Both epidemiological and experimental studies increasingly show that exposure to ambient fine particulate matter (PM2.5) is related to the occurrence and development of chronic diseases, such as metabolic diseases. However, whether PM2.5 has "exposure memory" and how these memories affect chronic disease development like hepatic metabolic homeostasis are unknown. Therefore, we aimed to explore the effects of exposure transition on liver cholesterol and bile acids (BAs) metabolism in mice. In this study, C57BL/6 mice were exposed to concentrated ambient PM2.5 or filtered air (FA) in a whole-body exposure facility for an initial period of 10 weeks, followed by another 8 weeks of exposure switch (PM2.5 to FA and FA to PM2.5) comparing to non-switch groups (FA to FA and PM2.5 to PM2.5), which were finally divided into four groups (FF of FA to FA, PP of PM2.5 to PM2.5, PF of PM2.5 to FA, and FP of FA to PM2.5). Our results showed no significant difference in food intake, body composition, glucose homeostasis, and lipid metabolism between FA and PM2.5 groups after the initial exposure before the exposure switch. At the end of the exposure switch, the mice switched from FA to PM2.5 exposure exhibited a high sensitivity to late-onset PM2.5 exposure, as indicated by significantly elevated hepatic cholesterol levels and disturbed BAs metabolism. However, the mice switched from PM2.5 to FA exposure retained a certain memorial effects of previous PM2.5 exposure in hepatic cholesterol levels, cholesterol metabolism, and BAs metabolism. Furthermore, 18-week PM2.5 exposure significantly increased hepatic free BAs levels, which were completely reversed by the FA exposure switch. Finally, the changes in small heterodimeric partner (SHP) and nuclear receptor subfamily 5 group A member 2 (LRH1) in response to exposure switch mechanistically explained the above alterations. Therefore, mice switching from PM2.5 exposure to FA showed only a weak memory of prior PM2.5 exposure. In contrast, the early FA caused mice to be more susceptible to subsequent PM2.5 exposure.