Toll pathway modulates TNF-induced JNK-dependent cell death in Drosophila

Abstract

Signalling networks that control the life or death of a cell are of central interest in modern biology. While the defined roles of the c-Jun N-terminal kinase (JNK) pathway in regulating cell death have been well-established, additional factors that modulate JNK-mediated cell death have yet to be fully elucidated. To identify novel regulators of JNK-dependent cell death, we performed a dominant-modifier screen in Drosophila and found that the Toll pathway participates in JNK-mediated cell death. Loss of Toll signalling suppresses ectopically and physiologically activated JNK signalling-induced cell death. Our epistasis analysis suggests that the Toll pathway acts as a downstream modulator for JNK-dependent cell death. In addition, gain of JNK signalling results in Toll pathway activation, revealed by stimulated transcription of Drosomycin (Drs) and increased cytoplasm-to-nucleus translocation of Dorsal. Furthermore, the Spätzle (Spz) family ligands for the Toll receptor are transcriptionally upregulated by activated JNK signalling in a non-cell-autonomous manner, providing a molecular mechanism for JNK-induced Toll pathway activation. Finally, gain of Toll signalling exacerbates JNK-mediated cell death and promotes cell death independent of caspases. Thus, we have identified another important function for the evolutionarily conserved Toll pathway, in addition to its well-studied roles in embryonic dorso-ventral patterning and innate immunity.

Keywords: Drosophila; Eiger; Toll; c-Jun N-terminal kinase; cell death.

Figure 1.

A genetic screen for dominant modifiers of GMR>Egr-induced eye-ablation phenotype. (a and j) Schematic of the genomic region surrounding Toll or Dif and dorsal locus. The deleted regions uncovered by deficiencies are indicated in grey (no effect) or red (suppressor). (b–i) Light micrographs of Drosophila adult eyes are shown. Compared with GMR-GAL4 control (b), GMR>Egr-triggered small eye phenotype (c) remains unaffected by Df(3R)ED6232 (d), but is partially inhibited by deficiencies Df(3R)BSC496 (e), Df(3R)BSC524 (f), Df(3R)ED6255 (g), Df(3R)BSC294 (h) or Toll^r3mutation (i). See the electronic supplementary material for detailed genotypes.

Figure 2.

Loss of Toll signalling inhibits GMR>Egr-triggered cell death. (a,b) Diagrams for the key components of Toll and Imd signalling. Light micrographs of Drosophila adult eyes (c–l) and fluorescence micrographs of third-instar larval eye discs (c′–l′) are shown. Compared with control (c and c′), GMR>Egr-induced small eye phenotype (d) and cell death in eye discs (d′) remain unaffected by expression of LacZ (e and e′) or mutation in imd (f and f′), but are partially suppressed by mutation in Dif (g and g′) or RNAi-mediated downregulation of Toll pathway components: Toll, tube, pelle, dorsal and Dif (h–l and h′–l′). (m) Statistic analysis of cell death in eye discs shown in (c′–l′). Average number of dying cells labelled by AO staining are counted. Error bars indicates standard deviation. One-way ANOVA with Bonferroni multiple comparison test was used to compute p-values, significance was indicated with asterisks (***p < 0.001, n = 10 in each group); n.s., not significant. See the electronic supplementary material for detailed genotypes.

Figure 3.

Toll signalling acts downstream of JNK in eye development. (a–o) Light micrographs showing Drosophila adult eyes. The small and rough eye phenotype resulting from ectopic expression of dTAK1 (a) or Hep^CA (f) is suppressed partially by RNAi-mediated knocking-down of bsk (c and h), pelle (d and i) or dorsal (e and j), but not of LacZ (b and g). The rough eye phenotype produced by GMR>Bsk (k) is obviously suppressed by RNAi-mediated inactivation of Toll pathway components: pelle, dorsal and Dif (m–o), but not by the expression of GFP (l). (p) A diagram for the key components of the Egr–JNK pathway. See the electronic supplementary material for detailed genotypes.

Figure 4.

Toll signalling acts downstream of Hep in wing development. Light micrographs of Drosophila adult wings (a–m). Compared with ptc-GAL4 control (a), ectopic expression of Egr^w (b) or Hep (h) or Toll (m) driven by ptc-GAL4 generates loss of ACV phenotype. This phenotype, produced by ptc>Egr^W (b) or ptc>Hep (h), is strongly suppressed by a dominant-negative form of Bsk (d and j) or by depletion of Toll signal (e–g, k and l), but not that of LacZ (c and i). The lower panels show high magnification view of the boxed area in upper panels (a–m). (n) Quantification of the ACV phenotype as shown in panels (a–m) (for each genotype, n = 20). Error bars indicates standard deviation. One-way ANOVA with Bonferroni multiple comparison test was used to compute p-values, significance was indicated with asterisks (***p < 0.001); n.s., not significant. See the electronic supplementary material for detailed genotypes.

Figure 5.

Toll signalling acts downstream of FoxO and mediates caspases-independent cell death. (a–h and j–m) Light micrographs of Drosophila adult wings are shown. Compared with Sd-GAL4 control (a), ectopic expression of dFoxO results in an evident reduction in wing size (d), which is suppressed significantly by loss of Toll signalling (f–h), but not that of GFP (e). Expression of dJun or dFos driven by Sd-GAL4 does not produce any visible defects (b and c). The small wing phenotype induced by Sd>Hep (j) is strongly suppressed by Df(3L)H99 that deletes one copy of the apoptotic genes reaper, hid and grim (k), the expression of DIAP1 (l) or Dronc^DN (m). (i and n) Quantifications of adult wing size/wild-type (WT) ratio shown in figures a–h and j–m, respectively (n = 10). One-way ANOVA with Bonferroni multiple comparison test was used to compute p-values, significance was indicated with asterisks (***p < 0.001); n.s., not significant. (o–r) X-Gal staining of a hid-LacZ reporter in wing discs. Compared with the control (o), expression of Hep in the wing pouch driven by Sd-GAL4 induces hid transcription (p), which cannot be suppressed by knocking-down dorsal or Dif (q and r). See the electronic supplementary material for detailed genotypes.

Figure 6.

Loss of Toll signalling suppresses physiological JNK-induced cell death. Light micrographs of Drosophila adult wings (a–g) and fluorescence micrographs of third-instar larval wing discs (h–n) are shown. RNAi-mediated downregulation of puc along the A/P compartment boundary by ptc-GAL4 produces the loss-of-ACV phenotype in adult wings (a), which results from strong cell death in larval wing discs (h). Both phenotypes depends on endogenous JNK (c and j) and the Toll pathway (d–g and k–n), but not on LacZ (b and i). The bottom panels show high magnification views of the boxed area in upper panels (a–g). (o and p) Statistical analysis of ACV phenotype (n = 20 for each genotype) and cell death in wing discs (n = 10) as shown in figures a–g and h–n respectively. Error bars indicate standard deviation. One-way ANOVA with Bonferroni multiple comparison test was used to compute p-values, significance was indicated with asterisks (***p < 0.001); n.s., not significant. See the electronic supplementary material for detailed genotypes.

Figure 7.

Gain of JNK signalling promotes Drs expression and dorsal nuclear localization. (a–c) Fluorescent microscope images showing third-instar larvae. Compared with control (a), expression of dTAK1 (b) or RNAi inactivation of puc (c) in fat body upregulates Drs-GFP expression. (d) Fluorescent microscope images showing fat body dissected from third-instar larvae stained with anti-dorsal (red). dTAK1-expressing clones were tagged by GFP (green) and induced for 1 h by heat shock at 37°C and recovered for 24 h at 25°C. Nuclei were labelled with DAPI (blue). Endogenous Dorsal protein displays nuclear localization in cells expressing high level of dTAK1 (arrows). (d′) shows high magnification views of the boxed area in (d). DAPI, 4, 6-diamidino-2-phenylindole. See the electronic supplementary material for detailed genotypes.

Figure 8.

Gain of JNK signalling upregulates the expression of Spz family ligands. (a) Histogram showing the levels of Spz1–6 mRNAs measured by quantitative RT-PCR. Total RNA of Drosophila adult eyes was extracted and normalized for cDNA synthesis. Error bars represent standard deviation from three independent experiments. (b–f) Fluorescence micrographs of Drosophila third-instar larval eye discs are shown. Compared with the GMR-GAL4 control (c), Spz6-GFP expression was evidently increased by expression of Egr (d), dTAK1 (e) or Hep (f), mRFP marking the GMR-GAL4 expression region (b). The right panels show views of vertical cross sections corresponding to the left panels (b–f). Nuclei were labelled with DAPI (blue), cell membranes were stained by anti-Dlg antibody (red). Imaging of prepared samples was conducted by a Leica confocal microscope (Leica SP5). See the electronic supplementary material for detailed genotypes.

Figure 9.

Gain of Toll signalling enhances GMR>Egr-induced cell death. Light micrographs of Drosophila adult eyes (a–j) and fluorescence micrographs of third-instar larval eye discs (a′–j′) are shown. Compared with control (a and a′), expression of Toll^10B, Pelle, a cactus RNAi or a weak Egr line (Egr^W) promotes little or mild cell death in eye discs and adult eyes (c–f and c′–f′). Co-expression of Egr^W with Toll^10B, Pelle or a cactus RNAi displays synergistic enhancement of cell death in eye discs and adult eyes (h–j and h′–j′). As a negative control, expression of LacZ neither triggers cell death by itself (b and b′) nor aggravates Egr^W-induced cell death (g and g′). (k) Statistical analysis of cell death in eye discs shown in (a′–j′). Average number of dying cells labelled by AO staining are counted. Error bars indicate standard deviation. One-way ANOVA with Bonferroni multiple comparison test was used to compute p-values, significance was indicated with asterisks (**p < 0.01, ***p < 0.001, n = 10 in each group); n.s., not significant. (l) A model for the role of the Spz/Toll/NF-κB pathway in modulating Egr-triggered JNK-mediated cell death. See the electronic supplementary material for detailed genotypes.