Commensal Escherichia coli reduces epithelial apoptosis through IFN-alphaA-mediated induction of guanylate binding protein-1 in human and murine models of developing intestine

Abstract

Appropriate microbial colonization protects the developing intestine by promoting epithelial barrier function and fostering mucosal tolerance to luminal bacteria. Commensal flora mediate their protective effects through TLR9-dependent activation of cytokines, such as type I IFNs (alpha, beta) and IL-10. Although IFN-beta promotes apoptosis, IFN-alpha activates specific antiapoptotic target genes whose actions preserve epithelial barrier integrity. We have recently identified guanylate binding protein-1 (GBP-1) as an antiapoptotic protein, regulated by both type I and type II IFNs, that promotes intestinal epithelial barrier integrity in mature intestine. However, the mechanisms by which commensal bacteria regulate epithelial apoptosis during colonization of immature intestine and the contributions of GBP-1 are unknown. The healthy newborn intestine is initially colonized with bacterial species present in the maternal gastrointestinal tract, including nonpathogenic Escherichia coli. Therefore, we examined the influence of commensal E. coli on cytokine expression and candidate mediators of apoptosis in preweaned mice. Specifically, enteral exposure of 2 wk-old mice to commensal E. coli for 24 h selectively increased both IFN-alphaA and GBP-1 mRNA expression and prevented staurosporine-induced epithelial apoptosis. Exogenous IFN-alphaA treatment also induced GBP-1 expression and protected against staurosporine-induced apoptosis in a GBP-1 dependent manner, both in vitro and ex vivo. These findings identify a role for IFN-alphaA-mediated GBP-1 expression in the prevention of intestinal epithelial apoptosis by commensal bacteria. Thus IFN-alphaA mediates the beneficial effects of commensal bacteria and may be a promising therapeutic target to promote barrier integrity and prevent the inappropriate inflammatory responses seen in developing intestine as in necrotizing enterocolitis.

FIGURE 1

Vulnerability to induced apoptosis emerges between 1 and 2 weeks postnatal age in immature mouse epithelium. C57Bl6 mouse pups were sacrificed on embryonic day 18 (E18) or postnatal day 2, 8 (1w), 14 (2w), or 21 (3w). Small intestines were harvested and exposed to DMEM with (Ctrl) or without serum for 1 (DMEM1) or 2 hours (DMEM1) and apoptosis was quantitated by TUNEL staining. The number of TUNEL positive cells was analyzed by confocal microscopy. Representative images depict the effects of serum-starvation after 1 hour (A). The number of apoptotic cells were counted for a total of 10 high power fields (HPFs) and depicted as the average per 10 HPF following the indicated intervals of serum starvation (B). Alternatively, intestines were incubated in the presence of staurosporine (1 μg/ml) for 1 (STS1) or 2 (STS2) hours and the number of apoptotic cells determined by TUNEL staining relative to intestines maintained in serum-containing media (C). (n = 5 mice per condition, *p < 0.0001 for serum starvation and for STS by 2-way ANOVA. Effects of age were significant in both cases by 2-way ANOVA, *p < 0.0001)

FIGURE 2

Commensal E. coli protects against STS-induced apoptosis in immature small and large intestine. 2 week-old mouse pups were orally gavage fed 1–2× 10⁷ CFU of commensal E. coli (E. coli) or equal volumes of media (Ctrl), then returned to their mothers for 24 hours prior to sacrifice. Small (A) and large intestine (B) were harvested and incubated in the absence (vehicle) or presence of staurosporine (STS) for 2 hours then analyzed for apoptosis by TUNEL staining with the results depicted as the number of TUNEL positive cells per 10 HPF. (n =6–9 mice per condition, *p < 0.0001)

FIGURE 3

Commensal E. coli selectively induces IFNαA and GBP-1 in immature mouse colon. Colons were isolated from E. coli-fed 2 week-old mice, processed for RNA and analyzed by real-time PCR for levels of cytokine (A) and GBP-1 (B) expression. Results are depicted as fold change relative to controls and normalized for expression of the housekeeping gene SRP-14. (n = 3 – 6, *p < 0.05, **p < 0.01)

FIGURE 4

IFNαA induces GBP-1 expression in immature model human intestinal epithelia. FHs 74 Int cells were grown on 24 well plates and treated with 0, 10 or 100 U/ml of IFNαA for indicated time intervals and processed for analysis of GBP-1 mRNA expression (A) by real-time PCR or for GBP-1 protein expression (B) by western blot. Fold change in mRNA level is depicted as the mean ±s.e.m. (n = 3, *p < 0.05). Protein expression is depicted relative to tubulin expression, as a control for protein loading, in this representative blot from a series of 3 replicate experiments.

FIGURE 5

IFNαA protects against STS-induced apoptosis in immature human intestinal epithelia, in vitro. FHs 74 Int cells were treated with IFNαA, 10 or 100 U/ml, for 24 or 48 hours then treated with STS, 1 μg/ml for 2 hours, fixed and analyzed for apoptotic cells by TUNEL (A) and immunofluorescent staining for activated caspase 3 (B). Results depict the mean number of positive cells ±s.e.m, determined from 10 high power fields. (n = 3, *p < 0.05)

FIGURE 6

GBP-1 is required for the anti-apoptotic effects of IFNαA in immature human intestinal epithelia, in vitro. FHs 74 Int cells were loaded with sequence specific siRNA directed against GBP-1 or an off-target siRNA against cyclophilin B (CyB), then treated with indicated concentrations of IFNαA for 24 hours. GBP-1 protein expression was then analyzed by western blot (A). Results are depicted by representative blot relative to the results from the same blot stripped and reprobed for tubulin as a control for the amount of total protein per lane (n = 3). FHs 74 Int cells loaded with either CyB specific (B) or GBP-1-specific siRNA (C) were then treated with indicated concentrations of IFNαA for 24 hours, followed by exposure to STS for 2 hours. Cells were then stained for activated caspase 3 and the number of positive cells counted per 10 HPFs. Results are expressed as the average number of cells positive for activated caspase 3 ± s.e.m. (n = 4, *p < 0.001).

FIGURE 7

Exogenous IFNαA prevents STS-induced apoptosis and induces GBP-1 in 2 week-old mouse intestines. 2 week-old mice were injected with IFNαA, 40 U/g baseline weight, i.p. or an equal volume of PBS (Ctrl) then sacrificed. Isolated small intestines (A) and colon (B) were then incubated for 2 hours in media alone or media containing 1 μg/ml STS. Tissues were then fixed and processed for TUNEL staining. The number of TUNEL positive cells for each condition were counted and plotted as the mean ± s.e.m., determined per 10 HPFs. Colons isolated from the same animals were also processed for RNA analysis by real-time PCR and the level of GBP-1 expression in IFNαA-treated animals is plotted as fold change relative to controls (C). Results are normalized according to levels of expression of murine SRP-14. (n = 5 – 6, *p < 0.01).