Loss- and gain-of-function analyses of vacuolar protein sorting 2 in Notch signaling of Drosophila melanogaster

ABSTRACT

Notch signaling is an evolutionarily conserved mechanism that controls many cell-fate specifications through local cell-cell interactions. The core mechanisms of Notch activation and its subsequent intracellular signaling are well understood. Various cellular functions are required for the activation and regulation of Notch signaling. Among them, the endocytosis of Notch and its ligands is important for the activation and suppression of Notch signaling. The endosomal sorting complex required for transport (ESCRT) proteins are required to sort ubiquitinated membrane proteins, such as Notch, into early endosomes. A loss-of-function allele of vacuolar protein sorting 2 (vps2), which encodes a component of ESCRT-III, has been reported. However, this vps2 mutant still produces the N-terminal half of the protein, and its phenotypes were studied in only a few organs. Here, we generated the first null mutant allele of Drosophila vps2, designated vps2², to better understand the function of this gene. In Drosophila wing imaginal discs homozygous for the vps2² allele, early endosomes and multivesicular bodies (MVBs) were enlarged, and Notch and Delta accumulated inside them. As reported for the previous vps2 mutant, the epithelium grew excessively under this condition. We further studied the roles of vps2 by RNA interference-knockdown. These experiments revealed that a partial reduction of vps2 attenuated Notch signaling; in contrast, the loss-of-function vps2 mutant is reported to up-regulate the Notch signaling in eye imaginal disc cells. These results suggest that Notch signaling can be up- or down-regulated, depending on the level of vps2 expression. Finally, we found that vps2 overexpression also resulted in early-endosome enlargement and the accumulation of Notch and Delta. In these cells, a portion of the Vps2 protein was detected in MVBs and colocalized with Notch. These data indicate that the expression of vps2 must be precisely regulated to maintain the normal structure of early endosomes.

INTRODUCTION

Cell-cell interactions play essential roles in the development and homeostasis of multicellular organisms. Notch is a single-span transmembrane receptor that mediates such cell-cell interactions through direct contacts between adjacent cells (Artavanis-Tsakonas et al., 1999; Fortini, 2009; Andersson et al., 2011). The canonical ligands for Notch in Drosophila, Delta and Serrate, are also single-span transmembrane proteins (Artavanis-Tsakonas et al., 1999; D’Souza et al., 2008). The interaction between the extracellular domains of Notch and Delta or Serrate triggers the activation of Notch (Fehon et al., 1990; Kopan and Ilagan, 2009), which is initiated by cleavage of the extracellular domain of Notch by Kuzbanian (Kuz)/ADAM10 (referred to as the S2 cleavage) (Brou et al., 2000; Mumm et al., 2000; Lieber et al., 2002). This cleavage removes most of the Notch extracellular domain and produces a membrane-tethered form of the Notch intracellular domain (NEXT) (Zolkiewska, 2008). Subsequently, NEXT is cleaved within its transmembrane domain by γ-secretase (the S3 cleavage), which liberates the intracellular domain, termed NICD (De Strooper et al., 1999; Struhl and Greenwald, 1999; Ye et al., 1999).

NICD is then translocated to the nucleus and interacts with Suppressor of Hairless and Mastermind at the transcription regulatory element of Notch target genes (Bray, 2006; Bray and Bernard, 2010), to regulate their transcription. These downstream genes include wg and cut, which are expressed in the wing imaginal disc of Drosophila (Neumann and Cohen, 1996; de Celis and Bray, 1997). These core components and their functions are conserved evolutionarily (Kopan and Ilagan, 2009).

In addition to the core components of Notch signaling, other processes are required for the activation and regulation of the Notch signaling pathway. For example, the endocytosis of Notch and its ligands plays crucial roles in the activation and down-regulation of Notch signaling. In signal-sending cells, ubiquitin-dependent endocytosis of the ligands is essential for the activation of Notch in neighboring cells (Seugnet et al., 1997; Parks et al., 2000; Itoh et al., 2003; Le Borgne et al., 2005). Notch signaling activity is reduced in mutants in which the entry of Notch into the early endosomes is disrupted; these mutants include avalanche and Rab5 in Drosophila (Lu and Bilder, 2005). On the other hand, in mutants of genes encoding ESCRT components, Notch signaling is hyper-activated in a ligand-independent manner, and it accumulates in early endosomes (Moberg et al., 2005; Thompson et al., 2005; Vaccari and Bilder, 2005; Herz et al., 2006; Vaccari et al., 2009). These results suggest that the entry of Notch into endosomes regulates its activation and its subsequent breakdown in lysosomes (Fortini and Bilder, 2009).

Early endosomes mature into multivesicular bodies (MVBs) (Babst, 2011; Huotari and Helenius, 2011). MVBs contain intraluminal vesicles (ILVs), which are formed by the invagination and detachment of their limiting membrane (Piper and Katzmann, 2007). The ubiquitination of cell-surface receptors, including Notch, causes them to be sorted into ILVs. The MVBs then fuse with the lysosomes, and the proteins in the ILVs are degraded (Luzio et al., 2010).

ESCRT consists of the ESCRT-0, ESCRT-I, ESCRT-II and ESCRT-III complexes together with Vps4, an ATPase and other associated proteins (Henne et al., 2011). ESCRT-0, ESCRT-I and ESCRT-II bind to the Ubiquitin attached to membrane receptors and are involved in sorting ubiquitinated cargo and in recruiting and activating ESCRT-III (Shields and Piper, 2011). Vps20, Snf7, Vps24 and Vps2 are soluble cytosolic proteins that are subunits of ESCRT-III. Activated ESCRT-III proteins assemble into a detergent-insoluble complex on the limiting membrane of early endosomes or MVBs. Vps20, Snf7, and Vps24 then induce membrane scission to detach the ILVs (Babst et al., 2002; Adell and Teis, 2011). Vps2 is responsible for recruiting Vps4, which is required for ESCRT-III recycling.

The functions of ESCRT-encoding genes have been extensively studied in yeast (Henne et al., 2011). However, it was recently found that ESCRT proteins may have more divergent roles in metazoans (Rusten et al., 2012). In Drosophila, null mutations of many genes that encode ESCRT proteins were isolated by a systematic genetic screen for the overgrowth phenotype (Vaccari et al., 2009). These analyses revealed that all the ESCRT-encoding genes are required for the trafficking of ubiquitinated proteins and for preventing the hyper-activation of Notch signaling. However, unlike yeast ESCRTs, the findings suggested that Drosophila ESCRT-I, ESCRT-II and ESCRT-III have some functional differences.

A mutant of vps2 was previously reported in Drosophila (Vaccari et al., 2009). This vps2 allele is a nonsense mutant that is predicted to produce the amino-terminal half of the deduced Vps2 protein. Furthermore, the developmental defects associated with this vps2 mutant were only studied in a limited number of tissues (Vaccari et al., 2009). To investigate the roles of vps2 in Notch signaling, we first created a null allele by generating a deletion mutant that lacks the entire coding region of vps2. We also studied the roles of vps2 in various developmental contexts by using RNAi lines to knock down vps2 and by overexpressing it.

MATERIALS AND METHODS

Fly strains and genetics

All experiments were performed at 25°C on a standard Drosophila culture medium. Canton-S and dpp-GAL4/+ were used as wild-type controls (Klein and Arias, 1998).

The FLP/FRT system was used to generate somatic mutant clones (Xu and Rubin, 1993). To induce somatic clones homozygous for vps2² at high efficiency in wing imaginal discs, we used Ubx-FLP and FRT82B chromosome combined with either P{A92}RpS3^Plac92 or vps2², as described previously (Hutterer and Knoblich, 2005).

The following RNA interference (RNAi) lines were used: for vps2, vps2^GD8363 (Dietzl et al., 2007; Vienna Drosophila RNAi Center, Vienna, Austria) and 14542R-1 (National Institute of Genetics, Mishima, Japan). The GAL4 drivers used were dpp-GAL4 (Klein and Arias, 1998) and actin-GAL4 (Adachi-Yamada et al., 1999).

Generation of the vps2 deletion mutant

The deletion mutant of vps2 was generated by imprecise excision of a P-element (Hummel and Klämbt, 2008). To mobilize the P-element, P{GSV6} vps2^GS11024 virgin females were crossed with w*;;Dr¹/TMS, P{ry[+t7.2] = Δ2-3}99B males (Toba et al., 1999; FlyBase: http://flybase.org/). Potential deletion lines were crossed with Df(3R)ED6232, a deletion uncovering the vps2 locus, and selected based on the lethality of their progenies (Ryder et al., 2004, 2007).

Genomic DNA samples were obtained from these potential vps2 mutants and used as templates for PCR amplification of a fragment encompassing the vps2 locus (forward primer, 5’-TCGAACTGAACGGCTACGTC-3’; reverse primer, 5’-GCGGAGGTACACACACACTT-3’). The size of the PCR fragments was analyzed by agarose gel electrophoresis, and the lines with shorter fragments were selected as candidates for vps2 deletion mutants. To determine the DNA sequence deleted from the vps2 locus in these candidates, the PCR fragment amplified from their genomic DNA was sequenced using a standard protocol.

Construction of UAS-vps2

The full-length cDNA of vps2 was excised by EcoRI and KpnI digestion from the RH72336 clone (Stapleton et al., 2002) and sub-cloned into the EcoRI and KpnI sites in the pUAST vector (Brand and Perrimon, 1993). The structure of the vps2 cDNA was confirmed by DNA sequencing analysis.

In situ hybridization

In situ hybridization with a wingless (wg) digoxigenin-labeled RNA antisense probe was carried out as described previously (González-Crespo and Levine, 1993).

Generation of an anti-Vps2 antibody

A polypeptide (KISPDEMLRKNQRAL) corresponding to the deduced amino acid sequence of Vps2 from position 8 to 22 was used as the antigen (Tanpaku Seisei Kogyo, Takasaki, Japan). This polypeptide was conjugated with Keyhole limpet hemocyanin through a cysteine added to the peptide’s N-terminus, and the conjugate was injected intradermally into guinea pigs along with Freund's complete adjuvant. The antibodies raised against this Vps2 polypeptide were affinity purified (Tanpaku Seisei Kogyo, Takasaki, Japan).

Immunohistochemistry

The primary antibodies used in this study, their final dilutions, and sources were as follows: mouse anti-Notch C17.9C6 (1:300, Development Studies Hybridoma Bank [DSHB], Iowa, USA), mouse anti-Dl C594.9B (1:100, DSHB, Iowa, USA), rabbit anti-Rab7 (1:5000) (Tanaka and Nakamura, 2008), guinea pig anti-Hrs (1:500) (Lloyd et al., 2002), rabbit anti-Rab5 ab31261 (1/500, Abcam, Cambridge, UK), mouse anti-Cut 2B10 (1:100, DSHB, Iowa, USA), rabbit anti-GFP 598 (1:500, MBL, Nagoya, Japan), rat anti-GFP GF090R (1:1000, Nacalai Tesque, Kyoto, Japan), mouse anti-Fz2 12A.7 (1/50, DSHB, Iowa, USA), guinea pig anti-Vps2 (1/500, this study). DAPI D212 (1/500, Dojindo, Kamimashiki, Japan) was used for nuclear staining. Rhodamine phalloidin R415 (1/500, Life Technologies, Carlsbad, USA) was used to stain F-actin. Immunostaining of wing imaginal discs was performed as previously described (Matsuno et al., 2002). Confocal images were taken with an LSM5 PASCAL, LSM700 and LSM510 META (Carl Zeiss, Oberkochen, Germany).

RESULTS

A novel null allele of vps2 caused defects in the trafficking of Notch and Delta, similar to those caused by a previously reported vps2 allele

The deduced wild-type Vps2 protein of Drosophila is composed of 256 amino acid residues (Fig. 1A). The previously reported vps2 allele produces a predicted Vps2 polypeptide consisting of the first 110 amino acids (Vaccari et al., 2009). Because this allele might have residual or novel functions, we induced a deletion mutation by the imprecise excision of a P-element, GS11024, and used it to analyze the defects caused by the absence of the Vps2 protein (Fig. 1A).

Fig. 1.

Notch and Delta accumulated in cells homozygous for a vps2-null mutant. (A) Diagram showing the vps2 locus of the null mutation allele, vps2². GS11024 (top) had a P-element insertion (triangle) 28-bp downstream of the vps2 transcription initiation site. Open pointed boxes denote the exons of the vps2 gene. vps2² (bottom) lacks 1,187 bp of the vps2 locus, including all of its exons (indicated by parentheses). (B–E’’’) Wing imaginal discs of the third instar including mosaic clones homozygous for vps2² (enclosed by white lines). (B–B’’’) Double-staining with anti-Notch (magenta, B, B’’ and B’’’) and anti-Hrs (green, B’, B’’ and B’’’) antibodies. (C–C’’’) Double-staining with anti-Notch (magenta, C, C’’ and C’’’) and anti-Rab7 (green, C’, C’’ and C’’’) antibodies. (D–D’’’) Double-staining with anti-Delta (magenta, D, D’’ and D’’’) and anti-Hrs (green, D’, D’’ and D’’’) antibodies. (E–E’’’) Double-staining with anti-Delta (magenta, E, E’’ and E’’’) and anti-Rab7 (green, E’, E’’ and E’’’) antibodies. B’’, C’’, D’’ and E’’ are merged images of B and B’, C and C’, D and D’, and E and E’, respectively. B’’’, C’’’, D’’’ and E’’’ show higher magnifications of B’’, C’’, D’’ and E’’, respectively. White arrowheads indicate vesicles containing Notch or Delta double-stained with anti-Hrs or anti-Rab7 antibodies. (F–G’’’) Wing imaginal discs double-stained with anti-Hrs (green, F, F’’, F’’’, G, G’’ and G’’’) and anti-Rab7 (magenta, F’, F’’, F’’’, G’, G’’ and G’’’) antibodies. (F–F’’’) Wild-type. (G–G’’’) Wing imaginal discs including cells homozygous for vps2² (enclosed by white lines). F’’’ and G’’’ show higher magnifications of F’’ and G’’, respectively. White arrowheads indicate vesicles stained with anti-Hrs and anti-Rab7 antibodies. (H–I) Wing imaginal discs stained with an anti-Fz2 antibody. (H) Wild-type. (I) Wing imaginal disc including mosaic clones homozygous for vps2². Scale bars: 20 μm in B–E’’’, H and I; 10 μm in F–G’’’.

Our sequencing analysis revealed that GS11024 has a P-element insertion 28-bp downstream of the vps2 transcriptional initiation site. The GS11024 line was recessive lethal. The heterozygote of GS11024 and Df(3R)ED6232, which uncovers the vps2 locus, was also lethal, suggesting that GS11024 was a mutant of vps2.

Through the imprecise deletion of GS11024, we obtained a null mutant allele of vps2, named vps2², which lacked a 1,187-bp genomic region, from 64-bp upstream of the initiation codon to 292-bp downstream of the stop codon (Fig. 1A). Thus, the vps2² genome lacked the entire coding region of vps2 (Fig. 1A). We also found a relic P-element sequence (CATGATGAAATAAGTATACTTG) derived from the GSV6 vector, next to the deleted region. vps2² homozygotes were embryonic lethal (data not shown), like the previously reported mutant (Vaccari et al., 2009). The vps2²/Df(3R)ED6232 heterozygote was also embryonic lethal (where Df(3R)ED6232 uncovers the vps2 locus), suggesting that the lethality was associated with the vps2² mutation (data not shown).

We also generated an overexpression construct, UAS-vps2, which drove the overexpression of vps2 under the control of Gal4 (Brand and Perrimon, 1993). However, our attempts to rescue the recessive lethal phenotypes associated with vps2² by overexpressing UAS-vps2 under the control of actin-GAL4 failed, because the overexpression of UAS-vps2 induced developmental defects, including lethality (data not shown).

In normal development, Notch signaling is activated and induces the expression of its downstream target genes, such as wg and cut, along the dorsal-ventral compartment boundary in the wing imaginal discs of third-instar larvae (Neumann and Cohen, 1996; de Celis and Bray, 1997). To study the roles of vps2 in Notch signaling, we generated somatic mosaic clones using the FLP/FRT system in the wing imaginal discs (Xu and Rubin, 1993). We could generate only small clones of cells homozygous for vps2² in this organ (Fig. 1, B–E’’). As reported previously, we observed that Notch accumulated in the vps2 mutant clones (Vaccari et al., 2009). At the intracellular level, we found that, compared to the surrounding cells heterozygous for vps2², Notch and Delta accumulated at higher levels in the endocytic vesicles of the homozygous mutant cells (Fig. 1, B–E).

To study the endocytic vesicles in the homozygous mutant cells in more detail, we used antibodies to the Hrs and Rab7 proteins, which are, respectively, markers for early endosomes and MVBs (Lloyd et al., 2002; Tanaka and Nakamura, 2008). We found that both the early endosomes and MVBs were enlarged in the mosaic clones homozygous for vps2² (Fig. 1, B’, C’, D’, E’, G and G’). Notch and Delta accumulated predominantly in the early endosomes of these cells, although they were also found occasionally in MVBs (Fig. 1, B’’, B’’’, C’’, C’’’, D’’, D’’’, E’’ and E’’’). These results are consistent with the previous finding that Notch accumulates in early endosomes labeled by anti-Avalanche antibody staining (Vaccari et al., 2009).

In wild-type cells, little overlap was seen between Hrs-containing and Rab7-containing vesicles (Fig. 1, F’’ and F’’’). However, in mosaic clones homozygous for vps2², staining for Hrs and Rab7 overlapped frequently (Fig. 1, G’’ and G’’’). These results suggested that the integrity of the endosomal compartments may be disrupted in these cells.

We also found that Frizzled2 (Fz2), a transmembrane receptor for Wingless, accumulated in intracellular vesicles (Fig. 1I). Fz2’s function is reported to be regulated by endosomal recycling and lysosomal degradation (Rives et al., 2006). This result suggested that the trafficking defect associated with the vps2 mutation was not specific for Notch and Delta. Although the mechanisms that induce the enlargement of early endosomes are still unclear in Drosophila, this result is consistent with the known roles of Vps2 in promoting ESCRT recycling in yeast and mammals (Shim et al., 2008; Teis et al., 2008; Saksena et al., 2009).

vps2² showed defects in epithelial integrity similar to those of the previously reported vps2 allele

In our mosaic analysis, we did not observe large clones homozygous for vps2² (Fig. 1, B–E, G and I). This result is consistent with previous reports in which cells homozygous for mutant alleles of genes encoding other ESCRT components also failed to form large clones (Moberg et al., 2005; Thompson et al., 2005; Vaccari and Bilder, 2005; Herz et al., 2006). This phenomenon was shown to be due to cell competition, because under conditions in which apoptosis in these cells was suppressed, the clones over-proliferated (Thompson et al., 2005; Herz et al., 2006, 2009). Thus, it is likely that the small clones homozygous for vps2² resulted from their selective removal by cell competition in our experiments.

An over-proliferation phenotype was reported in mutants of ESCRT-I-, ESCRT-II- and ESCRET-III-encoding genes, under conditions in which the apoptosis associated with cell competition was suppressed in the homozygous mutant cells (Thompson et al., 2005; Herz et al., 2006, 2009; Vaccari et al., 2009). The previously isolated vps2 allele also induces over-proliferation in the eye imaginal discs under such conditions (Vaccari et al., 2009). To suppress cell competition, we induced somatic mosaic clones homozygous for vps2² at a very high frequency in the wing imaginal discs of RpS3 heterozygotes, in which FLP/FRT-mediated recombination was driven by Ubx-FLP (Hutterer and Knoblich, 2005).

In these wing imaginal discs, most of the cells were homozygous for vps2², as evidenced by the small areas expressing GFP, which indicated vps2² heterozygous cells (Fig. 2, B, H, K and L). We often observed a marked over-proliferation of cells in these wing imaginal discs, which fused with the haltere and leg discs (Fig. 2B). Even in wing imaginal discs that maintained a normal size, the epithelial structure was disorganized (Fig. 2, H–L). The apicobasal polarity of the cortical actin distribution was also severely disrupted in these cells (Fig. 2, J and L). These results are consistent with observations of the previous vps2 mutant in follicle cells (Vaccari et al., 2009). We also found that the position of the nuclei along the apicobasal axis of these cells was disorganized (Fig. 2L). The nuclei were occasionally fragmented, suggesting that apoptosis was induced (Fig. 2I).

Fig. 2.

Defects in the epithelial integrity of wing imaginal discs mostly composed of vps2² homozygous cells. (A–L) Wing imaginal discs composed of wild-type cells (A and C–G) or mostly of vps2² homozygous cells [indicated by the absence of GFP (green) expression in B and H–L]. These wing imaginal discs were stained with DAPI (blue, A, B, C, D, F, G, H, I, K and L) and Phalloidin (red, A, B, C, E, F, G, H, J, K and L). A and B show low-power field images of wing imaginal discs with leg and haltere discs (indicated in A). D, E, F and I, J, K show higher magnifications of the area enclosed by white squares in C and H, respectively. GFP-expressing cells (green) were heterozygous for the RpS3 mutation. Scale bars: 20 μm.

In these wing imaginal discs, it was difficult to investigate whether the Notch signaling activity was up-regulated or abolished, because the tissue integrity and structure were disrupted too severely to analyze the expression of the Notch signaling target genes. Therefore, we decided to reduce the vps2 function partially, by knocking down vps2 gene expression in wing imaginal discs by using RNAi.

Knockdown of vps2 by RNAi leads to the accumulation of Notch and Delta similar to the effect of the null allele of vps2

We next examined whether Notch and Delta would accumulate in cells producing double-stranded RNA corresponding to vps2 mRNA, as observed in the vps2² homozygous cells. The production of vps2 double-stranded RNA was driven in wing imaginal discs under the control of dpp-Gal4 (Klein and Arias, 1998). In these discs, vps2 was knocked down in a stripe across the dorsal-ventral compartment boundary, which was labeled by UAS-GFP expression (Fig. 3, A’–D’). Although the vps2 gene function was probably not abolished completely in these cells, judging from the almost normal epithelial morphology of this region, we found that Notch and Delta did accumulate intracellularly in this region (Fig. 3, C, C’, D and D’).

Fig. 3.

RNAi against vps2 results in the accumulation of Notch and Delta in early endosomes. (A–D’) Wing imaginal discs stained with antibody against Notch (magenta, A, A’, C and C’) or Delta (magenta, B, B’, D and D’). (A–B’) Wild-type wing imaginal discs expressing UAS-GFP alone (green) driven by dpp-GAL4 as a control. (C–D’) Wing imaginal discs producing double-stranded RNA of vps2 and expressing two copies of UAS-GFP under the control of dpp-GAL4 (green). A–D show single-channel images (magenta) of A’–D’, respectively. (E–I’’’) Higher magnification of wing imaginal disc epithelium expressing double-stranded RNA of vps2 and UAS-GFP (left of white lines; GFP staining is not shown) under the control of dpp-GAL4. (E–F’’’) Staining with anti-Notch (magenta, E, E’’, E’’’, F, F’’ and F’’’) and anti-Hrs (green, E’–E’’’) or anti-Rab7 (green, F’–F’’’) antibodies. (G–H’’’) Staining with anti-Delta (magenta, G, G’’, G’’’, H, H’’ and H’’’) and anti-Hrs (green, G’–G’’’) or anti-Rab7 (green, H’–H’’’) antibodies. E’’, F’’, G’’ and H’’ are merged images of E and E’, F and F’, G and G’, and H and H’, respectively. E’’’, F’’’, G’’’ and H’’’ show higher magnifications of the area enclosed by white squares in E’’, F’’, G’’ and H’’, respectively. White arrowheads indicate vesicles containing Notch or Delta stained with anti-Hrs or anti-Rab7 antibodies. (I–I’’’) Staining with anti-Hrs (green, I, I’’ and I’’’) and anti-Rab7 (magenta, I’–I’’’) antibodies. I’’’ shows higher magnification of the area enclosed by a white square in I’’. White arrowheads indicate vesicles stained with anti-Hrs and anti-Rab7 antibodies. Scale bars: 20 μm.

To confirm that the effect of the RNAi mimicked the effect of the null allele, we next determined the intracellular compartment in which the two proteins accumulated (Fig. 3, E–H’’’). Most of the vesicles containing Notch and Delta were enlarged and co-localized with Hrs (Fig. 3, E’’, E’’’, G’’ and G’’’) and not with Rab7 (Fig. 3, H’’ and J’’), indicating that these proteins accumulated primarily in the early endosomes when vps2 function was reduced by RNAi, as in cells homozygous for vps2². In these cells, the Hrs-positive and Rab7-positive vesicles overlapped more frequently than in wild-type cells (Fig. 3, I–I’’’). This finding suggests that the integrity of the endosomal compartment may have been disrupted, as found in the vps2 mutant cells. In addition, cells expressing UAS-GFP occasionally infiltrated the wild-type tissues (Fig. 3, C’ and D’), suggesting that the epithelial integrity might have been partly disrupted by the vps2 knockdown.

Knockdown of vps2 results in reduced Notch signaling

Given that the RNAi against vps2 disrupted the trafficking of Notch and Delta, we next examined the expression of the Notch signaling target genes cut and wg in the vps2-knockdown cells (Fig. 4). We used an anti-Cut antibody to detect the protein (Fig. 4, A and B) (Blochlinger et al., 1990). In wild-type wing imaginal discs, Cut protein was detected in a few rows of cells along the dorsal-ventral compartment boundary, as reported previously (Fig. 4A) (Blochlinger et al., 1993). However, in 50% of the wing imaginal discs, we did not detect anti-Cut labeling in regions expressing the anti-vps2 RNAi (50%, n = 10) (Fig. 4, B and B’). These results suggested that Notch signaling was reduced by the knockdown of vps2.

Fig. 4.

Notch signaling is reduced by the knockdown of vps2. (A–B’) Wing imaginal discs of the third instar larvae stained with an anti-Cut antibody (magenta). (A–A’) Wild-type wing imaginal disc expressing UAS-GFP (green) alone, driven by dpp-GAL4, as a control. (B–B’) Wing imaginal disc producing double-stranded RNA of vps2 in the region expressing two copies of UAS-GFP (green) driven by dpp-GAL4. White arrowhead indicates the region where cut expression was reduced. (C–D) The expression of wg in wing imaginal discs was detected by in situ hybridization. (C) Wild-type. (D) Wing imaginal disc producing double-stranded RNA of vps2 in the region expressing UAS-GFP driven by dpp-GAL4. Black arrowhead indicates the region where wg expression was reduced. Scale bars: 50 μm.

The expression of wg was also detected along the dorsal-ventral compartment boundary in wild-type wing imaginal discs, by in situ hybridization (Fig. 4C). Its expression, like that of Cut, was also abolished where vps2 was knocked down (50%, n = 12) (Fig. 4D). The shape of these wing imaginal discs was mostly normal, although slight deformation was occasionally observed (data not shown). Based on our findings on wg and cut expression, we concluded that the knockdown of vps2 resulted in the reduction of Notch signaling.

It was previously reported that, in eye imaginal discs, Notch signaling is ectopically activated in somatic mosaic clones homozygous for mutant vps2 (Vaccari et al., 2009). This observation appeared inconsistent with our present results, which demonstrated that the knockdown of vps2 by RNAi led to the attenuation of Notch signaling in wing imaginal discs. This discrepancy could be explained if different levels of reduced vps2 function have differential effects on the activity of Notch signaling. That is, a severe reduction or complete absence of vps2 function could result in the hyper-activation of Notch signaling, whereas a partial reduction in vps2 function might reduce Notch signaling.

Overexpression of vps2 also disrupts the trafficking of Notch and Delta

Vps2 is required for the disassembly of the ESCRT-III complex, via Vps4, which in turn is required for the recycling of ESCRT-III components (Teis et al., 2008; Saksena et al., 2009). Thus, the expression level of vps2 might need to be tightly regulated for normal endocytosis. To test this possibility, we overexpressed wild-type UAS-vps2 under the control of dpp-Gal4 in wing imaginal discs. To detect the wild-type Vps2 protein, we generated a polyclonal antibody against a Vps2 peptide.

By immunostaining wing imaginal discs with the anti-Vps2 antibody, we detected wild-type Vps2 in the region where the expression of UAS-vps2 was driven by dpp-Gal4 (Fig. 5, A and A’). The overexpressed Vps2 protein was occasionally detected in early endosomes, which were labeled with an anti-Rab5 antibody (Pfeffer, 2003; Wucherpfennig et al., 2003), and in MVBs, which were labeled with an anti-Rab7 antibody (Fig. 5, B–C’’’). Notch accumulates in MVBs, and its expression overlapped with that of Vps2 (Fig. 5, D–D’’’). However, our antibody failed to detect the endogenous Vps2 protein (Fig. 5, A and A’ and data not shown).

Fig. 5.

Overexpressed Vps2 localizes to early endosomes and MVBs. (A–D’’) Wing imaginal discs expressing UAS-vps2 and UAS-GFP under the control of dpp-GAL4. (A and A’) Staining with an anti-Vps2 antibody (magenta, A and A’). GFP (green, A’) marks the region overexpressing UAS-vps2. (B–D’’’) Higher magnification of wing imaginal disc epithelium expressing UAS-vps2 and UAS-GFP (left of white lines; GFP-staining is not shown) under the control of dpp-GAL4. Double-staining with anti-Vps2 (magenta, B, B’’, B’’’, C, C’’, C’’’, D, D’’ and D’’’) and anti-Rab5 (green, B’, B’’ and B’’’), anti-Rab7 (green, C’, C’’ and C’’’) or anti-Notch (green, D’, D’’ and D’’’) antibodies. B’’, C’’ and D’’ are merged images of B and B’, C and C’, D and D’, respectively. B’’’, C’’’ and D’’’ show higher magnifications of the area enclosed by white squares in A’’, B’’, C’’ and D’’, respectively. White arrowheads indicate vesicles containing Notch or Delta and stained with anti-Hrs or anti-Rab7 antibodies. Scale bars: 20 μm.

To examine whether vps2 overexpression affected the endocytic compartments, we immunostained vps2-overexpressing wing imaginal discs with antibodies against Hrs and Rab7 (Fig. 6, A–D), and found that the early endosomes were enlarged in the vps2-overexpressing region (Fig. 6, A and C). In addition, the number of MVBs was increased, and they too were slightly enlarged (Fig. 6, B and D). In these cells, the Hrs-positive and Rab7-positive vesicles showed little overlap, as found in wild-type cells (Fig. 1, F–F’’’; Fig. 6, E–E’’’), suggesting that the integrity of the endosomal compartments was largely maintained in the vps2-overexpressing cells.

Fig. 6.

Overexpression of vps2 results in enlarged early endosomes, in which Notch and Delta accumulate. (A–E’’’) Wing imaginal disc epithelium overexpressing UAS-vps2 and UAS-GFP (left of white lines; GFP staining is not shown) under the control of dpp-GAL4. (A–B’’’) Staining with anti-Hrs (green, A, A’’ and A’’’) or anti-Rab7 (green, B, B’’ and B’’’) and anti-Notch (magenta, A’, A’’, A’’’, B’, B’’ and B’’’) antibodies. (C–D’’) Staining with anti-Hrs (green, C, C’’ and C’’’) or anti-Rab7 (green, D, D’’ and D’’’) and anti-Delta (magenta, C’, C’’, C’’’, D’, D’’ and D’’’) antibodies. A’’, B’’, C’’, D’’ are merged images of A and A’, B and B’, C and C’, and D and D’, respectively. A’’’, B’’’, C’’’ and D’’’ show higher magnifications of the area enclosed by a white square in A’’, B’’, C’’ and D’’, respectively. White arrowheads indicate vesicles containing Notch or Delta and stained with anti-Hrs or anti-Rab7 antibodies. (E–E’’’) Staining with anti-Hrs (green) and anti-Rab7 (magenta) antibodies. E’’ is a merged image of E and E’. E’’’ shows higher magnification of the area enclosed by a white square in E’’. White arrowheads indicate vesicles double-stained with anti-Hrs and anti-Rab7 antibodies. (F and F’) Wing imaginal discs overexpressing UAS-vps2 and UAS-GFP (green) under the control of dpp-GAL4. Cut expression (magenta) was detected by anti-Cut antibody staining. Scale bars: 20 μm.

We then examined the subcellular distribution of Notch and Delta in cells overexpressing vps2 and found that both of these proteins accumulated in vesicles (Fig. 6, A–D). The vesicles that accumulated large amounts of Notch and Delta were strongly labeled by the anti-Hrs antibody, indicating that these proteins mostly localized to the early endosomes (Fig. 6, A’’, A’’’, C’’ and C’’’). Thus, excess Vps2 affected the structure of the early endosomes and led to Notch’s accumulation in them.

We also examined whether Notch signaling was affected in cells overexpressing vps2. The expression of Cut along the dorsal-ventral compartment boundary was examined in cells overexpressing vps2 under the control of dpp-Gal4. However, no effect on Cut expression was observed in these wing imaginal discs (Fig. 6, F and F’). Therefore, although the trafficking of Notch and Delta was affected by both the loss of vps2 function and its overexpression, the effects on Notch-signaling activity were different (Fig. 4, B, D ; Fig. 6F).

DISCUSSION

Vps2 controls the trafficking of Notch and Delta

In this study, we generated the first null mutant allele of Drosophila vps2, designated vps2², and used it to examine the contribution of this gene to the regulation of Notch signaling. It was previously shown that Vps2 controls the trafficking of Notch and the apicobasally polarized structure of the epithelium (Vaccari et al., 2009). We confirmed that the vps2² allele showed defects similar to those of the previously described vps2 mutant, which has a nonsense mutation (Vaccari et al., 2009).

We found that Notch and Delta accumulated primarily in early endosomes and less frequently in MVBs, in the absence of vps2 function. Vps2 is required for the disassembly of the ESCRT-III complex, which is composed of Vps20, Snf7 and Vps24, and promotes the recycling of this complex (Teis et al., 2008; Saksena et al., 2009). We speculate that the disruption of ESCRT-III recycling may interfere with the sorting of Notch and Delta into ILVs for their breakdown in lysosomes. However, it remains to be studied whether the recycling of ESCRT-III is indeed attenuated in the absence of vps2 in Drosophila.

Notch signaling activity is differentially affected by the extent of vps2 reduction

It was previously shown that Notch signaling is ectopically activated in mutants of ESCRT-encoding genes in eye imaginal discs (Vaccari et al., 2009). Furthermore, Vaccari et al. (2008) reported that a mutant of tsg101, a component of the ESCRT-I complex, shows ligand-independent activation of Notch signaling. Thus, it was proposed that Notch is hyper-activated if it accumulates ligand-independently within the limiting membrane of early endosomes (Fortini and Bilder, 2009).

In our analysis using vps2², it was difficult to test whether Notch signaling was up-regulated in cells homozygous for this mutation, because the overgrowth of the epithelium and disruption of its apicobasal polarity made it difficult to interpret the expression pattern of Notch target genes (Fig. 2). However, interestingly, we found that Notch signaling was decreased by a partial reduction of vps2 function by RNAi, in wing imaginal discs. Under these conditions, the epithelial integrity appeared mostly normal, although some infiltration of vps2-knockdown cells was occasionally observed. Thus, we speculate that Notch signaling activity can be up- or down-regulated in response to different levels of reduction in vps2 function. Alternatively, it is possible that the effect of partial reduction in vps2 function may be tissue-specific.

To explain the attenuation of Notch signaling when vps2 function was partially reduced, we need to introduce the idea that vps2 directly and/or indirectly contributes to the ligand-dependent activation of Notch, because the endogenous activation of Notch signaling, which is known to depend on Delta and Serrate, was reduced in these cases (Fig. 4, B and D) (Seugnet et al., 1997). The ligand-dependent activation of Notch signaling is proposed to occur at the cell surface or during an early step of endocytosis (Tagami et al., 2008; Vaccari et al., 2008). Therefore, we speculate that the vps2-mediated recycling of ESCRT-III may influence these early endocytic events of Notch, although the molecular mechanism of this potential effect still needs to be addressed.

Expression level of vps2 affects the structure of early endosomes

In this study, we found that the absence of vps2 function resulted in the enlargement of early endosomes and MVBs. However, the partial reduction of vps2 activity by RNAi led to the enlargement of early endosomes, but not MVBs. Thus, the contribution of Vps2 to the morphogenesis of early endosomes and MVBs has distinct thresholds. Given that both an excess and a deficit of Vps2 affected the structure of early endosomes and MVBs, we speculate that the expression level of vps2 needs to be tightly regulated to maintain the normal structure of these vesicles. This finding also suggests that the amount of Vps2 may be a crucial parameter controlling the recycling of ESCRT-III.

Although Notch accumulated in early endosomes in vps2-overexpressing cells, Notch signaling activity was not reduced in these cells. However, the knockdown of vps2 by RNAi resulted in the accumulation of Notch in early endosomes and in reduced Notch signaling. Thus, the accumulation of Notch in early endosomes does not necessarily mean that Notch signaling is reduced. We speculate that the properties of the enlarged early endosomes may be different under conditions involving an excess versus a deficit of Vps2, even though Notch and Delta accumulated in them in both situations.

ACKNOWLEDGMENT

We thank J. Knoblich for the Ubx-FLP stock, A. Nakamura for the anti-Rab7 antibody, and H. Bellen for the anti-Hrs antibody. We also thank the Bloomington Drosophila Stock Center at Indiana University, the Drosophila Genetic Resource Center at Kyoto Institute of Technology, and the Vienna Drosophila RNAi Center for fly stocks, the Development Studies Hybridoma Bank at the University of Iowa for antibodies, and the Drosophila Genomics Resource Center at Indiana University for DNA clones. We thank M. M. Taketo and the members of his laboratory for helpful discussion. This work was supported by Grant-in-Aid for a Japan Society for the Promotion of Science fellow.

REFERENCES

•  Adachi-Yamada,  T.,  Gotoh,  T.,  Sugimura,  I.,  Tateno,  M.,  Nishida,  Y.,  Onuki,  T., and  Date,  H. (1999) De novo synthesis of sphingolipids is required for cell survival by down-regulating c-Jun N-terminal kinase in Drosophila imaginal discs. Mol. Cell. Biol. 19, 7276–7286.
•  Adell,  M. A., and  Teis,  D. (2011) Assembly and disassembly of the ESCRT-III membrane scission complex. FEBS Lett. 585, 3191–3196.
•  Andersson,  E. R.,  Sandberg,  R., and  Lendahl,  U. (2011) Notch signaling: simplicity in design, versatility in function. Development 138, 3593–3612.
•  Artavanis-Tsakonas,  S.,  Rand,  M. D., and  Lake,  R. J. (1999) Notch signaling: cell fate control and signal integration in development. Science 284, 770–776.
•  Babst,  M. (2011) MVB vesicle formation: ESCRT-dependent, ESCRT-independent and everything in between. Curr. Opin. Cell Biol. 23, 452–457.
•  Babst,  M.,  Katzmann,  D. J.,  Estepa-Sabal,  E. J.,  Meerloo,  T., and  Emr,  S. D. (2002) Escrt-III: an endosome-associated heterooligomeric protein complex required for mvb sorting. Dev. Cell 3, 271–282.
•  Blochlinger,  K.,  Bodmer,  R.,  Jan,  L. Y., and  Jan,  Y. N. (1990) Patterns of expression of Cut, a protein required for external sensory organ development in wild-type and cut mutant Drosophila embryos. Genes Dev. 4, 1322–1331.
•  Blochlinger,  K.,  Jan,  L. Y., and  Jan,  Y. N. (1993) Postembryonic patterns of expression of cut, a locus regulating sensory organ identity in Drosophila. Development 117, 441–450.
•  Brand,  A. H., and  Perrimon,  N. (1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415.
•  Bray,  S. J. (2006) Notch signalling: a simple pathway becomes complex. Nat. Rev. Mol. Cell Biol. 7, 678–689.
•  Bray,  S., and  Bernard,  F. (2010) Notch targets and their regulation. Curr. Top. Dev. Biol. 92, 253–275.
•  Brou,  C.,  Logeat,  F.,  Gupta,  N.,  Bessia,  C.,  LeBail,  O.,  Doedens,  J. R.,  Cumano,  A.,  Roux,  P.,  Black,  R. A., and  Israël,  A. (2000) A novel proteolytic cleavage involved in Notch signaling: the role of the disintegrin-metalloprotease TACE. Mol. Cell 5, 207–216.
•  D’Souza,  B.,  Miyamoto,  A., and  Weinmaster,  G. (2008) The many facets of Notch ligands. Oncogene 27, 5148–5167.
•  de Celis,  J. F., and  Bray,  S. (1997) Feed-back mechanisms affecting Notch activation at the dorsoventral boundary in the Drosophila wing. Development 124, 3241–3251.
•  De Strooper,  B.,  Annaert,  W.,  Cupers,  P.,  Saftig,  P.,  Craessaerts,  K.,  Mumm,  J. S.,  Schroeter,  E. H.,  Schrijvers,  V.,  Wolfe,  M. S.,  Ray,  W. J., et al. (1999) A presenilin-1-dependent gamma-secretase-like protease mediates release of Notch intracellular domain. Nature 398, 518–522.
•  Dietzl,  G.,  Chen,  D.,  Schnorrer,  F.,  Su,  K. C.,  Barinova,  Y.,  Fellner,  M.,  Gasser,  B.,  Kinsey,  K.,  Oppel,  S.,  Scheiblauer,  S., et al. (2007) A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448, 151–156.
•  Fehon,  R. G.,  Kooh,  P. J.,  Rebay,  I.,  Regan,  C. L.,  Xu,  T.,  Muskavitch,  M. A., and  Artavanis-Tsakonas,  S. (1990) Molecular interactions between the protein products of the neurogenic loci Notch and Delta, two EGF-homologous genes in Drosophila. Cell 61, 523–534.
•  Fortini,  M. E. (2009) Notch signaling: the core pathway and its posttranslational regulation. Dev. Cell 16, 633–647.
•  Fortini,  M. E., and  Bilder,  D. (2009) Endocytic regulation of Notch signaling. Curr. Opin. Genet. Dev. 19, 323–328.
•  González-Crespo,  S., and  Levine,  M. (1993) Interactions between dorsal and helix-loop-helix proteins initiate the differentiation of the embryonic mesoderm and neuroectoderm in Drosophila. Genes Dev. 7, 1703–1713.
•  Henne,  W. M.,  Buchkovich,  N. J., and  Emr,  S. D. (2011) The ESCRT pathway. Dev. Cell 21, 77–91.
•  Herz,  H. M.,  Chen,  Z.,  Scherr,  H.,  Lackey,  M.,  Bolduc,  C., and  Bergmann,  A. (2006) vps25 mosaics display non-autonomous cell survival and overgrowth, and autonomous apoptosis. Development 133, 1871–1880.
•  Herz,  H. M.,  Woodfield,  S. E.,  Chen,  Z.,  Bolduc,  C., and  Bergmann,  A. (2009) Common and distinct genetic properties of ESCRT-II components in Drosophila. PLoS ONE 4, e4165.
•  Hummel,  T., and  Klämbt,  C. (2008) P-element mutagenesis. Methods Mol. Biol. 420, 97–117.
•  Huotari,  J., and  Helenius,  A. (2011) Endosome maturation. EMBO J. 30, 3481–3500.
•  Hutterer,  A., and  Knoblich,  J. A. (2005) Numb and alpha-Adaptin regulate Sanpodo endocytosis to specify cell fate in Drosophila external sensory organs. EMBO Rep. 6, 836–842.
•  Itoh,  M.,  Kim,  C. H.,  Palardy,  G.,  Oda,  T.,  Jiang,  Y. J.,  Maust,  D.,  Yeo,  S. Y.,  Lorick,  K.,  Wright,  G. J.,  Ariza-McNaughton,  L., et al. (2003) Mind bomb is a ubiquitin ligase that is essential for efficient activation of Notch signaling by Delta. Dev. Cell 4, 67–82.
•  Klein,  T., and  Arias,  A. M. (1998) Different spatial and temporal interactions between Notch, wingless, and vestigial specify proximal and distal pattern elements of the wing in Drosophila. Dev. Biol. 194, 196–212.
•  Kopan,  R., and  Ilagan,  M. X. (2009) The canonical Notch signaling pathway: unfolding the activation mechanism. Cell 137, 216–233.
•  Le Borgne,  R.,  Bardin,  A., and  Schweisguth,  F. (2005) The roles of receptor and ligand endocytosis in regulating Notch signaling. Development 132, 1751–1762.
•  Lieber,  T.,  Kidd,  S., and  Young,  M. W. (2002) kuzbanian-mediated cleavage of Drosophila Notch. Genes Dev. 16, 209–221.
•  Lloyd,  T. E.,  Atkinson,  R.,  Wu,  M. N.,  Zhou,  Y.,  Pennetta,  G., and  Bellen,  H. J. (2002) Hrs regulates endosome membrane invagination and tyrosine kinase receptor signaling in Drosophila. Cell 108, 261–269.
•  Lu,  H., and  Bilder,  D. (2005) Endocytic control of epithelial polarity and proliferation in Drosophila. Nat. Cell Biol. 7, 1232–1239.
•  Luzio,  J. P.,  Gray,  S. R., and  Bright,  N. A. (2010) Endosome-lysosome fusion. Biochem. Soc. Trans. 38, 1413–1416.
•  Matsuno,  K.,  Ito,  M.,  Hori,  K.,  Miyashita,  F.,  Suzuki,  S.,  Kishi,  N.,  Artavanis-Tsakonas,  S., and  Okano,  H. (2002) Involvement of a proline-rich motif and RING-H2 finger of Deltex in the regulation of Notch signaling. Development 129, 1049–1059.
•  Moberg,  K. H.,  Schelble,  S.,  Burdick,  S. K., and  Hariharan,  I. K. (2005) Mutations in erupted, the Drosophila ortholog of mammalian tumor susceptibility gene 101, elicit non-cell-autonomous overgrowth. Dev. Cell 9, 699–710.
•  Mumm,  J. S.,  Schroeter,  E. H.,  Saxena,  M. T.,  Griesemer,  A.,  Tian,  X.,  Pan,  D. J.,  Ray,  W. J., and  Kopan,  R. (2000) A ligand-induced extracellular cleavage regulates gamma-secretase-like proteolytic activation of Notch1. Mol. Cell 5, 197–206.
•  Neumann,  C. J., and  Cohen,  S. M. (1996) A hierarchy of cross-regulation involving Notch, wingless, vestigial and cut organizes the dorsal/ventral axis of the Drosophila wing. Development 122, 3477–3485.
•  Parks,  A. L.,  Klueg,  K. M.,  Stout,  J. R., and  Muskavitch,  M. A. (2000) Ligand endocytosis drives receptor dissociation and activation in the Notch pathway. Development 127, 1373–1385.
•  Pfeffer,  S. (2003) Membrane domains in the secretory and endocytic pathways. Cell 112, 507–517.
•  Piper,  R. C., and  Katzmann,  D. J. (2007) Biogenesis and function of multivesicular bodies. Annu. Rev. Cell Dev. Biol. 23, 519–547.
•  Rives,  A. F.,  Rochlin,  K. M.,  Wehrli,  M.,  Schwartz,  S. L., and  DiNardo,  S. (2006) Endocytic trafficking of Wingless and its receptors, Arrow and DFrizzled-2, in the Drosophila wing. Dev. Biol. 293, 268–283.
•  Rusten,  T. E.,  Vaccari,  T., and  Stenmark,  H. (2012) Shaping development with ESCRTs. Nat. Cell Biol. 14, 38–45.
•  Ryder,  E.,  Blows,  F.,  Ashburner,  M.,  Bautista-Llacer,  R.,  Coulson,  D.,  Drummond,  J.,  Webster,  J.,  Gubb,  D.,  Gunton,  N.,  Johnson,  G., et al. (2004) The DrosDel collection: a set of P-element insertions for generating custom chromosomal aberrations in Drosophila melanogaster. Genetics 167, 797–813.
•  Ryder,  E.,  Ashburner,  M.,  Bautista-Llacer,  R.,  Drummond,  J.,  Webster,  J.,  Johnson,  G.,  Morley,  T.,  Chan,  Y. S.,  Blows,  F.,  Coulson,  D., et al. (2007) The DrosDel deletion collection: a Drosophila genomewide chromosomal deficiency resource. Genetics 177, 615–629.
•  Saksena,  S.,  Wahlman,  J.,  Teis,  D.,  Johnson,  A. E., and  Emr,  S. D. (2009) Functional reconstitution of ESCRT-III assembly and disassembly. Cell 136, 97–109.
•  Seugnet,  L.,  Simpson,  P., and  Haenlin,  M. (1997) Requirement for dynamin during Notch signaling in Drosophila neurogenesis. Dev. Biol. 192, 585–598.
•  Shields,  S. B., and  Piper,  R. C. (2011) How ubiquitin functions with ESCRTs. Traffic 12, 1306–1317.
•  Shim,  S.,  Merrill,  S. A., and  Hanson,  P. I. (2008) Novel interactions of ESCRT-III with LIP5 and VPS4 and their implications for ESCRT-III disassembly. Mol. Biol. Cell 19, 2661–2672.
•  Stapleton,  M.,  Liao,  G.,  Brokstein,  P.,  Hong,  L.,  Carninci,  P.,  Shiraki,  T.,  Hayashizaki,  Y.,  Champe,  M.,  Pacleb,  J.,  Wan,  K., et al. (2002) The Drosophila gene collection: identification of putative full-length cDNAs for 70% of D. melanogaster genes. Genome Res. 12, 1294–1300.
•  Struhl,  G., and  Greenwald,  I. (1999) Presenilin is required for activity and nuclear access of Notch in Drosophila. Nature 398, 522–525.
•  Tagami,  S.,  Okochi,  M.,  Yanagida,  K.,  Ikuta,  A.,  Fukumori,  A.,  Matsumoto,  N.,  Ishizuka-Katsura,  Y.,  Nakayama,  T.,  Itoh,  N.,  Jiang,  J., et al. (2008) Regulation of Notch signaling by dynamic changes in the precision of S3 cleavage of Notch-1. Mol. Cell. Biol. 28, 165–176.
•  Tanaka,  T., and  Nakamura,  A. (2008) The endocytic pathway acts downstream of Oskar in Drosophila germ plasm assembly. Development 135, 1107–1117.
•  Teis,  D.,  Saksena,  S., and  Emr,  S. D. (2008) Ordered assembly of the ESCRT-III complex on endosomes is required to sequester cargo during MVB formation. Dev. Cell 15, 578–589.
•  Thompson,  B. J.,  Mathieu,  J.,  Sung,  H. H.,  Loeser,  E.,  Rørth,  P., and  Cohen,  S. M. (2005) Tumor suppressor properties of the ESCRT-II complex component Vps25 in Drosophila. Dev. Cell 9, 711–720.
•  Toba,  G.,  Ohsako,  T.,  Miyata,  N.,  Ohtsuka,  T.,  Seong,  K. H., and  Aigaki,  T. (1999) The gene search system: A method for efficient detection and rapid molecular identification of genes in Drosophila melanogaster. Genetics 151, 725–737.
•  Vaccari,  T., and  Bilder,  D. (2005) The Drosophila tumor suppressor vps25 prevents nonautonomous overproliferation by regulating notch trafficking. Dev. Cell 9, 687–698.
•  Vaccari,  T.,  Lu,  H.,  Kanwar,  R.,  Fortini,  M. E., and  Bilder,  D. (2008) Endosomal entry regulates Notch receptor activation in Drosophila melanogaster. J. Cell Biol. 180, 755–762.
•  Vaccari,  T.,  Rusten,  T. E.,  Menut,  L.,  Nezis,  I. P.,  Brech,  A.,  Stenmark,  H., and  Bilder,  D. (2009) Comparative analysis of ESCRT-I, ESCRT-II and ESCRT-III function in Drosophila by efficient isolation of ESCRT mutants. J. Cell Sci. 122, 2413–2423.
•  Wucherpfennig,  T.,  Wilsch-Brauninger,  M., and  González-Gaitán,  M. (2003) Role of Drosophila Rab5 during endosomal trafficking at the synapse and evoked neurotransmitter release. J. Cell Biol. 161, 609–624.
•  Xu,  T., and  Rubin,  G. M. (1993) Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 117, 1223–1237.
•  Ye,  Y.,  Lukinova,  N., and  Fortini,  M. E. (1999) Neurogenic phenotypes and altered Notch processing in Drosophila Presenilin mutants. Nature 398, 525–529.
•  Zolkiewska,  A. (2008) ADAM proteases: ligand processing and modulation of the Notch pathway. Cell. Mol. Life Sci. 65, 2056–2068.