Exacerbation of oxazolone colitis by infection with the helminth Hymenolepis diminuta: involvement of IL-5 and eosinophils

Abstract

Substantial data show that infection with helminth parasites ameliorates colitis; however, oxazolone-induced colitis is exaggerated in mice infected with the tapeworm, Hymenolepis diminuta. We tested the hypothesis that the IL-5 response to helminth infection enhances the severity of oxazolone-induced colitis. Mice were infected with H. diminuta and 8 days later were treated with oxazolone ± anti-IL-5 antibodies. Colitis was assessed 72 hours postoxazolone treatment by disease activity scores, myeloperoxidase activity, and histopathology. Other mice received injections of a replication-deficient adenovirus that carried the IL-5 (Ad.IL-5) gene or a control adenovirus (Ad.delete) ± oxazolone. The effect of H. diminuta+oxazolone in CCL11/CCL22 (eotaxin-1 and 2) knockout (KO) mice was determined. Helminth infection and Ad.IL-5 treatment increased IL-5 and eosinophil numbers. In vivo neutralization of IL-5 significantly reduced the severity of colitis in H. diminuta+oxazolone-treated mice, and H. diminuta did not exaggerate oxazolone-induced colitis in CCL11/CCL22 KO mice. Mice receiving Ad.IL-5 only had no colitis, while oxazolone-induced colitis was more severe in animals cotreated with Ad.IL-5 (Ad.delete + oxazolone was not significantly different from oxazolone only). Thus, while there is much to be gleaned about antiinflammatory mechanisms from rodent-helminth model systems, these data illustrate the caveat that infection with helminth parasites as a therapy could be contraindicated in patients with eosinophilia or elevated IL-5 unless coupled to appropriate measures to block IL-5 and/or eosinophil activity.

Figure 1

Bar graphs showing increased IL-5 production by ConA (2 μg/ml)-stimulated spleen cells and lamina propria lymphocytes (LPLs) in H. diminuta (Hd)-infected mice [A (representative of three experiments) and B; n = 3). Panels C–F illustrate numbers of bone marrow eosinophils, circulating eosinophils, colonic eosinophils, and colonic eosinophil peroxidase (EPO) activity in control (con), oxazolone (OX; 3 mg ir, 72 hours)-treated mice, and OX+Hd mice (infected 8 days before OX) ± treatment with anti–IL-5 AB or an isotype-matched irrelevant AB (iAB) (mean ± SEM; n = 4 in B and D and 6–8 in E and F; both antibodies delivered ip. in three doses for a total of 200 μg; *P < 0.05 compared to control).

Figure 2

Bar graphs showing reduced colitic disease in mice treated with H. diminuta (Hd; five cysticercoids 8 days before oxazolone)+oxazolone (3 mg, ir.)+anti–IL-5 antibody (total 200 μg), as assessed by change in body weight (A), colon length (B), and disease activity scores (C) [mean ± SEM; n values are shown in parentheses in A and are from 1 to 3 experiments; assessment conducted 72 hours after oxazolone (OX); *P < 0.05 compared to control, **P < 0.05 compared to oxazolone, and ***P < 0.05 compared to H. d+oxazolone+anti–IL-5 AB; iAB, irrelevant antibody (IgG₁)].

Figure 3

A: histology damage scores in oxazolone (OX) ± H. diminuta ± anti–IL-5 antibody-treated mice [mean ± SEM; n = 7–8 mice from two experiments; assessment conducted 72 hours after oxazolone (OX); *P < 0.05 compared to control, **P < 0.05 compared to oxazolone, and ^†P < 0.05 compared to and H. d+oxazolone+anti–IL-5 AB, respectively; iAB, irrelevant antibody (IgG₁)]. B: depicts the number of mice/group with the most severe histological damage as assessed by histological damage scores greater than or equal to 9.

Figure 4

Infection with H. diminuta partially protects CCL11/CCL22 (eotaxin 1/2) KO mice from the colitis-induced instillation of oxazolone (OX; 3 mg, 72 hours before autopsy) as assessed by (A) disease activity (DAS) and (C) histological damage scores [mean ± SEM; n = 6–7 mice from two experiments; *P < 0.05 compared to control and **P < 0.05 compared to control and oxazolone, respectively; five cysticercoids of H. diminuta (H. d) given by oral gavage 8 days before oxazolone]. Panel B is representative images from H&E sections of the colon (E, edema; I, inflammatory infiltrate; L, lumen of colon; M, outer muscle layers; arrow indicates major architectural disruption; original magnification, ×200; note dilation of the lumen in the second example of H.d+OX-treated mice).

Figure 5

Bar graph showing serum levels of IL-5 four days after ip injection of Ad.IL-5 (10⁸ pfu, ip.) and 3 days postoxazolone (3 mg, ir) (mean ± SEM; n = 4; *P < 0.05 compared to control and **P < 0.05 compared to Ad.IL-5, respectively).

Figure 6

Bar graphs showing the exaggeration of colitis in Ad.IL-5 [10⁸ pfu, ip. 24 hours before oxazolone (OX)]+oxazolone (3 mg ir.) treated mice as assessed by change in body weight (A), colon length (B), and disease activity scores (C) [mean ± SEM; n values shown in parentheses in panel A (from 3 to 5 experiments); assessment conducted 72 hours after oxazolone; *P < 0.05 compared to control and **P < 0.05 compared oxazolone, respectively].

Figure 7

Representative images of H&E sections of the colon from the various groups of mice with the resultant histological damage scores from control (con), oxazolone (3 mg ir, +72 hours) and oxazolone+Ad.delete or Ad.IL-5 (10⁸ pfu, ip.)-treated mice (mean ± SEM; n = 9–18 from 3 to 5 experiments; *P < 0.05 compared to control and **P < 0.05 compared to oxazolone, respectively; m, muscle; l, lumen; arrowhead, friable epithelium; arrow, inflammatory infiltrate; asterisk, ulceration; original magnification, ×200).

Figure 8

Coculture with HL-60 differentiated eosinophils (1 × 10⁵) and LPS (100 ng/ml) results in increased permeability of monolayers of the human colon-derived T84 epithelial cell line as measured by (A) transepithelial electrical resistance (TER: a measure of the paracellular barrier to the passive flux of ions) and (B) the apical to basolateral flux of horse-radish peroxidase (HRP, a 44-kDa protein) 48 hours later [mean ± SEM; n = 6 epithelial monolayers from two experiments; starting TER = 790–2460 Ω/cm²; *P < 0.05 compared to control (ie, T84 cells only); **P < 0.05 compared to other groups].

Figure 9

Bar graph showing IL-10 production from spleen cells and lamina propria lymphocytes (LPLs) 48 hours after ConA (2 μg/ml) stimulation (mean ± SEM; n = 3 from a representative experiment; *P < 0.05 compared to control and oxazolone only; nd, not detected).