Protocol for the purification of Wnt proteins

April 21, 2003, Written by Karl Willert

Additions on storage of tissue culture medium and purified protein, Sept 10, 2003

See also: Willert, K., Brown, J. D., Danenberg, E., Duncan, A. W., Weissman, I. L., Reya, T., Yates, J. R., and Nusse, R. (2003). Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature 423, 448-452. Download a PDF file

Some modifications introduced by Kishida et al, (2004)

Note that Janda et al. (2012), and Chu et al. (2013) have obtained crystal structes of Wnt proteins. 

Note: Several Wnt proteins are being sold by R&D systems.

Necessary reagents
beta-catenin stabilization assay                                                                                                                                                                                            
Cells producing Wnt3A
Conditioned Medium
Blue Sepharose Fractionation
Heparin Cation Exchange
Purification in detail
The old text on Wnt production, references to earlier work

Necessary reagents:

  • Cell line producing and secreting Wnt protein (LWnt3A cell, ATCC#CRL-2647). LWnt5A ATCC# CRL-2814). The medium is DMEM plus 10% FBS, omitting the serum lowers the levels of Wnt protein in the medium. We attribute the serum requirement to the lipid modification of Wnt; in the absence of serum, the protein tends to precipitate due to its hydrophobic nature.
  • Blue Sepharose column:

Small scale: HiTrap Blue, 1ml and 5ml (Amersham Biosciences)

Large scale: column with a bed volume of 100 to 120ml Blue Sepharose HP (Amersham Biosciences, cat# 90-1000-22)

  • Gel Filtration column:

HiLoad 16/60 or 26/60 Superdex 200 prep grade (Amersham Biosciences)

  • Cation Exchange column:

HiTrap Heparin, 1ml (Amersham Biosciences)

  • Chemicals: CHAPS, Triton X-100, KCl, Trizma, PBS, NaCl
  • AcroPak 200 Supor 0.2 micrometer filter units (Pall) (for large scale purification)
  • Ultrafiltration devices for ~20 to 60 ml, 10-30kD MWCO (e.g. Amicon Centricon 30 or Amicon Ultra 30k)
  • Optional: anti-Wnt antibody for monitoring Wnt purification steps by immuno-blotting prior to detection by Coomassie. This is particularly valuable for the Blue Sepharose step since the concentration of Wnt is borderline for detection by Coomassie.
  • mouse anti-√ü-catenin antibody (available commercially through Santa Cruz Biotech or Transduction Laboratories).



Description of b-catenin stabilization assay: Mouse L cells seeded in 96 well plates.
Add Wnt protein (dilution of conditioned medium or dilution of purified protein in complete medium).
Incubate cells 2 hours in 37C/CO2 incubator.
Aspirate medium.
Wash cells once with 1X PBS.
Aspirate PBS.
Lyse cells by adding 30 ul lysis buffer (1% Triton X-100, 150mM NaCl, 50mM Tris-HCl, pH8).
Add lysate to Laemmli buffer and boil 5 minutes.
Resolve 20ul of each sample by SDS-10% PAGE.
Transfer proteins to nitrocellulose and blot with mouse anti-b-catenin antibody.



  • Cell lines producing Wnt. The L cells we have deposited with the ATCC producing and secreting Wnt3A protein (LWnt3A cell, ATCC#CRL-2647) express a non-tagged form of Wnt3A protein. We have tried to insert epitopes at several sites and found that the activity of the secreted protein is dimished. In these cells, Wnt3A is expressed from the PGK promoter, which works very well (Shibamoto et al (1998), We did select clones that express and secrete highest levels of protein; by collecting all transfected cells we obtained much lower levels. Thus, for generating other Wnts, we recommend the PGK promoter driving the native Wnt gene, and selection of highest expressing clones of c ells.
  • Conditioned Medium: We found that we obtained highest amounts of Wnt-3A from LWnt-3A cells by growing them adherent for 4 days from a 1:10 to 1:20 split. The medium is DMEM plus 10% FBS, omitting the serum lowers the levels of Wnt protein in the medium. We attribute the serum requirement to the lipid modification of Wnt; in the absence of serum, the protein tends to precipitate due to its hydrophobic nature.
  • After this first batch we replenish the media and incubate the cells for an additional 3 days. After this the cells become too dense and lyse. After harvesting, filter the medium through 0.2 micrometer; at this stage the medium can be stored for several months at 4 degrees C without appreciable loss in activity. After addition of detergent prior to fractionation on Blue Sepharose, re-filter the medium through 0.2mm.
  • Blue Sepharose Fractionation: A large-scale purification (2-4L Wnt3A CM on a 120ml Blue Sepharose HP column) yields the purest Wnt protein. Blue Sepharose HP column
  • Wnt elution from Blue Sepharose is done in a single step from 150mM to 1.5M KCl; approximately half of the Wnt protein elutes immediately with the salt and the majority of contaminants, but the other half is retained and elutes later in a second pool. This second pool contains much less total protein and consequently contains a higher proportion of Wnt. At this time we don't know the reason for this slow elution profile but suspect hydrophobic interactions with the resin that retard the elution of Wnt but not of the other contaminating proteins. As a result of this slow elution, some Wnt protein will likely remain on the column even after extensive washing with elution buffer. To avoid contaminating a new Wnt protein preparation with the Wnt from a previous preparation, we recommend dedicating a single Blue Sepharose column to the purification of a single Wnt protein. Alternatively, the column can be wasked with 0.2M NaOH.
  • Heparin Cation Exchange: This step serves the purpose of concentrating the protein and removing a predominant contaminant (most likely BSA). The final concentration of Wnt3A is about 0.1 mg/ml. If the concentration is higher, a precipitate will form which is predominately Wnt protein. This precipitate can be pelleted easily; the remaining supernatant will contain Wnt3A at about 0.1 mg/ml while the pellet contains largely inactive Wnt protein. Therefore, the present conditions (1X PBS, 1M NaCl, 1% CHAPS, pH7.3) can maintain the solubility and activity of Wnt3A at a maximum of 0.1 mg/ml.



Starting Material: Four day Wnt-3A conditioned medium from L-Wnt-3A; this medium contains about 100-200 ng Wnt-3A/ml. Filter, add detergent (Triton X-100 or CHAPS) to 1% final concentration, and re-filter just before applying the material to Blue Sepharose.


All purification steps are carried out at 4C. Do not freeze any fractions as this may be detrimental to the protein.


Small scale purification: 10-100 ml Wnt-3A CM on 1 ml or 5 ml HiTrap Blue Sepharose column. For the 5 ml HiTrap Blue column the binding capacity is about 100 ml CM for Wnt-3A.


Large scale purification: >100 ml Wnt-3A CM on 5ml HiTrap Blue columns in series or using a home packed column with Blue Sepharose HP (APB cat# 90-1000-22). For very large volumes (up to 4L) we use a sample pump for sample injections and a column packed to a bed volume of about 120ml Blue Sepharose HP. The Sepharose FF does not perform as efficiently as Sepharose HP.


1. Blue Sepharose

Small scale:

1-5ml HiTrap Blue column with up to 50ml Wnt CM + 1% Triton X-100 or CHAPS.

Eluent A: 1% CHAPS, 150mM KCl, 20mM Tris-HCl, pH7.5

Eluent B: 1% CHAPS, 1.5M KCl, 20mM Tris-HCl, pH7.5

Flow rate:1-2ml/min

Equilibrate column in Eluent A.

Inject Sample using Superloop.

Wash out unbound sample with 5 column volumes.

Elute bound sample with step gradient to 100% eluent B for 10 column volumes.

Collect 1-5ml elution fractions.


Large Scale

100-120 ml Blue Sepharose column with up to 4L Wnt CM + 1% Triton X-100 or CHAPS.

Eluent A: 1% CHAPS, 150mM KCl, 20mM Tris-HCl, pH7.5

Eluent B: 1% CHAPS, 1.5M KCl, 20mM Tris-HCl, pH7.5

Flow rate: 2-5ml/min

Equilibrate column in Eluent A.

Inject sample using sample pump.

Wash column with Eluent A with at least 4 column volumes.

Elute with Eluent B and collect 5-15 ml fractions.


Note: On a large Blue Sepharose column (CV ~120ml), Wnt3A elutes slowly despite the step elution, with several Wnt3A fractions trailing behind the main protein peak. The later fractions contain about half of the total Wnt3A protein but much less total protein. This material is much cleaner and will eventually produce the purest Wnt protein prep.


2. Gel Filtration

Sample: pooled Wnt containing fractions from Step 1 concentrated to 5 or 10 ml volume using Centricon or Amicon Ultra 30 ultrafiltration device (Amicon) depending on size of GF column.

Column: HiLoad 16/60 or 26/60 Superdex 200 prep grade

Eluent: 1X PBS, 1% CHAPS

Flow rate: 2ml/min

Inject Sample using Superloop.

Collect 5ml fractions for 16/60 and 10ml fractions for 26/60.

On HiLoad16/60 column Wnt3A will elute at 65-80 ml following a 5ml injection.

On HiLoad26/60 column Wnt3A will elute at 180-210 ml following a 10ml injection.


Note: The fraction numbers may change with varying buffer conditions and with performance/age of column. Especially note that lower concentrations of salt will cause the Wnt to be retained longer; at very low levels of salt (>50mM NaCl) the Wnt will emerge in the column volume fractions.


3. Heparin Cation Exchange Chromatography

Sample: pooled Wnt containing fractions from Step 2

Column: HiTrap Heparin, 1ml

Eluent A: 1X PBS + 1% CHAPS

Eluent B: Eluent A + 1 M NaCl.

Step elution from 0 to 100% Eluent B

Collect 1ml elution fractions.

We store the purified protein at 4 degrees C. It is then stable, although we sometimes add BSA to 100 micrograms per ml. We have also been able to freeze-dry the protein with retention of activity.

(what follows is the old and somewhat outdated text on this topic. It still contains some useful information)

There are numerous unpublished tales of failed attempts to produce secreted Wnt proteins in cell culture. In general, overexpression of the genes in cultured cells results in accumulation of misfolded protein in the ER [Kitajewski, 1992]. Secreted forms of Wnts can be found in the extracellular matrix or the cell surface [Bradley, 1990 ; Papkoff, 1990 ; Burrus, 1995], but efforts to solubilize this material have not been successful. Addition of suramin or heparin to cells can lead to a significant increase of Wnt protein in the medium [Bradley, 1990; Papkoff, 1990 , but this protein has not been shown to be biologically active [Burrus, 1995].

  • It should in particular be noted that attempts to produce Wnts in Baculovirus systems have NOT worked, nor have Yeast and E. coli overexpression systems (unpublished misery in many labs).

While under any circumstance most Wnt protein is cell bound, several systems have been developed that produce soluble forms.

  • The Drosophila Wg [Van Leeuwen, 1994 ] and DWnt-3 [Fradkin, 1995] proteins and the mouse Wnt-1 protein [Bradley, 1995] have been recovered form the medium of cultured cells. The Drosophila proteins are made by S2 cells, which may turn out to be the best system to make Wnts. Recent work in Nathan's lab has indicated that these cells can also produce soluble Xenopus Wnt-8 that is active (Hsieh JC, 1999). The amounts secreted are not overwhelming, but using in vitro assays for activity, these soluble forms have been shown to be biologically active. Wg protein can be tested for the stabilization of the Armadillo protein [Van Leeuwen, 1994 ) and Wnt-1 protein can induce morphological transformation of target cells [Jue, 1992; Bradley, 1995].
  • Using a hematopoietic stem cell proliferation assay, several Wnts have been shown to be active in solution, and one of these, Wnt-5A, has been partially purified while retaining activity [Austin, 1997].
  • Another useful system to work with soluble Wnt proteins was developed by Shibamoto et al (1998), showing the soluble Wnt3A can affect the cytoskeleton in target cells. This soluble Wnt-3A can also accumulate b-catenin in target cells (Kishida et al, 1999).