Finding. The first bite of a mitochondrion can happen on the endoplasmic reticulum. Ortiz-Silva, Nabi and colleagues put the ER E3 ligase Gp78 on DFCP1-positive omegasomes and show it is required to make those omegasomes and to degrade mitochondrial proteins there. Super-resolution STED says DFCP1 and LC3 structures hug intact mitochondrial fragments that then empty. Acute ivermectin mitophagy that strips TOMM20 needs the Gp78 RING and the BAG6-UBL4A-USP13 module. It does not need PINK1 or Parkin.
Why this paper matters
Textbook mitophagy starts with PINK1 on the outer membrane and Parkin painting ubiquitin. Plenty of tissues keep eating mitochondria without that pair. Gp78 was already on the Parkin-independent list. The missing geometry was where the ER, which births omegasomes, meets the cargo. This paper supplies the geometry.
The intact-fragment observation matters. You do not have to pulverize the organelle before the autophagosome arrives. Content loss is progressive on pieces that still look like mitochondria.
How to read the score
High eighties. Parkin-independent path, named ligase module, omegasome locale, STED. Confidence is high for the ivermectin/RING/BAG6 stack, medium for how general the trigger is.
Caveats
Ivermectin is a hammer. ERAD and mitophagy clients of Gp78 could compete. Do not declare PINK1 optional in substantia nigra from this brief.
What to do with it
If you map mitophagy, add an ER-omegasome, Gp78-positive start site. If you screen Parkin-null turnover, include BAG6-UBL4A-USP13. Pull the 3D STED object tables.