Molecular Cell
Article Systematic In Vivo RNAi Analysis Identifies IAPs as NEDD8-E3 Ligases Meike Broemer,1,* Tencho Tenev,1 Kristoffer T.G. Rigbolt,3 Sophie Hempel,2 Blagoy Blagoev,3 John Silke,4 Mark Ditzel,1,2,5 and Pascal Meier1,5,* 1The Breakthrough Toby Robins Breast Cancer Research Centre, Institute of Cancer Research, Mary-Jean Mitchell Green Building, Chester Beatty Laboratories, Fulham Road, London SW3 6JB, UK 2Institute of Genetics and Molecular Medicine, Edinburgh Cancer Research Centre, Crewe Road South, Edinburgh EH4 2XR, UK 3Department Biochemistry & Molecular Biology, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark 4Department of Biochemistry, Level 4 RL Reid Building, La Trobe University, Victoria 3086, Australia 5These authors contributed equally to this work *Correspondence:
[email protected] (M.B.),
[email protected] (P.M.) DOI 10.1016/j.molcel.2010.11.011
SUMMARY
The intimate relationship between mediators of the ubiquitin (Ub)-signaling system and human diseases has sparked profound interest in how Ub influences cell death and survival. While the consequence of Ub attachment is intensely studied, little is known with regards to the effects of other Ub-like proteins (UBLs), and deconjugating enzymes that remove the Ub or UBL adduct. Systematic in vivo RNAi analysis identified three NEDD8-specific isopeptidases that, when knocked down, suppress apoptosis. Consistent with the notion that attachment of NEDD8 prevents cell death, genetic ablation of deneddylase 1 (DEN1) suppresses apoptosis. Unexpectedly, we find that Drosophila and human inhibitor of apoptosis (IAP) proteins can function as E3 ligases of the NEDD8 conjugation pathway, targeting effector caspases for neddylation and inactivation. Finally, we demonstrate that DEN1 reverses this effect by removing the NEDD8 modification. Altogether, our findings indicate that IAPs not only modulate cellular processes via ubiquitylation but also through attachment of NEDD8, thereby extending the complexity of IAP-mediated signaling. INTRODUCTION A major mechanism for regulating protein function involves the covalent attachment of Ubiquitin (Ub) and Ubiquitin-like proteins (UBLs). Bound Ub and UBLs thereby influence protein function either directly via conformational changes or indirectly through mediating interactions with other proteins (Dikic et al., 2009). Ub/UBL modifications regulate a multitude of cellular processes, including cell survival and apoptosis (Haglund and Dikic, 2005). Over the recent years, several Ubiquitin (Ub)-E3 ligases have emerged as key regulators of the apoptosis program (Broemer and Meier, 2009). Protein levels and activity of many pro- and
antiapoptotic molecules are controlled by E3-mediated conjugation of Ub. However, Ub-mediated regulation of cell survival is not just a mere consequence of Ub-directed proteasomal degradation. Nondegradative ubiquitylation events play also important roles (Ditzel et al., 2008; Jin et al., 2009; Schile et al., 2008), and, in combination with deubiquitylation, may allow a further level of apoptotic regulation. Ubiquitylation of caspases by inhibitor of apoptosis (IAP) proteins has been shown to play an important part in regulating caspase activity (Choi et al., 2009; Ditzel et al., 2008; Jin et al., 2009; Schile et al., 2008; Shapiro et al., 2008). In Drosophila, DIAP1 acts as RING Ub-E3 ligase that promotes the ubiquitylation of the initiator caspase DRONC (Chai et al., 2003; Herman-Bachinsky et al., 2007; Wilson et al., 2002) and the effector caspases drICE and DCP-1 (Ditzel et al., 2008). Although DIAP1 reportedly inhibits caspases in vitro (Kaiser et al., 1998; Yan et al., 2004), physical interaction alone seems does not seem to be sufficient to maintain cell viability in vivo. Mutations of DIAP1’s RING-finger domain, which abrogate its E3 activity but not caspase binding, cause a loss-of-function phenotype (Lisi et al., 2000; Wilson et al., 2002). In addition, the protein levels of DIAP1 are also regulated in an Ub-dependent manner (Ditzel and Meier, 2002). Specialized IAP-antagonist proteins, such as Reaper (Rpr) and Head involution defective (Hid) (Grether et al., 1995; White et al., 1994), adjust DIAP1 protein levels by inducing its autoubiquitylation and degradation (Holley et al., 2002; Ryoo et al., 2002; Yoo et al., 2002), leading to activation of apoptotic caspases and cell death. While the consequence of Ub conjugation is intensely studied (Hoeller and Dikic, 2009), much less is known with regards to the effects of UBL proteins such as SUMO, NEDD8, ATG12, ATG8, URM1, ISG15, and FAT10. To date, 17 different human UBLs have been identified (Schulman and Harper, 2009). Besides adopting a similar overall structure (Vijay-Kumar et al., 1987), UBLs share surprisingly limited amino acid sequence similarity with Ub. The most studied among the UBLs are SUMO (small Ub-related modifier) and NEDD8 (neural precursor cell expressed developmentally downregulated protein 8) (Kerscher et al., 2006). While SUMO modifications are most frequently associated with transcriptional suppression (Garcia-Dominguez and Reyes, 2009), conjugation of NEDD8 is best known for its
810 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
Molecular Cell IAPs as NEDD8 E3 Ligases
role in regulating cullin-type E3 ligases (Merlet et al., 2009). Although cullins are the best-studied substrates, they are not the only class of proteins modified by NEDD8 (Xirodimas, 2008). Recent studies have uncovered p53 (Xirodimas et al., 2004) and EGFR (Oved et al., 2006) as targets of the neddylation pathway. Ribosomal proteins reportedly are also modified by NEDD8, which seems to protect them from destabilization (Xirodimas et al., 2008). pVHL (von Hippel-Lindau tumor suppressor protein), BCA (breast cancer-associated protein) and APP intracellular domain (AICD) are further proteins that are modified by neddylation (Gao et al., 2006; Lee et al., 2008; Stickle et al., 2004). In these cases, neddylation seems to affect their interaction with binding partners. Ub/UBLs are attached to target proteins either as a single moiety or as polymeric chains of variable length (Komander, 2009). Ub/UBLs are transferred to lysine (K) residues of substrates in a stepwise process that involves activating enzymes (E1), conjugating enzymes (E2), and protein ligases (E3) (Hochstrasser, 2009). E3s thereby determine which substrate is modified, as they bind to both E2 and substrate, bringing the E2 in position for Ub/UBL transfer. While the E3 provides substrate specificity, the E2 determines which type of modification is formed. Importantly, each UBL employs its own specific E1/E2 cascade (Hochstrasser, 2009). While UBLs tend to use UBL-specific E1-E2 cascades, it seems that the last step, selection of the E3, is more flexible (Brzovic and Klevit, 2006). E2s bind to a wide range of different E3s, predominantly via an interaction with an E3’s HECT (homologous to E6-associated protein C terminus), RING (really interesting new gene), or U-box domain (Deshaies and Joazeiro, 2009; Hatakeyama and Nakayama, 2003; Rotin and Kumar, 2009). The covalent attachment of Ub/UBLs to target proteins is a reversible process. Specialized deconjugating enzymes (referred to as deubiquitylating enzymes [DUBs]) remove the Ub/UBL message (Reyes-Turcu et al., 2009). The human genome encodes approximately 100 predicted DUBs capable of deconjugating Ub, NEDD8, and SUMO from target proteins (Mukhopadhyay and Dasso, 2007; Nijman et al., 2005). DUBs are involved in the maturation of Ub/UBLs from precursor peptides, as well as in constitutive or regulated removal of the Ub/UBL modification from target proteins. Together with E3 ligases, DUBs are crucial regulators of many cellular processes, determining stability of proteins and influencing signaling events (Reyes-Turcu et al., 2009). Here, we have identified three NEDD8-specific isopeptidases that, when knocked down, suppress Rpr- and Hid-induced cell death. This suggests that the NEDD8 modification prevents apoptosis signaling. Genetic validation, using null mutant animals, confirmed the involvement of deneddylase 1 (DEN1) in regulating apoptosis. Consistent with the notion that conjugation of NEDD8 protects from apoptosis, we found that endogenous drICE is neddylated in healthy cells and that the NEDD8 modification reduces the proteolytic activity of drICE. Surprisingly, conjugation of NEDD8 to effector caspases was mediated by IAPs, in both Drosophila and mammals. Therefore, IAPs not only function as E3 ligases of the Ub conjugation system, but also take part in the NEDD8-specific cascade of protein modification. Consequently, IAPs can influence cellular
processes by conjugating Ub as well as NEDD8 to target substrates. RESULTS A Systematic In Vivo RNAi Screen Identifies NEDD8-Specific Proteases Involved in Apoptosis To identify DUBs and deconjugating enzymes for UBL that regulate programmed cell death, we conducted a systematic in vivo RNA interference (RNAi)-modifier screen in which we selectively knocked down individual DUBs in the developing eye of flies ectopically expressing the IAP antagonists Reaper (Rpr) and Hid (M.B. and P.M., unpublished data). This identified three NEDD8-specific proteases that, when knocked down, suppressed Rpr and Hid killing (Figure 1A): DEN1, CSN5, and CG1503, a predicted deneddylase. The observation that knockdown of these three deneddylases suppress Rpr and Hid killing suggests that conjugation of NEDD8 suppresses the activity of proapoptotic proteins, or, alternatively, enhances the antiapoptotic potential of cell death inhibitors. Of particular interest was Deneddylase 1 (DEN1), which reportedly removes NEDD8 from non-Cullin proteins in vivo (Chan et al., 2008). To corroborate the involvement of DEN1 in the regulation of apoptosis, we assessed genetically whether reduction of DEN1 suppressed cell death induced by IAP antagonists. Consistently, a null allele of DEN1 (DEN1EX9) (Chan et al., 2008) suppressed both Rpr- and Hid-induced cell death, establishing DEN1 as a potential proapoptotic regulator of cell death pathways (Figure 1B). The Effector Caspase drICE Is Neddylated In Vivo To decipher how NEDD8 and DEN1 might regulate apoptosis, we first examined whether proteins of the cell death machinery are conjugated with NEDD8. To test whether the Drosophila initiator caspase DRONC and the effector caspases drICE and DCP1 are subject to NEDD8 conjugation, we immunoprecipitated caspases from cellular extracts and analyzed the eluates for the presence of NEDD8 conjugates using a NEDD8-specific antibody. In agreement with endogenous drICE being targeted for neddylation, we detected several distinct bands corresponding to potentially mononeddylated (46 kDa) and polyneddylated drICE, or drICE that is conjugated with NEDD8 and Ub chains (Figure 2A). The detected bands were specific for neddylated drICE since immunoprecipitation using preimmune serum of the anti-drICE antibody, or cell extracts in which drICE was knocked down via RNAi, reduced the amount of neddylated drICE detected. To calculate the proportion of drICE that is modified, we quantified the different areas of the anti-drICE immunoblot (Figure 2A) using LI-COR Odyssey technology (Figure S1B available online). This indicated that approximately 23% of total drICE is modified. The effector caspase DCP-1 is also subject to neddylation; however, we found no evidence for robust neddylation of DRONC (Figure S2; M.B. and P.M., unpublished data). Next, we knocked down the E1/E2s for Ub and NEDD8, respectively. RNAi-mediated knockdown of the NEDD8-selective E1/E2 APPBP1/Ubc12(CG7375) almost completely abrogated neddylation of drICE (Figure 2B). Intriguingly, knockdown of the Ub-E1/E2 Uba1/UbcD1 also affected the NEDD8-specific
Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc. 811
Molecular Cell IAPs as NEDD8 E3 Ligases
A Control
CSN5
DEN1
CG1503
A’
B’
C’
D’
A’’
B’’
C’’
D’’
Figure 1. Knockdown or Mutation of Deneddylating Enzymes that Modify Rpr- and Hid-Induced Cell Death (A) dsRNA of the indicated deneddylases was expressed with the eye-specific driver GMR-GAL4, and its effect on GMR-rpr and GMR-hid-mediated eye phenotypes was analyzed by light microscopy of whole mounts. Unmodified phenotypes of Rpr and Hid are shown in A0 and A00 . Genotypes in B0 –D00 are GMR-GAL4,GMR-rpr or GMR-hid/ UAS-dsRNA. (B) Genetic validation with DEN1EX9 mutant flies. DEN1EX9 mutant flies (B and D) exhibit a reduced Rpr (compare A with B) and Hid (compare C with D) eye phenotype. The graphs depict the eye size of (GMR-GAL4,GMR-rpr,DEN1EX9/DEN1EX9, left graph) and (GMR-GAL4,GMR-hid/DEN1EX9, right graph) flies. Shown is the average eye size of at least ten animals ± standard error (SE).
Rpr
Hid
B Rpr den1EX9/den1EX9
control
control
Hid
den1EX9/+
proteins were affinity purified under denaturing conditions using nickel columns to avoid isolation of protein complexes. The presence of modified drICE was assessed by immunoblot analysis of the eluate. Under these conditions, DIAP1 readily ubiquitylated drICE, as previously reported (Ditzel et al., 2008) (Figure 3A). Interestingly, DIAP1 also neddylated 80000 30000 Rpr Hid drICE but was unable to promote the 60000 25000 conjugation of SUMO to drICE. The notion that DIAP1 functions as NEDD840000 20000 E3, but not SUMO-E3, is further corrobo20000 15000 rated by the observation that DIAP1 physically interacted with the NEDD8-E2 0 10000 den1EX9/+ control den1EX9/den1EX9 control conjugating enzyme UbcD12/CG7375 (Figure 3B) but not the SUMO-E2 lesswright/UbcD9 (data not shown). Under ‘‘smearing’’ pattern of drICE, albeit less prominently than was the same conditions, DIAP1 also interacted with the Ub-E2 achieved by knockdown of APPBP1/UbcD12, suggesting that UbcD1, as previously reported (data not shown) (Ryoo et al., endogenous drICE is both neddylated as well as ubiquity- 2002). lated—either by mixed chains or via separate chains on To determine whether endogenous DIAP1 is responsible for individual K residues. This is also consistent with a correspond- neddylating drICE, we depleted DIAP1 protein levels using ultraing change in the overall smearing pattern of total modified drICE violet (UV) treatment. Exposure to UV causes rapid proteasomal (Figure 2B). Of note, the weaker reduction in the neddylation degradation of DIAP1 (Ditzel et al., 2003). While under nontreated pattern following knockdown of Uba1/UbcD1 might be due to conditions drICE was readily neddylated, UV-mediated deplea less efficient knockdown of Uba1/UbcD1 (Figure S1A). Taken tion of DIAP1 abrogated the appearance of neddylated forms together, these data indicate that endogenous drICE is neddy- of drICE (Figure 3C). Since UV induces apoptosis under these lated in living cells and that the conjugation of NEDD8 to drICE conditions, these results also indicate that the neddylation status is catalyzed by the sequential action of the bona fide NEDD8 of drICE drastically changes during cell death. This may be achieved not only by DIAP1 depletion (removal of the NEDD8activation and conjugation cascade. E3) but also through UV-mediated activation of the deneddylase DIAP1 Can Function as NEDD8-E3 Ligase DEN1. Consistent with this view, a recent publication reports that To identify the E3 ligase that promotes drICE neddylation, we genotoxic stress induces NEDP1 (Watson et al., 2010), the first focused on DIAP1. To examine whether DIAP1 targets drICE mammalian homolog of DEN1. Therefore, drICE deneddylation for neddylation, we coexpressed drICE and DIAP1 [DIAP1(21–438)] upon cell death insult might be the result of coordinated DIAP1 (Ditzel et al., 2008) in the presence of His-tagged NEDD8. As depletion and DEN1 activation. Taken together, these data indicontrols, we also included His-Ub and His-SUMO. Conjugated cate that endogenous DIAP1 functions as an NEDD8-E3 ligase pixels
C
D
pixels
B
A
812 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
Molecular Cell
α-drICE - +
(kDa)
Blot: α-N8
neddylated drICE
150 100 75 50
Blot: α-N8
IP: α-drICE
IP: α-drICE
20 modified drICE
100 75 50 Blot: α-drICE 37
drICE
-
50
20
150 100 75 Blot: α-drICE 50 37
drICE
Lysate
1 Blot: α-drICE 37 37
+
(A) drICE was immunoprecipitated from S2 cell lysates with preimmune serum (lane 1) or a-drICE antibodies (lanes 2 and 3). Immunoprecipitates were analyzed by immunoblotting with the indicated antibodies. Bands correspond to potentially mononeddylated (46 kDa) and polyneddylated forms of drICE, or drICE that is conjugated with NEDD8 and Ub chains. RNAi-mediated knockdown of drICE also reduced the amount of neddylated drICE detected (lane 3). (B) dsRNA-mediated knockdown of the NEDD8E1 dAPPBP1 and NEDD8-E2 UbcD12 (lane 4) abrogates drICE neddylation. Similarly, knockdown of Ub-E1 Uba1 and Ub-E2 UbcD1 (lane 3) reduced the NEDD8 signal, indicating that drICE is modified with both NEDD8 and Ub. Endogenous drICE was immunoprecipitated and analyzed as in (A). The knockdown was verified by RT-PCR (Figure S1A). See also Figures S1 and S2.
25
20
Blot: α-Actin
+ -
37
37 25
-
α-drICE
neddylated drICE
-
IP: Uba1/UbcD1 dsRNA: APPBP1/UbcD12 dsRNA: (kDa) 150 100 75
Figure 2. Endogenous drICE Is Conjugated with the Ubiquitin-like Modifier NEDD8
modified drICE
IP: drICE dsRNA:
B
pre-immune
A
pre-immune
IAPs as NEDD8 E3 Ligases
drICE Actin 1
2
3
for drICE in vivo and that drICE becomes deneddylated during cell death. To compare the extent of drICE ubiquitylation versus neddylation, we made use of His-tagged Ub and NEDD8 (Figure 3D). Note that anti-Ub- and anti-NEDD8-specific antibodies cannot be used to determine the proportion of drICE that is modified with either adduct because these antibodies harbor different affinities for their respective antigens, which precludes a direct comparison. Purification of total neddylated and/or ubiquitylated drICE under denaturing conditions indicated that drICE was modified by both Ub and NEDD8, although longer and more prominent chains were formed in the presence of His-Ub (Figure 3D). The same could be observed for DIAP1 automodification (Figure 3D). The ability of DIAP1 to promote neddylation of drICE was dependent on a functional RING finger, as the E3 mutants DIAP1(21–438/C406Y) and DIAP1(21–438/CD6) (Ditzel et al., 2008) failed to neddylate drICE (Figure 4A). Previous work has indicated that DIAP1 also requires (1) N-terminal cleavage and (2) binding to UBR domain-bearing N-end-rule E3s (UBR-E3s) to maximally ubiquitylate drICE (Ditzel et al., 2008). Therefore, we established the determinants for DIAP1-mediated neddylation of drICE. As shown in Figure 4A, N-terminal cleavage was necessary for DIAP1 to act as a NEDD8-E3 for drICE. DIAP1(D20A), which carries a point mutation in the caspase cleavage site and hence resides in its full-length form, failed to neddylate drICE efficiently. This is most likely due to the observation that cleaved DIAP1(21–438) interacts with drICE far better than full-length DIAP1, which binds caspases only weakly (Ditzel et al., 2008). While N-terminal cleavage was required, binding to UBR-E3s was dispensable for neddylation of drICE. M-DIAP1(21–438), in
2
3
4
which N(21) was replaced by methionine (M), an amino acid that does not allow UBR binding (Tasaki et al., 2005), was as efficient in neddylating drICE as wildtype DIAP1(21–438). Therefore, DIAP1mediated neddylation of drICE seems to occur independently of the N-end-rule machinery. We also tested whether neddylation of drICE required direct binding of drICE to DIAP1. Accordingly, ALG-drICE, which exposes an IAP-binding motif (IBM) at its neo-N terminus and binds to DIAP1 (Tenev et al., 2005), was neddylated by DIAP1(21–438). In contrast, LG-drICE, which lacks a functional IBM and fails to associate with DIAP1 (Tenev et al., 2005), also failed to be neddylated by DIAP1(21–438) (Figure 4B). This indicates that a physical interaction between drICE and DIAP1 is required for DIAP1-mediated neddylation of drICE. DIAP1 not only promoted neddylation of drICE, but also stimulated the conjugation of NEDD8 to itself (autoneddylation). This is evident as the RING mutants DIAP1(21–438/C406Y) and DIAP1(21–438/CD6) failed to become neddylated (Figure 4A), indicating that DIAP1’s own RING finger is required for the conjugation of NEDD8 to itself. Moreover, cleavage at position D20 of DIAP1, which is known to enhance its E3 ligase activity (Ditzel et al., 2008), seemed to boost DIAP1’s ability to autoneddylate. Accordingly, cleaved DIAP1 (N-DIAP121–438) was significantly more effective in autoneddylation than the noncleavable DIAP1(D20A) mutant (Figure 4A). Given that neddylation of cullin-type E3 complexes activates their ligase activity (Duda et al., 2008; Furukawa et al., 2000; Morimoto et al., 2000; Podust et al., 2000; Read et al., 2000; Saha and Deshaies, 2008; Wu et al., 2000), we next examined whether neddylation of DIAP1 similarly stimulates its E3 ligase activity. To test this unambiguously, we devised an in vitro assay (Figure 4C) in which we first incubated DIAP1 with a conjugation mixture in the presence or absence of NEDD8 (first step, top panel, lanes A and B) or Ub (lane C). Unmodified, neddylated, and ubiquitylated forms of DIAP1 were purified and
Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc. 813
Molecular Cell IAPs as NEDD8 E3 Ligases
drICE N8
-
-
+
V5-tagged : UbcD12
+
GST-tagged:
50 Blot: α-V5 37
(kDa)
45 Blot: α-V5 30
*
25 20
4
5
6
GST
Blot: α-V5 30
UbcD12 1
D
2
drICE-V5 DIAP1
α-drICE
His-N8: His-Ub:
- - +
(kDa)
250 150 Blot: α-N8 100 75 50
(kDa)
37 Blot: α-drICE
drICE
37 Blot: α-drICE
processed drICE
Blot: α-DIAP1
Lysate
drICE
Blot: α-V5
- + - + - + + -
150 102 76 52
modified drICE
IP: UV:
3
DHFR-HA-Ub (DIAP1 reference protein)
neddylated drICE
C
IP: α-drICE
Blot: α-GST
38 31 24 17 150 102 Blot: α-DIAP1 76 52
modified DIAP1
2
pre-immune
1
GST-DIAP1
30 Lysate
Blot: α-HA
UbcD12
97 66 45
drICE
His-purification
Lysate
Blot: α-V5 37
DIAP1
SUMO
+
modified drICE
His-purification
-
control
Ub
His-tagged : DIAP121-438 : (kDa) 100 75
Lysate
Figure 3. DIAP1 Functions as NEDD8-E3 Ligase for drICE
B
V5-tagged :
GST-purification
A
Blot: α-V5
drICE
38 Blot: α-DIAP1 52
(A) drICE is ubiquitylated and neddylated in a DIAP1-dependent manner. However, DIAP1 does not SUMOylate drICE. drICE-V5 was expressed in S2 cells in the presence or absence of DIAP1(21–438), and His-Ub, -SUMO, or -NEDD8, respectively. Note that DIAP1 was expressed as HA-DHFR/Ub-DIAP1 fusion in which the reference protein DHFR/Ub is cotranslationally cleaved off (Varshavsky, 2000). Expression of the reference protein HA-DHFR/Ub indirectly indicates the expression level of the protein of interest (DIAP1). Purification of His-tagged proteins was performed under denaturing conditions, and the presence of modified drICE was detected with a-V5 antibody. An asterisk marks unmodified drICE, which is detected in all lanes as a result of nonspecific drICE:matrix interaction. (B) DIAP1 binds to the NEDD8-E2 UbcD12. UbcD12-V5 was expressed together with GST (lane 1) or GST-DIAP1 (lane 2) in S2 cells. GSTand DIAP1-bound protein complexes were purified and analyzed by immunoblotting. (C) Endogenous drICE is neddylated in a DIAP1dependent manner. S2 cells were either left untreated or exposed to UV. The presence of neddylated forms of drICE was determined by immunoblotting. Note that UV treatment causes depletion of DIAP1 protein levels (bottom panel, compare lanes 1 and 2 with lane 3) (Ditzel et al., 2008). The top panel is a reblot with an a-NEDD8 antibody of the experiment shown in Ditzel et al. (2008), while the input controls (bottom panels) are the same as in Ditzel et al. (2008). (D) Comparison between drICE neddylation and ubiquitylation. drICE was coexpressed with the indicated constructs and analyzed as in A.
DIAP1 1 2 3
4
DIAP1 1
2
3
subsequently assayed for their ability to promote ubiquitylation and neddylation of drICE, respectively (second step). Incubation of DIAP1 with drICE in the presence of E1, E2, and NEDD8 leads to the appearance of several slow migrating species of drICE (bottom panel, lanes 4 and 5), demonstrating that DIAP1 is able to act as NEDD8-E3 ligase for drICE in vitro. DIAP1 appeared to preferentially modify drICE with mono- or polymono-NEDD8 conjugates. In contrast, DIAP1 promoted the conjugation of poly-Ub chain to drICE. Comparison between unmodified and neddylated DIAP1 indicates that neddylated DIAP1 is no more active than unmodified DIAP1 in ubiquitylating drICE (4C). Likewise, ubiquitylation of DIAP1 did not significantly affect the ability of DIAP1 to neddylate drICE under these conditions. This demonstrates that neddylation or ubiquitylation of DIAP1 itself does not modulate its E3 ligase activity. DEN1 but Not CSN5 Removes NEDD8 from drICE Since the deneddylases DEN1 and CSN5 were both identified as suppressors of IAP antagonist-induced eye phenotypes
(Figure 1), we examined whether DEN1 and CSN5 removed NEDD8 conjugates from drICE. Expression of DEN1 efficiently removed NEDD8 conjugates from drICE (Figure 5A). In contrast, CSN5 failed to cleave NEDD8 from drICE, despite being expressed to similar levels. DEN1-mediated deconjugation of NEDD8 required a functional protease domain since the catalytically inactive mutant DEN1(C165A) failed to remove NEDD8 from drICE (Figure 5B). Of note, DEN1-mediated deconjugation of NEDD8 is unlikely to be due to an effect of DEN1 on DIAP1’s E3 ligase activity, since DIAP1-mediated ubiquitylation of drICE still occurred under these conditions (data not shown). This indicates that DEN1, but not CSN5, acts as deneddylase for drICE. Conjugation of NEDD8 Inhibits Active drICE To test whether neddylation directly affects the proteolytic activity of drICE, we devised an in vitro neddylation assay followed by a cleavage reaction using caspase substrates (Figure 6). Several slow-migrating drICE species were detected when recombinant, active drICE was incubated with E1, E2, DIAP1, and increasing amounts of NEDD8 (Figure 6A), which
814 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
Molecular Cell IAPs as NEDD8 E3 Ligases
step 1:
+
(DIAP1 auto-Ub/N8ylation)
LG
+ ALG
ALG
21-438
M
V5-tagged drICE :
-
(kDa)
* 25 20
Blot: α-V5
Blot: α-V5 37 36 Blot: α-HA 28
250 130 95 72 Blot: α-DIAP1 55
*
37
N8 Ub
(kDa)
neddylated drICE
His-purification
Blot: α-V5
100 75 50
Lysate
neddylated drICE
100 75 50 37
DIAP1
25 36
1
2
3
28
DHFR-HA-Ub (DIAP1 reference protein)
A
B
C
step 2: (drICE Ub/neddylation)
)
100 75 50 37
drICE
-m un od -m . ( ne od A) dd . (A un yl ) -m ate ub od d ( iq . (A B) ui ) ty la te d (C
neddylated DIAP1
25 20
DIAP1: step2: -
un
His-purification His-purification
N8
His-tagged : N-DIAP121-438: -
(kDa)
Blot: α-DIAP1
C
B N21-438/CΔ6
21-438
N
-
DIAP1 :
N21-438/C406Y
N8 drICE
His-tagged : V5-tagged :
D20A
A
Ub Ub N8 N8
Lysate
37
drICE
25 Blot: α-V5 20
* *
modified drICE
(kDa)
250 130 95 72 55 Blot: α-drICE 36
15
drICE p20 17
Blot: 50 α-DIAP1
N8-drICE
28
drICE p10
1
2
3
4
5
DIAP1 1
2
3
4
5
6
Figure 4. DIAP1 Neddylates drICE in a RING- and Binding-Dependent Manner (A) DIAP1 requires N-terminal cleavage and a functional RING finger for its ability to conjugate NEDD8 to drICE and itself. drICE was coexpressed with a noncleavable DIAP1 mutant (D20A, lane 2), ‘‘active’’ DIAP(21–438) (lane 3) starting with N(21) (UBR-E3-binding proficient) (Ditzel et al., 2008), RING mutant DIAP(21–438/C406Y) (lane 4), DIAP(21–438/CD6) (lane 5), or cleaved DIAP1(21–438) (lane 6) that carries an M instead of N and therefore is defective in its ability to recruit UBR-E3s. Purification of His-tagged neddylated proteins was performed under denaturing conditions, and the presence of neddylated drICE and DIAP1 was detected by immunoblotting. (B) Physical interaction between DIAP1 and drICE is required for drICE neddylation. ALG-drICE but not LG-drICE, which lacks an IBM and is impaired in DIAP1 binding, is neddylated by active DIAP(21–438). The experiment was carried out as in (A). Note that DIAP1 was expressed as HA-DHFR/Ub-DIAP1 fusion. (A and B) An asterisk marks unmodified drICE which is detected in all lanes due to nonspecific drICE:matrix interaction. (C) Sequential in vitro neddylation and ubiquitylation assay. DIAP1 autoneddylation is not required for DIAP1’s ability to function as an Ub-E3. Step 1 (top): autoneddylation (lane B) or autoubiquitylation (lane C) of DIAP1. Step 2 (bottom): nonmodified DIAP1 (from A) or neddylated DIAP1 (from B) was used to ubiquitylate drICE (lanes 2 and 3). Vice versa, nonmodified DIAP1 (from A) or ubiquitylated DIAP1 (from C) was used to neddylate drICE (lanes 4 and 5). A purification step was performed after step 1 to remove any free NEDD8 and Ub from the reaction mix. Asterisks indicate nonspecific background bands from the recombinant drICE preparation.
were not detected in the presence of the E3 mutant DIAP1(F437A), indicating that they are neddylated products. To assess the effect of neddylation on the activity of drICE, we incubated in vitro neddylated and/or ubiquitylated forms of drICE with recombinant PARP1 protein or DEVD-AMC as caspase substrates (Figures 6A–6C and Figure S3). The proteolytic activity of modified drICE was compared with the one of nonmodified drICE that was incubated with the DIAP1 RING finger mutant DIAP1(F437A). drICE protein that had been modified with increasing amounts of NEDD8 was strongly impaired in its ability to cleave PARP1 (Figure 6A). NEDD8-mediated inhibition of drICE required the conjugation of NEDD8 to drICE and was not the result of free NEDD8 poisoning the caspase cleavage assay nonspecifically. This is evident because increasing amounts of
free NEDD8, in the presence of the RING mutant DIAP1(F437A), did not reduce the catalytic activity of drICE (Figure 6A). Quantification of PARP1 cleavage showed that while unmodified drICE cleaved 36% of PARP1, neddylated drICE was significantly less active in processing PARP1 (Figure 6B). Only 6% of PARP1 was cleaved in the mixture that carries the highest levels of neddylated drICE (Figure 6B, compare columns 1 and 3; Figure S3). Ubiquitylation of drICE similarly impaired its catalytic activity (Figure 6B; Figure S3) (Ditzel et al., 2008), but not as efficiently as neddylation (Figure 6B, compare lanes 2 and 4 and lanes 3 and 6). The observation that DIAP1 suppresses the catalytic activity of drICE in a Ub- and NEDD8-dependent manner is further supported by the finding that coexpression of DIAP1 and drICE prevents appearance of the p10 subunit of
Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc. 815
Molecular Cell IAPs as NEDD8 E3 Ligases
B
His-N8 drICE-V5
50
*
37
-
(kDa)
His-purification
150 100 75 Blot: α-V5
HA-DEN1: neddylated drICE
His-purification
(kDa)
neddylated drICE
- + + -
WT
-
(A) DEN1 acts as deneddylase for drICE. drICE-V5 was coexpressed with DIAP1(21-438) and HisNEDD8 in the absence (lane 1) or presence of HA-DEN1 (lane 2) or HA-CSN5 (lane 3). Coexpression of DEN1 but not CSN5 abolishes neddylation of drICE. (B) The catalytic mutant of DEN1 [DEN1(C165A)] fails to deneddylate. An asterisk marks unmodified drICE, which is detected in all lanes as a result of nonspecific drICE:matrix interaction.
DIAP1
DIAP1 HA-CSN5: HA-DEN1:
Figure 5. DEN1 Reverses DIAP1-Mediated Neddylation of drICE
His-N8 drICE-V5 C165A
A
150 100 75 Blot: α-V5 50
*
37
25
Blot: α-HA
Lysate
150 100 75 50
Lysate
HA-CSN5 37
Blot: α-V5 37 37 Blot: α-HA 25 1
2
3
HA-DEN1 25 20 Blot: α-DIAP1
DIAP1
50 1
2
3
drICE in vivo (Figure 4A, lane 3). It is important to note that cleavage of drICE in this system requires an input from active drICE itself (Ditzel et al., 2008). Accordingly, z-VAD-fmk also suppresses the appearance of the small subunit of drICE (Ditzel et al., 2008). Therefore, after DIAP1-mediated modification of drICE, drICE’s catalytic potential is inhibited, and hence no further drICE processing is possible. Concentration-dependent cleavage assays with DEVD-AMC further corroborate the notion that nonneddylated and neddylated drICE significantly differ in their processivity (Figure 6C). Since DIAP1 and DIAP1(F437A) bind to drICE equally well, but DIAP1(F437A) fails to neddylate drICE, this result indicates that neddylation reduces the catalytic potential of drICE. Nonlinear regression using the Michaelis-Menten equation showed a reduction in Vmax (62195 ± 6845 RFU/min for nonmodified drICE to 40669 ± 4536 RFU/min for neddylated drICE) but not KM. This suggests that conjugation of NEDD8 functions as noncompetitive inhibitor that suppresses caspase activity via a conformational change of the caspase, reducing its catalytic processivity. Importantly, the mechanism through which NEDD8 suppresses caspase activity is distinct from the one of Ub, which affects both Vmax and KM of drICE and, therefore, acts as ‘‘mixed’’ inhibitor (Ditzel et al., 2008). Taken together, these results suggest that drICE is regulated by the conjugation of both NEDD8 and Ub, which together cooperate to lower the activity of drICE. Note that immunoprecipitated drICE is modified by both NEDD8 and Ub chains (Figure 2B). Structural prediction indicates that drICE carries nine surface exposed lysine (K) residues (Ditzel et al., 2008) (Figure S4). Using mass spectrometric analysis of neddylated drICE, we identified that K142 (Figure S4) was neddylated. As a result of limited coverage, no data were obtained on the neddylation status of other surface exposed K residues. Although K142 can function
HA-DEN1 HA-DHFR-Ub (DIAP1 reference protein)
as acceptor K for NEDD8, it seems not to be the only K being neddylated, or else other Ks are used when K142 is mutated. This is evident because mutation of K142 did not abrogate DIAP1-mediated neddylation of drICE (Figure S4, compare lanes 4 and 5), and mutation of all nine surface exposed K residues in the p20 [drICE(9K > R)] was required to fully abrogate drICE neddylation (Figure S4). This is reminiscent to the conjugation of Ub, where also all nine surface-exposed residues of drICE needed to be mutated to abrogate its modification by Ub. Therefore, drICE mutants that retain a single K at various positions are ubiquitylated and neddylated as efficiently as WT drICE (data not shown) (Ditzel et al., 2008). This indicates that any of the nine K residues can serve as acceptor site for Ub and NEDD8, and that conjugation of these UBLs to drICE does not occur on a specific K residue at a fixed position, a phenomena that is frequently observed (also seen in p53, IkBa, c-jun, and cyclinB1), and contributes to the tremendous plasticity and flexibility of the Ub/UBL-system (Kirkpatrick et al., 2006; Rodriguez et al., 1996, 2000; Scherer et al., 1995; Treier et al., 1994; Xirodimas et al., 2004). Structural studies show that K residues positioned less than 50 A˚ away from the active site cysteine of the E2 can serve as acceptor sites (Duda et al., 2008). In this regard, it is interesting to note that the majority of the surface exposed K residues of drICE are positioned in a ring-like orientation (Figure S4D). Since they all can serve as acceptor Ks, they all must be in close proximity to the E2s, making it difficult to generate Ub- and NEDD8-selective drICE mutants. IAP-Mediated Neddylation Is Evolutionarily Conserved Next, we wished to establish whether mammalian IAPs can also function as NEDD8-E3 ligases. To this end, we studied mammalian XIAP, which has previously been shown to function as Ub-E3 ligase for itself and caspases (Morizane et al., 2005; Schile et al., 2008; Suzuki et al., 2001). XIAP readily promoted autoneddylation and neddylation of caspase-7 (Figure 7A). Similar to DIAP1, XIAP-mediated neddylation was RING dependent since XIAP(DC8), which lacks the last eight amino acids essential for its E3 activity (Silke et al., 2005), failed to promote conjugation of NEDD8 to itself and caspase-7. Further, Smac
816 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
Molecular Cell IAPs as NEDD8 E3 Ligases
A
1. Step:
DIAP1: WT N8 (μg): 0 1 2 4
1. Step:
in vitro Ubylation and N8ylation assay
Figure 6. Neddylation of drICE Suppresses Its Catalytic Potential through Noncompetitive Inhibition
in vitro neddylation assay F>A 0
1 2
4
(kDa)
modified drICE
130 95 72 55
DIAP1 drICE
modified drICE
36 28
drICE p20 (unmodified)
Blot: α-drICE 17 250 130
modified DIAP1
95 Blot: α-DIAP1
72 55
36 28 1
2
3
4
5 6
7
8
2. Step: +
2. Step:
PARP1
PARP cleavage assay
full length PARP1 cleaved PARP1
Blot: α-PARP1
drICE
1 2 PARP1
B
4
5
6
7
8
40
C
DEVDase assay non-modified drICE
Odyssey protein quantification (LI-COR)
50 % PARP processed
3
cleaved PARP1
1. step: DIAP1+drICE+Ubl 2.step: +PARP
N8ylated drICE
30 20 10
Ub
Ub
2
4
We also examined whether cIAP1 is capable of promoting autoneddylation and the conjugation of NEDD8 to RIP1, a bona fide target of cIAP1-mediated ubiquitylation (Broemer and Meier, 2009). cIAP1 readily targeted itself and RIP1 for neddylation (Figure 7D). While the physiological role of XIAP and cIAP1-mediated neddylation remains to be explored, our data clearly indicate that IAPs not only function as E3 ligases of the Ub conjugation system, but also are capable of promoting the conjugation of NEDD8 to target substrates, thereby extending the complexity of IAP-mediated signaling.
g
g
Ub
N8 g
g
DISCUSSION
2
g
N8
+
2
g 2
4
Ub
N8
l
0 no
(A) In vitro neddylation of recombinant drICE suppresses its catalytic ability to cleave PARP1. A schematic representation of the assay procedure is shown in the leftmost panels. In vitro neddylation assay of active drICE with DIAP1 or DIAP1(F437A) and the indicated amount of NEDD8. Modification of drICE (upper panel) and automodification of DIAP1 (middle panel) was assessed by immunoblot analysis. Bottom panel, 2 step depicting in vitro PARP1 cleavage assay: the neddylation reactions from step 1 were incubated with recombinant PARP1. Shown is immunoblot analysis with a-PARP1 antibodies. (B) In vitro ubiquitylation/neddylation was performed as in (A). The percentage of PARP1 cleavage was determined by immunoblotting and quantification with LI-COR-Odyssey. The mean ± SE of three independent experiments is shown. Values and immunoblot analysis of a representative experiment are shown in Figure S3. (C) Neddylation of drICE affects kinetic parameters of substrate cleavage. In vitro neddylation assay of active drICE with DIAP1 and DIAP1(F437A), respectively. Reactions were subsequently incubated with increasing concentrations of the caspase substrate DEVD-AMC. Curves were fitted with nonlinear regression with the Michaelis-Menten equation. Shown is the mean of three independent experiments ± SE. See also Figure S3.
mimetic treatment, which targets XIAP as well as other IAPs (Gaither et al., 2007), resulted in a significant reduction of caspase-7 neddylation (Figure 7B). Since XIAP strongly binds and inactivates caspase-7 and is the only physiological caspase inhibitor in mammals (Eckelman et al., 2006), it is likely that neddylation of caspase-7 is mediated by XIAP. However, since Smac mimetics also target other IAPs, such as cIAP1 and cIAP2, it is formally possible that these IAPs also contribute to neddylation of caspase-7. Consistent with the notion that neddylation suppresses caspase activity, treatment with the compound MLN4924, an inhibitor of the NEDD8 conjugation pathway (Soucy et al., 2009), resulted in increased caspase-3 and -7 dependent DEVDase activity (Figure 7C). Taken together, these results suggest that also mammalian IAPs act as NEDD8-E3 ligases for effector caspases.
Using an unbiased modifier screen, we identified three NEDD8-specific isopeptidases that, when knocked down, suppress Rpr- and Hid-induced cell death. Genetic validation confirmed the involvement of DEN1, and therefore NEDD8, in regulating apoptosis. Importantly, NEDD8 maturation still occurs in DEN1 mutant animals (Chan et al., 2008), indicating that DEN1 is not the only NEDD8-processing enzyme and that in the absence of DEN1, the conjugation of NEDD8 to target substrate occurs normally. This is in contrast to APPBP1 mutants that lack the NEDD8-E1 enzyme and in which no neddylation is seen (Chan et al., 2008). Consistent with the notion that conjugation of NEDD8 protects from apoptosis, we found that endogenous drICE is neddylated in healthy cells and that the conjugation of NEDD8 directly reduces the proteolytic activity of drICE. Neddylated drICE is less active toward caspase substrates than nonmodified drICE. NEDD8-mediated suppression of drICE occurs via a mechanism
Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc. 817
Molecular Cell IAPs as NEDD8 E3 Ligases
A
B His-tagged :
-
Smac mimetic:
-
ΔN-Casp-7 HA-XIAP :
-
Figure 7. Evolutionarily Conservation of IAP-Mediated Conjugation of NEDD8
ΔN-Casp-7
N8
His-tagged :
wt ΔC8
N8
-
+
(kDa)
neddylated Casp-7
His-purification
37
100 75 His-purification
50 Blot: α−Casp-7
(kDa)
neddylated Casp-7
100 75
50 Blot: α-Casp-7 37 25
25
100 75
Blot: α−HA
50
37 Lysate
neddylated XIAP
20 150
Blot: α-Casp-7 Blot: α-XIAP
Casp-7
25 50
XIAP 1
37
2
3
Blot: α−HA
XIAP 50
cIAP1 : 1
2
3
(kDa)
RFU
6x105
DEVDase assay
His-purification
Blot: α−RIP1
C
4x105
-
250 150 100 75
WT
50 37
150 100 Blot: α−cIAP1 75 50 37
0
- +
- +
- +
Lysate
2x105
MLN4924:
N8 RIP1
His-tagged :
D
neddylated RIP1
ΔN-Casp-7
neddylated cIAP1
Lysate
end. Casp-7 Blot: α−Casp-7 25
(A) The mammalian IAP XIAP targets caspase-7 and itself for neddylation in a RING finger-dependent manner. DN-caspase-7 was coexpressed with His-NEDD8 in 293T cells in the absence (lane 1) or presence of HA-XIAP(wt) (lane 2) or the RING mutant HA-XIAP(DC8) (lane 3). The experiment was performed as described in Figure 3A. (B) Inhibition of endogenous IAPs impairs caspase-7 neddylation. DN-caspase-7 was coexpressed with His-NEDD8 in 293T cells and incubated for 6 hr in the presence or absence of 1 mM Smac-mimetic (Gaither et al., 2007) before lysis. Neddylation of caspase-7 was determined by immunodetection with caspase-7 antibodies after His-purification under denaturing conditions (as above). Note: neddylation of caspase-7 is, in contrast to (A), also visible in the absence of XIAP overexpression as a result of stronger exposure of the immunoblot. (C) Inhibiting the NEDD8 conjugation pathway by the NEDD8-E1 inhibitor MLN4924 (Soucy et al., 2009) increases caspase activity. 293T cells were transfected with active caspase-3 or caspase-7 or control vector and treated with MLN4924 for 16 hr. Caspase activity in cell lysates was determined by cleavage of DEVD-AMC. The experiment was performed in triplicates and mean ± standard deviation is shown. (D) cIAP1 can target RIP1 and itself for neddylation. RIP1 and cIAP1 were expressed together with His-tagged N8 in 293T cells and the experiment performed as in (A).
Blot: α−RIP1 75 75 Blot: α−cIAP1
control Casp-3 Casp-7
that relies on noncompetitive inhibition, most likely through a NEDD8-induced conformational change of the caspase. Similar to the effect of NEDD8 on drICE, allosteric regulation is also seen in caspase-1 (Scheer et al., 2006), indicating that caspases can be regulated through modification of regulatory residues outside the catalytic pocket. Although conjugation of Ub and/or NEDD8 reduces the catalytic activity of drICE from 100% (unmodified drICE) to 16% (modified drICE), it is important to note that neddylation or ubiquitylation of drICE does not completely inhibit the caspase in vitro. Moreover, combined modification of NEDD8 and Ub also does not further inhibit drICE under these conditions. However, it is likely that in vivo additional mechanisms are in place to limit drICE-mediated proteolysis. A common mode of regulation involves the recognition of the Ub and NEDD8-adducts by proteins with Ub- or NEDD8-binding domains (UBDs) (Haglund and Dikic, 2005). Therefore, Ub- and NEDD8-binding proteins may contribute to the regulation of modified caspases in vivo, perhaps by spatial sequestration.
Our data indicate that DIAP1 can promote the covalent attachment of NEDD8 RIP1 to the effector caspase drICE and cIAP1 DCP-1. This is surprising, since DIAP1 1 2 reportedly functions as an Ub-specific E3 (Vaux and Silke, 2005). Although NEDD8 is homologous to ubiquitin (57% identity), it has unique E1 and E2 enzymes that specifically handle NEDD8 (Lee et al., 2008; Souphron et al., 2008; Whitby et al., 1998). The NEDD8 E1 is a heterodimer composed of the amyloid precursor protein-binding protein (APPBP1) and Uba3, while the NEDD8-E2 is Ubc12 (Gong and Yeh, 1999; Lammer et al., 1998; Liakopoulos et al., 1998; Osaka et al., 1998; Pozo et al., 1998). Moreover, until recently, the only known substrates for NEDD8 modification were the six members of the cullin family of proteins that are components of SCF (SKP-Cullin-F-Box) Ub-E3 ligase complexes. Lately, three other substrates of neddylation have been discovered (Oved et al., 2006; Xirodimas et al., 2004; Xirodimas et al., 2008). This study now identifies IAPs as versatile E3 ligases that bind to E2s of the Ub and NEDD8 conjugation pathways and can stimulate the conjugation of Ub and NEDD8 to target substrates. Several lines of evidence indicate that IAPs can function as NEDD8-E3 in the NEDD8 conjugation cascade. First, in vivo DIAP1 readily promotes the covalent attachment of NEDD8 to
818 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
Molecular Cell IAPs as NEDD8 E3 Ligases
itself and drICE. This is mediated by DIAP1 itself because neddylation of DIAP1 and drICE is dependent on DIAP1s RING finger domain. The RING mutants DIAP1(21–438/C406Y) and DIAP1(21–438/CD6) (Ditzel et al., 2008) fail to promote the conjugation of NEDD8 to DIAP1 and drICE. Likewise, in mammals XIAP is also capable of neddylating caspase-7 in a RING finger-dependent manner, indicating that IAP-mediated conjugation of NEDD8 to target substrates is evolutionarily conserved. Second, DIAP1 physically interacts with the NEDD8-E2 conjugating enzyme UbcD12. Under the same conditions, DIAP1 did not bind to the SUMO conjugating enzyme lesswright/UbcD9 and promoted neither SUMOylation of itself or target substrates, indicating that DIAP1 selectively interacts with the Ub and NEDD8 conjugation cascades. Third, DIAP1-mediated neddylation is diminished in cells in which components of the NEDD8 activation and conjugation cascade are knocked down by RNAi. This indicates that DIAP1 not only physically interacts with UbcD12 but also uses UbcD12 for the conjugation of NEDD8. Finally, neddylation of drICE appears to be DIAP1dependent since depletion of DIAP1 reduces the amount of neddylated forms of drICE. Therefore, DIAP1 seems to function as a versatile E3 ligase that is capable of interacting with Ub-specific E2s, such as UbcD1 (Ryoo et al., 2002) and Ubc13/Uev1a (Herman-Bachinsky et al., 2007) (M.B. and P.M., unpublished data), and the NEDD8-selective E2 UbcD12/ CG7375. Consistent with the notion that DIAP1 can handle both these conjugation systems, we find that endogenous drICE is modified with both Ub and NEDD8. Accordingly, knockdown of the respective E1 and E2 enzymes results in a corresponding reduction in the overall smearing pattern of total modified drICE. In healthy cells, only a relatively small fraction of total drICE is modified. This is not entirely surprising because healthy cells only harbor a small, yet detectable, amount of processed caspases. Importantly, caspase cleavage is a prerequisite for DIAP1 binding and DIAP1-mediated ubiquitylation and neddylation. After their proteolytic cleavage, effector caspases expose an evolutionarily conserved IAP-binding motif that is required for their recruitment to DIAP1 (Zachariou et al., 2003; Ditzel et al., 2008). Hence, DIAP1 does not bind or modify full-length drICE. Consistent with the view that deneddylation is required to execute apoptosis, we find that DEN1 efficiently removes NEDD8 conjugates from drICE. Since NEDD8 suppresses the catalytic processivity of drICE, DEN1 seems to reverse NEDD8-mediated inhibition of drICE. Consistently, the neddylation status of drICE drastically changes upon UV-induced cell death. While the loss of NEDD8 modifications on drICE may be achieved by UV-mediated depletion of DIAP1 alone, it is highly likely that UV also activates the deneddylase DEN1. In agreement with this view, a recent publication reports that genotoxic stress induces NEDP1 (Watson et al., 2010), the mammalian homolog of DEN1. Therefore, drICE deneddylation might be the result of coordinated action of DIAP1 depletion and DEN1 activation. While DEN1 is highly effective in removing NEDD8 from drICE, CSN5 failed to deneddylate drICE efficiently. CSN5 acts as part of the COP9 signalosome that regulates the activity of SCF E3 ligases (Cope et al., 2002; Lyapina et al., 2001; Schwechheimer et al., 2001). Therefore, it is likely that CSN5
regulates cell death indirectly via modulating the activity of SCF E3 ligases. It is interesting to note that the F-box protein Morgue and SkpA, which are likely to form a SCF complex, have previously been reported to regulate apoptosis via modulating DIAP1 levels (Hays et al., 2002; Wing et al., 2002), leaving open the possibility that CSN5 might mediate its proapoptotic effect through Morgue/SkpA-SCF. Taken together, our data identify Drosophila and mammalian IAPs as versatile E3 ligases that promote the conjugation of Ub and NEDD8 to target substrates. Since IAPs act as multifunctional signaling devices that also exert caspase-independent functions, it is likely that IAPs modulate such processes via conjugation of both Ub and NEDD8. Clearly, both modifications occur in vivo and appear to be required for proper regulation of cell death. Our data indicate that neddylation and ubiquitylation affect effector caspases through different mechanisms. While attachment of Ub acts as a mixed type inhibitor for caspase activity, acting through a competitive and a noncompetitive mode, neddylation inhibits caspases in a noncompetitive manner, most likely through conformational changes. Differential neddylation/deneddylation and/or ubiquitylation/deubiquitylation may well be used to fine-tune caspase activity to allow them to take part in nonapoptotic signaling processes. Furthermore, Ub- and NEDD8-specific binding proteins (Di Fiore et al., 2003) may also contribute to the suppression of modified caspases. Ultimately, a deeper understanding of how conjugation and deconjugation of Ub and UBLs determine cellular phenotypes will be key since aberrant Ub/UBL and apoptosis signaling sits at the heart of many human pathologies and may have significant impact on the development of new therapeutic strategies. EXPERIMENTAL PROCEDURES Fly Work GMR-GAL4,GMR-rpr, GMR-GAL4,GMR-hid, and DEN1EX9 flies, lacking the complete coding region of DEN1, were previously described (Chan et al., 2008; Leulier et al., 2006). UAS-DUB-dsRNA (RNAi) lines were obtained from the Vienna Drosophila RNAi Center (VDRC) and the National Institute of Genetics (NIG-fly), Japan. All crosses were performed at 25 C. Eye sizes were captured with the ‘‘Quick Selection Tool’’ of Adobe Photoshop, and the numbers of pixels were measured. Constructs, Antibodies, and Recombinant Proteins Constructs were cloned into pAc, pMT, pMT-GTC (Zachariou et al., 2003), or pcDNA3 (Invitrogen) and verified by sequencing. DHFR-HA-Ub-based diap1 constructs, full-length DRONC-V5/FLAG, and Ub-ALG-drICE(WT)V5/FLAG constructs were described previously (Ditzel et al., 2003, 2008; Tenev et al., 2007). Recombinant proteins were produced as described previously (Ditzel et al., 2008). Antibodies were as follows: a-V5 (AbD Serotec), a-GST (GE Healthcare), a-HA (Roche), a-PARP1 (F-2, Santa Cruz), a-Actin (C-11, Santa Cruz), a-NEDD8 (Alexis, 210-194 and Cell Signaling, #2745), a-Caspase-7 (BD PharMingen), a-cIAP1 (Alexis), and a-RIP1 (BD PharMingen). a-DIAP1 and a-drICE antibodies were described previously (Zachariou et al., 2003). RNAi in S2 Cells Production of double-stranded DNA (dsRNA) and transfection were performed as described previously (Wilson et al., 2002). The following coding regions were used to produce dsRNA: APPBP1, nt 94-760; UbcD12 (CG7375), nt 15–525; Uba1, nt 40–546; UbcD1, nt 297–808; and drICE, nt 84–763.
Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc. 819
Molecular Cell IAPs as NEDD8 E3 Ligases
Immunoprecipitations, Ubiquitylation, and Neddylation Assays Coimmunoprecipitation and immunoblot assays were performed as described previously (Tenev et al., 2005). Endogenous drICE was immunoprecipitated with guinea pig-derived a-drICE antibody that was crosslinked to protein A agarose beads (Affi-Gel, Biorad). For detection of neddylated drICE, 50 mM Chloroacetamide was added to the lysis buffer. Where indicated, S2 cells were irradiated with 150K mJ UV (UV Stratalinker, Stratagene) and lysed 4 hr after treatment. For the ubiquitylation and neddylation assays, cells were cotransfected with the indicated constructs and harvested as described previously (Ditzel et al., 2008; Rodriguez et al., 1999). For in vitro neddylation, PARP1 cleavage, and DEVDase, 4 mg recombinant drICE was incubated in the presence or absence of 1 mg DIAP1, 1 mg E2 (UbcD1), 80 ng E1, and 2 mg Ub or 4 mg NEDD8 (or as indicated) in a final volume of 30 ml reaction buffer. Reactions were carried out at 37 C for 90 min. Note that in vitro neddylation was carried out in the presence of Ub-E1 and Ub-E2, which have been shown to handle NEDD8 in vitro (Whitby et al., 1998). PARP1 cleavage and DEVDase assays were carried out as previously described (Ditzel et al., 2008). For the two-step in vitro neddylation/ubiquitylation assay, DIAP1 was incubated with E1, E2, and either Ub or N8 for 1 hr. Unincorporated Ub and NEDD8 was removed with Amicon Centrifugal Filter Units (30K cutoff, Millipore). For the second step, new reaction mixture, E1, E2, recombinant drICE, and either Ub or N8 were added and incubated for 90 min at 37C. Cell-Based Caspase Activity Assay 293T cells were transfected with pcDNA3-based C-terminally, V5-tagged caspase-3 and -7 constructs, or control vector. Twenty-four hours after transfection, cells were incubated with 5 mM MLN4924 (Millennium Pharmaceuticals) for 16 hr or left untreated. Cells were lysed, and 20 ml supernatant was added to 350 ml DEVDase assay mixture (Ditzel et al., 2008). The reaction was incubated at RT and analyzed at 380 nM excitation/460 nM emission. LI-COR Odyssey Analysis and Enzyme Kinetics For quantitation of expression levels, immunoblots were incubated with Odyssey-compatible secondary antibodies (mouse IR680 [Invitrogen], rat IR680 [Invitrogen], and guinea pig IR800 [Rockland]) and analyzed at 700 nm and 800 nm excitation, respectively. Odyssey software was used to determine the integrated intensities. Experiments for determining KM and Vmax were performed in triplicates and analyzed with Prism software.
SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures and four figures and can be found with this article online at doi:10.1016/ j.molcel.2010.11.011. ACKNOWLEDGMENTS We would like to thank Millenium Pharmaceuticals for MLN4924 compound. We thank members of the Meier Lab for support and critical reading of the manuscript. In particular, we would like to thank Mariam Orme for advice with fly genetics, Marketa Zvelebil for help with bioinformatics, and Helen Walden for support in structural information. M.B. is supported by a fellowship of the Deutsche Forschungsgemeinschaft and Breakthrough. We acknowledge National Health Service funding to the National Institute for Health Research Biomedical Research Centre. Received: January 25, 2010 Revised: August 2, 2010 Accepted: September 13, 2010 Published: December 9, 2010
Brzovic, P.S., and Klevit, R.E. (2006). Ubiquitin transfer from the E2 perspective: why is UbcH5 so promiscuous? Cell Cycle 5, 2867–2873. Chai, J., Yan, N., Huh, J.R., Wu, J.W., Li, W., Hay, B.A., and Shi, Y. (2003). Molecular mechanism of Reaper-Grim-Hid-mediated suppression of DIAP1-dependent Dronc ubiquitination. Nat. Struct. Biol. 10, 892–898. Chan, Y., Yoon, J., Wu, J.T., Kim, H.J., Pan, K.T., Yim, J., and Chien, C.T. (2008). DEN1 deneddylates non-cullin proteins in vivo. J. Cell Sci. 121, 3218–3223. Choi, Y.E., Butterworth, M., Malladi, S., Duckett, C.S., Cohen, G.M., and Bratton, S.B. (2009). The E3 ubiquitin ligase cIAP1 binds and ubiquitinates caspase-3 and -7 via unique mechanisms at distinct steps in their processing. J. Biol. Chem. 284, 12772–12782. Cope, G.A., Suh, G.S., Aravind, L., Schwarz, S.E., Zipursky, S.L., Koonin, E.V., and Deshaies, R.J. (2002). Role of predicted metalloprotease motif of Jab1/ Csn5 in cleavage of Nedd8 from Cul1. Science 298, 608–611. Deshaies, R.J., and Joazeiro, C.A. (2009). RING domain E3 ubiquitin ligases. Annu. Rev. Biochem. 78, 399–434. Di Fiore, P.P., Polo, S., and Hofmann, K. (2003). When ubiquitin meets ubiquitin receptors: a signalling connection. Nat. Rev. Mol. Cell Biol. 4, 491–497. Dikic, I., Wakatsuki, S., and Walters, K.J. (2009). Ubiquitin-binding domains from structures to functions. Nat. Rev. Mol. Cell Biol. 10, 659–671. Ditzel, M., and Meier, P. (2002). IAP degradation: decisive blow or altruistic sacrifice? Trends Cell Biol. 12, 449–452. Ditzel, M., Wilson, R., Tenev, T., Zachariou, A., Paul, A., Deas, E., and Meier, P. (2003). Degradation of DIAP1 by the N-end rule pathway is essential for regulating apoptosis. Nat. Cell Biol. 5, 467–473. Ditzel, M., Broemer, M., Tenev, T., Bolduc, C., Lee, T.V., Rigbolt, K.T., Elliott, R., Zvelebil, M., Blagoev, B., Bergmann, A., and Meier, P. (2008). Inactivation of effector caspases through nondegradative polyubiquitylation. Mol. Cell 32, 540–553. Duda, D.M., Borg, L.A., Scott, D.C., Hunt, H.W., Hammel, M., and Schulman, B.A. (2008). Structural insights into NEDD8 activation of cullin-RING ligases: conformational control of conjugation. Cell 134, 995–1006. Eckelman, B.P., Salvesen, G.S., and Scott, F.L. (2006). Human inhibitor of apoptosis proteins: why XIAP is the black sheep of the family. EMBO Rep. 7, 988–994. Furukawa, M., Zhang, Y., McCarville, J., Ohta, T., and Xiong, Y. (2000). The CUL1 C-terminal sequence and ROC1 are required for efficient nuclear accumulation, NEDD8 modification, and ubiquitin ligase activity of CUL1. Mol. Cell. Biol. 20, 8185–8197. Gaither, A., Porter, D., Yao, Y., Borawski, J., Yang, G., Donovan, J., Sage, D., Slisz, J., Tran, M., Straub, C., et al. (2007). A Smac mimetic rescue screen reveals roles for inhibitor of apoptosis proteins in tumor necrosis factor-alpha signaling. Cancer Res. 67, 11493–11498. Gao, F., Cheng, J., Shi, T., and Yeh, E.T. (2006). Neddylation of a breast cancer-associated protein recruits a class III histone deacetylase that represses NFkappaB-dependent transcription. Nat. Cell Biol. 8, 1171–1177. Garcia-Dominguez, M., and Reyes, J.C. (2009). SUMO association with repressor complexes, emerging routes for transcriptional control. Biochim. Biophys. Acta 1789, 451–459. Gong, L., and Yeh, E.T. (1999). Identification of the activating and conjugating enzymes of the NEDD8 conjugation pathway. J. Biol. Chem. 274, 12036– 12042. Grether, M.E., Abrams, J.M., Agapite, J., White, K., and Steller, H. (1995). The head involution defective gene of Drosophila melanogaster functions in programmed cell death. Genes Dev. 9, 1694–1708.
REFERENCES
Haglund, K., and Dikic, I. (2005). Ubiquitylation and cell signaling. EMBO J. 24, 3353–3359.
Broemer, M., and Meier, P. (2009). Ubiquitin-mediated regulation of apoptosis. Trends Cell Biol. 19, 130–140.
Hatakeyama, S., and Nakayama, K.I. (2003). U-box proteins as a new family of ubiquitin ligases. Biochem. Biophys. Res. Commun. 302, 635–645.
820 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
Molecular Cell IAPs as NEDD8 E3 Ligases
Hays, R., Wickline, L., and Cagan, R. (2002). Morgue mediates apoptosis in the Drosophila melanogaster retina by promoting degradation of DIAP1. Nat. Cell Biol. 4, 425–431.
Osaka, F., Kawasaki, H., Aida, N., Saeki, M., Chiba, T., Kawashima, S., Tanaka, K., and Kato, S. (1998). A new NEDD8-ligating system for cullin-4A. Genes Dev. 12, 2263–2268.
Herman-Bachinsky, Y., Ryoo, H.D., Ciechanover, A., and Gonen, H. (2007). Regulation of the Drosophila ubiquitin ligase DIAP1 is mediated via several distinct ubiquitin system pathways. Cell Death Differ. 14, 861–871.
Oved, S., Mosesson, Y., Zwang, Y., Santonico, E., Shtiegman, K., Marmor, M.D., Kochupurakkal, B.S., Katz, M., Lavi, S., Cesareni, G., and Yarden, Y. (2006). Conjugation to Nedd8 instigates ubiquitylation and down-regulation of activated receptor tyrosine kinases. J. Biol. Chem. 281, 21640–21651.
Hochstrasser, M. (2009). Origin and function of ubiquitin-like proteins. Nature 458, 422–429. Hoeller, D., and Dikic, I. (2009). Targeting the ubiquitin system in cancer therapy. Nature 458, 438–444. Holley, C.L., Olson, M.R., Colo´n-Ramos, D.A., and Kornbluth, S. (2002). Reaper eliminates IAP proteins through stimulated IAP degradation and generalized translational inhibition. Nat. Cell Biol. 4, 439–444. Jin, Z., Li, Y., Pitti, R., Lawrence, D., Pham, V.C., Lill, J.R., and Ashkenazi, A. (2009). Cullin3-based polyubiquitination and p62-dependent aggregation of caspase-8 mediate extrinsic apoptosis signaling. Cell 137, 721–735. Kaiser, W.J., Vucic, D., and Miller, L.K. (1998). The Drosophila inhibitor of apoptosis D-IAP1 suppresses cell death induced by the caspase drICE. FEBS Lett. 440, 243–248. Kerscher, O., Felberbaum, R., and Hochstrasser, M. (2006). Modification of proteins by ubiquitin and ubiquitin-like proteins. Annu. Rev. Cell Dev. Biol. 22, 159–180. Kirkpatrick, D.S., Hathaway, N.A., Hanna, J., Elsasser, S., Rush, J., Finley, D., King, R.W., and Gygi, S.P. (2006). Quantitative analysis of in vitro ubiquitinated cyclin B1 reveals complex chain topology. Nat. Cell Biol. 8, 700–710.
Podust, V.N., Brownell, J.E., Gladysheva, T.B., Luo, R.S., Wang, C., Coggins, M.B., Pierce, J.W., Lightcap, E.S., and Chau, V. (2000). A Nedd8 conjugation pathway is essential for proteolytic targeting of p27Kip1 by ubiquitination. Proc. Natl. Acad. Sci. USA 97, 4579–4584. Pozo, J.C., Timpte, C., Tan, S., Callis, J., and Estelle, M. (1998). The ubiquitinrelated protein RUB1 and auxin response in Arabidopsis. Science 280, 1760– 1763. Read, M.A., Brownell, J.E., Gladysheva, T.B., Hottelet, M., Parent, L.A., Coggins, M.B., Pierce, J.W., Podust, V.N., Luo, R.S., Chau, V., and Palombella, V.J. (2000). Nedd8 modification of cul-1 activates SCF(beta(TrCP))-dependent ubiquitination of IkappaBalpha. Mol. Cell. Biol. 20, 2326–2333. Reyes-Turcu, F.E., Ventii, K.H., and Wilkinson, K.D. (2009). Regulation and cellular roles of ubiquitin-specific deubiquitinating enzymes. Annu. Rev. Biochem. 78, 363–397. Rodriguez, M.S., Wright, J., Thompson, J., Thomas, D., Baleux, F., Virelizier, J.L., Hay, R.T., and Arenzana-Seisdedos, F. (1996). Identification of lysine residues required for signal-induced ubiquitination and degradation of I kappa B-alpha in vivo. Oncogene 12, 2425–2435.
Komander, D. (2009). The emerging complexity of protein ubiquitination. Biochem. Soc. Trans. 37, 937–953.
Rodriguez, M.S., Desterro, J.M., Lain, S., Midgley, C.A., Lane, D.P., and Hay, R.T. (1999). SUMO-1 modification activates the transcriptional response of p53. EMBO J. 18, 6455–6461.
Lammer, D., Mathias, N., Laplaza, J.M., Jiang, W., Liu, Y., Callis, J., Goebl, M., and Estelle, M. (1998). Modification of yeast Cdc53p by the ubiquitin-related protein rub1p affects function of the SCFCdc4 complex. Genes Dev. 12, 914–926.
Rodriguez, M.S., Desterro, J.M., Lain, S., Lane, D.P., and Hay, R.T. (2000). Multiple C-terminal lysine residues target p53 for ubiquitin-proteasomemediated degradation. Mol. Cell. Biol. 20, 8458–8467.
Lee, M.R., Lee, D., Shin, S.K., Kim, Y.H., and Choi, C.Y. (2008). Inhibition of APP intracellular domain (AICD) transcriptional activity via covalent conjugation with Nedd8. Biochem. Biophys. Res. Commun. 366, 976–981. Leulier, F., Ribeiro, P.S., Palmer, E., Tenev, T., Takahashi, K., Robertson, D., Zachariou, A., Pichaud, F., Ueda, R., and Meier, P. (2006). Systematic in vivo RNAi analysis of putative components of the Drosophila cell death machinery. Cell Death Differ. 13, 1663–1674. Liakopoulos, D., Doenges, G., Matuschewski, K., and Jentsch, S. (1998). A novel protein modification pathway related to the ubiquitin system. EMBO J. 17, 2208–2214. Lisi, S., Mazzon, I., and White, K. (2000). Diverse domains of THREAD/DIAP1 are required to inhibit apoptosis induced by REAPER and HID in Drosophila. Genetics 154, 669–678. Lyapina, S., Cope, G., Shevchenko, A., Serino, G., Tsuge, T., Zhou, C., Wolf, D.A., Wei, N., Shevchenko, A., and Deshaies, R.J. (2001). Promotion of NEDD-CUL1 conjugate cleavage by COP9 signalosome. Science 292, 1382–1385.
Rotin, D., and Kumar, S. (2009). Physiological functions of the HECT family of ubiquitin ligases. Nat. Rev. Mol. Cell Biol. 10, 398–409. Ryoo, H.D., Bergmann, A., Gonen, H., Ciechanover, A., and Steller, H. (2002). Regulation of Drosophila IAP1 degradation and apoptosis by reaper and ubcD1. Nat. Cell Biol. 4, 432–438. Saha, A., and Deshaies, R.J. (2008). Multimodal activation of the ubiquitin ligase SCF by Nedd8 conjugation. Mol. Cell 32, 21–31. Scheer, J.M., Romanowski, M.J., and Wells, J.A. (2006). A common allosteric site and mechanism in caspases. Proc. Natl. Acad. Sci. USA 103, 7595–7600. Scherer, D.C., Brockman, J.A., Chen, Z., Maniatis, T., and Ballard, D.W. (1995). Signal-induced degradation of I kappa B alpha requires site-specific ubiquitination. Proc. Natl. Acad. Sci. USA 92, 11259–11263. Schile, A.J., Garcı´a-Ferna´ndez, M., and Steller, H. (2008). Regulation of apoptosis by XIAP ubiquitin-ligase activity. Genes Dev. 22, 2256–2266. Schulman, B.A., and Harper, J.W. (2009). Ubiquitin-like protein activation by E1 enzymes: the apex for downstream signalling pathways. Nat. Rev. Mol. Cell Biol. 10, 319–331.
Merlet, J., Burger, J., Gomes, J.E., and Pintard, L. (2009). Regulation of cullin-RING E3 ubiquitin-ligases by neddylation and dimerization. Cell. Mol. Life Sci. 66, 1924–1938.
Schwechheimer, C., Serino, G., Callis, J., Crosby, W.L., Lyapina, S., Deshaies, R.J., Gray, W.M., Estelle, M., and Deng, X.W. (2001). Interactions of the COP9 signalosome with the E3 ubiquitin ligase SCFTIRI in mediating auxin response. Science 292, 1379–1382.
Morimoto, M., Nishida, T., Honda, R., and Yasuda, H. (2000). Modification of cullin-1 by ubiquitin-like protein Nedd8 enhances the activity of SCF(skp2) toward p27(kip1). Biochem. Biophys. Res. Commun. 270, 1093–1096.
Shapiro, P.J., Hsu, H.H., Jung, H., Robbins, E.S., and Ryoo, H.D. (2008). Regulation of the Drosophila apoptosome through feedback inhibition. Nat. Cell Biol. 10, 1440–1446.
Morizane, Y., Honda, R., Fukami, K., and Yasuda, H. (2005). X-linked inhibitor of apoptosis functions as ubiquitin ligase toward mature caspase-9 and cytosolic Smac/DIABLO. J. Biochem. 137, 125–132.
Silke, J., Kratina, T., Chu, D., Ekert, P.G., Day, C.L., Pakusch, M., Huang, D.C., and Vaux, D.L. (2005). Determination of cell survival by RING-mediated regulation of inhibitor of apoptosis (IAP) protein abundance. Proc. Natl. Acad. Sci. USA 102, 16182–16187.
Mukhopadhyay, D., and Dasso, M. (2007). Modification in reverse: the SUMO proteases. Trends Biochem. Sci. 32, 286–295. Nijman, S.M., Luna-Vargas, M.P., Velds, A., Brummelkamp, T.R., Dirac, A.M., Sixma, T.K., and Bernards, R. (2005). A genomic and functional inventory of deubiquitinating enzymes. Cell 123, 773–786.
Soucy, T.A., Smith, P.G., Milhollen, M.A., Berger, A.J., Gavin, J.M., Adhikari, S., Brownell, J.E., Burke, K.E., Cardin, D.P., Critchley, S., et al. (2009). An inhibitor of NEDD8-activating enzyme as a new approach to treat cancer. Nature 458, 732–736.
Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc. 821
Molecular Cell IAPs as NEDD8 E3 Ligases
Souphron, J., Waddell, M.B., Paydar, A., Tokgo¨z-Gromley, Z., Roussel, M.F., and Schulman, B.A. (2008). Structural dissection of a gating mechanism preventing misactivation of ubiquitin by NEDD8’s E1. Biochemistry 47, 8961–8969. Stickle, N.H., Chung, J., Klco, J.M., Hill, R.P., Kaelin, W.G., Jr., and Ohh, M. (2004). pVHL modification by NEDD8 is required for fibronectin matrix assembly and suppression of tumor development. Mol. Cell. Biol. 24, 3251– 3261. Suzuki, Y., Nakabayashi, Y., Nakata, K., Reed, J.C., and Takahashi, R. (2001). X-linked inhibitor of apoptosis protein (XIAP) inhibits caspase-3 and -7 in distinct modes. J. Biol. Chem. 276, 27058–27063. Tasaki, T., Mulder, L.C., Iwamatsu, A., Lee, M.J., Davydov, I.V., Varshavsky, A., Muesing, M., and Kwon, Y.T. (2005). A family of mammalian E3 ubiquitin ligases that contain the UBR box motif and recognize N-degrons. Mol. Cell. Biol. 25, 7120–7136. Tenev, T., Zachariou, A., Wilson, R., Ditzel, M., and Meier, P. (2005). IAPs are functionally non-equivalent and regulate effector caspases through distinct mechanisms. Nat. Cell Biol. 7, 70–77.
Whitby, F.G., Xia, G., Pickart, C.M., and Hill, C.P. (1998). Crystal structure of the human ubiquitin-like protein NEDD8 and interactions with ubiquitin pathway enzymes. J. Biol. Chem. 273, 34983–34991. White, K., Grether, M.E., Abrams, J.M., Young, L., Farrell, K., and Steller, H. (1994). Genetic control of programmed cell death in Drosophila. Science 264, 677–683. Wilson, R., Goyal, L., Ditzel, M., Zachariou, A., Baker, D.A., Agapite, J., Steller, H., and Meier, P. (2002). The DIAP1 RING finger mediates ubiquitination of Dronc and is indispensable for regulating apoptosis. Nat. Cell Biol. 4, 445–450. Wing, J.P., Schreader, B.A., Yokokura, T., Wang, Y., Andrews, P.S., Huseinovic, N., Dong, C.K., Ogdahl, J.L., Schwartz, L.M., White, K., and Nambu, J.R. (2002). Drosophila Morgue is an F box/ubiquitin conjugase domain protein important for grim-reaper mediated apoptosis. Nat. Cell Biol. 4, 451–456. Wu, K., Chen, A., and Pan, Z.Q. (2000). Conjugation of Nedd8 to CUL1 enhances the ability of the ROC1-CUL1 complex to promote ubiquitin polymerization. J. Biol. Chem. 275, 32317–32324. Xirodimas, D.P. (2008). Novel substrates and functions for the ubiquitin-like molecule NEDD8. Biochem. Soc. Trans. 36, 802–806.
Tenev, T., Ditzel, M., Zachariou, A., and Meier, P. (2007). The antiapoptotic activity of insect IAPs requires activation by an evolutionarily conserved mechanism. Cell Death Differ. 14, 1191–1201.
Xirodimas, D.P., Saville, M.K., Bourdon, J.C., Hay, R.T., and Lane, D.P. (2004). Mdm2-mediated NEDD8 conjugation of p53 inhibits its transcriptional activity. Cell 118, 83–97.
Treier, M., Staszewski, L.M., and Bohmann, D. (1994). Ubiquitin-dependent c-Jun degradation in vivo is mediated by the delta domain. Cell 78, 787–798.
Xirodimas, D.P., Sundqvist, A., Nakamura, A., Shen, L., Botting, C., and Hay, R.T. (2008). Ribosomal proteins are targets for the NEDD8 pathway. EMBO Rep. 9, 280–286.
Varshavsky, A. (2000). Ubiquitin fusion technique and its descendants. Methods Enzymol. 327, 578–593. Vaux, D.L., and Silke, J. (2005). IAPs, RINGs and ubiquitylation. Nat. Rev. Mol. Cell Biol. 6, 287–297.
Yan, N., Wu, J.W., Chai, J., Li, W., and Shi, Y. (2004). Molecular mechanisms of DrICE inhibition by DIAP1 and removal of inhibition by Reaper, Hid and Grim. Nat. Struct. Mol. Biol. 11, 420–428.
Vijay-Kumar, S., Bugg, C.E., and Cook, W.J. (1987). Structure of ubiquitin refined at 1.8 A resolution. J. Mol. Biol. 194, 531–544.
Yoo, S.J., Huh, J.R., Muro, I., Yu, H., Wang, L., Wang, S.L., Feldman, R.M., Clem, R.J., Mu¨ller, H.A., and Hay, B.A. (2002). Hid, Rpr and Grim negatively regulate DIAP1 levels through distinct mechanisms. Nat. Cell Biol. 4, 416–424.
Watson, I.R., Li, B.K., Roche, O., Blanch, A., Ohh, M., and Irwin, M.S. (2010). Chemotherapy induces NEDP1-mediated destabilization of MDM2. Oncogene 29, 297–304.
Zachariou, A., Tenev, T., Goyal, L., Agapite, J., Steller, H., and Meier, P. (2003). IAP-antagonists exhibit non-redundant modes of action through differential DIAP1 binding. EMBO J. 22, 6642–6652.
822 Molecular Cell 40, 810–822, December 10, 2010 ª2010 Elsevier Inc.
