Desensitizing mouse cardiac troponin C to calcium converts slow muscle towards a fast muscle phenotype

Abstract

Key points

• The Ca²⁺‐desensitizing D73N mutation in slow skeletal/cardiac troponin C caused dilatated cardiomyopathy in mice, but the consequences of this mutation in skeletal muscle were not known.

• The D73N mutation led to a rightward shift in the force versus pCa (‐log [Ca]) relationship in slow‐twitch mouse fibres.

• The D73N mutation led to a rightward shift in the force–stimulation frequency relationship and reduced fatigue resistance of mouse soleus muscle.

• The D73N mutation led to reduced cross‐sectional area of slow‐twitch fibres in mouse soleus muscle without affecting fibre type composition of the muscle.

• The D73N mutation resulted in significantly shorter times to peak force and to relaxation during isometric twitches and tetani in mouse soleus muscle.

• The D73N mutation led to major changes in physiological properties of mouse soleus muscle, converting slow muscle toward a fast muscle phenotype.

Abstract

The missense mutation, D73N, in mouse cardiac troponin C has a profound impact on cardiac function, mediated by a decreased myofilament Ca²⁺ sensitivity. Mammalian cardiac muscle and slow skeletal muscle normally share expression of the same troponin C isoform. Therefore, the objective of this study was to determine the consequences of the D73N mutation in skeletal muscle, as a potential mechanism that contributes to the morbidity associated with heart failure or other conditions in which Ca²⁺ sensitivity might be altered. Effects of the D73N mutation on physiological properties of mouse soleus muscle, in which slow‐twitch fibres are prevalent, were examined. The mutation resulted in a rightward shift of the force–stimulation frequency relationship, and significantly faster kinetics of isometric twitches and tetani in isolated soleus muscle. Furthermore, soleus muscles from D73N mice underwent a significantly greater reduction in force during a fatigue test. The mutation significantly reduced slow fibre mean cross‐sectional area without affecting soleus fibre type composition. The effects of the mutation on Ca²⁺ sensitivity of force development in soleus skinned slow and fast fibres were also examined. As expected, the D73N mutation did not affect the Ca²⁺ sensitivity of force development in fast fibres but resulted in substantially decreased Ca²⁺ sensitivity in slow fibres. The results demonstrate that a point mutation in a single constituent of myofilaments (slow/cardiac troponin C) led to major changes in physiological properties of skeletal muscle and converted slow muscle toward a fast muscle phenotype with reduced fatigue resistance and Ca²⁺ sensitivity of force generation.

Key points

• The Ca²⁺‐desensitizing D73N mutation in slow skeletal/cardiac troponin C caused dilatated cardiomyopathy in mice, but the consequences of this mutation in skeletal muscle were not known.

• The D73N mutation led to a rightward shift in the force versus pCa (‐log [Ca]) relationship in slow‐twitch mouse fibres.

• The D73N mutation led to a rightward shift in the force–stimulation frequency relationship and reduced fatigue resistance of mouse soleus muscle.

• The D73N mutation led to reduced cross‐sectional area of slow‐twitch fibres in mouse soleus muscle without affecting fibre type composition of the muscle.

• The D73N mutation resulted in significantly shorter times to peak force and to relaxation during isometric twitches and tetani in mouse soleus muscle.

• The D73N mutation led to major changes in physiological properties of mouse soleus muscle, converting slow muscle toward a fast muscle phenotype.

Introduction

Skeletal muscle plays a crucial role in multiple functions, including movement, maintenance of posture and respiration, and has a large impact on energy metabolism (for review see Gehlert et al. 2015; Argiles et al. 2016). Vertebrate skeletal muscles consist of heterogeneous mixtures of fibres with differences in contractile and metabolic properties (for reviews see Schiaffino & Reggiani, 2011; Luna et al. 2015). Skeletal muscle fibres can be divided into two major groups: slow‐twitch (type I) and fast‐twitch (type II) (reviewed by Schiaffino & Reggiani, 2011). Fast‐twitch fibres can be further divided into several subgroups (IIa, IIb and IIx), based primarily upon myosin heavy chain (MHC) isoform composition. Compared to fast‐twitch fibres, slow‐twitch fibres contract and relax more slowly, and are more resistant to fatigue (reviewed by Baylor & Hollingworth, 2012). The heterogeneous fibre type composition of skeletal muscles supports a wide range of functional demands, from maintenance of posture for prolonged periods to ballistic, high‐power physical activities. In addition, skeletal muscle can adapt to changes in functional demands by changing fibre type composition and/or fibre size (see reviews by Harridge, 2007; Blaauw et al. 2013).

Differences in contractile kinetics between slow‐twitch and fast‐twitch fibres are largely due to differences in biochemical properties of myosin isoforms expressed in these fibres (reviewed by Baylor & Hollingworth, 2012). Some of the contractile differences between slow‐twitch and fast‐twitch fibres could also result from differences in troponin (Tn) isoforms expressed in these fibres (reviewed by Baylor & Hollingworth, 2012). Contraction of striated muscle is triggered by binding of Ca²⁺ ions to the regulatory N‐domain of troponin C (TnC), a member of an EF‐hand family of Ca²⁺‐binding proteins (see reviews by Farah & Reinach, 1995; Filatov et al. 1999; Li & Hwang, 2015). Mammalian slow skeletal muscle and cardiac muscle normally express the same isoform of troponin C (ss/cTnC), while a different isoform (fsTnC) is expressed in fast skeletal muscle (reviewed by Sheng & Jin, 2014). While ss/cTnC and fsTnC each have four EF‐hands, the first EF‐hand of ss/cTnC lost the ability to bind Ca²⁺ due to several evolutionary residue substitutions (van Eerd & Takahshi, 1976). Thus, Ca²⁺ binding and exchange with only the second EF‐hand of ss/cTnC regulates contractility of slow skeletal and cardiac muscle. TnC regulates muscle contraction as a constituent of the thin filament‐bound Tn complex, consisting also of an inhibitory subunit, troponin I (TnI), and a tropomyosin‐binding subunit, troponin T (TnT) (reviewed by Filatov et al. 1999; Gordon et al. 2000). Separate genes encode cardiac, fast skeletal and slow skeletal isoforms of TnI and TnT (reviewed by Sheng & Jin, 2014).

The overriding objective of our work is to gain a deeper insight into the role of Ca²⁺ binding and exchange with TnC in the regulation of striated muscle contractility. Recently, we evaluated the effects of desensitizing the regulatory N‐domain of ss/cTnC to Ca²⁺ (by the D73N mutation) on cardiac muscle function (McConnell et al. 2015). The D73N mutation is not known to be associated with striated muscle diseases in humans, and was rationally designed by us to test whether desensitizing ss/cTnC to Ca²⁺ can trigger the pathogenesis of dilatated cardiomyopathy (DCM). Our results indicated that heterozygous knock‐in mice harbouring the D73N mutation in ss/cTnC develop cardiac muscle abnormalities similar to that of human patients with DCM, including higher mortality, increased heart weight‐to‐body weight ratio, increased left ventricular internal dimensions with thinner walls, reduced ejection fraction and fractional shortening, and electrophysiological abnormalities (McConnell et al. 2015). The consequences of the D73N mutation were consistent with results from a number of studies showing that desensitizing cardiac muscle to Ca²⁺ could be detrimental to the physiological function of the heart (see reviews by Kimura, 2010; Willott et al. 2010; Spudich, 2014). However, the physiological effects of desensitizing skeletal muscle to Ca²⁺ remain incompletely understood. The objective of our current study was to test the hypothesis that desensitizing the regulatory N‐domain of ss/cTnC to Ca²⁺ leads to alterations in the contractile properties of skeletal muscle, as potential contributions to morbidity when cardiac function is altered due to modified interactions between Ca²⁺ and Tn.

Methods

Ethical approval

All of the mice that were used in this study were provided standard daily care and monitoring to avoid unnecessary pain or suffering and were provided with unrestricted access to food and water at all times. Mice were killed by inhalation of carbon dioxide, followed by rapid removal of the heart. The care and use of all of the animals were in accordance with a protocol (2014A00000070‐R1) which was approved by the Institutional Animal Care and Use Committee at The Ohio State University and were consistent with the Guide for the Care and Use of Laboratory Animals which was released by the National Academy of Sciences (US) in 2011. All of the authors understand the ethical principles under which the journal operates and that their work complies with the animal ethics checklist.

Animals

Adult male mice, 8–9 weeks old, were used for this study. C57BL/6J mice served as controls. The targeting vector and the D73N knock‐in mice were generated by the Gene Targeting Mouse Service Core at the University of Cincinnati (Cincinnati, OH, USA). The D73N knock‐in mice were back‐crossed into the C57BL/6J background, using the Speed Congenic Service at the Jackson Laboratory (Bar Harbor, ME, USA). Previously, we determined that none of the homozygous D73N knock‐in mice survived longer than 1 day (McConnell et al. 2015). Thus, all of the D73N knock‐in mice in this study were heterozygous. Mice were genotyped as described previously (McConnell et al. 2015).

RT‐PCR analysis

Total RNA was extracted from soleus muscle from either leg of four WT and four D73N mice using RNeasy Fibrous Tissue Mini Kit (Qiagen, Venlo, the Netherlands) according to the manufacturer's instructions. Fragments of cDNA were amplified from total RNA using Superscript III One Step RT PCR system (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The ratio of ss/cTnC transcript with the D73N mutation to total ss/cTnC transcript in the soleus muscle of D73N mice was estimated by RT‐PCR using TNNC1 sequence‐specific primers followed by digestion of RT‐PCR product by the restriction enzyme KpnI, as previously described for cardiac muscle (McConnell et al. 2015).

Measurements of isolated muscle contractile properties

The soleus muscle from one leg of eight WT mice and eight D73N mice was isolated immediately following death of the animal and was placed in oxygenated (95% O₂/5% CO₂) Krebs‐Ringer's solution (137 mm NaCl, 5 mm KCl, 13 mm NaHCO₃, 1.8 mm KH₂PO₄, 2 mm CaCl₂, 11 mm glucose, 1 mm MgSO₄ and 0.025 mm tubocurarine chloride). All chemicals were purchased from Sigma‐Aldrich Inc. (St. Louis, MO, USA). Rigid stainless steel wires were glued (Loctite Superglue, Henkel Consumer Adhesives, Inc., Avon, OH, USA) to the distal and proximal tendons of the soleus. The muscle was connected, with one of the wires, to a fixed post in the tissue chamber of a model 801C in vitro muscle apparatus (Aurora Scientific, Inc., Aurora, Canada) and, with the other wire, to a 300B dual‐mode muscle lever system (Aurora). The lever system was used in the length‐control (isometric) mode and the force output from this system was analysed to measure muscle contractile properties. Force records were captured and analysed using DASYLab software (DASYLab 5.5, IO Tech, Cleveland, OH, USA) at a sampling rate of 10 kHz. All measurements were conducted at room temperature (mean 22.2°C; range 21.4–24.1°C). The muscle was then stimulated with a series of single stimuli and the muscle length was varied until the peak twitch force was maximal. The muscle was maintained at this length for all of the measurements. Length was measured with a Vernier caliper. The pulse duration for all stimulations was 0.1 ms. The twitch stimulus strength was initially varied to ensure that the same supramaximal intensity was used for all of the measurements in each muscle. Stimulation was provided by an Aurora Scientific High‐Power Bi‐Phase Current Stimulator (model 701B) which was gated with a Grass Instruments S48 stimulator (Grass Medical Instruments Co., West Warwick, RI, USA). Peak force, the time to peak force, the time to one‐half relaxation and the time to 90% of complete relaxation were then measured in a series of five twitches with 2–3 min between stimulations. The measured values of these parameters were averaged for each muscle. The muscle was then stimulated with a series of stimuli applied at multiple increasing frequencies, beginning with 10 Hz, to establish the force/frequency relationship. The stimulus duration for each contraction in this series was manually maintained until steady peak force was generated. The lowest frequency at which maximal force was generated was then used to stimulate the muscle tetanically. The tetanus stimulus duration was also maintained until steady peak force was generated (typically ∼500 ms). The maximal tetanic force and the times to one‐half relaxation and to 90% relaxation were measured for each of five tetani, spaced 5 min apart, and the average values were calculated. The muscle was allowed to rest for 5 min before starting a fatigue test. The muscle was stimulated with trains of stimuli, each at 70 Hz for 400 ms each second, for 2 min. The peak forces during the initial stimulus train and the final stimulus train were measured and a fatigue index was calculated as Force_final/Force_initial. The muscle was detached from the motor and transducer and the tendons were removed. Wet muscle mass was determined after gentle blotting. Peak forces were normalized with muscle cross‐sectional area, calculated as mass/length, assuming a tissue density of 1.00 g cm⁻³. The muscle was frozen until analysis of MHC and TnT isoform composition.

Measurements of isolated single fibre contractile properties

The preparation and isolation of skinned muscle fibres and the measurements of contractile properties were identical to those described previously (Reiser et al. 2013; McConnell et al. 2015). Single fibres from the soleus muscle were used for all measurements. The bundles from which single fibres were isolated were soaked in 1% Triton X‐100 for 30 min prior to isolation. All measurements were made at 15°C. Fibres were randomly selected for measurements and were subsequently identified, as fast‐type or slow‐type, using gel electrophoresis (described below). The resting sarcomere length was set to 2.40–2.50 μm. The maximal force‐generating ability, i.e. peak force generated in maximally activating solution (pCa 4.0), normalized with fibre cross‐sectional area, and the force/pCa relationship were determined in each fibre. Each fibre was exposed to a series of solutions with varying pCa and the total force generated in solution was recorded. Resting force was measured in a solution with pCa 9.0, and was subtracted from the total force in each activating solution to determine the active force generated at each pCa. These forces were normalized with the peak force generated at pCa 4.0. The total numbers of fibres studied were 16 WT slow, 14 WT fast, 16 D73N slow and 18 D73N fast. These were isolated from eight WT mice and seven D73N mice. All of the fibres in each of these groups were pooled for statistical analysis. The force/pCa data were fit with the logistic sigmoid function, mathematically equivalent to the Hill equation.

Gel electrophoresis and immunoblots

The MHC isoform composition of single fibres and of whole muscles in which contractile properties were measured was determined by gel electrophoresis (7% acrylamide, 50:1 acrylamide/bis‐acrylamide crosslinking, and 30% glycerol), as previously described (Young et al. 2016). One soleus muscle was isolated from three additional WT mice and three additional D73N mice and these muscles were included in the MHC isoform analysis. The same 22 samples were also loaded on gels (12% acrylamide, 200:1 acrylamide/bis‐acrylamide crosslinking, without glycerol) that were silver‐stained or were transferred for troponin T (TnT) western blots, as previously described (Brundage et al. 2015). The blots were probed with anti‐slow/cardiac TnT CT3 antibody from the Developmental Studies Hybridoma Bank (DSHB) (DSHB Cat. No. ct3, RRID:AB_528495; hybridoma culture supernatant, diluted 1:200) and with anti‐fast TnT JLT12 antibody (DSHB Cat. No. JLT12, RRID:AB_2618103; hybridoma culture supernatant, diluted 1:1000). Quantification of bands on gels and on blots was performed with ImageJ software from the National Institutes of Health (available at https://imagej.nih.gov/ij/).

Immunohistochemistry

The right or left soleus muscle was isolated from four WT and five D73N mice, immediately following death of the animal, and placed in a small inverted tent of aluminium foil. The length of each muscle was set such that there was no slack. The end‐to‐end length of the muscle was measured with calipers prior to application of OCT embedding medium (Tissue‐Tek; Sakura Finetek USA, Torrance, CA, USA). Muscle length was re‐measured and it was determined that there was no significant change in muscle length associated with the embedding step. The muscles were stored at −80°C until sectioning. The muscles were transferred to a cryostat chamber on dry ice. Sections, 14 μm thick, were cut at −20°C and mounted on a glass slide. The sections were allowed to dry at room temperature for 20–30 min and were then blocked with 5% normal goat serum (005‐000‐121, Jackson ImmunoResearch, West Grove, PA, USA) for 30 min at room temperature. The sections were incubated with the following primary antibodies, diluted with PBS, at 4°C overnight: anti‐laminin antibody (Sigma‐Aldrich Cat. No. L9393, RRID:AB_477163; diluted 1:1000), anti‐slow myosin heavy chain (DSHB Cat. No. BA‐F8, RRID:AB_10572253; hybridoma culture supernatant diluted 1:50) and anti‐fast MHC IIA (DSHB Cat. No. SC‐71, RRID:AB_2147165; hybridoma culture supernatant diluted 1:600). The sections were then washed three times (5 min each wash) in PBS. The sections were incubated with the following secondary antibodies diluted in PBS at 4°C overnight: Alexa 647 (far red, blue pseudocolour) goat anti‐mouse IgG2b (A‐21242, ThermoFisher Scientific) for slow MHC, diluted 1:500; Alexa 488 (green) goat anti‐mouse IgG1 (A‐21121, ThermoFisher Scientific) for MHC‐IIA, 1:500; and Alexa 594 (red) donkey anti‐rabbit (A21207, ThermoFisher Scientific) for laminin, 1:500. The sections were washed with PBS three times (5 min each wash) and covered with a cover slip (Vectashield H‐1400, Vector Laboratories, Inc., Burlingame, CA, USA). The sections were viewed with a fluorescence Olympus FV 1000 Filter Confocal system and images were obtained with a 20× objective for most sections. Sections from one mouse were viewed and imaged with a 10× objective. Image resolution was 512 × 512 pixels (0.62 μm/pixel with the 20× objective and 1.24 μm/pixel with the 10× objective). Quantification of slow and fast fibre cross‐sectional area was performed with ImageJ. Fibres that were only partially included on the edges of the image were not included in the analysis. Each fibre was manually outlined in ImageJ and the area was determined. The laminin stain was used to confirm fibre boundaries. All of the fibres that were stained blue were considered to be slow type. All other fibres were considered to be fast type and the vast majority of these were stained green, indicating the expression of MHC‐IIA. Some fibres were not stained and were assumed to be fast fibres expressing MHC‐IID/X and/or MHC‐IIB. These were included in the analysis and were pooled as fast fibres, along with the IIA fibres. Three sections were analysed for each muscle. The average number of fibres analysed per image was 67 with the 20× objective and 276 with the 10× objective.

Tests for statistical significance

Analysis of variance and the Bonferroni comparison were used when comparing more than two groups, with the level of significance (α) set at 0.05. Student's t test was used to test the significance of differences between mean values when only two groups were compared, also with α set at 0.05. Values are expressed as mean ± SEM.

Results

Physical characteristics of WT and D73N mice

Physical characteristics (body mass, heart mass and soleus mass) of WT and D73N mice are summarized in Table 1. Body mass and soleus muscle mass did not differ between WT and D73N mice. Consistent with the results from our previous study (McConnell et al. 2015), mean heart mass was significantly greater (46%) in D73N mice than in WT mice. The mean heart mass/body mass ratio was also significantly greater (48%) in D73N mice. The amount of ss/cTnC transcript with the D73N mutation in soleus muscle of D73N mice was estimated at 31 ± 4% of total ss/cTnC transcript (data not shown), similar to the value estimated for left ventricular muscle (McConnell et al. 2015).

Table 1.

Physical characteristics of the wild‐type and D73N mice

	WT	D73N
Body mass (g)	22.0 ± 0.4	21.6 ± 0.4
Soleus mass (mg)	8.4 ± 0.4	8.1 ± 0.6
Heart mass (mg)	122.3 ± 4.1	179.1 ± 5.8*
Heart mass/body mass	5.6 ± 0.2	8.3 ± 0.2*

^*Significant difference from the WT value.

Soleus isometric contractile properties

Twitch and tetanus contractile parameters (illustrated in Fig. 1), as well as results from the fatigue test, are presented in Table 2. Peak twitch and tetanus forces, normalized with muscle cross‐sectional area, did not differ between WT and D73N mice. All measured kinetic properties of the twitch (time to peak force and the times to 50% and 90% relaxation) and tetanus (the times to 50% and 90% relaxation) were significantly shorter in the D73N soleus. The tetanus force/frequency relationship was significantly different between the WT and D73N soleus, with the D73N soleus generating less force when stimulated at frequencies from 10 to 50 Hz (Fig. 2). There was a greater reduction in peak force generation during the fatigue test (examples shown in Fig. 3) in the D73N soleus, indicating lower fatigue resistance. There was a large elevation in baseline force during the fatigue test in the WT soleus, due to incomplete relaxation between stimulus trains (Fig. 3). The magnitude of this elevation (Table 2) was significantly lower in the D73N soleus, as in WT fast‐twitch extensor digitorum longus muscle (data not shown).

Figure 1. Examples of twitch and tension records and their analysis.

Left, illustration of the measurements of the time to peak force (T_peak), time to 50% relaxation (T_0.5R) and time to 90% relaxation (T_0.9R) during isometric twitches. Inset: examples of isometric twitches in D73N and WT soleus muscles. Right, illustration of the measurements of the time to 50% relaxation (T_0.5R) and time to 90% relaxation (T_0.9R) during isometric tetani.

Table 2.

Contractile properties of wild‐type and D73N soleus muscles

	WT	D73N
Peak twitch force (kN/m2)	91.4 ± 9.0	76.1 ± 6.0
Twitch, time to peak (s)	0.057 ± 0.002	0.047 ± 0.003*
Twitch, time to 50% relaxation (s)	0.101 ± 0.005	0.071 ± 0.004*
Twitch, time to 90% relaxation (s)	0.337 ± 0.031	0.163 ± 0.014*
Peak tetanus force (kN/m2)	313.2 ± 30.4	288.6 ± 13.5
Tetanus, time to 50% relaxation (s)	0.127 ± 0.005	0.099 ± 0.004*
Tetanus, time to 90% relaxation (s)	0.259 ± 0.014	0.166 ± 0.009*
Fatigue baseline shift (%)	12.1 ± 1.2	0.9 ± 0.4*
Fatigue Index (Forcefinal/Forceinitial)	0.38 ± 0.02	0.29 ± 0.02*

^*Significant difference from the WT value.

Figure 2. Force/frequency relationship in WT and D73N soleus muscles.

N = 8 in both groups. Force was normalized with the greatest force generated at any frequency for each muscle. D73N soleus muscles generated significantly less relative force from 10 to 50 Hz.

Figure 3. Examples of force records during fatigue tests in WT and D73N soleus muscles.

The muscle was stimulated at 70 Hz for 400 ms every second, for about 2 min. The initial peak force and the peak force at 2 min were measured. The fatigue index was calculated as Force_2 min./Force_initial.

Myosin heavy chain isoform composition

The amount of slow‐type MHC (MHC‐I) in the soleus, relative to total muscle MHC, differed significantly between WT mice (31.4 ± 2.6%) and D73N mice (20.6 ± 1.6%) (Fig. 4). Whereas the relative amount of total fast‐type MHC was correspondingly greater in the D73N mice, there were no significant differences between WT and D73N mice with respect to the relative amounts of individual fast MHC isoforms (IIA, IIB or IID/X).

Figure 4. Relative amount of slow‐type myosin heavy chain (MHC‐I) in WT and D73N soleus muscles.

The top panel shows the MHC region of a silver‐stained gel loaded with WT and D73N soleus homogenates. The bottom panel shows the mean relative amount of MHC‐I, as a per cent of total MHC in each sample. The mean (± SEM) values differ significantly.

Troponin T isoform composition

The same samples were run on 12% acrylamide gels to examine possible differences in proteins with lower molecular weight between WT and D73N soleus. No differences between groups were detected on silver‐stained gels (data not shown). Nevertheless, immunoblots for slow‐ and fast‐type TnT were run to assess whether there were any differences in TnT expression between the two groups that were not indicated on the silver‐stained gels (Fig. 5). Three slow‐type TnT isoforms were detected on immunoblots which were reacted with the anti‐slow/cardiac CT3 antibody (DSHB Cat. No. ct3, RRID:AB_528495) and all three isoforms were detected in WT and D73N soleus. There were very small, but significant, differences in the relative amounts of two of the bands that were recognized by this antibody (Fig. 5). Several prominent and several minor bands were consistently observed in the samples probed on immunoblots with the anti‐fast‐type TnT antibody, JLT12 (DSHB Cat. No. JLT12, RRID:AB_2618103). As with slow‐type TnT, there were several very small but statistically significant differences in the fast‐type TnT isoform composition of the soleus muscle between WT and D73N mice (Fig. 5).

Figure 5. Representative immunoblots probed with slow/cardiac TnT antibody (left) and anti‐fast TnT antibody (right).

The same samples were probed on both blots, in the same order from left to right. The lower panel represents quantification of all the blots. Black bars are WT and grey bars are D73N. Asterisks indicate significant differences between WT and D73N mice.

Soleus fibre type composition and fibre cross‐sectional area

Representative soleus muscle cross‐sections, from WT and D73N mice, which were stained to identify slow‐type and fast‐type fibres, are shown in Fig. 6. The fibre type composition of the soleus in WT and D73N mice did not differ, with slow fibres constituting about one‐third of the fibre population in both groups (Table 3). The mean cross‐sectional area of slow fibres was significantly lower in the D73N soleus, compared to the WT soleus. As a consequence of a lower mean cross‐sectional area of slow fibres and no difference in fibre type composition, the mean soleus muscle cross‐sectional area that was occupied by slow fibres was significantly lower in D73N mice. Whereas the mean cross‐sectional areas of WT slow and WT fast fibres did not differ from each other, D73N fast fibres were significantly larger (∼60%) than D73N slow fibres.

Figure 6. Representative cross‐sections of soleus muscles from WT (left) and D73N mutant (right) mice.

Green, fast fibres; blue, slow fibres; red, laminin.

Table 3.

Immunohistochemical results from soleus muscle of wild‐type and D73N mice

	WT	D73N
Slow fibres (%)	35.4 ± 1.6	34.3 ± 1.5
Slow fibre CSA (μm2)	1164 ± 130	697 ± 76*
Fast fibre CSA (μm2)	883 ± 77	1102 ± 126
Cumulative slow area (% of total area)	41.8 ± 2.4	23.5 ± 1.7*

^*Significant difference from the WT value.

Force/pCa relationship in soleus skinned fibres

Contractile properties of skinned single fibres are summarized in Table 4. The mean maximal force generating ability (P _o/CSA) of slow fibres from D73N mice was significantly lower than of slow fibres from WT mice. There was a significant difference in P _o/CSA between WT fast and slow fibres but not between D73N fast and slow fibres. The force/pCa relationship was different between WT and D73N slow fibres (Fig. 7). D73N slow fibres had a mean pCa₅₀ (the pCa at which 50% of maximal force is generated) that was significantly lower than that of WT slow fibres. That is, the D73N slow fibres had a significantly lower Ca²⁺ sensitivity. The mean pCa₅₀ did not differ between slow and fast fibres from WT mice, but it differed significantly between slow and fast fibres from D73N mice. The Hill coefficient (i.e. the steepness of the force/pCa curve and frequently interpreted as an index of the cooperativity of activation) was significantly greater in fast fibres, compared to slow fibres, in WT mice and in D73N mice. The mean Hill coefficient was greater in the D73N slow fibres, compared to WT slow fibres, but did not differ between WT fast fibres and D73N fast fibres.

Table 4.

Contractile properties of wild‐type and D73N soleus fibres

	WT		D73N
	Slow fibres	Fast fibres	Slow fibres	Fast fibres
P o/CSA (kN/m2)	95.1 ± 5.8†	72.9 ± 5.2*	58.2 ± 6.9*	62.7 ± 3.6*
pCa50	6.39 ± 0.04†	6.31 ± 0.06†	6.02 ± 0.02*	6.30 ± 0.02†
Hill coefficient	2.26 ± 0.05†	3.67 ± 0.10*†	2.74 ± 0.10*	3.87 ± 0.19*†

^*Significant difference from WT slow fibres. ^†Significant difference from D73N slow fibres.

Figure 7. Force/pCa relationship in slow fibres (left) and fast fibres (right) from the soleus of WT and D73N mice.

Mean ± SEM. N = 14–18 fibres in each group. The difference in pCa₅₀ between WT and D73N slow fibres is significant (P = 0.001).

Discussion

The results of this study indicate that a relatively small chemical change (substitution of Asp with Asn), in a single constituent of thin filaments (ss/cTnC), causes profound physiological, biochemical and structural alterations in skeletal muscle. Our study examined the impact of a single missense mutation which, when present simultaneously in the heart and in skeletal muscle, causes major alterations in skeletal muscle that are concurrent with changes in the heart (McConnell et al. 2015). The resultant changes in the heart and skeletal muscle, especially reduced cross‐sectional area of slow‐twitch fibres and transformation of slow skeletal muscle toward a fast muscle phenotype, could exacerbate the morbidity associated with heart failure due to mutations in ss/cTnC.

DCM is a disease of cardiac muscle that frequently progresses to heart failure (for reviews see Wexler et al. 2009; Jefferies & Towbin, 2010). The results of several studies indicate that skeletal muscle, including the diaphragm, is altered in heart failure patients (Tikunov et al. 1996; Szentesi et al. 2005; Weiss et al. 2017). Skeletal muscle abnormalities observed in human patients and animals with chronic heart failure include atrophy, decrease in oxidative capacity, shift in fibre type composition from slow‐twitch, type I fibres toward fast‐twitch, type II fibres, and increased tendency to fatigue (reviewed by Lunde et al. 2001; Adams et al. 2017). Exercise intolerance is one of the hallmark features of heart failure (reviewed by Okita et al. 2013). Skeletal muscle abnormalities are believed to be major contributors to exercise intolerance in heart failure patients (see reviews by Piepoli et al. 2010; Okita et al. 2013; Piepoli & Coats, 2013). It is likely that some of the changes in skeletal muscle associated with heart failure are secondary consequences of cardiac alterations, such as decreased physical activity, reduced perfusion of skeletal muscle and altered neuroendocrine status. Despite an increased proportion of fast‐twitch fibres, contraction and/or relaxation rates of skeletal muscle isolated from animals with heart failure tend to become slower, rather than faster (review by Lunde et al. 2001). Therefore, the faster kinetics of soleus isometric contractions, observed in the present study, appear to not be secondary consequences of coincident heart failure associated with the D73N mutation in the heart. Skeletal muscle atrophy is common among heart failure patients (Anker et al. 1997), but was not observed in the D73N mice in the present study. This further supports the conclusion that the alterations of contractile properties associated with the expression of the ss/cTnC D73N mutation are a direct consequence of the mutation, and not a secondary consequence of heart failure.

Consistent with earlier studies (Wigston & English, 1992; Denies et al. 2014), slow‐twitch fibres constitute ∼35% of the total population of fibres in mouse soleus muscle. Our results show that the D73N mutation led to a ∼40% decrease in cross‐sectional area of slow‐twitch fibres without affecting fibre type composition. An approximately 40% decrease in cross‐sectional area of slow‐twitch fibres without a change in fibre type composition was observed in the soleus muscle of transgenic mice with partial or total loss of slow skeletal TnT (ssTnT), a model of nemaline myopathy (Wei et al. 2014). Like the soleus muscles of ssTnT‐deficient mice, soleus muscles of the D73N mice had an altered force/frequency relationship, and lower fatigue resistance. In addition, similar changes in contractile parameters of twitch and tetanic contractions (with either significantly or trending shorter times to peak and half‐maximally relaxation) were observed in soleus muscles of ssTnT‐deficient mice (Wei et al. 2014) and D73N mice (present study). While the numbers of fast and slow fibres appear not to be different between WT and D73N soleus muscles, the changes in contractile parameters in soleus muscle of D73N mice could be a sole consequence of the increase in fractional content of total fast‐twitch fibres, based on fibre area, versus slow‐twitch fibres. However, additional studies are warranted to determine whether the D73N mutation causes alterations in contractile parameters of slow‐twitch fibres, contributing directly to the phenotype observed in soleus muscle of D73N mice. The differences in TnT isoform composition between WT and D73N mice were very minor and, therefore, seem unlikely to be the basis for the observed alterations in skeletal muscle phenotype between the two strains. Thus, the D73N mutation led to a skeletal muscle phenotype with similarities to that of nemaline myopathy.

Several ss/cTnC mutations have been linked to DCM (Mogensen et al. 2004; Lim et al. 2008; Hershberger et al. 2010). To date, there have been no reports describing the effects of these mutations on physiological properties of skeletal muscle. DCM‐linked mutations in ss/cTnC and other sarcomeric proteins tend to desensitize cardiac muscle to Ca²⁺ (for reviews see Kalyva et al. 2014; Spudich, 2014). Recently, we demonstrated that knock‐in mice expressing the D73N ss/cTnC mutation developed cardiac muscle abnormalities comparable to those in human patients with DCM (McConnell et al. 2015). Similar abnormalities of cardiac muscle were observed in mice after adeno‐associated virus‐mediated expression of ss/cTnC with the D73N mutation (Shettigar et al. 2016). While the D73N mutation is not currently known to be associated with DCM in humans, its effect on the Ca²⁺ sensitivity of reconstituted thin filaments was similar to that of the D75Y mutation found in site II of ss/cTnC isolated from a human patient diagnosed with idiopathic DCM (Lim et al. 2008; Dweck et al. 2010). The Ca²⁺ desensitizing effect of the D73N mutation, like that of the D75Y mutation, is probably due to the loss of a negative charge from the second Ca²⁺ binding loop of ss/cTnC. In addition, like the known DCM‐linked mutations in ss/cTnC (Biesiadecki et al. 2007; Pinto et al. 2011), the D73N mutation blunted the extent of Ca²⁺ desensitization due to cTnI phosphorylation, which occurs during β‐adrenergic stimulation (McConnell et al. 2015). Thus, the effect of the D73N mutation on the physiological properties of skeletal muscle is expected to be similar to that of known mutations in ss/cTnC associated with DCM in human patients.

Our earlier study showed that the Ca²⁺ sensitivity of force developed by trabeculae isolated from the ventricles of D73N mice was substantially reduced compared to that of WT mice (McConnell et al. 2015). Because ss/cTnC is expressed in cardiac and slow skeletal muscle, but not fast skeletal muscle, we expected that only slow‐twitch fibres isolated from the soleus muscle of the D73N mice would display reduced Ca²⁺ sensitivity of force development. Indeed, whereas Ca²⁺ sensitivity of fast‐twitch fibres was not affected, Ca²⁺ sensitivity of slow‐twitch fibres isolated from the soleus muscle of the D73N mice was reduced compared to that of WT mice. In addition, slow‐twitch fibres isolated from the D73N mice displayed reduced maximal force‐generating ability. Thus, the D73N mutation led to weakness of slow‐twitch muscle fibres. Whereas the D73N mutation has not been reported in humans, the impact of ss/cTnC mutations with similar consequences on Ca²⁺ sensitivity could have very significant consequences on skeletal muscle, as well as the heart, because human limb muscles consist of a greater percentage of slow fibres compared to mice in which fast fibres predominate in virtually all limb muscles. Furthermore, the reduction in Ca²⁺ sensitivity of slow skeletal muscle fibres (i.e. shift in pCa₅₀) due to the ss/cTnC D73N mutation was greater in slow fibres (0.36 pCa unit, this study) than that observed in cardiac trabeculae (0.23 pCa unit, McConnell et al. 2015). Consistent with our findings, hypertrophic cardiomyopathy‐linked mutations differentially affected Ca²⁺ sensitivity of force development in slow‐twitch and cardiac muscle fibres (Veltri et al. 2017). It is possible that interactions of ss/cTnC with the other constituents of the cTn complex in the heart (i.e. cTnI and cTnT) blunt the consequences of the D73N mutation, compared to the impact on Ca²⁺ sensitivity in slow skeletal muscle fibres. It is noteworthy that in the heart there is hypertrophy associated with the D73N mutation, but slow skeletal muscle fibres, with the same ss/cTnC mutation, are smaller than WT slow fibres. Thus, cardiac and skeletal muscle appear to utilize fundamentally different strategies in response to reduced Ca²⁺ sensitivity.

The lower force‐generation ability (force/cross‐sectional area) and smaller cross‐sectional area of D73N slow fibres, compared to the WT slow fibres, are unexpected findings from this study and the mechanisms underlying both are not known. It is possible that the reduced Ca²⁺ sensitivity that is associated with the D73N mutation results in a blunted growth and/or maturation of fibres expressing the mutant ss/cTnC with reduced myofibrillar protein content per unit of fibre volume. The combination of smaller than normal fibre size and lower force‐generating ability of the D73N slow fibres is similar to previous observations in vertebrate skeletal muscle at an early postnatal stage, in which both parameters are reduced, compared to adult values (e.g. Reiser et al. 1985, 1988, 1992). This similarity suggests that the reduced force generation and fibre size of D73N fibres are due to a delayed or blunted maturation of slow fibres that is associated with reduced Ca²⁺ sensitivity. It is possible that TnC has a role in muscle fibre size determination, perhaps involving regulation of transcription as has been reported for TnT (Zhang et al. 2013; reviewed by Johnston et al. 2108), but we are not aware of any evidence to directly support this. Related to this possibility, ss/cTnC has been found in the nuclei of cultured neonatal rat cardiomyocytes (Asumda and Chase, 2012). It also seems plausible that altered free Ca²⁺ transients, due to the altered affinity of ss/cTnC in the D73N mice for Ca²⁺, might have a Ca²⁺‐mediated impact on gene expression in slow fibres. The consequent lower force generated by slow fibres might, in turn, be compensated for by hypertrophy of fast fibres to maintain normal force generating ability of whole muscle, as was observed. The plateau in the force–pCa relationship clearly indicates that all of the studied fibres were maximally activated. A similarly reduced maximal force generation was observed previously in fast skeletal muscle which was reconstituted with fsTnC mutants with reduced Ca²⁺ sensitivity (Davis et al. 2004). We theorize that TnC mutants with lower Ca²⁺ sensitivity are not able to fully turn on the thin filament. Additional biochemical and physiological studies are needed to provide insight into the reasons behind the reduced force generation associated with the D73N mutation and other mutations that reduce the Ca²⁺ sensitivity of TnC.

While reduced Ca²⁺ sensitivity of skeletal muscle myofilaments is not always associated with fibre size differences, compared to normal muscle, there are several reports which indicate that alterations of thin filament proteins in skeletal muscle, which result in reduced Ca²⁺ sensitivity of force generation, are associated with reduced slow fibre size. For example, patients with mutations of tropomyosin 3 (expressed in slow skeletal muscle fibres) have smaller slow fibres and some of the same studies have shown that this phenomenon is associated with reduced Ca²⁺ sensitivity (e.g. Lawlor et al. 2010; Yuen et al. 2015).

Whereas the force‐generating ability (P _o/CSA) of whole soleus muscle did not differ between WT and D73N (Table 2), there were differences in P _o/CSA between groups of skinned fibres (Table 4). We do not have a clear explanation for this apparent discrepancy. However, in intact soleus muscles of D73N mice, the fast‐twitch fibres contribute more to force development than in WT mice. Because there were no significant differences in maximal force developed by fast‐twitch fibres in D73N mice compared to WT mice, the differences in force produced by intact muscles would be minimized, potentially to the point where the differences are no longer significant.

In conclusion, we utilized D73N knock‐in mice to examine the effect of desensitizing ss/cTnC to Ca²⁺ on physiological properties of soleus muscle, in which slow fibres are prevalent. There were four major findings from this study. First, all of the measured kinetics of isometric contraction (time to peak) and relaxation were significantly faster in the soleus muscle of D73N mice. Secondly, the muscles that expressed D73N ss/cTnC had significantly greater reduction in force during the fatigue test (hence, lower fatigue resistance). Thirdly, whereas the relative numbers of fast and slow fibres were not different between WT and D73N mice, the slow fibres in the soleus muscle from D73N mice had significantly smaller average cross‐sectional area; consequently, the per cent of the muscle cross‐section occupied by slow fibres was significantly lower in the D73N soleus. Consistent with the latter finding, there was a significantly lower relative amount of slow‐type myosin heavy chain in the D73N soleus muscle. Finally, the force/pCa relationship was significantly right‐shifted in slow fibres from D73N mice, indicating a significantly reduced Ca²⁺ sensitivity of force generation. Therefore, the consequences of the ss/cTnC D73N mutation significantly altered structural and functional properties of limb muscle slow fibres. We conclude that the D73N mutation in ss/cTnC transforms slow skeletal muscle toward the phenotype of fast muscle. The results demonstrate that a mutation in ss/cTnC that leads to DCM can also induce profound alterations in skeletal muscle function, potentially exacerbating the morbidity associated with cardiomyopathy.

Additional information

Competing interests

The authors declare that they have no competing interests.

Author contributions

ST and PJR designed the study. ST, NB, KD and PJR performed the study and analysed the data. ST and PJR drafted the manuscript. ST, NB, KD and PJR critically read and revised the manuscript. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

The authors thank The Ohio State University Office of Research and College of Dentistry for funding that supported this project. This project was also supported, in part, with funding from NHLBI of NIH under award number R15 HL117034 (to ST). The content is solely the responsibility of the authors, and does not necessarily represent the official views of the NIH.

Biography

Svetlana Tikunova received a PhD in Biophysics from The Ohio State University. Her research interests include studying regulation of muscle contraction by calcium‐binding proteins, and investigating consequences of mutations in thin filament proteins. She utilizes a variety of molecular, biochemical and physiological approaches to achieve her research goals and employs genetically modified mouse models in her research.