Malondialdehyde (MDA) is a widely used marker of oxidative lipid injury whose concentration varies in response to biotic and abiotic stress. Commonly, MDA is quantified as a strong light-absorbing and fluorescing adduct following reaction with thiobarbituric acid (TBA). However, plant tissues in particular contain many compounds that potentially interfere with this reaction and whose concentrations also vary according to the tissue type and stress conditions. As part of our studies into the stress responses of plant tissues, we were interested in developing a rapid, accurate, and robust protocol for MDA analysis using reverse-phased HPLC to avoid these problems with reaction specificity. We demonstrate that a partitioning step into n-butanol during sample preparation is essential and that gradient HPLC analysis is necessary to prevent sample carryover between injections. Furthermore, the starting composition of the mobile phase must be sufficiently hydrophobic to allow direct injection of the n-butanol extracts without peak splitting, tailing, and other artifacts. To minimize analysis times, we used a short, so-called "Rocket" HPLC column and high flow rates. The optimized HPLC separation has a turnaround time of 2.5 min per sample. Butanolic extracts of MDA(TBA)2 were stable for at least 48 h, and recoveries were linear between 0.38 and 7.5 pmol NIDA added. Importantly, this procedure proved to be compatible with existing extraction procedures for L-ascorbate and glutathione analysis in different plant species, allowing multiple "stress metabolite" analyses to be carried out on a single tissue extract. (c) 2005 Elsevier Inc. All rights reserved.