) and the E1~Atg8 thioester is placed [schematic] at the active site, not out at the far Cys. Binding order is compiled staging, not a kinetic fit. The protein mesh is static.
Disagreements on record (a) Species. Human O95352 has no PDB; 3VH4 is yeast Atg7 CTD. (b) LC3 vs Atg8. Teaching cargo is LC3; this fold is yeast Atg8. (c) Adenylation occupancy ≠ thioester. Crystal holds MgATP + free Gly116, not E1~Atg8 and not intact Atg8-AMP. (d) Cys507 is far. SG ⋯ Gly116 C ≈ 14 Å on this snapshot, so the catalytic Cys507 C-domain must swing in to the adenylate site; the drawn attack is local (short arrow) and the thioester is placed [schematic] at the active site (the thioester geometry itself is schematic). (e) Concerted vs split. Compiled E1 writes two half-reactions; this page splits them (③ then ⑤) for 1-event/1-step teaching. (f) ATG12 is a different cargo and is not this page. (g) 3T7H NTD is a companion, not this mesh. (h) Atg8 K26P is an engineered mutation on this deposit. (i) No IUBMB EC assigned (UBA-like).
Regulation — starvation · ATG12 cargo This mesh is yeast Atg7 CTD activating Atg8. Human ATG7 also activates ATG12 — that cargo is not this page (ATG12 is not an enzyme; the conjugate lives on the ATG5 occupancy page). Starvation / mTOR / ULK1 gating of autophagy is drawer, not motion. Homodimeric E1 architecture is named from Noda 2011, not a second chain on this cartoon.Source: UniProt P38862 · O95352 · Noda 2011
Kinetics scope (no measured timecourse on this page) One Atg8 + ATP → E1~Atg8 + AMP + PPi event is drawn, not a steady-state run. No measured k_cat/K_m is re-asserted here — no measured k_cat/K_m on this page [schematic]. Pair with a kinetics chapter.Source: Noda 2011 — no k_cat/K_m re-asserted
Sequence Yeast Atg7 UniProt P38862 · human ATG7 O95352 (no PDB) · NCBI Gene 10533. This screen uses yeast numbering (Cys507 on 3VH4). Chain A Cα 294–613. Atg8 chain B Cα 4–115 offset +1000. Gly116 is the ligand, not on the tube.
Simulation map — what you see is what is written Setup = 3VH4 pocket empty, Mg and Zn visible (opaque, no autoplay). ① crystal Gly116 moves in. ② crystal ATP moves in. ③ slow-mo: white Gly-OXT → ATP PA; Gly116+ATP fade to Atg8-AMP analog + PPi. ④ schematic PPi leaves. ⑤ slow-mo: blue Cys507-S swings in to the carbonyl (local attack); analog fades in place to E1~Atg8 (at the active site) + AMP. ⑥ AMP leaves; thioester stays; ATG3 next. Only these events move.
Sources — bibliography of this page S1 — PDB 3VH4: yeast Atg7 CTD·Atg8·MgATP, 2.65 Å — public coordinates; measured OXT⋯PA = 2.83 Å · O2B⋯Mg = 2.26 Å · OD2⋯Mg = 2.62 Å S2 — Noda et al., Mol Cell 2011, PMID:22055191 — Atg8 activation by homodimeric E1 Atg7; companion 3T7H NTD S3 — Sequence: UniProt P38862 (yeast Atg7) · O95352 (human ATG7, no PDB) · NCBI Gene 10533 S4 — Kinetics: no k_cat/K_m re-asserted on this page

› Backbones and ligand poses are taken directly from the listed PDB entries, superposed into one common frame. Where no crystalline state exists for a step (e.g., Atg8-AMP, PPi, AMP, E1~Atg8 thioester, nucleophile flight, product release), the pose is drawn schematically and labeled as such in the narration. For ⑤ the Cys507 C-domain is far (~14 Å) on the crystal and must swing in to the adenylate site, so the drawn nucleophile attack is local (short arrow) and the E1~Atg8 thioester is placed at the active site. PE (the membrane-lipid anchor LC3 is finally conjugated to) has no coordinates on 3VH4 or any ATG page, so it is named, never drawn.

Atg7 CTDAtg8ONphosphoryl (white)nucleophile (blue)