Protection against high fat diet induced obesity in mdx
mice: is sarcolipin involved?
by
Frenk Kwon
A thesis
presented to the University of Waterloo in fulfillment of the
thesis requirement for the degree of Master of Science
in Kinesiology
Waterloo, Ontario, Canada, 2018 © Frenk Kwon 2018
ii AUTHOR’S DECLARATION
I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final reversions, as accepted by my examiners.
iii Abstract
Sarcolipin (SLN) is a 31 amino acid proteolipid and one of the regulators of sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA) pumps. Previous research has shown that sarcolipin levels in muscles from wild type (WT) mice increase 2-3 fold in response to consuming a high fat diet, possibly to help mitigate obesity by increasing energy expenditure and burning excess calories. The role of SLN in diet-induced thermogenesis was shown in subsequent studies via high fat feeding of Sln-null mice. In response to consuming a high-fat diet (HFD), Sln-null animals have significantly greater mass and adiposity, in addition to poorer glucose handling, compared with WT mice. Higher than normal levels of SLN have also been found in mdx mice, a murine model of duch*enne muscular dystrophy. Interestingly, it was reported that mdx mice are protected from obesity after high fat feeding. The purpose of this thesis was two-fold: (1) to determine if mdx mice fed a Westernized diet are protected from diet-induced obesity, and (2) to determine the role of SLN in protecting mdx mice against diet-induced obesity. WT, mdx, and mdx/Sln-null mice consumed a Westernized high-fat diet (42% kcal from fat) for 8 weeks while measurements of body weight and food consumption were made on a weekly basis. Glucose tolerance testing (GTT) and measurements of whole body daily metabolic rate with the comprehensive laboratory animal monitoring system (CLAMS) were performed before and after the diet protocols. Western blotting was conducted to determine the relative expression levels of SLN in diaphragm (DIA) and uncoupling protein (UCP)-1 in brown adipose tissue (BAT) from all animals. Fat pads were collected to calculate relative adiposity to determine the level of obesity in all animals. As expected, in response to the high-fat diet, mdx but not mdx/Sln-null mice gained less weight and adiposity (P< 0.05) in comparison to WT mice. Compared with mdx mice, mdx/Sln-null mice did not gain more weight but were more obese (P< 0.05) following the high-fat diet. Whole body metabolic rate (ml O2/kg/hr) was higher (P< 0.05) in mdx mice in comparison
iv
to WT and mdx/Sln-null mice before and after the high-fat diet, which may explain the lower diet-induced obesity in mdx mice. With increased obesity, mdx/Sln-null mice became extremely (P<0.05) glucose intolerant, more so than WT and mdx mice. Interestingly, mdx mice saw no change in glucose tolerance when compared with pre-HFD values, suggesting that mdx mice were protected from HFD-induced glucose intolerance. Western blotting analyses revealed SLN content was 2-fold higher in chow-fed mdx mice compared to chow-fed WT mice. Although SLN expression in DIA remained higher in mdx compared with WT post-HFD, it decreased in response to the high-fat diet in mdx mice and was unchanged in WT mice. UCP-1 expression was found to be higher (P<0.05) after the high-fat diet in all three genotypes, but there were no significant genotype differences in UCP-1 expression despite mdx/Sln-null mice having greater BAT weights compared with mdx mice. Altogether, these results suggest that SLN could play a role in adaptive diet-induced thermogenesis in mdx mice and provide protection against diet-induced obesity and glucose intolerance. However, despite the higher (P<0.05) daily VO2 in mdx compared to mdx/Sln-null mice post-HFD, both saw a non-significant differences in absolute sleeping VO2 comparison to pre-diet values. Thus, it is possible that other thermogenic mechanisms may have caused the increase in resting VO2 of mdx/Sln-null mice.
v
Acknowledgements
First and foremost, I must sincerely thank my supervisor, Dr. A. Russell Tupling, for giving me an opportunity to be part of such a wonderful laboratory. Your support and patience the last three years, through all the complications and tribulations has been greatly appreciated. Your support and mentorship have gotten me through this challenging journey and I cannot thank you enough for that.
I would also like to thank the rest of the Tupling laboratory for their support. This thesis would not have been completed without the guidance of Dr. Dan Gamu. Dan, your encouragement and “tough love” helped me get my project off the ground and your work ethic and determination in the two years we were working together provided me with a role model to hold myself to. Thank you for friendship and support and making things interesting and fun at the lab. To Cat, Emma, Paige, Reilly, Brad, Gabbi, Val, and Khan, you made me feel less like a cog in the laboratory machine but more as a member of an extended family. Your friendship is greatly appreciated, and I wish you all the best in your future endeavours.
Finally, I would like to thank my family for their support and love. This has been a difficult but rewarding journey and you have stood by me with words of encouragement as I pursued this Master’s degree. I am truly lucky to have such a large and supportive family.
vi
Dedication
This thesis is dedicated to my father Jerry, mother Young Ju, uncle Joon, aunt Akane, and my dog Dol Dol. This experience has been at times stressful and difficult at times, but your love and support have helped get through this journey. This work is as much yours as it is mine.
vii Table of Contents AUTHOR’S DECLARATION……….ii ABSTRACT………..iii ACKNOWLEDGEMENTS………..v DEDICATION………..vi List of Tables………...ix List of Figures………...x Introduction………....1
Calcium Regulation and Thermogenesis ………...………..1
Adaptive Thermogenesis and Sarcolipin………..3
duch*ene Muscular Dystrophy and Sarcolipin………..7
High Fat Feeding and mdx mice………...10
Statement of the Problem………...11
Research Objectives………...11
Hypotheses……….11
Methods……….12
Experimental Animals………13
Experimental Design………..14
Whole Body Metabolic Rate Measures………..14
Whole body glucose tolerance tests………15
Tissue Collection………....15
Western Blot Analyses………...16
Statistical Analyses………....…16
Results………...17
viii
Adiposity and Anthropometric Measures………19
CLAMS Measurements………...22
Metabolic Efficiency………...…...31
Glucose Tolerance Tests……….….27
Western Blotting Analysis………....…...………..29
Discussion……….31 Study Limitations………..42 Future Directions………..43 References………...45 Appendix A……….………59 Appendix B………..62 Appendix C………...…63
ix List of Tables
Table 1. CLAMS Measurements of Wild type (WT), mdx mice (mdx), and mdx/Sln-null mice (mdx SLNKO) Pre and Post Chow/HFD………..23
x
List of Figures
Figure 1. Experimental Diet Design; HFD: High Fat Diet; WT: wild type; mdx: murinic model of duch*enne muscular dystropohy………..14 Figure 2. Greater change in mass in WT with no differences in final bodyweight at sacrifice………..18 Figure 3. Total Food Consumption across an 8 week high fat diet (HFD) for wild-type (WT), mdx, and mdx/Sln-null mice………18 Figure 4. Fat pad mass, brown adipose tissue (BAT) weight, and adiposity index for chow fed control and HFD, wild type (WT), DMD mice (mdx), DMD Sln-null (mdx SLNKO) mice….20 Figure 5. Soleus and liver weights for chow fed control and HFD, wild type (WT), mdx, and mdx/Sln-null (mdx SLNKO) mice………..21 Figure 6. Metabolic efficiency (M.E) pre and post HFD ……….26 Figure 7. Glucose tolerance curves and area under the curve (AUC) pre and post 8-week HFD ……….………...28 Figure 8. SLN expression in diaphragm (DIA)……….30 Figure 9. UCP-1 expression in brown adipose tissue (BAT)………30
1
Introduction
Calcium Regulation and Thermogenesis
Calcium ions play a vital role in the physiology of living organisms. With respect to skeletal muscle, calcium regulates excitation-contraction coupling (ECC) (Stammers et al., 2015), apoptosis (Berchtold et al., 2000), secondary messenger signaling (Gailly, 2002), and energy expenditure (Bombardier et al., 2013a; Bombardier et al., 2013b). The precise regulation of intracellular free calcium ([Ca2+]f) is essential for proper skeletal muscle function and health. The sarcoplasmic reticulum (SR) is the primary organelle responsible for the dynamic control of both cytoplasmic and SR calcium levels in striated muscle (Martonosi and Pikula, 2003). The SR maintains a resting [Ca2+]f within a concentration of 20-60 nM while maintaining a relatively greater luminal concentration creating a large Ca2+ gradient (1:10, 000) across the SR membrane (Schiaffino & Reggiani, 2011).
One of the integral SR membrane proteins that contributes to SR [Ca2+]f regulation is the sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA) pump. SERCAs are a 110kDa membrane proteins responsible for the transport of calcium and the maintenance of [Ca2+]f (MacLennan, Asahi, and Tupling, 2003). SERCA is a member of the protein family known as P-type ATPases, which utilize the energy from ATP to act as cation transporters (Toyoshima et al., 2000). The structural organization of SERCA is divided into three domains, a cytoplasmic head region, a transmembrane region containing the calcium binding sites, and luminal loops (Toyoshima et al., 2000; Toyoshima & Inesi, 2004; Stammers et al., 2015). The transmembrane region is comprised of 10 transmembrane domains (M1-M10) and the cytoplasmic head region is characterized by three cytoplasmic domains: the actuator domain, nucleotide binding domain, and the
phosphorylation domain (Toyoshima et al., 2000). Although Ca2+: ATP stoichiometry can be variable, the two calcium binding sites and one ATP binding site on SERCA corresponds to an
2
optimal coupling ratio of 2 Ca2+ for 1ATP hydrolyzed (Toyoshima, 2007). The two major conformational states of SERCA are the E1 and E2 states. Assuming an optimal ratio of 2 Ca2+for 1 ATP hydrolyzed, the reaction cycle for SERCA mediated Ca2+ transport would proceed with 2 Ca2+ binding to the two Ca2+ binding sites on SERCA (Toyoshima & Inesi, 2004). The two major conformational states of SERCA are the E1 and E2 states. During the E1 state, the Ca2+ binding sites (M4, M5, M6, and M8 transmembrane proteins) have a high affinity for cytoplasmic calcium (Toyoshima, 2007). Once calcium ions are bound to SERCA in its E1 conformation, SERCA undergoes autophosphorylation by ATP at Asp351 at its phosphorylation domain forming a high energy phosphoprotein intermediate (E1P) (Toyoshima, 2007). SERCA transitions from E1P to a low energy phosphoprotein intermediate known as E2P (Toyoshima, 2007). The low energy state, E2P, is characterized by low SERCA calcium affinity, resulting in the release of calcium into the SR lumen (Toyoshima, 2007). Although Ca2+: ATP stoichiometry can be variable, the two calcium binding sites and one ATP binding site on SERCA corresponds to an optimal coupling ratio of 2 Ca2+ for 1 ATP hydrolyzed (Toyoshima, 2007). This optimal ratio can be influenced by changes in physiological conditions, whereby the coupling ratio can be altered. A reduction in SERCA coupling ratio can occur as result of alterations in ADP/ATP ratios, [Ca2+]f, and sarcolipin (SLN) (Smith et al., 2002; Mall et al., 2006; Reis, Farage, & de Meis, 2002). There are three sources of reduced SERCA coupling ratio. First is a process known as a passive leak or uncoupled Ca2+ efflux whereby high concentrations of luminal SR Ca2+results in the passive efflux of Ca2+ from the lumen of the SR back into the muscle cytoplasm, this occurs due to the binding of Ca2+ to SERCA in its E2 state before it can revert to its E1 state (Berman, 2001; de Meis, 2001b). The second is known as uncoupled ATPase activity, where the forward conformational transition from E1-Ca2-P to E1-Ca2-P is slowed down due to the high
3
luminal concentrations of Ca2+, this increases the number of SERCA pumps in the E1-Ca2-P state resulting in the cleavage of Pi before Ca2+ is translocated to the SR lumen (Berman, 2001; de Meis, 2001b). Finally, in a process known as slippage where the Ca2+ ion is prematurely released back into the muscle cytoplasm instead of the SR lumen during the conformational transition from E1-Ca2+-P to E2-Ca2+-P. A decrease in SERCA Ca2+ affinity when it is in the E1-Ca2+-P is believed to be the cause of slippage (Smith et al., 2002; Berman, 2001). This reduction in affinity can be the result of high luminal Ca2+ concentrations or an interaction with SERCA regulatory proteins such as SLN and phospholamban (PLN). Although the three modalities of reducing Ca2+: ATP stoichiometry are different, all three have the same result that most of the energy released from the hydrolysis of ATP is released as heat rather than the successful translocation of Ca2+ into the SR (Mall et al., 2006; Reis, Farage, and de Meis, 2002). This reduction in SERCA efficiency as a means to increase energy expenditure has been postulated as novel mechanism of skeletal muscle based adaptive thermogenesis.
Adaptive Thermogenesis and Sarcolipin
Adaptive thermogenesis refers to an increase in energy expenditure due to perturbations in either temperature or diet. It is important for a living organism to possess the ability to
regulate energy expenditure during times of either food shortage or excess since it helps preserve energy stores and maintain body temperature. An inability to regulate either energy stores or body temperature can result in metabolic disorders such as obesity or even death. There are two mechanisms of adaptive thermogenesis: shivering thermogenesis and non-shivering
thermogenesis.
Shivering thermogenesis involves the contraction of skeletal muscles, which generate metabolic heat (Cannon & Nedergaard, 2004). Skeletal muscle contraction/relaxation is a
4
process that is energy dependent requiring ATP at the level of the myosin ATPase and SERCA, two of the major energy consuming complexes during contraction. A complex series of both electrical and mechanical steps are involved in a process known as excitation-contraction (EC) coupling, a term linking the depolarization of the sarcolemma to the release of intracellular calcium from the SR, resulting in the formation of actin-myosin cross-bridges, and force production (Dulhunty, 2006). The formation of cross-bridges between actin and myosin by myosin ATPase, and the eventual re-uptake of calcium by SERCA, consume ATP and generate heat. Skeletal muscle represents the most direct avenue in which to increase energy expenditure either to help maintain body temperature or through exercise activity, prevent obesity. However, the elevated metabolic rate persists despite cessation of shivering during prolonged cold
exposure, which underlies the importance of other mechanisms of thermogenesis, specifically non-shivering thermogenesis.
Non-shivering thermogenesis is the heat generated as result of mitochondrial proton leak seen in brown adipose tissue (BAT) (Cannon & Nedergaard, 2004). Mitochondrial proton leak occurs due to the presence of uncoupling protein 1 (UCP-1) which disrupts the proton (H+) gradient by allowing protons to leak back into the mitochondrial matrix from the intermembrane space (Cannon & Nedergaard, 2004). The disruption of the proton motive force results in the uncoupling of ATP production from the movement of H+ into the intermembrane space, as a result there is an increase in the break down and oxidation of carbon energy substrates (sugar, starch, protein, fat) in order to synthesize more reducing agents (NADH and FADH2) in an effort to restore the disrupted mitochondrial H+ gradient (Cannon & Nedergaard, 2004). Other isoforms of uncoupling proteins exist, UCP-2 is another uncoupling protein seen in BAT (Vidal-Puig et al., 1997), and UCP-3 a skeletal muscle-based isoform (Vidal-Puig et al., 2000) where all three
5
proteins share very similar sequence hom*ologies (Feldmann, 2009). A study examining the effects of the ablation of UCP-1 in mice have found that at thermoneutrality (~30°C; a
temperature at which no further compensatory thermogenesis mechanisms are needed to regulate body temperature), mice fed either a chow or a caloric dense high fat diet (HFD) are obese (Feldmann, 2009). The ability of UCP-1 to uncouple ATP production underlies an increase in energy expenditure which helps mitigate obesity pathogenesis. In a subsequent study, mice with overexpression of UCP-3 in their skeletal muscles were found to be protected from an obesity phenotype despite being fed a HFD (Son et al., 2004). These studies suggest that mice can be protected from obesity due to the existence of non-productive energy expending mechanisms. It has been suggested that other mechanisms may exist which help “waste energy” in an effort to attenuate excess caloric consumption and subsequent weight gain. One such mechanism may be SLN mediated slippage of Ca2+.
SERCA function is regulated by several integral membrane proteins, including SLN (Asahi et al., 2003; Bal et al., 2012). SLN is a 31-amino acid protein, organized into three distinct domains, a cytoplasmic domain comprised of 7 hydrophilic amino acids, a hydrophobic 19 amino acid transmembrane domain, and a 6 amino acid luminal domain (MacLennan, Asahi, & Tupling, 2003). The expression of SLN protein in rodent tissue has been found to be abundant in the atria, the soleus (SOL), and the diaphragm (Vangheluwe et al., 2005a). SLN physically interacts with SERCA at its transmembrane domains (M4, M5, M6, and M8) (Toyoshima et al., 2000; Toyoshima & Inesi, 2004). The M4 and M6 transmembrane domains make up the calcium binding sites, whereby the physical presence of SLN can reduce SERCA calcium affinity and result in the premature release of calcium into the muscle cytoplasm instead of being translocated into the SR lumen during its catalytic cycle. Thus, SLN reduces the apparent affinity for Ca2+ in
6
both cardiac and skeletal muscle, as well as reducing maximal Ca2+ uptake activity (Toyoshima et al., 2000; Toyoshima & Inesi, 2004; Stammers et al., 2015). The uncoupling of ATP
hydrolysis from SERCA Ca2+ uptake is another physiological role that SLN is responsible for. The amount of heat generated per molecule of ATP was increased as a result of SLN
reconstitution in artificial membrane experiments (Smith et al., 2002). A greater rate of slippage may be the potential explanation for SLN ability to uncouple ATP hydrolysis from Ca2+ uptake. As mentioned in the first section, slippage occurs when Ca2+ ions are released prematurely while SERCA undergoes a conformational change between the E1 and E2 states. The unsuccessful translocation of Ca2+ from the muscle cytoplasm to the SR and the fact that ATP hydrolysis still occurs, results in reduced fraction of energy used to transport Ca2+ into the SR thereby reducing the coupling ratio (Ca2+ transported: ATP hydrolyzed) (Smith et al., 2002; Mall et al., 2006; de Meis, 2001a).
The interaction between SLN and SERCA has been postulated as a novel mechanism of diet induced thermogenesis (DIT), a form of non-shivering thermogenesis in response to
excessive calorie consumption. In the soleus (SOL) of SLN ablated (Sln-null) mice, the apparent coupling ratios are higher than in WT counterparts (Bombardier et al., 2003b). A higher coupling ratio signifies a more energetically efficient transport of Ca2+ into the SR lumen, since Sln-null mice are able to transport a given amount of Ca2+ into the SR lumen at a lower energy cost. However, greater efficiency of Ca2+ transport can predispose Sln-null mice to obesity when fed an 8 week HFD (Bombardier et al., 2003b). This has been attributed to the absence of SLN mediated skeletal muscle adaptive thermogenesis and is characterized by greater weight gain, greater adiposity, and poorer glucose tolerance (Bal et al., 2012). In WT mice, a 3-5 fold increase in SLN is also observed in response to high fat feeding, further suggesting that SLN upregulation
7
is a mechanism of preventing excessive weight gain by promoting a decrease in SERCA energy efficiency (Bal et al., 2012; Bombardier et al., 2003a). The upregulation of SLN can also be observed in skeletal muscle disease states, such as duch*enne muscular dystrophy (Schneider et al., 2013).
duch*ene Muscular Dystrophy and Sarcolipin
duch*enne muscular dystrophy (DMD) is an X-linked disease caused by a frame shift mutation in the dystrophin gene, abolishing the production of functional dystrophin protein (Blake et al., 2002; Bulfield et al., 1984; Dupont-Versteegden et al., 1994; Schneider et al, 2013; Sicinski et al., 1989). Dystrophin is a 427 kDa cytoplasmic protein that plays a key role in assembling several cytosolic and transmembrane proteins into the dystrophin associated glycoprotein complex (Blake et al., 2002; Bonilla et al., 1988; Garcia-Pelagio et al., 2011). Although the precise function of dystrophin remains unknown, its role in muscle function has been explored by its absence in animal models.
mdx mice (C57/BL10 background), the murine model of DMD, are genetically similar to DMD, with a nonsense point mutation centered on exon 23 of the dystrophin gene leading to an absence of dystrophin protein (Bonilla et al., 1988; Bulfield et al., 1984; Turner et al., 1988), and is a common model used to understand the DMD pathology (Bulfield et al., 1984; Turner et al., 1988). mdx mice display progressive diaphragm degeneration and cycles of skeletal muscle degeneration and regeneration analogous to DMD (Bonilla et al., 1988, Schneider et al., 2013; Turner et al., 1998; Willmann et al., 2009). The lack of dystrophin has been associated with membrane fragility as indicated by the high concentration of endogenous extracellular proteins albumin, IgG, and IgM, in the muscle cells (McGreevy et al., 2015; Mokhtarian et al., 1996). This has been attributed to the inability of cytoskeletal γ-actin to bind to the sarcolemma, leaving
8
the muscle vulnerable to mechanical stress and muscle damage (Dupont-Versteegden et al., 1994; Garcia-Pelagio et al., 2011; McGreevy et al., 2015). An absence of dystrophin has also been linked to membrane fragility which is associated with aberrant calcium handling and persistent increases in [Ca2+]f (Culligan et al., 2002; McGreevy et al., 2015; Turner et al., 1988; Woods et al., 2004).
The increased influx of Ca2+ is initially started by mechanosensitive cation channels or stretch activated channels, preceding any signs of muscle damage and degeneration in young mdx mice, and is believed to be the primary source of Ca2+ entry into dystrophic muscle (Franco-Obregon Jr & Lansman, 1994). During subsequent contractile activity, transient membrane tears can occur which allow for localized influxes of calcium, a subsequent exocytotic transport of Ca2+ leak channels to the sarcolemma occurs which exacerbates the increased [Ca2+]f (Turner et al., 1991). High [Ca2+]f in skeletal muscle have been associated with degeneration and necrosis in mdx mice muscle (Gailly, 2002; Whitehead et al., 2006). Ca2+-activated proteases, most notably calpains are responsible for promoting the degradation of numerous proteins which include membrane and cytoskeletal proteins. Calpain activity is higher in mdx mice than in WT mice, and remains elevated during periods of regeneration and degeneration (Spencer et al., 1995). The high [Ca2+]f is also responsible for increased reactive oxygen species (ROS) production due to mitochondrial Ca2+ overload which impairs oxidative phosphorylation and normal mitochondrial Ca2+ handling (Kuznetsov, et al., 1998).
With respect to treatment, DMD at present has no known cure, however several treatment options have been explored. Genetic treatments include gene therapy aimed at either expressing a truncated, yet still in frame, dystrophin gene or restoring the dystrophin in its entirety (McGreevy et al., 2015). The goal of such therapy is to restore a functional gene or repair the gene through
9
targeted correction to restore dystrophin in DMD subjects (McGreevy et al., 2015).
Pharmaceutical treatments have primarily focussed on glucocorticoids which help mitigate DMD pathology by reducing the breakdown of protein and increasing muscle mass (McGreevy et al., 2015). Another potential target of treatment has been the aforementioned calcium dysregulation observed in mdx mice. A recent study examining the downregulation of SLN in mdx utrophin knockout (utr -/-) mice, a more severe form of muscular dystrophy that mimics human DMD, saw reduced pathophysiology and improved muscle function (Voit, 2017). One of the major characteristics of mdx mice is their high energy expenditure and inability to meet that energy demand, possibly due to cycles of skeletal muscle degeneration and regeneration (Radley et al., 2014). As a result, dietary interventions have been explored in addition to pharmaceutical and genetic treatment options. Potential dietary interventions aimed specifically at protein/amino acid supplementation have been examined extensively (Radley et al., 2007). However, their benefits remain mixed as some studies have shown potential benefits while others have shown no improvement (for review see Radley et al., 2007). Due to the high energy demand of mdx mice, interventions using a caloric dense HFD have been explored as a therapeutic option (Radley-Crabb et al., 2011).
High fat feeding and mdx mice
Previous work has shown that mdx mice are in a constant state of caloric deficit, where their energy intake is unable to meet their energy demands (Crabb et al., 2011; Radley-Crabb et al., 2014). This calorie deficit has been postulated as an exacerbating complication of the mdx disease, as muscles and protein must be broken down and used as energy substrates to meet mdx energy demands. As a result, Radley-Crabb (2011) fed mdx mice a calorie rich HFD (32% kcal from fat) in an effort to compensate for high energy expenditure to mitigate the
10
severity of mdx mice dystropathology (Radley-Crabb et al., 2011). After 24 weeks, the
bodyweights of the mdx mice did not significantly change after consuming a HFD, whereas WT mice had significantly higher body weights (Radley-Crabb et al., 2011). The differences in bodyweight between mdx and WT mice could not be accounted for by differences in food consumption or by differences in the amount of activity (Radley-Crabb et al., 2011). The body composition comparisons between the two strains revealed that the mdx mice also had
significantly smaller fat pad masses in contrast to WT. High fat feeding also had a profound impact on the dystropathology of mdx mice improving muscle fibre integrity (Radley-Crabb et al., 2011). It should be noted that the diet used in the study was not using a Westernized high fat diet. Although, the diet (32% kcal from fat) is relatively higher in fat in comparison to the chow diet given (5% from fat; 22/5 Rodent Diet (w) 8640; Harlan Teklad, Madison WI), it is
considerably lower than the Westernized high fat diet (42% kcal from fat; product TD 88137; Harlan Teklad, Madison, WI) used in previous studies (Bombardier, 2010; Gamu, 2012). Pilot work has shown that mdx mice fed a Westernized HFD are still protected from obesity. Even with the higher percentage HFD, mdx mice had lower weight gain in comparison to WT mice, had lower adiposity, and displayed lower metabolic efficiency. Potential mechanisms of higher energy expenditure in mdx mice have often focussed on the cycles of degeneration and
regeneration and protein synthesis (Dupont-Versteegden et al., 1994; Mokhatrian et al., 1996; Radley et al., 2007; Radley-Crabb et al., 2014). However, protection against obesity as a possible consequence of SLN upregulation in mdx mice has not been explored, despite SLN’s established role in adaptive thermogenesis.
11
Statement of the Problem
The increase in SLN and protection from diet-induced obesity in mice has lead us to believe that SLN may be a major protective mechanism against diet-induced obesity in mdx mice. To investigate the potential protective role of SLN against obesity in mdx mice, a transgenic line of mdx/Sln-null mice were examined and compared with mdx and WT mice. Research Objectives:
General Objective:
• Determine if SLN mediated diet induced thermogenesis is a major mechanism of obesity protection in mdx mice.
Specific Objectives:
i) To compare relative sarcolipin levels in mdx and WT mice after HFD. ii) To determine if mdx/Sln-null mice are protected against obesity.
iii) To determine if differences in VO2, activity Levels, and food Consumption have any effect on obesity protection in mdx mice.
Hypotheses:
i) SLN levels in mdx mice will be higher in comparison to WT mice. Since HFD increases the expression of SLN, SLN will increase in both strains, but the SLN levels will be higher in mdx mice due to its already high expression in skeletal muscle.
ii) In comparison to WT mice, mdx mice will be protected from obesity after consuming a Western high fat diet, which will be denoted by lower body weight, lower adiposity, and lower blood glucose measures.
12
iii) mdx/Sln-null mice will not be as obese as WT, but will not be protected from obesity after consuming a Western high fat diet with a higher body weight, higher adiposity, and higher blood glucose measures in comparison to mdx mice.
iv) Energy expenditure in mdx mice will be higher in response to the elevated levels of sarcolipin and thermogenesis as a result of high fat feeding, with no significant differences in activity levels.
13
Methods
Experimental Animals
mdx mice were purchased from Jackson Laboratories with intention of breeding an mdx SLN and mdx/Sln-null colony. The mdx mice were crossed with a Sln-null colony maintained at the University of Waterloo in order to create a line of mdx/Sln-null mice. Animals for the
proposed project were hemizygous mdx males and WT littermates. A wild type comparison was considered a necessary group in order to observe the degree of protection from obesity in mdx mice and mdx/Sln-null mice. For the duration of the study, animals were individually housed in an environmentally controlled room on a regular 12:12 light dark cycle and allowed access to food and water ad libitum. Analyses were conducted on soleus and diaphragm, since SLN are normally observed in these tissues. All experiments were reviewed and approved by the University of Waterloo Animal Care Committee in accordance with the Canadian Council on Animal Care.
Experimental Design
WT, mdx, and mdx/Sln-null were randomly assigned to dietary treatments as outlined below (Figure 1). The experimental diets were administered for a period of 8 weeks and comprised of ad libitum access to water and HFD (42% of kcal from fat, product TD 88137; Harlan Teklad, Madison, WI). Body mass was measured on a weekly basis during the
administration of the experimental diets. The experimental design included 8-week chow fed control animals, whose data is included in Appendix A.
14
Whole-Body Metabolic Rate Measures
WT, mdx, and mdx/Sln-null littermates were acclimated in clear plastic cages for one week and given ad libitum access to water and rodent chow (22/5 Rodent Diet (w) 8640; Harlan Teklad, Madison WI), prior to commencing whole-body metabolic measurements. Whole-body metabolic measurements were made using a Comprehensive Laboratory Animal Monitoring System (CLAMS; Oxymax Series; Columbus Instruments, Columbus OH). Animals were housed individually in clear plexiglass cages (20 cm x 10 xm x 12.5 cm) in a room that was temperature controlled to (~22° Celsius), on a reverse light dark cycle and allowed access to water and powdered food (as above) ad libitum. Twenty-four hours were allocated to acclimate the mice to the CLAMS before commencing measurements and data collection for a period of 24 hours thereafter. CLAMS measurements were completed two times during the same week, these measurements will be taken both before and after the experimental diets. Those mice who were allocated to the HFD will be given powdered high-fat food while in the CLAMS instead of the standard chow diet. The percent O2 and CO2 gas levels of the cages (20 cm x 10 cm x 12.5 cm) were measured periodically between reference readings of room air and were used to compute
Figure 1. Experimental Diet Design; Pre: pre-HFD; Post: post-HFD; WT: wild type; mdx: murinic model of duch*enne muscular dystrophy.
15
the O2 consumption (VO2) of the animal and used to calculate the daily metabolic rate. Relative VO2 measurements will be recorded in ml O2/kg/hr, whereas absolute VO2 will factor in the body weights of the animals and be recorded as ml O2/hr. In addition to metabolic rate, the CLAMs are equipped with a scale for monitoring mass of food consumed over time. There were also X and Z activity sensors for monitoring ambulatory activity (dual beam) counts (when 2 adjacent X axis beams are broken in succession) in addition to both total activity and rearing activity.
Whole-Body Glucose Tolerance Tests
Pre- and post-diet Whole body glucose tolerance tests (GTT) were conducted on all mice as outlined in Gamu (2012). Mice were fasted overnight (16 hrs); following the fast, each mice was administered an intraperitoneal injection of 20% D-Glucose (1g/1kg body mass), a 5-10ul sample of venous blood was drawn from the tail vein and blood glucose measured using a glucometer (Contour Next; Bayer Diabetes Canada). Timepoints of blood glucose measure occurred immediately before intraperitoneal injection, after 30, 60, and 120 minutes post-injection.
Tissue Collection
Prior to commencing tissue collection, mice were placed on a four hour fast.
Experimental animals were euthanized by cervical dislocation, the soleus and diaphragm were collected. All tissue samples were frozen in liquid N2 and stored at -80°C for future analyses. The retroperitoneal and epididymal fat pads were excised, cleaned of extraneous tissue and weighed. The weights of the epididymal, and retroperitoneal fat pads were used to calculate the adiposity index as:
16
Western Blotting Analysis
Soleus and diaphragm muscles from WT, mdx, and mdx/Sln-null mice were collected and hom*ogenized 10:1 (volume/weight). Brown adipose tissue (BAT) was also collected and later hom*ogenized 4:1 (volume/weight). Proteins will be separated in glycine based SDS-PAGE (13% Total AB6/AB3) for SLN. Separated proteins will be transferred onto a nitrocellulose membrane for SLN and a PVDF membrane for UCP-1 and immunoprobed with their corresponding primary antibodies and subsequently immunoprobed with the appropriate horseradish
peroxidase-conjugated secondary antibodies. Luminata ForteTM was used to detect SLN and UCP-1 antigen-antibody complexes. Quantification of optical densities was determined using GeneTools (Syngene, MD, USA) and all values were normalized to whole protein content.
Statistical Analyses
One-way ANOVAs were used to test for differences between WT, mdx, and mdx/Sln-null groups with respect to changes in body weight post-HFD. A two-way ANOVA, examining the factors of genotype and diet (pre/post-HFD), was used to detect differences in body weight, food consumption, all CLAMS measures, anthropometric measures such as fat pad masses, muscles and liver weights, as well as protein expression and glucose area under the curve (AUC) across all three genotypes, pre- and post-HFD. Analyses were performed using Statistica 12 (Statsoft. Inc., Tulsa, OK, USA). Data are presented as means ± SE. The significance level was set at 0.05, and when appropriate, a Newman-Keuls post hoc test was used to compare specific means.
17 RESULTS
Body Weight and Food Consumption
Pre-diet body weight was higher (P < 0.0001) in mdx and mdx/Sln-null mice in
comparison to WT mice with no differences observed between mdx groups (Fig.2b). In response to the HFD, body weight of all three genotypes increased significantly (P < 0.0001), with WT gaining more mass versus both mdx and mdx/Sln-null at week 5 (P < 0.03), week 6 (P < 0.02), week 7 (P < 0.02), and week 8 (P < 0.01) of the HFD (Fig. 2a). However, at the time of sacrifice (8 weeks Post-HFD), there were no differences in body weight between genotypes post-HFD (Fig.2b). Weight gain was also not different at any time point between high fat fed mdx and mdx/Sln-null mice (Fig. 2a).
Total food consumption was measured across all 8 weeks of the HFD, and across all three genotypes. The total food consumption was significantly higher (P < 0.05) in mdx mice in
comparison to WT and mdx/Sln-null mice (Figure 3). There was no difference (P = 0.99) in food consumption between WT and mdx/Sln-null mice.
18
Figure 2. Greater change in mass in WT with no differences in final bodyweight at sacrifice. A) Change in body mass of wild-type (WT), mdx, and mdx/Sln-null mice during 8 weeks of a high-fat diet (HFD). B) Pre- and Post-HFD bodyweights of WT, mdx, and mdx/Sln-null mice. Values are mean ± S.E. * Significantly different than corresponding mdx and mdx/Sln-null (P < 0.05). (WT: n = 9, mdx: n = 12, mdx/Sln-null; n = 8)
Figure 3. Total Food Consumption across an 8 week high fat diet (HFD) for wild-type (WT), mdx, and mdx/Sln-null mice. *Significantly different from mdx mice. Values are mean ± S.E. * Significantly different than mdx (P < 0.05). (WT: n = 9, mdx: n = 12, mdx/Sln-null; n = 8) Food Consumption (g)W T m d x m d x /S ln -n u l l05 01 0 01 5 02 0 02 5 0* *W e e k s Mass (g)1 2 3 4 5 6 7 8051 01 52 0W Tm d xm d x /S L N -n u ll*** *Body Weight (g)P r e -H F D P o s t-H F D01 02 03 04 05 06 07 0W Tm d xm d x /S L N - n u ll*A B
18 Adiposity and Anthropometric Measures
After consuming an 8-week high fat diet, an interaction (P < 0.02) was observed with mdx/Sln-null mice having higher epididymal fat pad masses, higher BAT fat pad masses, and a higher adiposity index compared to mdx mice post-HFD (Figure 4). An interaction close to statistical significance (P = 0.07) was observed with mdx/Sln-null mice having greater retroperitoneal fat pad masses in comparison to mdx mice post-HFD. Planned comparisons (Newman-Keuls) revealed that WT mice had greater epididymal fat pad masses (P < 0.001; Figure 4a), greater retroperitoneal fat masses (P < 0.001; Figure 4b), and higher adiposity index (P < 0.01; Figure 4d) than mdx/Sln-null and mdx mice regardless of diet. Post-HFD WT mice had greater (P < 0.001) BAT masses than post-HFD mdx/Sln-null and mdx mice, there were no differences between chow fed WT and mdx (P = 0.45) and mdx/Sln-null mice (P = 0.24). A main effect of diet (P < 0.0001) was observed with fat pad masses and adiposity indexes being higher post-HFD compared to chow.
When comparing soleus weights between groups, there were significant main effects of diet and genotype with soleus mass being heavier post-HFD compared with pre-HFD (P < 0.02) and in both mdx and mdx/Sln-null mice compared with WT (P< 0.0001; Figure 5a). The same main effects were observed when soleus weights were expressed as a percentage of bodyweight (Figure 5c). With respect to liver weights, an interaction (P < 0.05) was observed with WT mice having greater post-HFD liver weights than mdx and mdx/Sln-null mice, despite having the lowest liver mass pre-diet (Figure 5b). When liver weights were normalized to body weight, an interaction (P < 0.05) was observed with greater liver % BW post-HFD in WT mice and mdx/Sln-null mice in comparison to mdx mice, despite WT mice starting out with a low liver % BW and mdx/Sln-null mice having similar liver % BW pre-diet to mdx (Figure 5d). Planned
19
comparisons (Newman-Keuls) revealed that post-HFD mdx/Sln-null mice had significantly (P < 0.05) higher liver % BW in comparison to post-HFD mdx mice, with no differences (P = 0.79) observed when compared to WT mice. Planned comparisons further revealed a significant increase in liver % BW post-HFD in WT mice in comparison to pre-HFD (P< 0.02). An
interaction that was close to statistical significance (P= 0.08) was observed with greater liver % BW in HFD fed mdx/Sln-null mice in comparison to chow controls. No differences were
20
Figure 4. Fat pad mass, brown adipose tissue (BAT) weight, and adiposity index for chow fed control and HFD, wild type (WT), mdx mice,
mdx/Sln-null mice. (A) epididymal fat pad mass; (B) retroperitoneal fat pad mass; (C) BAT mass; (D) adiposity index. All showed a main effect
(P< 0.0001) of diet with HFD > Chow and main effect (P< 0.0001) of genotype with WT > mdx SLNKO > mdx. *Significantly different (P< 0.05) than mdx. #Significantly different from mdx/Sln-null. Values are means±SE (N=11 for WT HFD, N=9 for WT Post-HFD, N=13 for mdx Pre-Chow, N=12 for mdx Post-HFD, N=8 for mdx mdx/Sln-null Pre-HFD, N=8 for mdx/Sln-null Post-HFD).
Epidydmal Fat Pad (mg)P r e - H F D P o s t - H F D05 0 01 0 0 01 5 0 0W Tm d xm d x / S L N - n u ll* #*RetroperitonealFat Pad (mg)P r e - H F D P o s t - H F D01 0 02 0 03 0 04 0 05 0 0W Tm d xm d x / S l n - n u llP = 0 . 0 7* #Brown AdiposeTissue (mg)P r e - H F D P o s t - H F D01 0 02 0 03 0 0W Tm d xm d x / S l n - n u ll* #*Adiposity IndexP r e - H F D P o s t - H F D02468W Tm d xm d x / S l n - n u ll ** #A BC D*
21
Figure 5. Soleus and liver weights for chow fed control and HFD, wild type (WT), mdx, and mdx/Sln-null mice. (A) Soleus mass (main effect of genotype: P
< 0.0001 mdx/ mdx/Sln-null > WT; main effect of diet: P < 0.02, HFD > Chow); (B) liver weight (main effect of diet: P < 0.0001, HFD > Chow); (C) soleus % BW (main effect of genotype: P < 0.004, mdx/ mdx/Sln-null > WT; main effect of diet: P < 0.05); (D) liver % BW (main effect of diet: P<0.01, HFD >
Chow). * Significantly different from mdx (P < 0.05). #Significantly different (P < 0.05) than mdx/Sln-null mice. Values are means±SE (N=11 for WT
Pre-HFD, N=9 for WT Post-HFD, N=13 for mdx Pre-Chow, N=12 for mdx Post-HFD, N=8 for mdx null Pre-HFD, N=8 for mdx/Sln-null Post-HFD).Soleus Mass (mg)P r e - H F D P o s t - H F D051 01 52 0 W Tm d xm d x / S l n - n u ll* #Liver Mass (g)P r e - H F D P o s t - H F D01234 W Tm d xm d x / S l n - n u ll * #Soleus % BWP r e - H F D P o s t - H F D0 . 0 00 . 0 10 . 0 20 . 0 30 . 0 40 . 0 5 W Tm d xm d x / S l n - n u ll* #Liver % BWP r e - H F D P o s t - H F D02468W Tm d xm d x / S l n - n u ll * *A BC D
22 CLAMS Measurements
CLAMS data are presented in Table 1. Pre-HFD relative total daily VO2, relative waking VO2 and relative sleeping VO2, were significantly lower in mdx/Sln-null mice compared to both WT (P< 0.01) and mdx (P< 0.02) mice. No differences (P > 0.5) in these pre-HFD relative VO2measures were observed between mdx and WT mice. However, for all pre-HFD absolute (ml O2/hr) VO2 measures (i.e. total, waking and sleeping), values were lower in WT mice compared with both mdx (P < 0.0001) and mdx/Sln-null (P < 0.01) mice. No differences (P > 0.1) in these pre-HFD absolute VO2 measures were observed between mdx and mdx/Sln-null mice. Planned comparisons revealed that mdx mice had 4% greater pre-HFD absolute daily VO2 compared to mdx/Sln-null mice, but the difference was not significant (P = 0.26). Pre-HFD total and dual cage activity were higher (P < 0.001) in WT mice in comparison to mdx and mdx/Sln-null mice. Both pre-diet total and dual beam activity were not different (P > 0.05) in mdx versus mdx/Sln-null mice.
Following the 8-week HFD, relative total daily, waking and sleeping VO2 were higher (P < 0.05) in mdx mice in comparison to both WT and mdx/Sln-null mice. Furthermore, post-HFD relative total daily, waking and sleeping VO2 were higher in WT compared to mdx/Sln-null mice. A main effect of diet (P < 0.0001) was also observed with all three animal groups having lower relative total daily, waking and sleeping VO2 post-HFD compared with pre-HFD. In contrast to relative VO2, all absolute VO2 values were higher (P < 0.05) post-HFD in comparison to pre-HFD. Similar to relative VO2, all absolute VO2 values were higher (P< 0.0001) in mdx mice in comparison to WT and mdx/Sln-null mice, with the exception of absolute sleeping VO2, which was not different (P = 0.24) between mdx and mdx/Sln-null mice. Food intake (g) also
pre-23
HFD. No significant differences were found in food consumption between the three groups either pre- or post-HFD. The smaller quantity (g) of high fat food eaten by all three groups of mice was calculated to have the same amount of metabolizable energy as the standard rodent chow eaten during pre-HFD CLAMS measurements (3.0 kcal/g for standard chow versus 4.5 kcal/g for the HFD; Table 1). For activity levels there was a main effect (P < 0.01) of genotype, whereby WT mice had greater total activity and dual beam activity than both mdx and mdx/Sln-null mice both pre- and post-HFD. There was no main effect of diet for either total activity (P = 0.67) or dual beam activity (P = 0.79). Finally, a main effect of diet (P < 0.0001) was found for all three-respiratory exchange ratio (RER) values with pre-HFD values being higher than post-HFD values. There was also a main effect of genotype (P < 0.05) for sleeping RER, with WT mice having higher values than both mdx and mdx/Sln-null mice.
24
Table 1. CLAMS Measurements of Wild type (WT), mdx, and mdx/Sln-null mice Pre and Post-HFD
Pre Post WT mdx mdx/Sln-null WT mdx mdx/Sln-null Body Weight (g) 30.6 ± 1.3 34.5 ± 1.1# 34.8 ± 0.77# 44.6 ± 2.4* 44.3 ± 1.7* 45.3 ± 1.1* Waking VO2(ml O2/kg/hr) 3457± 55 3461 ± 55 3250 ± 59#$ 2909 ± 52* 3349 ± 113*# 2805 ± 42*#$ Sleeping VO2(ml O2/kg/hr) 2668 ± 80 2781 ± 68 2630 ± 26#$ 2587± 62* 2733 ± 59*# 2456 ± 46#$ Total Daily VO2 (ml O2/kg/hr) 3203 ± 51 3240 ± 55 3038 ± 70#$ 2838 ± 52* 3135 ± 76*# 2689 ± 37*#$ Food Intake (g) 3.83 ± 0.31 3.97 ± 0.22 4.20 ± 0.25 2.47 ± 0.19* 2.57 ± 0.24* 2.57 ± 0.31* Metabolizable Energy (kcal) 11.49 ± 0.93 11.9 ± 0.66 12.61 ± 0.76 11.13 ± 0.19* 11.55 ± 0.52* 11.56 ± 0.49* Total Activity 10776 ± 718 7624 ± 215# 7072 ± 174# 9852 ± 450 8257 ± 209# 7750 ± 132# Dual Beam Activity 3632 ± 349 2601 ± 238# 2515 ± 198# 3358 ± 198 2728 ± 134# 2471 ± 128#
25 Waking VO2 (ml O2/hr)103 ± 4.7 119 ± 2.8# 113 ± 1.2# 116 ± 6.3* 140 ± 5.1*# 126 ± 2.7*#$ Sleeping VO2 (ml O2/hr) 79 ± 3.8 95 ± 2.0# 91 ± 1.2# 104 ± 6.4* 119 ± 5.1*# 111 ± 3.5* Daily VO2 (ml O2/hr) 95 ± 4.1 111 ± 2.3# 105 ± 1.4# 116 ± 5.8* 137 ± 4.7*# 118 ± 5.4*$ Awake RER 0.921 ± 0.007 0.920 ± 0.009 0.916 ± 0.014 0.815 ± 0.007* 0.806 ± 0.007* 0.794 ± 0.011* Sleep RER 0.908 ± 0.009 0.891 ± 0.007 0.8907 ± 0.005 0.798 ± 0.009* 0.774 ± 0.009* 0.768 ± 0.013* Total RER 0.912 ± 0.010 0.906 ± 0.009 0.906 ± 0.01 0.807 ± 0.008* 0.794 ± 0.009* 0.784 ± 0.01*
Main effect of diet (Pre < Post) for Body Weight (P<.0001). Main effect of diet (Pre > Post) Waking, Sleeping, and Daily VO2 (P<0.0001), Food
Intake (P<0.0001), Total Activity (P<0.05), Dual Beam Activity (P<0.05), and Waking, Sleeping, Total RER (P<0.0001). Main effect of genotype (WT>mdx and mdx/Sln-null) for sleeping RER (P<0.05).*Significantly different from Pre (P<0.05). #Significantly different from WT. $
Significantly different than mdx. VO2, oxygen consumption, RER, respiratory exchange ratio. Values are Mean ± SEM (N=9 for WT, N=12 for
26 Metabolic Efficiency
Metabolic efficiency (M.E) was calculated by dividing the change in body weight by the amount of energy derived from either chow or high fat diet converted to kJ. An interaction (P < 0.0001) was found, with WT mice having significantly greater metabolic efficiency post-HFD than mdx and mdx/Sln-null mice (Figure 6). Planned comparisons revealed a trend (P = 0.07) towards higher metabolic efficiency in mdx/Sln-null mice in comparison to mdx mice post-HFD (Figure 8). There were no differences in metabolic efficiency pre-HFD.
Figure 6. Metabolic efficiency (M.E) pre- and post- HFD. M.E. increases in response to high fat diet (main effect of diet: P < 0.0001). Trend towards higher metabolic efficiency in mdx/Sln-null vs mdx mice (P=0.07). *Significantly greater than mdx mice (P < 0.0001). # Significantly greater than mdx/Sln-null mice (P < 0.0001). Values are mean ± S.E. (N=8 for WT, N=12 for mdx, N=8 for mdx SLNKO)
Metabolic Efficiency (g/kJ)P r e - H F D P o s t - H F D02468
W T
m d x
m d x / S ln -n u ll
P = 0 .0 7
* #
27 Glucose tolerance tests
GTT were performed on WT, mdx, and mdx/Sln-null mice pre- and post-HFD (Figure 5a). There were large differences between pre- and post-HFD responses with the glucose values being lower (P < 0.001) at all time points during pre-diet tests compared with post-diet tests in WT (Fig. 7a) and at two time points in mdx/Sln-null mice (Fig. 7b). There were no differences observed between pre- and post-diet responses with mdx mice (Fig. 7a). A trend towards an interaction (P = 0.06) was observed showing greater glucose intolerance (higher blood glucose levels) post-HFD in mdx/Sln-null mice compared with WT (Fig. 7b) and mdx mice (Fig. 7c). Planned comparisons (Neuman-Keuls) between mdx and mdx/Sln-null at the 60 and 120 min time points showed that mdx/Sln-null were more (P < 0.05) glucose intolerant post-HFD in comparison to mdx mice. A comparison of pre-diet values revealed no significant differences between mdx and mdx/Sln-null mice.
Whole glucose excursions after an intraperitoneal injection of glucose are a good
representation of glucose tolerance. Another index of glucose tolerance is the glucose area under the curve (AUC), which represents the total amount of blood glucose throughout the two-hour glucose tolerance test, where greater values will represent poorer glucose tolerance. When expressed as an AUC, an interaction was observed (P< 0.02) with mdx/Sln-null mice showing greater blood glucose post-HFD in comparison to mdx and WT mice (Fig. 7d). Planned comparisons (Neuman-Keuls) showed that post-HFD blood glucose levels were greater (P < 0.05) in mdx/Sln-null mice in comparison to mdx mice; WT mice tended (P = 0.08) to have lower AUC values post-HFD in comparison to mdx/Sln-null mice.
28
Figure 7. Glucose tolerance curves and area under the curve (AUC) pre and post 8-week HFD. (A) Wild type (WT) vs. mdx, (B) WT vs. mdx/Sln-null, (C)
mdx vs mdx/Sln-null, (D) Glucose AUC. Main effect of genotype (P < 0.0001), WT < mdx/ mdx/Sln-null. Main effect of time (P < 0.0001), 0 < 120 < 30
< 60. *Significantly different from Pre (P < 0.05). # Significantly different from WT (P < 0.05). $ Significantly different from mdx (P < 0.05). Values are Mean ± SEM (N=11 for WT, N=12 for mdx, N=8 for mdx/Sln-null)
T i m e ( m i n )Blood Glucose (mM)0 3 0 6 0 9 0 1 2 05 . 07 . 51 0 . 01 2 . 51 5 . 01 7 . 52 0 . 02 2 . 52 5 . 0W T P o s t H F Dm d x P o s t H F Dm d x P re H F DW T P re H F D*** ##T i m e ( m i n )Blood Glucose (mM)0 3 0 6 0 9 0 1 2 05 . 07 . 51 0 . 01 2 . 51 5 . 01 7 . 52 0 . 02 2 . 52 5 . 0W T P re H F DW T P o s t H F Dm d x S L N K O P r e H F Dm d x S L N K O P o s t H F D#**##***T i m eBlood Glucose (mM)0 3 0 6 0 9 0 1 2 05 . 07 . 51 0 . 01 2 . 51 5 . 01 7 . 52 0 . 02 2 . 52 5 . 0 m d x P re H F Dm d x p o s t H F Dm d x S L N K O P r e H F Dm d x S L N K O P o s t H F D$** $AUCP r e P o s t05 0 01 0 0 01 5 0 02 0 0 02 5 0 0 W Tm d xm d x / S l n - n u l lP = 0 . 0 8$A BC D
29 Western Blotting Analysis
Comparisons between chow fed WT, mdx mice, and mdx/Sln-null revealed that SLN is elevated in the mdx diaphragm (P < 0.0001) by a factor of 2, SLN was not detected in the mdx/Sln-null samples (Figure 8). Due to “technical issues”, SLN expression could not be accurately detected in the soleus. Comparisons between HFD fed WT and mdx mice shows that SLN remains elevated in the mdx diaphragm (P < 0.02) by a factor of 1.5. Interestingly, SLN levels decreased (P < 0.05) in the diaphragms of mdx mice post-HFD. UCP-1 was elevated (P < 0.02; Figure 9) post-HFD across all three genotypes, WT (30%), mdx (29%), and mdx/Sln-null (25%). However, UCP-1 content was not different (P > 0.05; Figure 9) between WT, mdx, and mdx/Sln-null post-HFD.
30
Figure 9. UCP-1 expression in brown adipose tissue (BAT). Post-HFD, UCP-1 expression increased
in WT (30%), mdx (29%), and mdx/Sln-null (25%). No significant differences between all three genotypes. *Significantly different from Chow (P < 0.05). Values are mean ± S.E (N=8 for WT Pre-HFD, N =8 for WT Post-Pre-HFD, N=8 for mdx Pre-Pre-HFD, N=8 for mdx Post-Pre-HFD, N=8 for mdx/Sln-null Pre-HFD, N=8 for mdx/Sln-null Post-HFD).
Arbritrary UnitsWT mdx0 .00 .51 .01 .52 .0W Tm d x**W T m d xP r e - H F D P o s t - H F DS L N5 k D aP o n c e a u SP r e - H F D P o s t - H F D
Figure 8. SLN expression in diaphragm (DIA). SLN is higher in mdx diaphragm by a factor of 2.
Post-HFD the SLN expression remained higher in mdx diaphragm in comparison to WT mice by factor of 1.75. *Significantly different than WT (P < 0.05). Values are mean ± S.E (N=9 for WT Pre-HFD, N=8 for WT Post-Pre-HFD, N=9 for mdx Pre-Pre-HFD, N=9 for mdx Post-HFD).
Arbritrary UnitsWT mdxmdx/Sln-null0 .00 .10 .20 .30 .4 W Tm d xm d x /S ln - n u ll* * *W T m d x m d x /S ln - n u llC h o w H F D C h o w H F D C h o w H F DU C P - 13 3 k D aP o n c e a u S
31
Discussion
This study set out to assess the role of SLN in providing protection against high fat diet induced obesity in mdx mice. The main findings of this study were: 1) SLN content in the diaphragm was higher in mdx mice compared to WT mice; 2) mdx/Sln-null mice had statistically significantly greater adiposity post-HFD, despite a non-statistical significant increase in body weight in comparison to mdx mice; 3) in accordance with the ablation of SLN, mdx/Sln-null mice demonstrated a lower whole body metabolic rate in comparison to mdx mice. Altogether these results further support SLN’s thermogenic role, wherein the absence of SLN is a major mechanism of diet induced obesity in mdx/Sln-null mice.
A previous study saw an upregulation of SLN in the diaphragm and soleus of mdx mice (Schneider et al., 2013). As hypothesized, there was a significant difference in the expression of SLN between WT and mdx mice, with a 2-fold higher expression in the diaphragm of mdx mice vs WT mice (Fig. 8). In relation to the Schneider study, a similar difference was observed with mdx diaphragm expressing a 3-fold increase in comparison to WT mice (Schneider et al., 2013). Schneider et al. (2013) also observed a 10-fold difference in the expression of SLN in the soleus muscle of mdx mice in comparison to WT mice. This finding could not be corroborated during this study due to the lack of viable results that could be systematically discerned due to technical issues. However, the finding of higher expressions of SLN in the diaphragm of mdx mice helps corroborate the finding that SLN is upregulated in mdx mice.
A novel finding was that after consuming a high fat diet, SLN in the diaphragm declined in mdx mice by 30% (Fig. 8). This decrease in SLN expression may seem paradoxical as SLN has been found to increase in soleus muscles of high fat fed animals (Bombardier, 2010). However, the thermogenic contribution of diaphragm to skeletal muscle non-shivering
32
thermogenesis is at this time unknown and the expression of SLN in diaphragm post-HFD has not been shown to change (Bal et al., 2012). This was corroborated in this study as SLN expression in the diaphragm of WT mice did not change after an 8-week HFD (Fig. 8). The upregulation in SLN seen in the muscles of mdx has been proposed as a contributing factor to it’s pathology (Schneider et al., 2013). However, SLN upregulation has also been proposed as a compensatory mechanism to combat mdx pathology by upregulating utrophin, a dystrophin hom*ologue, through increased NFAT translocation to the muscle nucleus via calcineurin activation (Fajardo, 2015). A previous study examining the quadriceps of mdx mice post-HFD have found that those muscles have significantly reduced myofibre necrosis when measured as a % of cross sectional area (% CSA), almost half the amount observed in chow fed mdx mice (Radley-Crabb et al., 2011). Furthermore, voluntarily exercised mdx mice on a high fat diet were able to run 50% further than their chow counterparts (Radley-Crabb et al., 2014).
CLAMs data from this study revealed total and dual beam activity was always lower in mdx mice and mdx/Sln-null mice in comparison to WT mice with mdx mice having greater movement than mdx/Sln-null mice (Table 1). Despite no statistical significance, post-HFD mdx mice saw an 8.3% increase in cage activity; there was also an 9.5% increase in cage activity observed in mdx/Sln-null mice. Two possible explanations may account for the increase in activity in both mdx strains, (1) the high fat diet may be better suited to meeting the energy demands of the mdx and mdx/Sln-null mice and/or (2) the high fat diet may have altered the phospholipid membrane composition and reduced calcium leak from the SR/ER. Diets that are high in saturated fat and cholesterol can change the composition of the phospholipid membrane of skeletal muscles, particularly the incorporation of a higher percentage of saturated fatty acids in the muscle phospholipid membrane (Janovska et al., 2010). Alterations in the form of
33
increased saturated fatty acids and cholesterol in the phospholipid membrane of the SR have been found to reduce membrane fluidity thus reduce the rate of Ca2+ leak from the SR/ER (Vangheluwe et al., 2005). This reduction in Ca2+ could help reduce the dystropathology of mdx mice and improve muscle function. This could also help explain the decline of SLN in the diaphragm of HFD fed mdx mice. Calcium dysregulation in the muscle is one of the chief
symptoms of duch*enne muscular dystrophy which contributes to its pathology (Franco-Obregon & Lansman, 1994; Turner, et al., 1991; Whitehead et al., 2006). The findings of the present study would suggest that the high fat diet may cause a compensatory decrease of SLN in mdx mice that could alleviate muscle necrosis and improve function. Furthermore, SLN expression remained upregulated despite a 30% decrease in comparison to pre-diet mdx mice. This suggests that despite the decline in SLN expression in response to high fat feeding, the already upregulated expression of SLN in mdx mice is large enough to maintain a significant difference in
comparison to WT mice thereby maintaining SLN mediated thermogenesis in mdx mice. A future study may be needed to examine if the same decrease in expression occurs in mdx soleus due to it’s role in adaptive thermogenesis. These findings pose interesting future questions for high fat diet’s therapeutic role in DMD treatment.
The ablation of sarcolipin results in a more efficient pumping of calcium into the SR (Bal et al., 2012), thus it was hypothesized that an overexpression of SLN would result in higher levels of SERCA uncoupling, increased energy expenditure, and protection against obesity (Maurya et al., 2015). After consuming an 8-week high fat diet, mice overexpressing SLN (SLNOE) were found to have lost weight and maintain a normal metabolic profile in comparison to WT mice and KO mice despite consuming significantly more calories than WT mice (Maurya et al., 2015). For this study, it was hypothesized that the upregulated expression of SLN in mdx
34
mice would enhance adaptive thermogenesis and increase protection against diet-induced obesity and glucose intolerance. The body weights of WT and mdx mice were different pre-HFD with mdx mice (33.8 ± 0.58g) being heavier than WT mice (29.9 ± 0.73g). The significant differences in body weight pre-HFD can be attributed to pseudohypertrophy, a compensatory muscle
hypertrophy due to muscle necrosis and replacement by connective tissue and fat in the muscles of mdx mice (Mokhatarian et al., 1996; Connolly et al., 2001; Coulton et al., 1988; Anderson et al, 1987). In contrast, post-HFD body weights of WT and mdx mice were not significantly different (43.1 ± 1.98g for WT vs 43.7 ± 1.98g for mdx) in accordance to previous findings which also observed no differences in final body weights (Radley-Crabb et al., 2011). The most likely explanation for a lack of a difference in bodyweight between WT and mdx mice after an 8-week high fat diet is the increase in body fat in the WT mice (Figure 4d).
Despite the lack of differences in final body weight post-HFD, as hypothesized WT mice experienced a 31% increase in body weight over the 8-week period. This demonstrates that the WT mice experienced greater weight gain (13.1 ± 1.15g) than mdx (9.0 ± 1.0g) or mdx/Sln-null (9.59 ± 0.92g) mice from week 2 of the high fat diet until its eventual completion (Fig. 2a). The weight gain observed in WT mice was in accordance with other studies examining weight gain over an 8-week high fat diet (Li et al., 2000; Collins et al., 2004; Bombardier, 2010). However, the weight gains observed in mdx mice were higher post HFD (171%) than the previously reported 137% increase (Radley-Crabb et al., 2014). One potential explanation could be a difference in the composition of the diets. The present study used a Western diet (TD 88137; Harlan Teklad, Madison, WI; 21.2% fat by weight; 42% of kcal from fat) whereas Radley-Crabb (2011) used a different high fat diet (SF06-040, Specialty Feeds, Glenn Forest, WA;16% fat by weight and 32% of kcal from fat) (Radley-Crabb, 2014). Furthermore, the diets varied in their
35
fatty acid profiles, with the Westernized high diet having 61.8% comprised of saturated fatty acids, whereas the high fat diet only had 3.26% composition of saturated fatty acids (Radley-Crabb et al., 2011). Higher fat composition may have exacerbated weight gain in the mdx mice, thereby resulting in greater weight gain. This has been demonstrated across a variety of species and with various diets, with specific diets causing some species to gain more weight in
comparison to other diets (Buettner et al., 2007). These findings further support that the composition of diet plays an important role in the manifestation of obesity.
The ablation of sarcolipin has been shown to increase the efficiency of the SERCA pumps, predisposing SLN ablated mice to obesity after consuming a high fat diet (Bombardier, 2010). In this study, it was hypothesized that the ablation of SLN in mdx mice would have similar effects to mdx/Sln-null mice, namely that they would be more susceptible to diet induced obesity and glucose intolerance when compared to mdx mice. Despite no statistical difference in weight gain between mdx and mdx/Sln-null mice, food consumption was significantly lower in mdx/Sln-null mice in comparison to mdx mice (Fig. 3). Thus, the mdx/Sln-null mice gained the same amount of weight as mdx mice despite consuming less of the high fat diet. The ablation of SLN blunts adaptive thermogenesis, resulting in higher metabolic efficiency, wherein less of the energy derived from ATP is dissipated as heat but rather stored and used as meaningful work. Higher metabolic efficiency makes it less likely that excess energy consumed from diet will be expended through thermogenic means, but instead stored as adipose tissue. Previous studies have shown that mice with increased metabolic efficiency, either through weight loss or the ablation of thermogenic proteins such as UCP-1, are more likely to gain weight and exhibit greater obesity (Maclean et al., 2004; Feldmann et al., 2007). Furthermore, Sln-null mice have exhibited enhanced metabolic efficiency, which was reflected in greater weight gain and adiposity in
