HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: M500192-JLR200v1 46/10/2265    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, B. Articles by Saku, K. Search for Related Content PubMed PubMed Citation Articles by Zhang, B. Articles by Saku, K. Social Bookmarking                      What's this?

Journal of Lipid Research, Vol. 46, 2265-2277, October 2005
Copyright © 2005 by American Society for Biochemistry and Molecular Biology

Relation between insulin resistance and fast-migrating LDL subfraction as characterized by capillary isotachophoresis

Bo Zhang^1,*, Takuya Kaneshi, Takao Ohta and Keijiro Saku^*

^* Department of Cardiology, Fukuoka University School of Medicine, Fukuoka, Japan
Department of Pediatrics, Faculty of Medicine, University of the Ryukyus, Okinawa, Japan

Published, JLR Papers in Press, August 1, 2005. DOI 10.1194/jlr.M500192-JLR200

¹ To whom correspondence should be addressed. e-mail: bozhang{at}fukuoka-u.ac.jp

in re-

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The proportion of the electronegative low density lipoprotein [LDL(–)] subfraction, which is atherogenic, is increased in type 2 diabetes but is not reduced by glycemic control. Therefore, we evaluated the ability of a new technique, capillary isotachophoresis (cITP), to quantify charge-based LDL subfractions and examined the relation between insulin resistance and the cITP fast-migrating (f) LDL levels. Seventy-five 10-year-old boys were included. The two cITP LDL subfractions, fLDL and major LDL subfractions, were proportional to the LDL protein content within the range of 0.1–0.8 mg/ml LDL protein. Levels of cITP fLDL were positively correlated with triglyceride (TG) levels and negatively correlated with LDL size. Insulin resistance as assessed by the homeostasis model assessment (HOMA-IR) was positively correlated (P < 0.01) with cITP fLDL levels (r = 0.41). The relation between HOMA-IR and cITP fLDL levels depended on TG levels but was independent of body mass index and LDL size.

cITP lipoprotein analysis is an accurate and sensitive method for quantifying charge-based LDL subfractions in human plasma, and insulin resistance is related to cITP fLDL independent of LDL size.

Supplementary key words children • plasma lipoprotein subfraction • low density lipoprotein size • electronegative low density lipoprotein

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Electronegative low density lipoprotein [LDL(–)] subfraction in plasma has been shown to have various atherogenic properties [reviewed by Sánchez-Quesada, Benitez, and Ordonez-Llanos (1)]. Although in vitro oxidized LDL can only be taken up by scavenger receptors, LDL(–) can also be taken up by LDL receptors to induce vascular cell adhesion molecule-1 expression through the activation of nuclear factor B and adaptor protein 1 (2).

Both type 1 and type 2 diabetic patients have been shown to have an increased proportion of LDL(–) (3, 4). However, LDL(–) in type 1 and type 2 diabetes seems to be of different origins. In type 1 diabetes, nonenzymatic glycosylation has been shown to contribute to the increased proportion of LDL(–): glycemic optimization decreased both the glycated LDL and the proportion of LDL(–) (3, 4). However, in type 2 diabetes, glycemic control decreased glycated LDL but had no significant effects on the proportion of LDL(–) (4, 5). It is not clear whether or not insulin resistance contributes to LDL(–) generation in type 2 diabetes.

Insulin resistance is known to be associated with increased levels of triglycerides (TGs) (6). Because LDL(–) separated by anion-exchange chromatography techniques has been shown to contain higher TG content than the major LDL subfraction (7–9), it is possible that there may be a relation between insulin resistance and LDL(–). However, this has not yet been examined. In addition, it would be interesting to know whether or not the relation between insulin resistance and LDL(–) depends on TG levels.

Insulin resistance is also linked to plasma levels of small, dense LDLs (pattern B lipoprotein phenotype), which are associated with an increased risk of coronary heart disease (CHD) (6). In normolipidemic (NL) subjects, LDL(–) is distributed predominantly in small, dense LDL subfractions (10). Therefore, it would also be interesting to determine whether or not the relation between insulin resistance and LDL(–) depends on the size of LDL. Clarifying these points should be important considering that glycemic control failed to decrease the proportion of LDL(–) in type 2 diabetic patients (4).

Chromatography has the advantage that the separated LDL(–) fraction can be collected to characterize the composition of LDL(–) (11–13). However, for routine analysis, it has the disadvantage that LDL needs to be separated from other plasma proteins (e.g., by ultracentrifugation) for analysis, because protein absorption is monitored by ultraviolet light detection at 280 nm (12, 13). Therefore, it is time-consuming, and although it determines the proportion of LDL(–), it is not able to determine the absolute amount of LDL(–) in plasma.

Capillary isotachophoresis (cITP) is a new technique for separating plasma lipoprotein subfractions based on their electric charges (14–18). The two cITP LDL subfractions, fast-migrating (f) and slow-migrating (s) LDL, represent the LDL(–) and major LDL subfractions, respectively. We and others (14, 15, 19–22) have previously shown that the absolute amount of lipoprotein subfraction can be determined by reference to an internal marker. In cITP analysis, because lipoproteins are stained by the lipophilic dye 7-nitrobenz-2-oxa-1,3-diazol-4-yl (NBD)-ceramide and monitored by laser-induced fluorescence detection (excitation, 488 nm; emission, 510 nm), lipoproteins can be analyzed without prior separation from other plasma proteins. However, it is not clear whether or not cITP LDL subfractions in plasma are equivalent to those in LDL separated by ultracentrifugation. Therefore, in the present study, we compared cITP LDL subfractions in plasma with those in LDL separated by ultracentrifugation. It is also not clear whether or not fLDL as determined by the method used in cITP analysis is related to LDL(–) as determined by the detection method used in chromatography. Therefore, we examined the linearity of the relation between levels of cITP LDL subfractions and protein contents of LDL to evaluate the ability of cITP lipoprotein analysis to quantify LDL subfractions. In addition, although cITP separates plasma lipoproteins into subfractions based on their electric charge, there is still no direct evidence that cITP fLDL represents an electronegative fraction of LDL. Therefore, we examined the changes in the distribution of cITP LDL subfractions during the in vitro oxidation of LDL to clarify whether or not cITP fLDL is related to the electronegativity of LDL, because oxidation is known to increase the negative charge of LDL.

Many studies have shown that atherosclerosis may start in childhood and is related to blood lipid levels measured in early life (23, 24), and both insulin resistance and lipid levels are related to age and gender. Therefore, to clarify whether or not insulin resistance contributes to the generation of LDL(–), we examined the relation between insulin resistance and cITP fLDL subfraction and the interaction of TG levels and LDL size in 10 year old nondiabetic boys, who have no CHD and have fewer conventional risk factors such as smoking, hypertension, etc.

Since Sánchez-Quesada et al. (10) reported that LDL(–) as measured by ion-exchange chromatography is distributed predominantly in the dense LDL subclass in NL subjects but is increased in the light LDL subclass in patients with hyperlipidemia, it would be interesting to know the relationship between cITP fLDL and LDL subclasses as measured by density. Therefore, we examined the distribution of cITP fLDL in light and dense LDL subclasses in NL, hypercholesterolemic (HC), and hypertriglyceridemic (HTG) subjects.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Subjects
Seventy-five 10-year-old Japanese boys who underwent screening for lifestyle-related diseases in Okinawa, Japan, were included in the study. The study was approved by the Review Board of the University of the Ryukyus and the Ethics Committee of Fukuoka University. Informed consent was obtained from the parents of all of the children. Fasting blood was drawn from the vein. Serum was separated by low-speed centrifugation (1,000 g, 20 min, 4°C). Levels of lipids, lipoproteins, and apolipoproteins were measured in fresh serum. For lipoprotein particle and cITP analysis, aliquots of serum in small volumes were protected with N₂ gas, snap-frozen in liquid nitrogen, and stored at –80°C for <2 months before analysis.

Determination of serum levels of lipids, lipoproteins, and apolipoproteins, LDL size, and insulin resistance
Serum total cholesterol (TC), TG, and high density lipoprotein-cholesterol (HDL-C) levels were measured by enzymatic methods. Low density lipoprotein-cholesterol (LDL-C) was calculated as TC – (HDL-C + TG/5). Serum levels of apolipoprotein A-I (apoA-I), apoA-II, and apoB were determined by the turbidity immunoassay method. LDL size was evaluated by electrophoresis on nondenaturing polyacrylamide gradient gels on precast MULTIGEL-LP (2–15%) according to the procedure specified by the manufacturer (Daiichi Pure Chemicals Co., Ltd., Tokyo, Japan), as described previously (25). Insulin resistance as assessed by homeostasis model assessment (HOMA-IR) was calculated according to the equation HOMA-IR = fasting glucose (mg/dl) x fasting insulin (µU/ml)/405 (26). A quantitative insulin sensitivity check index (QUICKI), which correlates well with insulin sensitivity from the glucose clamp technique, was determined according to the equation QUICKI = 1/[log(fasting glucose) + log(fasting insulin)] (27).

Determination of lipoprotein subfractions by cITP
cITP of lipoproteins in serum, to which was added EDTA-Na₂ at a final concentration of 1 mM before analysis, was performed on a Beckman P/ACE MDQ system (Beckman-Coulter, Inc., Tokyo, Japan) according to the method of Bottcher et al. (14) with some modifications, as described previously (19–22). All of the reagents used for cITP analysis were purchased from Sigma-Aldrich (Tokyo, Japan) unless indicated otherwise. For routine analysis, 6 µl of serum was diluted with 14 µl of leading buffer (LB) consisting of 10 mM HCl (product number 84428; Fluka, Tokyo, Japan) and 18 mM ammediol (2-amino-2-methyl-1,3-propanediol; product number A9074) (LB1, pH 8.8), prestained with 10 µl of 0.1 mg/ml NBD C6-ceramide (product number N1154; Molecular Probes, Inc., Eugene OR), which was prepared by dissolving 1 mg of NBD C6-ceramide in 1 ml of anhydrous methanol (product number 322415) and diluting 10-fold with anhydrous ethylene glycol (product number 324558) for 1 min at room temperature, and mixed with 50 µl of a mixture containing LB with 0.35% hydroxypropylmethylcellulose (product No. H4649) (LB2), spacers, and 5-carboxy-fluorescein (product number C0537) as an internal marker. For experiments to examine the ability of cITP to quantify LDL subfractions, 10 µl of LDL isolated by ultracentrifugation containing 1–8 µg of LDL protein was added to 6 µl of apoB-depleted EDTA plasma (plasma proteins and HDL) and 4 µl of LB1 and used for prestaining. The spacers included were N-(2-acetamido)-2-aminoethanesulfonic acid (product number A7949), D-glucuronic acid (product number 271632), 1-octanesulfonic acid sodium salt (product number O0133), 3-(N-tris[hydroxymethyl]methylamino)-2-hydroxypropanesulfonic acid (product number T0432), N-tris(hydroxymethyl)methyl-3-aminopropanesulfonic acid (product number T9659), L-serine (product number S260-0), L-glutamine (product number G320-2), L-methionine (product number M6039), and glycine (product number G7403). The terminating buffer (TB) contained 24 mM ß-alanine (product number 239720) and 13 mM ammediol (TB1) and was adjusted to pH 10.5 (TB2) by adding 47 µl of saturated barium hydroxide (product number 11783) solution to 2 ml of TB1. Saturated barium hydroxide solution was prepared under an atmosphere of N₂, kept at 30°C in a water bath, and used within 1 week of preparation. A dimethylpolysiloxane-modified fused silica capillary (AT-1^TM; part number 14130) was purchased from Alltech Japan, Inc. (Tokyo, Japan). The sample was injected for 18 s at 20 pounds per square inch into a 30 cm long capillary (inner diameter, 180 µm), and separation was performed at a constant 30 µA for 1 min and 10 kV for 7 min with pressure on both sides (10 pounds per square inch). The separated zones were monitored with argon-laser-induced fluorescence detection (excitation, 488 nm; emission, 520 nm). Each peak was identified, and the peak area in relative fluorescence units was analyzed using 32 Karat Software version 5.0 (Beckman-Coulter, Inc., Tokyo, Japan). The peak area for each cITP lipoprotein subfraction relative to that of the internal marker was presented as the level of cITP lipoprotein subfraction (19–22), unless indicated otherwise.

Preparation of apoB-depleted plasma and isolation of LDL
Plasma containing 1 mM EDTA from a healthy volunteer (female, 32 years old) was used for experiments to examine the ability of cITP to quantify LDL subfractions. Whole plasma was used for cITP analysis and LDL isolation by ultracentrifugation immediately after separation, and some was divided into aliquots in small volumes (150 µl), covered with N₂ gas, and stored at –80°C for 1 day. Plasma depleted of apoB-containing lipoproteins was obtained from frozen whole plasma and used within the day of separation without storage.

ApoB-containing lipoproteins in EDTA plasma were precipitated by the phosphotungstate-Mg²⁺ method, as described previously (19). LDL was isolated from fresh EDTA plasma from the same volunteer by sequential ultracentrifugation. The plasma was first adjusted to a density of 1.019 g/ml with solid KBr and subjected to ultracentrifugation in a TLA-100.3 rotor in a Beckman TL-100 Tabletop Ultracentrifuge for 3 h at 100,000 rpm (541,000 g) and 10°C. The bottom fraction was collected by cutting the tubes, overlaid with liquid KBr (d = 1.019 g/ml), and ultracentrifuged again to remove any contaminating lipoproteins. The density of the bottom fraction was then increased to 1.063 g/ml with solid KBr, and the sample was subjected to ultracentrifugation for 3 h at 100,000 rpm (541,000 g) and 10°C. The top fraction was collected by cutting the tubes and dialyzed in a Slide-A-Lyzer dialysis cassette (Pierce, Rockford, IL) at 4°C against 0.85% NaCl and 0.01% EDTA overnight to remove KBr. LDL was used the day after separation without storage.

Oxidative modification of LDL by incubation with copper
Ultracentrifugally isolated LDL was dialyzed against PBS (pH 7.4) to remove EDTA before oxidation. Oxidation of LDL (0.5 mg/ml) was started by incubation with 10 µM freshly prepared CuSO₄ (with water) at 37°C in a shaking water bath. Aliquots were withdrawn at 0, 0.5, 1, 2, and 3 h of oxidation, and oxidation was stopped by cooling the aliquots on ice and adding EDTA-Na₂ (1 mM) and butylated hydroxytoluene (1 mM). Ten microliters of oxidized LDL was immediately used for cITP separation without removing CuSO₄, because cITP patterns were not affected by CuSO₄ (data not shown).

Agarose gel electrophoresis of LDL was performed as described by Noble (28) using commercial kits (Gel Universal/8; Corning, Chiba, Japan). One microliter of plasma control and 0.5 mg/ml oxidized LDL were applied and subjected to 1% agarose gel electrophoresis for 40 min at 90 V. Lipoproteins were stained with fat red 7B.

Conjugated dienes were determined by measuring absorbance at 234 nm using an ultraviolet-visible recording spectrophotometer (UV-160A; Shimadzu, Kyoto, Japan). Thiobarbituric acid-reactive substances (TBARS) were assayed by the fluorometric method of Yagi (29). ApoB fluorescence was measured at 430 nm (excitation, 360 nm) using a spectrofluorophotometer (RF-5000; Shimadzu).

Separation of light and dense LDL subfractions by heparin-Mg²⁺ precipitation
To examine the distribution of cITP fLDL in light and dense LDL subclasses, blood was drawn from 8 NL, 8 HC (LDL-C 140 mg/dl), and 8 HTG (TG 150 mg/dl) subjects. Serum was frozen with liquid nitrogen immediately after separation and preserved at –80°C under N₂ gas. Serum levels of TC, TG, HDL-C, and LDL-C were measured by enzymatic methods. Six microliters of serum was used to measure lipoprotein subfractions by cITP as described above. The light LDL subclass (d = 1.019–1.044 g/ml) was separated from the dense LDL subclass (d = 1.044–1.063 g/ml) using the heparin-Mg²⁺ precipitation method as described by Hirano, Saegusa, and Yoshino (30), with modifications. Briefly, one part of 300 U/ml heparin sodium (Novo-Heparin Injection 1000; Mochida Pharmaceutical Co. Ltd., Tokyo, Japan) was mixed with one part of 180 mM MgCl₂. One part of the heparin-Mg²⁺ solution was then mixed with one part of serum. After the mixture was put on ice for 25 min, it was centrifuged at 4°C and 15,000 rpm for 15 min in a high-speed microrefrigerated centrifuge (MTX-150; Tomy Seiko Co., Ltd., Fukuoka, Japan). Twelve microliters of the supernatant was subjected to cITP analysis to determine the levels of cITP fLDL and sLDL in the small, dense LDL subclass contained in serum depleted of the large, light LDL subclass. Levels of cITP fLDL and sLDL in the large, light LDL subclass were calculated from those determined in whole serum and serum containing the light LDL subclass. The proportion of cITP fLDL was calculated from levels of cITP fLDL and sLDL.

Statistical analysis
All statistical analyses were performed using the SAS (Statistical Analysis System) software package (version 8.2; SAS Institute) at the Fukuoka University. Correlations between variables were examined by Spearman correlations and regression analysis. Variables among tertiles of HOMA-IR and among NL, HC, HTG subjects were compared by ANOVA and Scheffe's multiple comparison test (31). Differences in variables between large and dense LDL subclasses were examined by ANOVA. All P values are two-tailed. The significance level was considered to be 5% unless indicated otherwise.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Table 1 shows the anthropometric and biochemical characteristics of the 10 year old boys studied. To show the distribution of continuous variables, mean values, median values, and lower and upper quartiles are given. As shown, the mean serum TG level was much higher than the median value, indicating that serum TG levels were not normally distributed.

View larger version (25K): [in this window] [in a new window]   Fig. 1. Lipoprotein profiles as determined by capillary isotachophoresis (cITP) in plasma from a boy with low insulin resistance as assessed by homeostasis model assessment (HOMA-IR; 1.5) (A) and a boy with high HOMA-IR (6.5) (B). Peaks 1–3, fast (f)-, intermediate (i)-, and slow (s)-migrating HDL; peak 4; chylomicron/remnant fraction; peak 5, VLDL/intermediate density lipoprotein (IDL); peaks 6 and 7, fLDL and sLDL; peak 8, a minor LDL fraction.

View larger version (25K): [in this window] [in a new window]   Fig. 2. Electropherograms of plasma lipoproteins (A), HDL fraction obtained by precipitation of apolipoprotein B (apoB)-containing lipoproteins from plasma (B), and ultracentrifugally isolated LDL containing 1 mg (C), 2 mg (D), 4 mg (E), 6 mg (F), and 8 mg (G) of protein in 10 ml of solution that was added to the HDL fraction for cITP separation.

To prove that cITP fLDL is related to the electronegativity of LDL, we tested the hypothesis that the changes in cITP fLDL paralleled the changes in the negative charge of LDL, as indicated by the electrophoretic mobility of LDL in agarose electrophoresis during the in vitro oxidation of LDL. Subjecting LDL to lipid peroxidation by incubation at 37°C in PBS containing 10 µM CuSO₄ resulted in a rapid increase in conjugated dienes and TBARS (data not shown) and a progressive increase in the electrophoretic mobility on agarose gels (Fig. 3A) and apoB fluorescence (data not shown). As shown in Fig. 3A, the increase in the negative charge of oxidized LDL was visible on agarose gels after an oxidation period of 1 h. Changes in cITP LDL subfractions during oxidative modification were monitored by taking aliquots of oxidized LDL at different times and adding them to apoB-depleted EDTA plasma for cITP separation. The electropherograms of oxidized LDLs are shown in Fig. 3B. cITP sLDL (peak 7) gradually decreased from an oxidation time of 0.5 h (Fig. 3Bb–e), whereas cITP fLDL (peak 6) gradually increased from 0.5 to 1.5 h (Fig. 3Bb–d). This result indicates that at the initial stage of oxidation, cITP sLDL subfraction (peak 7) was converted to cITP fLDL subfraction (peak 6), and the increase in cITP fLDL paralleled that in the electrophoretic mobility on agarose gels. From 1.5 to 3 h of oxidation, cITP fLDL (peak 6) decreased, whereas a new cITP LDL subfraction (peak 5), which appeared after 1 h of oxidation (Fig. 3Bc), increased progressively (Fig. 3Bd–f). This result indicates that during the oxidative modification of LDL, although sLDL (peak 7) was converted to fLDL (peak 6), fLDL (peak 6) was further converted to a new, more negatively charged LDL subfraction (peak 5). Therefore, the distribution of cITP LDL subfractions changed gradually during the oxidative modification of LDL, with LDL subfractions shifting toward more negative charges. These results indicate that cITP fLDL represents an in vivo LDL(–) subfraction.

View larger version (22K): [in this window] [in a new window]   Fig. 3. A: Agarose electrophoresis of the time course to the oxidation of LDL (0.5 mg/ml) catalyzed by Cu2+ (10 µM). Lane C represents plasma of the female volunteer who donated blood for the isolation of LDL. B: Electropherograms of ultracentrifugally isolated LDL that was subjected to Cu2+-catalyzed oxidation for 0 h (a), 0.5 h (b), 1 h (c), 1.5 h (d), 2 h (e), and 3 h (f) and added to apoB-depleted plasma (HDL fraction) for cITP separation.

As shown in Table 2, body mass index (BMI), fasting glucose concentration, fasting insulin concentration, and HOMA-IR were significantly higher and QUICKI was significantly lower in the middle and high HOMA-IR tertiles than in the low HOMA-IR tertile, as assessed by ANOVA and Scheffe's multiple comparison test. Also, levels of TG and cITP VLDL/IDL and cITP fLDL fractions were significantly higher, and HDL-C levels, cITP fHDL levels, and LDL size were significantly lower, in the high HOMA-IR tertile than in low HOMA-IR tertile (Table 2). These results indicate that HOMA-IR is related to the altered distribution of not only cITP HDL subfractions but also LDL subfractions.

View larger version (20K): [in this window] [in a new window]   Fig. 4. Correlations between cITP fLDL and triglyceride (A) or LDL size (B) in the low (open circles) and high (closed circles) HOMA-IR groups.

Figure 5 shows typical cITP lipoprotein profiles in whole serum and serum depleted of large, light LDLs by heparin-Mg²⁺ (30) in NL, HC, and HTG subjects. As shown, cITP sLDL (peak 7) in serum depleted of large, light LDLs (Fig. 5B, D, F) was greatly reduced compared with that in whole serum LDL (Fig. 5A, C, E) in NL (Fig. 5A, B), HC (Fig. 5C, D), and HTG (Fig. 5E, F) subjects, whereas cITP fLDL (peak 6) in serum depleted of large, light LDLs was only slightly reduced (Fig. 5). These results indicate that the distribution of cITP LDL subfractions was different between whole serum and light LDL-depleted serum. cITP HDL subfractions (peaks 1–3) were similar in whole serum and serum depleted of large, light LDLs, as expected (Fig. 5). As shown in Fig. 5E, F, the VLDL/IDL subfraction (peak 5) in the HTG subject was also reduced in serum depleted of large, light LDLs compared with that in whole serum, indicating that heparin-Mg²⁺ precipitation also affects TRL. Hirano, Saegusa, and Yoshino (30) did not address the effects of heparin-Mg²⁺ precipitation on TRL because changes in TRL did not affect their measurement of LDL-C levels in whole serum or serum depleted of large, light LDLs. Similarly, because cITP completely separates HDL, TRL, and LDL subfractions, the effects of heparin-Mg²⁺ precipitation on TRL also do not interfere with the measurement of cITP LDL subfractions. Therefore, distributions of cITP fLDL subfractions in different LDL density subclasses can be compared among NL, HC, and HTG groups while ignoring the effects of heparin-Mg²⁺ precipitation on lipoprotein particles other than LDL.

View larger version (50K): [in this window] [in a new window]   Fig. 5. Typical lipoprotein profiles as determined by cITP in whole serum (A, C, E) and serum-depleted of large, light LDLs (d = 1.019–1.044 g/ml) by heparin-Mg2+ precipitation (30) (B, D, F) from a normolipidemic (NL) subject (A, B), a hypercholesterolemic (HC) subject (C, D), and a hypertriglyceridemic (HTG) subject (E, F). Peaks 1–3, HDL subfractions; peak 4; chylomicron/remnant fraction; peak 5, VLDL/IDL; peaks 6 and 7, fLDL and sLDL; peak 8, a minor LDL fraction.

View larger version (30K): [in this window] [in a new window]   Fig. 6. Distributions of levels of cITP fLDL expressed as the peak area relative to that of an internal marker (A) and the proportion of cITP fLDL in total cITP LDL (B) in whole serum (left panel) and in light and dense LDL subclasses separated by heparin-Mg2+ (30) (right panel) in NL (open circles), HC (closed circles), and HTG (closed triangles) subjects. * P < 0.05 versus NL subjects, assessed by ANOVA and Scheffe's multiple comparison test; P < 0.05, dense LDLs versus light LDLs, assessed by ANOVA.

View larger version (47K): [in this window] [in a new window]   Fig. 7. A: Effects of snap-freezing serum with liquid nitrogen and short-term storage of serum at –80°C on cITP lipoprotein subfractions in serum (a, c) and serum depleted of large, light LDLs by heparin-Mg2+ precipitation (b, d). a: Fresh whole serum; b: fresh serum depleted of large, light LDLs; c: frozen, preserved (for 1 day) whole serum; d: frozen serum that was depleted of large, light LDLs after thawing. B: Effects of intermediate-term storage of serum at –80°C on cITP lipoprotein subfractions. a: Fresh serum without being preserved (0 Mo.); b, c, d: serum that had been stored at –80°C under N2 gas for 1, 2, and 3 months, respectively. Peaks 1–3, HDL subfractions; peak 4; chylomicron/remnant fraction; peak 5, VLDL/IDL; peaks 6 and 7, fLDL and sLDL; peak 8, a minor LDL fraction.

View larger version (39K): [in this window] [in a new window]   Fig. 8. A: Effects of long-term storage of serum at –80°C (not protected by N2 gas) on cITP lipoprotein subfractions in a 12-year-old boy. a: Fresh serum without being preserved; b: serum that had been stored at –80°C for 9 months. B: Effects of very-long-term storage of serum at –80°C (not protected by N2 gas) on cITP lipoprotein subfractions in a 12-year-old girl. a: Fresh serum without being preserved; b: serum that had been stored at –80°C for 18 months. Peaks 1–3, HDL subfractions; peak 4; chylomicron/remnant fraction; peak 5, VLDL/IDL; peaks 6 and 7, fLDL and sLDL; peak 8, a minor LDL fraction.

The impact of long-term and very-long-term storage of serum was assessed using serum samples from a 12-year-old boy (Fig. 8A) and a 12 year old girl (Fig. 8B). Fig. 8Aa, b shows cITP lipoprotein profiles in serum from the boy that was not stored and stored at –80°C for 9 months (not protected by N₂ gas), respectively. As shown, the distribution patterns of cITP HDL (peaks 1–3), VLDL/IDL (peak 5), and LDL subfractions (peaks 6–8) were not markedly different in serum samples that were not stored (Fig. 8Aa) and stored for 9 months (Fig. 8Ab). Fig. 8Ba, b shows cITP lipoprotein profiles in serum from the girl that was not stored and stored at –80°C for 18 months (not protected by N₂ gas), respectively. As shown, VLDL/IDL (peak 5) and fLDL subfractions (peak 6) were apparently different in serum samples that were not stored (Fig. 8Ba) and stored for 18 months (Fig. 8Bb). cITP HDL (peaks 1–3) and sLDL subfractions (peak 7) were less affected. These results indicate that 9 months should be the maximum duration of storage of serum samples at –80°C for cITP analysis, if not protected by N₂ gas. Very-long-term storage of serum samples is not recommended for cITP analysis.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The LDL(–) subfraction in plasma includes various forms of modified LDL, and the proportion of LDL(–) to major LDL has been used as a marker for LDL modification (1, 4, 7, 11, 12). Both type 1 and type 2 diabetes have been shown to have increased glycated LDL and a greater proportion of the LDL(–) fraction (3, 4). However, glycemic control reduced LDL(–) in type 1 diabetes but not in type 2 diabetes (3–5). This encouraged us to examine the possibility of an association between insulin resistance and LDL(–).

In the present study, we used the cITP method, originally developed by Böttcher et al. (14) and Schmitz, Mollers, and Richter (15), to separate and quantify the LDL(–) subfraction (cITP fLDL). It is not clear whether or not the proportion of cITP fLDL that is measured directly in plasma is equivalent to that in LDL separated by ultracentrifugation. Therefore, we compared cITP LDL subfractions measured directly in plasma and those measured in ultracentrifugally separated LDL. We found that the proportion of cITP fLDL in ultracentrifugally separated LDL was much lower than that in plasma (Fig. 2). Therefore, separation of LDL by ultracentrifugation may reduce the accuracy of the measurement of the proportion of LDL(–) because LDL(–) is not distributed uniformly in fractions of LDL with different densities.

We examined whether or not the two different methods for monitoring LDL subfractions used in the cITP and chromatography methods are related. We found that levels of cITP fLDL and sLDL, prestained by the lipophilic dye NBD-ceramide, are proportional to the protein content of LDL between 0.1 and 0.8 mg/ml LDL protein. This finding indicates that cITP is an accurate and sensitive method for quantifying charge-based LDL subfractions. It has the advantage that LDL subfractions can be measured directly in plasma. We also examined changes in cITP LDL subfractions during the in vitro oxidation of LDL to prove that fLDL is related to the electronegativity of LDL. We found that cITP sLDL was converted to fLDL during the oxidation of LDL and that the progressive increase in fLDL at the initial stage of oxidation paralleled that in the negative charge of LDL, as indicated by the electrophoretic mobility of LDL on agarose gels. These findings indicate that cITP fLDL is related to the electronegativity of LDL and represents an electronegative subfraction of LDL.

We found that cITP fLDL is strongly and positively correlated with TG levels. This finding supports those of other authors that LDL(–) separated by anion-exchange fast protein liquid chromatography has a higher TG content than LDL(+) (7–9). Therefore, an increased TG level in plasma contributes to the qualitative modification of LDL. However, our finding disagrees with that of Cazzolato, Avogaro, and Bittolo-Bon (32), who separated LDL(–) using anion-exchange high-pressure liquid chromatography and found no significant correlation between the percentage concentration of LDL(–) in total native LDL and TG levels. Considering that the TG level was not correlated with the major LDL subfraction (cITP sLDL; data not shown) in the present study, this discrepancy suggests that the proportion of LDL(–) determined by anion-exchange chromatography may not be sensitive enough to detect changes in LDL(–) related to TG levels.

Our finding that cITP fLDL is negatively correlated with the size of LDL agrees with the finding of Sánchez-Quesada et al. (10) that most of the LDL(–) in NL subjects was contained in dense LDL subclasses. They also reported that half of the LDL(–) from patients with hypertriglyceridemia was contained in dense LDL subclasses, whereas most of the LDL(–) from patients with familial hypercholesterolemia was contained in light LDL subclasses (10). Because LDL(–) from both NL and familial hypercholesterolemia subjects has been shown to have proinflammatory activity on human endothelial cells (7, 11), it could be important to establish the relationship between LDL subfractions as measured by density and charge. Therefore, we examined the distribution of the cITP fLDL subfraction in light and dense LDL subclasses in NL, HC, and HTG subjects. Hirano, Saegusa, and Yoshino (30) recently reported a novel and simple method for the quantification of small, dense LDLs in serum depleted of large, light LDLs (d > 1.044 g/ml) by heparin-Mg²⁺ precipitation. We used their method to separate large, light LDL subclasses from small, dense LDL subclasses. Our findings that HC and HTG subjects had increased levels and a higher proportion of cITP fLDL agree with those of Sánchez-Quesada et al. (10), who reported that the proportion of LDL(–) determined by ion-exchange chromatography was increased in both patients with familial hypercholesterolemia and those with HTG. Our finding that cITP fLDL was distributed more in small, dense LDL subclasses in NL, HC, and HTG subjects is consistent with that of Sánchez-Quesada et al. (10) in NL subjects. Our finding that HC and HTG subjects had increased cITP fLDL in large, light LDL subclasses supports the suggestion of Sánchez-Quesada et al. (10) that hyperlipidemia could promote the formation of light LDL(–). We found that both the proportion and the levels of cITP fLDL were also increased in small, dense LDL subclasses in HC and HTG subjects. This finding disagrees with the finding of Sánchez-Quesada et al. (10), who reported that NL subjects had a higher percentage of LDL(–) in dense LDL fractions (LDL4–6 by gradient centrifugation) than patients with familial hypercholesterolemia and HTG. This discrepancy may be attributable to the different methods used to separate LDL density subclasses and quantify charge-based LDL subfractions. However, our findings and those of Sánchez-Quesada et al. (10) consistently indicate that the quantification of the minimally modified negatively charged LDL subfraction in LDL density subclasses could be clinically important. Further investigations are needed to clarify whether or not the combination of charge-based LDL subfractions and LDL density subclasses, which can be routinely measured with the cITP method and the precipitation method, respectively, could be a novel and potentially useful marker for atherosclerosis.

This is the first study to show that HOMA-IR is associated with increased cITP fLDL levels. The origin of the association between cITP fLDL and insulin resistance is not clear. However, our finding that the relation between HOMA-IR and cITP fLDL depended on TG levels indicates that an increased TG level is a contributing factor to cITP fLDL associated with insulin resistance. Because increased TG levels are preceding factors in the development of insulin resistance in type 2 diabetic patients (33, 34), our finding suggests that an increased TG level may be an important contributor to the risk of CHD in type 2 diabetes.

We found that the relation between insulin resistance and cITP fLDL levels was independent of LDL size. This novel finding suggests that insulin resistance may contribute to the risk of CHD independent of small, dense LDLs, considering that LDL with an increased negative charge is atherogenic (1). However, because the present study did not directly compare cITP fLDL with LDL(–) isolated by anion-exchange HPLC, further investigation will be needed to determine whether or not an association between insulin resistance and cITP fLDL may explain the failure of glycemic control to reduce the proportion of LDL(–) in type 2 diabetes (4). It would also be interesting to determine whether or not antidiabetic drugs such as thiazolidinediones, which are high-affinity peroxisomal proliferator-activated receptor (PPAR) ligands, may reduce cITP fLDL or LDL(–) in type 2 diabetes. However, it is possible that these drugs may at least limit the inflammatory response to LDL(–), because pioglitazone, a synthetic PPAR activator, has been shown to inhibit the expression of vascular cell adhesion molecule-1 on human umbilical vein endothelial cells after interleukin-1ß stimulation (2).

In conclusion, cITP lipoprotein analysis is an accurate and sensitive method for quantifying the charge-based LDL subfractions in human plasma, and insulin resistance is associated with an increased cITP fLDL fraction independent of LDL size. Further studies are needed to clarify whether or not cITP fLDL is useful for controlling the risk factors of CHD in type 2 diabetes.

	ACKNOWLEDGMENTS

 
The authors thank Ms. Yuri Saito for excellent technical assistance. This work was supported by Grants-in-Aid from the Ministry of Education, Science, and Culture of Japan (Grants 12670712, 15790403, and 16590806), by a research grant from the Clinical Research Foundation (2003), by research grants from the Ministry of Health and Welfare, by a grant from the Uehara Memorial Foundation (2002), and by Research Grants 996006 and 026001 from the Central Research Institute of Fukuoka University.

Submitted on September 30, 2004
Revised on May 12, 2005 and in re-revised form June 17, 2005 and in re-re-revised form July 12, 2005

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

1. Sánchez-Quesada, J. L., S. Benitez, and J. Ordonez-Llanos. 2004. Electronegative low-density lipoprotein. Curr. Opin. Lipidol. 15: 329–335.[CrossRef][Medline]

2. Ziouzenkova, O., L. Asatryan, D. Sahady, G. Orasanu, S. Perrey, B. Cutak, T. Hassell, T. E. Akiyama, J. P. Berger, A. Sevanian, et al. 2003. Dual roles for lipolysis and oxidation in peroxisome proliferation-activator receptor responses to electronegative low density lipoprotein. J. Biol. Chem. 278: 39874–39881.[Abstract/Free Full Text]

3. Sánchez-Quesada, J. L., A. Perez, A. Caixas, J. Ordonmez-Llanos, G. Carreras, A. Payes, F. Gonzalez-Sastre, and A. de Leiva. 1996. Electronegative low density lipoprotein subform is increased in patients with short-duration IDDM and is closely related to glycaemic control. Diabetologia. 39: 1469–1476.[CrossRef][Medline]

4. Sánchez-Quesada, J. L., A. Perez, A. Caixas, M. Rigla, A. Payes, S. Benitez, and J. Ordonez-Llanos. 2001. Effect of glycemic optimization on electronegative low-density lipoprotein in diabetes: relation to nonenzymatic glycosylation and oxidative modification. J. Clin. Endocrinol. Metab. 86: 3243–3249.[Abstract/Free Full Text]

5. Gambino, R., B. Uberti, N. Alemanno, E. Pisu, G. Pagano, and M. Cassader. 2004. In vivo oxidizability of LDL in type 2 diabetic patients in good and poor glycemic control. Atherosclerosis. 173: 103–107.[Medline]

6. Reaven, G. M. 2002. Multiple CHD risk factors in type 2 diabetes: beyond hyperglycaemia. Diabetes Obes. Metab. 4 (Suppl. 1): 13–18.

7. De Castellarnau, C., J. L. Sánchez-Quesada, S. Benitez, R. Rosa, L. Caveda, L. Vila, and J. Ordonez-Llanos. 2000. Electronegative LDL from normolipemic subjects induces IL-8 and monocyte chemotactic protein secretion by human endothelial cells. Arterioscler. Thromb. Vasc. Biol. 20: 2281–2287.[Abstract/Free Full Text]

8. Chappey, B., I. Myara, M. O. Benoit, C. Maziere, J. C. Maziere, and N. Moatti. 1995. Characteristics of ten charge-differing subfractions isolated from human native low-density lipoproteins (LDL). No evidence of peroxidative modifications. Biochim. Biophys. Acta. 1259: 261–270.[Medline]

9. Demuth, K., I. Myara, B. Chappey, B. Vedie, M. A. Pech-Amsellem, M. E. Haberland, and N. Moatti. 1996. A cytotoxic electronegative LDL subfraction is present in human plasma. Arterioscler. Thromb. Vasc. Biol. 16: 773–783.[Abstract/Free Full Text]

10. Sánchez-Quesada, J. L., S. Benitez, C. Otal, M. Franco, F. Blanco-Vaca, and J. Ordonez-Llanos. 2002. Density distribution of electronegative LDL in normolipemic and hyperlipemic subjects. J. Lipid Res. 43: 699–705.[Abstract/Free Full Text]

11. Sánchez-Quesada, J. L., M. Camacho, R. Anton, S. Benitez, L. Vila, and J. Ordonez-Llanos. 2003. Electronegative LDL of FH subjects: chemical characterization and induction of chemokine release from human endothelial cells. Atherosclerosis. 166: 261–270.[CrossRef][Medline]

12. Vedie, B., I. Myara, M. A. Pech, J. C. Maziere, C. Maziere, A. Caprani, and N. Moatti. 1991. Fractionation of charge-modified low density lipoproteins by fast protein liquid chromatography. J. Lipid Res. 32: 1359–1369.[Abstract]

13. Hodis, H. N., D. M. Kramsch, P. Avogaro, G. Bittolo-Bon, G. Cazzolato, J. Hwang, H. Peterson, and A. Sevanian. 1994. Biochemical and cytotoxic characteristics of an in vivo circulating oxidized low density lipoprotein (LDL–). J. Lipid Res. 35: 669–677.[Abstract]

14. Böttcher, A., J. Schlosser, F. Kronenberg, H. Dieplinger, G. Knipping, K. J. Lackner, and G. Schmitz. 2000. Preparative free-solution isotachophoresis for separation of human plasma lipoproteins: apolipoprotein and lipid composition of HDL subfractions. J. Lipid Res. 41: 905–915.[Abstract/Free Full Text]

15. Schmitz, G., C. Mollers, and V. Richter. 1997. Analytical capillary isotachophoresis of human serum lipoproteins. Electrophoresis. 18: 1807–1813.[CrossRef][Medline]

16. Schmitz, G., U. Borgmann, and G. Assmann. 1985. Analytical capillary isotachophoresis: a routine technique for the analysis of lipoproteins and lipoprotein subfractions in whole serum. J. Chromatogr. 320: 253–262.[CrossRef][Medline]

17. Schmitz, G., and C. Mollers. 1994. Analysis of lipoproteins with analytical capillary isotachophoresis. Electrophoresis. 15: 31–39.[CrossRef][Medline]

18. Böttcher, A., C. Mollers, K. J. Lackner, and G. Schmitz. 1998. Automated free-solution isotachophoresis: instrumentation and fractionation of human serum proteins. Electrophoresis. 19: 1110–1116.[CrossRef][Medline]

19. Zhang, B., P. Fan, E. Shimoji, H. Xu, K. Takeuchi, C. Bian, and K. Saku. 2004. Inhibition of cholesteryl ester transfer protein activity by JTT-705 increases apolipoprotein E-containing high-density lipoprotein and favorably affects the function and enzyme composition of high-density lipoprotein in rabbits. Arterioscler. Thromb. Vasc. Biol. 24: 1910–1915[Abstract/Free Full Text]

20. Zhang, B., A. Matsunaga, K. Saku, S. Nakano, and T. Yamada. 2004. Associations among plasma lipoprotein subfractions as characterized by analytical capillary isotachophoresis, apolipoprotein E phenotype, Alzheimer disease, and mild cognitive impairment. Arterioscler. Thromb. Vasc. Biol. 24: e144–e146.[Free Full Text]

21. Zhang, B., K. Noda, and K. Saku. 2003. Effect of atorvastatin on total lipid profiles assessed by analytical capillary isotachophoresis. Cardiology. 99: 211–213.[CrossRef][Medline]

22. Zhang, B., H. Tomura, A. Kuwabara, T. Kimura, S. Miura, K. Noda, F. Okajima, and K. Saku. 2005. Correlation of high density lipoprotein (HDL)-associated sphingosine 1-phosphate with serum levels of HDL-cholesterol and apolipoproteins. Atherosclerosis. 178: 199–205.[CrossRef][Medline]

23. Klag, M. J., D. E. Ford, L. A. Mead, J. He, P. K. Whelton, K. Y. Liang, and D. M. Levine. 1993. Serum cholesterol in young men and subsequent cardiovascular disease. N. Engl. J. Med. 328: 313–318.[Abstract/Free Full Text]

24. Berenson, G. S., W. A. Wattigney, R. E. Tracy, W. P. Newman 3rd, S. R. Srinivasan, L. S. Webber, E. R. Dalferes, Jr., and J. P. Strong. 1992. Atherosclerosis of the aorta and coronary arteries and cardiovascular risk factors in persons aged 6 to 30 years and studied at necropsy. The Bogalusa Heart Study. Am. J. Cardiol. 70: 851–858.[CrossRef][Medline]

25. Shimabukuro, T., M. Sunagawa, and T. Ohta. 2004. Low-density lipoprotein particle size and its regulatory factors in school children. J. Clin. Endocrinol. Metab. 89: 2923–2927.[Abstract/Free Full Text]

26. Matthews, D. R., J. P. Hosker, A. S. Rudenski, B. A. Naylor, D. F. Treacher, and R. C. Turner. 1985. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 28: 412–419.[CrossRef][Medline]

27. Katz, A., S. S. Nambi, K. Mather, A. D. Baron, D. A. Follmann, G. Sullivan, and M. J. Quon. 2000. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J. Clin. Endocrinol. Metab. 85: 2402–2410.[Abstract/Free Full Text]

28. Noble, R. P. 1968. Electrophoretic separation of plasma lipoproteins in agarose gel. J. Lipid Res. 9: 693–700.[Abstract]

29. Yagi, K. 1976. A simple fluorometric assay for lipoperoxide in blood plasma. Biochem. Med. 15: 212–216.[CrossRef][Medline]

30. Hirano, T., Y. Ito, H. Saegusa, and G. Yoshino. 2003. A novel and simple method for quantification of small, dense LDL. J. Lipid Res. 44: 2193–2201.[Abstract/Free Full Text]

31. SAS/STAT User's Guide. Version 6, 4th edition, Vol. 2. Cary, NC, SAS Institute, Inc., 1990.

32. Cazzolato, G., P. Avogaro, and G. Bittolo-Bon. 1991. Characterization of a more electronegatively charged LDL subfraction by ion exchange HPLC. Free Radic. Biol. Med. 11: 247–253.[CrossRef][Medline]

33. Sane, T., and M. R. Taskinen. 1993. Does familial hypertriglyceridemia predispose to NIDDM? Diabetes Care. 16: 1494–1501.[Abstract]

34. Mingrone, G., F. L. Henriksen, A. V. Greco, L. N. Krogh, E. Capristo, A. Gastaldelli, M. Castagneto, E. Ferrannini, G. Gasbarrini, and H. Beck-Nielsen. 1999. Triglyceride-induced diabetes associated with familial lipoprotein lipase deficiency. Diabetes. 48: 1258–1263.[Abstract]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				B. Zhang, A. Matsunaga, D. L. Rainwater, S.-i. Miura, K. Noda, H. Nishikawa, Y. Uehara, K. Shirai, M. Ogawa, and K. Saku Effects of rosuvastatin on electronegative LDL as characterized by capillary isotachophoresis: the ROSARY Study J. Lipid Res., September 1, 2009; 50(9): 1832 - 1841. [Abstract] [Full Text] [PDF]

				B. Zhang, Y. Uehara, S. Hida, S.-i. Miura, D. L. Rainwater, M. Segawa, K. Kumagai, K.-A. Rye, and K. Saku Effects of reconstituted HDL on charge-based LDL subfractions as characterized by capillary isotachophoresis J. Lipid Res., May 1, 2007; 48(5): 1175 - 1189. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: M500192-JLR200v1 46/10/2265    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, B. Articles by Saku, K. Search for Related Content PubMed PubMed Citation Articles by Zhang, B. Articles by Saku, K. Social Bookmarking                      What's this?