Identification of Rac guanine nucleotide exchange factors promoting Lgl1 phosphorylation in glioblastoma.

The protein Lgl1 is a key regulator of cell polarity. We previously showed that Lgl1 is inactivated by hyperphosphorylation in glioblastoma as a consequence of PTEN tumour suppressor loss and aberrant activation of the PI 3-kinase pathway; this contributes to glioblastoma pathogenesis both by promoting invasion and repressing glioblastoma cell differentiation. Lgl1 is phosphorylated by atypical protein kinase C that has been activated by binding to a complex of the scaffolding protein Par6 and active, GTP-bound Rac. The specific Rac guanine nucleotide exchange factors that generate active Rac to promote Lgl1 hyperphosphorylation in glioblastoma are unknown. We used CRISPR/Cas9 to knockout PREX1, a PI 3-kinase pathway-responsive Rac guanine nucleotide exchange factor, in patient-derived glioblastoma cells. Knockout cells had reduced Lgl1 phosphorylation, which was reversed by re-expressing PREX1. They also had reduced motility and an altered phenotype suggestive of partial neuronal differentiation; consistent with this, RNA-seq analyses identified sets of PREX1-regulated genes associated with cell motility and neuronal differentiation. PREX1 knockout in glioblastoma cells from a second patient did not affect Lgl1 phosphorylation. This was due to overexpression of a short isoform of the Rac guanine nucleotide exchange factor TIAM1; knockdown of TIAM1 in these PREX1 knockout cells reduced Lgl1 phosphorylation. These data show that PREX1 links aberrant PI 3-kinase signaling to Lgl1 phosphorylation in glioblastoma, but that TIAM1 is also to fill this role in a subset of patients. This redundancy between PREX1 and TIAM1 is only partial, as motility was impaired in PREX1 knockout cells from both patients.

Engineering PTEN-L for Cell-Mediated Delivery.

The tumor suppressor PTEN is frequently inactivated in glioblastoma. PTEN-L is a long form of PTEN produced by translation from an alternate upstream start codon. Unlike PTEN, PTEN-L has a signal sequence and a tract of six arginine residues that allow PTEN-L to be secreted from cells and be taken up by neighboring cells. This suggests that PTEN-L could be used as a therapeutic to restore PTEN activity. However, effective delivery of therapeutic proteins to treat CNS cancers such as glioblastoma is challenging. One method under evaluation is cell-mediated therapy, where cells with tumor-homing abilities such as neural stem cells are genetically modified to express a therapeutic protein. Here, we have developed a version of PTEN-L that is engineered for enhanced cell-mediated delivery. This was accomplished by replacement of the native leader sequence of PTEN-L with a leader sequence from human light-chain immunoglobulin G (IgG). This version of PTEN-L showed increased secretion and an increased ability to transfer to neighboring cells. Neural stem cells derived from human fibroblasts could be modified to express this version of PTEN-L and were able to deliver catalytically active light-chain leader PTEN-L (lclPTEN-L) to neighboring glioblastoma cells.

PREX1 integrates G protein-coupled receptor and phosphoinositide 3-kinase signaling to promote glioblastoma invasion.

A defining feature of the brain cancer glioblastoma is its highly invasive nature. When glioblastoma cells are isolated from patients using serum free conditions, they accurately recapitulate this invasive behaviour in animal models. The Rac subclass of Rho GTPases has been shown to promote invasive behaviour in glioblastoma cells isolated in this manner. However the guanine nucleotide exchange factors responsible for activating Rac in this context have not been characterized previously. PREX1 is a Rac guanine nucleotide exchange factor that is synergistically activated by binding of G protein αγ subunits and the phosphoinositide 3-kinase pathway second messenger phosphatidylinositol 3,4,5 trisphosphate. This makes it of particular interest in glioblastoma, as the phosphoinositide 3-kinase pathway is aberrantly activated by mutation in almost all cases. We show that PREX1 is expressed in glioblastoma cells isolated under serum-free conditions and in patient biopsies. PREX1 promotes the motility and invasion of glioblastoma cells, promoting Rac-mediated activation of p21-associated kinases and atypical PKC, which have established roles in cell motility. Glioblastoma cell motility was inhibited by either inhibition of phosphoinositide 3-kinase or inhibition of G protein ßγ subunits. Motility was also inhibited by the generic dopamine receptor inhibitor haloperidol or a combination of the selective dopamine receptor D2 and D4 inhibitors L-741,626 and L-745,870. This establishes a role for dopamine receptor signaling via G protein ßγ subunits in glioblastoma invasion and shows that phosphoinositide 3-kinase mutations in glioblastoma require a context of basal G protein-coupled receptor activity in order to promote this invasion.

PKCι depletion initiates mitotic slippage-induced senescence in glioblastoma.

Cellular senescence is a tumor suppressor mechanism where cells enter a permanent growth arrest following cellular stress. Oncogene-induced senescence (OIS) is induced in non-malignant cells following the expression of an oncogene or inactivation of a tumor suppressor. Previously, we have shown that protein kinase C iota (PKCι) depletion induces cellular senescence in glioblastoma cells in the absence of a detectable DNA damage response. Here we demonstrate that senescent glioblastoma cells exhibit an aberrant centrosome morphology. This was observed in basal levels of senescence, in p21-induced senescence, and in PKCι depletion-induced senescence. In addition, senescent glioblastoma cells are polyploid, Ki-67 negative and arrest at the G1/S checkpoint, as determined by expression of cell cycle regulatory proteins. These markers are all consistent with cells that have undergone mitotic slippage. Failure of the spindle assembly checkpoint to function properly can lead to mitotic slippage, resulting in the premature exit of mitotic cells into the G1 phase of the cell cycle. Although in G1, these cells have the replicated DNA and centrosomal phenotype of a cell that has entered mitosis and failed to divide. Overall, we demonstrate that PKCι depletion initiates mitotic slippage-induced senescence in glioblastoma cells. To our knowledge, this is the first evidence of markers of mitotic slippage directly in senescent cells by co-staining for senescence-associated ß-galactosidase and immunofluorescence markers in the same cell population. We suggest that markers of mitotic slippage be assessed in future studies of senescence to determine the extent of mitotic slippage in the induction of cellular senescence.

Inhibition of glioblastoma malignancy by Lgl1.

lethal giant larvae (lgl) was first identified as a tumor suppressor in Drosophila, where its loss repressed the differentiation and promoted the invasion of neuroblasts, the Drosophila equivalent of the neural stem cell. Recently we have shown that a human homolog of Lgl, Lgl1 (LLGL1), is constitutively phosphorylated and inactivated in glioblastoma cells; this occurs as a downstream consequence of PTEN loss, one of the most frequent genetic events in glioblastoma. Here we have investigated the consequences of this loss of functional Lgl1 in glioblastoma in vivo. We used a doxycycline-inducible system to express a non-phosphorylatable, constitutively active version of Lgl1 (Lgl3SA) in either a glioblastoma cell line or primary glioblastoma cells isolated under neural stem cell culture conditions from patients. In both types of cells, expression of Lgl3SA, but not wild type Lgl1, inhibited cell motility in vitro. Induction of Lgl3SA in intracerebral xenografts markedly reduced the in vivo invasion of primary glioblastoma cells. Lgl3SA expression also induced the differentiation of glioblastoma cells in vitro and in vivo along the neuronal lineage. Thus the central features of Lgl function as a tumor suppressor in Drosophila are conserved in human glioblastoma.

PTEN loss represses glioblastoma tumor initiating cell differentiation via inactivation of Lgl1.

Glioblastoma multiforme is an aggressive and incurable type of brain tumor. A subset of undifferentiated glioblastoma cells, known as glioblastoma tumor initiating cells (GTICs), has an essential role in the malignancy of this disease and also appears to mediate resistance to radiation therapy and chemotherapy. GTICs retain the ability to differentiate into cells with reduced malignant potential, but the signaling pathways controlling differentiation are not fully understood at this time. PTEN loss is a very common in glioblastoma multiforme and leads to aberrant activation of the phosphoinositide 3-kinase pathway. Increased signalling through this pathway leads to activation of multiple protein kinases, including atypical protein kinase C. In Drosophila, active atypical protein kinase C has been shown to promote the self-renewal of neuroblasts, inhibiting their differentiation along a neuronal lineage. This effect is mediated by atypical protein kinase c-mediated phosphorylation and inactivation of Lgl, a protein that was first characterized as a tumour suppressor in Drosophila. The effects of the atypical protein kinase C/Lgl pathway on the differentiation status of GTICs, and its potential link to PTEN loss, have not been assessed previously. Here we show that PTEN loss leads to the phosphorylation and inactivation of Lgl by atypical protein kinase C in glioblastoma cells. Re-expression of PTEN in GTICs promoted their differentiation along a neuronal lineage. This effect was also seen when atypical protein kinase C was knocked down using RNA interference, and when a non-phosphorylatable, constitutively active form of Lgl was expressed in GTICs. Thus PTEN loss, acting via atypical protein kinase C activation and Lgl inactivation, helps to maintain GTICs in an undifferentiated state.

Coordination of glioblastoma cell motility by PKCι.

BACKGROUND: Glioblastoma is one of the deadliest forms of cancer, in part because of its highly invasive nature. The tumor suppressor PTEN is frequently mutated in glioblastoma and is known to contribute to the invasive phenotype. However the downstream events that promote invasion are not fully understood. PTEN loss leads to activation of the atypical protein kinase C, PKCι. We have previously shown that PKCι is required for glioblastoma cell invasion, primarily by enhancing cell motility. Here we have used time-lapse videomicroscopy to more precisely define the role of PKCι in glioblastoma. RESULTS: Glioblastoma cells in which PKCι was either depleted by shRNA or inhibited pharmacologically were unable to coordinate the formation of a single leading edge lamellipod. Instead, some cells generated multiple small, short-lived protrusions while others generated a diffuse leading edge that formed around the entire circumference of the cell. Confocal microscopy showed that this behavior was associated with altered behavior of the cytoskeletal protein Lgl, which is known to be inactivated by PKCι phosphorylation. Lgl in control cells localized to the lamellipod leading edge and did not associate with its binding partner non-muscle myosin II, consistent with it being in an inactive state. In PKCι-depleted cells, Lgl was concentrated at multiple sites at the periphery of the cell and remained in association with non-muscle myosin II. Videomicroscopy also identified a novel role for PKCι in the cell cycle. Cells in which PKCι was either depleted by shRNA or inhibited pharmacologically entered mitosis normally, but showed marked delays in completing mitosis. CONCLUSIONS: PKCι promotes glioblastoma motility by coordinating the formation of a single leading edge lamellipod and has a role in remodeling the cytoskeleton at the lamellipod leading edge, promoting the dissociation of Lgl from non-muscle myosin II. In addition PKCι is required for the transition of glioblastoma cells through mitosis. PKCι therefore has a role in both glioblastoma invasion and proliferation, two key aspects in the malignant nature of this disease.

Interaction of Hsp90 with the nascent form of the mutant epidermal growth factor receptor EGFRvIII.

EGFRvIII is a mutant epidermal growth factor that promotes aggressive growth of glioblastomas. We made a plasmid that directed the expression of an EGFRvIII with three copies of the Flag epitope at its amino terminus. Flag-tagged EGFRvIII was expressed at the same levels as unmodified EGFRvIII, and showed the same subcellular localization. However, the Flag epitope could only be detected on EGFRvIII present in the endoplasmic reticulum; the epitope was covalently modified during trafficking of the receptor through the Golgi so that it was no longer recognized by anti-Flag antibody. This property was exploited to selectively purify nascent EGFRvIII from glioblastoma cells. Nascent EGFRvIII was found to copurify with a set of other proteins, identified by mass spectrometry as the two endoplasmic reticulum chaperones Grp94 and BiP, and the two cytosolic chaperones Hsc70 and Hsp90. The Hsp90-associated chaperone Cdc37 also co-purified with EGFRvIII, suggesting that Hsp90 binds EGFRvIII as a complex with this protein. Geldanamycin and radicicol, two chemically unrelated inhibitors of Hsp90, decreased the expression of EGFRvIII in glioblastoma cells. These studies show that nascent EGFRvIII in the endoplasmic reticulum associates with Hsp90 and Cdc37, and that the Hsp90 association is necessary to maintain expression of EGFRvIII.

Induction of apoptosis in glioblastoma cells by an atypical protein kinase C pseudosubstrate peptide.

BACKGROUND: Glioblastoma responds poorly to standard chemotherapy agents. The expression of a mutant, constitutively-active EGF receptor (EGFRvIII) is common in glioblastoma and contributes to chemotherapy resistance. We have assessed the cytotoxicity of an inhibitor of atypical protein kinase C on glioblastoma cells expressing EGFRvIII. MATERIALS AND METHODS: Glioblastoma cells were treated with a peptide-based atypical protein kinase C inhibitor. Apoptosis was assessed by morphological criteria, TUNEL assays, annexin V staining, Hoechst staining and colorimetric assays for cell viability. RESULTS: The atypical protein kinase C inhibitor induced rapid apoptosis in glioblastoma cells expressing EGFRvIII and killed these cells with an IC50 of 16 microM. Glioblastoma cells which do not express EGFRvIII were less sensitive. Apoptosis was not affected by caspase inhibitors and occurred without detectable caspase activation. CONCLUSION: An atypical protein kinase C inhibitor induces rapid apoptosis in glioblastoma cells by a caspase-independent mechanism that is enhanced, rather than inhibited, by EGFRvIII.