A specific, accurate, and sensitive measure of total plasma malondialdehyde by HPLC

Malondialdehyde (MDA) is one of the most commonly reported biomarkers of lipid peroxidation in clinical studies. The reaction of thiobarbituric acid (TBA) with MDA to yield a pink chromogen attributable to an MDA-TBA2 adduct is a common assay approach with products being quantified by ultraviolet-Vis assay as nonspecific TBA-reactive substances (TBARS) or chromatographically as MDA. The specificity of the TBARS assay was compared with both chromatographic assays for total plasma MDA. The levels of total plasma MDA were significantly lower than the plasma TBARS in each of the samples examined, and interestingly, the interindividual variation apparent in the level of plasma MDA was not evident in the plasma TBARS assay. Each of the four online chromatographic detectors yielded a precise, sensitive, and accurate determination of total plasma MDA, and selected-ion monitoring was the most-accurate assay (101.3%, n = 4). The online diode array detectors provided good assay specificity (peak purity index of 999), sensitivity, precision, and accuracy. This research demonstrates the inaccuracy that is inherent in plasma TBARS assays, which claim to quantify MDA, and it is proposed that the TBARS approach may limit the likelihood of detecting true differences in the level of lipid peroxidation in clinical studies. Malondialdehyde (MDA) is one of the most commonly reported biomarkers of lipid peroxidation in clinical studies. The reaction of thiobarbituric acid (TBA) with MDA to yield a pink chromogen attributable to an MDA-TBA2 adduct is a common assay approach with products being quantified by ultraviolet-Vis assay as nonspecific TBA-reactive substances (TBARS) or chromatographically as MDA. The specificity of the TBARS assay was compared with both chromatographic assays for total plasma MDA. The levels of total plasma MDA were significantly lower than the plasma TBARS in each of the samples examined, and interestingly, the interindividual variation apparent in the level of plasma MDA was not evident in the plasma TBARS assay. Each of the four online chromatographic detectors yielded a precise, sensitive, and accurate determination of total plasma MDA, and selected-ion monitoring was the most-accurate assay (101.3%, n = 4). The online diode array detectors provided good assay specificity (peak purity index of 999), sensitivity, precision, and accuracy. This research demonstrates the inaccuracy that is inherent in plasma TBARS assays, which claim to quantify MDA, and it is proposed that the TBARS approach may limit the likelihood of detecting true differences in the level of lipid peroxidation in clinical studies. diode array detection limit of detection limit of quantitation malondialdehyde relative standard deviation single ion monitoring thiobarbituric acid thiobarbituric acid-reactive substance tetraethoxypropane ultraviolet Lipid peroxidation is one of the most-commonly reported indices of oxidative stress and has been implicated as a contributing factor in a range of degenerative diseases, including diabetes (1Pasupathi P. Chandraseker V. Kumar U.S. Evaluation of oxidative stress, antioxidant and thyroid hormone status in patients with diabetes mellitus.J. Med. 2009; 10: 60-66Crossref Scopus (12) Google Scholar), cardiovascular disease, Parkinson's disease, Alzheimer's disease (2Adly A.A.M. Oxidative stress and disease: an updated review.Res. J. Immunol. 2010; 3: 129-145Crossref Scopus (103) Google Scholar), and psychiatric disorders, including schizophrenia (3Tsaluchidu S. Cocchi M. Tonnello L. Puri B.K. Fatty acids and oxidative stress in psychiatric disorders.BMC Psychiatry. 2008; 8 (Suppl): 1-3Crossref PubMed Scopus (98) Google Scholar). The process of lipid peroxidation results in a range of intermediates and end products including lipid hydroperoxides, aldyehydes, and malondialdehyde (MDA). These aldehydes and lipid hydroperoxides form DNA adducts and may result in extensive single-strand and double-strand breaks (4Devipriya N. Sudheer A.R. Vishwanathan P. Menon V.P. Modulatory potential of ellagic acid, a natural plant, on altered lipid profile and lipid peroxidation status during alcohol-induced toxicity: a pathohistological study.J. Biochem. Mol. Toxicol. 2008; 22: 101-112Crossref PubMed Scopus (32) Google Scholar). Various intermediates and end products generated during the lipid peroxidation cascade have been assayed, but the most-commonly employed approach continues to be the thiobarbituric acid (TBA) test (5Baron C.P. Bro R. Skibsted L.H. Andersen H.J. Direct measurement of lipid oxidation in oil in water emulsions using multiwavelength derivate UV-spectroscopy.J. Agric. Food Chem. 1997; 45: 1741-1745Crossref Scopus (17) Google Scholar). Although it is widely acknowledged that TBA reacts with a range of oxidized lipids, both saturated and unsaturated aldehydes, (6Alessio, H. M., 2000. In Handbook of Oxidants and Antioxidants in Exercise. O. Hanninen, L. Packy, and C. K. Sen, editors. Elsevier, Amsterdam. 115–128.Google Scholar, 7Kosugi H. Kikugawa K. Thiobarbituric acid reaction of aldehydes and oxidised lipids in glacial acetic acid.Lipids. 1985; 20: 915-921Crossref Scopus (123) Google Scholar) sucrose, and urea (8Buttkus H. Bose R.J. Amine-malonaldehyde condensation products and their relative color contribution in the thiobarbituric acid test.J. Am. Oil Chem. Soc. 1972; 50: 387-391Google Scholar), to form various chromogens, referred to as TBA-reactive substances (TBARS), it is the reaction of TBA with MDA to produce a pink pigment (9Bernheim F. Berheim M.L. Wilbur K.M. The reaction between thiobarbituric acid and the oxidation products of certain lipids.J. Biol. Chem. 1948; 174: 257-264Abstract Full Text PDF PubMed Google Scholar) with an absorption maximum at 532 nm (10Yagi K. A simple fluorometric assay for lipoperoxide in blood plasma.Biochem. Med. 1976; 15: 212-216Crossref PubMed Scopus (2050) Google Scholar) and mass ion at 323 amu (11Jardine D. Antoloovich M. Prenzler P.D. Robards K. Liquid chromatography-mass spectrometry (LCMS) investigation of the thiobarbituric acid reactive substances (TBARS) reaction.J. Agric. Food Chem. 2002; 50: 1720-1724Crossref PubMed Scopus (61) Google Scholar) that is a true indicator of lipid peroxidation. Historically, the TBARS test has been assayed by ultraviolet (UV) spectrophotometry or fluorescence assays, but the specificity of the TBA test, enabling the selective determination of MDA, may be achieved via chromatographic separation of the pink TBA2-MDA adduct (12Bird R.P. Hung S.O.S. Hadley M. Draper H.H. Determination of malondialdehyde in biological matrices by high pressure liquid chromatography.Anal. Biochem. 1983; 128: 240-244Crossref PubMed Scopus (201) Google Scholar). The adoption of HPLC techniques improves assay specificity and the sensitivity of MDA determination, and we have previously confirmed the validity of quantifying the TBA2-MDA adduct as a specific measure of lipid peroxidation in human biological fluids (13Boyle S.P. Dobson V.L. Duthie S.J. Hinselwood D.C. Kyle J.A.M. Collins A.R. Bioavailability and efficiency of rutin as an antioxidant: a human supplementation study.Eur. J. Clin. Nutr. 2000; 54: 774-782Crossref PubMed Scopus (159) Google Scholar). Despite the widely acknowledged limitations of the TBARS test (14Wheatley A.R. Some recent trends in the analytical chemistry of lipid peroxidation.TRAC. 2000; 19: 617-628Google Scholar), and in particular its lack of specificity, it continues to be reported as a true measure of MDA in clinical disease (1Pasupathi P. Chandraseker V. Kumar U.S. Evaluation of oxidative stress, antioxidant and thyroid hormone status in patients with diabetes mellitus.J. Med. 2009; 10: 60-66Crossref Scopus (12) Google Scholar, 15Zortea K. Fernandes B.S. Guimaraes L.R. Francesconi L.P. Lersch C. Gama C.S. Schroeder R. Zanotto-Filho A. Moreira J.C. Lobato M.I.R. et al.Reduced serum non-enzymatic antioxidant defense and lipid peroxidation in schizophrenic patients on a hypocaloric diet.Neurosci. Lett. 2012; 512: 43-47Crossref PubMed Scopus (4) Google Scholar, 16Karacay O. Sepici-Dincel A. Kacaaltincaba D. Sahin D. Yalvac S. Akyol H. Kanemir O. Altan N. A quantitative evaluation of total antioxidant status and oxidative stress markers in pre eclampsia and gestational diabetic patients in 24–36 weeks of gestation.Diabetes Res. Clin. Pract. 2010; 89: 231-238Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). It is proposed that the poor assay specificity associated with TBARS assays may lead to an overestimation of the levels of MDA in human plasma and other biological tissues and fluids, and this, in turn, may limit the likelihood of detecting true differences in the level of lipid peroxidation in clinical studies. Employing two chromatographic assays HPLC-diode array detection (DAD)-fluoro and LC/MS-DAD, this research investigated the absolute levels of plasma MDA and compared these with the level of TBARS determined by UV spectrophotometry. Within physiological systems, MDA may exist in a free state or in a protein-bound form; various studies have looked at the relative proportions of MDA in human plasma and have reported that between 83% and 92% of the MDA is protein bound (17Carbonneau M.A. Penchant E. Sess D. Caloni P. Clerc M. Free and bound malondialdehyde measured as thiobarbituric acid adduct by HPLC in serum and plasma.Clin. Chem. 1991; 37: 1423-1429Crossref PubMed Scopus (152) Google Scholar, 18Hong Y-L. Yeh S-L. Chiang C-Y. Hu M.L. Total plasma malondialdehyde levels in 16 Tiawanese college students determined by various thiobarbittiric acid tests and an improved high performance liquid chromatography-based method.Clin. Biochem. 2000; 33: 619-625Crossref PubMed Scopus (117) Google Scholar). Hydrolysis of the protein-bound MDA fraction may be achieved by either an alkaline treatment (18Hong Y-L. Yeh S-L. Chiang C-Y. Hu M.L. Total plasma malondialdehyde levels in 16 Tiawanese college students determined by various thiobarbittiric acid tests and an improved high performance liquid chromatography-based method.Clin. Biochem. 2000; 33: 619-625Crossref PubMed Scopus (117) Google Scholar, 19Mateos R. Lecumberri E. Ramos S. Goya L. Bravo L. Determination of malondialdehyde (MDA) by high-performance liquid chromatography in serum and liver as a biomarker for oxidative stress. Application to a rat model for hypercholesterolemia and evaluation of the effect of diets rich in phenolic antioxidants from fruits.J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2005; 827: 76-82Crossref PubMed Scopus (302) Google Scholar) or an acid treatment (20Londero D. Lo Greco P. Automated high-performance liquid chromatographic separation with spectrofluorimetric detection of a malondialdehyde-thiobarbituric acid adduct in plasma.J. Chromatogr. A. 1996; 729: 207-210Crossref PubMed Scopus (97) Google Scholar) of the sample prior to the acid-catalyzed TBA reaction. An approximately two-fold increase in plasma MDA levels determined by HPLC-UV was observed following alkaline hydrolysis of human plasma by Hong et al. (18Hong Y-L. Yeh S-L. Chiang C-Y. Hu M.L. Total plasma malondialdehyde levels in 16 Tiawanese college students determined by various thiobarbittiric acid tests and an improved high performance liquid chromatography-based method.Clin. Biochem. 2000; 33: 619-625Crossref PubMed Scopus (117) Google Scholar), and it was concluded that this additional sample pretreatment was a useful step that enhanced assay reliability, ensuring a more-complete and uniform release of the MDA. In light of this and similar studies (21Grotto D. Santa Maria L.D. Boeira S. Valentini J. Charao M.F. Moro A.M. Nascimento P.C. Pomblum V.J. Garcia S.C. Rapid quantification of malondialdehyde in plasma by high performance liquid chromatography-visible detection.J. Pharm. Biomed. Anal. 2007; 43: 619-624Crossref PubMed Scopus (258) Google Scholar) that report good assay precision when measuring total plasma MDA, a similar alkaline hydrolysis method was employed herein. Reagent-grade sodium hydroxide, sulphuric acid, trichloroacetic acid, and HPLC-grade solvents were purchased from Fisher Scientific. 1,1,3,3 Tetraethoxypropane (TEP) and 2-TBA were purchased from Sigma Chemical Co. Acetic acid (99.5% pure) was purchased from Arcos Organics. HPLC-grade water (H2O) was used throughout. Participant recruitment for this collaborative research study was managed by clinicians at the University of the United Arab Emirates, and the study protocol was approved by the Research Ethics Committee, Faculty of Medicine and Health Sciences, United Arab Emirates University. The study was conducted in accordance with good clinical practice, and all participants gave their informed prior to was from each by The blood was at for the plasma was and at for plasma samples were used for assay and samples from two were used to test the of the assay. was used as the MDA and of acid and with acid to form a adduct An alkaline hydrolysis step was employed (21Grotto D. Santa Maria L.D. Boeira S. Valentini J. Charao M.F. Moro A.M. Nascimento P.C. Pomblum V.J. Garcia S.C. Rapid quantification of malondialdehyde in plasma by high performance liquid chromatography-visible detection.J. Pharm. Biomed. Anal. 2007; 43: 619-624Crossref PubMed Scopus (258) Google Scholar), enabling a measure of total plasma MDA to be determined by HPLC or were in to which were human plasma or and of The were and in a water for alkaline of sulphuric acid and of were The were and and of was to to which was of The were and to for the were at for and an of the was chromatographically for MDA or by UV spectrophotometry for determination of TBARS (10Yagi K. A simple fluorometric assay for lipoperoxide in blood plasma.Biochem. Med. 1976; 15: 212-216Crossref PubMed Scopus (2050) Google Scholar). TBARS were using a were and were measured at 532 nm (10Yagi K. A simple fluorometric assay for lipoperoxide in blood plasma.Biochem. Med. 1976; 15: 212-216Crossref PubMed Scopus (2050) Google Scholar). HPLC separation was at using a liquid chromatography with an diode array and a fluorescence was achieved using a and a as in A of was and the was at 532 nm (10Yagi K. A simple fluorometric assay for lipoperoxide in blood plasma.Biochem. Med. 1976; 15: 212-216Crossref PubMed Scopus (2050) Google Scholar) by were the range of to The fluorescence used an of nm and an of nm (8Buttkus H. Bose R.J. Amine-malonaldehyde condensation products and their relative color contribution in the thiobarbituric acid test.J. Am. Oil Chem. Soc. 1972; 50: 387-391Google Scholar). An to an was with a analytical at and a of The in The was to measure in the and were a mass range of amu with monitoring at 323 other for the were standard for the chromatographic quantification was by standard using as were the range of MDA, and of MDA (peak were Determination of total TBARS used the assay and of UV absorption at 532 nm were and precision were determined in accordance with the on for of C. H. of Text and at Scholar). purity was by C. H. of Text and at Scholar), and sensitivity of the chromatographic assays was determined in of of detection and of quantitation from the method of et al. D. A. of to determination and for the of detection and quantification for by 2009; PubMed Scopus Google Scholar) using the standard deviation of the and the was determined by a process of of human plasma = and by HPLC and were to test for differences between the various chromatographic TBARS of the a pink chromogen that a absorption at 532 The of the assay was the range of and the precision of assay was with a of standard deviation = n = A similar pigment was observed following TBARS of the but in this the UV a maximum absorption at 532 nm and a absorption at nm of the MDA standard by and yielded a single HPLC with a in the range of and was n = for each of the four online a of 532 and the mass ion was at 323 of the MDA-TBA2 adduct assay of the plasma samples and by and yielded a single HPLC with a in the range of and was n = for each of the four online a of 532 nm and the mass ion was at 323 of the MDA-TBA2 adduct The of MDA determination the range of was on all four online chromatographic as in and the precision of assay was good n = on each of the four detectors = of a plasma sample in of determination of total plasma MDA The assay to be with = and in the absolute of MDA determined by each of total plasma MDA = in a The sensitivity of the assay was in a human plasma and the was in the range of for the array detectors and an in the range of An and an of and were achieved by fluorescence at 323 nm and of and The following of the plasma samples was from a mass of and were observed for other at and The of the assay was by a method of in the human plasma The were and yielded in the range of for each of the four detectors yielded the most-accurate the for were a high purity was following of both the standard and the human plasma samples by that assay specificity and were using this The assay was to human to assay performance in of precision, specificity, and for = of each plasma sample were the total MDA assay and by and The precision of assay was good n = on each of the four detectors The plasma MDA was and in and The assays were employed to a of the levels of TBARS and plasma MDA using test The test two of human plasma and two of MDA at 532 nm that the of TBARS in the plasma from in = to in = The plasma TBARS level in and was with and as in The of plasma samples was by and the level of plasma MDA was The levels of plasma MDA were significantly lower than the plasma levels of TBARS in all four of the test MDA levels from in plasma to in was a in the level of plasma MDA measured in and an level of plasma TBARS been determined A between the plasma MDA and TBARS was observed = not n = plasma It has been reported that acid of TBARS and a specific determination of lipid peroxidation the for HPLC D. R. The of free lipid antioxidant and by oxidative stress as to the clinical Med. Biol. PubMed Scopus Google Scholar). it is acknowledged that the specificity of the MDA assay as a true indicator of lipid peroxidation in biological matrices (13Boyle S.P. Dobson V.L. Duthie S.J. Hinselwood D.C. Kyle J.A.M. Collins A.R. Bioavailability and efficiency of rutin as an antioxidant: a human supplementation study.Eur. J. Clin. Nutr. 2000; 54: 774-782Crossref PubMed Scopus (159) Google Scholar) is achieved when a chromatographic assay is employed (12Bird R.P. Hung S.O.S. Hadley M. Draper H.H. Determination of malondialdehyde in biological matrices by high pressure liquid chromatography.Anal. Biochem. 1983; 128: 240-244Crossref PubMed Scopus (201) Google Scholar). This research investigated the absolute levels of plasma MDA by two chromatographic assays and compared this with the level of TBARS determined by UV and the and sensitivity of each of the detectors A between the plasma MDA and TBARS was observed = not n = plasma A to in the of total MDA was when determined by HPLC or measured as TBARS in human This with that reported by Hong et al. (18Hong Y-L. Yeh S-L. Chiang C-Y. Hu M.L. Total plasma malondialdehyde levels in 16 Tiawanese college students determined by various thiobarbittiric acid tests and an improved high performance liquid chromatography-based method.Clin. Biochem. 2000; 33: 619-625Crossref PubMed Scopus (117) Google Scholar) and that TBARS determination by UV spectrophotometry to an overestimation of the levels of plasma MDA. in this the interindividual differences observed in plasma MDA levels were not apparent in plasma TBARS This may be to the poor assay of the TBARS which to an assay The of the was evident from the that two absorption and that in a human plasma the TBARS method specificity for MDA a chromatographic separation of the MDA-TBA2 adduct is for a true measure of lipid peroxidation to be The in human plasma may be attributable to the reaction of TBA with H. Kikugawa K. Thiobarbituric acid reaction of aldehydes and oxidised lipids in glacial acetic acid.Lipids. 1985; 20: 915-921Crossref Scopus (123) Google Scholar) the of acid adducts with MDA (11Jardine D. Antoloovich M. Prenzler P.D. Robards K. Liquid chromatography-mass spectrometry (LCMS) investigation of the thiobarbituric acid reactive substances (TBARS) reaction.J. Agric. Food Chem. 2002; 50: 1720-1724Crossref PubMed Scopus (61) Google Scholar). is to be common in MDA assays, but may be from the adduct of by of the HPLC the HPLC assay in an on previously reported assays for total MDA in of efficiency of sample with for the sample step of et al. (21Grotto D. Santa Maria L.D. Boeira S. Valentini J. Charao M.F. Moro A.M. Nascimento P.C. Pomblum V.J. Garcia S.C. Rapid quantification of malondialdehyde in plasma by high performance liquid chromatography-visible detection.J. Pharm. Biomed. Anal. 2007; 43: 619-624Crossref PubMed Scopus (258) Google Scholar). The of total plasma MDA was to that in previously (21Grotto D. Santa Maria L.D. Boeira S. Valentini J. Charao M.F. Moro A.M. Nascimento P.C. Pomblum V.J. Garcia S.C. Rapid quantification of malondialdehyde in plasma by high performance liquid chromatography-visible detection.J. Pharm. Biomed. Anal. 2007; 43: 619-624Crossref PubMed Scopus (258) Google Scholar). of MDA were by all four online HPLC and was of or on that the acid and alkaline hydrolysis the Although the method with monitoring a specific and accurate measure n = of the TBA2-MDA it is that the of may be for the or method a selective determination of total plasma MDA, which sensitivity, that the plasma MDA was in the range of to and a assay of = a similar of in the TBARS method has been reported by et al. F. N. of serum malondialdehyde levels determined by two in patients with HPLC or TBARS 2005; Scholar) in a study of patients with disease in which reported a increase in levels of free MDA determined by TBARS compared with that determined by by Hong et al. (18Hong Y-L. Yeh S-L. Chiang C-Y. Hu M.L. Total plasma malondialdehyde levels in 16 Tiawanese college students determined by various thiobarbittiric acid tests and an improved high performance liquid chromatography-based method.Clin. Biochem. 2000; 33: 619-625Crossref PubMed Scopus (117) Google Scholar), reported an increase in total plasma MDA in a the study by et al. N. L. M. F. S. Bravo L. R. et of of lipid peroxidation measurement in human plasma Res. 2010; PubMed Scopus Google Scholar), which investigated the and performance of a range of lipid peroxidation assays in human a study of that measured MDA using a range of It reported precision of between and MDA when by HPLC in that the method with a precision of = being Although the study of et al. N. L. M. F. S. Bravo L. R. et of of lipid peroxidation measurement in human plasma Res. 2010; PubMed Scopus Google Scholar) not the performance of a it report on a assay the method of et al. D. K. N. J.C. J. of with malondialdehyde and to a assay of lipid Res. Toxicol. PubMed Scopus Google Scholar). This assay the MDA plasma level and was approximately than the MDA level determined by an of assay when assays employed for plasma or serum MDA a of research (18Hong Y-L. Yeh S-L. Chiang C-Y. Hu M.L. Total plasma malondialdehyde levels in 16 Tiawanese college students determined by various thiobarbittiric acid tests and an improved high performance liquid chromatography-based method.Clin. Biochem. 2000; 33: 619-625Crossref PubMed Scopus (117) Google Scholar, F. N. of serum malondialdehyde levels determined by two in patients with HPLC or TBARS 2005; Scholar, D. K. N. J.C. J. of with malondialdehyde and to a assay of lipid Res. Toxicol. PubMed Scopus Google Scholar) have HPLC as the MDA assay for human plasma but these the continues to the of research TBARS as a specific measure of MDA peroxidation (12Bird R.P. Hung S.O.S. Hadley M. Draper H.H. Determination of malondialdehyde in biological matrices by high pressure liquid chromatography.Anal. Biochem. 1983; 128: 240-244Crossref PubMed Scopus (201) Google Scholar, and altered lipid peroxidation in of schizophrenia with and Res. PubMed Scopus Google M. S. Gama C. M. K.M. M. F. serum and thiobarbituric acid reactive substances in of and in Biol. Psychiatry. 2008; PubMed Scopus Google Scholar). In this research demonstrates the inaccuracy that is inherent in TBARS assays, which claim to quantify MDA in biological tissues and fluids, and it is proposed that the of may limit the likelihood of detecting true differences in the level of lipid peroxidation in clinical studies. with