Human LRIG2 was cloned as a 1065-amino-acid protein with the predicted type-I
single-pass organization expected for the LRIG family: signal peptide, an
extracellular region containing 15 LRRs and three immunoglobulin-like domains, a
transmembrane segment, and a cytoplasmic tail PMID:15145052.
Heterologously expressed human LRIG2 was N-glycosylated, and cell-surface
biotinylation plus microscopy detected it at the cell surface and in the
cytoplasm PMID:15145052. This is direct human-protein evidence, but the
cellular assays used heterologous expression rather than an endogenous normal
human tissue model.
In engineered human U87 and U251 glioblastoma cells, full-length LRIG2 and an
expressed LRIG2 ectodomain physically associated with EGFR PMID:25353163. Conditioned
medium contained soluble LRIG2 ectodomain PMID:25353163, but the paper states that the mechanism of release was not addressed;
therefore, release is supported, while proteolytic shedding of LRIG2 itself is
not established.
The same engineered glioblastoma systems linked LRIG2/ectodomain expression to
increased EGFR abundance and activation, PI3K/Akt signaling, proliferation, and
reduced apoptosis PMID:25353163.
The “in vivo” arm was a glioblastoma xenograft, so these findings support a
tumor-model function and do not by themselves establish normal human physiology.
Reciprocal co-immunoprecipitation in engineered U87 cells supports physical
association with PDGFRB PMID:30015847. LRIG2 manipulation
altered PDGFRB abundance/activation, downstream Akt and STAT3, cell-cycle
effectors, and PDGF-BB-dependent U87 proliferation PMID:30015847. Again, “in vivo” means U87
xenografts in mice, not an endogenous normal-human setting.
The cell context matters: the authors explicitly contrast their U87 result with
mouse embryonic fibroblasts, where Lrig2 did not change PDGFR abundance or
PDGFR/Akt phosphorylation PMID:30015847. A general LRIG2-to-PDGFRB
regulatory rule should therefore not be inferred beyond the tested tumor model.
The paper has two corrigenda. The 2019 notice states that "the name of the second
author was misspelt" and corrects only the author name PMID:30942453. The 2022
notice identifies "replacement data for the p-Akt and Cyclin D1 experiments" in
Figure 7 after the experiment was repeated PMID:35014687. It does not correct
the Figure 8 colocalization or reciprocal co-immunoprecipitation used for the
LRIG2-PDGFRB association, and the notice states that the main conclusions remain
supported.
Biallelic LRIG2 variants cause a subset of human urofacial syndrome
PMID:23313374. LRIG2 and heparanase-2 were also detected in nerve fascicles
between muscle bundles in human fetal bladder PMID:23313374. These observations establish
human developmental importance and a neural anatomical context, but not a
specific biochemical activity.
Mouse Lrig2 is broadly expressed in the developing inner ear and becomes
enhanced in neurons and sensory epithelia PMID:24086156. Lrig2-null mice retained normal
sound-detection sensitivity but had attenuated auditory neuronal responses
PMID:24086156. This supports same-ortholog transfer of an auditory role to
human LRIG2, with the necessary bounds that the experiment is mouse and the
single-mutant defect is response amplitude rather than hearing threshold.
In neuronal models, mouse Lrig2 binds Neogenin and prevents premature
ADAM17-mediated Neogenin ectodomain shedding PMID:26651291. This mechanism was linked to
RGMa-dependent neurite-growth inhibition and cortical-neuron migration
PMID:26651291. The cached source is abstract-only and does not provide
direct normal-human evidence, so it supports the mouse ortholog's neuronal
function rather than an unqualified human mechanism.
LRIG1, LRIG2, and LRIG3 cannot be treated as functionally interchangeable.
Mouse inner-ear genetics found both redundancy and independence, with Lrig2 the
most distinct family member PMID:24086156. The glioblastoma ectodomain paper likewise reports
that the LRIG2 ectodomain phenotype was opposite to LRIG1 PMID:25353163. LRIG1 receptor-regulatory
mechanisms and LRIG3 developmental effects should therefore not be transferred
to LRIG2 without same-ortholog or direct LRIG2 evidence.
The two human protein-interaction annotations in the current LRIG2 source set
come from targeted EGFR and PDGFRB studies, not from broad discovery screens.
Interaction evidence establishes specific physical association in the tested
glioblastoma systems; signaling direction and physiological relevance require
the separate perturbation evidence and remain model-bounded.