Fig 2: Increased EC migration in HUVEC cultures exposed to CM of NEO1 KO astrocytes and hepcidin expression needs NEO1.A–C Increased EC proliferation in HUVEC cocultured with CM of NEO1-KO astrocytes. EDU incubated for 4 h (n = 6 experiments of astrocytes from NEO1 GFAP-Cre and its control group). D–F Increased EC migration in HUVEC cocultured with CM of NEO1-KO astrocytes (n = 6 experiments of astrocytes from NEO1GFAP-Cre and its control group). G The permeability of HUVEC monolayers (n = 3 experiments of astrocytes from NEO1GFAP-Cre and its control group). H–J Transwell assay to access HUVECs migration. K–Q Representative immunoblotting (K–M) and quantification (L, M, O, and Q) of FTH, FTL, FPN1, and hepcidin of cortex endothelial cells from mice subjected to SAH for different time points and sham surgery (n = 6 mice/group). PECAM-1 was used as a loading control. R The Quantification analysis of hepcidin in CM of primary astrocytes and primary neurons by ELISA assay (n = 6 experiments/group). S, T FISH analysis of hepcidin mRNA in primary astrocytes and primary neurons (n = 6 experiments/group). Scale bars: 20 μm. U Quantification analysis of hepcidin in CM of astrocytes from NEO1GFAP-Cre and its control group by ELISA assay (n = 6 experiments/group). V, W Representative immunoblotting (V) and quantification (W) of hepcidin in astrocytes from NEO1GFAP-Cre and its control group (n = 6 experiments of astrocytes from NEO1GFAP-Cre and its control group). Data are presented as mean ± SD. *P < 0.05; **P < 0.01, ***P < 0.01. 1-way ANOVA with Tukey’s correction for multiple comparisons (B, C, E, and G). Student’s t-test.

Fig 3: NEO1 is decreased in the CSF of SAH patients and increased blood vessel density but increased BBB leakage in the SAH mouse cortex.A Heatmap reveals the differentially expressed proteins from the comparisons of SAH (day 1–3) vs. control and SAH (day 4–7) vs. control, the expression of which was normalized, the high expression displayed in red and low in blue. B Quantification of NEO1 expression in CSF of control, SAH patients with Fisher1-2, and SAH patients with Fisher3-4. C Quantification of NEO1 expression in CSF of control, SAH patients with mRS ≤2, and SAH patients with mRS >2. D Analysis of the correlation between mRS in 6 months of SAH patients and the expression of NEO1 in CSF of SAH patients at the initial 7 days were performed with Pearson’s correlation test. E, F Representative images of EB (red) and dextran-10 kDa (green) coimmunostained with PECAM-1 antibody in sham and SAH cortex. G, H Quantitative analysis of data in (A) and (B) (n = 6 mice/group). I Representative images of dextran-3 kDa coimmunostained with PECAM-1 antibody in sham and SAH cortex. J, K Quantitative analysis of data in (C) (n = 5 mice/group). Scale bars: 20 μm. Data are presented as mean ± SD and were analyzed by student’s t-test, *P < 0.05; **P < 0.01; ns not significant.

Fig 4: Hepcidin alleviation of EC iron storage decreases, proliferation, migration, and leakage induced by CM of astrocytes from NEO1GFAP-Cre.A–C Representative immunoblotting (A) and quantification (B, C) of hepcidin effect on FTH and FTL of MBMEC in the presence of CM of astrocytes. Data are presented as mean ± SD (n = 6 experiments/group). D–F Hepcidin (800 nM) inhibition of MBMEC proliferation. D Schematic of hepcidin administration in MBMEC cultures in the presence of CM of NEO1 KO astrocytes. EDU incubated for 4 h. E Representative images of EDU + MBMEC. F Quantitative analysis of data in (E). Data are presented as mean ± SD (n = 6 experiments/group). G–I Hepcidin inhibition of MBMEC migration. G, H Schematic of hepcidin administration in MBMEC cultures in the presence of CM of NEO1 KO astrocytes to access migration by wound-healing assay. I Representative images of MBMEC migration 24 h after wound scratching. J Quantification analysis of data in (H). Data are presented as mean ± SD (n = 6 experiments/group). K Representative images of caveolae of MBMEC. L Hepcidin inhibition of permeability of MBMEC monolayers. Data are presented as mean ± SD (n = 3 experiments/group). *P < 0.05; **P < 0.01; ns not significant. 1-way ANOVA with Tukey’s correction.