It has been well documented that both gene and environment contribute to and usually interact to affect the development of depression. Despite the fast accumulating evidence for G × E on depression, there are some important issues are less well understood. First, the majority of studies on the G × E have focused on genes of serotonergic and dopaminergic systems which were guided by the monoamine-deficiency hypothesis. However, genes involved in the function of hypothalamic–pituitary–cortisol (HPA) also may underlie depression. Second, compared with identifying genetic variations, the measurement of SLE is readily overshadowed by molecular genetic techniques. Researchers have measured participants’ life stress across varying time periods. As a consequence of these inadequacies, mixed findings regarding G × E were obtained. Third, the observed G × E effects could possibly be an artifact of gene–environment correlation (rGE), but fewer studies sought to control for it. In this study, one of the most widely studied functional polymorphism (Bcl1) in the glucocorticoid receptor gene (NR3C1) was used to test G × E effects. To our knowledge, a few studies investigated the interaction between Bcl1 polymorphism and adversities, however, mixed findings were obtained. Besides, the rGE effects should be ruled out before solid conclusions can be drawn. Therefore, SLEs were differentiated as independent SLEs if they had a nonsignificant genetic underlying as prior study. In addition, we also hypothesized that, if the measure of SLEs represents genetic risk, then SLEs would interact with gene even if they occurred after the depression. One thousand and eighty one adolescents (mean age 15.23 years at T3) were assessed three times with an interval of three years. During each assessment, the participants completed self-reported questionnaire on depressive symptoms. At the second assessment, a self-reported SLE questionnaire was completed. All measures showed good reliability. DNA was extracted from saliva. Genotyping at Bcl1 polymorphism was performed for each participant in real time with MassARRAY RT software version 3.0.0.4 and analyzed using the MassARRAY Typer software version 3.4 (Sequenom). To examine whether Bcl1 polymorphism interacts with SLE in predicting depression and whether it is a true G × E effect after controlling for rGE, hierarchical regression analyses were conducted. The findings indicated that no main effect of Bcl1 polymorphism on depression was found. SLE was associated with an increase in depressive symptoms. Both total SLEs by Bcl1 polymorphism interaction and independent SLEs by Bcl1 polymorphism interaction were significantly predicted T3 depression. C allele carriers were more susceptible to the detrimental effects of SLEs. However, such an interactive effect was not observed when predicting T1 depression. These findings highlight there is a true G × E effect underlying the observed interaction between Bcl1 polymorphism and SLEs on depression rather than an artifact of rGE. Besides, it’s important to test and report rGE in G × E studies in future.