Year: 2018
Type: article
Abstract: The chromosomal translocation-generated gene fusion of FGFR3 and TACC3 (F3-T3) was previously described in human glioblastoma (GBM) and others cancers. F3-T3 fusions are oncogenic, but the downstream signaling pathways that mediate their effects are not known. A recent study published in the journal Nature uncovered the mode of action of this fusion in transformed astrocytes and brain tumors.1 The study showed that F3-T3 acts as an oncogene by activating mitochondrial metabolism and uncovered the underlying molecular mechanism. Analysis of the transcriptional profiles of glioma cells expressing F3-T3 or immortalized astrocytes overexpressing the gene fusion followed by functional assays showed that F3-T3 activated oxidative phosphorylation, mitochondrial biogenesis, oxygen consumption rate and ATP production and induced tumor growth. Phosphopeptide screening identified tyrosine-phosphorylated PIN4 [PIN4(Y122)], a homologue of the peptidyl-prolyl-trans-isomerase 1 cancer-driver, as a direct substrate of F3-T3. PIN4(Y122) was elevated in human GBM expressing F3-T3 and mediated the metabolic and oncogenic effects of F3-T3 in transformed astrocytes and glioma cells. To identify the transcription factors that might mediate the effects of F3-T3 in glioma, transcriptional networks from public data sets were analyzed and uncovered the transcriptional coactivator PCG1α and its nuclear estrogen-related receptor ERRγ as being the most activated in F3-T3-positive tumors. Subsequent functional and rescue experiments demonstrated that PCG1α mediated the activation of mitochondrial metabolism and tumor growth induced by F3-T3. Further investigation of F3-T3/PIN4 signaling identified the peroxisomal biogenesis complex proteins PEX1 and PEX6 as PIN4 interactors that mediated an induction of peroxisome biogenesis by F3-T3. The study then showed that peroxisome biogenesis generated reactive oxygen species (ROS) and that ROS induced PCG1α. In summary, the study showed that the F3-T3 oncogenic gene fusion signals through PIN4(Y122) to trigger peroxisome biogenesis resulting in the accumulation of ROS and increase in PCG1α-ERRγ and mitochondrial metabolism. Based on the above, the authors of the study argued that the dependency on mitochondrial metabolism suggests that GBM patients with F3-T3 tumors might benefit from inhibitors of oxidative phosphorylation. To date, therapies modulating the immune system have not hit a home run in treating glioblastoma. Oncolytic viruses, one of immunotherapy approaches being explored, are designed to bind specific tumor antigens and cause direct tumor cell death, local cytokine production, and enhanced T cell-mediated immunity. Another goal of this approach is to have a sustained immune system response to the tumor. With FDA approval of talimogene laherparepvec (T-VEC) to treat advanced melanoma in 2015, there has been increased interest in using oncolytic viruses to fight cancer. In a small study, Lang et al used an oncolytic adenovirus (DNX-2401) in patients with recurrent high-grade glioma.1 While the small sample size and single arm design limit conclusions, the study is promising, warranting further exploration of oncolytic viruses in GBM. Two groups were enrolled: one group received a single intratumoral injection of DNX-2401 and the second group received a DNX-2401 injection via an intratumoral catheter followed 2 weeks later by en bloc tumor resection and repeat injection. The post-treatment tissue showed evidence of viral infection, cell lysis, and viral replication in some, but not all, patients. In addition there seemed to be an increase in CD3 and CD4 cells. Median overall survival for both cohorts was 13 months, and 3 patients had a complete radiographic response with a duration of response >3 years. These patients were young and one harbored an IDH mutation. This initial report is intriguing but it is hard to draw firm conclusions with these small patient numbers. Larger studies are needed to confirm the findings and the next steps will include optimization of oncolytic virus delivery, as not all patients with post-treatment tissue showed evidence of viral effects. Chordoid gliomas are unusual tumors thought to arise from specialized ependymal cells in the anterior third ventricle. Despite their low-grade histopathology, the problematic localization of chordoid gliomas can render curative resection problematic. Moreover, effective non-surgical regimens for chordoid glioma treatment have not been established, in part due to significant knowledge gaps regarding the molecular pathogenesis of this rare neoplasm. In a recent Nature Communications report, Goode et al. performed targeted next-generation sequencing, consisting of a ~500-gene panel, on 13 chordoid gliomas, two of which were associated with patient-matched normal DNA.1 This analysis revealed a unique somatic mutation in PRKCA involving aspartate 463 (D463H) that was present in all sequenced tumors. No other genomic alterations involving established glioma- and/or ependymoma-relevant genes were identified. PRKCA encodes protein kinase C alpha (PKCα), a MAP kinase pathway effector, and D463 is an essential residue within the PKCα kinase domain. Subsequent in vitro analysis using immortalized normal human astrocytes (NHAs) and NIH-3T3 cells demonstrated that D463H-mutant PKCα enhances anchorage-independent growth over baseline, an effect not seen with a kinase-dead mutation (D463A). Moreover, PKCα D463H increased MAP kinase signaling in the setting of serum starvation. Consistent with this latter finding, chordoid gliomas exhibited high levels of phospho-ERK immunostaining, comparable to those associated with other MAP kinase pathway-activated tumors (e.g. pilocytic astrocytomas and pleomorphic xanthoastrocytomas). Finally, the authors demonstrated that MEK inhibition with trametinib abrogated the oncogenic effects of PKCα D463H in NHAs while exerting only modest effects on non-PRKCA-altered malignant glioma cell lines. This work provides yet another example of how large-scale molecular profiling can lead to striking and therapeutically relevant insights into the pathogenesis of rare tumors. While new immunotherapies are making dramatic advances in oncology, it has also become clear that standard therapies may also be acting in part through immunologic mechanisms. Given early evidence that the most prominent new immunotherapies may have less efficacy against challenging brain tumors such as glioblastoma (GBM), it may be especially important to understand the immunologic consequences of neuro-oncologic mainstay therapies such as radiation and temozolomide chemotherapy. Furthermore, there is a need to think outside the usual T cell focus of most studies; a new report from Weiss et al. indicates that radiation and temozolomide may act in GBM to a large part through up-regulation of ligands for an activating receptor found on natural killer (NK) and other immunologic cells.1 Natural-killer group 2 member D (NKG2D) is a major activator for NK cells, although other immune cells also express it. Its NKG2DL ligands, including MICA, MICB, and UL16 binding proteins 1–6, were previously shown to be expressed in gliomas and other cancers. Malignant cell transformation and DNA damage have been shown to increase expression of NKG2DL, which prompted the authors to assess whether radiation and temozolomide trigger expression of NKG2DL in glioma. They indeed found evidence of marked up-regulation of NKG2DL in gliomas exposed to either radiation or temozolomide, both in vitro and in vivo. Importantly, this therapy-triggered up-regulation of NKG2DL was also observed in GBM from human patients. Another alkylating chemotherapy agent besides temozolomide, CCNU, was also able to induce the NKG2DL up-regulation, but cytotoxic treatment with the multi-kinase inhibitor staurosporine did not. DNA damage seemed to be important from a mechanistic perspective, as the DNA damage mediator ATM played a key regulatory role in the effect. Remarkably, negating NKG2D effects in vivo with either a blocking antibody or the use of NKG2D knockout mice substantially reduced—if not ablated—the therapeutic efficacy of both radiation and temozolomide. These dramatic and surprising results supplement and even challenge our previous notions of the mechanisms of our central therapies for GBM, in addition to raising a host of questions. They indicate the importance of NKG2D in anti-GBM immunity, and further suggest testing combinations to leverage the NKG2DL-boosting effects of radiation and temozolomide. It will be critical to determine how other approved therapies affect NKG2DL expression in gliomas, but also how all of these therapies affect activity of the NK cells and other immune cells stimulated by NKG2D. Other factors should also be studied, such as whether dose and schedule of these agents make a substantial difference in the effect. Answering the questions raised by this report could have major implications for our treatment of GBM and other cancers. Isocitrate dehydrogenase (IDH) mutations seem to confer a better natural course of the disease in gliomas1,2 and leukemia3,4 irrespective of treatment potentially reflecting the growth- and glycolysis-inhibitory action,5,6 although a better response to treatment as a confounder can still not be excluded. Mutant IDH has a neomorphic enzymatic function and catalyzes the conversion of α-KG to the R enantiomer of 2-hydroxyglutarate (R-2HG). Tumor-promoting properties of R-2HG led to its classification as a tumor-promoting metabolite. Recent data suggest that R-2HG, despite its activity as an oncometabolite in IDH-mutant tumors, has antitumor activity in leukemia and glioma.7 This is mediated through inhibiting the enzymatic activity of fat mass- and obesity-associated protein (FTO). FTO effectively demethylates internal N6-methyladenosine (m6A). The R-2HG inhibition results in increased global m6A and accumulation of m6A on MYC transcripts, which leads to decreased MYC mRNA stability and downregulation of MYC and CCAAT/enhancer-binding protein alpha (CEBPA) signaling. CEBPA is a transcription factor described to coordinate differentiation in blood cells and recently reported to be a prognostic factor in several solid tumors. It is very likely that the inhibitory effects of R-2HG or IDH mutations on TET2 and FTO impact different stages or different types of cancer initiation and progression. In gliomas, R-2HG may serve as both an oncometabolite to facilitate initiation of cancer as well as a tumor suppressor to suppress further progression of initiated tumors. In a series of experiments, the authors demonstrate that high levels of FTO sensitize leukemic cells to R-2HG, whereas hyperactivation of MYC signaling confers resistance that can be reversed by the inhibition of MYC signaling.7 R-2HG also inhibits glioblastoma cell proliferation and viability of eight cell lines and acts synergistically with temozolomide. Conceptually, the intrinsic inhibition of FTO by endogenous R-2HG may prevent rapid malignant progression of glioma precursors. On the translational level, these data suggest that “2HG-type molecules” or FTO inhibitors, primarily when combined with chemotherapy, may be attractive options for IDH wild-type tumors with FTO expression, whereas inhibitors of mutant IDH and FTO inhibitors may be attractive for IDH mutant tumors. Here, FTO inhibitors would oppose the activity of rebounded expression and function of FTO with the reduced 2-HG production.7 In summary, this report not only helps to improve the understanding of IDH biology in gliomas and offers an additional therapeutic target in these medically responsive tumor groups, but also defines options for potentially 2-HG-sensitive, IDH wild-type tumors that constitute the majority of gliomas. It proposes stage-dependent activity of IDH-derived 2-HG and opens a perspective to reconsider the consequences of IDH-mutation and also to address IDH wild-type disease.
Source: Neuro-Oncology
Institution: Center for Neuro-Oncology
Language: English
Subfield: Genetics
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4240548788