Functional Validation of a Novel Homozygous TTN Splice‐Site Variant Reveals Aberrant Splicing in Hypertrophic Cardiomyopathy Authors: Xiaoyun Sun, Dandan Wang, Shujuan Wang, Rong Wan, Zhenhong Jiang, Wenqiang Zhang, Jin Ouyang, Yang Shen, Kui Hong Published: 2026 ABSTRACT The TTN gene encodes a crucial structural protein within cardiac sarcomeres, and its variants may contribute to hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy; however, phenotype and genotype are different. Whole‐exome sequencing (WES) was conducted on a Chinese proband diagnosed with HCM. In silico splicing prediction tools and minigene assays were employed to investigate the impact of the identified variant on mRNA splicing. A literature review was performed to retrieve and analyze previously reported splicing variants in the TTN gene associated with HCM. A 42‐year‐old male proband presented with nonobstructive HCM and paroxysmal atrial arrhythmias. A novel homozygous TTN variant was found, predicted to cause a 14‐base pair deletion at a splice acceptor site. Two asymptomatic offspring were found to carry the heterozygous variant. Based on variant interpretation guidelines, the variant met the PM2_supporting criterion and was classified as a variant of uncertain significance (VUS). The predicted aberrant splicing effect was subsequently confirmed by the minigene splicing assay, demonstrating altered pre‐mRNA splicing leading to an in‐frame insertion/deletion (p.Arg32498_Glu32504delinsGln). Functional verification confirmed that this mutation conforms to the PM4 criterion, suggesting that it can be reclassified as a tepid VUS (scoring 3 points). The functional result might provide pathogenic evidence. We also reviewed 28 previously reported splicing variants in TTN associated with HCM. Of these, 57.1% (16/28) localized to the I‐band region, whereas 21.4% (6/28) were situated in the A‐band domain of titin. Notably, 21.4% (6/28) co‐occurred with pathogenic variants in other sarcomeric genes (MYH7 or MYBPC3), correlating with more severe clinical phenotypes. Reclassification and reinterpretation of the variants revealed that none met the level of likely pathogenic or higher. The present case contributes a homozygous splice‐site variant with experimentally confirmed aberrant splicing and an in‐frame protein alteration in titin. The focus of this report is the Mendelian genetic basis of the proband′s cardiomyopathy phenotype, and our data provide additional case‐level and functional evidence for a possible role of specific TTN splicing defects in HCM. 1. Introduction Hypertrophic cardiomyopathy (HCM) is the most prevalent inherited cardiomyopathy, characterized by left ventricular hypertrophy that is not solely explained by hemodynamic load. Current estimates indicate a prevalence of approximately 1 in 200 individuals [1]. HCM is typically inherited in an autosomal dominant (AD) pattern with incomplete penetrance. Approximately 60% of patients with HCM harbor pathogenic variants, primarily in genes encoding myocardial sarcomere proteins [2, 3]. Variants in MYH7 and MYBPC3 represent the most frequent genetic etiologies, accounting for over 50% of cases, followed by TNNT2 and TNNI3 [4–7]. The Clinical Genome Resource (ClinGen) categorizes HCM‐associated genes based on the strength of supporting evidence. Genes such as MYH7, MYBPC3, TNNT2, TNNI3, TMP1, MYL2, MYL3, and ACTC1 are classified as having “definitive” evidence, whereas TTN is currently categorized as having only “limited” evidence [8]. Titin is a large, multidomain protein crucial for sarcomere structural integrity and passive stiffness, modulating active contractile force generation [9, 10]. It consists of repeating immunoglobulin (Ig) and Fibronectin type III (FN‐III) modules interspersed with nonrepetitive sequences that contain phosphorylation sites, PEVK motifs, and a terminal kinase. Alternative splicing of TTN pre‐mRNA generates several cardiac isoforms, including N2B, N2BA, and Novex‐3 [11]. TTN is the most common definitive pathogenic gene in dilated cardiomyopathy (DCM), accounting for 25% of cases [12, 13]. Importantly, TTN has also been identified as a candidate gene for HCM, left ventricular noncompaction cardiomyopathy (LVNC), and arrhythmogenic right ventricular cardiomyopathy (ARVC) [14–16]. TTN‐truncating variants (TTNtv), especially those enriched in the A‐band region, are known pathogenic drivers of DCM. These typically include frameshift, nonsense, and essential splice‐site variants [13, 17]. In contrast, the association between TTN and HCM is supported by limited and inconclusive evidence. In this study, we report a patient with nonobstructive HCM and atrial fibrillation (AF)/flutter carrying a novel homozygous TTN splicing variant and investigate its effect on pre‐mRNA splicing using a minigene assay. Our findings aim to enhance a more comprehensive understanding of the role of this specific TTN variant, thereby informing the ongoing evaluation of TTN′s contribution to the genetic basis of HCM. 2. Materials and Methods 2.1. Clinical Evaluation and Ethical Compliance A 42‐year‐old male patient presented with recurring chest tightness and palpitations, subsequently receiving medical care at The Second Affiliated Hospital of Nanchang University. Various clinical diagnostic examinations were conducted, including electrocardiogram (ECG), transthoracic echocardiography (TTE), computed tomography (CT) and laboratory tests. The patient was prescribed metoprolol. The Ethics Committee of the Second Affiliated Hospital of Nanchang University approved this study, and we obtained the patient′s written informed consent. 2.2. Whole Exome Sequencing (WES) Genomic DNA was extracted from a peripheral blood sample of the patient. All steps of WES were performed by the Department of Medical Genetics of The Second Affiliated Hospital of Nanchang University in Jiangxi, China. The process included genomic DNA extraction, exome library preparation, exome capture, and sequencing on the Nextseq 550DX platform (Illumina, United States) following alignment to the human reference genome (GRCh37/hg19), variants such as single nucleotide variants (SNVs), insertions and/or deletions (indels), and others were detected using Dynamic Read Analysis for GENomics software, with annotation performed using relevant databases. Finally, variant validation and cosegregation analysis were confirmed by Sanger sequencing in the core family. The sequences of the primer were 5 ′‐GCACATGTACCCCTGAAGCT‐3 ′ and 5 ′‐ATCCATGGCAACGATGGGTT‐3 ′. The 3500DX XL platform (Life, United States) was employed for this purpose. 2.3. Bioinformatics Analysis WES was performed for the proband. Variant analysis was initially focused on a virtual gene set of cardiomyopathies‐ and inherited arrhythmia‐associated genes with ClinGen‐curated or ClinGen‐based gene‐disease validity classifications of limited or higher. The genes included in this analysis, together with their associated phenotypes and evidence classifications, are listed in Table S2. Multiple online in silico splice‐site predictions were used to evaluate the potential pathogenicity of the splicing variants, including RDDCSC (https://rddc.tsinghua-gd.org/), VarSEAK (https://varseak.bio/index.php), and SpliceAI (https://spliceailookup.broadinstitute.org/). These predictions suggested potential aberrant splicing caused by the identified variant (seen in the Figure S1). 2.4. The Minigene Splicing Assay The minigene assay had been extensively validated in previous studies and was considered an effective, reliable, and relatively simple tool for evaluating potential RNA splicing alterations. We used the pcMINI vector that we developed previously, which contained a multiple cloning site (Figure 1A). We identified the genomic region surrounding the c.97493‐2_97504del variant and designed primers to amplify this region from the genomic DNA of controls using DNA Polymerase‐PrimeSTAR MAX DNA Polymerase (R045A, TaKaRa, Japan). Subsequently, the PCR products were cloned into pcMINI vectors using EcoRI/XhoI digestion and a recombination reaction. Mutant constructs containing the c.97493‐2_97504del variant were generated by PCR using the mutant primers (Method S1 and Table S1). The constructed vectors were then transformed into Escherichia coli DH5α‐competent cells and selected for PCR amplification. Following direct confirmation of wild type/mutant (wt/mut) plasmids by Sanger sequencing, both types of plasmids were transfected into HEK293T and HeLa cells. After 48 h, total RNA was extracted using the RNA extraction kit (Trizol RNAiso PLUS, 9109, TaKaRa, Japan) and reverse transcribed with Hifair 1st Strand cDNA Synthesis SuperMix for qPCR (gDNA digester plus) (11123ES70, YEASEN, China). The resulting cDNA was then PCR‐amplified. Finally, the amplified products were analyzed through 1.5% agarose gel electrophoresis, purified with the DNA Gel Extraction Kit (2001250, SIMGEN, China), and subsequently sequenced by Sanger sequencing. The effect of TTN c.97493‐2_97504del variant on splicing. (A) Schematic diagram of vector construction used in minigene assay. (B) The Sanger sequencing of the minigene plasmid shows the successful site‐directed mutagenesis of the variant, wt, wild type; mut, mutant. (C) Agarose gel electrophoresis for RT‐PCR analysis, bands in HeLa and 293T cells are labeled a and b, respectively, M indicates the ladder. It shows a qualitative RT‐PCR analysis of minigene‐derived transcripts. Band intensity was not used for quantitative comparison of transcript abundance. (D) Schematic diagram of the minigene plasmid showed 18‐bp deletion in Exon 350. (E) Sanger sequencing of the transcript of the minigene plasmid showed 18‐bp deletion in Exon 350. 3. Result 3.1. Clinical and Follow‐Up Results The 42‐year‐old male patient presented with recurrent chest tightness and palpitations. The patient exhibited clear consciousness and clear breath sounds in both lungs. Heart rate was elevated at 168 beats per minute with arrhythmia noted, characterized by unequal intensity of the first heart sound, and the intensity of the second heart sound in the aortic valve area was greater than in the pulmonary valve area. No other abnormal physical signs were observed. The proband did not report skeletal muscle weakness, myalgia, recurrent falls, respiratory muscle involvement, or other neuromuscular symptoms, and no obvious skeletal muscle abnormalities were documented during routine physical examination. The proband had no severe valvular disease, thyroid dysfunction, ischemic heart disease, diabetes mellitus, hypertension, or obesity. However, he had a history of smoking, which may represent a potential risk factor for AF/flutter [18]. ECG revealed paroxysmal atrial flutter (Figure 2A). TTE confirmed a diagnosis of nonobstructive HCM with preserved left ventricular ejection fraction (58%) (Figure 2C). Chest CT showed mild left atrial enlargement, left ventricular hypertrophy, and pericardial effusion (Figure 2B). Twenty‐four–hour ECG monitoring detected episodes of paroxysmal AF and supraventricular tachycardia. Limb peripheral nerve electromyography showed no characteristic abnormalities of peripheral nerves or skeletal muscles, and muscle MRI revealed no abnormal findings. HCM and AF/flutter were recognized at the same clinical presentation during this hospitalization. After further evaluation, the patient was diagnosed with HCM and paroxysmal AF. Clinical evaluation of the patient′s family members, including his parents, wife, 18‐year‐old son, and 14‐year‐old daughter, revealed they were asymptomatic. Based on the available family history, no known consanguinity was reported in the proband′s parents or wider family. The clinical analysis of the proband. (A) The 12‐lead electrocardiogram (ECG) shows 2:1 atrial flutter. (B) Left atrium CT suggests mild enlargement of the left atrium, myocardial hypertrophy of the left ventricle, and pericardial effusion. (C) The echocardiographic image shows symmetrical nonobstructive left ventricular (LV) hypertrophy. The red arrow indicates the interventricular septum with 15–22 mm, and the blue arrow indicates the left ventricular free wall with 19 mm. To manage the condition, the patient was treated with metoprolol for ventricular remodeling and rivaroxaban as anticoagulant therapy. The follow‐up results showed a persistently hypertrophic interventricular septum (IVS) and a reduction in the thickness of the left ventricular posterior wall (LVPW) and the diameter of the left atrial diameter (LAD) (Table 1). Table 1Follow‐up TTE results of the proband.TimeIVS (mm) (8–11)LVPW (mm) (8–11)LVEF (%) (50–70)LAD (mm) (< 35)2022/0315–221958532022/0914–231755462023/0712–241856432024/1212–23145743Abbreviations: IVS, interventricular septum thickness; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; LVPW, left ventricular posterior wall. Follow‐up TTE results of the proband. Abbreviations: IVS, interventricular septum thickness; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; LVPW, left ventricular posterior wall. 3.2. Identification of a Homozygous TTN Splice‐Site Variant WES identified a novel homozygous 14‐base pair deletion in the TTN gene (chr2: g.179406300_179406313del; NM_001267550.2: c.97493‐2_97504del) in the proband (Figure 3A), which was subsequently confirmed by Sanger sequencing (Figure 3B). The variant involves the canonical splice acceptor region of TTN meta‐transcript Exon 350, corresponding to LRG Exon 351 according to CardioDB/LRG exon annotation. This exon is located within the A‐band region of titin and encodes the FN‐III 125 domain (Figure 3C). The affected exon is highly included in cardiac TTN transcripts, with percentage spliced‐in (PSI) values of 100% in DCM left ventricular tissue and 98% in GTEx left ventricular tissue. No additional rare pathogenic or likely pathogenic (LP) variants in established HCM‐associated genes or major cardiomyopathy/arrhythmia‐related genes were identified after variant filtering; the remaining variants were classified as benign (B) or likely benign (LB) or were considered clinically unrelated. Cascade genetic screening via Sanger sequencing showed the proband′s wife was wild‐type at this locus, whereas his asymptomatic 18‐year‐old son and 14‐year‐old daughter were heterozygous for the variant (Figure 3D). The family pedigree is illustrated in Figure 3E. The genetic analysis of proband and cascade family screening. (A) Integrated genome view of the TTN variant (chr2: g.179406300_179406313del) with IGV software. (B) The Sanger sequencing data of the variant in the proband. (C) Schematic representation of the fibronectin type‐III 125 domain affected by the splice variant in titin. (D) The Sanger sequencing data of the variant in the proband′s wife and children. (E) The pedigree of the family. The arrow indicates the proband; the black square indicates the male patient; +/+ indicates the homozygous variant; +/− indicates the heterozygous variant carrier; and −/− indicates the wild type. 3.3. Effect of the TTN c.97493‐2_97504del Variant on Splicing Prediction tools suggested that the c.97493‐2_97504del variant disrupts the splice acceptor site. Specifically, SpliceAI (Δscore 0.99) predicted activation of a cryptic acceptor site. However, bioinformatic predictions alone were insufficient to establish the variant′s impact on RNA processing, in line with American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) guidelines. To assess its functional consequence, we constructed a minigene vector containing _TTN_ Exon 349, Intron 349, and Exon 350, harboring the c.97493‐2_97504del variant (Figure 1A,B). Transfection of this construct into HEK‐293T and HeLa cell lines, followed by RT‐PCR analysis, revealed no exon skipping (Figure 1C). Sequencing of the cDNA revealed that the variant led to the utilization of a cryptic splice‐site located 18 bp downstream of the authentic acceptor site (Figure 1D,E). This aberrant splicing event resulted in the deletion of amino acids Arg32498‐Glu32504 and the insertion of a glutamine residue (c.97493_97510del; p.Arg32498_Glu32504delinsGln). Based on the ACMG/AMP variant interpretation criteria [19], the variant was classified as a variant of uncertain significance (VUS), supported by PM2_supporting and PM4 evidence codes. 4. Literature Review of TTN Splicing Variants in HCM A comprehensive review of PubMed, ClinVar, and the Human Gene Mutation Database (HGMD) identified 28 reported TTN splicing variants associated with HCM (Table 2). These variants were re‐evaluated according to ACMG/AMP guidelines [19]. Of the analyzed variants, 57.1% (16/28) localized in the I‐band region, whereas 21.4% (6/28) were situated in the A‐band domain of titin. Notably, 21.4% (6/28) of the variants co‐occurred with pathogenic variants in sarcomeric genes (MYH7 or MYBPC3), correlating with severe clinical phenotypes, such as unexplained syncope, sudden cardiac death, and nonsustained ventricular tachycardia. Of the 28 variants, 46.4% (13/28) were classified as VUS, underscoring the need for functional studies to clarify their role as disease modifiers or contributors to compound heterozygosity. In addition, 53.6% (15/28) variants were classified as LB or B. Most existing evidence is derived from familial case reports using next‐generation sequencing, which systematically excluded variants with high population frequency, localization to nonconstitutive exons, or co‐occurrence with other known pathogenic variants. However, few of these variants have been functionally validated using minigene assays or RNA sequencing. Table 2Summary of TTN splicing variants associated with HCM.IDTranscriptcDNATypeIntronCardiac PSI%gnomAD (FAF)Affected domainSpliceAI predictionVariants classificationPhenotypesCo‐occurred variantsReference1NM_001267550.2c.1398+4C>TCryptic splicing81000.00023450Z‐disk△score < 0.20LBHCM—[20]2NM_001267550.2c.22816+1G>TCanonical splicing7835NAIg‐like 54 (I‐band)Donor loss:△score 0.98VUSHCM—[21]3NM_001267550.2c.31763‐1G>ACanonical splicing122780.00055920—Acceptor loss:△score 1; acceptor gain:△score 0.88VUSHCM—[13]4NM_001267550.2c.34612+1G>TCanonical splicing15020NA—Donor loss:△score 0.99VUSHCM MYBPC3:p.R495G[22]5NM_001267550.2c.35308+1G>TCanonical splicing1565NAPEVK 5 (I‐band)Donor loss:△score 0.98VUSHCM—[23]6NM_001267550.2c.55432+5G>CCryptic splicing2861000.00004868Ig‐like 101 (A‐band)△score < 0.20VUSHCM MYBPC3: p. E474X[15]7NM_001267550.2c.59344+3G>ACryptic splicing3001000.00174000FN‐III 31 (I‐band)△score < 0.20LBHCM, syncope, SCD MYBPC3: c.1928‐2A>G[15]8NM_001267550.2c.98098+3G>ACryptic splicing3511000.00111300FN‐III 126 (A‐band)△score < 0.20LBHCM, syncope, NSVT MYBPC3:p.I807N[15] TNNT2: p. I100M9NM_001267550.2c.106374+5C>ACryptic splicing358100NAIg‐like 149 (M‐line)△score < 0.20VUSHCM, LVOTO MYH7: p. F46I[15]10NM_001256850.1c.28091‐2A>CCanonical splicing99690.00127500—Acceptor loss:△score 1LBHCM—[20]11NM_001256850.1c.34186+1G>TCanonical splicing15248NAPEVK 15 (I‐band)Donor loss:△score 0.98VUSHCM—[13]12NM_133378.4c.10114+5G>ACryptic splicing431000.00923200Ig‐like 20 (I‐band)△score < 0.20BHCM—[23]13NM_133378.4c.25310‐2A>CCanonical splicing97720.00127500Ig‐like 77 (I‐band)Acceptor loss:△score 1VUSHCM MYH7: p.R723C. PKP4: p.T319M[24]14NM_133378.4c.28031‐1G>ACanonical splicing118780.00055920PEVK 3 (I‐band)Acceptor loss:△score 1VUSHCM—[24]15NM_133378.4c.30094+4C>TCryptic splicing140740.00001154PEVK 12 (I‐band)△score < 0.20VUSHCM—[25]16NM_133378.4c.32077+20C>GCryptic splicing15950.00000437I‐band△score < 0.20LBHCM—[26]17NM_133378.4c.32407+6C>TCryptic splicing16370.00795300PEVK 23 (I‐band)△score < 0.20LBHCM, LQTS, SUD—[23]18NM_133378.4c.43733‐2A>TCanonical splicing221100NAI‐bandAcceptor loss:△score 1VUSHCM—[26]19NM_133378.4c.49408‐4C>TCryptic splicing2411000.00249700FN‐III 26 (I‐band)△score < 0.20LBHCM—[23, 25]20NM_133378.4c.50143+6 T>ACryptic splicing2441000.00558200FN‐III 27 (I‐band)△score < 0.20LBHCM—[23]21NM_133378.4c.59066‐4A>TCryptic splicing266100NANonconstitutiveAcceptor loss:△score 0.09LBHCM—[26]22NM_133378.4c.62012‐5C>GCryptic splicing2741000.00373400FN‐III 57 (A‐band)△score < 0.20LBSUD, SUD/SIDS, HCM—[23]23NM_133378.4c.80003‐4G>TCryptic splicing2771000.00274100FN‐III 101 (A‐band)△score < 0.20LBHCM—[23]24NM_133378.4c.89200+4 T>CCryptic splicing2961000.00023870A‐band△score < 0.20VUSHCM—[25]25NM_133378.4c.98670+4A>GCryptic splicing307100NAIg‐like 149 (A‐band)△score < 0.20LBHCM—[26]26NM_133437c.9977‐4G>ACryptic splicing421000.00008225Ig‐like 209 (I‐band)△score < 0.20LBHCM DSP:c.1_2insA[24]27NM_133437c.43097‐5C>GCryptic splicing1541000.00373400FN‐III 57 (I‐band)△score < 0.20LBHCM—[24]28NM_133437c.70574‐3T>CCryptic splicing1771000.00018980FN‐III 124 (I‐band)△score < 0.20VUSHCM SCN5A: p.D1819N[24]Abbreviations: B, benign; FAF, filtering allele frequency; FN‐III, fibronectin type‐III; HCM, hypertrophic cardiomyopathy; Ig‐like, immunoglobulin (Ig)‐like domain; LB, likely benign; LQTS, long QT syndrome; LVOTO, left ventricular outflow tract obstruction; NA, not available; NSVT, nonsustained ventricular tachycardia; PEVK, the proline‐, glutamate‐, valine‐, and lysine‐rich; PSI, percentage spliced in; SIDS, sudden infant death syndrome; SUD, sudden unexplained death; VUS, variant of uncertain significance. Summary of TTN splicing variants associated with HCM. Abbreviations: B, benign; FAF, filtering allele frequency; FN‐III, fibronectin type‐III; HCM, hypertrophic cardiomyopathy; Ig‐like, immunoglobulin (Ig)‐like domain; LB, likely benign; LQTS, long QT syndrome; LVOTO, left ventricular outflow tract obstruction; NA, not available; NSVT, nonsustained ventricular tachycardia; PEVK, the proline‐, glutamate‐, valine‐, and lysine‐rich; PSI, percentage spliced in; SIDS, sudden infant death syndrome; SUD, sudden unexplained death; VUS, variant of uncertain significance. 5. Discussion To our knowledge, this study described the first Chinese pedigree with HCM associated with the homozygous TTN variant. Functional analysis via the minigene splicing assay revealed that cryptic splice‐site activation induced the in‐frame indels (p.Arg32498_Glu32504delinsGln). TTN variants have been reported in various cardiomyopathies including HCM, LVNC, and ARVC [13–16]. Although TTN has definitive gene‐disease validity for DCM, its role in HCM remains controversial due to limited evidence. Since the initial report linking TTN and HCM, TTNtv have been detected in 1.3%–4% of HCM cases [13, 15, 24, 27]. Some variants in TTN were reported in sporadic individuals with HCM. For instance, a heterozygous missense variant (rs201381085) was identified in a patient with nonobstructive HCM [28]. In a cohort of 242 HCM patients, a TTNtv (c.70437delG) located in the A‐band was detected in one patient, classified as LP, and categorized as a non–HCM‐associated variant [29]. Additionally, a pathogenic splice‐site variant (c.22816+1G>T) near the I‐band terminus was documented in a Romanian HCM patient [21]. However, case‐control studies have failed to provide compelling evidence for a strong association between TTN and HCM. This is because TTN variants are detected in both patient and control groups, with no significant enrichment of TTN variants in the case cohorts [13, 15]. These conflicting results challenge the association between the TTN gene and HCM and suggest that this relationship needs further investigations. TTNtv represent a prominent genetic alteration in cardiomyopathies [13, 15]. Although canonical splice‐site variants are often grouped with truncating variants when they are predicted to cause frameshift or nonsense‐mediated decay, the variant reported here activates a cryptic acceptor site and produces an in‐frame indel. Several variants were reclassified as LB or B due to high population frequencies, thereby excluding their potential pathogenicity. The remaining variants demonstrated insufficient evidence for classification as LP or higher. Most identified variants localized to the I‐band region, affecting exons within low PSI transcripts in myocardial tissue [12]. Furthermore, several splicing variants situated within the A‐band region were predicted to induce in‐frame exon skipping, thereby preserving the reading frame, and not resulting in a frameshift. Compared with many reported TTN splicing variants in HCM, which either affect low‐PSI exons, lack RNA validation, or co‐occur with pathogenic variants in established sarcomeric genes, the present variant affects a highly included A band exons and produces a minigene‐confirmed in‐frame indel. The minigene result predicts preservation of the reading frame and does not support a classical truncating mechanism in this assay. Although the aberrant transcript is predicted to preserve the reading frame, we did not assess endogenous TTN transcript abundance, titin protein expression, protein stability, sarcomere incorporation, or binding to interacting partners. Therefore, we cannot exclude altered protein stability or functional impairment of the affected titin domain. Specific in‐frame deletions affecting functional amino acid motifs in A‐band domains may exhibit HCM association potential, though this hypothesis requires validation through expanded cohort analyses and mechanistic functional studies. The inheritance pattern of HCM is primarily AD. However, variants in TTN and other genes typically associated with dominant inheritance can also manifest through recessive mechanisms. In the present study, familial analysis revealed autosomal recessive (AR) inheritance. Although rare, homozygous variants in sarcomeric genes are associated with severe cardiac phenotypes. For instance, a homozygous MYBPC3 in‐frame deletion (chr11:g.47357432_47357458del) has been reported to result in combined HCM/DCM phenotypes [30], and a pathogenic MYH7 in‐frame deletion (c.2791_2793delGAG) has been linked to infantile‐onset HCM [31]. A novel homozygous TPM1 variant (p.Gly3Arg) has been identified as a cause of AR pediatric HCM [32]. Rare TTN variants have previously been reported in sporadic HCM cases [21, 28, 29]. These reports provide context for considering TTN variants in HCM. In the present family, the proband was homozygous for a splice‐altering TTN variant predicted to preserve the reading frame. The heterozygous offspring were clinically asymptomatic at their current ages. This observation is compatible with a possible dosage‐dependent or recessive model. We hypothesize that a subset of rare HCM cases may result from homozygous variants, which are transmitted through AR inheritance within families. Therefore, our findings raise the possibility that homozygosity for TTN variants predicted to retain partial protein function may contribute to HCM through AR inheritance. Such variants could include missense variants, small in‐frame deletions, or splice‐altering variants that preserve the reading frame. The residual functional capacity preserved in such variants might account for their recessive inheritance pattern while still maintaining sufficient pathogenicity to manifest clinical phenotypes under homozygous conditions. Furthermore, previous studies utilizing mouse models carrying TTN variants have demonstrated HCM phenotypes. Specifically, adult homozygous PEVK knockout mice exhibited diastolic dysfunction and cardiac hypertrophy accompanied by upregulation of FHL proteins [33]. Similarly, titin M‐line–deficient mice developed cardiac hypertrophy and heart failure [34]. Although the relationship between TTN and HCM remains classified as limited in the recent ClinGen, the evidence score was close to the threshold for moderate, with the final classification remaining limited mainly because of insufficient case‐level evidence [8]. Several functional studies nevertheless support the biological plausibility of TTN involvement in HCM. For example, variant p.Arg740Leu in TTN was demonstrated to increase the binding affinity of titin to alpha‐actinin in a yeast two‐hybrid assay [27]. Investigating novel pathogenic pathways, a study identified a missense variant in an Ig domain near the M‐line–A‐band transition zone of TTN in a medaka mutagenesis fish model exhibiting diastolic dysfunction [35]. Additionally, the authors of this study identified two missense variants in titin′s Ig domains in patients with HCM and provided functional evidence that these variants enhance binding to muscle‐specific ring finger protein 1 (MURF1), leading to increased titin degradation through ubiquitination [35]. Current variant detection methodologies likely underestimate exonic and noncanonical splicing alterations, as comprehensive RNA analyses remain technically challenging in clinical practice. Minigene assays offer a practical method for in vitro splicing analysis. Functional verification confirmed that this variant conforms to the PM4 standard, suggesting that it be reclassified as a tepid VUS (scoring 3 points). Consistent with current ESC consensus on cardiovascular genomics, warm/hot VUS carriers require management analogous to LP variants [36]. However, because the present variant remains a tepid VUS after functional verification, clinical management in this family should be driven primarily by the proband′s phenotype, the inheritance model, and continued familial reassessment. TTN variants have been reported in patients with early‐onset AF. However, the strongest evidence in AF relates mainly to heterozygous truncating or loss‐of‐function variants located in highly expressed cardiac exons. In the UK Biobank study, high‐PSI TTN loss‐of‐function variants were associated with AF, but AF was present in only 14% of carriers, and polygenic risk substantially modified penetrance [37]. Similarly, studies of early‐onset AF have shown enrichment of pathogenic/LP variants in cardiomyopathy genes, with TTN being common among positive findings, but most such variants were interpreted in the broader context of inherited cardiomyopathy rather than isolated AF [38]. In cohorts of TTNtv carriers, AF is frequently observed together with, or in the setting of progression toward, left ventricular dysfunction; although AF may occasionally precede overt LV systolic dysfunction, this does not exclude latent or evolving cardiomyopathy [39]. A recent study of AF patients with pathogenic TTN variants also focused on high‐PSI truncating variants and found that nearly half‐developed LV systolic dysfunction and/or malignant ventricular arrhythmias during follow‐up [40]. Although the literature supports TTNtv as important contributors to AF susceptibility, particularly in early‐onset AF and AF associated with cardiomyopathy, these data do not directly establish that a nontruncating homozygous splice‐site variant such as the one reported here is primarily an AF‐only Mendelian cause. Furthermore, major gene‐disease curation resources, including ClinGen, GenCC, Gene2Phenotype, and Morbid, have not systematically curated a Mendelian association between TTN and isolated AF. This absence of formal curation further supports a cautious interpretation: Rare TTN variants are credible contributors to AF risk, especially when embedded within a cardiomyopathy phenotype, but they do not yet fulfill the evidentiary criteria for a standalone Mendelian AF gene. In this context, the present case may contribute additional clinical and functional evidence to this evolving gene–disease relationship. The homozygous splice‐site variant identified here affects a highly cardiac‐included TTN exon and was shown by minigene assay to generate an in‐frame deletion/insertion, suggesting a potential mechanism by which disruption of specific cardiac TTN exons, particularly in a biallelic state, may contribute to HCM or an HCM‐arrhythmia overlap phenotype. We interpreted the variant as a candidate contributor to the patient′s cardiomyopathy‐centered phenotype, with AF/flutter potentially representing part of a broader cardiomyopathy‐arrhythmia presentation rather than definitive evidence for an isolated AF mechanism. 5.1. Limitations TTN is characterized by highly complex and tissue‐specific alternative splicing, and minigene assays cannot fully recapitulate the endogenous regulatory context of TTN pre‐mRNA processing in human cardiomyocytes. Therefore, patient‐derived RNA analysis would, in principle, provide more direct evidence for the endogenous splicing consequence of this variant. However, the feasibility and interpretability of blood‐based RNA analysis for TTN are limited by both tissue expression and exon‐level splicing context. TTN is expressed predominantly in striated muscle, and its transcript abundance in peripheral blood is very low. In the available blood transcriptomic data, the expression level of the relevant TTN region was low (TPM = 0.2919), suggesting that peripheral blood RNA would have limited sensitivity for detecting aberrant splicing at this locus. In addition, because TTN has an exceptionally large transcript size and complex isoform structure, short‐read RNA sequencing from a low‐expression tissue may provide uneven coverage and may not reliably resolve the relevant full‐length splicing event. Thus, even if blood RNA were available, a negative result would not necessarily exclude an endogenous splicing defect in cardiac tissue. Unfortunately, we were unable to obtain an additional blood sample from the patient for RNA analysis. DNA samples from the proband′s parents and siblings were not available. Parental carrier status and broader familial segregation could not be assessed. Future research aimed at fully characterizing the role of TTN variants in HCM and developing novel diagnostics and therapeutics will probably require integrating multiomics functional analyses with studies of larger cohorts. 6. Conclusion To our knowledge, this study is the first to describe a Chinese pedigree with HCM linked to a homozygous TTN variant. The minigene assay demonstrated that this deletion caused aberrant pre‐mRNA splicing, resulting in the complex in‐frame indels (c.97493_97510del; p.Arg32498_Glu32504delinsGln) within the FN III domain. This alteration may be related to the less severe HCM phenotype. While describing a single variant cannot fully elucidate the correlation between genotype and phenotype, the results of this study provide new insights into the molecular genetic mechanisms underlying HCM. Author Contributions Xiaoyun Sun and Dandan Wang carried out all the data analyses, participated in the design of the work, and wrote the original draft. Xiaoyun Sun and Rong Wan designed the experiments. Shujuan Wang, Zhenhong Jiang, and Jin Ouyang participated in molecular genetic test. Wenqiang Zhang and Jin Ouyang participated in data analyses. Yang Shen and Kui Hong reviewed and revised manuscripts and provided financial support. Xiaoyun Sun and Dandan Wang contributed equally. Funding This study was supported by the National Natural Science Foundation of China (No. 82360048), Jiangxi Province Natural Science Foundation (20223BBG71011), and the Molecular Medicine Jiangxi Provincial Key Laboratory Support Program (2024SSY06231). Disclosure All authors read and approved the final manuscript. Ethics Statement This study was approved by the Medical Ethics Committee of The Second Affiliated Hospital of Nanchang University. The informed consent from the patient′s parents prior to conducting the genetic testing had been obtained. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting Information Additional supporting information can be found online in the Supporting Information section. Method S1 (minigene assay): To create hybrid minigene constructs, we used the pcMINI vector that we developed previously. We cloned DNA fragments containing a couple of exons and introns around the target variant in the TTN gene using classical restriction enzyme digestion and ligation methods. We described the detailed steps for constructing the mini gene carrying the c.97493‐2_97504del variant. Figure S1: The results of in silico splice‐site predictions. (A) SpliceAI predicts that this variant the disruption of the original acceptor site, and the creation of a new acceptor site may affect splicing (△score 0.99); (B) RDDCSC predicts that the disruption of the original acceptor site, and the creation of a new acceptor site may affect splicing; (C) VarSEAK predicts that the original acceptor site is disrupted, creating a new acceptor site, and that the confidence score is elevated by 74.93%, affecting the results of splicing. Table S1: Primer sequences in the minigene constructs carrying the c.97493‐2_97504del variant. Table S2: Cardiomyopathy‐ and inherited arrhythmia‐associated genes prioritized from WES.