External Ca²⁺-dependent excitation–contraction coupling in a population of ageing mouse skeletal muscle fibres

Abstract

In the present work, we investigate whether changes in excitation–contraction (EC) coupling mode occur in skeletal muscles from ageing mammals by examining the dependence of EC coupling on extracellular Ca²⁺. Single intact muscle fibres from flexor digitorum brevis muscles from young (2–6 months) and old (23–30 months) mice were subjected to tetanic contractile protocols in the presence and absence of external Ca²⁺. Contractile experiments in the absence of external Ca²⁺ show that about half of muscle fibres from old mice are dependent upon external Ca²⁺ for maintaining maximal tetanic force output, while young fibres are not. Decreased force in the absence of external Ca²⁺ was not due to changes in charge movement as revealed by whole-cell patch-clamp experiments. Ca²⁺ transients, measured by fluo-4 fluorescence, declined in voltage-clamped fibres from old mice in the absence of external Ca²⁺. Similarly, Ca²⁺ transients declined in parallel with tetanic contractile force in single intact fibres. Examination of inward Ca²⁺ current and of mRNA and protein assays suggest that these changes in EC coupling mode are not due to shifts in dihydropyridine receptor (DHPR) and/or ryanodine receptor (RyR) isoforms. These results indicate that a change in EC coupling mode occurs in a population of fibres in ageing skeletal muscle, and is responsible for the age-related dependence on extracellular Ca²⁺.

The role of external Ca²⁺ ions in mammalian skeletal muscle contraction is not completely understood. Controversy exists over whether external Ca²⁺ is necessary for adult mammalian skeletal muscle excitation–contraction (EC) coupling. EC coupling is the series of ionic and molecular events whereby membrane depolarization is transmitted to contraction in striated muscles. Two protein complexes particularly important in EC coupling are the dihydropyridine receptor (DHPR) and the ryanodine receptor (RyR). These two proteins align during development at the junction of the t-tubule with the terminal cisternae of the sarcoplasmic reticulum (SR) (Flucher & Franzini-Armstrong, 1996) – a collection of structures termed the triad. The α1 subunit of the DHPR is a voltage-gated L-type Ca²⁺ channel in the t-tubule membrane that also serves as the voltage sensor (Rios & Brum, 1987). The DHPR is responsible for activating Ca²⁺ release from the SR via the RyR (Schneider & Chandler, 1973; Marty et al. 1994) into the cytoplasm to trigger muscular contraction. The DHPR and RyR isoforms in cardiac and skeletal muscle are different, with cardiac muscle expressing the DHPR α1_C subunit (Mikami et al. 1989) and RyR2 (Nakai et al. 1990), and skeletal muscle expressing the DHPR α1_S subunit (Tanabe et al. 1988, 1990a) and RyR1 (Takeshima et al. 1989). In cardiac muscle, RyR2 is activated to release Ca²⁺ from the SR by Ca²⁺ influx through the DHPR α1_C (Fabiato, 1985; Nabauer et al. 1989). This Ca²⁺ influx is necessary for cardiac EC coupling, as indicated by the complete elimination of contraction and Ca²⁺ release transients in cells where external Ca²⁺ has been removed or Ca²⁺ entry has been blocked (Tanabe et al. 1990b; García et al. 1994). By contrast, RyR1 in skeletal muscle does not depend on Ca²⁺ influx through the DHPR α1_S to activate SR Ca²⁺ release and skeletal muscle contraction. Experiments in which external Ca²⁺ has been removed from (Armstrong et al. 1972; Dulhunty & Gage, 1988; Tanabe et al. 1990b) or calcium channel blockers are added to (González-Serratos et al. 1982; Dulhunty & Gage, 1988) the bathing medium show that EC coupling and contraction persist in skeletal muscle cells. These experiments suggest a mechanical coupling between the α1_S and RyR1 (Rios & Brum, 1987; Tanabe et al. 1990a).

While adult skeletal muscle does not seem to depend on external Ca²⁺ for EC coupling, developing skeletal muscle does (Dangain & Neering, 1991; Cognard et al. 1992; Pereon et al. 1993), exhibiting some aspects of cardiac-like EC coupling. This is presumably due to the temporary expression of the cardiac DHPR α1_C during development (Chaudhari & Beam, 1993). As muscle cells develop, the expression of α1_S increases while α1_C decreases (Chaudhari & Beam, 1993), shifting to a more skeletal type EC coupling (Cognard et al. 1992). Ca²⁺-dependent EC coupling in developing muscle could also be due to transient expression of RyR3 (Flucher et al. 1999; Chun et al. 2003) or RyR1 splice variants (Futatsugi et al. 1995) or a combination of all these different isoforms, together allowing the RyR to be responsive to Ca²⁺ influx through the DHPR at normal cytoplasmic Mg²⁺ concentrations. Regenerating skeletal muscle, which goes through many stages similar to developing skeletal muscle, is at least partially dependent on external Ca²⁺ for effective EC coupling and contraction (Louboutin et al. 1995, 1996; Pereon et al. 1997a). This is also presumably due to the transient expression of α1_C during the regeneration period (Pereon et al. 1997a). Over time, expression of α1_C diminishes while α1_S increases, decreasing the dependence of regenerating muscle on external Ca²⁺ for effective EC coupling (Pereon et al. 1997a).

Prompted by the changes in EC coupling mode during development and muscle injury in adult rodents described above, we tested the hypothesis that skeletal muscle fibres from senescent mice undergo a ‘shift’ towards cardiac-like EC coupling (dependence on external Ca²⁺). Therefore, in the present study, we examined the effects of Ca²⁺-free solution on the contractile properties of fibres from flexor digitorum brevis (FDB) muscle in young and old mice, and whether differences were due to shifts in DHPR and/or RyR isoforms. To that end, we employed a combination of contraction and intracellular Ca²⁺ recordings in single intact muscle fibres, and measurements of Ca²⁺ current, charge movement and intracellular Ca²⁺ in patch-clamped fibres, together with molecular techniques to address this issue.

Methods

Animals

Flexor digitorum brevis (FDB) muscles were dissected from 3- to 6-month-old (young; n = 30) and 22- to 24-month-old FVB (our colony) or 21 month-old DBA or 26–27 month-old CB6F1 (Harlan-NIA colonies) (old; n = 23) mice. FVB and DBA strains have been successfully used in the study of ageing muscle in previous work from our laboratory (Renganathan et al. 1998; González et al. 2000, 2003) and the CB6F1 strain has been used by other investigators (Miller et al. 1997). The availability of the oldest animals determined the inclusion of three mouse strains in the present work. The findings reported in this study are independent of the mouse strain (see below). Animals were housed at Wake Forest University School of Medicine (WFUSM). Mice were killed by cervical dislocation. Animal handling and procedures were approved by the Animal Care and Use Committee of WFUSM.

Single intact fibre contraction experiments

The technique for dissecting single intact fibres followed procedures previously described (Lannergren & Westerblad, 1987; González et al. 2000). Two physiological buffering solutions were used for contraction experiments: Ca²⁺-containing (recording) and Ca²⁺-free solutions. The recording solution consisted of (mm): NaCl 121, KCl 5, CaCl₂ 1.8, MgCl₂ 0.5, NaH₂PO₄ 0.4, NaHCO₃ 24, and glucose 5.5. The Ca²⁺-free solution was identical to the recording solution, except that the MgCl₂ concentration was increased to 2.3 mm in place of CaCl₂. Both solutions also contained 10⁻⁵ g ml⁻¹ of d-tubocurarine chloride and were bubbled continuously with a mixture of 5% CO₂–95% O₂ to achieve a pH of 7.4. Fibres were stimulated by an electrical field generated between two parallel silver electrodes connected to a Grass S48 stimulator (Astro-Medical, Inc., West Warwick, RI, USA). Fibre length was adjusted until maximum force was elicited by a single twitch contraction (L_O) under isometric conditions. Suprathreshold square wave pulses of 0.5 ms duration were delivered to elicit twitch contractions. Tetanic contractions were elicited with 0.5 ms square wave pulses delivered in 300-ms trains. Frequency was increased until maximum force was attained. All subsequent tetanic contractions were elicited with the frequency that elicited maximal force, as described (González et al. 2003). All experiments were carried out at room temperature (20–21°C).

The first set of contraction experiments in single intact FDB fibres consisted of two prolonged contractile sequences: the first (reference trial) to assess any degree of force decline for that fibre, the second (test trial) to assess the effects of the Ca²⁺-free solution on tetanic force. The experimental sequence was as follows. (1) reference trial consisting of repeated maximal tetanic contractions, set at a 10-s interval, persisting for 25 min, in recording solution (total of 150 contractions). (2) Ten min of rest in recording solution. (3) Test trial repeating the reference trial contraction procedure, replacing recording solution with Ca²⁺-free solution from min 5 until min 15 of the 25-min procedure (contractions 31–90 of 150). (4) Up to 15 min of rest in recording solution, with single tetanic contractions at 5-min intervals. Force was normalized to baseline values for each fibre. Baseline was defined as an average of the five contractions immediately preceding the start of Ca²⁺-free solution flow. Force decline was assessed during min 5–15 of the reference trial, the same time frame during which Ca²⁺-free solution was perfused during the test trial. Only those experiments in which the fibre force recovered to at least 90% of baseline force were included for analysis.

The second set of contraction experiments in single intact FDB fibres also consisted of two contractile sequences: the first to assess the effects of the Ca²⁺-free solution on tetanic force, the second to record simultaneous force and Ca²⁺ fluorescence (see below). The experimental sequence was as follows. (1) Single tetanic contractions elicited at 2-min intervals for 20 min. From 0 to 10 min, Ca²⁺-free solution was perfused; from 10 min, recording solution was perfused. (2) Incubation for 40–50 min with the acetoxymethyl ester of fluo-4 (fluo-4 AM; 5 μm), added as a 1 mm stock in DMSO, followed by washout. (3) Readjustment of L_O, if necessary, and focusing of the laser scanning confocal microscope (Noran, OZ, Middleton, WI, USA). (4) A repeat of step (1) with simultaneous recording of force and fluorescence. Only those experiments in which the fibre force recovered to at least 90% of baseline force were included for analysis.

Data were acquired with a personal computer, an A–D converter (Digidata 1200, Axon Instruments, Union City, CA, USA) and pCLAMP software (Axon Instruments). The pulse waveform and the contraction signal were acquired and digitized together and stored for later analysis.

Charge movement, Ca²⁺ current and intracellular Ca²⁺ recordings

Single dissociated fibres were obtained from FDB muscles. FDB muscles were dissected in a solution containing (mm): caesium aspartate 155, magnesium aspartate 5 and Hepes 10; pH adjusted to 7.4 with CsOH (Beam & Franzini Armstrong, 1997). Muscles were treated for 3 h with 2 mg ml⁻¹ collagenase in a shaking bath at 37°C. Then, fibres were dissociated with Pasteur pipettes of different tip sizes. Fibres were transferred to a small flow-through chamber set on an inverted microscope stage (Axiovert S100 2TV, Zeiss, Göttingen, Germany). Fibres were continuously perfused with external solution (see below) using a push–pull pump (WPI, Sarasota, FL, USA). Muscle fibres were voltage-clamped using an Axopatch 200B amplifier (Axon Instruments, Union City, CA, USA) in the whole-cell configuration of the patch-clamp technique, following published procedures (Wang et al. 2000, 2002). The voltage clamp of short FDB fibres utilizing the whole-cell configuration of the patch-clamp technique together with low resistance pipette tips allows for a higher seal resistance, better control of the space clamp, and more prolonged and stable recordings of membrane currents and intracellular Ca²⁺ compared to previous procedures applied to longer muscle fibres (for discussion see Wang et al. 1999). Patch pipettes were pulled from borosilicate glass using a Flaming Brown micropipette puller (P97, Sutter Instrument Co., Novato, CA, USA) and then fire-polished to obtain an electrode resistance ranging from 450 to 650 kΩ. The pipette was filled with the following solution (mm): caesium aspartate 145, EGTA 10, MgCl₂ 5 and Hepes 10; pH adjusted to 7.4 with CsOH.

Inward Ca²⁺ currents were evoked with 350-ms depolarizing pulses from the holding potential (−80 mV) to command potentials ranging from −70 to 50 mV with 10 mV intervals. The Ca²⁺ current–voltage relationship was fitted to the following equation:

where G_max is the maximum conductance, V is the membrane potential, V_r is the reversal potential, V_½ is the half-activation potential, z is the valence of the mobile charge, F is the Faraday constant, R is the gas constant and T is the absolute temperature (Wang et al. 1999). The rising phase of the calcium current, elicited in response to command pulses of −50 to 50 mV, was adequately fitted to a single-exponential function as described (Delbono, 1992). The activation time constant (τ_a) was determined by fitting the current trace to an exponential function from the beginning of the pulse to the point where the current reached a steady-state level (∼30 ms). The data were fitted to a single barrier Eyring model assuming both first- and second-order terms to be present in the rate constant–voltage function:

where τ_max is the activation time constant at the voltage of equal charge distribution, V_½ is the half activation potential, and K = kT/aez (where k is the steepness of the curve, T is the temperature, a and b are constants that express the dependence of the rates on the first and second powers of the electric potential, e is the elemental charge, and z is valence of the mobile charge) (Brum & Rios, 1987). Double-pulse experiments were also performed to examine Ca²⁺-dependent inactivation of Ca²⁺ current. Inward Ca²⁺ and Ba²⁺ currents were evoked with 300-ms depolarizing pulses from the holding potential (−80 mV) to command potentials ranging from −70 to 60 mV with 10 mV intervals (first pulse). The second depolarizing pulse (also 300-ms duration) was held at a constant 20 mV potential, and followed the first pulse by a 100-ms interval. Relative current inactivation was calculated as current in the second pulse relative to maximum current attained in the second pulse (I_Ca/I_Ca,max). Ba²⁺ currents were measured in an external solution identical to the Ca²⁺-containing external solution, except that the 2 mm CaCl₂ was replaced by 2 mm BaCl₂. Data were fitted according to a Boltzmann distribution using the following equation:

where I is current, I_max is the maximum current amplitude, V_½ is the half-inactivation potential, V_m is membrane voltage, k is the steepness of the curve, and A is the amplitude factor.

Intramembrane charge movements were elicited with 25-ms depolarizing pulses from the holding potential (−80 mV) to command potentials ranging from −70 to 50 mV with 10 mV intervals. Intramembrane charge movement was calculated as the integral of the current in response to depolarizing pulses (charge on, Q_on) and is expressed per membrane capacitance (coulombs per farad). Charge movement was recorded in two different external solutions to assess the influence of external Ca²⁺ ions. Fibres were patch-clamped using a Ca²⁺-containing external solution containing (mm): tetraethlyammonium hydroxide (TEA-OH) 150, CaSO₄ 2, MgSO₄ 2, 3,4-diaminopyridine (DAP) 2, Hepes 5 and tetrodotoxin 0.001; pH adjusted to 7.4 with CH₄SO₃. Ca²⁺ current was blocked with the addition of 0.5 mm CdCl₂ and 0.3 mm LaCl₃ to record charge movement in the presence of external Ca²⁺ ions. Ca²⁺-free external solution containing (mm): TEA-OH 150, MgSO₄ 4, DAP 2, Hepes 5 and tetrodotoxin 0.001 (pH adjusted to 7.4 with CH₄SO₃) was perfused to record charge movement in the absence of external Ca²⁺ ions. Mean data points were fitted to a Boltzmann distribution of the form:

where Q_max is the maximum charge, V_m is the membrane potential, V_Q½ is the charge movement half-activation potential, and k is the steepness of the curve (Zheng et al. 2002a).

For intracellular Ca²⁺ recordings, FDB fibres were loaded with 5 μm fluo-4 AM (Molecular Probes, Eugene, OR, USA) for 15–20 min. After washout with the Ca²⁺-containing solution, fibres were patch- and voltage-clamped (see above). In these experiments, the pipette solution was identical to that used for charge movement recordings, except that the EGTA concentration was reduced to 0.2 mm. Ca²⁺ transients were elicited with 20-ms depolarizing pulses from the holding potential (−80 mV) to command potentials ranging from −40 to 30 mV. Ca²⁺ transients were recorded in both the Ca²⁺-containing and the Ca²⁺-free solutions used for charge movement recordings as described above.

Fluorescent image recording

For Ca²⁺ fluorescence recordings in contraction and patch-clamp experiments, the fibres were illuminated with a laser beam at 488 nm wavelength. The beam passed through an OZ Scan module (Noran Instruments) and through a 20 X Fluar objective (Zeiss) before reaching the fibre. Emitted fluorescence was collected by the objective and directed to the OZ scan module, in a non-slit mode, through the emission filter at 525 nm wavelength before being collected by a photomultiplier tube and digitized. Hardware control, image acquisition and processing were performed with Intervision Software (Noran Instruments) run on a Silicon Graphics O₂ Workstation (Mountain View, CA, USA). Sequences of 150 images at 8 ms intervals were collected for each contraction. For data analysis, several regions of interest (ROIs) were selected for each cell and the maximum fluorescence deflection was used. Fluorescence data are reported as a percentage change in fluorescence normalized to basal fluorescence (%ΔF/F) (Finch & Augustine, 1998).

Ribonuclease protection assay

Ribonuclease protection assay (RPA) technique followed procedures previously described (Zheng et al. 2001). Briefly, total RNA was extracted from mouse skeletal muscle and heart by using TRI reagent (Molecular Research Center, Inc., Cincinnati, OH, USA). The probe for mouse DHPR α1_C is generated by RT-PCR using sense primer 5′TACGGACTTCTCTTCCACCC-3′, anti-sense primer 5′TCCCTCCTAGAGCATTGGCC-3′, corresponding to mRNA sequence (Accession number NM_009781) from 1707 to 1845. Its PCR fragment is cloned in TA-easy vector (Promega, WI, USA) and confirmed by DNA sequencing, the construct is linearized by Sal I digestion and purified. In vitro transcription was performed using the Maxiscript kit from Ambion (Austin, TX, USA). Briefly, the in vitro transcription reaction was performed by mixing linearized plasmid, 10 × transcription buffer, 10 mm ATP/CTP/GTP/UTP, [³²P]UTP, T7 RNA polymerase plus ribonuclease inhibitor. The transcription was followed by removal of template DNA with DNase and gel purification of the probe. The cRNA probe is 229 bp (139 plus 90-bp vector fragment), and protected fragment after RNase digestion in RPA is 139 bp. RPA was performed using RPA II reagents (Ambion). Total RNA 25 μg was hybridized with labelled probes at 56°C overnight. This was followed by RNase digestion of non-hybridized probes and sample RNA, separation and detection of the protected fragments in a 10% urea denaturing polyacrylamide gel and gel exposure to an X-ray film.

DHPR protein detection

Immunoprecipitation

Mouse skeletal and heart muscles were prepared for immunoprecipitation as previously described (Zheng et al. 2002b) with modification. Briefly, muscles were homogenized with a blender homogenizer (Kinematica, Switzerland) in 1% digitonin buffer (1% digitonin, 185 mm KCl, 1.5 mm CaCl₂ and 10 mm Hepes; pH 7.4) on ice. Cellular debris was pelleted by centrifugation at 10 000 g for 10 min at 4°C. Protein concentration was measured by bicinchoninic acid (BCA) protein assay (Pierce Biotechnology, Rockford, IL, USA). The lysate (500 μg total cellular protein) was pre-cleared by adding 0.5 μg of the appropriate control IgG (normal rabbit IgG), together with 20 μl of resuspended volume of the appropriate agarose conjugate (protein G-agarose). Samples were then incubated at 4°C for 30 min. After centrifugation at 500 g for 5 min at 4°C, supernatant was transferred to a fresh tube on ice. Rabbit anti-DHPR α1_C primary antibody (Sigma, St Louis, MO, USA; 1 μg) was added and incubated overnight at 4°C on a rotating device. The control tube received only rabbit IgG. Then 20 μl of resuspended volume of the Protein G-Agarose was added to each tube and incubated at 4°C for 2 h. After centrifugation at 500 g for 5 min at 4°C, the pellets were washed in PBS three times and resuspended in 20 μl of 1 × electrophoresis sample buffer. All samples were boiled for 2–3 min and separated by 10% denaturing SDS-PAGE at 100 V for 4–5 h. Rainbow Molecular Weight Marker (Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA) was loaded for reference. Proteins were transferred from the gel to nitrocellulose membrane (Amersham Pharmacia Biotech Inc.).

Microsome preparation

Microsomes from mouse heart and skeletal muscle were prepared as described (Saito et al. 1984; Inui et al. 1988) with modification. Mouse skeletal (5 g) and heart (3 g) muscles were finely cut, and were homogenized with a blender homogenizer (Kinematica) in a homogenization buffer containing 5 mm imidazol (pH 7.4) and 300 mm sucrose with complete protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA). The homogenate was centrifuged at 5000 g for 15 min and the supernatants were filtered through four layers of cheesecloth and centrifuged at 15 000 g 15 min, then the second supernatants were centrifuged at 100 000 g for 90 min. Microsome pellets were resuspended in 1% digitonin buffer. After the measurement of protein concentration by BCA protein assay, samples were mixed with 1 × sample buffer. All samples were boiled for 2–3 min, separated on 10% denaturing SDS-PAGE gels, and transferred to nitrocellulose membranes.

Immunoblot

Nitrocellulose membranes for microsome preparations and immunoprecipitation were blocked by incubating membranes in 150 mM NaCl, 10 mM Tris-HCl, pH 7.4 (TBS), 0.1% Tween-20, 5% milk for 30–60 min at room temperature. Membranes were then incubated with rabbit anti-DHPR α1_C primary antibody (Sigma), diluted 1: 100 in TBS, 5% milk, 0.1% Tween-20, for 1 h at room temperature and washed three times for 5 min each with TBS, 0.05% Tween-20. After incubation for 30 min at room temperature with peroxidase-conjugated anti-rabbit lgG (diluted 1: 3000 in TBS, 0.1% Tween-20, 5% milk), membranes were washed three times for 5 min each with TBS, 0.05% Tween-20, and once for 5 min with TBS. Finally, the membrane was incubated in ECL Reagent (Pierce Biotechnology) and visualized by X-ray film.

RyR protein detection

Crude membrane preparation

Crude membrane fractions from pooled hindlimb muscles from adult frog and young and old mice were prepared as previously described (Chun et al. 2003). Briefly, frozen muscles were chopped into small pieces with scissors and homogenized in 500 μl ice-cold homogenization buffer (20 mm Hepes, pH 7.4, 250 mm sucrose, 0.2% sodium azide, with complete protease inhibitor cocktail) with a blender homogenizer (Kinematica, Switzerland). Crude membrane fraction was extracted by adding 500 μl extraction buffer (5 mm NaPO₃, pH 7.4, 75 mm NaCl, 1% Triton X-100, 1% deoxycholate, 0.1% SDS, with complete protease inhibitor cocktail) and centrifuging at 10 000 g for 15 min. Proteins in the supernatant were separated on 4% SDS-PAGE gels, and transferred to polyvinylidene difluoride membranes (Amersham Biosciences, UK) in transfer buffer (12 mm Tris-base, 92 mm glycine, 0.02% SDS and 20% methanol) at 25 V overnight at 4°C.

Immunoblot

Membranes were blocked for 3 h in 5% milk TBST at room temperature. Membranes were then incubated in monoclonal anti-RyR (Affinity Bioreagents, Golden, CO, USA) diluted at 1:5000 in 5% milk TBST at room temperature for 60 min, and washed three times for 5 min each in TBS and 0.05% Tween-20. This antibody reacts with both RyR1 and RyR3 (Chun et al. 2003). After incubation for 60 min in peroxidase-conjugated anti-mouse IgG (Amersham Biosciences) diluted 1:10 000 in 5% milk TBST, membranes were washed again three times for 5 min each in TBS and 0.05% Tween 20, and once in TBS. Finally the membranes were incubated in ECL reagent (Pierce Biotechnology) and visualized on X-ray film.

Statistics

Values reported are mean ± s.e.m. Statistical analysis was performed using analysis of variance (ANOVA) followed by multiple comparisons tests (Tukey's HSD). An alpha value of P < 0.05 was considered significant.

Results

FDB single intact fibre contractility in Ca²⁺-free solution

The first set of experiments was performed in order to determine whether the absence of external Ca²⁺ ions affects young or old muscle contractility. To assess the effects of external Ca²⁺ on young and old muscle fibre contractility, force was measured in single intact FDB fibres from young (n = 11 fibres) and old (n = 14 fibres) mice in the presence of both normal recording solution and Ca²⁺-free solution. The single intact fibre preparation allows for several advantages over a multifibre preparation. First, the contracting fibre is very rapidly exposed to the perfusion solution. Second, whereas fibres can heterogeneously contribute to whole muscle force production due to differences in pennation and fibre length (Sugi & Tsuchiya, 1998), one can be assured the single intact fibre is functioning at optimal length (L_O) (González et al. 2000). Force was normalized to baseline values for each fibre. Figure 1A shows that the reference trial contraction protocol induced no decline in peak force in single intact fibres from young and old mice (P > 0.05). The same contraction protocol in the absence of external Ca²⁺ (the test trial) produced no significant force decline in fibres from young mice (P > 0.05), while force declined significantly in the fibres from old mice (35% decline, P < 0.05), compared to the reference trial. The effects of Ca²⁺-free solution on old FDB fibres was greater than on young fibres (P < 0.05).

Figure 1. Absence of external Ca²⁺ ions causes peak tetanic force to decline in single intact FDB fibres from old mice.

Single intact FDB fibres from young (n = 11) and old (n = 14) mice underwent the same two contractile protocols (see Methods). A, peak force in fibres from young mice did not significantly decline during the test trial compared to the reference trial (P > 0.05), while force declined significantly in fibres from old mice during the test trial compared to the reference trial (*P < 0.05). Force decline was significantly different in fibres from old and young mice (#P < 0.05). B, two populations of muscle fibres were distinguishable in old mice: fibres in which peak force did not decline in Ca²⁺-free solution (n = 8, Old Non-Affected, P > 0.05) similar to fibres from young mice, and fibres in which force greatly declined in Ca²⁺-free solution (n = 6, Old Affected, **P < 0.001 versus reference trial, ##P < 0.001 versus Young and Old Non-Affected test trial).

Upon further examination of the old FDB fibres, two clearly delineated populations of fibres from old mice were distinguished. Within the group there exists a population of fibres (Old Non-Affected, n = 8 of 14) that were not affected by Ca²⁺-free solution (P > 0.05), similar to young fibres. There is also a population of fibres from old mice (Old Affected, n = 6 of 14) that showed dramatic force decline in Ca²⁺-free solution compared to the reference trial (P < 0.001) and compared to both young (P < 0.001) and Old Non-Affected (P < 0.001) fibres in the absence of external Ca²⁺ (Fig. 1B). Of the fibres from young mice, only one fibre declined substantially in Ca²⁺-free solution (∼35% decline). Even though this fibre exists more than two standard deviations outside the mean for this group, it was included in the data because force recovered completely by the end of the experiment. The presence of this fibre, combined with some depression in force recorded in fibre bundles (n = 5–10 fibres) from young mice (data not shown), indicates that some fibres from young, although rare, may also be partially dependent on external Ca²⁺ ions for EC coupling.

Figure 2A shows the time course of the reference (left) and test (right) trials and a recovery tetanus (5 min after the end of test trial, far right) in one fibre from an old mouse that was unaffected by Ca²⁺-free solution. Peak force is maintained throughout the reference trial, indicating that the protocol induced no decline in peak tetanic force in this fibre population. Similarly, peak force does not significantly decline during the test trial when Ca²⁺-free solution is perfused. Individual force traces from the same fibre during each contraction protocol are shown in Fig. 2B (a–f). Figure 2C shows the two contraction protocols in one fibre from an old mouse that was significantly affected by Ca²⁺-free solution. Peak force was maintained during the reference trial (left), indicating that the protocol also induced no fatigue in this fibre population. However, during the test trial (right), peak force significantly declined (approximately 60% for this fibre) during perfusion of Ca²⁺-free solution. Upon return to normal 1.8 mm Ca²⁺ solution, force recovered to near baseline. Given more time (5 min after the end of test trial; far right), force recovered to initial levels. This, taken together with the lack of force decline in the reference trial, indicates that force decline during the test trial is due to the absence of external Ca²⁺ ions. Individual force traces from the same fibre during each contraction protocol are shown in Fig. 2D (a–f).

Figure 2. Two populations of single intact FDB fibres from old mice.

A, time course of reference (left traces) and test (right traces) trials for one representative ‘unaffected’ single intact FDB fibre from an old mouse. Top bar shows solution changes during the trials. Peak force in this fibre did not decline during the reference trial, and peak force in this fibre minimally declined in the absence of external Ca²⁺ ions during the test trial. C, time course of reference (left traces) and test (right traces) trials for one representative ‘affected’ single intact FDB fibre from an old mouse. Top bar shows solution changes during the trials. Force declined by approximately 60% for this fibre in Ca²⁺-free solution, then recovered to ∼90% of baseline once normal physiological solution was returned. Given enough recovery time, force recovered to initial levels (far right trace). B and D, tetanic force traces corresponding to selected traces in A and C, repectively: contraction 30 of 150 (a and d, immediately preceding change to Ca²⁺-free solution in test trial), contraction 90 of 150 (b and e, immediately preceding return to normal recording solution in test trial), contraction 150 of 150 (c, end of the reference trial), and the recovery tetanus (f, 5 min after the end of test trial).

DHPR function

A key step in EC coupling is the voltage sensing function of the DHPR resulting in intramembrane charge movements. Decreases in charge movements would probably decrease force generation in a muscle fibre. Therefore, to determine whether the absence of external Ca²⁺ ions affected the function of the DHPR, single dissociated FDB fibres were voltage-clamped in the whole cell configuration of the patch-clamp technique (Wang et al. 2000). Intramembrane charge movements were measured in FDB fibres from young (n = 16) and old (n = 11) mice after blocking the inward Ca²⁺ current (see Methods). Maximum charge movement (Q_max) was lower in old compared to young fibres (P < 0.05), which agrees with previously published data (Wang et al. 2000, 2002). However, no charge movement difference was found between 2 mm and 0 mm Ca²⁺ external solutions in either young (P > 0.05, Fig. 3A and B) or old (P > 0.05, Fig. 3C and D) FDB fibres. Mean data points were fitted to a Boltzmann distribution (eqn (4), see Methods). The best-fitting parameters for Q_max, V_Q½, and k for fibres from young mice are: 54.9 nC nF⁻¹, 16.9 mV, and 18.5, respectively, in Ca²⁺-free conditions and 47.1 nC nF⁻¹, 18.6 mV, and 17.7, respectively, in 2 mm Ca²⁺solution. The best-fitting parameters for Q_max, V_Q1/2, and k for fibres from old mice are: 34.1 nC nF⁻¹, 19.2 mV, and 18.5, respectively, in Ca²⁺-free conditions, and 31.1 nC nF⁻¹, 15.4 mV, and 18.8, respectively, in 2 mm Ca²⁺ solution. These data suggest that fibre force decline in Ca²⁺-free solution is not due to any inhibition of DHPR function in Ca²⁺-free solution. These results are supported by the fact that Ca²⁺-free solution did not have a general effect on contraction in all fibres. Indeed, a general deleterious effect of Ca²⁺-free solution on charge movements in all fibres would probably lead to force decline in all fibres. Additionally, two populations of fibres were not found in fibres from old mice (Fig. 3E), further suggesting that alteration in DHPR function in Ca²⁺-free solution does not contribute to force decline in Ca²⁺-free solution.

Figure 3. Intramembrane charge movement (Q_on) measured in the presence and the absence of external Ca²⁺.

Q_on measured over a range of stimulation voltages in FDB fibres from both young (A and B, n = 16) and old (C and D, n = 11) mice. Single dissociated FDB fibres were voltage-clamped in the whole-cell configuration of the patch-clamp technique. Charge movements were evoked with 25-ms depolarizing voltage steps from the holding potential (−80 mV) to command potentials ranging from −70 to 60 mV. Q_on–V_m relationship in fibres from young (A) and old (C) mice in 2 mm Ca²⁺ external solution (□) and in Ca²⁺-free external solution (•). Data points were fitted to a Boltzmann distribution (eqn (4)) (continuous line, 0 mm Ca²⁺ solution; dashed line, 2 mm Ca²⁺ solution). Sample charge movement recordings in fibres from a young mouse (B) and an old mouse (D). Numbers on the right indicate the membrane potential. Dotted lines represent the baseline. E, individual maximum Q_on data points for fibres from young and old mice in 2 mm and 0 mm Ca²⁺ solution show single populations.

Intracellular Ca²⁺ -transients

Following DHPR activation, the next key step in EC coupling is SR Ca²⁺ release via RyR1. Impairment of SR Ca²⁺ release from RyR1 results in a decreased intracellular Ca²⁺ transient and decreased force. Therefore, in order to determine whether intracellular Ca²⁺ was reduced in muscle fibres while in Ca²⁺-free solution, intracellular Ca²⁺ transients were measured in FDB fibres under two conditions: in voltage-clamped fibres under whole cell patch-clamp and in single intact fibres undergoing electrically elicited contractions. Intracellular Ca²⁺ transients, across a range of depolarization potentials in voltage-clamped fibres, are not affected by the absence of external Ca²⁺ ions in fibres from young mice (n = 15). Similar to the contractile data, the results in Fig. 4 show that two populations of fibres exist in old mice: fibres in which Ca²⁺ transients were not reduced while in Ca²⁺-free solution similar to fibres from young (P > 0.05, Old Non-Affected, n = 8 of 11) and fibres in which Ca²⁺ transients were greatly reduced in Ca²⁺-free solution compared to fibres from both young and old (P < 0.01, Old Affected, n = 3 of 11). These results confirm, with a different technique, the results in contracting fibres and suggest that a population of fibres from old mice are at least partially dependent upon external Ca²⁺ for efficient excitation–Ca²⁺ release coupling.

Figure 4. Intracellular Ca²⁺ measured in voltage-clamped FDB fibres in whole-cell patch clamp.

Single dissociated FDB fibres (young, n = 12; old n = 11) were voltage-clamped in the whole-cell configuration of the patch-clamp technique. Cells were loaded with the fluorescent Ca²⁺ indicator fluo-4. Intracellular Ca²⁺ release was evoked with 20-ms depolarizing voltage steps from the holding potential (−80 mV) to command potentials ranging from −40 to 30 mV in both 2 mm and 0 mm external Ca²⁺. The ratio of fluorescence measured in 0 mm external Ca²⁺ to fluorescence measured in 2 mm external Ca²⁺ was calculated. Two populations of fibres were identified from old mice: those in which Ca²⁺ transients were unaffected by the absence of external Ca²⁺ (n = 8 of 11, Old Non-Affected) similar to fibres from young mice and those in which Ca²⁺ transients were reduced in Ca²⁺-free solution (n = 3 of 11, Old Affected). *P < 0.05, Old Affected versus Old Non-Affected; **P < 0.01 Old Affected versus Young and Old Non-Affected.

In order to determine whether the lower intracellular Ca²⁺ transient recorded in Ca²⁺-free external solution is the cause of the force decline in fibres from old mice, intracellular Ca²⁺ transients were recorded simultaneously with force in fibres from young (n = 3) and old (n = 3) mice in normal recording solution and in Ca²⁺-free solution. These experiments also contained two contraction sequences: the first to assess the effects of the Ca²⁺-free solution on elicited force for that fibre, the second to record simultaneous force and Ca²⁺ transients (see Methods).

Ca²⁺-free solution did not cause force or intracellular Ca²⁺ transients to decline in fibres from young mice (Fig. 5A). However, in fibres from old mice, Ca²⁺-free solution caused both intracellular Ca²⁺ transients and force to decline in parallel (Fig. 5B). Within these three fibres from old mice, there was a large range of force decline (fibre 1021 A showed ∼92% force decline; fibre 1105 B ∼51% force decline; and fibre 1124 A ∼15% force decline). The range of force decline in these three fibres was within the range reported for the original experiments shown in Fig. 1, and covers the range of force decline of the two populations of fibres: fibres 1031 A and 1105 B fall into the ‘Old Affected’ group while fibre 1124 A falls into the ‘Old Non-Affected’ group. The data illustrated in Fig. 5B are from fibre 1105 B (∼51% force decline). Individual data points for force and intracellular Ca²⁺ from all time points during experiments in all fibres from young and old mice were analysed with regression analysis. These data indicate that the decreased force and decreased intracellular Ca²⁺ transient induced by Ca²⁺-free solution are directly related (r² = 0.829, P < 0.001). These data support the concept that some muscle fibres from old mice are partially dependent upon external Ca²⁺ for efficient EC coupling and maintenance of tetanic force.

Figure 5. Simultaneous force and Ca²⁺ traces from single FDB fibres.

Fibres were loaded with fluorescent Ca²⁺ indicator fluo-4. Maximal tetanic contractions were elicited every 2 min for 20 min; Ca²⁺-free solution was perfused from start time to min 10, and normal 1.8 mm Ca²⁺ solution was perfused from min 10–20. ‘Baseline’ and ‘Recovery’ traces were recorded in the presence of normal 1.8 mm external Ca²⁺, while all other traces were recorded in the absence of external Ca²⁺. A, compared to baseline, neither force nor Ca²⁺ fluorescence declined in the absence of external Ca²⁺ in a fibre from a young mouse. B, compared to baseline, both force and Ca²⁺ fluorescence declined in parallel in the absence of external Ca²⁺ in a fibre from an old mouse. Vertical scale bars represent both 100% force and 100% ΔF/F normalized to baseline values.

DHPR isoforms in skeletal muscle from young and old mice

Ca²⁺ current recordings

The current–voltage relationship, the activation time constant and the Ca²⁺ -dependence of inactivation of inward Ca²⁺ current (I_Ca) were examined to explore the possibility of an age-related shift from DHPR α1_S to DHPR α1_C. Figure 6A shows the whole cell Ca²⁺ current–voltage (I–V) relationship for fibres from young (n = 11) and old (n = 10) mice. The data were fitted to eqn (1) (see Methods). The best fitting parameters for G_max, V_r, V_½ and z determined from mean data points are 73 nS nF⁻¹, 66 ± 4.2 mV, − 1.9 ± 0.02 mV and 6.5 ± 2.1, respectively, for fibres from young mice; and 66 nS nF⁻¹, 63 ± 3.7 mV, 0.5 ± 0.01 mV and 4.8 ± 0.02, respectively, for fibres from old mice. The I–V relationship for DHPR α1_C is shifted leftwards compared to α1_S (García et al. 1994). Here, we show no differences in I–V curve orientation in fibres from young compared to old mice. Individual Ca²⁺ current data at 0 and + 10 mV for fibres from both young and old mice are evenly distributed, suggesting that two populations do not exist in this measure (Fig. 6B). The activation time constant (τ_a) was determined by fitting the current trace to an exponential function from the beginning of the pulse to the point where the current reached a steady-state level. As DHPR α1_C calcium current has a faster τ_a than that of α1_S (Tanabe et al. 1991; Nakai et al. 1994), an age-related subunit shift should be evident in the τ_a values. Figure 6C shows τ_a values corresponding to a range of test potentials. At 10 mV, τ_a is significantly faster in fibres from old (n = 7) than from young (n = 8) FDB fibres, suggesting the possibility of the presence of DHPR α1_C subunits in old muscle. The one barrier Eyring rate model was fitted to the mean time constant of I_Ca current activation. The best fitting parameters for τ_max, V_½, and K (see eqn (2), Methods) were: 16.4 ms, 7.7 mV and 10.7 mV, respectively, for fibres from young mice; and 11.2 ms, 5.5 mV and 15.4 mV, respectively, for fibres from old mice. Despite the difference in τ_a between young and old, individual data for τ_a at +10 mV also do not display two populations in fibres from young of old mice, suggesting that this may not be a mechanism to explain force decline in Ca²⁺-free solution. Representative Ca²⁺ current traces from young and old at a 10 mV command potential are shown in Fig. 6E. The raw Ca²⁺ current recording is shown (dashed line) with a single exponential fitting to the rising phase (Delbono, 1992) overlaid onto the traces. The rising phase, and hence τ_a, is faster in old compared to young at 10 mV.

Figure 6. Current–voltage relationship and activation time constant of Ca²⁺ current measured in patch-clamped fibres from young and old mice.

Fibres were voltage-clamped, and inward Ca²⁺ currents (I_Ca) were evoked with 350-ms depolarizing voltage steps from the holding potential (−80 mV) to command potentials ranging from −70 to 50 mV. A, Ca²⁺ current–voltage (I–V) relationship. Data from young (n = 11) and old (n = 10) were fitted to eqn (1) (continuous lines, see Methods). I–V relationship shows no age-related shift in the voltage-dependence of the Ca²⁺ current. B, individual data points for fibres from young and old mice at 0 and + 10 mV show single populations of data. C, the rising phase of the inward Ca²⁺ current was fitted to a single exponential function (see Methods). The activation time constant (τ_a) was determined by fitting the current trace to an exponential function from the beginning of the pulse to steady state (∼30 ms). The data were fitted to a single barrier Eyring model (continuous lines, eqn (2)). τ_a–V_m relationship in fibres from young (•) and old (○) animals. At 10 mV, τ_a is significantly slower in fibres from young than from old mice. D, individual τ_a data points for fibres from young and old mice at the + 10 mV command potential show single populations of data. E, sample I_Ca traces from representative fibres from young and old mice at 10 mV.

I_Ca inactivation of DHPR α1_S is voltage-dependent (Cota et al. 1984), whereas I_Ca inactivation of DHPR α1_C is voltage and Ca²⁺ dependent (De Leon et al. 1995; Zuhlke et al. 1999). Double-pulse experiments were performed in both Ca²⁺-and Ba²⁺-containing solutions (see Methods) to examine the Ca²⁺ dependence of inactivation of I_Ca. Relative current inactivation (I_Ca/I_Ca,max) was determined by normalizing the current during the test pulse to the maximum current measured during the set of test pulses. Current inactivation was measured with Ba²⁺ because Ba²⁺ current (I_Ba) inactivation is sensitive only to voltage inactivation. An I_Ca inactivation curve that is shifted leftwards in relation to I_Ba inactivation indicates Ca²⁺ dependence of I_Ca current inactivation. Figure 7 shows no Ca²⁺ dependence of I_Ca inactivation in either young (Figs 7A and B; n = 6) or old (Figs 7C and D; n = 7) fibres (P > 0.05). The best fitting parameters for the mean values of Ca²⁺ current (see eqn (3), Methods) for A, V_½ and k were: 0.8, −6.9 mV and 13.1 mV, respectively, for fibres from young mice; and 0.73, −7.4 mV and 8.5 mV, respectively, for fibres from old mice. The best fitting parameters for the mean values of Ba²⁺ current for A, V_½ and k were: 0.93, −17 mV and 13.6 mV for fibres from young mice; and 0.75, −12.2 mV and 11.0 mV, respectively, for fibres from old mice. These data disagree with the τ_a data and the suggestion of the presence of DHPR α1_C in old muscle fibres.

Figure 7. Ca²⁺ dependence of inactivation of Ca²⁺ current (I_Ca) measured in patch-clamped fibres from young and old mice.

Fibres were voltage-clamped, and I_Ca was evoked with two 300-ms depolarizing voltage steps from the holding potential (−80 mV) in double-pulse experiments. The first pulse ranged from −70 to 60 mV. The second (test) pulse was held constant at 20 mV, with a 100-ms interpulse interval (B, top illustration). Relative current inactivation (I_Ca/I_Ca,max) was determined by subtracting the current during a given test pulse from the highest current measured during the set of test pulses for that fibre. Experiments were repeated replacing Ca²⁺ in solution with Ba²⁺. Data were fitted to a Boltzmann equation (eqn (3)). I_Ca and I_Ba inactivation curves in fibres from young (A) and old (C) mice show no Ca²⁺ dependence of I_Ca inactivation, as no leftward shift was found for the I_Ca inactivation curve compared to the I_Ba curve. Sample I_Ca traces from double-pulse experiments from single FDB fibres from young (B) and old (D) mice display time course and magnitude of I_Ca responses over a range of command potentials.

DHPR α1_C detection

To clarify whether the dependence of EC coupling on external Ca²⁺ ions in old fibres is due to an age-related shift in DHPR isoforms from the skeletal α1_S subunit to the cardiac α1_C subunit, we performed ribonuclease protection assay (RPA) and immunoblot analysis. RPA could not detect the presence of DHPR α1_C mRNA in either young or old skeletal muscle (Fig. 8A). Similarly, immunoblot analysis following immunoprecipitation (Fig. 8B) and following isolation of muscle membrane fraction (Fig. 8C) did not detect the presence of DHPR α1_C protein in young or old skeletal muscle, suggesting that a shift between these two specific DHPR subunit isoforms does not occur with ageing. Samples from heart muscle were used as positive control for DHPR α1_C.

Figure 8. Detection of DHPR α1_C mRNA and protein and RyR protein isoforms.

A, ribonuclease protection assay did not reveal the presence of DHPR α1_C mRNA in hindlimb skeletal muscle from young (Y) or old (O) mice. Size marker (M) and undigested probe (P) are shown. Mouse heart muscle was used as a positive control (H). B, immunoblot following IgG immunopreciptiation (see Methods) failed to show the presence of DHPR α1_C protein in hindlimb muscle from young (Y) or old (O) mice. Mouse heart muscle was used as a positive control (H). C, immunoblot following membrane isolation failed to reveal the presence of DHPR α1_C protein in hindlimb muscle from either young (Y) or old (O) mice. Mouse heart muscle was used as a positive control (H). D, immunoblot following membrane fraction isolation revealed the presence of RyR1, but failed to reveal the presence of RyR3 in hindlimb skeletal muscle from young or old mice. Frog hindlimb muscle was used as a positive control for both RyR1 (RyRα) and RyR3 (RyRβ).

RyR isoforms in skeletal muscle from young and old

As no shift from DHPR α1_S to α1_C is evident, we examined a potential change in RyR isoforms expressed in ageing skeletal muscle. Embryonic skeletal muscle co-expresses RyR1 and RyR3 (Flucher et al. 1999; Chun et al. 2003), whereas adult fast-twitch skeletal muscle does not express RyR3 (Flucher et al. 1999). A shift in expression of RyR isoforms in ageing mammalian skeletal muscle towards RyR3 expression may also induce EC coupling dependence on external Ca²⁺ ions, due to the fact that DHPR and RyR3 do not physically couple (Fessenden et al. 2000; Protasi et al. 2000). In this scenario, RyR3 would only be able to contribute to the SR Ca²⁺ release transient via Ca²⁺ -induced Ca²⁺ release (CICR) activated by Ca²⁺ influx through the DHPR and Ca²⁺ release from neighbouring RyR1s. Figure 8D shows immunoblot results from young and old mouse pooled hindlimb skeletal muscle. No RyR3 protein could be detected in either young or old skeletal muscle, suggesting no age-induced shift in RyR isoforms. Adult frog muscle was used as a positive control for both RyR1 and RyR3, as frog muscle has been previously shown to express both proteins in relatively equal amounts (Chun et al. 2003).

Discussion

This work supports the conclusion that ageing leads to dependence on external Ca²⁺ ions for maintenance of intracellular Ca²⁺ release and force generation in old skeletal muscle fibres, and that this age-related dependence is not due to a switch in DHPR or RyR subtype.

Necessity of external Ca²⁺ ions to maintain force

In this work, we investigated the effects of the absence of external Ca²⁺ ions on single intact FDB muscle fibre contractile force. We hypothesized that EC coupling in muscle fibres from old animals becomes dependent on external Ca²⁺. Controversy exists over whether external Ca²⁺ is necessary for adult mammalian skeletal muscle EC coupling. Some previous studies have shown a need for external Ca²⁺ to maintain tetanic force in adult skeletal muscle (Anwyl et al. 1984; Kotsias et al. 1986; Oz & Frank, 1991), while others have shown no need (Dulhunty & Gage, 1988; Tanabe et al. 1990b; García et al. 1994). These differences may exist due to the inclusion of metal ion chelators, such as EGTA, in experiments that indicate the necessity of external Ca²⁺ for maintenance of tetanic force (Anwyl et al. 1984; Kotsias et al. 1986; Oz & Frank, 1991). No examination of EC coupling dependence on extracellular Ca²⁺ ions has been conducted in skeletal muscle from ageing animals. This study is the first to examine the importance of external Ca²⁺ ions for EC coupling in ageing skeletal muscle fibres. This work examined the question without EGTA, and shows that a population of muscle fibres from old mice is, indeed, dependent on the presence of external Ca²⁺. Force decline in Ca²⁺-free solution was not due to fatigue, as the reference contraction protocol in normal 1.8 mm Ca²⁺ solution did not cause force decline in fibres from either young or old mice.

Mechanisms of decreased force in the absence of extracellular Ca²⁺

To ascertain the mechanism(s) of decreased force in fibres from old mice induced by the removal of external Ca²⁺, we examined the activation of the DHPR. Intramembrane charge movement in fibres from young or old mice showed no difference in Ca²⁺-free compared to Ca²⁺-containing solution, and showed a single population in Ca²⁺-free solution. This indicates that the activation of DHPR is unaffected by a reduction in external Ca²⁺ ions, and is still capable of coupling with RyR1 and inducing SR Ca²⁺ release.

Intracellular Ca²⁺ release was examined in two experimental preparations: in voltage-clamped fibres and in electrically stimulated contracting fibres. Voltage-clamp experiments revealed reduced intracellular Ca²⁺ transients in a population of fibres from old mice, while contraction experiments showed parallel reduction in force and intracellular Ca²⁺ in the absence of external Ca²⁺ ions. The direct relationship between Ca²⁺ transient amplitude and force generation shown here supports the conclusion that impaired force generation in fibres from old mice upon removal of external Ca²⁺ is due to impaired SR Ca²⁺ release.

Mechanisms of decreased SR Ca²⁺ release

If ageing causes a subunit switch from α1_S to α1_C similar to regenerating skeletal muscle (Pereon et al. 1997a), DHPR–RyR coupling would become cardiac-like (dependent upon influx of external Ca²⁺ ions) similar to regenerating skeletal muscle (Louboutin et al. 1995, 1996; Pereon et al. 1997b). In this case, the absence of external Ca²⁺ would prevent Ca²⁺ influx through the α1_C to trigger CICR from the RyR1. The analysis of the activation phase of the Ca²⁺ current suggests a possible age-related DHPR subunit shift. However, the I–V curves for fibres from young and old show that activation does not occur at more negative potentials for old versus young as expected for α1_C expression. The inward Ca²⁺ current becomes apparent around −20 mV and the half-activation potential for young and old is similar. Additionally, Ca²⁺ current inactivation in response to double-pulse experiments in voltage-clamped cells showed no Ca²⁺ dependence in fibres from either young or old mice. Also, none of these measures displayed two populations in fibres from old mice. These data suggest no shift in DHPR isoform from α1_S to α1_C with age. Therefore, RPA and two immunoblot techniques were performed to clarify this. These assays failed to detect DHPR α1_C mRNA or protein in muscle from either young or old mice, again suggesting no age-related shift in DHPR isoforms from α1_S to α1_C.

Faster τ_a in old muscle fibres compared to young cannot be explained here by the presence of DHPR α1_C in aged skeletal muscle. The faster current activation may be due to a number of factors. Repeat I of DHPR is very important for L-type Ca²⁺ current activation kinetics (Tanabe et al. 1991; Nakai et al. 1994). Age-related alterations (splice variations, post-translational processing and phosphorylation) in this portion of the α1 subunit of the DHPR could, conceivably, alter the Ca²⁺ current kinetics. Similarly, alterations in expression of accessory subunits may play a role in age-related alterations of Ca²⁺ current activation kinetics. For example, the β subunit increases τ_a when co-expressed with the α1 subunit in a cell transfection system (Varadi et al. 1991).

The second major protein complex involved in EC coupling at the triad is the RyR. Adult skeletal muscle expresses RyR1 almost exclusively (Marks et al. 1989; Takeshima et al. 1989). Given that a shift from RyR1/RyR3 co-expression during development to exclusive expression of RyR1 in adult skeletal muscle occurs (Flucher et al. 1999; Chun et al. 2003), it is conceivable that a shift in the opposite direction may occur with ageing. A shift in expression from RyR1 to RyR1/RyR3 co-expression would require Ca²⁺ influx through the DHPR to activate RyR3, via CICR, to release Ca²⁺ from the SR, since RyR3 is not directly coupled to DHPR (Fessenden et al. 2000; Protasi et al. 2000). However, no expression of RyR3 protein was found in either young or old mouse hindlimb skeletal muscle, suggesting that no shift occurs in RyR isoform expression with ageing.

Another potential mechanism for impaired SR Ca²⁺ release in the absence of external Ca²⁺ ions is SR Ca²⁺ depletion. If the SR becomes depleted of Ca²⁺ during the contractile protocol in the absence of external Ca²⁺ ions, the SR Ca²⁺ release and force will decline. Skeletal muscle from MG29 knockout mice has been shown to contain two apparently different Ca²⁺ storage pools – one voltage-sensitive, the other non-voltage-sensitive – as shown by caffeine contracture after a SR Ca²⁺ depletion protocol. This secondary, voltage-insensitive, caffeine-sensitive Ca²⁺ pool has been hypothesized to result from fragmentation of the SR, leaving fragments of Ca²⁺-containing SR disconnected from the t-tubule membranes (Kurebayashi et al. 2003). A smaller voltage-sensitive Ca²⁺ storage pool could conceivably lead to depletion of available Ca²⁺ during contraction in Ca²⁺-free solution. Although alterations in MG29 have not been explored in ageing skeletal muscle, similar changes in the conformation of the intracellular membranes may explain the decreased tetanic force in muscle fibres from old mice in Ca²⁺-free solution.

Physical changes to the muscle fibre membrane structure with age could also possibly lead to age-related Ca²⁺-dependent EC coupling. Certain types of heart failure show expansion in the space between t-tubules and SR terminal cisternae (Gomez et al. 1997), interfering with the efficiency of normal CICR necessary for cardiac EC coupling. A similar spacing increase in skeletal muscle could disrupt the mechanical coupling of DHPR α1_S and RyR1, changing the coupling mechanism to Ca²⁺ dependent in muscle fibres from old mice. Altered membrane structure/spacing can result from decreased expression of putative triad membrane anchoring proteins, such as MG29 (Takeshima et al. 1998), which has been shown to alter t-tubule structure and arrangement and cause EC coupling to be Ca²⁺ dependent (Nishi et al. 1999). Recent evidence shows that these membrane structures in ageing human skeletal muscle may, indeed, be altered. The number of contact points between t-tubule and SR membranes is decreased with age, and the t-tubule network shows signs of disarrangement similar to developing skeletal muscle – greater numbers of dyads and longitudinal t-tubules than in normal skeletal muscle (F. Protasi, personal communication). Similar changes to intracellular membrane structures have been shown in junctophilin-1 knockout mice – reduced number of triads in favour of dyad formation and vacuolated terminal cisternae of SR (Komazaki et al. 2002). Whether ageing leads to alterations in MG29 and junctophilin-1 protein expression, as well as other triad proteins, such as JP-45 (Anderson et al. 2003), and whether these functional effects are due to changes in expression of these or other triad proteins with age is unknown at this time. However, age-related decreased expression and/or alterations in structure, function or trafficking of MG29 and similar triad proteins may be candidate mechanisms to explain the results reported here.

In order for uncoupled DHPRs and RyR1s in aged skeletal muscle fibres to initiate Ca²⁺ release from the SR, the RyR1 must be activated, via CICR, by the Ca²⁺ influx through the DHPR. Normally, Ca²⁺ and Mg²⁺ inactivation of the RyR1 in skeletal muscle is about 20-fold more sensitive than for RyR2 in cardiac muscle (Laver et al. 1995). This allows the Ca²⁺ influx through the DHPR α1_C to activate Ca²⁺ release from RyR2 in cardiac muscle. In skeletal muscle, the cytoplasmic [Mg²⁺] may inhibit Ca²⁺ release from a normal RyR1 that is not undergoing physical interaction with (and activation by) DHPR α1_S, undermining EC coupling in these fibres. However, EC coupling does occur in these fibres in the presence of external Ca²⁺. Splice variants of the RyR1 lacking one or both of 5- and 6-amino acid segments in the modulatory region of the RyR1 responsible for Ca²⁺ or Mg²⁺ and calmodulin binding are expressed developmentally (Futatsugi et al. 1995). If the DHPR α1_S and RyR1 are indeed physically uncoupled in some aged skeletal muscle fibres, expression of one or more of these splice variants may possibly reduce the sensitivity of RyR1 Mg²⁺ inactivation allowing EC coupling to occur in a cardiac-like manner.

Age-related alterations to the DHPR α1_S molecule may also explain the Ca²⁺ dependence of EC coupling and possibly changes in Ca²⁺ current activation. Recent evidence shows that there is a splice variant of the DHPR α1_S isoform which has a 19 amino acid deletion in the repeat IV S3–S4 extracellular loop (Jurkat-Rott & Lehmann-Horn, 2004). This splice variant has been found in adult human skeletal muscle at levels at or below 10% of total DHPR transcripts. However, in myotubes regenerating from human satellite cells – a condition known to induce Ca²⁺-dependent EC coupling (Louboutin et al. 1995, 1996; Pereon et al. 1997a) – this 19 amino acid deletion transcript has been found to make up more than 66% of total DHPR transcript (K. Jurkat-Rott and F. Lehmann-Horn, personal communication). The functional significance of this splice variant and its expression level in ageing muscle have not yet been examined. If ageing muscle fibres display a shift in DHPR expression such that the deletion variant becomes dominant in some fibres, and the deleted sequence plays a role in overall DHPR function, this could help explain our findings. Previous work has shown that the addition of divalent cations such as Mg²⁺, Co²⁺ and Cd²⁺ can alter EC coupling (Dulhunty & Gage, 1989; Mould & Dulhunty, 1999, 2000), all of which may affect the binding of external Ca²⁺ by sites on the sarcolemmal or t-tubule membrane (Mould & Dulhunty, 2000). Whether this site or some other potential age-related DHPR alteration plays a role in ‘sensing’ the presence or absence of external divalent cations remains unknown at this time.

Other DHPR subunits may play a role in inducing Ca²⁺-dependent EC coupling in skeletal muscle too. Expression of the cardiac DHPR β2 subunit in skeletal DHPR β1 knockout myotubes conferred Ca²⁺-dependent EC coupling to those fibres (Sheridan et al. 2003a). Similarly, truncation of the C-terminus of the skeletal DHPR β1 subunit also caused Ca²⁺-dependent EC coupling (Sheridan et al. 2003b). Whether ageing causes a shift in the expression of DHPR β subunit isoforms or a splice variant of the DHPR β subunit exists is unknown at this time.

The question remains: what is the consequence for old muscle fibres of becoming dependent on external Ca²⁺ ions for EC coupling? Cardiac myocytes are dependent on the influx of external Ca²⁺ for EC coupling (Fabiato, 1985; Nabauer et al. 1989) and undergo billions of contractions throughout a lifetime. As the t-tubules of skeletal muscle fibres are much smaller in diameter (approximately 20 nm; Franzini-Armstrong et al. 1975) than cardiac muscle fibres (approximately 250 nm; Soeller & Cannell, 1999), the Ca²⁺ buffering ability of the skeletal muscle t-tubule lumen is very low (Almers et al. 1981), and the coefficient for diffusion of Ca²⁺ in the t-tubule is very low (Almers et al. 1981), the possibility exists that Ca²⁺ is depleted from the t-tubule during a series of tetanic contractions. A muscle fibre that is dependent upon the influx of these Ca²⁺ ions for EC coupling would thus be at a force-generating disadvantage compared to normal skeletal muscle fibres that are not dependent upon the influx of Ca²⁺ from the t-tubule. Therefore, we propose that the Ca²⁺ dependence of EC coupling in a population of skeletal muscle fibres from ageing mice described herein contributes to a decline in force production in fibres undergoing repetitive contraction.