Diverse Domains of THREAD/DIAP1 Are Required to Inhibit Apoptosis Induced by REAPER and HID in Drosophila

Corresponding author: Kristin White, CBRC, Massachusetts General Hospital, Bldg. 149, 13th St., Charlestown, MA 02129., kristin.white{at}cbrc2.mgh.harvard.edu (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Significant amounts of apoptosis take place during Drosophila development. The proapoptotic genes reaper (rpr), grim, and head involution defective (hid) are required for virtually all embryonic apoptosis. The proteins encoded by these genes share a short region of homology at their amino termini. The Drosophila IAP homolog THREAD/DIAP1 (TH/DIAP1), encoded by the thread (th) gene, negatively regulates apoptosis during development. It has been proposed that RPR, GRIM, and HID induce apoptosis by binding and inactivating TH/DIAP1. The region of homology between the three proapoptotic proteins has been proposed to bind to the conserved BIR2 domain of TH/DIAP1. Here, we present an analysis of loss-of-function and gain-of-function alleles of th, which indicates that additional domains of TH/DIAP1 are necessary for its ability to inhibit death induced by RPR, GRIM, and HID. In addition, that analysis of loss-of-function mutations demonstrates that th is necessary to block apoptosis very early in embryonic development. This may reflect a requirement to block maternally provided RPR and HID, or it may indicate another function of the TH/DIAP1 protein.

THE active elimination of cells by apoptosis is a fundamental aspect of normal development and homeostasis in multicellular organisms. In the embryo of Drosophila, apoptosis is controlled by three genes that are clustered in a small region of the third chromosome (WHITE et al. 1994 ). These genes, head involution defective (hid), grim, and reaper (rpr), act to regulate all embryonic apoptosis (WHITE et al. 1994 ; GRETHER et al. 1995 ; CHEN et al. 1996 ). In the absence of all three genes, embryonic apoptosis is virtually eliminated. Overexpression of any one of these genes is sufficient to induce apoptosis in a wide variety of cell types (GRETHER et al. 1995 ; CHEN et al. 1996 ; WHITE et al. 1996 ).

Apoptosis is negatively regulated in Drosophila by the gene thread (th; HAY et al. 1995 ). This gene encodes a protein, TH/DIAP1, that is homologous to the inhibitor of apoptosis or IAP proteins, first identified in baculovirus (CROOK et al. 1993 ). In the virus, these proteins act to block the death of the infected host cell (CLEM and MILLER 1994 ). The viral proteins have been shown to block apoptosis in a wide variety of systems (HAY et al. 1995 ; ROTHE et al. 1995 ; UREN et al. 1996 ; VUCIC et al. 1997B ). IAP homologs have been found in both Drosophila and mammals (HAY et al. 1995 ; ROTHE et al. 1995 ; ROY et al. 1995 ; DUCKETT et al. 1996 ; UREN et al. 1996 ; AMBROSINI et al. 1997 ). The mammalian homologs have been proposed to provide survival signals in the tumor necrosis factor signaling pathway and in oncogenic transformation (ROTHE et al. 1995 ; UREN et al. 1996 ; AMBROSINI et al. 1997 ).

The highly conserved baculovirus IAP repeats (BIRs) play a major role in the antiapoptotic function of the IAPs. IAPs may have one, two, or three BIRs. There is also commonly a carboxyl-terminal "ring" domain consisting of four pairs of cysteines and histidines (DEVERAUX and REED 1999 ). Both of these domains are predicted to bind zinc. In the viral proteins, both the BIR and ring domains are required to inhibit apoptosis under some, but not all, conditions (CLEM and MILLER 1994 ; HARVEY et al. 1997 ; VUCIC et al. 1998B ), while in the Drosophila and mammalian proteins, the BIR domains alone are sufficient to inhibit apoptosis (HAY et al. 1995 ; DEVERAUX et al. 1997 ; ROY et al. 1997 ).

In transgenic Drosophila, ectopic expression of TH/DIAP1 has been shown to block apoptosis induced by overexpression of RPR and HID (HAY et al. 1995 ). RPR and TH/DIAP1 interact physically when overexpressed in lepidopteran SF-21 cells (VUCIC et al. 1997A ). Similarly, both HID and GRIM each can physically interact with TH/DIAP1 in this overexpression assay (VUCIC et al. 1998A ). In these studies, the binding of RPR, GRIM, and HID to TH/DIAP1 appears to require the BIR domains of TH/DIAP1 and a short stretch of conserved residues at the amino-terminus of the death inducers (VUCIC et al. 1998A ).

IAPs have also been shown to bind to and inhibit the activity of caspases, a class of proteases that effect cell death (reviewed in DEVERAUX and REED 1999 ). In particular, TH/DIAP1 binds to a Drosophila caspase, drICE, and inhibits apoptosis induced by activated drICE and by mammalian caspase 3 in SF-21 cells (KAISER et al. 1998 ). TH/DIAP1 also binds to the caspase DCP-1 and inhibits its activity (HAWKINS et al. 1999 ; WANG et al. 1999 ). Thus, one model for the proapoptotic action of RPR, GRIM, and HID is that they bind to TH/DIAP1, inhibiting the ability of TH/DIAP1 to block caspase activity (WANG et al. 1999 ). The activated caspases would then cause apoptosis. In this model, TH/DIAP1 acts downstream of RPR, GRIM, and HID.

In this article, we report the identification of a potent gain-of-function mutation in th. This mutation was unique in that it protected against RPR- and GRIM-induced cell death, but not against HID-induced cell death. The specificity of this protection prompted us to investigate the ability of other th alleles to inhibit apoptosis. We found that several other alleles showed specificity in their interactions with RPR, GRIM, and HID. We have identified the molecular lesions in these mutations. Our data support the model that TH/DIAP1 inhibits HID-induced death differently than it inhibits RPR- and GRIM-induced death.

In addition, we examined the lethal phenotype of th alleles. As described below, the analysis of embryos homozygous for loss-of-function alleles of th indicates that this gene is required to block apoptosis very early in embryonic development.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly stocks:
Fly matings were conducted using standard procedures, and the cultures were maintained at 25°. The following alleles of thread were used: Df(3L)brm11, th⁴, th⁵, th^6B, th⁷, th⁸ (BRIZUELA et al. 1994 ), th^81.03, and th^109.07 (MOORE et al. 1998 ). th^SL and th⁹ were generated by chemical mutagenesis (see below).

Mutagenesis for modifiers of GMRrpr:
To identify suppressors or enhancers of the GMRrpr eye phenotype, GMRrpr 81 (WHITE et al. 1996 ) homozygous male flies were starved for 2 hr and exposed overnight to 25 mM EMS in 1% sucrose. Males were then mated to GMRrpr81/GMRrpr81; GMRrpr97/GMRrpr97 females.The progeny of this cross were screened for changes in eye size and roughness. All flies with eyes that were larger or smaller than the eyes of siblings (carrying three copies of GMRrpr) were retested by crossing to CyO2xGMRrpr; TM2/MKRS (KURADA and WHITE 1998 ). Those that showed consistent suppression were mapped by segregation and by recombination with the rucuca mapping chromosome (ASHBURNER 1989B ).

Reversion of th^SL:
Homozygous th^SL males were mutagenized with EMS and crossed to GMRrpr81/GMRrpr81; GMRrpr97/GMRrpr97 females. The progeny were screened for loss of the suppressing capability of th^SL by identifying flies that showed eyes at least as small as those of GMRrpr81/+; GMRrpr97/+ flies.

Sequencing of the th alleles:
DNA was isolated from adult flies homozygous for viable alleles (th^SL) or from dead embryos homozygous for lethal alleles (th⁴, th⁵, th^6B, th⁷, th⁹, th^81.03, and th^109.07). Homozygous dead embryos were obtained from matings of th/+ flies. Collections of embryos 0–16 hr after egg laying (AEL) were aged for at least 24 hr, and the dead embryos were picked under a microscope and processed for DNA isolation. DNA extraction from 200 adults or 400 embryos was performed using QIAmp blood kit (Qiagen, Chatsworth, CA) according to the manufacturer's instructions. The th open reading frame was then amplified by polymerase chain reaction (PCR) under standard conditions using the Expand Long Template PCR system (Boehringer Mannheim, Indianapolis). The following oligonucleotide primers were used: 5'-CGACTTCAGAGGAAAGGAGCCAGA-3' and 5'-CGCTTATAATTAACACATTCCGCCCAC-3'. DNA sequences were obtained using the Thermo Sequenase radiolabeled terminator cycle sequencing system (Amersham, Arlington Heights, IL). The sequencing of the entire coding region of th was performed on both strands for all alleles.

Embryonic protection experiments:
hs-rpr (WHITE et al. 1996 ) flies were crossed to th^SL/TM2 flies, and embryos were collected on molasses plates, aged for 3–8 hr, and heat shocked for 1 hr at 37°. The embryos were transferred to standard fly food and scored for adult viability. TM3hs-hid (FLYBASE 1994 ) flies were crossed to th^SL/D gl³ or th^6B/D gl³. At 8–10 hr post-collection, embryos were heat shocked for 20 min at 37°. The embryos were transferred to standard fly food and scored for adult viability.

Anti-TH/DIAP1 antibody production and immunostaining:
The TH/DIAP1 polyclonal antibody was raised in rabbits (Poconos Rabbit Farm) against full-length TH/DIAP1 expressed and purified using the Xpress System protein purification kit (Invitrogen, San Diego, CA). The antibody was subsequently affinity purified with TH/DIAP1 His-tagged fusion protein covalently bound to cyanogen-bromide-activated Sepharose (Pharmacia, Piscataway, NJ). This antibody recognized a single band of the correct size on Western blots of eye discs. For immunostaining, eye discs were fixed in 2% paraformaldehyde, washed in BSS (ASHBURNER 1989B ), preincubated in BSS with 5% goat serum and 0.3% Triton X-100 (BSN), and incubated overnight at 4° in preabsorbed anti-TH/DIAP1 antibody diluted 1:1000 in BSN. Preabsorbed Texas red-conjugated anti-rabbit secondary antibody (Cappel, Aurora, OH) was used at a concentration of 1:1000. Tissues were mounted in Fluoromount-G (Southern Biotechnology, Birmingham, AL) and viewed on a TCS NT4D confocal microscope (Leica, Deerfield, IL). The specificity of the antibody was confirmed by preincubating the antibody with TH/DIAP1 fusion protein, which eliminated all staining.

pUAS constructs and transformation:
Full-length pUASthV85M and pUASthwt were generated in a two-part cloning: 3' fragments of the genes were obtained by SalI digestion of pHSP70PLVI⁺EpiD-iap1V85M and pHSP70PLVI⁺EpiD-iap1, which were kindly provided by Dr. Lois Miller (University of Georgia, Athens, GA). The SalI fragments were purified and redigested with MfeI, and the SalI-MfeI fragments were purified. These fragments contain the 3' end of the cDNA, starting at base 617 (HAY et al. 1995 ). The G-to-A mutation at base 684 found in the SL mutation was introduced by site-directed mutagenesis by the Miller lab, using the Clontech (Palo Alto, CA) Transformer system. The 5'fragment was generated by PCR amplification from genomic DNA extracted from yw flies, as described above. The following oligonucleotide primers were used: 5'-CGACTTCAGAGGAAAGGAGCCAGA-3' and 5'TAATGCTTCTTCGGCATTGATCGGCACA-3'. The PCR products were gel purified and cut with EcoRI and MfeI, followed by further gel purification. This fragment contains bases 357–616 of the cDNA. The vector GAL UAS (pUAST) was cut with EcoRI and XhoI and gel purified. The three DNA pieces were ligated together using T4 DNA ligase (Boehringer Mannheim), and constructs were confirmed by sequencing. Flies bearing transgenes were generated by P-element-mediated germline transformation (ASHBURNER 1989A ).

TUNEL staining:
Embryos were dechorionated and fixed for 30 min in 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. The vitelline membrane was removed with methanol. The embryos were rehydrated through 75, 50, and 25% methanol/PBT (PBS with 0.1% Tween-20) and treated with proteinase K (10 µg/ml in PBS) for 5 min, washed twice in PBT, postfixed for 20 min in 4% paraformaldehyde in PBS, washed five times for 5 min in PBT, and incubated for 1 hr at 37° in the equilibration buffer provided in the Apoptag kit from Intergen. The embryos were incubated overnight at 37° with TdT (reaction buffer with TdT 2:1 from the Apoptag kit, with 0.3% Triton X-100). The reaction was stopped by incubating the embryos for 4 hr at 37° with stop solution (from the Apoptag kit). The embryos were washed three times for 5 min in PBT and blocked for 1 hr with 2 mg/ml BSA and 5% goat serum in PBT. They were incubated overnight at 4° with 1:2000 preabsorbed antidigoxygenin antibody (alkaline phosphate conjugated, Boehringer Mannheim). The embryos were washed four times for 20 min in PBT, rinsed twice for 20 min in NTMT buffer (0.1 M Tris-HCl, pH 9.5, 50 mM MgCl₂, 0.1 M NaCl, 0.1% Tween-20), and incubated for 10 min with 3.5 µl of NBT and 4.5 µl of X-phosphate provided in the DIG Nucleic Acid detection kit (Boehringer Mannheim) in 1 ml of NTMT. The reaction was stopped by washing in PBT. The embryos were mounted in Vectashield (Vector Laboratories, Burlingame, CA).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of a gain-of-function allele of th as a dominant suppressor of GMRrpr:
To understand the mechanisms underlying apoptosis in Drosophila, we took advantage of the cell death phenotype caused by RPR overexpression. Directed overexpression of RPR in the developing eye using the GMR vector (GMRrpr) results in a dose-dependent elimination of the eye due to ectopic apoptosis (WHITE et al. 1996 ). We and others have shown that this phenotype is sensitive to dominant modification by mutations in genes that act to inhibit apoptosis (HAY et al. 1995 ; BERGMANN et al. 1998A ; KURADA and WHITE 1998 ).

A screen was carried out to look for modifiers of the GMRrpr phenotype. We mutagenized flies carrying two copies of the GMRrpr transgene and crossed them to flies carrying four copies of the transgene. Among the progeny, which showed a highly reproducible three-copy GMRrpr phenotype, we looked for flies that showed either a larger eye, indicating suppression or the GMRrpr phenotype, or a smaller eye, indicating enhancement. From the 64,000 haploid genomes scored, we identified 19 mutations that strongly suppressed GMRrpr. To investigate whether these mutations suppressed the killing activity of RPR or the expression of the GMRrpr transgene, we tested these mutations for the ability to suppress the rough-eye phenotype of GMRp21 (DE NOOIJ and HARIHARAN 1995 ). All but one of the strong suppressor mutations also suppressed the GMRp21 phenotype, indicating that these mutations probably inhibited expression from the GMR promoter. The remaining mutation, preliminarily called Su(GMRrpr)SL, was quite striking, as it restored the GMRrpr eye to a wild-type appearance (Figure 1A and Figure B).

View larger version (89K): In this window In a new window Download PPT slide   Figure 1. A gain-of-function allele of th suppresses GMRrpr- and GMRgrim-induced cell death. The rough-eye phenotype of flies expressing two copies of a GMRrpr transgene (A) or one copy of a GMRgrim transgene (D) can be rescued to wild type by Su(GMRrpr)SL (B and E). The small- and rough-eye phenotype of flies carrying one copy of the GMRhid transgene (G) is slightly enhanced by the presence of one copy of Su(GMRrpr)SL (H). A second mutation in Su(GMRrpr)SL, th9, completely reverts its ability to suppress GMRrpr- and GMRgrim-induced cell death. This allele dominantly enhances the rough-eye phenotypes of GMRrpr, GMRgrim, and GMRhid (C, F, and I).

Preliminary recombination mapping placed Su(GMR rpr)SL very closely to the th gene (HAY et al. 1995 ). As loss-of-function (LOF) mutations in th have been previously shown to dominantly enhance RPR killing, we speculated that Su(GMRrpr)SL might be a gain-of-function (GOF) th mutation. To explore this possibility, we tested the proximity of Su(GMRrpr)SL to th by further recombination mapping. No recombinants were obtained between Su(GMRrpr)SL and th¹ in >1000 chromosomes scored.

If Su(GMRrpr)SL was a GOF th mutation, then reversion of the dominant phenotype should generate LOF th alleles. We mutagenized the Su(GMRrpr)SL stock and tested for reversion of the Su(GMRrpr)SL phenotype. Out of 42,000 mutagenized chromosomes scored, we obtained 3 revertants of the Su(GMRrpr)SL phenotype. Two of these mutations reverted Su(GMRrpr)SL to the baseline GMRrpr phenotype, while one line (now called th⁹) enhanced the GMRrpr phenotype (Figure 1C). The mutations that did not enhance the GMRrpr phenotype complemented th LOF alleles, and they may represent second site suppressors of the mutation, or weak th LOF alleles. The th⁹ allele did not complement other th alleles when tested for lethality and, thus, was likely to be a th LOF allele. Sequencing has confirmed that th⁹ is a mutation in the th gene (see below). To eliminate the concern that th⁹ was not a revertant of Su(GMRrpr)SL, but a th allele acting as a closely linked second site suppressor, we tested whether another th LOF allele was able to suppress the phenotype of Su(GMRrpr)SL. We found that the phenotype of CyO2xGMRrpr; Su(GMRrpr)SL/th⁵ was the same as that of CyO2xGMRrpr; Su(GMRrpr) SL/+ (data not shown). Thus, it is highly likely that Su(GMRrpr)SL, now called th^SL, is a GOF mutation in th, which is reverted by a second mutation in the same gene, the th⁹ mutation.

Apoptosis in Drosophila embryos is regulated by three genes, rpr, grim, and hid. All of these genes are able to induce apoptosis when overexpressed, and LOF th alleles enhance killing by all three genes (HAY et al. 1995 ; BERGMANN et al. 1998B ). This supports a central role for TH/DIAP1 in all three apoptotic pathways. It therefore seemed likely that th^SL would suppress apoptosis induced by GRIM and HID in the eye. As expected, apoptosis induced by GRIM was strongly suppressed by th^SL (Figure 1E). To our surprise, th^SL did not suppress HID-induced apoptosis, but rather, slightly enhanced this apoptosis (Figure 1H). This suggested that different alleles of th might affect killing by RPR, GRIM, and HID differently, and that different parts of the TH/DIAP1 protein might be important for interaction with these different apoptosis inducers.

This theory provoked our interest in looking at the effects of other mutations in the th gene to determine if these showed a preferential interaction with either RPR-, GRIM-, or HID-induced cell death. We have found that certain alleles do show interactions with only one of these inducers of apoptosis (Figure 2). On the basis of these results, we can classify these mutations as LOF or GOF. The GOF alleles fall into two categories: those that suppress RPR- and GRIM-, but not HID-induced death, and those that suppress HID-, but not RPR- and GRIM-induced death.

View larger version (49K): In this window In a new window Download PPT slide   Figure 2. Different th alleles enhance or suppress RPR-, GRIM-, or HID-induced cell death in the eye. Overexpression of RPR (top row), GRIM (middle row), and HID (bottom row) in the eye produces a rough and small eye (leftmost column). Heterozygosity for a deficiency that removes the th gene [Df(3L)brm11] enhances all three of these eye phenotypes (second column). Similar enhancement is achieved in the heterozygous loss-of-function alleles th109.07 (third column), th4 (fourth column), th5 (fifth column), and th7 (sixth column). Enhancement of the GMRrpr and GMRgrim phenotype is also observed in th6B (seventh column) and th81.03 (last column) heterozygotes. Surprisingly, these two alleles suppress the GMRhid eye phenotype and, therefore, are defined as gain-of-function alleles (GOF1).

We have gone on to identify the molecular lesions for many of these alleles of th (Figure 3A). We looked at both the dominant phenotypes of these alleles in the eye and the recessive phenotypes in the embryo. Below is a summary of our molecular and phenotypic characterization of each of these classes of alleles. These data are summarized in graphic format in Figure 3B.

View larger version (56K): In this window In a new window Download PPT slide   Figure 3. Molecular characterization of th alleles. (A) The homology shared by the BIR and ring domains of TH/DIAP1 with other members of the IAP protein family. Both BIR domains of TH/DIAP1 are shown, as well as another Drosophila IAP (DIAP2; HAY et al. 1995 ), the baculoviral protein OpIAP (BIRNBAUM et al. 1994 ), and the mammalian family members C-IAP1 (ROTHE et al. 1995 ), NAIP (ROY et al. 1995 ), and X-IAP (LISTON et al. 1996 ). Single-amino-acid changes in the first BIR for th109.07, th9, and thSL alleles are marked in red, while in green are the point mutations in the second BIR for th4 and th5 and the end of the in-frame deletion of th7. The th7 deletion starts in the spacer between the first and second BIR domain, at proline 130. The single cysteine-to-tyrosine mutations present in both th6B and th81.03, located in highly conserved residues of the ring, are marked in blue. (B) A graphic summary of all th alleles characterized with regard to their effect on the GMRrpr, GMRgrim, and GMRhid eye phenotypes. The relative location of the mutations in the DIAP1 protein are also shown, using the following color code: purple for LOF alleles, green for GOF1 alleles, and red for the GOF2 allele.

Gain-of-function alleles:
These are the most interesting mutations that we tested. We have classified them as GOF mutations on the basis of their ability to dominantly suppress apoptosis, in contrast to the enhancement seen with LOF alleles, described below. We have divided these mutations into two classes. GOF1 is represented by two alleles, th^6B and th^81.03. In contrast to what we saw with th^SL, both of these mutations suppress killing by HID, and they have no effect or slightly enhance killing by RPR and GRIM (Figure 2). The second class of GOF mutation (GOF2) consists of the th^SL allele. As described above, this mutation suppresses killing by RPR and GRIM, but slightly enhances killing by HID (Figure 1).

th^6B and th^81.03: Both of these alleles result from mutations that change canonical cysteines in the ring domain to tyrosines. These mutations are likely to disrupt the secondary structure of the ring domain, as these residues are thought to coordinate zinc. Previous data have suggested that the BIR domains of TH/DIAP1 alone are more potent inhibitors of apoptosis induced by both RPR and HID than by the full-length molecule (HAY et al. 1995 ; VUCIC et al. 1998A ). However, our data suggest that these mutations disrupt the ring structure in a way that only activates the protein in the context of HID killing. The differences in these conclusions may reflect differences in the level of protein expression, as the previous results were based on overexpression, while our data reflects changes in the endogenous gene.

th^SL: We considered whether the ability of th^SL to suppress RPR- and GRIM-induced death might result from generally increased levels of TH/DIAP1. To examine this question further, we looked at TH/DIAP1 protein levels by immunostaining and by Western blot. Confocal analysis of wild-type and th^SL eye discs stained with a polyclonal anti-TH/DIAP1 antibody showed that TH/DIAP1 levels and distribution are not detectably affected by the th^SL mutation (Figure 4). In addition, previous experiments indicated that overexpression of wild-type TH/DIAP1 results in the suppression of apoptosis induced by HID, as well as by RPR and GRIM (HAY et al. 1995 ). Therefore, we believe that increased expression cannot explain the specific protection th^SL offers against RPR- and GRIM-induced death.

View larger version (173K): In this window In a new window Download PPT slide   Figure 4. The thSL allele does not increase TH/DIAP1 protein levels. Eye discs from (A and B) wild-type and (C and D) homozygous thSL larvae were stained with a polyclonal antibody to the TH/DIAP1 protein and then visualized by confocal microscopy. All panels show projections of staining throughout the disc. The protein is expressed ubiquitously. There is no evidence that the thSL mutation modifies either the levels (A and C) or the subcellular localization (B and D) of the protein.

Sequencing of the open reading frame of th^SL has revealed a valine-to-methionine change in the first BIR domain. This residue is not highly conserved, although Op-IAP also has a valine in this position (Figure 3A). However, the recently published solution structure of BIR3 from c-IAP-1 places this residue very close to the zinc-binding site of the BIR (HINDS et al. 1999 ).

To confirm that this was in fact the mutation that was responsible for the th^SL phenotype, we generated transgenic flies that expressed wild-type TH/DIAP1 or the V85M th^SL allele, using the gal4/upstream activating sequence (UAS) binary expression system. As expected, the overexpression of the wild-type protein in the eye was able to suppress GMRrpr-induced apoptosis. However, we found that in two independent insertion lines, the V85M mutant form of the protein resulted in stronger suppression of GMRrpr than the wild-type form (Figure 5). This indicated that the V85M mutation was responsible for the th^SL phenotype, and suggests that this mutant form of the protein specifically inhibits the proapoptotic effects of RPR.

View larger version (117K): In this window In a new window Download PPT slide   Figure 5. Expression of thSL from a transgene can suppress GMRrpr-, GMRgrim-, and GMRhid-induced cell death in the eye. Transgenic flies for either the wild-type or the V85M mutant version of TH/DIAP1 were generated using the UAS expression system. These flies were tested for their ability to suppress GMRrpr-, GMRgrim-, and GMRhid-induced cell death in the eye. Eye-specific expression was activated using a GMRgal4 driver. Two independent lines for the wild type and two for the GOF (V85M) constructs were examined. Control phenotypes for GMRrpr, GMRgrim, and GMRhid (A). The V85M transgene completely protects against RPR- and GRIM-induced death (B), while the wild-type transgene cannot completely rescue the GMRrpr eye phenotype (C). In contrast, the wild-type transgene can better protect against GMRhid-induced cell death, compared to the V85M transgene.

We also found that the V85M transgene was also able to suppress GMRhid-induced death, although not as strongly as the wild-type transgene (Figure 5). This corresponds to our finding that th^SL behaves as a weak LOF mutation in hid killing. In the context of overexpression, this slightly weakened protein can still protect against hid activity, but not as well as the wild-type protein.

The effect of th^SL on RPR-induced death was not confined to the eye; th^SL was able to protect against organismal death resulting from RPR overexpression in the embryo. Like HID-induced death in the eye, HIDinduced embryonic death was not inhibited by the th^SL mutation, but rather, was slightly enhanced. In contrast, th^6B of the GOF1 class protected against HID-induced death in the embryo (Figure 6).

View larger version (42K): In this window In a new window Download PPT slide   Figure 6. The GOF alleles also protect against apoptosis in the embryo. Flies carrying a transgene that expressed RPR under the heat shock promoter (hsrpr) were crossed to thSL/TM2 flies. Embryos were collected, heat shocked at 5–10 hr AEL and allowed to develop to adults (see MATERIALS AND METHODS). The results are expressed as the ratio of hsrpr with the thGOF allele (hsrpr/+; thGOF/+) to those without it (hsrpr/+; TM2/+). TM3-hshid flies were crossed to thSL/D gl3 or th6B/D gl3 flies. Embryos were collected and heat shocked as described in MATERIALS AND METHODS and were allowed to develop to adults. The results are expressed as the ratio of hshid flies with the thGOF allele (TM3-hshid/thGOF) to those without it (TM3-hshid/D gl3). The results shown are the mean and standard error of two or more experiments. thSL increases the viability of hsrpr flies more than threefold, but slightly enhances hshid-induced death, while th6B protects against hshid-induced death, also by about threefold.

Loss-of-function alleles:
We have found that five of the alleles we tested, th⁴, th⁵, th⁹, th^109.07, and th⁷, can be classified genetically as LOF. All of these alleles dominantly enhance the ability of RPR, GRIM, and HID to kill in the eye, approximately to the same extent as a deletion that takes out the gene (Figure 2). Sequencing has revealed that all of these alleles are due to changes in conserved residues in the BIR domains (Figure 3A). The molecular lesions found in these mutations are summarized below.

th⁴: One of the canonical histidine residues in the second BIR is mutated to a tyrosine in th⁴. A histidine-to-alanine mutation at this site has been shown to inhibit the ability of Op-IAP to inhibit HID-induced apoptosis in SF-21 cells (VUCIC et al. 1998B ).

th⁵: There is a stop mutation near the middle of the second BIR of th⁵. This mutation acts as a strong LOF mutation. GMRgrim-induced apoptosis is enhanced by th⁵ mutations to a slightly greater extent than by the deletion, suggesting that a single complete BIR may have slight dominant-negative properties (Figure 2). On the basis of work done with Op-IAP in Lois Miller's lab, this mutation would be predicted to prevent binding of DIAP1 to HID (VUCIC et al. 1998B ).

th⁹: This allele was isolated as a revertant of the th^SL allele and, thus, contains two mutations: the original V85M mutation in BIR1 seen in th^SL (see above), as well as a second mutation in BIR1 at Pro 55. This residue is conserved in almost all BIRs, except BIR1 of NAIP and BIR3 of X-IAP (Figure 3A).

th^109.07: This is likely to be a complete null, as there is a stop at the beginning of BIR1 at amino acid 54. It shows a slightly weaker ability to enhance RPR killing than a deletion of the gene (Figure 2). This may reflect additional enhancing activities on the deletion chromosome.

th⁷: This allele appears to show dominant-negative properties, as it enhances killing by all three inducers to a greater extent than the deletion of the region (Figure 2). This allele is an in-frame deletion that removes the first half of BIR2, but leaves BIR1 and the ring intact. This suggests that the presence of BIR1 alone may have dominant-negative properties, or that the close proximity of the ring to BIR1 interferes with IAP function.

Embryonic phenotypes:
All the LOF alleles described above are lethal when crossed to Df(3L)brm11, a deletion of the region, and when crossed to each other (data not shown). TUNEL staining was used to assess the role of apoptosis in the lethality. Embryos were collected from th⁵, th⁷, th⁹, and th^109.07 stocks. In collections of embryos from 0 to 5 hr AEL, no ectopic apoptosis could be seen. At this stage, normal, developmentally regulated apoptosis had not yet begun.

In collections of embryos where the youngest embryos were 5 hr old, we found a striking phenotype. In a subset of embryos, the vast majority of nuclei stain with TUNEL (Figure 7B). A total of 20% of the embryos showed this phenotype in a collection from a th⁷/TM2 stock. These embryos appear to have been arrested prior to gastrulation, as the majority of labeled nuclei appear at the surface of the embryo. Gastrulation occurs at 3 hr AEL. We did not see any TUNEL labeling in mutant embryos until 5 hr AEL. As the onset of apoptosis is several hours after these embryos have arrested, it is possible that the ectopic apoptosis is secondary to the developmental defects. However, it is also possible that zygotic th is required to suppress maternally supplied rpr, grim, and hid activity (see DISCUSSION). Early activation of apoptosis in th mutant embryos has also been reported recently by Hay (WANG et al. 1999 ).

View larger version (91K): In this window In a new window Download PPT slide   Figure 7. Early embryonic cell death phenotype of th mutants, as assayed by TUNEL staining. Normally occurring embryonic cell death is not present in wild-type blastoderm stage embryos. None of the nuclei of wild-type embryos 2–4 hr old AEL stains with TUNEL (A). In contrast, in a collection of th7/TM2 embryos at 5–11.5 hr AEL, 20% of the embryos show a blastoderm morphology and massive TUNEL labeling. All the labeled nuclei appear to be at the surface of the embryo (B). Embryos homozygous for a GOF1 allele (th81.03) are also heavily labeled with TUNEL at 5–12 hr AEL (D). These embryos, however, appear to progress further in development than the LOF embryos. Wild-type embryos at the same stage have significantly lower amounts of apoptosis (C).

Both GOF1 allele classes are also lethal in combination with LOF th alleles. This indicates that an intact ring structure is essential for full activity of the TH/DIAP1 protein. However, we have found that embryos homozygous for these alleles progress further in development than the LOF mutants before widespread apoptosis begins (Figure 7D). This may reflect the distinct activity of these alleles with regard to RPR, GRIM, and HID killing. It is also possible that these alleles maintain residual activity in a developmentally required pathway independent of rpr, grim, and hid function.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Several mechanisms of action have been suggested for the antiapoptotic properties of the IAP family of proteins. Among these are the binding of the Drosophila IAPs to the proapoptotic proteins RPR, GRIM, and HID (VUCIC et al. 1997A , VUCIC et al. 1998A ). This interaction has been demonstrated in overexpression systems, and has been proposed to involve the homologous amino-terminal 14 amino acid sequences of the apoptosis initiators with the second BIR domain of the IAPs. The data presented here suggest that this is an oversimplification.

Another mechanism that has been proposed for IAP antiapoptotic activity is the direct binding and inhibition of caspases (reviewed in DEVERAUX and REED 1999 ). TH/DIAP1 has been shown to bind to the Drosophila caspases drICE and DCP-1 and to inhibit their ability to induce apoptosis (KAISER et al. 1998 ; HAWKINS et al. 1999 ; WANG et al. 1999 ). Here again, this binding activity appears to rest within BIR2.

These physical interactions support a simple model of IAP action. In this model, IAPs act within viable cells to inhibit caspase function. The action of RPR, HID, and GRIM is to interfere with the ability of IAPs to inhibit caspases, thus inducing apoptosis.

On the basis of the model, the LOF mutations we have identified would be predicted to interfere with the ability of the TH/DIAP1 protein to inhibit caspase function. This is likely to be true for th^109.07, which lacks most of the protein, as well as for th⁵ and th⁴, which affect conserved residues in BIR2. BIR2 has been shown to be sufficient to inhibit apoptosis induced by the active form of the Drosophila caspase drICE. The th⁹ mutation in BIR1 suggests that this BIR is also important for the full function in caspase inhibition. Alternatively, this change in BIR1 might have long-range effects on BIR2 structure or on protein stability.

It is interesting to note that th⁷, which acts as a very strong LOF mutation and seems to show some dominant-negative properties, has only the BIR1 attached to the spacer and ring domains. Thus, despite the extensive homologies between the two BIR domains of the protein, a single BIR is not sufficient for TH/DIAP1 function, at least in the presence of an attached ring domain. Others have shown that BIR2 of TH/DIAP1 and Op-IAP, as well as the single BIR of survivin, are able to inhibit apoptosis (AMBROSINI et al. 1997 ; VUCIC et al. 1998B ).

Again, on the basis of the model above, the GOF mutations we have identified would be predicted to bind to caspases, but not to the inducers. The th^SL mutation maps to a weakly conserved residue in BIR1 and does not result in increased th protein levels. This suggests that BIR1 is important for RPR and GRIM binding, but not for HID binding, as HID activity is unaffected in this mutation. Even in the context of overexpression in the eyes of transgenic flies, this mutant IAP retains some specificity for RPR and GRIM killing. This implies that the simple model of BIR2 binding to the conserved NH₂-terminal sequences of RPR, GRIM, and HID is not accurate, and that other residues in the protein are differentially important for RPR and GRIM vs. HID binding.

The importance of regions outside of BIR2 for DIAP1 activity is supported by the analysis of the GOF1 class of mutations, th^6B and th^81.03. Both of these mutations suppress HID killing and would be predicted to inhibit HID binding. These mutations change conserved cysteines in the ring domain to tyrosines. This suggests that the ring is important for HID/DIAP1 interaction. However, VUCIC et al. 1998B have mapped the region of HID binding to DIAP1 and Op-IAP to BIR2, while the ring does not show any ability to bind to HID. In addition, these authors report that mutations in the ring, including those in conserved cysteines, had little effect on the ability of Op-IAP to protect against HID killing (VUCIC et al. 1998B ). These data, together with the finding that both GOF1 mutations are cysteine-to-tyrosine changes, suggest that these mutations might have a novel ability to interfere with binding of HID to BIR2. In addition, the observation that the GOF1 mutations slightly enhance RPR and GRIM killing suggests that these mutants are less potent inhibitors of caspases. This might result from weaker binding to caspases or from proteins that are slightly less stable. This second attribute would be predicted to enhance killing by any inducer that binds the IAP, but not to have an effect on HID, which is unable to bind.

In conclusion, our data support a model where RPR, GRIM, and HID interact with TH/DIAP1 to induce apoptosis. We have found that mutations that affect killing by RPR and GRIM or by HID can be isolated, indicating that these inducers interact with TH/DIAP1 in different ways. The GOF mutations that we have identified also provide us with useful tools to examine the roles of IAPs, rpr, grim, and hid during Drosophila development. The other Drosophila IAP homolog, DIAP2, has been shown to selectively inhibit RPR- and HID-induced but not GRIM-induced death (WING et al. 1998 ).

Finally, we have shown that the TH/DIAP1 protein is required very early in Drosophila development. This has also been reported recently by others (WANG et al. 1999 ). In LOF th alleles, we see a developmental arrest at the blastoderm stage and, subsequently, a synchronous apoptosis of all the nuclei. Earlier reports that homozygous th embryos showed no ectopic apoptosis probably reflects the very early stage at which this apoptosis occurs.

At this time, we cannot distinguish a direct requirement for th to block apoptosis or a requirement for th in another developmental process. This developmental defect could then result in secondary apoptosis. The latter possibility is reasonable, as many failures in development result in ectopic apoptosis (ABRAMS et al. 1993 ). A BIR containing protein from Caenorhabditis elegans is required for cytokinesis in embryos (FRASER et al. 1999 ). However, it is also possible that developmental arrest occurs as a result of the initiation of apoptosis, which is manifest only as DNA damage several hours later.

Does this early requirement for th reflect a need to inhibit apoptosis induced by rpr, grim, and hid? Double mutants of th and Df(3L)H99, the deletion that removes rpr, grim, and hid, show a phenotype similar to th alone (WANG et al. 1999 ). This indicates that TH/DIAP1 is not required to suppress zygotic RPR, GRIM, and HID activity. However, hid and rpr mRNA can be seen in a subset of cells in the blastoderm embryo, as judged by in situ analysis (NORDSTROM et al. 1996 ; NASSIF et al. 1998 ; and K. WHITE, unpublished observations). This may indicate that these gene products are supplied maternally. TH/DIAP1 may be required to suppress maternally supplied RPR, GRIM, or HID. We found that allelic differences in the stage at which apoptosis begins in the th mutants parallel the general ability of the alleles to inhibit apoptosis induced by RPR, HID, and GRIM. The strong LOF alleles arrest at the blastoderm stage, the GOF1 alleles arrest much later and the GOF2 allele is completely viable.

Many questions remain about how RPR, GRIM, and HID initiate apoptosis and how TH/DIAP1 protects cells. Our analysis has suggested that current models may be oversimplified. While BIR2 is certainly important for TH/DIAP1 antiapoptotic activity, BIR1 and the ring domain play a role in specific interactions with RPR and HID. In addition, our work indicates that overexpression of TH/DIAP1 does not completely reflect the activity of the gene in vivo. We expect that further investigation of these mutants should provide interesting insight into the functional mechanisms and regulation of th.

	ACKNOWLEDGMENTS

We thank Peter Bangs for advice on many aspects of the work, Laurel Raftery for comments on the manuscript, Yimin Ge for assistance with confocal microscopy, as well as Jim Kennison and Ruth Lehmann for providing us with th mutants. This work was supported by National Institutes of Health grant GM-55568 to K.W. Early stages of this work were supported by a grant from the Shiseido Company of Japan to Massachusetts General Hospital/Harvard Medical School.

Manuscript received June 25, 1999; Accepted for publication October 6, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABRAMS, J. M., K. WHITE, L. I. FESSLER, and H. STELLER, 1993  Programmed cell death during Drosophila embryogenesis. Development 117:29-43[Abstract].

AMBROSINI, G., C. ADIDA, and D. C. ALTIERI, 1997  A novel antiapoptosis gene, survivin, expressed in cancer and lymphoma. Nature Med. 3:917-921[Medline].

ASHBURNER, M., 1989a Drosophila: A Laboratory Handbook. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

ASHBURNER, M., 1989b Drosophila: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BERGMANN, A., J. AGAPITE, K. MCCALL, and H. STELLER, 1998a  The Drosophila gene hid is a direct molecular target of Ras-dependent survival signaling. Cell 95:331-341[Medline].

BERGMANN, A., J. AGAPITE, and H. STELLER, 1998b  Mechanisms and control of programmed cell death in invertebrates. Oncogene 17:3215-3223[Medline].

BIRNBAUM, M., R. CLEM, and L. MILLER, 1994  An apoptosis inhibiting gene from a nuclear polyhedrosis virus encoding a polypeptide with cys/his sequence motifs. J. Virol. 68:2521-2528[Abstract/Free Full Text].

BRIZUELA, B. J., L. ELFRING, J. BALLARD, J. W. TAMKUN, and J. A. KENNISON, 1994  Genetic analysis of the brahma gene of Drosophila melanogaster and polytene chromosome subdivisions 72AB. Genetics 137:803-813[Abstract].

CHEN, P., W. NORDSTROM, B. GISH, and J. M. ABRAMS, 1996  grim, a novel cell death gene in Drosophila.. Genes Dev. 10:1773-1782[Abstract/Free Full Text].

CLEM, R. J. and L. K. MILLER, 1994  Control of programmed cell death by the baculovirus genes p35 and iap.. Mol. Cell. Biol. 14:5212-5222[Abstract/Free Full Text].

CROOK, N. E., R. J. CLEM, and L. K. MILLER, 1993  An apoptosis-inhibiting baculovirus gene with a zinc finger-like motif. J. Virol. 67:2168-2174[Abstract/Free Full Text].

DE NOOIJ, J. C. and I. K. HARIHARAN, 1995  Uncoupling cell fate determination from patterned cell division in the Drosophila eye. Science 270:983-985[Abstract/Free Full Text].

DEVERAUX, Q. L. and J. L. REED, 1999  IAP family proteins-suppressors of apoptosis. Genes Dev. 13:239-252[Free Full Text].

DEVERAUX, Q. L., R. TAKAHASHI, G. S. SALVESEN, and J. C. REED, 1997  X-linked IAP is a direct inhibitor of cell death proteases. Nature 388:300-304[Medline].

DUCKETT, C. S., V. E. NAVA, R. W. GEDRICH, R. J. CLEM, and J. L. VAN DONGEN et al., 1996  A conserved family of cellular genes related to the baculovirus iap gene and encoding apoptosis inhibitors. EMBO J. 15:2685-2694[Medline].

FLYBASE,, 1994  The Drosophila genetic data base. Nucleic Acids Res. 22:3456-3458[Abstract/Free Full Text].

FRASER, A. G., C. JAMES, G. I. EVAN, and M. O. HENGARTNER, 1999  Caenorhabditis elegans inhibitor of apoptosis protein (IAP) homologue BIR-1 plays a conserved role in cytokinesis. Curr. Biol. 9:292-301[Medline].

GRETHER, M. E., J. M. ABRAMS, J. AGAPITE, K. WHITE, and H. STELLER, 1995  The head involution defective gene of Drosophila melanogaster functions in programmed cell death. Genes Dev. 9:1694-1708[Abstract/Free Full Text].

HARVEY, A., A. BIDWAI, and L. MILLER, 1997  Doom, a product of the Drosophila mod(mdg4) gene, induces apoptosis and binds to baculovirus inhibitor of apoptosis proteins. Mol. Cell. Biol. 17:2835-2843[Abstract].

HAWKINS, C. J., S. L. WANG, and B. A. HAY, 1999  A cloning method to identify caspases and their regulators in yeast: identification of Drosophila IAP1 as an inhibitor of the Drosophila caspase DCP-1. Proc. Natl. Acad. Sci. USA 96:2885-2890[Abstract/Free Full Text].

HAY, B. A., D. A. WASSARMAN, and G. M. RUBIN, 1995  Drosophila homologs of baculovirus inhibitor of apoptosis proteins function to block cell death. Cell 83:1253-1262[Medline].

HINDS, M. G., R. S. NORTON, D. L. VAUX, and C. L. DAY, 1999  Solution structure of a baculoviral inhibitor of apoptosis repeat. Nat. Struct. Biol. 6:648-651[Medline].

KAISER, W. J., D. VUCIC, and L. K. MILLER, 1998  The Drosophila inhibitor of apoptosis D-IAP1 suppresses cell death induced by the caspase drICE. FEBS Lett. 440:243-248[Medline].

KURADA, P. and K. WHITE, 1998  Ras promotes cell survival in Drosophila by downregulating hid expression. Cell 95:319-329[Medline].

LISTON, P., N. ROY, K. TAMAI, C. LEFEBVRE, and S. BAIRD et al., 1996  Suppression of apoptosis in mammalian cells by NAIP and a related family of IAP genes. Nature 379:349-353[Medline].

MOORE, L. A., H. T. BROIHIER, M. VAN DOREN, L. B. LUNSFORD, and R. LEHMANN, 1998  Identification of genes controlling germ cell migration and embryonic gonad formation in Drosophila.. Development 125:667-678[Abstract].

NASSIF, C., A. DANIEL, J. A. LENGYEL, and V. HARTENSTEIN, 1998  The role of morphogenetic cell death during Drosophila embryonic head development. Dev. Biol. 197:170-186[Medline].

NORDSTROM, W., P. CHEN, H. STELLER, and J. M. ABRAMS, 1996  Activation of the reaper gene during ectopic cell killing in Drosophila.. Dev. Biol. 180:213-226[Medline].

ROTHE, M., M.-G. PAN, W. HENZEL, T. AYRES, and D. GOEDDEL, 1995  The TNFR2-TRAF signaling complex contains two novel proteins related to baculoviral inhibitor of apoptosis proteins. Cell 83:1243-1252[Medline].

ROY, N., M. S. MAHADEVAN, M. MCLEAN, G. SHUTLER, and Z. YARAGHI et al., 1995  The gene for neuronal apoptosis inhibitory protein is partially deleted in individuals with spinal muscular atrophy. Cell 80:167-178[Medline].

ROY, N., Q. L. DEVERAUX, R. TAKAHASHI, G. S. SALVESEN, and J. C. REED, 1997  The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16:6914-6925[Medline].

UREN, A. G., M. PAKUSCH, C. J. HAWKINS, K. L. PULS, and D. L. VAUX, 1996  Cloning and expression of apoptosis inhibitory protein homologs that function to inhibit apoptosis and/or bind tumor necrosis factor receptor-associated factors. Proc. Natl. Acad. Sci. USA 93:4974-4978[Abstract/Free Full Text].

VUCIC, D., W. J. KAISER, A. J. HARVEY, and L. K. MILLER, 1997a  Inhibition of Reaper-induced apoptosis by interaction with inhibitor of apoptosis proteins (IAPs). Proc. Natl. Acad. Sci. USA 94:10183-10188[Abstract/Free Full Text].

VUCIC, D., S. SESHAGIRI, and L. K. MILLER, 1997b  Characterization of reaper and FADD-induced apoptosis in a lepidopteran cell line. Mol. Cell. Biol. 17:667-676[Abstract].

VUCIC, D., W. J. KAISER, and L. K. MILLER, 1998a  Inhibitor of apoptosis proteins physically interact with and block apoptosis induced by Drosophila proteins HID and GRIM. Mol. Cell. Biol. 18:3300-3309[Abstract/Free Full Text].

VUCIC, D., W. J. KAISER, and L. K. MILLER, 1998b  A mutational analysis of the baculovirus inhibitor of apoptosis Op-IAP. J. Biol. Chem. 273:33915-33921[Abstract/Free Full Text].

WANG, S. L., C. J. HAWKINS, S. J. YOO, H.-A. J. MÜLLER, and B. A. HAY, 1999  The Drosophila caspase inhibitor DIAP1 is essential for cell survival and is negatively regulated by HID. Cell 98:453-463[Medline].

WHITE, K., M. E. GRETHER, J. M. ABRAMS, L. YOUNG, and K. FARRELL et al., 1994  Genetic control of programmed cell death in Drosophila.. Science 264:677-683[Abstract/Free Full Text].

WHITE, K., E. TAHAOGLU, and H. STELLER, 1996  Cell killing by the Drosophila gene reaper.. Science 271:805-807[Abstract].

WING, J. P., L. ZHOU, L. M. SCHWARTZ, and J. R. NAMBU, 1998  Distinct killing properties of the Drosophila reaper, head involution defective, and grim genes. Cell Death Differ. 5:930-939[Medline].

This article has been cited by other articles:

				L. Wang, J. Evans, H. K. Andrews, R. B. Beckstead, C. S. Thummel, and A. Bashirullah A Genetic Screen Identifies New Regulators of Steroid-Triggered Programmed Cell Death in Drosophila Genetics, September 1, 2008; 180(1): 269 - 281. [Abstract] [Full Text] [PDF]

				A. J. Schile, M. Garcia-Fernandez, and H. Steller Regulation of apoptosis by XIAP ubiquitin-ligase activity Genes & Dev., August 15, 2008; 22(16): 2256 - 2266. [Abstract] [Full Text] [PDF]

				C. D. Freel, D. A. Richardson, M. J. Thomenius, E. C. Gan, S. R. Horn, M. R. Olson, and S. Kornbluth Mitochondrial Localization of Reaper to Promote Inhibitors of Apoptosis Protein Degradation Conferred by GH3 Domain-Lipid Interactions J. Biol. Chem., January 4, 2008; 283(1): 367 - 379. [Abstract] [Full Text] [PDF]

				P. S. Ribeiro, E. Kuranaga, T. Tenev, F. Leulier, M. Miura, and P. Meier DIAP2 functions as a mechanism-based regulator of drICE that contributes to the caspase activity threshold in living cells J. Cell Biol., December 31, 2007; 179(7): 1467 - 1480. [Abstract] [Full Text] [PDF]

				V. P. Yin, C. S. Thummel, and A. Bashirullah Down-regulation of inhibitor of apoptosis levels provides competence for steroid-triggered cell death J. Cell Biol., October 3, 2007; 178(1): 85 - 92. [Abstract] [Full Text] [PDF]

				D. A. Primrose, S. Chaudhry, A. G. D. Johnson, A. Hrdlicka, A. Schindler, D. Tran, and E. Foley Interactions of DNR1 with the apoptotic machinery of Drosophila melanogaster J. Cell Sci., April 1, 2007; 120(7): 1189 - 1199. [Abstract] [Full Text] [PDF]

				M. Balakireva, C. Rosse, J. Langevin, Y.-c. Chien, M. Gho, G. Gonzy-Treboul, S. Voegeling-Lemaire, S. Aresta, J.-A. Lepesant, Y. Bellaiche, et al. The Ral/Exocyst Effector Complex Counters c-Jun N-Terminal Kinase-Dependent Apoptosis in Drosophila melanogaster Mol. Cell. Biol., December 1, 2006; 26(23): 8953 - 8963. [Abstract] [Full Text] [PDF]

				A. Singh, X. Shi, and K.-W. Choi Lobe and Serrate are required for cell survival during early eye development in Drosophila. Development, December 1, 2006; 133(23): 4771 - 4781. [Abstract] [Full Text] [PDF]