Oxygen Club
of California


2000 World Congress

Oxidants and Antioxidants in Biology

March 1-4, 2000

Fess Parker's Double Tree Resort
Santa Barbara, California


Book of Abstracts


Lester Packer
Department of Molecular and Cell Biology
University of California, Berkeley, CA, USA

Enrique Cadenas
Department of Molecular Pharmacology & Toxicology
School of Pharmacy, University of Southern California, Los Angeles, CA, USA

Kelvin J. A. Davies
School of Gerontology
University of Southern California, Los Angeles, CA, USA

Catherine Rice-Evans
Wolfson Centre for Age-Related Diseases, Guy's, King's and St Thomas's School of Biomedical Sciences, King's College London, United Kingdom

Chandan K. Sen
Environmental Energy Technologies, Lawrence Berkeley National Laboratory
University of California, Berkeley, CA, USA

Helmut Sies
Institut für Physiologische Chemie I, Heinrich-Heine-Universität
Düsseldorf, Germany


Keynote Lecture 5
Session I. Nitric Oxide
Session II. Peroxynitrite 19
Session III. Polyphenols and Flavonoids 25
Session IV. Redox Regulation of Cell Signaling 37
Session V. Vitamin E and Lipoic Acid in Health and Disease 47
Session VII. Skin Disorders and Aging 67
Science and Humanity Prize 81
Nutraceutical Award 83
Session VIII. Oxidative Stress, Inflammation, and Neurodegeneration 85
Posters 103

Author Index 239

Keynote Lecture

Oxidants, antioxidants, and the modulation of cell death

Sten Orrenius

Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden

Among the multiple factors required for cell survival are molecular oxygen and various antioxidants in the right proportions to maintain a delicate intracellular redox balance. A sustained perturbation of the latter may result in either hypoxic or oxidative cell injury and death. Lethal oxidative stress may cause apoptotic or necrotic cell death, depending on the severity of the insult. Our early studies of menadione-treated hepatocytes established some critical events involved in oxidative necrotic cell death (1,2).
In the current paradigm for apoptotic cell death, the activity of a family of caspases (cysteine proteases with a stringent requirement for Asp in the P1 position of the substrate) orchestrates the multiple downstream events that comprise apoptosis. In TNF- and CD95-mediated apoptosis the proteolytic cascade is believed to be triggered directly by pro-caspase-8 binding to the activated plasma membrane receptor complex. In other forms of apoptosis the mechanisms of activation of the proteolytic cascade are less well established but may involve other proteases, e.g. granzyme B, calpain, or factors released from the mitochondria. Recently, the possibility that cytochrome c released from the mitochondria may serve to activate dormant caspases in the cytosol, and thereby to amplify the apoptotic process, has attracted considerable attention.
One of the most reproducible inducers of apoptosis is mild oxidative stress. Although it is yet unclear exactly how an oxidative stimulus can activate the caspase cascade, there is emerging evidence that this may occur by either upregulation of the CD95 system or stimulation of cytochrome c release from the mitochondria (3). Conversely, a variety of antioxidants are known to inhibit the death process (4). Further, oxidative modification of proteins and lipids have been observed also in cells undergoing apoptosis in response to non-oxidative stimuli, suggesting that intracellular oxidation may be a general feature of the effector phase of apoptosis. It now appears that both O2_-generation caused by auto-oxidation of reduced components of the respiratory chain following the depletion of cytochrome c and caspase activation of a glutathione (GSH) efflux mechanism may contribute to this shift in the intracellular redox state during apoptosis (5).
Although oxidative stress can induce apoptosis, this is slow to develop and usually occurs some time after the original source of the oxidative stress has decayed (6). During the acute phase of oxidative stress, cells appear to be partially resistant to a secondary induction of apoptosis. There is accumulating evidence that this inhibition operates via redox inactivation of caspases, and that caspase activity may be modulated by fluctuations in intracellular redox state. Our recent studies with disulfiram (7) and hydrogen peroxide demonstrate the toxicological significance of this mechanism which, however, may also apply to modulation of caspase activity by nitric oxide or alterations in the glutathione redox state under physiological conditions.
Finally, another mechanism by which oxidative stress can modulate cell death is by induction of heat shock proteins (Hsps). Thus, increased expression of Hsp27 and Hsp72 in thermotolerant T-lymphocytes protects the cells from apoptosis induced by several triggers (8). Whereas Hsp27 appears to exert its antiapoptotic effect by preventing the release of mitochondrial cytochrome c, Hsp72 interferes with the subsequent apoptosomal activation of caspases.
From these and other studies it is clear that oxidants and antioxidants can not only determine cell fate but can also modulate the mode of cell death. This modulation may be mediated by effects of oxidative stress on components of the apoptotic machinery.

1. Thor, H., Smith, M.T., Hartzell, P., Bellomo, G., Jewell, S. and Orrenius, S. The metabolism of menadione (2-methyl-1,4-naphthoquinone) by isolated hepatocytes. A study of the implications of oxidative stress in intact cells. J. Biol. Chem., 257, 12419-12425 (1982)
2. Jewell, S., Bellomo, G., Thor, H., Orrenius, S. and Smith, M.T. Bleb formation in hepatocytes during drug metabolism is caused by disturbances in thiol and calcium ion homeostasis. Science, 217, 1257-1259 (1982)
3. Hampton, M.B. and Orrenius, S.: Dual regulation of caspase activity by hydrogen peroxide: Implications for apoptosis FEBS Lett., 414, 552-556 (1997)
4. Slater, A.F.G., Nobel, S., Maellaro, E., Bustamante, J., Kimland, M. and Orrenius, S.: Nitrone spin traps and a nitroxide antioxidant inhibit a common pathway of thymocyte apoptosis. Biochem. J. 306, 771-778 (1995)
5. van den Dobbelsteen, D.J., Nobel, C.S.I., Schlegel, J., Cotgreave, I.A., Orrenius, S., and Slater, A.F.G. Rapid and specific efflux of reduced glutathione during apoptosis induced by anti-Fas/APO-1 antibody. J. Biol. Chem., 271, 15420-15427 (1996)
6. Nobel, C.S., Kimland, M., Lind, B., Orrenius, S. and Slater, A.F.G. Dithiocarbamates induce apoptosis in thymocytes by raising the intracellular level of redox-active copper. J. Biol. Chem., 270, 26202-26208 (1995)
7. Nobel, C.S.I., Kimland, M., Nicholson, D.W., Orrenius, S. and Slater, A.F.G.: Disulfiram is a potent inhibitor of proteases of the caspase family. Chem. Res. Toxicol., 10, 1319-1324 (1997)
8. Samali, A., Holmberg, C.I., Sistonen, L. and Orrenius, S.Thermotolerance and cell death are distinct cellular responses to stress: dependence on heat shock proteins. FEBS Lett., 461, 306-310 (1999)

Session I
Nitric Oxide

Nitric oxide and metal metabolism:
An examination using the yeast model system

Jon M. Fukuto, Ken Chiang, Masaru Shinyashiki and Chris Switzer

Department of Pharmacology, School of Medicine, University of California,
Los Angeles, CA, USA

The uptake, utilization and sequestration of metals such as iron and copper are important aspects of normal cell growth and viability. That is, the transport, sequestration and utilization of metals is required to perform a number of important biochemical processes. However, in spite of their chemical and biochemical utility, biological metals can be toxic if they are not handled properly by cells. Thus maintaining metal homeostasis is important from toxicological, physiological and pathophysiological perspectives. Alterations or disruptions in metal homeostasis can lead to cellular degeneration and death. Metal metabolism is well defined in the yeast Saccharomyces cerevisiae model system. Moreover, significant homology exists between the metal metabolism systems of humans and yeast. Using the yeast model system, we find that NO is capable of dramatically altering/disrupting metal homeostasis in yeast depending on NO levels and cellular conditions. Based on these studies, and those of others, it is becoming increasingly clear that NO plays a significant role in maintaining metal homeostasis under normal physiological circumstances. Furthermore, we suspect that NO can disrupt metal homeostasis leading to pathophysiological states.

The perivascular dynamics of nitric oxide and oxygen

Jack R. Lancaster, Jr.

Department of Physiology, LSU Health Sciences Center New Orleans,
New Orleans, LA 70112

Endothelial nitric oxide (nitrogen monoxide) is synthesized at the intravascular/extravascular interface. We have previously reported the intravascular half-life of NO, as a result of consumption by erythrocytes, as approximately 2 milliseconds (J. Biol. Chem. 273:18709(98)). We report here studies of the lifetime of NO in the parenchymal (extravascular) tissue and describe the implications of these results for the distribution of NO and oxygen concentration gradients away from the blood vessel.
The consumption of NO by parenchymal cells (hepatocytes) is linearly dependent on both NO and O2 concentration. We estimate that the extravascular half-life of NO will range from 0.05 to > 2 seconds, depending on O2 concentration and thus distance from the vessel. Computer modeling reveals that this phenomenon, coupled with reversible NO inhibition of cellular mitochondrial oxygen consumption, substantially extends the zone of tissue cellular oxygenation away from the blood vessel, with an especially dramatic effect during conditions of increased tissue work (oxygen consumption). This represents a second action of NO, in addition to vasodilation, in enhancing tissue cellular respiration and provides a possible physiological function for the known reversible inhibition of mitochondrial respiration by low concentrations of NO.

Reactions of nitric oxide with zinc finger structures:
Impact on gene-regulation by nitrosative stress

K.-D. Kröncke1, V. Kolb-Bachofen1, D. Berendji1 & C. Carlberg2

1Research Group Immunobiology, and 2Institute of Physiological Chemistry, Medical Department, Heinrich-Heine-University of Düsseldorf, Germany

We have shown that NO induces Zn2+ release from the zinc-st oring protein metallothionein in vitro but also in vivo in the nuclei and cytoplasm of cells. This suggests that proteins containing zinc finger structures represent primary targets of NO-induced stress. To investigate this, we used the model system lymphocytic IL-1? dependent IL-2 gene expression, which is dominantly regulated by the zinc finger transcription factor Sp1, and by the non-zinc finger transcription factor NFAT. NO, applied as subtoxic concentrations of the NO donor SNOC, inhibited IL-1?-dependent IL-2 mRNA expression in a concentration-dependent and reversible manner. In contrast, subtoxic concentrations of H2O2 slightly enhanced IL-2 mRNA expression. NO-dependent inhibition of IL-2 mRNA expression correlated with inhibition of Sp1 DNA binding activity, while NFAT was not affected. Next, we studied the effects of NO towards the specific interaction of the heterodimeric complex of two zinc finger transcription factors, the 1a,25-dihydroxyvitamin D3 receptor (VDR) and the retinoid X receptor (RXR), with vitamin D3 response elements (VDRE). NO caused a dose-dependent inhibition of VDR-RXR-VDRE complex formation, which in vitro was found to be reversible. Functional assays with transiently transfected cells indicated that in vivo NO indeed acts as a repressor of vitamin D3 signaling. These findings suggest that NO has a modulatory role on transcription factor activities. As zinc finger transcription factors are particularly sensitive towards NO, they appear to represent molecular targets for altered gene expression mediated by NO.

Role of endothelial NOS in modulating inflammatory tissue damage

Matthew B. Grisham, David Lefer, Steven Jones, Stephen Laroux
and Paul Huang*

Department of Molecular and Cellular Physiology, LSU Health Sciences Center, Shreverport, LA. 71130 and *Massachusetts General Hospital,
Cambridge, MA. 02115

The recruitment of leukocytes from the circulation to the venular endothelial cell surface and ultimately into the tissue constitutes the fundamental mechanism for the inflammatory response and tissue injury. Although exogenous NO donors have been shown to attenuate leukocyte-endothelial cell interactions in vitro and in vivo, little is known regarding the role that endogenous NO may play in modulating the inflammatory response and tissue injury in vivo. Thus, the objectives of this study were to assess the role of eNOS in modulating endothelial cell adhesion molecule expression, leukocyte adhesion and extravasation as well as post-ischemic myocardial injury in wild type (wt; C57Bl/6 x 129) vs eNOS-deficient (eNOS-/-) mice in vivo. Using the dual radiolabel monoclonal antibody technique to quantify ECAM expression in vivo (Am J. Path 151:205, 1997), we observed significant increases in constitutive expression of ICAM-1 (1.3-1.6-fold) and P-selectin (4-6-fold) in the vasculature of a variety of different tissues including intestine, mesentery and heart from eNOS-/- mice compared to their wt controls. These data correlated well with large and significant increases in basal leukocyte rolling (68
10 vs 12 2 cells/min for eNOS-/ vs wt, respectively; p < 0.05) and adhesion (14 4 vs 2 1 cells/ 100mm length for eNOS-/- vs wt, respectively; p < 0.05) in the post-capillary venules of eNOS-/- mice compared to their wt controls using intravital microscopy. Furthermore, the recruitment of leukocytes (i.e., PMNs) out of the circulation and into the peritoneum 6 hrs following a single injection (i.p.) of oyster glycogen, was significantly enhanced in eNOS-/- compared to wt mice (16 2 vs 8 1 x106 PMNs/mouse for eNOS-/- vs wt, respectively; p < 0.05). Finally, 20 min of coronary artery occlusion followed by 120 min of reperfusion in eNOS-/- mice enhanced dramatically leukocyte infiltration (29.5 2.5 vs 5.0 0.9 PMNs/field for eNOS-/- vs wt, respectively; p < 0.05) and infarct size (46 3.8% vs 20.9 2.9% of the ischemic zone for eNOS-/- vs wt, respectively; p < 0.01) compared to their wt controls. These data suggest that eNOS-derived NO plays an important role in regulating ECAM expression, leukocyte adhesion and extravasation as well as providing a cardioprotective effect in the post-ischemic heart (Supported by DK47663).

NO synthesis and the arginine paradox: evolution of a
specific thiol in mammalian argininosuccinate synthetases that bestows interim regulation of substrate recycling by NO

L. Xie, G. Ho and S.S. Gross

Weill Medical College of Cornell University, NY, NY 10021, USA

NO-producing cells regenerate substrate L-arginine (Arg) from the NO synthase co-product, L-citrulline (Cit). This gives rise to an Arg/Cit cycle that sustains high-output NO production even in circumstances where Arg is unavailable. The Arg/Cit cycle comprises three enzymes whose expression is co-induced by immunostim uli: the inducible isoform of NO synthase (iNOS), argininosuccin ate synthase (AS) and argininosuccinate lyase (AL). Evidence suggests that Arg regeneration form Cit provides the preferred source of substrate for iNOS. This knowledge may explain the arginine paradox, i.e. the finding that NO synthesis rate can be limited by substrate availability, even when Arg is present at 10-100 fold excess over its Km for NOS. We now reveal that thiol depletion of vascular smooth muscle cells specifically reduces the ability of Cit to serve as indirect source of substrate for NO synthesis. This immediately suggested that AS or AL possesses one or more redox sensitive thiols. In vitro studies confirmed that recombinant huAS is reversibly inactivated by NO-donors, via nitrosylation of a specific Cys residue that is protected in the ATP-bound enzyme. A different site on AS is additionally found to mediate inactivation by peroxide; this latter site is protected by bound citrulline. Identity of the Cys residue targeted by S-nitrosylation was specified by a novel mass spectrometric technique. Mutagenesis of this NO-sensitive Cys residue to Ala resulted in neither a loss of enzyme activity, nor a change in Km for substrates, however, inactivation by NO required higher concentrations of NO-donors. In contrast, inactivation of mutant AS by peroxide was unchanged from wild type AS. Sequence alignments reveal that the preferred Cys target of NO is present in all mammalian ASs, but absent from ASs of species apparently devoid of NOS. We propose that inactivation of AS by NO evolved to prevent toxic NO overproduction, by limiting the access of NOS to Arg substrate.

Nitric oxide and mitochondria:
modulation of cytochrome c release

Victor M. Darley-Usmar, Paul S. Brookes,
Peter G. Anderson, Bruce Freeman, Jason Eiserich,
Kenta Darley-Usmar, Doug Moellering & Rakesh Patel

Department of Pathology, University of Alabama at Birmingham,
Birmingham, AL 35294

Opening of the mitochondrial permeability transition pore (PTP) and associated release of cytochrome c are important factors in apoptosis. Conflicting data exist regarding the effects of nitric oxide (NO·) on apoptosis and the PTP. To resolve these issues Ca2+ induced PTP, a process implicated in ischemia/reperfusion injury, was modeled by studying Ca2+ induced swelling of isolated rat liver mitochondria. The effects of NO· released from NONOate compounds on this process were determined. NO· reversibly inhibited mitochondrial swelling through a mechanism dependent on the concentration and duration of NO· exposure. The minimum NO· release rate that inhibited swelling was 5nM/sec, with an IC50 of 25 nM/sec, which is within the range reported for maximal stimulation of iNOS and nNOS.
Supernatants from these experiments were assayed for released cytochrome c by western blotting and mass spectrometry. Cytochrome c was present in supernatants of swollen mitochondria, but undetectable in those treated with NO· donors or cyclosporin A (an inhibitor of the PTP). Our results suggest that inhibition of PTP and cytochrome c release are possible anti-apoptotic mechanisms of NO·.

Session II

Generation and decomposition of peroxynitrite at
physiological pH: Implications for biological activity

Sara Goldstein1, Gidon Czapski1, Johan Lind2 and Gabor Merenyi2

1The Hebrew University of Jerusalem, Israel, and
2The Royal Institute of Technology, Stockholm, Sweden

The reaction between ONOO_ and CO2 is relatively fast, and is most probably the main process that peroxynitrite undergoes under physiological conditions. It has been shown that this reaction proceeds via the formation of 33% NO2 and CO3._ in the bulk of the solution. The fast reaction between NO and O2._ is currently accepted as the main biological source of peroxynitrite. Therefore, it is of biological interest to compare the reactivity of authentic peroxynitrite to that of continuous generation of NO and O2._, which will be done for the nitration of tyrosine and oxidation of NADH.
The nitration of tyrosine occurs via the reaction of tyrosyl radical with NO2, and therefore the nitration yield obtained with equal fluxes of NO and O2._ is considerably lower than that obtained with authentic peroxynitrite, it decreases with decreasing the flux of the radicals, and is completely inhibited upon excess production of O2._ or NO. The oxidation of NADH by peroxynitrite takes place via the oxidation of NADH by CO3._ to NAD., which in turn reacts with O2 to form NAD+ and O2._. The oxidation of NADH by O2._ to yield NAD. is relatively slow, but becomes important under continuous generation of NO and O2._, where it competes efficiently with the dismutation of O2._. Under these conditions, the oxidation of NADH takes place via a chain reaction, and the yield of NAD+ exceeds 33% and increases upon decreasing the flux of the radicals. In conclusion, the reactivity of authentic peroxynitrite is not necessarily similar to that of continuous generation of NO and O2._, and peroxynitrite will play a minor role in almost any radical-radical process.

Peroxynitrite reactivities and biological implications

Giuseppe L. Squadrito and William A. Pryor

Biodynamics Institute, Louisiana State University, Baton Rouge, Louisiana

The mechanism for the decomposition of peroxynitrite and for the reactions of peroxynitrite with biological target molecules has received much attention during the past decade. Rapid development in this field of research has lead to divergent opinions; however, it is now possible to propose reaction mechanisms that are consistent with most reported data. In the absence of CO2 and at physiological pH, peroxynitrite decomposes via the homolytic dissociation of HO-ONO. However, the free radicals HOÄ and ÄNO2 are formed in relatively low yields because cage recombination yields nitrate (and a proton) and predominates over diffusive separation that yields the free radicals. The self-decomposition of peroxynitrite has no biological importance because this reaction is too slow compared to other reactions of peroxynitrite, for example, the reaction of peroxynitrite with CO2. The reaction of peroxynitrite with CO2 is specially important in biological systems due to the high concentration of CO2 and the relatively large rate constant for this reaction. Moreover, the reaction of peroxynitrite with CO2 yields the free radicals CO3Ä_ and ÄNO2 and the toxicity of peroxynitrite must be interpreted as arising in part from the reactions of CO3Ä_ and ÄNO2. The use of scavengers specific for these free radicals may offer a viable approach towards ameliorating the adverse effects of peroxy nitrite. The reaction of peroxynitrite with CO2 can affect the results in studies of peroxynitrite because working solutions are almost invariably contaminated with atmospheric CO2. Adventitious concentrations of CO2 can modulate the reactivity of peroxynitrite when using low (submillimolar) concentrations of peroxynitrite and results must be interpreted with care.

DNA damage by peroxynitrite

Steven R. Tannenbaum, Jacquin Niles and Natalia Tretyakova

Division of Bioengineering and Environmental Health and Department of
Chemistry, Massachusetts Institute of Technology, Cambridge, MA

We have been investigating the chemical interactions of peroxy nitrite with and without bicarbonate on DNA, oligonucleotides and nucleosides. Parallel work on plasmids, in collaboration with Gerald Wogan, has led to a generalized picture of the relationship between DNA damage and mutagenesis in the supF gene. Peroxyni trite is highly specific for guanine bases, yielding products that are the result of multiple one-electron oxidations plus interactions of peroxynitrite as a nucleophile. Several major products accumulate in DNA, some of which are repaired by known DNA repair enzymes. A major target for reaction is 8-oxoguanine, but all oxidation products are thermodynamically downhill. I will give an overview of these reactions and their possible consequences.

Defense against peroxynitrite:
Selenoprotein P binding to heparin

Helmut Sies and Gavin Arteel

Institut für Physiologische Chemie I, Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany

The concerted generation of nitric oxide and superoxide by inflammatory cells leads to peroxynitrite which is utilized for inactivation of bacteria, viruses and other invading microorganisms. The organism needs defense against peroxynitrite to protect its own cells. We have identified selenium-containing compounds as effective molecules for intercepting peroxynitrite. The present work addresses the question whether selenoprotein P, a protein containing 8-10 selenocysteines, might be useful in human blood plasma for such purposes. The free concentration of selenoprotein P in plasma is low, about 50 nM. We characterized the binding of seleno protein P to heparin, as a model for cellular glycosaminoglycans. The isolated protein was studied for binding properties to heparin using surface plasmon resonance techniques. Heparin contains two binding sites for selenoprotein P, a high-affinity, low-capacity site (Kd 1 nM) and a low-affinity high-capacity site (Kd 140 nM). The high-affinity site is abolished by histidine carbetho-xylation with diethylpyrocarbonate. It is concluded on the basis of these model studies that binding to proteoglycans on the vascular endothelium makes selenoprotein P suitable for the formation of a protective barrier against oxidants.

Supported by Deutsche Forschungsgemeinschaft

Session III
Polyphenols and Flavonoids

Peroxynitrite protection by cocoa polyphenol oligomers

Gavin E. Arteel*, Peter Schröder, and Helmut Sies

Institut für Physiologische Chemie I, Heinrich-Heine-Universität Düsseldorf,
Düsseldorf, Germany

Peroxynitrite is a mediator molecule in inflammation, and its biological properties are being studied extensively. Flavonoids, natural plant constituents, protect against peroxynitrite and could thereby play an antiinflammatory role. Procyanidin oligomers of different size (monomer through nonamer), isolated from the seeds of Theobroma cacao, were recently examined for their ability to protect against peroxynitrite-dependent oxidation of dihydrorho damine 123 and nitration of tyrosine (FEBS Lett.
462, 167-170, 1999) and were found to be effective in attenuating these reactions. The tetramer was particularly efficient at protecting against oxidation and nitration reactions. Epicatechin oligomers found in cocoa powder and chocolate may be a potent dietary source for defense against peroxynitrite.

*Current address: Laboratory of Hepatobiology and Toxicology, Department of Pharmacology, the University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

Licorice derived glabridin is a potent antioxidant
against LDL oxidation

Michael Aviram, Paula Belinki, Bianca Fuhrman, Mira Rosenblat, and Jacob Vaya

The Lipid Research Laboratory, Technion Faculty of Medicine,
Haifa, Israel

Extract of the Glycyrrhiza glabra (licorice) root is a potent antioxidant both in vitro and in vivo, against low density lipoprotein (LDL) oxidation. Licorice root extract inhibited the oxidation of LDL, induced by 2,2'-azobis (2-amidinopropane) dihydrochloride (AAPH), 2,2'-azobis (2,4-dimethyl) valeronitrile (AMVN) or by copper ions, in a dose-dependent manner, as was determined by the measurement of conjugated dienes, thiobarbituric acid reactive substances (TBARS), and lipid peroxides assays. LDL isolated from the plasma of 10 normolipidemic subjects who were orally supplemented for 2 weeks with 100 mgb of licorice/day was more resistant to oxidation than LDL isolated before licorice supplementation. Dietary supplementation of atherosclerotic apolipoprotein E deficient mice (E° mice) with licorice (200 g/g) resulted in a substantial reduction in the susceptibility of their LDL to oxidation along with a reduction in the atherosclerotic lesion area. Seven flavonoid constituents with antioxidant capacity were isolated from licorice roots. The isolated compounds were identified as the isoflavans hispaglabridin A, hispaglabridin B, glabridin, and 4'-O methylglabridin, the two chalcones: isoprenylchalcone and isoliquiritigenin, and the isoflavone, formononetin. Among these compounds, glabridin constituted the major amount in the licorice crude extract (11.6%, w/w). The antioxidative capacities of the isolated compounds were tested against b-carotene destruction and LDL oxidation. The isoflavans at a concentration of 50 M inhibited b-carotene consumption, following 90 min of incubation at 5 0°C, similar to the inhibitory effect of the whole licorice crude extract (at 16 mg/l). The chalcones exhibited moderate inhibition and the isoflavone was inactive. The inhibitory effect of the above constituents (at a concentration of 30 M) on AAPH-induced LDL oxidation was determined by measuring the amount of TBARS and lipid peroxides formed. While most of the above compounds exhibited substantial inhibitory activity, formononetin and also vitamin E were not active.
The isoflavan glabridin was the most abundant and potent antioxidant. The effect of glabridin consumption on the susceptibility of LDL to oxidation was studied in E° mice and was compared with that of a known flavonoids such as quercetin and catechin (which are present in red wine). Determination of the extent of LDL oxidation by measuring the formation of TBARS after 2 h of LDL incubation with CuSO4 (10 M) or AAPH (5 mM), revealed that glabridin or quercetin consumption resulted in a 53% and 54% reduction in copper ion-induced oxidation, respectively, and a 95% and 83% reduction in AAPH-induced LDL oxidation, respectively.
About 80% of added glabridin was found to bind to LDL. Glabridin-enriched LDL was less susceptible to copper ion-induced or to macrophage-mediated oxidation (in a glabridin dose- dependent manner),in comparison to native LDL. Glabridin inhibitory activity on in vitro oxidation of human LDL was also investigated, by determining the formation lipid peroxides and oxysterols as well as the consumption of LDL-associated lipophilic antioxidants. During LDL oxidation induced by AAPH (5 mM), glabridin inhibited the formation of TBARS, lipid peroxides and cholesteryl linoleate hydroperoxide (CLOOH) at all the concentrations tested (5-60 M).During oxidation induced by copper ions (10 M) however, glabridin exhibited antioxidant activity at concentrations greater than 20 M, but a pro-oxidant activity at concentrations lower than 20 M. Glabridin (30 M) inhibited the formation of cholest 5-ene-3,7-diol (7-hydroxycholesterol), cholest-5-ene-3-ol-7-one (7 ketocholesterol) and cholestan-5,6-epoxy-3-ol (5,6-epoxycholester ol) after 6 h of AAPH-induced LDL oxidation, by 55%, 80% and 40%, respectively, and also after 6 h of copper ion-induced LDL oxidation, by 73%, 94% and 52%, respectively.
Glabridin also inhibited the consumption of
b-carotene and lycopene by 38% and 52%, respectively, after 30 minutes of LDL oxidation with AAPH, but glabridin failed to protect vitamin E consumption under these conditions.
The in vivo and in vitro effect of glabridin on the susceptibility of LDL to oxidation, may be related to the glabridin absorption or binding to the LDL particle, leading to the subsequent protection of LDL from oxidation by inhibiting the formation of lipid peroxides and oxysterols, and by protecting LDL-associated carotenoids from consumption.
In order to evaluate the contribution of the different parts of the isoflavan molecule to its antioxidant activity against LDL oxidation, we modified the glabridin molecule and also used several related compounds. The peak potential (E1/2) of the isoflavan derivatives, their free radical scavenging capacity toward 1,1-diphenyl-2 picryl-hydrazyl (DPPH) and their ability to chelate transition metals were also analyzed in parallel to LDL oxidation studies.
In copper ion-induced LDL oxidation, glabridin, 4'-O-methylgl abridin, hispaglabridin A, and hispaglabridin B inhibited the formation of LDL associated conjugated dienes, TBARS and lipid peroxides, and reduced the electrophoretic mobility of LDL. In a macrophage-mediated oxidation of LDL system, LDL oxidation was inhibited by all studied isoflavans, similarly to the effects obtained in the copper ion-induced LDL oxidation system. In contrast, 2'-O methylglabridin showed only minor inhibitory effect in both LDL oxidation systems and 2',4'-O-dimethylglabridin was completely inactive. Resorcinol which is identical to the phenolic B ring in glabridin, presented very low inhibitory activity in the above oxidation systems. The licorice isoflavene glabrene was the most potent antioxidant compound of all flavonoid derivatives tested.
None of the isoflavan derivatives were able to chelate iron, or copper ions. These results suggest that the antioxidant effect of glabridin on LDL oxidation resides mainly in the hydroxyl at 2'-position of B ring, and that the hydrophobic moiety of the isoflavan is also essential for glabridin antioxidant activity. The 2'-position of the hydroxyl group in the B ring significantly contributed to the inhibitory effect on LDL oxidation, but did not influence the isoflavan's ability to donate an electron to DPPH, or their peak potential values.
Glabridin could possibly affect macrophage-mediated oxidation of LDL not only as a result of its association with LDL, but also by its possible accumulation in, and effect on, macrophage oxidative status. Glabridin was able to accumulate in mouse peritoneal macrophages (MPM), as well as in J-774 A.1 macrophage-like cell line. Both cells accumulated up to about 1.5 g of glabridin/mg cell protein after 2h of incubation, in a time- and glabridin dose-dependent manner. Most of the glabridin was accumulated within 30 minutes of incubation and macrophage-mediated oxidation of LDL was inhibited by 80% on using 20 M glabridin. Although similar cellular uptake of glabridin and of 2',4'-O-dimethylglabridin were obtained, the accumulation of the modified glabridin had no effect on macrophage-mediated oxidation of LDL. Glabridin, but not the modified derivative, was able to inhibit the release of superoxides from MPM that were stimulated with phorbol myristate acetate (PMA), or with LDL and copper ions, by up to 60% in a time- and glabridin dose-dependent manner. Glabridin inhibits the activation of the macrophage NADPH-oxidase as the translocation of the cytosolic component of this enzyme, P-47, in response to PMA or to LDL+ copper ions, was completely inhibited in the glabridin-enriched MPM, in comparison to control cells. The inhibitory effect of glabridin on macrophage-mediated oxidation of LDL via NADPH-oxidase superoxide production and release was found to be mediated by protein kinase C (PKC) signal transduction pathway.
Finally, the ability of MPM derived from mice that consumed glabridin (25 g/mouse/day for a 6weeks period) to oxidize LDL was also analyzed. MPM from mice that consumed glabridin, oxidized LDL 88% less than MPM derived from placebo-treated mice.
We conclude that glabridin is a potent lipophilic antioxidant against LDL oxidation, both in vitro and in vivo, and it can provides its antioxidant activity by both interacting with LDL directly, and also indirectly, via its effect on arterial wall macrophage oxygenases.
These properties of glabridin may significantly contribute to its anti-atherogenicity.

Flavonoid intake and risk of chronic diseases

Paul Knekt, Jorma Kumpulainen, Ritva Järvinen, Harri Rissanen, Sari Isotupa, Antti Reunanen, Markku Heliövaara, Lyly Teppo,
and Arpo Aromaa

National Public Health Institute, Helsinki, Agriculture Research Centre of Finland, Jokioinen,University of Kuopio, Kuopio, and Finnish Cancer Registry, Helsinki, Finland

Flavonoids may because of their antioxidant properties protect tissues against oxygen free radicals, and accordingly provide protection against several chronic diseases. This issue was studied in a Finnish cohort of 10 000 men and women, which was followed up for 26 years. The total dietary intake was determined with a dietary history interview method. The intake of the flavonoids quercetin, kaempferol, naringenin, hesperetin and myricetin was estimated mainly based on flavonoid concentrations in Finnish foods. Individuals with a higher flavonoid intake had a significantly reduced incidence of several diseases. The relative risk of lung cancer (RR) between the highest and lowest quartiles of quercetin intake was 0.42 and the relative risk of prostate cancer between the highest and lowest quartiles of myricetin was 0.43. The coronary heart disease mortality was reduced at higher quercetin (RR=0.79) and kaempferol (RR=0.82) levels and the incidence of cerebrovascular disease was reduced at higher levels of quercetin (RR=0.86), kaempferol (RR=0.70), naringenin (RR=0.79) and hesperetin (RR=0.80). The incidence of asthma was reduced at higher quercetin (RR=0.76), naringenin (RR=0.69) and hesperetin (RR=0.64) levels. A reduced non-insulin dependent diabetes mellitus risk was associated with higher quercetin (RR=0.81) and myricetin (RR= 0.79) levels. Contrary an elevated risk of rheumatoid arthritis was observed at higher intakes of quercetin (RR=2.64) and kaempferol (RR=1.91), and an elevated risk of cataract at higher intake of hesperetin (RR=1.66). The results suggest an apparent existence of several chronic diseases that potentially may be preventable by dietary flavonoids. Further research will be needed.

Xanthine oxidase inhibition and protein-binding action of the procyanidin-rich French maritime pine bark extract, Pycnogenol®

Hadi Moini, Qiong Guo, and Lester Packer

Department of Molecular and Cell Biology, 251 Life Sciences Addition, University of California at Berkeley, Berkeley, CA 94720-3200, USA

Pycnogenol® (PYC), an extract from the bark of the French maritime pine, is a complex mixture of bioflavonoids with reported protective effects against disease. Since PYC is an effective scavenger of reactive oxygen species, and its main constituents are procyanidins of various chain lengths, involvement of its redox activity and/or direct binding to the enzyme protein in its subsequent action on enzyme activity were of interest. Therefore in this study, the molecular basis of the effect of PYC on enzyme activities was investigated. PYC dose-dependently inhibited the activities of oxidative enzymes such as xanthine oxidase, xanthine dehydrogenase, horseradish peroxidase, and lipoxygenase, but it did not affect the activities of glucose oxidase, ascorbate oxidase, or elastase. To characterize the mechanism of PYC action, studies were focused on xanthine oxidase and glucose oxidase. Under non-denaturating conditions, PYC changed the electrophoretic mobility of xanthine oxidase but not of glucose oxidase. Gel filtration chromatography confirmed a change in the molecular weight of both xanthine oxidase and xanthine dehydrogenase in the presence of PYC. Hydrophobic binding was the dominant mode of interaction between PYC and xanthine oxidase. The importance of PYC binding in the modulation of enzyme activity was verified by showing that PYC binds and inhibits catalase, but not superoxide dismutase. However, no correlation was found between superoxide/hydroxyl radical scavenging activity and the inhibitory effect on xanthine oxidase activity of PYC, various purified procyanidins, or other complex mixtures of flavonoids.
Conclusion: PYC selectively inhibits xanthine oxidase through binding to the enzyme protein rather than by redox activity.

Modulation of Gene Expression in Human Cancer (T-24) Cells by Ginkgo biloba extract, EGb 761

Kishorchandra Gohil, Ronald K. Moy, Sahar Farzin,
John J. Maguire and Lester Packer

Department of Molecular and Cell Biology, Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory,
University of California, Berkeley, CA 94720

DNA microarray and oligonucleotide array GeneChips offer new opportunities to monitor the effects of phytochemicals on human cells. We have applied the Affymetrix hu6800 GeneChips to define the transcriptional response of human bladder cancer cells to the extract of Ginkgo biloba leaves, EGb 761. The extract treated cells showed increased levels of peroxides and decreased synthesis of DNA. Quantitative analysis of hybridization signals from Gene Chips revealed net induction of transcription in the T-24 cells over a period of 72 hours in the presence of the extract. Functional classification of affected transcripts revealed changes in the expression of genes encoding antioxidant proteins, vesicular membrane proteins and transport, DNA synthesis-repair, cell cycle and transcription. The expression profile may define, for the first time, the effects of flavonoid-glycosides and terpenoids, the pharmacologically active components of the extract, on gene expression of a human cancer cell.

Unraveling the mysteries of red wine

Fulvio Ursini

Department of Biological Chemistry, University of Padova, Italy

Drinking suitable amounts of wine and being a pre-menopausal female lowers, according to epidemiological studies, the risk of cardiovascular disease. Among the different biological mechanisms proposed to account for this, the decreased oxidizability of LDL is common to both conditions. The antioxidant activity of grape seed polyphenols accounts for the prevention of experimental atherosclerosis in rabbits. Although positive evidence for bioavailability is still missing, the protective effect has been attributed to the procyanidin fraction. In fact, it has been reported that catechins do not protect cholesterol fed rabbits from atherosclerosis, while polymers are much better antioxidants on both kinetic and thermodynamic bases. By contrast, the major estrogen, 17b-estradiol (E2), has a trivial free radical scavenging capacity, while it strongly inhibits LDL oxidation. Puzzled by this intriguing evidence, we investigated the mechanism of this unusual antioxidant effect of E2, as a model for protection shared with other polyphenols. E2 is structurally similar to isoflavones or stilbenes such as genistein or resveratrol. The increased oxidative resistance of E2-treated LDL challenged with copper was accompanied by a delay of LDL surface polarity increase, as measured by the generalized polarization of Laurdan. This indicates that in the presence of E2, lipid hydroperoxides are produced at a slower rate during the lag phase. These hydroperoxides, in addition to generate new radicals, perturb the water lipid boundary thus facilitating further massive peroxidation of the LDL lipid core. Circular dichroism and tryptophan fluorescence lifetime revealed that E2 increases apoB-100 secondary structure and folding of the protein, thus making apoB resistant to oxidative unfolding and the LDL resistant to peroxidation. Apparently, the conformation of the apoprotein and its interaction with the outer layer of the LDL particle contributes to lipid packing, thus increasing the overall oxidative resistance. The observation that in in vivo oxidatively modified LDL- apoB is fully unfolded supports this conclusion. The role of surface lipid packing in the overall oxidative resistance is also in agreement with previously observed antioxidant effects of cholesterol. This antioxidant effect is not related to the E2 hormonal effect, but due to conformational effect on apoB, and is reproduced by the soybean iso flavone, genistein. Further studies will clarify if other structurally similar polyphenols present in wine, but also in tea or other natural sources, have a similar effect. This will reveal a new and appealing common motif shared by women and wine.

In collaboration with the Department of Molecular Toxicology USC; Instutute for Experimental Mecicine C.N.R, Rome and Indena s.p.a. Milan

Session IV
Redox Regulation of Cell Signaling

Selenium-dependent antioxidant functions of mammalian thioredoxin reductase

Arne Holmgren, Liangwei Zhong, Elias Arner, Sergei Kuprin and Tahir Kerimov

Medical Nobel Institute for Biochemistry, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-171 77 Stockholm, Sweden

Thioredoxin reductase (TrxR) catalyzes reduction of oxidized thioredoxin (Trx) by NADPH.Thioredoxin has a large number of functions including antioxidant functions as an electron donor to thioredoxin peroxidases (peroxiredoxins).The Se-dependent high molecular weight TrxR enzymes present in C. elegans and mammals are dimers of 55 kDa subunits or higher, and have a wide substrate specificity also involving lipid hydroperoxides and hydrogen peroxide reducing activity.We have expressed recombinant mutants of the rat cytosolic TrxR in E.coli having replacements of the essential SeCys residue penultimate to the C-terminus in the conserved sequence: -Gly-Cys-SeCys-Gly. Replacement of SeCys by Cys generated a mutant enzyme with a 100-fold reduction in Kcat for thioredoxin and also a 10-fold lower Km but lacking hydrogen peroxide reducing activity. The mutant enzyme had a pH optimum of 9 as opposed to 7 for the wild type enzyme consistent with a catalytic function of a low pKa selenol in the enzyme mechanism. The truncated enzyme lacking the SeCys-Gly dipeptide expected in Se deficiency lacked catalytic activity in reduction of Trx or hydroperoxides although it was folded and contained FAD and had an intact NADPH-dependent reductive half-reaction generating the thiolate-flavin charge transfer complex characteristic of mammalian TrxR. We have postulated a structure and mechanism for the mammalian TrxR based on the identification of a selenylsulfide involving SeCys498 and Cys497 in the oxidized enzyme. By EPR spin trap methods we have discovered that the native Se-containing enzyme produces less reactive oxygen species compared with the mutant enzymes or the E.coli enzyme which suggests a Se-dependent antioxidant defense mechanism of putative relevance for the effects of selenium as an anticancer and antioxidant factor.

Redox paradox of HIV-1 infection

Takashi Okamoto

Department of Molecular Genetics, Nagoya City University Medical School, Nagoya 467-8601, Japan

This world is full of paradoxes particularly when it comes to some fatal illnesses like HIV/AIDS and more specifically when it comes to "redox" phenomenon. There have been increasing reports regarding the redox disturbances during HIV infection: such as low plasma glutathione levels in HIV infected people and most impressively in CD4 T cells, and paradoxically high serum TRX level in such plasma. These observations lead scientists to recommend the use of anti-oxidant compounds as supportive therapeutic measures. However, insufficient prescription of antioxidants can be dangerous: for example, excessively administered vitamin C or beta-carotene could be converted to prooxidant compounds. With regard to TRX actions, we previously found that TRX activates the NF-kB DNA binding and stimulate HIV-1 replication, and recently reported that extracellular TRX could activate NF-kB pathway presumably by lowering the threshold of the signaling cascade (J. Immunol. 163: 351, 1999). In another line of our basic HIV studies, we have recently found that in CD4 T cells and macrophage/monocyte cells, constitutive expression of HIV-1 Tat transactivator could induced TRX. When we tested these Tat-expressing cells for their apoptotic responses to oxidative stress, the net output was distinct between CD4 T (CEM) cells and macrophage/monocytes (THP-1): whereas THP-1/Tat cells became resistant to apoptosis, CEM/Tat became more sensitive than parental cells. We found that these apparent paradoxical responses could be ascribed to the differences in modulation of cellular antioxidant system, which will be described more precisely in a separate presentation at this congress (Inuzuka et al.). However, these differential effects of Tat clearly explains why latent HIV-1 infection preferably occurs in the monocyte/macrophage cell lineage.

Thioredoxin control of the p53 tumor suppressor protein

Gary F. Merrill, Neil A. Bersani, Jason R. Merwin,
Christopher S. Stoner, Nathan I. Lopez and George D. Pearson

Department of Biochemistry and Biophysics, Oregon State University

The ability of human p53 to stimulate reporter gene expression is inhibited in êtrr1 yeast lacking thioredoxin reductase. The êtrr1 yeast accumulate large amount of oxidized glutathione (GSSG), but this is not responsible for p53 inhibition because restoring glutathione to the reduced state (by overexpressing glutathione reductase) does not restore p53 activity. Also, deleting the yeast glutathione reductase gene (GLR1) does not inhibit p53 activity, even though it causes GSSG accumulation. A protein electrophoretic mobility shift assay was developed to measure the in vivo redox state of thioredoxin, and showed that oxidized thioredoxin accumulates in yeast lacking thioredoxin reductase. Accumulation of oxidized thioredoxin rather than loss of reduced thioredoxin is probably responsible for p53 inhibition, because deleting the yeast thioredoxin genes does not cause p53 inhibition and suppresses the inhibitory effect of deleting thioredoxin reductase on p53 activity. The yeast results suggest that oxidized thioredoxin actively oxidizes thiols on p53 or a protein required for p53 activity. To ask whether accumulation of oxidized thioredoxin inhibits p53 in human cells, the effects of selenium depletion on thioredoxin reductase activity and thioredoxin redox state in human MCF7 cells was determined. Although selenium depletion caused a modest decrease in thioredoxin reductase levels, it did not affect the redox state of thioredoxin, which remained primarily in the reduced state. To address the question of thioredoxin involvement in controlling p53 activity in human cells, a means of significantly inhibiting basal thioredoxin reductase levels in mammalian cells needs to be developed.

Thioredoxin in development and aging

Junji Yodoi, Mikiko Kobayashi, Kohei Shiota, Akira Mitsui,
and Akira Nishiyama

Department of Biological Responses, Institute for Virus Research,
Kyoto University, Department of Anatomy and Developmental Biology,
Faculty of Medicine, Kyoto University, Pharmaceutical Research Laboratories, Ajinomoto Co., Inc.

Thioredoxin (TRX) is one of the major components of the thiol reducing system and plays multiple regulatory roles in cellular processes such as proliferation, apoptosis and gene expression. We have previously shown that in early development, two-cell block was caused by reactive oxygen species and released by anti-oxidant, such as TRX. Targeted disruption of thioredoxin resulted in embryonic lethality shortly after implantation in mouse. The localization of TRX in developing mouse shows some spatial heterogeneity and temporal sequence, suggesting a role of TRX in cell differentiation and development.
Recently, thioredoxin binding protein-
2 (TBP-2) was identified by yeast two-hybrid system. TBP-2 was identical to vitamin D3 up-regulated protein 1 (VDUP1). Overexpression of TBP-2/VDUP1 diminished insulin reducing activity of TRX as well as TRX expression. Treatment of HL-60 cells with vitamin D3 caused an increase of TBP-2/VDUP1 expression and down-regulation of the expression and the reducing activity of TRX, suggesting that the TRX TBP-2/VDUP1 interaction may be an important redox regulatory mechanism in differentiation of myeloid and macrophage lineages.
Transgenic mice overexpressing
TRX show resistance against ischemic neuronal injury in a middle cerebral artery occlusion model. Bone marrow cells from TRX transgenic mice also demonstrated resistance against UV- or X-ray irradiation (in collaboration with Drs. Inoue and Hirabayashi, National Institute of Health Sciences, Tokyo, Japan). In addition, TRX transgenic mice exhibit up to on forth extension of median life span and one-third of maximum life span. Overexpressed TRX may have protected mice from oxidative stress-induced tissue damage during aging process.

A role for methallothionein in nitric oxide signaling
in vascular tissue

Bruce R. Pitt

Department of Pharmacology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261

Recent in vitro studies suggest that the oxidoreductive capacity of metal thiolate clusters in metallothionein (MT) contributes to intracellular zinc homeostasis. We used fluorescence based techniques to address this hypothesis in intact endothelial cells focusing on the contributory role of the important redox signaling molecule, nitric oxide (NO). Microspectrofluorometry with Zinquin revealed that exposure of cultured sheep pulmonary artery endothelial cells (SPAEC) to S-nitrosocysteine resulted in release of TPEN (N,N,N',N'-tetrakis(2-pyridyl-methyl)-ethylenediamine) chelatable zinc. SPAEC were transfected with a plasmid expression vector suitable for fluorescence resonance energy transfer (FRET) containing cDNA of MT sandwiched between two mutant green fluorescent proteins. Exposure of SPAEC transfected with this chimera to NO donors or to agents that increased cytoplasmic Ca2+ via endogenously generated NO decreased efficiency of FRET in a manner consistent with release of metal (Zn) from MT. A physiologic role for this interaction in intact tissue was supported by the lack of myogenic reflex in resistance arteries of MT knockout mice unless endogenous NO synthesis was blocked. These data suggest an important role for metal thiolate clusters of MT in NO signaling in the vascular wall.

Role of reactive oxygen intermediates in cytokine signaling

Bharat B. Aggarwal

Cytokine Research Section , Department of Bioimmunotherapy,
The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030

Although protein-protein interaction has been shown to be critical for cytokine signaling, evidence is emerging which suggest that reactive oxygen intermediates (ROI) play a major role in cell proliferation and apoptosis induced by various cytokines. We present evidence, using three independent approaches, that the pathways required for the cytokine-induced activation of apoptosis, nuclear transcription factors, NF-kB and AP-1 and stress-activated kina ses, all require the generation of ROI. The first approach involved overexpression of manganese superoxide dismutase and g-glutamyl cysteine synthetase, which led to suppression of TNF-induced activation of apoptosis (cytotoxicity and caspase activation) and anti-apoptotic signals (NF-kB, AP-1, JNK, and MEK). The second approach involved mitochondria ( a major source of ROI) depletion, which also abolished TNF-induced apoptosis and NF-kB activation. The third approach involved the use of ROI quenchers, which produced mixed results. All these approaches suggested a critical role of ROI in induction of apoptosis, NF-kB, AP-1, JNK, and MEK by cytokines. The data from these experiments will be presented in detail.

Higuchi M., Aggarwal B. B., and Yeh E. T. H. J. Clin. Invest. 99: 1751-1758 1997.
Giri D. K., and Aggarwal B. B. J. Biol. Chem. 273, 14008-14014, 1998.
Manna S., Zhang, H. J., Yan T., Oberley L. W., and Aggarwal B. B. J. Biol. Chem. 273, 13245-13254, 1998.
Estrov Z., Manna S.K., Harris D., Van Q., Kantarjian H.M., Talpaz M., and Aggarwal B. B. BLOOD 95, 2844 2853, 1999 .
Shrivastava A., and Aggarwal B. B. Anti. & Redox Signaling. 1, 181-191, 1999
Manna S. K., Kuo M.T., and Aggarwal B. B. Oncogene. 18, 4371-4382, 1999.
Manna S., and Aggarwal B. B. J. Immunol. 162, 1510-1518, 1999

Regulation of agonist-induced ICAM-1 and VCAM-1 gene expression by thiol oxidation/reduction
dependent mechanisms

Sashwati Roy, Chandan K. Sen, Hao G. Nguyen,
and Lester Packer

Molecular and Cell Biology and Lawrence Berkeley National Laboratory, University of California, Berkeley, CA 94720

Signal transduction processes are sensitive to intracellular thiol redox state. Studies have shown that thiol antioxidants (e. g., N-acetylcysteine, pyrrolidine dithiocarbamate) down-regulate agonist-induced expression of adhesion molecules. At the same time it has been reported that cellular thiol depletion, using the glutathione-conjugating agent diethyl maleate, inhibits inducible endothelial cell activation via down-regulation of adhesion molecule expression. The molecular sites and mechanisms involved in such regulation of inducible adhesion molecule (ICAM-1 & VCAM-1) gene expression by thiol redox state are yet unclear. We have observed that the thiol antioxidant lipoate differentially down-regulates agonist-induced VCAM-1 but not ICAM-1 expression in primary human umbilical vein endothelial cells (HUVEC). When supplemented to cells, lipoate (LA) is readily taken up and a part of it is reduced to dihydrolipoate (DHLA), which is a strong cellular reductant. To assess whether the oxidized form (LA) or reduced form (DHLA) of lipoate is responsible to regulate inducible VCAM 1 expression, the effect of various thiol-status modulating agents on inducible ICAM-1 and VCAM-1 gene expression was investigated. Thiol reducing (dithiothreitol) and blocking (N-ethyl maleimide) agents did not influence inducible VCAM 1 expression. The thiol oxidizing agent diamide, however, strongly inhibited inducible VCAM-1 expression. Cellular thiol oxidation suppressed inducible VCAM-1 expression by inhibiting the binding of NF-kB to kB sites.

Oxidative stress disrupts glucocorticoid hormone-dependent transcription of the amiloride-sensitive epithelial sodium channel a-subunit in lung epithelial cells through erk-dependent and thioredoxin-sensitive pathways

Hao-Chien Wang1, Mark D. Zentner1, Kwang-Jin Kim1, Reen Wu2 and David K. Ann1

1Departments of Molecular Pharmacology and Toxicology, University of Southern California, Los Angeles, CA 90033, and 2Center for Comparative Respiratory Biology and Medicine, University of California, Davis, CA 95616

The amiloride-sensitive epithelial Na+ channel (ENaC) plays a critical role by mediating Na+ transport across lung alveolar epithelial barrier and thus maintains alveolar fluid balance. It is generally accepted that reactive oxygen and nitrogen species can inhibit ENaC activity and aggravate acute lung injury; however, the molecular mechanism for free radical-mediated ENaC inhibition is unclear. Here, we investigated whether exogenous H2O2 modulates a-ENaC gene expression in lung epithelial cells through a similar molecular mechanism by utilizing transient transfection reporter assays with homologous a-ENaC promoter constructs and site-directed mutagenesis analyses. Together with results from studies using a heterologous promoter construct, we demonstrate that the presence of an intact GRE is necessary and sufficient for oxidants to repress a-ENaC expression. In addition, overexpression of thioredo-xin (TRX) restored glucocorticoid receptor (GR) action on the a-ENaC GRE in the presence of exogenous H2O2. Taken together, we hypothesize that oxidative stress impairs Na+ transport activity by inhibiting dexamethasone (Dex)-dependent a-ENaC GRE activation via both ERK-dependent and TRX-sensitive pathways. These results suggest a putative mechanism whereby cellular redox potentials modulate the GR/Dex effect on a-ENa C expression in lung and possibly other tight epithelia as well.

Session V
Vitamin E and Lipoic Acid in Health and Disease

Vitamin E prevents the increase of intracellular calcium elicited by reactive oxygen species

H.K. Biesalski and D. Mayer

Department of Biological Chemistry, University of Hohenheim,
70593 Stuttgart, Germany

It has been demonstrated that oxidative stress results in an increase of intracellular free calcium [Ca2+]i in EC. It is argued that this increase might be an important reason for ROS induced stress response. To elucidate the role of vitamin E on [Ca2+]i HMEC-a cells either pre-incubated with 25 M vitamin E or vehicle (ethanol) and [Ca2+]i was measured by means of a fluorescence-imaging system with a special perfusion chamber. Oxidative stress was performed with hypoxanthin/xanthine oxidase (HX/XO) perfusion (1000 sec). This treatment results in a 5-fold increase of [Ca2+]i compared to vitamin E-treated cells. Incubation with cell membrane Ca2+ channel blockers (SKF, Ni2+) or DTT (to protect SH groups) did not significantly reduce [Ca2+]i in untreated cells, in contrast to vitamin E-treated cells where [Ca2+]i mobilization was nearly completely abolished. From this we assume that vitamin E protects i.c. Ca2+ stores. This is strengthened by the fact that o-phenanthroline incubation (iron chelator) reduced [Ca2+]i release in untreated cells moderatly, but completly in vitamin E-treated cells. Thapsigargin pre-treatment (depletes [Ca2+]i) of EC in Ca2+ free medium and HX/XO application did not increase [Ca2+]i. If, however, these cells were reperfused with Ca2+-containing medium, untreated but not vitamin E-treated cells showed a strong Ca2+ increase and oscillation. Activation of PKC increases i.c. Ca2+ release. To exclude that the proposed action of vitamin E on PKC results in the reduction of ROS provoked i.c. Ca2+ we carried out experiments with a PKC-inhibitor (Staurosporine). Inhibition of PKC did not result in any additional effect on ROS provoked Ca2+ release. Taken together these data show for the first time, that vitamin E interferes with [Ca2+]i mobilization induced via oxidative stress. Consequently, vitamin E effects (beside its well known antioxidant properties) on stress-induced signal transduction might be mediated via effects on [Ca2+]i mobilization.

Lipid peroxidation, metabolism of
lipid peroxidation products, and aging

Werner Siems 1,Thomas von Zglinicki2, Violeta Serra2,
and Tilman Grune3

1Herzog-Julius-Hospital for Rheumatology and Orthopedics, Bad Harzburg, 2Institute of Pathology, and 3Clinics of Physical Medicine and Rehabilitation, Medical Faculty (Charité), Humboldt-University Berlin, Germany

4-Hydroxynonenal (HNE) plasma levels in healthy persons aged between 5 and 89 years increase with rising age. That is accompanied by increasing plasma malondialdehyde (MDA) and by declining intraerythrocytic glutathione (GSH) levels.
The metabolism of lipid peroxidations products such as HNE is an important part of the intracellular secondary antioxidative protective system. The question whether the age dependent increased levels of lipid peroxidation products such as HNE are due to age-dependent changes in the rate of the metabolism of those compounds was investigated in fibroblasts. In primary cultures of synovial fibroblasts an HNE consumption rate of 27.2 nmoles/min x 106 cells was measured. HNE was metabolized in those cells mainly oxidatively to 4-hydroxynone noic acid, intermediates of the tricarboxylic acid cycle and water and to the glutathione-HNE conjugate. The share of protein-bo und HNE was about up to 8% of the total HNE after 10 min of incubation. In human fibroblasts of subjects aged between 30 and 86 years the capacity of the HNE metabolism decreased by about 70%, i.e. the HNE metabolism is strongly age-dependent. The age-dependent reduction of the capacity of the HNE metabolism partially is due to the decreased flux rate of glutathione transferase reaction leading to the formation of the GSH HNE conjugate. Furthermore, a decrease of the mitochondrial 4 hydroxynonenoic acid formation may be responsible for the age-dependent decline of that secondary antioxidative pathway.

Doxorubicin-induced apoptosis is reversed by
nitrone spin traps and ebselen

Srigiridhar K., Konorev, Eugene , Joseph, Joy,
and Kalyanaraman, B

Biophysics Research Institute, Medical College of Wisconsin, Milwaukee

Doxorubicin (DOX) is a broad-spectrum antitumor anthra-cycl ine antibiotic that is being used to treat a variety of cancers. The efficacy of DOX is however limited owing to the severe cardiotoxici ty that leads to the development of cardiomyopathy and heart failure. Redox-activation of DOX to reactive oxygen species appears to be responsible for DOX-induced cardiotoxicity. In this work we investigated DOX-induced apoptosis in cultured bovine aortic endothelial cells (BAEC) and ventricular myocytes isolated from adult rat heart. Exposure of BAEC or myocytes to submicromolar levels of DOX induced significant apoptosis (70%) as measured by DNA fragmentation and TUNEL assays which were quantitated using a Sigma Scan Image analysis program. Pretreatment of cells with 100 M nitrone spin traps, N-t-butyl-(-phenylnitrone (PBN) or (-(4-pyridyl-1-oxide)-N-t-butylnitrone (POBN) inhibited DOX induced apoptosis (7%). Ebselen (50M), a glutathione peroxidase mimetic also inhibited apoptosis (8%). At these concentrations, these compounds did not affect DOX uptake into BAEC. DOX (0.5-1M) inactivated mitochondrial aconitase and complex I. PBN (100M), POBN (100M) and Ebselen (50M) restored these activities. These compounds inhibited DOX-induced caspage-3 activation and cytochrome c release. PBN and Ebselen also restored the glutathione levels in DOX-treated BAEC. We conclude that nitrone spin traps and Ebselen inhibit DOX-induced apoptotic signaling mechanism and that this antiapoptotic mechanism may be linked in part to the inhibition of hydrogen peroxide formation.

Redox balance in health and disease

Barry Halliwell

Department of Biochemistry, National University of Singapore,
Singapore 119260

In vivo, cells function in a largely-reduced state, with areas of local transient oxidation, e.g., to allow protein disulphide bridge formation and activation of certain transcription factors. During oxidative stress the oxidation level becomes abnormally high. The implications of this for cell proliferation, oxidative damage, cell death by apoptosis or necrosis, and the effects of administered antioxidants on these parameters, will be examined. Diabetes is an example of a disease where multiple parameters indicative of oxidative stress are elevated. The value of oxidative damage parameters as a predictor of disease development will be discussed

Current perspectives in diabetic research:
the AGE concept and implications for antioxidant therapy

A. Bierhaus, K. M. Haslbeck°, M. Morcos*, S. Schiekofer,
M. Andrassy, J. Chen, T. Dehmer, B. Neundörfer°, W. Möller#,
H. Tritschler#, M. Schwaninger*, D. Heuss°, H.-U. Häring,
E. Schleicher, and P. P. Nawroth

Sekt. Vaskuläre Medizin, Med. Klinik IV und Inst. für Neuropharm. Univ. Tübingen; *Med. Klinik I und Neurolog. Klinik, Univ. Heidelberg; °Neurol. Klinik Universität Erlangen-Nürnberg;#ASTA-Medica GmbH Frankfurt

Genes controlled by NF-kB have been shown to be involved in pain control (Machelska et al; Nat. Med. 1998;4: 1425-8). We studied whether upregulation of RAGE, deposition of the defined AGE-adduct CML and NF-kB activation is evident in human and animal models of diabetes mellitus. Immunhistology of sural nerve biopsies from diabetic patients demonstrated prominent upregulati on of RAGE, local accumulation of CML and activated NF-kB in the perineurium and the vascular endothelium. This colocalisation was not observed in controls. Schwann´ cells also demonstrated increased RAGE expression, but no staining for CML and NF-kB. In mouse models, histological examination confirmed upregulation of NF-kB in the perineurium, but was also evident in Schwann´ cells. NF-kB activation could be reversed by thioctic acid. Thus, NF-kB activation occurs in hyperglycemia, however different neuronal cells are involved in humans and mice. Activation of NF-kB results in enhanced NF-kB dependent gene expression, since transgenic mice in which a b-globin reporter gene is driven by NF-kB, demonstrated increased b-globin expression after inducing hypeglycemia by streptozotocin (STZ). Involvement of RAGE was shown, since injection of AGEs resulted in an RAGE-dependent induction of the b-globin-gene, that could be blocked in the presence of soluble RAGE, neutralizing anti-RAGE-antibodies and thioctic acid. These studies provide evidence that (1) perineural and endothelial cells are targets for oxidative stress in diabetic neuropathy, (2) that the AGE-RAGE-NF-kB system is upregulated in diabetic neuropathy and (3) that upregulation of cell activation can be reversed by treatment with thioctic acid or competitors of the AGE/RAGE pathway.

a-Lipoic acid in the treatment of experimental diabetic and other neuropathies

Phillip A. Low

Department of Neurology, Mayo Foundation. Rochester, Minnesota

Oxidative stress is present in the diabetic state. Our work has focused on its role in the pathogenesis of diabetic neuropathy. Antioxidant enzymes are reduced in peripheral nerve and are further reduced in diabetic nerves. That lipid peroxidation will cause neuropathy is supported by evidence of the development of neuropathy de novo when normal nerve is rendered a-tocopherol deficient and augmentation of the conduction deficit in diabetic nerves subjected to this insult. The mechanism of oxidative stress appears to be primarily due to the processes of nerve ischemia and hyperglycemic auto-oxidation. The indices of oxidative stress include an increase in nerve, dorsal root and sympathetic ganglia lipid peroxidation. Lipolysis is increased, resulting in increased arachidonic acid and the formation of 4-hydroxynonenal, which is toxic to neurons and causes apoptosis.
a-Lipoic acid is a potent antioxidant that prevents lipid peroxidation in vitro and in vivo. We demonstrated in vitro that lipid peroxidation of peripheral nerve occurred in a high glucose medium and that a-lipoic acid dose-dependently prevented its generation. In the in vivo system, we evaluated the efficacy of the drug in doses of 20, 50 and 100 mg/kg, administered intraperitoneally in preventing the biochemical, electrophysiologic and nerve blood flow deficits in peripheral nerve of experimental diabetic neuropathy. a-Lipoic acid dose- and time-dependently prevented the deficits in nerve conduction, nerve blood flow and biochemical abnormalities (of a reduction in reduced glutathione and lipid peroxidation). The nerve blood flow deficit was 50% (P < 0.001). Supplementation prevented the deficit; at the highest concentration, nerve blood flow was not different to control nerves. Digital nerve conduction underwent a dose-dependent improvement at 1 months (P < 0.05). By 3 months, all treated groups had lost their deficit.
In chronic experimental diabetic neuropathy (>6 months) a myelinopathy of dorsal roots and a vacuolar neuronopathy of dorsal root ganglion (DRG) occurs. The vacuoles are mitochondrial; we demonstrated Histochemical studies on the ischemic DRG under oxidative stress shows caspase-3 immunostaining, invoking an apoptotic mechanism. We have quantitated vacuolar degeneration and pigmentary degeneration of mitochondria and demonstrated that these are increased in the diabetic state and reduced by treatment with
a-lipoic acid
Reperfusion aggravates nerve ischemic fiber degeneration, by the generation of reduced oxygen species. We evaluated if racemic
a-lipoic acid, a potent antioxidant, will protect peripheral nerve from reperfusion injury, using our established model of ischemia reperfusion injury in the rat. Ischemia was produced by the ligature of each the supplying arteries to the sciatic-tibial nerve of the right hind-limb for a predetermined periods of time (either 3 or 5 hours), followed by the release of the ligatures, resulting in reperfusion. a Lipoic acid LA was given intraperitoneally daily for 3 days for both pre and post surgery. Treatment resulted in improvement in nerve electrophysiology, and histologically for the nerve subjected to 3 but not 5 hours of ischemia.
There is also preliminary data on the efficacy of
a-lipoic acid in the treatment of other experimental neuropathies, such as hexacarbon neuropathy and acrylamide induced degeneration. While the mechanism of improvement in these "toxic" neuropathies are uncertain, potential mechanisms include its effect in enhancing glucose transport and reducing the energy crisis induced by the actions of these drugs on glycolysis are candidate mechanisms.

Metabolic and vascular abnormalities in diabetic ocular
disorders: Implications for antioxidant therapy

Irina Obrosova*, Robert Frank#, Martin Stevens*, Douglas Greene*

*University of Michigan, Ann Arbor, and Wayne State University,
Detroit, MI, USA

Oxidative stress has an important role in diabetic retinopathy and cataract. In the diabetic retina, oxidative stress has been implicated in capillary cell loss, decreased blood flow, increased vascular permeability, and upregulation of vascular endothelial growth factor (VEGF). In the diabetic lens, oxidative stress has a key role in post translational modification of crystallins and their crosslinking ultimately resulting in opacification. These irreversible changes are preceded by metabolic abnormalities (i.e. sorbitol acumulation and loss of major non-enzymatic antioxidants, GSH, ascorbate and taurine, NAD- and NADP-redox imbalances, energy deficiency, down-regulation of membrane transporters and ATP-ases) most of which can be prevented or delayed by antioxidant therapy. Our studies indicate that the potent antioxidant, DL-(-lipoic acid (100 mg.kg-1.d-1, intraperitoneally), effectively counteracts oxidative stress in the lenses and retinas of diabetic rats. In the lens, DL-(-lipoic acid prevented diabetes-induced loss of GSH, ascorbate and taurine, the decrease in free cytosolic NAD+/NADH ratio, inhibition of glycolysis and energy failure. In the retina, DL-a-lipoic acid arrested diabetes associated 4-hydroxyalkenal accumulation and downregulation of superoxide dismutase activity, without affecting either GSH and GSSG concentrations or enzymes of glutathione metabolism. In addition, the antioxidant prevented decrease in free mitochondrial (a metabolic correlate of retinal hypoxia) and cytosolic NAD+/NADH ratios. Another antioxidant, taurine (5 % of the diet), reduced (lens) or prevented (retina) lipid peroxidation in diabetic ocular tissues. In contrast to DL-(-lipoic acid, taurine did not affect lens GSH levels and did not prevent a number of metabolic imbalances associated with diabetic cataract formation. Of particular interest is the observation that taurine (1% of the diet), that completely arrested retinal accumulation of malondialdehyde(MDA) plus 4-HA, attenuated, but did not prevent, diabetes-induced upregulation of vascular endothelial growth factor. This suggests that oxidative stress is important, but is not the only mechanism for increased VEGF formatiob in diabetes. Vitamin E (1% of the diet, which significantly reduced lipid peroxidation in the diabetic nerve) did not prevent accumulation of MDA plus 4-HA or VEGF in the diabetic retina.

Role of a-lipoic acid in the treatment of diabetic
polyneuropathy: Current evidence from the clinical trials

Dan Ziegler

Diabetes Research Institute at the Heinrich Heine University,
Düsseldorf, Germany

Near-normoglycaemia is now generally accepted as the primary approach to prevention of diabetic neuropathy, but is not achievable in a considerable number of patients. In the past two decades several medical treatments that exert their effects despite hyperglycaemia have been derived from the experimental pathogenetic concepts of diabetic neuropathy. Such compounds including a-lipoic acid have been designed to improve or slow the progression of the neuropathic process and are being evaluated in clinical trials. We review the current evidence from the clinical trials that assessed the therapeutic efficacy and safety of a-lipoic acid in diabetic polyneuropathy. Thus far, 15 clinical trials have been completed using different study designs, durations of treatment, doses, sample sizes, and patient populations. Within this variety of clinical trials, those with beneficial effects of a-lipoic acid on either neuropathic symptoms and deficits due to polyneuropathy or reduced heart rate variability resulting from cardiac autonomic neuropathy used doses of at least 600 mg per day. The following conclusions can be drawn from the recent controlled clinical trials. 1) Short-term treatment for 3 weeks using 600 mg of a-lipoic acid i.v. per day appears to reduce the chief symptoms of diabetic polyneuropathy. A 3-wee k pilot study of 1800 mg per day indicates that the therapeutic effect may be independent of the route of administration. 2) This effect is accompanied by an improvement of neuropathic deficits. 3) Oral treatment for 4-7 months tends to ameliorate neuropathic deficits and cardiac autonomic neuropathy. 4) Preliminary data over 2 years indicate possible long-term improvement in motor and sensory nerve conduction in the lower limbs. 5) Clinical and postmarket ing surveillance studies have revealed a highly favourable safety profile of the drug. Based on these findings, a pivotal long-term multicenter trial of oral treatment with a-lipoic acid is being conducted in North America and Europe aimed a slowing the progression of diabetic polyneuropathy using a clinically meaningful and reliable primary outcome measure that combines clinical and neurophysiological assessment.

Insulin signals regulating glucose uptake: towards understanding acquired insulin resistance in diabetes

Amira Klip

Research Institute, The Hospital for Sick Children, Toronto

Insulin resistance is a primary cause of type 2 diabetes, but the molecular defects that collude to bring about this condition are still unknown. In addition, once hyperglycemia occurs (in either type 1 or type 2 diabetes), there is a further, secondary or acquired insulin resistance. How hyperglycemia begets insulin resistance is unknown, but the oxidative stress caused by high glucose levels is one suspected cause. Therefore, there is a need to understand which steps in insulin action are prone to oxidative damage, and to explore strategies to relieve oxidative stress in general and the specific molecular defects in particular. Skeletal muscle is the primary tissue responsible for glucose uptake and therefore a main contributor to the maintenance of glycemia. Insulin resistance of muscle glucose uptake is well documented in both type 1 and type 2 diabetes. Over the past five years we have contributed to the understanding of insulin action leading to stimulation of glucose uptake in muscle cells. We defined that the enzyme phosphatidylinositol 3-kinase is required to mobilize glucose transporters in muscle cells from intracellular stores to the plasma membrane, and to maintain the insulin stimulated state. We further observed that the products of this enzyme, phosphatidylinositol phospholipids, cause glucose transporter recruitment to the cell surface, and that this action involves the serine/threonine kinase Akt/protein kinase B. More recently, we uncovered a new action of insulin, that of increasing the activity of the glucose transporters translocated to the plasma membrane. This action appears to require the p38 MAPK kinase. Lipoic acid, an agent with antioxidative properties, was able to mimic certain actions of insulin in muscle and fat cells in culture, specifically the phosphorylation of the insulin receptor substrate-1, the activation of phosphatidylinositol 3-kinase and of Akt, to result in an increase in glucose uptake. We are currently testing the effect of lipoic acid on the p38 MAPK pathway and glucose transporter activation. The studies to date suggest that lipoic acid has the capacity, at least in cell culture, to promote insulin-like actions, and may therefore be an important adjunct in the treatment of insulin resistance accompanying diabetes.

Increased oxidative stress in the aging rat heart is reversed by dietary supplementation with (R)-a-lipoic acid

Tory M. Hagen, Jung H. Suh and Eric T. Shigeno

Linus Pauling Institute, OSU, Corvallis, Oregon 97331

Little is known about the extent or nature of mitochondrial decay in the aging heart or whether this decay results in increased oxidative stress. Cardiac myocytes from young (2 months) and old (28 months) F344 rats were incubated with Rhodamine 123 (R123) to determine whether mitochondrial membrane potential becomes altered with age. Cardiac myocytes isolated from old rats exhibited 54 12% lower R123 staining than cells from young rats (p= 0.03), suggesting that mitochondrial electron transport efficiency declines with age. To determine whether this age-related change results in increased myocardial oxidative stress, oxidant production, antioxidant status and oxidative DNA damage were measured. Cardiac myocytes isolated from old rats showed a nearly 2-fold increase in the rate of oxidant production and a 2-fold decline in ascorbate levels (p=0.03) compared to young rats. No age-related change in alpha-tocopherol status was evident. Oxidative DNA damage, as measured by 8-oxo-2'-deoxyguanosine increased significantly (p=0.054). To investigate whether (R)-alpha-lipoic acid (LA) was effective at reducing oxidative stress, young and old rats were fed an AIN-93M diet with or without 0.2% (w/w) LA for two weeks prior to sacrifice. Cardiac myocytes from old, LA supplemented rats exhibited a markedly decreased rate of oxidant production, increased ascorbate levels, and a concomitant decline in oxidative DNA damage, which was no longer significantly different from that seen in cells from unsupplemented or supplemented young rats. Our data indicate that the aging rat heart is under increased mitochondrial-induced oxidative stress, which is significantly attenuated by lipoic acid supplementation.

Workshop on
Oxidative and Nitrative Stress in Inflammation

Organized by
Angelo Azzi
Lester Packer
Catherine Pasquier
Helmut Sies

Co-Sponsored by
UNESCO_MCBN (Global Network of Molecular & Cell Biology)

Session VII
Skin Disorders and Aging

Redox regulation of T-cell recruitment into the skin

M. Podda, H.A. Beschmann, R. Kaufmann, U.H. von Andrian¶, and T.M. Zollner

Dept. of Dermatology, University of Frankfurt, Germany and ¶CBR, Harvard Medical School, Boston, USA

Upon activation in lymphoid tissues associated with specific organs, lymphocytes acquire a predilection to migrate preferentially to these organs. The interaction of selectins with their carbohydrate ligands is responsible for initial tethering of blood leukocytes to the vessel wall and a central regulatory step in tissue specific T cell recruitment. Although leukocyte recirculation functions in immune surveillance, excessive trafficking to extravascular locations can lead to serious tissue injury and destruction. With respect to psoriasis vulgaris and atopic dermatitis, we and others have suggested a role of inappropriate T-cell skin homing, via induction of T-cell cutaneous lymphocyte-associated antigen (CLA). CLA is produced by post-translational glycosylation of the constitutively expressed P-selectin glycoprotein ligand-1 (PSGL-1) through fucosyltransferase VII (FucT-VII). FucT-VII and CLA expression can be upregulated by various specific T-cell stimuli. N-acetylcyst eine and various other antioxidants are capable to inhibit this upregulation. This inhibition is of functional relevance as shown by in vitro E-selectin binding, the ligand for CLA. Furthermore, using the in vivo mouse ear model a marked inhibition of T-cell rolling in postcapillary venules could be demonstrated by intravital microscopy. The antioxidants act presumably via interaction with the activation of the transcription factor NF-kB as other chemically unrelated NF-kB inhibitors showed comparable effects. We then asked if oxidants would be capable to induce CLA expression by T-cells. Indeed, of all oxidants tested only 3-morpholinosydnonimi ne (SIN-1) showed a marked induction of CLA on T-cells and an increase in E-selectin binding. The redox-regulation of CLA expression on T-cells could be a promising site for therapeutic intervention in skin inflammatory processes.

Photobiological and molecular mechanisms of ultraviolet A radiation-induced gene regulation

Jean Krutmann

Clinical and Experimental Photodermatology, Department of
Dermatology, Heinrich-Heine University, D-40225 Düsseldorf, Germany

Solar UVA radiation contributes to photoageing and photocar cinogenesis of human skin and is responsible for triggering the most frequent photodermatoses, that is polymorphous light eruption (PLE). Human skin is additionally exposed to UVA radiation from artificial sources including high-intensity UVA irradiation devices, which in former years have mainly been used for cosmetical purposes, but are currently employed at an increasing rate for the treatment of inflammatory skin diseases, such as atopic dermatitis, urticaria pigmentosa, and connective tissue diseases. In order to better understand the impact of UVA radiation on human skin, it is necessary to analyze the photobiological and molecular mechanisms responsible for UVA radiation-induced gene regulation.
The growing list of UVA radiation-inducible genes includes the heme-oxygenase-1 gene (HO-1), various cytokines, the adhesion molecule Intercellular Adhesion Molecule-1 (ICAM-1), matrix metalloproteinases, the FAS-Ligand molecule (CD95-L), and CL100, a nonreceptor type protein-tyrosine phosphatase. Ultraviolet A radiation-induced expression of the HO-1, MMP-1, ICAM 1, FAS-L involve the generation of singlet oxygen. For example, UVA radiation-induced ICAM-1 expression in human epidermal keratinocytes and UVA radiation-induced FAS-L upregulation in skin-infiltrating human T-helper cells were inhibited by singlet oxygen quenchers, enhanced under conditions that increased singlet oxygen half-life, and could be mimicked by using a singlet oxygen-generating system. Singlet oxygen thus seems to be the primary effector in UVA radiation-induced gene expression.
Ultraviolet A radiation and singlet oxygen-induced promoter activation have recently been studied by using promoter constructs based on the human ICAM-1 promoter. In these studies, we have found that transcription factor AP-2 mediates UVA- and the singlet oxygen-induced gene expression in human keratinocytes. The critical role of AP-2 in UVA radiation-induced gene expression is further emphasized by studies in which the functional role of AP-2 and its alternative splice product, AP2-B, was studied for UVA radiation-induced gene expression. Studies addressing the functional relevance of AP-2B have revealed that it serves as a dominant negative regulator of AP2 transactivation. This concept is in line with the recent observation that cells that had been stably transfected to overexpress AP-2 showed a 1000-fold increased ICAM-1 expression, as compared to the parental clone. Increased and sustained expression of proinflammatory genes, such as ICAM-1, has been observed immunohistochemically in photoprovocation-induced PLE lesions. We have recently observed that the development of skin lesions in PLE patients was associated with a sustained (> 3 days) and dramatically increased upregulation of ICAM-1 mRNA (up to 40-fold), as compared to normal skin (maximum: 4-fold, downregulated after 1 day). Transient ICAM-1 induction in normal skin was accompanied by a transient increase in both AP2 and AP2B expression. In striking contrast, in PLE skin, AP2B expression was completely absent from unirradiated as well as UVA-irradiated skin areas, whereas AP2 was present constitutively and increased after UVA radiation. These studies demonstrate the first qualitative difference between PLE and normal skin. We propose that an imbalance in the AP2/ AP2B system forms the pathogenetic basis of PLE. Recent studies have also demonstrated that ceramides are part of the signal transduction pathway operative in UVA radiation-induced, AP-2-mediated gene expression. We have demonstrated a previously unrecognized role of ceramides in AP-2 activation and UVA radiation-induced gene expression and have discovered a novel, nonenzymatic pathways, by which ceramides can be generated from sphingomyelin.

Bioflavonoid-rich extracts modify gene expression
in keratinocytes

Claude Saliou1*, Gerald Rimbach1, Bertrand Rihn1,
Laura McLaughlin1 and Lester Packer1

1Membrane Bioenergetics Group, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720

Several bioflavonoid-rich plant extracts have been used for centuries in traditional medicine to prevent and treat skin disorders. The objective of the present study was to seek evidence that flavonoid extracts interfere with intracellular transducers, such as Nuclear Factor (NF)-kB, that lead to changes in gene expression in keratinocytes exposed to UVR. HaCaT keratinocytes were exposed to a solar UV simulated light, gel retardation assays were performed to analyze NF-kB and AP-1 DNA binding activities, a reporter gene assay was used to monitor their transactivating activities, and mRNA levels for proinflammatory cytokines were measured with a quantitative PCR assay. A UV dose of 150 mJ/cm2 was used to induce NF-kB and AP-1 activation. Silymarin and the procyanidin-rich French maritime pine bark extract Pycnogenol (PBE), two flavonoid extracts, were able to abrogate UVR-induced NF-kB transactivating activity at the respective concentrations of 12.5 and 25 g/ml. Furthermore, UVR-induced increases in inter leukin-6 and interleukin-8 mRNA levels were thoroughly reduced by 25 g/ml of silymarin. In parallel, a cDNA array study allowed the identification of the calgranulin genes involved in chronic inflammatory disorders. The expression of both calgranulin A and B genes was strongly decreased by PBE.
Consequently, these results provide a new rationale, based on the changes in gene expression, for the chemoprevention of UVR induced skin diseases by natural flavonoid extracts.
*Current address: Johnson & Johnson Consumer Companies, Skillman, NJ 08558

Expression of inducible nitirc oxide synthase in human skin disease: adverse reaction or helping hand?

C.V. Suschek1, K. Briviba2, D. Bruch-Gerharz3, H. Sies,
K.-D. Kröncke1, and V. Kolb-Bachofen1

1Research Group Immunobiology, 2,Institute of Physiological Chemistry I, 3Department of Dermatology, Heinrich-Heine-University, D-40001 Düsseldorf, Germany

In a number of human skin diseases the expression of the inducible nitric oxide synthase (iNOS) is found. In addition, UVA irradiation of human skin will also lead to iNOS expression in keratinocytes and vascular endothelium. Due to its known cytotoxic activity NO may act as an inflammatory reaction contributing to UV-induced skin damage. To the contrary, we recently found that iNOS activation or exogenously applied NO administered 24 h prior to the light stimulus, protected endothelial cells from UVA-induced apoptosis and protection tightly correlated to NO-mediated increases in Bcl-2 mRNA and protein expression.
We now looked for the effect of NO - when present during or after the toxic insult - on cell death of endothelial cells (EC) induced by reactive oxygen species (ROS) generated by UVA- or Rose Bengal-treatment (RB) resulting in apoptosis (postUVA) or necrosis (postRB). We find that at concentrations
100 M the NO-donor S-nitroso cysteine (SNOC) significantly protects from apoptosis as well as necrosis (80% without vs. 10% apoptosis with 1 mM SNOC and 85% without vs. 5% necrosis with 1 mM SNOC). The NO-mediated protection is not due to scavenging of ROS but tightly correlates with complete inhibition of ROS-induced lipid peroxidation (6-fold increase in MDA formation without vs. 1.2-fold with 1 mM SNOC). In the case of apoptosis postUVA, cell death strongly correlates with caspase- and Bcl-2 independent inhibition of mitochondrial cytochrome c leakage into the cytoplasm, as key signal in apoptosis initiation.
The experiments presented here demonstrate a new protective mechanism of NO upstream of caspase activity and underline the protective capacity of NO against ROS as formed during UVA-irradiation as well as inflammatory processes.

Tocotrienols and tocopherols in the protection of the skin against environmental stressors

Stefan U. Weber, Giong Guo, Giuseppe Valacchi, Nancy Han,
Caroll E. Cross and Lester Packer

251 LSA, Department of Molecular and Cell Biology,
University of California at Berkeley, Berkeley, CA 94720-3200, USA

The skin is chronically exposed to environmental stressor such as ultraviolet radiation (UV) and ozone (O3). Oxidative damage to critical biomolecules is believed to be causally involved in skin aging and photocarcinogenesis. Vitamin E in the form of (-tocopherol is a commonly used antioxidant to protect cutaneous tissues from oxidants. Aims of the current studies were to investigate the mechanisms involved in O3-induced skin damage, and to explore the protective potential of vitamin E isoforms in this scenario.
It could be demonstrated, that O3 depletes antioxidants in the murine stratum corneum (SC), which forms the outer skin barrier. Since skin may be exposed to O3 and UV simultaneously in real life situations, a model of combined exposure was developed. It was shown that O3 potentiates the UV-induced vitamin E depletion in SC. A carbon centered lipid radical could be observed in SC upon UV and O3 exposure. The transepidermal water loss was not changed. However, O3 treated SC exhibited a reduced resistance to progressive tape stripping, indicating that O3 may latently compromise the mechanical integrity of the skin barrier. In the epidermis the proinflammatory cytokine IL-1
( was increased moderately in response to O3ess. Since vitamin E is usually applied as an acetate to increase its stability, it was analyzed if isoforms of vitamin E acetates are hydrolyzed in skin to release the active free vitamin E. In murine and human skin significant hydrolysis of d-,g , a-tocotrienyl acetate and g-, a-tocotrienyl acetate was observed using a newly developed HPLC method. The ability of (-tocotrien ol and (-tocopherol to protect skin from oxidative damage was tested in response to O3 and UV. In both cases the two isoforms reduced lipid peroxidation in the murine SC. (-Tocotrienol was as effective as (-tocopherol. Therefore, tocotrienols may be of interest for future applications as topical antioxidants.

Topically applied antioxidants in skin protection against
oxidative stress

Franz Stäb

PGU Skin Research Center, R&D Cosmed, Beiersdorf AG, Hamburg, Germany

For strategies of active skin protection against UV-light and ROS, the topical application of antioxidants becomes more and more prominent. However, only some of the in vitro effective antioxidants appear to be functional in vivo too. In our studies we compared physiological activities of diverse known antioxidants (e. g., propylgallate, pine bark extract, lipoic acid, vit. C, vit. E) versus the highly water soluble flavonoid alpha glucosylrutin (AGR). The studies were conducted in vivo on human skin and in vitro using primary cells from biopsies of human skin.
In vitro results. 1) AGR can bolster and protect the intracellular thiol level and the mitochondrial membrane potential of primary keratinocytes and fibroblasts against depletion by oxidative stress. 2) The oxidative stress associated ferritin synthesis and the UV induced intracellular generation of NO can be reduced by AGR. 3) AGR reduces the level of tyrosine phosphorylation in primary skin cells challenged by oxidative stressors (H2O2, UVA). 4) AGR protects primary human skin cells against both, UV-induced reduction in DNA- and hyaluronan- synthesis as well as against UV-induced intracellular hydroperoxide formation.
In vivo results. 1) Our studies on normal human skin using UVA-induced ultraweak photoemission measurements (UPE) confirm the antioxidative potential of topical AGR in vivo. 2) After UVA-irradiation in vivo, the phosphorylation of intracellular tyrosine residues and the hydroperoxide levels are significantly enhanced in keratinocytes isolated from donors suffering on polymorphous light eruption (PLE) compared with keratinocytes from healthy donors, but can be reduced by topical pretreatment with AGR-preparations. 3) Several clinical trials on PLE prophylaxis showed a significant improvement of the PLE score at skin sites pretreated with AGR compared with untreated or placebo treated skin sites.

Iron and heme release as key factors in the response of skin cells to UVA radiation

Tyrrell, R.M., Pourzand, C.A., Brown, J., Hejmadi, V.,
Kvam, E., Ryter, S. and R. Watkin

Department of Pharmacy and Pharmacology University of Bath,
Bath BA2 7AY UK

The UVA (320 _ 380nm) component of sunlight or sunbeds acts as an oxidising carcinogen and has been clearly implicated in skin cancer. Since UVA radiation interacts with cells by generating active oxygen species, the damaging effects of this radiation will be exacerbated by the presence of catalytically reactive iron in cells. We have now shown by two independent techniques (dequenching of metal-quenched calcein fluorescence in cells and changes in the binding activity of the iron responsive protein IRP1) that UVA radiation causes an immediate release of "free" iron in human skin fibroblasts via the proteolysis of ferritin(1). Within minutes of exposure to a range of doses of UVA at natural exposure levels, the binding activity of IRP-1, as well as the levels of intact ferritin complexes, decrease in a dose-dependent manner. This decrease coincides with a significant leakage of the lysosomal components into the cytosol. Stabilisation of Ft molecules occurs only when cells are pre-treated with lysosomal protease inhibitors after UVA treatment. We propose that the oxidative damage to lysosomes that leads to Ft degradation and the consequent release of potentially harmful "free" iron to the cytosol might be a major factor in UVA induced damage to the skin. UVA radiation also breaks down heme-containing proteins in the microsomal membrane to release free heme as an additional photosensitising component(2). This will provide another source of enhanced free iron in skin cells since constitutive heme oxygenase 2 (in keratinocytes) and UVA-inducible heme oxygenase-1 (in fibroblasts) are likely to break down any free heme to biliverdin and release iron and carbon monoxide in the process. Indeed heme oxygenase over-expressing cell lines are hypersensitive to UVA radiation when pre-incubated with heme (unpublished). We postulate that, in skin fibroblasts, this free heme release and the enhanced free iron pools will lead to an adaptive response involving heme oxygenase (with a maximum about 10h) and ferritin (in 24-48h) that will scavenge the heme and iron released by subsequent oxidising (UVA) treatments.

1. Pourzand, C. et al (1999). Proc. Natl.Acad.Sci.USA 96, 6751-6756.
2. Kvam, E. et al (1999) Free Radic.Biol.Med. 26, 511-517

This research has been supported by the UK Department of Health and the Association for International Cancer Research (UK)

Clinical implications for oxidative stress in the
human skin barrier

Jens Thiele

Department of Dermatology, Friedrich-Schiller-Universität, Jena, Germany

The stratum corneum (SC) is the outermost layer of human skin and thus is constantly exposed to a pro-oxidative environment. Previously, we have demonstrated that 1) the murine and human SC is equipped with a variety of antioxidants, of which vitamin E predominates1, 2) SC lipids are targets of oxidative stress induced by ozone and by UVA and UVB exposure2, and 3) sebaceous gland secretion is a major physiological route of vitamin E delivery to human skin3. In order to investigate the physiological role of protein oxidation in the skin barrier, we employed an immunoblotting technique to detect protein oxidation in human SC obtained by tape stripping4. After lysis, protein carbonyl groups were measured by derivatization with dinitrophenylhydrazine, separation by SDS-PAGE, and immunoblotting using antibodies against dinitrophenyl groups. Keratin 10, identified by use of specific antibodies and by microsequencing, was demonstrated in vitro to be oxidizable by UVA irradiation, hypochlorite, and benzoyl peroxide. In vivo, a keratin 10 oxidation gradient with low levels in the lower stratum corneum layers, and about 3-fold higher contents of carbonyl groups towards the outer layers was demonstrated in forehead stratum corneum of healthy volunteers. Since protein oxidation can be associated with an increased susceptibility to proteases, our findings may be important for a better understanding of the process of desquamation. Furthermore, due to the interactivity of the SC with subjacent cell layers chronic oxidative stress in the cutaneous barrier may be implicated in the pathophysiology of skin diseases and skin aging.

1. Thiele, J.J., Traber, M.G., Polefka, T.G., Cross, C.E. & Packer, L.P. J. Invest. Dermatol. 108, 753-757 (1997).
2. Thiele, J.J., Traber, M.G. & Packer, L. J. Invest. Dermatol. 110, 756-761 (1998).
3. Thiele, J.J., Weber, S.U. & Packer, L. J. Invest. Dermatol. 113, 1006-1010 (1999).
4. Thiele, J.J., Hsieh, S.N., Briviba, K. & Sies, H. J. Invest. Dermatol. 113, 335-339 (1999).

Science & Humanity Prize

Mitochondrial Cu,Zn superoxide dismutase revisited

Irwin Fridovich

Department of Biochemistry, Duke University Medical Center, Durham, North Carolina

In 1993 we examined chicken liver mitochondria and found a MnSOD in the matrix and a Cu,ZnSOD in the intermembrane space. In 1982, the presence of a Cu,ZnSOD in mitochondria was questioned by Geller & Winge who concluded that our results were due to contamination with lysosomes. To achieve assays of Cu, ZnSOD and MnSOD in mitochondrial extracts, we used CN_ to selectively suppress Cu,ZnSOD; while Geller & Winge used 2% SDS to selectively inactivate MnSOD. The cytosolic Cu,ZnSOD is stable to SDS. Could there be an SDS-sensitive Cu,ZnSOD in mitochondria? We now report data showing that there is and conclude that this represents the fourth eukaryotic SOD.

Nutraceutical Award

Dietary flavone is a potent apoptosis-inducer in
human colon carcinoma cells

U. Wenzel*, S. Kuntz, M. D. Brendel#, and H. Daniel

Institute of Nutritional Sciences, University of Giessen, 35392 Giessen, FRG, and #Third Department of Internal Medicine, University of Giessen, 35385 Giessen, FRG

Flavonoids are polyphenolic compounds, occuring ubiquitously in plants. They are discussed to represent cancer preventive food components in a human diet rich in fruits and vegetables. To understand the molecular basis of the putative anticancer activity of flavonoids, we investigated whether and how the core structure of the flavones, 2-phenyl-4H-1-benzopyran-4-one (flavone) affects proliferation, differentiation and apoptosis in HT-29 human colon cancer cells. Moreover, the effects of flavone in transformed epithelial cells were compared to those obtained in non-transformed primary mouse colonocytes. Proliferation, differentiation and apoptosis in transformed as well as non-transformed colon cells were measured by fluorescence based techniques. Apoptosis was in addition determined by changes in membrane permeability, FACScan analysis and detection of DNA fragmentation. Semi-quan titative RT-PCR was performed to assess the effects of flavone on transcript levels. Flavone was found to reduce cell proliferation in HT-29 cells with an EC50-value of 54.8 1.3 M and induced potently differentiation as well as apoptosis. The flavonoid proved to be a stronger apoptosis inducer than the clinically established anti tumor agent camptothecin. The effects of flavone in HT-29 cells were associated with changed mRNA levels of cell-cycle and apoptosis related genes including COX-2, NF-B and bcl-XL. Moreover, flavone, but not camptothecin, displayed a high selectivity for induction of apoptosis and growth-inhibition only in the transformed colonocytes.
In conclusion, the plant polyphenol flavone induces effectively programmed cell death, differentiation and growth-inhibition in transformed colonocytes by acting at the mRNA levels of genes involved in these processes. Since these genes play a crucial role in colon carcinogenesis, flavone might prove a potent new cytostatic compound with improved selectivity towards transformed cells.

Session VIII
Oxidative Stress, Inflammation, and Neurodegeneration

Regulation of cyclooxygenase activity by nitric oxide and nitric oxide-derived species

Lawrence J. Marnett, Lisa Landino, Brenda Crews,
Doug Goodwin, Jason Morrow, Theresa Wright,
and Steven Tannenbaum

Departments of Biochemistry and Medicine, Vanderbilt University School of Medicine, Nashville, TN 37232 and the Division of Bioengineering and
Environmental Health, Massachusetts Institute of Technology,
Cambridge MA 02139

Nitric oxide and nitric oxide-derived species have been reported to exert a range of effects on arachidonic acid metabolism to prostaglandins. The nature of the effect and its magnitude varies with the experimental system under investigation. Several groups have reported that NO or NO-derived species interact directly with the cyclooxygenase (COX) enzyme that oxidizes arachidonic acid to prostaglandin endoperoxide intermediates. We have found that NO reacts with tyrosyl radicals generated on COX during turnover and established by peptide mapping that the trapped radical arises from Tyr385, the catalytic tyrosine residue that oxidizes arachidonic acid. We also have demonstrated that peroxynitrite, the couping product of NO and superoxide, activates the cyclooxygenase activities of constitutive (COX-1) and inducible (COX-2) cyclooxygenases by serving as a substrate for the enzymes' peroxidase activities. Activation of purified enzyme is induced by direct addition of peroxynitrite or by in situ generation of peroxynitrite from NO and superoxide. Cu,Zn-superoxide dismu tase completely inhibits COX activation in systems where peroxy nitrite is generated in situ from superoxide. In the murine macrophage cell line RAW264.7, the lipophilic superoxide dismutase-mimetic agents, Cu(II) (3,5-diisopropylsalicylic acid)2, and Mn(III) tetrakis(1-methyl-4-pyridyl)porphyrin dose-dependently decrease the synthesis of prostaglandins without affecting the levels of inducible-NO synthase (iNOS) or COX proteins or by inhibiting the release of arachidonic acid.
We have extended these experiments to intact animals by monitoring the effects of targeted deletion of the iNOS gene on the formation of prostaglandins in vivo and ex vivo. Peritoneal macropha ges were obtained from control and iNOS-deficient mice and PGE2 quantified after stimulation with g-interferon and lipopolysaccharide to induce COX-2. Total nitrate and nitrite production was completely abolished in cells from iNOS-deficient animals compared to control cells. PGE2 formation by cells from iNOS-deficient animals was decreased 80% at 12 hr (0.85 + 0.90 ng/106 cells vs. 15.4 + 2.1 ng/106 cells, p < 0.01) and 74% (9.4 + 4.3 ng/106 cells vs. 36.8 + 4.1 ng/106 cells, p < 0.01) at 24 hr compared to cells from control animals. COX-2 protein expression was not significantly different in cells from control or knockout animals. Levels of PGE2 in the urine of iNOS-deficient mice were decreased 78% (0.24 + 0.14 ng/mg creatinine vs 1.09 + 0.66 ng/mg creatinine; p < 0.01) compared to control animals. In addition, the levels of urinary F2-isoprostanes, an index of endogenous oxidant stress, were significantly decreased in iNOS-deficient animals. In contrast, the levels of thromboxane B2 derived from platelets allowed to aggregate ex vivo were significantly increased in iNOS-deficient mice compared to wild-type mice. These studies support the hypothesis that NO and/or NO-derived species modulate cyclooxygenase activity and eicosanoid production in vivo.

Oxidative stress and inflammatory gene expression
alterations in neurodegeneration and neuroprotection:
Implications for Parkinson's and Alzheimer's diseases

Moussa B.H. Youdim, Edna Grunblatt, Yona Royak and
Silvia Mandel

Eve Topf and National Parkinson Foundation (US) Centers of Excellence For Neurodegenerative Diseases Research and Teaching and Department of Pharmacology, Technion-Faculty of Medicine, Haifa, Israel

The etiology of Parkinson's (PD) and Alzheimer's (AD) diseases remain elusive. However, a large body of biochemical evidence point to a pivotal role for oxidative stress (OS) and inflammatory processes. In PD the nigroiatal dopamine neurons of substantia nigra pars compacta (SNPC) degenerate progressively.We have shown that in brains obtained at autopsy from PD that the balance between reactive oxygen species (ROS) generation and their disposition in SNPC demised in favor of ROS accumulation. This would be expected considering that there is a progressive accumulation of iron, lipid peroxidation, and cytotoxic cytokines (TNF-alpha, IL1 beta and IL6) and reduction of mitochondrial complex I and depletion of reduced glutathione (GSH). Animal (rats and mice) models of PD employing 6-hydroxydopamine or MPTP confirm these findings and support the OS hypothesis. Furthermore we have shown that antioxidants and iron chelators pretreatment induce neuroprotection in these models . In order to have a better insight we have extended these studies to possible alterations in gene expression in brains of PD and of MPTP treated mice using cDNA expression array membranes (Clontech) and RT-PCR method for confirmation . MPTP induced increases in expression of some 40 genes and reduction of 5 genes associated with OS , inflammatory processes, glutamate and nitric oxide. Pretreatment of mice with the antiparkinson drug, R-apomorphine , a dopamine D1-D2 receptor agonist, shown by us to posses potent antioxidant and iron chalating properties, protected against MPTP induced neurodegen eration of dopamine neurons and normalized the MPTP induced gene expression . These data provide evidence for MPTP to induce gene expression and support our hypothesis for involvement of iron release which can initiate OS and inflammatory reactions leading to neuronal death in PD. Our studies go some way to implicate the clinical use of antioxidants, radical scavengers, iron chelators, non-steroid anti inflammatory drugs and monoamine oxidase B inhibitors as neuroprotective agents for PD and AD.

NF-kB role in stress reponse in CNS

K. Werrbach-Perez, K. Cole, J. Glasgow, O. Nesic, J. Qiu,
T. Toliver-Kinsky, M. Spies, and R. Perez-Polo

Department of Human Biological Chemistry & Genetics, University of Texas Medical Branch, Galveston, Texas, USA.77555-0652

It is our working hypothesis that both acute (head and spinal cord trauma and stroke) and chronic (aging and Alzheimers) impairments to the central nervous system (CNS) act in part via increased oxidative stress. Elevated oxidative stress in turn triggers stress response signal transduction pathways that may result in neuronal recovery or commitment to apoptosis. We have explored the NF-
kB transcription factor and its regulation in the CNS and found that basal levels of NF-kB are selectively elevated in the compromised CNS. We hypothesize that deficits result from transcriptional regulation of the ChAT, Bcl-family, and APE/Ref-1 genes via transcription factors. One such factor is NF-kB, an acute phase response factor particularly sensitive to oxidative stress. NF-kB DNA-binding consensus sequences are present in the promoters for the above-named genes. We have shown that NF-kB has anti-apoptotic and ChAT-suppressing properties in the CNS. We have shown that there are alterations in NF-kB-binding activity in aged brain and that the ChAT and Bcl-x genes have NF-kB sites in the promoters that acts as a suppressor and enhancer respectively.

Supported in part by NINDS grant NS-33228 and grants from the Shriners Burn Institute, the Clayton Foundation, the Spinal Cord Research Foundation and The Institute for Rehabilitation and Research.

Oxidative stress induces the transcription factors Sp1 and Sp3 prior to apoptotic cell death in cortical neurons

H. Ryu, K. Zaman, S. Chatterjee, and R.R. Ratan

Department of Neurology, Harvard Institutes of Medicine, Rm. 857,
77 Avenue Louis Pasteur. Boston, Mass 01225

Glutamate or glutamate analogs such as homocysteate (HCA) induce glutathione depletion and apoptotic cell death in embryonic cortical neurons. In this paradigm, apoptotic cell death can be suppressed by inhibitors of macromolecular synthesis. Here we report that glutamate (or HCA) induces three Sp1 family DNA binding complexes in immature embryonic cortical neurons. Gel shift assays using a consensus Sp DNA binding sequence and supershift analysis using subunit specific antibodies to distinct Sp family members revealed that one of these complexes contains Sp1 and the other two contain Sp3. Western Blot analysis and immunohistochemistry confirmed an increase in Sp1 and Sp3 protein levels in the nucleus of neurons in response to glutathione depletion. The increases in Sp1/Sp3 DNA binding, which were observed subsequent to glutathione depletion could be abrogated by a host of antioxidants. Of note, other methods of inducing oxidative stress, including peroxide addition, also induced Sp1/Sp3 DNA binding in cortical neurons. Altogether, these results suggest that Sp1/Sp3 may redox regulated transcription factors in cortical neurons.

Monoamine oxidase-knockout mice:
A model for oxygen toxicity and behavior

Jean C. Shih

Department of Molecular Pharmacology and Toxicology, School of Pharmacy, and Department of Cell and Neurobiology, School of Medicine,
University of Southern California, Los Angeles, CA

Monoamine oxidase (MAO) catalyzes the oxidative deamina tion of a number of biogenic amines with the production of hydrogen peroxide (H2O2). Two forms of MAO, A and B, have been found. MAO A prefers substrates such as serotonin (5-HT) and norepinephrine (NE), and is sensitive to the inhibitor clorgyline. MAO B prefers substrates such as phenylethylamine (PEA) and benzylamine and is sensitive to inhibitor deprenyl. Dopamine (DA) , tyramine, and tryptamine are common substrates for both isoenzymes. MAO A and B are made of two different polypep tides with 70% amino acid identity. The two genes are closely linked and they are located on X-chromosome. Both genes consist of 15 exons and they have an identical intron-exon organization, suggesting they are derived from same ancestral gene.
Studies of MAO A and B knock-out mice have clearly shown that MAO A and B have distinct functions in neurotransmitter metabolism and behavior. MAO A-deficient mice have elevated brain levels of 5-HT and NE and manifest aggressive behavior (Cases et al. Science 268 , 1763 (1995)) similar to men with deletion of MAO A. In contrast , MAO B-deficient mice do not exhibit aggressive behavior and only PEA levels are increased. MAO B-deficient mice are resistant to the neurodegenerative effects of the Parkinsonism-producing neurotoxin MPTP (Grimsby et al., Nature Genetics 17, 206 (1997)).
Recently , we have generated MAO A and B double knock-out mice. Interestingly, these mice showed a different phenotype, they are smaller at birth, and in the first month of their life. The levels of 5-HT, NE, DA, and PEA are increased. These mice show severe anxiety.
We have previously shown that H2O2 generated during MAO catalyzed oxidation of neurotransmitters may cause damage to mitochondrial DNA. MAO A and B have different distributions in brain and they have different development curves and change different with aging. These three types of MAO knock-out mice may provide a valuable model to study H2O2 toxicity.

Supported by NIMH grants R37 MH39085 (MERIT Award) RO1 MH37020, KO MH00796 (Research Scientist Award) and the Welin Professorship.

Air pollution induced-pulmonary inflammation:
desirable or not?

Frank J. Kelly

Centre for Cardiovascular Biology & Medicine, King's College London,
London, SE1 7EH, UK

There is growing concern about the health effects associated with increased levels of air pollution, particularly particulate matter with an aerodynamic diameter less than 10 mm. Exposure to such small particles is associated with a range of pulmonary responses including decreased lung function and respiratory symptoms. The mechanisms underlying these responses are unclear. A significant proportion of particulate matter in the urban environment is derived from diesel exhaust (DE). We therefore undertook a human exposure study with DE using investigative bronchoscopy to examine airway cellular responses. 15 healthy non-atopic volunteers were exposed to 300mg/m3 particles and air for one hour on separate occasions at least 4 weeks apart. Bronchoscopy was undertaken 6 hours post-exposure to obtain biopsy and lavage fluid samples. Exposure to DE was associated with increased expression of a range of adhesion molecules and an influx of neutrophils and mast cells to the lung wall and airways. An interesting, but unresolved, issue is whether such pulmonary inflammation is a detrimental or beneficial response. That is, is the recruitment of activated immune cells to the lung an unregulated step which contributes to tissue injury or is the arrival of these cells to be welcomed as they direct, or are part of, a defence response? A finding in favour of a protective benefit is the reduced glutathione (GSH) response in the lung lining fluid (LLF) compartment. Although particles react with, and deplete, GSH in an in vitro setting, the concentration of GSH in LLF increased following DE exposure. This increase in GSH may be part of a host defence mechanism in response to DE, mediated via the immune system.

Redox regulation of iNOS expression in
experimental septic shock

Anna Iannone, Cristina Rota, Stefania Bergamini and Aldo Tomasi

Department of Biomedical Sciences, University of Modena, Modena, Italy

Following the administration to rats of purified Escherichia coli lipopolysaccharide (LPS), nitric oxide (NO) production increases within three hours This model system mimics the severe pathology observed in human septic shock, whose mortality rate is higher than 50 per cent of the cases.
Methods: Using electron spin resonance spectroscopy- we followed the kinetic of NO increase in the blood and liver for 6 hours. The test takes advantage of NO specific and diffusion-limited reactivity towards chelated iron (including heme) and SH groups.
Inducible NO synthase (iNOS) the probable source of NO overproduction was measured on circulating monocytic cells (PBMC) using specific antibodies and flow cytometry and in hepatocytes, using specific antibodies and confocal microscopy. INOS RNA messenger and NF-B, a redox regulated factor that activates many inflammatory cytokines, was also determined-
Results: At the onset of shock, a massive increase in NO-Hb was observed NO blood levels increase up to 12 hours following LPS.
The hypothesis that iNOS synthesis is regulated by intracellular redox equilibria, in particularly by the activation of NF
kB, we modified the intracellular redox environment by administering n-ace tyl-cysteine (NAC). This resulted in the diminution of the NO-HB adduct and prevented the formation of hepatic dinitrosyl-iron complexes, indicating an effective inhibition of NO synthesis. Consistently, PBMC iNOS, iNOS RNA messenger and NF-B resulted significantly increased in shock, and inhibited by NAC administration. We have demonstrated that treatment affecting the redox equilibrium causes a diminution of NO level, acting by inhibiting NO synthases. However, NAC will also react directly with NO giving rise to long-lived nitroso-thiols. IN fact, when NAC was administered after LPS administration, no NO inhibition was noted.
These two different pathways will probably give rise to contrasting effects: NO inhibition will prevent peroxynitrite formation, while nitrosothiols will prolong NO availability for reaction with superoxide. A possible mechanism of protection exerted by SH-group manipulation, has to take into account these two different mechanisms of action.

Modified lipoproteins, antioxidants, and
endothelial activation

Weijian Zhang and Balz Frei

Linus Pauling Institute, Oregon State University,
571 Weniger Hall, Corvallis OR 97331

Endothelial activation leading to increased expression of cellular adhesion molecules and monocyte chemotactic peptide-1 (MCP-1) is one of the earliest events in atherosclerosis. Modified lipopro teins and inflammatory cytokines have been suggested to activate endothelial cells by a redox-sensitive mechanism involving activation of the nuclear transcription factor NFkB. Therefore, we investigated the role of the small-molecule antioxidants ascorbate, glutathione (GSH) and a-lipoic acid (LA) in tumor necrosis factor a-induced NFkB activation and expression of E-selectin, vascular cell adhesion molecule-1, intercellular adhesion molecule-1 and MCP-1 in human aortic endothelial cells (HAEC). Neither ascorbate loading nor manipulation of cellular GSH levels or redox status (GS H:GSSG ratio) significantly affected NFkB activation and cellular adhesion molecule expression. In contrast, pre-incubation of HAEC with LA (50 M to 1 mM) inhibited adhesion molecule and MCP-1 expression in a dose-dependent manner, as well as adhesion of monocytic THP-1 cells. Furthermore, LA dose-dependently inhibited NFkB activation by preventing degradation of its inhibitory subunits, IkB-a and -b. LA also inhibited the activity of IkB kinase. These data indicate that LA, in contrast to other aqueous, small-molecule antioxidants, inhibits cytokine-induced endothelial activation by inhibiting phosphorylation and degradation of IkB. We speculate that in these experiments, LA acted primarily as a metal chelator, not as an antioxidant. Preliminary data will be presented indicating that chelation of intracellular iron and, in particular, copper strongly inhibits tumor necrosis factor a-induced activation of HAEC.

Activation by Akt and NADPH binding studies of the leukocyte NADPH oxidase

Dang, M.-C., Hoyal, C., and Babior, B.M.

The Scripps Research Institute, La Jolla CA 92037

Activation of the leukocyte NADPH oxidase is accompanied by the phosphorylation of the oxidase subunit p47PHOX. Earlier work showed that the activating phosphorylation can be accomplished by protein kinase C. We now show that it can also be accomplished by Akt (protein kinase B). Incubation of cytosolic oxidase subunit p47PHOX with Akt results in the phosphorylation of serines 304 and 328. If Akt-phosphorylated p47PHOX is added to neutrophil membranes in the presence of cytosol and GTPgS, O2_ is produced at twice the rate seen with protein kinase C-phosphorylated p47PHOX. O2_ production is transient, however, falling to 20% of the original rate after 1 minute. If phosphorylation is carried out with both Akt and protein kinase C, the rate of O2_ production is the same as seen with protein kinase C alone. This suggests that one or more of the serines phosphorylated by protein kinase C plays an inhibitory role.
Another cytosolic subunit is p67PHOX. This subunit binds NADPH, as shown by labelling experiments with NADPH dialde hyde and by the demonstration that the dialdehyde-labelled p67 PHOX is unable to support oxidase activity. Furthermore, p67 PHOX has a weak but real NADPH dehydrogenase activity. Recent experiments have shown that NADPH dialdehyde binds to the N-terminal TPR domains of p67PHOX. The functional relationship between NADPH binding and the activity of the oxidase remains to be determined.

Models for oxidative stress-induced carcinogenesis

Peter L. Gutierrez

Biological Chemistry and Molecular Biology, and Program of Oncology. University of Maryland Greenebaum Cancer Center. University of Maryland School of Medicine

Widespread genomic instability is a hallmark of tumor cells. Oxidative stress caused by reactive oxygen species can lead to genomic instability resulting in the formation of a mutator phenotype. Ames has estimated that 1/3 of the cancers in third world countries can be attributed to chronic infections. Examples of oxidativeess induced cell transformation and malignancy will be discussed. In our laboratory we have attempted to develop a model for human breast cancer based on oxidative stress. Breast cancer is not commonly associated with pre-existing chronic immflamatory conditions, unlike bowel cancer where a causal relationship between ulcerative colitis and bowel cancer has been suggested. Breast cancers are, however, commonly infiltrated with macro-pha ges, which could be responsible for the genetic changes that would transform a lesion from benign to malignant. We used H2O2 to stress the benign breast cell line MCF-10A with 450 M bolus exposures, or by gradually increasing the concentration at 50 M increments up to 450 M, or by chronically exposing the cells to superoxide generated by the Xanthine/Xanthine Oxidase system. These cell lines show typical transformation characteristics such as loss of contact inhibition, very large nucleated cells, aberrant mitotic figures, vacuoles and anchorage-independent growth in soft agar. These transformed cells have different characteristics in terms of resistance to H2O2, degrees of catalase activity and amount of the mutagenic DNA lesion 8-OHdG. There are also a variety of genetic changes including enzymes that repair DNA mismatches and genes that inhibit apoptosis. These latter changes contribute to H2O2 resistance, and thus are involved in propagating the transformed phenotype. Consequences of this transformation will be discussed.

Tetramer-destabilizing mutations in human manganese superoxide dismutase

Daniel Hernandez, Elizabeth LaGuardia, and Joe M. McCord

Webb-Waring Institute, University of Colorado Health Sciences Center, Denver, CO 80262

In studying the Jurkat human leukemia cell line it was noted that the cells contained much less MnSOD than normal lymphocytes. The electrophoretic mobility was different, suggesting the possibility of a mutation. A mutation was confirmed by cloning and sequencing the cDNA. Jurkat cells are heterozygous for a previously undescribed L60F mutation. The L60F cDNA was introduced into an expression vector, and the recombinant mutant enzyme was expressed in E. coli and purified. The L60F enzyme is active, but thermally less stable due to a weakened dimer-dimer interaction. We believe that the oxidative stress caused by this mutation may contribute to the transformed phenotype of Jurkat cells. The properties of L60F are similar to those of an I58T mutant described by Borgstahl et al.1 We proposed that similar tetramer-des tabilizing mutations might exist in the region including residues 40 65, especially in cancer cells, as a mechanism of providing the oxidative stress necessary for proliferation. Using denaturing HPLC we screened PCR products from genomic DNA of 50 human lung cancers. A third mutation was found, Y45C, confirmed by cloning and sequencing. Thus, three tetramer-destabilizing mutations are now known in human MnSOD, and two are confirmed to be from cancer cells. These mutations may reflect one mechanism whereby cancer cells maintain a degree of oxidative stress to drive their continued proliferation.

1. G. E. Borgstahl et al. Biochemistry 35:4287-4297, 1996.

Immune activation and apoptosis in HIV infection

Luc Montagnier

Institut Pasteur, Paris and Queens College, CUNY, New York

HIV infection and AIDS can be seen as a chronic viral infection of lymphoid tissues leading to a profound immune depression.
The main targets of HIV infection are CD4 T helper lymphocytes and monocytes, and one of the main and early feature of the infection is the rapid drop of the number of T helper cells secreting interleukine 2.
Another well established observation is the preapoptotic state of many lymphocytes, particularly, but not only, CD4+ T lymphocytes. This state, which can be detected by the presence of several markers (transglutaminase, Fas / Fas ligand) may be caused by the summation of several factors: interaction of viral proteins with uninfected cells, oversecretion of inflammatory cytokines, oxidative stress and shortage of interleukine 2.
The decrease of viral load that follows active antiretroviral therapy allows partial restoration of the immune system, with a rapid disappearance of preapoptotic cells, but some persistence of other activation markers.
It will be therefore, necessary to add some complementary treatments to achieve a complete (or almost complete) restoration of the immune system, while infection will persist at a low level.