Objective To investigate the effects of short-chain fatty acids (SCFAs) on the cognitive function, cerebral cortex, and colon of mice chronically exposed to benzoapyrene (BaP), and to explore the underlying mechanisms.

Methods Thirty 8-week-old male ICR mice were randomly divided into control group, BaP group, and BaP+SCFAs group, with 10 mice in each group. The BaP group was intragastrically given BaP at a dose of 10.0 mg/kg every other day for 45 times, while the control group was given an equal volume of peanut oil, with free access to purified water for both groups. The BaP+SCFAs group was exposed to BaP at 10.0 mg/kg and at the same time, which was provided with a SCFAs solution (67.5 mmol/L sodium acetate, 40 mmol/L sodium butyrate, and 25.9 mmol/L sodium propionate) for drinking. After exposure, the variation of cognitive function of the mice was tested using the T-maze test, novel location recognition test, hole-board test, and pole test. The pathological changes in the cortex and colon were observed with HE staining. The expression levels of the inflammatory cytokines interleukin 1β (IL-1β) and tumor necrosis factor-α (TNF-α) in the cortex and colon were determined by enzyme-linked immunosorbent assay. The mRNA and protein expression levels of the tight junction proteins zonula occluden-1 (ZO-1) and occludin in the colon were measured using RT-PCR and Western blot, respectively.

Results Compared with the control group, the BaP group showed significantly decreased spatial memory and motor ability and significantly increased anxiety-like responses (P < 0.05), suggesting the occurrence of cognitive dysfunction; developed a large number of red neurons in the brain, disorganized intestinal cells, incomplete and sparse intestinal villi, and reduced and unevenly distributed goblet cells; showed significantly increased IL-1β levels (P < 0.05) and insignificantly increased expression of TNF-α in the cortex and colon (P>0.05); and showed significant reductions in the mRNA and protein expression of ZO-1 and occludin in the colon (P < 0.05). Compared with the BaP group, the BaP+SCFAs group showed significantly improved spatial memory and motor ability and significantly resolved anxiety-like responses (P < 0.05), suggesting that SCFAs could significantly alleviate BaP-induced cognitive dysfunction in mice; at the same time, SCFAs significantly reduced the number of red neurons in the brain, recovered the number of goblet cells in the colon, significantly decreased IL-1β levels in the cortex and colon and TNF-α levels in the cortex (P < 0.05), and significantly upregulated the mRNA and protein expression of ZO-1 and occludin in the colon (P < 0.05).

Conclusion SCFAs treatment can effectively improve BaP-induced cognitive dysfunction in mice, which may be through increasing the mRNA and protein expression of tight junction proteins in the colon, reducing intestinal permeability, promoting the repair of the intestinal mucosal barrier, and thus alleviating intestinal and neural inflammation.