Objective  To observe the effects of astragaloside （Ast）-incubated human adipose derived stem cell （hADSC） on diabetic kidney disease （DM） after caudal vein transplantation. examine the repair status of diabetic kidney disease （DKD） on renal injury in rats and elucidate its possible mechanism.

Methods  A total of 21 Sprague-Dawley rats were randomized into four groups of normal （n = 5）, model （n = 6）, hADSC （n = 5） and Ast-hADSC （n = 5）. DKD model was established in the last 3 groups. Both hADSC and Ast-hADSC groups were injected 6 times with hADSC or Ast-hADSC suspension through tail vein once biweekly. The model group received the same volume of normal saline. Renal function, blood lipid, urinary microalbumin/urinary creatinine and urinary protein quantification were detected. Renal pathology was observed after hematoxylin-eosin and Masson stains and renal colonization of hADSC by confocal laser. Immunohistochemistry and immunoblot were utilized for observing the voltage-dependent anion channel （VDAC） and microtubule-associated protein-light-chain-3. LC3Ⅱ/LC3Ⅰ, autophagy substrate P62 protein, phosphatase and tensin homology induced putative kinase 1 （Pink1） and E3 ubiquitin ligase （Parkin） protein expression. Mitochondria and autophagosomes were observed by transmission electron microscopy.

Results  （1） As compared with model group, total serum cholesterol （2.65 ± 0.04） mmol/L, （2.30 ± 0.29） mmol/L vs （3.31 ± 0.91） mmol/L, ratio of urinary microalbumin/urinary creatinine （58.91 ± 11.86） mg, （44.00 ± 23.82） mg vs （101.03 ± 46.88） mg and 24 h urinary protein quantity （321.18 ± 88.71） mg, （136.35 ± 66.01） mg vs （408.67 ± 113.95） mg （P＜0.05） declined in hADSC and Ast-hADSC groups. Body weight gained （240.20 ± 26.43） g, （274.00±17.89） g vs （193.60 ± 18.57） g （P＜0.05）. The effects of body weight and 24 h urinary protein were more significant in Ast-hADSC group than those in hADSC group （P＜0.05）. （2） As compared with model group, no significant differences existed in triglyceride, low density lipoprotein, urea nitrogen or high density lipoprotein in hADSC and Ast-hADSC groups （P＞0.05）. （3） Both hADSC and Ast-hADSC groups had lesser renal injury while renal index and glomerular cross-sectional area spiked （P＜0.05）; Improvement was more significant in Ast-hADSC group. （4） Under laser confocal microscope, minimal red fluorescence was observed in hADSC and Ast-hADSC groups. （5） Immunohistochemistry and immunoblotting revealed that protein expressions of LC3, LC3Ⅱ/LC3Ⅰ, Pink1 and Parkin spiked in hADSC and Ast-hADSC groups （P＜0.05） while protein expression of P62 declined （P＜0.05）. The effect of Ast-hADSCs group was more significant. （6） In model group, mitochondria were swollen, deformed and fragmented and the number of autophagosomes was fewer. Both hADSC and Ast-hADSC groups showed more autophagosomes and fewer swollen deformed mitochondria and mitochondrial fragments. The effect of Ast-hADSC group was more significant.

Conclusions  hADSC may restore mitochondrial autophagy in kidney tissue of DKD rats and reduce kidney injury. The effect is more significant after Ast incubation.