IAP-Targeted Therapies for Cancer

EC LaCasse1,2,3, DJ Mahoney1,2,3, HH Cheung1,2,3, S Plenchette1,2, S Baird1,2 and RG Korneluk1,2

1Apoptosis Research Centre, Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada and 2Department of Biochemistry, Microbiology, and Immunology, University of Ottawa, Ottawa, Ontario, Canada

DNA damage, chromosomal abnormalities, oncogene to daily stresses and insults. In this review, we activation, viral infection, substrate detachment and provide an overview of the mammalian IAPs, with hypoxia can all trigger apoptosis in normal cells. particular emphasis on their many cellular roles in However, cancer cells acquire mutations that allow them apoptosis suppression, signal transduction and to survive these threats that are part and parcel of the proliferation. In addition, we address the relationship transformation process or that may affect the growth and of the IAPs with cancer, and the various therapeutic dissemination of the tumor. Eventually, cancer cells modalities targeting these gene products for clinical accumulate further mutations that make them resistant benefit. to apoptosis mediated by standard cytotoxic chemo- The first discovered cellular, non-viral, IAP is the therapy or radiotherapy. The inhibitor of apoptosis mammalian gene NAIP (Roy et al., 1995). The human (IAP) family members, defined by the presence of a IAP family has rapidly expanded to include seven other baculovirus IAP repeat (BIR) protein domain, are key members: XIAP, cIAP1, cIAP2 (Rothe et al., 1995; regulators of cytokinesis, apoptosis and signal transduc- Duckett et al., 1996; Liston et al., 1996; Uren et al., tion. Specific IAPs regulate either cell division, caspase 1996), ILP2 (Lagace et al., 2001; Richter et al., 2001), activity or survival pathways mediated through binding to BRUCE (Hauser et al., 1998; Chen et al., 1999), survivin their BIR domains, and/or through their ubiquitin-ligase (Ambrosini et al., 1997) and livin (Vucic et al., 2000; RING domain activity. These protein–protein interactions Kasof and Gomes, 2001) (Figure 1 and Table 1). The and post-translational modifications are the subject of IAPs, deserving of their name, effectively suppress intense investigations that shed light on how these proteins apoptosis induced by a variety of stimuli, including contribute to oncogenesis and resistance to therapy. In death receptor activation, growth factor withdrawal, the past several years, we have seen multiple approaches ionizing radiation, viral infection, endoplasmic reticu- of IAP antagonism enter the clinic, and the rewards of lum stress and genotoxic damage (LaCasse et al., 1998; such strategies are about to reap benefit. Significantly, Liston et al., 2003; Cheung et al., 2006b; Hunter et al., small molecule pan-IAP antagonists that mimic an 2007). endogenous inhibitor of the IAPs, called Smac, have The defining characteristic of the IAPs is a baculo- demonstrated an unexpected ability to sensitize cancer virus IAP repeat (BIR) domain. IAPs contain 1–3 BIR cells to tumor necrosis factor-a and to promote autocrine domains of 70–80 amino acids encoding a C2HC-type or paracrine production of this cytokine by the tumor cell zinc-finger motif that tetrahedrally chelates one zinc and possibly, other cells too. This review will focus on atom, and forms a globular structure consisting of four these and other developmental therapeutics that target the or five a-helices and a variable number of antiparallel IAPs in cancer. b-pleated sheets (Hinds et al., 1999; Sun et al., 1999). Oncogene (2008) 27, 6252–6275; doi:10.1038/onc.2008.302 The two human IAPs, survivin and BRUCE, both with a role in cell mitosis (Ruchaud et al., 2007; Keywords: XIAP; cIAP1; cIAP2; survivin; livin; birc Pohl and Jentsch, 2008), have an earlier evolutionary origin compared with the type 1 human members with roles in apoptosis and immunity (Robertson et al., 2006). In addition to BIR domains, IAPs contain various Introduction other domains that are summarized in Figure 1. Several mammalian IAP family members contain a carboxy- The inhibitor of apoptosis (IAP) gene family terminal RING (really interesting new gene) zinc-finger regulates the cell’s decision to live or die in response domain. In XIAP, cIAP1 and cIAP2, this RING domain has been shown to possess E3 ubiquitin ligase activity, directly regulating auto- or trans-ubiquitination Correspondence: Dr EC LaCasse, Apoptosis Research Centre, and protein degradation (Yang et al., 2000; Suzuki et al., Children’s Hospital of Eastern Ontario, CHEO RI2, 401 Smyth Road, Ottawa, Ontario, Canada K1H 8L1. 2001; Silke et al., 2005; Cheung et al., 2008). E-mail: [email protected] Within the IAP family, the presence of a CARD 3These three authors contributed equally to this work. domain is unique to cIAP1 and cIAP2 (Figure 1). The IAP-targeted therapeutics EC LaCasse et al 6253 role of the CARD domain in the IAPs is unknown but still exist. NAIP is unique among the IAPs in that it they undoubtedly function in protein–protein inter- possesses a nucleotide-binding and oligomerization actions (Martin, 2001). Notably, a functional nuclear domain as well as leucine-rich repeats that classify this export signal exists within the CARD domain of cIAP1, IAP along with other proteins involved in innate which appears to be important for cell differentiation immunity, called the nucleotide-binding and oligomer- (Plenchette et al., 2004), revealing one such role for ization domain-like receptors (Kanneganti et al., 2007; the CARD domain. However, other functions likely Mariathasan and Monack, 2007; Ting et al., 2008). BRUCE, the largest of the IAPs, lacks a RING domain, but instead possesses an ubiquitin-conjugating domain NAIP BIR1 BIR2 BIR3 NOD LRR capable of performing a similar function. Survivin, the smallest IAP, possesses a coiled-coil domain required for cIAP1 BIR1 BIR2 BIR3 CARD RING its interaction with chromosomal passenger proteins, INCENP and borealin, and for the maintenance of cIAP2 BIR1 BIR2 BIR3 CARD RING residency in the nucleus (Chantalat et al., 2000; Engelsma et al., 2007; Jeyaprakash et al., 2007) XIAP BIR1 BIR2 BIR3 RING (Figure 1). Several alternatively spliced mRNA products for the ILP2 BIR RING survivin, livin and cIAP2 genes exist that lead to various protein isoforms of differing size and function (Conway livin BIR RING et al., 2000; Ashhab et al., 2001; Crnkovic-Mertens et al., 2006; Mosley and Keri, 2006; Noton et al., 2006; Kappler et al., 2007; Knauer et al., 2007; Mola et al., survivin BIR CC 2007; Nachmias et al., 2007; Hu et al., 2008) (Table 1). BRUCE BIR UBC The presence of the two closely related genes, cIAP1 and cIAP2, in mammals is confusing and has led to some Figure 1 Domain structure of the mammalian inhibitor of misidentification errors in databases and has produced apoptosis (IAP) family. The domain organization is shown for some confusion in the published literature. This gene the eight mammalian IAPs. The presence of at least one baculovirus IAP repeat (BIR) domain is the defining characteristic duplication event is fairly recent, as zebrafish only of the IAP family. The human IAPs possess either one (survivin, possess one cIAP gene (Santoro et al., 2007). Ts-IAP/ BRUCE, livin and ILP2) or three tandem amino-terminal BIR hILP2 gene is present on an autosomal chromosome domains (XIAP, cIAP1, cIAP2 and NAIP). Several IAPs contain (Lagace et al., 2001) (Table 1 and Figure 1) and found an E3 ubiquitin ligase zinc-finger (RING) domain at the carboxy terminus. Both cIAP1 and cIAP2 possess a caspase-recruitment only in great apes (Richter et al., 2001). Ts-IAP (CARD) domain in the linker region between the BIR and the produces a mutated, truncated and unstable version of RING domains. Uniquely, NAIP possesses a nucleotide-binding XIAP that, because of its autosomal location, is not and oligomerization (NOD) domain, as well as a leucine-rich repeat subject to X-chromosome inactivation (Lagace et al., (LRR) domain seen in other NOD-like receptor (NLR) proteins 2001; Shin et al., 2005). involved in innate immunity. BRUCE contains an ubiquitin- conjugating (UBC) domain but no RING domain. Survivin contains a coiled-coil (CC) domain that binds chromosomal passenger proteins, INCENP and borealin. The BIR domains IAP proteins: multifaceted inhibitors of apoptosis shown in green are longer type 2 domains found in the IAPs with roles in mitosis, compared with the type 1 BIR domains, in black, with roles in signal transduction and apoptosis suppression. Note Inhibition of caspase function that the drawings are not to scale and that the length of the IAP All apoptotic signaling pathways converge on the proteins, and their splice isoforms, is shown in Table 1. caspases making these proteases the ultimate effectors

Table 1 Properties of the human (and mouse) IAP genes and proteins IAP DNA locus Full length (aa) Splice isoformsa (aa) Principal function Cancer targetb

NAIP (birc1) 5q13.2 (13 D1) 1403 (1403, 1447) Specialized (innate immunity, neuroprotection) cIAP1 (birc2) 11q22 (9 A1) 618 (612) (>450) Signal transduction (TNFR, NF-kB activation) Yes cIAP2 (birc3) 11q22 (9 A1) 604 (600) (496, 516) Signal transduction (TNFR, NF-kB activation) Yes XIAP (birc4) Xq25 (X A3.2) 497 (496) Apoptosis inhibition (general apoptosis suppression) Yes Survivin (birc5) 17q25 (11 E2) 142 (140) 74, 137, 165 (48, 121) Mitosis (cytokinesis) Yes BRUCE (birc6) 2p22 (17 E2) 4857 (4854) Mitosis (cytokinesis) Livin (birc7) 20q13.3 (2 H4) 298 (271) 280 Specialized (developmental, other?) Yes ILP2 (birc8) 19q13.42 (NA) 236 (NA) Specialized (spermatogenesis?)

Abbreviations: aa, amino acid; IAP, inhibitor of apoptosis; NF-kB, nuclear factor-kB; TNFR, tumor necrosis factor receptor. aValues derived from searches with the ASD bioinformatics resource on alternative splicing (Stamm et al., 2006), or publications cited in the text. bIAPs for which antisense, small molecule or immunotherapy approaches are under clinical development for the potential treatment of cancer (see Table 3). Entries within brackets in DNA locus, protein and splice isoforms are for mouse. NA means the gene does not exist.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6254 of apoptotic cell death. Not surprisingly, caspases need domain of XIAP. It is evident that caspase-9 is inhibited to be tightly regulated, as their inappropriate activation in direct response to BIR3 binding, thereby preventing can have severe consequences. The first level of caspase further caspase-9 activation as well as suppressing regulation is seen in their structure and activation. downstream effector caspase activity. Thus far, none Caspases are synthesized as inactive zymogens and are of the IAP BIR1 domains has been shown to have any only activated through signaling through strictly con- caspase-inhibiting activity; however, Akt/PKB (protein trolled pathways. A second level of regulation involves kinase B)-mediated phosphorylation at serine 87 within the specific inhibition of active caspases by naturally BIR1 is suggested to reduce auto-ubiquitination and occurring cellular inhibitors. stabilizes XIAP (Dan et al., 2004; but see Cheung et al., The IAPs are the only known endogenous proteins 2008). BIR1 therefore, appears to play a regulatory role that regulate the activity of both initiator and effector rather than directly participating in apoptosis suppres- caspases. Controlled expression of the IAPs has been sion. The BIR1 domains of cIAP1 and cIAP2, for shown to influence cell death in a variety of contexts and example, are proposed to play a role in tumor necrosis is believed to have important consequences with respect factor (TNF) receptor-associated factor (TRAF) inter- to human cancer (LaCasse et al., 1998). The mechanism actions and ubiquitination reactions (Vaux and Silke, by which the IAPs inhibit apoptosis was first inter- 2005; Samuel et al., 2006). Furthermore, the BIR1 rogated in the laboratory of John Reed (Deveraux et al., domain of XIAP interacts with TAB1 (KD ¼ 14.3 mM) 1997, 1998). In these early studies, XIAP was found to and induces nuclear factor-kB (NF-kB) activation prevent caspase-3 processing in response to caspase-8 through the TAK1 pathway (Lu et al., 2007). activation. Therefore, XIAP was suggested to inhibit the extrinsic apoptotic signaling by blocking the activity of the downstream effector caspases, as opposed to Ubiquitination and RING E3 ligase function of the IAPs interfering directly with caspase-8 activation (Deveraux The post-translational modification of proteins with et al., 1997, 1998). Supporting this concept, XIAP was ubiquitin chains is an important regulatory mechanism shown to specifically bind to caspase-3 and -7, but not to that can result in their proteasomal degradation or allow caspase-1, -6, -8 or -10 (Deveraux et al., 1997, 1998). In signal transduction to occur, depending on the type of vitro assays confirmed that XIAP, as well as cIAP1 and ubiquitin linkage used. Typically, a K48 ubiquitin- cIAP2, could prevent downstream proteolytic proces- linkage destines the protein to the proteasome, whereas sing of pro-caspase-3, -6 and -7 by blocking cytochrome a polyubiquitin chain based on K63 is recognized c-induced activation of pro-caspase-9 in the intrinsic for cellular signal transduction (Hochstrasser, 2006; pathway (Deveraux et al., 1998). Hunter, 2007). The ubiquitination and subsequent The role of the IAPs in apoptosis suppression is proteasomal degradation or signaling of the IAPs may continuously being evaluated. The Salvesen’s laboratory be a key regulatory event in the apoptotic program. The (Eckelman and Salvesen, 2006; Eckelman et al., 2006) carboxy-terminal RING domain of the IAPs mediates suggests that only XIAP is a direct inhibitor of caspases, both ubiquitin-dependent and -independent degradation and that other IAPs simply bind caspases but do not of the IAPs, as well as that of their substrates (Yang inhibit them. This suggestion is based on the fact that et al., 2000; Suzuki et al., 2001; Silke et al., 2005; Cheung XIAP is relatively stable and exhibits the greatest et al., 2008). RING domain-containing proteins possess potency for caspase inhibition compared with the E3 ubiquitin ligase activity and function as specific other IAPs. adapters by recruiting target proteins to a multi- Importantly, although cIAP1 and cIAP2 can bind to component complex. Earlier studies on the RING caspases, their ability to inhibit caspases in vitro has domains of XIAP and cIAP1 show that this domain is been attributed to an artifact of the glutathione S- involved in the ubiquitination and degradation of the transferase fusion moiety, which allows aggregation and IAPs in response to apoptotic triggers (Yang et al., steric inhibition of the caspases by the cIAP fusions at 2000). Other studies show that XIAP and cIAP1 are able concentrations that are not achieved physiologically to promote the ubiquitination and proteasomal degra- (Eckelman and Salvesen, 2006). Nevertheless, cIAP1 dation of caspase-3 and -7, thereby enhancing the and cIAP2 can be considered true antagonists of caspase antiapoptotic effect of the IAPs (Suzuki et al., 2001). function. For example, cIAP1 and cIAP2 antagonize Although the RING domain is involved in the caspase activity when co-expressed in yeast (Wright ubiquitination and degradation of the IAPs, it remains et al., 2000), rescuing them from cell death. Yeast unclear whether this activity enhances the antiapoptotic represents a simple genetic system devoid of many activity of the IAPs, or actually antagonizes the activity. complicating factors (Jin and Reed, 2002) and is more Studies using RING deletion mutants have provided physiologically relevant compared with in vitro enzyme evidence for both scenarios (Yang et al., 2000; Suzuki systems. However, this yeast system does not address et al., 2001). the issue of ‘direct’ caspase inhibition, as cIAPs may K322 and K328 are the sites of ubiquitination in inactivate caspase through the ubiquitin–proteasomal XIAP, and the role of XIAP ubiquitination has been system, or by other direct or indirect means. examined by site-directed mutagenesis (Shin et al., 2003; In summary, caspase-3 is inhibited exclusively by the Cheung et al., 2008). Compared with XIAP wild type, linker region between BIR1 and BIR2, whereas caspase- XIAP K322/328R mutant protein in cultured cells 7 inhibition requires both the linker region and the BIR2 does not lead to XIAP accumulation and reveal no

Oncogene IAP-targeted therapeutics EC LaCasse et al 6255 differences in their ability to protect against Bax- or complex, TRAF1 and TRAF2 appear to function as Fas-induced apoptosis (Shin et al., 2003). In response to adaptors, bringing cIAP1 or cIAP2 together, allowing Sindbis virus infection, mammalian cells produce a pro- cIAP1 to ubiquitinate cIAP2 (Conze et al., 2005). The apoptotic carboxy-terminal fragment of cIAP1 contain- upregulation of cIAP2 in the absence cIAP1 was ing the CARD and RING domains (cIAP1-CR) (Clem similarly seen in mouse embryonic fibroblasts (MEFs) et al., 2001). Ectopic expression of cIAP1-CR (Cheung (Mahoney et al., 2008) and certain tumor cells (Cheung et al., 2008), as well as cIAP1-RING domain fused to a et al., 2008) transiently silenced by cIAP1-targeted carrier (Silke et al., 2005), targets XIAP for degradation. RNAi. Collectively, these observations support the However, the role of XIAP ubiquitination in this concept that IAP–IAP or IAP–partner interactions play processing appears minimal as XIAP mutants that were roles in regulating their relative abundance, through not ubiquitinated were also readily degraded by cIAP1- post-translational modifications, to impact on the CR (Cheung et al., 2008). These findings suggest that progression of apoptosis. ubiquitin-mediated destruction of the XIAP may not be as significant as believed. Similarly, ubiquitination of livin is likewise not a prerequisite for RING-mediated Modulation of survival signal-transduction pathways turnover (Cheung et al., 2008). In contrast, for the Over the past year, a series of landmark papers (Gaither metabolism of cIAP1 and cIAP2, RING-mediated et al., 2007; Petersen et al., 2007; Santoro et al., 2007; ubiquitination remains essential (Silke et al., 2005; Varfolomeev et al., 2007; Vince et al., 2007; Bertrand Cheung et al., 2008). These divergent responses of IAPs et al., 2008; Mahoney et al., 2008; Wang et al., 2008) to ubiquitination might in part explain the rapid have demonstrated that the cellular IAP proteins (cIAP1 degradation of cIAP1 and cIAP2, but not XIAP, in and 2) are critical regulators of the NF-kB signal response to Smac mimetic compounds (SMCs, see transduction. NF-kB is a pleiotropic signaling pathway below) (Vince et al., 2007; Vucic and Fairbrother, involved in a diverse range of biological processes, 2007). Taken together, these findings indicate that particularly in innate and adaptive immunity, as well as ubiquitination of XIAP and livin, but not cIAP1 and in proliferation and survival. In response to receptor cIAP2, is uncoupled from the regulation of its abun- stimulation by ligands of the TNF family or to various dance, and might yet serve an undetermined role. intracellular stressors (for example, DNA damage or Although the fate of the ubiquitinated cIAPs appears viral infection), a series of signaling events are initiated certain, notwithstanding de-ubiquitination, the ultimate that culminate in the activation of heterodimeric NF-kB destination for the destruction of the ubiquitinated transcription factors. On activation, NF-kB hetero- species depends on the initial trigger. The landmark dimers translocate into the nucleus and transactivate a study of Yang et al. (2000) has shown that in response to large number of target genes, which in turn elicit the glucocorticoid or etoposide, cIAP1 degradation is particular biological response. Depending on how the proteasome dependent. More recently, Vince et al. NF-kB pathway is activated, the response generated can (2008) have shown that TWEAK, a ligand for the be classified into the so-called classical or alternative TNF superfamily receptor FN14, can induce lysosomal NF-kB pathways. Although cross-talk exists, these two degradation of cIAP1 in a complex with TRAF2. pathways generally have distinct functions and exert a Ubiquitin-K29 linkage is known to mediate lysosomal unique biological footprint. Typically, the classical degradation (Chastagner et al., 2006). Whether or not pathway is stimulated by TNFa, DNA damage or cIAP1 degradation in response to TWEAKis facilitated viruses, and occurs in nearly all cell types, exerting its by K29-type linkage, some other exotic type of mixed effect through p65/p50 heterodimers. In contrast, the linkage or other undefined molecular interactions, alternative pathway is often stimulated by CD40-L, remains to be seen. Nonetheless, although ubiquitina- B-cell activation factor (BAFF)-L or TWEAK, occur- tion of the cIAPs is not necessary as a signal to the ring predominantly in lymphoid cells to elicit function proteasome, it does appear to be a common tag for their through p52/RelB heterodimers. destruction. A role for cIAP1 and cIAP2 in NF-kB signaling has Typically, cIAP2 is present in low abundance in the long been implied, although this has only been cell. The levels of cIAP2 are presumably maintained conclusively demonstrated recently. The earliest evi- through constitutive ubiquitination and subsequent dence was generated in David Goeddel’s laboratory, degradation by cIAP1 (Conze et al., 2005; Cheung where cIAP1 and cIAP2 were shown to interact with et al., 2008; Mahoney et al., 2008). Insight into cIAP1 TRAF2 (Rothe et al., 1995; Shu et al., 1996). TRAF2 is E3 ligase function came from the generation of the an adapter protein that functions in both the classical cIAP1-null mouse (Conze et al., 2005). Although these and alternative NF-kB pathways, as well as in mitogen- mice appeared to be both viable and fertile, they had activated protein kinase (MAPK) signaling pathways significantly elevated cIAP2 protein levels with normal (that is, p38 and Jun N-terminal kinase (JNK) signal- mRNA levels, suggesting that in a normal setting, cIAP1 ing). Goeddel’s group also showed that cIAP1 and ubiquitination of cIAP2 leads to low protein levels of cIAP2 get recruited to TNF receptors in response to cIAP2 (Conze et al., 2005). Furthermore, co-expression TNFa, a process that is dependent on their interaction and in vitro binding studies of cIAP1 and cIAP2 (along with TRAF2. Accordingly, it was hypothesized that with TRAF1 or TRAF2), demonstrate that the cIAPs cIAP1 and cIAP2 participate in TNFa-mediated NF-kB and TRAF proteins form a multimeric complex. In this activation. Subsequent overexpression studies supported

Oncogene IAP-targeted therapeutics EC LaCasse et al 6256 this notion, demonstrating that cIAP2 has the (Mahoney et al., 2008) have shown that dual loss of capacity to activate NF-kB (Chu et al., 1997) and cIAP1 and cIAP2 dramatically attenuates TNFa- that a RING-deficient version of cIAP1 can cooperate induced RIP1 ubiquitination, even though both RIP1 in the activation of NF-kB by TRAF2 (Samuel et al., and TRAF2 are still recruited to TNFR1. Moreover, 2006). Bertrand et al. (2008) have demonstrated that cIAP1 Unexpectedly, when the cIAP1 and cIAP2 knockout and cIAP2 directly target RIP1 for K63 (and to a lesser mice were generated, they were found to be asympto- extent K48)-linked polyubiquitination in vitro, whereas matic (Conze et al., 2005; Conte et al., 2006). TRAF2 does neither. Taken together, these data suggest Furthermore, their primary cells displayed normal a model in which TRAF2 serves as an adapter protein TNFa-induced NF-kB activation and were not sensi- that bridges the gap between cIAP1/cIAP2 and RIP1 at tized to TNFa-mediated cell death (Conze et al., 2005). TNFR1, which enables cIAP1/cIAP2 to target RIP1 for As such, the physiological involvement of cIAP1 and K63-linked polyubiquitination. cIAP2 in NF-kB signaling was called into question. Significantly, a recent article by Wang et al. (2008) However, a confounding variable in both of these reported that cIAP1 and cIAP2 negatively regulate knockout mice is the presence of the other highly similar RIP1 de-ubiquitination at TNFR1, thereby protecting cIAP, which may be able to compensate. Unfortunately, cells from TNFa-mediated apoptosis. In this study, the a cIAP1/cIAP2 double knockout mouse has not yet small compound-mediated loss of cIAP1 and cIAP2 been generated, as the chromosomal proximity of these sensitized cancer cells to TNFa-mediated apoptosis, two genes precludes a traditional cross-breeding strat- although oddly there was no effect on NF-kB signaling. egy. To circumvent this, we (Mahoney et al., 2008) and This may have been due to the cell lines used, or because others (Gaither et al., 2007; Petersen et al., 2007; the loss of cIAP1 and cIAP2 in these experiments was Santoro et al., 2007; Varfolomeev et al., 2007; Vince only partial. Wang et al. (2008) also demonstrated et al., 2007; Bertrand et al., 2008; Wang et al., 2008) decreased RIP1 ubiquitination at TNFR1 in response to have combined genetic knockout and siRNA-mediated TNFa when cIAP1 and cIAP2 were partially degraded, knockdown methodologies to generate various versions consistent with a role for cIAP1 and cIAP2 in the of cIAP1 and cIAP2 double ‘null’ cells. Additionally, ubiquitination of RIP1. However, in contrast to our these studies have used potent small molecule com- interpretation (Mahoney et al., 2008) and that of others pounds that induce the specific degradation of cIAP1 (Bertrand et al., 2008), these authors suggested that and cIAP2, thereby rendering cells ‘null’ for their cIAP1 and cIAP2 ‘protect’ RIP1 from de-ubiquitination expression. Using these approaches, cIAP1 and cIAP2 at TNFR1 rather than promote its ubiquitination. have been conclusively shown to be critical regulators of Although further investigation into the role of cIAP1 both classical and alternative NF-kB signal transduction and cIAP2 in the de-ubiquitination of RIP1 is and are functionally redundant in these roles. Impor- warranted, the fact that they can directly ubiquitinate tantly, cIAP1 and cIAP2 regulate these pathways in RIP1 in vitro strongly suggests that the primary function normal as well as cancer cells. of the cIAPs is TNFR1 signaling. In the classical pathway, the best characterized role of In addition to regulating TNFa-mediated NF-kB cIAP1 and cIAP2 is as an essential positive regulator of signaling, recent data have also shown that cIAP1 and TNFa-mediated activation (Santoro et al., 2007; Ber- cIAP2 also regulate constitutive or ligand-independent trand et al., 2008; Mahoney et al., 2008). In the absence classical NF-kB signaling (Varfolomeev et al., 2007; of both cIAP1 and cIAP2, TNFa-mediated NF-kB Vince et al., 2007). Intriguingly, in contrast to their role signaling is dramatically attenuated in many cells, as positive regulators of TNFa-mediated signaling, they including normal primary cells as well as transformed act as negative regulators of ligand-independent signal- cancer cells. Consequently, the dual loss of cIAP1 and ing. Using small molecule cIAP antagonists, both Vince cIAP2 greatly sensitizes cells to TNFa-mediated apop- et al. (2007) as well as Varfolomeev et al. (2007) have tosis. Mechanistically, TNFa signals through its recep- shown that dual loss of cIAP1 and cIAP2 leads to tor TNFR1. On TNFa occupancy, TNFR1 rapidly elevated constitutive classical NF-kB signaling in a recruits the TNFR-associated death domain (TRADD) variety of cancer cells. Vince et al. (2007) also reported protein as well as the receptor-interacting protein (RIP) elevated ligand-independent classical NF-kB signaling 1. TRADD binding in turn recruits TRAF2 and cIAP1/ in MEFs derived from cIAP1À/À mice, which suggested a cIAP2 to form a large membrane complex. When this non-redundant role for cIAP1 in this capacity. However, occurs, RIP1 gets modified with long K63-linked these MEFs were SV-40 transformed, and we have polyubiquitin chains, which serves as a docking site observed that the expression of cIAP2 is silenced therein for the downstream kinases that propagate the NF-kB (unpublished observation) making them a functional signal (Ea et al., 2006; Wu et al., 2006) (Figure 2). double knockout cell. Mechanistically, preliminary data Originally, the E3 ligase responsible was thought to be suggested that loss of cIAP1 and cIAP2 induced the TRAF2. TRAF2 is a bona fide E3 ubiquitin ligase, and recruitment of RIP1 to TNFR1, although these data in its absence TNFa-mediated ubiquitination of RIP1 were somewhat difficult to interpret. Regardless, it was does not occur (Lee et al., 2004). However, TRAF2 has proposed that cIAP1 and cIAP2 regulate the degree of never been shown to directly target RIP1 for ubiquitina- constitutive NF-kB activation by controlling RIP1 tion in vitro, which suggests that the loss of TRAF2 migration to TNFR1. Recent data demonstrate that indirectly affects RIP1 ubiquitination. Recently, we cIAP1 and cIAP2 constitutively interact with RIP1, lend

Oncogene IAP-targeted therapeutics EC LaCasse et al 6257

TNFR family receptors

TRAF2/3 NIK RIPK1 cIAP1/2

P P

NEMO NEMO IKKα IKKα IKK complex IKKα IKKβ

P P P proteasomal p100 RelB P degradation p100 p50 RelA processing

p52 RelB p50 RelA

nucleus nucleus p52 p50 RelA RelB target genes target genes

Figure 2 cIAP1 and cIAP2 regulation of classical and alternative nuclear factor-kB (NF-kB) activation pathways. The two major activation pathways for NF-kB are shown. Those tumor necrosis factor (TNF) superfamily receptors that utilize TNF receptor- associated factor 2 (TRAF2) as an adaptor can recruit cIAP1 or cIAP2, which can then positively modulate the classical pathway, or negatively regulate the alternative pathway. In the classical NF-kB pathway, NF-kB dimers (for example, p50/RelA) are retained in the cytoplasm by interaction with IkB. In response to the binding of a ligand to a TNF receptor (TNFR)-related protein, TRAF proteins are recruited to the receptor, and through the ubiquitination of receptor-interacting protein (RIP) kinase (RIP1/RIPK1) suggested to be mediated by the RING domain E3 ligase of cIAP1 and cIAP2, the IKK complex containing IKKa,IKKb and NEMO (IKKg)is recruited and activated. The activated IKK complex in turn phosphorylates the inhibitor of kB(IkB), leading to its degradation, thereby releasing NF-kB (p50/RelA) to translocate into the nucleus for transcriptional activation of targeted genes. By contrast, cIAP1 and cIAP2, TRAF2 and TRAF3 represses the alternative NF-kB pathway by promoting the ubiquitination and degradation of NF- kB-inducing kinase (NIK), which is required for phosphorylation of p100 in the p100/RelB complex. Phosphorylated p100 is processed into p52, and NF-kB (that is, the p52/RelB dimer) translocates into the nucleus for transcriptional activation of targeted genes. Hence, the E3 ubiquitin ligase function of cIAP1 or cIAP2, as well as their binding through BIR1 to TRAF2, plays a central role in the regulation of both NF-kB pathways.

support to this model (Bertrand et al., 2008; Wang et al., and TRAF3 proteins have RING zinc-ﬁngers with E3 2008). Clearly, more work is needed to fully substantiate ligase activity, and that loss of either TRAF2 or TRAF3 this mechanism, or to propose a new model of how results in NIKaccumulation and constitutive activation cIAP1 and cIAP2 regulated ligand-independent NF-kB of alternative NF-kB signaling, TRAF2 and/or TRAF3 signaling. were believed to target NIKfor ubiquitination and As mentioned above, cIAP1 and cIAP2 have also proteasomal degradation. Recent studies, however, have recently been shown to modulate the activity of the revealed that cIAP1 and cIAP2 are the true E3 ligases alternative NF-kB pathway (Varfolomeev et al., 2007; that constitutively target NIKfor K48-linkedpolyubi- Vince et al., 2007) (Figure 2). Normally inactive, quitination and proteasomal degradation (Varfolomeev alternative NF-kB signaling gets activated through the et al., 2007; Vince et al., 2007). Using small compounds ligation of various TNF receptors such as CD40-R, targeting cIAP1 and cIAP2, several groups have now lymphotoxin b-R and BAFF-R. On receptor ligation, shown that the loss of these IAP proteins leads to the the levels of NF-kB-inducing kinase (NIK) accumulate. accumulation of NIKand the activation of the NIKis normally co-translationally degraded in a alternative pathway. Again, cIAP1 and cIAP2 are complex containing TRAF2 and the related protein functionally redundant in this role, as we have observed TRAF3. Following sufﬁcient accumulation, NIKin that numerous cell lines rendered null for the individual turn phosphorylates downstream kinases that propagate cIAPs using siRNA do not have accumulated levels of the alternative NF-kB signal. Given that both TRAF2 NIK, whereas those rendered null for both cIAPs do

Oncogene IAP-targeted therapeutics EC LaCasse et al 6258 (unpublished observation). Moreover, both cIAP1 and regulator of the stress responsive kinases JNKand p38, cIAP2 were shown to directly target NIKfor K48- which are activated by TNFa but must be de-activated mediated ubiquitination in vitro. quickly to prevent apoptosis. Primary B cells Given the known roles for TRAF2 and TRAF3 in this from cIAP1-null mice revealed that cIAP1 targets pathway, a possible model for the control of NIK ASK1 for K48-linked ubiquitination and degradation protein levels is that the TRAF proteins serve as in response to TNFR2 ligation. As such, cIAP1 is adapters that bring the cIAP proteins into a complex responsible for regulating the duration of TNFa- with NIK. Indeed, using a variety of experimental induced JNKand p38 signaling in TNFR2-expressing approaches, we have seen that TRAF3 interacts with cells (Zhao et al., 2007). In the absence of cIAP1, JNK NIK, whereas TRAF2 interacts with the cIAPs, and and p38 signaling are prolonged and cells are sensitized that TRAF3 and TRAF2 interact with one another toward apoptosis. (unpublished observation). These data suggest that Significantly, XIAP is also known to influence signal- TRAF2 and TRAF3 bridge the gap between cIAP1/ transduction pathways, particularly the TAB1/TAK1 cIAP2 and NIK, bringing them into a large protein complex through interaction with its BIR1 domain. This complex (Figure 2). Colocalization by TRAF2 and TAB1/TAK1 complex impacts on NF-kB signaling TRAF3 allows cIAP1 and/or cIAP2 to constitutively pathways, as well as JNKpathways (Wang et al., target NIKfor K48-linked ubiquitination and protea- 2001; Shim et al., 2005). XIAP has been shown to inhibit somal degradation, thereby repressing alternative NF- c-JNK1 (JNK1) activation by transforming growth kB activity. factor b1 (TGF-b1) through ubiquitin-mediated protea- For the most part, NF-kB signaling strongly favors somal degradation of the TGF-b1-activated kinase 1 survival in both the classical and alternative pathways. (TAK1) (Kaur et al., 2005). XIAP’s antiapoptotic However, cIAP1 and cIAP2 positively regulate TNFa- activity has been proposed to be achieved by two mediated and yet negatively regulate ligand-independent separate mechanisms: one requiring TAK1-dependent classical NF-kB signaling (Santoro et al., 2007; Bertrand JNK1 activation and the second involving caspase et al., 2008; Mahoney et al., 2008), and the alternative inhibition (Sanna et al., 1998, 2002a, b). pathway (Varfolomeev et al., 2007; Vince et al., 2007). Hence, the intriguing possibility exists that cIAP1 and cIAP2 may have both anti- and pro-survival roles depending on the cell type and context. Indeed, TNFa Negative regulation of the IAPs strongly induces apoptosis in the absence of both cIAP1 and cIAP2 in most cell lines tested to date. In contrast, Multiple proteins play critical roles in maintaining an primary B cells, which rely on alternative NF-kB adequate and fine-tuned balance between too much or signaling for survival, undergo less spontaneous apop- too little apoptosis. Intrinsic antagonists of the IAPs tosis in the absence of cIAP1 and cIAP2 (unpublished exist, which help maintain and regulate this apoptotic observation). balance. To complicate matters further, cIAP1 has been shown to facilitate TNFa-mediated apoptosis when TNFR1 Smac/DIABLO, an IAP antagonist and a caspase and TNFR2 are simultaneously engaged (Li et al., 2002; activator Wu et al., 2005a). Unlike TNFR1, which is ubiquitously Smac/DIABLO (second mitochondria-derived activator and constitutively expressed, TNFR2 expression is of caspases, direct IAP-binding protein with low PI) is a largely restricted to lymphoid cells. Although the precise 25-kDa mitochondrial protein that promotes apoptosis biological role of TNFR2 remains unclear, co-stimula- through its ability to antagonize IAP-mediated caspase tion of TNFR2 sensitizes cells to TNFa-mediated inhibition once released into the cytoplasm. The ability apoptosis through TNFR1. Elegant work in Jonathan of Smac to prevent the IAP inhibition of caspases makes Ashwell’s laboratory has demonstrated two mechanisms this protein a functional mammalian equivalent to responsible for such ‘cross-talk’, both involving cIAP1- the Drosophila death proteins, Reaper, Grim and Hid mediated regulation of signal transduction. The first (Du et al., 2000; Verhagen et al., 2000). mechanism is that cIAP1 negatively regulates TNFa- Initial mitochondrial targeting of Smac or DIABLO mediated NF-kB signaling when TNFR2 is engaged (Li depends on a 53–55 amino-acid leader signal sequence et al., 2002). How this occurs is somewhat controversial, found at its amino terminus. This sequence is proteo- but it is known that TRAF2 recruits cIAP1 to TNFR2 lytically cleaved within the mitochondria to generate an on exposure to TNFa. Following such recruitment, AVPI-containing amino terminus polypeptide (Burri cIAP1 can target TRAF2 for K48-linked ubiquitination et al., 2005), sequestered until an apoptogenic stimulus is and proteasomal degradation, which likely prevents its sensed. On receiving an apoptotic signal through the own recruitment to TNFR1. As such, the pro-survival intrinsic mitochondrial stress pathway, both Smac and NF-kB signaling pathway, which requires cIAP1 or cytochrome c are released with similar, but not cIAP2 for activation, is not engaged and apoptosis necessarily identical, kinetics. Although the exact ensues. The second mechanism involves the ability of mechanism of Smac release from the mitochondria is cIAP1 to target apoptosis signal-regulating kinase 1 not known, one study suggests that, despite their (ASK1) for K48-linked ubiquitination and proteasomal significant size difference, the Smac dimer (100 kDa) degradation (Zhao et al., 2007). ASK1 is a key negative and cytochrome c (12 kDa) are released through the

Oncogene IAP-targeted therapeutics EC LaCasse et al 6259 same Bax- or Bak-formed membrane pores. This measure to ensure that the apoptotic program can mechanism is supported by the fact that Bcl-2 over- proceed. expression can inhibit both cytochrome c and Smac Smac/DIABLO appears to function as a general IAP release (Adrain et al., 2001), whereas Bid-induced inhibitor in that it is shown to bind to XIAP, cIAP1, permeabilization of the outer mitochondrial membrane cIAP2, survivin, livin and BRUCE, but not NAIP (Du induces the rapid and complete release of cytochrome c et al., 2000; Verhagen et al., 2000; Vucic et al., 2002; and Smac from the intermembrane space (Madesh et al., Davoodi et al., 2004; Hao et al., 2004; Qiu and 2002). Goldberg, 2005). The first four amino-terminal residues Other studies, however, suggest that Smac and of mature Smac/DIABLO, Ala-Val-Pro-Ile, are required cytochrome c release is mediated through different for Smac/DIABLO function, and deletion of this mechanisms. Although TRAIL (TNF-related apopto- sequence abolishes Smac/DIABLO–IAP interaction sis-inducing ligand) induces cytochrome c release and (Chai et al., 2000; Liu et al., 2000; Wu et al., 2000). apoptosis in wild-type, Bid-null, Bax-null and Bak-null The crystal structure of Smac/DIABLO indicates that it MEFs, in contrast, Bax/Bak double knockout MEFs are consists of three extended a-helices bundled together to resistant (Kandasamy et al., 2003). Moreover, TRAIL- form an arc-shaped structure, exposing an unstructured induced mitochondrial Smac release is blocked in all of amino terminus. Smac/DIABLO homodimerizes the single knockout and the double knockout MEFs. through an extensive hydrophobic interface, which is Therefore, it is concluded from these experiments that essential for its activity (Chai et al., 2000), forming an the release of Smac and cytochrome c from mitochon- extremely stable protein dimer (Goncalves et al., 2008). dria is, in fact, differentially regulated in receptor- The newly generated amino terminus in mature Smac/ mediated pathways of apoptosis (Kandasamy et al., DIABLO makes contacts with XIAP BIR3 and 2003). Differences in the mechanism of Smac and mediates XIAP inhibition. The co-crystal structure of cytochrome c release occur in the presence of caspase XIAP BIR3 and Smac/DIABLO indicates that the inhibitors. Under these conditions, Smac release is amino-terminal four residues AVPI in Smac/DIABLO prevented, whereas cytochrome c is permitted, suggest- recognize a surface groove on BIR3, with the alanyl ing that Smac efflux from the mitochondria is a caspase- residue bound within a hydrophobic pocket (Wu et al., catalysed event (Adrain et al., 2001). It may be that 2000). The amino-terminal four residues of the caspase- Smac release simply requires further membrane perme- 9 linker peptide (Ala316-Thr-Pro-Phe) share significant ability due to increased or persistent mitochondrial homology to the amino-terminal tetrapeptide in damage. mature Smac/DIABLO and Drosophila death proteins. Several Smac isoforms exist and studies using these Initially, it was believed that binding of the caspase-9 proteins suggest that the pro-apoptotic function of Smac linker peptide and Smac/DIABLO to the BIR3 domain may be mediated by additional, non-IAP mechanisms. of XIAP is mutually exclusive, suggesting a competition Smacb, an alternatively spliced cytoplasmic isoform of model in which Smac/DIABLO displaces XIAP Smac, lacks the mitochondrial targeting sequence found from caspase-9 (Srinivasula et al., 2001). Smac/DIA- in full-length Smac. In vitro experiments show that BLO is also predicted to bind to XIAP BIR2 and Smacb interacts with purified XIAP protein; however, in disrupt caspase-3 and -7 inhibition, possibly by steric intact cells, XIAP binding is not observed (Roberts hindrance (Chai et al., 2000). More recent experiments et al., 2001). Smacb is still considered to be pro- suggest that Smac/DIABLO is unable to remove apoptotic due to its ability to potentiate apoptosis caspase-3 and -7 inhibition by the linker-BIR2 domains following death receptor and chemical stimuli (Roberts of XIAP and inefficiently relieves caspase-9 inhibition et al., 2001). Unlike Smacb, the Smac3 isoform contains by BIR3. However, when constructs expressing the both an amino-terminal mitochondrial targeting se- XIAP-BIR2 and -BIR3 domains in tandem, Smac/ quence and an IBM sequence. Following an apoptotic DIABLO effectively prevents IAP inhibition of stress, Smac3, similar to Smac, is released from the both initiator and effector caspases. Furthermore, the mitochondria into the cytoplasm where it interacts with affinities of the BIR2 and BIR3 domains of XIAP the BIR2 and BIR3 domains of XIAP. There are some for Smac/DIABLO were shown to be almost identical reports that Smac3 is unique in that it is able to induce with each Smac dimer interacting with the BIR2 and the acceleration of XIAP auto-ubiquitination and BIR3 domains of one XIAP molecule to form a 2:1 destruction, whereas Smac only seems to have this stoichiometric complex (Huang et al., 2003). Although effect on cIAP1 and cIAP2 (Fu et al., 2003; Yang and individual BIR domains of XIAP are adequate for Du, 2004). However, another study describes the ability inhibiting in vitro caspase activity, these findings suggest of Smac to antagonize both XIAP auto-ubiquitination that both BIR2 and BIR3 are required not only for and XIAP-dependent ubiquitination of caspase-7 XIAP-Smac/DIABLO interaction but also for subse- (Creagh et al., 2004). By disrupting IAP–caspase quent liberation of caspase inhibition (Huang et al., interactions and repressing the ubiquitin ligase activities 2003). of the IAPs, Smac may effectively prevent caspase The structural analysis of Smac binding to XIAP demise through ubiquitination (Creagh et al., 2004). indicates that the amino-terminal tetrapeptide recog- Although Smac and its isoforms may differ in how they nizes a surface groove on the BIR3 domain, implying potentiate apoptosis, the involvement of all Smac that peptides or small molecules modeled on this proteins with the IAPs, at multiple levels, is an effective binding motif might serve as prototypical drugs the

Oncogene IAP-targeted therapeutics EC LaCasse et al 6260 activity of which might complement that of Smac (Kipp and BIR3 regions to liberate caspase-3, -7 and -9 et al., 2002). When Smac peptides composed of the first (Li et al., 2004; Bockbrader et al., 2005). Alternatively, 4–8 amino-terminal residues are delivered into MCF-7 as recently reported for several different SMCs, their breast cancer cells, they are capable of interacting with mechanism of action may be due to the induction of XIAP, cIAP1 (Arnt et al., 2002; Fulda et al., 2002) and auto- or trans-ubiquitination of the IAPs, primarily the single BIR domain of livin (Vucic et al., 2002). These cIAP1 and cIAP2, which results in their rapid loss from peptides enhance apoptosis and long-term antiprolifera- the cell within minutes of compound addition and tive effects of a range of chemotherapeutics, including sensitization to TNFa-induced killing (Varfolomeev paclitaxel, etoposide and doxorubicin (Arnt et al., 2002; et al., 2007; Vince et al., 2007). Fulda et al., 2002). Smac-like peptides delivered in combination with Other IBM-containing proteins as potential IAP cancer therapeutics such as Apo2 L/TRAIL appears to antagonists be a promising method to reduce tumor burden Following the identification of Smac, another mitochon- (Arnt et al., 2002; Fulda et al., 2002). The in vivo drial IAP-binding protein called Omi, or HtrA2, was delivery of Smac peptide, in conjunction with Apo2 L/ discovered (Hegde et al., 2002; Martins et al., 2002; TRAIL, completely eradicates established intracranial Verhagen et al., 2002). Omi/HtrA2 is a mammalian malignant glioma xenograft tumors and increases homolog of the Escherichia coli bacterium heat-induci- the survival time of treated mice (Fulda et al., 2002). ble serine protease, known as HtrA. Another identified Apo2 L/TRAIL induces apoptosis through the extrin- IBM-containing protein is an isoform of the polypeptide sic/death receptor pathway by binding to the transmem- chain-releasing factor GSPT1/eRF3 protein that is brane receptors TRAIL-R1/DR4 and TRAIL-R2/DR5. localized to the endoplasmic reticulum (Hegde et al., The delivery of Smac peptides in the presence of 2003). Checkpoint kinase 1 (Chk1), a dual function Apo2 L/TRAIL appears to have a significant impact kinase that is active during the S- to –M-phase transition on sensitizing otherwise resistant cells to apoptosis of the cell cycle, regulating Cdc25A function, and (Fulda et al., 2002; Guo et al., 2002; Ng and Bonavida, prevents mitotic progression in the presence of DNA 2002). damage, contains an IBM sequence (Galvan et al., Some studies have shown that amino-terminal Smac 2004). More IBM-containing proteins, all mitochondria- tetrapeptides bind only to the BIR3 domain of XIAP derived, exist (Verhagen et al., 2007), and potentially with low affinity, are sensitive to proteolytic degradation more are yet to be found. If we extrapolate from the and have a poor capacity to penetrate cells (Li et al., situation in Drosophila, with four known IAPs (DIAP1/ 2004). Efforts to circumvent the limitations of Smac thread, DIAP2, deterin and dBruce) versus eight human peptides led to the development of small molecule Smac IAPs, and five known Drosophila IBM-containing mimetics, compounds that inhibit the IAPs with higher antagonists (reaper, hid, grim, sickle and jafrac), then affinities than Smac peptides (Li et al., 2004). Such we would predict that a number of more mammalian submicromolar, small molecule, non-peptidic antago- IBM-containing proteins remain to be discovered. It is nists are reported by Park et al. (2005). These suggested that of all the mitochondrial apoptogenic compounds are created by individually replacing several proteins, Smac (an IBM-containing protein) and cyto- amino-acid residues while still maintaining the essential chrome c remain as the only likely mammalian ‘killer’ interactions necessary for binding in the Smac-binding proteins (Ekert and Vaux, 2005). site on XIAP BIR3 (Park et al., 2005). Another identified class of compounds is a series of proteolytically stable, capped tripeptides consisting of unnatural XAF1 is an interferon-inducible IAP antagonist, and a amino acids that bind to XIAP BIR3 with high candidate tumor suppressor nanomolar affinities. These compounds are cytotoxic Using a yeast two-hybrid screen, our laboratory in cancer cells and delay tumor growth in breast cancer identified XAF1 (XIAP-associated factor 1) as a xenograft models (Oost et al., 2004). XIAP-binding partner (Liston et al., 2001). The ability An additional compound class, a modified oxazoline of XAF1 to bind to XIAP is confirmed by an in vitro molecule, has been shown after several modifications to expression system (Liston et al., 2001), as well as by bind to the IAPs with a higher affinity than Smac immunoprecipitation of endogenous proteins (Arora tetrapeptides, and block IAP interaction with caspase-9. et al., 2007). XAF1 directly associates with XIAP to This compound also acts synergistically with Apo2 L/ antagonize XIAP-mediated caspase-3 inhibition, there- TRAIL and various chemotherapeutic agents (Li et al., by reversing XIAP-mediated protection against che- 2004). Further studies using a high IAP-expressing motherapeutic drugs such as cisplatin and etoposide. breast cancer cell line MDA-MB-231 show that at low Although the exact molecular interaction between nanomolar concentrations, the Smac mimetic com- XAF1 and XIAP remains unknown, it appears to be pound significantly sensitizes these cells to both Apo2 multipartite based on domain analysis (unpublished L/TRAIL- and etoposide-induced apoptosis through observations). In contrast to the original report, two caspase-3 activation (Bockbrader et al., 2005). The groups failed to detect any interaction between XAF1 effectiveness of this dimeric compound is due to its and XIAP under the conditions tested (Verhagen et al., ability to mimic the dimeric structure of Smac protein, 2000; Xia et al., 2006). Unexpectedly, one of these which, as mentioned previously, acts at the IAP BIR2 studies demonstrated that XAF1 exerted pro-apoptotic

Oncogene IAP-targeted therapeutics EC LaCasse et al 6261 effects in cells devoid of XIAP (Xia et al., 2006). DNA hypermethylation at the CpG sties (reviewed in In response to TNFa, the induction of XAF1 can Plenchette et al., 2007), and also by HSF1-mediated dramatically increase the level of apoptosis regardless transcriptional repression (Wang et al., 2006a). XAF1 of the status of XIAP. This is in contrast to the view expression can be restored experimentally on inhibition that XAF1 antagonism of XIAP is its quintessential of DNA methylation, as well as with the addition of pro-apoptotic function. However, these ostensibly IFN as discussed above (Reu et al., 2006; Zou et al., contradictory findings may be reconciled by the fact 2006; Arora et al., 2007; Micali et al., 2007). Expectedly, that XAF1 also binds, and perhaps antagonizes, because DNA methyltransferases (DNMTs) are in- cIAP1 and cIAP2 (and other IAPs as well), with at volved in the epigenetic silencing of various genes (Rhee least equal, if not greater affinity than XIAP et al., 2002), the inhibition of DNMT1 by either (Arora et al., 2007). Given that cIAP1 and cIAP2 oligonucleotide antisense or methylation inhibitor are critical negative regulators of TNFa-mediated cell 50aza-dC can lead to XAF1 induction (Reu et al., death (Mahoney et al., 2008), it should not be surprising 2006). Importantly, this enhanced induction of XAF1 is at all that the potential antagonism of these proteins by a crucial factor for apoptosis mediated by IFN in XAF1 can impact on cellular survival, in the absence melanoma cells (Reu et al., 2006). Interestingly, of XIAP. although methylation appears to play a role in the Although XAF1 appears to have an affinity toward regulation of XAF1 expression, IFN reactivation of many IAPs, XAF1 does not bind to survivin directly. XAF1 in certain tumor cells can occur in spite of a Interestingly, despite the lack of direct interaction, persistently hypermethylated promoter (Micali et al., XAF1 can destabilize survivin through their mutual 2007), suggesting that additional regulatory mechanisms intermediate, XIAP (Arora et al., 2007). Dohi et al. are involved. (2004) have shown that the binding of survivin to XIAP For tumor cells, XAF1 can be a pro-apoptotic increases the stability of latter, and reinforces the stimulus either by itself, or in combination with antiapoptotic activity of this IAP-IAP complex. We additional triggers (Liston et al., 2001; Leaman et al., have, in turn, demonstrated that the expression of 2002; Qi et al., 2006). In xenograft nude mouse models XAF1 can activate the RING E3 ligase of XIAP, in which gastric and colon cancers are established, subjecting survivin to the ubiquitin–proteasomal path- XAF1 overexpression alone suppresses tumor forma- way for degradation (Arora et al., 2007). In effect, tion. Further studies show that adenovirus XAF1’s tumor suppressor role may be due to its encoding xaf1 delivered by intratumoral injections in antagonism of the potential tumorigenic function of combination with Apo2 L/TRAIL or chemotherapeu- the XIAP-survivin complex. tics enhances apoptosis leads to the eventual complete XAF1 is ubiquitously expressed in normal tissues, but eradication of established tumors (Qi et al., 2006). is absent or detected at only extremely low levels in the Significantly, apoptosis is not induced in normal cells majority of the NCI 60 cell line panel of cancer cells surrounding the tumor (Qi et al., 2006), suggesting that (Fong et al., 2000), human colorectal cancer cell lines tumor cells have a lower tolerance for XAF1 than and primary carcinomas (Zou et al., 2006; Chung et al., normal tissues. XAF1 also shows selective killing 2007). The loss of heterozygosity of xaf1 is a frequent potential in other human tumor cells, including pan- occurrence in tumors (Fong et al., 2000; Liston et al., creatic and breast cancer cell lines. When these human 2001; Chung et al., 2007). In addition, a recent study tumor cell lines and normal cell lines are transduced shows that XAF1 is a predictive and prognostic factor with adenoviral xaf1, approximately 80% of the tumor in bladder cancer patients (Pinho et al., 2008). The cells underwent apoptosis compared with 5–10% in the downregulation or loss of xaf1 expression in cancer cells normal cell lines (Yang et al., 2003). The ability of may indeed contribute to apoptosis suppression and XAF1 to selectively induce apoptosis in tumor cells tumorigenesis through unrestricted IAP activity. Accu- while leaving healthy cells unaffected makes this mulating evidence indicates that restoring XAF1 ex- candidate tumor suppressor potentially effective for pression in cancer cells increases their sensitivity to cancer therapy. apoptotic triggers (reviewed in Plenchette et al., 2007). A strategy for inducing XAF1 is by exposing tumor cells to interferon-b (IFN-b) (Leaman et al., 2002; Wang et al., 2006b; Arora et al., 2007; Micali et al., 2007). Melanoma The IAPs in cancer and treatment cell lines that are responsive to IFN-b-mediated XAF1 upregulation are predisposed to Apo2 L/TRAIL-in- Convincing data for the involvement of the IAPs in duced apoptosis in a manner dependent on the XAF1 cancer and other human diseases now exist. Direct induction (Leaman et al., 2002). These studies suggest genetic evidence, as demonstrated by IAP gene ampli- that the re-expression of XAF1 strongly influences fications, gene mutations or deletions, chromosomal cellular sensitivity to cell death initiated by the extrinsic translocations, indicates causality in cancer and pro- pathway. liferative autoimmune disorders (Table 2). These find- In the past few years, we have made some appreciable ings are supported by animal studies and cell culture progress in determining the mechanism controlling experiments, demonstrating the immunomodulatory XAF1 expression levels. In both cell lines and tumor and oncogenic potential of the IAPs, as well as tissues, the XAF1 promoter is primarily silenced by identifying their normal physiological roles.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6262 Table 2 Summary of IAP mutations and deletions in mammals leading to proliferative disorders or cancer Gene Disease Comments

Birc2 (cIAP1) Multiple myeloma (MM) cIAP1 and cIAP2 are deleted in some MM cancers, as part of a larger trend of activating mutations for the alternative NF-kB pathway (Annunziata et al., 2007; Keats et al., 2007). Hence, the negative regulatory role of cIAP1 or cIAP2 as E3 ubiquitin ligases for NIKis removed, allowing for NIKstabilization and activation of the alternative NF- kB pathway. Various carcinomas cIAP1 and cIAP2 genes, at 11q22, are commonly amplified in many cancers, with visibly increased expression of cIAP1. This phenomenon is also conserved in some murine tumors that also lead to co-amplification of YAP. Both cIAP1 and YAP have been shown to cooperate in transformation (Zender et al., 2006). cIAP1 can also cooperate with Myc to transform. This results from cIAP1’s ability to induce the proteosomal degradation of the Myc antagonist Mad1 (Xu et al., 2007). Birc3 (cIAP2) MALT lymphoma cIAP2 is translocated in t(11;18)(q21;q21)-bearing lymphomas (Dierlamm et al., 1999). The chromosomal translocation fuses the three BIR domains of cIAP2 with the paracaspase domain of MALT1, while preserving the cIAP2, NF-kB-responsive, promoter. Multiple myeloma (MM) cIAP1 and cIAP2 genes are deleted in some MM cancers, as part of a larger trend of activating mutations for the alternative NF-kB pathway (Annunziata et al., 2007; Keats et al., 2007). Hence, the negative regulatory role of cIAP1 or cIAP2 as E3 ubiquitin ligases for NIKis removed, allowing for NIKstabilization and activation of the alternative NF- kB pathway. Various carcinomas cIAP1 and cIAP2 genes, at 11q22, are commonly amplified in many cancers, with visibly increased expression of cIAP1. While cIAP2 expression is often overlooked. This DNA amplification is also conserved in some murine tumors, which also leads to co-amplification of YAP (Zender et al., 2006). cIAP2 may also cooperate in transformation in these cases but this remains to be conclusively proven. Birc4 (XIAP) X-linked lymphoproliferative Apoptosis of lymphocytes from XIAP-deficient patients is enhanced in response to various disorder (XLP) stimuli including CD3, Fas and TRAIL. In addition, XIAP-deficient XLP patients (XLP-2), like SAP-deficient XLP patients (XLP-1), have low numbers of NKT cells. XLP patients can exhibit a fatal response to EBV, the viral cause of infectious mononucleosis (presumably due to a failure(s) in NKand lymphocyte homeostasis and signaling) (Rigaud et al., 2006).

Abbreviations: BIR, baculovirus inhibitor of apoptosis repeat; EBV, Epstein–Barr virus; MALT, mucosa-associated lymphoid tissue; NF-kB, nuclear factor-kB; NIK, NF-kB-inducing kinase; NK, natural killer; SAP, SLAM-associated protein; TRAIL, TNF-related apoptosis-inducing ligand; YAP, Yes-associated protein.

Pathological roles of the IAPs in cancer ical survey of tumor tissue shows that XIAP immuno- A variety of cancer cell lines and primary tumor biopsy staining patterns allow for the ready distinction of samples show elevated IAP expression levels (Ambrosini malignant from benign populations in effusion washes et al., 1997; LaCasse et al., 1998; Fong et al., 2000; (Wu et al., 2005b). In this survey, strong XIAP staining Tamm et al., 2000; Vucic et al., 2000; Li et al., 2001). was most prevalent in ovarian carcinoma effusions and The most dramatic example of IAP overexpression in less prevalent in mammary carcinoma effusions. XIAP tumors is seen with survivin. Survivin expression is expression increases with advancing stage in ovarian limited to embryonic tissues and many different tumor carcinoma (Mao et al., 2007). XIAP is also identified as types, but is absent in most adult differentiated tissues part of a progression signature in prostate cancer using (Ambrosini et al., 1997). Furthermore, the presence of an immunoblot approach to characterize proteomic survivin in patient tumor biopsy samples correlates with alterations in prostate tumors. XIAP protein is seen to poor prognosis, increased rates of treatment failure and increase in expression between benign prostatic tissue, relapse (Mesri et al., 2001). The prognostic significance and clinically localized prostate cancer, or metastatic of IAP overexpression is less clear for some of the other prostate cancer (Varambally et al., 2005). In addition, IAPs. For example, XIAP protein levels correlate with XIAP protein expression is a strong predictor of disease severity and prognosis in acute myelogenous prostate cancer recurrence (Seligson et al., 2007). leukemia (AML) (Tamm et al., 2000) and renal cell Collectively, these results suggest that the expression carcinoma (Ramp et al., 2004; Yan et al., 2004; levels of the IAPs could be expected to have a significant Mizutani et al., 2007), but not in non-small cell lung impact in the development and maintenance of cancer carcinoma (Ferreira et al., 2001a, b) or cervical carcino- due to their central role in the regulation of apoptosis. ma (Liu et al., 2001). XIAP levels have also been shown There is direct genetic evidence for the IAPs as to increase in leukocytes with the transformation of proto-oncogenes (Table 2). This is seen in cases with myelodysplastic syndromes to overt AML (Yamamoto chromosomal amplification of the 11q21–q23 region, et al., 2004). High expression of XIAP or cIAP2 is encompassing both cIAP1 and cIAP2, which is observed associated with shorter overall survival, and lower in a variety of malignancies, including medulloblasto- complete response rates for AML (Hess et al., 2007). mas, renal cell carcinomas, glioblastomas, gastric Other cross-validation and forward selection studies carcinomas and non-small cell lung carcinomas. Eso- reveal a three-gene signature, consisting of cIAP2, Bax(l) phageal squamous cell carcinomas frequently display and BMF, to optimally predict overall survival in AML this amplification, and transcriptional profiling has (Hess et al., 2007). Furthermore, an immunocytochem- identified cIAP1 as the sole target that is consistently

Oncogene IAP-targeted therapeutics EC LaCasse et al 6263 overexpressed in these tumors (Imoto et al., 2001). More 93% recent findings show amplification of 11q22 in primary 2% 1% 4% tumors of non-small cell lung and small cell lung cancers. Both cIAP1 and cIAP2 are overexpressed in cIAP2 BIR1 BIR2 BIR3 CARD RING these primary carcinomas (Dai et al., 2003). Comparison TRAF2 binding of human and mouse tumors reveal that cIAP1 and Fusion oligomerization cIAP2 are also amplified in the corresponding 11q22 bcl-10 binding synteneic region in mouse at 9qA1 (Zender et al., 2006). RipK1 binding caspase binding, Smac binding More importantly, when cIAP1 and another gene from Bcl-10, RipK1, NIK and cIAP2 ubiquitination 9qA1, YAP, are introduced into murine ES cells and injected into mice, they form hepatomas (Zender et al., 42% 32% 2006). In addition, cIAP2 could also transform p53-null 9% 17% hepatoblasts in combination with c-Myc overexpression. MALT1 DD Ig Ig paracaspase Ig-like These examples provide concrete proof of the role of cIAP1 as an oncogene (Zender et al., 2006). Further- bcl-10 binding more, cIAP1 cooperates with the Myc oncogene in A20 and bcl-10 cleavage transformation, by acting as a ubiquitin ligase for Nuclear export CARMA1 binding (bcl-10 independent) et al Mad1, an antagonist of Myc (Xu ., 2007). A recent TRAF6 binding report also suggests that cIAP1 and cIAP2 promote cancer cell survival by ubiquitinating RIP1, leading to constitutive RIP1 and NF-kB activity (Bertrand et al., cIAP2-MALT1 fusion examples 2008). BIR1 BIR2 BIR3 Ig Ig paracaspase Ig-like Further genetic evidence implicating the IAPs as BIR1 BIR2 BIR3 paracaspase Ig-like potential oncogenes is found in the generation of extranodal marginal zone mucosa-associated lymphoid Figure 3 cIAP2 is translocated and involved in mucosa-associated tissue (MALT) B-cell lymphomas. The translocation lymphoid tissue (MALT) lymphomagenesis. The t(11;18)(q21;q21) MALT lymphoma translocation disrupts the cIAP2 gene, also events t(1;14)(p22;q32) and t(11;18)(q21;q21) are well called api2, on chromosome 11q22 and the MALT1 gene, mlt,on documented in MALT lymphomas and involve both chromosome 18q21. The resulting novel cIAP2-MALT1 fusion NF-kB activation and cIAP2 expression. The less proteins possess the cIAP2 baculovirus IAP repeat (BIR) domains frequently observed translocation t(1;14)(p22;q32) re- (prior to the indicated triangles) linked to the carboxy terminus of sults in Bcl-10 overexpression (Vega and Medeiros, MALT1 at one of four different points (indicated by triangles on the protein diagram). This schematic diagram represents two such 2001). A common feature of these MALT lymphomas is resulting fusion proteins; more variants exist based on the location the increased nuclear expression of Bcl-10. Under of the chromosomal breakpoints. The breakpoint frequencies (in normal circumstances, cIAP2 exerts immunoregulatory percentage) are given for the two affected genes based on a review effects on anigen receptor signaling by ubiquitinating by Du (2007). The many functions and binding partners attributed to the various domains of both parental proteins are shown to help and targeting Bcl-10 for degradation (Hu et al., understand the properties of the resulting fusions. MALT1 2006a, b). Bcl-10 forms a ‘signalosome’ complex with possesses a death domain (DD) of unknown function, as well as CARMA1 and MALT1 that triggers NF-kB activation immunoglobulin (Ig) or Ig-like domains. MALT1 also possesses a in response to T-cell and B-cell receptor signaling caspase-like domain, referred to as a paracaspase, with protease (reviewed in Schulze-Luehrmann and Ghosh, 2006; activity. These fusion proteins are both antiapoptotic and nuclear factor-kB (NF-kB) activating either through TRAF6 or TRAF2 Wegener and Krappmann, 2007). Thus, it appears that binding, Bcl-10 stabilization and nuclear localization, and/or A20 cIAP family members function as ubiquitin protein cleavage. A positive feedback amplification loop is established in ligases and are capable of regulating not only one MALT lymphoma in which the fused gene is expressed under the another, but components of the TNFR and signalosome control of the ciap2 promoter, which contains several NF-kB- signaling complexes as well. responsive sites (not shown). A more common translocation event, present in approximately 50% of extranodal MALT lymphomas, MALT translocation establishes a positive feedback involves a gene rearrangement of the MALT1 locus loop for cIAP2 by activating NF-kB. This in turn, 18q21 with the cIAP2 gene (also known as API2, transcriptionally activates the promoter for cIAP2 (Chu located at 11q22) resulting in the cIAP2-MALT fusion et al., 1997; Erl et al., 1999; Hong et al., 2000). In this product t(11;18)(q21;q21) (Dierlamm et al., 1999; Baens way, the cIAP2-MALT1 fusion induces NF-kB activa- et al., 2000; Remstein et al., 2000; Hosokawa, 2005). The tion and contributes to an antiapoptotic phenomenon encoded chimeric protein consists of the amino-terminal by upregulating transcription of antiapoptotic genes cIAP2 BIR1–3 domains (excluding the RING domain) such as cIAP2, cFlip and MnSOD. As a result, the fused in-frame to the carboxy terminus of MALT1 lymphoma cells are no longer dependent on the (Uren et al., 2000) (Figure 3). Typically, gastric MALT inflammatory mechanism of NF-kB activation and lymphomas are associated with chronic inflammation eliminating the H. pylori is no longer sufficient to shut due to Helicobacter pylori infection, and can be treated down inappropriate cell proliferation. with antibiotic therapy. Significantly, the majority of The exact mechanism of NF-kB activation by the these lymphomas that do not respond to antibiotics cIAP2-MALT1 fusion is debated, and the subject of exhibit the cIAP2-MALT1 translocation. The cIAP2- much interest and investigation. The cIAP2-MALT1

Oncogene IAP-targeted therapeutics EC LaCasse et al 6264 fusion targets NEMO, the regulatory subunit of IKK, neoplasm), as deletions of cIAP1/2 in these cancer cells, for polyubiquitination and activates NF-kB (Zhou or mutations in other genes regulating the alternative et al., 2005). The E3 ligase function responsible for this NF-kB pathway, are commonly found (Annunziata activity is unclear as the cIAP2-MALT1 fusion lacks the et al., 2007; Keats et al., 2007) (Table 2). The deletion of cIAP2 E3 function, and the direct role of two other cIAP1 and cIAP2 allows for NIKstabilization and candidates, Bcl-10 and TRAF6, was dismissed although activation of NF-kB through the alternative pathway, not ruled out completely. TRAF6 represents the most and presumably aids in tumor cell transformation and/ likely candidate as a fusion binder with E3 ligase activity or increased cell survival. Thus, based on the positive and NF-kB-activating properties. This is shown by role of cIAP1 and cIAP2 in classical NF-kB signaling, Noels et al. (2007), and explains why TRAF1 and and their negative role in alternative NF-kB pathways, TRAF2 are not necessary for this NF-kB activation the cIAPs can be considered to be either oncogenes or pathway (Varfolomeev et al., 2006). However, TRAF2 tumor suppressors, depending on the cell type and binding of BIR1 for fusion contributes also to NF-kB context of NF-kB activation. However, in all cases activation (Lucas et al., 2007; Noels et al., 2007). Bcl-10 involving deregulated cIAP1 or cIAP2 expression (such lacks E3 ligase activity and serves as an adaptor in the as carcinoma, MALT lymphoma and other non- CARMA1-Bcl-10-MALT1 signaling complex. It has Hodgkin’s lymphoma, and multiple myeloma), NF-kB been shown that cIAP2 acts as an ubiquitin ligase for activation is increased and this is associated with Bcl-10 (Hu et al., 2006b). This activity is lost in the carcinogenesis and increased tumor cell survival. cIAP2-MALT1 fusion, and concomitantly the Bcl-10 protein is stabilized in these MALT lymphomas. Furthermore, Bcl-10 and cIAP2-MALT1 synergistically activate NF-kB(Huet al., 2006b). In another twist to Targeting the IAPs for cancer therapy this story, it has been recently shown that the paracaspase domain of MALT1 (a caspase-like domain The acquired resistance of cancer cells to apoptosis is long thought to have protease activity but never proven one of the defining hallmarks of cancer (Hanahan and until now) can cleave the NF-kB inhibitor A20 Weinberg, 2000), necessitated in part to suppress cell (Coornaert et al., 2008), as well as Bcl-10 (Rebeaud death induced by oncogene activation (Evan and et al., 2008). The cIAP2-MALT fusion can also cleave Vousden, 2001). The issue of primary or acquired A20 providing another means of sustained NF-kB resistance to current chemotherapeutic-based treatments activity (Coornaert et al., 2008). Some of the discre- is a major impediment to effective cancer therapy. pancies in defining the mode of action of the cIAP2- Although there are many genetic and biochemical in vitro MALT1 fusions may be due to differences in the fusion alterations that occur in cancer cells, experiments constructs used in experiments based on the different demonstrate that the upregulation of IAP expression patient cases (see Du (2007) and Figure 3 for transloca- increases resistance to chemotherapeutic and radiation tion breakpoints and frequencies). These different resistance. The fundamental role of the IAPs in fusions incorporate different domains, such as the apoptosis regulation suggests that there is value in immunoglobulin domains, that may allow for differ- exploiting the inhibition of IAP expression and function ential binding of Bcl-10 or other factors. as a direct therapeutic strategy for cancer, and possibly The oncogenic potential of the cIAP2-MALT1 fusion other proliferative disorders as well. has been further revealed with the generation of an Em- API2-MALT1 transgenic mouse, which also exemplifies Antisense oligonucleotides to XIAP or survivin as the effect of NF-kB activation due to the cIAP2-MALT therapeutics fusion protein. In this study, results with the Em-API2- Applications of single-stranded antisense oligodeoxynu- MALT1 transgenic model show that the expression of cleotides (AS ODNs) as selective inhibitors of gene the cIAP2-MALT1 fusion protein is not adequate for expression are being studied for efficacy in treating the formation of lymphomas during the lifespan of the particular genetic disorders. In addition, many AS mouse (Baens et al., 2006), unless chronic antigenic ODNs are being assessed both preclinically and in early stimulation is provided (Sagaert et al., 2006). Expression clinical trials for several cancer types. The AS ODNs are of the MALT1 fusion did affect B-cell maturation in the short stretches of synthetic DNA, approximately 12–30 bone marrow and triggered the expansion of splenic nucleotides long and are complementary to a specific marginal zone B cells. The survival of B cells in this mRNA strand. Hybridization of the AS ODNs to the transgenic model, therefore, is thought to be promoted mRNA by Watson–Crick base pairing prevents the through enhanced NF-kB activation (Baens et al., 2006). target gene from being translated into protein, thereby It is noteworthy that the cIAP2-MALT1 fusion in the blocking the action of the gene, and resulting in the transgenic model is under control of the Em promoter degradation of the target mRNA (Galderisi et al., 1999; and not the NF-kB-responsive promoter of API2/ Gleave et al., 2002; Jansen and Zangemeister-Wittke, cIAP2. This may explain the need for continued antigen 2002). The specificity in the AS ODN approach is stimulation for lymphomagenesis to occur in this based on the fact that any sequence of approximately transgenic model. 13 bases in RNA and 17 bases in DNA is estimated to Surprisingly, cIAP1 and cIAP2 play a tumor sup- be represented only once in the human genome. A pressor role in multiple myeloma (a plasma B-cell multitude of candidate genes, involved in apoptosis

Oncogene IAP-targeted therapeutics EC LaCasse et al 6265 regulation, represents potential targets for antisense- tion identify this approach as a viable strategy to inhibit based therapies. tumor growth (Tu et al., 2003; Ansell et al., 2004; Cao Additional proof-of-principle studies performed with et al., 2004). Current phase II clinical trials of a survivin newer double-stranded RNAi-based technologies show AS ODN are underway by Eli Lilly and Company that downregulating XIAP gene expression through (Table 3). siRNA increases apoptosis in cultured MCF-7 breast In contrast to survivin, the importance of XIAP in cancer cells and subsequently enhances the killing effects cancer prognosis is less clear; however, XIAP is of etoposide and doxorubicin (Lima et al., 2004). arguably the most potent of the IAPs with respect to Another study using short-hairpin RNAs as an RNAi its ability to inhibit caspase activation and to suppress approach directed against XIAP shows that XIAP apoptosis. Survivin, on the other hand, is a weak mRNA can be reduced by as much as 85% in some apoptotic inhibitor and its death-suppressing effects breast carcinoma cell lines. This reduction in XIAP may be mediated by XIAP stabilization (Dohi et al., dramatically sensitizes these cell lines to TRAIL- and to 2004). Furthermore, XIAP is highly overexpressed in taxane-induced killing (McManus et al., 2004). many tumor cell lines of the NCI panel (LaCasse et al., Survivin is expressed in fetal tissues, becomes 1998) and its expression has been correlated with poor restricted during development, and is absent in most prognosis or advancing stage in AML and other healthy, differentiated adult tissues (Adida et al., 1998), cancers, as noted earlier. These factors indicate that with some notable exceptions, such as stem cells, XIAP is an attractive target for improving treatment thymus, testes, regenerating hepatocytes and endothelial responses for a variety of tumor types. cells (Kobayashi et al., 1999; reviewed in Fukuda and XIAP AS ODNs effectively downregulate both Pelus, 2006). Significantly, survivin is re-expressed specific mRNA and protein levels in human non-small during malignant transformation and is found in nearly cell lung cancer growth both in vitro and in vivo. XIAP all tumor types, including neuroblastomas (Islam et al., AS ODNs effectively induce apoptosis on their own and 2000), pancreatic, prostate, gastric, colorectal, hepato- sensitize the tumor cells to the cytotoxic effects of cellular and breast carcinomas, as well as lung and several chemotherapeutics, including Taxol, etoposide bladder cancers, melanomas, B-cell lymphomas and and doxorubicin (Hu et al., 2003). Furthermore, the esophageal cancer (Altieri, 2001). The expression of administration of XIAP AS ODNs in a xenograft model survivin in cancer is a predictor of both poor prognosis of human non-small cell lung cancer results in a and decreased survival time, and is implicated in significant downregulation of XIAP protein. Tumor conferring chemo- and radio-resistance phenotypes to establishment is delayed when AS ODNs are combined tumor cells. Numerous studies have addressed the with vinorelbine treatments (Hu et al., 2003). Other diagnostic and prognostic potential of survivin expres- studies show that XIAP AS ODNs enhance tumor sion, as well as that of its nuclear versus cytoplasmic regression in conjunction with radiotherapy in a mouse localization, phosphorylation status and the presence of model of lung cancer (Cao et al., 2004). When XIAP is splice isoforms (Li, 2005; Span et al., 2006; Yin et al., downregulated in the pancreatic carcinoma cell line, 2006; Dohi et al., 2007; Kleinberg et al., 2007; Pannone Panc-1, using a second generation, mixed backbone AS et al., 2007; Piras et al., 2007; Stauber et al., 2007; ODN (AEG35156/GEM640), the treated cells are Brennan et al., 2008; Bria et al., 2008). sensitive to TRAIL-induced killing (McManus et al., On account of the ubiquitous nature of survivin 2004). Similarly, TRAIL, in combination with another expression in human cancer, one therapeutic focus is the XIAP AS ODN (in phosphorodiamidate morpholino targeting of this IAP in the attempt to downregulate and chemistry), potentiates both cisplatin sensitivity and eliminate its expression. Interestingly, EPR-1 (effector TRAIL killing in androgen-refractory prostate cancer cell protease receptor-1) and survivin are encoded by cells (Amantana et al., 2004). Overall, these promising mRNAs transcribed from opposite strands of the same data indicate that when XIAP protein expression levels chromosome locus (Ambrosini et al., 1998). When epr-1 are reduced through RNAi or antisense approaches, a cDNA is overexpressed in HeLa cells, survivin is decreased apoptotic threshold to an array of chemother- downregulated, the rate of spontaneous apoptosis apeutics can be achieved (LaCasse et al., 2005). Notably, increases and cell proliferation is inhibited (Ambrosini a XIAP AS ODN (AEG35156) is being evaluated for et al., 1998). These experiments indicate that down- cancer treatment (Cummings et al., 2006; Cheung et al., regulating survivin expression can be beneficial. AS 2006a) and is in multiple phase I or II clinical trials in ODNs targeting survivin expression in human lung the United Kingdom, Canada and the United States for adenocarcinoma cell lines decrease survivin protein solid tumors, lymphoma and leukemia (AML) (Table 3). levels in a dose-dependent manner, induce apoptosis, Initial AEG35156 trial results appear promising stimulate higher levels of caspase-3 activation, and (Cheung et al., 2006a; Danson et al., 2007; Dean et al., increase the sensitivity of cells to chemotherapeutics 2007). (Olie et al., 2000; Jiang et al., 2001). Inhibition of survivin expression, using different AS ODNs spanning the entire survivin gene, induces cell death through either SMCs as promising antagonists of XIAP, livin, cIAP1 a caspase-dependent or -independent pathway depend- and cIAP2 ing on the neuronal tumor type (Shankar et al., 2001). In Small molecule IAP antagonists have been designed to vivo models showing the effect of survivin downregula- mimic the interaction between XIAP and its antagonist

Oncogene IAP-targeted therapeutics EC LaCasse et al 6266 Table 3 IAP small molecule antagonists as research tools or clinical development candidates Target Compound Class Developmental stage Company

XIAP AEG35156/GEM640 Antisense Phase 2 Aegera, Montreal, Quebec, Canada (Idera, Cambridge, MA, USA) Xantags (e.g. 1396-11, 1396-12) Caspase-3 de-repressor Preclinical TWX006, TWX024 Caspase-3 de-repressor Preclinical Embelin and derivatives Natural product Preclinical

cIAP1, cIAP2, AEG40826/HGS1029 Smac mimetic Phase 1 Aegera (Human Genome Sciences, XIAP, livin Rockville, MD, USA) LBW242 Smac mimetic Preclinical Novartis, Cambridge, MA, USA Compound 3 Smac mimetic Preclinical Joyant, Dallas, TX, USA Compound 11 Smac mimetic Preclinical Pﬁzer, New York, NY, USA (Idun, La Jolla, CA, USA/Abbott, Abbott Park, IL, USA) Compound C, compound 8, Smac mimetic Phase 1, Genentech, San Francisco, CA, USA BV6 preclinical GT-T, compound A Smac mimetic Preclinical Tetralogic, Malvern, PA, USA (Amgen, Thousand Oaks, CA, USA) SM-164, SM-122 Smac mimetic Preclinical Ascenta, Malvern, PA, USA

cIAP1 Bestatin methyl ester cIAP1 degradation promotera Preclinical Nippon Kayaku Co., Tokyo, Japan Bestatin–actinonin hybrid, cIAP1 degradation promotera Preclinical Nippon Kayaku Co. HAB-5(30b)

cIAP2 Ro106-9920 IKK ubiquitination inhibitor Preclinical Roche, Palo Alto, CA, USA

Survivin LY2181308/ISIS23722 Antisense Phase 2 Eli Lilly, Indianapolis, IN, USA (Isis, Carlsbad, CA, USA) YM155 Transcriptional repressor Phase 2 Astellas, Tokyo, Japan Terameprocol Transcriptional repressor Phase 2 Erimos, Raleigh, NC, USA (EM-1421, M4N) Shepherdin Hsp90 antagonist Preclinical AICAR Hsp90 antagonist, AMPKagonist Phase 2 Immunotherapy Phase 2

Abbreviations: AICAR, 5-aminoimidazole-4-carboxamide-1-b-D-ribofuranoside; AMPK, AMP-activated protein kinase; HSP90, heat stock protein 90; M4N (Terameprocol), tetra-O-methyl nordihydroguaiaretic acid; IAP, inhibitor of apoptosis. aThese bestatin (Ubenimex) analogs are referred to as degradation promoters of cIAP1, and may be Smac mimetics. However, these analogs were not rationally designed on the AVPI tetrapeptide structure, so it is unclear whether they are bona ﬁde Smac mimetics. Only co-crystallization studies will reveal whether they bind to the same pocket (IBM groove) on the IAPs as do the Smac mimetics.

Smac, and these show great promise for the treatment of activation. Accordingly, in the absence of both cIAP1 cancer (Varfolomeev et al., 2007). SMCs were intended and cIAP2, TNFa induces apoptosis because the pro- to de-repress XIAP-mediated inhibition of apoptosis survival NF-kB pathway is blunted. On the other hand, in cancer cells. However, a number of groups (Gaither recent data show that cIAP1 and cIAP2 regulate RIP1 et al., 2007; Petersen et al., 2007; Varfolomeev et al., ubiquitination in response to TNFa, by targeting RIP1 2007; Vince et al., 2007; Bertrand et al., 2008; Wang directly or acting as a ‘brake’ on its de-ubiquitination. et al., 2008) have now demonstrated quite unexpectedly This suggests that cIAP1 and cIAP2 may regulate that the ability of SMCs to kill cancer cells is due to TNFa-mediated apoptosis in an NF-kB-independent cIAP1 and cIAP2 degradation rather than XIAP manner (O’Donnell et al., 2007). The dual loss of cIAP1 antagonism. Mechanistically, SMCs induce auto-ubi- and cIAP2, therefore, appears to sensitize cells to quitination and degradation of cIAP1 and cIAP2, which TNFa-mediated apoptosis by attenuating RIP1 ubiqui- in turn relieves the repression by cIAP1 and cIAP2 on tination, with unmodified RIP1 having the capacity to ligand-independent, classical and alternative NF-kB interact with caspase-8 and induce apoptosis (Wang signaling. Consequently, these NF-kB pathways are et al., 2008). It is noteworthy that although normal cells activated, which leads to the production of TNFa in a do not die in response to dual cIAP1 and cIAP2 subset of the cancer cells. The newly synthesized TNFa knockdown (presumably because they do not generate is subsequently secreted to activate TNF receptors TNFa), they are greatly sensitized to apoptosis on through autocrine or paracrine pathways. This TNFa exposure to exogenous TNFa (Mahoney et al., 2008). signaling loop strongly induces caspase-8-dependent As such, the pursuit of this very promising new apoptosis in the absence of cIAP1 and cIAP2, although treatment strategy should nevertheless proceed cau- the mechanism of action is somewhat controversial. On tiously, to better define the therapeutic window and the one hand, our recent data (Mahoney et al., 2008) as delineate the systemic and localized levels of TNFa and well as others (Santoro et al., 2007; Bertrand et al., the tissue-specific toxic effects. 2008), have demonstrated that cIAP1 and cIAP2 are Several biotechnological and pharmaceutical compa- critical positive regulators of TNFa-mediated NF-kB nies now have active programs developing SMCs as

Oncogene IAP-targeted therapeutics EC LaCasse et al 6267 INTRAMOLECULAR BRIDGING unexpectedly, dimeric SMCs are far more potent than monomeric SMCs, as they faithfully reproduce the BIR1 BIR2 action of endogenous Smac polypeptides that antagonize the IAPs as homodimer complexes.

XIAP

RING BIR3 Mechanistically, SMCs are thought to be apoptosis sensitizers, not apoptosis inducers, and therefore are postulated to require an additional stimulus such as chemotherapy. However, several cancer cell lines are killed outright by SMCs at low nanomolar concentra- INTERMOLECULAR BRIDGING tions; but the additional stimulus is likely the autocrine TNFa loop. Interestingly, cancer cells that are resistant BIR1 BIR2 BIR3 CARD RING to single-agent SMC treatment may be sensitized to TRAIL or chemotherapy, increasing the usefulness of

cIAP1 these compounds. However, caveats and concerns BIR1 CARD BIR3 BIR2

RING surrounding SMCs are already being raised with the few studies done so far. For example, SMC treatment may not be appropriate for B- or T-cell neoplasms, such BIR RING as non-Hodgkin’s lymphoma, Hodgkin’s disease or multiple myeloma, in which activation of the alternative livin pathway is common and leads to a survival signal. That BIR RING is, the SMC-induced loss of cIAP1 and cIAP2 may not lead to cell death by TNFa in these cancers but rather Figure 4 Mechanism of action for dimeric Smac mimetic stimulate tumor cell survival. Moreover, the activation compounds (SMCs): intramolecular versus intermolecular bridging of the alternative NF-kB pathway may lead to models. Two models are shown to explain the increased potency of dimeric SMCs versus monomeric SMCs. The small red barbell autoimmune disease, in a manner similar to cells that structure represents a dimeric SMC. The model either involves receive a stimulus from CD40-L, CD30-L, BAFF, internal ligation of baculovirus inhibitor of apoptosis repeat LIGHT or lymphotoxin-b, members of the TNFR (BIR3)–BIR2 domains within the same IAP (intramolecular superfamily of ligand that are capable of providing a bridging, shown for example for XIAP) versus ligation of BIR3 domains (the higher affinity site) between two separate IAPs survival signal in lymphocytes. Additionally, depending (intermolecular bridging, shown for example for cIAP1). Livin, on dosing, tumor burden and tumor sites, coupled with with a single BIR domain, is not capable of internal ligation and the local and systemic levels of TNFa produced in thus only SMC-induced protein dimers are possible. The bivalent response to SMCs, the normal surrounding or distant binding of a dimeric SMC increases its affinity compared with a tissue(s) may be negatively affected, such as that seen monovalent binding of a monomeric SMC. In addition, the forced dimerization or internal ligation of an IAP produced by the dimeric with cancer cachexia, anemia or sepsis. Therefore, SMCs enhances auto- or trans-ubiquitination leading to IAP patients treated with SMCs should be monitored for degradation, or effectively blocks caspase-3, -7 and -9 entry such possibilities and related adverse events. Clearly, simultaneously (not shown). The two models are not mutually many specific aspects of SMC mechanism of action exclusive within the same cell, and other configurations can be envisaged. That is, head-to-head or head-to-tail protein dimers are remain to be determined to maximize their impact in possible, allowing for additional interactions besides BIR3–BIR3 the clinic. pairings, such as BIR2–BIR2 or BIR3–BIR2. In addition, the polarity of the interaction may place the RING domain in a more Additional XIAP small molecule antagonists favorable ubiquitination context closer to its target residues. Much Other small molecule approaches targeting XIAP have remains to be determined for the mechanism of action of these dimeric SMCs and how they stimulate tumor necrosis factor-a lead to the identification of novel compounds (Table 3). (TNFa) production and subsequent cell death. For example, a series of polyphenylurea-based compounds, now called Xantags, were identified that lead to anticancer agents (Table 3). Two such compounds have caspase-3/-7 activation prior to the activation of up- already entered the clinic with more SMCs not far stream caspase-8 and -9 (Schimmer et al., 2004). behind. Dimeric SMCs have been found to exhibit a Specifically, these Xantags dissociated caspase-3 from 2- to 3-log increase in potency over their monomeric XIAP in vitro, but not caspase-9, in contrast to SMCs counterparts (Li et al., 2004). This finding suggests that (Wang et al., 2004). These XIAP antagonists demon- anchoring of one side of an SMC dimer into BIR3 will strate anticancer activity in several different models favor its interaction of its other half with the lower (Schimmer et al., 2004; Wang et al., 2004; Carter et al., affinity BIR2 site, and/or that the dual inhibition of 2005; Karikari et al., 2007; Cillessen et al., 2008). BIR2 and BIR3 results in greater caspase activation by Similarly, Wu et al. (2003) undertook a high-throughput simultaneously blocking and releasing caspase-3 or -7, biochemical screen of a combinatorial chemical library and -9. Alternatively, the SMC dimer may induce that led to the discovery of a novel non-peptidic small ligation of BIR2 and BIR3 within an IAP, or the molecule, TWX006, that has the ability to disrupt the ligation of BIR3–BIR3 domains between two IAPs, XIAP–caspase-3 interaction. resulting in greater stimulation of auto- or trans- A structure-based computational screening of a ubiquitination and enhanced protein degradation natural product structure database identified embelin, through the proteasome (Figure 4). Therefore, not from the Japanese Ardisia herb, as a small molecule

Oncogene IAP-targeted therapeutics EC LaCasse et al 6268 inhibitor that binds to XIAP BIR3 and leads to caspase- proteasomal degradation of survivin, mitochondrial- 9 activation in cancer cells with high XIAP and nominal dependent apoptosis and cell cycle arrest with mitotic effects in normal cells with low XIAP (Nikolovska- defects (Fortugno et al., 2003). On the basis of these Coleska et al., 2004). Derivatives of embelin have been findings, a small molecule peptidomimetic, shepherdin, designed with higher affinity for XIAP BIR3 (for was rationally designed to disrupt the hsp90–survivin example, compound 6 g, Ki ¼ 180 nM) to provide leads interaction (Plescia et al., 2005). Shepherdin not only for further optimization (Chen et al., 2006). Nuclear affects survivin expression but also destabilizes other magnetic resonance and molecular docking analyses hsp90 client proteins. Newer, non-peptidic, small confirmed that embelin interacts with several residues in molecules such as AICAR, similar in structure to the XIAP BIR3 domain with which Smac and caspase-9 shepherdin (Meli et al., 2006), are being evaluated. bind (Nikolovska-Coleska et al., 2004; Obiol-Pardo AICAR acts as an AMPKagonist, and thus affects et al., 2008). It remains to be determined whether other pathways such as protein translation and glucose embelin and its derivatives possess properties similar to metabolism (Koh et al., 2008). AICAR is being SMCs in their ability to produce TNFa, or sensitize investigated primarily as a treatment for type II diabetes cancer cells to exogenous TNFa, and result in the (Towler and Hardie, 2007) and its effects may pertain proteosomal loss of cIAP1 and cIAP2. primarily to AMPKactivation. AICAR and shepherdin do not show any deleterious effects in mice; however, AMPKactivation by AICAR is reported to produce an Additional cIAP1 and cIAP2 modulators acute drop in plasma potassium levels (Zheng et al., Another class of cIAP1 protein destabilizers exists in the 2008). Both AICAR and shepherdin appear to have form of analogs of bestatin (Ubenimex), an aminopep- anticancer effects (Buzzai et al., 2007; Guan et al., 2007; tidase inhibitor used in the treatment of leukemia in Isakovic et al., 2007), including cell death by apoptotic Japan (Bauvois and Dauzonne, 2006) (Table 3). The and non-apoptotic means. However, there is still cause bestatin analogs degrade cIAP1 in the micromolar for concern with survivin antagonism. Importantly, cell range, and may, in fact, be Smac mimetics as they bind cycle disruption effects and polyploidy are observed to BIR3 of cIAP1 (Sato et al., 2008). They are referred with some of the survivin antagonists. This is reminis- to as degradation promoters of cIAP1, and they cent of the multitude of survivin ablation studies were not rationally designed on the AVPI tetrapeptide performed with antisense or RNAi, which have not structure. Therefore, it is unclear whether they are only lead to the stimulation of apoptosis but also an bona fide Smac mimetics. Ultimately, co-crystallization increase in multinucleated cells with aberrant chromo- studies should reveal whether they bind to the same some numbers due presumably to cytokinesis defects pocket (IBM groove) on the IAPs as do the Smac resulting from survivin’s loss (Li et al., 1999; Yang et al., mimetics. 2004). Survivin is a well-established chromosomal Another small molecule inhibitor of NF-kB activa- passenger protein involved in chromosome alignment tion, Ro106-9920, that was identified in a cell-free and segregation during mitosis and cytokinesis system as an inhibitor of IkBa ubiquitination (Swinney (Ruchaud et al., 2007). This function of survivin is et al., 2002) requires cIAP2 or a cIAP2-associated conserved throughout evolution. Hence, there is concern protein for activity. However, it has not been definitively that survivin antagonism may increase the aggressive- demonstrated that cIAP2 E3 ligase is directly inhibited ness of the surviving tumor cells by altering or by Ro106-9920 and mechanism remains speculative. increasing their ploidy, leading to increased dosage of oncogenes or reductions in tumor suppressor gene Survivin small molecule antagonists number. Additionally, specialized pools of normal Several small molecule antagonists of survivin are under survivin-expressing cells, such as stem cells, may be development, or in the clinic, all with very different affected by these antagonists in ways that are not mechanisms of action and chemistries (Table 3). YM155 initially apparent. Further studies will be needed to is a small molecule identified to suppress the activity of address these concerns. the survivin promoter (Nakahara et al., 2007). YM155 is currently undergoing phase II clinical evaluation in Adenoviral expression of dominant-negative mutants of patients with cancer (Altieri, 2008). Terameprocol (EM- survivin 1421, M4N), a semisynthetic derivative of a naturally Adenoviral vectors targeting regulators of both cell cycle occurring plant lignan, is another small molecule that and/or apoptosis are being examined for potential suppresses survivin gene expression (Chang et al., 2004). applications in cancer gene therapy alone, either as However, it is known that Terameprocol affects global stand-alone treatment or in combination with che- Sp1-dependent gene expression, and therefore not only motherapy drugs (Mesri et al., 2001). One such targets survivin but other genes as well, such as the cell application is the intratumor adenoviral delivery of cycle regulator Cdc2 (Chang et al., 2004; Huang et al., wild-type p53 in patients with non-small cell lung 2006). Previously, the molecular chaperone protein, cancer, thereby restoring checkpoint functions of hsp90, was demonstrated to interact with and stabilize apoptosis and/or cell cycle arrest (Swisher et al., 1999). survivin (Fortugno et al., 2003). Global suppression of Targeting the survivin pathway through dominant- hsp90 chaperone function, or antibody-mediated dis- negative mutants in an adenoviral gene therapy ruption of the survivin-hsp90 complex, results in approach may also be beneficial for cancer treatment.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6269 Transfection of a survivin dominant-negative mutant, Approaches targeting survivin epitopes are currently carrying a cysteine 84-alanine mutation in the BIR undergoing preclinical and clinical evaluation (reviewed domain or a Thr34-Ala mutation, into melanoma cell in Altieri, 2008). Several phase I trials, involving either lines significantly increases the number of apoptotic administration of survivin peptides or survivin-directed cells. The Thr34-Ala mutation destroys a phosphoryla- autologous CTLs, have concluded and progressed on to tion site for the cyclin-dependent kinase p34cdc2, required larger phase II trials. for survivin’s association with pro-caspase-9 and inhibition of apoptosis (O’Connor et al., 2000). Conditional expression of the Thr34-Ala survivin mutant has been shown to prevent tumor formation in a subcutaneous Conclusions xenograft model of melanoma (Grossman et al., 2001). A replication-deficient adenovirus (pAd-T34A), used Apoptosis is controlled at multiple intracellular nodes, to overexpress the Thr34-Ala survivin dominant mu- each of which is influenced by pro- and antiapoptotic tant, causes spontaneous apoptosis in breast, cervical, proteins. The equilibrium between the cell death that prostate, lung and colorectal cancer cell lines. In induces the caspase cascade and inhibition of that process contrast, pAd-T34A has no observable effects on the by the IAPs constitutes a fundamental decision point. proliferation or viability of healthy cells that do not The recurrent deregulation of IAP activity in cancer cell express survivin (Mesri et al., 2001). Mitochondrial lines and tumors indicates that this decision point is release of cytochrome c, cleavage of caspase-9 and -3, important in determining cell fate. The IAPs not only and the catalytic enhancement of caspase-3 activity are control cell death but also influence signal-transduction observed in the apoptotic pAd-T34A-infected cells. In pathways, differentiation, protein turnover and progres- combination with chemotherapeutics such as Taxol, the sion through the cell cycle. Still, many aspects of IAP extent of apoptosis in HeLa and MCF-7 cells was also function in all these processes remain to be elucidated. found to be increased (Mesri et al., 2001). Experiments With the recognition of apoptosis suppression as a using the MCF-7 human breast cancer xenograft model fundamental aspect of human cancer, the IAPs and other shows that pAd-T34A transduction suppresses de novo antiapoptotic proteins are acknowledged as being out- tumor formation, inhibits the growth of pre-established standing therapeutic targets with several small molecules tumors, reduces intraperitoneal tumor dissemination now undergoing intense clinical investigation. and induces massive apoptosis in all transduced cells (Mesri et al., 2001). Thus, viral targeting of the survivin Acknowledgements pathway using dominant-negative mutants may be a powerful tool for selective cancer gene therapy. RGKis a Howard Hughes Medical Institute (HHMI) International Research Scholar and a Fellow of the Royal Survivin immunotherapy Society of Canada. RGKis supported by funds from the Cancer vaccines targeting cancer-restricted epitopes Canadian Institutes of Health Research (CIHR), and the HHMI. HHC is the recipient of a Michael Cuccione represent a novel approach to eradicate tumors through Foundation and CIHR Institute of Cancer Research post- stimulation of the host’s immune system. Survivin is a doctoral fellowship. SP and DJM are recipients of a Natural shared tumor-associated antigen expressed in a variety Sciences and Engineering Research Council of Canada of malignancies. Sera from cancer patients contain postdoctoral fellowships. ECL is supported by funds from antibodies and cytolytic T-cells (CTLs) against survivin. the Optimist Club’s Campaign against Childhood Cancers.

References

Adida C, Crotty PL, McGrath J, Berrebi D, Diebold J, Altieri D. Annunziata CM, Davis RE, Demchenko Y, Bellamy W, Gabrea A, (1998). Developmentally regulated expression of the novel cancer Zhan F et al. (2007). Frequent engagement of the classical and anti-apoptosis gene survivin in human and mouse differentiation. alternative NF-kappaB pathways by diverse genetic abnormalities Am J Pathol 152: 43–49. in multiple myeloma. Cancer Cell 12: 115–130. Adrain C, Creagh EM, Martin SJ. (2001). Apoptosis-associated release Ansell SM, Arendt BK, Grote DM, Jelinek DF, Novak AJ, Wellik LE of Smac/DIABLO from mitochondria requires active caspases and et al. (2004). Inhibition of survivin expression suppresses the growth is blocked by Bcl-2. EMBO J 20: 6627–6636. of aggressive non-Hodgkin’s lymphoma. Leukemia 18: 616–623. Altieri DC. (2001). The molecular basis and potential role of survivin Arnt CR, Chiorean MV, Heldebrant MP, Gores GJ, Kaufmann SH. in cancer diagnosis and therapy. Trends Mol Med 7: 542–547. (2002). Synthetic Smac/DIABLO peptides enhance the effects of Altieri DC. (2008). Survivin, cancer networks and pathway-directed chemotherapeutic agents by binding XIAP and cIAP1 in situ. J Biol drug discovery. Nat Rev Cancer 8: 61–70. Chem 277: 44236–44243. Amantana A, London CA, Iversen PL, Devi GR. (2004). X-linked Arora V, Cheung HH, Plenchette S, Micali OC, Liston P, Korneluk inhibitor of apoptosis protein inhibition induces apoptosis and RG. (2007). Degradation of survivin by the X-linked inhibitor of enhances chemotherapy sensitivity in human prostate cancer cells. apoptosis (XIAP)-XAF1 complex. J Biol Chem 282: 26202–26209. Mol Cancer Ther 3: 699–707. Ashhab Y, Alian A, Polliack A, Panet A, Ben Yehuda D. (2001). Two Ambrosini G, Adida C, Altieri D. (1997). A novel anti-apoptosis gene, splicing variants of a new inhibitor of apoptosis gene with different survivin, expressed in cancer and lymphoma. Nat Med 3: 917–921. biological properties and tissue distribution pattern. FEBS Lett 495: Ambrosini G, Adida C, Sirugo G, Altieri DC. (1998). Induction of 56–60. apoptosis and inhibition of cell proliferation by survivin gene Baens M, Fevery S, Sagaert X, Noels H, Hagens S, Broeckx V et al. targeting. J Biol Chem 273: 11177–11182. (2006). Selective expansion of marginal zone B cells in Em-API2-

Oncogene IAP-targeted therapeutics EC LaCasse et al 6270 MALT1 mice is linked to enhanced IkappaB kinase gamma Chu ZL, McKinsey TA, Liu L, Gentry JJ, Malim MH, Ballard DW. polyubiquitination. Cancer Res 66: 5270–5277. (1997). Suppression of tumor necrosis factor-induced cell death by Baens M, Maes B, Steyls A, Geboes K, Marynen P, De Wolf-Peeters inhibitor of apoptosis c-IAP2 is under NF-kappaB control. Proc C. (2000). The product of the t(11;18), an API2-MLT fusion, marks Natl Acad Sci USA 94: 10057–10062. nearly half of gastric MALT type lymphomas without large cell Chung SK, Lee MG, Ryu BK, Lee JH, Han J, Byun DS et al. (2007). proliferation. Am J Pathol 156: 1433–1439. Frequent alteration of XAF1 in human colorectal cancers: Bauvois B, Dauzonne D. (2006). Aminopeptidase-N/CD13 (EC implication for tumor cell resistance to apoptotic stresses. Gastro- 3.4.11.2) inhibitors: chemistry, biological evaluations, and thera- enterology 132: 2459–2477. peutic prospects. Med Res Rev 26: 88–130. Cillessen SA, Reed JC, Welsh K, Pinilla C, Houghten R, Hooijberg E Bertrand MJ, Milutinovic S, Dickson KM, Ho WC, Boudreault A, et al. (2008). Small-molecule XIAP antagonist restores caspase-9 Durkin J et al. (2008). cIAP1 and cIAP2 facilitate cancer cell mediated apoptosis in XIAP-positive diffuse large B-cell lymphoma survival by functioning as E3 ligases that promote RIP1 ubiquitina- cells. Blood 111: 369–375. tion. Mol Cell 30: 689–700. Clem RJ, Sheu TT, Richter BW, He WW, Thornberry NA, Duckett Bockbrader KM, Tan M, Sun Y. (2005). A small molecule Smac-mimic CS et al. (2001). c-IAP1 is cleaved by caspases to produce a compound induces apoptosis and sensitizes TRAIL- and etoposide- proapoptotic C-terminal fragment. J Biol Chem 276: 7602–7608. induced apoptosis in breast cancer cells. Oncogene 24: 7381–7388. Conte D, Holcik M, Lefebvre CA, LaCasse E, Picketts DJ, Wright KE Brennan DJ, Rexhepaj E, O’Brien SL, McSherry E, O’Connor DP, et al. (2006). Inhibitor of apoptosis protein cIAP2 is essential for Fagan A et al. (2008). Altered cytoplasmic-to-nuclear ratio of lipopolysaccharide-induced macrophage survival. Mol Cell Biol 26: survivin is a prognostic indicator in breast cancer. Clin Cancer Res 699–708. 14: 2681–2689. Conway EM, Pollefeyt S, Cornelissen J, DeBaere I, Steiner-Mosonyi Bria E, Visca P, Novelli F, Casini B, Diodoro MG, Perrone-Donnorso M, Ong K et al. (2000). Three differentially expressed survivin R et al. (2008). Nuclear and cytoplasmic cellular distribution of cDNA variants encode proteins with distinct antiapoptotic func- survivin as survival predictor in resected non-small-cell lung cancer. tions. Blood 95: 1435–1442. Eur J Surg Oncol 34: 593–598. Conze DB, Albert L, Ferrick DA, Goeddel DV, Yeh WC, Mak T et al. Burri L, Strahm Y, Hawkins CJ, Gentle IE, Puryer MA, Verhagen A (2005). Posttranscriptional downregulation of c-IAP2 et al. (2005). Mature DIABLO/Smac is produced by the IMP by the ubiquitin protein ligase c-IAP1 in vivo. Mol Cell Biol 25: protease complex on the mitochondrial inner membrane. Mol Biol 3348–3356. Cell 16: 2926–2933. Coornaert B, Baens M, Heyninck K, Bekaert T, Haegman M, Staal J Buzzai M, Jones RG, Amaravadi RK, Lum JJ, DeBerardinis RJ, et al. (2008). T cell antigen receptor stimulation induces MALT1 Zhao F et al. (2007). Systemic treatment with the antidiabetic drug paracaspase-mediated cleavage of the NF-kappaB inhibitor A20. metformin selectively impairs p53-deficient tumor cell growth. Nat Immunol 9: 263–271. Cancer Res 67: 6745–6752. Creagh EM, Murphy BM, Duriez PJ, Duckett CS, Martin SJ. (2004). Cao C, Mu Y, Hallahan DE, Lu B. (2004). XIAP and survivin as Smac/Diablo antagonizes ubiquitin ligase activity of inhibitor of therapeutic targets for radiation sensitization in preclinical models apoptosis proteins. J Biol Chem 279: 26906–26914. of lung cancer. Oncogene 23: 7047–7052, 9448. Crnkovic-Mertens I, Semzow J, Hoppe-Seyler F, Butz K. (2006). Carter BZ, Gronda M, Wang Z, Welsh K, Pinilla C, Andreeff M et al. Isoform-specific silencing of the Livin gene by RNA interference (2005). Small-molecule XIAP inhibitors derepress downstream defines Livin beta as key mediator of apoptosis inhibition in HeLa effector caspases and induce apoptosis of acute myeloid leukemia cells. J Mol Med 84: 232–240. cells. Blood 105: 4043–4050. Cummings J, Ranson M, LaCasse E, Ganganagari JR, St-Jean M, Chai J, Du C, Wu JW, Kyin S, Wang X, Shi Y. (2000). Structural and Jayson G et al. (2006). Method validation and preliminary biochemical basis of apoptotic activation by Smac/DIABLO. qualification of pharmacodynamic biomarkers employed to evaluate Nature 406: 855–862. the clinical efficacy of an antisense compound (AEG35156) targeted Chang CC, Heller JD, Kuo J, Huang RC. (2004). Tetra-O-methyl to the X-linked inhibitor of apoptosis protein XIAP. Br J Cancer 95: nordihydroguaiaretic acid induces growth arrest and cellular 42–48. apoptosis by inhibiting Cdc2 and survivin expression. Proc Natl Dai Z, Zhu WG, Morrison CD, Brena RM, Smiraglia DJ, Raval A Acad Sci USA 101: 13239–13244. et al. (2003). A comprehensive search for DNA amplification in lung Chantalat L, Skoufias DA, Kleman JP, Jung B, Dideberg O, Margolis cancer identifies inhibitors of apoptosis cIAP1 and cIAP2 as RL. (2000). Crystal structure of human survivin reveals a bow tie- candidate oncogenes. Hum Mol Genet 12: 791–801. shaped dimer with two unusual alpha-helical extensions. Mol Cell 6: Dan HC, Sun M, Kaneko S, Feldman RI, Nicosia SV, Wang HG et al. 183–189. (2004). Akt phosphorylation and stabilization of X-linked inhibitor Chastagner P, Israel A, Brou C. (2006). Itch/AIP4 mediates Deltex of apoptosis protein (XIAP). J Biol Chem 279: 5405–5412. degradation through the formation of K29-linked polyubiquitin Danson S, Dean E, Dive C, Ranson M. (2007). IAPs as a target for chains. EMBO Rep 7: 1147–1153. anticancer therapy. Curr Cancer Drug Targets 7: 785–794. Chen J, Nikolovska-Coleska Z, Wang G, Qiu S, Wang S. (2006). Davoodi J, Lin L, Kelly J, Liston P, MacKenzie AE. (2004). Neuronal Design, synthesis, and characterization of new embelin derivatives apoptosis-inhibitory protein does not interact with Smac and as potent inhibitors of X-linked inhibitor of apoptosis protein. requires ATP to bind caspase-9. J Biol Chem 279: 40622–40628. Bioorg Med Chem Lett 16: 5805–5808. Dean EJ, Ranson M, Blackhall F, Holt SV, Dive C. (2007). Novel Chen Z, Naito M, Hori S, Mashiima T, Yamori T, Tsuruo T. (1999). A therapeutic targets in lung cancer: inhibitor of apoptosis proteins human IAP-family gene, apollon, expressed in human brain cancer from laboratory to clinic. Cancer Treat Rev 33: 203–212. cells. Biochem Biophys Res Commun 264: 847–854. Deveraux QL, Roy N, Stennicke HR, Van Arsdale T, Zhou Q, Cheung HH, LaCasse EC, Korneluk RG. (2006a). X-linked inhibitor Srinivasula SM et al. (1998). IAPs block apoptotic events induced of apoptosis antagonism: strategies in cancer treatment. Clin Cancer by caspase-8 and cytochrome c by direct inhibition of distinct Res 12: 3238–3242. caspases. EMBO J 17: 2215–2223. Cheung HH, Kelly LN, Liston P, Korneluk RG. (2006b). Involvement Deveraux QL, Takahashi R, Salvesen GS, Reed JC. (1997). X-linked of caspase-2 and caspase-9 in endoplasmic reticulum stress-induced IAP is a direct inhibitor of cell-death proteases. Nature 388: 300–304. apoptosis: a role for the IAPs. Exp Cell Res 312: 2347–2357. Dierlamm J, Baens M, Wlodarska I, Stefanova-Ouzounova M, Cheung HH, Plenchette S, Kern CJ, Mahoney DJ, Korneluk RG. Hernandez JM, Hossfeld DK et al. (1999). The apoptosis inhibitor (2008). The RING domain of cIAP1 mediates the degradation of gene API2 and a novel 18q gene, MLT, are recurrently rearranged RING-bearing inhibitor of apoptosis proteins by distinct pathways. in the t(11;18)(q21;q21) associated with mucosa-associated lym- Mol Biol Cell 19: 2729–2740. phoid tissue lymphomas. Blood 93: 3601–3609.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6271 Dohi T, Okada K, Xia F, Wilford CE, Samuel T, Welsh K et al. Gleave M, Miyake H, Zangemeister-Wittke U, Jansen B. (2002). (2004). An IAP-IAP complex inhibits apoptosis. J Biol Chem 279: Antisense therapy: current status in prostate cancer and other 34087–34090. malignancies. Cancer Metastasis Rev 21: 79–92. Dohi T, Xia F, Altieri DC. (2007). Compartmentalized phosphoryla- Goncalves RB, Sanches D, Souza TL, Silva JL, Oliveira AC. (2008). tion of IAP by protein kinase A regulates cytoprotection. Mol Cell The proapoptotic protein Smac/DIABLO dimer has the highest 27: 17–28. stability as measured by pressure and urea denaturation. Biochem- Du C, Fang M, Li Y, Li L, Wang X. (2000). Smac, a mitochondrial istry 47: 3832–3841. protein that promotes cytochrome c-dependent caspase activation Grossman D, Kim PJ, Schechner JS, Altieri DC. (2001). Inhibition of by eliminating IAP inhibition. Cell 102: 33–42. melanoma tumor growth in vivo by survivin targeting. Proc Natl Du MQ. (2007). MALT lymphoma: recent advances in aetiology and Acad Sci USA 98: 635–640. molecular genetics. J Clin Exp Hematop 47: 31–42. Guan TJ, Qin FJ, Du JH, Geng L, Zhang YY, Li M. (2007). AICAR Duckett CS, Nava VE, Gedrich RW, Clem RJ, Van Dongen JL, inhibits proliferation and induced S-phase arrest, and promotes Gilfillan MC et al. (1996). A conserved family of cellular genes apoptosis in CaSki cells. Acta Pharmacol Sin 28: 1984–1990. related to the baculovirus iap gene and encoding apoptosis Guo F, Nimmanapalli R, Paranawithana S, Wittman S, Griffin D, inhibitors. EMBO J 15: 2685–2694. Bali P et al. (2002). Ectopic overexpression of second mitochondria- Ea CK, Deng L, Xia ZP, Pineda G, Chen ZJ. (2006). Activation of derived activator of caspases (Smac/DIABLO) or cotreatment with IKK by TNFalpha requires site-specific ubiquitination of RIP1 and N-terminus of Smac/DIABLO peptide potentiates epothilone B polyubiquitin binding by NEMO. Mol Cell 22: 245–257. derivative-(BMS 247550) and Apo-2 L/TRAIL-induced apoptosis. Eckelman BP, Salvesen GS. (2006). The human anti-apoptotic proteins Blood 99: 3419–3426. cIAP1 and cIAP2 bind but do not inhibit caspases. J Biol Chem 281: Hanahan D, Weinberg RA. (2000). The hallmarks of cancer. Cell 100: 3254–3260. 57–70. Eckelman BP, Salvesen GS, Scott FL. (2006). Human inhibitor of Hao Y, Sekine K, Kawabata A, Nakamura H, Ishioka T, Ohata H apoptosis proteins: why XIAP is the black sheep of the family. et al. (2004). Apollon ubiquitinates SMAC and caspase-9, and has EMBO Rep 7: 988–994. an essential cytoprotection function. Nat Cell Biol 6: 849–860. Ekert PG, Vaux DL. (2005). The mitochondrial death squad: hardened Hauser HP, Bardroff M, Pyrowolakis G, Jentsch S. (1998). A giant killers or innocent bystanders? Curr Opin Cell Biol 17: 626–630. ubiquitin-conjugating enzyme related to IAP apoptosis inhibitors. Engelsma D, Rodriguez JA, Fish A, Giaccone G, Fornerod M. (2007). J Cell Biol 141: 1415–1422. Homodimerization antagonizes nuclear export of survivin. Traffic 8: Hegde R, Srinivasula SM, Datta P, Madesh M, Wassell R, Zhang Z 1495–1502. et al. (2003). The polypeptide chain-releasing factor GSPT1/eRF3 is Erl W, Hansson GK, de Martin R, Draude G, Weber KS, Weber C. proteolytically processed into an IAP-binding protein. J Biol Chem (1999). Nuclear factor-kappa B regulates induction of apoptosis and 278: 38699–38706. inhibitor of apoptosis protein-1 expression in vascular smooth Hegde R, Srinivasula SM, Zhang Z, Wassell R, Mukattash R, Cilenti muscle cells. Circ Res 84: 668–677. L et al. (2002). Identification of Omi/HtrA2 as a mitochondrial Evan GI, Vousden KH. (2001). Proliferation, cell cycle and apoptosis apoptotic serine protease that disrupts inhibitor of apoptosis in cancer. Nature 411: 342–348. protein–caspase interaction. J Biol Chem 277: 432–438. Ferreira CG, van der Valk P, Span SW, Jonker JM, Postmus PE, Hess CJ, Berkhof J, Denkers F, Ossenkoppele GJ, Schouten JP, Kruyt FA et al. (2001a). Assessment of IAP (inhibitor of apoptosis) Oudejans JJ et al. (2007). Activated intrinsic apoptosis pathway is a proteins as predictors of response to chemotherapy in advanced key related prognostic parameter in acute myeloid leukemia. J Clin non-small-cell lung cancer patients. Ann Oncol 12: 799–805. Oncol 25: 1209–1215. Ferreira CG, van der Valk P, Span SW, Ludwig I, Smit EF, Kruyt FA. Hinds MG, Norton RS, Vaux DL, Day CL. (1999). Solution structure (2001b). Expression of X-linked inhibitor of apoptosis as a novel of a baculoviral inhibitor of apoptosis (IAP) repeat. Nat Struct Biol prognostic marker in radically resected non-small cell lung cancer 6: 648–651. patients. Clin Cancer Res 7: 2468–2474. Hochstrasser M. (2006). Lingering mysteries of ubiquitin-chain Fong WG, Liston P, Rajcan-Separovic E, St Jean M, Craig C, assembly. Cell 124: 27–34. Korneluk RG. (2000). Expression and genetic analysis of Hong SY, Yoon WH, Park JH, Kang SG, Ahn JH, Lee TH. (2000). XIAP-associated factor 1 (XAF1) in cancer cell lines. Genomics Involvement of two NF-kappa B binding elements in tumor necrosis 70: 113–122. factor alpha -, CD40-, and Epstein–Barr virus latent membrane Fortugno P, Beltrami E, Plescia J, Fontana J, Pradhan D, Marchisio protein 1-mediated induction of the cellular inhibitor of apoptosis PC et al. (2003). Regulation of survivin function by Hsp90. Proc protein 2 gene. J Biol Chem 275: 18022–18028. Natl Acad Sci USA 100: 13791–13796. Hosokawa Y. (2005). Anti-apoptotic action of API2-MALT1 fusion Fu J, Jin Y, Arend LJ. (2003). Smac3, a novel Smac/DIABLO splicing protein involved in t(11;18)(q21;q21) MALT lymphoma. Apoptosis variant, attenuates the stability and apoptosis-inhibiting activity 10: 25–34. of X-linked inhibitor of apoptosis protein. J Biol Chem 278: Hu H, Shikama Y, Matsuoka I, Kimura J. (2008). Terminally 52660–52672. differentiated neutrophils predominantly express Survivin-2 alpha, Fukuda S, Pelus LM. (2006). Survivin, a cancer target with an a dominant-negative isoform of survivin. J Leukoc Biol 83: 393–400. emerging role in normal adult tissues. Mol Cancer Ther 5: Hu S, Alcivar A, Qu L, Tang J, Yang X. (2006a). CIAP2 inhibits 1087–1098. anigen receptor signaling by targeting Bcl10 for degredation. Cell Fulda S, Wick W, Weller M, Debatin KM. (2002). Smac agonists Cycle 5: 1438–1442. sensitize for Apo2 L/TRAIL- or anticancer drug-induced apoptosis Hu S, Du MQ, Park SM, Alcivar A, Qu L, Gupta S et al. (2006b). and induce regression of malignant glioma in vivo. Nat Med 8: cIAP2 is a ubiquitin protein ligase for BCL10 and is dysregulated in 808–815. mucosa-associated lymphoid tissue lymphomas. J Clin Invest 116: Gaither A, Porter D, Yao Y, Borawski J, Yang G, Donovan J et al. 174–181. (2007). A Smac mimetic rescue screen reveals roles for inhibitor of Hu Y, Cherton-Horvat G, Dragowska V, Baird S, Korneluk RG, apoptosis proteins in tumor necrosis factor-alpha signaling. Cancer Durkin JP et al. (2003). Antisense oligonucleotides targeting XIAP Res 67: 11493–11498. induce apoptosis and enhance chemotherapeutic activity against Galderisi U, Cascino A, Giordano A. (1999). Antisense oligonucleo- human lung cancer cells in vitro and in vivo. Clin Cancer Res 9: tides as therapeutic agents. J Cell Phys 181: 251–257. 2826–2836. Galvan V, Kurakin AV, Bredesen DE. (2004). Interaction of Huang RC, Chang CC, Mold D. (2006). Survivin-dependent and checkpoint kinase 1 and the X-linked inhibitor of apoptosis during -independent pathways and the induction of cancer cell death by mitosis. FEBS Lett 558: 57–62. tetra-O-methyl nordihydroguaiaretic acid. Semin Oncol 33: 479–485.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6272 Huang Y, Rich RL, Myszka DG, Wu H. (2003). Requirement of both can function as a chromosomal passenger complex protein. Cell the second and third BIR domains for the relief of X-linked Cycle 6: 1502–1509. inhibitor of apoptosis protein (XIAP)-mediated caspase inhibition Kobayashi K, Hatano M, Otaki M, Ogasawara T, Tokuhisa T. (1999). by Smac. J Biol Chem 278: 49517–49522. Expression of a murine homologue of the inhibitor of apoptosis Hunter AM, LaCasse EC, Korneluk RG. (2007). The inhibitors of protein is related to cell proliferation. Proc Natl Acad Sci USA 96: apoptosis (IAPs) as cancer targets. Apoptosis 12: 1543–1568. 1457–1462. Hunter T. (2007). The age of crosstalk: phosphorylation, ubiquitina- Koh HJ, Brandauer J, Goodyear LJ. (2008). LKB1 and AMPK and tion, and beyond. Mol Cell 28: 730–738. the regulation of skeletal muscle metabolism. Curr Opin Clin Nutr Imoto I, Yang ZQ, Pimkhaokham A, Tsuda H, Shimada Y, Imamura Metab Care 11: 227–232. M et al. (2001). Identification of cIAP1 as a candidate target gene LaCasse EC, Baird S, Korneluk RG, MacKenzie AE. (1998). The within an amplicon at 11q22 in esophageal squamous cell inhibitors of apoptosis (IAPs) and their emerging role in cancer. carcinomas. Cancer Res 61: 6629–6634. Oncogene 17: 3247–3259. Isakovic A, Harhaji L, Stevanovic D, Markovic Z, Sumarac- LaCasse EC, Kandimalla ER, Winocour P, Sullivan T, Agrawal S, Dumanovic M, Starcevic V et al. (2007). Dual antiglioma action Gillard JW et al. (2005). Therapeutic oligonucleotides: transcrip- of metformin: cell cycle arrest and mitochondria-dependent tional and translational strategies for silencing gene expression. Ann apoptosis. Cell Mol Life Sci 64: 1290–1302. NY Acad Sci 1058: 215–234. Islam A, Kageyama H, Hashizume K, Kaneko Y, Nakagawara A. Lagace M, Xuan JY, Young SS, McRoberts C, Maier J, Rajcan- (2000). Role of survivin, whose gene is mapped to 17q25, in human Separovic E et al. (2001). Genomic organization of the X-linked neuroblastoma and identification of a novel dominant-negative inhibitor of apoptosis and identification of a novel testis-specific isoform, survivin-beta/2B. Med Ped Oncol 35: 550–553. transcript. Genomics 77: 181–188. Jansen B, Zangemeister-Wittke U. (2002). Antisense therapy for Leaman DW, Chawla-Sarkar M, Vyas K, Reheman M, Tamai K, cancer––the time of truth. The Lancet 3: 672–683. Toji S et al. (2002). Identification of X-linked inhibitor of Jeyaprakash AA, Klein UR, Lindner D, Ebert J, Nigg EA, Conti E. apoptosis-associated factor-1 as an interferon-stimulated gene that (2007). Structure of a Survivin-Borealin-INCENP core complex augments TRAIL Apo2 L-induced apoptosis. J Biol Chem 277: reveals how chromosomal passengers travel together. Cell 131: 28504–28511. 271–285. Lee TH, Shank J, Cusson N, Kelliher MA. (2004). The kinase activity Jiang X, Wilford C, Duensing S, Munger K, Jones G, Jones D. (2001). of Rip1 is not required for tumor necrosis factor-alpha-induced Participation of Survivin in mitotic and apoptotic activities of IkappaB kinase or p38 MAP kinase activation or for the normal and tumor-derived cells. J Cell Biochem 83: 342–354. ubiquitination of Rip1 by Traf2. J Biol Chem 279: 33185–33191. Jin C, Reed JC. (2002). Yeast and apoptosis. Nat Rev Mol Cell Biol 3: Li F. (2005). Role of survivin and its splice variants in tumorigenesis. 453–459. Br J Cancer 92: 212–216. Kandasamy K, Srinivasula SM, Alnemri ES, Thompson CB, Li F, Ackermann EJ, Bennett CF, Rothermel AL, Plescia J, Tognin S Korsemeyer SJ, Bryant JL et al. (2003). Involvement of proapopto- et al. (1999). Pleiotropic cell-division defects and apoptosis tic molecules Bax and Bak in tumor necrosis factor-related induced by interference with survivin function. Nat Cell Biol 1: apoptosis-inducing ligand (TRAIL)-induced mitochondrial disrup- 461–466. tion and apoptosis: differential regulation of cytochrome c and Li J, Feng Q, Kim JM, Schneiderman D, Liston P, Li M et al. (2001). Smac/DIABLO release. Cancer Res 63: 1712–1721. Human ovarian cancer and cisplatin resistance: possible role of Kanneganti TD, Lamkanfi M, Nunez G. (2007). Intracellular inhibitor of apoptosis proteins. Endocrinology 142: 370–380. NOD-like receptors in host defense and disease. Immunity 27: Li L, Thomas RM, Suzuki H, De Brabander JK, Wang X, Harran PG. 549–559. (2004). A small molecule Smac mimic potentiates TRAIL- and Kappler M, Rot S, Taubert H, Greither T, Bartel F, Dellas K et al. TNFalpha-mediated cell death. Science 305: 1471–1474. (2007). The effects of knockdown of wild-type survivin, survivin-2B Li X, Yang Y, Ashwell JD. (2002). TNF-RII and c-IAP1 or survivin-delta3 on the radiosensitization in a soft tissue sarcoma mediate ubiquitination and degradation of TRAF2. Nature 416: cells in vitro under different oxygen conditions. Cancer Gene Ther 345–347. 14: 994–1001. Lima RT, Martins LM, Guimaraes JE, Sambade C, Vasconcelos MH. Karikari CA, Roy I, Tryggestad E, Feldmann G, Pinilla C, Welsh K (2004). Specific downregulation of bcl-2 and xIAP by RNAi et al. (2007). Targeting the apoptotic machinery in pancreatic enhances the effects of chemotherapeutic agents in MCF-7 human cancers using small-molecule antagonists of the X-linked inhibitor breast cancer cells. Cancer Gene Ther 11: 309–316. of apoptosis protein. Mol Cancer Ther 6: 957–966. Liston P, Fong WG, Kelly NL, Toji S, Miyazaki T, Conte D et al. Kasof GM, Gomes BC. (2001). Livin, a novel inhibitor of apoptosis (2001). Identification of XAF1 as an antagonist of XIAP anti- protein family member. J Biol Chem 276: 3238–3246. caspase activity. Nat Cell Biol 3: 128–133. Kaur S, Wang F, Venkatraman M, Arsura M. (2005). X-linked Liston P, Fong WG, Korneluk RG. (2003). The inhibitors of inhibitor of apoptosis (XIAP) inhibits c-Jun N-terminal kinase 1 apoptosis: there is more to life than Bcl2. Oncogene 22: 8568–8580. (JNK1) activation by transforming growth factor beta1 (TGF- Liston P, Roy N, Tamai K, Lefebvre C, Baird S, Cherton-Horvat G beta1) through ubiquitin-mediated proteosomal degradation et al. (1996). Suppression of apoptosis in mammalian cells by NAIP of the TGF-beta1-activated kinase 1 (TAK1). J Biol Chem 280: and a related family of IAP genes. Nature 379: 349–353. 38599–38608. Liu S, Tsang BK, Cheung AN, Xue WC, Cheng DK, Ng TY. (2001). Keats JJ, Fonseca R, Chesi M, Schop R, Baker A, Chng WJ et al. Anti-apoptotic proteins, apoptotic and proliferative parameters and (2007). Promiscuous mutations activate the noncanonical their prognostic significance in cervical carcinoma. Eur J Cancer 37: NF-kappaB pathway in multiple myeloma. Cancer Cell 12: 131–144. 1104–1110. Kipp RA, Case MA, Wist AD, Cresson CM, Carrell M, Griner E et al. Liu Z, Sun C, Olejniczak ET, Meadows RP, Betz SF, Oost T et al. (2002). Molecular targeting of inhibitor of apoptosis proteins based (2000). Structural basis for binding of Smac/DIABLO to the XIAP on small molecule mimics of natural binding partners. Biochemistry BIR3 domain. Nature 408: 1004–1008. 41: 7344–7349. Lu M, Lin SC, Huang Y, Kang YJ, Rich R, Lo YC et al. (2007). XIAP Kleinberg L, Florenes VA, Silins I, Haug K, Trope CG, Nesland JM induces NF-kappaB activation via the BIR1/TAB1 interaction and et al. (2007). Nuclear expression of survivin is associated with BIR1 dimerization. Mol Cell 26: 689–702. improved survival in metastatic ovarian carcinoma. Cancer 109: Lucas PC, Kuffa P, Gu S, Kohrt D, Kim DS, Siu K et al. (2007). Dual 228–238. role for the API2 moiety in API2-MALT1-dependent NF-kappaB Knauer SK, Bier C, Schlag P, Fritzmann J, Dietmaier W, Rodel F activation: heterotypic oligomerization and TRAF2 recruitment. et al. (2007). The survivin isoform survivin-3B is cytoprotective and Oncogene 26: 5643–5654.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6273 Madesh M, Antonsson B, Srinivasula SM, Alnemri ES, Hajno´ ckzy G. Noton EA, Colnaghi R, Tate S, Starck C, Carvalho A, Ko Ferrigno P (2002). Rapid kinetics of tBid-induced cytochrome c and Smac/ et al. (2006). Molecular analysis of survivin isoforms: evidence that DIABLO release and mitochondrial depolarization. J Biol Chem alternatively spliced variants do not play a role in mitosis. J Biol 277: 5651–5659. Chem 281: 1286–1295. Mahoney DJ, Cheung HH, Lejmi Mrad R, Plenchette S, Simard C, O’Connor DS, Grossman D, Plescia J, Li F, Zhang H, Villa A et al. Enwere E et al. (2008). Both cIAP1 and cIAP2 regulate TNFa- (2000). Regulation of apoptosis at cell division by p34cdc2 mediated NF-kB activation. Proc Natl Acad Sci USA 105: phosphorylation of survivin. Proc Natl Acad Sci USA 97: 11778–11783. 13103–13107. Mao HL, Liu PS, Zheng JF, Zhang PH, Zhou LG, Xin G et al. (2007). O’Donnell MA, Legarda-Addison D, Skountzos P, Yeh WC, Transfection of Smac/DIABLO sensitizes drug-resistant tumor cells Ting AT. (2007). Ubiquitination of RIP1 regulates an NF- to TRAIL or paclitaxel-induced apoptosis in vitro. Pharmacol Res kappaB-independent cell-death switch in TNF signaling. Curr Biol 56: 483–492. 17: 418–424. Mariathasan S, Monack DM. (2007). Inflammasome adaptors and Obiol-Pardo C, Granadino-Roldan JM, Rubio-Martinez J. (2008). sensors: intracellular regulators of infection and inflammation. Nat Protein–protein recognition as a first step towards the inhibition Rev Immunol 7: 31–40. of XIAP and Survivin anti-apoptotic proteins. J Mol Recognit 21: Martin SJ. (2001). Dealing the CARDs between life and death. Trends 190–204. Cell Biol 11: 188–189. Olie RA, Simoes-Wust AP, Baumann B, Leech SH, Fabbro D, Stahel Martins LM, Iaccarino I, Tenev T, Gschmeissner S, Totty NF, RA et al. (2000). A novel antisense oligonucleotide targeting Lemoine NR et al. (2002). The serine protease Omi/HtrA2 regulates survivin expression induces apoptosis and sensitizes lung cancer apoptosis by binding XIAP through a reaper-like motif. J Biol cells to chemotherapy. Cancer Res 60: 2805–2809. Chem 277: 439–444. Oost TK, Sun C, Armstrong RC, Al-Assaad AS, Betz SF, Deckwerth McManus DC, Lefebvre CA, Cherton-Horvat G, St-Jean M, TL et al. (2004). Discovery of potent antagonists of the anti- Kandimalla ER, Agrawal S et al. (2004). Loss of XIAP protein apoptotic protein XIAP for the treatment of cancer. J Med Chem expression by RNAi and antisense approaches sensitizes cancer cells 47: 4417–4426. to functionally diverse chemotherapeutics. Oncogene 23: 8105–8117. Pannone G, Bufo P, Serpico R, Rubini C, Zamparese R, Corsi F et al. Meli M, Pennati M, Curto M, Daidone MG, Plescia J, Toba S et al. (2007). Survivin phosphorylation and M-phase promoting factor in (2006). Small-molecule targeting of heat shock protein 90 chaperone oral carcinogenesis. Histol Histopathol 22: 1241–1249. function: rational identification of a new anticancer lead. JMed Park CM, Sun C, Olejniczak ET, Wilson AE, Meadows RP, Betz SF Chem 49: 7721–7730. et al. (2005). Non-peptidic small molecule inhibitors of XIAP. Mesri M, Wall NR, Li J, Kim RW, Altieri D. (2001). Cancer Bioorg Med Chem Lett 15: 771–775. gene therapy using a survivin mutant adenovirus. J Clin Invest 108: Petersen SL, Wang L, Yalcin-Chin A, Li L, Peyton M, Minna J et al. 981–990. (2007). Autocrine TNFalpha signaling renders human cancer cells Micali OC, Cheung HH, Plenchette S, Hurley SL, Liston P, LaCasse susceptible to Smac-mimetic-induced apoptosis. Cancer Cell 12: EC et al. (2007). Silencing of the XAF1 gene by promoter 445–456. hypermethylation in cancer cells and reactivation to TRAIL- Pinho MB, Costas F, Sellos J, Dienstmann R, Andrade PB, Herchen- sensitization by IFN-beta. BMC Cancer 7: 52. horn D et al. (2008). XAF1 mRNA expression improves progres- Mizutani Y, Nakanishi H, Li YN, Matsubara H, Yamamoto K, Sato sion-free and overall survival for patients with advanced bladder N et al. (2007). Overexpression of XIAP expression in renal cell cancer treated with neoadjuvant chemotherapy. Urol Oncol; carcinoma predicts a worse prognosis. Int J Oncol 30: 919–925. e-pub June 16. Mola G, Vela E, Fernandez-Figueras MT, Isamat M, Munoz-Marmol Piras F, Murtas D, Minerba L, Ugalde J, Floris C, Maxia C et al. AM. (2007). Exonization of Alu-generated splice variants in the (2007). Nuclear survivin is associated with disease recurrence and survivin gene of human and non-human primates. J Mol Biol 366: poor survival in patients with cutaneous malignant melanoma. 1055–1063. Histopathology 50: 835–842. Mosley JD, Keri RA. (2006). Splice variants of mIAP1 have an Plenchette S, Cathelin S, Rebe C, Launay S, Ladoire S, Sordet O et al. enhanced ability to inhibit apoptosis. Biochem Biophys Res Commun (2004). Translocation of the inhibitor of apoptosis protein c-IAP1 348: 1174–1183. from the nucleus to the Golgi in hematopoietic cells undergoing Nachmias B, Mizrahi S, Elmalech M, Lazar I, Ashhab Y, Gazit R differentiation: a nuclear export signal-mediated event. Blood 104: et al. (2007). Manipulation of NKcytotoxicity by the IAP family 2035–2043. member Livin. Eur J Immunol 37: 3467–3476. Plenchette S, Cheung HH, Fong WG, LaCasse EC, Korneluk RG. Nakahara T, Takeuchi M, Kinoyama I, Minematsu T, Shirasuna K, (2007). The role of XAF1 in cancer. Curr Opin Investig Drugs 8: Matsuhisa A et al. (2007). YM155, a novel small-molecule 469–476. survivin suppressant, induces regression of established human Plescia J, Salz W, Xia F, Pennati M, Zaffaroni N, Daidone MG et al. hormone-refractory prostate tumor xenografts. Cancer Res 67: (2005). Rational design of shepherdin, a novel anticancer agent. 8014–8021. Cancer Cell 7: 457–468. Ng CP, Bonavida B. (2002). X-linked inhibitor of apoptosis (XIAP) Pohl C, Jentsch S. (2008). Final stages of cytokinesis and midbody ring blocks Apo2 ligand/tumor necrosis factor-related apoptosis-indu- formation are controlled by BRUCE. Cell 132: 832–845. cing ligand-mediated apoptosis of prostate cancer cells in the Qi R, Gu J, Zhang Z, Yang K, Li B, Fan J et al. (2006). Potent presence of mitochondrial activation: sensitization by overexpres- antitumor efficacy of XAF1 delivered by conditionally replicative sion of second mitochondria-derived activator of caspase/direct adenovirus vector via caspase-independent apoptosis. Cancer Gene IAP-binding protein with low pl (Smac/DIABLO). Mol Cancer Ther Ther 14: 82–90. 1: 1051–1058. Qiu XB, Goldberg AL. (2005). The membrane-associated inhibitor Nikolovska-Coleska Z, Xu L, Hu Z, Tomita Y, Li P, Roller PP et al. of apoptosis protein, BRUCE/Apollon, antagonizes both the (2004). Discovery of embelin as a cell-permeable, small-molecular precursor and mature forms of Smac and caspase-9. J Biol Chem weight inhibitor of XIAP through structure-based computational 280: 174–182. screening of a traditional herbal medicine three-dimensional Ramp U, Krieg T, Caliskan E, Mahotka C, Ebert T, Willers R et al. structure database. J Med Chem 47: 2430–2440. (2004). XIAP expression is an independent prognostic marker in Noels H, van Loo G, Hagens S, Broeckx V, Beyaert R, Marynen P clear-cell renal carcinomas. Hum Pathol 35: 1022–1028. et al. (2007). A novel TRAF6 binding site in MALT1 defines distinct Rebeaud F, Hailfinger S, Posevitz-Fejfar A, Tapernoux M, Moser R, mechanisms of NF-kappaB activation by API2 Á MALT1 fusions. Rueda D et al. (2008). The proteolytic activity of the paracaspase J Biol Chem 282: 10180–10189. MALT1 is key in T cell activation. Nat Immunol 9: 272–281.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6274 Remstein ED, James CD, Kurtin PJ. (2000). Incidence and subtype a strong predictor of human prostate cancer recurrence. Clin Cancer specificity of API2-MALT1 fusion translocations in extranodal, Res 13: 6056–6063. nodal, and splenic marginal zone lymphomas. Am J Pathol 156: Shankar SL, Mani S, O’Guin KN, Kandimalla ER, Agrawal S, Shafit- 1183–1188. Zagardo B. (2001). Survivin inhibition induces human neural tumor Reu FJ, Bae SI, Cherkassky L, Leaman DW, Lindner D, Beaulieu N cell death through caspase-independent and -dependent pathways. et al. (2006). Overcoming resistance to interferon-induced apoptosis J Neurochem 79: 426–436. of renal carcinoma and melanoma cells by DNA demethylation. Shim JH, Xiao C, Paschal AE, Bailey ST, Rao P, Hayden MS et al. J Clin Oncol 24: 3771–3779. (2005). TAK1, but not TAB1 or TAB2, plays an essential role in Rhee I, Bachman KE, Park BH, Jair KW, Yen RW, Schuebel KE et al. multiple signaling pathways in vivo. Genes Dev 19: 2668–2681. (2002). DNMT1 and DNMT3b cooperate to silence genes in human Shin H, Okada K, Wilkinson JC, Solomon KM, Duckett CS, Reed JC cancer cells. Nature 416: 552–556. et al. (2003). Identification of ubiquitination sites on the X-linked Richter BW, Mir SS, Eiben LJ, Lewis J, Reffrey SB, Frattini A et al. inhibitor of apoptosis protein. Biochem J 373: 965–971. (2001). Molecular cloning of ILP-2, a novel member of the inhibitor Shin H, Renatus M, Eckelman BP, Nunes VA, Sampaio CA, Salvesen of apoptosis protein family. Mol Cell Biol 21: 4292–4301. GS. (2005). The BIR domain of IAP-like protein 2 is conforma- Rigaud S, Fondaneche MC, Lambert N, Pasquier B, Mateo V, Soulas tionally unstable: implications for caspase inhibition. Biochem J P et al. (2006). XIAP deficiency in humans causes an X-linked 385: 1–10. lymphoproliferative syndrome. Nature 444: 110–114. Shu HB, Takeuchi M, Goeddel DV. (1996). The tumor necrosis factor Roberts DL, Merrison W, MacFarlane M, Cohen GM. (2001). The receptor 2 signal transducers TRAF2 and c-IAP1 are components of inhibitor of apoptosis protein-binding domain of Smac is not the tumor necrosis factor receptor 1 signaling complex. Proc Natl essential for its proapoptotic activity. J Cell Biol 153: 221–227. Acad Sci USA 93: 13973–13978. Robertson AJ, Croce J, Carbonneau S, Voronina E, Miranda E, Silke J, Kratina T, Chu D, Ekert PG, Day CL, Pakusch M et al. McClay DR et al. (2006). The genomic underpinnings of apoptosis (2005). Determination of cell survival by RING-mediated regula- in Strongylocentrotus purpuratus. Dev Biol 300: 321–334. tion of inhibitor of apoptosis (IAP) protein abundance. Proc Natl Rothe M, Pan MG, Henzel WJ, Ayres TM, Goeddel DV. (1995). Acad Sci USA 102: 16182–16187. The TNFR2-TRAF signaling complex contains two novel Span PN, Tjan-Heijnen VC, Heuvel JJ, de Kok JB, Foekens JA, Sweep proteins related to baculoviral inhibitor of apoptosis proteins. Cell FC. (2006). Do the survivin (BIRC5) splice variants modulate or 83: 1243–1252. add to the prognostic value of total survivin in breast cancer? Clin Roy N, Mahadevan MS, McLean M, Shutler G, Yaraghi Z, Farahani Chem 52: 1693–1700. R et al. (1995). The gene for neuronal apoptosis inhibitory protein is Srinivasula SM, Hegde R, Saleh A, Datta P, Shiozaki E, Chai J et al. partially deleted in individuals with spinal muscular atrophy. Cell (2001). A conserved XIAP-interaction motif in caspase-9 and 80: 167–178. Smac/DIABLO regulates caspase activity and apoptosis. Nature Ruchaud S, Carmena M, Earnshaw WC. (2007). Chromosomal 410: 112–116. passengers: conducting cell division. Nat Rev Mol Cell Biol 8: Stamm S, Riethoven JJ, Le Texier V, Gopalakrishnan C, Kumanduri 798–812. V, Tang Y et al. (2006). ASD: a bioinformatics resource on Sagaert X, Theys T, De Wolf-Peeters C, Marynen P, Baens M. (2006). alternative splicing. Nucleic Acids Res 34: D46–D55. Splenic marginal zone lymphoma-like features in API2-MALT1 Stauber RH, Mann W, Knauer SK. (2007). Nuclear and cytoplasmic transgenic mice that are exposed to antigenic stimulation. Haema- survivin: molecular mechanism, prognostic, and therapeutic poten- tologica 91: 1693–1696. tial. Cancer Res 67: 5999–6002. Samuel T, Welsh K, Lober T, Togo SH, Zapata JM, Reed JC. (2006). Sun C, Cai M, Gunasekera AH, Meadows RP, Wang H, Chen J et al. Distinct BIR domains of cIAP1 mediate binding to and ubiquitina- (1999). NMR structure and mutagenesis of the inhibitor-of- tion of tumor necrosis factor receptor-associated factor 2 apoptosis protein XIAP. Nature 401: 818–822. and second mitochondrial activator of caspases. J Biol Chem 281: Suzuki Y, Nakabayashi Y, Takahashi R. (2001). Ubiquitin-protein 1080–1090. ligase activity of X-linked inhibitor of apoptosis protein promotes Sanna MG, da Silva Correia J, Ducrey O, Lee J, Nomoto K, Schrantz proteasomal degradation of caspase-3 and enhances its anti- N et al. (2002a). IAP suppression of apoptosis involves distinct apoptotic effect in Fas-induced cell death. Proc Natl Acad Sci mechanisms: the TAK1/JNK1 signaling cascade and caspase USA 98: 8662–8667. inhibition. Mol Cell Biol 22: 1754–1766. Swinney DC, Xu YZ, Scarafia LE, Lee I, Mak AY, Gan QF et al. Sanna MG, da Silva Correia J, Luo Y, Chuang B, Paulson LM, (2002). A small molecule ubiquitination inhibitor blocks NF-kappa Nguyen B et al. (2002b). ILPIP, a novel anti-apoptotic protein that B-dependent cytokine expression in cells and rats. J Biol Chem 277: enhances XIAP-mediated activation of JNK1 and protection 23573–23581. against apoptosis. J Biol Chem 277: 30454–30462. Swisher SG, Roth JA, Nemunaitis J, Lawrence DD, Kemp BL, Sanna MG, Duckett CS, Richter BW, Thompson CB, Ulevitch RJ. Carrasco CH et al. (1999). Adenovirus-mediated p53 gene (1998). Selective activation of JNK1 is necessary for the anti- transfer in advanced non-small-cell lung cancer. J Natl Cancer Inst apoptotic activity of hILP. Proc Natl Acad Sci USA 95: 6015–6020. 91: 763–771. Santoro MM, Samuel T, Mitchell T, Reed JC, Stainier DY. (2007). Tamm I, Kornblau SM, Segall H, Krajewski S, Welsh K, Kitada S Birc2 (cIap1) regulates endothelial cell integrity and blood vessel et al. (2000). Expression and prognostic significance of IAP-family homeostasis. Nat Genet 39: 1397–1402. genes in human cancers and myeloid leukemias. Clin Cancer Res 6: Sato S, Aoyama H, Miyachi H, Naito M, Hashimoto Y. (2008). 1796–1803. Demonstration of direct binding of cIAP1 degradation-promoting Ting JP, Willingham SB, Bergstralh DT. (2008). NLRs at the bestatin analogs to BIR3 domain: synthesis and application intersection of cell death and immunity. Nat Rev Immunol 8: of fluorescent bestatin ester analogs. Bioorg Med Chem Lett 18: 372–379. 3354–3358. Towler MC, Hardie DG. (2007). AMP-activated protein kinase in Schimmer AD, Welsh K, Pinilla C, Wang Z, Krajewska M, Bonneau metabolic control and insulin signaling. Circ Res 100: 328–341. MJ et al. (2004). Small-molecule antagonists of apoptosis Tu SP, Jiang XH, Lin MC, Cui JT, Yang Y, Lum CT et al. (2003). suppressor XIAP exhibit broad antitumor activity. Cancer Cell 5: Suppression of survivin expression inhibits in vivo tumorigenicity 25–35. and angiogenesis in gastric cancer. Cancer Res 63: 7724–7732. Schulze-Luehrmann J, Ghosh S. (2006). Antigen-receptor signaling to Uren AG, O’Rourke K, Aravind LA, Pisabarro MT, Seshagiri S, nuclear factor kappa B. Immunity 25: 701–715. Koonin EV et al. (2000). Identification of paracaspases and Seligson DB, Hongo F, Huerta-Yepez S, Mizutani Y, Miki T, Yu H metacaspases: two ancient families of caspase-like proteins, one of et al. (2007). Expression of X-linked inhibitor of apoptosis protein is which plays a key role in MALT lymphoma. Mol Cell 6: 961–967.

Oncogene IAP-targeted therapeutics EC LaCasse et al 6275 Uren AG, Pakusch M, Hawkins CJ, Puls KL, Vaux DL. (1996). Wright ME, Han DK, Hockenbery DM. (2000). Caspase-3 and Cloning and expression of apoptosis inhibitory protein homologs inhibitor of apoptosis protein(s) interactions in Saccharomyces that function to inhibit apoptosis and/or bind tumor necrosis factor cerevisiae and mammalian cells. FEBS Lett 481: 13–18. receptor-associated factors. Proc Natl Acad Sci USA 93: 4974–4978. Wu CJ, Conze DB, Li T, Srinivasula SM, Ashwell JD. (2006). Sensing Varambally S, Yu J, Laxman B, Rhodes DR, Mehra R, Tomlins SA of Lys 63-linked polyubiquitination by NEMO is a key event in et al. (2005). Integrative genomic and proteomic analysis of prostate NF-kappaB activation. Nat Cell Biol 8: 398–406. cancer reveals signatures of metastatic progression. Cancer Cell 8: Wu CJ, Conze DB, Li X, Ying SX, Hanover JA, Ashwell JD. (2005a). 393–406. TNF-alpha induced c-IAP1/TRAF2 complex translocation to a Varfolomeev E, Blankenship JW, Wayson SM, Fedorova AV, Ubc6-containing compartment and TRAF2 ubiquitination. EMBO Kayagaki N, Garg P et al. (2007). IAP antagonists induce J 24: 1886–1898. autoubiquitination of c-IAPs, NF-kappaB activation, and TNFal- Wu G, Chai J, Suber TL, Wu JW, Du C, Wang X et al. (2000). pha-dependent apoptosis. Cell 131: 669–681. Structural basis of IAP recognition by Smac/DIABLO. Nature 408: Varfolomeev E, Wayson SM, Dixit VM, Fairbrother WJ, Vucic D. 1008–1012. (2006). The inhibitor of apoptosis protein fusion c-IAP2.MALT1 Wu M, Yuan S, Szporn AH, Gan L, Shtilbans V, Burstein DE. stimulates NF-kappaB activation independently of TRAF1 AND (2005b). Immunocytochemical detection of XIAP in body cavity TRAF2. J Biol Chem 281: 29022–29029. effusions and washes. Mod Pathol 18: 1618–1622. Vaux DL, Silke J. (2005). IAPs, RINGs and ubiquitylation. Nat Rev Wu TY, Wagner KW, Bursulaya B, Schultz PG, Deveraux QL. (2003). Mol Cell Biol 6: 287–297. Development and characterization of nonpeptidic small molecule Vega F, Medeiros LJ. (2001). Marginal-zone B-cell lymphoma of inhibitors of the XIAP/caspase-3 interaction. Chem Biol 10: extranodal mucosa-associated lymphoid tissue type: molecular 759–767. genetics provides new insights into pathogenesis. Adv Anat Pathol Xia Y, Novak R, Lewis J, Duckett CS, Phillips AC. (2006). Xaf1 can 8: 313–326. cooperate with TNFalpha in the induction of apoptosis, indepen- Verhagen AM, Ekert PG, Pakusch M, Silke J, Connolly LM, Reid GE dently of interaction with XIAP. Mol Cell Biochem 286: 67–76. et al. (2000). Identification of DIABLO, a mammalian protein that Xu L, Zhu J, Hu X, Zhu H, Kim HT, LaBaer J et al. (2007). c-IAP1 promotes apoptosis by binding to and antagonizing IAP proteins. cooperates with Myc by acting as a ubiquitin ligase for Mad1. Cell 102: 43–53. Mol Cell 28: 914–922. Verhagen AM, Kratina TK, Hawkins CJ, Silke J, Ekert PG, Vaux DL. Yamamoto K, Abe S, Nakagawa Y, Suzuki K, Hasegawa M, Inoue M (2007). Identification of mammalian mitochondrial proteins that et al. (2004). Expression of IAP family proteins in myelodysplastic interact with IAPs via N-terminal IAP binding motifs. Cell Death syndromes transforming to overt leukemia. Leuk Res 28: 1203–1211. Differ 14: 348–357. Yan Y, Mahotka C, Heikaus S, Shibata T, Wethkamp N, Liebmann J Verhagen AM, Silke J, Ekert PG, Pakusch M, Kaufmann H, Connolly et al. (2004). Disturbed balance of expression between XIAP and LM et al. (2002). HtrA2 promotes cell death through its serine Smac/DIABLO during tumour progression in renal cell carcinomas. protease activity and its ability to antagonize inhibitor of apoptosis Br J Cancer 91: 1349–1357. proteins. J Biol Chem 277: 445–454. Yang D, Welm A, Bishop JM. (2004). Cell division and cell survival in Vince JE, Chau D, Callus B, Wong WW, Hawkins CJ, Schneider P the absence of survivin. Proc Natl Acad Sci USA 101: 15100–15105. et al. (2008). TWEAK-FN14 signaling induces lysosomal degrada- Yang L, Cao Z, Yan H, Wood WC. (2003). Coexistence of high levels tion of a cIAP1-TRAF2 complex to sensitize tumor cells to of apoptotic signaling and inhibitor of apoptosis proteins in human TNFalpha. J Cell Biol 184: 171–184. tumor cells: implication for cancer specific therapy. Cancer Res 63: Vince JE, Wong WW, Khan N, Feltham R, Chau D, Ahmed AU et al. 6815–6824. (2007). IAP antagonists target cIAP1 to induce TNFalpha- Yang QH, Du C. (2004). Smac/DIABLO selectively reduces the levels dependent apoptosis. Cell 131: 682–693. of c-IAP1 and c-IAP2 but not that of XIAP and livin in HeLa cells. Vucic D, Deshayes K, Ackerly H, Pisabarro MT, Kadkhodayan S, J Biol Chem 279: 16963–16970. Fairbrother WJ et al. (2002). SMAC negatively regulates the anti- Yang Y, Fang S, Jensen JP, Weissman AM, Ashwell JD. (2000). apoptotic activity of melanoma inhibitor of apoptosis (ML-IAP). Ubiquitin protein ligase activity of IAPs and their degradation in J Biol Chem 277: 12275–12279. proteasomes in response to apoptotic stimuli. Science 288: 874–877. Vucic D, Fairbrother WJ. (2007). The inhibitor of apoptosis Yin W, Chen N, Zhang Y, Zeng H, Chen X, He Y et al. (2006). proteins as therapeutic targets in cancer. Clin Cancer Res 13: Survivin nuclear labeling index: a superior biomarker in superficial 5995–6000. urothelial carcinoma of human urinary bladder. Mod Pathol 19: Vucic D, Stennicke HR, Pisabarro MT, Salvessen GS, Dixit VM. 1487–1497. (2000). ML-IAP, a novel inhibitor of apoptosis that is preferentially Zender L, Spector MS, Xue W, Flemming P, Cordon-Cardo C, Silke J expressed in human melanomas. Curr Biol 10: 1359–1366. et al. (2006). Identification and validation of oncogenes in Wang C, Deng L, Hong M, Akkaraju GR, Inoue J, Chen ZJ. (2001). liver cancer using an integrative oncogenomic approach. Cell 125: TAK1 is a ubiquitin-dependent kinase of MKK and IKK. Nature 1253–1267. 412: 346–351. Zhao Y, Conze DB, Hanover JA, Ashwell JD. (2007). Tumor necrosis Wang J, He H, Yu L, Xia HH, Lin MC, Gu Q et al. (2006a). HSF1 factor receptor 2 signaling induces selective c-IAP1-dependent down-regulates XAF1 through transcriptional regulation. J Biol ASK1 ubiquitination and terminates mitogen-activated protein Chem 281: 2451–2459. kinase signaling. J Biol Chem 282: 7777–7782. Wang J, Peng Y, Sun YW, He H, Zhu S, An X et al. (2006b). All-trans Zheng D, Perianayagam A, Lee DH, Brannan MD, Yang LE, retinoic acid induces XAF1 expression through an interferon Tellalian D et al. (2008). AMPKactivation with AICAR provokes regulatory factor-1 element in colon cancer. Gastroenterology 130: an acute fall in plasma [K+]. Am J Physiol Cell Physiol 294: 747–758. C126–C135. Wang L, Du F, Wang X. (2008). TNF-alpha induces two distinct Zhou H, Du MQ, Dixit VM. (2005). Constitutive NF-kappaB activation caspase-8 activation pathways. Cell 133: 693–703. by the t(11;18)(q21;q21) product in MALT lymphoma is linked to Wang Z, Cuddy M, Samuel T, Welsh K, Schimmer A, Hanaii F et al. deregulated ubiquitin ligase activity. Cancer Cell 7: 425–431. (2004). Cellular, biochemical, and genetic analysis of mechanism of Zou B, Chim CS, Zeng H, Leung SY, Yang Y, Tu SP et al. (2006). small molecule IAP inhibitors. J Biol Chem 279: 48168–48176. Correlation between the single-site CpG methylation and expression Wegener E, Krappmann D. (2007). CARD-Bcl10-Malt1 signalosomes: silencing of the XAF1 gene in human gastric and colon cancers. missing link to NF-kappaB. Sci STKE 2007: pe21. Gastroenterology 131: 1835–1843.

Oncogene