HRAS germline mutations impair LKB1/AMPK signaling and mitochondrial homeostasis in Costello syndrome models
Laetitia Dard, … , Didier Lacombe, Rodrigue Rossignol
Laetitia Dard, … , Didier Lacombe, Rodrigue Rossignol
Published March 1, 2022
Citation Information: J Clin Invest. 2022;132(8):e131053. https://doi.org/10.1172/JCI131053.
View: Text | PDF
Research Article Metabolism

HRAS germline mutations impair LKB1/AMPK signaling and mitochondrial homeostasis in Costello syndrome models

Abstract

Germline mutations that activate genes in the canonical RAS/MAPK signaling pathway are responsible for rare human developmental disorders known as RASopathies. Here, we analyzed the molecular determinants of Costello syndrome (CS) using a mouse model expressing HRAS p.G12S, patient skin fibroblasts, hiPSC-derived human cardiomyocytes, a HRAS p.G12V zebrafish model, and human fibroblasts expressing lentiviral constructs carrying HRAS p.G12S or HRAS p.G12A mutations. The findings revealed alteration of mitochondrial proteostasis and defective oxidative phosphorylation in the heart and skeletal muscle of CS mice that were also found in the cell models of the disease. The underpinning mechanisms involved the inhibition of the AMPK signaling pathway by mutant forms of HRAS, leading to alteration of mitochondrial proteostasis and bioenergetics. Pharmacological activation of mitochondrial bioenergetics and quality control restored organelle function in HRAS p.G12A and p.G12S cell models, reduced left ventricle hypertrophy in CS mice, and diminished the occurrence of developmental defects in the CS zebrafish model. Collectively, these findings highlight the importance of mitochondrial proteostasis and bioenergetics in the pathophysiology of RASopathies and suggest that patients with CS may benefit from treatment with mitochondrial modulators.

Authors

Laetitia Dard, Christophe Hubert, Pauline Esteves, Wendy Blanchard, Ghina Bou About, Lyla Baldasseroni, Elodie Dumon, Chloe Angelini, Mégane Delourme, Véronique Guyonnet-Dupérat, Stéphane Claverol, Laura Fontenille, Karima Kissa, Pierre-Emmanuel Séguéla, Jean-Benoît Thambo, Lévy Nicolas, Yann Herault, Nadège Bellance, Nivea Dias Amoedo, Frédérique Magdinier, Tania Sorg, Didier Lacombe, Rodrigue Rossignol

×

Figure 4

AMPKα2 expression is inhibited by mutant HRAS.

Options: View larger image (or click on image) Download as PowerPoint
AMPKα2 expression is inhibited by mutant HRAS. (A) Mitochondrial respira...
(A) Mitochondrial respiration determined in skin fibroblasts from patients with CS and in WT human fibroblasts transduced with an empty plasmid (pV) or mutant forms of HRAS (n ≥ 3 for each condition). (B) Mitochondrial ATP content (expressed as a percentage of the total ATP) (n ≥ 3). (C) Total AMPK α subunits (α1+α2) and T172_phospho-AMPK protein content were quantified in cells from patients with CS (n = 6) and WT controls (n = 6). (Phospho-AMPK/[phospho-AMPK+total AMPK]) was denominated “AMPK P172/T_ratio.” (D and E) AMPK and T172_phospho-AMPK protein content determined in mutant HRAS-expressing human skin fibroblasts (pV, empty plasmid; HRAS, WT gene; HRAS p.G12A and HRAS p.G12S, mutated forms of the gene) (n ≥ 3). (F) mRNA content of 2 AMPK subunits (α1 and α2) in different CS models: CS mouse heart, fibroblasts from patients with CS, and mutant HRAS cell models (n = 3). (G and H) Effect of HRAS mutations on HRAS activity (n = 3). (I) LKB1 expression level and its S428 phosphorylation status on control fibroblasts (n = 7) and fibroblasts of patients with CS (n = 6). (J) Protein content of AMPKα1+α2, T172_phospho-AMPK, LKB1, and S428_P-LKB1 in heart samples from Costello (HRAs p.G12S) or WT (HRAS) mice. (K) Protein levels of ATG7 and CLPX determined by mass spectrometry in cells from patients with CS (n = 3) as compared with control cells (n = 3). (L and M) LC3B-I and LC3B-II levels were determined in skin fibroblasts from controls or patients in presence or absence of 300 nM of bafilomycin A1 (n = 3). (N) LC3 activation level was expressed as (LC3-II/[LC3I+LC3II]). (O) Relative normalized expression of TFAM mRNA in cells from patients with CS (n = 3). Data are expressed as the mean ± SEM. One-way ANOVA with Dunnett’s correction for multiple testing was used to compare the 3 groups of cells expressing HRAS p.G12A or HRAS p.G12S with the empty plasmid control, while a t test was used to compare the 2 groups of cells obtained from patients with CS and controls or the 2 groups of mice (Costello or WT). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.