AIMS: Accurate analysis of dinucleotide redox cofactors nicotinamide adenine dinucleotide phosphate reduced (NADPH), nicotinamide adenine dinucleotide phosphate (NADP), nicotinamide adenine dinucleotide reduced (NADH), and nicotinamide adenine dinucleotide (NAD) from biological samples is important to understanding cellular redox homeostasis. In this study, we aimed to develop a simple protocol for quenching metabolism and extracting NADPH that avoids interconversion among the reduced forms and the oxidized forms.

RESULTS: We compared seven different solvents for quenching and extraction of cultured mammalian cells and mouse tissues: a cold aqueous buffer commonly used in enzyme assays with and without detergent, hot aqueous buffer, and cold organic mixtures (80% methanol, buffered 75% acetonitrile, and acidic 40:40:20 acetonitrile:methanol:water with either 0.02 M or 0.1 M formic acid). Extracts were analyzed by liquid chromatography-mass spectrometry (LC-MS). To monitor the metabolite interconversion, cells were grown in C-glucose medium, and unlabeled standards were spiked into the extraction solvents. Interconversion between the oxidized and reduced forms was substantial except for the enzyme assay buffer with detergent, 80% methanol and 40:40:20 acetonitrile:methanol:water, with the 0.1 M formic acid mix giving the least interconversion and best recoveries. Absolute NAD, NADH, NADP, and NADPH concentrations in cells and mouse tissues were measured with this approach.

INNOVATION: We found that the interconversion between the reduced and oxidized forms during extraction is a major barrier to accurately measuring NADPH/NADP and NADH/NAD ratios. Such interconversion can be monitored by isotope labeling cells and spiking NAD(P)(H) standards.

CONCLUSION: Extraction with 40:40:20 acetonitrile:methanol:water with 0.1 M formic acid decreases interconversion and, therefore, is suitable for measurement of redox cofactor ratios using LC-MS. This solvent is also useful for general metabolomics. Samples should be neutralized immediately after extraction to avoid acid-catalyzed degradation. When LC-MS is not available and enzyme assays are accordingly used, inclusion of detergent in the aqueous extraction buffer reduces interconversion. Antioxid. Redox Signal. 28, 167-179.