v468.Chapter 8 采用 b-Catenin/TCF基本转录通讯结构对b-Catenin/TCF转录进行分析

Assaying b-Catenin/TCF Transcription with b-Catenin/TCFTranscription-Based Reporter Constructs

Methods in Molecular Biology v468. chapter 8

Abstract

Transcription-based reporters have been instrumental in characterizing the Wnt/b-catenin signalingpathway and will be essential in the search for therapeutics aimed at combating diseases linked to aberrantsignaling. In this chapter, we introduce a new improved Wnt/b-catenin reporter system, b-catenin-activatedreporter (BAR), and its accompanying control reporter system, found unresponsive BAR (fuBAR). Itsenhanced sensitivity, increased dynamic range, and lentiviral platform provide a reporter system that willkeep pace with the needs of scientists in the field.

Key words: Wnt , b-Catenin , Luciferase , Transcription , Reporter , BAR , TCF , LEF .




1. Introduction

The Wnt/b-catenin pathway is the best-studied Wnt pathway inpart due to robust tools for measuring pathway activation bothin vivo and in vitro. Among the earliest and still commonly usedassays of Wnt/b-catenin signaling include phenotypic assays inDrosophila (1) , dorsal axis duplication in Xenopus (2) , and proliferationof C57MG mammary epithelial cells (3) . These assayswere crucial for a substantial amount of the early characterizationof the pathway. Much of the more recent characterizationof the pathway has relied on the convenience of transcriptionbasedreporter systems. The first transcription-based luciferasereporter of Wnt/b-catenin signaling, TOPFlash, was designedby Korinek et al. (4) . The TOPFlash reporter contains threeTCF response elements (CCTTTGATC) upstream of a basal c-fos promoter while the control reporter, FOPFlash, containsthree mutant TCF response elements (CCTTTG GC C). TOPFlashwas later modified by Upstate Biotechnology to containthree TCF response elements upstream of a minimal thymidinekinase (TK) promoter. The TOPFlash reporter system provided areliable assay of pathway activation and was crucial for the identificationand characterization of several pathway components.

As the characterization of the Wnt/β-catenin pathway turnedto identifying modifiers of the core pathway components, theneed for a more sensitive reporter developed. This niche wasfilled by Ajamete Kaykas in the Moon lab with the constructionof the SuperTOPFlash reporter. SuperTOPFlash contains eightTCF response elements upstream of Clontech’s minimal TA promoter(5) . This modification greatly enhances the sensitivity anddynamic range of the reporter (Fig. 8.1a ) providing a better toolfor characterizing modifiers of the pathway as well as the abilityto identify new components in Drosophila genome-wide RNAinterference (RNAi) screens (6) . This was significant as Super-TOPFlash was the first Wnt/β-catenin reporter responsive towingless in Drosophila cells.

The necessity to monitor Wnt/β-catenin signaling in nontransfectablecells and achieve even greater sensitivity for high throughput screening inspired the construction of the β-cateninactivated reporter (BAR) system. The BAR system contains aconcatemer of 12 TCF response elements separated by uniquefive-nucleotide linkers specifically designed to minimize recombinationthat can lead to loss of TCF binding sites. This series of TCFresponse elements is inserted upstream of Promega’s minP minimalpromoter, completing a functional promoter that drives the transcriptionof either Firefly luciferase (pBARL), renilla luciferase(pBARRen), or β-globin intron-linked Venus (pBARV) (Venusis a variant of EYFP (7) ). These reporters were inserted betweenthe long terminal repeats (LTRs) of a lentiviral-transducing plasmid. The result is a highly sensitive luciferase reporter with anunmatched dynamic range and Venus reporter that allows a spatialreport of pathway activation (Fig. 8.1 ).

Fig. 8.1. The BAR system has enhanced sensitivity and dynamic range when directly compared with TOPFlash andSuperTOPFlash. A 10 ng of TOPFlash, SuperTOPFlash, or pGL3BARL were transfected along with 10 ng of pRLTK inHEK293T cells seeded in a 48-well plate. HEK293T cells stably expressing pBARLS were generated as described inSection 3.4 . Cells were treated with specified doses of Wnt3a-conditioned media ( CM ) for 18 h. Luciferase activity wasmeasured as described in Section 3.5.2 and data are presented as fold activation over control conditioned media-treatedcells. B A monoclonal HEK293T cell line stably expressing pBARVS was generated as described in Section 3.4 . Cells weretreated with either control conditioned media or Wnt3a-conditioned media for 30 h.Control reporters, found unresponsive BAR (fuBAR), wereconstructed using the same strategy. They are identical to theirrespective parent reporter with the exception that each TCFDNA binding element contains a two-base substitution conferringa non-functional element (pfuBARL and pfuBARV).The essentially identical nature of the control reporters provides themost optimal experimental control, as well as allowing for identicallentiviral titer production when generated side by side with theresponsive reporter.

A second version of the reporter constructs containing aPGK promoter driving a puromycin- or hygromycin-resistancegene was constructed for antibiotic selection in mammalian cells(pBARLS, pfuBARLS, pBARLHyg, pfuBARLHyg, pBARVS,pfuBARVS, pBARVHyg, and pfuBARLHyg). A third versioncontaining a PGK or EF1a promoter driving dsRed (pBARVRand pfuBARVR) was constructed for visual detection of cellscontaining the reporter independent of reporter activation(Fig. 8.2 ).

It should be noted that although the Wnt/β-catenin reportersappear to be very specific readouts of signaling, they are infact artificial promoters that may not faithfully reflect the activityof endogenous TCF/LEF response elements (8) . Thereforeit is important to complement reporter data with a measureof the transcription profile of known Wnt/β-catenin targetgenes (see http://www.stanford.edu/~rnusse/pathways/targets.html and Note 1 for target genes). It is also importantto note that TCF/LEF-independent β-catenin-mediated transcriptionalactivation will not be detected with these reportersystems (9) .

In the following sections, we outline protocols for usingBAR transiently, generating stable BAR cell lines, and measuring BARluciferase activity.

Fig. 8.2. Multiple platforms of the BAR system make it a versatile reporter system.




2. Materials



2.1. Transient Transfection of Reporter for Complementary DNA (cDNA) Overexpression or Small Interfering RNA (siRNA) Knockdown

1. 48-well cell culture plate.

2. HEK293T or other transfectable cells.

3. Lipofectamine 2000 (Invitrogen , Carlsbad, CA; cat. #11668-027) or transfection reagent of choice.

4. Optimem (Invitrogen; cat. #31985-088).

5. Plasmids (see Notes 2 and 3 ): pGL3BARL, pGL3fuBARL, pRLTK(Promega, Madison, WI; cat. #E2241), cDNA of interest.

6. siRNA, shRNA, and carrier plasmid (backbone of cDNAexpression plasmid or empty vector).

7. L-cell control and Wnt3a-conditioned media or purifiedWnt3a (ref. (10) ; see Chapter 2; ATCC, Manassas, VA;CRL-2648 and CRL-2647) or (R&D Systems, Minneapolis,MN; cat. #1324-WN-002).



2.2. Lentivirus Production

1. HEK293T cells.

2. Media: DMEM supplemented with 10% (v/v) fetal bovineserum (FBS) and 1% (v/v) penicillin/streptomycin.

3. 2× HEPES-buffered saline (HBS) pH 7.1 (0.22-μm filtered).

To prepare 40 mL of 2× HBS, pH 7.1, add 5.6 mLof 2 M NaCl, 4 mL of 0.5 M HEPES, pH 7, 60 μL of 1 MNa 2 HPO 4 , and 30.4 mL of dH 2 O.

4. 2.5 M CaCl 2 (0.22-μm filtered).

5. Sterile water (0.22-μm filtered).

6. Plasmids (see Note 3 )—pSL3, pSL4, pSL5, pSL9/rLuc,BAR, and fuBAR in lentiviral platform.



2.3. Lentivirus Concentration

1. 150 mL Millipore Stericup-GP PES filters (Millipore, Billerica,MA; cat. #SCGP U01 RE).

2. Beckman ultracentrifuge tubes (Beckman Coulter, Fullerton,CA; cat. #344058).

3. Beckman SW-28 swinging bucket rotor.

4. Pasteur pipettes.

5. 1× Tris-buffered saline (TBS): 50 mM Tris-HCl, pH 7.5, and 150mM NaCl. To prepare, dissolve 6.05 g Tris and 8.76 g NaCl in800 mL of ddH 2 O. Adjust pH to 7.5 with 1 M HCl and make volumeup to 1 L with ddH 2 O. TBS is stable at 4°C for 3 months.



2.4. Generating Stable Reporter Cell Lines

2.4.1. Stable Luciferase Reporter Cell Line

1. pBARLS and pfuBARLS or pBARLHyg and pfuBARLHygvirus.

2. pSL9/rLuc virus.

3. Puromycin or hygromycin.

4. 6-well and 48-well cell culture plates.

5. L-cell control and Wnt3a-conditioned media or purifiedWnt3a (ref. (10) ; see Chapter 2 ; ATCC, CRL-2648 andCRL-2647) or (R&D Systems; cat. #1324-WN-002).


2.4.2. Stable Venus Reporter Cell Line

1. pBARVS and pfuBARVS.

2. Puromycin.

3. 6-well and 100-mm cell culture plates.

4. L-cell control and Wnt3a-conditioned media or purifiedWnt3a (ref. (10) ; see Chapter 2; ATCC, CRL-2648 andCRL-2647) or (R&D Systems; cat. #1324-WN-002).



2.5. Luciferase Assay


2.5.1. Low-Throughput Assay

1. 1× Passive lysis buffer.

2. Firefly luciferase reagent.

3. Stop & Glo ? reagent (Renilla luciferase substrate and Fireflyluciferase antagonist; Promega).

4. 96-well plate with white wells.


2.5.2. High-Throughput Assay

1. Dual-Glo TM Firefly luciferase reagent (Promega).

2. Dual-Glo TM Stop & Glo ? reagent (Renilla luciferase reagentand Firefly luciferase antagonist).




3. Methods



3.1. Transient Transfection of Reporter for cDNA Overexpression or siRNA Knockdown


Prior to the BAR system, the majority of Wnt/b-catenin luciferasereporter assays were performed by transiently transfecting cellswith the Firefly luciferase reporter, a Renilla luciferase normalizationplasmid, and cDNAs/siRNA/shRNA to be analyzed.Although the transient reporter assay has a decreased dynamicrange compared with the stably integrated reporter, it is still veryrobust and alleviates the production of lentivirus. The followingmethod is based on a 48-well plate format and can be modifiedfor other plate formats by scaling based on the surface area of thewell. Each experimental condition is performed in triplicate. Specifictransfection details for the transfection reagent used shouldbe followed according to manufacturer’s specifications (Lipofectamine2000 protocol: http://www.invitrogen.com/content/sfs/manuals/lipofectamine2000_man.pdf ).


3.1.1. Transiently Transfecting Reporter with cDNA Expression Plasmids

1. Day 1: Plate cells at a density such that they will be 80% confluentthe following day.

2. Day 2: Transfect cells with 10 ng pGL3BARL or 10 ngpGL3fuBARL, 10 ng pRLTK, your construct(s) of interest,and the appropriate amount of carrier plasmid using themanufacturer’s protocol.

3. Day 3: If the cells will not be treated with a source of Wnt3aor other modulators, then proceed to Section 3.5 to readluciferase activity. Otherwise, treat the cells with Wnt3a orother modulators (see Note 4 ).

4. Day 4: Proceed to Section 3.5 for measuring luciferase activity.


3.1.2. Transiently Transfecting Reporter with shRNA or siRNA

1. Day 1: Plate cells at a density such that they will be 40% confluentthe following day.

2. Day 2: Transfect cells with siRNA or shRNA using themanufacturer’s protocol for the transfection reagent used(Lipofectamine 2000 protocol: http//www.invitrogen.com/content/sfs/manuals/lipofectamine2000_man.pdf ).

3. Day 3: Transfect cells with 10 ng pGL3BARL or 10 ngpGL3fuBARL, 10 ng pRLTK, and the appropriate amountof carrier plasmid using the manufacturer’s protocol.

4. Day 5: Treat cells with Wnt or other modulator if necessary.

5. Day 6: Proceed to section Section 3.5 for measuring luciferaseactivity.



3.2. Production of Lentivirus- Containing BAR and Stable Cell Line Production


The transduction plasmid backbone used for the lentiviral-compatibleBAR constructs and the lentiviral helper plasmids were providedby the Naldini lab, Vita-Salute San Raffaele University, Milan,Italy. This lentivirus is replication incompetent and does notcarry oncogenic cDNAs, which makes it Biosafety Level 2. Thevirus is, however, competent for human infection, requiring theuse of personal protective guidelines, including double gloving,the use of barrier tips, and collection of all liquids in a non-aspiratingsystem for inactivation with 10% bleach. The following protocol willyield high-titer virus that can be used to generate stable reportercell lines, assay Wnt/b-catenin signaling in cells that are difficultto transfect such as primary cultures, or assay signaling in vivo.

BAR and fuBAR virus is made at the same time to ensure equaltiter. The pSL9/rLuc plasmid can be used to generate lentiviruscontaining a constitutive EF1a promoter driving Renilla luciferasefor reporter assay normalization.

1. Day 1: Seed a 100-mm dish with HEK 293T cells such thatthey will be 70–80% confluent the next day. If very hightiter virus is needed, scale up production to several 150-mmdishes and adjust the transfection suggested guidelines basedon dish surface area.

2. Day 2: Prepare DNA cocktails for transfection as in Table 8.1.

Add 500 μL (1,250 μL for 150-mm dish) of 2× HBS dropwiseto the above cocktail and bubble with 10 strokes ofyour pipette. Add drop-wise to you cells, gently mix, andreturn to incubator.

3. Day 3: Remove media and dispose of media following properprocedures for inactivation in 10% bleach. Replace with freshmedia.

4. Day 4: Collect media and centrifuge for 5 min at 3,000× g toremove cellular debris. This media may now be used to infectcells or can be concentrated to achieve higher viral titer.

Table 8.1

DNA cocktails for transfection

 

3.3. Lentivirus Concentration


There are two methods for concentrating virus. Concentratingvirus with 30-kDa molecular weight cut-off centrifugation filters(Millipore Amicon Ultra cat. #UFC903024) is a simple approachto yield a 50× concentration. A limitation to this approachis concurrent concentration of other components in the mediaincluding serum and this may have deleterious affects onthe cell line to be infected. A second approach involves pelleting thevirus by ultracentrifugation. This technique is slightly more laborintensive but allows you to completely exchange the media andyield a 500× concentration.

1. Aliquot 30–35 mL of viral containing media into Beckmanultracentrifuge tubes, match tubes by weight (use freshmedia to balance the tubes), and spin at 50,000× g for 2 h at4°C in the SW28 swing bucket rotor.

2. Carefully decant the supernatant and invert the tube on apaper towel for 5 min (will have ~50 μL supernatant plusvirus left in the tube).

3. Add 50 μL (or desired volume) of 1× TBS or 1× PBS to eachtube, seal with paraffin, and leave at 4°C overnight with noshaking.

4. Pipette up and down three to five times and combine theresuspended virus from each tube. Filter pooled virus with a0.45-μm filter.

5. Aliquot, snap-freeze in liquid nitrogen, and store at –80°C.Virus should only be freeze–thawed once, dictating the sizeof the aliquots.



3.4. Generating Stable Reporter Cell Lines


For assays that do not require stable cell lines or the DsRed tracer,the reporters without a selectable marker are recommended, asthey will produce higher titer virus. In this section, we covermethods for generating stable luciferase reporter cell lines as wellas stable Venus reporter cell lines. The volume of virus used willvary depending on the cell line and viral titer.


3.4.1. Stable Luciferase Reporter Cell Line

The following method describes the production of a polyclonalreporter line. We want to stress that this is a general protocoland variable factors such as a cell line’s responsiveness to Wnt andthe sensitivity of your luminometer will determine the amountof virus needed to generate a perfect reporter line. Infecting thereporter line with pSL9/rLuc virus provides constitutive expressionof Renilla luciferase providing normalization for siRNAexperiments or assays that do not involve transfection. To date,we have generated over 40 stable reporter lines in vastly differentcell types. Although rare, we have found cell line exceptions inwhich the reporter is not responsive to pathway activation.

1. Day 1: Seed a 6-well plate such that the cells will be 50%confluent the following day.

2. Add three different doses of reporter virus and matchingdoses of control reporter virus to the 6-wells. We typicallystart with 200 μL, 50 μL, and 10 μL of virus that has beenconcentrated 50×.

3. Day 2: Replace the media with fresh media. As always, inactivateviral containing media in 10% bleach.

4. Day 3: Transfer cells from each well to a 100-mm dish containingthe appropriate concentration of puromycin or hygromycinfor selection.

5. Allow several days for selection and repopulation of the cells.

6. Test each reporter line by seeding each line in several wells ofa 48-well cell culture plate. Treat each line with several dosesof L-cell control or Wnt3a-conditioned media and measureluciferase activity the following day (see Section 3.5 ).

7. Choose the best reporter line and corresponding controlreporter line based on the dynamic range, expand the cells,and freeze back several vials, as reporter activity has beenfound in some cases to diminish over several passages.

8. For constitutive Renilla luciferase expression, seed thereporter cells in a 6-well plate such that they will be 50%confluent the following day and treat the cells with differentdoses of pSL9/rLuc virus.

9. Repeat steps 6 and 7.


3.4.2. Stable Venus Reporter Cell Line

A stable polyclonal Venus reporter line can be generated using anidentical approach as the stable luciferase reporter line. The onlydifference is that reporter activity is measured by fluorescenceusing a microscope or plate reader. The following protocol detailsthe use of fluorescence-activated cell sorting (FACS) to refine theheterogeneity of the line. Briefly, a stable pBARVS virus infectedcell line is generated. A monoclonal or polyclonal line with thehighest possible dynamic range is generated with two rounds ofFACS. In the first round, cells are stimulated with an EC50 doseof Wnt3a-conditioned and a population of high Venus-expressingcells are collected. The population is cultured for several dayswithout Wnt3a-conditioned medium and then resorted for cellsthat are not expressing Venus. This protocol yields a reporterline with very low basal activity and robust response to pathwayactivation (Fig. 8.1b ).

1. Perform steps 1–5 from Section 3.4.1 .

2. Test each reporter line by seeding each individual line in severalwells of a 48-well cell culture plate. Treat each line withseveral doses of L cell control or Wnt3a-conditioned mediaand visualize or measure Venus fluorescence the followingday. Choose the best cell line based on dynamic range anddetermine the EC 50 dose of the Wnt3a-conditioned media.

3. Seed a 100-mm culture dish with reporter cells such thatthey will be 70% confluent the next day.

4. The following day, treat the cells with the EC 50 dose of conditionedmedia.

5. 18–24 h following treatment, sort the cells by FACS with anarrow gate of the highest Venus-expressing cells.

6. Replate the sorted cells in standard growth media for atleast 4 days to allow Venus expression to return to baseline.

Before proceeding to step 7, you can restimulate a fractionof the cells with the EC 50 dose of Wnt3a-conditioned mediato check the integrity of the FACS.

7. Repeat the FACS and collect a narrow window of the lowestexpressing cells. The entire sorted population can be collectedas a single population or plated individually in a 96-well plate to create a monoclonal line (Fig. 8.1b ).

8. Expand the line(s) and freeze back several vials of cells.



3.5. Luciferase Assay


The sensitivity and robustness of the BAR reporter allows formeasuring luciferase activity in a broad range of luminometersand plate formats. BAR activity has been measured in luminometersranging from single-tube luminometers to high-throughputplate readers. A greater than 1,000-fold dynamic range wasachieved in a 384-well plate format and it is foreseeable that thiscan be achieved in a 1,536-well format as well. The luciferaseassay reagent to be used depends on throughput of the assay. Thestandard low-throughput reagent is Promega’s Dual-Luciferase ^®reporter assay system (cat. #E1910). For high-throughputassays, Promega’s Dual-Glo ^TM (cat. #E2940) is recommended.The robustness of the BAR reporter allows you to use a fractionof Promega’s suggested volume of reagent. The following lowthroughputmethod has been optimized for use on a BertholdMitras LB940 luminometer and the high-throughput methodhas been optimized on the Perkin Elmer Envision plate reader.


3.5.1. Low-Throughput Assay

In the following method, the cells were plated and treated in a48-well plate and the luciferase activity was measured in a 96-wellplate.

1. Aspirate cell culture media from each well.

2. Add 50 μL of 1× passive lysis buffer and moderately rotatefor 20 min at room temperature.

3. Transfer 5 μL of each sample in duplicate to a 96-well whitewell plate.

4. Program your luminometer with the following settings:

    (a) Inject 10 μL of Firefly luciferase reagent.

    (b) Read total luminescence.

    (c) Inject 10 μL of Stop & Glo ^® reagent.

    (d) Read total luminescence.

5. Express data as a ratio of Firefly relative light units to Renillarelative light units.


3.5.2. High-Throughput Assay

In the following method, the cells were plated, treated, and theluciferase activity measured in a 384-well plate. The volume ofculture media in each well prior to measuring luciferase activityis 40 μL.

1. Add 10 μL of Dual-Glo TM Firefly luciferase reagent using aliquid dispenser and incubate for 10 min at room temperature(if using clear bottom 384-well plates, bare nuclei willcan be visualized if lysis is complete) (see Note 5 ).

2. Read total luminescence.

3. Add 10 μL of Dual-Glo TM Stop & Glo ? luciferase reagentusing a liquid dispenser and incubate for 10 min at roomtemperature (see Note 5 ).

4. Read total luminescence.

5. Express data as a ratio of Firefly relative light units to Renillarelative light units.




4. Notes



1. Two common Wnt target genes are axin2 and lef1. RealtimePCR primers for analyzing the human transcripts are asfollows:

Axin2 forward: CTCCCCACCTTGAATGAAGA.

Axin2 reverse: TGGCTGGTGCAAAGACATAG.

Lef1 forward: GACGAGATGATCCCCTTCAA.

Lef1 reverse: AGGGCTCCTGAGAGGTTTGT.

2. The reporters in the lentiviral backbones cannot be used fortransient reporter assays as the episomal form contains a constitutivepromoter upstream of the TCF response elements thatdrives transcription independent of Wnt/b-catenin signaling.

3. All reporter plasmids and lentiviral helper plasmids can beobtained from the Moon lab by contacting either author.

4. The dose of Wnt3a-conditioned media and incubation time willvary based on the potency of the conditioned media. We havetreated cells with Wnt3a-conditioned media for as little as 4h and measured reporter activity above baseline. Commonincubation times are 12–24 h.

5. The 10 μL of Dual-Glo TM luciferase and Stop & Glo ? reagent usedin Section 3.5.2 may be reduced even further. The only concernis incomplete cell lysis, which may be overcome by supplementingthe Firefly reagent with Promega’s passive lysis buffer.




Acknowledgments

We thank the Howard Hughes Medical Institute for funding.




References


1. Nusslein-Volhard, C., and Wieschaus, E. (1980). Mutations affecting segment number and polarity in Drosophila. Nature 287 , 795–801.

2. McMahon, A.P. and Moon, R.T. (1989). Ectopic expression of the proto-oncogene int-1 in Xenopus embryos leads to duplicationof the embryonic axis. Cell 58 , 1075–1084.

3. Jue, S.F., Bradley, R.S., Rudnicki, J.A., Varmus, H.E., and Brown, A.M. (1992).The mouse Wnt-1 gene can act via a paracrine mechanism in transformation of mammary epithelial cells. Mol Cell Biol 12 , 321–328.

4. Korinek, V., Barker, N., Morin, P.J., van Wichen, D., de Weger, R., Kinzler, K.W., Vogelstein, B., and Clevers, H. (1997). Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC –/– colon carcinoma. Science 275 , 1784–1787.

5. Veeman, M.T., Slusarski, D.C., Kaykas, A.,Louie, S.H., and Moon, R.T. (2003). Zebrafish prickle, a modulator of noncanonicalWnt/Fz signaling, regulates gastrulation movements. Curr Biol 13 , 680–685.

6. DasGupta, R., Kaykas, A., Moon, R.T., and Perrimon, N. (2005). Functional genomic analysis of the Wnt-wingless signaling pathway. Science 308 , 826–833.

7. Rekas, A., Alattia, J.R., Nagai, T., Miyawaki,A., and Ikura, M. (2002). Crystal structure of Venus, a yellow fluorescent protein with improved maturation and reduced environmentalsensitivity. J Biol Chem 277 , 50573–50578.

8. Barolo, S., and Posakony, J.W. (2002). Three habits of highly effective signaling pathways: principles of transcriptional controlby developmental cell signaling. Genes Dev 16 , 1167–1181.

9. Olson, L.E., Tollkuhn, J., Scafoglio, C., Krones, A., Zhang, J., Ohgi, K.A., Wu, W., Taketo, M.M., Kemler, R., Grosschedl, R., et al. (2006). Homeodomain-mediated beta-catenin-dependent switching events dictate cell-lineage determination. Cell 125 ,593–605.

10. Willert, K., Brown, J. D., Danenberg, E.,Duncan, A. W., Weissman, I. L., Reya, T., et al. (2003) Wnt proteins are lipid-modifiedand can act as stem cell growth factors. Nature , 423 , 448–452.