Objective To investigate the feasibility of using the cell-free DNA(cfDNA)carrying tissue-specific mutations as biomarkers for assessing the severity of exercise-induced tissue injury.Methods Based on a public gene expression database,ten tissue-specific and highly-expressed genes were selected for each of ten different human tissues.A total of 34 young healthy volunteers were recruited,and their peripheral blood samples were collected after running 5 km per day for one week.Genomic DNA from leukocytes,plasma exosomal RNA,and plasma cfDNA were extracted and subjected to high-throughput sequencing respectively.Tissue-specific somatic mutations in plasma exosomal RNA were identified,and the relationship between cfDNA carrying these mutations and traditional protein biomarkers was analyzed.Results Tissue-specific cfDNA mutations related to five tissues(myocardium,skeletal muscle,intestine,stomach,and kidney)were detected in more than five volunteers.The correlation coefficients between total plasma cfDNA levels and the levels of tissue-specific protein biomarkers associated with these five tissues were less than 0.3(n=34,r=-0.51-0.28,P=0.0022-0.65).Notably,for each specific tissue,the levels of tissue-specific cfDNA mutations were positively correlated with the corresponding protein biomarker levels,and correlation coefficients were over 0.8(n=7-13,r=0.81-0.92,P=9.0×10-4-0.020).Conclusion Tissue-specific cfDNA mutations may serve as novel potential biomarkers for non-invasive evaluation of exercise-related tissue injury.