Isoprostanes as a tool to investigate oxidative stress in scleroderma spectrum disorders—advantages and limitations

Sir, Scleroderma spectrum disorders represent heterogeneous disease with a large variation of clinical manifestations in individual patients. At one end of the spectrum is the limited systemic sclerosis according to LeRoy [1], which includes the presence of Raynaud's phenomenon (RP) and a typical capillaroscopy scleroderma pattern or positivity for Scl-70 or anti-centromere antibodies. At the other end of the spectrum are limited and diffuse cutaneous systemic sclerosis, which associates cutaneous changes and visceral involvement. The major hallmarks of these diseases are vascular dysfunction, immunological activation and tissue fibrosis. Increased oxidative stress is one of the phenomenons that may explain the link between the three faces of the disease. Quantification of oxidative stress in vivo is an important issue that can be approached by measuring F₂-isoprostanes. F₂-isoprostanes are a complex family of compounds produced from arachidonic acid via a free-radical-catalysed mechanism. The first demonstration that these compounds were produced in humans was shown in 1990 by Morrow et al. [2], who reported the discovery of prostaglandin-F₂-like compounds generated by free-radical-induced peroxidation of arachidonic acid. Because these compounds are isomeric to prostaglandins and have an F-type cyclopentane (prostane) ring, these compounds were termed F₂-isoprostanes. Since that time, F₂-isoprostanes have been used extensively as clinical markers of lipid peroxidation in vascular disorders [3]. Several favourable attributes make measurement of F₂-isoprostanes a reliable biomarker of oxidative stress in vivo. Isoprostanes are stable in urine, where levels are present in detectable quantities, their formation increases in models of oxidant injury and are modulated by anti-oxidant status, but their levels are not affected by lipid content of the diet [4]. Among the numerous F₂-isoprostanes stereoisomers, 15-F_2t-IsoP is currently mostly used as accurate clinical biomarker of lipid peroxidation [5].

Urinary and F₂-isoprostane levels are increased in patients with systemic sclerosis [6]. Their levels are increased in patients at the early stages of the disease [7], but not in patients with primary (RP) [8]. Furthermore, 15-F_2t-IsoP urinary levels correlate to the lung involvement and the nailfold videocapillaroscopy pattern [9], and inversely correlate to the skin post-occlusive hyperaemia [10]. These data strongly suggest that lipid peroxidation is increased in scleroderma spectrum disorders. Lipid peroxidation level relates to the extent of pulmonary and vascular damage, and may represent a biomarker of aggressive disease. Among the biological fluids available, most studies were performed on urine because of the non-invasiveness of the procedure and the lack of artefactual generation. The recent data from Ogawa et al. [11] appear interesting at first sight: free F₂-isoprostane levels would be increased in serum as well as urine. However, several major limitations prevent drawing any conclusion from this data. The authors show that free serum F₂-isoprostane levels are elevated by 75-fold compared with controls, with 99% of the patients’ values being higher than the mean + 3 s.d. of the control group. This clear-cut result would be amazing for such a non-specific biomarker. This should have risen questions about potential bias. F₂-isoprostanes were initially discovered as an auto-oxidation product of plasma samples kept at −20°C [12]. There is no rationale for the use of serum samples for measuring compounds that may be produced ex vivo. Indeed, ex vivo generation of isoprostane is highly susceptible to occur during clot formation, which is why most investigators using blood samples quantify isoprostane on plasma samples. Blood is collected with anti-coagulants, and promptly centrifuged at 4°C. The half-life of F₂-isoprostane is short, being <20 min in the rat. Therefore, the choice of plasma vs urinary measurement is debatable for quantifying lipid peroxidation in chronic disease. The authors also specify that serum samples were obtained over the last 7 years. This implicates that serum F₂-isoprostanes are stable over this large period of time, ensuring that spontaneous auto-oxidation does not occur, and we have no evidence that this is validated. Therefore, a different time between blood sampling and storage at −80°C, and a different storage time is likely to explain the difference observed between controls and patients with scleroderma. Another hint that artefactual generation occurred is the tremendous dispersion of values, as some patients exhibited isoprostane levels more than 10 000 times higher than the control levels. Another pitfall that should have been discussed is renal failure, a common comorbidity in patients with severe systemic sclerosis. Clearance of F₂-isoprostanes is due both to renal filtration and metabolism. The most commonly quantified isoprostane 15-F_2t-IsoP leads to two major metabolites in humans: 2,3-dinor-15-F_2t-IsoP and 2,3-dinor-5,6-dihydro-15-F_2t-IsoP [13, 14], but this compound is also filtrated and quantified in urine. Therefore, renal failure, by itself may contribute to the increased plasma levels via decreased excretion. Finally, the authors specify that they followed the specifications of Cayman® enzyme immunoassay kit. However, there is no indication that any purification step was applied prior to the measurement, while the lack of purification may lead to inconsistent results.

From the previous available data on urine, there is a consistent increase in isoprostane levels in patients with scleroderma, ranging from 1.25 to 3 times compared with controls [6–10]. The 75-fold increase in serum levels does not compare with it, and is likely due in most part to an artefactual generation or a decreased clearance. Further studies are required to determine the link between oxidative stress and vascular damage in systemic sclerosis. F₂-isoprostane quantification exhibits many advantages that promote it as a good tool to investigate lipid peroxidation in humans. However, urinary measurements, using an initial purification step, provide a time-integrated index of systemic F₂-isoprostane formation, and do not exhibit the pitfall of blood measurements. They should be favoured by investigators in an attempt to quantify F₂-isoprostane as a tool to investigate oxidative stress in patients with scleroderma.

The author has declared no conflicts of interest.

References