Identification of a nonsense mutation in TNNI3K associated with cardiac conduction disease

Abstract Background Cardiac conduction disease (CCD) is a common cardiovascular disease which can lead to life‐threatening conditions. The importance of heredity in CCD has been realized in recent years. Several causal genes have been found to be implicated in CCD such as SCN5A, TRPM4, SCN1B, TNNI3K, LMNA, and NKX2.5. To date, only four genetic mutations in TNNI3K have been identified related to CCD. Methods Whole‐exome sequencing (WES) was carried out in order to identify the underlying disease‐causing mutation in a Chinese family with CCD. The potential mutations were confirmed by Sanger sequencing. Real‐time qPCR was used to detect the level of TNNI3K mRNA expression. Results A nonsense mutation in TNNI3K (NM_015978.2: g.170891C > T, c.1441C > T) was identified in this family and validated by Sanger sequencing. Real‐time qPCR confirmed that the level of TNNI3K mRNA expression was decreased compared with the controls. Conclusions This study found the first nonsense TNNI3K mutation associated with CCD in a Chinese family. TNNI3K harboring the mutation (c.1441C > T) implicated a loss‐of‐function pathogenic mechanism with an autosomal dominant inheritance pattern. This research enriches the phenotypic spectrum of TNNI3K mutations, casting a new light upon the genotype‐phenotype correlations between TNNI3K mutations and CCD and indicating the importance of TNNI3K screening in CCD patients.

Interaction partners of TNNI3K embody cardiac troponin I (cTnI), anti-oxidant protein 1 (AOP-1), and p38, which make TNNI3K an important factor in cardiovascular diseases. 7,8 TNNI3K was related to heart failure and hypertrophy, ischemia/reperfusion injury, cardiac conduction, and heart regeneration. [9][10][11][12][13][14][15] Rare mutations related to cardiovascular diseases have been identified in more than 100 genes which encode proteins belonging to different cellular structures and pathways such as cytoskeleton, sarcomere, nuclear membrane, ion channel, mitochondria, sarcoplasmic reticulum, and desmosomes. To date, there are only four mutations in TNNI3K that have been found to be relevant with cardiovascular diseases, including three missense mutations and one splicing mutation. [16][17][18][19] Here, we found that a nonsense variant in TNNI3K (NM_015978.2:c.1441C > T), which was not seen in the human gene mutation database (HGMD), may be the cause of cardiac abnormalities.

| Patients enrollment and ethical approval of the study
A Chinese family with cardiovascular diseases was enrolled in our studies. Parents of the proband are not related biologically. The pedigree of the family is shown in Figure 1A. Written informed consent was obtained from each individual, and the investigation was approved by the Ethics Committee of The Second Xiangya Hospital of Central South University.

| DNA extraction and whole-exome sequencing
DNA was extracted from blood samples of the proband (III-10) and family members (II-7, II-9, III-12, and IV-1 in ExAC, ESP6500,1000 Genome Project, and gnomAD) in general population were eliminated from raw data. SIFT, PolyPhen-2, MutationTaster, and PROVEAN were used for predicting pathogenicity of mutations. Interpretation of mutations pathogenicity was guided by American College of Medical Genetics and Genomics guideline (ACMG). 20 The filter strategies are shown in Figure 1B and Table 1. The validation of potential mutations in the proband and his family members was done by means of Sanger sequencing. Primer 3 was used to design the primer pairs (primers Tm and sequences that have been used in co-segregation were showed in Table. S1). ABI 3730 sequencer (Applied Biosystems) was used for Sanger sequencing.  available). Owing to the implantation of PM, cardiac magnetic resonance is not performed in this family (Table 2 and Figure S1A-D).

| Genetic testing
A nonsense mutation in TNNI3K (c.1441C > T, p.R481X) was found through WES and confirmed by Sanger sequencing (Figure 1E-F and Figure S1E-F). This variant produces a truncated protein with 480 amino acids which is significantly shorter than the wildtype protein (835 amino acids). It is a rare mutation with MAF of 0.0017%(2/119514 alleles) in ExAC database. In addition, this mutation was not seen in our 200 local control cohorts. 21 The c.1441C > T mutation in the TNNI3K gene co-segregated with the clinical phenotype in this family. Thus, we believed that this mutation seems to be the causative mutation of cardiac abnormalities in this family.

| Real-time qPCR analysis
The nonsense mutation results in the early appearance of stop codon which is anticipated to activate a process called nonsense-mediated mRNA decay (NMD). 22 We isolated the mRNA from the mononu-

clear cells in the patients and healthy family members. Real-time
qPCR regarded the healthy control levels of mRNA in TNNI3K as "1." The results revealed a decreased level of TNNI3K mRNA expression in patients compared with the controls (P < .0001; Figure 1G).

| D ISCUSS I ON
Herein, using WES we have identified a nonsense mutation of the  16 Xi, et al (2015) 17 Fan, et al (2018) 18 Podliesna, et al (2019) 19 Podliesna, et al (2019) 19 Podliesna, Nonsense mutation has larger effects in protein function in comparison with missense mutation. Thus, it is anticipated to have more serious phenotype than missense mutation correspondingly.
The clinical presentation of cases with TNNI3K mutation has been summarized in Table 3. Conduction abnormalities of TNNI3K mutation carriers in our case seem to be more serious than previously reported cases based on the fact that all of them had PMs implanted.
Cardiac structures were all normal in them except for II-5 and II-7 whose heart was enlarged in follow-up period. Based on the fact that However, there are several limitations in our research. This study lacks direct evidence in elucidating the underlying pathways between TNNI3K and CCD. An animal model harboring specific mutation will be needed to elucidate the underlying signaling pathways.
In conclusion, our study successfully identified the first nonsense mutation in TNNI3K (c.1441C > T) that was associated with CCD, paying the way for genetic diagnosis for CCD. Considering the obscure pathophysiological mechanisms and complicated variant spectrum of CCD, we suggest that WES could be applied in potential inherited patients to identify novel mutations and improve our understanding of CCD etiology.

ACK N OWLED G M ENTS
The authors thank all study participants.