A Novel Homozygous Intronic Variant in TNNT2 Associates With Feline Cardiomyopathy

Background Hypertrophic cardiomyopathy (HCM) is a genetic disease of the heart and the most common cause of sudden cardiac death in the young. HCM is considered a disease of the sarcomere owing to the large number of mutations in genes encoding sarcomeric proteins. The riddle lies in discovering how these mutations lead to disease. As a result, treatments to prevent and/or treat HCM are limited to invasive surgical myectomies or ablations. The A31P variant of cardiac myosin binding protein-C, encoded by MYBPC3, was found to be more prevalent in a cohort of Maine Coon cats with HCM. However, other mutations in MYBPC3 and MYH7 have also been associated with HCM in cats of other breeds. In this study, we expand the spectrum of genes associated with HCM in cats. Results Next Generation Whole Genome sequencing was performed using DNA isolated from peripheral blood of a Maine Coon with cardiomyopathy that tested negative for the MYBPC3 A31P variant. Through risk stratification of variants, we identified a novel, homozygous intronic variant in cardiac troponin T (TNNT2). In silico analysis of the variant suggested that it may affect normal splicing of exon 3 of TNNT2. Both parents tested heterozygous for the mutation, but were unaffected by the disease. Echocardiography analyses revealed that the proband had shown early onset congestive heart failure, which is managed with a treatment regime including ACE and aldosterone inhibitors. Conclusion In summary, we are the first to demonstrate the association between TNNT2 mutations and HCM in felines, suggesting that this gene should be included in the testing panel of genes when performing genetic testing for HCM in cats.


INTRODUCTION
Hypertrophic cardiomyopathy (HCM) is the most common form of genetic heart disease, affecting at least 1 in 500 humans (Ho, 2011). Defined by an unexplained abnormal thickening of the ventricular walls and possible outflow tract obstruction, no cure is currently available for HCM. HCM is primarily caused by mutations in genes that encode sarcomeric proteins. In particular, mutations in MYH7 and MYBPC3, encoding beta-cardiac myosin heavy chain and cardiac myosin binding protein-C, respectively, are the most commonly mutated genes (Viswanathan et al., 2017), although other sarcomeric genes, including TNNI3, TNNT2, and MYL2, have also been implicated in humans (Brouwer et al., 2011). How mutations in these sarcomeric genes result in the development of HCM remains poorly understood.
In cats, cardiomyopathy is the main cause of cardiovascular disease, with HCM presenting as the most common form, suggesting that cats may be an excellent non-rodent animal model of HCM (Ferasin et al., 2003). A variant in MYBPC3, A31P, was previously linked to the development of HCM in Maine Coon cats (Meurs et al., 2005). This variant occurs within the C0 domain of cardiac myosin binding protein-C (cMyBP-C), and it is expected to cause disease via structural defects in this domain important for regulation of actomyosin regulation (Van Dijk et al., 2016). Another variant in MYBPC3, A74T, has also been described across multiple cat breeds. However, indepth follow-up studies demonstrated that this polymorphism is unrelated to cardiomyopathy (Wess et al., 2010;Longeri et al., 2013). A third variant in MYBPC3, R820W, has been associated with HCM, most specifically in the Ragdoll (Meurs et al., 2007). More recently, the first variant in MYH7, E1883K, has been associated with HCM in a Domestic Shorthair cat, expanding the spectrum of genes associated with feline cardiomyopathy (Schipper et al., 2019). A full list of mutations associated with feline cardiomyopathy is shown in Table 1.
In the present study, a male Maine Coon cat with cardiomyopathy and early congestive heart failure was studied. After initially screening negative for the aforementioned A31P mutation, we performed WGS to identify potential mutations and, thus, pinpoint the genetic cause of this cat's cardiomyopathy through study of the diseased cat's family tree. This study expands upon recent Next Generation Sequencing performed in cats (Genova et al., 2018;Ontiveros et al., 2018). As a result, we have identified a novel autosomal recessive mutation in the sarcomeric cardiac troponin T (TNNT2) gene that appeared to be associated with feline cardiomyopathy. This is the first known record of such mutation in felines, suggesting that TNNT2 should be included as a candidate gene in the genetic screenings.

Phenotyping, Blood Collection and DNA Extraction
All procedures followed the protocol approved by the Institutional Animal Care and Use Committee of the University of Cincinnati and complied with the Guide for the Use and Care of Laboratory Animals published by the National Institutes of Health. The proband's owner established contact to determine if the animal's disease could be linked to a genetic mutation. To establish a full research study, a Material Transfer Agreement was initiated between the proband's owner and the University of Cincinnati. Researchers had no contact with any animals beyond receipt of blood samples. All studies involving contact with animals were performed by board certified veterinarians. Diagnosis and treatment of the proband, regular checkups and echocardiography were performed by a board certified veterinary cardiologist using standard echocardiographic techniques and references from a population of healthy Maine Coon cats (Thomas et al., 1993;Drourr et al., 2005). Peripheral blood was collected by a certified veterinarian or veterinarian nurse at each animal's local hospital following a routine check-up. Blood collected locally was transported back to the laboratory on wet ice. Blood that was collected at veterinary clinics farther from the lab was shipped on cold blocks overnight. Upon receipt at the laboratory, all blood samples were aliquoted into 200 µl samples and stored at −80 • C until use. DNA was extracted from a single 200 µl aliquot using the Qiagen QIAamp DNA Mini and Blood DNA isolation kit, following the manufacturer's protocol. Concentration and purity of DNA was analyzed by NanoDrop. For samples sent for WGS, DNA integrity and concentration were also determined using the Agilent 2100 Bioanalyzer and Qubit Fluorometer, respectively.

Whole Genome Sequencing and Analysis
DNA extracted from the whole blood of the proband was submitted to Novogene Co., Ltd., for WGS. DNA quality and quantity were confirmed by Agarose gel electrophoresis and Qubit Fluorometer. A total amount of 1.5 µg high-quality genomic DNA was randomly sheared into short fragments of approximately 350 bp and used for library construction using the NEBNext R DNA Library Prep Kit. Briefly, following end repairing, dA-tailing, and ligation with NEBNext adapter, the fragments were PCR-enriched by P5 and indexed P7 oligos. The concentration of the DNA library was quantified using the Qubit 2.0 fluorometer and diluted to 1 ng/µl. Following dilution, the insert size of the library was assessed with the Agilent 2100 bioanalyzer, and quantitative real-time PCR (qPCR) was performed to detect the effective concentration of the prepared library. Following enrichment and indexing, pair-end sequencing of the qualified library was performed on an Illumina HiSeq X Ten with a read length of PE150bp at each end.

Bioinformatics Analysis
The original sequencing data acquired by high-throughput sequencing platforms (e.g., Illumina HiSeq TM /NovaSeq TM ) and recorded in image files were first transformed to sequence reads by base calling with the CASAVA software. Information on sequences and corresponding sequencing quality was stored in a FASTQ file. Following quality control of raw sequencing data for clean data filtration, each clean read was mapped to the reference genome (Felis_Catus_9.0, Ensembl release 93) using the BWA software (Li and Durbin, 2009), and the mapping rate and coverage were counted according to the alignment results. Duplicates were removed by SAMTOOLS . Single nucleotide polymorphism (SNP) and InDel variants were detected using the GATK software (Depristo et al., 2011). These SNPs were annotated using ANNOVAR as previously described (Wang et al., 2010).

Sanger Sequencing
To test, or confirm, the presence of a genetic variant in parents of the proband, PCR was performed using a highfidelity Taq polymerase to specifically amplify the sequence around the variant. The PCR product was purified and sent for Sanger sequencing using Cincinnati Children's Hospital DNA Sequencing and Genotyping Core Facility.

RESULTS
The proband was a privately-owned male pure-bred Maine Coon. At 8 months of age, he presented with left ventricular, right atrial, and borderline left atrial dilatation and borderline septal hypertrophy, with preserved-to-elevated systolic function ( Figures 1A,B and Table 2), when compared to reference echocardiography values for the Maine Coon (Drourr et al., 2005). Diastolic function could not be quantified as a result of fusion of E and A waves. Progressive enlargement of all four chambers of the heart was noted at 14 months of age, while borderline septal hypertrophy and preserved systolic function was still noted. This was diagnosed as a primary unclassified cardiomyopathy with possible early congestive heart failure. Furosemide, Benazepril, Spironolactone, and Clopidogrel treatments were then initiated by the veterinarian. This combination appeared to stabilize the progression of the heart disease at 18 months ( Table 2). DNA was isolated from whole peripheral blood from the proband. PCR was performed to produce a 251 amplicon around the A31P variant in MYBPC3 previously associated with HCM in Maine Coon cats (primer sequences in Table 3). Sanger sequencing of the PCR product showed no mutation at codon 31, indicating that the proband did not carry the pathogenic A31P variant in MYBPC3 ( Figure 1C).
As the proband was negative for the A31P mutation in MYBPC3, we asked if any other variants could be traced to the proband's cardiomyopathy. Accordingly, DNA was freshly isolated for WGS to identify potential variants present in the proband. High-quality DNA with an approximate size of 22944bp   (Figures 2A-C). The detected GC content was 43.06% compared to the reference genome of 41.73%. Together, these data demonstrated sufficiently high quality of sequencing data.
Initial analysis reported more than 20,000 genetic variants identified from the proband, compared to the reference genome. We employed a cardiovascular disease risk stratification step in order to enrich for variants associated with cardiovascular disease. This stratification was with 174 genes covered by the Illumina TruSight Cardio Panel (Pua et al., 2016). This approach reduced the number of variants to 305 (Figure 2D). Of these 305 potential variants, only one had been annotated as having "high risk" impact. This particular variant was a single base pair substitution (G to A) within intron 3, corresponding to c.95-108G > A of ENSFCAG00000004613. This variant resulted in the potential inclusion of a de novo splice acceptor site ( Figure 3A). We then used in silico analysis to determine the likelihood that this intronic variant indeed acts as a novel splice acceptor site (Wang and Marín, 2006). This in silico method correctly identified greater than 96% of canonical splice donors and acceptors within TNNT2 (Table 4). Of note, in silico analysis of the c.95-108G > A variant in the proband supported   the hypothesis that this variant alters TNNT2 splicing. The consequence of this aberrant splicing was investigated at the protein level, which revealed that the variant could cause the loss of 223 carboxyl terminal amino acids and incorporation of only five novel amino acids, rendering the protein nonfunctional ( Figure 3B). Sanger sequencing was performed to validate the WGS annotation of the c.95-108G > A variant in the proband (primer sequence in Table 3). Strikingly, this sequencing revealed homozygosity of the c.95-108G > A variant in the proband, further suggesting its disease association ( Figure 3C). Following this finding, blood samples from both parents of the proband were obtained. Echocardiography did not suggest any cardiac pathology in either of these parents (data not shown). DNA was extracted and PCR performed for Sanger sequencing of the TNNT2 variant. Strikingly, both parents harbored a single copy of the c.95-108G > A variant ( Figure 3C). This finding led to investigations into the pedigree of the proband. It was found that the proband was born from consanguineous breeding (Figure 4). Taken together, these results implicate a novel intronic mutation which results in the truncation of the sarcomeric protein, cardiac troponin T, as the cause of cardiomyopathy in this Maine Coon cat. FIGURE 4 | Family pedigree of affected proband. Circles and squares represent males and females, respectively, while blue and red represent unaffected and affected individuals, respectively. A male Maine Coon, heterozygous for the TNNT2 c.95-108G > A variant, was mated to a female Maine Coon, which resulted in one female offspring. This female offspring, also heterozygous, was then bred with the original male, producing two offspring. The male offspring was the cardiomyopathic proband in this study and was homozygous for the TNNT2 c.95-108G > A variant. * indicates that samples from these cats were not available at the time of the study.

DISCUSSION
In this study, we identified a novel homozygous intronic variant in TNNT2 which was associated with a case of feline cardiomyopathy and early heart failure. Mutations in MYBPC3 and MYH7 have previously been described. However, to the best of our knowledge, this is the first report of a thin filament mutation associated with feline cardiomyopathy (Meurs et al., 2005;Schipper et al., 2019). Mutations in TNNT2 have been strongly implicated in the development of HCM and dilated cardiomyopathy (Watkins et al., 1995;Pasquale et al., 2012). In humans, these TNNT2 mutations are described as phenotypically mild compared to MYH7 mutations, but with a higher incidence of sudden cardiac death (Watkins et al., 1995). Importantly, mutations in intronic splice sites have been identified as causative of cardiomyopathy in humans (Thierfelder et al., 1994). Taken together, these data support our finding that the TNNT2 intronic mutation we describe is the most likely cause of this case of feline cardiomyopathy.
The TNNT2 gene encodes the cardiac-specific sarcomeric protein cardiac troponin-T (cTnT) (Wei and Jin, 2016). Cardiac troponin-C and -I (cTnC and cTnI, respectively), together with cTnT, make up the troponin complex of the heart (Gomes et al., 2002). The troponin complex is associated with the thin filaments, and cTnT is understood to play an important role in anchoring the complex to both actin and tropomyosin (Gomes et al., 2002). When calcium is released from the sarcoplasmic reticulum, it binds to and induces a structural change in cTnC (Vinogradova et al., 2005), ultimately leading to the movement of tropomyosin, allowing the formation of cross-bridges (Boussouf et al., 2007). cTnT has been demonstrated to regulate the calcium sensitivity of actomyosin ATPase and force (Gomes et al., 2005). Thus, cTnT has a central role in regulating contraction and relaxation of the heart.
Homozygous knockout of the TNNT2 gene in mice is embryonically lethal (Ahmad et al., 2008;Nishii et al., 2008). As such, it is unlikely that the de novo splice acceptor site in the mutant TNNT2 is fully penetrant. Rather, it is more likely that the aberrant splicing of TNNT2 is a limited event, resulting in haploinsufficiency of the cTnT protein. This variant results in the significant truncation of the cTnT protein. It is well known that both the middle carboxyl-terminal regions of cTnT are required for its interaction with tropomyosin, cTnI, and cTnC (Wei and Jin, 2016). Therefore, the loss of these regions would likely prevent the incorporation of the truncated protein into the sarcomere. The resultant reduction in cTnT levels would result in an abnormal stoichiometry of thin filament proteins in the sarcomere, which would be sufficient to cause cardiomyopathy. Indeed, haploinsufficiency of cTnT has previously been associated with cardiomyopathy (Bollen et al., 2017). Furthermore, the incorporation of less than 5% of C-terminally truncated cTnT is sufficient to cause cardiomyopathy in mice (Tardiff et al., 1998). Since the proband's symptoms are currently managed via treatment, consisting of Furosemide, Spironolactone, Benazepril and Clopidogrel, we cannot confirm the exact level of haploinsufficiency. Unfortunately, DNA from the proband's littermate was not available, but we predict it would be either negative or heterozygous for the TNNT2 c.95-108G > A variant. While no variants were identified in MYH6 or MYH7, two single nucleotide variants were also identified in MYBPC3 that were classified as having moderate risk. These variants were a proline substituted to a leucine at position 922 (P922L) and an alanine for a threonine at amino acid 1037 (A1037T), occurring within domains C7 and C8 of cMyBP-C, respectively. However, further investigation of these variants revealed that these amino acids are not evolutionarily conserved, and, furthermore, the substituted amino acids matched those of either human or mouse, effectively ruling out these MYBPC3 variants as disease-causing.

CONCLUSION
We have identified a novel, homozygous mutation in the sarcomeric gene TNNT2, which is associated with cardiomyopathy in a Maine Coon cat. The homozygosity of this mutation resulted from inbreeding. This study is the first to describe a mutation in TNNT2 that is associated with feline cardiomyopathy, suggesting that this gene should be included in routine genetic testing in felines. Additionally, breeders should be mindful of the dangers associated with close generational consanguineous breeding.

DATA AVAILABILITY STATEMENT
The sequencing data has been deposited into the BioProject database (accession: PRJNA671288).

ETHICS STATEMENT
The animal study was reviewed and approved by Ref: AM07-19-10-03-01 Institutional Animal Care and Use Committee OLAW Assurance D16-00190, AAALAC 00278 University of Cincinnati PO Box 210572, Cincinnati, OH 45267-0572. Written informed consent was obtained from the owner for the participation of their animals in this study.

AUTHOR CONTRIBUTIONS
JM and SS designed the research and wrote the manuscript. JM and MS performed the research. JM, MS, and SS analyzed data. MS and RB edited the manuscript. RB provided project oversight for clinical data. All authors read and approved the final manuscript.

FUNDING
Whole genome sequencing of the proband was funded by a donation from Kathleen and Michael Janson, owners of the proband. JM was supported by an American Heart Association Postdoctoral Fellowship (17POST33630095). SS received funding from the following grants: National Institutes of Health grants R01 AR078001, R01 HL130356, R56 HL139680, R01 AR067279, R01 HL105826, and R01 HL143490; and transformation (19TPA34830084) awards; and MyoKardia, AstraZeneca, Merck, and Amgen.