Increased Ca_Vβ_1a Expression with Aging Contributes to Skeletal Muscle Weakness

Summary

Ca²⁺ release from the sarcoplasmic reticulum (SR) into the cytosol is a crucial part of excitation-contraction (E-C) coupling. E-C uncoupling, a deficit in Ca²⁺ release from the SR, is thought to be responsible for at least some of the loss in specific force observed in aging skeletal muscle. E-C uncoupling may be caused by alterations in expression of the voltage-dependent calcium channel α_1s (Ca_V1.1) and β_1a (Ca_Vβ_1a) subunits, both of which are necessary for E-C coupling to occur. While previous studies have found Ca_V1.1 expression declines in old rodents, Ca_Vβ_1a expression has not been previously examined in aging models. Western blot analysis shows a substantial increase of Ca_Vβ_1a expression over the full lifespan of FVB mice. To examine the specific effects of Ca_Vβ_1a overexpression, a Ca_Vβ_1a -YFP plasmid was electroporated in vivo into young animals. The resulting increase in expression of Ca_Vβ_1a corresponded to decline of Ca_V1.1 over the same time period. YFP fluorescence, used as a measure of Ca_Vβ_1a -YFP expression in individual fibers, also showed an inverse relationship with charge movement, measured using the whole-cell patch-clamp technique. Specific force was significantly reduced in young Ca_Vβ_1a - YFP electroporated muscle fibers compared to sham-electroporated, age-matched controls. siRNA interference of Ca_Vβ_1a in young muscles reduced charge movement, while charge movement in old was restored to young control levels. These studies imply Ca_Vβ_1a serves as both a positive and negative regulator Ca_V1.1 expression, and that endogenous overexpression of Ca_Vβ_1a during old age may play a role in the loss of specific force.

INTRODUCTION

Depolarization of the sarcolemma leads to muscle fiber contraction and the generation of mechanical force in a process called excitation-contraction (E-C) coupling. Two key proteins involved in E-C coupling are: dihydropyridine receptor (DHPR) and ryanodine receptor (RyR). DHPR serves as a modest L-type Ca²⁺ channel but is primarily known for its function as a voltage sensor. DHPRs are located within invaginations of the sarcolemma called t-tubules, and are arranged into clusters of four known as tetrads. Each tetrad is positioned directly across from a single RyR, which are embedded within the membrane of the adjacent sarcoplasmic reticulum (SR). DHPRs contain four domains, each composed of six-transmembrane spanning segments. The S4 segment of each domain contains charged amino acid residues, and these residues respond to membrane depolarization by undergoing a conformational shift. This shift results in a proposed physical interaction of DHPR with RyR, causing RyR to open and release intracellular Ca²⁺ stores from the SR, allowing muscle contraction to occur (for review, see Melzer et al. 1995).

The decline in muscular strength with age, known as sarcopenia, is caused largely by a loss of total muscle mass - but also a disproportionate loss of strength. This loss of specific force (total force/cross sectional area) in old age (Brooks & Faulkner 1988; Gonzalez et al. 2000) is characterized in part by a deficit in Ca²⁺ release following depolarization (Delbono et al. 1995; Jimenez-Moreno et al. 2008), a phenomenon known as E-C uncoupling. E-C uncoupling is not a result of decreased Ca²⁺ stores or RyR release function (Jimenez-Moreno et al, 2008), and therefore may be caused by alterations in the functionality and expression of DHPR and its subunits with aging. The primary DHPR subunit in skeletal muscle is Ca_V1.1, previously known as DHPRα_1s (Catterall et al. 2005). Ca_V1.1 is a large transmembrane protein which contains both the Ca²⁺ conducting pore and the voltage sensing S4 domain. Four other auxiliary subunits bind Ca_V1.1 to make up DHPR (for review, see Flucher et al. 2005), with the most widely studied being the cytosolic Ca_Vβ_1a subunit. Ca_Vβ_1a, a muscle specific member of the Ca_Vβ family of proteins, binds to a region of the I–II intracellular loop of Ca_V1.1 known as the alpha interaction domain (AID) (Chen et al. 2004). Ca_Vβ_1a is classically described by its role in chaperoning Ca_V1.1 to the plasma membrane and regulating L-type Ca²⁺ current (Gregg et al. 1996; Strube et al. 1996; Beurg et al. 1997; Neuhuber et al. 1998). Most notably, E-C coupling cannot occur without Ca_Vβ_1a (Gregg et al. 1996). Ca_Vβ_1a binds to charged residues on RyR (Cheng et al. 2005) and neutralization of these residues reduces E-C coupling, suggesting a direct interaction with RyR. The correct organization of Ca_V1.1 into tetrads within the t-tubule membrane is also a specific function of the Ca_Vβ_1a isoform (Schredelseker et al. 2005).

Although classically known for augmenting the expression and function of Ca_V1 subfamily of calcium channels, the Ca_Vβ family of subunits may contribute to the down-regulation of Ca_V1 as well. A family of Ras-related G-proteins (RGKs) mediate the down-regulation of several Ca_V1 isoforms in a Ca_Vβ dependent manner (Beguin et al. 2001). Additionally, the previously uncharacterized SH3 domain of Ca_Vβ was shown to bind dynamin and mediate endocytosis of Ca_V1.2 (Gonzalez-Gutierrez et al. 2007).

As previous studies have shown that the Ca_V1.1 subunit declines in old rodents (Renganathan et al. 1997; Moreno et al. 2006; O'Connell et al. 2008) and this causes an impairment of E-C coupling (Renganathan et al. 1997), we wanted to investigate what effects aging had on Ca_Vβ_1a expression, as this subunit is also critical for E-C coupling. We have found that Ca_Vβ_1a expression is highly increased in old mice, and that experimental overexpression of Ca_Vβ_1a reduces both the expression of Ca_V1.1 and specific force in dissociated single fibers of young mice. Additionally, siRNA inhibition of Ca_Vβ_1a restores charge movement in aged muscle. These findings suggest that overexpression of Ca_Vβ_1a with aging contributes to E-C uncoupling by reducing the level of Ca_V1.1.

RESULTS

Ca_Vβ_1a subunit expression increases with age

As our laboratory had earlier shown that Ca_V1.1 expression declines in old animals (Renganathan et al. 1997), we investigated whether Ca_Vβ_1a might also change with age. Hindlimb muscles from FVB mice were harvested at 2, 6, 14, and 24 months of age and subjected to a membrane extraction protocol (see Methods) for use in western blot analyses. Gluteus and hamstring muscles (designated “pool”) contain a mixture of type I (slow) and type II (fast) fiber types (Fig. 1A), while tibialis anterior (TA) and extensor digitorum longus (EDL) muscles contain only type II fibers (Manttari and Jarvilehto, 2005; Augusto et al., 2004) (Fig. 1B). Actin was used as a loading control as it appears to remain relatively stable throughout the lifespan of FVB mice. In order to precisely compensate for any variations in loading or possible global alterations in protein expression with aging, the optical density of each Ca_V1.1 band was divided by that of the corresponding actin band. The resulting values confirm a significant increase of Ca_Vβ_1a with age. Relative Ca_Vβ_1a expression increased continuously during aging, approximately doubling at each time point. This culminated in a substantial increase between very young and very old age (Fig. 1C) in both muscle groups, though the effect was less pronounced in the TA/EDL muscle groups (~7-fold), than in pool (~10-fold). The mean normalized intensity values (arbitrary units, AU) for each age group (n=4) are as follows: 0.20 ± 0.02 (2 months), 0.49 ± 0.06 (6 months), 1.14 ± 0.25 (14 months), and 2.34 ± 0.49 (24 months) for pool muscles and 0.29 ± 0.09 (2 months), 0.51 ± 0.1 (6 months), 0.84 ± 0.27 (14 months), and 1.81 ± 0.41 (24 months) for TA/EDL muscles.

Fig. 1. Relative Ca_Vβ_1a expression across the lifespan of FVB mice.

(A) Representative immunoblot of Ca_Vβ_1a (52 kDa) expression in P100 fractions collected from gluteus and hamstring muscles (pool) at 2, 6, 14, and 24 months of age (24 months represents the very old age of this strain). Below is the mean Ca_Vβ_1a band intensity for each age group (n=4), normalized to the intensity of the corresponding actin band. Each age group reached a statistically significant difference from all of the others, with the exception of 6 and 14 month age groups. Data represent mean ± SE. *P ≤ 0.01 (one-way ANOVA). (B) The same as A, except representative of TA and EDL muscle groups (n=4). The 24 month age group reached significant difference from all other ages. *P < 0.01(one-way ANOVA). (C) Mean fold increase of pool and TA/EDL samples by age. Values represent normalized sample values from each age group divided by the corresponding 2 month normalized value.

Overexpression of Ca_Vβ_1a results in decreased Ca_V1.1 expression

We next sought to determine if the observed increase in Ca_Vβ_1a expression was directly involved in the age-related decline in Ca_V1.1. In order to separate the effects of Ca_Vβ_1a overexpression from all other age-related changes, a Ca_Vβ_1a-YFP plasmid was electroporated into the TA muscle of young (4–5 month old) FVB mice in vivo. For these experiments we used TA because it is a superficial muscle suitable for electroporation in vivo, and large enough to provide tissue for protein analysis. Western blots of membrane fractions from electroporated muscles at 3 days (n=4), 7 days (n=3), and 2 weeks (n=4) confirm an increase in expression of endogenous Ca_Vβ_1a, which continued to rise during the two weeks following in vivo electroporation (Fig 2A). Conversely, Ca_V1.1 expression levels from the same samples declined steadily over the two week time course. Quantification of Ca_V1.1 band optical density shows an approximately 50% decrease compared to control at two weeks following Ca_Vβ_1a-YFP electroporation (Fig 2B). Sham electroporated muscles (n=4 for each time point) exhibited no significant changes in Ca_V1.1 or Ca_Vβ_1a expression relative to non-electroporated controls. Additionally, quantitative real time RT-PCR was used to assess Ca_V1.1 transcript levels at 2 weeks after Ca_Vβ_1a-YFP electroporation (Fig 2C). No difference was seen in Ca_Vβ_1a-YFP electroporated (n=4) vs. controls (n=3), suggesting that Ca_V1.1 down regulation during Ca_Vβ_1a overexpression occurs only at the protein level. These results are significant for two reasons. First, they show use of the Ca_Vβ_1a-YFP plasmid to artificially overexpress endogenous Ca_Vβ_1a in vivo. Second, Ca_Vβ_1a overexpression directly correlates to a decline in Ca_V1.1 protein levels in young mice, thus providing evidence for Ca_Vβ_1a involvement in Ca_V1.1 down-regulation; namely when present in high levels such as those observed during senescence.

Fig. 2. Ca_V1.1 and Ca_Vβ_1a expression in TA muscles following in vivo electroporation of a Ca_Vβ_1a-YFP plasmid.

(A) Representative immunoblots from control muscle (lane 1) and from muscles harvested 3 days (lane 2), 1 week (lane 3), and 2 weeks (lane 4) post-electroporation. (B) Ca_V1.1 and endogenous Ca_Vβ_1a protein expression in muscle following Ca_Vβ_1a-YFP electroporation (n=4 for 3 days and 2 weeks, n=3 at 1 week), normalized to control (non-electroporated) muscle. Sham electroporation (n=4 at each time point) produced no significant difference in relative Ca_V1.1 or Ca_Vβ_1a expression. Data represent mean ± SE. *P <0.001 (one-way ANOVA). (C) Ca_V1.1 mRNA levels measured using qRT-PCR in control (non electroporated, n=3) and Ca_Vβ_1a-YFP electroporated (2 weeks, n=4) animals.

Ca_Vβ_1a -YFP intensity corresponds to reduced charge movement

The intensity of YFP fluorescence (F, arbitrary units) varied between individual FDB fibers following in vivo electroporation with the Ca_Vβ_1a-YFP plasmid. Thus, the relative level of YFP intensity can be used to represent differences in individual fiber’s Ca_Vβ_1a-YFP expression level. Similarly, charge movement (Q) is a measure of single cell Ca_V1.1 expression at the plasma membrane (Wang et al. 2000; for review, see Rios & Pizarro 1991). Maximal charge movement (Q_max) was plotted against the corresponding fiber’s YFP intensity (Fig 3A). The resulting plot shows an inverse relationship between Ca_Vβ_1a-YFP fluorescence level and Ca_V1.1 membrane expression (r=0.91, n=10), where electroporated fibers with high YFP fluorescence (F > 40 AU) show increasingly impaired Q_max. Fibers with relatively low YFP fluorescence (F ≤ 40 AU) are presumed to exhibit minimal overexpression of Ca_Vβ_1a -YFP, and subsequently showed no loss of Q_max. Fibers from muscles electroporated with a GFP plasmid (n=10) showed no difference in Q_max, regardless of relative F intensity. Figure 3B and Figure 4 display an example of two fibers with different YFP intensity, and the resulting difference in Q. Fiber D0912, a fiber with lower Ca_Vβ_1a -YFP expression (F=42 AU) exhibited unimpaired maximal charge movement (Q_max =52.6 nC/µF). Conversely, fiber C0925 showed high levels of Ca_Vβ_1a -YFP (F=71.3 AU) and substantially reduced maximal charge movement (Q_max =26.6 nC/µF). While the absolute Q-V relationship was reduced in fiber C0925 and others showing high AU (Fig. 4D), the relative Q-V relationship was unchanged (Fig. 4E). Therefore overexpression of Ca_Vβ_1a -YFP does not shift the Q-V curve in the voltage axis, reaffirming the notion that the effects of Ca_Vβ_1a-YFP are on Ca_V1.1 membrane expression and not other alterations of the channel’s function.

Fig. 3. Ca_V1.1 charge movement (Q) and YFP (Ca_Vβ_1a -YFP) fluorescence (F) in electroporated FDB fibers.

(A) Q_max-F relationship of single fibers electroporated with Ca_Vβ_1a –YFP (●)(n=10) or GFP (○)(n=10). Detection of F above ~40 AU correlates with decreased charge movement (r²=0.91). (B) Top - two representative fibers from A. Fiber D0912 exhibits F=42 AU, Q=52.6 nC/µF. Fiber C0925 exhibits F=71.3 AU, Q=26.6 nC/µF. (Also see Fig. 4 for more detail on these two fibers). Fluorescence acquisition parameters on confocal microscope: 20x objective, zoom = 3.5x, Iris = 3.8mm, Gain = 80, box size = 512×512 pixels (172.6×172.6 µm), pixel dwell time 37.76 µs. Excitation wavelength = 488 nm, Emission = 528 nm. Bottom – Illustration of the fluorescence intensity across the fibers.

Fig. 4. Charge movement from FDB fibers expressing Ca_Vβ_1a-YFP.

(A – B) Charge movement recordings from fiber D0912 (A), representing a fiber expressing low levels of Ca_Vβ_1a -YFP (indicated by F intensity in Figure 3), and from fiber C0925 (B), representing a fiber expressing high levels of Ca_Vβ_1a-YFP (indicated by F intensity in Figure 3). (C) The voltage step protocol. The holding potential was −80 mV. A 2-sec pre-pulse at −30 mV was followed by a 15-ms repolarization to −50 mV, followed by a 25-ms test pulse of varying membrane voltage and repolarization to −50 mV. (D) Absolute Q-V relationship of fibers D0912 (●) and C0925 (○). Note that the fiber which expresses higher levels of Ca_Vβ_1a-YFP displays lower charge movement. (E) Relative Q-V relationship of the same fibers in D. Q values were normalized to Q_max for each fiber. Overexpression of Ca_Vβ_1a -YFP does not shift the Q-V curve in the voltage axis or modify the steepness of the curve. Data points were fitted to a Boltzmann equation of the form: Q = Q_max/ [1+ exp(V_1/2 − V_m)/k], where Q_max is the maximal charge; V_1/2 is the charge half-activation potential; V_m is the membrane potential; and k is the steepness of the curve. Q_max, V_1/2, and K values were: 56.9 nC/µF, 12.3 mV and 3.37 for fiber D0912, and 30.2 nC/µF, 17.1 mV, and 14.9 for fiber C0925.

Ca_Vβ_1a -YFP causes a decline in specific force

Loss of muscle strength with aging is thought to be partially caused by reduction of Ca_V1.1, specifically those coupled to RyR in the T-tubule membrane, thereby resulting in an impairment of E-C coupling. As Ca_Vβ_1a overexpression lead to a marked reduction Ca_V1.1, shown by both western blot and charge movement studies, it seemed likely that Ca_Vβ_1a overexpression would also result in a loss of specific force. Tetanic specific force (kPa) was measured 14 days after single fibers from FDB muscles were electroporated in vivo with either Ca_Vβ_1a-YFP (n=7), GFP (n=7), or non-electroporated (n=3) (Fig 5A). The electroporation process itself causes short-term damage to the muscle fiber in the days immediately following the procedure (Schertzer et al. 2006) and as such the GFP group showed a significant reduction in specific force compared to controls. However, the Ca_Vβ_1a-YFP group showed an even greater and statistically significant decline in specific force compared to the YFP electroporated fibers. Figure 5B shows representative tetanic contraction traces from Ca_Vβ_1a-YFP, GFP, and control groups. These results support the idea that overexpression of Ca_Vβ_1a impairs single fiber specific force by reducing the amount of Ca_V1.1 expressed at the plasma membrane. No significant difference was found on average time to peak torque (ms), or half relaxation time (ms), between control, GFP, and Ca_Vβ_1a-YFP electroporated groups (data not shown), indicating no alterations in calcium buffering activity or rate of cross bridge formation. Twitch specific force (kPA) appeared to decline in Ca_Vβ_1a-YFP electroporated fibers compared to GFP and non-electroporated controls, although the reduction did not reach significance (p=0.081).

Fig. 5. Specific force of single intact FDB fibers from young mice electroporated and expressing Ca_Vβ_1a-YFP or GFP.

(A) Specific force (kPa) of single intact FDB fibers from Ca_Vβ_1a-YFP (n=7), GFP (n=7) and non-electroporated controls (n=3). Ca_Vβ_1a-YFP expressing fibers display significantly lower specific force compared to GFP expressing fibers. Data represent mean ± SE. *P<0.05 (one-way ANOVA) (B) Representative tetanic contraction traces from fibers expressing Ca_Vβ_1a–YFP, GFP, and control groups.

siRNA inhibition of Ca_Vβ_1a

If overexpression of Ca_Vβ_1a is responsible for muscle impairment with aging, then inhibition of Ca_Vβ_1a may provide a way to reverse these effects. We created several siRNA sequences against Ca_Vβ_1a (Fig 6A). The effectiveness of these sequences was determined by transfecting each of them into separate C2C12 cell cultures and performing western blot analyses on the harvested cell lysates. Each sample was probed for Ca_Vβ_1a, and the optical density of each band was compared to that of a control, untransfected sample. Clones 69051 and 69052 caused the greatest decline in Ca_Vβ_1a expression (37% and 28% decline, respectively). Both of these clones were then electroporated in vivo into the FDB muscle of young and old FVB mice (Fig. 6B). In young animals, siRNA against Ca_Vβ_1a caused a significant reduction of charge movement in dissociated FDB cells, recorded 7–11 days post electroporation. As demonstrated in Fig 3, electroporation of GFP alone does not result in a loss of charge movement. Old (24 month) mice naturally exhibit reduced charge movement, which was confirmed by our results. Interestingly, siRNA against Ca_Vβ_1a restored charge movement of old mice to young control levels. Sham electroporation of old muscles did not result in a further decline in charge movement, suggesting that the deleterious effect of electroporation on force reported above is due primarily to damage to contractile proteins and not due to reduced Cav1.1 expression.

Fig. 6. siRNA restoration of charge movement in vivo.

(A) Effect of five different siRNA sequences on Ca_Vβ_1a expression in C2C12 myotubes. Data are presented as percent change in immunoblot band intensity versus control. Clones 69051 and 69052 cause the greatest decline in endogenous Ca_Vβ_1a expression (37% and 28% decline, respectively). (B) Charge movement from FDB muscle fibers 7–11 days post-electroporation. FDB muscles were electroporated with 20 µg of siRNA sequence 69051 and 10 µg of siRNA sequence 69052. Fibers were dissociated, patch-clamped, and subjected to the same protocol as in Figure 4. The number of fibers examined for young control, sham GFP-electroporated and Ca_Vβ_1a siRNA electroporated was 13, 10, and 17, respectively, and for old control, Ca_Vβ_1a siRNA and sham electroporated was 21, 16 and 11, respectively. Data represent mean ± SE. *P < 0.05 (one-way ANOVA).

DISCUSSION

Here we show that Ca_Vβ_1a is significantly overexpressed in aging muscle, and present a model by which this phenomenon may contribute to loss of specific force with aging. In young mice, overexpression of Ca_Vβ_1a corresponds to decreased Ca_V1.1 expression at the sarcolemma, as shown by both western blot and charge movement studies. Ca_Vβ_1a overexpression also results in a loss maximal specific force, presumably by reducing the number of Ca_V1.1 channels in the t-tubule membrane coupled to RyRs. Because Ca_V1.1 subunits are critical for the transduction of sarcolemmal depolarizations into Ca²⁺ release from the SR, a decreased number of Ca_V1.1 subunits in the membrane would cause less Ca²⁺ to be released from intracellular stores, resulting in weakened contractions (Delbono, 2002). To further implicate overexpression of Ca_Vβ_1a directly with decreased Ca_V1.1, we show that using siRNA to partially inhibit Ca_Vβ_1a in old muscle restores charge movement to young control levels. Expectedly, inhibition of Ca_Vβ_1a in young muscle fibers significantly reduced charge movement. Thus, both the under and overexpression of Ca_Vβ_1a results in reduced expression of Ca_V1.1.

Ca_Vβ_1a Overexpression with Aging

While the most striking increase in Ca_Vβ_1a expression appears to be between middle aged and very old mice, the rate of Ca_Vβ_1a increase is surprisingly uniform, with the normalized expression level roughly doubling at each time point. Relative Ca_Vβ_1a level in very young mice is substantially lower than the level seen even at 6 months of age (young adulthood). Thus, increasing Ca_Vβ_1a expression may also be a necessary component of muscle development. It is therefore tempting to speculate that the extreme levels of Ca_Vβ_1a seen in very old animals may reflect a developmental program hyperfunction, a common attribute of aging (Blagosklonny, 2006). Interestingly, the rate of increase of Ca_Vβ_1a expression did not seem to be as high in the TA/EDL muscle group compared to that of the Pool group. TA and EDL muscles are of exclusively type II composition, while there is a mixture of type I and type II fibers found in the gluteus and hamstring muscles used for our pool group. The possible resistance to Ca_Vβ_1a overexpression seen in the type II muscle group is perplexing, as type I fibers are thought to be more resistant to age related changes such as denervation (for review, see Larsson, 1995). One possible explanation of this could be due to the increased cell turnover seen in type II fibers, as indicated by increased apoptotic signaling (Phillips and Leeuwenburgh, 2005) and depletion of satellite cells with age (Verdijk et al., 2007) compared to type I fibers, thus somewhat limiting the accumulation of Ca_Vβ_1a with age. However, the mouse is not a preferred model to examine fiber type due to the lack of pure type I muscles as seen in the rat, and thus any conclusions on the fiber type-specific rate of Ca_Vβ_1a increase are limited. Still, the notion that Ca_Vβ_1a is highly overexpressed in both fiber types is strongly supported by these experiments.

The two most likely explanations of Ca_Vβ_1a overexpression with age are an increase in its transcription levels, or a failure in its ability to be properly degraded. In regards to the latter, impairment of proteolysis is already known to be a hallmark of aging muscle (for review, see Combaret et al. 2009). Further experiments examining Ca_Vβ_1a mRNA levels during aging should shed light on whether transcriptional upregulation plays any role. These explanations are not mutually exclusive and indeed it would seem probable that both contribute to the accumulation of Ca_Vβ_1a in old muscle. Although we present a model by which Ca_Vβ_1a overexpression causes deleterious effects on muscle function by reducing the number of voltage sensing Ca_V1.1 subunits in the t-tubule membrane, the relative expression of Ca_Vβ_1a in old muscle is so high that it may also interfere with other key processes in a nonspecific manner. Ca_Vβ subunits contain two conserved protein interaction domains: GK and SH3 (Chen et al., 2004) and have recently been shown to interact with several other proteins besides Ca_V1, such as RGKs (kir/Gem, Rad, Rem), dynamin (for review, see Hidalgo and Neely, 2007), chromobox protein 2/heterochromatin protein 1γ (Hibino et al., 2003), and Akt (Catalucci et al., 2009). It is therefore possible that Ca_Vβ_1a may interact with additional proteins in a currently uncharacterized manner.

Potential mechanisms of Ca_Vβ_1a involvement in Ca_V1.1 down-regulation

The notion that Ca_V1.1 expression declines during old age is supported by several studies (Renganathan et al.1997; Ryan et al. 2000; Moreno et al. 2006; Wang et al. 2007; O’Connell et al. 2008). Importantly, Ca_V1.1 mRNA does not decrease significantly with age (Zheng et al. 2001), implicating some other mechanism responsible for the decline in its protein level. Here we demonstrate that Ca_V1.1 expression also declines following induced overexpression of Ca_Vβ_1a. Thus, Ca_Vβ_1a overexpression, both during natural aging and experimental overexpression in young cells, coincides with a decline in Ca_V1.1 expression. While we present evidence that Ca_Vβ_1a overexpression causes a decline of Ca_V1.1 at the protein level, the precise mechanism behind this occurrence is not known. Ca_V1.1 mRNA does not decline in young mice following Ca_Vβ_1a overexpression, suggesting Ca_Vβ_1a does not regulate Ca_V1.1 gene expression, in agreement with the aforementioned aging studies on Ca_V1.1 mRNA. Due to the multiple regulatory functions that Ca_Vβ_1a exerts on Ca_V1.1, both classically and those discovered more recently, there are several hypothetical mechanisms by which overexpression of Ca_Vβ_1a could lead to a reduction of Ca_V1.1. The two most important classical functions of Ca_Vβ_1a in skeletal muscle are increasing the trafficking of newly formed Ca_V1.1 subunit to the t-tubule membrane (Neuhuber et al. 1998; Bichet et al. 2000), and arranging DHPRs into orthogonal arrays, or tetrads (Schredelseker et al. 2005), which are presumed to be necessary for skeletal muscle E-C coupling. Interfering with tetrad formation is one possible way that Ca_Vβ_1a overexpression could reduce Ca_V1.1 insertion into the membrane. Tetrads are critical for the precise alignment of DHPRs with RyR, which is required for proper transduction during E-C coupling. Because Ca_Vβ_1a also binds to RyR (Cheng et al. 2005), it seems possible that overexpression of Ca_Vβ_1a could interfere with the alignment and coupling of DHPR tetrads to RyR. Also, our group in collaboration with others has found that Ca_Vβ_1a interacts with the junctional protein JP-45 (Anderson et al. 2006). Like Ca_Vβ_1a, JP-45 is localized to the t-tubule/SR junction (triad) and interacts with Ca_V1.1 via the AID. In co-immunoprecipitation experiments, treatment with exogenous purified Ca_Vβ_1a reduces the ability of JP-45 to pull down Ca_V1.1, suggesting Ca_Vβ_1a interferes with this interaction. As JP-45 KO mice have impaired muscle strength due to reduced levels of Ca_V1.1 (Delbono et al. 2007), disruption of the JP-45-Ca_V1.1 interaction associated with excess Ca_Vβ_1a is another possible mechanism by which Ca_Vβ_1a overexpression reduces the level of Ca_V1.1 at the membrane. Disruption of the DHPR tetrad – RyR complex, through any of the mechanisms mentioned above, could in turn make Ca_V1.1 more susceptible to endocytosis and proteolytic degradation.

Much attention has been given recently to the involvement of the RGK family of Ras-related GTP-binding proteins in Ca_V1 channel function and expression. Ca_Vβ isoforms are necessary for the RGK (kir/Gem) mediated down-regulation of Ca_V1.2 (Beguin et al. 2001), and Ca_Vβ_2a forms a trimeric complex with the RGK Rem and the Ca_V1 AID (Finlin et al., 2006). In skeletal muscle, overexpression of Rem leads to a decline in the number of Ca_V1.1 channels in the membrane (Bannister et al. 2008). Thus overexpression of Ca_Vβ_1a may contribute to Ca_V1.1 down-regulation via interaction with RGKs, although the necessity for up-regulation of Ca_Vβ_1a itself is unclear. Interestingly, Beguin and colleagues (2006) have recently shown that the RGKs Rad and Rem may sequester Ca_Vβ3 subunits into the nucleus in a calmodulin and 14-3-3 regulated manner. A previous study by Colecraft et al. (2002) also showed the presence of Ca_Vβ_1b, Ca_Vβ_2a, and Ca_Vβ₄ isoforms in the nucleus of cardiomyocytes. In additional to nuclear localization, Hibino et al. (2003) demonstrated that the Ca_Vβ₄ splice variant Ca_Vβ_4c acts as a transcriptional regulator by binding to and inhibiting the gene silencing ability of the nuclear protein CHCB2/HP1γ. Although we found no influence of Ca_Vβ_1a on Ca_V1.1 mRNA levels, based on these observations it is worth speculating on the possibility of Ca_Vβ_1a acting as transcription factor, perhaps even in a self regulating fashion.

The GK domain of Ca_Vβ subunits binds with high affinity to the AID of Ca_V1 channels (Chen et al.2004). While this is the consensus binding site of the classically defined Ca_V1- Ca_Vβ complex, there is evidence that an additional Ca_Vβ subunit can bind to a region on the C-terminus of Ca_V1 channels (in addition to the high affinity AID) and that this secondary binding modulates the functional properties of Ca_V1 (Qin et al. 1996; Birnbaumer et al. 1998; Gerster et al. 1999; Canti et al. 2001; García et al. 2005). The question of Ca_V1 containing multiple Ca_Vβ binding sites is particularly relevant to our present results. Colecraft’s group present model (Dalton et al. 2005) in which Ca_Vβ initially traffics Ca_V1 to the plasma membrane, then either remains bound to the AID or dissociates, producing two populations of Cav1 with either high (Ca_Vβ associated) or low (Ca_Vβ-less) gating activity. However Colecraft’s model does not exclusively rule out the possibility of one or more additional binding sites on Ca_V1. The possibility of Ca_V1 containing multiple Ca_Vβ binding sites of differing affinities offers a logical explanation for the pleiotropic effect of Ca_Vβ_1a shown here. If Ca_Vβ_1a is present in high concentrations (eg; exogenously applied or due to overexpression with aging), it may then be able to bind the proposed secondary, low affinity binding site with increasing frequency. In additional support of this hypothesis, the SH3 domain of Ca_Vβ2 has recently been shown to interact with dynamin to promote endocytosis of Ca_V1.2 (Gonzalez-Guiterrez et al,. 2007). Interestingly, this group found that disruption of the AID was necessary for the Ca_Vβ2-dynamin mediated endocytosis, further supporting the possibility that the presence of Ca_Vβ at a secondary binding site results in Ca_V1 endocytosis. Alternatively, rather than directly mediating channel endocytosis via protein-protein interactions, Ca_Vβ overexpression may alter Ca_V1 expression indirectly via Ca²⁺ dependent mechanisms. Garcia et al. (2005) showed that pressure injection of purified Ca_Vβ_1a into dissociated muscle fibers produces a rapid increase in both L-type Ca²⁺ current and intracellular Ca²⁺ release, without altering charge movement. Endogenous overexpression of Ca_Vβ_1a in the short term may produce similar physiological effects. In turn, a chronic increase in intracellular Ca²⁺ may activate some form of negative feedback, perhaps culminating in Ca_V1.1 endocytosis and proteolysis. Indeed, Sanchez’s group has also shown that long-term increase of L-type Ca²⁺ current in skeletal muscle leads to proteolytic down-regulation of Ca_V1.1, likely via a local Ca²⁺-dependent protease such as calpain (Carrillo et al. 2004).

Sarcopenia is a major cause of loss of independence in the elderly and presents a substantial public health cost (Janssen et al. 2004). As sarcopenia is caused in part by muscle dysfunction beyond the obvious loss in mass, E-C uncoupling may be a significant contributor to sarcopenia in aging humans. E-C uncoupling appears to be primarily caused by a decline in functional Ca_V1.1 subunits, which leads to an impairment of the mechanical coupling between neural signals and Ca²⁺ release necessary for contraction. Here we have shown that another protein, Ca_Vβ_1a, with several regulatory influences on Ca_V1.1, also exhibits significant changes in expression level with aging. Combined with evidence that Ca_Vβ_1a overexpression causes Ca_V1.1 decline in young muscle, our findings present a potentially novel and physiologically significant contributing factor to the loss of skeletal muscle strength with aging.

Experimental Procedures

Animals

Muscles were dissected from FVB (Friend Virus B, our colony) mice between 1.5 and 24 months of age. FVB mice have a maximum lifespan of 25 months and have been used previously as a model of aging skeletal muscle in our laboratory (Renganathan et al. 1998; Payne et al. 2004). 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.

Microsome preparation

Isolation of the t-tubule membrane was performed using a modified version of the protocol by (Knudson et al. 1989). Briefly, whole muscles were dissected and pulverized in liquid nitrogen and then homogenized in ice-cold Buffer A (20 mM sodium pyrophosphate, 20 mM sodium phosphate monobasic, 1 mM MgCl₂, 0.5 mM EDTA, 303 mM sucrose with complete protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA) using a handheld tissue tearor. Homogenate was centrifuged at 7,000 g for 20 min at 4°C, and the pellet discarded. Supernatant was filtered through four layers of cheesecloth and centrifuged at 100,000 g for 90 min at 4°C in a Beckman Type Ti.70i rotor. The pellet was rinsed with ice cold PBS and resuspended with a glass homogenizer in fresh Digitonin buffer (1% digitonin (w/v), 185 mM KCl, 1.5 mM CaCl₂, 10 mM HEPES pH 7.4 with complete protease inhibitor cocktail). Samples were left on ice for 1h and then vortexed. Protein concentration was measured using bicinchoninic protein assay using BSA digitonin standards.

Antibodies

Primary antibodies used for immunoblots were monoclonal VD2₁ to Ca_Vβ_1a (Developmental Studies Hybridoma Bank, Iowa City, IA, USA), monoclonal IIF7 to Ca_V1.1; a generous gift from Dr. Kevin P. Campbell of the University of Iowa, and actin (Chemicon International, Temecula, CA, USA). NA931V goat anti-mouse (Amersham Health, Little Chalfont, Buckinghamshire, UK) was used as a secondary antibody.

Western blots

For Ca_Vβ_1a subunit, microsomes were mixed with equal volume β-mercaptoethanol buffer and boiled for 5 minutes. For Ca_V1.1, microsomes were mixed with equal volume double strength Urea buffer and incubated at room temperature for 30 min (Murray & Ohlendieck 1998).SDS-PAGE was conducted using a 4.5 % stacking gel with a 10% resolving gel in a Mini-Protean gel system (BioRad Laboratories, Hemel-Hemptstead, Herts., UK). Gels were transferred to PVDF membranes (Amersham Health, Little Chalfont, Buckinghamshire, UK) overnight at 4°C. Blots were blocked in 5% non-fat dry milk with 0.1% Tween in TBS for Ca_Vβ_1a and PBS for all other antibodies. Primary antibody concentrations were as follows: 1:1000 (Ca_Vβ_1a), 1:5000 (Ca_V1.1) and 1:250,000 for actin. Secondary antibodies were used at a 1:5000 dilution. Band intensity was measured using Kodak Gel Doc imaging system.

Quantitative Real Time RT-PCR

Total RNA was isolated using TriReagent according to the manufacturers protocol (Molecular Research Center, Cincinnati, OH), treated with DNAse I (New England Biolabs, Ipswich, MA) and reverse transcribed to cDNA using random hexamers and reverse transcriptase (Promega, Madison, WN) according to manufacturers protocol. Following reverse transcription, cDNA expression was assessed by quantitative real time RT-PCR (Taqman Gene Expression Master Mix kit, Applied Biosystems, Foster City, CA, USA) on a Stratagene MX3000P. Primers and probes for Ca_V1.1 were purchased from Applied Biosystems.

Muscle Electroporation

Intramuscular plasmid injection and electroporation were performed according to Schertzer et al. (2006) and DiFranco et al. (2006). Briefly, FDB or TA muscles were injected with 30 Al of 0.5 U/Al hyaluronidase and injected 1 hr later with 20 µg Ca_Vβ_1a–YFP or 20 µg Ca_Vβ_1a siRNA equal volume saline solution. A pair of platinum plate electrodes was placed under the skin on adjacent sides of the muscle. Eight, 150 V, 20-ms square-wave pulses of 1-Hz frequency were generated using a Grass stimulator (Grass S48; W. Warwick, RI, USA) and delivered to the muscle. The polarity was then reversed and a further 8 pulses were delivered to the muscle. For sham electroporations, the same protocol was followed with a saline-only injection.

siRNA sequences

C2C12 cells were transfected with five siRNA sequences (Open Biosystems, Huntsville, AL, USA) using FuGENE 6 (Roche Diagnostics, Indianapolis, IN, USA). The source of shRNA has accession # NM_031173. The following five siRNA sequences were used:

• Seq. 69048: CCGGCCAGTGGTAATGAAATGACTACTCGAGTAGTCATTTCATTACCACTGGTTTTTG

• Seq. 69049: CCGGCCCAGCAAACACATCATCATTCTCGAGAATGATGATGTGTTTGCTGGGTTTTTG

• Seq. 69050: CCGGCGAGGGAAGTCTCAATCCAAACTCGAGTTTGGATTGAGACTTCCCTCGTTTTTG

• Seq. 69051: CCGGCCTCGGATACAACATCCAACACTCGAGTGTTGGATGTTGTATCCGAGGTTTTTG

• Seq. 69052: CCGGGCTCAGGAGAAATCTCAGCTTCTCGAGAAGCTGAGATTTCTCCTGAGCTTTTTG

Charge movement recordings

Enzymatically dissociated flexor digitorum brevis (FDB) fibers were transferred to a small, flow-through Lucite chamber positioned on a microscope stage. Fibers were continuously perfused with the external solution using a push-pull syringe pump (WPI). Only fibers exhibiting a clean surface and no contracture were used for electrophysiological recordings. Muscle fibers were voltage-clamped using an Axopatch-200B amplifier (Molecular Devices, Sunnyvale, CA, USA) in the whole-cell configuration of the patch-clamp technique (Hamill et al. 1981; Wang et al. 1999). Patch pipettes were pulled from borosilicate glass (Boralex, WPI, Sarasota, FL, USA) using a Flaming Brown micropipette puller (P97, Sutter Instrument Co., Novato, CA, USA) and then fire-polished to obtain electrode resistances ranging from 450 to 650 kΩ. In the cell-attached configuration, the seal resistance was in the range of 1–4.5 GΩ, and in the whole-cell configuration, values ranged between 75 and 120 MΩ (Wang et al. 1999). The pipette was filled with the following solution: 140 mM Cs-aspartate, 5 mM Mg-aspartate₂, 20 mM Cs₂EGTA (ethylene glycol-bis(aminoethyl ether)-N,N,N’N’-tetraacetic acid), and 10 mM HEPES (N-[2-hydroxyethyl]piperazine-N’-[2-ethanesulfonic acid]), and pH was adjusted to 7.4 with CsOH (Adams et al., 1990; Wang et al., 1999). The external solution contained: 150 mM TEA(tetraethylammonium hydroxide)-CH₃SO₃, 2 mM MgCl₂, 2 mM CaCl₂, 10 mM Na-HEPES, 0.05 mM BTS(N-benzyl-p-toluenesulfonamide), and 0.001 mM tetrodotoxin (Delbono 1992; Delbono et al. 1997). Solution pH was adjusted to 7.4 with CsOH. All the experiments were conducted at room temperature (21–22°C).

Confocal microscopy

FDB fibers fluorescence was analyzed using a Radiance 2100 confocal microscope (Bio-Rad/Zeiss, Thornwood, NY, USA). YFP fluorescence was detected at 488nm. Confocal microscope fluorescence acquisition parameters were maintained constant across recordings are described above.

Single Intact Muscle Fiber Contraction

At time of sacrifice, FDB muscles were carefully dissected and pinned into a Petri dish lined with Sylgard (Dow Corning, Auburn, MI, USA) in a Ca²⁺-containing physiological solution (see below). All contraction experiments were carried out at room temperature (21–22°C). Single intact fiber dissection followed procedures previously published (Lannergren & Westerblad 1987; Gonzalez et al. 2000). Following dissection, tendons of single intact fibers were placed in custom-made micro-clips, and these clips were connected to a force transducer and a micropositioner for length control. Fibers were adjusted to optimum length (L_O) by using single twitches, elicited by 0.5-ms square wave pulses at 10 V. Once at L_O, fibers were stimulated with 350-ms trains of pulses, using frequencies varying from 50 to 100 Hz. The stimulation frequency that elicited maximum force was used for the remainder of the experiment (Payne et al. 2004).

Statistical analysis

All data are presented as means ± SE. Data were analyzed with Student t-test or one-way repeated measures ANOVA, with Tukey’s multiple comparisons test applied post hoc when appropriate. An alpha value of P < 0.05 was considered significant.