Effect of extracellularly generated free radicals on the plasma membrane permeability of isolated pancreatic B-cells

Ultrasound-enhanced transgene expression in vascular cells is not dependent upon cavitation-induced free radicals

Although acoustic cavitation is clearly important in ultrasound (US)-enhanced gene delivery (UEGD), the relative importance of mechanical and sonochemical (free radical) bioeffects remains unclear, as does the mechanism of gene delivery at the cellular level. Porcine vascular smooth muscle cells (VSMC) were transfected with luciferase or green fluorescent protein (GFP) plasmid +/- pulsed 956 kHz US (2.0 mechanical index (MI), 128 W cm(-2) spatial peak pulse average intensity, ISPPA) for 60 s, in the presence or absence of 20 mM cysteamine or N-acetyl-L-cysteine. Both compounds effectively scavenged free radical production following US, leaving unaffected the 50- to 100-fold enhancements in luciferase expression seen in US-treated VSMC. US exposure enhanced plasmid uptake (25 +/- 4.6 vs. 3 +/- 1.9 cells/field, n=4, p<0.05), most likely directly into the cytoplasm, and increased both the total number (>sevenfold) and average fluorescence intensity (>sixfold) of GFP-transfected cells. UEGD is not dependent upon cavitation-induced free radical generation and has potential for use with a wide range of therapeutic transgenes.

Free radical modulation of insulin release in INS-1 cells exposed to alloxan

Generation of free radicals is thought to mediate the cytotoxic action of alloxan on the pancreatic beta-cell. In this investigation, the early effects of alloxan on cell function were studied. When INS-1D insulinoma cells were exposed to alloxan (1 mM) for 45 min followed by a 3-hr recovery period, the drug increased basal insulin release while abolishing the effect of glucose in static incubations. This was associated with impaired stimulation of cellular metabolism by glucose and reduced viability, both monitored colorimetrically with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). These alterations were largely counteracted by the antioxidant butylated hydroxyanisol (BHA). Similar changes occurred when glucose was added directly after 5 min of alloxan treatment, whereas KCl-induced secretion was only partially inhibited. In perifusion, alloxan caused transient insulin secretion to 50% of the rates obtained with glucose 30 min later. Under these conditions, epinephrine abolished the stimulation due to both agents. Membrane potential and cytosolic calcium concentrations ([Ca2+]i) were recorded to clarify the action of alloxan. Alloxan-induced insulin release correlated with depolarization of INS-1D cells and a rise in [Ca2+]i. Alloxan did not augment [Ca2+]i in the presence of BHA or the absence of extracellular calcium. Nickel chloride blocked the effect of alloxan on [Ca2+]i, whereas verapamil was ineffective. This suggests that alloxan promotes Ca2+ influx through channels distinct from L-type channels, perhaps through non-selective cation channels. Thus, alloxan causes changes in INS-1D cells prevented by antioxidant treatment, suggesting that free radicals may modulate the ionic permeability leading to functional activation.

The pyridoindole antioxidant stobadine prevents alloxan-induced lipid peroxidation by inhibiting its propagation

Under in vitro conditions, the pyridoindole stobadine inhibited alloxan-induced lipid peroxidation in a model biological membrane with the efficacy comparable with that of the standard Trolox. Intermediary alloxan radicals and hydroxyl radicals were not directly involved in the process of lipid peroxidation, however, the presence of iron chelate was a necessary prerequisite. Since stobadine did not affect the kinetics of alloxan redox-cycling in the presence of GSH, we suggest that the protective action of stobadine against the alloxan-induced lipid peroxidation was mediated predominantly by its ability to quench peroxyl radicals, inhibiting thus the propagation stage of the oxidative damage. The results also indicate that toxic effects of alloxan may well be mediated by mechanism(s) not involving hydroxyl radicals.

Insulinoma cells in culture show pronounced sensitivity to alloxan-induced oxidative stress

In a model system of cultured J-774 cells we have previously shown that alloxan in moderate concentration is toxic only in the presence of a reducing agent with the production of hydrogen peroxide. The cytotoxicity was found to involve lysosomal destabilization. In the present study the cytotoxic effects of (i) alloxan alone, (ii) a combination of alloxan and cysteine or (iii) hydrogen peroxide were investigated in two established insulinoma cell lines (HIT-T15 and RINm5F), and compared with the effects on J-774 cells. The protective effects of desferrioxamine and catalase, and the intracellular levels of reduced glutathione and activities of the enzymes glutathione peroxidase, glutathione reductase and catalase were also studied. HIT and RIN cells showed about 10 times greater sensitivity than J-774 cells against exposure to either alloxan and cysteine, or hydrogen peroxide. All cell types were relatively insensitive to alloxan alone. Preincubation with desferrioxamine and addition of catalase provided efficient protection against cytotoxicity and lysosomal destabilization. HIT and RIN cells has less capacity to degrade hydrogen peroxide and lower levels of glutathione peroxidase than J-774 cells. The lysosomal stability in all three cell lines was directly correlated to their viability. We conclude that HIT and RIN cells have weak antioxidative defence systems resulting in enhanced lysosomal vulnerability when they are exposed to alloxan and cysteine, which produce hydrogen peroxide extracellularly. The degree of cytotoxicity seems to be dependent on cellular capacity to degrade hydrogen peroxide and the lysosomal content of reactive iron.

Alloxan cytotoxicity is highly potentiated by plasma membrane- and lysosomal-associated iron--a study on a model system of cultured J-774 cells

Pancreatic islet beta cells, and some other cell types, are sensitive to the damaging effects of alloxan. The mechanisms behind the cytotoxicity have not been fully elucidated, although they are considered to be mediated by the formation and effects of reactive oxygen metabolites. In the present study, the cytotoxic effects of alloxan/cysteine at high and low concentrations were investigated on a model system of cultured J-774 cells. Viability was estimated by the trypan blue dye exclusion test, plasma membrane permeability by a modified microfluorometric fluorescein diacetate technique and lysosomal membrane stability by a microfluorometric acridine orange method. The results showed: (a) hydrogen peroxide, readily diffusing through cellular membranes and produced extracellularly in large amounts by alloxan/cysteine at high concentrations, enters the secondary lysosomes if not previously degraded by cellular anti-oxidant systems. Intralysosomal Fenton reactions, with the formation of hydroxyl radicals, may be induced provided catalytically active lysosomal iron is present. This would result in lysosomal membrane damage followed by leakage of lysosomal contents to the cell sap and cell degeneration. (b) Alloxan/cysteine at low concentrations induced production of superoxide and hydrogen peroxide in low amounts which caused almost no lysosomal damage and appeared to be non-toxic unless there was some plasma membrane-associated iron. Consequently, cells initially allowed to endocytose iron during culture, or briefly exposed to iron just before exposure to alloxan and cysteine, showed greatly enhanced sensitivity. In this case iron, in combination with superoxide and hydrogen peroxide, is believed to give rise to plasma membrane-associated hydroxyl radical production (Fenton reaction) with resultant loss of membrane integrity.(ABSTRACT TRUNCATED AT 250 WORDS)

Delayed onset and decreased incidence of diabetes in BB rats fed free radical scavengers

We tested the hypothesis that free radicals play a role in the selective destruction of pancreatic beta-cells in BB/Wor rats. Diabetes-prone BB rats of both sexes and 40 days of age were divided into three groups. The control group was fed ad libitum Purina rat chow powder, while the experimental group was fed ad libitum the rat chow powder blended with a mixture of four known free radical scavengers: allopurinol, mercaptopropionylglycine, dimethylthiourea and Vitamin E. A third group was pair-fed 10 g chow powder/rat/day, since in earlier experiments we observed that rats on the experimental diet consumed only about 10 g/rat/day. All rats were studied up to age 120 days. Body weight and food intake were measured daily. Urine was tested for glucose beginning at age 60 days. When glucosuria appeared, blood glucose and urinary ketones were measured. Body weight gain in the experimental and pair-fed groups was similar, but lower than the control group. Life table analysis of the data showed a decreased and a delayed onset of diabetes in the rats fed free radical scavengers. Thus, the results of this study demonstrated that calorie restriction and the related impaired growth did not affect the incidence of diabetes in the BB rat. In addition, the results suggested a role for free radicals in the spontaneous destruction of pancreatic beta-cells in the BB rat.

Alloxan cytotoxicity involves lysosomal damage

The cytotoxic effects of alloxan are not understood in any great detail, although they are considered to involve reactions mediated by oxygen-derived free radicals. These reactive species may form extra-or intracellularly following alloxan reduction, and result in cell damage through a number of complex interactions with a variety of macromolecules. The purpose of the present study was to elucidate further the early intracellular effects of alloxan on a model system of macrophage-like cells in culture. Addition of alloxan (15 mM), without reducing agents, to the medium surrounding the cells (phosphate-buffered saline, PBS, 37 degrees C, pH 7.4) resulted in rapid lysosomal damage (disappearance of the proton gradient over the membrane) followed by severe cellular degeneration (swelling and blebbing) and 50% cell death (trypan blue dye exclusion test) within fifty min. Cells pretreated with the gamma-glutamyl cysteine synthetase-inhibiting agent BSO, to decrease levels of intracellular glutathione, showed enhanced sensitivity to alloxan. The results are interpreted as indicating the cytotoxicity to result from intracellular formation of superoxide radicals, hydrogen peroxide and hydroxyl radicals, the latter within secondary lysosomes containing trace amounts of reactive iron (inducing Fenton reactions). The ensuing lysosomal membrane damage may result in leakage of lysosomal hydrolases and further cellular degeneration.

Extracellular reduction of alloxan results in oxygen radical-mediated attack on plasma and lysosomal membranes

Alloxan participation in extracellular redox processes results in the formation of the reactive oxygen species (ROS) superoxide anions (O2-), hydroxyl radical (OH.) and hydrogen peroxide (H2O2), causing cell damage through a number of complex interactions probably involving several different cellular structures. These involve the plasma membrane, and we have recently presented evidence for lysosomal interference. The present study elucidates the early (within 15 min) events in a model system of macrophage-like cells (J-774) in culture. Addition of 2 mM alloxan and 1 mM cysteine to the medium surrounding the cells (phosphate-buffered saline, PBS, 37 degrees C, pH 7.4) resulted in rapid lysosomal membrane damage with disappearance of the proton gradient as visualized by acridine orange relocalization, as well as plasma membrane alterations leading to increased leakage of fluorescein after fluorescein diacetate staining. These events were later (greater than 30 min) followed by cellular degeneration in the form of blebbing. Mitochondrial damage (rhodamine 123 relocalization) was a late event. Cells pretreated with desferrioxamine (Des) and superoxide dismutase (SOD) or Des, SOD and catalase (CAT) to induce partial (H2O2 formation only) or almost full protection (no ROS formation) showed about the same reactions as when cells were exposed to alloxan and cysteine without scavengers (O2-, H2O2 and OH. formation) or with PBS only, respectively. The results are interpreted as indicating that the cytotoxicity is a consequence mainly of H2O2 involvement and probably of lysosomal influx of H2O2 with ensuing OH.formation within secondary lysosomes containing trace amounts of reactive iron. It is suggested that the resultant lysosomal membrane damage is followed by leakage of lysosomal hydrolases and ensuing cellular degeneration.

Effects of alloxan and reducing agents on macrophages in culture

The diabetogenic effect of the quinonoid compound alloxan is not understood in detail although it supposedly involves reactions mediated by alloxan and oxygen radicals. These reactive species may form extra- or intracellularly and cause cell damage through a variety of complex interactions with several macromolecules. The purpose of this study was to elucidate early (less than or equal to 60 min) effects of alloxan and reducing agents (cysteine and ascorbic acid) on cultured macrophages, as assayed by the trypan blue dye exclusion test and the sensitive fluorescein diacetate and propidium iodide (FDA/PI) double staining technique. During the reactions between alloxan and reducing agents, oxygen was consumed as a sign of superoxide anion radical formation. When alloxan alone was added to two different culture media without serum, oxygen was still consumed, indicating formation of oxygen radicals due to the occurrence of reducing substances in cell culture media. This finding demonstrated the necessity of performing further studies in solutions without reducing capacity, e.g. in phosphate-buffered saline. The experiments showed that exposure of normal and malignant macrophages to alloxan and reducing substances resulted in rapidly occurring plasma membrane damage and ensuing cell death. Separate addition of catalase, desferrioxamine or superoxide dismutase resulted in evident, slight and no protection, respectively. The combinations of (i) catalase and desferrioxamine, and (ii) catalase, desferrioxamine and superoxide dismutase, however, inhibited cell damage in a pronounced and complete way, respectively. The results are interpreted as indicating cell damage due to the extracellular formation of hydrogen peroxide and hydroxyl radicals. The latter in close proximity to the cells and acting on the plasma membrane, while the former, after diffusing into the cell, may have several intracellular targets. The FDA/PI technique proved its value as a quantifiable method for the evaluation not only of cell death but also of cell damage with computer-based fluorometry.

Biochemical basis for the specificity of alloxan inactivation of calmodulin-dependent protein kinase II

The specificity and biochemical basis of inactivation of calmodulin-dependent protein kinase II by alloxan was studied in dispersed rat brain cells and a partially purified kinase preparation from an insulin-secreting tumor-cell line, RINm5f. When mechanically dispersed rat brain cells were incubated with [32P]-phosphate to label endogenous ATP, depolarization with 44 mM KCl produced a significant (P = 0.03) increase in phosphorylation of endogenous synapsin (132 +/- 8% of basal). Pre-treatment of the brain cells with 1.5 mM alloxan reduced depolarization-sensitive synapsin phosphorylation (109 +/- 5%). Phosphopeptide mapping of depolarization-phosphorylated synapsin showed that alloxan pre-treatment reduced phosphorylation specifically at synapsin sites phosphorylated by calmodulin-dependent protein kinase II. The results demonstrate selective inactivation of calmodulin-dependent protein kinase II activity by alloxan in an intact cell system, which may be useful in the study of the Type II kinase in cells and tissues. Using a partially purified kinase preparation from RINm5f cells, alloxan (100 microM) inactivated 76 +/- 1% calmodulin-dependent protein kinase II activity in 5 min at 37 degrees C. Subsequent incubation with dithiothreitol restored most of the activity. 5,5'-Dithiobis (2-nitrobenzoic acid) (I50 = 2.5 microM) also inactivated the kinase. These results suggested that a sulfhydryl group was involved at the inactivation site. Iodoacetamide (1.0 mM) had no inhibitory effect; however, preincubation with iodoacetamide protected the kinase activity from subsequent inactivation by alloxan. Covalent binding of [14C]-alloxan to calmodulin-dependent protein kinase was demonstrated.(ABSTRACT TRUNCATED AT 250 WORDS)

Studies on lipid peroxidation in pancreatic tissue. In vitro formation of thiobarbituric-acid-reactive substances (TBRS)

In pathogenic studies on acute pancreatitis the importance of a temporary ischemia on induction of autodigestion was demonstrated. Because of the involvement of oxygen-derived free radicals in the ischemia/reperfusion injury of other tissues we have investigated the influence of artificial oxidants, as FeCl3 and H2O2, on pancreatic tissue and isolated pancreatic acinar cells. Lipid peroxidation was determined as thiobarbituric-acid-reactive substances (TBRS). In these experiments the TBRS concentration was elevated within the first min of incubation with FeCl3. The exposure of pancreas homogenates and intact acinar cells to H2O2 had no remarkable effect on formation of TBRS. Under this condition the survival of cells was strongly reduced, while cells exposed to FeCl3 revealed a remarkably slower rate of cytolysis. The missing correlation between cell lysis and elevation of TBRS suggests that lipid peroxidation might not be essential process in pancreatic acinar cell damage.

Phagocyte oxidative metabolism in cyclosporine- or placebo-treated patients with insulin-dependent (type I) diabetes mellitus of recent onset

Several lines of evidence suggest that phagocyte-mediated oxidative processes are involved in damage to pancreatic islet cells of Type I insulin-dependent diabetes mellitus (IDDM). This hypothesis, however, has not yet been explored at the clinical onset of IDDM. Similarly, the possibility that cyclosporine A (Cy-A) might exert a selective influence on these phagocyte-mediated oxidative reactions has also not yet been investigated as compared to a placebo. The present study tested both hypotheses in 32 patients with recently diagnosed IDDM who were part of the recent French multicenter randomized therapeutic trial of Cy-A. The production of reactive oxygen intermediates (ROI) by circulating polymorphonuclear (PMN) and mononuclear (MN) phagocytes was determined by luminol-dependent chemiluminescence (CL), both directly within microamounts of whole blood and in purified PMN or MN phagocyte suspensions. Lastly, CL production was measured in the absence (resting CL) and the presence of a panel of particular and soluble phagocyte membrane-stimulating agents. We found that on entry into the trial, i.e. within less than 2 months of the clinical onset of IDDM, patients had normal whole blood CL production in the absence of a stimulating agent and upon phagocytic challenge with latex or opsonised zymosan particles. By contrast, whole blood CL responses to soluble stimuli such as phorbol myristate acetate (PMA), concanavalin A (Con-A) and F Met-Leu-Phe (FMLP) were significantly higher than in the control group of 52 normal subjects (P less than 0.01). In purified PMN and MN phagocyte suspensions, both resting and stimulated CL productions were normal, regardless of the type of stimulating agent. After 3 months of treatment, whole blood CL responses to Con-A and FMLP returned to almost normal levels in patients treated with Cy-A (15 cases) but not in those receiving the placebo (17 cases); PMA-induced CL responses were also decreased, but this was found in both groups of patients. In purified phagocyte suspensions we detected no effect of Cy-A on PMN, whereas MN phagocytes from Cy-A-treated patients showed reduced CL responses to FMLP but not to other stimuli. Altogether, these results demonstrate for the first time that the capacity of circulating PMN and MN phagocytes to generate ROI is normal at the clinical onset of IDDM and suggest that circulating substances increase oxidative responses to soluble, but not particulate, stimuli.(ABSTRACT TRUNCATED AT 400 WORDS)

Alloxan-induced alterations in composition and dynamics of red blood cell membranes. I. Effect of alloxan on intact red blood cells and isolated erythrocyte membranes

Changes of dynamics and chemical composition in membranes of intact red blood cells and isolated erythrocyte membranes treated with alloxan were investigated in order to assess whether alloxan-induced generation of active forms of oxygen may be critical for erythrocyte destroying. In vitro incubation of native red blood cells or prepared erythrocyte membrane ghosts with various concentrations of alloxan gave rise both to levels of membrane TBA-reacting substance and lipid membrane microviscosity both in the deeper and surface regions of lipid bilayer, as evidenced by fluorescence polarization technique. The amount of membrane phospholipid decreased upon alloxan action and that of membrane cholesterol remained rather unchangeable, thus resulting in significant elevation of membrane cholesterol:phospholipid (C:PL) ratio. Both time course and concentration effect of alloxan were found to change exponentially with the different rates of the reaction. There was a linear correlation between 1,6-diphenylhexatriene-1,3,5 (DPH) and 1-anilinonaphthalene-8-sulfonate (ANS) anisotropy coefficients and C:PL ratio (respectively r = 0.697 and r = 0.580) as well as TBARS levels (r = 0.386 for rDPH and r = 0.324 for rANS), thus implying the possible effect of membrane dialdehydes on bilayer components immobilization. Regression coefficients significance testing showed reaction rates of TBARS and C:PL changes to be significantly parallel, contrary to those of fluorescence anisotropy coefficients assessing considerably slower dynamics of alloxan-induced changes. The relevance of changes induced by alloxan in isolated erythrocyte ghosts and intact red blood cells and the compatibility of the present results with several previous studies support the widespreading idea pointing the cell membrane as a main target of damage during alloxan action.

Involvement of O2 radicals in 'autoimmune' diabetes

Spontaneous diabetes in the non-obese diabetic (NOD) mice is a CD4 T cell-dependent process. We have suggested that specific beta cell destruction results from free radical production at the site of islet inflammation; oxygen radicals are produced by activated inflammatory cells. We reported here that in vivo treatment of spontaneously diabetic NOD mice with the enzyme superoxide dismutase (2000 U for seven injections) and catalase (40,000 U for seven injections) protects islet tissue from disease recurrence following transplantation into spontaneously diabetic mice. Similar results were obtained when animals were treated with either enzyme alone. This effect was dose-dependent and little protection was observed when the dose of enzyme was reduced four-fold. These results indicate that oxygen metabolites, specially superoxide and hydrogen peroxide, are directly involved in the pathogenesis of immunology mediated diabetes.

Iron release from ferritin by alloxan radical

Alloxan in the presence of reduced glutathione released iron from ferritin which is the major intracellular iron storage protein. Superoxide dismutase inhibited by only about 30% the alloxan-dependent iron release from ferritin but completely inhibited the iron release from ferritin induced by hypoxanthine-xanthine oxidase. Under anaerobic conditions, the ESR spectrum of alloxan radical was obtained and interaction with ferritin resulted in a marked diminution of the alloxan radical signal. These results indicate that alloxan radical rapidly releases iron from ferritin.

Free radicals and diabetes

The role of active oxygen species in diabetes is discussed in this review. Type I diabetes is caused by destruction of the pancreatic beta cells responsible for producing insulin. In humans, the diabetogenic process appears to be caused by immune destruction of the beta cells; part of this process is apparently mediated by white cell production of active oxygen species. Diabetes can be produced in animals by the drugs alloxan and streptozotocin; the mechanism of action of these two drugs is different, but both result in the production of active oxygen species. Scavengers of oxygen radicals are effective in preventing diabetes in these animal models. Not only are oxygen radicals involved in the cause of diabetes, they also appear to play a role in some of the complications seen in long-term treatment of diabetes. Changes in antioxidants in the diabetic state and their consequences are discussed.

Peroxidative membrane damage in human erythrocytes induced by a concerted action of iodoacetate, vanadate and ferricyanide

Human erythrocytes incubated without substrate in the presence of iodoacetate (0.2 mM), vanadate (0.5 mM) and ferricyanide (5 mM) form aqueous membrane leaks of equivalent radii of 0.5-0.8 nm leading to complete colloid-osmotic lysis within 180 min. All three components are indispensable for the effect. Inosine but not glucose markedly enhances the rate of hemolysis. These effects are due to oxidative damage, as indicated by concomitant destruction of polyunsaturated fatty acids and suppression of both effects by radical scavengers. Hemoglobin is not oxidized under these conditions. GSH and membrane SH levels remain almost normal, and no crosslinking or irreversible aggregation of membrane proteins is observed. In the absence of O2 no membrane damage can be observed. It is proposed that radical formation originates from reduction of O2 by NADPH, analogous to processes described in microsomal membranes. NADH seems not to be involved, since leak formation occurs in spite of the blockage of NADH formation by iodoacetate. Vanadate and ferricyanide are probably required to amplify the peroxidative reaction sufficiently to overcome the cellular antioxidative capacity.

Free radical-mediated membrane depolarization in renal and cardiac cells

Cell membrane potential was measured with a flow cytometer by quantitating the intracellular accumulation of a fluorescent cationic carbocyanine dye. We used this system to demonstrate depolarization upon the addition of hydrogen peroxide (10-1,000 microM) and ferrous chloride (25-100 microM) to cultures of either neonatal rat myocardial or LLC-PK1 renal epithelial cells. Ferrous chloride-induced depolarization was prevented by superoxide dismutase, catalase and dimethyl sulfoxide, suggesting roles for the superoxide anion, hydrogen peroxide and the hydroxyl radical in effecting this depolarization, possibly through a Fenton-type reaction mechanism. Supplementation of either cell type with 2 microM tocopherol acid succinate during growth in tissue culture, prior to exposure to the oxidizing agent, decreased the magnitude of the depolarization in both cell types. The results are consistent with a role for tocopherols in scavenging free radical species responsible for the depolarization of the cell membrane.