Turning on Death in the Fly: Regulation of Apoptosis in Drosophila melanogaster Citation Wang, Susan Leishua (2000) Turning on Death in the Fly: Regulation of Apoptosis in Drosophila melanogaster. Dissertation (Ph.D.), California Institute of Technology. https://resolver.caltech.edu/CaltechTHESIS:08182025-182257078 Abstract Apoptosis, an evolutionarily conserved form of cell suicide, is implemented by a family of cysteine proteases termed caspases. Caspases are constitutively expressed in most cells and function in a proteolytic cascade activated by diverse extracellular and intracellular death stimuli. What molecules govern caspase activity so that it is limited to doomed cells? What mechanisms do these regulators employ? Caspases and their regulators have been identified in a wide variety of species. This dissertation describes the analysis of caspase regulators in Drosophila melanogaster, which has led to the elucidation of some mechanisms that control caspase activity. In Drosophila, three genes, reaper (rpr), head involution defective (hid), and grim, are essential for most normally occurring cell death, and are sufficient to initiate caspase-dependent death when expressed in cells that typically live. The Drosophila IAP homologue DIAP1 is a dosage-dependent suppressor of normally occurring, as well as rpr-and hid-dependent cell death. We show that DIAP1 physically interacts with RPR, HID, and GRIM, and binds to and inhibits the activity of Drosophila caspases. Using a yeast-based caspase activity reporter system and an in vitro reconstitution assay, we find that HID blocks DIAP1's ability to inhibit caspase activity and provide evidence that RPR and GRIM can behave similarly to activate caspases. These observations define a novel point at which caspase activity can be regulated and suggest that one mechanism by which RPR, HID, and GRIM can promote apoptosis is by disrupting productive IAP-caspase interactions. Supporting this hypothesis we show that DIAP1 is required to block apoptosis-inducing caspase activity during Drosophila embryonic development. Elimination of DIAPl function results in global early embryonic cell death and a large increase in DIAP1-inhibitable caspase activity. DIAP1 is still required for cell survival when expression of rpr, hid, and grim is eliminated. Since the death program is constitutively expressed in most cells, the mechanism of cell death activation defined by RPR, HID, and GRIM may be quite general. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Biology) Degree Grantor: California Institute of Technology Division: Biology Major Option: Biology Thesis Availability: Public (worldwide access) Research Advisor(s): Wold, Barbara J. Thesis Committee: Hay, Bruce A. (chair) Wold, Barbara J. Deshaies, Raymond Joseph Dunphy, William G. Fraser, Scott E. Patterson, Paul H. Defense Date: 18 October 1999 Record Number: CaltechTHESIS:08182025-182257078 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:08182025-182257078 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 17632 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 18 Aug 2025 21:18 Last Modified: 18 Aug 2025 21:18 Thesis Files PDF - Final Version See Usage Policy. 49MB Repository Staff Only: item control page

TURNING ON DEATH IN THE FLY: REGULATION OF APO PTO SIS IN DROSOPHILA MELANOGASTER Thesis by Susan Leishua Wang In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 2000 (Submitted October 18, 1999) 11 ©2000 Susan L. Wang All rights reserved Ill All accumulated wealth is eventually dispersed; What goes up must come down. All meetings end in partings; Every life concludes in death. -The Dharmapada IV ACKNOWLEDGMENTS I thank my advisor, Bruce Hay, for making this thesis possible. He was an ideal P.I. I thank faculty members Ray Deshaies, Bill Dunphy, Scott Fraser, Paul Patterson, and Paul Sternberg for their support and advice throughout my years at Caltech. I am grateful to Chris Hawkins and Jeong Yoon, with whom I have enjoyed many enlightening scientific discussions in addition to friendship and encouragement. I thank my teachers Myung Shin, Sarah Gaffen, J. Tso, and Ming-Ji Fann. Finally, I thank my family for always being there for me. V ABSTRACT Turning on Death in the Fly: Regulation of Apoptosis in Drosophila Melanogaster Apoptosis, an evolutionarily conserved form of cell suicide, is implemented by a family of cysteine proteases termed caspases. Caspases are constitutively expressed in most cells and function in a proteolytic cascade activated by diverse extracellular and intracellular death stimuli. What molecules govern caspase activity so that it is limited to doomed cells? What mechanisms do these regulators employ? Caspases and their regulators have been identified in a wide variety of species. This dissertation describes the analysis of caspase regulators in Drosophila melanogaster, which has led to the elucidation of some mechanisms that control caspase activity. In Drosophila, three genes, reaper (rpr), head involution defective (hid), and grim, are essential for most normally occurring cell death, and are sufficient to initiate caspasedependent death when expressed in cells that typically live. The Drosophila IAP homologue DIAPI is a dosage-dependent suppressor of normally occurring, as well as rprand hid-dependent cell death. We show that DIAPI physically interacts with RPR, HID, and GRIM, and binds to and inhibits the activity of Drosophila caspases. Using a yeastbased caspase activity reporter system and an in vitro reconstitution assay, we find that HID blocks DIAPI 's ability to inhibit caspase activity and provide evidence that RPR and GRIM can behave similarly to activate caspases. These observations define a novel point at which caspase activity can be regulated and suggest that one mechanism by which RPR, HID, and GRIM can promote apoptosis is by disrupting productive IAP-caspase interactions. VI Supporting this hypothesis we show that DIAPl is required to block apoptosisinducing caspase activity during Drosophila embryonic development. Elimination of DIAPl function results in global early embryonic cell death and a large increase in DIAPlinhibitable caspase activity. DIAPl is still required for cell survival when expression of rpr, hid, and grim is eliminated. Since the death program is constitutively expressed in most cells, the mechanism of cell death activation defined by RPR, HID, and GRIM may be quite general. Vil TABLE OF CONTENTS Acknowledgments IV Turning on Death in the Fly: Regulation of Apoptosis in Drosophila Melanogaster V Chapter 1: Introduction: Mechanisms of Caspase Regulation 1 Chapter 2: A Cloning Method to Identify Caspases and Their Regulators in Yeast: Identification of Drosophila IAPI as an Inhibitor of the Drosophila Caspase DCP-1 42 The Drosophila Caspase Inhibitor DIAPI Is Essential for Cell Survival and Is Negatively Regulated by HID 81 Conclusion: Future Directions 125 Abstract Chapter 3: Chapter 4: Appendix I 135 Appendix II 142 CHAPTER 1 Introduction : Mechanisms of Caspase Regulation Susan L. Wang 2 Apoptosis is an evolutionarily conserved, intrinsic process by which the organism abolishes undesired cells. The most common form of cell death, apoptosis is indispensable during development as well as in the adult metazoan animal. Its roles include: structure remodeling, lymphocyte control, tissue size homeostasis, and removal of abnormal and potentially harmful cells (reviewed in Jacobson et al., 1997). Apoptosis is executed by specialized proteolytic machinery involving a family of proteases termed caspases (Alnernri et al., 1996). Caspases are the essential components of many, if not all apoptotic pathways. They exist in most living cells as zymogens which are activated by proteolytic processing in response to death-inducing stimuli. Caspase activation is mediated either autocatalytically or by another protease, triggered by cofactor binding or inhibitor removal. Regulated caspase activation permits accumulation of large amounts of precursor that can be rapidly activated on demand (Cryns and Yuan , 1998 ; Nicholson and Thornberry , 1997; Salvesen and Dixit, 1997; Thornberry and Lazebnik, 1998). Since procaspases are widely expressed at high levels and have low but significant protease activity, efficient mechanisms must exist to prevent inappropriate caspase activation. Caspases: effectors of apoptosis Studies of developmental cell death in the nematode Caenorhabditis elegans have provided important clues about the molecular nature of the death program. A group of genes , termed ced for cell death-defective, regulates cell death in worms (Ellis et al., 1991). One of these genes, ced-3, is required for carrying out cell death and is a member of a family of cysteine proteases that form the core of the apoptotic machinery (Yuan et al., 1993). These proteases effect death by cleaving cellular substrates at specific aspartate residues and are thus called caspases (Alnemri et al., 1996). Since the discovery that CED-3 protein bears marked sequence similarity to interleukin- I ~-converting enzyme (ICE, 3 caspase-1 ), a mammalian protease responsible for proteolytic maturation of prointerleukin-1 ~, many additional caspase family members have been isolated in a variety of organisms (Thornberry et al., 1992). Although caspases are critical for the execution of many apoptotic pathways, their exact physiological roles have not been well defined. Gene targeting studies have shed light on the distinctive role of individual caspases in cell death as well as in other biological processes (reviewed in Los et al., 1999). In particular there are caspases that, rather than regulate apoptosis, primarily mediate immune responses by controlling proinflammatory cytokine processing. Caspases may also be involved in the regulation of cell proliferation, a process which is likely to share some machinery involved in programmed cell death. For example, cell division during proliferation requires enzymes involved in cytoskeletal rearrangement and the disruption of cell adhesion, two events that function during apoptosis. A primary phenotype of flies lacking zygotic Drosophila caspase-1 (DCP-1) is the absence of imaginal discs and gonads, hinting at a possible role in regulating cell proliferation in these tissues (Song et al., 1997). Several lines of evidence indicate that caspases are important for apoptosis. First, worms without functional CED-3 have a complete absence of developmental programmed cell death (Yuan et al., 1993). Second, targeted deletion of caspase and caspase regulator genes in mice have shown a definitive role for caspases in apoptosis. Caspase-3, caspase8, and caspase-9 knockout animals all die at birth due to serious defects in developmental apoptosis (Los et al., 1999). Ablation of the caspase-8 regulator, FADD, is embryonic lethal with mutant mice exhibiting a phenotype similar to that of caspase-8 deficient animals (Yeh et al., 1998). Mice lacking apoptotic protease-activating factor 1 (Apaf-1 ), a caspase-9 regulator, have a phenotype even more severe than caspase-9 knockout mice. Both Apaf-1 and caspase-9 knockout mice exhibit brain hyperplasia, craniofacial alterations, and persistence of interdigital webs (Cecconi et al. , 1998; Hakem et al., 1998; 4 Kuida et al., 1995; Yoshida, 1998). Third, loss of a functional caspase in flies also indicates caspase involvement in cell death regulation: flies lacking DCP-1 develop melanotic tumors and die as larvae (Song et al., 1997). Fourth, ectopic expression of the viral protein caspase inhibitor CrmA blocks apoptosis in a number of different systems (Ray et al., 1992). Addition of the cell-permeable peptide caspase inhibitor Z-V AD-fmk [benzyloxycarbonyl-Val-Ala-Asp(OMe) fluoromethyl ketone] to cultured cells, including thymocytes, hepatocytes, and neuronal cells, potently inhibits apoptosis induced by a wide range of stimuli (Armstrong et al., 1996; Chow et al., 1995). Similarly, the pan-caspase inhibitor baculovirus P35 blocks normally occurring developmental cell death in the Drosophila embryo and eye when ectopically expressed in these tissues (Hay and Rubin, 1994). Caspase substrates Identification of caspase substrates has provided insight into how caspases promote the cellular changes that occur during programmed cell death. Pro- and anti-apoptotic regulatory proteins are likely to be the first caspase targets following a death stimulus. Cleavage of these targets results in signal amplification and caspase inhibitor inactivation. For example, caspase-3 cleaves apoptotic regulators Bcl-2 (Bel, B cell lymphoma) and Bel-XL, destroying their anti-apoptotic function and releasing proapoptotic fragments (reviewed in Gross et al., 1999). Bcl-2 and Bel-XL act as death inhibitors upstream in pathways resulting in caspase activation. Proapoptotic Bcl-2 family member BID is cleaved by caspase-8, releasing a fragment that induces an efflux of cytochrome c , as well as other apoptotic effectors, from the mitochondria. Cytochrome c is required for the activation of Apaf-1, which promotes activation of procaspase-9 (Li et al., 1997). 5 Another category of caspase substrates includes housekeeping or structural proteins whose cleavage is required for the ordered disassembly of the cell. Apoptotic cells die a morphologically distinguished death characterized by condensation and fragmentation of nuclear chromatin, compaction of cytoplasmic organelles, a decrease in cell volume, and alterations to the plasma membrane resulting in recognition and phagocytosis of the dying cell (Kerr et al., 1972). Caspases cleave several proteins involved in cytoskeleton regulation, including gelsolin and focal adhesion kinase (Kothakota et al., 1997; Wen et al., 1997). Cleavage of these proteins results in deregulation of their activity. A caspase-3-activated fragment of gelsolin severs actin filaments, promoting both cytoplasmic and nuclear apoptosis, including DNA fragmentation. Caspases directly cleave lamins, causing lamina to collapse and contributing to chromatin condensation (Orth et al., 1996; Takahashi et al., 1996). Proteolysis of adhesion molecules such as ~-catenin by caspases may lead to disruption of cell-cell contacts (Brancolini et al., 1997). Altogether, cleavage of these structurally related substrates is believed to promote cellular packaging and subsequent engulfment of dead cell debris by phagocytes. The nuclear changes during apoptosis are associated with internucleosomal cleavage of DNA, recognized as a "DNA ladder" by agarose gel electrophoresis. Although DNA fragmentation now appears to be a relatively late event in the apoptotic process and in some models may be separated from early critical steps, its measurement is simple and it is often used as a major criterion to determine if a cell is apoptotic. Many proteins involved in DNA synthesis and repair, such as the nuclear replication factor MCM3, the large subunit of DNA replication factor C and human Rad5 l, have been identified as caspase substrates in apoptotic cells (Flygare et al., 1998; Rheaume et al., 1997; Schwab et al., 1998; Ubeda and Habener, 1997). Moreover, caspase-3 cleaves the 6 caspase-activated DNase (CAD) inhibitor ICAD, allowing the released nuclease to cleave DNA and promote nuclear condensation (Liu et al., 1997; Sakahira et al., 1998). Caspase stmcture Caspases are formed as latent precursors consisting of a prodomain and a region containing the two subunits that comprises the catalytic domain. These zymogens are activated in vivo following proteolytic cleavages after specific aspartate residues. In general, there is an initial cleavage or two that separates the small C-terminal subunit from the rest of the molecule and allows formation of an active protease that cleaves off its own prodomain. Based on the crystal structure of caspase-1 and caspase-3, mature caspases are heterotetramers composed of two large subunits and two small subunits (Mitt! et al., 1997; Rotonda et al., 1996; Walker et al., 1994; Wilson et al., 1994). The caspase active site is composed of residues from both subunits and contains a positively charged S 1 subsite that binds the P 1 aspartate of the substrate. Caspases contain an active site cysteine within a conserved QACXG motif (where Xis R, Q or G) . Substrate specificity varies with individual caspases depending on the composition of their residues present in the substrate binding site. The tertiary structure and residues within the active site can vary significantly among different caspases, consistent with the diversity of their biological functions. The prodomain and large and small subunits are derived from the proenzyme by cleavage following aspartates in regions that often resemble consensus caspase target sites. This observation suggests that caspases can be activated either autocatalytically or in a cascade by enzymes with similar specificity. Based on their structure and order in cell death pathways, caspases can be classified as either initiator or effector caspases. Initiator caspases, which often contain large prodomains, can promote activation of other family 7 members by cleaving them, generating a proteolytic cascade of caspase activation. The prodomains of several apical caspases contain protein modules that are conserved in many apoptosis regulatory molecules. Death effector domains (DEDs) and caspase recruitment domains (CARDs) are two types of motifs found in prodomains of initiator caspases. Both domains consist of six anti-parallel a-helices contained in a similar three-dimensional fold. These motifs associate with homologous regions in other proteins, thus prompting binding of proteins to one another (Eberstadt et al., 1998; Huang et al., 1996). The DED and CARD have been proposed to play a regulatory role in apoptosis by coupling caspase regulators to procaspases. Mammalian caspases Mammalian caspases comprise a group of at least fourteen members that can promote apoptosis. Based on substrate specificity studies and phylogenetic analysis, caspases can be classified into three subfamilies (reviewed in Cohen, 1997; Cryns and Yuan, 1998; Nicholson and Thornberry , 1997). The ICE-like family consists of caspase-1, -4, -5, 13, and -14 as well as murine caspase-11 and - I2. The CED-3 subfamily includes caspase-3 , -6, -7, -8, -9, and -10. A third subfamily has only one member, caspase-2. Members of a caspase subfamily recognize similar peptide sequence target sites within their substrates, and some can cleave the same molecules, indicating functional redundancy within the caspase family. Drosophila caspases The apoptosis death program is highly conserved among vertebrates and invertebrates. Several Drosophila caspases have been identified recently. They include DCP-1, 8 DREDD/DCP-2, DRICE, DRONC, DECAY (Chen et al. , 1998; Dorstyn et al. , 1999; Fraser and Evan, 1997; Song et al., 1997). Caspase activation There is strong evidence that the death program is constitutively expressed in most cells (Jacobson et al., 1997; Shaham and Horvitz, 1996; Steller, 1995). Various receptormediated apoptosis occur in the presence of inhibitors of either RNA or protein synthesis (Itoh et al., 1991 ). Even enucleated cells undergo apoptosis upon Fas activation suggesting that all of the components necessary for apoptotic signal transduction are present and that Fas activation simply triggers this machinery (Schulze-Ostoff et al., 1994). Initiation of the apoptotic program involves the activation of latent caspase zymogens. In the cascade model for effector caspase activation, a proapoptotic signal culminates in the activation of an initiator caspase which, in turn, activates effector caspases. Effector caspases then carry out proteolytic cleavages of cellular proteins ultimately resulting in cell death. Events likely to be involved in caspase activation include: the removal of inhibitors, association with proapoptotic cofactors, conformational changes, and proteolytic processing. Current evidence suggests that caspase processing may occur by autoactivation, transactivation, or proteolysis by other proteases such as granzyme B (Greenberg, 1996). Proapoptotic signals can originate from outside the cell through means such as receptor signaling, granzyme B uptake, and death peptide ingestion. Alternatively, the cell can sense internal stress, such as damaged DNA, and choose to commit suicide. Different upstream death signals are mediated by distinct initiator caspases. Three major caspase activation pathways have been elucidated to some extent. One pathway, which involves a 9 complex called the "apoptosome," functions similarly in organisms ranging from nematodes to mammals. Another route, featured in mammalian cell death, is triggered by and directly linked to "death receptors." A third apoptotic pathway operates through the inhibition of caspase inhibitors. Models for caspase autoactivation 1. The "facilitated autocatalysis" model The "facilitated autocatalysis" model is based on the hypothesis that procaspases are present in cells in an inactive conformation or complex that precludes autocatalysis. Cofactors promote activation by changing the conformation of the precursors either directly or by removing an inhibitor. Induction of apoptosis by arginine-glycine-aspartate (RGD) peptides may utilize such a mechanism. In a cell culture model, soluble RGD peptides cause apoptosis by activating procaspase-3 in cells (Buckley et al., 1999). In this scenario, the RGD peptide may out-compete an RGD motif in procaspase-3 and disrupt an intramolecular interaction with a DDM sequence. When this bond is blocked, the procaspase changes conformation, allowing it to become activated. The structure of the unprocessed pro-form of caspase-3 is needed to verify the existence of an intramolecular interaction between RGD and DDM. Experiments using purified components and sitedirected mutagenesis may aid in understanding the exact mechanism of caspase-3 activation by RGD peptides. 2. The "induced proximity" model A caspase cascade may initiate by autoproteolysis when two or more zymogen molecules aggregate. The activation of molecules by their clustering is a mechanism utilized in other biological processes, such as the activation of tyrosine kinases and complement cascade 10 proteins (Davie et al., 1991; Ullrich and Schlessinger, 1990). Overexpression of wildtype caspases, but not catalytically inactive mutants, results in caspase processing and activation (Orth et al., 1996). This suggests that autoactivation may be facilitated by high proenzyme concentration. Affinity-labeling experiments demonstrate that procaspase molecules have weak proteolytic activity (Muzio et al., 1998; Yamin et al., 1996). When zymogens are brought together in close contact, their low activity may be sufficient for trans-proteolytic generation of active enzyme. Once formed, active enzyme may cleave and activate downstream procaspases, thereby initiating a proteolytic cascade leading to apoptosis. In vivo, adapter molecules mediate oligomerization of long prodomain containing initiator procaspases. Using specific domains, adapter molecules connect procaspases to apoptotic sensors such as mitochondria and death receptors. CARDs, DEDs, and death domains (DDs) are three types of domains found in adapter molecules. Comparable to DEDs and CARDs, DDs associate via like-like interactions. A CARD domain can be found in many apoptotic molecules, including Apaf-1 and the prodomain of caspase-9. DDs are present in several death-inducing receptors of the tumor necrosis factor (TNF)-receptor family, including Fas, TNF-Rl, and others. They are also present in many cytoplasmic adapter proteins that interact with the DD-containing receptors. The adapter molecule FADD (Fas-associated death domain protein) contains both a DD and DED. Two DEDs are present in tandem in both caspase-8 and caspase-10 (reviewed in Wolf and Green, 1999). Pathways of caspase activation 1. The apoptosome Genetic studies of developmental cell death in C. elegans have been very successful in identifying the basic components of the apoptotic machinery. Through the use of genetic 1I screens, a pathway for caspase activation involving several worm proteins has been delineated. All of these proteins have homologues in mammals and fly. They include: CED-9, a Bcl-2-like protein; EGL-I, a BH3 (Bcl-2 Homology region 3)-only protein; CED-4, an Apaf-1-like protein which contains both a CARD and an ATPase domain; and CED-3 (Conradt and Horvitz, 1998; Hengartner and Horvitz, 1994; Yuan and Horvitz, 1992; Yuan et al., 1993; Zou et al. , 1997). In a process requiring the activity of the CED4 A TPase domain, the CED-4 CARD domain binds to a homologous region in the prodomain of CED-3. Although CED-4 and CED-3 associate, in cells that live, the CED-9 protein blocks CED-4 from aggregating and activating the CED-3 procaspase. It is thought that CED-9 localizes the complex (dubbed the "apoptosome") to intracellular membranes. In cells that die, apoptotic cofactor EGL-I binds to CED-9, thereby releasing the complex of CED-4 and CED-3. The CED-4 can now aggregate, which brings the bound CED-3 procaspase molecules close enough together so that they can activate each other (reviewed in Hengartner, 1997). The exact mechanism of CED-3 activation is not clear. The induced proximity resulting from CED-4 clustering may be sufficient for CED3 autoactivation. Conceivably, in a reaction inhibited by CED-9 binding, CED-4 catalyzes the hydrolysis of ATP, which could provide energy required for a conformational change and autocatalytic activation of CED-3. In mammals, CED-3 and CED-4 homologous molecules caspase-9 and Apaf-1 are important for developmental cell death and oncogene- and p53-dependent apoptosis (Cecconi et al., 1998; Fearnhead et al., 1998; Soengas et al., 1999; Yoshida, 1998). The mammalian apoptosome is more complicated than the worm machinery in that it appears to have an additional regulatory component involving the mitochondria as a death sensor. Apaf-1, like CED-4, consists of a CARD and a CED-4-like (Walker A and B loops composing the ATPase) domain (Zou et al., 1997). However, unlike CED-4, Apaf-1 also contains a WD40-repeat region required for cytochrome c binding. An apoptotic stimulus 12 releases cytochrome c from the mitochondrial intermembrane space into the cytoplasm. Proapoptotic BH3-containing molecules, such as Bax and Bak, are believed to be involved in this process (Shimizu et al., 1999). They stimulate opening of the voltage-dependent anion channel (VDAC), a mitochondrial channel through which the cytochrome c permeates. Conversely, anti-apoptotic protein Bel-XL closes VDAC by binding to it directly and blocking cytochrome c release. Apaf-1 is activated through a conformational change induced by the binding of ATP and cytochrome c. Activated Apaf-1 molecules oligomerize, providing a platform for caspase-9 activation. The oligomerized Apaf-1 complex recruits procaspase-9 to the apoptosome via its CARD domain. Consequently, the "induced proximity" of procaspase-9 molecules by cytochrome c-activated Apaf-1 stimulates their proteolytic processing to mature caspases. Once activated, caspase-9 can cleave and activate procaspase-3 and other downstream caspases. 2. "Death receptors" and DISC operation The TNFR fami ly systems regulate immune-related cell death and survival . They allow the cell to adjust to extracellular insult through brief, intense responses which are characterized by a remarkable duality. On one hand, they can induce the cell to commit suicide. On the other hand, in the appropriate environment, they promote repair and growth. Two of the best characterized TNFR family members are Fas and TNF-receptor1 (TNF-Rl). The Fas ligand, FasL, rapidly signals apoptosis by triggering the formation of a death-inducing signaling complex (DISC) (Chinnaiyan et al., 1995). When Fas is ligated, its clustered DDs bind a homologous domain in FADD. Subsequently, FADD recruits a complex consisting of procaspase-8 and a DED-like (DRD)-domain containing molecule, FLASH ["FLICE (caspase-8)-associated huge" protein] via homophilic interactions requiring DEDs (Imai et al. , 1999). FLASH's DRD binds to the DEDs in procaspase-8. 13 Figure I. Simplified diagram of the DISC. 14 .,..,. DED procaspase-8 t N caspase-8 15 Interestingly , FLASH contains a CED-4-like domain through which it can self-associate and bind ATP. In the DISC, FLASH may function in a similar manner to CED-4/Apaf-1 in an apoptosome-like complex. Following recruitment to the DISC, procaspase-8 is cleaved and processed, generating active caspase-8. As mentioned above, one caspase-8 substrate is the Bcl-2 family member BID. Cleavage of BID results in cytochrome c release, leading to apoptosome formation and caspase-9 activation (Li et al., 1998). Caspase-8 also directly activates caspase-3, -6, and -7. Caspase-3 can cleave and convert Bcl-2 to a BH3-only death effector, amplifying the caspase activation cascade by feeding into the vicious "circle of death" (Cheng et al., 1997). The apoptotic signaling pathway triggered by ligation of TNF shares some aspects with that of FasL (Ashkenazi and Dixit, 1998). TNF-R 1 signal transduction can lead to apoptosis in some cell types ; however, in most cells TNF-induced apoptosis is blocked by suppressive factors believed to be controlled by pathway involving NFKB. When TNFRl is ligated, it trimerizes. Subsequently, DD-containing adapter TRADD (7NFRassociated death domain) binds to the clustered receptors' DDs. Oligomerized TRADD recruits various signaling molecules to the receptor complex, depending on the repertoire of proteins available in the cell. For example, the DD-containing Ser/Thr kinase receptorinteracting protein (RIP) is required in the complex for the activation of NFKB (Kelliher et al., 1998; Ting, 1996). In a death-conducive environment FADD couples the TNFRITRADD complex to activation of caspase-8. Many questions regarding DISC operation remain . For example, it is unclear whether active caspase-8 must remain complexed with, or if it is released from, the DISC in order to be fully active. Other questions include: how many procaspase molecules are required for activation? Does activation require dimerization? Does processing occur in 16 cis or trans? Are there other molecules involved in DISC operation? What is FLASH's role in caspase-8 processing? Answers to these questions will help to clarify the molecular mechanisms involved in caspase activation downstream of death receptor signaling. The "induced proximity" hypothesis argues that caspases remain dormant in the cell until activated because they are present at low concentration as monomers. Whether caspase zymogens are indeed monomers in living cells remains to be tested. Analysis of nonprocessable mutant procaspases, generated by replacement of the two Asp cleavage sites separating the large and small subunits with Ala, revealed that procaspases vary in their level of intrinsic enzymatic activity (Muzio et al., 1998; Stennicke et al., 1999). For example, procaspase-8 possesses approximately I% of the proteolytic activity of mature caspase-8. Procaspase-9 is nearly as active before as it is after processing; however, there is evidence that purified, recombinant, processed caspase-9 is approximately two thousand-fold more active when associated with cytosolic factors supplemented with cytochrome c and dATP than in isolation (Stennicke et al., 1999). This and other data suggest that caspase-9 requires complex formation with cytosolic factors, possibly the apoptosome, to be fully active (Lazebnik, in press). 3. REAPER, HID, and GRIM In Drosophila melanogaster, caspases are important for cell death, although the role of particular caspases in apoptotic pathways is currently unknown (Hay and Rubin , 1994 ). The fly death activators REAPER (RPR), head involution defective (HID), and GRIM act upstream in pathways resulting in caspase-dependent cell death (Bump et al., 1995; Xue and Horvitz, 1995). Chapter 3 of this thesis describes a "facilitated autocatalysis"-like model in which RPR, HID, and GRIM promote apoptosis by directly inhibiting a caspase inhibitor (Wang et al., 1999). In this paradigm, the caspase inhibitor blocks death at the 17 level of either pro- or mature caspase. Thus, in certain situations, RPR, HID, and GRIM act to directly activate already processed mature caspases. In other cases, they liberate inhibitor-bound procaspases which subsequently undergo processing via other caspase activation pathways, such as the DISC or apoptosome. The Drosophila homologue of Apaf-1 is Dark (Drosophila Apaf-1-related killer) (Rodriguez et al. , 1999). RPR-, HID-, and GRIM-induced death is partly suppressed by a loss of Dark in embryos, suggesting that the three activators act as upstream regulators of apoptosome-mediated caspase activation. Caspase inhibition Cells employ a variety of methods to ensure that caspase activation is restricted to doomed cells. Inappropriate triggering of caspase activation is controlled at many stages; there are inhibitory mechanisms that function: 1) upstream of caspase activation, 2) at the procaspase level, and 3) directly on mature caspases. Decoy receptors and DED/CARDcontaining inhibitory molecules block signal transduction of death receptors. Compartmentalization of death-inducing factors is another way to control caspase activation: the apoptotic cofactor cytochrome c is kept sequestered in the mitochondrial inner membrane space until appropriate death stimuli promote its release into the cytosol, where it activates Apaf-1 . Procaspases and mature caspases can be directly regulated by protein phosphorylation. There are several known caspase inhibitor proteins. Work on viruses, which attenuate apoptosis to circumvent the normal host response to infection, has led to the identification of several caspase inhibitors. These include CrmA, P35 and SLP49, and a family of IAP (inhibitor of apoptosis) proteins. Of the viral caspase inhibitors, only the IAPs are known to have mammalian members thus far. 18 Decoy receptors A subgroup of TNF receptor homologues consists of decoy receptors, which function as inhibitors, rather than transducers of signaling. This subgroup includes two members that are cell-surface molecules, decoy receptor (DcR) 1 and DcR2 (Ashkenazi and Dixit, 1998). Two additional members, osteoprotegerin (OPG) and DcR3 are secreted, soluble proteins. A death ligand closely related to FasL called Apo2L was discovered through DNA database screening (Pitti et al., 1996). Like FasL, Apo2L potently induces apoptosis in tumor cells. It can bind several cell-associated receptors, including DR4, DRS, DcRl, DcR2, as well as the secreted OPG. DR4 and DRS cytoplasmic regions both contain DDs, and signal apoptosis upon ligation. DcRI is a glycosyl phosphatidylinositol-anchored cell surface protein that resembles DR4 and DRS but lacks a cytoplasmic tail. DcR2 has a short cytoplasmic domain that does not signal apoptosis. The extracellular domains of DcRI or DcR2 compete with those of DR4 and DRS for ligand binding; thus the DcRs appear to function as decoys that block Apo2L from triggering apoptosis through DR4 and DRS. DcR3 competes with Fas for FasL binding and inhibits apoptosis induction by Fas (Pitti et al. , 1998). Therefore, DcR3 is a decoy receptor for FasL. By modulating the ratio of decoy receptors to death receptors, a cell can regulate its response to death ligands. Expressing decoy receptors at high levels may be a mechanism to protect against Apo2L toxicity. FLIPs and ARC The multiplicity of interacting proteins in DISC formation provides points of ramification, allowing induction of different effects stemming from one receptor. Adapter-like molecules can block death receptor signaling in a manner similar to decoy receptors. These include the FLIPs (FLICE-inhibitory proteins), and ARC (apoptosis repressor with CARD) . FLIPs exist in a short and long form, referred to as FLIP5 and FLIPL, 19 respectively. FLIPs contains two DEDs, much like its viral counterparts, vFLIPs. vFLIPs block death induction by interfering with the binding of OED-containing caspases to FADD. Similarly, FLIPs can effectively inhibit apoptosis triggered by all known death receptors . FLIPL contains the same two DEDs found in FLIPs plus a caspase-like domain in which a tyrosine has been substituted for the cysteine normally found in the active site. The exact role FLIP Lplays in caspase regulation is controversial; some studies show that it inhibits while others indicate a proapoptotic function for FLIP L (Goltsev et al., 1997; Hu et al., 1997; Irmler et al., 1997; Shu et al., 1997). Possible explanations for the discrepancy in activity are that FLIPL's function may vary depending on its expression level or the differentiation state of the cell. As an apoptotic inhibitor, FLIPLmay act similarly to FLIPs and vFLIPs. In addition, FLIPL may bind to prodomains of caspase-8 and caspase-10 via its caspase-like domain and prevent proteolytic self-processing of the caspases (lrmler et al., 1997). Another decoy adapter molecule, ARC, is a human protein expressed in skeletal muscle and heart (Koseki et al., 1998). It contains a CARD that interacts with the prodomains of caspase-2 and caspase-8, and blocks caspase-8 processing. The factors that determine how procaspases select between their cofactors and their decoys, or how cofactors select between procaspases and their decoys, are not known. Compartmentalization Compartmentalization of caspases and their cofactors is another way of regulating caspase activation. For example, cytochrome c is kept sequestered in the intermembrane space of the mitochondria, where it is an essential component of the respiratory electron-transport chain. As mentioned above, it also functions as a potent activator of apoptosis when released into the cytosol. In addition to cytochrome c, there are some procaspases that reside within the intermembrane space in certain tissues (Susin et al., 1999). For 20 example, procaspase-9 can be released from the mitochondria and activated in the cytosol of certain neuronal cell lines and canine ischaemic brain tissue (Krajewski et al., 1999). Once activated, it facilitates apoptosis by activating downstream caspases. Neurons and myocardiocytes have a high requirement for ATP and may therefore be more susceptible to apoptosis, as ATP is a driving force behind apoptosome formation. Localization of caspase-9 zymogen to the mitochondria provides extra protection against accidental activation of the caspase cascade. Phosphorylation The serine/threonine kinase known as Akt can exert its anti-apoptotic effects in a variety of ways. PI3K, an effector of Ras, generates second messengers that activate Akt (Downward, 1998). Akt phosphorylation of Ser 196 of mature caspase-9 inhibits protease activity directly (Cardone et al., 1998). Furthermore, it blocks procaspase-9 autocatalytic self-processing. It is not known how phosphorylation inhibits caspase activity. Phosphorylation may allosterically affect subunit dimerization or alter the catalytic machinery of the substrate cleft through conformational changes. Akt also phosphorylates the proapoptotic BH3-only protein Bad, preventing Bad from binding and inhibiting BelXL (Datta et al., 1997). Consequently, Akt indirectly blocks caspase-9 processing by precluding apoptosome formation. Even more upstream in another caspase-activation pathway, Akt phosphorylates mammalian members of the Forkhead family, causing them to be retained in the cytoplasm where they cannot impel transcription of the gene encoding FasL (Brunet et al., 1999; Kops et al., 1999). CrmA, P35, SLP49 Viruses can block caspase activity to inhibit host cell death. They do this with genes whose products directly bind and block caspases. The cowpox virus-encoded cytokine 21 response inhibitor A (CrmA) forms a stable complex with caspases and can effectively block both TNF- and Fas-mediated apoptosis (Gagliardini et al., 1994; Ray et al., 1992; Tewari et al., 1995). The product of the p35 gene of Autographa californica multiply embedded nucleopolyhedrovirus (AcMNPV) inhibits a broad range of caspases, acting as a "suicide" inhibitor of ICE-like and CED-3 family caspases (Bump et al., 1995; Xue and Horvitz, 1995). P35 is cleaved by caspases, generating two polypeptides that form a stable inhibitory complex with the caspase. There are two homologues of AcMNPV P35, Bombyx mori NPV (BmNPV) P35, and SLP49 from Spodoptera littoralis NPV (SINPV) (Du et al., 1999). Thus far, p35-like genes have only been identified in viruses, but it would not be surprising if they are found to exist in metazoan genomes as well. Inhibitor of apoptosis protein (IAP) The IAPs constitute a subfamily of the BIR (baculovirus /AP repeat)-containing proteins (BIRPs) (reviewed in Miller, 1999). BIRPs are found in viruses, yeast, and metazoans. Evidence suggests that some BIRPs regulate cytokinesis and/or mitotic spindle function during cell division while others regulate apoptosis. IAPs were originally identified in baculovirus, where they were found to block host cell death induced by viral infection. Subsequently, the Drosophila IAP homologue DIAPl was identified in a genetic screen as a modifier of RPR-induced cell death (Hay et al., 1995). DIAPl and another fly IAP homologue, DIAP2, act as dosage-dependent suppressors of RPR- and HID-induced cell death in transgenic flies. Overexpression of DIAPl can block death upon coexpression with the apoptotic inducer RPR in insect cell cultures (Vucic et al., 1997). Ectopic expression of certain mammalian IAPs in cultured cells can suppress apoptosis induced by a variety of stimuli including TNF and Fas (Ambrosini et al., 1997; Duckett et al., 1996). IAPs block cell death at distinct steps in apoptotic pathways induced by Fas or another pro-apoptotic molecule, Bax (Deveraux et al., 1998). The mammalian IAPs 22 XIAP, cIAPl, and cIAP2 can directly bind to and inhibit procaspase-9 as well as mature caspase-9 (Deveraux et al., 1998). Since these IAPs can inhibit caspase-9, Bax-mediated caspase-3 processing through cytochrome c release is also blocked. XIAP and other IAPs can also directly inhibit mature caspase-3 and caspase-7 activity (Deveraux et al., 1997; Roy et al., 1997). Studies using purified, recombinant proteins in vitro have shown that DIAPl and several mammalian IAPs, including XIAP, cIAP, and cIAP2, can directly inhibit procaspase processing and/or mature caspase activity by direct interaction (Deveraux et al., 1997; Hawkins et al., 1999; Roy et al., 1997). Although some studies indicate that certain BIR domains of IAPs are sufficient for this binding, the exact nature of the interaction between IAPs and pro- and mature caspases has not been characterized. Many IAPs contain a RING finger C-terminal to one or more BIRs. However, the requirement of the RING finger for anti-apoptotic activity differs depending on the IAP and/or the nature of the apoptotic stimulus (Clem and Miller, 1994; Takahashi et al., 1998; Vucic et al., 1997). There is evidence that the different BIR domains within an IAP can perform distinct functions. During Fas-mediated apoptosis in cultured cells, ectopically expressed XIAP is cleaved by caspase (Deveraux et al., 1999). After cleavage, XIAP is separated into two molecules: one that contains two BIR domains, BIR 1-2, and another part consisting of the third BIR domain and the RING finger (BIR3-RING). BIR 1-2 inhibits Fas-mediated apoptosis, although with slightly less potency than full-length XIAP. In contrast, ectopic expression of BIR3-RING enhances Fas-induced cell death while potently inhibiting Bax-induced cytochrome c-mediated apoptosis. Studies using purified proteins revealed that BIR 1-2 can directly inhibit caspase-3 and caspase-7 while BIR3-RING only inhibits caspase-9. Therefore, IAPs have potential for inhibiting different caspases depending on the repertoire of their domains. In addition to the BIRs and RING, cIAPl and cIAP2 also contain an internal CARD domain . Although its function has not been determined, the IAP CARD can bind to ARC through a homophilic 23 interaction (Koseki et al., 1998). Some RING finger-containing proteins have the ability to facilitate ubiquitin-conjugating enzyme (UBC or E2)-dependent ubiquitination (Lorick et al. , 1999). Since many IAPs contain RING fingers, it is conceivable that they play a role in modulating protein levels via ubiquitination. An attractive possibility is that IAPs target proapoptotic molecules, such as caspases, for degradation. Conversely, IAPs may target themselves for degradation upon receipt of appropriate death stimuli. IAPs can bind other proteins, including several molecules implicated in cell death and survival. Among these are TNFR-associated factors (TRAFs), RPR, HID, GRIM, and DOOM (Harvey et al., 1997; Rothe et al., 1995; Vucic et al. , 1997). In some cases, IAP may act to suppress cell death via their association. Other interactions may serve to disrupt IAP inhibition of caspases, resulting in caspase activation and, subsequently, cell death. How are caspase inhibitors involved in the regulation of apoptosis? Caspase inhibitors may function to establish thresholds that determine the concentration of active effector caspases required to initiate cell disassembly. Consequently, they avert the dire consequences of accidental proenzyme activation. Overview of thesis The core components of the death machinery and their regulators have been identified in wide variety of species, including C.elegans, Drosophila, and mammals. Despite these major findings, our knowledge of the central mechanisms that underlie apoptosis is still rudimentary. Given the complexity of regulatory circuits that modulate caspase activity, there are likely to be many more molecules involved in the processes described above. Because the yeast Saccharomyces cerevisiae lack many of the specialized proteolytic 24 systems found in cells of higher eukaryotes, it provides an ideal environment to carry out screens for proteases, their regulators and their targets. Chapter 2 of this thesis describes a yeast-based system to identify caspases and their regulators. Chapter 2 also describes the isolation and in vitro characterization of a Drosophila caspase inhibitor. How is the cell death program activated in unwanted cells and kept quiescent in others? Given that the basic mechanism of apoptosis is widely conserved, Drosophila provides an attractive system to address this question because it is highly tractable for experimental analysis. The strength of the Drosophila model system lies in its usefulness in assessing the requirement for a gene in normal apoptosis, and in placing the gene in pathways relative to previously characterized functions. How do the Drosophila death activators RPR, HID, and GRIM promote apoptosis? Chapter 3 describes a proposed mechanism by which RPR, HID, and GRIM trigger death: by disrupting !AP-mediated inhibition of caspase activity. Furthermore, chapter 3 also describes the role of DIAPl as a caspase inhibitor that is required for cell survival even in the absence of RPR, HID, and GRIM. 25 References Alnemri, E. S., Livingston, D. J., Nicholson, D. W., Salvesen, G., Thornberry, N. A., Wong, W.W., and Yuan, J. (1996). Human ICE/CED3 protease nomenclature. Cell 87, 171. Ambrosini, G., Adida, C., and Altieri, D. C. (1997). A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nature Medicine 3, 917-921. Armstrong, R. C., Aja, T., Xiang, J., Gaur, S., Krebs, J. F., Hoang, K., Bai, X., Korsmeyer, S. J., Karanewsky, D. S., Fritz, L. C., and Tomaselli, K. J. (1996). Fasinduced activation of the cell death-related protease CPP32 Is inhibited by Bcl-2 and by ICE family protease inhibitors. J. Biol. Chem. 271, 16850-16855. Ashkenazi, A., and Dixit, V. M. ( 1998). Death Receptors: Signaling and Modulation. Science 28 I, 1305-1308. Brancolini, C., Lazarevic, D., Rodriguez, J., and Schneider, C. (1997). Dismantling cellcell contacts during apoptosis is coupled to a caspase-dependent proteolytic cleavage of beta-catenin. J. Cell Biol. 139, 759-771. Brunet, A., Bonni, A., Zigmond, M., Lin, M., Juo, P., Hu, L., Anderson, M. , Arden, K. , Blenis, J., and Greenberg, M. (1999). Akt promotes cell survival by phosphorylating and inhibiting a forkhead transcription factor. Cell 96, 857-868. 26 Buckley, C., Pilling, D., Henriquez, N., Parsonage, G., Threlfall , K., Scheel-Toellner, D., Simmons, D., Albar, A., Lord, J., and Salmon, M. (1999). RGD peptides induce apoptosis by direct caspase-3 activation. Nature 397, 534-539. Bump, N. J., Hackett, M., Hugunin, M ., Seshagiri, S., Brady, K., Chen, P., Ferenz, C., Franklin, S., Ghayur, T., Li, P., Licari, P., Mankovich, J., Shi, L. F., Greenberg, A.H., Miller, L. K., and Wong, W.W. (1995). Inhibition of ICE family proteases by baculovirus antiapoptotic protein p35. Science 269, 1885-1888. Cardone, M., Roy, N., Stennicke, H., Salvesen, G., Franke, T., Stanbridge, E., Frisch, S., and Reed, J. (1998). Regulation of cell death protease caspase-9 by phosphorylation. Science 282, 1318-1321. Cecconi, F., Alvarez-Bolado, G., Meyer, B., Roth, K., and Gruss, P. (1998). Apafl (CED-4 homolog) regulates programmed cell death in mammalian development. Cell 94, 727-737. Chen, P., Rodriguez, A., Erskine, R., Thach, T., and Abrams, J. M. (1998). Dredd, a novel effector of the apoptosis activators Reaper, Grim, and Hid in Drosophila. Dev. Biol. 201, 202-216. Cheng, E. H.-Y., Kirsch, D. G., Clem, R. J., Ravi, R., Kastan, M. B., Bedi, A., Ueno, K., and Hardwick, J.M. (1997). Conversion of Bcl-2 to a Bax-like Death Effector by Caspases. Science 278, 1966-1968. 27 Chinnaiyan, A., O'Rourke, K. , Tewari, M. , and Dixit, V. (1995). FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81, 505-512. Chow, S. C., Weis, M., Kass, G. E. N., Holstrom, T. H., Eriksson , J. E., and Orrenius, S. ( 1995). Involvement of multiple proteases during Fas-mediated apoptosis in T lymphocytes . FEBS Lett. 364, 134-138. Clem, R. J., and Miller, L. K. ( 1994). Control of programmed cell death by the baculovirus genes p35 and IAP. Mol. Cell. Biol. 14, 5212-22. Cohen, G. M. (1997). Caspases: the executioners of apoptosis. Biochem. J. 326, 1-16. Conradt, B., and Horvitz, H. R. (1998). The C. elegans protein EGL-I is required for programmed cell death and interacts with the Bcl-2-like protein CED-9. Cell 93, 519-529. Cryns, V. , and Yuan , J. (1998). Proteases to die for. Genes Dev. 12, 1551-1570. Datta, S., Dudek, H., Tao, X. , Masters, S. , Fu, H. , Gotoh , Y., and Greenberg, M. (1997). Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91 , 231 -241. Davie, E.W., Fujikawa, K. , and Kisiel, W. (1991 ). The coagulation cascade: initiation, maintenance, and regulation. Biochemistry 30, 10363-10370. 28 Deveraux, Q., Roy, N. , Stennicke, H., Van Arsdale, T., Zhou , Q., Srinivasula, S. , Alnemri, E., Salvesen, G., and Reed, J. (1998) . IAPs block apoptotic events induced by caspase-8 and cytochrome c by direct inhibition of distinct caspases. EMBO J. 17, 22152223. Deveraux, Q. L., Leo, E., Stennicke, H. R., Welsh, K., Salvesen, G. S., and Reed, J. C. ( 1999). Cleavage of human inhibitor of apoptosis protein XIAP results in fragments with distinct specificities for caspases. EMBO J. 18, 5242-5251. Deveraux, Q. L., Takahashi , R. , Salvesen, G. S., and Reed, J.C. (1997). X-lined IAP is a direct inhibitor of cell-death proteases. Nature 388, 300-304. Dorstyn, L. , Colussi, P. A., Quinn, L. M., Richardson , H ., and Kumar, S. (1999) . DRONC, an ecdysone-inducible Drosophila caspase. Proc. Natl. Acad. Sci . USA 96, 4307-4312. Downward, J. (1998). Mechanisms and consequences of activation of protein kinase B/Akt. Current Opinion in Cell Biology 10, 262-267. Du, Q., Lehavi, D., Faktor, 0., Qi , Y. , and Chejanovsky, N. (1999) . Isolation of an apoptosis suppressor gene of the Spodoptera littoralis nucleopolyhedrovirus. J. Virol. 73, 1278-1285. Duckett, C. S., Nava, V. E., Gedrich, R. W., Clem, R. J., Vandongen, J. L. , Gilfillan, M. C., Shiels, H., Hardwick, J. M., and Thompson, C. B. (1996). A conserved family 29 of cellular genes related to the baculovirus iap gene and encoding apoptosis inhibitors . EMBO J. 15, 2685-2694. Eberstadt, M., Huang, B., Chen, Z., Meadows, R., Ng, S., Zeng, L., Lenardo, M., and Fesik, S. (1998). NMR structure and mutagenesis of the FADD (Mortl) death- effector domain. Nature 392, 941-945. Ellis, R. E., Yuan, J. Y., and Horvitz, H. R. (1991 ). Mechanisms and functions of cell death. Annu . Rev . Cell Biol. 7, 663-698. Fearnhead, H. , Rodriguez, J., Govek, E. , Guo, W., Kobayashi , R. , Hannon , G., and Lazebnik, Y. (1998). Oncogene-dependent apoptosis is mediated by caspase-9. Proc. Natl. Acad. Sci. USA 95, 13664-13669. Flygare, J., Armstrong, R. C., Wennborg, A., Orsan, S., and Hellgren, D. (1998). Proteolytic cleavage of HsRad5 l during apoptosis. FEBS Lett. 427. Fraser, A. G., and Evan, G. I. ( 1997). Identification of a Drosophila melanogaster ICE/CED-3-related protease, drICE. EMBO J. 16, 2805-2813 . Gagliardini, V., Fernandez, P., Lee, R. K. K., Drexler, H. C. A., and Rotello, R. J. (1994). Prevention of vertebrate neuronal death by the crmA gene. Science 263, 826-828. Goltsev, Y. V. , Kovalenko, A. V., Arnold, E., Varfolomeev, E. E., Brodianskii, V. M., and Wallach, D. (1997). CASH, a novel caspase homologue with death effector domains. J. Biol. Chem. 272 , 19641-19644. 30 Greenberg, A.H. (1996). Activation of apoptosis pathways by granzyme B. Cell Death Differ. 3, 269-274. Gross, A., McDonnell, J., and Korsmeyer, S. (1999). BCL-2 family members and the mitochondria in apoptosis. Genes Dev. 13, 1899-1911. Hakem, R. , Hakem, A., Duncan, G., Henderson, J. , Woo, M. , Soengas, M., Elia, A., de la Pompa, J., Kagi, D., Khoo, W. , Potter, J., Yoshida, R., Kaufman, S., Lowe, S., Penninger, J., and Mak, T. (1998). Differential requirement for Caspase 9 in apoptotic pathways in vivo. Cell 94, 339-352. Harvey, A. J., Bidwai, A. P., and Miller, L. K. (1997). Doom, a product of the drosophila mod(mdg4) gene, induces apoptosis and binds to baculovirus inhibitor-ofapoptosis proteins. Mol. Cell. Biol. l 7, 2835-2843. Hawkins, C. J., Wang, S. L., and Hay, B. A. (1999). A cloning method to identify caspases and their regulators in yeast: Identification of Drosophila IAPl as an inhibitor of the Drosophila caspase DCP-1 . Proc. Natl. Acad. Sci. USA 96, 2885-2890. Hay, B. A., and Rubin, G. M. (1994). Expression of baculovirus P35 prevents cell death in Drosophila. Development 120, 2121-2129. Hay, B. A., Wassarman, D. A., and Rubin , G. M. (1995). Drosophila homologs of baculovirus inhibitor of apoptosis proteins function to block cell death . Cell 83, 12531262. 31 Hengartner, M. D. (1997). CED-4 is a stranger no more. Nature 388, 714-715. Hengartner, M. 0 ., and Horvitz, H. R. ( 1994). C.elegans cell survival gene ced-9 encodes a functional homolog of the mammalian proto-oncogene bcl-2. Cell 76, 665-676. Hu, S., Vincenz, C., Ni, J., Gentz, R., and Dixit, V. M. (1997). I-FLICE, a novel inhibitor of tumor necrosis factor receptor 1- and CD95-induced apoptosis. J. Biol. Chem. 272, 17255-17258. Huang , B., Eberstadt, M., Olejniczak, E ., Meadows, R. , and Fesik, S. (1996). NMR structure and mutagenesis of the Fas (APO-1/CD95) death domain. Nature 384, 638-641 . Imai, Y. , Kimura, T., Murakami, A., Yajima, N., Sakamaki, K. , and Yonehara, S. (1999). The CED-4-homologous protein FLASH is involved in Fas-mediated activation of caspase-8 during apoptosis. Nature 398, 777-785. Irmler, M., Thome, M., Hahne, M., Schneider, P., Hofmann, B., Steiner, V., Bodmer, J., Schroter, M., Burns, K., Mattmann, C. , Rimoldi, D., French, L., and Tschopp, J. ( 1997). Inhibition of death receptor signals by cellular FLIP. Nature 388, _190-195. Itoh, N., Yonehara, S., Ishii, A., Yonehara, M., Mizushima, S., Sameshima, M., Hase, A., Seto, Y., and Nagata, S. (1991). The polypeptide encoded by the cDNA for human cell-surface antigen Fas can mediate apoptosis. Cell 66, 233-243. 32 Jacobson, M. D., Weil, M ., and Raff, M. C. (1997). Programmed cell death in animal development. Cell 88, 347-354. Kelliher, M. A., Grimm, S., Ishida, Y., Kuo, F., Stanger, B. Z., and Leder, P. (1998). The death domain kinase RIP mediates theTNF-induced NF-kappaB signal. Immunity 8, 297-303. Kerr, J. F. R. , Wyllie, A. H., and Currie, A. R. (1972). Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics . Br. J. Cancer 26, 239257. Kops, G., de Ruiter, N., De Vries-Smits , A., Powell, D ., Bos, J., and Burgering, B. (1999). Direct control of the Forkhead transcription factor AFX by protein kinase B. Nature 398, 630-634. Koseki, T., Inohara, N., Chen, S., and Nunez, G. (1998) . ARC, an inhibitor of apoptosis expressed in skeletal muscle and heart that interacts selectively with caspases. Proc. Natl. Acad. Sci. USA 95, 5156-5160. Kothakota, S., Azuma, T., Reinhard, C., Klippel, A., Tang, J., Chu, -K., McGarry, T. , Kirschner, M., Koths, K., Kwiatkowski , D., and Williams , L. (1997). Caspase-3generated fragment of gelsolin: Effector of morphological change in apoptosis. Science 278, 294-298 . Krajewski, S., Krajewska, M., Ellerby, L., Welsh, K., Xie, Z., Deveraux, Q., Salvesen, G. , Bredesen, D., Rosenthal , R., Fiskum, G. , and Reed, J. (1999). Release of caspase-9 33 from mitochondria during neuronal apoptosis and cerebral ischemia. Proc. Natl. Acad. Sci . USA 96, 5752-5757. Kuida, K., Lippke, J. A. , Ku, G ., Harding, M . W., Livingston , D . J., Su, M. S. , and Flavell, R. A. (1995). Altered cytokine export and apoptosis in mice deficient in interleukin-I beta converting enzyme. Science 267, 2000-3 . Li , H., Zhu, H., Xu , C. , and Yuan , J. (1998). Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94, 491-501. Li, P., Nijhawan, D., Budihardjo, I. , Srinivasula, S. M., Ahmad, M., Alnemri , E. S., and Wang, X. (1997). Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 479-489. Liu , X., Zou, H., Slaughter, C. , and Wang, X. (1997). DFF, a heterodimeric protein that functions downstream of caspase-3 to trigger DNA fragmentation during apoptosis. Cell 89, 175-184. Lorick, K. L., Jensen, J. P., Fang, S., Ong, A. M., Hatakeyama, S., and Weissman, A. M. ( 1999). RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc. Natl. Acad. Sci . USA 96, 11364-11369. Los , M ., Wesselborg, S. , and Schulze-Osthoff, K . (1999). The Role of Caspases in Development, Immunity, and Apoptotic Signal Transduction: Lessons from Knockout Mice. Immunity JO. 34 Miller, L. K. (1999). An exegesis ofIAPs: salvation and surprises from BIR motifs. Trends in Cell Biology 9, 323-328. Mittl, P. R., Di Marco, S., Krebs, J. F., Bai, X., Karanewsky, D. S., Priestle, J. P., Tomaselli, K. J., and Grutter, M. G. ( 1997). Structure of recombinant human CPP32 in complex with the tetrapeptide acetyl-Asp-Val-Ala-Asp fluoromethyl ketone. J. Biol. Chem. 272, 6539-6547. Muzio, M., Stockwell, B. R., Stennicke, H. R., Salvesen, G. S., and Dixit, V. M. (1998). An induced proximity model for Caspase-8 activation. J. Biol. Chem. 273, 29262930. Nicholson, D. W., and Thornberry, N. A. (1997). Caspases: killer proteases. T.I.B.S. 22. Orth, K., Chinnaiyan, A. M., Garg, M., Froelich, C. J., and Dixit, V. M. (1996). The ced-3/ice-like protease mch2 is activated during apoptosis and cleaves the death substrate Jamin a. J. Biol. Chem. 271, 16443-16446. Orth, K., O'Rourke, K., Salvesen, G., and Dixit, V. (1996). Molecular ordering of apoptotic mammalian CED-S/ICE-like proteases. J. Biol. Chem. 271, 20977-20980. Pitti, R., Marsters, S., Lawrence, D., Roy, M., Kischkel, F., Dowd, P., Huang, A., Donahue, C., Sherwood, S., Baldwin, D., Godowski, P., Wood, W., Gurney, A., Hillan, K., Cohen, R., Goddard, A., Botstein, D., and Ashkenazi, A. (1998). Genomic 35 amplification of a decoy receptor for Fas ligand in lung and colon cancer. Nature 396, 699-703. Pitti, R. M., Marsters, S. A., Ruppert, S., Donahue, C. J. , Moore, A., and Ashkenazi , A. ( 1996). Induction of apoptosis by Apo-2 Ligand, a new member of the tumor necrosis factor receptor family. J. Biol. Chem. 271, 12687-12690. Ray, C. A., Black, R. A. , Kronheim, S. R., Greenstreet, T. A., and Sleath, P.R. ( 1992). Viral inhibition of inflammation: cowpox virus encodes an inhibitor of the interleukin-I~ converting enzyme. Cell 69, 597-604. Rheaume, E., Cohen, L. Y., Uhlmann, F., Lazure, C., Alam, A., Hurwitz, J., Sekaly, R. P., and Denis, F . ( 1997). The large subunit of replication factor C is a substrate for caspase-3 in vitro and is cleaved by a caspase-3-like protease during Fas-mediated apoptosis . EMBO J 16, 6346-6354. Rodriguez, A., Oliver, H., Zou, H., Chen, P., Wang, X., and Abrams, J. M. (1999). Dark is a Drosophila homologue of Apaf-1/CED-4 and functions in an evolutionarily conserved death pathway. Nature Cell Biology 1, 272-279. Rothe, M., Pan, M. G. , Henzel, W. J., Ayres , T. M., and Goedde], D. V. (1995). The TNFR2-TRAF signaling complex contains two novel proteins related to baculoviralinhibitor of apoptosis proteins. Cell 83, 1243-1252. Rotonda, J. , Nicholson, D. W., Fazil, K. M., Gallant, M., Gareau, Y., Labelle, M ., Peterson, E. P., Rasper, D . M., Ruel, R., Vaillancourt, J.P. , et al. (1996). The three- 36 dimensional structure of apopain/CPP32, a key mediator of apoptosis. Nat. Struct. Biol. 3, 619-625. Roy, N., Deveraux, Q. L., Takahashi, R., Salvesen, G. S., and Reed, J. C. (1997). The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16, 6914-6925. Sakahira, H., Enari, M., and Nagata, S. ( 1998). Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis. Nature 391, 96-99. Salvesen, G. S., and Dixit, V. M. (1997). Caspases: intracellular signaling by proteolysis. Cell 91, 443-446. Schulze-Ostoff, K., Krammer, P.H., and Droge, W. (1994). Divergent signalling via APO 1/Fas and the TNF receptor, two homologous molecules involved in physiological cell death. EMBO J. 13, 4587-4596. Schwab, B. L., Leist, M., Knippers, R., and Nicotera, P. (1998). Selective proteolysis of the nuclear replication factor MCM3 in apoptosis. Exp. Cell Res. 238, 415-421. Shaham, S., and Horvitz, H. R. (1996). Developing Caenorhabditis elegans neurons may contain both cell-death protective and killer activities. Genes Dev. JO, 578-91. Shimizu, S., Narita, M., and Tsujimoto, Y. (1999). Bcl-2 family proteins regulate the release of apoptogenic cytochrome c by the mitochondrial channel VDAC. Nature 399, 483-487. 37 Shu, H.-B., Halpin, D.R. , and Goeddel, D. V. (1997). Casper is a FADD- and caspaserelated inducer of apoptosis . Immunity 6, 715-763. Soengas, M., Alarcon, R., Yoshida, H., Giaccia, A., Hakem, R., Mak, T., and Lowe, S. (1999). Apaf-1 and caspase-9 in p53-dependent apoptosis and tumor inhibition. Science 284, 156-159. Song, Z., McCall, K., and Steller, H. (1997). DCP-1, a Drosophila cell death protease essential for development. Science 275, 536-540. Steller, H. (1995). Mechanisms and genes of cellular suicide. Science 267, 1445-1449. Stennicke, H. R., Deveraux, Q. L., Humke, E.W., Reed, J.C., Dixit, V. M. , and Salvesen, G. S. (1999). Caspase-9 can be activated without proteolytic processing. J. Biol. Chem. 274, 8359-8362. Susin, S., Lorenzo, H., Zamzami, N., Marzo, I., Brenner, C., Larochette, N., Prevost, M., Alzari, P., and Kroemer, G. (1999). Mitochondrial release of caspase-2 and -9 during the apoptotic process. Journal of Exp. Med. 189, 381-393. Takahashi, A., Alnemri , E. S., Lazebnik, Y. A., Fernandesalnemri, T. , Litwack, G., Moir, R. D., Goldman, R. D., Poirier, G. G. , Kaufmann , S. H., and Earnshaw, W. C. ( 1996). Cleavage of Jamin a by mch2-alpha but not CPP32 - multiple interleukin 1-betaconverting enzyme-related proteases with distinct substrate recognition properties are active in apoptosis. Proc. Natl. Acad. Sci. USA 93, 8395-8400. 38 Takahashi, R., Deveraux, Q., Tamm, I., Welsh, K., Assa-Munt, N., Salvesen, G.S., and Reed, J.C. ( 1998). A single BIR domain of XIAP sufficient for inhibiting caspases. J. Biol. Chem. 273, 7787-7790. Tewari, M., Telford, W. G., Miller, R. A., and Dixit, V. M. (1995). CrmA, a poxvirusencoded serpin, inhibits cytotoxic T-lymphocyte-mediated apoptosis. J. Biol. Chem. 270, 22705-22708. Thornberry, N. A., Bull, H. G., Calaycay, J. R., Chapman, K. T., Howard, A. D., Kostura, M. J., Miller, D. K., Molineaux, S. M., et al. (1992). A novel heterodimeric cysteine protease is required for interleukin- I beta processing in monocytes. Nature 356, 768-774. Thornberry, N. A., and Lazebnik, Y. (1998). Caspases: enemies within . Science 281, 1312-1316. Ting, A. T., Pimentel-Muinos, F.X., and Seed, B. (1996). RIP mediates tumor necrosis factor receptor I activation of NF-kappaB but not Fas/APO- I-initiated apoptosis. EMBO J I 5, 6189-6196. Ubeda, M., and Habener, J. F. (1997) . The large subunit of the DNA replication complex C (DSEB/RF-Cl40) cleaved and inactivated by caspase-3 (CPP32/Y AMA) during Fasinduced apoptosis. J. Biol. Chem. 272, 19562-19568. 39 Ullrich, A., and Schlessinger, J. (1990). Signal transduction by receptors with tyrosine kinase activity. Cell 61, 203-212. Vucic, D., Kaiser, W. J., Harvey, A. J., and Miller, L. K. (1997). Inhibition of Reaperinduced apoptosis by interaction with inhibitor of apoptosis proteins (IAPs). Proc. Natl. Acad. Sci. USA 94, 10183-10188. Walker, N. P., Talanian, R. V., Brady, K. D., Dang, L. C., Bump, N. J., Ferenz, C. R., Franklin, S., Ghayur, T., Hackett, M. C., Hammill, L. D., et al. (1994). Crystal structure of the cysteine protease interleukin- I~ -converting enzyme: a (p20/p 10)2 homodimer. Cell 78, 343-352. Wang, S., Hawkins, C., Yoo, S., Muller, H., and Hay, B. (1999). The Drosophila caspase inhibitor DIAPI is essential for cell survival and is negatively regulated by HID. Cell 98, 453-463. Wen, L. P., Fahrni, J. A., Troie, S., Guan, J. L., Orth, K., and Rosen, G. D . (1997). Cleavage of focal adhesion kinase by caspases during apoptosis. J. Biol. Chem. 272, 26056-26061. Wilson, K. P., Black, J. A., Thomson, J. A., Kim, E. E., Griffith, J. D., Navia, M. A., Murcko, M. A., Chambers, S. P., Aldapo, R. A., Raybuck, S. A., et al. (1994). Structure and mechanism of interleukin-I~ converting enzyme. Nature 370, 270-275. 40 Wolf, B. B., and Green, D.R. (1999). Suicidal Tendencies: Apoptotic Cell Death by Caspase Family Proteinases. J. Biol. Chem. 274, 20049-20052. Xue, D., and Horvitz, H. R. (1995). Inhibition of the caenorhabditis elegans cell-death protease ced-3 by a ced-3 cleavage site in baculovirus p35 protein. Nature 377, 248-251. Yamin, T., Ayala, J., and Miller, D. (1996). Activation of the native 45-kDa precursor form of interleukin- I-converting enzyme. J. Biol. Chem. 271, 13273-13282. Yeh, W. C., Pompa, J. L., McCurrach, M. E., Shu, H.B., Elia, A. J., Shahinian, A., Ng, M., Wakeham, A., Khoo, W., Mitchell, K., et al. (1998). FADD: essential for embryo development and signaling from some, but not all, inducers of apoptosis. Cell 279, 1954-1958. Yoshida, H., Kong, Y. Y., Yoshida, R., Elia, A. J., Hakem, A., Hakem, R., Penninger, J. M., and Mak, T. W. (1998). Apafl is required for mitochondrial pathways of apoptosis and brain development. Cell 94, 739-750. Yuan, J., and Horvitz, H. R. ( 1992). The Caenorhabditis elegans cell death gene ced-4 encodes a novel protein and is expressed during the period of extensive programmed cell death . Development 116, 309-320. Yuan, J. Y., Shaham, S., Ledoux, S., Ellis, H. M., and Horvitz, H. R. (1993). The C. elegans cell-death gene ced-3 encodes a protein similar to mammalian interleukin-lBeta converting enzyme. Cell 75, 641-652. 41 Zou, H., Henzel, W. J., Liu, X., Lutschg, A., and Wang, X. (1997). Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell 88. 42 CHAPTER 2 A Cloning Method to Identify Caspases and Their Regulators in Yeast: Identification of Drosophila IAP 1 as an Inhibitor of the Drosophila Caspase DCP-1 Christine J. Hawkins*, Susan L. Wang* and Bruce A. Hay *These two authors contributed equally to this work. Published in PNAS 96, 2885-2890, March 1999 43 Abstract Site-specific proteases play critical roles in regulating many cellular processes. To identify novel site-specific proteases, their regulators and substrates, we have designed a general reporter system in Saccharomyces cerevisiae, in which a transcription factor is linked to the intracellular domain of a transmembrane protein by protease cleavage sites. Here we explore the efficacy of this approach using caspases, a family of aspartate-specific cysteine proteases, as a model. Introduction of an active caspase into cells that express a caspasecleavable reporter results in the release of the transcription factor from the membrane, and subsequent activation of a nuclear reporter. We show that known caspases activate the reporter, that an activator of caspase activity stimulates reporter activation in the presence of an otherwise inactive caspase, and that caspase inhibitors suppress caspase-dependent reporter activity. We also find that whereas low or moderate levels of active caspase expression do not compromise yeast cell growth, higher level expression leads to lethality. We have exploited this observation to isolate clones from a Drosophila embryo cDNA library that block DCP-1 caspase-dependent yeast cell death. Among these clones we identified the known cell death inhibitor DIAPl. We showed, using bacterially synthesized proteins, that GST-DIAPl directly inhibits DCP-1 caspase activity, but that it had minimal effect on the activity of a second Drosophila caspase, drICE. 44 Introduction Site-specific proteolysis plays a critical role in regulating a number of cellular processes. An important class of site-specific proteases are a group of cysteine proteases known as caspases (Alnemri et al., 1996). Extensive genetic and biochemical evidence indicates that caspases play roles as cell death signaling and effector molecules in a number of different contexts, thus making them attractive potential therapeutic targets. Caspases identified to date have been found primarily based on homology to the C. elegans caspase CED-3 and mammalian caspase 1, and through biochemical purification (reviewed in Nicholson and Thornberry, 1997; Salvesen and Dixit, 1997; Thornberry and Lazebnik, 1998). Viral and cellular activators and inhibitors of caspase function have also been identified in genetic and biochemical screens for regulators of apoptosis (Teodoro and Branton, 1997; Villa et al., 1997). These approaches to isolating caspases and their regulators are limited by the fact that some proteases that cleave a caspase target site and their regulators may not share primary sequence homology with the proteins identified to date, or they may be expressed only in specific tissues with limited availability for biochemical purification. Furthermore, it is clear that caspases regulate processes other than cell death, including cytokine secretion in mammals (Cerretti et al., 1992; Thornberry et al., 1992; Ghayur et al. , 1997; Zhang et al., 1998; Wang et al., 1998) and cell proliferation and oogenesis in Drosophila (Song et al., 1997; McCall and Steller, 1998). It seems likely, given the early stage of the field, that more roles exist. Caspases and caspase regulators involved in these processes may be missed in screens that focus strictly on cell death related phenotypes. Thus, molecules that possess caspase or caspase regulatory activity may not have been identified yet. As an alternative approach to identifying novel caspases or caspase regulators it would be useful to have assays for caspase function that are based strictly on protease activity. 45 Because of the importance of site-specific proteolysis, we sought to develop a versatile system that would allow the identification of novel site-specific proteases, regulators of the activity of known site-specific proteases, or their substrates. Because caspase cleavage sites have been well defined, and activators and inhibitors of caspases have been identified, we set out to establish a prototype system that would allow positive selection for caspase-like proteases, their activators, and their inhibitors. Our approach to identifying these molecules employs reporters for caspase activity that function in living cells. Yeast, though eukaryotic, lacks many of the specialized proteolytic systems found in cells of higher eukaryotes. Thus it constitutes an ideal background in which to carry out function-based screens for these proteases, their regulators and their targets. Reporters for the activity of specific proteases in bacteria and eukaryotes have been developed using several strategies that involve cleavage-dependent alterations in the activity of specific proteins (McCall et al., 1994; Xu et al., 1998; Sices and Kristie, 1998; Struhl and Adachi, 1998; Lecourtois and Schweisguth, 1998; Stagljar et al., 1998). To visualize caspase activity , we created a fusion protein in which a transcription factor is linked to the intracellular domain of a transmembrane protein by caspase cleavage sites. Expression of this protein in yeast, in the presence of an active caspase, should result in release of the transcription factor from the membrane, followed by transcriptional activation of a reporter. As described below, using such a reporter system, we can visualize caspase activity in yeast, and identify proteins that act as caspase activators and inhibitors. Caspase inhibitors can also be identified by virtue of their ability to suppress caspase overexpression-dependent yeast cell death. Materials and Methods 46 Constructs Yeast strains. The W303a strain (MATa, canl-100, leu2-3,-l 12, his3-l l,-15, trpl-1 , ura3-J, ade2-I) was used to monitor caspase activity using the lacZ reporter system. EGY48 (MATa, ura3, trpl, his3, LexAop 6 -LEU2) (Invitrogen) was used to monitor caspase-dependent cell killing. Construction of caspase target site fusion proteins. The reporter, CLBDG6, was generated using PCR and standard techniques (details provided on request). This protein consists of, from N- to C-terminus, amino acids 1-401 of a type 1 transmembrane protein, human CD4 (Maddon et al., 1985), a linker consisting of 6 tetrapeptide caspase target sites that bracket the specificity of known caspases and granzyme B (Thornberry et al., 1997) - DEVDG-WEHDG-IEHDG-IETDG-DEHDG-DQMDG -, each of which is followed by a glycine residue, which acts as a stabilizing residue in the N-end rule degradation pathway in yeast (reviewed in Varshavsky, 1996), and finally, a transcription factor containing the LexA DNA binding domain (Horii et al., 1981 ). A second construct, designated CLBGG6, was generated that encodes a protein identical to CLBDG6 except that the essential Pl aspartates of the six caspase cleavage sites are replaced with glycines, rendering them nonfunctional. Construction of yeast expression plasmids. Plasmids for expression of genes in yeast were derived from the pRS series (Sikorski and Hieter, 1989). To express genes in yeast under galactose-inducible control we utilized two GALI promoter fragments: a long version, extending from base 1-815, called GALL, and a shorter and somewhat weaker version, extending from base 406-815, called GALS (Johnston and Davis, 1984). The yeast actin terminator, bases 2107-2490 (Genbank accession L00026), was used for all constructs. GALL promoter and actin terminator fragments were inserted into pRS313 (H!S3), pRS314 (TRP I) and pRS3 l 5 (LEU2), generating pGALL-(H/S3), pGALL- 47 (TRP I) and pGALL-(LEU2). CLBDG6 and CLBGG6 coding regions were cloned into pGALL-(TRP ]). GALS promoter and actin terminator fragments were inse1ted into pRS315, generating pGALS (LEU2). To express genes under the control of the copper inducible CUPl promoter, we utilized a promoter fragment extending from base 10791533 of the CUPl locus (Genbank accession K02204). Site-directed mutagenesis was used to mutate the CUP 1 promoter to prevent activation in response to glucose starvation . The mutated CUP 1 promoter and actin terminator fragments were inserted into pRS3 l 5 (LEU2), generating pCUP1-(LEU2) . The coding region for the C. elegans caspase CED-3 (Miura et al. , 1993) was introduced into pCUP 1-(LEU2) . Site-directed mutagenesis was used to generate an inactive version of CED-3 in which the active site cysteine was changed to serine (CED3CS). Full length human caspase 7 (caspase 7FL) (Duan et al., 1996a), caspase 7 lacking the N-terminal 53 amino acid prodomain (caspase i\ and the caspase 8 isoform corresponding to MACHa2 (Boldin et al., 1996) or Mch5 (Fernandes-Alnernri et al., 1996) (caspase 8FL), were introduced into pCUP1-(LEU2). Full length DCPl (Song et al. , 1997) was introduced into pGALS-(LEU2) and pGALL-(LEU2). Site-directed mutagenesis was used to change the active site cysteine to serine (DCP-1 CS). The resulting full length coding region was introduced into pGALL-(LEU2) . Full length drICE (Fraser and Evan, 1997) was inserted into pGALL-(LEU2). Full length human caspase 9 (Duan et al., 1996b; Srinivasula et al., 1996) was introduced into pGALL-(LEU2). The region encoding amino acids 1-530 of Apaf-1 (Zou et al. , 1997) (Apaf-1 530 ) was introduced into pGALL-(H/S3). Full length DIAPl (FLDIAPl) and a version of DIAPl C-terminally truncated following residue 381 (DIAPlBIR) (Hay et al. , 1995) were introduced into pGALL- 48 (HIS3). The mouse IAP MIHA (Uren et al., 1996) and baculovirus p35 (Clem et al., 1991) were introduced into pGALL-(H/SJ). Yeast transformation and characterization Plasmids were introduced into yeast by lithium acetate transformation. For caspase activity assays in which lacZ expression was monitored, yeast were transformed with pSH18-34 (URA3) (Invitrogen), which carries a LexA-responsive lacZ gene. These cells were then transformed with either pGALL-CLBDG6-(TRP]) or pGALL-CLBGG6-(TRP ]). Caspase expression plasmids or an empty expression vector were introduced into these backgrounds and characterized as described below. For caspase activation and inhibition assays, a fourth plasmid was also introduced, either expressing Apaf-1 530 , p35, MIHA, or with no insert. To carry out X-gal filter assays for P-galactosidase (P-gal) activity, transformants were plated on selective plates with glucose (2%) as the sugar source. After three days, duplicate colonies were picked and resuspended in 1ml of sterile Tris/EDTA, pH 8.0. One µl of each sample was streaked on a minimal medium glucose plate. After two days, a nylon membrane was used to lift the streaked yeast (yeast side upwards) onto a complete medium plate containing 2% galactose and 1% raffinose (gal/raf media) and 3µM copper sulfate. After various periods of induction the filters were processed for Xgal staining (Breeden and Nasmyth, 1985). To quantitate p-gal activity, three tubes of liquid selective gal/raf medium were inoculated with single colonies from each transformation plate, grown for 24 hours, then diluted I: 10 into fresh gal/raf selective medium containing the indicated concentration of copper sulfate, and grown for a further IO hours. ONPG assays were performed as described by Miller (Miller, 1972). To assay 49 caspase-dependent cell death and protection by inhibitors, colonies carrying the relevant plasmids were streaked from 2% glucose selective media plates onto gal/raf selective media plates. The plates were photographed after three days . Expression and purification of recombinant Drosophila IAPs and caspases The DIAPl coding region was amplified by PCR using primers that generated an Nterminal myc epitope (EQKLISEEDL) and introduced into the GST expression vector pGEX4T-1 (Pharmacia). The GST-myc-DIAPl fusion protein was expressed in E.coli strain BL21 (DE3)pLysS (Novagen) and affinity purified on glutathione-Sepharose by standard methods. The eluted protein was dialyzed against buffer A [25 mM Tris (pH 8.0), 50 mM NaCl, 10 mM DTT] . Following dialysis, the protein was frozen in aliquots after addition of glycerol to 10%. DCP-1, initiating at codon 31 (DCP-1 31 ), was introduced into pET23a( +) 31 (Novagen), generating a DCP-1 -His 6 fusion . A similar procedure was used to generate a 81 drICE-His 6 fusion protein, in which DrICE initiates at codon 81 (drICE -His6). DCP31 81 1 -His 6 and drICE -His 6 were expressed in the E. coli strain BL21(DE3)pLysS. Protein expression and affinity purification from the soluble fraction were carried out using ProBond resin (Invitrogen), by standard methods. Eluted protein was dialyzed against buffer A and subsequently snap frozen in buffer A containing 10% glycerol. Drosophila cDNA library construction, and DCP-1 inhibitor screening Drosophila embryonic poly A+ mRNA (Clontech) was converted into cDNA using a Superscript cDNA synthesis kit (GIBCO). cDNAs larger than 600 bp were ligated into a 50 modified version of pGALL-(H/S3) in which the poly linker was expanded to contain Xhol and Notl sites. ElectroMax DH I OB cells (GIB CO) were transformed with the ligation mix and used to amplify the library, which contained 5 X 106 primary transformants. W303a yeast carrying pGALL-DCP-1-(LEU2) were transformed with 26ug of library plasmid DNA, grown in YPD for 3 hrs, washed twice to remove glucose, and plated on gal/raf selective plates. A total of 140,000 transformants were screened. Colonies were picked after 4 days of growth at 30°C. PCR was carried out on DNA isolated from individual colonies using DIAP I-specific primers. In vitro protease assays 1 81 DCP-I 3 -His6 and DrICE -His 6 caspase activity were measured fluorometrically by following the release of 7-amino-4-trifluoromethyl-coumarin (AFC) from Ac-DEVD-AFC (Enzyme Systems Products) using the fmax fluorescence microplate reader (Molecular Devices) with an excitation wavelength of 405 nm and an emission wavelength of 510 nm. The ability of GST-DIAPI to inhibit caspase activity was determined from caspase activity 31 assay progress curves, in which substrate hydrolysis (I 00 µM) by DCP-1 -His 6 (0.2 81 nM) or drICE -His 6 (0.62 nM) was measured in the presence of GST (0.48 µM) or GSTmycDIAPI (0.16 µM), in caspase activity buffer (50 mM HEPES pH 7.5, 100 mM NaCl, I mM EDT A, 0.1 % CHAPS, 10% sucrose and 5 mM dithiothreitol). 51 Results and Discussion Our approach to monitoring caspase activity in vivo was to create cells in which caspase activity stimulates transcriptional activation of a reporter. We created a fusion protein substrate for caspase cleavage in which the transcription factor LexA-B42 (LB) is linked to the truncated cytoplasmic domain of a membrane protein, CD4 (C), by a short linker (DG6) consisting of six different caspase cleavage sites that bracket the specificities of known caspases and the serine protease granzyme B, which cleaves caspases and other targets at sites of similar sequence (Material and Methods for details). When this molecule, referred to as CLBDG6, is expressed in a reporter strain in which a LexA-dependent promoter drives lacZ expression (LexA/~-gal reporter), levels of ~-gal activity should depend on the presence of an active caspase able to cleave one or more of the introduced target sites, thereby releasing LexA-B42 from membrane association (Figure IA, B). A reporter for caspase activity in yeast We introduced CLBDG6 into the LexA/~-gal reporter strain in a plasmid, pGALLCLBDG6, in which expression is induced in response to galactose. We introduced into this background a copper-inducible expression plasmid, pCUPl, containing either no insert or different versions of the caspase CED-3. Transformants were initially streaked on glucose medium. Colonies from these streaks were then replica plated onto gal/raf medium containing 3µM copper to induce expression of CLBDG6, and from the pCUPl plasmid. After 12 hours of induction, levels of ~-gal activity were determined using an X-gal assay in which cells that do not express ~-gal remain white, while those that do turn shades of 52 Figure 1. A genetic system for monitoring caspase activity in yeast using a transcriptional reporter. Yeast were created that express a chimeric type-! transmembrane protein (CLBDG6) in which the N-terminal signal sequence and transmembrane domain (CD4) is followed by a linker consisting of 6 tetrapeptide caspase target sites (indicated in bold) that bracket the specificity of known caspases and granzyme B (Thornberry et al., 1997) - DEVDG-WEHDG-IEHDG-IETDG-DEHDG-DQMDG -, each of which is followed by a glycine residue, which acts as a stabilizing residue in the N-end rule degradation pathway in yeast (Varshavsky, 1996). C-terminal to the caspase target site linker is a transcription factor domain, LexA-B42. The LexA-dependent transcriptional reporter consists of LexA binding sites (LexA UAS) and a promoter (P) upstream of the bacterial lacZ gene (lacZ) (A). The cells in (A) act as caspase activity reporters since expression of an active caspase results in CLBDG6 cleavage at the caspase target sites, releasing LexAB42, which enters the nucleus and activates lacZ transcription (B). A version of CLBDG6 in which the Pl aspartates are changed to glycines (CLBGG6) cannot be cleaved by caspases. Cells expressing CLBGG6 act as false positive reporters for molecules that activate lacZ expression independent of cleavage at caspase target site (C). If cells in (B) express a caspase inhibitor as well as an active caspase, caspase activity, and thus caspasedependent release of LexA-B42 is inhibited, and ~-gal levels are decreased compared to cells that express the caspase alone (D). 53 (') 0 ""' C 54 blue. Reporter cells that expressed CLBDG6 alone remained white in this assay (Figure 2A), indicating that yeast contains negligible amounts of proteases capable of cleaving caspase target sites under standard growth conditions. However, when expression of the C.elegans caspase CED-3 (pCUPI-CED-3) was induced, a high level of ~-gal activity was observed (Figure 2A), which increased in a copper concentration-dependent manner (Figure 2B). Importantly, caspase activity was required for reporter activation, because expression of an inactive CED-3 mutant in which the active site cysteine had been changed to serine (CED-3CS) did not result in ~-gal expression (Figure 2A). Finally, expression of wildtype CED-3 in a reporter strain in which the essential Pl aspartates of the caspase target sites in CLBDG6 had been mutated to glycines (CLBGG6) (Figure 1C), did not result in ~-gal activity (Figure 2A), arguing that the CED-3dependent induction of ~-gal activity was a direct result of cleavage of CLBDG6 at the caspase target sites. These results establish that yeast can be used as a cell-based reporter system for caspase activity. In order for a caspase to be identified in this assay, the caspase must be active in yeast. Physiological activation of caspases occurs through multiple mechanisms, including recruitment and oligomerization at the plasma membrane, cleavage by caspases or other proteases able to recognize a caspase target site, interactions with members of the CED-4/Apaf-1 family of proteins, and autoactivation. In some cases overexpression alone is sufficient to induce autoactivation, while in other cases significant activation requires interactions with other proteins (reviewed in Cryns and Yuan, 1998; Nicholson and Thornberry, 1997; Salvesen and Dixit, 1997). Thus it is likely that only proteases in which the primary translation product is active, or in which the protease is able to autoactivate, will be identified in the simplest reporter-based caspase screen. However, 55 Figure 2. Yeast expressing CLBDG6 act as reporters for CED-3 caspase activity. W303cx yeast were transformed with pSH 18-34, which carries a LexA-responsive lacZ transcriptional cassette (the LexA/~-gal reporter strain). These cells were transformed with pGALL expression plasmids carrying CLBDG6 (DG6) or CLBGG6 (GG6). These cells also carry a copper-inducible pCUPl plasmid, which contains either wildtype CED-3 (CED-3), an inactive C to S mutant version of CED-3 (CED-3 C-S), or nothing. Duplicate colonies from each transformation were streaked onto gal/raf medium to induce GALldependent expression of the caspase substrates, and then lifted onto complete media plates with 3µM copper sulfate to induce caspase expression. After a 12 hour induction an X-gal assay was performed on the filter. Only cells expressing CLBDG6 and wild-type CED-3 have significant ~-gal activity (A). Cultures from three transformants carrying pSHl 8-34, pGAL-CLBDG6 and either the empty pCUPl vector or pCUPl-CED-3 were grown to stationary phase, then diluted into medium containing the indicated levels of copper sulfate and grown for a further 10 hours. ONPG assays were performed and ~-gal activity was determined. ~-gal activity in the CED-3-expressing cells increased as a function of copper concentration (filled circles, CED-3). No ~-gal activity was found in the cultures carrying only the empty pCUPl vector (open boxes, vector) (B). 56 B (/) so :!:: □ vector ::::, 40 • CED-3 C: <1) • • • (/) ro 30 '"C ·- • (/) .S20 u ro ro0) 10 I cc. •• • •• 0 00.3 1.0 [Cu] (µM) 3.0 57 more complex screens for caspases that can activate following forced oligomerization or association with potential caspase activators (Li et al., 1997; Srinivasula et al., 1998; Yang et al., 1998b; Yang et al., 1998a) can be envisioned. We have tested several other caspases in this reporter system. Expression of mammalian caspase 7 53 (below) and full length caspase 8 (data not shown) resulted in reporter-dependent lacZ expression. Expression of human caspase 3, caspase 9 or Drosophila drICE failed to activate reporter expression (data not shown), even though active forms of these caspases are known to efficiently cleave peptides with the same sequence as the target sites introduced into the CLBDG6 (Thornberry, et al., 1997; Fraser and Evan, 1997). Moreover, although overexpression of wild-type, but not an inactive mutant of CED-3 induced yeast cell death (below), similar overexpression of caspase 3, caspase 9 or drICE had no effect on cell growth. Based on these observations, it is likely that in yeast the procaspase forms of these caspases do not autoprocess to generate active caspase heterodimers. This result is expected: caspase 9 is thought to function as an upstream caspase, in which a major mechanism of activation requires association with Apaf-1 (Li et al., 1997; Srinivasula et al., 1998), while caspase-3 is thought to act as a downstream caspase, in which a principal mechanism of activation is cleavage by other caspases (reviewed in Cryns and Yuan, 1998; Nicholson and Thornberry, 1997; Salvesen and Dixit, 1997). drICE activation may be regulated by either of these mechanisms. Activators of caspase-dependent reporter activation The fact that certain caspases do not activate in yeast suggests that it should be possible to screen for their activators as molecules that induce reporter expression in the presence of an otherwise inactive caspase. To demonstrate this we carried out an experiment in which caspase activity was monitored in yeast that expressed full length caspase 9, alone or in 58 combination with a fragment of Apaf-1 that is constitutively active with respect to caspase 9 processing activity in vitro (Srinivasula et al., 1998). Transformants of the pGALLCLBDG6 LexA/~-gal reporter strain were generated that carried either two empty vectors, an empty vector and pGALL-Apaf-1 530 pGALL-caspase 9 and pGALL-Apaf-1 , an empty vector and pGALL-caspase 9, or 530 . Transformants were initially streaked on glucose medium. Colonies from these streaks were then replica plated onto gal/raf medium to induce GALL-dependent expression. After 16 hrs of induction, levels of ~-gal activity were determined using an X-gal assay. Reporter cells that expressed either nothing, caspase 9, or Apaf-1 530 alone remained white in this assay, indicating that caspase activity was not induced (Figure 3). However, colonies that expressed both caspase 9 and Apaf530 1 530 showed robust ~-gal activity, suggesting the occurrence of Apaf- 1 -mediated activation of the otherwise inactive procaspase 9 (Figure 3). We used ~-gal activity, as assayed following replica plating of colonies, as the basis for our caspase reporter assay because caspases are conditionally expressed following replica plating. Thus, their identification is feasible even if their expression is toxic to cells. However, caspase activity screens can also be adapted to positive, survival-based screening assays by requiring LexA-dependent expression of a yeast auxotrophic marker such as HIS3 or URA3. False positives in these caspase reporter assays could arise because introduced proteins bind to the LexA binding sites and activate transcription directly, or because they encode proteases that cleave CLBDG6, but not at the caspase target sites. Both classes of false positives can be identified by the fact that they should still activate lacZ expression when introduced into a LexA/~-gal reporter strain that expresses CLBGG6, the false positive reporter strain (Figure 1C). 59 Figure 3. Expression of Apaf-1 induces caspase 9-dependent reporter activation. W303a containing pSH18-34 and pGALL-CLBDG6 were transformed with pGALL plasmids so as to carry either two empty pGALL vectors, Apaf-1 530 and an empty pGALL vector, caspase 9 and an empty pGALL vector, or Apaf-1 and caspase 9. Two colonies from each transformation were streaked onto selective glucose medium plates, grown for several days, and then replica plated onto complete gal/raf medium. After 16 hours an X-gal assay was performed. Only cells expressing both Apaf-1 and caspase 9 show significant ~-gal activity. 60 + DG6 Apaf-153) Casp-9 - I I II1, I! - + + - + - /' + + + + E, ,l 61 Suppression of caspase-dependent reporter activation by caspase inhibitors Once a reporter for the activity of a specific caspase has been established in yeast, it should be possible to screen for inhibitors of that activity by identifying cells that express the protease, but in which reporter activity is repressed (Figure 1D). Here we demonstrate the feasibility of this approach by showing that two different families of caspase inhibitors, exemplified by baculovirus p35, and the murine IAP MIHA, suppress caspasedependent reporter activation. p35 is a broad specificity caspase inhibitor in which inhibition is associated with cleavage (reviewed in Teodoro and Branton, 1997). The IAPs comprise a second class of caspase inhibitors distinct from p35, in which cleavage of the inhibitor does not play a role. IAPs were originally identified in the baculovirus system by virtue of their ability to substitute for baculovirus p35 as suppressers of viral infectioninduced cell death (Crook et al., 1993). IAP homologs have now been found in other viruses, Drosophila, C. elegans, mammals and yeast, and are characterized by one or more N-terminal repeats known as baculovirus IAP repeats (BIRs). Many also have a Cterminal RING finger motif (Teodoro and Branton, 1997; Villa et al., 1997; Clem et al., 1991; Uren et al., 1998). Several mammalian IAPs: XIAP, cIAPl, and cIAP2, bind to and directly inhibit caspases -3 and -7 in vitro (Deveraux et al., 1997; Roy et al., 1997). We introduced pGALL-CLBDG6 and either an empty pCUPl plasmid or a pCUP plasmid 3 containing CED-3 or caspase ?5 into the LexA/~-gal reporter strain along with either an empty pGALL expression vector, a pGALL expression vector carrying p35, or a pGALL expression vector carrying MIHA (murine XIAP). Transformants were grown in galactose-containing medium to induce expression of CLBDG6 and the caspase inhibitor, and then transfetTed to medium containing galactose and 3µM copper, to induce expression of the caspase. ~-gal activity was determined following a 10 hr copper 62 induction . As shown in Figure 4 (Figure 4A, B), expression of p35 inhibited both CED-3 and caspase i 3 activity roughly fivefold. A similar, though somewhat weaker (approximately twofold) inhibition of caspase 7 53 activity was seen in the presence of MIHA (Figure 4B), indicating that caspase-IAP interactions can be detected in this assay. Caspase overexpression-dependent yeast cell death In the above assay, the presence of a caspase inhibitor is indicated by a decrease in ~-gal activity. However, in many situations it would be useful if caspase inhibition was coupled to a positive reporter output. A direct approach to identifying caspase inhibitors rests on the observation that high level expression of active caspase causes yeast cell death . Low level expression of the Drosophila caspase DCP-1 from the induced CUP-1 promoter did not significantly compromise yeast cell growth. Higher level expression from the GAL promoter, however, did result in cell lethality. Cells were able to grow on galactosecontaining media if they carried an empty pGALL expression vector, but not if they carried a pGALL-DCP-1 expression construct (Figure 5A). This effect of DCP-1 expression was due to cell death since a greater than 250-fold decrease in the number of colony forming units was seen when cells carrying the pGALL-DCP-1 expression plasmid were grown for 12 hours in liquid gal/raf medium, thus inducing high level DCP-1 expression , and then plated on glucose-containing medium (data not shown). Importantly, GALL-DCP-1expression-dependent cell killing depended on caspase activity because expression of an inactive mutant form of DCP-1 (DCP-1 C285S) did not cause cell death (Figure SA). Caspase-mediated cell death in yeast may be a general phenomenon because other caspases that are active in yeast, including full length CED-3, full length caspase 8, and caspase i 3, 63 Figure 4. Expression of caspase inhibitors suppresses caspase-dependent reporter activation in yeast. The LexA/p-gal reporter strain carrying pGALL-CLBDG6 was transformed with either an empty pCUPl plasmid or pCUPl-CED-3, and either an empty pGALL vector or pGALL-p35. Three colonies from each transformation were grown for 24 hours in selective gal/raf medium. Cultures were diluted I: 10 into fresh gal/raf medium containing 3µM copper sulfate, grown for a further 10 hours, after which ONPG assays for P-gal activity were performed. Cultures from caspase transformants showed significant p-gal activity, which was suppressed by GALL-dependent expression of baculovirus p35 (A). In an experiment similar to that described in (A), expression of caspase ?5 was induced in cells that express baculovirus p35 or the mouse IAP MIHA. 3 Expression of caspase ?5 3 resulted in a significant increase in cellular p-gal activity, which was suppressed by GALL-dependent expression of p35 or MIHA (B). 64 A B (/) 25- 7 (/) +J ::, (l) (/) -~ • ·2 6 5 <l> (/) •• n:s 32 4 (/) ~ 15(/) B 10u 83 u ro ~ rnC) 2 I • ~1 + p35 • • • § 20- - + + ro 5- I C> I a.. • + Casp-7 co. p35 MIHA - - - • • • • + + - + + 65 Figure 5. High level expression of the Drosophila caspase DCP-1 kills yeast, and is prevented by coexpression of baculovirus p35 or DIAP 1. EGY48 yeast were transformed with pGALL plasmids containing either full length DCP-1 (FLDCP-1 ), an active site C-S mutant of DCP-1 (DCP-1 C-S), or no insert (vector). Transformants were streaked from selective glucose-containing medium onto selective gal/raf inducing medium. Cells expressing either an empty pGALS vector or the DCP-1 C-S active site mutant grow on galactose-containing medium, while cells expressing full length DCP-1 do not (A). EGY48 yeast carrying pGALS-FLDCP-1 were transformed with pGALL vectors carrying full length DIAPI (DIAPl), the DIAPl BIR repeats ( DIAPl BIR), baculovirus p35 (p35), or nothing (vector). GALL-dependent expression of p35, full length DIAPl, and to a somewhat lesser extent, the DIAPI BIR repeats, block DCP-1-dependent cell death (B). 66 A B 67 block colony formation when expressed under control of the strong GALL promoter (data not shown). Caspase inhibitors suppress caspase overexpression-dependent cell death To demonstrate that caspase inhibitors can be identified as proteins that restore cell viability to yeast expressing an active caspase, we carried out an experiment in which the broad specificity caspase inhibitor p35 was coexpressed with full length DCP-1. We introduced pGALL expression plasmids that either had no insert, or that carried p35, into cells carrying a pGALS-DCP-1 plasmid. Cells from colonies carrying these plasmids were grown on glucose-containing media and then streaked onto gal/raf media to induce expression of the caspase and the potential inhibitor. As shown in Figure SB, pGALSdependent expression of DCP-1 in the presence of an empty pGALL expression vector resulted in no cell growth. In contrast, coexpression of baculovirus p35 with DCP-1 resulted in a dramatic rescue of cell growth. To determine if a yeast survival-based screen can be used to identify novel caspase inhibitors, we transformed yeast carrying pGALS-DCP-1 with a Drosophila pGALL embryonic cDNA expression library and plated these cells on gal/raf medium. In a screen 5 of about 1.4 X 10 transformants, approximately 50 positives were obtained. These were tested by PCR and all found to correspond to DIAP 1, which was originally identified as an inhibitor of reaper or hid overexpression-induced, caspase-dependent cell death in the fly eye (Hay et al., 1995). In the fly eye, and in cell culture (Vucic et al., 1998; Vucic et al., 1997; Harvey et al., 1997; Hawkins et al. , 1998), it was also found that cell death could be suppressed by expression of an N-terminal DIAPl fragment containing the two BIR repeats, but lacking the C-terminal RING finger domain. To determine if the same fragment of DIAP 1 was sufficient to block DCP-1-dependent cell killing in yeast, we 68 carried out an experiment in which full length DIAPl , an N-terminal fragment of DIAPl , or an empty vector, was expressed under GALL control, in the presence of GALS-driven DCP-1. As shown in Figure 5B, inhibition of DCP-1-dependent cell death was seen when DIAPl or the N-terminal fragment of DIAPl containing only the DIAPl BIR repeats was coexpressed with DCP-1. To determine if the observed interaction between DIAPl and DCP-1 was direct, we generated bacterially synthesized GST-DIAPl, as well as His 6 31 tagged versions of prodomainless DCP-1 (DCP-1 -His 6) and drICE (dr1CE 81 -His 6) . As 31 shown in Figure 6, GST-DIAPl inhibited DCP-1 -His 6 caspase activity, but had little, if 81 any, effect on that of drlCE -His 6 . Thus, these results demonstrate that caspase inhibitors can be identified as molecules that block GAL-driven, caspase-dependent cell death. Concluding Remarks We have developed a yeast cell-based assay for the activity of one group of proteases, the caspases, in which caspase activity is monitored either by the cleavage-dependent release of a transcription factor from its transmembrane anchor and subsequent activation of a reporter, or by induction of cell killing. Both reporter activation and cell killing are suppressed by known caspase inhibitors. We have exploited this fact to directly isolate caspase inhibitors from a Drosophila embryo cDNA library. Yeast carrying the transcription-based caspase reporter should be useful as a background in which to carry out screens for proteins that cleave a caspase target site and for their regulators. Because yeast can be transformed with high efficiency, it also constitutes an ideal system in which to carry out large scale mutagenesis studies of particular proteases or their regulators. It may also be possible to screen for cellular targets of specific caspases by using artificial substrate libraries in which cDNA fragments substitute for the caspase target site linker in the caspase substrate fusion protein 69 Figure 6. GST-DIAPl inhibits the caspase activity of bacterially synthesized DCP-1, but not drICE. Purified GST-DIAPl (0. 16µM) (open triangles) or GST (.48µM) (open 1 squares) was incubated with a fixed amount of DCP-l 3 His 6 (0.2nM) in caspase activity assay buffer containing 100µM of the Ac-DEVD-AFC substrate. Release of AFC was monitored fluorometrically over time. GST-DIAP 1 inhibits DCP-1-dependent caspase 81 activity (A) . In similar experiments in which 0.62nM drICE His 6 was incubated in caspase activity buffer with GST (0.48µM), or GST-DIAPl, (0.16µM), no inhibition of drICE activity by GST-DIAPl was seen (B). 70 A DCP-1 B drlCE 0.8 -~ 0.2 0.6 ~ 0.15 ::J.. 3, 0.4 ........... LL <( 0.1 0 0 LL 0 .2 <( 0 0.05 0 0 20 40 60 Time (minutes) 0 20 40 60 Time (minutes) 71 CLBDG6. A transcription-based reporter strategy similar to that described here may also provide a way to monitor caspase activity in cells of higher eukaryotes. Finally, substituting caspase cleavage sites for those of other site-specific proteases in the CLBDG6 reporter should enable the identification and study of these proteins and their regulators. 72 Acknowledgments We thank the following for providing plasmids and strains: J. Adams (CD4), J. Yuan (p~actced38Z), L. K. Miller (pBS p35), H. Steller (DCP-1 ), D . Vaux (MIHA and caspases 3 and 9), X. Wang (Apaf-1) and R . Deshaies (yeast strains and plasmids). R. Deshaies, P. Sternberg, M . Guo and W. Y. Shou provided helpful discussions and comments. C. J. H. is supported by a fellowship from the Human Frontiers Science Program. B . A.His a Searle Scholar. This work was supported by NIH grant GM057422-0l, and a Burroughs Wellcome Fund New Investigator Award in the Pharmacological Sciences to B. A. H. 73 References Alnemri, E., Livingston , D. , Nicholson, D., Salvesen, G., Thornberry , N., Wong, W. and Yuan, J. (1996) . Human ICE/CED-3 protease nomenclature. Cell 87, 171-171. Boldin, M., Goncharov, T., Goltsev, Y. and Wallach, D. (1996) . Involvement of MACH, a novel MORTl/FADD-interacting protease, in Fas/APO-1- and TNF receptorinduced cell death. Cell 85, 803-815 . Breeden, L. and Nasmyth, K. (1985). Regulation of the yeast HO gene. Cold Spring Harbor Symposia on Quantitative Biology 50, 643-650. Cerretti, D ., Kozlosky , C., Mosley , B., Nelson, N., Vanness, K. , Greenstreet, T. , March, C. , Kronheim, S., Druck, T. , Cannizzaro, L. , Huebner, K. and Black, R. ( 1992). Molecular-cloning of the interleukin-I-beta converting enzyme. Science 256, 97100. Clem, R. , Fechheimer, M. and Miller, L. (1991). Prevention of apoptosis by a baculovirus gene during infection of insect cells. Science 254, 1388-1390. Crook, N., Clem, R. and Miller, L. (1993). An apoptosis-inhibiting baculovirus gene with a zinc finger- like motif. Journal of Virology 67, 2168-2174. Cryns, V. and Yuan , J. (1998) . Proteases to die for. Genes Dev. 12, 1551-1570. 74 Deveraux, Q., Takahashi, R. , Salvesen, G. and Reed, J. (1997). X-linked IAP is a direct inhibitor of cell-death proteases. Nature 388, 300-304. Duan, H., Chinnaiyan, A. , Hudson, P., Wing , J. , He, W. and Dixit, V. (1996a). ICELAP3, a novel mammalian homologue of the Caenorhabditis elegans cell death protein ced-3 is activated during fas- and tumor necrosis factor-induced apoptosis. Journal of Biological Chemistry 27 I, 1621-1625. Duan, H ., Orth, K., Chinnaiyan, A., Poirier, G. , Froelich, C., He, W. and Dixit, V. (1996b) . ICE-LAP6, a novel member of the ICE/Ced-3 gene family, is activated by the cytotoxic T cell protease granzyme B. Journal of Biological Chemistry 271 , 16720-16724. Fernandes-Alnemri, T. , Armstrong, R., Krebs, J., Srinivasula, S., W ang, L., Bullrich, F., Fritz, L. , Trapani , J., Tomaselli , K., Litwack, G. and Alnemri , E. (1996). In vitro activation of CPP32 and Mch3 by Mch4, a novel human apoptotic cysteine protease containing two FADD-like domains . Proceedings of the National Academy of Sciences USA 93 , 7464-7469. Fraser, A. and Evan, G. (1997). Identification of a Drosophila melanogaster ICE/CED3related protease, drICE. EMBO J. 16, 2805-2813. Ghayur, T. , Banerjee, S. , Hugunin, M., Butler, D., Herzog, L., Carter, A., Quintal, L., Sekut, L. , Talanian, R., Paskind, M., Wong, W., Kamen, R., Tracey , D. and Allen, H. ( 1997). Caspase-1 processes IFN-gamma-inducing factor and regulates LPS-induced IFN-gamma production. Nature 386, 619-623 . 75 Harvey, A., Soliman, H. , Kaiser, W. and Miller, L. (1997) . Anti-and pro-apoptotic activities of baculovirus and Drosophila IAPs in an insect cell line. Cell Death and Differentiation 4, 733-744. Hawkins, C., Ekert, P. , Uren , A., Holmgreen, S. and Vaux, D. (1998). Anti-apoptotic potential of insect cellular and viral IAPs in mammalian cells. Cell Death and Differentiation 5, 569-576. Hay, B. , Wassarman, D. and Rubin, G. ( 1995). Drosophila homologs of baculovirus inhibitor of apoptosis proteins function to block cell death. Cell 83, 1253-1262. Horii, T., Ogawa, T. and Ogawa, H. (1981). Nucleotide Sequence of the LexA Gene of E. Coli. Cell 23, 689-697. Johnston, M . and Davis, R. (1984). Sequences that regulate the divergent gal 1-gal 10 promoter in saccharomyces-cerevisiae. Molecular and Cellular Biology 4, 1440-1448. Lecourtois, M. and Schweisguth, F. (1998). Indirect evidence for Delta-dependent intracellular processing of notch in Drosophila embryos. Current Biology 8, 771-774. Li , P., Nijhawan, D ., Budihardjo, I., Srinivasula, S., Ahmad, M., Alnemri, E. and Wang, X. (1997). Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 4 79-489. 76 Maddon , P. , Littman, D., Godfrey, M., Maddon , D., Chess, L. and Axel, R. (1985). The isolation and nucleotide-sequence of a cDNA-encoding the T- cell surface protein-T4 a new member of the immunoglobulin gene family. Cell 42, 93-104. McCall, J. , Kadam, S. and Katz, L. (1994) . A high-capacity microbial screen for inhibitors of human rhinovirus protease 3c. Biotechnology 12, 1012-1016. McCall, K. and Steller, H. (1998). Requirement for DCP-1 caspase during Drosophila oogenesis. Science 279, 230-234. Miller, J. Experiments in Molecular Genetics (Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y., 1972). Miura, M., Zhu, H., Rotello, R., Hartwieg, E. and Yuan, J. (1993). Induction of apoptosis in fibroblasts by ii-I-beta-converting enzyme, a mammalian homolog of the celegans cell-death gene ced-3. Cell 75, 653-660. Nicholson, D. and Thornberry, N. (1997). Caspases: killer proteases. Trends in Biochemical Sciences 22, 299-306. Roy, N., Deveraux, Q., Takahashi, R., Salvesen, G. and Reed, J. (1997). The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16, 6914-6925 . Salvesen, G. and Dixit, V. (1997). Caspases: Intracellular signaling by proteolysis. Cell 91 , 443-446. 77 Sices, H.J. and Kristie, T.M. ( 1998). A genetic screen for the isolation and characterization of site-specific proteases. Proc. Natl. Acad. Sci. 95, 2828-2833. Sikorski, R. and Hieter, P. (1989). A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in saccharomyces-cerevisiae. Genetics 122, 19-27. Song, Z., McCall, K. and Steller, H. (1997). DCP-1, a Drosophila cell death protease essential for development. Science 275, 536-540. Srinivasula, S., Ahmad, M., Fernandes-Alnemri, T. and Alnemri, E. (1998). Autoactivation of procaspase-9 by Apaf-1-mediated oligomerization. Molecular Cell 1, 949-957. Srinivasula, S., FernandesAlnemri, T., Zangrilli, J., Robertson, N., Armstrong, R., Wang, L., Trapani, J., Tomaselli, K., Litwack, G. and Alnemri, E. (1996). The Ced3/interleukin 1 beta converting enzyme-like homolog Mch6 and the Jamin-cleaving enzyme Mch2 alpha are substrates for the apoptotic mediator CPP32. Journal of Biological Chemistry 271, 27099-27106. Stagijar, I., Korostensky, C., Johnsson, N. and te Heesen, S. (1998). A genetic system based on split-ubiquitin for the analysis of interactions between membrane proteins in vivo. Proceedings of the National Academy of Sciences USA 95, 5187-5192. Struhl, G. and Adachi, A. (1998). Nuclear access and action of notch in vivo. Cell 93, 649-660. 78 Teodoro, J. and Branton, P. ( 1997). Regulation of apoptosis by viral gene products. Journal of Virology 71, 1739-1746. Thornberry, N., Bull, H., Calaycay, J., Chapman, K., Howard, A., Kostura, M., Miller, D., Molineaux, S., Weidner, J., Aunins, J., Elliston, K., Ayala, J., Casano, F., Chin, J., Ding, G., Egger, L., Gaffney, E., Limjuco, G., Palyha, 0., Raju, S., Rolando, A., Salley, J., Yamin, T., Lee, T., Shively, J., Maccross, M., Mumford, R., Schmidt, J. and Tocci, M. (1992). A novel heterodimeric cysteine protease is required for interleukin1-beta processing in monocytes. Nature 356, 768-774. Thornberry, N. and Lazebnik, Y. (1998). Caspases: Enemies within. Science 281, 13121316. Thornberry, N., Ranon, T., Pieterson, E., Rasper, D., Timkey, T., GarciaCalvo, M., Houtzager, V., Nordstrom, P., Roy, S., Vaillancourt, J., Chapman, K. and Nicholson, D. ( 1997). A combinatorial approach defines specificities of members of the caspase family and granzyme B - Functional, relationships established for key mediators of apoptosis. Journal of Biological Chemistry 272, 17907-17911. Uren, A., Coulson, E. and Vaux, D. (1998). Conservation of baculovirus inhibitor of apoptosis repeat proteins (BIRPs) in viruses, nematodes, vertebrates and yeasts. Trends in Biochemical Sciences 23, 159-162. Uren, A., Pakusch, M., Hawkins, C., Puls, K. and Vaux, D. (1996). Cloning and expression of apoptosis inhibitory protein homologs that function to inhibit apoptosis 79 and/or bind tumor necrosis factor receptor-associated factors. Proceedings of the National Academy of Sciences USA 93, 4974-4978. Varshavsky, A. ( 1996). The N-end rule: Functions, mysteries , uses. Proceedings of the National Academy of Sciences USA 93, 12142-12149. Villa, P., Kaufmann, S. and Earnshaw, W. (1997). Caspases and caspase inhibitors. Trends in Biochemical Sciences 22, 388-393. Vucic, D., Kaiser, W., Harvey, A. and Miller, L. (1997). Inhibition of Reaper-induced apoptosis by interaction with inhibitor of apoptosis proteins (IAPs). Proceedings of the National Academy of Sciences USA 94, 10183-10188. Vucic, D., Kaiser, W. and Miller, L. (1998). Inhibitor of apoptosis proteins physically interact with and block apoptosis induced by Drosophila proteins HID and GRIM. Mol Cell Biol 6, 3300-3309. Wang, S., Miura, M. , Jung, Y., Zhu, H., Li, E. and Yuan, J. (1998). Murine caspase11 , an ICE-interacting protease, is essential for the activation oflCE. Cell 92, 501-509. Xu, X., Gerard, A., Huang, B., Anderson , D., Payan, D. and Luo, Y. (1998). Detection of programmed cell death using fluorescence energy transfer. Nucleic Acids Research 26, 2034-2035 . Yang, X. , Chang, H. and Baltimore, D. (1998a). Autoproteolytic activation of procaspases by oligomerization. Molecular Cell 1, 319-325. 80 Yang, X., Chang, H. and Baltimore, D. (1998b). Essential role of CED-4 oligomerization in CED-3 activation and apoptosis. Science 281, 1355-1357. Zhang, Y., Center, D., Wu, D., Cruikshank, W., Yuan, J., Andrews, D. and Kornfeld, H. (1998). Processing and activation of pro-interleukin-16 by caspase-3. Journal of Biological Chemistry 273, 1144-1149. Zou, H., Henzel, W., Liu, X., Lutschg, A. and Wang, X. (1997). Apaf-1, a human protein homologous to C-elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell 90, 405-413. 81 CHAPTER 3 The Drosophila Caspase Inhibitor DIAPl Is Essential for Cell Survival and Is Negatively Regulated by HID Susan L. Wang, Christine J. Hawkins, Soon Ji Yoo, H.-Arno J. Mtiller, and Bruce A. Hay Published in Cell 98, 453-463, August 1999 82 Summary Drosophila REAPER (RPR), HID and GRIM induce caspase-dependent cell death and physically interact with the cell death inhibitor DIAP I. Here we show that HID blocks DIAPI 's ability to inhibit caspase activity, and provide evidence suggesting that RPR and GRIM can act similarly. Based on these results we propose that RPR, HID and GRIM promote apoptosis by disrupting productive IAP-caspase interactions, and that DIAPI is required to block apoptosis-inducing caspase activity. Supporting this hypothesis we show that elimination of DIAPI function results in global early embryonic cell death and a large increase in DIAPl-inhibitable caspase activity, and that DIAPl is still required for cell survival when expression of rpr, head involution defective and grim is eliminated. 83 Introduction Programmed cell death, or apoptosis, is an evolutionarily conserved process by which organisms remove damaged or unwanted cells (reviewed in Wyllie et al. , 1980; Raff, 1992). Central components of the machinery that carries out this process are caspases, a family of aspartate-specific, cysteine-dependent proteases (Alnernri et al., 1996). Caspases are made as zymogens comprising three functional modules: a prodomain and two catalytic subunits known as the large, or p20, and the small, or pl 0, subunits. In general caspases are activated in vivo following cleavage at aspartate residues that separate the prodomain from the catalytic region, and that separate the large and small subunits of the catalytic domain. The aspartates cleaved in caspase zymogens often resemble consensus target sites of known caspases. This has suggested that caspases may function in a cascade in which initiator caspases, activated by upstream death signals, cleave and activate a set of executioner caspases that carry out proteolytic cleavages of cellular proteins, leading ultimately to cell death and phagocytosis (reviewed in Nicholson and Thornberry, 1997; Salveson and Dixit, 1997; Cryns and Yuan, 1998; Thornberry and Lazebnik, 1998). Because caspases have the potential to initiate cascades of proteolysis it is important that their activity be tightly regulated. The only cellular caspase inhibitors identified to date all share homology with a family of cell death inhibitors identified in baculoviruses, known as inhibitors of apoptosis, or IAPs (Crook et al., 1993; Birnbaum et al., 1994; reviewed in Deveraux and Reed, 1999; LaCasse et al., 1998). IAP homologous proteins have been identified in other viruses, Drosophila, C. elegans, mammals and yeast (reviewed in Uren et al., 1998). These proteins are characterized by one or more N-terminal repeats of a motif known as the baculovirus IAP repeat (BIR). Many also have a C-terminal RING finger motif. Some IAP homologous proteins have 84 been shown to act as cell death inhibitors. In several cases this activity is localized to the BIR repeat-containing region of the protein (reviewed in Lacasse et al., 1998; Uren et al., 1998; Deveraux and Reed, 1999), which has also been shown to mediate caspase binding and inhibition (Deveraux et al. , 1997; Roy et al., 1997; Deveraux et al., 1998; Takahashi et al., 1998; Hawkins et al., 1999). Proteins that initiate cell death by disrupting IAP-caspase interactions have not been described. However, the products of the Drosophila reaper (rpr), head involution defective (hid) and grim genes are interesting candidates. All three loci are located in the 75C region of the Drosophila genome, and deletion of this region eliminates death induced by many different stimuli, indicating that important cell death activators reside within (White et al., 1994). Furthermore, individual overexpression of any one of these genes is sufficient to initiate caspase-dependent cell death in many cells that normally live (Grether et al., 1995; Hay et al., 1995; Chen et al., 1996; White et al., 1996; Vuvic et al., 1997, 1998; Zhou et al., 1997; Wing et al., 1998). Two Drosophila IAPs, DIAPl, the product of the th locus (Hay et al., 1995), and DIAP2 (Hay et al., 1995; Ducket et al., 1996; Uren et al., 1996), suppress rpr-, hid- and grim-dependent cell death when overexpressed (Hay etal., 1995;Vucicetal., 1997, 1998;Wingetal., 1998),andDIAPl andDIAP2can immunoprecipitate RPR, HID and GRIM from insect cells (Vucic et al., 1997, 1998). Furthermore, DIAPl binds a processed form of the caspase drICE in insect cells (Kaiser et al., 1998), and inhibits the activity of a second caspase, DCP-1 (Hawkins et al., 1999) . These observations suggest several models of how Drosophila IAPs, RPR, HID, and GRIM, and Drosophila caspases might interact to regulate cell death (Kaiser et al., 1998). In one model, RPR, HID or GRIM activate caspases through an !AP-independent pathway. In this model Drosophila IAPs suppress apoptosis by acting as a sink for these proteins and by inhibiting caspase activation or activity initiated by their action. In a second model DIAPl functions primarily as a caspase inhibitor, and RPR, HID and/or 85 GRIM initiate caspase-dependent cell death by preventing IAPs from productively interacting with caspases. The yeast Saccharomyces cerevisiae provides an ideal system to study interactions between caspases and molecules that regulate their activity (Hawkins et al., 1999; Kang et al., 1999). While yeast cells do not exhibit apoptosis, and database searches have failed to uncover yeast homo logs of the core apoptosis regulators identified in worms, flies and mammals, overexpression of DCP-1 or drICE results in yeast cell death that is blocked by DIAPI. Here we used yeast expressing these proteins as a system to assay the ability of RPR, HID and GRIM to disrupt IAP-caspase interactions. We found that all three proteins, while nontoxic on their own, killed yeast coexpressing DCP-1 or drICE, and DIAPl, suggesting that they were blocking DIAPl 's ability to function as a caspase inhibitor. We pursued the basis for this activity further with HID and found, both in yeast and in vitro, that proteins containing the N-terminal 37 residues of HID, which are sufficient to induce apoptosis in insect cells (Vucic et al., 1998), suppressed DIAPl 's ability to inhibit DCP-1 activity. We also found that the DIAPl loss-of-function phenotype consists of an embryo-wide set of cellular changes reminiscent of apoptotic cell death, and that these were associated with the activation of DIAPl-inhibitable caspase activity. Furthermore, double mutants that remove zygotic rpr, hid, grim, and DIAPl function showed phenotypes similar to those of the DIAPl loss-of-function mutant alone. These observations suggest that a principal function of DIAPl is to promote cell survival by blocking caspase activity, and that at least one mechanism by which RPR, HID and GRIM promote apoptosis is by disrupting IAP-caspase interactions. Results and Discussion 86 RPR, HID and GRIM block DIAPl's ability to suppress caspase-dependent cell death in yeast Yeast were transformed with two plasmids: one in which DCP- I expression was driven by the inducible GALI promoter, and a second in which DIAPI expression was driven by the constitutive Adh promoter (hereafter referred to as DCP-I-DIAPI yeast). These cells were then transformed with a second GALI expression plasmid that was either an empty vector, or that carried GALI-driven RPR, HID or GRIM. Transformants were spotted as a series of tenfold serial dilutions onto glucose plates to indicate the number of cells, and onto galactose plates to induce expression of DCP- I and RPR, HID or GRIM. DCP- IDIAP I yeast carrying an empty vector survived on galactose, but DCP-I-DIAPI yeast expressing HID or GRIM did not (Figure IA). Expression of RPR under GALI control did not kill DCP-I-DIAPI yeast in which DIAPI expression was driven by the Adh promoter, but it was able to kill DCP-1-DIAPI yeast in which DIAPI expression was driven by a weaker promoter, the copper-inducible CUP-I promoter in the presence of 30µM Cu++ (Figure I A). Importantly, expression in yeast of RPR, HID or GRIM alone under GALI control had no effect on cell growth as compared with yeast expressing only the empty vector (Figure 1A). These results suggest that RPR, HID and GRIM are able to block DIAPl 's ability to inhibit DCP-I caspase activity. Three other Drosophila caspases, drICE and DCP-2/Dredd, and DRONC have been implicated as effectors of apoptosis (Fraser et al., 1997, 1998; Chen et al., 1998; Dorstyn et al., 1999). We wanted to determine if RPR, HID and GRIM played a similar role in regulating their activity. We focused our attention on drICE. A version of drICE lacking the prodomain is not inhibited by DIAPl (Hawkins et al., 1999), but a version of drICE that contains prodomain sequences is (Hawkins and Hay, unpublished observations). Full length drICE expressed in yeast under GALI control neither kills nor 87 Figure J. RPR, HID and GRIM block DIAPl 's ability to suppress caspase-dependent yeast cell death. (A) RPR, HID and GRIM kill yeast kill yeast expressing DCP- I and DIAPI. Top: GALI-driven DCP-I causes yeast cell death which is prevented by Adh-dependent expression of DIAPI (DCP-1 + Adh-DIAPl + vector). GALI-driven HID (DCP-I + Adh-DIAPI + HID) or GRIM (DCP-I + Adh-DIAPI + GRIM) results in yeast cell death . Middle: GALI-driven RPR kills yeast expressing DCP-I and CUPl-driven DIAPI DCP-1 + CUP-DIAPI + RPR) in the presence of 30mM Cu++. Bottom: Expression of RPR, HID or GRIM in the presence of empty vectors does not affect yeast cell viability. (B) RPR, HID and GRIM kill yeast expressing rev-drICE and DIAPl Top: drICE contains prodomain (P), large (LS) and small (SS) subunits. In reversed drlCE (rev-drICE) the predicted drICE small subunit was placed N-terminal to drICE prodomain and large subunit sequences. Middle: Expression of HID or GRIM kills yeast expressing DCP-1 and Adh-DIAPI (rev-drICE + Adh-DIAPl + HID; rev-drICE + AdhDIAPl + GRIM) . Bottom: Expression of RPR kills yeast expressing rev-drICE and CUP-DIAPI (rev-drlCE + CUP-DIAPI + RPR) in the presence of no added Cu++. (C) Diagram of HID constructs. Numbers represent amino acid positions in full length HID. Solid lines indicate regions of HID present in the fusion protein, dashed lines sequences that are deleted. A polyoma epitope tag present at the C-terminus of each coding region is indicated by a black bar. (D) Proteins containing the first 37 residues of HID block DIAPI 's ability to inhibit DCP1 activity in yeast. Top: Effects of expression of HID fu sion proteins on the viability of DCP-1 + Adh-DIAPI yeast. Bottom: Effects of expression of HID fusion proteins and empty vectors on yeast viability. 88 A GLUCOSE DCP-1 + vecta-s DC P-1 + Adh-DIAP1 + vector DCP-1 + Ach-DIAP1 + HID DCP-1 + Adh-DIAP1 + GRIM vectcn; DCP-1 • vectors DCP-1 + CUP-OIAP1 + vector DCP-1 + CUP-OIAP1 + RPR vectors vectors + RPR vectors + HID vectors+ GRIM C ■ • •• 410 ~ HD HIDl\(38-335) GAuRAF 37 336 d - : _- _- _-_-_- _- --t::=11 37 B LS GLUCOSE rev-drlCE + vectorn rev-drlCE + AdMJIAP1 + vector rev-<lrlC E + Adh-DIAP1 + HID r<>v-drlCE • Adh-OIAP1 + GRIM rev-drlCE + vectors rev-drlCE + CUP-OIAP1 + vector relMlrlCE • CUP-DIAP1 + RPR D DCP-1 +Adh-OIAP1 • HID DCP-1 + Ml-DIAP1 + HIOl\(38-335) OCP-1 + Adh-DIAP1 + HIDl\(1-335) HIOt,(1-36) : lt=========:::J■: HID,',(1-335) 336 ~----_-_-_-_-_-_-_-_-.!:::=. vectorn HID + vectors HIDl\(38-335) + vectors HIDl\(1-335) + vectors HIDl\(1-36) + vectorn GALJRAF ■ • GLUCOSE DCP-1 + Adi-DIAP1 + vector ss LS relMlrlCE DCP-1 +Adh-OIAP1 + HIDt,(1-36) ■ Polyoma epitope tag Id wildtype drlCE •• •• •• • ., GAL/RAF ••" ·'.i 0 • (j 0 e ·- • • • " ,, •• t;. fl • • • • • .. • • " •• :,- 6) ·> C, ,;'! 89 activates a cleavage-dependent reporter, probably because the caspase does not significantly autoactivate (Hawkins et al., 1999). We created a form of drICE that is active in yeast and that contains prodomain sequences by generating a "reversed" form of drICE, rev-drICE, in which the drICE p!O domain was placed N-terminal to prodomain and p20 sequences (Figure IB). Reversed forms of mammalian caspases that are not otherwise active are constitutively active (Srinivasula et al., 1998). As shown in Figure IB, expression of rev-drICE under GALI control kills yeast and this death is prevented by coexpression of DIAPI. Importantly, as with DCP-1, expression of HID or GRIM or RPR killed yeast coexpressing rev-drICE and DIAPI. The N-terminus of HID mediates its ability to disrupt IAP-caspase interactions in yeast In insect cells the N-terminal 37 amino acids of HID, expressed in the context of a larger fusion protein, are necessary and sufficient to induce apoptosis and to mediate HID's binding to DIAPI (Vucic et al., 1998). To test the activity of these sequences in yeast we generated GALI expression constructs similar to those of Vucic et al., (1998) in which the first 37 amino acids of HID were either present or absent (Figure IC). HID6(38-335) encodes a protein in which the first 37 amino acids of HID are fused to the C-terminal 74; HID6(1-335) consists of only the last 74 residues of HID, and HID6(1-36) encodes a protein that lacks the first 36 residues of HID, but contains the rest of the full length HID coding sequence. GALI-dependent expression of HID6(38-335) in DCP-1-DIAPI yeast resulted in no growth. In contrast, DCP-1-DIAPI yeast expressing HID6(1-335) grew normally. HID6(1-36) had weak killing activity in this assay (Figure ID), but not in insect 90 cell death assays (Vucic et al., 1998), or in other in vitro assays described below. As with full length HID, expression of HID~(38-335), HID~(l-335) or HID~(l-36) in isolation had no effect on yeast cell growth (Figure ID). Thus, proteins carrying the N-terminal 37 residues of HID, which are sufficient to kill insect cells, are also able to kill DCP-1-DIAPl yeast. HID suppresses DIAPl's ability to inhibit DCP-1 caspase activity in vitro The results of our yeast experiments suggested that HID directly inhibits DIAPl. We tested this idea in vitro using purified proteins. Bacterially expressed and purified versions of DCP-1, DIAP 1 and HID (Figure 2A) were mixed in various combinations, and DCP-1 caspase activity measured fluorometrically (Figure 2B, C). His6-tagged DCP-1 (DCP-1His6) is active as a caspase, and this activity was inhibited by DIAPl but not by GST (Hawkins et al., 1999; Figure 2B). Addition of HID alone to a reaction containing DCP-lHis6 did not alter DCP-1-His6 activity, indicating that HID has no direct effect on DCP-1His6 (Figure 2B). However, when full length HID was incubated with DIAPl and DCP1-His6, DIAPl-dependent inhibition of DCP- 1-His6 activity was lost (Figure 2B). Incubation of DIAP 1 and DCP-1-His6 with HID~(l-36)-His6 had little or no effect on DIAPl 's ability to inhibit DCP-1-His6 activity, indicating that the first 37 residues of HID are necessary for this activity. To determine if the first 37 residues of HID were sufficient to mediate this effect, we carried out experiments using a version of HID that consists of the first 37 residues of HID fused to GST and a C-terminal His6 tag (HID(l-37)GSTHis6). As shown in Figure 2C, addition of HID( l-37)GST-His6 to an assay containing DCP-1-His6 and DIAPl blocked DIAPl 's ability to inhibit DCP-1-His6 activity, while 91 Figure 2. Proteins containing the first 37 residues of HID block DIAPl 's ability to inhibit DCP-1 activity in vitro. (A) Schematic of HID constructs. HID coding regions are represented by open boxes, deleted regions by dashed lines. (B) The N-terminal 36 residues of HID are required to inhibit DIAPl function. DCP-lHis6 (0.2 nM) activity was measured by following the release of AFC fluorometrically from the Ac-DEVD-AFC substrate over time. The total protein concentration in each incubation was kept constant, with GST (0.2µM or 0.77µM) being used to replace DIAPl (0.2µM), HID-His6 (0.44µM) or HID~( l-36)-His6 (0.44µM), as the assay required. (C) A protein containing the N-terminal 37 residues of HID, HID(l-37)GST-His6, is sufficient to block DIAPI 's caspase inhibitory activity. Assays were performed as in (B). 92 A 410 HID-His 6 37 HIDA(1 ·36)-His 6 : lt==========:::,~si 37 HIO(1-37)GST-His 6 l.,....,...❖...·.N GST-His 6 t->:•:•:·:•·•:•,:•M l.'S His 6 epitope t~ □ GST sequence B 0 20 40 60 Time(min) DCP-1-His 6+GST+GST ■ ■ DCP-1 -His 6+DIAP1 +GST DCP-1-Hls 6+GST +HID-His 6 DCP-1-Hls 6+OIAP1 +HID-His 6 DCP-1-His 6+DIAP1+HIDA(1·36)-Hls 6 C 0 20 40 60 Time (min) - DCP-1-His6+GST+GST-His6 - Cl- DCP-1-His 6+GST+HID(1-37)GST-His 6 -o- DCP-1-His 6+0\AP1+GST-His6 -6.- DCP-1-His 6+OIAP1+HID(1-37)GST-His 6 93 addition of an equivalent amount of GST-His6 had no effect. Thus, consistent with our observations in yeast, purified HID directly blocks DIAP 1's ability to inhibit DCP-1 caspase activity, and the first 37 amino acids of HID, at least in the context of a larger fusion protein, are necessary and sufficient for this activity. To determine if HID directly interacts with DIAP I or DCP-1 , we carried out binding experiments. As expected from the results of our yeast and in vitro activity assays, GST-DIAPl coupled to glutathione-Sepharose bound full length HID-His6 and DCP-1His6, but did not appreciably bind HIDL1(1-36)-His6. GST alone showed no interaction with any of these proteins (Figure 3A). Also, a maltose binding protein-DIAPl fusion protein (MBP-DIAPl) bound to amylose resin was able to bind HID(l -37)GST-His6, while MBP bound to the same resin was not (Figure 3B). Furthermore, HID(l-37)GSTHis6, bound to glutathione Sepharose beads was able to bind 35s-methionine labeled, in vitro translated DIAPl, but not in vitro translated DCP-1 (data not shown). These observations, in conjunction with the fact that HID-His6 or HID(l-37)GST-His6 alone had no effect on DCP-l-His6 caspase activity (Figure 2B,C), suggest that HID' s effects on caspase activity are mediated through DIAPl. To explore the mechanism by which HID blocks DIAP 1's ability to inhibit DCP-1, we asked if HID( 1-37)GST-His6 was able to displace DCP-1-His6 from DIAPl. MBPDIAPI bound to amylose resin was preincubated with increasing amounts of GST or HID(l-37)GST-His6 followed by addition of DCP-l-His6. Binding reactions were followed by washes and analysis of the bead bound proteins by Western blotting. As shown in Figure 3C, DCP-1 binding to MBP-DIAPl was not appreciably decreased by the presence of up to a 20-fold molar excess of GST-His6 or HID(1-37)GST-His6 over MBP-DIAPl. As expected, MBP beads incubated with HID(l-37)GST-His6 showed no binding to DCP-1-His6. Thus , we see no evidence that HID(l -37)GST-His6, at molar ratios with respect to DIAP I greater than those used to demonstrate inhibition of DIAP 1 94 Figure 3. Physical interactions between HID, DIAPl and DCP-1 (A) DIAPl binds DCP-1 and HID, and the first 37 residues of HID are required for this interaction. Glutathione Sepharose-bound GST or GST-DIAPl was incubated in binding buffer with DCP-l-His6, HID-His6 or HIDL1(1-36)-His6 (Figure 2A). Binding reactions were followed by washes and Western blotting. A fraction of the input of DCP- l-His6, HID-His6 and HIDL1(1-36)-His6 is shown. (B) DIAPl binds proteins containing only the first 37 residues of HID. Amy lose resinbound MBP or MBP-DIAPl was incubated with GST-His6 or HID(l-37)GST-His6 and processed as above. (C) HID(l-37)GST-His6 does not prevent DCP-l-His6 from binding MBP-DIAPl. MBP-DIAPl bound to amylose resin was incubated with increasing amounts of HID(l37)GST-His6 or GST-His6. Following this preincubation a constant amount of DCP-lHis6 was introduced into the mixture. The beads were washed and processed for Western blotting using an anti-His antibody to detect DCP-l-His6. MBP beads incubated with the highest concentration of HID(l-37)GST-His6 and equivalent amounts of DCP- l-His6 were processed in a similar manner. 95 A +GST +GST-OIAP1 INPUT kDa 60 30- - - 20 102 8 4 3 5 +MBP +MBPDIAP1 1 3 7 6 ♦ DCP-1-His6 (p10) 8 9 --- C 2 4 5 6 - ----------- +MBP Add: INPUT +MBP-DIAP1 HID(1 -37) GST·His6 GST-His 6 ....-:::::::::J ________, ♦ DCP-1-H i s6 (p10) 2 3 4 5 6 7 96 function in caspase activity assays (Figure 2C), prevents binding of DCP-1-His6 to MBPDIAPI. These observations suggest that HID(l-37)GST-His6 does not prevent DIAPl from inhibiting DCP-1 activity by blocking DCP-1 binding. It is important to note, however, that full length HID, which has a large C-terminal domain that can regulate HID activity (Bergmann et al., 1998), may behave differently. Our observations suggest that RPR, HID and GRIM can promote caspasedependent cell death by blocking DIAPl 's ability to inhibit caspase activity. These conclusions are supported by the observations that in insect cells RPR, HID and GRIM promote caspase-dependent cell death which is sensitive to the levels of DIAPl (Hay et al., 1995; Vucic et al., 1997, 1998; Wing et al., 1998), that these proteins associate with DIAPl in vivo (Vucic et al., 1997, 1998), and that DIAPl binds and inhibits caspases (Kaiser et al., 1998; Hawkins et al., 1999; this work). Furthermore, since proteins that contain only the first 37 amino acids of HID are sufficient to initiate the caspase-dependent death of insect cells and to block DIAP 1's ability to inhibit caspase activity, but other regions of HID lack activity in insect cells, it is likely that disrupting DIAPl 's ability to inhibit caspase activity is an important mechanism by which HID promotes cell death. Because of difficulties with solubility of RPR and GRIM under conditions that allowed productive DIAPl-caspase interactions, we were unable to carry out in vitro experiments similar to those carried out with HID. However, based on the fact that they behaved similarly to HID in yeast expressing DCP-1 or rev-drICE, and DIAPl, it seems likely that they also are able to block DIAPl 's ability to inhibit caspase activity. Our observations do not exclude the possibility that RPR, HID and GRIM promote apoptosis through other pathways as well. For example, RPR has been shown to promote apoptosis in a Xenopus system (Evans et al., 1997) through association with a novel protein, SCYTHE (Thress et al., 1998), and peptides corresponding to N-terminal sequences of RPR and GRIM block K+ channel function (Avdonin et al., 1998). 97 Furthermore, as described below, removal of DIAPl in the embryo results in massive cell death, but the DIAPl mutant cells do not show all the features of apoptosis seen in wildtype embryos, consistent with the idea that other pathways may be necessary to create a complete apoptotic response. DIAPl mutant embryos show global early morphogenetic arrest and cell death associated with an increase in DIAPl-inhibitable caspase activity If disruption of IAP-caspase interactions is an important mechanism by which cell death is initiated in Drosophila, then caspase activity must be continually held in check by DIAPl and/or other IAPs in order to block an ever present death signal. To explore this possibility we characterized the embryonic phenotype of mutations (alleles of th) that disrupt DIAPl function. Homozygous th mutant embryos exhibit a severe terminal embryonic lethal phenotype and do not form a cuticle (Figure 50, and data not shown). We monitored early development of loss of function th embryos from th5, a strong loss-of-function allele, and a deficiency for the region, Df(3R)brml 1 (Figure 4). th5 and Df(3R)brml 1 homozygous embryos undergo normal cellularization and initial gastrulation movements such as ventral furrow formation, cephalic furrow formation and posterior midgut invagination (Figure 4A-C). However, during germband extension, within 15 to 30 min after the onset of gastrulation movements, virtually all morphogenetic movements cease (Figure 4A-C). While the stage of initial arrest is somewhat variable in time (compare Figure 4B and C with respect to extent of ventral furrow formation and posterior midgut invagination), within the next hour all the cells adopt a rounded morphology and the yolk cell fragments and eventually moves to the surface of the embryo. The disruption of embryonic integrity is seen particularly clearly in Figures 4D-G, in which wildtype and th5mutant embryos are 98 Figure 4. Removal of zygotic DIAPI leads to morphogenetic arrest during germband extension and global cell shape changes. (A) Wild type developmental series of living embryos derived from ath5fTM3[ftz::lacZ] stock during gastrulation and germband extension (stages 6 through 9). Arrowheads mark the position of the cephalic furrow and arrows indicate the front of the extending germband. (B), (C) th5 and Df(3L)brml 1 homozygous embryos, respectively, at the same developmental stages as in (A). (D) , (F) Scanning electron micrographs of balancer bearing embryos from th5/TM3[ftz::lacZ] stock. The right-hand panels show higher magnification of the general morphology of the embryonic surface. (D) shows an embryo at stage 8, (F) shows an embryo at stage 9. (E), (G) Similar staged embryos homozygous for th5. Scale bars are 100 µmin (C) and (G) (left-hand panel), and lOµm in (G) (right-hand panel). 99 . _,·' ~- , .,..,._-..•;1ii:~ / 15' •:,, ....,,.••, - ... 100 Figure 5. Cells in embryos homozygous mutant for DIAPI show premature and increased DNA fragmentation and an increase in DIAPl-inhibitable caspase activity. (A-F) Confocal images of staged embryos from stocks of th5ffM3[ftz::lacZ], and (0, H)rh6ffM3[ftz::lacZ] double labeled for TUNEL (green) and ~-galactosidase (red). Scalebar is 100µm. (A) th5ffM3[ftz::lacZ], and (B) th5/th5 embryos at early germband extension (stage 8). (C) rh5ffM3[ftz::lacZ], and (D) th5/th5 embryos at extended germband stage (early stage 9). (E) th5ffM3[ftz::lacZ], and (F) th5/th5 embryos at extended germband stage (later stage 9). (0) th6ffM3[ftz::lacZ], and (H) th6/th6 embryos at retracted germband stage (stage 14). (I) Caspase activity of wildtype and th mutant embryo extracts in the presence of OST or DIAPl. lµg of 2-4 hour embryo extract from stocks of w-, th5ffM3 or th6ffM3 w- was introduced into caspase activity buffer containing lO0µM of the Ac-DEVD-AFC substrate and 0.5µg of either OST or DIAPI . Release of AFC was monitored fluorometrically over time. 10 l I 6 o w+GST X th6+GST a th5+GST • th5+ OIAP1 0 0 15 30 Tlme(min) 45 102 visualized using scanning electron microscopy. While some th5 homozygous embryos showed normal morphology at the beginning of germband extension (data not shown), other embryos with a similar degree of extension movements showed extensive cell rounding, suggesting a disruption of intercellular adhesion (Figure 4D, E). All th5 embryos also show a regression of the cephalic furrow (compare Figure 4A with Figure 4B and C, and Figure 4D with 4E). At the extended germband stage (stage 9), the surface of cells in control embryos is smooth and the cells are closely apposed to each other (Figure 4F). In contrast, by this time th5 homozygotes have adopted a uniform cellular morphology in which all cells have a round shape, the surface of the cells contains many membrane blebs, and vesicular cell fragments are present similar to those seen in dying cells (Figure 4G). The yolk cell, normally located interiorly, is visible at the surface. Thus, zygotic loss of DIAPl function results in early morphogenetic arrest followed by severe morphological abnormalities at the beginning of germband extension. We used TUNEL labeling to determine whether the massive DNA fragmentation characteristic of cells undergoing apoptosis had occurred in cells of th mutants (Figure 5). Wildtype embryos during germband extension showed significant DNA fragmentation in only a few cells (Figure 5 A,C, E). In contrast, comparably staged th5 embryos labeled extensively in most if not all cells. Initially, the TUNEL signal is relatively weak (Figure SB, D), but the intensity increases within the following hour of development to a strong uniform labeling (Figure 5B,D,F). In contrast, embryos homozygous for the weaker allele th6 do not show a significant increase in TUNEL positive cells as compared to comparably staged control embryos at the retracted germband stage (stage 14) (Figure 5G,H). To determine if loss of DIAPI in fact results in an increase in caspase activity, we measured caspase activity in extracts of 3-5 hour old embryos from wildtype, th5 and th6 stocks by following the release of Ac-DEVD-AFC fluorometrically (Figure 51) . Extracts 103 from wildtype and th6 embryos had very low levels of caspase activity. In contrast, extracts from th5 embryos had very large amounts of caspase activity. Importantly, the increased caspase activity present in the th 5 embryonic extracts was inhibited by the addition of purified DIAPl (Figure SJ), consistent with the idea that DIAPl normally represses these caspases. The caspase targets of DIAPl in the Drosophila embryo or other tissues have not been identified. Five caspases have been identified in Drosophila: DCP-1 (Song et al., 1997), drlCE (Fraser and Evan, 1997), DCP-2/Dredd (Inohara et al., 1997; Chen et al., 1998), DRONC (Dorstyn et al., 1999), and a fifth gene (C. J. H. and B. A. H., unpublished) . DIAPl is able to inhibit the activity of the three caspases we have tested: DCP-1 (Hawkins et al., 1999), drICE (this report) and DRONC (C. Hawkins, unpublished). The DIAPl embryonic phenotype may result from the unrestrained activity of zygotically expressed versions of these caspases. Alternatively, DIAPl may also function in the early embryo to block the activity of caspases introduced late during oogenesis into the developing oocyte from the degenerating nurse cells. (Cavaliere et al., 1998; Chen et al. , 1998; Foley and Cooley, 1998; McCall and Steller, 1998; Dorstyn et al., 1999). DIAPl loss-of-function mutant embryos do not stain with acridine orange, a marker of apoptotic cell death Acridine orange (AO) is commonly used as a marker for apoptotic cell death. It appears to enter all cells, but is only retained (through unknown mechanisms) by cells undergoing apoptosis (c.f. Abrams et al., 1993). In light of the strong defects in th mutant embryos reported here, we repeated experiments described in Hay et al., ( 1995) reporting that th embryos did not show an increase in AO staining. Figure 6A shows a brightfield image of 104 Figure 6. Cells in DIAPl loss-of-function embryos do not show an increase in acridine orange staining characteristic of cells undergoing apoptosis in wildtype embryos. (A), (B) stage 14 rh5ffM3[ftz::lacZ] or TM3[ftz::lacZ]/ TM3[ftz::lacZ] embryo visualized using Nomarski optics (A) and confocal microscope following acridine orange staining (B). (C), (D) th5 homozygous embryo exhibiting the terminal phenotype. At this stage acridine orange staining is never observed in th5 homozygotes (D). Scale bar is 50µm . 105 106 a stage 14 th5/TM3 embryo. Figure 6B shows the same embryo stained with AO. Dying cells show bright green fluorescence and living cells do not stain. A th5 homozygous embryo displaying the th terminal phenotype is shown in Figure 6C. All cells have rounded up and the yolk sac has moved to the surface. However, as shown in Figure 6D, no AO staining is seen, despite the fact that embryos at this stage show cell rounding, membrane blebbing, increased caspase activity and DNA fragmentation. Thus, the th phenotype, while showing some features of apoptosis, lacks others. One explanation for this is that the early embryo simply lacks the cellular machinery required to carry out apoptotic steps leading to AO retention . A second possibility is that the expression of other IAPs , such as DIAP2, suppresses some of the caspase-dependent events necessary for a full apoptotic cell death phenotype. A third possibility (discussed further below) is that loss of DIAPI function, while sufficient to kill cells, is not sufficient to confer all aspects of apoptosis. In this scenario, proteins such as RPR, HID or GRIM, that are sufficient to trigger a full apoptotic response when overexpressed, may play dual roles: inhibiting IAP function and in addition activating other apoptotic pathways (Figure 7). DIAPl is epistatic to rpr, hid and grim To explore further the relationship between rpr, hid, grim and DIAPl, we made a double mutant with the H99 deficiency, which includes rpr, hid and grim, and th5. Homozygous H99 embryos essentially lack apoptosis (White et al. , 1994). However, double mutants show a terminal embryonic phenotype similar to that of th5 alone (data not shown). This, in conjunction with the fact that DIAPI is a caspase inhibitor, argues against models in which DIAPI 's sole antiapoptotic role is to bind to and suppress RPR, HID and GRIMdependent death pathways, since in this scenario the double mutant phenotype would be 107 Figure 7. Model for how DIAPl regulates apoptosis. DIAPl inhibits caspase activity (--1) and is essential for cell survival. DIAPl 's ability to inhibit caspase activity is suppressed by RPR, HID and GRIM (--1), thus promoting caspase-dependent cell death. RPR and GRIM have other activities (Evans et al., 1997; Thress et al., 1998; Avdonin et al., 1999) suggesting that these proteins may also promote apoptosis through other pathways (dashed arrows). Free DIAPl may inhibit apoptosis by binding these proteins, sequestering them from their targets (dashed-------!). Thus DIAPl may act both upstream and downstream of RPR, HID and GRIM to prevent cell death. Genes within the H99 interval are not required for some cell deaths (White et al., 1994; Foley and Cooley, 1998), indicating that other death pathways exist (dashed arrow) . It is not known if these pathways act through DIAPl or through other pathways 108 / - ♦ other effector pathways RE~~ER _ _ _.,.I DIAP1 genes GRIM 1---·· : H99 H99-independent death pathways - - - - - .... I caspase activity ___. apoptosis ,,,!'_1 ------ A 109 expected to be similar to that of the H99 mutant alone. Instead, our observations are consistent with models in which DIAPl has an important antiapoptotic function as a caspase inhibitor and RPR, HID and GRIM function, at least in part, through DIAPl to promote apoptosis (Figure 7). DIAPl may, however, have antiapoptotic functions in addition to caspase inhibition. IAPs in insects and mammals have been shown to bind a number of different proteins through the BIR repeats in addition to caspases (Lacasse et al., 1998). An attractive hypothesis is that some of these proteins have proapoptotic functions that do not involve disrupting IAP-caspase interactions, and that death preventing IAPs are able to suppress apoptosis induced by their expression by binding and sequestering them, preventing access to their targets. Candidates for such proteins in Drosophila are RPR, HID, GRIM and DOOM (Harvey et al., 1997), as well as uncharacterized molecules that mediate normally occurring apoptotic cell death that occurs independently of expression of genes in the H99 interval (White et al., 1994; Foley and Cooley, 1998) (Figure 7). DIAP 1, and by extension other IAPs, may function to block death at multiple steps: at an upstream point free DIAP 1 may titrate apoptosis inducers away from their targets, DIAP 1caspase complexes and components of other effector pathways, while at a downstream point it inhibits caspase activation or activity induced by these proteins (Figure 7). Concluding remarks IAPs are the only cellular caspase inhibitors identified to date. We showed that RPR, GRIM, HID and HID fragments sufficient to cause the death of insect cells block DIAPl 's ability to inhibit caspase activity. These observations provide a mechanism for RPR, HID and GRIM action and define a new point at which caspase activity can be regulated. The finding that the DIAPl loss-of-function phenotype involves cell death associated with 110 increased caspase activity strongly suggests that the interactions we described between RPR, HID, GRIM, DIAPI and caspases are physiologically important. Since most mammalian cells constitutively express caspases sufficient to carry out cell death (Weil et al., 1996), the mechanism of cell death activation defined by these proteins may be quite general. A prediction of our model of RPR, HID and GRIM function is that their ability to kill will depend on the status of a cell's caspases. In cells in which caspases are activated, or spontaneously undergoing autoactivation at some level, their expression may be sufficient to kill. However, in other cells in which caspase activity is either absent or more tightly regulated, they may function primarily to create a permissive condition for caspasedependent cell death. Thus, just as IAPs in their role as caspase inhibitors may function as cellular buffers that create a requirement for a certain level of caspase activation in order to induce apoptosis, so RPR-, HID- and GRIM-like proteins may function to titrate this buffering activity so that cells are more or less sensitive to caspase-dependent death signals. III Experimental Procedures Yeast Strain Constructions All yeast strains used in this study are in the W303a background (MATa, can1-100, leu23,-112, his3-11,-15, trp1-1 ,ura3-1, ade2-J). Construction of pGALL-(H/S3), pGALL(LEU2), and pGALL-DCP-1 was described previously (Hawkins et al., 1999). The p!O and prodomain-p20 fragments of drICE were amplified by PCR and inserted in tandem into pGALL-(LEU2) to make pGALL-rev-drICE. pAdh-(TRP I) was made by subcloning the promoter-terminator region from pHybZeo (Invitrogen) into pRS314. pCUPl-(IRP J) was constructed by replacing the GAL I promoter of pGALL-(IRP /) (Hawkins et al. , 1999) with the CUPl promoter. The DIAPI coding region was inserted into pAdh(TRP I) and pCUPl-(TRP J) to generate pAdh-DIAPl and pCUP-DIAPl. The HID coding region was amplified by PCR using primers that appended a C-terminal double polyoma epitope tag, and was inserted into pGALL-(H/S3) to generate pGALL-HID. pGALL-HID was digested with XhoI and religated to make pGALL-HIDLi (38-335). Cterminally polyoma tagged versions of HID initiating at codon 37 (HIDLi(l-36)) and HID lacking the first 335 amino acids (HIDLi (1-335)) were amplified by PCR and inserted into pGALL-(H/S3). The coding regions of RPR and GRIM were amplified by PCR and inserted into pGALL-(H/S3) to generate pGALL-RPR and pGALL-GRIM. Cell Death Assays 112 Overexpression of active caspases in yeast results in cell death that is inhibited by coexpression of caspase inhibitors (Hawkins et al., 1999; Kang et al., 1999). Because expression of RPR, HID or GRIM alone has no effect on yeast growth (below) we infer that the yeast growth suppression seen in cells that express caspases, DIAP 1 and RPR, HID or GRIM is due to cell death. To assay caspase-dependent cell death and protection by inhibitors, exponentially growing cultures of individual W303a transformants containing the relevant constrncts were serially diluted tenfold and spotted onto 2% glucose selective medium plates or 2% galactose and 1% raffinose (gal/raf)-inducing selective medium plates. Plates were incubated at 25°C for 12 h, then 30°C and photographed after 2 (glucose) or 3-5 (gal/raf) days. Expression and Purification of Recombinant Drosophila IAPs, Caspases and HID DCP-1 3 1-His 6 (DCP-1-His6) was prepared as described previously (Hawkins et al., 1999) and diluted into buffer A (50 mM Hepes [pH 7.5] , 100 mM NaCl , 1 mM EDTA, 0.1 % CHAPS, 10% sucrose, 5 mM DTT). The DIAPI coding region was amplified by PCR using primers that generated an N-terminal TEV Protease recognition site (ENL YFQG) and was introduced into the glutathione S-transferase (GST) expression vector pGEX4T-1 (Amersham Pharmacia). The GST-TEV-DIAPl fusion protein was expressed in Escherichia coli strain BL21 (DE3)pLysS and was affinity-purified on glutathioneSepharose. The bound fusion protein was cleaved with TEV Protease in buffer A. The coding region of HID was introduced into pET23a( +) (Novagen), generating a HID-His6 fusion. A similar procedure was used to generate a version of HID initiating at codon 37 (HID~ (l-36)-His6) . The first 37 amino acids of HID and the coding region of GST were 113 PCR amplified and introduced in tandem into pET23a( +) to generate HID(l-37)GSTHis6. GST was introduced into pET23a(+) to generate GST-His6. These vectors were introduced into BL21 (DE3)pLysS , protein expressed and affinity purified from the soluble fraction using ProBond resin (Invitrogen). Proteins were eluted and stored at -80°C in buffer D (50 mM Hepes [pH 8.0], 100 mM NaCl , I mM EDTA, 0 .1% CHAPS, 10% sucrose, 20 mM 2-mercaptoethanol , I% Triton X-100, 20% glycerol , 0.55 M imidazole). A caspase cleavage site (DQVQ) at residue 20 of DIAPl (S.L. Wang, unpublished) was altered to DQVE in GST-mycDIAPl (Hawkins et al. , 1999), generating GST-DIAPI . GST and GST-DIAPI were purified on glutathione sepharose beads. The mutated myc-DIAPI coding sequence was inserted into pMAL-c2 (New England Biolabs), generating MBP-DIAPI . MBP and MBP-DIAPI were purified on amylase resm. In vitro Protease Assays HID's ability to inhibit DIAPI was determined from caspase activity assay progress curves in which the release of 7-amino-4-trifluoromethyl-coumarin (AFC) from the synthetic substrate Ac-DEVD-AFC (100 µM) by DCP-1-His6 (0.2 nM) was measured in the presence of GST (0.2 µM) or DIAPl (0.1 µM) , and HID-His6 (0.44 µM), HID11 (136)-His6 (0.44 µM), or GST (0.77 µM). Caspase activity assays were performed in triplicate and prepared as follows: 2 µI DCP-1-His6 was incubated with 5 µI DIAPl or GST for 5 minutes at 25°C. 71 µI buffer X (70.4 mM Hepes [pH 7.5], 140.8 mM NaCl , 1.4 mM EDTA, 0.14% CHAPS, 14.1 % sucrose, 3.5 mM DTT, 5.6% glycerol, and 0.7 % Triton X-100) and 20 µI HID-His6 , HID11 (l-36)-His6, or GST-Hi s6 was subsequently added and the samples were further incubated 23 minutes. After addition of 2 µl 5 mM 114 Ac-DEVD-AFC in DMSO to each sample, caspase activities were assayed at 27°C using a fluorometric plate reader (Molecular Devices, fmax) in the kinetic mode with excitation and emission wavelengths of 405 and 510 nm, respectively. HID(l-37)GST-His6 activity, used at a concentration of 1.6µM, was tested in a similar manner. Caspase activity of embryos was monitored as follows : Oto 2 hr old embryos were collected and aged for 3 hr at 25°C. Embryos were dechorionated with 50% bleach, rinsed, suspended in an equal volume of buffer E (buffer A containing 0.5% Triton X-100 and 4% glycerol), and homogenized. 1 µI of this extract (approximately lOmg/ml) was diluted into 100µ1 of buffer E (buffer A containing 0.5 % Triton X-100 and 4 % glycerol) in the presence of 100 µM Ac-DEVD-AFC. Assays were performed in triplicate with freshly prepared extracts. Protein Binding Assays Purified recombinant proteins ( 150 ng each) were incubated with either OST beads, MBP beads, GST-DIAPl beads, or MBP-DIAPI beads (200 ng) for 30 minutes on a rotator at 25°C in buffer C (100 mM Hepes [pH 7.5], 200 mM NaCl , 2 mM EDTA, 0 .2% CHAPS , 20% sucrose, 40 mM 2-mercaptoethanol, 1% Triton X-100, 8% glycerol). Beads were then washed three times for five minutes with 1 ml buffer D lacking imidazole. Beadprotein complexes were subjected to 15% SDS-PAGE and immunoblotting. Proteins were detected with an anti-His antibody (Qiagen) and HRP-conjugated secondary antibodies followed by ECL (Amersham). HID-DCP-1 Competition Assay 115 MBP beads or MBP-DIAPl beads (0.5 µg) were mixed with 1, 5, or 20 µg GST-His6 or HID(l-37)GST-His6 for 1 hat 4°C in buffer A. 5 µg DCP-1-His6 was added and the samples further incubated ten minutes. After extensive washing the bead-protein complexes were subjected to 15% SDS-PAGE and immunoblotting. Proteins were detected with an anti-His antibody and HRP-conjugated secondary antibodies followed by ECL. Drosophila Strains A w- Oregon R strain was used as wildtype. The deficiency DJ(3L)brml 1 (72A3-4; 72D 15) uncovers DIAPl. DIAPI mutants , alleles of the thread (th) locus, and DJ( 3L)brml 1 were kept as balanced stocks over a version of the TM3 balancer that carries a ftz::lacZ transgene (TM3[ftz::lacZ]). th5 is a strong loss of function mutation in DIAPl since Df(3L)brml 1 shows a GMR-rpr enhancer phenotype similar to that of th5 (Hay et al. , 1995), and the embryonic lethal phenotype of th5/ Df(3L)brml 1 (this work, data not shown) is similar to that of th5/th5. By these same criteria, th6 is a much weaker allele. Cell Death Assays and Microscopy Embryos were collected on yeasted apple juice plates. For observation of living embryos, syncytial blastoderm stages were selected under halocarbon oil 27 (Sigma) and mounted individually on a slide using 10 mm coverslips. Images were taken on a Zeiss Axiophot, from sets of 6 embryos per experiment every 15 min at 25°C. For scanning electron microscopy staged collections of embryos were processed for staining with anti-~-galactosidase (~-gal) antibodies (Cappel) as described elsewhere 116 (Muller and Wieschaus, 1996). Homozygotes were isolated under the dissecting scope; the remaining embryos served as controls. The embryos were then processed for scanning electron microscopy as described in Muller and Weischaus (1996). Samples were viewed under a Hitachi S800 SEM. TUNEL labeling was performed using an in situ Cell Death Detection kit (Roche). Embryos were fixed as described above, rinsed in PTX (PBS , 0.5% Triton X-100), and immunolabeled with antibodies against ~-gal (Cappel) and Cy3-conjugated goat anti rabbit (Jackson) secondary antibodies. Embryos were then incubated in 100 rnM sodium-citrate, 0.1 % Triton X-100 at 65°C for 30 min, followed by rinsing twice in PTX and twice in TUNEL dilution buffer (Roche). Embryos incubated in TUNEL assay buffer (Roche) (30 min to 2 hrs) were supplemented with TUNEL enzyme (Roche) and the embryos were incubated a further 2 to 3 hrs. Embryos were rinsed in PTX and mounted in MOWIOL (Polysciences) containing DABCO (Sigma). Images were taken on a Leica TSC-NT confocal microscope. A range of 50 µm in depth was scanned in 16 optical sections and projected into one focal plane using the TCS-NT software. Acri dine orange staining was carried out as described in Hay et al. , (1994 ). Confocal and Nomarksi images were taken on the confocal microscope. 117 Acknowledgments We thank R. Deshaies and members of his lab for advice with protein binding and yeast experiments. We also thank R. Deshaies, P. Sternberg and M . Guo for comments on the manuscript. C. J. H. is supported by a Human Frontiers postdoctoral fellowship. B. A. H. is a Searle Scholor. This work was supported by grants to B. A.H. from the Gustavus and Louise Pfeiffer Research Foundation, a Burroughs Wellcome Fund New Investigator Awards in the Pharmacological Sciences, the Ellison Medical Foundation and National Institutes of Health Grant GM057422-0l. 118 References Abrams, J.M., White, K., Fessler, L.I., and Steller, H. (1993) . Programmed cell death during Drosophila embryogenesis. Development 117, 29-43 . Alnemri, E. S. , Livingston, D. J., Nicholson, D. W., Salvesen, G., Thornberry, N. A. , Wong, W.W., and Yuan , J. (1996). Human ICE/CED3 protease nomenclature. Cell 87, 171. Avdonin , V., Kasuya, J., Ciorba, M.A., Kaplan, B., Hoshi , T ., and Iverson, L. (1998). Apoptotic proteins Reaper and Grim induce stable inactivation in voltage gated K+ channels. PNAS 95, 11703-11708. Bergmann, A., Agapite, J., McCall , K. , and Steller, S. (1998a). The Drosophila gene hid is a direct molecular target of Ras-dependent survival signaling. Cell, 95, 331-341. Birnbaum, M. J., Clem, R. J., and Miller, L. K. (1994). An apoptosis-inhibiting gene from a nuclear polyhedrosis virus encoding a peptide with Cys/His sequence motifs. J. Virol. 68, 2521-2528. Cavaliere, V., Taddei, C., and Gargiulo, G. (1998). Apoptosis of nurse cells at the late stages of oogenesis of Drosophila melanogaster. Dev. Genes and Evol. 208, 106-112. Chen, P., Nordstrom, W., Gish, B., and Abrams, J.M. (1996). grim , a novel cell death gene in Drosophila. Genes Dev. 10, 1773-1782. 119 Chen, P., Rodriguez, A., Erskine, R., Thach, T., and Abrams, J.M. (1998). Dredd, a novel effector of the apoptosis activators reaper, grim and hid in Drosophila. Dev. Biol. 201, 202-216. Crook, N. E. , Clem, R. J. , and Miller, L. K. ( 1993). An apoptosis-inhibiting baculovirus gene with a zinc finger-like motif. J. Virol. 67, 2168-2174. Cryns, V. and Yuan, J. (1998). Proteases to die for. Genes Develop. 12, 1551-1570. Deveraux, Q.L. and Reed, J.C. (1999). IAP family proteins--suppressors of apoptosis . Genes and Develop. 13, 239-252. Deveraux, Q. L., Roy, N., Stennicke, H. R., Arsdale, T. V., Zhou, Q., Srinivasula, S. M., Alnemri, E . S., Salveson, G. S., and Reed, J. C. (1998). IAPs block apoptotic events induced by caspase-8 and cytochrome c by direct inhibition of distinct caspases. EMBO J. 17, 2215-2223. Deveraux, Q. L., Takahashi, R., Salvesen, G. S., and Reed, J. C. ( 1997). X-linked IAP is a direct inhibitor of cell-death proteases. Nature 388, 300-304. Dorstyn, L., Colussi , P.A., Quinn, L.M. , Richardson, H., and Kumar, S. (1999). DRONC, an ecdysone-inducible Drosophila caspase. Proc. Natl. Acad. Sci. USA 96, 4307-4312. Duckett, C.S., Nava, V.E., Gedrich, R.W ., Clem, R.J. , Van Dongen, J.L., Gilfillan , C., Shiels, H. , Hardwick, J. M., and Thompson, C.B. (1996) . A conserved family of 120 cellular genes related to the baculovirus iap gene and encoding apoptosis inhibitors. EMBO J. (1996). 15, 2685-2694. Evans, E.K., Kuwana, T. , Strum, S.L., Smith, J.J., Newmeyer, D.D. , and Kornbluth , S. (1997) . Reaper-induced apoptosis in a vertebrate system. EMBO J. 16, 7372-7381. Foley, K. and Cooley, L. (1998) . Apoptosis in late stage Drosophila nurse cells does not require genes within the H99 deficiency. Development 125, 1075-1082. Fraser, A. G. and Evan, G. I. (1997). Identification of a Drosophila melanogaster ICE/CED-3-related protease, drICE. EMBO J. 16, 2805-2813 . Fraser, A.G., McCarthy, N.J. , and Evan , G.I. (1997). drICE is an essential caspase required for apoptotic activity in Drosophila cells . EMBO J. 20, 6192-6199. Grether, M.E., Abrams, J.M., Agapite, J., White, K. , and Steller, H. (1995). The head involution defective gene of Drosophila melanogaster functions in programmed cell death . Genes and Develop. 9, 1694-1708. Hawkins, C.J., Wang, S.L., and Hay, B.A. (1999). A cloning method to identify caspases and their regulators in yeast: Identification of Drosophila IAPl as an inhibitor of the Drosophila caspase DCP- 1. Proc. Natl. Acad. Sci. USA 96, 2885-2890. Hay, B. A., Wolff, T. , and Rubin, G. M. (1994). Expression of baculovirus p35 prevents cell death in Drosophila . Development. 120, 2121-2129. 121 Hay, B. A., Wassarman, D. A. , and Rubin , G. M. (1995). Drosophila homologs of baculovirus inhibitor of apoptosis proteins function to block cell death. Cell. 83, 12531262. Inohara, N., Koseki , T., Hu, Y. M., Chen, S. , and Nunez, G. (1997). CLARP, a death effector domain-containing protein interacts with caspase-8 and regulates apoptosis. Proc. Natl. Acad. Sci. USA 94, 10717-10722. Kaiser, W. J., Vucic, D., and Miller, L. K. ( 1998). The Drosophila inhibitor of apoptosis D-IAPl suppresses cell death induced by the caspase drICE. FEBS Letters 440, 243-248. Kang, J. K., Schaber, M. D., Srinivasula, S. M., Alnemri, E. S. , Litwack, G., Hall, D. J., and Bjornsti, M.A. (1999). Cascades of mammalian caspase activation in the yeast Saccharomyces cerevisiae. J. Bio. Chem. 274, 3189-3198. Lacasse, E. C., Baird, S. , Korneluk, R. G. , and MacKenzie, A. E. ( 1998). The inhibitors of apoptosis (IAPs) and their emerging role in cancer. Oncogene 17, 32473259. McCall, K. and Steller, H. ( 1998). Requirement for DCP-1 caspase during Drosophila oogenesis. Science 279, 230-234. Muller, H. A. and Wieschaus, E. ( 1996). Armadillo, bazooka, and stardust are critical for early stages in formation of the zonula adherens and maintenance of the polarized blastoderm epithelium in Drosophila. J. Cell. Biol. 134, 149-63. 122 Nicholson, D. W. and Thornberry, N. (1997) Caspases: killer proteases. Trends Biochem. Sci. 22, 299-306. Raff, M.C. (1992). Social controls on cell survival and cell death. Nature 356, 397-400 Roy, N., Deveraux, Q.L., Takahashi, R., Salvesen, G.S., and Reed, J.C. (1997). The cIAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16, 69146925. Salvesen, G. S. and Dixit, V. M. (1997). Caspases: Intracellular signaling by proteolysis. Cell 91, 443-446 Song, Z. W ., McCall, K., and Steller, H. (1997). DCP-1, a Drosophila cell death protease essential for development. Science 275, 536-540. Srinivasula, S. M., Ahmed, M., Macfarlane, A. M., Luo, Z. W ., Fernandes-Alnemri, T., and Alnemri, E. S. (1998). Generation of constitutively active recombinant caspases-3 and -6 by rearrangement of their subunits. J. Biol. Chem. 273, 10107-1011 I. Takahashi, R. , Deveraux, Q., Tamm, I., Welsh, K., Assa-Munt, N., Salvesen G., and Reed, J. C. (1998). A single BIR domain of XIAP sufficient for inhibiting caspases. J. Biol. Chem. 273, 7787-7790. Thornberry, N. A. and Lazebnik, Y. (1998). Caspases: enemies within. Science 281, 1312-1316. 123 Thress, K., Henzel, W., Shillinglaw, W., and Kornbluth , S. (1998). Scythe: a novel reaper-binding apoptotic regulator. EMBO J. 17, 6135-6143. Uren, A.G., Pakusch, M., Hawkins, C. J., Puls, K. L., and Vaux, D. L. (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. Uren, A. G. , Coulson, E. J., and Vaux, D. L. (1998). Conservation of baculovirus inhibitor of apoptosis repeat proteins (BIRPs) in viruses, nematodes, vertebrates and yeasts. TIBS. 23, 159-162. Vucic, D., Kaiser, W. J., Harvey, A. J., and Miller, L. K. (1997) Inhibition of Reaperinduced apoptosis by interaction with inhibitor of apoptosis proteins (IAPs). Proc. Natl. Acad. Sci. USA 94, 10183-10188. Vucic, D., Kaiser, W. J., and Miller, L. K. (1998). Inhibitor of apoptosis proteins physically interact with and block apoptosis induced by Drosophila proteins HID and GRIM. Mo!. Cell. Biol. 17, 667-676. Weil, M., Jacobson, M. D ., Coles, H. S. R., Davies, R. L., Gardner, R. L., Raff, K. D., and Raff, M. C. (1996). Constitutive expression of the machinery for programmed cell death. J. Cell Biol. 133, 1053-1059. White, K., Grether, M. E., Abrams , J.M. , Young, L., Farrell, K. and Steller, H. (1994) . Genetic control of programmed cell death in Drosophila. Science 264, 677-683. 124 White, K., Tahaoglu, E., and Steller, H. (1996). Cell killing by the Drosophila gene reaper. Science 271, 805-807. Wing, J. P., Zhou, L., Schwartz, L. M., and Nambu, J. R. (1998) . Distinct cell killing properties of the Drosophila reaper, head involution defective, and grim genes. Cell Death and Diff. 5, 930-939. Wyllie, A. H., Kerr, J. F. R., and Currie, A. R. (1980). Cell death: the significance of apoptosis. Int. Rev. Cytol. 68, 251-303. Zhou, L., Schnitzler, A., Agapite, J., Schwartz, L.M., Steller, H., and Nambu, J.R. (1997). Cooperative functions of the reaper and head involution defective genes in the programmed cell death of Drosophila central nervous system midline cells. Proc. Natl. Acad. Sci. USA 94, 5131-5136. 125 CHAPTER 4 Conclusion: Future Directions 126 Studies with Drosophila proteins in a yeast assay system and by in vitro reconstitution experiments have led to the isolation of a caspase inhibitor and the elucidation of a general mechanism by which it may modulate apoptosis. This work defines a novel point at which caspase activity can be regulated: caspases may be activated by the direct inhibition of a caspase inhibitor. In support of this proposed mechanism, we find that elimination of DIAPl function results in global early embryonic cell death and a large increase in DIAPl-inhibitable caspase activity, and thatDIAPl is still required for cell survival when expression of rpr, hid, and grim is eliminated. Among many questions that arise from this work are the following: I) What is the nature of the interaction between IAPs and pro- and mature caspases, and which domains or residues in DIAPl facilitate interaction with HID? 2) Can these regions be altered such that DIAPl still functions as a caspase inhibitor yet is rendered insensitive to regulation by RPR, HID, and GRIM? 3) What is the mechanism of full-length HID action? 4) At what level(s) of caspase activation are IAP/HID acting? 5) Although HID can activate caspases by inhibiting DIAPl from blocking their activity in yeast and in vitro, does this in fact occur in vivo? 6) Are there other molecules involved in the "double inhibition" model of caspase activation? 7) On a more general level, what regulates RPR, HID, and GRIM? 8) Do these three activators have distinct cell-killing activities? 9) How do they activate the Apaf1 homologue Dark? 10) Do the IAPs have a role in the Dark/Apaf-1 pathway? 11) Do sequence and/or functional RPR, HID, and GRIM homologues exist in mammals? A few key issues regarding questions 1-7 and 11 will be addressed in this essay. A number of different approaches may be taken to obtain information on IAPcaspase and IAP-HID interactions, including mutational analysis, structural studies, and genetic screens performed in yeast or Drosophila. Solution structure and mutational analysis of IAP BIR domains indicate that particular linker sequences flanking the BIR domains are important for IAP-caspase interactions (Hinds et al. , 1999; Sun et al., 1999). 127 Mutational analysis of IAPs, HID, and caspases should help to identify key residues mediating their interactions. Once minimal interaction domains are delineated by such studies, x-ray crystallography and solution structure analysis of these domains may provide information on the structural basis for their activity. As an alternative approach, one can utilize yeast-based screens using libraries of mutagenized molecules (i.e., DIAPI, HID, etc.) to identify residues required for IAP-HID interactions. Drosophila screens have uncovered a variety of DIAP mutants that exhibit varied phenotypes (K. White, unpublished data). Analysis of DIAPI point mutants isolated as modifiers of HIDinduced death but not RPR- or GRIM-mediated apoptosis is likely to provide information on the nature of HID-IAP interactions. Although we have demonstrated that a fusion protein containing the first 37 amino acids of HID is sufficient to inhibit DIAP 1 without displacing DCP-1, it is unknown whether the full-length protein behaves similarly. One could carry out binding/competition experiments with full-length HID to determine if it, too, fails to displace caspase from IAP. It is possible that full-length HID interferes with DIAPl inhibition of caspase by both inducing a conformational change in the IAP, as well as catalyzing the release of caspase from IAP. BIAcore analysis measures, in real time, the association and dissociation of unlabeled ligand to an immobilized receptor, or vice versa, by changes in the adjacent refractive index. BIAcore technology can be used to examine the kinetics of caspase-IAP-HID complex formation . Generation of a version of HID which can be cleaved after its first 37 amino acids would allow one to test the functionality of a 37-mer peptide of HID. Is the 37-mer sufficient for IAP inhibition, or does activity require fusion to a larger molecule? IAPs have the ability to inhibit both pro- and mature forms of certain caspases. In some cases, IAP can bind to and inhibit only the zymogen and not the active caspase. For example, while DIAPI can inhibit active DCP-1, it appears to block DRICE-induced 128 apoptosis by inhibiting zymogen processing at a particular cleavage step (C. Hawkins & S. L. W. , unpublished data; L. Miller, pers. comm.). The stage at which IAPs block caspase activity has implications for HID' s mode of action. In situations where IAPs directly inhibit mature caspases, HID relieves IAP inhibition of already processed activated caspase. In other cases, HID may liberate inhibitor-bound procaspases which subsequently undergo processing via other caspase activation pathways. In Drosophila, the role of particular caspases in apoptotic pathways is currently unknown. The Drosophila caspases DCP-1 , DREDD/DCP-2, DRICE, DRONC, and DECAY all have caspase-like activation and/or proteolytic activity in experimental settings, but their order of activation in proteolytic cascades has not been determined (Chen et al., 1998; Dorstyn et al., 1999; Fraser and Evan, 1997; Song et al., 1997). DREDD and DRONC are candidate initiator caspases, since their prodomains contain a DED and CARD, respectively. DCP-1 and DRICE are structurally and biochemically similar to CED-3, indicating that they may be effector caspases. Loss of zygotic DCP-1 function results in melanotic tumor development and larval lethality (Song et al., 1997). DCP-1 appears to play a role during nurse cell death (McCall, 1998). During oogenesis, nurse cells transfer their cytoplasmic contents to the developing oocytes and then die. Females carrying homozygous dcp-J· clones in their germline are sterile and their nurse cells show defects in cytoskeletal reorganization and nuclear breakdown that occurs during the transfer process. Normal nurse cells transfer the majority of their cytoplasm and presumably, active DCP-1, to the oocyte just prior to dying. However, the oocytes escape damage, suggesting the presence of a protective mechanism, feasibly IAP activity. DREDD may also participate in the execution of nurse cell death, as the accumulation of dredd mRNA in nurse cells coincides with nurse cell death (Chen et al., 1998). Dredd mRNA also accumulates in embryonic cells undergoing apoptosis. DREDD 129 has a unique active site containing a glutamic acid in the position typically occupied by a glycine in the consensus sequence for caspases (QACXG). Interestingly, studies with DRONC indicate that it has a unique substrate cleavage ability in that it can cleave after glutamate residues (C. Hawkins, pers. comm.). DRONC's catalytic cysteine residue is contained within a unique sequence, PFCRG, which may contribute to DRONC's ability to recognize and cleave glutamate residues. DIAPI can inhibit DRONC-induced yeast cell death, suggesting a role for the IAP as an upstream inhibitor of caspase cascade activation (C. Hawkins, pers. comm.). The expression patterns and levels of RPR, HID, and GRIM, DIAPl, and caspases in the developing Drosophila embryo are currently unknown . Once the stoichiometric ratios at which HID inhibits DIAPI from blocking caspase activity in vitro have been determined, immunohistochemical analysis could be carried out to verify whether or not these proteins are present at the appropriate amounts within cells at the time of apoptosis. Coimmunoprecipitation of IAP-caspase and HID-IAP-caspase complexes from Drosophila would support the hypothesis that IAP acts as a direct caspase inhibitor and HID is an IAP inhibitor in vivo. These types of experiments would help to clarify if HID inhibits DIAPI activity in vivo, as it can in yeast and in vitro. In vivo activation of the apoptotic program by HID inhibition of DIAPl is likely to be more complicated than the simple interaction of these three components. What regulates HID? Does DIAPI require any cofactors for maximal activity? Hid function is negatively coupled to Ras activation at both the RNA and protein levels (Bergmann et al., 1998; Kurada and White, 1998). Hyperactivation of the Ras-MAP kinase pathway downregulates Hid expression and blocks Hid-induced cell death. Site-directed mutation of MAPK consensus sites in Hid renders its killing properties insensitive to MAPK. Phosphorylation could inactivate HID in a number of ways which await evaluation. An attractive hypothesis is that phosphorylation induces a conformational change in HID 130 resulting in occlusion of DIAPl binding sites. Alternatively, phosphorylation may mark HID for proteolytic degradation. Yet another hypothesis is that phosphorylation may allow HID inhibitors to recognize, bind, and inhibit it. Similarly, DIAPl may be regulated by cofactors prior to association with DCP-1 or HID, or in the complex itself. Although the in vitro experiments demonstrate that no other cofactors are required for DIAPl inhibition of caspase activity, these results do not exclude the possibility that IAPs may have additional properties that modify caspase activity. Since many IAPs contain RING fingers, it is conceivable that they play a role in modulating protein levels via ubiquitination. An attractive possibility is that IAPs target proapoptotic molecules, such as caspases, for degradation. Conversely, IAPs may target themselves for degradation upon receipt of appropriate death stimuli. In looking for additional regulatory molecules, the Drosophila model system provides a sophisticated range of tools for gene discovery and has already proven very useful in isolating cell death regulators. Yeast-based screens described in Chapter 2 provide a way of looking for caspase regulators strictly based on their protease activity. To date, neither sequence nor functional mammalian homologues of RPR, HID, or GRIM have been identified. As these molecules are potent apoptosis inducers in vertebrate systems, it is conceivable that homologues exist (Evans et al., 1997; Raining et al., 1999; McCarthy and Dixit, 1998). Furthermore, the IAP-caspase interaction is conserved across species and is likely to have shared regulatory motifs. The yeast-based assay system described in this thesis could be modified and employed to search for mammalian RPR, HID, and GRIM-like molecules on functional grounds rather than by sequence homology. A very attractive candidate IAP to target in such a screen is the human IAP survivin. Notably, survivin is expressed in a high proportion of the most common human cancers but not in normal, terminally differentiated adult tissues (Ambrosini et al., 1997). 131 Antisense-mediated reduction in survivin expression effected apoptosis in some tumor cell lines. Expression of survivin is enhanced at the G2/M phase of the cell cycle, and survivin protein associates with microtubules of the mitotic spindle during metaphase. It has been proposed that survivin may act to suppress a default death pathway that is activated during a G2/M checkpoint (Li et al., 1998). Survivin has been shown to suppress apoptosis induced by Fas, Bax, Taxol or overexpression of procaspases-3 , 7, or 9 in cultured cells, and it can be coimmunoprecipitated with the processed forms of these caspases (Ambrosini et al., 1997; Tamm et al., 1998). These results imply that survivin is a caspase inhibitor. With the sequencing of the Drosophila genome complete, fly homologues of mammalian genes are being identified at a breakneck pace. Not surprisingly, among these are several Drosophila homologues of mammalian and nematode genes encoding cell death regulators such as caspases, CED-4/Apaf-1 and Bcl-2-like proteins. The identification of these molecules in a number of organisms underscores the conserved nature of the death program and indicates the likelihood of the existence of RPR-, HID-, and GRIM-like death activators in other species in addition to Drosophila. 132 References Ambrosini, G., Adida, C., and Altieri, D. C. (1997). A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nature Medicine 3, 917-921. Bergmann, A., Agapite, J. , McCall, K., and Steller, H. (1998) . The Drosophila gene hid is a direct molecular target of Ras-dependent survival signaling. Cell 95, 331-341. Chen, P., Rodriguez, A., Erskine, R., Thach, T., and Abrams, J.M. (1998). Dredd, a novel effector of the apoptosis activators Reaper, Grim, and Hid in Drosophila. Dev. Biol. 201, 202-216. Dorstyn, L. , Colussi, P. A., Quinn, L. M., Richardson, H., and Kumar, S. (1999). DRONC, an ecdysone-inducible Drosophila caspase. Proc . Natl. Acad. Sci. USA 96, 4307-4312. Evans, E. K., Kuwana, T., Strum, S. L., Smith, J. J., Newmeyer, D. D., and Kornbluth, S. (1997). Reaper-induced apoptosis in a vertebrate system. EMBO J. 16, 7372-7381. Fraser, A. G., and Evan, G. I. (1997). Identification of a Drosophila melanogaster ICE/CED-3-related protease, drICE. EMBO J. 16, 2805-2813. Haining, W. , Carboy-Newcomb, C., Wei, C., and Steller, H. (1999). The proapoptotic function of Drosophila Hid is conserved in mammalian cells. Proc. Natl. Acad. Sci. USA 96, 4936-4941. 133 Hinds, M. G., Norton, R. S., Vaux, D. L., and Day, C. L. (1999). Solution structure of a baculoviral inhibitor of apoptosis (IAP) repeat. Nat. Struct. Biol. 6, 648-651. Kurada, P., and White, K. ( 1998). Ras promoted cell survival in Drosophila by downregulating hid expression. Cell 95, 319-329. Li, F., Ambrosini , G., Chu, E. Y., Plescia, J., Tognin, S., Marchisio, P. C., and Altieri , D. C. ( 1998). Cell cycle control of apoptosis and mitotic spindle checkpoint by survivin. Nature 396, 580-584. McCall, K. and Steller, H. ( 1998). Requirement for DCP-1 caspase during Drosophila oogenesis. Science 279, 230-234. McCarthy , J. V., and Dixit, V. M. (1998). Apoptosis induced by Drosophila reaper and grim in a human system. Attenuation by inhibitor of apoptosis proteins (cIAPs). J. Biol. Chem. 2 73, 24009-24015. Song, Z., McCall , K., and Steller, H. (1997). DCP-1, a Drosophila cell death protease essential for development. Science 275, 536-540. Sun, C., Cai, M., Gunasekera, A. H., Meadows, R. P., Wang, H., Chen, J., Zhang, H., Wu, W., Xu, N., Ng, S.-C., and Fesik, S. W. (1999). NMR structure and mutagenesis of the inhibitor of apoptosis protein XIAP. Nature 401, 818-822. 134 Tamm, I., Want, Y., Sausville, E., Scudiero, D. A., Vigna, N., Oltersdorf, T., and Reed, J. C. (1998). !AP-family protein survivin inhibits caspase activity and apoptosis induced by Fas(CD95), Bax, caspases, and anticacer drugs. Cancer Res. 59, 5315-5320. 135 Appendix I Proc. Natl. Acad. Sci. USA Vo l. 96, pp. 2885- 2890, March 1999 Ge net ics 136 A cloning method to identify caspases and their regulators in yeast: Identification of Drosophila IAPI as an inhibitor of the Drosophila caspase DCP-1 C HRISTINE J. H AWK INS* , SUSAN L. W ANG*, AND B RUCE A. H AY"I° D ivisio n of Biology, MC 15(l·29, Ca li for ni a Insti tut e of Tcdrnology, 1200 East Ca lifo rni a Bou lcvanl , Pasadena, CA 91125 Comm unicated by Alexander Varshavsky, California Institute of Technology, Pasadena, CA, Ja11ua1y 12, 1999 (received for review November 18, 1998) ABSTRACT Site-specific proteases play critical roles in regulating many cellular processes. To identify novel sitespecific proteases, their regulators , and substrates, we have designed a general reporter system in Saccharomyces cerevisiae in which a transcription factor is linked to the intracellular domain of a transmembrane protein by protease cleavage sites. Here, we explore the efficacy of this approach by using caspases, a family of aspartate-specific cysteine proteases, as a model. Introduction of an active caspase into cells that express a caspase-cleavable reporter results in the release of the transcription factor from the membrane and subsequent activation of a nuclear reporter. We show that known caspases activate the reporter, that an activator of caspase activity stimulates reporter activation in the presence of an otherwise inactive caspase, and that caspase inhibitors suppress caspase-dependent reporter activity. We also find that, although low or moderate levels of active caspase expression do not compromise yeast cell growth, higher level expression leads to lethality. We have exploited this observation to is olate clones from a Drosophila embryo cDNA library that block DCP-1 caspase-dependent yeas t cell death. Among these clones, we identified the known cell death inhibitor DIAPl. We showed, by using bacterially synthesized proteins, that glutathione S-transferase-DIAPI directly inhibits DCP-1 caspase activity but that it had minimal effect on the activity of a predomainless version of a second Drosophila caspase, drICE. li kely, given th e ea rl y stage of th e fie ld, th at more ro les ex ist. Cas pases and caspase regul a tors involved in th ese processes may be missed in scree ns t hat focus strictly on cell deat hrelated phenotypes. Thus, molecul es that possess caspase or cas pase regul atory act ivit y may not have bee n identifi ed yet. As an alte rn ative approach to identify ing novel cas pases or cas pase regul ators, it would be useful to have assays fo r cas pase fun ctio n th at are based strictly on protease activity. Because of th e importance of site-specific proteo lys is, we so ught to develop a ve rsatile system that would a llow the ide ntifi cati on of novel site-specific pro teases, regul ato rs of the activity of kn ow n site-specific proteases, or th eir substrates. Because caspase cleavage sites have been well defin ed, and activators and inhi bi tors of caspases have bee n ident ified, we set out to es tablish a pro totype system th at would allow positive selecti on for cas pase- li ke pro teases, th eir activators, and th eir inhibitors. Our app roach to ide nti fy ing th ese mol ecul es uses reporters for cas pase activit y th at function in living cells. Yeas t, th ough euk aryot ic, lacks many of th e speciali zed proteolytic sys te ms fo un d in cells of higher euk a ryo tes . T hu s, it co nstitut es an ideal backgro und in which to carry out fu nctionbased screens fo r these pro teases, th eir regul ators, and their ta rge ts. Reporte rs for the act ivity of specific proteases in bacteria and euk aryotes have been developed by using several strateg ies th at invo lve cleavage-dependent alterations in the act ivit y of specific pro teins (16-21). T o visualize caspase activity, we created a fus ion protein in which a transcri ption facto r is linked to th e int race llul a r do mai n of a transmemb rane protein by cas pase cleavage sites. Ex press ion of this prote in in yeast, in th e prese nce of an active caspase, should result in release of th e transcri pti on fac to r fr om th e membrane, fo llowed by t ra nscripti onal act ivat io n of a re porter. As described below, using such a repo rter sys tem, we ca n v isualize caspase activity in yeast and ca n identify pro teins th at act as caspase activators and inhi bitors. Caspase inhibito rs also can be iden tifi ed by virtue of th e ir abili ty to su ppress caspase overex p ression-de pe nd ent yeast cell deat h. Site-specific proteo lys is plays a criti cal ro le in reg ul ating a numb er of cellul ar processes. A n impo rt ant cl ass of sitespecific proteases are a gro up of cysteine proteases kn own as cas pases (1). Extensive ge neti c and biochemi cal evid ence indicates th at cas pases play ro les as cell death signaling and effector mol ecul es in a nu mbe r of different co ntexts, thu s making th em attractive potential therapeutic targets. Caspases identi fie d to date have been fou nd pri ma rily base d on homology to th e Caenorhabditis elegans cas pase CED-3 and mammalian caspase 1 and th ro ugh b iochemi cal purifica tion (reviewed in refs. 2-5). V iral and cellul ar activators and inhib ito rs of caspase fun cti on also have bee n identified in ge neti c and biochemical scree ns for reg ul ators of apop tos is (rev iewed in refs. 6- 8). These approaches to isolatin g cas pases and th e ir regul ators are limited by th e fact th at so me proteases that cleave a caspase ta rget site and th eir reg ul ato rs may not share primary sequ ence homology with th e pro te ins identifi ed to date o r th ey may be expressed only in specific tissues with limited availability fo r biochemical puri fica ti on. Furthe rm ore, it is clear th at cas pases reg ul ate processes oth er th an cell dea th , including cytokin e secreti on in mamm als (9 -1 3) and cell prolife rati on and ooge nesis in Drosophila ( 14, 15) . It see ms MATERIALS AND METHODS Constructs Yeast Strains. T he W303a strain (MA T a, canl- 100, leu23, -1 12, hisJ-1 J,-15, lrp 1-7, uraJ- 7, ade2-7 ) was used to monitor cas pase activit y by using t he lacZ repo rt e r sys tem. EGY48 (MAT a, ura3, trp1, his3, LexAop1,-LEU2) (Invit roge n ) was used to monitor caspase-dependent cell killing. Abb rev iat io ns: /3-gal, /3-ga lactos idase ; GST, glu tat hione S- lransferase; AFC, 7-amino-4-t riflu oro methyl-coumari n; BIR, bacu lovirus I AP re pea ts; gal/ raf med ium, medium co ntai ni ng 2% galactose a nd I % raffinose. *C. J.H . and S.L.W. co nt rib ut ed eq ua lly to this work. tTo whom repr int requests sho ul d be addressed . e-m ail: haybruce@ cco.ca ltech. ed u. T he publication costs of this art icle were defrayed in part by page charge payment. This art icle must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 so lely to indicate this fact. PNAS is ava ila ble o nlinc at www.pnas.o rg. 2885 2886 Genetics: H awk ins et al. Construction of Caspase Target Site Fusion Proteins. The repo rte r, CLBDG6, was generated by using PCR and stand ard techniques (details provided on requ es t). This prote in co nsists of, from N to C termini , amino acids 1-401 of a type 1 transmembrane prote in , hum an CD4 (22), a linker consisting of six tetrapeptide caspase ta rget sites that bracket th e specificity of known caspases and granzyme B (23)-DEVDGWEHDG-IEHDG-IETDG-DEHDG-DQMDG-, eac h of which is followed by a glyc ine residue, which acts as a stab ilizing residue in the N-end rul e degradation pathway in yeas t (rev iewed in refs. 24) , and fin ally, a transcription factor co ntaining th e LexA DNA binding domain (25). A seco nd construct, designate d CLBGG6, was ge ne rat ed that encodes a protein id entical to CLBDG6 exce pt that th e esse ntial Pl as partates of the six caspase cleavage sites are replaced with glycines, rendering th em nonfunction al. Construction of Yeast Expression Plasmids. Plasmids for ex pression of genes in yeast were derived from th e pRS se ri es (26). To express genes in yeast under galactose-inducibl e control, we used two GALI promoter fragments: a long version, extending from base 1-815, called GALL, and a short er and somewhat weaker version, extending from base 406-815, called GALS (27) . The yeast act in terminator, bases 2,107-2,490 (GenBank access io n no . L00026), was used fo r all constructs. GALL promoter a nd actin terminator fragments were insert ed into pRS313 (HIS3) , pRS314 (TRPI) , and pRS315 (LEU2) , ge ne rat ing pGALL-(H/SJ) , pGALL(TRPJ), and pGALL-(LEU2). CLBDG6 and CLBGG6 cod ing regions we re cloned into pGALL-(TRP/) . GALS promoter and actin terminator fragments were insert ed into pRS315 , generating pGALS-(LEU2). To exp ress genes und er th e control of the copper-inducibl e CUPl promote r, we used a promoter fragment extending from bases 1,079-1 ,533 of th e CUPl locus (GenBank access ion no. K02204). Site-directed mutagenesis was used to mutate th e CUP J promote r to prevent activation in response to glucose sta rvation (28). The mut ated CUPl promoter and act in ter minator fragments were inse rted into pRS315 (LEU2) , generating pCUP1-(LEU2). The coding region for the C. elegans caspase CED-3 (29) was introduced into pCUP1-(LEU2). Site-directed mut agenesis was used to generate an inact ive ve rsion of CED-3 in which th e active site cysteine was changed to se rin e (CED-3CS). Full length human caspase 7 (caspase 7FL) (30) , caspase 7 lack ing the N-terminal 53-aa prodoma in (caspase 753 ), and th e caspase 8 isoform correspo nding to MACHa2 (31) or Mch5 (32) (caspase 8FL) were introduced into pCUP1-(LEU2) . Full length DCPl (14) was introduced into pGALS-(LEU2) and pGALL-(LEU2). Site-directed mutagenes is was used to change the active site cys te ine to se rin e (DCP-lCS). The resulting full length cod ing region was introduced int o pGALL-(LEU2). Full length drICE (33) was inse rt ed into pGALL-(LEU2). Full length human caspase 9 (34, 35) was introduced into pGALL-(LEU2). The region encod ing amin o acids 1-530 of Apaf-1 (36) (Apaf-1 530 ) was introduced into pGALL-(H/SJ). Full lengt h DIAPl (FLDIAPl) and a version of DIAPl C-terminally trun cated following residu e 38 1 (DIAPlBIR) (37) were introduced into pGALL-(H/SJ). The mouse IAP MIHA (38) and baculovirus p35 (39) we re introduced into pGALL-(H/SJ ). Yeast Transformation and Characterization. Plasmids were introduced into yeast by lithium acetate transformation. For cas pase activity assays in which la cZ ex pression was monit ored, yeast were transform ed with pSH1 8-34 ( URA3) (Invitrogen), which carries a LexA-res po nsive /acZ ge ne. These cells th en were transformed with e ithe r pGALL-CLBDG6-(TRP/) o r pGALL-CLBGG6-(TRP/) . Caspase ex pression pl asmids or an empty expression vector were introdu ced into th ese backgrounds and were character ized as describ ed below. For cas pase activation and inhibition assays, a fourth pl as mid also was introduced, e ither ex pressing Apaf-1 530 , p35 , or MIHA o r 137 Proc. Natl. Acad. Sci. USA 96 ( 1999) with no insert. To carry out X-ga l filt er assays for /3-galactosidase (/3-gal) activit y, transfor m ants were plated on select ive plates with glucose (2%) as the sugar so urce. After 3 days, duplicate colonies were picked and res uspended in 1 ml of sterile Tris·EDTA (pH 8.0). One microliter of each sample was strea ked on a minimal med ium glucose plate. After 2 days, a nylon membran e was used to lift th e streake d yeast (yeast side upwards) onto a co mpl ete medium plate containing 2% galactose and 1 % raffinose (gal/raf media) and 3 µ,M copper sulfate. Afte r va rious periods of induction, th e filters were processed for X-gal sta ining (40). To quantitate /3-gal act ivity, three tubes of liquid selective gal / raf medium were inocul ated with single colonies from each transfo rm ation plate, were grown for 24 hr, then were diluted 1:10 into fresh gal/raf selective medium containing the indicated concentration of co pper sulfate and were grown for a further 10 hr. o -nitrophenyl-/3-0-galactoside assays were perform ed as describ ed by Miller (41) . To assay caspase-dependent cell death and protection by inhibitors, colonies carrying the relevant plasmids were streaked from 2% glucose selective media plates onto gal/raf selective me dia plates . The plates were photograp hed aft e r 3 days. Expression and Purification of Recombinant Drosophila IAPs and Caspases. The DIAPl coding region was a mplified by PCR using primers th at generated an N-terminal myc epitope (EQKLISEEDL) and was introduced into th e glutathione S-transferase (GST) ex press ion vector pGEX4T-1 (Amersham Ph armaci a). The GST-myc-DIAPl fusion protein was ex pressed in Escherichia coli strain BL21(DE3)pLysS (Novage n) a nd was aff inity-purifi e d on glutathioneSepharose. The eluted protein was dialyzed against buffer A (25 mM Tris, pH 8.0/50 mM NaCl/10 mM DTT). After di alysis, the protein was frozen in aliquots after addition of glycerol to 10%. DCP-1 , initiating at codon 31 (DCP-1 31 ), was introduced into pET23a( +) (Novagen), ge nerating a DCP-1 31-His6 fusion. A similar procedure was used to ge nerate a drICE-His6 fusion prote in in which DrICE initiates at codon 81 (drICE 81 -His6). DCP-1 31 -His 6 and drICE81 -His 6 we re ex pressed in the£. coli stra in BL21(DE3)pLysS. Prote in ex pression and affinity purification from the soluble fr action were carried out using ProBond res in (Invitrogen). Eluted protein was dia lyzed aga inst buffer A and subsequ entl y was snap frozen in buffer A co ntaining 10% glycerol. Drosophila cDNA Library Construction and DCP-1 Inhibitor Screening. Drosophila e mbryonic polyA + mRNA (CLONTECH) was converted into cDNA by using a Superscript cDNA synthesis kit (GIBCO). cDNAs larger th an 600 bp were ligated into a mod ified ve rsion of pGALL-(H/S3) in which the polylinke r was ex pand ed to co nt a in Xhol and Nati sites. ElectroMax DHlOB cells (GIBCO) we re transfor med with th e ligation mix and were used to amplify th e librar y, which contained 5 X 106 primary transfo rmants. W303a yeast ca rr ying pGALL-DCP-1-(LEU2) were transformed with 26 µ,g of librar y plasmid DNA, were grown in yeas t extract/ pe ptone/dextrose for 3 hr, were washed twice to remove glucose, and were plated on gal/raf selective plates. A total of 140,000 transfo rm ants were screened. Colonies were picked after 4 days of growth at 30°C. PCR was carried out on DNA isolated from individua l colonies by using DIAPlspecific prim e rs. In Vitro Protease Assays. D CP-1 31 -His6 and Dr!CE 81 -His6 cas pase activity were meas ured flu oro metrically by following the release of 7-amino-4-trifluoromethyl-coumarin (AFC) from Ac-DEVD-AFC (Enzyme Systems Products, Live rmore, CA) usi ng th e fm ax flu oresce nce microplate reader (Molecul a r Dev ices) with an excit ation wavelength of 405 nm and an emission wave length of 510 nm. The abi lit y of GST-DIAPl to inhibit cas pase act ivit y was det e rmin ed from caspase activit y assay progress curves in whi ch substr ate hydrolysis (100 µ,M ) G e netics: Hawkins et al. by DCP-1 3 I-His 6 (0.2 nM) or drICE 8 I-His 6 (0.62 nM) was measured in the presence of GST (0.48 µ,M) or GSTmycDIAPl (0.16 µ,M), in caspase act ivity buffer (50 mM Hepes, pH 7.5/100 mM NaCl/I mM EDTA/0.1 % 3-[(3chol amidopropyl )dim e thyl a mm on io ]-1-propan es u lfonat e / 10% sucrose/5 mM DTT). RESULTS AND DISCUSSION Our approach to monitoring caspase activity in vivo was to create cells in which caspase activity st imulates transc ription a l activation of a re porte r. W e created a fu sion protein substrat e for caspase cleavage in which th e transcription factor LexAB42 (LB) is link ed to th e truncated cytoplasmic domain of a membrane prote in, CD4 (C), by a short linker (DG6) consisting of six different caspase cl eavage sites that bracket th e specificities of known caspases and the serine protease granzyme B, which cleaves caspases and other targets at sites of similar sequence (Material and Methods for detai ls). When thi s mol ecule, referre d to as CLBDG6, is expressed in a re port e r strain in which a LexA-d e pe nd e nt promoter driv es la cZ express ion (LexA//3-gal reporte r) , leve ls of /3-gal activity should depend on th e presence of an active caspase a bl e to cl eave one or more of the introduced target s ites, th ereby re leas ing LexA-B42 from m e mbrane association. (Fig. 1 A and B). A Reporter for Caspase Activity in Yeast. We introduce d CLBDG6 into th e LexA/ /3-gal repo rter strain in a plas mid, pGALL-CLBDG6, in which ex press ion is induced in respo nse to galactose. We introdu ced into this background a copper- 138 Proc. Natl. Acad. Sci. USA 96 (1999) 2887 inducible exp ress ion pl asmid, pCUPl , containing eith e r no in se rt or differe nt ve rsions of th e caspase CED-3 . Transformants initially were streaked on glucose medium. Colonies from these streaks then were replica plated onto gal/raf me dium containing 3 µ,M copper to induce expression of CLBDG6 and from th e pCUPl plasmid. After 12 hr of induction, levels of /3-ga l activity were determ ined by using an X-gal assay in which cells that do not express /3-gal remain white whereas those that do turn sha des of blue. R eporter cells th a t ex pressed CLBDG6 a lon e remained white in this assay (Fig. 2A), ind icating th at yeast co nt a ins negligibl e amounts of prot eases capable of cleaving caspase target sites und er stand ard growth conditions. H owever, when expression of th e C. e/egans caspase CED-3 (pCUPl-CED-3) was induced , a high leve l of /3-gal activ ity was observ ed (Fig. 2A), which increased in a copper concentration-dependent manner (Fig. 2B). Of importance, caspase activity was req uired for reporter activation becaus e expression of an inactive CED-3 mutant in which the active site cyste ine had been changed to serine (CED-3CS) did not result in /3-ga l ex press ion (Fig. 2A). Finally, ex press ion of wild-type CED-3 in a reporte r strain in which the essential Pl aspartates of th e caspase target sites in CLBDG6 had been mut a ted to glycines (CLBGG6) (Fig. IC) did not resu lt in /3-gal activ ity (Fig. 2A), a rguing th at th e CED-3-dependent indu ction of /3-gal activity was a direct result of cleavage of CLBDG6 at the caspase target sites. Th ese resu lts estab lish that yeast can be used as a cell-based reporter syste m for cas pase activity. In order for a caspase to be identified in this assay, th e caspase mu st be active in yeast. Physiological activation of caspases occurs through multiple m echanisms, including recruitment and ol igomerization at the p lasma membran e, cleavage by caspases or othe r proteases ab le to recognize a caspase target site, interact ions with me mb ers of th e CED-4/ Apaf-1 family of proteins, and autoVl 50 B ...., A GG61--'-+-'-+-+ DG6 + + + ~ +--'-~ □ vector ::J 40 • CEO-3 CED-3 + + CED-31---+-'-+-~+--+- ; C-S ,___~-~-~--< \~ 4. 1 ~i~ :1 V " T ·c (I) V) ro 30 :-g V) 320 u . .. • •• C1l ro01 10 •• = 0 00.3 1.0 I 3.0 [C u] (µM) FIG. 1. A genetic system for mon itoring caspase activi ty in yeas t using a transcriptio nal reporter. Yeast were created th at express a chimeric type-1 transme mb rane protein (CLBDG6) in which the N-terminal signal sequence and transmembrane domain (CD4) is followed by a linker consisting of six tetrapeptide caspase targe t si tes (indicated in bold) that bracket the specificity of known caspases and gra nzy me B (23)-DEVDG-WEHDG-IEHDG-IETDG-DEHDGDQMDG-each of which is followed by a glycine residue, which acts as a stabilizing residue in th e N-end rule degradation pathway in yeas t (reviewed in ref. 24). C-terminal to th e caspase target site linker is a transcription factor dom ain , LexA-B42. The LexA-depend ent transcriptiona l reporter consists of LexA binding sites (LexA UAS) and a promoter (P) upstrea m of the bacterial lacZ gene (/acZ) (A). The cells in A act as caspase act ivit y reporters because express ion of an act ive caspase results in CLBDG6 cleavage at th e caspase target sites, releasing LexA-B42, which enters the nucleus and activates lacZ transcript ion (B). A versio n of CLBDG6 in which the Pl asparta tes are changed to glycines (CLBGG6) ca nn ot be cleaved by caspases. Ce lls expressing CLBGG6 act as fa lse positive reporters for molecules that act ivate lacZ ex pression independent of cleavage at caspase target si te (C). If cells in B express a caspase inhi bit or as well as an active caspase, caspase activity, and thus caspase-dependent release of LexA-B42, is inhibited, and /3-gal levels are decreased compared with cells th at exp ress th e caspase alone (D). FI G. 2. Yeast express ing CLBDG6 act as reporters for CED-3 caspase activity. W303a yeast were transformed with pSH18-34, which carries a LexA-respo nsive lacZ transcr iptional cassette (the LexA//3ga l reporter strain). T hese ce lls were transformed with pGALL express ion plasmids carryi ng CLBDG6 (DG6) or CLBGG6 (GG6). These cells also ca rry a copper-inducible pCUPl plasmid, which co ntains either wild-type CED-3 (CED-3), an inactive C to S mutant ve rsion of CED-3 (CED-3 C-S ), or nothing. Duplicate colonies from each transformation were streaked onto ga l/raf medium to induce GA LI-depend ent exp ress ion of the caspase substrates and then were lifted onto complete med ia plates with 3 µM copper sulfate to induce cas pase ex pression. After a 12-hr inducti on, an X-ga l assay was performed on the filter. Only ce lls express ing CLBDG6 and wild -type CE D-3 have signifi cant /3-gal activ ity (A). Cu ltures from three transformants carrying pSH1 8- 34, pGA L-CLBDG6, and either th e empty pCUP I vector or pCUP J-CED-3 we re grown to station ary phase, then we re diluted into medium containing th e indica ted levels of copper sulfate and grown for a furth er 10 hr. o-nitrophenyl-/3- 0-ga lactos ide assays were performed, and /3-ga l activity was determined. /3-gal activi ty in the CED-3-expressing cells in creased as a function of copper concentration (filled circles, CED-3). No /3-gal activity was found in the cultures carrying on ly the empty pCUPl vector (ope n boxes, vector) (B). 2888 G enetics: Hawkins et al. activation. In some cases, overexpression alone is sufficient to induce autoactivation whereas, in other cases, significant activation requires interactions with other proteins (reviewed in refs. 2-5). Thus, it is likely that only proteases in which th e primary translation product is active, or in which th e protease is able to autoactivate, will be identified in th e simplest reporter-base d caspase screen. Howeve r, more complex screens for caspases that can activate after forced oligomerization or association with potenti al caspase activators (42-46) can be envisioned. We have tes ted several other caspases in this reporter system. Expression of mammalia n caspase 7 53 (below) and full length caspase 8 ( data not shown) resulted in reporterdependent lacZ ex pression. Expression of human caspase 3, caspase 9, or Drosophila drICE failed to activate reporter expression ( data not shown), even though active forms of these caspases are known to efficiently cleave peptides with the sa me sequence as the target sites introduced into CLBDG6 (23, 33). Moreover, although overexpression of wild-type but not an inactive mutant of CED-3 induced yeast cell death (below), similar overexpression of caspase 3, caspase 9, or drICE had no effect on cell growth. Based on these observations, it is likely that, in yeast, the procaspase forms of these caspases do not autoprocess to generate active caspase heterodimers. This result is expected: Caspase 9 is thought to function as an upstream caspase, in which a major mechanism of activation requires association with Apaf-1 (42, 43), whereas caspase-3 is thought to act as a downstream caspase, in which a principal mechanism of activation is cleavage by other caspases (reviewed in refs. 2-5). drICE activation may be regulated by either of these mechanisms. Activators of Caspase-Dependent Reporter Activation. The fact that certain caspases do not activate in yeast suggests that it should be possible to screen for their activators as molecules th at induce report er exp ression in the prese nce of an otherwise inactive caspase. To demonstrate this, we carried out an experiment in which caspase activity was monitored in yeast that expressed full length caspase 9, alone or in combination with a fragment of Apaf-1 that is constitutively active with respect to caspase 9 processing activity in vitro (43) . Transformants of th e pGALL-CLBDG6 LexA//3-gal reporter strain were generated that carried either two empty vectors, an empty vector and pGALL-Apaf-1 530 , an empty vector and pGALLcaspase 9, or pGALL-caspase 9 and pGALL-Apaf-1 530 Transformants initially were streaked on glucose medium. Colonies from these streaks then were replica plated onto gal/raf medium to induce GALL-dependent expression. After 16 hr of induction, levels of /3-gal activity were determin ed by using an X-gal assay. Reporter cells th at expressed either nothing, caspase 9, or Apaf-1 530 alone remained white in this assay, indicating that caspase activity was not induced (Fig. 3). However, colonies that expressed both caspase 9 and Apaf-1 530 showed robust /3-gal activity, suggesting the occurrence of Apaf-1 530 -mediated activation of the otherwise inactive procaspase 9 (Fig. 3). We used /3-gal activity, as assayed after replica plating of colonies, as th e basis for our caspase reporter assay because caspases are expressed conditionally after replica plating. Thus, their identification is feasible even if their expression is toxic to cells. However, caspase activity screens also can be adapted to positive, survival-based screening assays by requiring LexAdependent expression of a yeast auxotrophic marker such as HIS3 or URA3. False positives in these caspase reporter assays could arise because introduced proteins bind to the LexA binding sites and activate transcription directly or because they encode proteases that cleave CLBDG6, but not at the caspase target sites. Both classes of false positives can be identified by the fact that th ey should still activate lacZ expression when introduced into a LexA//3-gal reporter strain that expresses CLBGG6, the false positive reporter strain (Fig. lC). 139 Proc. Natl. Acad. Sci. USA 96 (1999) DG6 + + + + Apat-1=>-----l-_...!+_+-+--t--,++,--Casp-91,__ _L_--r-1--'--'--'-- F1G. 3. Expression of Apaf-1 530 induces caspase 9-dependen t reporter activation . W3O3a containing pSH18-34 and pGALLCLBDG6 were transformed with pGALL plasmids to carry ei ther two empty pGALL vectors, Apaf-J 53o and an empty pGALL vector, caspase 9 and an e mpty pGALL vector, or Apaf-1 530 and caspase 9. Two colonies from each transformation were streaked onto selective glucose medium plates, were grown for several days, and then were replica plated onto comp lete gal/raf medium . After 16 hr , a n X-gal assay was perform ed. Only cells expressing both Apaf-1 530 and caspase 9 show significant /3-gal activity. Suppression of Caspase-Dependent Reporter Activation by Caspase Inhibitors. Once a reporter for the activity of a specific caspase has been established in yeast, it should be possible to screen for inhibitors of that activity by identifying cells that ex press the protease but in which reporter activity is repressed (Fig. lD). H ere, we demonstrate th e feasibility of this approach by showing th at two different families of caspase inhibitors, exe mplified by baculovirus p35 and th e murine IAP MIRA, suppress caspase-dependent reporter activation . p35 is a broad specificity caspase inhibitor in which inhibition is associated with cleavage (reviewed in ref. 6). The IAPs comprise a second class of caspase inhibitors distinct from p35 in which cleavage of the inhibitor does not play a rol e . IAPs originally were identified in the baculovirus system by virtue of their ability to substitute for baculovirus p35 as suppressers of viral infection-induced cell dea th (47). IAP homologs have now been found in other viruses, Drosophila , C. elegans, mammals and yeast and are characterized by one or more N-terminal repeats known as baculovirus IAP repeats (BIRs). Many also have a C-terminal RING fing er motif (reviewed in refs. 6-8 and 48). Several mammalian IAPs-XIAP, cIAPl, and clAP2-bind to and directly inhibit caspases-3 and -7 in vitro (49, 50). We introduced pGALL-CLBDG6 and either an empty pCUPl plasmid or a pCUP plasmid containing CED-3 or caspase 753 into the LexA/ /3-gal reporter strain along with either an empty pGALL expression vector, a pGALL ex pression vector carrying p35 , or a pGALL expression vector carrying MIRA (murine XIAP) . Transfo rmants were grown in galactose-containing me dium to induce ex pression of CLBDG6 and th e caspase inhibitor and then were transferred to medium containing galactose and 3 µ,M copper to induce ex pression of th e caspase. /3-gal activity was determin e d after a 10-hr copper induction. As shown in Fig. 4 A and B, ex pression of p35 inhibited both CED-3 and caspase 753 activity = 5-fold. A similar though so mewhat weaker ( =2-fold) inhibition of caspase 753 activity was see n in the presence of MIRA (Fig. 4B), indicat ing that caspase-IAP interactions can be detected in this assay. Caspase Overexpression-Dependent Yeast Cell Death. In th e above assay, th e presence of a caspase inhibitor is indicated by a decrease in /3-gal act ivity. However, in many situations, it would be useful if cas pase inhibition was coupled to a positive repo rte r output. A direct approach to identifying caspase inhibitors rests on th e observation that high level expression of active caspase causes yeast cell death. Low level expression of the Drosophila caspase DCP-1 from the induced CUP-1 promoter did not significantly compromise yeast cell growth. High er level expression from the GAL promoter, however, did 140 Genetics: Hawkins et al. A B 7 .t: 6 (1) (J') ..., • C ::::i (1) 5 (J') •• rt! :Q 4 (J') _,gJ 15'ui B 10 u 0 ..., 3 u ~ rt! 01 2 I ah. 1 - + ~ rt! 01 5 I •Casp-7 + + + p35 - + + - - I =o MIHA - • • • • • 0 CED-3 p35 • • • .§ 20 + + FIG. 4. Expression of caspase inhibitors suppresses caspasedependent reporter activation in yeast. The LexA/ /3-gal reporter stra in carrying pGALL-CLBDG6 was transformed with either an empty pCUPl plasmid or pCUPl-CED-3 and either an empty pGALL vector or pGALL-p35. Three colonies from each transformation were grown for 24 hr in selective gal/raf medium. Cultures were diluted 1:10 into fresh gal/ raf medium containing 3 µ,M copper sulfate and were grown for a further 10 hr, after which o-n itrophenyl-/3-n-galactoside assays for /3-gal activity were performed. Cultures from caspase transformants showed significant /3-gal activity, which was suppressed by GALL-dependent expression of baculovirus p35 (A). In an experiment simi lar to that described in A , expression of caspase 753 was induced in cells that express baculovirus p35 or the mouse !AP MIHA. Expression of caspase 753 resulted in a significant increase in cell ular /3-gal activity, which was suppressed by GALL-dependent express ion of p35 or MIHA (B). result in ce ll letha lity. Cells were ab le to grow on ga lactosecontaining media if they carried an empty pGALL expression vector but not if they carried a pGALL-DCP-1 expressio n construct (Fig. SA) . This effect of DCP-1 expression was attributab le to ce ll death because a > 250-fold decrease in the number of co lony forming units was seen when ce lls carrying the pGALL-DCP-1 express ion plasmid were grown for 12 hr in li qu id gal/ raf medium, thus inducing high leve l DCP-1 express ion, and the n were plated on glucose-contain ing medium (dat a not shown). Of importance, GALL- D CP-1expression-dependent cell killing depended on caspase act ivity because express ion of a n inactive mutant form of DCP-1 (DCP-1 C285S) did not cause ce ll death (Fig. SA) . Caspase- A 2889 mediated cell death in yeast may be a general phenomenon beca use other caspases that are active in yeast, including full length CED-3, ful l length caspase 8, and caspase 7 53 , block colony formation when ex pressed under control of the strong GALL promote r (data not shown). V1 25 (J') Proc. Natl. Acad. Sci. USA 96 (1999) B Caspase Inh ibitors Suppress Caspase OverexpressionDependent Cell Death. To de monstrate that caspase inh ibitors can be identified as prote ins that restore ce ll viabi lity to yeast express ing an active caspase, we carried out an experiment in wh ich th e broad specificity caspase inh ibitor p35 was coexpressed with full length DCP-1. We introduced pGALL expression plasmids that either had no insert, or that carried p35, into cells carrying a pGALS-DCP-1 plasmid. Cells from colon ies carry ing th ese plasmids were grown on glucosecontai ni ng med ia and then were streaked o nto gal / raf media to induce expression of the caspase and the potential inhibitor. As shown in Fig. 5B, pGALS-dependent expression of DCP-1 in the presence of an empty pGALL express ion vector resu lted in no cell growth. In contrast, coexpression of baculovirus p35 with DCP-1 resulted in a dramatic rescue of cell growth. To de term ine whether a yeast survival-based screen can be us ed to identify novel caspase in h ibitors, we transformed yeast carrying pGALS-DCP-1 with a Drosophila pGALL embryonic cDNA expression library and plated these ce lls on gal/ raf medium. In a screen of = 1.4 x 105 transformants, =50 posit ives were obtained. Th ese were tested by PCR, and all were found to correspond to DIAPl, wh ich was originally id e nt ified as an inhibitor of reaper or hid overexpressioninduced, caspase-dependent cell death in the fly eye (37). In the fly eye, and in cell cu lture (51-54), it also was found that cell death cou ld be suppressed by expression of an N-terminal DIAPl fragment containing the two BIR repeats but lack ing the C-term inal RING finger doma in . To determine whether the same fragment of DIAPl was sufficient to block DCP-1dependent cell ki llin g in yeast, we carr ied out an experiment in which full length DIAPl, an N -termina l fragment of DIAPl, or an empty vector was expressed under GALL control in the presence of GALS-driven DCP-1. As shown in Fig. 5B, inhibition of DCP-1-dependent cell death was seen when DIAPl or the N-terminal fragment of DIAPl containing only the DIAPl BIR repeats was coexpressed with DCP-1. To determine whether the obse rved interact ion be tween DIAPl and DCP-1 was direct, we generated bacterially synthesized GSTDIAPl as well as His 6 -tagged versions of prodoma inless DCP-1 (DCP-1 3 1-His6) and drI CE (drICE 81 -His 6 ). As shown in Fig. 6, GST-DIAPl inhib ited DCP-1 31 -His 6 caspase activity but had litt le if any effect on that of drICE 81 -His 6. Thus, these resu lts demo nstrate that caspase inhibitors can be identifie d as A DCP-1 B drlCE 0 .8 0.2 ~ 0 .6 ~ 0 .15 ~ ~ 0.4 FIG. 5. High level expression of the Drosophila caspase DCP-1 kills yeast and is prevented by coexpression of baculovirus p35 or DIAPl. EGY48 yeast were transformed with pGALL plasmids co ntaining either fu ll length DCP-1 (FLDCP-1), an active site C-S mutant of DCP-1 (DCP-1 C-S), or no insert (vector). Transformants were streaked from se lective glucose-containing medium onto selective gal/raf-inducing med ium. Cells expressing either an empty pGALS vector or the DCP-1 C-S active site mutant grow on ga lactosecontaining medium whereas cells express ing full length DCP-1 do not (A). EGY48 yeast carrying pGALS-FLDCP-1 were transformed with pGALLvectors carrying full length DIAPl (DIAPl), the DIAPl BIR repeats (DIAPl BIR), baculovirus p35 (p35), or nothing (vector). GALL-dependent expression of p35, full length DIAPl, and, to a somewhat lesser extent, the DIAPl BIR repea ts block DCP-1dependent cell death (B). ~ 0.1 0 ~ 0 .05 0 0 .2 0 0 0 20 40 60 Time (minutes) 0 20 40 60 Time (minutes) FI G. 6. GST-DIAPl inhibits the caspase activity of bacterially synthesized DCP-13 1-His6 , but not drICE 8 1-HIS 6 . Purified GSTDIAPl (0.16 µ,M) (open triangles) or GST (0. 48 µ,M) (open squares) was incub ated with a fixed amount ofDCP-1 31 His 6 (0.2nM) in caspase activity assay buffer containing 100 µ,M of the Ac-DEVD-AFC su bstrate. Release of AFC was monitored fluorometrically over time. GST-DIAPl inh ibits DCP-1-depe ndent caspase activity (A). In similar experiments in which 0.62nM dr!CE 81 His 6 was incub ated in caspase activity buffer with GST (0.48 µ,M) or GST-DIAPl, (0.16 µ,M), no inhibition of dr!CE activity by GST-DIAPlwas seen (B). 2890 Genetics: Hawkins et al. molecules that block GAL-driven, caspase-dependent cell death. Concluding Remarks. We have developed a yeast cell-based assay for the activity of one group of proteases, the caspases, in which caspase activity is monitored either by the cleavagedependent release of a transcription factor from its transmembrane anchor and subsequent activation of a reporter or by induction of cell killing. Both reporter activation and cell killing are suppressed by known caspase inhibitors. We have exploited this fact to directly isolate caspase inhibitors from a Drosophila embryo cDNA library. Yeast carrying the transcription-based caspase reporter should be useful as a background in which to carry out screens for proteins that cleave a caspase target site and for their regulators. Because yeast can be transformed with high efficiency, it also constitutes an ideal system in which to carry out large scale mutagenesis studies of particular proteases or their regulators. It may also be possible to screen for cellular targets of specific caspases by using artificial substrate libraries in which cDNA fragments substitute for the caspase target site linker in the caspase substrate fusion protein CLBDG6. A transcription-based reporter strategy similar to that described here also may provide a way to monitor caspase activity in cells of higher eukaryotes. Finally, substituting caspase cleavage sites for those of other site-specific proteases in the CLBDG6 reporter should enable the identification and study of these proteins and their regulators. We thank the following for providing plasmids and strains: J. Adams (CD4), J. Yuan (p/3actced38Z), L. K. Miller (pBS p35), H. Steller (DCP-1), D. Vaux (MIHA and caspases 3 and 9), X. Wang (Apaf-1), and R. Deshaies (yeast strains and plasmids). R. Deshaies, P. Sternberg, M. Guo, and W. Y. Shou provided helpful discussions and comments. C.J.H. is supported by a fellowship from the Human Frontiers Science Program. B.A.H. is a Searle Scholar. This work was supported by National Institutes of Health Grant GM057422-01 and a Burroughs Wellcome Fund New Investigator Award in the Pharmacological Sciences to B.A.H. I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. l 7. Alnemri , E. S. , Livingston, D. J., Nicholson, D. W., Salvesen, G., Thornberry, N. A., Wong, W.W. & Yuan, J. (1996) Cell 87, 171. Nicholson, D. W. & Thornberry, N. (1997) Trends Biochem. Sci. 22, 299-306. Salvesen, G. S. & Dixit, V. M. (1997) Cell 91, 443-446. Thornberry, N. A. & Lazebnik , Y. (1998) Science 281, 1312-1316. Cryns, V. & Yuan, J. (1998) Genes Dev. 12, 1551-1570. Teodoro, J. G. & Branton, P. E. (1997) J. Viral. 71, 1739-1746. Villa, P., Kaufmann , S. H. & Earnshaw, W. C. (1997) Trends Biochem. Sci. 22, 388-393. Clem, R. J. & Duckett, C. S. (1997) Trends Biochem. Sci. 7, 337-339. Cerretti, D. P., Kozlosky, C. J., Mosley, B., Nelson, N., Ness, K. V., Greenstreet, T. A, March, C. J., Kronheim, S. R., Druck, T., Cannizzaro, L. A, et al. (1992) Science 256, 97-100. Thornberry, N. A., Bull, H. G ., Calaycay, J. R. , Chapman, K. T., Howard, A D., Kostura, M. J., Miller, D. K. , Molineaux, S. M., Weidner, J. R. & Aunins, J. (1992) Nature (London) 356, 768-774. Ghayur, T., Bane rjee, S., Hugunin, M., Butler, D., Herzog, L., Carter, A, Quintal, L., Sekut, L., Talanian , R. , Paskind, M., et al. (1997) Nature (London) 386, 619-623. Zhang, Y. , Center, D. M., Wu, D. M. H., Cruikshank, W. W., Yuan, J., Andrews, D. W. & Kornfeld, H. (1998) J. Biol. Chem. 273, 1144-1149. Wang, S., Miura, M. , Jung, Y.-K., Zhu, H., Li, E. & Yuan, J. (1998) Cell 92, 501-509. Song, Z. W. , McCall, K. & Steller, H. (1997) Science 275, 536-540. McCall , K. & Steller, H. (1998) Science 279, 230-234. McCall, J. 0., Kadam, S. & Katz, L. (1994) Bio/Technology 12, 1012-1016. Xu , X., Gerard, A L. V., Huang, B. C. B., Anderson, D. C., Payan, D. G. & Luo, Y. (1998) Nucleic Acids Res. 26, 2034-2035. 141 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45 . 46. 47. 48. 49. 50. 51. 52. 53. 54. Proc. Natl. Acad. Sci. USA 96 (1999) Sices, J. J. & Kristie, T. M. (1998) Proc. Natl. Acad. Sci. USA 95, 2828-2833. Struhl, G. & Adachi, A. (1998) Cell 93, 649-660. Lecourtis, M. & Schweisguth, F. (1998) Curr. Biol. 8, 771-774. Stagljar, I. , Korostensky, C. Johnsson, N. & te Heesen , S. (1998) Proc. Natl. Acad. Sci. USA 95, 5187-5192. Madde n, P. J., Littman, D. R ., Godfrey, M., Maddon , D. E. , Chess, L. & Axel, R. (1985) Cell 42, 93-104. Thornberry, N. A., Rano, T. A., Peterson, E. P. , Rasper, D. M. , Timkey, T., Garcia-Calvo, M., Houtzager, V. M., Nordstrom, P.A., Roy, S., Vaillancourt, J.P. , et al. (1997) J. Biol. Chem. 272, 17907-17911. Varshavsky, A (1996) Proc. Natl. Acad. Sci. USA 93, 1214212149. Horii, T., Ogawa, T. & Ogawa, H. (1981) Cell 23, 689-697. Sikorski, R. S. & Hi eter, P. (1989) Genetics 122, 19-27. Johnston , M. & Davis, R. W. (1984) Mo/. Cell. Biol. 4, 1440-1448. Tamai, K. T. , Liu, X. L., Silar, P., Sosinowski, T. & Thiele, D. J. (1994) Mo/. Cell. Biol. 14, 8155-8165. Miura, M., Zhu, H., Rotello, R., Hartweig, E. A & Yuan, J. (1993) Cell 75, 653-660. Duan, H. J. , Chinnaiyan, A M., Hudson , P. L., Wing, J.P., He, W. W. & Dixit, V. M. (1996) J. Biol. Chem. 271, 1621-1625. Boldin , M. P., Goncharov, T. M. , Goltsev, Y. V. & Wallach, D. (1996) Cell 85, 803-815. Fernandes-Alnemri, T., Armstrong, R. C., Krebs, J., Srinivasula, S. M. , Wang, L., Bullrich, F., Fritz, L. C., Trapani, J. A , Tomaselli, K. J., Litwack, G., et al. (1996) Proc. Natl. Acad. Sci. USA 93, 7464-7469. Fraser, A G. & Evan, G. I. (1997) EMBO J. 16, 2805-2813. Duan, H. , Orth, K., Chinnaiyan, A M., Poirier, G. G., Froelich, C. J., He, W. W. & Dixit, V. M. (1996) J. Biol. Chem. 271, 16720-16724. Srinivasula, S. M., Fernandes-Aln emri, T., Zangrilli, J ., Robertson, N., Armstrong, R. C., Wang, L., Trapani, J. A, Tomaselli, K. J., Litwack, G. & Alnemri, E. S. (1996) J. Biol. Chem. 271, 27099-27106. Zou, H., Henzel , W. J., Liu, X., Lutschg, A. & Wang, X. (1997) Cell 90, 405-413. Hay, B. A. , Wassarman, D. A & Rubin, G. M. (1995) Cell 83, 1253-1262. Uren, A G. , Pakusch, M. , Hawkins, C. J., Puls, K. L. & Vaux, D. L. (1996) Proc. Natl. Acad. Sci. USA 93, 4974-4978. Clem, R., Fechheimer, M. & Miller, L. K. (1991) Science 254, 1388-1390. Breeden, L. & Nasmyth, K. (1985) Cold Spring Symp. Quant. Biol. so, 643-650. Miller, J. H. (1972) in Experiments in Molecular Genetics (Cold Spring Harbor Lab. Press, Plainview, NY). Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S. M., Ahmad, M. , Alnemri, E. S. & Wang, X. (1997) Cell 91, 479-489. Srinivasula, S., Ahmed, M. , Fernandes-Alnemri, T. & Alnemri, E. S. (1998) Mo/. Cell 1, 949-957. Yang, X., Chang, H. Y. & Baltimore, D. (1998) Science 281, 1355-1357. Yang, X., Chang, H. Y. & Baltimore, D. (1998) Mo/. Cell 1, 319-325. Muzio, M., Stockwell, B. R. , Stennicke, H. R., Salvesen, G. S. & Dixit, V. M. (1998) J. Biol. Chem. 273, 2926-2930. Crook, N. E. , Clem , R. J. & Miller, L. K. (1993) J. Viral. 67, 2168-2174. Uren, AG., Coulson, E. J. & Vaux, D. L. (1998) Trends Biochem. Sci. 23, 159-162. Deveraux, Q. L., Takahashi, R., Salvesen, G. S. & Reed, J. C. (1997) Nature (London) 388, 300-304. Roy, N., Deveraux, Q. L., Takahashi, R. , Salvesen, G. S. & Reed , J. C. (1997) EMBO J. 16, 6914-6925. Harvey, A. J., Soliman, H., Kaiser, W. J. & Miller, L. K. (1997) Cell Death Differ. 4, 733-744. Vucic, D., Kaiser, W. J., Harvey, A. J. & Miller, L. K. (1997) Proc. Natl. Acacl. Sci. USA 94, 10183-10188. Vucic, D., Kaiser, W. J. & Miller , L. K. (1998) Mo/. Cell. Biol. 17, 667-676. Hawkins, C. J., Ekert, P. G. , Uren, A. G., Holmgreen, S. P. & Vaux, D. (1998) Cell Death Differ. 5, 569-576. 142 Appendix II 143 Cell, Vol. 98, 453-463, August 20, 1999, Copyright ©1999 by Cell Press The Drosophila Caspase Inhibitor DIAP1 Is Essential for Cell Survival and Is Negatively Regulated by HID Susan L. Wang,' Christine J. Hawkins,' Soon Ji Yoo,' H.-Arno J. MOiier, U and Bruce A. Hay•t 'Divi sion of Biology MC 156-29 California Institute of Technology Pasadena, California 91125 t lnstitut fOr Genetik Heinrich-Heine Universitat Do sseldorf Universitatsstrasse 1 D-40225 Dusseldorf Germany Summary Drosophila Reaper (RPR), Head Involution Defective (HID), and GRIM induce caspase-dependent cell death and physically interact with the cell death inhibitor DIAP1 . Here we show that HID blocks DIAP1's ability to inhibit caspase activity and provide evidence suggesting that RPR and GRIM can act similarly. Based on these results, we propose that RPR, HID, and GRIM promote apoptosis by disrupting productive IAP-caspase interactions and that DIAP1 is required to block apoptosis-inducing caspase activity. Supporting this hypothesis, we show that elimination of DIAP1 function results in global early embryonic cell death and a large increase in DIAP1-inhibitable caspase activity and that DIAP1 is still required for cell survival when expression of rpr, hid, and grim is eliminated . Introduction Programmed cell death, or apoptosis, is an evolutionarily conserved process by which organisms remove damaged or unwanted cells (reviewed in Wyllie et al., 1980; Raff, 1992). Central components of the machinery that carries out this proces s are caspa ses, a family of aspartate-specific, cysteine-dependent proteases (Alnemri et al., 1996). Caspases are made as zymogens, comprising three functional modules: a prodomain and two catalytic subunits known as the large, or p20, and the small, or p10, subunits . In general, caspases are activated in vivo following cleavage at aspartate residues that separate the prodomain from the catalytic region and separate the large and small subunits of the cata lytic domain. The aspartates cleaved in caspase zymogens often resemble consensus target sites of known caspases. This has suggested that caspases may function in a cascade in which initiator caspases, activated by upstream death signals, cleave and activate a set of executioner caspases that carry out proteolytic cleavages of cellular proteins, leading ultimately to cell death (reviewed in Nicholson and Thornberry, 1997; Salvesen and Dixit, 1997; Cryns and Yuan, 1998; Thornberry and Lazebnik, 1998). t To whom correspondence should be addressed (e-mail: muellear@ uni-duesseldorf.de [H .-A. J. M.], haybruce@its.caltech.edu [B. A. H.]). Because ca spases have the potential to initiate cascades of proteolysis, it is important that their activity be tightly regul ated. The only cellular caspase inhibitors identified to date all share homology with a family of cell death inhibitors identified in baculoviruses, known as inhibitors of apoptosis, or IAPs (Crook et al., 1993; Birnbaum et al., 1994; reviewed in Lacasse et al., 1998 and Deveraux and Reed, 1999). IAP homologous proteins have been identified in other viruses, Drosophila melanogaster, Caenorhabditis elegans, mammals, and yea st (reviewed in Uren et al., 1998). These proteins are characterized by one or more N-terminal repeats of a motif known as the baculovirus !AP repeat (BIR) . Many also have a C-terminal RING finger motif. Some !AP homologous proteins have been shown to act as cell death inhibitors. In several cases, this activity is localized to the BIR-containing region of the protein (reviewed in Lacasse et al., 1998; Uren et al., 1998; Deveraux and Reed, 1999), which has also been shown to mediate caspase binding and inhibition (Deveraux et al., 1997, 1998; Roy et al., 1997; Takahashi et al., 1998; Hawkins et al., 1999). Proteins that initiate cell death by disrupting IAPcaspase interactions have not been described. However, the products of the Drosophila reaper (rpr), head involution defective (hid), and grim genes are interesting candidates. All three loci are located in the 75C region of the Drosophila genome, and deletion of this region eliminates death induced by many different stimuli, indicating that important cell death activators reside within (White et al., 1994). Furthermore, individual overexpression of any one of these genes is sufficient to initiate caspase-dependent cell death in many cells that normally live (Grether et al., 1995; Hay et al., 1995; Chen et al., 1996; White et al., 1996; Vucic et al., 1997, 1998; Zhou et al., 1997; Wing et al., 1998). Two Drosophila IAPs, DIAP1, the product of the thread (th) locus (Hay et al., 1995), and DIAP2 (Hay et al., 1995; Duckett et al., 1996; Uren et al., 1996), suppress rpr-, hid-, and grimdependent cell death when overexpressed (Hay et al., 1995; Vucic et al., 1997, 1998; Bergmann et al., 1998; Wing et al., 1998), and DIAP1 and DIAP2 can immunoprecipitate RPR, HID, and GRIM from insect cells (Vucic et al., 1997, 1998). Furthermore, DIAP1 binds a processed form of the caspase drlCE in insect cells (Kaiser et al., 1998) and inhibits the activity of a second caspase, DCP-1 (Hawkins et al., 1999). These observations suggest several models of how Drosophila IAPs, RPR, HID, GRIM, and Drosophila caspases might interact to regulate cell death (Kaiser et al., 1998). In one model, RPR, HID, or GRIM activates caspases through an !AP-independent pathway. In this model, Drosophila IAPs suppress apoptosis by acting as a sink for these proteins and by inhibiting caspase activation or activity initiated by their action. In a second model, DIAP1 functions primarily as a caspase inhibitor, and RPR, HID, and/or GRIM initiate caspase-dependent cell death by preventing IAPs from productively interacting with caspases. The yeast Saccharomyces cerevisiae provides an Cell 454 ideal system to study interactions between caspases and molecules that regulate their activity (Hawkins et al., 1999; Kang et al., 1999). While yeast cells do not exhibit apoptosis, and database searches have failed to uncover yeast homologs of the core apoptosis regulators identified in worms, flies, and mammals, overexpression of DCP-1 or drlCE results in yeast cell death that is blocked by DIAP1. Here we used yeast expressing these proteins as a system to assay the ability of RPR. HID, and GRIM to disrupt IAP-caspase interactions. We found that all three proteins, while nontoxic on their own, killed yeast coexpressing DCP-1 or drlCE and DIAP1, suggesting that they were blocking DIAP1 's ability to function as a caspase inhibitor. We pursued the basis for this activity further with HID and found, both in yeast and in vitro, that proteins containing the N-terminal 37 residues of HID, which are sufficient to induce apoptosis in insect cells (Vucic et al., 1998), suppressed DIAP1 's ability to inhibit DCP-1 activity. We also found that the DIAP1 loss-of-function phenotype consists of an embryo-wide set of cellular changes reminiscent of apoptotic cell death and that these were associated with the activation of DIAP1-inhibitable caspase activity. Furthermore, double mutants that remove zygotic rpr, hid, grim, and DIAP1 function showed phenotypes similar to those of the DIAP1 loss-of-function mutant alone. These observations suggest that a principal function of DIAP1 is to promote cell survival by blocking caspase activity and that at least one mechanism by which RPR. HID, and GRIM promote apoptosis is by disrupting IAPcaspase interactions. Results and Discussion RPR, HID, and GRIM Block DIAP1's Ability to Suppress Caspase-Dependent Cell Death in Yeast Yeast were transformed with two plasmids: one in which DCP-1 expression was driven by the inducible GAL 1 promoter, and a second in which DIAP1 expression was driven by the constitutive Adh promoter (hereafter referred to as DCP-1-DIAP1 yeast). These cells were then transformed with a second GAL 1 expression plasmid that was either an empty vector or carried GAL 1-driven RPR, HID, or GRIM. Transformants were spotted as a series of 10-fold serial dilutions onto glucose plates to indicate the number of cells and onto galactose plates to induce expression of DCP-1 and RPR. HID, or GRIM. DCP-1-DIAP1 yeast carrying an empty vector survived on galactose, but DCP-1-DIAP1 yeast expressing HID or GRIM did not (Figure 1A). Expression of RPR under GAL 1 control did not kill DCP-1-DIAP1 yeast in which DIAP1 expression was driven by the Adh promoter, but it was able to kill DCP-1-DIAP1 yeast in which DIAP1 expression was driven by a weaker promoter, the copper-inducible CUP1 promoter, in the presence of 30 µ,M Cu 1 + (Figure 1A). Importantly, expression in yeast of RPR, HID, or GRIM alone under GAL 1 control had no effect on cell growth as compared with yeast expressing only the empty vector (Figure 1A). These results suggest that RPR, HID, and GRIM are able to block DIAP1 's ability to inhibit DCP-1 caspase activity. Three other Drosophila caspases, drlCE, DCP-2/ 144 Dredd, and DRONC, have been implicated as effectors of apoptosis (Fraser and Evan, 1997; Fraser et al., 1997; Chen et al., 1998; Dorstyn et al., 1999). We wanted to determine if RPR, HID, and GRIM played a similar role in regulating their activity. We focused our attention on drlCE. A version of drlCE lacking the prodomain is not inhibited by DIAP1 (Hawkins et al., 1999), but a version of drlCE that contains prodomain sequences is (C. J. H. and B. A. H., unpublished observations). Full -length drlCE expressed in yeast under GAL 1 control neither kills nor activates a cleavage-dependent reporter, probably because the caspase does not significantly autoactivate (Hawkins et al., 1999). We created a form of drlCE that is active in yeast and contains prodomain sequences by generating a "reversed" form of drlCE, rev-drlCE, in which the drlCE p10 domain was placed N-terminal to prodomain and p20 sequences (Figure 1 B). Reversed forms of mammalian caspases that are not otherwise active are constitutively active (Srinivasula et al., 1998). As shown in Figure 1B, expression of rev-drlCE under GAL 1 control kills yeast, and this death is prevented by coexpression of DIAP1. Importantly, as with DCP-1, expression of HID, GRIM, or RPR killed yeast coexpressing rev-drlCE and DIAP1. The N Terminus of HID Mediates Its Ability to Disrupt IAP-Caspase Interactions in Yeast In insect cells, the N-terminal 37 amino acids of HID, expressed in the context of a larger fusion protein, are necessary and sufficient to induce apoptosis and to mediate HID's binding to DIAP1 (Vucic et al., 1998). To test the activity of these sequences in yeast, we generated GAL 1 expression constructs similar to those of Vucic et al. (1998) in which the first 37 amino acids of HID were either present or absent (Figure 1C). HIDt.(38-335) encodes a protein in which the first 37 amino acids of HID are fused to the C-terminal 74; HIDt.(1-335) consists of only the last 74 residues of HID; and HIDt.(1-36) encodes a protein that lacks the first 36 residues of HID but contains the rest of the full-length HID coding sequence. GAL1-dependent expression of HIDt.(38-335) in DCP1-DIAP1 yeast resulted in no growth. In contrast, DCP1-DIAP1 yeast expressing HIDt.(1-335) grew normally. HIDt.(1-36) had weak killing activity in this assay (Figure 1 D) but not in insect cell death assays (Vucic et al., 1998) or other in vitro assays described below. As with fulllength HID, expression of HIDt.(38-335). HIDt.(1-335), or HIDt.(1-36) in isolation had no effect on yeast cell growth (Figure 1D). Thus, proteins carrying the N-terminal 37 residues of HID, which are sufficient to kill insect cells, are also able to kill DCP-1-DIAP1 yeast. HID Suppresses DIAP1's Ability to Inhibit DCP-1 Caspase Activity In Vitro The results of our yeast experiments suggested that HID directly inhibits DIAP1. We tested this idea in vitro using purified proteins. Bacterially expressed and purified versions of DCP-1, DIAP1, and HID (Figure 2A) were mixed in various combinations and DCP-1 caspase activity measured fluorometrically (Figures 2B and 2C). His6tagged DCP-1 (DCP-1-His6) is active as a caspase, and this activity was inhibited by DIAP1 but not GST 145 DIAP1 Is Essential and Is Inhibited by HID 45 5 A GLUCOSE GAI.JRAF DCP· 1 + vectcrs DC P.1 • Adh-OIAP l + ved<)r DCP·1 • Alf>-DIAPI • HID DCP·l + A<:h-OIAPI + GRIM vectcrs DCP·1 • vecta"S DCP·l • CUP-OIAP1 • vector DCP.1 • CUP-OIAP1 • APR ve<IQ"O vectcxs • APR vectors • HID vectors • GRIM C ■ •• -. Wlldtype drlCE rev-drlCE re',<!rlCE + Adh-OIAP1 • vector rev-<irlCE • Alf>-OIAP1 • HID re'-'<irlCE • Act>-D IAP1 • GRIM re\"drlCE • vect«s re~ICE + CUP-OIAP1 • vector re.,..;,ICE + CUP-OIAPl • APR OCP.1 • Act>-OIAP1 + vector d·::::: : ::!:::::a DCP.1 • Act>-0lAP1 • HID!l(J&-335) DCP.l • A<i>-OIAP1 • HID 336 DCP.1 • Alf>-OIAPl • HIDL',(1 ·335) HIDt,(1-36) : Jr HIDt,(1·335) ,••·--... - -.-.. ----..~ b ■ ss LS ■ I GLUCOSE GAI.JRAF D 410 37 LS !~ 98 I I~ GLUCOSE GAl.JRAF rev-dr!G E • vectxx's ■ HIO HI011(3&-335l B DCP-1 + Adtt-OIAP1 + HIO.l( 1·36I vectors ~ HlO + vect«s ■ Polyoma epitope tag HID,'(3&-33S) • voctors HI0 )(1-33S) • vecte<s HID<'.l(1·36J • veaors Figure 1. RPR, HID , and GRIM Block DIAP1 's Ability to Suppress Ca spase-Dependent Yeast Cell Death (A) RPR, HID, and GRIM kill yeast expressing DCP-1 and DIAP1 . Top: GAL 1-driven DCP-1 causes yeast cell death that is prevented by Adh dependent expression of DIAP1 (DCP-1 + Adh-DIAP1 + vector). GAL1-driven HID (DCP-1 + Adh-DIAP1 + HID) or GRIM (DCP-1 + AdhDIAP1 + GRIM) results in yeast cell death. Middle: GAL1-driven RPR kills yeast expressing DCP-1 and CUP1-driven DIAP1 (DCP-1 + CUPDIAP1 + RPR) in the presence of 30 µM cu >+ . Bottom: Expression of RPR, HID, or GRIM in the presence of empty vectors does not affect yeast cell viability. (B) RPR, HID, and GRIM kill yea st expressing rev-drlCE and DIAP1 . Top: drlCE contains prodomain (P), large (LS), and small (SS) subunits. In reversed drlCE (rev-drlCE), the predicted drlCE small subunit was placed N-terminal to drlCE prodomain and large subunit sequences. Middle: Expression of HID or GRIM kills yeast expressing rev -drlCE and Adh-DIAP1 (rev-drlCE + Adh-DIAP1 + HID; rev-drlCE + Adh-DIAP1 + GRIM). Bottom: Expression of RPR kills yeast expressing rev-drlCE and CUP-DIAP1 (rev-drlCE + CUP -DIAP1 + RPR) in the presence of no added Cu>+. (C) Diagram of HID constructs. Numbers represent amino acid positions in full-length HID . Solid lines indicate regions of HID present in the fusion protein; dashed lines indicate sequences that are deleted. A polyoma epitope tag present at th e C terminus of each coding region is indicated by a black bar. (D) Proteins containing the first 37 residues of HID block DIAP1 's ability to inhibit DCP-1 activity in yeast. Top: Effects of expression of HID fusion proteins on the viability of DCP-1 + Adh-DIAP1 yeast. Bottom: Effects of expression of HID fu sion proteins and empty vectors on yeast viability. (Hawkins et al., 1999; Figure 2B). Addition of HID alone to a reaction containing DCP-1-His6 did not alter DCP1-His6 activity, indicating that HID has no direct effect on DCP-1-His6 (Figure 2B). However, when full-length HID was incubated with DIAP1 and DCP-1-His6, DIAP1dependent inhibition of DCP-1-His6 activity was lost (Figure 2B). Incubation of DIAP1 and DCP-1-His6 with HID~(1-36)-His6 had little or no effect on DIAP1 's ability to inhibit DCP-1-His6 activity, indicating that the first 37 residues of HID are necessary for this activity. To determine if the first 37 residues of HID were sufficient to mediate this effect, we carried out experiments using a version of HID that consists of the first 37 residues of HID fused to GST and a C-terminal His6 tag [HID(137)GST-His6] . As shown in Figure 2C, addition of HID(137)GST-His6 to an assay containing DCP-1-His6 and DIAP1 blocked DIAP1 's ability to inhibit DCP-1-His6 activity, while addition of an equivalent amount of GSTHis6 had no effect. Thus, consistent with our observations in yeast, purified HID directly blocks DIAP1 ' s ability to inhibit DCP-1 caspase activity, and the first 37 amino acids of HID, at least in the context of a larger fusion protein, are necessary and sufficient for this activity. To determine if HID directly interacts with DIAP1 or DCP-1, we carried out binding experiments. As expected from the results of our yeast and in vitro activity 146 Cell 456 A A 4 10 HID-Hls6 •GST - +GST-OIAPi INPUT 37 HID~(1·36)•H1s6 " " 37 H10( 1-37)GST-H 1s6 ~ GST-Hls6 == !i,l His 6 epllope lag □ - - GST sequence B 10- 7 0.9 i' ~ -·· . 0 .6 (.) u.. <t B ......... - .. ,, .. :,,,"'•• 0.3 tMBP. DIAP1 INPUT 3 5 + DCP+H1s6 (ptO) 8 9 --+ MBP , ...... ....... .--~ 0 0 20 40 Time (min) 60 2 4 6 DCP· 1-H is6+GST-+GST C DCP- 1-His6+0IAP1+GST DCP- 1-His6+G ST +HIO-His6 • ~ _ _..,;•,.M.;; .B.,P,.-D;,;:IA_,P,.1...,_ _ DCP-1- His6+DIAP1 +HIO-His6 DCP- 1-His6+ 01AP l +HIOl\(1 ·36)-H is6 C ~~+~Hf;~ GST-+Is 6 ,,............. .....-:::::::J 0.9 , ------ i' 0 . 6 ~ 2 (.) u.. <t ~ 0 .3 0 0 - 20 40 Time (min) 60 DCP-1 -H1s6+GST+GST•His6 - □- OCP- 1-Hls6+GST+Hl0(1-37)GST-His 6 - 0 - DCP-1-His6+0IAP1+GST-H1s6 -l!J. - DCP- 1-HisG+DIAP i +HID( 1-37)GST- H1s6 Figure 2. Proteins Containing the First 37 Residues of HID Block DIAP1 's Ability to Inhibit DCP-1 Activity In Vitro (A) Schematic of HID constructs. HID coding regions are represented by open boxes and deleted regions by dashed lines. (B) The N-terminal 36 residues of HID are required to inhibit DIAP1 function. DCP-1-His6 (0.2 nM) activity was measured by following the release of AFC fluorometrica lly from the Ac-DEVD-AFC substrate over time. The total protein concentration in eac h incubation was kept constant, with GST (0 .2 µM or 0.77 µM) being used to replace DIAP1 (0.2 µM), HID-His6 (0 .44 µM), or HIDCi(1-36)-His6 (0.44 µM) as the assay required. (C) A protein containing th e N-terminal 37 residues of HID, HID(137)GST- His6, is sufficient to block DIAP1 's caspase inhibitory activity. Assays were performed as in (B). assays, GST-DIAP1 coupled to glutathione-Sepharose bound full-length HID-His6 and DCP-1 -His6 but did not appreciably bind HID6(1-36)- His6. GST alone showed no interaction with any of these proteins (Figure 3A). Also , a maltose-binding protein-DIAP1 fusion protein 3 4 5 6 ♦ DCP-1·His6 (ptO) 7 Figure 3. Physical Interacti ons between HID, DIAP1 , and DCP-1 (A) DIAP1 binds DCP-1 and HID, and the first 37 residues of HID are required for thi s interaction. Glutathione-Sepharose-bound GST or GST-DIAP1 was incubated in binding buffer with DCP-1-His6, HID-His6, or HIDCi(1-36)-His6 (Figure 2A) . Binding reactions were followed by washes and We stern blotting. A fraction of the input of DCP-1-His6, HID-His6, and HIDCi(1-36)-His6 is shown. (B) DIAP1 binds proteins containing only the first 37 residues of HID. Amylose resin-bound MBP or MBP-DIAP1 was incubated with GST-His6 or HID(1-37)GST-His6 and processed as above. (C) HID(1-37)GST-His6 does not prevent DCP-1- His6 from binding MBP-DIAP1 . MBP-DIAP1 bound to amylose resin was incubated with increasing amounts of HID(1-37)GST-His6 or GST-His6. Following this preinc ubation, a constant amount of DCP-1-His6 was introduced into th e mixture. The beads were washed and processed for Western blotting using an anti-His antibody to detect DCP-1-His6. MBP beads incuba ted with the highest concentration of HID(137)GST-His6 and equival ent amounts of DCP -1-His6 were processed in a similar manner. (MBP-DIAP1) bound to amylose resin was able to bind HID(1-37)GST-His6, while MBP bound to the same resin was not (Figure 3B). Furthermore, HID(1-37)GST-His6, bound to glutathione-Sepharose beads, was able to bind 35 S-methionine-labeled, in vitro-translated DIAP1 but not in vitro-translated DCP-1 (data not shown) , These observations, in conjunction with the fact that HID-His6 or HID(1-37)GST-His6 alone had no effect on DCP-1-His6 caspase activity (Figures 2B and ZC), suggest that HID's effects on caspase activity are mediated through DIAP1. DIAP1 Is Essential and Is Inhibited by HID 457 To explore the mechanism by which HID blocks DIAP1 's ability to inhibit DCP-1, we asked if HID(1-37)GST-His6 was able to displace DCP-1-His6 from DIAP1 . MBPDIAP1 bound to amylase resin was preincubated with increasing amounts of GST or HID(1-37)GST-His6 followed by addition of DCP-1-His6. Binding reactions were followed by washes and analysis of the beadbound proteins by Western blotting. As shown in Figure 3C, DCP-1 binding to MBP-DIAP1 was not appreciably decreased by the presence of up to a 20-fold molar excess of GST-His6 or HID(1-37)GST-His6 over MBPDIAP1. As expected, MBP beads incubated with HID(137)GST-His6 showed no binding to DCP-1-His6. Thus, we see no evidence that HID(1-37)GST-His6, at molar ratios with respect to DIAP1 greater than those used to demonstrate inhibition of DIAP1 function in caspase activity assays (Figure 2C), prevents binding of DCP-1His6 to MBP-DIAP1. These observations suggest that HID(1-37)GST-His6 does not prevent DIAP1 from inhibiting DCP-1 activity by blocking DCP-1 binding. It is important to note, however, that full-length HID, which has a large C-terminal domain that can regulate HID activity (Bergmann et al., 1998), may behave differently. Our observations suggest that RPR, HID, and GRIM can promote caspase-dependent cell death by blocking DIAP1 's ability to inhibit caspase activity. These conclusions are supported by the observations that in insect cells, RPR, HID, and GRIM promote caspase-dependent cell death that is sensitive to the levels of DIAP1 (Hay et al., 1995; Vucic et al., 1997, 1998; Bergmann et al., 1998; Wing et al., 1998), that these proteins associate with DIAP1 in vivo (Vucic et al., 1997, 1998), and that DIAP1 binds and inhibits caspases (Kaiser et al., 1998; Hawkins et al., 1999; this work). Furthermore, since proteins that contain only the first 37 amino acids of HID are sufficient to initiate the caspase-dependent death of insect cells and to block DIAP1 's ability to inhibit caspase activity, but other regions of HID lack activity in insect cells, it is likely that disrupting DIAP1 's ability to inhibit caspase activity is an important mechanism by which HID promotes cell death. Because of difficulties with solubility of RPR and GRIM under conditions that allowed productive DIAP1-caspase interactions, we were unable to carry out in vitro experiments similar to those carried out with HID. However, based on the fact that they behaved similarly to HID in yeast expressing DCP-1 or rev-drlCE and DIAP1, it seems likely that they also are able to block DIAP1 's ability to inhibit caspase activity. Our observations do not exclude the possibility that RPR, HID, and GRIM promote apoptosis through other pathways as well. For example, RPR has been shown to promote apoptosis in a Xenopus Jaevis system (Evans et al., 1997) through association with a novel protein, SCYTHE (Thress et al., 1998), and peptides corresponding to N-terminal sequences of RPR and GRIM block K + channel function (Avdonin et al., 1998). Furthermore, as described below, removal of DIAP1 in the embryo results in massive cell death, but the DIAP1 mutant cells do not show all the features of apoptosis seen in wildtype embryos, consistent with the idea that other pathways may be necessary to create a complete apoptotic response. 147 DIAP1 Mutant Embryos Show Global Early Morphogenetic Arrest and Cell Death Associated with an Increase in DIAP1 -lnhibitable Caspase Activity If disruption of IAP-caspase interactions is an important mechanism by which cell death is initiated in Drosophila, then caspase activity must be continually held in check by DIAP1 and/or other IAPs in order to block an everpresent death signal. To explore this possibility, we characterized the embryonic phenotype of mutations (alleles of th) that disrupt OIAP1 function. Homozygous th mutant embryos exhibit a severe terminal embryonic lethal phenotype and do not form a cuticle (Figure 4G and data not shown) . We monitored early development of loss-of-function th embryos from th', a strong loss-of-function allele, and a deficiency for the region, Of(L)brm11 (Figure 4). th' and Df(3R)brm11 homozygous embryos undergo normal cellularization and initial gastrulation movements such as ventral furrow formation, cephalic furrow formation, and posterior midgut invagination (Figures 4A-4C). However, during germband extension, within 15 to 30 min after the onset of gastrulation movements, virtually all morphogenetic movements cease (Figures 4A-4C). While the stage of initial arrest is somewhat variable in time (compare Figures 4B and 4C with respect to the extent of ventral furrow formation and posterior midgut invagination), within the next hour all the cells adopt a rounded morphology and the yolk cell fragments and eventually moves to the surface of the embryo. The disruption of embryonic integrity is seen particularly clearly in Figures 4D-4G, in which wild-type and th' mutant embryos are visualized using scanning electron microscopy. While some th 5 homozygous embryos showed normal morphology at the beginning of germband extension (data not shown), other embryos with a similar degree of extension movements showed extensive cell rounding, suggesting a disruption of intercellular adhesion (Figures 4D and 4E). All th5 embryos also show a regression of the cephalic furrow (compare Figure 4A with Figures 4B and 4C, and Figure 4D with 4E) . At the extended germband stage (stage 9), the surface of cells in control embryos is smooth and the cells are closely apposed to each other (Figure 4F). In contrast, by this time th 5 homozygotes have adopted a uniform cellular morphology in which all cells have a round shape, the surface of the cells contains many membrane blebs, and vesicular cell fragments similar to those seen in dying cells are present (Figure 4G). The yolk cell , normally located interiorly, is visible at the surface. Thus, zygotic loss of DIAP1 function results in early morphogenetic arrest followed by severe morphological abnormalities at the beginning of germband extension. We used TUNEL labeling to determine whether the massive DNA fragmentation characteristic of cells undergoing apoptosis had occurred in cells of th mutants (Figure 5) . Wild-type embryos during germband extension showed significant DNA fragmentation in only a few cells (Figures SA, SC, and SE). In contrast, comparably staged th' embryos labeled extensively in most if not all cells. Initially, the TUNEL signal is relatively weak (Figures SB and SD), but the intensity increases within the following hour of development to a strong uniform labeling (Figures SB, SD, and SF). In contrast, Cell 458 148 Figure 4. Removal of Zygoti c DIAP1 Lead s to Morph ogenetic Arrest during Germband Extension and Global Cell Sha pe Changes (A) Wild -type developmental seri es of living embryos derived from a th'/TM 3[ftz:: lac Z) stock during gastrulation and germ band ext ension (st ages 6 through 9). Arrowh eads mark th e position of the cephalic furrow, and arrow s indicate th e front of th e extend ing germba nd . (B and C) th' and Df(3L/brm 11 homozygous embryos, respectively, at th e sa me developmental stages as in (A). (D and F) Scanning electron micrograph s of balancer-bearing embryos from th'/TM3[ftz:: lacZ) stock. The right -hand panels show higher magnification of th e g eneral morphology of the embryonic surface. (D) shows an emb ryo at stag e 8, and (F) shows an embryo at stag e 9. (E and G) Similarly staged embryos homozygou s for th5. Scale bars are 100 µ m in (C) and (G) (l eft -hand panel) and 10 µ m in (G) (ri ght-hand panel). embryos homozygous for the weaker allele th' do not show a significant increase in TUNEL-positive cells as compared to comparably staged control embryos at the retracted germband stage (stage 14; Figures SG and SH). To determine if loss of DIAP1 in fact results in an increase in caspase activity, we measured caspase activity in extracts of 3- to 5-hr-old embryos from wild type, th 5, and th' stocks by following the release of AcDEVD-AFC fluorometrically (Figure SI). Extracts from wild-type and th' embryos had very low levels of caspa se activity. In contrast, extracts from th' embryos had very large amounts of caspase activity. Importantly, the increased caspase activity present in the th' embryonic extracts was inhibited by the addition of purified DIAP1 (Figure 51), consistent with the idea that DIAP1 normally represses these caspases. The caspase targets of DIAP1 in the Drosophila em bryo or other tissues have not been identified. Five caspases have been identified in Drosophila: DCP-1 (Song et al., 1997), drlCE (Fraser and Evan, 1997), DCP- 2/ Dredd (lnohara et al., 1997; Chen et al. , 1998), DRONC (Dorstyn et al., 1999), and a fifth gene (C . J. H. and B. A. H. , unpublished). DIAP1 is able to inhibit the activity of the three caspases we have tested : DCP-1 (Hawkins et al., 1999), drlCE (this report), and DRONC (C . J . H., unpublished). The DIAP1 loss-of-function embryonic phenotype may result from the unrestrained activity of zygotically expressed versions of th ese caspases. Alternatively, DIAP1 may also function in the early embryo to block the activity of caspases introduced late during oogenesis into the developing oocyte from the degenerating nurse cells (Cavaliere et al. , 1998; Chen et al. , 1998; Foley and Cooley, 1998; McCall and Steller, 1998; Dorstyn et a I. , 1999) . DIAP1 Loss-of-Function Mutant Embryos Do Not Stain with Acridine Orange, a Marker of Apoptotic Cell Death Acridine orange (AO) is commonly used as a marker for apoptotic cell death. It appe ars to enter all cells but is DIAP1 Is Essential and Is Inhibited by HID 459 149 Figure 5. Cells in Embryos Homozygous Mutant for DIAP1 Show Premature and Increased DNA Fragmentation and an Increase in DIAP1 -lnhibitabl e Caspase Activity (A-F) Confoca l images of staged embryos from stocks of th'/TM3[ftz::lacZ) and (G and H)th'/TM3 [ftz::lacZ) double labeled for TUNEL (green) and ~-galactosidase (red). Scale bar is 100 µ m. (A) th'/TM3[ftz::lacZ) and (B) th'lth' embryos at ea rly germband extension (stage 8). (C) th5/ TM 3[ftz:: lacZ) and (D) th 5/th 5 embryos at extend ed germband stage (early stage 9). (E) th'/TM3[ftz::lacZ) and (F) th'!th' embryos at extended germband stag e (late stage 9). (G) th'/TM3[ftz::lacZ) and (H) th'lth 6 embryos at retra cted germband stage (stage 14). (I) Caspase activity of wild-type and th mutant embryo extracts in the presence of GST or DIAP1 . One microgram of extract of 3- to 5-hr-old embryos from stocks of w , th'! TM3, or th6/ TM3 was introduced into caspase activity buffer con t aining 100 µM of the AcDEVD-AFC substrate and 0.5 µg of eith er GST or DIAP1. Release of AFC was monitored fluorometrically over time. 6 o w-+GST X th6+ GST • lh5+ GST • th5+ DIAPl 0 0 15 30 45 Time (rnin) only retained (through unknown mechanisms) by cells undergoing apoptosis (see Abrams et al., 1993) . In light of the strong defects in th mutant embryos reported here, we repeated experiments described in Hay et al. {1995) reporting that th embryos did not show an increase in AO staining. Figure 6A shows a bright-field image of a stage 14 th'!TM3 embryo. Figure 6B shows the same embryo stained with AO. Dying cells show bright green fluorescence, and living cells do not stain. A th' homozygous embryo displaying the th terminal phenotype is shown in Figure 6C. All cells have rounded up, and the yolk sac has moved to the surface. However, as shown in Figure 6D, no AO staining is seen despite the fact that embryos at this stage show cell rounding, membrane blebbing, increased caspase activity, and DNA fragmentation . Thus, the th phenotype, while showing some features of apoptosis, lacks others. One explanation for this is that the early embryo simply lacks the cellular machinery required to carry out apoptotic steps leading to AO retention . A second possibility is that the Figure 6. Cell s in DIAP1 Loss-of- Function Embryos Do Not Show an Increase in Acridine Orang e Staining Charact eristic of Cells Underg oing Apoptosis in Wild-Type Embryos (A and B) stage 14 th'/TM3[ftz::lacZ) or TM3[ftz::lacZ]/TM3[ftz::lacZ] embryo sta ined with acridin e orange and viewed und er the confocal microscope using Nomarski (A) and fluorescence channels (B). (C and D) th' homozygous embryo exhibiting the terminal phenotype. At this stage, acridine ora nge sta ining is never observed in th' homozygotes (D). Scale bar is 50 µm. 150 Cell 460 ,,... _.. oth oroltector pa1 hwoys Figure 7. Model for How DIAP1 Regulates Apoptosi s / I H99 genes ' REAPER {I HID ~ DIAP 1 GRIM I- .... : "?- ""·'"''P'""M' ' ---f caspase activity ,;I __., apoptosis DIAP1 inhibits caspase activity and is essential for cell survival. DIAPl's ability to inhibit caspa se activity is suppressed by RPR, HID, and GRIM , thu s promoting caspase-depen- !- - - - - - - - - - dent cell death . RPR and GRIM have other activities (Evans et al., 1997; Avdonin et al. , _ _ 1998; Thress et al. , 1998), suggesting that th ese proteins _may al so promote apoptos,s through other pathways. Free DIAP1 may inhibit apoptosis by binding th ese proteins, sequestering th em fmm their targ ets . Thu s, DIAP1 may act both upstream and dow nstream of RPR, HID, and GRIM to prevent cell d eath . Genes within th e H99 mterval _are not required for some cell deaths (White et al., 1994; Foley and Cooley, 1998), indicating that other death pathw a s exist. It ,s not known ,f th ese pathways act through DIAP1. y dea th palhway, expression of other IAPs, such as DIAP2, suppresses some of the caspase-dependent events necessary for a full apoptotic cell death phenotype. A third possibility (discussed further below) is that loss of DIAP1 function, while sufficient to kill cells, is not sufficient to confer all aspects of apoptosis. In this scenario, proteins such as RPR, HID, or GRIM that are sufficient to trigger a full apoptotic response when overexpressed may play dual roles: inhibiting IAP function and, in addition, activating other apoptotic pathways (Figure 7). DIAP1 Is Epistatic to rpr, hid, and grim To explore further the relationship between rpr, hid, grim, and DIAP1, we made a double mutant with the H99 deficiency (which includes rpr, hid, and grim) and th5. Homozygous H99 embryos essentially lack apoptosis (White et al. , 1994). However, double mutants show a terminal embryonic phenotype similar to that of th5 alone (data not shown). This, in conjunction with the fact that DIAP1 is a caspase inhibitor, argues against models in which DIAP1 ' s sole antiapoptotic role is to bind to RPR, HID, and GRIM, thereby suppressing death pathways they activate, since in this scenario, the double mutant phenotype would be expected to be similar to that of the H99 mutant alone. Instead, our observations are consistent with models in which DIAP1 has an important antiapoptotic function as a caspase inhibitor and RPR, HID, and GRIM function, at least in part, through DIAP1 to promote apoptosis (Figure 7). DIAP1 may, however, have antiapoptotic functions in addition to caspase inhibition. IAPs in insects and mammals have been shown to bind a number of different proteins through the BIRs in addition to caspases (reviewed in Lacasse et al., 1998). An attractive hypothesis is that some of these proteins have proapoptotic functions that do not involve disrupting IAP-caspase interactions and that death-preventing IAPs are able to suppress apoptosis induced by their expression by binding and sequestering them, preventing access to their targets. Candidates for such proteins in Drosophila are RPR, HID, GRIM, and DOOM (Harvey et al., 1997), as well as uncharacterized molecules that mediate normally occurring apoptotic cell death that occurs independently of expression of genes in the H99 interval (White et al., 1994; Foley and Cooley, 1998; Figure 7). DIAP1, and by extension other IAPs, may function to block death at multiple steps : at an upstream point, free DIAP1 may titrate apoptosis inducers away from their targets, DIAP1-caspase complexes and components of other effector pathways, while at a downstream point it inhibits caspase activation or activity induced by these proteins (Figure 7). Concluding Remarks IAPs are the only cellular caspase inhibitors identified to date. We showed that RPR, GRIM, HID, and HID fragments sufficient to cause the death of insect cells block DIAP1 's ability to inhibit caspase activity. These observations provide a mechanism for RPR, HID, and GRIM action and define a novel point at which caspase activity can be regulated. The finding that the DIAP1 loss-of-function phenotype involves cell death associated with increased caspase activity strongly suggests that the interactions we described between RPR HID GRIM, DIAP1, and caspases are physiologically i~por'. tant. Since most mammalian cells constitutively express caspases sufficient to carry out cell death (Weil et al., 1996), the mechanism of cell death activation defined by these proteins may be quite general. A prediction of our model of RPR, HID, and GRIM function is that their ability to kill will depend on the status of a cell's caspases. In cells in which caspases are activated or spontaneously undergoing autoactivation at some level, their expression may be sufficient to kill. However, in other cells in which caspase activity is either absent or more tightly regulated, they may function primarily to create a permissive condition for caspase-dependent cell death. Thus, just as IAPs in their role as caspase inhibitors may function as cellular buffers that create a requirement for a certain level of caspase activation in order to induce apoptosis, so RPR-, HID-, and GRIM-like proteins may function to titrate this buffering activity so that cells are more or less sensitive to caspase-dependent death signals. Experimental Procedures Yeast Strain Constructions All yeast strains used in thi s study are in th e W303 a background (MA Ta , can1-100, /eu2-3 and -112, his3 -11 and -15, trp1-1, ura3-1, ade2-1) . Construction of pGALL-(H/5 3), pGALL-(LEU2), and pGALLDCP-1 wa s described previously (Hawkin s et al., 1999). The p10 and prodomain-p20 fragments of drlCE were amplified by PCR and inserted in tandem into pGALL-(LEU2) to make pGALL-rev-drlCE. pAdh-(TRP7) was made by subcloning the promoter-terminator region from pHybZeo (lnvitrogen) into pRS314 . pCUP1 -( TRP1) was con structed by replacing th e GAL1 promot er of pGALL-(TRP7) (Hawkins et al. , 1999) with th e CUP1 promoter. The DIAP1 coding region was inserted into pAdh-(TRP7) and pCUP1 -(TRP7) to generate pAdh -DIAP1 and pCUP-DIAP1 . The HID coding region was amplified by PCR using primers that appended a C-terminal double DIAP1 Is Essential and Is Inhibited by HID 461 151 polyoma epitope tag and was inserted into pGALL-(HIS3) to generate pGALL-HID. pGALL-HID was digested with Xhol and religated to make pGALL-HIDil(38-335). C-terminally polyoma-tagged versions of HID initi ating at codon 37 [HIDil(l-36)] and HID lacking the first 335 amino acids [HID~(l-335)) were amplified by PCR and inserted into pGALL-(HIS3). The coding regions of RPR and GRIM were amplified by PCR and inserted into pGALL-(Hl53) to generate pGALL-RPR and pGALL-GRIM. Cell Death Assays Overexpression of active caspases in yeast results in cell death that is inhibited by coexpression of caspase inhibitors (Hawkins et al., 1999; Kang et al., 1999). Because expression of RPR, HID, or GRIM alone has no effect on yeast growth (below), we infer that the yeast growth suppression seen in cells that express caspases, DIAP1 and RPR, HID, or GRIM is due to cell death. To assay caspase-dependent cell death and protection by inhibitors, exponentially growing cultures of individual W303cx transformants containing the relevant constructs were serially diluted 10-fold and spotted onto 2% glucose selective medium plates or 2% galactose and 1% raffinose (gal/ rat)-indu c ing selective medium plates. Plates were incubated at 25°C for 12 hr and then at 30°C and photographed after 2 (glucose) or 3-5 (gal/rat) days . Expression and Purification of Recombinant Drosophila IAPs, Caspases, and HID DCP-1"-His, (DCP-1-His6) was prepared as described previously (Hawkins et al., 1999) and diluted into buffer A (50 mM HEPES [pH 7.51, 100 mM NaCl, 1 mM EDTA, 0.1% CHAPS, 10% sucrose, 5 mM OTT). The DIAP1 coding region was amplified by PCR usi ng primers that generated an N-terminal TEV protease recognition site (ENLYFQG) and was introduced into the glutathione S-transferase (GST) expression vector pGEX4T-1 (Amersham Pharmacia). The GSTTEV-DIAP1 fusion protein was expressed in Escherichia coli strain BL21(DE3)plysS and was affinity purified on glutathioneSepharose. The bound fusion protein was cleaved with TEV protease in buffer A. The coding region of HID was introduced into pET23a( + ) (Novagen), generating a HID-His6 fusion. A similar procedure was used to generate a version of HID initiating at codon 37 [HIDil(1-36)His6]. The first 37 amino acids of HID and the coding region of GST were PCR amplified and introduced in tandem into pET23a( + ) to generate HID(1-37)GST-His6. GST was introduced into pET23a( + ) to generate GST-His6. These vectors were introduced into BL21(DE3)plysS, protein expressed, and affinity purified from the soluble fraction using ProBond resin (lnvitrogen). Proteins were eluted and stored at - 80°C in buffer D (50 mM HEPES [pH 8.0), 100 mM NaCl, 1 mM EDTA, 0.1 % CHAPS, 10% sucrose, 20 mM 2-mercaptoethanol, 1% Triton X- 100, 20% glycerol, 0.55 M imidazole). A caspase cleavage site (DQVQ) at residue 20 of DIAP1 (S. L. W., unpublished) was altered to DQVE: in GST-mycDIAPl (Hawkins et al., 1999), generating GST-DIAP1. GST and GST-DIAP1 were purified on glutathione-Sepharose beads. The mutated myc -DIAP1 cod ing sequence was inserted into pMAL-c2 (New England Biolabs), generating MBP-DIAP1. MBP and MBP-DIAP1 were purified on amylose resin. In Vitro Protease Assays HID's ability to inhibit DIAP1 was determined from caspase activity assay progress curves in which the release of 7-amino-4-trifluoromethyl-coumarin (AFC) from the synthetic substrate Ac-DEVD-AFC (100 µM) by DCP-1-His6 (0.2 nM) was measured in the presence of GST (0.2 µM) or DIAP1 (0.1 µM) and HID-His6 (0.44 µM), HIDil(1-36)His6 (0.44 µM), or GST (0.77 µM). Caspase activity assays were performed in triplicate and prepared as follows. Two microliters DCP-1-His6 was incubated with 5 µI DIAP1 or GST for 5 min at 25°C. Seventy-one microliters buffer X (70.4 mM HEP ES [pH 7.51, 140.8 mM NaCl, 1.4 mM EDTA, 0.14% CHAPS, 14.1% sucrose, 3.5 mM OTT, 5.6 % glycerol, and 0.7% Triton X- 100) and 20 µI HID His6, HID~(1-36)-His6, or GST-His6 was subsequently added and th e samples further incubated 23 min. After addition of 2 µI 5 mM Ac-DEVD-AFC in DMSO to eac h sample, caspase activities were assayed at 27°C using a fluorometric plate reader (Molecular Devices, fmax) in the kinetic mode with excitation and emission wavelengths of 405 and 510 nm, respectively. HID(1-37)GST-His6 activity, used at a concentration of 1.6 µ.M, was tested in a similar manner. Caspase activity of embryos was monitored as follows. Zero to two-hr-old embryos were collected and aged for 3 hr at 25°C. Embryos were dechorionated with 50% bleach, rinsed , suspended in an eq ual volume of buffer E (buffer A containing 0.5% Triton X-100 and 4% glycerol), and homogenized . One microliter of this extra ct (approximately 10 mg/ml) was diluted into 100 µI of buffer E in the presence of 100 µM Ac-DEVD-AFC. Assays were performed in triplicate with freshly prepared extracts. Protein Binding As says Purified re combinant proteins (150 ng each) were incubated with either GST beads, MBP beads, GST-DIAP1 beads, or MBP-DIAP1 beads (200 ng) for 30 min on a rot ator at 25°C in buffer C (100 mM HEPES [pH 7.51, 200 mM NaCl, 2 mM EDTA, 0.2% CHAPS, 20% sucrose, 40 mM 2-mercaptoethanol, 1% Triton X-100, 8% glycerol). Beads were then washed three times for 5 min with 1 ml buffer D lacking imidazole. Bead-protein complexes were subjected to 15% SOS-PAGE and immunoblotting. Proteins were detected with an anti-His antibody (Qiagen) and HRP-conjugated secondary antibodies followed by ECL (Amersham). HID-DCP-1 Competition Assay MBP or MBP-DIAP1 beads (0.5 µg) were mixed with 1, 5, or 20 µg GST-His6 or HID(1-37)GST-His6 for 1 hr at 4°C in buffer A. Five micrograms of DCP-1-His6 was added and the samples further incubated for 10 min. After extensive washing, the bead-protein complexes were subjected to 15% SOS -PAGE and immunoblotting . Proteins were detected with an an ti-Hi s antibody and HRP-conjugated secondary antibodies followed by ECL. Drosophila Strains A w Oregon R strain was used as wild type. The deficiency Df(3L)brm11 (72A3-4; 72D1-5) uncovers DIAP1. DIAP1 mutants, alleles of the th locus, and Df(3L}brm 11 were kept as balanced stocks over a version of the TM3 balancer that carries an ftz::lacZ transgene (TM3[ftz::lacZ)). th' is a loss-of-function mutation in DIAP1, since Df(3L}brm11 shows a GMR-rpr enhancer phenotype similar to that of th' (Hay et al., 1995), and the embryonic lethal phenotype of th' ! Df(3L}brm11 (this work, data not shown) is similar to that of th'lth'. By these same criteria, th' is a much weaker allele. Cell Death Assays and Microscopy Embryos were collected on yeasted apple juice plates. For observation of living embryos, syncytial blastoderm stages were selected under halocarbon oil 27 (Sigma) and mounted individua lly on a slide using 1O mm coverslips. Images were taken on a Zeiss Axiophot from sets of six embryos per experim ent every 15 min at 25°C. For scanning electron microscopy, staged collections of embryos were processed for staining with anti-~-galactosidase (~-gal) antibodies (Cappel) as described elsewhere (MOiier and Wieschaus, 1996). Homozygotes were isolated under the dissecting scope; the remaining embryos served as controls. The embryos were then processed for scanning electron microscopy as described in MOiier and Wieschaus (1996). Samples were viewed under a Hitachi S800 SEM. TUNEL labeli ng was performed using an in situ cell death detection kit (Roche). Embryos were fixed as described above, rinsed in PTX (PBS, 0.5% Triton X-100), and immunolabeled with antibodies against ~-ga l (Cappel) and Cy3-conjugated goat anti-rabbit (Jackson) secondary antibodies. Embryos were then incubated in 100 mM sodium-citrate, 0.1 % Triton X-100 at 65°C for 30 min, followed by rinsing twice in PTX and twic e in TUNEL dilution buffer (Roche). Embryos incubated in TUNEL assay buffer (Roche; 30 min to 2 hr) were supplemented with TUNEL enzyme (Roche) and incubated a furth er 2 to 3 hr at 37°C. Embryos were rin sed in PTX and mounted in MOWIOL (Polysciences) containing DABCO (Sigma). Images were taken on a Leica TSC-NT confocal microscope . A range of 50 µm in depth was scanned in 16 optica l sections and projected into one focal plane using th e TCS-NT software. Cell 462 152 Acridine orange staining was carried out as described in Hay et al. (1994). Confocal and Nomarksi images were taken on th e confocal microscope. Evans, E.K., Kuwana, T. , Strum, S.L., Smith , J.J., Newmeyer, D.D., and Kornbluth, S. (1997). Reaper-induced apoptosis in a vertebrate system. EMBO J. 76, 7372-7381 . Acknowledgments Foley, K., and Cooley, L. (1998). Apoptosis in late stage Drosophila nurse cel ls does not require genes within the H99 deficiency. Development 725, 1075-1082. We thank R. Deshaies and members of his lab for advice with protein binding and yeast experiments and H. Schwarz and J. Berger (MPI TObingen, Germany) for the use of their SEM. We also thank R. Deshaies, P. Sternberg, and M. Guo for comments on the manu script. H.-A. J. M. th anks E. Wieschaus for discussions. C. J. H. is supported by a Human Frontiers postdoctoral fellowship. B. A. H. is a Searle Scholar. Thi s work was supported by a DFG research fellowship (MUl 168/1-1) to H.-A. J. M. and by grants to B. A. H. from the Gustavus and Louise Pfeiffer Research Foundation, a Burroughs Wellcome Fund New Investigator Award in the Pharmacologi ca l Sciences, the Ellison Medical Foundation, and National Institutes of Health Grant GM057422-01. Received May 28, 1999; revised July 13, 1999. References Abrams, J.M., White, K., Fessler, L.I., and Steller, H. (1993). Programmed cell death during Drosophila embryogenesis. Development 117, 29-43. Alnemri, E.S. , Livingston, D.J., Nicholson, D.W., Salvesen, G., Thornberry, N.A., Wong, W.W. , and Yua n, J. (1996). Human ICE/ CED3 protease nomenclature. Cell 87, 171. Avdonin, V., Kasuya, J., Ciorba, M.A., Kaplan, B., Hoshi , T., and Iverson, L. (1998). Apoptotic proteins Reaper and Grim induce stable inactivation in voltage gated K• channels. Proc. Natl. Acad. Sci. USA 95, 11703-11708. Bergmann, A., Agapite, J., McCall, K., and Steller, S. (1998). The Drosophila gene hid is a direct molecular target of Ras-dependent survival signaling. Cell 95, 33 1-341. Birnbaum, M.J., Clem, R.J., and Miller, L.K. (1994). An apoptosisinhibiting gene from a nuclear polyhedrosis virus encoding a peptide with Cys/His seq uence motifs. J . Viral. 68, 2521-2528. Cavaliere, V., Taddei, C., and Gargi ulo, G. (1998). Apoptosis of nurse cells at the late stages of oogenesis of Drosophila melanogaster. Dev. Genes Evol. 208, 106-112. Chen, P., Nordstrom, W., Gish, B., and Abrams, J.M. (1996). grim, a novel cell death gene in Drosophila. Genes Dev. 10, 1773-1782. Chen, P., Rodriguez, A., Erskine, R., Thach, T., and Abrams, J.M. (1998). Dredd, a novel effector of th e apoptosis activators reaper, grim and hid in Drosophila. Dev. Biol. 201, 202-216. Crook, N.E., Clem, R.J., and Miller, L.K. (1993). An apoptosis-inhibiting baculovirus gene with a zinc finger-like motif. J. Viral. 67, 21682174. Cryns, V., and Yuan, J. (1998). Proteases to die for. Genes Dev. 72, 1551-1570. Deveraux, Q.L. , and Reed, J.C. (1999). IAP family proteinssuppressors of apoptosis. Genes Dev. 73, 239-252. Deveraux, Q.L. , Takahashi, R., Salvesen, G.S., and Reed, J.C. (1997). X-linke d IAP is a direct inhibitor of cell-d eath proteases. Nature 388, 300-304. Deveraux, Q.L., Roy, N., Stennicke, H.R., Arsdale, T.V., Zhou, Q., Srinivasula, S.M., Alnemri, E.S., Sa lvesen, G.S., and Reed, J.C. (1998). IAPs block apoptotic events induced by caspase-8 and cytochrome c by direct inhibition of distinct caspases. EMBO J. 77, 2215-2223. Dorstyn, L., Colussi, P.A., Quinn, L.M., Richardson, H., and Kumar, S. (1999). DRONC, an ecdysone-inducible Drosophila caspase. Proc. Natl. Acad. Sci. USA 96, 4307-4312. Duckett, C.S., Nava, V.E., Gedrich, R.W., Clem, R.J., Van Dongen, J.L., Gilfillan, C., Shiels, H. , Hardwick, J.M ., and Thompson, C.B. (1996). A conserved family of cellular genes related to the baculovirus iap gene and encoding apoptosis inhibitors. EMBO J. 75, 26852694. Fraser, A.G., and Evan , G.I. (1997). Identification of a Drosophila melanogaster ICE/CED-3-related protease, drlCE. EMBO J. 76, 2805-2813. Fra ser, A.G., McCarthy, N.J., and Evan, G.I. (1997). drlCE is an essential caspase required for apoptotic activity in Drosophila cells. EMBO J. 20, 6192-6 199. Grether, M.E., Abrams, J.M ., Agapite, J., White, K., and Steller, H. (1995). The head involution defective gene of Drosophila melanogasrer functions in programmed cell death. Genes Dev. 9, 1694-1708. Harvey, A.J., Bidwai, A.P., and Miller, L.K. (1997). Doom, a product of the Drosophila mod(mdg4) gene, induces apoptosis and binds to baculovirus inhibitor-of-apoptosis proteins. Mol. Cell. Biol. 77, 2835-2843. Hawkins, C.J., Wang, S.L., and Hay, B.A. (1999). A cloning method to identify caspases and their regulators in yeast: identification of Drosophila IAPl as an inhibitor of the Drosophila caspase DCP -1 . Proc. Natl. Acad. Sci. USA 96, 2885-2890. Hay, B.A., Wolff, T., and Rubin, G.M. (1994). Expression of baculovirus p35 prevents cell death in Drosophila. Development 120, 21212129 _ Hay, B.A., Wassarman, D.A. , and Rubin, G.M. (1995). Drosophila homologs of baculovirus inhibitor of apoptosis proteins function to block cell death. Cell 83, 1253-1262. lnohara, N., Koseki, T., Hu, Y.M., Chen, S., and Nunez, G. (1997). CLARP, a death effector domain-containing protein interacts with caspase-8 and regulates apoptosis. Proc. Natl. Acad. Sci. USA 94, 10717-10722. Kaiser, W.J., Vucic, D., and Miller, L.K. (1998). The Drosophila inhibitor of apoptosis D-IAPl suppresses cell death induced by the caspase drlCE. FEBS Lett. 440, 243- 248. Kang, J.K., Schaber, M.D., Srinivasula, S.M., Alnemri , E.S., Litwack, G., Hall, D.J., and Bjornsti, M.A. (1999). Cascades of mammalian caspase activation in the yeast Saccharomyces cerevisiae. J. Biol. Chem. 274, 3189-3 198. Lacasse, E.C., Baird, S., Korneluk, R.G., and MacKenzie, A.E. (1998). The inhibitors of apoptosis (IAPs) and th eir emerging role in cancer. Oncogene 17, 3247-3259. McCall, K., and Steller, H. (1998). Requirement for DCP-1 caspase during Drosophila oogenesis. Science 279, 230-234. MOiier, H.A., and Wieschau s, E. (1996). armadillo, bazooka, and stardust are critical for early stages in formation of the zonula adherens and maintenance of the polarized blastoderm epithelium in Drosophila. J. Cell Biol. 134, 149-163. Nicholson, D.W., and Thornberry, N. (1997) . Caspases: killer proteases. Trends Biochem. Sci . 22, 299-306. Raff, M.C. (1992). Social controls on cell survival and cell death. Nature 356, 397-400. Roy, N., Deveraux, Q.L., Takahashi, R., Salvesen, G.S., and Reed, J.C. (1997). The c-lAP-1 and c-lAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 76, 6914-6925. Salvesen, G.S., and Dix it, V.M. (1997). Caspases: intracellular signaling by proteolysis. Cell 91, 443-446. Song, Z.W., McCall, K., and Steller, H. (1997). DCP-1, a Drosophila cell death protease essential for development. Science 275, 536-540. Srinivasu la , S.M., Ahmed, M., MacFarlane, A.M., Luo, Z.W., Fernandes-Alnemri, T., and Alnemri, E.S. (1998). Generation of constitutively active recombinant caspases-3 and -6 by rearrangement of their subunits. J. Biol. Chem. 273, 10107-10111. Takahashi, R., Deveraux, Q.L., Tamm, I., Welsh, K., Assa-Munt, N., Salvesen G., and Reed, J.C. (1998). A single BIR domain of XIAP sufficient for inhibiting caspases. J. Biol. Chem. 273, 7787-7790. Thornberry, N.A. , and Lazebnik , Y. (1998). Caspases: enemies within. Science 281, 1312-1316. DIAP1 Is Essential and Is Inhibited by HID 463 153 Thress. K., Henzel, W., Shillinglaw, w., and Kornbluth, S. (1998). Scythe: a novel reaper-binding apoptotic regulator. EMBO J. 77, 6135-6143. Uren, A.G., Pakusch, M., Hawkins, C.J., Puls, K.L., and Vaux, D.L. (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. Uren, A.G .. Coulson, E.J., and Vaux, D.L. (1998). Conservation of baculovirus inhibitor of apoptosis repeat proteins (BIRPs) in viru ses, nematodes, vertebrates and yeasts. Trend s Biochem . Sci. 23, 159-162. Vucic, D., Kaiser, W.J., Harvey, A.J., and Miller, L.K. (1997). Inhibition of Reaper-induced apoptosis by interaction with inhibitor of apoptosis proteins (IAPs). Proc. Natl. Acad. Sci. USA 94, 10183-10188. Vucic, D .. Kaiser, W.J., and Miller, L.K. (1998). Inhibitor of apoptosis proteins physically interact with and block apoptosis induced by Drosophila proteins HID and GRIM. Mal. Cell. Biol. 17, 667-676. Weil, M ., Jacobson, M.D .. Coles, H.S.R., Davies, R.L., Gardner, R.L., Raff, K.D., and Raff, M.C. (1996). Constitutive expression of th e machinery for programmed cell death. J. Cell Biol. 133, 1053-1059. White, K. , Grether, M.E., Abrams, J.M., Young, l., Farrell, K. , and Steller, H. (1994). Genetic control of programmed cell death in Drosophila. Science 264, 677-683. White, K., Tahaoglu, E., and Steller, H. (1996). Cell killing by the Drosophila gene reaper. Science 271, 805-807. Wing, J.P., Zhou, L., Schwartz, L.M. , and Nambu, J.R. (1998) . Distinct cell killing properties of the Drosophila reaper, head involution defective, and grim genes. Cell Death Differ. 5, 930-939. Wyllie, A.H., Kerr, J.F.R .. and Currie, A.R. (1980). Cell death: th e significance of apoptosis. Int. Rev. Cytol. 68, 251-303. Zhou, L., Schnitzler, A .. Agapite, J., Schwartz, L.M ., Steller, H., and Nambu, J.R. (1997). Cooperative functions of the reaper and head involution defective genes in the programmed cell death of Drosophila central nervous system midline cells. Proc. Natl. Acad. Sci. USA 94, 5131-5136.