OBJECTIVE:

To investigate the involvement of the high mobility group box protein B1 (HMGB1)-Toll-like receptor 2 (TLR2)/TLR4-nuclear factor κB (NF-κB) pathway in the intestinal mucosal injury induced by Cryptosporidium parvum infection, and to examine the effect of oxymatrine (OMT) on C. parvum infection in mice.

METHODS:

Forty SPF 4-week-old BALB/c mice were randomly divided into four groups, including the control group, infection group, glycyrrhizin (GA) group and OMT group. Each mouse was orally administered with 1 × 105 C. parvum oocysts one week in the infection, GA and OMT groups following dexamethasone-induced immunosuppression to model C. parvum intestinal infections in mice. Upon successful modeling, mice in the GA group were intraperitoneally injected with GA at a daily dose of 25.9 mL/kg for successive two weeks, and animals in the OMT group were orally administered OMT at a daily dose of 50 mg/kg for successive two weeks, while mice in the control group were given normal food and water. All mice were sacrificed two weeks post-treatment, and proximal jejunal tissues were sampled. The pathological changes of mouse intestinal mucosal specimens were observed using hematoxylin-eosin (HE) staining, and the mouse intestinal villous height, intestinal crypt depth and the ratio of intestinal villous height to intestinal crypt depth were measured. The occludin and zonula occludens protein 1 (ZO1) expression was determined in mouse intestinal epithelial cells using immunohistochemistry, and the relative expression of HMGB1, TLR2, TLR4, myeloid differentiation primary response gene 88 (MyD88) and NF-κB p65 mRNA was quantified in mouse jejunal tissues using quantitative real-time PCR (qPCR) assay.

RESULTS:

HE staining showed that the mouse intestinal villi were obviously atrophic, shortened, and detached, and the submucosal layer of the mouse intestine was edematous in the infection group as compared with the control group, while the mouse intestinal villi tended to be structurally intact and neatly arranged in the GA and OMT groups. There were significant differences among the four groups in terms of the mouse intestinal villous height (F = 6.207, P = 0.000 5), intestinal crypt depth (F = 6.903, P = 0.000 3) and the ratio of intestinal villous height to intestinal crypt depth (F = 37.190, P < 0.000 1). The mouse intestinal villous height was lower in the infection group than in the control group [(321.9 ± 41.1) µm vs. (399.5 ± 30.9) µm; t = 4.178, P < 0.01] and the GA group [(321.9 ± 41.1) µm vs. (383.7 ± 42.7) µm; t = 3.130, P < 0.01], and the mouse intestinal crypt depth was greater in the infection group [(185.0 ± 35.9) µm] than in the control group [(128.4 ± 23.6) µm] (t = 3.877, P < 0.01) and GA group [(143.3 ± 24.7) µm] (t = 2.710, P < 0.05). The mouse intestinal villous height was greater in the OMT group [(375.3 ± 22.9) µm] than in the infection group (t = 3.888, P < 0.01), and there was no significant difference in mouse intestinal villous height between the OMT group and the control group (t = 1.989, P > 0.05). The mouse intestinal crypt depth was significantly lower in the OMT group [(121.5 ± 27.3) µm] than in the infection group (t = 4.133, P < 0.01), and there was no significant difference in mouse intestinal crypt depth between the OMT group and the control group (t = 0.575, P > 0.05). The ratio of the mouse intestinal villous height to intestinal crypt depth was significantly lower in the infection group (1.8 ± 0.2) than in the control group (3.1 ± 0.3) (t = 10.540, P < 0.01) and the GA group (2.7 ± 0.3) (t = 7.370, P < 0.01), and the ratio of the mouse intestinal villous height to intestinal crypt depth was significantly higher in the OMT group (3.1 ± 0.2) than in the infection group (t = 15.020, P < 0.01); however, there was no significant difference in the ratio of the mouse intestinal villous height to intestinal crypt depth between the OMT group and the control group (t = 0.404, P > 0.05). Immunohistochemical staining showed significant differences among the four groups in terms of occludin (F = 28.031, P < 0.000 1) and ZO1 expression (F = 14.122, P < 0.000 1) in mouse intestinal epithelial cells. The proportion of positive occluding expression was significantly lower in mouse intestinal epithelial cells in the infection group than in the control group [(14.3 ± 4.5)% vs. (28.3 ± 0.5)%; t = 3.810, P < 0.01], and the proportions of positive occluding expression were significantly higher in mouse intestinal epithelial cells in the GA group [(30.3 ± 1.3)%] and OMT group [(25.8 ± 1.5)%] than in the infection group (t = 7.620 and 5.391, both P values < 0.01); however, there was no significant differences in the proportion of positive occluding expression in mouse intestinal epithelial cells between the GA or OMT groups and the control group (t = 1.791 and 2.033, both P values > 0.05). The proportion of positive ZO1 expression was significantly lower in mouse intestinal epithelial cells in the infection group than in the control group [(14.4 ± 1.8)% vs. (24.2 ± 2.8)%; t = 4.485, P < 0.01], and the proportions of positive ZO1 expression were significantly higher in mouse intestinal epithelial cells in the GA group [(24.1 ± 2.3)%] (t = 5.159, P < 0.01) and OMT group than in the infection group [(22.5 ± 1.9)%] (t = 4.441, P < 0.05); however, there were no significant differences in the proportion of positive ZO1 expression in mouse intestinal epithelial cells between the GA or OMT groups and the control group (t = 0.037 and 0.742, both P values > 0.05). qPCR assay showed significant differences among the four groups in terms of HMGB1 (F = 21.980, P < 0.000 1), TLR2 (F = 20.630, P < 0.000 1), TLR4 (F = 17.000, P = 0.000 6), MyD88 (F = 8.907, P = 0.000 5) and NF-κB p65 mRNA expression in mouse jejunal tissues (F = 8.889, P = 0.000 7). The relative expression of HMGB1 [(5.97 ± 1.07) vs. (1.05 ± 0.07); t = 6.482, P < 0.05] 、TLR2 [(5.92 ± 1.29) vs. (1.10 ± 0.14); t = 5.272, P < 0.05] 、TLR4 [(5.96 ± 1.50) vs. (1.02 ± 0.03); t = 4.644, P < 0.05] 、MyD88 [(3.00 ± 1.26) vs. (1.02 ± 0.05); t = 2.734, P < 0.05] and NF-κB p65 mRNA [(2.33 ± 0.72) vs. (1.04 ± 0.06); t = 2.665, P < 0.05] was all significantly higher in mouse jejunal tissues in the infection group than in the control group. A significant reduction was detected in the relative expression of HMGB1 (0.63 ± 0.01), TLR2 (0.42 ± 0.10), TLR4 (0.35 ± 0.07), MyD88 (0.70 ± 0.11) and NF-κB p65 mRNA (0.75 ± 0.01) in mouse jejunal tissues in the GA group relative to the control group (t = 8.629, 5.830, 11.500, 4.729 and 6.898, all P values < 0.05), and the relative expression of HMGB1, TLR2, TLR4, MyD88 and NF-κB p65 mRNA significantly reduced in mouse jejunal tissues in the GA group as compared to the infection group (t = 7.052, 6.035, 4.084, 3.165 and 3.274, all P values < 0.05). In addition, the relative expression of HMGB1 (1.14 ± 0.60), TLR2 (1.00 ± 0.24), TLR4 (1.14 ± 0.07), MyD88 (0.96 ± 0.25) and NF-κ B p65 mRNA (1.12 ± 0.17) was significantly lower in mouse jejunal tissues in the OMT group than in the infection group (t = 7.059, 5.320, 3.510, 3.466 and 3.273, all P values < 0.05); however, there were no significant differences between the OMT and control groups in terms of relative expression of HMGB1, TLR2, TLR4, MyD88 or NF-κB p65 mRNA in mouse jejunal tissues (t = 0.239, 0.518, 1.887, 0.427 and 0.641, all P values > 0.05).

CONCLUSIONS:

C. parvum infection causes intestinal inflammatory responses and destruction of intestinal mucosal barrier through up-regulating of the HMGB1-TLR2/TLR4-NF-κB pathway. OMT may suppress the intestinal inflammation and repair the intestinal mucosal barrier through inhibiting the activity of the HMGB1-TLR2/TLR4-NF-κB pathway.