Treatment of osteosarcoma, a cancer type mostly affecting children and young adults, is complicated by the relative drug resistance of this disease. Basic fibroblast growth factor (b-FGF2, FGF2) signalling is involved in the development of drug resistance in various cancers, including osteosarcoma. Hence, inhibitors of the FGF2 receptors (FGFRIs) were expected to provide ways to re-sensitize resistant tumours to therapy. Unfortunately, the degree of enhanced sensitivity of tumours to conventional chemotherapy when used in combination with FGFRIs in mouse models has not so far been reproduced in a clinical environment. These results highlight the importance of appropriate patient stratification and the associated need to identify novel biomarkers to predict tumour response to FGFR-directed therapy. Previous studies in lung cancer from our lab revealed part of the underlying mechanism of FGF2-mediated cell resistance to chemotherapy, which involves the binding and cytoplasmic export of mRNAs encoding survival proteins by the RNA-binding protein heterogeneous nuclear ribonucleoproteins A1 (hnRNPA1).
In this study, based on this existing knowledge, we first validated and further explored the molecular mechanisms underlying FGF2 signalling in osteosarcoma cells. We identified ribosomal s6 kinase4 (RSK4) as a new upstream regulator for hnRNPA1 localization in response to FGF2, providing new potential druggable avenues to overcome FGF2-mediated resistance in the clinic. We then revealed the full repertoire of mRNAs binding to hnRNPA1 in response to FGF2 stimulation using RNA immunoprecipitation followed by sequencing (RIP-Seq). Pathway enrichment analysis of the RIP-Seq data highlighted the potential involvement of the FGF2/hnNRPA1 axis in the regulation of cytokine signalling and the crosstalk between cancer cells and the tumour micro-environment. Indeed, validation of some of these findings in our in vitro cell systems revealed that this axis-controlled activation of cancer-associated fibroblasts, as hnRNPA1 depletion prevented this process. In addition, our data indicated that hnRNPA1 is a prominent regulator for IFNγ signalling in osteosarcoma cells both in an FGF2-dependent and -independent manner. Of particular interest, both FGF2 and hnRNPA1 silencing sensitized U2OS osteosarcoma cells to IFNγ-induced growth arrest, a process that appeared dependent on the expression of wild-type p53. This occurred partly through disruption of cell cycle progression, as IFNγ induced an G2/M arrest in hnRNPA1-downregulated cells. Furthermore, hnRNPA1 is required for the canonical STAT1/IRF1 axis downstream of IFNγ signalling as hnRNPA1 knockdown dampened the IFNγ-induced overexpression of its key downstream mediators, an effect promoted by FGF2 co-stimulation. Finally, we discovered a novel role of hnRNPA1 in regulating the expression of the immune checkpoint protein programmed cell death 1 ligand 1 (PDL1) and its receptor, programmed cell death 1 (PD1) in osteosarcoma cells, with FGF2 further promoting IFNγ-induced PDL1 overexpression in hnRNPA1-silenced cells.
In short, the obtained data provided invaluable insight into the signalling pathways modulated by the FGF2/hnRNPA1 axis and propose novel therapeutic strategies to tackle FGF2-associated therapy resistance that, once further validated, could ultimately be tested in the clinic.