Effects of singlet oxygen and UVA on MAP kinases in human dermal fibroblasts

Lars-Oliver Klotz, 1Freimut Schliess, Karlis Briviba, and Helmut Sies
Heinrich-Heine-Universität Düsseldorf, Institut für Physiologische Chemie I and 1Klinik für Gastroenterologie, Hepatologie und Infektiologie, Moorenstr. 5, D-40225 Düsseldorf, Germany

Singlet molecular oxygen (1O2) can be produced in biological systems by photoexcitation upon exposure of endogenous photo sensitisers, such as flavins or porphyrins to UVA, or by chemi-excitation. Apart from being involved in the damaging effects of UVA radiation (320-400 nm) on cellular systems, 1O2 modulates gene expression and herein mimics UVA radiation: Heme oxyge-nase-1, interstitial collagenase, and intercellular adhesion molecule-1 have been shown to be induced by UVA and 1O2.

We here demonstrate that UVA enhances c-Jun-N-terminal kinase (JNK) activity in human skin fibroblasts. Exposure of cells to UVA (300 kJ/m2) led to a 5-fold induction of JNK-activity which was significantly increased in the presence of D2O, an enhancer of the lifetime of 1O2. Sodium azide and imidazole, quenchers of 1O2, abolished the activation of JNK. The hydroxyl radical scavengers mannitol and DMSO had no effect when present during irradiation. Furthermore, photochemically produced 1O2 (Rose Bengal (RB) plus white light) was found to induce JNK activity. Again, this was enhanced by D2O and inhibited by azide (1). Likewise, p38/HOG1, is dually phosphorylated upon treatment with UVA or RB + light. The ERK 1 and 2 MAP kinases, however, are neither phosphorylated nor activated as measured by Western blotting and in immuno-complex kinase assays. A hydrophilic source of 1O2, the 1,4-endo-peroxide of 3,3“-(1,4-naph thylidene) dipropionate, neither leads to the activation of JNKs, p38 nor of ERKs, suggesting the activation is by intracellular 1O2 (2). JNK and p38 activation are known to lead to the recruitment of transcription factor AP-1, which may therefore play a pivotal role in 1O2- and UVA-signalling.

(1) Klotz, L.O., Briviba,K., and Sies,H. (1997). FEBS Lett. 408, 289-91.
(2) Briviba, K., Klotz, L.O., and Sies, H. (1997). Biol. Chem. 378, 1259-65.


Serum a-tocopherol and selenium and the risk of coronary heart disease

Paul Knekt1, Antti Reunanen1, Arpo Aromaa1, and Jukka Marniemi2
1National Public Health Institute, Helsinki, 2Social Insurance Institution, Turku, Finland

Oxidation of lipoproteins is an important step in the pathogenesis of atherosclerosis and therefore a sufficiently high antioxidant status may protect against atherosclerotic complications. Serum levels of a-tocopherol and selenium were studied for their prediction of coronary heart disease in a case-control study nested into a longitudinal population based survey of 2 587 Finnish men, 45 - 64 years of age, and free from heart disease. Altogether 279 incident cases of coronary heart disease occurred during a maximum follow-up of 13 years. Two controls per case, altogether 547 men, were chosen by individual matching for age and municipality. Serum a-tocopherol and selenium concentrations were determined from serum samples stored at the baseline examination at -20šC. Serum level of a-tocopherol was non significantly inversely associated with the risk of coronary hearth disease. The lipid-adjusted relative risk (95% confidence interval) between the highest and lowest quintiles of the distribution was 0.71 (0.36 - 1.38). No association was observed for serum selenium; the corresponding relative risk was 1.10 (0.58 - 2.06). Among men with a high level of serum cholesterol (> 300 mg/dl), however, there was a significant inverse association also for serum selenium. In that sub population, the relative risk between individuals in the highest tertiles of both micronutrients in comparison with those in the lowest one was 0.17 (0.03 - 0.86). The findings are in line with the hypothesis that the antioxidants a-tocopherol and selenium at high levels of low density lipoprotein provide protection against atherosclerosis.


Ginkgo biloba extract (EGb 761) protects HT-4 neuronal cells against glutamate induced loss of cell viability

Michael S. Kobayashi, Derick Han, Chandan K. Sen, Sashwati Roy, and Lester Packer
Department of Molecular and Cell Biology, University of California at Berkeley

EGb 761, an extract of Ginkgo biloba, has proven useful in some cerebral pathologies, and helps alleviate the symptoms associated with numerous cognitive disorders. EGb 761 also has been shown to have potent antioxidant properties, both in vivo and in vitro perhaps because of its high flavonoid, terpenoid, and organic acid composition. Elevated levels of extracellular glutamate is known to have cytotoxic effects on neuronal cells, and has been linked to several brain disorders. We have recently shown (Am J Physiol, 273: R1771, 1997) that in C6 glial cells 10 mM glutamate treatment decreases cellular glutathione levels and increases the level of intracellular peroxides. Antioxidants such as a-tocopherol and a-lipoic acid protected cells against glutamate cytotoxicity. Thus, the effects of EGb 761 treatment on glutamate cytotoxicity was investigated.

We observed that glutamate (10 mM, 12h) induces loss (90%) of viability in HT-4 hippocampal neuronal cells. Simultaneous co-treat ment of cells with EGb 761 and glutamate protected against glutamate induced cell death in a dose-dependent manner. In cells treated with 100 µg/ml EGb 761, 70-80 % of cells maintained viability even when challenged with 10 mM glutamate. Reduced glutathione (GSH) levels were measured from cells using HPLC-EC. Glutamate treatment for 12 h resulted in a 90-95% decrease in the level of intracellular GSH compared to glutamate non-treated controls. Although treatment of cells with EGb 761 protected against glutamate induced loss of viability, the treatment did not influence the loss of intracellular GSH in response to the challenge. This suggests that EGb 761 protects HT-4 neuronal cells against glutamate induced cytotoxicity by a GSH- independent mechanism.


Regulation of Inducible nitric oxide synthase expression in IFN-g-treated macrophages by Bio-Normalizer

Hirotsugu Kobuchi and Lester Packer
Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory, Membrane Bioenergetics Group, Department of Molecular and Cell Biology, University of California, Berkeley, California 94720-3200 USA

Bio-Normalizer, a health food supplement prepared from Carica papaya has been reported to have therapeutic effects in a variety of human pathologies. We recently reported that it protected isolated rat hearts from either ischemia-reperfusion injury or from the effect of a peroxyl radical initiator in a supplementation study. However, despite accumulating reports on the practical advantages of Bio-Nor malizer, little is known concerning the molecular mechanisms involved in its action. In this study, we investigated whether Bio-Normalizer could affect nitric oxide (NO) production, measured as accumulated nitrite in medium, and inducible nitric oxide synthase (iNOS) expression, capable of affecting a multitude of mechanisms in the immune response. The efficacy of Bio-Normalizer on NO production was investigated in a chemical and cellular system. Bio-Normalizer failed to interact with NO produced by sodium nitroprusside. Mouse RAW 264.7 macrophages did not produce NO constitutively, nor was NO production induced by treatment with Bio-Normalizer alone. However, a major increase of NO production from macrophages was observed after treatment with Bio-Norma lizer plus interferon (IFN)-g. This effect of Bio-Normalizer was dose-dependent. Such effects of Bio-Normalizer on NO production were not due to changes in the activity of iNOS. Reverse transcrip tion-polymerase chain reaction analysis revealed that the increase in NO production corresponded to an increase in the accumulation of iNOS mRNA without any changes in mRNA stability. Furthermore, Bio-Normalizer augmented the IFN-g-dependent expression of mRNA both of tumor necrosis factor-a and interleukin-1b which might also be implicated in the immune system. These results suggest that Bio-Normalizer itself does not provide a signal that triggers induction of the NO pathway, but shows synergistic interaction with IFN-g to induce NO synthesis. These properties of Bio-Normalizer may be in part responsible for its reported therapeutic activity against various diseases.


Aging versus the concept of reliability: Qualification and qantification of the antioxidant intervention

Vitali Koltover
Institute of Chemical Physics, RAS, Chernogolovka, Moscow Region, Russia

In spite of the vast complexity of the aging mechanisms, the entire aging process is governed by relatively simple quantitative laws: (i) there is the exponential mortality growth with age; (ii) there is a maximum life-span potential for each species; (iii) the species-specific life-span potentials strongly correlate with the resting specific metabolic rates. These laws are easy explained on the basis of the theory of reliability, suffice it to arrange the limit reliability levels for the cell molecular machines. The random malfunctions of the mitochonrial redox enzymes are of the first importance since they produce active free radicals of oxygen. In part, the reliability theory predicts an essential increase in production of oxyradicals when catalytic conditions for the mitochondrial enzymes tune away from the physiological optimum. The heart mitochondria indeed become intensive generators of superoxide after transient anoxia/ischaemia conditions. The loss of control of the electron flow was found to be caused by increase in the membrane lipid fluidity. The concept of reliability also serves for searching the life-prolongation substances and for studying their real action mechanisms. The capability of the simple antioxidants, like well-known gero-protector BHT (butylated hydroxytoluene), to compete with anti-peroxidative enzymes (like SOD) for reactive forms of oxygen seems to be unthinkable, considering the real antioxidant concentrations and their rate constants in situ. Investigating the BHT-induced changes in the whole-tissue EPR signals of rats of different ages and in the concentrations of hormones (ACTH, 11-OHCS, TSH, T3) in the rats¹ blood, we suggested a new antioxidant mechanism. Namely, the antioxidants can indirectly decrease the level of oxygen radicals in situ by improving the supply of mitochondria with oxygen via the NO/cGMP/endocrine system.


Superoxide and nitric oxide production by leukocytes of diabetic patients. Effects of natural antioxidant bio-normalizer (BN)

L. Korkina*, E. Ostrachovitch§, E. Michal'chik*, and I. Afanas'ev§
*Russian Institute of Pediatric Hematology; §Vitamin Research Institute, Moscow, Russia

This study was designed to determine the effects of BN, a natural functional food produced by fermentation of papaya fruits and some tropical herbs, on clinical conditions, metabolic parameters, and oxidant/antioxidant status of diabetic patients. The randomized, double blind, placebo-controlled clinical trials have been carried out in Russian Center of Diabetes. At the first stage of the study, two groups of infection free patients with insulin-dependent diabetes mellitus of > 10 years duration (group 1, IDDM, n = 30) and insulin-independent DM of < 2 years duration (group 2, IIDM, n = 26) were investigated for evidence of imbalance of O2.­/NO production by circulating polymorphonuclear and mononuclear leukocytes. Healthy controls (group 3, n = 20) were matched for age and sex. We examined the NO release from freshly isolated and PMA- or zymosan-activated leukocytes by measuring the nitrite (NO2­) levels by Griess reaction. O2.­ production was examined by the ferrocytochrome c and lucigenin chemiluminescence assays. Activities of leukocyte CuZnSOD, MnSOD and catalase were determined in the ethanol-chloroform cell extracts. Results were analyzed according to type of diabetes, glucose control, duration of disease, and presence of vascular complications. The diabetic patients had an impaired oxidant/antioxidant balance, which depended mainly on type of diabetes and duration of disease. Thus, IDDM characterized by suppressed NO and O2.­ production and MnSOD and catalase activities. IIDM leukocytes produced either highly elevated levels of both O2.­ and NO or decreased level of O2.­, which had a negative correlation with duration of disease. We found that the BN but not placebo administration (6 g a day for 1 month) improved significantly clinical conditions (especially, vascular complications and wound healing) in a more than 80% patients and their metabolic parameters decreasing the main DM markers in patients' blood and urine. A daily requirement of insulin decreased in average by 20-25%. Besides, the course of BN therapy led to the restoration of practically normal level of O2.­ and NO production by cells. BN increased substantially activities of IDDM MnSOD and catalase and did not change CuZnSOD. To evaluate the mechanism of clinical efficacy, BN was used in the experimental model of streptozotocin-induced DM as well. Using specific antibodies and immunohistochemical methods, we found that BN normalized frequency of iNOS-positive cells in islets and brain from acutely diabetic rats.


Critical interactions of natural antioxidants during low density lipoprotein oxidation

Joćo Laranjinha*,+, Otilia Vieira*, and Enrique Cadenas+
*Center for Neurosciences and Fac. Pharmacy, University of Coimbra, Portugal, +University of Southern California, Los Angeles, CA, USA

Strong evidence suggest that oxidized low density lipoproteins (LDL) critically contribute to atherogenesis and that lipid peroxidation plays a key role in its development. This ³LDL oxidation theory² led to the search for efficient antioxidants obtained either from natural or synthetic sources. Although much work has been done with individual compounds, the potential dynamic interactions among antioxidants is still unclear and more relevant in a biological milieu.

We have previously shown that caffeic acid (dietary dihydroxy-cinna mic acid derivative) was able to regenerate a-tocopherol at LDL surface, supporting a synergistic inhibition of LDL oxidation. On the other hand, data from HPLC and UV absorption spectroscopy revealed that addition of ascorbate during caffeic acid oxidation by ferrylmyoglobin resulted in the regeneration of caffeic acid. In the LDL system, caffeic acid and ascorbate synergistically inhibited lipid oxidation, as inferred from the time courses of cholesteryl linoleate hydroperoxide and 7-ketocho lesterol formation. Moreover, when present simultaneously in the LDL reaction mixture, caffeic acid, a-tocopherol, and ascorbate afforded a protection against LDL oxidation stronger than the sum of the individual effects of the antioxidants, as evaluated by oxygen consumption. Although, independent reactions of caffeic acid and ascorbate with ferrylmyoglobin or a-tocopherol cannot be ruled out, these results pointed to an interaction of caffeic acid with ascorbate and a-toco pherol, encompassing the redox cycles of a-tocopherol and caffeic acid during LDL oxidation, being ascorbate the ultimate reductant.

EPR data obtained in an a-tocopherol-containing SDS micellar system oxidized by UV radiation also support the aforementioned redox cycles. Tocopheroxyl radical is quenched by addition of caffeic acid to the medium. The caffeic radical is not observed suggesting that the radical is short-lived. The interaction of caffeic phenoxyl radical (produced either by autoxidation at pH 9.3 or by reaction of caffeic with ferrylmyoglobin) with ascorbate, as expected from the respective reduction potentials, results in the reduction of caffeic radical by ascorbate.

Finally, two important implications can be surmised from the participation of caffeic acid in reactions with a-tocopherol and ascorbate: chemically, there is an efficient transfer of the radical character from LDL particle to the aqueous medium; biologically, a strong synergistic antioxidant effect prevents LDL oxidation. These results may have implications for preventing atherosclerotic disease.

Acknowlegments: Joćo Laranjinha acknowledges grants PRAXIS XXI/BPD/11855/97 from JNICT and 419/97 from Fundaēćo Luso-Americana.


Oxidative stress and LDL oxidation in patients with non-insulin-dependent diabetes mellitus (NIDDM)

Janne Leinonen1, Terho Lehtimäki2, Vappu Rantalaiho2, Shinya Toyokuni3, Tiina Solakivi2, and Hannu Alho4
1University of Tampere, Tampere, Finland; 2Tampere University Hospital, Tampere, Finland; 3Kyoto University, Kyoto, Japan; 4National Public Health Institute, Helsinki, Finland

Oxidative stress and oxidation of low-density lipoprotein (LDL) may be involved in the development of diabetic macroangiopathy and kidney disease. Therefore, we determined the urinary excretion of 8 hydroxydeoxyguanosine (8-OHdG), plasma concentrations of ascorbic acid, uric acid, a-tocopherol, and protein thiols, the plasma total peroxyl radical-trapping potential (TRAP), the level of autoantibodies against oxidized LDL (oxLDL-Ab), and the susceptibility of LDL to oxidation in a cohort of 87 NIDDM patients after nine years of disease duration. NIDDM patients had higher 24-h urinary excretion of 8-OHdG (68.2 ± 39.4 vs. 49.6 ± 37.7 mg, p=0.001), and a moderately higher level of oxLDL-Ab (0.083 ± 0.051 vs. 0.062 ± 0.045, optical density units, p=0.04) than their matched non-diabetic control subjects. There were no significant differences in the other main outcome variables between NIDDM patients and control subjects. NIDDM patients with coronary heart disease (CHD) had lower urinary 8-OHdG and higher plasma TRAP than patients without CHD, but there were no such differences in the parameters of LDL oxidation. Low glomerular filtration rate was associated with high plasma uric acid and high plasma TRAP, but there were no other significant associations between indices of oxidative stress, LDL oxidation, and renal function.

In conclusion, systemic oxidative stress, assessed by urinary 8 OHdGwas increased in patients with NIDDM, but this was not reflected in the antioxidant activity of plasma. High oxidative stress or LDL oxidation were not associated with the presence of CHD or diabetic kidney disease.


Stable carbon isotope analysis of Lutein using gas isotope ratio mass spectrometry

Y. Lihang, W.S. Trahanovsky, Y. Liang, R.E. Serfass, and W.S. White
Department of Food Science & Human Nutrition and Department of Chemistry, Iowa State University, Ames, IA 50011

High-precision stable carbon isotope ratio analysis of carotenoids using gas chromatography-combustion interfaced-isotope ratio mass spectrometry (GC-C-IRMS) requires hydrogenation to thermally stable perhydro-analogs. Catalytic activities for hydrogenation of b-carotene (bC) and for alcohol hydrogenolysis and hydrogenation of lutein (b,e-carotene-3,3¹-diol) were compared on palladium (Pd) and platinum (Pt) catalysts. The catalyst in cyclohexane and glacial acetic acid containing 0.02 M p-toluenesulfonic acid (50:50, v/v) was reduced by hydrogen gas prior to addition of bC or lutein in cyclohexane; the reaction proceeded at 60°C and 10 psi hydrogen pressure for 16 h. The reaction products were washed with water, extracted twice with hexane, and the perhydro-bC product was quantified by using gasliquid chromatography and squalane as internal standard. For the respective catalysts, respective recoveries of perhydro-bC (mol%SD) from bC and lutein were: Pt oxide, 102.8±0.5 and 21.7±3.9; 5% Pt-on-carbon, 101.9±3.0 and trace; 5% Pd-on- carbon, 45.9±1.0 and 55.3±4.8. The stable carbon isotope ratio (expressed as d13C versus the international standard, Pee Dee Belemnite, in per mil units, denoted ”) of the perhydro-bC product of lutein hydrogenation with Pd catalyst was determined to be -30.97±0.27” by using GC-C-IRMS; the stable carbon isotope ratio of lutein without hydrogenation was determined to be -30.82±0.10” by using elemental analyzer (EA)-IRMS. Hydrogenation of lutein with Pd catalyst does not result in carbon isotopic fractionation, and thus the per hydro-bC product is suitable for GC-C-IRMS analysis.

Supported by USDA ISU/CDFIN 94-34115-2835


Coordinate regulation of g-glutamylcysteine synthetase catalytic and regulatory subunit gene expression in rat lung epithelial L2 cells by 4-hydroxynonenal

Rui-Ming Liu, Lin Gao, Jinal Choi, and Henry Jay Forman
Department of Molecular Pharmacology & Toxicology, School of Pharmacy, University of Southern California, Los Angeles, California

Rationale: g-Glutamylcysteine synthetase (GCS), the rate-limiting enzyme in de novo GSH synthesis, is composed of two subunits, one catalytic (GCS-HS) and one regulatory (GCS-LS). Although both subunits have been shown to be inducible by oxidants, it is not clear whether they are regulated through the same mechanism.

Methods and Results: Rat lung epithelial L2 cells were treated with 4-hydroxynonenal (4HNE), a lipid peroxidation product. The GCS protein contents, the steady state contents and the stabilities of GCS mRNAs, as well as the transcription rates of both GCS subunits were measured. It was found that 4HNE, at concentrations of 5-20 µM, caused an initial decrease and then an increase in the intracellular GSH content in a concentration-dependent manner. There was no measurable cytotoxicity. The steady state levels and the stabilities of both GCS subunit mRNAs were increased along with the GCS protein contents. Actinomycin D, an inhibitor of transcription, completely abolished these increases in GCS mRNAs suggesting that, in addition to an increased stability of mRNAs, transcriptional activation also contributed to the increased steady state GCS mRNAs contents. This conclusion was confirmed by nuclear run-on experiments. In contrast, protein synthesis inhibitors blocked the increase in GCS-LS mRNA but not the increase in GCS-HS mRNA caused by 4HNE.

Conclusion: Although the expression of both GCS-HS and GCS LS subunits were coordinately up-regulated by 4HNE, the mechanism was not the same. The increase in mRNA content of GCS-LS appears to be a secondary response while that of GCS-HS may be a primary response to 4HNE.

Suppoted by Grant ES05511 from NIH