DeSUMOylation of IGF2BP2 Promotes Neuronal Differentiation of OM-MSCs by Stabilizing SOX11 to Ameliorate Brain Injury After Intracerebral Hemorrhage

Abstract

Background: Our previous study demonstrated that olfactory mucosa mesenchymal stem cell (OM-MSC) neuronal differentiation can reduce neural damage following intracerebral hemorrhage (ICH). However, the mechanisms that regulate OM-MSC neuronal differentiation to mitigate ICH-induced brain injury remain unclear.

Methods: The ICH model was established through autologous blood injection to evaluate the impact of OM-MSCs on brain injury, using the mNSS method, TUNEL, and Nissl staining, as well as the Western blot assay. qPCR, Western blotting, flow cytometry, and immunofluorescence assays were employed to assess the neuronal differentiation of OM-MSCs. The SUMOylation assay was conducted to investigate the relationship between IGF2BP2 and SENP1. RIP, RNA pull-down, and mRNA stability assays were performed to analyze the molecular interaction network involving SENP1, IGF2BP2, and SOX11.

Results: IGF2BP2 enhanced the protective effects of OM-MSCs against ICH-induced brain injury, as demonstrated by a significant reduction in brain edema, mNSS scores, and apoptosis, along with improved neuronal survival. Furthermore, the overexpression of IGF2BP2 increased the expression of Tuj-1, MAP2, NF200, and NeuN, while decreasing GFAP and ALDH1L1 levels, suggesting the stimulatory effects of IGF2BP2 on the neuronal differentiation of OM-MSCs. Mechanistically, SENP1 enhanced IGF2BP2 expression through SUMO1-induced IGF2BP2 SUMOylation. Additionally, IGF2BP2 functioned as an RNA-binding protein for SOX11, thereby increasing SOX11 levels. The depletion of IGF2BP2 negated the SENP1-induced neuronal differentiation of OM-MSCs. The overexpression of SOX11 mitigated the inhibitory effects of IGF2BP2 silencing on OM-MSC neuronal differentiation.

Conclusion: The SENP1/IGF2BP2/SOX11 axis played a crucial role in the neuronal differentiation of OM-MSCs and ameliorated brain damage caused by ICH.

Keywords: ICH; IGF2BP2; OM‐MSCs; SENP1; SOX11; neuronal differentiation.

FIGURE 1

Identification of the isolated OM‐MSCs. (A) OM‐MSCs mainly exhibited spindle‐shaped and a radial arrangement under a light microscope. Scale bar, 50 μm. (B) Specific markers STRO‐1 and Nestin of OM‐MSCs were identified by immunofluorescence. Scale bar, 50 μm. (C) Surface markers (CD34, CD44, CD45, CD73, CD90, and CD105) of OM‐MSCs were measured by flow cytometry assay. n = 3.

FIGURE 2

IGF2BP2 expedited the protective effect of OM‐MSCs against ICH‐caused brain injury. (A) Analysis of brain water content in brain region at 72 h after ICH, which was performed in mice in sham (76.25 ± 1.66), ICH (83.48 ± 1.49), ICH + OM‐MSCs (80.63 ± 1.42), ICH + OM‐MSCs oe‐NC (79.93 ± 1.71), and ICH + OM‐MSCs oe‐IGF2BP2 (76.92 ± 1.05) groups. n = 6. (B) mNSS scoring (ICH: 11.67 ± 2.16, ICH + OM‐MSCs: 9.00 ± 1.41, ICH + OM‐MSCs oe‐NC: 8.67 ± 1.12, ICH + OM‐MSCs oe‐IGF2BP2: 4.83 ± 1.84) of mice after 72 h of ICH modeling. n = 6. (C) Representative images of TUNEL in the perihematoma area of mice after 72 h of ICH (scale bar, 100 μm) and TUNEL positive cells (sham: 1.62 ± 0.55, ICH: 34.15 ± 1.16, ICH + OM‐MSCs: 15.63 ± 1.49, ICH + OM‐MSCs oe‐NC: 12.97 ± 2.09, ICH + OM‐MSCs oe‐IGF2BP2: 7.65 ± 0.93) were counted. (D) Nissl staining was conducted in the perihematoma area of mice after 72 h of ICH (cale bar, 100 μm) and the number of survival neurons (sham: 42.67 ± 4.73, ICH: 10.00 ± 1.00, ICH + OM‐MSCs: 18.67 ± 1.53, ICH + OM‐MSCs oe‐NC: 18.67 ± 2.08, ICH + OM‐MSCs oe‐IGF2BP2: 29.00 ± 4.36) was counted. (E) Bax (sham: 0.25 ± 0.08, ICH: 1.02 ± 0.09, ICH + OM‐MSCs: 0.81 ± 0.06, ICH + OM‐MSCs oe‐NC: 0.84 ± 0.05, ICH + OM‐MSCs oe‐IGF2BP2: 0.54 ± 0.07), cleaved Caspase3 (sham: 0.20 ± 0.05, ICH: 0.76 ± 0.06, ICH + OM‐MSCs: 0.61 ± 0.03, ICH + OM‐MSCs oe‐NC: 0.59 ± 0.03, ICH + OM‐MSCs oe‐IGF2BP2: 0.45 ± 0.06), and Bcl‐2 (sham: 0.87 ± 0.03, ICH: 0.39 ± 0.05, ICH + OM‐MSCs: 0.57 ± 0.06, ICH + OM‐MSCs oe‐NC: 0.56 ± 0.06, ICH + OM‐MSCs oe‐IGF2BP2: 0.69 ± 0.03) protein levels were measured by western blot assay in the above groups. (F) Immunofluorescence was used to observe the number of NeuN‐positive cells (sham: 65.54 ± 5.84, ICH: 10.11 ± 2.16, ICH + OM‐MSCs: 41.70 ± 3.53, ICH + OM‐MSCs oe‐NC: 41.23 ± 3.12, ICH + OM‐MSCs oe‐IGF2BP2: 54.90 ± 3.32) in the perihematoma area of mice after 72 h of ICH. Scale bar, 100 μm. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3

IGF2BP2 facilitated neuronal differentiation of OM‐MSCs. (A) OM‐MSCs were infected with adenoviral vectors (oe‐IGF2BP2) at the MOIs of 25–200 (control: 1.00 ± 0.04, oe‐NC: 1.07 ± 0.10, oe‐IGF2BP2–25: 1.59 ± 0.14, oe‐IGF2BP2–50: 2.47 ± 0.25, oe‐IGF2BP2–100: 3.63 ± 0.16, oe‐IGF2BP2–200: 2.69 ± 0.15) for 48 h and the mRNA expression of IGF2BP2 was detected using qPCR in control (1.00 ± 0.11), oe‐NC (1.09 ± 0.18), and oe‐IGF2BP2 (3.70 ± 0.19) groups. (B) OM‐MSCs were infected with adenoviral vectors at an optimal multiplicity of 100 for 48 h and IGF2BP2 protein levels were measured by western bolt assays in oe‐NC (0.47 ± 0.05) and oe‐IGF2BP2 (0.90 ± 0.05) groups. (C) OM‐MSCs were infected with Ad and incubated with differentiation medium for four days. The mRNA levels of Tuj‐1 (oe‐NC: 1.00 ± 0.08, oe‐IGF2BP2: 2.50 ± 0.18), MAP2 (oe‐NC: 1.00 ± 0.18, oe‐IGF2BP2: 3.31 ± 0.28), NF200 (oe‐NC: 1.00 ± 0.07, oe‐IGF2BP2: 3.75 ± 0.29), GFAP (oe‐NC: 1.00 ± 0.06, oe‐IGF2BP2: 0.33 ± 0.05), and ALDH1L1 (oe‐NC: 1.00 ± 0.12, oe‐IGF2BP2: 0.46 ± 0.06) was measured by qPCR assay. (D) Flow cytometry assay was used to detect Tuj‐1 positive cells (oe‐NC: 8.19 ± 1.33, oe‐IGF2BP2: 16.50 ± 1.15). (E) Immunofluorescence was used to observe the number of GFAP (oe‐NC: 28.55 ± 3.16, oe‐IGF2BP2: 12.14 ± 3.26) and NeuN (oe‐NC: 20.07 ± 2.27, oe‐IGF2BP2: 49.14 ± 7.57)‐positive cells. Scale bar, 20 μm. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4

SENP1 regulated the SUMOylation of IGF2BP2. (A) OM‐MSCs were infected with adenoviral vectors at an optimal multiplicity of 100 for 48 h. Western blot analysis of SENP1 (oe‐NC: 0.39 ± 0.04, oe‐SENP1: 0.87 ± 0.11, sh‐NC: 0.50 ± 0.08, sh‐SENP1: 0.20 ± 0.05) and IGF2BP2 (oe‐NC: 0.44 ± 0.03, oe‐SENP1: 0.77 ± 0.04, sh‐NC: 0.54 ± 0.07, sh‐SENP1: 0.31 ± 0.07) protein levels in OM‐MSCs treated with oe‐NC, oe‐SENP1, sh‐NC, or sh‐SENP1. (B) OM‐MSCs were infected with adenoviral vectors and IGF2BP2 SUMOylation was assessed by immunoprecipitation/western blot. Cell lysates were collected and subjected to immunoprecipitation using anti‐SUMO1 or anti‐IGF2BP2. Immunoblot analysis was then performed using the specified antibodies. n = 3. **p < 0.01, ***p < 0.001.

FIGURE 5

IGF2BP2 was implicated in SENP1‐mediated OM‐MSC neuronal differentiation. (A and B) qPCR (sh‐NC: 1.00 ± 0.13, sh‐IGF2BP2: 0.30 ± 0.05) and western blot (sh‐NC: 1.05 ± 0.06, sh‐IGF2BP2: 0.46 ± 0.07) analysis of IGF2BP2 mRNA and protein levels in OM‐MSCs treated with sh‐NC or sh‐IGF2BP2 for 48 h. (C) OM‐MSCs were infected with Ad and incubated with differentiation medium for four days. The mRNA levels of Tuj‐1 (oe‐NC: 1.00 ± 0.14, oe‐SENP1: 2.33 ± 0.27, oe‐SENP1 + sh‐NC: 2.27 ± 0.26, oe‐SENP1 + sh‐IGF2BP1: 1.40 ± 0.23), MAP2 (oe‐NC: 1.00 ± 0.12, oe‐SENP1: 2.19 ± 0.32, oe‐SENP1 + sh‐NC: 2.08 ± 0.32, oe‐SENP1 + sh‐IGF2BP1: 1.43 ± 0.07), NF200 (oe‐NC: 1.00 ± 0.13, oe‐SENP1: 3.92 ± 0.48, oe‐SENP1 + sh‐NC: 3.94 ± 0.69, oe‐SENP1 + sh‐IGF2BP1: 1.30 ± 0.16), GFAP (oe‐NC: 1.00 ± 0.07, oe‐SENP1: 0.70 ± 0.06, oe‐SENP1 + sh‐NC: 0.62 ± 0.08, oe‐SENP1 + sh‐IGF2BP1: 0.87 ± 0.04), and ALDH1L1 (oe‐NC: 1.00 ± 0.12, oe‐SENP1: 0.49 ± 0.07, oe‐SENP1 + sh‐NC: 0.47 ± 0.03, oe‐SENP1 + sh‐IGF2BP1: 0.90 ± 0.14) were measured by qPCR assay. (D) Flow cytometry assay was used to detect Tuj‐1 positive cells (oe‐NC: 7.05 ± 1.29, oe‐SENP1: 14.26 ± 2.16, oe‐SENP1 + sh‐NC: 15.35 ± 2.09, oe‐SENP1 + sh‐IGF2BP1: 9.01 ± 1.77). (E) Immunofluorescence was used to observe the number of GFAP (oe‐NC: 30.84 ± 6.19, oe‐SENP1: 12.04 ± 0.90, oe‐SENP1 + sh‐NC: 14.24 ± 1.17, oe‐SENP1 + sh‐IGF2BP1: 23.79 ± 3.00) and NeuN (oe‐NC: 11.39 ± 2.14, oe‐SENP1: 37.68 ± 5.30, oe‐SENP1 + sh‐NC: 35.24 ± 4.29, oe‐SENP1 + sh‐IGF2BP1: 15.05 ± 1.67)‐positive cells. Scale bar, 20 μm. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 6

IGF2BP2 interacted with SOX11 to promote its expression. (A and B) RIP (IgG: 1.00 ± 0.13, IGF2BP2: 9.02 ± 2.23) and RNA pull‐down assay verified IGF2BP2 binding to SOX11 mRNA. (C) Western blot analysis of SOX11 protein level in OM‐MSCs treated with oe‐NC (0.46 ± 0.03), oe‐IGF2BP2 (0.80 ± 0.08), sh‐NC (0.49 ± 0.02), or sh‐IGF2BP2 (0.32 ± 0.05). (D) SOX11 mRNA stability in OM‐MSCs after oe‐NC (0 h: 100.00 ± 2.32, 3 h: 66.73 ± 6.26, 6 h: 52.90 ± 5.57), oe‐IGF2BP2 (0 h: 100.00 ± 4.61, 3 h: 87.57 ± 4.73, 6 h: 76.45 ± 3.76), sh‐NC (0 h: 100.00 ± 6.74, 3 h: 70.51 ± 3.59, 6 h: 56.78 ± 4.93), or sh‐IGF2BP2 (0 h: 100.00 ± 8.57, 3 h: 53.49 ± 3.88, 6 h: 38.57 ± 4.23) treatment and incubation with actinomycin D for 0, 3, and 6 h was analyzed by qPCR. (E) RIP assay was performed using an anti‐IGF2BP2 antibody in OM‐MSCs with or without SENP1 knockdown. The enrichment of SOX11 mRNA was determined by qPCR (sh‐NC‐IgG: 0.48 ± 0.22, sh‐NC‐IGF2BP2: 5.15 ± 1.06, sh‐SENP1‐IgG: 0.40 ± 0.18, sh‐SENP1‐IGF2BP2: 2.18 ± 0.68). (F) RNA pull‐down assay was conducted using biotin‐labeled SOX11 RNA probes (sense and antisense) in OM‐MSCs transfected with sh‐NC or sh‐SENP1. IGF2BP2 binding was detected by western blot. β‐actin was used as a loading control. (G) OM‐MSCs with indicated treatments (sh‐NC (0 h: 100.00 ± 1.40, 3 h: 67.98 ± 3.76, 6 h: 53.00 ± 4.26), sh‐SENP1 (0 h: 100.00 ± 7.98, 3 h: 44.65 ± 4.82, 6 h: 30.64 ± 6.52), sh‐SENP1 + oe‐NC (0 h: 100.00 ± 3.01, 3 h: 45.24 ± 2.43, 6 h: 25.08 ± 2.08), sh‐SENP1 + oe‐IGF2BP2 (0 h: 100.00 ± 7.90, 3 h: 57.71 ± 2.91, 6 h: 47.08 ± 3.30)) were treated with actinomycin D for 0, 3, and 6 h. SOX11 mRNA levels were measured by qPCR and normalized to the 0‐h time point. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 7

IGF2BP2 facilitated OM‐MSC neuronal differentiation through increasing SOX11. (A and B) qPCR and western blot analysis of SOX11 mRNA (oe‐NC: 1.00 ± 0.14, oe‐SOX11: 2.89 ± 0.15) and protein levels (oe‐NC: 0.39 ± 0.01, oe‐SOX11: 0.82 ± 0.06) in OM‐MSCs treated with oe‐NC or oe‐SOX11 for 48 h. (C) OM‐MSCs were infected with Ad and incubated with differentiation medium for four days. The mRNA levels of Tuj‐1 (sh‐NC: 1.00 ± 0.11, sh‐IGF2BP2: 0.47 ± 0.09, sh‐IGF2BP2 + oe‐NC: 0.44 ± 0.06, sh‐IGF2BP2 + oe‐SOX11: 0.85 ± 0.11), MAP2 (sh‐NC: 1.00 ± 0.16, sh‐IGF2BP2: 0.21 ± 0.08, sh‐IGF2BP2 + oe‐NC: 0.24 ± 0.07, sh‐IGF2BP2 + oe‐SOX11: 0.78 ± 0.15), NF200 (sh‐NC: 1.00 ± 0.10, sh‐IGF2BP2: 0.35 ± 0.08, sh‐IGF2BP2 + oe‐NC: 0.33 ± 0.05, sh‐IGF2BP2 + oe‐SOX11: 0.73 ± 0.13), GFAP (sh‐NC: 1.00 ± 0.11, sh‐IGF2BP2: 1.92 ± 0.14, sh‐IGF2BP2 + oe‐NC: 1.79 ± 0.18, sh‐IGF2BP2 + oe‐SOX11: 1.39 ± 0.08), and ALDH1L1 (sh‐NC: 1.00 ± 0.06, sh‐IGF2BP2: 2.10 ± 0.21, sh‐IGF2BP2 + oe‐NC: 2.12 ± 0.13, sh‐IGF2BP2 + oe‐SOX11: 1.32 ± 0.18) was measured by qPCR assay. (D) Flow cytometry assay was used to detect Tuj‐1 positive cells (sh‐NC: 12.05 ± 0.99, sh‐IGF2BP2: 5.92 ± 1.07, sh‐IGF2BP2 + oe‐NC: 5.71 ± 1.18, sh‐IGF2BP2 + oe‐SOX11: 9.78 ± 1.34). (E) Immunofluorescence was used to observe the number of GFAP (sh‐NC: 22.68 ± 3.19, sh‐IGF2BP2: 46.90 ± 3.96, sh‐IGF2BP2 + oe‐NC: 50.26 ± 4.25, sh‐IGF2BP2 + oe‐SOX11: 28.34 ± 3.20) and NeuN (sh‐NC: 32.35 ± 4.40, sh‐IGF2BP2: 14.91 ± 2.24, sh‐IGF2BP2 + oe‐NC: 12.52 ± 2.97, sh‐IGF2BP2 + oe‐SOX11: 22.82 ± 1.13)‐positive cells. Scale bar, 20 μm. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 8

Schematic diagram of the mechanism in this study. The mechanism schematic diagram illustrates how the SENP1/IGF2BP2/SOX11 axis regulates neuronal differentiation in OM‐MSCs. It highlights the cascade of molecular interactions that promote neuronal differentiation and mitigate brain damage following ICH.