*2.1. TTN Truncation Variations in Highly Constitutively Expressed Exons Cause DCM*

*TTN* mutations are the most common cause of genetic DCM due in part to *TTN's* large gene size as well as its essential role in the sarcomere and cardiomyocyte function. There are numerous point mutations of *TTN* leading to missense changes in single amino acids that are associated with diverse phenotypes including several types of cardiomyopathies [19,20]. However, pathogenic *TTN* mutations most commonly result from nonsense mutations leading to frameshifts and incorporation of early stop codons predicting a truncated, shortened titin protein, termed titin truncation variations (TTNtv) [21]. Interestingly, TTNtv, mostly restricted to the I-band, are found in 2–3% of the general population [13,22,23] and there is substantial interest in determining the associated impact of TTNtvs in ostensibly healthy persons. While phenotypes of TTNtvs can be diverse and may present with restrictive cardiomyopathy, they are mostly associated with DCM that presents later in life [24,25]. There is a relationship between mutation location and pathogenicity of TTNtv phenotype. Given the size of *TTN*, there are many alternative splicing events during transcription and hundreds of unique isoforms expressed in cardiac tissue. The key predictor of pathogenicity of TTNtv is whether the mutated exon is expressed in a high proportion of the total cardiac-expressed isoforms. In a population-based study with genome sequencing, RNA sequencing, and cardiac phenotyping, pathogenicity of a given TTNtv was correlated with how frequently the mutated exon is expressed. This was assigned a value, proportion spliced in (PSI) from 0 to 1, with a PSI > 0.9 imparting a 93% probability of pathogenicity [23]. The association of PSI and phenotype is further supported by pathogenicity data from the Clinvar (https://www.ncbi.nlm.nih.gov/clinvar/) database where exon locations of high PSI mutations are correlated with known pathogenic mutations [17]. This is an essential finding because *TTN* mutations are common, therefore predicting whether a given cardiac phenotype is attributable to a TTNtv mutation or an alternative cause is helpful for clinical prognostication. The PSI score as a predictor of pathogenicity fits with previous observations that A-band mutations are more pathogenic, as A-band exons are more constitutively expressed. As a corollary, Z-disk mutations are less pathogenic, likely to be due to lower constitutive expression of Z-disk exons. In addition, there is a common TTN isoform, cronos, that has an alternative downstream translational start site that can compensate for upstream truncation mutations [26]. Recently there have examples of TTNtv with high PSI without cardiac phenotypes. There is evidence that in these circumstances, exons with TTNtv can be translated either through utilization of internal ribosomal entry site or stop codon suppression [27].
