Background

Myostatin is a protein produced by the MSTN gene that stimulates ActRII signaling, thereby inhibiting the growth and differentiation of skeletal muscle. In mice, deletion of MSTN more than doubles muscle mass, but in humans, evidence has been limited to a few cases of congenital muscle hypertrophy. Myostatin-blocking antibodies in adults have only increased muscle mass by 2-4%, a significant difference from animal studies.

Recently, the glucagon-like peptide-1 receptor (GLP-1R) agonists have brought this pathway back into focus. Up to one-third of the weight loss achieved with GLP-1R drugs may be due to skeletal muscle loss. In particular, older adults with poor muscle recovery may be at increased risk of sarcopenia after obesity treatment. However, evaluating the human effects and long-term safety of rare MSTN variants requires a much larger sample size than previous studies.

Key Findings

The researchers analyzed exomes from 11 cohorts, totaling 1.1 million individuals. Among the 13,454 individuals carrying 506 rare MSTN variants (allele frequency <0.5%), 38 were predicted to be loss-of-function variants, and 188 were classified as impairing protein function. Individuals with loss-of-function variants did not differ in overall weight, but had lower body fat percentage and higher lean mass, as measured by bioelectrical impedance analysis. The statistical significance was P=2.2×10⁻⁷ for body fat percentage and P=1.5×10⁻⁷ for grip strength.

Whole-body Dixon MRI scans from 77,572 participants in the UK Biobank were analyzed using two types of deep learning segmentation models. The results showed that individuals with variants had increased muscle water volume in several muscle groups and decreased fat tissue volume and intramuscular fat infiltration. Individuals heterozygous for variants close to loss-of-function had 5-8% more total muscle mass, and some muscle groups, such as the gluteus maximus, had more than 10% more.

Arg65His, Thr115Met, Ile225Thr, and Arg283Cys were the four variants that drove the muscle-increasing signal. The effect of the relatively common Ile225Thr variant was about 20% of that of Arg283Cys, supporting a dose-response relationship between the degree of MSTN inhibition and muscle mass. One individual heterozygous for Arg65His had 31.8% lower body fat percentage and 12.3% higher lean mass than expected, but this is only a single case and not sufficient to confirm the effect size.

Significance and Outlook

These results provide large-scale genetic evidence that partial myostatin inhibition throughout life can increase muscle mass and strength while reducing fat mass in humans. Cardiac MRI showed no signs of ventricular wall hypertrophy, and there was no strong association with heart failure, hypertrophic cardiomyopathy, or reproductive markers. This indirectly supports the long-term safety of drug inhibition, but lifelong exposure in individuals with genetic variants cannot be equated with drug administration in adulthood.

There was a trend toward protection in waist-to-hip ratio and glycated hemoglobin, but statistical evidence was not sufficient to confirm a reduction in the risk of diabetes or cardiovascular disease. The frequency of the three variants was also very low, less than 0.01%. The protein stabilization and aggregation mechanisms proposed by the researchers using AlphaFold2 also need to be confirmed by cell and biochemical experiments. This study is more of a human genetic map for determining the appropriate degree of myostatin inhibition and evaluation parameters for myostatin inhibitors, rather than a clinical trial that proves efficacy. Original article