A Versatile New Model of Chemically Induced Chronic Colitis Using an Outbred Murine Strain

Abstract

Murine colitis models are crucial tools for understanding intestinal homeostasis and inflammation. However, most current models utilize a highly inbred strain of mice, and often only one sex is employed to limit bias. This targeted approach, which in itself is biased, means that murine genetic diversity and sex-related differences are ignored, making it even more difficult to extend findings to humans, who are highly heterogeneous. Furthermore, most models do not examine the chronic form of colitis, an important fact taking into account the chronic nature of the inflammatory bowel diseases (IBD). Here, we attempted to create a more realistic murine colitis model by addressing these three issues. Using chemically induced chronic colon inflammation in an outbred strain of mice (RjOrl:SWISS [CD-1]), we (i) mimicked the relapsing nature of the disease, (ii) better represented normal genetic variability, and (iii) employed both female and male mice. Colitis was induced by intrarectal administration of dinitrobenzene sulfonic acid (DNBS). After a recovery period and 3 days before the mice were euthanized, colitis was reactivated by a second administration of DNBS. Protocol length was 24 days. Colitis severity was assessed using body mass, macroscopic scores, and histological scores. Myeloperoxidase (MPO) activity, cytokine levels, and lymphocyte populations were also characterized. Our results show that the intrarectal administration of DNBS effectively causes colitis in both female and male CD-1 mice in a dose-dependent manner, as reflected by loss of body mass, macroscopic scores and histological scores. Furthermore, colon cytokine levels and mesenteric lymph node characteristics indicate that this model involves immune system activation. Although some variables were sex-specific, most of the results support including both females and males in the model. Our ultimate goal is to make this model available to researchers for testing candidate anti-inflammatory agents, such as classical or next-generation probiotics; we also aim for the results to be more easily transferrable to human trials.

Keywords: CD-1 mice; DNBS; colitis; gut inflammation; murine IBD model.

FIGURE 1

Experimental design (A) and key methodological steps (B). Mice were anesthetized with an intraperitoneal (i.p.) injection of ketamine and xylazine (B1). The DNBS solution or the control solution was administered by injection, using a tuberculin syringe and a flexible plastic tube inserted 3.5 cm into the colon (B2). Following the treatment, all mice received a subcutaneous injection of 1 ml of saline solution to prevent dehydration (B3); they were kept in a horizontal position until they awoke (B4).

FIGURE 2

Survival rate (A) and body mass trends in CD-1 (B) and Black-6 (D) mice, and loss of body mass after second DNBS injection (C). For the survival rate analysis Logrank test (Mantel Cox) was performed. For weight curves, a multiple unpaired T-test was performed per day with fewer assumptions corrected for multiple comparison with Holm–Sidak method, (^∗) indicates significance vs. vehicle group and (#) significance between female and male individuals in DNBS treated groups. n = 10; p < 0.05. For the weight loss analyses, due to the lack of uniform variances when included the vehicle groups, two-way ANOVA was performed only in inflamed groups with strain and sex as factors followed by a Tukey test (results indicated as ^∗). In order to compare the effect of the DNBS vs. the vehicle groups, a non-parametric Kruskal–Wallis test followed by a Dunn’s test was performed inside CD-1 and Black 6 groups separately (results indicated as +). n = 10; ^∗p < 0.05, ^∗∗∗∗p < 0.0001, ⁺p < 0.05, ⁺⁺p < 0.01, ⁺⁺⁺⁺p < 0.0001. The black arrows indicate the moment when mice started to recover weight after the first DNBS injection. B6M, Black-6 males; B6F, Black-6 females; CD1M, CD-1 males; CD1F, CD-1 females; B6 M+F, Black-6 mice; CD-1 M+F, CD-1 mice.

FIGURE 3

Macroscopic (A) and histological scores (B) and representative images of Black-6 mice (C) and CD-1 mice (D). As both scores do not follow a Gaussian distribution, in order to compare the effect of the DNBS vs. the vehicle groups, a non-parametric Kruskal–Wallis test followed by a Dunn’s test was performed inside CD-1 and Black 6 groups separately (results indicated as ^∗). The same test was performed for testing differences among inflamed groups (results indicated as +). n = 10; ^∗∗∗∗p < 0.0001, ⁺p < 0.05. The black arrows indicate eosinophils. B6M, Black-6 males; B6F, Black-6 females; CD1M, CD-1 males; CD1F, CD-1 females; B6 M+F, Black-6 mice; CD-1 M+F, CD-1 mice.

FIGURE 4

Response of CD-1 mice to colitis induced by different doses of DNBS. Body mass trends (A), loss of body mass after the first DNBS injection and the second DNBS injection (B), macroscopic (C) and histological scores (D) and in vivo permeability (E). For weight curves, a multiple unpaired T-test was performed per day with fewer assumptions corrected for multiple comparison with Holm–Sidak method, (^∗) indicates significance vs. vehicle group. n = 10; p < 0.05. No Gaussian data comparisons (loss of body mass and macro and histological scores) were performed using a non-parametric Kruskal–Wallis test followed by a Dunn’s test (results indicated as ^∗). For the permeability analyses, two-way ANOVA was performed with dose and sex as factors followed by a Tukey test (results indicated as +). CD-1 males (M, in blue) and CD-1 females (F, in orange). n = 10. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ⁺p < 0.05.

FIGURE 5

Colon levels of cytokines induced by different doses of DNBS in CD-1 mice. Analyses were performed by two-way ANOVA with dose and sex as factors followed by a Tukey test. CD-1 males (M, in blue) and CD-1 females (F, in orange) injected with vehicle (v), 1.5 mg (1.5), 2.5 mg (2.5), or 3.5 mg (3.5) of DNBS. n = 10; ⁺p < 0.05, ⁺⁺p < 0.01, ⁺⁺⁺p < 0.001.

FIGURE 6

Serum levels of cytokines induced by different doses of DNBS in CD-1 mice. Analyses were performed by two-way ANOVA with dose and sex as factors followed by a Tukey test. CD-1 males (M, in blue) and CD-1 females (F, in orange) injected with vehicle (v), 1.5 mg (1.5), 2.5 mg (2.5), or 3.5 mg (3.5) of DNBS. n = 10; ⁺p < 0.05, ⁺⁺p < 0.01, ⁺⁺⁺p < 0.001.

FIGURE 7

Levels of lymphocytes induced by different doses of DNBS in CD-1 mice. Lymphocyte populations were characterized using flow cytometry. Analyses were performed by two-way ANOVA with dose and sex as factors followed by a Tukey test. CD-1 males (M, in blue) and CD-1 females (F, in orange) injected with vehicle (v), 1.5 mg (1.5), 2.5 mg (2.5), or 3.5 mg (3.5) of DNBS. n = 10; ⁺p < 0.05, ⁺⁺p < 0.01, ⁺⁺⁺p < 0.001.