ANGIOTENSIN-DEPENDENT AUTONOMIC DYSREGULATION PRECEDES DILATED CARDIOMYOPATHY IN A MOUSE MODEL OF MUSCULAR DYSTROPHY

Abstract

Sarcoglycan mutations cause muscular dystrophy. Patients with muscular dystrophy develop autonomic dysregulation and dilated cardiomyopathy (DCM), but the temporal relationship and mechanism of autonomic dysregulation are not well understood. We hypothesized that activation of the renin-angiotensin system (RAS) causes autonomic dysregulation prior to development of DCM in sarcoglycan-delta (Sgcd) deficient mice, and that the severity of autonomic dysfunction at a young age predicts the severity of DCM at older ages. At 10-12 weeks of age, when left ventricular function assessed by echocardiography remained normal, Sgcd−/− mice exhibited decreases in arterial pressure, locomotor activity, baroreflex sensitivity (BRS) and cardiovagal tone, and increased sympathetic tone compared with age-matched C57BL/6 control mice (P<0.05). Systemic and skeletal muscle RAS were activated, and angiotensin II type 1 receptor (AT₁R) expression, superoxide and fibrosis were increased in dystrophic skeletal muscle (P<0.05). Treatment with the AT₁R blocker losartan for 7-9 weeks beginning at 3 weeks of age prevented or strongly attenuated the abnormalities in Sgcd−/− mice (P<0.05). Repeated assessment of phenotypes between 10 and 75 weeks of age demonstrated worsening of autonomic function, progressive cardiac dysfunction and DCM, and increased mortality in Sgcd−/− mice. High sympathetic tone predicted subsequent left ventricular dysfunction. We conclude that RAS activation causes severe autonomic dysregulation in young Sgcd−/− mice, which portends a worse long-term prognosis. Therapeutic targeting of RAS at a young age may improve autonomic function and slow disease progression in muscular dystrophy.

INTRODUCTION

Muscular dystrophies are inherited myopathic disorders accompanied by progressive skeletal muscle weakness and wasting (Wallace & McNally, 2009). Mutations in genes encoding components of the dystrophin glycoprotein complex (DGC) located at the myocyte plasma membrane cause muscular dystrophy (Ervasti & Campbell, 1991; Wallace & McNally, 2009). Cardiac muscle and vascular smooth muscle are also often affected in resulting in dilated cardiomyopathy (DCM) and premature death due to heart failure or cardiac arrhythmias (Coral-Vazquez et al. 1999; Hack et al. 2000; Lanza et al. 2001; Politano et al. 2001; Jefferies et al. 2005; Spurney et al. 2011).

Autonomic dysregulation is common in muscular dystrophy, but its temporal relationship with DCM and underlying mechanisms are not well understood. As expected, the sympathetic nervous system and renin-angiotensin system (RAS) are activated in patients with left ventricular dysfunction, and treatment with angiotensin converting enzyme (ACE) inhibitors improves left ventricular function and survival (Lanza et al. 2001; Vita et al. 2001; Duboc et al. 2005; Jefferies et al. 2005). Evidence of autonomic dysregulation in muscular dystrophy at a young age comes primarily from studies demonstrating decreased heart rate variability (HRV) in patients with Duchenne muscular dystrophy (Yotsukura et al. 1995; Lanza et al. 2001; Inoue et al. 2009) and animal models (Chu et al. 2002; Hampton et al. 2012). The results point to a substantial decrease in cardiovagal tone, while evidence of increased cardiac sympathetic tone is less compelling, particularly prior to development of left ventricular dysfunction. In addition, disruption of nitric oxide synthase-1 in dystrophic muscle abrogates sympatholysis resulting in exaggerated sympathetic-mediated vasoconstriction in contracting muscle in patients (Sander et al. 2000; Martin et al. 2012) and mice (Thomas et al. 1998).

Interestingly, skeletal muscle RAS is activated in dystrophic myoblasts which promotes fibrosis via activation of transforming growth factor-beta 1 (Dietze & Henriksen, 2008; Sun et al. 2009; Cabello-Verrugio et al. 2012; Morales et al. 2012). Treatment with the Ang II type 1 receptor (AT₁R) blocker losartan (Cohn et al. 2007; Cabello-Verrugio et al. 2012) or the protective peptide Ang-(1-7) (Acuna et al. 2014) attenuates fibrosis and inflammation and improves muscle function in muscular dystrophy. While activation of skeletal muscle RAS clearly contributes to skeletal muscle pathology and dysfunction, its possible role in autonomic and cardiovascular dysregulation in muscular dystrophy has not been investigated. Furthermore, we are unaware of any studies that have measured circulating levels of renin, Ang II or other RAS peptides in patients or animal models of muscular dystrophy prior to development of left ventricular dysfunction.

We recently reported that AT₁R expression and oxidative stress are increased in dystrophic skeletal muscle in young Sgcd deficient mice (10-13 wks) (Sabharwal et al. 2014), a model of limb girdle muscular dystrophy 2F (Nigro et al. 1996; Coral-Vazquez et al. 1999; Hack et al. 2000; Politano et al. 2001). These mice also exhibit decreased baroreflex sensitivity, decreased cardiovagal tone, and increased sympathetic tone (Sabharwal et al. 2014). Furthermore, oral administration of Ang-(1-7) in hydroxypropyl β-cyclodextrin abrogated the abnormalities in skeletal muscle and autonomic regulation (Sabharwal et al. 2014). In the present study we tested the hypothesis that RAS activation occurs prior to development of cardiac dysfunction in Sgcd−/− mice, and causes autonomic dysregulation that predicts subsequent development of DCM at older ages. The results of our studies are summarized in a symposium report recently published in Experimental Physiology (Sabharwal & Chapleau, 2014).

METHODS

Animals and Ethical Approval

Male and female C57BL/6 (Jackson Laboratory, Bar Harbor, ME, USA) and Sgcd−/− (University of Iowa) mice were studied. As we did not find any significant differences in any of the measured variables between male and female mice in either genotype, the results obtained from males and females were combined. Generation of homozygous Sgcd−/− mice has been described previously (Coral-Vazquez et al. 1999). The mice were maintained in a 12:12 hr light-dark cycle (6:00 – 18:00 light), fed normal mouse chow, and had access to water ad libitum. All procedures were performed in accordance with American Physiological Society and institutional guidelines, and approved by the University of Iowa Animal Care and Use Committee. Kaplan-Meier survival curves were generated from birth-death records of mice in our C57BL/6 and Sgcd−/− colonies that had not undergone any experimental procedures.

Study Design

In group 1 experiments, the major objectives were to determine if skeletal muscle pathology and autonomic dysregulation in young, 10-12 week old Sgcd−/− mice is accompanied by cardiac dysfunction and/or activation of the RAS, and whether chronic administration of the AT₁R blocker losartan (Sigma) abrogates the abnormalities observed at this young age. Arterial blood pressure, heart rate (HR) and locomotor activity were measured in untreated control C57BL/6 (n = 5, 3 males and 2 females), untreated Sgcd−/− (n = 7, 4 males and 3 females), losartan-treated C57BL/6 (n = 6, 3 males and 3 females), and losartan-treated Sgcd−/− (n = 6, 3 males and 3 females) mice by radiotelemetry. Left ventricular systolic function was measured in the same mice by echocardiography. These and other methods used to assess autonomic function, RAS activation, and skeletal muscle pathology are described below under the corresponding headings.

At the end of experimental protocols, mice were euthanized, telemeters were removed, and skeletal muscle tissues were prepared for histology, Western blots and/or immunofluorescence analyses.

In group 2 experiments, we determined the time course of disease progression in Sgcd−/− mice, focusing on the temporal relationship between autonomic and cardiac dysfunction. Radiotelemeters were implanted in C57BL/6 (n = 6, 3 males, 3 females) and Sgcd−/− (n = 8, 4 males, 4 females). Cardiovascular, locomotor, autonomic, and left ventricular functions were measured every 4 weeks in C57BL/6 and Sgcd−/− mice between 10 and 75 weeks of age. At the end of protocols, mice were euthanized with sodium pentobarbital (1.5 mg/10 g of body weight, IP). In separate groups of mice, skeletal muscle (quadriceps, gastrocnemius) and heart were removed at various ages between 10 and 75 weeks (~6 wk intervals) and prepared for histological assessment. Heart, lung and body weight were also measured.

Cardiovascular/Autonomic Assessment by Radiotelemetry

A telemetry probe (PC10, DSI) was inserted into the thoracic aorta via the left common carotid artery in mice anaesthetized with ketamine (91 mcg/g, IP) and xylazine (9.1 mcg/g, IP) at 9-10 weeks of age for measurement of blood pressure as described previously (Lu et al. 2009; Sabharwal et al. 2010; Sabharwal et al. 2014). Mice were allowed to recover from the surgery for 7 days before data were collected. Diurnal variations in blood pressure, HR and locomotor activity were measured at a sampling rate of 500 Hz from data collected for 10 second periods every 5 minutes using the Dataquest ART Acquisition software (DSI).

Blood pressure was also continuously recorded at 2000 Hz for 1 hour between 10:00 – 13:00 on four separate days to enable collection of beat-to-beat data for assessment of baroreflex sensitivity (BRS), cardiac vagal and sympathetic tone, and vasomotor sympathetic tone (Laude et al. 2008; Lu et al. 2009; Sabharwal et al. 2010; Sabharwal et al. 2014). Cardiovagal and cardiac sympathetic tone were estimated from the HR responses to methylatropine (1 mcg/g, IP, Sigma) and propranolol (1 mcg/g, IP, Sigma), respectively. Sympathetic vasomotor tone was estimated from the decrease in mean arterial pressure in response to the ganglionic blocker chlorisondamine (12 mcg/g, IP, Tocris, Minneapolis, MN, USA). Spontaneous BRS was measured using the sequence technique. Measurements of spontaneous locomotor activity were derived from changes in transmitter signal strength associated with movement of the mouse.

Echocardiography

Echocardiograms were obtained in mice implanted with blood pressure telemeters using a 30 MHz (VisualSonics, Toronto, Canada) linear-array probe (Coral-Vazquez et al. 1999; Hill et al. 2000). Mice were mildly sedated with midazolam (0.15 mg, SC). Both two-dimensional parasternal short axis and long axis frames were captured at a rate >200/s. Images were archived off-line and analyzed using companion software. Left ventricular mass, end systolic volume (ESV), end diastolic volume (EDV) and ejection fraction (EF) were calculated as validated previously (Hill et al. 2000).

Measurement of RAS Peptides

RAS peptides were measured in blood plasma and skeletal muscle collected from young (10-12 wks), age-matched C57BL/6 control (n= 4) and Sgcd−/− (n=4) male mice. The mice were anaesthetized with 5% isoflurane (closed chamber) followed by thoracotomy, and blood (0.5 ml) and both quadriceps muscles were collected. Angiotensin peptides were stabilized in blood by immediate addition of a protease inhibitor cocktail (Attoquant Diagnostics GmbH, Vienna, Austria) and plasma was separated by cooled centrifugation for 10 min at 3000 × g (Poglitsch et al. 2012). The samples were snap frozen in liquid nitrogen immediately after collection and stored at −80 °C for shipment to Attoquant Diagnostics GmbH (Vienna, Austria) for analysis.

RAS peptides were extracted from samples by mechanical homogenization and detergent-based extraction (Attoquant Diagnostics GmbH, Vienna, Austria; Poglitsch et al. 2012). Frozen tissue slices (<50 mg) were transferred to a t-Prep vial containing 100 microliters of ice-cold 4 M aqueous guanidine thiocyanate (Sigma) supplemented with 1 % (v/v) trifluoroacetic acid (Sigma), smashed once by the t-Prep device and homogenized by subjecting the sample to 5 sonication cycles of 10 seconds high energy (Power: 500 Watts, Duty Factor 50 %) followed by 20 seconds of low energy sonication (Power: 80 Watts, Duty Factor 10 %) using the focused ultrasonicator S220X (Covaris®; Woburn, MA, USA). The tissue concentration in the extraction buffer was adjusted to 100 mg/ml by addition of buffer.

Concentrations of RAS peptides in plasma and tissue samples were measured by liquid chromatography-mass spectrometry (Poglitsch et al. 2012). Stable-isotope-labeled internal standards for individual angiotensin metabolites were added to the samples previously thawed on ice. Following C18 based solid-phase-extraction, samples were subjected to the analysis using a reversed-phase analytical column (Acquity UPLC® C18; Waters, Milford, MA, USA) operating in line with a XEVO TQ-S triple quadrupole mass spectrometer (Waters) in MRM mode. Two different mass transitions were measured per peptide and angiotensin concentrations were calculated by relating endogenous peptide signals to internal standard signals, provided that the integrated signals exceeded a signal-to-noise ratio of 10, as described previously (Poglitsch et al. 2012). The minimum quantification limits for individual peptides in plasma were 4 pg/ml (Ang 1-10), 1 pg/ml (Ang 1-8), 2 pg/ml (Ang 1-7), 1 pg/ml (Ang 1-5), 3 pg/ml (Ang 2-8), 1 pg/ml (Ang 3-8), 5 pg/ml (Ang 2-10), 4 pg/ml (Ang 2-7), 5 pg/ml (Ang 1-9), and 1 pg/ml (Ang 3-7).

Blockade of Ang II AT₁ Receptors

In subsets of mice, losartan (Sigma) was administered to Sgcd−/− and control mice in drinking water for 7-9 weeks at a concentration of 0.1 mg/ml, beginning at 3 weeks of age. Based on measurements of water intake, the estimated delivered dose of losartan was ~10 mg/kg/day. Blood pressure, heart rate, locomotor activity, left ventricular function and autonomic indices were measured in both losartan-treated and untreated mice at 10-12 weeks of age.

Preparation of Tissues for Histology, Immunofluorescence and Western Blot

C57BL/6 and Sgcd−/− mice were anaesthetized with sodium pentobarbital (1.5 mg/10 g of body weight, IP) and perfused with 15 ml of PBS and zinc formalin fixative solution. Skeletal muscle (quadriceps and gastrocnemius) and hearts were embedded in paraffin, sectioned (7 μm) and stained with Haematoxylin and Eosin or Masson’s trichrome to delineate dystrophic features (centralized nuclei and fibrosis). The area of fibrosis was calculated in tissue slices obtained from separate groups of C57BL/6 (n = 4-5, 2-3 males, 2 females) and Sgcd−/− (n = 4-7, 2-4 males, 2-3 females) mice over a range of ages (10-75 wks at ~6 wk intervals).

Separate groups of young mice at 10 weeks of age were anaesthetized with sodium pentobarbital (1.5 mg/10 g of body weight, IP) to prepare frozen sections of quadriceps and gastrocnemius muscle for detection of AT₁R via immunostaining and for measurement of superoxide, a marker of oxidative stress. The tissues collected from untreated and losartan-treated C57BL/6 and Sgcd−/− mice were embedded with OCT compound and rapidly frozen in isopropanol-chilled liquid nitrogen. They were stored at −80°C until 7 μm sections were made using a cryostat. Skeletal muscle from untreated C57BL/6 (n = 6, 3 males, 3 females) and Sgcd−/− (n = 6, 3 males, 3 females) mice were directly snap frozen (without OCT) for Western blot analysis.

Measurement of AT₁R Protein Expression

AT₁R protein expression was measured in quadriceps muscles by Western blot and immunofluorescence. For Western blot, tissue was ground, homogenized, and centrifuged, and membranes blocked and incubated overnight at 4°C with an anti-AT₁R antibody (Santa Cruz Biotechnology, Dallas, TX, USA; catalogue no. sc-1173) as described previously (Sabharwal et al. 2014). Bands were visualized with an ECL reagent (Amersham Biosciences, Piscataway, NJ, USA) and films were scanned and analyzed using ImageJ software (NIH, Bethesda, MD, USA). AT₁R protein was normalized to total protein measured by Bradford assay (Bio-Rad Laboratories, Hercules, CA, USA).

For immunofluorescence, 7 μm tissue samples were stained with a rabbit polyclonal antibody (Santa Cruz Biotechnology, catalog no. sc-1173) and Alexa Fluor® 488 goat anti-rabbit IgG (Invitrogen, Grand Island, NY, USA; catalogue no. A-11070) as described previously (Sabharwal et al. 2014). DAPI (4’,6-diamidino-2-phenylindole; Invitrogen) was used to identify cell nuclei. Images were captured on a confocal microscope (Zeiss model 710). Immunofluorescent intensity was quantified using NIH Image J software. AT₁R expression in muscle from each animal was obtained by averaging the fluorescence intensity of AT₁R immunoreactivity in six representative transverse sections.

Oxidative Stress

Superoxide levels measured by dihydroethidium (DHE, Sigma) fluorescence were used as a marker of oxidative stress in skeletal muscle. DHE stock solution was prepared and applied on tissue sections for 15 min at 37°C, as described previously (Sabharwal et al. 2014). Images were captured on a confocal microscope (BioRad 1024) at excitation 488 nm and emission wavelength of 568 nm. An average of six 7-μm sections of skeletal muscle were obtained from each mouse. Intensity of fluorescence was quantified using NIH ImageJ software and normalized to a percentage of the average fluorescence intensity measured in untreated C57BL/6 mice at 10 weeks of age.

Measurement of Fibrosis in Skeletal Muscles and Heart

To quantify fibrosis, tissues embedded in paraffin were cut into serial sections. A random number chart determined the starting point and every 10th section was stained with freshly prepared Masson's Trichrome stain. Masson Trichrome stains nuclei black, cytoplasm in muscle fibers red, and collagen blue. The stained sections were imaged with an Olympus BX- 51 Light Microscope equipped with a DP-71 digital camera. The images were overlaid on a counting grid containing points evenly spaced according to the rules outlined previously (Gundersen & Jensen, 1985). The space between these course points was used to calculate area of fibrosis. This was done for each section stained. Six to eight sections from each mouse were analyzed, and the average percent fibrosis was calculated for each mouse.

For Group 1, fibrosis was calculated in skeletal muscle removed from untreated (n = 6, 3 males, 3 females) and losartan-treated (n = 6, 3 males, 3 females) C57BL/6 control mice and untreated (n = 8, 4 males, 4 females) and losartan-treated (n = 8, 4 males, 4 females) Sgcd−/− mice at 10 weeks of age. To define the time course of disease progression in Group 2, fibrosis was quantified in skeletal muscle and heart tissues of C57BL/6 (n = 4-5, 2-3 males, 2 females) and Sgcd−/− (n = 4-7, 2-4 males, 2-3 females) mice at different ages between 10 and 75 weeks (~6 wk intervals). Furthermore, we also calculated the percentage of nuclei located centrally within skeletal muscle myocytes at each age. The presence of centralized nuclei is an established feature of dystrophic muscle.

Data Analysis

The results are expressed as means ± standard error of the mean (SEM). Significant differences were defined at P<0.05. Differences between two independent groups of mice were compared using unpaired t-test. Because of substantial age-related mortality in the Sgcd−/− mice, effects of genotype and age on cardiovascular, autonomic and locomotor activity parameters were analyzed by 2-factor ANOVA (unpaired, non-repeated measures) followed by Fishers PLSD post-hoc test (StatView SAS Institute, Cary, NC). The relationships between autonomic indices (sympathetic and vagal tone, BRS) and left ventricular function (EF, EDV) or mortality were analyzed by linear regression analysis.

RESULTS

Autonomic Regulation is Impaired in Young Sgcd−/− Mice while Cardiac Function is Preserved

At 10-12 weeks of age, mean arterial pressure (MAP) and locomotor activity (24 hr averages) were significantly lower in Sgcd−/− mice compared with C57BL/6 control mice, whereas mean HR did not differ between the genotypes (Fig 1). The normally occurring diurnal variations in MAP, HR and activity were markedly attenuated in Sgcd−/− mice (Fig 1A, Table 1). As a result, MAP was significantly lower in Sgcd−/− vs. control mice during the night (dark phase), but not lower during the day when activity was low and not different between the genotypes (Fig 1A, Table 1). In addition, while the 24 hour average HR did not differ between genotypes, HR was significantly higher in Sgcd−/− vs. control mice during the day (Table 1).

Figure 1. MAP, HR and locomotor activity in young (10-12 wks), untreated and losartan-treated C57BL/6 and Sgcd−/− mice.

Panel A. Diurnal changes in MAP, HR and activity measured over 24 hours in untreated control C57BL/6 (black, n = 5) and Sgcd−/− (blue, n = 7) mice. Sgcd−/− mice exhibited less diurnal variability in MAP, HR and activity than C57BL/6 mice. Panel B. 24-hr averages of MAP, HR and activity in untreated and losartan-treated mice. Compared with untreated C57BL/6 mice (n = 5), untreated Sgcd−/− mice (n = 7) exhibited lower MAP, similar HR, and reduced activity. Treatment with losartan decreased HR and increased activity without lowering MAP in Sgcd−/− mice (n = 6), but did not affect these variables in C57BL/6 mice (n = 6). *P<0.05, vs. C57BL/6 mice; †P<0.05, vs. untreated mice.

Table 1.

Diurnal variations in mean arterial pressure (MAP), heart rate (HR), and locomotor activity in untreated and losartan-treated C57BL/6 and Sgcd−/− mice.

Measurements	C57BL/6 Mice		Sgcd−/− Mice
	Untreated (n=5)	Losartan (n=6)	Untreated (n=7)	Losartan (n=6)
24-hr avg MAP (mmHg)	119±4	112±3	100±3*	97±2*
Night-time MAP	136±3	129±4†	105±4*	113±1*†
Day-time MAP	105±4	100±2	97±3	94±3
Diurnal Variation (Night – Day)	31±2	29±4	7±1*	19±2*†
24-hr avg HR (bpm)	591±12	599±9	612±15	553±10*†
Night-time HR	644±24	659±16	623±12	595±18*†
Day-time HR	530±13	554±24	598±15*	522±18
Diurnal Variation (Night – Day)	114±21	105±11	25±4*	73±13*†
24-hr avg Activity (c/min)	11±2	10±1	4±1*	7±1†
Night-time Activity	23±5	22±4	6±1*	10±2*
Day-time Activity	3±1	5±1	2±1	2±1
Diurnal Variation (Night – Day)	21±4	17±3	4±1*	7±1*†

^*Sgcd−/− vs. C57BL/6, P<0.05

^†Losartan-treated vs. Untreated, P<0.05

Night Time 6:00pm – 12:00am (6hr avg)

Day Time 7:00am – 1:00pm (6hr avg)

Measurements of autonomic indices during the day revealed severe autonomic dysregulation in young Sgcd−/− mice including significant decreases in spontaneous BRS and cardiac vagal tone, and an increase in sympathetic vasomotor tone (Fig 2A). The increase in cardiac sympathetic tone did not reach statistical significance (Fig 2A). Left ventricular systolic function assessed by echocardiography was not significantly different in young Sgcd−/− and control mice at 10-12 weeks of age (Fig 2B).

Figure 2. Autonomic indices and cardiac function in young (10-12 wks), untreated and losartan-treated C57BL/6 and Sgcd−/− mice.

Panel A. Sgcd−/− mice exhibited severe autonomic dysfunction at a young age, which was prevented by treatment with losartan. Results from untreated (n = 5) and treated (n = 6) C57BL/6 mice, and untreated (n = 7) and treated (n = 6) Sgcd−/− mice are shown. *P<0.05, Sgcd−/− vs. C57BL/6 mice; †P<0.05, vs. untreated mice. Panel B. Left ventricular ejection fraction and end diastolic volume were not significantly different in Sgcd−/− vs. C57BL/6 mice, and were not affected by losartan treatment.

Losartan Abrogates Autonomic Dysregulation and Increases Locomotor Activity in Young Sgcd−/− Mice

Administration of losartan in drinking water to Sgcd−/− mice for approximately 8 weeks beginning at 3 weeks of age prevented or strongly attenuated the cardiovascular, locomotor and autonomic dysregulation observed in untreated mice (Figs 1 and 2, Table 1). Losartan significantly decreased mean HR, increased locomotor activity, and increased the diurnal variations in MAP, HR and activity in Sgcd−/− mice, without affecting the 24 hr average MAP (Fig 1B, Table 1). Furthermore, losartan significantly increased BRS and cardiac vagal tone, and decreased both cardiac and vasomotor sympathetic tone in Sgcd−/− mice (Fig 2A). The autonomic indices in treated Sgcd−/− mice were restored to levels not significantly different than those measured in untreated control mice (Fig 2A). Losartan treatment had essentially no effect on the variables measured in C57BL/6 control mice (Figs 1 and 2A, Table 1), and did not affect left ventricular function significantly, in either Sgcd−/− or control mice (Fig 2B).

Circulating and Skeletal Muscle RAS are Activated in Young Sgcd−/− Mice

The concentrations of RAS peptides were increased in plasma of young, 10-12 week old Sgcd−/− mice, most notably angiotensin I (Ang-(1-10)) and Ang II (Ang-(1-8)) (Fig 3A). Levels of the angiotensin peptide Ang-(1-7) did not differ in plasma from Sgcd−/− vs. control mice (Fig 3A).

Figure 3. Quantification of RAS peptides.

RAS peptides in plasma and skeletal muscle (quadriceps) measured by liquid chromatography-mass spectrometry were increased in untreated, 10-week old Sgcd−/− (n = 4) vs. C57BL/6 (n = 4) mice. Panel A. RAS peptides in plasma. Diameter of spheres reflects the mean concentration of the respective peptides in plasma from C57BL/6 (left) and Sgcd−/− mice (right). Peptide levels (pg/ml) are provided next to each sphere. The amino acid sequence annotation of each angiotensin metabolite (in parentheses) is based on the decapeptide Ang I (1-10) that is cleaved by the proteases (blue arrows) connecting their substrates and products (AP, aminopeptidases; NEP, neutral endopeptidase; DAP, di-aminopeptidase; ACE, angiotensin converting enzyme; ACE2, angiotensin converting enzyme 2). Bar graph shows means ± SEM for Ang I (1-10), Ang II (1-8), and Ang-(1-7). Ang I and Ang II were increased significantly in plasma from Sgcd−/− vs. C57BL/6 mice (P<0.05). Panel B. RAS peptides in skeletal muscle. Ang I and Ang II were increased significantly in Sgcd−/− (blue) vs. C57BL/6 (black) mice. The ratio of Ang II (1-8)/Ang-(1-7) was markedly increased in Sgcd−/− vs. C57BL/6 mice (right). *P<0.05, Sgcd−/− vs. C57BL/6 mice.

Ang-(1-10) and Ang II (Ang-(1-8)) were also increased in skeletal muscle (quadriceps) of Sgcd−/− mice (Fig 3B, left panel). While the decrease in Ang-(1-7) in skeletal muscle of Sgcd−/− did not reach statistical significance (Fig 3B, left panel), the ratio of Ang II to Ang-(1-7) was markedly increased (Fig 3B, right panel).

Losartan Abrogates AT₁R Expression, Oxidative Stress and Fibrosis in Dystrophic Skeletal Muscle

Ang II binding to AT₁R generates superoxide and the resulting oxidative stress promotes upregulation of AT₁R expression in states of chronic RAS activation (Gao et al. 2005). We confirmed our previous findings (Sabharwal et al. 2014) that AT₁R protein expression, superoxide, and fibrosis are all increased in quadriceps skeletal muscle of young (10-12 wks) Sgcd−/− mice (Fig 4). Treatment with losartan decreased AT₁R-imunoreactivity in quadriceps of Sgcd−/− mice by 48%, whereas it had no effect in muscle of C57BL/6 mice (Fig 4B). Losartan treatment also decreased superoxide (Fig 5A) and fibrosis (Fig 5B) in quadriceps of Sgcd−/− mice. Similar changes in AT₁R expression, oxidative stress and fibrosis from untreated and losartan-treated Sgcd−/− mice were also observed in gastrocnemius muscle (data not shown).

Figure 4. Expression of Ang II AT₁R in skeletal muscle of young (10-12 wks), untreated and losartan-treated C57BL/6 and Sgcd−/− mice.

Panel A. Western blots showing increased AT₁R protein expression in quadriceps from untreated Sgcd−/− (n = 6) vs. C57BL/6 (n = 6) mice. Panel B. Immunofluorescence showing increased AT₁R-immunoreactivity in quadriceps of untreated Sgcd−/− vs. C57BL/6 mice, but normal expression in losartan-treated Sgcd−/− mice. Left: Confocal images of muscle with AT₁R staining in green and nuclear staining in blue (magnification, 63X; scale bar, 100 microns). Right: Quantification of AT₁R-immunoreactivity in arbitrary units (AU) in muscle from untreated (n = 5) and treated (n = 6) C57BL/6 mice, and untreated (n = 4) and treated (n = 5) Sgcd−/− mice. *P<0.05, Sgcd−/− vs. C57BL/6 mice; †P<0.05, vs. untreated mice.

Figure 5. Oxidative stress and fibrosis in skeletal muscle of young (10-12 wks), untreated and losartan-treated C57BL/6 and Sgcd−/− mice.

Panel A. Left: Superoxide indicated by dihydroethidium (DHE) fluorescence (red) in 7 micron sections of quadriceps from C57BL/6 and Sgcd−/− mice (magnification, 20X; scale bar, 50 microns). Right: Quantification of superoxide in muscle from untreated (n = 5) and treated (n = 4) C57BL/6 mice, and untreated (n = 4) and treated (n = 4) Sgcd−/− mice. Fluorescence is expressed as a percentage of the mean level of fluorescence measured in muscle from untreated C57BL/6 mice. Panel B. Left: Fibrosis indicated by blue staining (Masson Trichrome) in tissue sections of quadriceps from C57BL/6 and Sgcd−/− mice. Right: Fibrosis is expressed as a percentage of the total area examined in skeletal muscle from untreated (n = 7) and treated (n = 8) C57BL/6 mice, and untreated (n = 6) and treated (n = 7) Sgcd−/− mice. Losartan markedly attenuated both oxidative stress and fibrosis in Sgcd−/− mice, with no effect on either measurement in C57BL/6 mice. *P<0.05, Sgcd−/− vs. C57BL/6 mice; †P<0.05, vs. untreated mice.

Autonomic Regulation Worsens with Age in Sgcd−/− Mice leading to DCM and Premature Death

To determine the time course of disease progression in Sgcd−/− mice and define the temporal relationship between autonomic and cardiac dysfunction we repeatedly measured cardiovascular, locomotor, autonomic, and left ventricular function phenotypes in C57BL/6 and Sgcd−/− mice between 10 and 75 weeks of age. Mean 24 hr-average values of MAP and locomotor activity remained significantly lower in Sgcd−/− vs. control mice over the entire 75-week age range (Fig 6A,B). The very low BRS and cardiac vagal tone observed in young Sgcd−/− mice (10 wks) persisted at older ages with values decreasing further with increasing age (Fig 6C,D). The increase in cardiac sympathetic tone in Sgcd−/− mice reached statistical significance at 18 weeks of age and continued to rise to very high levels over the 75 week period (Fig 6E). Sympathetic vasomotor tone in Sgcd−/− mice showed a similar, pronounced increase with age (Fig 6F). Sympathetic tone did not increase over the age-range studied in C57BL/6 control mice (Fig 6E,F). While the general pattern of the progression of autonomic dysregulation was present in each of the 8 Sgcd−/− mice, the severity of the autonomic dysfunction and the rate of progression with age varied significantly between mice (Table 2).

Figure 6. Age-dependent changes in mean arterial blood pressure, locomotor activity, and autonomic regulation in C57BL/6 and Sgcd−/− mice.

Mean blood pressure, locomotor activity, baroreflex sensitivity, cardiac vagal tone, cardiac sympathetic tone, and vasomotor sympathetic tone were measured once every 4 weeks, from 10 weeks to up to 75 weeks of age in C57BL/6 (n = 6, black) and Sgcd−/− (n = 4-8, blue) mice. Lower levels of MAP and locomotor activity, and autonomic dysregulation were evident in Sgcd−/− mice at 10 weeks of age. Activity and autonomic regulation continued to worsen with age in Sgcd−/− mice while MAP remained stable. *P<0.05, Sgcd−/− vs. C57BL/6. †P<0.05 vs. 10 wks of age.

In contrast to the autonomic and locomotor phenotypes, left ventricular systolic function was normal in most (six of eight) of the Sgcd−/− mice until 30 weeks of age (Fig. 7 and Table 2). Cardiac function deteriorated in some Sgcd−/− mice between 30 and 50 weeks, but remained relatively normal in others (Table 2). At older ages (>50 wks), decreases in EF and increases in EDV, ESV, and EDV/mass were consistently observed, albeit the severity of dysfunction varied between mice (Table 2, Fig 7A-7D). Furthermore, 4 of the 8 Sgcd−/− mice (50%) died spontaneously before the end of the study (Table 2). All of the 6 control C57BL/6 mice survived until the end of the study. Analysis of birth-death records of mice in our colony not involved in any study confirmed that Sgcd−/− mice died at significantly younger ages than C57BL/6 control mice (Fig 7E). Significant mortality in Sgcd−/− mice began at ~60 weeks of age, and 50% mortality occurred at ~100 weeks of age in Sgcd−/− mice compared with ~135 weeks in control mice (Fig 7E). Interestingly, the spontaneous deaths of the 4 Sgcd−/− mice in this study occurred at a younger age than predicted by the Kaplan-Meier plot of the colony mice (Table 2, Fig 7E).

Figure 7. Age-dependent changes in left ventricular (LV) function and Kaplan-Meier survival analysis in C57BL/6 and Sgcd−/− mice.

Panels A-D. LV ejection fraction, end diastolic volume, end systolic volume, and end diastolic volume/mass ratio were measured by echocardiography once every 4 weeks, from 10 weeks to up to 75 weeks of age in C57BL/6 (n = 6, black) and Sgcd−/− (n = 4-8, blue) mice. LV function did not differ in Sgcd−/− vs. C57BL/6 mice at younger ages (less than ~50 wks). While LV function was well preserved in C57BL/6 mice throughout the age range studied, it deteriorated at older ages (>50 wks) in Sgcd−/− mice. *P<0.05, Sgcd−/− vs. C57BL/6. †P<0.05 vs. 10 wks of age. Panel E. Sgcd−/− mice (n = 260, blue) exhibited increased mortality compared with C57BL/6 mice (n = 508, black) (Kaplan-Meier analysis, Logrank Test, Chi Square = 104.1, df = 1, P<0.0001).

Table 2.

Progression of autonomic dysregulation and left ventricular dysfunction in individual Sgcd−/− mice studied at 10, 30, 50, and 75 weeks of age.

Sgcd−/− Mice	BRS (ms/mmHg)				Vagal Tone (Δbpm)				Sympathetic Tone (Δbpm)				EF (%)				EDV (μl)
	10	30	50	75	10	30	50	75	10	30	50	75	10	30	50	75	10	30	50	75
Mouse #1	0.62	0.29	x	x	+16	+4	x	x	−212	−301	x	x	72	44	x	x	30	39	x	x
Mouse #2	0.69	0.36	x	x	+30	+18	x	x	−171	−230	x	x	79	45	x	x	29	34	x	x
Mouse #3	1.38	1.23	0.70	x	+51	+46	+21	x	−166	−256	−304	x	86	62	40	x	28	22	46	x
Mouse #4	0.99	0.95	0.49	x	+100	+60	+12	x	−101	−139	−271	x	86	84	66	x	26	25	42	x
Mouse #5	1.65	1.50	0.93	0.53	+40	+30	+25	+10	−158	−164	−180	−307	86	73	72	34	21	20	30	42
Mouse #6	0.77	1.11	0.71	0.56	+53	+45	+40	+39	−96	−114	−150	−182	86	85	76	47	22	20	27	35
Mouse #7	0.85	0.93	0.62	0.33	+26	+17	+9	−1	−92	−106	−156	−281	86	85	80	37	28	20	29	53
Mouse #8	0.80	0.92	0.88	0.70	+45	+40	+10	+17	−87	−90	−104	−130	82	82	79	58	25	20	26	38

The mice are arranged in order of lifespan and cardiac sympathetic tone. Sgcd−/− mice that died (indicated with an ‘x’) during the study were Mouse #1 at 31 weeks of age, Mouse #2 at 37 weeks, Mouse #3 at 53 weeks, and Mouse #4 at 56 weeks. BRS, baroreflex sensitivity; Vagal Tone, HR response to methylatropine; Sympathetic Tone, HR response to propranolol; EF, left ventricular ejection fraction; EDV, left ventricular end-diastolic volume

Histological analysis of skeletal muscle and heart confirmed the presence of dystrophic features (centralized nuclei and fibrosis) in skeletal muscle of Sgcd−/− mice at a young age and progressive worsening of fibrosis in Sgcd−/− skeletal muscle with age (Fig 8). Fibrosis (% area) in heart was negligible in young Sgcd−/− mice but increased significantly with age, averaging 1.2 ± 0.1%, 3.8 ± 0.8% and 8.9 ± 2.1% at 10, 34 and 75 weeks of age (n = 4-7). Corresponding values of fibrosis in C57BL/6 control mice at the same ages averaged 0.9 ± 0.2%, 2.9 ± 0.6% and 4.0 ± 1.0% (n = 4-5).

Figure 8. Presence of centralized nuclei (A) and area of fibrosis (B) in quadriceps skeletal muscle of mice between 10 and 75 weeks of age.

The percentage of nuclei that were centrally located within myocytes was markedly increased whereas fibrosis (% area) was more modestly increased in young (10 wk) Sgcd−/− vs. C57BL/6 mice. Fibrosis increased markedly with age in Sgcd−/− mice, while the number of centralized nuclei increased only slightly at older ages in Sgcd−/− mice. Data at each age were collected from C57BL/6 (black, n = 4-5), and Sgcd−/− (blue, n = 4-7) mice. * Sgcd−/− vs. C57BL/6, P<0.05. † vs. 10wks of age, P<0.05. Bottom: Examples of tissue sections stained with Masson Trichrome to show age-dependent increase in fibrosis in Sgcd−/− mice muscle (blue color) (magnification, 20X; scale bar, 100 microns).

Older Sgcd−/− mice (75 wks) also exhibited increases in heart weight to body weight ratio (measured post-mortem) and left ventricular mass to body weight ratio (calculated from echocardiograms) compared with age-matched C57BL/6 control mice and young Sgcd−/− mice (10 wks) (Table 3). Lung weight to body weight ratios were not significantly different in Sgcd−/− vs. C57BL/6 control mice at young or older ages (Table 3).

Table 3.

Body weights, heart weights, lung weights, and left ventricular (LV) mass in C57BL/6 and Sgcd−/− mice at 10 and 75 weeks of age.

Measurements	C57BL/6 Mice		Sgcd−/− Mice
	10 wks (n=9)	75 wks (n=8)	10 wks (n=8)	75 wks (n=7)
Body Wt (g)	21±3	29.3±2†	20±4	23±2*
Heart Wt (mg)	117±1.7	175±3.8†	104±1.4	189±0.4*†
Heart Wt/Body Wt (mg/g)	5.6±0.8	6.0±0.6	5.2±1.0	8.2±1.2*†
Lung Wt/Body Wt (mg/g)	6.4±1.0	7.3±0.8	6.9±1.3	7.8±1.1
LV Mass/Body Wt (mg/g)	3.1±1.1	3.4±0.9	3.0±0.7	5.4±0.2*†

^*Sgcd−/− vs. C57BL/6, P<0.05

^†75 wks vs. 10 wks, P<0.05

LV mass calculated from echocardiograms.

Autonomic Dysregulation in Young Sgcd−/− Mice Predicts Later Development of DCM

By measuring autonomic indices and systolic left ventricular function repeatedly at four week intervals in the same mice, we were able to assess whether the severity of autonomic dysregulation observed at a young age in Sgcd−/− mice predicts the severity of left ventricular dysfunction measured in the same mice at older ages. Indeed, cardiac sympathetic tone measured in Sgcd−/− mice at 30 weeks of age was inversely related to left ventricular EF (r² = 0.916, P<0.05) and directly related to left ventricular EDV (r² = 0.671, P<0.05) measured at 50 weeks of age (Fig 9A). Thus, higher levels of cardiac sympathetic tone at a young age predicted more severe left ventricular dysfunction measured 20 weeks later.

Figure 9. Associations between autonomic indices and left ventricular (LV) function in Sgcd−/− mice.

Panel A. Cardiac sympathetic tone (HR response to propranolol) measured at a young age (30 wks) was inversely related to LV ejection fraction (EF) and positively related to LV end diastolic volume (EDV) measured at an older age (50 wks) in Sgcd−/− mice (P<0.05). Data points from individual C57BL/6 (open triangles, n = 6) and Sgcd−/− (filled circles, n = 6) mice are plotted in Panels A and B. The correlation between sympathetic tone and ejection fraction remained significant after excluding the data from the Sgcd−/− mouse with the highest sympathetic tone (r² = 0.547, P<0.05). Panel B. In contrast, neither cardiac vagal tone (HR response to methyl-atropine, left) nor spontaneous baroreflex sensitivity (BRS, right) measured at a young age (10 wks) predicted the severity of LV dysfunction (EF at 50 wks).

Interestingly, neither cardiac vagal tone nor BRS measured at a young age predicted the severity of left ventricular dysfunction measured at older age (Fig 9B). Vagal tone and BRS tended to correlate with longer lifespan for the four Sgcd−/− mice that died before the end of the study (r² = 0.529 and 0.692), but the low number of mice that died precluded a meaningful statistical analysis. While Sgcd−/− mice with the most severe autonomic impairment at a younger age developed severe DCM and died prematurely, the Sgcd−/− mice with modest autonomic dysfunction at a younger age exhibited less left ventricular dysfunction and survived until completion of the study (Table 2).

DISCUSSION

The results reported here confirm our recently published findings showing that Sgcd−/− mice exhibit increases in AT₁R expression, oxidative stress and fibrosis in skeletal muscle, and severe autonomic dysregulation at a young age (10-12 wks) (Sabharwal et al. 2014). The major new findings of this study are: 1) Left ventricular systolic function is normal in young Sgcd−/− mice with autonomic dysregulation; 2) Circulating and skeletal muscle RAS are activated in young Sgcd−/− mice; 3) Treatment of young Sgcd−/− mice with the AT₁R blocker losartan attenuates the pathology in dystrophic skeletal muscle, abrogates the autonomic dysregulation, and increases locomotor activity; 4) Autonomic regulation deteriorates further with age in Sgcd−/− mice culminating in DCM and premature death; and 5) Autonomic dysregulation in young Sgcd−/− mice predicts later development of DCM.

We discuss below the mechanisms underlying the changes in blood pressure, HR and autonomic regulation that occur early in the progression of disease in muscular dystrophy with an emphasis on the role of the RAS; the temporal relationship between autonomic and cardiac dysfunction; and therapeutic implications of our findings.

Mechanisms of Changes in MAP, HR and Autonomic Regulation—Role of the RAS

Patients with Duchenne muscular dystrophy exhibit normal MAP and increased HR (Boas & Lowenburg, 1931; Sander et al. 2000). Our results are consistent with this pattern in that MAP did not differ and HR was higher in Sgcd−/− vs. control mice when measured during the day when activity levels were low in both groups of mice (Fig 1, Table 1). The increased HR measured during the day reflects the high cardiac sympathetic tone and low cardiac vagal tone measured in Sgcd−/− mice when they were inactive (Fig 2A). Thus, the low level of locomotor activity in Sgcd−/− vs. C57BL/6 mice during the night (active dark phase) and markedly reduced diurnal variations in activity, HR and MAP explain our findings of a lower MAP and lack of an increase in HR in Sgcd−/− mice when measured as 24 hour averages (Fig 1A, Table 1). Activity level is a major determinant of MAP and HR in mice (Van Vliet et al. 2003).

Decreased diurnal variations in MAP and HR as well as decreased HRV have been observed in boys with Duchenne muscular dystrophy (Yotsukura et al. 1995; Sander et al. 2000; Lanza et al. 2001; Inoue et al. 2009), both of which have been associated with increased risk of cardiovascular events (Lanza et al. 2001; Portaluppi et al. 2012). Increased HR and decreased HRV have also been demonstrated in the mdx mouse model of Duchenne muscular dystrophy and the hamster model of sarcoglycanopathy (Giudice et al. 2000; Chu et al. 2002; Hampton et al. 2012). To our knowledge we are the first to show increased resting cardiac and vasomotor sympathetic tone in muscular dystrophy prior to development of left ventricular dysfunction (Figs 2,6,7). The combination of high sympathetic tone and impaired sympatholysis during exercise (Thomas et al. 1998; Sander et al. 2000) likely contributes to skeletal muscle ischemia and worsening of muscle pathology in muscular dystrophy.

Losartan treatment decreased 24-hr average HR, increased 24-hr average activity, and restored diurnal variations in MAP, HR and activity in Sgcd−/− mice without affecting these measures in control mice (Fig 1B, Table 1). Losartan also abolished abnormalities in vagal and sympathetic tone and BRS in Sgcd−/− mice, without affecting these autonomic indices in control mice (Fig 2A). The favorable cardiovascular, autonomic and locomotor responses to losartan in young Sgcd−/− mice were accompanied by decreases in fibrosis, AT₁R expression and superoxide in skeletal muscle, consistent with previous reports of activation of the RAS in dystrophic muscle in both mice and patients (Dietze & Henriksen, 2008; Sun et al. 2009; Cabello-Verrugio et al. 2012; Morales et al. 2012) and decreases in fibrosis and oxidative stress in skeletal muscle of mdx mice treated with either losartan or the ACE inhibitor enalapril (Cohn et al. 2007; Cozzoli et al. 2011; Spurney et al. 2011). Both positive (Cohn et al. 2007; Cozzoli et al. 2011) and negative results (Spurney et al. 2011; Bish et al. 2011) results have been reported in regards to whether blockade of AT₁R or inhibition of ACE can improve skeletal muscle function in mdx mice. In two studies, losartan failed to improve skeletal muscle function despite a clear improvement in cardiac function and decrease in mortality (Spurney et al. 2011; Bish et al. 2011). Our finding that losartan treatment increases spontaneous locomotor activity in Sgcd−/− mice (Fig 1, Table 1) is consistent with improved skeletal muscle function. Importantly, the beneficial effects of losartan occurred without lowering MAP in Sgcd−/− mice (Fig 1B, Table 1), which might have otherwise compromised blood flow to skeletal muscle further.

The mechanism(s) underlying autonomic dysregulation in young Sgcd−/− mice is unknown. We speculate that activation of Group III and IV muscle afferents by inflammatory mediators produced in dystrophic muscle and/or by abnormal mechanical coupling between muscle and sensory nerve endings may evoke reflex increases in sympathetic activity, inhibition of parasympathetic activity, and decreased BRS. Activation of this reflex during exercise (Delliaux et al. 2009; Kaufman, 2012) is enhanced in heart failure (Piepoli et al. 2008; Koba et al. 2009; Piepoli & Crisafulli, 2014). Similar mechanisms including activation of the RAS in skeletal muscle may enhance this sympathoexcitatory reflex in muscular dystrophy as a result of oxidative stress, acidosis and/or inflammation. We demonstrated increases in Ang II and AT₁R expression in dystrophic skeletal muscle of Sgcd−/− mice (Figs 3B and 4). AT₁R are expressed in several cell types in dystrophic muscle including sensory nerve terminals (Pavel et al. 2008; Sun et al. 2009; Patil et al. 2010). AT₁R may function as mechanosensors (Mederos y Schnitzler et al. 2011), and therefore may contribute to mechanically-induced damage of myocytes and/or activation of mechanosensitive nerve terminals. Proinflammatory cytokines released from dystrophic muscle and increased circulating levels of Ang II (Fig 3A) may also cause autonomic dysregulation via actions at central and peripheral nervous system sites (Reid, 1992; Ferguson & Bains, 1997; O’Callaghan et al. 2013).

Temporal Relationship between Autonomic and Cardiac Dysfunction

In primary heart disease, the severity of autonomic dysregulation predicts occurrence of arrhythmias and mortality (Cohn et al. 1984; La Rovere et al. 1998). We observed a similar relationship in older Sgcd−/− mice (75wks) where increases in cardiac sympathetic tone were associated with decreases in left ventricular ejection fraction (Fig 10, right panel). In contrast to primary heart disease where cardiac and autonomic dysfunction occur concurrently, the onset of left ventricular dysfunction in Sgcd−/− mice was delayed reaching statistical significance at ~50 weeks of age (Fig 7A-D), well beyond the age when autonomic dysregulation was first observed (Figs 2A and 6). The age of onset and severity of cardiac dysfunction varied substantially between individual Sgcd−/− mice (Table 2). Remarkably, cardiac sympathetic tone measured in Sgcd−/− mice at a young age predicted the severity of left ventricular dysfunction measured 20 weeks later in the same mice (Fig 9A).

Figure 10. Relationships between cardiac sympathetic tone and left ventricular ejection fraction (EF) in C57BL/6 and Sgcd−/− mice at 10 and 75 weeks of age.

At a young age (10 wks, left panel), there was no relationship between cardiac sympathetic tone (HR response to beta adrenergic receptor blocker propranolol) and left ventricular EF despite a variable increase in sympathetic tone in Sgcd−/− mice (y = −0.0007x + 0.929 for Sgcd−/− mice, r² = 0.467). In contrast, at an older age (75 wks, right panel), higher sympathetic tone was associated with decreased EF in Sgcd−/− mice (y = −0.0013x + 0.7173, r² = 0.972, P<0.05). C57BL/6 mice (triangles): 10 wks, n = 6; 75 wks, n = 6. Sgcd−/− mice (circles):10 wks, n = 8; 75 wks, n = 4.

The strength of our repeated measures experimental design was tempered by the relatively small number of mice studied and the substantial age-related mortality in Sgcd−/− mice (Figs 7E and 9A). Moreover, Sgcd−/− mouse #3 exhibited by far the highest sympathetic tone and the most severe left ventricular dysfunction (Table 2, Fig 9A). To determine the influence of this single mouse on our conclusions, we repeated the regression analyses after excluding the data from Sgcd−/− mouse #3. The correlation between the cardiac sympathetic tone at 30 weeks of age and EF at 50 weeks of age remained significant (r² = 0.547, P<0.05) after excluding this mouse, and the correlation between sympathetic tone and EDV was close to reaching statistical significance (r² = 0.257, P=0.065).

It is most important to note that the regression analyses presented in Figure 9 do not take into account the two Sgcd−/− mice that died before reaching 50 weeks of age (Mouse #1 and #2, Table 2). These two mice exhibited the highest sympathetic tone of all the mice at 10 weeks of age and the lowest left ventricular ejection fraction at 30 weeks of age (Table 2). Thus, the results from these two mice strongly reinforce the positive correlation between sympathetic tone and later left ventricular dysfunction.

Interestingly, decreases in vagal tone and BRS in Sgcd−/− at a young age failed to predict left ventricular dysfunction (Fig 9B), but did tend to correlate with lifespan, consistent with earlier studies showing increased susceptibility to ventricular arrhythmias and mortality post-myocardial infarction in subjects with decreased HRV and/or BRS (La Rovere et al. 1998). Precipitous, age-related decreases in BRS and vagal tone were observed preceding the spontaneous death of the four Sgcd−/− mice that died (Table 2). Our results therefore encourage assessment of both sympathetic and parasympathetic function in patients with muscular dystrophy as each may provide unique prognostic value. The reason that the four Sgcd−/− mice died at younger ages than one would have predicted from the Kaplan-Meier analysis (Fig 7E) is not known. We speculate that the stress associated with experimental interventions may have been a contributing factor.

While our results are applicable to limb-girdle muscular dystrophy-2F (Nigro et al. 1996; Coral-Vazquez et al. 1999; Hack et al. 2000; Politano et al. 2001), additional studies are needed to determine if our results will hold true for other types of muscular dystrophy. We suspect that our findings will be relevant to other types of muscular dystrophy in which skeletal muscle wasting and weakness, resting tachycardia, and decreased HRV have been shown to be present, such as Duchenne and facioscapulohumeral muscular dystrophy (Yotsukura et al. 1995; Thomas et al. 1998; Sander et al. 2000; Lanza et al. 2001; Chu et al. 2002; Inoue et al. 2009; Della Marca et al. 2010). While Ang II and exaggerated sympathetic-mediated vasoconstriction appear to contribute to Becker’s muscular dystrophy (Jefferies et al. 2005; Martin et al. 2012), the available evidence suggests that HR and HRV are not consistently affected in this disease (Vita et al. 2001; Ammendola et al. 2006; Martin et al. 2012). The latter findings suggest that parasympathetic control of HR may be relatively normal compared with age-matched controls, perhaps related to the older age and variable pathology present in patients with Becker’s muscular dystrophy.

Perspectives and Significance

The results of the present study demonstrate that RAS activation causes severe autonomic dysregulation in young Sgcd−/− mice, which portends a worse long-term prognosis. Therefore, therapeutic targeting of RAS and/or autonomic dysregulation at a young age may improve sympathovagal balance and slow disease progression in patients with muscular dystrophy. The significant delay in the onset of left ventricular dysfunction in muscular dystrophy provides a unique opportunity for therapeutic intervention. The results of this study and our recently published results (Sabharwal et al. 2014) support the early use of RAS antagonists, Ang-(1-7) analogs, and/or β-adrenergic receptor blockers in muscular dystrophy. Others have also reported promising therapeutic benefits of AT₁R antagonists, ACE inhibitors, and Ang-(1-7) in muscular dystrophies (Duboc et al. 2005; Jefferies et al. 2005; Cohn et al. 2007; Cabello-Verrugio et al. 2012; Acuna et al. 2014). In the present study, losartan was administered to young mice. It will be important to determine if treatment of Sgcd−/− mice for longer durations can reduce the severity and/or delay the onset of DCM.

NEW FINDINGS.

• What is the central question of this study?

  Is autonomic dysregulation in a mouse model of muscular dystrophy dependent on left ventricular systolic dysfunction and/or activation of the renin-angiotensin system (RAS), and does it predict development of dilated cardiomyopathy (DCM)?

• What is the main finding and its importance?

  The results demonstrate that autonomic dysregulation precedes and predicts left ventricular dysfunction and DCM in sarcoglycan delta (Sgcd−/−) mice. The autonomic dysregulation is prevented by treatment of young Sgcd−/− mice with the angiotensin II type 1 receptor blocker losartan. Measurements of RAS activation and autonomic dysregulation may predict risk of DCM, and therapies targeting RAS and autonomic dysregulation at a young age may slow disease progression in patients.