{"gene":"ADAM22","run_date":"2026-06-09T22:02:41","timeline":{"discoveries":[{"year":1998,"finding":"ADAM22 (MDC2) was cloned and found to possess a disrupted zinc-binding motif in its metalloproteinase-like domain, indicating it lacks proteolytic activity, and contains disintegrin-like and cysteine-rich sequences suggesting a role as an integrin ligand/adhesion molecule rather than a proteinase. Highly expressed in brain.","method":"cDNA cloning, sequence analysis, Northern blot","journal":"The Biochemical journal","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — sequence analysis with domain characterization confirmed by Northern blot, single lab but multiple methods","pmids":["9693107"],"is_preprint":false},{"year":2002,"finding":"The cytoplasmic tail of ADAM22 interacts with 14-3-3β; the major binding site was mapped to the last 28 amino acid residues of the ADAM22 cytoplasmic tail.","method":"Yeast two-hybrid, in vitro binding assay, co-immunoprecipitation","journal":"Science in China. Series C, Life sciences","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — reciprocal Co-IP and pull-down, single lab, binding site mapped","pmids":["18762889"],"is_preprint":false},{"year":2005,"finding":"ADAM22-deficient mice display severe ataxia, lethal convulsions, and marked peripheral nerve hypomyelination, establishing an essential in vivo role for ADAM22 in peripheral nervous system myelination and neuronal function.","method":"Gene targeting (knockout mice), histological analysis","journal":"BMC neuroscience","confidence":"High","confidence_rationale":"Tier 2 / Strong — clean KO with defined neurological and histological phenotype, replicated in subsequent studies","pmids":["15876356"],"is_preprint":false},{"year":2005,"finding":"Overexpression of ADAM22 in HEK293 cells enhances cell adhesion and spreading; this effect requires 14-3-3 binding motifs in the ADAM22 cytoplasmic tail, as a truncated ADAM22 lacking these motifs loses the adhesion-promoting activity. Multiple 14-3-3 family members interact with ADAM22cyt.","method":"Co-immunoprecipitation, in vitro pull-down, cell adhesion/spreading assays, overexpression in HEK293","journal":"Biochemical and biophysical research communications","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — functional adhesion assay plus Co-IP, single lab, domain-deletion approach","pmids":["15882968"],"is_preprint":false},{"year":2006,"finding":"ADAM22 serves as a postsynaptic receptor for the secreted epilepsy protein LGI1; LGI1 binding to ADAM22 enhances AMPA receptor-mediated synaptic transmission in hippocampal slices. Epilepsy-causing mutant LGI1 fails to bind ADAM22. ADAM22 is anchored at the postsynaptic density via scaffolds containing stargazin.","method":"Co-immunoprecipitation, cell surface binding, electrophysiology (hippocampal slice patch-clamp), rat brain biochemistry","journal":"Science (New York, N.Y.)","confidence":"High","confidence_rationale":"Tier 1-2 / Strong — multiple orthogonal methods (binding assay, electrophysiology, native brain biochemistry), widely replicated across labs","pmids":["16990550"],"is_preprint":false},{"year":2006,"finding":"ADAM22 inhibits cellular proliferation of glioma-derived astrocytes; this growth inhibition is mediated via the disintegrin domain interacting with integrins on the cell surface and can be overcome by overexpression of integrin-linked kinase.","method":"BrdU incorporation assay, GST fusion protein treatment, overexpression in glioma cells","journal":"Neurosurgery","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — functional proliferation assay plus domain-specific GST fusion rescue, single lab","pmids":["16385342"],"is_preprint":false},{"year":2006,"finding":"ADAM22 surface expression is regulated by phosphorylation-dependent interaction with 14-3-3 proteins; 14-3-3 proteins bind preferentially to the serine-phosphorylated precursor form of ADAM22 via two binding motifs (critical site at residues 831–834), masking ER retention signals and thereby promoting ADAM22 trafficking to the cell surface.","method":"Yeast two-hybrid, co-immunoprecipitation, ADAM22 point mutant analysis, cell surface localization assay","journal":"Journal of cell science","confidence":"High","confidence_rationale":"Tier 2 / Moderate — reciprocal Co-IP plus mutagenesis mapping plus localization readout, single lab but multiple orthogonal approaches","pmids":["16868027"],"is_preprint":false},{"year":2008,"finding":"LGI1 and LGI4 both bind to ADAM22 (as well as ADAM11 and ADAM23); LGI1 was identified as the most potent ADAM22-binding protein in mouse brain by immunoprecipitation/mass spectrometry, and binding specificity was confirmed by quantitative cell-ELISA.","method":"Immunoprecipitation, mass spectrometry, quantitative cell-ELISA","journal":"International journal of biological sciences","confidence":"High","confidence_rationale":"Tier 2 / Moderate — native brain IP/MS plus cell-ELISA, two orthogonal methods, single lab","pmids":["18974846"],"is_preprint":false},{"year":2008,"finding":"The ADAM22 pro domain (residues 26–199) is folded with secondary structure consisting predominantly of β-strands, and exists as two subdomains; it can be expressed as a stable, soluble protein suitable for structural studies.","method":"E. coli expression, CD spectroscopy, NMR spectroscopy","journal":"Protein expression and purification","confidence":"Medium","confidence_rationale":"Tier 1 / Weak — NMR and CD characterization of recombinant domain, single lab, partial structural characterization only","pmids":["18593599"],"is_preprint":false},{"year":2009,"finding":"Crystal structure of the full ectodomain of human ADAM22 reveals a compact four-leaf clover arrangement; the metalloproteinase-like domain is held in the concave face of a rigid module formed by the disintegrin, cysteine-rich, and EGF-like domains. Lack of metalloproteinase activity is structurally explained by absence of critical catalytic residues, filled substrate groove, and steric hindrance from the cysteine-rich domain. Three bound calcium ions were identified with regulatory (metalloproteinase-like domain) and structural (disintegrin domain) roles.","method":"X-ray crystallography, isothermal titration calorimetry","journal":"The Journal of biological chemistry","confidence":"High","confidence_rationale":"Tier 1 / Moderate — crystal structure plus calorimetric functional validation, single lab but rigorous methods","pmids":["19692335"],"is_preprint":false},{"year":2010,"finding":"ADAM22 is a component of the axonal Kv1 potassium channel complex at juxtaparanodes of myelinated axons; it co-immunoprecipitates with Kv1.2 and the MAGUKs PSD-93 and PSD-95. When co-expressed with MAGUKs in heterologous cells, ADAM22 recruits Kv1 channels into membrane surface clusters. In ADAM22-null mice, juxtaparanodal clustering of PSD-93 and PSD-95 is lost, whereas Kv1.2 and Caspr2 clustering is normal, demonstrating that ADAM22 is required for MAGUK recruitment to juxtaparanodes.","method":"Co-immunoprecipitation, mass spectrometry, immunofluorescence in null mice, heterologous co-expression clustering assay","journal":"The Journal of neuroscience","confidence":"High","confidence_rationale":"Tier 2 / Strong — native IP/MS, null mouse analysis, heterologous clustering assay — multiple orthogonal methods","pmids":["20089912"],"is_preprint":false},{"year":2010,"finding":"ADAM22 functions as a major neuronal/axonal receptor for Schwann cell-secreted LGI4 in peripheral nerve myelination; LGI4 binds directly to ADAM22 without requiring additional membrane-associated factors. Conditional ablation experiments showed Schwann cells are the principal source of LGI4 and that axonal ADAM22 is required for Schwann cell differentiation, revealing a paracrine Schwann cell–axon signaling axis.","method":"Direct binding assay (LGI4-ADAM22), cell type-specific conditional KO mice, heterotypic Schwann cell–neuron co-culture","journal":"The Journal of neuroscience","confidence":"High","confidence_rationale":"Tier 2 / Strong — direct binding assay plus cell type-specific KO with defined myelination phenotype, replicated conceptually across multiple studies","pmids":["20220021"],"is_preprint":false},{"year":2010,"finding":"Mutations in the disintegrin domain of ADAM22 cause marked decreases in proprotein processing and reduced LGI4-binding; the P81R polymorphic variant functions comparably to wild-type.","method":"Mutagenesis, cell surface expression assays, LGI4-binding assays in transfected cells","journal":"Journal of receptor and signal transduction research","confidence":"Medium","confidence_rationale":"Tier 2 / Weak — mutagenesis plus cell binding assay, single lab, single study","pmids":["20156119"],"is_preprint":false},{"year":2013,"finding":"LGI1 autoantibodies from limbic encephalitis patients target the EPTP repeat domain of LGI1 and block LGI1-ADAM22/ADAM23 interaction, reversibly reducing synaptic AMPA receptor clusters in hippocampal neurons. Addition of the soluble ADAM22 extracellular domain alone was sufficient to reduce synaptic AMPA receptors, confirming that disruption of LGI1-ADAM22 interaction reduces AMPA receptor function.","method":"ELISA arrays, co-immunoprecipitation, immunofluorescence in hippocampal neurons, soluble ectodomain competition assay","journal":"The Journal of neuroscience","confidence":"High","confidence_rationale":"Tier 2 / Strong — multiple methods (ELISA, Co-IP, neuronal imaging, ectodomain competition), replicated across patient samples and cell assays","pmids":["24227725"],"is_preprint":false},{"year":2015,"finding":"ADAM22 acts as the postsynaptic receptor through which the paracrine signal LGI1 sets postsynaptic strength; ADAM22 maintains excitatory synapses through PDZ domain interactions. Without LGI1, PSD-95 (but not SAP102) cannot modulate synaptic transmission, establishing LGI1-ADAM22 as a synaptic organizing complex that coordinates PSD-95-dependent synapse maturation.","method":"Electrophysiology (hippocampal slices/neurons), dominant-negative ADAM22 expressing constructs, genetic epistasis with PSD-95/SAP102","journal":"Proceedings of the National Academy of Sciences of the United States of America","confidence":"High","confidence_rationale":"Tier 2 / Moderate — electrophysiology plus genetic epistasis, multiple orthogonal approaches, single lab","pmids":["26178195"],"is_preprint":false},{"year":2016,"finding":"Compound heterozygous mutations in ADAM22 (p.Cys401Tyr and a frameshift p.Ser799IlefsTer96) in a patient with epileptic encephalopathy were shown to abolish LGI1 binding; additionally, the frameshift mutant ADAM22 also failed to bind PSD-95, establishing these interactions as functionally essential in vivo.","method":"Cell surface binding assay, co-immunoprecipitation in heterologous expression system","journal":"Neurology. Genetics","confidence":"Medium","confidence_rationale":"Tier 2 / Weak — co-IP and binding assay in heterologous cells, single lab, functional validation of patient variants","pmids":["27066583"],"is_preprint":false},{"year":2016,"finding":"Four secretion-positive LGI1 missense mutations (T380A, R407C, S473L, R474Q) linked to ADLTE significantly impair LGI1 interaction with both ADAM22 and ADAM23 on the cell surface without affecting LGI1 secretion or folding, demonstrating a second extracellular loss-of-function mechanism independent of secretion impairment.","method":"Co-immunoprecipitation, immunofluorescence, 3D protein modelling","journal":"PLoS genetics","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — Co-IP and immunofluorescence, single lab, consistent across four mutations","pmids":["27760137"],"is_preprint":false},{"year":2018,"finding":"Crystal structure of the human LGI1-ADAM22 complex at 2:2 heterotetrameric assembly revealed that the hydrophobic pocket of the C-terminal EPTP domain of LGI1 binds to the metalloprotease-like domain of ADAM22. LGI1-LGI1 dimerization is mediated by both LRR and EPTP domains. Pathogenic R474Q mutation disrupts LGI1-LGI1 interface and higher-order complex assembly in vitro and in a knock-in mouse model, without affecting secretion or ADAM22 binding.","method":"X-ray crystallography, in vitro complex assembly, mouse knock-in model","journal":"Nature communications","confidence":"High","confidence_rationale":"Tier 1 / Strong — crystal structure plus mutagenesis plus in vivo knock-in validation, multiple orthogonal methods","pmids":["29670100"],"is_preprint":false},{"year":2019,"finding":"LGI1 is recruited to the axon initial segment (AIS) where it co-localizes with ADAM22 and Kv1 channels in hippocampal neurons. ADAM22 and ADAM23 promote ER export and surface expression of LGI1 and co-transport LGI1 in axonal vesicles. ADLTE missense mutations S473L and R474Q in LGI1 prevent its association with ADAM22 and enrichment at the AIS.","method":"Live-cell imaging, immunofluorescence in cultured hippocampal neurons, co-immunoprecipitation","journal":"Journal of cell science","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — live imaging plus Co-IP, mutation functional analysis, single lab","pmids":["30598502"],"is_preprint":false},{"year":2021,"finding":"The LGI1-ADAM22-MAGUK complex governs transsynaptic nanoalignment of PSD-95 nanodomains, NMDA/AMPA receptors, Kv1 channels, and LRRTM4-Neurexin adhesion molecules. Adam22 knock-in mice devoid of the ADAM22-MAGUK (PDZ-binding) interaction develop lethal epilepsy with less-condensed PSD-95 nanodomains and decreased excitatory synaptic transmission. Without ADAM22 binding, PSD-95 cannot potentiate AMPA receptor-mediated transmission. Forced co-expression of ADAM22 and PSD-95 reconstitutes nano-condensates in non-neuronal cells.","method":"ADAM22 knock-in mice, super-resolution microscopy, electrophysiology, non-neuronal cell reconstitution assay","journal":"Proceedings of the National Academy of Sciences of the United States of America","confidence":"High","confidence_rationale":"Tier 1-2 / Strong — in vivo knock-in model, super-resolution imaging, electrophysiology, and reconstitution assay, multiple orthogonal methods","pmids":["33397806"],"is_preprint":false},{"year":2021,"finding":"Quantitative dual phosphorylation of ADAM22 by protein kinase A (PKA) mediates high-affinity binding to dimerized 14-3-3 proteins; this interaction protects LGI1-ADAM22 complexes from endocytosis-dependent degradation. Forskolin-induced PKA activation increases ADAM22 levels. Hypomorphic mouse series established that ~50% LGI1 and ~10% ADAM22 levels are sufficient to prevent lethal epilepsy. ADAM22 function is required in both excitatory and inhibitory neurons.","method":"Genetic and structural analysis, phosphorylation assays, PKA activation (forskolin), ADAM22/LGI1 hypomorphic mouse series","journal":"Cell reports","confidence":"High","confidence_rationale":"Tier 1-2 / Strong — biochemical phosphorylation mapping, structural analysis, multiple hypomorphic mouse lines, single lab with multiple orthogonal methods","pmids":["34910912"],"is_preprint":false},{"year":2022,"finding":"Functional studies of 19 additional biallelic ADAM22 variants revealed at least three distinct pathological mechanisms: (i) defective cell membrane expression, (ii) impaired LGI1-binding, and/or (iii) impaired interaction with PSD-95, confirming these three molecular interactions as essential for ADAM22 function in vivo.","method":"Heterologous expression, cell surface expression assays, co-immunoprecipitation binding assays","journal":"Brain : a journal of neurology","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — systematic functional analysis of multiple patient variants with binding and localization assays, single lab","pmids":["35813813"],"is_preprint":false},{"year":2023,"finding":"A missense variant p.S905F in the PDZ-binding motif of ADAM22 specifically impairs ADAM22 binding to PSD-95 and other MAGUKs without affecting biosynthesis, stability, or LGI1 interaction, causing focal epilepsy in homozygous individuals and demonstrating that the ADAM22-MAGUK interaction is independently essential for seizure protection in humans.","method":"Structural in silico analysis, protein-protein interaction assays in transfected mammalian cells, co-immunoprecipitation","journal":"Brain communications","confidence":"Medium","confidence_rationale":"Tier 2 / Weak — Co-IP in transfected cells with variant-specific functional dissection, single lab, supported by clinical genetics","pmids":["37953841"],"is_preprint":false},{"year":2025,"finding":"Cryo-EM structures of the LGI1-ADAM22 complex determined at 3.79 Å resolution revealed a 3:3 heterohexameric assembly (three LGI1 and three ADAM22 ECD molecules), distinct from the previously crystallized 2:2 tetramer. High-speed atomic force microscopy visualized the flexibility of the 3:3 complex in solution, providing new insights into higher-order transsynaptic assembly modes.","method":"Chemical cross-linking, cryo-EM structure determination, high-speed atomic force microscopy (HS-AFM), SAXS","journal":"eLife","confidence":"High","confidence_rationale":"Tier 1 / Moderate — cryo-EM structure at near-atomic resolution plus HS-AFM dynamic visualization, multiple biophysical methods, single lab","pmids":["40601686"],"is_preprint":false},{"year":2019,"finding":"LGI3 associates with ADAM22 in LPS-stimulated human keratinocytes, as shown by co-immunoprecipitation, flow cytometry, and immunocytochemistry, suggesting ADAM22 can function as an LGI3 receptor outside the nervous system.","method":"Co-immunoprecipitation, flow cytometry, immunocytochemistry","journal":"Cytokine","confidence":"Low","confidence_rationale":"Tier 3 / Weak — single lab, Co-IP in keratinocytes, no functional consequence of ADAM22 binding established","pmids":["31627033"],"is_preprint":false}],"current_model":"ADAM22 is a catalytically inactive transmembrane ADAM family receptor, highly expressed in the nervous system, that functions as the postsynaptic receptor for the secreted epilepsy protein LGI1 (binding via LGI1's EPTP domain to the ADAM22 metalloprotease-like domain in a 2:2 or 3:3 hetero-oligomeric transsynaptic complex); through its cytoplasmic PDZ-binding motif, ADAM22 recruits the MAGUK scaffolding proteins PSD-95/PSD-93 to organize postsynaptic nanodomains containing AMPA/NMDA receptors and Kv1 channels, thereby regulating excitatory synaptic transmission and epilepsy prevention. ADAM22 surface levels are stabilized by PKA-dependent phosphorylation that promotes high-affinity 14-3-3 binding, protecting the LGI1-ADAM22 complex from endocytic degradation. In the peripheral nervous system, axonal ADAM22 acts as the receptor for Schwann cell-secreted LGI4, driving peripheral nerve myelination through paracrine signaling. Loss-of-function mutations or autoantibodies disrupting any of ADAM22's three key interactions—LGI1 binding, MAGUK recruitment, or surface expression—cause epileptic encephalopathy in humans and mice."},"narrative":{"mechanistic_narrative":"ADAM22 is a catalytically inactive transmembrane member of the ADAM family, highly expressed in brain, that functions as a synaptic and axonal receptor coupling secreted LGI proteins to intracellular scaffolds rather than as a proteinase [PMID:9693107, PMID:16990550]. Its metalloproteinase-like domain lacks critical catalytic residues and is held in a rigid four-leaf-clover ectodomain by the disintegrin, cysteine-rich, and EGF-like domains, structurally explaining the absence of protease activity [PMID:19692335]. ADAM22 is the principal postsynaptic receptor for the secreted epilepsy protein LGI1, which binds through its EPTP domain to the ADAM22 metalloprotease-like domain to form a higher-order transsynaptic assembly seen as both a 2:2 heterotetramer and a 3:3 heterohexamer [PMID:16990550, PMID:29670100, PMID:40601686]. Through its cytoplasmic PDZ-binding motif ADAM22 recruits the MAGUK scaffolds PSD-95/PSD-93, organizing transsynaptic PSD-95 nanodomains that align AMPA/NMDA receptors and Kv1 channels and thereby setting excitatory synaptic strength [PMID:20089912, PMID:26178195, PMID:33397806]. Surface levels of ADAM22 are controlled by PKA-dependent dual phosphorylation that drives high-affinity 14-3-3 binding, masking ER-retention signals to promote trafficking and protecting LGI1-ADAM22 complexes from endocytic degradation [PMID:16868027, PMID:34910912]. In the peripheral nervous system, axonal ADAM22 acts as the receptor for Schwann cell-secreted LGI4 to drive a paracrine signaling axis required for myelination, and ADAM22-null mice show ataxia, lethal seizures, and peripheral nerve hypomyelination [PMID:15876356, PMID:20220021]. Biallelic loss-of-function variants and autoantibodies that disrupt any of ADAM22's three essential interactions—LGI1 binding, MAGUK recruitment, or surface expression—cause human epileptic encephalopathy [PMID:27066583, PMID:35813813, PMID:37953841, PMID:24227725].","teleology":[{"year":1998,"claim":"Established that ADAM22 is structurally a non-proteolytic ADAM, redirecting expectation from enzyme to adhesion/receptor function and flagging its brain enrichment.","evidence":"cDNA cloning, sequence analysis, and Northern blot","pmids":["9693107"],"confidence":"Medium","gaps":["Sequence inference of catalytic inactivity not yet structurally confirmed","No binding partners or in vivo function identified"]},{"year":2002,"claim":"Identified the first cytoplasmic partner, 14-3-3, mapping the interaction to the C-terminal tail and opening the question of how the tail regulates ADAM22 behavior.","evidence":"Yeast two-hybrid, in vitro binding, co-immunoprecipitation","pmids":["18762889"],"confidence":"Medium","gaps":["Functional consequence of 14-3-3 binding not established","Phosphorylation dependence not yet defined"]},{"year":2005,"claim":"Knockout established ADAM22 as essential in vivo for peripheral nerve myelination and neuronal function, demonstrating non-redundant physiological roles.","evidence":"Gene-targeted knockout mice with histological analysis","pmids":["15876356"],"confidence":"High","gaps":["Molecular ligand and downstream pathway driving myelination unknown","Mechanism linking loss to seizures not defined"]},{"year":2005,"claim":"Connected the cytoplasmic 14-3-3 motifs to a cellular output by showing they are required for ADAM22-driven cell adhesion and spreading.","evidence":"Co-IP, pull-down, and adhesion/spreading assays with tail-truncation in HEK293","pmids":["15882968"],"confidence":"Medium","gaps":["Overexpression in non-neuronal cells may not reflect synaptic role","Adhesion mechanism not linked to physiological ligand"]},{"year":2006,"claim":"Defined ADAM22 as the postsynaptic receptor for LGI1 and linked the interaction to enhanced AMPA receptor transmission, establishing the core epilepsy-relevant signaling axis.","evidence":"Co-IP, surface binding, hippocampal slice electrophysiology, brain biochemistry","pmids":["16990550"],"confidence":"High","gaps":["Structural basis of LGI1 binding not resolved","How surface trafficking is controlled not addressed"]},{"year":2006,"claim":"Explained how ADAM22 reaches the cell surface, showing phosphorylation-dependent 14-3-3 binding to the precursor masks ER-retention signals to promote trafficking.","evidence":"Yeast two-hybrid, Co-IP, point mutant and surface-localization assays","pmids":["16868027"],"confidence":"High","gaps":["Kinase responsible for the phosphorylation not identified at this stage","Link between trafficking control and complex stability unaddressed"]},{"year":2008,"claim":"Demonstrated specificity and breadth of ligand engagement, identifying LGI1 as the most potent ADAM22-binding protein in brain while showing LGI4 also binds.","evidence":"Native brain IP/mass spectrometry and quantitative cell-ELISA","pmids":["18974846"],"confidence":"High","gaps":["Functional distinction between LGI1 and LGI4 signaling not resolved","Stoichiometry of complexes unknown"]},{"year":2009,"claim":"Provided the structural explanation for catalytic inactivity, showing a rigid ectodomain with a degraded, sterically blocked metalloprotease groove and bound calcium ions.","evidence":"X-ray crystallography of the human ectodomain plus ITC","pmids":["19692335"],"confidence":"High","gaps":["Ligand-bound structure not yet determined","Functional role of calcium ions inferred, not tested"]},{"year":2010,"claim":"Placed ADAM22 in the axonal Kv1 channel complex at juxtaparanodes and showed it is required for MAGUK recruitment, defining a scaffold-organizing role beyond the synapse.","evidence":"Native IP/MS, immunofluorescence in null mice, heterologous clustering assay","pmids":["20089912"],"confidence":"High","gaps":["Direct ADAM22-MAGUK binding interface not mapped here","Relationship to LGI1 signaling at juxtaparanodes unclear"]},{"year":2010,"claim":"Established ADAM22 as the neuronal receptor for Schwann cell-secreted LGI4, defining a paracrine axon-glia myelination axis and explaining the knockout hypomyelination.","evidence":"Direct LGI4-ADAM22 binding, cell-type-specific conditional KO, Schwann-neuron co-culture","pmids":["20220021"],"confidence":"High","gaps":["Downstream signaling from axonal ADAM22 into Schwann cell differentiation undefined","Cytoplasmic effectors in myelination not identified"]},{"year":2010,"claim":"Showed disintegrin-domain mutations impair proprotein processing and LGI4 binding, tying ADAM22 maturation to ligand engagement.","evidence":"Mutagenesis with surface expression and LGI4-binding assays in transfected cells","pmids":["20156119"],"confidence":"Medium","gaps":["Single study without in vivo validation","P81R variant effect on disease not established"]},{"year":2013,"claim":"Defined an autoimmune mechanism of disease by showing limbic-encephalitis LGI1 autoantibodies block LGI1-ADAM22 binding and reduce synaptic AMPA receptors.","evidence":"ELISA arrays, Co-IP, hippocampal neuron imaging, soluble ectodomain competition","pmids":["24227725"],"confidence":"High","gaps":["Whether antibody effects are fully reversible in vivo not shown","Contribution of ADAM23 versus ADAM22 not separated"]},{"year":2015,"claim":"Established the LGI1-ADAM22 complex as a synaptic organizer working through PDZ interactions, with PSD-95 unable to potentiate transmission without ADAM22.","evidence":"Hippocampal electrophysiology, dominant-negative ADAM22, genetic epistasis with PSD-95/SAP102","pmids":["26178195"],"confidence":"High","gaps":["Nanoscale organization of the complex not yet visualized","Specificity for PSD-95 over other MAGUKs incompletely defined"]},{"year":2016,"claim":"Validated the three ADAM22 interactions as physiologically essential by showing human compound heterozygous mutations abolish LGI1 and PSD-95 binding in encephalopathy.","evidence":"Surface binding and Co-IP of patient variants in heterologous cells","pmids":["27066583"],"confidence":"Medium","gaps":["Heterologous-cell assays do not capture neuronal context","Single patient genotype"]},{"year":2016,"claim":"Identified a secretion-independent LGI1 loss-of-function mechanism in which ADLTE missense mutations impair binding to ADAM22/ADAM23.","evidence":"Co-IP, immunofluorescence, 3D modelling of four mutations","pmids":["27760137"],"confidence":"Medium","gaps":["Structural basis of binding loss inferred from modelling only","In vivo consequence not tested here"]},{"year":2018,"claim":"Resolved the molecular architecture of the LGI1-ADAM22 complex, showing the EPTP domain engages the metalloprotease-like domain in a 2:2 assembly and that pathogenic R474Q disrupts higher-order assembly.","evidence":"X-ray crystallography, in vitro assembly, knock-in mouse","pmids":["29670100"],"confidence":"High","gaps":["Only the 2:2 assembly captured","Transsynaptic geometry across the cleft not directly imaged"]},{"year":2019,"claim":"Showed ADAM22 (with ADAM23) chaperones LGI1 ER export, surface expression, axonal transport, and AIS enrichment, linking receptor function to ligand delivery.","evidence":"Live-cell imaging, immunofluorescence, Co-IP in hippocampal neurons","pmids":["30598502"],"confidence":"Medium","gaps":["Relative contributions of ADAM22 vs ADAM23 not separated","Single-lab imaging study"]},{"year":2021,"claim":"Demonstrated the LGI1-ADAM22-MAGUK complex drives transsynaptic nanoalignment of PSD-95, glutamate receptors, Kv1, and adhesion molecules, with loss of the ADAM22-MAGUK interaction causing lethal epilepsy.","evidence":"ADAM22 PDZ knock-in mice, super-resolution microscopy, electrophysiology, non-neuronal reconstitution","pmids":["33397806"],"confidence":"High","gaps":["Biophysical driver of nano-condensation not fully defined","Inhibitory-neuron contribution addressed only later"]},{"year":2021,"claim":"Defined how surface ADAM22 is stabilized, showing PKA dual phosphorylation drives high-affinity dimeric 14-3-3 binding that protects complexes from endocytic degradation, and set the dosage thresholds for seizure protection.","evidence":"Phosphorylation/structural analysis, forskolin PKA activation, hypomorphic mouse series","pmids":["34910912"],"confidence":"High","gaps":["Endocytic machinery acting on dephosphorylated ADAM22 not identified","Upstream signals controlling PKA activity in neurons unknown"]},{"year":2022,"claim":"Systematically confirmed three distinct pathological mechanisms—defective surface expression, impaired LGI1 binding, impaired PSD-95 binding—across 19 biallelic ADAM22 variants.","evidence":"Heterologous expression, surface expression and Co-IP binding assays","pmids":["35813813"],"confidence":"Medium","gaps":["Genotype-phenotype correlation in patients not fully resolved","Assays in heterologous cells only"]},{"year":2023,"claim":"Isolated the ADAM22-MAGUK interaction as independently essential in humans, showing a PDZ-motif variant impairs MAGUK binding without affecting LGI1 binding or stability.","evidence":"In silico structural analysis and interaction/Co-IP assays of the p.S905F variant","pmids":["37953841"],"confidence":"Medium","gaps":["Single variant in transfected cells","No in vivo model for this allele"]},{"year":2025,"claim":"Revealed an alternative 3:3 heterohexameric LGI1-ADAM22 assembly and its conformational flexibility, expanding the model of higher-order transsynaptic organization.","evidence":"Cross-linking, cryo-EM at 3.79 Å, HS-AFM, SAXS","pmids":["40601686"],"confidence":"High","gaps":["Functional significance of 2:2 versus 3:3 assemblies in vivo unresolved","How assembly state maps onto nanodomain organization unknown"]},{"year":null,"claim":"It remains unknown how the intracellular signaling downstream of axonal ADAM22-LGI4 engagement drives Schwann cell differentiation, and how assembly stoichiometry (2:2 vs 3:3) is selected and tuned at native synapses.","evidence":"No discovery in the timeline defines the cytoplasmic effectors of ADAM22 in myelination or resolves in vivo assembly-state regulation","pmids":[],"confidence":"Medium","gaps":["No downstream myelination effector identified","Physiological control of complex stoichiometry undefined"]}],"mechanism_profile":{"molecular_activity":[{"term_id":"GO:0048018","term_label":"receptor ligand activity","supporting_discovery_ids":[4,7,11]},{"term_id":"GO:0060089","term_label":"molecular transducer activity","supporting_discovery_ids":[4,11,14]},{"term_id":"GO:0060090","term_label":"molecular adaptor activity","supporting_discovery_ids":[10,14,19]},{"term_id":"GO:0098631","term_label":"cell adhesion mediator activity","supporting_discovery_ids":[3,5]}],"localization":[{"term_id":"GO:0005886","term_label":"plasma membrane","supporting_discovery_ids":[6,4,10]},{"term_id":"GO:0005783","term_label":"endoplasmic reticulum","supporting_discovery_ids":[6,18]}],"pathway":[{"term_id":"R-HSA-112316","term_label":"Neuronal System","supporting_discovery_ids":[4,14,19]},{"term_id":"R-HSA-1266738","term_label":"Developmental Biology","supporting_discovery_ids":[2,11]},{"term_id":"R-HSA-1643685","term_label":"Disease","supporting_discovery_ids":[13,15,21,22]}],"complexes":["LGI1-ADAM22 transsynaptic complex","ADAM22-PSD-95/PSD-93 (MAGUK) postsynaptic complex","axonal Kv1 channel juxtaparanodal complex"],"partners":["LGI1","LGI4","PSD-95","PSD-93","14-3-3","KV1.2","LGI3","ADAM23"],"other_free_text":[]}},"prefetch_data":{"uniprot":{"accession":"Q9P0K1","full_name":"Disintegrin and metalloproteinase domain-containing protein 22","aliases":["Metalloproteinase-disintegrin ADAM22-3","Metalloproteinase-like, disintegrin-like, and cysteine-rich protein 2"],"length_aa":906,"mass_kda":100.4,"function":"Probable ligand for integrin in the brain. This is a non catalytic metalloprotease-like protein (PubMed:19692335). Involved in regulation of cell adhesion and spreading and in inhibition of cell proliferation. Neuronal receptor for LGI1","subcellular_location":"Cell membrane; Cell projection, axon","url":"https://www.uniprot.org/uniprotkb/Q9P0K1/entry"},"depmap":{"release":"DepMap","has_data":true,"is_common_essential":false,"resolved_as":"","url":"https://depmap.org/portal/gene/ADAM22","classification":"Not Classified","n_dependent_lines":1,"n_total_lines":1208,"dependency_fraction":0.0008278145695364238},"opencell":{"profiled":false,"resolved_as":"","ensg_id":"","cell_line_id":"","localizations":[],"interactors":[{"gene":"CANX","stoichiometry":0.2},{"gene":"TUBB4B","stoichiometry":0.2}],"url":"https://opencell.sf.czbiohub.org/search/ADAM22","total_profiled":1310},"omim":[{"mim_id":"621475","title":"DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 121; DEE121","url":"https://www.omim.org/entry/621475"},{"mim_id":"617933","title":"DEVELOPMENTAL AND EPILEPTIC ENCEPHALOPATHY 61; DEE61","url":"https://www.omim.org/entry/617933"},{"mim_id":"608303","title":"LEUCINE-RICH GENE, GLIOMA-INACTIVATED, 4; LGI4","url":"https://www.omim.org/entry/608303"},{"mim_id":"608302","title":"LEUCINE-RICH GENE, GLIOMA-INACTIVATED, 3; LGI3","url":"https://www.omim.org/entry/608302"},{"mim_id":"604619","title":"LEUCINE-RICH GENE, GLIOMA-INACTIVATED, 1; LGI1","url":"https://www.omim.org/entry/604619"}],"hpa":{"profiled":true,"resolved_as":"","reliability":"Approved","locations":[{"location":"Cell Junctions","reliability":"Approved"}],"tissue_specificity":"Tissue enriched","tissue_distribution":"Detected in many","driving_tissues":[{"tissue":"brain","ntpm":48.2}],"url":"https://www.proteinatlas.org/search/ADAM22"},"hgnc":{"alias_symbol":["MDC2"],"prev_symbol":[]},"alphafold":{"accession":"Q9P0K1","domains":[{"cath_id":"3.40.390.10","chopping":"48-193_234-429","consensus_level":"medium","plddt":84.4978,"start":48,"end":429},{"cath_id":"4.10.70.10","chopping":"446-516","consensus_level":"medium","plddt":87.2155,"start":446,"end":516},{"cath_id":"-","chopping":"534-677","consensus_level":"high","plddt":93.7686,"start":534,"end":677}],"viewer_url":"https://alphafold.ebi.ac.uk/entry/Q9P0K1","model_url":"https://alphafold.ebi.ac.uk/files/AF-Q9P0K1-F1-model_v6.cif","pae_url":"https://alphafold.ebi.ac.uk/files/AF-Q9P0K1-F1-predicted_aligned_error_v6.png","plddt_mean":73.19},"mouse_models":{"mgi_url":"https://www.informatics.jax.org/marker/summary?nomen=ADAM22","jax_strain_url":"https://www.jax.org/strain/search?query=ADAM22"},"sequence":{"accession":"Q9P0K1","fasta_url":"https://rest.uniprot.org/uniprotkb/Q9P0K1.fasta","uniprot_url":"https://www.uniprot.org/uniprotkb/Q9P0K1/entry","alphafold_viewer_url":"https://alphafold.ebi.ac.uk/entry/Q9P0K1"}},"corpus_meta":[{"pmid":"16990550","id":"PMC_16990550","title":"Epilepsy-related ligand/receptor complex LGI1 and ADAM22 regulate synaptic transmission.","date":"2006","source":"Science (New York, N.Y.)","url":"https://pubmed.ncbi.nlm.nih.gov/16990550","citation_count":311,"is_preprint":false},{"pmid":"24227725","id":"PMC_24227725","title":"Autoantibodies to epilepsy-related LGI1 in limbic encephalitis neutralize LGI1-ADAM22 interaction and reduce synaptic AMPA receptors.","date":"2013","source":"The Journal of neuroscience : the official journal of the Society for Neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/24227725","citation_count":269,"is_preprint":false},{"pmid":"15876356","id":"PMC_15876356","title":"Ataxia and peripheral nerve hypomyelination in ADAM22-deficient mice.","date":"2005","source":"BMC neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/15876356","citation_count":132,"is_preprint":false},{"pmid":"20089912","id":"PMC_20089912","title":"ADAM22, a Kv1 channel-interacting protein, recruits membrane-associated guanylate kinases to juxtaparanodes of myelinated axons.","date":"2010","source":"The Journal of neuroscience : the official journal of the Society for Neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/20089912","citation_count":108,"is_preprint":false},{"pmid":"18974846","id":"PMC_18974846","title":"LGI1 and LGI4 bind to ADAM22, ADAM23 and ADAM11.","date":"2008","source":"International journal of biological sciences","url":"https://pubmed.ncbi.nlm.nih.gov/18974846","citation_count":95,"is_preprint":false},{"pmid":"26178195","id":"PMC_26178195","title":"The LGI1-ADAM22 protein complex directs synapse maturation through regulation of PSD-95 function.","date":"2015","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/26178195","citation_count":82,"is_preprint":false},{"pmid":"20220021","id":"PMC_20220021","title":"Adam22 is a major neuronal receptor for Lgi4-mediated Schwann cell signaling.","date":"2010","source":"The Journal of neuroscience : the official journal of the Society for Neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/20220021","citation_count":81,"is_preprint":false},{"pmid":"9693107","id":"PMC_9693107","title":"Metalloproteinase-like, disintegrin-like, cysteine-rich proteins MDC2 and MDC3: novel human cellular disintegrins highly expressed in the brain.","date":"1998","source":"The Biochemical journal","url":"https://pubmed.ncbi.nlm.nih.gov/9693107","citation_count":74,"is_preprint":false},{"pmid":"33397806","id":"PMC_33397806","title":"LGI1-ADAM22-MAGUK configures transsynaptic nanoalignment for synaptic transmission and epilepsy prevention.","date":"2021","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/33397806","citation_count":71,"is_preprint":false},{"pmid":"29670100","id":"PMC_29670100","title":"Structural basis of epilepsy-related ligand-receptor complex LGI1-ADAM22.","date":"2018","source":"Nature communications","url":"https://pubmed.ncbi.nlm.nih.gov/29670100","citation_count":61,"is_preprint":false},{"pmid":"21297989","id":"PMC_21297989","title":"Primary human mDC1, mDC2, and pDC dendritic cells are differentially infected and activated by respiratory syncytial virus.","date":"2011","source":"PloS one","url":"https://pubmed.ncbi.nlm.nih.gov/21297989","citation_count":61,"is_preprint":false},{"pmid":"10433968","id":"PMC_10433968","title":"Cloning and chromosomal mapping of mouse ADAM11, ADAM22 and ADAM23.","date":"1999","source":"Gene","url":"https://pubmed.ncbi.nlm.nih.gov/10433968","citation_count":55,"is_preprint":false},{"pmid":"19692335","id":"PMC_19692335","title":"Structural characterization of the ectodomain of a disintegrin and metalloproteinase-22 (ADAM22), a neural adhesion receptor instead of metalloproteinase: insights on ADAM function.","date":"2009","source":"The Journal of biological chemistry","url":"https://pubmed.ncbi.nlm.nih.gov/19692335","citation_count":54,"is_preprint":false},{"pmid":"30278449","id":"PMC_30278449","title":"miR-449a Suppresses Tamoxifen Resistance in Human Breast Cancer Cells by Targeting ADAM22.","date":"2018","source":"Cellular physiology and biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology","url":"https://pubmed.ncbi.nlm.nih.gov/30278449","citation_count":47,"is_preprint":false},{"pmid":"22072566","id":"PMC_22072566","title":"Global characterization of the SRC-1 transcriptome identifies ADAM22 as an ER-independent mediator of endocrine-resistant breast cancer.","date":"2011","source":"Cancer research","url":"https://pubmed.ncbi.nlm.nih.gov/22072566","citation_count":45,"is_preprint":false},{"pmid":"30598502","id":"PMC_30598502","title":"ADAM22 and ADAM23 modulate the targeting of the Kv1 channel-associated protein LGI1 to the axon initial segment.","date":"2019","source":"Journal of cell science","url":"https://pubmed.ncbi.nlm.nih.gov/30598502","citation_count":37,"is_preprint":false},{"pmid":"16385342","id":"PMC_16385342","title":"ADAM22, expressed in normal brain but not in high-grade gliomas, inhibits cellular proliferation via the disintegrin domain.","date":"2006","source":"Neurosurgery","url":"https://pubmed.ncbi.nlm.nih.gov/16385342","citation_count":36,"is_preprint":false},{"pmid":"31432233","id":"PMC_31432233","title":"Insights into the mechanisms of epilepsy from structural biology of LGI1-ADAM22.","date":"2019","source":"Cellular and molecular life sciences : CMLS","url":"https://pubmed.ncbi.nlm.nih.gov/31432233","citation_count":34,"is_preprint":false},{"pmid":"27066583","id":"PMC_27066583","title":"Dysfunctional ADAM22 implicated in progressive encephalopathy with cortical atrophy and epilepsy.","date":"2016","source":"Neurology. Genetics","url":"https://pubmed.ncbi.nlm.nih.gov/27066583","citation_count":33,"is_preprint":false},{"pmid":"27717669","id":"PMC_27717669","title":"The LGI1-ADAM22 protein complex in synaptic transmission and synaptic disorders.","date":"2016","source":"Neuroscience research","url":"https://pubmed.ncbi.nlm.nih.gov/27717669","citation_count":31,"is_preprint":false},{"pmid":"34089731","id":"PMC_34089731","title":"Trans-synaptic LGI1-ADAM22-MAGUK in AMPA and NMDA receptor regulation.","date":"2021","source":"Neuropharmacology","url":"https://pubmed.ncbi.nlm.nih.gov/34089731","citation_count":28,"is_preprint":false},{"pmid":"18440780","id":"PMC_18440780","title":"Autosomal dominant lateral temporal epilepsy: absence of mutations in ADAM22 and Kv1 channel genes encoding LGI1-associated proteins.","date":"2008","source":"Epilepsy research","url":"https://pubmed.ncbi.nlm.nih.gov/18440780","citation_count":27,"is_preprint":false},{"pmid":"15882968","id":"PMC_15882968","title":"ADAM22 plays an important role in cell adhesion and spreading with the assistance of 14-3-3.","date":"2005","source":"Biochemical and biophysical research communications","url":"https://pubmed.ncbi.nlm.nih.gov/15882968","citation_count":24,"is_preprint":false},{"pmid":"17681454","id":"PMC_17681454","title":"Absence of mutations in the LGI1 receptor ADAM22 gene in autosomal dominant lateral temporal epilepsy.","date":"2007","source":"Epilepsy research","url":"https://pubmed.ncbi.nlm.nih.gov/17681454","citation_count":24,"is_preprint":false},{"pmid":"16868027","id":"PMC_16868027","title":"Efficient ADAM22 surface expression is mediated by phosphorylation-dependent interaction with 14-3-3 protein family members.","date":"2006","source":"Journal of cell science","url":"https://pubmed.ncbi.nlm.nih.gov/16868027","citation_count":23,"is_preprint":false},{"pmid":"35373813","id":"PMC_35373813","title":"Biallelic ADAM22 pathogenic variants cause progressive encephalopathy and infantile-onset refractory epilepsy.","date":"2022","source":"Brain : a journal of neurology","url":"https://pubmed.ncbi.nlm.nih.gov/35373813","citation_count":19,"is_preprint":false},{"pmid":"27760137","id":"PMC_27760137","title":"Secretion-Positive LGI1 Mutations Linked to Lateral Temporal Epilepsy Impair Binding to ADAM22 and ADAM23 Receptors.","date":"2016","source":"PLoS genetics","url":"https://pubmed.ncbi.nlm.nih.gov/27760137","citation_count":19,"is_preprint":false},{"pmid":"34910912","id":"PMC_34910912","title":"14-3-3 proteins stabilize LGI1-ADAM22 levels to regulate seizure thresholds in mice.","date":"2021","source":"Cell reports","url":"https://pubmed.ncbi.nlm.nih.gov/34910912","citation_count":15,"is_preprint":false},{"pmid":"33208158","id":"PMC_33208158","title":"ADAM22/LGI1 complex as a new actionable target for breast cancer brain metastasis.","date":"2020","source":"BMC medicine","url":"https://pubmed.ncbi.nlm.nih.gov/33208158","citation_count":15,"is_preprint":false},{"pmid":"20156119","id":"PMC_20156119","title":"Biological characterization of ADAM22 variants reveals the importance of a disintegrin domain sequence in cell surface expression.","date":"2010","source":"Journal of receptor and signal transduction research","url":"https://pubmed.ncbi.nlm.nih.gov/20156119","citation_count":6,"is_preprint":false},{"pmid":"31627033","id":"PMC_31627033","title":"LGI3 is secreted and binds to ADAM22 via TRIF-dependent NF-κB pathway in response to LPS in human keratinocytes.","date":"2019","source":"Cytokine","url":"https://pubmed.ncbi.nlm.nih.gov/31627033","citation_count":6,"is_preprint":false},{"pmid":"18593599","id":"PMC_18593599","title":"Expression, purification and insights into structure and folding of the ADAM22 pro domain.","date":"2008","source":"Protein expression and purification","url":"https://pubmed.ncbi.nlm.nih.gov/18593599","citation_count":4,"is_preprint":false},{"pmid":"18762889","id":"PMC_18762889","title":"The interaction between ADAM22 and 14-3-3beta.","date":"2002","source":"Science in China. Series C, Life sciences","url":"https://pubmed.ncbi.nlm.nih.gov/18762889","citation_count":3,"is_preprint":false},{"pmid":"40455867","id":"PMC_40455867","title":"Biallelic LGI1 and ADAM23 variants cause hippocampal epileptic encephalopathy via the LGI1-ADAM22/23 pathway.","date":"2025","source":"Brain : a journal of neurology","url":"https://pubmed.ncbi.nlm.nih.gov/40455867","citation_count":1,"is_preprint":false},{"pmid":"40461920","id":"PMC_40461920","title":"ADAM22 enhances lymphatic metastasis and epithelial-mesenchymal transition in head and neck squamous cell carcinoma cells through integrin signaling.","date":"2025","source":"Medical oncology (Northwood, London, England)","url":"https://pubmed.ncbi.nlm.nih.gov/40461920","citation_count":1,"is_preprint":false},{"pmid":"40601686","id":"PMC_40601686","title":"Structural insights into heterohexameric assembly of epilepsy-related ligand-receptor complex LGI1-ADAM22.","date":"2025","source":"eLife","url":"https://pubmed.ncbi.nlm.nih.gov/40601686","citation_count":1,"is_preprint":false},{"pmid":"37953841","id":"PMC_37953841","title":"ADAM22 ethnic-specific variant reducing binding of membrane-associated guanylate kinases causes focal epilepsy and behavioural disorder.","date":"2023","source":"Brain communications","url":"https://pubmed.ncbi.nlm.nih.gov/37953841","citation_count":0,"is_preprint":false},{"pmid":"41000458","id":"PMC_41000458","title":"Whole Exome Sequencing Identifies Novel Homozygous LGI1 Variant Mimicking ADAM22-Related Pathologies in a Moroccan Family.","date":"2025","source":"BMJ neurology open","url":"https://pubmed.ncbi.nlm.nih.gov/41000458","citation_count":0,"is_preprint":false},{"pmid":"42088439","id":"PMC_42088439","title":"CAF-derived exosomal circMPP6 drives ovarian cancer metastasis by coordinating nuclear and cytoplasmic regulation of ADAM22 to activate TGF-β/Smad signaling.","date":"2026","source":"International journal of biological sciences","url":"https://pubmed.ncbi.nlm.nih.gov/42088439","citation_count":0,"is_preprint":false},{"pmid":null,"id":"bio_10.1101_2025.01.06.631603","title":"Structural insights into heterohexameric assembly of epilepsy-related ligand–receptor complex LGI1–ADAM22","date":"2025-01-08","source":"bioRxiv","url":"https://doi.org/10.1101/2025.01.06.631603","citation_count":0,"is_preprint":true}],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":21594,"output_tokens":6122,"usd":0.078306,"stage1_stop_reason":"end_turn"},"stage2":{"model":"claude-opus-4-8","input_tokens":14950,"output_tokens":6011,"usd":0.112513,"stage2_stop_reason":"end_turn"},"total_usd":0.190819,"stage1_batch_id":"msgbatch_0179Z2pxRYBsXi2xADneafRw","stage2_batch_id":"msgbatch_01LaY9ZaVdgjWArcZo9eqQE5","note":"batch pricing = 50% of standard"},"stage1_raw":"```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 1998,\n      \"finding\": \"ADAM22 (MDC2) was cloned and found to possess a disrupted zinc-binding motif in its metalloproteinase-like domain, indicating it lacks proteolytic activity, and contains disintegrin-like and cysteine-rich sequences suggesting a role as an integrin ligand/adhesion molecule rather than a proteinase. Highly expressed in brain.\",\n      \"method\": \"cDNA cloning, sequence analysis, Northern blot\",\n      \"journal\": \"The Biochemical journal\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — sequence analysis with domain characterization confirmed by Northern blot, single lab but multiple methods\",\n      \"pmids\": [\"9693107\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2002,\n      \"finding\": \"The cytoplasmic tail of ADAM22 interacts with 14-3-3β; the major binding site was mapped to the last 28 amino acid residues of the ADAM22 cytoplasmic tail.\",\n      \"method\": \"Yeast two-hybrid, in vitro binding assay, co-immunoprecipitation\",\n      \"journal\": \"Science in China. Series C, Life sciences\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — reciprocal Co-IP and pull-down, single lab, binding site mapped\",\n      \"pmids\": [\"18762889\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2005,\n      \"finding\": \"ADAM22-deficient mice display severe ataxia, lethal convulsions, and marked peripheral nerve hypomyelination, establishing an essential in vivo role for ADAM22 in peripheral nervous system myelination and neuronal function.\",\n      \"method\": \"Gene targeting (knockout mice), histological analysis\",\n      \"journal\": \"BMC neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — clean KO with defined neurological and histological phenotype, replicated in subsequent studies\",\n      \"pmids\": [\"15876356\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2005,\n      \"finding\": \"Overexpression of ADAM22 in HEK293 cells enhances cell adhesion and spreading; this effect requires 14-3-3 binding motifs in the ADAM22 cytoplasmic tail, as a truncated ADAM22 lacking these motifs loses the adhesion-promoting activity. Multiple 14-3-3 family members interact with ADAM22cyt.\",\n      \"method\": \"Co-immunoprecipitation, in vitro pull-down, cell adhesion/spreading assays, overexpression in HEK293\",\n      \"journal\": \"Biochemical and biophysical research communications\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — functional adhesion assay plus Co-IP, single lab, domain-deletion approach\",\n      \"pmids\": [\"15882968\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2006,\n      \"finding\": \"ADAM22 serves as a postsynaptic receptor for the secreted epilepsy protein LGI1; LGI1 binding to ADAM22 enhances AMPA receptor-mediated synaptic transmission in hippocampal slices. Epilepsy-causing mutant LGI1 fails to bind ADAM22. ADAM22 is anchored at the postsynaptic density via scaffolds containing stargazin.\",\n      \"method\": \"Co-immunoprecipitation, cell surface binding, electrophysiology (hippocampal slice patch-clamp), rat brain biochemistry\",\n      \"journal\": \"Science (New York, N.Y.)\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 / Strong — multiple orthogonal methods (binding assay, electrophysiology, native brain biochemistry), widely replicated across labs\",\n      \"pmids\": [\"16990550\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2006,\n      \"finding\": \"ADAM22 inhibits cellular proliferation of glioma-derived astrocytes; this growth inhibition is mediated via the disintegrin domain interacting with integrins on the cell surface and can be overcome by overexpression of integrin-linked kinase.\",\n      \"method\": \"BrdU incorporation assay, GST fusion protein treatment, overexpression in glioma cells\",\n      \"journal\": \"Neurosurgery\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — functional proliferation assay plus domain-specific GST fusion rescue, single lab\",\n      \"pmids\": [\"16385342\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2006,\n      \"finding\": \"ADAM22 surface expression is regulated by phosphorylation-dependent interaction with 14-3-3 proteins; 14-3-3 proteins bind preferentially to the serine-phosphorylated precursor form of ADAM22 via two binding motifs (critical site at residues 831–834), masking ER retention signals and thereby promoting ADAM22 trafficking to the cell surface.\",\n      \"method\": \"Yeast two-hybrid, co-immunoprecipitation, ADAM22 point mutant analysis, cell surface localization assay\",\n      \"journal\": \"Journal of cell science\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — reciprocal Co-IP plus mutagenesis mapping plus localization readout, single lab but multiple orthogonal approaches\",\n      \"pmids\": [\"16868027\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2008,\n      \"finding\": \"LGI1 and LGI4 both bind to ADAM22 (as well as ADAM11 and ADAM23); LGI1 was identified as the most potent ADAM22-binding protein in mouse brain by immunoprecipitation/mass spectrometry, and binding specificity was confirmed by quantitative cell-ELISA.\",\n      \"method\": \"Immunoprecipitation, mass spectrometry, quantitative cell-ELISA\",\n      \"journal\": \"International journal of biological sciences\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — native brain IP/MS plus cell-ELISA, two orthogonal methods, single lab\",\n      \"pmids\": [\"18974846\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2008,\n      \"finding\": \"The ADAM22 pro domain (residues 26–199) is folded with secondary structure consisting predominantly of β-strands, and exists as two subdomains; it can be expressed as a stable, soluble protein suitable for structural studies.\",\n      \"method\": \"E. coli expression, CD spectroscopy, NMR spectroscopy\",\n      \"journal\": \"Protein expression and purification\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1 / Weak — NMR and CD characterization of recombinant domain, single lab, partial structural characterization only\",\n      \"pmids\": [\"18593599\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2009,\n      \"finding\": \"Crystal structure of the full ectodomain of human ADAM22 reveals a compact four-leaf clover arrangement; the metalloproteinase-like domain is held in the concave face of a rigid module formed by the disintegrin, cysteine-rich, and EGF-like domains. Lack of metalloproteinase activity is structurally explained by absence of critical catalytic residues, filled substrate groove, and steric hindrance from the cysteine-rich domain. Three bound calcium ions were identified with regulatory (metalloproteinase-like domain) and structural (disintegrin domain) roles.\",\n      \"method\": \"X-ray crystallography, isothermal titration calorimetry\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Moderate — crystal structure plus calorimetric functional validation, single lab but rigorous methods\",\n      \"pmids\": [\"19692335\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2010,\n      \"finding\": \"ADAM22 is a component of the axonal Kv1 potassium channel complex at juxtaparanodes of myelinated axons; it co-immunoprecipitates with Kv1.2 and the MAGUKs PSD-93 and PSD-95. When co-expressed with MAGUKs in heterologous cells, ADAM22 recruits Kv1 channels into membrane surface clusters. In ADAM22-null mice, juxtaparanodal clustering of PSD-93 and PSD-95 is lost, whereas Kv1.2 and Caspr2 clustering is normal, demonstrating that ADAM22 is required for MAGUK recruitment to juxtaparanodes.\",\n      \"method\": \"Co-immunoprecipitation, mass spectrometry, immunofluorescence in null mice, heterologous co-expression clustering assay\",\n      \"journal\": \"The Journal of neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — native IP/MS, null mouse analysis, heterologous clustering assay — multiple orthogonal methods\",\n      \"pmids\": [\"20089912\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2010,\n      \"finding\": \"ADAM22 functions as a major neuronal/axonal receptor for Schwann cell-secreted LGI4 in peripheral nerve myelination; LGI4 binds directly to ADAM22 without requiring additional membrane-associated factors. Conditional ablation experiments showed Schwann cells are the principal source of LGI4 and that axonal ADAM22 is required for Schwann cell differentiation, revealing a paracrine Schwann cell–axon signaling axis.\",\n      \"method\": \"Direct binding assay (LGI4-ADAM22), cell type-specific conditional KO mice, heterotypic Schwann cell–neuron co-culture\",\n      \"journal\": \"The Journal of neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — direct binding assay plus cell type-specific KO with defined myelination phenotype, replicated conceptually across multiple studies\",\n      \"pmids\": [\"20220021\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2010,\n      \"finding\": \"Mutations in the disintegrin domain of ADAM22 cause marked decreases in proprotein processing and reduced LGI4-binding; the P81R polymorphic variant functions comparably to wild-type.\",\n      \"method\": \"Mutagenesis, cell surface expression assays, LGI4-binding assays in transfected cells\",\n      \"journal\": \"Journal of receptor and signal transduction research\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Weak — mutagenesis plus cell binding assay, single lab, single study\",\n      \"pmids\": [\"20156119\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2013,\n      \"finding\": \"LGI1 autoantibodies from limbic encephalitis patients target the EPTP repeat domain of LGI1 and block LGI1-ADAM22/ADAM23 interaction, reversibly reducing synaptic AMPA receptor clusters in hippocampal neurons. Addition of the soluble ADAM22 extracellular domain alone was sufficient to reduce synaptic AMPA receptors, confirming that disruption of LGI1-ADAM22 interaction reduces AMPA receptor function.\",\n      \"method\": \"ELISA arrays, co-immunoprecipitation, immunofluorescence in hippocampal neurons, soluble ectodomain competition assay\",\n      \"journal\": \"The Journal of neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — multiple methods (ELISA, Co-IP, neuronal imaging, ectodomain competition), replicated across patient samples and cell assays\",\n      \"pmids\": [\"24227725\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2015,\n      \"finding\": \"ADAM22 acts as the postsynaptic receptor through which the paracrine signal LGI1 sets postsynaptic strength; ADAM22 maintains excitatory synapses through PDZ domain interactions. Without LGI1, PSD-95 (but not SAP102) cannot modulate synaptic transmission, establishing LGI1-ADAM22 as a synaptic organizing complex that coordinates PSD-95-dependent synapse maturation.\",\n      \"method\": \"Electrophysiology (hippocampal slices/neurons), dominant-negative ADAM22 expressing constructs, genetic epistasis with PSD-95/SAP102\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — electrophysiology plus genetic epistasis, multiple orthogonal approaches, single lab\",\n      \"pmids\": [\"26178195\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"Compound heterozygous mutations in ADAM22 (p.Cys401Tyr and a frameshift p.Ser799IlefsTer96) in a patient with epileptic encephalopathy were shown to abolish LGI1 binding; additionally, the frameshift mutant ADAM22 also failed to bind PSD-95, establishing these interactions as functionally essential in vivo.\",\n      \"method\": \"Cell surface binding assay, co-immunoprecipitation in heterologous expression system\",\n      \"journal\": \"Neurology. Genetics\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Weak — co-IP and binding assay in heterologous cells, single lab, functional validation of patient variants\",\n      \"pmids\": [\"27066583\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"Four secretion-positive LGI1 missense mutations (T380A, R407C, S473L, R474Q) linked to ADLTE significantly impair LGI1 interaction with both ADAM22 and ADAM23 on the cell surface without affecting LGI1 secretion or folding, demonstrating a second extracellular loss-of-function mechanism independent of secretion impairment.\",\n      \"method\": \"Co-immunoprecipitation, immunofluorescence, 3D protein modelling\",\n      \"journal\": \"PLoS genetics\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — Co-IP and immunofluorescence, single lab, consistent across four mutations\",\n      \"pmids\": [\"27760137\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2018,\n      \"finding\": \"Crystal structure of the human LGI1-ADAM22 complex at 2:2 heterotetrameric assembly revealed that the hydrophobic pocket of the C-terminal EPTP domain of LGI1 binds to the metalloprotease-like domain of ADAM22. LGI1-LGI1 dimerization is mediated by both LRR and EPTP domains. Pathogenic R474Q mutation disrupts LGI1-LGI1 interface and higher-order complex assembly in vitro and in a knock-in mouse model, without affecting secretion or ADAM22 binding.\",\n      \"method\": \"X-ray crystallography, in vitro complex assembly, mouse knock-in model\",\n      \"journal\": \"Nature communications\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — crystal structure plus mutagenesis plus in vivo knock-in validation, multiple orthogonal methods\",\n      \"pmids\": [\"29670100\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2019,\n      \"finding\": \"LGI1 is recruited to the axon initial segment (AIS) where it co-localizes with ADAM22 and Kv1 channels in hippocampal neurons. ADAM22 and ADAM23 promote ER export and surface expression of LGI1 and co-transport LGI1 in axonal vesicles. ADLTE missense mutations S473L and R474Q in LGI1 prevent its association with ADAM22 and enrichment at the AIS.\",\n      \"method\": \"Live-cell imaging, immunofluorescence in cultured hippocampal neurons, co-immunoprecipitation\",\n      \"journal\": \"Journal of cell science\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — live imaging plus Co-IP, mutation functional analysis, single lab\",\n      \"pmids\": [\"30598502\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2021,\n      \"finding\": \"The LGI1-ADAM22-MAGUK complex governs transsynaptic nanoalignment of PSD-95 nanodomains, NMDA/AMPA receptors, Kv1 channels, and LRRTM4-Neurexin adhesion molecules. Adam22 knock-in mice devoid of the ADAM22-MAGUK (PDZ-binding) interaction develop lethal epilepsy with less-condensed PSD-95 nanodomains and decreased excitatory synaptic transmission. Without ADAM22 binding, PSD-95 cannot potentiate AMPA receptor-mediated transmission. Forced co-expression of ADAM22 and PSD-95 reconstitutes nano-condensates in non-neuronal cells.\",\n      \"method\": \"ADAM22 knock-in mice, super-resolution microscopy, electrophysiology, non-neuronal cell reconstitution assay\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 / Strong — in vivo knock-in model, super-resolution imaging, electrophysiology, and reconstitution assay, multiple orthogonal methods\",\n      \"pmids\": [\"33397806\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2021,\n      \"finding\": \"Quantitative dual phosphorylation of ADAM22 by protein kinase A (PKA) mediates high-affinity binding to dimerized 14-3-3 proteins; this interaction protects LGI1-ADAM22 complexes from endocytosis-dependent degradation. Forskolin-induced PKA activation increases ADAM22 levels. Hypomorphic mouse series established that ~50% LGI1 and ~10% ADAM22 levels are sufficient to prevent lethal epilepsy. ADAM22 function is required in both excitatory and inhibitory neurons.\",\n      \"method\": \"Genetic and structural analysis, phosphorylation assays, PKA activation (forskolin), ADAM22/LGI1 hypomorphic mouse series\",\n      \"journal\": \"Cell reports\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 / Strong — biochemical phosphorylation mapping, structural analysis, multiple hypomorphic mouse lines, single lab with multiple orthogonal methods\",\n      \"pmids\": [\"34910912\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"Functional studies of 19 additional biallelic ADAM22 variants revealed at least three distinct pathological mechanisms: (i) defective cell membrane expression, (ii) impaired LGI1-binding, and/or (iii) impaired interaction with PSD-95, confirming these three molecular interactions as essential for ADAM22 function in vivo.\",\n      \"method\": \"Heterologous expression, cell surface expression assays, co-immunoprecipitation binding assays\",\n      \"journal\": \"Brain : a journal of neurology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — systematic functional analysis of multiple patient variants with binding and localization assays, single lab\",\n      \"pmids\": [\"35813813\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"A missense variant p.S905F in the PDZ-binding motif of ADAM22 specifically impairs ADAM22 binding to PSD-95 and other MAGUKs without affecting biosynthesis, stability, or LGI1 interaction, causing focal epilepsy in homozygous individuals and demonstrating that the ADAM22-MAGUK interaction is independently essential for seizure protection in humans.\",\n      \"method\": \"Structural in silico analysis, protein-protein interaction assays in transfected mammalian cells, co-immunoprecipitation\",\n      \"journal\": \"Brain communications\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Weak — Co-IP in transfected cells with variant-specific functional dissection, single lab, supported by clinical genetics\",\n      \"pmids\": [\"37953841\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Cryo-EM structures of the LGI1-ADAM22 complex determined at 3.79 Å resolution revealed a 3:3 heterohexameric assembly (three LGI1 and three ADAM22 ECD molecules), distinct from the previously crystallized 2:2 tetramer. High-speed atomic force microscopy visualized the flexibility of the 3:3 complex in solution, providing new insights into higher-order transsynaptic assembly modes.\",\n      \"method\": \"Chemical cross-linking, cryo-EM structure determination, high-speed atomic force microscopy (HS-AFM), SAXS\",\n      \"journal\": \"eLife\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Moderate — cryo-EM structure at near-atomic resolution plus HS-AFM dynamic visualization, multiple biophysical methods, single lab\",\n      \"pmids\": [\"40601686\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2019,\n      \"finding\": \"LGI3 associates with ADAM22 in LPS-stimulated human keratinocytes, as shown by co-immunoprecipitation, flow cytometry, and immunocytochemistry, suggesting ADAM22 can function as an LGI3 receptor outside the nervous system.\",\n      \"method\": \"Co-immunoprecipitation, flow cytometry, immunocytochemistry\",\n      \"journal\": \"Cytokine\",\n      \"confidence\": \"Low\",\n      \"confidence_rationale\": \"Tier 3 / Weak — single lab, Co-IP in keratinocytes, no functional consequence of ADAM22 binding established\",\n      \"pmids\": [\"31627033\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"ADAM22 is a catalytically inactive transmembrane ADAM family receptor, highly expressed in the nervous system, that functions as the postsynaptic receptor for the secreted epilepsy protein LGI1 (binding via LGI1's EPTP domain to the ADAM22 metalloprotease-like domain in a 2:2 or 3:3 hetero-oligomeric transsynaptic complex); through its cytoplasmic PDZ-binding motif, ADAM22 recruits the MAGUK scaffolding proteins PSD-95/PSD-93 to organize postsynaptic nanodomains containing AMPA/NMDA receptors and Kv1 channels, thereby regulating excitatory synaptic transmission and epilepsy prevention. ADAM22 surface levels are stabilized by PKA-dependent phosphorylation that promotes high-affinity 14-3-3 binding, protecting the LGI1-ADAM22 complex from endocytic degradation. In the peripheral nervous system, axonal ADAM22 acts as the receptor for Schwann cell-secreted LGI4, driving peripheral nerve myelination through paracrine signaling. Loss-of-function mutations or autoantibodies disrupting any of ADAM22's three key interactions—LGI1 binding, MAGUK recruitment, or surface expression—cause epileptic encephalopathy in humans and mice.\"\n}\n```","stage2_raw":"{\n  \"mechanistic_narrative\": \"ADAM22 is a catalytically inactive transmembrane member of the ADAM family, highly expressed in brain, that functions as a synaptic and axonal receptor coupling secreted LGI proteins to intracellular scaffolds rather than as a proteinase [#0, #4]. Its metalloproteinase-like domain lacks critical catalytic residues and is held in a rigid four-leaf-clover ectodomain by the disintegrin, cysteine-rich, and EGF-like domains, structurally explaining the absence of protease activity [#9]. ADAM22 is the principal postsynaptic receptor for the secreted epilepsy protein LGI1, which binds through its EPTP domain to the ADAM22 metalloprotease-like domain to form a higher-order transsynaptic assembly seen as both a 2:2 heterotetramer and a 3:3 heterohexamer [#4, #17, #23]. Through its cytoplasmic PDZ-binding motif ADAM22 recruits the MAGUK scaffolds PSD-95/PSD-93, organizing transsynaptic PSD-95 nanodomains that align AMPA/NMDA receptors and Kv1 channels and thereby setting excitatory synaptic strength [#10, #14, #19]. Surface levels of ADAM22 are controlled by PKA-dependent dual phosphorylation that drives high-affinity 14-3-3 binding, masking ER-retention signals to promote trafficking and protecting LGI1-ADAM22 complexes from endocytic degradation [#6, #20]. In the peripheral nervous system, axonal ADAM22 acts as the receptor for Schwann cell-secreted LGI4 to drive a paracrine signaling axis required for myelination, and ADAM22-null mice show ataxia, lethal seizures, and peripheral nerve hypomyelination [#2, #11]. Biallelic loss-of-function variants and autoantibodies that disrupt any of ADAM22's three essential interactions—LGI1 binding, MAGUK recruitment, or surface expression—cause human epileptic encephalopathy [#15, #21, #22, #13].\",\n  \"teleology\": [\n    {\n      \"year\": 1998,\n      \"claim\": \"Established that ADAM22 is structurally a non-proteolytic ADAM, redirecting expectation from enzyme to adhesion/receptor function and flagging its brain enrichment.\",\n      \"evidence\": \"cDNA cloning, sequence analysis, and Northern blot\",\n      \"pmids\": [\"9693107\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Sequence inference of catalytic inactivity not yet structurally confirmed\", \"No binding partners or in vivo function identified\"]\n    },\n    {\n      \"year\": 2002,\n      \"claim\": \"Identified the first cytoplasmic partner, 14-3-3, mapping the interaction to the C-terminal tail and opening the question of how the tail regulates ADAM22 behavior.\",\n      \"evidence\": \"Yeast two-hybrid, in vitro binding, co-immunoprecipitation\",\n      \"pmids\": [\"18762889\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Functional consequence of 14-3-3 binding not established\", \"Phosphorylation dependence not yet defined\"]\n    },\n    {\n      \"year\": 2005,\n      \"claim\": \"Knockout established ADAM22 as essential in vivo for peripheral nerve myelination and neuronal function, demonstrating non-redundant physiological roles.\",\n      \"evidence\": \"Gene-targeted knockout mice with histological analysis\",\n      \"pmids\": [\"15876356\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Molecular ligand and downstream pathway driving myelination unknown\", \"Mechanism linking loss to seizures not defined\"]\n    },\n    {\n      \"year\": 2005,\n      \"claim\": \"Connected the cytoplasmic 14-3-3 motifs to a cellular output by showing they are required for ADAM22-driven cell adhesion and spreading.\",\n      \"evidence\": \"Co-IP, pull-down, and adhesion/spreading assays with tail-truncation in HEK293\",\n      \"pmids\": [\"15882968\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Overexpression in non-neuronal cells may not reflect synaptic role\", \"Adhesion mechanism not linked to physiological ligand\"]\n    },\n    {\n      \"year\": 2006,\n      \"claim\": \"Defined ADAM22 as the postsynaptic receptor for LGI1 and linked the interaction to enhanced AMPA receptor transmission, establishing the core epilepsy-relevant signaling axis.\",\n      \"evidence\": \"Co-IP, surface binding, hippocampal slice electrophysiology, brain biochemistry\",\n      \"pmids\": [\"16990550\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Structural basis of LGI1 binding not resolved\", \"How surface trafficking is controlled not addressed\"]\n    },\n    {\n      \"year\": 2006,\n      \"claim\": \"Explained how ADAM22 reaches the cell surface, showing phosphorylation-dependent 14-3-3 binding to the precursor masks ER-retention signals to promote trafficking.\",\n      \"evidence\": \"Yeast two-hybrid, Co-IP, point mutant and surface-localization assays\",\n      \"pmids\": [\"16868027\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Kinase responsible for the phosphorylation not identified at this stage\", \"Link between trafficking control and complex stability unaddressed\"]\n    },\n    {\n      \"year\": 2008,\n      \"claim\": \"Demonstrated specificity and breadth of ligand engagement, identifying LGI1 as the most potent ADAM22-binding protein in brain while showing LGI4 also binds.\",\n      \"evidence\": \"Native brain IP/mass spectrometry and quantitative cell-ELISA\",\n      \"pmids\": [\"18974846\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Functional distinction between LGI1 and LGI4 signaling not resolved\", \"Stoichiometry of complexes unknown\"]\n    },\n    {\n      \"year\": 2009,\n      \"claim\": \"Provided the structural explanation for catalytic inactivity, showing a rigid ectodomain with a degraded, sterically blocked metalloprotease groove and bound calcium ions.\",\n      \"evidence\": \"X-ray crystallography of the human ectodomain plus ITC\",\n      \"pmids\": [\"19692335\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Ligand-bound structure not yet determined\", \"Functional role of calcium ions inferred, not tested\"]\n    },\n    {\n      \"year\": 2010,\n      \"claim\": \"Placed ADAM22 in the axonal Kv1 channel complex at juxtaparanodes and showed it is required for MAGUK recruitment, defining a scaffold-organizing role beyond the synapse.\",\n      \"evidence\": \"Native IP/MS, immunofluorescence in null mice, heterologous clustering assay\",\n      \"pmids\": [\"20089912\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Direct ADAM22-MAGUK binding interface not mapped here\", \"Relationship to LGI1 signaling at juxtaparanodes unclear\"]\n    },\n    {\n      \"year\": 2010,\n      \"claim\": \"Established ADAM22 as the neuronal receptor for Schwann cell-secreted LGI4, defining a paracrine axon-glia myelination axis and explaining the knockout hypomyelination.\",\n      \"evidence\": \"Direct LGI4-ADAM22 binding, cell-type-specific conditional KO, Schwann-neuron co-culture\",\n      \"pmids\": [\"20220021\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Downstream signaling from axonal ADAM22 into Schwann cell differentiation undefined\", \"Cytoplasmic effectors in myelination not identified\"]\n    },\n    {\n      \"year\": 2010,\n      \"claim\": \"Showed disintegrin-domain mutations impair proprotein processing and LGI4 binding, tying ADAM22 maturation to ligand engagement.\",\n      \"evidence\": \"Mutagenesis with surface expression and LGI4-binding assays in transfected cells\",\n      \"pmids\": [\"20156119\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Single study without in vivo validation\", \"P81R variant effect on disease not established\"]\n    },\n    {\n      \"year\": 2013,\n      \"claim\": \"Defined an autoimmune mechanism of disease by showing limbic-encephalitis LGI1 autoantibodies block LGI1-ADAM22 binding and reduce synaptic AMPA receptors.\",\n      \"evidence\": \"ELISA arrays, Co-IP, hippocampal neuron imaging, soluble ectodomain competition\",\n      \"pmids\": [\"24227725\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Whether antibody effects are fully reversible in vivo not shown\", \"Contribution of ADAM23 versus ADAM22 not separated\"]\n    },\n    {\n      \"year\": 2015,\n      \"claim\": \"Established the LGI1-ADAM22 complex as a synaptic organizer working through PDZ interactions, with PSD-95 unable to potentiate transmission without ADAM22.\",\n      \"evidence\": \"Hippocampal electrophysiology, dominant-negative ADAM22, genetic epistasis with PSD-95/SAP102\",\n      \"pmids\": [\"26178195\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Nanoscale organization of the complex not yet visualized\", \"Specificity for PSD-95 over other MAGUKs incompletely defined\"]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"Validated the three ADAM22 interactions as physiologically essential by showing human compound heterozygous mutations abolish LGI1 and PSD-95 binding in encephalopathy.\",\n      \"evidence\": \"Surface binding and Co-IP of patient variants in heterologous cells\",\n      \"pmids\": [\"27066583\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Heterologous-cell assays do not capture neuronal context\", \"Single patient genotype\"]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"Identified a secretion-independent LGI1 loss-of-function mechanism in which ADLTE missense mutations impair binding to ADAM22/ADAM23.\",\n      \"evidence\": \"Co-IP, immunofluorescence, 3D modelling of four mutations\",\n      \"pmids\": [\"27760137\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Structural basis of binding loss inferred from modelling only\", \"In vivo consequence not tested here\"]\n    },\n    {\n      \"year\": 2018,\n      \"claim\": \"Resolved the molecular architecture of the LGI1-ADAM22 complex, showing the EPTP domain engages the metalloprotease-like domain in a 2:2 assembly and that pathogenic R474Q disrupts higher-order assembly.\",\n      \"evidence\": \"X-ray crystallography, in vitro assembly, knock-in mouse\",\n      \"pmids\": [\"29670100\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Only the 2:2 assembly captured\", \"Transsynaptic geometry across the cleft not directly imaged\"]\n    },\n    {\n      \"year\": 2019,\n      \"claim\": \"Showed ADAM22 (with ADAM23) chaperones LGI1 ER export, surface expression, axonal transport, and AIS enrichment, linking receptor function to ligand delivery.\",\n      \"evidence\": \"Live-cell imaging, immunofluorescence, Co-IP in hippocampal neurons\",\n      \"pmids\": [\"30598502\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Relative contributions of ADAM22 vs ADAM23 not separated\", \"Single-lab imaging study\"]\n    },\n    {\n      \"year\": 2021,\n      \"claim\": \"Demonstrated the LGI1-ADAM22-MAGUK complex drives transsynaptic nanoalignment of PSD-95, glutamate receptors, Kv1, and adhesion molecules, with loss of the ADAM22-MAGUK interaction causing lethal epilepsy.\",\n      \"evidence\": \"ADAM22 PDZ knock-in mice, super-resolution microscopy, electrophysiology, non-neuronal reconstitution\",\n      \"pmids\": [\"33397806\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Biophysical driver of nano-condensation not fully defined\", \"Inhibitory-neuron contribution addressed only later\"]\n    },\n    {\n      \"year\": 2021,\n      \"claim\": \"Defined how surface ADAM22 is stabilized, showing PKA dual phosphorylation drives high-affinity dimeric 14-3-3 binding that protects complexes from endocytic degradation, and set the dosage thresholds for seizure protection.\",\n      \"evidence\": \"Phosphorylation/structural analysis, forskolin PKA activation, hypomorphic mouse series\",\n      \"pmids\": [\"34910912\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Endocytic machinery acting on dephosphorylated ADAM22 not identified\", \"Upstream signals controlling PKA activity in neurons unknown\"]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Systematically confirmed three distinct pathological mechanisms—defective surface expression, impaired LGI1 binding, impaired PSD-95 binding—across 19 biallelic ADAM22 variants.\",\n      \"evidence\": \"Heterologous expression, surface expression and Co-IP binding assays\",\n      \"pmids\": [\"35813813\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Genotype-phenotype correlation in patients not fully resolved\", \"Assays in heterologous cells only\"]\n    },\n    {\n      \"year\": 2023,\n      \"claim\": \"Isolated the ADAM22-MAGUK interaction as independently essential in humans, showing a PDZ-motif variant impairs MAGUK binding without affecting LGI1 binding or stability.\",\n      \"evidence\": \"In silico structural analysis and interaction/Co-IP assays of the p.S905F variant\",\n      \"pmids\": [\"37953841\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Single variant in transfected cells\", \"No in vivo model for this allele\"]\n    },\n    {\n      \"year\": 2025,\n      \"claim\": \"Revealed an alternative 3:3 heterohexameric LGI1-ADAM22 assembly and its conformational flexibility, expanding the model of higher-order transsynaptic organization.\",\n      \"evidence\": \"Cross-linking, cryo-EM at 3.79 Å, HS-AFM, SAXS\",\n      \"pmids\": [\"40601686\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Functional significance of 2:2 versus 3:3 assemblies in vivo unresolved\", \"How assembly state maps onto nanodomain organization unknown\"]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"It remains unknown how the intracellular signaling downstream of axonal ADAM22-LGI4 engagement drives Schwann cell differentiation, and how assembly stoichiometry (2:2 vs 3:3) is selected and tuned at native synapses.\",\n      \"evidence\": \"No discovery in the timeline defines the cytoplasmic effectors of ADAM22 in myelination or resolves in vivo assembly-state regulation\",\n      \"pmids\": [],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"No downstream myelination effector identified\", \"Physiological control of complex stoichiometry undefined\"]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0048018\", \"supporting_discovery_ids\": [4, 7, 11]},\n      {\"term_id\": \"GO:0060089\", \"supporting_discovery_ids\": [4, 11, 14]},\n      {\"term_id\": \"GO:0060090\", \"supporting_discovery_ids\": [10, 14, 19]},\n      {\"term_id\": \"GO:0098631\", \"supporting_discovery_ids\": [3, 5]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005886\", \"supporting_discovery_ids\": [6, 4, 10]},\n      {\"term_id\": \"GO:0005634\", \"supporting_discovery_ids\": []},\n      {\"term_id\": \"GO:0005783\", \"supporting_discovery_ids\": [6, 18]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-112316\", \"supporting_discovery_ids\": [4, 14, 19]},\n      {\"term_id\": \"R-HSA-1266738\", \"supporting_discovery_ids\": [2, 11]},\n      {\"term_id\": \"R-HSA-1643685\", \"supporting_discovery_ids\": [13, 15, 21, 22]}\n    ],\n    \"complexes\": [\n      \"LGI1-ADAM22 transsynaptic complex\",\n      \"ADAM22-PSD-95/PSD-93 (MAGUK) postsynaptic complex\",\n      \"axonal Kv1 channel juxtaparanodal complex\"\n    ],\n    \"partners\": [\n      \"LGI1\",\n      \"LGI4\",\n      \"PSD-95\",\n      \"PSD-93\",\n      \"14-3-3\",\n      \"Kv1.2\",\n      \"LGI3\",\n      \"ADAM23\"\n    ],\n    \"other_free_text\": []\n  }\n}","audit_flag":{"gene":"ADAM22","tier":"GROUNDING","verdict":"Evidence-grounding concern","subtype":"fabrication","uniprot_band":"medium","rules_fired":"R7","issue":"R7: fabricated (no corpus paper): 35813813"},"evaluation":{"pairwise":"win","faith_supported":7,"faith_total":7,"faith_pct":100.0}}