Electron paramagnetic resonance (EPR) spin trapping of biological nitric oxide

Nitric oxide (NO) is a free radical species with multiple physiological functions. Because of low concentrations and short half-life of NO, its direct measurement in living tissues remains a difficult task. Electron paramagnetic resonance (EPR) spin trapping is probably one of the best suitable platforms for development of new methods for quantification of biological NO. The most reliable EPR-based approaches developed so far are based on the reaction of NO with various iron complexes, both intrinsic and exogenously applied. This review is focused on the current state and perspectives of EPR spin trapping for experimental and clinical NO biology.     

Detection of nitric oxide in tissue by spin trapping EPR spectroscopy and triacetylglycerol extraction

A modified method based on EPR spin trapping and triacetylglycerol extraction was used for tissue nitric oxide (NO) detection at room temperature. NO signal intensity was stable for about 1.5 h and the detection limit of this method was less than 200 pmol g^–1 tissue. Using this method, we report evidence that NO production in vivo can be inhibited by adriamycin in mice livers.     

Redox cycling of iron complexes of N-(dithiocarboxy)sarcosine and N-methyl-D-glucamine dithiocarbamate

Iron(II)-dithiocarbamate complexes are used to trap nitrogen monoxide in biological samples, and the resulting nitrosyliron(II)-dithiocarbamate is detected and quantified by ESR. As the chemical properties of these compounds have been little studied, we investigated whether iron dithiocarbamate complexes can redox cycle. The electrode potentials of iron complexes of N-(dithiocarboxy)sarcosine (dtcs) and N-methyl-d-glucamine dithiocarbamate (mgd) are 56 and -25 mV at pH 7.4, respectively, as measured by cyclic voltammetry. The autoxidation and Fenton reaction of iron(II)-dtcs and iron(II)-mgd were studied by stopped-flow spectrophotometry with both iron(II) complexes and dioxygen or hydrogen peroxide in excess. In the case of excess iron(II)-dtcs and -mgd complexes, the rate constants of the autoxidation and the Fenton reaction are (1.6-3.2) x 10(4) and (0.7-1.1) x 10(5) M(-1) s(-1), respectively. In the presence of nitrogen monoxide, the oxidation of iron(II)-dtcs and iron(II)-mgd by hydrogen peroxide is significantly slower (ca. 10-15 M(-1) s(-1)). The physiological reductants ascorbate, cysteine, and glutathione efficiently reduce iron(III)-dtcs and iron(III)-mgd. Therefore, iron bound to dtcs and mgd can redox cycle between iron(II) and iron(III). The ligands dtcs and mgd are slowly oxidized by hydrogen peroxide with rate constants of 5.0 and 3.8 M(-1) s(-1), respectively.     

Redox cycling of iron complexes of N-(dithiocarboxy)sarcosine and N-methyl-d-glucamine dithiocarbamate

Iron(II)–dithiocarbamate complexes are used to trap nitrogen monoxide in biological samples, and the resulting nitrosyliron(II)–dithiocarbamate is detected and quantified by ESR. As the chemical properties of these compounds have been little studied, we investigated whether iron dithiocarbamate complexes can redox cycle. The electrode potentials of iron complexes of N-(dithiocarboxy)sarcosine (dtcs) and N-methyl-d-glucamine dithiocarbamate (mgd) are 56 and −25 mV at pH 7.4, respectively, as measured by cyclic voltammetry. The autoxidation and Fenton reaction of iron(II)–dtcs and iron(II)–mgd were studied by stopped-flow spectrophotometry with both iron(II) complexes and dioxygen or hydrogen peroxide in excess. In the case of excess iron(II)–dtcs and –mgd complexes, the rate constants of the autoxidation and the Fenton reaction are (1.6–3.2) × 10⁴ and (0.7–1.1) × 10⁵ M⁻¹ s⁻¹, respectively. In the presence of nitrogen monoxide, the oxidation of iron(II)–dtcs and iron(II)–mgd by hydrogen peroxide is significantly slower (ca. 10–15 M⁻¹ s⁻¹). The physiological reductants ascorbate, cysteine, and glutathione efficiently reduce iron(III)–dtcs and iron(III)–mgd. Therefore, iron bound to dtcs and mgd can redox cycle between iron(II) and iron(III). The ligands dtcs and mgd are slowly oxidized by hydrogen peroxide with rate constants of 5.0 and 3.8 M⁻¹ s⁻¹, respectively.     

EPR spin trapping detection of carbon-centered carotenoid and beta-ionone radicals

Free radical intermediates were detected by the electron paramagnetic resonance spin trapping technique upon protonation/deprotonation reactions of carotenoid and beta-ionone radical ions. The hyperfine coupling constants of their spin adducts obtained by spectral simulation indicate that carbon-centered radicals were trapped. The formation of these species was shown to be a result of chemical oxidation of neutral compounds by Fe(3+) or I(2) followed by deprotonation of the corresponding radical cations or addition of nucleophilic agents to them. Bulk electrolysis reduction of beta-ionone and carotenoids also leads to the formation of free radicals via protonation of the radical anions. Two different spin adducts were detected in the reaction of carotenoid polyenes with piperidine in the presence of 2-methyl-2-nitroso-propane (MNP). One is attributable to piperidine radicals (C(5)H(10)N*) trapped by MNP and the other was identified as trapped neutral carotenoid (beta-ionone) radical produced via protonation of the radical anion. Formation of these radical anions was confirmed by ultraviolet-visible spectroscopy. It was found that the ability of carotenoid radical anions/cations to produce neutral radicals via protonation/deprotonation is more pronounced for unsymmetrical carotenoids with terminal electron-withdrawing groups. This effect was confirmed by the radical cation deprotonation energy (H(D)) estimated by semiempirical calculations. The results indicate that the ability of carotenoid radical cations to deprotonate decreases in the sequence: beta-ionone > unsymmetrical carotenoids > symmetrical carotenoids. The minimum H(D) values were obtained for proton abstraction from the C(4) atom and the C(5)-methyl group of the cyclohexene ring. It was assumed that deprotonation reaction occurs preferentially at these positions.     

Spinnokinetic analyses of blood disposition and biliary excretion of nitric oxide (NO)-Fe(II)-N-(dithiocarboxy)sarcosine complex in rats: BCM-ESR and BEM-ESR studies

Nitric oxide (NO) is well known to have a wide variety of biological and physiological functions in animals. On the basis of the fact that Fe(II)-dithiocarbamates react with NO, a Fe(II)-N-(dithiocarboxy)sarcosine complex (Fe(II)-DTCS) was proposed as a trapping agent for endogenous NO. However, quantitative pharmacokinetic investigation for NO-Fe(II)-dithiocarbamate complexes in experimental animals has been quite limited. This paper describes the results on the quantitative pharmacokinetic features of a NO-Fe(II)-N-DTCS in both the blood and bile of rats following intravenous (i.v.) administration of the complex. For this purpose, we applied two in vivo methods, i.e. (1) in vivo blood circulation monitoring-electron spin resonance (BCM-ESR) which previously developed, and (2) in vivo biliary excretion monitoring-electron spin resonance (BEM-ESR). We monitored real-time ESR signals due to nitrosyl-iron species in the circulating blood and bile flow. The ESR signal due to NO-Fe(II)-DTCS was stable in biological systems such as the fresh blood and bile. In in vivo BCM- and BEM-ESR, the pharmacokinetic parameters were calculated on the basis of the two-compartment and hepatobiliary transport models. The studies also revealed that the compound is widely distributed in the peripheral organs and partially excreted into the bile. We named a kinetic method to follow spin concentrations as spinnokinetics and this method will be useful for detecting and quantifying the endogenously generated NO in Fe(II)-DTCS administered animals.     

Spin trapping of vascular nitric oxide using colloid Fe(II)-diethyldithiocarbamate

Currently available EPR spin-trapping techniques are not sensitive enough for quantification of basal vascular nitric oxide (NO) production from isolated vessels. Here we demonstrate that this goal can be achieved by the use of colloid Fe(DETC)(2). Rabbit aortic or venous strips incubated with 250 microM colloid Fe(DETC)(2) exhibited a linear increase in tissue-associated NO-Fe(DETC)(2) EPR signal during 1 h. Removal of endothelium or addition of 3 mM N(G)-nitro-l-arginine methyl ester (L-NAME) inhibited the signal. The basal NO production was estimated as 5.9 +/- 0.5 and 8.3 +/- 2.1 pmol/min/cm(2) in thoracic aorta and vena cava, respectively. Adding sodium nitrite (10 microM) or xanthine/xanthine oxidase in the incubation medium did not modify the intensity of the basal NO-Fe(DETC)(2) EPR signal. Reducing agents were not required with this method and superoxide dismutase activity was unchanged by the Fe(DETC)(2) complex. We conclude that colloid Fe(DETC)(2) may be a useful tool for direct detection of low amounts of NO in vascular tissue.     

Differential effect of buffer on the spin trapping of nitric oxide by iron chelates.

Nitric oxide synthase (NOS) generates nitric oxide (NO*) by the oxidation of l-arginine. Spin trapping in combination with electron paramagnetic resonance (EPR) spectroscopy using ferro-chelates is considered one of the best methods to detect NO* in real time and at its site of generation. The spin trapping of NO* from isolated NOS I oxidation of L-arginine by ferro-N-dithiocarboxysarcosine (Fe(DTCS)2) and ferro-N-methyl-d-glucamide dithiocarbamate (Fe(MGD)2) in different buffers was investigated. We detected NO-Fe(DTCS)2, a nitrosyl complex, resulting from the reaction of NO* and Fe(DTCS)2, in phosphate buffer. However, Hepes and Tris buffers did not allow formation of NO-Fe(DTCS)2. Instead, both of these buffers reacted with Fe2+, generating sparingly soluble complexes in the absence of molecular oxygen. Fe(DTCS)2 and Fe(MGD)2 were found to inhibit, to a small degree, NOS I activity with a greater effect observed with Fe(MGD)2. In contrast, Fe(MGD)2 was more efficient at spin trapping NO* from the lipopolysaccharide-activated macrophage cell line RAW264.7 than was Fe(DTCS)2. Data suggested that Fe(DTCS)2 and Fe(MGD)2 are efficient at spin trapping NO* but their maximal efficiency may be affected by experimental conditions.     

Role of nitric oxide in murine cytomegalovirus (MCMV) infection

Cytomegalovirus (CMV) is a typical pathogen of an opportunistic infection. In this review article, various roles of nitric oxide (NO) in murine CMV (MCMV) infections, including acute, persistent and latent infections, are discussed. In the acute phase of MCMV infection, NO plays a protective role against MCMV infection. In contrast, NO has been proven to act as a pathogenic factor in a model of MCMV pneumonitis. In MCMV persistent infection, when MCMV was detected only in the salivary gland, T cells of mice were modified to produce a massive amount of such cytokines as TNF-alpha and IFN-gamma upon in vivo stimulation with anti-CD3. These cytokines then induced mRNA for inducible NO synthase (iNOS), thus resulting in the production of a large amount of NO. A histochemical study demonstrated that NO damaged bronchial epithelial cells, and thereby apparently inducing pneumonitis. In the case of a latent infection, when viral DNA was detected in the host in spite of the absence of any infectious particle, NO increased the amount of persistently-infected MCMV-DNA. As a result, NO was found to act as "a double edged sword" in the CMV-host relationship.     

Inactivation of monomeric sarcosine oxidase by reaction with N-(cyclopropyl)glycine

Monomeric sarcosine oxidase (MSOX) catalyzes the oxidative demethylation of sarcosine (N-methylglycine) and contains covalently bound flavin adenine dinucleotide (FAD). The present study demonstrates that N-(cyclopropyl)glycine (CPG) is a mechanism-based inhibitor. CPG forms a charge transfer complex with MSOX that reacts under aerobic conditions to yield a covalently modified, reduced flavin (lambda(max) = 422 nm, epsilon(422) = 3.9 mM(-1) cm(-1)), accompanied by a loss of enzyme activity. The CPG-modified flavin is converted at an 8-fold slower rate to 1,5-dihydro-FAD (EFADH(2)), which reacts rapidly with oxygen to regenerate unmodified, oxidized enzyme. As a result, CPG-modified MSOX reaches a CPG-dependent steady-state concentration under aerobic conditions and reverts back to unmodified enzyme upon removal of excess reagent. No loss of activity is observed under anaerobic conditions where EFADH(2) is formed in a reaction that goes to completion at low CPG concentrations. Aerobic denaturation of CPG-modified enzyme yields unmodified, oxidized flavin at a rate similar to the anaerobic denaturation reaction, which yields 1,5-dihydro-FAD. The CPG-modified flavin can be reduced with borohydride, a reaction that blocks conversion to unmodified flavin upon removal of excess CPG or enzyme denaturation. The possible chemical mechanism of inactivation and structure of the CPG-modified flavin are discussed.     

Alternative pathways of nitric oxide formation in Lactobacilli: Evidence for nitric oxide synthase activity by EPR

The study of the ability of Lactobacillus plantarum 8P-A3 to synthesize nitric oxide (NO) showed that this strain lacks nitrite reductase. However, analysis by the EPR method revealed the presence of nitric oxide synthase activity in this strain. Like mammalian nitric oxide synthase, lactobacillar NO synthase is involved in the formation of nitric oxide from L-arginine. L. plantarum 8P-A3 does not produce NO in the denitrification process. The regulatory role of NO in symbiotic bacteria is emphasixed.     

In vivo monitoring of hydroxyl radical generation caused by x-ray irradiation of rats using the spin trapping/EPR technique

Measurement of hydroxyl radical (*OH) in living animals irradiated with ionizing radiation should be required to clarify the mechanisms of radiation injury and the in vivo assessment of radiation protectors, because generation of *OH is believed to be one of the major triggers of radiation injury. In this study, *OH generation was monitored by spin trapping the secondary methyl radical formed by the reaction of *OH with dimethyl sulfoxide (DMSO). Rats were injected intraperitoneally with a DMSO solution of alpha-phenyl-N-tert-butylnitrone (PBN). X-irradiation of the rats remarkedly increased the six-line EPR signal in the bile. The strengthened signal was detectable above 40 Gy. Use of 13C-substituted DMSO revealed that the signal included the methyl radical adduct of PBN as a major component. The EPR signal of the PBN-methyl radical adduct was completely suppressed by preadministration of methyl gallate, a scavenger of *OH but not of methyl radical. Methyl gallate did not reduce the spin adducts to EPR-silent forms. These observations indicate that what we were measuring was *OH generated in vivo by x-irradiation. This is the first report of the in vivo monitoring of *OH generation at a radiation dose close to what people might receive in the case of radiological accident or radiation therapy.     

Spin trapping of nitric oxide by ferro-chelates: kinetic and in vivo pharmacokinetic studies

Biologically generated nitric oxide appears to play a pivotal role in the control of a diverse series of physiologic functions. Iron-chelates and low-frequency EPR spectroscopy have been used to verify in vivo production of nitric oxide. The interpretation of in vivo identification of nitric oxide localized at the site of evolution in real time is complicated by the varied kinetics of secretion. The quantitative efficiency of the spectroscopic measurement, so important in understanding the physiology of nitric oxide, remains elusive. The development of a more stable iron-chelate will help better define nitric oxide physiology. In this report, we present data comparing the commonly used ferro-di(N-methyl-D-glucamine-dithiocarbamate) (Fe2+(MGD)2) and the novel chelate ferro-di(N-(dithiocarboxy)sarcosine) (Fe2+(DTCS)2) quantifying the in vitro and in vivo stability of the corresponding spin trapped adducts, NO-Fe(MGD)2 and NO-Fe(DTCS)2. Finally, very low frequency EPR spectroscopy has been used to evaluate the pharmacokinetics of NO-Fe(MGD)2 and NO-Fe(DTCS)2 in mice in real time.     

EPR detection of heme and nonheme iron-containing protein nitrosylation by nitric oxide during rejection of rat heart allograft.

The paramagnetic molecule nitric oxide (NO), produced from L-arginine by a specific enzyme (NO synthase), has been shown to be involved in a surprising variety of mammalian cellular responses, including the regulation of T cell immunity to alloantigens in vitro. In cytotoxic activated macrophages, NO production results in a characteristic pattern of alteration of iron-containing enzyme function that is mimicked by exposure to NO. Electron paramagnetic resonance (EPR) studies have shown the formation of iron-nitrosyl species during macrophage activation and also during sepsis, indicating that alteration of iron-containing protein function may be the result of the well-documented tendency of NO to bind to metal ions. We have recently shown that the NO synthesis induced during alloantigenic activation of rat splenocytes inhibits lymphocyte proliferation and cytotoxic T-lymphocyte generation. This report demonstrates that iron-nitrosyl EPR signals similar to those observed in macrophages and during sepsis are present in the blood and in the grafted tissue of rats during the rejection of allogeneic (but not syngeneic) heart grafts. These signals are found in the blood and at the site of allograft rejection, but are not found in other tissues (such as spleen and lung), and are obliterated by administration of the immunosuppressant FK506. These results directly demonstrate the formation of iron-nitrosyl complexes during vascularized allograft rejection and suggest that consequent destruction of iron-containing protein function plays an important role in the rejection response.     

Formation of methyl radicals derived from cumene hydroperoxide in reconstructed human epidermis: an EPR spin trapping confirmation by using 13C-substitution

Dermal exposure to cumene hydroperoxide (CumOOH) during manufacturing processes is a toxicological issue for the industry. Its genotoxicity, mutagenic action, ability to promote skin tumour, capacity to induce epidermal hyperplasia, and aptitude to induce allergic and irritant skin contact dermatitis are well known. These toxic effects appear to be mediated through the activation to free radical species such as hydroxyl, alkoxyl, and alkyl radicals characterised basically by electron paramagnetic resonance (EPR) and spin-trapping (ST) techniques. To be a skin sensitiser CumOOH needs to covalently bind to skin proteins in the epidermis to form the antigenic entity triggering the immunotoxic reaction. Cleavage of the O–O bond allows formation of unstable CumO^?/CumOO^? radicals rearranging to longer half-life specific carbon-centred radicals R^? proposed to be at the origin of the antigen formation. Nevertheless, it is not still clear which R^? is precisely formed in the epidermis and thus involved in the sensitisation process. The aim of this work was to elucidate in conditions closer to real-life sensitisation which specific R^? are formed in a 3D reconstructed human epidermis (RHE) model by using ¹³C-substituted CumOOH at carbon positions precursors of potentially reactive radicals and EPR-ST. We demonstrated that most probably methyl radicals derived from β-scission of CumO^? radicals occur in RHE through a one-electron reductive pathway suggesting that these could be involved in the antigen formation inducing skin sensitisation. We also describe a coupling between nitroxide radicals and β position ¹³C atoms that could be of an added value to the very few examples existing for the coupling of radicals with ¹³C atoms.     

Continuous monitoring of in vivo nitric oxide formation using EPR analysis in biliary flow.

A method to continuously monitor the nitric oxide (NO) level in anesthetized rats, using an in vivo trapping reaction of NO by iron-dithiocarbamate complex, is reported. Previously, we developed a method of monitoring NO in bile samples containing an NO complex excreted from the liver (Anal. Biochem. 243, 8-14, 1996). In the present study, we modified the method so that the bile flows directly through the EPR sample cell. Rats were injected with low doses of lipopolysaccharide (LPS) to induce NO formation and were later anesthetized. After cannulation, the bile duct was connected to the inlet of the EPR sample cell and the trapping agent iron complex of D-N-methylglucamine dithiocarbamate (MGD-Fe) was administered. The EPR signal level from NO complex of MGD-Fe in the flowing bile was continuously monitored. Using this method, immediate changes in in vivo NO level in rats were observed following administration of drugs that can affect NO formation. In addition, a continuous intravenous saline containing MGD-Fe made the EPR signal level stable and improved animal condition as well as survival time. Therefore, this method has two merits; (1) one can continuously monitor NO formation until it reaches the maximum level; (2) a rapid change in NO level after intervention can be followed. Using this method, we tested the effect of the substrate L-arginine and inhibitors for NO synthase activity and NO synthase induction. The sensitivity of the present method was tested by monitoring NO formation in rats after exposure to ionizing radiation.     

Autoinhibition of endothelial nitric oxide synthase (eNOS) in gut smooth muscle by nitric oxide

Nitric oxide in the gut is produced by nNOS in enteric neurons and by eNOS in smooth muscle cells. The eNOS in smooth muscle is activated by vasoactive intestinal peptide (VIP) released from enteric neurons. In the present study, we examined the effect of nitric oxide on VIP-induced eNOS activation in smooth muscle cells isolated from human intestine and rabbit stomach. NOS activity was measured as formation of the 1:1 co-product, l-citrulline from l-arginine. VIP caused an increase in l-citrulline production that was inhibited by NO in a concentration dependent manner (IC(50)~25 microM; maximal inhibition 72% at 100 microM NO). Basal l-citrulline production, however, was unaffected by NO. The effect was not mediated by cGMP/PKG since the PKG inhibitor KT5823 had no effect on eNOS autoinhibition. The autoinhibition was selective for NO since the co-product l-citrulline had no effect on VIP-induced NOS activation. Similar effects were obtained in rabbit gastric and human intestinal smooth muscle cells. The results suggest that NO produced in smooth muscle cells as a result of the activation of eNOS by VIP exerts an autoinhibitory restraint on eNOS thereby regulating the balance of the VIP/cAMP/PKA and NO/cGMP/PKG pathways that regulate the relaxation of gut smooth muscle.     

Real-time monitoring of nitric oxide in ischemic myocardium using an NO-selective electrode calibrated by electron spin resonance

Using a Langendorff-perfused rat heart preparation and selective electrodes, we determined nitric oxide (NO) and oxygen levels in cardiac tissue. An NO-selective electrode that was calibrated by electron spin resonance (ESR) spectroscopy was inserted into the middle of the myocardium in the left ventricle. Simultaneously, we used an O2-selective electrode to measure the partial pressure of oxygen (pO2) in the perfusate, Krebs-Henseleit (K-H) solution, that was ejected from the heart. After 30 min of aerobic control perfusion, hearts were subjected to 30 min of global ischemia followed by 30 min of reperfusion. Under ischemic conditions, with a gradually decreasing pO2, NO detected by an NO-sensitive electrode within the myocardium was gradually increased. The maximum concentration increases in NO and decreases in pO2 during global ischemia were +10.200 +/- 1.223 microM and -58.608 +/- 4.123 mmHg, respectively. NO and pO2 levels both recovered to pre-ischemia baseline values when perfusion was restarted after global ischemia (reperfusion). The presence of Nomega-nitro-L-arginine methyl ester (L-NAME, 10 mM), a NOS inhibitor, prevented ischemia/reperfusion-induced changes in NO. This study shows that an NO-selective electrode that is calibrated by ESR can provide accurate, real-time monitoring of cardiac NO in normal and ischemic myocardium.     

Immunological detection of nitric oxide synthase(s) in human tissues using heterologous antibodies suggesting different isoforms

Summary Nitric oxide (NO) is generated from l-arginine by NO synthases. Localization of the brain enzyme has been carried out in the rat; however, despite data suggesting that NO is a major regulator of vascular and neural functions in man, there is no information about the localization of NO synthase in human tissues. Rabbit antisera to NO synthase purified from rat brain (antisera A and B) were raised, tested by Western blotting, affinity purification and enzyme immunoprecipitation assay, and used to investigate the distribution of the enzyme in a variety of human tissues by immunohistochemistry. Antisera to two synthetic peptides from cloned neural NO synthase were used to aid specificity testing. Antisera A and B reacted with a 160-kDa protein in Western blots of human brain extracts, gave immunostaining of nerves, and precipitated enzyme activity from rat brain homogenates. Antiserum B to NO synthase also reacted with proteins of M_r between 125 and 140 kDa in extracts of well-vascularised tissues, and immunostained vascular endothelium; the neural and vascular immunoreactivity persisted after affinity purification of antiserum B with the 160 kDa protein. Endothelial staining with antiserum B was seen in respiratory tract, liver, skin and umbilicus; syncytial trophoblasts stained in the placenta. Neural staining with antiserum A and B was seen in the myenteric and submucous plexus, and in nerve fibres in smooth muscle of the gut and in many areas of the central nervous system, particularly cortex, hippocampus, hypothalamus, cerebellum, brain stem and spinal cord. Therefore, antibodies to rat brain enzyme react with the human equivalent and also with other NO synthase isoforms in human endothelium. These findings support the contention that the endothelial enzyme is a different form with partial homology to that in nerves and also provide an anatomical distribution of NO synthase isoforms.Abbreviations NADPH -Nicotinamide adenine dinucleotide phosphate, reduced - FAD flavin adenine dinucleotide - FMN flavin mononucleotide - KLH keyhole limpet haemocyanin - G-MDP N-acetylmuramyl-l-ananyl-d-isoglutamine - TLCK N-p-tosyl-l-lysine chloro-methylketone - DRG dorsal root ganglia