Lisa K. Folkes

Reactivity of hydrogen sulfide with peroxynitrite and other oxidants of biological interest

Hydrogen sulfide (H(2)S) is an endogenously generated gas that can also be administered exogenously. It modulates physiological functions and has reported cytoprotective effects. To evaluate a possible antioxidant role, we investigated the reactivity of hydrogen sulfide with several one- and two-electron oxidants. The rate constant of the direct reaction with peroxynitrite was (4.8±1.4)×10(3)M(-1) s(-1) (pH 7.4, 37°C). At low hydrogen sulfide concentrations, oxidation by peroxynitrite led to oxygen consumption, consistent with a one-electron oxidation that initiated a radical chain reaction. Accordingly, pulse radiolysis studies indicated that hydrogen sulfide reacted with nitrogen dioxide at (3.0±0.3)×10(6)M(-1) s(-1) at pH 6 and (1.2±0.1)×10(7)M(-1) s(-1) at pH 7.5 (25°C). The reactions of hydrogen sulfide with hydrogen peroxide, hypochlorite, and taurine chloramine had rate constants of 0.73±0.03, (8±3)×10(7), and 303±27M(-1) s(-1), respectively (pH 7.4, 37°C). The reactivity of hydrogen sulfide was compared to that of low-molecular-weight thiols such as cysteine and glutathione. Considering the low tissue concentrations of endogenous hydrogen sulfide, direct reactions with oxidants probably cannot completely account for its protective effects.

Thiyl radicals react with nitric oxide to form S-nitrosothiols with rate constants near the diffusion-controlled limit

A possible route to S-nitrosothiols in biology is the reaction between thiyl radicals and nitric oxide. D. Hofstetter et al. (Biochem. Biophys. Res. Commun.360:146-148; 2007) claimed an upper limit of (2.8+/-0.6)x10(7) M(-1)s(-1) for the rate constant between thiyl radicals derived from glutathione and nitric oxide, and it was suggested that under physiological conditions S-nitrosation via this route is negligible. In the present study, thiyl radicals were generated by pulse radiolysis, and the rate constants of their reactions with nitric oxide were determined by kinetic competition with the oxidizable dyes 2,2-azino-bis(3-ethylbenzothiazoline-6-sulfonate) and a phenothiazine. The rate constants for the reaction of nitric oxide with thiyl radicals derived from glutathione, cysteine, and penicillamine were all in the range (2-3) x10(9) M(-1)s(-1), two orders of magnitude higher than the previously reported estimate in the case of glutathione. Absorbance changes on reaction of thiyl radicals with nitric oxide were consistent with such high reactivity and showed the formation of S-nitrosothiols, which was also confirmed in the case of glutathione by HPLC/MS. These rate constants imply that formation of S-nitrosothiols in biological systems from the combination of thiyl radicals with nitric oxide is much more likely than claimed by Hofstetter et al.

Kinetics of reduction of tyrosine phenoxyl radicals by glutathione

Modification of tyrosine (TyrOH) is used as a marker of oxidative and nitrosative stress. 3,3-Dityrosine formation, in particular, reflects oxidative damage and results from the combination of two tyrosyl phenoxyl radicals (TyrO·). This reaction is in competition with reductive processes in the cell which repair tyrosyl radicals: possible reductants include thiols and ascorbate. In this study, a rate constant of 2 x 10⁶ M⁻¹ s⁻¹ was estimated for the reaction between tyrosyl radicals and glutathione (GSH) at pH 7.15, generating the radicals by pulse radiolysis and monitoring the tyrosyl radical by kinetic spectrophotometry. Earlier measurements have suggested that this repair reaction could be an equilibrium, and to investigate this possibility the reduction (electrode) potential of the (TyrO·,H+/TyrOH) couple was reinvestigated by observing the fast redox equilibrium with the indicator 2,2-azinobis(3-ethylbenzothiazoline-6-sulphonate). Extrapolation of the reduction potential of TyrO· measured at pH 9-11 indicated the mid-point reduction potential of the tyrosyl radical at pH 7, E(m₇)(TyrO·,H+/TyrOH) = 0.93 ± 0.02 V. This is close to the reported reduction potential of the glutathione thiyl radical, E(m₇) = 0.94 ± 0.03V, confirming the repair equilibrium constant is of the order of unity and suggesting that efficient reduction of TyrO· by GSH might require removal of thiyl radicals to move the equilibrium in the direction of repair. Loss of thiyl radicals, facilitating repair of TyrO·, can arise either via conjugation of thiyl with thiol/thiolate or oxygen, or unimolecular transformation, the latter important at low concentrations of thiols and oxygen.

Kinetics of reaction of nitrogen dioxide with dihydrorhodamine and the reaction of the dihydrorhodamine radical with oxygen: implications for quantifying peroxynitrite formation in cells.

Dihydrorhodamine 123 (RhH2) has been used to detect reactive nitrogen species, including peroxynitrite and its radical decomposition products, peroxynitrite probably oxidizing RhH2 to rhodamine (Rh) via radical products rather than directly. In this study, the radical intermediate (RhH(.)) was generated by pulse radiolysis, and shown to react with oxygen with a rate constant k approximately 7 x 10(8) M(-1) s(-1). This fast reaction was exploited in experiments observing Rh being formed slowly (k approximately 4-7 x 10(5) M(-1) s(-1)) from oxidation of RhH2 by nitrogen dioxide in a rate-limiting step, >1000-fold slower than the corresponding oxidation by carbonate radicals. The time-dependent uptake of RhH2 into mammalian cells was measured, with average intracellular levels reaching only approximately 10 microM with the protocol used. The combination of low loading and relatively low reactivity of oxidants towards RhH2 compared to competing cellular nucleophiles suggests rather a small fraction of peroxynitrite-derived radicals (mainly CO3(.-)) may be scavenged intracellularly by RhH2.

Modification of DNA damage mechanisms by nitric oxide during ionizing radiation

Nitric oxide ((•)NO) is a very effective radiosensitizer of hypoxic mammalian cells. In vivo (•)NO may have effects on tumor vasculature and hence on tumor oxygenation and it may also interact with radiation-produced radicals to modify DNA lesions. Few studies have addressed this last aspect, and we report here specific base modifications that result from reaction of (•)NO with radicals in DNA bases and in plasmid DNA after irradiation. 2-Deoxyxanthosine monophosphate and 2-deoxy-8-azaguanosine monophosphate (8azadGMP) are formed upon γ-irradiation of 2-deoxyguanosine monophosphate (dGMP) in the presence of micromolar levels of (•)NO in anoxia. In addition, the presence of (•)NO at physiological pH inhibits the formation of the well-described (•)OH-induced oxidation product of dGMP, 8-oxo-2-deoxyguanosine monophosphate. Single-strand breaks are induced in plasmid DNA when γ-irradiated in anoxia, whereas in the presence of (•)NO the number of breaks is reduced by approximately threefold, and evidence is shown for the formation of 8azadGMP in these plasmids. The consequence of the base modifications by (•)NO are as yet unknown although additional breaks are revealed in irradiated plasmid DNA after treatment with glycosylases involved in base excision repair. V79-4 cells irradiated in anoxia show an enhancement in the number of γH2AX foci when (•)NO is present, particularly evident a few hours postirradiation, indicative of the formation of replication-induced DNA damage. We propose that the consequence of (•)NO-induced base modifications in anoxia contributes to its radiosensitization of cells.

Recombinant horseradish peroxidase variants for targeted cancer treatment

Cancer is a major cause of death. Common chemo‐ and radiation‐therapies damage healthy tissue and cause painful side effects. The enzyme horseradish peroxidase (HRP) has been shown to activate the plant hormone indole‐3‐acetic acid (IAA) to a powerful anticancer agent in in vitro studies, but gene directed enzyme prodrug therapy (GDEPT) studies showed ambivalent results. Thus, HRP/IAA in antibody directed enzyme prodrug therapy (ADEPT) was investigated as an alternative. However, this approach has not been intensively studied, since the enzyme preparation from plant describes an undefined mixture of isoenzymes with a heterogenic glycosylation pattern incompatible with the human system. Here, we describe the recombinant production of the two HRP isoenzymes C1A and A2A in a Pichia pastoris benchmark strain and a glyco‐engineered strain with a knockout of the α‐1,6‐mannosyltransferase (OCH1) responsible for hypermannosylation. We biochemically characterized the enzyme variants, tested them with IAA and applied them on cancer cells. In the absence of H2O2, HRP C1A turned out to be highly active with IAA, independent of its surface glycosylation. Subsequent in vitro cytotoxicity studies with human T24 bladder carcinoma and MDA‐MB‐231 breast carcinoma cells underlined the applicability of recombinant HRP C1A with reduced surface glycoslyation for targeted cancer treatment. Summarizing, this is the first study describing the successful use of recombinantly produced HRP for targeted cancer treatment. Our findings might pave the way for an increased use of the powerful isoenzyme HRP C1A in cancer research in the future.

DNA damage induced by nitric oxide during ionizing radiation is enhanced at replication.

Nitric oxide (NO) is a very effective radiosensitizer of hypoxic mammalian cells, at least as efficient as oxygen in enhancing cell death in vitro. NO may induce cell death through the formation of base lesions which are difficult to repair, and if they occur within complex clustered damage common to ionizing radiation, they may lead to replication-induced DNA strand breaks. It has previously been shown that 8-azaguanine and xanthine result from the reaction of guanine radicals with nitric oxide. We have now shown that adenine radicals also react with NO to form hypoxanthine and 8-azaadenine. Cells irradiated in exponential growth in the presence of NO are twice as radiosensitive compared to those irradiated in anoxia alone, whereas confluent cells are less radiosensitive to (•)NO. In addition, the numbers of DNA double strand breaks observed as γH2AX staining following radiosensitization by NO, are higher in exponential cells than in confluent cells. DNA damage, detected as 53BP1 foci, is also higher in HF-19 cells expressing Cyclin A, a marker for cells in S and G2 phases of the cell cycle, following radiosensitization by NO. RAD51 foci are highest in V79-4 cells irradiated in the presence of NO compared to in anoxia, 24h after radiolysis. This work presents evidence that radiosensitization of cells by NO is in part through the formation of specific DNA damage, difficult to repair, which in dividing cells may induce the formation of stalled replication forks and as a consequence replication-induced DNA strand breaks which may lead to cell death.

Validation of a method for the determination of the anticancer agent Combretastatin A1 phosphate (CA1P, OXi4503) in human plasma by HPLC with post-column photolysis and fluorescence detection.

A validated method for the determination of Combretastatin A1 phosphate (CA1P, OXi4503), a bisphosphate prodrug of the vascular disrupting agent Combretastatin A1 in human plasma has been developed using fluorescence detection after post-column photolysis. The separation used the ion-pairing agent tetrabutylammonium hydrogen sulphate, and this agent was also required to give consistently high recovery from plasma. Initially, the range was shown to be linear (r(2)>0.995) from the LOQ of 0.025 μM to 5 μM, but as the trial progressed to much higher doses, using a lower injection volume, the assay was subsequently subject to limited revalidation to cover the range from 0.05 to 50 μM. Intra-assay precision and accuracy ranged from 2.2 to 11.8% and 1.8 to 13% respectively, and for inter-assay from 4.4 to 14.9% and 1.7 to 6.5%. Mean recovery of OXi4503 from plasma was 80.2%.

A validated HPLC method with fluorescence detection for the glucuronides of Combretastatin A1 in human plasma, and studies on their cis-trans isomerisation.

Two monoglucuronides (CA1G1 and CA1G2) of the catecholic cis-stilbene Combretastatin A1 (CA1, OXi4500), have been identified in a clinical trial of the bisphosphate prodrug of OXi4500, OXi4503. A validated assay for the two glucuronides in human plasma using HPLC with fluorescence detection after post-column photolysis is described. The assay was linear over the range 25 nM (CA1G1) or 50 nM (CA1G2) - 5000 nM, R(2)≥ 0.996. The intra-day precision for CA1G1 was better than 8.7% RSD (19.4% at the LLOQ), and the inter-day precision was better than 5.5% RSD (7.6% at the LLOQ). The intra- and inter-day accuracies were better than ± 12.6% relative error (14.8% at the LLOQ) and 4.8% (5.4% at the LLOQ) respectively. For CA1G2, the intra-day precision was better than 5.7% RSD (7.5% at the LLOQ), and the inter-day precision was better than 4.8% RSD (11.9% at the LLOQ). The intra- and inter-day accuracies were better than ± 10.1% relative error (12.6% at the LLOQ) and 2.2% (3.8% at the LLOQ) respectively. Recovery from plasma was measured at three concentrations (125, 625 and 2500nM). Mean recovery of CA1G1 was 94.5% and ranged from 94.4 to 99.2%. Mean recovery of CA1G2 was 90.7%, range 88-92%. During the validation process, one of the isomers was unexpectedly found to be unstable. CA1G1, substituted ortho to the stilbene, was relatively stable, but the meta-substituted CA1G2 readily converted from the cis-stilbene conformation to the trans isomer. This was catalysed by acid and heavy metals, and could be inhibited by antioxidants such as ascorbic acid. Isomerisation could also be induced by one-electron oxidation processes such as horseradish peroxidase and azide radicals.

Set2 Methyltransferase Facilitates DNA Replication and Promotes Genotoxic Stress Responses through MBF-Dependent Transcription

Summary Chromatin modification through histone H3 lysine 36 methylation by the SETD2 tumor suppressor plays a key role in maintaining genome stability. Here, we describe a role for Set2-dependent H3K36 methylation in facilitating DNA replication and the transcriptional responses to both replication stress and DNA damage through promoting MluI cell-cycle box (MCB) binding factor (MBF)-complex-dependent transcription in fission yeast. Set2 loss leads to reduced MBF-dependent ribonucleotide reductase (RNR) expression, reduced deoxyribonucleoside triphosphate (dNTP) synthesis, altered replication origin firing, and a checkpoint-dependent S-phase delay. Accordingly, prolonged S phase in the absence of Set2 is suppressed by increasing dNTP synthesis. Furthermore, H3K36 is di- and tri-methylated at these MBF gene promoters, and Set2 loss leads to reduced MBF binding and transcription in response to genotoxic stress. Together, these findings provide new insights into how H3K36 methylation facilitates DNA replication and promotes genotoxic stress responses in fission yeast.