| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Medical Chemistry, Biochemistry and Biotechnology, School of Medicine, University of Milan, Via Saldini 50, 20133 Milan, Italy;
2 Dipartimento di Medicina, Chirurgia e Odontoiatria, Università di Milano, H. San Paolo, Via Di Rudini 8, 20142 Milan, Italy;
3 Department of Chemistry, Purdue University, West Lafayette, IN 47907-1393;
4 IRCCS H San Raffaele, Via Olgettina 60, 20132 Milan, Italy;
aauthor for correspondence: fax 39-2-50316040, e-mail giuliana.cighetti{at}unimi.it
Malondialdehyde (MDA), a compound derived from lipid peroxidation and from eicosanoid biosynthesis, exists in biological matrices both in the free form and bound to SH and/or NH2 groups of various biomolecules (1). Although other compounds (isoprostanes) have been proposed as more reliable indicators of oxidative damage (2), MDA is still widely used in clinical chemistry laboratories to monitor oxidative stress (3). Several methods have been developed to evaluate MDA in biological samples (1)(4), but the different analytical conditions used and the lack of a suitable internal standard have led to large discrepancies in measurements even at physiologic MDA concentrations in human plasma (5).
We (6) recently reported a "reference method" for free and total plasma MDA quantification, as the phenylpyrazole derivative, by isotope-dilution gas chromatography–mass spectrometry (ID-GC-MS) with dideuterated MDA (d2-MDA) as internal standard. This method, used for clinical MDA detection (7)(8)(9), offers the possibility of validating other proposed internal standards that differ from MDA in structure, stability, and reactivity. Unfortunately, the major limitations of d2-MDA include its difficult synthesis (10) and that it is detectable only by GC-MS, a method not always available in clinical laboratories. A compound that appears to be more suitable as an internal standard is methyl malondialdehyde (MMDA) because it is structurally close to MDA, is absent from biological matrices, is easily obtainable from a commercial compound, and is detectable by common methods such as HPLC, GC, and capillary electrophoresis.
MMDA was first evaluated as an internal standard for MDA determinations by Bull and Marnett (11), who unfortunately experienced difficulties in resolving the underivatized MDA and MMDA by HPLC. Recently, Claeson et al. (12) reported the use of MMDA as an internal standard for measurement of MDA in rat brain by capillary electrophoresis, thus avoiding the derivatization step. We successfully adapted this method for detection of MDA in rat liver microsomes and human plasma (13) and tested it against the ID-GC-MS method.
Claeson et al. (12) also proposed the use of MMDA for free MDA detection by GC with prior derivatization with pentafluorophenylhydrazine. However, some problems remained unresolved, such as the performance of MMDA in human plasma, the most frequently used clinical biological matrix, and the validation of results by ID-GC-MS.
In the present study, we reconsidered the use of MMDA to quantify free and total MDA in human plasma after phenylhydrazine derivatization, focusing our attention on its use in ID-GC-MS. In addition, we established the optimum conditions (pH value and reaction time) to maximize simultaneous derivatization of MMDA and MDA.
MMDA was easily prepared by alkaline hydrolysis of 2-methyl-3-ethoxyprop-2-enal (Sigma) (13) and purified as the sodium salt. The melting point of our compound was 365 °C (in a sealed tube); the melting point reported in the literature is 360 °C (14). The calculated composition for C4H5O2Na, based on combustion analysis, is C, 44.45%; H, 4.66%. The composition of our compound, based on combustion analysis, was C, 44.50%; H, 4.60%. The proton nuclear magnetic resonance results were as follows (500 MHz, D2O, 
): 8.90 (s, 2H, CHO), 2.00 (s, 3H, CH3). Pure MDA and d2-MDA, as the corresponding sodium salts, were obtained as reported previously (10)(15). Stock solutions (11 µmol/L) of MMDA, MDA, and d2-MDA in 50 mmol/L potassium phosphate buffer (pH 7.0) were stored at -85 °C. Their concentrations were calculated based on the molar absorptivity of MMDA at 274 nm (
= 2.990 x 104 M-1 cm-1) and of MDA and d2-MDA at 267 nm ( = 3.180 x 104 M-1 cm-1).
GC-MS analyses in electron impact mode were carried out on a Hewlett-Packard 5890 gas chromatograph, equipped with a HP-5 fused-silica capillary column, coupled to a 5988A mass spectrometer as described previously (6). Derivatization of MMDA and MDA with phenylhydrazine yielded 4-methyl-1-phenylpyrazole and 1-phenylpyrazole, respectively. The mass spectra of these pure derivatives showed specific molecular ions at m/z 158 and 144 as the most intense peaks, respectively (Fig. 1
); these ions were selected for detection of the MMDA and MDA derivatives. The chromatographic profiles of the MDA and MMDA derivatives showed well-resolved peaks. No endogenous interfering ions at m/z 158 (at the retention time of the MMDA derivative) were observed in plasma, and no ions at m/z 144 (at the retention time of the MDA derivative) were present in the 4-methyl-1-phenylpyrazole GC-MS analysis.
|
Preliminary experiments carried out to convert MMDA to 4-methyl-1-phenylpyrazole, as established previously for the parent MDA (6) (25 °C; pH 2.0–4.0; reaction time, 30–60 min), showed low derivatization. Better results with quantitative derivatization yields were obtained at higher pH values (pH 4.0–5.0) for reaction times of 60 and 90 min. We therefore studied the derivatization of both MMDA and MDA after adjusting the pH of the reaction mixture to 4.0. The optimized conditions [60-min reaction time and a strictly controlled pH (4.0)] afforded higher derivatization yields for a mixture of both aldehydes.
The preanalytical treatments for free and total MDA evaluations included a modification of those reported for d2-MDA (6). For free MDA, 0.2 mL of plasma enriched with MMDA (0.26 nmol; 20 µL of stock solution) was mixed with butylated hydroxytoluene (BHT; 5 nmol; 10 µL of 0.5 mmol/L in ethanol) and citrate buffer (0.4 mol/L, pH 4.0; up to a final reaction volume of 0.5 mL), and MDA was derivatized with phenylhydrazine (1 µmol; 20 µL of 50 mmol/L aqueous solution). After 60 min at room temperature, the samples were extracted with 1 mL of hexane. The organic phases were concentrated to 50 µL under nitrogen and analyzed by GC-MS (1 µL injected). For total MDA, 0.2 mL of plasma enriched with MMDA (0.26 nmol; 20 µL of stock solution) was mixed with BHT (8 nmol; 16 µL of 0.5 mmol/L ethanol solution), NaOH (0.5 mmol; 100 µL of 5 mol/L aqueous solution), and water up to 0.5 mL and hydrolyzed at 60 °C for 60 min. The pH was then adjusted to 4.0 by adding citrate buffer (1.2 mL of 2 mol/L solution), derivatization was carried out as reported for free MDA detection, and the samples were extracted with 1.5 mL of hexane. Citrate buffer (2 mol/L; pH 4.0) was prepared by dissolving citric acid monohydrate (5.50 g; 28.6 mmol) and citric acid trisodium salt dihydrate (5.26 g; 20.4 mmol) in water (25 mL).
Fresh plasma EDTA obtained from healthy individuals (n = 8) was pooled, immediately stored at -85 °C, and used to determine the linearity of the method. Calibration curves were prepared by adding MMDA (0.26 nmol; 20 µL of stock solution) and increasing amounts of MDA (0–1.6 nmol of stock solution) to water or pooled plasma (0.2 mL). After the addition of BHT, the samples were treated as reported above for both free and total MDA measurements. Regression analysis was performed by plotting the integrated peak-area ratios of the MDA and MMDA derivatives against the ratios of the known added amounts of MDA and MMDA. The results are reported as the mean ± SD (n = 10). The regression lines for the free MDA calibration curves in water [y = 0.98(± 0.02)x + 0.002; r2 = 0.999] and plasma [y = 0.98(± 0.03)x + 0.37; r2 = 0.999] indicated no interference of the biological matrix during the preanalytical steps. The within- and between-day CVs were obtained by injecting derivatized stock MDA solutions (0.2 and 2 µmol/L) seven times on the same day and twice daily for 5 consecutive days, respectively. The within-day CVs were 1.2% (n = 7) for 0.2 µmol/L MDA and 0.8% (n = 10) for 2 µmol/L MDA. The between-day CVs were 1.5% and 1.2%, respectively.
For total MDA evaluation, the regression lines of the calibration curves prepared in water and pooled plasma were: y = 0.98(± 0.02)x + 0.002 (r2 = 0.999) and y = 0.92(± 0.06)x + 1.32 (r2 = 0.999), respectively. Initial attempts to adjust the pH to 4.0, after alkaline hydrolysis by HCl addition, gave calibration curve slopes from 0.45 to 1.4, showing the critical importance of the pH. In fact, small variations in pH caused irreproducible results because of the incomplete derivatization of MDA (slope <1; pH >4.0) or MMDA (slope >1; pH <4.0). The best and most reproducible way to precisely adjust the pH to 4.0 after alkaline hydrolysis was the addition of 2 mol/L citrate buffer. The similarity of the slopes for calibration curves obtained for free and total MDA determinations indicated the chemical stability of MMDA during the hydrolysis conditions used and the absence of interference from 2 mol/L citrate buffer. The within-day CVs were 2.0% (n = 7) for 1 µmol/L MDA and 1.6% (n = 10) for 5 µmol/L MDA. The between-day CVs were 2.1% and 2.3%, respectively.
To validate the use of MMDA, free and total MDA were quantified in plasma from healthy individuals (n = 50) and patients with cardiovascular disease (6 patients with unstable angina and 6 with stable angina) recruited randomly from a large group of individuals participating in a previous study (7). The MDA concentrations were independently quantified by the method reported here (GC-MS with MMDA as internal standard) and by GC-MS with d2-MDA. The obtained results (Table 1
) were superimposable and not significantly different (by Student t-test), confirming that MMDA fulfills the requirements for its use in plasma MDA detection after derivatization with phenylhydrazine, provided that the conditions reported here (pH 4.0 and 60-min reaction time) are strictly followed.
|
In conclusion, we have shown that MMDA is a suitable internal standard for MDA detection in human plasma, validating its use in ID-GC-MS. More importantly, we optimized the reaction conditions for the derivatization of MDA and of MMDA to provide the corresponding phenylpyrazoles in comparable yields.
Acknowledgments
We wish to thank Andrea Lorenzi for technical assistance in MS. This work was supported by grants from the Italian Ministry of University and Scientific Research (MURST; 60%)
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  G. Cighetti, R. Paroni, S. Marzorati, E. Borotto, R. Giudici, G. Magnanini, and G. Iapichino Evaluation of Oxidative Stress in Serum of Critically Ill Patients by a Commercial Assay and Gas Chromatography-Mass Spectrometry Clin. Chem., August 1, 2005; 51(8): 1515 - 1517. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
