Comparison between the binding of [3H]misonidazole and AF-2 in Chinese hamster V79 spheroids

Binding of two hypoxia probes, [3H]misonidazole and AF-2 (2-(2-Furyl)-3-(5-nitro-2-furyl)acrylamide), was compared in Chinese hamster V79 spheroids incubated under different oxygen concentrations. Fluorescence-activated cell sorting based on Hoechst 33342 penetration was used to obtain populations of cells from different depths within the spheroid, and sorted cells were analyzed by cytofluorometry for AF-2 content and by liquid scintillation counting for [3H]misonidazole content. The patterns of AF-2 and misonidazole binding were very similar, with about 20-fold more localization of both drugs in anoxic compared to aerobic cells. Similar results were obtained when cells were sorted on the basis of AF-2 rather than Hoechst 33342 fluorescence. When mean cellular fluorescence of AF-2 was plotted versus cpm misonidazole/cell for different oxygen tensions, it appeared that oxygen was equally effective in inhibiting AF-2 and misonidazole binding. Internal cells of anoxic spheroids bound about twice as much AF-2 and misonidazole as external cells, apparently due to an increased rate of nitroreduction by chronically hypoxic cells. AF-2 was found to enhance the retention of misonidazole in oxic and hypoxic spheroids when both drugs were present.     

Modification of cell survival and transformation by exposure to anisotonic solutions during irradiation

C3H 10T 1/2 cells were exposed to a wide range of anisotonic NaCl solutions and were irradiated during the last few minutes of the salt treatment. Radiosensitization in terms of cell killing and transformation was observed for hypotonic NaCl treatment. Hypertonic treatment with 0.5 mol/liter NaCl also caused radiosensitization for cell killing and transformation, while extreme hypertonic treatment (1.5 mol/liter) resulted in radioprotection for both end points. Radiosensitization for both end points declined as the hypotonic treatment (0.05 mol/liter) was prolonged. To a lesser extent, a decline in radioprotection by 1.5 mol/liter treatment occurred as exposure time increased.     

Radiosensitizing and cytotoxic properties of misonidazole on glutathione synthetase deficient human fibroblasts

The cytotoxic and radiosensitizing effects of misonidazole have been studied on glutathione synthetase deficient fibroblasts and on their controls. At any concentration from 0.1 to 4 mM, deficient cells are more sensitive to the cytotoxic effect of misonidazole than the control cells. The differential effect between the two cell strain concerns both the shoulder and the slope of the survival curve, thus suggesting that NPSH play a role in the determination of misonidazole cytotoxicity. Like oxygen, misonidazole clearly sensitizes deficient cells to a lesser extent than control cells. For both cell strains, the maximum sensitizing effect of misonidazole is very close to that of oxygen (1.5 and 1.5 for deficient cells, 2.8 and 2.9 for control cells, respectively). The sensitizing effect of misonidazole appears in the same concentration range for both cell strains, with a maximal effect at lower concentrations for deficient cells.     

Sensitivity of a human hybrid cell line (HeLa X skin fibroblast) to radiation-induced neoplastic transformation in G2, M, and mid-G1 phases of the cell cycle

The dependence of gamma-radiation-induced neoplastic transformation frequency on position in the cell cycle was measured for a human hybrid cell line (HeLa X skin fibroblast). The end point used was the induction of a tumor-associated antigen which in these cells correlates with tumorigenicity. Induction was measured in cells at G2, M, and mid-G1 phases and compared with the frequency induced in asynchronous cells. For studies of cells in G2 phase, the cells of an asynchronous population were collected for 3 h post-irradiation using the mitotic shake-off technique. For studies of cells in M and mid-G1 phases, cells were collected by mitotic harvest and then treated at the appropriate time. The data show that cells in G2 and M phase are very radiosensitive in terms of both cell killing and induction of neoplastic transformation compared to cells in mid-G1 or asynchronous populations. At a dose of 1 Gy, the transformation frequency was 10- to 20-fold higher for cells in M and G2 phase than for cells in mid-G1 or for asynchronous cells. However, the data indicate that the transformation frequencies were similar in the different phases of the cell cycle when correlated with surviving fraction. The results indicate that transformation frequency is more sensitive to changes in dose than is cell survival.     

Postirradiation modification of tumor blood flow: a method to increase the effectiveness of chemical radiosensitizers

The effect of postirradiation hypoxia induced by administration of the vasodilator hydralazine on the efficacy of misonidazole and RSU-1069 used in combination with radiation has been evaluated. Studies with the Lewis lung carcinoma indicate that hydralazine at a dose of 5 mg/kg reduces tumor blood flow and consequently increases the amount of hypoxia in the tumor tissue. Administration of hydralazine immediately after radiation treatment increased the amount of cell kill. However, the increase in cell kill was more pronounced when hydralazine was used in treatment regimes in which misonidazole (0.2 mg/g) or RSU-1069 (0.02 mg/g) was administered pre- or postirradiation. The finding that similar effects are observed if the nitroimidazoles were administered either before or after radiation in the regimes involving hydralazine suggests that the enhanced cell killing observed is due to hypoxic cell cytotoxicity. In contrast to the effects of hydralazine on the response of tumors to radiation plus misonidazole or RSU-1069, it has no effect on the response of mouse intestine to such treatment regimes. Thus therapeutic gain may accrue from the use of hydralazine in radiation treatments which incorporate the nitroimidazole radiosensitizers misonidazole and RSU-1069.     

Radiosensitizing and cytocidal effects of misonidazole: evidence for separate modes of action

Hypoxic BP-8 murine sarcoma cells were exposed to misonidazole and/or radiation and the kinetics and extent of cell death were evaluated with the [125I]iododeoxyuridine-prelabeling assay. Cell death after treatment with lethal doses of misonidazole was rapid and essentially complete within 2 or 3 days after drug exposure. In contrast, radiation death became apparent only after a delay period of 4 days and was complete by Day 10 after irradiation. Radiosensitization by short exposures to sublethal doses of misonidazole affected only the delayed component of cell death, that is, the radiation component of death. In experiments involving sequential radiation and drug treatment, prior irradiation of cells did not enhance the direct cytocidal effects of misonidazole, as evidenced by the fact that the early component of cell death was equal in control and preirradiated cells. However, postirradiation treatment with misonidazole did enhance the delayed radiation component of cell death. These results suggest that radiosensitization and direct killing by misonidazole are two distinct phenomena mediated by different cellular mechanisms, and radiosensitization by misonidazole represents a two-component effect composed of true dose modification and dose additive damage interactions, but these additive effects must occur at a site different from the cellular structure responsible for direct drug-induced cell death.     

Measurements of DNA double-strand break yields in E. coli after rapid irradiation and cell inactivation: the effects of inactivation technique and anoxic radiosensitizers

A study has been made of methods for rapidly inactivating cells of E. coli at neutral pH to prevent enzymatic, chemical, or physical modification to DNA damaged by irradiation. The radiation was delivered in a fraction of a second using an electron accelerator. Cell inactivation was with ethanol or a solution (CSE) containing detergent, EDTA, and chloroform. It was found that DNA could be released from irradiated and inactivated cells simply by incubating them with the protease Pronase, and this DNA appeared to be in a form suitable for centrifugational analysis in neutral sucrose gradients. When cells were irradiated in the presence of oxygen, inactivation with ethanol gave a radiation-induced double-strand break (dsb) yield 1.8-fold higher than when inactivation was with CSE. Possible explanations of this are discussed. Using CSE inactivation, an oxygen enhancement ratio for dsb formation of 4.3 was observed and yields of dsb per Gray were in good agreement with results from other laboratories. At concentrations which enhanced cell killing 2.6-fold the "electron-affinic" type anoxic radiosensitizer misonidazole enhanced dsb formation 2.0-fold, whereas the nitroxyl free radical anoxic radiosensitizer norpseudopelletierine-N-oxyl had no significant effect on dsb yield although there was a possible slight enhancement at the higher doses used.     

Mutational analysis of Vpr-induced G2 arrest, nuclear localization, and cell death in fission yeast

Cell cycle G2 arrest, nuclear localization, and cell death induced by human immunodeficiency virus type 1 Vpr were examined in fission yeast by using a panel of Vpr mutations that have been studied previously in human cells. The effects of the mutations on Vpr functions were highly similar between fission yeast and human cells. Consistent with mammalian cell studies, induction of cell cycle G2 arrest by Vpr was found to be independent of nuclear localization. In addition, G2 arrest was also shown to be independent of cell killing, which only occurred when the mutant Vpr localized to the nucleus. The C-terminal end of Vpr is crucial for G2 arrest, the N-terminal alpha-helix is important for nuclear localization, and a large part of the Vpr protein is responsible for cell killing. It is evident that the overall structure of Vpr is essential for these cellular effects, as N- and C-terminal deletions affected all three cellular functions. Furthermore, two single point mutations (H33R and H71R), both of which reside at the end of each alpha-helix, disrupted all three Vpr functions, indicating that these two mutations may have strong effects on the overall Vpr structure. The similarity of the mutant effects on Vpr function in fission yeast and human cells suggests that fission yeast can be used as a model system to evaluate these Vpr functions in naturally occurring viral isolates.     

A model of cell killing by low-dose-rate radiation including repair of sublethal damage, G2 block, and cell division

A computer model that simulates the killing of exponentially growing cells by low-dose-rate radiation is described. The model incorporates cell killing by single-hit damage and double-hit (sublethal) damage, as well as repair of sublethal damage, delay of cell cycle progression, blockage and increased sensitivity of cells in the G2 phase of the cell cycle, and cell division. Seven cellular parameters determine the rate of cell killing. Initial estimates of most of these parameters can be made from independent experiments. Parameters were obtained that gave the best fit to the data for four cell lines, using constant or variable dose rates, and using as end points either the fraction of single cells forming colonies or the total number of clonogenic cells in a mass culture. Some of the parameters were determined to be insignificant or similar for the four cell lines. The main differences between the cell lines in patterns of cell killing over a range of dose rates appeared to be determined by differences in the values of four of the parameters.     

Induction and repair of DNA damage in gamma-irradiated human lymphoblasts: irradiation in the presence and absence of misonidazole

The effects of oxygen and misonidazole on the induction of DNA lesions were examined in human TK6 lymphoblasts irradiated with 60Co gamma rays. We have investigated both the formation and subsequent repair of two classes of DNA damage, single-strand breaks and lesions recognized by the gamma endonuclease activity in a cell-free extract of Micrococcus luteus. Relative to irradiation under hypoxia, single-strand break yields were increased by the presence of either oxygen or misonidazole at the time of irradiation. In contrast, M. luteus enzyme-sensitive site yields were unaffected by the presence of either oxygen or misonidazole. No significant differences in single-strand break or enzyme-sensitive site repair kinetics were observed for lesions induced under any of the irradiation conditions employed. These results confirm the sensitizing effects of oxygen and oxygen-mimetic drugs on the induction of single-strand breaks but provide no support for their ability to enhance the induction of enzyme-sensitive sites.     

Radiolabeled hypoxic cell sensitizers: tracers for assessment of ischemia

Hypoxic, non-functional, but viable, tissue may exist in heart and brain following an arterial occlusion. Identification of such tissue in vivo is crucial to the development of effective treatment strategies. It has been suggested that certain compounds capable of sensitizing hypoxic tumor cells to killing by x-rays (i.e., misonidazole) might serve as in vivo markers of hypoxic tissue in ischemic myocardium or brain if properly radiolabeled. To this end we have radiolabeled two fluorinated analogs of nitroimidazole based hypoxic cell sensitizers with the 110 minute half-lived positron-emitting fluorine-18. The ability of these tracers to quantitate the presence of hypoxic tissue has been studied in a gerbil stroke model. The in vivo uptake of one of these tracers [F-18]-fluoronormethyoxymisonidazole is dependent on the extent of tissue hypoxia, and thus, appears to have potential as a diagnostic indicator of non-functional but viable tissue when the tracer is used in conjunction with positron emission tomography.     

Reduced bone marrow pO2 following treatment with radioprotective drugs

The sensitizer adduct technique [( 3H]misonidazole binding) was used to assess the extent of murine bone marrow hypoxia following treatment with a variety of radioprotectors. The binding rates previously determined in vivo were compared to those obtained by incubating marrow cells in atmospheres of varying oxygen content. Parallel experiments demonstrated that the oxygen dependence of [3H]misonidazole binding (Km approximately 0.15% oxygen) was similar to the oxygen dependence of marrow radiosensitivity (Km approximately 0.2% oxygen). Maximally radioprotective doses of several drugs have been shown to increase the binding of [3H]misonidazole significantly in vivo. A comparison to the in vitro binding rates suggests that the average oxygen concentration in the marrow at times associated with radioprotection was on the order of 0.5 to 0.8% oxygen. The relative importance of marrow hypoxia to the overall radioprotective effects of different drugs may vary considerably. However, these results have demonstrated that certain radioprotective drugs can induce marrow hypoxia and this reduced pO2 may contribute to the efficacy of these agents.     

Generation of radical anions of nitrofurantoin, misonidazole, and metronidazole by ascorbate

Nitrofurantoin, misonidazole, and metronidazole were reduced to their corresponding nitro anion radicals by ascorbate in anaerobic solutions at high pH. The nitrofurantoin anion radical could be detected at neutral pH. In neutral solutions, the nitro anion radicals of misonidazole and metronidazole were too unstable to be observed by electron spin resonance spectroscopy. At neutral pH, solutions containing ascorbate, nitrofurantoin, or misonidazole consumed oxygen. The addition of superoxide dismutase, catalase, or both superoxide dismutase and catalase decreased the rate of oxygen consumption. These results show that nitro anion radicals are formed by reduction with ascorbate, and superoxide anion radical and hydrogen peroxide are produced by reactions of these radicals with oxygen.     

Inhibitors of repair of potentially lethal damage and DNA polymerases also influence the recovery of potentially neoplastic transforming damage in C3H10T-1/2 cells

The effects of aphidicolin and beta Ara A on radiation sensitivity were evaluated in terms of cell killing, recovery, and neoplastic transformation in the C3H10T-1/2 cell system. When cells were held in plateau phase, recovery of potentially lethal damage (PLD) and potentially transforming damage (PTD) occurred. The addition of beta Ara A resulted in reduced PLD recovery for both the survival and neoplastic transformation end points. The addition of aphidicolin did not affect recovery of PLD or PTD. These data show that the inhibition of polymerase alpha by aphidicolin does not affect recovery of damage leading to cell death or neoplastic transformation. However, the inhibition of both polymerase alpha and beta by beta Ara A resulted in inhibition of recovery of damage leading to both cell death and neoplastic transformation. These data indicated that polymerase beta may be involved in both PLD and PTD recovery.     

Killing of EMT-6 cells by decays from isotopes incorporated on sensitizer adducts.

EMT-6 tumor cell killing by decays from 3H and 125I incorporated by adduct formation of radiolabeled sensitizers was studied in vitro. Hypoxic radiosensitizers become covalently bound to cellular molecules after metabolic reduction, and EMT-6 tumor cells can tolerate over 10(9) adducts/cell of misonidazole without loss of colony-forming ability. Cells were incubated under hypoxic conditions in the presence of [3H]misonidazole or [125I]iodoazomycinriboside for various times and the amounts of bound 3H and 125I were determined. Cells were stored as monolayers at 22 degrees C, in suspension culture at 4 degrees C, and frozen in complete medium plus 8% DMSO at -196 degrees C for various times to facilitate the accumulation of radioactive decays before plating in vitro for colony-forming assays at 37 degrees C. At 22 degrees C in monolayer culture, EMT-6 tumor cells tolerated 950 and 1720 decays/cell of 3H and 125I, respectively, without evidence of radiotoxicity. This number of decays/cell over the exposure times used represents 1.54 x 10(6) 3H/cell and 8.4 x 10(4) 125I/cell, respectively. Significant cell killing was detected after similar amounts of isotope decay when cells were held at 4 degrees C. When cells were frozen in the presence of 8% DMSO, they were more resistant to inactivation by isotope decays or by gamma rays than cells in liquid phase at 4 degrees C. These data suggest that selective hypoxic tumor cell suicide by 3H or 125I decays from bound sensitizer at 37 degrees C will be an inefficient process, at least for drugs with specific activities as tested. These data are consistent with data on cell inactivation by isotopes incorporated into cells by other procedures.     

Failla Memorial Lecture. From beans to genes--back to the future.

Developments in radiation biology have inevitably paralleled the evolution of systems and end points in the wider field of biology. I review the development of three areas that have interested me during the 35 years that I have been involved with radiation research. (1) The dose-rate effect. My first graduate student, Joel Bedford, and I demonstrated the dependence of mammalian cell killing on the dose rate at which gamma rays are delivered. These studies led to the recent development of pulsed low-dose-rate brachytherapy and the design of a new machine for clinical use. (2) Hypoxic radiosensitizers and bioreductive drugs. The hypothesis that the presence of foci of hypoxic cells in human tumors could limit their curability by gamma rays led to the development of nitromidazales that preferentially sensitize hypoxic cells. Our contributions included the observation that misonidazole was also cytotoxic to hypoxic cells, that sensitization was increased by prolonged preincubation, and that the mechanism involved depletion of thiols. More recently we have collaborated with Ged Adams and Bob Sutherland in the development of a new generation of hypoxic cell cytotoxins. In particular, we have been concerned with the assessment of oncogenicity in relation to drug structure. (3) Oncogenic transformation in vitro. The use of in vitro assays for oncogenic transformation has represented a major interest for this laboratory largely due to collaborations first with Carmia Borek, and more recently with Richard Miller and Tom Hei. We published the first dose-response curves for gamma rays, and more recently for a range of neutron energies and for charged particles of defined LET. Radiation-induced oncogenic transformation in C3H 10T1/2 cells has been shown to be due to a dominant acting transforming gene, which has been isolated and is in the process of being characterized and sequenced by Greg Freyer.     

Influence of cellular environment on toxicity of nitroheterocycles.

The preferential sensitivity of hypoxic cells to nitroheteroxycles is thought to result from the actions of toxic intermediates of drug reduction produced under hypoxic conditions. However, a lack of oxygen also alters the biochemical state of the cell and may indirectly enhance the sensitivity, of hypoxic cells to these drugs. This hypothesis was tested by 'conditioning' mouse L-929 cells in oxygen-free buffer, then exposing the cells to nitrofurazone under both aerobic and anaerobic conditions. After conditioning, the rate of cell inactivation by nitrofurazone was equal in air or nitrogen-equilibrated buffer. Pretreatment of cells in 1 muM rotenone or 0.5 mM 2,4-dinitrophenol for one hour under aerobic conditions increased the sensitivity of the cells to nitrofurazone under aerobic conditions. Similar rates of cell killing were obtained when mouse L-cells were heated in buffer for 30 min at 43 degrees before incubation with nitrofurazone in either air or nitrogen. Also, incubation of cells with nitrofurazone in the presence of 0.1% glucose, or at a cell density less than 10(5) cells/ml significantly enhanced cell killing, especially under aerobic conditions. Thus, the intracellular state of the cell, manipulated by altering the cellular environment, influenced the cellular sensitivity to nitrofurazone. Similar results were not, however, obtained with the nitroimidazoles, dimetronidazole and misonidazole; pretreatment for 2 h in buffer under anaerobic conditions did not increase the sensitivity of L cells to subsequent drug treatment in air-equilibrated buffer.     

Differential sensitivities of transformed and untransformed murine cell lines to DNA cross-linking agents relative to repair of O6-methylguanine

Sensitivities of several murine cell strains to killing by the DNA cross-linking agents 1-(2-chloroethyl)-1-nitrosourea (CNU), cis-diamminedichloroplatinum (II) (Cis-Pt) and mitomycin C (MMC) were measured by post-treatment colony-formation. Virally-transformed murine cells were usually more sensitive to cell killing by these agents than were the parental 3T3 cell strains. The hypersensitivity to CNU of some virally-transformed murine cell strains correlated well with the reduced ability to repair O6-methylguanine (O6mGua), a phenomenon similar to that in human cells. The loss of ability to repair O6mGua, as well as the increased sensitivity of transformed strains to cell killing, may not be due to a mutation but rather due to a change of gene expression associated with transformation by viruses or activation of oncogenes.     

Factors associated with the preincubation effect of hypoxic cell sensitizers in vitro and their possible implications in chemosensitization

The enhancement of melphalan toxicity was observed by preincubation of V-79- 379A cells in spinner culture with multiple doses of misonidazole (miso) or SR-2508 under hypoxic conditions. Chemosensitization was shown to be a function of sensitizer concentration and duration of exposure to the alkylating agent. A preincubation exposure of cells with 5 mM miso reduced endogenous cell thiols to less than 5% of controls and enhanced melphalan toxicity by a factor of 4.7. Cells preincubated with miso not only had lower levels of nonprotein thiols, but also were shown to have altered levels of intracellular calcium and a lower threshold to oxidative stress as measured by toxicity to cysteamine or H2O2. Preincubated cells, hypoxic cells, and cells receiving moderate hyperthermia (42.5 degrees C for 3 hr) all showed increased sensitivity to either cysteamine or H2O2. The increased killing of preincubated cells by cysteamine was shown to be similar to that of H2O2, and the dramatic reduction of cysteamine toxicity by catalase indicated H2O2 was the major reaction associated with this effect. These results indicate that preincubated cells exhibit a variety of biological effects that may significantly influence their response to further treatment with drugs or radiation, especially where peroxidative and free radical mechanisms are involved. The depletion of endogenous thiols, calcium disturbance, and vulnerability to oxidative stress are factors to be considered when interpreting mechanisms of combined drug action and effects that may potentially be exploited in terms of therapeutic gains.     

Changes in the pigment patterns and the photosynthetic activity during a light-induced cell cycle of the green alga Scenedesmus armatus

The photosynthetic oxygen evolution as well as the chlorophyll and carotenoid patterns were studied during the light phase (14 h) of the Scenedesmus armatus cell cycle. The alga was synchronised by the light/dark regime (14/10 h). In this publication, the term “cell cycle” refers to this period of light only. The oxygen evolution measured by a Clark-type electrode and expressed per cell, gradually increased from the beginning of the cell cycle, reaching its maximum at 12 h and then slowly declined towards the end of the cell cycle. This pattern reflects the final reproductive events of the cell cycle consisting of the third mitotic division, chloroplast and protoplast fission, followed by the formation of the autospores. When the same amount of oxygen was expressed per chlorophyll a content, we observed a rapid increase just after the onset of light, which reached a maximum at the third hour, after which, this slowly declined until the end of the cell cycle. A similar pattern for the relative quantum yield of oxygen evolution was obtained when a photobaric component of a photoacoustic signal was analysed by the photoacoustic spectroscopy method. The most abundant carotenoid was lutein. Much smaller amounts of α-carotene, β-carotene, loroxanthin, violoxanthin and neoxanthin were noted; traces of zeaxanthin and antheraxanthin were also identified. The photosynthetic efficiency and the ratio of lutein/α-carotene followed the same patterns during the cell cycle and similar relationships were also observed in the ratio changes of violoxanthin/β-carotene and violoxanthin/neoxanthin. The most photosynthetic-efficient cells contained the highest level of lutein, and had a much lower violoxanthin content. The content of neoxanthin and β-carotene found was lower, with both pigments still being present in similar amounts. These results suggest that the molecular organisation of LHC IIb mainly determined the photosynthetic efficiency of algae during its light-induced cell cycle.