big bang gene modulates gut immune tolerance in Drosophila

Abstract

Chronic inflammation of the intestine is detrimental to mammals. Similarly, constant activation of the immune response in the gut by the endogenous flora is suspected to be harmful to Drosophila. Therefore, the innate immune response in the gut of Drosophila melanogaster is tightly balanced to simultaneously prevent infections by pathogenic microorganisms and tolerate the endogenous flora. Here we describe the role of the big bang (bbg) gene, encoding multiple membrane-associated PDZ (PSD-95, Discs-large, ZO-1) domain-containing protein isoforms, in the modulation of the gut immune response. We show that in the adult Drosophila midgut, BBG is present at the level of the septate junctions, on the apical side of the enterocytes. In the absence of BBG, these junctions become loose, enabling the intestinal flora to trigger a constitutive activation of the anterior midgut immune response. This chronic epithelial inflammation leads to a reduced lifespan of bbg mutant flies. Clearing the commensal flora by antibiotics prevents the abnormal activation of the gut immune response and restores a normal lifespan. We now provide genetic evidence that Drosophila septate junctions are part of the gut immune barrier, a function that is evolutionarily conserved in mammals. Collectively, our data suggest that septate junctions are required to maintain the subtle balance between immune tolerance and immune response in the Drosophila gut, which represents a powerful model to study inflammatory bowel diseases.

Fig. 1.

BBG is required for normal lifespan and immune tolerance in the gut. (A and C) Lifespan experiments on regular cornmeal medium (A) and on medium containing antibiotics (C). In normal conditions, the lifespan of bbg^B211/B211 flies is reduced compared with WT flies (A). Premature death of mutant flies is rescued by a treatment with antibiotics (C). Dashed line indicates the LT₅₀. Each curve represents the mean of three independent experiments with three groups of 20 flies. Error bars are SD. (B and D) Diptericin-lacZ reporter activity in absence of infection with (D) or without (B) antibiotic treatment. IMD pathway activation increases with time on normal cornmeal medium but remains stronger in bbg^B211/B211 mutant flies (B). Antibiotic treatment abolishes IMD pathway activation observed in B (D). Age of the flies is indicated. Representative image from an experimental sample of 15 guts. (Scale bar, 300 μm.) (E) Mitotic count in gut epithelial cells from WT or bbg^B211/B211 mutant flies raised on regular (normal) or antibiotic-treated cornmeal medium (AB) for either 30–35 d or 45–50 d after hatching. The mean of PH3-positive cells per condition is represented by a black line. A minimum of 15 guts per condition were analyzed. *P value < 0.05; **P value < 0.01; ***P value < 0.001.

Fig. 2.

BBG is localized at the apical and lateral sides of gut epithelial cells. BBG is localized at the apical (arrowhead) and lateral (arrow) sides of the proventriculus (A–C) and of the midgut (D–L) epithelial cells. BBG is distributed as an apical ring in these cells (E) and matches the localization of the septate junction-associated protein Coracle (H and L, merge). (A–C) Immunolocalization of BBG (green) and Adducin (red) on whole gut or (D) on paraffin section. (C) Square, lower magnification of proventriculus; arrowhead, region of A–C. (E–L) Immunolocalization of BBG (red) and Coracle (green) on whole gut. (G and K) DAPI staining. (H) Merge of E–G. (L) Merge of I–K. Representative image from an experimental sample of 10 guts. [Scale bars (A–D) 20 μm; (C, square) 50 μm; (E–H) 30 μm; (I–L) 15 μm.]

Fig. 3.

Lack of BBG results in septate junctions and permeability defects in the gut. (A) BBG participates in septate junctions structure. TEM micrographs of transversal sections through the anterior midgut of WT or bbg^B211/B211. In WT fly midgut, the paracellular space at the level of the septate junctions spans 20 nm, whereas it reaches 30 nm in flies defective for bbg. Magnification, 120,000×. Squares are numeric magnifications of original pictures. Representative image from an experimental sample of five guts. (Scale bar, 20 nm.) (B and C) Survival curves for WT, kenny^C02831 (key ^C02831), bbg null mutants (bbg^B211/B211, bbg^P-15602/B211), bbg^{P-15602/def-3125} or bbg^{ex-15602/def-3125} flies fed on (C) P. aeruginosa PA14 (OD₆₀₀ = 0.25), or (B) S. marcescens DB11 (OD₆₀₀ = 1) at 25 °C. Each curve is representative of three independent experiments with three groups of 20 flies; statistics are detailed in Fig. S6. (D) Gut permeability toward P. aeruginosa PA14. Flies were preinjected (Δphagocytosis) or not with latex beads and fed 1 d later on a solution of P. aeruginosa PA14 (OD₆₀₀ = 0.25) at 25 °C. Bacterial counts from hemolymph or dissected guts of the same genotypes were evaluated 2 d after oral infection. Data are representative of three independent experiments performed with three groups of 10 flies for each condition. *P value < 0.05.

Fig. 4.

SJs are required to prevent oral infection by invasive bacteria. (A) coracle knockdown. Immunolocalization of Coracle (Green) in Drosophila anterior midgut isolated from WT or NP³⁰⁸⁴-Gal4/UAS-RNAi-coracle (NP/RNAi coracle) flies. (Scale bar, 20 μm.) (B) Coracle participates in septate junctions structure. TEM micrograph of transversal section through the anterior midgut of flies knocked down for coracle in the midgut (NP/RNAi coracle). In WT flies midgut the paracellular space at the level of the septate junctions spans 20 nm, whereas it reaches 30 nm in flies defective for Coracle (NP/RNAi coracle). Magnification, 120,000×. Square is numeric magnification of original picture. Representative image from an experimental sample of five guts. (Scale bar, 20 nm.) (C) Knockdown of Coracle in the gut impairs survival to P.aeruginosa. WT, bbg^B211/B211, NP³⁰⁸⁴/+, NP³⁰⁸⁴-Gal4/UAS-RNAi-gfp (NP/RNAi-GFP), NP³⁰⁸⁴-Gal4/UAS-RNAi-coracle (NP/RNAi-Coracle), flies were challenged at 25 °C with P.aeruginosa PA14. Data are representative of three independent experiments performed with three groups of 20 flies; statistics are detailed in Fig. S6.