Pedigree structures for kindreds DC-12 (A) and DC-35 (B) are shown. Square = male; circle = female; solid = affected; open = unaffected; gray = clinical status unknown; parallel diagonal lines = suspected dilated cardiomyopathy (DCM) on the basis of family history; slash through the symbol = deceased, with cause of/age at death indicated. The gene for ribonucleic acid binding motif protein 20 (RBM20) is located at chromosome 10q25.2. Markers that were tested for this region of chromosome 10 are listed in order from centromere to q-telomere, with map locations according to the National Center for Biotechnology Information website and given in megabases and centimorgans. The haplotypes for these markers are shown in columns beneath family members who underwent genetic evaluation; the disease-associated haplotypes are boxed. Question marks indicate genotypes that could not be scored from paraffin-embedded samples. Recombination events account for inheritance of portions of the disease haplotype in some affected subjects (DC-12: III.2, III.3, III.11, III.15, III.16, IV.1; DC-35: III.6, IV.1, IV.5, IV.9, IV.10) and enable critical regions to be defined by the minimal region of overlap. The RBM20 missense mutations (RBM20 mut), which cosegregate with DCM, are indicated by plus symbols; minus symbols indicate wild-type sequence. *Proband. ALZ = Alzheimer's disease; CA = cancer; CHF = congestive heart failure; CVA = cerebrovascular accident; MI = myocardial infarction; MS = multiple sclerosis; MVA = motor vehicle accident; Pn = pneumonia; SD = sudden death; Tx = cardiac transplantation.

Diamonds = 2 or more family members of both sexes; parentheses = inferred RBM20 mutation status. Symbols and abbreviations as in Figure 1.

(A) Genomic deoxyribonucleic acid (DNA) mutation scans with denaturing high-performance liquid chromatography revealed abnormal heteroduplex profiles (red), compared with the control wild-type profile (gray), in exon 9 of RBM20 in 8 dilated cardiomyopathy families. Abnormal profiles indicated DNA sequence alterations, which were absent in 480 control samples. Below the heteroduplex profiles, DNA sequencing revealed corresponding heterozygous missense mutations. The location of each mutation and its resultant amino acid substitution are based on predicted reference RBM20 complementary deoxyribonucleic acid (cDNA) and protein sequences and are indicated below each chromatogram. Mutation c.1906 C>A, R636S was shared by 3 families, and c.1913 C>T, P638L was shared by 2 families. (B) The predicted genomic structure of RBM20, consisting of 14 exons, is depicted to scale. Exons that encode peptides homologous to highly conserved functional domains—ribonucleic acid (RNA) recognition motif 1 (RRM-1, in purple), arginine/serine-rich region (RS-rich, in blue), and U1 zinc finger (zf-U1, in green)—are indicated. Putative start (ATG) and stop (TGA) codons are located in exons 1 and 14, respectively. A polyadenylation signal (AATAAA) is located at the 3' end of exon 14. Directly below the RBM20 genomic structure, cDNA amplification and sequencing confirmed transcription of messenger RNA from exons 2 to 14 in human heart tissue, as depicted by parallel alignment. The cDNA transcript contains the complete RS domain and identified RBM20 mutations in the 5' region of exon 9. (C) Alignment of homologous RBM20 protein sequences that flank the amino acid substitutions is shown. The RS domain spans residues 632 to 654, with arginine (R) and serine (S) residues highlighted in yellow. Residues conserved between human RBM20 and another species are indicated by (•) and amino acid deletions by (–). Amino acids that are altered by the identified RBM20 missense mutations (residues 634, 636, 637, and 638, indicated with red bars) are conserved among all 8 species. Accession numbers: NP_001127835.1 for human, XP_508032 for chimpanzee, XP_544017 for dog, XP_603772 for cow, BAE24961 for mouse, NP_001101081 for rat, XP_421755 for chicken, XP_683222 for zebrafish, and CAG01297 for pufferfish.