Structure before function: myosin binding protein-C slow is a structural protein with regulatory properties

Abstract

Myosin binding protein-C slow (sMyBP-C) comprises a family of accessory proteins in skeletal muscles that bind both myosin and actin filaments. Herein, we examined the role of sMyBP-C in adult skeletal muscles using in vivo gene transfer and clustered regularly interspaced short palindromic repeats technology to knock down all known sMyBP-C variants. Our findings, confirmed in two different skeletal muscles, demonstrated efficient knockdown (KD) of sMyBP-C (>70%) resulting in notably decreased levels of thick, but not thin, filament proteins ranging from ∼50% for slow and fast myosin to ∼20% for myomesin. Consistent with this, A bands were selectively distorted, and sarcomere length was significantly reduced. Contrary to earlier in vitro studies showing that addition of recombinant sMyBP-C slows down the formation of actomyosin crossbridges, our work demonstrates that KD of sMyBP-C in intact myofibers results in decreased contraction and relaxation kinetics under no-load conditions. Similarly, KD muscles develop markedly reduced twitch and tetanic force and contraction velocity. Taken together, our results show that sMyBP-C is essential for the regular organization and maintenance of myosin filaments into A bands and that its structural role precedes its ability to regulate actomyosin crossbridges.—Geist, J., Ward, C. W., Kontrogianni-Konstantopoulos, A. Structure before function: myosin binding protein-C slow is a structural protein with regulatory properties.

Myosin binding protein-C (MyBP-C) comprises a family of thick filament accessory proteins within striated muscles. There are three isoforms, including slow skeletal (sMyBP-C), fast skeletal (fMyBP-C), and cardiac (cMyBP-C) encoded by MYBPC1, MYBPC2, and MYBPC3, respectively (1–3).

MyBP-C localizes in the thick filament within the crossbridge-bearing zone (C zone) of the A band in seven to nine transverse stripes that have a periodicity of 43 nm (4). The main binding partners of MyBP-C are myosin and actin. MyBP-C binds to light meromyosin via its COOH-terminus and to subfragment 2 (S2) of myosin via its NH₂ terminus (5, 6). The NH₂ terminus also supports binding to actin (1, 7, 8), although the COOH terminus may contain actin binding sites as well (9). Whereas the interaction of MyBP-C with light meromyosin is constant, its association with S2 and actin via the NH₂ terminus is dynamic and regulated through phosphorylation (10, 11).

Structural and regulatory roles have been attributed to MyBP-C. Support for a structural role is evidenced by early work showing that purified skeletal MyBP-C reduces the critical concentration required for myosin polymerization in vitro (12) and that cMyBP-C is required for the regular assembly of synthetic myosin filaments in regards to thickness, length, formation of bare zones, and head distribution (13, 14). Support for a regulatory role comes from evidence showing that addition of purified skeletal MyBP-C in skinned myofibers slows down the shortening velocity of actomyosin crossbridges (15) and that recombinant sMyBP-C reduces the sliding velocity of actin filaments past myosin heads in vitro in a variant-specific manner (5). Moreover, cMyBP-C has been suggested to act both as a “brake” and as an “accelerator” to regulate the rate of formation of actomyosin crossbridges in response to Ca²⁺ concentration and depending on its phosphorylation status (10, 16–20). Less work has focused on fMyBP-C, but a structural and functional impact was recently seen in skeletal muscles of zebrafish larvae depleted of fMyBP-C that exhibited shorter sarcomeres, wider interfilament space, reduced force production, and suppressed contraction kinetics (21).

Consistent with its key roles in striated muscle physiology, MyBP-C has been causatively linked to the pathogenesis of disease, with >500 mutations in MYBPC3 linked to hypertrophic and dilated cardiomyopathy (13, 22, 23) and an increasing number of mutations in MYBPC1 resulting in severe and lethal forms of distal arthrogryposis (24–28). Distal arthrogryposis is characterized by joint contractures primarily affecting the muscles of the hands and feet and is often accompanied by moderate to severe craniofacial anomalies and scoliosis.

In contrast to MYBPC2 and MYBPC3, MYBPC1 is heavily spliced, giving rise to 14 variants in humans encoding proteins between 126 and 131.5 kDa. The sMyBP-C variants are coexpressed in variable amounts and in combination in both slow and fast twitch skeletal muscles where they may coexist with fMyBP-C (25). Given the molecular complexity of MYBPC1 (1), its early expression during fetal development preceding that of MYBPC2 (29), and the neonatal lethality of patients having a recessive nonsense mutation (27), it becomes apparent that sMyBP-C is essential for skeletal muscle structure and function. In agreement with these results, a sMyBP-C knockout model is postnatally lethal, further highlighting the pathophysiological importance of sMyBP-C and distinguishing it from the cardiac and fast isoforms (30).

Herein we report on our effort to delineate the structural and functional roles of sMyBP-C in adult skeletal muscles in vivo. To this end, we used in vivo gene transfer (IVGT) combined with electroporation to introduce a clustered regularly interspaced short palindromic repeat (CRISPR) plasmid that specifically targets sMyBP-C. We show that depletion of sMyBP-C from mouse flexor digitorum brevis (FDB) and lumbricalis muscles resulted in decreased expression and impaired organization of myosin thick filaments. Consistent with this major structural alteration, sMyBP-C–depleted muscles exhibited impaired contractile kinetics under no-load conditions and reduced force production and delayed contractile kinetics after twitch or tetanic stimulation. Taken together, our studies demonstrate the crucial role of sMyBP-C in thick filament organization and stability, which appears to precede its regulatory capability to modulate crossbridge cycling.

MATERIALS AND METHODS

Mice

All animal work was conducted under protocols approved by the Institutional Animal Care and Use Committee of the University of Maryland, School of Medicine using C57Bl/6J mice from The Jackson Laboratory (Bar Harbor, ME, USA).

IVGT via electroporation and CRISPR plasmid injection

Male, 2-mo-old mice were anesthetized, and 20 μg of hyaluronidase (MilliporeSigma, St. Louis, MO, USA) was injected subcutaneously into each footpad. After 1 h, mice were anesthetized again, and 20 μg of CRISPR/Cas9 plasmid [CRISPR Universal Negative Control or custom-designed sMyBP-C CRISPR knockdown (KD) CGGAGTACTATGTGACAGCTGG; MilliporeSigma] was injected into the same footpad. Electrodes (Model 531, 2 Needle Array; BTX, Holliston, MA, USA) were inserted under the skin of the foot below the toes and at the heel. Pulses were administered at 150 V/cm (1 Hz, 20 ms pulses, 20 s total) by an ECM 830 Square Wave Electroporation System (BTX). Although CRISPR is a gene-editing method, because not all myofibers take up the indicated plasmid via IVGT/electroporation, complete knockout is not achieved. Therefore, we refer to our treated muscles as “sMyBP-C KD” rather than “sMyBP-C knockout.” For all experiments, FDB and lumbricalis muscles were harvested 2 wk after IVGT/electroporation, when KD was experimentally determined to be optimal.

Antibodies

The antibodies used for Western blotting and immunofluorescence were as follows: rabbit polyclonal, actin (A2066; MilliporeSigma), sMyBP-C (SAB3501005; MilliporeSigma), and fMyBP-C (PAB19214; Abnova, Walnut, CA, USA); mouse monoclonal, myosin (fast skeletal, M1570; MilliporeSigma), myosin (slow skeletal, clone NOQ7.5.4D; MilliporeSigma), GAPDH (G7895; MilliporeSigma), α-actinin (A7811; MilliporeSigma), and myomesin (mMac myomesin B4, Developmental Studies Hybridoma Bank; this antibody was developed by Jean-Claude Perriard and obtained from Developmental Studies Hybridoma Bank under the auspices of NICHD and maintained by the University of Iowa, Department of Biology, Iowa City, IA, USA). Phalloidin (A12379; Thermo Fisher Scientific, Waltham, MA, USA) was used to stain F-actin only in immunofluorescence studies.

Generation of protein lysates and Western blotting

Freshly isolated FDB and lumbricalis muscles were collected from mice and flash frozen in liquid nitrogen. FDB lysates were prepared with TissueLyser LT (Qiagen, Germantown, MD, USA) for 2 min at 50 Hz in modified NP-40 lysis buffer [10 mM NaPO₄ (pH 7.2), 2 mM EDTA, 10 mM NaN₃, 120 mM NaCl, 0.5% deoxycholate, 0.5% NP-40) supplemented with complete protease inhibitors (Roche, Indianapolis, IN, USA). Lumbricalis lysates were prepared in modified NP-40 lysis buffer that contained 1% NP-40 and 1% deoxycholate. After the addition of Nupage LDS sample buffer (Invitrogen, Carlsbad, CA, USA) and boiling at 95°C for 5 min, 20 μg of FDB lysates or 10 μg of lumbricalis lysates were fractionated by 4–12% SDS-PAGE, transferred to nitrocellulose, and probed with the indicated primary antibodies and the appropriate alkaline phosphatase–conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). Immunoreactive bands were visualized with the Tropix chemiluminescence detection kit (Thermo Fisher Scientific). Densitometry was performed with ImageJ software using replica blots of lysates from at least three independent experiments. Statistical significance was calculated with Student’s t test. All values are expressed as means ± sd.

Immunofluorescence staining and confocal microscopy

Mice were deeply anesthetized and euthanized by exsanguination, and FDB and lumbricalis muscles were rapidly harvested. For preparation of cell cultures, FDB muscles were placed into DMEM containing 4.5 g/L glucose, l-glutamine, sodium pyruvate (Mediatech, Inc., Manassas, VA, USA), 10% FBS, 1% penicillin-streptomycin, and 0.5 mg/ml collagenase A (Roche) for overnight digestion. The next day, single fibers were separated by trituration, fixed with 2% paraformaldehyde, permeabilized with 0.1% Triton-X, and blocked for several hours in PBS containing 1% bovine serum albumin (BSA), 1 mM NaN₃, and 1% goat serum at room temperature before immunolabeling with the indicated primary antibodies diluted in PBS/BSA/NaN₃ overnight at 4°C. Samples were washed with PBS/BSA/NaN₃, counterstained with secondary antibodies conjugated with Alexa-488 or Alexa-568 in PBS/BSA, mounted with Vectashield (Vector Laboratories, Burlingame, CA, USA), and analyzed under a confocal laser microscope (510; Carl Zeiss, Tarrytown, NY, USA) equipped with a ×63, 1.4 numerical aperture objective (Carl Zeiss) under the same laser settings.

Electron microscopy

Muscle specimens were fixed in 2% paraformaldehyde, 2.5% glutaraldehyde, and 0.1 M PIPES buffer (pH 7.4); washed; and postfixed with 1% osmium tetroxide and 1.5% potassium ferrocyanide in 0.1 M PIPES buffer for 1 h at 4°C. Specimens were then treated with 1% tannic acid in H₂O for 15 min, followed by en bloc staining with 1% (w/v) uranyl acetate and dehydration using 30, 50, 70, 90, and 100% ethanol in series. After dehydration, specimens were infiltrated and embedded in Araldite-Epoxy resin (Araldite, EMbed 812; Electron Microscopy Sciences, Hatfield, PA, USA) following the manufacturer’s recommendations. Ultrathin sections at ∼70 nm thickness were cut on a Leica UC6 ultramicrotome (Leica Microsystems, Bannockburn, IL, USA) and examined in an FEI Tecnai T12 transmission electron microscope (FEI, Hillsboro, OR, USA) operated at 80 kV. Images were acquired by using an AMT bottom mount CCD camera and AMT600 software (AMT, Woburn, MA, USA), and exposure was digitally adjusted. Sarcomere length was quantified with ImageJ software (National Institutes of Health, Bethesda, MD, USA). Three regions of interest were selected within 10 random fibers from three different muscles per group. Distances between adjacent z disks (troughs) were used to generate average sarcomere lengths. Statistical significance was determined by Student’s t test (P < 0.01), and values are expressed as means ± sd.

Contractile kinetics in single FDB fibers

Isolated FDB fibers were prepared as above. After trituration, isolated fibers were suspended in Hepes-buffered Ringer solution containing 140 NaCl mM, 4 KCl mM, 1 MgSO₄ mM, 5 NaHCO₃ mM, 10 mM glucose, and 10 mM HEPES (pH 7.3) at room temperature in a custom rotating glass-bottomed perfusion chamber (Four-Hour Day Foundation; Towson, MD, USA) and mounted over an inverted microscope (IX-70 ×40 H₂O 1.4NA objective; Olympus, Tokyo, Japan). Twitch contractions were elicited with field pulses (0.2 μs square pulse at 1 Hz) under no-load conditions, and sarcomere length was monitored and recorded using a high-speed video sarcomere length system (HSVL 901B; Aurora Scientific, Aurora, ON, Canada). Shortening and relaxation velocities were calculated as the first-derivative of recorded sarcomeric length (SL) dynamics (>500 Hz) across a 5-s period of contractions. Significance was determined by Mann-Whitney U test, and values are expressed as mean ± sem (control fibers, n = 44; KD fibers, n = 40).

Assay of FDB function in vivo

We developed an assay to assess FDB function in vivo using the Aurora 1300A in vivo system (Aurora Scientific). In isoflurane-anesthetized mice, the hindfoot was immobilized, and the toe was secured to the force transducer with a loop of 3-0 braided silk suture around the pad of the fourth digit. Using percutaneous nerve stimulation across the ankle, brief trains (200 ms) of pulses (0.2 ms, 30 Hz) elicited isometric contractions. The angle of FDB extension was adjusted to achieve maximal isometric force. After a demonstration that the prototypic force vs. stimulation frequency relationship (250 ms trains of 0.2 μs pulses at 1–300 Hz) was achieved with this preparation single twitch (1 Hz) and tetany (150 Hz) were used as our outcome variables. Statistical significance was determined by paired Student’s t test or Wilcoxon signed ranked test if normality test failed, n = 8 animals/group.

Assay of lumbricalis function ex vivo

The lumbricalis is comprised of four small skeletal muscles, deep in the foot, that are an established model for ex vivo skeletal muscle physiology. All four lumbricalis muscles were dissected and immersed in cold Ringer solution, and single lumbricalis samples were dissected from the medial side of digit 2 and 3. Single muscles were secured at each tendon with silk suture (6-0) between a fixed post and an Aurora 300C Dual model transducer mounted over an ex vivo perfusion bath (Aurora 1300A). Experiments were under temperature control (25°C) and constant perfusion with bicarbonate containing HEPES-buffered oxygenated Ringer solution (95% O₂, 5% CO₂) at a rate of one exchange per minute. Muscle activation was achieved by electrical stimulation via platinum plate electrodes placed parallel to each side of the muscle. Optimal length was determined by iteratively adjusting muscle length to achieve maximal twitch force (single 0.2-ms pulse), which was verified with a brief tetany (200 ms, 80 Hz). After a demonstration that the prototypic force vs. stimulation frequency relationship (250 ms trains of 0.2 μs pulses at 1–150 Hz) was achieved with this preparation, single twitch (1 Hz) and tetany (150 Hz) were used as our outcome variables. Statistical significance was determined by paired Student’s t test or Wilcoxon signed rank test if normality test failed, n = 4–6 animals/group.

RESULTS

Use of IVGT and CRISPR/Cas9 technology to knock down sMyBP-C in adult murine skeletal muscles

To study the roles of sMyBP-C in adult skeletal muscle, we used IVGT combined with electroporation and CRISPR/Cas9 technology to knock down its expression in FDB and lumbricalis muscles of young adult (∼2-mo-old) mice. The targeting sequence contained in the CRISPR sMyBP-C KD plasmid is located within Ig domain C2, which is constitutively expressed among all variants (Fig. 1A, red arrowhead). For ease of visualization, both the control and the sMyBP-C KD plasmids were tagged with GFP to allow identification of successfully transduced FDB (Fig. 1B, B′) and lumbricalis (Fig. 1C, C′) muscles and fibers.

Figure 1.

Structural organization of sMyBP-C and IVGT of CRISPR/Cas9 plasmids in mouse skeletal muscles. A) Schematic representation of variant 1, which is the most complete mammalian variant of sMyBP-C. Black and gray horizontal rectangles correspond to the Pro/Ala rich region and the M motif, and the white and blue boxes represent the Ig and FnIII domains, respectively. Colored vertical rectangles denote alternatively spliced insertions. The red arrow indicates the constitutively expressed region within Ig domain C2 that is targeted by the sMyBP-C KD plasmid. B–C′) The CRISPR/Cas9 plasmids were tagged with GFP for ease of visualization of transduced FDB (B, B′) and lumbricalis (C, C′) muscles.

Effects of sMyBP-C KD on thick and thin filaments

Immunoblot analysis confirmed that the sMyBP-C KD plasmid effectively reduced the levels of the protein in both FDB (Fig. 2, ∼70%) and lumbricalis (Supplemental Fig. 1A, B, ∼70%) muscles 2 wk after IVGT, whereas the control plasmid had no effect. To investigate the possible structural role of sMyBP-C, we examined the expression levels of additional myofibrillar proteins. We observed a consistent and significant reduction in the expression levels of thick filament proteins (fast and slow myosin and myomesin) in FDB (Fig. 2) and lumbricalis (Supplemental Fig. 1A, B). This was not the case, however, for fMyBP-C because its levels remained unaltered in FDB (Fig. 2) but were significantly increased in lumbricalis (Supplemental Fig. 1A, B). The expression of thin filament proteins, including actin and α-actinin, was not significantly altered in either FDB (Fig. 2) or lumbricalis (Supplemental Fig. 1A, B).

Figure 2.

Depletion of sMyBP-C from FDB muscle leads to reduced levels of thick filament proteins. A) Western blot analysis of protein lysates prepared from FDB muscles expressing control or sMyBP-C KD CRISPR plasmids 2 wk after IVGT; the presence of solid vertical black lines indicates that lanes were on the same gel but not continuous. B) Quantification of the percent expression of myofibrillar proteins in KD relative to control lysates after normalization to the expression levels of GAPDH was performed from two replica blots of at least three independent experiments. Statistical significance was calculated with a Student’s t test; all values are expressed as means ± sd. *P < 0.01, ^#P < 0.05.

We complemented these studies with assessment of the sarcomeric organization of treated muscles using immunofluorescence and confocal microscopy. After sMyBP-C KD in FDB (Fig. 3) and lumbricalis (Supplemental Fig. 2), immunolocalization revealed that residual sMyBP-C was diffusely distributed in cytoplasmic puncta, with occasional accumulation in striated-like structures reminiscent of A bands (Fig. 3A3–E3 and Supplemental Fig. 2A3–E3). In contrast, sMyBP-C was integrated normally into A bands in FDB (Fig. 3A1–E1) and lumbricalis (Supplemental Fig. 2A1–E1) treated with control plasmid. Similar to sMyBP-C, both fast (Fig. 3A4 and Supplemental Fig. 2A4) and slow (Fig. 3B4 and Supplemental Fig. 2B4) myosin isoforms exhibited minimal staining within the myoplasm and failed to assemble into periodic A bands in KD muscles when compared with controls (Fig. 3A, B and Supplemental Fig. 2A2, B2). The organization of myomesin in M bands was mostly unaffected in KD muscles (Fig. 3C4 and Supplemental Fig. 2C4), although its levels appeared consistently reduced compared with controls (Fig. 3C2 and Supplemental Fig. 2C2), in accordance with the immunoblotting experiments (Fig. 2 and Supplemental Fig. 1). The distribution of thin actin filaments in I-bands (Fig. 3D4 and Supplemental Fig. 2D4) and α-actinin in z disks (Fig. 3E4 and Supplemental Fig. 2E4) was indistinguishable from controls (Fig. 3D2 and Supplemental Fig. 2D2).

Figure 3.

sMyBP-C KD in FDB muscle results in impaired organization of thick filaments. Control and KD FDB fibers were costained for sMyBP-C (columns A1–E1 and A3–E3), the thick filament proteins fast skeletal myosin (A2, A4), slow skeletal myosin (B2, B4), and myomesin (C2, C4), and the thin filament proteins actin (D2, D4) and α-actinin (E2, E4). All proteins tested assumed their typical distribution in control fibers. In contrast, thick, but not thin, filament proteins exhibited impaired organization of different extents with fast and slow myosin failing to organize in A bands and myomesin staining being less intense. We used three different antibodies to immunostain control fibers for fMyBP-C along with various fixation, permeabilization, and antigen retrieval techniques, but the resulting staining was unreliable and inconsistent.

To gain further insight into the structural defects due to sMyBP-C depletion, we examined sarcomeric organization at the ultrastructural level using electron microscopy. Electron micrographs of longitudinal sections of FDB (Fig. 4A) and lumbricalis (Supplemental Fig. 3A) treated with control plasmid showed regularly organized sarcomeres and internal membranes. In contrast, longitudinal sections of FDB (Fig. 4B, C) and lumbricalis (Supplemental Fig. 3B) treated with sMyBP-C KD plasmid revealed the presence of distorted A bands containing a sparser distribution of misaligned myosin filaments and occasionally undefined M bands. Although I bands and z disks assumed their typical distribution in KD muscles, significant z-line streaming was observed in FDB muscles (Fig. 4D), in which z discs appeared misaligned and extended into the adjacent I bands (31). Consistent with these findings, trough-to-trough measurements corresponding to adjacent z disks showed significantly shorter sarcomere lengths in both FDB (2.15 vs. 1.69 μm; Fig. 4A–G) and lumbricalis (2.06 vs. 1.86 μm; Supplemental Fig. 3A–E) KD muscles compared with controls. We obtained similar findings when we performed these measurements in immunofluorescent images of control and KD fibers using α-actinin or myomesin to mark z disks or M bands, respectively (data not shown). In addition to longitudinal sections, we evaluated cross sections of KD muscles by electron microscopy. In agreement with our results in longitudinal sections, KD muscles contained fibers that displayed either minimal myosin filament levels (Fig. 4H, I and Supplemental Fig. 3F, G, single asterisk) or sparser and thinner myosin filaments that failed to organize in canonical hexagonal arrays (Fig. 4H, I and Supplemental Fig. 3F, G, double asterisk). KD muscles also exhibited altered internal membranes, which appeared fragmented and enlarged, suggestive of a myopathic phenotype; this morphologic alteration was more pronounced in KD FDB (Fig. 4I, arrowheads) than KD lumbricalis muscle.

Figure 4.

The ultrastructure of FDB muscle is altered after sMyBP-C KD. A–D) Evaluation of longitudinal sections of control (A) and KD (B, C) FDB muscles using electron microscopy revealed the presence of sparser and disorganized myosin filaments in the latter along with significant z-line streaming (D). E–G) Measurements of average sarcomere length determined by the distance of successive troughs representing neighboring z disks indicated significantly shorter sarcomere length in KD FDB muscles. H, I) Examination of cross sections showed normal hexagonal arrays of myosin filaments in control (H) but not KD (I) muscles, which displayed either minimal levels of myosin filaments (single asterisk) or sparse and thin (double asterisk) myosin filaments. KD muscles also contained fragmented and enlarged internal membranes (I, arrowheads). Significance was calculated via Student’s t test (P < 0.01); all values are expressed as means ± sd.

Impact of sMyBP-C KD on contractility

Given our evidence that sMyBP-C KD significantly affected the organization of thick filaments and the expression of several thick filament proteins, we determined the impact of these changes on function. We used field stimulation (0.2 μs square pulse) in single isolated FDB myofibers to elicit single action potential–driven contractions under no-load conditions. With this approach, we revealed that sMyBP-C KD significantly decreased the rates of contraction and relaxation (Fig. 5A, B). Furthermore, we developed an assay to assess the functional impact of sMyBP-C KD in FDB muscles contracting in vivo. To this end, we immobilized the hind foot of an anesthetized mouse and secured the toe to a force transducer with a loop of 3-0 braided silk suture around the pad of the fourth digit (Fig. 5C). Using percutaneous stimulation across the ankle, we delivered brief (200 ms) trains of pulses (0.2 μs, 30 Hz) to interactively determine the optimal angle of FDB extension to generate the maximal isometric force. After demonstrating that the prototypic force vs. stimulation frequency relationship (250 ms trains of 0.2 μs pulses at 1–300 Hz) could be achieved (Fig. 5D), we collected a single twitch (1 Hz) and tetany (150 Hz) as outcome variables. sMyBP-C KD significantly decreased the peak isometric force (Fig. 5E, F), the peak rate of contraction (Fig. 5E′, F′), and the peak rate of relaxation under twitch stimulation (Fig. 5E″); however, the peak rate of relaxation was not significantly affected under tetanic stimulation (Fig. 5F″, P = 0.604). To benchmark these effects against an established whole muscle contractility assay, we examined the lumbricalis muscle ex vivo. We observed a qualitatively similar force vs. frequency relationship (250 ms trains of 0.2 μs pulses at 1–150 Hz; Fig. 6A) to that in the FDB in vivo. Again, using the twitch (1 Hz) and tetanic (150 Hz) stimulations, we show a significant decrease in the peak force (Fig. 6B, C) and the peak rate of contraction (Fig. 6B′, C′), with no difference in the peak rate of relaxation (Fig. 6B″, P = 0.375; Fig. 6C″, P = 0.336) in sMyBP-C KD muscles vs. controls. The qualitative agreement of the in vivo and ex vivo assays validates our new in vivo functional assay and strengthens the conclusion that sMyBP-C KD affects contractility.

Figure 5.

sMyBP-C KD in FDB muscle results in reduced contractile kinetics and force production in vitro and in vivo. A, B) Sarcomere shortening of isolated, single FDB fibers was assayed under single action potential stimulation (1 Hz; field stimulation) under no-load conditions. A) Representative raw velocity measures of control (black) and KD (gray) fibers are shown as a second derivative of sarcomere length. B) sMyBP-C KD significantly decreased both contraction and relaxation velocities. Statistical significance was determined by Mann-Whitney U test (n = 44 for control, n = 40 for KD fibers). Values are expressed as means ± sem. *P < 0.01. C, D) At 2 wk after IVGT, mice were anesthetized, and the hindfoot was immobilized at the distal tarsometatarsus. C) The fourth digit was secured by a suture loop at the distal phalange to a tension transducer. D) Using percutaneous nerve stimulation across the ankle, a force vs. stimulation frequency relationship (250 ms trains at 1–300 Hz) was generated with representative raw traces for all stimulation frequencies shown. E–F″) sMyBP-C KD significantly decreased peak force production at both twitch (E) and tetanic (F) stimulations. Contraction velocity was also significantly reduced in KD muscles under both twitch (E′) and tetanic (F′) stimulations, whereas relaxation velocity was only significantly altered under twitch (E″) but not tetanic stimulation (F″; P = 0.604). Statistical significance was determined by paired Student’s t test or Wilcoxon signed ranked test if normality test failed. Values are expressed as means ± sem. *P < 0.01, ^#P < 0.05.

Figure 6.

sMyBP-C KD in lumbricalis muscle results in decreased force production and contractility kinetics ex vivo. Control and KD lumbricalis muscles were dissected 2 wk after IVGT and attached through the tendons to a force transducer and length controller to generate a force vs. stimulation frequency relationship (250 ms trains at 1–150 Hz). A) Representative raw traces for all stimulation frequencies are shown. B–C″) KD muscles exhibited significantly decreased peak force production at both twitch (B) and tetanic (C) stimulations. Contraction velocity was also significantly reduced in KD muscles at both twitch (B′) and tetanic (C″) stimulations. However, relaxation velocity was unaltered in KD lumbricalis muscles (B″, P = 0.375; C″, P = 0.336). Statistical significance was determined by paired Student’s t test or Wilcoxon signed rank test if normality test failed. All values are expressed as means ± sem. *P < 0.01, ^#P < 0.05.

DISCUSSION

Using IVGT and CRISPR/Cas9 technology, we knocked down all known sMyBP-C variants in two different adult skeletal muscles. Our findings show that sMyBP-C KD has structural and regulatory effects, manifested as decreased levels of thick filament proteins, distorted A bands, reduced sarcomere length, impaired force production, and delayed contractility kinetics. Our studies are therefore the first to demonstrate that, in addition to playing key roles in regulating crossbridge cycling (5, 32), sMyBP-C is essential in maintaining the regular organization and levels of thick filaments at maturity, which appears to be a unique property of the slow isoform not shared by the cardiac (14) and fast skeletal (33) proteins.

Sarcomeric assembly is a complex and highly regulated process relying on the blueprint properties of template and scaffolding proteins (34, 35). Earlier work has shown that a multiprotein complex consisting of the COOH-terminus of titin, myomesin, M protein, and obscurin aids the integration of myosin filaments into regular A bands with their rod domains occupying M bands (34, 36). sMyBP-C expression is concurrent with the appearance of these proteins, approximately at gestational d 14 during mouse embryogenesis, preceding the organization of myosin filaments into A bands (29, 37, 38). Although the contribution of sMyBP-C in thick filament assembly during embryogenesis has not been interrogated, its direct binding to titin (39) and obscurin (40) suggests that it may also assist in the incorporation of myosin into A bands. This notion is in agreement with the neonatal lethal phenotype associated with the R318Stop recessive mutation present in Ig domain C2 of sMyBP-C that has been causatively linked to lethal congenital contracture syndrome type-4 (27). Given that heterozygous lethal congenital contracture syndrome type-4 carriers are asymptomatic, it is highly possible that the R318Stop mutation results in a complete lack of sMyBP-C expression in homozygous patients rather than a truncated (yet stable) poisonous protein. Thus, depletion of sMyBP-C from skeletal muscles may be incompatible with life, signifying its crucial role in muscle development. Consistent with this, deletion of sMyBP-C in a mouse model is postnatally lethal (30).

sMyBP-C KD in adult skeletal muscles led to markedly reduced levels of both the slow and fast myosin isoforms, with residual proteins failing to remain organized in A bands. Similar to myosin, the levels of myomesin were significantly decreased in sMyBP-C–depleted muscles, albeit to a lesser extent. However, contrary to myosin, residual myomesin maintained its typical distribution at M bands, which appeared undisturbed. This milder effect is consistent with the lack of a direct interaction between sMyBP-C and myomesin and the assembly of M bands prior to A bands (40). Examination of the levels of fMyBP-C revealed no alteration in KD FDB muscle but considerably increased expression in KD lumbricalis muscle, possibly due to a compensatory response. We were unable to examine the subcellular localization of fMyBP-C due to technical issues (unpublished observations). However, given the distorted A bands that we observed, we presume that the protein might associate with filamentous or striated-like structures, which would lack the periodicity of typical A bands. Our measurements demonstrating reduced sarcomere length in KD muscles are in agreement with this speculation. In contrast to thick filament proteins, the expression levels and distribution of thin filament proteins were indistinguishable between KD and control muscles. Thus, our findings indicate that, in addition to its role in the regulation of actomyosin binding and sliding (5, 32), sMyBP-C contributes to thick, but not thin, filament stabilization and maintenance at maturity.

This structural role appears to be unique to sMyBP-C among the MyBP-C isoforms. Adult cMyBP-C–null hearts from two independent animal models exhibited gross morphologic alterations consistent with the development of hypertrophy, including fibrosis and occasional myocyte disarray (14, 41). Although occasional z-disk streaming was reported (14), these changes were independent of alterations in sarcomeric assembly during embryogenesis, with adult hearts having regularly spaced sarcomeres. Both cMyBP-C null models displayed diastolic dysfunction characterized by impaired relaxation (14, 41), whereas one of the two models exhibited reduced myofilament Ca²⁺ sensitivity (14). Moreover, skinned myocardium from a third cMyBP-C–null model exhibited an accelerated rate of force decay and delayed force transient compared with wild type (42). Thus, in contrast to sMyBP-C, which appears to assume both structural and regulatory roles, cMyBP-C mainly has a regulatory role in crossbridge cycling modulation.

In further support of the unique structural role of sMyBP-C in mammals, extensor digitorum longus and soleus muscles from a fMyBP-C KO mouse model did not display morphologic or structural alterations but exhibited increased Ca²⁺ sensitivity of force development (33). In contrast to the mouse model, however, KD of fMyBP-C in zebrafish led to the development of a myopathic phenotype characterized by shorter sarcomeres that generated reduced force and slower rates of contraction and relaxation (21). The development of disparate phenotypes by the two models is surprising, especially given that in the zebrafish model only the dominating gene, MYBPC-2B, was knocked down by 50%, whereas in the mouse model the single MYBPC2 gene was completely ablated. The levels of MYBPC1 were unaltered in the zebrafish model, indicating the absence of a compensatory response. Although it is unknown if the levels of MYBPC1 are altered in the fMyBP-C knockout mouse model, it is possible that MYBPC1 is up-regulated, leading to the development of a mild phenotype. Conversely, it is likely that, similar to cMyBP-C, fMyBP-C may primarily play a regulatory role in mammals. This would be consistent with the delayed appearance of fMyBP-C during mouse embryogenesis at gestational d 18 that follows sMyBP-C expression and myosin incorporation into A bands (29).

Few studies have examined the impact of skeletal MyBP-C on crossbridge regulation. In the early 1990s, Hofmann et al. (15) reported that addition of purified MyBP-C from rabbit psoas muscle to skinned myofibers slowed down the shortening velocity of actomyosin crossbridges in vitro. Moreover, our laboratory demonstrated that addition of the NH₂ terminus of sMyBP-C reduced the sliding velocity of actin filaments past myosin heads in vitro in a variant-specific manner (5), and Lin et al. (32) showed that sMyBP-C activates thin filament motility at low [Ca²⁺]. Because its NH₂ terminus undergoes complex phosphorylation (11, 25, 32, 43), it is highly likely that its activities are modulated via phosphorylation similar to cMyBP-C (44, 45).

Our findings show that sMyBP-C KD reduces force production and slows down contractile kinetics, which contradicts previous in vitro studies with isolated proteins. A major caveat of in vitro studies is the use of either purified or recombinant proteins outside the context of the cell. Given the drastic decrease in the levels of fast and slow myosin due to sMyBP-C KD and that alterations in protein stoichiometry and organization are not captured in in vitro assays, a meaningful comparison of these results in not possible. We therefore postulate that, in the context of our in vitro single cell and in vivo/ex vivo whole muscle studies, the presence of a significantly smaller number of functional contractile units may underscore the deficiencies in force production and contractility in sMyBP-C KD muscles.

Taken together, our studies demonstrate that sMyBP-C may be unique within the MyBP-C family not only because of its molecular complexity and multifaceted character but also because of its essential structural role in thick filament organization and maintenance, which appears to precede its regulatory role in modulating crossbridge cycling. The obvious question is what makes sMyBP-C unique among the MyBP-C isoforms. Earlier work from our group has shown that sMyBP-C undergoes extensive exon shuffling within the NH₂ and COOH termini, resulting in the inclusion or exclusion of small amino acid segments (1, 46). It is therefore plausible that the presence or absence of these inserts may alter the binding affinity of the resulting variants to myosin and actin and/or mediate unique binding interactions. Consistent with this, sMyBP-C variants exhibit differential abilities to bind myosin and actin and to regulate actomyosin sliding in vitro (5). Moreover, select sMyBP-C variants (v1/v2) preferentially target to the periphery of the M band, instead of the C zone, where they bind obscurin via their COOH terminus (39). Neither cMyBP-C nor fMyBP-C bind obscurin via their respective COOH termini in vitro (39). Because combinations of sMyBP-C variants are coexpressed within an individual muscle, myofiber, and possibly sarcomere (1), it is likely that some variants have structural or stabilizing roles, depending on their location and binding interactions, whereas others may have regulatory roles. Further work is needed to dissect the functional complexity of the sMyBP-C subfamily, focusing on the spatiotemporal expression profile, regulation, binding interactions, and roles of individual variants in slow- and fast-twitch muscles using sophisticated biochemical, biophysical, and targeted genetic approaches.

Supplementary Material

This article includes supplemental data. Please visit http://www.fasebj.org to obtain this information.

Glossary

BSA

bovine serum albumin

cMyBP-C

cardiac myosin binding protein-C

CRISPR

clustered regularly interspaced short palindromic repeats

FDB

flexor digitorum brevis

fMyBP-C

fast myosin binding protein-C

IVGT

in vivo gene transfer

KD

knockdown

MyBP-C

myosin binding protein-C

S2

subfragment 2

sMyBP-C

slow myosin binding protein-C

AUTHOR CONTRIBUTIONS

J. Geist designed and performed the experiments and wrote and edited the manuscript; C. W. Ward designed and performed the experiments and edited the manuscript; and A. Kontrogianni-Konstantopoulos designed the experiments and wrote and edited the manuscript.