Drosophila acinus encodes a novel regulator of endocytic and autophagic trafficking

Abstract

Endosomal trafficking affects many cellular pathways from cell signaling to metabolism, but little is known about how these effects are coordinated. In a genetic screen for mutants affecting endosomal trafficking, we identified Drosophila acinus (dacn; hook-like). Its mammalian homolog Acinus has been implicated in RNA processing and chromatin fragmentation during apoptosis. Loss-of-function analysis of dacn revealed two distinct functions. First, dacn is required for stabilization of early endosomes, thus modulating levels of Notch and Egfr signaling. Second, loss of dacn interferes with cellular starvation responses by inhibiting autophagosome maturation. By contrast, overexpression of dacn causes lethality due to enhanced autophagy. We show that this enhanced autophagy is independent of the Tor pathway. Taken together, our data show that dacn encodes a regulator of endosomal and autophagosomal dynamics, modulating developmental signaling and the cellular response to starvation.

Fig. 1.

dacn²⁷ was found in a screen for mutations affecting trafficking. (A-C) Light micrographs of adult Drosophila eyes. The red wild-type whole-eye clone (A), is distinct from the rust-colored dacn²⁷ whole-eye clone, which lacks a pseudopupil (B). A dacn²⁷ whole-eye clone in a fly carrying a genomic dacn transgene has wild-type eye color (C). (D-F) Eye discs from wild type (D), dacn¹/dacn²⁷ (E) and dacn¹/dacn²⁷; P[dacn] (F) stained for Boss. Endosomal accumulation of Boss in R7 cells (arrowheads) is lost in dacn cells (E) and is restored by a genomic dacn transgene. (G-I′) In eye discs carrying mitotic clones, dacn²⁷ cells were marked with a double dose of GFP (green) carried on the mutant chromosome (G,H,I). (G) Eye disc stained for Boss (magenta). Arrows point to individual dacn²⁷ R7 cells that lack Boss staining, even when R8 is wild-type (yellow arrows). (H) Eye disc stained for Dl (red) and Avl (cyan). Higher magnification in the inset reveals partial colocalization. (I) Eye disc stained for Dl after a 2-hour incubation in 50 μg/ml chloroquine (CQ), which restored Delta staining in dacn²⁷ cells (arrows) to the level of surrounding wild-type cells. In G′, H′ and I′, the GFP channel was removed for clarity. (J) dAcn and human Acinus share regions of high homology as indicated in green, among them the P17 domain, which shows 53% identity. Two mutations, dacn¹ and dacn²⁷, truncate the protein before the conserved domains. Genotypes are: (A) nGFP³³ nGFP³⁸ FRT40A/FRT40 GMR-Hid l(2)CL-L¹; eyFLP; (B) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40 GMR-Hid l(2)CL-L¹; eyFLP; (C) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40 GMR-Hid l(2)CL-L¹; eyFLP/P[genomic dacn]; (D) Oregon R; (E) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹; (F) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹; P[genomic dacn]; (G-I) yw eyFLP/+; dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40A.

Fig. 2.

dAcn is expressed in the nuclei of eye disc cells. (A) In western blots of Drosophila third instar larval lysates, dAcn levels were reduced in dacn¹/dacn²⁷, but the levels of the other indicated proteins were similar to those in wild type. (B-D) Eye discs expressing nuclear GFP (green) were stained for dAcn (magenta). Posterior is to the right. dAcn accumulates in nuclei at the morphogenetic furrow and in a dynamic pattern in photoreceptor and cone cells. (B,C) Micrographs show cells either close to the furrow (B, white arrow) or cells more apical and more posterior (C). Asterisks mark examples of cells with cytoplasmic staining. The three ommatidia next to the yellow arrow are magnified in D, as labeled by the specific rows of their location. Photoreceptor or cone cells (c) with strongest staining in each ommatidium are indicated. (E) Schematic illustrating the relative positions of cells in ommatidia. (F) Eye disc expressing Myc-tagged dAcn was stained with anti-Myc antibodies to reveal dAcn expression in ommatidia of the indicated row or the peripodial membrane (PM). Genotypes are: (B-D) nGFP³³ nGFP³⁸ FRT40A/CyO; (F) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹; P[genomic Myc-dacn].

Fig. 3.

dacn is not required for apoptotic chromatin destruction. (A-I) Micrographs of hemocytes isolated from Drosophila third instar larvae and cultured for 2 hours (A-C). After apoptosis was induced using Smac mimetic (Smac) (Li et al., 2004), wild-type cells (D-F) and dacn^1/27 cells (G-I) had condensed and fragmented chromatin, as visualized by Hoechst staining (B,E,H) and TUNEL labeling (C,F,I). Genotypes: (A-F) w¹¹¹⁸; (G-I) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹.

Fig. 4.

Endocytosis is not impaired in dacn, but dwell time in endosomes is reduced. (A-C′) Live Drosophila eye discs with mitotic clones of dacn²⁷ (–/–) were labeled with an antibody to the extracellular domain of Dl. Clones were marked with a double dose of nuclear GFP (A,B,C, outlined). Immediately after labeling, Dl labeling was indistinguishable in apical sections of wild-type and dacn²⁷ cells (A′) and not detected in basal sections (A′). After a 20-minute chase, Dl antibody signal was reduced in subapical regions of dacn clones compared with wild-type cells (B). After a 60-minute chase, Dl antibody signal was indistinguishable in wild-type and dacn²⁷ cells (C′). (D-D′) Texas-Red dextran was endocytosed into eye discs containing dacn²⁷ clones for 5 minutes and then chased for 20 minutes. Dextran trafficking was unchanged in dacn²⁷ cells. Genotypes: yw eyFLP/+; dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40A. Scale bar: 10 μm.

Fig. 5.

Endosomes are mislocalized in dacn²⁷ cells. (A-E′) Drosophila third instar eye discs carrying mitotic clones of dacn²⁷ (–/–) were stained with antibodies marking endocytic compartments. Clones were marked with a double dose of nuclear GFP (A-E′, outlined in yellow). Entire stacks of z-sections encompassing all nuclei in a given region were used to evaluate individual cells as mutant or wild-type as reflected in the complex clone boundaries. Levels of Avl (A′), Rab5 (B′), Hrs (C′), which label early endosomes and MVBs, are reduced in dacn²⁷ clones. Levels of Rab7 (D′) and Rab11 (E′) were only mildly affected in dacn²⁷ clones. (F) The relative fluorescence for the indicated endosomal markers in dacn²⁷ cells normalized to that in wild-type cells. Genotypes: yw eyFLP/+; dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40A. Scale bar: 10 μm.

Fig. 6.

Egfr and Notch signaling are reduced in dacn mutants. (A,B) Sections of adult Drosophila eyes showed stereotypical patterning in a wild-type whole-eye clone (A), but 41% of ommatidia in dacn²⁷ whole-eye clones had patterning defects (B). (C-F) Scanning electron micrographs of adult eyes showing regular patterning in wild-type (C) and roughness in Egfr^E1/+ (D) eyes. Removing one copy of dacn from Egfr^E1/+ flies restored eye patterning to close to that of the wild type (E), and a dacn genomic rescue construct abrogated the effect of the dacn²⁷ (F). (G,H) N^263-39/+ wings had small wing notches (G) due to reduced Notch signaling. Removal of a copy of dacn (H) enhanced the wing notch phenotype. (I) Measurement of wing notch area showed that removal of one copy of dacn doubled the size of the notches. This effect is suppressed by a genomic rescue construct. Notch area: wild type, 1870±151.2 μm², n=105; dacn¹, 3391±417.9 μm², n=82, P=0.002; dacn²⁷, 3637±233.9 μm², n=150, P<0.0001; dacn¹ plus rescue, 2306±233.1 μm², n=72, P=0.0306 versus dacn¹ only; dacn²⁷ plus rescue, 1990±151.6 μm², n=126, P<0.0001 versus dacn²⁷ only. Genotypes: (A,C) Oregon R; (B) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40 GMR-Hid l(2)CL-L¹; eyFLP; (D) Egfr^E1/+; (E) Egfr^E1, +/+, dacn²⁷ nGFP³³ nGFP³⁸ FRT40A; (F) Egfr^E1/+; (E) Egfr^E1, +/+, dacn²⁷ nGFP³³ nGFP³⁸ FRT40A; P[genomic dacn]; (G) N^263-39/+; (H) N^263-39/+; dacn²⁷ nGFP³³ nGFP³⁸ FRT40A.

Fig. 7.

Impaired autophagosome maturation in dacn mutants. (A,B) TEM of adult Drosophila eyes. Compared with wild-type eyes (A), dacn²⁷ eyes (B) contained high levels of glycogen-like particles in pigment cells (those not pseudo-colored yellow). Inset shows a higher magnification image of the glycogen-like particles. (C,D) Sections of dacn²⁷ whole-eye clones (D), but not wild-type clones (C), tested positive for glycogen by PAS staining. (E-L) Micrographs of fat body cells stained with Lysotracker (red) and Hoechst (blue). Four hours of starvation induced Lysotracker-positive autolysosomes in the wild-type fat body (E,F). Autolysosome formation was impaired in dacn¹/dacn²⁷ fat bodies (G), but restored by the addition of a dacn genomic rescue construct (H). Autolysosome formation induced by expression of tor^TED (I,J) or Pten (K,L) was also blocked in dacn¹/dacn²⁷ fat bodies (J,L). (M,N) TEM of fat bodies revealed autophagosomes (AP) and autolysosomes (AL) in wild-type (M) and dacn¹/dacn²⁷ (N) starved fat bodies. (O) Quantification of organelle size and quantity in electron micrographs. Autophagosome area as a percentage of total tissue increased in dacn, whereas autolysosome area was reduced. All phenotypes were restored to wild-type by the addition of a genomic dacn rescue construct. ^*, P≤0.0001 compared with wild-type or dacn plus rescue. Genotypes: (A,C) Oregon R; (B,D) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/FRT40 GMR-Hid l(2)CL-L¹; eyFLP; (E,F,M) w¹¹¹⁸; (G,H,N) dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹; (I) Lsp2-Gal4; UAS-tor^TED; (J) Lsp2-Gal4, dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹; UAS-Tor^TED; (K) Lsp2-Gal4; UAS-Pten; (L) Lsp2-Gal4, dacn²⁷ nGFP³³ nGFP³⁸ FRT40A/dacn¹; UAS-Pten. Scale bars: 25 μm in E for E-L; 1 μm in N for M,N.

Fig. 8.

Overexpression of dAcn in fat body-induced autophagy. (A-D) Micrographs of Drosophila third instar fat bodies stained with Lysotracker (red) and Hoechst (blue). Expression of dAcn induced the formation of Lysotracker-positive autolysosomes in the absence of protein starvation (A). Co-expression of p110 (B) or Rheb (C) did not block dAcn-induced autophagy. Atg5-RNAi co-expression suppressed the induction of Lysotracker puncta by dAcn (D). (E-G) Atg8-GFP labeling of autophagic vacuoles in third instar fat bodies (green). Atg8-GFP labels autophagosomes and autolysosomes in fat bodies from starved (F), but not from fed (E), third instar larvae. Expression of dAcn induced the formation of Atg8-GFP-labeled structures in fed larvae (G). (H) TEM reveals dAcn-induced structures as small multi-lamellar organelles. (I) Quantification of lethality induced by dAcn overexpression under control of Tubulin-Gal4. Bars indicate percentage of flies eclosed for the indicated genotypes compared with control flies. Genotypes: (A,H) Lsp2-Gal4; UAS-dacn; (B) Lsp2-Gal4; UAS-p110/UAS-dacn; (C) Lsp2-Gal4; UAS-Rheb/UAS-dacn; (D) Lsp2-Gal4; UAS-UAS-Atg5IR/UAS-dacn; (E,F) Lsp2-Gal4 UAS-Atg8a/+; (G) Lsp2-Gal4 UAS-Atg8a/UAS-dacn; (I) tub-Gal4/UAS-dacn | UAS-Atg5IR/+;; tub-Gal4/+ | tub-Gal4/UAS-dacn; UAS-CD8-GFP | Atg5IR/+; tub-Gal4/UAS-dacn. Scale bars: 25 μm in A for A-D; 10 μm in E for E-G; 500 nm in H.