Design, Synthesis, and Biological Activity of a Potent Smac Mimetic

Sep 15, 2006 - Brice H. CurtinFrancesco ManoniJiyong ParkLuke J. SistoYu-hong LamMichel .... Stacy Frankovitz Reisner , Lewis Gazzard , Jason Halladay...
0 downloads 0 Views 1MB Size
Download PDF
ARTICLE

Design, Synthesis, and Biological Activity of a Potent Smac Mimetic That Sensitizes Cancer Cells to Apoptosis by Antagonizing IAPs

Kerry Zobel†,‡, Lan Wang‡, Eugene Varfolomeev†, Matthew C. Franklin†, Linda O. Elliott‡, Heidi J. A. Wallweber†, David C. Okawa‡, John A. Flygare‡, Domagoj Vucic†, Wayne J. Fairbrother†, and Kurt Deshayes†,‡,* †

Departments of Protein Engineering and ‡Medicinal Chemistry, Genentech, Inc., South San Francisco, California 94080

T

he proper regulation of apoptosis is crucial for development and sustained health (1). Many disease states result from over-sensitivity or resistance to apoptotic stimuli (2). Of special interest is the role of apoptotic resistance in aggressive cancers, since a lack of death response often permits cancers to tolerate conventional treatment (3). The inhibitor of apoptosis (IAP) proteins are key components of the apoptotic cascade (4) and are believed to prevent cell death through interactions between their baculoviral IAP repeat (BIR) domains and caspase-3, -7, or -9, which are critical for the initiation and execution phases of apoptosis (Figure 1). Overexpression of IAP proteins in human cancers has been shown to suppress apoptosis induced by a variety of stimuli (5–7). X-chromosome-linked IAP (XIAP) is a ubiquitously expressed IAP protein and a potent inhibitor of caspases that plays a critical role in resistance to chemotherapeutic agents and other proapoptotic stimuli (8–11). Melanoma inhibitor of apoptosis (ML-IAP) is upregulated in a number of melanomas but not expressed in most normal adult tissues (12, 13), and down regulation of ML-IAP by RNA interference leads to induction of apoptosis in tumor cells (14, 15). Cellular Inhibitors of apoptosis 1 and 2 (cIAP1 and cIAP2) are unique among IAP proteins for their ability to interact with tumor necrosis factor receptor-associated factors 1 and 2 (TRAF1 and 2) (16). In addition, cIAP1 and cIAP2 are targets of genetic amplification, which is potentially correlated with resistance to antitumor agents (17–19). Taken together, IAP proteins are attractive targets for anticancer therapeutic intervention (20). One mechanism by which XIAP inhibits apoptosis is the interaction between its BIR3 domain and caspase-9, which prevents the enzyme from adopting the catalyti-

www.acschemicalbiology.org

A B S T R A C T Designed second mitochondrial activator of caspases (Smac) mimetics based on an accessible [7,5]-bicyclic scaffold bind to and antagonize protein interactions involving the inhibitor of apoptosis (IAP) proteins, X-chromosomelinked IAP (XIAP), melanoma IAP (ML-IAP), and c-IAPs 1 and 2 (cIAP1 and cIAP2). The design rationale is based on a combination of phage-panning data, peptide binding studies, and a survey of potential isosteres. The synthesis of two scaffolds is described. These compounds bind the XIAP-baculoviral IAP repeat 3 (BIR3), cIAP1-BIR3, cIAP2-BIR3, and ML-IAP-BIR domains with submicromolar affinities. The most potent Smac mimetic binds the cIAP1–BIR3 and ML-IAP–BIR domains with a Ki of 50 nM. The X-ray crystal structure of this compound bound to an MLIAP/XIAP chimeric BIR domain protein is compared with that of a complex with a phage-derived tetrapeptide, AVPW. The structures show that these compounds bind to the Smac-binding site on ML-IAP with identical hydrogen-bonding patterns and similar hydrophobic interactions. Consistent with the structural data, coimmunoprecipitation experiments demonstrate that the compounds can effectively block Smac interactions with ML-IAP. The compounds are further demonstrated to activate caspase-3 and -7, to reduce cell viability in assays using MDA-MB-231 breast cancer cells and A2058 melanoma cells, and to enhance doxorubicin-induced apoptosis in MDA-MB-231 cells.

*Corresponding author, [email protected]. Received for review June 27, 2006 and accepted August 25, 2006. Published online September 15, 2006 10.1021/cb600276q CCC: $33.50 © 2006 by American Chemical Society

VOL.1 NO.8 • ACS CHEMICAL BIOLOGY

525

Figure 1. Apoptotic pathway. The extrinsic apoptotic pathway is triggered when death receptors, such as Fas, DR4, or TNF receptor 1, are engaged by their cognate ligands, resulting in recruitment of the adaptor protein Fas-Associated Death Domain (FADD) and the apical caspase, caspase-8. This leads to activation of caspase-8 and subsequent activation of the effector caspases, caspase-3 and -7. The intrinsic apoptotic pathway is triggered by stimuli such as irradiation, chemotherapeutic agents, or growth factor withdrawal. Activation of pro-apoptotic BH3only members of the Bcl-2 family neutralizes the anti-apoptotic proteins Bcl-2, Bcl-xL, and Mcl-1, leading to disruption of the mitochondrial membrane potential and the release of cytochrome c and Smac from the mitochondria into the cytoplasm. These events result in Apaf-1-mediated activation of caspase-9 and subsequent activation of the effector caspases, caspase-3 and -7, and culminate in dismantling of the cell. The IAP proteins are the last line of defense against cellular suicide and act by inhibiting caspases (XIAP) and by preventing Smac from blocking XIAP-mediated caspase inhibition (ML-IAP, cIAP1, and cIAP2). Smac mimetics bind to IAP proteins and block their inhibitory activity by antagonizing the critical IAP– caspase and IAP–Smac interactions.

cally active homodimer conformation (21). The four N-terminal residues of the small subunit of caspase-9, ATPF, make critical interactions with a peptide-binding groove on XIAP-BIR3 (21, 22). The endogenous IAP antagonist protein second mitochondrial activator of caspases (Smac) is released from the mitochondria in response to pro-apoptotic stimuli (23, 24). This mature, processed form of Smac has been demonstrated to bind to the same peptide-binding groove of XIAP-BIR3 via its four N-terminal residues, AVPI, thus releasing caspase-9 and promoting apoptosis (22, 25, 26). Recent data suggest that ML-IAP, cIAP1, and cIAP2 may inhibit apoptosis primarily by binding Smac and thereby preventing it from antagonizing the ubiquitously expressed XIAP, rather than by directly inhibiting caspases (27–29). Regardless of the specific mode of action, it is reasonable to believe that binding a Smacbased peptidomimetic to the relevant BIR domains of these IAP proteins will have a pro-apoptotic effect. Indeed, Smac-derived peptides have been demonstrated to sensitize a number of different tumor cell lines 526

VOL.1 NO.8 • 525–533 • 2006

ZOBEL ET AL.

to apoptosis induced by a variety of pro-apoptotic drugs (20, 30–33). Pro-apoptotic activity has also been reported for peptidomimetics, both in vitro and in vivo (34–37). Such agents hold the promise of reducing therapeutic resistance in cancers where IAP expression precludes apoptosis. We report here the development of Smac mimetics based on an easily accessible [7,5]bicyclic scaffold, including design rationale, synthesis, binding affinities, and an X-ray crystal structure of an isostere–IAP-BIR domain complex. Evidence for isosteredriven activation of caspase-3 and -7 and induction of apoptosis in cancer cells is also presented.

RESULTS AND DISCUSSION Structural Analysis of Peptide–BIR Domain Complexes. Structure/activity relationships of peptide binding to the BIR domain of ML-IAP and the BIR2 and BIR3 domains of XIAP were investigated previously using peptide-phage display technology and peptide library positional scanning (38). Most dramatically, for peptides selected to bind ML-IAP-BIR and XIAP-BIR3, alanine was found in the N-terminal (P1) position in 100% of the selectants. Other positions tolerated some variation, although preferences for valine at P2, proline at P3, and the aromatic residues phenylalanine or tryptophan at P4 were apparent. Of these, the P2 position could tolerate the most substitutions without significant losses in affinity. Examination of the crystal structure of the most potent peptide, AVPW (Table 1), in complex with MLXBIR3SG (an ML-IAP/XIAP chimeric BIR domain that preserves the native ML-IAP peptide-binding site) (27) reinforces the key points of the phage selection experiments (Figure 2, panel a). The structure of the chimeric BIR domain in the 1.6 Å resolution structure of the MLXBIR3SG–AVPW complex is essentially identical to that reported previously for the MLXBIR3SG–Smac 9-mer peptide complex (27) with a C␣ root mean square deviation of 0.115 Å for 180 atom pairs. In the P1 position, the alanine methyl group is buried in a hydrophobic pocket formed by the side chains of Leu131, Trp134, and Glu143 (ML-IAP residue numbering) of the protein. Thus, substituting this methyl group with anything larger leads to a large decrease in binding affinity. The N-terminus of the peptide is in an acidic environment with the amino group of Ala1 hydrogen bonded to the side chain carboxylates of Asp138 and Glu143. Others have noted that N-methylation of www.acschemicalbiology.org

ARTICLE sequence overlapped with that of known dipeptide isosteres. A series of five Y dipeptide isosteres were O S X substituted into positions P2 N HN N X H and P3 of the Smac-based O NH O peptide AVPIAQKSE, and Ki N values were determined for N binding to ML-IAP-BIR and XIAP-BIR3. Incorporation of Ki (␮M) conformationally rigid dipeptide isosteres such as bena Compound X Y cIAP1-BIR3 cIAP2-BIR3 XIAP-BIR3 ML-IAP-BIR zodiazepine or 3-amino-1b b AVPIAQKSE (1) 0.67 0.5 carboxymethylvalerolactone b b 2 39.8 15.3 in positions P2 and P3 AVPW 0.05 0.04 0.03 0.07 resulted in complete loss of 8 CH3 0.05 0.05 0.13 H (R) 0.77 activity. In both cases, b b 9 CH3 0.19 H (S) 2.34 manual docking into the 10 H 0.28 0.11 0.49 H (R) 0.27 peptide-binding site of b b 11 H 4.19 H (S) 3.00 ML-IAP-BIR suggested unfaa vorable steric clashes with Binding affinity for ML-IAP-BIR was determined using the chimeric protein MLXBIR3SG; binding affinity for cIAP1-BIR3 was determined using the chimeric protein cIAP1XBIR3; Trp147 as the cause of this binding affinity for cIAP2-BIR3 was determined using the chimeric protein cIAP2XBIR3. loss in binding affinity. b Not determined. Molecular modeling suggested that a more flexible 7-membered ring might the alanine is tolerated ithout loss of binding affinity adopt a conformation that allows favorable interactions (34, 37). The relative promiscuity at the P2 position is ex- with Trp147, while maintaining the P2 hydrogen plained by the lack of substantial interaction between bonding with the protein as well as optimal binding orithe Val2 side chain and the protein, although this entations at positions P1 and P4. Thus, substituting residue does form two main chain–main chain hydro3-amino-1-carboxymethylcaprolactam into positions P2 gen bonds with the protein. Pro3 makes extensive van der Waals contact with Trp147 and helps position the P1 and P4 residues for optimal binding. The P4-binding pockets of the BIR domains were shown to accommodate a number of large aromatic groups, as illustrated by the deep pocket binding the tryptophan side chain in the crystal structure of the MLXBIR3SG–peptide complex (Figure 2, panel a). Despite its high binding affinity, AVPW has no measurable biological activity in cell-based assays (data not shown). Therefore, to obtain active agents, it is necessary to translate the key components of the AVPW strucFigure 2. Structure of the AVPW peptide and isostere 8 bound to the Smac binding ture onto a non-peptide scaffold with more druglike site. Solvent-accessible surface representation of the peptide-binding site of properties. MLXBIR3SG in complex with a) the phage-derived peptide AVPW and b) peptide Isostere Design. The first step toward evolving the isostere 8. The protein surface is color-coded according to electrostatic surface peptide into a more druglike molecule was to determine potential: red is negatively charged; blue is positively charged. This figure was produced using the program PyMOL (www.pymol.org) (46). whether the bioactive conformation of the peptide TABLE 1.

Structure of isosteres 8–11 and binding affinities for XIAP-BIR3, ML-IAP-BIR, cIAP1-BIR3, and cIAP2-BIR3

www.acschemicalbiology.org

VOL.1 NO.8 • 525–533 • 2006

527

Figure 3. Conversion of peptide to isostere. Mutation of peptide 1 to give either the caprolactam 2 or the peptide 3, where proline is substituted with alanine, results in a 30-fold loss in binding affinity for the BIR domain of ML-IAP. Fusion of a five-membered ring to the sevenmembered ring of 2 gives 4, which has a binding affinity comparable to the starting peptide 1.

and P3 of the Smac-derived 9-mer peptide resulted in compound 2 with binding affinity (Ki) to the ML-IAP-BIR and XIAP-BIR3 domains of 15 and 40 ␮M, respectively. While this represents a 30–60-fold loss in affinity relative to the starting peptide, prior peptide positional scanning data showed similar losses in affinity when proline was substituted with an alanine at position P3 (compound 3) (38). We reasoned that introduction of a fused 5-membered ring within the context of the caprolactam to yield a [7,5]-bicyclic compound (4) should retain the high affinity of the endogenous peptide sequence, while altering the peptide character of the P2 and P3 positions (Figure 3). Similar bicyclic lactams have been used previously to conformationally constrain a peptide backbone (39). For synthetic ease, a sulfur was chosen for the 3 position of the 5-membered ring. Since we had discovered previously that substitution of proline in the tetrapeptide AVPI with ␥-thioproline was tolerated without loss of affinity (data not shown), this scaffold was chosen as our initial target (Figure 3). The synthesis of Patchett et al. (40) was used to construct the bicyclic [7,5]-lactam 4 as a 1:1 mixture of the two ring-junction diastereomers (Figure 3). The diastereomeric mixture was determined to have a Ki value for binding to ML-IAP-BIR of 1 ␮M. 528

VOL.1 NO.8 • 525–533 • 2006

ZOBEL ET AL.

Depending on the potency of each diastereomer, the addition of the 5-membered ring to the caprolactam peptide isostere results in a 15–30-fold improvement in affinity for ML-IAP-BIR and thus gives a conformationally constrained Smac mimetic with an affinity comparable to that of the 9-mer peptide 1 (Table 1). This, in turn, indicates that the [7,5]-bicyclic lactam effectively constrains the middle portion of the molecule (P2–P3) and can be expected to provide a scaffold that can be further optimized at positions P1 and P4. N-Methyl alanine in the P1 position was found to impart proteolytic stability and, as noted previously, is tolerated without loss of binding affinity. Given the preference for aromatic groups in the P4 position, we focused on incorporating heterocyclic rings that were able to place an aromatic group in the correct orientation to interact with the P4-binding pocket of the protein. A variety of 5-membered ring heterocycles were synthesized, with pyrazole 6 (Scheme 1) used to construct isosteres providing optimal positioning of the aromatic ring in P4. Synthesis. A convergent synthetic approach was required to allow access to a variety of analogues (Scheme 1). The aldehyde 5 was prepared in three steps from diphenylmethylene glycine ethyl ester (synthetic details are given in Supporting Information). The key step of the synthesis is the condensation of 5 with primary amines 6a,b, derived from either cysteine or penicillamine, to give thiazolidines 7a,b. Saponification followed by ring closure yielded peptide isosteres 8–11 (Table 1). Discrimination between the BIR Domains. The Ki values for binding of the different Smac mimetics to the relevant BIR domains of XIAP, cIAP1, cIAP2, and ML-IAP were determined using a fluorescence polarization assay as described in Methods (Figure 4). The isosteres with the preferred bridgehead stereochemistry bind to the BIR domains in the submicromolar range with the highest affinity observed for isostere 8 binding to MLXBIR3SG (which can be considered equivalent to ML-IAP-BIR, with regard to Smac peptide binding) (27) and cIAP1XBIR3 (which can be considered equivalent to cIAP1-BIR3, with regard to Smac peptide binding) (Table 1). In the case of the penicillamine derivatives, the nonpreferred ring-junction stereochemistry (isostere 9) results in a 3–4-fold decrease in binding affinity for both XIAP-BIR3 and MLXBIR3SG relative to isostere 8. Similarly, for the cysteine derivatives, reversing the ring-junction stereochemistry of the preferred isostere 10 to give the diasteromer 11 results in a greater than 10-fold decrease www.acschemicalbiology.org

ARTICLE

Scheme 1. Synthesis of isosteres 8 –11.

in affinity for both XIAP-BIR3 and MLXBIR3SG. As anticipated (38), the penicillamine derivatives bind more tightly to MLXBIR3SG than to XIAP-BIR3, while the cysteine-derived isosteres 10 and 11 bind with essentially equal affinity to both XIAP-BIR3 and MLXBIR3SG. These data agree with a model in which the hydroxyl group of Tyr324 of XIAP-BIR3 (the residue corresponding to Phe148 in ML-IAP, Phe330 in cIAP1, and Phe316 in cIAP2) has a steric clash with the (pro-R)-methyl groups in the P3 positions of isosteres 8 and 9, while the aromatic side chains of the corresponding phenylalanine residues in the other BIR domains investigated have favorable interactions with the (pro-R)-methyl groups of these compounds. In these examples, the selectivity for ML-IAP-BIR over XIAP-BIR3 is on the order of 15-fold, compared with the ⬃100-fold discrimination observed previously when (3S)-methyl-proline was substituted into the AVPIAQKSE peptide (38). Structure of MLXBIR3SG–Isostere 8 Complex.To gain a more detailed understanding of the interactions between the Smac mimetics and the BIR domain of www.acschemicalbiology.org

ML-IAP, the crystal structure of the complex between MLXBIR3SG and peptide isostere 8 was determined to a resolution of 2.3 Å and compared with that of the AVPW peptide complex (Figure 2, panel b). X-ray data collection and refinement statistics are listed in Supplementary Table 1. The key contacts seen in the structure of the MLXBIR3SG–AVPW peptide complex (Figure 2, panel a) are conserved in the structure of the complex with peptide isostere 8 (Figure 2, panel b). For instance, the main chain–main chain hydrogen bonds noted in the AVPW complex structure are completely conserved in the structure of the complex with isostere 8. The alanine groups are also superimposable in the two structures, with the side chain methyl groups being buried in a hydrophobic pocket and the N-methyl group of isostere 8 being exposed to the solvent. The [7,5]-bicyclic moiety of isostere 8 forms a slightly larger hydrophobic interface with the protein, through contacts with Trp147 and Phe148, than the corresponding valine and proline residues in the AVPW peptide complex. Overall, however, the two complexes bury approximately the same total surface area, with the AVPW peptide complex burying 722 Å2 and the isostere 8 complex burying a total of 747 Å2. Mechanism of Action. To determine whether potent peptide isosteres 8 and 10 ( and the weak binding 11 as a control) can prevent physical interaction between ML-IAP and Smac, 293T cells were transiently transfected with Flag-tagged ML-IAP or empty vector and myctagged Smac (Figure 5). Following transfection, Smac autoprocesses to generate the mature processed form that is capable of binding to IAP proteins (Figure 5). Incubation of lysates with isosteres 8, 10, and 11 reduced ML-IAP–Smac binding in a dose-dependent manner (Figure 5). Incubation of cellular lysates with isosteres 8 or 10 was more efficient at blocking the ML-IAP–Smac interaction than was addition of isostere 11, in agreement with their respective binding affinities (Table 1). These results demonstrate that the peptide isosteres are capable of competitively blocking protein–protein interactions involving ML-IAP. Caspase Activation and Cell Killing. Antagonism of IAP proteins has been shown to stimulate apoptosis in cancer cell lines. To determine whether the peptide isosteres can induce apoptosis in cancer cells, MDA-MB231 breast cancer and A2058 melanoma cells were treated with isosteres 8, 10, and 11 (Figure 6, panels a VOL.1 NO.8 • 525–533 • 2006

529

increasing doses of doxorubicin was investigated (Figure 6, panel d). Isostere 8 shows potent single-agent activity and additivity with doxorubicin. Isostere 10 shows modest single-agent activity that also appears to be additive with the effect of doxorubicin. No significant activity or additivity with doxorubicin is seen with isostere 11.

Figure 4. Competition binding assays. Competition binding curves for Smac mimetics binding to the a) MLXBIR3SG and b) XIAP-BIR3 domains. See Methods for details of the fluorescence polarization-based binding assays. The Ki values (Table 1) were calculated from IC50 values as described (43).

and b). Similar to other structurally distinct Smac mimetics (34, 37), isosteres 8 and 10 exhibit single-agent killing of the treated cells with approximate IC50 values of 0.1 and 2 ␮M, respectively. Isostere 11 shows no detectable single-agent activity, in agreement with its lower affinity for the IAP-BIR domains. In addition, treatment of MDA-MB-231 cells with isostere 8 induced nuclear condensation and fragmentation further supporting its pro-apoptotic activity (data not shown). Treatment of the susceptible A2058 melanoma cells with isosteres 8 and 10 also results in activation of caspase-3 and -7 (Figure 6, panel c), suggesting that the single-agent cell killing observed for these compounds is due to the induction of apoptosis. Isostere 8 induced significant caspase-3/7 activation in the submicromolar concentration range, consistent with its higher activity in the cell viability assay. Isostere 10 shows significant caspase-3/7 activation in the low micromolar concentration range, which is again consistent with their observed potency in the cell viability assay. Smac-based peptides and Smac mimetics have been shown to sensitize cancer cells to a variety of chemotherapeutic agents. To determine whether the IAPbinding peptide isosteres might also collaborate with chemotherapeutic agents to stimulate cell death, cell viability of MDA-MB-231 cells treated with isosteres 8, 10, and 11 at 1 ␮M concentrations in the presence of 530

VOL.1 NO.8 • 525–533 • 2006

ZOBEL ET AL.

CONCLUSIONS In this study, a discrete and tight-binding peptide motif has been translated into a peptide isostere that has comparable binding affinity for relevant BIR domains of IAP proteins. The peptide isostere contains a [7,5]bicyclic lactam that comformationally constrains the P2–P3 portion of the molecule and a 5-membered heterocycle that optimally positions an aromatic ring in the P4 position. Whereas the tightest binding peptide discovered, AVPW, shows no cellular biological activity, the peptide isosteres with the preferred ring-junction stereochemistry disrupt key IAP protein–protein interactions, cause activation of effector caspases, and exhibit both single-agent cell killing and additivity with doxorubicin in cancer cell lines. Similar conformationally constrained Smac mimetics have been reported previously, although

Figure 5. Immunoprecipitation shows isosteres compete with Smac for the ML-IAP binding site. HEK293T cells were transiently transfected with Flag-tagged ML-IAP or vector plasmid and myc-tagged Smac. After 40 h transfection, cells were lysed in NP40 lysis buffer. Lysates were incubated with 0, 1, or 50 ␮M concentrations of the indicated peptide isosteres for 2 h, followed by immunoprecipitation with anti-Flag antibodies for 3 h, SDS-PAGE, and immunoblotting with anti-Flag (lower panel) and anti-Smac (upper panel) antibodies. The righthand panel shows the input amount of Smac protein from HEK293T cells co-transfected with ML-IAP or the empty vector. The right-most lane of each panel represents cells co-transfected with myc-tagged Smac and the empty vector. www.acschemicalbiology.org

ARTICLE

Figure 6. Cell killing and caspase activation experiments show single-agent activity and cooperation with doxorubicin. a) MDA-MB-231 breast carcinoma and b) A2058 melanoma cells were treated with indicated amounts of isosteres 8, 10, and 11 or DMSO control. Cell death was assessed by neutral red staining 30 h after the start of the treatment. c) A2058 cells were treated with indicated amounts of isosteres 8, 10, and 11 or DMSO control for 30 h, and caspase-3 and -7 activation was assessed using Z-DEVD-R110 as the fluorogenic substrate. d) MDA-MB-231 cells were treated with 1 ␮M of isosteres 8, 10, and 11 or DMSO control in the presence of indicated amounts of doxorubicin. Cell death was assessed as in panels a and b. NT ⴝ nontreated cells.

single-agent cell killing activity was not demonstrated for these compounds (36). In summary, these results demonstrate that small-molecule Smac mimetics can

overcome the inhibitory effects of the IAP proteins and that such compounds have promise as potential cancer therapeutics.

METHODS

Purified MLXBIR3SG protein (20 –25 mg mL⫺1) was mixed with peptide (AVPW) in a 1:2 protein/peptide ratio. The MLXBIR3SG– peptide complex was mixed with crystallization well solution (100 mM Bis–tris(hydroxymethyl)aminomethane (tris), pH 6; 200 mM lithium sulfate; 20 –25% (w/v) poly(ethylene glycol) 3350) in a ratio of 1 mL of protein complex to 1 mL of well solution. Hanging or sitting drops of the mixed solutions were then allowed to equilibrate by vapor diffusion against a reservoir of the well solution. Tetragonal bipyramidal crystals appear in a few days and grow to full size (typically 0.1– 0.2 mm long) in a week. Crystals of MLXBIR3SG complexed with AVPW were removed from the crystallization drop and transferred to a stabilizer drop (5 ␮L) containing 100 mM Bis–tris, pH 6; 200 mM lithium sulfate; 30% (w/v) poly(ethylene glycol) 3350; and 0.5–1.0 mM of compound 8. Crystals were left in the soaking solution overnight as a hanging drop over a reservoir of the same solution (to

Protein Production. The ML-IAP-BIR/XIAP-BIR3 chimeric protein, MLXBIR3SG, and the BIR3 domain of XIAP (XIAP-BIR3) were expressed and purified as described previously (27). cIAP1-BIR3/XIAP-BIR3 and cIAP2-BIR3/XIAP-BIR3 chimeric proteins, cIAP1XBIR3 and cIAP2XBIR3, respectively, were engineered, expressed, and purified in a similar fashion to MLXBIR3SG. The amino acid sequences of these BIR domains retain the native cIAP peptide-binding sites (Supplementary Figure 1). Purified proteins were concentrated and stored at ⫺80 °C. Compound Preparation. Isosteres 8–11 were synthesized as described in Supporting Information text and stored as solids at 4 °C. X-ray Crystallographic Analysis of MLXBIR3SG–Antagonist Complexes. The peptide was reconstituted from lyophilized powder in 10 mM 2-(N-morpholino)ethanesulfonic acid, pH 5.5; final peptide concentration (25 mg mL⫺1) was verified by A280.

www.acschemicalbiology.org

VOL.1 NO.8 • 525–533 • 2006

531

prevent evaporation). Crystals were then transferred to a cryostabilizer containing 100 mM Bis–tris, pH 6, 200 mM lithium sulfate; 30% (w/v) poly(ethylene glycol) 3350; 15% (v/v) ethylene glycol; and 0.5–1.0 mM of the same compound used in the soaking stage. After 15–20 min in the cryostabilizer, the crystals were frozen in liquid nitrogen. Crystals of the parent complex (AVPW) were prepared in the same manner, except that the AVPW peptide was substituted for the inhibitor compound in the soaking and cryostabilizer solutions. Data for the AVPW peptide complex were collected at the Advanced Photon Source (Argonne, IL), while data for the compound 8 complex were collected using an in-house X-ray source. Data statistics are listed in Supplementary Table 1. The starting models for refinement of these structures were derived from a 1.3 Å resolution structure of a different peptidomimetic complex (details to be published elsewhere), which was stripped of the peptidomimetic antagonist molecule and all water molecules within 10 Å of it. After one round of refinement of the antagonist-free model, the new antagonists were built into clear difference electron density visible in Fo – Fc maps. New water molecules were picked automatically using the program Arp/wArp (41), and the entire new complex models were subjected to several rounds of positional, anisotropic B-factor, and translation–libration–screw refinement using Refmac5 (42). Refinement statistics for the complex structures are in Supplementary Table 1. Binding Assays. Initial polarization experiments were performed in order to determine dissociation constants (Kd) between IAP protein BIR domains and fluorescent probes. Samples for fluorescence polarization affinity measurements were prepared by addition of serial dilutions of MLXBIR3SG, XIAP-BIR3, cIAP1XBIR3, or cIAP2XBIR3 in polarization buffer (50 mM Tris [pH 7.2], 120 mM NaCl, 1% bovine globulins, 5 mM dithiothreitol, and 0.05% octylglucoside) to 5 nM 5-carboxyflourescein (5-FAM)-conjugated AVP-diphenylalanine-AKK (AVPdiPhe-FAM). The reactions were read after an incubation time of 30 min at RT with standard cut-off filters for the fluorescein fluorophore (␭ex ⫽ 485 nm; ␭em ⫽ 530 nm) in 384-well black HE96 plates (Molecular Devices Corp.). Fluorescence polarization values were plotted as a function of the protein concentration, and the effective concentration 50 (EC50) values were obtained by fitting the data to a four-parameter equation using KaleidaGraph software (Synergy software, Reading, PA). The apparent Kd values were determined from the EC50 values. Inhibition constants (Ki) for the antagonists were determined by addition of 0.06 ␮M MLXBIR3SG, 0.5 ␮M XIAP-BIR3, 0.2 ␮M cIAP1XBIR3, or 0.4 ␮M cIAP2XBIR3 to wells containing 1:3 serial dilutions of the antagonists and 5 nM AVP-diPhe-FAM probe in the polarization buffer. Samples were read after a 30-min incubation. Fluorescence polarization values were plotted as a function of the antagonist concentration, and the IC50 values were obtained by fitting the data to a four-parameter equation using KaleidaGraph software. Ki values for the antagonists were determined from the IC50 values (43). Cellular Coimmunoprecipitation. HEK293T cells were transiently transfected with Flag-tagged ML-IAP or vector plasmid and myc-tagged Smac. Forty hours after transfection, cells were lysed in NP40 lysis buffer (44). Lysates were incubated with 0, 1, or 50 ␮M of the indicated compounds for 2 h, followed by immunoprecipitation with anti-Flag antibodies for 3 h, SDS-PAGE, and immunoblotting with anti-Flag and anti-Smac antibodies. Analysis of Apoptosis. Human breast carcinoma MDA-MB-231 and human melanoma A2058 cells were obtained from ATCC. Cells were grown in 50:50 Dulbecco’s modified Eagle’s and FK12 medium supplemented with 10% fetal bovine serum (FBS), penicillin, and streptomycin. Doxorubicin was purchased from Sigma. Cells ((1.5–2) ⫻ 104 cells per well) were seeded into

532

VOL.1 NO.8 • 525–533 • 2006

ZOBEL ET AL.

96-well dishes in media containing 5% heat-inactivated FBS. The medium was changed 8 –12 h later, and cells were treated with different compounds, either alone or in combination with doxorubicin, for 30 h. Cell viability was measured by neutral red uptake as described previously (45). Caspase Activation. Cells were seeded and treated with compounds, either alone or in combination with doxorubicin (as described above). Caspase-3/7 activity was measured 30 h later using the Apo-ONETM Homogeneous Caspase-3/7 assay kit (Promega) according to the manufacturer’s instructions. Accession codes: The coordinates have been deposited with the Protein Data Bank and are listed as follows: The AVPW structure is listed as 2I3H and the isostere 8 structure as 2I3I. Acknowledgments: We thank C. Quan, J. Tom, M. Stuble, and J. Dority and the Genentech DNA synthesis group for their valuable assistance. Note added after print publication: Because of a production error, the following references were misformatted: 1–39, 41– 44. These errors do not affect the scientific integrity of the article. This paper was originally posted September 15, 2006, and the electronic version was corrected and reposted to the web on October 20, 2006. An Addition and Correction may be found in ACS Chem. Biol. 1(9). Supporting information available: This material is available free of charge via the Internet.

REFERENCES 1. Aravind, L., Dixit, V. M., and Koonin, E. V. (1999) The domains of death: evolution of the apoptosis machinery, Trends Biochem. Sci. 24, 47–53. 2. Thompson, C. B. (1995) Apoptosis in the pathogenesis and treatment of disease, Science 267, 1456–1462. 3. Salvesen, G. S., and Dixit, V. M. (1997) Caspases: intracellular signaling by proteolysis, Cell 91, 443–446. 4. Salvesen, G. S., and Duckett, C. S. (2002) IAP proteins: blocking the road to death’s door, Nat. Rev. Mol. Cell. Biol. 3, 401–410. 5. Tamm, I., Kornblau, S. M., Segall, H., Krajewski, S., Welsh, K., Kitada, S., Scudiero, D. A., Tudor, G., Qui, Y. H., Monks, A., Andreeff, M., and Reed, J. C. (2000) Expression and prognostic significance of IAPfamily genes in human cancers and myeloid leukemias, Clin. Cancer Res. 6, 1796–1803. 6. Tamm, I., Richter, S., Scholz, F., Schmelz, K., Oltersdorf, D., Karawajew, L., Schoch, C., Haferlach, T., Ludwig, W. D., and Wuchter, C. (2004) XIAP expression correlates with monocytic differentiation in adult de novo AML: impact on prognosis, Hematol. J. 5, 489–495. 7. LaCasse, E. C., Baird, S., Korneluk, R. G., and MacKenzie, A. E. (1998) The inhibitors of apoptosis (IAPs) and their emerging role in cancer, Oncogene 17, 3247–3259. 8. Stennicke, H. R., Ryan, C. A., and Salvesen, G. S. (2002) Reprieval from execution: the molecular basis of caspase inhibition, Trends Biochem. Sci. 27, 94–101. 9. Li, J., Feng, Q., Kim, J. M., Schneiderman, D., Liston, P., Li, M., Vanderhyden, B., Faught, W., Fung, M. F., Senterman, M., Korneluk, R. G., and Tsang, B. K. (2001) Human ovarian cancer and cisplatin resistance: Possible role of inhibitor of apoptosis proteins, Endocrinology 142, 370–380. 10. Fong, W. G., Liston, P., Rajcan-Separovic, E., St Jean, M., Craig, C., and Korneluk, R. G. (2000) Expression and genetic analysis of XIAPassociated factor 1 XAF1) in cancer cell lines, Genomics 70, 113–122. 11. Cummins, J. M., Kohli, M., Rago, C., Kinzler, K. W., Vogelstein, B., and Bunz, F. (2004) X-linked inhibitor of apoptosis protein (XIAP) is a nonredundant modulator of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-mediated apoptosis in human cancer cells, Cancer Res. 64, 3006–3008.

www.acschemicalbiology.org

ARTICLE 12. Vucic, D., Stennicke, H. R., Pisabarro, M. T., Salvesen, G. S., and Dixit, V. M. (2000) ML-IAP, a novel inhibitor of apoptosis that is preferentially expressed in human melanomas, Curr. Biol. 10, 1359–1366. 13. Gong, J., Chen, N., Zhou, Q., Yang, B., Wang, Y., and Wang, X. (2005) Melanoma inhibitor of apoptosis protein is expressed differentially in melanoma and melanocytic naevus, but similarly in primary and metastatic melanomas, J. Clin. Pathol. 58, 1081–1085. 14. Crnkovic-Mertens, I., Hoppe-Seyler, F., and Butz, K. (2003) Induction of apoptosis in tumor cells by siRNA-mediated silencing of the livin/ML-IAP/KIAP gene, Oncogene 22, 8330–8336. 15. Crnkovic-Mertens, I., Semzow, J., Hoppe-Seyler, F., and Butz, K. (2006) Isoform-specific silencing of the Livin gene by RNA interference defines Livin beta as key mediator of apoptosis inhibition in HeLa cells, J. Mol. Med. 84, 232–240. 16. Rothe, M., Pan, M. G., Henzel, W. J., Ayres, T. M., and Goeddel, D. V. (1995) The TNFR2-TRAF signaling complex contains two novel proteins related to baculoviral inhibitor of apoptosis proteins, Cell 83, 1243–1252. 17. Dai, Y., Rahmani, M., and Grant, S. (2003) Proteasome inhibitors potentiate leukemic cell apoptosis induced by the cyclin-dependent kinase inhibitor flavopiridol through a SAPK/JNK- and NF-kappaBdependent process, Oncogene 22, 7108–7122. 18. Imoto, I., Tsuda, H., Hirasawa, A., Miura, M., Sakamoto, M., Hirohashi, S., and Inazawa, J. (2002) Expression of cIAP1, a target for 11q22 amplification, correlates with resistance of cervical cancers to radiotherapy, Cancer Res. 62, 4860–4866. 19. Imoto, I., Yang, Z. Q., Pimkhaokham, A., Tsuda, H., Shimada, Y., Imamura, M., Ohki, M., and Inazawa, J. (2001) Identification of cIAP1 as a candidate target gene within an amplicon at 11q22 in esophageal squamous cell carcinomas, Cancer Res. 61, 6629–6634. 20. Yang, L., Cao, Z., Yan, H., and Wood, W. C. (2003) Coexistence of high levels of apoptotic signaling and inhibitor of apoptosis proteins in human tumor cells: implication for cancer specific therapy, Cancer Res. 63, 6815–6824. 21. Shiozaki, E. N., Chai, J., Rigotti, D. J., Riedl, S. J., Li, P., Srinivasula, S. M., Alnemri, E. S., Fairman, R., and Shi, Y. (2003) Mechanism of XIAP-mediated inhibition of caspase-9, Mol. Cell 11, 519–527. 22. Srinivasula, S. M., Hegde, R., Saleh, A., Datta, P., Shiozaki, E., Chai, J., Lee, R. A., Robbins, P. D., Fernandes-Alnemri, T., Shi, Y., and Alnemri, E. S. (2001) A conserved XIAP-interaction motif in caspase-9 and Smac/DIABLO regulates caspase activity and apoptosis, Nature 410, 112–116. 23. Du, C., Fang, M., Li, Y., Li, L., and Wang, X. (2000) Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition, Cell 102, 33–42. 24. Verhagen, A. M., Ekert, P. G., Pakusch, M., Silke, J., Connolly, L. M., Reid, G. E., Moritz, R. L., Simpson, R. J., and Vaux, D. L. (2000) Identification of DIABLO, a mammalian protein that promotes apoptosis by binding to and antagonizing IAP proteins, Cell 102, 43–53. 25. Liu, Z., Sun, C., Olejniczak, E. T., Meadows, R. P., Betz, S. F., Oost, T., Herrmann, J., Wu, J. C., and Fesik, S. W. (2000) Structural basis for binding of Smac/DIABLO to the XIAP BIR3 domain, Nature 408, 1004–1008. 26. Wu, G., Chai, J., Suber, T. L., Wu, J. W., Du, C., Wang, X., and Shi, Y. (2000) Structural basis of IAP recognition by Smac/DIABLO, Nature 408, 1008–1012. 27. Vucic, D., Franklin, M. C., Wallweber, H. J., Das, K., Eckelman, B. P., Shin, H., Elliott, L. O., Kadkhodayan, S., Deshayes, K., Salvesen, G. S., and Fairbrother, W. J. (2005) Engineering ML-IAP to produce an extraordinarily potent caspase 9 inhibitor: implications for Smacdependent anti-apoptotic activity of ML-IAP, Biochem. J. 385, 11–20.

www.acschemicalbiology.org

28. Wilkinson, J. C., Wilkinson, A. S., Scott, F. L., Csomos, R. A., Salvesen, G. S., and Duckett, C. S. (2004) Neutralization of Smac/ Diablo by inhibitors of apoptosis (IAPs). A caspase-independent mechanism for apoptotic inhibition, J. Biol. Chem. 279, 51082–51090. 29. Eckelman, B. P., and Salvesen, G. S. (2006) The human antiapoptotic proteins cIAP1 and cIAP2 bind but do not inhibit caspases, J. Biol. Chem. 281, 3254–3260. 30. Vucic, D., Deshayes, K., Ackerly, H., Pisabarro, M. T., Kadkhodayan, S., Fairbrother, W. J., and Dixit, V. M. (2002) SMAC negatively regulates the anti-apoptotic activity of melanoma inhibitor of apoptosis (ML-IAP), J Biol Chem 277, 12275–12279. 31. Guo, F., Nimmanapalli, R., Paranawithana, S., Wittman, S., Griffin, D., Bali, P., O’Bryan, E., Fumero, C., Wang, H. G., and Bhalla, K. (2002) Ectopic overexpression of second mitochondria-derived activator of caspases (Smac/DIABLO) or cotreatment with N-terminus of Smac/DIABLO peptide potentiates epothilone B derivative-(BMS 247550) and Apo-2L/TRAIL-induced apoptosis, Blood 99, 3419–3426. 32. Arnt, C. R., Chiorean, M. V., Heldebrant, M. P., Gores, G. J., and Kaufmann, S. H. (2002) Synthetic Smac/DIABLO peptides enhance the effects of chemotherapeutic agents by binding XIAP and cIAP1 in situ, J Biol Chem 277, 44236–44243. 33. Fulda, S., Meyer, E., and Debatin, K. M. (2002) Inhibition of TRAILinduced apoptosis by Bcl-2 overexpression, Oncogene 21, 2283–2294. 34. Oost, T. K., Sun, C., Armstrong, R. C., Al-Assaad, A. S., Betz, S. F., Deckwerth, T. L., Ding, H., Elmore, S. W., Meadows, R. P., Olejniczak, E. T., Oleksijew, A., Oltersdorf, T., Rosenberg, S. H., Shoemaker, A. R., Tomaselli, K. J., Zou, H., and Fesik, S. W. (2004) Discovery of potent antagonists of the antiapoptotic protein XIAP for the treatment of cancer, J. Med. Chem. 47, 4417–4426. 35. Sun, H., Nikolovska-Coleska, Z., Chen, J., Yang, C.-Y., Tomita, Y., Pan, H., Yoshioka, Y., Krajewski, K., Roller, P. P., and Wang, S. (2005) Structure-based design, synthesis and biochemical testing of novel and potent Smac peptido-mimetics, Bioorg. Med. Chem. Lett. 14, 793–797. 36. Sun, H., Nikolovska-Coleska, Z., Yang, C.-Y., Xu, L., Liu, M., Tomita, Y., Pan, H., Yoshioka, Y., Krajewski, K., Roller, P. P., and Wang, S. (2004) Structure-based design of potent, conformationally constrained Smac mimetics, J. Am. Chem. Soc. 126, 16686–16687. 37. Li, L., Thomas, R. M., Suzuki, H., De Brabander, J. K., Wang, X., and Harran, P. G. (2004) A small molecule Smac mimic potentiates TRAIL- and TNFalpha-mediated cell death, Science 305, 1471–1474. 38. Franklin, M. C., Kadkhodayan, S., Ackerly, H., Alexandru, D., Distefano, M. D., Elliott, L. O., Flygare, J. A., Mausisa, G., Okawa, D. C., Ong, D., Vucic, D., Deshayes, K., and Fairbrother, W. J. (2003) Structure and function analysis of peptide antagonists of melanoma inhibitor of apoptosis (ML-IAP), Biochemistry 42, 8223–8231. 39. Alberg, D. A., and Schreiber, S. L. (1993) Structure-based design of a cyclophilin-calcineurin bridging ligand, Science 262, 248–250. 40. Patchett, A. A., and Wyvratt, M. J. (1986) U.S. Patent 4,617,301. 41. Perrakis, A., Harkiolaki, M., Wilson, K. S., and Lamzin, V. S. (2001) ARP/wARP and molecular replacement, Acta Crystallogr. D57, 1445–1450. 42. Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997) Refinement of macromolecular structures by the maximum-likelihood method, Acta Crystallogr. D53, 240–255. 43. Keating, S. M., Marsters, J., Beresini, M., Ladner, C., Zioncheck, K., Clark, K., Arellano, F., and Bodary, S. (2000) Putting the pieces together: Contribution of fluorescence polarization assays to smallmolecule lead optimization, Proc. SPIE 3913, 128–137. 44. Vucic, D., Kaiser, W. J., Harvey, A. J., and Miller, L. K. (1997) Inhibition of Reaper-induced apoptosis by interaction with inhibitor of apoptosis proteins (IAPs), Proc. Natl. Acad. Sci. U.S.A. 94, 10183–10188. VOL.1 NO.8 • 525–533 • 2006

533

45. Johnston, M. D., Finter, N. B., and Young, P. A. (1981 Methods Enzymol. 78, 394–399. 46. DeLano, W. L. (2002) The PyMOL Molecular Graphics System, DeLano Scientific, San Carlos, CA.

534

VOL.1 NO.8 • 525–533 • 2006

ZOBEL ET AL.

www.acschemicalbiology.org