Titin

MYBPC3, MYH7, and Titin (TTN) encode the primary structural and mechanochemical proteins of the vertebrate cardiac sarcomere, and mutations across these three genes account for the majority of inherited hypertrophic cardiomyopathy (HCM, prevalence ~1 in 200 to 1 in 500) and dilated cardiomyopathy (DCM, prevalence ~1 in 250). Within the 2.2-μm contractile unit of ventricular cardiomyocytes, β-cardiac myosin heavy chain (MYH7 on chromosome 14q11.2) forms the bipolar thick-filament motor that hydrolyzes ATP to generate piconewton power strokes along actin thin filaments; cardiac myosin-binding protein C (MYBPC3 on Chr11p11.2) acts as a phosphorylation-sensitive molecular brake tethering the myosin S1/S2 neck to the thick-filament backbone; and titin (TTN on Chr2q31.2) spans a full 1.1-μm half-sarcomere from the Z-disc to the M-band as the largest polypeptide in nature (~3.8 MDa), providing passive diastolic viscoelasticity and length-dependent Frank–Starling activation.

High-resolution cryo-electron microscopy (cryo-EM) of native cardiac thick filaments and single-molecule optical trapping biophysics have unified the opposing phenotypes of HCM and DCM around a single thermodynamic equilibrium: the ratio of myosin heads parked in the energy-conserving, auto-inhibited Interacting-Heads Motif (IHM) / Super-Relaxed State (SRX) versus the catalytically active Disordered-Relaxed State (DRX). Pathogenic HCM mutations in MYH7 (p.Arg403Gln, p.Arg453Cys) and truncations in MYBPC3 destabilize the IHM/SRX state, forcing an excess of myosin heads (DRX) onto actin to drive hypercontractility, impaired diastolic relaxation, and energetic exhaustion—a defect reversed by the first-in-class allosteric cardiac myosin inhibitors mavacamten and aficamten. Conversely, truncating variants in constitutive A-band exons of TTN (TTNtv) and hypocontractile MYH7 variants depress sarcomeric force generation (tension-time integral), triggering eccentric chamber remodeling and dilated cardiomyopathy (DCM).

The Interacting-Heads Motif (IHM) and ATPase Turnover Rates

Encoded by the MYH7 gene on chromosome 14q11.2 (40 exons, 1,935 amino acids, ~223 kDa heavy chain arranged in tandem with the fast α-myosin heavy chain gene MYH6), β-cardiac myosin assembles into hexamers (two MYH7 heavy chains + two essential light chains MYL3 + two regulatory light chains MYL2) comprising an N-terminal globular subfragment-1 (S1) motor head (residues 1–847), an α-helical subfragment-2 (S2) coiled-coil neck (residues 848–1,270), and a C-terminal light meromyosin (LMM) rod (residues 1,271–1,935) that polymerizes into the thick-filament shaft.

During diastole (and across ~40% to 50% of myosin heads even during baseline systole), cardiac myosin heads do not swing freely toward the actin thin filament. Instead, the two S1 heads of a single myosin dimer fold backward onto their own proximal S2 coiled-coil tail in an asymmetric, auto-inhibited structure termed the Interacting-Heads Motif (IHM). In the IHM, the actin-binding surface of the "blocked head" docks against the converter domain and catalytic pocket of the adjacent "free head", locking both active sites in the Super-Relaxed State (SRX):

1. Super-Relaxed State (SRX / Folded IHM): Exhibits an ultra-slow basal ATP turnover rate of k_SRX ≈ 0.003 to 0.005 s⁻¹ (1 ATP hydrolyzed every ~200 to 300 seconds), conserving myocardial oxygen consumption and allowing rapid ventricular filling during diastole.
2. Disordered-Relaxed State (DRX / Unfolded Heads): Myosin heads released from the IHM extend radially outward toward the thin filament, hydrolyzing ATP at a 10-fold higher basal rate (k_DRX ≈ 0.03 to 0.05 s⁻¹) and standing primed to enter the actin-activated cross-bridge cycle (k_cat ≈ 3 to 5 s⁻¹) as soon as cytosolic Ca²⁺ binds troponin C (TNNC1).

The Myosin Mesa, HCM Missense Variants (R403Q), and Mavacamten Allostery

When the distribution of more than 300 pathogenic MYH7 HCM missense variants (~30% to 35% of all HCM cases) is mapped onto the 3D IHM cryo-EM structure (Spudich, Ruppel, and colleagues), a striking cluster emerges along a flat, positively charged surface of the S1 catalytic domain termed the "myosin mesa", together with the converter domain and proximal S2 tail—precisely the intramolecular interfaces that hold the IHM folded. Canonical HCM mutations such as p.Arg403Gln (R403Q), p.Arg453Cys (R453C), p.Arg719Trp (R719W), and p.Val606Met (V606M) disrupt electrostatic salt bridges stabilizing the blocked-head:S2 interface. This shifts the conformational equilibrium from SRX toward DRX (reducing the SRX fraction from ~50% down to ~20–25%), recruiting 25% to 50% more myosin heads onto actin during each heartbeat. The resulting ensemble hypercontractility increases left ventricular outflow tract (LVOT) pressure gradients, prevents complete diastolic cross-bridge detachment (diastolic dysfunction), and depletes the myocardial phosphocreatine-to-ATP (PCr/ATP) ratio, triggering compensatory concentric septal hypertrophy and myocardial disarray.

Conversely, a distinct subset of MYH7 missense variants (p.Glu525Lys, p.Ser532Pro, p.Phe764Leu) either hyper-stabilizes the IHM/SRX state or directly impairs the converter-domain power stroke (reducing intrinsic force or actin-activated ATPase), producing hypocontractility and familial DCM. The therapeutic breakthrough for obstructive and non-obstructive HCM came with mavacamten (Camzyos) and aficamten: small-molecule allosteric modulators that bind the MYH7 S1 motor pocket (IC₅₀ ≈ 0.3–0.5 μM), slow phosphate (P_i) release, and refold excess DRX heads back into the energy-sparing IHM/SRX state, normalizing LVOT gradients and diastolic wall stress in Phase 3 trials (EXPLORER-HCM and SEQUOIA-HCM).

Encoded on chromosome 11p11.2 (35 exons, 1,274 amino acids, ~140 kDa), cardiac myosin-binding protein C (cMyBP-C / MYBPC3) is the single most frequently mutated gene in HCM, accounting for 40% to 50% of all genotyped cases. Structurally, cMyBP-C consists of eight immunoglobulin (Ig)-like domains and three fibronectin type-III (Fn3) domains designated C0 through C10, plus a cardiac-specific regulatory M-domain (between C1 and C2) containing four serine phosphorylation sites (Ser275, Ser284, Ser304, Ser311) targeted by protein kinase A (PKA) and Ca²⁺/calmodulin-dependent protein kinase II (CaMKII). Within the central C-zone of the sarcomere A-band, cMyBP-C is arranged in 9 transverse stripes spaced 43 nm apart: its C-terminal C8–C10 domains anchor tightly to the thick-filament LMM backbone and titin, while its N-terminal C0–C2 domains bind the myosin S1 mesa and S2 neck, physically clamping myosin heads in the IHM/SRX state.

When sympathetic β₁-adrenergic stimulation activates PKA, phosphorylation of the cMyBP-C M-domain neutralizes its binding to myosin S2, releasing myosin heads from SRX into DRX to accelerate lusitropy and inotropy. Unlike MYH7 (where >95% of pathogenic variants are missense alleles acting as poison polypeptides), ~70% of pathogenic MYBPC3 mutations are heterozygous frameshift, nonsense, or splice-site variants (including a 25-bp intronic deletion present in ~4% of South Asian populations). Truncated MYBPC3 transcripts are eliminated by nonsense-mediated mRNA decay (NMD) and the ubiquitin-proteasome system, reducing total myocardial cMyBP-C protein stoichiometry by ~30% to 40% (haploinsufficiency). With fewer cMyBP-C tethers in the C-zone, unrestrained myosin heads escape from SRX into DRX, phenocopying MYH7 mesa mutations and explaining why both MYBPC3 and MYH7 HCM respond equivalently to mavacamten (while inspiring TN-201, an AAV9 gene-replacement therapy delivering full-length MYBPC3 cDNA).

The 363-Exon Giant Spring and RBM20 Alternative Splicing

Spanning 294 kb across 363 coding exons on chromosome 2q31.2, Titin (TTN) encodes the largest protein in the human proteome (up to 34,350 amino acids, ~3.8 MDa), comprising four structural zones across each half-sarcomere: the Z-disc N-terminal anchor (Tcap/α-actinin binding), the elastic I-band spring (tandem Ig domains, the N2B and N2A elements, and the unstructured proline-glutamate-valine-lysine PEVK coil), the rigid A-band scaffold (11-domain super-repeats of Ig and Fn3 modules that template exact thick-filament length at 1.6 μm), and the M-band C-terminal pseudokinase (TK) domain. In adult human myocardium, the splicing repressor RBM20 excises scores of extensible I-band exons to generate two principal cardiac isoforms: the short, stiff N2B isoform (~3.0 MDa) and the longer, more compliant N2BA isoform (~3.2–3.3 MDa), whose ratio (~65:35 in healthy human left ventricle) dictates passive diastolic ventricular stiffness.

Why Constitutive A-Band TTNtv Cause DCM While Low-PSI I-Band TTNtv Are Benign

A landmark puzzle in clinical genomics arose when exome sequencing revealed that heterozygous titin-truncating variants (TTNtv—nonsense, frameshift, or canonical splice variants) occur in ~20% to 25% of familial dilated cardiomyopathy (DCM) and ~12% to 15% of sporadic DCM, yet are also found in ~0.5% to 1.0% of the general population. Roberts, Cook, Ware, and colleagues resolved this paradox via the Percent Spliced In (PSI) rule:

1. Low-PSI (<50%) I-Band Exons (Clinically Benign): Background population TTNtv cluster inside alternatively spliced fetal or skeletal-muscle-specific I-band exons that are naturally spliced out (PSI < 15–50%) by RBM20 in adult cardiomyocytes. Because the mutant exon is excluded from mature cardiac TTN transcripts, full-length cardiac titin synthesis is unaffected.
2. High-PSI (>90%, Constitutive) Exons, Especially the A-Band (OR ≈ 30–50 for DCM): Pathogenic DCM TTNtv occur almost exclusively in constitutive exons expressed in >90% of adult cardiac N2B/N2BA transcripts, with the highest penetrance in the A-band and distal I-band. Recent ribosome profiling and quantitative proteomics of human DCM explanted hearts demonstrate a dual pathogenic mechanism: A-band TTNtv cause both full-length titin haploinsufficiency (~15–25% deficit) and accumulation of stable, C-terminally truncated titin "poison peptides" that incorporate into the Z-disc and I-band but fail to anchor at the M-band, disrupting mechanosensing, blunting length-dependent Frank–Starling activation, and triggering metabolic shift and chamber dilation.

How do mavacamten and aficamten reverse the molecular defect of hypertrophic cardiomyopathy (HCM)?

In HCM caused by MYH7 missense variants or MYBPC3 haploinsufficiency, the auto-inhibited Interacting-Heads Motif (IHM) / Super-Relaxed State (SRX) is destabilized, releasing excess myosin heads into the active DRX state and causing hypercontractility and diastolic stiffness. Mavacamten and aficamten bind allosterically to the β-cardiac myosin S1 motor domain, slowing phosphate release and shifting myosin heads back into the folded, energy-sparing IHM/SRX conformation, which reduces LVOT obstruction and restores diastolic relaxation.

Why do truncating variants in MYBPC3 cause hypertrophic cardiomyopathy (HCM), whereas truncating variants in TTN cause dilated cardiomyopathy (DCM)?

MYBPC3 functions as a mechanical tether (brake) in the thick-filament C-zone that holds myosin heads parked in the inactive SRX state; thus, a 30% to 40% reduction in cMyBP-C protein (haploinsufficiency) releases myosin heads onto actin, causing hypercontractility and HCM. Conversely, TTN is the longitudinal viscoelastic spring required to transmit passive tension and activate thick filaments via length-dependent Frank–Starling strain; truncating TTN (TTNtv) reduces sarcomeric force generation (hypocontractility), driving compensatory chamber dilation and DCM.

Why is the Percent Spliced In (PSI) metric essential when interpreting a TTN truncating variant on a genetic test report?

Of the 363 exons in the TTN gene, more than 150 I-band exons are alternatively spliced out of adult human heart transcripts (PSI < 50%) by RBM20. A truncating variant (TTNtv) located in a low-PSI exon is skipped during cardiac mRNA processing and is benign, whereas a TTNtv located in a constitutively expressed exon (PSI > 90%), particularly within the A-band, disrupts mature cardiac N2B/N2BA titin and confers a 30- to 50-fold increased risk of DCM.

1. Alamo, L., Ware, J. S., Pinto, A., et al. (2017). "Effects of myosin variants on interacting-heads motif explain distinct hypertrophic and dilated cardiomyopathy phenotypes." eLife, 6, e24634. https://doi.org/10.7554/eLife.24634
2. Roberts, A. M., Ware, J. S., Herman, D. S., et al. (2015). "Integrated allelic, transcriptional, and phenomic dissection of the cardiac effects of titin truncations in health and disease." Science Translational Medicine, 7(270), 270ra6. https://doi.org/10.1126/scitranslmed.3010134
3. Olivotto, I., Oreziak, A., Barriales-Villa, R., et al. (2020). "Mavacamten for treatment of symptomatic obstructive hypertrophic cardiomyopathy (EXPLORER-HCM): a randomised, double-blind, placebo-controlled, phase 3 trial." The Lancet, 396(10253), 759–769. https://doi.org/10.1016/S0140-6736(20)31792-X31792-X)
4. McAfee, Q., Chen, C. Y., Yang, Y., et al. (2021). "Truncated titin proteins in dilated cardiomyopathy." Science Translational Medicine, 13(618), eabd7287. https://doi.org/10.1126/scitranslmed.abd7287