This study aims to clarify the causal relationship between gut microbiota and phenotype indicators of sarcopenia, grip strength and skeletal muscle mass, based on Mendelian randomization (MR) study design.

According to the data derived from the British Biological Bank database and the published genome-wide association study with available data, the genetic variation of 131 specific intestinal bacteria was selected as the tool variable. The Single Nucleotide Polymorphism (SNP) selection criteria include being significantly associated with gut microbiota at the whole-genome level, mutually independent among SNPs, and having a sufficiently high instrumental variable strength. An instrumental variable is considered sufficiently strong when F>10. The inverse variance method (IVW) was used as the main analysis method, as well as the weighted median method and MR Pleiotropy RESidual Sum and Outlier (MR-PRESSO) method were used as the sensitivity analysis methods, to evaluate the causal relationship among 131 gut microbiota species and grip strength and skeletal muscle mass. The MR-PRESSO method is used to test for horizontal pleiotropy and address heterogeneity.

A total of 324, 976 subjects were included in this study. The number of SNPs ranges from 3 to 22, with a minimum F value of 14.6. The IVW results suggested a causal relationship between the 9 genetically predicted bacterial genera and host grip strength (P<0.05). Among them, Alloprevotella (β=0.012 kg, 95% CI: 0.002, 0.022) and Sellimonas (β=0.014 kg, 95% CI: 0.006, 0.022) showed a positive causal effect, while Olsenella (β=-0.012 kg, 95% CI: -0.023, 0.001) and Paraprevotella (β=-0.014 kg, 95% CI: -0.023, 0.004) showed negative causality. The causal associations between these four gut specific bacteria and host grip strength were statistically different in different MR sensitivity analyses, suggesting that the results were stable and reliable. In addition, a total of 7 genetically predicted gut specific bacterial genera had causal associations with host skeletal muscle mass (P<0.05), among which Eubacterium nodatum group (β=0.069 kg, 95% CI: 0.012, 0.125) showed a positive causal association with host skeletal muscle mass, while Erysipelatoclostridium (β=-0.090 kg, 95% CI: -0.162, 0.019) and Ruminococcaceae UCG011 (β=-0.104 kg, 95% CI: -0.199, 0.010) showed a negative causal association with host skeletal muscle mass. The causal associations between these three specific gut bacteria and host skeletal muscle mass were statistically different in different MR sensitivity analyses, suggesting that the results were stable and reliable. The results of MR-PRESSO suggest that the potential influence of horizontal pleiotropy and heterogeneity may be relatively small.

This study investigated specific gut microbiota have a causal relationship with host grip strength and skeletal muscle mass, such as Alloprevotella, Selemonas, and Eubacterium nodatum group, which have a positive causal relationship with phenotype indicators of sarcopenia, while Olsenella, Paraprevotella, Erysipelatoclostridium and Ruminococcaceae UCG011 have a negative causal relationship with phenotype indicators of sarcopenia, so as to provide theoretical references for elucidating the pathogenesis of sarcopenia and developing new treatment methods.