Differential SKIP expression in PTEN-deficient glioblastoma regulates cellular proliferation and migration.

Glioblastoma is the most common and lethal primary malignant brain tumor in adults. The tumor suppressor gene PTEN is deleted, mutated or hypermethylated in more than 60% of glioblastoma cases resulting in hyperactivation of the phosphoinositide 3-kinase pathway, which leads to sustained PI(3,4,5)P3 signaling, and thereby hyperactivation of Akt and other effectors. PI(3,4,5)P3 is also hydrolyzed to PI(3,4)P2 by inositol polyphosphate 5-phosphatases such as SKIP, but the role this pathway has in glioblastoma is unknown. Microarray expression profiling of SKIP in human glioblastoma has revealed both increased and decreased SKIP gene expression. Here we have screened PTEN-deficient glioblastoma for SKIP protein expression by immunohistochemistry and report that SKIP expression is increased in some cases or decreased relative to normal brain. Using the U-87MG PTEN-deficient cell line we show that SKIP knockdown did not further enhance cell proliferation or survival. However, SKIP overexpression in U-87MG cells suppressed anchorage-independent cell growth and growth factor-induced PI(3,4,5)P3/Akt signaling. Although, SKIP knockdown did not affect cell proliferation or survival, cell migration was significantly retarded, associated with significantly increased PI(4,5)P2 signals, and decreased phosphorylation of the actin-regulatory protein cofilin, a PI(4,5)P2-binding protein. Notably, overexpression of SKIP also inhibited migration of U-87MG cells to a similar degree as observed with PTEN reconstitution, however, via distinct mechanisms. PTEN reconstitution promoted sustained lamellipodia generation and focal adhesion formation. In contrast, SKIP overexpression reduced sustained lamellipodia formation, talin incorporation into focal adhesions and recruitment of PI(4,5)P2-binding proteins to the plasma membrane. Notably, analysis of two independent ONCOMINE microarray data sets revealed a significant correlation between increased SKIP mRNA expression in glioblastoma and improved long-term survival. Therefore, SKIP expression in glioblastoma may affect the local invasion of PTEN-deficient tumors.

Chloromethyl-X-rosamine (MitoTracker Red) photosensitises mitochondria and induces apoptosis in intact human cells.

We report that chloromethyl-X-rosamine (MitoTracker Red), a mitochondrion-selective fluorescent probe, has a strong photosensitising action. Photoirradiation of intact cells loaded with chloromethyl-X-rosamine induces depolarisation of the inner mitochondrial membrane and swelling of mitochondria, subsequently resulting in apoptosis. We have studied human osteosarcoma 143B TK-(rho+) cells and the derived (rho)0 206 cell line devoid of mitochondrial DNA. Colony formation tests revealed that chloromethyl-X-rosamine itself has no toxicity to either cell line in the concentration range 100-250 nM (unless photoirradiated). Chloromethyl-X-rosamine has potent phototoxicity such that almost quantitative cell killing was achieved at light doses of >2 J/cm2. These photodamaged cells initially showed swollen degenerative mitochondria and, later, uptake of propidium iodide in their apoptotic nuclei was observed. When cells were loaded with chloromethyl-X-rosamine (100 nM) and imaged by laser scanning confocal microscopy, photoirradiation by the laser beam under routine scanning conditions was sufficient to induce mitochondrial damage in both cell lines. This was evidenced by a rapid decrease of fluorescence intensity of co-loaded rhodamine 123 (indicative of mitochondrial depolarisation). Globular swelling of mitochondria took place within 15 minutes, imaged by the residual fluorescence of chloromethyl-X-rosamine itself, which also markedly decreased in intensity after imaging. Mitochondrial membrane depolarisation of cells loaded with chloromethyl-X-rosamine after photoirradiation using a measured dose of visible light was independently confirmed in 143B TK- and (rho)0 206 cells, by the significant decrease in uptake into cells of [3H]methyltriphenylphosphonium ions. Photoactivation of chloromethyl-X-rosamine in 143B TK-(rho+) cells, whose mitochondria had previously been loaded with calcein, caused rapid release of the mitochondrially entrapped calcein into the cytosol and nucleus. This major change in permeability of the mitochondrial inner membrane could not be prevented by cyclosporin A. Immunohistochemical study of cytochrome c revealed its diffuse redistribution into the cytoplasm in chloromethyl-X-rosamine-loaded cells after irradiation, as opposed to its specific mitochondrial localisation in non-irradiated cells. As a photosensitiser specifically targeted to mitochondria, and also a reporter of membrane potential and morphology, chloromethyl-X-rosamine may provide versatile new applications in studies of mitochondrial roles in cell death.

Photosensitisation properties of mitochondrially localised green fluorescent protein.

The photosensitisation properties of a mitochondrially localised green fluorescent protein (GFP) variant were established in cultured monkey kidney cells. We first cloned into a mammalian expression vector the thermostable variant GFP5, fused at its N-terminus to the 16-amino acid mitochondrial targeting sequence of human 3-oxoacyl-CoA thiolase. The recombinant plasmid thus constructed, pCZ34, was transfected into COS7 cells, under conditions for transient expression. GFP5 was shown to have a mitochondrial localisation by confocal microscopy, confirmed by its almost identical pattern with that of MitoTracker Red which selectively stains mitochondria. After photoirradiation of such transfected COS7 cells with 390-570 nm light to excite the mitochondrially localised GFP5, significant cell killing was observed. DAPI-staining of the dead cells detached from the growth support revealed cells to have undergone apoptosis. Thus, GFP5 can be used in the development of an organelle-specific photosensitiser since it can be targeted to different subcellular locations by protein engineering of the signal sequences.

Mitochondria are the functional intracellular target for a photosensitizing boronated porphyrin.

A photosensitizing boron-containing porphyrin derivative denoted BOPP, which is selectively localised into mitochondria, has been tested on Namalwa cells, in each of two genetic configurations: rho+ cells containing normal mtDNA and mitochondrial respiratory functions, or rho0 cells lacking mtDNA and devoid of mitochondrial oxidative phosphorylation. After short-term cellular uptake for 18 h, BOPP (30 micrograms/ml) was not cytotoxic, but did show marked phototoxicity in Namalwa rho+ cells, concomitant with substantial reduction of mitochondrial respiratory activity. After long-term (3 days or more) exposure to BOPP without light, growth of Namalwa rho+ cells was inhibited at concentrations significantly above 30 micrograms/ml. At such concentrations BOPP was shown to have direct inhibitory effects on mitochondrial azide-sensitive respiration of p+ cells. By contrast, BOPP showed neither cytotoxic nor phototoxic effects in rho0 cells. These results indicate functional mitochondria to be a major cellular target in vivo after BOPP uptake and photoactivation.

Polymorphonuclear neutrophils release 35S-labelled proteoglycans into cartilage during frustrated phagocytosis.

Rabbit peritoneal polymorphonuclear neutrophils (PMN), incubated in medium containing [35S]sulphate, incorporated 35S into proteoglycan and protein fractions. Approximately 46% of the 35S-labelled macromolecules associated with the PMN cells after 1 h of incubation were recovered in a cytoplasmic granule extract, the majority being present in azurophil granules. Analysis of the azurophil granule fraction showed that approximately 90% of the 35S-labelled macromolecules were proteoglycans. When challenged with heat-aggregated rabbit gamma-globulin in the presence of cytochalasin B and cGMP, PMN were induced to release granular enzymes but did not release 35S-labelled proteoglycans into the incubation medium. When incubated with articular cartilage slices, PMN released their granule 35S-labelled proteoglycan into the medium and into the cartilage matrix. Granule enzymes and 35S-labelled granule proteoglycan were extracted from the cartilage tissue after incubation and 35S-labelled macromolecules were detected in the cartilage tissue by autoradiography.

Synthesis of 35S-labelled macromolecules by polymorphonuclear neutrophils. Evidence for the production of [35S]sulphite which can modify both endogenous and exogenous proteins.

The incorporation of [35S]sulphate into macromolecules by rabbit peritoneal polymorphonuclear neutrophils (PMN) in vitro revealed that two major groups of 35S-labelled macromolecules were synthesized by these cells. The first group did not bind to anion-exchange columns at pH 6.0 and contained 60-80% of the total incorporated radiolabel. The second group did bind to anion-exchange columns at pH 6.0 and eluted as a single peak of radioactivity at an ionic strength characteristic of sulphated proteoglycans; it accounted for the remaining incorporated radiolabel. Analysis of this material on Sepharose CL-6B demonstrated that 35S-labelled macromolecules isolated from the cell extract migrated with Kav. of 0.36, while corresponding material isolated from the medium migrated with Kav. of 0.51. When subjected to electrophoresis on SDS/polyacrylamide gels the intact proteoglycan had a molecular mass of approx. 90 kDa and yielded two core proteins of molecular mass 31 kDa and 28 kDa after digestion with chondroitinase ABC. The peak of labelled macromolecules which did not bind to the anion-exchange column was found, by SDS/PAGE, to comprise 35S-labelled proteins of various molecular masses. The 35S label was displaced from this fraction by treatment with 0.1 M-sodium sulphite, suggesting that the radiolabel was in the form of an S-sulpho sulphite derivative. Using the sulphite-trapping agents N-2,4-dinitroanilinomaleimide and cyst(e)ine, [35S]sulphite was detected in the incubation medium of PMN, indicating that these cells were able to synthesize [35S]sulphite from [35S]sulphate. The release of [35S]sulphite from neutrophil cultures was calculated to be 78 pmol/h per 10(6) cells. When exogenous proteins were included in the incubation medium of cell cultures, the [35S]sulphite reacted with these proteins to form a stable 35S-labelled conjugate.

The role of serine proteinase in cartilage damage.

Arthritic cartilage from experimental animals has been shown to have a decreased proteoglycan content, a decreased rate of proteoglycan synthesis, and a marked increase in an active serine proteinase when compared with normal articular cartilage. The serine proteinase is transferred from PMN cells into cartilage during the inflammatory response where it increased the rate of proteoglycan degradation and is eventually removed by interaction with the chondrocyte surface. The interaction also results in an inhibition of proteoglycan synthesis by the chondrocytes. Both these factors contribute to the loss of proteoglycan from arthritic cartilage.

Carrageenin-induced arthritis. VI. Alterations in amino acid transport by articular cartilage in acute inflammatory arthritis.

The mechanism of transport of alanine and aminoisobutyric acid into chondrocytes in rabbit articular cartilage was shown to be mediated by transport systems similar to that described for other eukaryotic cells namely the A, ASC, and L systems. Three days after the initiation of an acute inflammatory arthritis by the intra-articular injection of carrageenin into one knee joint the rate of transport of both these amino acids was decreased. Although all three transport systems were depressed, it appeared that the A and ASC systems were partially susceptible to damage by the induced inflammation. The rate of amino acid transport by the affected cartilage had recovered by 28 days after carrageenin treatment. This depression in amino acid transport is discussed in relation to a decrease in general metabolic processes in chondrocytes as a consequence of inflammation.

Carrageenin-induced arthritis: V. A morphologic study of the development of inflammation in acute arthritis.

Following a single injection of the polysaccharide carrageenin into the rabbit knee joint, a rapid inflammatory process occurs in the joint space and synovial membrane, followed by changes in the articular cartilage. Initially there is an influx of cells, mainly PMNs, into the synovial fluid, accompanied by proliferation of the synovial lining cells and infiltration of the synovial membrane. The numbers of synovial fluid cells decline gradually after 24 hr. The reaction in the synovial membrane is greatest at day 7, and inflammation is still evident at day 21. Initially, the infiltrate consists mainly of PMNs, but by day 7 it is predominantly mononuclear, with small clusters of lymphocytes. The articular cartilage shows loss of metachromasia with toluidine blue at 3-14 days after injection, but stains normally after day 21. Electron microscopy shows damage to the chondrocytes at day 1 and 7, with complete destruction of cells in the surface layer. At day 7 cells in the deeper layers have lost the apparatus required for proteoglycan synthesis, but at day 21 the cells appear virtually normal. There was no evidence for a direct inhibitory effect of carrageenin on proteoglycan biosynthesis. Most labeled carrageenin was rapidly cleared from the joint space, but about 10% was retained in the synovial membrane and 0.6% in articular cartilage at 48 hr after injection. Since the increase and decline in PMN numbers respectively precede the cartilage damage and recovery, it is suggested that there may be a correlation between the clinical activity of arthritis and the number of PMNs in the synovial fluid.

Evidence for polymorphonuclear-leucocyte-derived proteinases in arthritic cartilage.

1. An enzyme that degrades proteoglycan at neutral pH was extracted with 4 M-guanidine hydrochloride from the articular cartilage of rabbits with antigen-induced arthritis. 2. The enzyme had an apparent molecular weight on Ultrogel AcA 54 of about 8000 and was optimally active at pH 7.5 in Tris/HCl buffer containing 0.2 M-NaCl. The partially purified preparation was totally inhibited by 0.01 mM-N-acetyldialanylprolylvalylchloromethane, severely inhibited by 2 mM-phenylmethanesulphonyl fluoride and soya-bean trypsin inhibitor (200 microgram/ml) and slightly inhibited by 10 mM-EDTA. Marked inhibition was also obtained with a cytosolic fraction prepared from rabbit polymorphonuclear leucocytes. 3. All properties of the enzyme were virtually identical with those of an 'elastase-like' proteinase that was isolated from rabbit polymorphonuclear-leucocyte granules. 4. The results are consistent with the idea that cartilage proteoglycan degradation in acute joint inflammation is due at least partly to the diffusion into the cartilage of proteinases derived from synovial-fluid polymorphonuclear leucocytes.