[Fexolone inhibits neuronal ferroptosis through the Nrf2/HO-1/GPX4 pathway to alleviates sepsis-associated brain injury].

Objective: To observe the protective effect of Fisetin on sepsis-associated brain injury and explore its possible mechanism from the perspective of ferroptosis.

Methods: Sprague-Dawley (SD) rats (6-8-week-old male) were randomly divided into three groups: sham operation group (Sham group), colonic ligation and puncture (CLP) induced sepsis model group (CLP group) and Fisetin preprocessing group (CLP+Fisetin group), with 18 rats in each group (12 for observing survival rate and 6 for indicator testing). The CLP+Fisetin group was given Fisetin solution 50 mg×kg^-1×d^-1 by gavage continuously for 5 days before CLP, with dimethyl sulfoxide (DMSO) as the solute, while Sham group and CLP group were given the same dose of DMSO. The model was established at 2 hours after the last gavage. The general condition of each group of rats were observed, and the 10-day mortality were record. The behavioral testing (new object recognition experiment, elevated cross maze experiment) were performed after 7 days of modeling. After 24 hours of modeling, nerve reflex scoring was performed, and then the rats were euthanized and brain tissue was collected. The pathological changes of brain tissue were observed under a microscope by hematoxylin-eosin (HE) staining, the deposition of iron ion in brain tissue was observed by Prussian blue staining. The content of iron in brain tissue was determined by tissue iron kit, and the content of malondialdehyde (MDA) in brain tissue was determined by colorimetry. The expressions of tumor necrosis factor-α (TNF-α), neuron damage marker S100β, nuclear factor E2-related factor 2 (Nrf2), heme oxygenases-1 (HO-1) and glutathione peroxidase 4 (GPX4) were detected by Western blotting.

Results: On day 10 post-operation, 12, 3, and 7 animals survived in the Sham group, CLP group, and CLP+Fisetin group, respectively. Compared with the Sham group, rats in the CLP group showed significantly decreased nerve reflex score, new object discrimination index and open arm dwell time. HE staining showed arranged disorderly of neuronal cells, cytoplasm deep staining, nuclear condensation, unclear structures, neuron loss, and significant inflammation in the hippocampus in the hippocampus. Prussian blue staining showed iron ion deposition in the brain tissue. The contents of iron and MDA in brain tissue were elevated, and the expressions of TNF-α and S100β were up-regulated, while the expressions of Nrf2, HO-1, and GPX4 were down-regulated. Compared with the CLP group, the CLP+Fisetin group showed significantly increased neurological reflex score (7.33±1.15 vs. 4.67±1.53), improved new object discrimination index (0.44±0.02 vs. 0.32±0.04), and longer open arm dwell time (minutes: 78.33±9.29 vs. 41.15±9.64). Neuronal cells in the hippocampus were more organized, with less cytoplasmic staining, nuclear condensation, reduced neuronal loss, and fewer inflammatory cells. Iron ion deposition was reduced, and the contents of iron ions and MDA in brain tissue were decreased [iron ion (μg/g): 151.27±14.90 vs. 224.69±17.64, MDA (μmol/g): 470.0±44.3 vs. 709.3±65.4]. The expressions of TNF-α and S100β were significantly decreased (TNF-α/GAPDH: 0.651±0.060 vs. 0.896±0.022, S100β/GAPDH: 0.685±0.032 vs. 0.902±0.014), while the expressions of Nrf2, HO-1, and GPX4 were significantly increased (Nrf2/GAPDH: 0.708±0.108 vs. 0.316±0.112, HO-1/GAPDH: 0.694±0.022 vs. 0.538±0.024, GPX4/GAPDH: 0.620±0.170 vs. 0.317±0.039). All differences were statistically significant (all P < 0.05).

Conclusion: Fisetin pretreatment can inhibit ferroptosis and reduce sepsis-associated brain injury by Nrf2/HO-1/GPX4 pathway.