Letters to the Editor THE USE OF PHENYLALANINE TO DETECT HYDROXYL RADICAL PRODUCTION IN VIVO: A CAUTIONARY NOTE To the Editors: Phenylalanine (both L- and D-isomers) reacts with hy- droxyl radical (OH • ) to form three hydroxylated prod- ucts: para-, meta-, and ortho tyrosine. These are usually formed in comparable amounts, although the fate of the initial adduct radicals can vary depending on the reaction conditions. Formation of p-tyrosine from L-phenylala- nine can be catalyzed by phenylalanine hydroxylase, whereas m- and o-tyrosine are not regarded as normal metabolites. Hence changes in these isomers have been used as an index of the production of OH • in several systems including human lens proteins, rhematoid arthri- tis, ozone exposure, and ischemia–reperfusion [1]. We previously reported that levels of p-, m-, and o-tyrosine increased in plasma of cats subjected to myo- cardial ischemia–reperfusion [2]. However, we have identified an artifact that can occur during high-perfor- mance liquid chromatography analysis of m-tyrosine. Arterial plasma samples collected from cats were ana- lyzed for tyrosines using an absorbance detector as pre- viously [2], except that an electrochemical detector (ECD; ESA, Coulochem II, guard electrode = 0.7 V, first electrode = 0.4 V, second electrode = 0.65 V) was attached in tandem. The detection limit for the tyrosine isomers was 10 times lower for the ECD. In some sam- ples collected from certain animals, the m-tyrosine quan- titated by the absorbance detector was 100 times more than the ECD. We suspect coelution of a compound with the m-tyrosine, which absorbs at 274 nm but is not ECD sensitive. The fraction was collected and analyzed by photodiode array. The m- and o-tyrosine have absorption maxima at 214 and 272 nm, respectively, but the fraction collected had an absorbance maxima at 248 nm, that is, it is not exclusively m-tyrosine. Attempts to identify components in the fraction by derivatization and GC/MS showed numerous compounds, none as yet conclusively identified. The control samples obtained from animals subjected to surgical manipulation and administration of phenylalanine but no ischemia–reperfusion did not con- tain the unknown compound(s), suggesting formation and release during ischemia–reperfusion injury. As with any high-performance liquid chromatography technique, one must carefully check for overlapping peaks. In these studies, a peak(s) was identified overlap- ping with m-tyrosine, and spiking failed to show that it was not m-tyrosine. Hence, diode array and ECD are recommended when aromatic hydroxylation of phenyl- alanine is used to detect OH • in vivo. One should be especially cautious when the m-tyrosine peak is much larger than that of o-tyrosine. It is possible that this artifact may have confused some earlier results [2]. Acknowledgements — We thank Dr. A. van der Vliet for helpful dis- cussions. We also thank Roberta H. Holt and Kameron Chun for their help and patience. This study was supported by the PPG, HL 52165-04, and HL 36527. Sharanya Reddy Center for Comparative Respiratory Biology and Medicine University of California Davis, CA, U.S.A. Barry Halliwell Department of Biochemistry National University of Singapore Singapore A. D. Jones Department of Chemistry Pennsylvania State University University State Park, PA, U.S.A. John C. Longhurst Department of Medicine University of California Irvine, CA, U.S.A. PII S0891-5849(99)00165-3 REFERENCES [1] Kaur, H.; Halliwell, B. Detection of hydroxyl radicals by aromatic hydroxylation. Methods Enzymol 233:67– 82; 1994. [2] O’Neill, C. A.; Fu, L.-W.; Halliwell, B.; Longhurst, J. C. Hydroxyl radical production during myocardial ischemia and reperfusion in cats Am. J. Physiol. 271:H660 –H666; 1996. Free Radical Biology & Medicine, Vol. 27, Nos. 11/12, p. 1465, 1999 Copyright © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0891-5849/99/$–see front matter 1465