IGFBP5 is an ROR1 ligand promoting glioblastoma invasion via ROR1/HER2-CREB signaling axis

Abstract

Diffuse infiltration is the main reason for therapeutic resistance and recurrence in glioblastoma (GBM). However, potential targeted therapies for GBM stem-like cell (GSC) which is responsible for GBM invasion are limited. Herein, we report Insulin-like Growth Factor-Binding Protein 5 (IGFBP5) is a ligand for Receptor tyrosine kinase like Orphan Receptor 1 (ROR1), as a promising target for GSC invasion. Using a GSC-derived brain tumor model, GSCs were characterized into invasive or non-invasive subtypes, and RNA sequencing analysis revealed that IGFBP5 was differentially expressed between these two subtypes. GSC invasion capacity was inhibited by IGFBP5 knockdown and enhanced by IGFBP5 overexpression both in vitro and in vivo, particularly in a patient-derived xenograft model. IGFBP5 binds to ROR1 and facilitates ROR1/HER2 heterodimer formation, followed by inducing CREB-mediated ETV5 and FBXW9 expression, thereby promoting GSC invasion and tumorigenesis. Importantly, using a tumor-specific targeting and penetrating nanocapsule-mediated delivery of CRISPR/Cas9-based IGFBP5 gene editing significantly suppressed GSC invasion and downstream gene expression, and prolonged the survival of orthotopic tumor-bearing mice. Collectively, our data reveal that IGFBP5-ROR1/HER2-CREB signaling axis as a potential GBM therapeutic target.

Fig. 1. IGFBP5 expression is associated with GSCs invasion and patient survival in glioma.

a Hematoxylin and eosin (H&E) staining of the whole brain from mice implanted with GSCs (448, X01, 131, and 83). Scale bar, 100 μm. b Invasion assays of four GSCs (448, X01, 131, and 83) after 48 h. Images are representative of three independent experiments (Scale bar, 100 μm; n = 3), the bar graph shows the average numbers of invasive cells. All error bars represent means ± standard error of the mean (SEM), ^***P < 0.001, two-tailed Student’s t-test. c Venn diagram of Differentially-Expressed Genes (DEGs) identified in our RNA-seq analysis and the TCGA low-grade glioma (LGG) RNA-seq dataset (left) and the putative candidate genes associated with GSC invasion and the survival of patients with glioma (right). d IGFBP5 read distribution (RNA-seq) in the four types of GSCs. e RT-qPCR validation of IGFBP5 expression in GSCs. Data are represented as mean ± SEM (n = 3 independent experiments), ^**P < 0.01, two-tailed Student’s t-test. f ELISA analysis of secreted IGFBP5 protein in conditioned media (CM) from GSCs cultured for 3 days. Data are presented as mean ± SEM (n = 3 independent experiments), ^***P < 0.001, two-tailed Student’s t-test. g Kaplan–Meier survival curves for LGG patients with high or low IGFBP5 expression based on the median expression in the TCGA dataset. P = 0.000431, log-rank test. h Kaplan–Meier survival curves for GBM patients with high or low IGFBP5 expression based on the median expression in the TCGA dataset. P = 0.0031, log-rank test. Source data and exact P values are provided as the Source Data file.

Fig. 2. Inhibition of IGFBP5 impairs GSCs invasion and tumorigenesis.

a RT-qPCR analysis of IGFBP5 mRNA expression in X01 GSCs infected with lentivirus expressing shIGFBP5 or shCtrl. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, ^***P = 0.0008, ^***P = 0.0007). b ELISA analysis of IGFBP5 in CM from X01 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, ^****P = 0.00001, ^****P = 0.00001). c, d Invasion assays with X01 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. c Images taken after 48 h of invasion are representative of three independent experiments (scale bar, 100 μm; n = 3), and d the graph shows the mean number of invasive cells ± SEM, two-tailed Student’s t-test (left to right, ^**P = 0.003, ^**P = 0.007). e RT-qPCR analysis of IGFBP5 mRNA expression in 448 GSCs infected with shCtrl, shIGFBP5-1, and shIGFBP5-2 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, ^***P = 0.0006, ^**P = 0.002). f ELISA analysis of IGFBP5 in CM from 448 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, ^***P = 0.0002, ^***P = 0.0002). g, h Invasion assays using 448 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. g Images taken after 48 h of invasion are representative of three independent experiments (scale bar, 100 μm; n = 3), and h the graph shows the mean number of invasive cells ± SEM, two-tailed Student’s t-test (left to right, ^***P = 0.0004, ^***P = 0.0004). i H&E staining of the whole brains of mice bearing orthotopic xenografts of X01 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Scale bar, 100 μm. j Kaplan–Meier survival curves of mice implanted with X01 GSCs infected with shCtrl (n = 10), shIGFBP5-1 (n = 8), or shIGFBP5-2 (n = 8) lentivirus (1 × 10⁴ cells/mouse). P < 0.0001, log-rank test. Source data are provided as the Source Data file.

Fig. 3. Ectopic IGFBP5 overexpression enhances invasion and tumorigenesis of non-invasive GSCs.

a RT-qPCR analysis of IGFBP5 expression in 83 GSCs infected with IGFBP5 or vector control lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (^***P = 0.0009). b ELISA analysis of IGFBP5 in CM from 83 GSCs infected with IGFBP5 or vector control lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (^**P = 0.003). c, d Invasion assays with 83 GSCs infected with vector control or IGFBP5 lentivirus. c Images taken after 24 h of invasion are representative of three independent experiments (scale bar, 100 μm; n = 3), and d the graph shows the mean number of invasive cells ± SEM, two-tailed Student’s t-test (^***P = 0.0007). e RT-qPCR analysis of IGFBP5 expression in 131 GSCs infected with IGFBP5 or vector control lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (^***P = 0.0002). f ELISA analysis of IGFBP5 in CM from 131 GSCs infected with IGFBP5 or vector control lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (^****P = 0.00002). g, h Invasion assays using 131 GSCs infected with vector control or IGFBP5 lentivirus. g Images taken after 24 h of invasion are representative of three independent experiments (scale bar, 100 μm; n = 3), and h the graph shows the mean number of invasive cells ± SEM, two-tailed Student’s t-test test (^**P = 0.004). i H&E staining of the whole brains of mice bearing orthotopic xenografts of 83 GSCs infected with vector control or IGFBP5 lentivirus. Scale bar, 50 μm. j Kaplan–Meier survival curves of mice implanted with 83 GSCs infected with vector control or IGFBP5 lentivirus (n = 6 in each group, 1 × 10³ cells/mouse). P = 0.0051, log-rank test. Source data are provided as the Source Data file.

Fig. 4. IGFBP5 activates HER2 and ROR1 signaling.

a Human phospho-RTK array of 83 GSCs treated with 100 ng/ml recombinant IGFBP5 (rIGFBP5) or vehicle control for 6 h. b Human phospho-kinase array of 83 GSCs treated with 100 ng/ml rIGFBP5 or vehicle control for 6 h. c–e Immunoblot (IB) analysis of pROR1, pHER2, pCREB, ROR1, HER2, and CREB in c non-invasive GSCs (83 and 131) treated with rIGFBP5 (100 ng/ml) or vehicle control for 6 h, d 83 and 131 cells infected with vector control or IGFBP5 lentivirus and e 448 and X01 GSCs infected with shIGFBP5 or shCtrl lentivirus. β-actin was used as a loading control. f, g Co-IP of X01 cells with antibodies targeting HER2, ROR1 or normal IgG. h Co-IP analysis for the interaction of IGFBP5 and ROR1 in 293 T cells transfected with Flag-tagged IGFBP5 and HA-tagged ROR1. Cell lysates were precipitated with anti-Flag antibody. i Co-IP analysis for the interaction of IGFBP5 and ROR1 in 293 T cells transfected with HA-tagged ROR1 and Flag-tagged IGFBP5. Cell lysates were precipitated with anti-HA antibody. j In vitro binding affinity between IGFBP5 and ROR1 tested by MST assay. The concentration of IGFBP5 proteins is kept constant at 50 nM, while the ROR1 concentration varies from 1.45 µM to 0.04 nM. The binding curve yields a Kd of 157.5 nM. Inset, thermophoretic movement of fluorescently labeled proteins. Fnorm = F₁/F₀ (Fnorm: normalized fluorescence; F₁: fluorescence after thermodiffusion; F₀: initial fluorescence or fluorescence after T-jump). Kd, dissociation constant. k IB analysis of pROR1, pHER2, pCREB, ROR1, HER2, and CREB in X01 GSCs infected with shCtrl, shHER2-1, or shHER2-2 lentivirus. GAPDH was used as a loading control. l IB analysis of pROR1, pHER2, pCREB, ROR1, HER2, and CREB in X01 GSC infected with shCtrl, shROR1-1, or shROR1-2 lentivirus. GAPDH was used as a loading control. m IHC analysis of pHER2, pROR1, and pCREB in orthotopic xenografts of X01 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Scale bar, 100 μm. n IHC analysis of pHER2, pROR1, and pCREB in orthotopic xenografts of 83 GSCs infected with vector control or IGFBP5 lentivirus. Scale bar, 100 μm. All the immunoblots were representative data from three independent experiments. Source data are provided as the Source Data file.

Fig. 5. IGFBP5 activates transcription of ETV5 and FBXW9 via CREB.

a Venn diagram showing the overlap of upregulated DEGs in invasive vs non-invasive GSCs and downregulated DEGs in 448 and X01 GSCs infected with shIGFBP5 versus GSCs infected with shCtrl based on RNA-seq data. The 22 selected genes are putative invasion-related target genes of IGFBP5. b Heatmap of the log2-fold change in putative invasion-related target genes of IGFBP5 identified by RNA-seq analysis of invasive GSCs infected with shIGFBP5 or shCtrl. c Kaplan–Meier survival curves for LGG patients with high or low ETV5 (left, P = 0.0004) or FBXW9 expression (right, P = 0.0005), log-rank test. d, e RT-qPCR analysis of ETV5 and FBXW9 expression in 83 GSCs infected with vector control or IGFBP5 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (d, e, ^**P = 0.008, ^****P = 0.00004). f, g RT-qPCR analysis of ETV5 and FBXW9 expression in X01 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, (f) ^**P = 0.001, ^**P = 0.008; (g) ^***P = 0.0003, ^***P = 0.0002). h, i RT-qPCR analysis of ETV5 and FBXW9 expression in invasive GSCs (X01 and 448) and non-invasive GSCs (131 and 83). Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (compare with 131 group, left to right, ^***P = 0.0002, ^****P = 0.00001, ^*P = 0.03; ^***P = 0.0006, ^***P = 0.0001, ^**P = 0.003). j ChIP-qPCR analysis of CREB binding to the FBXW9 TSS in 448 GSCs infected with shCtrl or shIGFBP5-1 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (^**P = 0.002). k, l ChIP-qPCR analysis of CREB binding to the ETV5 promoter in 448 GSCs infected with shCtrl or shIGFBP5-1 lentivirus or in 83 GSCs infected with vector control or IGFBP5 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (k, l, ^*P = 0.02, ^**P = 0.006). m Kaplan–Meier survival curves for LGG patients stratified by IGFBP5 and ETV5 (left, P = 0.00000006) or IGFBP5 and FBXW9 expression (right, P = 0.00004), log-rank test. Source data are provided as the Source Data file.

Fig. 6. IGFBP5 knockdown suppresses GSC invasion and tumorigenesis in patient-derived xenografts.

a Magnetic resonance imaging (MRI) of mice bearing orthotopic xenografts of patient-derived 772 GSCs. b Categorization of patient-derived 772 GSCs based on the RNA-seq analysis. c RT-qPCR analysis of IGFBP5 expression in patient-derived 772 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Data are presented as mean ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, ^****P = 0.00004, ^****P = 0.00005). d Invasion assays with patient-derived 772 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. The graph shows the mean number of invasive cells ± SEM (n = 3 independent experiments), two-tailed Student’s t-test (left to right, ^***P = 0.0004, ^**P = 0.001). e H&E staining of the whole brains of mice bearing orthotopic xenografts of patient-derived 772 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Scale bar, 100 μm. f Kaplan–Meier survival curves of mice implanted with patient-derived 772 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus (n = 6 in each group, 1 × 10⁵ cells/mouse). shCtrl vs shIGFBP5-1, P = 0.0275; shCtrl vs shIGFBP5-2, P = 0.0005; log-rank test. g IHC analysis of pHER2, pROR1, and pCREB in orthotopic xenografts of patient-derived 772 GSCs infected with shCtrl, shIGFBP5-1, or shIGFBP5-2 lentivirus. Scale bar, 100 μm. Source data are provided as the Source Data file.

Fig. 7. Nanocapsule-mediated delivery of Cas9/sgIGFBP5 suppresses GSC invasion and tumorigenesis.

a Schematic illustration of Ang-SS-Cas9/sgRNA nanocapsule preparation. b Size distribution of Ang-SS-Cas9/sgRNA nanocapsules. c TEM images of Ang-SS-Cas9/sgRNA nanocapsules treated with or without 10 mM GSH at pH 7.4 for 12 h. Scale bar, 50 nm. d, e Indel detection by T7 endonuclease I (T7EI) of X01 and 448 GSCs treated by Ang-SS-Cas9/sgIGFBP5. T7EI cleavage assays were representative data from three independent experiments. f, g Invasion assays with X01 GSCs treated with Ang-SS-Cas9/sgIGFBP5 and Ang-SS-Cas9/sgScramble. f Images taken after 48 h of invasion are representative of three independent experiments (scale bar, 100 μm; n = 3), and g the graph shows the mean number of invasive cells ± SEM, two-tailed Student’s t-test (^***P = 0.0001). h, i Invasion assays with 448 GSCs treated with Ang-SS-Cas9/sgIGFBP5 and Ang-SS-Cas9/sgScramble. h Images taken after 48 h of invasion are representative of three independent experiments (scale bar, 100 μm; n = 3), and i the graph shows the mean number of invasive cells ± SEM, two-tailed Student’s t-test (^***P = 0.0004). j Luminescence images of orthotopic X01-Luc human glioblastoma tumor-bearing nude mice following treatment with Ang-SS-Cas9/sgIGFBP5, Ang-SS-Cas9/sgScramble or PBS. Mice were intravenously injected at a dose of 1.5 mg Cas9 equiv./kg on day 10, 12, 14, 16, and 18 post tumor implantations. k IB analysis of IGFBP5, pROR1, pHER2, pCREB, ROR1, HER2 and CREB in tumor tissues taken from mice treated with Ang-SS-Cas9/sgIGFBP5, Ang-SS-Cas9/sgScramble or PBS. β-actin was used as the loading control. The immunoblots were representative data from three independent experiments. l Body weight changes in mice following treatment with Ang-SS-Cas9/sgIGFBP5, Ang-SS-Cas9/sgScramble or PBS. Data are presented as mean ± SEM (n = 6 in each group), one-way ANOVA (day 18, ^***P = 0.0006) and two-tailed Student’s t-test (day 20 sgIGFBP5 vs sgScramble, ^****P = 0.00008). m Kaplan–Meier survival curves of mice implanted with 1 × 10⁵ X01-Luc GSCs and treated with Ang-SS-Cas9/sgIGFBP5, Ang-SS-Cas9/sgScramble or PBS (n = 6 in each group). PBS vs sgIGFBP5, P = 0.0006, sgScramble vs sgIGFBP5, P = 0.001, log-rank test. Source data are provided as the Source Data file.