Summary. Malonyldialdehyde (MDA) formation, a measure of polyunsaturated fat autoxidation, was estimated in normal human red cells incubated in vitro. Exposure to oxygen under a variety of conditions did not induce autoxidation. Exposure to hydrogen peroxide was either by the addition of a hydrogen peroxide solution or by incubation in an atmosphere saturated with hydrogen peroxide vapour. A pattern of MDA formation was established with both methods. Lack of reproducibility, a feature of previous methods of measuring the susceptibility of red cells to exogenous peroxide, could be overcome by (i) catalase inhibition, (ii) attention to the non‐linear relation between packed‐cell volume and MDA formation, and (iii) elimination of potentially misleading coloured complexes on spectroscopy. A number of recognized antioxidants inhibited peroxide‐induced MDA formation. Inhibition was proportional to the logarithm of the antioxidant concentration. Under certain conditions pre‐incubation with nucleosides and with lecithin protected the cells against autoxidation on subsequent exposure to peroxide. Peroxide‐induced cell autoxidation was also influenced by plasma, bovine albumin, and ascorbic acid. Haemolysis was an invariable consequence of autoxidation. Similarities and dissimilarities between autoxidation and antioxidation in free lipid emulsions and in organized biological structures were examined and discussed in the light of the experimental findings.

1. Fresh ox-brain homogenate prepared under standard conditions at 4°C and stored at -20°C autoxidizes spontaneously and reproducibly when re-heated to 37°C.2. The preparation can be used for assaying the antioxidant activity of serum and of other biological fluids. The preparatory and assay procedures are described.3. The antioxidant activity of normal serum, cerebrospinal fluid, synovial fluid and of various known antioxidants have been compared.Key words : antioxidant assay.The powerful antioxidant effect of serum is now well recognized (Barber, 1961;Vidlakova, Erazimova, Horki & Placer, 1972), and recent work suggests that variations in serum antioxidant activity (AOA) may have considerable clinical relevance (Dormandy, Hoare, Gutteridge, Stocks & Dormandy, 1974). These findings have created the need for a standard biological antioxidant assay, i.e. an autoxidizing system against which the inhibitory effect of serum and of other biological fluids can be measured. The assay described in the present paper is based on the spontaneous autoxidation of a standard ox-brain homogenate. MATERIALS A N D METHODSOx brains were obtained from recently slaughtered animals and conveyed to the laboratory packed in ice. Human serum was separated from blood collected from healthy laboratory staff. It was allowed to clot at room temperature. Serum samples were stored for up to 48 h at 4°C or for up to 4 weeks at -20°C. Synovial fluid samples were from patients with osteoarthritis. Cerebrospinal fluid samples were taken from specimens which had been sent to the hospital service laboratory for analysis and which had proved normal. Butylated hydroxyani-

Summary. The susceptibility of human red cells to autoxidation was measured (i) in various groups of normal subjects, (ii) in haemolytic states, and (iii) in non‐haemolytic disease. Many types of haemolytic disease—notably thalassaemia major and autoimmune haemlytic anaemia—were associated with a greatly and persistently increased susceptibility of the red cell lipids to autoxidation. In two patients with haemolytic disease this was the only biochemical abnormality found. Susceptibility to autoxidation is about three times higher in the newborn period than it is in adult life; and there is a comparatively wide scatter in old age.

No abstract

SUMMARY Using recently developed methods for measuring free-radical oxidation products in biological material, plasma extracts were studied in 24 women in the first two trimesters of pregnancy, in 124 women in the third trimester of pregnancy, in 20 women with pre-eclamptic toxaemia (PET), and in a control group. There was a significant progressive rise of two groups of free-radical oxidation products throughout pregnancy and a significantly greater rise in PET. In women whose diastolic blood pressure rose to above 70 mmHg there was a highly significant relation between two groups of free-radical reaction products and blood pressure.Methods weeks in 97 cases and at 40 weeks in 33 cases. A further study was carried out on eight women who were admitted with PET, the diagnosis being based on a rise in diastolic blood pressure (BP) to 90 mmHg or over with proteinuria or oedema or both. A control series of 14 non-pregnant women (aged 26·4 ± 7 years) included at least six on various oral contraceptive agents. The results in this group did not differ significantly from those in women during the first trimester of pregnancy, all falling within a very narrow range. Although a correlation between BP and some FR reaction products was not anticipated, the BP was measured and recorded at the time the blood samples were collected.Free-radical reaction products were measured by methods based on those described by Lunec and Dormandy," A scanning fluorescence spectroscope (MPF-3L, Perkin-Elmer, Beaconsfield, Bucks, UK) was used for all fluorescence measurements: 8 ml chloroform/methanol (2:1 v/v) was added to 1 ml plasma. The mixture was shaken for 2 minutes followed by centrifugation for 10 minutes at 1000 g. A preliminary study was carried out on blood from Of the resultant lower phase 5 ml was removed, and 36 women, 12 each in the first, second, and third 2 m1 deionised water was added. After vortex trimesters of pregnancy. The last eight weeks of mixing for 1 minute the mixture was spun for pregnancy were then studied in greater detail. Blood 10 minutes at 1000g. The lower phase formed will be was collected from 112 consecutive primigravid referred to as the 'chloroform' phase, and the upper women (aged 24· 6 ± 4· 6 years) attending the as the 'aqueous-methanol' phase. Two measurements antenatal clinic at the Whittington Hospital. Repeat were performed on the chloroform phase: 'diene samples from the same women were collected at 36 conjugation' measured by absorption at 240 nm; and 158