Secreted frizzled related protein 2 protects cells from apoptosis by blocking the effect of canonical Wnt3a

Abstract

We have demonstrated that mesenchymal stem cells overexpressing the survival gene Akt can confer paracrine protection to ischemic myocytes both in vivo and in vitro through the release of secreted frizzled related protein 2 (Sfrp2). However, the mechanisms mediating these effects of Sfrp2 have not been fully elucidated. In this study, we studied rat cardiomyoblasts subjected to hypoxia reoxygenation (HR) injury to test the hypothesis that Sfrp2 exerts anti-apoptotic effect by antagonizing pro-apoptotic properties of specific Wnt ligands. We examined the effect of Wnt3a and Sfrp2 on HR-induced apoptosis. Wnt3a significantly increased cellular caspase activities and TUNEL staining in response to HR. Sfrp2 attenuated significantly Wnt3a-induced caspase activities in a concentration dependent fashion. Using a solid phase binding assay, our data demonstrates that Sfrp2 physically binds to Wnt3a. In addition, we observed that Sfrp2 dramatically inhibits the beta-catenin/TCF transcriptional activities induced by Wnt3a. Impressively, Dickkopf-1, a protein that binds to the Wnt coreceptor LRP, significantly inhibited the Wnt3a-activated caspase and transcriptional activities. Similarly, siRNA against beta-catenin markedly inhibited the Wnt3a-activated caspase activities. Consistent with this, significantly fewer TUNEL positive cells were observed in siRNA transfected cells than in control cells. Together, our data provide strong evidence to support the notion that Wnt3a is a canonical Wnt with pro-apoptotic action whose cellular activity is prevented by Sfrp2 through, at least in part, the direct binding of these molecules. These results can explain the in vivo protective effect of Sfrp2 and highlight its therapeutic potential for the ischemic heart.

Figure 1

The expression levels of Sfrp2 and the effects of Wnt3a/Sfrp2 on apoptosis. (A) The mRNA and protein expression levels of Sfrp2 in sham, permanent ligation and reperfusion rat hearts. GAPDH was used as RT-PCR internal control; while CBB staining to show the protein loading. (B) and (C), H9C2 cells were exposed to hypoxia (24hrs) and reoxygenation (24 hrs) with Wnt3a or Sfrp2 or Wnt3a with the addition of different doses of Sfrp2. Then the cells were assayed for caspases activities or TUNEL staining. The data of caspase are expressed as mean ± SD (B). In Fig C, hypoxic (24hrs) and reoxygenated (24hrs) cells, which were cultured in 4-well slide-chambers with 3nM Wnt3a or Wnt3a(3nM)/Sfrp2(300nM), were stained with DeadEnd Colorimetric TUNEL System. The blue color indicated the nuclei of TUNEL positive cells. Quantification of TUNEL positive cells were presented as the percentage of TUNEL positive nuclei in total nuclei.

Figure 2

The effect of Wnt3a and Sfrp2 on beta-catenin/TCF transcriptional activities. (A) Nuclear beta-catenin levels. H9C2 cells were treated with Wnt3a alone or Wnt3a in combination of increasing dose of Sfrp2 during 24-hour hypoxia and 24-hour reoxygenation. Then nuclear beta-catenin was isolated with FractionPREP Cell Fractionation System, and the protein samples were separated by SDS-PAGE. Protein loading was shown by CBB staining. (B) Beta-catenin/TCF transcriptional activities. The cells were transfected by Top or Fop Flash with pRL-CMV renilla. 24 hours later, transfected cells were treated by Wnt3a with addition of Sfrp2 or not. They were then exposed to 12-hour hypoxia and 12-hour reoxygenation, following by dual-luciferase reporter assay. The data are expressed as mean ± SD.

Figure 3

The ELISA binding of Sfrp2 with Wnt3a. (A) ELISA wells were coated with different doses (from 3nM to 30nM) of Sfrp2 or BSA alone and incubated with 15nM of Wnt3a. The bound Wnt3a protein was detected with anti-Wnt3a and secondary immune reagents as described in Materials and Methods. (B) Sfrp2 (from 0.6nM to 3.6nM) were coated on 96-well Falcon ELISA plates and incubated with 3nM of Wnt3a. Absorbance at 450nm was measured with an ELISA plate reader. (C) ELISA plates were pre-incubated with 3~18nM of Sfrp3, a member of Sfrps, and then 15nM of Wnt3a was added on Sfrp3. In contrast, no increase in absorbance was observed with any concentration and the readings are comparable to basal BSA levels. Each panel is the representative of three experiments. (D) In the immunoprecipitation, the binding proteins were precipitated by Wnt3a antibody and were detected by Sfrp2 antibody in SDS page.

Figure 4

The effect of DKK1 on caspases and transcriptional activities. (A) caspase activities. Wnt3a or DKK1 or Wnt3a with increasing concentration of DKK1 was added to H9C2 cells, which then exposed to 24-hour hypoxia and 24-hour reoxygenation. Dose-response inhibitory effect of DKK1 was observed. All data are represented as means ± SD. (B) transcriptional activities. Co-transfection of Top or Fop Flash with pRL-CMV renilla was performed in H9C2 cells, following by 12-hour hypoxia and 12-hour reoxygenation with Wnt3a or DKK1 or Wnt3a plus indicated amounts of DKK1. Dkk1-mediated repression of TCF activity was shown dose dependently and dramatically. The data are expressed as mean ± SD.

Figure 5

The effect of Wnt3a on apoptosis in beta-catenin knock down cells. (A) caspase activities and beta-catenin. H9C2 cells were electroporated by beta-catenin siRNA and incubated for 24 hours following by 24-hour hypoxia and 24-hour reoxygenation with or without Wnt3a. Cells were harvested for caspase and Western assays. Protein loading was shown by CBB staining. The results are expressed as mean ± SD. (B) TUNEL staining and cell images. Beta-catenin siRNA was transfected into H9C2 cells with nucleofector technology following by hypoxia (24 hours) and reoxygenation (24 hours). In TUNEL staining, the blue color indicated the nuclei of TUNEL positive cells. Quantification of TUNEL positive cells were presented as the percentage of TUNEL positive nuclei in total nuclei. In the pictures of cell image, the survival of cells is indicated by increasing of cell number and spread morphology.