Logo: to the web site of Uppsala University

Glioblastoma is the most aggressive primary brain tumor in adults. Androgens are reported to influence the development of glioblastoma. Dihydrotestosterone (DHT) is formed from testosterone by action of the enzyme steroid 5α-reductase and is the most potent growth-inducing androgen metabolite. 24-Hydroxycholesterol, another steroid in the brain, is pivotal for brain cholesterol homeostasis and has been suggested to influence glioblastoma cells. However, a connection between 24-hydroxycholesterol and androgen metabolism related to glioblastoma has not previously been reported. The present study reports that human T98G glioblastoma cells metabolize testosterone into DHT, 3αandrostanediol and androstenedione. The 5α-reductase pathway converted testosterone to DHT and further to 3α-androstanediol. The 17β-hydroxysteroid dehydrogenase pathway metabolized testosterone to androstenedione. Results indicated that the 5α-reductase pathway is the major pathway for testosterone metabolism in this cell line. 24-Hydroxycholesterol significantly suppressed the conversion of testosterone to DHT and 3α-androstanediol, to a similar degree as the synthetic 5α-reductase inhibitor finasteride. Suppression of DHT formation resulted in increased metabolism to androstenedione. Similar effects on DHT formation were observed with the LXR agonist T0901317 as with 24-hydroxycho-lesterol. In addition, 24-hydroxycholesterol suppressed DHT formation in patient-derived primary GB cell lines U3009 and U3013, indicating that the observed connection between 24-hydroxycholesterol and androgen metabolism is not unique for T98G cells. Furthermore, 24-hydroxycholesterol-mediated suppression of DHT formation was also observed in human neuroblastoma SH-SY5Y cells. To summarize, the present data provide information on androgen metabolism in glioblastoma cells and indicate a previously unknown link between cholesterol homeostasis and growth-inducing androgens in glioblastoma and potentially other cell types.

The active form of vitamin D, 1,25-dihydroxyvitamin D₃, is known to act via VDR (vitamin D receptor), affecting several physiological processes. In addition, PDIA3 (protein disulphide-isomerase A3) has been associated with some of the functions of 1,25-dihydroxyvitamin D₃. In the present study we used siRNA-mediated silencing of PDIA3 in osteosarcoma and prostate carcinoma cell lines to examine the role(s) of PDIA3 for 1,25-dihydroxyvitamin D₃-dependent responses. PDIA3 silencing affected VDR target genes and significantly altered the 1,25-dihydroxyvitamin D₃-dependent induction of CYP24A1, essential for elimination of excess 1,25-dihydroxyvitamin D₃. Also, PDIA3 silencing significantly altered migration and proliferation in prostate PC3 cells, independently of 1,25-dihydroxyvitamin D₃. 1,25-Dihydroxyvitamin D₃ increased thermostability of PDIA3 in cellular thermal shift assay, supporting functional interaction between PDIA3 and 1,25-dihydroxyvitamin D₃-dependent pathways. In summary, our data link PDIA3 to 1,25-dihydroxyvitamin D₃-mediated signalling, underline and extend its role in proliferation and reveal a novel function in maintenance of 1,25-dihydroxyvitamin D₃ levels.

This study explored the signaling interplay between the vitamin D receptor (VDR) and receptor tyrosine kinases (RTKs). Epidermal growth factor (EGF) and platelet-derived growth factor (PDGF)-BB promotes cell proliferation in normal and cancer cells. At the same time, the active form of vitamin D (1,25(OH)₂-vitamin D₃) inhibits proliferation in some cells. Although EGF receptors (EGFR) and PDGF receptors (PDGFR) activate similar downstream pathways, we found that they interact with VDR signaling in distinct ways. We confirmed that 1,25(OH)₂-vitamin D₃ induces CYP24A1 gene expression in U2OS, T98G, and U251 cells. We found this to be potentiated when combined with EGF. In contrast, PDGF-BB did not impact 1,25(OH)₂-vitamin D₃-induced CYP24A1 expression in U2OS cells. The increase in CYP24A1 expression due to the combined action of EGF and 1,25(OH)₂-vitamin D₃ was dependent on AKT and ERK1/2 activation. Another VDR-responsive gene, CYP27B1, was unaffected by the addition of EGF, suggesting that EGF may have gene-specific effects on VDR signaling. While PDGF-BB did not influence CYP24A1 expression, 1,25(OH)₂-vitamin D₃ significantly influenced PDGF-BB-induced receptor phosphorylation and cell proliferation. In summary, we found that EGF, but not PDGF-BB, influenced the expression of the VDR-dependent gene CYP24A1, while 1,25(OH)₂-vitamin D₃ had an inhibitory effect on PDGFR signaling and proliferation. These findings highlight unique crosstalk between 1,25(OH)₂-vitamin D₃ signaling and EGF or PDGF-BB.

Platelet-derived growth factor (PDGF)-induced signalling via PDGF receptor β (PDGFRβ) leads to activation of downstream signalling pathways which regulate multiple cellular responses. It is unclear how PDGFRβ is degraded; both lysosomal and proteasomal degradation have been suggested. In this study, we have characterised the proteolytic cleavage of ligand-activated PDGFRβ, which results in two fragments: a larger fragment containing the extracellular domain, the transmembrane segment, and a part of the intracellular juxtamembrane region with a molecular mass of ∼130 kDa, and an intracellular ∼70 kDa fragment released into the cytoplasm. The proteolytic processing did not take place without internalisation of PDGFRβ. In addition, chelation of intracellular Ca²⁺ inhibited proteolytic processing. Inhibition of the proteasome affected signal transduction by increasing the phosphorylation of PDGFRβ, PLCγ, and STAT3 while reducing it on Erk1/2 and not affecting Akt. The proteolytic cleavage was observed in fibroblasts or cells that had undergone epithelial-mesenchymal transition.

The interplay between membrane subregions and receptor tyrosine kinases (RTK) will influence signaling in both normal and pathological RTK conditions. In this study, epidermal growth factor receptor (EGFR) and platelet-derived growth factor receptor β (PDGFR-β) internalizations were investigated by immunofluorescent microscopy following simultaneous treatment with EGF and PDGF-BB. We found that the two receptors utilize separate routes of internalization, which merges in a common perinuclear endosomal compartment after 45 min of stimulation. This is further strengthened when contrasting the recruitment of either EGFR or PDGFR-β to either clathrin or caveolin-1: PDGFR-β dissociates from caveolin-1 upon stimulation, and engages clathrin, whilst an increased recruitment of EGFR, to both clathrin and caveolin-1, was observed upon EGF stimulation. The association between EGFR and caveolin-1 is supported by the observation that EGFR was localized in lipid raft associated fractions, whereas PDGFR-β was not. We also found that disruption of lipid rafts using MβCD led to an increased EGFR dimerization and phosphorylation in response to ligand, as well as a dramatic decrease in AKT- and a smaller but robust decrease in ERK1/2 phosphorylation. This suggest that lipid rafts may be important to effectively connect the EGFR with downstream proteins to facilitate signaling. Our data implies that cholesterol depletion of the plasma membrane affect the signaling of EGFR and PDGFRβ differently.

It has become clear that lipid rafts functions as signaling hotspots connecting cell surface receptors to intracellular signaling pathways. However, the exact involvement of lipid rafts in receptor tyrosine kinase signaling is still poorly understood. In this study, we have analyzed platelet-derived growth factor (PDGF) receptor β (PDGFR-β) signaling in two different cell lines depleted of cholesterol, and as a consequence, disruption of lipid rafts. Cholesterol depletion of BJ-hTERT fibroblasts using methyl-β-cyclodextrin (MβCD) did not affect PDGFR-β activation as measured by its tyrosine phosphorylation. However, we did observe a small reduction in AKT phosphorylation and a more robust decrease of ERK1/2 activation. In contrast, in the osteosarcoma cell line U2OS, we noticed a deficient receptor activation. Interestingly, in U2OS cells, the ERK1/2 pathway was unaffected, but instead AKT and SRC signaling was reduced. These results suggest that cell type specific wiring of signaling pathways can lead to differential sensitivity to cholesterol depletion. Furthermore, MβCD treatment had a much more pronounced morphological effect on U2OS compared to BJ-hTERT cells. This is consistent with a previous report claiming that cancer cells are more sensitive to cholesterol depletion than normal cells. Our data supports the possibility that cholesterol lowering drugs may impede tumor growth.

The active hormonal form of vitamin D, 1 alpha,25-dihydroxyvitamin D-3, is reported to have 1000s of biological targets. The growth-suppressive properties of 1 alpha,25-dihydroxyvitamin D-3 and its synthetic analogs have attracted interest for the development of treatment and/or prevention of cancer. We examined effects of 1 alpha,25-dihydroxyvitamin D-3 and the vitamin D analog tacalcitol on signaling pathways and anchorage-independent growth in T98G and U251 glioblastoma cells. Assay of signaling proteins important for cellular growth indi-cated suppression of p70-S6 kinase levels by 1 alpha,25-dihydroxyvitamin D-3 and tacalcitol in T98G cells, whereas the levels of PLC gamma, a target for phospholipid signaling, was slightly increased. Activation of STAT3, an important regulator of malignancy, was suppressed by 1 alpha,25-dihydroxyvitamin D-3 and tacalcitol in T98G and U251 cells. However, despite the close structural similarity of these compounds, suppression was stronger by tacalcitol (1 alpha,24-dihydroxyvitamin D-3), indicating that even minor modifications of a vitamin D analog can impact its effects on signaling. Experiments using soft agar colony formation assay in T98G and U251 cells revealed significant suppression by 1 alpha,25-dihydroxyvitamin D-3 and tacalcitol on anchorage-independent growth, a property for cancer invasion and metastasis known to correlate with tumor-igenicity. These findings indicate that vitamin D and its analogs may be able to counteract the oncogenic transformation, invasion and metastatic potential of glioblastoma and prompt further study of these compounds in the development of improved therapy for brain cancer.