Abstract: Objective To explore the causal relationship between brain-derived neurotrophic factor (BDNF) and depression through the two-sample Mendelian randomization(TSMR). Methods The open field test (OFT) and forced swimming test (FST) were used to detect the behavior of the control group and the chronic unpredictable mild stress (CUMS) group. Western blot was used to measure the expression of BDNF in the hippocampus of mice. The single-nucleotide polymorphisms (SNPs) closely related to BDNF were extracted from the whole genome-wide association study of BDNF as instrumental variables. The causal relationship between BDNF and depression was evaluated using the MR-Egger regression, weighted median (WM¹), weighted model (WM²) and inverse variance weighting (IVW) methods of TSMR. Results The CUMS group showed remarkable shorter exploration time in the central area of OFT, longer time of immobility in FST, and a lower level of BNDF protein in the hippocampus than the control group. According to TSMR, no causal relationship was found between BDNF and depression (IVW, OR=1.007, 95%CI=0.962~1.054, P=0.760; WM¹, OR= 1.011, 95%CI=0.985~1.038, P=0.411 and WM², OR= 1.028, 95%CI=0.996~1.026, P=0.173). However, through IVW analysis, significant heterogeneity was observed among the three SNPs (Q = 10.4972, P=0.005 2). Conclusions Treatment with CUMS resulted in depression-like behavior. The expression of hippocampal BDNF protein was down-regulated. SNPs as a tool variable predicts that there is no causal relationship between BDNF and depression.