TNFRSF10B / DR5 notes

Core role

• TNFRSF10B is DR5/TRAIL-R2, the second signaling receptor for TNFSF10/TRAIL.
  It is a type I plasma-membrane TNF receptor with extracellular cysteine-rich
  TRAIL-binding repeats and a cytoplasmic death domain.
• The structural DR5 paper provides the clearest ligand-level evidence:
  Apo2L/TRAIL and the DR5 ectodomain form a 3:3 complex, and receptor
  ligation triggers apoptosis by oligomerizing intracellular death domains
  PMID:10549288.
• DR5 signaling is FADD dependent. The FADD-deficient MEF study showed that
  cells stably expressing human DR5 are resistant to TRAIL without FADD and
  regain sensitivity after FADD reconstitution PMID:10862756. Rows to
  generic apoptosis, regulation of apoptosis, death-domain-receptor apoptosis,
  and cell-surface signaling should therefore be tightened to
  GO:0036462 TRAIL-activated apoptotic signaling pathway.
• GO:0036463 TRAIL receptor activity is the best molecular-function term for
  DR5 because it captures both ligand recognition and receptor signaling.
  Generic death receptor activity, signaling receptor activity,
  TRAIL binding, and TNFSF10 protein binding rows should be folded into it.

Contextual and indirect rows

• DR5 is frequently an induced effector in stress-sensitization papers.
  CHOP/ER stress can enhance DR5 expression, and DR5 knockdown blocks
  thapsigargin-induced apoptosis in some carcinoma cells PMID:15322075.
  That supports an ER-stress-to-DR5 route, but the receptor itself is not an
  intrinsic ER-stress apoptotic effector, so the ER-stress rows are
  over-annotations.
• The PU.1 AML paper is similar: PU.1 supports TRAIL-mediated apoptosis partly
  by inducing DR5 expression PMID:28362429. The paper supports the
  TRAIL/DR5 axis but not the immune-process term
  GO:0002357 defense response to tumor cell on the tumor-cell receptor.
• The NF-kappaB rows are secondary. The 1997 DR5/DR4 paper supports
  contextual NF-kappaB activation PMID:9430227, but the pancreatic
  adenocarcinoma paper is mixed CD95/TRAIL-receptor evidence PMID:11464292,
  and the large-scale cDNA screen cache does not expose its TNFRSF10B-specific
  result PMID:12761501. Those latter two rows should remain unresolved from
  local evidence.
• The recurring mechanical-stimulus row has the same problem as in DR4. It
  derives from expression-pattern evidence in a BAD/prostate-cancer study
  PMID:19593445; altered DR5 expression during a stretch experiment does not
  make DR5 part of the mechanical-stimulus response.

Interaction imports

• Direct TNFSF10 GO:0005515 protein binding rows from the Apo2L/TRAIL:DR5
  crystal structure and the Shilts et al. surface-protein screen should be
  narrowed to GO:0036463 TRAIL receptor activity.
• TNFSF10 imports from TRAIL sensitization studies with resveratrol,
  bortezomib, HDAC inhibition, or V-ATPase inhibition should be left
  UNDECIDED where the local cache is abstract-only and does not expose the
  exact TNFRSF10B-TNFSF10 interaction assayed.
• The DDX3X, GSK3B, BIRC2, and FADD/CASP8 rows from the anti-apoptotic
  death-receptor complex paper should also be left UNDECIDED. The cache only
  has title-level evidence for an anti-apoptotic complex at death receptors, so
  it cannot verify the individual DR5 IntAct edges.
• HuRI imports for MUC1, ASPH, and ZDHHC15 are useful interaction-screen leads
  but have no DR5-specific apoptotic mechanism in the cached article
  PMID:32296183, so they should not remain as GO:0005515 annotations.
• The Shilts et al. surface-protein screen recovered the extracellular
  TNFSF10-TNFRSF10B interaction and is aligned with TRAIL receptor activity
  PMID:35922511.

Reactome and GO-CAM exports

• The TRAIL/TRAILR2 GO-CAM places TNFRSF10B in
  GO:0036462 TRAIL-activated apoptotic signaling pathway at the plasma
  membrane. The HPV E6/MARCHF8 GO-CAM uses the newer
  GO:0036463 TRAIL receptor activity term for TNFRSF10B at the plasma
  membrane while modelling viral down-regulation of TRAIL-R1/R2.
• Unlike TNFRSF10A, the Reactome procaspase-8 dimerization event
  R-HSA-141156 is directly DR5-specific: its summary places the event at a
  "TRAIL:TRAIL receptor-2:FADD" complex. It should therefore be accepted for
  TNFRSF10B.
• Reactome rows for TNFSF10 binding, receptor trimerization, FADD recruitment,
  procaspase-8/procaspase-10 recruitment, and DcR2 heteromerization are close
  enough to DR5 to keep as plasma-membrane assertions.
• Reactome exports from TP53 transcription of the TNFRSF10B gene and from
  downstream c-FLIP/CASP8 checkpoint events are too indirect for a
  TNFRSF10B component row, even though they are correctly inside broad TRAIL
  or death-receptor apoptosis models.