Sarcolipin deletion exacerbates soleus muscle atrophy and weakness in phospholamban overexpressing mice

Sarcolipin (SLN) and phospholamban (PLN) are two small proteins that regulate the sarco(endo)plasmic reticulum Ca2+-ATPase pumps. In a recent study, we discovered that Pln overexpression (PlnOE) in slow-twitch type I skeletal muscle fibers drastically impaired SERCA function and caused a centronuclear myopathy-like phenotype, severe muscle atrophy and weakness, and an 8 to 9-fold upregulation of SLN protein in the soleus muscles. Here, we sought to determine the physiological role of SLN upregulation, and based on its role as a SERCA inhibitor, we hypothesized that it would represent a maladaptive response that contributes to the SERCA dysfunction and the overall myopathy observed in the PlnOE mice. To this end, we crossed Sln-null (SlnKO) mice with PlnOE mice to generate a PlnOE/SlnKO mouse colony and assessed SERCA function, CNM pathology, in vitro contractility, muscle mass, calcineurin signaling, daily activity and food intake, and proteolytic enzyme activity. Our results indicate that genetic deletion of Sln did not improve SERCA function nor rescue the CNM phenotype, but did result in exacerbated muscle atrophy and weakness, due to a failure to induce type II fiber compensatory hypertrophy and a reduction in total myofiber count. Mechanistically, our findings suggest that impaired calcineurin activation and resultant decreased expression of stabilin-2, and/or impaired autophagic signaling could be involved. Future studies should examine these possibilities. In conclusion, our study demonstrates the importance of SLN upregulation in combating muscle myopathy in the PlnOE mice, and since SLN is upregulated across several myopathies, our findings may reveal SLN as a novel and universal therapeutic target.


Mice
The Pln OE and Sln KO mice have been described previously [3,21]. To direct type I fiber specific Pln overexpression, the Pln transgene was attached to the β-myosin heavy chain (MHC) promoter [22], which directs high levels of type I skeletal-muscle specific transgene expression [23][24][25]. FVB/N Pln OE mice were resuscitated from cryopreserved embryos by the mmRRC (000067-MU) to generate a breeding colony with WT FVB/N mice in our facility. Since the Sln KO mice were generated onto a C57BL/6J background [21], we backcrossed heterozygous male FVB/N Pln OE animals with either C57BL/6J Sln KO or WT littermate females for 6 generations to generate Pln OE /Sln KO , Pln OE , and WT control lines on a C57BL/6J background. All Pln overexpressing, with or without Sln, were heterozygous for Pln. With this breeding strategy, WT and Pln OE littermate mice were cousins to Pln OE /Sln KO mice. All animals used in the study were adult mice (WT, 6.2 ± 0.1 months; Pln OE , 6.1 ± 0.7 months; Pln OE /Sln KO , 6.1 ± 0.6 months). Animals were housed in an environmentally controlled room with a standard 12:12hour light-dark cycle and allowed access to food and water ad libitum. All animal procedures were reviewed and approved by the Animal Care Committee of the University of Waterloo and are consistent with the guidelines established by the Canadian Council on Animal Care.
In vitro muscle contractility WT, Pln OE , and Pln OE /Sln KO mice were sacrificed by cervical dislocation, and the intact soleus muscles were removed and placed into a bath with oxygenated Tyrode solution (95% O 2 , 5% CO 2 ) containing 121 mM NaCl 2 , 5 mM KCl, 24 mM NaHCO 3 , 1.8 mM CaCl 2 , 0.4 mM NaH 2 PO 4 , 5.5 mM glucose, 0.1 mM EDTA, and 0.5 mM MgCl 2 , pH 7.3, that was maintained at 25˚C. Muscles were situated between two platinum electrodes and force was electrically evoked and assessed across a range of stimulation frequencies from 1 to 100 Hz using a biphasic stimulator (Model 710B, Aurora Scientific, Inc., ON, Canada). Data were analyzed using Dynamic Muscle Control Data Acquisition software (Aurora Scientific). Specifically, peak isometric force amplitude (mN) and the maximal rates of force development (+dF/dt) and relaxation (−dF/dt) were determined during a twitch and across the range of stimulation frequencies. Peak isometric force was then normalized to muscle weight (mN/g).

Histological, histochemical and immunofluorescent staining
Soleus samples embedded in O.C.T. compound were cut into 10 μm thick serial cross sections with a cryostat (Thermo Electronic, MA, USA) maintained at -20˚C. Histological staining included H&E and Van Gieson, and histochemical staining included succinate dehydrogenase (SDH) activity. Images were acquired with a brightfield Nikon microscope linked to a PixeLink digital camera and quantified with ImageJ software (NIH, MA, USA). Immunofluorescence analysis of MHC expression was previously described [30] and performed with primary antibodies against MHCI, MHCIIa, and MHCIIb. Fibers that were not positively stained with MHCI, IIa, or IIb antibodies were considered as type IIX fibers. Slides were visualized with an Axio Observer Z1 fluorescent microscope equipped with standard red, green, blue filters, an AxioCam HRm camera, and AxioVision software (Carl Ziess, Oberkochen, Germany). Quantification of fiber distribution and cross-sectional area (CSA) was performed using imageJ software.

Electron microscopy
To examine the triad structures in the soleus muscles of WT, Pln OE , and Pln OE /Sln KO mice, we performed transmission electron microscopy (TEM). Briefly, muscle samples (1 cubic mm each) were fixed immediately in 2.5% glutaraldehyde (EM grade) in 0.2M phosphate buffer (pH 7.4), overnight at 4˚C. Subsequently, samples were postfixed in 1% osmium tetraoxide in the same buffer, dehydrated in a graded acetone series, and embedded in an epon-aldarite resin. Semithin sections (0.74 μm) were stained with toluidine blue and placed on a hot plate for 30 seconds to determine orientation of samples. Ultrathin longitudinal sections (90 nm) were cut on a Leica Reichart Ultracut E microtome (Leica Microsystem, Wetzlar, Germany), double contrasted with uranyl acetate (50% in methanol) and lead citrate (0.4% w/v in boiled H 2 O), and viewed in a CM 10 Philips TEM.

Proteolytic enzyme activity assays
Caspase-3 and cathepsin-B/L activity were determined in soleus muscle homogenates using the substrates Ac-DEVD-AMC (Alexis Biochemicals) and z-FR-AFC (Enzo Life Sciences, NY, USA), respectively, whereas calpain and 20S proteasome activity were determined using Suc-LLVY-AMC (Enzo-Life Sciences) [3]. These fluorogenic substrates are weakly fluorescent but yield highly fluorescent products following proteolytic cleavage by their respective proteases. Fluorescence was measured using a SPECTRAmax Gemini XS microplate spectrofluorometer (Molecular Devices, Sunnyvale, CA) with excitation and emission wavelengths of 360 nm and 440 nm (caspase-3), 380 nm and 460 nm (calpain and 20S proteasome), or 400 nm and 505 nm (cathepsin), respectively. To distinguish between calpain and 20S proteasome activity, the specific inhibitors Z-Leu-Leu-CHO (calpain; Enzo-Life Sciences) and epoxomicin (20S proteasome; Cayman Chemical, MI, USA) were incubated and the difference in fluorescence from homogenates incubated with and without the respective inhibitors was measured. All proteolytic activities were normalized to total protein content and expressed as fluorescence intensity in arbitrary units per mg of protein.

Daily activity and food intake measurements
Daily food consumption (g/day) and cage activity were measured using a comprehensive laboratory animal monitoring system (CLAMS; Oxymax series; Columbus Instruments, Columbus, OH, USA) as previously described [31]. This system is equipped with a feed scale for monitoring mass of food consumed, as well as X-and Z-axes infrared photocell detectors that allow for monitoring ambulatory activity (when 2 adjacent x-axis beams were broken in succession) and total cage activity. Mice were placed in the CLAMS for a 3-day period for 3 separate trials and had free access to food and water.

Statistics
All values are presented as means ± standard error (SEM). Statistical significance was set to p 0.05. Most comparisons between WT, Pln OE , and Pln OE /Sln KO mice were performed using a one-way ANOVA with a Tukey's post-hoc test or a Games-Howell post-hoc test to control for unequal variances. A two-way ANOVA was used to examine the force frequency curves.

Effect of Sln ablation on SERCA function
Similar to the FVB/N line [3] we observed an 8-fold upregulation of SLN protein in the Pln OE soleus, and we confirmed the lack of SLN protein after genetic deletion of Sln ( Fig 1A). Removal of Sln did not alter the level of monomeric or pentameric PLN overexpression ( Fig   Fig 1. Sln deletion does not rescue SERCA function in Pln OE mice. (A) Western blot analyses showing successful removal of the 8-fold upregulation in SLN protein (n = 6 per genotype). Ponceau staining ensured equal loading. (B) Pln OE and Pln OE /Sln KO soleus muscles displayed similar levels of monomeric (m) and pentameric (p) PLN overexpression compared with WT (n = 6 per genotype). Values were normalized to actin. (C) Rates of Ca 2+ uptake were similarly reduced in the Pln OE and Pln OE /Sln KO soleus muscles compared with WT (n = 6-7, per genotype). (D) Representative SERCA activity-pCa curves measured in soleus homogenates (n = 6-7, per genotype). (E) SERCA's apparent affinity for Ca 2+ , presented as pCa 50 , was similarly reduced in the Pln OE and Pln OE /Sln KO soleus muscles compared with WT (n = 6-7, per genotype). pCa 50 is the negative logarithm of the Ca 2 + concentration required to attain half-maximal SERCA activity rate. *Significantly different from WT using a Student's t-test (A), and a one-way ANOVA with a Tukey's post-hoc (B, C and E), p 0.05. All values are means ± SEM. 1B). Unexpectedly, the rates of Ca 2+ uptake, and SERCA's apparent affinity for Ca 2+ were similarly reduced in the Pln OE and Pln OE /Sln KO soleus muscles compared with WT (Fig 1C-1E).

PLN phosphorylation and expression of other Ca 2+ regulatory proteins
As with total PLN expression, only 2.5 μg of protein from soleus homogenates was required to detect phosphorylated PLN (p-PLN) in the Pln OE and Pln OE /Sln KO mice compared to the 25 μg of protein required from WT mice (Fig 2A). The p-PLN/PLN ratio was significantly lower in both Pln OE and Pln OE /Sln KO mice compared with WT, but there were no differences between Pln OE and Pln OE /Sln KO mice (Fig 2A). Assessment of the other major Ca 2+ regulatory proteins including SERCA1a, SERCA2a, DHPR, RyR, and CSQ did not reveal any significant differences (Fig 2B-2F).

CNM phenotype in Pln OE and Pln OE /Sln KO mice
To determine whether genetic deletion of SLN would influence the CNM phenotype, we performed: 1) H&E staining to examine the percentage of central nuclei, 2) SDH staining to look for central aggregation of oxidative activity, and 3) immunofluorescent staining to assess type I fiber distribution and size. Both the Pln OE and Pln OE /Sln KO soleus muscles exhibited a CNMlike phenotype with an increase in central nuclei, central aggregation of oxidative activity, and type I fiber predominance and hypotrophy (Fig 3A-3F). Consistent with other models of CNM and biopsies from CNM patients [32][33][34][35], TEM analyses revealed triad abnormalities with swollen sarcoplasmic reticulum membranes in both the Pln OE and Pln OE /Sln KO soleus muscles (Fig 3G). In addition to this and in agreement with our initial characterization of the Pln OE mice [3], endomysial fibrosis and core-like lesions were evident in the Pln OE and Pln OE / Sln KO soleus (Fig 3H-3J). Collectively, these data suggest that Sln deletion had very little effect on the overall CNM phenotype that occurs with PLN overexpression. Effect of Sln deletion on muscle size and contractile function The absence of SLN exacerbated the soleus muscle atrophy seen with PLN overexpression [3,22] as the soleus:body weight ratios were smallest in the Pln OE /Sln KO mice compared with both Pln OE and WT (Fig 4A). Furthermore, the type IIA fibers from Pln OE /Sln KO mice failed to undergo the typical compensatory hypertrophy observed in the Pln OE mice (Fig 3F and reference [3]), and the type IIX fibers showed a similar pattern (Fig 3F); however the one-way ANOVA did not indicate statistical significance (p = 0.11). There was also a significant reduction in total fiber count in the Pln OE /Sln KO soleus muscles compared with both Pln OE and WT mice (Fig 4B). Soleus muscles from Pln OE /Sln KO mice displayed significant reductions in mass-specific force from 50-100Hz stimulation compared with WT ( Fig 4C). In contrast, the Pln OE soleus muscles were only significantly weaker than WT at 100 Hz (Fig 4C). Furthermore, comparisons between Pln OE and Pln OE /Sln KO soleus force-frequency curves using a two-way ANOVA reveals a significant main effect of genotype (p = 0.02), which suggests that the Pln OE /Sln KO soleus muscles are weaker than the Pln OE soleus muscles. With respect to twitch kinetics, the rates of relaxation ( Fig 4D) and force development ( Fig 4E) were reduced in the Pln OE and Pln OE /Sln KO soleus muscles compared with WT, with the rates of force development being significantly slower in Pln OE /Sln KO compared with Pln OE . Importantly, body weight, daily food intake, and daily activity were not different between groups (Table 1), and therefore cannot explain the worsened soleus muscle atrophy and weakness.

Fiber type distribution
In response to Pln overexpression in the slow-twitch type I fibers, the type II fibers decrease in proportion while the type I fibers increase [3,22]. A similar result was observed in this study with significant reductions in both type IIA and type IIX fiber percentages in Pln OE mice compared with WT ( Fig 3E). In contrast, the type IIA fiber distribution in the Pln OE /Sln KO was not different from WT (Fig 3E) suggestive of a delayed fast-to-slow fiber type shift.

NFATc1 phosphorylation and stabilin-2 expression as markers of calcineurin signaling
To explain the lack of type II fiber hypertrophy and the delay in the fast-to-slow fiber type shift in the absence of Sln, we focused on calcineurin, the Ca 2+ -calmodulin dependent serine/threonine phosphatase known to regulate these cellular adaptations [36][37][38] and, perhaps more importantly, shown to be responsive to SLN [39]. When examining the phosphorylation status of nuclear factor of activated T-cells cytoplasmic 1 (NFATc1) as a marker of calcineurin signaling, compared with WT we found that NFATc1 phosphorylation relative to total NFATc1 was significantly lower in Pln OE but not Pln OE /Sln KO soleus (Fig 5A). Calcineurin-mediated activation of NFATc1 increases the expression of stabilin-2; a phosphatidylserine receptor required for myoblast fusion [40]. Corresponding well with the significant dephosphorylation of NFATc1 only in the Pln OE mice, we found a significant elevation in stabilin-2 expression only in Pln OE soleus compared with WT ( Fig 5B). Stabilin-2 expression was also higher in the Pln OE soleus compared to the Pln OE /Sln KO soleus.

Assessment of the major proteolytic pathways and autophagy
Next, we tested whether elevated proteolytic activities could also contribute to the exacerbated soleus muscle atrophy seen in the Pln OE /Sln KO mice. Consistent with our previous study [3], calpain, caspase-3, 20S proteasome and cathepsin activities were all significantly elevated in the Pln OE soleus compared with WT; however, none of these were augmented further in the Pln OE /Sln KO soleus (Fig 6A-6D). In fact, only the 20S proteasome was significantly elevated in the Pln OE /Sln KO soleus compared with WT. Moreover, a trending reduction in p62 content (p = 0.10) and a significant increase in the LC3-II/I ratio in the Pln OE soleus, combined with Values are means ± SEM (n = 18-23, for the average body weight of Pln WT /Sln WT , Pln OE /Sln WT , and Pln OE /Sln KO mice; n = 6 per genotype for daily food consumption, ambulation, and total activity). Ambulatory activity counts were when the mouse crossed 2 adjacent x-axis infrared beams in succession. One-way ANOVA revealed no significant effect of genotype for body weight (p = 0.37), food intake (p = 0.48), total activity (p = 0.12), or dual beam activity (p = 0.17). doi:10.1371/journal.pone.0173708.t001 The Effects of Sln deletion in Pln OE mice PLOS ONE | DOI:10.1371/journal.pone.0173708 March 9, 2017 the elevated cathepsin activity, indicate greater autophagic signaling compared with WT ( Fig  6E and 6F); however, we did not observe this in the Pln OE /Sln KO mice. Taken together, these results suggest that proteolytic activity and autophagic signaling were not enhanced in Pln OE / Sln KO soleus and cannot explain the exacerbated muscle atrophy caused by Sln ablation.

Discussion
In this study, we examined the physiological role of SLN upregulation in the Pln OE mouse model of CNM by generating the Pln OE /Sln KO mouse line. We hypothesized that SLN ablation would alleviate the CNM pathology, muscle atrophy, and weakness in this model through improvements in SERCA function. However, in contrast to our hypotheses, we did not observe any improvements in SERCA function or CNM phenotype in the Pln OE /Sln KO soleus, and found that Sln deletion actually exacerbated the soleus muscle atrophy and weakness by preventing the compensatory type II fiber hypertrophy and reducing the total myofiber count. In addition, rates of force development were slower in the Pln OE /Sln KO soleus compared with those from Pln OE mice, which could be due to excitation-contraction (E-C) coupling impairments related to triad structural defects.
In a previous study with Sln overexpressing mice, SLN was shown to stimulate calcineurin [39], a Ca 2+ -dependent phosphatase that is implicated in muscle growth [36,38,41] and promotion of the slow-oxidative phenotype [36,[41][42][43][44][45]. Pharmacological inhibition of calcineurin prevents myofiber hypertrophy and the fast-to-slow fiber type transition in plantaris muscles that have been mechanically overloaded after synergist ablation of the soleus and gastrocnemius [36,41,46]. We propose that the Pln OE soleus represents a model of intramuscular overload, whereby the type I fiber hypotrophy results in overload of the type II fibers within the soleus causing them to hypertrophy and transition towards the slow oxidative phenotype [3]. However, in the absence of SLN upregulation, the fast-to-slow fiber type shift and type II fiber hypertrophy are attenuated in the Pln OE mice.
A failure to dephosphorylate NFAT and increase expression of stabilin-2 are indicative of impaired calcineurin signaling in the Pln OE /Sln KO mice. It is well accepted that calcineurin activation is important for controlling muscle fibre type in adaptive muscle remodeling so lowered calcineurin activation can at least partly explain the blunted fast-to-slow fiber type shift in The Effects of Sln deletion in Pln OE mice the Pln OE /Sln KO soleus. On the other hand, there is some discrepancy regarding calcineurin's role in stimulating muscle growth. Specifically, some studies have failed to replicate the hypertrophic effect of calcineurin in the overloaded plantaris [47,48], and genetically modified mice, characterized with either calcineurin inhibition or over-activation, display no effects on myofiber size or muscle mass [45,[49][50][51]. Nonetheless, calcineurin's role in myoblast fusion is well established [52][53][54] and recent evidence has shown that calcineurin activation, specifically through NFATc1, increases the expression of a phosphatidylserine receptor, stabilin-2, and that without stabilin-2, myoblast fusion, myofiber size, and muscle mass are severely reduced [40]. According to the myonuclear domain theory, an increase in the relative amount of nuclei per fibre, via enhanced myoblast fusion to pre-existing myofibers, will increase myofiber size and overall muscle mass [54,55]. Collectively, our results suggest that the blunted calcineurin signalling in the Pln OE /Sln KO mice reduces stabilin-2 expression and likely impairs myoblast fusion, thereby contributing to the reductions in soleus size, type II fiber CSA, and force generation.
Impairments in stabilin-2 and myoblast fusion could also contribute to the reduction in total fiber count we observed in the Pln OE /Sln KO mice, possibly via impaired muscle regeneration. While calcineurin's role in muscle regeneration has been well established [56][57][58], it is important to note that CNMs, including the Pln OE mouse, generally do not display signs of extensive muscle regeneration [3,59]. Alternatively, although excessive autophagy could be detrimental and may lead to muscle atrophy [60], impaired autophagy also results in extensive myofiber degeneration [61]. Therefore, the impaired activation of autophagic signaling in the Pln OE /Sln KO soleus could also contribute to the reductions in total fiber count. Collectively, these results indicate that failure to promote both calcineurin and autophagic signaling in the Pln OE /Sln KO soleus, unlike the Pln OE soleus, likely contribute to the hypoplasia and the exacerbated muscle atrophy and weakness in this model. Upregulated SLN is a hallmark of several muscle diseases [5][6][7][8][9][10][11][12] but it's unknown whether that response is adaptive or pathogenic. This is the first study to examine the role of SLN upregulation in any form of muscle myopathy and our results indicate that SLN upregulation is an adaptive response that may be required for optimal activation of calcineurin to combat the muscle atrophy and weakness that occurs in the Pln OE soleus. In the dystrophic mdx mouse model, the affected skeletal muscles show large upregulation of SLN [7], and given the data presented here, this increase in SLN could potentially aid in activating calcineurin and autophagic signaling, both of which have been shown to be promising therapeutic strategies [57,[62][63][64][65][66]. In addition, corticosteroid [63] and high-fat feeding [67] can ameliorate dystrophic pathology in mdx mice, and both treatments lead to an increase in SLN expression [9,31,68]. Future studies aimed at examining the role of SLN across these different myopathies could further reveal the importance of SLN in overall muscle health.
An unresolved issue is why Sln deletion did not result in increased Ca 2+ uptake and SERCA activity in soleus homogenates, given that Sln deletion alone, without Pln overexpression, increased the apparent affinity of SERCA for Ca 2+ by 0.1 pCa units in soleus homogenates compared with WT [20], an effect that we reproduced here (data not shown). Further analysis of PLN's phosphorylation status or the expression of other major Ca 2+ regulatory proteins such as SERCA1a, SERCA2a, RyR, DHPR, and CSQ do not provide a likely explanation. Sln deletion in Pln OE mice primarily affected the type II fibers (ie. attenuated type II fiber hypertrophy and fast-to-slow fiber type shift). Therefore, it is possible that SLN upregulation in soleus from Pln OE mice is occurring primarily in the overloaded type II fibers that do not overexpress PLN. In support of this view, previous findings from our laboratory show that SLN is predominantly expressed in type II fibers from human vastus lateralis [28]. If SLN upregulation occurs primarily in the type II fibers, there would be less potential for superinhibition to occur, and any improvement in SERCA function expected in the type II fibers after Sln deletion could be masked by the level of PLN overexpression that occurs in the type I fibers, which comprise more than 55% of the muscle fibers. Thus, our study is limited in measuring SERCA function in whole homogenates, but if it was possible to measure SERCA activity specifically in type II fibers then we would expect increased activity in Pln OE /Sln KO compared with Pln OE , which would correspond with the blunted calcineurin activity we observed in Pln OE /Sln KO soleus.
Our work also sheds light on the differences between PLN and SLN function in skeletal muscle. Our findings showing that PLN overexpression causes myopathy and muscle atrophy whereas SLN upregulation counters it, adds to the notion that PLN and SLN may not be functionally redundant [69]. Indeed, unlike the muscle weakness observed in Pln OE mice [3,22], Sln OE mice present with enhanced contractility [70]. It also appears that PLN is different from SLN in its ability to regulate calcineurin signaling since PLN overexpression alone without SLN upregulation resulted in greatly reduced calcineurin signaling. Given that both PLN and SLN are capable of inhibiting SERCA [13,71,72], it is surprising that only SLN seems to be a significant regulator of calcineurin. Similarly, it appears that SLN but not PLN is capable of altering autophagy in skeletal muscle since PLN overexpression alone without SLN upregulation did not increase autophagic signaling. We speculate that SLN's influence on energy expenditure [73] may have a role in activating autophagy, whereas PLN may not have the same influence on ATP consumption in muscle [74], but this needs to be explored further.
In summary, we have found that SLN is upregulated in the soleus muscles from PLN overexpressing mice to combat muscle atrophy and weakness by promoting the compensatory type II fiber hypertrophy and maintaining total myofiber number through calcineurin activation and autophagic signaling. We did not find any alterations in the CNM phenotype in the Pln OE /Sln KO soleus, which suggests that SLN likely has no contribution to CNM pathology. Increased SLN expression is a hallmark of muscle disease, which may represent an adaptive response that is necessary for muscle growth and remodeling and future studies should determine whether SLN provides a novel therapeutic target for other neuromuscular disorders.
Supporting information S1 Dataset. Minimal dataset for the study. (XLSX)