- Email: [email protected]
Simple and rapid assessment of lipoprotein susceptibility to oxidation in the macromolecule fraction of plasma
Clinica Chimica Acta 316 (2002) 19 – 24 www.elsevier.com/locate/clinchim
Simple and rapid assessment of lipoprotein susceptibility to oxidation in the macromolecule fraction of plasma W. Grady Smith a,*, Charles Reeves a, David Bibbs a, Fred H. Faas b a
Department of Biochemistry and Molecular Biology, McClellan VA Hospital and University of Arkansas for Medical Sciences, Slot 516, 4301 W. Markham, Little Rock, AR 72205, USA b Department of Medicine, McClellan VA Hospital and University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA Received 30 June 2001; received in revised form 8 August 2001; accepted 9 August 2001
Abstract Background: Susceptibility of lipoprotein to oxidation is usually studied using purified lipoproteins. However, for large clinical studies or routine clinical assessment, a rapid, less time-consuming method that minimizes hydroperoxide accumulation in lipoprotein during preparation is desirable. Methods using whole plasma are complicated by the presence of an anticoagulant that may affect oxidation and the presence of antioxidants in the plasma aqueous phase. Lipoproteins in serum are relatively unprotected from initiation of oxidation. Methods: We studied copper-mediated oxidation of a macromolecule fraction of plasma prepared by simple molecular sieve chromatography. Results: LDL was more susceptible to oxidation than plasma macromolecules. The lag times approached zero as the copper concentration increased. The propagation rate was linear for PM vs. the copper concentration while LDL became saturated at 10 – 20 mmol/l. The interassay CVs for the lag phase rate, lag time, PR and maximal DAmax were 23%, 7.7%, 4.9%, and 3.3%, respectively. Conclusions: This procedure should be applicable to large numbers of individuals in investigations regarding the effects of drugs and diets on lipid composition and oxidative susceptibility. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Plasma; Oxidation; Lipoprotein; Copper; Hyperlipidemia; Macromolecules
1. Introduction The oxidative stress hypothesis for the development of atherosclerosis and coronary heart disease has spawned numerous investigations related to
Abbreviations: EDTA, ethylenediaminetetracetate; LDL, lowdensity lipoprotein; VLDL, very low-density lipoprotein; HDL, highdensity lipoprotein; TBARS, thiobarbituric acid-reactive substances. * Corresponding author. Tel.: +1-501-686-5842; fax: +1-501603-1146. E-mail address: [email protected] (W.G. Smith).
pathophysiological mechanisms [1,2]. Because of the relation between atherosclerosis and hyperlipoproteinemia, many laboratory studies have focused on the oxidation of purified lipoprotein, mainly LDL. Hence, an individual’s susceptibility to LDL oxidation is a clinically relevant parameter. Lipoprotein fatty acid composition and antioxidants have been identified as factors affecting lipoprotein oxidation [3 –6]. Because diet and drugs are potential modifiers of fatty acid composition and antioxidant concentration, there is considerable interest in both as modifiers of oxidative susceptibility. Such studies require
0009-8981/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 8 9 8 1 ( 0 1 ) 0 0 7 2 6 - 4
20
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24
clinical trials with human subjects. As noted by others, large scale clinical studies in this area are hampered by the time and labor intensive nature of lipoprotein isolation and the real possibility that lipoproteins may suffer oxidation during isolation [7]. The possibility of doing large clinical trials using plasma to assess susceptibility or to evaluate diet or pharmacological effects on exposure to oxidative stress has prompted studies using plasma or serum directly. Hydroperoxides formed in plasma exposed to copper are mainly found in LDL [8] and lipid-free serum does not display an increase in absorbance at 234 nm when exposed to copper, indicating a lack of conjugated diene formation [9]. These observations strongly suggest that clinical trials using plasma can assess lipoprotein oxidation. Several studies using plasma or serum have been published [3,7,9 – 12]. These studies used whole serum or plasma diluted from 2-fold [11] to 150-fold [3] and copper concentrations ranging from 50 mmol/l [3,9,10] to 3.75 mmol/l [11]. Oxidation measures included absorbance at 234 [3,9,10] or 245 nm [12], and TBARS [11]. Together, these studies establish that oxidation measured in serum or plasma is a consequence of lipoprotein oxidation that correlates well with oxidation of LDL. Serum or plasma oxidation is thus synonymous with lipoprotein oxidation. However, whole plasma or serum contains antioxidants not found in lipoproteins, such as ascorbate and urate, in addition to proteins (albumin, for example), which also act as antioxidants and will influence results. Anticoagulants such as EDTA and citrate can interfere with oxidation measures and must be removed for any meaningful study of plasma oxidation [11,13]. Whole plasma studies have been conducted using heparinized plasma [3,9 – 11]. While whole plasma or serum methods can assess oxidation, their direct relevance to lipoprotein susceptibility is clouded by the water-soluble antioxidants in the aqueous phase. Ascorbate and urate contribute half of the radical trapping activity of plasma [14]. Vitamin C may also protect vitamin E from oxidative destruction [15 – 18]. This complication can be avoided by removal of the water-soluble antioxidants by a simple ‘‘desalting’’ step (molecular sieve chromatography), which also removes the EDTA anticoagulant.
2. Materials and methods 2.1. Preparation of plasma macromolecule fraction An aliquot (0.5 ml) of EDTA-anticoagulated plasma was loaded on a disposable 15 50 mm PD-10 column (9.1-ml volume, Sephadex G25, Sigma-Aldrich) after the column was extensively washed and equilibrated with PBS. The column was eluted with PBS at a flow rate of about 1 ml/ min and 0.5-ml fractions collected to separate macromolecules from low-molecular weight water-soluble components including the EDTA anticoagulant. The great majority of 240-nm-absorbing material ( > 98%) is eluted in the void volume peak. The void volume contains material with a molecular weight >5000. The macromolecule fraction in PBS was designated plasma macromolecules (PM). LDL samples were freed of EDTA anticoagulant by the same procedure. 2.2. Copper-mediated oxidation Oxidation was measured essentially as described by Puhl et al. [19] for lipoproteins. Assays were performed at 37 C using a Shimadzu model UV2401 PC spectrophotometer. We followed oxidation by monitoring absorbance at 240 nm rather than the 234 nm used for lipoproteins because of the greater background absorbance of plasma at 234 nm. The PM fraction was aliquotted into PBS to give a desired initial absorbance at 240 nm. The reaction was initiated by addition of a volume of a concentrated copper solution to give the desired copper concentration and a final volume of 1 ml. The absorbance vs. time readings were recorded at appropriate intervals ranging from 20 s to 2 min for a period of up to 10 h. The files were transferred to a spreadsheet and lag phase rate, lag time, propagation phase rate and maximum absorbance change were calculated. The lag phase rate was the slope of the initial linear phase of absorbance change. The propagation phase rate was the slope of the linear mid region of the propagation phase. The lag time was obtained by extrapolating the linear propagation phase of the DA vs. time plot to the time axis. The maximal change was the difference between initial absorbance and maximum absorbance attained.
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24
21
2.2.1. Copper oxidation difference spectra Absorption spectra were recorded by scanning a sample of PM or LDL from 340 to 220 nm in PBS before addition of copper and again after 4 or 5 h of exposure to copper (specific times and concentrations in figure legends). The pre copper scan was subtracted from the post copper scan to produce the difference spectrum.
3. Results Fig. 1. Oxidation Time Course for LDL and PM: Sigma LDL with EDTA removed by PD-10 desalting was diluted to 44 mg/ ml, which gave an initial absorbance of 0.41 OD units at 234 nm, and a PM sample similarly prepared was diluted to an OD of 0.92 at 240 nm with PBS, and oxidation initiated by addition of concentrated copper sulfate solution to final concentrations of 1 mmol/l for LDL and 20 mmol/l for PM, and OD change recorded at 20-s intervals for 200 min. Controls having no copper are also shown.
A major consideration in developing this method was to decrease time of exposure of the sample to air prior to initiating the oxidation assay. We therefore wanted to avoid any time-consuming steps to standardize regarding lipid, cholesterol or apolipoprotein concentrations as is usually done with purified lipoproteins. Therefore, we related the initial concentration in the assay to the absorbance at 240 nm, the wavelength chosen for following oxidation. An initial absorbance of 0.8 OD units at 240 nm is equivalent to 0.5 mg/ml protein based on bovine albumin standard as determined by a modified Lowry assay [20]. This was roughly equivalent to a 100-fold dilution of plasma. The equivalency with respect to protein was determined by assaying plasma macromolecules from 12 individuals at various dilutions by both the Lowry assay and UV absorbance. Absorbance at 240 nm and Lowry protein concentration were then subjected to linear regression analysis. The correlation coefficient was >0.99 and the standard error of regression was < 3% of the regression coefficient with an intercept of zero. In the routine procedure for determination of PM oxidation parameters 0.5 mg/ml protein was used, and the reaction was initiated by addition of 20 ml of 1 mmol/l CuSO4 to 980 ml of PM solution, giving a final concentration of 20 mmol/l.
Fig. 1 shows the comparison of PM oxidation and LDL oxidation as a function of time. Both show typical lag, propagation and stationary phases. The concentrations were chosen to give approximately the same maximal change in absorbance. PM was monitored at 240 nm and LDL at 234 nm. The greater susceptibility of LDL to oxidation by 1 mmol/l copper as compared to the plasma sample with 20 mmol/l copper was obvious. Fig. 2 shows lag time for LDL oxidation and PM oxidation as functions of copper concentration. Both lag times approached zero as the copper concentration increased. However, PM was clearly more resistant to oxidation by this measure than was LDL; 50 mmol/l of copper obliterated the lag phase of PM as nearly does 20 mmol/l for LDL. The propagation phase rate for LDL and PM as functions of copper concentration are shown in Fig. 3. LDL was clearly saturated at 10 and 20 mmol/l copper
Fig. 2. Lag time as a function of Cu + 2 concentration: lag times were determined for PM from 2 to 50 mmol/l Cu + 2 and from 1 to 20 mmol/l Cu + 2 for LDL. Squares, PM, concentration 0.5 mg protein/ ml; triangles, LDL, concentration 44mg protein/ml.
22
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24 Table 1 Intra-assay and inter-assay coefficients of variation for replicate determinations of oxidation parameters on a single plasma sample ASSAY #
LR
LT
PR
Amax
1 2 3 Inter-assay
11.5 17.2 8.4 23.1
8.0 7.8 9.0 7.7
5.9 5.5 5.3 4.9
1.0 4.1 1.8 3.3
The intra-assay coefficient of variation for each parameter is given across from the assay number. The inter-assay coefficient is given on the last line in each column. Coefficient of variation is the standard deviation divided by the mean expressed as a percent. Fig. 3. Propagation rate for LDL and PM as functions of Cu + 2 concentration: conditions same as for Fig. 2. Squares, PM; triangles, LDL.
whereas PM oxidation rates were much slower and essentially linear with copper concentration in this range. The difference spectra of oxidized plasma macromolecules were very similar to those shown for serum in Ref. [9] (Fig. 4). The peak difference occurred at 234 nm for LDL and at 236 –237 nm for PM. The absorbance change at 240 nm was approximately 105% of that at 234, i.e. slightly greater at 240 nm.
The background absorbance of unoxidized plasma at 240 nm was about half of that at 234 nm. In comparison for purified LDL (from Sigma), the absorbance change at 240 nm was 90% of that observed at 234 nm (Fig. 4). The assay reproducibility is shown in Table 1. Three different assays each with five replicates were conducted on a plasma sample to estimate intra-assay and inter-assay variation. The measure of lag phase rate is highly variable as would be expected from the small absorbance changes. Intra-assay coefficients of variation ranged from 8% to 17% and the inter-assay CV was 23%. Clearly, this was not a useful parameter in assessing oxidation status and is not generally used in such studies even with purified lipoprotein. Considerably less variation was apparent in the other parameters. The lag time intra-assay CVs ranged from 5% to 9% with an inter-assay CV of 7.7%. In view of the linear extrapolation employed to estimate lag time, this degree of variation does not seem surprising. Less variation in PR and DAmax was observed. The intra-assay CVs for PR ranged from 1.8% to 5.4% with an inter-assay CV of 4.9%. Maximal DA intra-assay CVs ranged from 1% to 4% with an interassay CV of 3.3%.
4. Discussion
Fig. 4. Oxidation difference spectra for PM and LDL: LDL concentration was 44 mg/ml and PM concentration was 0.54 mg/ml. Initial absorbance of LDL at 240 nm was 0.41 OD units, and for PM 0.86 OD units. Spectra recorded after 4.5-h exposure to 1 mmol/l Cu + 2 (LDL) or 20 mmol/l Cu + 2 (PM).
The body of literature cited in the introduction establishes that conjugated diene formation in plasma or serum is due to lipoprotein oxidation and is probably predominantly due to LDL. It is reported that mixtures of LDL with different lag times display biphasic propagation phase kinetics, that is, the pro-
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24
file observed is the sum of the two species combined [21]. Therefore, the observation of a monophasic propagation phase profile in PM oxidation suggests that one or very similar species of lipoprotein is/are undergoing oxidation. Since treatment of plasma with copper results in hydroperoxides predominantly in LDL, the kinetic profile observed in plasma is logically due mainly to LDL [8]. The difference spectra for LDL and PM, while very similar, are certainly not identical at the shorter wavelengths. Products of lipid oxidation can interact with other proteins, forming adducts that could alter spectral properties. It must also be noted that the spectra at any time is a quasi steady-state observation. With prolonged incubation or higher copper concentration, the conjugated diene absorption returns to zero as the oxidative process continues. Copper-mediated oxidation of LDL involves saturable binding of copper to initiation sites [21,22]. Other proteins also may bind copper and compete with the saturable lipoprotein site(s) involved in initiation of oxidation. In plasma, much of the copper is conceivably bound to nonlipoprotein proteins and copper available for binding to lipoprotein may be much less than the 1 mmol/l, accounting for the generally longer lag times and slower propagation rates. Twenty micromoles per liter of copper is a suitable concentration for PM oxidation whereas pure LDL requires only 1 mmol/l or less. This situation may account for the linear dependence of plasma oxidation on copper concentration. The CVs for oxidation parameters of PM appeared reasonable for the processes measured. Maximal change which really only depends on two OD numerical values had the smallest coefficient. Propagation rate, which is measured as the slope in the steepest part of the curve, had the next smallest coefficient. Lag time, which depends on linear extrapolation of the linear region of the propagation phase to baseline, was the next smallest but was still in a reasonable range. Lag rate, due to small numerical changes in OD over a restricted time period, had the most variability. This procedure merits further evaluation in studies involving numbers of individuals in investigations regarding relations of oxidative parameters to plasma lipid content and composition. This topic will be explored in a subsequent paper.
23
Acknowledgements This work was supported by American Heart Association, Local Affiliate.
References [1] Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witzum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenecity. N Engl J Med 1989; 320:918 – 24. [2] Steinberg D. Oxidative modification of LDL and atherogenesis: the 1995 Lewis A. Conner memorial lecture. Circulation 1997;95:1062 – 71. [3] Spranger T, Finckh B, Fingerhut R, Kohlschutter A, Beisiegel U, Kontush A. How different constituents of human plasma and low density lipoprotein determine plasma oxidizability by copper. Chem Phys Lipids 1998;91:39 – 52. [4] Croft KD, Williams P, Dimmit S, Abu-Amsha R, Beilin U. Oxidation of low-density lipoproteins: effect of antioxidant content, fatty acid composition and intrinsic phospholipase activity on susceptibility to metal ion-induced oxidation. Biochim Biophys Acta 1995;1254:250 – 6. [5] Reaven P, Parthasarathy S, Grasse B, Miller E, Alamazon F, Mattson F, et al. Feasibility of using an oleate-rich diet to reduce the susceptibility of low density lipoprotein to oxidative modification in humans. Am J Clin Nutr 1991;54:701 – 6. [6] Kontush A, Hubner C, Finckh B, Kohlschutter A, Beisiegel U. Low density lipoprotein oxidizability by copper correlates to its initial ubiquinol-10 and polyunsaturated fatty acid content. FEBS Lett 1994;341:69 – 73. [7] Kontush A, Hubner C, Finckh B, Kohlschutter A, Beisiegel U. How different constituents of low density lipoprotein determine its oxidizability by copper: a correlational approach. Free Radical Res 1996;24:135 – 57. [8] Nourooz-Zadeh J, Tajaddini-Sarmadi J, King KLE, Wolff SP. Low density lipoprotein is the major carrier of lipid hydroperoxides in plasma. Biochem J 1996;313:781 – 6. [9] Regnstrom J, Strom K, Moldeus P, Nilson J. Analysis of lipoprotein diene formation in human serum exposed to copper. Free Radical Res Commun 1993;19:177 – 278. [10] Kontush A, Spranger T, Reich A. Whole plasma oxidation assay as a measure of lipoprotein oxidizability. BioFactors 1997;6:99 – 109. [11] Karten B, Beisiegel U, Gercken G, Kontush A. Mechanisms of lipid peroxidation in human blood plasma: a kinetic approach. Chem Phys Lipids 1997;88:83 – 96. [12] Schnitzer E, Pinchuk I, Fainaru M, Schafer Z., Lichtenberg D. Copper-induced oxidation in unfractionated plasma: the lag preceding oxidation as a measure of oxidation-resistance. Biochem Biophys Res Commun 1995;216:854 – 61. [13] Halliwell B, Gutteridge JMC. Free radicals biology and medicine. 3rd edn. Oxford, UK: Oxford Univ. Press; 1999. p. 58. [14] Wayner DDM, Burton GW, Ingold KU, Barclay LRC, Locke SJ. The relative contributions of vitamin E, urate, ascorbate
24
[15]
[16]
[17]
[18]
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24 and proteins to the total peroxyl radical trapping antioxidant activity of human blood plasma. Biochim Biophys Acta 1987;924:408 – 19. Bowry VW, Ingold KU. The unexpected role of vitamin E (alpha tocopherol) in the peroxidation of human low-density lipoprotein. Acc Chem Res 1999;32:27 – 34. May JM, Qu ZC, Morrow JD. Interaction of ascorbate and alpha tocopherol in resealed human erythrocyte ghosts: transmembrane electron transfer and protection from lipid peroxidation. J Biol Chem 1996;271:10577 – 82. Frei B. Ascorbic acid protects lipids in human plasma and lowdensity lipoprotein against oxidative damage. Am J Clin Nutr 1991;54:1113S – 8S. Pryor WA. Vitamin E and heart disease: basic science to
[19]
[20]
[21]
[22]
clinical intervention. Free Radical Biol Med 2000;13A:141 – 64. Puhl H, Wag G, Esterbauer H. Methods to determine oxidation of low density lipoproteins. Methods Enzymol 1994;233:425 – 41. Hartree EF. Determination of protein: a modification of the Lowry method that gives a linear photometric response. Anal Biochem 1972;48:422 – 7. Thomas MJ, Chen Q, Franklin C, Rudel LL. A comparison of the kinetics of low-density lipoprotein oxidation initiated by copper or by azobis(2-amidinopropane). Free Radical Biol Med 1997;23:927 – 35. Geisig SP, Esterbauer H. Low density lipoprotein is saturable by pro-oxidant copper. FEBS Lett 1994;343:188 – 94.
Simple and rapid assessment of lipoprotein susceptibility to oxidation in the macromolecule fraction of plasma W. Grady Smith a,*, Charles Reeves a, David Bibbs a, Fred H. Faas b a
Department of Biochemistry and Molecular Biology, McClellan VA Hospital and University of Arkansas for Medical Sciences, Slot 516, 4301 W. Markham, Little Rock, AR 72205, USA b Department of Medicine, McClellan VA Hospital and University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA Received 30 June 2001; received in revised form 8 August 2001; accepted 9 August 2001
Abstract Background: Susceptibility of lipoprotein to oxidation is usually studied using purified lipoproteins. However, for large clinical studies or routine clinical assessment, a rapid, less time-consuming method that minimizes hydroperoxide accumulation in lipoprotein during preparation is desirable. Methods using whole plasma are complicated by the presence of an anticoagulant that may affect oxidation and the presence of antioxidants in the plasma aqueous phase. Lipoproteins in serum are relatively unprotected from initiation of oxidation. Methods: We studied copper-mediated oxidation of a macromolecule fraction of plasma prepared by simple molecular sieve chromatography. Results: LDL was more susceptible to oxidation than plasma macromolecules. The lag times approached zero as the copper concentration increased. The propagation rate was linear for PM vs. the copper concentration while LDL became saturated at 10 – 20 mmol/l. The interassay CVs for the lag phase rate, lag time, PR and maximal DAmax were 23%, 7.7%, 4.9%, and 3.3%, respectively. Conclusions: This procedure should be applicable to large numbers of individuals in investigations regarding the effects of drugs and diets on lipid composition and oxidative susceptibility. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Plasma; Oxidation; Lipoprotein; Copper; Hyperlipidemia; Macromolecules
1. Introduction The oxidative stress hypothesis for the development of atherosclerosis and coronary heart disease has spawned numerous investigations related to
Abbreviations: EDTA, ethylenediaminetetracetate; LDL, lowdensity lipoprotein; VLDL, very low-density lipoprotein; HDL, highdensity lipoprotein; TBARS, thiobarbituric acid-reactive substances. * Corresponding author. Tel.: +1-501-686-5842; fax: +1-501603-1146. E-mail address: [email protected] (W.G. Smith).
pathophysiological mechanisms [1,2]. Because of the relation between atherosclerosis and hyperlipoproteinemia, many laboratory studies have focused on the oxidation of purified lipoprotein, mainly LDL. Hence, an individual’s susceptibility to LDL oxidation is a clinically relevant parameter. Lipoprotein fatty acid composition and antioxidants have been identified as factors affecting lipoprotein oxidation [3 –6]. Because diet and drugs are potential modifiers of fatty acid composition and antioxidant concentration, there is considerable interest in both as modifiers of oxidative susceptibility. Such studies require
0009-8981/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 8 9 8 1 ( 0 1 ) 0 0 7 2 6 - 4
20
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24
clinical trials with human subjects. As noted by others, large scale clinical studies in this area are hampered by the time and labor intensive nature of lipoprotein isolation and the real possibility that lipoproteins may suffer oxidation during isolation [7]. The possibility of doing large clinical trials using plasma to assess susceptibility or to evaluate diet or pharmacological effects on exposure to oxidative stress has prompted studies using plasma or serum directly. Hydroperoxides formed in plasma exposed to copper are mainly found in LDL [8] and lipid-free serum does not display an increase in absorbance at 234 nm when exposed to copper, indicating a lack of conjugated diene formation [9]. These observations strongly suggest that clinical trials using plasma can assess lipoprotein oxidation. Several studies using plasma or serum have been published [3,7,9 – 12]. These studies used whole serum or plasma diluted from 2-fold [11] to 150-fold [3] and copper concentrations ranging from 50 mmol/l [3,9,10] to 3.75 mmol/l [11]. Oxidation measures included absorbance at 234 [3,9,10] or 245 nm [12], and TBARS [11]. Together, these studies establish that oxidation measured in serum or plasma is a consequence of lipoprotein oxidation that correlates well with oxidation of LDL. Serum or plasma oxidation is thus synonymous with lipoprotein oxidation. However, whole plasma or serum contains antioxidants not found in lipoproteins, such as ascorbate and urate, in addition to proteins (albumin, for example), which also act as antioxidants and will influence results. Anticoagulants such as EDTA and citrate can interfere with oxidation measures and must be removed for any meaningful study of plasma oxidation [11,13]. Whole plasma studies have been conducted using heparinized plasma [3,9 – 11]. While whole plasma or serum methods can assess oxidation, their direct relevance to lipoprotein susceptibility is clouded by the water-soluble antioxidants in the aqueous phase. Ascorbate and urate contribute half of the radical trapping activity of plasma [14]. Vitamin C may also protect vitamin E from oxidative destruction [15 – 18]. This complication can be avoided by removal of the water-soluble antioxidants by a simple ‘‘desalting’’ step (molecular sieve chromatography), which also removes the EDTA anticoagulant.
2. Materials and methods 2.1. Preparation of plasma macromolecule fraction An aliquot (0.5 ml) of EDTA-anticoagulated plasma was loaded on a disposable 15 50 mm PD-10 column (9.1-ml volume, Sephadex G25, Sigma-Aldrich) after the column was extensively washed and equilibrated with PBS. The column was eluted with PBS at a flow rate of about 1 ml/ min and 0.5-ml fractions collected to separate macromolecules from low-molecular weight water-soluble components including the EDTA anticoagulant. The great majority of 240-nm-absorbing material ( > 98%) is eluted in the void volume peak. The void volume contains material with a molecular weight >5000. The macromolecule fraction in PBS was designated plasma macromolecules (PM). LDL samples were freed of EDTA anticoagulant by the same procedure. 2.2. Copper-mediated oxidation Oxidation was measured essentially as described by Puhl et al. [19] for lipoproteins. Assays were performed at 37 C using a Shimadzu model UV2401 PC spectrophotometer. We followed oxidation by monitoring absorbance at 240 nm rather than the 234 nm used for lipoproteins because of the greater background absorbance of plasma at 234 nm. The PM fraction was aliquotted into PBS to give a desired initial absorbance at 240 nm. The reaction was initiated by addition of a volume of a concentrated copper solution to give the desired copper concentration and a final volume of 1 ml. The absorbance vs. time readings were recorded at appropriate intervals ranging from 20 s to 2 min for a period of up to 10 h. The files were transferred to a spreadsheet and lag phase rate, lag time, propagation phase rate and maximum absorbance change were calculated. The lag phase rate was the slope of the initial linear phase of absorbance change. The propagation phase rate was the slope of the linear mid region of the propagation phase. The lag time was obtained by extrapolating the linear propagation phase of the DA vs. time plot to the time axis. The maximal change was the difference between initial absorbance and maximum absorbance attained.
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24
21
2.2.1. Copper oxidation difference spectra Absorption spectra were recorded by scanning a sample of PM or LDL from 340 to 220 nm in PBS before addition of copper and again after 4 or 5 h of exposure to copper (specific times and concentrations in figure legends). The pre copper scan was subtracted from the post copper scan to produce the difference spectrum.
3. Results Fig. 1. Oxidation Time Course for LDL and PM: Sigma LDL with EDTA removed by PD-10 desalting was diluted to 44 mg/ ml, which gave an initial absorbance of 0.41 OD units at 234 nm, and a PM sample similarly prepared was diluted to an OD of 0.92 at 240 nm with PBS, and oxidation initiated by addition of concentrated copper sulfate solution to final concentrations of 1 mmol/l for LDL and 20 mmol/l for PM, and OD change recorded at 20-s intervals for 200 min. Controls having no copper are also shown.
A major consideration in developing this method was to decrease time of exposure of the sample to air prior to initiating the oxidation assay. We therefore wanted to avoid any time-consuming steps to standardize regarding lipid, cholesterol or apolipoprotein concentrations as is usually done with purified lipoproteins. Therefore, we related the initial concentration in the assay to the absorbance at 240 nm, the wavelength chosen for following oxidation. An initial absorbance of 0.8 OD units at 240 nm is equivalent to 0.5 mg/ml protein based on bovine albumin standard as determined by a modified Lowry assay [20]. This was roughly equivalent to a 100-fold dilution of plasma. The equivalency with respect to protein was determined by assaying plasma macromolecules from 12 individuals at various dilutions by both the Lowry assay and UV absorbance. Absorbance at 240 nm and Lowry protein concentration were then subjected to linear regression analysis. The correlation coefficient was >0.99 and the standard error of regression was < 3% of the regression coefficient with an intercept of zero. In the routine procedure for determination of PM oxidation parameters 0.5 mg/ml protein was used, and the reaction was initiated by addition of 20 ml of 1 mmol/l CuSO4 to 980 ml of PM solution, giving a final concentration of 20 mmol/l.
Fig. 1 shows the comparison of PM oxidation and LDL oxidation as a function of time. Both show typical lag, propagation and stationary phases. The concentrations were chosen to give approximately the same maximal change in absorbance. PM was monitored at 240 nm and LDL at 234 nm. The greater susceptibility of LDL to oxidation by 1 mmol/l copper as compared to the plasma sample with 20 mmol/l copper was obvious. Fig. 2 shows lag time for LDL oxidation and PM oxidation as functions of copper concentration. Both lag times approached zero as the copper concentration increased. However, PM was clearly more resistant to oxidation by this measure than was LDL; 50 mmol/l of copper obliterated the lag phase of PM as nearly does 20 mmol/l for LDL. The propagation phase rate for LDL and PM as functions of copper concentration are shown in Fig. 3. LDL was clearly saturated at 10 and 20 mmol/l copper
Fig. 2. Lag time as a function of Cu + 2 concentration: lag times were determined for PM from 2 to 50 mmol/l Cu + 2 and from 1 to 20 mmol/l Cu + 2 for LDL. Squares, PM, concentration 0.5 mg protein/ ml; triangles, LDL, concentration 44mg protein/ml.
22
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24 Table 1 Intra-assay and inter-assay coefficients of variation for replicate determinations of oxidation parameters on a single plasma sample ASSAY #
LR
LT
PR
Amax
1 2 3 Inter-assay
11.5 17.2 8.4 23.1
8.0 7.8 9.0 7.7
5.9 5.5 5.3 4.9
1.0 4.1 1.8 3.3
The intra-assay coefficient of variation for each parameter is given across from the assay number. The inter-assay coefficient is given on the last line in each column. Coefficient of variation is the standard deviation divided by the mean expressed as a percent. Fig. 3. Propagation rate for LDL and PM as functions of Cu + 2 concentration: conditions same as for Fig. 2. Squares, PM; triangles, LDL.
whereas PM oxidation rates were much slower and essentially linear with copper concentration in this range. The difference spectra of oxidized plasma macromolecules were very similar to those shown for serum in Ref. [9] (Fig. 4). The peak difference occurred at 234 nm for LDL and at 236 –237 nm for PM. The absorbance change at 240 nm was approximately 105% of that at 234, i.e. slightly greater at 240 nm.
The background absorbance of unoxidized plasma at 240 nm was about half of that at 234 nm. In comparison for purified LDL (from Sigma), the absorbance change at 240 nm was 90% of that observed at 234 nm (Fig. 4). The assay reproducibility is shown in Table 1. Three different assays each with five replicates were conducted on a plasma sample to estimate intra-assay and inter-assay variation. The measure of lag phase rate is highly variable as would be expected from the small absorbance changes. Intra-assay coefficients of variation ranged from 8% to 17% and the inter-assay CV was 23%. Clearly, this was not a useful parameter in assessing oxidation status and is not generally used in such studies even with purified lipoprotein. Considerably less variation was apparent in the other parameters. The lag time intra-assay CVs ranged from 5% to 9% with an inter-assay CV of 7.7%. In view of the linear extrapolation employed to estimate lag time, this degree of variation does not seem surprising. Less variation in PR and DAmax was observed. The intra-assay CVs for PR ranged from 1.8% to 5.4% with an inter-assay CV of 4.9%. Maximal DA intra-assay CVs ranged from 1% to 4% with an interassay CV of 3.3%.
4. Discussion
Fig. 4. Oxidation difference spectra for PM and LDL: LDL concentration was 44 mg/ml and PM concentration was 0.54 mg/ml. Initial absorbance of LDL at 240 nm was 0.41 OD units, and for PM 0.86 OD units. Spectra recorded after 4.5-h exposure to 1 mmol/l Cu + 2 (LDL) or 20 mmol/l Cu + 2 (PM).
The body of literature cited in the introduction establishes that conjugated diene formation in plasma or serum is due to lipoprotein oxidation and is probably predominantly due to LDL. It is reported that mixtures of LDL with different lag times display biphasic propagation phase kinetics, that is, the pro-
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24
file observed is the sum of the two species combined [21]. Therefore, the observation of a monophasic propagation phase profile in PM oxidation suggests that one or very similar species of lipoprotein is/are undergoing oxidation. Since treatment of plasma with copper results in hydroperoxides predominantly in LDL, the kinetic profile observed in plasma is logically due mainly to LDL [8]. The difference spectra for LDL and PM, while very similar, are certainly not identical at the shorter wavelengths. Products of lipid oxidation can interact with other proteins, forming adducts that could alter spectral properties. It must also be noted that the spectra at any time is a quasi steady-state observation. With prolonged incubation or higher copper concentration, the conjugated diene absorption returns to zero as the oxidative process continues. Copper-mediated oxidation of LDL involves saturable binding of copper to initiation sites [21,22]. Other proteins also may bind copper and compete with the saturable lipoprotein site(s) involved in initiation of oxidation. In plasma, much of the copper is conceivably bound to nonlipoprotein proteins and copper available for binding to lipoprotein may be much less than the 1 mmol/l, accounting for the generally longer lag times and slower propagation rates. Twenty micromoles per liter of copper is a suitable concentration for PM oxidation whereas pure LDL requires only 1 mmol/l or less. This situation may account for the linear dependence of plasma oxidation on copper concentration. The CVs for oxidation parameters of PM appeared reasonable for the processes measured. Maximal change which really only depends on two OD numerical values had the smallest coefficient. Propagation rate, which is measured as the slope in the steepest part of the curve, had the next smallest coefficient. Lag time, which depends on linear extrapolation of the linear region of the propagation phase to baseline, was the next smallest but was still in a reasonable range. Lag rate, due to small numerical changes in OD over a restricted time period, had the most variability. This procedure merits further evaluation in studies involving numbers of individuals in investigations regarding relations of oxidative parameters to plasma lipid content and composition. This topic will be explored in a subsequent paper.
23
Acknowledgements This work was supported by American Heart Association, Local Affiliate.
References [1] Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witzum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenecity. N Engl J Med 1989; 320:918 – 24. [2] Steinberg D. Oxidative modification of LDL and atherogenesis: the 1995 Lewis A. Conner memorial lecture. Circulation 1997;95:1062 – 71. [3] Spranger T, Finckh B, Fingerhut R, Kohlschutter A, Beisiegel U, Kontush A. How different constituents of human plasma and low density lipoprotein determine plasma oxidizability by copper. Chem Phys Lipids 1998;91:39 – 52. [4] Croft KD, Williams P, Dimmit S, Abu-Amsha R, Beilin U. Oxidation of low-density lipoproteins: effect of antioxidant content, fatty acid composition and intrinsic phospholipase activity on susceptibility to metal ion-induced oxidation. Biochim Biophys Acta 1995;1254:250 – 6. [5] Reaven P, Parthasarathy S, Grasse B, Miller E, Alamazon F, Mattson F, et al. Feasibility of using an oleate-rich diet to reduce the susceptibility of low density lipoprotein to oxidative modification in humans. Am J Clin Nutr 1991;54:701 – 6. [6] Kontush A, Hubner C, Finckh B, Kohlschutter A, Beisiegel U. Low density lipoprotein oxidizability by copper correlates to its initial ubiquinol-10 and polyunsaturated fatty acid content. FEBS Lett 1994;341:69 – 73. [7] Kontush A, Hubner C, Finckh B, Kohlschutter A, Beisiegel U. How different constituents of low density lipoprotein determine its oxidizability by copper: a correlational approach. Free Radical Res 1996;24:135 – 57. [8] Nourooz-Zadeh J, Tajaddini-Sarmadi J, King KLE, Wolff SP. Low density lipoprotein is the major carrier of lipid hydroperoxides in plasma. Biochem J 1996;313:781 – 6. [9] Regnstrom J, Strom K, Moldeus P, Nilson J. Analysis of lipoprotein diene formation in human serum exposed to copper. Free Radical Res Commun 1993;19:177 – 278. [10] Kontush A, Spranger T, Reich A. Whole plasma oxidation assay as a measure of lipoprotein oxidizability. BioFactors 1997;6:99 – 109. [11] Karten B, Beisiegel U, Gercken G, Kontush A. Mechanisms of lipid peroxidation in human blood plasma: a kinetic approach. Chem Phys Lipids 1997;88:83 – 96. [12] Schnitzer E, Pinchuk I, Fainaru M, Schafer Z., Lichtenberg D. Copper-induced oxidation in unfractionated plasma: the lag preceding oxidation as a measure of oxidation-resistance. Biochem Biophys Res Commun 1995;216:854 – 61. [13] Halliwell B, Gutteridge JMC. Free radicals biology and medicine. 3rd edn. Oxford, UK: Oxford Univ. Press; 1999. p. 58. [14] Wayner DDM, Burton GW, Ingold KU, Barclay LRC, Locke SJ. The relative contributions of vitamin E, urate, ascorbate
24
[15]
[16]
[17]
[18]
W.G. Smith et al. / Clinica Chimica Acta 316 (2002) 19–24 and proteins to the total peroxyl radical trapping antioxidant activity of human blood plasma. Biochim Biophys Acta 1987;924:408 – 19. Bowry VW, Ingold KU. The unexpected role of vitamin E (alpha tocopherol) in the peroxidation of human low-density lipoprotein. Acc Chem Res 1999;32:27 – 34. May JM, Qu ZC, Morrow JD. Interaction of ascorbate and alpha tocopherol in resealed human erythrocyte ghosts: transmembrane electron transfer and protection from lipid peroxidation. J Biol Chem 1996;271:10577 – 82. Frei B. Ascorbic acid protects lipids in human plasma and lowdensity lipoprotein against oxidative damage. Am J Clin Nutr 1991;54:1113S – 8S. Pryor WA. Vitamin E and heart disease: basic science to
[19]
[20]
[21]
[22]
clinical intervention. Free Radical Biol Med 2000;13A:141 – 64. Puhl H, Wag G, Esterbauer H. Methods to determine oxidation of low density lipoproteins. Methods Enzymol 1994;233:425 – 41. Hartree EF. Determination of protein: a modification of the Lowry method that gives a linear photometric response. Anal Biochem 1972;48:422 – 7. Thomas MJ, Chen Q, Franklin C, Rudel LL. A comparison of the kinetics of low-density lipoprotein oxidation initiated by copper or by azobis(2-amidinopropane). Free Radical Biol Med 1997;23:927 – 35. Geisig SP, Esterbauer H. Low density lipoprotein is saturable by pro-oxidant copper. FEBS Lett 1994;343:188 – 94.