ObjectiveTo investigate the mechanism of Buyang Huanwutang in treating diabetic peripheral neuropathy （DPN） via mitochondrial transport. MethodDiabetes in SD rats was induced by a high-carbohydrate/high-fat diet and intraperitoneal injection of streptozotocin （STZ）. The 45 diabetic rats were randomly assigned into a DPN group， an alpha-lipoic acid （60 mg·kg^-1·d^-1） group， and a Buyang Huanwutang （15 g·kg^-1·d^-1） group， with 15 rats in each group. Fifteen normal SD rats were fed with the standard diet and set as the control group. The rats were administrated with corresponding drugs by gavage for 12 weeks. The paw withdraw threshold （PWT） and motor nerve conduction velocity （MNCV） were measured at the end of medication， and the sciatic nerve and the bilateral dorsal root ganglia of L4-5 were collected. The injury model of NSC34 cells was established by treating with 50 mmol·L^-1 glucose and 250 μmol·L^-1 sodium palmitate. The NSC34 cells were then randomly assigned into a blank （10% blank serum） group， a DPN （10% blank serum） group， an apha-lipoic acid （10% apha-lipoic acid-containing serum） group， a Buyang Huanwutang （10% Buyang Huanwutang-containing serum） group， and a Buyang Huanwutang + Compound C （CC） （10% Buyang Huanwutang-containing serum + 10 μmol·L^-1 CC） group. The cell intervention lasted for 24 h. The immunofluorescence method， immunohistochemistry， and Western blot were employed to determine the expression levels of phosphorylated adenosine monophosphate-activated protein kinase （p-AMPK）， phosphorylated cAMP-response element binding protein （p-CREB）， kinesin family member 5A （KIF5A）， and dynein cytoplasmic 1 intermediate chain 2 （DYNC1I2）. ResultCompared with the control group， the DPN group of rats showed increased fasting blood glucose （P<0.01）， decreased MNCV and PWT （P<0.01）， down-regulated expression of KIF5A， p-AMPK/AMPK， and p-CREB/CREB （P<0.01）， and up-regulated expression of DYNC1I2 （P<0.01）. Compared with the DPN group， drug intervention groups showed increased MNCV and PWT （P<0.01）， up-regulated expression of KIF5A， p-AMPK/AMPK， and p-CREB/CREB （P<0.05， P<0.01）， and down-regulated expression of DYNC1I2 （P<0.05， P<0.01）. The Buyang Huanwutang group had higher levels of MNCV and KIF5A （P<0.05） and lower level of DYNC1I2 （P<0.01） than the apha-lipoic acid group. Compared with the blank group， the DPN group of NSC34 cells showed decreased levels of KIF5A， p-AMPK/AMPK， and p-CREB/CREB （P<0.01） and increased level of DYNC1I2 （P<0.01）. The apha-lipoic acid group and Buyang Huanwutang group had higher levels of KIF5A， p-AMPK/AMPK， and p-CREB/CREB （P<0.05， P<0.01） and lower level of DYNC1I2 （P<0.01） in NSC34 cells than the DPN group. Buyang Huanwutang group had higher KIF5A level （P<0.05） in NSC34 cells than the apha-lipoic acid group. Moreover， the Buyang Huanwutang + CC group had lower levels of KIF5A， DYNC1I2， p-AMPK/AMPK， and p-CREB/CREB （P<0.01） in NSC34 cells than the Buyang Huanwutang group. ConclusionBuyang Huanwutang may regulate mitochondrial anterograde transport via the AMPK/CREB pathway to prevent and treat DPN.