Abstract: Objective: Pyrazinamide (PZA), an important first-line drug for tuberculosis, can cause liver injury and intestinal flora disorder in patients. In this study, we intended to explore the effect of dietary probiotic (Lactobacillus casei, LcS) supplementation on PZA-induced liver injury and intestinal flora disorder in rats. Methods: A total of 40 adult male SD rats were randomly divided into 4 groups: normal control group (NC), PZA group (L0), low-dose LcS group (L1) and highdose LcS group (L2). Liver injury was induced in rats in the L0, L1 and L2 groups, and those in the L1 and L2 groups were subjected to LcS (108 CFU/mL) intervention at a dose of 10 and 20 mL/(kg·d) by gavage daily for 10 consecutive weeks. Hematoxylin-eosin (HE) staining was used to observe the pathological changes of liver tissues in rats. The serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were detected. Quantitative analysis of Bifidobacteria, Lactobacillus and Escherichia coli in rat feces was conducted by real-time fluorescence quantitative polymerase chain reaction (qPCR) based on 16S rDNA V3 variable region. Results: After 10 weeks of treatment with pyrazinamide, HE staining of liver tissues revealed that liver cells showed moderate edema and marked ballooning degeneration with the disappearance of hepatic cord accompanied by the occurrence of inflammatory cell infiltration. The pathological score was 3.20 points, and serum ALT and AST levels were significantly increased to 95.90 and 188.60 U/L compared with those in the control group (73.90 and 139.20 U/L), suggesting successful establishment of the liver injury model. After LcS intervention at different doses for 10 weeks, examination of liver tissue sections showed the structure of hepatic lobules was obviously improved compared with the PZA group and pathological score as well as serum ALT and AST levels were reduced to normal. The quantities of 3 representative intestinal strains were statistically different between different groups and changed with prolonged intervention time. At the end of the sixth and the tenth week of intervention, the amounts of Bifidobacteria and Lactobacillus in the high-dose LcS group were significantly increased compared with the PZA group, while the number of Escherichia coli was significantly decreased (P < 0.05). At the end of the tenth week, the amounts of Bifidobacteria and Lactobacillus in the high-dose LcS group were significantly increased by 1.18 and 1.03 times compared with those before intervention, respectively. Conclusions: LcS time- and dose-dependently has a protective effect on liver injury and intestinal flora disturbance induced by pyrazinamide. The underlying mechanism may be related to the maintenance of intestinal homeostasis and the modulation of the intestinal microbiota in the presence of LcS.

Key words: pyrazinamide, probiotics, liver injury, gastrointestinal microbiome