Hepatocyte growth factor (HGF) ligand and its receptor tyrosine kinase (RTK) mesenchymal-epithelial transition factor (MET) are important regulators of cellular processes such as proliferation, motility, angiogenesis, and tissue regeneration. In healthy adult somatic cells, this ligand and receptor pair is expressed at low levels and has little activity except when tissue injuries arise. In cancer cells, HGF/MET are often overexpressed, and this overexpression is found to correlate with tumorigenesis, metastasis, and poorer overall prognosis. This review focuses on the signaling of these molecules in the context of malignant brain tumors. RTK signaling pathways are among the most common and universally dysregulated pathways in gliomas. We focus on the role of HGF/MET in the following primary malignant brain tumors: astrocytomas, glioblastomas, oligodendrogliomas, ependymomas, and embryonal central nervous system tumors (including medulloblastomas and others). Brain metastasis, as well as current advances in targeted therapies, are also discussed.

Glioblastoma (GBM) is the most frequent and lethal primary malignant brain tumor. Despite decades of research, therapeutic advances that significantly prolong life are non-existent. In recent years, microRNAs (miRNAs) have been a focus of study in the pathobiology of cancer because of their ability to simultaneously regulate multiple genes. The aim of this study was to determine the functional and mechanistic effects of miR-3928 in GBM both in vitro and in vivo. To the best of our knowledge, this is the first article investigating the role of miR-3928 in GBM. We measured endogenous miR-3928 expression levels in a panel of patient-derived GBM tissue samples and cell lines. We found that GBM tissue samples and cell lines express lower levels of miR-3928 than normal brain cortex and astrocytes, respectively. Therefore, we hypothesized that miR-3928 is a tumor suppressive microRNA. We verified this hypothesis by showing that exogenous expression of miR-3928 has a strong inhibitory effect on both cell growth and invasiveness of GBM cells. Stable ex vivo overexpression of miR-3928 in GBM cells led to a reduction in tumor size in nude mice xenografts. We identified many targets (MDM2, CD44, DDX3X, HMGA2, CCND1, BRAF, ATOH8, and BMI1) of miR-3928. Interestingly, inhibition of the oncogene MDM2 also led to an upregulation of wild-type p53 expression and phosphorylation. In conclusion, we find that miR-3928, through the downregulation of several oncogenes and upregulation and activation of wild-type p53, is a strong tumor suppressor in GBM. Furthermore, the fact that miR-3928 can target many important dysregulated proteins in GBM suggests it might be a “master” regulatory microRNA that could be therapeutically exploited.

Solid tumors are often infiltrated with nerve endings. This infiltration promotes tumor growth and is correlated to a poor prognosis. Despite this well-observed phenomenon, the mechanisms underlying this process are not clear. In this paper, Amit et al. further elucidate the causal mechanisms of innervation in head and neck cancer and find that it involves an interplay between deficient p53, a canonical tumor suppressor gene, and secreted microRNA miR-34a. First, the authors looked for a link between p53 and neuritogenesis. Using The Cancer Genome Atlas, they analyzed survival data of oral cavity squamous cell carcinoma and found a correlation between poorer overall survival, high neural density in the oral cavity and the presence of TP53 abnormalities. To test this correlation experimentally, they co-cultured nerve cells with head and neck cancer cells that either expressed wild type (WT) or mutant p53 and observed that neurite outgrowth was significantly enhanced when p53 was either absent or mutated. This effect was also tested in vivo and showed a similar trend. In other words, only null or mutant p53 cells seemed to increase neuritogenesis. Second, the authors showed that the chemical cues that promoted neuritogenesis originated from the extracellular vesicles (EVs) secreted by head and neck cancer cells. To determine this, nerve cells were incubated with conditioned media collected from head and neck cancer cell lines with different p53 status (WT, null, or mutant). The conditioned media from the cells expressing the WT p53 gave rise to the least amount of neurofilament growth in nerve cell cultures. To confirm that the signals are indeed coming from secreted EVs, the authors deleted important GTPases involved in exocytosis and found that the mutant and null p53 head and neck cancer cells no longer promote neuritogenesis when their secretory pathways are blocked. This showed that neuritogenesis is enhanced specifically by chemical cues within secreted EVs. It is known that EVs contain microRNAs and that certain microRNAs are important regulators of neuronal development, growth, and function. Therefore, to determine which specific microRNAs are involved in neuritogenesis, the authors compared the microRNAs from the EVs of WT p53 cells to the EVs of deficient p53 cells and found 17 miRNAs that were differentially expressed. Further experiments narrowed this cohort to miR-34a and miR-141. When the antagomiRs of these two miRs were added to nerve cells, only the antagomiR of miR-34a was shown to have a significant effect on neurofilament growth. To distinguish between whether neurofilament growth arises from pre-existing filaments or occurs de novo, the authors first ablated adrenergic nerves in mice by chemical sympathectomy and then injected these mice with head and neck cancer cell xenografts. The results show no difference in the amount of nerve fibers in the ablated mice versus the controls. Furthermore, when the head and neck cancer cells were deficient in p53 (as compared to having WT p53), the number of nerve cell bodies markedly increased. Demonstrably, WT p53 cells can also grow de novo nerve cell bodies under the condition that miR-34a is first knocked down. Transcription factors POU5F1 and KLF4 are validated direct targets of miR-34a. Immunofluorescence experiments showed that both proteins are upregulated in mouse nerve cells when deficient p53 head and neck cancer cells were regionally inoculated. Together, these transcription factors are sufficient to promote neurofilament growth from mouse adult neural stem cells. Lastly, Kaplan-Meier curves of a cohort of patients being treated at MD Anderson Cancer Center show that nerve density alone is strong enough to predict the aggressiveness of oral cancer. Taken together, these results clarify the mechanisms of how innervation of solid tumors can promote tumorigenesis. The EVs secreted by deficient p53 head and neck cancer cells contain low amounts of miR-34a, which targets neuritogenesis pathways and inhibits growth of neurofilaments. In conclusion, this paper is informative and enhances the understanding of innervation mechanisms within solid tumors. However, more needs to be done to shed light on the precise molecular mechanisms of the link between deficient p53, miR-34a, and neuritogenesis.

Solid tumors are often infiltrated by immune cells that function to either support or debilitate tumor growth. The glioblastoma (GBM) tumor microenvironment contains a large number of macrophages known as tumor-associated macrophages (TAMs). Their presence has been associated with poor patient prognosis due to the supportive role they seem to play in tumor expansion and invasion. However, it is not clear what mediates the interactions between glioma cells and macrophages. In this paper, Chen et al. uncover the signaling pathway that connects glioblastoma cells to macrophages and find it to be associated with the loss of the tumor suppressor PTEN. By examining The Cancer Genome Atlas datasets, the authors observed an association between macrophage enrichment and GBMs that are specifically PTEN-mutated or deleted. To prove that this association is more than just a coincidence, they compared conditioned media collected from PTEN-knockout cells to conditioned media collected from PTEN-wildtype cells and showed that the former are capable of greatly and significantly enhancing macrophage migration in transwell assays. Furthermore, this enhancing effect can be reversed if PTEN is re-expressed in PTEN-deficient cell lines. To find out what is secreted into the media by cells with a deficient PTEN that is causing macrophages to migrate, they performed a set of screening experiments using a secreted protein database, qRT-PCR, and immunoblot analysis. They discovered that the secreted chemoattractant is lysyl oxidase (LOX), which was vastly and specifically secreted by PTEN-deficient glioma cells. When LOX is added to regular growth media, macrophage migration is enhanced. Next, using biochemical (immunoblots, qRT-PCR, transwell migration assays, and immunofluorescence) and in silico (gene set enrichment analysis) experiments, Chen et al. established the following link between PTEN-deficiency, LOX secretion, macrophage recruitment, and tumor growth. YAP1, a transcriptional coactivator, is activated in PTEN-deficient glioma cells as a consequence of upregulated SRC and AKT. YAP1 activates LOX expression. Secreted LOX from GBM cells binds to surface integrins on macrophages and is internalized. Once inside the macrophage, the PYK2 pathway is activated, leading to macrophage migration. SPP1 is also secreted by the macrophages upon LOX uptake and migration. SPP1 is a factor that promotes angiogenesis and tumor survival. Based on these findings, the authors tested the efficacy of LOX inhibition on tumor growth and survival in various GBM models. They reasoned that if LOX is inhibited, which can be easily achieved through a variety of techniques (shRNA, small molecule, or antibody), some of which can even cross the blood and brain barrier, macrophages will not be recruited and therefore, the tumor will lose this crucial support and slow in growth. This was indeed the case. Both immune deficient as well as immune intact murine models showed a decrease in tumor growth and exhibited better survival outcomes when LOX was inhibited. In summary, this interesting and well-performed study identifies key macrophage-cancer cell interdependencies, uncovers the underlying mechanism, and validates LOX inhibition as a promising therapeutic strategy in PTEN-deficient GBM.