Divergent effects of Porcupine and Wntless on WNT1 trafficking, secretion, and signaling

Abstract

Loss-of-function studies have identified Porcupine (PORCN) and Wntless (WLS) as essential mediators of Wnt secretion and signaling. Whereas PORCN is thought to palmitoylate Wnt proteins, WLS is believed to transport palmitoylated Wnt proteins to the cell surface. However, little is known about how these two proteins cooperate to regulate Wnt palmitoylation, trafficking, secretion, and signaling. We first investigated possible interactions between PORCN, WLS, and WNT1, by carrying out co-immunoprecipitation studies. These studies demonstrate the existence of a complex containing PORCN and WLS. They further show that PORCN and WLS compete for binding to WNT1. Then, we used gain-of-function studies to investigate the cooperation between PORCN and WLS as well as possible biochemical interactions between PORCN, WLS, and WNT1. Consistent with the proposed roles for PORCN and WLS, we show that overexpression of PORCN promotes palmitoylation of WNT1 while overexpression of WLS does not. Overexpression of PORCN enhances the ability of WLS to promote WNT1 trafficking to the cell surface as well as secretion, but decreases the ability of WLS to activate WNT1 signaling in target cell. These observations suggest that the levels of WNT1 on the cell surface and in the media are not the sole determinants of the activation of Wnt signaling in target cells.

Keywords: Palmitoylation; Porcupine; Secretion; WNT1; Wntless; β-catenin dependent signaling.

FIGURE 1

WNT1, PORCN and WLS co-immunoprecipitate with one another. HEK293T cells were transiently transfected with constructs encoding GFP, WNT1, PORCN, and/or WLS as indicated. The amount of total DNA was held constant using pcDNA3.1.GFP as a filler. Cell lysates were immunoprecipitated (IP) with anti-HA (PORCN) in A and D or anti-H6 (WLS) in B, C, and E and subjected to SDS-PAGE. Gels were immunoblotted (WB) with A) anti-HA (PORCN), B) anti-H6 (WLS), C) anti-GA (PORCN), or D–E) anti-WNT1. Secondary antibody conjugated with Alexa Fluor® 680 was used for detection with an Odyssey CLX infrared scanner. Proteins present in total lysates and on the bead are shown. The images shown are representative of those from 4 independent experiments.

FIGURE 2

PORCN and WLS bind to each other and compete for binding to WNT1. Quantification of the results from Figure 1 is shown. Each data point represents three independent replicates. Error bars show +/− standard error.

FIGURE 3

PORCN, but not WLS, promotes the palmitoylation of WNT1. A) PORCN and WLS increase the apparent hydrophobicity of WNT1 in a TX-114 phase separation assay. HEK293T cells were transiently transfected with WNT1 along with GFP, PORCN or WLS. Cell lysates were subjected to extraction in TX-114 as described in Experimental Procedures. Aqueous (Aq) and TX-114 (TX) fractions were subjected to SDS-PAGE and immunoblotted with antibodies against WNT1. This experiment was carried out in triplicate. B) Only PORCN promotes the palmitoylation of WNT1. HEK293T cells were transiently transfected with WNT1:F_c along with GFP (control), PORCN, or WLS. Cells were metabolically labeled with DMSO (D, control) or Alkyne-palmitate (A). Wnt1:F_c was precipitated from cell lysates using A/G Agarose beads and subjected to click chemistry to covalently attach biotin azide to the alkyne palmitate. Proteins were separated by SDS-PAGE and analyzed by probing immunoblots with anti-WNT1 followed by an Alexa Fluor 680 labeled secondary antibody and Streptavidin IR 800 dye. C) Fluorescence was quantitated using a Licor Odyssey CLX infrared scanner. Data points reflect the average of 3 independent replicates. Errors bars represent +/− standard error. The Alk-C16 data with PORCN data for each independent replicate were normalized to 1. Hence, there are no error bars for that data point.

FIGURE 4

PORCN and WLS promote the association of WNT1 with detergent resistant membranes. HEK293T cells were transiently transfected with WNT1 along with GFP, PORCN or WLS. Cell lysates were separated by density gradient centrifugation. Fractions 1–10 were collected from top to bottom and then separated by SDS-PAGE. Gels were immunoblotted and analyzed with PCNA (non-raft associated nuclear protein), Flotillin1 (lipid raft control) and anti-WNT1 antibodies. Arrows mark the presence of WNT1 in lipid raft fractions. NIH Image J was used for quantification of the results. Data points reflect the average of 3–6 independent replicates. Error bars represent +/− standard error.

FIGURE 5

PORCN and WLS are localized to lipid rafts. HEK293T cells were transiently transfected with PORCN (or HA-PORCN) or WLS-H6. Cell lysates were separated by density gradient centrifugation in Optiprep density gradient medium. Fractions 1–10 were collected from top to bottom and then separated by SDS-PAGE. Gels were immunoblotted and analyzed with PCNA (non-raft associated nuclear protein), anti-Flotillin1 (lipid raft control) as well as anti-PORCN, anti-HA, or anti-H6. Asterisks (*) demarcate non-specific bands. NIH Image J was used for quantification of the results. Data points reflect the average of 2–3 independent replicates. Error bars represent +/− standard error.

FIGURE 6

PORCN and WLS cooperate to regulate the trafficking of WNT1 to the cell surface. HEK293T cells were transiently transfected with the indicated constructs. Cells were incubated in the presence (bottom) or absence (top) of sulfo-NHS-LC-biotin. After quenching the reaction, cells were lysed and cell surface proteins were precipitated with NeutrAvidin Agarose. Proteins were separated by SDS-PAGE and analyzed via Western blot with antibodies against β-tubulin (cytoplasmic control) and WNT1. This experiment was carried out 4 times. Representative results are shown.

FIGURE 7

PORCN and WLS are present on the cell surface. HEK293T cells were transiently transfected with HA-PORCN, WLS-MYC/H6 or WLS-MYC/H6 Y425A. Proteins on the cell surface were covalently labeled with Sulfo-NHS-SS-Biotin. After cell lysis, cell surface proteins were precipitated using NeutrAvidin Agarose. A) The amount of PORCN, WLS and WLS Y425A in the original lysates and on the bead was visualized by immunoblotting with anti-HA (PORCN) or anti-H6 (WLS and WLS Y425A). B) A duplicate blot with lysates was probed with anti-β-tubilin and Streptavidin-IRdye800 as loading and labeling controls, respectively. C) The levels of cell surface PORCN (bead) as compared to the total levels of PORCN (lysate) are shown in a graphical format. The data for PORCN+GFP in the presence of Sulfo-NHS-SS-Biotin was normalized to a value of 1. D) The levels of cell surface WLS or WLS Y425A (bead) as compared to the total levels of WLS or WLS Y425A (lysate) are shown in a graphical format. The data for WLS+GFP in the presence of Sulfo-NHS-SS-biotin was set to a value of 1. Values in C–D represent the average of 2 independent replicates. Error bars reflect +/− standard error. E) Additional controls show that β-tubulin (a cytoplasmic protein) and spGFP-KDEL (an ER resident protein) are not detected on the cell surface using this method.

FIGURE 8

PORCN and WLS cooperate to promote WNT1 secretion from HEK293T cells. HEK293T cells were transiently transfected with the constructs indicated. Media was collected and WNT1 was precipitated using Sepharose blue beads. A) Proteins in the (partially purified) media and cell lysates were subjected to SDS-PAGE and immunoblotted with antibodies against WNT1 and β-tubulin (as a loading control). Western blots were quantified using NIH Image J. B) This graph shows the relative levels of WNT1 protein in the media/total WNT1 protein (media + cell lysates). C) This graph shows the relative levels of WNT1 protein in cell lysates/total WNT1 protein (media + cell lysates). Samples transfected with WNT1 and GFP were further normalized to a value of 1. Data points reflect the average of 6 independent replicates. Error bars represent +/− standard error.

FIGURE 9

Autocrine 8xSuperTopFlash assay reveals the effects of PORCN and WLS on WNT1 signaling. The autocrine SuperTopFlash assay was carried out as described in Experimental Procedures. HEK293T cells were transfected with GFP along with the indicated plasmid. GFP levels were varied such that total DNA levels were held constant. Data from cells transfected with GFP and WNT1 were normalized to a value of 1. Each data point represents the average of 12–15 independent replicates carried out on 3 different days. Errors bars indicate +/− standard error.

FIGURE 10

Paracrine 8xSuperTopFlash assay reveals the effects of PORCN and WLS on WNT1 and WNT3A signaling. The paracrine SuperTopFlash assay was carried out as described in Experimental Procedures. Cells were transfected with GFP along with the indicated plasmid. GFP levels were varied such that total DNA levels were held constant. For “high concentration” transfections, 183 ng of the Wnt expression construct was transfected into cells seeded in 24 well plates. For “low concentration” transfection, 10 ng of the Wnt expression construct was used in each well. The transfected HEK293T cells were mixed with LSL reporter cells before measuring luciferase activity. Data from HEK293T cells transfected with GFP and WNT1 (or WNT3A) were normalized to a value of 1. Each data point represents the average of 12–15 independent replicates carried out on 3 different days. Errors bars indicate +/− standard error.