| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Institute of Community Health, Odense University, Odense, Denmark.
a Address correspondence to this author at: Department of Environmental Medicine, Odense University, Winsløwparken 17, DK-5000 Odense C, Denmark. Fax +45 65 91 14 58; e-mail F.Nielsen{at}Winsloew.ou.dk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Key Words: indexing terms: lipid peroxidation • reference intervals • HPLC
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One of the prominent risk factors for increased lipid peroxidation is smoking. Because of the presence of free radicals in cigarette smoke (1)(2), increases in P-MDA may occur (3)(4)(5). Nevertheless, others failed to identify any correlation between smoking status and P-MDA (6)(7)(8). In interpreting this evidence as well as other published P-MDA results, the size of the studies, the selection of subjects, and the quality of the P-MDA assay must be taken into account.
We have quantified P-MDA by the thiobarbituric acid (TBA) test. TBA-reactive substances (TBARS) formed in plasma, urine, or tissue samples after a calibrated sample pretreatment procedure primarily consist of MDA, which forms a red adduct with two molecules of TBA (MDA-TBA2) (9). The adducts are separated by an HPLC method originally described by Wong et al. (10) and Carbonneau et al. (11), but modified for improvement of the selectivity from nonidentified substances in plasma. Reference values for P-MDA for several individuals, with a selection between different age groups as well as gender, have been reported only in a selected group of blood donors (12). The aim of this study was to establish and validate an HPLC-based method for analysis of P-MDA, to determine a reference interval in a population-based sample group, and to assess the possible influence of smoking on P-MDA.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Analysis of the questionnaires indicated that the participants did not deviate in any significant way from the general Danish population. Approximately 60% of our reference population was employed (the remaining participants were attending school or university, retired, or unemployed). Forty-three percent were smokers, 23% had stopped smoking, and 32% had never smoked.
statistics
All statistical analyses of relations between P-MDA and age,
gender, and life-style factors were performed by the statistical
package SPSS for Windows, Version 6.1.3 (SPSS, Chicago, IL). Analysis
was performed on loge-normalized data. Correlations were
calculated as Pearson's correlation coefficients. For comparison of
means we used an independent t-test. Reference intervals
were calculated as recommended by IFCC (13), with REFVAL
3.43 (14). The data were standardized to zero mean and
unit variance before exponential and modulus transformation to adjust
for skewness and kurtosis. The final distribution was not significantly
different from gaussian (Anderson–Darling's A2 = 0.255,
P 
1.0).
preanalytical factors
Each participant was invited to come to either the Department of
Environmental Medicine at Odense University, or the local health center
in Søndersø. A home visit was also offered as an alternative. We
reviewed the completed questionnaire. All sampling of blood was
performed between 1400 and 1800 h during the months of September
and October, 1994. All blood samples were obtained by the same
phlebotomist. The participants were placed in a supine posture and
blood was drawn from a cubital vein into 10-mL Venoject tubes with 0.1
mL of 0.47 mol/L EDTA as anticoagulant (Terumo Europe, Leuven,
Belgium). Blood samples were kept at 5 °C until centrifugation at
1000g (3 h at maximum). Plasma was distributed into sterile
Cryo Vials (Greiner labortechnik, Frickenhausen, Germany) in volumes of
500 µL and was immediately frozen to -80 °C until analysis.
chemicals and reagents for analytical hplc
2-TBA, potassium dihydrogen phosphate, potassium hydroxide, sodium
hydroxide, and orthophosphoric acid 85% were of analytical grade and
purchased from Merck (Darmstadt, Germany). 1,1,3,3-Tetraethoxypropane
(TEP) was purchased from Sigma (St. Louis, MO). Methanol was of
HiPerSolv grade and obtained by BDH Laboratory Supplies (Poole, UK).
All water used was demineralized twice and filtered through a Milli-RO
10 Plus and a Milli-Q Plus plant (final pore size 0.2 µm; Millipore,
Bedford, MA). A 10 mmol/L potassium dihydrogen phosphate solution was
prepared and adjusted to pH 6.8 with 2.0 mol/L potassium hydroxide. The
solution was filtered by vacuum through a 0.45-µm nylon 66 membrane
(Supelco 5–8060; Bellefonte, PA). A 42 mmol/L 2-TBA solution was
prepared by dissolving 0.6 g of 2-TBA in 80 mL of water, and then
stirring and heating to 35–40 °C. The solution was cooled to room
temperature and filled with water to 100 mL. The solution was stable
for 2 days at 5 °C. The mobile phase consisted of 60:40 (by vol) 10
mmol/L potassium dihydrogen phosphate, pH 6.8:methanol. The mobile
phase was degassed by vacuum and sonification before use.
hplc instrumentation and conditions
A Kontron HPLC system (Kontron Instruments, Zürich,
Switzerland) consisting of a Kontron HPLC pump 420, a Kontron HPLC 360
autosampler with a 50-µL injection loop, and a Kontron UV detector
430 equipped with a 3-µL flow cell was used. The system was
controlled through a Kontron Multiport Module and a personal computer
(Victor 433D). The column was a LiChroCART® 250–4 packed
with 5-µm LiChrospher® 100 RP-18 (Merck). The column was
equipped with a guard column: LiChroCART 4–4 packed with 5-µm
LiChrospher 100 RP-18. The elution was carried out at a flow rate of
0.5 mL/min. The column effluent was quantified at a wavelength of 532
nm.
sample pretreatment
We added 100 µL of plasma to a 10-mL Pyrex centrifugation tube
containing 700 µL of 1% orthophosphoric acid, and vortex-mixed for
10 s. We then added 200 µL of 42 mmol/L 2-TBA solution, screwed
the Teflon-lined cap on tightly, and vortex-mixed the sample for
10 s, and then heated it for 60 min in a water bath at 100 °C.
Hereafter, the sample was kept on ice until 10 min before HPLC
analysis. At that time the sample was vortex-mixed for 10 s, and
200 µL was transferred into a 2.0-mL Micrewtube (Simport Plastics,
Québec, Canada) containing 200 µL of 1:12 (by vol) 2 mol/L
sodium hydroxide:methanol. The sample was vortex-mixed for 10 s
and centrifuged for 3 min at 13 000g. We transferred 200
µL of the supernatant to a 300-µL glass vial, and injected a
50-µL aliquot onto the column. A calibration solution was prepared
from TEP solubilized in water. TEP undergoes hydrolysis to liberate
stoichiometric amounts of MDA. Calibration curves from supplemented
water samples were produced for each day of analysis. Calibration
samples can be neutralized until 10 h before analysis. The 42
mmol/L 2-TBA was stable for 4 months, but was freshly prepared every
second day. To avoid interfering peaks, after each use the tubes were
immediately placed in a detergent solution and machine-washed and
rinsed in Milli-Q-treated water. The tubes were then soaked in 1%
HNO3, rinsed with Milli-Q water, flushed with 960 mL/L
ethanol, then oven-dried. To maintain optimal separation performance
and avoid buffer precipitation, the column was regenerated with 15 mL
of Milli-Q water followed by 15 mL of methanol at a flow rate of 0.5
mL/min after each day of analysis. The guard column was replaced after
~300 injections.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Baseline separation of MDA-TBA2 and two
unidentified peaks (1 and 3) was achieved with the applied conditions
(see Fig. 1A
). About 15 min was required for each analysis. The
retention times in minutes were 8.52 for peak 1, 9.67 for the
MDA-TBA2 complex 2, and 10.77 for peak 3. No correlation
was found between any of the unidentified peaks and the
MDA-TBA2 peak. The peak in the blank water sample was
present in all blank samples, and the mean area of this peak was
subtracted from the area of the plasma samples.
|
The linearity of the detector response to different concentrations of
the compound was determined at plasma concentrations of 0.59, 1.09,
1.59, 2.09, and 2.59 µmol/L for the MDA-TBA2
complex. The calibration curve for the compound added to water and
plasma was linear over the range investigated when peak areas were
plotted against concentrations and applied to a least-squares
regression equation (see Fig. 2
). The regression coefficient r was 0.999 for the
plasma samples and 0.998 for the water samples.
|
Before analysis of each series, calibration curves from supplemented water samples were prepared for three calibration concentrations covering the expected concentration range. The linear calibration curve was fitted through the data points by linear regression.
The within-day repeatability of the method was evaluated by repeated
analysis (n = 10) of samples of supplemented plasma. Four
concentrations of MDA were investigated: 0.5, 1.0, 1.5, and 2.0
µmol/L. The mean background of MDA-TBA2 response in
blank samples was subtracted from all the supplemented samples. The
average CVs were 1.6% (range 1.3–2.1%). The between-day
reproducibility and accuracy of the method was assessed five
consecutive days at three different concentrations in plasma (0.49,
0.75, and 1.32 µmol/L). Accuracy is calculated as deviation in
percent of the mean estimate from the 5 days of analysis with the
supplemented value. The mean estimates, SDs, and CVs and deviations
from supplemented values are given in Table 1
.
|
The relative recovery for the MDA-TBA2 complex was assessed (n = 10) at four concentrations: 0.5, 1.0, 1.5, and 2.0 µmol/L. The peak area from a supplemented plasma sample was compared, after subtraction of the baseline response, with the peak area from a supplemented water sample. The plasma sample was prepared with equal amounts of the TEP calibrator. The average recovery was 99% (range 96.7–103.4%). The limit of determination for P-MDA on the basis of a signal-to-noise ratio of 10:1 was 0.05 µmol/L. The limit of detection, from a signal-to-noise ratio of 3:1, was 0.02 µmol/L.
Stability of EDTA-treated plasma was investigated in a plasma pool used for control samples. The samples were stable at -80 °C for 6 months. We also compared the response from serum and from plasma collected in Venoject tubes with 0.13 mol/L sodium citrate, sodium heparin, and 0.47 mol/L tripotassium EDTA. The mean response of the MDA-TBA2 complex from 13 fasting volunteers (7 men and 6 women, ages 20–33 years) was 1.16, 0.89, 1.27, and 0.26 µmol/L, respectively. Reproducibility calculated as CV was <10% only with EDTA as anticoagulant (data not shown).
Intraindividual variations were investigated in six healthy volunteers
(3 women and 3 men, ages 21–53 years). Blood samples were obtained in
EDTA-treated Venoject tubes from nonfasting individuals between 0800
and 0830 h for 6 consecutive days. To eliminate analytical
variations, analysis was performed for all individuals in a single run.
Within-subject variations are shown in Table 2
and the day-to-day variability for the group is shown in Table 3
.
|
|
The reference intervals are defined by the 0.025 and 0.975 fractals,
and the 0.90 confidence interval is calculated for each of these
fractals. Additional reference intervals were calculated for each age
decade and for women and men separately (Table 4
).
|
Analysis of variance was used to reveal relations between normalized P-MDA concentration in the three 20-year age groups and gender. The analysis revealed a significant effect from gender (P = 0.033), but apparently not from age (P = 0.109). No major interaction occurred between gender and age (P = 0.103). Men had slightly but significantly higher P-MDA concentrations than women. In the reference sample group, 92 were smokers and 122 were nonsmokers. Smokers had a significantly higher P-MDA concentration (mean 0.66 µmol/L) than nonsmokers (mean 0.60 µmol/L) (P = 0.05). Correlation analysis revealed an association between P-MDA and the number of hours of daily exposure to cigarette smoke (r = 0.162, P = 0.03), but we found no clear correlation between P-MDA and the number of cigarettes smoked (r = -0.065, P = 0.55). P-MDA was significantly correlated with weekly alcohol consumption (r = 0.153, P = 0.03). Weekly alcohol consumption was defined as units consumed during 1 week before completing the questionnaire.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), eluting 1 min
after the MDA-TBA2 complex. No correlation was found
between the areas of background peaks 1 and 3 and the area of the
MDA-TBA2 adduct. The sensitivity of the method is similar
to the one reported by Young and Trimble and by Richard et al.
(17)(18). Further advantages of this method
are that no extraction into organic solvents and tedious evaporation
procedures are required, and the use of an isocratic eluant allows a
higher number of samples to be analyzed per day. The previously reported findings of effects on P-MDA from use of anticoagulant by blood sampling has been somewhat contradictory. Richard et al. (18) found no differences in amount of MDA-TBA2 adduct formed during the sample preparation by addition of EDTA as antioxidant to blood samples collected with heparin as anticoagulant. Our findings of lowest P-MDA concentrations in EDTA compared with heparin-treated plasma are in agreement with the findings of Knight et al. (12). Carbonneau et al. (11) found only minor differences in the concentrations of MDA in serum or plasma with EDTA/heparin as anticoagulants. The consistently lower MDA-TBA2 concentrations observed in EDTA-treated plasma are probably related to EDTA chelation of iron in the TBA assay as well as its weak activity as an antioxidant (10). The difference in findings of effects of EDTA could reflect the importance of having EDTA in the blood sampling tube to ensure immediate reduction in iron-initiated lipoperoxidation generated from the platelets. The higher content of MDA in serum vs plasma could be explained by lipoperoxides being formed during coagulation (10). Thus, MDA measured in EDTA-treated plasma seems to generate the least interfered indication of the degree of lipid peroxidation at the time of blood sampling.
The within-subject variations (CV range 5.9–30%) revealed that P-MDA probably cannot be used as a biomarker or a diagnostic test on an individual basis. However, on a group basis, the small day-to-day variability seems more promising. We therefore suggest that P-MDA be used only as a biomarker for the degree of lipid peroxidation on a group basis.
The reference sample group in the study of Knight et al. (12) consists of blood donors with a carefully screened medical history. Our study deviates from this by its selection criteria, random selection from the general population, instead of healthy subjects generally used as controls in clinical studies. Other studies reporting total P-MDA collected in EDTA sample tubes and measured as a TBA adduct by HPLC do, however, report a similar group mean of 0.6 µmol/L (10)(17). Jiun and Hsien (19) reported a mean of 0.9 µmol/L in a control group consisting of 26 men and 40 women, ages 19–69 years. Carbonneau et al. (11) reported a group mean for healthy controls of 0.43 µmol/L in a group of 30 individuals, ages 23–70 years.
We found a significant correlation between P-MDA and the number of hours of exposure to cigarette smoke, but we found no correlation between P-MDA and the number of cigarettes smoked by the individual (inhalation was not taken into account). These findings, although supporting that P-MDA is a weak biomarker for individual exposure, may also indicate that the recorded number of cigarettes smoked by an individual may be a poor estimate for the actual exposure to the smoke toxins. Smokers are often exposed for longer periods to cigarette smoke from other smokers than are nonsmokers. Also, some smokers do not inhale the smoke from their own cigarettes. These factors may affect the relation between P-MDA and the exposure indicators. On a group basis, however, our findings of a significantly increased P-MDA in smokers is supported by the findings of Kalra et al. (4).
Cigarette smoke is known to increase production of oxygen free radicals by polymorphonuclear leukocytes (20), and to decrease activities of some free radical scavengers (21)(22). Alcohol's ability to induce lipid peroxidation has been related to hypotheses concerning damages caused directly or indirectly by ethanol or the major metabolite acetaldehyde. Current hypotheses include the direct impact of the free radicals derived by ethanol; ethanol's ability to generate formation of oxygen free radical species, which are able to start lipid peroxidation either directly or by exhausting antioxidative defense substances; and acetaldehyde's ability to stimulate lipid peroxidation either directly through free-radical formation or through depletion of the concentration of antioxidative substances (23). The weak correlation seems in accordance with Vendemiale et al. (24): increased plasma concentrations of glutathione and MDA after acute ethanol ingestion in humans (although our parameters are based on total consumption during 1 week before the blood sampling). Adjustment of the correlation analysis for smoking, by "number of daily smoked cigarettes," did not change the outcome.
In future studies, P-MDA may be used as a biomarker of oxidative stress in exposed groups, but smoking and alcohol consumption should be taken into account as possible potential confounders.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. J. Bouwstra, R. M. A. Goselink, P. Dobbelaar, M. Nielen, J. R. Newbold, and T. van Werven The Relationship Between Oxidative Damage and Vitamin E Concentration in Blood, Milk, and Liver Tissue from Vitamin E Supplemented and Nonsupplemented Periparturient Heifers J Dairy Sci, March 1, 2008; 91(3): 977 - 987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fraczek, D. Sanocka, M. Kamieniczna, and M. Kurpisz Proinflammatory Cytokines as an Intermediate Factor Enhancing Lipid Sperm Membrane Peroxidation in In Vitro Conditions J Androl, January 1, 2008; 29(1): 85 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z Yetkin-Ay, B Cadir, E Uskun, F. Bozkurt, N Delibas, F. Gultepe, and I Ergurhan-Ilhan The periodontal status of indirectly lead-exposed apprentices working in autorepair workshops Toxicology and Industrial Health, November 1, 2007; 23(10): 599 - 606. [Abstract] [PDF]
|
|
|
|
|
|
K. R. Babu, H. R. R. Rajmohan, B. K. M. Rajan, and K. K. M Plasma lipid peroxidation and erythrocyte antioxidant enzymes status in workers exposed to cadmium. Toxicology and Industrial Health, September 1, 2006; 22(8): 329 - 335. [Abstract] [PDF]
|
|
|
|
 |
|
 A. A. El Solh, R. Saliba, T. Bosinski, B. J. B. Grant, E. Berbary, and N. Miller Allopurinol improves endothelial function in sleep apnoea: a randomised controlled study. Eur. Respir. J., May 1, 2006; 27(5): 997 - 1002. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Howarth, W. B. Wiehler, M. Pannirselvam, Y. Jiang, J. P. Berger, D. Severson, T. J. Anderson, and C. R. Triggle A Nonthiazolidinedione Peroxisome Proliferator-Activated Receptor {gamma} Agonist Reverses Endothelial Dysfunction in Diabetic (db/db-/-) Mice J. Pharmacol. Exp. Ther., January 1, 2006; 316(1): 364 - 370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Manuel-y-Keenoy, A. Van Campenhout, P. Aerts, J. Vertommen, P. Abrams, L. F. Van Gaal, C. Van Gils, and I. H. De Leeuw Time Course of Oxidative Stress Status in the Postprandial and Postabsorptive States in Type 1 Diabetes Mellitus: Relationship to Glucose and Lipid Changes J. Am. Coll. Nutr., December 1, 2005; 24(6): 474 - 485. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Stuart, G. S. Evans, Y. S. Lin, and H. J. Powers Reduced Collagen and Ascorbic Acid Concentrations and Increased Proteolytic Susceptibility with Prelabor Fetal Membrane Rupture in Women Biol Reprod, January 1, 2005; 72(1): 230 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.G. Rumley, M. Woodward, A. Rumley, J. Rumley, and G.D.O. Lowe Plasma lipid peroxides: relationships to cardiovascular risk factors and prevalent cardiovascular disease QJM, December 1, 2004; 97(12): 809 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Koechlin, A. Couillard, J.P. Cristol, P. Chanez, M. Hayot, D. Le Gallais, and C. Prefaut Does systemic inflammation trigger local exercise-induced oxidative stress in COPD? Eur. Respir. J., April 1, 2004; 23(4): 538 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Del Rio, N. Pellegrini, B. Colombi, M. Bianchi, M. Serafini, F. Torta, M. Tegoni, M. Musci, and F. Brighenti Rapid Fluorimetric Method to Detect Total Plasma Malondialdehyde with Mild Derivatization Conditions Clin. Chem., April 1, 2003; 49(4): 690 - 692. [Full Text] [PDF]
|
|
|
|
|
|
A. Couillard, C. Koechlin, J.P. Cristol, A. Varray, and C. Prefaut Evidence of local exercise-induced systemic oxidative stress in chronic obstructive pulmonary disease patients Eur. Respir. J., November 1, 2002; 20(5): 1123 - 1129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Bolisetty, D Naidoo, K Lui, T H H G Koh, D Watson, R Montgomery, and J Whitehall Postnatal changes in maternal and neonatal plasma antioxidant vitamins and the influence of smoking Arch. Dis. Child. Fetal Neonatal Ed., January 1, 2002; 86(1): F36 - 40. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Totan, O. Cekic, M. Borazan, E. Uz, S. Sogut, and O. Akyol Plasma malondialdehyde and nitric oxide levels in age related macular degeneration Br. J. Ophthalmol., December 1, 2001; 85(12): 1426 - 1428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Doshi, I. F. W. McDowell, S. J. Moat, D. Lang, R. G. Newcombe, M. B. Kredan, M. J. Lewis, and J. Goodfellow Folate Improves Endothelial Function in Coronary Artery Disease : An Effect Mediated by Reduction of Intracellular Superoxide? Arterioscler. Thromb. Vasc. Biol., July 1, 2001; 21(7): 1196 - 1202. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Cavalca, G. Cighetti, F. Bamonti, A. Loaldi, L. Bortone, C. Novembrino, M. De Franceschi, R. Belardinelli, and M. D. Guazzi Oxidative Stress and Homocysteine in Coronary Artery Disease Clin. Chem., May 1, 2001; 47(5): 887 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. ROOB, G. KHOSCHSORUR, A. TIRAN, J. H. HORINA, H. HOLZER, and B. M. WINKLHOFER-ROOB Vitamin E Attenuates Oxidative Stress Induced by Intravenous Iron in Patients on Hemodialysis J. Am. Soc. Nephrol., March 1, 2000; 11(3): 539 - 549. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.A. Livrea, L. Tesoriere, A. Maggio, D. D'Arpa, A.M. Pintaudi, and E. Pedone Oxidative Modification of Low-Density Lipoprotein and Atherogenetic Risk in beta -Thalassemia Blood, November 15, 1998; 92(10): 3936 - 3942. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
