Background and Aims A large number of studies have been carried out on the regulatory action of adenosine monophosphate-activated protein kinase (AMPK) on circulation system, and it is confirmed that AMPK can modulate the relaxation and contraction of arterioles by regulating the concentration of free calcium ions in the vascular smooth muscle cells (VSMCs). Previous studies found that AMPK can help the arterial relaxation through increasing cofilin activity that can cause the depolymerization of cytoskeletal protein filamentous actin (F-actin) into the monomer, globular actin (G-actin). Therefore, this study was conducted to further investigate the molecular mechanism for AMPK-induced cofilin activation.Methods The second- and third-order branches of the superior mesenteric artery of C57BL6/N mice were harvested. Using the pressure myograph model, the responses of the isolated arteries to epinephrine and acetylcholine were determined, and the vessels with a contraction rate >30% and a relaxation rate >90% were used for the experiment. Then, the vessels were divided into two groups and added with AMPK activator PT1 (PT1 group) or vehicle DMSO (control group) after pretreatment and pre-contraction with high potassium MOPS solution. The vascular relaxation responses of the two groups of vessels were compared. After that, the expressions of phosphorylated AMPK (p-AMPK) and other relevant downstream proteins of AMPK in the vascular tissues of the two groups were detected by Western blot or immunofluorescent Western blot.Results The vascular activity of all experimental arteries met the requirements. There were no significant differences in contraction and relaxation abilities between the two groups of vessels (both P>0.05), and there were no significant differences in blood vessel diameters during pretreatment and pre-contraction between the two groups of vessels (both P>0.05). After corresponding treatment, the blood vessels in PT1 group were gradually relaxed, while those in control group showed no obvious change, and 60 min later, the average vessel diameter in PT1 group expanded to (196.6±11.5) μm, while it in control group was (136.1±8.1) μm, and the difference had statistical significance (P<0.001). In PT1 group compared with control group, the expressions of p-AMPK and G-actin in vascular tissue were increased, which were (3.25±0.52) and (2.26±0.64) folds of those in control group, while the expression of phosphorylated cofilin (p-cofilin) was decreased, which was (0.48±0.19) folds of that in control group, and the differences had statistical significance (all P<0.05); the expression of total heat shock protein 20 (t-HSP20) showed no obvious change (P>0.05), but the expression of phosphorylated HSP20 (p-HSP20) was increased, which was (2.45±0.52) folds of that in control group, and the difference had statistical significance (P<0.001).Conclusion The results of this study suggest that HSP20 participates in assisting AMPK activating the cofilin, and the mechanism may be associated with the activated AMPK increasing the level of p-HSP20 that competes the binding site of p-cofilin, and then causes its dephosphorylation and increased activity, and thereby reducing cytoskeletal actin homeostasis and promoting vascular relaxation.