LEF1 and B9L Shield b-Catenin from Inactivation by Axin,Desensitizing Colorectal Cancer Cells to TankyraseInhibitors
Marc de la Roche, Ashraf E.K. Ibrahim, Juliusz Mieszczanek, and Mariann Bienz
AbstractHyperactive b-catenin drives colorectal cancer, yet inhibiting its activity remains a formidable challenge.
Interest is mounting in tankyrase inhibitors (TNKSi), which destabilize b-catenin through stabilizing Axin. Here,we confirm that TNKSi inhibit Wnt-induced transcription, similarly to carnosate, which reduces the transcrip-tional activity of b-catenin by blocking its binding to BCL9, and attenuates intestinal tumors in ApcMin mice. Bycontrast, b-catenin's activity is unresponsive to TNKSi in colorectal cancer cells and in cells after prolonged Wntstimulation. This TNKSi insensitivity is conferred by b-catenin's association with LEF1 and BCL9-2/B9L, whichaccumulate during Wnt stimulation, thereby providing a feed-forward loop that converts transient into chronicb-catenin signaling. This limits the therapeutic value of TNKSi in colorectal carcinomas, most of which expresshigh LEF1 levels. Our study provides proof-of-concept that the successful inhibition of oncogenic b-catenin incolorectal cancer requires the targeting of its interaction with LEF1 and/or BCL9/B9L, as exemplified bycarnosate. Cancer Res; 74(5); 1495–505. �2014 AACR.
IntroductionWnt/b-catenin signaling plays pivotal roles in animal devel-
opment and tissue homeostasis, and in human cancer (1). Inthe absence of Wnts, b-catenin is continually earmarked forproteasomal degradation by the Axin complex: Axin providesscaffolding for glycogen synthase kinase 3 (GSK3) to phos-phorylate the N-terminus of b-catenin (after priming by caseinkinase 1a, CK1a), thus generating a phosphodegron recog-nized by the ubiquitin ligase adaptor b-TrCP (2). This processrelies on the adenomatous polyposis coli (APC) tumor sup-pressor, which promotes Axin complex assembly (3), releasesphosphorylated b-catenin (to be called PBC) from the complex(4), and/or promotes PBC recognition by b-TrCP and subse-quent ubiquitylation (5). Wnt stimulation blocks the activity ofthe Axin complex, thereby causing accumulation of unpho-sphorylated b-catenin (equivalent to activated b-catenin,ABC). ABC thus binds to the T-cell factor/lymphoid enhancer
factor (TCF/LEF) DNA-binding proteins to operate a tran-scriptional switch, recruiting various chromatin modifiers andremodelers to TCF/LEF target genes (6).
Awide range of cancers exhibit hyperactive b-catenin, eitherdue to oncogenic mutations in its N-terminal phosphodegron,or through mutational inactivation of its negative regulatorsAPC or Axin (1). Similarly, inactivation of Apc, or activation ofb-catenin, initiates tumorigenesis in themurine intestine (7, 8),in which the normal crypt stem cell compartment depends onWnt/b-catenin signaling (1). In mice, b-catenin is continuallyrequired for growth and progression of Apc-dependentadenomas and APC-mutant human xenografts (9), and theprogressive accumulation of nuclear b-catenin in colorectalcarcinomas also implies their continual reliance on oncogenicb-catenin through cancer progression (9, 10).
The case for b-catenin as a target for therapeutic interven-tion in colorectal cancer is thus overwhelming. However,developing b-catenin inhibitors has proved to be a consider-able challenge (11): b-Catenin is an intracellular protein whoseoncogenic activity is little affected by upstream Wnt signalingcomponents, and its inhibition therefore requires cell-perme-able agents. Furthermore, its activity depends primarily on itsbinding to TCF/LEF through the same extensive molecularinterface that also binds its negative regulators Axin and APC(4, 12, 13). Nevertheless, small-molecule antagonists have beenreported that target the b-catenin–TCF interaction, or regu-lators of b-catenin's activity or stability (Supplementary Fig.S1). There has been a recent boom of interest in a highlypromising group of compounds that inhibit tankyrase inhibi-tors (TNKSi), which destabilize b-catenin by blocking theturnover of Axin (14–18).
We recently identified a natural compound (carnosate) thatdestabilizes ABC in colorectal cancer cells, apparently by
Authors' Affiliation: MRC Laboratory of Molecular Biology, CambridgeBiomedical Campus, Francis Crick Avenue, Cambridge, United Kingdom
Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).
Current address for M. de la Roche: Department of Biochemistry,University of Cambridge, 80 Tennis Court Road, Cambridge CB2 1GA,United Kingdom.
Corresponding Author: Mariann Bienz, MRC Laboratory of MolecularBiology, Francis Crick Avenue, Cambridge CB2 0QH, United Kingdom.Phone: 44-1223-267-093; Fax: 44-1223-268-305; E-mail:[email protected]; andMarc de laRoche, Department of Biochem-istry,University ofCambridge, 80TennisCourtRoad,CambridgeCB21GA,United Kingdom. E-mail: [email protected]
promoting its aggregation through an intrinsically labilea-helix in its N-terminus, which prevents binding to its cofac-tor BCL9 (19). Carnosate is the only compound known to targetABC directly, and we thus set out to compare its efficacy withthat of indirect b-catenin inhibitors. Here, we show that mostof these elicit unspecific off-target effects, except for carnosateand TNKSi, which specifically reduce b-catenin–dependenttranscription in Wnt-stimulated cells. In APC-mutant colorec-tal cancer cells, carnosate proved the most effective b-catenininhibitor, and it also attenuated the levels and transcriptionaloutputs of ABC in the murine intestine, and intestinal tumor-igenesis in ApcMin mice. In contrast, although TNKSi stabilizeAxin and thus reduce ABC to low levels in colorectal cancercells, they fail to block its transcriptional activity. Notably, inAPC-wt cells, b-catenin also becomes TNKSi unresponsiveafter prestimulation with Wnt3a for 4 to 6 hours. This TNKSiinsensitivity is conferred by LEF1 and B9L (the nuclear paralogof BCL9, also called BCL9–2; refs. 20, 21). Both factors are Wntinducible, accumulating to high levels in cells with chronicWntpathway activity, which enables them to divert b-catenin fromthe Axin complex. Finally, most colorectal carcinomas expresshigh levels of LEF1, which could render them TNKSi insensi-tive. Our results highlight a key requirement for effectiveb-catenin inhibitors, namely their ability to block b-catenin'sassociation with LEF1 and B9L—a complex capable of limitingthe Axin-dependent inhibition of b-catenin in cells with chron-ic Wnt/b-catenin pathway activity.
Materials and MethodsPlasmids, antibodies, and chemicals
The following reagents were used: FLAG-b-catenin, TCF1(p45; provided by H. Clevers Hubrecht Institute, Utrecht, TheNetherlands); TOP-GFP/CMV-dsRFP (provided by C. Gottardi,Northwestern University Feinberg School of Medicine, Chi-cago, IL); FLAG-BCL9, FLAG-BCL9L366K, FLAG-BCL9DC,FLAG-BCL9DCL366K (22); pcDNA-Myc-TCF4 (23); pcDNA-HA-LEF1; SuperTOP (24); dimethyl sulphoxide, carnosate,XAV939, IWR-1, pyrvinium pamoate, iCRT3 (Sigma Pharma-ceuticals); PKF115-584 (Enzo Life Sciences); a-b-catenin (BDTransduction Laboratories); a-ABC, a-PBC (phospho-Ser33/37/Thr41), a-LEF1 (C18A7), a-TCF1, a-TCF4 (Cell SignalingTechnology);a-BCL9, a-B9L (R&D Systems); a-actin (Abcam);a-FLAG (Sigma-Aldrich). ICG-001 and 16kwere synthesized byMRC Technology.
Cell-based assaysCell lines were purchased from the European Collection of
Cell Cultures (HEK293T and HCT-116 in 2007; COLO320 in2011; SW480 and DLD1 in 2013). RKO cells were kindly pro-vided by Doug Winton (University of Cambridge, Cambridge,UK; in 2012). All cell lines were authenticated by short tandemrepeat (STR) DNA profiling. Upon receipt, cells were frozen,and individual aliquots were taken into culture, typically foranalysis within <10 passages. For SW480 and COLO320 cells,truncated APCproteinwasmonitored byWestern blot analysis(see Results). Cells were grown and transfected for Wntreporter assays and indirect immunofluorescence as described(22). Cytotoxicity assays were done as described (19). An
SW480 cell line with integrated TOP-GFP reporter (see sectionPlasmids, antibodies, and chemicals; ref. 25) was isolated bynegative selection and cloning of stable transfectants, and GFPwas monitored by fluorescence-activated cell sorting (FACS).Standard inhibitor treatment was for 24 hours (2.5 mmol/LXAV939, or 25 mmol/L carnosate), unless specified otherwise.
Quantitative real-time PCR and Western blot analysiscDNA was synthesized, and quantitative real-time PCR
(qRT-PCR) reactions were carried out with the ABI7900 Taq-man Thermocycler (Applied Biosystems), with primers andgene expression assays for human Wnt target genes (22), andthe following murine Wnt target genes (26): Tnfrsf12a,Mm00489103; Tbp1, Mm00446971; Bcl9l, Mm00518807; Axin2,Mm00443610; c-Myc, Mm00487804 (Applied Biosystems).Western blots were done as described (22).
Animal experimentsAnimal care and procedures were done according to the
standards set by the United Kingdom Home Office. Adminis-tration of single doses of carnosate (dissolved in British Phar-macopoeia compliant olive oil) to C57BL/6Jmice by gavage, andpreparation of lysates from isolated intestinal epithelia weredone essentially as previously described (10, 27). ApcMin/þ
control mice were fed AIN-76A, whereas treatment cohortswere fed AIN-76A pelleted with 0.1% carnosol or carnosate, orwith 1% carnosate from weaning, as described (28). Weightswere checked twice weekly, to monitor growth and food intake.Intestinal tumors were "blind" scored in methacarn-fixed smallintestines upon dissection, as described (10, 29). Proliferationand apoptosis was monitored by immunofluorescence usingantibodies against Ki67 and cleaved caspase-3 (Asp175 and 8D5,respectively; Cell Signaling Technology).
Tissue microarray analysisTissue microarrays (TMA) were processed for antibody
staining as described (10, 29), except that indirect immuno-fluorescencewas used. Scoring of protein expression levels wasdone blind (by A.E.K. Ibrahim, an experienced histopathologistspecializing in colorectal cancer), classifying LEF1 staininglevels of individual sections as negative (0), weak and patchy(1), moderate and widespread (2), or universally strongthroughout the core (3), whereby each tissue core was repre-sented by 2 to 3 nonadjacent cores.
ResultsFor a side-by-side comparison of previously reported b-cate-
nin antagonists, we conducted functional tests in Wnt3a-stimulated HEK293T cells treated with inhibitors, using aTCF-specific reporter (SuperTOP) as a readout of b-catenin–dependent transcription. The TNKSi XAV939 (15) and IWR-1(14) inhibited SuperTOP (IC50 0.3 and 0.1 mmol/L, respectively)more potently than carnosate (IC50 7 mmol/L; Fig. 1A). How-ever, the other compounds that reduced SuperTOP (e.g.,pyrvinium and ICG-001) also reduced the internal (cytomeg-alovirus-based) control reporter and elicited pronounced celltoxicity at their IC50, indicating significant off-target effects(Supplementary Fig. S2).
de la Roche et al.
Cancer Res; 74(5) March 1, 2014 Cancer Research1496
We further tested these agonists in SW480 colorectal cancercells, which express an APC truncation lacking its Axin-bindingsites (30), and thus accumulate high levels of ABC, as detectableby an antibody specific for this unphosphorylated form (31).Again, most inhibitors showed high cell toxicity and unspecificside effects (Supplementary Fig. S2). Of the nontoxic com-pounds, carnosate reduced SuperTOP to 40% of mock-treatedSW480 cells (19); however, TNKSi had very little effect (Fig. 1A),even in combination with carnosate (Supplementary Fig. S3).Notably, this was true for both XAV939 and IWR-1, whichrepresent different classes of TNKSi (binding to the nicotin-amide and adenosine pocket of TNKS, respectively; ref. 17),arguing that the inability of these inhibitors to reduce theb-catenin–dependent transcriptional in these cells is not limitedto a single TNKSi class. We also tested TNKSi on DLD1 cells(another APC-mutant colorectal cancer cell line commonlyused; ref. 15), which were only marginally more TNKSi respon-sive than SW480 cells (Supplementary Fig. S3). Our data areconsistent with previous reports that TNKSi are more potent inWnt-stimulated compared with APC-mutant cells (14–16).
TNKSi reduce the levels but not the activity of ABC inAPC-mutant colorectal cancer cellsWe confirmed that carnosate reduces ABC levels in SW480
cells (19; Fig. 1B), explaining why it attenuates SuperTOP (seeFig. 1A) and expression of endogenous AXIN2 (Fig. 1B), a well-established b-catenin target gene (32). TNKSi had an evenmore profound effect, reducing the levels of total b-catenin,
and of ABC, to <10% of mock-treated controls (Fig. 1B). Incontrast, the PBC levels remained high, and were even slightlyincreased (Supplementary Fig. S3), supporting the notion thatTNKSi deplete ABC by promoting its phosphorylation. BecausePBC is the substrate for b-TrCP recognition and subsequentdegradation (see Introduction), this explains why TNKSireduce total b-catenin through stabilizing Axin, as previouslyshown (15). It is well known that overexpressed Axin promotesb-catenin degradation in SW480 cells, despite their dysfunc-tional APC (e.g. 3, 33).
We also assessed the levels of b-catenin and its regulators inAPC-wt cells after inhibitor treatment—namely in Wnt-stim-ulated HEK293T cells (Fig. 1B), and in the colorectal cancer celllines HCT116 (in which ABC is high, due to a mutation in theCK1a phosphorylation site) and RKO (in which ABC is unde-tectable because its Wnt pathway is inactive; SupplementaryFig. S3). XAV939 increased the levels of AXIN1 and tankyrase inthese cells, but the levels of total b-catenin and ABC wereessentially unaffected.
ABC is destabilized by Axin degradasomes in TNKSi-treated SW480 cells
Immunofluorescence confirmed that overall b-cateninstaining was reduced in TNKSi-treated SW480 cells, consistentwith our Western blots (see Fig. 1B), though many cellsretained substantial levels of nuclear b-catenin (Fig. 2A andB), which could account for their sustained b-catenin–depen-dent transcription. In contrast, the nuclear b-catenin staining
A
Rel. a
ctivity (
%)
Concentration of added compound (μmol/L)
54321107.55.02.5
IWR-1XAV939
25
50
75
100
5040302010
CA125
B
α-ABC
α-Actin
α-AXIN1
α-TNKS
α-β-cat
α-AXIN2
α-BCL9
α-B9L
HEK293T cellsSW480 cells
M P M P
Figure 1. Responses to differentb-catenin antagonists. A,SuperTOP assays in Wnt3a-stimulated HEK293T (*) or SW480cells (*) treated with inhibitorconcentrations as indicated;values are shown as a percentageof mock-treated controls; errorbars, SEM (in all figures unlessotherwise specified). B, Westernblots of lysates from SW480 orHEK293T cells (as indicated)treated with inhibitors for 24 hours[25 mmol/L carnosate (CA) oriCRT3, 5mmol/L IWR-1 or ICG-001,2.5 mmol/L XAV939, 1 mmol/L 16k,25 nmol/L pyrvinium].
Acquiring Resistance to Tankyrase Inhibitors
www.aacrjournals.org Cancer Res; 74(5) March 1, 2014 1497
was reduced in carnosate-treated SW480 cells, which alsoshowed less AXIN2 staining (19; Fig. 2A), reflecting reducedAXIN2 expression. Thus, the nuclear pool of b-catenin seemsdepleted by carnosate but less so by TNKSi.
We noticed discrete cytoplasmic puncta of b-catenin inTNKSi-treated SW480 cells (Fig. 2B, arrows), which are neithervisible in carnosate-treated nor in control cells. These punctaalso contain Axin, and GSK3b, tankyrase (Fig. 2) and APC (seebelow). Given that they also containPBC (Fig. 2C), they are likelyto represent functionalAxin degradasomes (3) that promote thephosphorylation and subsequent degradation of b-catenin.TNKSi-induced Axin degradasomes do not contain other Axin-or APC-interacting proteins such as phosphorylated LRP6(signifying activated Wnt coreceptor; ref. 2), nor markers forendosomes or autophagosomes (Supplementary Fig. S4).
Axin degradasomes have been observed following Axinoverexpression (3, 33), but endogenous Axin degradasomesare neither detectable in untreated SW480 cells (Fig. 2) nor inAPC-wt cells (Supplementary Fig. S4), probably because theendogenous Axin levels are low inmammalian cells (34). TNKSithus enabled us for the first time to observe endogenous Axindegradasomes, likely because Axin is stabilized (AXIN1 3-5�,AXIN2 5-20�; Fig. 1B).
We also examined COLO320 cells, which express a rare APCtruncation without any b-catenin and Axin-binding sites (30).
These cells also exhibit Axin puncta, which are however negativefor APC, as expected. In contrast with SW480 cells, TNKSi-treated COLO320 cells did not show reduced ABC levels (unlikecarnosate-treated cells) but, instead, vastly increased PBC levels(Supplementary Fig. S5). This indicates that the Axin degrada-somes in these cells actively promoteb-catenin phosphorylation(consistent with their PBC reactivity; Supplementary Fig. S5); inother words, they are fully functional with regard to scaffoldingof GSK3. However, they seem unable to promote the ubiquityla-tion and/or proteasomal degradation of PBC, likely due to thecomplete lack of interaction between APC and b-catenin. Theythus seem to be stalled degradasomes.
ABC activity in SW480 cells remains refractory to TNKSieven during prolonged treatment
Our immunofluorescence indicated persistence of thenuclear b-catenin pool in TNKSi-treated SW480 cells throughthe 24-hour treatment. We thus extended the treatment to 5days (replenishing XAV939 daily), but found that the effects ofTNKSi plateaued within 2 days, with the levels of total b-cate-nin and ABC no longer reducing, and those of tankyrase andAxin no longer increasing (Supplementary Fig. S6). Indeed, allTNKSi-induced changes in the levels and subcellular distribu-tions of these proteins were observed after the first day oftreatment, and persisted thereafter.
BA XAV939CAMock
α-AXIN2 α-AXIN2 α-AXIN2
α-β-cat α-β-cat α-β-cat
α-β-cat
α-TNKS α-TNKS
α-β-cat α-GSK3β
α-AXIN2
α-AXIN1
MockXAV939
α-PBC
C
α-β-cat α-AXIN1
Figure 2. Axin degradasomes in TNKSi-treated colorectal cancer cells. A and B, confocal sections through inhibitor-treated SW480 cells, costained withantibodies as indicated; arrows, degradasomes containing Axin (green in merges) and b-catenin (red in merges), magnified in B; blue, 40,6-diamidino-2-phenylindole (DAPI). C, confocal sections through XAV939-treated SW480 cells, stained with antibodies as indicated. Size bars, 10 mm.
de la Roche et al.
Cancer Res; 74(5) March 1, 2014 Cancer Research1498
We also monitored the effects of TNKSi on b-catenin–dependent transcription over a 5-day treatment, using anintegrated TCF reporter based on destabilized enhanced greenfluorescent protein (25). SW480 cells remained unresponsive toXAV939 over 3 days, whereas carnosate reduced reporteractivity after the first day, and further still by the third dayof treatment. Likewise, carnosate reduced AXIN2 and B9Lexpression within 24 hours to approximately 20% and 45%,respectively (19), whereas TNKSi only modestly reduced theexpression of these target genes (to 75%–90%), even after 5days (Supplementary Fig. S6). Thus, b-catenin remains tran-scriptionally active in TNKSi-treated SW480 cells duringextended treatment, despite the TNKSi-induced depletion oftheir ABC.
Prolonged Wnt stimulation renders b-catenin activityunresponsive to TNKSiWe asked whether b-catenin activity would also become
refractory to TNKSi in APC-wt cells after prolonged Wntstimulation. We thus stimulated HEK293T cells with Wnt3afor various periods before TI treatment (Fig. 3A), and moni-tored their TCF-dependent transcription. As expected, Super-
TOP activity was much reduced if the cells were exposedsimultaneously to TNKSi and Wnt3a, but became increasinglyTNKSi insensitive with longer Wnt prestimulation, and wascompletely refractory 6 hours after Wnt3a stimulation (Fig.3B), accompanied by a slight progressive increase of ABC anddecrease of AXIN1 (Fig. 3C). The same was also seen in otherAPC-wt cell lines such as HeLa (Supplementary Fig. S7). Incontrast, HEK293T cells remained fully carnosate-responsive,even after 6 hours ofWnt prestimulation (Fig. 3B). Therefore, 4to 6 hours ofWnt stimulation of APC-wt cells suffices to rendertheir b-catenin activity refractory to TNKSi, mimicking thesituation in APC-mutant cells.
LEF1- and B9L-associated b-catenin is protected fromTNKSi-induced Axin degradasomes
b-Catenin equilibrates rapidly between nucleus and cyto-plasm (35, 36), and it is therefore unlikely that the observedTNKSi insensitivity of the transcriptionally active b-catenin inchronically Wnt-stimulated cells is due to its insulation fromthe cytoplasmic pool. Indeed, we estimate that the nuclearb-catenin in unstimulated HEK293T cells turns over with a t1/2of approximately 60 minutes (Supplementary Fig. S6). Wetherefore surmised that transcriptionally active b-catenin isshielded by a factor that limits its access toAxin degradasomes.Because 6 hours ofWnt stimulation suffices to render this poolrefractory to TNKSi, we further surmised that this factor wouldaccumulate during this period, and that it would bind tob-catenin in competition with Axin. This identifies BCL9/B9L and TCF/LEF as potential candidates.
Examining the expression levels of these candidates inWnt-stimulated HEK293T and HeLa cells, we found that only LEF1and B9L are Wnt inducible (Fig. 4A). Furthermore, amongstfour tested colorectal cancer cell lines with chronic Wntpathway activity, each expressed high levels of at least oneTCF/LEF and one BCL9/B9L family member, with SW480 cellsexpressing high levels of both LEF1 and B9L (Fig. 4A). Impor-tantly, coimmunoprecipitation revealed that TNKSi treatmentreduced the b-catenin associated with TCF and BCL9, but notwith LEF1 and B9L (Fig. 4B). Thus, the LEF1- and B9L-asso-ciated b-catenin is protected fromTNKSi-induced degradationin SW480 cells.
LEF1 and B9L confer TNKSi insensitivity on b-catenin incells with chronic Wnt pathway activity
To test the ability of our candidates to confer TNKSiinsensitivity on b-catenin, we overexpressed them atmoderatelevels in HEK293T cells (<5� over endogenous; Fig. 4C) beforestimulation with Wnt3a. Overexpression of B9L, TCF1, LEF,and b-catenin increased SuperTOP activity in unstimulatedand Wnt3a-treated cells (Supplementary Fig. S8), but thisactivity was strongly reduced if the cells were simultaneouslytreated with TNKSi (as shown in Fig. 3B), even in cells over-expressing BCL9, TCF4, or TCF1 (Fig. 4C). By contrast, approx-imately 30% of the SuperTOP activity was retained in cellsoverexpressing B9L, but not in cells expressing a B9L mutantunable to bind b-catenin (22). Strikingly, cells overexpressingLEF1 retained >95% of their SuperTOP activity despite simul-taneous exposure to TNKSi and Wnt3a (Fig. 4C). Thus, LEF1
α-AXIN1
α-ABC
α-TNKS
XAV939 − + + + + + +−− − 1 2 4 60
WCM
prestimulation 0
α-Actin
WCM prestimulation (h)6 4 2 1
100
75
50
25
0
CAXAV939
B
C
* *
*
*
*
*
*
*
Re
l. a
ctivity (
%)
A
0 h6 4 2 1
24 h of XAV939
WCM stimulation
Figure 3. ProlongedWnt stimulation renders cells unresponsive to TNKSi.A, Wnt pretreatment regime. B, SuperTOP assays of HEK293T cellstreated as outlined in A; shown are relative activities (%) of Wnt3a-conditioned medium (WCM)-treated controls (dotted line); �, P < 0.001.C, corresponding Western blots.
Acquiring Resistance to Tankyrase Inhibitors
www.aacrjournals.org Cancer Res; 74(5) March 1, 2014 1499
renders b-catenin completely unresponsive to TNKSi inHEK293T cells, even at moderate overexpression levels(approximately 4� above normal) comparable with endoge-nous LEF1 in SW480 cells (Fig. 4A).
As a further test, we asked whether reducing LEF1 levelswould restore TNKSi responsiveness in HEK293T cells presti-mulated with Wnt3a for 6 hours. We used two different LEF1sequences (an exon 1-specific siRNA and an exon 5-specificshort hairpin RNA), both of which depleted LEF1 2-3� (Sup-plementary Fig. S8) approximately to the levels of uninducedcells (Fig. 5A). As shown in Fig. 3B, prestimulation with Wnt3afor 6 hours rendered SuperTOP in mock-depleted HEK293Tcells refractory to TNKSi, in contrast with LEF1-depleted cells,which recovered a near-complete TNKSi response (Fig. 5Aand B). Interestingly, depletion of BCL9 and B9L also restoredTNKSi sensitivity under these conditions (Fig. 5B), whereasdepletion of TCF1 and of TCF4 did not.
Finally, we tested whether LEF1 depletion in SW480 cellswould render its ABC TNKSi-responsive. This was the case:
LEF1 depletion reduced b-catenin–dependent transcription inTNKSi-treated SW480 cells to 50% to 60% compared withmock-depleted cells (Fig. 5C). We conclude that their elevatedLEF1 is a crucial determinant ofb-catenin's TNKSi insensitivityin these APC-mutant colorectal cancer cells.
Carnosate reduces ABC in the normal murine intestineand the tumor numbers of ApcMin mice
Given the poor TNKSi response of cells with chronic Wntpathway activity, there was little incentive for testing TNKSi intissues with sustained Wnt signaling, e.g., normal intestinalcrypts or Apc-mutant intestinal tumors (see Introduction).Intrinsic in vivo toxicity of some TNKS inhibitors (see alsoSupplementary Fig. S2; ref. 16) further argues against their useon animals. However, we decided to test carnosate in theApcMin model, given its inhibitory activity in COLO320 cellswhose APC mutation resembles that of the ApcMin mutation(truncating all b-catenin and Axin-binding sites). Moreover, itwas previously shown that orally administered carnosate
Figure 4. Overexpressed LEF1 and B9L attenuate the TNKSi response of b-catenin. A, Western blots of lysates from various cells as indicated above panel[�24 hours Wnt3a-conditioned medium (WCM)], probed with antibodies as indicated (arrowheads point to long and short isoforms of TCF4). B, Westernblots of lysates from inhibitor-treated SW480 cells, probed for proteins coimmunoprecipitating with b-catenin; inputs on the left. C, SuperTOP assays ofHEK293T cells transfected for 24 hours to express proteins indicated below corresponding Western blot, treated as indicated below panel; relativeactivities were normalized (%) to Wnt3a-conditioned medium (WCM)-treated controls (dotted line).
de la Roche et al.
Cancer Res; 74(5) March 1, 2014 Cancer Research1500
spreads rapidly to various murine tissues including the brain,in which it exhibits bioactivity (27).We thus adopted this experimental regime (27), adminis-
tering a single dose of 1 mg carnosate to individual mice, andmonitored the transcript levels of different b-catenin targetgenes (22, 26) in cell lysates from intestinal epithelial prepara-tions (Supplementary Fig. S9). Indeed, the Bcl9l and Tnfrsf12atranscripts were reduced significantly throughout the moni-toring period; the Axin2 transcripts were initially reduced (upto 4 hours post carnosate), but recovered subsequently(Fig. 6A). We could not detect a statistically significant effecton c-Myc transcripts, possibly because the b-catenin–depen-dent modulation of c-Myc transcription is subtle (37). Asexpected, the ABC levels were also reduced upon carnosatetreatment, whereas the total b-catenin levels were not affected
We also tested whether carnosate attenuates intestinaltumorigenesis in ApcMin mice, which develop multiple intes-tinal neoplasms, driven by b-catenin activation followingsporadic Apc loss (7). Of note, a previous study showed thatcarnosol, a close chemical relative of carnosate, reduced thetumor burden of ApcMinmice if administered in their diet (28).We thus adopted the same experimental design, administering0.1% carnosol to ApcMin mice in their diet, or 0.1% or 1%carnosate (because carnosate is less toxic than carnosol;ref. 19). At 105 days, control mice showed 19 � 7.5 tumors,whereas all three treatment cohorts had significantly reducedtumor numbers (Fig. 6C). The tumor volume was also reducedapproximately 2� in the treatments groups comparedwith thecontrol (Supplementary Fig. S9). Thus, carnosate is as effectiveas carnosol in attenuating intestinal tumorigenesis in thismouse model.
LEF1 overexpression is prevalent in colorectalcarcinomas
A recent analysis of LEF1 expression in colorectal carcino-mas, based on immunohistochemistry, concluded that LEF1protein is detectable only in 26% of carcinomas (38). Accord-ingly, most carcinomas would therefore be potentially respon-sive to TNKSi. However, these results contrasted with thosefrom an earlier analysis, demonstrating that LEF1 transcriptsare highly abundant in colorectal cancer cell lines and carci-nomas (39). Indeed, evidence frommurine models and humancancers indicates a key role of LEF1 during cancer progression(6).
To resolve this controversy, and to examine the potentialtherapeutic value of TNKSi in colorectal cancer, we decidedto reexamine LEF1 expression in tissue specimens frompatients with cancer. We thus screened a TMA containingtissue cores from normal colonic mucosa, adenomatouspolyps, and colon carcinomas by immunofluorescence, andusing a different LEF1 antibody, the combination of whichimproved the sensitivity of endogenous LEF1 detectionconsiderably (Supplementary Fig. S10). Costaining this TMAfor LEF1 and b-catenin, we found that most epithelial cells ofthe normal mucosa were only weakly positive for bothproteins, except for a small number of cells near the bottomof crypts, which exhibit high levels of LEF1 and b-catenin,coinciding in each case (Fig. 7A)—likely marking LGR5-positive intestinal progenitor cells (40). However, virtuallyall cores from adenomas (n ¼ 21) show elevated levels ofboth proteins, and carcinomas (n ¼ 32) showed even higherlevels of LEF1 and nuclear b-catenin throughout, in eachcase strikingly coinciding at the cellular level (Fig. 7B and C).Semiquantitative analysis of the immunofluorescence signalintensity of each core indicates higher levels of LEF1 incarcinomas compared with adenomas (Fig. 7D), correlatingwith nuclear b-catenin (r ¼ 0.78; P < 0.0001; Pearson cor-relation test) whose levels also increase from adenoma tocarcinoma (9, 10). Our data are fully consistent with the RNAexpression data (39), showing that high levels of LEF1expression are prevalent in colorectal carcinomas.
C
125
25
50
75
100
Rel. a
ctivity (
%)
+ XAV939CTRL
siCTRLsiLEF1shCTRLshLEF1**
**
B
25
50
75
100
Rel. a
ctivity (
%)
60
siCTRLsiBCL9siB9LsiLEF1siβ-cat
**
WCM prestimulation
A125
25
50
75
100
Rel. a
ctivity (
%)
60
shScramshLEF1
WCM prestimulation
**
shScram shLEF1
0 6Preincubation (h)
0 6
α-LEF1
α-Actin
150
175
Figure 5. LEF1 depletion renders b-catenin sensitive to TNKSi. A and B,SuperTOP assays of HEK293T cells, treated as in Fig. 3A, followingtransfection for 24 hours with shLEF1 or shScram (A; control) for 24 hours(inset, correspondingWestern blot), or siRNAs (B) as indicated.C,SW480cells treated with XAV939, following LEF1 depletion as in A and B asindicated; values were normalized to mock-treated siCTRL. WCM,Wnt3a-conditioned medium.
Acquiring Resistance to Tankyrase Inhibitors
www.aacrjournals.org Cancer Res; 74(5) March 1, 2014 1501
DiscussionWhile comparing the potencies of recently identified b-cate-
nin inhibitors in cell-based assays, we encountered substantialcell toxicity and unspecific off-target effects at their IC50 for
most of them, whose inhibitory activity toward b-catenin istherefore unlikely to be specific. However, carnosate andTNKSi behaved as specific inhibitors of b-catenin in our hands,reducing its transcriptional activity in Wnt-stimulated cells.
Expre
ssio
n r
ela
tive
to T
BP
Bcl9-2
** **
20
10
5
15
Axin2
**
**
**
5
4
3
2
1
1 0.1 0.01
0 1 4 24 4 Time (h)
CA (mg) 1 0.1 0.01
0 1 4 24 4
Rel. β
-cat expre
ssio
n ** *
1 0.1 0.01
0 1 4 24 4
1 0.1 0.01
0 1 4 24 4 Time (h)
CA (mg)
125
100
75
50
25
α-ABCα-β-catenin40
30
20
10
Tum
or
co
unt
Mock 0.1% CO 0.1% CA
0.005P value 0.018 0.009
1% CA
c-Myc
0.4
0.3
0.2
0.1
Tnfrsf12a
** ***
20
10
5
C1 0.1 0.01
0 1 4 24 4
1 0.1 0.01
0 1 4 24 4
0.600.65
A
B
Figure 6. Carnosate (CA) reduces b-catenin levels and outputs in the normal and neoplastic mouse intestine. A, qRT-PCR of transcripts in lysates frommurineintestinal preparations at various times after carnosate administration, as indicated belowpanels; each symbol refers to one animal (fromcohorts of 4–5mice),from three independent experiments (see Supplementary Fig. S9; control values are from experiments II and III); statistical significance: �, P < 0.025;��, P < 0.0025. B, quantification of Western blots (Supplementary Fig. S9) by densitometry of intestinal lysates obtained as in A; symbols andstatistical significance as in A. C, numbers of intestinal tumors in 105-day old ApcMin mice fed with control or supplemented diet; P values (from t tests)are relative to controls.
β-cat
LEF1LEF1
β-cat
LEF1
β-cat
DA Normal Adenoma CarcinomaB C
1
3
2
LE
F1 e
xpre
ssio
n
Normal Adenoma Carcinoma0
****
*
Figure 7. Overexpression of LEF1 is prevalent and progressive in colorectal cancer. A–C, immunofluorescence of representative tissue cores from normalmucosa (A), adenomas (B), and carcinomas (C), costained for b-catenin (red), LEF1 (green), and DAPI (blue in merge, to mark nuclei) as indicated;magnifications of boxed areas are on the right. Arrows in A indicate putative crypt progenitor cells; size bars, 25 mm. D, boxplots of the TMA scoring results,indicating LEF1 expression levels (see Materials and Methods); statistical significance: �, P < 0.001; ��, P < 0.0001 (Wilcoxon rank sum tests).
de la Roche et al.
Cancer Res; 74(5) March 1, 2014 Cancer Research1502
They also destabilized ABC in APC-mutant colorectal cancercells, but despite this, the activity ofb-catenin in these cells wasbarely responsive to TNKSi. Importantly, b-catenin activityalso became refractory to TNKSi in APC-wt cells followingWntstimulation for 4 to 6 hours. We presented evidence that thisTNKSi insensitivity in cells with chronic Wnt pathway activityis conferred predominantly by high levels of LEF1 and, to alesser degree, of B9L—both products of Wnt target genesthat accumulate in these cells. Our data imply that LEF1 andB9L cooperate to lock a transient burst of b-catenin–depen-dent signaling into a stable state of chronic Wnt/b-cateninpathway activity.
High LEF1 and B9L levels divert b-catenin from TNKSi-induced Axin degradasomesOur experimental evidence, based on preexpressing or
depleting TCF/LEF and B9L/BCL9 factors, indicates that theTNKSi insensitivity of b-catenin in cells with chronic Wntpathway activity is determined primarily by high levels of LEF1and, to a lesser degree, of B9L. Both factors are unique amongsttheir family members in that they are Wnt inducible and thustend to accumulate in cells with chronicWnt pathway activity.They also show a marked tendency to be overexpressed incolorectal carcinomas (20; Fig. 7) and in colorectal cancer celllines (Fig. 4A) albeit to varying degrees. It is possible that theobserved TNKSi insensitivity of b-catenin in these cell lines isdue to the cumulative expression levels of all their LEF/TCFand BCL9/B9L family members (some of which can also beoverexpressed in carcinomas, e.g., BCL9; ref. 41).How do LEF/TCF and BCL9/B9L factors protect the activity
of ABC despite its continued conversion to PBC by the Axincomplex and its consequent degradation? It seems likely thatthis is due to direct competitive binding: TCF/LEF exhibit a 20–50 times higher affinity for b-catenin than (unphosphorylated)Axin andAPC (42), and thus have a competitive advantage overthe latter in binding to b-catenin, as previously shown (43).This advantage could be increased considerably in the ternarycomplex with BCL9/B9L. TCF4 and BCL9 can bind simulta-neously to b-catenin, together occupying a surface on b-cate-nin (44) larger than that occupied by Axin or APC (4, 13), and sothe combined affinity of b-catenin for LEF and B9L is likely toexceed its affinity to the Axin complex by at least two orders ofmagnitude. We thus propose that high levels of LEF1 and B9Lbind to and divert a significant fraction of the de novo syn-thesized b-catenin to the nucleus, before its access to Axin,thereby creating a continuous pipeline that fuels the pool oftranscriptionally active b-catenin.LEF1 and B9L, despite being predominantly nuclear at
steady state (21, 45), are likely to shuttle rapidly in and outof the nucleus, like b-catenin itself (35, 36) and, on overexpres-sion, shift cytoplasmic b-catenin into the nucleus (21, 45).Furthermore, these factors are partially cytoplasmic in APC-mutant cancer cell lines and in colorectal carcinomas (Fig. 7). Itis therefore plausible that they can access de novo synthesizedb-catenin in the cytoplasm, in competition with Axin. Of note,the nuclear export function of APC is disabled by most APCtruncations found in colorectal cancer cell lines and carcino-mas (46) and, therefore, the APC-mediated conveyance of
nuclear b-catenin to the cytoplasmic Axin complex is atten-uated in these APC-mutant cells.
Is LEF1 unique amongst TCF in conferring TI insensitivityonb-catenin? Although this has not been assessed side-by-side,it seems that LEF1 has a slightly higher binding affinity tob-catenin than TCF4 (42, 47). Furthermore, LEF1 exhibits thesame key (aspartic acid to glutamic acid) substitution as TCF3that allows formation of a hairpin in its b-catenin–bindingdomain, thereby increasing its interface with b-catenin (12).Indeed, TCF3was found to shieldb-catenin fromAxin by directcompetition for binding in early Xenopus embryos (48). How-ever, neither TCF3 nor other TCFs are likely candidates forconferring TNKSi insensitivity on b-catenin in cells withchronic Wnt pathway activity because (i) none of them accu-mulate in response to Wnt stimulation in APC-wt cells (Fig.4A), (ii) neither TCF1 nor TCF4 protects b-catenin from Axin-dependent degradation in cells with chronic Wnt pathwayactivity (Fig. 4B and C), and (iii) TCF3 is not expressed in any ofthe APC-mutant cells we tested (Fig. 4A).
TNKSi-induced destabilization of b-catenin occursdownstream of its nuclear conveyance by LEF1 and B9L
Carnosate and TNKSi both destabilize ABC in colorectalcancer cells—TNKSi considerably more so than carnosate—but they achieve this by distinct mechanisms. TNKSi increaseAxin degradasome activity, thus depleting ABCby converting itto PBC, the substrate for b-TrCP recognition and proteasomaldegradation. In contrast, carnosate promotes selectively theproteasomal degradation of ABC (19), without affecting thelevels of PBC or total b-catenin (Fig. 1B), suggesting that thisroute of ABC destabilization does not involve Axin. Impor-tantly, only the carnosate- but not the TNKSi-induced desta-bilization of ABC proved effective in reducing its transcrip-tional activity.
The likely reason for this is that carnosate blocks the bindingof BCL9/B9L to ABC, apparently by altering the conformationof a structurally labile N-terminal a-helix of b-catenin (abut-ting its BCL9-binding site); thisa-helix constitutes an "Achilles'heel," which renders b-catenin aggregation-prone when disor-dered (19). Therefore, carnosate acts upstream of, or in parallelto, the nuclear conveyance of ABC by LEF1 and B9L. Incontrast, the TNKSi-induced destabilization of ABC seems tooccur downstream of this conveyance, following b-catenin'snuclear exit.
Carnosate attenuates b-catenin activity and intestinaltumorigenesis in mice
We have shown inhibitory effects of carnosate on transcrip-tional outputs of b-catenin in the normal intestine, and onintestinal tumorigenesis, confirming its bio-activity in murinetissues (27). The tumor-attenuating effects of carnosate in theApcMin model are relatively modest, but they are equivalent tothose of its chemical relative, carnosol (28). The latter havebeen attributed to reduced phosphorylation of tyrosine 142within b-catenin's "Achilles' heel" (ref. 28; see also ref. 19),broadly consistent with our own evidence that carnosate actsthrough this structurally labile a-helix of b-catenin to interferewith its binding to BCL9 (19). Targeting this interaction thus
Acquiring Resistance to Tankyrase Inhibitors
www.aacrjournals.org Cancer Res; 74(5) March 1, 2014 1503
seems a promising strategy for developing inhibitors of b-cate-nin–driven intestinal neoplasia.
Limited application of TNKSi in b-catenin–dependentneoplasia
Our study confirms that TNKSi are highly effective inblocking b-catenin–dependent transcription in transientlyWnt-stimulated cells. However, the latter become refractoryto TNKSi after 4 to 6 hours of prestimulation with Wnt, onceLEF1 has accumulated sufficiently. Intriguingly, a similar lagperiod of approximately 4 hours following Wnt stimulationwas observed before theAxin complex plateaued and stabilizedat its inhibited state (49). It thus seems that the activity of theAxin complex is only susceptible to perturbations, such asTNKSi-induced Axin levels, during this initial period of ree-quilibration after Wnt stimulation.
Our data imply that TNKSi are only effective in blockingb-catenin activity in tissues that experience transient bursts ofWnt signaling, and/or express low levels of LEF1 and B9L. Theysuggest that the combined levels of LEF1 and B9L overexpres-sion in normal and cancerous tissues determine the TNKSiresponsiveness of their b-catenin. However, most colorectalcarcinomas express high levels of both proteins (20, 41; Fig. 7),which are expected to render their oncogenic b-catenin unre-sponsive toTNKSi. Therefore, the therapeutic value of TNKSi incolorectal cancer is somewhat limited, and crucially dependson identifying those carcinomas with low levels of theseprotective factors (e.g., those resembling DLD1 cells, whichexhibit a partial response to TNKSi).
Implications for targeting oncogenic b-cateninOur study provides a proof-of-concept that oncogenic
b-catenin can be targeted directly by a small inhibitory mol-ecule not just in cell assays (19), but also in an animal model.
Importantly, we have shown that merely destabilizing onco-genic b-catenin is not sufficient for inhibiting its activity. Ourstudy highlights the importance of targeting the transcription-ally active b-catenin directly or its interface with LEF1, and/orwith BCL9/B9L, as successfully achieved recently (50). Theseinsights should guide the future development or application ofsmall-molecule inhibitors of oncogenic b-catenin.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: M. de la Roche, M. BienzDevelopment of methodology: M. de la Roche, J. MieszczanekAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M. de la Roche, A.E.K. Ibrahim, J. MieszczanekAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): M. de la Roche, A.E.K. Ibrahim, M. BienzWriting, review, and/or revisionof themanuscript:M. de la Roche,M. BienzAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): M. de la Roche, A.E.K. Ibrahim,J. MieszczanekStudy supervision: M. de la Roche, M. Bienz
AcknowledgmentsThe authors thank Cara Gottardi and Hans Clevers for plasmids, Andrew
Merritt and his staff at MRC Technology for chemical synthesis, and TraceyButcher and her staff for help with the animal experiments.
Grant SupportThis work was supported by the Medical Research Council (U105192713) and
Cancer Research UK (grant C7379/A8709 to M. Bienz; Clinical FellowshipC10112/A11388 to A.E.K. Ibrahim). The Human Research Tissue Bank is sup-ported by the NIHR Cambridge Biomedical Research Centre.
The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received September 19, 2013; revised November 20, 2013; accepted December11, 2013; published OnlineFirst January 13, 2014.
References1. Clevers H, Nusse R. Wnt/b-catenin signaling and disease. Cell 2012;
149:1192–205.2. MacDonald BT, Tamai K, He X. Wnt/b-catenin signaling: components,
mechanisms, and diseases. Dev Cell 2009;17:9–26.3. Mendoza-Topaz C, Mieszczanek J, Bienz M. The adenomatous poly-
posis coli tumour suppressor is essential for Axin complex assemblyand function and opposes Axin's interaction with dishevelled. OpenBiol 2011;1:110013.
4. Ha NC, Tonozuka T, Stamos JL, Choi HJ, Weis WI. Mechanism ofphosphorylation-dependent binding of APC to b-catenin and its role inb-catenin degradation. Mol Cell 2004;15:511–21.
5. Su Y, Fu C, Ishikawa S, Stella A, Kojima M, Shitoh K, et al. APC isessential for targeting phosphorylated b-catenin to the SCFb-TrCPubiquitin ligase. Mol Cell 2008;32:652–61.
6. Arce L, Yokoyama NN, Waterman ML. Diversity of LEF/TCF action indevelopment and disease. Oncogene 2006;25:7492–504.
7. Su LK, Kinzler KW, Vogelstein B, Preisinger AC, Moser AR, Luongo C,et al. Multiple intestinal neoplasia caused by a mutation in the murinehomolog of the APC gene. Science 1992;256:668–70.
8. Harada N, Tamai Y, Ishikawa T, Sauer B, Takaku K, Oshima M, et al.Intestinal polyposis in mice with a dominant stable mutation of theb-catenin gene. EMBO J 1999;18:5931–42.
9. Scholer-Dahirel A, Schlabach MR, Loo A, Bagdasarian L, Meyer R,Guo R, et al. Maintenance of adenomatous polyposis coli (APC)-
mutant colorectal cancer is dependent on Wnt/b-catenin signaling.Proc Natl Acad Sci U S A 2011;108:17135–40.
10. Metcalfe C, Ibrahim AE, GraebM, de la RocheM, Schwarz-Romond T,Fiedler M, et al. Dvl2 promotes intestinal length and neoplasia in theApcMin mouse model for colorectal cancer. Cancer Res 2010;70:6629–38.
11. Polakis P. Drugging Wnt signalling in cancer. EMBO J 2012;31:2737–46.
12. GrahamTA,WeaverC,MaoF,KimelmanD, XuW.Crystal structure of ab-catenin/Tcf complex. Cell 2000;103:885–96.
13. Xing Y, Clements WK, Kimelman D, Xu W. Crystal structure of ab-catenin/axin complex suggests a mechanism for the b-catenindestruction complex. Genes Dev 2003;17:2753–64.
14. ChenB,DodgeME, TangW, Lu J,MaZ, FanCW, et al. Small molecule-mediateddisruption ofWnt-dependent signaling in tissue regenerationand cancer. Nat Chem Biol 2009;5:100–7.
15. Huang SM, Mishina YM, Liu S, Cheung A, Stegmeier F, Michaud GA,et al. Tankyrase inhibition stabilizes axin and antagonizes Wnt signal-ling. Nature 2009;461:614–20.
16. Lau T, Chan E, Callow M, Waaler J, Boggs J, Blake RA, et al. A noveltankyrase small-molecule inhibitor suppresses APC mutation-drivencolorectal tumor growth. Cancer Res 2013;73:3132–44.
17. Lehtio L, Chi NW, Krauss S. Tankyrases as drug targets. FEBS J2013;280:3576–93.
de la Roche et al.
Cancer Res; 74(5) March 1, 2014 Cancer Research1504
18. Shultz MD, Cheung AK, Kirby CA, Firestone B, Fan J, Chen CH, et al.Identification of NVP-TNKS656: the use of structure-efficiency rela-tionships to generate a highly potent, selective, and orally activetankyrase inhibitor. J Med Chem 2013;56:6495–511.
19. de laRocheM,RutherfordTJ,GuptaD,VeprintsevDB,SaxtyB, FreundSM, et al. An intrinsically labilea-helix abutting theBCL9-bindingsite ofb-catenin is required for its inhibition by carnosic acid. Nat Commun2012;3:680.
20. Adachi S, Jigami T, Yasui T, Nakano T, Ohwada S, Omori Y, et al. Roleof a BCL9-related b-catenin-binding protein, B9L, in tumorigenesisinduced by aberrant activation of Wnt signaling. Cancer Res 2004;64:8496–501.
21. Brembeck FH, Schwarz-Romond T, Bakkers J, Wilhelm S, Ham-merschmidt M, Birchmeier W. Essential role of BCL9–2 in the switchbetween b-catenin's adhesive and transcriptional functions. GenesDev 2004;18:2225–30.
22. de la Roche M, Worm J, Bienz M. The function of BCL9 in Wnt/b-catenin signaling and colorectal cancer cells. BMC Cancer 2008;8:199.
24. Veeman MT, Slusarski DC, Kaykas A, Louie SH, Moon RT. Zebrafishprickle, a modulator of noncanonical Wnt/Fz signaling, regulatesgastrulation movements. Curr Biol 2003;13:680–5.
25. Maher MT, Flozak AS, Stocker AM, Chenn A, Gottardi CJ. Activity ofthe b-catenin phosphodestruction complex at cell-cell contacts isenhanced by cadherin-based adhesion. J Cell Biol 2009;186:219–28.
26. Sansom OJ, Meniel VS, Muncan V, Phesse TJ, Wilkins JA, Reed KR,et al. Myc deletion rescues Apc deficiency in the small intestine. Nature2007;446:676–9.
27. SatohT,KosakaK, ItohK,Kobayashi A, YamamotoM,ShimojoY, et al.Carnosic acid, a catechol-type electrophilic compound, protects neu-rons both in vitro and in vivo through activation of the Keap1/Nrf2pathway via S-alkylation of targeted cysteines onKeap1. JNeurochem2008;104:1116–31.
28. Moran AE, Carothers AM, Weyant MJ, Redston M, Bertagnolli MM.Carnosol inhibits b-catenin tyrosine phosphorylation and preventsadenoma formation in the C57BL/6J/Min/þ(Min/þ) mouse. CancerRes 2005;65:1097–104.
29. Sansom OJ, Berger J, Bishop SM, Hendrich B, Bird A, Clarke AR.Deficiency of Mbd2 suppresses intestinal tumorigenesis. Nat Genet2003;34:145–7.
30. Rowan AJ, Lamlum H, Ilyas M, Wheeler J, Straub J, Papadopoulou A,et al. APC mutations in sporadic colorectal tumors: a mutational"hotspot" and interdependence of the "two hits". Proc Natl Acad SciU S A 2000;97:3352–7.
31. Staal FJ, Noort Mv M, Strous GJ, Clevers HC. Wnt signals aretransmitted through N-terminally dephosphorylated b-catenin. EMBORep 2002;3:63–8.
32. Lustig B, Jerchow B, Sachs M, Weiler S, Pietsch T, Karsten U, et al.Negative feedback loop of Wnt signaling through upregulation ofconductin/axin2 in colorectal and liver tumors. Mol Cell Biol 2002;22:1184–93.
33. FauxMC, Coates JL, Catimel B, Cody S, Clayton AH, Layton MJ, et al.Recruitment of adenomatous polyposis coli and b-catenin to axin-puncta. Oncogene 2008;27:5808–20.
35. Townsley FM,CliffeA, BienzM.Pygopus andLegless target Armadillo/b-catenin to the nucleus to enable its transcriptional co-activatorfunction. Nat Cell Biol 2004;6:626–33.
36. Krieghoff E, Behrens J, Mayr B. Nucleo-cytoplasmic distribution ofb-catenin is regulated by retention. J Cell Sci 2006;119:1453–63.
37. van de Wetering M, Sancho E, Verweij C, de LauW, Oving I, HurlstoneA, et al. The b-catenin/TCF-4 complex imposes a crypt progenitorphenotype on colorectal cancer cells. Cell 2002;111:241–50.
38. Kriegl L, Horst D, Reiche JA, Engel J, Kirchner T, Jung A. LEF-1 andTCF4 expression correlate inversely with survival in colorectal cancer.J Transl Med 2010;8:123.
39. Hovanes K, Li TW, Munguia JE, Truong T, Milovanovic T, LawrenceMarsh J, et al. b-catenin-sensitive isoforms of lymphoid enhancerfactor-1 are selectively expressed in colon cancer. Nat Genet 2001;28:53–7.
40. Fan XS, Wu HY, Yu HP, Zhou Q, Zhang YF, Huang Q. Expression ofLgr5 in human colorectal carcinogenesis and its potential correlationwith b-catenin. Int J Colorectal Dis 2010;25:583–90.
41. ManiM, Carrasco DE, Zhang Y, Takada K, Gatt ME, Dutta-Simmons J,et al. BCL9 promotes tumor progression by conferring enhancedproliferative, metastatic, and angiogenic properties to cancer cells.Cancer Res 2009;69:7577–86.
42. Choi HJ, Huber AH, Weis WI. Thermodynamics of b-catenin-ligandinteractions: the roles of the N- and C-terminal tails in modulatingbinding affinity. J Biol Chem 2006;281:1027–38.
43. vonKries JP,WinbeckG, AsbrandC,Schwarz-RomondT,SochnikovaN, Dell'Oro A, et al. Hot spots in b-catenin for interactions with LEF-1,conductin and APC. Nat Struct Biol 2000;7:800–7.
44. Sampietro J, Dahlberg CL, Cho US, Hinds TR, Kimelman D, Xu W.Crystal structure of a b-catenin/BCL9/Tcf4 complex. Mol Cell 2006;24:293–300.
45. Behrens J, von Kries JP, Kuhl M, Bruhn L, Wedlich D, Grosschedl R,et al. Functional interaction of b-catenin with the transcription factorLEF-1. Nature 1996;382:638–42.
46. Rosin-Arbesfeld R, Cliffe A, Brabletz T, Bienz M. Nuclear export of theAPC tumour suppressor controls b-catenin function in transcription.EMBO J 2003;22:1101–13.
47. Knapp S, Zamai M, Volpi D, Nardese V, Avanzi N, Breton J, et al.Thermodynamics of the high-affinity interaction of TCF4 with b-cate-nin. J Mol Biol 2001;306:1179–89.
48. Lee E, Salic A, Kirschner MW. Physiological regulation of b-cateninstability by Tcf3 and CK1e. J Cell Biol 2001;154:983–93.
49. Hernandez AR, Klein AM, Kirschner MW. Kinetic responses ofb-catenin specify the sites of Wnt control. Science 2012;338:1337–40.
50. TakadaK, ZhuD,BirdGH, SukhdeoK, Zhao JJ,ManiM, et al. Targeteddisruption of the BCL9/b-catenin complex inhibits oncogenic Wntsignaling. Sci Transl Med 2012;4:148ra17.
Acquiring Resistance to Tankyrase Inhibitors
www.aacrjournals.org Cancer Res; 74(5) March 1, 2014 1505
2014;74:1495-1505. Published OnlineFirst January 13, 2014.Cancer Res Marc de la Roche, Ashraf E.K. Ibrahim, Juliusz Mieszczanek, et al. Desensitizing Colorectal Cancer Cells to Tankyrase Inhibitors
-Catenin from Inactivation by Axin,βLEF1 and B9L Shield
Updated version
10.1158/0008-5472.CAN-13-2682doi:
Access the most recent version of this article at:
