A Heterozygous Mutation in Cardiac Troponin T Promotes Ca2+ Dysregulation and Adult Cardiomyopathy in Zebrafish

Cardiomyopathies are a group of heterogeneous diseases that affect the muscles of the heart, leading to early morbidity and mortality in young and adults. Genetic forms of cardiomyopathy are caused predominantly by mutations in structural components of the cardiomyocyte sarcomeres, the contractile units of the heart, which includes cardiac Troponin T (TnT). Here, we generated mutations with CRISPR/Cas9 technology in the zebrafish tnnt2a gene, encoding cardiac TnT, at a mutational “hotspot” site to establish a zebrafish model for genetic cardiomyopathies. We found that a heterozygous tnnt2a mutation deleting Arginine at position 94 and Lysine at position 95 of TnT causes progressive cardiac structural changes resulting in heart failure. The cardiac remodeling is presented by an enlarged atrium, decreased ventricle size, increased myocardial stress as well as increased fibrosis. As early as five days post fertilization, larvae carrying the TnT RK94del mutation display diastolic dysfunction and impaired calcium dynamics related to increased Ca2+ sensitivity. In conclusion, adult zebrafish with a heterozygous TnT-RK94del mutation develop cardiomyopathy as seen in patients with TnT mutations and therefore represent a promising model to study disease mechanisms and to screen for putative therapeutic compounds.


Introduction
Cardiomyopathies are a heterogeneous group of diseases that affect the myocardium, causing sudden cardiac death and early mortality in children and adults [1]. Patients within the same family can range from being asymptomatic to having mild-severe symptoms [2]. There are several types of cardiomyopathies that are categorized based on their morphological and pathophysiological differences. These include hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), and restrictive cardiomyopathy (RCM) [3]. Initially, it was thought that these individual diseases have distinctive features where HCM has a thickened left ventricle wall and septum, while DCM causes dilation and thinning of the left ventricle wall, and RCM promotes stiffening of the ventricle walls and restrictive filling. However, in recent years, it is becoming more prominent that these cardiomyopathy types can come as a spectrum of disease with overlapping features [4,5]. These features can include the progression of fibrosis, changes in the ventricle wall structure, disruption in the contractility and intracellular calcium (Ca 2+ ), and dysregulation in the electrical functions of the heart. Although these diseases can have various underlying factors, many progress as a result of mutations in proteins of the sarcomere, the contractile unit of the striated muscles. Hence, they are referred to as sarcomeric cardiomyopathies.

Zebrafish Husbandry
Fish used in this study were housed under standard conditions as described previously [21]. All experiments were conducted in accordance with the ethical guidelines and approved by the local ethics committee of the Royal Dutch Academy of Sciences (KNAW).

Generation of Mutant Lines
Tnnt2a lines were generated using CRISPR/Cas9 genome editing technology. Open reading frame of tnnt2a was chosen using CHOPCHOP (http://chopchop.cbu.uib.no/, accessed on 1 May 2014 [22]). sgRNA template was purchased from Integrated DNA Technologies (IDT) as standard desalted same-day oligo, and synthesis was carried out using MEGAshortscript T7 kit (Ambion). Purification of the in vitro synthesized mRNA was performed using RNeasy Mini Kit (Qiagen). sgRNA and Cas9-encoding mRNA were co-injected into one-cell stage wild-type (TL) zebrafish embryos at a concentration of 12.5 ng/µL sgRNA and 150 ng/µL Cas9-encoding mRNA. Injected embryos were grown to adulthood to identify F0 founder fish with indel mutations and screen for germline transmission. Each putative founder adult zebrafish was crossed with a wild-type adult fish (F1). Details of sequencing primers and sgRNA sequences can be found in Figure S1. Sequencing analysis of the targeted region revealed mutation carrier fish with either 5 or 6 nucleotide deletion (referred to as tnnt2a +/hu11260 and tnnt2a +/RK94del ).

Adult Zebrafish Heart Isolations and Preparation
For paraffin sections, adult zebrafish hearts were dissected and fixed in 4% paraformaldehyde (dissolved in phosphate buffer containing 4% sucrose) at 4 • C overnight, washed twice in PBS, dehydrated in EtOH, and embedded in paraffin. Serial sections were made using a microtome (Leica RM2035, Leica, Wetzlar, Germany). Different section thicknesses were used, depending on the size/age of the sample: 10 µm for adult hearts, 8 µm for juvenile hearts, and 6 µm for embryos. For cryosections, zebrafish hearts were extracted and fixed in 4% paraformaldehyde (in phosphate buffer) for 4 h at room temperature (RT). Three washes of 30 min were performed using 4% sucrose (in phosphate buffer) followed by an overnight incubation at 4 • C in 30% sucrose (in phosphate buffer). Hearts were embedded in tissue-freezing medium (Leica, Lot# 03811456), frozen on dry ice, and kept at −80 • C. Cryo-sectioning using Cryostar NK70 (Thermo Scientific, Breda, The Netherlands) was performed to obtain 10 µm thin sections. Images of extracted whole hearts were acquired using a Leica M165 FC stereomicroscope.

Adult Zebrafish Heart Staining
In situ hybridization (ISH) was performed on paraffin sections as described previously [23], with the exception that the hybridization buffer did not contain heparin and yeast total RNA. vmhcl, amhc, nppb, tnnt2a, and col6a1 probes were previously generated [24]. Primers for the postnb probe generation were forward primer AGAGGTTCTG-GACAGGCTCA and reverse primer AAGGCACCATTTTTCACCAG. Haematoxylin and eosin (H&E) staining and acid fuchsin-orange staining (AFOG) were performed on paraffin and cryosections, respectively, in accordance with standard laboratory protocols. Imaging of stained sections was performed using Leica DM4000 B LED upright automated microscope and Olympus Slideview VS200 digital slide scanner.

High-Speed Brightfield Imaging
Embryos were placed in 200 µM 1-phenyl-2-thiourea (PTU) 20-24 hpf to prevent pigmentation. At 5 dpf, embryos were embedded in 0.3% agarose prepared in E3 medium containing 16 mg/mL MS-222. Recordings were performed at 150 frames per second (fps) using a high-speed inverted light microscope at 28 • C. Heart rate measurements and contractility parameters were analyzed using ImageJ (USA National Institutes of Health, Bethesda, MD, USA). Contraction time, relaxation time, and systolic duration were analyzed by going through the recording frame by frame manually and observing the moments when (1) the ventricular wall moved inward, and the ventricle started expelling blood (start of contraction), (2) the ventricle wall moved back outward (start of relaxation), and (3) the ventricle wall was back at its most dilated position (maximum relaxation). After repeating the process 3 times for each heart, values were averaged, and contractile parameters were calculated. The contraction and relaxation time were corrected for heart rate (HR) by using the formula: time/ √ (beat to beat interval). The systolic fraction was calculated from the systolic time by using: (systolic time × R)/60.000. The hemodynamic parameters such as surface area and volumes were analyzed using ImageJ by drawing an ellipse on top of the ventricle at end-diastole and end-systole. Six ellipses were analyzed per heart: 3 at diastole and 3 at systole. Values were averaged. ImageJ provided the values for the minor and major axis of each ellipse. End-diastolic and end-systolic volume (EDV/ESV) were calculated by: (1/6) × (π) × (major axis) × (minor axis 2 ). Stroke volume (SV) by: EDV-ESV. Ejection fraction (EF) by: SV/EDV. Cardiac output (CO) by: SV × Heart rate. Surface area was calculated using the following formula: (0.5 × major axis) × (0.5 × minor axis) × π. Fractional area change (FAC) was calculated by: (area diastole − area systole)/(area diastole).

Statistical Analysis
Statistical analysis and drawing of graphs and plots were carried out in GraphPad Prism (version 7 for Windows, GraphPad Software, San Diego, CA, USA). Differences between two groups were analyzed using the unpaired Student's t-test, with the exception of the Kaplan-Meier curve, where Log-rank (Mantel-Cox) test was used. All data are presented as mean ± SEM, and p < 0.05 was considered significant. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001, n.s. p > 0.05. N denotes the number of fish used per dataset.

Targeting Exon 9 of tnnt2a via CRISPR/Cas9 Promotes Changes in Heart Morphology of Adult Zebrafish
Zebrafish tnnt2a is located on chromosome 23 and consists of 15 exons. Exon 9 encodes for a highly conserved region within the Tm binding region which is often mutated in patients with familial cardiomyopathy, including the Arginine 92 site (corresponding to Arginine 94 in zebrafish) [9] ( Figure 1A). We therefore designed a guide RNA to target this region for mutagenesis with CRISPR/Cas9. From this, we recovered two F1 lines with different mutations. In one of the lines (hu11260), we identified a deletion of five nucleotides causing a frameshift followed by a stop codon resulting in a truncated protein lacking most of the Tm binding region and the entire TnC-TnI binding region. In the second line (hu11320), we identified an in-frame deletion of six nucleotides resulting in the deletion of an Arginine at position 94 and a Lysine at position 95, which hereafter is referred to as tnnt2a +/RK94del ( Figure 1A,B).
Homozygous tnnt2a hu11260/hu11260 embryos at two days post fertilization (dpf) showed normal embryo morphology but lacking heart contraction, which resembles the previously reported tnnt2a loss-of-function mutation [20]. Heterozygous tnnt2a +/hu11260 zebrafish were viable with no apparent phenotypic changes, including normal heart rate at 5 dpf, and looked similar to their wild-type siblings in embryonic and adult stages ( Figure 1C,D and Figure S2). Moreover, extracted hearts of adult tnnt2a +/hu11260 zebrafish appeared normal, with no visible morphological changes ( Figure 1C ,C",D ,D"). Homozygous tnnt2a RK94del/RK94del embryos were suffering from defects after heart looping, evident by the lack of heart chamber ballooning, pericardial edema, and weak heart contraction. Interestingly, also heterozygous tnnt2a +/RK94del embryos showed heart defects with variable penetrance. Survival count revealed that tnnt2a RK94del/RK94del embryos die within the first two weeks ( Figure 1F, Figure S3). Strikingly, 70% of tnnt2a +/RK94del zebrafish survive the first two weeks but suffer from heart failure at ages ranging from 10 to 20 weeks post-fertilization (wpf) and have a higher susceptibility for bacterial infection. Only around 30% of tnnt2a +/RK94del survive the first four months of life (120 dpf) ( Figure 1F). Females are noticed to have a decline in their egg production as early as 13 wpf. From 10 wpf onward, tnnt2a +/RK94del mutants develop protruding hearts combined with reduced body length ( Figure 1G). Hearts extracted from 12-weeks-old tnnt2a +/RK94del zebrafish were enlarged, and it was often hard to distinguish the atrium and the ventricle ( Figure 1E-E"). This heart remodeling was observed in 100% of the tnnt2a +/RK94del fish (n = 78) with variable degrees of severity, compared to no remodeling in their wild-type siblings (n = 109). In addition, the phenotype was consistent in more than 15 generations in which the tnnt2a +/RK94del zebrafish were outcrossed with wild-type TL fish, indicating a strong genotype/phenotype correlation. Fish with severely remodeled hearts suffered from breathing difficulties, evident by swimming at the surface gasping for air. These fish also showed a significant decrease in size and length, compared to their wild-type siblings ( Figure 1E,G). In conclusion, by targeting exon9 in tnnt2a, we generated different alleles that display very different phenotypes. While the 5 bp out-of-frame deletion in tnnt2a acts as a recessive mutation and does not cause disease in heterozygous fish, the 6 bp deletion that removes Arg94 and Lys95 is dominant and causes phenotypes in heterozygous fish, including heart failure in adults. 4-month-old adult zebrafish with the genotypes tnnt2a +/+ , tnnt2a +/hu11260 , or tnnt2a +/RK94del , respectively. Scale bar of 0.5 cm. (C'-E') Extracted hearts of the respective fish. Scale bar of 100 µm. (C"-E") Haematoxylin and Eosin (H&E) stainings of heart sections from tnnt2a +/+ , tnnt2a +/hu11260 , or tnnt2a +/RK94del , respectively. Images were taken at a magnification of 20×, and scale bars are 100 µm. A: atrium, V: ventricle, BA: bulbus arteriosus. (F) Kaplan-Meier curve, displaying the survival of tnnt2a +/+ , tnnt2a +/RK94del , and tnnt2a RK94del/RK94del fish during the first 120 days of development (n = 7, 17, and 17, respectively). Statistics: Log-rank (Mantel-Cox) test, mean ± SEM, * p ≤ 0.05, **** p ≤ 0.0001. (G) Measurement of the length of adult fish from tnnt2a +/+ and tnnt2a +/RK94del . Statistics: mean ± SEM, **** p ≤ 0.0001, tnnt2a +/+ n = 26, tnnt2a +/RK94del n = 21, unpaired Students t-test.

Heart Morphological Changes in tnnt2a +/RK94del Can Be Observed at Early Larval Stages of Development
As tnnt2a +/RK94del fish developed phenotypes such as a protruding heart and altered swimming behavior at 10 weeks of age, we wanted to investigate whether morphological differences in the heart occurred and when these started to develop. To achieve this, live imaging on 5 dpf larvae was performed, as well as histological analysis at three different time points being 2, 4, and 7 wpf. Surprisingly, changes were observed at 5 dpf in the atrium. The atrium appeared larger in 20% of tnnt2a +/RK94del larvae, while the ventricle seemed to be unaffected (n = 24) (Figure 2A,B). Histological analysis confirmed this finding and showed that by 2 wpf, the size of the atrium was slightly increased in all tnnt2a +/RK94del fish, while the size of the ventricle seemed unaffected (n = 6) ( Figure 2C,D). Remodeling becomes more apparent at 4 wpf, when the atrium of tnnt2a +/RK94del fish has enlarged further, and the ventricle appears slightly smaller in comparison to the wild-type siblings (n= 6) ( Figure 2E,F). By 7 wpf, the size of the atrium in tnnt2a +/RK94del fish is grossly enlarged, and a clear decrease in the size of the ventricle is now apparent (n = 6) ( Figure 2G,H). To better distinguish the atrium from the ventricle histologically, we analyzed the expression of atrial myosin heavy chain (amhc) and ventricle myosin heavy chain-like (vmhcl) via mRNA in situ hybridization. Both amhc and vmhcl are expressed at comparable levels in hearts of adult tnnt2a +/+ and tnnt2a +/RK94del zebrafish, but the atrium of tnnt2a +/RK94del zebrafish is much larger and contains a thicker outer layer compared to wild-type controls (n = 6) ( Figure 3I,J). The nuclei in the atria of the tnnt2a +/RK94del appear abundant but similarly spaced as in wild-types, suggesting that hyperplasia (an increase in cell number) rather than hypertrophy (an increase in cell size) causes the enlargement of the atrium. The ventricle of tnnt2a +/RK94del fish appears to be morphologically similar to wild-type but is strongly reduced in size (n = 6) ( Figure 3K,L). Together, these results demonstrate that cardiac remodeling in tnnt2a +/RK94del fish starts as early as the larval stage and is characterized by a progressive enlargement of the atrium and a reduction of the ventricle.

Heart of Adult tnnt2a +/RK94del Displays Increased Myocardial Stress
As we observed a strong remodeling in tnnt2a +/RK94del hearts, we wondered whether genes that are upregulated in failing mammalian hearts, such as natriuretic peptides [26], are also upregulated in tnnt2a +/RK94del hearts. Indeed, we detected a higher expression of Natriuretic peptide type B (nppb) in the atrium of the tnnt2a +/RK94del hearts and some sparse expression in the outer edges of the ventricle (n = 6) ( Figure 3A,B, arrowheads). As previous results indicate that the atrium is more severely affected than the ventricle, we investigated whether tnnt2a is differentially expressed between the two heart chambers. In situ hybridization revealed no difference in tnnt2a expression between the atrium and ventricle of wild-type and tnnt2a +/RK94del hearts (n = 6) ( Figure 3D,E). As tnnt2a +/RK94del hearts express only one wild-type copy of the tnnt2a gene, we wondered whether the loss of one tnnt2a wild-type copy is sufficient to induce the observed cardiac remodeling. However, as the hearts of tnnt2a +/hu11260 fish, which also have only one wild-type tnnt2a copy, show normal morphology (n = 6) and lack upregulation of nppb ( Figure 3C), remodeling has to be due to the RK94del mutation. Based on these findings, we concluded that tnnt2a +/RK94del mutants develop evident morphological changes of the heart, characterized by an enlarged atrium and a smaller ventricle. These changes are accompanied by myocardial stress throughout the mutant heart. In situ hybridization on 4-month-old adult heart sections of tnnt2a +/+ , tnnt2a +/RK94del , and tnnt2a +/hu11260 stained for nppb, a marker of myocardial stress. In situ hybridization on adult heart sections of tnnt2a +/+ and tnnt2a +/RK94del stained for (D,E) tnnt2a, to mark cardiac Troponin T expression, and (F,G) col6a1, which marks collagen. (H,I) AFOG staining, which marks collagen in blue, muscle in orange, and fibrin in red of tnnt2a +/+ and tnnt2a +/RK94del . (J,K) In situ hybridization on adult heart sections of tnnt2a +/+ and tnnt2a +/RK94del stained for postnb, to mark fibroblasts. Images were taken at a magnification of 20×, except for (H,I) where magnification was at 60×. Scale bars are 200 µm for whole-heart tile scans, 50 µm for zoom-in regions, and 20 µm for zoom-in regions of H, I. A: atrium, V: ventricle, BA: bulbus arteriosus.

Detection of Increased Collagen and Fibrin Deposition in Atrium of Adult tnnt2a +/RK94del Hearts
Myocardial stress and remodeling are often accompanied by increased fibrosis. To investigate whether this also occurred in tnnt2a +/RK94del hearts, col6a1 expression was analyzed. col6a1 is a commonly used fibrosis marker and is implicated in ECM remodeling. It is elevated during fibrotic conditions, such as cardiac fibrosis, via its role in myofibroblast differentiation [27]. While minimal col6a1 expression was detected in wild-type siblings ( Figure 3F), we observed patches of col6a1 expression around the outer layer of both the atrium and ventricle of adult tnnt2a +/RK94del zebrafish (n = 6) ( Figure 3G, arrowheads). Next, we stained for collagen and fibrin using acid fuchsin orange G (AFOG) staining, where we observed increased collagen levels (in blue) and fibrin deposition (in red) within and around the atrium of tnnt2a +/RK94del , unlike their wild-type siblings (n = 6) ( Figure 3H,I, arrowhead). We then stained for fibroblast marker, periostin (postnb), and we observed an increased expression in tnnt2a +/RK94del fish, which overlapped with the areas of elevated col6a1 expression (n = 6) ( Figure 3J,K, arrowheads). Taken together, these results indicate a moderate increase in fibrosis, mostly in the outer layer of the atrium in tnnt2a +/RK94del fish.

Embryonic tnnt2a +/RK94del fish Show Decreased Heart Rate Accompanied by Changes in Contractile and Hemodynamic Parameters
Structural changes were detected early on in the developmental stages of the tnnt2a +/RK94del hearts, which suggests that very early functional defects may underlie the cardiac remodeling in these mutants. We demonstrated a consistent enlargement of the atrium and a reduction of the ventricle in tnnt2a +/RK94del hearts, which appears similar to ventricular dysfunction with atrial compensation [28]. To investigate the role of ventricular dysfunction, we measured early heart rhythm and ventricular contractility in 5 dpf larvae, using brightfield high-speed video recording combined with image analysis. The highspeed video recording was performed on tnnt2a +/RK94del embryos without an apparent phenotype and on their wild-type siblings ( Figure 4A,B). Image analysis revealed that the heart rate of tnnt2a +/RK94del embryos was significantly decreased compared to wild-type siblings ( Figure 4C). Next, we measured different contractile parameters to observe the ability of the ventricle to contract and relax. Contraction time (time from maximum relaxation to maximum contraction), relaxation time (time from maximum contraction to maximum relaxation), and systolic fraction (the fraction of time the heart is in systole) were measured and corrected for heart rate. We observed a prolonged relaxation time in the ventricle of tnnt2a +/RK94del embryos, indicating that it takes longer before cardiomyocytes return to their uncontracted state ( Figure 4E). Contraction time and systolic fraction were not affected ( Figure 4D,F). Next, we measured hemodynamic parameters to identify the effect of RK94del mutation on cardiac pump function. An ellipse shape was fitted on the ventricle of brightfield high-speed recordings of embryonic zebrafish hearts ( Figure 4G). Using established mathematical formulas, we measured the volume and surface area of each ventricle during diastole and systole. A decrease in end-diastolic volume (EDV) and end-systolic volume (ESV) was observed in tnnt2a +/RK94del embryos ( Figure 4H,I). Ejection fraction (EF), stroke volume (SV), and cardiac output (CO) remained unaffected ( Figure 4J-L). Surface areas were also affected, where there was a significant decrease in systolic and diastolic surface areas, while fractional area change (FAC) remained unaffected ( Figure 4M-O). The results shown here demonstrate that seemingly healthy tnnt2a +/RK94del embryos already display minor changes in heart rhythm and contractility. These changes likely give rise to the remodeling at adult stages. (C-F) Bar graph of contractility parameters, including heart rate, contraction time, relaxation time, and systolic fraction for tnnt2a +/+ and tnnt2a +/RK94del . (G) For hemodynamic parameters, we used ImageJ to fit an ellipse over the ventricle in every recording, both at end-diastole (left) and end-systole (right). (H-L) Bar graphs of hemodynamic parameters, including end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), stroke volume (SV), and cardiac output (CO). (M-O) Bar graphs of surface area parameters: diastolic surface area, systolic diastolic surface area, and fractional area change (FAC). Scale bar: 100 µm. Statistics: mean ± SEM, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001, tnnt2a +/+ n = 19, tnnt2a +/RK94del n = 24, unpaired Students t-test.

The tnnt2a +/RK94del Mutation Causes Impaired Calcium Dynamics in Embryos
While isolating ventricular cardiomyocytes to observe potential structural and functional changes, we observed that cells from tnnt2a +/RK94del fish were highly sensitive and always died during the process of isolation, where wild-type cells remained intact. Since cardiomyopathies progress over a longer time span due to a spectrum of defects, including impaired intracellular Ca 2+ handling, we investigated whether we could observe Ca 2+ dysfunction well before the end-stages of RK94del remodeling. tnnt2a +/RK94del fish were crossed with a tg (myl7:Gal4FF UAS:GCaMP6f) line, expressing a cytosolic cardiac Ca 2+ sensor [25]. Analysis of GCaMP6f (cpEGFP) signal intensity over time allows the examination of Ca 2+ transient amplitudes, the speed of intracellular Ca 2+ release, and the speed of Ca 2+ reuptake/clearance, as shown earlier [25] (Figure 5A,B). We observed changes in Ca 2+ parameters in both the ventricle and atrium of 5 dpf tnnt2a +/RK94del embryos. No significant differences were observed in Ca 2+ transient upstroke time between wild-type siblings and tnnt2a +/RK94del in both heart chambers ( Figure 5C), indicating that the speed of cytosolic Ca 2+ influx is unaffected in both heart chambers. Ca 2+ recovery time was significantly prolonged, indicating slower clearance of Ca 2+ during diastole, and signal amplitudes were significantly increased in the atrium of tnnt2a +/RK94del embryos ( Figure 5D,E). Atrial diastolic Ca 2+ levels and peak Ca 2+ levels were not significantly affected in tnnt2a +/RK94del compared to wild-type siblings ( Figure 5F,G); however, a significant decrease in both parameters was detected in the ventricle of the tnnt2a +/RK94del embryos. This suggests that the ventricle of tnnt2a +/RK94del embryos has a reduced amount of cytosolic Ca 2+ during contraction and relaxation compared to wild-type siblings. Ca 2+ transient frequency was significantly decreased in both the atrium and ventricle of tnnt2a +/RK94del , which is likely linked to the decrease in heart rate that we have shown earlier (Figures 5H and 4C). Overall, the atrium of tnnt2a +/RK94del embryos shows an increase in Ca 2+ amplitude and prolonged recovery time, while the ventricle has lower cytosolic Ca 2+ levels during contraction and relaxation. Therefore, these findings imply that the Ca 2+ dynamics within the hearts of tnnt2a +/RK94del fish are affected as early as 5 dpf.

Discussion and Conclusions
Previously characterized sih/tnnt2 mutant alleles in zebrafish result in either no protein or a severely truncated protein only 11-amino-acids-long [20]. In this study, we targeted a location within the tnnt2a gene that is well known in human patients as the mutational "hotspot" for cardiomyopathies [9]. This location is within the TNT1 region of the protein, which is involved in the binding of cardiac TnT to Tm and is essential for the function of the TnT protein. The TNT1 region is a highly helical domain that stabilizes the TNT1-Tm interaction and increases the affinity of Tm to actin [29][30][31]. Earlier work showed that mutations in the region between amino acids 92-110 of TNNT2 impairs Tm binding or the ability of TnT to enhance tropomyosin-actin binding, leading to improper muscle contraction [31,32]. Flexibility of the TNT1 domain is important in allowing the efficient binding to Tm. Mutations in TNT1 that alter its flexibility lead to changes in Ca 2+ activation in the thin filaments and propagation of the Ca 2+ signal [33]. The regulatory element of Ca 2+ is partially retained in the TNT2 domain of TnT; however, for regulation of contractility, Ca 2+ needs Tm-TNT1 binding to propagate its signal [29]. The changes in TNT1 flexibility promote Ca 2+ alterations, which have been shown to correspond to Ca 2+ sensitivity, where increased flexibility leads to increased Ca 2+ sensitivity [33][34][35]. Indeed, force measurements on detergent-skinned fiber bundles from transgenic mouse cardiomyocytes with mutations in TNT1 found in human cardiomyopathy patients (e.g., R92Q, I79N, and F110I) display increased Ca 2+ sensitivity [36][37][38][39][40][41]. The observed early defects in Ca 2+ dynamics and contractile dynamics presented in our study may very well be linked to changes in flexibility of TNT1 due to the deletion of Arginine and Lysine in the mutational hotspot of TnT, leading to inefficient propagation of Ca 2+ signal to Tm.
The troponin complex is an important regulator of Ca 2+ dynamics as TnC typically binds 50% of the Ca 2+ that is released from the SR [42]. An increase in Ca 2+ sensitivity of the troponin complex, by an increase in Ca 2+ affinity of TnC, is predicted to result in a decrease in peak [Ca 2+ ] [40,43]. This is consistent with our observation of decreased peak Ca 2+ levels in the ventricles of tnnt2a +/RK94del embryos measured with the GCaMP6 sensor. A limitation of the GCaMP sensor is that it does not allow measurement of exact [Ca 2+ ]. This would require measurements on isolated cells, which is hampered by the difficulty to isolate intact cardiomyocytes from embryos and from adult tnnt2a +/RK94del hearts. A higher affinity of the troponin complex for Ca 2+ would also result in a slower release of Ca 2+ into the cytoplasm and a slower clearance from the cell, which translates to a slower Ca 2+ recovery and prolonged relaxation time. Indeed, we observed prolonged relaxation times in tnnt2a +/RK94del embryos. In addition, a decrease in diastolic Ca 2+ levels in the ventricle of tnnt2a +/RK94del embryos was detected, suggesting that more Ca 2+ was bound to troponin when TnT-RK94del is present in the complex. More Ca 2+ bound to troponin at diastole is also consistent with the observed reduced EDV of the ventricles in tnnt2a +/RK94del embryos. Taken together, the reduced EDV and ESV combined with the reduction in cytoplasmic Ca 2+ levels at diastole and systole are consistent with an increase in Ca 2+ sensitivity of the troponin complex due to the TnT-RK94del mutation.
Structural remodeling observed in tnnt2a +/RK94del adult zebrafish is characterized by a reduced ventricular size together with a severely enlarged atrium. The model presented here appears similar to heart failure with preserved EF (HFpEF), as we observed diastolic dysfunction with atrial enlargement while maintaining EF level, a classic appearance of HFpEF [44,45]. The atrium, however, demonstrates increased myocardial stress, collagen deposition, and fibrosis during later stages, causing loss of elasticity and eventual heart failure. Interestingly, transgenic mice expressing a truncated form of Troponin T or mutated form of Troponin T (R92Q mutation) show a similar diastolic dysfunction combined with a decrease in ventricle size and a larger hypertrophic atrium [46,47]. A similar diastolic dysfunction combined with enlargement of the atria has been documented in patients with HCM, including familial cases with mutations in TNNT2 affecting the TNT1 domain [11,12,28,48]. Left atrium dilation is associated with ventricular diastolic dysfunction, which comes as a result of abnormal left ventricular relaxation and changes in actin-myosin interaction [49]. The diastolic dysfunction causes the left atrium pressure to continually increase in order to maintain a proper filling, leading to atrial wall tension and causing dilation and stretching of the atrial myocardium.
In summary, we suggest that the RK94del mutation in TnT affects the ability of Tm to bind TnT, leading to changes in Ca 2+ dynamics, increased Ca 2+ sensitivity, and contractility defects. We showed that the ventricle displays a form of HFpEF, where diastolic function is dysregulated while EF is maintained. The atrium, in this case, is enlarged to relieve the pressure and promote the proper ventricular filling. At a point where the atrium is unable to efficiently maintain the heart function, remodeling, involving increased collagen deposits and fibrosis, progresses ultimately leading to heart failure. The tnnt2a +/RK94del fish model we described here may help identify potential therapeutics in the future, which have the potential to reduce Ca 2+ sensitivity and prevent cardiac remodeling in patients. As zebrafish have a highly conserved drug metabolism, physiology, and pharmacology, and the larvae are small enough to grow in a 96-well plate, they present an ideal model for drug discovery [50,51].
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/jcdd8040046/s1, Figure S1: Sequences for primers and oligonucleotides used in this study, Figure S2: Heart rate from a brightfield high-speed video imaging of tnnt2a+/hu11260 at 5 dpf, Figure S3: Representative embryonic pictures of the different genotypes at 3 dpf and 5 dpf embryos.
Author Contributions: S.M.K.: study design, experiments, data analysis and interpretation, manuscript preparation, expansion, and care of the animal colony. C.D.K.: assistance in high-speed fluorescent imaging recordings, interpretation of calcium experiment findings, manuscript reviewing and editing. F.K.: care of animal colony and initial heart isolations, consultation for techniques, and experimental advice. S.W. and S.C.: design, injections, and generation of the zebrafish F0 generation. J.B.: study conceptualization, main funding, and design, data interpretation, study coordination, and manuscript reviewing and editing. All authors have read and agreed to the published version of the manuscript.