Introduction

Gliomas are one of the most common primary malignant brain tumors diagnosed in adults and arises as a consequence of uncontrolled growth of glial cells or their precursors. The World Health Organization (WHO) classifies gliomas into four grades based upon growth patterns and detected mutations to isocitrate dehydrogenase, among which glioblastoma (GBM) is recognized as the highest grade (IV) glioma (Lapointe et al. 2018; Louis et al. 2016). GBM is an invasive solid tumor, and the main treatment methods include surgery, radiotherapy, temozolomide chemotherapy (Lapointe, et al. 2018; Weller et al. 2017). In addition, GBM treatments with the anti-vascular drug bevacizumab also show promise, and patients prescribe with this treatment show reductions in peritumor edema and corticosteroid use, and a significant increase in progression-free survival, yet overall survival did not improve. The median survival time of GBM patients is less than two years, while the five-year survival rate is only 5% (Lofgren et al. 2022; Yang et al. 2022).

In GBM, tumors can form unique patterns of tube-like formations in which tumor cells replace the endothelium and form a structure that appears similar to blood vessels, a cellular phenomenon described as vasculogenic mimicry (VM). The cellular features of VM include the prominence of matrix protein-rich cells that do not express endothelial markers such as CD31 and CD34, while showing positive histochemical staining for periodic acid–Schiff (PAS) that detects the abundant carbohydrate stores (predominantly glycogen) in tumor cells, and the presence of blood components (Maniotis et al. 1999; ** vasculature. Cell. 2001;105(7):863–75. https://doi.org/10.1016/s0092-8674(01)00396-8 ." href="/article/10.1007/s10565-024-09890-5#ref-CR16" id="ref-link-section-d114376091e1914">2001). Our study shows that in GBM cells, an upregulation of NFATC3 likely leads to binding of the VEGFR2 promoter to transcriptionally activate VEGFR2 expression which, in turn, leads to an increase in VM tubes. Guided by the findings from previous studies and our current results, we speculate that there is a NFATC3-VEGFR2-NFATC3 regulatory loop that modulates VM formation in GBM. This is reminiscent of a postulated NFATC3-VEGFR2-NFATC3 signaling loop in breast cancer (Zhao et al. 2016). Furthermore, we verified that knockdown of NFATC3 inhibited the proliferation and migration capacities of GBM cells, which was consistent with previous findings (Urso et al. 2019).

FOSL1 belongs to the FOS family of transcription factors, and binds JUN to form the AP1 transcription factor complex, comprising a basic leucine zipper structure. This complex recognizes and bind DNA regulatory sequences known as TPA responsive elements (TRE) (Jiang et al. 2020), and FOSL1 can also dimerize with other transcription factors that are not AP1 members, but its function in those signaling contexts are less clear (Sobolev et al. 2022). We detected the binding of FOSL1 to NFATC3 in GBM and predicted possible binding sites for FOSL1 and NFATC3. However, in GBM cells that stably expression NFATC3 shRNA to suppress the expression of this gene, we found no evidence of binding by FOSL1 to the VEGFR2 promoter region. Furthermore, knockdown of FOSL1 in GBM cells led to reduced binding of NFATC3 to the VEGFR2 promoter. Furthermore, disruption to HNRNPA2B1 expression had no effect on FOSL1 mRNA levels. These lines of evidence indicate that FOSL1 and NFATC3 interact to transcriptionally co-regulate VEGFR2 expression in an NFATC3-dependent manner. This is consistent with the finding that NFAT interacts with the AP1 complex to promote transcription of another target, IL-13Ra2, in GBM cells (Wu et al. 2010).

HNRNPA2B1 is an RNA-binding protein that is prominent in cancer research. Elevated levels of HNRNPA2B1 are observed in gastric cancer (Hu et al. 2023), pancreatic cancer (Dai et al. 2017) and lung cancer (Yu et al. 2018), and this gene influences the proliferation and migration of cancer cells. Our knockdown studies with GMB cells also found that HNRNPA2B1 was necessary for the proliferation and migration of GBM cells. We detected that HNRNPA2B1 enriched NFATC3 mRNA rather than VEGFR2 mRNA, and knockdown of HNRNPA2B1 reduced the half-life of NFATC3 mRNA and not VEGFR2 mRNA. When we suppressed NFATC3 and VEGFR2 in stable transfected GBM cell lines using shRNAs, we found that this curtailed VM-like behaviors in GBM. These results suggest that HNRNPA2B1 influences VM-like behaviors in GBM by stabilizing NFATC3 mRNA. This is consistent with a previous study that reported that HNRNPA2B1 increases the expression of the VM marker MMP2 in GBM (Deng et al. 2016), and suggests that HNRNPA2B1 promotes VM formation in GBM through multiple pathways.

Through our in vivo experiments, we demonstrated that knockdown of NFATC3, FOSL1, and HNRNPA2B1 inhibited VM-like behaviors and resulted in a significant decrease in xenografted GBM tumors, both their sizes and volumes. Within GBM cell lines U251 and U373, we found that the regulation of VEGFR2 by HNRNPA2B1 and FOSL1 required the participation of NFATC3, such that NFATC3 was downstream of the signaling pathway. Our results are reminiscent of a study which reported that mice lacking NFATC3 showed decreased colorectal tumor numbers and reduced tumor sizes (Lin, et al. 2020). Drugs that inhibit NFAT, such as tacrolimus and cyclosporin A (CSA) that inhibit phosphatases, have demonstrated anticancer activity that is accompanied by immunosuppressive side effects (Durnian et al. 2007; Azzi et al. 2013). Considering the presence of the blood–brain barrier and phosphatase inhibition of many types of NFAT, the search for small molecule drugs that are more specific to NFATC3 could be an appropriate future direction for novel GBM treatments. This is one direction of our future research.

Conclusion

We find that HNPNPA2B1, NFATC3 and FOSL1 are highly expressed in GBM cells and are associated with their cellular behaviors including cell proliferation, migration and the formation of tube-like structures reminiscent of VM. We find that NFATC3 promotes VM formation and transcriptionally upregulates VEGFR2 transcription. FOSL1 interacts with NFATC3 to further facilitate VEGFR2 gene expression and VM. Furthermore, the half-life of NFATC3 mRNA is prolonged by HNRNPA2B1, and NFATC3 expression is increased, ultimately leading to enhanced VM. Taken altogether, our findings suggest that targeting NFATC3 and its interacting partners could represent a viable strategy to develop novel treatments for GBM with VM features.