Mouse Wnt3a coding sequences were PCR amplified and cloned into a pCS2+ expression vector using BamHI and XhoI restriction sites. All point mutations were introduced into full-length Wnt3a using PCR. All DNA fragments derived from PCR were automatically sequenced to confirm the presence of the desired mutation and to rule out spurious mutations. The NTD (1 to 276) and CTD (277 to 352) of mouse Wnt3a were cloned into pCS2+ using the same restriction sites as above. To generate a C-terminal His-tagged version of mouse NTD and CTD, a histidine tag was introduced by PCR amplification and cloned into the pCS2 vector. To produce the Fz8-CRD-Fc fusion protein, the CRD (i.e., the 135-amino-acid region extending from the first to the tenth conserved CRD cysteine) was inserted between EcoRI and BglII in the pFUSE-hIgG3*01-FC2 vector (InvivoGen, San Diego, CA, USA), which contains an IL-2 signal sequence and a Cytomegalovirus immediate-early gene 1 enhancer and promoter, followed upstream by the hinge region of a human IgG heavy chain.
Cell culture, DNA transfection and cell lysis
HEK293T cells and mouse L cells were routinely maintained in Dulbecco’s modified Eagle’s medium (Invitrogen, Waltham, MA, USA) containing 10% fetal bovine serum (FBS) (PAA, Freiburg, Germany) and 1% penicillin/streptomycin in 5% CO2 at 37°C. Cells were split one day prior to transient transfection and grown to 60 to 70% confluence. HEK293T cells were transfected using jetPRIME™ (Peqlab, Erlangen, Germany) reagent (Polyplus Transfection, USA) according to the manufacturer’s protocol. To detect Wnt3a proteins in cell lysates, HEK293T cells were transfected with wild-type Wnt3a or its mutant isoforms. The cells were harvested after 48 hr of transfection, washed with ice-cold 1× PBS and disrupted in lysis buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1% NP-40, 10 mM ethylenediaminetetraacetic acid (EDTA) protease inhibitor cocktail complete (Roche Diagnostics GmbH, Mannheim, Germany). The resulting supernatant was transferred to a fresh tube. Proteins were analyzed by Western blot using anti-Wnt3a antibody. For loading control, the membranes were probed with anti-α-tubulin.
Rabbit polyclonal antibodies against mouse Wnt3a were obtained from Abcam (Camebridge, UK) . Penta-His antibodies were obtained from Qiagen (Hilden, Germany). Mouse monoclonal antibodies against α-tubulin and goat anti-human IgG (Fc specific) antibodies were obtained from Sigma (St. Louis, MD, USA). Goat anti-rabbit and rabbit anti-mouse secondary antibodies were obtained from Jackson ImmunoResearch (Bar Harbor, ME, USA).
Dual luciferase assay
HEK293T cells were seeded in 384-well plates 24 hr prior to transfection. Cells were transfected with the reporter construct TCF-firefly (10 ng) , actin-Renilla (10 ng) and Wnt3a (5 ng) or mutant isoforms, and balanced with an empty expression vector (pRL-CMV) using TransIT (Mirus Bio LLC, Madison, WI, USA) according to the manufacturer’s protocol. The activity of the firefly luciferase and Renilla luciferase was measured using the dual luciferase assay kit (Promega, Madison, WI, USA) and detected using a Mithras LB940 Multimode Microplate Reader Luminometer (Berthold Technologies, Bad Wildbad, Germany). The activity of the Renilla luciferase was measured as an internal control of transfection efficiency. The activity of the Wnt-induced Wnt reporter is depicted as the ratio of the activity of the firefly to Renilla luciferase. All assays were performed in triplicate and repeated in three independent experiments.
Production of mouse Wnt3a and mutant isoforms in HEK293T cells
To collect a conditioned medium of wild-type mouse Wnt3a and the indicated mutants, HEK293T cells were transiently transfected with the corresponding cDNAs under control of a Cytomegalovirus immediate early gene enhancer and promoter. For protein expression, cells were incubated at 37°C for 48 hr in DMEM (Invitrogen, Waltham, MA, USA) containing 10% FBS and penicillin/streptomycin. The conditioned media were harvested, centrifuged at 2,000 g for 10 min, and directly used or stored in aliquots at −80°C. The control medium was prepared from untransfected HEK293T cells following the same protocol. For the solution binding assay, the conditioned media containing wild-type Wnt3a or mutant isoforms were produced in a low serum (3% FBS) condition and adjusted to give equal Wnt3a protein levels by quantifying Western blot bands as described below. To detect the secretion levels of wild-type Wnt3a and mutant isoforms, equal amounts of conditioned medium were resolved on 12% SDS-PAGE and analyzed by Western blot using anti-Wnt3a antibody.
To produce the NTD-His and CTD-His conditioned medium, HEK293T cells were transiently transfected with 10 μg of the relevant plasmid. One day after transfection, cells were transferred to advanced low serum DMEM/F-12 (GIBCO, Invitrogen Corporation, Waltham, MA, USA), and secreted proteins were harvested after an additional 48 hr. Proteins were purified by incubating the conditioned medium with Ni-NTA agarose beads (Qiagen, Hilden, Germany) for 1 hr at 4°C. Bound proteins were solubilized with elution buffer (300 mM imidazole, 250 mM NaCl, 1× PBS). Purified proteins were resolved on SDS-PAGE and analyzed by Western blot using anti-His antibodies.
Production of recombinant Fz8-CRD proteins
The mouse Fz8-CRD-Fc and Fc were produced in HEK293T cells that were transiently transfected by using jetPRIME™ reagent (Polyplus Transfection, USA). One day after transfection, cells were transferred to advanced low serum DMEM/F-12 (GIBCO, Invitrogen corporation), and secreted protein was harvested after an additional 48 hr. The control conditioned medium was obtained from untransfected HEK293T cells. To purify the Fz8-CRD-Fc and Fc, the conditioned medium was incubated with Protein G Agarose (Roche Diagnostics GmbH) for 3 hr at 4°C, and washed three times with washing buffer (10 mM Tris-Cl (pH 7.8), 0.2 mM EDTA, 250 mM NaCl, 0.1% Tween). Proteins were eluted in low pH elution buffer (0.1 M glycine-HCl (pH 2.5), 0.1% Tween) and immediately neutralized with 1 M Tris–HCl (pH 9.0). Purified protein was resolved on SDS-PAGE and analyzed by Western blot and Coomassie blue staining.
Mouse Fz8-CRD-His was cloned into the pCEP-PU vector (Invitrogen). To produce the Fz8-CRD-His protein, HEK293T cells were transiently transfected with 10 μg of plasmid. The conditioned medium was harvested after 48 hr of transfection. Fz8-CRD-his was further purified by incubating 50 ml of conditioned medium with 200 μl Ni-NTA agarose beads (Qiagen) for 1 hr at 4°C. The beads were washed three times with 1× PBS, 0.1% Tween. Bound protein was eluted with elution buffer (300 mM imidazole, 250 mM NaCl, 1× PBS (pH 7.5)).
Solution binding assay
First 12 μg recombinant mouse Fz8-CRD-Fc or Fc alone as a negative control were pre-incubated with Protein G Agarose (Roche Diagnostics GmbH) for 3 hr at 4°C, then the beads were washed twice with 0.01% Triton X-100, 1× PBS, and twice with 1× PBS. The beads were then incubated with 1 to 2 ml of Wnt3a conditioned medium for 3 to 4 hr. The beads were separated by centrifugation and washed four times with 0.01% Triton X-100, 1× PBS buffer and boiled in 2× SDS sample buffer containing beta-mercaptoethanol. The supernatants were analyzed by Western blot analysis using anti-Wnt3a primary antibodies. The membranes were further re-probed with anti-human IgG antibodies.
Blue Sepharose pull-down of Wnt3a proteins from conditioned medium
Wnt3a-conditioned medium (2 to 5 ml) was collected from HEK293T cells transiently transfected with different Wnt3a expression constructs. The Wnt3a protein levels were adjusted according to the quantitative band intensities after Western blotting with mouse ant-Wnt3a antibodies. The conditioned medium was incubated with 30 μl Blue Sepharose beads (GE Healthcare Life Sciences, Chalfont St. Giles, UK) on a rocking platform for 2 hr at 4°C. The beads were washed four times with washing buffer (1× PBS, 0.1% Tween-20). A protein sample buffer containing SDS and 150 mM DTT was added to the beads and the mixture samples were heated at 95°C for 5 min. Proteins were resolved on 12% SDS-PAGE and analyzed by Western blot using anti-Wnt3a antibodies.
Modulation of Wnt3a activity and solubility by soluble Fz8-CRD protein
To investigate the effect of recombinant Fz8-CRD-Fc on Wnt activity, HEK293T cells were transfected with reporter construct TCF-firefly (10 ng) , actin-Renilla (10 ng) and Wnt3a (5 ng). After 24 hr of transfection, the indicated amount of Fz8-CRD-Fc was added to the cells and the relative luciferase activity was measured after another 24 hr .
To investigate the effect of recombinant Fz8-CRD protein on Wnt solubility, HEK293T cells were transfected with wild-type or W333A mutant Wnt3a plasmid in 24-well plates and cultivated in 500 μl DMEM medium supplemented with 10% FCS. After 24 hr, the indicated amount of recombinant Fz8-CRD-Fc was added to the Wnt3a-producing cells, which were further incubated for 16 to 20 hr. The supernatant containing wild-type or mutant Wnt3a protein was analyzed by Western blot using the anti-Wnt3a antibody.
For the dual luciferase chamber assay, HEK293T cells were seeded on thinCerts TC chambers with a pore size of 8.0 μm (Greiner, Frickenhausen, Germany), transfected with 0.25 μg CMV-β-galactosidase and 0.4 μg Super TOPFlash by Ca++-phosphate precipitation and cultivated in 1 ml DMEM supplemented with 10% FCS. Then 40 hr before analysis, the inlets with the transfected cells were transferred to mWnt3a-producing cells (ATCC, Manassas, VA, USA: CRL-2647) seeded in 500 μl DMEM supplemented with 10% FCS in 12-well plates. Then, 16 hr before analysis the indicated amount of recombinant Fz-CRD protein was added to the Wnt-producing cells.
mRNA microinjections in frog (Xenopus) and zebrafish (Danio rerio) embryos
Xenopus eggs were obtained from females injected with 500 IU of human chorionic gonadotropin (Sigma-Aldrich, USA) and fertilized in vitro. Embryos were dejellied in 2% cysteine hydrochloride (pH 8) and microinjected into 1× MBS-H (88 mM NaCl, 1 mM KCl, 2.4 mM Na2HCO3, 0.82 mM MgSO4, 0.33 mM Na(NO3)2, 0.41 mM CaCl2, 10 mM HEPES pH 7.4, 10 μg/ml streptomycin sulfate, and 10 μg/ml penicillin). The embryos were cultured in 0.1× MBS-H and staged according to Nieuwkoop . Capped mRNAs coding for wild-type mWnt3a, NTD and CTD were synthesized from linearized pCS2+ plasmids using the Message Machine Kit (Ambion, Grand Island, NY, USA). Plasmids were linearized with NotI and transcribed with SP6. Embryos for the axis-duplication assay were injected at the ventral marginal zone with the indicated amount of mRNAs and screened for axis-duplication after 48 hr with an Olympus (Hamburg, Germany) SZX12 stereomicroscope and a Soft Imaging System CC-12 digital camera.
Zebrafish of the AB wild-type line were raised and staged as described previously . For zebrafish injection experiments, a series of 1.5, 3, 33, 100 and 300 pg of wild-type mWnt3a mRNA was first injected and dose-dependence evaluated. Since 3 pg of mWnt3a or more caused very severe abnormalities at 1 dpf, therefore 2 pg of each mRNA was microinjected into one-cell-stage embryos in experiments as shown in Figure 6. Embryos were incubated in E3 medium (with additional PTU treatment after 20 hr post fertilization) at 28°C and examined at 1, 1.5 and 3 dpf. Zebrafish embryos were either photographed live or fixed with 4% PFA and stained with Alexa-488 conjugated Phalloidin (Molecular Probes, Grand Island, NY, USA), mounted in 1% low melting point agarose and imaged on an A1R (Nikon, Tokyo, Japan) confocal microscope using a 40× LWD water immersion objective.
To monitor expression levels of the mouse Wnt3a proteins, 25 pg of the His-tagged wild-type mouse Wnt3a capped mRNA or its mutant versions were injected into zebrafish embryos at the one-cell stage. After 8 to 9 hr (at the gastrula stage), 20 embryos were lysed in 100 μl SDS-PAGE sample buffer (100 mM Tris–HCl (pH 6.8), 20% glycerol (v/v), 4% SDS (w/v), 100 mM DTT) and homogenized by passing through a fine needle. The embryo lysate was further centrifuged at 12,000g for 10 min at 4°C to remove debris. The supernatants were transferred to a new tube and the proteins were separated on 12% SDS-PAGE and analyzed by Western blotting using anti-His5 antibodies.
Western blot quantification
Chemiluminescence was detected with a photographic film (Amersham Hyperfilm ECL, GE Healthcare, Chalfont St. Giles, UK) in a Curix 60 processor (Agfa, Mortsel, Belgium). Films of immunoblots were scanned into TIF format using a CanoScan LiDE 35 (Canon, Tokyo, Japan) and digital images were imported and quantified using ImageJ software (National Institutes of Health, Berthesda, MD, USA) following the method outlined in . The quantitative band intensities of Wnt3a mutant proteins were normalized relative to the wild-type Wnt3a protein band intensity.
Modeling of Wnt3a structure
We performed homology modeling based on the most recently determined crystal structure of the XWnt8-Fz8-CRD complex  (pdb-code: 4F0A). Both ligand and receptor had homology scores of >75%. For the subsequent modeling steps, two criteria were important: the e-value as returned by BLASTP against the PDB dataset and the 3D-Jury score from metaserver, which is an approximation of the number of Cα atoms within 3.5 Å from the PDB template compared to the current theoretical model. Since the secondary structure coverage of PDB corresponded to the secondary structure elements predicted for Wnt3A by PSIPRED , DSSP , STRIDE  and PROFSEC (B. Rost, unpublished data), it seemed reasonable to try a homology modeling protocol to the PDB from XWnt8-Fz8-CRD as templates using alignment information provided by metaserver. Modeller9v2’s  script for automated model generation using homology modeling was run with successive molecular dynamics simulated for annealing refinement (refine.slow). Optimization was repeated four times to get the first impression of the model. Finally, the model was optimized with the variable target function method (VTFM), which tries to minimize constraint violations (e.g., torsion angles) with a maximum of 300 iterations.