Secreted Frizzled-related protein potentiation versus inhibition of Wnt3a/β-catenin signaling

Abstract

Wnt signaling regulates a variety of cellular processes during embryonic development and in the adult. Many of these activities are mediated by the Frizzled family of seven-pass transmembrane receptors, which bind Wnts via a conserved cysteine-rich domain (CRD). Secreted Frizzled-related proteins (sFRPs) contain an amino-terminal, Frizzled-like CRD and a carboxyl-terminal, heparin-binding netrin-like domain. Previous studies identified sFRPs as soluble Wnt antagonists that bind directly to Wnts and prevent their interaction with Frizzleds. However, subsequent observations suggested that sFRPs and Frizzleds form homodimers and heterodimers via their respective CRDs, and that sFRPs can stimulate signal transduction. Here, we present evidence that sFRP1 either inhibits or enhances signaling in the Wnt3a/β-catenin pathway, depending on its concentration and the cellular context. Nanomolar concentrations of sFRP1 increased Wnt3a signaling, while higher concentrations blocked it in HEK293 cells expressing a SuperTopFlash reporter. sFRP1 primarily augmented Wnt3a/β-catenin signaling in C57MG cells, but it behaved as an antagonist in L929 fibroblasts. sFRP1 enhanced reporter activity in L cells that were engineered to stably express Frizzled 5, though not Frizzled 2. This implied that the Frizzled expression pattern could determine the response to sFRP1. Similar results were obtained with sFRP2 in HEK293, C57MG and L cell reporter assays. CRDsFRP1 mimicked the potentiating effect of sFRP1 in multiple settings, contradicting initial expectations that this domain would inhibit Wnt signaling. Moreover, CRDsFRP1 showed little avidity for Wnt3a compared to sFRP1, implying that the mechanism for potentiation by CRDsFRP1 probably does not require an interaction with Wnt protein. Together, these findings demonstrate that sFRPs can either promote or suppress Wnt/β-catenin signaling, depending on cellular context, concentration and most likely the expression pattern of Fzd receptors.

Keywords: BSA; CRD; Cysteine-rich domain; Frizzled; Fzd; PBS; Wnt; bovine serum albumin; cysteine-rich domain; phosphate-buffered saline; sFRP; secreted Frizzled-related protein; β-Catenin.

Fig. 1

sFRP1 modulation of Wnt3a/β-catenin promoter reporter activity varies with cell context. (A) HEK293 cells stably expressing a SuperTopFlash reporter, and (B) C57MG and (C) L cells transiently expressing the Super8XTopFlash reporter were incubated overnight with the indicated concentrations of sFRP1 in the presence or absence of Wnt3a (2.5 nM). In the absence of Wnt3a, BSA was added as a vehicle control. The luminescence of cell lysates was measured and normalized to total cellular protein concentration. Relative luminescence was expressed in arbitrary units (a.u.) with 1 a.u. = luminescence observed in the absence of Wnt3a and sFRP1. Results are expressed as the mean ± S.D. of biological triplicate samples from a representative experiment (at least 3 experiments were performed with each cell line). Statistical significance was determined with Student's t-test, comparing results to those obtained in the presence of Wnt3a alone. *P < 0.05; **P < 0.005; ***P < 0.0005. (D) HEK293/STF and (E) L cells stably expressing pBARLS were incubated with the indicated concentrations of Wnt3a in the absence of sFRP1 (circles) or the presence of 3 nMsFRP1 (squares) or 300 nMsFRP1 (triangles). Results from representative experiments (3 performed with each cell line) were obtained and expressed as described in (A–C).

Fig. 2

sFRP1 regulation of Wnt3a/β-catenin signaling in L cells expressing HA tagged Fzd2 vs. Fzd5. (A) Reporter assay. L cells stably expressing the pBARLS reporter were stably transfected with empty lentiviral vector (L/eV) or vector encoding HA-tagged Fzd2 (L/Fzd2) or Fzd5 (L/Fzd5). Cells were incubated overnight with the indicated concentrations of sFRP1 andWnt3a. Relative luminescence was measured as described in the legend in Fig. 1. Results are expressed as the mean ± S.D. of biological triplicate samples from a representative experiment (at least 3 experiments were performed with each cell line). Statistical significance was determined with Student's t-test, comparing results to those obtained in the presence of Wnt3a alone. *P < 0.05; **P < 0.005. (B) Western blot analysis. Equivalent amounts (30 µg/lane) of cell lysates from the three L cell transfectants were immunoblotted with antibody to the HA epitope. The positions of molecular mass markers are indicated. (C) β-catenin stabilization assay. L/eV and L/Fzd5 cells were treated for 3 h with the indicated concentrations of sFRP1 and/orWnt3a. In the absence of Wnt3a, BSA was added as a vehicle control. Equivalent amounts of cell lysates (30 µg/lane) were immunoblotted for β-catenin and HSP70, the latter serving as a loading control. (D) Intracellular distribution of β-catenin. L/eV and L/Fzd5 cells were treated withWnt3a and 300 nMsFRP1 for 3 h. Cells were fixed, permeabilized and stained with β-catenin antibody, phalloidin to visualize polymerized actin and DAPI to detect nuclear β-catenin by confocal microscopy. Bars = 20 µm. (E) Quantitative analysis of experiments illustrated in (D), indicating the number of cells surveyed that displayed nuclear β-catenin.

Fig. 3

sFRP1 and sFRP2 regulation of Wnt3a/β-catenin-dependent reporter activity in HEK293/STF, C57MG and L/Fzd5 cells. (A) HEK293/STF, (B) C57MG and (C) L/eV and L/Fzd5 cells were incubated overnight with the indicated concentrations of sFRP1 or sFRP2 andWnt3a. Relative luminescence was measured and statistical significance was determined with Student's t-test, comparing results to those obtained in the presence of Wnt3a alone. *P < 0.01; **P < 0.001.

Fig. 4

Comparison of CRD_sFRP1 and sFRP1 effects onWnt3a/β-catenin-dependent transcription. (A)HEK293/STF cells were treated overnight with the indicated concentrations of CRD_sFRP1 or sFRP1 and Wnt3a, and subsequently processed for measurement of relative luminescence as previously described. (B) C57MG cells were treated overnight with the indicated concentrations of sFRP1 or CRD_sFRP1 andWnt3a, followed by quantitative PCR analysis of Axin2 and β-actin. Results in (A) and (B) are expressed as the mean ± S.D. of biological triplicate samples from a representative experiment [3 experiments were performed for (A) and 2 for (B)]. Statistical significance of data in (A) and (B) was determined with Student's t-test, comparing results to those obtained in the presence of Wnt3a alone. *P < 0.05; **P < 0.005.

Fig. 5

Co-immunoprecipitation of Wnt3a with Myc/His-tagged sFRP1 and CRD_sFRP1 derivatives. Wnt3a was incubated with epitope-tagged full-length sFRP1 or sFRP1-Δ1, an analog of CRD_sFRP1, and subsequently co-immunoprecipitated with Myc antibody. Pelleted proteins were immunoblotted with antibodies toWn3a (upper panel) and Myc (lower panel). The lower panel also includes aliquots corresponding to 12% of sFRP1 and 29% of sFRP1-Δ1 used in the co-immunoprecipitation experiment. Below the panels is a schematic diagram of sFRP1-Δ1 highlighting the amino acid residues that define its boundaries.