Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nature. Author manuscript; available in PMC 2010 January 2.
Published in final edited form as:
PMCID: PMC2777527

Genome-Wide Silencing in Drosophila Captures Conserved Apoptotic Effectors


Apoptosis is a conserved form of programmed cell death (PCD) firmly established in the etiology, pathogenesis and treatment of many human diseases. Central to the core machinery of apoptosis are the caspases and their proximal regulators. Current models for caspase control envision a balance of opposing elements, with variable contributions from positive regulators and negative regulators among different cell types and species1. To advance a comprehensive view of components that support caspase-dependent cell death, we conducted a genome-wide silencing screen in the Drosophila model. Our strategy combined a library of dsRNAs together with a chemical antagonist of Inhibitor of Apoptosis Proteins (IAPs) that simulates the action of native regulators in the Reaper/Smac family2. A highly validated set of targets necessary for death provoked by multiple stimuli was identified. Among these, Tango7 is advanced here as a novel effector. Cells depleted for this gene resisted apoptosis at a step prior to induction of effector caspase activity and directed silencing of Tango7 in the animal prevented caspase-dependent PCD. Unlike known apoptosis regulators in this model3, Tango7 activity did not influence stimulus-dependent loss of Drosophila IAP1 (DIAP1) but, instead, regulated levels of the apical caspase Dronc. Likewise, the human Tango7 counterpart, PCID1, similarly impinged on caspase 9, revealing a novel regulatory axis impacting the apoptosome.

Mammalian Smac proteins are thought to be functional orthologues of the Drosophila IAP antagonists referred to as RHG proteins (Reaper, Head Involution Defective, Grim, Sickle, Jafrac)3. Therefore, we tested the possibility that a small molecule Smac-mimetic (SM)4 might simulate the action of RHG proteins in Drosophila cells. Like RHG proteins5, the SM compound specifically bound DIAP1, a central brake against caspases3 (Fig. 1a). In Drosophila cell lines, the SM compound induced stereotypical apoptosis (Fig. 1b, supplementary movies), that was completely reversed either by peptide caspase inhibitors or RNAi depletion of the apical caspase Dronc (Fig. 1c), mirroring haemocytes lacking apoptosomal genes6,7. Together, the findings establish the SM compound as a molecular mimetic of RHG proteins.

Figure 1
A smac-mimetic (SM) compound exerts broad cross-species IAP antagonist activity

We conducted an RNAi-based genome scale screen to identify essential apoptotic determinants using the SM compound as a proximal apoptogenic stimulus (Fig. 2a, methods). Our platform used S2R+ cells and a library of dsRNAs targeting 13,071 annotated genes in Drosophila genome build 3 (Fig. 2a, methods) to capture targets that prevent SM-induced killing when silenced (Fig. 2c). Each dsRNA (supplemental Table 4) was tested in triplicate. Effects were typically reproducible (Fig. 2b) and assay quality, measured by Z-factor values, was consistently high8. Using both plate and position mean centering analyses (methods), we identified 42 candidate genes with protective activity meeting or exceeding a stringent z-score ≥3.1 threshold (P<0.001). This collection includes the expected benchmark targets Dronc, Dark and the effector caspase Drice (Fig. 2c, Table 1, supplemental Table1).

Figure 2
A genome-wide screen captures apoptogenic effectors
Table 1
Characterization of High Rank Candidate Genes

33 of the 42 candidate genes with a z-score greater than 3.1 were retested using a different dsRNA amplicon (Fig. 2d)9. To examine the biological relevance of this threshold, 42 targets conferring less potent activity (2.1<z<3.1) were also retested (supplemental Table 2). 19 candidates retested at least 3.1 standard deviations above the control mean, 16 of which had z>3.1 in the original screen, while only 3 that survived this retest standard originally had 2.1<z<3.1. Hence, our initial z threshold was experimentally relevant and predictive for reproducibility across different dsRNA amplicons.

We selected a set of 13 priority targets (Table 1) for further characterization based on conservation and/or implied links to human disease or development. For these, our validation strategy applied different assay methods, distinct apoptotic triggers and alternative non-overlapping dsRNAs10. We assessed new amplicons targeting each candidate by measuring effector caspase activity after UV or cycloheximide challenge, two well-established apoptotic stimuli11. These data (Table 1) discriminate several target groups including: 1) genes that exert substantial activity across multiple stimuli (Drice, Hrb27C, enok, Tango7), 2) stimulus-dependent genes (CG32626, CG7275) and 3) genes with mild but consistently significant activity (SNF4Agamma, eIF5, sktl).

Depletion of Tango7 (CG8309) prevented killing induced by multiple apoptotic stimuli. Non-overlapping dsRNAs that target this gene (Tango7#1 and Tango7#2) conferred protective activity comparable to benchmark amplicons that silence Dronc and Dark (Table 1, Fig. 3b). When tracked over a 22 hour time course following UV challenge, more than 95% of control cells initiated blebbing and died. However, like samples depleted for Dronc, cells silenced for Tango7 were similarly prevented from apoptotic blebbing when directly visualized over this same period (Fig. 3a, supplementary movies). Together with DEVDase assays in Figure 3b, our results indicate that Tango7 acts prior to effector caspase activation and prior to steps involved in apoptotic blebbing.

Figure 3
Tango7 is an effector of apoptosis

The combined results from multiple amplicons and three distinct assays validate Tango7 as an apoptogenic effector in S2R+ cells. To test the relevance of Tango7 during PCD in vivo, Tango7 dsRNAs were targeted to the pupal retina where excessive interommatidal cells are eliminated through caspase-dependent cell death3. Two different insertions of the 8309R1 transgene (Tango7R1 and Tango7R3) were compared to an irrelevant control dsRNA transgene (methods). Like retinas mutated for apical or effector caspases12,13, retinas expressing either Tango7R1 or Tango7R3 retained significant numbers of extra interommatidial cells (Figure 3d–f). This effect was similar to animals lacking drice13 and was not seen with an irrelevant dsRNA strain from the same collection. Depletion of Tango7 in the wing also prompted a range of defects which were consistent with failures in post-eclosion cell death and similar to canonical PCD mutants6,12,14,15 (supplemental Figure 1a–d). Low-penetrance phenotypes not shared with other PCD mutants suggest that Tango7 may not be dedicated to apoptotic functions. Unlike canonical cell death mutants, occasional scars appeared in a minority of adult retinas silenced for Tango7 (supplemental Figure 1e, f) and, hence, extra interommatidial cells could conceivably arise through activity unrelated to cell death. However, since scars are not evident in pupal stages, these infrequent defects manifest long after pruning of interommatidial cells and, more likely, they reflect pleiotropic roles for Tango7 in cells that are not fated to die.

To investigate the mode of apoptosis regulation by Tango7, core determinants were investigated. Surprisingly, Tango7 did not affect basal or stimulus-dependent loss of DIAP1 protein levels (Fig. 4a), a central point of control commonly observed during canonical apoptosis signaling in Drosophila3. Levels of the apoptosomal protein Dark were also unaffected (Fig. 4a). By contrast, silencing of Tango7 caused a marked decline in pro-Dronc levels (Fig. 4b). This relationship was also observed in the Dark depleted state (Figure 4b) which causes pro-Dronc accumulation to unusually high steady state concentrations16. Therefore, under a wide dynamic range, Tango7 positively regulates levels of this caspase. The findings also implicate a post-transcriptional mechanism linking Tango7 activity to Dronc protein levels since, within the limits of our semi-quantitative rtPCR assay, the treatments had little or no effect on Dronc mRNA abundance (Fig. 3c).

Figure 4
Tango7 and its human ortholog regulate apoptosomal caspases

To examine if the relationship between Tango7 and Dronc (Fig 4b) reflects a conserved regulatory axis, we tested whether the human ortholog of Tango7, PCID1, might similarly impact caspase 9. Human T98G cells17 or IMR90E1A cells18 were treated with gene-specific siRNAs and steady state levels of pro-caspase 9 were determined. As in Drosophila, a similar relationship between human counterparts was exposed in cell lines of distinct origins (Fig 4c). Here, PCID1 siRNAs caused a repression of pro-caspase 9 levels, with effects that were analogous to siRNAs that directly target caspase 9. To further explore links between PCID1 and apoptosis, siRNA-treated cells were challenged with an apoptogenic stimulus (UV) and tested for induction of effector caspase activity. As seen in Figure 4d, silencing either PCID1 or caspase 9 attenuated stimulus-dependent DEVDase. Furthermore, co-silencing of both targets reduced activity comparable to depletion of the pro-apoptotic effector, Bak19. Hence, in cells where levels of caspase 9 are rate-limiting, PCID1 is a relevant determinant of caspase activity. We also measured ATP in these experiments (not shown) and, by this criterion, neither caspase 9 siRNAs nor PCID1 siRNAs appeared cytoprotective. However, this result was anticipated since, unlike caspase-inhibited fly cells3, mammalian cells often die by alternate forms of cell death when caspase activity is blocked19,20.

The highly validated gene set in Table 1 presents new opportunities for understanding how cells integrate death signals and execute apoptogenic responses. Like all RNAi screens, only rate-limiting targets were captured and Drosophila effectors not present in S2R+ cells were missed. Of note, Dark, Dronc and Drice dsRNAs were recovered among the top candidates, providing reassurance for the logistics of our platform. Targets recovered through our screen exposed previously unknown modalities that support apoptosis but the extent to which most of these generalize beyond the original Drosophila culture system is not known. In principle, even the effector featured here, Tango7, might regulate caspase-dependent cell death and Dronc in just S2R+ cells. However, this scenario is quite unlikely for at least two reasons. First, when Tango7 is silenced in the retina, the exact same cells which persist are those which are known to die via caspase-dependent, canonical pathways. Second, our Drosophila studies successfully predicted cross-species apoptogenic activity for the counterpart of Tango7 in human cells of distinct origin.

Tango7 depletion uniquely illustrates how cell survival can be sustained (despite stimulus-dependent loss of DIAP1, Fig. 4a), exposing a novel axis of regulation that bypasses a common apoptotic switch in flies and, instead, impacts the apoptosome3. We found no evidence for a physical association between Tango7 and Dronc protein (not shown) and, despite less severe effects upon pro-Dronc levels, rescue by dsRNAs targeting Tango7 were generally comparable to those targeting Dronc. Hence, Tango7 could affect the post-translational processing or sub-cellular localization of pro-Dronc. Alternatively, Tango7 could impact other apoptotic effectors, exerting affects beyond simply Dronc regulation. Consistent with these possibilities, Tango7 was linked to golgi activity21 but, since genes needed for general golgi functions were not recovered here, links between Tango7 and Dronc evidently do not reflect compromised golgi. Whatever the mechanism, this regulatory axis appears to be conserved since the Tango7 counterpart, PCID1, similarly regulated caspase 9 levels and DEVDase in human cells (Fig 4C). In a recent study of pancreatic cancers22, PCID1 expression was repressed at least six-fold in all five patient tumors profiled. Furthermore, PCID1 was also implicated as a determinant of Herpes Simplex Virus susceptibility23. It will be important to examine whether these possible links to human disease reflect apoptosis related activity or, perhaps, non-apoptotic functions reported for caspase 924.


RNAi screen and validation

18,000 cells in serum/antibiotic-free Schneider's media were seeded in each well containing 0.5μg dsRNA (Ambion Silencer RNAi library or synthesized dsRNA) and incubated for 1 hour before adding 100μl serum-supplemented media. 3 days later, the media was replaced with SM-containing media. Cell viability was assayed 2 days later using CellTiter-Glo (Promega) in a plate reader (Envision multimode). Every assay plate included 4 control wells containing Dronc dsRNA. Of 429 sample plates assayed in the screen, 76% (325 plates) had excellent assay quality (Z≥0.5) and only 2% were unacceptable (Z<0). The Z-factor for each plate8 and plate mean centered z-score for each well (n=92 sample wells) were calculated. To correct for systematic bias/edge effects25, the position mean centered z-score for each well is calculated by its plate mean centered z-score minus the position average, divided by the position standard deviation (n=143 plates per triplicate). Genes with z>3.1 (P<0.001) from either plate mean or position mean centered normalizations were considered together as primary candidates. For secondary screens and subsequent RNAi experiments, dsRNA synthesis and treatment was as described9,26 (Supplementary Table 3). Briefly, cells were cultured 72 hours to deplete target gene product before analysis. Effector caspase (DEVDase) activity was measured using Caspase3/7-Glo (Promega) 6 hours after apoptosis stimulation.


Fly stocks

Tango7R1 denotes the GAL4-responsive the RNAi line targeting CG8309 with a genotype w; P{UAS-8309R-1}2 on the 2nd chromosome (NIG, Japan). This transgene was mobilized by standard methods and an insertion, designated TangoR3, was recovered on the 3rd chromosome. 3523R2 denotes the GAL4-responsive CG3523 RNAi line w; P{UAS-3523R-2}3 (stock#3523R-2, NIG). 5887R1 denotes the GAL4-responsive CG5887 RNAi line w; P{UAS-5887R-1} (stock#5887R-1, NIG). For pupal eye analyses, genotypes for Tango7 are UAS-dicer2/+; gmr-GAL4/UAS-Tango7R1, or UAS-dicer2/+; gmr-GAL4/+; UAS-Tango7R3/+; or, as control, UAS-dicer2/+; gmr-GAL4/UAS-5887R1. The 3 independent wing-specific vestigial-GAL4 driver lines, vg-GAL4#1 denotes y1 w1118; P{w+mC=vgM-GAL4.Exel}3 (stock#8223, Bloomington), vg-GAL4#2 denotes P{w+mC=vgMQ-GAL4.Exel}1, y1 w1118 (stock#8231, Bloomington) and vg-GAL4#3 denotes y w hsp70-flp; P{mini w+, GAL4}vgBE P{mini w+, UAS-flp}; P{y+, hsp70-cd2} P{mini w+, FRT}79 P{ry+, hsp70-neo, FRT}82 P{mini w+, hsp70-HA-gfp, smo+}.

Cell culture, UV challenge and western analyses

S2R+ cells were cultured in Schneider's media with 10% Fetal Bovine Serum, 25U/mL penicillin, 25μg/mL streptomycin at 25°C (all cell culture reagents from Invitrogen except FBS, Atlas Biologicals). Unless otherwise indicated, media in all procedures refer to the above. T98G and IMR90E1A cells were cultured in DMEM supplemented with 10% FBS. For western analysis, samples were lysed in buffer A with 1% TX-100. Blots were probed with 1:1000 α-DIAP1 (gift from Kristin White), 1:1000 α-Dronc (gift from Bruce Hay), 1:1000 α-Dark7, 1:5000 α-tubulin (E7, Developmental Studies Hybridoma Bank, University of Iowa). Drosophila cells treated with two dsRNAs were exposed to 15μg/ml of each indicated dsRNA, for a total dsRNA concentration of 30μg/ml. AmpRx2 denotes cells exposed to 30μg/ml of control AmpR dsRNA. Cells were analyzed either untreated or 6 hours after UV challenge (100 mJ/cm2). Relative changes in band intensity were obtained by comparing integrated densitometric values (IDV). Human cells were harvested five days after siRNA transfection, and 10μg total protein were separated by 10% SDS-PAGE. Caspase-9 and PCID1 levels were analyzed using anti-caspase 9 (Cell Signaling) and α-PCID1 (Proteintech Group, Inc.).

Smac-mimetic binding

Cells were lysed in 20 mM HEPES-KOH [pH 7.5], 10 mM KCl, 1.5 mM MgCl2, 1 mM sodium EDTA, 1 mM sodium EGTA, 1 mM dithiothreitol (buffer A), 1% Triton-X100 buffer with protease inhibitors (#1836170 Roche). Biotinylated-SM or Biotin was incubated with avidin-conjugated beads and 10mg/ml BSA in PBS for 4 hours at 4 °C. The beads were then incubated with precleared lysate (~1–2 mg protein) overnight before analysis.

Primary and secondary RNAi Screen

The primary screen was optimized for S2R+ cells and conducted using a `direct soaking' protocol, together with the Silencer Drosophila RNAi library (Ambion/Applied Biosystems) targeting 13,071 genes from Drosophila genome build 3. Sequences for all dsRNAs in the library are available in Table 4 of supplemental materials. The RNAi library or synthesized dsRNAs were plated in 96-well microplates (Corning) using Beckman FX liquid handlers. Every assay plate included 4 control wells containing Dronc dsRNA. Each well contained 0.5μg dsRNA in 20μl of SF900 media. 18,000 cells in 30μl serum/antibiotic-free Schneider's media were seeded in each well and incubated for 1 hour before adding 100μl media. 3 days later, the media was replaced with 80μl 2.5μM SM media. Cell viability was assayed 2 days later using CellTiter-Glo (Promega) in a plate reader (Envision multimode). The plate mean centered z-score for each well is its luminescence value minus the plate average, divided by the plate standard deviation (n=92 sample wells). The Z-factor for each plate was calculated as described8. Low quality (Z<0) plates were excluded from subsequent secondary retests and analyses. To correct for systematic bias/edge effects25, the position mean centered z-score for each well is calculated by its plate mean centered z-score minus the position average, divided by the position standard deviation (n=143 plates per triplicate). Average position mean centered z-score for each amplicon is calculated from the individual position mean centered z-scores of the triplicates. Genes with z>3.1 from either plate mean or position mean centered normalizations were considered together as primary candidates. For secondary screens and subsequent RNAi experiments, dsRNAs synthesis and treatment was as described 9,26 (Supplementary Table 3). Cells were cultured 72 hours to deplete target gene product before analysis. Effector caspase (DEVDase) activity was measured using Caspase3/7-Glo (Promega) 6 hours after apoptosis stimulation according to manufacturer's instructions.

Time-lapse microscopy

Cells pretreated with dsRNAs were tracked by time-lapse microscopy for 22 hrs after UV exposure (90mJ/cm2 UVC). Photomicrographs were captured at 6-minute intervals on a Zeiss Axiovert 200M inverted microscope equipped with a Marzhauser programmable stage, Nikon DXM1200F camera and controlled by Metamorph software (Molecular Devices). Movie images were assembled and analysed using ImageJ (NIH). At least 100 cells were tracked in each treatment. Dead cells are scored by morphology of rounding up, blebbing or fragmentation into corpses.

Pupal eye analyses

White prepupae were isolated at 0hr and aged 46–48hr. at 25°C. To outline cell borders, pupal eyes were dissected and stained with mouse α-DiscsLarge (Developmental Studies Hybridoma Bank, 1:1000) and α-mouse Fluorecein (Vector Laboratories, 1:250) as described in6,14. Interommatidial cell counts were modified from27. Hexagonal units connecting six `ommatidia centers' that completely surround one ommatidium were counted. Secondary or tertiary pigment cells inside or partly inside hexagonal assemblies were counted as a single cell.

siRNA Transfection

siRNAs to CASP9 and PCID1 were ON-TARGETplus SMARTpool siRNA, and the control siRNA used was ON-TARGETplus Non-targeting pool (Dharmacon). T98G and IMR90E1A cells were transfected with 64 nM siRNAs targeting the genes indicated using Dharmafect Duo (Dharmacon) and Optimem according to manufacturer's instructions. Pools contain 4 siRNAs for each indicated gene. Five days later, transfected cells were challenged with UV (100 mJ/cm2). Effector caspase (DEVDase) activity was measured using Caspase3/7-Glo (Promega) 18 hours after UV stimulation.

Supplementary Material

Supplementary Information

Supplementary Table 4

Supplementary movie 1

Supplementary movie 2

Supplementary movie 3

Supplementary movie 4


We thank Xiaoqin Tu, Leslie Durham, Lawrence Lum, Phil Beachy, Michael Roth, Xiaodong Wang, Patrick Harran, Yuri Lazebnik, David Dorris, the UTSW High Throughput Screening lab and the Live Cell Imaging facility for technical and material support; Bloomington Stock Center, Konrad Basler, National Institute of Genetics (NIG) stock center (Japan) for fly lines; Kristin White, Bruce Hay and the Hybridoma Bank for antibodies; Vivek Malhotra for pMT-Tango7 plasmid; Joachim Seeman and the Wang lab for discussions; and Margaret Allen for administrative support. This work was supported by grants from NIGMS, NIAAA and the UTSW High Impact/High Risk Grant Program. N.L is supported by an NRSA.


Author information The authors declare no competing financial interests. Reprints and permissions information is available at

Supplementary Information is linked to the online version of the paper at


1. Salvesen GS, Abrams JM. Caspase activation - stepping on the gas or releasing the brakes? Lessons from humans and flies. Oncogene. 2004;23(16):2774. [PubMed]
2. Verhagen AM, Vaux DL. Cell death regulation by the mammalian IAP antagonist Diablo/Smac. Apoptosis. 2002;7(2):163. [PubMed]
3. Hay BA, Guo M. Caspase-dependent cell death in Drosophila. Annu Rev Cell Dev Biol. 2006;22:623. [PubMed]
4. Li L, et al. A small molecule Smac mimic potentiates TRAIL- and TNFalpha-mediated cell death. Science. 2004;305(5689):1471. [PubMed]
5. Shi Y. Caspase activation, inhibition, and reactivation: a mechanistic view. Protein Sci. 2004;13(8):1979. [PubMed]
6. Chew SK, et al. The apical caspase dronc governs programmed and unprogrammed cell death in Drosophila. Dev Cell. 2004;7(6):897. [PubMed]
7. Akdemir F, et al. Autophagy occurs upstream or parallel to the apoptosome during histolytic cell death. Development. 2006;133(8):1457. [PubMed]
8. Zhang JH, Chung TD, Oldenburg KR. A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays. J Biomol Screen. 1999;4(2):67. [PubMed]
9. Boutros M, et al. Genome-wide RNAi analysis of growth and viability in Drosophila cells. Science. 2004;303(5659):832. [PubMed]
10. Kulkarni MM, et al. Evidence of off-target effects associated with long dsRNAs in Drosophila melanogaster cell-based assays. Nat Methods. 2006;3(10):833. [PubMed]Ma Y, Creanga A, Lum L, Beachy PA. Prevalence of off-target effects in Drosophila RNA interference screens. Nature. 2006;443(7109):359. [PubMed]
11. Zimmermann KC, Ricci JE, Droin NM, Green DR. The role of ARK in stress-induced apoptosis in Drosophila cells. J Cell Biol. 2002;156(6):1077. [PMC free article] [PubMed]
12. Xu D, et al. The CARD-carrying caspase Dronc is essential for most, but not all, developmental cell death in Drosophila. Development. 2005;132(9):2125. [PMC free article] [PubMed]
13. Muro I, et al. The Drosophila caspase Ice is important for many apoptotic cell deaths and for spermatid individualization, a nonapoptotic process. Development (Cambridge, England) 2006;133(17):3305. [PubMed]
14. Link N, et al. A collective form of cell death requires homeodomain interacting protein kinase. The Journal of cell biology. 2007;178(4):567. [PMC free article] [PubMed]
15. Rodriguez A, et al. Dark is a Drosophila homologue of Apaf-1/CED-4 and functions in an evolutionarily conserved death pathway. Nature Cell Biology. 1999;1(5):272. [PubMed]
16. Muro I, Hay BA, Clem RJ. The Drosophila DIAP1 protein is required to prevent accumulation of a continuously generated, processed form of the apical caspase DRONC. J Biol Chem. 2002;277(51):49644. [PubMed]
17. Weinmann L, et al. A novel p53 rescue compound induces p53-dependent growth arrest and sensitises glioma cells to Apo2L/TRAIL-induced apoptosis. Cell Death Differ. 2008 [PubMed]
18. Matapurkar A, Lazebnik Y. Requirement of cytochrome c for apoptosis in human cells. Cell Death Differ. 2006;13(12):2062. [PubMed]
19. Kroemer G, Martin SJ. Caspase-independent cell death. Nat Med. 2005;11(7):725. [PubMed]
20. Colell A, et al. GAPDH and autophagy preserve survival after apoptotic cytochrome c release in the absence of caspase activation. Cell. 2007;129(5):983. [PubMed]
21. Bard F, et al. Functional genomics reveals genes involved in protein secretion and Golgi organization. Nature. 2006;439(7076):604. [PubMed]
22. Jones S, et al. Core signaling pathways in human pancreatic cancers revealed by global genomic analyses. Science (New York, N.Y. 2008;321(5897):1801. [PMC free article] [PubMed]
23. Perez A, et al. A new class of receptor for herpes simplex virus has heptad repeat motifs that are common to membrane fusion proteins. J Virol. 2005;79(12):7419. [PubMed]Perez-Romero P, et al. Herpes simplex virus entry mediator associates in infected cells in a complex with viral proteins gD and at least gH. J Virol. 2005;79(7):4540. [PubMed]
24. Murray TV, et al. A non-apoptotic role for caspase-9 in muscle differentiation. J Cell Sci. 2008;121(Pt 22):3786. [PubMed]
25. Malo N, et al. Statistical practice in high-throughput screening data analysis. Nat Biotechnol. 2006;24(2):167. [PubMed]
26. Clemens JC, et al. Use of double-stranded RNA interference in Drosophila cell lines to dissect signal transduction pathways. Proc Natl Acad Sci U S A. 2000;97(12):6499. [PubMed]
27. Wolff T, Ready DF. Cell death in normal and rough eye mutants of Drosophila. Dev. Biol. 1991;113:825. [PubMed]
28. Mendes CS, et al. Cytochrome c-d regulates developmental apoptosis in the Drosophila retina. EMBO Rep. 2006;7(9):933. [PubMed]
29. Ditzel M, et al. Degradation of DIAP1 by the N-end rule pathway is essential for regulating apoptosis. Nat Cell Biol. 2003;5(5):467. [PubMed]
30. Kimura K, Kodama A, Hayasaka Y, Ohta T. Activation of the cAMP/PKA signaling pathway is required for post-ecdysial cell death in wing epidermal cells of Drosophila melanogaster. Development. 2004;131(7):1597. [PubMed]