- Email: [email protected]
Estimation of lipoprotein profile in patients with type II diabetes and its relevance to remnant lipoprotein cholesterol levels
Atherosclerosis 222 (2012) 541–544
Contents lists available at SciVerse ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Short communication
Estimation of lipoprotein profile in patients with type II diabetes and its relevance to remnant lipoprotein cholesterol levels Hiroshi Yoshida a,∗∗ , Yuji Hirowatari b,∗ , Hideo Kurosawa a , Daisuke Manita b , Hidekatsu Yanai c , Kumie Ito d , Norio Tada d a
Department of Laboratory Medicine, Jikei University Kashiwa Hospital, 163-1 Kashiwashita Kashiwa, Chiba 277-8567 Japan Bioscience Division, TOSOH CORPORATION, Japan c Department of Internal Medicine, Kohnodai Hospital, National Center for Global Health and Medicine, Japan d Department of Internal Medicine, Jikei University Kashiwa Hospital, Japan b
a r t i c l e
i n f o
Article history: Received 21 June 2011 Received in revised form 20 March 2012 Accepted 22 March 2012 Available online 30 March 2012 Keywords: Cholesterol Remnants Lipoprotein Coronary artery disease Type 2 diabetes
a b s t r a c t Background: Remnant lipoprotein (RLP), associated with atherosclerosis progression, is often elevated in diabetes mellitus. The RLP level is estimated by immune-separation method and agarose-gel electrophoresis (AGE). Methods: The patients were grouped into three groups according to tertile of RLP-cholesterol (RLP-C) levels. The lipoprotein profiles of type II diabetic patients (T2DM) (n = 194) were measured by an anionexchange liquid chromatography (AEX-HPLC) and an AGE with lipid-staining or cholesterol-staining. Results: IDL- and VLDL-cholesterol by the AEX-HPLC, and VLDL-levels by the AGE with lipid-staining and with cholesterol-staining were significantly different in the three groups. In all the subjects, IDLcholesterol (r = 0.531) and VLDL-cholesterol (r = 0.880) by the AEX-HPLC method were strongly correlated with RLP-C, but only VLDL levels were correlated with RLP-C in AGE, respectively. Conclusion: These results suggest that the AEX-HPLC, which can provide cholesterol levels of not only VLDL but also IDL, is helpful for estimation of lipid profiles in T2DM with high RLP-C. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Remnant lipoprotein (RLP), which increases in abnormality of lipoprotein metabolism, is associated with the progression of atherosclerosis and coronary artery disease (CAD) [1,2]. Cholesterol levels of RLP, which contains chylomicron remnant and VLDL remnant (IDL), can be measured by immunoseparation method as RLP-C [3]. It is known that RLP-C levels are closely associated with CAD risk in patients with normal triglyceride levels [4]. RLP-C may be more sensitive marker for CAD risk than triglyceride level. Diabetes mellitus, which frequently involves the impaired lipoprotein metabolism, is associated with an increased risk of CAD [5,6]. Therefore, it is likely important to measure RLP-C level for the assessment of CAD risk in diabetic patients. A study in Japan
∗ Corresponding author at: Bioscience Division, TOSOH CORPORATION, 2743-1 Hayakawa Ayase-shi, Kanagawa 252-1123, Japan. Tel.: +81 467 76 9911; fax: +81 467 76 9932. ∗∗ Corresponding author. Tel.: +81 471 64 1111; fax: +81 471 64 1126. E-mail addresses: [email protected] (H. Yoshida), [email protected] (Y. Hirowatari). 0021-9150/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2012.03.028
recently showed that high RLP-C levels (>0.12 mmol/L) are a significant risk factor for presence of CAD in type II diabetic patients (T2DM) [7]. IDL (VLDL remnant) can be estimated by using agarose-gel electrophoresis (AGE). In the pattern, the bunched peak which contains LDL, IDL and VLDL is showed with high IDL level. These methods, immunoseparation method and AGE, are used for measurement of RLP levels in clinical laboratories. However, chylomicron remnant and VLDL remnant (IDL) cannot be separated by immunoseparation method, and basically, AGE is a qualitative analysis for IDL. We previously developed an anion-exchange liquid chromatographic method (AEX-HPLC) which can measure cholesterol levels of triglyceride-rich lipoproteins (IDL, VLDL and chylomicron) in addition to HDL and LDL [8]. Previously, lipoprotein profiles in patients of hemodialysis and coronary artery disease were estimated by the AEX-HPLC [9,10]. It was reported that cholesterol levels of IDL and VLDL were high in haemodialysis patients [9]. The RLP level in normolipidemic CAD patients significantly correlated with IDL and VLDL [10]. Here, lipoprotein profiles in Japanese T2DM were assessed by using the new AEX-HPLC method, comparing with conventional AGE with lipid-staining or cholesterol-staining.
542
H. Yoshida et al. / Atherosclerosis 222 (2012) 541–544
Table 1 Basic data and lipoprotein profiles of diabetic patients. Total
Tertiles of RLP-C levels (range, mnol/L)
P value
T1 (0.052–0.106)
T2 (0.109–0.168)
T3 (0.171–0.957)
T1 vs. T2
T1 vs. T3
T2 vs. T3
Basic data n (male/female) Age (years) Body mass index (kg/cm2) Syematic blood pressure (mmHg) Diastolic blood pressure (mmHg) Glycated hemoglobin A1c (%) Fasting blood glucose (mmol/L) RLP-C (mmol/L) Total cholesterol (mmol/L) Triglyceride (mmol/L)
194 (117/77) 63 ± 13 23.1 ± 2.0 132.0 ± 9.5 78.3 ± 7.0 6.3 ± 0.95 7.60 ± 2.89 0.17 ± 0.13 5.10 ± 1.01 1.58 ± 0.98
65 (27/38) 63 ± 14 22.7 ± 1.7 130.1 ± 5.3 76.9 ± 5.2 6.1 ± 0.87 6.99 ± 1.72 0.080 ± 0.016 4.76 ± 0.85 0.95 ± 0.34
64 (25/39) 67 ± 11 23.4 ± 1.8 129.5 ± 9.2 77.2 ± 7.9 6.4 ± 0.98 8.10 ± 3.05 0.13 ± 0.019 5.04 ± 0.91 1.52 ± 0.38
65 (25/40) 61 ± 13 23.1 ± 2.4 136.3 ± 11.5 80.8 ± 6.8 6.4 ± 0.99 7.77 ± 3.61 0.30 ± 0.15 5.77 ± 1.06 2.76 ± 1.12
NS NS <0.05 NS NS NS NS <0.0001 NS <0.0001
NS NS NS <0.0001 <0.001 NS NS <0.0001 <0.0001 <0.0001
NS <0.05 NS <0.0001 <0.01 NS NS <0.0001 <0.0005 <0.0001
Medication use, n (%) Insulin Statin
51 (26) 51 (26)
14 (22) 22 (34)
20 (31) 12 (19)
17 (26) 17 (26)
NS NS
NS NS
NS NS
Lipoprotein profiles Anion-exchange liquid chromatography n (male/female) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) IDL cholesterol (mmol/L) VLDL cholesterol (mmol/L) Chylomicron cholesterol (mmol/L)
194 (117/77) 1.29 ± 0.43 3.16 ± 0.89 0.24 ± 0.12 0.52 ± 0.42 0.11 ± 0.080
65 (27/38) 1.46 ± 0.51 2.90 ± 0.58 0.17 ± 0.070 0.22 ± 0.11 0.12 ± 0.088
64 (25/39) 1.31 ± 0.36 3.10 ± 0.91 0.23 ± 0.078 0.42 ± 0.15 0.10 ± 0.091
65 (25/40) 1.09 ± 0.34 3.47 ± 0.91 0.32 ± 0.14 0.92 ± 0.47 0.11 ± 0.057
NS <0.05 NS <0.0001 <0.0001 NS
NS <0.0001 <0.001 <0.0001 <0.0001 NS
NS <0.001 <0.05 <0.0005 <0.0001 <0.005
Agarose-gel electrophoresis with lipid-staining n (male/female) 186 (112/74) 36 ± 6.7 HDL ratio (%) LDL ratio (%) 41 ± 6.9 21 ± 7.9 VLDL ratio (%)
63 (25/38) 41 ± 5.6 42 ± 7.7 14 ± 5.6
63 (24/39) 36 ± 4.7 41 ± 6.4 21 ± 5.8
60 (23/37) 30 ± 4.7 41 ± 6.4 27 ± 6.9
NS <0.0001 NS <0.0001
NS <0.0001 NS <0.0001
NS <0.0001 NS <0.0001
Agarose-gel electrophoresis with cholesterol-staining 167 (102/65) n (male/female) 1.42 ± 0.44 HDL cholesterol (mmol/L) 3.16 ± 0.91 LDL cholesterol (mmol/L) VLDL cholesterol (mmol/L) 0.59 ± 0.31
45 (18/27) 1.63 ± 0.49 2.74 ± 0.54 0.41 ± 0.16
60 (21/39) 1.47 ± 0.39 3.05 ± 0.88 0.57 ± 0.16
62 (25/37) 1.27 ± 0.36 3.54 ± 1.03 0.91 ± 0.36
NS NS NS <0.0001
NS <0.0001 <0.0001 <0.0001
NS <0.01 <0.005 <0.0001
Data are means ± SD. Not significant is indicated by NS.
2. Materials and methods 2.1. Study subjects The 194 Japanese T2DM were recruited in Jikei University Kashiwa Hospital. This study was approved by the ethics committee at the Jikei University, and written informed consent was obtained from all patients. The diagnosis of diabetes was defined as currently being in anti-diabetic drug therapy or the clinical criteria of Japanese Diabetes Society; fasting blood glucose above 7.0 mmol/L or glycated hemoglobin A1c above 6.1%. HbA1c levels are expressed according to the procedures outlined by the Laboratory Test Committee of the Japan Diabetes Society.
2.2. Measurement of lipoproteins The venous blood was taken in all patients after a 12-hour overnight fast. Serum was stored at 4 ◦ C and was used for assays within 3 days after sampling. Total cholesterol and triglyceride levels were determined by enzymatic reagent, Cholestest-CHO and Cholestest-TG, Sekisui Medical Corp., Japan, respectively. AGE was performed using Electrophoresis Processing Analyzer Epalyzer2 (Herena Laboratories, Japan). Fat-Red 7B agarose (Herena Laboratories, Japan), and QuickGel LIPO and enzymatic cholesterol reagent CholeTriComb CH (Herena Laboratories, Japan) were used for the methods for lipid-staining and cholesterol-staining, respectively. The AEX-HPLC reported previously [8] was partially modified, and used for measuring cholesterol levels in HDL, LDL, IDL, VLDL,
and chylomicrons. A column, which contained 2.5 m of nonporous polymer-based gal with diethylaminoethyl ligands, and 3.0 mm ID × 15 mm in size, and a post-column reactor, which contained an enzymatic cholesterol reagent (TCHO-CL, Serotec Corp., Japan), were used. It took 10 min to complete the assay of one sample. The coefficient data of variation for lipoprotein classes in between-day assay (n = 10) and within-day assay (n = 10) using hyperlipidemic serum samples were 0.57–2.93% and 0.85–5.60%, respectively. 2.3. RLP-C analysis RLP fraction in serum was obtained using a mixed immunoaffinity gel containing monoclonal anti-human apoA-I (H-12) and anti-human apoB-100 (JI-H) antibodies (Japan Immunoresearch Labolatories, Japan) [2]. 2.4. Statistical analysis The data were presented as mean ± standard deviation (SD). Between-group differences were evaluated by F-test, and the comparisons were analyzed by using unpaired t-test for normal distribution data or Mann–Whitney U-test for not normal distribution data. P values <0.05 were considered significant. Correlations were estimated using Pearson product-moment correlation test. 3. Results T2DM patients (n = 194) were enrolled in this study, and were divided into three groups according to tertile of RLP-C levels.
H. Yoshida et al. / Atherosclerosis 222 (2012) 541–544
B
0.8
VLDL cholesterol by AEX-HPLC (mmol / L)
IDL cholesterol by AEX-HPLC (mmol / L)
A
0.6 r = 0.531 p < 0.0001 n = 194
0.4
0.2
0.0
2.5 2.0 1.5 1.0 r = 0.880 p < 0.0001 n = 194
0.5 0.0
0.0
0.4
0.8
1.2
0.0
40
D
30 20 r = 0.591 p < 0.0001 n = 186
10 0 0.0
0.4
0.8
0.4
0.8
1.2
RLP-C (mmol / L) VLDL cholesterol by AGE with cholesterol-staining (mmol / L)
50
VLDL ratio by AGE with lipid-staining(%)
RLP-C (mmol / L)
C
543
1.2
2.5 2.0 1.5 1.0 r = 0.765 p < 0.0001 n = 167
0.5 0.0 0.0
RLP-C (mmol / L)
0.4 0.8 RLP-C (mmol / L)
1.2
Fig. 1. Correlation between lipoprotein profiles and RLP-C. A, B, C, and D show the correlations of IDL and VLDL cholesterol by AEX-HPLC, VLDL ratio by AGE with lipid-staining and VLDL cholesterol by AGE with cholesterol-staining, respectively. The data of T1, T2, and T3 are represented by closed circles, open circles, and triangles, respectively.
HDL-cholesterol by AEX-HPLC decreased in order of lowest (T1), middle (T2), and highest (T3) tertile (Table 1). LDL-, IDL-, and VLDL-cholesterol by AEX-HPLC increased in order of T1, T2, and T3 (Table 1). HDL-, IDL-, and VLDL-cholesterol were significantly different between each other in the three groups. LDL and VLDL peak in 2 sera in T1, 1 serum in T2 and 5 sera in T3 could not be separated by AGE with lipid-staining. Therefore the data of 8 samples were excluded from the analysis. HDL ratio obtained by AGE with lipid-staining decreased in order of T1, T2, and T3 (Table 1). VLDL ratio obtained by AGE with lipid-staining increased in order of T1, T2, and T3 (Table 1). LDL and VLDL peak in 20 sera in T1, 4 sera in T2, and 3 sera in T3 could not be separated by AGE with cholesterol-staining. Therefore the data of 27 samples were excluded from the analysis. HDLcholesterol obtained by AGE with cholesterol-staining decreased in order of T1, T2, and T3 (Table 1). LDL- and VLDL-cholesterol obtained by AGE with cholesterol-staining increased in order of T1, T2, and T3 (Table 1). LDL- and chylomicron-cholesterol by AEX-HPLC and LDL ratio by AGE with lipid-staining did not correlated significantly with RLP-C, and the others correlated significantly, as follows: HDL-cholesterol by AEX-HPLC, r = −0.348; IDL-cholesterol by AEX-HPLC, r = 0.531; VLDL-cholesterol by AEX-HPLC, r = 0.880; HDL ratio by AGE with lipid-staining, r = −0.622; VLDL ratio by AGE with lipid-staining, r = 0.591; HDL-cholesterol by AGE with cholesterol-staining, r = −0.299; VLDL-cholesterol by AGE with cholesterol-staining, r = 0.765; LDL-cholesterol by AGE with cholesterol-staining, r = 0.265. Therefore, among these data, IDLand VLDL-cholesterol by AEX-HPLC, VLDL ratio by AGE with lipidstaining and VLDL-cholesterol by AGE with cholesterol-staining showed highly positive correlations with RLP-C (Fig. 1). IDL-cholesterol levels obtained by the AEX-HPLC were compared between the samples with and without a bunch of peak
contained LDL and VLDL in AGE. IDL-cholesterol measured by the AEX-HPLC in the lipid-staining 8 samples was not significantly higher than in the other samples; 0.36 ± 0.23 mmol/L and 0.23 ± 0.11 mmol/L, respectively. Also, IDL-cholesterol in the cholesterol-staining 27 samples was similar to the others; 0.22 ± 0.16 mmol/L and 0.24 ± 0.11 mmol/L, respectively. 4. Discussion Our developed AEX-HPLC method can measure cholesterol levels in lipoprotein classes, HDL, LDL, IDL, VLDL and chylomicron [8]. Remnant lipoproteins are triglyceride rich lipoproteins, including chylomicron remnant and VLDL remnant (IDL) [1,2]. The major component in RLP was VLDL remnant (IDL) because the fasting samples were measured in this study. Fig. 1 showed that IDL- and VLDL-cholesterol were significantly correlated with RLPC levels, but chylomicron-cholesterol was not. This chylomicron fraction in the AEX-HPLC don’t only contain chylomicron-related lipoprotein but also lipoprotein (a) [11]. RLP-C was reported as a risk maker for CAD, described as above [1–7]. However, the lack of standardization of RLP-C assay may provide a less quality of the measured data. A previous report showed that the values of two RLP-C kits are different in a part of patients [12]. The sequential ultracentrifugation method is a reference method to analyze lipoprotein, which is inconvenient and time-consuming [13]. The AEX-HPLC used in this study conveniently can measure IDL-cholesterol in addition to HDL-, LDL-, and VLDL-cholesterol. It is conceivable that IDL-cholesterol value may be important information to provide insight on the impaired metabolism of remnant lipoproteins when the measured values of two RLP-C kits are different. IDL-cholesterol levels of the AEX-HPLC in the samples with or without a bunch peak contained LDL and VLDL in AGEs were
544
H. Yoshida et al. / Atherosclerosis 222 (2012) 541–544
similar. We thought that lipoproteins in the samples with the one bunch peak contained LDL and VLDL might be highly heterogenous, and were not always available for IDL level determination. In the AGE analysis with lipid or cholesterol staining, VLDL levels only correlated with RLP-C, but VLDL itself is not RLP. However, AEX-HPLC can provide data of IDL-cholesterol correlated with RLP-C, suggesting the clinical usefulness of the AEX-HPLC because IDL is one of RLPs. IDL-cholesterol is high level in type III hyperlipoproteinemia, and the IDL levels obtained by an ultracentrifugation method in Japanese patients were reported 1.1 ± 0.5 mmol/L [14]. We showed that the IDL-cholesterol values obtained by the AEX-HPLC and the ultracentrifugation method were nearly equal [8]. IDL-cholesterol levels of T2DM in this study are likely to be lower than in the previous study [14]. In conclusion, this study indicates that IDL and VLDL levels measured by the AEX-HPLC were significantly different between T2DM patients by RLP-C levels, and strongly correlated with RLPC. IDL and VLDL-cholesterol, which are difficult to be conveniently estimated by conventional methods, in addition to HDL and LDLcholesterol can be measured at a time by the AEX-HPLC [8]. These results suggest that the AEX-HPLC which can provide IDL- and VLDL-cholesterol is more helpful than the AGE which can only provide VLDL level for estimation of lipid profiles in T2DM with high RLP-C. Sources of funding Research funds were provided in part by the Jikei University Research Fund (to H. Yoshida) and also in part by Grant-in-Aid for Scientific Research (23234567) from Japan Ministry of Health, Labor and Welfare (to H. Yoshida). Conflict of interests There were no conflicts of interest in this study.
References [1] Havel JR. Determination and clinical significance of triglyceride-rich lipoprotein remnants. In: Rifai N, Warnick GR, Dominiczak MH, editors. Handbook of lipoprotein testing. 2nd ed. Washington, D.C.: The American Association for Clinical Chemistry, Inc. Press; 2000. p. 565–80. [2] Fujioka Y, Ishikawa Y. Remnant lipoproteins as strong key particles to atherogenesis. Journal of Atherosclerosis and Thrombosis 2009;16:145–54. [3] Nakajima K, Saito T, Tamura A, et al. Cholesterol in remnant-like lipoproteins in human serum using monoclonal anti apo B-100 and anti apo A-I immunoaffinity mixed gels. Clinica Chimica Acta 1993;223:53–71. [4] Masuoka H, Kamei S, Ozaki M, et al. Predictive value of remnant-like particle cholesterol as an indicator of coronary artery stenosis in patients with normal serum triglyceride levels. Internal Medicine 2000;39:540–6. [5] Goff Jr DC, D’Agostino Jr RB, Haffner SM, Saad MF, Wagenknecht LE. Lipoprotein concentrations and carotid atherosclerosis by diabetes status: results from the insulin resistance atherosclerosis study. Diabetes Care 2000;23:1006–11. [6] Milicevic Z, Raz I, Beattie SD, et al. Natural history of cardiovascular disease in patients with diabetes: role of hyperglycemia. Diabetes Care 2008;31:S155–60. [7] Fukushima H, Sugiyama S, Honda O, et al. Prognostic value of remnantlike lipoprotein particle levels in patients with coronary artery disease and type II diabetes mellitus. Journal of the American College of Cardiology 2004;43:2219–24. [8] Hirowatari Y, Yoshida H, Kurosawa H, Doumitu K, Tada N. Measurement of cholesterol of major serum lipoprotein classes by anion-exchange HPLC with perchlorate ion-containing eluent. Journal of Lipid Research 2003;44:1404–12. [9] Hirowatari Y, Yoshida H, Fueki Y, et al. Measurement of cholesterol concentrations of major serum lipoprotein classes in haemodialysis patients by anionexchange chromatography. Annals of Clinical Biochemistry 2008;45:571–4. [10] Nakada Y, Kurosawa H, Tohyama J, Inoue Y, Ikewaki K. Increased remnant lipoprotein in patients with coronary artery disease-evaluation utilizing a newly developed remnant assay, remnant lipoproteins cholesterol homogenous assay (RemL-C). Journal of Atherosclerosis and Thrombosis 2007;14:56–64. [11] Hirowatari Y, Yoshida H, Kurosawa H, et al. Analysis of cholesterol levels in lipoprotein(a) with anion-exchange chromatography. Journal of Lipid Research 2010;51:1237–43. [12] Yoshida H, Kurosawa H, Hirowatari Y, et al. Characteristic comparison of triglyceride-rich remnant lipoprotein measurement between a new homogenous assay (RemL-C) and a conventional immunoseparation method (RLP-C). Lipids Health Disease 2008;7:18–23. [13] Schumaker VN, Puppione DL. Sequential flotation ultracentrifugation. Methods in Enzymology 1986;128:155–70. [14] Todo Y, Kobayashi J, Higashikata T, et al. Detailed analysis of serum lipids and lipoproteins from Japanese type III hyperlipoproteinemia with apolipoprotein E2/2 phenotype. Clinica Chimica Acta 2004;348:35–40.
Contents lists available at SciVerse ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Short communication
Estimation of lipoprotein profile in patients with type II diabetes and its relevance to remnant lipoprotein cholesterol levels Hiroshi Yoshida a,∗∗ , Yuji Hirowatari b,∗ , Hideo Kurosawa a , Daisuke Manita b , Hidekatsu Yanai c , Kumie Ito d , Norio Tada d a
Department of Laboratory Medicine, Jikei University Kashiwa Hospital, 163-1 Kashiwashita Kashiwa, Chiba 277-8567 Japan Bioscience Division, TOSOH CORPORATION, Japan c Department of Internal Medicine, Kohnodai Hospital, National Center for Global Health and Medicine, Japan d Department of Internal Medicine, Jikei University Kashiwa Hospital, Japan b
a r t i c l e
i n f o
Article history: Received 21 June 2011 Received in revised form 20 March 2012 Accepted 22 March 2012 Available online 30 March 2012 Keywords: Cholesterol Remnants Lipoprotein Coronary artery disease Type 2 diabetes
a b s t r a c t Background: Remnant lipoprotein (RLP), associated with atherosclerosis progression, is often elevated in diabetes mellitus. The RLP level is estimated by immune-separation method and agarose-gel electrophoresis (AGE). Methods: The patients were grouped into three groups according to tertile of RLP-cholesterol (RLP-C) levels. The lipoprotein profiles of type II diabetic patients (T2DM) (n = 194) were measured by an anionexchange liquid chromatography (AEX-HPLC) and an AGE with lipid-staining or cholesterol-staining. Results: IDL- and VLDL-cholesterol by the AEX-HPLC, and VLDL-levels by the AGE with lipid-staining and with cholesterol-staining were significantly different in the three groups. In all the subjects, IDLcholesterol (r = 0.531) and VLDL-cholesterol (r = 0.880) by the AEX-HPLC method were strongly correlated with RLP-C, but only VLDL levels were correlated with RLP-C in AGE, respectively. Conclusion: These results suggest that the AEX-HPLC, which can provide cholesterol levels of not only VLDL but also IDL, is helpful for estimation of lipid profiles in T2DM with high RLP-C. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Remnant lipoprotein (RLP), which increases in abnormality of lipoprotein metabolism, is associated with the progression of atherosclerosis and coronary artery disease (CAD) [1,2]. Cholesterol levels of RLP, which contains chylomicron remnant and VLDL remnant (IDL), can be measured by immunoseparation method as RLP-C [3]. It is known that RLP-C levels are closely associated with CAD risk in patients with normal triglyceride levels [4]. RLP-C may be more sensitive marker for CAD risk than triglyceride level. Diabetes mellitus, which frequently involves the impaired lipoprotein metabolism, is associated with an increased risk of CAD [5,6]. Therefore, it is likely important to measure RLP-C level for the assessment of CAD risk in diabetic patients. A study in Japan
∗ Corresponding author at: Bioscience Division, TOSOH CORPORATION, 2743-1 Hayakawa Ayase-shi, Kanagawa 252-1123, Japan. Tel.: +81 467 76 9911; fax: +81 467 76 9932. ∗∗ Corresponding author. Tel.: +81 471 64 1111; fax: +81 471 64 1126. E-mail addresses: [email protected] (H. Yoshida), [email protected] (Y. Hirowatari). 0021-9150/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2012.03.028
recently showed that high RLP-C levels (>0.12 mmol/L) are a significant risk factor for presence of CAD in type II diabetic patients (T2DM) [7]. IDL (VLDL remnant) can be estimated by using agarose-gel electrophoresis (AGE). In the pattern, the bunched peak which contains LDL, IDL and VLDL is showed with high IDL level. These methods, immunoseparation method and AGE, are used for measurement of RLP levels in clinical laboratories. However, chylomicron remnant and VLDL remnant (IDL) cannot be separated by immunoseparation method, and basically, AGE is a qualitative analysis for IDL. We previously developed an anion-exchange liquid chromatographic method (AEX-HPLC) which can measure cholesterol levels of triglyceride-rich lipoproteins (IDL, VLDL and chylomicron) in addition to HDL and LDL [8]. Previously, lipoprotein profiles in patients of hemodialysis and coronary artery disease were estimated by the AEX-HPLC [9,10]. It was reported that cholesterol levels of IDL and VLDL were high in haemodialysis patients [9]. The RLP level in normolipidemic CAD patients significantly correlated with IDL and VLDL [10]. Here, lipoprotein profiles in Japanese T2DM were assessed by using the new AEX-HPLC method, comparing with conventional AGE with lipid-staining or cholesterol-staining.
542
H. Yoshida et al. / Atherosclerosis 222 (2012) 541–544
Table 1 Basic data and lipoprotein profiles of diabetic patients. Total
Tertiles of RLP-C levels (range, mnol/L)
P value
T1 (0.052–0.106)
T2 (0.109–0.168)
T3 (0.171–0.957)
T1 vs. T2
T1 vs. T3
T2 vs. T3
Basic data n (male/female) Age (years) Body mass index (kg/cm2) Syematic blood pressure (mmHg) Diastolic blood pressure (mmHg) Glycated hemoglobin A1c (%) Fasting blood glucose (mmol/L) RLP-C (mmol/L) Total cholesterol (mmol/L) Triglyceride (mmol/L)
194 (117/77) 63 ± 13 23.1 ± 2.0 132.0 ± 9.5 78.3 ± 7.0 6.3 ± 0.95 7.60 ± 2.89 0.17 ± 0.13 5.10 ± 1.01 1.58 ± 0.98
65 (27/38) 63 ± 14 22.7 ± 1.7 130.1 ± 5.3 76.9 ± 5.2 6.1 ± 0.87 6.99 ± 1.72 0.080 ± 0.016 4.76 ± 0.85 0.95 ± 0.34
64 (25/39) 67 ± 11 23.4 ± 1.8 129.5 ± 9.2 77.2 ± 7.9 6.4 ± 0.98 8.10 ± 3.05 0.13 ± 0.019 5.04 ± 0.91 1.52 ± 0.38
65 (25/40) 61 ± 13 23.1 ± 2.4 136.3 ± 11.5 80.8 ± 6.8 6.4 ± 0.99 7.77 ± 3.61 0.30 ± 0.15 5.77 ± 1.06 2.76 ± 1.12
NS NS <0.05 NS NS NS NS <0.0001 NS <0.0001
NS NS NS <0.0001 <0.001 NS NS <0.0001 <0.0001 <0.0001
NS <0.05 NS <0.0001 <0.01 NS NS <0.0001 <0.0005 <0.0001
Medication use, n (%) Insulin Statin
51 (26) 51 (26)
14 (22) 22 (34)
20 (31) 12 (19)
17 (26) 17 (26)
NS NS
NS NS
NS NS
Lipoprotein profiles Anion-exchange liquid chromatography n (male/female) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) IDL cholesterol (mmol/L) VLDL cholesterol (mmol/L) Chylomicron cholesterol (mmol/L)
194 (117/77) 1.29 ± 0.43 3.16 ± 0.89 0.24 ± 0.12 0.52 ± 0.42 0.11 ± 0.080
65 (27/38) 1.46 ± 0.51 2.90 ± 0.58 0.17 ± 0.070 0.22 ± 0.11 0.12 ± 0.088
64 (25/39) 1.31 ± 0.36 3.10 ± 0.91 0.23 ± 0.078 0.42 ± 0.15 0.10 ± 0.091
65 (25/40) 1.09 ± 0.34 3.47 ± 0.91 0.32 ± 0.14 0.92 ± 0.47 0.11 ± 0.057
NS <0.05 NS <0.0001 <0.0001 NS
NS <0.0001 <0.001 <0.0001 <0.0001 NS
NS <0.001 <0.05 <0.0005 <0.0001 <0.005
Agarose-gel electrophoresis with lipid-staining n (male/female) 186 (112/74) 36 ± 6.7 HDL ratio (%) LDL ratio (%) 41 ± 6.9 21 ± 7.9 VLDL ratio (%)
63 (25/38) 41 ± 5.6 42 ± 7.7 14 ± 5.6
63 (24/39) 36 ± 4.7 41 ± 6.4 21 ± 5.8
60 (23/37) 30 ± 4.7 41 ± 6.4 27 ± 6.9
NS <0.0001 NS <0.0001
NS <0.0001 NS <0.0001
NS <0.0001 NS <0.0001
Agarose-gel electrophoresis with cholesterol-staining 167 (102/65) n (male/female) 1.42 ± 0.44 HDL cholesterol (mmol/L) 3.16 ± 0.91 LDL cholesterol (mmol/L) VLDL cholesterol (mmol/L) 0.59 ± 0.31
45 (18/27) 1.63 ± 0.49 2.74 ± 0.54 0.41 ± 0.16
60 (21/39) 1.47 ± 0.39 3.05 ± 0.88 0.57 ± 0.16
62 (25/37) 1.27 ± 0.36 3.54 ± 1.03 0.91 ± 0.36
NS NS NS <0.0001
NS <0.0001 <0.0001 <0.0001
NS <0.01 <0.005 <0.0001
Data are means ± SD. Not significant is indicated by NS.
2. Materials and methods 2.1. Study subjects The 194 Japanese T2DM were recruited in Jikei University Kashiwa Hospital. This study was approved by the ethics committee at the Jikei University, and written informed consent was obtained from all patients. The diagnosis of diabetes was defined as currently being in anti-diabetic drug therapy or the clinical criteria of Japanese Diabetes Society; fasting blood glucose above 7.0 mmol/L or glycated hemoglobin A1c above 6.1%. HbA1c levels are expressed according to the procedures outlined by the Laboratory Test Committee of the Japan Diabetes Society.
2.2. Measurement of lipoproteins The venous blood was taken in all patients after a 12-hour overnight fast. Serum was stored at 4 ◦ C and was used for assays within 3 days after sampling. Total cholesterol and triglyceride levels were determined by enzymatic reagent, Cholestest-CHO and Cholestest-TG, Sekisui Medical Corp., Japan, respectively. AGE was performed using Electrophoresis Processing Analyzer Epalyzer2 (Herena Laboratories, Japan). Fat-Red 7B agarose (Herena Laboratories, Japan), and QuickGel LIPO and enzymatic cholesterol reagent CholeTriComb CH (Herena Laboratories, Japan) were used for the methods for lipid-staining and cholesterol-staining, respectively. The AEX-HPLC reported previously [8] was partially modified, and used for measuring cholesterol levels in HDL, LDL, IDL, VLDL,
and chylomicrons. A column, which contained 2.5 m of nonporous polymer-based gal with diethylaminoethyl ligands, and 3.0 mm ID × 15 mm in size, and a post-column reactor, which contained an enzymatic cholesterol reagent (TCHO-CL, Serotec Corp., Japan), were used. It took 10 min to complete the assay of one sample. The coefficient data of variation for lipoprotein classes in between-day assay (n = 10) and within-day assay (n = 10) using hyperlipidemic serum samples were 0.57–2.93% and 0.85–5.60%, respectively. 2.3. RLP-C analysis RLP fraction in serum was obtained using a mixed immunoaffinity gel containing monoclonal anti-human apoA-I (H-12) and anti-human apoB-100 (JI-H) antibodies (Japan Immunoresearch Labolatories, Japan) [2]. 2.4. Statistical analysis The data were presented as mean ± standard deviation (SD). Between-group differences were evaluated by F-test, and the comparisons were analyzed by using unpaired t-test for normal distribution data or Mann–Whitney U-test for not normal distribution data. P values <0.05 were considered significant. Correlations were estimated using Pearson product-moment correlation test. 3. Results T2DM patients (n = 194) were enrolled in this study, and were divided into three groups according to tertile of RLP-C levels.
H. Yoshida et al. / Atherosclerosis 222 (2012) 541–544
B
0.8
VLDL cholesterol by AEX-HPLC (mmol / L)
IDL cholesterol by AEX-HPLC (mmol / L)
A
0.6 r = 0.531 p < 0.0001 n = 194
0.4
0.2
0.0
2.5 2.0 1.5 1.0 r = 0.880 p < 0.0001 n = 194
0.5 0.0
0.0
0.4
0.8
1.2
0.0
40
D
30 20 r = 0.591 p < 0.0001 n = 186
10 0 0.0
0.4
0.8
0.4
0.8
1.2
RLP-C (mmol / L) VLDL cholesterol by AGE with cholesterol-staining (mmol / L)
50
VLDL ratio by AGE with lipid-staining(%)
RLP-C (mmol / L)
C
543
1.2
2.5 2.0 1.5 1.0 r = 0.765 p < 0.0001 n = 167
0.5 0.0 0.0
RLP-C (mmol / L)
0.4 0.8 RLP-C (mmol / L)
1.2
Fig. 1. Correlation between lipoprotein profiles and RLP-C. A, B, C, and D show the correlations of IDL and VLDL cholesterol by AEX-HPLC, VLDL ratio by AGE with lipid-staining and VLDL cholesterol by AGE with cholesterol-staining, respectively. The data of T1, T2, and T3 are represented by closed circles, open circles, and triangles, respectively.
HDL-cholesterol by AEX-HPLC decreased in order of lowest (T1), middle (T2), and highest (T3) tertile (Table 1). LDL-, IDL-, and VLDL-cholesterol by AEX-HPLC increased in order of T1, T2, and T3 (Table 1). HDL-, IDL-, and VLDL-cholesterol were significantly different between each other in the three groups. LDL and VLDL peak in 2 sera in T1, 1 serum in T2 and 5 sera in T3 could not be separated by AGE with lipid-staining. Therefore the data of 8 samples were excluded from the analysis. HDL ratio obtained by AGE with lipid-staining decreased in order of T1, T2, and T3 (Table 1). VLDL ratio obtained by AGE with lipid-staining increased in order of T1, T2, and T3 (Table 1). LDL and VLDL peak in 20 sera in T1, 4 sera in T2, and 3 sera in T3 could not be separated by AGE with cholesterol-staining. Therefore the data of 27 samples were excluded from the analysis. HDLcholesterol obtained by AGE with cholesterol-staining decreased in order of T1, T2, and T3 (Table 1). LDL- and VLDL-cholesterol obtained by AGE with cholesterol-staining increased in order of T1, T2, and T3 (Table 1). LDL- and chylomicron-cholesterol by AEX-HPLC and LDL ratio by AGE with lipid-staining did not correlated significantly with RLP-C, and the others correlated significantly, as follows: HDL-cholesterol by AEX-HPLC, r = −0.348; IDL-cholesterol by AEX-HPLC, r = 0.531; VLDL-cholesterol by AEX-HPLC, r = 0.880; HDL ratio by AGE with lipid-staining, r = −0.622; VLDL ratio by AGE with lipid-staining, r = 0.591; HDL-cholesterol by AGE with cholesterol-staining, r = −0.299; VLDL-cholesterol by AGE with cholesterol-staining, r = 0.765; LDL-cholesterol by AGE with cholesterol-staining, r = 0.265. Therefore, among these data, IDLand VLDL-cholesterol by AEX-HPLC, VLDL ratio by AGE with lipidstaining and VLDL-cholesterol by AGE with cholesterol-staining showed highly positive correlations with RLP-C (Fig. 1). IDL-cholesterol levels obtained by the AEX-HPLC were compared between the samples with and without a bunch of peak
contained LDL and VLDL in AGE. IDL-cholesterol measured by the AEX-HPLC in the lipid-staining 8 samples was not significantly higher than in the other samples; 0.36 ± 0.23 mmol/L and 0.23 ± 0.11 mmol/L, respectively. Also, IDL-cholesterol in the cholesterol-staining 27 samples was similar to the others; 0.22 ± 0.16 mmol/L and 0.24 ± 0.11 mmol/L, respectively. 4. Discussion Our developed AEX-HPLC method can measure cholesterol levels in lipoprotein classes, HDL, LDL, IDL, VLDL and chylomicron [8]. Remnant lipoproteins are triglyceride rich lipoproteins, including chylomicron remnant and VLDL remnant (IDL) [1,2]. The major component in RLP was VLDL remnant (IDL) because the fasting samples were measured in this study. Fig. 1 showed that IDL- and VLDL-cholesterol were significantly correlated with RLPC levels, but chylomicron-cholesterol was not. This chylomicron fraction in the AEX-HPLC don’t only contain chylomicron-related lipoprotein but also lipoprotein (a) [11]. RLP-C was reported as a risk maker for CAD, described as above [1–7]. However, the lack of standardization of RLP-C assay may provide a less quality of the measured data. A previous report showed that the values of two RLP-C kits are different in a part of patients [12]. The sequential ultracentrifugation method is a reference method to analyze lipoprotein, which is inconvenient and time-consuming [13]. The AEX-HPLC used in this study conveniently can measure IDL-cholesterol in addition to HDL-, LDL-, and VLDL-cholesterol. It is conceivable that IDL-cholesterol value may be important information to provide insight on the impaired metabolism of remnant lipoproteins when the measured values of two RLP-C kits are different. IDL-cholesterol levels of the AEX-HPLC in the samples with or without a bunch peak contained LDL and VLDL in AGEs were
544
H. Yoshida et al. / Atherosclerosis 222 (2012) 541–544
similar. We thought that lipoproteins in the samples with the one bunch peak contained LDL and VLDL might be highly heterogenous, and were not always available for IDL level determination. In the AGE analysis with lipid or cholesterol staining, VLDL levels only correlated with RLP-C, but VLDL itself is not RLP. However, AEX-HPLC can provide data of IDL-cholesterol correlated with RLP-C, suggesting the clinical usefulness of the AEX-HPLC because IDL is one of RLPs. IDL-cholesterol is high level in type III hyperlipoproteinemia, and the IDL levels obtained by an ultracentrifugation method in Japanese patients were reported 1.1 ± 0.5 mmol/L [14]. We showed that the IDL-cholesterol values obtained by the AEX-HPLC and the ultracentrifugation method were nearly equal [8]. IDL-cholesterol levels of T2DM in this study are likely to be lower than in the previous study [14]. In conclusion, this study indicates that IDL and VLDL levels measured by the AEX-HPLC were significantly different between T2DM patients by RLP-C levels, and strongly correlated with RLPC. IDL and VLDL-cholesterol, which are difficult to be conveniently estimated by conventional methods, in addition to HDL and LDLcholesterol can be measured at a time by the AEX-HPLC [8]. These results suggest that the AEX-HPLC which can provide IDL- and VLDL-cholesterol is more helpful than the AGE which can only provide VLDL level for estimation of lipid profiles in T2DM with high RLP-C. Sources of funding Research funds were provided in part by the Jikei University Research Fund (to H. Yoshida) and also in part by Grant-in-Aid for Scientific Research (23234567) from Japan Ministry of Health, Labor and Welfare (to H. Yoshida). Conflict of interests There were no conflicts of interest in this study.
References [1] Havel JR. Determination and clinical significance of triglyceride-rich lipoprotein remnants. In: Rifai N, Warnick GR, Dominiczak MH, editors. Handbook of lipoprotein testing. 2nd ed. Washington, D.C.: The American Association for Clinical Chemistry, Inc. Press; 2000. p. 565–80. [2] Fujioka Y, Ishikawa Y. Remnant lipoproteins as strong key particles to atherogenesis. Journal of Atherosclerosis and Thrombosis 2009;16:145–54. [3] Nakajima K, Saito T, Tamura A, et al. Cholesterol in remnant-like lipoproteins in human serum using monoclonal anti apo B-100 and anti apo A-I immunoaffinity mixed gels. Clinica Chimica Acta 1993;223:53–71. [4] Masuoka H, Kamei S, Ozaki M, et al. Predictive value of remnant-like particle cholesterol as an indicator of coronary artery stenosis in patients with normal serum triglyceride levels. Internal Medicine 2000;39:540–6. [5] Goff Jr DC, D’Agostino Jr RB, Haffner SM, Saad MF, Wagenknecht LE. Lipoprotein concentrations and carotid atherosclerosis by diabetes status: results from the insulin resistance atherosclerosis study. Diabetes Care 2000;23:1006–11. [6] Milicevic Z, Raz I, Beattie SD, et al. Natural history of cardiovascular disease in patients with diabetes: role of hyperglycemia. Diabetes Care 2008;31:S155–60. [7] Fukushima H, Sugiyama S, Honda O, et al. Prognostic value of remnantlike lipoprotein particle levels in patients with coronary artery disease and type II diabetes mellitus. Journal of the American College of Cardiology 2004;43:2219–24. [8] Hirowatari Y, Yoshida H, Kurosawa H, Doumitu K, Tada N. Measurement of cholesterol of major serum lipoprotein classes by anion-exchange HPLC with perchlorate ion-containing eluent. Journal of Lipid Research 2003;44:1404–12. [9] Hirowatari Y, Yoshida H, Fueki Y, et al. Measurement of cholesterol concentrations of major serum lipoprotein classes in haemodialysis patients by anionexchange chromatography. Annals of Clinical Biochemistry 2008;45:571–4. [10] Nakada Y, Kurosawa H, Tohyama J, Inoue Y, Ikewaki K. Increased remnant lipoprotein in patients with coronary artery disease-evaluation utilizing a newly developed remnant assay, remnant lipoproteins cholesterol homogenous assay (RemL-C). Journal of Atherosclerosis and Thrombosis 2007;14:56–64. [11] Hirowatari Y, Yoshida H, Kurosawa H, et al. Analysis of cholesterol levels in lipoprotein(a) with anion-exchange chromatography. Journal of Lipid Research 2010;51:1237–43. [12] Yoshida H, Kurosawa H, Hirowatari Y, et al. Characteristic comparison of triglyceride-rich remnant lipoprotein measurement between a new homogenous assay (RemL-C) and a conventional immunoseparation method (RLP-C). Lipids Health Disease 2008;7:18–23. [13] Schumaker VN, Puppione DL. Sequential flotation ultracentrifugation. Methods in Enzymology 1986;128:155–70. [14] Todo Y, Kobayashi J, Higashikata T, et al. Detailed analysis of serum lipids and lipoproteins from Japanese type III hyperlipoproteinemia with apolipoprotein E2/2 phenotype. Clinica Chimica Acta 2004;348:35–40.