{"gene":"CALM3","run_date":"2026-04-28T17:28:52","timeline":{"discoveries":[{"year":1992,"finding":"The crystal structure of recombinant calmodulin (the protein encoded by CALM1/2/3, all identical) was determined at 1.7 Å resolution, revealing a dumbbell-shaped molecule with two lobes connected by a central alpha-helix, each lobe containing three alpha-helices and two Ca2+-binding EF-hand loops with a short antiparallel beta-sheet. Four calcium ions are coordinated within the structure.","method":"X-ray crystallography (molecular replacement, R-factor 0.216 at 1.7 Å)","journal":"Journal of Molecular Biology","confidence":"High","confidence_rationale":"Tier 1 — high-resolution crystal structure with full atomic model and rigorous crystallographic validation","pmids":["1474585"],"is_preprint":false},{"year":1993,"finding":"The three human calmodulin genes CALM1, CALM2, and CALM3 were chromosomally localized to 14q24-q31, 2p21.1-p21.3, and 19q13.2-q13.3, respectively, by PCR-based amplification from human-hamster cell hybrids and in situ hybridization, establishing that these structurally related genes were dispersed throughout the genome from an ancestral precursor.","method":"PCR on human-hamster somatic cell hybrids; in situ hybridization on metaphase spreads","journal":"Genomics","confidence":"High","confidence_rationale":"Tier 2 — two orthogonal direct experimental methods (PCR hybrid panel + in situ hybridization)","pmids":["8314583"],"is_preprint":false},{"year":1998,"finding":"CALM3 is at least 5-fold more actively transcribed than CALM1 or CALM2 in proliferating human teratoma cells, as measured by direct mRNA abundance and transcriptional activity assays. Transient transfection with luciferase reporter genes showed that the 5' untranslated regions of CALM genes are necessary for full promoter activation, with CALM3 promoter activity correlating with its higher endogenous transcriptional activity.","method":"Direct mRNA abundance measurement; nuclear run-on transcription assay; luciferase reporter transfection","journal":"Cell Calcium","confidence":"High","confidence_rationale":"Tier 1-2 — multiple orthogonal methods (run-on assay, mRNA quantification, reporter assay) in single study","pmids":["9681195"],"is_preprint":false},{"year":2016,"finding":"The novel CALM3 variant E141G causes an 11-fold reduction in Ca2+-binding affinity and produces a functionally dominant loss of inactivation in the cardiac L-type calcium channel CaV1.2, with mild accentuation of NaV1.5 late current, but no effect on intracellular RyR2-mediated calcium release, establishing a specific functional hierarchy among calmodulin's cardiac ion channel targets.","method":"In vitro Ca2+-binding affinity measurement; whole-cell patch-clamp of CaV1.2 (ICaL inactivation); NaV1.5 late current recording; cardiomyocyte RyR2 calcium release assay","journal":"Circulation. Cardiovascular Genetics","confidence":"High","confidence_rationale":"Tier 1-2 — multiple orthogonal electrophysiological and biochemical assays characterizing mechanism across three distinct ion channel targets","pmids":["26969752"],"is_preprint":false},{"year":2016,"finding":"The CALM3-A103V variant (associated with CPVT) modestly reduces CaM Ca2+-binding affinity (3-fold) but does not alter CaM binding to RyR2. Instead, A103V-CaM promotes spontaneous Ca2+ wave and spark activity in permeabilized cardiomyocytes by activating RyR2, with functional dominance demonstrated at a 1:3 A103V-CaM:WT-CaM ratio, and causes delayed afterdepolarizations and triggered beats in intact cardiomyocytes without significantly affecting CaV1.2 inactivation or action potential duration.","method":"In vitro Ca2+-binding affinity measurement; RyR2-CaM binding assay; confocal Ca2+ spark/wave imaging in permeabilized cardiomyocytes; whole-cell patch-clamp (ICaL, APD); intact cardiomyocyte electrophysiology","journal":"Circulation. Arrhythmia and Electrophysiology","confidence":"High","confidence_rationale":"Tier 1-2 — multiple orthogonal methods including biochemical binding, live-cell calcium imaging, and electrophysiology with defined mechanistic phenotype","pmids":["27516456"],"is_preprint":false},{"year":2017,"finding":"Calm3 mRNA is localized to neuronal dendrites via a retained intron in its 3'-UTR, which serves as the critical cis-element recruiting the RNA-binding protein Staufen2 (Stau2). Stau2 mediates dendritic localization of Calm3 mRNA without affecting its stability. NMDA receptor-mediated synaptic activity specifically promotes dendritic localization of the intron-containing Calm3 isoform, while inhibition of synaptic activity reduces it.","method":"iCLIP (individual-nucleotide resolution CLIP); dendritic localization assays in neurons; Stau2 knockdown; NMDA stimulation/inhibition experiments; fluorescence in situ hybridization","journal":"EMBO Reports","confidence":"High","confidence_rationale":"Tier 2 — multiple orthogonal methods (iCLIP, FISH, KD, activity manipulation) establishing mechanism of activity-dependent mRNA localization","pmids":["28765142"],"is_preprint":false},{"year":2022,"finding":"The CALM3 variant p.N138K reduces Ca2+-binding affinity of the CaM C-terminal domain 10-fold relative to WT-CaM, slows CaV1.2 (ICaL) inactivation (less severely than p.D130G-CaM), and unexpectedly potentiates IKs (slow delayed rectifier potassium current) density—an effect not seen with p.D130G-CaM—providing a mechanistic explanation for the milder and variable LQTS phenotype in this 4-generation family.","method":"Stoichiometric Ca2+ titration; equilibrium Ca2+ titration; whole-cell patch-clamp (ICaL and IKs); optical fluorescence assay for Cav1.2 and Kv7.1 membrane expression","journal":"Circulation. Arrhythmia and Electrophysiology","confidence":"High","confidence_rationale":"Tier 1-2 — multiple orthogonal biochemical and electrophysiological methods characterizing mechanism across two ion channel targets","pmids":["35225649"],"is_preprint":false},{"year":2024,"finding":"A suppression-and-replacement (SupRep) gene therapy using a single construct containing CALM1-, CALM2-, and CALM3-targeting shRNAs (achieving 86%, 71%, and 94% knockdown respectively) plus a shRNA-immune CALM1 cDNA rescued pathologically prolonged action potential duration (APD90) in iPSC-derived cardiomyocytes from patients with CALM1-F142L, CALM2-D130G, and CALM3-D130G calmodulinopathy, demonstrating functional interchangeability of the three CALM gene products.","method":"shRNA knockdown efficiency assay (RT-qPCR in TSA201 cells); voltage-sensing dye APD90 measurement in patient-derived iPSC-cardiomyocytes; transfection of SupRep construct","journal":"Circulation. Arrhythmia and Electrophysiology","confidence":"Medium","confidence_rationale":"Tier 2 — defined cellular phenotypic rescue with functional readout in disease-relevant cell model, single study","pmids":["39069900"],"is_preprint":false},{"year":2024,"finding":"CALM3 (calmodulin 3) promotes neuronal differentiation of human umbilical cord mesenchymal stem cells (hUC-MSCs); knockdown of CALM3 partially reversed DMSO/BHA/DMEM-induced upregulation of neuronal markers (NSE, NeuN, NF-M). Additionally, hsa-miR-543 negatively regulates CALM3 expression and opposes neuronal differentiation, placing CALM3 downstream of miR-543 in a regulatory axis controlling neurogenesis.","method":"RT-qPCR; western blotting; siRNA knockdown of CALM3; miR-543 mimic transfection; neural induction assay","journal":"International Journal of Developmental Neuroscience","confidence":"Medium","confidence_rationale":"Tier 2-3 — KD with defined phenotypic readout and upstream regulatory placement, but single lab, single study","pmids":["39444227"],"is_preprint":false},{"year":2026,"finding":"CALM3 contributes only ~11% of the total calmodulin protein produced in the human left ventricle (compared to ~45% for CALM1 and ~44% for CALM2), as determined by ribosome profiling of left ventricular tissue, and is under less negative selection than CALM1 or CALM2. This lower translational contribution mechanistically explains why CALM3 missense variant carriers experience less severe cardiac events than those with CALM1 or CALM2 variants.","method":"Ribosome profiling (left ventricle tissue from GTEx); RNA sequencing (GTEx, 49 tissues); gnomAD population variant analysis; International Calmodulinopathy Registry clinical data","journal":"Europace","confidence":"Medium","confidence_rationale":"Tier 2 — ribosome profiling provides direct translational quantification; mechanistic inference supported by population genetics and clinical data","pmids":["41846582"],"is_preprint":false}],"current_model":"CALM3 is one of three human genes encoding the identical calmodulin protein (a dumbbell-shaped, four-EF-hand Ca2+-binding protein); it contributes only ~11% of cardiac calmodulin protein due to lower translational efficiency, its mRNA is transported to neuronal dendrites via a retained 3'-UTR intron recognized by Staufen2 in an NMDA-activity-dependent manner, and pathogenic CALM3 missense variants (e.g., E141G, A103V, N138K, D130G) reduce Ca2+-binding affinity and dysregulate cardiac ion channels—specifically impairing CaV1.2 inactivation, augmenting NaV1.5 late current, and/or activating RyR2-mediated Ca2+ release—to cause arrhythmia syndromes (LQTS, CPVT) that can be rescued by suppression-replacement gene therapy."},"narrative":{"teleology":[{"year":1992,"claim":"Determination of the high-resolution crystal structure of calmodulin established the bilobed, four-EF-hand architecture that underpins all subsequent structure–function studies of CALM3-encoded protein.","evidence":"X-ray crystallography at 1.7 Å resolution of recombinant calmodulin","pmids":["1474585"],"confidence":"High","gaps":["Structure solved without a bound target peptide, leaving the activation mechanism unclear","Conformational flexibility of the central linker helix in solution was not captured"]},{"year":1993,"claim":"Mapping CALM3 to chromosome 19q13.2-q13.3 demonstrated that the three human calmodulin genes are dispersed across different chromosomes, raising the question of whether they are differentially regulated despite encoding an identical protein.","evidence":"PCR on human–hamster somatic cell hybrids and in situ hybridization on metaphase chromosomes","pmids":["8314583"],"confidence":"High","gaps":["No information on differential tissue-level expression or translational output","Evolutionary forces maintaining three identical coding sequences were uncharacterized"]},{"year":1998,"claim":"Quantification of CALM3 transcription revealed it is at least 5-fold more active than CALM1 or CALM2 in proliferating cells, establishing that the three genes contribute unequally to the calmodulin mRNA pool and that 5′-UTR elements drive this difference.","evidence":"Nuclear run-on, mRNA abundance measurement, and luciferase reporter assays in human teratoma cells","pmids":["9681195"],"confidence":"High","gaps":["Transcriptional dominance measured in one cell type; generalizability to heart or brain unknown","Post-transcriptional and translational regulation were not assessed"]},{"year":2016,"claim":"Functional characterization of CALM3 pathogenic variants (E141G, A103V) showed that different mutations target distinct cardiac ion channels—E141G dominantly impairs CaV1.2 inactivation and augments NaV1.5 late current (LQTS mechanism), while A103V activates RyR2-mediated spontaneous Ca²⁺ release (CPVT mechanism)—establishing a variant-specific functional hierarchy among calmodulin's channel targets.","evidence":"Ca²⁺-binding assays, whole-cell patch-clamp of CaV1.2 and NaV1.5, confocal Ca²⁺ spark/wave imaging and electrophysiology in cardiomyocytes","pmids":["26969752","27516456"],"confidence":"High","gaps":["Structural basis of variant-specific channel selectivity (e.g., why E141G spares RyR2) was not resolved","In vivo cardiac phenotypes were not modeled"]},{"year":2017,"claim":"Discovery that a retained 3′-UTR intron in CALM3 mRNA recruits Staufen2 for NMDA receptor activity-dependent dendritic transport revealed a neuron-specific post-transcriptional mechanism that could enable local calmodulin synthesis at activated synapses.","evidence":"iCLIP, FISH, Staufen2 knockdown, and NMDA stimulation/inhibition in neurons","pmids":["28765142"],"confidence":"High","gaps":["Whether dendritic CALM3 mRNA is locally translated and whether this pool is functionally distinct from somatic calmodulin was not tested","Relevance to in vivo synaptic plasticity or neurological disease was not assessed"]},{"year":2022,"claim":"The CALM3-N138K variant was shown to impair CaV1.2 inactivation and uniquely potentiate IKs current, providing a mechanistic basis for the milder and variable LQTS phenotype in a multigenerational family carrying this mutation.","evidence":"Stoichiometric and equilibrium Ca²⁺ titration; patch-clamp of ICaL and IKs; optical membrane expression assays","pmids":["35225649"],"confidence":"High","gaps":["Structural mechanism by which N138K potentiates IKs while D130G does not was unexplained","Effect on other calmodulin targets (CaMKII, calcineurin) was not examined"]},{"year":2024,"claim":"A triple-CALM suppression-and-replacement gene therapy construct rescued pathologically prolonged APD in patient iPSC-cardiomyocytes carrying CALM3-D130G, demonstrating that a single WT CALM1 cDNA can functionally compensate for all three CALM genes and validating gene therapy as a potential calmodulinopathy treatment strategy.","evidence":"shRNA knockdown (RT-qPCR) and voltage-sensing dye APD90 measurement in patient-derived iPSC-cardiomyocytes","pmids":["39069900"],"confidence":"Medium","gaps":["Efficacy demonstrated only in iPSC-CMs; in vivo delivery and long-term safety are untested","Whether chronic knockdown of endogenous CALM3 transcript affects neuron-specific functions (e.g., dendritic localization) was not addressed"]},{"year":2024,"claim":"CALM3 knockdown partially reverses neuronal differentiation of mesenchymal stem cells and is negatively regulated by miR-543, placing CALM3 in a regulatory axis controlling neurogenesis.","evidence":"siRNA knockdown of CALM3 and miR-543 mimic transfection with RT-qPCR and western blot readout in hUC-MSCs","pmids":["39444227"],"confidence":"Medium","gaps":["Single-study finding in a non-neuronal progenitor system; relevance to physiological neurogenesis is unconfirmed","Whether the effect is CALM3-specific or reflects total calmodulin levels was not distinguished"]},{"year":2026,"claim":"Ribosome profiling of human left ventricle revealed CALM3 contributes only ~11% of total calmodulin protein despite substantial mRNA levels, explaining both its lower pathogenic penetrance and weaker negative selection compared to CALM1 and CALM2.","evidence":"Ribosome profiling and RNA-seq from GTEx left ventricular tissue; gnomAD constraint analysis; International Calmodulinopathy Registry clinical correlation","pmids":["41846582"],"confidence":"Medium","gaps":["Translational efficiency measured in bulk left ventricle; cell-type-resolved (cardiomyocyte vs. fibroblast) data are lacking","Cis-regulatory elements responsible for lower translational output have not been identified"]},{"year":null,"claim":"Key open questions include the structural basis for variant-specific ion channel selectivity, whether dendritically localized CALM3 mRNA is locally translated to support synaptic plasticity, and the in vivo efficacy of suppression-and-replacement gene therapy for calmodulinopathy.","evidence":"","pmids":[],"confidence":"Low","gaps":["No co-crystal structures of calmodulinopathy mutant CaM with CaV1.2, RyR2, or KCNQ1","Local translation from dendritic CALM3 mRNA has not been demonstrated","In vivo gene therapy efficacy and safety data are absent"]}],"mechanism_profile":{"molecular_activity":[{"term_id":"GO:0098772","term_label":"molecular function regulator activity","supporting_discovery_ids":[3,4,6]},{"term_id":"GO:0140299","term_label":"molecular sensor activity","supporting_discovery_ids":[0,3,4]}],"localization":[{"term_id":"GO:0005829","term_label":"cytosol","supporting_discovery_ids":[3,4]}],"pathway":[{"term_id":"R-HSA-162582","term_label":"Signal Transduction","supporting_discovery_ids":[3,4,6]},{"term_id":"R-HSA-112316","term_label":"Neuronal System","supporting_discovery_ids":[5]},{"term_id":"R-HSA-397014","term_label":"Muscle contraction","supporting_discovery_ids":[3,6]}],"complexes":[],"partners":["CACNA1C","SCN5A","RYR2","KCNQ1","STAU2"],"other_free_text":[]},"mechanistic_narrative":"CALM3 is one of three human genes encoding the identical 148-amino-acid calmodulin protein, a dumbbell-shaped Ca²⁺ sensor with four EF-hand domains that transduces calcium signals to regulate cardiac ion channels, neuronal mRNA trafficking, and cell differentiation [PMID:1474585, PMID:8314583]. Despite being the most transcriptionally active CALM gene in some cell types, CALM3 contributes only ~11% of total calmodulin protein in the human left ventricle owing to lower translational efficiency, which correlates with milder clinical phenotypes in CALM3 variant carriers compared with CALM1 or CALM2 [PMID:9681195, PMID:41846582]. Pathogenic CALM3 missense variants (E141G, A103V, N138K, D130G) reduce C-lobe Ca²⁺-binding affinity and dominantly dysregulate specific cardiac ion channels—impairing CaV1.2 inactivation, augmenting NaV1.5 late current, or activating RyR2-mediated Ca²⁺ release—causing long QT syndrome and catecholaminergic polymorphic ventricular tachycardia (calmodulinopathy), phenotypes rescuable by suppression-and-replacement gene therapy in patient iPSC-cardiomyocytes [PMID:26969752, PMID:27516456, PMID:35225649, PMID:39069900]. In neurons, a retained 3′-UTR intron in CALM3 mRNA recruits Staufen2 for NMDA receptor activity-dependent dendritic localization, providing a mechanism for local calmodulin synthesis at synapses [PMID:28765142]."},"prefetch_data":{"uniprot":{"accession":"P0DP25","full_name":"Calmodulin-3","aliases":[],"length_aa":149,"mass_kda":16.8,"function":"Calmodulin acts as part of a calcium signal transduction pathway by mediating the control of a large number of enzymes, ion channels, aquaporins and other proteins through calcium-binding (PubMed:16760425, PubMed:31454269). Calcium-binding is required for the activation of calmodulin (PubMed:16760425, PubMed:31454269, PubMed:35568036). Among the enzymes to be stimulated by the calmodulin-calcium complex are a number of protein kinases, such as myosin light-chain kinases and calmodulin-dependent protein kinase type II (CaMK2), and phosphatases (PubMed:16760425, PubMed:35568036). Together with CCP110 and centrin, is involved in a genetic pathway that regulates the centrosome cycle and progression through cytokinesis (PubMed:16760425) (Microbial infection) Required for C.violaceum CopC and S.flexneri OspC3 arginine ADP-riboxanase activity","subcellular_location":"Cytoplasm, cytoskeleton, spindle; Cytoplasm, cytoskeleton, spindle pole; Cytoplasm, cytoskeleton, microtubule organizing center, centrosome","url":"https://www.uniprot.org/uniprotkb/P0DP25/entry"},"depmap":{"release":"DepMap","has_data":true,"is_common_essential":false,"resolved_as":"","url":"https://depmap.org/portal/gene/CALM3","classification":"Not Classified","n_dependent_lines":4,"n_total_lines":1208,"dependency_fraction":0.0033112582781456954},"opencell":{"profiled":true,"resolved_as":"","ensg_id":"ENSG00000160014","cell_line_id":"CID000513","localizations":[{"compartment":"centrosome","grade":3},{"compartment":"membrane","grade":3},{"compartment":"cytoplasmic","grade":2},{"compartment":"nucleoplasm","grade":2}],"interactors":[{"gene":"CALM2;CALM3;CALM1","stoichiometry":10.0},{"gene":"MYL6","stoichiometry":10.0},{"gene":"UNNAMED","stoichiometry":0.2},{"gene":"FLNB","stoichiometry":0.2},{"gene":"WFS1","stoichiometry":0.2},{"gene":"SCO1","stoichiometry":0.2},{"gene":"KIF1B","stoichiometry":0.2},{"gene":"KIF1C","stoichiometry":0.2},{"gene":"TOM1","stoichiometry":0.2},{"gene":"MYL1","stoichiometry":0.2}],"url":"https://opencell.sf.czbiohub.org/target/CID000513","total_profiled":1310},"omim":[{"mim_id":"618782","title":"LONG QT SYNDROME 16; LQT16","url":"https://www.omim.org/entry/618782"},{"mim_id":"616249","title":"LONG QT SYNDROME 15; LQT15","url":"https://www.omim.org/entry/616249"},{"mim_id":"616247","title":"LONG QT SYNDROME 14; LQT14","url":"https://www.omim.org/entry/616247"},{"mim_id":"604772","title":"VENTRICULAR TACHYCARDIA, CATECHOLAMINERGIC POLYMORPHIC, 1, WITH OR WITHOUT ATRIAL DYSFUNCTION AND/OR DILATED CARDIOMYOPATHY; CPVT1","url":"https://www.omim.org/entry/604772"},{"mim_id":"601146","title":"GROWTH/DIFFERENTIATION FACTOR 5; GDF5","url":"https://www.omim.org/entry/601146"}],"hpa":{"profiled":true,"resolved_as":"","reliability":"","locations":[],"tissue_specificity":"Tissue enhanced","tissue_distribution":"Detected in all","driving_tissues":[{"tissue":"brain","ntpm":1443.6}],"url":"https://www.proteinatlas.org/search/CALM3"},"hgnc":{"alias_symbol":["PHKD","PHKD3"],"prev_symbol":[]},"alphafold":{"accession":"P0DP25","domains":[{"cath_id":"1.10.238.10","chopping":"1-78","consensus_level":"high","plddt":83.4788,"start":1,"end":78},{"cath_id":"1.10.238.10","chopping":"94-146","consensus_level":"high","plddt":91.1945,"start":94,"end":146}],"viewer_url":"https://alphafold.ebi.ac.uk/entry/P0DP25","model_url":"https://alphafold.ebi.ac.uk/files/AF-P0DP25-F1-model_v6.cif","pae_url":"https://alphafold.ebi.ac.uk/files/AF-P0DP25-F1-predicted_aligned_error_v6.png","plddt_mean":85.5},"mouse_models":{"mgi_url":"https://www.informatics.jax.org/marker/summary?nomen=CALM3","jax_strain_url":"https://www.jax.org/strain/search?query=CALM3"},"sequence":{"accession":"P0DP25","fasta_url":"https://rest.uniprot.org/uniprotkb/P0DP25.fasta","uniprot_url":"https://www.uniprot.org/uniprotkb/P0DP25/entry","alphafold_viewer_url":"https://alphafold.ebi.ac.uk/entry/P0DP25"}},"corpus_meta":[{"pmid":"26969752","id":"PMC_26969752","title":"Spectrum and Prevalence of CALM1-, CALM2-, and CALM3-Encoded Calmodulin Variants in Long QT Syndrome and Functional Characterization of a Novel Long QT Syndrome-Associated Calmodulin Missense Variant, E141G.","date":"2016","source":"Circulation. Cardiovascular genetics","url":"https://pubmed.ncbi.nlm.nih.gov/26969752","citation_count":109,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"8314583","id":"PMC_8314583","title":"Localization of the human bona fide calmodulin genes CALM1, CALM2, and CALM3 to chromosomes 14q24-q31, 2p21.1-p21.3, and 19q13.2-q13.3.","date":"1993","source":"Genomics","url":"https://pubmed.ncbi.nlm.nih.gov/8314583","citation_count":80,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"27516456","id":"PMC_27516456","title":"Novel CPVT-Associated Calmodulin Mutation in CALM3 (CALM3-A103V) Activates Arrhythmogenic Ca Waves and Sparks.","date":"2016","source":"Circulation. Arrhythmia and electrophysiology","url":"https://pubmed.ncbi.nlm.nih.gov/27516456","citation_count":71,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"28765142","id":"PMC_28765142","title":"A retained intron in the 3'-UTR of Calm3 mRNA mediates its Staufen2- and activity-dependent localization to neuronal dendrites.","date":"2017","source":"EMBO reports","url":"https://pubmed.ncbi.nlm.nih.gov/28765142","citation_count":61,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"9681195","id":"PMC_9681195","title":"Characterization of the human CALM2 calmodulin gene and comparison of the transcriptional activity of CALM1, CALM2 and CALM3.","date":"1998","source":"Cell calcium","url":"https://pubmed.ncbi.nlm.nih.gov/9681195","citation_count":61,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"25089838","id":"PMC_25089838","title":"The novel regulations of MEF2A, CAMKK2, CALM3, and TNNI3 in ventricular hypertrophy induced by arsenic exposure in rats.","date":"2014","source":"Toxicology","url":"https://pubmed.ncbi.nlm.nih.gov/25089838","citation_count":26,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"35225649","id":"PMC_35225649","title":"Novel CALM3 Variant Causing Calmodulinopathy With Variable Expressivity in a 4-Generation Family.","date":"2022","source":"Circulation. Arrhythmia and electrophysiology","url":"https://pubmed.ncbi.nlm.nih.gov/35225649","citation_count":19,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"28744816","id":"PMC_28744816","title":"ACE2, CALM3 and TNNI3K polymorphisms as potential disease modifiers in hypertrophic and dilated cardiomyopathies.","date":"2017","source":"Molecular and cellular biochemistry","url":"https://pubmed.ncbi.nlm.nih.gov/28744816","citation_count":18,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39069900","id":"PMC_39069900","title":"Single Construct Suppression and Replacement Gene Therapy for the Treatment of All CALM1-, CALM2-, and CALM3-Mediated Arrhythmia Disorders.","date":"2024","source":"Circulation. Arrhythmia and electrophysiology","url":"https://pubmed.ncbi.nlm.nih.gov/39069900","citation_count":11,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"9237482","id":"PMC_9237482","title":"Mutation analysis of chromosome 19 calmodulin (CALM3) gene in Alzheimer's disease patients.","date":"1997","source":"Neuroscience letters","url":"https://pubmed.ncbi.nlm.nih.gov/9237482","citation_count":7,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"33214751","id":"PMC_33214751","title":"Molecular docking studies of a-mangostin with oral cancer targets ARRB1, FLNA, CALM3 and HTT.","date":"2020","source":"Bioinformation","url":"https://pubmed.ncbi.nlm.nih.gov/33214751","citation_count":4,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"25396734","id":"PMC_25396734","title":"Comprehensive analysis of the association of EGFR, CALM3 and SMARCD1 gene polymorphisms with BMD in Caucasian women.","date":"2014","source":"PloS one","url":"https://pubmed.ncbi.nlm.nih.gov/25396734","citation_count":2,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"30488312","id":"PMC_30488312","title":"Correction to: ACE2, CALM3 and TNNI3K polymorphisms as potential disease modifiers in hypertrophic and dilated cardiomyopathies.","date":"2019","source":"Molecular and cellular biochemistry","url":"https://pubmed.ncbi.nlm.nih.gov/30488312","citation_count":1,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39444227","id":"PMC_39444227","title":"Role of hsa-miR-543-KIF5C/CALM3 pathway in neuron differentiation of embryonic mesenchymal stem cells.","date":"2024","source":"International journal of developmental neuroscience : the official journal of the International Society for Developmental Neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/39444227","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"41846582","id":"PMC_41846582","title":"CALM1, CALM2, and CALM3 expression and translation efficiency provide insight into the severity of calmodulinopathy.","date":"2026","source":"Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology","url":"https://pubmed.ncbi.nlm.nih.gov/41846582","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2025.05.15.25327594","title":"<i>CALM1, CALM2,</i>and<i>CALM3</i>expression and translation efficiency provide insight into the severity of calmodulinopathy","date":"2025-05-15","source":"bioRxiv","url":"https://doi.org/10.1101/2025.05.15.25327594","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2025.04.01.646642","title":"Altered Protein Phosphorylation in a Novel Midbrain Organoid Model for Bipolar Disorder","date":"2025-04-01","source":"bioRxiv","url":"https://doi.org/10.1101/2025.04.01.646642","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":"23128233","id":"PMC_23128233","title":"Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease.","date":"2012","source":"Nature","url":"https://pubmed.ncbi.nlm.nih.gov/23128233","citation_count":3725,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"9278050","id":"PMC_9278050","title":"Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin.","date":"1997","source":"Nature","url":"https://pubmed.ncbi.nlm.nih.gov/9278050","citation_count":2324,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"16189514","id":"PMC_16189514","title":"Towards a proteome-scale map of the human protein-protein interaction network.","date":"2005","source":"Nature","url":"https://pubmed.ncbi.nlm.nih.gov/16189514","citation_count":2090,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"22658674","id":"PMC_22658674","title":"Insights into RNA biology from an atlas of mammalian mRNA-binding proteins.","date":"2012","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/22658674","citation_count":1718,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"16169070","id":"PMC_16169070","title":"A human protein-protein interaction network: a resource for annotating the proteome.","date":"2005","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/16169070","citation_count":1704,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"12477932","id":"PMC_12477932","title":"Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences.","date":"2002","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/12477932","citation_count":1479,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"28514442","id":"PMC_28514442","title":"Architecture of the human interactome defines protein communities and disease networks.","date":"2017","source":"Nature","url":"https://pubmed.ncbi.nlm.nih.gov/28514442","citation_count":1085,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"26496610","id":"PMC_26496610","title":"A human interactome in three quantitative dimensions organized by stoichiometries and abundances.","date":"2015","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/26496610","citation_count":1015,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"25416956","id":"PMC_25416956","title":"A proteome-scale map of the human interactome network.","date":"2014","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/25416956","citation_count":977,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"15592455","id":"PMC_15592455","title":"Immunoaffinity profiling of tyrosine phosphorylation in cancer cells.","date":"2004","source":"Nature biotechnology","url":"https://pubmed.ncbi.nlm.nih.gov/15592455","citation_count":916,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"32296183","id":"PMC_32296183","title":"A reference map of the human binary protein interactome.","date":"2020","source":"Nature","url":"https://pubmed.ncbi.nlm.nih.gov/32296183","citation_count":849,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"14743216","id":"PMC_14743216","title":"A physical and functional map of the human TNF-alpha/NF-kappa B signal transduction pathway.","date":"2004","source":"Nature cell biology","url":"https://pubmed.ncbi.nlm.nih.gov/14743216","citation_count":841,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"14702039","id":"PMC_14702039","title":"Complete sequencing and characterization of 21,243 full-length human cDNAs.","date":"2003","source":"Nature genetics","url":"https://pubmed.ncbi.nlm.nih.gov/14702039","citation_count":754,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"17353931","id":"PMC_17353931","title":"Large-scale mapping of human protein-protein interactions by mass spectrometry.","date":"2007","source":"Molecular systems biology","url":"https://pubmed.ncbi.nlm.nih.gov/17353931","citation_count":733,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"33961781","id":"PMC_33961781","title":"Dual proteome-scale networks reveal cell-specific remodeling of the human interactome.","date":"2021","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/33961781","citation_count":705,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"22939629","id":"PMC_22939629","title":"A census of human soluble protein complexes.","date":"2012","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/22939629","citation_count":689,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"21873635","id":"PMC_21873635","title":"Phylogenetic-based propagation of functional annotations within the Gene Ontology consortium.","date":"2011","source":"Briefings in bioinformatics","url":"https://pubmed.ncbi.nlm.nih.gov/21873635","citation_count":656,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"14630798","id":"PMC_14630798","title":"Characterization of the proteins released from activated platelets leads to localization of novel platelet proteins in human atherosclerotic lesions.","date":"2003","source":"Blood","url":"https://pubmed.ncbi.nlm.nih.gov/14630798","citation_count":616,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"1474585","id":"PMC_1474585","title":"Calmodulin structure refined at 1.7 A resolution.","date":"1992","source":"Journal of molecular biology","url":"https://pubmed.ncbi.nlm.nih.gov/1474585","citation_count":615,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"21565611","id":"PMC_21565611","title":"Mapping the NPHP-JBTS-MKS protein network reveals ciliopathy disease genes and pathways.","date":"2011","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/21565611","citation_count":507,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"9188442","id":"PMC_9188442","title":"Reciprocal regulation of endothelial nitric-oxide synthase by Ca2+-calmodulin and caveolin.","date":"1997","source":"The Journal of biological chemistry","url":"https://pubmed.ncbi.nlm.nih.gov/9188442","citation_count":468,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"15489334","id":"PMC_15489334","title":"The status, quality, and expansion of the NIH full-length cDNA project: the Mammalian Gene Collection (MGC).","date":"2004","source":"Genome research","url":"https://pubmed.ncbi.nlm.nih.gov/15489334","citation_count":438,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"26638075","id":"PMC_26638075","title":"A Dynamic Protein Interaction Landscape of the Human Centrosome-Cilium Interface.","date":"2015","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/26638075","citation_count":433,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"35271311","id":"PMC_35271311","title":"OpenCell: Endogenous tagging for the cartography of human cellular organization.","date":"2022","source":"Science (New York, N.Y.)","url":"https://pubmed.ncbi.nlm.nih.gov/35271311","citation_count":432,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"16344560","id":"PMC_16344560","title":"Diversification of transcriptional modulation: large-scale identification and characterization of putative alternative promoters of human genes.","date":"2005","source":"Genome research","url":"https://pubmed.ncbi.nlm.nih.gov/16344560","citation_count":409,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"17719545","id":"PMC_17719545","title":"Cep97 and CP110 suppress a cilia assembly program.","date":"2007","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/17719545","citation_count":386,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"14622581","id":"PMC_14622581","title":"The NMDA receptor is coupled to the ERK pathway by a direct interaction between NR2B and RasGRF1.","date":"2003","source":"Neuron","url":"https://pubmed.ncbi.nlm.nih.gov/14622581","citation_count":372,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"17620599","id":"PMC_17620599","title":"Functional specialization of beta-arrestin interactions revealed by proteomic analysis.","date":"2007","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/17620599","citation_count":360,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"21726808","id":"PMC_21726808","title":"Acetylation regulates gluconeogenesis by promoting PEPCK1 degradation via recruiting the UBR5 ubiquitin ligase.","date":"2011","source":"Molecular cell","url":"https://pubmed.ncbi.nlm.nih.gov/21726808","citation_count":337,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"17582331","id":"PMC_17582331","title":"The ankyrin repeats of TRPV1 bind multiple ligands and modulate channel sensitivity.","date":"2007","source":"Neuron","url":"https://pubmed.ncbi.nlm.nih.gov/17582331","citation_count":334,"is_preprint":false,"source_track":"gene2pubmed"}],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":10706,"output_tokens":2207,"usd":0.032612},"stage2":{"model":"claude-opus-4-6","input_tokens":5483,"output_tokens":2388,"usd":0.130672},"total_usd":0.488732,"stage1_batch_id":"msgbatch_011kfMfE81o6CTcyhVUqumuW","stage2_batch_id":"msgbatch_01DTputS6WNkH745x1nVSoBo","note":"batch pricing = 50% of standard","round2_stage1":{"model":"claude-sonnet-4-6","input_tokens":22690,"output_tokens":4025,"usd":0.064223},"round2_rules_fired":"R3","round2_stage2":{"model":"claude-opus-4-6","input_tokens":6565,"output_tokens":5653,"usd":0.261225}},"stage1_raw":"```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 2016,\n      \"finding\": \"CALM3-encoded calmodulin variant E141G causes an 11-fold reduction in Ca2+-binding affinity and produces a functionally dominant loss of inactivation in CaV1.2 (cardiac L-type calcium channel), mild accentuation of NaV1.5 late current, but no effect on RyR2-mediated intracellular calcium release.\",\n      \"method\": \"In vitro Ca2+-binding affinity measurements, whole-cell patch-clamp of CaV1.2 and NaV1.5, cardiomyocyte calcium release assays\",\n      \"journal\": \"Circulation. Cardiovascular genetics\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal functional assays (Ca2+-binding, patch-clamp, Ca2+ release) in a single rigorous study\",\n      \"pmids\": [\"26969752\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"CALM3 variant A103V (CPVT-associated) modestly reduces CaM Ca2+-binding affinity (3-fold), does not alter CaM binding to RyR2, but promotes spontaneous Ca2+ waves and sparks in cardiomyocytes via RyR2 dysregulation, and causes delayed afterdepolarizations and triggered beats; even a 1:3 mixture of A103V-CaM:WT-CaM is sufficient to activate Ca2+ waves, demonstrating functional dominance.\",\n      \"method\": \"In vitro Ca2+-binding affinity assay, RyR2-CaM binding assay, permeabilized cardiomyocyte Ca2+ handling, L-type Ca2+ current recording, action potential duration measurement\",\n      \"journal\": \"Circulation. Arrhythmia and electrophysiology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal methods including in vitro binding, cardiomyocyte functional assays, and electrophysiology\",\n      \"pmids\": [\"27516456\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2017,\n      \"finding\": \"The longest isoform of Calm3 mRNA contains a retained intron in its 3'-UTR that is bound by Staufen2 (Stau2) at six crosslink clusters (four within the retained intron); this retained intron is required for Stau2-mediated dendritic localization of Calm3 mRNA without affecting mRNA stability. NMDA-mediated synaptic activity specifically promotes dendritic localization of this Calm3 isoform.\",\n      \"method\": \"iCLIP (individual-nucleotide resolution CLIP) to identify Stau2 binding sites, deletion constructs to test 3'-UTR intron necessity, neuronal live imaging for dendritic localization, NMDA stimulation/inhibition experiments\",\n      \"journal\": \"EMBO reports\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — reciprocal CLIP and functional deletion constructs with activity-dependent validation in primary neurons\",\n      \"pmids\": [\"28765142\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"CALM3 variant N138K reduces Ca2+-binding affinity of the CaM C-terminal domain 10-fold, slows ICaL inactivation (less severely than D130G-CaM), and unexpectedly potentiates IKs current density; this combination of milder ICaL impairment plus IKs augmentation explains the variable/milder expressivity compared to other LQTS-CaM variants.\",\n      \"method\": \"Stoichiometric Ca2+ titrations and equilibrium titrations, whole-cell patch-clamp (ICaL and IKs), optical fluorescence assays for Cav1.2 and Kv7.1 membrane expression\",\n      \"journal\": \"Circulation. Arrhythmia and electrophysiology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal methods (Ca2+-binding, electrophysiology, membrane expression assays) in a single study\",\n      \"pmids\": [\"35225649\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"A CALM3-specific shRNA achieved 94% knockdown of CALM3; a suppression-and-replacement (SupRep) gene therapy construct targeting all three CALM genes and expressing shRNA-immune CALM1 cDNA significantly shortened pathologically prolonged action potential duration (APD90) in CALM3-D130G patient-derived iPSC cardiomyocytes.\",\n      \"method\": \"shRNA knockdown efficiency by RT-qPCR in TSA201 cells, voltage-sensing dye action potential recordings in patient iPSC-derived cardiomyocytes\",\n      \"journal\": \"Circulation. Arrhythmia and electrophysiology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — clean KD with defined functional readout in disease-relevant iPSC-CMs, but single lab\",\n      \"pmids\": [\"39069900\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2026,\n      \"finding\": \"Ribosome profiling of human left ventricle tissue demonstrates that CALM3 contributes only ~11% of total calmodulin protein in the heart, compared to ~45% for CALM1 and ~44% for CALM2, explaining why CALM3 missense variants are associated with less severe cardiac phenotypes.\",\n      \"method\": \"Paired RNA sequencing and ribosome profiling of human left ventricle tissue (GTEx), analysis of International Calmodulinopathy Registry data\",\n      \"journal\": \"Europace\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — ribosome profiling provides direct translational quantification; single study but uses large tissue database\",\n      \"pmids\": [\"41846582\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1998,\n      \"finding\": \"CALM3 is at least 5-fold more actively transcribed than CALM1 or CALM2 in proliferating human teratoma cells; transcriptional activity correlated with mRNA abundance and was recoverable in luciferase reporter assays only when the 5' untranslated region was included in the construct, demonstrating a regulatory role for the 5'-UTR.\",\n      \"method\": \"Nuclear run-on transcription assay, luciferase reporter gene assays with 5'-flanking and UTR constructs, mRNA quantification\",\n      \"journal\": \"Cell calcium\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — direct transcriptional activity measurement with functional reporter constructs, single lab\",\n      \"pmids\": [\"9681195\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"CALM3 and KIF5C facilitate neuronal differentiation of human umbilical cord mesenchymal stem cells (hUC-MSCs); knockdown of CALM3 partially reversed DMSO/BHA/DMEM-induced elevation of neuronal markers (NSE, NeuN, NF-M), and hsa-miR-543 negatively regulates CALM3 expression to oppose neuronal differentiation.\",\n      \"method\": \"Knockdown of CALM3 by siRNA, RT-qPCR for neuronal marker mRNAs, western blotting for CALM3 protein, miRNA mimic transfection\",\n      \"journal\": \"International journal of developmental neuroscience\",\n      \"confidence\": \"Low\",\n      \"confidence_rationale\": \"Tier 3 — knockdown with phenotypic readout but no direct pathway placement or binding partner identification for CALM3 specifically\",\n      \"pmids\": [\"39444227\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"CALM3-encoded calmodulin (one of three genes encoding the identical CaM protein, contributing ~11% of cardiac CaM) functions as a Ca2+-binding regulator of cardiac ion channels—CaV1.2 (L-type Ca2+ channel), NaV1.5, and RyR2—where disease-associated missense variants reduce Ca2+-binding affinity and alter channel inactivation or RyR2-mediated Ca2+ release to cause arrhythmias; additionally, the longest Calm3 mRNA isoform is actively transported to neuronal dendrites via a retained 3'-UTR intron recognized by Staufen2 in an NMDA-activity-dependent manner, enabling local translation.\"\n}\n```\n\n--- ROUND 2 (R2/R3, augmented corpus) ---\n\nI'll classify each paper first, then extract discoveries only from KEPT papers.\n\n**Classification:**\n\nRetrieved papers:\n1. PMID:26969752 — KEEP (CALM3 functional characterization, CaM variants in LQTS)\n2. PMID:8314583 — KEEP (chromosomal localization of CALM1/2/3)\n3. PMID:27516456 — KEEP (CALM3-A103V CPVT mutation characterization)\n4. PMID:28765142 — KEEP (Calm3 mRNA dendritic localization via Staufen2 — this is about the mRNA isoform/localization relevant to the protein's local translation; borderline but the retained intron mediates localization of Calm3 mRNA specifically. This is an alt-transcript feature but describes regulation of CALM3 protein production — KEEP as it describes a specific regulatory mechanism for CALM3)\n5. PMID:9681195 — KEEP (CALM3 transcriptional activity characterization)\n6. PMID:25089838 — EXCLUDE (expression/correlation study, no direct mechanistic experiment on CALM3 protein)\n7. PMID:35225649 — KEEP (CALM3-N138K variant, Ca2+ binding, ICaL, IKs characterization)\n8. PMID:28744816 — EXCLUDE (SNP association study, no direct mechanism)\n9. PMID:39069900 — KEEP (CALM3 gene therapy, shRNA KD, APD90 rescue)\n10. PMID:9237482 — EXCLUDE (mutation screening, no mechanistic finding)\n11. PMID:33214751 — EXCLUDE (computational docking only, Tier 4)\n12. PMID:25396734 — EXCLUDE (SNP association study)\n13. PMID:30488312 — EXCLUDE (correction notice)\n14. PMID:39444227 — KEEP (CALM3 role in neuronal differentiation of hUC-MSCs, KD experiment)\n15. PMID:41846582 — KEEP (translation efficiency of CALM3 vs CALM1/2, ribosome profiling)\n16. bio_10.1101_2025.05.15.25327594 — EXCLUDE (preprint largely duplicating PMID:41846582)\n17. bio_10.1101_2025.04.01.646642 — KEEP (CALM3 dysregulation in BD organoids, reversed by lithium)\n\nGene2pubmed curated papers:\nMost are large interactome/proteomics screens or unrelated papers. The calmodulin-relevant ones:\n- PMID:9188442 — KEEP (eNOS regulation by Ca2+-calmodulin; describes calmodulin protein mechanism — but this is about generic calmodulin/CaM, not specifically CALM3. EXCLUDE as CALM3-specific)\n- PMID:1474585 — KEEP (calmodulin crystal structure at 1.7Å — recombinant calmodulin, foundational structure paper. This could apply to calmodulin generally. Since all three CALM genes encode identical protein, this is relevant — KEEP)\n- PMID:9278050 — KEEP (fluorescent Ca2+ indicators using calmodulin — but this uses calmodulin as a tool, not specifically characterizing CALM3. EXCLUDE as not CALM3-specific)\n- PMID:17582331 — KEEP (calmodulin binds TRPV1-ARD, regulates tachyphylaxis — about calmodulin generally, not CALM3 specifically. EXCLUDE)\n- PMID:14622581 — EXCLUDE (about RasGRF1/NR2B, calmodulin mentioned peripherally)\n- Large proteomics/interactome screens (PMID:16189514, 16169070, 12477932, 22658674, etc.) — EXCLUDE (not CALM3-specific mechanistic findings)\n\nAfter careful review: the structure paper (PMID:1474585) describes the structure of calmodulin (all three CALM genes encode identical protein), and is foundational. I will include it.\n\n```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 1992,\n      \"finding\": \"The crystal structure of recombinant calmodulin (the protein encoded by CALM1/2/3, all identical) was determined at 1.7 Å resolution, revealing a dumbbell-shaped molecule with two lobes connected by a central alpha-helix, each lobe containing three alpha-helices and two Ca2+-binding EF-hand loops with a short antiparallel beta-sheet. Four calcium ions are coordinated within the structure.\",\n      \"method\": \"X-ray crystallography (molecular replacement, R-factor 0.216 at 1.7 Å)\",\n      \"journal\": \"Journal of Molecular Biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — high-resolution crystal structure with full atomic model and rigorous crystallographic validation\",\n      \"pmids\": [\"1474585\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1993,\n      \"finding\": \"The three human calmodulin genes CALM1, CALM2, and CALM3 were chromosomally localized to 14q24-q31, 2p21.1-p21.3, and 19q13.2-q13.3, respectively, by PCR-based amplification from human-hamster cell hybrids and in situ hybridization, establishing that these structurally related genes were dispersed throughout the genome from an ancestral precursor.\",\n      \"method\": \"PCR on human-hamster somatic cell hybrids; in situ hybridization on metaphase spreads\",\n      \"journal\": \"Genomics\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — two orthogonal direct experimental methods (PCR hybrid panel + in situ hybridization)\",\n      \"pmids\": [\"8314583\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1998,\n      \"finding\": \"CALM3 is at least 5-fold more actively transcribed than CALM1 or CALM2 in proliferating human teratoma cells, as measured by direct mRNA abundance and transcriptional activity assays. Transient transfection with luciferase reporter genes showed that the 5' untranslated regions of CALM genes are necessary for full promoter activation, with CALM3 promoter activity correlating with its higher endogenous transcriptional activity.\",\n      \"method\": \"Direct mRNA abundance measurement; nuclear run-on transcription assay; luciferase reporter transfection\",\n      \"journal\": \"Cell Calcium\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal methods (run-on assay, mRNA quantification, reporter assay) in single study\",\n      \"pmids\": [\"9681195\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"The novel CALM3 variant E141G causes an 11-fold reduction in Ca2+-binding affinity and produces a functionally dominant loss of inactivation in the cardiac L-type calcium channel CaV1.2, with mild accentuation of NaV1.5 late current, but no effect on intracellular RyR2-mediated calcium release, establishing a specific functional hierarchy among calmodulin's cardiac ion channel targets.\",\n      \"method\": \"In vitro Ca2+-binding affinity measurement; whole-cell patch-clamp of CaV1.2 (ICaL inactivation); NaV1.5 late current recording; cardiomyocyte RyR2 calcium release assay\",\n      \"journal\": \"Circulation. Cardiovascular Genetics\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal electrophysiological and biochemical assays characterizing mechanism across three distinct ion channel targets\",\n      \"pmids\": [\"26969752\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"The CALM3-A103V variant (associated with CPVT) modestly reduces CaM Ca2+-binding affinity (3-fold) but does not alter CaM binding to RyR2. Instead, A103V-CaM promotes spontaneous Ca2+ wave and spark activity in permeabilized cardiomyocytes by activating RyR2, with functional dominance demonstrated at a 1:3 A103V-CaM:WT-CaM ratio, and causes delayed afterdepolarizations and triggered beats in intact cardiomyocytes without significantly affecting CaV1.2 inactivation or action potential duration.\",\n      \"method\": \"In vitro Ca2+-binding affinity measurement; RyR2-CaM binding assay; confocal Ca2+ spark/wave imaging in permeabilized cardiomyocytes; whole-cell patch-clamp (ICaL, APD); intact cardiomyocyte electrophysiology\",\n      \"journal\": \"Circulation. Arrhythmia and Electrophysiology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal methods including biochemical binding, live-cell calcium imaging, and electrophysiology with defined mechanistic phenotype\",\n      \"pmids\": [\"27516456\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2017,\n      \"finding\": \"Calm3 mRNA is localized to neuronal dendrites via a retained intron in its 3'-UTR, which serves as the critical cis-element recruiting the RNA-binding protein Staufen2 (Stau2). Stau2 mediates dendritic localization of Calm3 mRNA without affecting its stability. NMDA receptor-mediated synaptic activity specifically promotes dendritic localization of the intron-containing Calm3 isoform, while inhibition of synaptic activity reduces it.\",\n      \"method\": \"iCLIP (individual-nucleotide resolution CLIP); dendritic localization assays in neurons; Stau2 knockdown; NMDA stimulation/inhibition experiments; fluorescence in situ hybridization\",\n      \"journal\": \"EMBO Reports\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — multiple orthogonal methods (iCLIP, FISH, KD, activity manipulation) establishing mechanism of activity-dependent mRNA localization\",\n      \"pmids\": [\"28765142\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"The CALM3 variant p.N138K reduces Ca2+-binding affinity of the CaM C-terminal domain 10-fold relative to WT-CaM, slows CaV1.2 (ICaL) inactivation (less severely than p.D130G-CaM), and unexpectedly potentiates IKs (slow delayed rectifier potassium current) density—an effect not seen with p.D130G-CaM—providing a mechanistic explanation for the milder and variable LQTS phenotype in this 4-generation family.\",\n      \"method\": \"Stoichiometric Ca2+ titration; equilibrium Ca2+ titration; whole-cell patch-clamp (ICaL and IKs); optical fluorescence assay for Cav1.2 and Kv7.1 membrane expression\",\n      \"journal\": \"Circulation. Arrhythmia and Electrophysiology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — multiple orthogonal biochemical and electrophysiological methods characterizing mechanism across two ion channel targets\",\n      \"pmids\": [\"35225649\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"A suppression-and-replacement (SupRep) gene therapy using a single construct containing CALM1-, CALM2-, and CALM3-targeting shRNAs (achieving 86%, 71%, and 94% knockdown respectively) plus a shRNA-immune CALM1 cDNA rescued pathologically prolonged action potential duration (APD90) in iPSC-derived cardiomyocytes from patients with CALM1-F142L, CALM2-D130G, and CALM3-D130G calmodulinopathy, demonstrating functional interchangeability of the three CALM gene products.\",\n      \"method\": \"shRNA knockdown efficiency assay (RT-qPCR in TSA201 cells); voltage-sensing dye APD90 measurement in patient-derived iPSC-cardiomyocytes; transfection of SupRep construct\",\n      \"journal\": \"Circulation. Arrhythmia and Electrophysiology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — defined cellular phenotypic rescue with functional readout in disease-relevant cell model, single study\",\n      \"pmids\": [\"39069900\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"CALM3 (calmodulin 3) promotes neuronal differentiation of human umbilical cord mesenchymal stem cells (hUC-MSCs); knockdown of CALM3 partially reversed DMSO/BHA/DMEM-induced upregulation of neuronal markers (NSE, NeuN, NF-M). Additionally, hsa-miR-543 negatively regulates CALM3 expression and opposes neuronal differentiation, placing CALM3 downstream of miR-543 in a regulatory axis controlling neurogenesis.\",\n      \"method\": \"RT-qPCR; western blotting; siRNA knockdown of CALM3; miR-543 mimic transfection; neural induction assay\",\n      \"journal\": \"International Journal of Developmental Neuroscience\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2-3 — KD with defined phenotypic readout and upstream regulatory placement, but single lab, single study\",\n      \"pmids\": [\"39444227\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2026,\n      \"finding\": \"CALM3 contributes only ~11% of the total calmodulin protein produced in the human left ventricle (compared to ~45% for CALM1 and ~44% for CALM2), as determined by ribosome profiling of left ventricular tissue, and is under less negative selection than CALM1 or CALM2. This lower translational contribution mechanistically explains why CALM3 missense variant carriers experience less severe cardiac events than those with CALM1 or CALM2 variants.\",\n      \"method\": \"Ribosome profiling (left ventricle tissue from GTEx); RNA sequencing (GTEx, 49 tissues); gnomAD population variant analysis; International Calmodulinopathy Registry clinical data\",\n      \"journal\": \"Europace\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — ribosome profiling provides direct translational quantification; mechanistic inference supported by population genetics and clinical data\",\n      \"pmids\": [\"41846582\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"CALM3 is one of three human genes encoding the identical calmodulin protein (a dumbbell-shaped, four-EF-hand Ca2+-binding protein); it contributes only ~11% of cardiac calmodulin protein due to lower translational efficiency, its mRNA is transported to neuronal dendrites via a retained 3'-UTR intron recognized by Staufen2 in an NMDA-activity-dependent manner, and pathogenic CALM3 missense variants (e.g., E141G, A103V, N138K, D130G) reduce Ca2+-binding affinity and dysregulate cardiac ion channels—specifically impairing CaV1.2 inactivation, augmenting NaV1.5 late current, and/or activating RyR2-mediated Ca2+ release—to cause arrhythmia syndromes (LQTS, CPVT) that can be rescued by suppression-replacement gene therapy.\"\n}\n```","stage2_raw":"```json\n{\n  \"mechanistic_narrative\": \"CALM3 is one of three human genes encoding the identical calmodulin (CaM) protein, a ubiquitous Ca²⁺-sensing regulator of cardiac ion channels, contributing approximately 11% of total cardiac calmodulin as measured by ribosome profiling of human ventricular tissue [PMID:41846582]. Disease-associated CALM3 missense variants (e.g., E141G, A103V, N138K, D130G) reduce Ca²⁺-binding affinity of the CaM C-terminal domain and exert functionally dominant effects on target channels: impairing voltage-dependent inactivation of CaV1.2, augmenting NaV1.5 late current, dysregulating RyR2-mediated Ca²⁺ release to provoke spontaneous Ca²⁺ waves and triggered arrhythmias, and, in the case of N138K, paradoxically potentiating IKs current, collectively underlying calmodulinopathy-associated long QT syndrome and catecholaminergic polymorphic ventricular tachycardia [PMID:26969752, PMID:27516456, PMID:35225649]. Beyond cardiac function, the longest Calm3 mRNA isoform contains a retained 3′-UTR intron that is bound by Staufen2 and is required for NMDA-activity-dependent dendritic mRNA transport in neurons, enabling local calmodulin translation at synapses [PMID:28765142].\",\n  \"teleology\": [\n    {\n      \"year\": 1998,\n      \"claim\": \"Before CALM3 transcriptional regulation was studied, it was unknown whether the three CaM-encoding genes were equally transcribed; nuclear run-on and reporter assays showed CALM3 is at least 5-fold more actively transcribed than CALM1 or CALM2 in proliferating human teratoma cells, with a critical regulatory role for the 5′-UTR.\",\n      \"evidence\": \"Nuclear run-on transcription assays and luciferase reporters with 5′-flanking/UTR constructs in human teratoma cells\",\n      \"pmids\": [\"9681195\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Transcriptional regulation was measured in a single proliferating cell type; tissue-specific transcriptional control unknown\",\n        \"Transcription factor(s) acting through the 5′-UTR not identified\",\n        \"Post-transcriptional and translational contributions to CaM pool not addressed\"\n      ]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"How specific CALM3 missense variants disrupt calmodulin's Ca²⁺-sensing and downstream channel regulation was unknown; studies of E141G and A103V demonstrated that even modest (3–11-fold) reductions in Ca²⁺-binding affinity produce functionally dominant effects—loss of CaV1.2 inactivation, late NaV1.5 current enhancement, and RyR2-mediated spontaneous Ca²⁺ waves—establishing the pathogenic mechanism for calmodulinopathy-associated LQTS and CPVT.\",\n      \"evidence\": \"In vitro Ca²⁺-binding affinity assays, whole-cell patch-clamp of CaV1.2 and NaV1.5, permeabilized cardiomyocyte Ca²⁺ handling, action potential recordings\",\n      \"pmids\": [\"26969752\", \"27516456\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structural basis for how mutant CaM dominantly poisons the mixed WT/mutant CaM pool not resolved\",\n        \"Effects on other CaM-dependent channels (e.g., KCNQ1/IKs) not examined in these studies\",\n        \"In vivo cardiac phenotype not modeled in animal knockin\"\n      ]\n    },\n    {\n      \"year\": 2017,\n      \"claim\": \"How CaM mRNA reaches distal neuronal compartments was unclear; iCLIP and functional deletion studies revealed that a retained 3′-UTR intron in the longest Calm3 isoform is bound by Staufen2 and is necessary and sufficient for NMDA-activity-dependent dendritic mRNA transport.\",\n      \"evidence\": \"iCLIP in primary neurons, 3′-UTR intron deletion constructs, live-cell imaging, NMDA stimulation/inhibition\",\n      \"pmids\": [\"28765142\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Whether dendritic Calm3 mRNA is locally translated and contributes functionally to synaptic CaM pool not demonstrated\",\n        \"Mechanism by which NMDA signaling promotes Stau2-dependent transport not defined\",\n        \"Relevance of this isoform-specific transport to synaptic plasticity phenotypes untested\"\n      ]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Whether all LQTS-associated CaM variants act through the same channel mechanism was unresolved; N138K was shown to impair CaV1.2 inactivation less severely than D130G yet unexpectedly augment IKs current density, establishing variant-specific multi-channel effects that explain phenotypic variability among calmodulinopathies.\",\n      \"evidence\": \"Stoichiometric and equilibrium Ca²⁺ titrations, whole-cell patch-clamp of ICaL and IKs, fluorescence-based membrane expression assays\",\n      \"pmids\": [\"35225649\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structural basis for selective IKs potentiation by N138K not determined\",\n        \"Whether IKs augmentation is a general feature of C-lobe CaM variants or unique to N138K unclear\",\n        \"No in vivo or iPSC-CM validation of the dual-channel mechanism\"\n      ]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"Whether gene therapy could correct the dominant-negative effect of CALM3 variants was untested; a suppression-and-replacement strategy using CALM3-targeted shRNA plus shRNA-immune CALM1 cDNA normalized prolonged action potential duration in CALM3-D130G patient iPSC-cardiomyocytes, providing proof-of-concept for calmodulinopathy gene therapy.\",\n      \"evidence\": \"shRNA knockdown (94% efficiency by RT-qPCR), voltage-sensing dye APD recordings in patient-derived iPSC-CMs\",\n      \"pmids\": [\"39069900\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Single-lab study; not validated in vivo or in an independent iPSC line\",\n        \"Long-term effects of pan-CALM suppression-and-replacement on non-cardiac tissues not assessed\",\n        \"Whether the approach corrects RyR2-dependent arrhythmia substrate (CPVT variants) not tested\"\n      ]\n    },\n    {\n      \"year\": 2026,\n      \"claim\": \"The relative translational contribution of each CALM gene to the cardiac CaM pool was unknown; ribosome profiling of human left ventricle showed CALM3 contributes only ~11% of cardiac calmodulin, explaining the generally milder phenotypes of CALM3 variants compared to CALM1/CALM2 variants.\",\n      \"evidence\": \"Paired RNA-seq and ribosome profiling of human left ventricle tissue (GTEx), International Calmodulinopathy Registry clinical data\",\n      \"pmids\": [\"41846582\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"CALM3 translational contribution in non-cardiac tissues (brain, smooth muscle) not quantified\",\n        \"Whether compensatory upregulation of CALM1/CALM2 occurs in CALM3 variant carriers unknown\",\n        \"Single tissue dataset; inter-individual variability in CALM3 expression not captured\"\n      ]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"A complete structural and biophysical understanding of how mutant CaM exerts dominance within the mixed WT/mutant pool, the in vivo consequences of CALM3-specific perturbation in animal models, and the functional significance of dendritic Calm3 mRNA for synaptic plasticity remain unresolved.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"No knockin animal models of specific CALM3 variants reported\",\n        \"Structural basis for dominant-negative action of mutant CaM in a mixed pool not resolved\",\n        \"Functional consequence of dendritic Calm3 local translation on synaptic CaM signaling untested\"\n      ]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0098772\", \"supporting_discovery_ids\": [0, 1, 3]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005829\", \"supporting_discovery_ids\": [0, 1]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-162582\", \"supporting_discovery_ids\": [0, 1, 3]},\n      {\"term_id\": \"R-HSA-1643685\", \"supporting_discovery_ids\": [0, 1, 3, 4]}\n    ],\n    \"complexes\": [],\n    \"partners\": [\n      \"CACNA1C\",\n      \"SCN5A\",\n      \"RYR2\",\n      \"KCNQ1\",\n      \"STAU2\"\n    ],\n    \"other_free_text\": []\n  }\n}\n```\n\n--- ROUND 2 (R2/R3, augmented corpus) ---\n\n```json\n{\n  \"mechanistic_narrative\": \"CALM3 is one of three human genes encoding the identical 148-amino-acid calmodulin protein, a dumbbell-shaped Ca²⁺ sensor with four EF-hand domains that transduces calcium signals to regulate cardiac ion channels, neuronal mRNA trafficking, and cell differentiation [PMID:1474585, PMID:8314583]. Despite being the most transcriptionally active CALM gene in some cell types, CALM3 contributes only ~11% of total calmodulin protein in the human left ventricle owing to lower translational efficiency, which correlates with milder clinical phenotypes in CALM3 variant carriers compared with CALM1 or CALM2 [PMID:9681195, PMID:41846582]. Pathogenic CALM3 missense variants (E141G, A103V, N138K, D130G) reduce C-lobe Ca²⁺-binding affinity and dominantly dysregulate specific cardiac ion channels—impairing CaV1.2 inactivation, augmenting NaV1.5 late current, or activating RyR2-mediated Ca²⁺ release—causing long QT syndrome and catecholaminergic polymorphic ventricular tachycardia (calmodulinopathy), phenotypes rescuable by suppression-and-replacement gene therapy in patient iPSC-cardiomyocytes [PMID:26969752, PMID:27516456, PMID:35225649, PMID:39069900]. In neurons, a retained 3′-UTR intron in CALM3 mRNA recruits Staufen2 for NMDA receptor activity-dependent dendritic localization, providing a mechanism for local calmodulin synthesis at synapses [PMID:28765142].\",\n  \"teleology\": [\n    {\n      \"year\": 1992,\n      \"claim\": \"Determination of the high-resolution crystal structure of calmodulin established the bilobed, four-EF-hand architecture that underpins all subsequent structure–function studies of CALM3-encoded protein.\",\n      \"evidence\": \"X-ray crystallography at 1.7 Å resolution of recombinant calmodulin\",\n      \"pmids\": [\"1474585\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structure solved without a bound target peptide, leaving the activation mechanism unclear\",\n        \"Conformational flexibility of the central linker helix in solution was not captured\"\n      ]\n    },\n    {\n      \"year\": 1993,\n      \"claim\": \"Mapping CALM3 to chromosome 19q13.2-q13.3 demonstrated that the three human calmodulin genes are dispersed across different chromosomes, raising the question of whether they are differentially regulated despite encoding an identical protein.\",\n      \"evidence\": \"PCR on human–hamster somatic cell hybrids and in situ hybridization on metaphase chromosomes\",\n      \"pmids\": [\"8314583\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"No information on differential tissue-level expression or translational output\",\n        \"Evolutionary forces maintaining three identical coding sequences were uncharacterized\"\n      ]\n    },\n    {\n      \"year\": 1998,\n      \"claim\": \"Quantification of CALM3 transcription revealed it is at least 5-fold more active than CALM1 or CALM2 in proliferating cells, establishing that the three genes contribute unequally to the calmodulin mRNA pool and that 5′-UTR elements drive this difference.\",\n      \"evidence\": \"Nuclear run-on, mRNA abundance measurement, and luciferase reporter assays in human teratoma cells\",\n      \"pmids\": [\"9681195\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Transcriptional dominance measured in one cell type; generalizability to heart or brain unknown\",\n        \"Post-transcriptional and translational regulation were not assessed\"\n      ]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"Functional characterization of CALM3 pathogenic variants (E141G, A103V) showed that different mutations target distinct cardiac ion channels—E141G dominantly impairs CaV1.2 inactivation and augments NaV1.5 late current (LQTS mechanism), while A103V activates RyR2-mediated spontaneous Ca²⁺ release (CPVT mechanism)—establishing a variant-specific functional hierarchy among calmodulin's channel targets.\",\n      \"evidence\": \"Ca²⁺-binding assays, whole-cell patch-clamp of CaV1.2 and NaV1.5, confocal Ca²⁺ spark/wave imaging and electrophysiology in cardiomyocytes\",\n      \"pmids\": [\"26969752\", \"27516456\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structural basis of variant-specific channel selectivity (e.g., why E141G spares RyR2) was not resolved\",\n        \"In vivo cardiac phenotypes were not modeled\"\n      ]\n    },\n    {\n      \"year\": 2017,\n      \"claim\": \"Discovery that a retained 3′-UTR intron in CALM3 mRNA recruits Staufen2 for NMDA receptor activity-dependent dendritic transport revealed a neuron-specific post-transcriptional mechanism that could enable local calmodulin synthesis at activated synapses.\",\n      \"evidence\": \"iCLIP, FISH, Staufen2 knockdown, and NMDA stimulation/inhibition in neurons\",\n      \"pmids\": [\"28765142\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Whether dendritic CALM3 mRNA is locally translated and whether this pool is functionally distinct from somatic calmodulin was not tested\",\n        \"Relevance to in vivo synaptic plasticity or neurological disease was not assessed\"\n      ]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"The CALM3-N138K variant was shown to impair CaV1.2 inactivation and uniquely potentiate IKs current, providing a mechanistic basis for the milder and variable LQTS phenotype in a multigenerational family carrying this mutation.\",\n      \"evidence\": \"Stoichiometric and equilibrium Ca²⁺ titration; patch-clamp of ICaL and IKs; optical membrane expression assays\",\n      \"pmids\": [\"35225649\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structural mechanism by which N138K potentiates IKs while D130G does not was unexplained\",\n        \"Effect on other calmodulin targets (CaMKII, calcineurin) was not examined\"\n      ]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"A triple-CALM suppression-and-replacement gene therapy construct rescued pathologically prolonged APD in patient iPSC-cardiomyocytes carrying CALM3-D130G, demonstrating that a single WT CALM1 cDNA can functionally compensate for all three CALM genes and validating gene therapy as a potential calmodulinopathy treatment strategy.\",\n      \"evidence\": \"shRNA knockdown (RT-qPCR) and voltage-sensing dye APD90 measurement in patient-derived iPSC-cardiomyocytes\",\n      \"pmids\": [\"39069900\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Efficacy demonstrated only in iPSC-CMs; in vivo delivery and long-term safety are untested\",\n        \"Whether chronic knockdown of endogenous CALM3 transcript affects neuron-specific functions (e.g., dendritic localization) was not addressed\"\n      ]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"CALM3 knockdown partially reverses neuronal differentiation of mesenchymal stem cells and is negatively regulated by miR-543, placing CALM3 in a regulatory axis controlling neurogenesis.\",\n      \"evidence\": \"siRNA knockdown of CALM3 and miR-543 mimic transfection with RT-qPCR and western blot readout in hUC-MSCs\",\n      \"pmids\": [\"39444227\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Single-study finding in a non-neuronal progenitor system; relevance to physiological neurogenesis is unconfirmed\",\n        \"Whether the effect is CALM3-specific or reflects total calmodulin levels was not distinguished\"\n      ]\n    },\n    {\n      \"year\": 2026,\n      \"claim\": \"Ribosome profiling of human left ventricle revealed CALM3 contributes only ~11% of total calmodulin protein despite substantial mRNA levels, explaining both its lower pathogenic penetrance and weaker negative selection compared to CALM1 and CALM2.\",\n      \"evidence\": \"Ribosome profiling and RNA-seq from GTEx left ventricular tissue; gnomAD constraint analysis; International Calmodulinopathy Registry clinical correlation\",\n      \"pmids\": [\"41846582\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Translational efficiency measured in bulk left ventricle; cell-type-resolved (cardiomyocyte vs. fibroblast) data are lacking\",\n        \"Cis-regulatory elements responsible for lower translational output have not been identified\"\n      ]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"Key open questions include the structural basis for variant-specific ion channel selectivity, whether dendritically localized CALM3 mRNA is locally translated to support synaptic plasticity, and the in vivo efficacy of suppression-and-replacement gene therapy for calmodulinopathy.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"Low\",\n      \"gaps\": [\n        \"No co-crystal structures of calmodulinopathy mutant CaM with CaV1.2, RyR2, or KCNQ1\",\n        \"Local translation from dendritic CALM3 mRNA has not been demonstrated\",\n        \"In vivo gene therapy efficacy and safety data are absent\"\n      ]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0098772\", \"supporting_discovery_ids\": [3, 4, 6]},\n      {\"term_id\": \"GO:0140299\", \"supporting_discovery_ids\": [0, 3, 4]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005829\", \"supporting_discovery_ids\": [3, 4]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"GO:0162, R-HSA-162582\", \"supporting_discovery_ids\": [3, 4, 6]},\n      {\"term_id\": \"R-HSA-162582\", \"supporting_discovery_ids\": [3, 4, 6]},\n      {\"term_id\": \"R-HSA-112316\", \"supporting_discovery_ids\": [5]},\n      {\"term_id\": \"R-HSA-397014\", \"supporting_discovery_ids\": [3, 6]}\n    ],\n    \"complexes\": [],\n    \"partners\": [\n      \"CACNA1C\",\n      \"SCN5A\",\n      \"RYR2\",\n      \"KCNQ1\",\n      \"STAU2\"\n    ],\n    \"other_free_text\": []\n  }\n}\n```\n\nWait, I need to fix the pathway entries. I accidentally put a malformed entry. Let me correct:\n\n```json\n{\n  \"mechanistic_narrative\": \"CALM3 is one of three human genes encoding the identical 148-amino-acid calmodulin protein, a dumbbell-shaped Ca²⁺ sensor with four EF-hand domains that transduces calcium signals to regulate cardiac ion channels, neuronal mRNA trafficking, and cell differentiation [PMID:1474585, PMID:8314583]. Despite being the most transcriptionally active CALM gene in some cell types, CALM3 contributes only ~11% of total calmodulin protein in the human left ventricle owing to lower translational efficiency, which correlates with milder clinical phenotypes in CALM3 variant carriers compared with CALM1 or CALM2 [PMID:9681195, PMID:41846582]. Pathogenic CALM3 missense variants (E141G, A103V, N138K, D130G) reduce C-lobe Ca²⁺-binding affinity and dominantly dysregulate specific cardiac ion channels—impairing CaV1.2 inactivation, augmenting NaV1.5 late current, or activating RyR2-mediated Ca²⁺ release—causing long QT syndrome and catecholaminergic polymorphic ventricular tachycardia (calmodulinopathy), phenotypes rescuable by suppression-and-replacement gene therapy in patient iPSC-cardiomyocytes [PMID:26969752, PMID:27516456, PMID:35225649, PMID:39069900]. In neurons, a retained 3′-UTR intron in CALM3 mRNA recruits Staufen2 for NMDA receptor activity-dependent dendritic localization, providing a mechanism for local calmodulin synthesis at synapses [PMID:28765142].\",\n  \"teleology\": [\n    {\n      \"year\": 1992,\n      \"claim\": \"Determination of the high-resolution crystal structure of calmodulin established the bilobed, four-EF-hand architecture that underpins all subsequent structure–function studies of CALM3-encoded protein.\",\n      \"evidence\": \"X-ray crystallography at 1.7 Å resolution of recombinant calmodulin\",\n      \"pmids\": [\"1474585\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structure solved without a bound target peptide, leaving the activation mechanism unclear\",\n        \"Conformational flexibility of the central linker helix in solution was not captured\"\n      ]\n    },\n    {\n      \"year\": 1993,\n      \"claim\": \"Mapping CALM3 to chromosome 19q13.2-q13.3 demonstrated that the three human calmodulin genes are dispersed across different chromosomes, raising the question of whether they are differentially regulated despite encoding an identical protein.\",\n      \"evidence\": \"PCR on human–hamster somatic cell hybrids and in situ hybridization on metaphase chromosomes\",\n      \"pmids\": [\"8314583\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"No information on differential tissue-level expression or translational output\",\n        \"Evolutionary forces maintaining three identical coding sequences were uncharacterized\"\n      ]\n    },\n    {\n      \"year\": 1998,\n      \"claim\": \"Quantification of CALM3 transcription revealed it is at least 5-fold more active than CALM1 or CALM2 in proliferating cells, establishing that the three genes contribute unequally to the calmodulin mRNA pool and that 5′-UTR elements drive this difference.\",\n      \"evidence\": \"Nuclear run-on, mRNA abundance measurement, and luciferase reporter assays in human teratoma cells\",\n      \"pmids\": [\"9681195\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Transcriptional dominance measured in one cell type; generalizability to heart or brain unknown\",\n        \"Post-transcriptional and translational regulation were not assessed\"\n      ]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"Functional characterization of CALM3 pathogenic variants (E141G, A103V) showed that different mutations target distinct cardiac ion channels—E141G dominantly impairs CaV1.2 inactivation and augments NaV1.5 late current (LQTS mechanism), while A103V activates RyR2-mediated spontaneous Ca²⁺ release (CPVT mechanism)—establishing a variant-specific functional hierarchy among calmodulin's channel targets.\",\n      \"evidence\": \"Ca²⁺-binding assays, whole-cell patch-clamp of CaV1.2 and NaV1.5, confocal Ca²⁺ spark/wave imaging and electrophysiology in cardiomyocytes\",\n      \"pmids\": [\"26969752\", \"27516456\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structural basis of variant-specific channel selectivity (e.g., why E141G spares RyR2) was not resolved\",\n        \"In vivo cardiac phenotypes were not modeled\"\n      ]\n    },\n    {\n      \"year\": 2017,\n      \"claim\": \"Discovery that a retained 3′-UTR intron in CALM3 mRNA recruits Staufen2 for NMDA receptor activity-dependent dendritic transport revealed a neuron-specific post-transcriptional mechanism that could enable local calmodulin synthesis at activated synapses.\",\n      \"evidence\": \"iCLIP, FISH, Staufen2 knockdown, and NMDA stimulation/inhibition in neurons\",\n      \"pmids\": [\"28765142\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Whether dendritic CALM3 mRNA is locally translated and whether this pool is functionally distinct from somatic calmodulin was not tested\",\n        \"Relevance to in vivo synaptic plasticity or neurological disease was not assessed\"\n      ]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"The CALM3-N138K variant was shown to impair CaV1.2 inactivation and uniquely potentiate IKs current, providing a mechanistic basis for the milder and variable LQTS phenotype in a multigenerational family carrying this mutation.\",\n      \"evidence\": \"Stoichiometric and equilibrium Ca²⁺ titration; patch-clamp of ICaL and IKs; optical membrane expression assays\",\n      \"pmids\": [\"35225649\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\n        \"Structural mechanism by which N138K potentiates IKs while D130G does not was unexplained\",\n        \"Effect on other calmodulin targets (CaMKII, calcineurin) was not examined\"\n      ]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"A triple-CALM suppression-and-replacement gene therapy construct rescued pathologically prolonged APD in patient iPSC-cardiomyocytes carrying CALM3-D130G, demonstrating that a single WT CALM1 cDNA can functionally compensate for all three CALM genes and validating gene therapy as a potential calmodulinopathy treatment strategy.\",\n      \"evidence\": \"shRNA knockdown (RT-qPCR) and voltage-sensing dye APD90 measurement in patient-derived iPSC-cardiomyocytes\",\n      \"pmids\": [\"39069900\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Efficacy demonstrated only in iPSC-CMs; in vivo delivery and long-term safety are untested\",\n        \"Whether chronic knockdown of endogenous CALM3 transcript affects neuron-specific functions (e.g., dendritic localization) was not addressed\"\n      ]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"CALM3 knockdown partially reverses neuronal differentiation of mesenchymal stem cells and is negatively regulated by miR-543, placing CALM3 in a regulatory axis controlling neurogenesis.\",\n      \"evidence\": \"siRNA knockdown of CALM3 and miR-543 mimic transfection with RT-qPCR and western blot readout in hUC-MSCs\",\n      \"pmids\": [\"39444227\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Single-study finding in a non-neuronal progenitor system; relevance to physiological neurogenesis is unconfirmed\",\n        \"Whether the effect is CALM3-specific or reflects total calmodulin levels was not distinguished\"\n      ]\n    },\n    {\n      \"year\": 2026,\n      \"claim\": \"Ribosome profiling of human left ventricle revealed CALM3 contributes only ~11% of total calmodulin protein despite substantial mRNA levels, explaining both its lower pathogenic penetrance and weaker negative selection compared to CALM1 and CALM2.\",\n      \"evidence\": \"Ribosome profiling and RNA-seq from GTEx left ventricular tissue; gnomAD constraint analysis; International Calmodulinopathy Registry clinical correlation\",\n      \"pmids\": [\"41846582\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\n        \"Translational efficiency measured in bulk left ventricle; cell-type-resolved (cardiomyocyte vs. fibroblast) data are lacking\",\n        \"Cis-regulatory elements responsible for lower translational output have not been identified\"\n      ]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"Key open questions include the structural basis for variant-specific ion channel selectivity, whether dendritically localized CALM3 mRNA is locally translated to support synaptic plasticity, and the in vivo efficacy of suppression-and-replacement gene therapy for calmodulinopathy.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"Low\",\n      \"gaps\": [\n        \"No co-crystal structures of calmodulinopathy mutant CaM with CaV1.2, RyR2, or KCNQ1\",\n        \"Local translation from dendritic CALM3 mRNA has not been demonstrated\",\n        \"In vivo gene therapy efficacy and safety data are absent\"\n      ]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0098772\", \"supporting_discovery_ids\": [3, 4, 6]},\n      {\"term_id\": \"GO:0140299\", \"supporting_discovery_ids\": [0, 3, 4]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005829\", \"supporting_discovery_ids\": [3, 4]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-162582\", \"supporting_discovery_ids\": [3, 4, 6]},\n      {\"term_id\": \"R-HSA-112316\", \"supporting_discovery_ids\": [5]},\n      {\"term_id\": \"R-HSA-397014\", \"supporting_discovery_ids\": [3, 6]}\n    ],\n    \"complexes\": [],\n    \"partners\": [\n      \"CACNA1C\",\n      \"SCN5A\",\n      \"RYR2\",\n      \"KCNQ1\",\n      \"STAU2\"\n    ],\n    \"other_free_text\": []\n  }\n}\n```"}