The heart is able to beat partly because of a large protein called Titin that helps to give heart muscle its elasticity. Mutations that shorten the gene that encodes Titin can cause part of the heart to become enlarged and weakened, a condition called dilated cardiomyopathy. Some people with shortened copies of this protein have a mild form of cardiomyopathy and are able to lead relatively normal lives. Others develop more severe symptoms that prevent the heart from pumping blood effectively and may even cause the individual to need a heart transplant.

Genetic studies have revealed that mutations that shorten the Titin protein by disrupting the portion of the gene corresponding to the latter two-thirds of the protein (which encodes the so-called “C-terminal” end of the protein) cause more severe symptoms than mutations that occur near the start of the gene. But it is not clear why the location of the mutation matters.

To investigate this problem, Zou et al. used a gene-editing tool called CRISPR to create genetically engineered zebrafish. These fish had mutations at one of six different points in the gene that encodes the zebrafish version of Titin. Just as with humans, mutations near the C-terminal end of the gene caused more severe muscle problems in the fish.

Specifically, Zou et al. found that the worst disease was associated with mutations that occurred at or after a “promoter” region within the gene and near this C-terminal end. Normally, the promoter produces an independent smaller form of the Titin protein, which helps to reduce the severity of muscle problems in zebrafish that have mutations near the start of the gene. However, mutations near the C-terminal end of the gene also damage this smaller form, preventing this failsafe from working, and so lead to more severe symptoms. Zou et al. also found this promoter to be active in both mouse and human hearts.

Future work will focus on learning how this smaller form of Titin works to help muscle develop and withstand stress and determine whether increasing its production can overcome the more severe forms of disease.

The use of genetics in clinical medicine depends on knowledge that an identified mutation confers risk of disease. Recent guidance issued by the American College of Medical Genetics emphasizes that the strongest form of evidence supporting causality of a mutation is whether it results in a truncation of the protein (nonsense, frameshift or canonical splice-site mutations), specifically for genes where loss-of-function mutations are known to cause disease (Richards et al., 2015). Truncating mutations within alternatively spliced exons and at the extreme C-terminus of the protein must be interpreted with caution, as they may have little to no impact on protein function. However, the guidelines do not address whether truncating mutations at different positions along the length of the protein might differ in phenotype severity or mechanism of action, a finding that could complicate interpretation of truncation mutations for a broad range of inherited diseases.

Truncations in the giant sarcomeric protein Titin (TTN) reveal patterns that are at odds with this uniform assignment of causality. TTN lies along the length of the sarcomere and is thought to mediate passive tension of the muscle fiber, while also serving as a binding scaffold for a large number of proteins, including actin and myosin (Lewinter and Granzier, 2013). Individual regions of TTN are annotated according to their localization to the Z-disc, I-band, A-band, and M-line regions of the sarcomere visualized on immuno-electron micrographs Fürst et al., 1988. Truncating and/or missense mutations in TTN result in cardiac (dilated cardiomyopathy [DCM]) and skeletal muscle (myofibrillar myopathy) disease (Chauveau et al., 2014). Recent estimates suggest that truncation mutations in TTN may explain as much as 25% of DCM (Herman et al., 2012), a disease which affects upwards of 1 in 2500 individuals (Codd et al., 1989). Although truncations are found along the length of TTN in DCM patients, in patients with the most severe forms of DCM, truncations reside within the C-terminal two-thirds of the protein (from amino acid 14,760 onwards, Figure 1A), specifically in the distal I-band and A-band (we will refer to these as C-terminal truncations) (Herman et al., 2012; Roberts et al., 2015). Multiple TTN exons are alternatively spliced (Bang et al., 2001), with many having little to no inclusion within cardiac isoforms, and splicing has been proposed to at least partially explain the exclusion of N-terminal mutations in patients with end-stage DCM (Herman et al., 2012). Nonetheless, over 5500 amino acids in the N-terminal region of the protein are constitutive (percent spliced in or PSI >0.9, Figure 1B), implying the importance of other mechanisms.

Figure 1 with 1 supplement see all

Download asset Open asset

We sought to address the basis of this variation of severity with mutation position. For a disease such as DCM, in which inheritance is most commonly observed as adult-onset autosomal dominant with variable penetrance, deciphering such mechanisms in mouse models can be challenging, as heterozygotes tend to show minimal phenotypes and homozygotes tend to be embryonic lethal (Gramlich et al., 2009). We selected the vertebrate zebrafish model for several reasons. Firstly, zebrafish has high sequence conservation of sarcomeric proteins with mammals and recapitulate disease phenotypes with loss-of-function of multiple established Mendelian cardiomyopathy genes (Dahme et al., 2009). In fact, forward genetic screens have generated zebrafish Titin mutants, although in only one case is the position of the mutation known (Xu et al., 2002; Steffen et al., 2007; Myhre et al., 2014). Secondly, zebrafish can live for up to two weeks with a non-contractile heart, thus allowing mechanistic studies of homozygote mutants. Thirdly, the relatively low cost and large clutch size of zebrafish allow the ability to generate multiple mutant lines and provide statistical power for robust conclusions.

As a result of an ancestral genome duplication event, zebrafish include two titin genes: ttna and ttnb (Xu et al., 2002; Steffen et al., 2007; Seeley et al., 2007). Ttna shares considerable similarity with the human TTN gene (Figure 1—figure supplement 1, showing ‘meta-transcripts’ with all possible exons included). The zebrafish and human proteins are similar in length (∼32,000–35,000 amino acids depending on the splice isoform) and share 56% identity and 82% homology. They have a nearly identical domain organization, with three immunoglobulin-like of domains (proximal, mid, and distal), an elastic PEVK region, N2A and N2B domains which distinguish cardiac and skeletal isoforms, a fibronectin-immunoglobulin repeat region that mediates binding to myosin, and a C-terminal kinase domain. In both species, alternative splicing targets the middle Ig-like and PEVK domains, which modulate passive tensile properties of the sarcomere (Granzier and Labeit, 2004). Cardiac isoforms typically include much of this elastic region as well as the N2B ± N2A domains, while skeletal isoforms include the N2A domain but exclude N2B as well as much of the elastic region. Multiple mutants isolated from forward genetic screens coupled with morpholino knockdown analysis have demonstrated that ttna is essential for cardiac sarcomere formation in zebrafish (Xu et al., 2002; Myhre et al., 2014; Seeley et al., 2007).

Although the paralagous zebrafish ttnb gene is dispensable for cardiac sarcomere development (Steffen et al., 2007; Seeley et al., 2007), it plays an essential role in skeletal muscle sarcomerogenesis. It is expressed at comparable levels to ttna in skeletal muscle (as opposed to ∼2/3 of ttna abundance in heart). In contrast to human TTN or zebrafish ttna, the ttnb protein has abbreviated versions of the middle Ig-like and PEVK elastic domains (Figure 1—figure supplement 1) and thus more closely resembles skeletal rather than cardiac muscle isoforms of ttna. Morpholino-based skipping of the N2A exon of either ttna or ttnb results in severe disruption of skeletal muscle sarcomeric architecture, highlighting the mutual contribution of the two genes (Seeley et al., 2007).

Using CRISPR/Cas9 technology (Jinek et al., 2012), we engineered six truncating mutations within the major cardiac TTN orthologue, ttna, (Figure 1A, Table 1), so that the resulting protein product would be truncated at either the Z-disk (n1), proximal or mid I-band (n2 and n3) or proximal, mid, or distal A-band (c1, c2, and c3). From wild-type (WT) zebrafish embryo transcriptome data, we anticipated that all exons would be constitutive in both cardiac and skeletal muscle (PSI >0.99, Table 1). Mutant lines were bred for two generations to minimize any off-target effects from the genome editing, and we studied the F3 generation for cardiac and skeletal muscle consequences.

As expected (Xu et al., 2002; Gramlich et al., 2015), ttna truncations showed no gross phenotype in heterozygotes, with all heterozygotes attaining sexual maturity. We thus focused on homozygote mutants, with the expectation that partial rescue from ttnb could potentially elicit phenotypic differences among mutants, especially in skeletal muscle where it is expressed at comparable levels to ttna and has a documented essential role. N-terminal (ttna^n1/n1, ttna^n2/n2, ttna^n3/n3, collectively referred to as ttna^n/n) and C-terminal (ttna^c/c) truncations all showed severe deficits in cardiac contractility (Video 1) and typically died within two weeks of birth. Immunostaining revealed severe cardiac sarcomeric disarray (Figure 2A).

Figure 2

Download asset Open asset

Video 1

Download asset

In contrast to their cardiac phenotypes, ttna mutants demonstrated a sharp division in development of skeletal muscle disease. All three ttna^n/n mutants were capable of grossly normal motion while the ttna^c/c truncations were entirely unable to move (Video 2). Immunostaining supported the distinction, with ttna^c/c mutants having highly distorted sarcomeric architecture while ttna^n/n architecture closely resembled wild-type (Figure 2B). To confirm transcription of all mutant isoforms, assess nonsense-mediated decay, and rule out selective alternative splicing of ttna^n/n mutant exons, we performed RNA-Seq of heart (cardiac muscle) and trunk (skeletal muscle) of 3 day old embryos, as well as adult fish. All mutated exons appeared constitutive in both heart and skeletal muscle, across all genotypes (Figure 2C). We next examined the extent of nonsense-mediated decay (NMD) for each mutant allele by performing targeted RNA sequencing of heterozygote embryos and comparing the number of reads for wild-type and mutant exons. In keeping with typical estimates for NMD efficiency (Zetoune et al., 2008), we saw ∼75–80% of mutant transcript being degraded (Figure 2D). However, no substantial variation was seen across the mutants.

Video 2

Download asset

Given that variability in alternative splicing or NMD could not explain the selective skeletal muscle disarray in ttna^c/c mutants, we hypothesized that this variation in severity could be due to ttna^c/c mutants interfering with the action of ttnb, although the ratio of mutant to wild-type transcript of ∼20% made this an unlikely scenario. Knockdown of ttnb by morpholino disrupted ttna^n/n skeletal muscle morphology indicating that the intact skeletal muscle sarcomeric architecture in ttna^n/n mutants arose at least in part from rescue by ttnb protein (Figure 3A), although the phenotype was not nearly as severe as that of ttna^c/c mutants. To investigate a dominant negative mechanism for ttna^c/c mutants, we performed a morpholino knockdown of ttna, targeting a splice site of an N-terminal constitutive exon (exon 4), which would be expected to introduce an N-terminal frameshift on top of the same C-terminal mutant allele. If the differences between ttna^n/n and ttna^c/c mutants arose simply from inhibitory effects of the ttna^c/c truncated product, then introducing an upstream frameshift should disrupt production of the dominant negative protein product even further and restore sarcomeric architecture. Knockdown by morpholino was effective (∼78% knockdown efficiency, data not shown) and the combined action of NMD and morpholino knockdown would be expected to reduce mutant ttna protein to only ∼5% of the level of ttnb protein. Nonetheless we were unable to rescue skeletal muscle architecture in ttna^c/c mutants (Figure 3B). We thus concluded that ttna^c/c mutants do not act as dominant negatives and that an alternative explanation was needed.

Figure 3

Download asset Open asset

Another possible explanation for the discrepancy in disease severity between ttna^n/n and ttna^c/c mutants could be an internal promoter that produces a C-terminal isoform, which could partially rescue N-terminal truncations, especially in conjunction with expression of ttnb. Even though N- and C-terminal truncating mutations would both disrupt full-length ttna, the latter would also disrupt a C-terminal isoform while the former would not. Interestingly invertebrates, including C. elegans and D. melanogaster, have distinct genes that encode for protein products corresponding to the N-terminal (Z-disc and proximal I-band) or C-terminal (distal I-band and A-band and M-line) portions of Titin (Bullard et al., 2002). The Novex-3 isoform in humans (discussed below) does in fact include only the Z-disc and proximal I-band portions of TTN, but no isoform corresponding only to the C-terminal portion of TTN has ever been reported in vertebrates.

Using the exon boundaries defined by ttna^n3/n3 and ttna^c1/c1, we examined zebrafish and mouse cardiac and skeletal muscle transcriptomic data for patterns of reads indicating a promoter upstream of one of the ttna exons. We noticed an accumulation of reads 5’ to exon 116 (e116, ENSDART00000109099) in zebrafish skeletal muscle (Figure 4A) and in the corresponding exon of mouse heart (e223, ENSMUST00000099981, Figure 4A). We hypothesized that these reads arose from transcription from an alternative promoter within the e115-e116 intron. 5’-RACE confirmed an alternative transcription start site (TSS), 110 bp upstream of e116 (within the e115-e116 intron) in zebrafish skeletal muscle, with an in-frame initiator methionine located 7 amino acids upstream of e116. To establish conservation in mammals, we repeated 5’-RACE on adult mouse and fetal human cardiac tissue and found a TSS in the upstream intron (257bp upstream of e223 in mouse; 256bp upstream of the orthologous exon, e240, in humans), with a putative initiator methionine 12 amino acids upstream of the relevant exon. Using PCR, we confirmed the presence of a multi-exonic transcript that includes this alternative promoter site in zebrafish, mouse, and human cardiac tissue (Figure 4B).

Figure 4 with 2 supplements see all

Download asset Open asset

We next consulted publicly available chromatin accessibility and activation data to find supporting evidence for an internal TTN promoter at this site. DNAse-Seq is a powerful tool to identify regions of accessible chromatin, which typically correspond to promoter or enhancer regions (Neph et al., 2013). Examination of multiple fetal human cardiac and skeletal muscle DNA-Seq datasets revealed a prominent peak overlying e240, the orthologous exon to e116 in humans (Figure 4C, Figure 4—figure supplement 1). A similar peak was found in H3K4me3 Chip-Seq data, a mark for active promoters (Zhou et al., 2010), from skeletal muscle and cardiac tissue (Figure 4C, Figure 4—figure supplement 1). Unbiased epigenetic data thus provides additional strong support for an internal promoter. For convenience, and as a counterpart to the N-terminal Novex-3 isoform, we hereafter refer to this C-terminal Titin isoform as the Cronos isoform.

If a protein product arising from the ttna internal promoter ameliorates disease severity in ttna^n/n but not ttna^c/c mutants, we should be able to pinpoint the phenotypic transition at e116 – i.e. truncation mutants N-terminal to e116 should have mild phenotypes while those at or C-terminal to e116 should have severe phenotypes. Interestingly, the corresponding human genomic position (the e239-e240 intron in ENST00000589042) represents the exact transition point for where disease severity markedly increases in DCM TTN truncation patients (Figure 1A). We used high dose guide RNA (gRNA) and Cas9 protein injections to exons 113, 114, 115, 116, and 122 and examined cardiac and skeletal muscle architecture and function in the resulting fish, focusing on embryos with nearly complete cessation of cardiac beating, as these would be expected to have a high likelihood of homozygous genomic disruption (given that only homozygote and not heterozygote ttna^n/n or ttna^c/c mutants have a cardiac phenotype). As expected, e113, e114 and e115 F0 mutants showed intact skeletal muscle architecture while e116 and e122 mutants had severe skeletal muscle disruption (Figure 4D, Figure 4—figure supplement 2). To confirm that the Cronos isoform protein product was actually translated, we used agarose protein electrophoresis and found that a higher mobility (smaller size) protein product was indeed found in wild-type and ttna^n/n mutants but not in ttna^c/c mutants (Figure 5A). Finally, a translation start site morpholino targeted upstream of the actual alternative initiator methionine (and 38 nucleotides upstream of the intron-exon junction) was sufficient to completely impair motility and disrupt skeletal muscle architecture in the ttna^n3/n3 mutant (Figure 4E, Video 3), further supporting the hypothesis that the Cronos product can rescue skeletal muscle architecture in ttna^n/n mutants. No obvious impact of the morpholino on alternative splicing of neighboring exons was seen (data not shown).

Figure 5

Download asset Open asset

Video 3

Download asset

Given that NMD substantially reduces the levels of ttna transcripts with premature termination codons (Figure 2D), we reasoned that we should be able to distinguish N- and C-terminal truncation mutants by the ratio of Cronos to full-length transcript. In ttna^n/n mutants, which would degrade full-length ttna but not Cronos, we would anticipate a higher Cronos:full-length ratio than in ttna^c/c mutants mutants, which should degrade both transcripts approximately equally. As expected, we see a much higher ratio of Cronos:full-length transcript levels in ttna^n/n mutants than in ttna^c/c mutants (Figure 5B).

We next surveyed the expression levels of Cronos in cardiac and skeletal muscle in zebrafish at different developmental stages using both in situ hybridization (Figure 6A) and real-time PCR (Figure 6B). Cronos is expressed at high levels in zebrafish skeletal muscle during development (∼2:1 ratio of Cronos to full-length ttna), but drops off sharply during adulthood (∼1:70 ratio). Although the Cronos expression in zebrafish heart also appears to be developmentally regulated, it is much lower than in skeletal muscle, with a ∼1:5 ratio at 72 hpf and 1:30 ratio in adulthood. This low cardiac Cronos expression (coupled with a minor role of ttnb in zebrafish hearts) is a likely reason for the severe and indistinguishable cardiac sarcomere disruption in ttna^n/n and ttna^c/c mutants.

Figure 6

Download asset Open asset

Given that the primary observation that motivated this work was variation in phenotypic severity in cardiomyopathy patients, we next looked to see if there was appreciable Cronos expression in mammalian hearts. We found that Cronos levels were much higher in developing mouse hearts than in zebrafish, with nearly equal transcript levels of Cronos and full-length Ttn (Figure 6C) at embryonic day 12.5. Thereafter mouse cardiac expression of Cronos decreases to ∼20–30% of full-length Titin. Skeletal muscle expression varies across development and in different muscle beds with a 1:5 ratio in mouse hindlimb at P2, and between 1:16 and ∼1:1000 ratios in adult extensor digitorum longus (fast-twitch) and soleus (slow-twitch) muscle, respectively.

Although our zebrafish work explains why a milder phenotype would be expected in individuals with truncation mutants upstream of the alternate promoter, it does not answer whether such N-terminal truncations would still result in deleterious manifestations in heterozygote patients. Specifically, an open question remains whether all TTN truncations lead to cardiac or skeletal muscle phenotypes, or conversely, do any TTN truncations lack cardiac or skeletal muscle consequences. In a disease such as dilated cardiomyopathy, with a variable age of onset, it can be challenging to derive general conclusions from observational data of control subjects, many of whom are young or middle-aged. Furthermore, self-reports of ‘health’ may not reveal limitations in an individual’s ability to attain full exercise potential, especially with aging.

To assess whether any TTN truncations are compatible with a high degree of cardiovascular fitness throughout life, we turned to a population genetics approach, and sequenced the complete TTN exome (as well as that of 104 other Mendelian cardiac disease genes, Supplementary file 1A) in 199 competitive senior athletes who entered as contestants in the Huntsman World Senior Games (median age 73, Table 2). We compared the frequency and distribution of TTN truncation mutations to those found in unselected controls and healthy volunteers from a recent sequencing study (CTL_lit) (Roberts et al., 2015). The overall rate of TTN truncations did not differ between senior athletes and CTL_lit (1.5 vs 1.2%, p = 0.91). However, the distribution of TTN truncations differed significantly between groups (Figure 7, Figure 7—figure supplement 1): TTN truncations in senior athletes mapped exclusively to the extreme C-terminal exon of the rare Novex-3 isoform (Bang et al., 2001), at a nearly 8-fold higher rate than CTL_lit (p = 0.003). This exon is found only in the Novex-3 isoform, which is expressed at low levels in the human heart (<10% of the levels of major TTN isoforms [Roberts et al., 2015]), and we would thus expect truncations in this exon would have little effect on TTN function. We next analyzed the ratio of mutants up- and downstream of Cronos in affected DCM patients, literature controls and senior athletes (Figure 7). As expected, in end-stage DCM we observed a much higher rate of mutants C-terminal to Cronos than N-terminal (30:1), followed by unselected DCM (37:11), literature controls (6:11, 2:2, 10:7 for FHS, healthy volunteers and JHS) and senior athletes (0:3).

Figure 7 with 1 supplement see all

Download asset Open asset

Loss of function experiments across a number of systems have demonstrated Titin’s importance in diverse aspects of sarcomere development and function, including the ordered assembly of sarcomeric constituents (Xu et al., 2002; van der Ven et al., 2000), cardiomyocyte systolic (Hinson et al., 2015) and diastolic function (Radke et al., 2007), and resistance to neurohormonal and vasoconstrictor stresses (Gramlich et al., 2009). Our work further contributes the discovery of a conserved internal promoter in TTN, whose protein product is essential for skeletal muscle sarcomere development in zebrafish. The Cronos promoter is prominently expressed in developing mouse heart, and its location coincides strikingly with the position of TTN truncations in patients with the most severe forms of dilated cardiomyopathy. Our working model is that disruption of both the full-length TTN and Cronos protein products results in more severe disease than disruption of the full-length product alone.

Throughout this work, we have focused on skeletal muscle architecture and function in zebrafish ttna homozygote mutants, given that this tissue demonstrates a sharp and rapidly developing phenotypic distinction between N- and C-terminal Titin truncations. The early onset of this unambiguous phenotypic difference combined with the ease of genetic and transcriptomic manipulation using CRISPR/Cas9 and morpholino technology allowed efficient mapping of the novel Cronos internal promoter as well as testing of alternative mechanistic hypotheses for this phenomenon. Nonetheless, implicit in our work is the assumption that findings from zebrafish skeletal muscle sarcomere development in ttna homozygote mutants (albeit with the paralogous ttnb present for rescue) are relevant towards understanding disease development in humans with TTN mutations. Prominent differences between these systems must be noted. With the exception of certain childhood-onset forms of skeletal and cardiac myopathy arising from homozygous TTN truncations (Carmignac et al., 2007; Ceyhan-Birsoy et al., 2013), the mode of inheritance in TTN truncation mutation patients is heterozygous with marked variable expressivity and incomplete penetrance (Herman et al., 2012; Roberts et al., 2015; Itoh-Satoh et al., 2002; Norton et al., 2013; van Spaendonck-Zwarts et al., 2014). Moreover, disease onset usually occurs in adulthood. Such an inheritance pattern is typical of many of the familial forms of cardiomyopathy, and is believed to imply additional modifying effects, potentially from diverse sources as autoimmunity and viral infection as well as contributions from additional genetic variants (Mrh et al., 2015). Nonetheless, taking all of our results together, we find it a highly plausible explanation that a superimposed deficiency in expression of the Cronos protein product would result in the more severe forms of human cardiac disease seen in DCM patients with C-terminal TTN truncations.

With these limitations noted, our results have two clinical implications for titinopathy patients. Firstly, by precisely mapping the location of the alternate promoter to the e239-e240 intron in humans, we are in a better position to give mutation-specific guidance to TTN truncation patients regarding the potential severity of their disease, with the caveat that additional factors (genetic and environmental) also modulate phenotypic severity. Along these lines, given the overlapping distribution of mutations in unselected cases and controls, it is unlikely that highly accurate patient prognoses would be realized from genotype information alone (Figure 7—figure supplement 1), although there does appear to be nearly complete separation of mutation distribution for the more extreme phenotypes (end-stage DCM and senior athletes). Secondly, our work implies that strategies that augment activity of the internal TTN promoter and levels of the Cronos isoform might provide therapeutic benefit in TTN truncation patients, although it is unclear whether there is a restricted developmental window where this might be possible.

In addition to clinical ramifications, our work may shed light on a long-standing mystery regarding sarcomeric architecture. Electron microscopy reconstructions of serial sections of skeletal muscle dating back over 50 years presented a puzzle: the thin/thick filament arrangement in the sarcomere alters its geometry, progressing from a tetragonal lattice arrangement at the Z-band to a trigonal arrangement in the A-band (Pringle, 1968), with the transition taking place near the I-A junction (Traeger et al., 1983). The implication for TTN would be that there would be a resulting increase in the copies of TTN protein, with four at the Z-band and six at the A-band and M-line (Liversage et al., 2001), which is difficult to reconcile with the fact that a single TTN molecule spans the entire length, from Z-band to M-line. Our data provides a simple explanation for this observation, with the Cronos protein product selectively increasing copies of TTN at the distal I-band, A-band and M-line. Our work also proposes an alternative explanation for the consistently observed ‘T2’ band on protein gels, which migrates at the expected molecular weight of Cronos. This has invariably been labeled a C-terminal degradation product, as it reacts with A-band antibodies (Opitz et al., 2004). However, in keeping with Cronos expression patterns in development, the T2 band was prominently seen in developing mouse (Lahmers et al., 2004) and rat hearts (Opitz et al., 2004), but was markedly reduced in adulthood. It is possible that, at least in some situations, T2 and Cronos represent the same isoform.

These results also raise a number of questions related to the function of the Cronos isoform. Given its more marked expression in early development, is it important only for the initial sarcomerogenesis or in subsequent aspects of muscle function? Is it likely to be reactivated under stress? And how fast does it turn over compared to full-length TTN? Models with selective disruption and tagging of this isoform should begin to answer these questions. Finally, it is possible that unrecognized internal promoters influence disease severity in truncating mutations of other genes. It is likely a more nuanced diagnostic and prognostic interpretation of truncating mutations will be needed.

1. 1. Liu Y
   2. Morley MP
   3. Brandimarto J
   4. Hannenhalli S
   5. Hu Y
   6. Ashley EA
   7. Tang WH
   8. Moravec CS
   9. Margulies KB
   10. Cappola T
   11. Li M

   (2015) RNA-Seq Identifies Novel Myocardial Gene Expression Signatures of Heart Failure [RNA-seq]

   Publicly available at the NCBI Gene Expression Omnibus (Accession no: GSE57344).

   http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE57344
2. 1. Stamatoyannopoulos J

   (2010) University of Washington Human Reference Epigenome Mapping Project

   Publicly available at the NCBI Gene Expression Omnibus (Accession no: GSE18927).

   http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE18927

1. 1. Bang M-L
   2. Centner T
   3. Fornoff F
   4. Geach AJ
   5. Gotthardt M
   6. Mcnabb M
   7. Witt CC
   8. Labeit D
   9. Gregorio CC
   10. Granzier H
   11. Labeit S

   (2001) The complete gene sequence of titin, expression of an unusual 700-kDa titin isoform, and its interaction with obscurin identify a novel z-line to i-band linking system

   Circulation Research 89:1065–1072.

   https://doi.org/10.1161/hh2301.100981

   • Google Scholar
2. 1. Bullard B
   2. Linke WA
   3. Leonard K

   (2002) Varieties of elastic protein in invertebrate muscles

   Journal of Muscle Research and Cell Motility 23:435–447.

   https://doi.org/10.1023/A:1023454305437

   • Google Scholar
3. 1. Carmignac V
   2. Salih MA
   3. Quijano-Roy S
   4. Marchand S
   5. Al Rayess MM
   6. Mukhtar MM
   7. Urtizberea JA
   8. Labeit S
   9. Guicheney P
   10. Leturcq F
   11. Gautel M
   12. Fardeau M
   13. Campbell KP
   14. Richard I
   15. Estournet B
   16. Ferreiro A

   (2007) C-terminal titin deletions cause a novel early-onset myopathy with fatal cardiomyopathy

   Annals of Neurology 61:340–351.

   https://doi.org/10.1002/ana.21089

   • Google Scholar
4. 1. Ceyhan-Birsoy O
   2. Agrawal PB
   3. Hidalgo C
   4. Schmitz-Abe K
   5. DeChene ET
   6. Swanson LC
   7. Soemedi R
   8. Vasli N
   9. Iannaccone ST
   10. Shieh PB
   11. Shur N
   12. Dennison JM
   13. Lawlor MW
   14. Laporte J
   15. Markianos K
   16. Fairbrother WG
   17. Granzier H
   18. Beggs AH

   (2013) Recessive truncating titin gene, TTN, mutations presenting as centronuclear myopathy

   Neurology 81:1205–1214.

   https://doi.org/10.1212/WNL.0b013e3182a6ca62

   • Google Scholar
5. 1. Chauveau C
   2. Rowell J
   3. Ferreiro A

   (2014) A rising titan: TTN review and mutation update

   Human Mutation 35:1046–1059.

   https://doi.org/10.1002/humu.22611

   • Google Scholar
6. 1. Codd MB
   2. Sugrue DD
   3. Gersh BJ
   4. Melton LJ

   (1989) Epidemiology of idiopathic dilated and hypertrophic cardiomyopathy. a population-based study in olmsted county, minnesota, 1975-1984

   Circulation 80:564–572.

   https://doi.org/10.1161/01.CIR.80.3.564

   • Google Scholar
7. 1. Dahme T
   2. Katus HA
   3. Rottbauer W

   (2009) Fishing for the genetic basis of cardiovascular disease

   Disease Models & Mechanisms 2:18–22.

   https://doi.org/10.1242/dmm.000687

   • Google Scholar
8. 1. Deo RC
   2. Musso G
   3. Tasan M
   4. Tang P
   5. Poon A
   6. Yuan C
   7. Felix JF
   8. Vasan RS
   9. Beroukhim R
   10. De Marco T
   11. Kwok P-Y
   12. MacRae CA
   13. Roth FP

   (2014) Prioritizing causal disease genes using unbiased genomic features

   Genome Biology 15:534.

   https://doi.org/10.1186/s13059-014-0534-8

   • Google Scholar
9. 1. Furst DO
   2. Osborn M
   3. Nave R
   4. Weber K

   (1988) The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the z line extends close to the m line

   The Journal of Cell Biology 106:1563–1572.

   https://doi.org/10.1083/jcb.106.5.1563

   • Google Scholar
10. 1. Gaudet M
    2. Fara A-G
    3. Sabatti M
    4. Kuzminsky E
    5. Mugnozza GS

    (2007) Single-reaction for SNP genotyping on agarose gel by allele-specific PCR in black poplar (populus nigra l.)

    Plant Molecular Biology Reporter 25:1–9.

    https://doi.org/10.1007/s11105-007-0003-6

    • Google Scholar
11. 1. Geoffrey Burns C
    2. Macrae C

    (2006) Purification of hearts from zebrafish embryos

    BioTechniques 40:278–282.

    https://doi.org/10.2144/000112135

    • Google Scholar
12. 1. Gramlich M
    2. Michely B
    3. Krohne C
    4. Heuser A
    5. Erdmann B
    6. Klaassen S
    7. Hudson B
    8. Magarin M
    9. Kirchner F
    10. Todiras M
    11. Granzier H
    12. Labeit S
    13. Thierfelder L
    14. Gerull B

    (2009) Stress-induced dilated cardiomyopathy in a knock-in mouse model mimicking human titin-based disease

    Journal of Molecular and Cellular Cardiology 47:352–358.

    https://doi.org/10.1016/j.yjmcc.2009.04.014

    • Google Scholar
13. 1. Gramlich M
    2. Pane LS
    3. Zhou Q
    4. Chen Z
    5. Murgia M
    6. Schötterl S
    7. Goedel A
    8. Metzger K
    9. Brade T
    10. Parrotta E
    11. Schaller M
    12. Gerull B
    13. Thierfelder L
    14. Aartsma-Rus A
    15. Labeit S
    16. Atherton JJ
    17. Mcgaughran J
    18. Harvey RP
    19. Sinnecker D
    20. Mann M
    21. Laugwitz KL
    22. Gawaz MP
    23. Moretti A

    (2015) Antisense-mediated exon skipping: a therapeutic strategy for titin-based dilated cardiomyopathy

    EMBO Molecular Medicine 7:562–576.

    https://doi.org/10.15252/emmm.201505047

    • Google Scholar
14. 1. Granzier HL
    2. Labeit S

    (2004) The giant protein titin: a major player in myocardial mechanics, signaling, and disease

    Circulation Research 94:284–295.

    https://doi.org/10.1161/01.RES.0000117769.88862.F8

    • Google Scholar
15. 1. Hazebroek MR
    2. Moors S
    3. Dennert R
    4. van den Wijngaard A
    5. Krapels I
    6. Hoos M
    7. Verdonschot J
    8. Merken JJ
    9. de Vries B
    10. Wolffs PF
    11. Crijns HJGM
    12. Brunner-La Rocca H-P
    13. Heymans S

    et al. (2015) Prognostic relevance of gene-environment interactions in patients with dilated cardiomyopathy

    Journal of the American College of Cardiology 66:1313–1323.

    https://doi.org/10.1016/j.jacc.2015.07.023

    • Google Scholar
16. 1. Heredia JE
    2. Mukundan L
    3. Chen FM
    4. Mueller AA
    5. Deo RC
    6. Locksley RM
    7. Rando TA
    8. Chawla A
    9. Mueller AA
    10. Deo RC
    11. Locksley RM

    (2013) Type 2 innate signals stimulate fibro/adipogenic progenitors to facilitate muscle regeneration

    Cell 153:376–388.

    https://doi.org/10.1016/j.cell.2013.02.053

    • Google Scholar
17. 1. Herman DS
    2. Lam L
    3. Taylor MR
    4. Wang L
    5. Teekakirikul P
    6. Christodoulou D
    7. Conner L
    8. Depalma SR
    9. Mcdonough B
    10. Sparks E
    11. Teodorescu DL
    12. Cirino AL
    13. Banner NR
    14. Pennell DJ
    15. Graw S
    16. Merlo M
    17. di Lenarda A
    18. Sinagra G
    19. Bos JM
    20. Ackerman MJ
    21. Mitchell RN
    22. Murry CE
    23. Lakdawala NK
    24. Ho CY
    25. Barton PJ
    26. Cook SA
    27. Mestroni L
    28. Seidman JG
    29. Seidman CE

    (2012) Truncations of titin causing dilated cardiomyopathy

    The New England Journal of Medicine 366:619–628.

    https://doi.org/10.1056/NEJMoa1110186

    • Google Scholar
18. 1. Hinson JT
    2. Chopra A
    3. Nafissi N
    4. Polacheck WJ
    5. Benson CC
    6. Swist S
    7. Gorham J
    8. Yang L
    9. Schafer S
    10. Sheng CC
    11. Haghighi A
    12. Homsy J
    13. Hubner N
    14. Church G
    15. Cook SA
    16. Linke WA
    17. Chen CS
    18. Seidman JG
    19. Seidman CE

    (2015) HEART DISEASE. titin mutations in iPS cells define sarcomere insufficiency as a cause of dilated cardiomyopathy

    Science 349:982–986.

    https://doi.org/10.1126/science.aaa5458

    • Google Scholar
19. 1. Hwang WY
    2. Fu Y
    3. Reyon D
    4. Maeder ML
    5. Tsai SQ
    6. Sander JD
    7. Peterson RT
    8. Yeh JR
    9. Joung JK

    (2013) Efficient genome editing in zebrafish using a CRISPR-cas system

    Nature Biotechnology 31:227–229.

    https://doi.org/10.1038/nbt.2501

    • Google Scholar
20. 1. Itoh-Satoh M
    2. Hayashi T
    3. Nishi H
    4. Koga Y
    5. Arimura T
    6. Koyanagi T
    7. Takahashi M
    8. Hohda S
    9. Ueda K
    10. Nouchi T
    11. Hiroe M
    12. Marumo F
    13. Imaizumi T
    14. Yasunami M
    15. Kimura A

    (2002) Titin mutations as the molecular basis for dilated cardiomyopathy

    Biochemical and Biophysical Research Communications 291:385–393.

    https://doi.org/10.1006/bbrc.2002.6448

    • Google Scholar
21. 1. Jinek M
    2. Chylinski K
    3. Fonfara I
    4. Hauer M
    5. Doudna JA
    6. Charpentier E

    (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity

    Science 337:816–821.

    https://doi.org/10.1126/science.1225829

    • Google Scholar
22. 1. Kakaradov B
    2. Xiong H
    3. Lee LJ
    4. Jojic N
    5. Frey BJ

    (2012) Challenges in estimating percent inclusion of alternatively spliced junctions from RNA-seq data

    BMC Bioinformatics 13:S11.

    https://doi.org/10.1093/bioinformatics/btp120

    • Google Scholar
23. 1. Lahmers S
    2. Wu Y
    3. Call DR
    4. Labeit S
    5. Granzier H

    (2004) Developmental control of titin isoform expression and passive stiffness in fetal and neonatal myocardium

    Circulation Research 94:505–513.

    https://doi.org/10.1161/01.RES.0000115522.52554.86

    • Google Scholar
24. 1. Lewinter MM
    2. Granzier HL

    (2013) Titin is a major human disease gene

    Circulation 127:938–944.

    https://doi.org/10.1161/CIRCULATIONAHA.112.139717

    • Google Scholar
25. 1. Li H
    2. Durbin R

    (2009) Fast and accurate short read alignment with burrows-wheeler transform

    Bioinformatics 25:1754–1760.

    https://doi.org/10.1093/bioinformatics/btp324

    • Google Scholar
26. 1. Liversage AD
    2. Holmes D
    3. Knight PJ
    4. Tskhovrebova L
    5. Trinick J

    (2001) Titin and the sarcomere symmetry paradox

    Journal of Molecular Biology 305:401–409.

    https://doi.org/10.1006/jmbi.2000.4279

    • Google Scholar
27. 1. Matz M
    2. Shagin D
    3. Bogdanova E
    4. Britanova O
    5. Lukyanov S
    6. Diatchenko L

    et al. (1999) Amplification of cDNA ends based on template-switching effect and step- out PCR

    Nucleic Acids Research 27:1558–1560.

    https://doi.org/10.1093/nar/27.6.1558

    • Google Scholar
28. 1. McKenna A
    2. Hanna M
    3. Banks E
    4. Sivachenko A
    5. Cibulskis K
    6. Kernytsky A
    7. Garimella K
    8. Altshuler D
    9. Gabriel S
    10. Daly M
    11. DePristo MA

    (2010) The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data

    Genome Research 20:1297–1303.

    https://doi.org/10.1101/gr.107524.110

    • Google Scholar
29. 1. Myhre JL
    2. Hills JA
    3. Prill K
    4. Wohlgemuth SL
    5. Pilgrim DB

    (2014) The titin a-band rod domain is dispensable for initial thick filament assembly in zebrafish

    Developmental Biology 387:93–108.

    https://doi.org/10.1016/j.ydbio.2013.12.020

    • Google Scholar
30. 1. Neph S
    2. Vierstra J
    3. Stergachis AB
    4. Reynolds AP
    5. Haugen E
    6. Vernot B

    (2013)

    An expansive human regulatory lexicon encoded in transcription factor footprints

    Nature 488:83–90.

    • Google Scholar
31. 1. Norton N
    2. Li D
    3. Rampersaud E
    4. Morales A
    5. Martin ER
    6. Zuchner S
    7. Guo S
    8. Gonzalez M
    9. Hedges DJ
    10. Robertson PD
    11. Krumm N
    12. Nickerson DA
    13. Hershberger RE
    14. National Heart, Lung, and Blood Institute GO Exome Sequencing Project and the Exome Sequencing Project Family Studies Project Team
    15. National Heart, Lung, and Blood Institute GO Exome Sequencing Project and the Exome Sequencing Project Family Studies Project Team

    (2013) Exome sequencing and genome-wide linkage analysis in 17 families illustrate the complex contribution of TTN truncating variants to dilated cardiomyopathy

    Circulation. Cardiovascular Genetics 6:144–153.

    https://doi.org/10.1161/CIRCGENETICS.111.000062

    • Google Scholar
32. 1. Oleykowski CA
    2. Bronson Mullins CR
    3. Godwin AK
    4. Yeung AT

    (1998) Mutation detection using a novel plant endonuclease

    Nucleic Acids Research 26:4597–4602.

    https://doi.org/10.1093/nar/26.20.4597

    • Google Scholar
33. 1. Opitz CA
    2. Leake MC
    3. Makarenko I
    4. Benes V
    5. Linke WA

    (2004) Developmentally regulated switching of titin size alters myofibrillar stiffness in the perinatal heart

    Circulation Research 94:967–975.

    https://doi.org/10.1161/01.RES.0000124301.48193.E1

    • Google Scholar
34. 1. Pervouchine DD
    2. Knowles DG
    3. Guigo R

    (2013) Intron-centric estimation of alternative splicing from RNA-seq data

    Bioinformatics. Oxford University Press 29:273–274.

    https://doi.org/10.1093/bioinformatics/bts678

    • Google Scholar
35. 1. Picelli S
    2. Faridani OR
    3. Askbo R
    4. Winberg GOS
    5. Sagasser S
    6. Sandberg R

    (2014)

    Full-length RNA-seq from single cells using smart-seq2

    Nature Publishing Group 9:171–181.

    • Google Scholar
36. 1. Pringle JW

    (1968)

    Mechano-chemical transformation in striated muscle

    Symposia of the Society for Experimental Biology 22:67–86.

    • Google Scholar
37. 1. Radke MH
    2. Peng J
    3. Wu Y
    4. Mcnabb M
    5. Nelson OL
    6. Granzier H
    7. Gotthardt M

    (2007) Targeted deletion of titin N2B region leads to diastolic dysfunction and cardiac atrophy

    Proceedings of the National Academy of Sciences of the United States of America 104:3444–3449.

    https://doi.org/10.1073/pnas.0608543104

    • Google Scholar
38. 1. Richards S
    2. Aziz N
    3. Bale S
    4. Bick D
    5. Das S
    6. Gastier-Foster J
    7. Grody WW
    8. Hegde M
    9. Lyon E
    10. Spector E
    11. Voelkerding K
    12. Rehm HL
    13. Gastier-Foster J

    (2015) Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the american college of medical genetics and genomics and the association for molecular pathology

    Genetics in Medicine 17:405–423.

    https://doi.org/10.1038/gim.2015.30

    • Google Scholar
39. 1. Roberts AM
    2. Ware JS
    3. Herman DS
    4. Schafer S
    5. Baksi J
    6. Bick AG
    7. Buchan RJ
    8. Walsh R
    9. John S
    10. Wilkinson S
    11. Mazzarotto F
    12. Felkin LE
    13. Gong S
    14. Macarthur JA
    15. Cunningham F
    16. Flannick J
    17. Gabriel SB
    18. Altshuler DM
    19. Macdonald PS
    20. Heinig M
    21. Keogh AM
    22. Hayward CS
    23. Banner NR
    24. Pennell DJ
    25. O'Regan DP
    26. San TR
    27. de Marvao A
    28. Dawes TJ
    29. Gulati A
    30. Birks EJ
    31. Yacoub MH
    32. Radke M
    33. Gotthardt M
    34. Wilson JG
    35. O'Donnell CJ
    36. Prasad SK
    37. Barton PJ
    38. Fatkin D
    39. Hubner N
    40. Seidman JG
    41. Seidman CE
    42. Cook SA

    (2015) Integrated allelic, transcriptional, and phenomic dissection of the cardiac effects of titin truncations in health and disease

    Science Translational Medicine 7:270ra6.

    https://doi.org/10.1126/scitranslmed.3010134

    • Google Scholar
40. 1. Robinson MD
    2. Mccarthy DJ
    3. Smyth GK

    (2010) EdgeR: a bioconductor package for differential expression analysis of digital gene expression data

    Bioinformatics (Oxford, England) 26:139–140.

    https://doi.org/10.1093/bioinformatics/btp616

    • Google Scholar
41. 1. Robu ME
    2. Larson JD
    3. Nasevicius A
    4. Beiraghi S
    5. Brenner C
    6. Farber SA
    7. Ekker SC

    (2007) P53 activation by knockdown technologies

    PLoS Genetics 3:e78.

    https://doi.org/10.1371/journal.pgen.0030078

    • Google Scholar
42. 1. Seeley M
    2. Huang W
    3. Chen Z
    4. Wolff WO
    5. Lin X
    6. Xu X

    (2007) Depletion of zebrafish titin reduces cardiac contractility by disrupting the assembly of z-discs and a-bands

    Circulation Research 100:238–245.

    https://doi.org/10.1161/01.RES.0000255758.69821.b5

    • Google Scholar
43. 1. Steffen LS
    2. Guyon JR
    3. Vogel ED
    4. Howell MH
    5. Zhou Y
    6. Weber GJ
    7. Zon LI
    8. Kunkel LM

    (2007) The zebrafish runzel muscular dystrophy is linked to the titin gene

    Developmental Biology 309:180–192.

    https://doi.org/10.1016/j.ydbio.2007.06.015

    • Google Scholar
44. 1. Thisse C
    2. Thisse B

    (2008) High-resolution in situ hybridization to whole-mount zebrafish embryos

    Nature Protocols 3:59–69.

    https://doi.org/10.1038/nprot.2007.514

    • Google Scholar
45. 1. Traeger L
    2. Mackenzie JM
    3. Epstein HF
    4. Goldstein MA

    (1983)

    Transition in the thin-filament arrangement in rat skeletal muscle

    Journal of Muscle Research and Cell Motility 4:353–366.

    • Google Scholar
46. 1. Trapnell C
    2. Pachter L
    3. Salzberg SL

    (2009) TopHat: discovering splice junctions with RNA-seq

    Bioinformatics 25:1105–1111.

    https://doi.org/10.1093/bioinformatics/btp120

    • Google Scholar
47. 1. van der Ven PF
    2. Bartsch JW
    3. Gautel M
    4. Jockusch H
    5. Fürst DO

    (2000)

    A functional knock-out of titin results in defective myofibril assembly

    Journal of Cell Science 113:1405–1414.

    • Google Scholar
48. 1. van Spaendonck-Zwarts KY
    2. Posafalvi A
    3. van den Berg MP
    4. Hilfiker-Kleiner D
    5. Bollen IA
    6. Sliwa K
    7. Alders M
    8. Almomani R
    9. van Langen IM
    10. van der Meer P
    11. Sinke RJ
    12. van der Velden J
    13. van Veldhuisen DJ
    14. van Tintelen JP
    15. Jongbloed JD

    (2014) Titin gene mutations are common in families with both peripartum cardiomyopathy and dilated cardiomyopathy

    European Heart Journal 35:2165–2173.

    https://doi.org/10.1093/eurheartj/ehu050

    • Google Scholar
49. 1. Warren CM
    2. Krzesinski PR
    3. Greaser ML

    (2003) Vertical agarose gel electrophoresis and electroblotting of high-molecular-weight proteins

    Electrophoresis 24:1695–1702.

    https://doi.org/10.1002/elps.200305392

    • Google Scholar
50. 1. Westerfield M

    (1993)

    The zebrafish book: a guide for the laboratory use of zebrafish (brachydanio rerio)

    • Google Scholar
51. Book

    1. Wickham H

    (2009)

    Ggplot2

    New York: NY: Springer Science & Business Media.

    • Google Scholar
52. 1. Xu X
    2. Meiler SE
    3. Zhong TP
    4. Mohideen M
    5. Crossley DA
    6. Burggren WW
    7. Fishman MC

    (2002) Cardiomyopathy in zebrafish due to mutation in an alternatively spliced exon of titin

    Nature Genetics 30:205–209.

    https://doi.org/10.1038/ng816

    • Google Scholar
53. 1. Zetoune AB
    2. Fontanière S
    3. Magnin D
    4. Anczuków O
    5. Buisson M
    6. Zhang CX
    7. Mazoyer S

    et al. (2008) Comparison of nonsense-mediated mRNA decay efficiency in various murine tissues

    BMC Genetics 9:83–11.

    https://doi.org/10.1186/1471-2156-9-83

    • Google Scholar
54. 1. Zhou VW
    2. Goren A
    3. Bernstein BE

    (2011) Charting histone modifications and the functional organization of mammalian genomes

    Nature Reviews Genetics 12:7–18.

    https://doi.org/10.1038/nrg2905

    • Google Scholar

Author details

1. Jun Zou

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   JZ, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

2. Diana Tran

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   DT, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

3. Mai Baalbaki

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   MB, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

4. Ling Fung Tang

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   LFT, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

5. Annie Poon

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   AP, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

6. Angelo Pelonero

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   AnP, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

7. Erron W Titus

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   EWT, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0001-6868-9121

8. Christiana Yuan

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   CY, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

9. Chenxu Shi

   Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

   Contribution

   CS, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

10. Shruthi Patchava

    Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

    Contribution

    SP, Acquisition of data, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

11. Elizabeth Halper

    Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

    Contribution

    EH, Acquisition of data, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

12. Jasmine Garg

    Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

    Contribution

    JG, Acquisition of data, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

13. Irina Movsesyan

    Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States

    Contribution

    IM, Acquisition of data, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

14. Chaoying Yin

    1. Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, Chapel Hill, United States
    2. McAllister Heart Institute, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    ChY, Conception and design, Acquisition of data

    Competing interests

    The authors declare that no competing interests exist.

15. Roland Wu

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Medicine, University of California, San Francisco, San Francisco, United States

    Contribution

    RW, Acquisition of data, Analysis and interpretation of data

    Competing interests

    The authors declare that no competing interests exist.

16. Lisa D Wilsbacher

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Medicine, University of California, San Francisco, San Francisco, United States

    Present address

    Department of Medicine and Feinberg Cardiovascular Research Institute, Northwestern University Feinberg School of Medicine, United States

    Contribution

    LDW, Acquisition of data, Drafting or revising the article

    Competing interests

    The authors declare that no competing interests exist.

17. Jiandong Liu

    1. Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, Chapel Hill, United States
    2. McAllister Heart Institute, University of North Carolina at Chapel Hill, Chapel Hill, United States

    Contribution

    JL, Acquisition of data, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

18. Ronald L Hager

    Department of Exercise Sciences, Brigham Young University, Provo, United States

    Contribution

    RLH, Drafting or revising the article, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

    "This ORCID iD identifies the author of this article:" 0000-0002-5128-4635

19. Shaun R Coughlin

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Medicine, University of California, San Francisco, San Francisco, United States

    Contribution

    SRC, Conception and design, Acquisition of data, Drafting or revising the article

    Competing interests

    The authors declare that no competing interests exist.

20. Martin Jinek

    Department of Biochemistry, University of Zurich, Zurich, Switzerland

    Contribution

    MJ, Conception and design, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

21. Clive R Pullinger

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Physiological Nursing, University of California, San Francisco, San Francisco, United States

    Contribution

    CRP, Drafting or revising the article, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

22. John P Kane

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Biochemistry and Biophysics, University of California, San Francisco, San Francisco, United States

    Contribution

    JPK, Conception and design, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

23. Daniel O Hart

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Biochemistry and Biophysics, University of California, San Francisco, San Francisco, United States

    Contribution

    DOH, Acquisition of data, Contributed unpublished essential data or reagents

    Competing interests

    The authors declare that no competing interests exist.

24. Pui-Yan Kwok

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Dermatology, University of California, San Francisco, San Francisco, United States
    3. Institute for Human Genetics, University of California, San Francisco, United States

    Contribution

    PYK, Conception and design, Acquisition of data

    Competing interests

    The authors declare that no competing interests exist.

25. Rahul C Deo

    1. Cardiovascular Research Institute, University of California, San Francisco, San Francisco, United States
    2. Department of Medicine, University of California, San Francisco, San Francisco, United States
    3. Institute for Human Genetics, University of California, San Francisco, United States
    4. California Institute for Quantitative Biosciences, San Francisco, United States

    Contribution

    RCD, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    For correspondence

    rahul.deo@ucsf.edu

    Competing interests

    The authors declare that no competing interests exist.

• 3,944

  views

• 843

  downloads

• 88

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.