Structural and functional analysis of Dickkopf 4 (Dkk4): New insights into Dkk evolution and regulation of Wnt signaling by Dkk and Kremen proteins

Dickkopf (Dkk) family proteins are important regulators of Wnt signaling pathways, which play key roles in many essential biological processes. Here, we report the first detailed structural and dynamics study of a full-length mature Dkk protein (Dkk4, residues 19–224), including determination of the first atomic-resolution structure for the N-terminal cysteine-rich domain (CRD1) conserved among Dkk proteins. We discovered that CRD1 has significant structural homology to the Dkk C-terminal cysteine-rich domain (CRD2), pointing to multiple gene duplication events during Dkk family evolution. We also show that Dkk4 consists of two independent folded domains (CRD1 and CRD2) joined by a highly flexible, nonstructured linker. Similarly, the N-terminal region preceding CRD1 and containing a highly conserved NXI(R/K) sequence motif was shown to be dynamic and highly flexible. We demonstrate that Dkk4 CRD2 mediates high-affinity binding to both the E1E2 region of low-density lipoprotein receptor–related protein 6 (LRP6 E1E2) and the Kremen1 (Krm1) extracellular domain. In contrast, the N-terminal region alone bound with only moderate affinity to LRP6 E1E2, consistent with binding via the conserved NXI(R/K) motif, but did not interact with Krm proteins. We also confirmed that Dkk and Krm family proteins function synergistically to inhibit Wnt signaling. Insights provided by our integrated structural, dynamics, interaction, and functional studies have allowed us to refine the model of synergistic regulation of Wnt signaling by Dkk proteins. Our results indicate the potential for the formation of a diverse range of ternary complexes comprising Dkk, Krm, and LRP5/6 proteins, allowing fine-tuning of Wnt-dependent signaling.

(A) Multiple sequence alignment of mammalian and reptilian Dkk4N (human, macaque, mouse, rat, dog, cow, turtle and alligator). Residues with completely conserved sequence identity are highlighted in red and those with conserved sequence similarity in 70% or more sequences are highlighted in yellow. The consensus sequence is shown below. Amino acids with completely conserved sequence identity are shown in uppercase; those with conserved sequence identity in over 70% of the sequences are shown in lowercase. Similar residues were grouped as follows: AVILM, FYW, KRH, DE, STNQ, PG and C. The symbol '!' is used to denote either I or V, '%' denotes F or Y, and '#' denotes anyone of NDQE. (B) and (C) surfaces views of Dkk4 CRD1, with residues highlighted on the basis of sequence conservation, with residues that are identical across all the representative Dkk4 species shown in red and those with conserved sequence identity in over 70% of the sequences shown in yellow. The structure in (C) has been rotated by 180° about the Y axis relative to (B).

Figure S4. Representative Biolayer Interferometry Sensorgrams and Steady-State Analysis Curves
Obtained for Dkk4FL and Dkk4N Binding to LRP6 E1E2-Fc (A)-(B) The Dkk protein used, the binding partner and pH are indicated in the title of both panels. In each, the normalised sensorgrams (top) observed for the association and dissociation phases of the binding and the steadystate equilibrium binding curve (bottom) are shown. The steady-state curves obtained were fitted to a single binding model using Prism 6.0 to determine the KDs reported.

Mammalian Expression Vectors
Constructs to express Dkk1FL-His or Dkk4FL-His were subcloned into a UCB proprietary expression vector following PCR amplification of the appropriate inserts from Origene ORF cDNA clones. A C-terminal His tag was engineered into the reverse primer for each insert. The Dkk1FL-His construct used the native signal sequence whilst Dkk4FL-His was cloned in-frame and downstream of the human VL signal sequence.
Sequences encoding Krm1 ECD (residues 20-326 and UniProtID Q96MU8-1 with the isoform 2 and 3 insertion at position 31 of E to GPE) and Krm2 ECD (residues 26-332 and UniProt ID Q8NCW0-1) were cloned into the pNAFH-hFc vector, which contained a TEV cleavable C-terminal human Fc tag.
LRP6-E1E2 consisting of the first two propeller and EGF domains (residues 1-637, UniprotID O75581) with a C-terminal TEV protease site was isolated by PCR and cloned into the pMH vector, which contained an inframe human Fc expression tag. All plasmid inserts were verified by DNA sequencing prior to transfection (Holdsworth et al., 2012).

Preparation of Dkk Proteins
For the production of unlabelled, 15 N, or 15 N/ 13 C labelled proteins a single E. coli Rosetta™ 2(DE3) colony transformed with pLEICS-05 encoding Dkk2FL, Dkk4FL or Dkk4N was grown in LB or minimal media (containing 1 g/L 15 N ammonium sulphate and 3 g/l 13 C glucose if required, as the sole nitrogen and carbon sources) at 37 °C to an OD600 of 0.6-0.8. Protein expression was induced by the addition of 0.4 mM IPTG and cultures incubated at 37 °C for 4 hrs. For 15 N/ 2 H or 15 N/ 13 C/ 2 H labelled Dkk4FL expression a single E. coli BL21 (DE3) colony was transformed with pET29-b(+) encoding codon-optimised Dkk4FL. Cells were grown in 15 N/ 2 H or 15 N/ 13 C/ 2 H minimal media (100% D20) at 37 °C to an OD600 of 0.8-1.0. Protein expression was induced by the addition of 0.4 mM IPTG and cultures incubated overnight at 18 °C.
Cell pellets containing insoluble expressed Dkk proteins were resuspended in lysis buffer (50 mM Tris, 2 mM EDTA, 0.1 % (v/v) Triton X-100, pH 8.0) and lysed using a French® Pressure Cell Press (Thermo Fisher Scientific) at 1000 psi with the 40 000 max psi French® Pressure Cell (Thermo Fisher Scientific) pre-chilled to 4 °C. To isolate the inclusion bodies the cell lysate was centrifuged (12000 x g for 20 mins at 4 °C) and the inclusion bodies were washed 2-3 times with wash buffer (50 mM Tris, 10 mM EDTA, 0.5 % (v/v) Triton X-100, 10 mM DTT, pH 8.65) using a glass homogeniser on ice. The homogenate was centrifuged (12000 x g for 15 mins at 4 °C) after each wash step. After the final wash step the pellet was resuspended in 20 mL resolubilisation buffer (50 mM Tris, 11 mM DTT, 5 M GdmHCl, pH 8.70) before being centrifuged (12000 x g for 20 mins at 4 °C). The supernatant was decanted and the protein concentration determined from the A280.
The resolubilised material was diluted to a final concentration of 2 mg/mL in resolubilisation buffer. This was diluted 100-fold by adding drop-wise to room temperature refolding buffer (50 mM Tris, 0.45 M GdmHCl, 0.78 mM GSH, 0.44 mM GSSG, pH 8.65), which was gently stirred. Once all the resolubilised material had been added to the refolding buffer it was stirred for a further 2 hrs before transferring to 4 °C where it was left unstirred overnight. Refolded material was filtered and concentrated to a volume of 50-100 mL using a Vivaflow 200 polyethersulfone (PES) membrane (5000 or 10 000 MWCO; Sartorius) attached to MasterFlex easy-load pump (Cole Parmer).
The refold mix was dialysed against 50 mM Tris, 100 mM NaCl, pH 7.5 and applied to a 5 mL NiNTA superflow cartridge (QIAGEN) pre-equilibrated with 10 column volumes (CV) of loading buffer (50 mM Tris, 100 mM NaCl, 20 mM imidazole, pH 7.5). The column was washed with 25 CV loading buffer and the bound Dkk proteins were eluted using a linear gradient of loading buffer supplemented with 500 mM imidazole over 10 CV. Fractions containing Dkk were pooled and dialysed against size exclusion chromatography (SEC) buffer (25 mM bis-Tris, 100 mM NaCl, pH 5.5). The dialysed protein solution was concentrated to 5-10 mL using an Amicon-Ultra15 centrifugal filter unit with a regenerated cellulose membrane (10 or 3 kDa NWCO; Merck Millipore).
SEC was performed using a pre-packed HiLoad™ 16/60 Superdex™ 75 prep grade column (GE Healthcare) at room temperature. The column was pre-equilibrated with 3 CV of SEC buffer prior to loading 5 mL of protein onto the column at a rate of 1 mL/min. The A280 was monitored throughout the run and 2 mL fractions were collected and analysed by SDS-PAGE. Fractions from under each elution peak were pooled separately and stored at 4 °C. A calibration was performed for the SEC column to allow estimations of molecular weights. The calibration was done using the Gel Filtration LMW Calibration Kit (GE Healthcare). Globular protein standards were made up in 50 mM Na2HPO4, 100 mM NaCl, pH 7.5 at the manufacturers recommended concentrations. A linear regression analysis was performed on the relationship between the logarithm of the molecular weight and elution volume for the globular protein standards and was used to calculate the apparent molecular weights of the species in all subsequent samples analysed by size exclusion chromatography.

Preparation of Krm Proteins
For the production of Krm ECD proteins 2 x10 8 CHO-SXE cells were resuspended in 10 mL Earle's balanced salt solution (EBSS) and 4 mg of recombinant plasmid DNA was added. The cells were transfected using the Gene Pulser Xcell™ system (BioRad) by adding 800 µL of the plasmid DNA/cell mixture into a Gene Pulser electroporation cuvette and applying ~280-300 V. This was added to 1 L CD Gibco® CHO medium (supplemented with 0.01 % (v/v) GlutaMAX™ supplement and 0.002 % (v/v) Gibco® Antibiotic Antimycotic), which was prewarmed to 37 °C in a water bath. Cultures were incubated at 37 °C with shaking overnight following which the temperature was reduced to 32 °C. Sodium butyrate (3 mM) was added on day 3. On day 7, the cell number and viability were checked.
The cultures were harvested by centrifugation (4000 x g for 30 mins at 4 °C) and the supernatant filtered using a Sartobran® P 0.2 µm filter (Sartorius) and a Millipak® 40 Gamma Gold 0.22 µm filter (Merck Millipore). A final concentration of 0.02 % (w/v) NaN3 was added to prevent microbial growth. For purification, 10 mL of MabSelect SuRe Protein A beads (GE Healthcare) were pre-equilibrated with 5 CV protein A purification buffer (50 mM Na2HPO4, 100 mM NaCl, pH 7.5) at room temperature. The filtered supernatant was loaded onto the medium at 3 mL/min and the A280 and conductivity measured throughout the run. The beads were subsequently washed with 3 CV protein A purification buffer and resuspended in 10 mL protein A purification buffer.
The hFc tag was cleaved using TEV protease whilst the Krm recombinant protein was still bound to the protein A beads. 0.01 % (v/v) TEV protease (3 mg/mL; UCB Pharma) was added to the protein A beads and left at 4 °C overnight with gentle rocking. The beads were centrifuged (4500 x g for 5 mins at 4°C) and the supernatant decanted. The protein A beads were gently resuspended in 10 mL protein A purification buffer and washed to remove all cleaved Krm ECD recombinant protein. The supernatant was combined and concentrated to a final volume of 5-10 mL using an Amicon Ultra15 centrifugal filter unit with a regenerated cellulose membrane (10 kDa NWCO; Merck Millipore) prior to SEC. SEC was performed as described above using protein A purification buffer.

Preparation of LRP6 Proteins
Expression and purification of LRP6 E1E2 were carried out as described previously (Holdsworth et al., 2012).

Analysis of Purified Proteins
All proteins were analysed by SDS-PAGE and mass spectrometry under reducing and non-reducing conditions. To obtain an intact mass for Dkk2FL, protein samples (100 µM) were submitted for liquid chromatography-mass spectrometry (LC-MS/MS) analysis. Experiments were conducted under non-reducing conditions using a LTQ Orbitrap spectrometer (Thermo Fisher Scientific). All other samples were analysed using electrospray ionisation mass spectrometry (ESI-MS) on an Acquity H-Class UPLC system with a Xevo G2 QTof detector.