hiPSC-derived neural stem cells from patients with schizophrenia induce an impaired angiogenesis

Abstract

Schizophrenia is a neurodevelopmental disease characterized by cerebral connectivity impairment and loss of gray matter. It was described in adult schizophrenia patients (SZP) that concentration of VEGFA, a master angiogenic factor, is decreased. Recent evidence suggests cerebral hypoperfusion related to a dysfunctional Blood Brain Barrier (BBB) in SZP. Since neurogenesis and blood-vessel formation occur in a coincident and coordinated fashion, a defect in neurovascular development could result in increased vascular permeability and, therefore, in poor functionality of the SZP's neurons. Here, we characterized the conditioned media (CM) of human induced Pluripotent Stem Cells (hiPSC)-derived Neural Stem Cells of SZP (SZP NSC) versus healthy subjects (Ctrl NSC), and its impact on angiogenesis. Our results reveal that SZP NSC have an imbalance in the secretion and expression of several angiogenic factors, among them non-canonical neuro-angiogenic guidance factors. SZP NSC migrated less and their CM was less effective in inducing migration and angiogenesis both in vitro and in vivo. Since SZP originates during embryonic brain development, our findings suggest a defective crosstalk between NSC and endothelial cells (EC) during the formation of the neuro-angiogenic niche.

Fig. 1. Profiling of NSC derived hiPSC obtained from Ctrl and SZP.

Immunostaining for Ctrl (a, c, and e) and SZP (b, d, and f) NSC specific markers; (a and b) Nestin (green) and nuclear marker DAPI (blue), (c and d) Pax6 (red) and corresponding nuclear marker DAPI (e and f); Scale bar = 100 um. Representative images from N = 3 (3 Ctrl NSC and 3 SZP NSC). g Percentage of positive cells, with 98% (±0.003) and 97% (±0.01) Nestin positive cells for Ctrl and SZP NSCs respectively; 99% (±0.002) and 99.7% (±0.002) Pax6 positive cells for Ctrl and SZP NSCs respectively. No Oct4 positive cells were detected in NSC of Ctrl and SZP. Data is shown as mean ± SD from N = 3. h Cell division was monitored for 12 h and no significant differences in proliferation were found between Ctrl and SZP NSC. Data for each cell line (Ctrl #1, #2, #3 and SZP #1, #2, #3) is shown in a correlative order control and graphed as mean ± SD. i, j Morphological parameters such as cell area (i) and cell roundness (j) did not show detectable differences among Ctrl and SZP groups. Data for each cell line (Ctrl #1, #2, #3 and SZP #1, #2, #3) is shown in a correlative order control and graphed as mean ± SD

Fig. 2. SZP NSC present an imbalance in neuro-angiogenic factor secretion.

a Representative pictures of angiogenic profile secretome from Ctrl NSC and SZP NSC. Red boxes indicate positive internal control spots; blue boxes indicate negative controls. b Quantification of angiogenic protein levels of (a) 3 Ctrl NSC CM versus 3 SZP NSC CM. Data are shown as level of expression relative to internal control and graphed as mean ± SD; *p < 0.05 according to Mann–Whitney test. c Heatmap depicting the levels of angiogenic proteins present in 3 Ctrl NSC CM (#1, #2 and #3), 3 SZP NSC CM (#1, #2 and #3), 3 Ctrl Nsp CM (#1, #2 and #3) and 3 SZP Nsp CM (#1, #2 and #3). Data are shown as average level relative to internal control. d–g Quantification of IGBP-2 (d), VEGFA (e), TIMP-1 (f), and NRG1-B1 (g), in each of the CM. Data for each cell line (Ctrl #1, #2, #3 and SZP #1, #2, #3) is shown in a correlative order. Values are shown as the level of protein expression relative to internal control and graphed as mean ± SD; *p < 0.05, **p ≤ 0.01 according to Mann–Whitney test. h–m Analysis of non-canonic neuroangiogenic factors expression by qPCR in 3 Ctrl NSC and 3 SZP NSC. Data for each cell line (Ctrl #1, #2, #3 and SZP #1, #2, #3) is shown in a correlative order. Data are shown as mean ± SD; *p < 0.05 according to Mann–Whitney test. (n) Expression of SEMA3A was evaluated in 3 Ctrl and 3 SZP NSC CM. Data for each cell line (Ctrl #1, #2, #3 and SZP #1, #2, #3) is shown in a correlative order. (o) Quantification of protein bands of (n) reveal increased SEMA3A expression in SZP NSC. Data are shown as mean ± SD; *p < 0.05 according to Mann–Whitney test

Fig. 3. SZP NSC CM induce less angiogenesis and migration in vitro than Ctrl NSC CM.

a Cartoon representing the experimental design of the tube formation assay. After 48 h, CM was collected from NSC cultures and applied on HUVEC seeded on matrigel coated wells. After 4 h, tubes (closed polygons) and sprouts were counted. b–d Representative images of tube formation assay when incubating HUVEC on Neural Expansion Media (NEM, used as negative control (b), Ctrl NSC CM (c) or SZP NSC CM (d); Scale bar = 30 µm. e–f Quantification of average number of sprouts (e) or tubes (f) formed in each condition. Data for each cell line (Ctrl #1,#2, #3 and SZP #1, #2, #3) is shown in a correlative order control and graphed as mean ± SD with *p < 0.05 according to Kruskal-Wallis test. g–n Ctrl NSC CM was incubated with 100 µg/ml of bevacizumab (Bvz) to inhibit VEGFA signaling. g–l Representative images of tube formation assay when incubating HUVEC on NEM (g), NEM plus 50 ng/ml VEGFA (h), NEM plus 50 ng/ml VEGFA with 100 µg/ml of bevacizumab (Bvz) inhibitor (i), Ctrl NSC CM (j), Ctrl NSC CM with 100 µg/ml Bvz (k) or SZP NSC CM (l). m–n Quantification of average number of sprouts (m) or tubes (n) formed in each condition. Data is shown as mean ± SD with *p < 0.05, **p ≤ 0.01 and ***p ≤ 0.001 according to Kruskal–Wallis test

Fig. 4. SZP NSC CM induce less angiogenesis in vivo than Ctrl NSC CM.

a To perform the in vivo angiogenesis assay in the CAM of chicken embryos we added NSC 48 h CM on top of a bio cellulose scaffold. Scaffold was positioned over the CAM at embryonic day 8 (E8). After 4 days, vessels in the perimeter of the scaffold were counted. b Representative images of CAM vessels as indicated. Scaffold was filled with NEM as negative control, Ctrl NSC CM and SZP NSC CM. Scale bar = 1 cm. c Quantification of number of vessels in each condition at E12, represented as fold change in vessel number compared to E8. Data for each cell line (Ctrl #1, #2, #3 and SZP #1, #2, #3) is shown in a correlative order control and graphed as mean ± SD; * p < 0.05 and ***p ≤ 0.001 according to Kruskal–Wallis test. d, e Ctrl NSC CM was incubated with 100 µg/ml of bevacizumab (Bvz) to inhibit VEGF-A signaling. d Representative images of CAM vessels as indicated. Scaffold was filled with NEM, NEM plus 50 ng/ml VEGFA, NEM plus 50 ng/ml VEGFA with 100 µg/ml Bvz, Ctrl NSC CM, Ctrl NSC CM with 100 µg/ml Bvz or SZP NSC CM. e Quantification of number of vessels in each condition at E12, represented as fold change in vessel number compared to E8. Data is shown as mean ± SD; *p < 0.05 and ***p ≤ 0.001 according to Kruskal–Wallis test