Recent research highlights the importance of muscular strength as a key factor in physical fitness, a strong indicator of overall mortality risk, and a vital target for preventing chronic diseases. This study used a proteome-wide Mendelian randomization analysis plus colocalization analysis for low hand grip strength to explore potential therapeutic targets for muscle weakness.We conducted two two-sample Mendelian randomization analyses from four cohorts to identify and validate the causal relationship between plasma proteins and low grip strength. We also employed bidirectional Mendelian randomization analysis with Steiger filtering, Bayesian co-localization, and phenotype scanning to detect reverse causality, thereby consolidating our Mendelian randomization findings. Downstream analyses were also undertaken of identified proteins, including knockout models, enrichment analyses, and protein-protein interaction networks. Finally, we assessed the druggability of the identified proteins.At Bonferroni significance (P < 6.82 × 10-5), Mendelian randomization analysis revealed that three proteins were causally associated with low grip strength. Increased MGP (OR = 0.85) and HP (OR = 0.96) decreased the risk of low grip strength, whereas elevated ART4 (OR = 1.06) increased the risk of low grip strength. None of the three proteins had reverse causality with low grip strength. Bayesian co-localization suggested that MGP shared the same variant with low grip strength (coloc.abf-PPH4 = 0.826). Further downstream analyses showed that MGP, which is highly expressed in musculoskeletal system, is a potential novel target for muscle weakness.The proteome-wide Mendelian randomization investigation identified three proteins associated with the risk of muscle weakness. MGP, HP, and ART4 deserve further investigation as potential therapeutic targets for muscle weakness.Copyright © 2024 The Authors. Published by Elsevier Masson SAS.. All rights reserved.