Oxidation of glycosaminoglycans by free radicals and reactive oxidative species: A review of investigative methods.

Glycosaminoglycans, in particular hyaluronan (HA), and proteoglycans are components of the extracellular matrix (ECM). The ECM plays a key role in the regulation of cellular behaviour and alterations to it can modulate both the development of human diseases as well as controlling normal biochemical processes such as cell signalling and pro-inflammatory responses. For these reasons, in vitro fragmentation studies of glycosaminoglycans by free radicals and oxidative species are seen to be relevant to the understanding of in vivo studies of damage to the ECM. A wide range of investigative techniques have therefore been applied to gain insights into the relative fragmentation effects of several reactive oxidative species with the ultimate goal of determining mechanisms of fragmentation at the molecular level. These methods are reviewed here.

One-electron oxidation and reduction of glycosaminoglycan chloramides: a kinetic study.

Hypochlorous acid and its acid-base counterpart, hypochlorite ions, produced under inflammatory conditions, may produce chloramides of glycosaminoglycans, these being significant components of the extracellular matrix (ECM). This may occur through the binding of myeloperoxidase directly to the glycosaminoglycans. The N-Cl group in the chloramides is a potential selective target for both reducing and oxidizing radicals, leading possibly to more efficient and damaging fragmentation of these biopolymers relative to the parent glycosaminoglycans. In this study, the fast reaction techniques of pulse radiolysis and nanosecond laser flash photolysis have been used to generate both oxidizing and reducing radicals to react with the chloramides of hyaluronan (HACl) and heparin (HepCl). The strong reducing formate radicals and hydrated electrons were found to react rapidly with both HACl and HepCl with rate constants of 1-1.7 × 10(8) and 0.7-1.2 × 10(8)M(-1)s(-1) for formate radicals and 2.2 × 10(9) and 7.2 × 10(8)M(-1)s(-1) for hydrated electrons, respectively. The spectral characteristics of the products of these reactions were identical and were consistent with initial attack at the N-Cl groups, followed by elimination of chloride ions to produce nitrogen-centered radicals, which rearrange subsequently and rapidly to produce C-2 radicals on the glucosamine moiety, supporting an earlier EPR study by M.D. Rees et al. (J. Am. Chem. Soc.125: 13719-13733; 2003). The oxidizing hydroxyl radicals also reacted rapidly with HACl and HepCl with rate constants of 2.2 × 10(8) and 1.6 × 10(8)M(-1)s(-1), with no evidence from these data for any degree of selective attack on the N-Cl group relative to the N-H groups and other sites of attack. The carbonate anion radicals were much slower with HACl and HepCl than hydroxyl radicals (1.0 × 10(5) and 8.0 × 10(4)M(-1)s(-1), respectively) but significantly faster than with the parent molecules (3.5 × 10(4) and 5.0 × 10(4)M(-1)s(-1), respectively). These findings suggest that these potential in vivo radicals may react in a site-specific manner with the N-Cl group in the glycosaminoglycan chloramides of the ECM, possibly to produce more efficient fragmentation. This is the first study therefore to conclusively demonstrate that reducing radicals react rapidly with glycosaminoglycan chloramides in a site-specific attack at the N-Cl group, probably to produce a 100% efficient biopolymer fragmentation process. Although less reactive, carbonate radicals, which may be produced in vivo via reactions of peroxynitrite with serum levels of carbon dioxide, also appear to react in a highly site-specific manner at the N-Cl group. It is not yet known if such site-specific attacks by this important in vivo species lead to a more efficient fragmentation of the biopolymers than would be expected for attack by the stronger oxidizing species, the hydroxyl radical. It is clear, however, that the N-Cl group formed under inflammatory conditions in the extracellular matrix does present a more likely target for both reactive oxygen species and reducing species than the N-H groups in the parent glycosaminoglycans.

Reaction of superoxide radicals with glycosaminoglycan chloramides: a kinetic study.

Hypochlorous acid and its acid-base counterpart, hypochlorite ions, produced under inflammatory conditions, may produce chloramides of glycosaminoglycans, perhaps through the binding of myeloperoxidase directly to the glycosaminoglycans. The N-Cl group in the chloramides is a potential target for reducing species such as Cu(I) and superoxide radicals. Laser flash photolysis has been used here to obtain, for the first time, the rate constants for the direct reaction of superoxide radicals with the chloramides of hyaluronan and heparin. The rate constants were in the range 2.2-2.7 × 10(3)M(-1)s(-1). The rate constant for the reaction with the amino acid taurine was found to be much lower, at 3.5-4.0 × 10(2)M(-1)s(-1). This demonstration that superoxide anion radicals react directly with hyaluronan and heparin chloramides may support the mechanism first proposed by M.D. Rees et al. (Biochem. J.381, 175-184, 2004) for an efficient fragmentation of these glycosaminoglycans in the extracellular matrix under inflammatory conditions.

Chain scission of hyaluronan by carbonate and dichloride radical anions: potential reactive oxidative species in inflammation?

The reactions of the carbonate and dichloride radical anions, CO3- and Cl2-, with the extracellular matrix glycosaminoglycan hyaluronan (HA) have been studied using the kinetic technique of pulse radiolysis and also by steady-state irradiation combined with gel permeation chromatography/multiangle laser light scattering(gpc/MALLS) to measure the rates of reaction with HA and the yield of HA chain scission, respectively. For comparison, the same measurements were made for the reactions of the free radicals *OH, Br2*-, and N3*. The carbonate and dichloride radical anions were found to react relatively quickly with HA (7.0 x 10(5) and 6.9 x 10(6) dm3 mol(-1) s(-1), respectively) although they are much less reactive than the hydroxyl radical, *OH. Significant yields (20 and 38%, respectively) of chain scission of HA by these radical anions were also determined from the gpc/MALLS experiments, providing some support for their potential participation in the depolymerization of HA in vivo. These results are compared with data obtained for the other free radicals (hydroxyl, azide radicals, and dibromide radical anions) investigated in this study in order to gain an insight into their mechanism of reaction with HA. Earlier chain scission yields of HA by hydroxyl radicals determined by the authors have also been revised using the gpc/MALLS technique employed in the current study. The yields of 52% (absence of air) and 44% (in air) are much lower than the previous values. In the current study, the effect of oxygen on the yields of HA chain breaks is discussed in terms of the reactivity of HA peroxyl radicals in the presence of superoxide radical anions. The relevance of the results of this study to mechanisms of inflammation is discussed.

Chain scission of hyaluronan by peroxynitrite.

The reaction of peroxynitrite with the biopolymer hyaluronan has been studied using stopped-flow techniques combined with detection of molecular weight changes using the combination of gel permeation chromatography and multiangle laser light scattering. From the effect of peroxynitrite on the yield of hyaluronan chain breaks, it was concluded that the chain breaks were caused by hydroxyl radicals which escape a cage containing the *OH NO*(2) radical pair. The yield of free hydroxyl radicals was determined as 5+/-1% (as a proportion of the total peroxynitrite concentration). At high peroxynitrite concentrations, it was observed that the yield of chain breaks leveled out, an effect largely attributable to the scavenging of hydroxyl radicals by nitrite ions present in the peroxynitrite preparation. These experiments also provided some support for a previous proposal that the adduct formed between ONOOH and ONOO(-) might itself produce hydroxyl radicals. The rate of this reaction would have to be of the order of 0.05 s(-1) to produce hydroxyl radical yields that would account quantitatively for chain break yields at high peroxynitrite concentrations. By carrying out experiments at higher hyaluronan concentrations, it was also concluded that an additional yield of chain breaks was produced by the bimolecular reaction of the polymer with ONOOH at a rate constant of about 10 dm(3)mol(-1)s(-1). At 5.3 x 10(-3)mol dm(-3) hyaluronan, this amounted to 3.5% chain breaks (per peroxynitrite concentration). These conclusions support the proposal that the yield of hydroxyl radicals arising from the isomerization of ONOOH to nitrate ions is relatively low.

The effect of hydroxyl radicals on the rheological performance of hylan and hyaluronan.

Shear flow, dynamic oscillation and extensional viscosity measurements were used to compare the rheological performance of several hylan samples (M(v) 1.6, 3.2, 3.7, 4.7 and 5.6x10(6)) and hyaluronan (M(v) 1.4 and 1.8x10(6)) before and after hydroxyl radicals (*OH) induced degradation. It was found that the higher molecular weight cross-linked structure of hylan was more resistant to degradation than hyaluronan and that this superior stability was reflected in various rheological parameters. The *OH degradation of the initial hylan and hyaluronan samples produced a range of polysaccharides based on hylan and hyaluronan with molecular weight covering a range from 0.5-5.6x10(6). The rheological parameters associated with the polysaccharides could then also be studied. Zero shear values of the complex viscosity (eta*), dynamic viscosity (eta') and shear viscosity (eta) were calculated using the method of Morris(1) and shown to approach the same value at zero shear or frequency. An adaptation of the method of Gibbs et al. gave a 'master curve' for the storage and loss modulus of hyaluronan and hylan, which encompasses a 10-fold molecular weight and a 5-fold concentration variation. In all instances for hylan, the storage modulus predominates over the loss modulus, whereas for hyaluronan, the reverse is true, demonstrating the greater elasticity of hylan throughout the whole experimental range of molecular weights and concentrations.

The incidence of drug-related problems as a cause of hospital admissions in children.

OBJECTIVES: To determine the incidence of hospital admissions for drug-related problems (DRPs) among children, and to examine cases for causality, preventability and clinical severity. DESIGN: Prospective assessment involving review of case notes and parent interview to determine if an admission was associated with a DRP. PATIENTS AND SETTING: All patients admitted to a large university-affiliated paediatric hospital in Melbourne, Victoria, for medical reasons (i.e., not surgical, trauma or oncology patients) during 56 consecutive days from 24 June to 19 August 1996 for which a DRP could be identified. Patients whose parents or guardians could not communicate adequately in English were excluded. MAIN OUTCOME MEASURES: The incidence, type, causality, preventability and clinical severity of DRPs associated with admission. RESULTS: Of 1682 eligible patients admitted to the Royal Children's Hospital during the study period, 58 admissions (3.4%) were associated with DRPs. Non-compliance was implicated in 50%. Causality was ranked as "definite" (34.5%), "possible" (56.9%) and "doubtful" (8.6). Two-thirds of admissions associated with DRPs were deemed preventable. Although no patients died from DRPs, four were admitted to the intensive care unit. CONCLUSIONS: The incidence of DRPs as a cause of hospital admission in this study falls within the range of incidences published for the Australian adult population (range, 2.4%-22%). In contrast to findings among Australian adults, a high proportion of admissions for DRPs in this study were associated with non-compliance. The high percentage of preventable admissions indicates that further study is necessary to characterise risk factors within this population and to test prevention strategies.

Intramolecular electron transfer in the dipeptide, histidyltyrosine: a pulse radiolysis study.

The technique of pulse radiolysis has been used to investigate the possibility of intramolecular charge transfer in the dipeptide histidyltyrosine, following one-electron oxidation of one of its amino acid residues. The radical anion, Br2.- was found to react with the dipeptide at pH 6.0 with a bimolecular rate constant of 2.3+/-0.2 x 10(7) dm3 mol(-1)s(-1) suggesting that it reacts very selectively with the histidine moiety. Spectral observations at, or close to the end of this reaction show only the presence of a tyrosinyl free radical (TyrO.), however, indicating that fast (>10(6) s(-1) intramolecular charge transfer has taken place between histidine radicals (His+.) and tyrosine (TyrOH). This finding was supported by the direct observation of the rate of formation of TyrO. in experiments with the free amino acids, histidine and tyrosine, under conditions where Br2.- reacted selectively with histidine. The bimolecular rate constant for the reaction between His+. and TyrOH was found to be 2.4+/-0.5 x 10(6) dm3 mol(-1)s(-1). Taken together, the results of the study indicate that His+. is a relatively strong oxidising agent where (E (His+./His) > 770 mV at pH 6.0.

Eucalyptus oil poisoning among young children: mechanisms of access and the potential for prevention.

We studied unintentional paediatric eucalyptus oil poisoning to identify potential intervention strategies. The epidemiology of paediatric eucalyptus oil poisoning in Victoria was determined by analysis of four databases. The sequence of events preceding ingestion was examined by a telephone survey involving 109 parents or guardians of children under five years involved in an actual or suspected ingestion of eucalyptus oil. Such children were identified prospectively from all callers during a nine-month period to the Victorian Poisons Information Centre and those presenting to the emergency departments of the participating hospitals of the Victorian Injury Surveillance System. Eucalyptus oil was a leading agent associated with hospitalisation for poisoning among Victorian children aged under five years. In the telephone survey, 90 incidents were found to involve vaporiser solutions, 15 eucalyptus oil preparations, and the remainder other eucalyptus-oil-containing products of a medicinal nature. Regardless of the type of product, 74 per cent gained access via a home vaporiser unit, most frequently placed at ground level. Although amounts ingested are usually small, the reported range of toxic doses is wide, necessitating at least a medical assessment following ingestion. Potential countermeasures identified in a consultative workshop included: discontinuing the use of eucalyptus oil as a therapeutic agent; confirmation that vaporiser-well residues are nontoxic; removal of barriers to product reregistration following safety-related modifications; improved child-resistant closures; discouraging vaporiser use for respiratory infections among young children; and development and dissemination of protocols for treatment of suspected ingestion.

Rodenticide poisoning among children.

We aimed to determine the nature, extent and sequence of events of accidental childhood poisoning with rodenticides and identify potential intervention strategies. Subjects were identified prospectively from callers to the Victorian Poisons Information Centre and those presenting to the Emergency Departments of hospitals participating in the Victorian Injury Surveillance System from 1 April to 31 December 1993. The events preceding rodenticide ingestion were examined via telephone questionnaire involving 128 parents or guardians of children under five years exposed to rodenticides. Rodenticides are not leading agents for severe poisoning but are a frequent and increasing cause of less severe poisoning. Most children (90 per cent) had obtained the rodenticide from the site at which it had been laid, usually by the caregiver (67 per cent), in the kitchen, lounge room or laundry, inside cupboards or-wardrobes. In 69 per cent of these cases, respondents thought the site would not normally allow access to children. Only 13 per cent of children were admitted to hospital, although 41 per cent sought medical attention. Most caregivers (90 per cent) were aware of some dangers associated with rodenticides. Children usually ingest insufficient amounts of rodenticide to cause serious effects. However, medical assessment and monitoring of prothrombin times is often indicated, with a consequent cost to the healthcare system. Potential countermeasures, focusing on packaging and positioning of rodenticide baits, product reformulation, and the distribution of management guidelines for health workers, were identified. Implementation of the identified countermeasures was initiated by a workshop involving stakeholders from industry, research and health professions.

Automatic dishwasher detergent poisoning: opportunities for prevention.

We investigated the antecedents of ingestion of dishwashing machine detergent to enable the development of effective countermeasures. Parents who had sought advice from the Victorian Poisons Information Centre about dishwasher detergent poisoning exposures of their children were interviewed by telephone. Almost all the children (94 per cent) were aged between 6 and 29 months. Of the 61 children included in the survey, 53 (87 per cent) gained access to the detergent from the dishwasher. Of these, 50 (94 per cent) took the detergent from the dispenser on the internal surface of the door of the machine, and 38 (76 per cent) of these ingested detergent remaining in the dispenser after operation of the machine. Parents were present in the room on 78 per cent of occasions at the time of ingestion. Most parents (72 per cent) were aware of the toxicity of the detergents. Relocation of the dispenser or redesigning it to prevent access both before and after operation would have prevented most of the exposures to detergent. Altering the detergent to prevent caking or sludging might prevent many of the exposures to detergent remaining in the dispenser after operation of the machine. The level of prior knowledge about toxicity suggests that education or additional warnings are unlikely to contribute substantially to prevention of poisoning. Telephone call-back to identified cases is a useful method of investigating complex poisoning problems and developing effective countermeasures.

Effect of an automated sink on handwashing practices and attitudes in high-risk units.

OBJECTIVE: To assess the effects of an automated sink on handwashing practices and attitudes of staff. DESIGN: Quasi-experimental crossover design. SETTING: Two high-risk patient care areas, one postanesthesia recovery room (Site 1), and one neonatal intensive care unit (Site 2) in two tertiary care hospitals. PARTICIPANTS: All patient care staff on study units; approximately 55 individuals. INTERVENTIONS: An automated sink was installed to replace one handwashing sink for about five weeks; the sink was then crossed-over for an equivalent time period to the other location. Handwashing practices of all unit staff were observed in three two-hour observation periods/week. Questionnaires were distributed to staff two weeks after sink installation and at the study's end. RESULTS: One thousand, six hundred ten handwashes were observed. Handwashing practices differed significantly by site. For both sites, hands were washed significantly better but significantly less often with the automated sink (all p less than .001). Staff expressed negative attitudes, however, about certain features of the sink, and these negative attitudes increased over the study period. CONCLUSIONS: Automated devices must be flexible enough to allow adjustments based on staff acceptance. Application of new technology to improve hand hygiene requires a multifaceted approach to behavior change.

One-electron reduction of 8-hydroxy-5-deazaisoalloxazine in aqueous solution: a pulse radiolysis study.

The one-electron reduction of 8-hydroxy-5-deazaisoalloxazine (HMDI) has been studied in aqueous solution in the acidity range pH 0 to 13 using the reducing species CO2.-, eaq- and (CH3)2COH radicals. The spectral and other properties of the HMDI radicals were found to be independent of the reductant used. Four protolytic forms of the radical were distinguished with associated pKa values of 2.3 +/- 0.3, 6.0 +/- 0.3 and 10.1 +/- 0.3.

The kinetics of hydroxyl-radical-induced strand breakage of hyaluronic acid. A pulse radiolysis study using conductometry and laser-light-scattering.

Hydroxyl radicals were generated radiolytically in N2O- and N2O/O2(4:1)-saturated aqueous solutions of hyaluronic acid. The hydroxyl radicals react rapidly with hyaluronic acid mainly by abstracting carbon-bound H atoms. As a consequence of subsequent free-radical reactions, chain breakage occurs the kinetics of which has been followed using the pulse radiolysis technique. In the absence of oxygen, strand breakage was followed by the change in conductivity induced by the release of cationic counterions condensed at the surface of hyaluronic acid which is a polyanion consisting of subunits of glucuronic acid alternating with N-acetyl-glucosamine. It appears that strand breakage is not due to one single first-order process, however, the contributions of the different components cannot be adequately resolved. At pH 7 the overall half-life is 1.4 ms, in both acid and basic solutions the rate of free-radical induced strand breakage is accelerated (at pH 4.8, t1/2 = 0.6 ms; at pH 10, t1/2 = 0.18 ms). In the absence of oxygen there is no effect of dose rate on the kinetics of strand breakage. In the presence of oxygen in addition to conductometric detection, strand breakage was also followed by changes in low-angle laser light-scattering. These two techniques are complementary in that in this system the conductometry requires high doses per pulse while the light-scattering technique is best operated in the low-dose range. In the presence of oxygen a pronounced dose-rate effect is observed, e.g. at pH 9.7 after a dose of 9.4 Gy the overall half-time is approx. 0.5 s, while after a dose of 6.6 Gy the half-time is approx. 0.23 s. Both the yield and the rate of strand breakage increase with increasing pH, e.g. at pH 7 G(strand breaks) = 0.7 x 10(-7) mol J-1 and at pH 10.4, 4.8 x 10(-7) mol J-7. The radiolytic yields of CO2, H2O2, organic hydroperoxides, O2.- and oxygen consumption have been determined in gamma-irradiated N2O/O2(4:1)-saturated solutions of both hyaluronic acid and beta-cyclodextrin.

One-electron reduction of 5-deazalumiflavin in aqueous solution: a pulse radiolysis study.

The one-electron reduction of 5-deazalumiflavin has been studied in aqueous solution in the acidity range H0 = -1 to pH 13 using the reducing species CO2-, e-aq and (CH3)2COH radicals. The spectral and other properties of the deazaflavin radicals formed were found to be independent of the reductant used. Four protolytic forms of the radical were distinguished with associated pKa values of 1.3 +/- 0.3, 6.0 +/- 0.3 and 10.7 +/- 0.3.

Superoxide radical reactions in aqueous solutions of pyrogallol and n-propyl gallate: the involvement of phenoxyl radicals. A pulse radiolysis study.

The reactions of O2-. in aqueous solutions of pyrogallol 1 and the antioxidant n-propyl gallate 2 have been studied. In both cases the initial reaction gives hydrogen peroxide and the corresponding phenoxyl radical (k(1 + O2-.) = 3.4 x 10(5), k(2 + O2-.) = 2.6 x 10(5) dm3 mol-1S-1). These phenoxyl radicals have been produced independently by reacting 1 and 2 with Br2-. and their spectra and first pKa values measured (pKa(phenoxyl radical from 1) = 5.1, pKa(phenoxyl radical from 2) = 4.1). It is necessary to correct the observed spectra for the contribution of the H-adducts, formed by the reaction of radiolytically produced H atoms with the substrates (k(1 + H) = 2.5 x 10(9), k(2 + H) = 3.8 x 10(9) dm3 mol-1 S-1). The H-adduct spectra are given. In the reactions of O2-. with the substrates the initial transient absorbances are characteristic of the phenoxyl radicals; however at longer times a new transient absorbing around 500 nm (epsilon congruent to 10(4) dm3 mol-1 cm-1) appears. This is believed to be the deprotonated hydroxy-orthoquinone, formed by the reaction of phenoxyl radicals with O2-. (k congruent to 1.5 x 10(8) dm3 mol-1 S-1, from kinetic curve-fitting). The absorbance due to the hydroxy-orthoquinones decays by first-order kinetics (1.6 x 10(2) in the case of 1 and 1.1 x 10(2) s-1 in the case of 2). This is thought to be mainly the result of the conversion of the hydroxy-orthoquinone into its hydrate. Similar experiments were carried out with catechol and ethyl protocatechuate. The chemistry appears to be similar to that of the pyrogallol derivatives. The rate constant for reaction of these compounds with O2-. is, however, only less than or equal to x 10(4) dm3 mol-1 s-1.

The possible involvement of changes in phosphoinositol turnover in the responses of renal sodium transport to noradrenaline.

Noradrenaline stimulated the formation of [3H]inositol monophosphates from 3H-inositol-phospholipids in rat kidney cortex slices. This response was inhibited by the alpha 1-adrenoceptor antagonist prazosin but not by the alpha 2-adrenoceptor antagonist yohimbine, suggesting that the response was mediated via an alpha 1-adrenoceptor. This is in keeping with the ability of noradrenaline to increase sodium transport in this tissue by alpha 1-adrenoceptor stimulation.

Long-term use of intranasal insulin in insulin-dependent diabetic patients.

This study, in 3 phases, compared the long-term acceptability and efficacy of insulin administered by nasal spray with an intensified subcutaneous regimen in nine type I (insulin-dependent) diabetic subjects on baseline therapy with ultralente insulin. In phase 1, patients were begun and stabilized on a regimen of ultralente daily and Actrapid insulin three times daily. Phase 2 consisted of 4 mo of this intensified subcutaneous regimen. In phase 3, intranasal administration of insulin, with 1% (wt/vol) sodium glycocholate, replaced Actrapid insulin for 4 mo. Glycemic control was compared in each of the three phases. It was possible to maintain the dose of ultralente insulin relatively constant in only six of the nine subjects during the intranasal phase of the study. The six subjects showed a significant rise in glycosylated hemoglobin during the intranasal phase (10.4 +/- 0.6% intranasal vs. 9.1 +/- 0.3% subcutaneous, P less than .05) but not in plasma or urinary glucose levels. There was no significant change in the incidence of hypoglycemic episodes during intranasal insulin therapy in this group. The other three subjects were considered treatment failures. Six of the nine original subjects expressed a preference for intranasal insulin, and one subject complained of mild nasal irritation insufficient to cease treatment. The intranasal route of administration of insulin has the potential to replace short-acting insulin as an adjunct to longer-acting insulin in some insulin-treated diabetic patients.