AWARD NUMBER : W 81 XWH - 14 - 2 - 0133 TITLE : Regulation of Heat Stress by HSF 1 and GR PRINCIPAL INVESTIGATOR

The regulation of mitochondrial morphology is closely coupled to cell survival during stress. We examined changes in the mitochondrial morphology of mouse C2C12 skeletal muscle cells in response to heat acclimation and heat shock exposure. Acclimated cells showed a greater survival rate during heat shock exposure than non-acclimated cells, and were characterized by long interconnected mitochondria and reduced expression of dynaminrelated protein 1 (Drp1) for their mitochondrial fractions. Exposure of C2C12 muscle cells to heat shock led to apoptotic death featuring activation of caspase 3/7, release of cytochrome c and loss of cell membrane integrity. Heat shock also caused excessive mitochondrial fragmentation, loss of mitochondrial membrane potential, and production of reactive oxygen species in C2C12 cells. Western blot and immunofluorescence image analysis revealed translocation of Drp1 to mitochondria from the cytosol in C2C12 cells exposed to heat shock. Mitochondrial division inhibitor 1 or Drp1 gene silencer reduced mitochondrial fragmentation and increased cell viability during exposure to heat shock. These results suggest that Drp1-dependent mitochondrial fission may regulate susceptibility to heatinduced apoptosis in muscle cells and that Drp1 may serve as a target for the prevention of heat-related injury. Abbreviations. AR, aspect ratio; Drp1, dynamin-related protein 1; DHE, dihydroethidium; FF, Form factor; GFP, green fluorescence protein; HA, heat acclimation; HSF1, heat shock transcription factor 1; HSPs, heat shock proteins; IMF, intermyofibrillar; Mfn1, mitofusin 1; Mfn2, mitofusin 2; Mdivi-1, mitochondrial division inhibitor 1; OPA1, optic atrophy 1; ROS, reactive oxygen species; SS, subsarcolemmal; TEM, transmission electron microscope; TMRE, tetramethylrhodamine ethyl ester; Introduction Skeletal muscle, which makes up ~40-50% of the body mass in mammals, generates significant amounts of heat during contraction (Block, 1994). Muscle temperature can exceed 40 °C during exercise (Taylor et al., 1998; Drust et al., 2005). Understanding the regulation of muscle function under high temperatures may help develop strategies for improving performance and preventing injury. Exposure to a temperature above the physiological range can produce both detrimental and beneficial effects on muscle health. In general, acute exposure to very high temperatures can cause apoptotic damage in muscle cells (Islam et al., 2013) and muscle injury in animals (Abdelnasir et al., 2014), whereas frequent exposures to moderately high temperatures or heat acclimation (HA) may protect muscle cells against a subsequent severe heat insult (Monastyrskaya et al., 2003; Liu & Brooks, 2012). In fact, HA has also been tested for cross-tolerance or protection against other pathological conditions (Horowitz et al., 2015). Interestingly, HA has been shown to protect against obesity-induced insulin resistance by increasing heat shock protein 72 in skeletal muscles of mice (Chung et al., 2008; Gupte et al., 2009). How muscle cells paradoxically respond to mild versus severe heat stress remains poorly understood. The majority of studies of the adaptation and resistance of muscle cells to heat have focused on mechanisms involving heat shock transcription factor 1 (HSF1) and heat shock proteins (HSPs) (Tetievsky et al., 2008; Abdelnasir et al., 2014). Emerging evidence suggests that cellular organelles, including mitochondria, endoplasmic reticulum, lysosomes and the Golgi apparatus, serve key roles in stress/damage-sensing and apoptosis signaling. How they respond to the disruption of cellular homeostasis by high temperatures may ultimately determine the fate of stressed cells (Ferri & Kroemer, 2001). Mitochondria regulate cell survival and coordinate cell-wide stress responses by constantly changing their morphology through fusion and fission in response to energy demands and environmental stimuli (Hoppins & Nunnari, 2012; Youle & van der Bliek, 2012). Disruption in these processes causes mitochondrial dysfunction, and leads to the pathogenesis of various acute and chronic diseases (Archer, 2013). High temperatures indeed affect mitochondrial function, and impair mitochondrial electron transport; this condition also induces the production of reactive oxygen species (ROS) under in vitro (Wang et al., 2013) and in vivo (Qian et al., 2004) conditions. Altered mitochondrial morphology has also been observed in heat shock-exposed cultured mouse embryonic fibroblasts (Sanjuan Szklarz & Scorrano, 2012). How changes in mitochondrial morphology might affect resistance of cultured muscle cells, in particular C2C12 myoblasts, to heat is largely unexplored. In mammals, mitochondrial morphology is coordinately regulated by several GTPases, dynamin-related protein 1 (Drp1) for fission and mitofusin isoforms (Mfn1 and 2), and optic atrophy1 (OPA1) for fusion, respectively (Ishihara et al., 2006). The overall morphology and shape of mitochondria are determined by the balance between mitochondrial fission and fusion activities (Hoppins & Nunnari, 2012; Youle & van der Bliek, 2012; Archer, 2013). The role of mitochondrial dynamics proteins in responses to HA and heat stress has yet to be determined. In this study, we investigated changes in mitochondrial morphology and dynamics of C2C12 mouse skeletal muscle cells exposed to HA or lethal heat shock. We discovered that HAinduced resistance of these cells against heat injury is associated with mitochondrial elongation mediated by down-regulation of the fission protein, Drp1. We demonstrate that exposing C2C12 cells to heat shock causes cell damage characterized by excessive mitochondrial fragmentation with increased recruitment of Drp1 to mitochondria. We further show that inhibition of mitochondrial fission by mitochondrial division inhibitor (Drp1 translocation blocker) Mdivi-1 or by Drp1 gene silencer protects mitochondrial morphology and prevents cell injury during exposure to heat shock. Together our work reveals a novel mechanism regulating resistance of C2C12 cells to heat stress, which may have relevance to heat injury prevention. Methods Cell culture, myogenic differentiation, heat acclimation and heat shock The mouse skeletal myoblast cell line C2C12 was recently purchased from ATCC (ATCC® CRL-1772TM) and was maintained at 37 °C in DMEM containing 10% fetal bovine serum, 100-units/ml penicillin and 100-μg/ml streptomycin. Myogenic differentiation was initiated upon reaching ~80% confluence by switching the cells to DMEM containing 2% horse serum supplemented with 1μM insulin. For heat acclimation, cells were placed in an incubator preset at 39.5 °C for 3 hours per day for 3 days. Control cells and HA cells (when not being incubated at 39.5 °C) were maintained at 37 °C for 3 days. After the 3-day incubation, cells were tested for cell viability during exposure to heat shock. For heat shock, cells were maintained in an incubator preset at 43 °C for indicated times. Immediately after heat shock, cells were harvested for subsequent assays. Mouse heat acclimation protocol Mice (6-8 weeks old male C57BL/6J) were purchased from Jackson Laboratories (Bar Harbor, ME) and maintained in a temperature-controlled (21 °C) animal facilities at the Uniformed Services University (USU), with 12-hour light/dark cycle and free access to food & water. After arrival, all the mice were given one week to recover. Heat acclimation was initiated by exposing the mice to 33 °C in an environmental chamber from 8-11 AM (3 hours/day) for consecutive 10 days. Control mice were maintained at room temperature (21 °C) all the time. All procedures were approved by the USU Institutional Animal Care and Use Committee. Transmission electron microscope (TEM) analysis of mitochondria in mouse skeletal muscle Immediately after heat or sham exposure, mice were perfusion fixed under isoflurane anesthesia. The gastrocnemius muscle was removed and incubated in fixative (5% formaldehyde, 2% glutaraldehyde in 0.1M PBS, pH7.4) overnight. The samples were then sent to the Biomedical Instrument Center at USU for a standard TEM preparation procedure. The gastrocnemius muscle comprises a mixture of type I (mitochondria-rich) and type II (mitochondria-poor) muscle fibers (Mishra et al., 2015). For comparison, only mitochondria enriched fibers were selected and areas presenting all the classical structures of sarcomere were imaged. For each animal, six to eight muscle fibers were analyzed, a total of 200 subsarcolemmal (SS) and 200 intermyofibrillar (IMF) mitochondria were individually traced by using NIH developed ImageJ software quantify the following morphological and shape descriptors (Picard et al., 2013): surface area, perimeter, circularity (4π·(surface area/perimeter2)) and Feret's diameter (longest distance between any two points within a given mitochondrion). Form factor (FF), a measure sensitive to the complexity and branching aspect of mitochondria, was calculated as perimeter/(4π·surface area) and aspect ratio (AR), a measure of the “length to width ratio”, was calculated as major axis/minor axis. Cell viability and cell death assays Cell viability was determined by trypan blue exclusion test with Bio-Rad TC20 automated cell counter per manufacturer's instruction. Caspase activity was measured by CellEventTM Caspase-3/7 Green detection reagent (Molecular Probes), and dead cells were detected by using an Annexin V Alexa Fluor® 488 apoptosis kit (Molecular Probes). Mitochondrial morphology analysis and membrane potential and ROS measurement Mitochondria were labeled with CellLight Mitochondria-GFP (Molecular Probes) or visualized bysCMXRos (Invitrogen) staining. Quantitative analyses of mitochondrial morphology and Drp1 subcellular distribution were performed using ImageJ software (NIH Image) (Yu et al., 2014). Mitochondrial membrane potential was evaluated with Tetramethylrhodamine ethyl ester (TMRE, Molecular Probes) and ROS levels were detected by using dihydroethidium, a fluorescent probe (DHE; Invitrogen), as described previously (Yu et al., 2014). Fluorescence microscopy Fluorescence images were viewed and acquired with a Leica AF6000 epifluorescence microscope, which was equipped with a digital microscope camera. Excitation/emission wavelengths were 358/461 nm for blue fluorescence (4',6-diamidino-2-phenylindole, DAPI), 480/535 nm for green fluorescence (GFP, annexin V, Alexa 488, MitoTracker Green FM and caspase-3/7 Green), and 555/613 nm for red fluorescence (MitoTracker Red, TMRE, ethidium, and Alexa 594). Mitochondria isolation Mitochondria were isolated as described previously (Yu et al., 2014). Briefly, the cells were suspended in cold isolation buffer (10 mM Hepes pH 7.2, 1 mM EDTA, 320 mM sucrose) containing protease inhibitor and homogenized in a Dounce homogenizer. The homogenate was centrifuged at 700 × g for 8 min. The first supernatant was saved, and the pellet was homogenized and centrifuged again. The two supernatants were pooled and centrifuged together at 17,000 × g for 15 min to obtain the mitochondrial pellet. Western blotting and immunofluorescence Western blotting was performed with 1:1000 of the following primary antibodies (Yu et al., 2014): mouse anti-Drp1 (BD Biosciences), rabbit antiphospho-Drp1-Ser616 (Cell Signaling Technology), rabbit anti-Mfn1 (Santa Cruz Biotech), rabbit anti-Mfn2 (Cell Signaling Technology), mouse anti-OPA1 (BD Biosciences), rabbit anti-VDAC (Cell Signaling Technology), mouse anti-cytochrome c (Santa Cruz Biotech) and mouse anti-actin (Santa Cruz Biotech). Horseradish peroxidase-conjugated anti-rabbit and anti-mouse antibodies were used as secondary antibodies. Indirect immunofluorescence was performed as described previously (Yu et al., 2014). Briefly, cells cultured on coverslip were fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X-100. Mouse anti-Drp1 (BD Biosciences) antibody was used for primary antibody and Alexa 594-conjugated antibodies (Invitrogen) were used for the secondary antibody. Fluorescence images were acquired and adjusted by using ImageJ software (NIH Image). Statistical analysis Data are expressed as mean ± standard deviations. Statistical significance was established by t-test, or oneor two-way ANOVA followed by post test for comparisons. Comparisons of mitochondrial morphological parameters were carried out using nonparametric Mann– Whitney test. Results Heat acclimation improved cell viability during heat shock exposure and modified mitochondrial morphology in C2C12 myoblasts Following treatment with HA, C2C12 myoblasts showed significantly higher survival rates during exposure to heat shock compared to control cells (Figure 1A). Morphological examination revealed that HA-treated myoblasts contained overly elongated mitochondrial networks, whereas control cells maintained predominantly tubular mitochondria (Figure 1B). Furthermore, expression of mitochondrial fission protein Drp1 was lower in the mitochondrial fractions from HA-treated cells compared to control cells (Figure 1C). We also performed western blot of VDAC and actin in whole cell lysate and found no differences in the ratio of VDAV to actin between control and HA cells (1.13 ± 0.05 versus 1.16 ± 0.04, p > 0.05). No changes in mitochondrial fusion proteins OPA1, and Mfn 1 and 2 were found in HA-treated cells (Figure 1D). Mitochondrial morphology and dynamics were further examined in the gastrocnemius muscles of control and HA-exposed mice. Transmission electron microscope (TEM) images revealed larger SS and IMF mitochondria in the muscles of HA-exposed mice, compared to control mice (Figure 2A). Further quantitative analysis of major mitochondrial shape parameters indicated an increase in size of SS and IMF mitochondria in the muscles of HA-exposed mice (Table 1 and Figure 2B). Finally, expression of Drp1 was lower in the mitochondrial fractions of the gastrocnemius muscles from HA-exposed mice compared to control mice (Figure 2C). Exposure to heat shock caused apoptotic cell death The survival rates of both C2C12 myoblast and myotubes decreased with duration of heat shock exposure in a similar manner: < 70% survived after exposure to heat shock for 4 hour (Figure 3A). We examined mitochondrial events, known to occur during programed cell death (apoptosis), in C2C12 myoblast following exposure to heat shock. Immunofluorescence staining and western blotting showed that cytochrome c was located inside mitochondria under normal incubation, but was released from mitochondria into the cytosol of C2C12 myoblasts during heat shock exposure (Figure 3B). Cell injury was further assessed by using Caspase 3/7 Green and Annexin V Apoptosis Detection kits. Heat shockexposed caused C2C12 myoblasts became Caspase 3/7 (Figure 3C) and Annexin V positive (Figure 3D). Exposure to heat shock caused mitochondrial fragmentation in C2C12 myoblasts We examined mitochondrial morphology in C2C12 cells after exposure to heat shock. Mitochondria in C2C12 myotubes were small and densely packed, and very difficult to discern. C2C12 myoblasts contained more conventional tubular mitochondria under our experimental conditions and thus were used for mitochondrial morphology analysis. Mitochondria form filamentous and often interconnected networks under normal incubation at 37 °C, but the mitochondrial networks became mostly small and punctate units in C2C12 myoblasts after exposure to heat shock (Figure 4A). We conducted time course experiments and found that mitochondria rapidly became fragmented when exposed to heat shock (Figure 4B): the number of cells containing fragmented mitochondria increased from ~10% at 0 min to ~ 47% at 15 min and reached a plateau of ~ 91% at 30 min and thereafter. Exposure to heat shock caused activation of mitochondrial fission in C2C12 myoblasts Mitochondrial structural dynamics are regulated by the fusion and fission of the organelles with an increase in fission activity, which results in mitochondrial fragmentation. Therefore, we assessed mitochondrial fission protein Drp1. Whole cell lysate immunoblotting showed no significant changes in total Drp1 after exposure to heat shock (Figure 5A). We further examined cytosolic and mitochondrial Drp1 levels in C2C12 cells after incubation at 43 °C for 0, 15, 30 and 60 min (Figure 5A). We found a time-dependent increase in the mitochondrial fractions of Drp1 and concurrent decrease in its cytosolic fractions. To verify heat-induced mitochondrial translocation of Drp1, we performed immunofluorescence image and intensity analysis of Drp1 and mitochondria in C2C12 cells (Figure 5B). Under normal incubation, Drp1 predominantly resided throughout the cytosol, with limited association with the tubular mitochondria. After exposure to heat shock, cytosol Drp1 decreased, whereas the majority of Drp1 puncta were co-localized with the fragmented mitochondria. Western blot analysis also showed significantly increased phosphorylation of Drp1 at serine-616 in heat shock-exposed C1C12 myoblasts (Figure 5C). Inhibition of mitochondrial fission protected cell viability and mitochondrial structural integrity against heat shock To determine whether resistance of cells to heat injury is mediated by mitochondrial dynamics, we tested the effects of inhibiting mitochondrial fission on cell viability by Mdivi1 and Drp1 shRNA, which inhibits Drp1 assembly and GTPase activity (Cassidy-Stone et al., 2008) and suppresses the transcription of Drp1 gene in infected cells, respectively. Pretreatment with Mdivi-1 or Drp1 shRNA significantly improved the survival rates of C2C12 myoblast during exposure to heat shock and significantly reduced the number of Annexin V positive apoptotic cells (Figure 6A). We found that Drp1 shRNA also prevented cytochrome c release and caspase activation in C2C12 myoblast during exposure to heat shock (Figure 6B). Mitochondrial morphology analysis revealed a significant reduction in the number of fragmented mitochondria in C2C12 myoblast pretreated with Mdivi-1 or infected with Drp1 shRNA following heat shock exposure (Figure 6C). Only ~5.3% of Mdivi-1-treated cells contained fragmented mitochondria compared to ~82.5% of cells pretreated with vehicle. A similar reduction in mitochondrial fragmentation (> 93% cells maintained tubular mitochondrial networks) was found in cells infected with Drp1 shRNA. Western analysis showed that expression of Drp1 protein was reduced by ~80% in Drp1 shRNA-treated cells, compared to cells infected with vehicle control, Lentiviral Particle carrying copGFP (Santa Cruz) (Figure 6D). Inhibition of mitochondrial fission reduced loss of mitochondrial membrane potential and production of ROS in C2C12 myoblasts caused by heat shock We measured mitochondrial membrane potential magnitude in C2C12 myoblasts labeled with MitoTracker Green and TMRE. Exposure to heat shock caused a significant decline of mitochondrial TMRE, not MitoTracker Green, fluorescence in C2C12 myoblasts (Figure 7A). Time-lapse experiments further revealed that the decrease in mitochondrial TMRE fluorescence occurred as early as ~ 20-25 min (Figure 7B). Pretreatment with Mdivi-1 or Drp1 shRNA prevented heat shock-induced loss of mitochondrial membrane potential in C2C12 cells (Figure 7C). Exposing C2C12 myoblasts to heat shock caused a time-dependent increase in DHE fluorescence, an indicator of ROS (Figure 7D). The fluorescence signal increased from baseline to ~2 fold at 1 hour and ~5 fold at 4 hours of exposure to heat shock. Pretreatment with Mdivi-1 or Drp1 shRNA prevented an increase in ROS in C2C12 cells induced by heat shock. Exposure to heat shock caused OPA1 cleavage and loss of mitochondrial membrane potential in C2C12 myoblasts in a time-dependent manner We examined mitochondrial fusion proteins in C2C12 myoblasts exposed heat shock. No significant changes in Mfn1 and Mfn2 levels were found in C2C12 cells following up to 60 minutes of heat shock exposure (Figure 8A). Western blot analysis of whole cell lysates with an OPA1 antibody (BD Biosciences) revealed at least one long and one short form of OPA1 in C2C12 cells before exposure to heat shock (Figure 8B). No changes in OPA1 levels were detected in cells 15 minutes into heat exposure. The long form of OPA1 significantly decreased and a new short form band appeared, which indicates cleavage of high molecular weight forms of OPA1 at 30 min into heat exposure. Discussion Fusion-fission dynamics adapt the morphology of the mitochondrial compartment to various metabolic needs of cells and enable them to regain homeostasis and survive under many stress conditions. The present study showed that HA and heat shock affects mitochondrial morphology in C2C12 cells primarily by reducing and increasing recruitment of the fission mediator Drp1 to mitochondria. Our results demonstrate key mitochondrial events involved in the HA and heat shock responses and provide evidence supporting a new mechanism Drp1-mediated mitochondrial fission underlying the susceptibility of muscle cells to heatinduced injury. Mitochondria typically form a reticular network radiating from the nucleus to create an interconnected system for ensuring essential energy supplies to the cell; they are able to adapt to environmental challenges by remodeling their morphology. The formation of elongated mitochondrial networks is a conserved mechanism engaged during chronic cellular stress (e.g., nutrient deprivation) as a means to maintain ATP production, avoid mitophagy, and improve tolerance of mtDNA mutations by diluting damaged mitochondrial contents across the mitochondrial network (Tondera et al., 2009; Rambold et al., 2011; Mishra et al., 2015; Senyilmaz et al., 2015). On the other hand, fragmented mitochondria are associated with various pathological conditions (Yoon et al., 2011; Galluzzi et al., 2012; Hoppins & Nunnari, 2012; Youle & van der Bliek, 2012; Archer, 2013). In the present study, we expanded previous findings by showing that HA-exposed mouse skeletal muscle tissue and cells shared a common feature of elongated mitochondria: HA resulted in increased resistance against heat shock-induced injury in muscle cells. Mitochondrial networks are maintained through the complex coordination of fission and fusion regulated by a number of key mitochondrial morphology proteins. Our results indicate that HA-induced mitochondrial morphological remodeling is due to an imbalance favoring fusion over fission, which is caused by downregulation of Drp1. The mechanisms behind these changes remain unclear. We are not aware of any studies linking HA to cellular energy stress, but HA may likely have an indirect effect on mitochondrial morphology. For instance, HA is known to cause an adaptation response with up-regulation of HSF1 and HSP70s (Tetievsky et al., 2008). Expression of HSP72 is critical in promoting mitochondrial fusion in skeletal muscle (Drew et al., 2014). A single bout of one-hour mild heat exposure (40 °C) is reportedly sufficient to cause up-regulation of HSP72 in C2C12 cells (Liu & Brooks, 2012). In the present study, HA-treated cells likely had upregulated HSP72, which helped maintain mitochondrial homeostasis and morphology during heat shock exposure. Although fused or elongated mitochondrial networks help cells survive starvation, it is unknown whether this morphological adaptation in mitochondria is associated with resistance of cells against heat insults. The present study suggests that elongation of mitochondria with decreased fission may be advantageous under lethal heat conditions. We showed that heat shock caused mitochondrial fragmentation and Drp1 translocation in control but not in HAtreated muscle cells. The homeostatic alterations in mitochondrial fission and morphology in HA-treated cells may limit recruitment of Drp1 and resist mitochondrial breakdown under heat stress. In addition, as discussed above, HA is known to cause overexpression of heat stress proteins. HSP70 can inhibit stress protein kinases (Gabai et al., 1997) that activate Drp1 under mitochondrial stress (Kashatus et al., 2015; Park et al., 2015). Thus, through complex mechanisms, HA may alter mitochondrial morphology and dynamics to protect cells against mitochondria-dependent apoptosis during heat exposure. Our results provide evidence that cells exposed to lethal heat shock undergo apoptotic death. Limited information is available on how heat stress initiates apoptosis signaling. Apoptotic cell death is initiated and activated by diverse internal or external pro-apoptotic insults. Depending on the type of stimulus, the apoptotic signaling cascades undergo a mitochondrial-dependent, ligand-mediated death receptor, or an endoplasmic reticulum stress-induced pathway (Adams, 2003). The execution phase of apoptosis usually involves activation of a family of protease proteins called caspases. Heat-induced apoptosis of C2C12 cells is likely activated primarily through an intrinsic, mitochondria-dependent pathway. Several events, including mitochondrial oxidative stress, observed in cells exposed to heat shock are linked to activation of caspases, which likely serve as the primary mediators of apoptosis (Primeau et al., 2002; Palmer et al., 2011; Rigoulet et al., 2011). These cellular events ultimately lead to changes in the integrity of the mitochondrial membrane (Westphal et al., 2011). Loss of mitochondrial integrity results in release of pro-apoptotic proteins, including cytochrome c, which facilitate recruitment of caspases to form the apoptosome. This downstream caspase activation results in subsequent apoptotic cell death under heat conditions (Katschinski et al., 2000; Wang et al., 2013). Thus, several primary pro-apoptotic factors clearly contributed to initiation and activation of heat-induced apoptosis of C2C12 cells. In the present study, we demonstrated that HA improved viability of C2C12 cells against heat. Reduced mitochondrial Drp1 and increased mitochondrial size were found in both C2C12 cells and mouse gastrocnemius muscles following HA. Whether these changes directly contributed to HA-induced heat tolerance of cells remains unclear. Mitochondrial apoptotic susceptibility to heat stress is likely dependent upon the ratio of proand antiapoptotic Bcl-2 family proteins, including pro-apoptotic Bax and anti-apoptotic Bcl-2 (Primeau et al., 2002). Further investigation of HA effects on Bcl-2 family proteins should provide information about improved ability of HA-treated cells to survive subsequent heat stress. Mitochondrial fission is involved in skeletal muscle differentiation and growth. Inhibition of mitochondrial fission has been shown to impair myogenesis (Kim et al., 2013), whereas severe burn injury has been shown to induce skeletal muscle regeneration (Fry et al., 2016). Whether heat shock would stimulate and HA would inhibit muscle regeneration process has yet to be examined. Our results revealed proteolytic conversion of larger into smaller isoforms of OPA1 in C2C12 cells during heat exposure. Whether this was caused directly by heat or indirectly by other mitochondrial events remains unclear. It is known that OPA1 plays a critical role in maintaining mitochondrial inner membrane integrity (Lee & Yoon, 2014). Disruption of OPA1 assembly can cause an increase of mitochondrial permeability (Frezza et al., 2006; Yamaguchi et al., 2008) and dissipation of mitochondrial membrane potential (Lee & Yoon, 2014). On the other hand, loss of mitochondrial membrane potential may also lead to the cleavage of OPA1 and fusion inhibition (Duvezin-Caubet et al., 2006; Ishihara et al., 2006; Liu et al., 2009). The fact that mitochondrial membrane potential (Figure 7A) in C2C12 cells declined well before, not concurrently with, proteolysis of OPA1 (Figure 8B) during heat exposure seems to negate a direct link between them. Instead, Drp1-dependent mitochondrial fission (Figure 4B) preceded all the mitochondrial events, including release of ROS, loss of membrane potential and cleavage of OPA1 during heat exposure (Figures 7A, 7D and 8B). Overall, our data support Drp1 over OPA1 as the primary mediator involved in the regulation of cell susceptibility to heat-induced apoptosis. First, translocation and phosphorylation of Drp1 and mitochondrial fragmentation preceded the cleavage of OPA1, and second, HA affected expression of Drp1, but not OPA1 in muscle cells and tissues. However, factors that may act as an upstream Drp1 signal in heat stress have yet to be determined. In conclusion, the findings of the present study provide novel insights into the role of mitochondria in the regulation of heat-induced apoptosis of muscle cells. Our results demonstrate that HA and heat shock cause changes in mitochondrial morphology of muscle cells by acting on Drp1, which results in favoring mitochondrial fusion and fission respectively. Inhibition of Drp1-dependent mitochondrial division improves mitochondrial morphology and cell viability during heat exposure. Thus, we propose that Drp1 may serve as a biomarker for susceptibility to heat stress and a candidate for preventing heat injury. Additional information Competing interests The authors declare that there are no competing interests. Author contributions T.Y. designed and performed the experiments, analyzed the data and helped prepare the manuscript. P.D. provided editorial advice. Y.C. developed the concept and project, analyzed the data and wrote the manuscript.