Onconase, an anti-tumor ribonuclease suppresses intracellular oxidative stress.

Onconase (ONC), an antitumor ribonuclease from oocytes of a frog Rana pipiens, capable of inducing apoptosis in many cell lines is synergistic with several other anticancer drugs. Since cytotoxic effects of numerous drugs are modulated by reactive oxygen intermediates (ROI), we have studied effects of ONC on the intracellular level of oxidants in several normal cell types as well as tumor cell lines. It is demonstrated for the first time that ONC substantially decreases the content of ROI in all cell lines studied. This effect depends on the ribonucleolytic activity of the enzyme and is due to both, decreased rate of ROI generation and accelerated rate of their degradation. Onconase decreases the mitochondrial transmembrane potential and consequently, generation of ATP. Simultaneously the enzyme decreases the expression of an antiapoptotic protein Bcl-2, and upregulates the proapoptotic Bax protein. These finding are consistent with the enzyme propensity to induce apoptosis. The observed antioxidant activity of ONC may be an important element of its cytotoxicity towards cancer cells. The enzyme seems to exert its biological activities by interfering with the redox system of cellular regulation.

Combined effects of onconase and IFN-beta on proliferation, macromolecular syntheses and expression of STAT-1 in JCA-1 cancer cells.

Prostate cancer (CaP), the most common malignancy in American men, presents its greatest challenge to clinicians when the cancer progresses to the hormone-refractory state. In the present investigation, we studied the combined effects of interferon (IFN) and onconase, each of which has reported antitumor activity, on growth and specific protein expression in JCA-1 cells. Cells were treated for up to 3 days with 1 and 5 microg/ml onconase, with and without concurrent addition of IFN-beta(ser) (10(3) IU/ml). Cell count and viability, and de novo RNA and protein synthesis were determined. Expression and subcellular distribution of STAT-1 were also assessed by immunoblot analysis. JCA-1 cells treated for 3 days with IFN or onconase showed a 15-30% reduction in cell proliferation, which was increased to 42-51% with both agents. Analysis of [35S]methionine incorporation into cells confirmed a more pronounced inhibitory effect elicited by IFN-beta and onconase; IFN-beta and 1 microg/ml onconase each decreased de novo protein synthesis by 23-25%, while the combination resulted in 59% suppression. Similar studies using incorporation of [3H]uridine into RNA yielded less significant effects. Further investigation using pre-labeled cellular RNA and proteins showed that either agent or their combination did not affect the turnover of macromolecules. To test whether the antiproliferative effects of IFN-beta and onconase were correlated with one or more specific gene changes, expression of an IFN-modulated protein, STAT-1, was determined. Both phosphorylated and unphosphorylated forms of STAT-1 and its subcellular distribution in the nucleus and cytoplasm, were increased 3-fold by IFN-beta. The IFN-elicited elevation of STAT-1 was not additionally augmented by onconase but was reduced 20-25% when onconase was simultaneously present as IFN-beta. These data show that the overall changes in STAT-1 did not correlate with the reduction in cell growth and the suppression of de novo protein/RNA synthesis elicited by these two agents, and imply that other target proteins are likely to be involved in the combined effects of IFN-beta and onconase in JCA-1 cells.

Morphometry of nucleoli and expression of nucleolin analyzed by laser scanning cytometry in mitogenically stimulated lymphocytes.

BACKGROUND: Various attributes of nucleoli, including abundance of the nucleolar product (rRNA), correlate with cell-proliferative status and are useful markers for tumor diagnosis and prognosis. However, there is a paucity of methods that can quantitatively probe nucleolus. The aim of the present study was to utilize the morphometric capacity of the laser scanning cytometer (LSC) to analyze nucleoli and measure expression of the nucleolar protein nucleolin (NCL) in individual cells and correlate it with their state of proliferation. MATERIALS AND METHODS: Human lymphocytes were mitogenically stimulated, and at different time points their nucleoli were detected immunocytochemically using NCL Ab. The frequency of nucleoli per nucleus, their area, and the level of expression of NCL, separately in the nuclear and nucleolar compartments, were estimated in relation to the G(0) to G(1) transition and the cell cycle progression. RESULTS: During the first 24 h of stimulation, when the cells underwent G(0) to G(1) transition, their RNA content was increased nearly 8-fold, the level of NCL per nucleus also increased 8-fold, the NCL per nucleolus increased 12-fold, nucleolear area increased 3-fold, and NCL/nucleolar area increased nearly 4-fold. During the subsequent 24-48 h of stimulation, when cells were progressing through S, G(2), and M and reentering the next cycle, the number of nucleoli per nucleus was increased and a massive translocation of NCL from nucleoli to nucleoplasm was observed; its overall level per nucleus, however, still remained high, at 6-fold above of that of G(0) cells. CONCLUSIONS: While high expression of NCL in the nucleolar compartment correlates with the rate of rRNA accumulation in the cell and is a sensitive marker of the G(0) to G(1) transition, the cells progressing through the remainder of the cycle are better distinguished from G(0) cells by high overall level of NCL within the nucleus. Such an analysis, when applied to tumors, may be helpful in obtaining the quantitative parameters related to the kinetic status of the tumor-cell population and tumor prognosis. The capability of LSC to measure the protein translocation between nucleolus and nucleoplasm can be used to study the function and regulatory mechanisms of other proteins that reside in these compartments.

Chlorophyllin protects cells from the cytostatic and cytotoxic effects of quinacrine mustard but not of nitrogen mustard.

Chlorophyllin (CHL), the sodium and copper salt of chlorophyll, is capable of inhibiting the mutagenic activity of many chemical compounds. Several mechanisms have been advanced to explain the antimutagenic activity of CHL, including its antioxidant properties and its ability to form complexes with mutagens. The present study was designed to reveal whether the heterocyclic aromatic nature of a potential mutagen is essential to its sensitivity to CHL. Toward this end, the inhibitory effect of CHL on two compounds of similar chemical reactivity (mustards), that either embodied an aromatic structure (quinacrine mustard; QM) or did not (nitrogen mustard; NM), were compared. Human leukemic HL-60 and breast carcinoma MCF-7 cells were treated with QM or NM in the absence or presence of various concentrations of CHL. Both QM and NM when administered for 1-2 h at micromolar concentrations exerted similar effects; both arrested cells in G2 phase of the cell cycle, induced apoptosis and reduced the clonogenicity of MCF-7 cells. The simultaneous addition of 0.22 M CHL to cultures receiving QM virtually abolished the QM-induced inhibition of cell growth and clonogenicity. In contrast, CHL had no effect on reducing the cytostatic or cytotoxic activity of NM. CHL alone, at a concentration of 0.22 M, had minimal effect on growth of HL-60 cells slightly perturbing their progression through G2. The results are consistent with the model that explains the inhibition of the activity of mutagens or antitumor drugs with aromatic structures by CHL as mediated by its ability to sequester these molecules within heterologous mutagen:CHL complexes that are maintained by stacking interactions. Therefore, excess of chlorophyll in the diet, by sequestering aromatic mutagens (or antitumor drugs with a heterocyclic structure, if taken orally), may inhibit their accessibility to cells, thereby reducing their activity.

G1 arrest of U937 cells by onconase is associated with suppression of cyclin D3 expression, induction of p16INK4A, p21WAF1/CIP1 and p27KIP and decreased pRb phosphorylation.

Onconase is a 12 kDa protein homologous to pancreatic RNase A isolated from amphibian oocytes which shows cytostatic and cytotoxic activity in vitro, inhibits growth of tumors in mice and is in phase III clinical trials. The present study was aimed to reveal mechanisms by which onconase perturbs the cell cycle progression. Human histiocytic lymphoma U937 cells were treated with onconase and expression of cyclins D3 and E, as well as of the cyclin-dependent kinase inhibitors (CKIs) p16INK4A, p21WAF1/CIP1 and p27KIP1 (all detected immunocytochemically) was measured by multiparameter flow cytometry, in relation to the cell cycle position. Also monitored was the status of phosphorylation of retinoblastoma protein (pRb) by a novel method utilizing mAb which specifically detects underphosphorylated pRb in individual cells. Cell incubation with 170 nM onconase for 24 h and longer led to their arrest in G1 which was accompanied by a decrease in expression of cyclin D3, no change in cyclin E, and enhanced expression of all three CKIs. pRb was underphosphorylated in the onconase arrested G1 cells but was phosphorylated in the cells that were still progressing through S and G2/M in the presence of onconase. The cytostatic effect of onconase thus appears to be mediated by downregulation of cyclin D3 combined with upregulation of p27KIP1, p16INK4A and p21WAF1/CIP1, the events which may prevent phosphorylation of pRb during G0/1 and result in cell arrest at the restriction point controlled by Cdk4/6 and D type cyclins.

Potentiation of tumor necrosis factor induced apoptosis by onconase.

Onconase (ONC) a ribonuclease from amphibian oocytes is cytostatic and cytotoxic to many human tumor lines, shows in vivo antitumor activity in mouse tumor models and is in Phase III clinical trials. The mechanism of antitumor activity of ONC is presumed to be due to its internalization, degradation of intracellular RNA and suppression of protein synthesis. Since apoptosis triggered by TNF-alpha is known to be potentiated by inhibitors of protein synthesis, we have hypothesized that it also may be potentiated by ONC. Indeed, preincubation of U-937 or HL-60 leukemic cells with 0.17 microM ONC rendered them more sensitive to induction of apoptosis by TNF-alpha or antibody to CD95 (Fas). The mechanism by which ONC amplifies the effect of TNF-alpha may involve suppression of induction of the survival genes whose expression is triggered by activation of NFkB by this factor.

Reactive oxygen species generated by cyanide mediate toxicity in rat pheochromocytoma cells.

Peroxide formation has been implicated in impairment of motor function by cyanide which occurs in both animals and man. The present study employs the neuronal model, rat pheochromocytoma (PC12) cells to evaluate peroxidation as a toxic mechanism of cyanide. Confocal imaging shows that peroxides form within a few seconds in cell cytoplasm after cyanide exposure and continue to accumulate over a period of several minutes. Peroxide generation by cyanide is decreased to about 50% by phospholipase A2 inhibitors indicating involvement of arachidonic acid in the oxidative process. Also antioxidant defense enzymes (CuZn superoxide dismutase and especially catalase) in PC12 cells are inhibited by cyanide. It appears that peroxide accumulation after cyanide treatment involves both inhibition of breakdown and increased production. Furthermore, both peroxide accumulation and cell death induced by cyanide in PC12 cells are blocked by an antioxidant (ascorbate). These data support the hypothesis that the cytotoxic action of cyanide is related in part to an oxidative process.

Apoptosis and cell cycle effects induced by extracts of the Chinese herbal preparation PC SPES.

A herbal preparation denoted PC SPES(R) is available in 'natural food' or 'health food' stores in the United States. This mixture (patent pending, US Serial number 08/697,920) consists of extracts from 8 different herbs, 7 originating from China and one from America, and is sold as a dietary supplement. Although several components of this herbal mixture were reported to have antiproliferative and/or antitumor activity little is known about the possible in vitro cytostatic or cytotoxic properties of the formulation. Composition of PC SPES is standardized by HPLC; the ethanol extract is characterized by the presence of 6 distinct components, reproducible from batch to batch. This extract suppressed cell proliferation and reduced the clonogenicity of a variety of human tumor cell lines, including PC-3 and LNCaP prostate carcinomas, MCF-7 and T47-D breast carcinomas, SK-N-MC neuroepithelioma, Cole 38 melanoma, U937 histiomonocytic lymphoma, as well as HL-60 and MOLT-4 leukemias. The sensitivity to PC SPES was different for particular cell lines, with MCF-7 cells being the most sensitive (IC50 = 20 nl/ml) and Cole 38 the most resistant (IC50 = 430 nl/ml) in clonogenicity assays. The predominant cell cycle effect induced by PC SPES was the prolongation of G(1) phase. Apoptosis was observed after exposure of tumor cells to PC SPES for 48 h and longer. PC SPES also downregulated expression of bcl-2, the gene protecting cells against apoptosis (studied in U937 cells) and sensitized these cells to gamma-irradiation. The cell cycle progression of mitogen stimulated human lymphocytes was not affected at PC SPES concentrations which induced cytotoxic and cytostatic effects in tumor cells. The data indicate that PC SPES is cytostatic and cytotoxic for different tumor cell lines and modulates the cell's propensity to undergo apoptosis.

Cytometry of cyclin proteins.

Cyclins are key components of the cell cycle progression machinery. They activate their partner cyclin-dependent kinases (CDKs) and possibly target them to respective substrate proteins within the cell. CDK-mediated phosphorylation of specific sets of proteins drives the cell through particular phases or checkpoints of the cell cycle. During unperturbed growth of normal cells, the timing of expression of several cyclins is discontinuous, occurring at discrete and well-defined periods of the cell cycle. Immunocytochemical detection of cyclins in relation to cell cycle position (DNA content) by multiparameter flow cytometry has provided a new approach to cell cycle studies. This approach, like no other method, can be used to detect the unscheduled expression of cyclins, namely, the presentation of G1 cyclins by cells in G2/M and of G2/M cyclins by G1 cells, without the need for cell synchronization. Such unscheduled expression of cyclins B1 and A was seen when cell cycle progression was halted, e.g., after synchronization at the G1/S boundary by inhibitors of DNA replication. The unscheduled expression of cyclins B1 or E, but not of A, was also observed in some tumor cell lines even when their growth was unperturbed. Likewise, whereas the expression of cyclins D1 or D3 in nontumor cells was restricted to an early section of G1, the presentation of these proteins in many tumor cell lines also was seen during S and G2/M. This suggests that the partner kinase CDK4 (which upon activation by D-type cyclins phosphorylates pRB committing the cell to enter S) is perpetually active throughout the cell cycle in these tumor lines. Expression of cyclin D also may serve to discriminate G0 vs. G1 cells and, as an activation marker, to identify the mitogenically stimulated cells entering the cell cycle. Differences in cyclin expression make it possible to discriminate between cells having the same DNA content but residing at different phases such as in G2 vs. M or G2/M of a lower DNA ploidy vs. G1 cells of a higher ploidy. The expression of cyclins D, E, A and B1 provides new cell cycle landmarks that can be used to subdivide the cell cycle into several distinct subcompartments. The point of cell cycle arrest by many antitumor agents can be estimated with better accuracy in relation to these compartments compared to the traditional subdivision into four cell cycle phases. The latter applications, however, pertain only to normal cells or to tumor cells whose phenotype is characterized by scheduled expression of cyclins. As sensitive and specific indicators of the cell's proliferative potential, the cyclins, in particular D-type cyclins, are expected to be key prognostic markers in neoplasia.

2-Deoxy-D-glucose enhances sensitivity of human histiocytic lymphoma U937 cells to apoptosis induced by tumor necrosis factor.

It has been reported that the cytotoxic effect of tumor necrosis factor (TNF) on cells of several tumor cell lines was potentiated in culture media lacking glucose. Also, the antitumor effect of TNF was shown to be enhanced in vivo in mice treated with insulin to reduce their blood glucose level. The present study was aimed to reveal whether (a) the administration of the glucose antimetabolite 2-deoxy-D-glucose (2DG) has an effect similar to that of reduction of the extracellular glucose concentration; (b) the combined treatment with TNF and 2DG, similar to TNF alone, leads to apoptosis; and (c) there is a preference of cells in a particular phase of the cell cycle to undergo apoptosis in the presence of these agents. Exponentially growing human histiocytic lymphoma U937 cells were exposed to 0.1-0.5 nM of recombinant human TNF-alpha in the absence and presence of 1.0-5.0 mM 2DG. Analysis of the cell proliferation rates and their viability revealed that cytotoxicity of TNF was markedly potentiated by 2DG. Thus, administration of 1.0 mM 2DG to the cultures treated with 0.3 nM recombinant human TNF-alpha increased by 2-3-fold the percentage of dead cells after 24-72 h. The antimetabolite alone, at that low concentration, showed minimal cytotoxicity. More than additive cytotoxic effects also were seen at 2.5 and 5.0 mM concentrations of 2DG. Apoptosis was identified by typical changes in cell morphology, preferential degradation of internucleosomal DNA, and in situ extensive DNA strand breakage. The number of cells with DNA strand breaks after 24-h incubation was increased from 13% (0.1 nM TNF alone) to 20 or 45% in the presence of 2.5 or 5.0 mM 2DG, respectively. There was no evidence of a significant cell cycle phase preference in induction of apoptosis by combined treatment with recombinant human TNF-alpha and 2DG, although 2DG alone reduced the percentage of cells in S and G2 + M, apparently by arresting cells in G1. These data, along with observations in other cell systems, suggest that simultaneous stimulatory signals for growth induction, presumed to be provided by TNF, and growth suppression (inhibition of glycolysis) may preferentially trigger apoptosis of transformed cells. The data also suggest that 2DG may be an effective adjunct to TNF in the clinic, increasing the antitumor potency of this cytokine.

Unscheduled expression of cyclin B1 and cyclin E in several leukemic and solid tumor cell lines.

Normal, nontumorous cells express cyclin proteins in an orderly, scheduled fashion, at a given phase of the cell cycle. Thus, cyclin B1 is synthesized during G2 and abruptly degraded during mitosis. The onset of cyclin E synthesis takes place in mid-G1, its maximal expression is at the time of cell entrance to S, and its degradation occurs during cell progression through S phase. In the present study, multiparameter flow cytometry was used to correlate expression of cyclin B1 or cyclin E with cell cycle position (estimated by cellular DNA content) in normal human proliferating lymphocytes as well as in T-cell MOLT-4 leukemia; promyelocytic HL-60 leukemia; histiocytic U937 lymphoma; MCF-7, T-47D, and Hs 587T breast carcinoma; Colo 320DM colon carcinoma; and the T-24 transitional cell carcinoma cell line. The scheduled expression of both cyclins, namely of cyclin B1 restricted to G2 + M cells and of cyclin E restricted to late G1 and early S cells, was observed only in normal lymphocytes and MOLT-4 cells. The cells of HL-60, U937, T-47D, and Hs 587T lines expressed both cyclins in an unscheduled ("ectopic") fashion, i.e., unrelated to cell cycle position. Colo 320DM cells showed unscheduled expression of cyclin E (i.e., during G2) but expression of cyclin B1 in this line was generally restricted to G2 + M cells. There were relatively few (10-12%) cells in MCF-7 and T-24 cell lines that expressed cyclin B1 or E in an unscheduled manner. It may be expected that the unscheduled expression of cyclins in tumor cells may lead to a loss of the regulatory mechanisms of cell cycle progression and that such feature of the tumor may be of prognostic value. There is a need, therefore, to conduct similar studies in primary tumor cells.

Cyanide-induced lipid peroxidation in different organs: subcellular distribution and hydroperoxide generation in neuronal cells.

To evaluate hydroperoxide generation as a potential mechanism of cyanide neurotoxicity, mice were treated with KCN (7 mg/kg, subcutaneously (s.c.)) and the level of lipid peroxidation (expressed as conjugated dienes) was measured later in various organs. Brain showed elevated conjugated diene levels after cyanide but the liver, which is not considered a target for cyanide toxicity, showed no increase. The heart also showed no increase, whereas kidney conjugated dienes slowly increased to a peak 1 h after cyanide. In vitro studies show elevation of peroxidized lipids in mouse brain cortical slices following incubation with KCN (0.1 mM). Omission of calcium from the medium or pretreatment of brain slices with diltiazem (a calcium channel blocker) prevented formation of conjugated dienes by KCN. Calcium thus appears to play a critical role in cyanide-induced generation of peroxidized lipids in neuronal cells. Subcellular fractionation of brains from mice treated with cyanide showed that lipid peroxidation increased in the microsomal fraction but not in the mitochondrial fraction. Fluorescent studies using 2,7-dichlorofluorescein (a hydroperoxide sensitive fluorescent dye) show that hydroperoxides are generated rapidly after cyanide treatment of PC12 cells, a neuron-like cell, and hydroperoxide levels remain elevated for many minutes in the presence of cyanide. These results suggest that hydroperoxide generation with subsequent peroxidation of lipids may lead to changes in structure and function of certain membranes and contribute to the neurotoxic damage produced by cyanide.

Effect of staurosporine on MOLT-4 cell progression through G2 and on cytokinesis.

Staurosporine (SSP) is an inhibitor of a variety of protein kinases with an especially high affinity towards protein kinase C. Whereas SSP has been shown to halt the cell cycle progression of various normal, nontransformed cell types in G1, most virus transformed or tumor cells are unaffected in G1 but arrest in G2 phase. SSP has also been observed to increase the appearance of cells with higher DNA content, suggestive of endoreduplication, in cultures of tumor cells. Using multivariate flow cytometry (DNA content vs. expression of cyclin B, nuclear p120 protein, or protein reactive with Ki-67 antibody) which makes it possible to discriminate cells with identical DNA content but at different phases of the cycle, we have studied the cell cycle progression of human lymphocytic leukemic MOLT-4 cells in the presence of 0.1 microM SSP. MOLT-4 cells did not arrest in G1 or G2 phase in the presence of the inhibitor. Rather, they failed to undergo cytokinesis, entering G1 phase at higher DNA ploidy (tetraploidy; G1T), and then progressed through ST (rereplication) into G2T and MT. The rates of entrance to G2 and G2T were essentially identical, indicating that the rates of cell progression through S and ST as well as through G2 and G2T, respectively, were similar. Cells entrance to mitosis and mitotic chromatin condensation were also similar at the diploid and tetraploid DNA content level and were unaffected by 0.1 microM SSP. No evidence of growth imbalance (altered protein or RNA to DNA ratio) was observed in the case of tetraploid cells. The data show that, in the case of MOLT-4 cells, all events associated with the chromosome or DNA cycle were unaffected by SSP; the only target of the inhibitor appears to be kinase(s) controlling cytokinesis.

Caffeine prevents apoptosis and cell cycle effects induced by camptothecin or topotecan in HL-60 cells.

Caffeine (3,7-dihydro-1,3,7,-trimethyl-1H-purine-6,6-dione; CAF) is known to potentiate the cytotoxic effects of DNA damaging agents such as ionizing radiation and alkylating agents. In contrast, however, the cytotoxic and cytostatic activity of aromatic, DNA-intercalating, DNA topoisomerase II inhibitors such as Adriamycin, ellipticine, or mitoxantrone are diminished in the presence of CAF. To resolve whether the protective effect of CAF is associated with a particular mechanism of drug interaction (e.g., intercalation into DNA, inhibition of DNA topoisomerase II), or the aromatic nature of the drug structure, per se, we have presently studied the effects of CAF on the cytostatic and cytotoxic action of camptothecin (CAM) and its less toxic but more water soluble derivative topotecan (TPT) on HL-60 human myelogenous leukemia cells: both drugs have aromatic structures but are nonintercalating inhibitors of DNA topoisomerase I. By using spectroscopy and titration microcalorimetry, we have also studied the direct interaction between CAF and TPT in solution. Low (20 nM) concentrations of CAM or TPT perturbed progression of HL-60 cells through S-phase, whereas higher concentrations (0.15 microM) of these drugs induced apoptosis; both effects were easily demonstrable after 4 h of treatment. When added simultaneously with CAM or TPT, CAF prevented both effects. The protective effect of CAF was concentration dependent and evident within the concentration range of 1-5 mM; nearly total protection was seen at a CAF concentration of 5 mM. The bathochromic and hypochromic shift in the absorption spectrum of the water soluble compound TPT upon addition of CAF indicated that CAF and TPT interact (stack) in a fashion similar to that previously observed for CAF and DNA intercalators. Microcalorimetric measurements of TPT titration with CAF indicate an exothermic reaction between these compounds (the enthalpy change was delta H degree = -4.2 kcal/mol), which is consistent with a stacking model of CAF-TPT interaction. Thus, the ability of CAF to protect HL-60 cells against the cell kinetic effects of CAM or TPT, as in the case of DNA intercalating topoisomerase II inhibitors, is most likely due to formation of complexes between CAF and these aromatic molecules, which result in reducing the effective concentration of the free form of these drugs available to the cells.

The cell cycle related differences in susceptibility of HL-60 cells to apoptosis induced by various antitumor agents.

The studies were aimed to detect the cell cycle-associated differences in the susceptibility of HL-60 cells to apoptosis induced by diverse agents. Exponentially growing HL-60 cells were treated with the DNA topoisomerase I inhibitor camptothecin; the DNA topoisomerase II inhibitors teniposide, m-AMSA, Mitoxantrone, or Fostriecin; the presumed tyrosine kinase inhibitor genistein; a serine/threonine kinase inhibitor H7; the protein synthesis inhibitor cycloheximide; the DNA replication inhibitor hydroxyurea; the nucleoside antimetabolites 1-beta-D-arabinofuranosylcytosine and 5-azacytidine; and the alkylating agent nitrogen mustard, cisplatin, hyperthermia, and gamma irradiation. Endonucleolysis, which accompanied apoptosis induced by these agents, was assessed by two different flow cytometric methods, one based on DNA content measurements following extraction of low molecular weight DNA, and another using exogenous terminal deoxynucleotidyl transferase to label in situ DNA strand breaks. Each method allowed for both identification of apoptotic cells and analysis of the cell cycle distribution of the unaffected cell population; the method using terminal transferase also allowed for identification of the cell cycle position of apoptotic cells. Confirmed by analysis of DNA degradation by gel electrophoresis and changes in cell morphology, apoptosis was observed as early as 3 h after administration of most drugs and for some drugs was cell cycle phase specific. Cells progressing through S phase were selectively susceptible when treated with camptothecin, teniposide, m-AMSA, Mitoxantrone, H7, hydroxyurea, and 1-beta-D-arabinofuranosylcytosine. Cells in G2-M preferentially underwent apoptosis in cultures treated with H7 or with gamma-irradiation. Cells in G1 phase were preferentially affected by 5-azacytidine, nitrogen mustard, and hyperthermia. No significant cell cycle specificity was observed in the case of Fostriecin, genistein, cycloheximide, or cisplatin. The cell cycle related difference in susceptibility to apoptosis may be a reflection of both the severity of the lesion induced by a given drug and the ability of the cells to repair that lesion; both can vary depending on the cell cycle phase.

Different rates of degradation of the proliferation-associated nuclear and nucleolar antigens during apoptosis of hl-60 cells induced by DNA topoisomerase inhibitors.

One of the early events of apoptosis is proteolysis. The enzyme(s) involved in this event appear to be the serine protease(s), inasmuch as the reversible or irreversible inhibitors of serine proteases prevent DNA degradation and abrogate other signs of apoptosis in several cell systems (Gorczyca et al, Int J Oncol 1: 639-648, 1992). In the present studies, using multiparameter flow cytometry, we attempted to characterize the rate of disappearance of the nuclear and nucleolar proteins [Proliferating Cells Nuclear Antigen (PCNA), Ki-67, p120 and another nucleolar protein] detected by mononuclear antibodies, during apoptosis of HL-60 cells. Apoptosis was induced by the topoisomerase I inhibitor camptothecin and by the intercalating and nonintercalating topoisomerase inhibitors m-AMSA and teniposide, respectively, and was specific to cells in S phase. The loss of the protein reactive with Ki-67 antibody was the most rapid: two hours after administration of the drugs few S phase cells that expressed this protein remained in the cultures. Nearly all cells which were more advanced in apoptosis, and, due to extensive DNA degradation, had an already diminished DNA content, were negative with respect to expression of Ki-67. The nucleolar proteins, especially the one detected by the antibody distributed by Chemicon, appeared to be more stable compared to Ki-67. The most stable was PCNA: expression of this protein was high even after 6 h of treatment with each of the drugs, even in cells which had reduced DNA content. The data indicate that the proteolytic step is not entirely random and may involve different proteases having different substrate preferences or specific targeting of individual proteins for the degradation. Because spontaneous apoptosis appears to be common in tumors, the phenomenon of rapid degradation of some of the proliferation-associated antigens during apoptosis should be taken into account in the analysis of expression of these proteins in tumors (e.g. for estimation of the tumor growth fraction).

The protein-kinase-C inhibitor h7 blocks normal human lymphocyte stimulation and induces apoptosis of both normal lymphocytes and leukemia molt-4 cells.

The isoquinoline sulfonamide (H7) is an inhibitor of protein kinase C (PKC) that also inhibits the activity of cyclic nucleotide-dependent protein kinases. The effect of H7 on mitogen stimulation (G0 to G1 transition) of normal human lymphocytes and on their subsequent progression through the cell cycle was investigated and compared with the effect of this inhibitor on proliferation of human lymphocytic leukemic MOLT-4 cells. At H7 concentrations of 10 and 50 muM, the transition of G0 lymphocytes to the cell cycle was suppressed by 45 and 98%, respectively. The cell cycle progression of stimulated lymphocytes was unaffected at 10 muM H7, whereas, at 50 muM, the overall rate of progression was reduced by 50% with no evidence of cell arrest at a specific phase of the cycle. Similar concentrations of H7 (45 muM) suppressed proliferation of MOLT-4 cells by 50%, though, in the latter case, cells underwent transition to higher DNA ploidy, most likely via endoreduplication. Thus, the G0 to G1 transition appears to be the event most sensitive to H7. Exposure of MOLT-4 cells to 100 muM H7 for 24 h induced extensive apoptosis: activation of an endogenous nuclease with preference to internucleosomal linker DNA sections resulted in DNA degradation (revealed by agarose gel electrophoresis and loss of DNA measured by flow cytometry), which was paralleled by intracellular proteolysis, while the integrity of the plasma membrane, mitochondria and lysosomes was preserved. Morphological examination of these apoptotic cells confirmed DNA degradation. However, the perinuclear and fine-granular localization of the remaining DNA and lack of typical chromatin condensation and nuclear fragmentation differed from the classical pattern of apoptosis observed in other cell systems, suggesting that some events of apoptosis (nuclear fragmentation) may be affected by H7. The observed effects are consistent with the possible role of H7 in inhibition of PKC or its direct effect on the ATP-binding domain of DNA topoisomerase II, which shares homology with the H7 binding sites on PKC and the cyclic nucleotide-dependent protein kinases.

Effects of genistein on the growth and cell cycle progression of normal human lymphocytes and human leukemic MOLT-4 and HL-60 cells.

Genistein (GEN) is an isoflavone known to inhibit both tyrosine protein kinases and DNA topoisomerase II. The effects of GEN on cell proliferation and cell cycle kinetics of human myelogenous leukemia HL-60 and lymphocytic leukemia MOLT-4 cell cultures were studied, and the data were compared to results obtained with normal human lymphocytes stimulated to proliferate with phytohemagglutinin. GEN concentrations greater than 50 micrograms/ml (185 microM) were cytotoxic to HL-60 and MOLT-4 cells following exposure for 24 h; in HL-60 cell cultures, a population of cells with decreased DNA content and nuclear fragmentation characteristic of apoptosis was observed within 8 h. The 50% inhibition concentration after 24 h of exposure for HL-60 and MOLT-4 cells was 8.5 and 13.0 micrograms/ml, respectively. Normal proliferating lymphocytes survived a 24-h exposure of up to 200 micrograms/ml GEN. Short-term (4-8 h) exposures of MOLT-4 or HL-60 cells to 5-20 micrograms/ml GEN resulted in a suppression of cell progression through S or through both S and G2 phases, respectively, while equivalent treatment had no effect on proliferating lymphocytes. A stathmokinetic experiment using MOLT-4 cells revealed that as little as 5 micrograms/ml GEN suppressed cell exit from S to G2 phase by 40%, with a terminal point of action at or near the S-G2 border. Cell progression through the very early portion of G1 phase (G1A, characterized by postmitotic chromatin decondensation) was also suppressed by approximately 40%, whereas cell advancement through the remainder of the G1 phase was not markedly affected. Longer (24 h) exposure of proliferating lymphocytes to 20 micrograms/ml GEN led to an S-phase arrest, while similar treatment of leukemic cells caused cell arrest in G2 phase and an increase in the number of cells entering the cycle at higher DNA ploidy. The mitogen-induced transition of lymphocytes from G0 to G1 phase was extremely sensitive to inhibition by GEN; the 50% inhibition concentration was 1.6 micrograms/ml. The chemotherapeutic value of GEN may be due to the fact that, in terms of cytotoxicity, this agent is more active against proliferating leukemic cells than against normal proliferating lymphocytes. The sensitivity of the G0 to G1 transition in normal lymphocyte cultures and the suppressive effect of GEN on the G1A exit in MOLT-4 cells both suggest that protein kinases involved in chromatin decondensation may be a target of this drug. In light of the observation that lymphocyte stimulation is sensitive to the presence of GEN, the drug is expected to be a strong immunosuppressant.

Different effects of staurosporine, an inhibitor of protein kinases, on the cell cycle and chromatin structure of normal and leukemic lymphocytes.

Staurosporine, a microbial alkaloid, is a strong inhibitor of protein kinases. The effects of staurosporine on the cell cycle progression and nuclear morphology of normal human lymphocytes stimulated to proliferate by phytohemagglutinin were studied and compared with the effects of this drug on human lymphocytic leukemic MOLT-4 cells. Exposure of normal lymphocytes to either 5-10 or 50-100 ng/ml of staurosporine resulted in the preferential accumulation of cells in G1 or G1 and G2 phases of the cell cycle, respectively. In contrast, regardless of the concentration (5-100 ng/ml), staurosporine arrested MOLT-4 cells initially in G2; these cells then initiated additional rounds of DNA replication, without division. Staurosporine (5-100 ng/ml) induced severe changes in the nuclear morphology of MOLT-4 cells, manifested as nuclear elongation, deep invaginations of the nuclear membrane, extensive fragmentation, and micronucleation. At concentrations of 5-10 ng/ml, staurosporine had no apparent effect on the nuclear morphology of normal lymphocytes and at 50-100 ng/ml it produced minor changes in the nuclear shapes of these cells. The data indicate that the kinase(s) involved in the regulation of cell exit from G1 and G2, respectively, in normal and leukemic lymphocytes may have different sensitivities to staurosporine, which suggests that the mechanisms controlling exit from G1 in these cells may be different. In MOLT-4 cells the staurosporine-sensitive kinase(s) appear to also be involved in phosphorylation of nuclear constituents essential for organization of gross chromatin structure. The different response of normal versus leukemic lymphocytes to staurosporine, if confirmed on clinical material, opens new strategies of tumor treatment.

Cyanide induces Ca(2+)-dependent and -independent release of glutamate from mouse brain slices.

The effect of cyanide-induced histotoxic hypoxia on endogenous glutamate release from mouse cortical, cerebellar and hippocampal slices was studied. Incubation of slices with cyanide over a 30 min period resulted in extracellular accumulation of glutamate which was decreased in the absence of Ca2+ in the incubation media. When glutamate release was continuously monitored by fluorometry, cyanide initiated a rapid release of glutamate. This initial release was found to be independent of extracellular calcium. Depolarizing concentrations of potassium chloride produced a predominantly Ca(2+)-dependent release. It is concluded that cyanide exposure induced a rapid release of endogenous glutamate mediated by both Ca(2+)-dependent and Ca(2+)-independent mechanisms.