Inhibition of topoisomerase II by antitumor agents bis(2,6-dioxopiperazine) derivatives.

Several recently developed derivatives of bis(2,6-dioxopiperazine) have been shown to be new antitumor agents and are currently under clinical trials. We found that the mother compound of the bis(2,6-dioxopiperazine)s, ICRF-154, and its derivatives, ICRF-159, ICRF-193, and MST-16, are all inhibitors of mammalian type II DNA topoisomerase. By decatenation assay using kinetoplast DNA from Crithidia fasciculata, inhibition of purified calf thymus topoisomerase II by these compounds was investigated. Potency of inhibition was in the following order: ICRF-193 greater than ICRF-154 = ICRF-159 greater than MST-16. The doses giving 50% inhibition were 2, 13, 30 and 300 microM, respectively, for these compounds. ICRF-193, the most potent inhibitor, however, did not inhibit topoisomerase I at concentrations up to 300 microM. Addition of excess enzyme, but not of the substrate DNA, overcame the inhibition by ICRF-193. The drug did not stimulate the formation of cleavable complex between DNA and the enzyme. Furthermore, ICRF-193 even inhibited the formation of enzyme-mediated DNA cleavage induced by etoposide or 4'-[9-acridinylamino)methanesulfon-m-anisidide. These observations, together with the finding that ICRF-193 did not intercalate into DNA, suggest that ICRF-154 and related compounds are specific inhibitors of topoisomerase II with different modes of action: i.e., they interfere with some step(s) before the formation of the intermediate cleavable complex in the catalytic cycle. This is a property quite distinct from previously known cleavable complex-forming type topoisomerase II-targeting antitumor agents such as acridines, anthracyclines, and epipodophyllotoxins, but rather, mechanistically similar to the recently reported group of inhibitors that includes merbarone, aclarubicin, and fostriecin.

Properties of a coenzyme, pyrroloquinoline quinone: generation of an active oxygen species during a reduction-oxidation cycle in the presence of NAD(P)H and O2.

The oxidation of NAD(P)H by pyrroloquinoline quinone (PQQ) was non-enzymatically carried out at physiological pH in the presence of O2. The PQQ-NAD(P)H system requires about 1 mol of O2 for the oxidation of 1 mol of NAD(P)H. The oxidation of NAD(P)H occurred at a pseudo-first-order rate with respect to NAD(P)H and was of zero order with respect to PQQ concentration in in the presence of O2: k0[PQQ] [NAD(P)H] = k1 [NAD(P)H], where k0[PQQ] = k1, in which [PQQ] represents the initial concentration of PQQ. k0 values for NADH and NADPH were 3.4.10(2) M-1.min-1 and 2.0.10(2) M-1.min-1, respectively, at 25 degrees C and at 258 microM O2 (initial concentration). The system produced O-2, probably by the interaction of PQQ.H and/or NAD(P).with O2, during the oxidation of NAD(P)H. PQQH2 and PQQ.H were easily oxidized to PQQ in the presence of O2, yielding H2O2.

Generation of hydroxyl radicals during the enzymatic reductions of the Fe3+-ADP-phosphate-adriamycin and Fe3+-ADP-EDTA systems. Less involvement of hydroxyl radical and a great importance of proposed perferryl ion complexes in lipid peroxidation.

A system which contains NADPH, purified cytochrome P-450 reductase (enzyme) and Fe3+-ADP-adriamycin complex in Tris-HCl buffer does not produce hydroxyl radical, but possesses a strong lipid peroxidation activity on exogenously added phospholipid micelles. Fe3+-ADP-adriamycin complex, a tightly coordinated complex in Tris-HCl buffer, could be dissociated to Fe3+-ADP-phosphate complex and adriamycin in phosphate buffer. Hydroxyl radical, which can be detected by a spin trapping method using N-tert-butyl-alpha-phenylnitrone, is produced during the enzymatic reduction of a mixture of Fe3+-ADP-phosphate complex and adriamycin or of Fe3+-ADP-EDTA complex while it is not involved in phospholipid peroxidation under the conditions used. With hydroxyl radical-generating systems, little or no quenching of hydroxyl radical in Tris-HCl buffer could be demonstrated. The oxidative cleavage of phospholipid is initiated by the proposed perferryl ion complex, which may be generated by the interaction of Fe2+-ADP-adriamycin complex with O2. A similar perferryl ion complex is also produced during the enzymatic reduction of Fe3+-ADP-EDTA complex with a molar ratio of 2 for [Fe3+]/[EDTA] in the presence of air. This is also able to catalyze lipid peroxidation.

Mechanism of O2- generation in reduction and oxidation cycle of ubiquinones in a model of mitochondrial electron transport systems.

O2- generation in mitochondrial electron transport systems, especially the NADPH-coenzyme Q10 oxidoreductase system, was examined using a model system, NADPH-coenzyme Q1-NADPH-dependent cytochrome P-450 reductase. One electron reduction of coenzyme Q1 produces coenzyme Q1-. and O2- during enzyme-catalyzed reduction and O2+ coenzyme Q1-. are in equilibrium with O2- + coenzyme Q1 in the presence of enough O2. The coenzyme Q1-. produced can be completely eliminated by superoxide dismutase, identical to bound coenzyme Q10 radical produced in a succinate/fumarate couple-KCN-submitochondrial system in the presence of O2. Superoxide dismutase promotes electron transfer from reduced enzyme to coenzyme Q1 by the rapid dismutation of O2- generated, thereby preventing the reduction of coenzyme Q1 by O2-. The enzymatic reduction of coenzyme Q1 to coenzyme Q1H2 via coenzyme Q1-. is smoothly achieved under anaerobic conditions. The rate of coenzyme Q1H2 autoxidation is extremely slow, i.e., second-order constant for [O2][coenzyme Q1H2] = 1.5 M-1.s-1 at 258 microM O2, pH 7.5 and 25 degrees C.

Importance of Fe2+-ADP and the relative unimportance of OH in the mechanism of mitomycin C-induced lipid peroxidation.

The mechanism of mitomycin C-induced lipid peroxidation has been studied at pH 7.5, using systems containing phospholipid membranes (liposomes) and an Fe3+-ADP complex with purified NADPH-cytochrome P-450 reductase. Both O2- and H2O2 are generated during the aerobic enzyme-catalyzed reaction in the presence of mitomycin C. Hydroxyl radical is formed in the reaction by the reduction of H2O2. This is catalyzed by the Fe2+-ADP complex in a phosphate buffer or to a lesser extent when in a Tris-HCl buffer. The reduction of Fe3+-ADP to Fe2+-ADP is mainly achieved by O2-. The resulting Fe2+-ADP in the presence of O2 forms a perferryl ion complex which is a powerful stimulator of lipid peroxidation. However, the formation of such an iron-oxygen complex is strongly inhibited by phosphate ions, which do not interfere with the generation of OH radicals. These findings suggest that, since lipid peroxidation occurs in a Tris-HCl buffer (but not in a phosphate buffer), the OH radical is unlikely to be involved in the observed lipid peroxidation process.

Fluorescence digital image analysis of serotonin-induced calcium oscillations in single blood platelets.

The intracellular concentration of Ca2+ [( Ca2+]i) was monitored continuously in single rabbit blood platelets by digital imaging microscopy in conjunction with Fura-2, a specific Ca(2+)-indicator dye. Ionomycin as well as aluminium fluoride caused sustained increase in [Ca2+]i in the platelet, but oscillations of [Ca2+]i were not observed. Serotonin (5-HT) induced oscillatory increases in [Ca2+]i in the presence of 1 mM CaCl2; these had not been detectable in cell populations because the oscillations were not in synchrony. This effect of 5-HT was diminished when CaCl2 was omitted from the medium, and was antagonized by 1 microM ketanserin, a specific 5-HT2 receptor antagonist. Furthermore, DOI, a specific 5-HT2 agonist, had the same effect as 5-HT at lower concentration. A specific effector mechanism, not fully understood at present, therefore appears to mediate 5-HT2 receptors thereby allowing rabbit platelets to generate [Ca2+]i oscillations. It is suggested that protein kinase C in platelets might play a key role in the regulation of [Ca2+]i, and possibly in [Ca2+]i oscillations.

Acidic phospholipids directly inhibit DNA binding of mammalian DNA topoisomerase I.

Inhibition of mammalian DNA topoisomerase I by phospholipids was investigated using purified enzyme. Acidic phospholipids inhibited the DNA relaxation activity of topoisomerase I whereas neutral phospholipid, phosphatidylethanolamine, did not. Accumulation of a protein-DNA cleavable complex, an intermediate which is known to accumulate upon inhibition by a specific inhibitor camptothecin, did not occur. The filter binding assay revealed that the DNA binding activity of the enzyme was inhibited by acidic phospholipids. Moreover, direct binding of phosphatidylglycerol to topoisomerase I was demonstrated. These results indicated that the inhibitory effect of acidic phospholipids on topoisomerase I was due to the loss of the DNA binding of the enzyme as a result of direct interaction between phospholipids and the enzyme.

Morphological and biochemical evidence that scrapie-associated fibrils are derived from aggregated amyloid-like filaments.

The membrane fraction from scrapie infected mouse brains was dissolved in saturated urea, centrifuged on a 10 to 50% glycerol gradient at 35,000 rpm for 24 h, and fractionated from the bottom of the tube into 11 fractions. PrP was detected throughout the gradient. However, the relative PrP concentrations of fractions 4 and 8 were the highest. The relative PrP concentration versus protein concentration of fractions 1 to 4 was higher than that of the other fractions. Scrapie infectivity also was detected in all fractions. Fractions 2, 3, 4, 7, and 8 produced the shortest incubation periods. Positively stained filamentous aggregates with sizes varying from about 40 x 60 nm to more than 4 microns were observed in fractions 2 and 4 by negative staining. These resembled amyloid filaments. Congo red-stained aggregates showed birefringence under polarized light. Aggregation of the filamentous aggregates was observed by incubation with anti-mouse SAF serum. Fine fibrils 10-18 nm in width were partially dissociated from the aggregates by brief exposure to the detergent Sarkosyl. These facts suggest that SAF are not products of self-assembly from subunit structures liberated from membranes by exposure to detergent, but exist as aggregates of amyloid-like filaments from which SAF are dissociated by detergent extraction.

Conversion of endoglycoceramidase-activator II by trypsin to the 27.9 kDa polypeptide possessing full activity: purification of activator for endoglycoceramidase by trypsin treatment followed by trypsin-inhibitor agarose column application.

Endoglycoceramidase (EGCase) cleaves the linkage between oligosaccharides and ceramides of various glycosphingolipids [Ito, M. & Yamagata, T. (1986) J. Biol. Chem. 261, 14278-14282]. A detergent was required for EGCase to express full activity, possibly due to its hydrophobic nature. Recently, activator proteins responsible for stimulating EGCase activity in the absence of detergents were isolated from the culture supernatant of Rhodococcus sp. [Ito, M., Ikegami, Y., & Yamagata, T. (1991) J. Biol. Chem. 266, 7919-7926]. The activity of activator II specific for EGCase II was heat-labile but insensitive to trypsin-treatment. This activator (69.2 kDa) was converted to the 27.9 kDa polypeptide via the 42 kDa intermediate by exhaustive trypsination, and the stimulatory activity of 27.9 kDa polypeptide on EGCase II was identical to that of the native form toward asialo GM1 and cell-surface GM3 of horse erythrocytes as substrates. This observation was successfully applied to obtain the purified activator without contamination with EGCase activity, which is abolished completely following treatment with trypsin.

Different behavior towards raw starch of three forms of glucoamylase from a Rhizopus sp.

Three forms of glucoamylase [EC 3.2.1.3] of a Rhizopus sp., Gluc1 (M.W. 74,000), Gluc2 (M.W. 58,600), and Gluc3 (M.W. 61,400), which have similar pH optima and specific activities towards soluble starch were studied as to their behavior towards raw starch. The pH optima for raw starch digestion were different, that is, 4.5 for Gluc1 and 5.0 for both Gluc2 and Gluc3. All the enzymes digested raw starch almost completely but at quite different rates; Gluc2 and Gluc3, which lack the N-terminal portions of Gluc1, were 22 and 25 times less effective, respectively, for raw starch digestion than Gluc1. Of the three enzymes, only Gluc1 tightly bound to raw starch. Binding of Gluc1 to raw starch occurred pH-dependently with a broad pH optimum of 4.5-5.5, but temperature and ionic strength affected it only slightly and little, respectively. The binding constant of Gluc1 for raw starch at pH 5.0 and 4 degrees C was estimated to be 1.2 X 10(5) M-1. Fragment H (M.W. 16,700), presumably released from the N-terminal part of Gluc1, not only bound to raw starch itself but also inhibited the binding of Gluc1 to raw starch. pap-Gluc (M.W. 57,000) and chymo-Gluc (M.W. 64,000), which are papain- and chymotrypsin-modified Gluc1, respectively, and lack the N-terminal portions of Gluc1, resembled Gluc2 and Gluc3 in raw starch binding as well as digestion. All these results indicate that Gluc1 has a raw starch-binding site, different from the active center, in the N-terminal region. Various substrates and analogs inhibited the binding of Gluc1 to raw starch, presumably due to steric hindrance.

Fluorescence digital image analysis of thrombin and ADP induced rise in intracellular Ca2+ concentration of single blood platelets.

Cytoplasmic free Ca2+ concentration, [Ca2+]i, was estimated in single rabbit blood platelets by digital imaging microscopy with the use of the specific Ca(2+)-indicator dye Fura-2. Uneven distribution and low level of [Ca2+]i was found in the resting platelet even in the presence of extracellular 1 mM Ca2+. Thrombin at 1 unit/ml immediately caused a transient increase in [Ca2+]i, which was followed by a secondary and sustained increase in [Ca2+]i. The distribution of increased levels of [Ca2+]i was also shown to be uneven within the cell. The presence of 1 mM EGTA in the medium only slightly decreased the initial rise in [Ca2+]i, but completely inhibited the latter phase, a sustained rise in [Ca2+]i. This result shows that the initial rise of [Ca2+]i might not be caused by Ca2+ influx, but might be induced by mobilization of Ca2+ from intracellular Ca2+ storage sites. This speculation is further supported by the fact that the elevated [Ca2+]i induced by thrombin immediately decreased to the base line value when 3 mM EGTA was applied. Thus, thrombin induced elevation of [Ca2+]i is suggested to consist of two different processes, namely the mobilization of Ca2+ from the intracellular storage sites and the successive Ca2+ influx through the receptor activated Ca2+ channels. Stimulation with ADP also caused a rapid elevation of platelet [Ca2+]i, but this effect of ADP was different form that of thrombin. Thus, the ADP induced rise in [Ca2+]i was accompanied by oscillation and was inhibited by extracellular EGTA. Our present experiment is the first report that clearly and directly reveals the differences between the effects of thrombin and ADP on [Ca2+]i of platelets.

Relationships between serotonin induced elevation of intracellular Ca2+ concentration and stimulation of Ca2+ influx in blood platelets.

Serotonin (5-HT) caused immediate elevation of intracellular Ca2+ concentration ([Ca2+]i) in blood platelets, and it was completely inhibited by 1 mM EGTA. In Ca2+ replenished platelets, however, 2 mM EGTA did not affect the 5-HT induced elevation of [Ca2+]i when EGTA was applied just before or during the stimulation by 5-HT. At the same concentration 5-HT was also found to enhance Ca2+ influx through the activation of 5-HT2 receptor, but with rather longer latent time. From these results it is suggested that 5-HT induced elevation of [Ca2+]i is caused by mobilization of Ca2+ from intracellular Ca2+ storage sites, but not by direct stimulation of Ca2+ influx. Depletion of such Ca2+ stores might impair the effect of 5-HT on [Ca2+]i. Thus, 5-HT induced augmentation of Ca2+ influx might be secondary to replenishment of the depleted Ca2+ stores which was caused by 5-HT induced internal release of Ca2+. It is concluded that the effects of 5-HT on [Ca2+]i and Ca2+ influx in platelets are manifested sequentially or independently.

Spinal cord lactate concentration during chemical stimulation of the nucleus tractus solitarii in anesthetized rats.

This study was conducted to determine the mechanism of spinal cord blood flow (SCBF) decrease following the nucleus tractus solitarii (NTS) activation. In urethane-anesthetized, paralyzed and artificially ventilated rats, neurons in the NTS were chemically stimulated by microinjection of L-glutamate (1.7 nmol; 50 nl) and the lactate concentration, one of indicators of local neuronal metabolism, in the spinal cord was monitored in real time using an enzyme electrode. Before the chemical stimulation study, the responses of the enzyme electrode and its specificity were tested in vitro and in vivo. The electrode responded to step changes in lactate concentration and a calibration plot and regression line were obtained in vitro. The lactate concentration was significantly (P < 0.01) increased during induced apnea in vivo (n = 8). The lactate concentration in the spinal cord was not significantly changed by chemical stimulation of the NTS when arterial blood pressure (ABP) remained above the lower limit of spinal cord autoregulation (n = 21). When chemical stimulation of the NTS decreased ABP to below the lower limit of autoregulation (n = 18), the lactate concentration in the spinal cord was significantly (P < 0.01) increased. This may only be due to hypotensive effects because the lactate concentration was also significantly (P < 0.01) increased when the ABP was passively decreased below the lower limit of autoregulation by controlled hemorrhage in intact (n = 11) and sinoaortic denervated rats (n = 10). Intravenous lactate injection produced no significant increase in the current from the enzyme electrode in the spinal cord (n = 4). Using the electrode with inactivated enzyme solution, the current from the electrode did not change with the increase in lactate in the spinal cord. These findings indicate that the enzyme electrode can detect rapid changes of lactate, a product of anaerobic metabolism. These results also indicate that the spinal cord vasoconstrictor response elicited by chemical stimulation of the NTS, which was performed above the lower limit of spinal cord autoregulation in our previous study, may be due to neurogenic regulatory mechanism, but not to the secondary effects of changes in metabolism.

Cerebral extracellular lactate concentration and blood flow during chemical stimulation of the nucleus tractus solitarii in anesthetized rats.

The extracellular lactate concentration and blood flow in the cerebral cortex of urethane-anesthetized, paralyzed and artificially ventilated rats were monitored continuously and simultaneously using an enzyme electrode and a laser Doppler flowmeter (LDF), respectively, during chemical stimulation of the nucleus tractus solitarii (NTS) by microinjection of L-glutamate (1.7 nmol 50 nl). Chemical stimulation of the NTS significantly decreased the arterial blood pressure (ABP) from 85 +/- 17 to 68 +/- 14 mmHg, heart rate from 418 +/- 13 to 402 +/- 19 beats x min(-1) and cerebral blood flow (CBF) by 17.9 +/- 6.2% (P < 0.001). However, chemical stimulation of the NTS significantly increased the lactate concentration by 58.9 +/- 17.3 microM (P < 0.001). Barostat maneuver, which held systemic ABP constant during chemical stimulation of the NTS attenuated the responses in CBF and lactate concentration by 30 and 27%, respectively. The onset of the increase in lactate concentration was delayed about 19 s after that of the CBF decrease. Circulatory lactate produced no significant change in the cerebral extracellular lactate concentration. These results indicate that chemical stimulation of the NTS induces an increase in extracellular lactate concentration in the cerebral cortex through a decrease in CBF via cerebral vasoconstriction.

K(+) channel blockers and cytochrome P450 inhibitors on acetylcholine-induced, endothelium-dependent relaxation in rabbit mesenteric artery.

Acetylcholine caused an endothelium-dependent relaxation in isolated rabbit mesenteric small artery in the presence of nitro L-arginine and indomethacin. The acetylcholine-induced relaxation was attenuated by high K(+) solution, suggesting that the response is mediated by a membrane potential-sensitive mechanism, presumably an endothelium-derived hyperpolarizing factor. The acetylcholine-induced relaxation was also inhibited with tetraethylammonium, 4-aminopyridine and charybdotoxin, but not with Ba(2+), apamin, iberiotoxin nor glibenclamide. The relaxation was abolished by a combination of apamin and charybdotoxin, but iberiotoxin could not replace charybdotoxin in this combination. The responses to charybdotoxin and 4-aminopyridine were synergistic but neither apamin nor iberiotoxin increased the effect of 4-aminopyridine. Clotrimazole and proadifen inhibited the acetylcholine-induced relaxation, but these drugs also inhibited the cromakalim-induced relaxation, while protoporphyrin IX inhibited the acetylcholine- but not cromakalim-induced relaxation. 17-Octadecynoic acid and 1-aminobenzotriazole did not affect the response to acetylcholine. Four regioisomers of epoxyeicosatrienoic acids did not relax endothelium-denuded artery. A gap junction inhibitor 18alpha-glycyrrhetinic acid attenuated the relaxation to acetylcholine. It is suggested that in rabbit mesenteric artery, the acetylcholine-induced, nitric oxide- and prostacyclin-independent relaxation is mainly mediated by 4-aminopyridine- and charybdotoxin-sensitive K(+) channels and that the relaxation is not mediated through cytochrome P450 enzyme metabolites. The contribution of heterocellular gap junctional communication to the relaxation is discussed.

Increased TUNEL positive cells in human alcoholic brains.

Alcohol-sensitive neuronal cell loss, which has been reported in the superior frontal cortex and hippocampus, may underlie the pathogenesis of subsequent cognitive deficits. In the present study, we have used the TUNEL labeling to detect the DNA damage in human alcoholic brains. Seven out of eleven alcoholics exhibited TUNEL-positive cells in both superior frontal cortex and hippocampus, which were co-localized with GFAP immunoreactivity. In contrast, almost no positive cells were detected in the non-alcoholic controls. None of the TUNEL-positive cells showed any typical morphological features of apoptosis or necrosis. TUNEL-positive cells observed in the present study may indicate DNA damage induced by ethanol-related overproduction of reactive oxygen species.

Chemical stimulation of the nucleus tractus solitarii decreases spinal cord blood flow in anesthetized rats.

L-Glutamate was microinjected into the nucleus tractus solitarii (NTS) in anesthetized (chloralose and urethane), paralyzed and artificially ventilated rats, and spinal cord blood flow (SCBF) was determined using a combination of labeled microspheres. Unilateral chemical stimulation of the NTS (n = 13) significantly decreased SCBF in the cervical cord from 43 +/- 6 (mean +/- SEM) to 28 +/- 4 (P < 0.05), in the thoracic cord from 35 +/- 3 to 24 +/- 4 (P < 0.01), and in the lumbar cord from 49 +/- 3 to 40 +/- 3 ml min-1 (100 g)-1 (P < 0.05). The decrease in SCBF was not due to the decrease in arterial blood pressure (ABP) because the SCBF during the chemical stimulation of the NTS was significantly smaller (P < 0.05) than the SCBF during controlled hemorrhagic hypotension (n = 11). Chemical stimulation of the NTS did not affect the reactivity of the spinal cord vessels to hypercapnia (n = 5). Microinjection of the vehicle solution into the NTS had no effects on spinal cord circulation (n = 9). These results suggest that the cell bodies within the NTS may play a role in the control of spinal cord circulation.

EPR and TREPR spectroscopic studies of antioxidant sesamolyl and related phenoxyl radicals.

Sesamolyl and related phenoxyl radicals were studied by conventional and time-resolved electron paramagnetic resonance (EPR) spectroscopic techniques. Continuous UV irradiation of sesamol in benzene produces two types of radicals. Based on the hyperfine coupling values obtained we determined that one is the neutral sesamolyl radical and the others are the dimer radicals. Comparison was made with related compounds, especially 3,4-dimethoxyphenol. We found that the 3,4-dimethoxyphenoxyl radical had a shorter lifetime than the neutral sesamolyl radical. The EPR results obtained suggest that a near perpendicular orientation of the oxygen p-orbitals with respect to the benzene ring of sesamol makes the radical more stable. This stability may be important for the antioxidant properties.