BAK1 Apoptosis Review Notes

2026-09-30

Reviewed the seeded human BAK1 GOA rows against cached primary literature,
Reactome events, UniProt-derived rows, the BAK-mediated apoptosis GO-CAM, the
completed BAX comparator review, and an annotation-reviewer audit.

Key curation decisions:

• Centered the review on BAK as a constitutively mitochondrial outer-membrane
  BCL2-family effector that changes conformation, self-associates, builds BAK
  oligomers, permeabilizes the outer membrane, and releases cytochrome c after
  BID or other BH3-only inputs [PMID:10950869, "Activated tBID results in an
  allosteric activation of BAK, inducing its intramembranous oligomerization
  into a proposed pore for cytochrome c efflux"; PMID:22006182, "active
  Bax/Bak, but not any other Bcl-2 family protein, displays holin behavior"].
• Accepted GO:0015267 channel activity, GO:0015288 porin activity,
  GO:0005741 mitochondrial outer membrane, GO:0097193 intrinsic apoptotic signaling pathway, GO:0001836 release of cytochrome c from mitochondria,
  and GO:0097145 BAK complex as core existing assertions. The poxvirus GO-CAM
  also models BAK porin activity at the mitochondrial outer membrane upstream of
  CASP9 in intrinsic apoptotic signaling.
• Removed nearly all generic GO:0005515 protein binding rows. Direct
  activator BH3 rows were tightened to GO:0051434 BH3 domain binding; the rest
  were uninformative duplicates of BCL2-family complexes, BAK self-binding,
  MCL1/BCL-XL sequestration, viral inhibitors, or proteome-scale interaction
  edges.
• Treated BAK self-association as a core pore-assembly function. The accepted
  GO:0042802 identical protein binding, GO:0042803 protein homodimerization activity, and BAK-complex rows reflect BH3/groove dimerization and higher
  oligomers rather than generic complex assembly [PMID:23782464, "BAK and BAX
  orchestrate outer mitochondrial membrane permeabilization (MOMP) during
  apoptosis by forming pores in the membrane"].
• Accepted the BCL2-family complex role because BAK is restrained by direct
  MCL1/BCL-XL sequestration in viable cells until BH3-only proteins displace it
  [PMID:15901672, "Here we show that Bak is subject to a distinctive mode of
  regulation involving its direct sequestration by two of its prosurvival
  relatives"].
• Tightened broad GO:0006915 apoptotic process, GO:0043065 positive regulation of apoptotic process, GO:0046902 regulation of mitochondrial membrane permeability, GO:0097190 apoptotic signaling pathway, and ARBA
  regulation rows to GO:0097345 mitochondrial outer membrane permeabilization
  when the row was really describing BAK's direct pore-forming step.
• Removed the PAINT GO:0043066 negative regulation of apoptotic process,
  ComplexPortal GO:0090201 negative regulation of release of cytochrome c from mitochondria, and ComplexPortal GO:1901029 negative regulation of mitochondrial outer membrane permeabilization rows. BAK is the effector held
  in check by MCL1 or BCL-XL, not the anti-apoptotic activity in those
  inhibited complexes.
• Kept ER calcium, ER-stress apoptosis, IRE1/UPR, UV-response, and chaperone
  rows as non-core where they were compatible with BAK/BAX side biology. Removed
  readout-derived or phenotype-level projections such as positive regulation of
  proteolysis from a DEVDG caspase reporter PMID:18387192, cellular response
  to mechanical stimulus from a BAD prostate-cancer study PMID:19593445, and
  generic endocrine-pancreas, xenobiotic, ethanol, regeneration, transport, or
  intracellular-signaling electronic imports.
• Replaced the 1998 permeability-transition-era GO:0046930 pore complex,
  membrane-potential, and electrochemical-gradient rows with BAK complex or
  negative regulation of mitochondrial membrane potential. The historical paper
  reported recombinant Bax and Bak causing cytochrome-c release and loss of
  mitochondrial potential PMID:9843949, but BAK should not be curated today as
  a component of a permeability-transition pore complex.
• Left only the VDAC1 GO:0044325 transmembrane transporter binding row
  undecided because the cached K5/VDAC1 paper is abstract-only and does not
  expose the BAK-specific evidence.

2026-11-04 ORGANELLES ER batch 1 (endoplasmic reticulum rows)

• Rows: one GO:0005783 row, IEA GO_REF:0000107 (Ensembl projection from mouse Bak O08734), located_in. No ER sub-compartment row before this batch.
• Source check (QuickGO): mouse Bak O08734 carries GO:0005783 IDA PMID:12847083 (MGI). Full text cached. Endogenous Bak in the ER-enriched fraction PMID:12847083, calnexin probed PMID:12847083, fraction free of mitochondria PMID:12847083. ER-targeted Bak depletes ER Ca2+ PMID:12847083.
• Outcome: GO:0005783 row kept KEEP_AS_NON_CORE, reason extended. NEW IC GO:0005789 ER membrane added (tail-anchor TM 188-205 + endogenous light-membrane fraction + ER-face Ca2+ function; paralogs BAX and BCL2 carry GO:0005789 by IDA). MAM not asserted. core_functions unchanged.