Homozygous TNNC1-A8V Causes Impaired Diastolic Function and Restrictive Cardiomyopathy in Children and Mice

Abstract

Background:

Pathologic variants in TNNC1 are associated with the development of hypertrophic cardiomyopathy (HCM). Unlike other sarcomeric genes, TNNC1 variants tend to cause early onset disease, and the most common variant identified in the gene, TNNC1-A8V, may also lead to severe diastolic dysfunction and restrictive cardiomyopathy (RCM). Yet, this variant-disease association, and the mechanism of diastolic disease development, remains poorly understood. To address this, we sought to determine whether the A8V variant was associated with development of RCM, and if so, to determine the frequency and age-specific spectrum of disease relative to HCM.

Methods:

A pediatric proband was identified carrying the TNNC1-A8V variant. Sanger sequencing and cardiac evaluations were performed on several of the proband’s family members. A literature search was performed to evaluate the cardiac penetrance of TNNC1-A8V. Additionally, we investigated cardiac contractility in a homozygous A8V mouse model. We assessed left ventricular (LV) function in 10–12-week-old mice via cardiac catheterization and LV global and segmental myocardial strain analyses. Electrophysiological studies were conducted to evaluate arrhythmic predisposition.

Results:

We report an 8-month-old female homozygous for p.A8V (c.C23T) who presented with bi-atrial enlargement, normal ventricular function, and ventricular tachycardia. Cardiac histology of the proband showed subendocardial vacuolization, endocardial fibrosis, and localized myocardial enhancement. The A8V mouse model showed elevated end-diastolic pressure-volume relationship (EDPVR), prolonged relaxation time (tau), and impaired ventricular relaxation rate (dP/dt min) compared with controls, consistent with increased myocardial stiffness. Strain analyses further highlighted LV dysfunction and dyssynchrony, and ECG analyses revealed atrial remodeling and enlargement at 10-weeks of age.

Conclusions:

Homozygous A8V mice recapitulate the severe diastolic dysfunction and early-onset RCM observed in the pediatric proband carrying this pathogenic variant, underscoring the utility of this model for advancing translational research. Identifying myofilament-specific defects opens the door for precision therapies tailored to the underlying molecular disarray.

Introduction

Inherited cardiomyopathies are primarily caused by mutations in sarcomeric genes that disrupt the structural and functional integrity of the myocardium^1,2. The underlying etiology is often more heterogeneous during childhood, making diagnosis challenging due to its diverse genetic basis and association with multiple myopathies^3–5. Among these diseases, hypertrophic cardiomyopathy (HCM) and restrictive cardiomyopathy (RCM) share significant phenotypic overlap and a similar genetic basis, both primarily driven by variants in sarcomeric genes^6–9. HCM is the most prevalent cardiomyopathy, affecting approximately 1:500 individuals, and remains the leading cause of sudden cardiac death (SCD) in the young^1–3. HCM is characterized by asymmetric myocardial hypertrophy, diastolic dysfunction, and a hypercontractile phenotype^4,5. In contrast, RCM is a rare and likely underdiagnosed condition associated with a poor prognosis, accounting for approximately 4.5% of pediatric cases; among these, 66% are classified as pure RCM, while 34% display a mixed HCM/RCM phenotype².

Patients with a pure RCM diagnosis typically have the poorest outcomes, with a 5-year survival rate from diagnosis^2,6. RCM is marked by impaired ventricular filling due to increased myocardial stiffness, despite preserved systolic function⁷. Diagnosis typically includes atrial enlargement relative to normal or small ventricles, with clear evidence of impaired diastolic filling. Mounting evidence suggests that RCM may exist along a spectrum of familial cardiomyopathies, particularly in cases with overlapping genetic, histological, and clinical features of HCM⁸. Notably, the presence of pure RCM does not appear to confer a higher risk of death compared with the mixed RCM/HCM phenotype².

While pathogenic variants in thick filament proteins such as β-myosin heavy chain (MYH7) and myosin-binding protein C (MYBPC3) are more commonly associated with cardiomyopathies, mutations in thin filament proteins, though rarer, are linked to significant morbidity and mortality^9,10. These include variants in the cardiac troponin complex (Tn) subunits TnT, TnI, and TnC (TNNT2, TNNI3, TNNC1, respectively), α-tropomyosin (TPM1), and cardiac actin-1 (ACTC1)^11–15. For example, thin filament mutations in HCM are often associated with mild left ventricular (LV) hypertrophy with an elevated risk of arrhythmias, a distinct clinical phenotype^6,10. Thin filament mutations in RCM, on the other hand, are associated with more severe phenotypes and increased mortality, necessitating early intervention (e.g., cardiac transplant)⁶.

Among sarcomeric genes implicated in pediatric cardiomyopathy, TNNC1 has emerged as a critical determinant of disease severity¹⁶. TNNC1-endoded cTnC functions as a sarcomeric Ca²⁺ sensor of the thin filament, regulating crossbridge formation and thus cardiac contraction. Patients diagnosed with a mutation in TNNC1 present with worse clinical outcomes, including earlier disease onset and more aggressive disease progression compared with other genetic cardiomyopathies^17–19. The expressed phenotype can evolve over time, and disease penetrance tends to increase with age, reflecting the complex and heterogeneous pathophysiology of TNNC1-related cardiomyopathy¹⁹. These clinical findings have contributed to consensus from the American College of Medical Genetics and Genomics for screening of TNNC1 variants when evaluating patients for heritable cardiomyopathic risk^20,21. Additionally, a comprehensive evaluation of gene-disease relationships conducted by the Clinical Genome Resource (ClinGen) Hereditary Cardiovascular Disease (HCVD) gene curation expert panel reclassified TNNC1 as a definitive gene with strong evidence for HCM.

Among TNNC1 variants, c.C23T encoding p.A8V is causative for both RCM and HCM^17,18, likely due to its location in the N-helix, which is known to modulate Ca²⁺ binding affinity of both the N- and C-terminal domains of cTnC^22–24. Its association to RCM was illustrated in a homozygous cTnC-A8V variant identified in an child with severe diastolic dysfunction¹⁷. To further understand this genotype-phenotype correlation, we leveraged a cTnC-A8V homozygous proband with a Tnnc1 mouse model bearing A8V²⁵. In our investigation, we aimed to characterize for the first time the hemodynamic status with the gold standard of pressure-volume (PV) loops, identify signs of early diastolic dysfunction using strain analysis as a link to early onset presentation of RCM in A8V mice, and to correlate these findings to a multi-generational family study.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request.

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Clinica Imbanaco and Duke University Health System Institutional Review Boards. Informed consent was obtained from all participants or their legal guardians, as appropriate. All protocols and experimental procedures using animal subjects followed NIH guidelines and were approved by the University of Miami, Florida State University, and Duke University Institutional Animal Care and Use Committee (ACUC).

Further details on specific methods are described in individual sections below.

Clinical Evaluation

The proband underwent cardiac evaluation consisting of 12-lead electrocardiogram, Holter ECG recording, two-dimensional Doppler echocardiogram, cardiovascular magnetic resonance (CMR) imaging, cardiac catheterization, and diagnostic genetic testing.

Histology of Explanted Heart

Myocardium from the proband’s explanted heart was formalin-fixed (FF) in 4% paraformaldehyde and paraffin-embedded (PE). FFPE-blocks were then subjected to H&E staining for further pathological analyses.

Sanger Sequencing

Sanger sequencing was performed to confirm the TNNC1 A8V mutation from samples collected from the proband, and 12 other family members of the proband. Genomic DNA (gDNA) was extracted from all family members blood samples using the ReliaPrep Blood gDNA system (Cat#A5082), and polymerase chain reaction (PCR) was performed with the purified gDNA. The following primers were utilized, (i) Forward: AAACCTGGTCCTTGTAGCCG; (ii) Reverse: GCTACTAACCCCGCACTCTC; (iii) along with 2.0X Taq Red Master Mix kit (42–138, Genesee). PCR products were confirmed via gel electrophoresis (2% agarose gel). PCR products were then excised from gels and underwent Sanger sequencing.

Cardiomyopathy Variant Penetrance Analysis

A comprehensive literature review was conducted to evaluate the penetrance of variant TNNC1-A8V, incorporating the proband’s family history (TNNC1 Family) into the final analysis. Only published cases reporting patients with the TNNC1-A8V variant were included in the analysis. Literature and TNNC1 family cases were excluded if they did not meet the following criteria: genetic test verification of the variant, cardiac evaluation, age information, and if they had another pathogenic/likely pathogenic (P/LP) variant in another cardiac disease-related gene.

Animal Studies

The homozygous A8V mouse model has been previously characterized in our laboratory^25,26. Both male and female wildtype (WT) controls and A8V mice were used at 10–12 weeks of age.

Murine Echocardiography

Transthoracic echocardiography was performed using the Vevo F2 Visualsonic imaging system (FUJIFILM VisualSonics, Toronto, ON, Canada) with a UHF57x linear transducer (57–25 MHz). The Vevo F2 imager was utilized to obtain parasternal short-axis (PSAX)- and long (PLAX) images at the level of the papillary muscles (sweep speed of 200 mm/s). Pulsed-wave doppler images of mitral inflow were acquired. Mice were lightly sedated via nose cone anesthesia (1.5–2% vaporized isoflurane delivered in 100% O₂). Short-axis, M-mode recordings were taken to obtain percent left ventricular (LV) ejection fraction (LVEF) and fractional shortening (LVFS), internal diameters at end-diastole (LVIDd) and end systole (LVIDs), and LV relative wall thickness (RWT), according to the American Society of Echocardiography guidelines for animals²⁷. All recordings were taken in triplicate and averaged per mouse. High-frequency 2D speckle-tracking echocardiography (STE) was assessed from PLAX images with 200 frames per second and analyzed offline using Vevo Strain 2.0 software (VisualSonics).

Murine Electrocardiography (ECG)

8-lead surface ECGs were performed using the iWorx BIO-8 small animal ECG System (Cat. No. TCS-100; iWorx Systems Inc., Dover, NH, USA). Mice were lightly anesthetized via nose cone anesthesia (1.5 – 2% isoflurane vaporized in 100% O₂) and electrodes were placed on all four paws while the mouse was in a supine position to obtain Lead I & II ECG recordings. ECGs were recorded for 5–10 minutes and the iWorx ECG Analysis Add-on Software was utilized to construct signal-averaged ECGs (SAECGs) to determine wave durations/intervals and amplitudes.

Pressure-Volume (PV) Loops

Hemodynamic assessments were conducted using a micro-tipped pressure-volume (PV) catheter (MPVR-1035; ADInstruments), as previously described [24]. Briefly, mice were anesthetized with isoflurane (2%−3%) and core body temperature was maintained at 37°C). The catheter was inserted into the right carotid artery and advanced retrograde into the LV. PV loops were recorded at steady state and during temporary inferior vena cava occlusion. The volumes were calibrated using volumes derived from echocardiographic measurements. All analyses were performed using LabChart Pro version 8.1.5 software (AD Instruments).

Intracardiac electrophysiology studies

Mice were anesthetized using a volatile anesthetic system with induction chamber (R5835, RWD Life Science, Dover, Delaware) with 2% isoflurane mixed with 1.5 L/min 100% O₂. Subdermal leads were placed for surface electrocardiographic (ECG) analysis which includes two total leads: I and II. Rhythm detection was captured by an iWorx-RA-834 Eight Channel 16-bit Data Acquisition System (iWorx, Dover New Hampshire, United States). Data were recorded at 1000 samples/second and viewed using a custom-built ECG Analysis Module software program for LabScribe v4. Once mice reached an internal temperature of 37C, RR, PR, QRS, QT, and corrected QT (Bazett’s QT correction) were measured. Jugular venous cutdown was performed using a dissection microscope (A60, Leica, Buffalo Grove, Illinois) and an 8-lead 1.1F, 8E, 1.0 mm octapolar catheter electrophysiology catheter (iWire-BIO8, ADInstruments, Colorado Springs, Colorado) was placed in the internal jugular vein and positioned in the right ventricle and right atrium (Au - Li and Au - Wehrens 2010 – PMID: 20512106). Electrical pacing was performed utilizing customized electrical stimulator to elicit ectopy and arrhythmia generation with progressive electrical challenge (Clasen, Eickholt et al. 2018). To induce ventricular arrhythmias, 30-pulse pacing episodes with pacing interval of 75 ms were applied followed by 10 pulses with pacing interval stepping from 55 ms to 15 ms with 2 ms decrement with a 10–15 sec inter-episode break. Non-sustained ventricular tachycardia (NSVT) was defined as 3–10 ventricular ectopic beats. Sustained VT was defined as greater than 10 ventricular ectopic beats. Arrhythmic burden was defined as the sum of the following events post-stimulation training: (i) NSVT, (ii) VT, (iii) premature ventricular complex (PVC) and PVC couplets. 1 μg/g isoproterenol was administered followed by a repeated the programmed electrical stimulation protocol as described above. Finally, 10 μg/g isoproterenol was administered. followed by programmed electrical stimulation.

Histological Analysis

Mouse hearts were harvested and FFPE. Blocks were cut at 5–6 μm (2–3 cuts per slide), deparaffinized and stained with Masson’s Trichrome (Sigma; Cat. No. HT15–1KT) following manufacturer’s protocol. The slides were imaged by light microscopy using a BZ-X800 Keyence Microscope (Leica Microsystems, Deerfield, IL, USA). Myocardial sections were traced using ImageJ to achieve total area. Then blue fibrotic sections were traced to record total fibrosis. Percent myocardial fibrosis was determined by the sum of all fibrotic areas (within one slice) divided by total myocardial area using ImageJ version 1.53e software. Each slice was then averaged to obtain percent myocardial fibrosis per mouse.

Statistical Analysis

Data are presented as medians [interquartile range, IQR] for the animal studies; sample sizes (n-values) and specific statistical tests performed are indicated within each figure legend. Due to the sample sizes, a normal (Gaussian) distribution was not assumed, and non-parametric statistical analyses were applied. Outliers were identified using Grubbs’ test with an alpha level of 0.05 and removed where appropriate. Statistical significance for echocardiography, ECG, and pressure-volume (PV) studies was assessed using the non-parametric Mann–Whitney U test. Significance levels are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001, with p < 0.05 considered statistically significant. All statistical analyses and data visualizations were performed using GraphPad Prism software (version 11.0.0).

Results

Clinical presentation of an individual diagnosed with RCM

An 8-month-old female with a history of recurrent respiratory infections was incidentally noted to have cardiomegaly on a chest X-ray. Subsequent echocardiograms revealed a ventricular septal defect (VSD), which was clinically managed with diuretics and angiotensin-converting enzyme inhibitors. At 19 months of age, the proband presented with dyspnea, ascites, and bradycardia. ECG showed sinus bradycardia with left atrial (LA) enlargement, reduced PR-interval, LV enlargement with abnormal repolarization, and marked ST depression concerning for hypertrophic cardiomyopathy (Figure S1A, B). The proband was transferred to a tertiary care center where she received an echocardiogram, cardiovascular magnetic resonance imaging (CMR), and targeted exome sequencing. Echocardiogram and MRI both showed restrictive physiology and spontaneous resolution of the VSD. The MRI also showed evidence of diffuse fibrosis of the myocardium. These findings prompted a diagnosis of heart failure secondary to RCM. The proband was discharged with a medical management plan including furosemide, spironolactone, enalapril, and carvedilol. Diagnostic genetic testing was performed using the cardiovascular disease panel provided by Gencell Pharma, which revealed a homozygous variant in the TNNC1 gene (NM_003280) resulting in p.A8V (c.C23T). This was predicted to be likely pathogenic due to its association with HCM and RCM by ClinVar and ACMG criteria. Variants of unknown significance were also reported in APOB (c.4056_4064dup, p.G1354_L1356dup) and TTN (c.8005T>A, p.S2669T). No other relevant mutations were identified.

ECG and echocardiogram

At 26 months of age, the proband was hospitalized with a UTI unrelated to her heart failure diagnosis. During this stay she had multiple episodes of marked pallor and hypotonia associated with desaturation and bradycardia. CMR confirmed bi-atrial enlargement with a LA volume of 57 cc/m² and a RA volume of 9 cc/m² with normal ventricular volume and function (LVEF: 51%; RVEF: 53%), consistent with RCM (Figure 1A). Holter data showed multiple events of monomorphic ventricular tachycardia (Figure S1A). Echocardiogram showed bi-atrial enlargement with normal ventricles and ejection fraction (LVEF: 55%). Doppler imaging of the mitral valve revealed an E/A of 0.67, mild mitral regurgitation, and a deceleration time of 155 ms (Figure 1B). LV mass was 41 g/m² (95^th percentile for BSA). Strain analysis demonstrated abnormal strain in the inferior and anterior segments (Figure 1C). This corresponded to T1-weighted cardiac MRI imaging demonstrating fibrosis with focal enhancement present on the lateral wall basal. Cardiac catheterization revealed a RV end diastolic pressure of 16 mmHg and a wedge end diastolic pressure of 7 mmHg. At that time, the proband was deemed a transplant candidate and was discharged on medical management while awaiting a heart transplant, and ultimately received an orthotopic heart transplant three months later.

Figure 1. TNNC1-A8V variant manifests into hallmark RCM phenotypes.

A. Patient cardiac CMR at end-systole and end-diastole, highlighting dilated atria with normal ventricular function; characteristic of RCM. B. Patient echocardiogram exhibiting E/A ratio of 0.67. C. LV Strain measurement demonstrated increased segmental strain of the inferior base. D-E. Representative H&E immunostained ventricular myocardium from the proband’s explanted heart. E. Note, patchy interstitial and replacement fibrosis with remodeling in the ventricle (highlighted by arrows).

Explanted heart showed endocardial fibrosis and inflammation

Gross examination of the explanted heart showed bi-atrial dilation with no other pathological abnormalities were noted. However, H&E staining of ventricular sections showed patterns suggestive of endocardial fibrosis and focal inflammation, consistent with RCM (Figure 1D). Additionally, subendocardial myocyte vacuolization, indicative of chronic ischemic injury and commonly seen in transplant candidates, was observed in the proband’s explanted heart (Figure 1E). Thus, pathology results from the proband’s explanted heart supports the clinical findings of RCM.

Spectrum and penetrance of cardiomyopathy development associated with the TNNC1-A8V variant

To further understand the disease spectrum and penetrance associated with the TNNC1-A8V variant, we performed Sanger sequencing of 12 family members of the proband and performed a comprehensive literature review of patients with TNNC1-A8V variants. Sanger sequencing of the family members revealed that 8 out of the 12 individuals were heterozygous for the c.C23T (p.A8V) genetic variant in TNNC1 (Figure 2, Table 1). All individuals who were either genotype WT or heterozygous for the TNNC1-A8V were asymptomatic and alive and did not have evidence of disease (Table 1). Further, we identified 13 additional cases from the literature with a verifiable genetic test confirming a TNNC1-A8V variant (Figure 3A). Among these 25 individuals, we found that heterozygous individuals for TNNC1-A8V, were susceptible to cardiomyopathy in adult life with an 24% chance of developing cardiomyopathy by 40 years of age (Figure 3B). Among these individuals, 6% were diagnosed with RCM, 41% with HCM while the remaining 53% were healthy at the time of last follow-up (Figure 3C). Conversely, individuals found to be homozygous for TNNC1-A8V were universally diagnosed with RCM within the neonatal and infant stage of life (Figure 3D). Collectively, these findings suggest homozygous variants are highly penetrant with infant-onset RCM, whereas heterozygous individuals demonstrated lower penetrance, adult-onset disease that can manifest as either RCM or HCM.

Figure 2. Autosomal recessive inheritance of TNNC1-A8V variant results in RCM and VT.

A. Sanger chromatograms of TNNC1 region of interest in the proband’s mother (upper) and half-sister (lower). Heterozygosity of the c.23C>T (p.A8V) variant is present in both subjects. B. Pedigree of proband. Phenotype of family was gathered from verbal history by the parents of the proband. No clinical testing was performed.

Table 1.

Summary of Sanger sequencing results of family members to identify TNNC1-A8V variants and mode of inheritance.

Patient ID	TNNC1 Genotype	Sex	Age at Diagnosis (years)	Age at Last Evaluation (years)	Cardiovascular Phenotype	Status at Last Follow-up
I.1	WT/WT	F	N/A	N/A	Asymptomatic without evaluation	Alive, 54 yrs
I.2	WT/A8V	M	N/A	N/A	Asymptomatic without evaluation	Alive, 62 yrs
I.3	WT/A8V	F	N/A	58 yrs	Asymptomatic without evaluation	Alive, 58 yrs
I.4			N/A	57 yrs	None, asymptomatic	Alive, 59 yrs
II.1	WT/A8V	F	N/A	N/A	Heart murmur at birth. Asymptomatic without evaluation	Alive, 28 yrs
II.2	WT/A8V	F	N/A	N/A	Asymptomatic without evaluation	Alive. 34 yrs
II.3			N/A	N/A	N/A	N/A
II.4	WT/A8V	M	N/A	N/A	Asymptomatic without evaluation	Alive, 36 yrs
II.5	WT/A8V	F	N/A	30 yrs	Healthy with cardiac evaluation	Alive, 34 yrs
II.6		M	N/A	N/A
II.7	WT/WT	M	N/A	N/A	Asymptomatic without evaluation	Alive, 31 yrs
II.8	WT/A8V	F	N/A	N/A	Asymptomatic without evaluation	Alive, 27 yrs
III.1	N/A	M	N/A	N/A	Asymptomatic without evaluation	Alive, 7 yrs
III.2	N/A	M	N/A	N/A	Asymptomatic without evaluation	Alive, 10 yrs
III.3	WT/WT	F	N/A	N/A	Asymptomatic without evaluation	Alive, 16 yrs
III.4	A8V/A8V	F	19 months	7 yrs	VSD, left arial enlargement and LV enlargement	Orthotopic Heart Transplant / Alive
III.5	WT/A8V	F	N/A	13 yrs	Asymptomatic with cardiac evaluation	Alive, 16 yrs
III.6	N/A	F	N/A	N/A	Asymptomatic without evaluation	Alive, 14 yrs
III.7	N/A	F	N/A	N/A	Asymptomatic without evaluation	Alive, 4 yrs
III.8	N/A	F	N/A	N/A	Asymptomatic without evaluation	Alive, 7 yrs

ventricular septal defect (VSD), not available (N/A).

Figure 3. Penetrance and phenotypic spectrum of cardiomyopathy associated with the TNNC1-A8V variant.

A. Cohort assembly. Individuals carrying the TNNC1-A8V variant were identified through literature review and evaluation of the proband’s family. Individuals with a confirmed TNNC1-A8V variant (n=20) were included in downstream analyses (B–D). Proband family members negative for TNNC1-A8V (n=3) served as gene-negative controls. B. Kaplan–Meier curve showing freedom from cardiomyopathy by age, stratified by genotype: wild-type (WT/WT), heterozygous (WT/A8V), and homozygous (A8V/A8V). Survival curves differed significantly among groups (log-rank test, P = 0.0003). C. Phenotypic distribution among heterozygous TNNC1-A8V carriers (n=17): 9 without cardiomyopathy, 7 with hypertrophic cardiomyopathy (HCM), and 1 with restrictive cardiomyopathy (RCM). D. Phenotypic distribution among homozygous TNNC1-A8V carriers (n=3), all with RCM. Data were analyzed using the log-rank test with Holm-Šídák’s adjustment to determine statistical significance.

Hemodynamic characterization of restrictive physiology by pressure–volume loops (PV loops) in mice

In vivo hemodynamic assessment of 12-week-old A8V male and female mice relative to wild-type (WT) controls showed phenotypic characteristics reminiscent of human RCM. Representative loops are illustrated in Figure 4A. The A8V mice showed a significant steeper end-diastolic pressure-volume (EDPVR) relative to the WT group, a key parameter reflecting increased ventricular stiffness and reduced compliance during filling (Figure 4B). Moreover, the mutant mice showed reduced minimum rate of pressure change during diastole (dP/dt min) and prolonged relaxation constant (Tau) compared to WT, indicating that the ventricle takes longer to reach pressures low enough for optimal filling. These changes were accompanied by a decrease in end-diastolic volume (Ved) and increase in mean ventricular pressures (P mean) corresponding to delayed relaxation (Figure 4B). In addition, arterial elastance (Ea) was significantly increased in mutant mice, indicating elevated afterload and an increased vascular burden on the LV (Table 2). Functional parameters such as stroke work (SW) and cardiac output (CO), key indicators of LV pump performance were significantly reduced in the A8V group (Figure 4B).

Figure 4. Cardiac function and hemodynamic changes in A8V mice by 12-weeks of age.

A. Representative pressure-volume (PV) loops from WT (top) and A8V (bottom) mice. B. Results for slope of end-diastolic pressure-volume relationship (EDPVR), minimum rate of pressure change (dP/dt min), relaxation constant (Tau), maximum rate of pressure rise (dP/dt max), volume at diastole (Ved), mean ventricular pressures (P mean), stroke work (SW) and cardiac output (CO). Statistical significance was determined using the non-parametric Mann–Whitney U test, * p<0.05, ** p<0.01, *** p<0.001, ****p<0.0001; (WT n=9, A8V n=10).

Table 2.

Summary table of echocardiographic data and key hemodynamic parameters of WT and A8V mice.

Parameters
Echocardiogram	WT (n=5)	A8V (n=5)
Heart rate (bpm)	576 [552,615]	483 [457, 549] ***
LV Mass (mg)	117 [72, 141]	84 [64, 99]
LVEF (%)	69.2 [41.9, 76.2]	84.2 [76.7, 87.1] *
LVFS (%)	37.7 [20.5, 43.7]	52.2 [44.2, 55.4] *
LVEDD (mm)	3.2 [3.2, 4.3]	3.2 [2.8, 3.5]
LVESD (mm)	2.4 [1.8, 3.4]	1.6 [1.4, 1.8]
EDV (uL)	42.5 [40.9, 83.1]	44.6 [31.2, 51.9]
ESV (uL)	12.5 [9.9, 48.0]	6.7 [4.9, 10.8]
MV E/A	1.3 [1.2, 1.4]	1.2 [1.1, 1.7]
IVCT (ms)	14.2 [13.3, 15.0]	17.5 [16.7, 19.6] *
IVRT (ms)	20.8 [17.3, 25.7]	26.7 [21.2, 29.2]
RWT (%)	50 [45, 64]	63 [58, 69]
Hemodynamics	WT (n=9)	A8V (n=10)
SW (mmHg × ml)	2692 [2074, 2972]	1786 [1385, 2179] *
SV (ul)	40.4 [32.4, 47.3]	27.5 [22.4, 38.5] *
CO (ml/min)	24.0 [19.3, 26.1]	15.8 [11.8, 18.7] ***
Ea/Ees	0.6 [0.5, 1.1]	1.0 [0.8, 1.3]
Afterload
ESP (mmHg) (Pes)	65.7 [60.6, 70.5]	69.8 [64.1, 79.9]
Ea (mmHg/ul)	1.6 [1.3, 2.1]	2.6 [2.0, 3.3] **
Preload
EDP (mmHg) (Ped)	5.2 [3.9, 5.6]	7.0 [3.5, 12.6]
EDV (ul)	52.0 [43.3, 65.1]	37.1 [29.7, 46.6] **
Contractility
dP/dT max (mmHg/s)	7408 [6621, 8693]	7603 [5886, 9193]
dP/dT max_EDV (mmHg/s per ml)	110.9 [73.5, 166.4]	100.4 [40.9, 141.4]
Ees (mmHg/ml)	2.9 [1.4, 3.3]	2.8 [2.1, 3.4]
PRSW (mmHg)	75.3 [58.5, 88.0]	71.2 [52.6, 73.1]
Lusitropy
dP/dT min (mmHg/s)	−7824 [−8083, 7365]	−4331 [5144, 3504] ***
EDPVR	0.08 [0.07, 0.1]	0.17 [0.12, 0.24] ***
Tau (ms)	4.8 [4.4, 5.0]	11.6 [8.5, 14.6] ****

Data are expressed as median [25th percentile, 75th percentile]. Statistical significance was determined using the non-parametric Mann–Whitney U test;

^*p < 0.05;

^*p<0.05,

^**p<0.01,

^***p<0.001,

^****p<0.0001.

ejection fraction (LVEF), fractional shortening (LVFS), LV end-diastolic diameter (LVEDD) and end-systolic diameter (LVESD), end-diastolic volume (EDV), end-systolic volume (ESV), mitral valve (MV) E/A ratio, isovolumetric contraction time (IVCT), isovolumetric relaxation time (IVRT), relative wall thickness (RWT), stroke work (SW), stroke volume (SV), cardiac output (CO), arterial elastance (Ea) and end-systolic elastance (Ees) ratio, LV end-systolic pressure (ESP), end-diastolic pressure (EDP), maximal rate of pressure rise (dP/dTmax), relation of maximal rate of pressure rise and end diastolic volume (dP/dTmax_EDV), preload recruitable stroke work (PRSW), maximal rate of pressure decline (dP/dTmin), end-diastolic pressure volume relationship (EDPVR), relaxation time constant calculated by Weiss method (Tau).

Altered myocardial strain and reverse strain in A8V mice

We assessed myocardial deformation using speckle tracking echocardiography (STE) in 10-week-old male and female A8V mice, along with age- and sex-matched WT controls. We found that at this time point, the A8V group showed a hyperdynamic systolic phenotype, characterized by significantly higher global longitudinal (GLS) and radial (GRS) strain compared to WT. Conversely, evaluation of diastolic function via reverse GLS revealed a significant decrease in the mutant group, indicating impaired early myocardial relaxation (Figure 5A). For the segmental LV analysis, the LV was divided into seven segments for the PLAX view (Figure 5B), we measured longitudinal strain and reverse peak strain (strain Pk), and time-to-peak (TTP) was used to measure the speed of myocardial deformation (Supplemental Table 1). Reverse longitudinal segmental strain (RS) TTP analysis revealed hallmarks of impaired relaxation in all LV segments except the apical lateral and apex region (Figure 5C). Representative strain and reverse strain traces are illustrated in Figure 5D. These findings suggest asymmetric regional wall remodeling and impaired LV wall synchrony for individual segments.

Figure 5. A8V mice exhibit LV segmental dyssynchrony and altered global myocardial strain.

A. Quantification of global myocardial strain in longitudinal (GLS), radial (GRS), and reverse strain. B. Schematic of the LV segmented into anatomical regions for strain analysis C. Quantification of reverse strain (RS) time-to-peak (TTP) in longitudinal direction for the mid-inferolateral, mid-anteroseptal, basal-anteroseptal, and apical-septal segments. D. Representative images of strain and reverse strain analysis for both radial and longitudinal myocardial orientations in WT (D) and A8V (E). Each line represents the average strain values of pre-defined LV segments of all myocardial layers between the endocardium and epicardium from three consecutive cardiac cycles. Statistical significance was determined using the non-parametric Mann–Whitney U test, * p<0.05, ** p<0.01, *** p<0.001, ****p<0.0001, (WT n=5, A8V n=5). Abbreviations: RS, reverse strain; TTP, time to peak.

A8V mice display both systolic and diastolic dysfunction

Echocardiography was employed to further evaluate cardiac structural or functional changes at 10-weeks of age. Cardiac function, assessed as percent LV ejection fraction (LVEF) and fractional shortening (LVFS) were both significantly higher in the A8V mice compared to WT mice, consistent with a hypercontractile phenotype often associated with more advanced HCM/RCM (Table 2). Structurally, the relative wall thickness (RWT) showed a trend toward an increased in the A8V mice compared with WT. However, LV end-systolic diameter (LVESD) and end-diastolic diameter (LVEDD) did not differ between groups. Representative PSAX M-mode traces are displayed in Supplemental Figure 2. Importantly, there were no significant differences in LV mass or volume at systole or diastole between A8V mice and WT, suggesting an absence of global hypertrophy at this time point. In addition, pulse-wave doppler of the mitral inflow showed that isovolumetric contraction time (IVCT) was significantly increased in the A8V, while the relaxation time (IVRT) showed a trend towards prolongation. Mitral valve (MV) E/A ratio showed a trend towards an increase, but this did not reach statistical significance (Table 2). Therefore, these findings indicate that there are signs of subclinical changes in diastolic and systolic function.

A8V mice display changes in atrial electrophysiology using electrocardiography (ECG)

Cardiac electrical activity of A8V mice was assessed by ECG under light anesthesia and compared with WT controls at 10-weeks of age. Representative ECG traces for both A8V and WT mice are illustrated in Figure 6A. Among the ECG parameters evaluated, only P-wave amplitude was significantly increased in A8V mice compared to WT controls. Trends toward RR-interval prolongation and increased R-wave amplitude were observed but did not reach statistical significance (Figures 6A and 6B). No significant differences were detected in QRS duration or any other ECG parameters between the two groups (Supplemental Table 2).

Figure 6. Abnormal 8-Lead surface ECGs and evidence of fibrosis in A8V mouse hearts.

A. Representative Lead II signal average ECGs (SAECGs) for WT and A8V mice. Note, abnormal P-wave morphology from A8V mice; indicative of bi-atrial enlargement. B. R-R interval (RR-I) prolongation, increased P-wave amplitude (P-Amp), and reduced R-wave amplitude in A8V mice; (WT n=5, A8V n=5). C. Representative Masson’s trichrome immunostained hearts from 10-week-old mice. Note, areas of interstitial fibrosis (yellow asterisk). Scale bar, 1 mm. D. Percent fibrosis from WT (n=3) and A8V mice (n=4). Statistical significance was determined using the non-parametric Mann–Whitney U test, * p<0.05, ** p<0.01.

Furthermore, heart sections stained with Masson’s trichrome showed collagen deposition scattered throughout the LV myocardium, suggesting mild-to-moderate fibrotic response in A8V mice (Figure 6C, D). Subendocardial fibrosis was observed in both the proband and the mice. We next evaluated whether A8V mice have an arrhythmic susceptibility that can be unmasked with intracardiac programed electrical stimulation and β-adrenergic agonism. Baseline electrophysiological testing using programmed electrical stimulation (PES) showed no significant differences in the inducibility of atrial or ventricular arrhythmias between A8V and WT mice (Supplemental Figure 3). To further assess arrhythmogenic potential under stress, the same cohort underwent two additional rounds of PES following administration of the β-adrenergic agonist, isoproterenol (1 mg/kg and 10 mg/kg via i.p.). Even under heightened sympathetic stimulation, no differences in arrhythmia susceptibility were observed between cohorts (Supplemental Figure 3A–D).

Discussion

Restrictive cardiomyopathy (RCM) is classically known as a disease affecting the ventricles’ ability to relax and fill properly resulting in diastolic dysfunction. Among the genetic variants associated with RCM, mutations in thin filament genes are a more rare cause of the disease⁶. Despite their lower prevalence, these variants are often associated with a particularly severe clinical phenotype^10,28. Patients with variants in thin filament genes tend to have milder hypertrophy and less often have LV outflow tract (LVOT) obstruction, a more aggressive disease course, with a higher risk of progression to heart failure (HF) with systolic dysfunction or restrictive physiology²⁹. This is especially notable in cases with pediatric onset, where the disease may progress rapidly and require advanced HF therapies, including early consideration for cardiac transplantation^29,30. Hypertrophic cardiomyopathy (HCM) is frequently characterized by prominent diastolic dysfunction, which in some cases can be severe enough to produce a restrictive filling pattern, complicating the distinction from primary RCM. This overlap reflects the well-recognized heterogeneity of sarcomeric cardiomyopathies, where variants in a single gene can give rise to multiple phenotypic expressions². In contrast, the defining feature of primary RCM is impaired ventricular filling due to increased myocardial stiffness, leading to elevated end-diastolic pressures and consequent atrial dilation. Together, these shared pathophysiological features highlight the challenge of differential diagnosis and support the concept of an HCM/RCM mixed phenotype in certain contexts.

Our findings provide evidence of early diastolic dysfunction in a A8V mouse model consistent with an RCM phenotype in a pediatric proband who is homozygous for a TNNC1 mutation, p.A8V (c.C23T). While TNNC1 is a well-established causative gene in DCM, recent findings have expanded its phenotypic spectrum to definitively include HCM²¹. To date, at least seven variants in TNNC1 have been identified in HCM and RCM patients^22,31,32. Our findings provide additional evidence that TNNC1 variants, particularly in an autosomal recessive state, can drive an RCM phenotype, potentially representing a more severe disease manifestation, especially in pediatric populations.

The proband presented with clinical features consistent with RCM, including LV enlargement and biatrial dilation, phenotypes closely mirroring previously reported TNNC1-related cases^17,25. In addition to biatrial dilation, histological analysis of the explanted heart from the proband also reveals interstitial fibrosis, and focal inflammation. Clinical data, including ECG and echocardiography, demonstrated hallmark features of RCM, reinforcing the diagnosis. The probands bull’s-eye strain map revealed impaired GLS, a sensitive marker of early contractile dysfunction, despite preserved ejection fraction. GLS correlates closely with the extent of LV hypertrophy and may be reduced even in early disease stages³³.

Cardiac remodeling and impaired diastolic function in A8V mice

Echocardiographic assessment revealed increased LVEF and LVFS, with no changes in LV mass or chamber dimensions, consistent with a hypercontractile phenotype and preserved ventricular size at this time point. It was previously reported that at 12-weeks of age, this mouse model presents with atrial enlargement, increased isovolumetric relaxation time (IVRT), and an elevated mitral valve E/A ratio, with significantly reduced LV dimensions by 14-months, consistent with progressive diastolic dysfunction²⁵. In contrast, our study included in vivo hemodynamic assessment by pressure-volume (PV) loop analysis, the gold standard for assessing diastolic function. While our echocardiography findings were consistent with those of Martins et al²⁵, the younger age of our animals and the smaller cohort may explain why we observed only a trend toward diastolic dysfunction that did not reach statistical significance.

Despite this preserved systolic performance, A8V mice showed hallmarks of diastolic dysfunction in the PV loops such as an elevated EDPVR and tau constant, consistent with the restrictive physiology observed in the human proband. Notably, myocardial fibrosis was observed in the A8V mice but not in WT controls, suggesting a direct link between the A8V variant and fibrotic remodeling. The presence of fibrosis likely contributes to the observed diastolic dysfunction by increasing myocardial stiffness, further supporting an RCM-like phenotype in this model. Importantly, while no arrhythmias were detected in mutant mice, which suggests that electrical instability is not an early feature of this model, A8V mice displayed a significant increase in P-wave amplitude compared to controls. In the context of surface electrocardiography, increased P-wave voltage is an indicator of atrial remodeling and enlargement^34,35.

Echocardiographic strain imaging has the advantage of detecting early cardiac involvement, even before thickened walls or symptoms are apparent in the conventional echocardiography measurements³⁶. Consistent with a hypercontractile phenotype, 10-weeks-old A8V mice showed increased global radial and longitudinal strain, while reverse strain was reduced in the longitudinal direction, indicating that the LV does not expand adequately during filling. Additionally, segmental analysis revealed mechanical dyssynchrony, with prolonged time-to-peak reverse segmental strain in the basal and mid segments with preservation at the apex, suggesting that the myocardium takes longer to reach its maximal relaxation during diastole. These findings align with the prolonged Tau and slower pressure decline (dP/dt_min) observed in PV loops analysis, supporting impaired relaxation, elevated filling pressure, and reduced filling efficiency. These regional abnormalities indicate spatially heterogeneous sarcomere function and abnormal mechanical activation, potentially predisposing to asymmetric remodeling, arrhythmogenesis, and progressive cardiac remodeling.

The homozygous Tnnc1 variant reproduced the key clinical manifestations of the human disease, early-onset RCM, severe diastolic dysfunction, and myocardial fibrosis, features that are rarely reported together in previous sarcomeric cardiomyopathy models.

Functional consequences of the A8V variant in cardiac muscle regulation

The A8V variant is located in the N-terminal helix of cTnC (Figure 7), a crucial Ca²⁺-binding subunit of the troponin complex that is involved in initiating muscle contraction. A recent structure of the native cardiac thin filament demonstrated that the troponin complex is a dynamic structure and that the N-helix of TnC adopts different positions within the troponin core of inhibited and activated thin filaments^37,38. Therefore, we predict this variant disrupts the molecular structure of cTnC, affecting its Ca²⁺-binding affinity and downstream contractile regulation²⁵. In vitro work on Ca²⁺ kinetics showed that incorporation of the A8V mutation into thin filaments predominantly slowed the Ca²⁺ dissociation rate (k_off), rather than affecting the association rate (k_on), thus increasing the apparent Ca²⁺ affinity relative to the WT protein³⁹. These findings suggest that the slower k_off contributes to impaired relaxation and diastolic dysfunction. Several sarcomeric mutations associated with HCM, including those in TnC, exert pathogenic effects by altering the myofilament Ca²⁺ sensitivity^40,41. These findings established cTnC as a promising target for drug development.

Figure 7. Location of A8V within the cTnC N-helix and the troponin core complex in the cardiac thin filament.

The troponin complex is comprised of three major units: troponin I (dark green), troponin T (light green), and the Ca²⁺-binding troponin C, which, in turn, is comprised of the N-terminal lobe (N-lobe, cyan and magenta) and C-terminal lobe (C-lobe, blue). The N-lobe contains the regulatory Ca²⁺-binding site that binds Ca²⁺ (red sphere) that primes the regulatory unit for activation. The N-helix (magenta) that harbors A8V (magenta spheres) is located between the N- and C-lobes and in close proximity to troponin I. The position of the N-helix in the center of the troponin complex suggests an important role in molecular mechanisms of cardiac thin filament regulation.

Clinical implications for early genetic diagnosis

As our understanding of the severe phenotypes linked to thin filament variants grow, such as the one described in this study, underscores the importance of implementing preventative measures before clinical symptoms appear. Genetic testing for instance remains one of the most accessible and rapid approaches for identifying genetic variants, enabling better informed and personalized patient care⁴². We demonstrate the early signs of RCM in both a patient and a genetic mouse model, revealing overlapping mechanistic defects. We evaluate the penetrance of the TNNC1-A8V variant using cases reported in the literature and from samples and outcomes provided by the probands family. Future studies will be necessary to model thin filament variants for potential drug therapeutic studies, which may advance the development of precision medicine approaches to help shed more light on possible precision medicine therapeutics.

Study Limitations/Future Directions

The relatively small sample size of a cohort of 25 individuals with the TNNC1-A8V variant limits the generalizability of the findings and may not capture the full variability of disease presentation. Future studies will require using larger sample sizes to further understand the penetrance of the TNNC1-A8V variant. Moreover, the mechanisms underlying why mutations in the same gene can result in distinct cardiomyopathy phenotypes remain unclear. This highlights the complexity of gene–phenotype relationships and suggests that additional genetic, epigenetic, or environmental modifiers may influence disease expression. The observed genetic overlap among cardiomyopathy subtypes further complicates variant interpretation and clinical classification. Finally, the broad phenotypic spectrum of RCM presents a challenge for establishing clear diagnostic boundaries.

Supplementary Material

Supplemental Tables 1–2

Supplemental Figures 1–3

Nonstandard Abbreviations and Acronyms

HCM

hypertrophic cardiomyopathy

RCM

restrictive cardiomyopathy

SCD

sudden cardiac death

PSAX

parasternal short axis

PLAX

parasternal long axis

STE

speckle-tracking echocardiography

NSVT

Non-sustained ventricular tachycardia

PVC

premature ventricular complex

VSD

ventricular septal defect

CMR

cardiovascular magnetic resonance

PES

programmed electrical stimulation

LVOT

Left ventricular outflow tract

GLS

global longitudinal strain