Objective To determine the effect of irisin on improving exertional heat stroke (EHS) and explore related mechanisms. Methods One hundred and nineteen wild‑type male C57BL/6 mice were used as research subjects. ① Divide 30 mice into 2 groups according to the random number table method (n=15): the SHE group and the EHS group. After 6 days of routine training and 2 days of rest, the SHE group ran at room temperature at (25.0±1.0) ℃ and (40±5)% humidity; after 6 days of routine training and 2 days of rest, the EHS group ran in a constant temperature and high humidity environment at (38.0±1.0) ℃ and (70±5)% humidity to create a model. Record the core body temperature, onset and recovery time, and 7‑day survival rate of each group of mice. Observe the pathological changes in brain, liver, and kidney tissues using H‑E staining and calculate the number of surviving cells using Nissl staining. Conduct behavioral experiments to detect autonomous activity and exploration ability. ② Divide 18 mice into 3 groups according to the random number table method (n=6): SHE group, EHS group, and heat acclimation (HA) group. The modeling methods for SHE group and EHS group were shown in ①. The HA group underwent 14 days of running training in a constant temperature and high humidity environment exercise chamber at (35.0±1.0) ℃ and (70±5)% humidity. Enzyme‑linked immunosorbent assay (ELISA) method was used to detect the serum concentration of irisin in each group of mice. ③ Divide 63 mice into 3 groups according to the random number table method (n=21): physiological saline+EHS group (NS+EHS group), irisin+EHS group (Irisin+EHS group), and HA+EHS group. The NS+EHS group received saline injection 7 days prior to EHS modeling. The Irisin+EHS group received irisin injection starting 7 days before modeling; The Irisin+EHS group received irisin injection starting 7 days before modeling. Record core body temperature, onset and recovery time, 7‑day survival rate, observe tissue pathological changes using H‑E staining, and conduct behavioral experiments to test autonomous activity and exploration ability. ④ Divide 8 mice into 2 groups according to the random number table method (n=4): NS+EHS group and Irisin+EHS group (the intervention methods are shown in ③). The NS+EHS group and Irisin+EHS group were used to detect the expression of c‑Fos in neurons using immunofluorescence staining at 30 min and 60 min of modeling, respectively. Electrophysiological detection of frequency and amplitude changes in excitatory postsynaptic current (EPSC) of thermosensitive neurons in the hypothalamic preoptic area (POA) before and after treatment with irisin. Results Compared with the SHE group, the EHS group showed increased core body temperature, and presented heat stroke within an average of 70 min, with a reduced 7‑day survival rate (P<0.05); showed pathological damage in various tissues, with a decreased number of survival cells (P<0.05); decreases in the number/time of entering the central area of the open field (P<0.05) and decreases in the time and percentages of entering the elevated cross arm (P<0.05). Compared with the SHE and EHS group, the levels of serum irisin in the HA group increased (P<0.05). Compared with the NS+EHS group, the Irisin+EHS group and the HA+EHS group showed a 60‒80 min decrease in core body temperature, prolonged onset of EHS，decreased in recovery time, and improved in 7‑day survival rate (P<0.05); showed relieved pathological damage in various tissues, increases in the number/time of entering the central area of the open field (P<0.05) and increases in the time and percentages of entering the elevated cross arm (P<0.05). There were no statistical difference in each indicator between the Irisin+EHS group and the HA+EHS group (P>0.05). For the NS+EHS group, compared with modeling for 30 min, 60 min of modeling resulted in decreases in the levels of c‑Fos in the POA region (P<0.05). For the Irisin+EHS group, compared with modeling for 30 min, 60 min of modeling resulted in increases in the levels of c‑Fos (P<0.05). Compared with those before irisin treatment, irisin treatment resulted in decreases in the frequency and amplitude of EPSC in POA thermosensitive neurons (P<0.05). Conclusions Irisin can stabilize the electrical activity of POA thermosensitive neurons, reduce their excitatory damage, improve EHS survival rate, and improve brain function, simulating the protective effect of HA on EHS.