KLF13 Loss-of-Function Mutations Underlying Familial Dilated Cardiomyopathy

Abstract

Background Dilated cardiomyopathy (DCM), characterized by progressive left ventricular enlargement and systolic dysfunction, is the most common type of cardiomyopathy and a leading cause of heart failure and cardiac death. Accumulating evidence underscores the critical role of genetic defects in the pathogenesis of DCM, and >250 genes have been implicated in DCM to date. However, DCM is of substantial genetic heterogeneity, and the genetic basis underpinning DCM remains elusive in most cases. Methods and Results By genome-wide scan with microsatellite markers and genetic linkage analysis in a 4-generation family inflicted with autosomal-dominant DCM, a new locus for DCM was mapped on chromosome 15q13.1-q13.3, a 4.77-cM (≈3.43 Mbp) interval between markers D15S1019 and D15S1010, with the largest 2-point logarithm of odds score of 5.1175 for the marker D15S165 at recombination fraction (θ)=0.00. Whole-exome sequencing analyses revealed that within the mapping chromosomal region, only the mutation in the KLF13 gene, c.430G>T (p.E144X), cosegregated with DCM in the family. In addition, sequencing analyses of KLF13 in another cohort of 266 unrelated patients with DCM and their available family members unveiled 2 new mutations, c.580G>T (p.E194X) and c.595T>C (p.C199R), which cosegregated with DCM in 2 families, respectively. The 3 mutations were absent from 418 healthy subjects. Functional assays demonstrated that the 3 mutants had no transactivation on the target genes ACTC1 and MYH7 (2 genes causally linked to DCM), alone or together with GATA4 (another gene contributing to DCM), and a diminished ability to bind the promoters of ACTC1 and MYH7. Add, the E144X-mutant KLF13 showed a defect in intracellular distribution. Conclusions This investigation indicates KLF13 as a new gene predisposing to DCM, which adds novel insight to the molecular pathogenesis underlying DCM, implying potential implications for prenatal prevention and precision treatment of DCM in a subset of patients.

Keywords: KLF13; biological assay; dilated cardiomyopathy; functional genomics; molecular genetics; transcriptional factor.

Figure 1. Pedigree and haplotypes for Family 1 affected with dilated cardiomyopathy.

The 4‐generation family with a high incidence of autosomal‐dominant dilated cardiomyopathy was designated as Family 1, with members identified by generations‐numbers (Roman‐Arabic numerals) shown below member symbols. A vertical bar beneath a member denotes the chromosomal region defined by genetic linkage analysis. Polymorphic markers spanning the linkage region on chromosome 15q13.1–q13.3 were displayed to the left of the pedigree, with members' genotypes (represented by numbers) for markers given next to the chromosome bars. The filled bars mean disease haplotype. Key recombination events occurred with marker D15S1019 in member III‐16, and with marker D15S1010 in member III‐3. “+” indicates a carrier of the heterozygous mutation of c.430G>T in the KLF13 gene; an “–” indicates a noncarrier.

Figure 2. New KLF13 mutations responsible for dilated cardiomyopathy.

Sequencing electropherograms showing 2 nonsense mutations and 1 missense mutation in the KLF13 gene as well as their corresponding wild‐type sequences were exhibited. Arrows direct to the heterozygous mutations or homozygous wild‐type sequences. A, Sequencing electropherograms of the affected proband and an unaffected member from Family 1, showing the heterozygous KLF13 mutation c.430G>T and its wild‐type control. B, Sequencing electropherograms of the affected proband and an unaffected member from Family 2, showing the heterozygous KLF13 mutation c.580G>T and its wild‐type control. C, Sequencing electropherograms of the affected proband and an unaffected member from Family 3, showing the heterozygous KLF13 mutation c.595T>C and its wild‐type control.

Figure 3. Pedigrees of Family 2 and Family 3 inflicted with dilated cardiomyopathy.

The 2 families with a high incidence of dilated cardiomyopathy were designated arbitrarily as Family 2 and Family 3, respectively, with members recognized by generations‐numbers. “+” a carrier of the heterozygous KLF13 mutation c.580G>T in Family 2 or c.595T>C in Family 3; “–”, a noncarrier.

Figure 4. Decreased transcriptional activity of the mutant KLF13 proteins.

A, In cultivated HeLa cells, WT KLF13 showed dose‐dependent transactivation of the ACTC1 promoter. B, Activation of the ACTC1 promoter‐driven firefly luciferase in Hela cells by WT, E144X, E194X, or C199R mutant, alone or together, showed significantly decreased transactivation by the mutant KLF13 proteins. C, WT KLF13 displayed dose‐dependent transcriptional activation of the MYH7 promoter in cultured Hela cells. D, Activation of the MYH7 promoter‐driven luciferase in Hela cells by WT or E144X, E194X, C199R, separately or in combination, showed significantly reduced transactivation by the mutants. All the experiments were performed in triplicate, with the results shown as mean±SD. Here, each * indicates P<0.005, when compared with the same amounts of WT. WT indicates wild type.

Figure 5. Disrupted synergistic transactivation between mutant KLF13 and GATA4.

A, In cultured Hela cells, the transactivation of the ACTC1 promoter by GATA4 in synergy with E144X, E194X or C199R mutant was abolished by each mutation when compared with that by GATA4 and WT KLF13. B, In the presence of GATA4, transactivation of the MYH7 promoter‐driven luciferase in Hela cells by WT, E144X, E194X, or C199R mutant revealed that the mutations ablated the synergistic transcriptional activation between each mutant and GATA4. All the experiments were performed in triplicate, with the results expressed as mean±SD. Here, each * indicates P<0.001, in comparison with its WT counterpart. WT indicates wild type.

Figure 6. Diminished DNA‐binding ability of the mutant KLF13 proteins.

A, The ability of WT KLF13, E144X, E194X or C199R mutant to bind to the ACTC1‐DNA probe. B, The ability of wild‐type KLF13 (WT), E144X, E194X, or C199R mutant to bind to the MYH7‐DNA probe. Electrophoretic mobility shift assay indicated that WT bound specifically to the ACTC1‐DNA probe (A) or the MYH7‐DNA probe (B), whereas all the mutants showed diminished DNA‐binding affinity to the ACTC1‐DNA (A) or MYH7‐DNA probe (B). WT indicates wild type.

Figure 7. Distinct subcellular distribution of the mutant KLF13 proteins.

Subcellular localization of WT KLF13, E144X, E194X, and C199R mutant in Hela cells were detected by immunofluorescence. The red color shows the Alexa‐Fluor 594 conjugated secondary antibodies against the anti‐KLF13 antibody and the nuclei were stained with 4′,6‐diamidino‐2‐phenylindole (DAPI; blue). WT, E194X, and C199R mutant were localized exclusively to the nuclei whereas E144X mutant was distributed not only in the nuclei but also in the cytoplasm. Bar=50 μm. WT indicates wild type.