HACD1, a regulator of membrane composition and fluidity, promotes myoblast fusion and skeletal muscle growth

The reduced diameter of skeletal myofibres is a hallmark of several congenital myopathies, yet the underlying cellular and molecular mechanisms remain elusive. In this study, we investigate the role of HACD1/PTPLA, which is involved in the elongation of the very long chain fatty acids, in muscle fibre formation. In humans and dogs, HACD1 deficiency leads to a congenital myopathy with fibre size disproportion associated with a generalized muscle weakness. Through analysis of HACD1-deficient Labradors, Hacd1-knockout mice, and Hacd1-deficient myoblasts, we provide evidence that HACD1 promotes myoblast fusion during muscle development and regeneration. We further demonstrate that in normal differentiating myoblasts, expression of the catalytically active HACD1 isoform, which is encoded by a muscle-enriched splice variant, yields decreased lysophosphatidylcholine content, a potent inhibitor of myoblast fusion, and increased concentrations of ≥C18 and monounsaturated fatty acids of phospholipids. These lipid modifications correlate with a reduction in plasma membrane rigidity. In conclusion, we propose that fusion impairment constitutes a novel, non-exclusive pathological mechanism operating in congenital myopathies and reveal that HACD1 is a key regulator of a lipid-dependent muscle fibre growth mechanism.

Labrador Retrievers were dogs maintained in our research colony, and samples used in this study were frozen samples obtained in the 1990s by one of the co-authors (SB). At the time they were sampled, there was no animal welfare committee at the Ecole nationale vétérinaire d'Alfort; however, SB is a certified veterinarian and was accredited by the Veterinary Division of the French Ministry of Agriculture to perform research on animals.

Cell culture
HEK 293T cells were grown in Dulbecco's modified Eagle's medium (DMEM; Sigma, St. Louis, MO), supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin, seeded in dishes coated with 0.3% collagen respectively. C2C12 myoblasts were grown in DMEM (PAA) containing 4.5 g/l glucose, supplemented with 15% FBS (PAA) and penicillin/streptomycin. Cells were maintained at 37 °C in a saturated humidity atmosphere containing 5% CO 2 . For the differentiation of C2C12 myoblasts, fetal bovine serum was replaced by 2% horse serum (Biowest) when myoblasts reached 70-80% confluence. For fatty acid supplementation, 10 mM fatty acids (Sigma) stock solution maintained at -20 °C in ethanol in glass tubes, were dissolved in DMEM supplemented with 0.25 mM fatty acid-free BSA (Sigma, A6003) at 37 °C for 5 min under strong agitation. Fatty acid solutions were filtrated through 0.2 µm filters (Dutscher, 146560) and diluted in differentiation medium (DMEM + 2% of horse serum) at a final concentration of 5 µM from the first day of differentiation (D0). The fusion index was calculated as the mean number of nuclei per myotube at day 5 of differentiation. Myotubes were defined as myosin heavy chain-positive cells containing at least 2 nuclei. Primary myoblasts were obtained by dissection of hindlimb muscles from 5-day-old wild type, Hacd1 +/and Hacd1 -/pups. Muscles were digested in PBS containing 0.5 mg/ml collagenase (GIBCO 17101) and 3.5 mg/ml dispase (GIBCO 17105) for 2 h at 37 °C. Cell suspension was filtered through a 40 µm cell strainer and pre-plated in DMEM + 15% FBS for 4 h at 37 °C, 5% CO 2 . Non-adherent myogenic cells were collected and plated in IMDM (GIBCO 31980) + 20% FBS onto Ibidi dishes (Biovalley 80426) coated with collagen (Sigma C7661). Differentiation was triggered by changing the culture media to IMDM + 2% of horse serum for 2 days.

Generation of sh-Hacd1 cells
Two shRNA pGIPZ lentiviral vectors designed to target Hacd1 exon 4 (V2LHS_5923 (GCTCATTACTCACAGTATA) and V2LHS_252516 (CTCATTACTCACAGTATAA)) and a control vector (RHS4349; OpenBiosystems) were transfected into C2C12 cells using Arrest-in (OpenBiosystems). After 48h, transfected cells were selected with puromycin (2 mg/ml; Invitrogen). Individual clones were picked after 10 days of selection and shRNA efficiency was checked by RT-qPCR. Initial experiments conducted in parallel in one V2LHS_5923 and one V2LHS_252516 clone proved phenotypic similarity. Clone V2LHS_252516 clone (named sh-Hacd1 cells) was then selected for the series of functional experiments. Control cells correspond to cells transduced with the control vector.

Transduction of HACD1 isoforms
For expression of shRNA-resistant isoforms in control and sh-Hacd1 cells, cDNAs from the three canine HACD1 isoforms were isolated by RT-PCR on mRNA extracted from muscles of wild type Tranduced cells were selected with blasticidin (5 µg/ml; Invitrogen). Plasmid expression was checked by RT-qPCR and immunostaining.

Extraction of total RNA, RT-PCR and RT-qPCR analysis
Mouse muscle samples were snap frozen in liquid nitrogen and stored at -80 °C. Total RNAs were isolated from C2C12 myoblasts and mouse samples with RNA Nucleospin Kit (Macherey-Nagel) according to the manufacturer's protocol. Purity of RNAs was assessed by a ratio of absorbance at 260 nm and 230 nm > 1.7. RNA quality was checked on agarose gel. One microgram of RNA was  Table S1. All PCR and qPCR products were examined qualitatively on agarose gels. All presented RT-qPCR results were normalized to Tbp1 (1:500) (Invitrogen) secondary antibodies were then used. Images were captured using an Axio Observer Z1 microscope (Zeiss) and analyzed using Photoshop CS3.
Routine muscle histopathological evaluation was conducted on the basis of the H&E staining.
Diameter and distribution of myofiber type were examined on ATPase 9.4 sections. Quantification of fiber number, size or nuclear content was performed on 7 randomly-selected fields (dog samples) or 3 sections separated by at least 30 µm (mouse samples). Morphometric quantification of the minimal Feret's diameter was done using the Visilog software (Noesis) and quantification of fiber number and nuclear content was done using ImageJ (1.47v). In dog and mouse samples, 500 to 1000 fibers and 700 to 1500 fibers, respectively, were analyzed for each section.
Expression vector was electroporated into tibialis cranialis (TA) muscles of 7-wk-old female C57BL/6 mice (2 mice bilaterally) according to (McMahon et al., 2001). Mice were anaesthetised with fentanyl/fluanisone (Hypnorm) and midazolam (Hynovel) and injected intramuscularly with bovine testes hyaluronidase (25 µl/muscle of 0.4 µg/µl solution in sterile saline) using an insulin syringe and 28 gauge needle. After 2 h, anaesthesia was deepened with isofluorane inhalation, and 25 µl plasmid solution (0.5 µg/µl) was injected into the TA prior to transcutaneous electroporation performed by applying a potential difference of 81V across the muscle in 10 x 20 ms pulses at 1 Hz frequency (ElectroSquare Porator ECM 830, BTX). After 6 days, mice were euthanized by cervical ! 7! dislocation and the TA muscles were dissected and treated as described in the previous section.

Western blot analysis
sh-Hacd1 cells expressing shRNA-insensitive isoforms were washed in PBS and put in lysis buffer Quantification was performed using Photoshop CS3.
Labeling was detected using Pierce Western Blotting Substrate (Thermo Fisher Scientific, Waltham, MA, USA).
Human EIF2B1 cDNA was cloned in the pDEST-Myc Gateway vector (gift from Marc Vidal's lab) as a control prey. For each interaction assay, GST-bait and Flag/Myc-prey expression vectors were co-transfected in HEK293T cells using calcium phosphate. Each Flag/Myc-prey vector was also co- Acquisition was done using Fusion Fx5 (Vilber Lourmat).

Lipids analyses
Cell preparation C2C12 cells were rinsed using 0.9% NaCl, trypsinated and resuspended with proliferation medium.
Cells were centrifuged at 4,000 g for 10 min and pellets were rinsed twice using 0.9% NaCl. Tubes containing dry pellets were filled with non-reactive argon to prevent aerial oxidation of samples and then immediately frozen at -80 °C. Finally fatty acids were dissolved in 100 µl prior to GC-MS quantification.

Reagents
Fatty acids esters (1 µl) were injected in split mode on a HP7890A Gas Chromatograph equipped with an HP7683 injector and a HP5975C Mass Selective Detector (Agilent Technologies).
Analysis of phospholipids was conducted by LCMS-MS in MRM mode as previously described (Vial et al., 2014).

Membrane fluidity assay
Plasma membrane fluidity was estimated by fluorescence anisotropy (r) as described (Bastiaanse et al., 1995;Le Borgne et al., 2012). (r) measurements were conducted using 1,6 diphenyl 1,3,5 hexatriene (DPH) as a probe. In polarized excitation light, a high r value reflects increased rotational lifetime of DPH that happens when DPH orientation is stabilized in less fluid membranes.
Cultured cells were rinsed, harvested and centrifuged at 1,000 g for 5 min at 4 °C. The resulting ! 10! pellets were resuspended in Opti-MEM® Reduced Serum Medium (Life Technologies, Saint Aubin, France) at a concentration of one million cells per ml. Two milliliters of cell suspension were placed into a 1 cm path length spectroscopic quartz cuvette (VWR International, Limonest, France). The cells were stirred and maintained in a thermostatically controlled chamber. After 6 min at 37 °C, 2 µl of the fluorescent probe DPH (1 mM in tetrahydrofuran) were added to label plasma membranes. r was measured at 37 °C 10 min after addition of the probe; after this first series of measurement, the temperature of the chamber was gradually reduced and anisotropy was measured at 35, 30, 25, 20, 15, 10 and 5 °C. Data were expressed as mean ± s.e.m. of three to five independent experiments.

Statistical analysis
ANOVA with repeated measure factor was applied to paired data, i.e. in the case of measurement of fiber number, size, nuclear content or PAX7-positive cell content on muscle sections, as well as mass measurements of paired muscles in mice. Student T test was used for all other analyses and Welch modification was applied when sample number was low (n < 4). Data are expressed as mean ± standard error of the mean and differences were considered statistically significant when P < 0.05.

14!
(PUFA) fatty acids in proliferation (n = 3 for each condition; error bars correspond to standard error of the mean). (B) Fluorescence anisotropy (r, inverse of fluidity) measured at Day 3 of differentiation from 5°C to 37°C. For each cell type, the measured anisotropy increased when the temperature decreased, demonstrating that the DPH probe was in a lipid environment (n = 3 for each condition). Error bars correspond to standard error of the mean. *: P < 0.05; ***: P < 0.001.