lag-2 (C. elegans) curation notes

UniProt P45442 "Protein lag-2". PANTHER (verbatim from UniProt): PTHR22669 (DELTA/SERRATE/LAG-2 DOMAIN
PROTEIN), PTHR22669:SF8 (PROTEIN LAG-2). IBA rows: PANTHER:PTN000505678 for Notch binding (GO:0005112),
cell projection membrane (GO:0031253) and cell fate specification (GO:0001708). Donors are lag-2 itself
(WB:WBGene00002246) plus WB:WBGene00001103 for Notch binding.

Ligand function

• Delta-like membrane protein expressed in the DTC PMID:7607081.
• Required for both LIN-12 and GLP-1 signaling PMID:1769331.
• ECD binds LIN-12 EGF repeats 1-6 (Y2H) PMID:18700817.
• The N-terminal region and DSL domain are required; the EGF repeats are dispensable; the intracellular
  region down-regulates activity PMID:9307971.
• APX-1 substitutes for LAG-2; truncated secreted ECDs activate LIN-12/GLP-1 ectopically PMID:8575327.

Variant biology

• Lacks a DOS motif. The DOS motif is proposed to be supplied in trans by OSM-11-family co-ligands
  PMID:18700817
• Germline stem-cell niche: LAG-2 on DTC processes PMID:16672375.
• AC/VU decision: feedback restricts lag-2 transcription to the presumptive AC (HLH-2 directly activates
  lag-2) PMID:14701877.

Flags

• GO:0032809 neuronal cell body membrane (IDA, PMID:16672375 and PMID:8575327): the cells assayed (DTC and
  gonadal cells) are not neurons, so I MODIFY this to plasma membrane.

Deep research status

The first falcon run (just deep-research-falcon worm lag-2 --fallback perplexity-lite) timed out at 600 s. The
perplexity fallback is unavailable in this environment. A rerun with --timeout 2400 succeeded:
lag-2-deep-research-falcon.md. It is consistent with the review and is cited in core_functions. Note that
falcon reports C. elegans LIN-12/GLP-1 are tuned to lower force thresholds for activation than Drosophila
Notch (Langridge et al. 2021 bioRxiv; not in the publications cache, so not used as evidence here).

Falcon also reports two points not verified against cached primary papers. (1) A soluble LAG-2 ectodomain
did not rescue lag-2 null lethality, whereas GPI tethering partly did (Post et al. 2025). This appears to be in
tension with PMID:8575327, where truncated secreted DSL forms substituted for lag-2. (2) The adhesion GPCR LAT-1
binds LAG-2 in cis on the DTC and enhances GLP-1 activation.