Effect of MyBP-C binding to actin on contractility in heart muscle

Abstract

In contrast to skeletal muscle isoforms of myosin binding protein C (MyBP-C), the cardiac isoform has 11 rather than 10 fibronectin or Ig modules (modules are identified as C0 to C10, NH2 to COOH terminus), 3 phosphorylation sites between modules C1 and C2, and 28 additional amino acids rich in proline in C5. Phosphorylation between C1 and C2 increases maximum Ca-activated force (Fmax), alters thick filament structure, and increases the probability of myosin heads on the thick filament binding to actin on the thin filament. Unphosphorylated C1C2 fragment binds to myosin, but phosphorylation inhibits the binding. MyBP-C also binds to actin. Using two types of immunoprecipitation and cosedimentation, we show that fragments of MyBP-C containing C0 bind to actin. In low concentrations C0-containing fragments bind to skinned fibers when the NH2 terminus of endogenous MyBP-C is bound to myosin, but not when MyBP-C is bound to actin. C1C2 fragments bind to skinned fibers when endogenous MyBP-C is bound to actin but not to myosin. Disruption of interactions of endogenous C0 with a high concentration of added C0C2 fragments produces the same effect on contractility as extraction of MyBP-C, namely decrease in Fmax and increase in Ca sensitivity. These results suggest that cardiac contractility can be regulated by shifting the binding of the NH2 terminus of MyBP-C between actin and myosin. This mechanism may have an effect on diastolic filling of the heart.

Figure 1.

Diagram of cardiac MyBP-C showing 11 modules, binding sites, phosphorylation sites, and regions specific for the cardiac isoform. Also shown are the fragments used in this study, an estimate of their length in numbers of amino acid residues and the position of the His tag used in preparing the fragments. The length of a skeletal MyBP-C is ∼35 nm (Offer et al., 1973).

Figure 2.

Gels of purified fragments and actin stained with Coomassie blue after SDS PAGE. (A) successive fractions of C0 during elution; standards at right. (B) Successive fractions of C0C2 during elution; standards at left. (C) C1C2 and C0C1; lane 1 contains markers and (D) actin.

Figure 3.

(A) Western blot with an antibody to cardiac C0C1 showing binding of actin to C0C2 measured by immunoprecipitation (See materials and methods). Lanes 1 and 2 show results with a mixture containing C0C2 and actin. Lane 3 contains only actin, and lanes 4 and 5 contain C1C2 and actin. In lanes 2 and 5 the fragments had been phosphorylated before incubation with actin. Labels mark expected positions of C0C2 and C1C2. (B) A Western blot with antiactin showing the binding of actin to natural C0 (lane 2) and natural C0C2 (no exposure to urea, lane 5), but the absence after denaturation with urea (C0, lane 1) and decrease (C0C2, lane 4) after renaturation by removal of the urea. Lane 3 contains only the resin. The lane at left contains only actin.

Figure. 4.

(Top) Coomassie blue–stained gels after SDS-PAGE in cosedimentation experiments. Lane 1 contains standards. Lanes 2, 5, and 8 contain the beginning solutions; lanes 3, 6, and 9 the supernatants after centrifugation; and lanes 4, 7, and 10 contain the pellets. In lanes 2–4 the starting solution contained only filamentous actin. In lanes 5–7 and 8–10 the starting solutions contained filamentous actin and C0C2. The actin concentration in all three runs was 4.1 μM. In lanes 5–7 and 8–10 the concentrations of C0C2 were, respectively, 0.3 and 0.1 μM. (Bottom) Cosedimentation showing saturation of actin with C0C2. Lanes 1 and 3 are supernatant and lanes 2 and 4 are pellet. Concentrations of C0C2 in initial solution in lanes 1 and 2 and 3 and 4 are, respectively, 0.6 and 0.9 μM. Lane 5 contains markers with values to the right.

Figure 5.

(A) IEF gel of lysate of two muscles stained with anti-C0C1, one soaked at rest in bathing solution containing 2.5 mM Ca, and a second soaked in a bathing solution of the same composition except that Ca was 1.25 mM. The different bands indicate MyBP-C containing 0, 1, 2, and 3 phosphates per molecule of MyBP-C. (B) The average distribution of phosphorylated forms of MyBP-C in eight experiments.

Figure 5.

(A) IEF gel of lysate of two muscles stained with anti-C0C1, one soaked at rest in bathing solution containing 2.5 mM Ca, and a second soaked in a bathing solution of the same composition except that Ca was 1.25 mM. The different bands indicate MyBP-C containing 0, 1, 2, and 3 phosphates per molecule of MyBP-C. (B) The average distribution of phosphorylated forms of MyBP-C in eight experiments.

Figure 6.

Western blots stained with anti-C0C1 showing reciprocal binding of C0C1 and C1C2 by skinned cardiac muscle occurs depending on the degree of phosphorylation of endogenous MyBP-C. (Right) The intact cardiac muscle was soaked in 1.25 mM Ca at rest for 2 h to produce a low level of phosphorylation after skinning. (Left) The intact cardiac muscle was soaked in 2.5 mM Ca at rest for 2 h to produce a high level of phosphorylation after skinning. Labels at the bottom indicate the fragment added to the skinned fiber preparation before lysing. Labels to the right indicate the position of a particular fragment on the gel. After skinning muscles were then soaked in relaxing solution containing a low concentration of the specified fragment (0.01 μM) and then briefly washed before lysing and electrophoresis. Note that C0C1 does not bind when phosphorylation is high (2.5 mM Ca). Presumably endogenous C0 is bound to actin under these conditions. C1C2 does not bind when phosphorylation is low (1.25 mM Ca), and endogenous C1C2 is bound to myosin. The faint bands present in gels of trabeculae soaked in 1.25 mM Ca are proteolytic fragments of MyBP-C, which are more common with unphosphorylated MyBP-C.

Figure 7.

Force tracing of a cardiac trabecula that had been skinned after a 2-h exposure to Krebs' solution containing 2.5 mM Ca. Note decrease in Fmax and increase in force at pCa 5.6 in presence of 3 μM C0. Changes are reversed within 10–15 min of bathing in a medium without C0. Calibration bars equal 15 min and 0.05 g. Thin arrows pointing down indicate pCa of 6.0, 5.6, and 5.2 in that order. Short thick arrows indicate 5.0 and longer thick arrows indicate 4.5. Short arrows pointing up indicate pCa of 9.0.

Figure 8.

Force/[Ca] relation of trabeculae skinned after 2 h soak in Krebs' solution containing either 1.25 mM (filled circles) or 2.5 mM (open circles) Ca. There is no significant difference.

Figure 9.

Addition of 3-μM C0 fragments to solution bathing skinned rat cardiac trabecula increases Ca sensitivity. The effect of C0 was compared on two different types of preparations: trabeculae with a high (left) and trabeculae with a low (right) degree of phosphorylation of endogenous MyBP-C. In the former, C0 of a majority of MyBP-C are bound to actin, and in the latter a majority of C1C2 are bound to myosin. Force is expressed as percent of maximum. Open squares indicate before addition of C0; x indicates with C0; and filled circles indicate after C0 removal.

Figure 10.

(A) Force tracing of a skinned cardiac trabecula before, during, and after soaking in 2 μM C0C2. Calibration lines equal 0.05 g and 20 min. Thin arrows pointing down indicate pCa of 6.0, 5.6, and 5.2 in that order. Short thick arrows indicate 5.0 and longer thick arrows indicate 4.5. Short arrows pointing up indicate pCa of 9.0. (B) Western blot of lysate of trabeculae stained with anti-His tag to demonstrate retention of C0C2 after 60 min in solutions without C0C2. In lanes 1 and 2 the trabeculae had been exposed to C0C2 for 75 min before the 60 min without C0C2. Lane 3 is the control, in which the 75 min soak solution did not contain C0C2. Lane 4 is C0C2.

Figure 10.

(A) Force tracing of a skinned cardiac trabecula before, during, and after soaking in 2 μM C0C2. Calibration lines equal 0.05 g and 20 min. Thin arrows pointing down indicate pCa of 6.0, 5.6, and 5.2 in that order. Short thick arrows indicate 5.0 and longer thick arrows indicate 4.5. Short arrows pointing up indicate pCa of 9.0. (B) Western blot of lysate of trabeculae stained with anti-His tag to demonstrate retention of C0C2 after 60 min in solutions without C0C2. In lanes 1 and 2 the trabeculae had been exposed to C0C2 for 75 min before the 60 min without C0C2. Lane 3 is the control, in which the 75 min soak solution did not contain C0C2. Lane 4 is C0C2.

Figure 11.

Comparison of the amino acid sequence of the proline-rich region of C0 with the sequence of the PEVK region of titin indicates a high degree of homology (58%).