Fig 2: Glypican-2 binds FGF2 and modulates FGF2-mediated progenitor cell proliferation.(A) Experimental work flow for immunocompetitive capture mass spectrometry (ICC-MS). (B) ICC-MS displacement profile of neural progenitor cells cultured for 21 days under differentiation conditions. FGF2 and GSLG1 show a statistically significant concentration-dependent decrease in signal. (C) Eluted fractions of the Gpc2 immunocompetitive precipitations. Cell lysates were pre-incubated with 0, 1, 2.5, 5 and 10 µg/ml of free Gpc2 antibody before immunoprecipitation with the Gpc2-specific antibody (clone 2/43). The detection of the FGF2 protein decreased as the concentration of free competitor antibody increased, similar to Gpc2, thereby confirming the specificity of the FGF2 co-immunoprecipitation. (D) FGF2 is detected in the bound fraction of the FGF2 immunoprecipitation. Gpc2 is co-captured with FGF2. An IP of a lysate pre-incubated with 10 µg/ml of free Gpc2 antibody is also shown as a control. Input: cell lysate. (E) Increasing amounts of FGF2 promote progenitor cell proliferation. Recombinant Gpc2 counteracts the proliferative effect of FGF2. (F) Representative images of proliferating neural precursor cells (arrows) cultured in the presence of increasing amounts of Gpc2. Error bars represent the SEM. Scale bars = 10 µm. p < 0.05 = *; p < 0.01 = **; p < 0.001 = ***.

Fig 3: Linear relation between log concentration of DCX and GPC2.The graph shows the linear mixed-effect regression model (blue line) and 95% credible interval (grey area). Blue dots represent values over, and red dots below the LOD for DCX. As in Figs 1 and 3, DCX values below the LOD are displayed as the mean LOD of the DCX assays.

Fig 4: Glypican-2 expression is restricted to the neurogenic regions in the adult rodent brain. Genetic ablation of neurogenesis results in a strong reduction of Glypican-2 expression.(A) Expression of Gpc2 mRNA in the micro-dissected dentate gyrus (DG) and residual hippocampus (RHp) was compared to expression in the cortex (Ctx) of adult mice. Gpc2 is expressed at the highest levels in regions with continuous adult neurogenesis.(B–D) In situ hybridization of Gpc2 RNA in the adult mouse brain (w7). Gpc2 expression is restricted to the subgranular zone of the DG (B), the lateral ventricles of the SVZ (C) and the rostral migratory stream entering the OB (D). (E) Experimental design for inducing a loss of neurogenesis by Hes5::CreERT2-mediated ablation of Rbpjfl/fl. (F–F’) Representative Dcx immunostaining in Rbpj+/+ Ctrl animals (F) and Rbpj-/- mutants (F’). (G–G’) Gpc2 in situ hybridization in Rbpj+/+ Ctrl animals (G) and Rbpj-/- mutants (G’). (H) Quantification of Dcx immunoreactivity. The ablation of Rbpj results in a reduction in Dcx immunoreactivity in vivo. (I) Relative expression of Dcx and Gpc2 mRNAs in micro-dissected hippocampi from the Rbpj-/- mutants. (J) CSF levels of Gpc2 protein. The ablation of Rbpj results in a significant decrease in the levels of Gpc2 in mouse CSF. Ctx, cortex; RHp, residual hippocampus; DG, dentate gyrus; SVZ, subventricular zone; OB, olfactory bulb. The error bars shown in J represent the SEM. The error bars shown in A, H and I represent CI. Scale bar = 200 µm. p < 0.05 = *; p < 0.01 = **; p < 0.001 = ***.

Fig 5: Gpc2 is present in human CSF and shows a moderate decrease in a aging longitudinal cohort.(A) Gpc2 protein was detected in human CSF samples grouped by age. Each dot represents an individual human donor. (B) Longitudinal analysis of the CSF levels of Gpc2 in individual samples. Baseline measurements were compared to a 3-year follow-up measurement. Error bars shown in C and D represent the SEM. Error bars shown in A and B represent CI. p < 0.05 = *; p < 0.01 = **; p < 0.001 = ***.