Objective To investigate the role and underlying mechanism of circular RNA tousled-like kinase 1(circTLK1) in myocardial ischemia/ reperfusion injury (MIRI). Methods Forty-eight C57BL/6 mice were divided into a sham operation group (Sham), MIRI group, sh-NC group, sh-circTLK1 group, sh-circTLK1+antagomir-NC group, and sh-circTLK1+antagomir-miR-26a-5p group with eight mice per group. The MIRI mouse model was established by ligation of the left anterior descending coronary artery. Left ventricular ejection fraction, left ventricular fractional shortening, left ventricular end-diastolic diameter, and end-systolic diameter were measured by echocardiography to assess cardiac functions. Serum levels of lactate dehydrogenase, cardiac troponin I and creatine kinase isoenzyme were measured by ELISA. Histopathological changes of myocardial tissue were assess by hematoxylin-eosin staining. Cardiomyocyte apoptosis was detected by TUNEL staining. mRNA and protein expression of circTLK1, miR-26a-5p, and YES1 in myocardial tissue were measured by real-time quantitative PCR and Western blot. An interaction between circTLK1/ YES1 and miR-26a-5p was verified by dual luciferase reporter assays and RNA-binding protein immunoprecipitation. Results Compared with the findings in the Sham group, the ejection fraction, fractional shortening, and the level of myocardial tissue miR-26a-5p in the MIRI group were significantly decreased, the left ventricular end-diastolic diameter, end-systolic diameter, activities of serum lactate dehydrogenase and creatine kinase isoenzyme, the level of cardiac troponin I, myocardial cell apoptosis rate, and mRNA and protein levels of circTLK1 and YES1 in myocardial tissue were significantly increased (all P<0. 05). Moreover, cardiomyocytes were disordered, and inflammatory cell infiltration was observed among interstitial cells. circTLK1 knockdown significantly upregulated miR-26a-5p expression, inhibited YES1 expression, improved the changes in the above indicators (all P<0. 05), and reduced myocardial injury. Inhibition of miR-26a-5p significantly attenuated the protective effect of circTLK1 knockdown against MIRI by upregulating YES1 expression. Dual luciferase reporter assays and RNA-binding protein immunoprecipitation confirmed the direct interaction between circTLK1 and miR-26a-5p, and YES1 was a target of miR-26a-5p. Conclusions Knockdown of circTLK1 may exert a protective effect against MIRI by regulating the miR-26a-5p / YES1 axis, and circTLK1 may be a potential therapeutic target for MIRI.