Fig 1: WNT5A secreted by HbChP controls morphogenesis of embryonic cerebellum. a In situ analysis of Wnt5a and Wls expression in the dorsal neuroepithelium adjacent to HbChP at E13.5 shows transcripts restriction to the HbChP epithelium, n = 4. Lower panel highlights absence of Wnt5a and low Wls expression in cerebellar ventricular zone (cVZ, empty arrowheads) compared to the HbChP (arrowheads). Scale bar: top 50 µm, bottom 10 µm. b WNT5A observed in the apical cVZ at E14.5 (arrowheads) is absent in Wnt5aKO embryos (empty arrowheads), n = 3. Scale bar: top 5 µm, inset 2 µm. c WNT5A and APOE (arrowheads) or APOE with APOJ (empty arrowhead) immunostaining in the apical cVZ (dotted line) at E14.5, n = 4. Scale bar: top 5 µm, bottom 2 µm. (d) Scheme of analyzed cerebellar parameters: total area (dash line), width (latero-medial, L-M, horizontal arrows) and length (dorso-ventral, D-V, vertical arrows). Scale bar: 200 µm. e Coronal sections of cerebellum in Wnt5aWT and Wnt5aKO embryos. Scale bar: 200 µm. f tdTomato staining in E14.5 FoxJ-CreERT2 embryo demonstrates recombination in HbChP epithelium, which is absent in the cVZ (inset, empty arrowhead) compared to HbChP (arrowheads), n = 3. Scale bar: top 50 µm, inset 20 µm. g Coronal sections of cerebellum in Wnt5aWT vs Wnt5acKO embryos. Scale bar: 200 µm. h–k Analysis of the h total area, i length, j width j, and k width/length ratio of cerebellum in Wnt5aWT, Wnt5aKO and Wnt5acKO embryos. Graphs show individual data points (dots) from n = 3 biologically independent samples; error bars represent mean ± s.d.; P-values (two-tailed Student’s t-test with unequal variance): * P < 0.05, ** P < 0.01, **** P < 0.0001. Corresponding P-values for differences between Wnt5aWT (WT), Wnt5aKO (KO) and Wnt5acKO (cKO)—h: WT vs. KO: 0.0004, WT vs. cKO: 0.0001; i WT vs KO: 0.0353, WT vs cKO: 0.0098; j WT vs. KO or cKO: – not significant; k WT vs KO: 0.016, WT vs cKO: 0.0115. l Schematic depiction of the model for WNT5A secretion by HbChP and its transventricular delivery to recipient regions. Source data are provided as a Source Data file
Fig 2: WNT5A is secreted and associated with lipoproteins in HbChP epithelium. a Schematic representation of the differential centrifugation protocol used for the isolation of exosomes and enrichment of lipoproteins from primary culture supernatants. b Western blot analysis of HbChP epithelial primary culture CM (Input), lipoprotein (100gSN) and exosomal (Exo) fractions, n = 3. Exosomal markers: FLOT2, HSP70, TSG101, CD63; lipoproteins: APOE and APOJ; negative control: Golgin97 (marker of Golgi) and epithelial markers: Claudin-1 and AQP1. c, d Immunofluorescent analysis of co-localization between markers of exosomes - CD63 and TSG101 c or various apolipoproteins d with WNT5A in E14.5 in HbChP epithelium, n = 3. WNT5A puncta only poorly overlap with exosomal markers CD63 and TSG101 (arrowheads) but show significant degree of overlap (arrowheads) with APOA1, APOB, APOE and APOJ. Scale bar: 5 µm. e Quantification of co-localization between WNT5A puncta and apolipoproteins (APOA1, APOB, APOE, APOJ) or exosomal markers (TSG101 and CD63). Graph shows n = 3 biologically independent samples. 3 representative images from 3 consecutive sections – 1 image per section - have been analyzed for each combination of markers and graphs show mean ± s.d. Quantification has been performed by IMARIS software and differences analysed by two-tailed Student’s t-test with unequal variance (*P < 0.05; **P < 0.01; ****P < 0.0001). WNT5A+/APOs+ puncta vs WNT5A+/CD63+ and WNT5A+/TSG101+, respectively: APOA1+ P = 0.0076 and P = 0.009; APOE+ P = 0.0014 and P = 0.009; APOB+ P = 0.0048 and P = 0.0094, APOJ+ P = < 0.0001 and P = < 0.0001. f Quantification of co-localization between puncta double-positive for WNT5A and APOJ which are also positive for APOA1, APOB and APOE, respectively. g Representative image directly illustrating the extent of immunofluorescent signal overlap between WNT5A and either APOJ as a marker of lipoproteins and CD63 as a marker of exosomes, n = 3. Scale bar: 5 µm. Source data are provided as a Source Data file
Fig 3: WNT5A is present in lipoprotein complexes. a CM from HbChP primary culture was subjected to KBr gradient-based ultracentrifugation and different isolated fractions (VLDL, LDL, and HDL) were analysed for the presence of WNT5A, APOE and APOJ by western blot, n = 3. (b–e) WNT5A-V5 and HA-tagged APOJ (b, d) or APOE (c, e) were overexpressed in HEK293T cells. WNT5A-V5 (b, d) and APOJ/E (c, e) were immunoprecipitated using anti-V5, anti-HA antibody and control IgG, respectively, and detected by western blotting, n = 3. Input is the loading control. Asterisk indicates non-specific immunoglobulin light chain. f Recombinant WNT5A (rcWNT5A) interacts with isolated APOJ protein. WNT5A and APOJ co-immunoprecipitated together. IgG served as a control, n = 3. Input represents the initial mixture. g Schematic depiction of the experimental design for delipidation and rescue experiments. h Delipidated FBS (Delipid) prevents production of WNT5A that is restored by relipidation (Relipid). Presence of WNT5A in cell lysates and CM from primary HbChP cultures has been analyzed using western blot, n = 3. Loading control: ß-actin. i WNT5A secretion was restored after the addition of different mouse lipoprotein fractions to primary HbChP cultures cultivated in presence of delipidated conditions. j Recombinant rcWNT5A was incubated with either 0.6% CHAPS or human LDL (hLDL)- lipoprotein fraction for 4 h at 37 °C and filtered through 150 kDa cut-off protein concentrator. Western blot confirmed presence of WNT5A only in >150 kDa fraction but not in the <150 kDa fraction, n = 3. k Separation of >150 kDa fraction into VLDL, LDL, and HDL fractions using KBr gradient confirmed presence of rcWNT5A in LDL and its co-fractionation with APOB and APOE specific for LDL fraction, n = 3. l Only >150 kDa fraction of LDL/rcWNT5A mixture can trigger DVL3 phosphorylation (indicated by arrowhead) in MEF cells, n = 3. Negative control: 10% FBS DMEM; positive control: rcWNT5A; loading control: ß-actin. m Statistical analysis of DVL3 shift assay. Graph shows n = 3 biologically independent samples; error bars represent mean ± s.d.; P-values (two-tailed Student’s t-test with unequal variance): * P < 0.05. Control CM vs >150 kDa hLDL + rcWNT5A: P = 0.0231. Biological replicates are indicated in the graph. Source data are provided as a Source Data
from BioVendor Laboratory Medicine, Inc. for Clusterin Human NATIVE, Human Serum