Free radical increases in cancer: evidence that there is not a real increase.

The controversial finding of an increase in free radicals with the development of cancer was restudied with Walker 256 carcinosarcoma cells. It was confirmed that such an increase appears to occur, but it was also demonstrated that it is not a real increase. With growth of the tumor, the electron spin resonance lines for lyophilized samples became narrower, resulting in an increase in peak-to-peak height measurements, but there was no change in the total number of spins. The signals for lyophilized tumor samples were different from those for the same samples before lyophilization. Changes in line shape indicated that lyophilization. Changes in line shape indicated that lyophilized tumor samples contain a different mixture of radicals than lyophilized samples of normal tissue.

Thioredoxin reductase mediates cell death effects of the combination of beta interferon and retinoic acid.

Interferons (IFNs) and retinoids are potent biological response modifiers. By using JAK-STAT pathways, IFNs regulate the expression of genes involved in antiviral, antitumor, and immunomodulatory actions. Retinoids exert their cell growth-regulatory effects via nuclear receptors, which also function as transcription factors. Although these ligands act through distinct mechanisms, several studies have shown that the combination of IFNs and retinoids synergistically inhibits cell growth. We have previously reported that IFN-beta-all-trans-retinoic acid (RA) combination is a more potent growth suppressor of human tumor xenografts in vivo than either agent alone. Furthermore, the IFN-RA combination causes cell death in several tumor cell lines in vitro. However, the molecular basis for these growth-suppressive actions is unknown. It has been suggested that certain gene products, which mediate the antiviral actions of IFNs, are also responsible for the antitumor actions of the IFN-RA combination. However, we did not find a correlation between their activities and cell death. Therefore, we have used an antisense knockout approach to directly identify the gene products that mediate cell death and have isolated several genes associated with retinoid-IFN-induced mortality (GRIM). In this investigation, we characterized one of the GRIM cDNAs, GRIM-12. Sequence analysis suggests that the GRIM-12 product is identical to human thioredoxin reductase (TR). TR is posttranscriptionally induced by the IFN-RA combination in human breast carcinoma cells. Overexpression of GRIM-12 causes a small amount of cell death and further enhances the susceptibility of cells to IFN-RA-induced death. Dominant negative inhibitors directed against TR inhibit its cell death-inducing functions. Interference with TR enzymatic activity led to growth promotion in the presence of the IFN-RA combination. Thus, these studies identify a novel function for TR in cell growth regulation.

Reductive activation of diaziquone and possible involvement of free radicals and the hydroquinone dianion.

To analyze the mode of action of diaziquone [AZQ] on DNA, we examined the activity of two AZQ analogues and N,N',N"-triethylenethiophosphoramide on three forms [supercoiled (Form I), open circular (Form II), and linear (Form III)] of PM-2 DNA. The AZQ analogues contained chlorine atoms which substituted either the carbethoxyamino groups or the aziridine groups of the parent compound. N,N',N"-triethylenethiophosphoramide is a triaziridine compound containing pentavalent phosphorus instead of a quinone group. We found that only when reduced with sodium borohydride did AZQ change the topology of the three forms of PM-2 DNA by introducing mainly single strand breaks. The AZQ analogue containing only aziridines (RQ2) was active in both its oxidized and its reduced forms, while the analogue containing only the carbethoxyamino groups (RQ14) or N,N',N"-triethylenethiophosphoramide were not active in either form. Under similar experimental conditions, Adriamycin alone altered the electrophoretic mobility of PM-2 DNA, while borohydride reduced Adriamycin did not. By using electron spin resonance spectroscopy, we showed that dihydrodiaziquone (AZQH2) oxidizes to the semiquinone in the presence of oxygen. Although AZQH2 was active against DNA, it was not active against cellular DNA synthesis as measured by [3H]thymidine incorporation into exponentially growing HEp-2 cells. However, AZQ alone prevented [3H]thymidine incorporation into HEp-2 cells. We found that HEp-2 cells have the ability to reduce AZQ to its free radical anion, but AZQH2 does not autoxidize to the free radical in the presence of cells. The reductive ability of HEp-2 cells may be responsible in part for preventing the oxidation of AZQH2 to the free radical. We found that under our conditions (1-h incubations) the aziridines are essential for the activity of aziridinyl quinones against PM-2 DNA and that in the case of AZQ the hydroquinone is also required.

Effects of pH and temperature on the stability and decomposition of N,N'N"-triethylenethiophosphoramide in urine and buffer.

N,N',N"-Triethylenethiophosphoramide (thiotepa) was dissolved at 100 micrograms/ml in urine or in 0.1 M sodium acetate buffer and incubated at 37 degrees or 22 degrees. After 0, 15, 30, 60, 90, and 120 min of incubation, 0.1-ml samples were extracted into ethyl acetate and analyzed by gas-liquid chromatography (1.8-m X 2-mm column packed with 3% OV225 on 100/120 Supelcoport; oven at 180 degrees; injection port and nitrogen-phosphorus detector at 230 degrees). Thiotepa was more stable at 22 degrees than at 37 degrees and at pH 6 to 7 than at pH 4 to 5.5. After 2 hr of incubation at 37 degrees, thiotepa concentrations decreased by 40% at pH 5.0 but only 10% at pH 6 or 7. Although thiotepa concentrations declined as described above, alkylating activity, as assessed by p-nitrobenzyl pyridine reactivity, was stable at all temperatures and pHs tested. Partition coefficients of thiotepa degradation products into toluene, ethyl acetate, diethyl ether, and hexane were determined after 0 and 120 min of incubation in urine at pH 4.0. The extractability of alkylating activity into these organic solvents decreased dramatically after 120 min. Thiotepa degradation products were extracted from urine at pH 4.0 after 0, 30, 60, and 120 min incubation at 37 degrees and were separated by thin-layer chromatography. In addition to thiotepa (Rf 0.15), 3 degradation products possessing p-nitrobenzyl pyridine alkylating activity (Rf 0.35, 0.52, and 0.60) were observed during the course of incubation. The structures of the materials with Rf 0.35 and 0.52 were identified by mass spectrometry and indicated that thiotepa degradation occurs by successive addition of HCl molecules with opening of the aziridine rings and conversion to 2-chloroethyl moieties.

Cellular pharmacology in murine and human leukemic cell lines of diaziquone (NSC 182986).

We investigated the in vitro interaction with and antitumor effect on several murine and human leukemic cell lines of diaziquone (AZQ). L1210 cells accumulated AZQ from Roswell Park Memorial Institute Medium 1640 with or without newborn calf serum by a temperature-dependent and sodium azide-resistant process. AZQ inhibited, in a dose-dependent fashion, [3H]thymidine incorporation into L1210 cells, but this inhibition was slow to develop, requiring approximately 6 hr to become apparent. The minimal inhibitory concentration of AZQ for this process was 0.05 to 0.25 nmol/ml. AZQ was a much less effective inhibitor of L1210 cell [3H]uridine and [14C]valine incorporation. In suspension cultures, AZQ inhibited growth of L1210 and HL-60 cells at minimal inhibitory concentrations of 0.5 to 1 nmol/ml. In soft agar cultures, AZQ inhibited HL-60 cell cloning at minimal inhibitory concentrations of 0.1 to 0.3 nmol/ml. AZQ provoked a dose-dependent increase in oxygen consumption when added to intact L1210, HL-60, and K562 cells and was converted to an AZQ anion free radical by these cells. When the aziridine rings of AZQ were opened by acid treatment, the resulting molecule was not accumulated by L1210 cells, did not provoke O2 consumption, did not form free radicals when added to L1210 cells, and was a much less effective inhibitor of [3H]thymidine incorporation by L1210 cells than was AZQ.

Free radicals in quinone containing antitumor agents. The nature of the diaziquone (3,6,-diaziridinyl-2,5-bis(carboethoxyamino)-1,4-benzoquinone) free radical.

The enzymatically generated free radical of the antitumor agent diaziquone is analyzed with the help of two analogs where either the aziridine rings (RQ14) or the carboethoxyamino groups (RQ2) were substituted by chlorine atoms. The hyperfine couplings observed in the diaziquone free radical are due to the nitrogens in the aziridine group. Unresolved coupling and hindered rotation contribute to line broadening. We find that diaziquone free radicals are more stable than RQ14 but less stable than RQ2 free radicals. The reason for this is that the carboethoxyamino groups make the aromatic ring unstable, while the aziridines contribute to its stability. The free radical observed in diaziquone is in all probability that of the parent compound and not that of an intermediate metabolite.

Mechanism(s) of bioreductive activation. The example of diaziquone (AZQ).

Bioreduction in the activation of diaziquone (2,5-diaziridinyl-3,6-bis (carboethoxyamino)-1,4-benzoquinone) has been investigated by exploring its reduction by whole cells, rat liver microsomes and purified enzymes. The mechanism of bioreduction was further investigated by exploring the chemical and electrochemical reduction of diaziquone as well as its photochemistry. Reduced diaziquone (by several means) was then tested for activity against parent compound. It appears that reduced diaziquone in most cases is more active than the oxidized form. Diaziquone redox cycles, but it is easily reduced to the hydroquinone which oxidizes to the semiquinone yielding free radicals under aerobiosis. The most probable mechanism of action is that of bioreductive alkylation where the alkylating aziridines are protonated after reduction facilitating the opening of the aziridine rings and thus alkylation.

The role of NAD(P)H oxidoreductase (DT-Diaphorase) in the bioactivation of quinone-containing antitumor agents: a review.

Bioactivation of quinone-containing anticancer agents has been studied extensively within the context of the chemistry and structure of the individual quinones which may result in various mechanisms of bioactivation and activity. In this review we focus on the two electron enzymatic reduction/activation of quinone-containing anticancer agents by DT Diaphorase (DTD). This enzyme has become important in oncopharmacology because its activity varies with tissues and it has been found to be elevated in tumors. Thus, a selective tumor cell kill can exist for agents that are good substrates for this enzyme. In addition, the enzyme can be induced by a variety of agents, a fact that can be used in chemotherapy. That is induction by a nontoxic agent followed by treatment with a good DT-Diaphorase substrate. A wide variety of anticancer drugs are discussed some of which are not good substrates such as Adriamycin, and some of which are excellent substrates. The latter category includes a variety of quinone containing alkylating agents.

The reductive metabolism of diaziquone (AZQ) in the S9 fraction of MCF-7 cells: free radical formation and NAD(P)H: quinone-acceptor oxidoreductase (DT-diaphorase) activity.

The S9 fraction of MCF-7 human breast carcinoma cells has NAD(P)H (quinone-acceptor) oxidoreductase activity as measured by the reduction of dichlorophenol-indophenol (DCPIP). This reduction is dependent on the activators Tween-20 and bovine serum albumin and it is inhibitable by dicumarol. The S9 fraction also has cytochrome c reductase activity which is approximately 29 times less than the two-electron reduction activity of NAD(P)H (quinone-acceptor) oxidoreductase. Diaziquone (AZQ) is a substrate for this NAD(P)H oxidoreductase active S9 fraction as judged by its enzymatic reduction detected spectrophotometrically and by electron spin resonance spectroscopy. Two-electron mediated enzymatic reduction of AZQ was evidenced by the formation of the colorless dihydroquinone (AZQH2) which could be followed at 340 nm. The production of the dihydroquinone was inhibitable by dicumarol implicating NAD(P)H oxidoreductase in its formation. Under aerobic conditions, electron spin resonance spectroscopy showed evidence for the production of AZQ semiquinone (AZQH) and oxygen radicals. Under anaerobic conditions no oxygen radicals were observed, but the semiquinone was stable for hours. These results are also inhibitable by dicumarol and suggest a two-step one-electron oxidation process of the dihydroquinone. The production of semiquinone and oxygen radicals as detected by electron spin resonance spectroscopy was more sensitive to dicumarol when NADPH was used as cofactor (68% inhibition of OH and 65% inhibition of AZQH) than when NADH was used (28% inhibition of OH and 5% inhibition of AZQH). This suggests that NADH flavin reductases play a more important role in the one-electron reduction pathway of AZQ in MCF-7 S9 fraction than NADPH reductases. The reduction of AZQ by NAD(P)H (quinone-acceptor) oxidoreductase may play an important role in the bioreductive alkylating properties of AZQ.

Free radical formation and DNA strand breakage during metabolism of diaziquone by NAD(P)H quinone-acceptor oxidoreductase (DT-diaphorase) and NADPH cytochrome c reductase.

One-electron reduction of diaziquone (AZQ) by purified rat liver NADPH cytochrome c reductase was associated with formation of AZQ semiquinone, superoxide anions, hydrogen peroxide, and hydroxyl radicals as indicated by ESR spin-trapping studies. Reactive oxygen formation correlated with AZQ-dependent production of single and double PM2 plasmid DNA strand breaks mediated by this system as detected by gel electrophoresis. Direct two-electron reduction of AZQ by purified rat liver NAD(P)H (quinone acceptor) oxidoreductase (QAO) was also associated with formation of AZQ semiquinone, superoxide anions, hydrogen peroxide, and hydroxyl radicals as detected by ESR spin trapping. Furthermore, PM2 plasmid DNA strand breaks were detected in the presence of this system. Plasmid DNA strand breakage was inhibited by dicumarol (49 +/- 5%), catalase (57 +/- 2.3%), SOD (42.2 +/- 3.6%) and ethanol (41.1 +/- 3.9%) showing QAO and reactive oxygen formation was involved in the PM2 plasmid DNA strand breaks observed. These results show that both one- and two-electron enzymatic reduction of AZQ give rise to formation of reactive oxygen species and DNA strand breaks. Autoxidation of the AZQ semiquinone and hydroquinone in the presence of molecular oxygen appears to be responsible for these processes. QAO appears to be involved in the metabolic activation of AZQ to free radical species. The cellular levels and distribution of this enzyme may play an important role in the response of tumor and normal cells to this antitumor agent.

Trace detection of hydroxyl radicals during the redox cycling of low concentrations of diaziquone: a new approach.

Quantifying oxygen radicals that arise during the redox cycling of quinone-containing anticancer agents such as diaziquone (AZQ) has been difficult, as has been their detection at low drug concentrations. This is due to the fact that EPR spin trapping, the method most often used for *OH detection, requires the use of high drug concentrations. Using a new highly sensitive technique that employs a fluorescamine-derivatized nitroxide, we show that low levels of NADPH-cytochrome P450 reductase (4.25 microg/ml) catalyze the production of hydroxyl radicals at very low, clinically relevant AZQ concentrations. Thus, at this enzyme concentration, we were able to detect a rate of 0.10 nM s(-1) hydroxyl radical production by 5 microM AZQ, a clinically relevant concentration. The Michaelis-Menten constants for AZQ-mediated hydroxyl radical production are: K(M) = 10.7 +/- 1.4 microM, and V(max) = 5.2 +/- 0.9 x 10(-8) M s(-1) (mg protein)(-1). Experiments employing catalase, superoxide dismutase, and NADPH-cytochrome P450 reductase, confirm the previously deduced conclusions from high drug concentrations, that is, that at low concentrations, AZQ acts to shuttle reducing equivalents from the enzyme to oxygen, thus generating the redox cycle. The data presented here suggest that the levels and locations of redox active metal ions may be the principal controlling factor in the pathway of AZQ activity that involves oxidative stress.

Modulation of streptonigrin cytotoxicity by nitroxide SOD mimics.

Nitroxides are cell-permeable, stable radicals that react readily with paramagnetic species such as transition metals or short-lived free radicals, though not generally with diamagnetic molecules. Nitroxides can undergo one-electron selective redox reactions and thereby potentially modify the activity of cytotoxic drugs. Streptonigrin (SN) toxicity requires bioreduction to yield the semiquinone radical, and the toxicity is reportedly mediated by transition metals and oxygen-derived reactive species via redox-cycling of the semiquinone intermediate. The present study shows that (1) nitroxides protected isolated DNA and also aerated or hypoxic bacterial cells from SN toxicity; (2) H2O2 potentiated the hypoxic cytotoxicity of the drug but inhibited the damage to aerated cells; (3) pretreatment of cells with H2O2 conferred some protection, but not when the drug alone was preexposed to H2O2; and (4) desferrioxamine and 2,2-dipyridyl, though neither diethylenetriamino pentaacetate, exogenous catalase, or superoxide dismutase, decreased SN-induced cell killing. The mechanisms by which nitroxides protect from SN toxicity involve both a selective radical-radical reaction with SN semiquinone and the reoxidation of reduced cellular transition metal ions. On the other hand, H2O2 appears to exert two opposing effects: (1) facilitation of cell killing by the Fenton reaction and (2) lowering the cellular level of reducing equivalents, thus inhibiting the bioreductive activation of SN.

The metabolism of quinone-containing alkylating agents: free radical production and measurement.

The metabolism of quinone-containing antitumor agents involves enzymatic reduction of the quinone by one or two electrons. This reduction results in the formation of the semiquinone or the hydroquinone of the anticancer drug. The consequence of these enzymatic reductions is that the semiquinone yields its extra electron to oxygen with the formation of superoxide radical anion and the original quinone. This reduction by a reductase followed by oxidation by molecular oxygen (dioxygen) is known as redox-cycling and continues until the system becomes anaerobic. In the case of a two electron reduction, the hydroquinone could become stable, and as such, excreted by the organism in a detoxification pathway. In some cases such as aziridine quinones, the hydroquinone can be oxidized by one electron at a time resulting in the production of superoxide, the semiquinone and the parental quinone. Quinone anticancer agents upon reduction can also set up an equilibrium between the hydroquinone, the parental quinone and the semiquinone which results in a long-lived semiquinone. Depending on the compound, aziridine quinones, for example, this equilibrium is long lasting thus allowing for the detection of the semiquinone under aerobic conditions. This phenomenon is known as comproportionation-disporportionation equilibrium. The series of reviews in this Special Issue address the consequences of bioreduction of quinone alkylators used in the treatment of cancer. In this particular review we are interested in describing the phenomenon of redox-cycling, how it is measured, and the biological consequences of the presence of the semiquinone and the oxygen radicals generated.

In the search for new anticancer drugs. XX: A comparison of the in vitro growth inhibition of P388 cells by TEPA and N,N;N',N'-bis (1,2-ethanediyl)-N''-(1-oxyl-2,2,6,6-tetramethyl-4- piperidinylaminocarbonyl) phosphoric triamide and congeners.

We tested the in vitro growth inhibitory activity of TEPA, and three analogs against P388 murine lymphocytic leukemia cells in culture. The analogs consist of spin-labeled TEPA and two reduced forms containing the NH and NOH groups instead of the nitroxyl function. Spin label TEPA was obtained by replacing one of the aziridine groups in TEPA with spin-labeled urea. In a concentration range of 10(-6)-10(-5) M, only the reduced analog containing the NH group was active. That is, to achieve a 50% inhibition of cell growth, a five-fold excess in concentration of this analog (IC50 = 10 X 10(-6) M) was needed as compared to the parent compound TEPA (IC50 = 2 X 10(-6) M). These results are in contrast with those obtained in vivo against the same leukemia cell line, indicating inherent discrepancies between in vivo and in vitro evaluation of antitumor agents.

Cellular pharmacology of N,N',N''-triethylene thiophosphoramide.

N,N',N''-triethylene thiophosphoramide (Thio-TEPA) is an alkylating agent whose antineoplastic activity has been known for nearly 30 years. Human plasma pharmacokinetic studies revealed the presence of TEPA, a Thio-TEPA metabolite which after 4 h achieved plasma concentrations equal to those of the parent compound. We studied the activity of both Thio-TEPA and TEPA against murine leukemia P388 cells in culture. We found that Thio-TEPA is approximately two-fold more active than TEPA in arresting cell growth (IC50 = 2.8 microM for TEPA and 1.5 microM for Thio-TEPA). In inhibiting [3H]thymidine incorporation, Thio-TEPA and TEPA have the same activity (IC50 = 2 microM for both compounds). Experiments in which drug was removed from cell cultures which were further incubated in drug-free media, revealed that the bulk of the cell damage occurs during the first 4 h of incubation. Cell cultures exposed to 0.5 microM Thio-TEPA for 22 h fully recovered their [3H]thymidine incorporation ability after 24 h of drug-free incubation. Cells exposed to 2.5 microM Thio-TEPA for 22 h partially recovered their ability to incorporate [3H]thymidine. Cells exposed to 10 microM Thio-TEPA for 22 h did not recover their ability to incorporate [3H]thymidine. Gas liquid chromatographic analysis of the media from incubated cells showed that the concentration of Thio-TEPA remained unchanged during the incubations and that TEPA was not present. In Thio-TEPA doses ranging from 0.1 microM to 100 microM, [3H]uridine and [3H]-leucine incorporation were less affected than [3H]thymidine incorporation. This may indicate that a longer observation time may be needed to allow the DNA damage to be expressed in terms of protein or RNA synthesis.

Reductive metabolism of diaziquone (AZQ) in the S9 fraction of MCF-7 cells. II. Enhancement of the alkylating activity of AZQ by NAD(P)H: quinone-acceptor oxidoreductase (DT-diaphorase).

The alkylating activity of reduced diaziquone was studied by the nitrobenzylpyridine (NBP) assay and was compared to those of the parent compound and aziridine-containing N,N',N"-triethylenethiophosphoramide (Thio-TEPA). Diaziquone (AZQ) was reduced enzymatically by 2e- using S9 cell fraction from MCF-7 cells which is rich in NAA(P)H:quinone-acceptor oxidoreductase (DT-diaphorase) (QAO) activity. One electron enzymatic reduction was performed with NADPH-cytochrome c reductase. The alkylating activity of AZQ increased 3-fold when reduced by 2e-. This increase was inhibited by dicumarol, an inhibitor of QAO. In contrast, the alkylating activity of AZQ did not increase beyond that of the parent compound when reduced by 1e- using purified NADPH-cytochrome c reductase. Similar results were obtained when AZQ was reduced chemically with borohydride (2e-) and with NADPH (1e-). Anaerobic incubations of AZQ with the S9 fraction of MCF-7 cells (2e- reduction) resulted in an increase in NBP alkylation over its aerobic counterpart (1.8-fold) while maintaining the near 3-fold increase in alkylation over untreated AZQ. In contrast, AZQ incubations with NADPH-cytochrome c reductase (1e- reduction) under the same conditions did not result in an NBP alkylation increase over untreated AZQ. These results indicate that AZQ hydroquinone is most likely the responsible species for the observed alkylation of this antitumor agent to DNA and other nucleophiles. The results also suggest that NAD(P)H:quinone-acceptor oxidoreductase is a very important enzyme in the bioactivation of AZQ.

NAD(P)H (quinone acceptor) oxidoreductase (DT-diaphorase)-mediated two-electron reduction of anthraquinone-based antitumour agents and generation of hydroxyl radicals.

The anthraquinone-based antitumour agents mitoxantrone, daunorubicin and ametantrone were found to be substrates for NAD(P)H (quinone acceptor) oxidoreductase (DT-diaphorase) [QAO] isolated from rat liver. This was indicated by the stimulation of QAO-dependent NADPH oxidation by these agents. This effect followed Michaelis-Menten kinetics and was dependent on the concentration of QAO, inhibited by the specific QAO inhibitor dicumarol (15 microM) and enhanced by the QAO activators bovine serum albumin (0.01%) and Triton X-100 (0.03%). As indicated by the Vmax/Km ratio, mitoxantrone (26.53) was considerably more active than ametantrone (11.25) or daunorubicin (7.35). Metabolism of these anthraquinones was associated with the formation of superoxide anions, hydrogen peroxide and hydroxyl radicals as indicated by electron spin resonance spin trapping studies with 5,5-dimethyl-1-pyrroline-N-oxide. This is likely to be due to the slow auto-oxidation of the respective dihydroquinones in the presence of molecular oxygen. QAO needs to be considered as a possible route of bioreductive activation of these agents.

Cellular activation of diaziquone [2,5-diaziridinyl-3,6-bis (carboethoxyamino)-1,4-benzoquinone] to its free radical species.

Human leukemic cell lines K562 and HL60, and the murine leukemic cell line L1210, reduce Diaziquone (AZQ) (NCS182986) to its free radical anion. With all cell lines, the free radical was observed immediately in both aerobic and anaerobic cell suspensions. The steady-state concentration of AZQ free radicals was approximately 1% of the total AZQ concentration. L1210 cells treated with azide reduced AZQ, but cells treated with diamide and N-ethylmaleimide did not. NADPH and L-cysteine reduced AZQ. The latter did so under anaerobic conditions; the former did so under both anaerobic and aerobic conditions.

Fragmentation characteristic of glutathione conjugates activated by high-energy collisions.

Product ion spectra of fifteen monoglutathione and diglutathione conjugates have been measured using activation by 6000-eV collisions with helium in the third field-free region of a four-sector tandem mass spectrometer of EBEB configuration. Fragmentation patterns in the cation spectra have been analyzed for decompositions of the glutathione moiety that would permit recognition of an unknown as a glutathione conjugate. Five spectra from an earlier study of high-energy collisional activation on a BEEB four-sector instrument have also been included in this analysis. A suite of appropriate ions was found to occur consistently,, including ions of m/z 307 comprising the glutathione tripeptide and the complementary ion [MH-307](+) or the ion radical [MH-306](+).

In vitro action of continuous-wave ultrasound combined with adriamycin, X rays or hyperthermia.

We compared the ability of continuous-wave ultrasound to enhance cytotoxicity from X irradiation, hyperthermia or exposure to adriamycin. The survival of CHO cells exposed in culture medium to these agents was determined with and without continuous-wave ultrasound (1.62 or 1.765 MHz). In water-filled transmission exposure vessels with 2-cm-diameter Mylar end windows, 10-min insonation not producing cytotoxicity could produce .OH radicals (measured by electron paramagnetic resonance) even at 0.4 W/cm2. Ultrasound at intensities ranging between 1 and 2.5 W/cm2 increased the clonogenic cytotoxicity of adriamycin (P = 0.0023 by paired t test) but not of X rays (2-10 Gy) or hyperthermia (44 degrees C for 10-50 min). The only significant action of continuous-wave ultrasound under similar test conditions was the potentiation of adriamycin-induced clonogenic cytotoxicity, possibly mediated by cavitational activity.