Annals ofthe Rheumatic Diseases 1996; 55: 383-387 383 Protection against peroxynitrite dependent Ann Rheum Dis: first published as 10.1136/ard.55.6.383 on 1 June 1996. Downloaded from tyrosine nitration and a1-antiproteinase inactivation by some anti-inflammatory drugs and by the antibiotic tetracycline Matthew Whiteman, Harparkash Kaur, Barry Halliwell Abstract physiological pH to give a range of noxious Objective-To examine in vitro the ability products with reactivities resembling those of of several drugs to protect against del- hydroxyl radicals (OH), nitrogen dioxide eterious effects of peroxynitrite, a cyto- (NO2%), and nitronium ion (NO2+).8 toxic agent formed by reaction of nitric Addition of peroxynitrite to tissues and oxide with superoxide radical, that may be biological fluids leads to nitration of aromatic generated in the rheumatoid joint and amino acid residues, and the presence of these could cause joint damage. may be a 'marker' of peroxynitrite mediated Methods-The ability of several drugs to (NO" dependent) damage in vivo.6 9-11 Indeed, protect against such possible toxic actions nitrotyrosine is present in plasma and synovial ofperoxynitrite as inactivation of Gl-anti- fluid from patients with RA, but it was not proteinase and nitration of tyrosine was detected in healthy control subjects.'2 Nitration evaluated. of tyrosine residues on proteins may interfere Results-Most non-steroidal anti-inflam- with cell signal transduction by tyrosine matory drugs were moderately (indo- phosphorylation/dephosphorylation.'3 Peroxy- methacin, diclofenac, naproxen, tolmetin) nitrite also inactivates otl-antiproteinase, the or only weakly (sulindac, ibuprofen, auro- major inhibitor of serine proteases (such as thioglucose, flurbiprofen, sulphasalazine, elastase) in human body fluids,7 14 and con- salicylate, penicillamine disulphide) ef- siderable inactivation of otl-antiproteinase has fective in preventing tyrosine nitration been shown to occur in the inflamed rheuma- and a,-antiproteinase inactivation by toid joint.'5 peroxynitrite, but 5-aminosalicylate and RA is treated by the use of a number penicillamine were much more effective, of drugs, including non-steroidal anti- http://ard.bmj.com/ as was the antibiotic tetracycline (but not inflammatory drugs (NSAIDs) such as aspirin ampicillin). Phenylbutazone and flu- and indomethacin,.and slower acting 'disease fenamic acid protected effectively against modifying' drugs, such as penicillamine and tyrosine nitration, but could not be tested sulphasalazine, the mechanism of action of in the a,-antiproteinase system. The anal- which is not always clear. gesic paracetamol was highly protective in We have examined the ability of a range of both assay systems. drugs used in the treatment of RA to protect on September 26, 2021 by guest. Protected copyright. Conclusion-Many drugs used in the against damage by ONOO-. Two measures of treatment of rheumatoid arthritis are un- damage that may be relevant to events in vivo likely to act by scavenging peroxynitrite. were used: ability of the drug to prevent The feasibility ofperoxynitrite scavenging inactivation of ca1-antiproteinase by ONOO-, as a mechanism of penicillamine, and ability of the drug to inhibit the nitration 5-aminosalicylate, and paracetamol of tyrosine by ONOO-. action in vivo is discussed. (Ann Rheum Dis 1996; 55: 383-387) Materials and methods REAGENTS N-succinyl (ala)3-p-nitroanilide (SANA), cata- In chronic inflammatory diseases such as rheu- lase (type C40), elastase (E0258), al-anti- matoid arthritis (RA) and inflammatory bowel proteinase (A9024), DL-tyrosine, and the anti- disease, there is known to be overproduction of inflammatory drugs were from Sigma oxygen derived species such as superoxide Chemical Corp, Poole, Dorset, UK. Solutions King's College London, radical, O2*.1 There is also considerable evi- of sulindac, paracetamol, ibuprofen, diclo- Pharmacology Group, dence that production of nitric oxide (NO") is fenac, ampicillin, and tetracycline were made Manresa Road, increased-for example from measurements of up in distilled water; those of indomethacin, London SW3 6LX, United Kingdom increased nitric oxide synthase activity and tolmetin, flurbiprofen, sulphasalazine, sulpha- M Whiteman demonstration of increased concentrations of pyridine, and piroxicam were made up in water H Kaur nitrite/nitrate in body fluids from patients with with minimum potassium hydroxide added to B Halliwell RA.2A Nitric oxide reacts with a very high rate ensure solution. Salicylate, 5-aminosalicylate, Correspondence to: B Halliwell. constant with 02-- to give peroxynitrite aurothioglucose, penicillamine, and penicilla- Accepted for publication (ONOO-).5 6 Peroxynitrite can be directly mine disulphide were dissolved in 500 mmol/l 6 February 1996 cytotoxic' 7 and it may also decompose at potassium dihydrogen phosphate-dipotassium 3.8434Whiteman, Kaur, Haliwell hydrogen phosphate (KH2PO4-K2HPO4) saline, pH 7-4 (140 mmol/l sodium chloride, buffer pH 7-4; phenylbutazone was dissolved 2-7 mmol/l potassium chloride, 16 mmol/I in 0 4% (w/v) sodium carbonate. Solutions disodium hydrogen phosphate, 2-9 mmol/l Ann Rheum Dis: first published as 10.1136/ard.55.6.383 on 1 June 1996. Downloaded from were made up freshly every day and diluted KH2PO4) to a concentration of 4 mg/ml and with distilled water to obtain the required drug elastase was dissolved in the same buffer to a concentrations. final concentration of 5 mg/ml. The volume of ot,-antiproteinase required to inhibit elastase by 80-90% (typically 60-70 gl) was added to PEROXYNITRITE SYNTHESIS6 buffer (500 mmol/I K2HPO4-KH2PO4 pH 7A4), Five millilitres of an acidic solution (0-6 mol/l with or without 0-1 ml of compound to be hydrochloric acid) of hydrogen peroxide tested, to give a volume of 0 945 ml, and (H202) 0 7 mol/l was mixed with 5 ml of incubated in a water bath at 37°C for 15 potassium nitrite 0-6 mol/l on ice for one minutes, when peroxynitrite (typically 5 j.l) second and the reaction quenched with 5 ml was added to give a final concentration of 0 5 of ice cold sodium hydroxide 1-2 moll. The mmol/l. The sample was vortexed for 10 stock was then frozen overnight (-20°C) and seconds and incubated for five minutes, the top layer of the solution collected for the elastase (usually 50 ,ul) was added and the experiment. Concentrations of stock ONOO- sample further incubated at 37°C for 15 min- were redetermined before each experiment utes, followed by addition of 2-0 ml of buffer. using a molar absorption coefficient of 1670 Then after 15 minutes, 0 1 ml of elastase (moll)-1 cm-' at 302 nm.6 substrate (SANA) was added and the rate of reaction followed at 410 nm for 30 seconds. MEASUREMENT OF TYROSINE NITRATION DL-Tyrosine solutions were made up to a final Results concentration of 10 mmolI by dissolving the INACTIVATION OF ctl-ANTIPROTEINASE: required amount in 8 ml of water with 250 ,l PROTECTION BY DRUGS of 10% (w/v) potassium hydroxide followed by As expected,7 14 addition of ONOO- to al-anti- 250 pl of 5% phosphoric acid and dilution to proteinase led to inactivation of the ability of 10 ml by the addition of 1-5 ml of water. otl-antiproteinase to inhibit elastase. The in- Tyrosine solution 0 1 ml together with 0 1 ml activation was complete within five minutes at of a solution of the compound to be tested was pH 7*4 (data not shown). added to a plastic test tube containing 0 795 The extent of inactivation increased with ml of buffer (500 mmol/l K2HPO4-KH2PO4 ONOO- concentration and a five minute incu- pH 7 4) and incubated in a water bath at 37°C bation time with 0 5 mmol/l ONOO- was for 15 minutes. After this time peroxynitrite selected for further studies. Investigation of the (typically 5 gl) was added to a final concen- possibility that other constituents of the tration of 1 mmol/I and the tubes were vortexed ONOO- solution (such as NO3- and H202, http://ard.bmj.com/ for 15 seconds and incubated for a further 15 which can be present in ONOO- preparations6) minutes. The pH was measured after the were involved in the inactivation demonstrated addition of peroxynitrite and found to be that catalase (final concentration 103 U/ml) between 7-4 and 7-5. had no effect on inactivation of otl-anti- Formation of 3-nitrotyrosine was measured proteinase by ONOO-, and that, when the by high performance liquid chromatography ONOO- solution was added to buffer and (HPLC), rather than spectrophotometrically, to incubated for five minutes at 37°C before the on September 26, 2021 by guest. Protected copyright. avoid interference as a result of the generation addition of aol-antiproteinase, the resulting of chromogen from the drugs. Measurement of 'decomposed ONOO-' solution (which still 3-nitrotyrosine was performed essentially as contains all the other contaminants6) had no described previously8 using a Spherisorb 5 ,um effect on oxt-antiproteinase (data not shown). ODS2 C18 column (25 cm X 4-6 mm) (HPLC Figure 1 summarises the ability of a range of Technology, Wellington House, Cheshire, UK) drugs to protect ac,-antiproteinase against in- with a guard column C,8 cartridge (Hibar, activation by ONOO-. There was wide vari- BDH, Poole, UK). The eluent was 500 mmolI ation in the protective effects seen. Significant K2HPO4-KH2PO4 pH 3-01, with 20% meth- protection was observed with penicillamine, anol (v/v) at a flow rate of 1 ml min-' through paracetamol, indomethacin, 5-aminosalicylate, a Polymer Laboratories pump (Essex Road, diclofenac, naproxen, and tolmetin, but flurbi- Church Stretten, Shropshire, UK); the ultra- profen, salicylate, sulphasalazine, sulindac, violet detector was set at 274 mm (sensitivity aurothioglucose, ibuprofen, and penicillamine 0-02). The 3-nitrotyrosine detected was con- disulphide were weakly effective. Sulpha- firmed by spiking with standards.