Developmentally regulated IkappaB expression in intestinal epithelium and susceptibility to flagellin-induced inflammation

Abstract

Necrotizing enterocolitis is a devastating inflammatory condition of the intestine that occurs almost exclusively in premature newborns. Although its exact pathogenesis is unclear, we have postulated that it may result from a predisposition of the immature intestine to mount an unusually robust and damaging response to microbial infection. In support of this idea, we report that the IL-8 response of an immature human enterocyte cell line to bacterial infection was significantly higher than that of a mature enterocyte cell line. The response in both cell lines was flagellin-dependent. Corresponding to the difference in IL-8 production, the immature enterocytes expressed appreciably lower levels of specific IkappaB genes when compared with the mature enterocytes. Similar developmentally regulated differences in cytokine response and IkappaB expression were also seen in primary rat enterocytes, indicating that these observations were not peculiarities of the cell lines. Furthermore, when the level of IkappaBalpha expression was increased in the immature cell line by transfection, the flagellin-dependent IL-8 response was attenuated. Thus, we have demonstrated a previously undescribed developmental regulation of IkappaB expression in the intestine involved in modulating the IL-8 response to bacterial infection, which may contribute to the pathogenesis of age-specific inflammatory bowel diseases such as necrotizing enterocolitis.

Fig. 1.

Fetal H4 cells secrete significantly more IL-8 in response to the pathogenic Salmonella strain SL3201 and the commensal E. coli strain F18 than do the adult T84 cells. Cells were infected with the indicated bacteria, and IL-8 was measured by ELISA in the 5-h cell supernatants. IL-8 values were normalized to protein concentrations from cell lysates and are presented as mean ± SEM (n ≥ 4). Asterisks denote significance at P < 0.05 for the comparison of the values indicated by the brackets.

Fig. 2.

Flagellin is required for IL-8 secretion in response to bacterial infection in both T84 and H4 cells. Cells were infected with the indicated bacteria, and IL-8 was measured by ELISA in the 5-h cell supernatants. IL-8 values were normalized to protein concentrations from cell lysates and are presented as mean ± SEM (n ≥ 4). Asterisks denote significance at P < 0.05 for the comparison of the values indicated by the brackets. IL-8 secretion to the pathogenic Salmonella SL3201 strain is below control levels in the flagellin mutant SL3201 FljB^-/FliC^- (SL3201 BC^-), and IL-8 secretion to the commensal E. coli strain F18 is below control levels in the E. coli flagellin mutant DH5α in T84 cells (Upper) and H4 cells (Lower).

Fig. 3.

Differential expression of IκB genes in immature and mature enterocytes. Cell lysates were obtained from human IEC cell lines (Left) or rat primary IECs (Right). The expression of IκBα, IκBβ, IκBε, and GAPDH was revealed by Western blotting with specific antibodies. The results shown are representative of four separate experiments in each category.

Fig. 4.

Increased intestinal IL-6 expression in response to bacteria in preweaned rats compared with postweaned rats. Closed ileal loops were created and injected with buffer (negative control) or SL3201. Intestinal RNA samples were prepared 4 h later, and IL-6 was measured by real-time quantitative PCR. The experiment was performed in triplicate, and the results are presented as the mean± SEM C_T for IL-6 normalized to the C_T value for GAPDH. An asterisk denotes significance at P < 0.05 as compared with buffer-treated animals.

Fig. 5.

Transfection of H4 cells with an IκBα expression construct attenuates IL-8 production in response to flagellin. H4 cells were transfected with an IκBα construct and an IL-8 promoter-driven firefly luciferase reporter construct along with a Renilla luciferase reporter as an internal control for transfection efficiency. Cells were then treated for 5 h with bacterial supernatants as indicated, and luminescence was measured. As described in Materials and Methods, results are presented as the mean ± SEM fold increase in firefly luminescence (normalized to Renilla luminescence) over control (n = 6). An asterisk denotes significance at P < 0.05 compared with control.