Investigation of some properties of oligodeoxynucleotides containing 4'-thio-2'-deoxynucleotides: duplex hybridization and nuclease sensitivity

The thermal stabilities of the duplexes formed between 4'-thio-modified oligodeoxynucleotides and their DNA and RNA complementary strands were determined and compared with those of the corresponding unmodified oligodeoxynucleotides. A 16mer oligodeoxynucleotide containing 10 contiguous 4'-thiothymidylate modifications formed a less stable duplex with the DNA target (deltaTm/modification -1.0 degrees C) than the corresponding unmodified oligodeoxynucleotide. However, when the same oligodeoxynucleotide was bound to the corresponding RNA target, a small increase in Tm was observed (deltaTm/modification +0.16 degrees C) when compared with the unmodified duplex. A study to identify the specificity of an oligodeoxynucleotide containing a 4'-thiothymidylate modification when forming a duplex with DNA or RNA containing a single mismatch opposite the modification found the resulting Tms to be almost identical to the wild-type duplexes, demonstrating that the 4'-thio-modification in oligodeoxynucleotides has no deleterious effect on specificity. The nuclease stability of 4'-thio-modified oligodeoxynucleotides was examined using snake venom phosphodiesterase (SVPD) and nuclease S1. No significant resistance to degradation by the exonuclease SVPD was observed when compared with the corresponding unmodified oligodeoxynucleotide. However, 4'-thio-modified oligodeoxynucleotides were found to be highly resistant to degradation by the endonuclease S1. It was also demonstrated that 4'-thio-modified oligodeoxynucleotides elicit Escherichia coli RNase H hydrolysis of the RNA target only at high enzyme concentration.

Role of lipid hydroperoxides in the activation of 15-lipoxygenase

We have used stopped-flow rapid reaction methods, employing both fluorescence and absorbance monitoring, together with HPLC analysis of the products to study the activation of soybean 15-lipoxygenase by 13(S)-hydroperoxy-9, 11(E,Z)-octadecadienoic acid (13-HPOD). When lipoxygenase is mixed with an equimolar concentration of 13-HPOD, the enzyme undergoes a rapid change in fluorescence. The rate of the change of fluorescence is dependent on the concentration of the 13-HPOD (k = 6.7 x 10(6) M-1 s-1) and is accompanied by activation of the enzyme. The fluorescence change is not accompanied by any change in the UV absorbance of the 13-HPOD, suggesting no loss of the conjugated diene during enzyme activation, and HPLC analysis of the products of the reaction confirms that the 13-HPOD can be recovered unchanged following this reaction. In the presence of an inhibitor (BWA4C, a hydroxamate inhibitor) that reduces the active-site iron, the 13-HPOD and the inhibitor are destroyed in a peroxidase-like reaction. On the basis of these observations we propose that 13-HPOD binds to the enzyme and facilitates activation of the enzyme, possibly through the formation of a protein radical, and that the 13-HPOD is not changed chemically in this process.

Dual action of tirapazamine in the induction of DNA strand breaks

Tirapazamine (3-amino-1,2,4-benzotriazine-1,4-dioxide, SR 4233) is the lead compound of a new class of hypoxic cell cytotoxins showing considerable antitumor activity. Hypoxic cytotoxicity of tirapazamine is believed to be mediated by free radical attack of its one-electron reduced metabolite on DNA, but little is known about the DNA lesions induced by the drug. Using the anoxic xanthine/xanthine oxidase system to effect one-electron reduction of tirapazamine under controlled conditions, we studied the action of the drug toward pUC18 and calf thymus DNA. Agarose gel electrophoresis indicated that tirapazamine causes substantially higher levels of single-strand breakage than double-stand breakage. The 5' DNA termini at the single-strand breaks were shown to be phosphorylated. Little, if any, base damage was observed when the damaged DNA was analyzed by a 32P-postlabeling assay. The major detectable lesion (comprising approximately 32% of the 3' ends of tirapazamine-induced single-strand breaks) was the phosphoglycolate moiety, which is caused by deoxyribose fragmentation. Since phosphoglycolate formation requires the addition of oxygen, we conclude that tirapazamine acts in a dual fashion to produce phosphoglycolates: (a) to generate a free radical in the deoxyribose ring (i.e., .C-4' and (b) then to donate an oxygen atom. The oxygen donation by tirapazamine was confirmed by anoxic irradiation of DNA in the presence of the unmetabolized drug. Increasing the concentration of the drug (up to 50 microM) led to a dramatic increase in the yield of phosphoglycolate.

Differentiation of the two forms of GPIb functioning as receptors for alpha-thrombin and von Willebrand factor: Ca2+ responses of protease-treated human platelets activated with alpha-thrombin and the tethered ligand peptide

Previous results have shown that both GPIb and the seven transmembrane domain receptor (STDR) are required for optimal thrombin-induced platelet activation (Greco et al., 1996). Limited degradation (approximately 10%) of GPIb and the STDR by elastase reduced the Ca2+ response to 0.5 nM alpha-thrombin by only 10% whereas Serratia marcescens metalloprotease reduced the Ca2+ response by 80% and fully abrogated high-affinity thrombin binding and aggregation. vWF/ristocetin-induced agglutination was only slightly reduced (20%) while Ca2+ and aggregation response to higher thrombin concentrations were retained. At increasing elastase and Serratia protease concentrations, degradation of the STDR proceeded from the amino-terminal domain, but Ca2+ responses to the tethered ligand peptide SFLLRNPNDKYEPF were not affected by either protease. These results show that both putative thrombin receptors are susceptible to protease degradation and suggest that Serratia protease is able to differentiate the GPIb-mediated events associated with thrombin activation from those associated with ristocetin-induced agglutination.

Contributions of glycoprotein Ib and the seven transmembrane domain receptor to increases in platelet cytoplasmic [Ca2+] induced by alpha-thrombin

The individual contributions of glycoprotein Ib (GPIb) and the seven transmembrane domain receptor (STDR) to increases in platelet [Ca2+]i induced by alpha-thrombin or the tethered ligand peptide (TLP; SFLLRNPNDKYEPF) have been determined in control platelets, in platelets where the thrombin binding site on GPIb was blocked with the monoclonal antibodies TM60 and LJ-Ib10, in platelets where access of thrombin to the STDR was blocked by polyclonal antipeptide antibodies, and in Bernard-Soulier platelets which constitutively lack GPIb. Curve-fitting analyses (LIGAND) showed that binding of PPACK-thrombin and alpha-thrombin to the moderate-affinity site was not detected in the best-fit model in the presence of anti-STDR antibodies although with alpha-thrombin there was also decreased binding at the high-affinity site. Conversely, TM60 blocked binding of alpha-thrombin to the high-affinity site but also decreased binding at the moderate affinity site. Separately, either TM60 or anti-TNA (150 micrograms/mL) reduced thrombin (0.5 nM)-induced elevations in [Ca2+]i to 50% of control values, but Ca2+ elevations were essentially abrogated (4.2 +/- 5%) when the two were added in combination. [Ca2+]i dose-response curves for alpha-thrombin were curvilinear and were only 50% of controls in the presence of anti-GPIb or anti-STDR antibodies at up to 10 nM alpha-thrombin, with their greatest sensitivity being below 2 nM. With Bernard-Soulier platelets, changes in [Ca2+]i were not detectable at < or = 0.5 nM alpha-thrombin but were also 50% of controls at 5-10 nM alpha-thrombin. [Ca2+]i responses to TLP (1-100 microM) of antibody-blocked platelets were identical to those of controls whereas responses were approximately 50% of controls in Bernard-Soulier platelets. The rate of increase in [Ca2+]i in controls was twice that seen in antibody-blocked platelets and about 5-fold greater than in Bernard-Soulier platelets. These results demonstrate that both GPIb and the STDR are required to ensure the optimal rate and extent of platelet activation over a range of alpha-thrombin concentrations (0.3-10 nM) and that the STDR corresponds to the previously described moderate-affinity thrombin receptor.

Mechanisms of radiosensitization in iododeoxyuridine-substituted cells

The radiosensitization caused by iododeoxyuridine (IdU)-substitution of thymidine in V79-171 cells is decreased by the presence of acetone during irradiation. Acetone, at 1 mol dm-3, removes almost all the increase in double strand breaks (dsbs) caused by IdU substitution, but removes only about two-thirds of the enhancement in killing. Similar observations were made with BrdU-substituted cells. The decrease in cell radiosensitization coincides with the removal of the additional dsbs. The protection afforded by acetone is assumed to be due to its scavenging of hydrated electrons, thought to be the active species causing enhanced DNA damage in the presence of halogenated pyrimidines. The residual component of IdU radiosensitization, which could not be removed by treatment with acetone, is manifest largely as a shoulder effect (Dq) and may be due to either a subset of non-scavengable, lethal dsbs and/or the influence of IdU on the fixation of potentially lethal damage. This study further demonstrates that halogenated pyrimidine-mediated radiosensitization consists of at least distinct components each associated with a different phenomenon.

Increase in copy number of an integrated vector during continuous culture of Hansenula polymorpha expressing functional human haemoglobin

Recombinant human haemoglobin A (rHbA) was produced by a leucine-requiring strain of Hansenula polymorpha which had been transformed with an integration vector containing the Saccharomyces cerevisiae LEU2 gene and cDNAs for the expression of alpha and beta globin each driven by the H. polymorpha MOX promoter. After 40 generations in a chemostat it was found that the integrated vector had become amplified in the host strain. In some cases this led to an increase in LEU2 gene dosage, but a loss of globin expression cassettes. In other cases the globin gene dosage also increased. These changes coincided with an increase in rHbA production in the culture, which was reversed when the dilution rate was increased. Isolates from a chemostat culture producing elevated levels of rHbA were grown in fed-batch fermentations, resulting in higher productivities than when inoculated with the parent strain. The rHbA produced was purified and characterized. Oxygen binding studies and electrospray mass spectrometry showed that the rHbA had been processed and assembled correctly, and behaved as a fully functional co-operative tetramer.

Assay and single dose pharmacokinetics of a novel systemic acyl coenzyme A cholesterol O-acyltransferase inhibitor, RP 73163, in rat plasma using automated solid-phase extraction with high-performance liquid chromatography

RP 73163, (S)-2-[5-(3,5-dimethylpyrazol-1-yl)pent-1-yl]sulphinyl-4,5 diphenylimidazole (I), is a highly potent in vitro and in vivo inhibitor of acyl coenzyme A cholesterol O-acyltransferase (ACAT) (E.C. 2.3.1.26), and as such it has potential therapeutic use as a cholesterol lowering agent in man. A method has been developed for the extraction and assay of I from rat plasma, using a fully automated solid-phase extraction column (ASPEC) technique, coupled to a reversed-phase HPLC system with detection by native fluorescence. The method has been validated over the concentration range 10-500 ng/ml, with demonstrated linearity, precision and accuracy, the mean limit of detection being 6.6 +/- 1.3 ng/ml. Application of the method to the assay of samples following administration of the compound to male and female rats is reported, together with determined pharmacokinetic parameters.

A comparison of DNA damages produced under conditions of direct and indirect action of radiation

Studies using a 32P-postlabelling assay reveal that at both low and high concentrations of hydroxyl radical scavenger, DNA damage from alpha-irradiation is similar to that from gamma-irradiation. At low concentrations of scavenger the identified damages are predominantly thymine glycol (Tg) and phosphoglycolate (pg) lesions, indicative of indirect hydroxyl radical (.OH) action. However, at high concentrations of scavenger (i.e. where direct effects are expected to dominate) a somewhat different pattern of damage is observed after alpha- and gamma-irradiation, with several novel lesions detected. Further experiments, in which the results of DMSO scavenging of irradiated DNA solutions are compared with glycerol scavenging and irradiation of 'dry' solid DNA, demonstrate that both direct effects and scavenger-derived secondary radicals contribute to the different spectra of lesions observed upon irradiation of highly scavenged DNA systems. Generally, irradiation under conditions that favour direct damage result in a lower relative yield of pg, whilst for the highly scavenged systems Tg damages persist. To account for this we propose that Tg are efficiently produced by direct action in these systems. Differences seen upon irradiation of the 'dry' system, compared with those of dilute aqueous solution, suggest that direct action yields a distinct spectrum of lesions.

Influence of nucleic acid base composition on radiation-induced strand breakage in single stranded DNA: a time resolved study

The following study investigates the pathways involved in the induction of single strand breaks (ssb) in various samples of single stranded (ss) DNA (calf thymus, Micrococcus lysodeikticus, Clostridium perfringens) with differing nucleic acid base composition. The time scale for the induction of ssb was determined from changes in the light scattering intensity following pulse irradiation of aqueous solutions containing these ssDNA samples at pH7.8 under either aerated or deaerated conditions. The induction of ssb under these conditions is predominantly by the hydroxyl radical and shows various kinetically distinct components. The immediate ssb (t < 0.02 s) account for approximately 40-60% of the total yield of ssb. The majority of these ssb are suggested to arise from the 'common' initial attack of the hydroxyl radicals at the sugar phosphate backbone for each of the three DNA samples. Furthermore, slower components for ssb formation (t > 0.02 s) were observed and are suggested to occur through base radical mediated H-atom abstraction from the sugar moiety. The half lives for formation of the majority of ssb, formed through this base radical-mediated H-atom abstraction(s), are in the range of 20-43 ms. The yields of these 'base-mediated' ssb vary markedly (under both aerobic and anaerobic conditions) and reflect the base composition of the DNA sample. It is suggested from these studies that the OH-induced base radicals of guanine/cytosine are more effective precursors for strand breakage than those from adenine/thymine in ssDNA.

Effects of postirradiation temperature on the yields of radiation-induced single- and double-strand breakage in SV40 DNA

The effects of postirradiation holding temperature on the yields of radiation-induced single- and double-strand breaks (SSBs and DSBs) in SV40 DNA have been measured by agarose gel electrophoresis. When the DNA is held at low temperatures (< or = 2 degrees C) before and during electrophoresis, the measured yields of radiation-induced SSBs and DSBs are twofold less than in samples exposed to room temperature. In contrast, if the DNA is incubated at 37 degrees C overnight, the yield of DSBs increases twofold over the room temperature assay, while the SSB yield increases only to a small extent (< or = 20%). From a comparison of the various yields, we suggest that low temperature stabilizes radiation-induced labile sites, and that the increased yield of DSBs at 37 degrees C is due either to the recruitment of spatially separate SSBs as DSBs by duplex melting, or to labile sites generating DSBs. The different routes to DSB formation are kinetically distinct. We conclude that room-temperature electrophoresis measures all SSBs including those from labile sites.

Electron transfer from Phanerochaete chrysosporium cellobiose oxidase to equine cytochrome c and Pseudomonas aeruginosa cytochrome c-551

The electron-transfer reactions of cellobiose oxidase (CBO) have been investigated by conventional and by rapid-scan stopped-flow spectroscopy at pH 6.0. Analysis of the absorbance/time/wavelength matrix by Singular Value Decomposition (SVD) confirms earlier studies showing that cellobiose rapidly reduces the flavin group (7.7 s-1; cellobiose, 100 microM) which in turn slowly (0.2 s-1) reduces the cytochrome b moiety. In the presence of CBO, cellobiose reduces cytochromes c in a reaction that does not depend on oxygen or superoxide. The rate limit for this process is independent of the source of the cytochromes c and is identical with the rate of cytochrome b reduction. Rapid-mixing experiments show that cytochrome b may donate electrons very rapidly to either mammalian cytochrome c or bacterial cytochrome c-551. The reactions were second-order (kc = 1.75 x 10(7) M-1 x s-1; kc-551 = 4.3 x 10(6) M-1 x s-1; pH 6.0, 21 degrees C and I0.064) and strongly ionic-strength (I)-dependent: kc decreasing with I and kc-551 increasing with I. These results suggest the electron-transfer site near cytochrome b bears a significant negative charge. Equilibrium gel chromatography confirms that CBO oxidase and positively charged mammalian cytochrome c make stable complexes. These results are discussed in terms of a model suggesting an electron-transfer role for cytochrome b in vivo, possibly connected with radical-mediated cellulose breakdown.

Mechanisms of radiosensitization in bromodeoxyuridine-substituted cells

The radiosensitization of exponentially-growing V79-171 cells whose DNA has been substituted by bromodeoxyuridine (BrdU) in place of thymidine is decreased if acetone is present during irradiation. Acetone, at a concentration of 1 mol dm-3, removes the majority of the increase in double-strand breaks (dsbs) caused by BrdU substitution, but only removes approximately half of the increase in cell killing. The decrease in cell radiosensitization coincides with the removal of the additional dsbs. The protection afforded by acetone against dsbs is assumed to be due to its ability to scavenge hydrated electrons, thought to be the active species causing the increased DNA damage in the presence of BrdU. The residual component of BrdU radiosensitization which could not be removed by treatment with acetone may be due to either a subset of nonscavengable, lethal dsbs or the influence of BrdU on the fixation of potentially-lethal damage (Iliakis et al. 1992). Cells substituted with BrdU are not sensitized to hydroxyl radicals (from hydrogen peroxide). Also, the enhanced levels of single-strand break (ssb) and dsb production in cells substituted with BrdU arise from analogous events (i.e. increases in the yield of ssbs). These studies support the locally multiply damaged site theory of lesion (dsb) production (Ward 1981) and, in the case of BrdU-substituted cells, the increase in dsbs appears to be due to the production of additional ssbs by hydrated electrons at sites of multiple damage.

Yield of strand breaks as a function of scavenger concentration and LET for SV40 irradiated with 4He ions

We have measured by gel electrophoresis the yields of single- and double-strand breaks (SSBs and DSBs) induced in aqueous solutions of SV40 DNA and the SV40 minichromosome by 137Cs gamma rays (mean LET 0.3 keV micron-1) and 4He ions (mean LETs 85, 102, and 152 keV microns-1). DNA SSBs are caused mainly by the hydroxyl radicals under these conditions and are reduced in yield as either the hydroxyl radical scavenger concentration or the LET is increased (over the range studied). The G(SSB) for 4He ion irradiation is less by a factor of up to 10 than the G(SSB) for gamma irradiation, depending upon the scavenger concentration. The difference in the yields of SSBs agrees well with the difference in the yields of hydroxyl radicals for the radiations in question. In contrast, the yields of DSBs are similar for gamma and 4He ion irradiation over much of the range of scavenging capacity studied. However, at the highest scavenger concentrations the yields of DSBs are greater for 4He ion irradiation. In addition, the yields of DSBs remain almost constant with increasing LET (over the range studied). Therefore the relative yield of DSBs per SSB increases with increasing LET, supporting the hypothesis that increasing LET leads to an increased clustering of damage in DNA.

Induction of strand breaks in polyribonucleotides and DNA by the sulphate radical anion: role of electron loss centres as precursors of strand breakage

The interaction of the sulphate radical anion, SO4.-, with the polyribonucleotides, poly U and poly C, in deaerated, aqueous solutions at pH 7.5 results in strand breakage (sb) with efficiencies of 57 and 23%, respectively, determined by time resolved laser light scattering (TRLS). Most sb are produced within 70 microseconds, the risetime of the detection system. Oxygen inhibits the induction of sb in poly U and poly C by SO4.- through its interaction with a radical precursor to sb. In contrast, the interaction of SO4.- with poly A and single stranded DNA does not lead to significant strand breakage (< or = 5% efficiency). From optical studies, the interaction of poly A and poly G with SO4.- radicals yields predominantly the corresponding one electron oxidized base radicals. With poly C and poly U, it is proposed that the SO4.- radical interacts predominantly by addition to the base moiety to produce the C(5)-yl and C(6)-yl sulphate radical adducts which react with oxygen. These base adducts subsequently interact with the sugar-phosphate moiety by H-atom abstraction to yield C(2)' sugar radicals with rate constants in the range 1.3-1.7 x 10(5) s-1. It is proposed that the C(2)' sugar radical leads to strand breakage within 70 microseconds, in competition with its transformation into the C(1)'-sugar radical involving base release. From optical studies on the interaction of SO4.- with double stranded DNA, it is suggested that the predominant radical species produced in DNA is the one-electron oxidized radical of guanine, consistent with positive charge migration in DNA. Since the efficiency of SO4.- to induce sb in single stranded DNA is low, it is concluded that the one-electron oxidized guanine radical does not effectively induce strand breakage in DNA.