Synthesis and cellular studies of an octa-anionic 5,10,15,20-tetra[3,5-(nido-carboranylmethyl)phenyl]porphyrin (H(2)OCP) for application in BNCT

The total synthesis of 5,10,15,20-tetra[3,5-(carboranylmethyl)phenyl]porphyrins 2-5 containing 36-43% boron by weight are reported. All compounds were characterized by spectroscopic methods and, in the case of 2, by X-ray crystallography. The water-soluble nido-carboranylporphyrin 5 (H(2)OCP) was found to have low dark toxicity toward V79 lung fibroblasts (CS(50) > or = 250 microM), to be readily taken up by human glioblastoma T98G cells in culture and to localize subcellularly preferentially in the cell lysosomes. In comparison with a known tetra(nido-carboranyl)porphyrin (6), H(2)OCP (5) is taken up slower and to a lower extent by T98G cells, possibly as a result of its higher hydrophilic character. The metal-free H(2)OCP (5) was also found to accumulate to a higher extent in T98G cells compared with its zinc(II) complex analog 4. Our studies show that carboranylporphyrins bearing eight nido-carborane cages can still accumulate intracellularly and have low dark toxicity toward cells in culture, and therefore might have promise for application in BNCT.     

Downregulation of the RECK-tumor and metastasis suppressor gene in glioma invasiveness

Invasive behavior is the pathological hallmark of malignant gliomas, being responsible for the failure of surgery, radiation, and chemotherapy. Matrix metalloproteinases (MMPs) are essential for proper ECM remodeling and invasion. The tumor and metastasis suppressor RECK protein regulates at least three members of the MMPs family: MMP-2, MMP-9, and MT1-MMP. In order to mimic the in vivo invasion process, A172 and T98G, respectively, non-invasive and invasive human glioblastoma cell lines, were cultured onto uncoated (control) or type I collagen gel-coated surface, and maintained for up to 7 days to allow establishment of the invasive process. We show that the collagen substrate causes decreased growth rates and morphological alterations correlated with the invasive phenotype. Electronic transmission microscopy of T98G cells revealed membrane invaginations resembling podosomes, which are typically found in cells in the process of crossing tissue boundaries, since they constitute sites of ECM degradation. Real time PCR revealed higher RECK mRNA expression in A172 cells, when compared to T98G cells and, also, in samples obtained from cultures where the invasive process was fully established. Interestingly, the collagen substrate increases RECK expression in A172 cells and the same tendency is displayed by T98G cells. MMPs-2 and -9 displayed higher levels of expression and activity in T98G cells, and their activities are also upregulated by collagen. Therefore, we suggest that: (1) RECK downregulation is critical for the invasiveness process displayed by T98G cells; (2) type 1 collagen could be employed to modulate RECK expression in glioblastoma cell lines. Since a positive correlation between RECK expression and patients survival has been noted in several types of tumors, our results may contribute to elucidate the complex mechanisms of malignant gliomas invasiveness.     

T98G: an anchorage-independent human tumor cell line that exhibits stationary phase G1 arrest in vitro.

T98 and T98G are two related cell lines that were derived from a human glioblastoma multiforma tumor. T98G has almost twice as many chromosomes as T98, suggesting that it is a polyploid variant of T98. Three aspects of control of cellular proliferation were studied in T98 and T98G cells in comparison to WI-38 normal human diploid cells. WI-38 cells have the following properties: (1) they can undergo only a limited number of population doublings in vitro; (2) they cannot proliferate without anchorage; and (3) they become arrested in G1 phase under stationary phase conditions. T98 cells differ from normal cells in all three of these properties, as do many other transformed cell lines. However, the derivative of T98, namely T98G, expresses an unique combination of normal and transformed aspects of the control of cellular proliferation. T98G cells are like normal cells in that they become arrested in G1 phase under stationary phase conditions, yet they also exhibit the transformed characteristics of anchorage independence and immortality. Thus, T98G cells demonstrate that transformation to immortality and anchorage independence can exist without concomitant loss of the normal mechanism for G1 arrest in response to stationary phase conditions. This result supports the hypothesis that each of these three aspects of control of cellular proliferation can be altered independently. Partially transformed cell lines, such as T98G, should be useful for sorting out the biochemical changes associated with transformation in each of these aspects.     

Suppression of cytokine-induced expression of adrenomedullin and endothelin-1 by dexamethasone in T98G human glioblastoma cells

There is accumulating evidence showing that glial cells and gliomas secrete some neuropeptides and vasoactive peptides, such as adrenomedullin and endothelin-1. We have previously shown that expression of these two peptides is induced by inflammatory cytokines in T98G human glioblastoma cells. Glucocorticoids are frequently used for the treatment of inflammatory diseases and glioblastomas. We therefore studied effects of dexamethasone on expression of adrenomedullin and endothelin-1 in T98G human glioblastoma cells. Dexamethasone dose-dependently increased adrenomedullin mRNA levels and immunoreactive-adrenomedullin levels in the medium in T98G cells, whereas it decreased immunoreactive-endothelin levels in the medium. A combination of three cytokines, interferon-gamma (100 U/ml), tumor necrosis factor-alpha (20 ng/ml) and interleukin-1beta (10 ng/ml) induced expression of adrenomedullin and endothelin-1 in T98G cells. Dexamethasone (10(-8) mol/l) suppressed increases in expression of both adrenomedullin and endothelin-1 induced by these three cytokines. Thus, dexamethasone alone increased adrenomedullin expression whereas it suppressed the cytokine-induced expression of adrenomedullin in T98G cells. These findings raised the possibility that effects of dexamethasone on brain inflammation and glioblastomas may be partly mediated or modulated by its effects on expression of adrenomedullin and endothelin-1.     

Selective potentiation of drug cytotoxicity by NSAID in human glioma cells: the role of COX-1 and MRP.

Here, we report that nonsteroidal anti-inflammatory drugs (NSAID) enhance the cytotoxic effects of doxorubicin and vincristine in T98G human malignant glioma cells. The cytotoxicity of BCNU, cisplatin, VM26, camptothecin, and cytarabine is unaffected by NSAID. No free radical formation is induced by doxorubicin or vincristine in the absence or presence of NSAID. Doxorubicin and vincristine cytotoxicity in the absence or presence of NSAID are unaffected by free radical scavengers. Functional inhibitors of phospholipase A2 (PLA2), such as dexamethasone and quinacrine, do not mimick the effects of NSAID. T98G cells, but not LN-18, LN-229, LN-308, or A172 glioma cells, express cyclooxygenase (COX-1) and NSAID do not modulate drug cytotoxicity in the other cell lines, except T98G. Thus, augmentation of doxorubicin and vincristine cytotoxicity by NSAID correlates with COX-1 expression. However, ectopic expression of COX-1 in LN-229 cells does not induce the phenotype of T98G cells, indicating that COX-1 inhibition does not mediate the effects of NSAID on drug cytotoxicity. In contrast, a multidrug resistance (MDR) phenotype due to expression of the multidrug resistance-associated protein (MRP) is most prominent in T98G cells and is amenable to modulation by indomethacin, suggesting that inhibition of MRP is at least in partly responsible for the potentiation of doxorubicin and vincristine cytotoxicity by NSAID.     

Arsenic trioxide induces different gene expression profiles of genes related to growth and apoptosis in glioma cells dependent on the p53 status

We have previously reported that As(2)O(3) affected cell cycle progression and cyclins D1 and B1 expression in two glioma cell lines differing in p53 status (U87MG-wt; T98G-mutated). In the present study, we further demonstrated that As(2)O(3) affected proliferation, viability and apoptosis of the two cell lines in a dose- and time-dependent manner, and T98G cells were more sensitive than U87MG cells to As(2)O(3) -induced apoptosis and inhibition of proliferation and viability. We further investigated the expression profiles of genes related with apoptosis and cell cycle in the two cell lines with a human cDNA-microarray (SuperArray) spotted with 267 genes of apoptosis and cell cycle. Thirty five genes were upregulated and 15 genes downregulated at least 2-fold by As(2)O(3) in U87-MG cells; whereas, 38 genes were upregulated and 21 genes downregulated at least 2-fold in T98G cells by As(2)O(3). After As(2)O(3) treatment, p53 expression was upregulated 56.5-fold in T98G cells, but only 6.0-fold in U87MG cells. The results indicate that As(2)O(3) suppresses the growth of U87MG cells mainly by regulating expression of genes of cell cycle arrest, stress and toxicity; whereas As(2)O(3) affects T98G cells mainly by regulating expression of genes belonging to Bcl-2, tumor necrotic factor receptor and ligand families. The data may be helpful for optimizing As(2)O(3) as an anti-cancer drug in the treatment of gliomas.     

Hypoxia induces expression of a GPI-anchorless splice variant of the prion protein

The human prion protein (PrP) is a glycoprotein with a glycosylphosphatidylinositol (GPI) anchor at its C-terminus. Here we report alternative splicing within exon 2 of the PrP gene (PRNP) in the human glioblastoma cell line T98G. The open reading frame of the alternatively spliced mRNA lacked the GPI anchor signal sequence and encoded a 230 amino acid polypeptide. Its product, GPI-anchorless PrP (GPI(-) PrPSV), was unglycosylated and soluble in non-ionic detergent, and was found in the cytosolic fraction. We also detected low levels of alternatively spliced mRNA in human brain and non-neuronal tissues. When long-term passaged T98G cells were placed in a low-oxygen environment, alternatively spliced mRNA expression increased and expression of normally spliced PrP mRNA decreased. These findings imply that oxygen tension regulates GPI(-) PrPSV expression in T98G cells.     

Expression of Heme Oxygenase Isozyme mRNAs in the Human Brain and Induction of Heme Oxygenase-1 by Nitric Oxide Donors

Abstract: Heme oxygenase is an essential enzyme in the heme catabolism that produces carbon monoxide (CO). This study was designed to examine the expression of two heme oxygenase isozyme mRNAs in the human brain and to explore the involvement of nitric oxide (NO) and various neuropeptides in the regulation of their expression. Northern blot analysis showed the expression of heme oxygenase-1 and heme oxygenase-2 mRNAs in every region of the brain examined, with the highest levels found in the frontal cortex, temporal cortex, occipital cortex, and hypothalamus. In a human glioblastoma cell line, T98G, treatment with any of three types of NO donors—sodium nitroprusside, 3-morpholinosydnonimine, and S -nitroso- l -glutathione—caused a significant increase in the levels of heme oxygenase-1 mRNA but not in the levels of heme oxygenase-2 and heat-shock protein 70 mRNAs. Sodium nitroprusside increased the levels of heme oxygenase-1 protein but not the levels of heat-shock protein 70 in T98G cells. The increase in content of heme oxygenase-1 mRNA caused by sodium nitroprusside was completely abolished by the treatment with actinomycin D. On the other hand, the levels of heme oxygenase isozyme mRNAs were not noticeably changed in T98G cells following the treatment with 8-bromo cyclic GMP, sodium nitrite, or various neuropeptides, such as calcitonin gene-related peptide, endothelin-1, and corticotropin-releasing hormone. The present study has shown the expression profiles of heme oxygenase-1 and -2 mRNAs in the human brain and the induction of heme oxygenase-1 mRNA caused by NO donors in T98G cells. These findings raise a possibility that the CO/heme oxygenase system may function in concert with the NO/NO synthase system in the brain.     

Entry into S phase is inhibited in two immortal cell lines fused to senescent human diploid cells.

Senescent human diploid cells (HDC) were fused to T98G human glioblastoma cells and to RK13 rabbit kidney cells, and DNA synthesis was analyzed in the heterodikaryons. T98G and RK13 cells are “partially transformed” cell lines that have some characteristics of normal cells, yet are transformed to immortality, i.e., they do not senesce. Previous experiments have shown that “fully transformed” HeLa and SV80 cells induce DNA synthesis in senescent HDC nuclei, whereas normal young HDC do not. Our experiments show that T98G and RK13 cells do not induce DNA synthesis in senescent HDC nuclei. These results demonstrate that the ability to induce DNA synthesis in senescent HDC is not correlated with immortality per se. Our results show further that a T98G cell in S phase at the time of fusion to a senescent HDC will continue to make DNA. However, a T98G cell in G1 phase at the time of fusion is prevented from initiating DNA synthesis. RK13 cells behave similarly to T98G. These results are consistent with the hypothesis that the molecular basis for the senescent phenotype involves a block that prevents cells in G1 phase from entering S phase. Thus, we conclude that the senescent phenotype can be dominant in heterokaryons composed of senescent HDC fused with certain immortal cell lines. To explain the different results obtained with various immortal cell lines, we present a model that suggests that T98G and RK13 cells are immortal because they have lost a normal regulatory factor, whereas HeLa and SV80 are immortal because they have gained a dominant transformation factor.     

Calcium fluxes modulate the radiation-induced bystander responses in targeted glioma and fibroblast cells

Bystander responses have been reported to be a major determinant of the response of cells to radiation exposure at low doses, including those of relevance to therapy. This study investigated the role of changes in calcium levels in bystander responses leading to chromosomal damage in nonirradiated T98G glioma cells and AG01522 fibroblasts that had been either exposed to conditioned medium from irradiated cells or co-cultured with a population where a fraction of cells were individually targeted through the nucleus or cytoplasm with a precise number of microbeam helium-3 particles. After the recipient cells were treated with conditioned medium from T98G or AG01522 cells that had been irradiated through either nucleus or cytoplasm, rapid calcium fluxes were monitored in the nonirradiated recipient cells. Their characteristics were dependent on the source of the conditioned medium but had no dependence on radiation dose. When recipient cells were co-cultured with an irradiated population of either T98G or AG01522 cells, micronuclei were induced in the nonirradiated cells, but this response was eliminated by treating the cells with calcicludine (CaC), a potent blocker of Ca(2+) channels. Moreover, both the calcium fluxes and the bystander effect were inhibited when the irradiated T98G cells were treated with aminoguanidine, an inhibitor of nitric oxide synthase (NOS), and when the irradiated AG01522 cells were treated with DMSO, a scavenger of reactive oxygen species (ROS), which indicates that NO and ROS were involved in the bystander responses generated from irradiated T98G and AG01522 cells, respectively. Our findings indicate that calcium signaling may be an early response in radiation-induced bystander effects leading to chromosome damage.     

Interaction between the pRb2/p130 C-terminal domain and the N-terminal portion of cyclin D3

An association between cyclin D3 and the C-terminal domain of pRb2/p130 was demonstrated using the yeast two-hybrid system. Further analysis restricted the epitope responsible for the binding within the 74 N-terminal amino acids of cyclin D3, independent of the LXCXE amino acid motif present in the D-type cyclin N-terminal region. In a coprecipitation assay in T98G cells, a human glioblastoma cell line, the C-terminal domain of pRb2/p130 was able to interact solely with cyclin D3, while the corresponding portion of pRb interacted with either cyclin D3 or cyclin D1. In T98G cells, endogenous cyclin D3-associated kinase activity showed a clear predisposition to phosphorylate preferentially the C-terminal domain of pRb2/p130, rather than that of pRb. This propensity was also confirmed in LAN-5 human neuroblastoma cells, where phosphorylation of the pRb2/p130 C-terminal domain and expression of cyclin D3 also decreased remarkably in the late neural differentiation stages.     

Monoclonal antibodies that inhibit activation and proliferation of lymphocytes. II. Requisite role of the monoclonal antibody-defined antigen systems in activation and proliferation of human and rat lymphocytes

A murine monoclonal antibody (MoAb) B3 to rat cells and MoAb HBJ127 and HBJ98 to human cells were found previously to recognize the homologous antigen systems (gp130 in the rat and gp125 in the human) which are predominantly distributed on the cell surface of proliferating cells of the respective species, and the expression of the antigen systems in lymphocytes were indicated previously to correlate closely with the activation and proliferation of the lymphocytes. In this respect, the in vitro effects of these MoAb on the nucleic acid synthesis, cell cycles, or proliferation of stimulated rat and human lymphocytes were examined by use of T cell-enriched and B cell-enriched cell populations. The addition of B3 MoAb to cultures diminished Con A-induced or allogeneic mixed lymphocyte culture-induced rat T cell proliferation and lipopolysaccharide-induced rat B cell proliferation, whereas B31 MoAb, which is unreactive with the gp130 antigen, did not inhibit these lymphocyte responses. Similarly, both HBJ127 and HBJ98 MoAb could inhibit the human lymphocyte proliferation in vitro, although HBJ127 MoAb showed about eight times greater inhibitory activity than did HBJ98 MoAb; HBJ127 MoAb almost completely inhibited the DNA synthesis of the Con A-stimulated lymphocytes at concentrations higher than 13 micrograms/ml. The flow cytometric analysis of the cellular nucleic acid contents with acridine orange-stained cells showed that when B3 MoAb and Con A were simultaneously added to unstimulated rat T cells, progression of the cell cycle was blocked at the G0 to G1 transition. In this culture condition, the appearance of the B3-defined antigen was arrested in a moderate level, as determined with fluorescein-stained cells. On the addition of B3 MoAb to the culture of the T cells after 24-hr Con A stimulation, the MoAb also strongly inhibited the cellular DNA synthesis, but it did not arrest the cell cycle at a certain phase and did not modulate the corresponding antigen. These data suggest that the B3 MoAb-defined antigen on the rat lymphocytes and the HBJ127/HBJ98 MoAb-defined antigen on the human lymphocytes may play some requisite roles not only in lymphocyte activation but also in the subsequent progression through the cell cycle to proliferate.     

Synthesis and in vitro PDT activity of miscellaneous porphyrins with amino acid and uracil

The synthesis of a series of modularized porphyrins bearing bioactive molecule is described. Starting with meso-tetraphenylporphyrin, the compounds with two nitro functional groups were synthesized via regiospecific nitration reaction. After reduction to the amino group and subsequent coupling with l-phenylalanine or 1-carboxylmethyl-5-fluorouracil (5-Fu acid), the functionalized porphyrins were metallized with Co(II) or Mn(II) to form miscellaneous porphyrins in good yields. The spectra of all the porphyrins were furnished. In vitro photodynamic therapy of the porphyrins against Ec9706 cell line was evaluated by standard cytotoxicity assays.     

In situ assessment of porphyrin photosensitizers in Propionibacterium acnes

Porphyrins are known to be efficient photosensitizer molecules and the combined action of light and porphyrins in Propionibacterium acnes have a lethal action on the cells. Identification and quantification of in situ porphyrins in P. acnes have been done using an integrating sphere connected to an ordinary absorption spectrophotometer, and the amounts of porphyrins in the cells were quantified by measuring scattering free absorption spectra of the cell suspensions. The concentration of porphyrins in P. acnes cells were increased in either of two ways; by the addition of delta-aminolevulinic acid (ALA), which lead to the formation of coproporphyrin III under the incubation conditions used in these experiments, or by the addition of protoporphyrin IX (PPIX) to the cell suspension. In the latter case, PPIX molecules are taken up by the cells in a membrane-mediated uptake mechanism, and accumulate in the cells either on a monomeric or a particular aggregate form. The fraction of porphyrins on aggregate form increased with increasing PPIX additions. In the case of ALA induced porphyrin production, only monomeric porphyrins were stored in the cells. In both cases, the cells have a limited binding capacity of monomeric porphyrins, which is estimated to be 3 x 10(5) molecules/cell, or one porphyrin molecule to every 100st lipid molecule in the cell membrane.     

The effect of folic acid on porphyrin synthesis in tumors and normal skin of mice treated with 5-aminolevulinic acid or methyl 5-aminolevulinate.

5-Aminolevulinic acid (ALA) or its derivative methyl 5-aminolevulinate (MAL) combined with folic acid was applied in nude mice bearing human colon adenocarcinoma. The aim of the study is to see whether folic acid may increase biosynthesis of porphyrins in tumor tissue after systemic or topical administration of ALA or MAL. The production of porphyrins was determined by spectrofluorometric measurements with an optical fibre probe. It was found that the porphyrin production after i.p injection of 200 mg kg(-1) ALA or MAL was significantly increased by i.p injection of 100 mg kg(-1) folic acid. However, in the case of topically applied 20% ALA, folic acid had no effect. In the case of topically applied 20% MAL, folic acid (i.p or topically applied) reduced the porphyrin synthesis. This might be used for the protection of normal skin against photosensitization. The effects of folic acid were similar in tumors and normal skin. Two mechanisms may explain the results: enhancement of the efficiency of the rate-limiting enzyme porphobilinogen deaminase by folic acid or interference of folic acid with the transport of ALA and MAL to and into the cells synthesizing porphyrins in the tissues. The present data seem to favour the latter mechanism. Folic acid may have a role as an adjuvant in photodynamic therapy with systemically administered ALA and its derivatives.