And GSK3-Mediated Degradation of P100 Is a Pro-Survival Mechanism in Multiple Myeloma

ARTICLES

Fbxw7α- and GSK3-mediated degradation of p100 is a pro-survival mechanism in multiple myeloma

Luca Busino1,5, Scott E. Millman1,5, Luigi Scotto1, Christos A. Kyratsous1,2,6, Venkatesha Basrur3, Owen O’Connor1, Alexander Hoffmann4, Kojo S. Elenitoba-Johnson3 and Michele Pagano1,2,7

Fbxw7α is a member of the F-box family of proteins, which function as the substrate-targeting subunits of SCF (Skp1/Cul1/F-box protein) ubiquitin ligase complexes. Using differential purifications and mass spectrometry, we identified p100, an inhibitor of NF-κB signalling, as an interactor of Fbxw7α. p100 is constitutively targeted in the nucleus for proteasomal degradation by Fbxw7α, which recognizes a conserved motif phosphorylated by GSK3. Efficient activation of non-canonical NF-κB signalling is dependent on the elimination of nuclear p100 through either degradation by Fbxw7α or exclusion by a newly identified nuclear export signal in the carboxy terminus of p100. Expression of a stable p100 mutant, expression of a constitutively nuclear p100 mutant, Fbxw7α silencing or inhibition of GSK3 in multiple myeloma cells with constitutive non-canonical NF-κB activity results in apoptosis both in cell systems and xenotransplant models. Thus, in multiple myeloma, Fbxw7α and GSK3 function as pro-survival factors through the control of p100 degradation.

Fbxw7 (F-box/WD40 repeat-containing protein 7; also known as members to inhibit their activity, primarily by sequestering them Fbw7, hCdc4 and hSel10) is a member of the F-box family of in the cytoplasm. proteins, which function as the substrate-targeting subunits of SCF p100 is the main inhibitor of the non-canonical, or alternative, ubiquitin ligase complexes1–3. Fbxw7 is an essential gene owing to its NF-κB pathway15. In response to developmental signals, such as 4–7 function in development and differentiation . Mammals express three lymphotoxin (LTα1β2), B-cell activating factor (BAFF) and CD40 alternatively spliced Fbxw7 isoforms (Fbxw7α, Fbxw7β and Fbxw7γ) ligand15–20, p100 is dissociated from NF-κB dimers (p50:RelA), that are localized in the nucleus, cytoplasm and nucleolus, respectively5. allowing their nuclear translocation21. Subsequent activation of the The SCFFbxw7 complex targets multiple substrates, including cyclin E, transcriptional response includes de novo synthesis of p100 (NFKB2 c-Myc, Jun, Mcl1 and Notch (refs 5,7–9). In T-cell acute lymphoblastic gene), leading to concomitant generation of p52 through a co- leukaemia (T-ALL), Fbxw7 is a tumour suppressor, and mutations in translational processing mechanism that requires IKKα-dependent the FBXW7 gene, as well as overexpression of microRNAs targeting its phosphorylation of p100 on Ser 866 and Ser 870, and the activity expression, have been reported7,10,11. Moreover, mutations of FBXW7 of SCFβTrCP (refs 22–24). p52 preferentially binds RelB to activate a have been found in a variety of solid tumours5. Interestingly, alterations distinct set of gene targets involved in lymphoid development25,26. of the FBXW7 gene have not been observed in multiple myelomas and Whether and how p100 is regulated by protein degradation have B-cell lymphomas12,13. not been investigated. The p100 protein belongs to the NF-κB family, which consists Constitutive activation of NF-κB is common in B-cell neoplasms27. of five evolutionarily conserved and structurally related activator Notably, many mutations in genes encoding regulators of non- proteins (RelA (p65), RelB, c-Rel (Rel), p50 and p52) and five canonical NF-κB activity have been identified in human multiple inhibitory proteins (p100, p105 and the IκB proteins IκBα,IκBβ myelomas28,29 (for example, loss-of-function mutations in TRAF2/3 and IκBε). The NF-κB activators share a 300-amino-acid Rel and cIAP1/2, gain-of-function mutations in NIK and C-terminal homology domain (RHD) that controls DNA binding, dimerization truncations of p100)13,28–30. These abnormalities result in a consti- and nuclear localization14. The five inhibitors are characterized by tutively elevated level of NF-κB signalling, which is associated with ankyrin repeat domains (ARDs) that bind the RHDs of NF-κB glucocorticoid resistance and proteasome inhibitor sensitivity. The

1NYU Cancer Institute, New York University School of Medicine, 522 First Avenue, SRB 1107, New York, New York 10016, USA. 2Howard Hughes Medical Institute, USA. 3Department of Pathology, University of Michigan, Ann Arbor, Michigan 48109, USA. 4Signaling Systems Laboratory, Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, California 92093, USA. 5These authors contributed equally to this work. 6Present address: Regeneron Pharmaceuticals, Inc., 777 Old Saw Mill River Road, Tarrytown, New York 10591, USA. 7Correspondence should be addressed to M.P. (e-mail: [email protected])

Received 19 October 2011; accepted 10 February 2012; published online 4 March 2012; DOI: 10.1038/ncb2463

abc la) a) IP: anti-FLAG ) α TrCP1 WCE EV β Fbxw2 Fbxw4 Fbxw5 Fbxw7 Fbxw8 Cdh1 Cdc20

p100 input EL(pSer93/94/98) p100 p100(Ser707/711AEV p100 p100(Ser707/711AlEV im p100 p100(pSer707)p100(pSer711)p100(pSer707/711 p100(pSer707) 10% B Fbxw7α (anti-HA) Fbxw7α p105 Skp1 βTrCP1 IκBα p100 (anti-FLAG) Skp1 IP: anti-FLAG WCE –+ FBPs, Cdh1 f GSK3i IX and Cdc20 p100(pSer707) (anti-FLAG) p100

e p100(pSer866/870)

d –+ GSK3β c-Myc(pThr58/Ser62) 11AIa) Fbxw7α c-Myc

p100 (anti-FLAG) β-catenin(pSer33/37/Thr41) p100 p100 (Ser707/7 Input IP: anti-FLAG +–+– GSK3β β-catenin p100(pSer707) Skp1 p100

g FLAG–p100 h p100 p100(Ser707/711AIa) p100 +SCFFbxw7α +SCFFbxw7α GSK3 inhibitior IX 0 30 60 90 0 30 60 90 0 30 60 90 Time (min) M M M M μ μ μ μ p100(Ub) n DMSO 0.1 1.0 5.0 10.0 FLAG–p100 (Ser707/711Ala) Fbxw7α p100(pSer707) p100 (anti-FLAG) IVT [ 35S]p100 IP: anti-FLAG

Figure 1 p100 interacts with Fbxw7α through a conserved degron absence of GSK3β. Reaction products were immunoblotted as indicated. phosphorylated by GSK3. (a) p100 binds Fbxw7α. HEK293 cells were (e) GSK3-mediated phosphorylation of p100 is required for p100 binding transfected with cDNAs encoding the indicated FLAG-tagged F-box proteins to Fbxw7α in vitro. In vitro-translated, FLAG-tagged p100 was incubated (FBPs), Cdh1 or Cdc20 and treated with the proteasome inhibitor MG132 with or without GSK3β before incubation with in vitro-translated Fbxw7α. for 6 h. FLAG-tagged immunoprecipitates (IPs) from cell extracts with FLAG-tagged immunoprecipitates were immunoblotted as indicated. 5% of anti-FLAG resin were immunoblotted as indicated. Lane 1 shows whole-cell inputs are shown. (f) GSK3 phosphorylates p100 in vivo. MEFs were treated extract (WCE) from empty vector (EV)-transfected cells. (b) Ser 707 and with dimethylsulphoxide (DMSO) or the GSK3 inhibitor GSK3i IX (5 µM). Ser 711 in p100 are required for the interaction with Fbxw7α. HEK293 Cell extracts were immunoblotted as indicated. (g) In vivo binding between cells were transfected with HA-tagged Fbxw7α and constructs encoding p100 and Fbxw7α depends on GSK3 activity. Nfκb2−/− MEFs were infected FLAG-tagged p100 or p100(Ser707/711Ala). FLAG-tagged p100 was with retroviruses expressing FLAG-tagged p100 or p100(Ser707/711Ala). immunoprecipitated from cell extracts, followed by immunoblotting as Cells were treated with the indicated concentrations of GSK3i IX for indicated. The right panel shows whole-cell extract. (c) The p100 degron 12 h. FLAG-tagged immunoprecipitates were immunoblotted as indicated. requires phosphorylation to bind Fbxw7α. In vitro-translated Fbxw7α and (h) p100 is ubiquitylated in vitro in a degron- and SCFFbxw7α-dependent βTrCP1 were incubated with beads coupled to the indicated p100 peptides or manner. [35S]-in vitro-translated (IVT) p100 or p100(Ser707/711Ala) was BimEL peptide. Beads were washed and eluted proteins were immunoblotted incubated at 30 ◦C with a ubiquitylation mix. SCFFbxw7α was added to the as indicated. The ﬁrst lane shows 10% of in vitro-translated protein reaction for the indicated times. Reactions were subjected to SDS–PAGE inputs. (d) GSK3 phosphorylates p100 in vitro. In vitro-translated p100 or and analysed by autoradiography. Uncropped images of blots are shown in p100(Ser707/711Ala) was incubated at 30 ◦C for 1 h in the presence or Supplementary Fig. S8. efficacy of the proteasome inhibitor bortezomib in multiple myeloma RESULTS patients and human multiple myeloma cell lines (HMMCLs) with Phosphorylation- and GSK3-dependent interaction of p100 inactivation of TRAF3 has been attributed in part to inhibition of with Fbxw7α the NF-κB pathway29. To identify previously unknown substrates of the SCFFbxw7α ubiquitin Here, we show that Fbxw7α constitutively targets nuclear p100 ligase, FLAG–HA-tagged Fbxw7α was immunopurified from HEK293 for proteasomal degradation on phosphorylation of p100 by GSK3. cells (Supplementary Fig. S1a) and analysed by mass spectrometry. Clearance of p100 from the nucleus is required for efficient activation As a negative control, we used FLAG–HA-tagged Fbxw7α (WD40), of the NF-κB pathway and the survival of multiple myeloma cells. a mutant that lacks the ability to bind substrates, but not Skp1

a HA–p100(ΔNES) b HA–p100(ΔNES; HA–p100(ΔNES) Ser707/711AIa) 1.3 LacZ Fbxw7 siRNA siRNA MG132 01 3 6 91201 3 6 912CHX (h) 1.2 p100(ΔNES) 0 3 6120361203612CHX (h) 1.1 (anti-HA) p100(ΔNES) (anti-HA) 1.0 Fold change Fbxw7α α-tubulin 0.9 β-actin NES) NES; Δ Δ

c p100 HA–p100(HA–p100( (anti-HA) DAPI Merge Ser707/711AIa) d Nfκ b2–/– MEFs NES) × 1.5 NES) NES; × ×

p100 (4 p100(4 1.0 HA–p100(NLS; HA–p100(4HA– HA–p100(NLS) Ser707/711AIa) Ser707/711AIa)

NES; 0136013601360136 0.5 × CHX (h) p100 (anti-HA) Fold change IκBα 0 p100 (4

Ser707/711AIa) Skp1 Φ xxxΦ xxΦ xΦ PKI NES consensus NES; NES) × L LEALSDMGL p100 (aa 831–840) × A A EALSDA GA p100 (aa 831–840; 4×NES) p100(4 HA–p100(NLS) HA– HA–p100(NLS; Φ = Hydrophobic amino acids HA–p100(4 Ser707/711AIa) (Leu, lle, Val, Met, Phe) Ser707/711AIa) x = small/polar/charged amino acids

Figure 2 Fbxw7α controls p100 stability in the nucleus. (a) A constitutively were infected with retroviruses expressing either HA-tagged p100(4×NES) nuclear mutant of p100 (p100(1NES)) is stabilized by proteasome inhibitor or HA-tagged p100(4×NES; Ser707/711Ala) and stained with an antibody treatment or silencing of Fbxw7. HeLa cells were infected with a retrovirus against HA. Scale bar, 20 µm. The identiﬁed consensus NES is shown expressing HA-tagged p100(1NES) and treated with either siRNAs or on the bottom. aa, amino acids. (d) A constitutively nuclear mutant of MG132. Cells were treated with cycloheximide (CHX) for the indicated p100 (p100(4×NES)) is stabilized by mutation of Ser 707 and Ser 711 times, and total cell lysates were analysed by immunoblotting as indicated. to alanine. Nfκb2−/− MEFs were infected with a retrovirus expressing (b) A constitutively nuclear mutant of p100 (p100(1NES)) is stabilized HA-tagged p100(4×NES), HA-tagged p100(4×NES; Ser707/711Ala), by mutation of Ser 707 and Ser 711 to alanine. HeLa cells were infected HA-tagged p100(NLS) or HA-tagged-p100(NLS; Ser707/711Ala), treated with retroviruses expressing either HA-tagged p100(1NES) or HA-tagged with cycloheximide for the indicated times and total cell lysates were p100(1NES; Ser707/711Ala) and treated with cycloheximide for the analysed by immunoblotting as indicated (left). IκBα is shown as a positive indicated times. Total cell lysates were analysed by immunoblotting as control for cycloheximide activity. mRNA levels of retrovirally expressed indicated (left). mRNA levels of retrovirally expressed p100 were analysed p100 were analysed by quantitative PCR (right). Uncropped images of blots by quantitative PCR (right). (c) Identiﬁcation of an NES in p100. MEFs are shown in Supplementary Fig. S8. and Cul1 (ref. 31 and Supplementary Fig. S1b). p100 peptides were To investigate whether serine phosphorylation plays a role in the identified in Fbxw7α immunoprecipitates, but not in Fbxw7α (WD40) interaction of p100 with Fbxw7α, we used immobilized, synthetic purifications (Supplementary Fig. S1c), indicating that p100 may be peptides spanning the candidate phospho-degron (amino acids a SCFFbxw7α substrate. 702–715 in human p100). Only a peptide doubly phosphorylated on To investigate whether the binding between p100 and Fbxw7α is Ser 707 and Ser 711 efficiently bound Fbxw7α (but not βTrCP; Fig. 1c), specific, we screened a panel of human WD40 domain-containing indicating that Fbxw7α binds doubly phosphorylated p100. F-box proteins, as well as Cdh1 and Cdc20 (WD40 domain- To further study the role of p100 phosphorylation, we generated containing subunits of an SCF-like ubiquitin ligase). Whereas a phospho-specific antibody against phospho-serine at position 707 p100, p105 and IκBα were detected in βTrCP immunoprecipi- (Supplementary Fig. S2a,b). Using this tool, we found that Fbxw7α, but tates, as previously reported24,32, Fbxw7α co-immunoprecipitated not βTrCP, co-immunoprecipitated p100 phosphorylated on Ser 707, only p100 (Fig. 1a). whereas βTrCP, but not Fbxw7α, co-immunoprecipitated p100 phos- Next, we systematically mapped the Fbxw7α-binding motif of phorylated at Ser 866 and Ser 870 (Fig. 1a and Supplementary Fig. S2c). human p100. A set of p100 mutants with serial deletions narrowed As previous studies have shown that Fbxw7 substrates are the binding motif to a C-terminal region between amino acids phosphorylated by GSK3 (ref. 5), we investigated whether this was the 702 and 720 (Supplementary Fig. S1d,e). This region contains a case for p100. GSK3 was able to phosphorylate p100 on Ser 707 in vitro conserved sequence resembling the canonical Fbxw7 degradation motif and promoted its binding, but not the binding of p100(Ser707/711Ala), (degron) S/TPPXS/E (Supplementary Fig. S1f). Further mutational to Fbxw7α (Fig. 1d,e and Supplementary Fig. S2d), confirming that analysis of p100 revealed Ser 707 and Ser 711 as residues contributing only p100 phosphorylated on Ser 707 associates with Fbxw7α. We to its interaction with Fbxw7α (Supplementary Fig. S1g). A p100 next asked whether p100 is a target of GSK3 in vivo. Treatment of mutant containing alanine substitutions at both Ser 707 and Ser 711 mouse embryo fibroblasts (MEFs) with a GSK3 inhibitor33 (GSK3i IX) [p100(Ser707/711Ala)] failed to bind Fbxw7α (Fig. 1b). markedly reduced the level of phosphorylation of Ser 707 on p100 to an

p100/p52 p100 (anti-N-term) MEFs p100 p100 + (anti-C-term) HA–p52 b p100 p52 DAPI Merge Fbxw7 flox/flox Fbxw7–/– Fbxw7 flox/flox Fbxw7–/– (anti-C-term) (anti-HA) 0 3 6 12 24 0 3 6 12 24 0 3 6 12 24 0 3 6 12 24 anti-LTβR (h) p100 e Tet-ON HA-p52 p100 LacZ Fbxw7 (anti-N-Term) siRNA siRNA p52 0 6 12 24 0 6 12 24 DOX (h) p100(pSer707) p100 p100(pSer866/870) p52 (anti-HA) RelB Cyclin E IKBα Skp1 Skp1 f Cytoplasmic fraction Nuclear fraction (15 μg) (15 μg) g Fbxw7flox/flox c DMSO GSK3i IX DMSO GSK3i IX 03612036120361203612Anti-LTβR (h) p100 R 0 h R 18 h β β p52 FLAG–p100 –/–

RelB -LT -LT

HA–p52 Fbxw7 anti anti FLAG–p100(Ser707/711Ala) IκBα EV - Fbxw7α p100 Skp1 Cytoplasmic fraction (15 μg) Nuclear fraction (15 μg) p52 (anti-HA) Fbxw7α

p100 (anti-FLAG) IP: anti-Fbxw7α

IP: anti-FLAG

Figure 3 p100 is degraded by Fbxw7α and GSK3 independently of NF-κB HA tag (red). DNA was stained with DAPI. Scale bar, 100 µm. (e) p52 signalling, but NF-κB proteins compete with Fbxw7α for binding to p100. induces p100 accumulation in control cells, but not in cells depleted (a) p100 accumulates in the nucleus on LTβR activation. MEFs were treated of Fbxw7. HeLa Tet-ON HA-tagged p52 cells were treated with siRNA with agonistic anti-LTβR antibodies, fixed and stained with antibodies against LacZ or FBXW7. Doxycycline (DOX) was added to cells at the against the N- or C-terminus of p100 (green). A similar pattern was indicated times. Total cell extracts were immunoblotted as indicated. observed with an antibody against either the N-terminus (recognizing (f) p52 competes with Fbxw7α for binding to p100 in vivo. HEK293 both p100 and p52) or the C-terminus (recognizing only p100). Scale cells were transfected with cDNAs encoding either FLAG-tagged p100 bar, 100 µm. (b) Levels of p100 are higher in Fbxw7 −/− MEFs than or FLAG-tagged p100(Ser707/711Ala). Increasing amounts of HA-tagged in Fbxw7flox/flox MEFs. MEFs were stimulated with agonistic anti-LTβR p52 plasmid were transfected. Proteins were immunoprecipitated from cell antibodies and collected at the indicated times. Nuclear and cytoplasmic extracts with anti-FLAG resin (anti-FLAG), and the immunoprecipitates (IPs) fractions were analysed by immunoblotting as indicated. (c) Levels of were immunoblotted as indicated. The first lane shows cells transfected with p100 are increased on GSK3 inhibition. MEFs were incubated with an empty vector (EV). (g) The interaction of Fbxw7 with p100 is disrupted GSK3i IX and stimulated with agonistic anti-LTβR antibodies. Nuclear after treatment of cells with anti-LTβR. MEFs were treated with anti-LTβR for and cytoplasmic fractions were analysed by immunoblotting as indicated. 18 h, collected, lysed and endogenous Fbxw7α was immunoprecipitated. (d) p52 induces the accumulation of nuclear p100. Nfκb2−/− MEFs were Immunocomplexes were analysed by western blotting for the indicated infected with retroviruses expressing p100 and HA-tagged p52, fixed and proteins. Fbxw7−/− cells were used as a negative control. Uncropped images stained with antibodies against the C-terminus of p100 (green) and the of blots are shown in Supplementary Fig. S8. extent similar to that observed for GSK3 phosphorylation sites in c-Myc Fbxw7α-mediated degradation of p100 occurs in the nucleus and β-catenin (Fig. 1f). In contrast, phosphorylation of Ser 866/870 independently of NF-κB signalling (the βTrCP degron) was unaffected by GSK3 inhibition. Furthermore, p100 is predominantly a cytoplasmic protein, but on a brief increasing doses of GSK3i IX decreased the affinity of Fbxw7α for p100 treatment of cells with the CRM1/exportin1 inhibitor Leptomycin (Fig. 1g). Taken together, these results indicate that GSK3 phosphory- B (LMB), p100 accumulated in the nucleus (Supplementary lates p100 on the Fbxw7α degron, promoting binding to Fbxw7α. Fig. S2f,g), indicating that it actively shuttles between the cytoplasm Last, we reconstituted the ubiquitylation of p100 in vitro. Wild-type and nucleus, as previously suggested34. Knockdown analysis of p100, but not p100(Ser707/711Ala), was efficiently ubiquitylated only the Fbxw7 isoforms using established siRNAs targeting Fbxw7α, when recombinant SCFFbxw7α was present in the reaction mix (Fig. 1h Fbxw7β and Fbxw7γ, either individually or all at once35,36, revealed and Supplementary Fig. S2e), supporting the hypothesis that Fbxw7α that the α (nuclear) isoform regulates p100 cellular abundance directly controls the ubiquitylation of p100. (Supplementary Fig. S2h).

378 NATURE CELL BIOLOGY VOLUME 14 | NUMBER 4 | APRIL 2012 © 2012 Macmillan Publishers Limited. All rights reserved. ARTICLES ac 140 ﬂox/ﬂox Fbxw7 120 Fbxw7 –/– 100 4 VCAM-1 6 NFKB2 80 2 60 3 40 0 0 R-responsive genes β 20 Fold change EV Fold change 0 1 3 12 24 0 1 3 12 24 0 p100

Anti-LTβR (h) Anti-LTβR (h) Percentage of activation

Anti-LT p100(Ser707/711Ala) NFKB2 MCP-1 RANTES VCAM1 NFKBIA p100(4×NES) 3 NFKBIA 4 MCP-1 R stimulation (12 h) p100(4×NES; Ser707/711Ala) 2 β 140 2 1 140 120 0 0 120

Anti-LT Anti-LT 100 Fold change

Fold change 0 1 3 12 24 0131224 100 80 80 Anti-LTβR (h) Anti -LTβR (h) 60 B target genes 60 40 κ 40

R-unresponsive genes R-unresponsive 20 20 8 β b NFKB2 15 RANTES 0 6 0 Percentage of activation Percentage 10 Non-NF- IL-6 Percentage of activation 18S

4 Anti-LT 2 5 CCND1 SPARC Fold change Fold change 0 0 0 1 6 12 24 0161224 d NFKB2 promoter MCP1 promoter Anti-LTβR (h) Anti-LTβR (h) 4 4 3 6 3 10 BAFF MCP1 2 2 4 1 1 Fold change

5 Fold change 0 0 2 0 6 12 24 0 6 12 24 Fold change Fold change 0 0 Anti-LTβR (h) Anti-LTβR (h) 01 61224 0161224 p100 MIP2 promoter IL-6 promoter Anti-LTβR (h) Anti-LTβR (h) 4 4 p100(Ser707/711Ala) 3 3 p100(4×NES) NFKBIA 4 BLC 4 2 2 1 2 2 EV 1 p100 Fold change 0 Fold change 0 0 6 12 24 0 6 12 24 Fold change 0 Fold change p100(Ser707/711Ala) 0 Anti-LTβR (h) Anti-LTβR (h) 0161224 0161224 Anti-LTβR (h) Anti-LTβR (h)

Figure 4 Clearance of p100 from the nucleus is crucial for non-canonical with retroviruses expressing empty vector, p100, p100(Ser707/711Ala), NF-κB signalling. (a) LTβR-dependent gene transcription is impaired in cells p100(4×NES) or p100(4×NES; Ser707/711Ala) and stimulated with lacking Fbxw7. Fbxw7flox/flox and Fbxw7 −/− MEFs were treated with agonistic agonistic anti-LTβR antibodies. After 12 h of stimulation, levels of the anti-LTβR antibodies and collected at the indicated times. Levels of the indicated mRNAs were determined by quantitative PCR (s.d., n = 3) and indicated mRNAs were determined by quantitative real-time PCR (s.d., normalized to the transcriptional activation measured in cells infected with n = 3). The value for the amount of PCR product present in Fbxw7flox/flox empty vector. (d) LTβR-dependent binding of RelB to NF-κB elements is MEFs was set as 1. (b) LTβR-dependent gene transcription is impaired in impaired in cells expressing p100(Ser707/711Ala) or p100(4×NES). MEFs cells expressing stable p100(Ser707/711Ala). MEFs were infected with were infected with retroviruses expressing p100, p100(Ser707/711Ala) or retroviruses expressing empty vector (EV), p100 or p100(Ser707/711Ala), p100(4×NES) and stimulated with agonistic anti-LTβR antibodies. DNA stimulated with agonistic anti-LTβR antibodies and processed as in a. Levels precipitated with an anti-RelB antibody was amplified by real-time PCR of the indicated mRNAs were determined by quantitative PCR (s.d., n = 3). using primers flanking the indicated gene promoters (s.d., n = 3). The The value for the amount of PCR product present in empty-vector-infected LTβR-unresponsive promoter of IL-6 was used as a negative control. The MEFs was set to 1. (c) LTβR-dependent gene transcription is impaired in value for the amount of PCR product present in MEFs infected with wild-type cells expressing constitutively nuclear p100(4×NES). MEFs were infected p100 was set as 1.

To expand our studies on the nuclear regulation of p100, we substituted for four residues within the NES (Leu 831, Leu 832, generated p100(1NES), a C-terminal deletion mutant of p100 that Met 838 and Leu 840), mislocalized to the nucleus (Fig. 2c). lacks the last 180 amino acids, but retains the Fbxw7α degron. p100(4×NES) exhibited a short half-life, and the Ser707/711Ala p100(1NES) constitutively localizes to the nucleus34 (Supplementary mutations stabilized the protein (Fig. 2d). In contrast, p100(NLS), Fig. S3a). Depletion of Fbxw7 or treatment of cells with the a mutant in which the amino acids of the p100 nuclear localization proteasome inhibitor MG132 increased the half-life of p100(1NES) signal (338KRRK341) were mutated to alanine (Supplementary Fig. S3b), (Fig. 2a). Moreover, mutation of Ser 707 and Ser 711 to alanine was stable and unaffected by mutations in the Fbxw7α degron in the context of p100(1NES) significantly stabilized this largely (Fig. 2d). We also found that retrovirally expressed wild-type p100 nuclear protein (Fig. 2b). Next, on the basis of homology with exhibited a short half-life in the nuclear fraction of Nfκb2−/− NES consensus sequences, we identified the NES sequence in MEFs, and the Ser707/711Ala mutations stabilized nuclear p100 p100 (Fig. 2c). p100(4×NES), a p100 mutant in which alanine was (Supplementary Fig. S3c). Furthermore, Fbxw7 silencing or treatment

Other B-cell 600 a bcKMS11 400 MM1.R Multiple myeloma malignancies 11Ala) 11Ala) 400 200 200

NES) NES) cell viability

× × 0 cell viability 0 Percentage of Percentage 1357 of Percentage 1357 Ly10 SudHL6 Farage Raji Ly10 SudHL6 Farage Raji ARP-1 ARP-1 KMS11 MM1.S U266B1 KMS11 MM1.S U266B1 Days p100 Days –p100 –p100 Fbxw7α 600 ARP-1 400 PCNY1 c-Myc EV HA HA–p100(Ser707/7HA–p100(4EV HA HA–p100(Ser707/7HA–p100(4 p100 (Anti-HA, s.e.) 400 IκBα 200 α-tubulin p100 200

(Anti-HA, l.e.) cell viability Skp1 0 cell viability 0 Percentage of Percentage p52 of Percentage Cytoplasmic Nuclear Cytoplasmic Nuclear 135 1357 Skp1 Days fraction fraction fraction fraction Days ARP-1 KMS11 EV p100 p100(Ser707/711Ala) d 1.2 KMS11 1.5 MM1.R e 1.0 h 0.8 1.0

0.6 100 50 shRNA1 shRN2 shRNA1 shRNA2 shRNA1 shRNA2

0.4 0.5 40 shRNA shRNA shRNA 0.2 30 Fold change Fold change

50 Luc Fbxw7 Fbxw7 Luc Fbxw7 Fbxw7 Luc Fbxw7 Fbxw7 0 0 20 10 p100 (pSer707) p100 α

0 Percentage of 0 Percentage of CD74 RELB CD74 IL2RG BIRC3 ICAM1 ICAM1 KMS11 MM1.R KMS11MM1.R Fbxw7 NFKBIE NFKBIE RANTES TNFAIP3 TNFAIP3 Lamin A/C Annexin V+ / 7-AAD+ cells 7-AAD+ / Annexin V+ Annexin V – /7-AAD– cells Annexin V – α IκB EV p100 p100(Ser707/711Ala) EV p100 p100(Ser707/711Ala) Skp1 T CN g 400 f KMS11 MM1.S 500 KMS11 MM.1R 500 600 400 300 1 1 400 300 300 400 200 200 200 200 100 shRNA shRNA2 shRNA shRNA2 cell viability cell viability 100 100 shRNA shRNA Percentage of Percentage of cell viability 0 cell viability 0 0 0 Percentage of 13579 Percentage of 13579 135 1357 Luc Fbxw7Fbxw7Luc Fbxw7Fbxw7 Fbxw7α Days Days Days Days Skp1 EV p100 p100(4×NES) Luc shRNA Fbxw7 shRNA1 Fbxw7 shRNA2 KMS11 MM1.R

Figure 5 p100 degradation promotes multiple myeloma cell survival. (a) In set as 1. (e) Expression of stable p100 promotes apoptosis of HMMCLs. B-cell lines, p100 is cytoplasmic, whereas Fbxw7α is largely nuclear. HMMCLs were infected as in c. At 72 h post-infection, cells were stained Subcellular fractionation was carried out on HMMCLs (ARP-1, KMS11, with Annexin V/7-AAD and analysed by ﬂow cytometry. Apoptosis and MM1.S and U266B1), DLBCL (Ly10, SudHL6 and Farage) and Burkitt’s cell viability were calculated as the percentage of Annexin V/7-AAD lymphoma (Raji) cell lines. Lysates were immunoblotted as indicated. IκBα double-positive and double-negative cells, respectively. (f) Forced nuclear and α-tubulin, cytoplasmic controls. c-Myc, nuclear control. (b) Stable localization of p100 inhibits HMMCLs proliferation. HMMCLs were infected p100(Ser707/711Ala) and p100(4×NES) are correctly processed in with retroviruses encoding p100, p100(4×NES) mutant or empty vector. HMMCLs. HMMCLs were infected with retroviruses encoding HA-tagged Cell viability was monitored as in c. Values were normalized on the empty p100, p100(Ser707/711Ala), p100(4×NES) or empty vector (EV). Cell vector at time 0. Error bars represent s.d., n = 4. (g) Fbxw7 depletion extracts were immunoblotted as indicated. Short exposure (s.e.) and long impairs HMMCL growth. HMMCLs were infected with the indicated exposure (l.e.) are shown. (c) Expression of stable p100 mutant impairs shRNA-encoding lentiviruses. Cell viability was monitored as in c (left HMMCL growth. HMMCLs were infected with retroviruses expressing p100, panels). Values were normalized on Luc shRNA at time 0. Error bars p100(Ser707/711Ala) or empty vector. Cell proliferation was monitored represent s.d., n =4. Fbxw7 protein levels were analysed by immunoblotting by MTS assay and normalized on empty vector at day 1, arbitrarily set (right panel). (h) Fbxw7 depletion stabilizes nuclear p100. HMMCLs were as 100%. Error bars represent s.d., n = 4. (d) Stable p100 expression infected as in g. Total (T), cytoplasmic (C) and nuclear (N) fractions were impairs NFκB-dependent transcription in HMMCLs. HMMCLs were infected immunoblotted as indicated. IκBα and Lamin A/C are cytoplasmic and as in c. Steady-state levels of the indicated mRNAs were analysed by nuclear markers, respectively. Uncropped images of blots are shown in quantitative PCR (s.d., n = 3). PCR product amount in empty vector was Supplementary Fig. S8. with MG132 stabilized nuclear p100 in cells pre-treated with LMB in Fbxw7flox/flox and Fbxw7−/− MEFs with similar kinetics (despite (Supplementary Fig. S3d). higher levels in knockout cells, possibly owing to its nucleus–cytoplasm To study how Fbxw7α influences levels of endogenous p100, shuttling). Pharmacologic inhibition of GSK3 produced results similar we activated the non-canonical NF-κB pathway in MEFs using an to loss of Fbxw7 (Fig. 3c). antibody agonistic to the lymphotoxin-β receptor21 (anti-LTβR). In Together, these data support the hypothesis that Fbxw7α constitu- the nuclei of Fbxw7flox/flox MEFs, levels of p100 increased in parallel with tively targets p100 for degradation in the nucleus. p52 generation (Fig. 3a–c). In comparison, unstimulated Fbxw7−/− MEFs exhibited higher levels of nuclear p100 and this difference was NF-κB proteins compete with Fbxw7α for p100 particularly evident for the fraction of p100 phosphorylated on Ser 707 The nuclear accumulation of p100 following LTβR ligation has been (Fig. 3b). On LTβR ligation, levels of cytoplasmic p100 decreased attributed to de novo synthesis37. We reasoned that stabilization of

a c KMS11 pBABE-luc ( × 10

3 ( ×

4 10 7 photons s 6 photons s Radiance p100 wild type p100 (Ser707/711Ala) p100 wild type p100 (Ser707/711Ala) 2 3 Radiance –1 EV p100 p100 (Ser707/711Ala) cm –1

2 cm 1 –2 sr –2 sr –1 ) 1 –1 Day 7 )

Colour scale 10 days 36 days Min = 1.70 × 106 Max = 3.40 × 107 Colour scale Min = 1.89 × 105 8 Max = 4.75 × 106 b ) –1 7 p100 Day 14 sr p100(Ser707/711Ala)

–2 6 5 cm

–1 4 3 2 Luminescence

photons s 1 8 Day 28 10 0 ×

( 72136 –1 Days

Figure 6 Expression of a stable p100 mutant impairs the growth of HMMCLs represent s.d., n = 3. (c) KMS11 cells were infected with retroviruses xenotransplanted into immunodeficient mice. (a) KMS11 cells were infected expressing a firefly luciferase reporter and an empty vector (EV), p100 with retroviruses expressing either p100 or p100(Ser707/711Ala) and a or p100(Ser707/711Ala). Three days after infection, cells were injected firefly luciferase reporter, before intraperitoneal injection of SCID Beige subcutaneously into SCID Beige mice. Cell growth was monitored by in vivo mice. Cell growth was monitored by in vivo imaging. (b) Bioluminescence imaging at 7, 14 and 28 days following injection. Bioluminescence was quantification of a with Living Image software (Xenogen). Error bars quantified with Living Image software (Xenogen). nuclear p100 also contributes to its elevated levels in the nucleus. As expression of several genes, including NFKB2, MCP-1, VCAM-1 and NF-κB proteins bind the ARD of p100, the Fbxw7α phospho-degron NFKBIA (Fig. 4a). This response was attenuated in MEFs lacking of p100 is located at the C-terminal end of the ARD (between the sixth Fbxw7, which exhibited a significant reduction in the amplitude and seventh ankyrin repeats) and NF-κB proteins (particularly p52 and of target gene transcription. Consistently, expression of the stable RelB) accumulate in the nucleus on LTβR ligation, we investigated p100(Ser707/711Ala) mutant, but not wild-type p100, inhibited the whether NF-κB proteins compete with Fbxw7α for p100 binding. transcription of NF-κB target genes (Fig. 4b). Furthermore, treatment Expression of exogenous p52 stabilized p100 (Fig. 3d,e), but this of cells with GSK3i IX also inhibited NF-κB activation following LTβR effect was no longer observed in cells depleted of Fbxw7 (Fig. 3e). stimulation (Supplementary Fig. S5a). Similarly to our observations Furthermore, p52 or RelB reduced the binding between p100 and with anti-LTβR, Fbxw7 silencing or GSK3 inhibition attenuated the Fbxw7α in a concentration-dependent manner both in vivo and in vitro NF-κB response to either CD40L or TNF (Supplementary Fig. S5b–e). (Fig. 3f and Supplementary Fig. S4a,b). Further evidence also supports The latter finding is in agreement with a role for p100 as a general a competition model. First, following LTβR ligation, the interaction of inhibitor of NF-κB signalling38. nuclear p100 with Fbxw7α decreased (Fig. 3g), whereas overall binding To further investigate the relevance of p100 accumulation in of RelA and RelB to p100 decreased in the cytoplasm and increased in the nucleus, we investigated whether a constitutively nuclear p100 the nucleus (Supplementary Fig. S4c). Second, deoxycholate analysis mutant is capable of inhibiting LTβR signalling. To this end, we of p100–Fbxw7 complexes revealed that the fraction of p100 bound expressed p100(4×NES) in MEFs and stimulated the cells with to Fbxw7α associates with RelB only through RHD:RHD interactions anti-LTβR. After 12 h of stimulation, the transcription of several (Supplementary Fig. S4d). Together, these results indicate that NF-κB LTβR-responsive genes (RANTES, NFKB2, MCP-1, VCAM-1 and proteins induce the stabilization of p100 by competing with Fbxw7α NFKBIA), but not LTβR-unresponsive genes (IL-6, CCND1) or for the ARD of p100. non-NF-κB target genes (18S, SPARC), was impaired by expression of the nuclear p100 mutant (Fig. 4c and Supplementary Fig. S6a). Clearance of nuclear p100 is required for efficient NF-κB The expression of stable p100(Ser707/711Ala) or a combined activation nuclear and stable p100 mutant (4×NES; Ser707/711Ala) impaired We next assessed the effect of Fbxw7 loss on the expression LTβR-dependent gene transcription to a similar extent (Fig. 4c), of NF-κB target genes with roles in lymphoid development and as did the expression of a constitutively cytoplasmic p100(NLS) inflammation. LTβR stimulation in Fbxw7flox/flox MEFs induced the mutant (Supplementary Fig. S6a). These results show that forced

abcTCN TCN TCN Untreated GSK3i IX 1.4 –+–+–+ –+–+–+ GSK3 IX –––+––+–+ MLN4924 KMS11 ––+––+ ––+ 1.2 p100 Bortezomib 1.0 p100 0.8 IκBα 0.6 c-Myc IκBα 0.4

c-Myc Fold change 0.2 Skp1 0 ARP-1 KMS11 Skp1 CD74 RELB KMS11 ICAM1 BIRC3 IL2RG NFKBIERANTES TNFAIP3 d 110 120 120 100 110 110 MM1.R KMS11 90 100 100 90 90 ARP-1 80 80 80 70 70 70 60 60 60 50 50 50 Percentage of viability Percentage of viability Percentage of viability 40 40 40 0 1 0 1 5 0 10 10 10 50 100 500 100 500 0.1 0.5 1.0 5.0 100 1,0002,0005,000 1,000 5,000 Bortezomib (nM) GSK3i IX (nM) MLN4924 (nM) f e 140 160 KMS11 TCN 140 120 MM1.R –––+––+–+ GSK3i XVI 120 ––+––+––+GSK3i XXII 100 ARP-1 100 80 p100 80 p100 (pSer707) 60 60 IκBα 40 40 20 20 Percentage of viability Lamin A/C Percentage of viability 0 0 0 0 1 10 10 Skp1 100 500 100 500 1,0002,0005,000 1,0002,0005,000 KMS11 10,000 GSK3i XVI (nM) GSK3i XXII (nM)

Figure 7 Pharmacologic inhibition of GSK3, Cullin–RING ligases or the or the proteasome decreases the viability of HMMCLs. KMS11, MM1.R proteasome impairs the growth of multiple myeloma cells. (a) GSK3 and ARP-1 cells were treated with increasing concentrations of GSK3i IX, inhibition induces nuclear accumulation of p100 in HMMCLs. KMS11 MLN4924 or bortezomib. Cell viability was measured by MTS assay at and ARP-1 cells were treated with GSK3i IX (2 µM for 8 h). Total (T), 72 h after treatment and individually normalized to untreated cells (0 nM), cytoplasmic (C) and nuclear (N) fractions were isolated and analysed by arbitrarily set as 100%. Error bars represent s.d., n = 4. (e) Next-generation immunoblotting as indicated. Cytoplasmic and nuclear markers, IκBα and GSK3 inhibitors induce accumulation of nuclear p100 in HMMCLs. KMS11 c-Myc, respectively, are also shown. (b) Proteasome or Cullin–RING ligase cells were treated with GSK3i XVI or GSK3i XXII (5 µM and 1 µM for 8 h, inhibitors induce nuclear accumulation of p100 in HMMCLs. KMS11 cells respectively). Total, cytoplasmic and nuclear fractions were isolated and were treated with either bortezomib or MLN4924 for 4 h. Total, cytoplasmic analysed by immunoblotting as indicated. Cytoplasmic and nuclear markers, and nuclear fractions were analysed by immunoblotting as indicated. IκBα and Lamin A/C, respectively, are also shown. (f) Next-generation GSK3 Cytoplasmic and nuclear markers, IκBα and c-Myc, respectively, are also inhibitors decrease the viability of HMMCLs. KMS11, MM1.R and ARP-1 shown. (c) NF-κB-dependent transcription is impaired in HMMCLs treated cells were treated with increasing concentrations of GSK3i XVI or GSK3i with a GSK3 inhibitor. KMS11 cells were treated with GSK3i IX and the XXII. Cell viability was measured by MTS assay at 72 h after treatment. Each steady-state levels of the indicated mRNAs were analysed by real-time value was individually normalized on the DMSO-treated cells, arbitrarily set PCR (s.d., n = 3). The value for the amount of PCR product present in as 100%. Error bars represent s.d., n = 4. Uncropped images of blots are DMSO-treated cells was set as 1. (d) Inhibition of GSK3, Cullin–RING ligases shown in Supplementary Fig. S8. localization of p100 to either the nucleus or cytoplasm inhibits Thus, inefficient clearance of nuclear p100 (due to lack of Fbxw7, NF-κB signalling. mutation of the p100 degron, GSK3 inhibition or mutation of To analyse whether the transcriptional effect is a consequence of the p100 NES) inhibits NF-κB target gene transcription on NF-κB reduced binding of NF-κB proteins at the promoters of endogenous pathway stimulation. target genes, we carried out chromatin immunoprecipitation assays using an anti-RelB antibody. We found that, compared with wild-type Fbxw7α-mediated degradation of p100 is necessary for the p100, expression of p100(Ser707/711Ala) or p100(4×NES) decreased proliferation and survival of multiple myeloma cells the LTβR-induced binding of RelB to NF-κB elements at the NFKB2, Multiple myeloma cells often exhibit and require elevated NF-κB MCP1 and MIP2 promoters (Fig. 4d). RelB association with the activity27. We observed that, in HMMCLs, most p100 is localized promoter of the LTβR-unresponsive gene IL-6 was unaffected by p100 in the cytoplasm, in contrast with Fbxw7α, which is exclusively mutants. The expression of a combined nuclear and stable mutant nuclear (Fig. 5a). To assess the relevance of the Fbxw7α–GSK3–p100 (4×NES; Ser707/711Ala) impaired RelB enrichment to a similar extent regulatory axis in this cancer, we expressed our p100 mutants in as p100(Ser707/711Ala) (Supplementary Fig. S6b). four HMMCLs (KMS11, MM1.R, ARP-1 and PCNY1) that are

ab120 120 120 120 110 110 110 110 100 100 100 shRNA1 shRNA2 shRNA3 shRNA4 shRNA 100 90 90 90 90 80 N.I. LacZ p100 p100 p100 p100 80 80 80 p100 70 70 70 70 60 60 60 60 Skp1 50 50 50 50 40 Percentage of viability Percentage of viability 40 Percentage of viability 40 Percentage of viability 40 0 1 1 5 0 0 10 0 10 50 10 100 500 100 500 0.5 5.0 0.1 0.5 1.0 5.0 100 1,0002,0005,000 1,0005,000 1,000 GSK3i IX (nM) MLN4924 (nM) Bortezomib (nM) Dexamethasone (nM)

LacZ shRNA p100 shRNA1 p100 shRNA2 p100 shRNA4

Figure 8 p100 contributes to the sensitivity of multiple myeloma cells the indicated lentiviruses and treated with increasing concentrations to GSK3 inhibition. (a) Efﬁcient knockdown of p100 in HMMCLs. of GSK3i IX, MLN4924, bortezomib or dexamethasone. Cell viability ARP-1 cells were infected with lentiviruses encoding shRNAs was measured by MTS assay at 72 h after treatment and normalized targeting p100 or a control shRNA against LacZ. Cell extracts were to the untreated LacZ shRNA (0 nM), arbitrarily set as 100%. Error then subjected to immunoblotting as indicated. (b) p100 knockdown bars represent s.d., n = 4. Uncropped images of blots are shown in desensitized ARP-1 cells to GSK3i IX. ARP-1 cells were infected with Supplementary Fig. S8. characterized by elevated NF-κB activity due to mutations in the activity in HMMCLs. Next, we investigated whether inhibition of non-canonical pathway (that is, TRAF3-inactivating mutations)28,29. GSK3 affected the viability of HMMCLs. Similarly to treatment Compared with wild-type p100 or an empty vector, expression with bortezomib or MLN4924, GSK3 inhibition results in the of p100(Ser707/711Ala) resulted in decreased proliferation of all death of HMMCLs (Fig. 7d). Comparable results were obtained four HMMCLs investigated (Fig. 5b,c). Moreover, NF-κB target using next-generation GSK3 inhibitors, such as GSK3i XVI and gene transcription was reduced in cells expressing the stable p100 GSK3i XXII (Fig. 7e,f). mutant (Fig. 5d). Accordingly, expression of p100(Ser707/711Ala) To assess whether the cytotoxic effect of GSK3i IX was attributable decreased cell viability and induced apoptosis (Fig. 5e). Similarly, either to p100 stabilization, we depleted p100 in ARP-1 cells using expression of p100(4×NES) or Fbxw7 knockdown (the latter inducing lentiviruses encoding three different short hairpin RNAs (Fig. 8a). accumulation of p100 in the nucleus) decreased the proliferation rate Cells were then treated with increasing concentrations of GSK3i IX, of HMMCLs (Fig. 5b,f–h). These effects were not due to impairment bortezomib, MLN4924 or dexamethasone and assayed for viability of p100 processing because p52 generation was comparable in cells (Fig. 8b). p100 knockdown desensitized ARP-1 cells to GSK3i IX, expressing wild-type p100 and the p100(Ser707/711Ala) mutant, indicating that the presence of p100 contributes to the cytotoxic depleted of Fbxw7 or treated with GSK3 inhibitors (Fig. 5b and effect of GSK3i IX. Instead, a milder desensitization to MLN4924 Supplementary Fig. S7). and, to an even lesser extent, bortezomib was detectable on p100 Finally, we transduced KMS11 cells with retroviruses encoding p100 knockdown. Finally, p100 depletion had no effect on cells treated with or p100(Ser707/711Ala) and a luciferase reporter cassette. Equal ratios dexamethasone, another clinically relevant compound used for the of relative light units per cell number were injected intraperitoneally treatment of multiple myeloma. or subcutaneously into SCID Beige mice (Fig. 6a–c). The expression of p100(Ser707/711Ala) in KMS11 cells inhibited tumour growth, as DISCUSSION revealed by in vivo imaging. In this study, we show that Fbxw7α promotes degradation of the NF-κB Thus, constitutive p100 degradation by Fbxw7α /GSK3 sustains the inhibitor protein p100. Phosphorylation of p100 on Ser 707 and Ser 711 survival of multiple myeloma cells with aberrant NF-κB activity. by GSK3 is a prerequisite for the p100–Fbxw7α interaction. Although predominantly cytoplasmic, p100 shuttles continuously between the p100 stabilization contributes to the cytotoxic effect of GSK3 cytoplasm and the nucleus. We show that p100 degradation by Fbxw7α inhibition in multiple myeloma cells and GSK3 occurs in the nucleus, independently from NF-κB activation. Bortezomib, an inhibitor of the 26S proteasome, is used for the However, although independent from NF-κB signalling, p100 clearance treatment of multiple myeloma39. Recently, the neddylation inhibitor from the nucleus (through Fbxw7α-mediated degradation or nuclear MLN4924, which blocks the function of all Cullin–RING ligases exclusion by a newly identified nuclear export signal in the C-terminus (including SCFFbxw7α), entered phase I clinical trials for multiple of p100) is a prerequisite for binding of RelB at the promoters myeloma40. As stabilization of nuclear p100 induces cell death in of target genes and efficient activation of NF-κB-dependent gene HMMCLs, we investigated whether chemical inhibition of GSK3 transcription. In fact, deletion of the Fbxw7 gene, inhibition of GSK3, could be used as a therapeutic strategy in multiple myeloma. First, expression of a stable p100 mutant or expression of a constitutively we treated ARP-1 and KMS11 cells with GSK3i IX. Cell fractionation nuclear p100 mutant attenuates the NF-κB response. The function revealed significant stabilization of p100 in the nucleus of these of Fbxw7α in the NF-κB response is consistent with mouse genetic cells, similar to what was observed after bortezomib or MLN4924 data showing that conditional knockout of Fbxw7 in haematopoietic treatment (Fig. 7a,b). Accordingly, NF-κB target gene transcription stem cells induces a significant decrease in the number of both was inhibited after treatment of KMS11 cells with GSK3i IX (Fig. 7c), immature and mature B cells41, which rely on BAFF-mediated NF-κB indicating an important role for GSK3 in sustaining aberrant NF-κB signalling for homeostasis42. Moreover, in agreement with our findings,

NATURE CELL BIOLOGY VOLUME 14 | NUMBER 4 | APRIL 2012 383 © 2012 Macmillan Publishers Limited. All rights reserved. ARTICLES fibroblasts from GSK3-deficient embryos exhibit reduced activation METHODS of NF-κB signalling43. Methods and any associated references are available in the online Whereas phosphorylation of p100 on Ser 707 is constitutive, version of the paper at www.nature.com/naturecellbiology phosphorylation of p100 on Ser 866/870 is induced by NF-κB stimuli, Note: Supplementary Information is available on the Nature Cell Biology website leading to p100 cleavage into p52 (refs 23,24). Accordingly, p52 can be generated from p100(Ser707/711Ala), but not from a p100(Ser866/870) ACKNOWLEDGEMENTS mutant. Interestingly, cytoplasmic p100 is phosphorylated on both The authors thank I. Aifantis, H. J. Cho, G. Franzoso, W. M. Kuehl and A. Ventura for reagents; B. Aranda-Orgilles for her contribution; and I. Aifantis, G. Franzoso, Ser 707 and Ser 866/870, whereas nuclear p100 is phosphorylated only K. Nakayama, S. Reed and J. R. Skaar for critically reading the manuscript. M.P. is on Ser 707. Furthermore, p100 phosphorylated on Ser 707 binds only grateful to T. M. Thor for continuous support. This work was supported in part nuclear Fbxw7α, whereas p100 phosphorylated on Ser 866/870 binds by grant 5P30CA016087-33 from the National Cancer Institute, a fellowship from the American Italian Cancer Foundation and NIH 5T32HL007151-33 to L.B., NIH only βTrCP. Thus, differentially localized and modified fractions of T32 CA009161 grant to S.M. and grants from the National Institutes of Health p100 undergo distinct regulation by the ubiquitin proteasome system. (R01-GM057587, R37-CA076584 and R21-CA161108) and the Multiple Myeloma From yeast to mammals, inducible phosphorylation is the dominant Research Foundation to M.P. M.P. is an Investigator with the Howard Hughes Medical Institute. mechanism controlling the recruitment of substrates to F-box proteins2,5. As phosphorylation of the p100 degron is constitutive, we AUTHOR CONTRIBUTIONS L.B. conceived and directed the project. L.B. and S.E.M. designed and carried out investigated how Fbxw7α-mediated proteolysis of p100 is regulated. most experiments. L.S. and O.O. helped with the mouse experiments. C.K. carried We found that NF-κB proteins are able to induce p100 stabilization out the in vitro experiments shown in Fig. 1d,e and Supplementary Figs S2d and S5b. through Fbxw7α displacement. These observations indicate that the V.B. and K.S.E-J. carried out the mass spectrometry analysis of the Fbxw7α complex purified by L.B. A.H. provided constructs, cell lines and advice. M.P. coordinated nuclear accumulation of NF-κB proteins (p52 and RelB) induced by the work and oversaw the results. L.B., S.M. and M.P. wrote the manuscript. NF-κB signalling may lead to p100 stabilization through competition COMPETING FINANCIAL INTERESTS with Fbxw7α. Thus, we propose a control mechanism, in which The authors declare no competing financial interests. recognition of a substrate by an F-box protein is negatively regulated by accumulation of the substrate’s interacting partners, physically Published online at www.nature.com/naturecellbiology Reprints and permissions information is available online at www.nature.com/reprints disrupting the ligase–substrate interaction. The mechanism of p100 degradation has implications for B-cell 1. Skaar, J. R., D’Angiolella, V., Pagan, J. K. & Pagano, M. S. F box proteins II. Cell 137, e1351–e1358 (2009). malignancies. Constitutive activation of the NF-κB pathway has 2. Cardozo, T. & Pagano, M. The SCF ubiquitin ligase: insights into a molecular machine. been found in many lymphoid diseases and contributes to the Nat. Rev. Mol. Cell Biol. 5, 739–751 (2004). 27–30 3. Petroski, M. D. & Deshaies, R. J. Function and regulation of cullin–RING ubiquitin malignant transformation of B-cell lineages, including plasma cells . ligases. Nat. Rev. Mol. Cell Biol. 6, 9–20 (2005). Pharmacologic inhibition of the proteasome28 or Cullin–RING ligases44 4. Hoeck, J. D. et al. Fbw7 controls neural stem cell differentiation and progenitor apoptosis via Notch and c-Jun. Nat. Neurosci. 13, 1365–1372 (2010). induces apoptosis in multiple myeloma and B-cell lymphoma cell 5. Welcker, M. & Clurman, B. E. FBW7 ubiquitin ligase: a tumour suppressor at lines, partially by inhibiting NF-κB. Here, we have shown that the crossroads of cell division, growth and differentiation. Nat. Rev. Cancer 8, inhibition of GSK3-mediated signalling (for example, by expressing 83–93 (2008). 6. Onoyama, I. et al. Conditional inactivation of Fbxw7 impairs cell-cycle exit a phospho-defective mutant of p100 or through pharmacologic during T cell differentiation and results in lymphomatogenesis. J. Exp. Med. 204, inhibition of GSK3) impairs the survival of HMMCLs both in 2875–2888 (2007). 7. Thompson, B. J. et al. The SCFFBW7 ubiquitin ligase complex as a tumor suppressor cell systems and xenotransplant models. Interestingly, the toxicity in T cell leukemia. J. Exp. Med. 204, 1825–1835 (2007). of GSK3 inhibition can be substantially reduced by knockdown 8. Rajagopalan, H. et al. Inactivation of hCDC4 can cause chromosomal instability. Nature 428, 77–81 (2004). of p100. Thus, the intersection of Fbxw7–GSK3–p100 may serve 9. Millman, S. E. & Pagano, M. MCL1 meets its end during mitotic arrest. EMBO Rep. as a potential intervention point for the treatment of multiple 12, 384–385 (2011). 10. Mavrakis, K. J. et al. A cooperative microRNA-tumor suppressor gene network in myeloma. Moreover, it is likely that the effects observed in multiple acute T-cell lymphoblastic leukemia (T-ALL). Nat. Genet. 43, 673–678 (2011). myeloma may be generalized to other B-cell neoplasms, especially 11. O’Neil, J. et al. FBW7 mutations in leukemic cells mediate NOTCH pathway those addicted to NF-κB. activation and resistance to γ-secretase inhibitors. J. Exp. Med. 204, 1813–1824 (2007). Together, our studies have revealed an unexpected pro-survival role 12. Akhoondi, S. et al. FBXW7/hCDC4 is a general tumor suppressor in human cancer. for Fbxw7. A number of cancers, including T-ALL, breast cancers, Cancer Res. 67, 9006–9012 (2007). 13. Chapman, M. A. et al. Initial genome sequencing and analysis of multiple myeloma. cholangiocarcinoma, gastric adenocarcinoma and head and neck Nature 471, 467–472 (2011). squamous carcinoma7,8,11,45, often carry hemizygous or homozygous 14. Ghosh, S. & Karin, M. Missing pieces in the NF-κB puzzle. Cell 109 Suppl., S81–S96 (2002). mutations in the FBXW7 gene. Most cancer-associated FBXW7 point 15. Senftleben, U. et al. Activation by IKKα of a second, evolutionary conserved, NF-κB mutations result in the accumulation of substrates (for example, Notch, signaling pathway. Science 293, 1495–1499 (2001). c-Myc, cyclin E and Mcl1) due to reduced binding. Interestingly, 16. Dejardin, E. et al. The lymphotoxin-β receptor induces different patterns of gene expression via two NF-κB pathways. Immunity 17, 525–535 (2002). mutations of Fbxw7 have not been detected in 38 multiple myeloma 17. Coope, H. J. et al. CD40 regulates the processing of NF-κB2 p100 to p52. EMBO J. samples13, 24 HMMCLs (L.B., W. M. Kuehl and M.P., unpublished 21, 5375–5385 (2002). 18. Weih, F. & Caamano, J. Regulation of secondary lymphoid organ development results), 70 primary B-ALLs (I. Aifantis, personal communication, by the nuclear factor-κB signal transduction pathway. Immunol. Rev. 195, and ref. 12), 20 B-CLLs (ref. 12) and 13 DLBCLs (ref. 46). Our 91–105 (2003). 19. Ramakrishnan, P., Wang, W. & Wallach, D. Receptor-speciﬁc signaling for both the findings indicate that, in contrast to T cells, B cells are uniquely alternative and the canonical NF-κB activation pathways by NF-κB-inducing kinase. dependent on Fbxw7 function for survival. Although characterized Immunity 21, 477–489 (2004). 20. Zarnegar, B. et al. Unique CD40-mediated biological program in B cell activation as a tumour suppressor, Fbxw7 therefore functions as a pro-survival requires both type 1 and type 2 NF-κB activation pathways. Proc. Natl Acad. Sci. gene in multiple myeloma through its control of NF-κB activity. USA 101, 8108–8113 (2004).

21. Basak, S. et al. A fourth IκB protein within the NF-κB signaling module. Cell 128, 34. Solan, N. J., Miyoshi, H., Carmona, E. M., Bren, G. D. & Paya, C. V. RelB cellular 369–381 (2007). regulation and transcriptional activity are regulated by p100. J. Biol. Chem. 277, 22. Mordmuller, B., Krappmann, D., Esen, M., Wegener, E. & Scheidereit, C. 1405–1418 (2002). Lymphotoxin and lipopolysaccharide induce NF-κB-p52 generation by a co- 35. van Drogen, F. et al. Ubiquitylation of cyclin E requires the sequential translational mechanism. EMBO Rep. 4, 82–87 (2003). function of SCF complexes containing distinct hCdc4 isoforms. Mol. Cell 23, 23. Xiao, G., Harhaj, E. W. & Sun, S. C. NF-κB-inducing kinase regulates the processing 37–48 (2006). of NF-κB2 p100. Mol. Cell 7, 401–409 (2001). 36. Sangfelt, O., Cepeda, D., Malyukova, A., van Drogen, F. & Reed, S. I. Both SCF(Cdc4 24. Fong, A. & Sun, S. C. Genetic evidence for the essential role of β-transducin α) and SCF(Cdc4 γ) are required for cyclin E turnover in cell lines that do not repeat-containing protein in the inducible processing of NF-κB2/p100. J. Biol. Chem. overexpress cyclin E. Cell Cycle 7, 1075–1082 (2008). 277, 22111–22114 (2002). 37. Derudder, E. et al. RelB/p50 dimers are differentially regulated by tumor necrosis 25. Bonizzi, G. et al. Activation of IKKα target genes depends on recognition of specific factor-α and lymphotoxin-β receptor activation: critical roles for p100. J. Biol. Chem. κB binding sites by RelB:p52 dimers. EMBO J. 23, 4202–4210 (2004). 278, 23278–23284 (2003). 26. Muller, J. R. & Siebenlist, U. Lymphotoxin β receptor induces sequential activation 38. Shih, V. F. et al. Kinetic control of negative feedback regulators of NF-κB/RelA of distinct NF-κB factors via separate signaling pathways. J. Biol. Chem. 278, determines their pathogen- and cytokine-receptor signaling specificity. Proc. Natl 12006–12012 (2003). Acad. Sci. USA 106, 9619–9624 (2009). 27. Staudt, L. M. Oncogenic activation of NF-κB. Cold Spring Harb. Perspect Biol. 2, 39. Richardson, P. G. et al. Bortezomib or high-dose dexamethasone for relapsed a000109 (2010). multiple myeloma. New Engl. J. Med. 352, 2487–2498 (2005). 28. Annunziata, C. M. et al. Frequent engagement of the classical and alternative NF-κB 40. Shah, J. J. et al. Phase 1 dose-escalation study of MLN4924, a novel NAE pathways by diverse genetic abnormalities in multiple myeloma. Cancer Cell 12, inhibitor, in patients with multiple myeloma and non-Hodgkin lymphoma. Blood 115–130 (2007). 114, 735–736 (2009). 29. Keats, J. J. et al. Promiscuous mutations activate the noncanonical NF-κB pathway 41. Thompson, B. J. et al. Control of hematopoietic stem cell quiescence by the E3 in multiple myeloma. Cancer Cell 12, 131–144 (2007). ubiquitin ligase Fbw7. J. Exp. Med. 205, 1395–1408 (2008). 30. Razani, B. et al. Negative feedback in noncanonical NF-κB signaling modulates NIK 42. Xie, P., Stunz, L. L., Larison, K. D., Yang, B. & Bishop, G. A. Tumor necrosis factor stability through IKKα-mediated phosphorylation. Sci. Signal. 3, ra41 (2010). receptor-associated factor 3 is a critical regulator of B cell homeostasis in secondary 31. Hao, B., Oehlmann, S., Sowa, M. E., Harper, J. W. & Pavletich, N. P. Structure of lymphoid organs. Immunity 27, 253–267 (2007). a Fbw7-Skp1-cyclin E complex: multisite-phosphorylated substrate recognition by 43. Hoeflich, K. P. et al. Requirement for glycogen synthase kinase-3β in cell survival SCF ubiquitin ligases. Mol. Cell 26, 131–143 (2007). and NF-κB activation. Nature 406, 86–90 (2000). 32. Orian, A. et al. SCF(β)(-TrCP) ubiquitin ligase-mediated processing of NF-κB 44. Milhollen, M. A. et al. MLN4924, a NEDD8-activating enzyme inhibitor, is active in p105 requires phosphorylation of its C-terminus by IκB kinase. EMBO J. 19, diffuse large B-cell lymphoma models: rationale for treatment of NF-κB-dependent 2580–2591 (2000). lymphoma. Blood 116, 1515–1523 (2010). 33. Sato, N., Meijer, L., Skaltsounis, L., Greengard, P. & Brivanlou, A. H. Maintenance 45. Stransky, N. et al. The mutational landscape of head and neck squamous cell of pluripotency in human and mouse embryonic stem cells through activation carcinoma. Science 333, 1157–1160 (2011). of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat. Med. 10, 46. Morin, R. D. et al. Frequent mutation of histone-modifying genes in non-Hodgkin 55–63 (2004). lymphoma. Nature 476, 298–303 (2011).

METHODS representing the spatial distribution of detected photon counts emerging from active Cell culture and drug treatment. MEF, HEK293 and HeLa cells were maintained luciferase within the animal. Images and measurements of bioluminescent signals in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum (FBS). were acquired using a 50% threshold and analysed using Living Image software KMS11, MM1.R, ARP-1, Ly10, SudHL6, Farage and Raji were maintained in RPMI (Xenogen). Image data are shown in photons per second per square centimetre per 1640 medium containing 10% FBS. PCNY1 cells were grown in X-VIVO15 with steradian. Camera settings (integration time, binning, f/stop, field of view) were kept 10% pooled human serum and 1 ng ml−1 of IL-6. To stimulate the LTβR, a mix of constant during all measurements. the 4H8 WH2 (Alexis) and 3C8 (eBiosciences; 10:1) monoclonal antibodies was used. Recombinant human soluble CD40 Ligand/TRAP (BioVision) was used at Purification and analysis of Fbxw7α interactors. HEK293 cells were transfected a final concentration of 50 ng ml−1. Recombinant mouse TNF (BD Pharmingen) with FLAG–HA-tagged Fbxw7α or FLAG–HA-tagged Fbxw7α(1WD40) mutant. was used at a final concentration of 10 ng ml−1. The following drugs were Approximately 5 × 109 cells were treated with 10 µM MG132 for 6 h. Cells were used: proteasome inhibitor MG132 (Peptide Institute; 10 µM final concentration), collected and subsequently lysed in lysis buffer (50 mM Tris–HCl at pH 7.5, 150 mM LMB (Sigma; 20 nM final concentration), BIO ((20Z,30E)-6-bromoindirubin-30- NaCl, 1 mM EDTA, 50 mM NaF, 0.5% NP-40, plus protease and phosphatase oxime; GSK3 inhibitor IX; Calbiochem; from 0.1 to 10 µM), dexamethasone inhibitors). Fbxw7α was immunopurified with anti-FLAG agarose resin (Sigma) and (Sigma; 0.1 nm–1 µm), MLN4924 (Active Biochem; 1 nm to 5 µm) and bortezomib washed five times with lysis buffer (15 min each). After washing, proteins were eluted (Velcade; Millennium Pharmaceuticals; 0.1–100 nm). MTS assays (Promega) and twice by competition with FLAG peptide (Sigma). Ninety per cent of the eluate was Annexin V/7AAD staining (BD) were carried out according to manufacturers’ then subjected to a second immunopurification with an anti-HA resin (Covance), protocols. For the MTS assay, 2×103 cells were plated three days after infection in before elution by competition with HA peptide (Roche). The two eluates (10% quadruplicate. EdU incorporation and click reaction with biotin–azide was carried of the first immunoprecipitate and 100% of the second immunoprecipitate) were out according to the manufacturer’s instruction (Invitrogen). separated by SDS–PAGE, and proteins were stained with colloidal blue (for mass spectrometry analysis) or silver stain. Bands from the single anti-FLAG purification Biochemical methods. Extract preparation, immunoprecipitation and im- and sequential anti-FLAG, anti-HA purification were excised from the gel and munoblotting were previously described47,48. Fractionation of MEFs was carried digested overnight in situ with trypsin (Sigma). The peptide solution was prepared out according to ref. 21. Fractionation of HMMCLs was carried out as described for mass spectrometry analysis on an Orbitrap LC–MS. in ref. 21, except that the cytoplasmic extract buffer contained 0.02% NP-40. The following antibodies were used: anti-p100 N-terminus (Cell Signaling Technology, Transient transfections and retrovirus-mediated gene transfer. HEK293 cells #18D10, and RICE1495 NCI-BRB preclinical repository), anti-p100 C-terminus were transfected using polyethylenimine (PEI). For retrovirus production, GP-293 (RICE1310 from NCI-BRB preclinical repository), anti-p105 (Santa Cruz, sc-8414), packaging cells (Clontech) were used. Forty-eight hours after transfection, the anti-Fbxw7α (Bethyl, A301-720A), anti-RelA (Santa Cruz, sc-372 and sc-8008), virus-containing medium was collected and supplemented with 8 µg ml−1 Polybrene anti-RelB (Santa Cruz, sc-226 and sc-48366), anti-IκBα (Santa Cruz, sc-371), anti- (Sigma). Cells were then infected by replacing the cell culture medium with the p100 phospho-Ser866/870 (Cell Signaling Technology, #4810), anti-Skp1 (Santa viral supernatant for 6 h. siRNA oligonucleotide transfection was carried out with Cruz, sc-5281), anti-c-Myc (Santa Cruz, sc-764), anti-c-Myc phospho-Thr58/Ser62 HiPerfect (Quiagen) according to the manufacturer’s procedure. hFBXW7 siRNA (Cell Signaling Technology, #9401), anti-β-catenin phospho-Ser33/37/Thr41 sequences are listed in Supplementary Table S1. (Cell Signaling Technology, #9561), anti-CyclinE (clone HE12), anti-FLAG M2 (Sigma, F3165), anti-Lamin A/C (Cell Signaling Technology, #2032) and anti-HA Immunofluorescence microscopy. Direct and indirect immunofluorescence (Covance, HMB-45). The anti-phospho-Ser707 antibody was produced in rabbits microscopy was carried out as described previously48. Primary antibodies (anti-N- by immunization with a phospho-Ser707 peptide corresponding to amino acids terminal p100, anti-C-terminal p100 and anti-HA) were used at a dilution of 1:1,000. 704–720 (C-PLPpSPPTSDSDSDSEGP) of human p100. Streptavidin–FITC was from Jackson ImmunoResearch, 016-010-084. Primary antibodies were used at a Plasmids. Generation of lentivirus encoding shRNAs targeting human Fbxw7 was dilution of 1:1,000. carried out as previously described48. The target sequences used to knockdown human Fbxw7 and p100 are described in Supplementary Table S1. In vitro ubiquitylation, kinase and binding assays. GST–p52 and p100 were expressed in Escherichia coli (BL-21) and SCFFbxw7α in Sf9 insect cells. Proteins were pu- Messenger RNA analysis. RNA was extracted using the RNeasy Kit (Qiagen). rified as described previously47,48. The ubiquitylation of in vitro-translated [35S]p100 Complementary DNA synthesis was carried out using Superscript III (Invitrogen). was carried out as described previously48. The ubiquitin reaction mix contains E1 Quantitative PCR analysis with SYBR green (Roche) was carried out according to (100 nM), UbcH3 (10 ng µl−1), UbcH5c (10 ng µl−1), GSK3 (10 ng µl−1), ubiquitin standard procedures. Primer sequences are listed in Supplementary Table S1. (2.5 µg µl−1) and ATP (2 mM). For in vitro kinase assays, in vitro-translated p100 was incubated at 30 ◦C for 30 min, with 2 mM ATP and 100 ng GSK3β, in 15 µl of kinase Chromatin immunoprecipitations. MEFs were washed twice with PBS and buffer (50 mM Tris at pH 7.5, 10 mM MgCl2 and 0.6 mM dithiothreitol). The reac- crosslinked with 1% formaldehyde at room temperature for 10 min. Cells then were tion was stopped by the addition of Laemmli buffer, and the products were subjected rinsed with ice-cold PBS twice and sonicated in lysis buffer (1% SDS, 10 mM EDTA, to immunoblotting. For in vitro binding assays, in vitro-translated, FLAG-tagged 50 mM Tris-HCl, at pH 8.1, and protease inhibitor cocktail) in a Bioruptor XL (Di- p100 and p100 mutants were immunopurified on anti-FLAG resin and subjected to agenode), followed by centrifugation for 10 min. Supernatants were collected and di- a kinase reaction (15 min) with purified GSK3β. Samples were washed three times in luted in buffer (1% Triton X-100, 2 mM EDTA, 150 mM NaCl and 20 mM Tris–HCl, lysis buffer (50 mM Tris–HCl at pH 7.5, 250 mM NaCl 0.1% Triton X-100 and 1 mM at pH 8.1) followed by immunoclearing with Rabbit IgG and protein A–Sepharose EGTA) to remove GSK3β, and in vitro-translated Fbxw7α was added to the beads. for 2 h at 4 ◦C. Immunoprecipitations were carried out by incubating the super- Samples were then washed three times with lysis buffer, and complexes were eluted natants overnight at 4 ◦C with an anti-RelB antibody (Santa Cruz, sc-226) used at in Laemmli buffer. Finally, the reactions were subjected to western blot analysis. For a dilution of 1:200. After immunoprecipitation, protein A–Sepharose was added, the p52 competition assay, increasing amounts of recombinant p52 were added to and the incubation was continued for 1 h. Precipitates were washed sequentially for phosphorylated p100 and incubated for 1 h. The p100:p52 complex was washed three 10 min each in buffer I (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris–HCl, times in lysis buffer, and in vitro-translated Fbxw7α was added to the beads. at pH 8.1, and 150 mM NaCl), buffer II (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris–HCl, at pH 8.1, and 500 mM NaCl) and buffer III (0.25 M LiCl, 1% Xenotransplantation experiments. Multiple myeloma cells were transplanted NP-40, 1% deoxycholate, 1 mM EDTA and 10 mM Tris–HCl, at pH 8.1). Precipitates into six- to eight-week-old recipient mice by intraperitoneal or subcutaneous were then washed three times with TE buffer and extracted three times with 1% SDS ∼ × 6 / µ ◦ injection of equal ratios of relative light units per cell number ( 1 10 cells 250 l). and 0.1 M NaHCO3. Eluates were pooled and heated at 65 C for at least 6 h to reverse In vivo bioluminescence imaging was conducted on a cryogenically cooled IVIS the formaldehyde crosslinking. DNA fragments were precipitated with a standard system (Xenogen). Ten to fifteen minutes before in vivo imaging, animals were phenol/chloroform protocol. Quantitative PCR analysis was carried out according − injected intraperitoneally with a saline solution of 15 mg ml 1 of d-luciferin to the standard procedure. Primer sequences are listed in Supplementary Table S1. (Caliper) at a dose of 150 mg kg−1 body weight and five minutes later were anaesthetized with a mixture of 2% isofluorane/air. During image acquisition, 47. Busino, L. et al. Degradation of Cdc25A by β-TrCP during S phase and in response isofluorane anaesthesia was maintained using a cone delivery system integrated with to DNA damage. Nature 426, 87–91 (2003). the IVIS system. A greyscale body surface image was collected in the chamber under 48. Busino, L. et al. SCFFbxl3 controls the oscillation of the circadian clock by directing dim illumination followed by acquisition and overlay of the pseudocolour image the degradation of cryptochrome proteins. Science 316, 900–904 (2007).

DOI: 10.1038/ncb2463

FLAG-HA! FLAG-HA! Fbxw7α! Fbxw7α(WD40)! Accession #! Protein! Fbxw7 Fbxw7(WD40) Fbxw7 Fbxw7(WD40)

A! ! C !

! ! ! ! ! ! ! Single IP! Single IP! Double IP! Double IP! IPI00056349! FBXW7! 0.0443! 0.0415! 0.0889! 0.0902! IPI00301364! SKP1! 0.0428! 0.052! 0.0842! 0.0596! IPI00014310! CUL1! 0.0006! 0.0015! 0.0088! 0.0069!

IPI00807463! NFKB2! 0.0002! 0! 0.0005! 0!

HA elution 2 (10%) 2 elution HA HA elution 2 (10%) 2 elution HA HA elution 3 (10%) 3 elution HA HA elution 3 (10%) 3 elution HA HA elution 1 (10%) 1 elution HA (10%) 1 elution HA FLAG elution (1%) elution FLAG (1%) elution FLAG

Mr (K)! 250! D! p100 mutants! 160! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! F!

105! !

70! *Fbxw7α! ! p100 (1-703) (1-618) (1-448) (1-400) (1-300) (50-900) (100-900) (150-900) (250-900) (300-900) (1-720) (1-780) (1-800) (1-840) (1-860) EV 50! Fbw7α

35! ! (α-HA)!

30! -FLAG

α p100 25! IP: (α-FLAG)!

10! p100 mutants! E! G! Fbxw7α binding! 1! 900! +! 50! 900! +! 100! 900! +! ! Fbw7α (α-HA)! B! 150! 900! +! +! Skp1! 250! 900! -FLAG 300! +! α 900! α

IP: p100 ( -FLAG)! 1! 860! +! 1! 840! +! Cyclin E! 1! 800! +! Fbw7α (α-HA)! +! ! Cleaved-Notch! 1! 780! 1! 740! +! Skp1! Input Skp1! 1! 720! +! α Cul1! 1! 703! -! p100 ( -FLAG)! 1! 618! -! Fbxw7a (α-FLAG)! 1! 448! -! IP: α-FLAG! 1! 300! -! Figure S1!

Figure S1 Purification of the Fbxw7a complex and identification of the intracellular Notch1), although it still interacted with core components of the p100 degron. (a) Purification of the Fbxw7a complex. HEK293 cells were SCF ligase, such as Skp1 and Cul1. (c) The table shows mass spectrometry transfected with constructs encoding either FLAG-HA tagged Fbxw7a or analysis of four Fbxw7a immunoprecipitations, listing normalized spectral FLAG-HA tagged Fbxw7a(WD40). Proteins were immunoprecipitated (IP) abundance factors for the proteins indicated. (d) p100 binds Fbxw7a with anti-FLAG resin (α-FLAG), and a second immunoprecipitation was through its C-terminal domain. HEK293 cells were transfected with performed with an anti-HA antibody, as indicated. Immunocomplexes were constructs encoding HA-tagged Fbxw7a and FLAG-tagged p100 deletion eluted with 1% SDS, precipitated, and resolved by SDS-PAGE. The gel was mutants. p100 was immunoprecipitated from cell extracts with anti-FLAG then stained with silver stain for protein visualization. (b) Fbxw7a binds to resin (α-FLAG), and immunocomplexes were probed with antibodies to the substrates through specific residues in the WD40 domain. HEK293 cells indicated proteins. (e) Schematic representation of p100 mutants. p100 were transfected with cDNAs encoding either FLAG-HA tagged Fbxw7a mutants that interact with Fbxw7a are designated with the symbol (+). (f) or FLAG-HA tagged Fbxw7a(WD40), a previously-established, substrate Alignment of the p100 degron with the degrons of other Fbxw7 substrates. binding mutant (Hao et al. Mol Cell 26, 131-143, 2007), in which three Conserved serine, threonine or negatively charged residues are highlighted residues (Ser462, Thr463, and Arg465) within one of the seven WD40 in gray. (g) p100 binds Fbxw7a through Ser707 and Ser711. HEK293 cells repeats in Fbxw7a are changed to Ala. Proteins were immunoprecipitated were transfected with constructs encoding HA-tagged Fbxw7a and FLAG- (IP) from cell extracts with anti-FLAG resin (α-FLAG), and immunoblotting tagged p100 mutants. p100 was immunoprecipitated from cell extracts was performed for the indicated proteins. In contrast to WT Fbxw7a, with anti-FLAG resin (α-FLAG), and immunocomplexes were probed with Fbxw7a(WD40) was unable to bind its substrates (e.g., cyclin E and antibodies to the indicated proteins.

IP: α-FLAG ! C! D! + GSK3β! B! p100 mutants! DMSO! MG132!

! ! α ! ! ! ! +! +! +! ! A! ! -! -! Fbxw7 -! +! -! +! -! FLAG-p100! IP: α-FLAG! ! +! +! FLAG-p100(Ser707/711Ala)!

WD40) WD40) -! -! -! λ + -! -PPase! ( (

! ! (S707A) p100 (T710A) (S711A) (S713A) (S715A) (S717A) p100 (pSer707)! α α α α Fbxw7α p100 (pSer707)! α p100 (α-FLAG)! p100 ( -FLAG)! TrCP2 TrCP2

Fbxw7 Fbxw7 p100 (pSer866/870)! Fbxw7 β Fbxw7 β α p100 (pSer707)! Input! IP: -FLAG! E! p100 (α-FLAG)! p100 (pSer866/870)! cMyc (pThr58/Ser62)!

GST-MYC! Cul1 F! IkBα! Fbxw76HIS-FLAG! WT MEFs! Nfκb2-/- MEFs! Skp1! Fbxw7α Green: α-p100/52 ! α-FLAG! Skp1GST-T7! (N-terminus)! βTrCP2! G! Blue: DAPI! untreated! + Leptomycin B! ! ! Green: α-p100!

(C-terminus)! p100

Rbx1! Blue: DAPI! -C-terminus) α (

H! siRNA! ! ! siLacZ! ! ! ! ! 100! α ! ! ! 1.8! siFbxw7 ! α β γ α β γ ! ! ! 80! G1! siFbxw7β ! 60! S-G2/M! 1.3! siFbxw7γ !

LacZ Fbxw7 Fbxw7 Fbxw7 Fbxw7(Pan) LacZ Fbxw7 Fbxw7 Fbxw7 Fbxw7(Pan) EdU positive! 40! siFbxw7(pan) ! p100! % of Cells 20! 0.8! ! Lamin A/C! ! ! Fold Change

0! α β γ ! ! κ α ! 0.3!

I B ! ! Cytoplasmic! Nuclear!

siLacZ -0.2! Fraction! Fraction! siLacZ siFbxw7 siFbxw7 siFbxw7 siFbxw7 (Pan) Serum Starved (72h)! 10% FBS! Serum Starved (72h)! Figure S2!

Figure S2 p100 is phosphorylated on Ser707 and Fbxw7a controls and Rbx1. Cell lysates were subjected to Ni-NTA affinity purification. An p100 stability in the nucleus. (a) The p100 (pSer707) antibody is aliquot of the purified complex was analyzed on SDS-PAGE, and the gel was phosphorylation specific. A phospho-specific antibody against the stained with Coomassie Blue. (f) Specificity of antibodies to the C-terminus 704PLPpSPPTSDSDSDSEGP720 peptide with a phospho-serine at position or N-terminus of p100. WT and Nfαb2 -/- MEFs were fixed and stained with 707 was generated. HEK293 cells were transfected with a construct the indicated antibodies. DAPI is shown in blue. The antibody specifically encoding FLAG-tagged p100. p100 was immunoprecipitated from recognizing a C-terminal epitope of p100 distinguishes p100 from p52 in cell extracts with anti-FLAG resin (α-FLAG). Immunocomplexes were immunofluorescence studies. Scale bar, 20 μm. (g) p100 shuttles between incubated for 1 hour at 30°C with or without l-phosphatase and blotted the cytoplasm and the nucleus. MEFs were treated with Leptomycin B (LMB) with the indicated antibodies. (b) Phosphorylation site specificity of the for 4 hours, fixed, and stained with an antibody against the C-terminus of p100 (pSer707) antibody. HEK293 cells were transfected with constructs p100 (green). Scale bar, 20 μm. (h) p100 is targeted by Fbxw7a in the encoding different FLAG-tagged p100 mutants. Cell extracts were nucleus of resting, primary human fibroblasts. IMR90 cells were transfected immunoblotted with the indicated antibodies. (c) Fbxw7a binds p100 with either an siRNA targeting all Fbxw7 isoforms (pan;), or isoform-specific phosphorylated on Ser707, but not Ser866 and Ser870. HEK293 cells Fbxw7a/b/g siRNAs. Cells were serum starved in 0.1% FBS for 72 hours. were transfected with the indicated FLAG-tagged constructs and treated Sub-cellular fractionation was performed and lysates were subjected to with MG132 (10 mM) or DMSO for 6 hours. FLAG-tagged proteins were immunoblotting for the indicated proteins (left panel). Growth arrest following immunoprecipitated from cell extracts and probed with antibodies to the serum starvation was confirmed by EdU incorporation and Propidium Iodide indicated proteins. (d) GSK3-mediated phosphorylation of p100 is required (PI) staining followed by FACS analysis (middle panel). The G1 bar indicates for the in vitro binding of p100 to Fbxw7a. In vitro-translated, FLAG-tagged the percentage of cells with a 2N DNA content and the S-G2/M bar indicates p100 or p100(Ser707/711Ala) was incubated with GSK3b before incubation the percentage of cells with a DNA content between >2N and <4N (as with in vitro-translated Fbxw7a. Proteins were immunoprecipitated (IP) determined by PI stain). The percentage of actively replicating cells (positive using anti-FLAG resin (α-FLAG), and immunoblotting for the indicated for EdU staining) is shown in green. Isoform-specific knockdown of Fbxw7 proteins was performed. 30% of inputs are also shown. (e) Insect cells (Sf9) was confirmed by qPCR (right panel) using amplification primers recognizing were co-infected with baculoviruses encoding His-Fbxw7a, Skp1, Cul1, the unique 5’ exon of each isoform.

Nfκb2-/- MEFs! B! Untreated! + LMB! A! p100 (α-HA)! DAPI! Merge! p100! HA-p100(ΔNES) !

p100(Ser707/711Ala)! HA-p100(ΔNES;! Ser707/711Ala) !

p100 (NLS)!

α-p100 (C-terminus) !

C! D!

Nfκb2-/- MEFs! HA-p100 + Leptomycin B!

! 1.2! p100! p100! siLacZ! MG132 ! siFbxw7! p100! (Ser707/711Ala)! p100! (Ser707/711Ala)! 1! 0.8! 0 1 2 4 0 1 2 4 0 1 2 4 !CHX (hours)! 0 3 6 12! 0 3 6 12! 0 3 6 12! 0 3 6 12! CHX (hours)! α 0.6! p100 ( -HA)! p100! 0.4! Fold Change Skp1! p100 (pSer707)! 0.2! ! 0! ! IκBα! p100 p100

Histone H3!

Skp1!

Cytoplasmic Fraction (15µg)! Nuclear Fraction (15µg)! (Ser707/711Ala) Figure S3!

Figure S3 p100 stability is dependent on its cellular localization. (a) and analyzed by immunoblotting for the indicated proteins. mRNA levels Mutation of Ser707 and Ser711 to Ala induces the accumulation of of retrovirally-expressed p100 were analyzed by qPCR (right panel). In p100(ΔNES), a constitutively nuclear mutant of p100. HeLa cells were the nucleus, p100 half-life was approximately three hours, whereas in infected with retroviruses expressing either HA-tagged p100(ΔNES) the cytoplasm p100 remained stable after 12 hours of CHX treatment. or HA-tagged p100(ΔNES;Ser707/711Ala). Cells were fixed and A similar difference was observed for p100 phosphorylated on Ser707, stained with an antibody against the HA tag (a-HA). Scale bar, 20 showing that, in contrast to nuclear p100, cytoplasmic p100 is stable μm. (b) p100 requires an intact NLS for nuclear-cytoplasmic shuttling. even when phosphorylated on the Fbxw7a degron. The Ser707/711Ala Nfkb2-/- MEFs were infected with retroviruses expressing either mutations increased both the steady state level (time 0) and the stability p100, p100(Ser707/711Ala), or p100(NLS). Cells were treated with of nuclear p100. IkBa half-life is included as control for CHX. IkBa Leptomycin B (LMB) for four hours, fixed, and stained with an antibody and Histone H3 are used as controls of fractionation. (d) The half-life of against the C-terminus of p100 (green). Scale bar, 20 μm. (c) Ser707 nuclear p100 is regulated by Fbxw7a and the proteasome. HeLa cells and Ser711 are required for the degradation of nuclear p100. Nfαb2 were infected with retroviruses expressing HA-tagged p100 2. Cells were -/- MEFs were infected with retroviruses expressing either p100 or treated with Leptomycin B (LMB) for four hours, and cycloheximide (CHX) p100(Ser707/711Ala). Cells were treated with cycloheximide (CHX) for was then added for the indicated times. Cell extracts were analyzed by the indicated times. Cytoplasmic and nuclear fractions were isolated immunoblotting for the indicated proteins.

C! N!

A! ! B! + GSK3β! C! ! 18h 18h ! ! β β NT NT IgG FLAG-p100! LT LT Mr (K)! p100 (pSer707)! HA-RelB! -! +! +! +! +! p52! 250! p100! Fbxw7α (short exp.)! FLAG-p100 (Ser707/711Ala) - 150! RelB! Fbxw7α! Fbxw7α (long exp.)! 100! IP: α-RelB! α 75! RelB ( -HA)! p100 (pSer707) C! N! p100 (α-FLAG)! p100 (α-FLAG)! 50! ! 18h 18h α ! ! IP -FLAG! β β 37!

α NT NT - IgG

IP: FLAG! LT LT p100 (pSer707)! 25! p100! D! 20! RelA! IP: α-RelA! ! Deoxycholate!

! Wash! ! ! ! ! ! 0 EV 0.3 0.5 0.8 IP: α-FLAG-p100! WCE p100 (α-Flag)

Deoxycholate! RelB wash! IP: α-Flag RelB bound to total p100 !

α IP: α-HA-Fbxw7! ! Deoxycholate!

! Wash! ! ! ! HA elution! ! ! 0 EV 0.3 0.5 0.8 WCE HA-elution

p100 (α-Flag) IP: α-Flag-p100! RelB

RelB bound to p100-Fbxw7 Deoxycholate! IP: α-Flag wash! Figure S4!

Figure S4 p52 and RelB compete with Fbxw7a for binding to p100. A similar result was observed for the fraction of p100 phosphorylated on (a) RelB competes with Fbxw7a for binding to p100 in vivo. HEK293 Ser707. These data indicate a distinct regulation of the cytoplasmic and cells were transfected with cDNAs encoding either FLAG-tagged p100 or nuclear pools of p100 with regard to its binding to NF-kB proteins. (d) p100 FLAG-tagged p100(Ser707/711Ala). Increasing amounts of a plasmid bound to Fbxw7a associates to NF-kB proteins via RHD:RHD interactions. encoding HA-tagged RelB were also transfected as indicated. Proteins were HEK293 cells were transfected with FLAG-tagged p100 or an empty vector immunoprecipitated (IP) from cell extracts with anti-FLAG resin (α-FLAG), (EV) either alone or in combination with HA-tagged Fbxw7a. To analyze total and immunoblotting was performed for the indicated proteins. The first lane RelB-p100 complexes (top panel), FLAG-p100 was immunoprecipitated with shows cells transfected with an empty vector (EV). (b) p52 competes with an a-FLAG resin. The IP was divided in four tubes and then washed with Fbxw7a for binding to p100 in vitro. In vitro translated, FLAG-tagged p100 lysis buffer containing the indicated concentrations of deoxycholate (DOC). To was immunoprecipitated and incubated at 30°C with purified, recombinant analyze RelB-p100 complexes bound to Fbxw7a (bottom panel), HA-tagged GSK3b. The indicated amounts of purified, recombinant p52 were added and Fbxw7a was immunoprecipitated with an a-HA resin. The IP was washed and allowed to bind immunoprecipitated p100. Unbound p52 was washed off, eluted using the HA peptide. 5% of the eluate was resolved by SDS-PAGE and in vitro translated Fbxw7a was incubated with the p100-p52 complexes. (designated “HA-elution”). The remaining eluate was subjected to a second After washing, immunoblotting for the indicated proteins was performed. round of immunoprecipitation with an a-FLAG resin to isolate FLAG-p100 In the right panel the purified, recombinant p52 is shown. Recombinant complexes. The IP was divided in four tubes and then washed with lysis buffer p52 was expressed as GST-p52 in E. coli. Purified GST-p52 was cleaved containing the indicated concentrations of DOC. IPs were resolved by SDS- with PreScission protease, and an aliquot of the eluted, purified protein PAGE and immunoblotted for the indicated proteins. The presence of DOC in was analyzed by SDS-PAGE and Coomassie Blue staining. Arrow indicates wash buffer has been demonstrated to interfere with RHD:ARD interactions, the expected size of full-length p52.(c) Following LTαR activation, p100 but not with stronger RHD:RHD interactions. The RHDs of NF-kB proteins binding to NF-kB proteins decreases in the cytoplasm and increases in the bind p100 either through its RHD (i.e. forming a complex that is DOC- nucleus. MEFs were treated with a-LTbR for 18 hours. Cells were harvested, insensitive) or its ARD (i.e. forming an inactive complex that is DOC-sensitive). fractionated, and endogenous RelA and RelB were immunoprecipitated from We found that, in contrast to total p100, which forms both DOC-sensitive both cytoplasmic and nuclear fractions. IgG was used as a negative control. and DOC-insensitive complexes, the fraction of p100 bound to Fbxw7a only Immunocomplexes were blotted for the indicated proteins. Following a-LTbR forms DOC-insensitive complexes with RelB, indicating that the latter binds ligation, p100 binding to RelA and RelB was decreased in cytoplasmic the RHD of p100 and not its ARD. A schematic representation of RelB:p100 fractions, whereas more p100 associated with RelA and RelB in the nucleus. complexes is shown in the right panels.

A! B! ! ! ICAM1! ! 6! NFKBIE!

! 10! 10! 15! MCP-1! RANTES! 4! 10! 5! 5! 2! 5!

Fold Change α Fold Change 0! 0! Fbxw7 Fold Change Fold Change 0! 0! 0! 1! 3! 6! 0! 1! 3! 6! Skp1! 0! 6! 12! 0! 6! 12! CD40L (hours)! CD40L (hours)! α β -LT R (hours)! α-LTβR (hours)! ! ! ! 6! 6! RELB! 3! CD74! VCAM-1! 4! shLacZ! 4! 2! DMSO! shFbxw7! 2! 1! 2! GSK3i IX! Fold Change Fold Change Fold Change 0! 0! 0! 0! 1! 3! 6! 0! 1! 3! 6! 0! 6! 12! CD40L (hours)! CD40L (hours)! α-LTβR (hours)!

DMSO! C! DMSO! D! Fbxw7 ﬂox/ﬂox ! E!

! GSK3i IX! GSK3i IX! Fbxw7 -/-! ! 2! VCAM1! 8! IL-6! ! ! 6! NFKB2! 1.5! 6! 6! TNF ! ! ICAM1! 4! 4! 4! 1! 4! 3! 0.5! 2! Fold Change Fold Change 2! 2! 2! 0! 0! Fold Change 1! Fold Change 0! 0! 0! 1! 6! 12! 0! 1! 6! 12! Fold Change 0! 0! 1! 6! 12! 24! 0h! 1h! 6h! 12h! TNF (hours)! TNF (hours)! 0! 1! 3! 6! TNF (hours)! TNF (hours)! ! 6! ! 8! CD40L (hours)! TNF! MCP1! 6! 4! ! ! ! TNFAIP3! 15! IL-6 ! 4! 4! 4! VCAM-1 ! 2! 2! 3! 10! Fold Change Fold Change 2! 2! 0! 0! 5! 0! 1! 6! 12! 0! 1! 6! 12! 1! Fold Change Fold Change Fold Change 0! 0! 0! TNF (hours)! TNF (hours)! 0! 1! 3! 6! 0! 1! 6! 12! 24! 0h! 1h! 6h! 12h! CD40L (hours)! TNF (hours)! TNF (hours)! Figure S5!

Figure S5 Clearance of nuclear p100 is required for efficient activation of RAMOS cells were treated with GSK3i IX (5 mM) stimulated with recombinant NF-kB signaling pathways. (a) LTbR-dependent gene transcription is impaired CD40L (50 nM), and processed as in (b). The levels of the indicated mRNAs upon GSK3 inhibition. MEFs were treated with GSK3i IX (5 mM), stimulated were determined by quantitative real time PCR (+ s.d., n=3). The value with agonistic a-LTbR antibodies, and harvested at the indicated times. The given for the amount of PCR product present in DMSO treated cells at time levels of the indicated mRNAs were determined by quantitative real time 0 was set to 1. (d) TNF-dependent gene transcription is inhibited by loss of PCR (+ s.d., n=3). The value given for the amount of PCR product present Fbxw7. Fbxw7 flox/flox and Fbxw7 -/- MEFs were stimulated with purified, in DMSO treated MEFs at time 0 was set to 1. (b) CD40-dependent gene recombinant murine TNF (10ng/ml) and harvested at the indicated times. The transcription is impaired upon Fbxw7 silencing. RAMOS cells were infected levels of the indicated mRNAs were determined by quantitative real time PCR with lentivirus encoding shRNA (targeting either LacZ or Fbxw7) and a GFP (+ s.d., n=3). The value given for the amount of PCR product present in Fbxw7 reporter. GFP-positive cells were sorted, stimulated with recombinant human flox/flox MEFs at time 0 was set to 1. (e) TNF-dependent gene transcription CD40L (50 nM), and finally harvested at the indicated times. The levels of the is impaired upon GSK3 inhibition. MEFs were treated with GSK3i IX (5 mM), indicated mRNAs were determined by quantitative real time PCR (+ s.d., n=3). stimulated with recombinant TNF, and harvested at the indicated times. The The value given for the amount of PCR product present in LacZ at time 0 was levels of the indicated mRNAs were determined by quantitative real time set to 1. Knockdown of Fbxw7a was confirmed by immunoblot (right panel). PCR (+ s.d., n=3). The value given for the amount of PCR product present in (c) CD40-dependent gene transcription is impaired upon GSK3 inhibition. DMSO treated cells at time 0 was set to 1.

A! ! 120! 100! !

! 80! 60! genes R responsive β 40! % of activation LT -

α 20! 0! MCP1! NFKB2! RANTES! BAFF! ! ! 150! ! R stimulation (12 hours) p100 ! β 100! B target LT

- p100(4xNES)! κ α

genes 50! p100(NLS)!

% of activation 0! non-NF- SPARC!

! 6! ! ! NFKB2 promoter! 3! MCP1 promoter! 6! MIP2 promoter!

4! 2! 4!

2! 1! 2! Fold Change Fold Change Fold Change

0! 0! 0! 0! 6! 0! 6! 0! 6! α-LTβR (hours) α β -LT R (hours) α-LTβR (hours)

! 2! IL-6 promoter! EV! 1.5! p100 ! 1! p100(Ser707/711Ala)! p100(4xNES)! 0.5! Fold Change p100(4xNES;Ser707/711Ala)! 0! 0! 6! α-LTβR (hours)

Figure S6 Stabilization or mislocalization of p100 inhibits NF-kB LTbR stimulation. MEFs retrovirally expressing empty vector (EV), p100, signaling. (a) Forced localization of p100 in either the nucleus or p100(Ser707/711Ala), p100(4xNES), or p100(4xNES;Ser707/711Ala) cytoplasm inhibits LTbR-dependent NF-kB target gene transcription. MEFs were stimulated for six hours with agonistic a-LTbR antibodies. Chromatin retrovirally expressing p100, p100(4xNES), or p100(NLS) were stimulated immunoprecipitation (ChIP) was performed using an anti-RelB antibody with agonistic a-LTbR antibodies and harvested 12 hours later. The and immunoprecipitated DNA was amplified by real-time PCR with primers levels of the indicated mRNAs were determined by quantitative real time flanking NF-kB elements in the indicated gene promoters (+ s.d., n=3). PCR (+ s.d., n=3) and normalized to percent of activation for p100 WT The LTbR unresponsive IL-6 promoter was used as negative control. The infected cells. (b) Stabilization or mislocalization of p100 in the nucleus value given for the amount of PCR product present in empty vector infected compromises RelB association with NF-kB target gene promoters upon MEF cells was set to 1.

A! GSK3i IX! GSK3i XXII! Fbxw7α! 0 6 12 24 6 12 24! hrs! p100! p100!

p52! p52! p100 (pSer707)!

IκBα IκBα Skp1! Skp1! KMS11! MM1.S! KMS11!

NIK (µg)!

p100!

p52! NIK!

Skp1!

Figure S7 Processing of p100 to p52 is independent of the Fbxw7a- Mutation of the Fbxw7a phospho-degron does not affect NIK-inducible GSK3 axis. (a) Processing of p100 to p52 in HMMCLs is unaffected by processing of p100 to p52. HEK293 cells were transfected with p100, knockdown of Fbxw7 or inhibition of GSK3. Left panel: KMS11 and p100(Ser707/711Ala), p100(Ser866Ala), or an empty vector (EV), along MM1.S cells were infected with lentiviruses encoding shRNAs targeting with the indicated amounts of NIK cDNA. Cells were harvested at 24 LacZ or FBXW7 (two different targeting sequences). After 72 hours, cells hours after transfection and western blot analysis was performed for the were harvested and analyzed by immunoblot for the indicated antibodies. indicated proteins. The results show that p100 and p100(Ser707/711Ala) Right panel: KMS11 were treated with either GSK3i IX or GSK3i XXII (2 are processed to a similar extent upon NIK expression. A p100 mutant μM and 1 μM, respectively). Cells were harvested at the indicated times, defective for inducible processing, p100(Ser866Ala), is shown as a and subjected to immunoblot analysis for the indicated proteins. (b) control.

Fig.1a! Fig. 1b! Fig. 1c!

Mr (K)! Mr (K)! Mr (K)! 150! 150! HA! 100! 100! 100! 75! 75! 75! p100! 150! Skp1! 50! 20! Fbxw7α! 100! FLAG! 75! 100! p105! 75! 50! 150! 37! 100! 100! HA! p100(pSer707)! β 75! 25! TrCP1!

37! 20! Skp1! IκBα! 100! FLAG!

25! 75!

20! Skp1!

250! 150! 100! 75!

50! 37!

25! 20! FLAG! Figure S8!

Figure S8 Full scans of the key immunoblots. Boxes indicate cropped images used in the figures and numbers indicate the molecular weight (KDa).

Fig. 1e! Fig. 1d! Mr (K)! FLAG! 150! 150! p100! 100! 100! 250! 75! 150! 150! 100! p100 100! 75! (pSer707)! 75! 50! Fbw7α

Fig. 1g! M (K)! Fig. 1f! r 150! 100! p100 p100 Mr (K)! (pSer707)! (pSer866/70)! c-Myc! p100! 75! 250! p100(pSer707)! 150! 75! 150! 50! 150! 100! 100! Fbxw7α! 100! 150! 75! 50! 75! 75! 100! 75!

75! 100! 100! 25! 75! 75! 20! 150! 50! 100! c-Myc β-catenin β-catenin! Skp1! 75! (pThr58/ (pSer33/37/ Ser62)! Thr41)! FLAG! 50!

Figure S8 continued!

Figure S8 continued

Fig.1h! Fig. 2a! Mr (K)! M (K)! r HA! 250! 150! 100! 75! 150! Fbw7α 100! 150! 100! 75! 75! β-Actin! 50! 50! 37! 37!

autoradiography!

Figure S8 continued!

Figure S8 continued

Fig. 2d!

Mr (K)! 150! Fig.2b! 150! 100! 100! M (K)! r 75! 75! 150! HA! HA! 100! 75! 50! 50! 37! HA! 37! 25! 75! 25! IκBα! IκBα! 50!

37! 50! 50! Tubulin! Tubulin! Tubulin!

Figure S8 continued!

Figure S8 continued

Fig.3b! Mr (K)! 150! 150! 100! 100! 75! 75! 50! 50! p100! p100!

250! 250! 150! 150! 100! 100! 75! 75! p100(pSer707)! p100(pSer707)!

250! 250! 150! 150! 100! 100! 75! 75!

p100(pSer866/870)! p100(pSer866/870)! 100! 100! 75! 75!

50! 50! RelB! RelB!

50! 50! 37! 37!

IκBα! IκBα!

Figure S8 continued

Fig. 3c! Fig. 3e!

Mr (K)! Mr (K)! 150! p100! p100! 150! p100! 150! 100! 100! 100! 75! 75! 75! p52! HA! 50! 75! 50!

RelB! 50! p52! 75! 150! 100! 50! Cyclin E! 50! 75! 37!

IκBα 50! 25! 37! 25! Skp1!

25! RelB! 20! 75! 25! Skp1! 20! IκBα 37!

25!

25! Skp1! 20!

Figure S8 continued!

Figure S8 continued

Fig.3f! Mr (K)! Fig.3g! 150!

100! Mr (K)! 75! 150! 100! 100! α Fbxw7 ! 75! 75! p100! 50! 250! HA! 150! 100! 150! 100! Fbxw7α! 75! FLAG!

Figure S8 continued

Fig. 5a!

Mr (K)! Mr (K)! p100! p100! 150! 150! 100! 100! Fig. 5b! 75! 75! Fbxw7α! Fbxw7α! 150! 150! Mr (K)! Mr (K)! 100! 100! 150! 150! 75! 75! 100! 100! c-Myc! c-Myc! 75! 75! HA! HA! 150! 150! 50! 50! 100! 100! IκBα IκBα 50! 75! 75! 37! 50! 37! 50! 25! 37! 37! 75! α-tubulin 75! α-tubulin HA! HA! 25! 25! 50! 50! 20! 20!

Skp1! Skp1! Skp1! 25! 25! Skp1!

20! 20!

Figure S8 continued!

Figure S8 continued

Mr (K)! 250! 150! Fig. 5h! 100! p100 (pSer707)!

150! 100! p100! 150! 100! Mr (K)! α 75! Fbxw7 ! 150! 75! 100! 75! Lamin A/C! Fig. 5g! 50! Fbxw7α! 37! 25! IκBα

20! 20! Skp1! 15! Skp1!

Figure S8 continued!

Figure S8 continued

Fig. 7a! M (K)! Fig. 7b! r Mr (K)! 150! 250! p100! 100! 150! 75! p100! 100! IκBα 75! 50! 37! 250! p100! 75! 150! 100! 50! c-Myc! 75! 25!

IκBα 20! Skp1! 37!

25! M (K)! Fig. 7e! r 100! c-Myc! 150! p100! 75! 100! 75! 50! 250! p100 (pSer707)! 150! Skp1! 25! 100!

20! IκBα 37!

25! Lamin A/C! 75!

50! 25! Skp1! 20!

Figure S8 continued!

Figure S8 continued

Fig.8a!

Mr (K)! 150! 100! 75!

p100!

37!

25! 20!

Skp1!

Figure S8 continued