A binding motif for Siah ubiquitin ligase - Semantic Scholar

A binding motif for Siah ubiquitin ligase Colin M. House*, Ian J. Frew*, Huei-Luen Huang†, Gerhard Wiche†, Nadia Traficante*, Edouard Nice‡, Bruno Catimel‡, and David D. L. Bowtell*§ *Trescowthick Research Laboratories, Peter MacCallum Cancer Institute, Melbourne 8006, Victoria, Australia; †Institute of Biochemistry and Molecular Cell Biology, Vienna Biocenter, A-1030 Vienna, Austria; and ‡Ludwig Institute for Cancer Research, Parkville 3152, Victoria, Australia Edited by Alexander Varshavsky, California Institute of Technology, Pasadena, CA, and approved January 6, 2003 (received for review August 8, 2002)

P

roteasomal degradation of proteins requires recognition of a polyubiquitin signal on the targeted protein. Ubiquitination is a multistep process that involves at least three classes of transfer proteins, E1 (ubiquitin activating proteins), E2 (ubiquitin conjugating proteins), and E3 (ubiquitin ligases) (1). A subset of E3s transfer ubiquitin from the E2 directly to the substrate. These comprise both single subunit and multiprotein complexes, which are characterized by the presence of a RING (really interesting new gene) domain. Recent structures of a cCbl-UbcH7 complex (2) and an SCF complex (3) suggest RING domain proteins function as part of the scaffold to optimally position substrate and E2 for transfer of ubiquitin. Members of the highly conserved SINA (seven in absentia)/ Siah (seven in absentia homologue) family of proteins contain a RING domain and function as E3 ligases (4). This protein family was first defined in Drosophila, where SINA is required for R7 cell determination in the developing eye, downstream of the Sevenless兾Ras pathway (5). Genetic and biochemical evidence support a model where an E3 comprising SINA兾PHYL兾EBI interacts with the transcriptional repressor TTK88 and the ubiquitin-conjugating enzyme UBCD1, leading to the ubiquitination and proteasomal degradation of TTK88 (6–9). The Drosophila SINA E3 complex is the best described, both genetically and biochemically, suggesting that it can provide clues to the function of mammalian Siah proteins. SINA兾Siah sequences are highly conserved from plants to mammals. Whereas the N terminus and RING domain of Siah bind E2 proteins (10) (11), the C terminus can be considered as a substrate- and cofactor-interaction domain (substrate-binding domain, SBD) that interacts with a number of proteins, some of which are degraded. Degraded proteins include netrin-1 receptor兾deleted in colorectal cancer, DCC (10); the nuclear receptor corepressor, N-CoR (12); the motor protein, Kid (13); the transcriptional activator, OBF-1 (14, 15); the developmental regulator, NUMB (16); the neural transmitter protein, synaptophysin (17); and the transcriptional repressor, TIEG-1 (18). In these cases, Siah may function alone as a targeting, single subunit E3 ligase, but Siah has also been shown to interact in an SCF-type complex including Skp1, Ebi, Siah interacting protein (SIP), and adenomatous polyposis coli protein (pAPC) to facilitate the degradation of ␤-catenin in a p53-dependent manner www.pnas.org兾cgi兾doi兾10.1073兾pnas.0534783100

(19, 20). Both SIP and pAPC interact with the C terminus of Siah, although no direct interaction with the substrate, ␤-catenin, was reported. Siah’s ability to act as single E3 ligase and also to participate in a variant SCF complex is very unusual (reviewed in ref. 21) and highlights the importance of understanding how Siah SBD interacts with its partners. We have previously focused on the Siah SBD and showed that the crystal structure of that domain displays a fold similar to the C-terminal domain of tumor necrosis factor receptor associated factor proteins (22). Given the diverse interactions of the Siah SBD with a range of cellular proteins, we have sought to define the molecular basis of these interactions. Here we describe a high-affinity binding peptide, present in the Drosophila protein PHYL, which binds with high affinity to the SINA and Siah SBDs. Mutagenesis of this peptide has revealed a binding motif that is conserved and functional in diverse Siah-interacting proteins. Materials and Methods Plasmid Construction. Mouse Siah1a, Siah2, and Drosophila SINA (full length and the SBDs, lacking the N termini and RING domains) were cloned into the bacterial expression vector pMalC2 (New England Biolabs) at the BamHI and HindIII sites, utilizing sequence specific oligonucleotide primers and PCR amplification. For Siah1a, the SBD consisted of residues 80–282, for Siah2, residues 116–325 and for SINA, residues 108–314. Fragments of PHYL, DCC, and SIP (defined in the text) were cloned and expressed by using pGEX2T (Amersham Pharmacia). The Kid H16 construct, in pGEX-4T2, was a gift from A. Germani and F. Calvo (Saint-Louis Hospital, Paris). GSTTIEG-1 expression vector was a gift from Steven Johnsen and Thomas Spelsberg (Mayo Clinic and Foundation, Rochester, NY). Mutagenesis of constructs was performed by using the QuikChange site-directed mutagenesis kit (Stratagene). Protein Expression. Proteins were expressed as GST or maltosebinding protein (MBP) fusions in Escherichia coli BL21(DE3) cells at 22°C for 5 h. Cells were lysed and sonicated (three times for 30 sec on ice) in 50 mM Tris, pH 8.0兾200 mM NaCl兾15 mM 2-mercaptoethanol (␤-ME)兾0.2 mg/ml lysozyme兾0.5% Triton X-100兾10 ␮g/ml leupeptin兾10 ␮g/ml aprotinin兾1 ␮g/ml pepstatin兾0.5 mM PMSF before purification with either amylose (for MBP proteins) or glutathione (for GST proteins) on Sepharose-4B solid supports. MBP-fusion proteins were eluted with 10 mM maltose in 50 mM Tris, pH 8.0兾200 mM NaCl兾15 mM ␤-ME. For Biacore analysis, MBP-Siah-SBD, MBP-Sina-SBD and Siah-SBD were further purified before kinetic studies by using size exclusion chromatography (Superose 12 HR 3.2兾30, Amersham Pharmacia) equilibrated in 10 mM Hepes, pH 7.4, containing 3.4 mM EDTA, 0.15 mM NaCl, and 0.005% (vol兾vol) Tween 20 (HBS). The protein concentration was determined by absorbance at 280 nm using an extinction coefficient calculated

This paper was submitted directly (Track II) to the PNAS office. Abbreviations: SBD, substrate-binding domain; MBP, maltose-binding protein; pAPC, adenomatous polyposis coli protein; SIP, Siah interacting protein. §To

whom correspondence should be addressed. E-mail: [email protected].

PNAS 兩 March 18, 2003 兩 vol. 100 兩 no. 6 兩 3101–3106

BIOCHEMISTRY

The Drosophila SINA (seven in absentia) protein and its mammalian orthologs (Siah, seven in absentia homolog) are RING domain proteins that function in E3 ubiquitin ligase complexes and facilitate ubiquitination and degradation of a wide range of cellular proteins, including ␤-catenin. Despite these diverse targets, the means by which SINA兾Siah recognize substrates or binding proteins has remained unknown. Here we identify a peptide motif (RPVAxVxPxxR) that mediates the interaction of Siah protein with a range of protein partners. Sequence alignment and mutagenesis scanning revealed residues that are important to this interaction. This consensus sequence correctly predicted a high-affinity interaction with a peptide from the cytoskeletal protein plectin-1 (residues 95–117). The unusually high-affinity binding obtained with a 23-residue peptide (KDapp ⴝ 29 nM with SINA) suggests that it may serve as a useful dominant negative reagent for SINA兾Siah proteins.

from the amino acid composition. GFP fusions of plectin exons 1 and 1c were stably expressed in Chinese hamster ovary cells. Total cell lysates, in a buffer of 50 mM Tris, pH 7.5兾0.1M NaCl兾5% glycerol兾2.5 mM sodium orthovanadate兾2.5 mM sodium metavanadate兾0.1% Triton X-100兾0.1% Nonidet P-40兾 0.1% sodium deoxycholate兾1 mM PMSF were used for interaction studies with GST-Siah proteins. Protein–Protein Interaction Studies (Coprecipitation). GST fusions of Siah-binding proteins were expressed as described above and used in binding assays with MBP-Siah-SBD. A total of 0.2–2 ␮g of each protein, bound to GSH beads, was incubated with 3 ␮g/ml MBP-Siah-SBD in 50 mM Tris-HCl, pH 8.0兾200 mM NaCl兾15 mM 2-mercaptoethanol兾0.1% Nonidet P-40 (TNBN) at 4°C for 1 h. In cases where binding was competed with free PHYL peptide, 20 ␮M peptide was added to the MBP-Siah-SBD for 15 min before the 1 h incubation. After washing four times in 1 ml of TNBN, the beads and bound proteins were recovered by centrifugation and boiled in 100 ␮l of Laemmli sample buffer before separation by SDS兾PAGE. For Siah interactions with plectin exons 1 and 1c (as GFP fusions), the cell lysis buffer (see Protein Expression) was used and interaction was overnight at 4°C. Detection of protein–protein interaction was achieved either by Coomassie staining of gels or Western blotting (using a rabbit anti-MBP polyclonal antibody raised in our laboratory or a monoclonal anti-GFP antibody from Roche Diagnostics). Westerns were developed by using horseradish peroxidaseconjugated secondary antibody and enhanced chemiluminescence (Amersham Pharmacia). Peptide Synthesis. Peptides were synthesized by using Fmoc

chemistry on a 96-well format (Mimotopes, Clayton, Victoria, Australia) and cleaved from the solid phase support. Peptides were dissolved in DMSO before dilution in H2O or PBS兾0.1% Tween 20. Purity and quantitation of peptides was assessed by mass spectrometry and reversed-phase HPLC. Purity of peptides was above 70% in all cases and consistent across the mutagenesis set. For Biacore analysis, peptides were purified immediately before immobilization by using a C18 micropreparative reversephase high-pressure liquid chromatography column and analyzed by matrix-assisted laser desorption兾ionization time-offlight-mass spectrometry (Kratos IV). Peptide Binding Assay. Synthetic peptides were synthesized with an

N-terminal biotin residue and captured (100 ng兾well) on ELISA plates coated with Neutravidin (Pierce, 0.5 ␮g兾well). MBP-SiahSBD (1 ␮g兾ml) was bound to the peptides for 20 min at room temperature and quantitated by using anti-MBP antibodies, horseradish peroxidase-goat-anti-rabbit secondary antibody (Sigma) and 2,2⬘-azino-bis(3-ethyl)benzthiazoline-6-sulfonic acid as substrate, monitored at 405 nm after 10–15 min. MBPSiah-SBD protein bound to mutant peptides was compared with the protein bound to the PHYL108 –130 parent peptide. Protein Alignment. Siah-binding fragments of various proteins,

corresponding to the published minimal interacting fragments, were aligned by using the CLUSTALW program (23). The alignment was initiated by using short PHYL and OBF-1 sequences.

Biacore Analysis. Surface plasmon resonance analyses were performed by using a BIAcore 2000 Biosensor (Uppsala). Purified, N-terminally biotinylated PHYL108 –130 and plectin-195–117 peptides were immobilized onto a Neutravidin sensor surface (N-hydroxy-succinimide-activated CM5 carboxymethylated dextran sensor chip) according to described protocols (24, 25). After peptide immobilization, the surface was washed with 10 mM HCl until a stable level of immobilized peptide was achieved. Binding data were generated by injecting 30 ␮l of varying 3102 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0534783100

concentrations of proteins (25–1,000 nM) over the sensor surface at a flow rate of 10 ␮l兾min. After completion of the injection phase, dissociation was monitored in HBS buffer for 300 s at the same flow rate. Bound proteins were eluted, and the surface was regenerated between injections with 10 mM HCl. Regeneration conditions did not denature the immobilized antigen as shown by equivalent signals on reinjection of a ligand-containing sample. The apparent association (ka) and dissociation (kd) rate constants were calculated as described previously (26) (27) by nonlinear least squares regression by using BIAEVALUATION version 3.0 software into which the appropriate iterative curvefitting equations have been installed. When applicable, the affinity constant KD was also determined by equilibrium binding analysis where KD is obtained from the reciprocal of the slope of the graph (KA) obtained by plotting the biosensor data in Scatchard format [(Req兾nC) versus Req, where Req is the biosensor response at equilibrium, n the valency and C the concentration] (26, 28). Calculations for the goodness of fit are described in Supporting Materials and Methods, which is published as supporting information on the PNAS web site, www.pnas.org. Results A 22-aa Peptide Is Sufficient to Mediate PHYL Binding to Siah and SINA. Although a number of binding partners have been defined

for SINA or Siah proteins, we found that fragments of PHYL bound avidly to both SINA and Siah in coprecipitation experiments (data not shown). Additionally, a previous report mapped a SINA-binding domain to the vicinity of residues 108–130 (29). The PHYL protein therefore served as a useful starting point for analysis of Siah and SINA binding partners. We tested the binding of a protein deletion series of PHYL, expressed as GST-fusions, to the Siah1a SBD by using an in vitro pull-down assay. A fragment corresponding to amino acids 108–130 of PHYL was sufficient for the Siah binding under these conditions (Fig. 1A). The MBP-Siah-SBD fused protein bound equally well to full-length PHYL (1–400) and fragments corresponding to residues 1–130, 1–198 or 108–130 (Fig. 1 A, lanes 9–12). MBP alone did not bind (Fig. 1 A, lanes 2–6). These results suggested that the peptide 108–130 is sufficient for maximal binding of PHYL to the Siah兾SINA family of proteins. Mapping the Siah Interaction Site in Kid. A C-terminal deletion analysis of the Kid protein was also performed, because it was previously reported that Siah-binding resided in the C-terminal residues 404–665 (13). Siah binding was observed for the constructs 1–665, 1–560, and 1–540, but was substantially reduced in the 1–530 construct and shorter fragments (Fig. 1B). This result suggests that at least part of the Siah binding is present in the Kid residues 531–540. Identification of a Functionally Conserved Binding Motif in Siah-SBD Binding Proteins. On the basis of the avid binding between Siah

SBD and the PHYL108 –130 peptide, and the mapping of the Siah-binding region in Kid, we sought to ascertain whether elements of these peptides were present in other Siah SBDbinding proteins. Beginning with PHYL peptide and the region around Val-51 in OBF-1, alignment with fragments of reported Siah SBD-binding partners demonstrated the presence of a possible binding motif present in many of these proteins (Fig. 2). The core sequence PxAxVxP was found in the Siah interacting proteins SIP, OBF-1, DCC, and TIEG1, with more degenerate consensus sequences found in NUMB, EF1-␦, Vav, Kid, N-CoR, and FIR. The remaining Siah-SBD-interacting proteins, including pAPC, synaptophysin, mGlutR1, and ␣-tubulin, did not align when this method was used. To investigate whether the putative binding motif was functional in these proteins, we expressed protein fragments of DCC House et al.

Fig. 1. (A) Binding of PHYL to Siah maps to amino acids 108 –130 of PHYL. Interaction between GST-PHYL fragments and a soluble fusion protein MBP-Siah-SBD was investigated in an in vitro GST pull-down. PHYL constructs were mixed with MBP alone (lanes 2– 6) and MBP-Siah-SBD (lanes 8 –12). Lane 1 shows the MBP alone input, and lane 7 shows the MBP-Siah-SBD input. The gels were stained with Coomassie blue. The MBP-Siah-SBD bound to the PHYL fragments is highlighted with an asterisk. (B) Binding of GST-Kid C-terminal deletion mutants to MBP-Siah SBD. Experiment was as in A, although bound proteins were detected by Western blotting using anti-MBP antibody. MBP alone did not bind Kid fragments (data not shown). The full-length GST-Kid is highlighted with an asterisk. The binding兾protein ratio was determined by using densitometry of the bands, excluding the nonspecific bands highlighted by dashes.

Scanning Mutagenesis of PHYL108 –130 Peptide Defines Residues Required for Interaction with Siah. Because PHYL108 –130 pep-

tide bound more strongly to Siah SBD than did any of the other Siah-interacting proteins, we sought to define further binding determinants within the PHYL peptide. A full alanine mutagenesis scan was performed to determine key residues in PHYL108 –130 required for the interaction with Siah SBD. A set

Fig. 2. Presence of a common motif in many Siah-SBD binding proteins. Alignment of PHYL108 –130 with fragments of proteins previously reported to interact with Siah-SBD. Totally conserved residues are highlighted in black and residues conserved in ⬎60% of sequences are in gray. ¶, Y. Hu and D.D.L.B., unpublished data.

House et al.

of biotinylated peptides was synthesized in which alanine was substituted at each position across the PHYL108 –130 peptide. Where alanine occurred in the native sequence, glycine was substituted. The peptide set was captured on a Neutravidincoated ELISA plate, MBP-Siah-SBD was bound directly and measured by using an anti-MBP antibody. Solution-phase binding of Siah-SBD was almost eliminated by substitution of either the Val-120 and Pro-122 residues (Fig. 4A). Whereas mutagenesis of the conserved Pro-116 and Ala-118 residues partially reduced binding, the effects were not as strong as those seen for mutagenesis of Arg-115, Val-117, and Arg-125, which reduced binding by ⬇50% (Fig. 4A). In addition to the alanine mutagenesis, residues in the central portion of the peptide were altered to residues with opposite characteristics but similar bulk, that is, charge reversal, or hydrophobic to hydrophilic etc. The results reinforced those obtained with alanine mutants, although in general the inhibition of binding was even more pronounced. For example, Met-119Gln, in which a large polar residue (Gln) was less well tolerated than Ala, and Arg-121Glu, where the negative Glu was more inhibitory than Ala substitution (Fig. 4B). As before, mutation of Val-120, Pro-122, Arg-115, and Val-117 were most important for strong binding. Database Searches with the Siah Binding Motif Correctly Predict Interaction with a Plectin-1 Peptide. PHYL protein has no mam-

malian ortholog, raising the possibility that another protein may form a high-affinity interaction with Siah in mammalian cells. Combining the information from the alignment of Siah-binding proteins (P116xAxVxP122) and the mutagenesis scan of the PHYL peptide (R115xVxxVxPxxR125), to generate a consensus motif, we searched protein databases to identify other potential Siahinteracting proteins. A large number of proteins were identified that matched the consensus to varying degrees, including the cytoskeletal linker protein, plectin-1 (31), which was the only protein with a perfect match (A95SLQRVRRPVAMVMPARRTPHVQ117) to the proposed motif. We synthesized the plectin-1 peptide (residues 95–117) and found that it bound Siah SBD efficiently. Moreover, binding was competed by free PHYL peptide, indicating that the plectin-195–117 peptide also recognized the same site on the Siah SBD as did PHYL (Fig. 6, which is published as supporting information on the PNAS web site). GST-Siah SBD efficiently bound a fused protein of mouse plectin-1 (32) residues 1-180(exon 1)-GFP expressed in Chinese hamster ovary cells but not plectin 1c residues 1-66(exon1c)GFP, a splice variant that lacks the consensus binding motif (Fig. 5K). The PHYL108 –130 and Plectin-195–117 Peptides Bind SINA-SBD and SiahSBD with Nanomolar Affinity. Biosensor analysis was used to

measure the affinity of interactions between peptides and Siah兾 PNAS 兩 March 18, 2003 兩 vol. 100 兩 no. 6 兩 3103

BIOCHEMISTRY

(1203–1364), Kid (404–665), and SIP (1–77) reported to contain necessary sequences for Siah SBD interaction (10, 13, 19) and tested their binding to Siah. Interaction of each fragment with Siah SBD was observed (Fig. 3A) and was competed with free PHYL peptide, suggesting that these bound the same site on Siah SBD. Demonstration that this motif was sufficient for binding was obtained by showing that a 24 residue peptide of DCC (1324–1347), which encompassed the motif present in DCC, efficiently bound Siah-SBD. This interaction was also abrogated by free competing PHYL peptide (Fig. 3A). To further investigate the role of the motif in these proteins, mutagenesis of the VxP triplet to NxN was performed. The results obtained were protein-dependent. Mutation of the Val and Pro residues abrogated binding of the full-length proteins PHYL and TIEG-1, and the short DCC peptide (1324–1347) (Fig. 3B), and binding was reduced for DCC (1203–1364) and SIP (1–77) fragments. Although interaction of Kid with Siah mapped to the consensus motif (Figs. 1B and 2), mutation of the VxP motif did not affect Siah-binding (Fig. 3B). These findings demonstrate the importance of the consensus binding motif in mediating interaction of a diverse range of proteins with Siah, but indicate that the contribution of specific residues to binding varies.

Fig. 3. Consensus motif confers binding of protein partners to Siah SBD. (A) Relative binding of Siah SBD to fragments of DCC (1203–1364), Kid (404 – 665), SIP (1–77), PHYL (108 –130), and DCC (1324 –1347) was assessed in an in vitro GST pull-down assay, in the absence and presence of 20 ␮M free PHYL108 –130 peptide. Note that the loading on to the gel for the Western detection of the PHYL interaction was 10% of that for the other proteins, as this interaction was avid and gave an extremely strong signal. For this reason the GST-PHYL108 –130 protein was undetectable when probed with anti-GST to show relative loading of solid-phase proteins. The amount of GST-PHYL108 –130 fusion protein used in the experiment was approximately equal to that of the GST-Kid (404 – 665). (B) Effect of mutation of the VxP motif to NxN in a number of GST-fused interacting proteins, assessed by the binding of these proteins to MBP-Siah-SBD. Loading of the PHYL interaction products is 10% of that loaded for the other interactions. The relative loading was assessed by anti-GST immunoblot or Coomassie staining (TIEG-1FL and KidFL). In both panels, the migration of the solid-phase binding proteins are highlighted by an asterisk in the loading control Westerns.

SINA proteins. The results show that both of the peptides, PHYL108 –130 and plectin-195–117, interacted with MBP-SINASBD with higher affinity than MBP-Siah-SBD (Fig. 5 and Table 1). The apparent KD values were as low as 29 nM for MBPSINA-SBD interacting with the plectin-195–117 peptide and 92– 123 nM for the MBP-Siah-SBD interaction with the same peptide. The peptide interactions with the isolated Siah-SBD were of a similar affinity to those seen for the fusion protein MBP-Siah-SBD, showing the validity of working with fusion proteins in these experiments and the mutagenesis study. Further information concerning the biosensor data (association and dissociation constants) can be found in Supporting Materials and Methods.

Fig. 4. Scanning mutagenesis of PHYL peptide identifies core residues required for Siah-SBD binding. Immobilized PHYL mutant peptides were tested for their ability to bind MBP-Siah-SBD in an ELISA-based assay. Binding is compared with the PHYL108 –130 parent peptide. The results are averages of five separate experiments, showing SEM. (A) Alanine mutagenesis of individual residues in the PHYL108 –130 peptide. (B) Altering the charge and hydrophobicity characteristics of individual residues in the core of the PHYL peptide. 3104 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0534783100

Discussion The Siah兾SINA family of proteins has been reported to interact, through their SBDs, with various proteins, some of which are degraded in Siah overexpression experiments. Although it is important to know how the Siah SBD recognizes target proteins, the mapping of interacting domains in these proteins has not elucidated any common domain or binding mechanism. It is not known whether substrates and other partners interact at similar sites on Siah. Only Siah SBD interactions with PHYL (29), mGlutR1 (31), and synaptophysin (17) have mapped the interaction to peptide fragments of the interacting proteins. We have not investigated binding to synaptophysin, but we could not detect Siah SBD binding to mGlutR1 peptide under ELISA or Biacore conditions (data not shown). In contrast, we have found that the PHYL108 –130 peptide interacts with SINA and Siah with a low nanomolar affinity. Alignment studies and a mutagenesis study of the PHYL peptide have revealed a potential Siah-binding motif, R115PVAxVxPxxR125. The most conserved residues in this motif appear to be Val-120 and Pro-122, and indeed mutagenesis of both of these residues abrogates Siah binding. Identification of the PHYL peptide sequence recognized by SINA兾Siah provided an essential guide to identify potential interacting domains in other known Siah-binding proteins. We investigated some of these interactions and found that the observed binding could be competed by free PHYL peptide and that mutagenesis of the VxP motif in the Siah-binding fragments reduced binding, although not in all proteins. These results suggest that the proteins, whether substrates or interacting partners, are binding at or near the PHYL binding site. It was previously reported that Siah SBD interacted with residues 1–101 of the transcriptional activator OBF-1 (14, 15). Moreover, in a random mutagenesis screen designed to investigate OBF-1 and Oct-1 binding, a Val-51 to Glu substitution abrogated binding to Siah SBD (14, 15). Val-51 lies within the sequence P47TAVV51LP53 (Fig. 2), suggesting that OBF-1 also binds Siah through a VxP motif. Also of interest is that whereas pAPC did not align with the PHYL motif when the CLUSTALW program was used, visual inspection has found a V2777AARVTPFNY2786 sequence, within the Siah SBD interacting region (20), that includes part of the PxAxVxP motif and is thus a potential binding site for Siah binding. Siah House et al.

Fig. 5. Biosensor analysis of the interactions of MBP-SINA-SBD, MBP-Siah-SBD, and Siah-SBD with immobilized PHYL108 –130 and plectin-195–117 peptides. Biotinylated peptides were immobilized onto a neutravidin sensor surface as described in Materials and Methods. Varying concentrations of MBP-SINA-SBD (25– 800 nM) were injected over immobilized PHYL108 –130 (A) and immobilized plectin-195–117 (B). MBP-Siah-SBD (50 –1000 nM) was injected over immobilized PHYL108 –130 (C) and immobilized plectin-195–117 (D). Siah-SBD (300 –1,000 nM) was injected over immobilized PHYL108 –130 (E) and immobilized plectin-195–117 (F). The sensorgrams shown have controls subtracted, in which the sample was passed over a control neutravidin surface. Not all traces are shown, to aid clarity. An equilibrium binding analysis of the interaction between MBP-Siah-SBD and Siah-SBD and immobilized peptides PHYL108 –130 and plectin-195–117 was performed for the following interactions: MBP-Siah-SBD and PHYL108 –130 (G), MBP-Siah-SBD and plectin-195–117 (H), Siah-SBD and PHYL108 –130 (I), Siah-SBD and plectin-195–117 (J). KD was obtained from the reciprocal of the slope of the graph (KA) obtained by plotting the biosensor data in Scatchard format [(Req兾nC) versus Req, where Req is the biosensor response at equilibrium, n is the valency, and C is the concentration). The KD values are tabulated in Table 1. (K) Plectin exon1 binds Siah in vitro. GST fusions of Siah1 and Siah2 were used to pull-down recombinant plectin exon 1 or exon 1c-GFP. Bound protein was detected by anti-GFP Western blotting. Lane 1, plectin exon1-GFP; lane 2, plectin exon1c-GFP; lane 3, GST plus exon1-GFP; lane 4, GST-Siah1a plus plectin exon1-GFP; lane 5, GST-Siah2 plus plectin exon1-GFP; lane 6, GST-Siah1a plus plectin exon1c-GFP; lane 7, GST-Siah plectin plus plectin exon1c-GFP. Loading controls are shown in Lower.

is a dimeric protein (22), and as such may be able to accommodate both SIP and pAPC binding to equivalent sites on each monomer. The Siah兾PHYL interaction that we observe is far stronger than the interaction between Siah and the other reported interacting proteins that we have tested, including DCC, Kid, OBF-1, SIP, mGlutR1 and pAPC. DCC, OBF-1 and SIP each have a perfect match to the alignment consensus PxAxVxP and this region has been shown to be important for interaction with Siah (this paper and refs. 14 and 15). It is possible that the binding of Siah to proteins such as SIP or OBF-1 may be stabilized by factors or modifications not found in the bacterial expression system used here. One possible modification of the PHYL peptide is proline hydroxylation, recently shown to be required in hypoxia inducible factor-1␣ for recognition by the E3 component, von Hippel-Lindau protein (32–34). We tested the effect of hydroxylation of each proline residue in the SIP peptide (P60 and P66 in AELLDNEKP60AAVVAP66ITTGYTVKI), but

this did not increase binding in the ELISA-based binding assay (data not shown). Clearly there are residues within PHYL108 –130, other than the core PxAxVxP, that confer high-affinity binding. Although the alanine mutagenesis scan did not identify individual residues outside the core PxAxVxP that make contributions greater than the VxP residues, mutation of several flanking residues, including Arg-115, Val-117, and Arg-125, did reduce binding substantially. The notion that the combined interaction of residues flanking the core consensus may contribute to a high-affinity interaction with Siah-SBD was supported by the very strong interaction we identified with the plectin-195–117 peptide (78–123 nM), which contains a perfect match with the RPVAxVxPxxR consensus. SINA is part of a complex with EBI and PHYL targeting TTK88, but not PHYL, for degradation. Similarly, Siah1 is part of an E2兾E3 complex involving SIP, pAPC and an F-Box protein, Ebi, to target the degradation of ␤-catenin (19). Although PHYL

Soluble analyte

Immobilized ligand

MBP-SINA-SBD

PHYL108–130 plectin-195–117 PHYL108–130

MBP-Siah-SBD

plectin-195–117 Siah-SBD

PHYL108–130 plectin-195–117

Analysis

ka, ⫻10⫺4 M⫺1䡠s⫺1

kd, ⫻103 s⫺1

KD, ⫻10⫺9 M

NLLS NLLS NLLS Equilibrium NLLS Equilibrium NLLS Equilibrium NLLS Equilibrium

4.43 4.59 2.82

2.16 1.32 5.65

4.43

4.09

48.8 28.7 203 226 92.3 123 176 183 89.7 78.2

6.30 7.40

11.1 6.64

NNLS, nonlinear least squares regression.

House et al.

PNAS 兩 March 18, 2003 兩 vol. 100 兩 no. 6 兩 3105

BIOCHEMISTRY

Table 1. Biosensor kinetic analysis of the interaction among MBP-SINA-SBD, MBP-Siah-SBD, and Siah-SBD with immobilized PHYL108 –130 and plectin-195–117

and SIP appear to bind to the same site on Siah, the affinities are very different. It may be that a strong interaction of Siah with its substrate is only achieved in the mammalian setting through the regulated formation of a multiprotein complex. The plectin-195–117 and PHYL108 –130 peptides bind to SiahSBD with apparent affinity constants in the low nanomolar range, several orders of magnitude lower than those observed between SH3 domains and polyproline sequences (35). Given that this interaction maps to a relatively short peptide and involves a small number of key residues it should be possible to generate small molecule inhibitors that block interaction with Siah and a range of protein partners. Such inhibitors would be of considerable value for investigating the biology of this novel protein family and may have therapeutic uses.

Note Added in Proof. During the preparation of this manuscript, Li et al. (38) reported a strong interaction between SINA and PHYL109–127, with mutagenesis results consistent with those reported in this manuscript.

1. Ciechanover, A. (1998) EMBO J. 17, 7151–7160. 2. Zheng, N., Wang, P., Jeffrey, P. D. & Pavletich, N. P. (2000) Cell 102, 533–539. 3. Zheng, N., Schulman, B. A., Song, L., Miller, J. J., Jeffrey, P. D., Wang, P., Chu, C., Koepp, D. M., Elledge, S. J., Pagano, M., et al. (2002) Nature 416, 703–709. 4. Lorick, K. L., Jensen, J. P., Fang, S., Ong, A. M., Hatakeyama, S. & Weissman, A. M. (1999) Proc. Natl. Acad. Sci. USA 96, 11364–11369. 5. Carthew, R. W. & Rubin, G. M. (1990) Cell 63, 561–577. 6. Tang, A. H., Neufeld, T. P., Kwan, E. & Rubin, G. M. (1997) Cell 90, 459–467. 7. Li, S., Li, Y., Carthew, R. W. & Lai, Z. C. (1997) Cell 90, 469–478. 8. Dong, X., Tsuda, L., Zavitz, K. H., Lin, M., Li, S., Carthew, R. W. & Zipursky, S. L. (1999) Genes Dev. 13, 954–965. 9. Boulton, S. J., Brook, A., Staehling-Hampton, K., Heitzler, P. & Dyson, N. (2000) EMBO J. 19, 5376–5386. 10. Hu, G., Zhang, S., Vidal, M., Baer, J. L., Xu, T. & Fearon, E. R. (1997) Genes Dev. 11, 2701–2714. 11. Hu, G. & Fearon, E. R. (1999) Mol. Cell. Biol. 19, 724–732. 12. Zhang, J., Guenther, M. G., Carthew, R. W. & Lazar, M. A. (1998) Genes Dev. 12, 1775–1780. 13. Germani, A., Bruzzoni-Giovanelli, H., Fellous, A., Gisselbrecht, S., VarinBlank, N. & Calvo, F. (2000) Oncogene 19, 5997–6006. 14. Tiedt, R., Bartholdy, B. A., Matthias, G., Newell, J. W. & Matthias, P. (2001) EMBO J. 20, 4143–4152. 15. Boehm, J., He, Y., Greiner, A., Staudt, L. & Wirth, T. (2001) EMBO J. 20, 4153–4162. 16. Susini, L., Passer, B. J., Amzallag-Elbaz, N., Juven-Gershon, T., Prieur, S., Privat, N., Tuynder, M., Gendron, M.-C., Israel, A., Amson, R., et al. (2001) Proc. Natl. Acad. Sci. USA 98, 15067–15072. 17. Wheeler, T. C., Chin, L. S., Li, Y., Roudabush, F. L. & Li, L. (2002) J. Biol. Chem. 277, 10273–10282. 18. Johnsen, J. A., Subramaniam, M., Monroe, D. G., Janknecht, R. & Spelsberg, T. C. (2002) J. Biol. Chem., 277, 30754–30759. 19. Matsuzawa, S. & Reed, J. C. (2001) Mol. Cell 7, 915–926. 20. Liu, J., Stevens, J., Rote, C. A., Yost, H. J., Hu, Y., Neufeld, K. L., White, R. L. & Matsunami, N. (2001) Mol. Cell 7, 927–936.

21. Conaway, R. C., Brower, C. S. & Conaway, J. W. (2002) Science 296, 1254–1258. 22. Polekhina, G., House, C. M., Traficante, N., Mackay, J. P., Relaix, F., Sassoon, D. A., Parker, M. W. & Bowtell, D. D. (2002) Nat. Struct. Biol. 9, 68–75. 23. Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994) Nucleic Acids Res. 22, 4673–4680. 24. Catimel, B., Teh, T., Fontes, M. R., Jennings, I. G., Jans, D. A., Howlett, G. J., Nice, E. C. & Kobe, B. (2001) J. Biol. Chem. 276, 34189–34198. 25. Mathieu, M. N., Wade, J. D., Catimel, B., Bond, C. P., Nice, E. C., Summers, R. J., Otvos, L., Jr., & Tregear, G. W. (2001) J. Pept. Res. 57, 374–382. 26. Nice, E. C. & Catimel, B. (1999) BioEssays 21, 339–352. 27. Catimel, B., Nerrie, M., Lee, F. T., Scott, A. M., Ritter, G., Welt, S., Old, L. J., Burgess, A. W. & Nice, E. C. (1997) J. Chromatogr. A 776, 15–30. 28. Catimel, B., Scott, A. M., Lee, F. T., Hanai, N., Ritter, G., Welt, S., Old, L. J., Burgess, A. W. & Nice, E. C. (1998) Glycobiology 8, 927–938. 29. Kauffmann, R. C., Li, S., Gallagher, P. A., Zhang, J. & Carthew, R. W. (1996) Genes Dev. 10, 2167–2178. 30. Germani, A., Romero, F., Houlard, M., Camonis, J., Gisselbrecht, S., Fischer, S. & Varin-Blank, N. (1999) Mol. Cell. Biol. 19, 3798–3807. 31. Liu, C.-G., Maercker, C., Castanon, M. J., Hauptmann, R. & Wiche, G. (1996) Proc. Natl. Acad. Sci. USA 93, 4278–4283. 32. Fuchs, P., Zorer, M., Rezniczek, G. A., Spazierer, D., Oehler, S., Castanon, M. J., Hauptmann, R. & Wiche, G. (1999) Hum. Mol. Genet. 8, 2461–2472. 33. Ishikawa, K., Nash, S. R., Nishimune, A., Neki, A., Kaneko, S. & Nakanishi, S. (1999) Genes Cells 4, 381–390. 34. Ivan, M., Kondo, K., Yang, H., Kim, W., Valiando, J., Ohh, M., Salic, A., Asara, J. M., Lane, W. S. & Kaelin, W. G., Jr. (2001) Science 292, 464–468. 35. Jaakkola, P., Mole, D. R., Tian, Y. M., Wilson, M. I., Gielbert, J., Gaskell, S. J., Kriegsheim, A., Hebestreit, H. F., Mukherji, M., Schofield, C. J., et al. (2001) Science 292, 468–472. 36. Yu, F., White, S. B., Zhao, Q. & Lee, F. S. (2001) Proc. Natl. Acad. Sci. USA 98, 9630–9635. 37. Lim, W. A. (1996) Structure (London) 4, 657–659. 38. Li, S., Xu, C. & Carthew, R. W. (2002) Mol. Cell. Biol. 22, 6854–6865.

3106 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0534783100

We thank Ken Mitchelhill for HPLC and mass spectrometry, Julie Rothacker (Ludwig Institute, Melbourne) for mass spectrometry and Biacore analysis, Drs. A. Germani and F. Calvo (Saint-Louis Hospital) for the gift of the Kid expression construct, Drs. Steven Johnsen and Thomas Spelsberg (Mayo Clinic and Foundation) for the TIEG-1 expression vector, and Ross Dickins, Peter Janes, and Richard Pearson for critical comments on the manuscript. D.D.L.B. and C.M.H. were supported by National Health and Medical Research Council (Australia). G.W. is supported by Austrian Science Research Fund Grant P14520.

House et al.