Word Search

A Chemical and Genetic Approach to the Mode of Action of Fumagillin

A Chemical and Genetic Approach to the Mode of Action of Fumagillin
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Chemistry & Biology  13 , 1001–1009, September 2006  ª 2006 Elsevier Ltd All rights reserved DOI 10.1016/j.chembiol.2006.07.010 A Chemical and Genetic Approachto the Mode of Action of Fumagillin Yi Zhang, 1 Jing Ruey Yeh, 1 Andrew Mara, 1 Rong Ju, 1 John F. Hines, 1 Pasquale Cirone, 1 Hilary L. Griesbach, 4 Igor Schneider, 4 Diane C. Slusarski, 4 Scott A. Holley, 1 and Craig M. Crews 1,2,3, * 1 Department of Molecular, Cell, and DevelopmentalBiology 2 Department of Chemistry 3 Department of Pharmacology Yale UniversityNew Haven, Connecticut 06520 4 Department of Biological SciencesUniversity of IowaIowa City, Iowa 52242 SummaryPrevious mode ofaction studies identified methionineaminopeptidase 2 (MetAP-2) as the target of the anti-angiogenic natural product fumagillin and its drugcandidate analog, TNP-470. We report here that TNP-470-mediated MetAP-2 inhibition blocks noncanonicalWnt signaling, which plays a critical role in develop-ment, cell differentiation, and tumsrcenesis. Consis-tent with this finding, antisense MetAP-2 morpholinooligonucleotide injectioninzebrafishembryospheno-copies gastrulation defects seen in noncanonicalWnt5 loss-of-function zebrafish mutants. MetAP-2 in-hibition or depletion blocks signaling downstream ofthe Wnt receptor Frizzled, but upstream of Calmodu-lin-dependent Kinase II, RhoA, and c-Jun N-terminalKinase. Moreover, we demonstrate that TNP-470 doesnot block the canonical Wnt/  b -catenin pathway. Thus,TNP-470 selectively regulates noncanonical over ca-nonical Wnt signaling and provides a unique meanstoexploreanddissectthebiologicalsystemsmediatedby these pathways.Introduction Targetingthetumorvasculaturebyusingantiangiogenictherapeutics has been an attractive strategy to limit thesize and metastases of tumors. One of the most potentsmall-molecule antiangiogenic agents is the naturalproductfumagillin(  1  ),whichwasisolatedintheFolkmanlab from a contaminating  Aspergillus fumagatus  colonyin an endothelial cell culture. Fumagillin was subse-quently chemically modified to create the drug candi-date TNP-470(  2  ) [1], which was tested inPhase I/II trialsfor Kaposi’s sarcoma, renal cell carcinoma, brain can-cer, breast cancer, cervical cancer, and prostate cancer [2]. Using a chemical genetics approach, we previouslyidentified methionine aminopeptidase 2 (MetAP-2) as adirectbindingproteinoffumagillin [3],andwepublishedthe structure of human MetAP-2 bound to fumagillin [4].Both fumagillin and TNP-470 specifically bind MetAP-2and inhibit its activity [3, 5]. However, a role for MetAP-2inanintracellularsignalingpathwayhasnotbeeniden-tified. Here, we continue our mode of action studies of TNP-470 by combining model organism (zebrafish)small-molecule chemical genetics with nucleic acid-based genetic techniques (MO and siRNA) to identifyTNP-470 as a selective inhibitor of noncanonical Wntsignaling.How cells coordinate polarity and movement within alarger population is a fundamental question in develop-mental biology. Genetic studies of the  Drosophila  wingepithelia have identified several genes that function toestablish asymmetric cell polarities by coordinatingthe organization of the cytoskeleton [6]. These genescomprise the planar cell polarity pathway (PCP), whichincludes some genes known to be components (i.e.,Frizzled, Dishevelled) of the canonical Wnt signalingpathway that leads to  b -catenin-mediated transcription( Figure 1 ). Additional core PCP proteins constituting a‘‘noncanonical’’ Wnt signaling pathway include Daam1,RhoA, c-Jun N-terminal Kinase (JNK), and CaMKII. Inaddition to  Drosophila  wing epithelia polarization, non-canonicalWntsignalinghasbeenimplicatedinthecom-plex coordination of cell polarity, adhesion, and move-ment required for vertebrate gastrulation. A key setof morphogenetic movements during gastrulation isthemediolateralnarrowing andanterior-posterior tissuelengthening necessary to form the embryonic axis [7].Studies have demonstrated that these ‘‘convergenceand extension’’ (CE) movements are regulated by non-canonical signaling [8, 9]. For example, Wnt5 has beenimplicated in the morphogenetic movements during  Xenopus  and zebrafish gastrulation [10–12]. Unlike ca-nonical Wnt signaling, which is fairly well characterized,noncanonicalWntsignalingpathwaysarenotfullyeluci-dated. For example, it remains to be determined if mul-tiplenoncanonicalpathwaysexistsincevarioussubsetsof genes reported to be involved in noncanonical Wntsignaling have been linked to biological systems as di-verse as neurogenesis [13], cell adhesion [14], and cell polarization [15, 16]. Alternatively, a ‘‘core’’ noncanoni-cal Wnt pathway may exist, and these observed differ-ences may simply reflect the differences inherent inthe various systems studied. Unlike canonical Wnt sig-naling, for which both small-molecule agonists andantagonists have been reported [17, 18], the study of noncanonical Wnt signaling has been hampered by thelack of specific small-molecule probes of this pathway.To follow up on our recent finding that MetAP-2knockout murine embryos fail to undergo gastrulation[19], here we explore the role of MetAP-2 in early zebra-fish development. We show that MetAP-2 is necessaryfornoncanonical Wnt signaling in vivo and in cell culturesystems.Moreover,wereportthattheMetAP-2inhibitor TNP-470 selectively and potently inhibits noncanonicalWnt signaling downstream or at the level of Frizzled re-ceptors and upstream of JNK and CaMKII activation.These studies demonstrate the first small-molecule in-hibitor fortheinvestigation ofthe diverse biologicalsys-tems mediated by these important signaling pathways. *Correspondence: craig.crews@yale.edu  ResultsMetAP-2 Genetically Interacts with Wnt5in Zebrafish Usingthezebrafish(  Daniorerio  )developmentalprocessas a chemical genetic model to elucidate the mode of action of the small-molecule compound TNP-470, wetreated zebrafish embryos with the MetAP-2 inhibitor fumagillin (  1  ) and TNP-470 (  2  ) ( Figure 2 ). While these ex-periments resulted in a truncated tail phenotype (datanot shown), the penetrance was low, possibly due topoor permeability across the chorionic membrane. Weidentified and cloned zebrafish MetAP-2 (chromosome25), and it has ubiquitous expression in early develop-ment. Interestingly, at the time of vascularization, thisMetAP-2 transcript is enriched in the intersomitic re-gions ( Figure S1; see the Supplemental Data available with this article online). Thus, we next knocked down MetAP-2  expression in zebrafish embryos by antisensemorpholino oligonucleotides (MO) injection. The result-ingzebrafishembryosdisplayedahighlypenetratetrun-cated posterior phenotype similar to the AP axis tailextension defect observed in  Wnt5  MO-injected zebra-fish, although both of these low-dose MOs show milder gastrulation defects than the  Wnt5  mutant  pipetail   (   ppt   )( Figures 3 A–3D). This tail phenotype was also observedby using a second MO targeting a different splice sitenot observed in 5 base pair mismatch  MetAP-2  MO-in- jected embryos (data not shown). Downregulation of MetAP-2proteinwasconfirmedbyimmunoblotanalysis( Figure S2 ). Interestingly, embryos coinjected with acombination of low-dose  MetAP-2  and  Wnt5  MOs re-sulted in a strong synergistic effect rather than an addi-tive effect, suggesting a genetic interaction betweenthese two molecules ( Figure 3E; Table 1 ). Furthermore, theseMetAP-2MO-injectedzebrafishdisplayedbroader MyoD somite staining consistent with the observedMyoD staining in  ppt   fish [20] ( Figure S3 ). This gastrula- tiondefectissimilartobutnotassevereasthecompleteloss of gastrulation in the murine MetAP-2 knockoutmouse [19]. This discrepancy between the two modelsystems may be due to (1) an incomplete antisenseMO MetAP-2 knockdown in the zebrafish system, (2)the significant contribution of presynthesized maternalMetAP-2 protein in the fish egg, which could partiallyrescue loss of embryonic MetAP-2 mRNA, or (3) geneticredundancy as a result of the teleost genomic duplica-tion (we recently identified a related MetAP-2 gene onchromosome 4).To evaluate if MetAP-2 has a general role in nonca-nonical signaling or is only specific for Wnt5 signaling,zebrafish embryos were coinjected with MOs targetingMetAP-2 and the noncanonical Wnt11. Injection of theWnt11 MO alone resulted in 12% of the embryos dis-playing a moderate or severe eye defect consistentwith the phenotype of the Wnt11 mutant  silberblick  (   slb  ). However, coinjection with MetAP-2 MO morethan tripled (38%) the observed moderate/severe eyephenotype ( Table S1 ).It has been shown that Wnt5a activates noncanonicalWnt signaling through intracellular Ca 2+ release [21] andtheactivationofCa 2+ -sensingenzymessuchasCamKII.In addition, it has been demonstrated that injection of truncated, constitutively activated CamKII (CamKIItr)mRNA is sufficient to rescue the  Wnt5  homozygous mu-tant tail defect [22]. Thus, we wanted to determine if ex-ogenous CamKII activity was also able to suppress the MetAP-2  MO-induced phenotypes. As demonstratedin a genetic rescue experiment ( Figure 3F; Table 1 ), Figure 1. Schematic of Canonical and Non-canonical Wnt Signaling Although not all Wnt signaling effectors/ transducers are included, key componentsrelevant to this work are depicted.Figure 2. Structure of the Antiangiogenic Natural Product Fumagil-lin,  1 , the Clinical Trial Drug Candidate,  2 , and the NonspecificMetAP-1/MetAP-2 Inhibitory Natural Product Bengamide E,  3 Chemistry & Biology1002  coinjectionofCamKIItrmRNAwith MetAP-2 MOpartiallysuppressed the MO-induced phenotype (from a 37%frequency of shortened AP axis in  MetAP-2  MO only toa 17% frequency in  MetAP-2  MO plus CamKIItr). Thus,these genetic experiments clearly demonstrate a re-quirement for MetAP-2 in noncanonical Wnt5 signaling. MetAP-2 Is Required for Noncanonical,but Not Canonical, Wnt Signaling Since the mouse teratocarcinoma F9 system played animportant role in elucidating the noncanonical Wnt5a/ Rfz2 pathway in mammalian cells and faithfully recapit-ulates what is observed in zebrafish [23, 24], we choseto utilize this system to confirm the genetic interactionbetween MetAP-2 and Wnt5a observed inzebrafish em-bryos. In this assay, the activation of either canonical or noncanonical Wnt receptors (e.g., Fz1 and Fz2, respec-tively) leads to differentiation into primitive endoderm(PE), as measured by the induction of several markerssuchastissueplasminogenactivator(tPA)andthecyto-keratinEndo-A [23,24].Wethereforetestedtheabilityof TNP-470 (  2  ) to disrupt Wnt5a-mediated induction of PEintheF9cellsystem.AsshowninFigure4 A,10nMTNP-470 was sufficient to inhibit PE induction, as assessedby tPA activity levels in F9 cells expressing the nonca-nonical Wnt receptor Fz2 when cocultured with Wnt5a-expressing HEK293 cells. In contrast, PE induction byretinoic acid was unaffected. This inhibition was notduetothecytostaticactivityofTNP-470observedinen-dothelial cells since 10 nM TNP-470 had a negligibleeffect on F9 cell proliferation ( Figure S4 ). In addition,we found that directly knocking down 90% of MetAP-2expression with MetAP-2 gene (  MAP2  )-specific siRNA resulted in a dramatic decrease in PE induction to levelssimilar to those seen in TNP-470-treated cells ( Fig-ure 4B). In parallel experiments in which cytokeratinEndo-A antibody (TROMA-1) staining was used to mon-itorPE induction,TNP-470 potentlyinhibited Wnt5a/Fz2signaling( Figure4C).Similarly,siRNA-mediatedMetAP-2 downregulation also blocked Wnt5a-induced cytoker-atin Endo-A induction ( Figure 3C). These data indicatethat MetAP-2 activity is required for Wnt5a function. Figure 3. Genetic Interactions between MetAP-2  and  Wnt5  in Zebrafish(A) Wild-type zebrafish at 24 hr postfertiliza-tion denoting normal anterior-posterior length.(B) Homozygous  ppt   embryo.(C) Coinjection of   Wnt5  MO and a missense MetAP-2  MO as control.(D)  MetAP-2  MO injection phenocopies  Wnt5 MO-injected zebrafish, as shown in (C).(E) Coinjection of   MetAP-2  and  Wnt5  MOs inzebrafish embryos at minimal concentrationsresulted in a strong synergistic effect rather than an additive effect (also see Table 1 ).(F)CoinjectionofactivatedCamKIItrRNAwith MetAP-2  MO partially rescues the  MetAP-2 MO phenotype (also see Table 1 ).Table 1. Wnt5 and MetAP-2 Interact Genetically% Wild-Type% ModeratelyMorphant% HighlyMorphant% SeverelyTruncated n150  m M Wnt5 and 300  m M MetAP-2 MO 1.8 54.1 31.2 12.9 109150  m M Wnt5 MO 89.3 8.0 2.7 0 75300  m M MetAP-2 MO 80.6 12.2 6.1 1.1 98600  m M MetAP-2 MO plus control 27 35 37 0 137600  m M MetAP-2 MO plus CamKIItr 53 30 17 0 182 Wnt5  (150  m M) and  MetAP-2  (300  m M) antisense morpholinos act synergistically to elicit tail extension defects, and activated CamKIItr partiallyrescues the  MetAP-2  MO phenotype. (‘‘Moderately Morphant’’ describes a weaker   ppt   mutant phenotype, ‘‘Highly Morphant’’ embryos havea strong  ppt   phenotype, and ‘‘Severely Truncated’’ describes embryos that fail to extend past the yolk.)Fumagillin, TNP-470 Block Noncanonical Wnt Signaling1003  While generally believed to be a noncanonical Wntfamily member  [25], Wnt5a has also beenreported toin-duceaxisduplicationin  Xenopus embryoswhenoverex-pressed with the canonical Frizzled Fz5 [26]. Since bothcanonical and noncanonical Wnt signaling can inducePE formation in the F9 cell system [23, 24], we investi-gatedwhetherbothpathwaysrequireMetAP-2.Toavoidany ambiguity regarding crosstalk between Wnt5a anddifferent Frizzled receptors, we used isoproterenol-re-sponsivechimericFrizzledreceptors,whichhaveprovenuseful in the study of signaling events downstream of canonical and noncanonical Frizzled proteins [27, 28].These chimeric receptors consist of the extracellular and transmembrane domains of   b 2 -adrenergic receptor (  b 2 -AR) and the putative intracellular loops of the ratFrizzled-1/Rfz1 (canonical) or Frizzled-2/Rfz2 (nonca-nonical) receptors. Activation with the  b 2 -AR agonistisoproterenol ofeither chimericreceptorindividually ex-pressed in F9 cells leads to PE induction ( Figure 5 A).Confirming the observed TNP-470 inhibition of Wnt5a-induced PE induction ( Figures 4 A and 4C), we observedthatisoproterenol-inducedPEformationthrough b 2 -AR/ Rfz2 in F9 cells was attenuated by TNP-470, as assayedby TROMA-1 staining. Similarly, knockdown of endoge-nous MetAP-2 by siRNA blocked isoproterenol-trig-gered PE formation in F9 cells expressing  b 2 -AR/Rfz2( Figure 5 A). In contrast, PE induction through  b 2 -AR/ Rfz1 was not affectedby TNP-470or bysiRNA targeting MAP2 ( Figure5 A).Thus,ourresultsstronglysuggestthatMetAP-2isrequiredforsignalingthroughthenoncanon-ical Wnt receptor Frizzled 2, but not the canonical Wntreceptor Frizzled 1.It is well established that activation of the canonicalWntpathwayleadsto b -cateninstabilizationandnuclear translocation, which subsequently activates lymphoid-enhancer factor (LEF)/T cell factor (TCF)-regulated tran-scription [29, 30]. To confirm that MetAP-2 function isnot involved in the canonical Wnt pathway, we testedwhether fumagillin (  1  ), the potent parent compound of TNP-470, regulates LEF/TCF transactivation by usingthe LEF/  b -catenin-responsive TOPflash-luciferase re-porter [31].Asshown inFigure5C,100 m Misoproterenolinduced a 6-fold increase in TCF-luciferase after 6 hr in b 2 -AR/Rfz1 cells. Preincubation with 10 nM fumagillinhad no effect on TCF transactivation. Thus, once again,ourresultsindicatethatMetAP-2activitydoesnotaffectthe Wnt/  b -catenin pathway. MetAP-2 Acts Upstream of CamKII, RhoA,and JNK Activation To study further whether MetAP-2 is required for theinduction of intracellular calcium, as suggested by our zebrafish experiments, we investigated the effect of MetAP-2 inhibition on CamKII in F9 cells. Treatment of  b 2 -AR/Rfz2-expressing F9 cells with 100  m M isoprotere-nol for 10 min increased CamKII activity 4-fold ( Fig-ure 6 A). However, this increase in CamKII activity wasinhibited in cells pretreated with 10 nM TNP-470. Atten-uationofMetAP-2proteinlevelsviasiRNAalsoinhibitedCamKII activation upon treatment with isoproterenol Figure 4. Wnt5a/Fz2-Induced F9 Differentia-tion Was Attenuated by TNP-470 or siRNA-Mediated MetAP-2 Downregulation(A) F9 cells expressing wild-type rat Frizzled2 (Rfz2) were cocultured with HEK293 cellsexpressingWnt5ainthepresenceorabsenceof 10 nM TNP-470 for 4 days. The primitiveendoderm (PE) marker tPA was measuredbyELISA.Retinoicacid-inducedF9celldiffer-entiation was unaffected by TNP-470 (insert).Means + SD are shown (n = 6).(B) F9 cells with siRNA-mediated  MAP2 knockdown showed asimilarinhibitoryeffectas TNP-470. Means + SD are shown (n = 6).(C) Wnt5a-induced F9 cell differentiation wasmonitored by immunofluorescence stainingwith TROMA-1 antibody against PE-specificcytokeratin Endo-A. Identical fields aredepicted showing phase contrast (left),HEK293 cells coexpressing GFP with Wnt5a(center), and TROMA induction in Rfz2-ex-pressing F9 cells (right).Chemistry & Biology1004  ( Figure6 A).ThesedataareconsistentwiththeactivatedCamKIItr rescue of MetAP-2 downregulation in zebra-fish ( Figure 3F) and confirm the requirement of MetAP-2 activity in the Wnt/Ca 2+ signaling pathway.SinceJNKisreportedtobedownstreamofnoncanon-ical Wnt signaling [32, 33] and has been implicated incontrolling endothelial cell proliferation [34], we nexttested whether JNK activity can be regulated by TNP-470 by using phospho-63-c-  jun  antibody. As shown inFigure 6B, Wnt5a-stimulated JNK activity in Rfz2-ex-pressing F9 cells was inhibited when these cells werepretreated with TNP-470 for 16 hr. This result indicatesthat MetAP-2 acts upstream of c-Jun activation.The small GTPase RhoA is another important down-stream component of the noncanonical Wnt signalingpathway. In pull-down assays with the Rho binding do-main (RBD) from the effector protein Rhotekin, Wnt5a-stimulated RhoA activity was blocked by pretreatmentof mouse pulmonary endothelial (MPE) cells with TNP-470 for 16 hr ( Figure 6C). It is interesting to note thatboth c-  jun  and RhoA are critical factors involved inangiogenesis [34, 35]. Activation of Noncanonical Wnt Signaling RescuesTNP-470 Sensitivity in Endothelial Cells Given our result that MetAP-2 acts downstream of non-canonical Wnt signaling, we next explored whether TNP-470-induced cell cycle arrest in endothelial cellscould be rescued by activation of the downstream Wnteffector Dishevelled. We used  D DIX-Dvl2 that containedan amino-terminal deletion of the DIX domain of Dishev-elled-2,whichpreferentiallyactivatednoncanonicalWntsignaling in a Fz-independent manner  [8, 36]. Stableintroduction of a  D DIX-Dvl2 construct into MPE cellsreduces endothelial cell sensitivity to TNP-470 by w 10-fold ( Figure 6D). This was also observed in another endothelial cell type, HUVE cells stably expressing D DIX-Dvl2(datanotshown).Thesepooledretrovirus-in-fected HUVE cells do not represent an individual clone,therebyprecludingthepossibilitythatthedistinctsensi-tivities are due to clonal differences in responsiveness.These data indicate that overexpression of activatedDishevelled-2 partially rescues the loss of noncanonicalWnt signaling upon TNP-470 treatment. In addition,these results suggest that inhibition of noncanonicalWnt signaling by TNP-470 may be partially responsiblefor its cytostatic effect. Taken together, our findingsplace the need for MetAP-2 activity in noncanonicalWnt signaling proximal to Frizzled activation of Dishev-elled and upstream of calcium signaling, RhoA, andJNK activation [36]. Discussion The identification of selective small-molecule inhibitorscangreatlyaidtheinvestigationofintracellularsignalingpathways.Here,wecontinueourmodeofactionstudiesof TNP-470 by combining model organism chemicalgenetic and traditional genetics approaches to demon-strate that the target of TNP-470, MetAP-2, is essentialfor noncanonical Wnt signaling, which plays a criticalrole in development, cell differentiation, and tumsrcen-esis. Given the several shared components betweennoncanonical and canonical Wnt/  b -catenin signalingpathways, our finding that TNP-470 selectively inhibitsnoncanonical Wnt signaling over canonical Wnt sig-naling in a MetAP-2-dependent manner will aid the Figure 5. MetAP-2 Is Essential in Noncanon-ical Wnt Signaling but Has a Negligible Effecton the Canonical Wnt/  b -Catenin Pathway(A) F9 cell differentiation induced by 10  m Misoproterenol in cells expressing chimeric b 2 -AR/Rfz2 and  b 2 -AR/Rfz1 was monitoredby immunofluorescence staining withTROMA-1 antibody.(B) MetAP-2 protein levels in F9 cells ex-pressing  b 2 -AR/Rfz2 or   b 2 -AR/Rfz1 with MAP2 -targeting siRNA.(C) F9 cells expressing chimeric  b 2 -AR/Rfz1transfected with the pTOPflash reporter plasmid and treated with 10  m M isoprotere-nol for 6 hr. Relative LEF/TCF-luciferaseactivities of triplicate samples are presentedas fold induction. Means + SD are shown(n = 6).Fumagillin, TNP-470 Block Noncanonical Wnt Signaling1005
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks