{"gene":"CD99L2","run_date":"2026-06-09T22:57:18","timeline":{"discoveries":[{"year":2003,"finding":"CD99L2 is a paralog of CD99 encoded on the X chromosome, with alternative splicing generating species-specific transcript variants; amino acid sequence alignment identified five conserved functional regions between CD99L2 and CD99, indicating close evolutionary relationship from a common ancestor gene.","method":"cDNA cloning, genomic organization analysis, comparative sequence alignment, alternative splicing analysis, in situ hybridization","journal":"Gene","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — multiple orthogonal methods (cloning, genomic analysis, in situ hybridization) in a single study establishing molecular characterization, but no direct functional assay of the conserved motifs","pmids":["12706889"],"is_preprint":false},{"year":2007,"finding":"CD99L2 mediates homophilic cell-cell adhesion in a divalent cation-dependent manner; transfection of L cell fibroblasts with CD99L2 conferred homophilic aggregation ability, and anti-CD99L2 antibody blocked neutrophil and monocyte influx into sites of inflammation in vivo.","method":"Cell transfection, homophilic aggregation assay, in vivo inflammation model with blocking antibodies","journal":"Cell communication & adhesion","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — direct functional assay in transfected cells plus in vivo blocking antibody experiment, two orthogonal methods in a single lab","pmids":["18163232"],"is_preprint":false},{"year":2010,"finding":"CD99 and CD99L2 act independently of PECAM-1 during leukocyte extravasation; blocking CD99 or CD99L2 in vivo traps neutrophils between endothelial cells and the underlying basement membrane, and CD99/CD99L2 are required even when the need for PECAM-1 is bypassed by TNF-alpha stimulation.","method":"Blocking antibodies in vitro and in vivo, PECAM-1 gene disruption, electron microscopy, 3D confocal fluorescence microscopy in inflamed cremaster tissue","journal":"Blood","confidence":"High","confidence_rationale":"Tier 2 / Strong — multiple orthogonal methods (antibody blockade, genetic KO, EM, 3D confocal), replicated across multiple inflammation models","pmids":["20479283"],"is_preprint":false},{"year":2013,"finding":"Endothelial-specific (but not myeloid-specific) gene ablation of CD99L2 impairs neutrophil extravasation in vivo, demonstrating that CD99L2 on endothelial cells, not on leukocytes themselves, is the critical functional pool for neutrophil recruitment; T cell recruitment into inflamed skin was also impaired by endothelial CD99L2 deletion.","method":"Conditional gene-deficient mice (endothelial-specific and myeloid-specific), multiple in vivo inflammation models (cremaster, peritoneum, skin)","journal":"Journal of immunology (Baltimore, Md. : 1950)","confidence":"High","confidence_rationale":"Tier 2 / Strong — cell-type-specific conditional knockout with clean phenotypic readout in multiple in vivo models, rigorously distinguishing endothelial vs. leukocyte contribution","pmids":["23293350"],"is_preprint":false},{"year":2013,"finding":"Mouse CD99 physically interacts with CD99L2 forming heterodimers; this interaction is dependent on the cytoplasmic domain of CD99, and heterodimers are more efficiently localized to the plasma membrane than homodimers, thereby positively regulating CD99L2 trafficking to cell surfaces.","method":"Bimolecular fluorescence complementation (BiFC), co-immunoprecipitation, FRET assays, cytoplasmic domain mutant analysis, analysis of endogenous CD99L2 surface levels in CD99-deficient cells and rescue by exogenous wild-type CD99","journal":"Journal of immunology (Baltimore, Md. : 1950)","confidence":"High","confidence_rationale":"Tier 1-2 / Strong — three independent methods (BiFC, Co-IP, FRET) plus domain mutagenesis and rescue experiments establishing the cytoplasmic-domain-dependent mechanism","pmids":["24133166"],"is_preprint":false},{"year":2013,"finding":"Knockdown of CD99L2 in murine B lymphoma (A20) cells results in G2 phase prolongation, reduced proliferative ability, H/RS-cell-like morphology, increased CD30 and CD15 expression, and upregulation of cytokines; NF-κB pathway activation is associated with mCD99L2 downregulation.","method":"shRNA knockdown, flow cytometry, MTT assay, western blot, antibody arrays, in vivo tumor growth assay","journal":"Oncology reports","confidence":"Low","confidence_rationale":"Tier 3 / Weak — single lab, multiple phenotypic readouts but NF-κB pathway involvement is correlative (western blot only), no direct mechanistic dissection","pmids":["23338758"],"is_preprint":false},{"year":2015,"finding":"CD99L2 knockout mice have a greater than 80% block in neutrophil infiltration and near-complete block in monocyte emigration in the thioglycollate peritonitis model; CD99L2 deficiency did not affect circulating leukocyte subset numbers, nor expression of ICAM-1, PECAM, or CD99 on endothelial cells.","method":"CD99L2 knockout mouse, thioglycollate peritonitis model, leukocyte counts, flow cytometry for endothelial adhesion molecule expression","journal":"Experimental and molecular pathology","confidence":"High","confidence_rationale":"Tier 2 / Strong — genetic knockout with defined in vivo inflammatory phenotype, replicated findings from a second independently generated KO line consistent with prior reports","pmids":["26321243"],"is_preprint":false},{"year":2018,"finding":"Tie-2-Cre-driven endothelial deletion of CD99L2 inhibits leukocyte entry into the CNS during EAE by blocking diapedesis through the endothelial basement membrane of BBB vessels, with accumulation of leukocytes between endothelial cells and the basement membrane, while luminal leukocyte docking was unaffected.","method":"Conditional knockout (Tie-2-Cre), EAE model (MOG35-55 immunization), 3D vibratome confocal microscopy, histopathological analysis of demyelination and inflammatory foci","journal":"Journal of leukocyte biology","confidence":"High","confidence_rationale":"Tier 2 / Strong — conditional genetic knockout with specific mechanistic localization of defect (basement membrane crossing, not luminal adhesion) using 3D imaging","pmids":["29791026"],"is_preprint":false},{"year":2022,"finding":"Human CD99L2 is constitutively expressed at endothelial cell borders and on leukocytes, and regulates a specific sequential step of transendothelial migration between PECAM and CD99; it promotes transmigration by recruiting the lateral border recycling compartment (LBRC) to TEM sites specifically downstream of PECAM initiation.","method":"Antibody blockade of human CD99L2, genetic knockdown in primary human endothelial cells, in vitro TEM assays with primary human neutrophils and monocytes, LBRC recruitment analysis","journal":"Journal of immunology (Baltimore, Md. : 1950)","confidence":"High","confidence_rationale":"Tier 2 / Strong — antibody blockade and genetic knockdown combined with LBRC recruitment assay, defined sequential pathway position between PECAM and CD99, using primary human cells with multiple orthogonal methods","pmids":["35914838"],"is_preprint":false},{"year":2022,"finding":"cd99l2 knockout in zebrafish (TALEN-mediated) reduces recruitment of granulocytes and macrophages to wound sites after caudal fin damage, and decreases expression of mfap4 (a macrophage marker), indicating a role for CD99L2 in leukocyte interstitial migration in addition to extravasation.","method":"TALEN knockout in zebrafish, caudal fin injury model, transgenic fluorescent reporter lines, RNA-seq, mfap4 expression analysis","journal":"Yi chuan = Hereditas","confidence":"Medium","confidence_rationale":"Tier 2 / Weak — genetic knockout with defined wound-response phenotype in zebrafish, single lab, one publication","pmids":["36384956"],"is_preprint":false},{"year":2023,"finding":"P. gingivalis gingipains downregulate CD99 and CD99L2 expression on endothelial cells and leukocytes through inhibition of the PI3K/Akt pathway, impairing monocyte transendothelial migration while promoting monocyte adhesion.","method":"In vitro gingipain stimulation, in vivo mouse infection model, TEM assay, PI3K/Akt pathway inhibitor analysis, flow cytometry for CD99/CD99L2 surface expression","journal":"Microbiology spectrum","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — in vitro and in vivo convergent evidence for gingipain-mediated downregulation through PI3K/Akt, but PI3K pathway involvement is pharmacological without direct mechanistic dissection of CD99L2","pmids":["37199607"],"is_preprint":false},{"year":2025,"finding":"CD99L2 positively regulates neurite outgrowth and excitatory synapse development in neurons; it inversely regulates immediate-early gene expression (Arc, Egr1, c-Fos) by inhibiting CREB and SRF transcription factor activity; neuronal inactivation promotes CD99L2 transport to the cell surface from recycling endosomes; CD99L2 knockout mice exhibit impaired excitatory synaptic transmission and plasticity in the hippocampus, and deficits in spatial memory and contextual fear conditioning.","method":"CD99L2 knockout mice, neurite outgrowth assays, synaptic transmission electrophysiology, immediate-early gene expression analysis, CREB/SRF activity assays, live-cell imaging of CD99L2 trafficking from recycling endosomes, behavioral tests","journal":"Cell reports","confidence":"High","confidence_rationale":"Tier 2 / Strong — genetic knockout with multiple orthogonal functional readouts (electrophysiology, gene expression, live imaging, behavior), single lab but convergent methods establishing mechanism","pmids":["39808524"],"is_preprint":false},{"year":2026,"finding":"CD99L2 is a mechanoresponsive adhesion regulator transiently induced in primitive erythrocytes by shear stress-activated Piezo1 signaling; CD99L2 binds and anchors β-catenin at the plasma membrane, and its loss leads to nuclear translocation of β-catenin, Rap1 signaling activation, persistent adhesion molecule expression, erythrocyte retention, impaired maturation, and hemolytic anemia.","method":"Zebrafish and mouse models, CD99L2 knockout, Piezo1 activation experiments, co-immunoprecipitation for CD99L2-β-catenin interaction, β-catenin nuclear translocation assays, Rap1 signaling analysis","journal":"Cell reports","confidence":"High","confidence_rationale":"Tier 1-2 / Strong — multiple orthogonal methods (KO in two species, binding assay, pathway analysis) demonstrating CD99L2-β-catenin physical interaction and downstream signaling consequences","pmids":["41915472"],"is_preprint":false},{"year":2026,"finding":"CD99L2 protein occurs mainly in a ubiquitinated form and physically interacts with and activates the calcium-dependent protease CAPN1; ablation of cytoplasmic or extracellular domains of CD99L2 leads to intracellular mislocalization and loss of CAPN1 interaction; loss-of-function variants in CD99L2 cause X-linked spastic ataxia, with transcriptome analysis in patient fibroblasts revealing synaptic function-specific disturbances.","method":"Exome/genome sequencing in patient cohort, cellular studies of CD99L2 domain-deletion mutants (localization by microscopy), co-immunoprecipitation for CD99L2-CAPN1 interaction, ubiquitination analysis, transcriptome analysis in patient-derived fibroblasts","journal":"Nature communications","confidence":"High","confidence_rationale":"Tier 1-2 / Strong — domain mutagenesis establishing domain requirements for localization and CAPN1 interaction, patient cellular studies, transcriptome analysis; multiple orthogonal methods in a single rigorous study","pmids":["41690933"],"is_preprint":false}],"current_model":"CD99L2 is a type I transmembrane glycoprotein that functions as a homophilic adhesion molecule and multi-functional signaling regulator: in the vasculature, endothelial CD99L2 is essential for leukocyte diapedesis through the endothelial basement membrane, acting sequentially between PECAM-1 and CD99 by recruiting the lateral border recycling compartment; CD99 physically interacts with CD99L2 via its cytoplasmic domain to promote CD99L2 trafficking to the plasma membrane; in neurons, CD99L2 promotes excitatory synapse development, anchors β-catenin at the membrane to suppress its nuclear translocation, and inversely regulates immediate-early gene expression via CREB/SRF inhibition; in erythroid development, shear stress-activated Piezo1 induces CD99L2 expression, which restrains β-catenin nuclear signaling and Rap1 activation to allow erythrocyte de-adhesion; and CD99L2 also serves as an activating interactor for the protease CAPN1, with loss-of-function variants causing X-linked spastic ataxia."},"narrative":{"mechanistic_narrative":"CD99L2 is a type I transmembrane glycoprotein and homophilic, divalent-cation-dependent cell-cell adhesion molecule that functions across vascular, neuronal, and erythroid contexts [PMID:18163232]. Its best-characterized role is in leukocyte extravasation: endothelial CD99L2 — not the leukocyte pool — is the critical functional protein required for neutrophil, monocyte, and T cell diapedesis, acting specifically at the step of crossing the endothelial basement membrane such that its blockade traps leukocytes between endothelial cells and the basement membrane [PMID:20479283, PMID:23293350, PMID:26321243, PMID:29791026]. In human transendothelial migration it operates as a sequential gateway downstream of PECAM and upstream of CD99, promoting transmigration by recruiting the lateral border recycling compartment to migration sites [PMID:35914838]. CD99L2 heterodimerizes with its paralog CD99 in a manner dependent on the CD99 cytoplasmic domain, and this interaction promotes CD99L2 trafficking to the plasma membrane [PMID:24133166]. In neurons, CD99L2 promotes neurite outgrowth and excitatory synapse development, traffics to the surface from recycling endosomes upon neuronal inactivation, and inversely regulates immediate-early gene expression by inhibiting CREB and SRF activity, with knockout mice showing impaired hippocampal plasticity and memory deficits [PMID:39808524]. In erythroid development it is a mechanoresponsive regulator induced by shear-stress-activated Piezo1 signaling that anchors β-catenin at the plasma membrane to restrain its nuclear translocation and Rap1 activation, enabling erythrocyte de-adhesion and maturation [PMID:41915472]. CD99L2 also physically interacts with and activates the calcium-dependent protease CAPN1, and loss-of-function variants in CD99L2 cause X-linked spastic ataxia [PMID:41690933].","teleology":[{"year":2003,"claim":"Established the molecular identity of CD99L2 as an X-linked CD99 paralog, framing it as a candidate adhesion molecule before any function was tested.","evidence":"cDNA cloning, genomic organization, comparative sequence alignment and in situ hybridization","pmids":["12706889"],"confidence":"Medium","gaps":["No functional assay of the conserved motifs","Surface adhesion behavior not yet tested"]},{"year":2007,"claim":"Demonstrated CD99L2 is a functional adhesion molecule, answering whether the CD99 paralog had adhesive activity, and linked it to inflammatory leukocyte recruitment.","evidence":"Homophilic aggregation assay in transfected L cell fibroblasts plus in vivo blocking antibody inflammation model","pmids":["18163232"],"confidence":"Medium","gaps":["Did not localize the step of extravasation affected","Did not distinguish endothelial vs leukocyte CD99L2"]},{"year":2010,"claim":"Positioned CD99/CD99L2 in the diapedesis cascade by showing they act independently of and downstream of PECAM-1, with blockade trapping neutrophils at the basement membrane.","evidence":"Blocking antibodies, PECAM-1 gene disruption, electron microscopy and 3D confocal imaging in inflamed cremaster tissue","pmids":["20479283"],"confidence":"High","gaps":["Did not resolve which cell type's CD99L2 is required","Molecular mechanism of basement membrane crossing unknown"]},{"year":2013,"claim":"Resolved that endothelial, not myeloid, CD99L2 is the functional pool for leukocyte recruitment, settling the cell-of-origin question.","evidence":"Endothelial- and myeloid-specific conditional knockout mice across cremaster, peritoneum and skin inflammation models","pmids":["23293350"],"confidence":"High","gaps":["Did not define molecular partners at the basement membrane step"]},{"year":2013,"claim":"Defined the CD99-CD99L2 physical interaction and its consequence, showing CD99 heterodimerizes with CD99L2 via its cytoplasmic domain to drive CD99L2 surface trafficking.","evidence":"BiFC, co-immunoprecipitation, FRET, cytoplasmic domain mutagenesis, and rescue of surface CD99L2 in CD99-deficient cells","pmids":["24133166"],"confidence":"High","gaps":["Trafficking machinery downstream of CD99 binding not identified","Stoichiometry in vivo not determined"]},{"year":2013,"claim":"Probed CD99L2 in a B lymphoma context, linking its downregulation to altered proliferation, Hodgkin/Reed-Sternberg-like phenotype and NF-κB activation.","evidence":"shRNA knockdown in murine A20 cells with flow cytometry, MTT, western blot, antibody arrays and in vivo tumor assay","pmids":["23338758"],"confidence":"Low","gaps":["NF-κB involvement is correlative (western blot only)","No direct mechanistic dissection","Single lab, not independently confirmed"]},{"year":2015,"claim":"Confirmed the magnitude of the extravasation defect with a second knockout line and excluded indirect effects on circulating leukocytes or other endothelial adhesion molecules.","evidence":"CD99L2 knockout mouse thioglycollate peritonitis model with leukocyte counts and flow cytometry for ICAM-1, PECAM, CD99","pmids":["26321243"],"confidence":"High","gaps":["Mechanism of basement-membrane-specific block still undefined"]},{"year":2018,"claim":"Extended the diapedesis role to the blood-brain barrier, showing endothelial CD99L2 is required for leukocyte entry into the CNS during neuroinflammation at the basement membrane step.","evidence":"Tie-2-Cre endothelial conditional knockout in the EAE model with 3D vibratome confocal imaging and histopathology","pmids":["29791026"],"confidence":"High","gaps":["Did not identify the basement membrane binding partner","Luminal docking explicitly unaffected, leaving the molecular trigger of crossing open"]},{"year":2022,"claim":"Placed human CD99L2 precisely in the TEM cascade between PECAM and CD99 and identified its mechanism as LBRC recruitment downstream of PECAM initiation.","evidence":"Antibody blockade and genetic knockdown in primary human endothelial cells with TEM assays and LBRC recruitment analysis","pmids":["35914838"],"confidence":"High","gaps":["Molecular link between CD99L2 and LBRC trafficking machinery unresolved"]},{"year":2022,"claim":"Showed a conserved CD99L2 role in interstitial leukocyte migration beyond extravasation using a non-mammalian system.","evidence":"TALEN cd99l2 knockout in zebrafish with caudal fin injury, fluorescent reporter lines, RNA-seq and mfap4 analysis","pmids":["36384956"],"confidence":"Medium","gaps":["Single lab, one publication","Mechanism of interstitial migration defect not dissected"]},{"year":2023,"claim":"Identified a pathogen-driven regulatory input, showing P. gingivalis gingipains suppress CD99/CD99L2 via the PI3K/Akt pathway to impair monocyte TEM.","evidence":"In vitro gingipain stimulation, in vivo infection, TEM assay, PI3K/Akt inhibitor analysis and flow cytometry","pmids":["37199607"],"confidence":"Medium","gaps":["PI3K pathway involvement is pharmacological without direct mechanistic dissection of CD99L2 regulation"]},{"year":2025,"claim":"Revealed a distinct neuronal function, showing CD99L2 promotes synapse development and inversely controls immediate-early gene expression via CREB/SRF inhibition, with behavioral consequences.","evidence":"CD99L2 knockout mice with neurite/synapse assays, electrophysiology, immediate-early gene and CREB/SRF activity assays, recycling-endosome trafficking imaging and behavioral tests","pmids":["39808524"],"confidence":"High","gaps":["Direct molecular link from membrane CD99L2 to CREB/SRF not defined","Neuronal adhesion partner not identified"]},{"year":2026,"claim":"Connected CD99L2 to mechanotransduction and Wnt signaling, showing Piezo1-induced CD99L2 anchors β-catenin at the membrane to enable erythrocyte de-adhesion and prevent hemolytic anemia.","evidence":"Zebrafish and mouse knockouts, Piezo1 activation, co-IP for CD99L2-β-catenin, β-catenin nuclear translocation and Rap1 signaling analysis","pmids":["41915472"],"confidence":"High","gaps":["Structural basis of CD99L2-β-catenin binding unknown","How the same protein anchors β-catenin across cell types not unified"]},{"year":2026,"claim":"Identified CD99L2 as a CAPN1 activator and the cause of X-linked spastic ataxia, linking domain requirements, ubiquitination, and a Mendelian disease.","evidence":"Patient exome/genome sequencing, domain-deletion mutant localization studies, co-IP for CD99L2-CAPN1, ubiquitination and patient fibroblast transcriptome analysis","pmids":["41690933"],"confidence":"High","gaps":["How CD99L2 activates CAPN1 enzymatically not resolved","Role of CD99L2 ubiquitination in function unclear","Substrates downstream of CAPN1 activation not defined"]},{"year":null,"claim":"It remains unknown how a single adhesion molecule mechanistically reconciles its diverse roles — basement-membrane crossing, LBRC recruitment, β-catenin anchoring, CREB/SRF control, and CAPN1 activation — and whether a unifying biochemical activity underlies them.","evidence":"No single study in the corpus reconciles the vascular, neuronal, erythroid and protease-activating functions","pmids":[],"confidence":"Low","gaps":["No structural model of CD99L2","Direct extracellular ligand at the basement membrane unidentified","Mechanism connecting surface adhesion to intracellular signaling outputs unknown"]}],"mechanism_profile":{"molecular_activity":[{"term_id":"GO:0098631","term_label":"cell adhesion mediator activity","supporting_discovery_ids":[1]},{"term_id":"GO:0098772","term_label":"molecular function regulator activity","supporting_discovery_ids":[12,13]},{"term_id":"GO:0060089","term_label":"molecular transducer activity","supporting_discovery_ids":[8,11]}],"localization":[{"term_id":"GO:0005886","term_label":"plasma membrane","supporting_discovery_ids":[4,8,12]},{"term_id":"GO:0005768","term_label":"endosome","supporting_discovery_ids":[11]}],"pathway":[{"term_id":"R-HSA-168256","term_label":"Immune System","supporting_discovery_ids":[2,3,6,7,8]},{"term_id":"R-HSA-112316","term_label":"Neuronal System","supporting_discovery_ids":[11]},{"term_id":"R-HSA-162582","term_label":"Signal Transduction","supporting_discovery_ids":[12]},{"term_id":"R-HSA-1643685","term_label":"Disease","supporting_discovery_ids":[13]}],"complexes":[],"partners":["CD99","CTNNB1","CAPN1"],"other_free_text":[]}},"prefetch_data":{"uniprot":{"accession":"Q8TCZ2","full_name":"CD99 antigen-like protein 2","aliases":["MIC2-like protein 1"],"length_aa":262,"mass_kda":28.0,"function":"Plays a role in a late step of leukocyte extravasation helping cells to overcome the endothelial basement membrane. Acts at the same site as, but independently of, PECAM1 (By similarity). Homophilic adhesion molecule, but these interactions may not be required for cell aggregation (By similarity)","subcellular_location":"Cell membrane; Cell junction; Secreted","url":"https://www.uniprot.org/uniprotkb/Q8TCZ2/entry"},"depmap":{"release":"DepMap","has_data":true,"is_common_essential":false,"resolved_as":"","url":"https://depmap.org/portal/gene/CD99L2","classification":"Not Classified","n_dependent_lines":0,"n_total_lines":1208,"dependency_fraction":0.0},"opencell":{"profiled":false,"resolved_as":"","ensg_id":"","cell_line_id":"","localizations":[],"interactors":[],"url":"https://opencell.sf.czbiohub.org/search/CD99L2","total_profiled":1310},"omim":[{"mim_id":"313470","title":"CD99 ANTIGEN, X CHROMOSOME; CD99","url":"https://www.omim.org/entry/313470"},{"mim_id":"300846","title":"CD99 ANTIGEN-LIKE 2; CD99L2","url":"https://www.omim.org/entry/300846"},{"mim_id":"173445","title":"PLATELET-ENDOTHELIAL CELL ADHESION MOLECULE 1; PECAM1","url":"https://www.omim.org/entry/173445"}],"hpa":{"profiled":true,"resolved_as":"","reliability":"Approved","locations":[{"location":"Plasma membrane","reliability":"Approved"}],"tissue_specificity":"Tissue enhanced","tissue_distribution":"Detected in all","driving_tissues":[{"tissue":"skeletal muscle","ntpm":170.4}],"url":"https://www.proteinatlas.org/search/CD99L2"},"hgnc":{"alias_symbol":["CD99B"],"prev_symbol":["MIC2L1"]},"alphafold":{"accession":"Q8TCZ2","domains":[],"viewer_url":"https://alphafold.ebi.ac.uk/entry/Q8TCZ2","model_url":"https://alphafold.ebi.ac.uk/files/AF-Q8TCZ2-F1-model_v6.cif","pae_url":"https://alphafold.ebi.ac.uk/files/AF-Q8TCZ2-F1-predicted_aligned_error_v6.png","plddt_mean":56.06},"mouse_models":{"mgi_url":"https://www.informatics.jax.org/marker/summary?nomen=CD99L2","jax_strain_url":"https://www.jax.org/strain/search?query=CD99L2"},"sequence":{"accession":"Q8TCZ2","fasta_url":"https://rest.uniprot.org/uniprotkb/Q8TCZ2.fasta","uniprot_url":"https://www.uniprot.org/uniprotkb/Q8TCZ2/entry","alphafold_viewer_url":"https://alphafold.ebi.ac.uk/entry/Q8TCZ2"}},"corpus_meta":[{"pmid":"20479283","id":"PMC_20479283","title":"CD99 and CD99L2 act at the same site as, but independently of, PECAM-1 during leukocyte diapedesis.","date":"2010","source":"Blood","url":"https://pubmed.ncbi.nlm.nih.gov/20479283","citation_count":70,"is_preprint":false},{"pmid":"12706889","id":"PMC_12706889","title":"Cloning, genomic organization, alternative transcripts and expression analysis of CD99L2, a novel paralog of human CD99, and identification of evolutionary conserved motifs.","date":"2003","source":"Gene","url":"https://pubmed.ncbi.nlm.nih.gov/12706889","citation_count":51,"is_preprint":false},{"pmid":"18163232","id":"PMC_18163232","title":"The murine CD99-related molecule CD99-like 2 (CD99L2) is an adhesion molecule involved in the inflammatory response.","date":"2007","source":"Cell communication & adhesion","url":"https://pubmed.ncbi.nlm.nih.gov/18163232","citation_count":41,"is_preprint":false},{"pmid":"22681640","id":"PMC_22681640","title":"Immune function genes CD99L2, JARID2 and TPO show association with autism spectrum disorder.","date":"2012","source":"Molecular autism","url":"https://pubmed.ncbi.nlm.nih.gov/22681640","citation_count":34,"is_preprint":false},{"pmid":"23293350","id":"PMC_23293350","title":"Cutting edge: Endothelial-specific gene ablation of CD99L2 impairs leukocyte extravasation in vivo.","date":"2013","source":"Journal of immunology (Baltimore, Md. : 1950)","url":"https://pubmed.ncbi.nlm.nih.gov/23293350","citation_count":26,"is_preprint":false},{"pmid":"37199607","id":"PMC_37199607","title":"Porphyromonas gingivalis Gingipains Destroy the Vascular Barrier and Reduce CD99 and CD99L2 Expression To Regulate Transendothelial Migration.","date":"2023","source":"Microbiology spectrum","url":"https://pubmed.ncbi.nlm.nih.gov/37199607","citation_count":17,"is_preprint":false},{"pmid":"29791026","id":"PMC_29791026","title":"CD99L2 deficiency inhibits leukocyte entry into the central nervous system and ameliorates neuroinflammation.","date":"2018","source":"Journal of leukocyte biology","url":"https://pubmed.ncbi.nlm.nih.gov/29791026","citation_count":12,"is_preprint":false},{"pmid":"24133166","id":"PMC_24133166","title":"Interaction of CD99 with its paralog CD99L2 positively regulates CD99L2 trafficking to cell surfaces.","date":"2013","source":"Journal of immunology (Baltimore, Md. : 1950)","url":"https://pubmed.ncbi.nlm.nih.gov/24133166","citation_count":11,"is_preprint":false},{"pmid":"26321243","id":"PMC_26321243","title":"CD99-like 2 (CD99L2)-deficient mice are defective in the acute inflammatory response.","date":"2015","source":"Experimental and molecular pathology","url":"https://pubmed.ncbi.nlm.nih.gov/26321243","citation_count":8,"is_preprint":false},{"pmid":"23338758","id":"PMC_23338758","title":"Effect of shRNA targeting mouse CD99L2 gene in a murine B cell lymphoma in vitro and in vivo.","date":"2013","source":"Oncology reports","url":"https://pubmed.ncbi.nlm.nih.gov/23338758","citation_count":6,"is_preprint":false},{"pmid":"35914838","id":"PMC_35914838","title":"Human CD99L2 Regulates a Unique Step in Leukocyte Transmigration.","date":"2022","source":"Journal of immunology (Baltimore, Md. : 1950)","url":"https://pubmed.ncbi.nlm.nih.gov/35914838","citation_count":4,"is_preprint":false},{"pmid":"39808524","id":"PMC_39808524","title":"Cd99l2 regulates excitatory synapse development and restrains immediate-early gene activation.","date":"2025","source":"Cell reports","url":"https://pubmed.ncbi.nlm.nih.gov/39808524","citation_count":3,"is_preprint":false},{"pmid":"41690933","id":"PMC_41690933","title":"Loss-of-function variants in the CAPN1 activator CD99L2 cause X-linked spastic ataxia.","date":"2026","source":"Nature communications","url":"https://pubmed.ncbi.nlm.nih.gov/41690933","citation_count":0,"is_preprint":false},{"pmid":"36384956","id":"PMC_36384956","title":"The role of the cd99l2 gene on leukocyte interstitial migration in zebrafish.","date":"2022","source":"Yi chuan = Hereditas","url":"https://pubmed.ncbi.nlm.nih.gov/36384956","citation_count":0,"is_preprint":false},{"pmid":"20501370","id":"PMC_20501370","title":"[Preparation of RNA probe for cd99l2 gene of zebrafish labeled with digoxingenin-UTP].","date":"2010","source":"Nan fang yi ke da xue xue bao = Journal of Southern Medical University","url":"https://pubmed.ncbi.nlm.nih.gov/20501370","citation_count":0,"is_preprint":false},{"pmid":"41915472","id":"PMC_41915472","title":"Shear stress-induced Piezo1 activates CD99L2 to facilitate the initiation of blood circulation.","date":"2026","source":"Cell reports","url":"https://pubmed.ncbi.nlm.nih.gov/41915472","citation_count":0,"is_preprint":false},{"pmid":"40692150","id":"PMC_40692150","title":"Loss of CD99L2 Contributed to Temozolomide Resistance and Glioblastoma Tumorigenesis Based on Genome-scale CRISPR/Cas9 Screening.","date":"2026","source":"Current pharmaceutical design","url":"https://pubmed.ncbi.nlm.nih.gov/40692150","citation_count":0,"is_preprint":false}],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":10217,"output_tokens":3873,"usd":0.044373,"stage1_stop_reason":"end_turn"},"stage2":{"model":"claude-opus-4-8","input_tokens":11716,"output_tokens":4443,"usd":0.084828,"stage2_stop_reason":"end_turn"},"total_usd":0.129201,"stage1_batch_id":"msgbatch_01ACAxJQabDsht39JXeWG22d","stage2_batch_id":"msgbatch_01QEU22A9U5sbPTgsMzTmvSf","note":"batch pricing = 50% of standard"},"stage1_raw":"```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 2003,\n      \"finding\": \"CD99L2 is a paralog of CD99 encoded on the X chromosome, with alternative splicing generating species-specific transcript variants; amino acid sequence alignment identified five conserved functional regions between CD99L2 and CD99, indicating close evolutionary relationship from a common ancestor gene.\",\n      \"method\": \"cDNA cloning, genomic organization analysis, comparative sequence alignment, alternative splicing analysis, in situ hybridization\",\n      \"journal\": \"Gene\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — multiple orthogonal methods (cloning, genomic analysis, in situ hybridization) in a single study establishing molecular characterization, but no direct functional assay of the conserved motifs\",\n      \"pmids\": [\"12706889\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"CD99L2 mediates homophilic cell-cell adhesion in a divalent cation-dependent manner; transfection of L cell fibroblasts with CD99L2 conferred homophilic aggregation ability, and anti-CD99L2 antibody blocked neutrophil and monocyte influx into sites of inflammation in vivo.\",\n      \"method\": \"Cell transfection, homophilic aggregation assay, in vivo inflammation model with blocking antibodies\",\n      \"journal\": \"Cell communication & adhesion\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct functional assay in transfected cells plus in vivo blocking antibody experiment, two orthogonal methods in a single lab\",\n      \"pmids\": [\"18163232\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2010,\n      \"finding\": \"CD99 and CD99L2 act independently of PECAM-1 during leukocyte extravasation; blocking CD99 or CD99L2 in vivo traps neutrophils between endothelial cells and the underlying basement membrane, and CD99/CD99L2 are required even when the need for PECAM-1 is bypassed by TNF-alpha stimulation.\",\n      \"method\": \"Blocking antibodies in vitro and in vivo, PECAM-1 gene disruption, electron microscopy, 3D confocal fluorescence microscopy in inflamed cremaster tissue\",\n      \"journal\": \"Blood\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — multiple orthogonal methods (antibody blockade, genetic KO, EM, 3D confocal), replicated across multiple inflammation models\",\n      \"pmids\": [\"20479283\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2013,\n      \"finding\": \"Endothelial-specific (but not myeloid-specific) gene ablation of CD99L2 impairs neutrophil extravasation in vivo, demonstrating that CD99L2 on endothelial cells, not on leukocytes themselves, is the critical functional pool for neutrophil recruitment; T cell recruitment into inflamed skin was also impaired by endothelial CD99L2 deletion.\",\n      \"method\": \"Conditional gene-deficient mice (endothelial-specific and myeloid-specific), multiple in vivo inflammation models (cremaster, peritoneum, skin)\",\n      \"journal\": \"Journal of immunology (Baltimore, Md. : 1950)\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — cell-type-specific conditional knockout with clean phenotypic readout in multiple in vivo models, rigorously distinguishing endothelial vs. leukocyte contribution\",\n      \"pmids\": [\"23293350\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2013,\n      \"finding\": \"Mouse CD99 physically interacts with CD99L2 forming heterodimers; this interaction is dependent on the cytoplasmic domain of CD99, and heterodimers are more efficiently localized to the plasma membrane than homodimers, thereby positively regulating CD99L2 trafficking to cell surfaces.\",\n      \"method\": \"Bimolecular fluorescence complementation (BiFC), co-immunoprecipitation, FRET assays, cytoplasmic domain mutant analysis, analysis of endogenous CD99L2 surface levels in CD99-deficient cells and rescue by exogenous wild-type CD99\",\n      \"journal\": \"Journal of immunology (Baltimore, Md. : 1950)\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 / Strong — three independent methods (BiFC, Co-IP, FRET) plus domain mutagenesis and rescue experiments establishing the cytoplasmic-domain-dependent mechanism\",\n      \"pmids\": [\"24133166\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2013,\n      \"finding\": \"Knockdown of CD99L2 in murine B lymphoma (A20) cells results in G2 phase prolongation, reduced proliferative ability, H/RS-cell-like morphology, increased CD30 and CD15 expression, and upregulation of cytokines; NF-κB pathway activation is associated with mCD99L2 downregulation.\",\n      \"method\": \"shRNA knockdown, flow cytometry, MTT assay, western blot, antibody arrays, in vivo tumor growth assay\",\n      \"journal\": \"Oncology reports\",\n      \"confidence\": \"Low\",\n      \"confidence_rationale\": \"Tier 3 / Weak — single lab, multiple phenotypic readouts but NF-κB pathway involvement is correlative (western blot only), no direct mechanistic dissection\",\n      \"pmids\": [\"23338758\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2015,\n      \"finding\": \"CD99L2 knockout mice have a greater than 80% block in neutrophil infiltration and near-complete block in monocyte emigration in the thioglycollate peritonitis model; CD99L2 deficiency did not affect circulating leukocyte subset numbers, nor expression of ICAM-1, PECAM, or CD99 on endothelial cells.\",\n      \"method\": \"CD99L2 knockout mouse, thioglycollate peritonitis model, leukocyte counts, flow cytometry for endothelial adhesion molecule expression\",\n      \"journal\": \"Experimental and molecular pathology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — genetic knockout with defined in vivo inflammatory phenotype, replicated findings from a second independently generated KO line consistent with prior reports\",\n      \"pmids\": [\"26321243\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2018,\n      \"finding\": \"Tie-2-Cre-driven endothelial deletion of CD99L2 inhibits leukocyte entry into the CNS during EAE by blocking diapedesis through the endothelial basement membrane of BBB vessels, with accumulation of leukocytes between endothelial cells and the basement membrane, while luminal leukocyte docking was unaffected.\",\n      \"method\": \"Conditional knockout (Tie-2-Cre), EAE model (MOG35-55 immunization), 3D vibratome confocal microscopy, histopathological analysis of demyelination and inflammatory foci\",\n      \"journal\": \"Journal of leukocyte biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — conditional genetic knockout with specific mechanistic localization of defect (basement membrane crossing, not luminal adhesion) using 3D imaging\",\n      \"pmids\": [\"29791026\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"Human CD99L2 is constitutively expressed at endothelial cell borders and on leukocytes, and regulates a specific sequential step of transendothelial migration between PECAM and CD99; it promotes transmigration by recruiting the lateral border recycling compartment (LBRC) to TEM sites specifically downstream of PECAM initiation.\",\n      \"method\": \"Antibody blockade of human CD99L2, genetic knockdown in primary human endothelial cells, in vitro TEM assays with primary human neutrophils and monocytes, LBRC recruitment analysis\",\n      \"journal\": \"Journal of immunology (Baltimore, Md. : 1950)\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — antibody blockade and genetic knockdown combined with LBRC recruitment assay, defined sequential pathway position between PECAM and CD99, using primary human cells with multiple orthogonal methods\",\n      \"pmids\": [\"35914838\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"cd99l2 knockout in zebrafish (TALEN-mediated) reduces recruitment of granulocytes and macrophages to wound sites after caudal fin damage, and decreases expression of mfap4 (a macrophage marker), indicating a role for CD99L2 in leukocyte interstitial migration in addition to extravasation.\",\n      \"method\": \"TALEN knockout in zebrafish, caudal fin injury model, transgenic fluorescent reporter lines, RNA-seq, mfap4 expression analysis\",\n      \"journal\": \"Yi chuan = Hereditas\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Weak — genetic knockout with defined wound-response phenotype in zebrafish, single lab, one publication\",\n      \"pmids\": [\"36384956\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"P. gingivalis gingipains downregulate CD99 and CD99L2 expression on endothelial cells and leukocytes through inhibition of the PI3K/Akt pathway, impairing monocyte transendothelial migration while promoting monocyte adhesion.\",\n      \"method\": \"In vitro gingipain stimulation, in vivo mouse infection model, TEM assay, PI3K/Akt pathway inhibitor analysis, flow cytometry for CD99/CD99L2 surface expression\",\n      \"journal\": \"Microbiology spectrum\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — in vitro and in vivo convergent evidence for gingipain-mediated downregulation through PI3K/Akt, but PI3K pathway involvement is pharmacological without direct mechanistic dissection of CD99L2\",\n      \"pmids\": [\"37199607\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"CD99L2 positively regulates neurite outgrowth and excitatory synapse development in neurons; it inversely regulates immediate-early gene expression (Arc, Egr1, c-Fos) by inhibiting CREB and SRF transcription factor activity; neuronal inactivation promotes CD99L2 transport to the cell surface from recycling endosomes; CD99L2 knockout mice exhibit impaired excitatory synaptic transmission and plasticity in the hippocampus, and deficits in spatial memory and contextual fear conditioning.\",\n      \"method\": \"CD99L2 knockout mice, neurite outgrowth assays, synaptic transmission electrophysiology, immediate-early gene expression analysis, CREB/SRF activity assays, live-cell imaging of CD99L2 trafficking from recycling endosomes, behavioral tests\",\n      \"journal\": \"Cell reports\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — genetic knockout with multiple orthogonal functional readouts (electrophysiology, gene expression, live imaging, behavior), single lab but convergent methods establishing mechanism\",\n      \"pmids\": [\"39808524\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2026,\n      \"finding\": \"CD99L2 is a mechanoresponsive adhesion regulator transiently induced in primitive erythrocytes by shear stress-activated Piezo1 signaling; CD99L2 binds and anchors β-catenin at the plasma membrane, and its loss leads to nuclear translocation of β-catenin, Rap1 signaling activation, persistent adhesion molecule expression, erythrocyte retention, impaired maturation, and hemolytic anemia.\",\n      \"method\": \"Zebrafish and mouse models, CD99L2 knockout, Piezo1 activation experiments, co-immunoprecipitation for CD99L2-β-catenin interaction, β-catenin nuclear translocation assays, Rap1 signaling analysis\",\n      \"journal\": \"Cell reports\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 / Strong — multiple orthogonal methods (KO in two species, binding assay, pathway analysis) demonstrating CD99L2-β-catenin physical interaction and downstream signaling consequences\",\n      \"pmids\": [\"41915472\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2026,\n      \"finding\": \"CD99L2 protein occurs mainly in a ubiquitinated form and physically interacts with and activates the calcium-dependent protease CAPN1; ablation of cytoplasmic or extracellular domains of CD99L2 leads to intracellular mislocalization and loss of CAPN1 interaction; loss-of-function variants in CD99L2 cause X-linked spastic ataxia, with transcriptome analysis in patient fibroblasts revealing synaptic function-specific disturbances.\",\n      \"method\": \"Exome/genome sequencing in patient cohort, cellular studies of CD99L2 domain-deletion mutants (localization by microscopy), co-immunoprecipitation for CD99L2-CAPN1 interaction, ubiquitination analysis, transcriptome analysis in patient-derived fibroblasts\",\n      \"journal\": \"Nature communications\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 / Strong — domain mutagenesis establishing domain requirements for localization and CAPN1 interaction, patient cellular studies, transcriptome analysis; multiple orthogonal methods in a single rigorous study\",\n      \"pmids\": [\"41690933\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"CD99L2 is a type I transmembrane glycoprotein that functions as a homophilic adhesion molecule and multi-functional signaling regulator: in the vasculature, endothelial CD99L2 is essential for leukocyte diapedesis through the endothelial basement membrane, acting sequentially between PECAM-1 and CD99 by recruiting the lateral border recycling compartment; CD99 physically interacts with CD99L2 via its cytoplasmic domain to promote CD99L2 trafficking to the plasma membrane; in neurons, CD99L2 promotes excitatory synapse development, anchors β-catenin at the membrane to suppress its nuclear translocation, and inversely regulates immediate-early gene expression via CREB/SRF inhibition; in erythroid development, shear stress-activated Piezo1 induces CD99L2 expression, which restrains β-catenin nuclear signaling and Rap1 activation to allow erythrocyte de-adhesion; and CD99L2 also serves as an activating interactor for the protease CAPN1, with loss-of-function variants causing X-linked spastic ataxia.\"\n}\n```","stage2_raw":"{\n  \"mechanistic_narrative\": \"CD99L2 is a type I transmembrane glycoprotein and homophilic, divalent-cation-dependent cell-cell adhesion molecule that functions across vascular, neuronal, and erythroid contexts [#1]. Its best-characterized role is in leukocyte extravasation: endothelial CD99L2 — not the leukocyte pool — is the critical functional protein required for neutrophil, monocyte, and T cell diapedesis, acting specifically at the step of crossing the endothelial basement membrane such that its blockade traps leukocytes between endothelial cells and the basement membrane [#2, #3, #6, #7]. In human transendothelial migration it operates as a sequential gateway downstream of PECAM and upstream of CD99, promoting transmigration by recruiting the lateral border recycling compartment to migration sites [#8]. CD99L2 heterodimerizes with its paralog CD99 in a manner dependent on the CD99 cytoplasmic domain, and this interaction promotes CD99L2 trafficking to the plasma membrane [#4]. In neurons, CD99L2 promotes neurite outgrowth and excitatory synapse development, traffics to the surface from recycling endosomes upon neuronal inactivation, and inversely regulates immediate-early gene expression by inhibiting CREB and SRF activity, with knockout mice showing impaired hippocampal plasticity and memory deficits [#11]. In erythroid development it is a mechanoresponsive regulator induced by shear-stress-activated Piezo1 signaling that anchors \\u03b2-catenin at the plasma membrane to restrain its nuclear translocation and Rap1 activation, enabling erythrocyte de-adhesion and maturation [#12]. CD99L2 also physically interacts with and activates the calcium-dependent protease CAPN1, and loss-of-function variants in CD99L2 cause X-linked spastic ataxia [#13].\",\n  \"teleology\": [\n    {\n      \"year\": 2003,\n      \"claim\": \"Established the molecular identity of CD99L2 as an X-linked CD99 paralog, framing it as a candidate adhesion molecule before any function was tested.\",\n      \"evidence\": \"cDNA cloning, genomic organization, comparative sequence alignment and in situ hybridization\",\n      \"pmids\": [\"12706889\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"No functional assay of the conserved motifs\", \"Surface adhesion behavior not yet tested\"]\n    },\n    {\n      \"year\": 2007,\n      \"claim\": \"Demonstrated CD99L2 is a functional adhesion molecule, answering whether the CD99 paralog had adhesive activity, and linked it to inflammatory leukocyte recruitment.\",\n      \"evidence\": \"Homophilic aggregation assay in transfected L cell fibroblasts plus in vivo blocking antibody inflammation model\",\n      \"pmids\": [\"18163232\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Did not localize the step of extravasation affected\", \"Did not distinguish endothelial vs leukocyte CD99L2\"]\n    },\n    {\n      \"year\": 2010,\n      \"claim\": \"Positioned CD99/CD99L2 in the diapedesis cascade by showing they act independently of and downstream of PECAM-1, with blockade trapping neutrophils at the basement membrane.\",\n      \"evidence\": \"Blocking antibodies, PECAM-1 gene disruption, electron microscopy and 3D confocal imaging in inflamed cremaster tissue\",\n      \"pmids\": [\"20479283\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Did not resolve which cell type's CD99L2 is required\", \"Molecular mechanism of basement membrane crossing unknown\"]\n    },\n    {\n      \"year\": 2013,\n      \"claim\": \"Resolved that endothelial, not myeloid, CD99L2 is the functional pool for leukocyte recruitment, settling the cell-of-origin question.\",\n      \"evidence\": \"Endothelial- and myeloid-specific conditional knockout mice across cremaster, peritoneum and skin inflammation models\",\n      \"pmids\": [\"23293350\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Did not define molecular partners at the basement membrane step\"]\n    },\n    {\n      \"year\": 2013,\n      \"claim\": \"Defined the CD99-CD99L2 physical interaction and its consequence, showing CD99 heterodimerizes with CD99L2 via its cytoplasmic domain to drive CD99L2 surface trafficking.\",\n      \"evidence\": \"BiFC, co-immunoprecipitation, FRET, cytoplasmic domain mutagenesis, and rescue of surface CD99L2 in CD99-deficient cells\",\n      \"pmids\": [\"24133166\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Trafficking machinery downstream of CD99 binding not identified\", \"Stoichiometry in vivo not determined\"]\n    },\n    {\n      \"year\": 2013,\n      \"claim\": \"Probed CD99L2 in a B lymphoma context, linking its downregulation to altered proliferation, Hodgkin/Reed-Sternberg-like phenotype and NF-\\u03baB activation.\",\n      \"evidence\": \"shRNA knockdown in murine A20 cells with flow cytometry, MTT, western blot, antibody arrays and in vivo tumor assay\",\n      \"pmids\": [\"23338758\"],\n      \"confidence\": \"Low\",\n      \"gaps\": [\"NF-\\u03baB involvement is correlative (western blot only)\", \"No direct mechanistic dissection\", \"Single lab, not independently confirmed\"]\n    },\n    {\n      \"year\": 2015,\n      \"claim\": \"Confirmed the magnitude of the extravasation defect with a second knockout line and excluded indirect effects on circulating leukocytes or other endothelial adhesion molecules.\",\n      \"evidence\": \"CD99L2 knockout mouse thioglycollate peritonitis model with leukocyte counts and flow cytometry for ICAM-1, PECAM, CD99\",\n      \"pmids\": [\"26321243\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Mechanism of basement-membrane-specific block still undefined\"]\n    },\n    {\n      \"year\": 2018,\n      \"claim\": \"Extended the diapedesis role to the blood-brain barrier, showing endothelial CD99L2 is required for leukocyte entry into the CNS during neuroinflammation at the basement membrane step.\",\n      \"evidence\": \"Tie-2-Cre endothelial conditional knockout in the EAE model with 3D vibratome confocal imaging and histopathology\",\n      \"pmids\": [\"29791026\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Did not identify the basement membrane binding partner\", \"Luminal docking explicitly unaffected, leaving the molecular trigger of crossing open\"]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Placed human CD99L2 precisely in the TEM cascade between PECAM and CD99 and identified its mechanism as LBRC recruitment downstream of PECAM initiation.\",\n      \"evidence\": \"Antibody blockade and genetic knockdown in primary human endothelial cells with TEM assays and LBRC recruitment analysis\",\n      \"pmids\": [\"35914838\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Molecular link between CD99L2 and LBRC trafficking machinery unresolved\"]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Showed a conserved CD99L2 role in interstitial leukocyte migration beyond extravasation using a non-mammalian system.\",\n      \"evidence\": \"TALEN cd99l2 knockout in zebrafish with caudal fin injury, fluorescent reporter lines, RNA-seq and mfap4 analysis\",\n      \"pmids\": [\"36384956\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Single lab, one publication\", \"Mechanism of interstitial migration defect not dissected\"]\n    },\n    {\n      \"year\": 2023,\n      \"claim\": \"Identified a pathogen-driven regulatory input, showing P. gingivalis gingipains suppress CD99/CD99L2 via the PI3K/Akt pathway to impair monocyte TEM.\",\n      \"evidence\": \"In vitro gingipain stimulation, in vivo infection, TEM assay, PI3K/Akt inhibitor analysis and flow cytometry\",\n      \"pmids\": [\"37199607\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"PI3K pathway involvement is pharmacological without direct mechanistic dissection of CD99L2 regulation\"]\n    },\n    {\n      \"year\": 2025,\n      \"claim\": \"Revealed a distinct neuronal function, showing CD99L2 promotes synapse development and inversely controls immediate-early gene expression via CREB/SRF inhibition, with behavioral consequences.\",\n      \"evidence\": \"CD99L2 knockout mice with neurite/synapse assays, electrophysiology, immediate-early gene and CREB/SRF activity assays, recycling-endosome trafficking imaging and behavioral tests\",\n      \"pmids\": [\"39808524\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Direct molecular link from membrane CD99L2 to CREB/SRF not defined\", \"Neuronal adhesion partner not identified\"]\n    },\n    {\n      \"year\": 2026,\n      \"claim\": \"Connected CD99L2 to mechanotransduction and Wnt signaling, showing Piezo1-induced CD99L2 anchors \\u03b2-catenin at the membrane to enable erythrocyte de-adhesion and prevent hemolytic anemia.\",\n      \"evidence\": \"Zebrafish and mouse knockouts, Piezo1 activation, co-IP for CD99L2-\\u03b2-catenin, \\u03b2-catenin nuclear translocation and Rap1 signaling analysis\",\n      \"pmids\": [\"41915472\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Structural basis of CD99L2-\\u03b2-catenin binding unknown\", \"How the same protein anchors \\u03b2-catenin across cell types not unified\"]\n    },\n    {\n      \"year\": 2026,\n      \"claim\": \"Identified CD99L2 as a CAPN1 activator and the cause of X-linked spastic ataxia, linking domain requirements, ubiquitination, and a Mendelian disease.\",\n      \"evidence\": \"Patient exome/genome sequencing, domain-deletion mutant localization studies, co-IP for CD99L2-CAPN1, ubiquitination and patient fibroblast transcriptome analysis\",\n      \"pmids\": [\"41690933\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"How CD99L2 activates CAPN1 enzymatically not resolved\", \"Role of CD99L2 ubiquitination in function unclear\", \"Substrates downstream of CAPN1 activation not defined\"]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"It remains unknown how a single adhesion molecule mechanistically reconciles its diverse roles — basement-membrane crossing, LBRC recruitment, \\u03b2-catenin anchoring, CREB/SRF control, and CAPN1 activation — and whether a unifying biochemical activity underlies them.\",\n      \"evidence\": \"No single study in the corpus reconciles the vascular, neuronal, erythroid and protease-activating functions\",\n      \"pmids\": [],\n      \"confidence\": \"Low\",\n      \"gaps\": [\"No structural model of CD99L2\", \"Direct extracellular ligand at the basement membrane unidentified\", \"Mechanism connecting surface adhesion to intracellular signaling outputs unknown\"]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0098631\", \"supporting_discovery_ids\": [1]},\n      {\"term_id\": \"GO:0098772\", \"supporting_discovery_ids\": [12, 13]},\n      {\"term_id\": \"GO:0060089\", \"supporting_discovery_ids\": [8, 11]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005886\", \"supporting_discovery_ids\": [4, 8, 12]},\n      {\"term_id\": \"GO:0005768\", \"supporting_discovery_ids\": [11]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-168256\", \"supporting_discovery_ids\": [2, 3, 6, 7, 8]},\n      {\"term_id\": \"R-HSA-112316\", \"supporting_discovery_ids\": [11]},\n      {\"term_id\": \"R-HSA-162582\", \"supporting_discovery_ids\": [12]},\n      {\"term_id\": \"R-HSA-1643685\", \"supporting_discovery_ids\": [13]}\n    ],\n    \"complexes\": [],\n    \"partners\": [\"CD99\", \"CTNNB1\", \"CAPN1\"],\n    \"other_free_text\": []\n  }\n}","audit_flag":null,"evaluation":{"pairwise":"win","faith_supported":7,"faith_total":7,"faith_pct":100.0}}