The primary objective of this project has been to obtain a better understanding of the factors that account for the potentiation or inhibition of Wnt3a/beta-catenin signaling by sFRP1. Such knowledge would be beneficial in clinical applications that involve the use of sFRP1 to control Wnt signaling. Cell context is a determinant of the effect sFRP1 has on Wnt3a/beta-catenin signaling. sFRP1 has biphasic activity in parental HEK293 cells and HEKSTF cells (a clonal line stably expressing a SuperTopFlash reporter construct), enhancing beta-catenin signaling at 1-10 nM and inhibiting it at 100-300 nM. In contrast, sFRP1 primarily stimulated Wnt3a activity in the mouse mammary epithelial line C57MG in this concentration range, but was a strict inhibitor in L929 fibroblasts (L cells) even at low concentrations. Receptor expression is a key factor in the response to sFRP1. Ectopic over-expression of Fzd5, but not Fzd2, in L cells enabled potentiation of Wnt3a activity. We had hypothesized that Fzd5 would contribute to a potentiating effect at low sFRP1 concentrations, because of its homology to DFz2, the Drosophila Fzd expressed in S2 cells that previously had shown a biphasic response to sFRP1 in the presence of Wingless. The potentiation of signaling correlated with an increase in soluble beta-catenin protein in HEKSTF cells and L/Fzd5 transfectants. The inhibitory effect of sFRP1 in parental L cells was associated with a decrease in beta-catenin protein. However, a different pattern was seen in HEKSTF cells: the suppression of reporter activity observed with 100 and 300 nM sFRP1 did not result in diminished levels of soluble beta-catenin. CRDsFRP1 potentiated Wnt3a activity in HEKSTF and C57MG cells, but had little or no inhibitory activity. As previously reported for Wg, CRDsFRP1 exhibited little avidity for Wnt3a binding in ELISA, Biacore or co-immunoprecipitation assays in contrast to full-length sFRP1. We hypothesized that CRDsFRP1 promoted Wnt signaling by interacting with Fzd, most likely via CRD-CRD binding. CRDsFRP1 co-immunoprecipitated with full-length Fzds when co-expressed in HEK293 cells, although there was not a significant difference in its association with Fzd2 or Fzd5. However, in cell-based and cell-free assays of Wnt3a binding to Fzds or CRDFzd-Fc fusion proteins, we observed little effect of sFRP1 either at low or high concentrations. In summary, we determined that sFRP1 had a variety of dose-dependent effects on Wnt3a/beta-catenin signaling depending on the cell model. The balance between potentiation and inhibition presumably depends on the relative expression of individual Fzds, with Fzd5 being able to support potentiation at low sFRP1 concentrations. While the enhancement of beta-catenin signaling was associated with increases in soluble beta-catenin protein, inhibition appears to depend on different mechanisms: either suppression of beta-catenin stabilization as seen in L cells or another process observed in HEKSTF cells that has not yet been elucidated. Because CRDsFRP1 mimicked the potentiating activity of sFRP1 but did not bind significantly to Wnt3a, we suspect that its effect is mediated by interaction with Fzds via the CRDFzd or by other, as yet uncharacterized mechanisms. The 32D cell is an attractive model for the study of signaling that results from specific Wnt/Fzd interactions because of the limited amount of endogenous Wnt ligand/receptor expression in these cells. We have stably transfected 32D cells with nine of the ten mammalian Fzds, all with N-terminal HA epitope tags. Thus far, attempts to introduce Fzd3 have been problematic: only low levels of expression have been detected and even this signal is transient, suggesting negative selection. Positive results have been obtained with a beta-catenin protein stabilization assay using Wnt3a and a subset of the 32D/Fzd transfectants.
Showing the most recent 10 out of 17 publications
