IGF2BP family of RNA-binding proteins regulate innate and adaptive immune responses in cancer cells and tumor microenvironment

Abstract

Insulin-like growth factor 2 mRNA-binding proteins (IGF2BP1, IGF2BP2, and IGF2BP3) are a family of RNA-binding proteins that play an essential role in the development and disease by regulating mRNA stability and translation of critical regulators of cell division and metabolism. Genetic and chemical inhibition of these proteins slows down cancer cell proliferation, decreases invasiveness, and prolongs life span in a variety of animal models. The role of RNA-binding proteins in the induction of tissues' immunogenicity is increasingly recognized, but, the impact of the IGF2BPs family of proteins on the induction of innate and adaptive immune responses in cancer is not fully understood. Here we report that downregulation of IGF2BP1, 2, and 3 expression facilitates the expression of interferon beta-stimulated genes. IGF2BP1 has a greater effect on interferon beta and gamma signaling compared to IGF2BP2 and IGF2BP3 paralogs. We demonstrate that knockdown or knockout of IGF2BP1, 2, and 3 significantly potentiates inhibition of cell growth induced by IFNβ and IFNγ. Mouse melanoma cells with Igf2bp knockouts demonstrate increased expression of MHC I (H-2) and induce intracellular Ifn-γ expression in syngeneic T-lymphocytes in vitro. Increased immunogenicity, associated with Igf2bp1 inhibition, "inflames" mouse melanoma tumors microenvironment in SM1/C57BL/6 and SW1/C3H mouse models measured by a two-fold increase of NK cells and tumor-associated myeloid cells. Finally, we demonstrate that the efficiency of anti-PD1 immunotherapy in the mouse melanoma model is significantly more efficient in tumors that lack Igf2bp1 expression. Our retrospective data analysis of immunotherapies in human melanoma patients indicates that high levels of IGF2BP1 and IGF2BP3 are associated with resistance to immunotherapies and poor prognosis. In summary, our study provides evidence of the role of IGF2BP proteins in regulating tumor immunogenicity and establishes those RBPs as immunotherapeutic targets in cancer.

Keywords: IGF2BP/IMP; RNA-binding protein (RBP); anti-PD-1; cancer stem cell (CSC); immunotherapy; interferon-stimulated genes (ISGs); leukemia; melanoma.

Figure 1

IGF2BP1 regulates type – I and type-II IFN-stimulated gene expression in mammalian cells (A) RNA-seq gene ontology (GO) and KEGG pathway analysis, by STRING software, of human B-ALL 697 (EU3) ALDH⁺ cells with a doxycycline-inducible IGF2BP1 knockdown; log2(fold) expression of several interferon-stimulated genes from B-ALL mRNA-seq data set; (B) HEK293T cells mRNA-seq analysis: a heat map of differentially expressed, type I interferon signaling genes in HEK293T cells expressing short-hairpin (sh) RNAs targeting IGF2BP messengers (shIGF2BP1, 2, 3) or non-targeting shRNA control (shControl, shCNTRL) before (left map), and after hrIFN-β (10ng/mL, 12 hours (hrs)) treatment (right map). (C) GSEA type I interferon pathway (left panel), interferon gamma signaling reactome (right panel) in unstimulated HEK293T cells expressing shIGF2BP1 (shIGF2BP1) or hrIFN-β-treated shIGF2BP1 (shIGF2BP1+hrIFN-β), versus to non-targeting corresponding (unstimulated and hrIFN-β-treated) shControls.

Figure 2

IGF2BP1 downregulation sensitizes human cancer cells to IFN-beta and IFN-gamma treatment. (A) IncuCyte™ cell imaging analysis of human melanoma Mel928 cells growth expressing shIGF2BP1 or shControl, stimulated with hrIFN-β (10ng/mL), hrIFN-γ (100ng/mL), and hrTNF-α (10ng/mL) and (B) SK-Mel-28 treated with hrIFN-β (10ng/mL); the summary of the unpaired t-tests at 60-hours of growth, approximately 50% cell confluency for most samples, is presented on right graphs; the summary of the unpaired t-tests at 100-hours of growth, the end time point at maximum cell confluency, is presented on the plots (left); (C) qPCR for RIG-1 expression in SK-Mel-28 and MEL928 cells, expressing shIGF2BP1 or shControl, treated with hrIFN-β (left graph), and hrIFN-γ (right graph), relative to unstimulated shRNA control. *p<0.05, **p<0.01, ***p<0.001; not significant (ns, p>0.05).

Figure 3

IGF2BPs downregulation sensitizes mouse cancer cells to IFN-beta and IFN-gamma treatment. (A) IncuCyte™ cell growth imaging analysis of SM1 mouse melanoma Igf2bp1-KO or non-targeting guide RNA (gRNA) CRISPR/Cas9 control (CNTRL), stimulated with mouse recombinant (mr) Ifn-β (1000Units/mL, left graph) or mrIfn-γ (50Units/mL, center graph). IncuCyte™ cell growth imaging analysis of PVMM (Ifnar1^−/− ) cell line Igf2bp1-KO or CRISPR/Cas9 control, stimulated with mrIfn-β (1000Units/mL) or mrIfn-γ (50Units/mL);the summaries of the unpaired t-tests at 120-hours of growth, the end time point at maximum cell confluency, are presented on the plots; (B) IncuCyte™ cell growth imaging analysis of mouse melanoma B16.F10-Tag cells genetically edited by CRISPR/Cas9 non-targeting gRNA (CNTRL), gRNA against Igf2bp2 (Igf2bp2-KO), Igf2bp3 (Igf2bp3-KO), and exposed to PBS (unstimulated control), or mrIfn-γ 50U/mL, or mrIfn--γ 100U/mL; the summaries of the unpaired t-tests at 120-hours of growth, the end time point at maximum cell confluency, are presented on the plots; (C) Endogenous expression of interferon-stimulated genes (ISGs) in unstimulated mouse Igf2bp1-3-KO clones relative to CRISPR/Cas9 non-targeting control assessed by qPCR (left); Normalized relative expression of endogenous interferon beta (Infb1) gene in mouse melanoma cells with CRISPR/Cas9-induced Igf2bp1, 2, and 3 knockouts relative to CRISPR/Cas9 gRNA non-targeting control after mrIfn-γ stimulation (50U/mL, 12 hrs), relative to mrIfn-γ stimulated CRISPR/Cas9 CNTRL. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; not significant (ns, p>0.05).

Figure 4

Downregulation of IGF2BP family of proteins increases immunogenicity of mouse melanoma cells and inflames tumor microenvironment. (A) Flow cytometric analysis of mouse H-2D^b surface expression in SM1 Igf2bp1-KO and B16.F10-Tag Igf2bp2-KO, Igf2bp3-KO cells compared to CRISPR/Cas9 control; *p<0.05, ****p<0.0001; (B) flow cytometric analysis of intracellular Ifn-γ expression in mouse CD8+ T-cell clone K-11, co-cultured for 5 hours with B16.F10-Tag Igf2bp2-KO, Igf2bp3-KO, or CRISPR/Cas9 CNTRL clones (n=3); (C) flow cytometric analysis of mouse tumor microenvironment in SM1 mouse melanoma tumors formed by Igf2bp1-KO or CRISPR/Cas9 CNTRL cells in syngeneic C57BL/6 mice (n=6); (D) quantification and representative flow cytometry plots for myeloid cells in doxycycline-inducible SW1 Igf2bp1-KD compared to SW1 shControl tumors in syngeneic C3H mice (n=7).

Figure 5

Downregulation of IGF2BP1 increases sensitivity to immunotherapy in mouse and human melanoma. (A) Kaplan-Meier survival analysis of C57BL/6 melanoma mouse model injected with 0.7x10⁶ SM1 cells carrying Igf2bp1-KO (n=20) or gRNA non-targeting CRISPR/Cas9 control (n=20), treated with either PBS (n=20) or anti-PD-1 antibodies (n=20); (B) a retrospective statistical analysis of human melanoma patient survival treated with immunotherapies, considering endogenous low or high levels of IGF2BP1 and IGF2BP3 expression (p<0.01, p<0.001 respectively).