Effects of HMG CoA reductase (HMGCR) deficiency on skeletal muscle development

Abstract

Pathogenic variants in HMGCR were recently linked to a limb-girdle muscular dystrophy (LGMD) phenotype. The protein product HMG CoA reductase (HMGCR) catalyzes a key component of the cholesterol synthesis pathway. The two other muscle diseases associated with HMGCR, statin-associated myopathy (SAM) and autoimmune anti-HMGCR myopathy, are not inherited in a Mendelian pattern. Statins inhibit HMGCR activity to generate their cholesterol-lowering effects and are known to cause multiple types of adverse effects on skeletal muscle, while the antibodies associated with anti-HMGCR myopathy specifically target this enzyme. The mechanism linking pathogenic variants in HMGCR with skeletal muscle dysfunction is unclear. We knocked down Hmgcr in mouse skeletal myoblasts, knocked down hmgcr in Drosophila, and expressed three pathogenic HMGCR variants (c.1327C>T, p.Arg443Trp; c.1522_1524delTCT, p.Ser508del; and c.1621G>A, p.Ala541Thr) in Hmgcr knockdown mouse myoblasts. Hmgcr deficiency was associated with decreased proliferation, increased apoptosis, and impaired myotube fusion. Transcriptome sequencing of Hmgcr knockdown versus control myoblasts revealed differential expression involving mitochondrial function, with corresponding differences in cellular oxygen consumption rates. Both ubiquitous and muscle-specific knockdown of hmgcr in Drosophila led to lethality. Overexpression of reference HMGCR cDNA rescued myotube fusion in knockdown cells, whereas overexpression of the pathogenic variants of HMGCR cDNA did not. These results suggest that the three HMGCR-related muscle diseases share disease mechanisms related to skeletal muscle development.

Keywords: HMGCR; muscular dystrophy; myoblast; skeletal muscle; statin myopathy.

Fig. 1

Diagram of the HMGCR metabolic pathway. 3‐hydroxy‐3‐methylglutaryl‐CoA (HMG‐CoA) is converted to mevalonate by HMG‐CoA reductase (HMGCR). Mevalonate then serves as a substrate for several downstream reactions, including cholesterol synthesis. Statins inhibit HMGCR, lowering cholesterol synthesis in hypercholesterolemia. Two additional enzymes associated with inherited skeletal muscle or musculoskeletal diseases are shown: HMGCS1 and GGPS1. Diagram was generated using BioRender under a license permitting its use in publications.

Fig. 2

Structural representation of the human HMGCR protein with pathogenic variants. (A) Diagram of the protein domains of human HMGCR. Positions of the three pathogenic variants in the current study are marked with yellow arrow heads. The p.Arg443 residue is located outside the catalytic domain, whereas p.Ser508 and p.Ala541 are located in the catalytic domain. Diagram was generated using BioRender under a license permitting its use in publications. (B) Diagram of the full HMGCR protein is shown with the three variants of interest. Insets of HMGCR are shown with the (C) p.Arg443, (D) p.Ser508, and (E) p.Ala541 residues noted. The amino acid residues of interest are colored blue and marked by red arrows. Diagrams (B–E) were generated using Pymol 2.5.0.

Fig. 3

shRNA‐mediated knockdown of Hmgcr in C2C12 myoblasts. (A) qPCR was performed in triplicate for each sample with n = 5 experiments. Transcript levels were normalized against Gapdh with fold change calculated as Hmgcr knockdown versus scrambled control. (B) shRNA mediated knockdown of Hmgcr in C2C12 myoblasts. Successful shRNA transfection was confirmed by detecting GFP‐positive cells. Scale bar, 100 μm. (C) Hmgcr knockdown efficiencies on day 3 and day 7 of differentiation were 53% and 61%, respectively, on qPCR. Results are from n = 3 independent experiments. Unpaired t‐test results showed ****P < 0.0001, **P < 0.01; and *P < 0.05. Error bars show standard error of the mean (SEM).

Fig. 4

Impact of Hmgcr deficiency on cell proliferation and apoptosis. (A) Hmgcr shRNA and scrambled shRNA control cells were seeded and counted daily for 3 days. Hmgcr knockdown cells showed decreased proliferation compared with scrambled controls. Unpaired t‐tests showed P values as follows: **P = 0.0032 (day 1); *P = 0.0214 (day 2); and *P = 0.0211 (day 3). N = 3 experiments. The mean proliferation rate over 72 h was 2685 cells·h⁻¹ for the Hmgcr knockdown cells and 5373 cells·h⁻¹ for the scrambled control cells. (B) Dead End Calorimetric TUNEL assay was used to detect apoptotic cells, identified as dark stained nuclei indicating nuclear fragmentation and condensation. Dnase 1 treatment served as a positive control, and absence of the TdT enzyme mix served as a negative control. Hmgcr knockdown is associated with an increased number of apoptotic cells compared to scrambled shRNA control (black arrows). Representative microscopic images of n = 3 experiments are shown. Scale bar, 100 μm. (C) Quantification of the TUNEL assay was conducted via manual counting of dark brown clumped cells. An unpaired t‐test showed *, P = 0.0488. Error bars show standard error of the mean (SEM).

Fig. 5

Impact of Hmgcr deficiency on myoblast differentiation. (A) Hmgcr knockdown and scrambled shRNA control myoblasts were seeded and switched to differentiation medium at ~80% confluence. Hmgcr‐deficient cells showed decreased differentiation potential at Days 3 and 7, shown in representative images from n = 3 independent experiments and quantified in panel D. Scale bar, 100 μm. (B) The graph shows significant downregulation of MyoG gene expression levels normalized to Gapdh in Hmgcr knockdown cells compared to scrambled control cells in n = 3 independent experiments. ANOVA analysis showed *P = 0.0416; **P = 0.0078. (C) Hmgcr shRNA and scrambled shRNA control myoblasts were differentiated and stained for desmin. Hmgcr knockdown cells showed impaired myotube formation compared to scrambled shRNA control cells. Scale bar, 100 μm. (D) The myotube fusion index calculated from 7 microscope fields in n = 3 independent experiments was significantly lower for Hmgcr knockdown cells compared to scrambled shRNA controls. An unpaired t‐test showed ****P < 0.0001. Error bars show standard error of the mean (SEM).

Fig. 6

Filipin staining in Hmgcr deficiency. (A) Hmgcr knockdown and shRNA myoblasts were stained with Filipin III to label total free cholesterol. The Hmgcr knockdown cells had decreased staining intensity compared to scrambled shRNA cells. Scale bar, 50 μm. (B) Fluorescence intensity was measured for at least 50 cells per sample. Results are from n = 3 independent experiments. An unpaired t‐test showed **P < 0.01. Error bars show standard error of the mean (SEM).

Fig. 7

Whole transcriptome analysis of Hmgcr deficiency. (A) Total RNA was isolated from Hmgcr shRNA knockdown and scrambled shRNA cells. Reverse transcription was conducted to generate cDNA libraries for two biological replicates of nanopore sequencing for each condition. A total of 299 significant differentially expressed genes were identified using Deseq2. These genes are represented in a volcano plot with the Ogdhl gene highlighted. The dotted line indicates P‐value = 0.05. (A). The outliers Nefl, Il33, Prl2c3, Myl9, Stmn2, Peg10 and Crabp2 were excluded from the volcano plot to facilitate visualization of the core pattern. Among the 299 genes differentially expressed between Hmgcr shRNA knockdown and scrambled shRNA control cells, 109 genes were downregulated and 190 were upregulated genes. These genes were analyzed using ShinyGO 0.77. Gene ontology categories are shown based on the (B) cellular components and (C) KEGG pathways for upregulated genes, and the (D) cellular components and (E) KEGG pathways for downregulated genes.

Fig. 8

Oxygen consumption and metabolic activity in Hmgcr deficiency. The oxygen consumption rate (OCR) of Hmgcr shRNA versus scrambled shRNA C2C12 myoblasts was determined using the RESIPHER system at (A) 24 h, (B), 48 h, and (C) 72 h. The OCR of the Hmgcr knockdown cells was increased compared to scrambled controls at 24 and 48 h. (D) Rotenone treatment at 72 h confirms that the increased oxygen consumption occurred via mitochondrial respiration. (E) A 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide (MTT) assay at 72 h shows that the total cellular metabolic activity is higher in Hmgcr knockdown cells compared to scrambled controls. Paired t‐tests were performed. *P < 0.05. (A–D) show results from n = 4 independent experiments and (E) shows results from n = 3 independent experiments. Error bars show standard error of the mean (SEM).

Fig. 9

NCBI blastp alignment between Homo sapiens 3‐hydroxy‐3‐methylglutaryl‐coenzyme A reductase (HMGCR) UniProtKB/Swiss‐Prot: P04035 and Drosophila melanogaster hmgcr NP_732900.1. The p.Arg443 and p.Ala541 residues targeted by human pathogenic variants are conserved in the fly (red ovals).

Fig. 10

RNAi‐mediated knockdown of the homologous fly gene hmgcr. (A) The hmgcr RNAi sequence is aligned on the hmgcr locus (arrow, FlyBase, GBrowse). Corresponding mutant fly lines were obtained. (B) On the left, ubiquitous downregulation of hmgcr (Act5C‐G4>UAS‐ds hmgcr Drosophila) leads to lethality while two sets of hmgcr ^+/+ control (Cy) siblings emerge normally (Act5c‐G4 and UAS‐hmgcr). Conversely on the right, ubiquitous hmgcr overexpression (Act5C‐G4>UAS hmgcr Drosophila) is well tolerated. RNAi and overexpression are accomplished via the Gal4/UAS binary system. Ds, double stranded; Act5C, cytosolic Actin5C. (C) The lethality phenotype is recapitulated when hmgcr knockdown is under the control of either the muscle Gal4 driver Mef2 (number of emerged progeny: control flies n = 80, experimental RNAi flies n = 0, one replicate), or the muscle Gal4 driver how (number of emerged progeny: control flies n = 101, experimental RNAi flies n = 0, three replicates).

Fig. 11

Stable expression of human reference HMGCR and three pathogenic variants of HMGCR in Hmgcr knockdown and scrambled shRNA cells. (A) On Day 3 of culture, there were moderate rescue effects with expression of reference HMGCR but not with the variant HMGCRs on proliferation patterns of Hmgcr knockdown cells. N = 6 experiments. (B) On Day 7 of differentiation, qPCR showed moderate rescue of mRNA MyoG expression in Hmgcr knockdown cells with expression of reference HMGCR but not with expression of the 3 variant HMGCRs. N = 3 experiments. (C) On Day 4 of differentiation, the myotube fusion index in Hmgcr knockdown cells showed recovery with expression of reference HMGCR but not with the 3 variant HMGCRs. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. One‐way ANOVA was conducted throughout. Error bars show standard error of the mean (SEM).