Dose-dependent and independent temporal patterns of gene responses to ionizing radiation in normal and tumor cells and tumor xenografts.

U87 cells derived from human malignant gliomas and growtharrested human embryonic lung (HEL) fibroblasts were examined with respect to their response to ionizing radiation by profiling their RNAs. In the first series of experiments, cells grown in vitro were harvested and the RNAs were extracted 5 h after exposure to 1, 3, or 10 Gy. In the second series of experiments the U87 tumors were implanted in nude mice and subjected to the same doses of irradiation. The xenografts were harvested at 1, 5, or 24 h after irradiation and subjected to the same analyses. We observed and report on (i) cell-type common and cell-type specific responses, (ii) genes induced at low levels of irradiation but not at higher doses, (iii) temporal patterns of gene response in U87 xenografts that varied depending on radiation dose and temporal patterns of response that were similar at all doses tested, (iv) significantly higher up-regulation of cells in xenografts than in in vitro cultures, and (v) genes highly up-regulated by radiation. The responding genes could be grouped into nine functional clusters. The representation of the nine clusters was to some extent dependent on dose and time after irradiation. The results suggest that clinical outcome of ionizing radiation treatment may benefit significantly by taking into account both cell-type and radiation-dose specificity of cellular responses.

Angiostatin effects on endothelial cells mediated by ceramide and RhoA.

Angiostatin is a cleavage product of plasminogen that has anti-angiogenic properties. We investigated whether the effects of angiostatin on endothelial cells are mediated by ceramide, a lipid implicated in endothelial cell signaling. Our results demonstrate that angiostatin produces a transient increase in ceramide that correlates with actin stress fiber reorganization, detachment and death. DNA array expression analysis performed on ceramide-treated human endothelial cells demonstrated induction of certain genes involved in cytoskeleton organization. Specifically, we report that treatment with angiostatin or ceramide results in the activation of RhoA, an important effector of cytoskeletal structure. We also show that treatment of endothelial cells with the antioxidant N-acetylcysteine abrogates morphological changes and cytotoxic effects of treatment with angiostatin or ceramide. These findings support a model in which angiostatin induces a transient rise in ceramide, RhoA activation and free radical production.

Angiostatin induces mitotic cell death of proliferating endothelial cells.

Angiostatin is an inhibitor of tumor angiogenesis that induces regression of experimental tumors and enhances the antitumor effects of radiation therapy. We report that the cytotoxic effects of angiostatin are restricted to the proliferating endothelial cell population. In addition, angiostatin and ionizing radiation (IR) interact by inducing death of dividing endothelial cells. We also show that angiostatin and IR interact to inhibit endothelial cell migration. These findings demonstrate that angiostatin targets the proliferating tumor vasculature and provide a mechanistic basis for the cytotoxic interaction of angiostatin and IR.

Down-regulation of ceramide production abrogates ionizing radiation-induced cytochrome c release and apoptosis.

Previous work has demonstrated that down-regulation of ceramide production after selection of cells with N-oleoylethanolamine (OE), an inhibitor of ceramidase, results in resistance to DNA damage-induced apoptosis. We report here that acute exposure of WEHI-231 cells (murine B-cell lymphoma) to OE activates neutral sphingomyelinase, induces ceramide production and increases intracellular reactive oxygen species. OE exposure also induces mitochondrial permeability, cytochrome c release, and apoptosis. Cells selected for resistance to OE exhibit little if any change in reactive oxygen species and cytochrome c release when exposed either to OE or to toxic doses of ceramide. Importantly, the OE resistant cells are also resistant to ionizing radiation-induced cytochrome c release and apoptosis. These findings demonstrate that down-regulation of neutral sphingomyelinase activity is associated with decreased DNA-damage-induced apoptosis. In addition, the data suggests that agents that modify extranuclear targets responsible for ceramide production select for cells resistant to ionizing radiation-induced apoptosis through alterations in mitochondrial function.

Modulation of apoptotic response of a radiation-resistant human carcinoma by Pseudomonas exotoxin-chimeric protein.

Strategies to sensitize human tumors that are resistant to apoptosis have been clinically unsuccessful. We demonstrate that a structurally modified chimeric Pseudomonas exotoxin, PEdelta53L/TGF-alpha/KDEL, with binding specificity for the epidermal growth factor receptor, markedly enhances sensitivity of human xenografts to radiation killing. Exposure to PEdelta53L/TGF-alpha/KDEL decreases the apoptotic threshold through protein synthesis inhibition and simultaneous production of ceramide in tumor cells that lack functional p53 protein. In contrast, no increase in local or systemic toxicity was observed with the chimeric toxin and radiation. We conclude that biochemical targeting of the chimeric toxin and physical targeting of ionizing radiation may increase the therapeutic ratio in the treatment of human cancers with alterations of p53 expression. This strategy offers a high therapeutic potential for Pseudomonas exotoxin A chimeric proteins and irradiation.

Human papillomavirus expression and p53 gene mutations in squamous cell carcinoma.

OBJECTIVES: To determine the incidence of human papillomavirus (HPV) infection and p53 gene mutation expression in squamous cell carcinomas (SCCs) of the oral cavity and tonsils, to correlate the presence of HPV and p53 gene mutation with known clinical and pathological features of SCC, and to determine whether infection with HPV or the presence of p53 gene mutations are independent prognosticators of patient survival. DESIGN: To accomplish this goal, 58 patients with SCCs of the oral cavity and 42 patients with SCCs of the tonsils were randomly examined. The cases examined met the criteria of 5-year clinical follow-up, availability of complete staging information and treatment history, and the presence of paraffin-embedded tumor specimens. Immunohistochemical tests were performed to identify the mutant p53 protein. Human papillomavirus identification was accomplished with polymerase chain reaction, with confirmation via restriction fragment length polymorphisms. RESULTS: The incidence of p53 gene mutation expression for this series was 66%. Human papillomavirus infection was found in 11 patients (11%). There was a trend toward increased p53 gene mutation expression with advancing stage of tumor in the oral cavity cancer group, although this was less evident in the tonsil cancer population. The p53 gene mutation status was found not to correlate with the histological grade of the tumor, patient age or sex, recurrence rates, or survival status. Like p53 expression, there were no correlations found between the presence of HPV and age, sex, histological grade, or recurrence rates. However, a correlation did exist between HPV and survival status in the tonsil cancer group, with improved survival noted among patients with tonsil cancers infected with HPV compared with those not infected with HPV. A significant correlation existed with both p53 gene mutation status and HPV status with respect to alcohol and tobacco use. The presence of the p53 gene mutation positively correlated with increased tobacco and alcohol use, whereas infection with HPV predicted a significantly lower rate of alcohol and tobacco consumption. CONCLUSIONS: Human papillomavirus infection is an independent risk factor for the development of oral cavity and tonsil SCCs in those patients with a relatively low alcohol and tobacco use history. Conversely, there is a strong association between heavy alcohol and tobacco use and mutation of the p53 gene. Neither p53 gene mutation nor HPV infection serve as prognosticators of tumor behavior in SCCs of the oral cavity or tonsils, with the exception of improved survival noted among patients with tonsil cancers infected with HPV.

Decreasing the apoptotic threshold of tumor cells through protein kinase C inhibition and sphingomyelinase activation increases tumor killing by ionizing radiation.

Approximately 30% of cancer deaths result from the failure to control local and regional tumors. The goal of radiotherapy is to maximize local and regional tumor cell killing while minimizing normal tissue destruction. Attempts to enhance radiation-mediated tumor cell killing using halogenated pyrimidines, antimetabolites, and other DNA-damaging agents or sensitizers of hypoxic tumor cells have met with only modest clinical success. In an unique strategy to modify tumor radiosensitivity, we used an inhibitor of the protein kinase C group A and B isoforms, chelerythrine chloride (chelerythrine), to enhance the killing effects of ionizing radiation (IR). Protein kinase C activity plays a central role in cellular proliferation, differentiation, and apoptosis. Chelerythrine increases sphingomyelinase activity and enhances IR-mediated cell killing through induction of apoptotic tumor cell death in a radioresistant tumor model both in vitro and in vivo. Although previous reports have suggested that IR-mediated apoptosis correlates with tumor volume reduction, we demonstrate for the first time that lowering the apoptotic threshold increases tumor cell killing in vivo.

Loss of ceramide production confers resistance to radiation-induced apoptosis.

Ionizing radiation mediates cell death, in part, through chromosomal damage following one or more cell divisions. X-rays also induce programmed cell death (apoptosis) in some cell types both in vitro and in vivo. Both neutral and acidic sphingomyelinases, which generate the lipid second messenger ceramide, are reported to induce apoptosis following ionizing radiation and other death signals such as tumor necrosis factor alpha and Fas ligand. Herein we report that a loss of ceramide production from a neutral sphingomyelinase generates a radioresistant phenotype as measured by a marked decrease in apoptosis. A WEHI-231 subline made deficient in ceramide production was found to be resistant to apoptosis compared with the parental subline following treatment with X-rays. The resistant subline underwent two to three subsequent cell divisions following X-irradiation, confirming that X-rays induce cell death through both mitotic and apoptotic mechanisms. These data suggest that loss of ceramide production following X-rays represents an extranuclear mechanism for the development of radioresistance. Modulation of extranuclear signals may increase tumor cell killing following radiation and represent new cellular targets for cancer therapy.

Protein kinase C inhibition induces apoptosis and ceramide production through activation of a neutral sphingomyelinase.

We report that WEHI-231 undergo apoptosis following exposure to the protein kinase C inhibitors chelerythrine chloride and calphostin C. Following the addition of chelerythrine or calphostin C to WEHI-231 cells, ceramide production increased over baseline levels with a concurrent decrease in sphingomyelin. More detailed examinations determined that the ceramide accumulation resulted from activation of neutral, but not acidic, sphingomyelinase. These results suggest an antagonistic relationship between protein kinase C activity and ceramide in the signaling events preceding apoptosis.

Human papilloma virus and p53 in head and neck cancer: clinical correlates and survival.

Recent studies have shown that p53 mutations are frequently found in cancer of the head and neck, whereas others have indicated that human papilloma virus (HPV) infection may be involved. Thus far, no studies have examined both p53 and HPV in the same patient population and correlated the results with clinical characteristics and outcome. The purpose of this study was to examine any interrelationship between p53 and HPV in patients with squamous cell carcinoma (SCC) of the head and neck. We also planned to correlate the experimental findings with clinical characteristics, known risk factors, and treatment outcome to determine whether any prognostic factors could be detected. Archival material from 66 patients with SCC of the head and neck were selected for study based on the availability of tissue from the primary tumors prior to treatment. A data base was constructed containing all clinical parameters at the time of diagnosis and risk factors. Genomic DNA was isolated and amplified using PCR, followed by SSCP analysis and direct genomic sequencing of all variants to detect p53 mutations. Two independent methods were used for HPV detection: (a) PCR amplification using primers homologous to the E6 region of HPV 16, 18, and 33, followed by RFLP analysis; and (b) PCR amplification with HPV L1 consensus primers, followed by triple restriction enzyme digestion. The results were entered into the data base for statistical analysis. Twenty-four percent of patients were found to have p53 mutations, and 18% were positive for HPV infection. Only one patient was positive for both. Tonsilar cancer was strongly correlated with HPV (P = 0.0001) and inversely correlated with p53 (P = 0.03). The only clinical parameter associated with p53 mutation was a trend toward a heavier smoking history. A subset analysis of the patients with tonsilar cancer revealed inverse correlations with smoking (P = 0. 015) and alcohol use (P = 0.05). Also, white patients with SCC of the tonsil were more likely to be HPV positive (P = 0.015). No significant relationships with outcome were detected with either p53 or HPV in the entire population. A subset analysis of patients with stage IV disease revealed that HPV infection was correlated with overall survival. This is the largest study to date to examine both p53 and HPV in patients with SCC of the head and neck. Our results suggest that HPV may be involved in the development of these cancers in patients without traditional risk factors and that HPV-related cancers are more prevalent in the white race.

Cross-talk between ceramide and PKC activity in the control of apoptosis in WEHI-231.

WEHI-231, a murine B-cell lymphoma, readily undergoes programmed cell death following surface immunoglobulin (Ig) cross-linking [1]. Ceramide has been shown to induce apoptosis in WEHI-231 following its exposure to anti-lg antibodies, dexamethasone, and irradiation [2]. Recently, Haimovitz-Friedman et al. have demonstrated in endothelial cells that PMA not only prevented ceramide mediated apoptosis, but inhibited the generation of ceramide following irradiation [3]. In this paper we use highly specific PKC inhibitors to explore the connection between PKC activity, ceramide signaling and apoptosis. Both chelerythrine chloride and calphostin C triggered rapid apoptosis in WEHI-231 and acted in synergy with exogenous ceramide to induce apoptosis. Detailed studies of chelerythrine's mechanism of action revealed that 30 minutes following addition of 10 microM chelerythrine, sphingomyelin and phosphatidylcholine (PC) mass decreased confirming our previous findings of neutral, but not acidic, sphingomyelinase activation following treatment with PKC inhibitors [4]. The novel observation that inhibition of PKC isoforms present in WEHI-231 leads to a rapid rise in cellular ceramide as a results of sphingomyelin hydrolysis further suggests an antagonistic relationship between PKC activity and ceramide in the signaling events preceding apoptosis.

Ribonucleotide reductase gene expression during cyclic AMP-induced cell cycle arrest in T lymphocytes.

In both 3T3 mouse fibroblasts and S49 mouse T lymphocytes the genes encoding both subunits of ribonucleotide reductase are expressed beginning in late G1 phase. In studies reported here, we compared the expression of the genes that code for the M1 and M2 subunits of ribonucleotide reductase in S49 cells, which are arrested in G1 phase by agents that increase cyclic AMP, with those from CEM human T lymphoma cells that are unaffected by exposure to dibutyryl cyclic AMP. Dibutyryl cyclic AMP treatment results in a prompt steady diminution of M2 mRNA concentration to levels at or below that of elutriated G1 cell-cycle-specific populations in S49 cells, in contrast to CEM cell M2 mRNA, which is unchanged. M1 mRNA concentration decreases more slowly than M2 mRNA in S49 cells and marginally, if at all, in CEM cells. The time course of diminution of the M2 message concentration by dibutyryl cyclic AMP in S49 cells is similar to that obtained when cells are treated with actinomycin D and to the combination of the two agents. This suggests that cyclic AMP and actinomycin D may act similarly on ribonucleotide reductase gene expression. Furthermore, cycloheximide pretreatment diminishes the effect of dibutyryl cyclic AMP, indicating that the effect might be mediated by a labile protein. Transcription runoff assays suggest a diminution of transcription rate for the M2 gene in S49 cells treated with dibutyryl cyclic AMP and a transient decline in the M1 transcription rate. These data suggest that dibutyryl cyclic AMP diminishes the transcription of ribonucleotide reductase genes in sensitive cells and that this and the short half-life of the M2 message are major factors in the disappearance of the M2 messenger RNA from dibutyryl cyclic AMP-treated cells although other mechanisms may also play a role. These events clearly precede any alteration in cell cycle distribution and thus they may contribute to G1 arrest.

Deoxyadenosine- and cyclic AMP-induced cell cycle arrest and cytotoxicity.

We compared deoxyadenosine (AdR)- and cyclic AMP (cAMP)-induced cell cycle arrest and cytotoxicity in wild type and mutant S49 cells to determine whether they resulted from the same or different mechanisms. Cyclic AMP and deoxyadenosine are synergistic rather than additive in cytotoxicity assays, suggesting different mechanisms of toxicity. Although cyclic AMP causes cell death after 72 h, in concentrations sufficient to result in cell cycle arrest it is reversible with virtually no cytotoxicity for at least 24 h, whereas AdR-induced cell cycle arrest is lethal and irreversible. AdR-induced G1 cell cycle arrest results in diminished ribonucleotide reductase activity but the kinetics of this inhibition differ from cyclic AMP-induced cell cycle arrest. Cyclic AMP arrest and cytotoxicity depend on cyclic AMP-dependent protein kinase (PKA) activity, whereas AdR toxicity does not differ between cell lines with or without PKA activity. Furthermore, deoxycytidine prevents AdR cell cycle arrest and cytotoxicity but has no effect on cyclic AMP G1 arrest. Finally, comparison of cytofluorographic patterns of G1-arrested cells suggests that the AdR block is later in G1 than cyclic AMP-induced cell cycle arrest. In summary, these data show that while the mechanisms of cell cycle arrest and cytotoxicity of cyclic AMP and deoxyadenosine are uncertain, they do appear to involve different pathways.

Effect of cyclic AMP on the cell cycle regulation of ribonucleotide reductase M2 subunit messenger RNA concentrations in wild-type and mutant S49 T lymphoma cells.

Ribonucleotide reductase activity in S49 T lymphoma cells is cell cycle regulated by de novo protein synthesis of the M2 subunit. There is maximal enzyme activity in S and G2/M phase with low activity and low concentrations of the M2 subunit in G1 phase. Pharmacologic concentrations of cyclic AMP arrest S49 cells in the G1 phase of the cell cycle. We investigated the effect of cyclic AMP on M2 messenger RNA concentrations using RNA from exponentially growing and elutriated, cell cycle-enriched populations. To discern whether cyclic AMP-induced G1 arrest was associated with low concentrations of M2-specific messenger RNA, we probed blots with a full-length cDNA for M2. Cell cycle variation in M2 messenger RNA concentrations was similar in wild-type, hydroxyurea-resistant cells with amplified M2 activity, and cyclic AMP-dependent protein kinase-deficient cell lines. All lines had low amounts of M2-specific mRNA in early G1, an increase at the late G1/early S phase interface, a decrease in mid S phase, and another increase in late S phase that continued through G2/M. These concentrations did not directly correlate with enzyme activity, suggesting other regulatory effects might participate in determining ribonucleotide reductase activity. Cyclic AMP exposure appeared to induce cell cycle arrest in early G1 with low M2-specific messenger RNA concentration. This effect reversed upon washout of the cyclic AMP and was dependent on functional cyclic AMP-dependent protein kinase (PKA). These results suggest that cyclic AMP arrests S49 mouse T lymphoma cells in early G1 prior to transcriptional activation of the M2 gene.

The dose dependent effect of cyclic AMP on ribonucleotide reductase in mitogen stimulated mononuclear cells.

Cyclic adenosine monophosphate arrests proliferating T lymphocytes in the G1 phase of the cell cycle. Here we demonstrate that exogenous and endogenous elevations in cyclic AMP concentration result in diminished mitogen stimulation, cell cycle arrest, and decreased ribonucleotide reductase messenger RNA concentrations in peripheral blood mononuclear cells. At lower concentrations (less than 1mM) of dibutyryl cyclic AMP that do not generate cell cycle arrest there is inhibition of ribonucleotide reductase activity without decreased messenger RNA concentration for the M2 subunit of ribonucleotide reductase. However, at higher concentrations of dibutyryl cyclic AMP there is G1 cell cycle arrest and reduced M2 specific messenger RNA concentration. Thus, cyclic AMP inhibition of lymphocyte activation may occur by different mechanisms that are dose dependent.

M2 subunit of ribonucleotide reductase is a target of cyclic AMP-dependent protein kinase.

Cyclic AMP arrests T lymphocytes in the G1 phase of the cell cycle, and prolonged exposure results in cytolysis. Both of these effects require cyclic AMP-dependent protein kinase. We recently observed that some S49 mouse T lymphoma cell lines selected for hydroxyurea resistance were not arrested in G1 by cyclic AMP. Further analysis revealed that these cell lines were cyclic AMP-dependent protein kinase deficient, and conversely, other cyclic AMP-dependent protein kinase deficient cell lines not selected for hydroxyurea resistance were two- to threefold more hydroxyurea resistant. However, hydroxyurea is a specific inhibitor of ribonucleotide reductase and does not inhibit this kinase. We subsequently showed that cyclic AMP-dependent protein kinase will phosphorylate the M2 but not the M1 subunit of ribonucleotide reductase in vitro, and this phosphorylation will diminish CDP reductase activity. In vivo phosphorylation of M2 occurred under conditions similar to those that generate cell cycle arrest. We conclude that the M2 subunit of ribonucleotide reductase can be a target of cyclic AMP-dependent protein kinase. The phosphorylated enzyme has diminished activity, and this may play a role in cyclic AMP-induced lymphocyte cell cycle arrest.

Deoxyadenosine toxicity and cell cycle arrest in hydroxyurea-resistant S49 T-lymphoma cells.

Hydroxyurea-resistant S49 T-lymphoma cells have increased ribonucleotide reductase activity and deoxyribonucleoside triphosphate pools when compared with wild-type cultures. If ribonucleotide reductase inhibition is the mechanism by which deoxyadenosine is cytotoxic, then hydroxyurea (HU)-resistant S49 cells might be more resistant to deoxyadenosine toxicity when adenosine deaminase is inhibited than wild-type cells. Five S49 cell lines resistant to varying concentrations of HU were compared with wild-type cells by measuring CDP reductase activity, deoxyribonucleoside triphosphate pools, and deoxyadenosine toxicity. All five cell lines resistant to increasing concentrations of HU exhibited a twofold increase in resistance to deoxyadenosine toxicity when compared to wild type, and the resistance was proportional to the twofold increased pools of dNTPs in these cell lines but was less than the six- to eight fold increase in ribonucleotide reductase activity. In both wild-type and mutant cell lines, deoxyadenosine toxicity was accompanied by the accumulation of deoxyadenosine triphosphate and reduction of the other dNTPs; however, only dGTP greatly diminished. Exogenous addition of deoxycytidine decreased the dATP accumulation by about 20%, but also resulted in increases in the dCTP, dTTP, and dGTP pools. The S49 cells arrested in G1 phase when exposed to dAdo, although hydroxyurea-resistant cells required higher dAdo concentrations to elicit G1-phase arrest than wild-type cells. Deoxycytidine prevented dAdo-induced G1 arrest in all cell types. In summary, these data support the hypothesis that deoxyadenosine-induced dATP accumulation results in inhibition of ribonucleotide reductase and that this may be the mechanism for both cell cycle arrest and cytotoxicity in S49 T-lymphoma cells.

Deoxyribonucleotide metabolism and cyclic AMP resistance in hydroxyurea-resistant S49 T-lymphoma cells.

We investigated the cell cycle regulation of deoxyribonucleoside triphosphate (dNTP) metabolism in hydroxyurea-resistant (HYUR) murine S49 T-lymphoma cell lines. Cell lines 10- to 40-fold more hydroxyurea-resistant were selected in a stepwise manner. These HYUR cells exhibited increased CDP reductase activity (5- to 8-fold) and increased dNTP pools (up to 5-fold) that appeared to result from increased activity of the M2 subunit (binding site of hydroxyurea) of ribonucleotide reductase. These characteristics remained stable when the cells were grown in the absence of hydroxyurea for up to 2 years. In both wild type and hydroxyurea-resistant cell populations synchronized by elutriation, dCTP and dTTP pools increased in S phase, whereas dATP and dGTP pools generally remained the same or decreased, suggesting that allosteric effector mechanisms were operating to regulate pool sizes. Additionally, CDP reductase activity measured in permeabilized cells increased in S phase in both wild type and hydroxyurea-resistant cells, suggesting a nonallosteric mechanism of increased ribonucleotide reductase activity during periods of active DNA synthesis. While wild type S49 cells could be arrested in the G1 phase of the cell cycle by dibutyryl cyclic AMP, hydroxyurea-resistant cell lines could not be arrested in the G1 phase by exogenous cyclic AMP or agents that elevate the concentration of endogenous cyclic AMP. These data suggest that cyclic AMP-generated G1 arrest in S49 cells might be mediated by the M2 subunit of ribonucleotide reductase.