Establishment and evaluation of a specific antibiotic-induced inflammatory bowel disease model in rats

Physical and chemical methods for generating rat models of enteritis have been established; however, antibiotic induction has rarely been used for this purpose. The present study aimed to establish and evaluate a rat model of inflammatory bowel disease (IBD) using antibiotics. A total of 84 Sprague-Dawley (SD) rats were divided into the following groups, according to the dosage and method of administration of the antibiotics: A, control; B, low-dose clindamycin; C, medium-dose clindamycin; D, high-dose clindamycin; E, low-dose clindamycin, ampicillin and streptomycin; F, medium-dose clindamycin, ampicillin and streptomycin; and G, high-dose clindamycin, ampicillin and streptomycin. Antibiotic administration was stopped on day 7; the modeling period covered days 1–7, and the recovery period covered days 8–15. Half of the animals were dissected on day 11, with the remaining animals dissected on day 15. Food and water intake, body weight and fecal weight were recorded. Intestinal flora was analyzed via microbial culture and quantitative PCR. The content of TNF-α, IL1-β, IL-6 and C-reactive protein (CRP) was assessed in abdominal aorta blood. Colonic and rectal tissues were examined pathologically via hematoxylin-eosin staining to assess leukocyte infiltration and intestinal mucosal changes as indicators of inflammation. Rat weight, food intake, water intake and 2-h fecal weight were significantly different across the experimental groups (P = 0.040, P = 0.016, P<0.001 and P = 0.009, respectively). Microbial cultures revealed no significant differences between group A and B,C (P = 0.546,0.872) but significant differences betwenn group A and the other experimental groups (all P<0.001). Furthermore, significant differences in the levels of Bacteroides, Faecalibacterium prausnitzii and Dialister invisus on day 4 between groups A, C and F (P = 0.033, P = 0.025 and P = 0.034, respectively). Significant differences were detected in the levels of TNF-α, IL1-β, IL-6 and CRP between the groups (all P<0.001). The colonic and rectal pathological inflammation scores of the experimental groups were significantly different compared with group A (B vs. A, P = 0.002; others, all P<0.001). These findings indicated that an antibiotic-induced IBD model was successfully established in SD rats; this animal model may serve as a useful model for clinical IBD research.