Several epidemiological studies have indicated that an increased risk of depression was associated with long-term exposure to PM2.5. The objective of our two-sample Mendelian randomization study was to determine the causal relationship between long-term exposure to particulate matter 2.5 and the risk of depression. A two-sample Mendelian randomization study was performed based on GWAS summary data. Forty-six PM2.5-related single nucleotide polymorphisms were suitable for the analysis as instrumental variables. The random-effect model of inverse-variance weighted and the other four methods (weighted median, MR-Egger, Simple mode and weighted mode) were all performed for the analysis. Additionally, multivariate Mendelian randomization analysis was also completed. Our two-sample Mendelian randomization study indicated that exposure to particulate matter 2.5 has a significantly positive impact on the risk of depression (P = 0.026, random-effect model of inverse-variance weighted). After adjusting for smoking and body mass index in our multivariate Mendelian randomization analysis separately, the relationship between exposure to PM2.5 and the risk of depression remained significant. Based on current GWAS data, our study supplies potential evidence that long-term exposure to PM2.5 is a risk factor for depression. The improvement in air quality may be conducive to reducing the risk of depression.