Heat shock protein A4 ablation leads to skeletal muscle myopathy associated with dysregulated autophagy and induced apoptosis

Male and female Hspa4-KO mice were generated on 129/Sv genetic background as described previously [17].

Protein lysates were extracted from frozen tibialis anterior (TA) muscles using RIPA lysis buffer (Millipore) containing protease and phosphatase inhibitor cocktail (Roche Diagnostics). Aliquots of 20 μg lysates were resolved on a NuPage 4–12% SDS-PAGE. Western blotting was carried out using the following primary antibodies: rabbit anti-LC3, anti-p62 (Cell signaling technology), anti-BCL-2 (Abcam), anti-HSPH1 (Sigma Aldrich), anti-HSPA4L, anti-HSPA4, mouse anti-BAX and anti-GAPDH (Santa Cruz Biotechnology). For quantification, an enhanced chemiluminescence detection system (Amersham Bioscience) and Image Lab software (Bio-Rad) were used according to the manufacturer’s instructions.

Muscles were collected and either paraffin-embedded, or immediately frozen in isopentane. Sections (6 µm) were stained with hematoxylin and eosin (H&E), and the number of centrally nucleated fibers was counted across 5 separate fields of view from at least three sections of each mouse. TUNEL assay was performed in paraffin-embedded sections using In Situ Cell Death Detection Kit (Roche Diagnostics). After fixation in ethanol–acetic acid, TA sections were treated with proteinase K and permeabilized with 0.5% Triton X-100. The sections were then incubated in the TUNEL reaction mixture containing terminal deoxynucleotidyl transferase and nucleotide mixture for 60 min at 37 °C in a dark humid chamber. TUNEL-positive cells were counted in 5–8 random fields/ muscle. For immunofluorescence, frozen sections were permeabilized using 0.2% Triton X-100 in phosphate-buffered saline (PBS), blocked with 5% bovine serum albumin in PBS and incubated with rabbit anti-LC3 (Cell signaling technology). Photomicrographs were captured using a microscope Olympus BX60 fluorescence microscope.

For real time PCR, cDNA synthesis was carried out with iScript cDNA synthesis kit (Bio-Rad). QRT-PCR was performed on a Biorad iQ-Cycler using SYBR Green Supermix (Bio-Rad). For Northern blot analysis, 20 μg of total RNA samples was size fractionated by electrophoresis, transferred onto nylon membrane (Amersham Bioscience) and hybridized with a 32P-labeled fragments. All the primers used are listed in the Additional file 1: Table S1.

Mouse C2C12 myoblasts [American Type Culture Collection (ATCC)] were cultured in growth media (GM) containing Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen), 10% fetal bovine serum and 1% penicillin–streptomycin (Sigma-Aldrich). Differentiation in C2C12 cultures was induced by replacing the growth with differentiation medium (2% horse serum in DMEM and 1% antibiotic mixture).

Using 20S Proteasome Assay Kit (10,008,041; Biomol), the 20S proteasome assay was carried out in a total volume of 100 μl in 96 well plates. Assays were initiated by addition of 100 μM of fluorescently labeled substrate, succinyl-Leu-Leu-Val-Tyr-7-amido-4-methylcoumarin (Suc-LLVY-AMC), to the protein lysates (50 μg) and incubation at 37 °C. These substrates are cleaved by the proteasome, releasing free AMC which was then measured spectrofluorometrically after one hour at an excitation wavelength of 360 nm and an emission wavelength of 480 nm. Each assay was conducted in duplicates and in the absence and presence of the specific proteasomal inhibitor, lactacystin (20 μM).

Adult mice were anesthetized with isoflurane and 50 μL of 10 μM cardiotoxin was injected into the left TA muscle. As a control 0.9% saline (vehicle) was injected in the contralateral side. Carprofen was used for post-treatment analgesia. Mice were sacrificed and TA muscles were dissected at various time points after injection.

Statistical analysis was carried out using GraphPad Prism 7.0 (GraphPad Software, Inc, California, USA) with two-tailed unpaired Student’s t-test or one-way ANOVA with Bonferroni post-test correction where appropriate. Kaplan–Meier survival analysis was performed, and a Log-rank test was used to determine significance.

Western blot analysis revealed that HSPA4 is ubiquitously expressed in different skeletal muscles Additional file 1: Fig. S1). Markedly increased Hspa4 mRNA and protein levels were detected in regenerating TA muscle after CTX injection, which induces muscle degeneration followed by regeneration (Fig. 1A–C). HSPA4 protein expression was also markedly upregulated in immortalized C2C12 myoblasts after induction of differentiation (Fig. 1D). Taken together, our data suggest a relevant role of HSPA4 in myogenesis, which prompted us to characterize skeletal muscles in Hspa4-KO mice.

Hspa4-KO pups were born from heterozygous breeding pairs at the expected Mendelian ratios (Table 1). In the first week postnatal, the Hspa4-KO pups were indistinguishable from their Hspa4^+/+ and Hspa4^± littermates. However, starting at postnatal day 8 (P8), Hspa4-KO pups gained less weight and showed clear growth retardation between P14 and P28 (Fig. 2A, B), likely the result of decreased milk intake caused by muscle weakness. Serum glucose levels and expression of the key gluconeogenic enzyme, lipid transport related-genes and growth hormone-responsive gene did not show any significant difference between Hspa4-KO and control animals (Additional file 1: Fig. S2). In accordance with stunted growth, Hspa4-KO mice also exhibited generalized dwarfism affecting all organs tested. However, the decrease in muscle mass was much more severe than other organs (Table 2). About 35% of the Hspa4-KO mice died during the peri-weaning period (Fig. 2C). After weaning, surviving Hspa4-KO mice normalized their body weight and were generally similar to wild-type (WT) littermates by two months of age (Fig. 2B). By the age of 12 months, Hspa4-KO mice developed spinal deformity in the form of kyphosis (Fig. 2D), indicative of paraspinal muscles weakness [23]. The protein level of HSPA4L and HSPH1, other members of HSP110 family, was not different between WT and Hspa4-KO muscles ruling out any compensatory upregulation of the studied proteins in the Hspa4-KO muscles (Fig. 2E, F).

By the age of 4 months, H&E-stained Hspa4-KO TA muscles exhibited myopathic changes including heterogeneous myofiber size distribution, numerous centrally nucleated fibers (CNF), and inflammatory cell infiltrates (Fig. 3A–C). The induction of the inflammatory response in Hspa4-KO TA muscles was confirmed at the transcript level by the detection of macrophage-specific markers, Cd68 and F4/80, and interleukins, Il6 and Il1β (Fig. 3D). TA muscles from Hspa4-KO mice experienced induction of embryonic (Myh3) and perinatal (Myh8) muscle myosin heavy chain genes, indicative of ongoing de- and regeneration (Fig. 3E, F). These data suggest that the Hspa4 deletion impairs the integrity of the myofibers resulting in the activation of a regenerative response. Interestingly, induction of Myh8 and Myh3 and increased percentage of CNF were also observed in Hspa4-KO muscles during the peri-weaning period (Fig. 3E, F and Additional file 1: Fig. S3). Myopathy was observed in other examined muscles including soleus, gastrocnemius and paraspinal muscles (Fig. 3A and Additional file 1: Fig. S4), indicating a generalized skeletal muscle involvement in Hspa4-KO mice.

Induced HSPA4 expression in muscles following muscle injury forced us to address the requirement of HSPA4 for normal muscle regeneration and recovery following muscle injury. We monitored skeletal muscle repair in CTX-injected TA muscles of 4-month-old WT and Hspa4-KO mice. One day following CTX injury muscles from both genotypes showed significant fiber degeneration and necrosis, as confirmed by eosinophilic staining and marked mononuclear cell infiltration. By 5 days following injury, degenerating muscle fibers were largely cleared in WT and Hspa4-KO muscles, and replaced by centrally nucleated nascent myoblasts and other mononuclear cells, indicative of the early phase of muscle regeneration. At 15 days of injury, the nascent myofibers in both genotypes were similarly enlarged (Fig. 4A–C). Complete restoration of the muscle architecture was observed in both genotypes at 30 days post-injury (Fig. 4A). The relative mRNA levels of Pax7, Myod1 and Myogenin, satellite cell-related markers for activation, fusion and differentiation, respectively, were not statistically different between WT and Hspa4-KO muscles at any time point post-injury (Fig. 4D–F). These data therefore suggest that the ability of Hspa4-KO mice to activate the myogenic program is not markedly altered.

Autophagy is initiated with the sequestration of cytoplasmic components by isolation membrane that expands to form double-membrane vesicles, the autophagosomes, which fuse with endosome/lysosome, followed by lysosomal hydrolysis of sequestered cytoplasmic components [24]. Autophagy can be assessed by detection of the modification of microtubule associated protein1 light chain 3 (LC3) from free mature LC3-I form to membrane-bound lipidated LC3-II form and by assessment of protein level of the autophagy adaptor, p62/sequestosome [25]. Hspa4-KO TA muscles showed marked increased LC3-II and p62 protein levels (Fig. 5A, B). Consistently, immunofluorescence revealed an abundant LC3 puncta in Hspa4-KO muscles (Fig. 5C), suggesting the dysregulated autophagy as a possible cause for skeletal muscle myopathy in Hspa4-KO mice.

Because an impaired proteasome pathway can also lead to an accumulation of p62 protein, we assessed proteasome activity in WT and Hspa4-KO TA muscles. No marked changes were found in chymotrypsin enzyme activity in total homogenates from WT and Hspa4-KO TA muscles (Fig. 5D). Treatment with known 20S proteasome inhibitors, lactacystin, resulted in a significant, but comparable, inhibition of 20S activity in WT and Hspa4-KO muscles (Fig. 5D). Moreover, the expression of atrogenes, MuRF1 and MAFbx, was not different between WT and Hspa4-KO muscles (Fig. 5E), indicating that the proteasome activity is not impaired in Hspa4-KO muscles.

An anti-apoptotic effect of HSPA4 was previously reported [18‐20]. We examined the apoptosis in the skeletal muscle of Hspa4-KO mice by performing TUNEL assay using TA sections from 4-month-old WT and Hspa4-KO mice. The number of apoptotic nuclei was markedly increased in Hspa4-KO compared to WT muscles (Fig. 6A, B). Additionally, the protein level of anti-apoptotic factor BCL-2 and consequently BCL-2/ BAX ratio were markedly decreased in Hspa4-KO TA muscles, suggesting that apoptosis may have exacerbated skeletal muscle myopathy in Hspa4-KO mice.