Objective  To explore the inhibitory effect of berberine （BBR） on contrast-induced nephropathy （CIN） and clarify the underlying mechanism.

Methods  Sprague-Dawley （SD） rats were utilized for constructing a rat CIN model. BBR treatment group （M+B） received a gavage of 200 mg/kg BBR while control group （Con） and model group （Mod） had an oral intake of the same volume of normal saline. Human kidney cortex proximal tubule epithelial cells （HK-2） were utilized for constructing a CIN cell model in vitro. No special treatment was offered to HK-2 cells in Con group. HK-2 cells in Mod group were treated with 50 g/L ioversol for 24 h and HK-2 cells in M+B group 50 g/L ioversol and 30 μmol/L BBR for 24 h. Akt agonist （SC79） and Akt inhibitor （MK2206） were utilized for confirming the effect of BBR on protein kinase B-forkhead box O3-nuclear factor erythroid 2-related factor 2 （Akt-Foxo3a-Nrf2） regulatory axis. Serum creatinine （Scr） detection and hematoxylin-eosin （HE） stain were employed for evaluating renal injury in rats. Annexin V-propidium iodide （annexin V-PI） stain was used for detecting cell apoptosis. And Western blot was utilized for detecting the protein expression of Akt-Foxo3a-Nrf2 axis.

Results  The level of Scr was higher in Mod group than that in Con group （58.83 ± 7.96）μmol/L vs （23.52 ± 0.78）μmol/L, P＜0.05. And it was lower in M+B group than that in Mod group （39.64 ± 2.52）μmol/L vs （58.83 ± 7.96）μmol/L, P＜0.05. HE stain indicated that kidney morphology was normal and renal tubular epithelial cells remained intact in Con group. In Mod group, structure of renal tubules was disarrayed with diffuse death of epithelial cells and obvious vacuolar degeneration of renal tubular epithelial cells. In M+B group, morphology of renal tubules was partially restored and vacuolar degeneration of renal tubule epithelial cells and cell death lessened. Western blot results indicated that, as compared with Con group, the expression of phospho-protein kinase B （p-Akt） was up-regulated in CIN rats and ioversol treated HK-2 cells. And the difference was statistically significant （P＜0.05）. In M+B group, the expression of p-Akt dropped as compared with Mod group （P＜0.05）. Annexin V-PI apoptosis assay indicated that, as compared with Con group, cell apoptosis spiked in Mod group （12.65 ± 1.25）% vs （27.44 ± 0.73）%, P＜0.05. As compared with Mod group, cell apoptosis declined in M+B group （27.44 ± 0.73）% vs （24.81 ± 0.49）%, P＜0.05. As compared with M+B group, cell apoptosis rose in BBR plus SC79 treatment group （24.81 ± 0.49）% vs （27.50 ± 0.35）%, P＜0.05. As compared with Mod group, cell apoptosis declined in MK2206 treatment group （27.65 ± 0.56）% vs （25.20 ± 0.64）%, P＜0.05. As compared with M+B group, MK2206 plus BBR further reduced ioversol-induced HK-2 cell apoptosis （24.51 ± 0.52）% vs （21.24 ± 0.99）%, P＜0.05.

Conclusion  BBR may suppress CIN in vivo and in vitro through regulating Akt-Foxo3a-Nrf2 axis in a p-Akt dependent mode.