Bioreductive metabolism of AF-2[2-(2-furyl)-3-(5-nitro-2-furyl)acrylamide] combined with 2-nitroimidazoles. Implications for use as hypoxic cell markers

Quantification of Tumor Hypoxic Fractions Using Positron Emission Tomography with [18F]Fluoromisonidazole ([18F]FMISO) Kinetic Analysis and Invasive Oxygen Measurements

PURPOSE

The purpose of this study is to use dynamic [¹⁸F]fluoromisonidazole ([¹⁸F]FMISO) positron emission tomography (PET) to compare estimates of tumor hypoxic fractions (HFs) derived by tracer kinetic modeling, tissue-to-blood ratios (TBR), and independent oxygen (pO₂) measurements.

PROCEDURES

BALB/c mice with EMT6 subcutaneous tumors were selected for PET imaging and invasive pO₂ measurements. Data from 120-min dynamic [¹⁸F]FMISO scans were fit to two-compartment irreversible three rate constant (K ₁, k ₂, k ₃) and Patlak models (K _i). Tumor HFs were calculated and compared using K _i, k ₃, TBR, and pO₂ values. The clinical impact of each method was evaluated on [¹⁸F]FMISO scans for three non-small cell lung cancer (NSCLC) radiotherapy patients.

RESULTS

HFs defined by TBR (≥1.2, ≥1.3, and ≥1.4) ranged from 2 to 85 % of absolute tumor volume. HFs defined by K _i (>0.004 ml min cm^-3) and k ₃ (>0.008 min^-1) varied from 9 to 85 %. HF quantification was highly dependent on metric (TBR, k ₃, or K _i) and threshold. HFs quantified on human [¹⁸F]FMISO scans varied from 38 to 67, 0 to 14, and 0.1 to 27 %, for each patient, respectively, using TBR, k ₃, and K _i metrics.

CONCLUSIONS

[¹⁸F]FMISO PET imaging metric choice and threshold impacts hypoxia quantification reliability. Our results suggest that tracer kinetic modeling has the potential to improve hypoxia quantification clinically as it may provide a stronger correlation with direct pO₂ measurements.

Dosimetry study of 18F-FMISO + PET/CT hypoxia imaging guidance on intensity-modulated radiation therapy for non-small cell lung cancer

OBJECTIVES

This study was to evaluate the feasibility of simultaneous integrated boost on tumor hypoxia area by studying the dosimetric change of hypoxia imaging guidance on intensity-modulated radiation therapy for non-small cell lung cancer (NSCLC).

METHODS

Five NSCLC patients with large hypoxic volume participated in this study. FDG PET/CT images were fused with CT localization images to delineate gross tumor volume. FMISO PET/CT images were fused with CT localization images to delineate hypoxic biological target volume (BTV) (tissue maximum ratio ≥ 1.3) by threshold. BTV was irradiated with 72, 78 and 84 Gy, respectively, 30 times. The dosimetry differences were compared in target volume and organ at risk between simultaneous integrated boost plans and conventional radiotherapy plans.

RESULTS

Dosages on BTV of NSCLC hypoxic area were increased to 72, 78 and 84 Gy, respectively, by simultaneous integrated boost intensity-modulated radiation therapy. There was no obvious difference in dosage distributions on original target volume compared with those in conventional radiotherapy. Dosages on main organ at risk in chest met the dosimetric constraint, and there was no significant difference compared with those in conventional radiotherapy.

CONCLUSION

It is feasible in dosiology that the dosages in NSCLC hypoxic area were added to 72, 78 and 84 Gy by simultaneous integrated boost with the guidance of ¹⁸F-FMISO PET/CT.

Patterns and levels of hypoxia in head and neck squamous cell carcinomas and their relationship to patient outcome

PURPOSE

EF5, a 2-nitroimidazole hypoxia marker, was used to study the presence, levels, and prognostic significance of hypoxia in primary head and neck squamous cell tumors.

METHODS AND MATERIALS

Twenty-two patients with newly diagnosed squamous cell carcinoma of the oral cavity, oropharynx, or larynx with at least 2 years of clinical follow-up were included in this study. Quantitative analyses of EF5 immunofluorescence was carried out, and these data were compared with patient outcome.

RESULTS

EF5 immunostaining showed substantial intra- and intertumoral hypoxic heterogeneity. The majority of cells in all tumors were well oxygenated. Three patterns of EF5 binding in cells were identified using criteria based on the cellular region that was stained (peripheral or central) and the relationship of binding to necrosis. We tested the association between EF5-binding levels with event-free and overall survival irrespective of the pattern of cellular binding or treatment regimen. Patients with tumors containing EF5-binding regions corresponding to severe hypoxia (< or =0.1% oxygen) had a shorter event-free survival time than patients with pO(2) values greater than 0.1% (p = 0.032). Nodal status was also predictive for outcome.

CONCLUSIONS

These data illustrate the potential utility of EF5 binding based on quantitative immunohistochemistry of tissue pO(2) and provide support for the development of noninvasive hypoxia positron emission tomographic studies with fluorine 18-labeled EF5.

Oxygen levels in normal and previously irradiated human skin as assessed by EF5 binding

The oxygen status of skin is a controversial topic. Skin is radiosensitive, suggesting it is well-oxygenated. However, it can be further sensitized with nitroimidazole drugs, implying that it is partially hypoxic. Skin oxygen levels are difficult to measure with either electrodes or the hypoxia-monitoring agent (3)H-misonidazole. For the latter, binding has previously been reported to be high in murine skin, but this could be attributed to either non-oxygen-dependent variations in nitroreductase activity, drug metabolism, and/or actual oxygen gradients. We obtained tumor and skin from patients given EF5, a 2-nitroimidazole tissue hypoxia monitor. We performed immunohistochemical studies using highly specific monoclonal antibodies for the hypoxia-dependent production of EF5 tissue adducts. Some tissue sections were counterstained using either Ki67 for proliferation or CD31 for vessels. We found that the human dermis is well-oxygenated, the epidermis is modestly hypoxic and portions of some sebaceous glands and hair follicles are moderately to severely hypoxic. Normal and irradiated skin had similar oxygenation patterns. Control studies demonstrated that these observations are not due to tissue variations in nitroreductase activity. The importance of the highly heterogeneous distribution of oxygen in skin requires further study, but recent investigations suggest that skin hypoxia may have important clinical ramifications including mediating cellular transformation.

Reduced Expression of Hypoxia-Inducible Factor-1α in Perinecrotic Regions of Solid Tumors

Hypoxia that develops in solid tumors stabilizes the hypoxia-inducible factor-1alpha (HIF-1alpha) subunit of the HIF-1 transcription factor, leading to up-regulation of dozens of hypoxia-regulated genes that increase glycolysis and oxygen delivery. HIF-1alpha and its downstream target gene CA9 have both been used as surrogate hypoxia markers, and, in general, high expression predicts for a poor response to treatment. Combinations of hypoxia markers offer the opportunity to measure changes in tumor oxygenation that may be relevant to tumor response to treatment. We compared the degree of colocalization of two endogenous markers for hypoxia, HIF-1alpha and carbonic anhydrase IX (CAIX), with a chemical marker for hypoxia, pimonidazole. Unexpectedly, expression of HIF-1alpha was reduced in the most hypoxic regions that border necrosis in xenograft tumors composed of SiHa cervical carcinoma, WiDr colon carcinoma, or M006 astrocytoma cells. Similar results were obtained for samples from three cervical cancer biopsies. However, CAIX was present in these perinecrotic cells that were also capable of metabolizing and binding a chemical marker for hypoxia, pimonidazole. In vitro experiments using tumor cells and tumor cubes incubated under anoxic conditions indicated that nutrient deprivation seems to be largely responsible for the lack of HIF-1alpha expression in perinecrotic regions. The half-life of CAIX was sufficiently long that, once formed, it remained for days in the absence of continued HIF-1alpha expression. These results have implications for the use of HIF-1alpha as an indicator of tumor hypoxia and aggressiveness as well as development of hypoxia-directed antitumor therapies based on the expression of HIF-1alpha.

Prospects for bioreductive drug development

Bioreductive drugs are inactive prodrugs that are converted into potent cytotoxins under conditions of either low oxygen tension or in the presence of high levels of specific reductases. The biochemical basis for selectivity relies on the ability of oxygen to reverse the activation process and the presence of elevated reductase levels in some tumour types. Key criteria for an ideal bioreductive drug should include poor activity against aerobic cells, activation over a broad range of oxygen tensions and, penetration through the aerobic fraction of cells. In addition, the active drug should be capable of killing non-proliferating cells. Numerous compounds are currently at various stages of drug development but Mitomycin C, which is generally considered to be the prototype bioreductive drug, is the only one in clinical use today. Of the drugs currently being evaluated clinically, tirapazamine has definite clinical activity against a variety of solid tumours when used in combination with cisplatin. Other drugs, such as EO9 and various nitroimidazoles, have not been impressive in the clinic and further development is required to improve properties such as drug delivery in the case of indoloquinones. A novel approach to exploiting tumour hypoxia is the development of a gene-directed enzyme prodrug therapy (GDEPT) strategy, where a gene encoding for a prodrug activating enzyme has been placed under the control of a hypoxia responsive promoter sequence. It is generally recognised that bioreductive drugs must be directed towards patients whose tumours have hypoxic regions or have appropriate enzymological characteristics. In terms of identifying tumour hypoxia, there has been considerable progress in the development of nitroimidazole based hypoxia markers that can be detected either via non-invasive or invasive procedures. Another strategy currently undergoing preclinical evaluation is the use of agents that modulate tumour blood flow and synergistic effects have been reported between bioreductive drugs and photodynamic therapy or inhibitors of nitric oxide synthase for example. The development of clinically useful bioreductive drugs depends therefore on the expertise of scientists and clinicians with varying backgrounds. The purpose of this review is to describe and critically assess recent developments in this field, with particular emphasis being placed on drug development and strategies aimed at optimising bioreductive drug activity.

Hypoxia and VEGF mRNA expression in human tumors

High expression of circulating plasma vascular endothelial growth factor (VEGF) in patients with cancer is an indicator of poor treatment response. Similarly, hypoxia in tumors, as measured by oxygen needle electrodes, has been found to predict for tumor-treatment failure. These two predictors may be related because hypoxia is a potent stimulator of VEGF expression in vitro. However, the demonstration of a relationship between hypoxia and VEGF in human tumors has, to date, been indirect or even negative. The purpose of this study was to test whether this unexpected result was caused by factors unique to human tumors, or whether the prior results could have been influenced by the known complexities of VEGF regulation. Therefore, we undertook a direct assessment of VEGF induction in human tumors using in situ hybridization and compared its distribution with that of hypoxia, as measured by the distribution of adducts of the hypoxia marker EF5. The distribution of both markers was assessed in relationship to the distribution of blood vessels, as measured by antibodies to CD31. Our hypothesis was that VEGF mRNA and hypoxia would colocalize, assuming that detectability of the former was not limiting. Four squamous cell carcinomas, three sarcomas and one glioblastoma multiforme were studied. When VEGF mRNA signal was detectable, its maxima colocalized with regional maxima of EF5 binding. The strongest levels of both signals were sometimes adjacent to regions of tissue necrosis. However, we were unable to predict absolute levels of EF5 binding based on absolute levels of VEGF mRNA. Conversely, for all tumors studied, regions with relatively low levels of EF5 binding had relatively low or undetectable VEGF mRNA. We found moderate EF5 binding in some keratinized cells but VEGF mRNA was not expressed by these differentiated cells. The paradigm that hypoxia and VEGF expression are linked in human tumors is supported by the data presented herein. A better understanding of the biology behind VEGF expression, including its modulation by hypoxia, is important for optimizing its use as a prognostic indicator and/or modulating its presence with biologic therapies.

Role of NADPH:cytochrome P450 reductase in the hypoxic accumulation and metabolism of BRU59-21, a technetium-99m-nitroimidazole for imaging tumor hypoxia

Nitroimidazoles labeled with technetium-99m are being investigated as non-invasive markers of tumor hypoxia. They are bioreductive compounds that require enzymatic reduction for retention in hypoxic cells, but little is known about the cellular factors affecting their accumulation in hypoxic cells. If the absolute accumulation of hypoxia markers is affected by enzyme levels, an inaccurate assessment of the hypoxic cell fraction in tumors may occur. BRU59-21, (99m)Tc-oxo[[3,3,9, 9-tetramethyl-6-[(2-nitro-1H-imidazol-1-yl)methyl]5-oxa-4, 8-diazadioximato]-(3-)-N,N',N",N"'] technetium (V), a technetium-99m-nitroimidazole that is being studied as a potential marker of tumor hypoxia, was used in the present study to evaluate the effect of NADPH:cytochrome P450 reductase (EC 1.6.2.4) levels on BRU59-21 accumulation and metabolism. Metabolism of BRU59-21 in hypoxic cellular lysates derived from Chinese hamster ovary cells overexpressing NADPH:cytochrome P450 reductase was 8-fold greater than in control cells. This effect required the presence of exogenous NADPH. The increased metabolism of BRU59-21 in lysates overexpressing NADPH:cytochrome P450 reductase was inhibited at 4 degrees and by the addition of NADPH:cytochrome P450 reductase inhibitors. The addition of inhibitors of other nitroreductase enzymes had no effect on BRU59-21 metabolism in these lysates. When the accumulation and metabolism of BRU59-21 were studied in stirred suspension cultures, it was found that cells overexpressing NADPH:cytochrome P450 reductase exhibited about a 3-fold increase in both the hypoxic metabolism and the accumulation of BRU59-21. These findings suggest that NADPH:cytochrome P450 reductase is an important enzyme in BRU59-21 metabolism in model systems of tumor hypoxia.

Modifying the in vitro accumulation of BMS181321, a technetium-99m-nitroimidazole, with unlabelled nitroaromatics

BMS181321, [99mTc]oxo[[3,3,9,9-tetramethyl-1-(2-nitro-1H-imidazol-1-yl)-4,8-diaz aundecane-2,10-dione dioximato]-(3)-N,N',N",N"']technetium, is a 99mTc-nitroimidazole that is being investigated as a hypoxic marker in tumors. Due to the high specific activity of 99mTc, the concentration of BMS181321 used in its applications is very low. Metabolic depletion and non-specific binding of the drug may limit its ability to fully map out hypoxic regions. An attempt has been made to modify the in vitro accumulation of BMS181321 in hypoxic Chinese hamster ovary (CHO) cells with unlabelled nitroaromatics. The 2-nitroimidazole etanidazole (0.08 to 8 mM) caused a concentration-dependent decrease in BMS181321 accumulation to 70-28% and metabolism to 70-40% of the control level in hypoxic cells at 4 hr. In contrast, the 5-nitroimidazole tinidazole (0.09 to 9 mM) caused a concentration-dependent increase in BMS181321 accumulation to 110-170% and metabolism to 100-150% of the control level in hypoxic cells at 4 hr. Nitroaromatics with an electron affinity similar to or greater than that of BMS181321 inhibited its accumulation and metabolism, and 5-nitroimidazoles, which have an electron affinity lower than that of BMS181321, enhanced its accumulation and metabolism. The enhanced accumulation with the addition of metronidazole was not observed in the presence of low oxygen levels or of a nitrofuran of higher electron affinity than BMS181321. These results suggest that a competition for reducing equivalents and/or for the BMS181321 radical anion itself can occur in cells, leading to the inhibition of BMS181321 reduction in the presence of nitroaromatics of similar or greater electron affinity. A transfer of electrons from the radical anion form of the reduced 5-nitroimidazole to the more electron affinic BMS181321 compound may occur, causing increased hypoxic accumulation of BMS181321.

Evaluation of the concept of "hypoxic fraction" as a descriptor of tumor oxygenation status

The presence and significance of tumor hypoxia has been recognized since the 1950's. Hypoxic cells in vitro and in animal tumors in vivo are documented to be three times more resistant to radiation-induced killing compared to aerobic cells. There is now evidence that tumor hypoxia is treatment-limiting in many human cancers. One common way to describe the extent of hypoxia in individual and groups of tumors is the "hypoxic fraction." This measurement infers that cells are present in only two radiobiologically significant states: oxygenated and hypoxic. In this paper, we demonstrate the qualitative and quantitative presence of hypoxic tumor cells using the oxygen dependent metabolism of the 2-nitroimidazole, EF5. Two assumptions concerning the calculation and interpretation of the hypoxic fraction are considered. The first is the use of multiple animals to describe the radiation response at a given radiation dose. We hypothesize that the presence of intertumor variability in radiation response due to hypoxia could negatively influenced the characterization of the change in slope required to calculate the hypoxic fraction. The studies presented herein demonstrate heterogeneity of radioresponse due to hypoxic fraction within and between tumor lines. The 9L subcutaneous tumor studied in air-breathing rats demonstrates a 2 log variation in surviving fraction at 17 Gy. The Morris 7777 hepatoma, in contrast, showed little variability of radiation response. Our second question addresses the limitations of using the "hypoxic fraction" to describe the radiation response of a tumor. This calculated value infers that radiobiological hypoxia is a binary measurement: that a tumor contains two cell populations, aerobic cells with maximal radiosensitivity and hypoxic cells with maximal radioresistance. The classic work of Thomlinson and Gray, however, implies the presence of an oxygen gradient from tumors vessel through the tissues. In both the 9L and Q7 tumors, flow cytometric analysis of EF5 binding demonstrates a continuous range of cellular pO2 levels. These studies suggest that: 1) there is extensive intertumor variability of radiation response in certain tumor lines; 2) the variability in radiation response between individual tumors in a group may affect the ability to describe a particular tumor type's "hypoxic fraction" and 3) The oxygen status of tumor cells is a continuum. This realization affects the ability to apply a binary concept such as the "hypoxic fraction" effectively in radiobiology.

Oxygen dependence of cellular uptake of EF5 [2-(2-nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl)a cet amide] : analysis of drug adducts by fluorescent antibodies vs bound radioactivity

The present studies were initiated to quantitate the oxygen dependence of bioreductive metabolism-induced binding of EF5, a pentafluorinated derivative of the 2-nitroimidazole, etanidazole. Two different assays were compared: first, radioactive drug incorporation into cell lysates, which provides a direct measure of drug metabolism or uptake; second, monoclonal antibody detection of cellular macromolecular adducts of EF5 after whole cell permeabilisation and fixing. The antibodies (a single clone designated ELK3-51) were conjugated with the fluorescent dye Cy3, with fluorescence determined by fluorescence microscopy and flow cytometry. For the two cell lines tested (V79 Chinese hamster fibroblasts and 9L rat glioma), the oxygen dependence of binding was found to be the same for the two techniques. Using the antibody binding technique, the fluorescence signal was highly reproducible between experiments, resistant to light or chemical bleaching and stable over time following cell or tissue staining. Flow cytometric analysis of cells from rat 9L tumours treated with EF5 in vivo or in vitro showed a distribution of fluorescent signal which was very compatible, on both a relative and absolute basis, with the in vitro results. Our results indicate that immunofluorescent techniques provide a quantitative assay for bioreductive drug adducts, and therefore may be able to measure the absolute oxygen concentration distribution in cell populations and tissues of interest.

Identification of hypoxia in cells and tissues of epigastric 9L rat glioma using EF5 [2-(2-nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide]

One of the most sensitive hypoxia detection methods is based on the observation that binding of nitroimidazoles to cellular macromolecules occurs as a result of hypoxia-dependent bioreduction by cellular nitroreductases. Nitroimidazole-binding techniques provide measurements of hypoxia to virtually any degree of spatial resolution and with a multiplicity of techniques. This paper demonstrates hypoxia imaging using in vivo EF5 binding with detection by a fluorochrome-conjugated monoclonal antibody. We investigated these techniques in the 9L glioma tumour, in part because the exact nature of the hypoxia in this tumour system is controversial. Our results demonstrate that following intravenous injection of EF5, binding and detection using a monoclonal antibody in 9L gliomas is specific and oxygen dependent. Detection of binding using fluorescence microscopy can be performed on frozen tissues; tissue sections can be counterstained with haematoxylin and eosin for light microscopic analysis. Alternatively, the distribution of hypoxia in a tumour can be inferred by examining individual tumour cells using flow cytometric techniques. Based upon the results presented herein, the radiation-resistant phenotype of 9L epigastric tumours grown in our laboratories can be associated with the presence of hypoxic cells.

The novel fluorinated 2-nitroimidazole hypoxia probe SR-4554: reductive metabolism and semiquantitative localisation in human ovarian cancer multicellular spheroids as measured by electron energy loss spectroscopic analysis

The novel fluorinated 2-nitroimidazole SR-4554 is undergoing preclinical development as a magnetic resonance spectroscopy and imaging probe for hypoxic tumour cells. We have used electron energy loss spectroscopic analysis (EELS) to show selective reduction and differential subcellular localisation of SR-4554 in human ovarian multicellular spheroids. SR-4554 was demonstrated to be metabolised by these A2780 cells under hypoxic but not under normal aerobic cell culture conditions. The EELS technique illustrated that the relative amount of drug within the cytoplasm of cells from both the inner region (150-160 microns from edge) and outer edge of the spheroid did not differ significantly after an initial 3 h incubation with drug. In contrast, an 8-fold differential between the amount of drug retained in the cytoplasm (primarily ribosomes and endoplasmic reticulum) of cells from the inner vs outer regions of the spheroids was observed following a subsequent 2 h 'chase' culture in drug-free medium. Within cells from the hypoxic region of the spheroid, SR-4554 was mainly associated with the endoplasmic reticulum, nucleus and the cytoplasmic side of intracellular vesicles and also to a lesser extent with the nuclear periphery. Interestingly, the drug was only weakly associated with the mitochondria and plasma membrane of the cells. The characteristics of cellular and subcellular distribution of SR-4554 are consistent with the hypothesis that 2-nitroimidazole compounds undergo hypoxia-mediated enzymatic reduction to reactive species. These reactive species are selectively retained in the cells in which they are metabolised through covalent association with subcellular components. These findings provide additional support for the clinical development of the drug as a non-invasive probe for tumour hypoxia and at the same time illustrate the utility of the EELS technique for examining the heterogeneity of drug distribution both between and within cells.

Enhanced radiation-sensitivity by preincubation with nitroimidazoles: effect of glutathione depletion

PURPOSE

The mechanism of enhanced radiosensitization by nitroheterocyclics after a preincubation period under hypoxic conditions was investigated. The hypothesis that this phenomenon was caused by glutathione depletion was tested.

METHODS AND MATERIALS

The phenomenon of enhanced radiosensitization by nitroheterocyclics after a preincubation period under hypoxic conditions is potentially of importance therapeutically because essentially nonlethal preradiation exposures to the electron affinic drugs cause a much larger radiation sensitization than would otherwise be expected. We have investigated this interesting property of several 2-nitroimidazoles to determine its possible cause and to test various hypotheses about maximizing its possible therapeutic benefit. In view of many observations that thiols are depleted by incubation of cells with nitroimidazoles under hypoxic conditions, we have specifically investigated this aspect of the preincubation effect. Depletion of glutathione was either enhanced by an overnight incubation with buthionine sulfoximine or minimized by preincubation with a 2-nitroimidazole which is sterically inhibited from causing thiol depletion.

RESULTS

When conditions were chosen which minimized variations in cellular glutathione content during the preincubation period, no preincubation effect was observed. At low, therapeutically relevant radiation doses, where 2-nitroimidazoles are less efficient sensitizers, the preincubation effect may be even more important, but thiol depletion still minimizes its impact in this region of the dose-response curve.

CONCLUSION

These results suggest that the preincubation effect is caused by a "self-sensitization" involving the known enhancement of radiation sensitization by thiol depletion.

Bioreductive metabolism of SR-4233 (WIN 59075) by whole cell suspensions under aerobic and hypoxic conditions: role of the pentose cycle and implications for the mechanism of cytotoxicity observed in air

PURPOSE

Measurement of pentose cycle (PC) activity is shown to be a noninvasive means for monitoring the reduction of SR-4233 in whole cells. Comparing these measurements to the actual measurements of drug loss under aerobic and hypoxic conditions helps to define the mechanism for the associated aerobic toxicity.

METHODS AND MATERIALS

SR-4233 is activated to a toxic species by bioreductive metabolism. NADPH is required for the activation of the drug by purified enzymes, cell homogenates and whole cells. In vivo the NADPH:NADP+ ratio is maintained by the oxidation of glucose via the oxidative limb of the pentose cycle. By measuring radiolabeled 14CO2 released as a product of this oxidation one can get an accurate measurement of the rate of drug metabolism in whole cells. These results are compared to measurements of drug consumption under aerobic and hypoxic conditions using an HPLC assay.

RESULTS

SR-4233 stimulates pentose cycle activity to a greater extent in air then under hypoxia, however, in the presence of added catalase, pentose cycle activity is stimulated to a similar extent under both conditions. The higher levels of PC activity observed in air are due to the production of hydrogen peroxide by the nitroxide free radical undergoing futile redox cycling. The contribution of H2O2 to the observed aerobic cytotoxicity of SR-4233 is minimal however, since toxicity is only slightly reduced in the presence of exogenous catalase and antioxidants such as vitamin E. The level of PC stimulation by SR-4233 suggests that the rate of electron addition to the drug is independent of O2 concentration. The loss of drug from the incubation medium, i.e., conversion to a stable intermediate species, occurs approximately five times faster under nitrogen than in air for A549 cells. It is the rate of drug loss from the cell and not the rate of reduction which best correlates with the observed aerobic and hypoxic toxicity.

CONCLUSION

Toxicity in air and in nitrogen is directly related to the rate of drug reduction, i.e., at equivalent levels of drug loss we observe equal levels of cytotoxicity.