Cryoballoon ablation of atrial fibrillation is effectively feasible without previous imaging of pulmonary vein anatomy: insights from the 1STOP project.

BACKGROUND: Pulmonary vein isolation by cryoablation (PVI-C) is a standard therapy for the treatment of atrial fibrillation (AF); however, PVI-C can become a challenging procedure due to the anatomy of the left atrium and pulmonary veins (PVs). Importantly, the utility of imaging before the procedure is still unknown regarding the long-term clinical outcomes following PVI-C. The aim of the analysis is to evaluate the impact of imaging before PVI-C on procedural data and AF recurrence. METHODS: Patients with paroxysmal AF underwent an index PVI-C. Data were collected prospectively in the framework of 1STOP ClinicalService® project. Patients were divided into two groups according to the utilization of pre-procedural imaging of PV anatomy (via CT or MRI) or the non-usage of pre-procedural imaging. RESULTS: Out of 912 patients, 461 (50.5%) were evaluated with CT or MRI before the PVI-C and denoted as the imaging group. Accordingly, 451 (49.5%) patients had no pre-procedural imaging and were categorized as the no imaging group. Patient baseline characteristics were comparable between the two cohorts, but the ablation centers that comprised the imaging group had fewer PVI-C cases per year than the no imaging group (p < 0.001). The procedure, fluoroscopy, and left atrial dwell times were significantly shorter in the no imaging cohort (p < 0.001). The rates of complications were significantly greater in the imaging group compared to the no imaging group (6.9% vs. 2.7%; p = 0.003); this difference was attributed to differences in transient diaphragmatic paralysis. The 12-month freedom from AF was 76.2% in the imaging group and 80.0% in the no imaging group (p = 0.390). CONCLUSIONS: In our analysis, PVI-C was effective regardless of the availability of imaging data on PV anatomy.

Splenic angiosarcoma and iron deficiency anemia in a 43-year-old man.

Thrombocytopenia and microcytic anemia are two laboratory findings that alone or together suggest an underlying disease process. Both are found throughout particular age groups and have broad differential diagnoses. Angiosarcomas are rare neoplasms from the lining of blood vessels. Primary splenic angiosarcoma is an even rarer neoplasm, first reported in the late 1870s. We report a case of primary splenic angiosarcoma in a patient with thrombocytopenia, microcytic anemia, and splenomegaly.

Aminobenzoic acid compounds as HOCl traps for activated neutrophils.

This study was designed to develop traps for hypochlorous acid (HOCl) which could be used to detect HOCl in the microenvironment of activated neutrophils. Reagent HOCl was found to react with para-aminobenzoic acid (PABA) in aqueous solution to produce a predominant metabolite detectable by high performance liquid chromatography (HPLC). Mass spectroscopy and nuclear magnetic resonance identified this metabolite as the ring addition product 3-chloro PABA. The related compound para-aminosalicylic acid (PAS) was also metabolized by HOCl to 3-chloro PAS. The formation of the 3-chloro metabolite was specific for reactions involving HOCl, since several other oxidants in chloride buffer failed to produce the metabolite. Human blood neutrophils activated by phorbol myristate acetate or zymosan in the presence of PABA (or PAS) used their HOCl to produce large amounts of the 3-chloro metabolite. The formation of 3-chloro PABA was inhibited by azide, catalase, and taurine, which is consistent with the production of the metabolite by the neutrophil myeloperoxidase (MPO) pathway. The reaction of HOCl with PABA and PAS was relatively slow as shown by competitive reactions with endogenous antioxidants like taurine, methionine, and glutathione. This was confirmed in reactions involving PABA/PAS and reagent HOCl or HOCl generated by the MPO enzyme system. In these in vitro systems, glutathione and serum completely inhibited the formation of the 3-chloro metabolite. In contrast, activated neutrophils metabolized PABA/PAS to the 3-chloro metabolite even in the presence of 1% serum. These findings demonstrate that PABA and PAS are specific trapping agents for HOCl produced by neutrophils in complex biological conditions.

Metabolism of phenytoin and covalent binding of reactive intermediates in activated human neutrophils.

ontivation of neutrophils by phorbol-12-myristate-13-acetate (PMA) causes rapid production of superoxide radical (O2-), leading to the formation of additional reactive oxygen species, including hydrogen peroxide (H2O2), hypochlorous acid (HOCl), and possibly hydroxyl radical (.OH). These reactive oxygen species have been associated with the oxidation of some drugs. We investigated the metabolism of phenytoin (5,5-diphenylhydantoin) and the covalent binding of reactive intermediates to cellular macromolecules in activated neutrophils. In incubations with 100 microM phenytoin, PMA-stimulated neutrophils from six human subjects produced p-, m-, and o-isomers of 5-(hydroxyphenyl)-5-phenylhydantoin (HPPH) in a ratio of 1.0:2.1:2.8, respectively, as well as unidentified polar products. Analysis of cell pellets demonstrated that phenytoin was bioactivated to reactive intermediates that bound irreversibly to macromolecules in neutrophils. Glutathione, catalase, superoxide dismutase, azide, and indomethacin all diminished the metabolism of phenytoin and the covalent binding of its reactive intermediates. The iron-inactivating chelators desferrioxamine and diethylenetriaminepentaacetic acid had little or no effect on the metabolism of phenytoin by neutrophils, demonstrating that adventitious iron was not contributing via Fenton chemistry. In an .OH-generating system containing H2O2 and Fe2+ chelated with ADP, phenytoin was oxidized rapidly to unidentified polar products and to p-, m-, and o-HPPH (ratio 1.0:1.7:1.5, respectively). Reagent HOCl and human myeloperoxidase (MPO), in the presence of Cl- and H2O2, both formed the reactive dichlorophenytoin but no HPPH. However, no chlorinated phenytoin was detected in activated neutrophils, possibly because of its high reactivity. These findings, which demonstrated that activated neutrophils biotransform phenytoin in vitro to hydroxylated products and reactive intermediates that bind irreversibly to tissue macromolecules, are consistent with phenytoin hydroxylation by .OH generated by a transition metal-independent process, chlorination by HOCl generated by MPO, and possibly cooxidation by neutrophil hydroperoxidases. Neutrophils activated in vivo may similarly convert phenytoin to reactive intermediates, which could contribute to some of the previously unexplained adverse effects of the drug.

Biologic effects of the adoptive transfer of cells depleted of monocytes with L-phenylalanine methyl ester.

Monocytes macrophages have negative regulatory effects on many immunologic responses. Depletion of monocytes from peripheral blood using the lysosomotropic agent, L-phenylalanine methyl ester (PME), has been shown to improve lymphokine activated killer (LAK) cell expansion in vitro. A pilot study of the adoptive transfer of LAK cells expanded with PME was performed in patients with metastatic renal cell carcinoma. Patients received interleukin-2 (IL-2) by continuous infusion for 5 days. Leukopheresis was performed daily for 4 days during the second week. Cells obtained from 8 patients were depleted of monocytes using PME in an one-step procedure; < or = 3% of the remaining cells were monocytes. All cells were expanded for 10 days in air-porous plastic bags with IL-2. Cells expanded 2.7-fold when depleted with PME and 1.7-fold when not depleted (P = 0.02). Expanded cells were administered together with IL-2. Patients received up to 60 x 10(10) PME-depleted cells (mean = 26 x 10(10)) with LAK activity (% lysis) of 60 +/- 12%. Lymphocyte phenotype and cytolytic activity were not modulated by PME-depletion, and clinical toxicities and systemic immunologic effects observed in patients receiving PME-depleted cells were similar to that of 5 patients receiving cells not expanded with PME. Thus, the use of PME to deplete monocytes ex vivo can result in the yield of large number of effectors that retain immunologic activity for potential clinical use. The process is convenient, efficient, and does not add clinical toxicity.

Biotransformation of para-aminobenzoic acid and salicylic acid by PMN.

Para-aminobenzoic acid (PABA) is an essential cofactor for the production of folic acid in bacteria and has mild anti-inflammatory activity. We have recently reported that salicylic acid and benzoic acid are oxidized by stimulated granulocytes Polymorphonuclear Neutrophils (PMN). The oxidation of salicylate appears mediated by a potent oxygen metabolite generated during the respiratory burst which is dependent primarily on superoxide (O2-) for its production. These background studies with the salicylate group of drugs suggested that PABA might be similarly metabolized by PMN. In these studies, we demonstrate that PABA is metabolized by stimulated PMN. However, in contrast to the biochemical mechanism involved in the metabolism of salicylate, our scavenger studies indicate that PABA is metabolized primarily by the myeloperoxidase pathway. Our results may explain the mild anti-inflammatory actions of the drug and suggest that the degradation of PABA by PMN at an inflammatory site may limit the availability of PABA for bacterial growth.

Glucose metabolism of hairy cells.

Hairy cell leukemia is a malignant B-cell disorder characterized by splenomegaly and pancytopenia. The malignant cell is morphologically unique and characterized by fine cytoplasmic projections. Although studies of the cell have revealed important information about its proliferative capacity, cell surface, and membrane composition, less is known about the metabolic characteristics of the cell. We have previously investigated the oxidative metabolism of the hairy cell and have suggested that hairy cells might have a unique glucose metabolism compared to normal lymphocytes. This is indicated by a high rate of [6-14C]glucose oxidation in short-term culture consistent with an active Kreb's cycle and a high ratio of [6-14C]glucose oxidation to [1-14C] glucose oxidation. In this study, we evaluated an additional group of patients with hairy cell leukemia prior to or after treatment with the experimental drug 2'-deoxycoformycin (dCF). We found that in seven of eight patients the leukemic cells had a pattern similar to that previously described and that all of these seven patients had a significant response to therapy. The cells of the eighth patient had minimal Kreb's cycle activity, and at the time of study the patient was resistant to therapy with dCF. The metabolic activity of hairy cells may distinguish them from other lymphoid populations and may be a marker for sensitivity to dCF.

Interleukin-2 therapy enhances salicylate oxidation by blood granulocytes.

These studies determined the effect of interleukin-2 (IL-2) immunotherapy on the oxidative metabolism of the blood granulocytes of eight patients with metastatic renal cancer. We quantitated the rate of the hexose monophosphate shunt activity (HMPS), hydrogen peroxide (H2O2) production, and salicylate oxidation of the unstimulated and phorbol myristate acetate (PMA)-stimulated granulocyte cultures before, during, and after a 5-day continuous infusion of IL-2. There was no change in the rate of HMPS activity. However, the rate of salicylate oxidation of the unstimulated and PMA-stimulated cultures of these patients was significantly increased after the therapy was complete. Overall, there was no increase in the rate of H2O2 production, although the PMA-stimulated cultures of three of eight patients had a twofold higher production of H2O2 after treatment compared with the pretreatment values. The enhanced rate of salicylate oxidation by the granulocytes after treatment indicates that these cells were "stimulated" in vivo to produce a potent oxidant, which is most likely hydroxyl radical or an oxidant of comparable activity. Further, the granulocytes were primed ("activated"), since they had an augmented response to PMA. IL-2 did not stimulate the oxidative metabolism of granulocyte cultures in vitro, suggesting that the IL-2 effect in vivo is not a direct one. Our results indicate that IL-2 immunotherapy is associated with the activation of blood granulocyte oxidative metabolism and that these activated granulocytes may be related to some of the toxic side effects of IL-2 therapy such as the capillary leak syndrome. Further oxidant injury to the granulocytes may explain the reported defect in chemotaxis.

Tumor necrosis factor primes neutrophils for hypochlorous acid production.

Tumor necrosis factor (TNF) has a weak direct effect on neutrophil oxidative metabolism and primes neutrophils for oxidant release in response to other stimuli. We examined the effect of recombinant human TNF alpha (rTNF alpha) on production of hypochlorous acid (HOCl) by human neutrophils. TNF alone, even at concentrations of 1,000 U/ml, did not stimulate HOCl production. In contrast, rTNF alpha, in a dose-dependent manner, primed neutrophils for HOCl production in response to the weak agent unopsonized zymosan. rTNF alpha concentrations as low as 10 U/ml resulted in a fivefold increase in HOCl in this system. rTNF alpha-primed cells also exhibited increased phagocytosis. Priming in this model system occurred regardless of whether cells were preincubated with rTNF alpha before addition of zymosan or coincubated with both rTNF alpha and zymosan. rTNF alpha priming for HOCl production could not be washed away and required a lag period of approximately 10 min. rTNF alpha priming was not dependent on extracellular Ca2+ and Mg2+. Preincubation experiments demonstrated that rTNF alpha priming was not inhibited by the microfilament blocker cytochalasin B. Although the mechanism remains unclear, these findings demonstrate that rTNF alpha has an important priming effect on the neutrophil myeloperoxidase pathway.

Hydroxylation of salicylate by activated neutrophils.

Salicylates are metabolized in vivo to hydroxylated compounds, including 2,3-dihydroxybenzoic acid and 2,5-dihydroxybenzoic acid (gentisic acid). The present study hypothesized that activated neutrophils represent one pathway for salicylate hydroxylation. Human neutrophils were incubated in medium containing 10 mM salicylate and stimulated with phorbol myristate acetate (PMA) for 1 hr. The cell-free supernatant fractions were analyzed by HPLC. Neutrophils (1 x 10(6) cells) produced 55 +/- 11 ng of gentisic acid. Neutrophils also produced smaller quantities of 2,3-dihydroxybenzoic acid. Antioxidant inhibitor experiments indicated that superoxide dismutase (SOD), heme protein inhibitors, and glutathione blocked gentisic acid formation, whereas catalase, mannitol, and deferoxamine failed to inhibit. Experiments with the reagent hypochlorous acid (HOCl) and the model myeloperoxidase (MPO) enzyme system did not support a role for the MPO pathway in gentisic acid formation. These findings demonstrate that activated neutrophils can hydroxylate salicylate by an unknown pathway. This pathway may contribute to the increased recovery of hydroxylated salicylates in patients with inflammatory disorders.

Effect of dimethylthiourea on the neutrophil myeloperoxidase pathway.

The sulfur-centered compound dimethylthiourea (DMTU) affords antioxidant protection in animal models of acute lung injury, an effect that has been attributed to its OH. scavenging properties. Although DMTU can also react with H2O2 in certain experimental systems, the effect of DMTU on the neutrophil myeloperoxidase (MPO) pathway has not been studied. DMTU (1-10 mM) completely blocked stable oxidants and hypochlorous acid formation by phorbol myristate acetate- and zymosan-stimulated neutrophils. DMTU also provided complete inhibition when incubated with cell-free supernatants after the formation of the MPO products. DMTU prevented the oxidative inactivation of alpha 1-antitrypsin by neutrophil-stable oxidants. Evidence that DMTU was oxidized by the MPO products was obtained by titration of oxidized DMTU with reduced glutathione. Surprisingly, supernatants from cells incubated with DMTU (10 mM) consumed two- to threefold higher amounts of reduced glutathione than supernatants from cells incubated with taurine (15 mM). Metabolic studies with stimulated neutrophils and experiments with the MPO enzyme system in a cell-free system suggested that DMTU acts by scavenging the products of the MPO pathway rather than by blocking H2O2 production in the intact cell. These findings demonstrate that DMTU blocks the neutrophil MPO pathway in addition to its known ability to scavenge other reactive O2 species. The capacity of DMTU to scavenge MPO products may explain some of its protective effects in acute lung injury.

The effects of the anti-tumor agent mezerein on the cytotoxic capacity and oxidative metabolism of human blood cells.

Mezerein, the most active antitumor compound isolated from the daphne species of plants, has a structural similarity to phorbol myristate acetate (PMA), the major active compound isolated from croton oil. PMA is known to have tumor promoting activity and is a potent inflammatory agent. Mezerein has similarly been reported to have potent inflammatory properties but appears to be a weaker tumor promoter than PMA. While the effect of PMA on the function and metabolism of human blood cells has been extensively studied, there is little similar information concerning mezerein. Therefore, in these studies, we have compared the capacities of mezerein and PMA to activate the cytotoxic capacity and oxidative metabolism of human granulocyte (PMNs), monocyte, lymphocyte, and mononuclear cell (lymphocytes and monocytes) cultures in vitro. Mezerein stimulated the oxidative metabolism of PMNs in an identical manner to PMA as indicated by a burst in the activity of the HMPS pathway, the production of H2O2, hydroxyl radical and stable oxidants. Mezerein also stimulated the release of thromboxane B2 from PMNs. Both compounds activated the oxidative metabolism of monocytes but not the oxidative metabolism of lymphocytes. The enhanced oxidative metabolism of the phagocytic cells was associated with an increased cytotoxicity against human red cells which are sensitive to oxidant damage but not against the NK resistant Raji lymphoblast cell line or the SW1116 colon tumor cell line. Of interest is that mezerein did not augment significantly the minimal cytotoxic capacity (NK activity) of mononuclear cells, monocytes or freshly isolated lymphocyte cultures against the tumor cell targets used in our experiments. However, lymphocyte cultures preincubated for 15 hours with mezerein had a marked enhancement of cytotoxicity against the tumor targets. This activation was not observed in similarly treated mononuclear cell cultures suggesting a suppressor activity of the monocytes. Our data suggest that the potent inflammatory activity of mezerein similar to PMA, may be related to its capacity to activate the oxidative and arachidonic metabolism of phagocytic cells. In addition, the capacity of mezerein to activate the cytotoxic capacity of lymphocytes may relate to its reported in vivo antitumor activity.

Activation and deactivation of guinea pig peritoneal eosinophils during chronic polymyxin B stimulation.

Eosinophils exhibit different levels of oxidative metabolism depending on their site of origin and various host factors that may influence metabolism. The present study examined the time course of eosinophil oxidative metabolism in animals undergoing chronic peritoneal stimulation. Eosinophils were purified from the peritoneal exudates of guinea pigs stimulated with weekly polymyxin B and saline peritoneal lavage. 14C-1- and 14C-6-glucose oxidation and H2O2 production were measured at week 0 and at various time points throughout 43 weeks of stimulation. Baseline oxidative metabolism of eosinophils was relatively high throughout the time course, but then declined sharply after 32 weeks. These "deactivated" cells that were recovered after 32 weeks also failed to respond to phorbol myristate acetate (PMA) or opsonized zymosan. Electron microscopy did not reveal significant differences between deactivated eosinophils and cells from earlier time points. These findings document the time course of eosinophil activation and deactivation in this model and suggest that metabolic heterogeneity of eosinophils can occur over time in response to a chronic stimulus.

Effect of O2 partial pressure on the myeloperoxidase pathway of neutrophils.

Neutrophils recruited to different tissues undergo respiratory burst activity at widely different PO2 levels. The present study investigated the in vitro effects of PO2 on neutrophil oxidative metabolism. When neutrophils were stimulated with either zymosan or phorbol myristate acetate (PMA) under different PO2's (0-700 Torr), hexose monophosphate shunt activity, H2O2, and hydroxyl radical (OH.) production were directly related to the level of PO2. Neutrophils functioned surprisingly well at PO2's as low as 10 Torr, where metabolic burst activity was prolonged and usually exceeded 50% of maximal values. The production of neutrophil stable oxidants and hypochlorous acid (HOCl) by zymosan-stimulated neutrophils was also directly related to PO2. In contrast, the production of stable oxidants and HOCl by PMA-stimulated neutrophils was significantly higher at 10 Torr compared with 700 Torr. The decrease in stable oxidant production by PMA-stimulated neutrophils at elevated PO2's was explained by both increased destruction of stable oxidant products and by decreased availability of the precursor HOCl. Superoxide dismutase and the OH. scavenger benzoate partially prevented the fall in stable oxidants at elevated PO2's. Measurements of stable oxidants in PMA-stimulated supernates generated at 10 and 700 Torr correlated with the ability of these supernates to decrease the elastase inhibitory capacity of the serum antiprotease alpha 1-antitrypsin. These findings suggest that different PO2's alter the magnitude and pattern of neutrophil oxidative metabolism.

Oxidation of salicylates by stimulated granulocytes: evidence that these drugs act as free radical scavengers in biological systems.

Although salicylates have been used for centuries as treatment of inflammatory diseases, the mechanism of action of these drugs is still not clear. Aspirin (acetylsalicylic acid) and other nonsteroidal anti-inflammatory drugs (NSAID) inhibit prostaglandin biosynthesis, a property that appears to explain part of their anti-inflammatory activity. However, this mechanism does not appear to explain the anti-inflammatory properties of salicylic acid, which is a major metabolite of ASA in vivo. Results of prior studies in our laboratory have established that benzoic acid, the parent compound of the salicylate group of drugs, is decarboxylated and hydroxylated by the hydroxyl free radical (OH.) produced by stimulated granulocytes. These observations suggested that salicylates might be similarly metabolized by granulocytes. If so, the capacity of salicylates to rapidly react with OH. might relate directly to their known anti-inflammatory properties. Preliminary experiments established that salicylic acid and aspirin were decarboxylated by the hydroxyl free radical generated by the enzyme system xanthine-xanthine oxidase. We then studied the metabolism of salicylates by human granulocytes. Unstimulated granulocyte suspensions did not oxidize ASA or salicylic acid. However, suspensions stimulated by opsonized zymosan particles rapidly oxidized both substrates in pharmacological concentrations. The rate of oxidation of salicylic acid was 16-fold higher than benzoic acid, whereas the rate of oxidation of ASA was four-fold higher. The reaction was oxygen dependent and could be inhibited by known hydroxyl scavengers, particularly dimethylthiourea. The reaction could also be inhibited by superoxide dismutase and azide, indicating that O-2 and heme or an iron-dependent enzyme were required for the reaction. The reaction was not impaired by compounds known to react with the HOCL and the chloramines generated by stimulated PMN. Furthermore, salicylic acid in high concentrations did not impair the HMPS pathway, the production of O-2 or the production of H2O2 by granulocytes. These data provide evidence that salicylates are rapidly oxidized by the hydroxyl free radical produced by granulocytes and not O-2, H2O2, or HOCL. This capacity of salicylates to react rapidly and selectively react with OH. may directly relate to their anti-inflammatory properties. In addition, results of our experiments indicate that stimulated granulocytes acquire the capacity to metabolize these drugs. Therefore, several metabolites of salicylates may be produced at a site of inflammation, all of which may have altered biological activity compared with the parent compound.

Effect of diethyldithiocarbamate (DDC) on human monocyte function and metabolism.

Recent evidence indicates that phagocytic cells play a major role in tissue inflammation. The release of enzymes, lipid metabolites such as prostaglandins, and reactive oxygen species by these cells appear to mediate the inflammatory process. In this study we have evaluated the effects of diethyldithiocarbamate (DDC) on human monocyte function and metabolism. We demonstrate that DDC impairs that antibody-dependent cytoxicity (ADCC) of monocytes to red cell targets. The concentration of DDC which caused maximal suppression of ADCC also prevented the burst of oxidative metabolism in monocytes stimulated by sensitized red cells targets or phorbol myristate acetate (PMA). DDC also impairs the lipid metabolism of these cells as indicated by a decrement in malonyldialdehyde (MDA) production. These data indicate that DDC impairs the activity of two major biochemical pathways in monocytes which are related to the inflammatory process, i.e., the release of oxygen metabolites and prostaglandins.

Effect of hyperoxia on the carbohydrate metabolism of human lymphocytes.

Lymphocytes incubated under high oxygen tensions have impaired lymphoblastic transformation to nonspecific mitogens. Since carbohydrate metabolism may be essential for an optimal proliferation of these cells following an antigenic stimulus, we have characterized the effect of hyperoxia on the glucose metabolism of human lymphocytes in vitro. The mean rates of glucose utilization, Krebs cycle activity, and hexose monophosphate shunt activity were determined for unstimulated and phytohemagglutinin (PHA)-stimulated cultures incubated for 3 days. In unstimulated cultures, hyperoxia stimulated glucose utilization and markedly impaired Krebs cycle activity but did not alter HMPS activity. These observations suggest that hyperoxia impairs the mitochondrial activity of lymphocytes and that increased glycolysis compensates for the loss of cellular energy normally produced by this organelle. Under high oxygen tension PHA-stimulated cultures failed to show the burst in HMPS activity and increased Krebs cycle activity characteristic of cultures incubated under physiological oxygen tensions and consumed less glucose. Our results indicate that hyperoxia alters the glucose metabolism of lymphocytes. This metabolic alteration may be related to the impaired function of the cells under hyperoxic conditions.

The effects of the quinone type drugs on hydroxyl radical (OH.) production by rat liver microsomes.

The quinone drugs are known to be metabolized to semiquinone free-radical intermediates and to enhance NADPH oxidation in microsomal system. The effect of adriamycin and mitomycin C on the decarboxylation of [14C] carboxyl benzoate via hydroxyl radical (OH.) production in the microsomal system was examined. The activity of these drugs was compared to 5-fluorouracil, cyclophosphamide, and methotrexate, which are inactive in oxygen consumption experiments and are non-quinone-type drugs. Adriamycin and mitomycin C stimulated decarboxylation of benzoate 100 and 50% above the controls, respectively, while 5-fluorouracil, cyclophosphamide, and methotrexate were not different from controls. Addition of superoxide dismutase increased benzoate decarboxylation with or without the drugs present, while catalase was inhibitory in both circumstances. These results suggest that the quinone drugs enhanced hydroxyl radical (OH.) production by liver microsomes, and offer a possible mechanism of cellular toxicity by these agents.